Importance of macrophage cholesterol content on the flux of cholesterol mass.

Net flux of cholesterol represents the difference between efflux and influx and can result in net cell-cholesterol accumulation, net cell-cholesterol depletion, or no change in cellular cholesterol content. We measured radiolabeled cell-cholesterol efflux and cell-cholesterol mass using cholesterol-normal and -enriched J774 and elicited mouse peritoneal macrophage cells. Net cell-cholesterol effluxes were observed when cholesterol-enriched J774 cells were incubated with 3.5% apolipoprotein (apo) B depleted human serum, HDL3, and apo A-I. Net cell-cholesterol influxes were observed when cholesterol-normal J774 cells were incubated with the same acceptors except apo A-I. When incubated with 2.5% individual sera, cholesterol mass efflux in free cholesterol (FC)-enriched J774 cells correlated with the HDL-cholesterol (HDL-C) concentrations (r2 = 0.4; P=0.003), whereas cholesterol mass influx in cholesterol-normal J774 cells correlated with the LDL cholesterol (LDL-C) concentrations (r2 = 0.6; P<0.0001) of the individual sera. A positive correlation was observed between measurements of [3H]cholesterol efflux and reductions in cholesterol mass (r2 = 0.4; P=0.001) in FC-enriched J774 cells. In conclusion, isotopic efflux measurements from cholesterol-normal or cholesterol-enriched cells provide an accurate measurement of relative ability of an acceptor to remove labeled cholesterol under a specific set of experimental conditions, i.e., efflux potential. Moreover, isotopic efflux measurements can reflect changes in cellular cholesterol mass if the donor cells are enriched with cholesterol.

scribed previously ( 13 ). Presence of PEG does not alter the effl ux capacity of the extracellular acceptors. Because of the dilution of the sera by the precipitation reagents, 3.5% of the apo B depleted serum is equivalent to 2.5% whole serum.

Cell culture
J774 cells were plated in growth medium in 24-well plates at a density of 150,000 cells/well. All the samples were assayed in triplicate. Cholesterol-normal cells were labeled for 24 h with 2 Ci/ ml [ 3 H]cholesterol in medium supplemented with 1% FBS and 2 g/ml ACAT inhibitor. Cholesterol-enriched cells were incubated with the same labeling medium plus 25 g/ml acLDL. In some experiments, the ACAT inhibitor was added to ensure pools of radiolabeled CE would not be present because this could complicate the determination of the fractional release of labeled cholesterol from the cells. After labeling plus or minus cholesterol enrichment, the cells were washed twice with MEM HEPES and equilibrated for 18 h in RPMI medium containing 0.2% BSA. In the case of enriched J774 cells without ACAT inhibitor, the equilibration period was 2 h. Following these treatments the cells were incubated with indicated exogenous acceptors for 8 h to measure the labeled FC effl ux and the cellular FC mass.
Elicited MPM cells were collected from C57Bl/6 mice and plated for 24 h to allow attachment of the MPM to the plastic. Nonadhering cells were removed and MPM cells were incubated with RPMI containing 2 Ci/ml [ 3 H]cholesterol, 1% FBS, and 50 µg/ml acLDL for 24 h. This treatment enriched the cells with both FC and CE and produced a foam cell phenotype (CE у FC). The washed monolayers were then equilibrated in RPMI medium containing 0.2% BSA for 18 h. Afterwards, the cells were incubated for 18 h with the indicated exogenous cholesterol acceptors.
The effl ux of labeled cholesterol was quantifi ed as the percentage of radiolabel in the media compared with that present in the cells prior to the incubation with cholesterol acceptors. The amount of radiolabel present in the cells was determined by extracting cell lipids with 2-propanol and measuring the [

Protein and cholesterol mass determination
At the end of the experiment, the cell monolayers were washed with DPBS and cell lipid was extracted using 2-propanol containing 5 µg/ml of cholesteryl methyl ether (CME; Sigma, St. Louis, MO) as an internal standard for gas liquid chromatography (GLC) analysis. A fraction of the extracted lipid was used to measure total cholesterol radioactivity incorporated into cellular lipids using scintillation counting. The remaining lipid was prepared for GLC to measure cholesterol mass (FC and CE) as previously described ( 14 ). The extent of loss of cell FC and CE mass was calculated based on the cholesterol content of the cells at time zero, prior to the effl ux phase. Cell protein was measured by the method described by Markwell et al. ( 15 ). Cell proteins were measured before the addition of acceptors (T0) and after the incubation period with serum or lipoproteins. There were no statistically signifi cant changes in cell protein measurement between cells at T0 and cells at the end of effl ux period.

Two-dimensional gel electrophoresis
Apo A-I containing subpopulations of HDL, particularly pre ␤ HDL, were measured using immunoblotting and image analysis after separation of the particles using nondenaturing two-dimensional gel electrophoresis as previously described ( 13 ).

Statistical analysis
All statistical analyses were performed using GraphPad Prism (San Diego, CA) software. Data were presented as mean ± SD. lesterol between cells and serum. Cholesterol mass and isotope fl ux studies were conducted on J774 (murine cell line) cells and mouse peritoneal macrophage (MPM) cells that contained both normal and elevated levels of cholesterol. The level of cell cholesterol infl uences fl ux in a number of ways, including modulating the expression level of cell proteins mediating effl ux such as ABCA1, ABCG1, and SR-BI, and proteins, such as LDL and scavenger receptors, which participate in lipoprotein uptake ( 7,10 ) or infl ux. The fractional effl ux values were determined using cells prelabeled with [ 3 H]cholesterol and net fl ux was measured directly by quantitating changes in cell cholesterol mass upon incubation with either human sera or apolipoprotein (apo)-B-depleted sera. Because a large body of data has been collected by following the movement of radiolabeled cholesterol, we compared values for isotopic effl ux to changes in cell cholesterol mass so as to determine the reliability of isotopic measurements in predicting actual mass fl ux.

Preparation of lipoproteins and human sera
Human HDL 3 and LDL were isolated by sequential ultracentrifugation from plasma obtained with approved consent from healthy, normolipemic individuals as previously described (HDL 3 , d = 1.12-1.21 g/mL and LDL, d = 1.006-1.063 g/mL) ( 11 ). The study protocol was reviewed and approved by the Institutional Review Board of The Children's Hospital of Philadelphia. To obtain acetylated low density lipoprotein (acLDL), LDL was modifi ed using acetic anhydride as previously described ( 12 ). Apo A-I was purifi ed from delipidated HDL using ethanol/diethyl ether followed by anion-exchange chromatography on a Q-Sepharose column. The apo A-I fractions were pooled, dialyzed against 5 mM NH 4 HCO 3 , lyophilized, and stored at Ϫ 20°C. The apo A-I was resolubilized in 6 M guanidine hydrochloride and dialyzed extensively against saline (0.15 M NaCl, pH 7.4) prior to incubating with the cells. The human sera used in these studies were collected, with informed consent, from a group of 22 normolipemic individuals consisting of 13 females and 9 males ranging in age from 23 to 70 years. The sera had an average HDL cholesterol (HDL-C) level of 58 ± 15 mg/dl (31-100 mg/dl), LDL cholesterol (LDL-C) of 113 ± 22 mg/dl (62-160 mg/dl) and triglycerides of 120 ± 66 mg/dl (52-294 mg/dl). The 22 human serum samples were used individually or combined to obtain a pool of serum. To isolate the serum HDL fraction, the pool of serum or individual serum samples were depleted of apo-B-containing lipoproteins by precipitating them with polyethylene glycol (PEG) solution as de-been a valuable tool in elucidating the pathways and mechanisms involved in the removal of cell cholesterol. In addition, recent studies have demonstrated a relationship between effl ux from macrophages and the deposition of lipids in vessels, as measured by intima media thickness (IMT) and angiography ( 17 ). Of prime importance with respect to understanding the process of reverse cholesterol transport (RCT) is net fl ux of cholesterol mass that occurs when cells are incubated with serum or isolated lipoproteins ( 18,19 ). In the present study, we have quantitated net cholesterol mass fl ux by directly measuring the change in cell cholesterol mass upon incubation of both cholesterol-normal and cholesterol-enriched J774 macrophages with different acceptors. Table 2 demonstrates the changes in cell cholesterol mass when normal and enriched cells were exposed to a pool of 3.5% apo-B-depleted human serum (equivalent to 2.5% serum), isolated HDL 3 (50 µg/ml), or apo A-I (25 µg/ml) for 8 h. As shown in Table 2 , incubating cholesterol-enriched cells for 8 h resulted in a signifi cant net reduction of cell cholesterol mass (net effl ux). In contrast, if the starting cells contained the level of sterol normally observed when the cells were grown in FBS, exposure to the same acceptors resulted in a small net increase in cell cholesterol mass (net infl ux). As expected, there was no infl ux of cholesterol mass when lipid-free apo A-I was present as acceptor ( Table 2 ).
Because the net movement of cholesterol mass, either in (net infl ux) or out (net effl ux) of cells, is a very relevant parameter resulting from exposure of cells to serum or lipoproteins, we determined the net cholesterol mass fl ux with both cholesterol-normal and cholesterol-enriched J774 cells upon exposure to a group of normal human sera. Net cholesterol mass fl ux values were determined directly by comparing the cell cholesterol content at the beginning of the incubation to that obtained following 8 h incubation with each serum sample. The results shown in Fig. 1 clearly illustrate that the fl ux of cell cholesterol mass is directly related to the starting cholesterol levels. With cells enriched with FC or both FC and CE there was a signifi cant depletion of total cell cholesterol mass, whereas with cholesterol-normal cells there was a signifi cant increase in total cell cholesterol mass upon exposure to 2.5% human serum.
As illustrated in

Relationship between effl ux of labeled cholesterol and reduction of cell cholesterol mass in J774 cells
As discussed earlier, the great majority of studies on cell cholesterol fl ux have followed the release of radiolabeled Statistical signifi cance was determined by unpaired t -tests. Pearson correlation (r 2 ) was used to assess the correlation between the serum components and the % effl ux of radiolabeled cholesterol or cholesterol mass. Signifi cance was assessed at P р 0.05.

RESULTS
The fl ux of cell cholesterol is not only linked to the type and concentration of lipoproteins in serum but it is also a function of the array of transport proteins expressed by the cells. Although many manipulations can be done with cells in culture that will infl uence the expression level of transport proteins, we have compared cells prepared under two common growth conditions: cholesterol-normal (grown in the presence of FBS, which yields cells with normal levels of cholesterol) and cholesterol-enriched (grown in the presence of acLDL, which yields cells with excess cholesterol). For both conditions, we grew cells in the presence or absence of an ACAT inhibitor. Thus, in the absence of ACAT inhibitor, cholesterol accumulated as FC and CE, whereas in the presence of this inhibitor only the FC pool was expanded and there was no deposition of cellular CE. The expression level of effl ux proteins is infl uenced by the cholesterol content of the cells. In cholesterol-normal J774 and MPM cells there are low levels of SR-BI, ABCA1, and ABCG1; however, enrichment of the cells with cholesterol produces an increase in the expression of both ABCA1 and ABCG1 together with a decrease in SR-BI ( 16 ). In addition to these pathways, our previous studies have demonstrated that the aqueous transfer pathway plays a large role in cholesterol effl ux from cholesterol-normal cells ( 16 ).

Correlation between cholesterol fl ux and serum components
In our initial studies, we examined the correlation between fractional effl ux and serum components ( Table 1 ). The correlations were obtained using J774 cells enriched with either FC or both FC and CE. A comparison of the correlation between percent cholesterol fl ux and serum components determined for radiolabeled cholesterol effl ux as well as cholesterol mass effl ux are presented in Table 1 . Although there is some similarity in correlations between the fl ux of either isotope or mass and serum components, the correlation patterns are not identical. It is probable that this is a refl ection of the fact that mass changes refl ect both the release of cell cholesterol and the uptake of lipoprotein cholesterol. At present, there is no information available on the effi ciency of HDL subfractions in delivering cholesterol to cells. It will require the isolation and testing of individual subfractions to obtain such information. Even though the correlation coeffi cient of some HDL subfractions against percent effl ux of radiolabeled cholesterol or cholesterol mass seems to be low, together, the HDL fraction contributes around 70-75% of the total labeled cholesterol effl ux of whole serum in J774 cells.

Net fl ux of cholesterol mass from J774 cells
The measurement of the effi ciency of serum or isolated lipoproteins to mediate cell cholesterol effl ux has

Net fl ux of cholesterol mass from MPM cells
To demonstrate that the results obtained with J774 cells were not unique to this cell line, we measured net cholesterol mass and label fl ux using MPM grown in media containing 50 µg/ml of acLDL in the absence of an ACAT inhibitor (FC-and CE-enriched). Similar to cholesterolenriched J774 cells, there was a signifi cant reduction of total cell cholesterol mass (net effl ux) when FC-and CEenriched MPM cells (169.4 ± 19.4 µg total cellular cholesterol/mg cell protein, with 60% ± 1% CE) were incubated with 3.5% apo-B-depleted pooled human serum, 50 µg/ml of isolated HDL 3 , or 50 µg/ml of apo A-I for 18 h ( Fig. 4 ). There was a signifi cant correlation between the fractional loss of cholesterol mass and the change in radiolabeled cholesterol (r 2 = 0.98, P < 0.001). The fractional loss was 23% when apo-B-depleted pooled serum was used and was similar for all of the cell cholesterol pools (TC, FC, and CE) because of their similar specifi c activity after 18 h of incubation. If an ACAT inhibitor was added together with the 3.5% apo-B-depleted pooled serum, the loss of total cell cholesterol mass was increased (data not shown).
cholesterol from the cells because the protocols for this type of experiment are straightforward and readily adapted to high-throughput assays ( 20 ). Using the data collected in the present study, we established the relationship between the fractional effl ux of labeled cholesterol to the net reduction of cholesterol mass in J774 macrophages enriched with FC. As illustrated in Fig. 3 , when the cells were enriched with FC and incubated with a group of 2.5% individual sera, there was a statistically signifi cant positive correlation between the fraction of labeled cholesterol that was released and the percent reduction in cell cholesterol mass (r 2 = 0.4, P = 0.001). There was also a good correlation between fractional effl ux of labeled cholesterol and percent decrease in cell cholesterol mass (r 2 = 0.4, P = 0.001) when 3.5% apo-B-depleted sera were used. However, the reductions in cell cholesterol mass upon exposure to the apo-Bdepleted sera were greater than observed with whole sera, a result that can be attributed to the reduced infl ux of exogenous cholesterol upon removal of apo-Bcontaining lipoproteins. J774 cells were enriched with FC or FC and CE by exposure to acLDL with or without the presence of ACAT inhibitor respectively. The cell cholesterol mass at T0 (before the effl ux phase) was 53.1 ± 1.9 µg total cholesterol/mg cell protein for FC-enriched cells, and 89.9 ± 4.2 µg total cholesterol/mg cell protein for FC-and CE-enriched cells. The experiment has been repeated twice in triplicate and the value used for correlation is the average of six values. Pearson correlations (r 2 ) were assessed between the HDL, apo A-I, and HDL subfractions and the % radiolabeled cholesterol effl ux or % cholesterol mass effl ux. The r 2 of the statistically signifi cant correlations are presented in the table. The statistical signifi cance of the correlations were assessed at P р 0.05. NS = nonsignifi cant. J774 cells were plated in growth medium for 24 h and labeled with medium containing 1% FBS and 2 µCi/ml ± 25 µg/ml acLDL for 24 h. The cells were equilibrated for 18 h in 0.2% BSA and the effl ux was initiated by incubating the cells with 3.5% pooled apo-B-depleted human serum, HDL 3 (50 µg/ml), and apo A-I (25 µg/ml). ACAT inhibitor was added in the medium for cholesterol-normal and FC-enriched cells. The cell cholesterol mass at T0 (before the effl ux phase) was 6.4 ± 0.3 µg total cholesterol/mg cell protein for cholesterol-normal cells, 53.1 ± 1.9 µg total cholesterol/mg cell protein for FC-enriched cells, and 89.9 ± 4.2 µg total cholesterol/mg cell protein for FC-and CE-enriched cells. Values are presented as the difference in cholesterol mass between the treatment at the end of effl ux period and their corresponding T0 values. Plus sign indicates increase in total cellular cholesterol content, whereas minus sign indicates decrease in total cellular cholesterol content. Each value represents the average of triplicate. The data are given as mean ± SD. a Statistically signifi cant from their corresponding T0 cholesterol mass content.
terol effl ux from macrophages in culture to sera from human subjects can be associated with measurements of plaque size as determined by measures of IMT and angiography ( 17 ). However, because lipoproteins contain cholesterol, the incubation of cells with serum or isolated lipoproteins results in bidirectional fl ux between the cell and medium compartments. Thus, the measurement of labeled cholesterol release, or effl ux, evaluates only one arm of this bidirectional process.
In the present study, we have measured the change in cell cholesterol mass when J774 macrophages or MPM were exposed to human serum or HDL obtained from normolipemic donors. We have used both J774 cells that were grown on FBS, and thus contain low levels of cholesterol, and J774 cells that have been exposed to acLDL to enrich the macrophages with both FC and CE, or only FC if an ACAT inhibitor was present. In additional experiments, we have used MPM enriched in both FC and CE.

Importance of cell cholesterol content on cholesterol mass fl ux
A number of both cellular and lipoprotein factors determine the net fl ux of cell cholesterol mass. Effl ux is mediated by different pathways such as aqueous diffusion, SR-BI, ABCA1, and ABCG1 and the contribution of these pathways to effl ux is infl uenced by cell type and cell cholesterol status. Even with a single cell type, the cholesterol status of the cells is a major determinant of the net fl ux of cholesterol mass. This is exemplifi ed by Fig. 1, which illustrates that exposure of cholesterol-enriched J774 to 22 normolipemic human sera for 8 h results in the net depletion of cell cholesterol mass. In contrast, exposure of cholesterol-normal J774 to the same sera produced a net mass accumulation with all serum samples. The importance of cell cholesterol content on net mass fl ux is further exemplifi ed by the data shown in Table 2 , in which cholesterolnormal and -enriched J774 cells were exposed to a pool of 3.5% apo-B-depleted human serum, human HDL 3 (50 µg/ ml), and human apo A-I (25 µg/ml). The pool of 3.5% apo-B-depleted human serum produced a substantial reduction in cell cholesterol mass in cholesterol-enriched DISCUSSION Various aspects of the movement of cholesterol between cells in culture and serum or isolated serum lipoproteins have been investigated. The majority of these investigations focused on the movement of FC (effl ux) using radiolabeled cholesterol previously incorporated into the cells ( 20 ). The protocols for assaying the release of labeled cholesterol are relatively simple because only FC undergoes effl ux and the assay can be easily adapted for high-throughput use. Data generated with this approach have been very valuable in identifying a number of pathways that participate in the release of cell cholesterol and the maintenance of cell cholesterol homeostasis. Thus, the measurement of the fractional release of labeled cell cholesterol can be used to assess the effi ciency of serum or lipoproteins in stimulating effl ux and to evaluate drugs or diets for their ability to modulate effl ux, a process that is thought to be the fi rst step in RCT. Indeed, recent investigations have provided evidence that the in vitro quantitation of choles-  produces a net loss of cell cholesterol mass and the loss occurs equally from the FC and CE pools. If ACAT is blocked and an extracellular acceptor is present, the loss of total cholesterol is increased due to the inhibition of cholesterol reesterifi cation and the accumulation and subsequent effl ux of the newly generated FC (data not shown). If no acceptor is present, there is relatively little change in cell total cholesterol during an 18 h incubation; however, introduction of the ACAT inhibitor during the effl ux phase shifts cholesterol mass from the CE pool into the FC pool without producing an appreciable change in the total cholesterol mass.

Relationship between percent labeled cholesterol effl ux and cholesterol mass fl ux
The measurement of the release of radiolabeled cholesterol from cells to serum or lipoproteins has been extensively used to estimate the effl ux effi ciency of the acceptor particles. The extent to which the values obtained from isotopic effl ux measurements refl ect changes in the mass of cell cholesterol has not been well established. As shown in Fig. 3 , there is a statistically signifi cant association between fractional effl ux of label and percent loss of cell cholesterol mass from enriched cells exposed to either 2.5% whole human sera or the apo-B-depleted sera. Although the apo B components of serum appear to be relatively ineffi cient in stimulating effl ux, they do contribute signifi cantly to the cellular uptake of both FC and CE, thus impacting on cell cholesterol mass. The agreement between measurement of fl ux based on labeled cholesterol versus cholesterol mass in J774 ( Fig. 3 ) supports the conclusion that the use of cellular labeled cholesterol to assess the fl ux of cholesterol mass does provide reliable data on the fl ux of cell cholesterol mass if the cells are fi rst cholesterol enriched.
Similar to the correlation between the fractional effl ux of radiolabeled cholesterol and the percent loss of FC mass in J774 cells, there was a signifi cant correlation between cholesterol mass and labeled cholesterol in the MPM exposed to apo-B-depleted pooled serum. In the case of MPM, the agreement between mass and isotope effl ux was due to the fact that the specifi c activity of labeled FC and CE pools had come to an equilibrium under the conditions that we employed to enrich, label, and equilibrate the cells. If cholesterol-specifi c activity in FC and CE are not similar, isotopic release will not accurately predict mass changes. In addition, effl ux of labeled cholesterol from cholesterol-normal cells will not predict mass fl ux because the difference between infl ux and effl ux is generally not large and would not produce an appreciable change in the cholesterol mass in either the cell or media compartments. However, the measurement of release of labeled cholesterol does allow a comparison of the effi ciency of various acceptors in mediating the effl ux arm of the bidirectional fl ux.

SUMMARY
There is recent evidence that the effl ux effi ciency of a given specimen of serum as measured in vitro using cholesterol-normal J774 macrophages is a refl ection of its ability to cells, whereas there was a small net increase in cell cholesterol mass with the cholesterol-normal cells. These data are consistent with the signifi cant association between net effl ux of cholesterol mass from FC-enriched cells and serum HDL-C levels ( Fig. 2B ) and the signifi cant association of net infl ux of cholesterol mass to serum LDL-C levels seen with cholesterol-normal J774 cells ( Fig. 2A ). At the concentrations used in these experiments, both HDL 3 and apo A-I produced net mass effl ux from enriched cells ( Table 2 ). However, apo A-I had no signifi cant impact on the cell cholesterol content of cholesterol-normal cells.
If MPM containing elevated levels of both FC and CE are used for cholesterol mass studies ( Fig. 4 ), the presence of the HDL fraction of serum during an 18 h incubation Fig. 3. Correlation between fractional effl ux of labeled cholesterol and percent reduction of total cell cholesterol mass in free cholesterol-enriched J774 cells. J774 cells enriched with FC were incubated for 8 h with 2.5% human sera or 3.5% of the same sera following the removal of apo-B-containing lipoproteins. Signifi cant correlation was seen between the fraction of labeled cholesterol that was released and the fractional reduction in total cell cholesterol mass when FC-enriched J774 cells were incubated with 2.5% serum (r 2 = 0.4, P = 0.001) or apo-B-depleted serum (r 2 = 0.4, P = 0.001) for 8 h. The fractional reduction in total cell cholesterol mass was greater upon exposure to apo-B-depleted serum compared with whole serum. Each 22 value is an average of six determinations-triplicate from two experiments. reduce cell cholesterol from vessel wall cells as measured by IMT and angiography ( 17 ). The results presented here further suggest that in vitro assays of effl ux effi ciency may refl ect the in vivo potential of a given serum to promote cholesterol homeostasis. If in vitro experiments are conducted exposing cholesterol-normal cells to HDL, cholesterol exchange will occur; however, if the cells contain excess cholesterol net movement of cholesterol mass will be achieved. In vivo this would result in the removal of cholesterol mass from foam cells, thus contributing to the regression of the atherosclerotic lesion. Enriching the macrophage cells with cholesterol allowed us to measure serum-induced cholesterol mass changes by 1 ) increasing the pool size of cellular cholesterol, and 2 ) altering the expression and array of cell cholesterol transporters. Thus, results from the present studies indicate that the fractional effl ux of labeled cholesterol refl ect the "effl ux effi ciency", or as sometimes termed, "effl ux capacity", of a particular extracellular acceptor. Effl ux effi ciency/capacity can be defi ned as the relative ability of an acceptor to remove labeled cholesterol under a specifi c set of experimental conditions. In most cases, the effl ux of labeled cholesterol does not represent the movement of cholesterol mass. The lack of an association between mass and isotopic values is a refl ection of an exchange of cholesterol driven by the bidirectional fl ux process. The exception to this general phenomenon is if the acceptor is cholesterol-free where there is no bidirectional cholesterol fl ux. The present study expands our understanding of the relationship between radioisotopic fl ux and mass fl ux by showing that there is no association between changes in cell mass and isotope fl ux if the cells contain normal levels of cholesterol. However, we also demonstrate that there is an association between effl ux of label and change in cell cholesterol mass if the cells are cholesterol enriched. This association is a refl ection of greatly enhanced effl ux from enriched cells while at the same time infl ux of exogenous cholesterol is relatively low. Thus, under most conditions, fractional release of label will provide an estimate of the "effl ux effi ciency" of an acceptor whereas actual mass movement has to be determined by measurement of changes in cell cholesterol mass in either the cells or incubation medium.