The uptake of the apoprotein and cholesteryl ester of high-density lipoproteins by the perfused rat liver

Biochimicu Elsevier

et Biophvsica Acts

917 (1987) 9-17

9

BBA 52315

The uptake of the apoprotein and cholesteryl ester of high-density lipoproteins by the perfused rat liver Cynthia

M. Arbeeny,

Department

Vincent A. Rifici and Howard

of Medicrne, Albert Einstein College of Medicine. Bronx, NY 10461 (U.S.A.) (Received

Key words:

HDL;

A. Eder

Hepatic

receptor:

Apolipoprotein

14 July 1986)

A-I; Cholesterol

ester; Lipoprotein

uptake;

(Rat liver)

The uptake of the 1251-labeled apolipoprotein and 3H-labeled cholesteryl ester components of rat apolipoprotein E-deficient HDL by the perfused liver was studied. The uptake of the cholesteryl ester moiety was 4-fold higher than that of apolipoprotein. The concentration-dependent uptake of labeled protein was saturable and competed for by an excess of unlabeled HDL. The uptake of cholesteryl ester was not saturable over the concentration range studied. In the presence of a 50-fold excess of unlabeled HDL, the uptake of both radiolabeled components was decreased by over 75%, indicating that three-quarters of the hepatic uptake of HDL is by a receptor-mediated process. After 15 min of perfusion, 37% of the apolipoprotein radioactivity that was initially bound at 5 min was released into the perfusate as a more dense particle. After 5, 15, 30 and 60 min of perfusion the subcellular distribution of the apolipoprotein and cholesteryl ester components was analyzed by Percoll density gradient centrifugation. Over the 60 min period, there appeared to be transfer of radioactivity from the plasma membrane fraction to the lysosomal fraction. However, the internalization and degradation of cholesteryl ester was more rapid than that of the apolipoprotein. Our findings indicate that there is preferential uptake of HDL cholesteryl ester relative to protein by the liver and that the internalization of these components may occur independently.

Introduction The liver is the major organ for the uptake of the cholesterol component of high-density lipoproteins [l]. The hepatic removal of cholesterol from HDL may be an important means of clearance of cholesterol from the plasma [2]. It has been shown that HDL cholesterol is preferentially incorpo-

Abbreviations: VLDL, very-low-density lipoproteins: LDL, low-density lipoproteins; HDL, high-density lipoproteins; EDTA, ethylenediaminetetraacetic acid; SDS. sodium dodecyl sulfate; LPDS, lipoprotein-deficient serum. Correspondence: Dr. Cynthia M. Arbeeny, Albert College of Medicine. 1300 Morris Park Avenue, Building. Room 511, Bronx, NY 10461, U.S.A.

0005-2760/87/$03.50

Einstein Ullmann

e 1987 Elsevier Science Publishers

rated into biliary sterols [3], and that HDL apolipoproteins are secreted into the bile [4]. A high-affinity binding site for HDL has been identified on the liver cell surface which is distinct from the hepatic LDL and chylomicron remnant receptors [5-71. Recent studies by Glass et al. [8] indicate that there is a dissociation between the tissue uptake of HDL cholesteryl ester and protein. They found that the uptake of the cholesteryl ether component of reconstituted HDL by cultured hepatocytes was over 4-fold greater than that of apolipoprotein A-I [9]. This finding is in contrast to what has been described for the internalization of LDL and remnant lipoproteins by the liver which involves endocytosis of intact particles and subsequent degradation in the lysosomes [lO,ll]. However, the

B.V. (Biomedical

Division)

10

processes mediating the internalization and degradation of HDL components by the liver have not been described and are the focus of this study. Several mechanisms for the selective uptake of cholesteryl ester have been suggested [12-161. It is possible that HDL binds to the cell surface and cholesteryl esters are preferentially transferred into the cell. Another possibility is that the entire HDL particle is internalized with the apoprotein component being released from the cell. Reaven et al. [15] have shown by morphological methods that in ovarian cells labeled HDL protein remains associated with the plasma membrane and that cholesteryl ester is transferred into the cell from the bound HDL particles. Schmitz et al. [16] have demonstrated that colloidal gold-labeled HDL was internalized by macrophages, but was not transported to lysosomes and was transported back to the cell surface. There is evidence that at least part of the HDL apoprotein is degraded by the liver. Have1 et al. [17] have shown that a small fraction of the administered ‘251-labeled HDL is degraded by the perfused liver. We have shown that the ionophore monensin, which prevents the movement of ligand from the cell surface to lysosomes, partially inhibits the degradation of HDL apoprotein by hepatocytes [7]. In the present study we have investigated the receptor-mediated uptake of the protein and cholesteryl ester components of apolipoprotein Edeficient HDL by the perfused liver. This system has the advantage of using an intact organ, where the cellular architecture is maintained. We have also studied the internalization of HDL components by subcellular fractionation of the liver in order to elucidate the mechanisms involved in the metabolism of HDL. Experimental procedures Animals. Male Sprague-Dawley Hewitt, NJ) that weighed 300-350 these studies. All animals were standard Purina rat chow. Animals tized with diethyl ether prior to and all surgical procedures.

rats (Marland, g were used in maintained on were anestheexsanguination

Preparation of radiolabeled HDL. Apolipoprotein E-deficient HDL was isolated from rat serum by a modification [18] of the polyanion

precipitation method [19]. The serum lipoproteins that contain apolipoprotein B and apolipoprotein E were removed by precipitation of the rat serum with sodium phosphotungstate (NaPhT) and magnesium chloride. To 20 ml serum, 2.0 ml 4% NaPhT and 0.5 ml 2 M MgCl, was added and this solution was mixed vigorously. After incubation for 1 h at 20°C the mixture was centrifuged at 1500 x g and 20°C for 1 h. The supernatant was removed and dialyzed against 0.85% NaCl and 0.01% EDTA. The density of the supernatnat was adjusted to 1.21 g/ml with KBr and then centrifuged at 100000 X g for 48 h. This HDL subfraction was then radiolabeled in the cholesteryl ester component by the procedure of Roberts et al. [20], in which radiolabeled ester is incorporated into the core of HDL by incubation with the lipid transfer protein present in the lipoprotein-free fraction. Specifically, 500 PCi of 1,2,6,7-3H cholesteryl oleate (2.2 TBq/mmol, NEN Research Products, Boston, MA) was incubated for 45 min at 23°C with 12 ml of human LPDS. This mixture was then incubated for 30 min at 4°C with 3 mg of apolipoprotein E-deficient HDL cholesteryl (1 mg/ml). The density was then adjusted to 1.21 g/ml and the 3H-labeled HDL was re-isolated by centrifugation. The HDL was dialyzed against 0.85% NaCl and 0.01% EDTA and radioiodinated [21] yielding apolipoprotein E-deficient HDL that contained 3H-cholesteryl ester and ‘251-labeled apolipoprotein. The distribution of 3H-radiolabel between cholesterol and cholesteryl ester was analyzed by thin layer chromatography [22] and indicated that 98% of the radiolabel was associated with the cholesteryl esters. The protein content of HDL was determined by a modification [23] of the method of Lowry et al. [24]. Aliquots (50 pg of protein) were dried under vacuum and delipidated for analysis by SDS polyacrylamide electrophoresis utilizing gels with a gradient of 3-27s acrylamide [25]. The SDS-gradient gel analysis is shown in Fig. 1. In lane 1 is the apolipoprotein E-deficient HDL that was labeled with 3H and lz51. This fraction contained apolipoprotein A-IV, apolipoprotein A-I and the C-apolipoproteins. In lane 2 is the total HDL fraction (isolated by centrifugation at a density of 1.063-1.21 g/ml) which contained apolipoprotein E.

11

analysis that the the lipid in these

AIV E

Fig. 1. SDS-gradient gel electrophoresis of rat HDL. An aliquot containing 50 pg of delipidated lipoprotein was applied to each position, except lane 3 which contained 20 pg of bovine serum albumin. Lane 1 is double-labeled apo E-deficient HDL. lane 2 is total rat HDL.

The lipid composition of the HDL which was incubated with LPDS or saline alone was determined. Free and esterified cholesterol were assayed by enzymatic methods (Finnsugar Biochemicals, Elks Grove Village, IL). Triglycerides [26] and phospholipids [27] were measured by chemical methods. The results of these determinations are shown in Table I. Both the electrophoretic

TABLE

and chemical composition data indicate incubation with LPDS had no effect on or protein components of the HDL used studies. Liver perfusion. Non-recirculating liver perfusions were performed in situ as previously reported [28]. After a 15 min stabilization period, a bolus of labeled lipoprotein (lo-500 pg of protein in 400 ~1) was injected into the portal vein cannula in order to measure the concentration-dependent uptake of 3H- and “‘I-labeled HDL. In competition studies, a bolus containing radiolabeled HDL (250 ~18 of protein in 200 ~1) and a separate bolus of unlabeled competitor (0.5-12.5 mg in 200 ~1) were simultaneously injected. The livers were perfused for an additional 5 min with lipoprotein-free perfusate and then flushed with saline. The livers were homogenized in ice-cold 0.15 M NaCl/O.Ol M Tris-HCl (pH 7.6) by six strokes in a Dounce homogenizer using the loosefitting pestle. The radioactivity from ‘*‘I-labeled apolipoprotein that was associated with the liver homogenates was then determined using a gamma counter. The 3H-labeled cholesteryl ester radioactivity was extracted from triplicate aliquots (0.5 ml) of the liver homogenates in 10 ml of chloroform methanol (3 : 1, v/v) and measured by liquid scintillation counting. The 3H radioactivity that was extracted by chloroform-methanol was analyzed by thin layer chromatography [22] to determine the distribution of radiolabel in cholesteryl ester and unesterified cholesterol associated with the liver. The catabolism of HDL protein and cholesteryl ester was studied over a 1 h interval. A bolus of radiolabeled HDL was injected into the portal vein cannula followed by a 10 min single-pass perfusion of lipoprotein-free perfusate to wash out

I

CHEMICAL COMPOSITION LPDS OR SALINE Values are percentages

HDL + LPDS HDL + saline

OF APOLIPOPROTEIN

E-DEFICIENT

HDL

FOLLOWING

INCUBATION

WITH

by weight. Unesterified cholesterol

Cholesteryl ester

Triglyceride

Phospholipid

Protein

2.5 3.1

31.3 32.0

8.0 6.6

17.8 17.9

40.4 40.3

HUMAN

12

any unbound lipoprotein. Then 80 ml of fresh perfusate was recirculated through the liver for 1 h. At 15 min intervals after the bolus injection, 250 ~1 aliquots of perfusate were removed from the reservoir, the erythrocytes were removed by centrifugation and the 1251 radioactivity that was soluble and insoluble in 10% trichloroacetic acid was determined. Bile samples were also collected during the entire period and assayed for trichloroacetic acid soluble and insoluble radioactivity from ‘251-labeled protein and radioactivity from 3Hlabeled cholesteryl ester. Perfusate was collected after 15 min of perfusion and was concentrated by membrane dialysis in Aquacide (Calbiochem, La Jolla, CA). The density of the solution was then adjusted to 1.21 g/ml using KBr and centrifuged at 100000 x g for 48 h. The radioactivities associated with both the d < 1.21 and d > 1.21 fractions were then measured to determine whether the radioactivity associated with the lipoproteins that were present in the perfusate was of the same density as the HDL that was originally injected into the liver. A control experiment was performed in which radiolabeled HDL was reisolated after circulation through the perfusion apparatus in the absence of the liver. The fraction that floated at d< 1.21 was then cleared of albumin by recentrifugation and the apolipoprotein constituents were analyzed by SDS gel electrophoresis. The intracellular distribution of radiolabeled apoprotein and cholesteryl ester in the liver following perfusion with HDL was studied by Perco11 (Pharmacia, Piscataway, NJ) density gradient centrifugation by the method of Harford et al. [29]. At 5, 15, 30 and 60 min of perfusion after injection of 3H and ‘251-HDL, the livers were rinsed with 50 ml of 0.15 M NaCl/O.Ol M TrisHCl (pH 7.6) and then excised, weighed and placed on ice. All subsequent procedures were performed at 4°C. The livers were homogenized in 3 vol (w/v) of ice-cold 0.28 M sucrose/O.002 M CaCl,/O.Ol M Tris-HCl (pH 7.6) using a Dounce homogenizer (12 strokes of the loose fitting pestle and three strokes of the tight fitting pestle). The homogenates were then centrifuged at 300 X g for 15 min and the supernatants were made to be 20% in Percoll. The samples were then centrifuged at 10000 X g in a Beckman SW 41 rotor for 45 min.

The gradients were fractionated into 0.6 ml fractions by displacement from the bottom with 50% sucrose. The 12’1 and 3H radioactivity in each of the gradient fractions was determined. The distribution of label between cholesteryl ester and unesterified cholesterol in each gradient fraction was detrmined by thin-layer chromatography. The location in the gradients of the plasma membrane marker enzyme 5’-nucleotidase [30] and the lysosome marker enzyme, /%hexosaminidase [31] were determined. Results

The concentration-dependent uptake of apolipoprotein E-deficient HDL by the liver was measured to determine saturability of uptake of the radiolabeled components of HDL (Fig. 2). Perfusion of the livers with increasing amounts of radiolabeled HDL resulted in the saturable uptake of the 1251-labeled apolipoprotein indicating that the uptake of HDL is mediated though a limited number of binding sites. The uptake of 3Hcholesteryl ester increased linearly and did not saturate over the concentration range studied. The 3H-labeled cholesteryl ester found in the liver was 4-fold higher than ‘251-labeled protein suggesting a dissociation of apolipoprotein and cholesteryl

5.0 -

I loo

I

I

200

I

I

I

300

HDL ( pg protein 1

I 400

I, 500

Fig. 2. Concentration dependent uptake of HDL protein and cholesteryl ester components. The uptake of the ‘251-labeled protein (O--O) and ‘H-labeled cholesteryl ester (O--O) of apolipoprotein E-deficient HDL by the perfused liver was determined. Each point is the mean of three perfusions.

13

ester uptake by the liver. In order to determine receptor-mediated binding, a competition study was performed. Livers were perfused with a constant amount of radiolabeled HDL protein and an increasing amount of unlabeled HDL (Fig. 3). At a 50-fold excess of unlabeled HDL the uptake of by 75% and apolipoprotein was decreased cholesteryl ester was decreased by 77%, indicating that three-quarters of the hepatic uptake of HDL is receptor mediated. In another study, a recirculating perfusion was used to mesure the degradation of HDL. After 15 min of perfusion 37% of the radioactivity that was initially bound at 5 min was present in the perfusate (Fig. 4). After 30 min this value represented 56% of that which was initially bound. At subsequent times there was no additional release of radioactivity into the perfusate. Of all the radioactivity present in the perfusate at 15 min, 93% was insoluble in 10% trichloroacetic acid, suggesting that this radioactivity represented undegraded HDL that was released from the liver. When the perfusate was centrifuged at a density of 1.21 g/ml, 32% of the protein radioactivity floated at this density. In a control experiment where HDL was recirculated through the perfusion apparatus

I

I

25

I

I

I

75 IO 0 Unlabe~do HDL (mg protein)

I

I

12.5

Fig. 3. Competition for the uptake of HDL protein and cholesteryl ester. Livers were perfused with 250 pg of HDL protein and the uptake of ‘2SI-labeled protein (O- -0) and 3H-cholesteryl ester (0- -0) was determined in the absence and presence of increasing amounts of unlabeled HDL. Each point is the mean of three perfusions.

30 Time(minutes1 Fig. 4. Release of ‘25I radioactivity into the perfusate. Livers were perfused with 250 pg of double labeled HDL and the accumulation of “‘1 radioactivity that was trichloroacetic acid-insoluble (O--O) and soluble (O--O) was determined over a 60 min period.

in the absence of the liver and then reisolated by ultracentrifugation, 92% of the radioactivity floated at a density of 1.21 g/ml. SDS gel electrophoresis showed that the HDL that was released by the liver and present in the perfusate had a similar apolipoprotein distribution as the reisolated control HDL, indicating that there is no selective removal of any individual apoprotein. During a 60 min recirculating perfusion, there was a linear increase of radioactivity in the perfusate that was soluble in 10% trichloroacetic acid. This finding suggests that some of the HDL protein is internalized and degraded within the lysosomes. After 15 min of perfusion less than 5% of the radiolabeled cholesteryl ester was present in the perfusate and remained constant over the next 45 min of perfusion. Thus it appears that more cholesteryl ester than protein is internalized and that the released HDL that was found in the perfusate is depleted in cholesteryl ester. After 60 min of perfusion the apoprotein and cholesteryl ester radioactivity that was present in the bile was minimal and represented 6.5% and 1.0% of the total uptake at 5 min of perfusion (data not shown). The association of the protein and cholesteryl ester components of HDL after internalization by the liver with the plasma membrane and lysosomes was determined after centrifugation on Perco11 density gradients (Fig. 5 and 6). The activity

50

Grad&t

30 Mmules

a

60 Minutes

6

Fraction-

Fig. 5. Percoil density gradient analysis of internalized cnmponcnts of HDL after 5 (A) and 15 (B) min of perfusion. The 1251-labeled protein (S -0) and 3H-labeled cholesterol (O- -0) radioactivity was determined for each gradient fraction. Each time point is representative of three perfusions.

for the plasma membrane marker, S-nucleotidase, was found in fraction 1, and the activity of the lysosome marker &hexosaminidase, was found in fraction 19. ‘Throughout the 60 min period, greater than 98% of the 125I radioactivity in the gradient fractions was insoluble in IO% t~~h~oroa~ti~ acid, indicating that the radioactivity was still protein bound. At all times there was more choiesteryl ester than protein present in the liver. During the 60 min of perfusion, there appeared to be transfer of radioactivity from the plasma membrane to the lysosome fraction. However, cholesteryl ester transferred from the plasma membrane to the lysosome fraction more rapidly. At 5 min after the injection of radiol~b~led HDL, both labels were largely associated with the plasma membrane fraction (Fig. 5A). At this time

Fig, 6. Percdl density gradient analysis of internalized components of HDL after 30 (A) and 60 (B) min of perfusion. The ‘“51-labeled protein (@- -0) and 3H-labeled cholesterol (0~~-0) radioactivity was determined. Each time point is representative of three perfusions.

there was 4-times more chofesteryl ester radioactivity than protein associated with the plasma membrane fraction. Lipid analysis by thin layer chromatography indicated that 97% of the 3H-iabel in a11 gradient fractions was present as cholesteryl ester. After ‘fS min of perfusion (Fig. 5B), ‘H radioactivity was associated with the plasma membrane fraction, an intermediate density fractions as well as with the lysosomal fraction. At this time 96% QE the 3H-label was found as cholesteryl ester. 1251-Iabeled protein radioactivity was decreased in the plasma membrane fraction and increased in the lysosomal fraction. Both the lz51 and ‘H radioactivity in the plasma membrane fraction was di~nish~d at 30 min (Fig. 6A), and at this time the 3H radioactivity in the lysosomes was 2-fold higher compared to 1251-

15

labeled apoprotein. Of the 3H radioactivity, 76% and 81% of the radioactivity associated with the plasma membrane fraction and lysosomal fraction was as cholesteryl ester. After 60 min of perfusion (Fig. 6B), there still appeared a small amount of 125I-labeled apolipoprotein that was associated with the lysosomes, while all of the 3H was associated with the plasma membrane fractions. Half of the 3H-label was as cholesteryl ester and half was found as unesterified cholesterol. These results indicate a preferential uptake of cholesteryl ester from HDL by the liver and a more rapid movement of cholesteryl ester to the lysosomes compared to HDL apoprotein. In addition, there appears to be hydrolysis of the cholesteryl ester and appearance of unesterified cholesterol in the plasma membrane. Discussion Our results indicate that there is selective hepatic uptake of HDL cholesteryl ester relative to apoprotein and that the internalization and degradation of these components may occur independently. The uptake of the radiolabeled apolipoprotein was saturable and competed for by an excess of unlabeled HDL. The data indicate that 77% of the uptake is by a receptor mediated process. The uptake of radiolabeled cholesteryl ester was 4-fold higher than the uptake of apoprotein. These results are consistent with other studies which indicate that there is selective uptake of HDL cholesteryl ester by hepatocytes [8,9] and steroidogenic cells [8,12]. In contrast to what was found for apolipoprotein uptake, the uptake of cholesteryl ester by the liver did not saturate over the concentration range studied. The results of the competition study show that in the presence of a l-fold excess of unlabeled HDL, competition for the uptake of protein and cholesteryl ester were similar and may be due to the contribution of the nonspecific binding to the total binding. At a l-25-fold excess of unlabeled HDL, the uptake of protein was competed for to a greater extent than was the cholesteryl ester, indicating that the uptake of protein saturates at a lower concentration than cholesteryl ester. At a 50-fold excess of unlabeled HDL, the competition for the uptake of both components was equal.

This result suggests that the saturation of the capacity for uptake of cholesteryl ester occurs at a higher concentration than protein. Glass et al. [9] have also shown using rat hepatocytes that although the absolute uptake of cholesteryl ester was higher than protein, the degree of competition was similar at higher concentrations of unlabeled HDL. It is still unclear whether the binding of HDL apolipoprotein to the receptor is required for the uptake of cholesteryl ester. In this study we observed that a fraction of the HDL that was initially bound was released from the surface of the hepatocyte into the perfusate. This particle was more dense than the initial starting material and may have been depleted of cholesteryl ester. This is consistent with the finding of Bachorik et al. [13] who showed that in cultured pig hepatocytes a portion of the ‘*‘Ilabeled HDL that was initially bound was internalized and a portion was modified at the cell surface and then released as a more dense lipoprotein. Bamberger et al. [32] has suggested that hepatic lipase, by its phosphoiipase activity, shifts the equilibrium of cholesterol between HDL and the plasma membrane resulting in the net delivery of cholesterol to the cell by a surface-transfer process. Reaven et al. [15] have shown using radioautography that in the luteunized ovary perfused in situ, HDL is not internalized as an intact particle. They showed that the protein moiety remained associated with the plasma membrane and speculate that the cholesterol is transferred to a specialized region of the plasma membrane. Others have suggested [16] that the HDL is internalized by the cell and the cholesterol is removed before the particle is returned to the exterior. This process of retroendocytosis has been visualized for the uptake of HDL colloidal gold conjugates by macrophages. Whether HDL is internalized by the liver as an intact particle is still not known. However, our results suggest that the dissociation of the HDL components may occur at the hepatocyte surface. In the present study we have used apolipoprotein E-deficient HDL in order to exclude binding of HDL by the apolipoprotein E receptor. However, Quarfordt et al. [33] have shown that there was substantially greater uptake of apolipoprotein E-rich HDL than apolipoprotein E-defi-

16

cient HDL by the perfused liver and suggest that apo E-rich HDL plays an important role in the return of cholesterol to the liver for subsequent metabolism. In a previous study [7] we have found that apo A-I was more effective than apo E in competing for the binding of HDL which contained apolipoprotein E, indicating that both apolipoprotein A-I and apolipoprotein E are ligands for the binding of HDL. It is possible that apolipoprotein E-rich HDL may be metabolized, in part, by a similar pathway mediating the intracellular processing of apolipoprotein E-deficient HDL. The results of the Percoll density gradient analyses show that there was transfer of radioactivity from the plasma membrane to the lysosome fraction and that the movement of HDL cholesteryl ester to lysosomes was more rapid than protein. This finding suggests that the internalization and degradation of HDL constitutents may occur independently. We have previously shown [7] that the presence of monensin, an ionophore that inhibits receptor recycling, inhibited the degradation of ‘251-labeled HDL protein by isolated hepatocytes. Pittman et al. [14] using adrenal cells, found that chloroquine and monensin reduced apolipoprotein A-I uptake by 62% but reduced cholesteryl ester uptake by less than 25%. This suggests that the selective uptake of cholesteryl ester is by a mechanism different from that for protein. Our results indicate that both HDL protein and cholesteryl ester are degraded within lysosomes, although the degradation of protein occurs more slowly. In contrast to our previous liver perfusion (Arstudies with 12SI-labeled VLDL remnants beeny, C.M., Rifici, V.A., Handley D.A. and Eder, H.A. unpublished observations), the rate of degradation of rat ‘251-labeled HDL was 4-fold slower than that of remnants. Using Percoll density gradient analyses to study the intracellular movement of 1251-labeled VLDL remnants, we found that the movement of remnants to lysosomes occurred more rapidly than that of 1251labeled HDL. In a recent publication Bachorik et al. [13] reported that the ‘251-labeled HDL protein that was internalized by cultured pig hepatocytes was degraded by lysosomes at a rate that was half that of LDL. The reason for the slower delivery of

HDL to lysosomes remains to be evaluated, but may explain the low rate of degradation of 1251labeled HDL by the perfused liver previously reported by Sigurdsson et al. [17]. Our findings suggest that internalized cholesteryl ester is processed in part by lysosomal hydrolases. However, since we found free cholesterol in the membrane fractions of the gradient, some cholesteryl ester may be metabolized at the plasma membrane and degraded by a cytosolic neutral sterol ester hydrolase [34]. Nestler et al. [35] found that treatment of ovarian cells with cholorquine and NH,Cl which inhibit lysosomal hydrolysis, had little effect on the metabolism of HDL sterol esters. It is possible that there may be an intracellular partitioning of the HDL cholesterol within the hepatocyte. Previous studies have suggested that HDL cholesterol in preference to LDL cholesterol is the source for biliary cholesterol [3]. Robins et al. [26] have suggested that lipoprotein cholesterol may be transported directly from the plasma membrane of the hepatocyte to the bile canaliculus without equilibrating with newly synthesized intracellular cholesterol. The concept that intracelluar lipid transfer may be a highly specific process has been put forth by several investigators [37-391 and a sterol carrier protein present in the rat liver cytosol has been implicated in the intracellular transfer of cholesterol among organelles [40]. The results of these studies indicate that the binding, internalization and degradation of the apoprotein and cholesteryl ester components of HDL by the liver is a more complex process than that described for LDL and remnants of triglyceride-rich lipoproteins which are transported to the lysosomes as intact particles. Further study is necessary to elucidate the pathways that mediate the hepatic uptake of HDL. Acknowledgements

The authors gratefully acknowledge the excellent technical assistance of Diane Edelstein and Craig Bliss. This study was supported by Research Grant HL 26817 from the National Institutes of Health and an Established Fellowship Award from the American Heart Association-New York City Affiliate (CMA).

17

References I Glass, C., Pittman. R.C., Keller, G.A. and Steinberg, D. (1968) J. Biol. Chem. 258, 7161-7167 2 Glomset, J.A. (1968) J. Lipid Res. 9, 155-167 3 Schwartz, C.C.. Halloran, L.G., Vlahcevic, Z.R., Gregory, D.H. and Swell, L. (1978) Science 200, 62-64 4 Sewell, R.B., Mao, S.J.T., Kawamoto, T. and LaRusso, N.F. (1983) J. Lipid Res. 24, 391-401 5 Bachorik, P.S., Franklin, F.A., Virgil, D.G. and Kwiterovich, P.O. (1982) Biochemistry 21, 5675-5684 6 Soltys, P.A., Portman, O.W. and O’Malley, J.P. (1982) Biochim. Biophys. Acta 713. 300-314 7 Rifici, V.A. and Eder, H.A. (1984) J. Biol. Chem. 259, 13814-13818 8 Glass, C.. Pittman, R.C., Weinstein, D.B. and Steinberg, D. (1983) Proc. Natl. Acad. Sci. USA 80, 5435-5439 9 Glass, C., Pittman, R.C.. Given, M. and Steinberg, D. (1985) J. Biol. Chem. 260, 744-750 10 Handley, D.A., Arbeeny, C.M., Eder, H.A. and Chien, S. (1981) J. Cell. Biol. 90, 778-787 11 Jones, A.L., Hradek, G.T., Homick, C., Renaud, G., Windler, E.E.T. and Havel, R.J. (1984) J. Lipid Res. 25, 1151-1158 12 Gwynne. J.T. and Hess, B. (1980) J. Biol. Chem. 255, 10875-10883 13 Bachorik, P.S., Franklin, F.A., Virgil, D.G. and Kwiterovich, P.O. (1985) Arteriosclerosis 5, 142-152 14 Pittman, R.C., Knecht, T.P. and Taylor, CA. (1985) Circulation 72, 377 15 Reaven. E., Chen, Y-D.I., Spicher, M. and Azhar, S. (1984) J. Clin. Invest. 74, 1384-1397 16 Schmitz, G., Robenek, H., Lohmann. U. and Assmann, G. (1985) EMBO J. 4, 613-622 17 Sigurdsson, G., Noel. S.P. and Havel, R.J. (1979) J. Lipid Res. 20, 316-324 18 Arbeeny. C.M. and Eder, H.A. (1983) J. Lipid Res. 24, 1457-1467 19 Burstein, M., Scholnik. H.R. and Morfin, R. (1980) J. Lipid Res. 11, 583-595 20 Roberts. D.C.K., Miller, N.E., Crook, D., Price, S.L., Cortese, C., LaVille. A.. Masan, L. and Lewis, B. (1985) Biochem. J. 226, 319-322

21 Bilheimer, D.W., Eisenberg, S. and Levy, R.J. (1972) Biochim. Biophys. Acta 260, 212-221 22 Chen, C.H. and Albers. J.J. (1983) Biochim. Biophys. Acta 753, 40-46 23 Sata, T., Havel, R.J. and 13, 757-768 24 Lowry, O.H., Rosebrough, R.J. (1951) J. Biol. Chem. 25 Swaney, J.B. and Kuehl, Acta 446, 561-565 26 Fletcher. M.J. (1968) Clin. 27 Norton, W.T. and Autilio, 213-222

Jones, A.L. (1972) J. Lipid

Res.

N.J., Farr, A.L. and Randall, 193, 265-275 K.S. (1976) Biochim. Biophys. Chim. Acta 22. 393-397 L.A. (1966) J. Neurochem.

13.

2X Arbeeny, C.M. and Rifici, V.A. (1984) J. Biol. Chem. 259, 9662-9666 29 Harford, J., Wolkoff, A.W., Ashwell, G. and Klausner, R.D. (1983) J. Cell Biol. 96, 1824-1828 30 Widnell. C.C. and Unkeless, J.C. (1968) Proc. Natl. Acad. Sci. USA 61, 1050-1057 31 Hall, C.W., Liebars, I., DiNatale. P. and Neufeld, E.F. (1978) Methods Enzymol. 50, 439-456 32 Bamberger, M., Lund-Katz, S., Phillips, M.C. and Rothblat, G.H. (1985) Biochemistry 24, 3693-3701 33 Quarfordt, S., Hanks, J., Jones, R.S. and Shelburne, F. (1980) J. Biol. Chem. 255. 2934-2937 34 Khoo. J.C., Drevon. C.A. and Steinberg, D. (1979) J. Biol. Chem. 254, 1785-1787 35 Nestler, J.E., Bamberger, M., Rothblat. G.H. and Strauss, III, J.F. (1985) Endocrinology 117. 502-510 36 Robins, S.J., Fasulo, J.M. Collins, M.A. and Patton, G.M. (1985) J. Biol. Chem. 260, 6511-6513 37 Dempsey, M.E., McCoy, K.E., Baker, H.N., DimitriadouVafiadou, A., Lorsback, T. and Howard, J.B., (1981) J. Biol. Chem. 256, 1867-1873 38 Gavey, K.L., Noland. B.J. and Scallen. T.J. (1981) J. Biol. Chem. 256, 2993-2999 39 Wetterau, J.R. and Zilversmit, D.B. (1984) J. Biol. Chem. 259, 10863-10866 40 Chanderbhan. R., Noland, B.J., Scallen, T.J. and Vahouny, G. (1982) J. Biol. Chem. 257, 8928-8934