Binding and degradation of human high-density lipoproteins by human hepatoma cell line HepG2

Binding and degradation of human high-density lipoproteins by human hepatoma cell line HepG2

100 Biochimica et Biophysics Acta 833 (1985) 100-110 Elsevier BBA 51831 Binding and degradation of human high-density lipoproteins by human hepatom...

1MB Sizes 0 Downloads 51 Views

100

Biochimica et Biophysics Acta 833 (1985) 100-110 Elsevier

BBA 51831

Binding and degradation of human high-density lipoproteins by human hepatoma cell line HepG2 Nassrin Dashti *, Gertrud

Wolfbauer

** and Petar Alaupovic

The Loboratory of Lipid and Lipoprotein Studies, Oklahoma Medical Research Foundation, 825 Northeast Street, Oklahoma City, OK 73104 (U.S.A.) (Received (Revised manuscript

Key words:

HDL, Lipoprotein

Thirteenth

February 27th, 1984) received September 17th, 1984)

receptor;

Apolipoprotein;

(Human

hepatoma

HepG,)

The catabolism of human HDL was studied in human hepatoma cell line HepG2. The binding of ‘251-labeled HDL at 4°C was time-dependent and reached completion within 2 h. The observed rates of binding of ‘251-labeled HDL at 4°C and uptake and degradation at 37°C indicated the presence of both high-affinity and low-affinity binding sites for this lipoprotein density class. The specific binding of ‘251-labeled HDL accounted for 55% of the total binding capacity. The lysosomal degradation of ‘251-labeled HDL was inhibited 25 and 60% by chloroguine at 50 and 100 PM, respectively. Depolymerization of microtubules by colchicine (1 PM) inhibited the degradation of ‘251-labeled HDL by 36%. Incubation of cells with HDL caused no significant change in the cellular cholesterol content or in the de novo sterol synthesis and cholesterol esterification. Binding and degradation of 1251-labeled HDL was not affected by prior incubation of cells with HDL. When added at the same protein concentration, unlabeled VLDL, LDL and HDL had similar inhibitory effects on the degradation of ‘251-labeled HDL, irrespective of a short or prolonged incubation time. Reductive methylation of unlabeled HDL had no significant effect on its capacity to inhibit the ‘251-labeled HDL degradation. The competition study indicated no correlation between the concentrations of apolipoproteins A-I, A-II, B, C-II, C-III, E and F in VLDL, LDL and HDL and the inhibitory effect of these lipoprotein density classes on the degradation of 1251-labeled HDL. There was, however, some association between the inhibitory effect and the levels of apolipoprotein D and C-I. Introduction Recent studies have shown that plasma lipoprotein participate not only in the transport but also in the cellular metabolism of lipids. Whereas the * To whom correspondence should be addressed. ** Present address: Department of Physiology and Biochemistry, Medical College of Pennsylvania, Philadelphia, PennSylvania 19129, U.S.A. Abbreviations: HDL, high-density lipoproteins (d 1.063-1.21 g/ml); LDL, low-density lipoproteins (d 1.019-1.063 g/ml); VLDL, very-low-density lipoproteins (d < 1.006 g/ml); Hepes, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; HMGCoA, P-hydroxy-b-methylghrtaryl coenzyme A. 0005-2760/85/$03.30

0 1985 Elsevier Science Publishers

B.V.

role of LDL in delivering esterified cholesterol to a variety of cells through a receptor-mediated process is well established [1,2], the exact function of HDL in cholesterol homeostasis is still not known. It has been suggested that HDL play a major role in the esterification of plasma cholesterol during the initial stages of catabolism of triacylglycerolrich lipoprotein [3]. Several other studies have shown that HDL stimulate removal of cholesterol from a variety of cells such as cultured skin fibroblasts and aortic smooth muscle cells [4,5], macrophages [6] and cultured rabbit hepatocytes [71. The role of liver in catabolism

of HDL has not

101

been clarified. The early in vivo studies [&lo] have shown that, in the rat, the liver is active in accumulating injected HDL predo~nantly in the parenchymal cells [9]. Hip-affinity binding and HDL have been redegradation of ‘25-i-labeled ported to take place in rat parenchymal [ll-131 and non-parenchymal [12,13] cells. It appears, however, that only a small fraction of HDL is degraded by the rat liver [14,15]. The recently established cell line from human liver tumor biopsies, designated HepG2, has been shown to have mo~holo~cal characteristics compatible with those of liver parenchymal cells [I@. It has also been demonstrated that this cell line is capable of synthesizing and secreting into the medium human plasma apolipoproteins A-I, A-II, B, C-II, C-III and E [17,18]. Furthermore, recent studies from this laboratory [19,20] and those by Havekes et al. (211 have established that HepG2 cells possess ~~-affinity receptors for binding and degradation of human LDL. Since these results indicated the usefulness of human hepatoma cell line HepG2 for studying the hepatic metabolism of human plasma lipoproteins in a homologous system, we have extended these investigations to studies on the mechanism of binding and degradation by HepG2 cells of human HDL. Preliminary results of this study have been reported in an abstract form [19]. Materials and Methods Tritiated water and [l-‘4C]oleic acid were purchased from New England Nuclear (Boston, MA) and sodium [‘251]iodide was obtained from Amersham Corp. (Arlington Heights, IL). IodoBeads were purchased from Pierce Chemical Co. (Rockford, IL). Modified minimum essential medium and fetal calf serum were obtained from Grand Island Biological Co. (Grand Island, NY). Sodium pyruvate, L-glutamine and minimum essential medium vitamin solution were from KC Biological Inc. (Lenexa, KS). Chloroquine, colchicine, digitonin, EDTA and Hepes were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals were reagent grade. Cell C’t&ure. Human hepatoma cell line HepG2 was obtained from Dr. Barbara Knowles, The Wistar Institute, P~ladelp~a, PA. Cells were cul-

tured in T-25 flasks (25 cm2 surface area) in 5 ml of minimum essential medium supplemented with 10 ml/l of 100 x minimum essential medium vitamins, 0.11 g/l of sodium pyruvate, 1.17 g/l of L-glutamine and 10% (v/v) fetal calf serum at 37°C in equilibration with 95% sir/5% CO,. The cells were routinely split 1: 6 every 5-6 days. 6-7 days prior to each experiment, cells were seeded in either multiwell dishes (16 mm diameter) or in 60 mm diameter tissue culture dishes ((0.2-0.3) - lo6 cells/well or (0.7-1.0). lo6 cells/dish) in 0.1-0.2 ml ~nimum essential medium supplemented with 10% fetal calf serum/cm* surface area. 5 days after plating, the maintenance medium was removed, cells were washed with 10 mM phosphate-buffered saline (pH 7.4) and minimum essential medium containing 5% (v/v) lipoprotein-deficient human plasma (d > 1.23 g/ml) was added to plates; all experiments were started 24-48 h thereafter as indicated in the Results section. At the time of assay, dishes were near confluence, Isolation of lipoprotein density classes and iodination and modification of HDL. Plasma was ob-

tained from normolipidemic subjects after an overnight fast. VLDL (d -C 1.006 g/ml), LDL, (d 1.019-1.063 g/ml) and HDL (d 1.063-1.21 g/ml) were isolated by sequential preparative ultracentrifugation [22] of plasma and washed under identical conditions until free of albumin. The isolated lipoproteins were dialyzed against phosphate-buffered saline containing 0.01% EDTA (pH 7.4). LDL and HDL were iodinated with NaiZSI using Iodo-Beads as described by Markwell [23]. The ‘251-labeled lipoproteins were dialyzed for 48 h at 4°C against phosphate-buffered saline containing 0.01% EDTA (pH 7.4) to remove free iodide. The properties of ‘251-labeled LDL have been reported previously [20]. In all HDL preparations approx. 96% of the radioactivity was precipitable in 10% trichloroacetic acid and 4% of the radioactivity of ‘251-labeled HDL was associated with lipids; HDL had a specific activity of 80 cpm/ng protein. Reductive methylation of HDL was carried out according to the procedure described by Weisgraber et al. [24]. Determination of lipids and apolipoproteins. Triacylglycerol and cholesterol contents of lipoprotein density classes and HepG2 cells were determined as described by Kuksis et al. 1251. The

102

concentrations of apolipoproteins A-I, A-II, B, C-I, C-II, C-III, D, E and F were measured by electroimmunoassays developed in this laboratory [26-321. Lipoprotein binding and degradation assays. For binding studies, cells in 16 mm diameter multiwell plates were precooled for 30 min at 4°C. The cells were then incubated with fresh cold medium (without bicarbonate) supplemented with 10 mM Hepes (pH 7.4), 5% lipoprotein-deficient plasma and ‘Z51-labeled lipoproteins in the presence or absence of an excess of unlabeled lipoproteins in the total volume of 0.5 ml. At the end of incubation, medium was removed at 4°C and the cells were washed twice each with 2 ml of ice-cold phosphate-buffered saline (pH 7.4), 2 ml of phosphate-buffered saline containing 0.2% bovine serum albumin and, finally, with 2 ml of phosphate-buffered saline, and then dissolved in 0.7 ml of 0.1 M NaOH. The radioactivity associated with the cells was determined in a gamma counter and the cell protein was measured by the method of Lowry et al. [33]. Binding, internalization and degradation of ‘251-labeled HDL at 37°C were performed under similar conditions except that cells were incubated with minimum essential medium containing bicarbonate and 5% lipoprotein-deficient plasma without Hepes. At the end of incubation, dishes were placed on ice, medium was removed and cell monolayers were washed as previously described. To determine the amount of ‘251-labeled HDL bound to the cell surface, monolayers were treated with 0.05% trypsin and incubated for 5 min at 37°C as described by Stein et al. [34]. The cells were centrifuged at 2000 rpm for 10 min at 4°C. The radioactivity released by trypsin is the measure of bound HDL and that remaining with the cells is the internalized HDL. Degradation of HDL was determined in the medium after precipitation of protein with ice-cold trichloroacetic acid. Trichloracetic acid at a final concentration of 10% (w/v) was added to the medium and the mixture was kept for 1 h in an ice bath; the precipitate was removed by centrifugation at 2000 rpm for 30 min at 4°C and the degradation was determined by measuring the radioactivity in the trichloroacetic acid-soluble supernatant fraction. Heating the supernatant fraction for 10 min at 100°C water bath

followed by centrifugation had no effect on the total trichloroacetic acid-soluble radioactivity. In some experiments, in order to remove free iodide, a 1.0 ml aliquot of acid-soluble supernatant fraction was treated with 0.01 ml of 40% (w/v) KI and 0.04 mi of 30% H,02 and, after exactty 5 min at room temperature, the mixture was extracted with 2.0 ml of chloroform 1351. The iodide-free trichloroacetic acid-soluble radioactivity was determined in the aqueous phase. To control for spontaneous generation of acid-soluble 1251, dishes without cells were incubated under identical conditions. All binding and degradation assays were performed in the presence of 1.8 mM CaCl,.

measurements of cholesterol e~teri~~ation and sterol synthesis. Cholesterol esterification was measured by determining in incorporation of [lr4C]oleic acid into cholesteryl oleate as described by Goldstein et al. [36]. Cells in 60 mm dishes were incubated with minimum essential medium containing 5% lipoprotein-deficient human plasma and increasing concentrations of unlabeled HDL for 18 h at 37°C. Monolayers were then supplied with [l- “C]oleate bound to albumin and incubated for an additional 4 h at 37°C. After removing the medium, cells were washed as previously described, then scraped off the plates with a rubber policeman after addition of two times 1 ml of phosphate-buffered saline and sonicated. l-ml ahquots of cell suspensions were extracted with 20 ml of chloroform/methanol (2: 1); extracts were washed and applied to thin-layer chromatography plates as described previously [37]. The cholesteryl ester bands were scraped into vials containing Instagel for liquid scintillation counting. Cholesterol synthesis was determined by measuring the incorporation of tritiated water (‘H,O) into cellular sterols. Ceils in 60 mm dishes were incubated with minimum essential medium containing 5% lipoprotein-deficient plasma and increasing concentrations of HDL. After incubation for 18 h at 37”C, 1 mCi of 3H20 was added to each plate and the cells were incubated for an additional 2 h. At the end of incubation, medium was removed and the cells were washed and harvested as described above. l-ml aliquots of cell suspensions were saponified by addition of 1.0 ml of 60% KOH and 2.0 ml of absolute ethanol followed by a 2 h incubation in a sand-bath at

103

90°C. The incorporation of tritium precipitable sterols was measured described [37].

into digitoninas previously

Results Effect of incubation time and substrate concentration on the binding of “‘I-labeled lipoproteins to HepG2 cells The binding of ‘251-labeled HDL (10.0 pg protein/ml of medium) to HepG2 cells at 4°C appeared to be time-dependent, reaching completion within 2 h (Fig. 1). When cells were incubated with increasing concentrations of 125I-labeled LDL (Fig. 2A) or ‘251-labeled HDL (Fig. 2B) for 2 h at 4°C the binding of these lipoproteins did not reach saturation but continued to increase up to 200 pg protein/ml tested. In the presence of excess unlabeled lipoproteins (2.0 mg protein/ml), the saturable component of the curve was abolished and only the nonspecific binding of ‘251-labeled



120 Time



(minutes)

1. Time-course of human ‘251-labeled HDL binding to human HepG2 cell line. Cells were incubated in minimum essential medium (without bicarbonate) with 10 mM Hepes (pH 7.4) containing 5% lipoprotein-deficient human plasma and constant concentration of ‘251-labeled HDL (10 pg protein/ml). After incubation at 4’C for the indicated time, the radioactivity associated with the cells was determined. Values are mean + S.E. of triplicate incubations.

lipoproteins to the cells was observed (Fig. 2). The lipoproteins to specific binding of ‘25I-labeled HepG2 cells was calculated by subtracting the nonspecific binding from the total binding (in the absence of excess unlabeled lipoproteins). As shown in Fig. 2A, the specific binding of ‘251labeled LDL in the presence of lo-200-fold excess unlabeled ligand accounted for 6580% of the total binding. Under similar conditions, the contribution of specific binding of ‘251-labeled HDL to the total binding was 45-65s (Fig. 2B). Thus, under conditions where unlabeled lipoproteins essentially occupied all the high-affinity binding sites, the specific binding of ‘251-labeled LDL and i2’Ilabeled HDL was calculated to be on the average 73 and 55% of total bindings, respectively. Uptake and degradation of “‘I-labeled HDL To ascertain the possible role of deiodinase in the formation of free ‘251- during the degradation of ‘251-labeled lipoproteins [38,39], 3-iodotyrosine, an inhibitor of deiodinase activity [39], was added to the incubation mixture. The results showed that the addition of 3-iodotyrosine at 0.05 or 0.5 mM concentrations had no effect on the recovery of radioactivity in the peroxidation-resistant fraction after 3-4 h incubation of ‘251-labeled HDL at 37°C. Even after 16 h incubation, the increase in

Fig. 2. Binding of human ‘251-labeled LDL (A) and ‘251-labeled HDL (B) to human HepG2 cell line as a function of concentration. Cells were incubated under conditions described in Fig. 1 with increasing concentrations of ‘251-labeled LDL (A) in the absence (0) or presence (0) of 2 mg unlabeled LDL protein/ml or with increasing concentration of ‘251-labeled HDL (B) in the absence (0) or presence (0) of 2 mg unlabeled HDL protein/ml. After 2 h incubation at 4°C. the radioactivity associated with the cells was determined. The high-affinity binding (A) was calculated by subtracting the nonspecific binding (0) from the total binding (0). Values are mean of triplicate incubations.

the radioactivity of this fraction was only 7-15%. Therefore, it seems more accurate to express the rate of degradation in terms of iodide-free acidsoluble radioactivity than in terms of total acidsoluble 12’I. Since even after 16-19 h incubation, the concentration of deiodinase to the free 1251formation is relatively small, the inclusion of lz51might result in the overestimation of the HDL degradation. Binding, internalization and degradation of ‘251-labeled HDL by HepG2 cells are shown in Figs. 3 and 4. When cells were incubated with increasing concentrations of “‘I-labeled HDL for

‘251-Lab&d HDL concentration

(9g

protein/ml)

Fig. 3. Binding, internalization and degradation of ‘251-labeled HDL by HepG2 cells as a function of concentration. After incubation of cells for 3 h at 37°C in minimum essential medium containing 5% lipoprotein-deficient plasma and increasing concentrations of 12s1-labeled HDL, the binding (0). internalization (0) and degradation (A) of HDL were measured. Degradation is expressed as total trichloroacetic acid-soluble radioactivity. Values are mean of triplicate incubations.

B

‘251-labeled

HDL Concentration

(yg

protein/ml)

Cells were incubated under conditions Fig. 4. Degradation of ‘2SI-labeled HDL by HepG2 cells as a function of concentration. was determined as described in Fig, 3 for either 3 h (A) or 16 h (B) with increasing concentrations of ‘2sI-labeled HDL. Degradation iodide-free trichloroacetic acid-soluble radioactivity. Values are mean k S.E. of triplicate incubations.

105

TABLE

I

EFFECTS OF CHLOROQUINE AND COLCHICINE THE DEGRADATION OF “‘I-LABELED HDL

ON

Cells were preincubated with medium containing 5% lipoprotein-deficient plasma and the indicated concentrations of chloroquine and colchicine for 2 h. After addition of fresh medium ‘251-labeled HDL (7.0 pg protein/ml), the incubacontaining tions were continued for 3 h in the presence or absence of chloroquine and colchicine. Degradation is expressed as total acid-soluble radioactivity. Values are mean + S.E. of triplicate incubations. Numbers in parentheses are percent of control values. Addition

Concentration (PM) _

None Chloroquine

50

Chloroquine

100

Colchicine

‘251-labeled HDL degraded (ng/mg cell protein) 519.3 + (100.0) 388.8 + (74.8) 210.9+ (40.6) 331.3 * (63.8)

1

34.6 32.2 6.4 15.7

4A) or 16 h (Fig. 4B) also displayed a two-component saturation curve. The nonsaturable nature of HDL uptake and degradation at 37°C confirms results obtained at 4°C indicating both the highand low-affinity binding of HDL to HepG2 cells. The roles of microtubules and lysosomes in the degradation of HDL were determined by studying the effects of colchicine, an inhibitor of microtubular polymerization [40], and chloroquine, a general inhibitor of lysosomal hydrolysis [41]. As shown in Table I, degradation of ‘251-labeled HDL, measured as total acid-soluble radioactivity after 3 h incubation, was decreased 25 and 60% in the presence of 50 I_IM and 100 PM chloroquine, respectively. Furthermore, colchicine at 1 PM caused a 36% decrease in ‘251-labeled HDL degradation (Table I). Similar effects were observed when the degradation of ‘251-labeled HDL was expressed in terms of iodide-free acid-soluble ‘251. The effect of lipoproteins

on the cellular cholesterol

metabolism

3 h at 37°C binding and internalization of “‘Ilabeled HDL did not reach a plateau and continued to increase up to 200 pg protein/ml tested. The degradation of ‘25I-labeled HDL expressed both as total acid-soluble 125I (Fig. 3) and as iodide-free acid-soluble radioactivity after 3 h (Fig. TABLE EFFECT

In contrast to an increased cholesterol accumulation in HepG2 cells incubated with LDL, addition of HDL to the medium had no effect on the total cholesterol accumulation in these cells (Table II). The lack of an HDL effect on the cellular content of cholesterol, in contrast to that of LDL,

II OF HUMAN

LDL AND HDL ON CHOLESTEROL

CONCENTRATION

OF HepG2

CELLS

After 30 h (in LDL experiment) or 40 h (in HDL experiment) preincubation with medium containing 5% lipoprotein-deficient plasma, fresh media containing LDL or HDL at the indicated concentrations were added and cells were incubated for an additional 16 h. Cells were washed with phosphate-buffered saline and free cholesterol (FC), cholesteryl esters (CE) and total cholesterol (TC) were determined by gas-liquid chromatography. Values for cellular cholesterol are averages of duplicate incubations. Lipoprotein density class

Lipoprotein Protein

concentration FC

(total pg added) CE

_

_

Cellular cholesterol TC

(W)

FC

CE

TC

100.0

100.0

100.0

None

_

LDL

31 61 122 306

20.9 41.5 83.1 207.6

64.4 127.9 255.7 639.4

59.3 117.7 235.4 588.5

125.9 134.0 130.6 136.7

192.2 194.8 252.3 330.6

135.7 143.0 148.5 165.3

HDL

23 47 117 233 467 1167

1.6 3.3 8.2 16.3 32.6 81.6

9.2 18.5 46.2 92.5 185.0 462.4

7.1 14.3 35.7 71.4 142.8 357.1

100.3 106.2 103.4 98.7 99.7 99.8

103.0 105.4 90.5 93.1 104.1 120.8

100.3 105.8 101.3 97.6 99.5 102.7

106

TABLE

the addition of HDL resulted in a 21% increase in cellular cholesteryl esters content. Degradation of LDL by HepG2 cells which markedly increased the cholesterol content of cells (Table II) also caused a stimulation of acylCoA : cholesterol acyltransferase and suppression of HMG-CoA reductase activities [20]. To determine if the activities of these two enzymes are influenced by HDL, cells were incubated for 1X h with increasing concentrations of HDL, and 3HZ0 incorporation into cellular sterols and [1-‘4C]oleate conversion into cholesteryl esters were measured. As shown in Table III, HDL had no significant effect on the de novo synthesis of sterols and, at the highest concentration tested, only caused a 15% increase in “C-labeled cholesteryl ester formation. These results indicate a lack of any substantial regulation of HMF-CoA reductase and marginal stimulation of acyl-CoA : cholesterol acyltransferase activities, respectively, by HDL.

III

EFFECT OF HDL ON THE DE NOVO SYNTHESIS STEROLS AND CHOLESTEROL ESTERIFICATION

OF

After a 36 h preincubation in medium containing 5% lipoprotein-deficient plasma, fresh medium and the indicated concentrations of HDL were added and incubations were continued for 18 h. Cells were then incubated either for 2 II with 1 mCi of 3H,0 or for 4 h with [I-“‘Cjoleate bound to albumin, and the incorporation of tritium into sterols and [l-‘4C]oieate conversion into cellular cholesteryl esters were measured. Values are means + S..E. of triplicate incubations. HDL added (pg protein/ ml medium)

0 10 20 50 100 200 300

3H,0 incorporation into sterols @pm/m8 cell protein)

2ooi:I2 2091 21ort: 196zk 205f 221+ 218rt19

[l-‘4C]Oleate conversion into cholesteryl esters (nmol/mg cell protein) 1.71 fO.10 1.69kO.17 1.78 k 0.09 1.85rtO.06 1.95 It 0.04 1.98 rt 0.06 1.97 * 0.05

7 5 9 5 8

C~nlpet~t~~n by unlabeled human Ii~o~ro~e~n~ for binding and deg~ffd~~io~ of ‘~‘~-i~be~ed HL)L by Hep G2 cells In order to determine whether HDL receptors are regulated by the uptake of HDL, HepG2 cells were preincubated with unlabeled lipoproteins for 24 h prior to the addition of ‘251-labeled HDL. As shown in Table IV, prior exposure of cells to HDL had no significant effect on the binding at 4°C or degradation at 37°C expressed either as total

was not due to the amount of cholesterol provided by these lipoproteins, As shown in Table II, LDL at the lowest cholesterol concentrations tested (59-118 I-18) doubled the cellular cholesteryl esters. However, similar levels of cholesterol (71-143 pg) provided in the form of HDL had no effect on the cellular content of cholesterol. Only when tested at the highest concentration (357 pg total cholesterol) TABLE

IV

EFFECT OF NORMAL AND MODIFIED DEGRADATION OF ‘=I-LABELED HDL

UNLABELED

HIGH

DENSITY

LIPOPROTEINS

ON

THE

BINDING

AND

After a 24 h preincubation with minimum essential medium (MEM) containing 5% lipoprotein-deficient (LPD) plasma with or without unlabeled HDL (200 pg protein/ml), medium was removed, cells were washed with phosphate-buffered saline and fresh medium and ‘2SI-labeled HDL (10 pg protein/ml) were added. The binding of ‘aSI-HDL was measured after a 2 h incubation at 4°C and degradation was measured after 4.5 h incubation at 37°C. Normal and modified (reductively methylated) HDL were added at 200 pg protein/ml. Values are mean i S.E. of triplicate incubations. Preincubation medium

MEM + 5% LPD-plasma MEM + 5% LPD-ptasma + unlabeled HDL MEM + 5% LPD-plasma MEM + 5% LPD-plasma

Unlabeled in the incubation

lipoproteins medium

‘2s1-labeled

None

255.2

None HDL modified

272.1 i4.8 127.5 _t1.6 158.22+ 1.1

HDL

HDL (ng/mg

cell protein)

Degraded

Bound

t4.1

Total “sl

Iodide-free

661.6+107.0

49.9 f 0.9

623.7 * 35.9 311.1& 42.2 308.6 k 48.9

49.9*1.2 33.9+1.9 32.8 f 1.4

‘*‘I

107

TABLE

V

CORRELATION TORY EFFECT

BETWEEN APOLIPOPROTEIN CONTENT ON 12’1-LABELED HDL DEGRADATION

OF LIPOPROTEIN

DENSITY

CLASSES

AND THEIR

INHIBI-

After a 24 h preincubation in medium containing 5% lipoprotein-deficient plasma, fresh medium containing unlabeled lipoproteins of medium) were added and degradation of ‘251-labeIed HDL was (200 pg protein/ml) and ‘251-labeled HDL (9.5 pg protein/ml measured after a 3.5 h incubation at 37°C. Degradation is expressed in terms of total acid-soluble radioactivity. Values for ‘25 I-labeled HDL degradation are the mean k S.E. of triplicate incubations of two different experiments. Lipoprotein density class

Apolipoproteins A-I

A-II

(Total pg added) B

C-I

C-II

C-III

D

E

F

None VLDL LDL HDL

2.64 3.33 3.09 81.15

0.00 0.55 0.48 40.73

0.00 47.09 101.97 6.55

0.00 4.41 2.28 3.39

0.00 8.68 0.04 0.51

0.00 15.97 0.44 1.92

0.00 1.65 0.52 3.84

0.00 7.16 0.73 1.42

0.00 0.00 0.33 0.86

acid-soluble or iodide-free acid-soluble ‘251. In addition, reductive methylation of unlabeled HDL did not abolish its ability to compete with ‘251labeled HDL for binding at 4°C or degradation at 37°C (Table IV). To study the effect of unlabeled VLDL, LDL and HDL on the degradation of ‘251-labeled HDL (9.5 pg protein/ml), cells were incubated for various time intervals at 37°C. All three lipoprotein density classes, added at the same protein concentration, competed similarly with 125I-labeled HDL for degradation, irrespective of whether they were incubated for 3.5 h (Table V) or 20 h (data not shown). The results expressed in terms of iodide-free acid-soluble 1251 were similar to those measured as total acid-soluble radioactivity shown in Table V. None of the neutral lipid constituents of major lipoprotein density classes correlated with their inhibitory effect on the degradation of ‘251-labeled HDL. Likewise, no correlation existed between the levels of apolipoproteins A-I, A-II, B, C-II, C-III, E and F in VLDL, LDL and HDL and the inhibitory effect of these lipoprotein density classes on the degradation of ‘251-labeled HDL. There was, however, a possible association between the concentration and inhibitory effect of apolipoprotein D; the increasing contents of apolipoprotein D in LDL, VLDL and HDL corresponded positively with the increasing inhibitory effect of these lipoprotein density classes (Table V). A similar, albeit less pronounced, effect was also exhibited by apolipoprotein C-I.

‘25 I-labeled HDL degraded (ng/mg cell protein) 640.79 k 40.35 305.12 +40.27 388.50 f 29.22 279.16k31.63

Discussion The results of this study have demonstrated that the human hepatoma cell line HepG2 is capable of binding and degrading human HDL. However, the catabolism of HDL by this cell line seems to be mediated through processes different from those characteristic of the HDL degradation by extrahepatic cells [5,42,53] or those involved in the hepatic removal of LDL [20,21]. When studied in the presence of excess unlabeled HDL, 55% of ‘251-labeled HDL was bound to HepG2 cells through specific, high-affinity receptors. When determined under identical experimentally conditions, the high-affinity binding of LDL to HepG2 cells accounted for 73% of the total binding capacity. A lower specific binding of LDL previously reported from this laboratory [20] was due most probably to the use of a lesser amount of excess unlabeled LDL than in the present study (50-fold vs. 200-fold). Similarly to LDL, the degradation of HDL apparently takes place in lysosomes, since chloroquine inhibited its degradation by 60%. Comparable results were reported in studies with hepatocytes from other species [11,12,39]. Colchicine, which was shown to inhibit the degradation of HDL by rat hepatocytes [44], also inhibited the degradation of human HDL by HepG2 cells. Whereas in extrahepatic cells HDL has been shown to stimulate the efflux of cholesterol [4-6,42,45], LDL seems to have only a slight effect

108

on this process [45]. In the present study, there was practically no change in the cellular cholesterol content of HepG2 cells exposed to HDL for 16 h. On the other hand, the exposure of HepG2 cells to LDL, at similar cholesterol concentrations, caused a marked accumulation of cellular cholesterol (Table II). This finding is in agreement with a report by Rothblat et al. [46] who have shown that incubation of rat hepatoma cells with normal HDL for 24 h had no effect on the cellular content of cholesterol. However, the addition of cholesterolenriched HDL to the medium resulted in the accumulation of cholesterol, primarily in the form of cholesteryl esters [46]. Although in the present study the cholesterol concentration of LDL was greater than that of HDL (Table II), at similar cholesterol concentrations, LDL caused a 100% increase in cellular cholesteryl esters while HDL was without any effect. Thus, the inability of HDL in modulating the cholesterol content of HepG2 cells cannot be entirely due to its cholesterol content. At the highest concentration of HDL tested (1167 pg protein), which resulted in a 21% increase in the cellular content of cholesteryl esters, the amount of added apolipoprotein B was 65 pg. However, the addition of a similar amount of apolipoprotein B (62 pg) in the form of LDL resulted in a 95% increase of cellular cholesteryl esters (Table II). Although the effect of high levels of HDL on the concentration of cellular cholesteryl esters may be due to its content of apolipoprotein B, dissimilar effects of LDL and HDL at equal concentrations of apolipoprotein B may possibly be explained by different lipoprotein forms of this apolipoprotein in LDL and HDL. In LDL, apolipoprotein B is present mainly in the form of lipoprotein B [47], while in HDL it occurs as lipoprotein (a) and lipoprotein B [48,49]. O’Malley et al. [7] have shown that, in rabbit hepatocytes, the influx of labeled free cholesterol from HDL was greater than that from LDL; similarly, the efflux of labeled cellular cholesterol was greater in the presence of HDL than LDL. It has been suggested [46,50] that, in many instances, the flux of free cholesterol from lipoproteins into the cells is balanced by its molecular transfer in the opposite direction, resulting in no net change in the cholesterol concentration; such a condition

may exist in the HepG2 cells. The lack of any substantial effect of HDL on the cholesterol content was verified by the observation that this lipoprotein neither promoted nor impaired cellular sterol synthesis, and at the highest concentration tested increased the formation of 14C-labeled cholesteryl esters by only 15%. The absence of this end-point regulation of HDL catabolism in HepG2 cells, under experimental conditions which favor such regulatory mechanism in fibroblasts and smooth muscle cells [5,42], might be due to an exchange reaction resulting in neither the loss nor the accumulation of cellular cholesterol [46]. This equilibrium may possibly be maintained by immediate utilization of HDL-cholesterol as a source for bile acid production and/or lipoprotein formation. It has been shown, for example, that the HDL-free cholesterol is preferentially utilized over LDL-free cholesterol for bile acid synthesis in man [51] and in squirrel monkeys [52]. In the present study, preincubation of cells with unlabeled HDL did not suppress the binding of ‘251-labeled HDL and only had a marginal effect on its degradation. The specificity of ‘251-labeled HDL binding and degradation in HepG2 cell was further evaluated by examining the effects of unlabeled lipoproteins on its catabolism. Although HDL seems to be the most effective inhibitor of ‘251-labeled HDL degradation, VLDL, LDL and reductively methylated HDL also competed efficiently with the labeled lipoprotein for binding and degradation. Thus, the binding and degradaHDL apparently proceed tion of ‘251-labeled through sites that are not sensitive to preincubation with unlabeled HDL, do not require intact lysine residues in the protein moiety and recognize VLDL and LDL as well. The role of apolipoproteins in regulating the binding and degradation of lipoproteins has received much attention in recent years. Results of several studies [53-561 have indicated that apolipoprotein C peptides inhibit, while apolipoprotein E promotes hepatic uptake of lipoproteins. The liver appears to possess an apolipoprotein E receptor which mediates the removal of chylomicron remnants [53,55,56] and canine HDL [53,57]. In the present study, one of the most significant and unexpected findings was the efficiency of VLDL and LDL in inhibiting the ‘251-labeled HDL de-

109

gradation. The VLDL preparation contained very small amounts of apolipoproteins A-I or A-II and yet caused a marked inhibition of ‘251-labeled HDL degradation. On the other hand, the HDL was the most effective inhibitor of the ‘251-iabeled HDL degradation despite its relatively small content of apolipoprotein B. Similarly, the higher contents of apolipoproteins C-II, C-III and E in VLDL or HDL were not commensurate with the inhibitory activity of this lipoprotein density class. The present finding that apolipoprotein A-I does not appear to be the recognition signal for HDL binding is in contrast to that reported for the rat. Binding and catabolism of HDL in intact rat [58], liver plasma membranes [59] and isolated hepatocytes [60] have been shown to be mediated through apolipoprotein A-I as the recognition signal for the receptor. However, our results are consistent with those reported by Bachorik et al. [39] who found that, in cultured porcine hepatocytes, the apolipoprotein A-I-free LDL and the apolipoprotein B-free HDL were equally efficient in competing with ‘251-labeled HDL for uptake and degradation. The finding that apolipoprotein E does not appear to play a significant role in the binding and degradation of human HDL by HepG2 cells agrees with similar results reported for the HDL metabolism in the rat [58,60] and porcine hepatocytes [39]. There was, however, some correlation between the apolipoprotein D and C-I contents of VLDL, LDL and HDL and their inhibitory effect on r2’I-labeled HDL degradation. Despite the indication that cells might recognize HDL by apolipoprotein D or apoiipoprotein C-I, further studies are required to elucidate the nature of recognition signal(s) for hepatic HDL catabolism. The main aim of such studies will be to identify lipoprotein forms of various apolipoproteins and to establish their effect on cellular processes. In summary, the present study has demonstrated that human hepatoma cell line HepG2 binds an degrades human HDL. Approx. 55% of HDL binding is mediated by high-affinity receptors different from those involved in the catabolism of LDL [20]. In contrast to LDL, the binding and uptake of HDL has no effect on the cellular cholesterol content and, although the degradation occurs in lysosomes, this process does not lead to any significant regulation of cholesterol metabo-

lism. Previous studies have indicated apolipoprotein B as the recognition signal for the specific binding and uptake of LDL particles f20]. The present findings seem to favor apolipoproteins D or C-I rather than A-I and E as the possible recognition signal(s) for the specific binding and degradation of HDL. However, due to the marked lipoprotein particle heterogeneity of HDL, further studies are required elucidate the exact nature of recognition signal(s). It also remains to be established what role, if any, HDL-cholesterol plays in the production of bile acids and nascent lipoproteins. The similarity between the HepG2 cells and rabbit and porcine hepatocytes [7,39] in binding and degrading HDL renders this cell line a suitable model for studying the metabolism of human lipoproteins in a homologous system.

We thank MS Carmen Quiroga, Ms. Melinda Heffron, Mr. J. Fesmire and Mr. R. Whitmer for their skillful technical assistance and Ms. Anne Harris for her secretarial assistance. This investigation was supported in part by research grant HL23181 from the United States National Institutes of Health and by the resources of the Oklahoma Medical Research Foundation. References 1 Brown, M.S. and Goldstein, J.L. (1976) Science 191, 150-154 2 Goldstein, J.L. and Brown, MS. (1977) Ann. Rev. Biochem. 46, 897-930 3 Glomset, J.A. and Norum, K.R. (1973) Adv. Lipid Res. 11, 1-65 4 Stein, Y., Glangeaud, M.C., Fainaru, M. and Stein, 0. (197s) Biochim. Biophys. Acta 380, 106-118 5 Stein, O., Vanderhoek, J. and Stein, Y. (1976) Biochim. Biophys. Acta 431, 347-358 6 Ho, Y.K., Brown, M.S. and Goldstein, J.L. (1980) J. Lipid Res. 21, 391-398 7 O’Mafley, J.P., Soltys. P.A. and Portman, O.W. (1981) J. Lipid Res. 22, 1214-1224 8 Roheim, P.S., Rachmilewitz, D., Stein, 0. and Stein, Y. (1971) Biochim. Biophys. Acta 248, 315-329 9 Rachmilewitz, D., Stein, O., Roheim. P.S. and Stein, Y. (1972) Biochim. Biophys. Acta 270, 414-425 10 Eisenberg, S., Windmuller, H.G. and Levy, R.I. (1973) J. Lipid Res. 14, 446-458 11 Nakai. T., Otto, P.S., Kennedy, D.L. and Whayne, T.F., Jr. (1976) J. Biol. Chem. 251,4914-4921

110

12 Ose, L., Ose, T., Norum, K.R. and Berg, T. (1979) Biochim. Biophys. Acta 574, 521-536 13 Van Berkel, T.J.C., Kruijt, J.K., Van Gent, T. and Van Tol, A. (1980) Biochem. Biophys. Res. Commun. 92, 1002-1008 14 Van Tol, A., Van Gent, T., Van? Hooft, F.M. and Vlaspolder, F. (1978) Atherosclerosis 29, 439-448 15 Sigurdsson, G., Noel, S.-P. and Have], R.J. (1979) J. Lipid Res. 20, 316-324 16 Aden, D.P., Fogel, A., Plotkin, S., Damjanov, I. and Knowles, B.B. (1979) Nature 282, 615-616 17 Rash, J.M., Rothblat, G.H. and Sparks, C.E. (1981) Biochim. Biophys. Acta 666, 294-298 18 Zannis, V.I., Breslow, J.L., San Giacomo, T.R., Aden, D.P. and Knowles, B.B. (1981) Biochemistry 20, 7089-7096 19 Dashti, N., Wolfbauer, G., Koren, E., Knowles, B.B. and Alaupovic, P. (1983) Arteriosclerosis 3, 498a 20 Dashti, N., Wolfbauer, G., Koren, E., Knowles, B. and Alaupovic, P. (1984) Biochim. Biophys. Acta 794, 373-384 21 Havekes, L., Van Hinsbergh, V., Kempen, J. and Emeis, J. (1983) Biochem. J. 214, 951-958 22 Alaupovic, P., Lee, D.M. and McConathy, W.J. (1972) B&him. Biophys. Acta 260, 689-707 23 Markwell, M.A.K. (1982) Anal. Biochem. 125, 427-432 24 Weisgraber, K.H., Innerarity, T.L. and Mahley, R.W. (1978) J. Biol. Chem. 253, 9053-9062 25 Kuksis, A., Myher, J.J., Marai, L. and Geher, K. (1975) J. Chromatogr. Sci. 13, 423-430 26 Curry, M.D., Alaupovic, P. and Suenram, C.A. (1976) Clin. Chem. 22, 315-322 27 Curry, M.D., Gustafson, A., Alaupovic, P. and McConathy, W.J. (1978) Clin. Chem. 24, 280-286 28 Curry, M.D., McConathy, W.J., Fesmire, J.D. and Alaupovic, P. (1981) Clin. Chem. 27, 543-548 29 Curry, M.D., McConathy, W.J., Fesmire, J.D. and Alaupovic, P. (1980) Biochim. Biophys. Acta 617, 503-513 30 Curry, M.D., McConathy, W.J. and Alaupovic, P. (1977) Biochim. Biophys. Acta 491, 232-241 31 Curry, M.D., McConathy, W.J., Alaupovic, P., Ledford, J.H. and Popovic, M. (1976) Biochim. Biophys. Acta 439, 413-425 32 Koren, E., McConathy, W.J. and Alaupovic, P. (1982) Biochemistry 21, 5347-5351 33 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 34 Stein, 0.. Weinstein, D.B., Stein, Y. and Steinberg, D. (1976) Proc. Natl. Acad. Sci. USA 73, 14-18 35 Bierman, E.L., Stein, 0. and Stein, Y. (1974) Circ. Res. 35, 136-150 36 Goldstein, J.L., Dana, S.E. and Brown, M.S. (1974) Proc. Natl. Acad. Sci. USA 71, 4288-4292.

37 Dashti, N., Kelley, J.L., Thayer, R.H. and Ontko, J.A. (1983) J. Lipid Res. 24, 368-380 38 Pangburn, S.H., Newton, R.S., Chang, C.-M., Weinstein, D.B. and Steinberg, D. (1981) J. Biol. Chem. 256,3340-3347 39 Bachorik, P.S., Franklin, F.A., Virgil, D.G. and Kwiterovich, P.O., Jr. (1982) Biochemistry 21, 5674-5684 40 Borisy, G.G. and Taylor, E.W. (1967) J. Cell Biol. 34, 525-533 41 Seglen, P.O., Grinde, B. and Solheim, A.E. (1979) Eur. J. Biochem. 95, 215-225 42 Biesbroeck, R., Oram, J.F., Albers. J.J. and Bierman, E.L. (1983) J. Clin. Invest. 71, 525-539 43 Oram, J.F., Brinton, E.A. and Bierman, E.L. (1983) J. Clin. Invest. 72, 1611-11621 44 Ose, L., Roken, I., Norum, K.R. and Berg, T. (1980) Exp. Cell Res. 130, 127-135 45 Henriksen, T., Evensen, S.A., Blomhoff, J.P., Torsvik, H. and Carlander. B. (1979) Biochim. Biophys. Acta 574, 312-320 46 Rothblat, G.H., Arbogast, L.Y. and Ray, E.K. (1978) J. Lipid Res. 19, 350-358 47 Lee, D.M. and Alaupovic, P. (1974) Biochem. J. 137, 155-167 48 Simons, K., Ehnholm, C., Renkonen. 0. and Bloth, B. (1970) Acta Path. Microbial. Stand. Section B 78, 459-466 49 Ehnholm, C., Garoff, H., Simons, K. and Aro, H. (1971) B&him. Biophys. Acta 236, 431-439 50 Bruckdorfer, K.R. and Green, C. (1967) Biochem. J. 104, 270-277 51 Schwartz, CC., Halloran, L.G., Vlahcevic, Z.R., Gregory, D.H. and Swell. L. (1978) Science 200, 62-64 52 Portman, O.W., Alexander, M. and O’Malley, J.P. (1980) Biochim. Biophys. Acta 619, 545-558 53 Windier, E., Chao, Y.-S. and Havel, R.L. (1980) J. Biol. Chem. 255, 5475-5480 54 Windier, E., Chao, Y.-S. and Havel, R.J. (1980) J. Biol. Chem. 255, 8303-8307 55 Sherrill, B.C., Innerarity, T.L. and Mahley, R.W. (1980) J. Biol. Chem. 255, 1804-1807 56 Shelburne, F., Hanks, J., Meyer, W. and Quarfordt, S. (1980) J. Clin. Invest. 65, 652-658 57 Mahley, R.W., Weisgraber, K.H. Innerarity, T.L. and Windmueller, H.G. (1979) Proc. Nat]. Acad. Sci. USA 76, 1746-1750 58 Sparks, C.E., Tennenberg, S.D. and Marsh, J.B. (1981) B&him. Biophys. Acta 665, 8-12 59 Chacko, G.K. (1982) Biochim. Biophys. Acta 712, 129-141 60 Rifici, V.A. and Eder, H.A. (1983) Fed. Proc. 42, 1059