This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Asztalos, B. Articles by Wong, L. Search for Related Content PubMed PubMed Citation Articles by Asztalos, B. Articles by Wong, L. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB CHOLESTEROL

(Arteriosclerosis, Thrombosis, and Vascular Biology. 1997;17:1630-1636.)
© 1997 American Heart Association, Inc.

Articles

Role of Free Apolipoprotein A-I in Cholesterol Efflux

Formation of Pre–-Migrating High-Density Lipoprotein Particles

Bela Asztalos; Wenwu Zhang; Paul S. Roheim; ; Laurence Wong

From the Division of Lipoprotein Metabolism and Pathophysiology, Department of Physiology, Louisiana State University Medical Center, New Orleans, La.

Correspondence to Laurence Wong, Division of Lipoprotein Metabolism and Pathophysiology, Department of Physiology, Louisiana State University Medical Center, New Orleans, LA 70112.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Abstract This article characterizes products formed by the interaction of purified apolipoprotein (apo) A-I and human fibroblasts. Fibroblasts were incubated with different concentrations of purified apoA-I (1 to 30 µg/mL) in tissue culture medium for different periods of time (0 to 24 hours). The medium was then characterized by one- (agarose) and two-dimensional (agarose : polyacrylamide nondenaturing gradient gel) electrophoresis. At any given concentration of apoA-I, the rate of cellular cholesterol efflux appeared linear over 24 hours. Incubating purified apoA-I with fibroblasts for 4 hours, we detected five pre- lipoproteins with particle sizes between 114 and 684 kDa. Formation of pre- lipoproteins was concentration-dependent. At low concentrations (below 5 µg/mL apoA-I), all purified apoA-I (with pre-ß mobility) was converted to pre- lipoproteins. At higher concentrations (greater than 5 µg/mL apoA-I), more apoA-I remained with pre-ß mobility. The pre- lipoproteins were characterized by colocalization of apoA-I particles with ¹⁴C-cholesterol and ³²P-phospholipids. Results showed that the pre- particle of lowest molecular weight contained phospholipid and apoA-I but no cholesterol. The remaining pre- particles contained all three substances. When pre- particles were subjected to ultracentrifugation, all particles floated at d<1.21 g/mL with some of the smallest phospholipid apoA-I only particles being present in the d>1.21 g/mL fraction. Based on these results, we postulated that in the first stages of reverse cholesterol transport, pre- lipoproteins are formed by the interaction of lipid free apoA-I and peripheral cells.

Key Words: high-density lipoproteins • pre- lipoproteins • pre-ß lipoproteins • cholesterol efflux • fibroblasts • cholesterol • two-dimensional electrophoresis • apolipoprotein A-I • phospholipids

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Peripheral cells cannot break down and metabolize cholesterol. To maintain cell homeostasis, peripheral cells transfer excess cholesterol and transport it to hepatocytes to be metabolized.¹ Numerous studies have examined the effect of cholesterol acceptors mediating cholesterol efflux in vitro from peripheral cells. The role of apolipoprotein (apo),² HDL,³ recombinant HDL,⁴ and phospholipid vesicles⁵ on cellular cholesterol efflux has been studied. In most of these studies, cells were labeled with radioactive cholesterol, incubated with medium containing one or another of the above particles, and radioactivity of transferred cholesterol examined as a function of time. In other studies, total cholesterol mass in the medium was determined after introduction of acceptors. The results of these complementary studies can generally be summarized as follows: (1) most apoprotein appear to be adept in acting as cholesterol acceptors^2,6; (2) in the presence of unlimited acceptors, the rate of cellular cholesterol efflux is limited by diffusion of cholesterol from the cell membrane³; (3) apoproteins may not be necessary for cholesterol efflux, and phospholipid vesicles are also effective²; and (4) artificial cholesterol acceptors, such as cyclodextrans,⁷ can also promote cholesterol efflux.

Currently, two major processes are generally accepted as being responsible for cholesterol efflux (reverse cholesterol transport). The first, a specific process of rapid movement of cholesterol from plasma membrane involves lipoproteins/apoprotein. The second, a nonspecific process, involves diffusion of a slowly exchanging pool of cholesterol from cell membrane.^1,8,9

Recently, we developed a two-dimensional agarose nondenaturing polyacrylamide gradient gel electrophoresis system (2DE) to analyze plasma HDL.¹⁰ Using this procedure, we have observed "free apoA-I–like" particles in plasma.¹¹ The purposes of the study presented here were to determine if these free apoA-I particles could accept cholesterol from peripheral cells and what type of a cholesterol-containing particle would result.

The study presented here differs from previous ones in its methodology of characterization of the newly formed cholesterol-apoprotein complex. To model this newly formed complex in vitro, we incubated fibroblasts with purified apoA-I and characterized the product of apoA-I cell interaction. Using the method of 2DE,¹² we demonstrated the presence of five different lipoprotein particles (containing apoA-I), all with pre- mobility but differing in size from 114 to 684 kDa. Particles with other than pre- mobility were not recovered.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
ApoA-I Purification
ApoA-I was purified from normolipemic human plasma (Southeast Blood Center) HDL (1.080<d<1.21 g/mL) by modification of a procedure described by Nichols et al¹³ and Forte et al.¹⁴ Solid guanidine HCl (ICN) was added to 3 mg/mL HDL to give a final concentration of 2 mol/L. This solution was incubated at 37°C for 3 hours. After incubation, the mixture was dialyzed four times against a solution containing 150 mmol/L NaCl, 10 mmol/L Tris-HCl, and 0.01% EDTA at pH 8.0 to remove the guanidine HCl. All chemicals were from Sigma unless otherwise noted. After dialysis, density of the solution was adjusted to 1.21 g/mL using solid KBr. The HDL solution was then transferred into an ultracentrifuge tube (Beckman) and overlaid with an equal volume of KBr solution of the same density. The sample was then subjected to ultracentrifugation at 50 000 revolutions per minute for 24 hours in a 50.3 Ti rotor (Beckman) at 10°C. The bottom 1.5 mL was collected by tube slicing and dialyzed extensively against 0.01 mol/L NH₄HCO₃. The apoA-I was stored sterile in solution for experimental use. No resolubilization was necessary. Purity of the isolated apoA-I was determined by SDS polyacrylamide gradient gel electrophoresis followed by Coomassie Blue and silver staining. A single band corresponding to apoA-I was detected in all cases. In addition, apoA-I did not cross-react to antibodies to any known apoprotein except to anti–apoA-I by double immunodiffusion.

Cell Culture and Lipid Labeling
Normal human fibroblasts GM 3468A (Coriel Institute) from passages 10 to 20 were grown in T75 plastic flasks in DMEM containing 10% fetal bovine serum (Life Science Technologies) at 37°C in a humidified incubator equilibrated with 5% CO₂ and 95% room air. Before the experiments, one million cells were trypsinized and seeded into 35-mm Petri dishes. The mean cell protein per dish was 160.1±14.7 µg.

Radioactive cholesterol [4-¹⁴C] (Amersham) was purified by thin-layer chromatography. After purification, cholesterol (161 µg) was dissolved in chloroform and dried in a stream of nitrogen. Thereafter, cholesterol was resuspended by addition of Molecusol HPB-20, 50 µL of 45% hydroxy propyl-cyclodextran (Molecusol HPB-20, Pharmatec Inc) in medium (DMEM, Gibco).¹⁵ Under these conditions, the cyclodextran solution is saturated with cholesterol and, therefore, cannot act as a net cholesterol acceptor.⁷ After stirring overnight, the Molecusol-cholesterol solution was sterile-filtered before addition to fibroblasts. Twenty-two microcuries of ¹⁴C-cholesterol was mixed with DMEM containing 10% fetal bovine serum, added to each 35-mm Petri dish of subconfluent fibroblasts, and incubated for 48 hours. After incubation, the labeled cells were washed four times with D-PBS and incubated with BioRich medium (ICN) for 24 hours before incubation with apoA-I. On the day of the experiment, the cells were washed three times with D-PBS, apoA-I (approximately 500 µg/mL) was diluted to 10 µg/mL with BioRich medium, and 1 mL was incubated with cells.

For phospholipid labeling, subconfluent fibroblasts in each 35-mm Petri dish were incubated for 72 hours with 100 µCi of ³²P-orthophosphate mixed with DMEM containing 10% fetal bovine serum. Labeled cells were washed four times with D-PBS before incubation with apoA-I.

Other Methods
Separation of lipoprotein subpopulations by the technique of 2DE, agarose electrophoresis (1D) in the first dimension, and nondenaturing PAGE in the second dimension (2D) has been previously described.¹⁰ It should be noted that the agarose did not contain albumin, permitting the separation of and pre-–migrating particles.^10,12 After transfer of lipoproteins to nitrocellulose membranes, apoA-I–containing lipoproteins were immunolocalized with monospecific goat anti-human apoA-I. Goat antibody was then immunodetected with ¹²⁵I-labeled anti-goat globulin. The radioactivity was quantified on the phosphorimaging device (Molecular Dynamics). For both the 1D and 2D analyses, 15 µL of unconcentrated medium was used. We defined the apoA-I–containing lipoproteins by their electrophoretic mobility on agarose gels as pre-ß, , and pre-.¹⁰ The size of the lipoproteins was then defined by their mobility by 2DE, compared with known molecular weight standards (Pharmacia HMW standard consisting of: thyroglobulin, 669 kDa; ferritin, 440 kDa; catalase, 232 kDa; lactate dehydrogenase 140 kDa; and albumin, 67 kDa) that were run simultaneously in the same gel. In some experiments designated herein as 2DE (3% to 16%), the 2DE was run with a 3% to 16% linear gradient gel. Otherwise, the gels have a 3% to 35% concave gradient, designated 2DE (3% to 35%). Molecular weight standards were not used in the 2DE (3% to 16%) because electrophoresis was not carried to completion, and migration of the proteins did not represent true molecular weight. As a control, purified apoA-I and human plasma pre-ß segment obtained from 1D were used as comparative standards.

Sequential colocalization of lipid label and apoA-I was carried out in a two-step process. After incubation of apoA-I with lipid-labeled cells, the lipoproteins in the medium were separated by either 1DE or 2DE. The lipoproteins were then transferred to nitrocellulose membrane and imaged by the phosphorimaging device. After imaging, the nitrocellulose membrane was washed with PBS (0.01 mol/L sodium phosphate, pH 7.4, 0.145 mol/L NaCl) containing 0.05% Tween 20. This effectively stripped the membrane of any lipids associated with the lipoproteins. The nitrocellulose membrane was then processed for immunolocalization as described above.

Medium content of free cholesterol was determined by gas-liquid chromatography using a Hewlett-Packard Model 5890 Gas Chromatography with automatic on-column sample injection (Hewlett-Packard) equipped with a DB17 capillary column (J&B Scientific). Injector temperature was 100°C. Temperature programming was as follows: 100°C for 5 minutes, then 30°C per minute until 280°C, and holding at 280°C for 9 minutes. Helium carrier pressure was 21.5 pounds per square inch at a constant flow of 40 mL/min. A known amount of internal standard (ß-stigmasterol) was added to the medium before the lipid was extracted for cholesterol determination. We assumed that the cholesterol specific radioactivity was constant over 24 hours. At times earlier than 24 hours, medium was collected, lipid extracted, and counted by liquid-scintillation counting. At the end of the experiment (24 hours), 0.5 mL of medium was extracted for cholesterol determination by GLC. An aliquot was also counted by liquid-scintillation counting to determine specific radioactivity at 24 hours. The concentration of cholesterol was then calculated from the radioactivity at each time point divided by the specific radioactivity at 24 hours.

Protein determination was completed according to the method of Lowry et al.¹⁶ Iodination of secondary antibodies was as previously described.¹⁷ For some experiments, iodinated apoA-I was used. The iodinated apoA-I was prepared from iodinated HDL using the method of MacFarlane,¹⁸ and purified apoA-I was prepared as described in the apoA-I purification section. HDL was chosen to protect from denaturation of apoA-I during iodination. Ultracentrifugation was carried out by the method of Havel et al.¹⁹

	Results

Top Abstract Introduction Methods Results Discussion References

 
To study the rate of cellular cholesterol efflux, fibroblasts were labeled with radioactive cholesterol and then 10 µg/mL purified apoA-I was added to the medium. Medium was incubated with fibroblasts for various times. Cholesterol appearance in the medium as a function of time was determined by liquid-scintillation counting divided by the cholesterol specific radioactivity at the end of the experiment. The result, summarized in Fig 1, suggested that the rate of cholesterol efflux was linear over the 24-hour incubation period (r=.99, P<.0001).

View larger version (18K): [in this window] [in a new window]   Figure 1. The rate of cholesterol efflux from fibroblasts in the presence or absence of apoA-I. The rate of cholesterol efflux was determined when purified apoA-I was used as a cholesterol acceptor. The cells were labeled with 3H-cholesterol as described in the "Methods" section. Purified apoA-I in medium was added to the dish, and samples were sequentially drawn; the radioactive lipid was then extracted and counted. As a control, medium without apoA-I was added to another dish and processed the same as medium with apoA-I. At the end of the incubation, lipid was extracted from the medium, and the cholesterol specific radioactivity was determined. From the cholesterol-specific radioactivity, the medium cholesterol quantity was calculated, assuming the specific radioactivity of cholesterol to be constant over this period. The rate of cholesterol desorption appears linear for 24 hours (r=.99, P<.0001).

We also determined the distribution of effluxed cholesterol on 1D agarose gel. In this experiment, 10 µg/mL apoA-I was used and ¹⁴C-cholesterol was substituted for ³H-cholesterol to allow detection by the phosphorimaging device. The result, presented in Fig 2, showed a steady increase of radioactivity from 4 to 24 hours in the pre- region. It should be noted that there was not any radioactivity in the pre-ß region. We also immunolocalized apoA-I on 1D gel (Fig 2). Incubation of apoA-I with medium, without cells, resulted in pre-ß mobility in 1D. When 10 µg apoA-I was incubated with fibroblasts, both pre- and pre-ß (free) apoA-I were found in the medium. The appearance of apoA-I with pre- mobility increased with the time of incubation from 4 to 24 hours.

View larger version (45K): [in this window] [in a new window]   Figure 2. The formation of pre- particles as a function of time. 14C-cholesterol–labeled cells were incubated with 10 µg/mL apoA-I. Samples were drawn at 4 hours, 12 hours, and 24 hours. Medium was subjected to 1D agarose gel electrophoresis. The lipoproteins were then transferred to nitrocellulose membrane, and the radioactivity was detected using a phosphorimaging device. After localization of 14C-cholesterol on the nitrocellulose membrane, the membrane was processed for apoA-I immunolocalization as described in the "Methods" section. This included removal of lipid from membrane with detergents. The apoA-I (125I) signal was at least 10x higher than the 14C radioactivity. For every time point, the 14C radioactivity is on the left, and the apoA-I immunodetected radioactivity is on the right. It should be noted that nonincubated apoA-I represents apoA-I stored at 4°C and had only pre-ß mobility. Taking 14C radioactivity in the pre- region at 24 hours to be 100%, the radioactivity at the pre- region at 12 hours and 4 hours was 47% and 6%, respectively.

To determine if the formation of pre- particles was dependent on apoA-I concentration, 1 to 20 µg/mL of ¹²⁵I-labeled apoA-I was used and analyzed by 1D. The result is summarized in Fig 3. At concentrations of 1 to 8 µg/mL, nearly all the purified apoA-I was converted to pre-–migrating particles. Above this concentration, the pre- formation approaches saturation. At 10 and 20 µg/mL, 80% and 50% of the pre-ß apoA-I was converted to pre-, respectively. The formation of pre- particles appeared to be dependent on the concentration of apoA-I present.

View larger version (17K): [in this window] [in a new window]   Figure 3. Pre- particle formation measured as a function of apoA-I concentration. Various concentrations of free apoA-I were incubated with fibroblasts for 24 hours. Medium was then subjected to 1D agarose gel electrophoresis and then immunodetected for apoA-I. These data showed that formation of pre- particles depended on the amount of apoA-I in medium.

Products of purified apoA-I incubations with fibroblasts were also analyzed on 2DE (3% to 16%) gels. The result of the 24-hour incubation of 10 µg/mL apoA-I in medium is shown in Fig 4. When apoA-I was incubated in medium without fibroblasts (Fig 4a), only a pre-ß mobility particle was found both in the 1DE (top insert) and 2DE (3% to 16%). Mobility of this particle in the second dimension was similar to that of purified apoA-I (standard 2) as well as that of free apoA-I–like particle of human plasma (standard 1) obtained from the pre-ß region of the 1D gel. When apoA-I was incubated in medium containing fibroblasts (Fig 4b), pre-ß and pre- mobility particles appeared in 1D and 2D as well. The mobility of pre-ß was similar to the "free apoA-I." In 2D, five distinct pre- particles of different sizes were detected. When no apoA-I was present in the medium, cholesterol was effluxed from the cells at a lower rate (Fig 1), but there were no distinct spots when the medium was analyzed by 2DE. Also, there were no distinct bands when the medium was analyzed by 1DE (data not shown).

View larger version (52K): [in this window] [in a new window]   Figure 4. Immunoblot of 2D gel electrophoresis (3% to 16%) of apoA-I. a, ApoA-I (10 µg/mL) in medium was incubated without fibroblasts. Standard 1 (st. 1) is obtained by cutting out the pre-ß region of 1D gel of plasma and electrophoresing in 2D in the same gel as the samples. B, ApoA-I was incubated with fibroblasts for 24 hours. Standard 2 (st. 2) is 10 µg/mL purified apoA-I electrophoresed in 2D. Note that mobility of purified apoA-I is the same as apoA-I incubated in the absence of fibroblasts for 24 hours. Plasma pre-ß1 particles migrate slower than purified apoA-I (compare st. 1 with st. 2). Insert on top represents 1D.

To demonstrate that pre- particles contained lipid, we incubated purified apoA-I with cells that were labeled with radioactive cholesterol (¹⁴C). To test whether labeled cholesterol would appear both in the pre-ß and pre- region, we incubated 30 µg/mL apoA-I with fibroblasts. The results are shown in Fig 5 (middle panel). Immunorecognition with anti–apoA-I demonstrated the presence of five particles with pre- mobility (Fig 5, left panel). Labeled cholesterol was found in four of the pre- particles but not in the smallest-sized particle. Importantly, there was no radioactivity in the pre-ß region.

View larger version (57K): [in this window] [in a new window]   Figure 5. Colocalization of cholesterol, phospholipid, and apoA-I. Combined detection of phospholipid, cholesterol, and immunoreactive apoA-I by 2DE. Details of the experiment are described in the "Results" section. For this experiment, 30 µg/mL apoA-I was used. ApoA-I immunodetection showed in addition to the five pre- particles, the presence of pre-ß1–like and free apoA-I particles in medium. The middle panel shows the 14C-cholesterol radioactivity, and the right panel shows the 32P-phospholipid radioactivity. The inserts on the top show location of radioactivity on agarose gel.

To examine the possibility that this small particle that did not contain cholesterol might contain phospholipid, fibroblasts were labeled with ³²P-orthophosphate. The smallest particle in the pre- region was detected with ³²P labeling in addition to four higher-molecular-weight particles (Fig 5, right panel). Immunorecognition by anti–apoA-I showed the presence of pre-ß particles in the medium (Fig 5, left panel). To further confirm that the radioactive counts were those of phospholipids, the region of pre- was extracted with chloroform:methanol. Radioactive counts were lipid-extractable and co-migrated on thin-layer chromatography with plasma phospholipids.

To understand the nature of the pre- particles, their molecular masses were determined by 2DE (3% to 35%). The results are summarized in the Table. The molecular weight of the five pre- particles ranged from 114 to 684 kDa. All particles contained cholesterol, phospholipid, and apoA-I except the smallest particle, which contained only phospholipid and apoA-I (Fig 5). The phospholipid-apoA-I particle had a molecular weight of 114 kDa. The higher-molecular-weight pre- particles appeared to be polymers of the lowest-molecular-weight phospholipid-apoA-I particle.

View this table: [in this window] [in a new window]   Table 1. Particle Molecular Weight of Pre-–Containing Lipoproteins

To determine if the pre- particles were lipoproteins, we incubated 5 µg/mL ¹²⁵I-labeled apoA-I with fibroblasts. This concentration was chosen to maximize formation of pre- particles and minimize the presence of pre-ß particles. At the end of incubation, medium density was adjusted to d=1.25 g/mL and transferred to an ultracentrifuge tube. A potassium bromide solution of d=1.21 g/mL was overlaid on the medium, and the solution ultracentrifuged. The resultant d<1.21 g/mL top and d>1.21 g/mL bottom fractions were collected and analyzed by 2DE (3% to 35%). The results showed that 85% of the radioactivity was in the d<1.21 g/mL fraction, and 15% was in the infranatant. The top 85% contained all five of pre- particles (Fig 6), whereas the bottom 15% contained some of the phospholipid-apoA-I–only particle. SDS PAGE of the pre- lipoproteins showed only apoA-I (data not shown).

View larger version (65K): [in this window] [in a new window]   Figure 6. 2DE (3% to 35%) of the d<1.21 and d>1.21 g/mL fraction of conditioned medium. 125I-labeled apoA-I was incubated with fibroblasts at 5 µg/mL concentration. At the end of incubation, medium was adjusted to 1.24 g/mL with solid KBr, and d=1.21g/mL KBr solution was overlaid onto the medium. Medium was ultracentrifuged, and the d<1.21 g/mL and d>1.21 g/mL fractions were collected. After dialysis, the fractions were subjected to 2DE. The band on the extreme left of the figure is the 125I-labeled molecular weight markers (Pharmacia HMW standards consisting of: thyroglobulin, ferritin, catalase, lactate dehydrogenase, and albumin; see "Methods" section). About 15% of radioactivity was in d>1.21 g/mL fraction, whereas 85% of radioactivity was in d<1.21 g/mL fraction. This figure should not be taken to represent actual radioactive distribution because different amounts of sample were applied.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Cellular cholesterol efflux occurs in the interstitial space. Lipoproteins and apoproteins must penetrate the interstitial space before reverse cholesterol transport can occur. Capillary endothelia sieve plasma proteins, allowing particles to penetrate the interstitial space in varying degrees, depending on their size. Lymph particles derived from plasma proteins are proportionally enriched in smaller particles.²⁰ For this reason, interstitial fluid contains fewer LDL particles than plasma (about 8%). In the prenodal peripheral lymph (an accepted model of interstitial fluid), apoA-I concentration is only 10% to 15% that of plasma.²⁰ ApoA-I–containing particles are heterogeneous, having several subpopulations that differ in size and charge both in plasma and lymph.^9,10,12 Fielding⁹ postulated that a specific subpopulation of HDL, the pre-ß₁ particles, participates in the initial stages of reverse cholesterol transport. Based on the properties of the interstitial space, we would expect that the lower-molecular-weight particles (pre-ß₁ particles) would cross the interstitial space much more readily than the larger particles.

The concentration of pre- particles in lymph greatly exceeded the pre- particles of plasma.^10,21 Because of the larger size of pre- particles, we expected a much lower concentration in the lymph than we observed. The higher-than-expected pre- concentration in lymph suggests generation of pre- particles in the interstitial space. The higher concentration of pre-ß₁ particles in lymph than in plasma^10,21 suggests that at least some of lymph pre-ß₁ particles are generated outside the plasma compartment.

Recently, we have demonstrated the presence of free apoA-I–like particles with pre-ß mobility, which differ from pre-ß₁ particles,¹¹ in plasma. Pre-ß particles may be generated either through lipolysis²² or through the action of cholesterol ester transfer protein on HDL.²³ It should also be noted that although free apoA-I has pre-ß mobility on 1D, not all pre-ß–migrating particles are "free apoA-I."¹¹ For example, the majority of dog lymph pre-ß particles have a molecular weight of 45 kDa²¹ and contain lipid.

The estimated molecular weight of free apoA-I–like particles is similar to purified apoA-I (Fig 4). The relatively small molecular size of these particles would increase their proportion passing into the interstitial space as well as their rate of diffusion within the interstitial space. We postulated that the interaction of free apoA-I with peripheral cells could result in cellular cholesterol efflux with the formation of pre- and pre-ß particles. In an attempt to understand this phenomenon, we studied lipoprotein formation in a tissue culture model in which all the components could be defined. We used our 1DE and 2DE methods, which allow clear differentiation between purified or free apoA-I–like particles and pre-ß₁ particles. Using this approach, we investigated the effect of purified apoA-I particles on cellular cholesterol efflux in a tissue culture model. In this article, we have shown that during incubation of apoA-I with fibroblasts, lipoproteins are formed with pre- mobilities.

It is estimated that the concentration of free apoA-I like particles in plasma is between 10 to 20 µg/mL. As shown in Fig 3, apoA-I concentration in the incubation medium at 8 µg/mL is almost maximal for pre- particle formation. Our observation is consistent with those of other investigators who found maximal cholesterol efflux by macrophages,²⁴ CHO cells,¹⁴ and fibroblasts²⁵ to be about 10 µg/mL apoA-I; thus, it is conceivable that some pre- particles found in lymph could be formed from "free apoA-I" pre-ß–migrating particles through their interaction with peripheral cells such as fibroblasts. Recently, Clay and Barter²⁶ have shown that lipoproteins could also be formed in vitro through the interaction of free apoA-I and lipolysis products.

The most important observation reported in this article is that the newly formed cholesterol–apoA-I complex has pre- mobilities (Figs 2 and 4-6). We have shown that this pre- particle can be separated on 2DE into five apoA-I–containing particles with distinctly different sizes. Cholesterol, phospholipid, and apoA-I colocalized with one another in the four larger particles. However, the smallest (114 kDa) particle contained only phospholipid and apoA-I. We have shown that these particles are lipoproteins (Fig 6). After ultracentrifugation, the four larger particles and a fraction of the smallest particles were recovered in the d<1.21 g/mL fraction. Some of the smallest pre- particles were also recovered in the d>1.21 g/mL fraction.

The presence of a phospholipid–apoA-I particle with pre- mobility that is devoid of cholesterol is of considerable interest (Fig 5). The phospholipid–apoA-I particle is the major pre- particle. This particle may be similar to that of the 7.9-nm particle reported by Forte et al.¹⁴ Assuming that the particle contained one molecule of apoA-I, the particle would have 76% lipid and 24% protein. If the particle contained two molecules of apoA-I, it would still contain 53% lipid. If we assume an average molecular weight of 645 for a phospholipid, the particle will have between 60 to 100 molecules of phospholipid. This amount of lipid is sufficient for it to behave as a lipoprotein. Indeed, our experiment showed that the smallest pre- particles were recovered both in the d<1.21 g/mL and the d>1.21 g/mL fractions (Fig 6). That this particle could be recovered in the d>1.21 g/mL fraction may be caused by ultracentrifugation. We have previously shown that some pre- and particles could be recovered from the d>1.21 g/mL bottom.¹⁰ The molecular weight of the other pre- particles suggests that they may be polymers of the 114 000 molecular weight particle. It is possible that this particle may be a precursor particle for accepting cholesterol. Forte et al¹⁴ reported a 7.9-nm particle that was not a good cholesterol acceptor, and it is possible that our phospholipid–apoA-I–only particle could be the 7.9-nm particle. Thus, the exact function of this particle requires further study.

Although we have provided evidence that pre- particles could be formed from free apoA-I, we have not clarified whether different pre- particles are precursors of one another. Two hypotheses are possible: (1) pre- particles are formed spontaneously and are independent of lower-molecular-weight pre- particles; or (2) lower-molecular-weight pre- particles are precursors of the higher-molecular-weight pre- particles. Evidence in support of the first hypothesis is our observation that at the earliest time when pre- particles are observed, all pre- particles have already formed.

The observation that the interaction of purified apoA-I with effluxed cholesterol and phospholipids led to the formation of lipoproteins with pre- mobilities may provide an explanation for our previous observation of increased pre- lipoproteins in lymph.¹² We postulate that as a result of passage of plasma apoA-I–containing particles through the capillary endothelium, the particles will be exposed to lipoprotein lipases and phospholipases. Interaction will increase the number of free apoA-I particles at the site of filtration. When these free apoA-I–like particles cross the endothelium and interact with peripheral interstitial cells, the effluxed cholesterol and phospholipid will transform the free apoA-I particles into pre- particles. It is postulated that this process is responsible for the increase in pre- particles in lymph. To understand the subsequent fate of these newly formed interstitial fluid pre- particles after they have returned to the plasma compartment would be an important avenue of future research.

In variance to other investigators, we were unable to find significant amounts of lipid associated with apoA-I in the pre-ß region. Castro and Fielding²⁷ and von Eckardstein et al²⁸ have both shown radioactive cholesterol associated with pre-ß₁–migrating lipoproteins when plasma was incubated with radioactive cholesterol–labeled cells for 1 to 2 minutes. We could not rule out the possibility that the amount of lipid incorporated into pre-ß₁ particles at 1- to 2-minute incubations was too low to be detected. Indeed, the first detectable amount of pre- particles appears at 4 hours, ample time to transfer radioactivity from pre-ß₁ to pre- particles. It should also be emphasized that both Castro and Fielding²⁷ and von Eckardstein et al²⁸ used whole plasma in their experiments. Under their conditions, approximately 10% of total medium cholesterol radioactivity was in the pre-ß₁ region at the end of 1 minute; the majority of the remaining 90% occurred in the region.²⁸

Hara and Yokoyama²⁴ used purified apoA-I on radioactive cholesterol–labeled macrophages and found radioactive cholesterol in pre-ß lipoproteins. In contrast, Bielicki et al²⁵ have found lipid labeled in near -migrating apoA-I particles. The difference may be attributed either to use of different cell lines (macrophages that secrete apoE versus fibroblasts that do not) or to differences in apoA-I purification.²⁵ Our observation is more consistent with that of Bielicki et al.²⁵ Difference in electrophoretic mobility could also be attributed to the method of 1D. Bielicki et al²⁵ used albumin in their agarose electrophoresis. We have shown previously¹⁰ that using the method of Noble²⁹ for agarose gel electrophoresis caused movement of pre-–migrating particles to the region.

In conclusion, we have shown that incubation of purified apoA-I with fibroblasts resulted in formation of lipoproteins with pre- mobility. These newly formed lipoproteins consist of five distinct particles ranging in molecular weight between 114 and 684 kDa. The particle of smallest molecular weight was an apoA-I–only phospholipid particle. The other four particles contained apoA-I, cholesterol, and phospholipids.

	Selected Abbreviations and Acronyms

 

apo = apolipoprotein DMEM = Dulbecco's modified Eagle's medium 2DE = two-dimensional electrophoresis PAGE = polyacrylamide gel electrophoresis

	Acknowledgments

 
We wish to thank our colleagues Drs Howard Eder, C.H. Sloop, and J.J. Thompson for their many suggestions and discussions; Dr Theda Foster for her editorial assistance; and Katalin Horvath and Colleen Tierney for their expert technical assistance. Wenwu Zhang is supported by a student fellowship from the American Heart Association Louisiana Affiliate. This research was supported by NIH program project grant HL-25596 and NIH grant HL56160.

	Footnotes

 
Drs Asztalos and Zhang contributed equally to the study.

Received October 17, 1996; accepted January 9, 1997.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Johnson WJ, Mahlberg FH, Rothblat GH, Phillips MC. Cholesterol transport between cells and high-density lipoproteins. Biochim Biophys Acta. 1991;1085:273-298.[Medline] [Order article via Infotrieve]
2. Stein O, Stein Y, Lefevre M, Roheim PS. The role of apolipoprotein A-IV in reverse cholesterol transport studied with cultured cells and liposomes derived from an ether analog of phosphatidylcholine. Biochim Biophys Acta. 1986;878:7-13.[Medline] [Order article via Infotrieve]
3. Bernard DW, Rodriguez A, Rothblat GH, Glick JM. Influence of high density lipoprotein on esterified cholesterol stores in macrophages and hepatoma cells. Arterioscler Thromb. 1990;10:135-144.[Abstract/Free Full Text]
4. Davidson WS, Rogrigueza WV, Lund-Katz S, Johnson WJ, Rothblat GH, Phillips MC. Effects of acceptor particle size on the efflux of cellular free cholesterol. J Biol Chem. 1995;270:17106-17113.[Abstract/Free Full Text]
5. Rothblat GH, Phillips MC. Mechanism of cholesterol efflux from cells. Effects of acceptor structure and concentration. J Biol Chem. 1982;257:4775-4782.[Free Full Text]
6. DeLamatre J, Wolfbauer G, Phillips MC, Rothblat GH. Role of apolipoproteins in cellular cholesterol efflux. Biochim Biophys Acta. 1986;875:419-428.[Medline] [Order article via Infotrieve]
7. Kilsdonk EPC, Yancey PG, Stoudt GW, Bangerter FW, Johnson WJ, Phillips MC, Rothblat GH. Cellular cholesterol efflux mediated by cyclodextrins. J Biol Chem. 1995;270:17250-17256.[Abstract/Free Full Text]
8. Barter PJ, Rye K-A. Molecular mechanisms of reverse cholesterol transport. Curr Opin Lipidol. 1996;7:69-76.[Medline] [Order article via Infotrieve]
9. Fielding CJ, Fielding PE. Molecular physiology of reverse cholesterol transport. J Lipid Res. 1995;36:211-228.[Abstract]
10. Asztalos BF, Sloop CH, Wong L, Roheim PS. Two-dimensional electrophoresis of plasma lipoproteins: recognition of new apoA-I containing subpopulations. Biochim Biophys Acta. 1993;1169:291-300.[Medline] [Order article via Infotrieve]
11. Asztalos B, Roheim PS. Presence and formation of `Free Apolipoprotein A-I-like' particles in human plasma. Arterioscler Thromb. 1995;15:1419-1423.[Abstract/Free Full Text]
12. Asztalos BF, Sloop CH, Wong L, Roheim PS. Comparison of apoA-I-containing subpopulations of dog plasma and prenodal peripheral lymph: evidence for alteration in subpopulations in the interstitial space. Biochim Biophys Acta. 1993;1169:301-304.[Medline] [Order article via Infotrieve]
13. Nichols AV, Gong EL, Blanche PJ, Forte TM, Anderson DW. Effects of guanidine hydrochloride on human plasma high density lipoproteins. Biochim Biophys Acta. 1976;446:226-239.[Medline] [Order article via Infotrieve]
14. Forte TM, Bielicki JK, Goth-Goldstein R, Selmek J, McCall MR. Recruitment of cell phospholipids and cholesterol by apolipoproteins A-II and A-I: formation of nascent apolipoprotein-specific HDL that differ in size, phospholipid composition, and reactivity with LCAT. J Lipid Res. 1995;36:148-157.[Abstract]
15. Caprio JD, Yun J, Javitt NB. Bile acid and sterol solubilization in 2-hydroxypropyl-ß-cyclodextrin. J Lipid Res. 1992;33:441-443.[Abstract]
16. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275.[Free Full Text]
17. Wong L, Torbati A. Differentiation of intrahepatic membrane-bound and secretory apolipoprotein B by monoclonal antibodies-membrane-bound apolipoprotein B is more glycosylated. Biochemistry. 1994;33:1923-1929.[Medline] [Order article via Infotrieve]
18. McFarlane AS. Efficient trace-labeling of proteins with iodine. Nature. 1958;182:53.[Medline] [Order article via Infotrieve]
19. Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345-1353.
20. Sloop CH, Dory L, Roheim PS. Interstitial fluid lipoproteins. J Lipid Res. 1987;28:225-237.[Abstract]
21. Wong L, Sivok B, Kurucz E, Sloop CH, Roheim PS, Asztalos B. Lipid composition of HDL subfractions in dog plasma and lymph. Arterioscler Thromb. 1995;15:1875-1881.[Abstract/Free Full Text]
22. Barrans A, Collet X, Barbaras R, Jaspard B, Manent J, Vieu C, Chap H, Perret B. Hepatic lipase induces the formation of pre-ß1 high density lipoprotein (HDL) from triacylglycerol-rich HDL2. J Biol Chem. 1994;269:11572-11577.[Abstract/Free Full Text]
23. Liang HQ, Rye K-A, Parter PJ. Dissociation of lipid-free apolipoprotein A-I from high density lipoproteins. J Lipid Res. 1994;35:1187-1199.[Abstract]
24. Hara H, Yokoyama S. Interaction of free apolipoproteins with macrophages—formation of high density lipoprotein-like lipoproteins and reduction of cellular cholesterol. J Biol Chem. 1991;266:3080-3086.[Abstract/Free Full Text]
25. Bielicki JK, Johnson WJ, Weinberg RB, Glick JM, Rothblat GH. Efflux of lipid from fibroblasts to apolipoproteins-dependence on elevated levels of cellular unesterified cholesterol. J Lipid Res. 1992;33:1699-1709.[Abstract]
26. Clay MA, Barter PJ. Formation of new HDL particles from lipid-free apolipoprotein A-I. J Lipid Res. 1996;37:1722-1731.[Abstract]
27. Castro GR, Fielding CJ. Early incorporation of cell derived cholesterol into pre-ß-migrating HDL. Biochemistry. 1988;27:25-29.[Medline] [Order article via Infotrieve]
28. von Eckardstein A, Huang Y, Wu S, Sarmadi AS, Schwarz S, Steinmetz A, Assmann G. Lipoproteins containing apolipoprotein A-IV but not apolipoprotein A-I take up and esterify cell-derived cholesterol in plasma. Arterioscler Thromb. 1995;15:1755-1763.[Abstract/Free Full Text]
29. Noble RP. Electrophoretic separation of plasma lipoproteins in agarose gel. J Lipid Res. 1968;9:693-700.[Abstract]

This article has been cited by other articles:

				M. P. Adorni, F. Zimetti, J. T. Billheimer, N. Wang, D. J. Rader, M. C. Phillips, and G. H. Rothblat The roles of different pathways in the release of cholesterol from macrophages J. Lipid Res., November 1, 2007; 48(11): 2453 - 2462. [Abstract] [Full Text] [PDF]

				A. Kontush, P. Therond, A. Zerrad, M. Couturier, A. Negre-Salvayre, J. A. de Souza, S. Chantepie, and M. J. Chapman Preferential Sphingosine-1-Phosphate Enrichment and Sphingomyelin Depletion Are Key Features of Small Dense HDL3 Particles: Relevance to Antiapoptotic and Antioxidative Activities Arterioscler. Thromb. Vasc. Biol., August 1, 2007; 27(8): 1843 - 1849. [Abstract] [Full Text] [PDF]

				A. Kontush and M. J. Chapman Functionally Defective High-Density Lipoprotein: A New Therapeutic Target at the Crossroads of Dyslipidemia, Inflammation, and Atherosclerosis Pharmacol. Rev., September 1, 2006; 58(3): 342 - 374. [Abstract] [Full Text] [PDF]

				P. T. Duong, H. L. Collins, M. Nickel, S. Lund-Katz, G. H. Rothblat, and M. C. Phillips Characterization of nascent HDL particles and microparticles formed by ABCA1-mediated efflux of cellular lipids to apoA-I J. Lipid Res., April 1, 2006; 47(4): 832 - 843. [Abstract] [Full Text] [PDF]

				P. G. Yancey, B. F. Asztalos, N. Stettler, D. Piccoli, D. L. Williams, M. A. Connelly, and G. H. Rothblat SR-BI- and ABCA1-mediated cholesterol efflux to serum from patients with Alagille syndrome J. Lipid Res., September 1, 2004; 45(9): 1724 - 1732. [Abstract] [Full Text] [PDF]

				K.-A. Rye and P. J. Barter Formation and Metabolism of Prebeta-Migrating, Lipid-Poor Apolipoprotein A-I Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 421 - 428. [Abstract] [Full Text]

				M. Denis, B. Haidar, M. Marcil, M. Bouvier, L. Krimbou, and J. Genest Jr. Molecular and Cellular Physiology of Apolipoprotein A-I Lipidation by the ATP-binding Cassette Transporter A1 (ABCA1) J. Biol. Chem., February 27, 2004; 279(9): 7384 - 7394. [Abstract] [Full Text] [PDF]

				W. Zhang, P. G. Yancey, Y. R. Su, V. R. Babaev, Y. Zhang, S. Fazio, and M. F. Linton Inactivation of Macrophage Scavenger Receptor Class B Type I Promotes Atherosclerotic Lesion Development in Apolipoprotein E-Deficient Mice Circulation, November 4, 2003; 108(18): 2258 - 2263. [Abstract] [Full Text] [PDF]

				W. Safi, J. N. Maiorano, and W. S. Davidson A proteolytic method for distinguishing between lipid-free and lipid-bound apolipoprotein A-I J. Lipid Res., May 1, 2001; 42(5): 864 - 872. [Abstract] [Full Text]

				A. von Eckardstein, J.-R. Nofer, and G. Assmann High Density Lipoproteins and Arteriosclerosis : Role of Cholesterol Efflux and Reverse Cholesterol Transport Arterioscler. Thromb. Vasc. Biol., January 1, 2001; 21(1): 13 - 27. [Abstract] [Full Text] [PDF]

				D. Rees, T. Sloane, W. Jessup, R. T. Dean, and L. Kritharides Apolipoprotein A-I Stimulates Secretion of Apolipoprotein E by Foam Cell Macrophages J. Biol. Chem., September 24, 1999; 274(39): 27925 - 27933. [Abstract] [Full Text] [PDF]

				G. H. Rothblat, M. de la Llera-Moya, V. Atger, G. Kellner-Weibel, D. L. Williams, and M. C. Phillips Cell cholesterol efflux: integration of old and new observations provides new insights J. Lipid Res., May 1, 1999; 40(5): 781 - 796. [Abstract] [Full Text]

				K. L. Gillotte, W. S. Davidson, S. Lund-Katz, G. H. Rothblat, and M. C. Phillips Removal of cellular cholesterol by pre-ß-HDL involves plasma membrane microsolubilization J. Lipid Res., October 1, 1998; 39(10): 1918 - 1928. [Abstract] [Full Text]

				W. Zhang, B. Asztalos, P. S. Roheim, and L. Wong Characterization of phospholipids in pre-{alpha} HDL: selective phospholipid efflux with apolipoprotein A-I J. Lipid Res., August 1, 1998; 39(8): 1601 - 1607. [Abstract] [Full Text]

				A. J. Mendez, G. Lin, D. P. Wade, R. M. Lawn, and J. F. Oram Membrane Lipid Domains Distinct from Cholesterol/Sphingomyelin-Rich Rafts Are Involved in the ABCA1-mediated Lipid Secretory Pathway J. Biol. Chem., January 26, 2001; 276(5): 3158 - 3166. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articl