Download - Regulation of cholesterol esterification and biosynthesis in monolayer cultures of normal adult rat hepatocytes

THE J O U R N A I . OF ~ I O L O G I C A I . CHEMISTRY Vol. 255. No. 19. Issue of October IO, pp. 9128-9137, 1980 Prrnled in I 1 S A

Regulation of Cholesterol Esterification and Biosynthesis in Monolayer Cultures of Normal Adult Rat Hepatocytes*

(Received for publication, May 25, 1979)

Christian A. Drevon,S David B. Weinstein, and Daniel Steinberg From the Division of Metabolic Disease, Department of Medicine, University of California, San Diego, School of Medicine, La Jolla, California 92093

Adult rat parenchymal liver cells were isolated and cultured in monolayers. Cholesterol esterification in the intact cultured cells was determined by measuring incorporation of tritiated oleic acid into cell cholesterol ester. Addition of 10 pg/ml of 25-hydroxycholestero1 to the medium gave a 3- to 6-fold increase in cholesterol esterification, while the incorporation of oleic acid into phospholipids and triglycerides remained unaltered. Similar stimulation of cholesterol esterification by 25- hydroxycholesterol was also found if [‘‘C]meva1ono1ac- tone or [3H]cholesterol (the latter presented to the cells in high density lipoproteins) were used as precursors. The stimulatory effect of 25-hydroxycholestero1 was maximal after only 15-min incubation and was inde- pendent of protein synthesis. After 4 to 6 h of incubation with 25-hydroxycholesterol, its stimulatory effect was reduced significantly, and after 18 h of incubation no stimulation was observed. Thus, the liver cells in some fashion adapt to the continuing presence of 25- hydroxycholesterol.

Isolated microsomes prepared from cells previously incubated with 25-hydroxycholesterol showed acyl- CoAxholesterol acyltransferase activity 3 times that of microsomes from control cells. Incubation of isolated microsomes with 25-hydroxycholestero1 increased acyl-CoAxholesterol acyltransferase activity 2-fold.

The net cellular content of ester cholesterol increased after 2 to 6 h incubation of hepatocytes with 25-hydroxycholesterol; there was a net decrease in cellular free cholesterol. Mevalonolactone (10 m ~ ) also stimulated cholesterol esterification and increased the cellular content of ester cholesterol (3- to 4-fold). The effective- ness of mevalonolactone did not diminish with longer periods of preincubation. Furthermore, the stimulatory effects of 25-hydroxycholesterol and mevalonolactone added together were at least 50 to 100% greater than the effects of either agent alone, suggesting that the mechanisms by which they increase cellular cholesterol esterification are different. Both 25-hydroxycholesterol and mevalonolactone rapidly inhibited microsomal 3- hydroxy-3-methylglutaryl-CoA reductase activity. Pure cholesterol had no effect on cellular cholesterol

* This investigation was supported by National Institutes of Health Research Grant HL 14197, awarded by the National Heart, Lung and Blood Institute, and a grant from the San Diego County Heart Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. + Recipient of a fellowship from the Royal Norwegian Council for Scientific and Industrial Research and a Public Health Service Inter- national Research Fellowship (TW 02609). Present address, Institute of Pharmacy, Department of Pharmacology, University of Oslo, P.O. Box 1068, Blindern, Oslo 3, Norway.

esterification or on 3-hydroxy-3-methylglutaryl-CoA reductase activity at concentrations up to 10 pg/ml.

It has been shown that liver cholesterol esterification (1,2) and cholesterol biosynthesis (3,4) are influenced significantly by cholesterol feeding and fasting. However, the interaction of multiple variables, notably the imposition of experimental perturbations on the diurnal variation in cholesterol biosynthesis, has made it difficult to assess the relative importance of factors regulating cholesterol metabolism (5). Study of control mechanisms using tissue slices or freshly isolated cells is difficult because some of the responses of interest take many hours to develop and these systems are not in a stable state for more than a few hours (6, 7).

The use of cell culture techniques has been fruitful in characterizing the regulation of cholesterol metabolism in peripheral cells, particularly skin fibroblasts. In fibroblasts there appears to be a reciprocal regulation of 3-hydroxy-3- methylglutaryl-CoA (HMG-CoA) reductase and acyl-CoA: cholesterol acyltransferase (ACAT) (8). A number of oxygenated sterols suppress HMG-CoA reductase activity (9) and stimulate acyl-CoA:cholesterol acyltransferase (10) in these cells. Analogous mechanisms may function in liver. It has been shown that oxygenated sterols decrease HMG-CoA reductase’ activity in hepatoma cells (ll), in fetal mouse liver cells in culture (13, in freshly isolated hepatocytes and in perfused liver (13). Furthermore, mevalonic acid or mevalonolactone have been shown to regulate rat liver HMG-CoA reductase activity both in vivo and in freshly isolated hepatocytes (13-15).

The present studies were undertaken primarily to explore the mechanisms involved in regulation of hepatic acyl-CoA: cholesterol acyltransferase. Using a system for maintaining adult rat hepatocytes in monolayer culture recently reported by this laboratory (16), it was possible to follow the time course of responses and to examine changes in cell content of free and esterified cholesterol. Changes in HMG-CoA reductase activity were followed concurrently.

MATERIALS AND METHODS

Chemi~als-[lcu,2a-~H]Cholesterol (specific activity, 43 Ci/mmok Amersham/Searle Corp., Chicago, IL) was purified by thin layer chromatography on Silica Gel G (Analtech, Inc., Newark, DE) in the developing system hexane/diethyl ether/acetic acid (80u):1, v/v/v).

methylglutaryl coenzyme A reductase; HDL, high density lipopro- ’ The abbreviations used are: HMC-CoA reductase, 3-hydroxy-3-

teins; FCS, fetal calf serum; DME medium, Dulbecco’s modified Eagle’s medium; Me2S0, dimethyl sulfoxide; TLC, thin layer chromatography; PBS, Dulbecco’s phosphate-buffered saline.

9128

Regulation of Cholesterol Metabolism in Liver Cells 9129

The cholesterol band contained over 99% of the radioactivity recovered.

[9,10-3H]Oleic acid (specific activity 8.26 Ci/mmol), [l-'4C]oleyl coenzyme A (specific activity 51.0 mCi/mmol), ~~-[2- '~C]mevalonolactone (specific activity 27.3 mCi/mmol), and [3-"C]hydroxymethylglutaric acid (specific activity 51.9 mCi/mmol) were obtained from New England Nuclear Corp. (Boston, MA). Routinely used chemicals were obtained from Sigma Chemical Co. (St. Louis, MO); p-sitosterol and 25-hydroxycholestero1 were from Steraloids (Wilton, NH). The purity of these sterols was checked by thin layer chromatography using either hexane/diethyl ether/acetic acid (80.20:1, v/v/ v), or n-heptane/ethyl acetate (L1, v/v). Purity was checked by gas- liquid Chromatography. p-Sitosterol, 25-hydroxycholestero1, and cholesterol were found to be =%, 988, and 98% pure, respectively. Tritiated 25-hydroxycholestero1 was prepared by exchange labeling (New England Nuclear Corp.) and repurified by repeated thin layer chromatography using the heptane/ethyl acetate system. Ninety- seven per cent of the radioactivity of the final preparation co-chromatographed with unlabeled 25-hydroxycholestero1. The other steroids tested were used as delivered from Sigma (ethinylestradiol, lanosterol, dexamethasone) or Steraloids (5-a-cholestane, 5,24-choles- tadiene-3b-01). All sterols were stored in 96% ethanol at +4OC or -2OOC. [3H]Oleic acid was mixed with unlabeled oleic acid, dissolved in ethanol, and converted to the potassium salt by adding an equi- molar amount of 1 N KOH. This was injected with a sterile syringe into a solution (previously sterilized and warmed to 37°C) containing 5 or 10% bovine serum albumin (fatty acid-free, Sigma Chemical Co.) dissolved in the standard medium used for hepatocyte culture (see below). The molar ratio of fatty acid to albumin was 1:l or 5 1 as indicated. Fetal calf serum (FCS) (Irvine Scientific Sales Co., Foun- tain Valley, CA) was heat-inactivated a t 56°C for 30 min before use. Penicillin G, streptomycin sulfate, and Dulbecco's phosphate-buffered saline (17) (PBS) were obtained from Grand Island Biological Co., Berkeley, CA.

Perfusion and Incubation Media-Rat livers were perfused with a modified calcium-free Hanks' solution (18) as previously described (16). An arginine-free Dulbecco's modified Eagle's medium (DME medium) (19) supplemented with glucose (40 mM), ornithine (0.4 mM), and 10 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (Hepes) from Grand Island Biological Co. (Santa Clara, CA) was used as the standard incubation medium for the liver cells. Collagenase (type 1) was purchased from Sigma (St. Louis, MO) and 60-mm diameter Contur culture dishes from Lux, Inc. (Newberg Park, CA).

Isolation a n d Cell Culture of Hepatocytes-The isolation of rat hepatocytes was based upon the method described by Berry and Friend (20) except that sterile conditions were maintained throughout. The detailed procedures have been described previously (16). The yield of hepatocytes was 2 to 4 X 10' cells/liver with 92 to 96% trypan blue (0.04%) dye exclusion. Cells were plated in dishes at 3 X lo6 cells/dish in 3 ml of arginine-free DME medium containing 20% FCS, insulin (10 pg/ml), penicillin C (100 units/ml), and streptomycin sulfate (100 pg/ml), and were maintained at 39°C in a 95% air, 5% CO, atmosphere. Cells were incubated 24 h under these conditions before changing the medium to the standard medium without serum. In some experiments the freshly isolated cells were placed directly into serum-free medium and used immediately. Morphological and biochemical characterization of these hepatocyte cultures has been reported (16). Each dish contained 1.5 to 2.5 mg of cell protein.

Preparation of Homogenates-The hepatocyte cultures were washed three to six times with ice-cold phosphate-buffered saline and the cells were scraped off the plates with a rubber policeman. The plates were then washed twice with 2 ml of the same buffer. The cells were sedimented a t 50 X g for 5 min and resuspended in 0.25 M sucrose. The suspension, chilled on ice, was sonicated three times for 10 s a t maximum power using a micro ultrasonic cell disrupter (Kontes), after which no intact cells were visible on microscopic examination. Samples for protein assay were taken and the homoge- nate was then centllfuged at 12,000 X g for 20 min a t 0°C. The supernatant was centrifuged a t 100,OOO X g for 60 min a t 0°C. The resulting microsomal pellet was resuspended in 0.1 M phosphate buffer, pH 7.4, containing fatty acid-free bovine serum albumin (1

Measurement of Acyl-CoA:cholesterol Acyltransferase Activity in Microsomes a n d in Intact Cells-The assay for this enzyme in microsomes was performed as described by Hashimoto et al. (21) in aortic microsomes except that assays were limited to 5 min since we found that with liver microsomes there was significant departure from linearity beyond 10 min. The incubation mixture included: 250 pl of

mg/ml).

0.1 M phosphate buffer, pH 7.4, containing albumin (1 mg/ml); 50 to 100 pg of microsomal protein in 25 p1 of the same buffer; 10 pI of [ l - "C]oleyl-CoA (about 100,000 cpm; final concentration, 5 0 ~ ~ ) . Studies of activity as a function of substrate concentration showed that maximal activity was reached at about 40 p~ oleyl-CoA. The incubation was carried out a t 37°C and the reaction was stopped a t 5 min by adding 20 volumes of chloroform/methanol(2:1, v/v), as described by Folch et al. (22).

Cholesterol esterification in intact hepatocytes was measured by adding to the culture dish either [3H]oleic acid bound to bovine serum albumin as described above or [2-'4C]mevalonolactone dissolved in ethanol. The incorporation of precursor was linear for at least 60 min under the conditions used and this was the standard incubation time used in subsequent studies. The cells were placed on ice, washed three times with phosphate-buffered saline, scraped off with a rubber policeman, and centrifuged as described above. The cell pellet was resuspended in 500 pl of distilled water and sonicated. A sample for protein determination was taken, and the remaining cells were ex- tracted with chloroform/methanol (2:1, v/v). After phase separation, the chloroform phase was washed with water and taken to dryness. The residual was redissolved in 200 p1 of hexane, applied to Silica Gel H TLC plates, and developed in hexane/diethyl ether/acetic acid (80: 201, v/v/v) (System A). Cholesterol esters, triglycerides, free cholesterol, and phospholipids were visualized using iodine vapor and scraped into counting vials containing 10 ml of a toluene-base liquid scintillation fluid. Methanol ( 8 0 pl) was added to the vials containing phospholipids to release the phospholipids from the silica gel. Quench- ing was evaluated using an internal standard. Overall recovery of labeled cholesterol and triglycerides through the above procedures was 90% and for phospholipid was 80 to 85%. Since monoglycerides (RF = 0.05 to 0.10) may not always be sufficiently well separated from the phospholipids at the origin in this system some of the lipid extracts were also run in diethyl ether/benzene/ethanoI/acetic acid (4050:2:0.2, System B) which provides better separation of these two lipid classes (monoglyceride RF = 0.20). Radioactivity in the monoglyceride fraction did not exceed 2% of the total amount of [ 'HI- glycerol incorporated into total lipids. In experiments utilizing ["C]- mevalonolactone to measure cholesterol ester production, the lipid extract was in some cases separated by TLC using System A run up to 4 cm from the top of the plate, followed by a second development in hexane to the top of the plate. Using this system, in which squalene (RF = 0.95) was totally separated from cholesterol esters ( R F = 0.8 to 0.85), essentially similar results were obtained. The rate of cholesterol esterification was calculated by dividing the amount of [ 'Hloleic acid or ['4C]mevalonolactone incorporated into esters by the specific radioactivity of the precursor in the medium. This value, of course, represents only the minimum rate of ester formation, neglecting intracellular dilution of the tracer.

Measurement of HMG-CoA Red~ctase-[3-'~C]Hydroxymethyl- glutaric acid anhydride was prepared according to the method of Goldfarb and Pitot (23). [3-I4C]HMG-CoA was formed by reacting the [3-'4C]hydroxymethylglutaric acid anhydride with reduced CoA on ice at pH 9.0, as described by Hilz et al. (24).

The microsomes used as enzyme source were prepared as described above, except that the isolation buffer contained 0.25 M sucrose, 0.3 mM EDTA, 50 mM potassium phosphate (pH 7.2), and 5 mM dithiothreitol. Microsomal protein (120 to 200 pg) in 100 pI of this buffer was preincubated 10 min a t 37°C. The final reaction mixture contained glucose 6-phosphate (30 mM), glucose-6-phosphate dehydro- genase (3.5 units/ml), NADP' (2.55 mM), and [3-"C]HMG-CoA (140 pM, specific activity 4.5 pCi/pmol) in a volume of 200 pl. The reaction was linear for a t least 2 h and assays were routinely carried out for 30 to 60 min. The reaction was stopped by adding 30 pl of 12 N HCI, and 3 pmol of ['Hlmevalonic acid as carrier and internal standard. The tubes were then incubated for 30 min at 37°C or overnight at room temperature to complete lactonization of mevalonic acid and centrifuged to remove precipitated protein. Aliquots (100 pl) were applied to Silica Gel G (260 pm, Analtech) TLC plates that had been activated at 110°C overnight and the applied samples were dried thoroughly before development. The developing system used was dry acetone/ benzene (l:l, v/v). with anhydrous NaySOr present in the chamber. Mevalonolactone (RF = 0.5 to 0.7) was visualized by spraying with 1': IZ in methanol and was well resolved from HMG (RF = 0 to 0.1 ). After evaporation of the 12, the gel was scraped directly into scintillation vials. One hundred microliters of 1 M acetic acid and 10 ml of scintillation fluid (5% BSS Biosolv) were added. Blank values were obtained from incubations stopped at zero time.

Preparation of Lipoproteins-Preparative ultracentrifugation of

9130 Regulation of Cholesterol Metabolism in Liver Cells

human plasma lipoproteins was performed as described by Have1 et

and dialyzed four times against 100 volumes of a 0.1 M phosphate- al. (25). The lipoprotein fractions were washed by recentrifugation

buffered saline containing 0.3 mM EDTA, pH 7.4. The purity of the lipoproteins was verified by agarose-electrophoresis using the method described by Noble (26).

Chemical Analyses-Protein ,was measured according to the method of Lowry et al. (27) using bovine serum albumin as standard. Free cholesterol and ester cholesterol were determined by two alter- native methods. Gas-liquid chromatography was used to determine free cholesterol before and after saponification using p-sitosterol as an internal standard. A Varian Aerograph, series 2100, equipped with a &foot, 2-mm inside diameter column, packed with 3 8 SP-2250 on 80/100 mesh Supelcoport (Supelco) was used at 295°C with argon (30 ml/min) as carrier gas. Sterols were also measured enzymatically in some experiments by the method of Gamble et al. (28). The extraction of lipids for this assay was adapted from Bligh and Dyer (29). Ester cholesterol concentration was calculated from the difference between the free cholesterol content of untreated samples and of samples fist subjected to cholesterol esterase activity as described (28).

RESULTS

Specific Stimulation of Cholesterol Ester Formation by 25- Hydroxycholesterol-The effect of 25-hydroxycholestero1 on the incorporation of ["Hloleic acid into cellular cholesterol esters, phospholipids, and triglycerides is shown in Fig. 1. Both the labeled oleic acid and the unlabeled 25-hydroxycholesterol were added a t zero time and the incubation was terminated after 60 min. Even at concentrations as low as 1 pg/ml, 25-hydroxycholestero1 caused a highly significant 2- fold increase in ester formation. At 10 pg/ml, the increase was 3- to &fold. This stimulation of esterification was specific, i.e. incorporation of oleic acid into other cellular lipids (phospholipids and triglycerides) was not affected a t any concentration of 25-hydroxycholestero1 tested. Increasing the concentration of 25-hydroxycholestero1 still further (to 50 pg/ml) increased the activity to a value 30% above that at 10 pg/ml.

The possibility that esterification of 25-hydroxycholesterol itself might contribute to the apparent increase in cholesterol

1 7 0

E l l , , 1 1

0 2

I u 1 2 5 IO

MEDIUM 25-HYDROXYCHOLESTEROL (pg/ml)

FIG. I. Effects of 25-hydroxycholesterol on the rate of incorporation of [3H]oleic acid into cellular cholesterol esters (A), triglycerides 0, and phospholipids (0). After initial plating, hepatocytes were incubated for 24 h in DME medium with 20% FCS, then incubated in DME medium for another 24 h. The incubation time with [3H]oleic acid and 25-hydroxycholesterol was 60 min and the final concentration of oleic acid was 50 ,UM, with a 1:l molar ratio between oleic acid and bovine serum albumin. The specific activity of ["Hloleic acid was 50 dpm/pmol. 25-Hydroxycholesterol was added as an ethanol solution and the same amount of ethanol was added to the control plates (1% final concentration). Incorporation in control plates without ethanol did not differ from that in the presence of 1% ethanol. Radioactivity incorporated into each lipid class was divided by the specific radioactivity of the ["Hloleic acid in the medium and expressed in picomoles per mg of cell protein/h. Data points represent the means -+ S.D. for three individual dishes.

TIME IN CULTURE (hr)

i 0 0 0 0 ~ m

I-

L I I 0 24 48 12

TIME IN CULTURE (hr)

I P I I I \

300

tL TIME IN CULTURE (hr)

FIG. 2. Rate of incorporation of [3H]oleic acid into cellular cholesterol esters (A), triglycerides (B), and phospholipids (C) in the presence of 25-hydroxycholestero1 in freshly plated cells and in cells in culture for 1 to 3 days. After 20 h incubation in DME medium with 20% FCS the cells were switched to serum-free medium. Freshly isolated cells did not adhere to the dish surface during the 60-min incubation; these cells came off the plates with PBS and were centrifuged and harvested as described under "Mate- rials and Methods" and then washed twice with PBS. Medium in the dishes used for measurement a t 1, 2, and 3 days was changed daily. Hepatocytes were exposed to 25-hydroxycholesterol (A, 10 pg/ml) and control medium (0) for 60 min along with the ['H]oleic acid as described in the legend to Fig. 1. Data points represent the mean * S.D. for three individual dishes.

ester synthesis was considered. The cholesterol ester fraction isolated by thin layer chromatography of the lipids from cells incubated for 1 h with 25-hydroxycholesterol (10 pg/ml) was saponified and the nonsaponifiable fraction was subjected to gas-liquid chromatography. No 25-hydroxycholesterol was de- tected. As a more sensitive test, hepatocytes were incubated for 1 h with [~H]25-hydroxycholesterol (10 pg/ml) and unlabeled oleic acid (50 p ~ ) under the same conditions used for the experiments summarized in Fig. 1. The cell lipids were fractionated by thin layer chromatography using hexane/diethyl ether/acetic acid (80:20:1, v/v/v) as developing solvent. In this system esters of cholesterol have an RF of about 0.8 to 0.9 and free cholesterol about 0.3 while free 25-hydroxycho-


lesterol remains at or near the origin. Of the total radioactivity recovered (54,200 cpm), 61% remained at the origin and 37% ran with an RF of 0.4. Less than 0.1% co-chromatographed with the cholesterol ester or hydrocarbon bands. When the labeled component with RF = 0.4 was eluted, saponified, and rechromatographed, 96% of the radioactivity now was recovered at the origin. Thus, we conclude that there is esterification of 25-hydroxycholestero1, but that because of its greater polarity, this ester is not included in the fraction of cholesterol esters recovered. The absence of any significant amount of 25-hydroxycholesterol radioactivity in less polar fractions implies that little or no diester is formed.

The cells used in the experiments shown in Fig. 1 had been plated for 24 h in FCS and then transferred to DME medium without FCS for 24 h prior to addition of the labeled oleic acid. A marked stimulation of cholesterol esterification by 25- hydroxycholesterol could also be demonstrated in freshly isolated hepatocytes (&fold) as well as in hepatocytes incubated for 48 h or 72 h prior to exposure to 25-hydroxycholesterol (3- to 4-fold) (Fig. 2 4 ) . In each case the incorporation of oleic acid was measured in 1-h incubations. The absolute rate of incorporation of labeled oleic acid into cholesterol esters in control cells increased during the first 48 h of incubation but further stimulation due to 25-hydroxycholesterol was evident a t all times. The absolute rate of oleic acid incorporation into triglycerides increased as a function of time during the first 48 h after plating but there was no stimulation by 25-hydroxycholesterol at any time (Fig. 2B). Incorporation into phospholipids over the entire 72-h period remained relatively constant and there was no stimulation by 25-hydroxycholesterol (Fig. 2C). The same stimulatory effect of 25-hydroxycholesterol on cholesterol esterification was observed in the presence or in the absence of lipoprotein-deficient serum.

Time Course of the 25-Hydroxycholesterot Effect on Cho- lesterol Ester Formation-Cells were preincubated with or without 25-hydroxycholesterol for periods ranging from 15 min to 18 h and then incorporation of [''Hloleic acid into cholesterol esters was measured (over a 60-min interval). As shown in Fig. 3, the greatest stimulation of cholesterol esterification was observed after the shortest preincubation times (400%, at 15 min) and then fell progressively. After 4 h of

1 4 0 0 9 4

PREINCUBATION TIME (hr) FIG. 3. Effect of 25-hydroxycholestero1 on cholesterol ester

formation with increasing time of exposure of hepatocytes. The hepatocytes were kept in DME medium with 20% FCS for 24 h after initial plating and then changed to serum-free medium. After another 24 h the experiment was started by adding fresh medium with or without 10 pg/rnl of 25-hydroxycholestero1 (see legend to Fig. 1). After the indicated preincubation times in the continued presence (A) and absence (0) of 25-hydroxycholesterol, [,'H]oleic was added to the medium and cholesterol ester formation over a 60-min interval was measured as described under "Materials and Methods." Data points represent the means C S.D. for three individual dishes.

INCUBATION TIME (min) FIG. 4. The effect of short term exposure of hepatocytes to

25-hydroxychoIestero1 on the incorporation of [3H]oleic acid into cholesterol esters. The cells were plated in DME medium with 20% FCS and the medium was changed to serum-free DME medium after 24 h. After another 24 h at 39°C the experiment was started by changing the medium to DME containing ['Hloleic acid (50 p ~ ) in the presence (A) and absence (0) of 10 pg/ml of 25-hydroxycholesterol (see legend to Fig. 1). Data points represent means -+ S.U. for three individual dishes.

preincubation, the stimulation due to 25-hydroxycholestero1 had dropped to 60%; after 18 h, there was no detectable 25- hydroxycholesterol effect. These findings suggested that in the longer incubations there might be metabolism of 25-hydroxycholesterol, leading to a decrease in its concentration in the medium. To test this possibility, we replaced the medium on cells previously incubated for 4 h with 25-hydroxycholesterol (10 pg/ml) with fresh medium containing the same concentration of 25-hydroxycholesterol. The addition of fresh 25-hydroxycholestero1 at 4 h along with labeled oleic acid increased cholesterol ester synthesis only 40% above that seen in cells to which no fresh 25-hydroxycholesteroI had been added. This contrasts with the 4-fold increase seen in the 1st h of exposure to 25-hydroxycholesterol.

The time course of the 25-hydroxycholesterol effect was examined more closely in shorter incubations. As shown in Fig. 4, the stimulating effect was evident and already essentially maximal at 15 min. The rapidity of this response and the fact that it did not increase progressively with time made it unlikely that new enzyme protein synthesis was involved. Further evidence supporting this conclusion was obtained by showing that pretreatment of the cells with cycloheximide (final concentration, 10 ~ L M ) did not block the response to 25- hydroxycholesterol (Fig. 5 ) . Even 18-h exposure to cycloheximide did not appreciably reduce acyl-CoA:cholesterol acyltransferase activity nor recovery of total cell protein. Incor- poration of labeled amino acids into total cellular protein in the presence of cycloheximide was inhibited by 85'% (measured during the final 1 h of incubation).

Stimulation of Esterification of Endogenously Synthesized Cholesterol a n d of Exogenous Cholesterol by 25-Hydroxy- cholesterol-In order to determine whether 25-hydroxycholesterol stimulated the esterification only of preformed cholesterol or also that of newly synthesized cholesterol, ~ ~ 4 2 -

C]mevalonolactone was used as precursor. As shown in Table I, the incorporation of labeled mevalonolactone into cholesterol esters at 1 h was increased almost 7-fold. The incorporation of labeled mevalonolactone into free cholesterol, in contrast, was lower in cells exposed to 25-hydroxycholesterol.

We also tested whether exogenous labeled free cholesterol, delivered to the cells in high density lipoprotein (HDL) added to the medium, was esterified more rapidly in the presence of

14

Regulation of Cholesterol Metabolism in Liver Cells

Lu X + z > v)

a Lu + 2 2 0 a LLI c

0 I u

C o n t r o l I C y c l o h e x i m i d e

T I I ( l X IO+ M ) I

I I

I 0 C o n t r o l

I T c h o l e s t e r o l 200 I

El 25- h y d r o x y -

(IO pg/mI )

FIG. 5. The effect of 25-hydroxycholestero1 on incorporation of [9,10-3H]oleic acid into cholesterol esters in hepatocytes incubated in the presence or in the absence of cycloheximide. Cells were initially plated in DME medium with 20% FCS and then changed to DME medium with or without cycloheximide (1 X M). After preincubation in this medium for 18 h, the medium was changed to DME medium containing 10 pg/ml of 25-hydroxycholesterol dissolved in ethanol (hatched bars) or control medium with only ethanol added (open bars). [JH]Oleic acid (WPM final concentration; oleic acid/bovine serum albumin ratio equal to 1:l; specific activity 100 dpm/pmol) was added at zero time to the dishes and incubation continued for 1 h. Radioactivity incorporated into cholesterol esters was divided by the specific radioactivity of the tritiated oleic acid in the medium. The means f S.D. from three individual dishes are given.

TABLE I The effect of 25-hydroxycholesterol and meualonolactone on

incorporation of [2-"CJmevalonolactone into cholesterol ester Hepatocytes were plated in DME medium with 20% FCS and after

24 h the medium was changed to DME medium containing 25- hydroxycholesterol (10 p g / d ) , DL-mevalonolactone (1 mM), or both. The media contained 1% ethanol (final concentration) and human lipoprotein-deficient serum (5 mg of protein/ml). The same amount of ~~-[2'~C]mevalonolactone was added to all dishes (1.09 X 10" dpm/ ml; 18 nmol/ml); final specific activity was thus 60.6 dpm/pmol in the absence of added unlabeled mevalonolactone and 1.07 dpm/pmol in its presence. Total radioactivity incorporated into sterol or sterol ester was divided by the specific radioactivity of mevalonolactone in the medium. Because of the unknown degree of dilution by endogenous intermediates this yields only a minimum estimate of true synthetic rate. The incubations lasted 60 min and the labeled precursor was present throughout the incubation. The means f S.D. for three individual dishes are given.

Incorporation of [Z-"Clmevalono- ["C]Choles- terol ester as

Free cholesterol Cholesterol es- of total ["C]- a percentage

lactone into Medium

ters cholesterol pmol/mgprotein w

Control 1,448 f 31 84.6 f 6.7 5.5 25-Hydroxycholes- 732 f 72 555 f 82 42.9

DL-Mevalonolac- 12,888 f 1360 14,687 f 2013 53.3

25-Hydroxycholes- 8,504 f 904 23,827 & 1178 73.7

terol

tone

terol + DL-meVa- lonolactone

TABLE I1 The effect of 25-hydroxycholesteroI and mevalonolactone on

esterification of [3HJcholesterol presented to cells in high density lipoproteins (HDL)

The hepatocytes were plated in DME medium with 20% FCS and changed to DME medium containing 400 p g / d of human HDL protein (1.09 < d < 1.21 g/ml), 0.1 mg/ml of albumin, and 5 mg protein/ml of human lipoprotein-deficient serum. [1,2-JH]Cholesterol (specific activity 43 Ci/nmol; 2.5 X IO6 dpm) was dissolved in 70 p1 of acetone and then injected into a solution of bovine serum albumin (1 mg/ml). The acetone was blown off with a stream of N2. Seven hundred microliters of the albumin-stabilized solution of labeled cholesterol was added to 5.5 mg of HDL protein and then incubated for 4 h a t 37°C in a shaking waterbath to equilibrate the labeled free cholesterol with the free cholesterol in the HDL. The ['H]cholesterol- labeled HDL was recentrifuged at d 1.21 prior to use. The final concentrations of 25-hydroxycholesterol and DL-mevalonolactone were 10 p g / d and 1 mM, respectively. The incubation lasted for 120 min with the labeled HDL present during the whole incubation. The means of three individual dishes f S.D. are given.

Incorporation of ['H]cholesterol , ~Hlcholes- into terol ester as

Hepatocyte free cholesterol es- of total ['HI- Hepatocyte a percentage

cholesterol cholesterol ters cpm/mg cell protein B

Control 19,379 f 1938 402 f 130 2.0 25-Hydroxycholes- 19,731 f 2722 1441 f 184 6.8

DL-Mevdonolactone 19.007 f 2027 1440 f 126 5.7 terol

25-hydroxycholesterol (Table 11). Both the [3H]cholesterol- labeled HDL and the 25-hydroxycholesterol were added at zero time and the incubation was terminated at 120 min. There was more than a 3-fold increase in esterification of exogenous cholesterol in the hepatocytes exposed to 25-hydroxycholesterol.

Stimulation of Cholesterol Ester Formation by Meualono- lactone-As shown in Tables I through 111, high concentrations of mevalonate increased rates of cholesterol esterification. It should be noted that an effect was observed with each of the three labeled precursors tested, oleic acid, cholesterol added in HDL, and mevalonolactone. The addition of high concentrations of mevalonolactone, of course, markedly increased synthesis of free cholesterol from labeled mevalonolactone but the distribution of incorporated radioactivity between free and ester cholesterol was shifted radically, indicating an effect on the esterification processper se (Table I). The incorporation of [3H]oleic acid into cholesterol esters as a function of mevalonolactone concentration increased rapidly up to about 0.1 mM and then more slowly but progressively up to 1.0 rnM; however, incorporation of [14C]mevalonolactone into cholesterol esters was unaffected by concentration of oleic acid (data not shown). At a concentration of 1 mM, the stimulation of ["Hloleic acid incorporation varied from 2- to 5-fold above the control (5-fold in the experiment summarized in Table 11). When [14C]mevalonolactone was used as the precursor (Fig. 6), there was also an increase in incorporation into cholesterol esters with increasing concentration of mevalonolactone. The apparent magnitude of the effect was much greater than that in the studies using [3H]oleic acid as precursor (Table 111). There was a 38- to 48-fold increase in incorporation of 'H into cholesterol esters between 25 p M and 10 mM mevalonolactone. The radioactivity in free cholesterol also increased markedly. Since the extent of dilution of labeled precursor by endogenous intermediates may be greater at the lower concentrations of added mevalonolactone, some of the increased incorporation could reflect such differential dilution. The key point is that with increasing concentrations of mevalonolactone, there was a greater increase in incorporation


TABLE I11 Effects of 25-hydroxycholesterol and mevalonolactone on

incorporation of C3H]oleic acid into cholesterol esters Hepatocytes were plated in DME medium with 20% FCS and after

24 h the medium was changed to DME medium containing 25- hydroxycholesterol (IO pg/ml), DL-mevdonolactone (1 mM), or both. The media contained 1% ethanol and 50 oleic acid bound to defatted bovine serum albumin (1:I molar ratio between the albumin and oleic acid). The specific activity of [9,10-3H]oleic acid was 50 dpm/pmol. The incubation lasted for 60 min and the label was present throughout the incubation. The means for three individual dishes & S.D. &e given.

19.10-’HlOIeic acid incomoration into cholesterol esters pmol/mgprotein

Control 123 k 19 25-Hydroxycholestero1 666 k 93 DL-Mevalonolactone 682 k 42 25-Hydroxycholestero1 + m-mevalon- 1209 k 141 olactone

12 -

IO

6

I I I I

0.2 0.4 0.6 0 8 I O IO

I /,A’ I I I I I

0.2 0.4 0.6 0 8 I O 1 /I IO

MEDIUM MEVALONOLACTONE (mM)

FIG. 6. The effect of increasing concentrations of mevalonolactone on incorporation of [”C]mevdonolactone into free cholesterol (0) and cholesterol esters (A). The hepatocytes were plated in DME medium with 20% FCS and after 24 h the medium was changed to DME medium. Forty-eight hours after plating the cells the medium was changed to DME medium with the indicated concentrations of DL-mevalonolactone (dissolved in ethanol) and labeled ~~-[2-’~C]mevalonolactone was added at the Same time. In- cubations lasted for 60 min. ~~-[2-’~C]Mevalonolactone had a specific activity ranging from 60.0 to 0.157 dpm/pmol. Total radioactivity incorporated into sterol or sterol ester was divided by the specific activity of mevalonolactone in the medium. Because of the unknown degree of dilution by endogenous intermediates, the incorporation rates represent only a minimum estimate of true sterol synthetic rates. Mean values for duplicate dishes are given.

into cholesterol esters than into free cholesterol, indicating stimulation of the rate of esterification.

Additive Effect of 25-Hydroxycholesterol a n d Mevalono- lactone-Having two compounds both active in increasing cholesterol esterifkation it was of interest to test whether or not their effects were additive. In Table I11 data are presented strongly suggesting an additive effect of these compounds using labeled oleic acid as precursor. Additivity was also shown when labeled mevalonolactone was used as precursor (Table I). As discussed above, the effect of 25-hydroxycholesterol at 10 p g / d is near-maximal. Increasing the concentrations to 50 p g / d only enhanced the effect by an additional 30%. As shown in Fig. 6, 1 m~ mevalonolactone also exerts a near-maximal effect. Increasing the concentration to 10 mM yielded only a 25% further enhancement. Thus, a 5- to 10-fold increase in the concentrations of the individual agonists

yielded only a 25 to 30% increase in effect. As shown in Tables I and 111, their combined effects were at least 50 to 100% above the effect of either one alone. An additive effect was also obtained using 50 pg/ml of 25-hydroxycholestero1 and 10 mM mevalonolactone (data not shown).

Changes in Free and Ester Cholesterol Mass in the He- patocytes-The apparent increase in the rate of cholesterol esteriiication induced by mevalonolactone or by 25-hydroxycholesterol might be expected to increase the hepatocyte content of esterified cholesterol (unless cholesterol ester hy- drolysis is also increased). The free cholesterol content of control cells and of cells incubated with mevalonolactone was about 12 pg/mg of cell protein and did not change significantly over a 20-h incubation (Fig. 7A). After 2 and 6 h of incubation with 25-hydroxycholesterol (IO pg/ml) there was a drop in free cholesterol content (to about 9 pg/mg of cell protein). The amount of esterified cholesterol in control cells increased about 2-fold over the 20 h of incubation (Fig. 7B) . The reason for this increase is presently unknown. However, the ester cholesterol content at 2 and 6 h was even greater in cells exposed to 10 pg/ml of 25-hydroxycholestero1 than in control cells. This difference in ester cholesterol content virtually disappeared by 20 h. This finding is consistent with the reduced rate of cholesterol esterification in longer incubations demonstrated in tracer studies (Fig. 3). In the presence of mevalonolactone, however, the mass of esterified cholesterol

0 0 2 5 10 20

INCUBATION TIME ( h r ) FIG. 7. The effect of 25-hydroxycholestero1 and mevalono-

lactone on cellular concentrations of free (A) and esterified cholesterol (B) as a function of time of exposure. Hepatocytes were plated in DME medium with 206 FCS and after 24 h the medium was changed to DME medium containing 50 PM oleic acid bound to albumin (molar ratio 1:l) and 10% FCS. Further additions: 10 pg/d of 25-hydroxycholesterol (A) or 10 m~ DL-mevalonolactone (U) (each dissolved in ethanol), or ethanol alone (0). After incubations over the indicated time periods, the cells were washed six times and harvested for analysis (see “Materials and Methods”). The data points represent means & S.D. for three individual dishes.


increased progressively over the entire 20-h incubation, rising from 3.1 to 18.9 pg/mg of cell protein at 20 h.

Zncrease in Acyl-CoA:choZesterol Acyltransferase Activity in Microsomes-An attempt was made to determine whether the apparent increase in cholesterol esterification in the intact hepatocytes could be correlated with increased activity of acyl-CoA:cholesterol acyltransferase in cell-free preparations. Microsomes were isolated from cells that had been previously exposed to 25-hydroxycholesterol for 60 min. In these experiments, freshly isolated liver cells were used since we find that the yield of microsomes is better from freshly isolated cells than from cells that have been in monolayer culture for 24 to 48 h. The latter are more difficult to disrupt by sonication. The incorporation of [I4C]oleic acid from [’4C]oleyl-CoA was increased about %fold in microsomes prepared from cells previously incubated with 25-hydroxycholestero1 (Fig. 8). It was also shown that direct addition of 25-hydroxycholesteroI to the microsomal incubation mixture promoted a significant stimulation of acyl-CoA:cholesterol acyltransferase activity and this was true both in microsomes from control cells and

Z0 r

1 5 ;”----

FIG. 8.

O O 6 1 5 10

CONCENTRATION OF 25-HYDROXYCHOLESTEROL IN MICROSOMAL INCUBATION (pg/ml)

Activity of acyl-CoA:cholesterol acyltransferase in microsomes isolated from hepatocytes previousiy incubated with 25-hydroxycholestero1. Freshly isolated liver hepatocytes were incubated in DME medium with antibiotics and insulin as described under “Materials and Methods.” In eight plates, 5 O p ~ oleic acid bound to 58 bovine serum albumin (ratio 1:1) was added along with 1% (v/v) ethanol (O), and in eight other plates, additions were the same plus 10 pg/ml of 25-hydroxycholestero1 (A). Cells were harvested after 60 min of incubation and washed twice. Sonication of the cell pellet was performed carefully in 0.25 M sucrose with a Bronson sonicator at 0°C for 15 s twice. The homogenates were centrifuged at 12,000 X g for 20 min at 0°C and then the supernatants were centrifuged at 10,ooO X g for 60 rnin. The microsomal pellets were resuspended in 0.25 M sucrose. Microsomal protein ( 1 0 0 to 150 p g ) in 25 pl was added to each tube containing 250 pi of 0.1 M potassium phosphate buffer with 0.1% (w/v) defatted bovine serum albumin and 45 +M [l-’4C]oleyl coenzyme A (specific activity 31 dpm/ pmol) in a final voiurne of 0.3 ml. To some of the microsomal incubation mixtures, 25-hydroxycholesterol was added as an ethanol solution to give final concentrations of 1.0 or 10 p g / m l . Assays were stopped at 5 min. The means of triplicates rt S.D. are given.

I I - - 90 - A

- -

Cr 60- -

t = 50- - I- V 4 ;;; 4 0 - -

V

0 W

- 3 30- a 2 20- - u r3

10-

I 1 1 I I 0.5 1.0 2.0 5.0 10.0

MEDIUM 25-HYDROXYCHOLESTEROL (Fg/ml)

I I 1

MEDIUM MEVALONDIACTONE ( pM) 25 50 250 1000

FIG. 9. HMG-CoA reductase activity in isolated microsomes prepared from cells previously incubated with 25-hydroxycholesterol (A) or mevalonolactone (B). Freshly isolated hepatocytes were plated in DME medium and incubated for 60 min with the indicated concentrations of 25-hydroxycholestero1 or m-mevalonolactone. The cells were harvested as described under “Materials and Methods,” sonicated, and microsomes were isolated as described in the legend to Fig. 8 (except that the buffer used contained 0.25 M sucrose, 0.3 mM EDTA, 50 mM K2P04, and 5 m~ dithiothreitol, pH 7.2). The incubation conditions for assaying HMG-CoA reductase are described under “Materials and Methods” and this incubation lasted for 40 min at 37°C on a shaking water bath. Means for duplicate samples are given.

in microsomes from cells previously incubated with 25-hydroxycholesterol (Fig. 8).

The Effect of Different Steroids on Acyl-CoA:cholesterol Acyltransferase in Cultured Hepatocytes-Several steroids were tested for their effects on liver cell cholesterol esterification (Table IV). In addition to 25-hydroxycholest~erol, 7- ketocholesterol and 5,24-cholestadien-3/?-01 promoted a sig- niftcant increase in oleic acid incorporation into cholesterol esters a t a final concentration of 10 pg/ml. Unexpectedly, there was no increase in cholesterol esterification when pure cholesterol (in ethanol) was added to the medium (Table IV). Cholesterol (10 pg/ml) was also added in acetone or in dimethylsulfoxide but failed to influence cholesterol esterification activity irrespective of the solvent used. However, on adding 50 p g / d of cholesterol in ethanol we observed a 20% increase in cellular cholesterol esterification (Table IV).

The Effect of 25-Hydroxycholesterol and Mevalonolactone


on HMG-CoA Reductase-HMG-CoA reductase activity was assayed in microsomes from freshly isolated liver cells previously exposed for 60 min to 25-hydroxycholesterol and/or mevalonolactone. With increasing concentrations of 25-hydroxycholesterol in the medium there was a progressive and highly significant decrease in HMG-GOA reductase activity (Fig. 9). At 10 p g / d of 25-hydroxycholestero1, activity was reduced by 60 to 90% relative to that measured in the control cells. Mevalonolactone also decreased HMG-GOA reductase activity but high concentrations were required (Fig. 9). At 1.0 mM DL-mevdonolactone, the activity fell to 27% of the control level. The time course for the effect of 25-hydroxycholestero1 on HMG-CoA reductase activity in freshly isolated rat hepatocytes is shown in Fig. 10. After just 15 min incubation, HMG-CoA reductase activity was down to 60% of the control level and the activity dropped to about 25% of the control level after 60 min. When the time course was extended (Fig.

TABLE IV The effect of different steroids on cholesterol esterification in

cultured heputocytes The hepatocytes were plated in DME medium with 20% FCS and,

after 24 h, changed to DME medium with human lipoprotein-deficient serum (5 mg of protein/ml) with the indicated steroids (added in ethanol unless otherwise indicated). The incubation with the steroids lasted 120 min and 50 p~ ['Hloleic acid bound to defatted bovine serum albumin (1:l molar ratio) was present for the last 60 min. The final concentrations of solvents were 1%. The number of dishes used for testing of the components are given in parentheses (N). Me2SO, dimethylsulfoxide.

Incorporation of [ 'Hloleic acid

Compounds Added N concentra- into cholesterol esters relative to

control values tion

Control (5) ( 100) 25-Hydroxycholestero1 (4) 10 pg /ml 435 7-Ketocholesterol (6) 1 O p g / d 260 P-Sitosterol (6) lOpg/ml 100 5-a-Cholestane (6) 10 p g / d 100 5,24-Cholestadien-3P-ol ( 6 ) 10 p g / d 190 Lanosterol (6) 10 pg/ml 75 Dexamethasone (3) 10pM 95 Ethinyl estradiol (3) 100Pg/ml 60 Cholesterol (3) lOpg/d 100 Cholesterol (in acetone) (3) 10 p g / d 100 Cholesterol (in Me,SO) (3) 10 p g / d 100 Cholesterol (3) 20 p g / d 100 Cholesterol (3) 50 pg/ml 120

I I I I

INCUBATION TIME (min)

FIG. 10. HMG-CoA reductase activity in microsomes from liver cells previously incubated for different times with 25- hydroxycholesterol. The conditions were as described in the legend to Fig. 9. The media added to the cells were with (A) or without (0) 25-hydroxycholesterol (10 p g / d ) . Means of duplicate samples are given.

Kx) -0-0 e 0

>

J 1 2 5 10 22

INCUBATION TIME (hr) FIG. 11. HMG-CoA reductase activity in microsomes iso-

lated from cultured liver cells previously incubated with 25- hydroxycholesterol for longer time intervals. The isolated hepatocytes were plated in DME medium with 20% FCS and after 24 h the medium was changed to DME medium with (A) or without (0) 25-hydroxycholestero1 (10 pg/ml). After the indicated times o f incubation the cells were harvested and microsomes were isolated for HMG-CoA reductase assay as described in the legend to Fig. 10. Data are related to the activity in control cells set equal to 100% (29.1 pmol of mevdonate formed/mg of protein/min). Cells from three individual plates were pooled and means of duplicate assays are given.

ll), the HMG-CoA reductase activity was shown to remain at a low level (about 15% of the control) for at least 6 h. Even after 20 h of incubation with 25-hydroxycholestero1 the activity of HMG-GOA reductase was less than 30% of that in the control cells. Liver cells incubated with cholesterol (10 pg/ml) for 60 min showed no change in microsomal HMG-CoA reductase activity (data not shown).

DISCUSSION

The conclusion from these studies that 25-hydroxycholesterol stimulates microsomal acyl-CoA:cholesterol acyltransferase activity in normal rat hepatocytes is supported by the following findings: 1) incorporation of labeled oleic acid into cholesterol esters, but not into triglycerides or phospholipids, was increased 3- to 6-fold in intact cells exposed to 25-hydroxycholesterol (Fig. 1); 2) incorporation of exogenously labeled cholesterol (presented in HDL) into cholesterol esters was increased (Table 11); 3) incorporation of newly synthesized endogenous cholesterol (from labeled mevalonolactone) into cholesterol esters was increased (Table I; Fig. 6); 4) during the fwst few hours of exposure of the cells to 25-hydroxycholesterol, when stimulation of cholesterol ester formation from labeled precursors was maximal, there was a net increase in cellular ester cholesterol content and a decrease in cellular free cholesterol content (Fig. 7 ) ; 5) isolated microsomes prepared from cells previously incubated with 25-hydroxycholesterol showed acyl-CoA:cholesterol acyltransferase activity 3 times that of microsomes from control cells (Fig. 8); 6) incubation of isolated microsomes with 25-hydroxycholestero1 increased acyl-CoA:cholesterol acyltransferase activity (Fig. 8).

As described above, 25-hydroxycholestero1 itself was esterified by hepatocytes under the conditions of these studies. However, the esterified 25-hydroxycholesterol was not isolated in the thin layer chromatographic band containing esters of cholesterol itself. Moreover, the effect of 25-hydroxycholesterol on cholesterol ester formation from exogenous labeled free cholesterol and from newly synthesized endogenous cholesterol was fully as great as its effect on ester formation from labeled oleic acid. In rat livers perfused with vH125-hydroxycholesterol, Erickson et al. (13) found that most of the liver- associated 'H was recovered in 25-hydroxycholesterol, but small amounts of a metabolite with polarity less than that of


cholesterol were also observed. Goldstein and Brown reported significant synthesis of esters of 25-hydroxycholestero1 in cultured skin fibroblasts ( 10).

The fact that the 25-hydroxycholesterol effect was maximal within 15 min of incubation (Fig. 4) and that it was not blocked by concentrations of cycloheximide that inhibited protein synthesis by 85% (Fig. 5) rules out mechanisms de- pendent on new protein synthesis. The finding that addition of 25-hydroxycholestero1 to isolated microsomes, without preincubation, stimulated acyl-CoA:cholesterol acyltransferase activity suggests that the effect is rather directly evoked by 25-hydroxycholesterol itself rather than by a metabolite or by some indirect metabolic action of the polar sterol. Taken together, the data support the interpretation that 25-hydroxycholesterol itself may affect the activity of acyl-CoA:cholesterol acyltransferase by changing its configuration, either by a direct interaction with the enzyme protein (allosteric modi- fication) or by a less specific effect on the lipid environment of the enzyme in the microsomal membrane. Changes in acyl- CoA:cholesterol acyltransferase activity in Ehrlich ascites tu- mor cells of rats given diets of different fatty acid composition may be effected by an analogous mechanism (30).

The stimulatory effect of 25-hydroxycholesterol on acyl- CoA:cholesterol acyltransferase activity appeared with almost no delay but even in the continued presence of 25-hydroxycholesterol the magnitude of the effect decreased steadily. By 18 h the effect had disappeared. The mechanism of this “adaptation” is not clear. The 25-hydroxycholesterol inhibition of HMG-CoA reductase (another microsomal enzyme) also was evident very quickly (Fig. 10) but persisted at a significant level for at least 22 h (Fig. 11) in agreement with the results of Bell et al. (11). This striking dissociation in the time course of the effects strongly suggests that the 25-hydroxycholesterol effect on the two systems occurs by mechanisms that must be different in some way, although they could share some common features. Inhibition of HMG-CoA reductase by cholesterol is believed to occur in two ways: 1) by an acute mechanism relating to the activity of a given amount of enzyme protein (31, 32); 2) by a long term mechanism, probably attributable to suppression of synthesis of the enzyme (5, 33). Our results suggest that 25-hydroxycholesterol stimulates acyl-CoA:cholesterol acyltransferase very directly, by altering the enzyme or its environment in the membrane. It may acutely alter HMG-CoA reductase by a similar mechanism. The adaptation (loss of the acyl-CoA:cholesterol acyltransferase stimulation) may reflect reshaping of the microsomal membrane as a result of shifts in lipid or protein content or at least changes in some domains within the membrane. If there were no suppression of HMG-CoA reductase synthesis, there might be a similar adaptation with respect to its activity. However, the fact that HMG-CoA reductase activity fell and stayed down may reflect inhibition of new enzyme synthesis in analogy with the demonstrated effect of cholesterol itself on reductase synthesis (33).

Mevalonolactone at high concentration also increased cholesterol esterification in the hepatocytes and the stimulatory effect was observed using labeled mevalonolactone, labeled free cholesterol, or labeled oleic acid as precursors (Fig. 6, Tables I and 11). Moreover, there was a striking net increase in the cellular content of ester cholesterol on incubation with mevalonolactone at a high concentration (10 mM) without any significant change in cellular free cholesterol content (Fig. 7). Presumably the effect is the result of a greatly increased rate of de novo synthesis of cholesterol, synthesis not inhibited by normal feedback control because it occurs beyond the HMG- CoA reductase step. The observed results with mevalonolactone undoubtedly are due a t least in part (or possibly in toto)

to the marked increases in the amount of free cholesterol made available as substrate for acyl-CoA:cholesterol acyltransferase and need not entail any changes in the activity state of the enzyme. On the other hand, increases in the cholesterol content of the microsomal membrane, like increases in its 25-hydroxycholesterol content, may contribute to the apparent increase in acyl-CoA:cholesterol acyltransferase. Hashimoto and Dayton (34) have shown that acyl-CoA: cholesterol acyltransferase activity in microsomes isolated from the rabbit aorta is related linearly to the concentration of free cholesterol in the microsomes. They concluded that the increased rate of cholesterol ester formation in athero- sclerotic rabbits could not be explained by increases in amount of enzyme but must reflect increases in available cholesterol substrate and/or increases in activity of the enzyme. Meva- lonate stimulation of cholesterol ester formation using tritiated water as labeled precursor has also been observed in freshly isolated rat hepatocytes (35). Mitropoulos et al. (15) found that mevalonolactone administered intravenously to rats failed to increase liver acyl-CoA:cholesterol acyltransferase activity. However, both Mitropoulous et al. (15) and Erickson (36) demonstrated increased acyl-CoA:cholesterol acyltransferase activity after intragastric administration of mevalonolactone. They suggested that transformation of the mevalonolactone to cholesterol had to take place in the intestine first, after which the increased amounts of lipoprotein cholesterol made in the intestine modified hepatic cholesterol metabolism. The present data and those of Nilsson (35), on the other hand, show that direct uptake of mevalonolactone by the liver cell can increase cholesterol ester formation. The mevalonolactone effect may in part be due to mechanisms different from those responsible for the 25-hydroxycholesterol effect. This is suggested by the additivity of the effects of maximally effective concentrations of the two (Tables I and 111).

In the present studies, we have used 25-hydroxycholesterol, because of its potency, as a tool to explore regulation of hepatic acyl-CoA:cholesterol acyltransferase and HMG-CoA reductase. Whether or not oxygenated sterols play a significant role under physiological conditions remains to be determined. Under some circumstances pure cholesterol itself has been reported to have little if any effect on endogenous cholesterol synthesis in fibroblasts (37) and in hepatoma cells (1 1) while a variety of oxygenated sterols were highly effective. These findings have led to the proposal that a molecule other than cholesterol itself may be of importance for intracellular regulation of cholesterol metabolism (11, 37, 38). In the adult rat hepatocytes used here, 25-hydroxycholesterol was again very effective both in inhibiting HMG-CoA reductase and stimulating acyl-CoA:cholesterol acyltransferase while cholesterol itself was without effect at the same low concentration (10 pg/ml). By using 50 p g / d of cholesterol (>98% pure) we observed a slight stimulation of cellular cholesterol esterification (20% above the control), which could be due to a small contamination with an oxygenated sterol. However, the greater potency of the more polar sterols, in this system as in others, does not rule out a regulatory role for cholesterol itself nor establish that polar sterols are involved in physiological regulation.

The physiological importance of intracellular cholesterol esterification in vivo (probably catalyzed predominantly by acyl-CoA:cholesterol acyltransferase) is not completely un- derstood. There is significant evidence indicating that cultured peripheral cell types such as fibroblasts, smooth muscle cells, and rat hepatoma cell lines have the capacity to respond with increased acyl-CoA:cholesterol acyltransferase activity to an increase in uptake of cholesterol in lipoproteins (8, 39). It has

Regulation of Cholesterol

also been demonstrated that cholesterol feeding of guinea pigs promotes a 5- to 10-fold increase in acyl-CoA:cholesterol acyltransferase activity in the liver and small intestine along with a large increase in cellular content of cholesterol, especially ester cholesterol (1). An important function in these cases is probably to “detoxify” free cholesterol in order to avoid alter- ations of vital membrane functions due to increases in membrane-free cholesterol content (40). In the liver (and in the intestine) cholesterol esterification could be of importance also for incorporating cholesterol esters into lipoproteins, even though much or most of ester formation may occur after lipoprotein secretion (41). Mahley et al. (42) have shown that lipoprotein particles in the Golgi apparatus of rat liver contained cholesterol esters. Their data and those of Hamilton et al. (43) and Davis et al. (16) suggest that very low density lipoproteins (VLDL) secreted from the liver already contain some cholesterol in ester form. Whether or not regulation of hepatic acyl-CoA:cholesterol acyltransferase plays a significant role in control of the amount and concentration of lipoprotein cholesterol ester remains to be determined.

Acknowledgments-We are indebted to Erika Sandford, Sheldon C. Engelhorn, and Joellen Barnett for their skilled technical assist- ance. The kind gift of purified mevalonolactone from Dr. Thomas s. Parker and valuable discussions with Dr. Ray C. Pittman are greatly appreciated.

REFERENCES

1. Drevon, C. A. (1978) Atherosclerosis 30, 123-136 2. Erickson, S. K., Shrewsbury, M. A., and Gould, R. G. (1978)

3. Taylor, G. B., and Gould, R. G. (1950) Circulation 2, 467-468 4. Bucher, N. L. R., Overath, P., and Lynen, F. (1960) Biochim.

Biophys. Acta 40,491-501 5. Edwards, P. A., and Gould, R. G. (1972) J. Biol. Chem. 247,1520-

1525 6. Beinert, H. (1964) in Manometric Techniques (Umbreit, W. W.,

Burris, R. H., and Stauffer, J. F., e&) 4th Ed., pp. 114-129, Burgess Publishing Co., Minneapolis, MN

7. Jeejeebhoy, K. N., and Phillips, M. J. (1976) Gastroenterology

8. Goldstein, J . L., and Brown, M. S. (1977) Annu. Reu. Biochem.

9. Brown, M. S., and Goldstein, J . L. (1974) J. Biol. Chem. 249,

10. Brown, M. S., Dana, S. E., and Goldstein, J. L. (1975) J. Biol.

11. Bell, J . J., Sargeant, T. E., and Watson, J. A. (1976) J. Biol.

Circulation 58, 11-169

71, 1086-1096

46,897-930

7306-7314

Chem. 250,4025-4027

Chem. 251, 1745-1758

Metabolism in Liver Cells 9137

12. Kandutsch, A. A,, and Chen, H. W. (1974) J. Biol. Chem. 249,

13. Erickson, S. K., Matsui, S. M., Shrewsbury, M. A., Cooper, A. D., and Gould, G. (1978) J. Biol. Chem. 253,4159-4164

14. Edwards, P. A., Popjak, G., Fogelman, A. M., and Edmond, J. (1977) J. Biol. Chem. 252, 1057-1063

15. Mitropoulos, K. A., Balasubramaniam, S., Venkatesan, S., and Reeves, B. E. A. (1978) Biochim. Biophys. Acta 530,99-11 I

16. Davis, R. A,, Engelhorn, S. C., Pangburn, S. H., Weinstein, D. B., and Steinberg, D. (1979) J. Biol. Chem. 254,2010-2016

17. Dulbecco, R., and Freeman, G. (1959) Virology 8, 396-397 18. Hanks, J . H., and Wallace, R. E. (1949) Proc. SOC. Exp. Biol.

19. Leffert, H. L., and Koch, K. S. (1978) CIBA Found. Symp. 55,

20. Berry, M. N., and Friend, D. S. (1969) J. Cell Biol. 43,506-520 21. Hashimoto, S., Dayton, S., Alfin-Slater, R. B., Bui, P. T., Baker,

22. Folch, J., Lees, M., and Sloane-Stanley, G. H. (1957) J. Biol.

23. Goldfarb, S., and Pitot, H. C. (1971) J. Lip& Res. 12, 512-515 24. Hilz, H., Knappe, J., Ringelmann, E., and Lynen, F. (1958) Bio-

25. Havel, R. J., Eder, H. A,, and Bragdon, J. H. (1955) J. Clin.

26. Noble, R. P. (1968) J. Lipid Res. 9,693-700 27. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

(1951) J. Biol. Chem. 193, 265-275 28. Gamble, W., Vaughan, M., Kruth, H. S., and Avigan, J. (1978) J.

Lipid Res. 19, 1068-1070 29. Bligh, E. G., and Dyer, W. J. (1959) Can. J. Biochem. Physiol.

37,911-917 30. Brenneman, D. E., Kaduce, T., and Spector, A. A. (1977) J. Lipid

Res. 18,582-591 31. Nordstrom, J. L., Rodwell, V. W., and Mitschelsen, J. J. (1977) J.

Biol. Chem. 252,8924-8934 32. Ingebritsen, T. S., Lee, H. S., Parker, R. A., and Gibson, D. M.

(1978) Biochem. Biophys. Res. Commun. 81, 1268-1277 33. Higgins, M., and Rudney, H. (1973) Nature New Biol. 246.60-61 3 4 . Hashimoto, S., and Dayton, S. (1977) Atherosclerosis 28,447-452 35. Nilsson, A. (1975) Eur. J. Biochem. 51, 337-342 36. Erickson, S. K. (1978) Fed. Proc. 37, 1428 37. Kandutsch, A. A., and Chen, H. W. (1973) J. Biol. Chem. 248,

38. Watson, J. A., Havel, C., Hansbury, E., and Scallen, T. J. (1979)

39. Rothblat, G. H., Naftulin, M., and Arbogast, L. Y. (1977) Proc.

40. Papahadjopoulos, D. (1974) J. Theor. Biol. 43, 329-337 41. Glomset, J. A., and Norum, K. R. (1973) Adu. Lipid Res. 11, 1-65 42. Mahley, R. W., Hamilton, R. L., and LeQuire, V. A. (1969) J.

43. Hamilton, R. L. Williams, M. C., Fielding, C. J., and Havel. R. J.

6057-6061

Med. 71,196-200

61-81

N., and Wilson, L. (1974) Circ. Res. 36, 176-183

Chem. 226,497-509

chem. Z.329,476-489

Inuest. 34, 1345-1353

8408-8417

Fed. Proc. 38, 2934

SOC. Exp. Biol. Med. 155, 501-506

Lipid Res. 10,433-439

(1976) J. Clin. Inuest. 58, 667-680