ABCA1-Induced Cell Surface Binding Sites for ApoA-I

【摘要】  Objective- The purpose of this study was to understand the interactions of apoA-I with cells expressing ABCA1.

Methods and Results- The binding of wild-type (WT) and mutant forms of human apoA-I to mouse J774 macrophages was examined. Analysis of total binding at 37°C of 125 I-WT apoA-I to the cells and specifically to ABCA1, as determined by covalent cross-linking, revealed saturable high affinity binding in both cases. Determination of the level of cell-surface expression of ABCA1 showed that only about 10% of the apoA-I associated with the cell surface was bound directly to ABCA1. Furthermore, when 125 I -apoA-I was cross-linked to ABCA1-upregulated cells and examined by SDS-PAGE, the major ( 90%) band migrated as monomeric apoA-I. In contrast to WT apoA-I, the C-terminal deletion mutants 190 to 243 and 223 to 243 that have reduced lipid affinity, exhibited marked reductions (50 and 70%, respectively) in their abilities to bind to the surface of ABCA1-upregulated cells. However, these C-terminal deletion mutants cross-linked to ABCA1 as effectively as WT apoA-I.

Conclusions- This study demonstrates that ABCA1 activity creates 2 types of high affinity apoA-I binding sites at the cell surface. The low capacity site formed by direct apoA-I/ABCA1 interaction functions in a regulatory role, whereas the much higher capacity site generated by apoA-I/lipid interactions functions in the assembly of nascent HDL particles.

15-LO-I is expressed in rabbit aorta. 15-LO regulates vasodilatory eicosanoid synthesis and vascular tone. The current study demonstrates that ABCA1 activity creates 2 types of high affinity apoA-I binding sites at the cell surface. The low capacity site formed by direct apoA-I/ABCA1 interaction functions in a regulatory role, whereas the much higher capacity site generated by apoA-I/lipid interactions functions in the assembly of nascent HDL particles.

【关键词】  ABCA apoAI phospholipids binding

Introduction

High density lipoprotein (HDL) is known to possess pleiotropic biological properties which are able to protect against the development of cardiovascular disease. 1-3 An important aspect of HDL function is its role in reverse cholesterol transport (RCT), a process that mediates the transport of cholesterol from peripheral tissues, including the arterial wall, to the liver for biliary secretion. 1,4,5 Cellular lipid efflux mediated by apolipoprotein (apo) A-I, the major HDL protein constituent, is one of the earliest events in RCT.

Recent studies have identified the ATP-binding cassette transporter A1 (ABCA1) as a critical factor in the formation of HDL, and lack of a functional ABCA1 leads to cholesterol accumulation in peripheral tissues and low plasma HDL. 6 The importance of ABCA1 has been demonstrated by the fact that mutations in the ABCA1 gene lead to Tangier disease and familial HDL deficiency. 7 These patients are generally characterized by an accumulation of excess cellular cholesterol, low levels of HDL, and an elevated risk of atherosclerosis. 8 Indeed, a direct correlation between cell surface expression of ABCA1 and apoA-I-mediated cellular lipid efflux has been revealed. 4

Although it is known that the interaction of apoA-I with functional ABCA1 results in the lipidation of apoA-I and formation of HDL, the mechanism of this interaction is not resolved. Prior studies 9-11 have established that the initial binding of apoA-I to the surface of cells in which ABCA1 is upregulated directs a sequence of events leading to the transfer of cellular phospholipids and cholesterol to the apolipoprotein. However, the exact nature of this ABCA1-induced interaction of apoA-I with the cell surface remains a matter of debate. Chemical cross-linking studies have shown that apoA-I forms a complex with ABCA1 in cells expressing this transporter. 12-14 The cell surface-associated apoA-I binds within 3Å of ABCA1, suggesting protein-protein interactions between apoA-I and ABCA1. 15 On the other hand, some studies 16-18 have proposed that apoA-I binds to a modified plasma membrane lipid domain created by ABCA1. An important limitation of the prior studies is that largely qualitative experiments were conducted to study apoA-I binding to ABCA1. The lack of quantitative comparison of the potential sites to which apoA-I binds in ABCA1-expressing cells has made it difficult to distinguish the relative importance of protein-protein and protein-lipid interactions during the reaction. Hence, it is apparent that there is unresolved complexity in understanding the mechanism of apoA-I interaction with ABCA1-expressing cells.

In the current study, we have addressed some of these questions by characterizing the sites to which apoA-I binds during its interaction with ABCA1-expressing cells. Cell surface binding sites activated by this interaction have been studied using engineered variants of apoA-I. Importantly, ABCA1 activity creates 2 binding sites on the cell surface for apoA-I: a low capacity site created by direct apoA-I/ABCA1 interaction and a much higher capacity site that involves apoA-I/lipid interactions. Studies of the cross-linking of apoA-I and its variants with ABCA1 have revealed that formation of the apoA-I/ABCA1 complex is not dependent on a particular structural domain within the apoA-I molecule. However, there is a strong correlation between the lipid binding affinity of the various mutants of apoA-I and effective binding to the major site on the cell membrane created by ABCA1 activity.

Methods

Preparation of ApoA-I

ApoA-I was isolated from fresh plasma of normolipidemic donors as described earlier. 19,20 The construction of plasmids for expressing wild-type (WT) apoA-I and its mutants and the isolation and purification of these proteins have been described in detail previously. 21

Cell Culture and Binding Experiments

ApoA-I was iodinated by the iodine monochloride method 22 to a specific activity of 250 to 1200 cpm/ng. Binding of apoA-I to ABCA1-expressing J774 and Baby hamster kidney (BHK) cells was determined as described previously. 14,18,23 Competition binding assays were performed as described in detail earlier. 14 Chemical cross-linking and immunoprecipitation of bound apoA-I to ABCA1 expressing J774 and BHK cells was carried out as described before. 14,15 Biotinylation of cell surface proteins in J774 cells was performed as described previously. 24

Please see the supplemental materials, available online at http://atvb.ahajournals.org, for more methodological details.

Results

Binding of ApoA-I to ABCA1-Upregulated Cells

In the current study, binding of 125 I apoA-I was examined in control and ABCA1-expressing J774 and BHK cells. As shown in Figure 1 A, increasing the concentration of apoA-I resulted in a marked increase in the apoA-I binding to stimulated J774 cells. Significant binding was also observed in unstimulated cells, possibly because of the basal level of ABCA1 expression and nonspecific binding of apoA-I to other cell surface sites such as the extracellular matrix. 25 The isotherm for specific binding of apoA-I to ABCA1 was obtained by subtracting the binding values for unstimulated cells from the corresponding values for stimulated cells. The binding parameters B max (105±12 ng apoA-I/mg cell protein) and K d (1.2±0.7 µg apoA-I/mL) were calculated from fitting the hyperbolic specific binding curve to a 1-site binding model. A 2-site binding model did not fit the data satisfactorily. It is apparent that apoA-I associates with the J774 cell surface with high affinity (K d 0.04 µmol/L) to give saturable binding (cf. 26 ). The specific binding was also evaluated by incubating stimulated cells with different concentrations of 125 I apoA-I for 2 hours at 37°C in the absence or presence of a 30-fold excess of unlabeled apoA-I to block nonspecific binding. The values of K d (2.6±0.9 µg apoA-I/mL) and B max (135±20 ng apoA-I/mg of cell protein) from this analysis were similar to those derived from Figure 1 A. The above binding parameters are consistent with the results of similar experiments conducted with different cell types. 14,27

Figure 1. Cell association and cross-linking of 125 I-apoA-I to ABCA1: J774 cells (A and B) stimulated or not with cAMP and human-ABCA1 expressing and mock -transfected BHK cells (C and D) were incubated for 2 hours at 37°C with increasing concentrations of 125 I-apoA-I (0.5, 1, 2, 5, 10, and 20 µg/mL). After binding, cells were incubated with the crosslinker (DSP) for 30 minutes at room temperature, lysed, and radioactivity determined from aliquots of the cell lysate. A and C, The specific binding curve (ABCA1 specific [-]) was determined by subtracting the binding values for the control cells ( ) from the corresponding values for ABCA1-expressing cells ( ). The values for total apoA-I binding are the mean±SD from at least 2 independent experiments performed in triplicate. B and D, Aliquots of the cell lysate were immunoprecipitated with an anti-ABCA1 antibody, separated on SDS-PAGE, transferred, and analyzed by phosphorimager. The values for apoA-I cross-linked to ABCA1 in D are representative from 2 independent experiments performed in triplicate, and the values shown in D are single values from 1 of 3 representative experiments.

To establish that the specific binding parameters obtained above are associated with an increase in binding of apoA-I to ABCA1 in stimulated cells, 125 I apoA-I was examined for its ability to crosslink to ABCA1. Stimulation of cells with cAMP increased both cellular ABCA1 expression and apoA-I/ABCA1 cross-linking compared with unstimulated cells (data not shown). Treatment with increasing concentrations of radioiodinated WT apoA-I enhanced the cross-linking of apoA-I, and maximum cross-linking ( 3 ng/mg cell protein) was attained at a concentration of 20 µg apoA-I/mL ( Figure 1 B). It is important to note that values for total (B max = 105±12 ng/mg cell protein) and cross-linked (B max =3.6±0.4 ng/mg cell protein) apoA-I binding are clearly markedly different by some 30-fold, indicating that only a minor part of the cell-associated apoA-I apparently binds directly to ABCA1. Analysis of the binding isotherm in Figure 1 B showed that the K d value for cross-linking with WT apoA-I was 5.3±1.5 µg apo/mL, which is higher than the apparent K d attained from Figure 1 A.

It is apparent from the above results that ABCA1 activity in J774 cells creates 2 types of binding sites for apoA-I. To establish that these binding sites were caused only by ABCA1 activity and were not attributable to any other factor induced by cAMP, the binding and cross-linking experiments described above were repeated in mock and human ABCA1-transfected BHK cells. As is evident from Figure 1 C, 125 I-apoA-I exhibited concentration-dependent binding to ABCA1-transfected BHK cells; the B max (70±3 ng/mg cell protein) and the K d (0.06±0.04 µg apoA-I/mL) values for the specific binding originating from ABCA1 activity were obtained as described above. The total amount of apoA-I bound to the cell and cross-linked to ABCA1 (3.0±0.3 ng/mg cell protein) ( Figure 1 D) were significantly different, indicating once again that a major part of the cell-associated apoA-I is bound to another site created by the activity of ABCA1. It is important to note that the reaction conditions (see Methods) were selected to give quantitative cross-linking of apoA-I/ABCA1 complexes present at the cell surface. This condition was essentially achieved because cross-linking and immunoprecipitation with anti-apoA-I removed about one-third of the total cell ABCA1 (data not shown) that is consistent with the fraction of transporter present at the cell surface at any given time (see below).

Cell Surface Expression of ABCA1

To determine the amount of ABCA1 present at the cell surface relative to the total amount of ABCA1 in the cells, biotinylation experiments were carried out on J774 cells upregulated or not with cAMP. Western blot analysis was conducted on whole cell protein extracts to detect ABCA1 which migrated at a molecular 220kDa ( Figure 2, lane C). Densitometric scanning showed that treatment with cAMP increased the expression of total cellular ABCA1 by almost 8-fold (cf. Figure 2, lanes C and D). As earlier studies have indicated that ABCA1 may be most active at the plasma membrane, 11,28 all the surface proteins in J774 cells were nonspecifically biotinylated using sulfo-NHS-SS-biotin. These biotinylated proteins were then isolated with streptavidin immobilized on agarose beads and analyzed by immunoblotting with a polyclonal antibody to ABCA1. It is evident from Figure 2 (lane B) that cAMP augments the expression of ABCA1 at the cell surface. The ABCA1 appeared as a doublet consistent with earlier reports. 29 Comparison of the expression of ABCA1 at the cell surface with that in the total cell lysate by densitometric scanning (cf. Figure 2, lanes B and C) indicates that about 30% to 40% of the total cellular ABCA1 is present at the cell surface at any given time.

Figure 2. Effect of cAMP on total and cell surface expression of ABCA1: J774 cells were incubated for 14 hours in the presence or absence of cAMP and treated with WT apoA-I (10 µg/mL) for 2 hours at 37°C to stabilize the ABCA1. After washing, cell surface proteins were biotinylated at 4°C for 30 minutes, the cells were lysed, and the protein concentration was determined. 5-µg aliquots of the cell lysate were used to determine total cellular ABCA1, whereas 100-µg aliquots were incubated with streptavidin beads to isolate biotinylated proteins for the determination of cell surface ABCA1. Western blot analysis was performed using anti-ABCA1 antibody to detect total cellular (lanes C and D) and surface (lanes A and B) ABCA1 for cells incubated with (lanes B and C) and without (lanes A and D) cAMP. The blot is a representative of 3 different experiments.

To further confirm the relative surface expression of ABCA1, displacement experiments were carried out with J774 cells stimulated with cAMP. After binding 125 I-apoA-I (2 µg/mL) for 2 hours at 37°C, the cell monolayers were kept on ice, washed, and treated with excess unlabeled apoA-I (100 µg/mL medium). The cells were then incubated for 6 hours at 4°C, and other cells were treated in parallel at 37°C. The displacement of cell-associated 125 I-apoA-I followed monoexponential decay kinetics and, as shown in Figure 3, about 30% of labeled apoA-I was displaced from cells placed at 4°C whereas about 80% was displaced from cells held at 37°C. Because internalization of the transporter was not possible for cells kept at 4°C, only labeled apoA-I bound to the cell surface as a result of ABCA1 activity was displaced by excess unlabeled apoA-I. The observation that only 30% of the 125 I-apoA-I was displaced is in good agreement with the finding from the biotinylation experiment that 30% to 40% of total cellular ABCA1 is present at the cell surface at any given time. On the other hand, for the cells held at 37°C, radioiodinated apoA-I associated with ABCA1 cycling between the cell interior and the cell surface was displaced by unlabeled apoA-I.

Figure 3. Displaceability of cell-associated apoA-I: J774 cells after upregulation with cAMP were incubated with 125 I-apoA-I (2 µg/mL) for 2 hours at 37°C, and the cells were split into 2 sets. One set was placed at 4°C ( ) while the other was placed at 37°C ( ), and both were incubated with excess unlabeled WT apoA-I (100 µg/mL) for up to 6 hours. At specific time points, cells were washed, lysed, and the cell-associated radioactivity determined from aliquots of the cell lysate. The symbols represent the mean of triplicate measurements ±SD.

Binding of ApoA-I to Other Sites in Cells

To further explore the nature of the binding sites of the cell-associated apoA-I, J774 cells expressing ABCA1 were treated with 20 µg/mL of 125 I-apoA-I for 2 hours to ensure saturation of all sites and then cross-linked with DSS. This nonreducible cross-linker was used so that the cross-links between apoA-I and any cellular protein were maintained during subsequent SDS-PAGE under reducing conditions. Aliquots of the cell lysate were either directly analyzed by SDS-PAGE or were first immunoprecipitated with anti-apoA-I antibody and then analyzed by SDS-PAGE; the resultant blots were analyzed by phosphorimager ( Figure 4 ). Immunoprecipitation with anti-apoA-I brought down a lot more 125 I counts ( 80%) of the total compared with precipitation with anti-ABCA1 ( 15% of the total) consistent with the relative bound apoA-I pool sizes in Figure 1A and 1 B. When the cell lysates were analyzed directly by SDS-PAGE, only a single major band representing the monomeric form of apoA-I was identified ( Figure 4 A). Also, compared with control cells there was a 2- to 3-fold increase in the intensity of the apoA-I band in ABCA1-expressing cells. Similar results were also obtained with mock-transfected and ABCA1-expressing BHK cells cross-linked to apoA-I (data not shown). It is important to note that after immunoprecipitation with anti-apoA-I, the major band (band 1, in Figure 4 B) migrated at the molecular weight (28 000) equivalent to that of monomeric apoA-I. The ABCA1/apoA-I complex migrated at the expected molecular weight ( 250 000; band 4 in Figure 4 B). Bands 2 and 3 corresponded to the cross-linked dimeric and trimeric forms of apoA-I. Thus, it is apparent from Figure 4 that the apoA-I bound to the J774 cells as a consequence of ABCA1 activity was not associated with any other cellular protein, as evaluated by cross-linking. Indeed, while ABCA1 expression increased the cell-association of apoA-I considerably to 100 ng/mg cell protein more than 90% of this apoA-I was present in its monomeric form. For comparison, using control cells in which ABCA1 was not upregulated, band 4 was not apparent (data not shown).

Figure 4. Immunoprecipitation of cell-associated apoA-I: J774 cells were either upregulated or not upregulated with cAMP and incubated with 125 I-apoA-I for 2 hours and treated with the nonreducible crosslinker DSS (500 µmol/L) for 30 minutes at room temperature as described in the legend to Figure 1. A, Aliquots of the cell lysate were analyzed directly by SDS-PAGE and the 125 I-apoA-I detected by phosphorimager analysis. B, Aliquots of the cell lysate were immunoprecipitated with anti-apoA-I antibody and analyzed by SDS-PAGE to determine the fraction of cell-associated apoA-I that was cross-linked to cellular proteins. Results are representative of three independent experiments.

Effects of N- and C-Terminal Mutations in ApoA-I on Cellular and Lipid Interactions

To explore the concept that the majority of the apoA-I bound to cells as a consequence of ABCA1 activity is located on lipid sites, we used apoA-I variants with altered lipid binding properties. Deletion and proline insertion mutants of apoA-I were incubated with either ABCA1-stimulated or -unstimulated J774 cells, and their efficacies of binding were compared. The specific binding curves for the various mutants were obtained as described earlier for WT apoA-I, and the results are depicted in Figure 5 A. Deletion of the C-terminal domain ( 190 to 243) and the C-terminal -helix ( 223 to 243) led to 50% and 70% reductions in B max ( Table ), respectively, relative to the total cell binding of WT apoA-I. These results are in agreement with an earlier study in which the C-terminal mutant ( 187 to 243) showed a 58% reduction in total binding to THP-1 macrophages compared with WT apoA-I. 30 In contrast, disruption of the N-terminal domain by inserting a proline residue in that region (apoA-I Y18P) had no effect on total cell binding ( Table ).

Figure 5. Influence of C-terminal deletions in apoA-I on binding to cells, ABCA1, and phospholipid SUV: The binding and cross-linking characteristics of the C-terminal mutants ( 190 to 243, 223 to 243) were determined with cAMP-stimulated or -unstimulated cells, as described in the legend to Figure 1. A, The specific binding isotherms for WT apoA-I ( ), 190 to 243 (-), and 223 to 243 ( ) at various concentrations. B, Concentration-dependence of the amounts of apoA-I crosslinked to ABCA1 for WT apoA-I ( ), 190 to 243 (-), and 223 to 243 ( ). C, Binding isotherms of WT apoA-I ( ), 190 to 243 (-), and 223 to 243 ( ) to egg PC SUV (data replotted from ref. 32 ). D, Comparison of the amounts of apoprotein bound to cells and to SUV at 10 µg/mL, as derived from their respective binding isotherms (A and C). The open bars represent apoA-I binding to cells, whereas the solid bars represent apoA-I binding to the phospholipid bilayer of SUV.

Binding and Cross-Linking Parameters for Interaction of ApoA-I Variants With Cells or ABCA1

To analyze the effects of the alterations in apoA-I structure on the ability of apoA-I to bind directly to ABCA1, the C-terminal truncation mutants of apoA-I were assessed for their capacity to crosslink to ABCA1. As depicted in Figure 5 B, these mutants and WT apoA-I cross-linked similarly to ABCA1. The B max values for the N- and C-terminal apoA-I mutants were around the value of 3 ng/mg cell protein observed with WT apoA-I ( Table ). These results show that the structural requirements for binding of apoA-I to ABCA1 are not highly stringent; apparently the presence of amphipathic -helices is sufficient to allow apoA-I to bind to ABCA1. In agreement with this idea, peptides of identical sequences containing either D or L enantiomers have been shown to have the same ability to bind to ABCA1 revealing that there is no stereoselective requirement for binding. 31

Figure 5 C compares the binding isotherms for WT apoA-I, 190 to 243 and 223 to 243 to egg PC small unilamellar vesicles (SUV). 32 It is evident from the binding isotherms that the abilities of the C-terminal deletion mutants to bind to a phospholipid bilayer are greatly reduced compared with WT apoA-I. This effect occurs because apoA-I initially binds to lipid surfaces via its C-terminal domain. 21,33 Importantly, deletion of the C-terminal region caused parallel reductions in the ability to bind to ABCA1-expressing cells and egg PC bilayers ( Figure 5 D). This correlation is consistent with the idea that the second site for the apoA-I bound to cells is a membrane lipid site that is created by the expression of active ABCA1. Indeed, earlier studies have shown that ABCA1 can modify the lipid distribution in a cell plasma membrane. 16,18

To further evaluate the importance of the lipid-binding properties of apoA-I in its ability to associate with ABCA1-expressing cells, we assessed the capabilities of the N- and C-terminal mutants of apoA-I to compete for the binding of 125 I apoA-I to ABCA1-upregulated cells. As expected, WT apoA-I competes effectively with 125 I apoA-I, such that 75% to 80% of the 125 I apoA-I binding was inhibited at concentrations of unlabeled apoA-I 50 µg/mL (supplemental Figure I). The lack of complete inhibition was probably attributable to a combination of nonspecific binding of the radiolabeled ligand to the cells and to the plastic tissue culture plates. The EC 50 value for WT apoA-I was 2.9±1.2 µg apo/mL (equivalent to about 0.1 µmol/L apoA-I; supplemental Table I). In contrast, the C-terminal deletion mutants ( 190 to 243, 223 to 243) were less effective competitors; the EC 50 values were almost 30-fold and 180-fold higher for the 190 to 243 and 223 to 243 variants, respectively (supplemental Table I). These results are consistent with the protein binding data shown in Figure 5 A and Table; the apoA-I C-terminal mutants with impaired lipid binding properties have reduced capacities both to bind to ABCA1-upregulated cells and compete with WT apoA-I for binding to such cells. Significantly, the N-terminal mutant (Y18P), which has similar lipid-binding properties to WT apoA-I, 32 competed for the binding of 125 I-apoA-I almost as effectively as WT apoA-I (supplemental Table I).

Discussion

The ABCA1 transporter is critical for the biogenesis of HDL through its ability to facilitate apoA-I-mediated removal of phospholipids and cholesterol from peripheral and hepatic cells. 1,3 ApoA-I is catabolized rapidly in the absence of functional ABCA1 leading to free cholesterol accumulation in peripheral tissues and low levels of plasma HDL. 4,34 Hence, it is apparent that ABCA1 plays a vital role in the lipidation of apolipoproteins, such as apoA-I. To achieve the important goal of elucidating the mechanism of interaction between apoA-I and ABCA1, it is necessary to understand the cellular binding sites for apoA-I created by ABCA1 activity. Many theories have been proposed to explain the mechanism of interaction of apoA-I with ABCA1-expressing cells. Some studies 35,36 have pointed to the importance of apoA-I/ABCA1 interactions, whereas others 16,17 have postulated the importance apoA-I/lipid interactions for the overall binding process. However, so far a consensus has not been reached and the mechanism of this reaction is still unresolved. Furthermore, the mechanism of formation of nascent HDL particles during ABCA1-mediated lipid efflux is still not clearly understood. Current thinking is that it involves 2 steps. 15,23 The first step involves the formation of a complex between apoA-I and ABCA1, and the second step involves the lipidation of apoA-I and dissociation of the nascent HDL particle from the transporter. This model implies that all of the cell-associated apoA-I is bound to ABCA1 where it becomes transformed into nascent HDL particles. In the current study, we present quantitative binding information showing that ABCA1 activity creates 2 binding sites for apoA-I.

Cell-Surface Pool of ApoA-I Bound to ABCA1

As summarized in the Results Section, apoA-I exhibits saturable high affinity binding to ABCA1. The B max for binding to ABCA1 in J774 cells is 3 ng apoA-I/mg of cell protein ( Table ); this interaction is not very specific and apparently involves interaction of apoA-I amphipathic -helices with the transporter. This concept is supported by the data in the Table showing that all of apoA-I mutants (both N- and C-terminal domain) studied bind equally well to ABCA1. This interaction of apoA-I with ABCA1 requires signaling responses mediated by Janus kinase 2 29 and is associated with stabilization of ABCA1 at the cell surface by protecting it from proteolytic degradation. 5,9,37 Thus, the low capacity apoA-I binding site on the ABCA1 molecule has regulatory functions ( Figure 6 ).

Figure 6. Schematic comparing the two types of apoA-I binding sites present at the surface of cells expressing functional ABCA1. The low capacity site involving direct apoA-I/ABCA1 interaction serves a regulatory function and stabilizes ABCA1. The high capacity site involves apoA-I binding to a membrane lipid domain created by ABCA1 activity.

Nature of Second ApoA-I Binding Site

The size of the second binding site equals the total amount of 125 I-apoA-I bound to the cell (100 ng/mg cell protein) minus the amount of protein bound specifically to ABCA1 (3 ng/mg of cell protein). However, from the biotinylation ( Figure 2 ) and apoA-I displacement ( Figure 3 ) experiments, it is apparent that only around 35% of the ABCA1 is present on the surface of the cells. Hence, the actual amount of apoA-I associated with the second binding site on the surface is equal to about 30 ng of apoA-I/mg cell protein [(100 ng · 0.35)-3 ng)]. It follows that the amount of apoA-I associated with the second site is some 10-fold greater than that bound to ABCA1 indicating that the second site is a high capacity binding site ( Figure 6 ), as has been postulated previously. 4

Cross-linking of bound apoA-I followed by immunoprecipitation with anti-apoA-I has demonstrated that, apart from ABCA1, apoA-I does not bind to any other cellular protein ( Figure 4 B). In fact, 90% of the bound apoA-I is monomeric after cross-linking, consistent with it being located on a lipid site. The K d value for the binding of WT apoA-I to the cell surface is around 1 µg/mL. Interestingly, the K d for binding of WT apoA-I to egg PC SUV, a surrogate for the phospholipid bilayer of a cell membrane, is in the same range. 32 This similarity in K d values lends strong support to the idea that the other binding site for apoA-I generated by ABCA1 activity is a lipid site. In support of this interpretation, ABCA1 expression in BHK cells redistributes cholesterol to cell-surface domains 18 and causes a change in the overall lipid packing of the plasma membrane. 38 Furthermore, removal of either the strongly hydrophobic C-terminal domain ( 190 to 243) or the last helix ( 223 to 243) of WT apoA-I reduces the level of binding to cells by 50 and 70% respectively, compared with WT apoA-I. The reason that these mutants associate poorly with the cells is their inability to bind well to the second site created by ABCA1 activity. This notion is further supported by the observation that the C-terminal mutants compete badly for binding to the cells compared with WT apoA-I (supplemental Table I). Deletion of the C-terminal region of apoA-I results in low affinity non-saturable binding to egg PC SUV as compared with WT apoA-I which exhibits high affinity saturable binding ( Figure 5 C, cf 32 ). It is apparent that the hydrophobic C-terminal -helices of apoA-I are needed for effective insertion into the phospholipid bilayer and high affinity binding. 32,33 The fact that removal of these functional -helices from the apoA-I molecule leads to similar reductions in binding to both phospholipid vesicles and the second cell surface binding site created by ABCA1 activity ( Figure 5 D) is consistent with the latter binding site being a lipid site. The affinity of the N-terminal apoA-I (Y18P) mutant for the second cell binding site is essentially the same as that of WT apoA-I ( Table ) implying that the N-terminal helix-bundle domain in apoA-I 33 does not play a major role in lipid binding. This finding is consistent with the fact that the binding affinity of apoA-I Y18P to phospholipid SUV is similar to that of WT apoA-I. 32 Because the high capacity apoA-I binding site created by ABCA1 activity is a lipid site it follows that this site functions in the assembly of nascent HDL particles.

It is generally thought that the ABCA1 molecule itself is the major binding site for apoA-I in ABCA1-expressing cells. However, our study shows that 2 binding sites for apoA-I are created at the surface of cells containing functional ABCA1. Furthermore, the major binding site is the modulated lipid site orchestrated by the ABCA1 activity. Hence, it is evident that there is a need to reevaluate the current theories on the formation of nascent HDL particles during apoA-I/ABCA1 interaction.

Acknowledgments

We thank Dr Eric Alexander for helpful discussions.

Sources of Funding

This work was supported by NIH grants HL22633, HL67093, and HL62542.

Disclosures

None.

【参考文献】
  Rader DJ. Regulation of reverse cholesterol transport and clinical implications. Am J Cardiol. 2003; 92: 42J-49J.

Yancey PG, Bortnick AE, Kellner-Weibel G, Llera-Moya M, Phillips MC, Rothblat GH. Importance of different pathways of cellular cholesterol efflux. Arterioscler Thromb Vasc Biol. 2003; 23: 712-719.

Barter PJ, Rye KA. Relationship between the concentration and antiatherogenic activity of high-density lipoproteins. Curr Opin Lipidol. 2006; 17: 399-403.

Krimbou L, Marcil M, Genest J. New insights into the biogenesis of human high-density lipoproteins. Curr Opin Lipidol. 2006; 17: 258-267.

Yokoyama S. Assembly of high-density lipoprotein. Arterioscler Thromb Vasc Biol. 2006; 26: 20-27.

Oram JF, Heinecke JW. ATP-binding cassette transporter A1: a cell cholesterol exporter that protects against cardiovascular disease. Physiol Rev. 2005; 85: 1343-1372.

Nofer JR, Remaley AT. Tangier disease: still more questions than answers. Cell Mol Life Sci. 2005; 62: 2150-2160.

Lee JY, Parks JS. ATP-binding cassette transporter AI and its role in HDL formation. Curr Opin Lipidol. 2005; 16: 19-25.

Wang N, Tall AR. Regulation and mechanisms of ATP-binding cassette transporter A1-mediated cellular cholesterol efflux. Arterioscler Thromb Vasc Biol. 2003; 23: 1178-1184.

Wang N, Chen W, Linsel-Nitschke P, Martinez LO, Agerholm-Larsen B, Silver DL, Tall AR. A PEST sequence in ABCA1 regulates degradation by calpain protease and stabilization of ABCA1 by apoA-I. J Clin Invest. 2003; 111: 99-107.

Oram JF, Lawn RM, Garvin MR, Wade DP. ABCA1 is the cAMP-inducible apolipoprotein receptor that mediates cholesterol secretion from macrophages. J Biol Chem. 2000; 275: 34508-34511.

Wang N, Silver DL, Costet P, Tall AR. Specific binding of ApoA-I, enhanced cholesterol efflux, and altered plasma membrane morphology in cells expressing ABC1. J Biol Chem. 2000; 275: 33053-33058.

Fitzgerald ML, Morris AL, Rhee JS, Andersson LP, Mendez AJ, Freeman MW. Naturally occurring mutations in the largest extracellular loops of ABCA1 can disrupt its direct interaction with apolipoproteinA-I. J Biol Chem. 2002; 277: 33178-33187.

Denis M, Haidar B, Marcil M, Bouvier M, Krimbou L, Genest J Jr. Molecular and cellular physiology of apolipoproteinA-I lipidation by the ATP-binding cassette transporter A1 (ABCA1) J Biol Chem. 2004; 279: 7384-7394.

Chroni A, Liu T, Fitzgerald ML, Freeman MW, Zannis VI. Cross-linking and lipid efflux properties of ApoA-I mutants suggest direct association between ApoA-I helices and ABCA1. Biochemistry. 2004; 43: 2126-2139.

Chambenoit O, Hamon Y, Marguet D, Rigneault H, Rosseneu M, Chimini G. Specific docking of apolipoproteinA-I at the cell surface requires a functional ABCA1 transporter. J Biol Chem. 2001; 276: 9955-9960.

Panagotopulos SE, Witting SR, Horace EM, Hui DY, Maiorano JN, Davidson WS. The role of apolipoproteinA-I helix 10 in apolipoprotein-mediated cholesterol efflux via the ATP-binding cassette transporter ABCA1. J Biol Chem. 2002; 277: 39477-39484.

Vaughan AM, Oram JF. ABCA1 redistributes membrane cholesterol independent of apolipoprotein interactions. J Lipid Res. 2003; 44: 1373-1380.

Lund-Katz S, Phillips MC. Packing of cholesterol molecules in human low-density lipoprotein. Biochemistry. 1986; 25: 1562-1568.

Weisweiler P. Isolation and quantitation of apolipoproteinsA-I andA-II from human high-density lipoproteins by fast-protein liquid chromatography. Clin Chim Acta. 1987; 169: 249-254.

Saito H, Dhanasekaran P, Nguyen D, Holvoet P, Lund-Katz S, Phillips MC. Domain structure and lipid interaction in human apolipoproteinsA-I and E, a general model. J Biol Chem. 2003; 278: 23227-23232.

Goldstein JL, Basu SK, Brown MS. Receptor-mediated endocytosis of low-density lipoprotein in cultured cells. Methods Enzymol. 1983; 98: 241-260.

Vedhachalam C, Liu L, Nickel M, Dhanasekaran P, Anantharamaiah GM, Lund-Katz S, Rothblat GH, Phillips MC. Influence of apoA-I structure on the ABCA1-mediated efflux of cellular lipids J Biol Chem. 2004; 279: 49931-49939.

Witting SR, Maiorano JN, Davidson WS. Ceramide enhances cholesterol efflux to apolipoproteinA-I by increasing the cell surface presence of ATP-binding cassette transporter A1. J Biol Chem. 2003; 278: 40121-40127.

Burgess JW, Kiss RS, Zheng H, Zachariah S, Marcel YL. Trypsin-sensitive and lipid-containing sites of the macrophage extracellular matrix bind apolipoproteinA-I and participate in ABCA1-dependent cholesterol efflux. J Biol Chem. 2002; 277: 31318-31326.

Sakr SW, Williams DL, Stoudt GW, Phillips MC, Rothblat GH. Induction of cellular cholesterol efflux to lipid-free apolipoproteinA-I by cAMP. Biochim Biophys Acta. 1999; 1438: 85-98.

Smith JD, Miyata M, Ginsberg M, Grigaux C, Shmookler E, Plump AS. Cyclic AMP induces apolipoprotein E binding activity and promotes cholesterol efflux from a macrophage cell line to apolipoprotein acceptors. J Biol Chem. 1996; 271: 30647-30655.

Neufeld EB, Remaley AT, Demosky SJ, Stonik JA, Cooney AM, Comly M, Dwyer NK, Zhang M, Blanchette-Mackie J, Santamarina-Fojo S, Brewer HB Jr. Cellular localization and trafficking of the human ABCA1 transporter. J Biol Chem. 2001; 276: 27584-27590.

Tang C, Vaughan AM, Anantharamaiah GM, Oram JF. Janus kinase 2 modulates the lipid-removing but not protein-stabilizing interactions of amphipathic helices with ABCA1. J Lipid Res. 2006; 47: 107-114.

Burgess JW, Frank PG, Franklin V, Liang P, McManus DC, Desforges M, Rassart E, Marcel YL. Deletion of the C-terminal domain of apolipoproteinA-I impairs cell surface binding and lipid efflux in macrophage. Biochemistry. 1999; 38: 14524-14533.

Remaley AT, Thomas F, Stonik JA, Demosky SJ, Bark SE, Neufeld EB, Bocharov AV, Vishnyakova TG, Patterson AP, Eggerman TL, Santamarina-Fojo S, Brewer HB. Synthetic amphipathic helical peptides promote lipid efflux from cells by an ABCA1-dependent and an ABCA1-independent pathway. J Lipid Res. 2003; 44: 828-836.

Tanaka M, Dhanasekaran P, Nguyen D, Ohta S, Lund-Katz S, Phillips MC, Saito H. Contributions of the N- and C-terminal helical segments to the lipid-free structure and lipid interaction of apolipoproteinA-I. Biochemistry. 2006; 45: 10351-10358.

Saito H, Lund-Katz S, Phillips MC. Contributions of domain structure and lipid interaction to the functionality of exchangeable human apolipoproteins. Prog Lipid Res. 2004; 43: 350-380.

Barter PJ, Rye KA. High density lipoproteins and coronary heart disease. Atherosclerosis. 1996; 121: 1-12.

Wang N, Silver DL, Thiele C, Tall AR. ATP-binding cassette transporter A1 (ABCA1) functions as a cholesterol efflux regulatory protein. J Biol Chem. 2001; 276: 23742-23747.

Fitzgerald ML, Morris AL, Chroni A, Mendez AJ, Zannis VI, Freeman MW. ABCA1 and amphipathic apolipoproteins form high-affinity molecular complexes required for cholesterol efflux. J Lipid Res. 2004; 45: 287-294.

Oram JF. HDL apolipoproteins and ABCA1: partners in the removal of excess cellular cholesterol. Arterioscler Thromb Vasc Biol. 2003; 23: 720-727.

Landry YD, Denis M, Nandi S, Bell S, Vaughan AM, Zha X. ATP-binding cassette transporter A1 expression disrupts raft membrane microdomains through its ATPase-related functions. J Biol Chem. 2006; 281: 36091-36101.

作者单位：Division of GI/Nutrition (C.V., S.L.-K., G.H.R., M.C.P.), The Children?s Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia; and the Department of Pathology and Laboratory Medicine (A.B.G., W.S.D.), University of Cincinnati, Ohio.