- Email: [email protected]
The effect of bezofibrate and clofibrate on cholesterol accumulation, esterification and removal in cultured 3T3 fibroblasts
53
Atherosclerosis, 40 (1981) 5343 Elsevier/North-Holland Scientific
Publishers,
Ltd.
THE EFFECT OF BEZOFIBRATE AND CLOFIBRATE ON CHOLESTEROL ACCUMULATION, ESTERIFICATION AND REMOVAL IN CULTURED 3T3 FIBROBLASTS
KATRINA
HUDSON
Department
and ALLAN
of Physiology,
J. DAY
University of Melbourne,
(Received 5 January, 1981) (Revised, received 28 January, (Accepted 3 February, 1981)
Parkville, Vie 3052 (Australia)
1981)
summary 3T3 mouse fibroblasts were used to determine the effect of bezafibrate and clofibrate on the cellular metabolism of cholesterol. In cells incubated in normal medium these agents decreased the incorporation of 3H-labelled oleic acid relative to 14C-labelled linoleic acid into the cholesterol ester fraction. When the 3T3 fibroblasts were incubated with cationised low density lipoprotein (LDL) the amount of esterified cholesterol which accumulated in the cells was greatly increased. This accumulation of cholesterol ester was reduced by bezafibrate and clofibrate. These agents decreased the incorporation of both 3H-labelled oleic acid and 14C-labelled linoleic acid into the cholesterol ester fraction of the cells, with a preferential effect on oleic acid as indicated by a reduction in the 3H/14C ratio. When cells which had been preincubated with cationised LDL were reincubated in normal medium, the removal of esterified cholesterol from the cells was increased by both bezafibrate and clofibrate. The mechanism of the effects of these agents on the metabolism of cellular cholesterol is discussed, Key words:
Bezafibrate - Cationised low density lipoprotein - Cholesterol esterification Cholesterol removal - Clofibrate
Introduction Clofibrate [ Atromid-S,2-(4-chlorophenoxy)-2-methylpropionate] is commonly used to decrease elevated serum cholesterol and triacylglycerol concenSupport for this work was graduate Research Award. 0021-9150/81/0000-0000/$02.50
received from
Boehringer Mannheim GmbH. and a Commonwealth
0 1981 Elsevier/North-Holland
Scientific
Publishers,
Post-
Ltd.
54
trations [ 11. An analogue, bezafibrate (2-( 4-[ 2-( 4-chlorobenzamido)-ethyl] phenoxy)-a-methyl propionate) is also effective at lower dosages in decreasing elevated serum lipid levels 123. Clofibrate has also been shown to decrease the cholesterol content and severity of atherosclerotic lesions in cholesterol-fed rabbits [3,4] and to decrease the accumulation of lipid in skin and tendon xanthomas of hyperlipemic patients [ 51. The reduction in atherosclerosis brought about by these agents does not always correlate with a reduction in serum lipids [ 1,5] and the possibility that these agents alter the metabolism of lipid by the arterial wall cells cannot be discounted. The inhibition by clofibrate of arterial wall cholesterol esterifying enzymes support this possibility [ 61. In the present paper therefore the effect of bezafibrate and clofibrate on the accumulation and removal of the free and esterified cholesterol was investigated using 3T3 cells as a model cell system. Materials and Methods [9,10-3H]Oleic acid, specific activity 2.1 Ci/mmol, and [ 1-14C]linoleic acid, specific activity 51 mCi/mmol, were obtained from the Radiochemical Centre, Amersham, U.K. Bezafibrate was obtained from Boehringer Mannheim GmbH, F.R.G., and clofibrate from Imperial Chemical Industries Ltd., U.K. These agents were dissolved in 0.5 M sodium hydroxide and added to Dulbecco’s Modified Eagle Medium (Grand Island Biological Co., New York, NY) and the pH adjusted to 7.4 with hydrochloric acid. Blood was obtained from the jugular veins of 2 female pigs, 7 months old, which had been fed for 2 months on a diet containing 0.6% cholesterol and 6% beef tallow [ 71. After centrifugation at 6000 r.p.m. for 5 min at 4”C, the cholesterol content of the plasma was measured [ 81 as 368 mg/dl. After addition of EDTA 1 mg/ml plasma, LDL was obtained by ultracentrifugation (d = 1.006-1.063) according to the method described by Hatch and Lees [9] and was cationised as described by Basu et al. [lo]. In this reaction the free carboxyl groups of the protein are blocked by small molecules containing a positive charge at the unattached end. 3T3 Balb/c mouse fibroblasts, clone A31 from the American Type Culture Collection [ 111, were obtained from St. Vincent’s Hospital, Melbourne. Stock cells were grown in a humidified 5% CO, incubator at 37”C, in 75 cm2 flasks (Falcon Plastics, Los Angeles, U.S.A.). These were passaged twice weekly in a ratio of 1 : 3 to 1 : 5, and distributed into Falcon petri plates of area 25 cm2 for experiments. The incubation medium comprised Dulbecco’s Modified Eagle Medium supplemented with 10% inactivated foetal calf serum (Medos Co. Pty. Ltd.), glutamine 525.6 pg/ml, sodium penicillin G 100 units/ml and streptomycin sulphate 100 pg/ml (Sigma Chem. Co., St. Louis, MO). Experimental procedure Experiments were performed on cells at confluence. Three experiments are described here: uptake by non-stimulated cells, uptake by stimulated cells and removal by stimulated cells. In the first experiment, confluent cells were incubated for 48 h in 5 ml incubation medium containing 1.8 &i of 3H-labelled oleic acid and 0.1 FCi of
55
14C-labelled linoleic acid. Five groups; control, bezafibrate and clofibrate were compared, each at 2 concentrations (50 and 250 I-1glml). In the second experiment, confluent cells were incubated for 48 h in 5 ml incubation medium containing 2.9 PCi of 3H-labelled oleic acid and 0.61 PCi of 14C-labelled linoleic acid together with cationised hyperlipemic LDL 50 ,ug cholesterol/ml. Three groups were compared: control, bezafibrate 100 pg/ml and clofibrate 250 E.cg/ml. In the third experiment, confluent cells were incubated in medium containing 2.9 PCi of 3H-labelled oleic acid and 0.63 PCi of 14C-labelled linoleic acid together with cationised hyperlipemic LDL 50 pg cholesterol/ml. After 48 h the cells were washed 5 times with Hank’s solution containing calcium and magnesium [ 123, then incubated for a further 5 days in normal incubation medium containing bezafibrate 100 pug/ml, clofibrate 250 pg/ml or no agent. Cells were taken down for analysis at the end of the 48-h uptake period (day 0), and at days 1, 3 and 5. The incubation medium of the remaining cells was changed on day 3. Cells were washed 5 times with Hank’s solution containing calcium and magnesium, then incubated with 2 ml 0.25% trypsin (Difco Labs. Inc., MI) in calcium- and magnesium-free Hank’s solution for 20 min at 37°C. The cell suspension was transferred to siliconised glass centrifuge tubes. In the second and third experiments, cells from 2 petri plates were combined. Ten ml of Hank’s solution, containing calcium and magnesium, were added and they were centrifuged at 1000 r.p.m. for 10 min at 4°C. After 2 washes with Hank’s solution the cells were suspended in Hank’s solution and aliquots were taken for protein assay 1131. The cells were recentrifuged and the pellet extracted with 5 ml chloroform/methanol (2 : 1, v/v). The cell lipid extract was washed according to’the method of Folch [ 14],50 pg cholestane added as internal standard and the extract taken to 5 ml. Aliquots were taken for counting directly, lipid phosphorous assay [15], and for separation and subsequent radioactive and chemical assay of the fractions. For radioactive quantitation, neutral lipids were separated by thin layer chromatography on Silica Gel G (Merck Pty. Ltd.) using petroleum ether/diethyl ether/ acetic acid (60 : 15 : 1, v/v/v) as the developing solvent. After visualisation with 0.2% dichlorofluorescein in ethanol, the spots were scraped off and counted in a Beckman LS 9000 Counter using Snyder’s scintillator [16]. For chemical determination, neutral lipids were separated on chromorods using the solvent system petroleum ether/diethyl ether/formic acid (60 : 1.2 : 0.7, V/V/V). The spots were quantitatively scanned by an Iatroscan TH-10 Analyser (Iatron Laboratories Inc ., Japan).
Results
The effect of bezafibrate and clofibrate on the uptake and incorporation of 3H-labelled oleic acid and i4C-labelled linoleic acid into cellular lipid is shown in Table 1. The protein content of the cells ranged from 385 to 512 pg per petri plate. Considerable uptake and incorporation into combined lipid of both fatty
0.15 0.35 0.60
37
23
f f f
0.028 0.011 0.003
0.06 0.21 0.59
f 100
f f f
f
i
1.084 f 0.556 * 0.350 f
1.10 3.58 90.26
1659
3.27 5.52 87.38
601
385
0.918 0.518 0.350
1.42 3.05 91.80
1634
3.55 4.36 88.87
587
415
0.08
0.44 0.28 * 0.60
47
* * *
+ f c 0.047 0.020 0.005
0.20 0.22 0.33 *
* 121
* + f
*
f
**
* f f
0.657 0.396
f f
0.016 0.021 0.021
** ** **
0.30 * 0.73 ** 0.83 *
1.50 1.55 3.26
* f f
0.59 * 0.72 1.55
f 67
f f f
* 25 *
*22***
OLEIC AND
0.525 0.308
f f
0.029 0.008
**
0.893 f 0.029 **
2.25 3.93 88.79
1537
5.99 6.53 82.85
508
512
50 ccg/ml)
Clofibrate
OF 3H-LABELLED
0.76 1.26 ** 1.49 **
42
f 108
f f +
f
0.948 f
1.79 6.18 87.57
1495
4.34 10.27 81.59
591
Bezafibrate (250 pg/ml)
(50 I.cg/mu
AND INCORPORATION
Bezafibrate
s Incubation period of 48 h, single petri plates. bMean*SEM.n=6. C Dpmlmg protein/1000 dpm of the fatty acid in the incubation medium. * P < 0.05, ** P < 0.01 by Student t-test.
3H114C Ratio Cholesterol ester Triacylglycerol Phospholipid
[14CILino&ate Total uptake R Distribution Cholesterol ester Triacylglycerol Phospholipid
[3H101eote Total uptake % Distribution Cholesterol ester Triaculglycerol Phospholipid
Protein (pg)
Control
THE EFFECT OF BEZAFIBRATE AND CLOFIBRATE ON THE UPTAKE LEIC ACIDS IN NON-STIMULATED 3T3 FIBROBLASTS a3b.c
TABLE 1
1.127 0.759 0.432
1.13 8.42 86.77
1403
2.72 13.64 80.68
650
404
** **
0.08 0.73 ** 0.75 ** f 0.053 f 0.037 _+ 0.014
f f f
LINO-
0.17 ** 0.97 ** 1.00 ** * 58 *
f * f
f 24
* 40
(250 Pglml)
Clofibrate
14C-LABELLED
57 TABLE 2 THE EFFECT OF BEZAFIBRATE FIBROBLASTS
STIMULATED
AND CLOFIBRATE
BY CATIONISED
Control Protein (pg) Cholesterol ester Free cholesterol
635 f 60 117 * 4.5 35.4 * 4.2
ON CHOLESTEROL
ACCUMULATION
IN 3T3
LDL a,b,c Bezafibrate
Clofibrate
(100 I.cg/mu
(250 fig/ml)
680 * 43 75.0 k 7.2 *** 34.3 f 3.0
565 f 71 73.3 * 6.0 *** 27.0 * 2.0
a Incubation period 48 h, double petri plates. bMean*SEM,n=6. c Data is expressed as wg sterol/mg protein. *** P < 0.001 by Student t-test.
acids occurred. The presence of bezafibrate or clofibrate in the incubation medium did not appreciably influence the total uptake. In the control group approximately 90% of the label was found in the phospholipid fraction, with 4--5% in the triacylglycerol fraction and only l-3% in the cholesterol ester fraction. Clofibrate at 250 pg/ml decreased the proportion of ‘H-labelled oleic acid incorporated into the cholesterol ester fraction. Bezafibrate at 250 pg/ml and clofibrate at 50 pg/ml increased the proportion of 14C-labelled linoleic acid incorporated into the cholesterol ester fraction. Although these differences were statistically significant (P < 0.05), they were numerically small, as was the level of cholesterol ester synthesis. At 250 M/ml both agents increased the incorporation of the 2 fatty acids into the triacylglycerol fraction and decreased the incorporation of the 2 fatty acids into the phospholipid fraction. In order to compare the incorporation of 3H-labelled oleic acid into the various lipid fractions with that of 14C-labelled linoleic acid, the ratios of [3H] oleate to [ 14C]linoleate ( 3H/14C) were calculated for each fraction and are also shown in Table 1. In the control group, it is evident that oleic acid (relative to linoleic acid) was preferentially incorporated into the cholesterol ester fraction. Bezafibrate at both concentrations decreased the 3H/14C ratio of the cholesterol ester fraction, indicating decreased esterification with oleic acid relative to linoleic acid. A similar change was evident for clofibrate at the lower concentration but not.at the higher concentration. Incubation of 3T3 cells with cationised LDL at 50 c(g cholesterol/ml resulted in the accumulation of many small vesicles within the cells and recoverable amounts of esterified cholesterol above 100 pg/mg cell protein. As shown in Table 2, this large accumulation was markedly reduced by bezafibrate 100 ,ug/ ml and clofibrate 250 lg/ml. Table 3 shows that the labelled oleic and linoleic acids were taken up to a considerable extent by the cells and this uptake was reduced by both bezafibrate and clofibrate. In the control group, a larger proportion of both the 3H-labelled oleic acid and 14C-labelled linoleic acid were incorporated into the cholesterol ester fraction than in the non-stimulated cells. Bezafibrate and clofibrate at the concentrations used sighificantly reduced the incorporation of both 3H-labelled oleic acid and 14C-labelled linoleic acid into the cholesterol ester fraction. Both agents increased the relative incorporation
58 TABLE 3 THE EFFECT OF BEZAFIBRATE AND CLOFIBRATE ON THE DISTRIBUTION OF JH-LABELLED OLEIC AND l”C-LABELLED LINOLEIC ACIDS IN LIPIDS IN 3T3 FIBROBLASTS STIMULATED BY CATIONISED LDL a,b,= Control
13HIOleate Total uptake c % Distribution ’ Cholesterol ester Triacylglycerol Phospholipid C14C/Linoleate Total uptake c % Distribution Cholesterol ester Triacylglycerol Phospholipid 3H/14C
Ratio Cholesterol ester Triacylglycerol Phospholipid
488 36.08 9.81 46.36 105 12.25 6.53 71.12 2.05 1.14 0.454
f 45 * f f
*
1.73 1.08 2.17
70
f f f
0.54 0.86 1.20
f i f
0.05 0.10 0.022
Bezafibrate
Clofibrate
(100 fig/ml)
(250 rgiml)
209 24.80 14.42 52.07 302 9.13 9.93 72.22 1.86 1.04 0.499
f 35 ** f f f *35
1.47 ;*** 2.22 * 0.84 * **
349 19.78 25.12 47.02 426
f 23 * f f *
2.32 *** 3.18 *** 1.15
f 35 ** f f f
0.36 *** 3.22 ** 3.05 *
f * f
0.20 *** 1.75 1.69
8.98 18.67 64.42
f f f
0.06 * 0.04 0.023
1.77 f 0.10 * 1.16 f 0.03 0.693 f 0.031 *
a Incubation period 48 h. double petri plates. bMean*SEM,n=6. c Dpm/mg protein/1000 dpm of the fatty acid in the incubation medium. * P < 0.05, ** P < 0.01. *** P < 0.001 by Student t-test.
of 3H-labelled oleic acid into the triacylglycerol fraction of the cell lipids. From comparison of the 3H/14C ratios of the various lipid fractions in the control group, it is seen that more oleic acid relative to linoleic acid was incorporated into the cholesterol ester fraction than was the case for the triacylglycerol and phospholipid fractions. Both bezafibrate and clofibrate significantly decreased the 3H/14C ratio of the cholesterol ester fraction indicating that, for both these agents, the incorporation of oleic acid into cholesterol ester was reduced relative to that of linoleic acid. Figure 1 shows the free and esterified cholesterol content of the cells expressed as pg/mg cell protein. During the incubation period the protein and phospholipid content of the cells remained constant at 610 k 31 pg and 139 f 8 E.cg/mgcell protein, respectively (n = 22). After the cells had been stimulated with cationised LDL at 50 pg cholesterol/ml, and then incubated in normal incubation medium, the free cholesterol content of the cells remained relatively constant and was not significantly affected by bezafibrate or clofibrate. On the other hand, the esterified cholesterol content of the cells decreased with time. Analysis of variance of each of the 3 lines shown indicates that the fitted linear regression lines were statistically significant (P < 0.01). The slope of the line corresponding to the bezafibrate group is greater (P < 0.05) than that of the control group. Although the slope of the line corresponding to the clofibrate group does not reach statistical significance against that of the control group, it is seen to be between the control group and the bezafibrate group. When a logarithmic transformation was performed on the data, both lines were signifi-
59
0
1
3
5
Fig. 1. Effect of clofibrate and bezafibrate on removal of free and esterified blasts stimulated with cationised low density lipoprotein. -0 control, ml. *- - - - - -* bezafibrate 100 ~g/ml.
cholesterol from 3T3 fibro* -. - .mclofibrate 250 ~g/
??
Fig. 2. Effect of clofibrate and bezafibrate on removal of [3H]oleic acid and [ 14C]linoleic acid labelled cholesterol ester from 3T3 fibroblasts stimulated with cationised low density lipoprotein, e-----o control, ??. - * - a clofibrate 250 /.&/ml. *- - - - - -* bezafibrate 100 fig/ml.
cantly different from the control (bezafibrate P < 0.01, clofibrate P < 0.05). Figure 2 shows the removal of 3H-labelled oleic acid and 14C-labelled linoleic acid from the cellular cholesterol ester fraction. The data is plotted semi-logarithmically against the period of incubation in normal medium. The 3H-labelled oleic acid was rapidly removed from the cholesterol ester fraction of these cells. Each of these regression lines calculated from the data fit a straight line whose slope is different from zero (P< 0.01). The slopes corresponding to the bezafibrate and clofibrate groups are different from that of the control group (P< 0.01and P < 0.05, respectively). The 14C-labelled linoleic acid was more slowly removed from the cellular cholesterol ester fraction. The data for the control group, however, does not fit a line whose slope is different from zero. The data for the bezafibrate and clofibrate groups do fit straight lines whose slope is different from zero (P< 0.01 and P < 0.05, respectively). For all 3 groups there was a preferential removal of oleate relative to linoleate from the cholesterol ester fraction of these cells, however bezafibrate and clofibrate had no effect on this preferential removal pattern. Discussion The concentrations of bezafibrate and clofibrate used in these experiments were chosen to be in the range of the therapeutically achieved levels of these agents [17,18]. 3T3 mouse fibroblasts were used as a model system to determine the effect of these agents on cholesterol metabolism at the cellular level. In these experiments, the confluent 3T3 fibroblasts were in a steady state with
60
respect to protein and phospholipid content. However, when cationised LDL was added to or removed from the incubation medium, the cells were not in a steady state with respect to total cholesterol content. Basu et al. have described the cellular metabolism of cationised LDL using human fibroblasts [10,19]. Nevertheless, cationised LDL was similarly metabolised by aortic smooth muscle cells [ 20,211 and endothelial cells in culture [ 221. In the non-stimulated 3T3 fibroblasts the extent of cholesterol esterification was quite small -- less than 6% of the labelled fatty acids taken up by the cells were incorporated into cholesterol ester. Similar levels of incorporation of 14Clabelled oleic acid into the cholesterol ester fraction have been observed in aortic intimal and medial cells of normal and cholesterol-fed rabbits [ 231. Bezafibrate at 50 ,ug/ml and 250 I.cg/ml, and clofibrate at 50 pg/ml decreased the incorporation of oleic acid relative to linoleic acid into the cellular cholesterol ester fraction. However, in these non-stimulated 3T3 cells the content of esterified cholesterol was too small to be determined accurately, so cationised LDL was used to produce cells containing large amounts of esterified cholesterol. These cells also demonstrated a marked stimulation of cholesterol esterification compared with cells incubated in normal medium. The incorporation of labelled oleic acid into the cholesterol ester fraction was similar to that observed in foam cells from atherosclerotic lesions [ 241. In the 3T3 fibroblasts stimulated with cationised LDL, bezafibrate and clofibrate reduced the accumulation of esterified cholesterol. These agents decreased the uptake of oleic and linoleic acids and their incorporation into cholesterol ester, They also reduced the incorporation of oleic acid relative to linoleic acid into the cellular cholesterol ester fraction. When the stimulated cells were incubated in normal medium, esterified cholesterol was removed from the cells and the removal rate was increased by both bezafibrate and clofibrate. While these agents increased the rate of removal of labelled oleic and linoleic acids from the cholesterol ester fraction, they did not affect the removal of labelled oleic acid relative to linoleic acid from cholesterol ester. The mechanism of these effects remains to be determined. Cationised LDL binds to non-specific negative charges on the cell membrane, undergoes adsorptive endocytosis and is transported to the cellular lysosomes where cholesterol ester is hydrolysed. The free cholesterol is then liberated into the cytoplasm where it may be transferred to cellular membranes or esterified While both the cholesterol ester and accumulates in lipid droplets [ 10,19-221. taken up and the free cholesterol esterified in the cell contribute to the cellular cholesterol ester, hydrolysis of cholesterol ester and any associated transport of free cholesterol out of the cell reduce the cellular content of cholesterol ester. The possible mechanisms of action of these agents, therefore, include reduction in the uptake of cationised LDL, suppression of esterification, stimulation of hydrolysis and of removal of the liberated free cholesterol. Although there is no information on the effects of these agents on the binding or internalisation of cationised LDL, clofibrate has no effect on the surface binding of normal LDL by human aortic smooth muscle cells in culture [25]. However, the reduction in the ‘H/14C ratio of the cholesterol ester fraction by bezafibrate and clofibrate suggests that the metabolism of endogenously esterified cholesterol is altered by these agents. The decreased incorporation of oleic
61
and linoleic acids in cholesterol ester may be due to decreased esterification or increased hydrolysis of endogenously synthesised cholesterol ester. Two different cholesterol esterifying enzyme systems have been described in foam cells from atherosclerotic rabbit aorta [26]. One has a pH optimum of 5 and does not require cofactors. This corresponds to lysosomal sterol ester hydrolase (E.C. 3.1.1.13). In rat liver lysosomes, the activity of this enzyme has been shown to be reversible, catalysing both the formation and hydrolysis of cholesterol ester [2’7]. The other enzyme system has a pH optimum of 7.5 and absolute requirement for ATP and coenzyme A. This involves the microsomal fatty acyl coenzyme A synthetase (E.C. 6.2.1.3) and fatty acyl coenzymeA: cholesterol acyl transferase (E.C. 2.3.1.26). The activity of a cytoplasmic cholesterol hydrolase enzyme has been demonstrated in human fibroblasts [28] and monkey aortic smooth muscle cells [ 291. The microsomal esterifying system and lysosomal cholesterol esterase have been demonstrated in human fibroblasts [30,31] aortic smooth muscle cells [29,32,33] and macrophages [34,35]. The microsomal esterifying system demonstrates a preference for oleic acid [ 361 while the microsomal cholesterol esterase hydrolyses cholesterol oleate and cholesterol linoleate at similar rates [ 371, as does the lysosomal cholesterol e&erase [ 33,381. Since oleic acid is preferentially esterified to cholesterol, the greater decrease in the incorporation of oleic acid relative to linoleic acid into the cholesterol ester fraction by bezafibrate and clofibrate in the stimulated cells, would be consistent with a suppression of microsomal esterification of cholesterol by these agents. In the atherosclerotic rabbit aorta, Brecher and Chobanian have demonstrated a marked inhibition of the microsomal esterifying system by clofibrate [6]. The mechanism for the reduction in cellular esterified cholesterol by bezafibrate and clofibrate may, therefore, be due to an inhibition in the activity of the microsomal esterifying system. The enzymes that could be inhibited are fatty acyl coenzyme A synthetase or fatty acyl coenzyme A: cholesterol acyl transferase. If there is no change in the uptake of cationised LDL by these agents, the free cholesterol liberated by lysosomal hydrolysis must be removed from the cells rather than esterified by the microsomes. Free cholesterol has been shown to be removed’ from cells in culture, however esterified cholesterol is not directly removed from cells but requires to be hydrolysed prior to removal [38,39]. The effect of bezafibrate and clofibrate on cholesterol ester hydrolysis is not known, however, stimulation of lysosomal or microsomal hydrolysis and removal of the liberated free cholesterol by these agents cannot be excluded as a mechanism contributing to the reduction in cellular esterified cholesterol. When 3T3 fibroblasts which had been stimulated with cationised LDL were incubated in normal medium, cholesterol ester was removed from the cells and there was a decrease with time in the relative amount of 3H-labelled oleic to 14C-labelled linoleic acid in the cholesterol ester fraction. This is consistent with a decrease in the microsomal esterifying activity during this period of removal. If this esterifying activity was further depressed by bezafibrate or clofibrate, a further decrease in the amount of ‘H-labelled oleic relative to “C-labelled linoleic acid would be expected. The fact that this did not occur may, there-
62
fore, indicate a stimulation of hydrolysis of endogenously esterified cholesterol by these agents. Further investigation is, however, required to determine the effects of these agents on cytoplasmic and lysosomal hydrolysis of esterified cholesterol. The removal of cellular free cholesterol is increased by serum high density lipoprotein (HDL) [ 401, so that a possible mechanism for the increased removal by bezafibrate and clofibrate may be that serum HDL is rendered a more effective acceptor for cholesterol. This possibility is currently being explored. Acknowledgements The authors wish to thank Miss Trudi Harris for excellent technical assistance and Dr. P. Singh for assistance with the Iatroscan determinations. References 1 2 3
4
6 6 7
8 9 10
11 12 13 14 16 16 17 18 19 20
Oliver, M.F., Heady, J.A. and Morris, J.N., A cooperative trial in the primary prevention of ischaemic heart disease using clofibrate, Brit. Heart J., 40 (1978) 1069. Olsson, A.G. and Lang, P.D.. Dose--response study of bezafibrate on serum lipoprotein concentrations in hyperhpoproteinaemia, Atherosclerosis, 31 (1978) 421. Kritchevsky, D.. Sailata, P. and Tepper. S.A.. Infiuence of ethyl pshlorophenoxyisobutyrate (CPIB) upon establishment and Progression of experimental atherosclerosis in rabbits, J. Atheroscler. Res., 8 (1968) 766. Sate. M. and Takano, T.. Fatty acid composition of cholesteryl esters in cholesterol-fed rabbits treated with lipid-lowering agent ethyl 2-@&lorophenoxy) isobutyrate, Jap. J. Pharmacol., 29 (1979) 261. Buxtorf, J.C.. Beaumont, V.. Jacotot. B. and Beaumont, J.-L., Regression de xsnthomes et medicamerits hypoIipidemiants, Atherosclerods, 17 (1974) 1. Brecher. P.I. and Chobanian, A.V., Cholesterol ester synthesis in normal and atherosclerotic aortas of rabbits and rhesus monkeys, C&c. Res., 35 (1974) 692. Lee, K.T., Jarmolych, J., Kim, D.N.. Grant, C., Krasnev, J.A., Thomas, W.A. and BNnO. A.M., Production of advanced coronary atherosclerosis, myocardial infarction and “sudden death” in swine, Exp. Mol. Path., 16 (1971) 170. Watkis. A.. Zak. B. and Boyle, A.J.. A new method for the direct determination of serum cholesterol, J. Lab. CIin. Med., 41 (1953) 486. Hatch, F.T. and Lees, R.S., Practical methods for plasma Iipoprotein analysis, Advances in Lipid Research, 6 (1968) 1. Bssu, S.K.* Goldstein, J.L., Anderson, R.G.W., and Brown, MS., Degradation of c&ionized low density lipoprotein and reguI&ion of cholesterol metabolism in homozygous familial hypercholesterolemia fibroblasts, Proc. Nat. Acad. Scl. (USA), 73 (1976) 3178. Goldberg. B.. Collagen synthesis as a marker for cell type in mouse ST3 lines, Cell, 11 (1977) 169. Hanks, J.H.. The longevity of chick tissue cultures without renewal of medium, J. CeU. Camp. Physiol., 31 (1948) 285. Lowry, O.H., Rosebrough. N.J., Farr. A.L. and Randall. R.J.. Protein measurement with the FoIin phenol reagent, J. Biol. Chem.. 193 (1961) 266. Folch, J., Lees, M. and Sloane Stanley, G.H., A simple method for the isolation and purification of totlil lipides from animal tissue, J. Biol. Chem., 226 (1961) 497. Itaya, K. and Ui. M.. A new micromethod for the calorimetric determination of inorganic phosphate, CIin. Chim. Acta, 14 (1966) 361. Snyder, F.. Radioassay of thin layer chromatogram - A high resolution zonal scraper for auantitative C14 and H3 scanning of thin layer chromatograms, Anal. Biochem., 9 (1964) 183. Thorp, J.M., Experimental evaluation of an orally active combination of androsterone with ethyl chlorophenoxyisobutyrate. Lance& i (1962) 1323. Endele, R.. A gas chromatographic method for the determination of bezafibrate in serum and urine, J. Chromatog., lb4 (1978) 261. Basu. S.K.. Anderson, R.G.W., Goldstein, J.L. and Brown, M.S., MetaboIism of cationized Bpoproteins by human fibroblasts, J. Cell Biol.. 74 (1977) 119. Stein, 0.. Goren, R. and Stein, Y.. Removal of cholesterol from fibroblasts and smooth muscle c&s in culture in the presence and absence of cholesterol esterification in the medium, Biochim. Biophys. Acta, 629 (1978) 309.
63 21
22
23 24 25 26 27 28 29
30 31 32 33 34 35 36 37 39 39 40
Goldstein, J.L.. smooth muscle
Anderson, R.G.W.. Buja. L.M.. Basu. SK. and Brown, MS.. Overloading cells with low density lipoprotein-cholesteryl esters reproduces features
human aortic of atheroscle-
rosis in vitro, J. Clln. Invest., 59 (1977) 1196. D. and Fielding, C.J., Effect of contact inhibition on Fielding, P.E.. Vlodavsky. I., Gospodarowicz, the regulation of cholesterol metabolism in cultured vascular endothelial cells, J. Biol. Chem., 254 (1979) 749. Day, A.J., Lipid metabolism by rabbit aortic intimsl and medial cells in tissue culture, Vircbows Arch. A. Path. Anat. Histol., 362 (1974) 89. Day, A.J. and Tume, R.K.. Incorporation of [l-14Cloleic acid into lipid by foam cells and by other fractions separated from rabbit atherosclerotic lesions, Atherosclerosis, 11 (1970) 291. Filipovic, I. and Buddecke, E., p-Chlorophenoxyisobutyrate enhanced retention of homologous lipoproteins by human aortic smooth muscle cells, Lipids, 12 (1977) 1069. Proudlock, J.W., Day. A.J. and Tume. R.K., Cholesterol esterifying enzymes of foam cells isolated from atherosclerotic rabbit intima. Atherosclerosis, 18 (1973) 451. Nilsson, A., Nordon. H. and Wilbehnsson, L., Hydrolysis and formation of cholesterol esters with rat liver lysosomes, Biochim. Biophys. Acta, 296 (1973) 593. Goldstein, J.L.. Dana, SE., Faust, J.R., Beaudet, A.L. and Brown, MS.. Role of lysosomal acid lipase in the metabolism of plasma low density lipoprotein. J. Biol. Chem.. 250 (1975) 8487. Riddle, MC., Fujimoto, W. and Ross, R., Two cholesterol ester hydrolases -Distribution in rat tissues and in cultured human fibroblasts and monkey arterial smooth muscle cells, Biochim. Biophys. Acta, 488 (1977) 359. Brown, MS., Dana, SE. and Goldstein, J.L., Cholesterol ester formation in cultured human fibroblasts, J. Biol. Chem., 250 (1975) 4025. Brown, M.S.. Dana, S.E. and Goldstein, J.L., Receptor-dependent hydrolysis of cholesteryl esters contained in plasma low density lipoprotein, Proc. Nat. Acad. Sci. (USA), 72 (1975) 2925. St. Clair. R.W.. Metabolism of the arterial wall and atherosclerosis. In: R. Paoletti and A.M. Gotto, Jr. (Eds.). Atherosclerosis Reviews, Vol. 1, Raven Press, New York, 1976, p. 61. Takano, T., Black, W.J., Peters, T.J. and de Duve. C., Assays, kinetics, and lysosomal localization of an acid cholesterase in rabbit aortic smooth muscle cells, J. Biol. Chem.. 249 (1974) 6732. Day, A.J. and Tume. R.K.. Cholesterol*sterifying activity of cell-free preparations of rabbit peritoneal macrophages. Biochim. Biophys. Acta. 176 (1969) 367. Werb, Z. and Cohn, Z.A.. Cholesterol metabolism in the macrophage. Part 3 (Ingestion and intracellular fate of cholesterol and cholesterol esters), J. EXP. Med., 136 (1972) 21. Bowyer. D.E., Howard, A.N. and Gresham, G.A., Lipid synthesis ln perfused normal and atherosclerotic rabbit aortas, Biochem. J., 103 (1967) 54. Deykin, D. and Goodman, D.S.. The hydrolysis of long-chain fatty acid esters of cholesterol with rat liver enzymes. J. Biol. Chem.. 237 (1962) 3649. Stein, Y., Halperin, G. and Stein, 0.. Intralysosomal hydrolysis of cholesterol esters of varying fatty acid CornPosition in cultured human skin fibroblasts, Biochim. Biophys. Acta, 530 (1978) 420. Chen, R.M. and Fischer-Dzoga, K.. Effect of byperlipemic serum lipoproteins on the lipid accumulation and cholesterol flux of rabbit aortic medial cells. Atherosclerosis, 28 (1977) 339. Stein, 0. and Stein, Y.. The removal of cholesterol from Landschiitz ascites cells by high-density apolipoprotein, Biocbim. Biophys. Acta, 326 (1973) 232.
Atherosclerosis, 40 (1981) 5343 Elsevier/North-Holland Scientific
Publishers,
Ltd.
THE EFFECT OF BEZOFIBRATE AND CLOFIBRATE ON CHOLESTEROL ACCUMULATION, ESTERIFICATION AND REMOVAL IN CULTURED 3T3 FIBROBLASTS
KATRINA
HUDSON
Department
and ALLAN
of Physiology,
J. DAY
University of Melbourne,
(Received 5 January, 1981) (Revised, received 28 January, (Accepted 3 February, 1981)
Parkville, Vie 3052 (Australia)
1981)
summary 3T3 mouse fibroblasts were used to determine the effect of bezafibrate and clofibrate on the cellular metabolism of cholesterol. In cells incubated in normal medium these agents decreased the incorporation of 3H-labelled oleic acid relative to 14C-labelled linoleic acid into the cholesterol ester fraction. When the 3T3 fibroblasts were incubated with cationised low density lipoprotein (LDL) the amount of esterified cholesterol which accumulated in the cells was greatly increased. This accumulation of cholesterol ester was reduced by bezafibrate and clofibrate. These agents decreased the incorporation of both 3H-labelled oleic acid and 14C-labelled linoleic acid into the cholesterol ester fraction of the cells, with a preferential effect on oleic acid as indicated by a reduction in the 3H/14C ratio. When cells which had been preincubated with cationised LDL were reincubated in normal medium, the removal of esterified cholesterol from the cells was increased by both bezafibrate and clofibrate. The mechanism of the effects of these agents on the metabolism of cellular cholesterol is discussed, Key words:
Bezafibrate - Cationised low density lipoprotein - Cholesterol esterification Cholesterol removal - Clofibrate
Introduction Clofibrate [ Atromid-S,2-(4-chlorophenoxy)-2-methylpropionate] is commonly used to decrease elevated serum cholesterol and triacylglycerol concenSupport for this work was graduate Research Award. 0021-9150/81/0000-0000/$02.50
received from
Boehringer Mannheim GmbH. and a Commonwealth
0 1981 Elsevier/North-Holland
Scientific
Publishers,
Post-
Ltd.
54
trations [ 11. An analogue, bezafibrate (2-( 4-[ 2-( 4-chlorobenzamido)-ethyl] phenoxy)-a-methyl propionate) is also effective at lower dosages in decreasing elevated serum lipid levels 123. Clofibrate has also been shown to decrease the cholesterol content and severity of atherosclerotic lesions in cholesterol-fed rabbits [3,4] and to decrease the accumulation of lipid in skin and tendon xanthomas of hyperlipemic patients [ 51. The reduction in atherosclerosis brought about by these agents does not always correlate with a reduction in serum lipids [ 1,5] and the possibility that these agents alter the metabolism of lipid by the arterial wall cells cannot be discounted. The inhibition by clofibrate of arterial wall cholesterol esterifying enzymes support this possibility [ 61. In the present paper therefore the effect of bezafibrate and clofibrate on the accumulation and removal of the free and esterified cholesterol was investigated using 3T3 cells as a model cell system. Materials and Methods [9,10-3H]Oleic acid, specific activity 2.1 Ci/mmol, and [ 1-14C]linoleic acid, specific activity 51 mCi/mmol, were obtained from the Radiochemical Centre, Amersham, U.K. Bezafibrate was obtained from Boehringer Mannheim GmbH, F.R.G., and clofibrate from Imperial Chemical Industries Ltd., U.K. These agents were dissolved in 0.5 M sodium hydroxide and added to Dulbecco’s Modified Eagle Medium (Grand Island Biological Co., New York, NY) and the pH adjusted to 7.4 with hydrochloric acid. Blood was obtained from the jugular veins of 2 female pigs, 7 months old, which had been fed for 2 months on a diet containing 0.6% cholesterol and 6% beef tallow [ 71. After centrifugation at 6000 r.p.m. for 5 min at 4”C, the cholesterol content of the plasma was measured [ 81 as 368 mg/dl. After addition of EDTA 1 mg/ml plasma, LDL was obtained by ultracentrifugation (d = 1.006-1.063) according to the method described by Hatch and Lees [9] and was cationised as described by Basu et al. [lo]. In this reaction the free carboxyl groups of the protein are blocked by small molecules containing a positive charge at the unattached end. 3T3 Balb/c mouse fibroblasts, clone A31 from the American Type Culture Collection [ 111, were obtained from St. Vincent’s Hospital, Melbourne. Stock cells were grown in a humidified 5% CO, incubator at 37”C, in 75 cm2 flasks (Falcon Plastics, Los Angeles, U.S.A.). These were passaged twice weekly in a ratio of 1 : 3 to 1 : 5, and distributed into Falcon petri plates of area 25 cm2 for experiments. The incubation medium comprised Dulbecco’s Modified Eagle Medium supplemented with 10% inactivated foetal calf serum (Medos Co. Pty. Ltd.), glutamine 525.6 pg/ml, sodium penicillin G 100 units/ml and streptomycin sulphate 100 pg/ml (Sigma Chem. Co., St. Louis, MO). Experimental procedure Experiments were performed on cells at confluence. Three experiments are described here: uptake by non-stimulated cells, uptake by stimulated cells and removal by stimulated cells. In the first experiment, confluent cells were incubated for 48 h in 5 ml incubation medium containing 1.8 &i of 3H-labelled oleic acid and 0.1 FCi of
55
14C-labelled linoleic acid. Five groups; control, bezafibrate and clofibrate were compared, each at 2 concentrations (50 and 250 I-1glml). In the second experiment, confluent cells were incubated for 48 h in 5 ml incubation medium containing 2.9 PCi of 3H-labelled oleic acid and 0.61 PCi of 14C-labelled linoleic acid together with cationised hyperlipemic LDL 50 ,ug cholesterol/ml. Three groups were compared: control, bezafibrate 100 pg/ml and clofibrate 250 E.cg/ml. In the third experiment, confluent cells were incubated in medium containing 2.9 PCi of 3H-labelled oleic acid and 0.63 PCi of 14C-labelled linoleic acid together with cationised hyperlipemic LDL 50 pg cholesterol/ml. After 48 h the cells were washed 5 times with Hank’s solution containing calcium and magnesium [ 123, then incubated for a further 5 days in normal incubation medium containing bezafibrate 100 pug/ml, clofibrate 250 pg/ml or no agent. Cells were taken down for analysis at the end of the 48-h uptake period (day 0), and at days 1, 3 and 5. The incubation medium of the remaining cells was changed on day 3. Cells were washed 5 times with Hank’s solution containing calcium and magnesium, then incubated with 2 ml 0.25% trypsin (Difco Labs. Inc., MI) in calcium- and magnesium-free Hank’s solution for 20 min at 37°C. The cell suspension was transferred to siliconised glass centrifuge tubes. In the second and third experiments, cells from 2 petri plates were combined. Ten ml of Hank’s solution, containing calcium and magnesium, were added and they were centrifuged at 1000 r.p.m. for 10 min at 4°C. After 2 washes with Hank’s solution the cells were suspended in Hank’s solution and aliquots were taken for protein assay 1131. The cells were recentrifuged and the pellet extracted with 5 ml chloroform/methanol (2 : 1, v/v). The cell lipid extract was washed according to’the method of Folch [ 14],50 pg cholestane added as internal standard and the extract taken to 5 ml. Aliquots were taken for counting directly, lipid phosphorous assay [15], and for separation and subsequent radioactive and chemical assay of the fractions. For radioactive quantitation, neutral lipids were separated by thin layer chromatography on Silica Gel G (Merck Pty. Ltd.) using petroleum ether/diethyl ether/ acetic acid (60 : 15 : 1, v/v/v) as the developing solvent. After visualisation with 0.2% dichlorofluorescein in ethanol, the spots were scraped off and counted in a Beckman LS 9000 Counter using Snyder’s scintillator [16]. For chemical determination, neutral lipids were separated on chromorods using the solvent system petroleum ether/diethyl ether/formic acid (60 : 1.2 : 0.7, V/V/V). The spots were quantitatively scanned by an Iatroscan TH-10 Analyser (Iatron Laboratories Inc ., Japan).
Results
The effect of bezafibrate and clofibrate on the uptake and incorporation of 3H-labelled oleic acid and i4C-labelled linoleic acid into cellular lipid is shown in Table 1. The protein content of the cells ranged from 385 to 512 pg per petri plate. Considerable uptake and incorporation into combined lipid of both fatty
0.15 0.35 0.60
37
23
f f f
0.028 0.011 0.003
0.06 0.21 0.59
f 100
f f f
f
i
1.084 f 0.556 * 0.350 f
1.10 3.58 90.26
1659
3.27 5.52 87.38
601
385
0.918 0.518 0.350
1.42 3.05 91.80
1634
3.55 4.36 88.87
587
415
0.08
0.44 0.28 * 0.60
47
* * *
+ f c 0.047 0.020 0.005
0.20 0.22 0.33 *
* 121
* + f
*
f
**
* f f
0.657 0.396
f f
0.016 0.021 0.021
** ** **
0.30 * 0.73 ** 0.83 *
1.50 1.55 3.26
* f f
0.59 * 0.72 1.55
f 67
f f f
* 25 *
*22***
OLEIC AND
0.525 0.308
f f
0.029 0.008
**
0.893 f 0.029 **
2.25 3.93 88.79
1537
5.99 6.53 82.85
508
512
50 ccg/ml)
Clofibrate
OF 3H-LABELLED
0.76 1.26 ** 1.49 **
42
f 108
f f +
f
0.948 f
1.79 6.18 87.57
1495
4.34 10.27 81.59
591
Bezafibrate (250 pg/ml)
(50 I.cg/mu
AND INCORPORATION
Bezafibrate
s Incubation period of 48 h, single petri plates. bMean*SEM.n=6. C Dpmlmg protein/1000 dpm of the fatty acid in the incubation medium. * P < 0.05, ** P < 0.01 by Student t-test.
3H114C Ratio Cholesterol ester Triacylglycerol Phospholipid
[14CILino&ate Total uptake R Distribution Cholesterol ester Triacylglycerol Phospholipid
[3H101eote Total uptake % Distribution Cholesterol ester Triaculglycerol Phospholipid
Protein (pg)
Control
THE EFFECT OF BEZAFIBRATE AND CLOFIBRATE ON THE UPTAKE LEIC ACIDS IN NON-STIMULATED 3T3 FIBROBLASTS a3b.c
TABLE 1
1.127 0.759 0.432
1.13 8.42 86.77
1403
2.72 13.64 80.68
650
404
** **
0.08 0.73 ** 0.75 ** f 0.053 f 0.037 _+ 0.014
f f f
LINO-
0.17 ** 0.97 ** 1.00 ** * 58 *
f * f
f 24
* 40
(250 Pglml)
Clofibrate
14C-LABELLED
57 TABLE 2 THE EFFECT OF BEZAFIBRATE FIBROBLASTS
STIMULATED
AND CLOFIBRATE
BY CATIONISED
Control Protein (pg) Cholesterol ester Free cholesterol
635 f 60 117 * 4.5 35.4 * 4.2
ON CHOLESTEROL
ACCUMULATION
IN 3T3
LDL a,b,c Bezafibrate
Clofibrate
(100 I.cg/mu
(250 fig/ml)
680 * 43 75.0 k 7.2 *** 34.3 f 3.0
565 f 71 73.3 * 6.0 *** 27.0 * 2.0
a Incubation period 48 h, double petri plates. bMean*SEM,n=6. c Data is expressed as wg sterol/mg protein. *** P < 0.001 by Student t-test.
acids occurred. The presence of bezafibrate or clofibrate in the incubation medium did not appreciably influence the total uptake. In the control group approximately 90% of the label was found in the phospholipid fraction, with 4--5% in the triacylglycerol fraction and only l-3% in the cholesterol ester fraction. Clofibrate at 250 pg/ml decreased the proportion of ‘H-labelled oleic acid incorporated into the cholesterol ester fraction. Bezafibrate at 250 pg/ml and clofibrate at 50 pg/ml increased the proportion of 14C-labelled linoleic acid incorporated into the cholesterol ester fraction. Although these differences were statistically significant (P < 0.05), they were numerically small, as was the level of cholesterol ester synthesis. At 250 M/ml both agents increased the incorporation of the 2 fatty acids into the triacylglycerol fraction and decreased the incorporation of the 2 fatty acids into the phospholipid fraction. In order to compare the incorporation of 3H-labelled oleic acid into the various lipid fractions with that of 14C-labelled linoleic acid, the ratios of [3H] oleate to [ 14C]linoleate ( 3H/14C) were calculated for each fraction and are also shown in Table 1. In the control group, it is evident that oleic acid (relative to linoleic acid) was preferentially incorporated into the cholesterol ester fraction. Bezafibrate at both concentrations decreased the 3H/14C ratio of the cholesterol ester fraction, indicating decreased esterification with oleic acid relative to linoleic acid. A similar change was evident for clofibrate at the lower concentration but not.at the higher concentration. Incubation of 3T3 cells with cationised LDL at 50 c(g cholesterol/ml resulted in the accumulation of many small vesicles within the cells and recoverable amounts of esterified cholesterol above 100 pg/mg cell protein. As shown in Table 2, this large accumulation was markedly reduced by bezafibrate 100 ,ug/ ml and clofibrate 250 lg/ml. Table 3 shows that the labelled oleic and linoleic acids were taken up to a considerable extent by the cells and this uptake was reduced by both bezafibrate and clofibrate. In the control group, a larger proportion of both the 3H-labelled oleic acid and 14C-labelled linoleic acid were incorporated into the cholesterol ester fraction than in the non-stimulated cells. Bezafibrate and clofibrate at the concentrations used sighificantly reduced the incorporation of both 3H-labelled oleic acid and 14C-labelled linoleic acid into the cholesterol ester fraction. Both agents increased the relative incorporation
58 TABLE 3 THE EFFECT OF BEZAFIBRATE AND CLOFIBRATE ON THE DISTRIBUTION OF JH-LABELLED OLEIC AND l”C-LABELLED LINOLEIC ACIDS IN LIPIDS IN 3T3 FIBROBLASTS STIMULATED BY CATIONISED LDL a,b,= Control
13HIOleate Total uptake c % Distribution ’ Cholesterol ester Triacylglycerol Phospholipid C14C/Linoleate Total uptake c % Distribution Cholesterol ester Triacylglycerol Phospholipid 3H/14C
Ratio Cholesterol ester Triacylglycerol Phospholipid
488 36.08 9.81 46.36 105 12.25 6.53 71.12 2.05 1.14 0.454
f 45 * f f
*
1.73 1.08 2.17
70
f f f
0.54 0.86 1.20
f i f
0.05 0.10 0.022
Bezafibrate
Clofibrate
(100 fig/ml)
(250 rgiml)
209 24.80 14.42 52.07 302 9.13 9.93 72.22 1.86 1.04 0.499
f 35 ** f f f *35
1.47 ;*** 2.22 * 0.84 * **
349 19.78 25.12 47.02 426
f 23 * f f *
2.32 *** 3.18 *** 1.15
f 35 ** f f f
0.36 *** 3.22 ** 3.05 *
f * f
0.20 *** 1.75 1.69
8.98 18.67 64.42
f f f
0.06 * 0.04 0.023
1.77 f 0.10 * 1.16 f 0.03 0.693 f 0.031 *
a Incubation period 48 h. double petri plates. bMean*SEM,n=6. c Dpm/mg protein/1000 dpm of the fatty acid in the incubation medium. * P < 0.05, ** P < 0.01. *** P < 0.001 by Student t-test.
of 3H-labelled oleic acid into the triacylglycerol fraction of the cell lipids. From comparison of the 3H/14C ratios of the various lipid fractions in the control group, it is seen that more oleic acid relative to linoleic acid was incorporated into the cholesterol ester fraction than was the case for the triacylglycerol and phospholipid fractions. Both bezafibrate and clofibrate significantly decreased the 3H/14C ratio of the cholesterol ester fraction indicating that, for both these agents, the incorporation of oleic acid into cholesterol ester was reduced relative to that of linoleic acid. Figure 1 shows the free and esterified cholesterol content of the cells expressed as pg/mg cell protein. During the incubation period the protein and phospholipid content of the cells remained constant at 610 k 31 pg and 139 f 8 E.cg/mgcell protein, respectively (n = 22). After the cells had been stimulated with cationised LDL at 50 pg cholesterol/ml, and then incubated in normal incubation medium, the free cholesterol content of the cells remained relatively constant and was not significantly affected by bezafibrate or clofibrate. On the other hand, the esterified cholesterol content of the cells decreased with time. Analysis of variance of each of the 3 lines shown indicates that the fitted linear regression lines were statistically significant (P < 0.01). The slope of the line corresponding to the bezafibrate group is greater (P < 0.05) than that of the control group. Although the slope of the line corresponding to the clofibrate group does not reach statistical significance against that of the control group, it is seen to be between the control group and the bezafibrate group. When a logarithmic transformation was performed on the data, both lines were signifi-
59
0
1
3
5
Fig. 1. Effect of clofibrate and bezafibrate on removal of free and esterified blasts stimulated with cationised low density lipoprotein. -0 control, ml. *- - - - - -* bezafibrate 100 ~g/ml.
cholesterol from 3T3 fibro* -. - .mclofibrate 250 ~g/
??
Fig. 2. Effect of clofibrate and bezafibrate on removal of [3H]oleic acid and [ 14C]linoleic acid labelled cholesterol ester from 3T3 fibroblasts stimulated with cationised low density lipoprotein, e-----o control, ??. - * - a clofibrate 250 /.&/ml. *- - - - - -* bezafibrate 100 fig/ml.
cantly different from the control (bezafibrate P < 0.01, clofibrate P < 0.05). Figure 2 shows the removal of 3H-labelled oleic acid and 14C-labelled linoleic acid from the cellular cholesterol ester fraction. The data is plotted semi-logarithmically against the period of incubation in normal medium. The 3H-labelled oleic acid was rapidly removed from the cholesterol ester fraction of these cells. Each of these regression lines calculated from the data fit a straight line whose slope is different from zero (P< 0.01). The slopes corresponding to the bezafibrate and clofibrate groups are different from that of the control group (P< 0.01and P < 0.05, respectively). The 14C-labelled linoleic acid was more slowly removed from the cellular cholesterol ester fraction. The data for the control group, however, does not fit a line whose slope is different from zero. The data for the bezafibrate and clofibrate groups do fit straight lines whose slope is different from zero (P< 0.01 and P < 0.05, respectively). For all 3 groups there was a preferential removal of oleate relative to linoleate from the cholesterol ester fraction of these cells, however bezafibrate and clofibrate had no effect on this preferential removal pattern. Discussion The concentrations of bezafibrate and clofibrate used in these experiments were chosen to be in the range of the therapeutically achieved levels of these agents [17,18]. 3T3 mouse fibroblasts were used as a model system to determine the effect of these agents on cholesterol metabolism at the cellular level. In these experiments, the confluent 3T3 fibroblasts were in a steady state with
60
respect to protein and phospholipid content. However, when cationised LDL was added to or removed from the incubation medium, the cells were not in a steady state with respect to total cholesterol content. Basu et al. have described the cellular metabolism of cationised LDL using human fibroblasts [10,19]. Nevertheless, cationised LDL was similarly metabolised by aortic smooth muscle cells [ 20,211 and endothelial cells in culture [ 221. In the non-stimulated 3T3 fibroblasts the extent of cholesterol esterification was quite small -- less than 6% of the labelled fatty acids taken up by the cells were incorporated into cholesterol ester. Similar levels of incorporation of 14Clabelled oleic acid into the cholesterol ester fraction have been observed in aortic intimal and medial cells of normal and cholesterol-fed rabbits [ 231. Bezafibrate at 50 ,ug/ml and 250 I.cg/ml, and clofibrate at 50 pg/ml decreased the incorporation of oleic acid relative to linoleic acid into the cellular cholesterol ester fraction. However, in these non-stimulated 3T3 cells the content of esterified cholesterol was too small to be determined accurately, so cationised LDL was used to produce cells containing large amounts of esterified cholesterol. These cells also demonstrated a marked stimulation of cholesterol esterification compared with cells incubated in normal medium. The incorporation of labelled oleic acid into the cholesterol ester fraction was similar to that observed in foam cells from atherosclerotic lesions [ 241. In the 3T3 fibroblasts stimulated with cationised LDL, bezafibrate and clofibrate reduced the accumulation of esterified cholesterol. These agents decreased the uptake of oleic and linoleic acids and their incorporation into cholesterol ester, They also reduced the incorporation of oleic acid relative to linoleic acid into the cellular cholesterol ester fraction. When the stimulated cells were incubated in normal medium, esterified cholesterol was removed from the cells and the removal rate was increased by both bezafibrate and clofibrate. While these agents increased the rate of removal of labelled oleic and linoleic acids from the cholesterol ester fraction, they did not affect the removal of labelled oleic acid relative to linoleic acid from cholesterol ester. The mechanism of these effects remains to be determined. Cationised LDL binds to non-specific negative charges on the cell membrane, undergoes adsorptive endocytosis and is transported to the cellular lysosomes where cholesterol ester is hydrolysed. The free cholesterol is then liberated into the cytoplasm where it may be transferred to cellular membranes or esterified While both the cholesterol ester and accumulates in lipid droplets [ 10,19-221. taken up and the free cholesterol esterified in the cell contribute to the cellular cholesterol ester, hydrolysis of cholesterol ester and any associated transport of free cholesterol out of the cell reduce the cellular content of cholesterol ester. The possible mechanisms of action of these agents, therefore, include reduction in the uptake of cationised LDL, suppression of esterification, stimulation of hydrolysis and of removal of the liberated free cholesterol. Although there is no information on the effects of these agents on the binding or internalisation of cationised LDL, clofibrate has no effect on the surface binding of normal LDL by human aortic smooth muscle cells in culture [25]. However, the reduction in the ‘H/14C ratio of the cholesterol ester fraction by bezafibrate and clofibrate suggests that the metabolism of endogenously esterified cholesterol is altered by these agents. The decreased incorporation of oleic
61
and linoleic acids in cholesterol ester may be due to decreased esterification or increased hydrolysis of endogenously synthesised cholesterol ester. Two different cholesterol esterifying enzyme systems have been described in foam cells from atherosclerotic rabbit aorta [26]. One has a pH optimum of 5 and does not require cofactors. This corresponds to lysosomal sterol ester hydrolase (E.C. 3.1.1.13). In rat liver lysosomes, the activity of this enzyme has been shown to be reversible, catalysing both the formation and hydrolysis of cholesterol ester [2’7]. The other enzyme system has a pH optimum of 7.5 and absolute requirement for ATP and coenzyme A. This involves the microsomal fatty acyl coenzyme A synthetase (E.C. 6.2.1.3) and fatty acyl coenzymeA: cholesterol acyl transferase (E.C. 2.3.1.26). The activity of a cytoplasmic cholesterol hydrolase enzyme has been demonstrated in human fibroblasts [28] and monkey aortic smooth muscle cells [ 291. The microsomal esterifying system and lysosomal cholesterol esterase have been demonstrated in human fibroblasts [30,31] aortic smooth muscle cells [29,32,33] and macrophages [34,35]. The microsomal esterifying system demonstrates a preference for oleic acid [ 361 while the microsomal cholesterol esterase hydrolyses cholesterol oleate and cholesterol linoleate at similar rates [ 371, as does the lysosomal cholesterol e&erase [ 33,381. Since oleic acid is preferentially esterified to cholesterol, the greater decrease in the incorporation of oleic acid relative to linoleic acid into the cholesterol ester fraction by bezafibrate and clofibrate in the stimulated cells, would be consistent with a suppression of microsomal esterification of cholesterol by these agents. In the atherosclerotic rabbit aorta, Brecher and Chobanian have demonstrated a marked inhibition of the microsomal esterifying system by clofibrate [6]. The mechanism for the reduction in cellular esterified cholesterol by bezafibrate and clofibrate may, therefore, be due to an inhibition in the activity of the microsomal esterifying system. The enzymes that could be inhibited are fatty acyl coenzyme A synthetase or fatty acyl coenzyme A: cholesterol acyl transferase. If there is no change in the uptake of cationised LDL by these agents, the free cholesterol liberated by lysosomal hydrolysis must be removed from the cells rather than esterified by the microsomes. Free cholesterol has been shown to be removed’ from cells in culture, however esterified cholesterol is not directly removed from cells but requires to be hydrolysed prior to removal [38,39]. The effect of bezafibrate and clofibrate on cholesterol ester hydrolysis is not known, however, stimulation of lysosomal or microsomal hydrolysis and removal of the liberated free cholesterol by these agents cannot be excluded as a mechanism contributing to the reduction in cellular esterified cholesterol. When 3T3 fibroblasts which had been stimulated with cationised LDL were incubated in normal medium, cholesterol ester was removed from the cells and there was a decrease with time in the relative amount of 3H-labelled oleic to 14C-labelled linoleic acid in the cholesterol ester fraction. This is consistent with a decrease in the microsomal esterifying activity during this period of removal. If this esterifying activity was further depressed by bezafibrate or clofibrate, a further decrease in the amount of ‘H-labelled oleic relative to “C-labelled linoleic acid would be expected. The fact that this did not occur may, there-
62
fore, indicate a stimulation of hydrolysis of endogenously esterified cholesterol by these agents. Further investigation is, however, required to determine the effects of these agents on cytoplasmic and lysosomal hydrolysis of esterified cholesterol. The removal of cellular free cholesterol is increased by serum high density lipoprotein (HDL) [ 401, so that a possible mechanism for the increased removal by bezafibrate and clofibrate may be that serum HDL is rendered a more effective acceptor for cholesterol. This possibility is currently being explored. Acknowledgements The authors wish to thank Miss Trudi Harris for excellent technical assistance and Dr. P. Singh for assistance with the Iatroscan determinations. References 1 2 3
4
6 6 7
8 9 10
11 12 13 14 16 16 17 18 19 20
Oliver, M.F., Heady, J.A. and Morris, J.N., A cooperative trial in the primary prevention of ischaemic heart disease using clofibrate, Brit. Heart J., 40 (1978) 1069. Olsson, A.G. and Lang, P.D.. Dose--response study of bezafibrate on serum lipoprotein concentrations in hyperhpoproteinaemia, Atherosclerosis, 31 (1978) 421. Kritchevsky, D.. Sailata, P. and Tepper. S.A.. Infiuence of ethyl pshlorophenoxyisobutyrate (CPIB) upon establishment and Progression of experimental atherosclerosis in rabbits, J. Atheroscler. Res., 8 (1968) 766. Sate. M. and Takano, T.. Fatty acid composition of cholesteryl esters in cholesterol-fed rabbits treated with lipid-lowering agent ethyl 2-@&lorophenoxy) isobutyrate, Jap. J. Pharmacol., 29 (1979) 261. Buxtorf, J.C.. Beaumont, V.. Jacotot. B. and Beaumont, J.-L., Regression de xsnthomes et medicamerits hypoIipidemiants, Atherosclerods, 17 (1974) 1. Brecher. P.I. and Chobanian, A.V., Cholesterol ester synthesis in normal and atherosclerotic aortas of rabbits and rhesus monkeys, C&c. Res., 35 (1974) 692. Lee, K.T., Jarmolych, J., Kim, D.N.. Grant, C., Krasnev, J.A., Thomas, W.A. and BNnO. A.M., Production of advanced coronary atherosclerosis, myocardial infarction and “sudden death” in swine, Exp. Mol. Path., 16 (1971) 170. Watkis. A.. Zak. B. and Boyle, A.J.. A new method for the direct determination of serum cholesterol, J. Lab. CIin. Med., 41 (1953) 486. Hatch, F.T. and Lees, R.S., Practical methods for plasma Iipoprotein analysis, Advances in Lipid Research, 6 (1968) 1. Bssu, S.K.* Goldstein, J.L., Anderson, R.G.W., and Brown, MS., Degradation of c&ionized low density lipoprotein and reguI&ion of cholesterol metabolism in homozygous familial hypercholesterolemia fibroblasts, Proc. Nat. Acad. Scl. (USA), 73 (1976) 3178. Goldberg. B.. Collagen synthesis as a marker for cell type in mouse ST3 lines, Cell, 11 (1977) 169. Hanks, J.H.. The longevity of chick tissue cultures without renewal of medium, J. CeU. Camp. Physiol., 31 (1948) 285. Lowry, O.H., Rosebrough. N.J., Farr. A.L. and Randall. R.J.. Protein measurement with the FoIin phenol reagent, J. Biol. Chem.. 193 (1961) 266. Folch, J., Lees, M. and Sloane Stanley, G.H., A simple method for the isolation and purification of totlil lipides from animal tissue, J. Biol. Chem., 226 (1961) 497. Itaya, K. and Ui. M.. A new micromethod for the calorimetric determination of inorganic phosphate, CIin. Chim. Acta, 14 (1966) 361. Snyder, F.. Radioassay of thin layer chromatogram - A high resolution zonal scraper for auantitative C14 and H3 scanning of thin layer chromatograms, Anal. Biochem., 9 (1964) 183. Thorp, J.M., Experimental evaluation of an orally active combination of androsterone with ethyl chlorophenoxyisobutyrate. Lance& i (1962) 1323. Endele, R.. A gas chromatographic method for the determination of bezafibrate in serum and urine, J. Chromatog., lb4 (1978) 261. Basu. S.K.. Anderson, R.G.W., Goldstein, J.L. and Brown, M.S., MetaboIism of cationized Bpoproteins by human fibroblasts, J. Cell Biol.. 74 (1977) 119. Stein, 0.. Goren, R. and Stein, Y.. Removal of cholesterol from fibroblasts and smooth muscle c&s in culture in the presence and absence of cholesterol esterification in the medium, Biochim. Biophys. Acta, 629 (1978) 309.
63 21
22
23 24 25 26 27 28 29
30 31 32 33 34 35 36 37 39 39 40
Goldstein, J.L.. smooth muscle
Anderson, R.G.W.. Buja. L.M.. Basu. SK. and Brown, MS.. Overloading cells with low density lipoprotein-cholesteryl esters reproduces features
human aortic of atheroscle-
rosis in vitro, J. Clln. Invest., 59 (1977) 1196. D. and Fielding, C.J., Effect of contact inhibition on Fielding, P.E.. Vlodavsky. I., Gospodarowicz, the regulation of cholesterol metabolism in cultured vascular endothelial cells, J. Biol. Chem., 254 (1979) 749. Day, A.J., Lipid metabolism by rabbit aortic intimsl and medial cells in tissue culture, Vircbows Arch. A. Path. Anat. Histol., 362 (1974) 89. Day, A.J. and Tume, R.K.. Incorporation of [l-14Cloleic acid into lipid by foam cells and by other fractions separated from rabbit atherosclerotic lesions, Atherosclerosis, 11 (1970) 291. Filipovic, I. and Buddecke, E., p-Chlorophenoxyisobutyrate enhanced retention of homologous lipoproteins by human aortic smooth muscle cells, Lipids, 12 (1977) 1069. Proudlock, J.W., Day. A.J. and Tume. R.K., Cholesterol esterifying enzymes of foam cells isolated from atherosclerotic rabbit intima. Atherosclerosis, 18 (1973) 451. Nilsson, A., Nordon. H. and Wilbehnsson, L., Hydrolysis and formation of cholesterol esters with rat liver lysosomes, Biochim. Biophys. Acta, 296 (1973) 593. Goldstein, J.L.. Dana, SE., Faust, J.R., Beaudet, A.L. and Brown, MS.. Role of lysosomal acid lipase in the metabolism of plasma low density lipoprotein. J. Biol. Chem.. 250 (1975) 8487. Riddle, MC., Fujimoto, W. and Ross, R., Two cholesterol ester hydrolases -Distribution in rat tissues and in cultured human fibroblasts and monkey arterial smooth muscle cells, Biochim. Biophys. Acta, 488 (1977) 359. Brown, MS., Dana, SE. and Goldstein, J.L., Cholesterol ester formation in cultured human fibroblasts, J. Biol. Chem., 250 (1975) 4025. Brown, M.S.. Dana, S.E. and Goldstein, J.L., Receptor-dependent hydrolysis of cholesteryl esters contained in plasma low density lipoprotein, Proc. Nat. Acad. Sci. (USA), 72 (1975) 2925. St. Clair. R.W.. Metabolism of the arterial wall and atherosclerosis. In: R. Paoletti and A.M. Gotto, Jr. (Eds.). Atherosclerosis Reviews, Vol. 1, Raven Press, New York, 1976, p. 61. Takano, T., Black, W.J., Peters, T.J. and de Duve. C., Assays, kinetics, and lysosomal localization of an acid cholesterase in rabbit aortic smooth muscle cells, J. Biol. Chem.. 249 (1974) 6732. Day, A.J. and Tume. R.K.. Cholesterol*sterifying activity of cell-free preparations of rabbit peritoneal macrophages. Biochim. Biophys. Acta. 176 (1969) 367. Werb, Z. and Cohn, Z.A.. Cholesterol metabolism in the macrophage. Part 3 (Ingestion and intracellular fate of cholesterol and cholesterol esters), J. EXP. Med., 136 (1972) 21. Bowyer. D.E., Howard, A.N. and Gresham, G.A., Lipid synthesis ln perfused normal and atherosclerotic rabbit aortas, Biochem. J., 103 (1967) 54. Deykin, D. and Goodman, D.S.. The hydrolysis of long-chain fatty acid esters of cholesterol with rat liver enzymes. J. Biol. Chem.. 237 (1962) 3649. Stein, Y., Halperin, G. and Stein, 0.. Intralysosomal hydrolysis of cholesterol esters of varying fatty acid CornPosition in cultured human skin fibroblasts, Biochim. Biophys. Acta, 530 (1978) 420. Chen, R.M. and Fischer-Dzoga, K.. Effect of byperlipemic serum lipoproteins on the lipid accumulation and cholesterol flux of rabbit aortic medial cells. Atherosclerosis, 28 (1977) 339. Stein, 0. and Stein, Y.. The removal of cholesterol from Landschiitz ascites cells by high-density apolipoprotein, Biocbim. Biophys. Acta, 326 (1973) 232.