Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Acid Sphingomyelinase-deficient Macrophages Have Defective Cholesterol Trafficking and Efflux

Acid Sphingomyelinase-deficient Macrophages Have Defective Cholesterol Trafficking and Efflux THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 48, Issue of November 30, pp. 44976 –44983, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Acid Sphingomyelinase-deficient Macrophages Have Defective Cholesterol Trafficking and Efflux* Received for publication, July 10, 2001, and in revised form, September 24, 2001 Published, JBC Papers in Press, September 28, 2001, DOI 10.1074/jbc.M106455200 Andrew R. Leventhal‡, Wengen Chen‡, Alan R. Tall‡, and Ira Tabas‡§¶ From the ‡Departments of Medicine and §Anatomy & Cell Biology, Columbia University, New York, New York 10032 Cholesterol efflux from macrophage foam cells, a key Cholesteryl ester (CE) -loaded macrophages, or foam cells, are prominent features of atherosclerotic lesions and play im- step in reverse cholesterol transport, requires traffick- ing of cholesterol from intracellular sites to the plasma portant roles in lesion progression (1, 2). During atherogenesis, membrane. Sphingomyelin is a cholesterol-binding mol- intimal macrophages internalize atherogenic lipoproteins, in- ecule that transiently exists with cholesterol in endo- cluding modified forms of LDL, that have been retained in the somes and lysosomes but is rapidly hydrolyzed by lyso- arterial subendothelium (1, 3, 4). This event directly leads to somal sphingomyelinase (L-SMase), a product of the esterification of cellular cholesterol by acyl-coenzyme A:choles- acid sphingomyelinase (ASM) gene. We therefore hy- terol O-acyltransferase (ACAT), resulting in “foam cell” forma- pothesized that sphingomyelin hydrolysis by L-SMase tion (3, 5). Foam cell formation can be prevented or reversed by enables cholesterol efflux by preventing cholesterol se- the process known as cellular cholesterol efflux (6). Cholesterol questration by sphingomyelin. Macrophages from wild- efflux is the initial step of reverse cholesterol transport, a type and ASM knockout mice were incubated with process whereby excess cholesterol in peripheral cells is deliv- [ H]cholesteryl ester-labeled acetyl-LDL and then ex- ered to the liver for excretion (6). Thus, the elucidation of posed to apolipoprotein A-I or high density lipoprotein. cellular molecules and pathways that facilitate and regulate In both cases, [ H]cholesterol efflux was decreased sub- cholesterol efflux is a major goal of research in the area of stantially in the ASM knockout macrophages. Similar atherosclerosis. results were shown for ASM knockout macrophages la- Atherogenic lipoproteins internalized by macrophages de- beled long-term with [ H]cholesterol added directly to liver their stores of cholesterol, which are mostly in the form of medium, but not for those labeled for a short period, CE, to late endosomes and/or lysosomes (3). Here, lysosomal suggesting defective efflux from intracellular stores but acid lipase hydrolyzes the CE to free cholesterol, which is then not from the plasma membrane. Cholesterol trafficking transported by poorly defined mechanisms to various sites in to acyl-coenzyme A:cholesterol acyltransferase (ACAT) the cell (3). A major site of transport is the plasma membrane, was also defective in ASM knockout macrophages. Using and from there the cholesterol can be effluxed to extracellular filipin to probe cholesterol in macrophages incubated acceptors, such as apoA-I and HDL, or transported to ACAT in with acetyl-LDL, we found there was modest staining in the endoplasmic reticulum for re-esterification (3, 6). Efflux to the plasma membrane of wild-type macrophages but apoA-I involves the initial formation of phospholipid-apoA-I bright, perinuclear fluorescence in ASM knockout macrophages. Last, when wild-type macrophages were particles by ABCA1-mediated phospholipid efflux, followed by incubated with excess sphingomyelin to “saturate” L- cholesterol efflux to these phospholipid-rich particles (7, 8). SMase, [ H]cholesterol efflux was decreased. Thus, Cholesterol efflux to HDL can be mediated by scavenger recep- sphingomyelin accumulation due to L-SMase deficiency tor type B1 (SR-B1) in those cell types that have relatively high leads to defective cholesterol trafficking and efflux, levels of this receptor, such as human monocyte-derived macro- which we propose is due to sequestration of cholesterol phages, but another mechanism must be involved in cells that by sphingomyelin and possibly other mechanisms. This have very low expression of SR-B1, such as mouse peritoneal model may explain the low plasma high density lipopro- macrophages (6, 9, 10). tein found in ASM-deficient humans and may implicate Given that all pathways of cholesterol efflux require choles- L-SMase deficiency and/or sphingomyelin enrichment of terol transport to the plasma membrane, the identification of lipoproteins as novel atherosclerosis risk factors. molecules mediating or regulating this transport process is an important goal. Thus far, only the molecules npc1, npc2 (HE1), and possibly lysobisphosphatidic acid have been shown to play a role in cholesterol transport to the plasma membrane, and * This work was supported by Specialized Center of Research in the molecular mechanisms are poorly understood (11–15). We Atherosclerosis Grant HL-56984 (to A. R. T. and I. T.) from the NHLBI, reasoned that another molecule, lysosomal sphingomyelinase National Institutes of Health and a research grant from Berlex Bio- (L-SMase), may also be involved in cholesterol transport from sciences (to I. T.). The Columbia University Confocal Microscope Facil- ity used for this study was established by National Institutes of Health lysosomes to the plasma membrane. L-SMase, a product of the Shared Instrument Grants S10 RR10506 and S10 RR13701 and the Lieber Foundation. The operation of the Facility is supported in part by National Institutes of Health Grant P30 CA13696 as part of the Herbert The abbreviations used are: CE, cholesteryl ester; ACAT, acyl-CoA: Irving Comprehensive Cancer Center at Columbia University. The cholesterol acyltransferase; apoA-I, apolipoprotein A-I; ASM, acid costs of publication of this article were defrayed in part by the payment sphingomyelinase; BSA, bovine serum albumin; DMEM, Dulbecco’s of page charges. This article must therefore be hereby marked “adver- modified Eagle’s medium; FBS, fetal bovine serum; HDL, high-density tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate lipoprotein; LDL, low-density lipoprotein; L-SMase, lysosomal sphingo- this fact. myelinase; NPC, Niemann-Pick C; PBS, phosphate-buffered saline; PS, ¶ To whom correspondence should be addressed: Dept. of Medicine, phosphatidylserine; SMase, sphingomyelinase; SR-BI, scavenger recep- Columbia University, 630 West 168th St., New York, NY 10032. Tel.: tor BI. 212-305-9430; Fax: 212-305-4834; E-mail: [email protected]. Y. Sun and A. R. Tall, unpublished data. 44976 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44977 tific Sonic Dismembranator-60 at level 5 for 5 min. [ H]Cholesterol (1 acid sphingomyelinase (ASM) gene, hydrolyzes sphingomyelin Ci/ml) was then added, and the solution was incubated overnight in a in late endosomes and lysosomes (16). Because SM avidly binds shaking water bath at 37 °C. The solution was then filtered through a cholesterol (17, 18), we hypothesized that sphingomyelin hy- 0.45-m filter. drolysis by L-SMase enables cholesterol transport by prevent- Cholesterol Efflux Assay—After labeling the cells, the medium was ing cholesterol sequestration by sphingomyelin. Of interest, changed to DMEM, 0.2% BSA containing 20 g/ml apoA-I, 15 g/ml humans with ASM deficiency (types A and B Niemann-Pick HDL ,or10 g/ml HDL . At the indicated time points, 100 l of media 2 3 was removed and spun for 5 min at 14,000 rpm in a microcentrifuge to disease) have low plasma HDL levels (19, 20), which could remove cellular debris, and the radioactivity in this fraction of media result from defective cholesterol efflux (cf. Ref. 21). was quantified by liquid scintillation counting. After the last time point, In this context, we show herein that macrophages from ASM the remainder of the media was removed, and the cells were dissolved knockout mice, which lack L-SMase (22, 23), have a defect in N NaOH at room temperature for 5 h. A 100-l aliquot of in1mlof0.1 cholesterol efflux to both apoA-I and HDL, a decrease in the the cell lysate was counted, and the percent efflux was calculated as esterification of cellular cholesterol, and an accumulation of [(media cpm)  (cell  media cpm)]  100. Note that there was no statistical difference in cellular counts/min between wild-type and cholesterol in perinuclear vesicles. Moreover, a defect in cho- ASM-deficient macrophages using any of the three [ H]cholesterol la- lesterol efflux was also observed in wild-type macrophages that beling methods. To obtain the value for acceptor-stimulated efflux, the internalized a large amount of sphingomyelin. These data sup- percent efflux in the absence of acceptor (i.e. DMEM, 0.2% BSA without port the hypothesis that intracellular accumulation of sphin- apoA-I or HDL) was subtracted from the percent efflux in the presence gomyelin due to L-SMase deficiency or to internalization of of acceptor; the basal efflux values were 10% of those in the presence excess sphingomyelin leads to cholesterol sequestration and of acceptor. Whole Cell Cholesterol Esterification Assay—In the first method, defective cholesterol trafficking and efflux. macrophages were incubated in DMEM, 0.2% BSA containing 0.1 m EXPERIMENTAL PROCEDURES C]oleate complexed with albumin and 3 g/ml acetyl-LDL. In the Materials—The Falcon tissue culture plasticware used in these stud- second method, cells were labeled with [ H]cholesterol by long-term ies was purchased from Fisher Scientific Co. Tissue culture media and incubation with [ H]cholesterol-labeled medium as described above, g/ml unlabeled acetyl-LDL or 5 M other tissue culture reagents were obtained from Life Technologies, Inc. followed by incubation with 3 25-hydroxycholesterol for up to 12 h. At the indicated time points, the Fetal bovine serum (FBS) was obtained from Hyclone Laboratories cells were washed two times with cold PBS, and the cell monolayers (Logan, UT). Alexa 543-labeled dextran-10,000 was purchased from were extracted twice with 0.5 ml of hexane/isopropyl alcohol (3:2, v/v) Molecular Probes, Inc. (Eugene, OR). All radiochemicals were pur- for 30 min at room temperature. Whole cell cholesterol esterification chased from Perkin-Elmer Life Sciences, Inc. (Boston, MA). All other activity was assayed by determining the cellular content of cholesteryl chemicals and reagents were from Sigma, and all organic solvents were 14 3 from Fisher Scientific Co. C]oleate or [ H]cholesteryl ester by thin-layer chromatography (28). N NaOH, and aliquots Macrophages—Littermate wild-type and homozygous ASM knockout The cell monolayers were dissolved in 1 ml of 0.1 were assayed for protein by the method of Lowry et al. (29). mice were obtained by breeding ASM heterozygous knockout mice, In Vitro ACAT Assay—Macrophages in 100-mm dishes were incu- which were provided by Dr. Edward Schuchman, Mt. Sinai School of bated for 24 h in DMEM, 10% lipoprotein-deficient serum and then Medicine (22, 23). Macrophages were harvested from the peritoneum of scraped in 2.5 ml of ice-cold 20 mM potassium phosphate buffer, 2 mM these mice 3 days after the intraperitoneal injection of 40 g of con- dithiothreitol, pH 7.4. This suspension was sonicated for five 3-s bursts canavalin A in 0.5 ml of PBS and then cultured as described previously on level 10 of a Fisher Scientific Sonic Dismembranator-60. Aliquots of (24). the cell lysates were removed for protein determination, and then Lipoproteins and Liposomes—LDL (d, 1.020 –1.063 g/ml), HDL (d, 130-l of the cell lysates were added to capped glass test tubes that 1.063–1.125 g/ml), HDL (d, 1.125–1.21 g/ml) from fresh human plasma contained 20 l of 0.5 mg/ml cholesterol-rich PS liposomes in PBS, were isolated by preparative ultracentrifugation as described (25). which were prepared as described previously (30), or 20 l of PBS alone. Acetyl-LDL was prepared by reaction with acetic anhydride (26) and After a 15-min incubation at 37 °C, 30 l of 40 mg/ml fatty acid-free labeled with [ H]CE as described (27). Briefly, label was transferred BSA in 100 mM potassium phosphate buffer, 2 mM dithiothreitol, pH from [ H]cholesteryl oleate-containing liposomes to HDL by CE trans- 7.4, and 20 lof25 M [ fer protein, followed by CE transfer protein-mediated transfer of the C]oleoyl-CoA (40 Ci/mol) were added. After [ H]CE from HDL to acetyl-LDL, which were then separated from each a 15-min incubation at 37 °C, the reaction was stopped by adding 2 ml g of unlabeled CE internal other by density ultracentrifugation. The final specific activity of the of 2:1 chloroform:methanol containing 250 standard. The mixture was vortexed and allowed to sit at room tem- labeled acetyl-LDL was 17 cpm/ng of protein. [ H]Cholesterol-contain- perature for 30 min. Four hundred l of 0.88% KCl were added to each ing liposomes were prepared by first mixing 100 Ci of [ H]cholesterol tube, and the tubes were then centrifuged at 1000 rpm for 10 min. The and 2.8 mg of phosphatidylserine (PS) in chloroform, in the absence or lower organic phase was collected and separated by thin layer chroma- presence of 1.25 mg of sphingomyelin. The chloroform was evaporated tography, and the CE spot was scraped and counted. under nitrogen, and the dried lipids were sonicated in 3 ml of PBS at Filipin Staining of Free Cholesterol and Fluorescence Microscopy— 4 °C under argon using 15 4-min bursts at setting number 2 on a For filipin staining, the method of Blanchette-Mackie et al. (31) was Branson 450 sonicator equipped with a tapered microtip. [ C]Sphin- employed. Briefly, a 0.05 mg/ml filipin solution was made from a 5 gomyelin-containing liposomes were made in the same manner except 14 3 mg/ml stock in Me 1.2 Ci of [ C]sphingomyelin was used instead of [ H]cholesterol. For SO by dilution with 10% FBS in PBS. Macrophages certain experiments in which we needed to fluorescently label PS- were cultured on polylysine-coated coverslip bottom dishes (32) and sphingomyelin liposomes, the PBS contained 0.5 mg/ml Alexa 543- preincubated for 24 h in DMEM, 10% LPDS. After the incubation labeled dextran-10,000; after sonication, the solution was dialyzed us- described in the legend to Fig. 5, the cells were washed with PBS and ing dialysis tubing with a 100,000-MW pore size. fixed with 3% paraformaldehyde in PBS for1hat room temperature. Labeling of Cells with [ H]Cholesterol—Cell monolayers were labeled The macrophages were subsequently rinsed three times with PBS and by one of three methods. For labeling by acetyl-LDL or PS liposomes, incubated with 1.5 mg of glycine/ml of PBS for 10 min at room temper- cells were incubated for4hin DMEM, 0.2% BSA containing 10 g/ml ature. Next, the cells were incubated with the 0.05 mg/ml filipin solu- 3 3 [ H]CE-acetyl-LDL or [ H]cholesterol-labeled phosphatidylserine-lipo- tion for2hat room temperature. The cells were then washed three somes such that the final PS concentration was 93 g/ml. For long-term times with PBS and viewed with a Zeiss Axiovert S100 epifluorescence labeling of cellular cholesterol, macrophages were grown for 24 h in microscope using a UV filter set (340 –380-nm excitation, 70-nm di- DMEM, 10% FBS containing 0.5 Ci/ml [ H]cholesterol, then equili- chroic, 430-nm long pass filter). For the double-label filipin/Alexa 543- brated for 12 h in DMEM, 0.2% BSA. The [ H]cholesterol-containing dextran study displayed in Fig. 6, dual-photon microscopy was con- medium was made by adding the labeled cholesterol, which was in ethanol, dropwise (0.5 l/ml) into DMEM, 10% FBS while stirring at 3 3 37 °C. After a 30-min incubation at 37 °C, the medium was passed In the experiments with apoA-I, the percent of [ H]cholesterol in the through a 0.45-m filter. For short-term labeling experiments, macro- medium represents net efflux, because the medium contains no unla- phages were incubated for 15 min at room temperature in DMEM beled cholesterol at the beginning of the experiment. In the experiments containing methyl--cyclodextrin-[ H]cholesterol complex. To prepare with HDL, however, this method used in this study does not distinguish this solution, methyl--cyclodextrin and unlabeled cholesterol (8:1 mo- between net cholesterol efflux and exchange of cholesterol between the lar ratio) were dissolved in DMEM and sonicated using a Fisher Scien- cells and HDL. 44978 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages ducted using an LSM 510 nonlinear optics (NLO) Zeiss dual photon confocal microscope equipped with a 100X/1.3 NA Plan-Neofluor objec- tive lens. For Alexa-546, the pinhole was adjusted to produce an optical section of 1.0 m. A helium-neon laser (543 nm) was used for excitation and a 560-nm long-pass was used as an emission filter. For filipin, the pinhole was completely open, and a Coherent titanium-sapphire laser tuned to 800 nm was used for excitation and a 390 – 465-nm band pass was used as an emission filter. Pilot studies indicated that this config- uration allowed good resolution of both Alexa-546 and filipin fluores- cence without cross-over fluorescence. Statistics—Results are given as mean  S.E. (n  3); absent error bars in the figures signify S.E. values smaller than the graphic symbols. For the data in Fig. 7A, the unpaired, two-tailed t test was used to determine statistically significance. RESULTS Efflux of Acetyl-LDL-derived [ H]Cholesterol and Cellular Cholesterol Labeled by Long-term Incubation with [ H]Choles- terol-containing Medium Is Defective in ASM Knockout Macro- phages—To assess the role of macrophage L-SMase in the efflux of acetyl-LDL-derived cholesterol, peritoneal macro- phages from wild-type and ASM knockout mice were incubated with [ H]CE-labeled acetyl-LDL for 4 h and then chased in serum-free medium containing lipid-free apoA-I, which medi- ates cholesterol efflux through the ABC1 pathway, and HDL , which involves mostly other pathways (33). [ H]Cholesterol derived from [ H]CE-acetyl-LDL traffics through late endo- somes and lysosomes, which are known sites of L-SMase activ- ity and SM accumulation in ASM-deficient cells (16). As shown in Fig. 1, cholesterol efflux to both apoA-I (panel A) and HDL (panel B) was reduced by 60 –70% in the ASM knockout macrophages. For example, efflux to apoA-I at 12 h was 4.3% in ASM knockout cells versus 13.4% in wild-type cells (Fig. 1A). Cellular cholesterol can also be labeled by long-term incuba- tion with [ H]cholesterol added directly to the medium. In this case, the plasma membrane is labeled first, followed by equil- ibration with intracellular stores that probably include recy- cling endosomes and the trans-Golgi network (34, 35). To as- sess the efflux or exchange of cellular cholesterol pools labeled in this manner, wild-type and ASM knockout macrophages were labeled for 24 h in the presence of 0.5 Ci/ml [ H]choles- terol. After a 12-h equilibration period, the cells were incubated with HDL, and [ H]cholesterol in the medium was measured. FIG.1. Efflux of acetyl-LDL-derived [ H]cholesterol from wild- As shown in Fig. 2A, there was an 50% decrease in the 3 type and ASM knockout macrophages. Monolayers of peritoneal percent [ H]cholesterol in the medium of ASM knockout macro- macrophages from wild-type (closed circles) or ASM knockout (open phages under these conditions. circles) mice were incubated with 10 g/ml [ H]CE-labeled acetyl-LDL One interpretation of the data in Fig. 2A is that ASM knock- in DMEM, 0.2% BSA for 4 h. The cells were then rinsed and incubated with fresh medium containing either 20 g/ml human apoA-I (A)or10 out macrophages have a defect in the efflux or exchange of g/ml human HDL (B) for the indicated times. H-Labeled cpm in the plasma membrane cholesterol. To address this issue, wild-type media and cells were measured to calculate percent [ H]cholesterol in and ASM knockout macrophages were incubated for 15 min at the medium. room temperature with [ H]cholesterol-charged methyl--cy- clodextrin, and then chased for 1 or3hin medium without macrophages was shown using HDL (Fig. 2B and data not label but containing HDL. According to Lange et al. (34), the shown), which mediates efflux by an ABCA1-independent 15-min labeling procedure labels mostly plasma membrane mechanism (7). SR-B1 almost certainly is not involved, because cholesterol, which then eventually equilibrates with intracel- its expression is very low in mouse peritoneal macrophages, lular pools. As shown in Fig. 2B, medium [ H]cholesterol at the and an SR-B1 neutralizing antibody does not block cholesterol early time point was the same in the two cell types, while after efflux from these cells to HDL. Given these data, we focused a 3-h chase it was less in the ASM knockout macrophages. our efforts on the hypothesis that decreased efflux in ASM These data suggest that transfer of cholesterol directly from knockout macrophages was caused by defective intracellular the plasma membrane to HDL is not defective in ASM knock- cholesterol trafficking. out macrophages but that efflux from the intracellular site(s) Trafficking of Cholesterol to ACAT Is Defective in ASM that accumulates nonlipoprotein cholesterol is affected. Knockout Macrophages—To determine if other cellular choles- In view of these data, we considered the possibility that a terol trafficking pathways were altered in ASM knockout decrease in expression of ABCA1 or SR-B1 could explain the macrophages, we assessed the ability of the cells to esterify defect in efflux. However, quantitative polymerase chain reac- cholesterol, which requires cholesterol transport to ACAT in tion showed no difference in ABCA1 mRNA expression between the endoplasmic reticulum (3). First, wild-type and ASM wild-type and ASM knockout macrophages (data not shown). Although there could be differences in ABCA1 localization or activity, a defect in cholesterol efflux from ASM knockout W. Chen and A. R. Tall, unpublished data. Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44979 FIG.3. Cholesterol esterification in wild-type and ASM knock- out macrophages incubated with acetyl-LDL. A, macrophages from wild-type (closed circles) or ASM knockout (open circles) mice were FIG.2. Efflux of cellular cholesterol labeled by long-term in- incubated for the indicated times in DMEM, 0.2% BSA containing 3 3 14 cubation of macrophages with [ H]cholesterol-containing me- g/ml acetyl-LDL and 0.1 mM [ C]oleate. Cellular lipids were then dium. A, macrophages from wild-type (closed circles) or ASM knockout extracted and assayed for cholesteryl [ C]oleate. B, the macrophages (open circles) mice were incubated for 24 h in DMEM, 10% FBS con- were incubated for 24 h in DMEM, 10% FBS containing 0.5 Ci/ml taining 0.5 Ci/ml [ H]cholesterol. The cells were then incubated for an [ H]cholesterol. The cells were then incubated for an additional 12 h in additional 12 h in DMEM, 0.2% BSA without label, and then finally DMEM, 0.2% BSA without label, and then finally incubated in the same incubated in the same medium containing 10 g/ml HDL for the medium containing 3 g/ml acetyl-LDL for the indicated times. Cellular 3 3 indicated times. Percent [ H]cholesterol in the medium was then deter- lipids were extracted and assayed for [ H]cholesteryl ester. mined. B, macrophages from wild-type (solid bars) or ASM knockout (hatched bars) mice were incubated for 15 min at room temperature in protocol, there was even a greater defect in cholesterol esteri- DMEM, 0.2% BSA containing 1 Ci/ml [ H]cholesterol-charged methyl- fication in the ASM knockout cells. -cyclodextrin. The cells were then rinsed and incubated with fresh A similar experiment to that in Fig. 3B was conducted, but medium containing 15 g/ml HDL for either 1 or 3 h, and then percent 25-hydroxycholesterol was used as the stimulator of ACAT [ H]cholesterol in the medium was determined. instead of acetyl-LDL (36). In this scenario, the formation of knockout macrophages were incubated with acetyl-LDL and [ H]cholesteryl ester was almost totally abolished in the ASM 14 14 [ C]oleate, and the formation of cholesteryl [ C]oleate was knockout macrophages (Fig. 4A). We next determined if these assayed. Under these conditions, acetyl-LDL-derived choles- data could be explained by decreased active ACAT enzyme in terol mixes with cellular cholesterol in the plasma membrane, the ASM knockout macrophages. Wild-type and ASM knockout and when a threshold level of cholesterol is reached, this cho- macrophages were incubated for 24 h in the absence of lipopro- lesterol is transported to ACAT and esterified (36, 37). As teins, and then lysates from these cells were assayed for ACAT shown in Fig. 3A, ASM knockout macrophages demonstrated a activity in vitro in the absence or presence of exogenous cho- 30 –50% decrease in CE formation. Next, the macrophage cho- lesterol (Fig. 4B). In the absence of exogenous cholesterol, lesterol pools were labeled by long-term incubation with ACAT activity was low and similar in both cell types; in the [ H]cholesterol-labeled medium (above), and the formation of presence of exogenous cholesterol, ACAT activity was higher [ H]cholesteryl ester in response to subsequent incubation with than that in the absence of cholesterol and, surprisingly, some- acetyl-LDL was assayed (Fig. 3B). Using this experimental what greater in the lysates of ASM knockout mice. Although we 44980 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages FIG.5. Free cholesterol distribution in wild-type and ASM knockout macrophages. Macrophages from wild-type (A)orASM knockout (B) mice were incubated with DMEM, 0.2% BSA containing 100 g of acetyl-LDL/ml for 4 h and then chased in medium without lipoproteins for an additional 4 h. The cells were then fixed, stained with filipin, and visualized by fluorescence microscopy. Bar,5 m. demonstrate intracellular sequestration of cholesterol in ASM knockout macrophages. We next determined if the free cholesterol that accumulates in ASM-deficient macrophages is localized solely or mostly in late endosomes and lysosomes. Macrophages from ASM knock- out mice were incubated with acetyl-LDL and Alexa 546-la- beled dextran for 4 h, and then chased in medium alone for an additional 4 h. The dextran is taken up by fluid-phase pinocy- tosis and, after a 4-h chase, labels late endosomes and lyso- somes; for example, in macrophages and other cells, the stain- ing pattern with dextran is essentially identical to that with the lysosomal marker LAMP1 (38). After fixing and staining with filipin to detect the cholesterol, the cells were viewed by dual-photon confocal fluorescence microscopy. An example of one ASM-deficient macrophage is shown in Fig. 6, A–C, and a cluster of macrophages is shown in Fig. 6, D–F. Panels A and D show the filipin staining pattern (pseudocolored green), panels B and E the Alexa-dextran pattern, and panels C and F the merged images. As in Fig. 5B, free cholesterol accumulated in a perinuclear distribution (A and D). Alexa-dextran also accu- mulated in a perinuclear pattern (B and E), which is consistent with a late endosome/lysosome pattern (38). Although there FIG.4. 25-Hydroxycholesterol-induced cholesterol esterifica- was some co-localization of cholesterol and Alexa-dextran (yel- tion and in vitro ACAT activity in wild-type and ASM knockout macrophages. A, macrophages from wild-type (closed circles)orASM low vesicles in panels C and F), there were also a number of knockout (open circles) mice were incubated for 24 h in DMEM, 10% vesicles that were distinct, as indicated by the pure red and FBS containing 0.5 Ci/ml [ H]cholesterol. The cells were then incu- pure green vesicles in panels C and F. Thus, the cholesterol that bated for an additional 12 h in DMEM, 0.2% BSA without label, and accumulates in ASM-deficient macrophages accumulates both then finally incubated in the same medium containing 5 g/ml 25- hydroxychoelsterol for the indicated times. Cellular lipids were ex- in dextran-containing late endosomes/lysosomes and in other tracted and assayed for [ H]cholesteryl ester. B, lysates of macrophages perinuclear sites (see “Discussion”). from wild-type (solid bars) or ASM knockout (hatched bars) mice were Delivery of Excess SM to Cells Can Decrease Cholesterol assayed for ACAT activity in the absence or presence of 125 M Efflux in Wild-type Macrophages—The data in this report sup- cholesterol. port the hypothesis that the accumulation of SM in late endo- somes, lysosomes, and probably other sites as a result of do not know the cause of this increase, the data strongly sug- L-SMase deficiency leads to a defect in cholesterol efflux. Ac- gest that the defect in cholesterol esterification in ASM knock- cording to this hypothesis, it might be possible to “saturate” out macrophages (Fig. 3) reflects a defect in cholesterol traf- L-SMase activity in wild-type cells by delivering excess SM, ficking to ACAT rather than decreased active ACAT enzyme. thus resulting in a decrease in cholesterol efflux. To test this ASM Knockout Macrophages Accumulate Free Cholesterol in prediction, wild-type macrophages were incubated for 4 h with Perinuclear Sites—To directly assess the intracellular distribu- PS liposomes that had tracer amounts of [ H]cholesterol and tion of free cholesterol, wild-type and ASM knockout macro- either no sphingomyelin or sphingomyelin (sphingomyelin:PS, phages were incubated with acetyl-LDL (4-h incubation fol- 1:2 molar ratio). Note that PS-containing liposomes, like acetyl- lowed by 4-h chase) and then stained with filipin, a fluorescent LDL, are endocytosed by macrophages through the type A free cholesterol-binding molecule. The filipin-staining pattern scavenger receptor pathway and possibly through a newly iden- in wild-type macrophages was modest in intensity, although tified PS receptor (39, 40). The cells were then rinsed and greater than in cells not exposed to acetyl-LDL, and the plasma incubated for 12 h with fresh medium containing HDL, after membrane was the principal site of staining (Fig. 5A). In strik- which the percent [ H]cholesterol in the medium was deter- ing contrast, ASM knockout macrophages showed bright, pe- mined. Using parallel experiments in which the sphingomye- rinuclear, punctate fluorescence (Fig. 5B). These data directly lin-PS liposomes contained tracer [ C]SM, we calculated that Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44981 FIG.6. Free cholesterol and late en- dosome/late lysosome distribution in wild-type and ASM knockout macro- phages. Macrophages from ASM knock- out were incubated with medium contain- ing 100 g of acetyl-LDL/ml and 0.5 mg/ml Alexa 546-labeled dextran-10,000 for 4 h and then chased in medium with- out lipoproteins or dextran for an addi- tional 4 h. The cells were then fixed, stained with filipin, and visualized by dual-photon confocal fluorescence micros- copy. An example of one ASM-deficient macrophage is shown in A-C, and a clus- ter of macrophages is shown in D-F. Pan- els A and D show the filipin staining pat- tern (pseudocolored green), panels B and E the Alexa-dextran pattern, and panels C and F the merged images. Bar,1 m. FIG.7. Effect of liposomal sphingomyelin content on the efflux of liposome-derived cholesterol from wild-type and ASM knockout macrophages. A, macrophages from wild-type (solid bars) or ASM knockout (hatched bars) mice were incubated for 4 h with DMEM, 0.2% BSA containing PS liposomes that had tracer amounts of [ H]cholesterol and either no sphingomyelin (SM) or sphingomyelin (sphingomyelin:PS, 1:2 molar ratio; SM). In all incubations, the final PS concentration was 93 g/ml, and the incubation volume was 0.5 ml. The cells were then rinsed and incubated for 12 h with fresh medium containing 15 g/ml HDL , after which the percent [ H]cholesterol in the medium was determined. The difference in efflux between SM and SM liposomes in the wild-type macrophages were statistically significant (p  0.001). B, to verify that the liposomes were internalized, macrophages were incubated with sphingomyelin/PS liposomes labeled with Alexa 543-dextran-10,000 for4hand then chased for an additional3hin medium without liposomes. C, for comparison of the image in B with a cell-surface pattern, macrophages were incubated with the Alexa-labeled liposomes for1hat4 °C. Bar,5 m. this protocol led to the delivery of 6.6  0.8 g of sphingomy- prior incubation in serum-containing medium) and endogenous elin/mg of cell protein. As expected for receptor-mediated en- SM (41). These data are consistent with a model in which the docytic targeting to late endosomes and lysosomes, the vesicles activity of L-SMase in normal macrophages can be partially were internalized and delivered to perinuclear organelles (Fig. “saturated” by the endocytosis of SM, leading to intracellular 7B); for comparison, Fig. 7C shows a typical cell-surface distri- SM accumulation and defective cholesterol trafficking. bution of macrophages incubated with vesicles at 4 °C. DISCUSSION As shown in the first pair of bars in Fig. 7A, there was a statistically significant decrease in efflux of [ H]cholesterol The results of this study have implications ranging from the delivered in the sphingomyelin-containing liposomes compared basic cellular biology of lipid trafficking to the important phys- with [ H]cholesterol delivered in liposomes without SM. As iologic area of macrophage cholesterol efflux in atherosclerosis. expected, [ H]cholesterol efflux from ASM knockout macro- The mechanisms and regulation of cholesterol trafficking from phages was decreased to an even greater degree (second pair of lysosomes to the plasma membrane are poorly understood. bars in Fig. 7A). Interestingly, the presence of excess sphingo- Data from mutant cells indicate that the proteins npc1 and myelin did not further decrease efflux in these cells, presum- npc2 (HE1) have functions along this pathway (11–15), and ably because intracellular SM levels were already very high antibody experiments may suggest a role for the lipid lyso- due to L-SMase deficiency in the face of exogenous SM (from bisphosphatidic acid (15). The mechanism of action of these 44982 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages molecules, however, is not known. The idea that the sphingo- (50) made the interesting observation that lysosomal choles- myelin content of lysosomes could be an important regulatory terol accumulation in Chinese hamster ovary cells induced by factor is based upon the ability of this lipid to bind cholesterol either the NPC mutation or progesterone causes a decrease in (17, 18). Indeed, Aviram and colleagues (42) suggested that the the enzymatic activity of L-SMase. Therefore, in view of our data, it is possible that the cholesterol trafficking defects in ability of oxysterols in oxidized LDL to inhibit L-SMase in murine and human macrophages may account for the accumu- NPC cells and progesterone-treated cells may be amplified by secondary inhibition of L-SMase. lation of lysosomal FC under these conditions, although molec- The findings in this report may have important implications ular genetic proof was not provided to support their hypothesis. for foam cell biology. Macrophages in advanced lesions are In contrast, cholesterol esterification and trafficking to the known to accumulate large amount of free cholesterol, much of plasma membrane were reported as being normal in several which appears to be in lysosomes (51–55). On the one hand, it lines of fibroblasts from humans with types A and B Niemann- is possible that exposure of these macrophages to oxidized LDL Pick disease (43, 44). It is possible that the human fibroblasts or oxysterols, by inhibiting L-SMase (42), or SM-rich lipopro- data could have been influenced by residual L-SMase activity teins, by “saturating” L-SMase (see Fig. 6), may contribute to in these cells (16) or by inherent differences in cholesterol this event. Regarding this latter possibility, Jiang et al. (56) trafficking between fibroblasts and macrophages. recently found that a high plasma SM level is an independent Our working hypothesis states that defective cholesterol risk factor for coronary artery disease in humans. On the other trafficking in ASM knockout macrophages is due to sequestra- hand, the accumulation of lysosomal cholesterol, even if caused tion of cholesterol by sphingomyelin. This model can readily by another process, might be expected to secondarily inhibit explain the acetyl-LDL-cholesterol trafficking data, because L-SMase (50), which could further exacerbate the accumulation acetyl-LDL-derived cholesterol traffics through lysosomes, of lysosomal cholesterol and inhibit cholesterol efflux. which is a known site of SM accumulation in ASM-deficient The findings in this study raise additional questions related cells (16). With regard to those experiments in which the to lipoprotein abnormalities and atherosclerotic risk in humans macrophages were labeled by long-term incubation with with ASM deficiency (types A and B Niemann-Pick disease). [ H]cholesterol-containing medium, it is possible that this These subjects have markedly low plasma HDL levels (19, 20). method also labels lysosomal pools of cholesterol. However, Given that low plasma HDL can result from defective cellular when a similar method was used in Chinese hamster ovary cholesterol efflux (21), our current data may provide a mecha- cells to incorporate the fluorescent sterol dehydroergosterol or nism that contributes to this lipoprotein abnormality. Regard- cholesterol itself, followed by filipin labeling, the major sites of ing atherosclerotic risk, one must focus on type B Niemann- accumulation were the endosomal recycling compartment and Pick patients, who survive to adulthood due to low levels of the trans-Golgi network (35). If this were the case in macro- residual ASM activity, and type A or type B obligate heterozy- phages, it would indicate defective trafficking and efflux of gotes, who are reported to be “normal” (16). In considering the non-lysosomal cholesterol and therefore might imply that ASM potential atherogenic effects of L-SMase deficiency in these deficiency leads to SM accumulation in the endosomal recycling subjects, it is interesting to consider that the ASM gene also compartment, trans-Golgi network, or other nonlysosomal sites gives rise to secretory SMase (57). Because secretory SMase (cf. Refs. 15, 34, and 45). This idea might also provide an promotes the subendothelial aggregation and retention of li- explanation for our finding that not all of the acetyl-LDL- poproteins, leading to enhanced foam cell formation, S-SMase derived free cholesterol that accumulates in ASM-deficient deficiency, unlike L-SMase deficiency, may decrease choles- macrophages is in dextran-containing late endosomes and ly- terol accumulation in lesional macrophages (57). Therefore, sosomes (Fig. 6). Detailed sphingomyelin localization studies in while the deficiency of S-SMase in these subjects might be ASM knockout macrophages will be required to sort out these protective, defective L-SMase activity, by inhibiting cholesterol possibilities. efflux from lesional macrophages and possibly by leading to low One must also consider the possibility that direct sequestra- HDL levels, may promote atherosclerotic vascular disease. tion of cholesterol by SM is not the only mechanism behind defective cholesterol trafficking in ASM knockout macro- Acknowledgments—We thank Dr. Edward Schuchman for providing the ASM heterozygote knockout mice that were used to generate the phages. In this context, there is evidence that initial accumu- wild-type and ASM homozygous knockout mice used in this study, Dr. lation of unesterified cholesterol in lysosomes or late endo- Anthony Johns (Berlex Biosciences) for performing the ABCA1 quanti- somes can lead to secondary defects in vesicular trafficking. For tative polymerase chain reaction assay, and Theresa Swayne for assist- ance with the fluorescence microscopy studies. example, the defective trafficking of lactosyl ceramide in ASM- deficient fibroblasts and the defects in late endosomal tubu- REFERENCES lovesicular trafficking observed in NPC cells can be corrected 1. Ross, R. (1995) Annu. Rev. Physiol. 57, 791– 804 by cellular cholesterol depletion (46, 47), and defects in the 2. Libby, P., and Clinton, S. K. (1993) Curr. Opin. Lipidol. 4, 355–363 3. Tabas, I. (2000) Biochim. Biophys. Acta 1529, 164 –174 trafficking of sucrose can be induced in normal cells by cellular 4. Williams, K. J., and Tabas, I. (1995) Arterioscler. Thromb. Vasc. Biol. 15, cholesterol enrichment (48). Thus, it is possible that abnormal 551–561 SM accumulation in ASM knockout macrophages causes an 5. Brown, M. S., and Goldstein, J. L. (1983) Annu. Rev. Biochem. 52, 223–261 6. Tall, A. R. (1998) Eur. Heart J. 19, A31-A35 initial accumulation of FC in lysosomes or some other site, but 7. Wang, N., Silver, D. L., Thiele, C., and Tall, A. R. (2001) J. Biol. Chem. 276, that at least some of the trafficking and efflux defects observed 23742–23747 8. Fielding, P. E., Nagao, K., Hakamata, H., Chimini, G., and Fielding, C. J. in our studies are due to membrane vesiculation defects sec- (2000) Biochemistry 39, 14113–14120 ondary to this initial cholesterol accumulation. 9. Ji, Y., Jian, B., Wang, N., Sun, Y., Moya, M. L., Phillips, M. C., Rothblat, G. H., Swaney, J. B., and Tall, A. R. (1997) J. Biol. Chem. 272, 20982–20985 The cholesterol trafficking defect described in this report is 10. Rothblat, G. H., Llera-Moya, M., Atger, V., Kellner-Weibel, G., Williams, D. L., similar to that which occurs in fibroblasts from patients with and Phillips, M. C. (1999) J. Lipid Res. 40, 781–796 Niemann-Pick C (NPC) disease (11–14). Moreover, recent work 11. Liscum, L., and Klansek, J. J. (1998) Curr. Opin. Lipidol. 9, 131–135 12. Blanchette-Mackie, E. J. (2000) Biochim. Biophys. Acta 1486, 171–183 from our groups has shown that macrophages from NPC mice 13. Ory, D. S. (2000) Biochim. Biophys. Acta 1529, 331–339 have a defect in cholesterol trafficking (49). Although the mo- 14. Naureckiene, S., Sleat, D. E., Lackland, H., Fensom, A., Vanier, M. T., Wattiaux, R., Jadot, M., and Lobel, P. (2000) Science 290, 2298 –2301 lecular etiologies are distinct, important interrelationships be- 15. Kobayashi, T., Beuchat, M. H., Lindsay, M., Frias, S., Palmiter, R. D., tween the cholesterol trafficking defects in ASM deficiency and Sakuraba, H., Parton, R. G., and Gruenberg, J. (1999) Nat. Cell Biol. 1, Niemann-Pick C disease may exist. For example, Reagan et al. 113–118 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44983 16. Schuchman, E. H., and Desnick, R. J. (1995) in The Metabolic and Molecular Chem. 275, 15717–15727 Basis of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W. S., and 39. Nishikawa, K., Arai, H., and Inoue, K. (1990) J. Biol. Chem. 265, 5226 –5231 Valle, D., eds) pp. 2601–2624, McGraw-Hill, New York 40. Fadok, V. A., Bratton, D. L., Rose, D. M., Pearson, A., Ezekewitz, R. A., and 17. Slotte, J. P. (1999) Chem. Phys. Lipids 102, 13–27 Henson, P. M. (2000) Nature 405, 85–90 18. Ridgway, N. D. (2000) Biochim. Biophys. Acta 1484, 129 –141 41. Levade, T., Salvayre, R., Maret, A., and Douste-Blazy, L. (1988) J. Inherit. 19. Viana, M. B., Giugliani, R., Leite, V. H., Barth, M. L., Lekhwani, C., Slade, Metab. Dis. 11, 151–157 C. M., and Fensom, A. (1990) J. Med. Genet. 27, 499 –504 42. Maor, I., Mandel, H., and Aviram, M. (1995) Arterioscler. Thromb. Vasc. Biol. 20. Worgall, T. S., McGovern, M. M., Jiang, X. C., Berglund, L., Shea, S., Tabas, I., 15, 1378 –1387 and Deckelbaum, R. J. (2001) Circulation 102, II–30 43. Liscum, L., Ruggiero, R. M., and Faust, J. R. (1989) J. Cell Biol. 108, 21. Tall, A. R., and Wang, N. (2000) J. Clin. Invest. 106, 1205–1207 1625–1636 22. Horinouchi, K., Erlich, S., Perl, D., Ferlinz, K., Bisgaier, C. L., Sandhoff, K., 44. Pentchev, P. G., Comly, M. E., Kruth, H. S., Vanier, M. T., Wenger, D. A., Desnick, R. J., Stewart, C. L., and Schuchman, E. H. (1995) Nature Gen. 10, Patel, S., and Brady, R. O. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 288 –293 8247– 8251 23. Otterbach, B., and Stoffel, W. (1995) Cell 81, 1053–1061 45. Cruz, J. C., Sugii, S., Yu, C., and Chang, T. Y. (2000) J. Biol. Chem. 275, 24. Tang, W., Walsh, A., and Tabas, I. (1999) Biochim. Biophys. Acta 1437, 4013– 4021 301–316 46. Puri, V., Watanabe, R., Dominguez, M., Sun, X., Wheatley, C. L., Marks, D. L., 25. Havel, R. J., Eder, H., and Bragdon, J. (1955) J. Clin. Invest. 34, 1345–1353 and Pagano, R. E. (1999) Nat. Cell Biol. 1, 386 –388 26. Basu, S. K., Goldstein, J. L., Anderson, R. G. W., and Brown, M. S. (1976) Proc. 47. Zhang, M., Dwyer, N. K., Neufeld, E. B., Love, D. C., Cooney, A., Comly, M., Natl. Acad. Sci. U. S. A. 73, 3178 –3182 Patel, S., Watari, H., Strauss, J. F., III, Pentchev, P. G., Hanover, J. A., and 27. Rinninger, F., Brundert, M., Jackle, S., Kaiser, T., and Greten, H. (1995) Blanchette-Mackie, E. J. (2001) J. Biol. Chem. 276, 3417–3425 Biochim. Biophys. Acta 1255, 141–153 48. Zhang, M., Dwyer, N. K., Love, D. C., Cooney, A., Comly, M., Neufeld, E., 28. Tabas, I., Boykow, G. C., and Tall, A. R. (1987) J. Clin. Invest. 79, 418 – 426 Pentchev, P. G., Blanchette-Mackie, E. J., and Hanover, J. A. (2001) Proc. 29. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Natl. Acad. Sci. U. S. A. 98, 4466 – 4471 Chem. 193, 265–275 49. Chen, W., Sun, Y., Welch, C., Gorelik, A., Leventhal, A. R., Tabas, I., and Tall, 30. Tabas, I., Marathe, S., Keesler, G. A., Beatini, N., and Shiratori, Y. (1996) A. R. (2001) J. Biol. Chem. 276, 43564 – 43569 J. Biol. Chem. 271, 22773–22781 50. Reagan, J. W., Jr., Hubbert, M. L., and Shelness, G. S. (2000) J. Biol. Chem. 31. Blanchette-Mackie, E. J., Dwyer, N. K., Amende, L. M., Kruth, H. S., Butler, 275, 38104 –38110 J. D., Sokol, J., Comly, M. E., Vanier, M. T., August, J. T., Brady, R. O., and 51. Lundberg, B. (1985) Atherosclerosis 56, 93–110 Pentchev, P. G. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 8022– 8026 52. Small, D. M., Bond, M. G., Waugh, D., Prack, M., and Sawyer, J. K. (1984) 32. Tabas, I., Lim, S., Xu, X., and Maxfield, F. R. (1990) J. Cell Biol. 111, 929 –940 J. Clin. Invest. 73, 1590 –1605 33. Wang, N., Silver, D. L., Costet, P., and Tall, A. R. (2000) J. Biol. Chem. 275, 53. Rapp, J. H., Connor, W. E., Lin, D. S., Inahara, T., and Porter, J. M. (1983) J. 33053–33058 Lipid Res. 24, 1329 –1335 34. Lange, Y., Ye, J., Rigney, M., and Steck, T. L. (2000) J. Biol. Chem. 275, 54. Shio, H., Haley, N. J., and Fowler, S. (1979) Lab. Invest. 41, 160 –167 17468 –17475 55. Tangirala, R. K., Mahlberg, F. H., Glick, J. M., Jerome, W. G., and Rothblat, 35. Mukherjee, S., Zha, X., Tabas, I., and Maxfield, F. R. (1998) Biophys. J. 75, G. H. (1993) J. Biol. Chem. 268, 9653–9660 1915–1925 56. Jiang, X. C., Paultre, F., Pearson, T. A., Reed, R. G., Francis, C. K., Lin, M., 36. Tabas, I., Rosoff, W. J., and Boykow, G. C. (1988) J. Biol. Chem. 263, 1266 –1272 Berglund, L., and Tall, A. R. (2000) Arterioscler. Thromb. Vasc. Biol. 20, 2614 –2618 37. Xu, X., and Tabas, I. (1991) J. Biol. Chem. 266, 17040 –17048 38. Botelho, R. J., Hackam, D. J., Schreiber, A. D., and Grinstein, S. (2000) J. Biol. 57. Tabas, I. (1999) Chem. Phys. Lipids 102, 131–139 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Acid Sphingomyelinase-deficient Macrophages Have Defective Cholesterol Trafficking and Efflux

Leventhal, Andrew R.; Chen, Wengen; Tall, Alan R.; Tabas, Ira
Journal of Biological Chemistry – Nov 1, 2001
Download PDF
8 pages

Loading next page...
 
/lp/unpaywall/acid-sphingomyelinase-deficient-macrophages-have-defective-cholesterol-c9dgHPbYCW

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.m106455200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 48, Issue of November 30, pp. 44976 –44983, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Acid Sphingomyelinase-deficient Macrophages Have Defective Cholesterol Trafficking and Efflux* Received for publication, July 10, 2001, and in revised form, September 24, 2001 Published, JBC Papers in Press, September 28, 2001, DOI 10.1074/jbc.M106455200 Andrew R. Leventhal‡, Wengen Chen‡, Alan R. Tall‡, and Ira Tabas‡§¶ From the ‡Departments of Medicine and §Anatomy & Cell Biology, Columbia University, New York, New York 10032 Cholesterol efflux from macrophage foam cells, a key Cholesteryl ester (CE) -loaded macrophages, or foam cells, are prominent features of atherosclerotic lesions and play im- step in reverse cholesterol transport, requires traffick- ing of cholesterol from intracellular sites to the plasma portant roles in lesion progression (1, 2). During atherogenesis, membrane. Sphingomyelin is a cholesterol-binding mol- intimal macrophages internalize atherogenic lipoproteins, in- ecule that transiently exists with cholesterol in endo- cluding modified forms of LDL, that have been retained in the somes and lysosomes but is rapidly hydrolyzed by lyso- arterial subendothelium (1, 3, 4). This event directly leads to somal sphingomyelinase (L-SMase), a product of the esterification of cellular cholesterol by acyl-coenzyme A:choles- acid sphingomyelinase (ASM) gene. We therefore hy- terol O-acyltransferase (ACAT), resulting in “foam cell” forma- pothesized that sphingomyelin hydrolysis by L-SMase tion (3, 5). Foam cell formation can be prevented or reversed by enables cholesterol efflux by preventing cholesterol se- the process known as cellular cholesterol efflux (6). Cholesterol questration by sphingomyelin. Macrophages from wild- efflux is the initial step of reverse cholesterol transport, a type and ASM knockout mice were incubated with process whereby excess cholesterol in peripheral cells is deliv- [ H]cholesteryl ester-labeled acetyl-LDL and then ex- ered to the liver for excretion (6). Thus, the elucidation of posed to apolipoprotein A-I or high density lipoprotein. cellular molecules and pathways that facilitate and regulate In both cases, [ H]cholesterol efflux was decreased sub- cholesterol efflux is a major goal of research in the area of stantially in the ASM knockout macrophages. Similar atherosclerosis. results were shown for ASM knockout macrophages la- Atherogenic lipoproteins internalized by macrophages de- beled long-term with [ H]cholesterol added directly to liver their stores of cholesterol, which are mostly in the form of medium, but not for those labeled for a short period, CE, to late endosomes and/or lysosomes (3). Here, lysosomal suggesting defective efflux from intracellular stores but acid lipase hydrolyzes the CE to free cholesterol, which is then not from the plasma membrane. Cholesterol trafficking transported by poorly defined mechanisms to various sites in to acyl-coenzyme A:cholesterol acyltransferase (ACAT) the cell (3). A major site of transport is the plasma membrane, was also defective in ASM knockout macrophages. Using and from there the cholesterol can be effluxed to extracellular filipin to probe cholesterol in macrophages incubated acceptors, such as apoA-I and HDL, or transported to ACAT in with acetyl-LDL, we found there was modest staining in the endoplasmic reticulum for re-esterification (3, 6). Efflux to the plasma membrane of wild-type macrophages but apoA-I involves the initial formation of phospholipid-apoA-I bright, perinuclear fluorescence in ASM knockout macrophages. Last, when wild-type macrophages were particles by ABCA1-mediated phospholipid efflux, followed by incubated with excess sphingomyelin to “saturate” L- cholesterol efflux to these phospholipid-rich particles (7, 8). SMase, [ H]cholesterol efflux was decreased. Thus, Cholesterol efflux to HDL can be mediated by scavenger recep- sphingomyelin accumulation due to L-SMase deficiency tor type B1 (SR-B1) in those cell types that have relatively high leads to defective cholesterol trafficking and efflux, levels of this receptor, such as human monocyte-derived macro- which we propose is due to sequestration of cholesterol phages, but another mechanism must be involved in cells that by sphingomyelin and possibly other mechanisms. This have very low expression of SR-B1, such as mouse peritoneal model may explain the low plasma high density lipopro- macrophages (6, 9, 10). tein found in ASM-deficient humans and may implicate Given that all pathways of cholesterol efflux require choles- L-SMase deficiency and/or sphingomyelin enrichment of terol transport to the plasma membrane, the identification of lipoproteins as novel atherosclerosis risk factors. molecules mediating or regulating this transport process is an important goal. Thus far, only the molecules npc1, npc2 (HE1), and possibly lysobisphosphatidic acid have been shown to play a role in cholesterol transport to the plasma membrane, and * This work was supported by Specialized Center of Research in the molecular mechanisms are poorly understood (11–15). We Atherosclerosis Grant HL-56984 (to A. R. T. and I. T.) from the NHLBI, reasoned that another molecule, lysosomal sphingomyelinase National Institutes of Health and a research grant from Berlex Bio- (L-SMase), may also be involved in cholesterol transport from sciences (to I. T.). The Columbia University Confocal Microscope Facil- ity used for this study was established by National Institutes of Health lysosomes to the plasma membrane. L-SMase, a product of the Shared Instrument Grants S10 RR10506 and S10 RR13701 and the Lieber Foundation. The operation of the Facility is supported in part by National Institutes of Health Grant P30 CA13696 as part of the Herbert The abbreviations used are: CE, cholesteryl ester; ACAT, acyl-CoA: Irving Comprehensive Cancer Center at Columbia University. The cholesterol acyltransferase; apoA-I, apolipoprotein A-I; ASM, acid costs of publication of this article were defrayed in part by the payment sphingomyelinase; BSA, bovine serum albumin; DMEM, Dulbecco’s of page charges. This article must therefore be hereby marked “adver- modified Eagle’s medium; FBS, fetal bovine serum; HDL, high-density tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate lipoprotein; LDL, low-density lipoprotein; L-SMase, lysosomal sphingo- this fact. myelinase; NPC, Niemann-Pick C; PBS, phosphate-buffered saline; PS, ¶ To whom correspondence should be addressed: Dept. of Medicine, phosphatidylserine; SMase, sphingomyelinase; SR-BI, scavenger recep- Columbia University, 630 West 168th St., New York, NY 10032. Tel.: tor BI. 212-305-9430; Fax: 212-305-4834; E-mail: [email protected]. Y. Sun and A. R. Tall, unpublished data. 44976 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44977 tific Sonic Dismembranator-60 at level 5 for 5 min. [ H]Cholesterol (1 acid sphingomyelinase (ASM) gene, hydrolyzes sphingomyelin Ci/ml) was then added, and the solution was incubated overnight in a in late endosomes and lysosomes (16). Because SM avidly binds shaking water bath at 37 °C. The solution was then filtered through a cholesterol (17, 18), we hypothesized that sphingomyelin hy- 0.45-m filter. drolysis by L-SMase enables cholesterol transport by prevent- Cholesterol Efflux Assay—After labeling the cells, the medium was ing cholesterol sequestration by sphingomyelin. Of interest, changed to DMEM, 0.2% BSA containing 20 g/ml apoA-I, 15 g/ml humans with ASM deficiency (types A and B Niemann-Pick HDL ,or10 g/ml HDL . At the indicated time points, 100 l of media 2 3 was removed and spun for 5 min at 14,000 rpm in a microcentrifuge to disease) have low plasma HDL levels (19, 20), which could remove cellular debris, and the radioactivity in this fraction of media result from defective cholesterol efflux (cf. Ref. 21). was quantified by liquid scintillation counting. After the last time point, In this context, we show herein that macrophages from ASM the remainder of the media was removed, and the cells were dissolved knockout mice, which lack L-SMase (22, 23), have a defect in N NaOH at room temperature for 5 h. A 100-l aliquot of in1mlof0.1 cholesterol efflux to both apoA-I and HDL, a decrease in the the cell lysate was counted, and the percent efflux was calculated as esterification of cellular cholesterol, and an accumulation of [(media cpm)  (cell  media cpm)]  100. Note that there was no statistical difference in cellular counts/min between wild-type and cholesterol in perinuclear vesicles. Moreover, a defect in cho- ASM-deficient macrophages using any of the three [ H]cholesterol la- lesterol efflux was also observed in wild-type macrophages that beling methods. To obtain the value for acceptor-stimulated efflux, the internalized a large amount of sphingomyelin. These data sup- percent efflux in the absence of acceptor (i.e. DMEM, 0.2% BSA without port the hypothesis that intracellular accumulation of sphin- apoA-I or HDL) was subtracted from the percent efflux in the presence gomyelin due to L-SMase deficiency or to internalization of of acceptor; the basal efflux values were 10% of those in the presence excess sphingomyelin leads to cholesterol sequestration and of acceptor. Whole Cell Cholesterol Esterification Assay—In the first method, defective cholesterol trafficking and efflux. macrophages were incubated in DMEM, 0.2% BSA containing 0.1 m EXPERIMENTAL PROCEDURES C]oleate complexed with albumin and 3 g/ml acetyl-LDL. In the Materials—The Falcon tissue culture plasticware used in these stud- second method, cells were labeled with [ H]cholesterol by long-term ies was purchased from Fisher Scientific Co. Tissue culture media and incubation with [ H]cholesterol-labeled medium as described above, g/ml unlabeled acetyl-LDL or 5 M other tissue culture reagents were obtained from Life Technologies, Inc. followed by incubation with 3 25-hydroxycholesterol for up to 12 h. At the indicated time points, the Fetal bovine serum (FBS) was obtained from Hyclone Laboratories cells were washed two times with cold PBS, and the cell monolayers (Logan, UT). Alexa 543-labeled dextran-10,000 was purchased from were extracted twice with 0.5 ml of hexane/isopropyl alcohol (3:2, v/v) Molecular Probes, Inc. (Eugene, OR). All radiochemicals were pur- for 30 min at room temperature. Whole cell cholesterol esterification chased from Perkin-Elmer Life Sciences, Inc. (Boston, MA). All other activity was assayed by determining the cellular content of cholesteryl chemicals and reagents were from Sigma, and all organic solvents were 14 3 from Fisher Scientific Co. C]oleate or [ H]cholesteryl ester by thin-layer chromatography (28). N NaOH, and aliquots Macrophages—Littermate wild-type and homozygous ASM knockout The cell monolayers were dissolved in 1 ml of 0.1 were assayed for protein by the method of Lowry et al. (29). mice were obtained by breeding ASM heterozygous knockout mice, In Vitro ACAT Assay—Macrophages in 100-mm dishes were incu- which were provided by Dr. Edward Schuchman, Mt. Sinai School of bated for 24 h in DMEM, 10% lipoprotein-deficient serum and then Medicine (22, 23). Macrophages were harvested from the peritoneum of scraped in 2.5 ml of ice-cold 20 mM potassium phosphate buffer, 2 mM these mice 3 days after the intraperitoneal injection of 40 g of con- dithiothreitol, pH 7.4. This suspension was sonicated for five 3-s bursts canavalin A in 0.5 ml of PBS and then cultured as described previously on level 10 of a Fisher Scientific Sonic Dismembranator-60. Aliquots of (24). the cell lysates were removed for protein determination, and then Lipoproteins and Liposomes—LDL (d, 1.020 –1.063 g/ml), HDL (d, 130-l of the cell lysates were added to capped glass test tubes that 1.063–1.125 g/ml), HDL (d, 1.125–1.21 g/ml) from fresh human plasma contained 20 l of 0.5 mg/ml cholesterol-rich PS liposomes in PBS, were isolated by preparative ultracentrifugation as described (25). which were prepared as described previously (30), or 20 l of PBS alone. Acetyl-LDL was prepared by reaction with acetic anhydride (26) and After a 15-min incubation at 37 °C, 30 l of 40 mg/ml fatty acid-free labeled with [ H]CE as described (27). Briefly, label was transferred BSA in 100 mM potassium phosphate buffer, 2 mM dithiothreitol, pH from [ H]cholesteryl oleate-containing liposomes to HDL by CE trans- 7.4, and 20 lof25 M [ fer protein, followed by CE transfer protein-mediated transfer of the C]oleoyl-CoA (40 Ci/mol) were added. After [ H]CE from HDL to acetyl-LDL, which were then separated from each a 15-min incubation at 37 °C, the reaction was stopped by adding 2 ml g of unlabeled CE internal other by density ultracentrifugation. The final specific activity of the of 2:1 chloroform:methanol containing 250 standard. The mixture was vortexed and allowed to sit at room tem- labeled acetyl-LDL was 17 cpm/ng of protein. [ H]Cholesterol-contain- perature for 30 min. Four hundred l of 0.88% KCl were added to each ing liposomes were prepared by first mixing 100 Ci of [ H]cholesterol tube, and the tubes were then centrifuged at 1000 rpm for 10 min. The and 2.8 mg of phosphatidylserine (PS) in chloroform, in the absence or lower organic phase was collected and separated by thin layer chroma- presence of 1.25 mg of sphingomyelin. The chloroform was evaporated tography, and the CE spot was scraped and counted. under nitrogen, and the dried lipids were sonicated in 3 ml of PBS at Filipin Staining of Free Cholesterol and Fluorescence Microscopy— 4 °C under argon using 15 4-min bursts at setting number 2 on a For filipin staining, the method of Blanchette-Mackie et al. (31) was Branson 450 sonicator equipped with a tapered microtip. [ C]Sphin- employed. Briefly, a 0.05 mg/ml filipin solution was made from a 5 gomyelin-containing liposomes were made in the same manner except 14 3 mg/ml stock in Me 1.2 Ci of [ C]sphingomyelin was used instead of [ H]cholesterol. For SO by dilution with 10% FBS in PBS. Macrophages certain experiments in which we needed to fluorescently label PS- were cultured on polylysine-coated coverslip bottom dishes (32) and sphingomyelin liposomes, the PBS contained 0.5 mg/ml Alexa 543- preincubated for 24 h in DMEM, 10% LPDS. After the incubation labeled dextran-10,000; after sonication, the solution was dialyzed us- described in the legend to Fig. 5, the cells were washed with PBS and ing dialysis tubing with a 100,000-MW pore size. fixed with 3% paraformaldehyde in PBS for1hat room temperature. Labeling of Cells with [ H]Cholesterol—Cell monolayers were labeled The macrophages were subsequently rinsed three times with PBS and by one of three methods. For labeling by acetyl-LDL or PS liposomes, incubated with 1.5 mg of glycine/ml of PBS for 10 min at room temper- cells were incubated for4hin DMEM, 0.2% BSA containing 10 g/ml ature. Next, the cells were incubated with the 0.05 mg/ml filipin solu- 3 3 [ H]CE-acetyl-LDL or [ H]cholesterol-labeled phosphatidylserine-lipo- tion for2hat room temperature. The cells were then washed three somes such that the final PS concentration was 93 g/ml. For long-term times with PBS and viewed with a Zeiss Axiovert S100 epifluorescence labeling of cellular cholesterol, macrophages were grown for 24 h in microscope using a UV filter set (340 –380-nm excitation, 70-nm di- DMEM, 10% FBS containing 0.5 Ci/ml [ H]cholesterol, then equili- chroic, 430-nm long pass filter). For the double-label filipin/Alexa 543- brated for 12 h in DMEM, 0.2% BSA. The [ H]cholesterol-containing dextran study displayed in Fig. 6, dual-photon microscopy was con- medium was made by adding the labeled cholesterol, which was in ethanol, dropwise (0.5 l/ml) into DMEM, 10% FBS while stirring at 3 3 37 °C. After a 30-min incubation at 37 °C, the medium was passed In the experiments with apoA-I, the percent of [ H]cholesterol in the through a 0.45-m filter. For short-term labeling experiments, macro- medium represents net efflux, because the medium contains no unla- phages were incubated for 15 min at room temperature in DMEM beled cholesterol at the beginning of the experiment. In the experiments containing methyl--cyclodextrin-[ H]cholesterol complex. To prepare with HDL, however, this method used in this study does not distinguish this solution, methyl--cyclodextrin and unlabeled cholesterol (8:1 mo- between net cholesterol efflux and exchange of cholesterol between the lar ratio) were dissolved in DMEM and sonicated using a Fisher Scien- cells and HDL. 44978 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages ducted using an LSM 510 nonlinear optics (NLO) Zeiss dual photon confocal microscope equipped with a 100X/1.3 NA Plan-Neofluor objec- tive lens. For Alexa-546, the pinhole was adjusted to produce an optical section of 1.0 m. A helium-neon laser (543 nm) was used for excitation and a 560-nm long-pass was used as an emission filter. For filipin, the pinhole was completely open, and a Coherent titanium-sapphire laser tuned to 800 nm was used for excitation and a 390 – 465-nm band pass was used as an emission filter. Pilot studies indicated that this config- uration allowed good resolution of both Alexa-546 and filipin fluores- cence without cross-over fluorescence. Statistics—Results are given as mean  S.E. (n  3); absent error bars in the figures signify S.E. values smaller than the graphic symbols. For the data in Fig. 7A, the unpaired, two-tailed t test was used to determine statistically significance. RESULTS Efflux of Acetyl-LDL-derived [ H]Cholesterol and Cellular Cholesterol Labeled by Long-term Incubation with [ H]Choles- terol-containing Medium Is Defective in ASM Knockout Macro- phages—To assess the role of macrophage L-SMase in the efflux of acetyl-LDL-derived cholesterol, peritoneal macro- phages from wild-type and ASM knockout mice were incubated with [ H]CE-labeled acetyl-LDL for 4 h and then chased in serum-free medium containing lipid-free apoA-I, which medi- ates cholesterol efflux through the ABC1 pathway, and HDL , which involves mostly other pathways (33). [ H]Cholesterol derived from [ H]CE-acetyl-LDL traffics through late endo- somes and lysosomes, which are known sites of L-SMase activ- ity and SM accumulation in ASM-deficient cells (16). As shown in Fig. 1, cholesterol efflux to both apoA-I (panel A) and HDL (panel B) was reduced by 60 –70% in the ASM knockout macrophages. For example, efflux to apoA-I at 12 h was 4.3% in ASM knockout cells versus 13.4% in wild-type cells (Fig. 1A). Cellular cholesterol can also be labeled by long-term incuba- tion with [ H]cholesterol added directly to the medium. In this case, the plasma membrane is labeled first, followed by equil- ibration with intracellular stores that probably include recy- cling endosomes and the trans-Golgi network (34, 35). To as- sess the efflux or exchange of cellular cholesterol pools labeled in this manner, wild-type and ASM knockout macrophages were labeled for 24 h in the presence of 0.5 Ci/ml [ H]choles- terol. After a 12-h equilibration period, the cells were incubated with HDL, and [ H]cholesterol in the medium was measured. FIG.1. Efflux of acetyl-LDL-derived [ H]cholesterol from wild- As shown in Fig. 2A, there was an 50% decrease in the 3 type and ASM knockout macrophages. Monolayers of peritoneal percent [ H]cholesterol in the medium of ASM knockout macro- macrophages from wild-type (closed circles) or ASM knockout (open phages under these conditions. circles) mice were incubated with 10 g/ml [ H]CE-labeled acetyl-LDL One interpretation of the data in Fig. 2A is that ASM knock- in DMEM, 0.2% BSA for 4 h. The cells were then rinsed and incubated with fresh medium containing either 20 g/ml human apoA-I (A)or10 out macrophages have a defect in the efflux or exchange of g/ml human HDL (B) for the indicated times. H-Labeled cpm in the plasma membrane cholesterol. To address this issue, wild-type media and cells were measured to calculate percent [ H]cholesterol in and ASM knockout macrophages were incubated for 15 min at the medium. room temperature with [ H]cholesterol-charged methyl--cy- clodextrin, and then chased for 1 or3hin medium without macrophages was shown using HDL (Fig. 2B and data not label but containing HDL. According to Lange et al. (34), the shown), which mediates efflux by an ABCA1-independent 15-min labeling procedure labels mostly plasma membrane mechanism (7). SR-B1 almost certainly is not involved, because cholesterol, which then eventually equilibrates with intracel- its expression is very low in mouse peritoneal macrophages, lular pools. As shown in Fig. 2B, medium [ H]cholesterol at the and an SR-B1 neutralizing antibody does not block cholesterol early time point was the same in the two cell types, while after efflux from these cells to HDL. Given these data, we focused a 3-h chase it was less in the ASM knockout macrophages. our efforts on the hypothesis that decreased efflux in ASM These data suggest that transfer of cholesterol directly from knockout macrophages was caused by defective intracellular the plasma membrane to HDL is not defective in ASM knock- cholesterol trafficking. out macrophages but that efflux from the intracellular site(s) Trafficking of Cholesterol to ACAT Is Defective in ASM that accumulates nonlipoprotein cholesterol is affected. Knockout Macrophages—To determine if other cellular choles- In view of these data, we considered the possibility that a terol trafficking pathways were altered in ASM knockout decrease in expression of ABCA1 or SR-B1 could explain the macrophages, we assessed the ability of the cells to esterify defect in efflux. However, quantitative polymerase chain reac- cholesterol, which requires cholesterol transport to ACAT in tion showed no difference in ABCA1 mRNA expression between the endoplasmic reticulum (3). First, wild-type and ASM wild-type and ASM knockout macrophages (data not shown). Although there could be differences in ABCA1 localization or activity, a defect in cholesterol efflux from ASM knockout W. Chen and A. R. Tall, unpublished data. Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44979 FIG.3. Cholesterol esterification in wild-type and ASM knock- out macrophages incubated with acetyl-LDL. A, macrophages from wild-type (closed circles) or ASM knockout (open circles) mice were FIG.2. Efflux of cellular cholesterol labeled by long-term in- incubated for the indicated times in DMEM, 0.2% BSA containing 3 3 14 cubation of macrophages with [ H]cholesterol-containing me- g/ml acetyl-LDL and 0.1 mM [ C]oleate. Cellular lipids were then dium. A, macrophages from wild-type (closed circles) or ASM knockout extracted and assayed for cholesteryl [ C]oleate. B, the macrophages (open circles) mice were incubated for 24 h in DMEM, 10% FBS con- were incubated for 24 h in DMEM, 10% FBS containing 0.5 Ci/ml taining 0.5 Ci/ml [ H]cholesterol. The cells were then incubated for an [ H]cholesterol. The cells were then incubated for an additional 12 h in additional 12 h in DMEM, 0.2% BSA without label, and then finally DMEM, 0.2% BSA without label, and then finally incubated in the same incubated in the same medium containing 10 g/ml HDL for the medium containing 3 g/ml acetyl-LDL for the indicated times. Cellular 3 3 indicated times. Percent [ H]cholesterol in the medium was then deter- lipids were extracted and assayed for [ H]cholesteryl ester. mined. B, macrophages from wild-type (solid bars) or ASM knockout (hatched bars) mice were incubated for 15 min at room temperature in protocol, there was even a greater defect in cholesterol esteri- DMEM, 0.2% BSA containing 1 Ci/ml [ H]cholesterol-charged methyl- fication in the ASM knockout cells. -cyclodextrin. The cells were then rinsed and incubated with fresh A similar experiment to that in Fig. 3B was conducted, but medium containing 15 g/ml HDL for either 1 or 3 h, and then percent 25-hydroxycholesterol was used as the stimulator of ACAT [ H]cholesterol in the medium was determined. instead of acetyl-LDL (36). In this scenario, the formation of knockout macrophages were incubated with acetyl-LDL and [ H]cholesteryl ester was almost totally abolished in the ASM 14 14 [ C]oleate, and the formation of cholesteryl [ C]oleate was knockout macrophages (Fig. 4A). We next determined if these assayed. Under these conditions, acetyl-LDL-derived choles- data could be explained by decreased active ACAT enzyme in terol mixes with cellular cholesterol in the plasma membrane, the ASM knockout macrophages. Wild-type and ASM knockout and when a threshold level of cholesterol is reached, this cho- macrophages were incubated for 24 h in the absence of lipopro- lesterol is transported to ACAT and esterified (36, 37). As teins, and then lysates from these cells were assayed for ACAT shown in Fig. 3A, ASM knockout macrophages demonstrated a activity in vitro in the absence or presence of exogenous cho- 30 –50% decrease in CE formation. Next, the macrophage cho- lesterol (Fig. 4B). In the absence of exogenous cholesterol, lesterol pools were labeled by long-term incubation with ACAT activity was low and similar in both cell types; in the [ H]cholesterol-labeled medium (above), and the formation of presence of exogenous cholesterol, ACAT activity was higher [ H]cholesteryl ester in response to subsequent incubation with than that in the absence of cholesterol and, surprisingly, some- acetyl-LDL was assayed (Fig. 3B). Using this experimental what greater in the lysates of ASM knockout mice. Although we 44980 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages FIG.5. Free cholesterol distribution in wild-type and ASM knockout macrophages. Macrophages from wild-type (A)orASM knockout (B) mice were incubated with DMEM, 0.2% BSA containing 100 g of acetyl-LDL/ml for 4 h and then chased in medium without lipoproteins for an additional 4 h. The cells were then fixed, stained with filipin, and visualized by fluorescence microscopy. Bar,5 m. demonstrate intracellular sequestration of cholesterol in ASM knockout macrophages. We next determined if the free cholesterol that accumulates in ASM-deficient macrophages is localized solely or mostly in late endosomes and lysosomes. Macrophages from ASM knock- out mice were incubated with acetyl-LDL and Alexa 546-la- beled dextran for 4 h, and then chased in medium alone for an additional 4 h. The dextran is taken up by fluid-phase pinocy- tosis and, after a 4-h chase, labels late endosomes and lyso- somes; for example, in macrophages and other cells, the stain- ing pattern with dextran is essentially identical to that with the lysosomal marker LAMP1 (38). After fixing and staining with filipin to detect the cholesterol, the cells were viewed by dual-photon confocal fluorescence microscopy. An example of one ASM-deficient macrophage is shown in Fig. 6, A–C, and a cluster of macrophages is shown in Fig. 6, D–F. Panels A and D show the filipin staining pattern (pseudocolored green), panels B and E the Alexa-dextran pattern, and panels C and F the merged images. As in Fig. 5B, free cholesterol accumulated in a perinuclear distribution (A and D). Alexa-dextran also accu- mulated in a perinuclear pattern (B and E), which is consistent with a late endosome/lysosome pattern (38). Although there FIG.4. 25-Hydroxycholesterol-induced cholesterol esterifica- was some co-localization of cholesterol and Alexa-dextran (yel- tion and in vitro ACAT activity in wild-type and ASM knockout macrophages. A, macrophages from wild-type (closed circles)orASM low vesicles in panels C and F), there were also a number of knockout (open circles) mice were incubated for 24 h in DMEM, 10% vesicles that were distinct, as indicated by the pure red and FBS containing 0.5 Ci/ml [ H]cholesterol. The cells were then incu- pure green vesicles in panels C and F. Thus, the cholesterol that bated for an additional 12 h in DMEM, 0.2% BSA without label, and accumulates in ASM-deficient macrophages accumulates both then finally incubated in the same medium containing 5 g/ml 25- hydroxychoelsterol for the indicated times. Cellular lipids were ex- in dextran-containing late endosomes/lysosomes and in other tracted and assayed for [ H]cholesteryl ester. B, lysates of macrophages perinuclear sites (see “Discussion”). from wild-type (solid bars) or ASM knockout (hatched bars) mice were Delivery of Excess SM to Cells Can Decrease Cholesterol assayed for ACAT activity in the absence or presence of 125 M Efflux in Wild-type Macrophages—The data in this report sup- cholesterol. port the hypothesis that the accumulation of SM in late endo- somes, lysosomes, and probably other sites as a result of do not know the cause of this increase, the data strongly sug- L-SMase deficiency leads to a defect in cholesterol efflux. Ac- gest that the defect in cholesterol esterification in ASM knock- cording to this hypothesis, it might be possible to “saturate” out macrophages (Fig. 3) reflects a defect in cholesterol traf- L-SMase activity in wild-type cells by delivering excess SM, ficking to ACAT rather than decreased active ACAT enzyme. thus resulting in a decrease in cholesterol efflux. To test this ASM Knockout Macrophages Accumulate Free Cholesterol in prediction, wild-type macrophages were incubated for 4 h with Perinuclear Sites—To directly assess the intracellular distribu- PS liposomes that had tracer amounts of [ H]cholesterol and tion of free cholesterol, wild-type and ASM knockout macro- either no sphingomyelin or sphingomyelin (sphingomyelin:PS, phages were incubated with acetyl-LDL (4-h incubation fol- 1:2 molar ratio). Note that PS-containing liposomes, like acetyl- lowed by 4-h chase) and then stained with filipin, a fluorescent LDL, are endocytosed by macrophages through the type A free cholesterol-binding molecule. The filipin-staining pattern scavenger receptor pathway and possibly through a newly iden- in wild-type macrophages was modest in intensity, although tified PS receptor (39, 40). The cells were then rinsed and greater than in cells not exposed to acetyl-LDL, and the plasma incubated for 12 h with fresh medium containing HDL, after membrane was the principal site of staining (Fig. 5A). In strik- which the percent [ H]cholesterol in the medium was deter- ing contrast, ASM knockout macrophages showed bright, pe- mined. Using parallel experiments in which the sphingomye- rinuclear, punctate fluorescence (Fig. 5B). These data directly lin-PS liposomes contained tracer [ C]SM, we calculated that Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44981 FIG.6. Free cholesterol and late en- dosome/late lysosome distribution in wild-type and ASM knockout macro- phages. Macrophages from ASM knock- out were incubated with medium contain- ing 100 g of acetyl-LDL/ml and 0.5 mg/ml Alexa 546-labeled dextran-10,000 for 4 h and then chased in medium with- out lipoproteins or dextran for an addi- tional 4 h. The cells were then fixed, stained with filipin, and visualized by dual-photon confocal fluorescence micros- copy. An example of one ASM-deficient macrophage is shown in A-C, and a clus- ter of macrophages is shown in D-F. Pan- els A and D show the filipin staining pat- tern (pseudocolored green), panels B and E the Alexa-dextran pattern, and panels C and F the merged images. Bar,1 m. FIG.7. Effect of liposomal sphingomyelin content on the efflux of liposome-derived cholesterol from wild-type and ASM knockout macrophages. A, macrophages from wild-type (solid bars) or ASM knockout (hatched bars) mice were incubated for 4 h with DMEM, 0.2% BSA containing PS liposomes that had tracer amounts of [ H]cholesterol and either no sphingomyelin (SM) or sphingomyelin (sphingomyelin:PS, 1:2 molar ratio; SM). In all incubations, the final PS concentration was 93 g/ml, and the incubation volume was 0.5 ml. The cells were then rinsed and incubated for 12 h with fresh medium containing 15 g/ml HDL , after which the percent [ H]cholesterol in the medium was determined. The difference in efflux between SM and SM liposomes in the wild-type macrophages were statistically significant (p  0.001). B, to verify that the liposomes were internalized, macrophages were incubated with sphingomyelin/PS liposomes labeled with Alexa 543-dextran-10,000 for4hand then chased for an additional3hin medium without liposomes. C, for comparison of the image in B with a cell-surface pattern, macrophages were incubated with the Alexa-labeled liposomes for1hat4 °C. Bar,5 m. this protocol led to the delivery of 6.6  0.8 g of sphingomy- prior incubation in serum-containing medium) and endogenous elin/mg of cell protein. As expected for receptor-mediated en- SM (41). These data are consistent with a model in which the docytic targeting to late endosomes and lysosomes, the vesicles activity of L-SMase in normal macrophages can be partially were internalized and delivered to perinuclear organelles (Fig. “saturated” by the endocytosis of SM, leading to intracellular 7B); for comparison, Fig. 7C shows a typical cell-surface distri- SM accumulation and defective cholesterol trafficking. bution of macrophages incubated with vesicles at 4 °C. DISCUSSION As shown in the first pair of bars in Fig. 7A, there was a statistically significant decrease in efflux of [ H]cholesterol The results of this study have implications ranging from the delivered in the sphingomyelin-containing liposomes compared basic cellular biology of lipid trafficking to the important phys- with [ H]cholesterol delivered in liposomes without SM. As iologic area of macrophage cholesterol efflux in atherosclerosis. expected, [ H]cholesterol efflux from ASM knockout macro- The mechanisms and regulation of cholesterol trafficking from phages was decreased to an even greater degree (second pair of lysosomes to the plasma membrane are poorly understood. bars in Fig. 7A). Interestingly, the presence of excess sphingo- Data from mutant cells indicate that the proteins npc1 and myelin did not further decrease efflux in these cells, presum- npc2 (HE1) have functions along this pathway (11–15), and ably because intracellular SM levels were already very high antibody experiments may suggest a role for the lipid lyso- due to L-SMase deficiency in the face of exogenous SM (from bisphosphatidic acid (15). The mechanism of action of these 44982 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages molecules, however, is not known. The idea that the sphingo- (50) made the interesting observation that lysosomal choles- myelin content of lysosomes could be an important regulatory terol accumulation in Chinese hamster ovary cells induced by factor is based upon the ability of this lipid to bind cholesterol either the NPC mutation or progesterone causes a decrease in (17, 18). Indeed, Aviram and colleagues (42) suggested that the the enzymatic activity of L-SMase. Therefore, in view of our data, it is possible that the cholesterol trafficking defects in ability of oxysterols in oxidized LDL to inhibit L-SMase in murine and human macrophages may account for the accumu- NPC cells and progesterone-treated cells may be amplified by secondary inhibition of L-SMase. lation of lysosomal FC under these conditions, although molec- The findings in this report may have important implications ular genetic proof was not provided to support their hypothesis. for foam cell biology. Macrophages in advanced lesions are In contrast, cholesterol esterification and trafficking to the known to accumulate large amount of free cholesterol, much of plasma membrane were reported as being normal in several which appears to be in lysosomes (51–55). On the one hand, it lines of fibroblasts from humans with types A and B Niemann- is possible that exposure of these macrophages to oxidized LDL Pick disease (43, 44). It is possible that the human fibroblasts or oxysterols, by inhibiting L-SMase (42), or SM-rich lipopro- data could have been influenced by residual L-SMase activity teins, by “saturating” L-SMase (see Fig. 6), may contribute to in these cells (16) or by inherent differences in cholesterol this event. Regarding this latter possibility, Jiang et al. (56) trafficking between fibroblasts and macrophages. recently found that a high plasma SM level is an independent Our working hypothesis states that defective cholesterol risk factor for coronary artery disease in humans. On the other trafficking in ASM knockout macrophages is due to sequestra- hand, the accumulation of lysosomal cholesterol, even if caused tion of cholesterol by sphingomyelin. This model can readily by another process, might be expected to secondarily inhibit explain the acetyl-LDL-cholesterol trafficking data, because L-SMase (50), which could further exacerbate the accumulation acetyl-LDL-derived cholesterol traffics through lysosomes, of lysosomal cholesterol and inhibit cholesterol efflux. which is a known site of SM accumulation in ASM-deficient The findings in this study raise additional questions related cells (16). With regard to those experiments in which the to lipoprotein abnormalities and atherosclerotic risk in humans macrophages were labeled by long-term incubation with with ASM deficiency (types A and B Niemann-Pick disease). [ H]cholesterol-containing medium, it is possible that this These subjects have markedly low plasma HDL levels (19, 20). method also labels lysosomal pools of cholesterol. However, Given that low plasma HDL can result from defective cellular when a similar method was used in Chinese hamster ovary cholesterol efflux (21), our current data may provide a mecha- cells to incorporate the fluorescent sterol dehydroergosterol or nism that contributes to this lipoprotein abnormality. Regard- cholesterol itself, followed by filipin labeling, the major sites of ing atherosclerotic risk, one must focus on type B Niemann- accumulation were the endosomal recycling compartment and Pick patients, who survive to adulthood due to low levels of the trans-Golgi network (35). If this were the case in macro- residual ASM activity, and type A or type B obligate heterozy- phages, it would indicate defective trafficking and efflux of gotes, who are reported to be “normal” (16). In considering the non-lysosomal cholesterol and therefore might imply that ASM potential atherogenic effects of L-SMase deficiency in these deficiency leads to SM accumulation in the endosomal recycling subjects, it is interesting to consider that the ASM gene also compartment, trans-Golgi network, or other nonlysosomal sites gives rise to secretory SMase (57). Because secretory SMase (cf. Refs. 15, 34, and 45). This idea might also provide an promotes the subendothelial aggregation and retention of li- explanation for our finding that not all of the acetyl-LDL- poproteins, leading to enhanced foam cell formation, S-SMase derived free cholesterol that accumulates in ASM-deficient deficiency, unlike L-SMase deficiency, may decrease choles- macrophages is in dextran-containing late endosomes and ly- terol accumulation in lesional macrophages (57). Therefore, sosomes (Fig. 6). Detailed sphingomyelin localization studies in while the deficiency of S-SMase in these subjects might be ASM knockout macrophages will be required to sort out these protective, defective L-SMase activity, by inhibiting cholesterol possibilities. efflux from lesional macrophages and possibly by leading to low One must also consider the possibility that direct sequestra- HDL levels, may promote atherosclerotic vascular disease. tion of cholesterol by SM is not the only mechanism behind defective cholesterol trafficking in ASM knockout macro- Acknowledgments—We thank Dr. Edward Schuchman for providing the ASM heterozygote knockout mice that were used to generate the phages. In this context, there is evidence that initial accumu- wild-type and ASM homozygous knockout mice used in this study, Dr. lation of unesterified cholesterol in lysosomes or late endo- Anthony Johns (Berlex Biosciences) for performing the ABCA1 quanti- somes can lead to secondary defects in vesicular trafficking. For tative polymerase chain reaction assay, and Theresa Swayne for assist- ance with the fluorescence microscopy studies. example, the defective trafficking of lactosyl ceramide in ASM- deficient fibroblasts and the defects in late endosomal tubu- REFERENCES lovesicular trafficking observed in NPC cells can be corrected 1. Ross, R. (1995) Annu. Rev. Physiol. 57, 791– 804 by cellular cholesterol depletion (46, 47), and defects in the 2. Libby, P., and Clinton, S. K. (1993) Curr. Opin. Lipidol. 4, 355–363 3. Tabas, I. (2000) Biochim. Biophys. Acta 1529, 164 –174 trafficking of sucrose can be induced in normal cells by cellular 4. Williams, K. J., and Tabas, I. (1995) Arterioscler. Thromb. Vasc. Biol. 15, cholesterol enrichment (48). Thus, it is possible that abnormal 551–561 SM accumulation in ASM knockout macrophages causes an 5. Brown, M. S., and Goldstein, J. L. (1983) Annu. Rev. Biochem. 52, 223–261 6. Tall, A. R. (1998) Eur. Heart J. 19, A31-A35 initial accumulation of FC in lysosomes or some other site, but 7. Wang, N., Silver, D. L., Thiele, C., and Tall, A. R. (2001) J. Biol. Chem. 276, that at least some of the trafficking and efflux defects observed 23742–23747 8. Fielding, P. E., Nagao, K., Hakamata, H., Chimini, G., and Fielding, C. J. in our studies are due to membrane vesiculation defects sec- (2000) Biochemistry 39, 14113–14120 ondary to this initial cholesterol accumulation. 9. Ji, Y., Jian, B., Wang, N., Sun, Y., Moya, M. L., Phillips, M. C., Rothblat, G. H., Swaney, J. B., and Tall, A. R. (1997) J. Biol. Chem. 272, 20982–20985 The cholesterol trafficking defect described in this report is 10. Rothblat, G. H., Llera-Moya, M., Atger, V., Kellner-Weibel, G., Williams, D. L., similar to that which occurs in fibroblasts from patients with and Phillips, M. C. (1999) J. Lipid Res. 40, 781–796 Niemann-Pick C (NPC) disease (11–14). Moreover, recent work 11. Liscum, L., and Klansek, J. J. (1998) Curr. Opin. Lipidol. 9, 131–135 12. Blanchette-Mackie, E. J. (2000) Biochim. Biophys. Acta 1486, 171–183 from our groups has shown that macrophages from NPC mice 13. Ory, D. S. (2000) Biochim. Biophys. Acta 1529, 331–339 have a defect in cholesterol trafficking (49). Although the mo- 14. Naureckiene, S., Sleat, D. E., Lackland, H., Fensom, A., Vanier, M. T., Wattiaux, R., Jadot, M., and Lobel, P. (2000) Science 290, 2298 –2301 lecular etiologies are distinct, important interrelationships be- 15. Kobayashi, T., Beuchat, M. H., Lindsay, M., Frias, S., Palmiter, R. D., tween the cholesterol trafficking defects in ASM deficiency and Sakuraba, H., Parton, R. G., and Gruenberg, J. (1999) Nat. Cell Biol. 1, Niemann-Pick C disease may exist. For example, Reagan et al. 113–118 Cholesterol Trafficking and Efflux in ASM-deficient Macrophages 44983 16. Schuchman, E. H., and Desnick, R. J. (1995) in The Metabolic and Molecular Chem. 275, 15717–15727 Basis of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W. S., and 39. Nishikawa, K., Arai, H., and Inoue, K. (1990) J. Biol. Chem. 265, 5226 –5231 Valle, D., eds) pp. 2601–2624, McGraw-Hill, New York 40. Fadok, V. A., Bratton, D. L., Rose, D. M., Pearson, A., Ezekewitz, R. A., and 17. Slotte, J. P. (1999) Chem. Phys. Lipids 102, 13–27 Henson, P. M. (2000) Nature 405, 85–90 18. Ridgway, N. D. (2000) Biochim. Biophys. Acta 1484, 129 –141 41. Levade, T., Salvayre, R., Maret, A., and Douste-Blazy, L. (1988) J. Inherit. 19. Viana, M. B., Giugliani, R., Leite, V. H., Barth, M. L., Lekhwani, C., Slade, Metab. Dis. 11, 151–157 C. M., and Fensom, A. (1990) J. Med. Genet. 27, 499 –504 42. Maor, I., Mandel, H., and Aviram, M. (1995) Arterioscler. Thromb. Vasc. Biol. 20. Worgall, T. S., McGovern, M. M., Jiang, X. C., Berglund, L., Shea, S., Tabas, I., 15, 1378 –1387 and Deckelbaum, R. J. (2001) Circulation 102, II–30 43. Liscum, L., Ruggiero, R. M., and Faust, J. R. (1989) J. Cell Biol. 108, 21. Tall, A. R., and Wang, N. (2000) J. Clin. Invest. 106, 1205–1207 1625–1636 22. Horinouchi, K., Erlich, S., Perl, D., Ferlinz, K., Bisgaier, C. L., Sandhoff, K., 44. Pentchev, P. G., Comly, M. E., Kruth, H. S., Vanier, M. T., Wenger, D. A., Desnick, R. J., Stewart, C. L., and Schuchman, E. H. (1995) Nature Gen. 10, Patel, S., and Brady, R. O. (1985) Proc. Natl. Acad. Sci. U. S. A. 82, 288 –293 8247– 8251 23. Otterbach, B., and Stoffel, W. (1995) Cell 81, 1053–1061 45. Cruz, J. C., Sugii, S., Yu, C., and Chang, T. Y. (2000) J. Biol. Chem. 275, 24. Tang, W., Walsh, A., and Tabas, I. (1999) Biochim. Biophys. Acta 1437, 4013– 4021 301–316 46. Puri, V., Watanabe, R., Dominguez, M., Sun, X., Wheatley, C. L., Marks, D. L., 25. Havel, R. J., Eder, H., and Bragdon, J. (1955) J. Clin. Invest. 34, 1345–1353 and Pagano, R. E. (1999) Nat. Cell Biol. 1, 386 –388 26. Basu, S. K., Goldstein, J. L., Anderson, R. G. W., and Brown, M. S. (1976) Proc. 47. Zhang, M., Dwyer, N. K., Neufeld, E. B., Love, D. C., Cooney, A., Comly, M., Natl. Acad. Sci. U. S. A. 73, 3178 –3182 Patel, S., Watari, H., Strauss, J. F., III, Pentchev, P. G., Hanover, J. A., and 27. Rinninger, F., Brundert, M., Jackle, S., Kaiser, T., and Greten, H. (1995) Blanchette-Mackie, E. J. (2001) J. Biol. Chem. 276, 3417–3425 Biochim. Biophys. Acta 1255, 141–153 48. Zhang, M., Dwyer, N. K., Love, D. C., Cooney, A., Comly, M., Neufeld, E., 28. Tabas, I., Boykow, G. C., and Tall, A. R. (1987) J. Clin. Invest. 79, 418 – 426 Pentchev, P. G., Blanchette-Mackie, E. J., and Hanover, J. A. (2001) Proc. 29. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Natl. Acad. Sci. U. S. A. 98, 4466 – 4471 Chem. 193, 265–275 49. Chen, W., Sun, Y., Welch, C., Gorelik, A., Leventhal, A. R., Tabas, I., and Tall, 30. Tabas, I., Marathe, S., Keesler, G. A., Beatini, N., and Shiratori, Y. (1996) A. R. (2001) J. Biol. Chem. 276, 43564 – 43569 J. Biol. Chem. 271, 22773–22781 50. Reagan, J. W., Jr., Hubbert, M. L., and Shelness, G. S. (2000) J. Biol. Chem. 31. Blanchette-Mackie, E. J., Dwyer, N. K., Amende, L. M., Kruth, H. S., Butler, 275, 38104 –38110 J. D., Sokol, J., Comly, M. E., Vanier, M. T., August, J. T., Brady, R. O., and 51. Lundberg, B. (1985) Atherosclerosis 56, 93–110 Pentchev, P. G. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 8022– 8026 52. Small, D. M., Bond, M. G., Waugh, D., Prack, M., and Sawyer, J. K. (1984) 32. Tabas, I., Lim, S., Xu, X., and Maxfield, F. R. (1990) J. Cell Biol. 111, 929 –940 J. Clin. Invest. 73, 1590 –1605 33. Wang, N., Silver, D. L., Costet, P., and Tall, A. R. (2000) J. Biol. Chem. 275, 53. Rapp, J. H., Connor, W. E., Lin, D. S., Inahara, T., and Porter, J. M. (1983) J. 33053–33058 Lipid Res. 24, 1329 –1335 34. Lange, Y., Ye, J., Rigney, M., and Steck, T. L. (2000) J. Biol. Chem. 275, 54. Shio, H., Haley, N. J., and Fowler, S. (1979) Lab. Invest. 41, 160 –167 17468 –17475 55. Tangirala, R. K., Mahlberg, F. H., Glick, J. M., Jerome, W. G., and Rothblat, 35. Mukherjee, S., Zha, X., Tabas, I., and Maxfield, F. R. (1998) Biophys. J. 75, G. H. (1993) J. Biol. Chem. 268, 9653–9660 1915–1925 56. Jiang, X. C., Paultre, F., Pearson, T. A., Reed, R. G., Francis, C. K., Lin, M., 36. Tabas, I., Rosoff, W. J., and Boykow, G. C. (1988) J. Biol. Chem. 263, 1266 –1272 Berglund, L., and Tall, A. R. (2000) Arterioscler. Thromb. Vasc. Biol. 20, 2614 –2618 37. Xu, X., and Tabas, I. (1991) J. Biol. Chem. 266, 17040 –17048 38. Botelho, R. J., Hackam, D. J., Schreiber, A. D., and Grinstein, S. (2000) J. Biol. 57. Tabas, I. (1999) Chem. Phys. Lipids 102, 131–139

Journal

Journal of Biological Chemistry – Unpaywall

Published: Nov 1, 2001

There are no references for this article.