Increased Cholesterol-Ester Formation during Forced Cholesterol Synthesis in Rat Hepatocytes

Transcription

1 Eur. J. Biochem. 51, (1975) Increased Cholesterol-Ester Formation during Forced Cholesterol Synthesis in Rat Hepatocytes Wke NILSSON Department of Physiological Chemistry, University of Lund (Received July 24/0ctober 18, 1974) By comparing the incorporation of 3H20 and [14C]mevalonate into cholesterol in suspensions of rat hepatocytes, it was calculated that the cholesterol biosynthesis could be stimulated 4-7-fold by addition of mevalonate. The addition of mm mevalonate also caused a 5-6-fold increase in the proportion of newly synthesized cholesterol that was esterified. The esterification of radioactive cholesterol, entering the cells by exchange with surrounding plasma lipoproteins was also increased, indicating that a true increase in the rate of cholesterol ester formation rather than a more selective utilization of newly synthesized cholesterol for esterification, occurred. The increase in cholesterol esterification was not abolished by cycloheximide, indicating that it did not require an increased synthesis of cholesterol esterifying enzyme. Instead the data suggest that the supply of cholesterol to the esterifiable pool may be an important factor, regulating the rate of cholesterol ester formation in rat liver. The addition of 0.5 mm oleic acid to the medium did not increase the degree of cholesterol esterification significantly, whereas 2 mm oleic acid bound to 1 % albumin increased the proportion of newly synthesized cholesterol that was esterified, by about 70 %. The cells secreted radioactive cholesterol esters into the medium. Cycloheximide inhibited this secretion to about 80 but did not affect the rate at which newly synthesized cholesterol was transferred to surrounding plasma lipoproteins. The factors controlling the rate of cholesterol ester formation and secretion in rat liver are poorly understood. The finding of Roheim and Eder [l] that perfused rat livers from cholesterol-fed animals secrete more cholesterol ester than normal as lipoproteins, suggest that the liver may react on an increased supply of cholesterol by increasing the formation of lipoprotein cholesterol esters. It is not known, however, whether the increased cholesterol ester formation requires induction of cholesterol esterifying enzymes or whether it is a direct effect of the increased substrate availability or other metabolic changes. In the present study the esterification of newly synthesized cholesterol was studied in suspended rat hepatocytes. By adding mevalonate the cholesterol biosynthesis could be increased significantly and the degree of esterification could therefore be studied in a situation, where the rate of cholesterol synthesis exceeded the physiological needs of the cells. In other experiments the effect of albumin-bound oleic acid on the esterification of newly synthesized cholesterol was studied, to examine the extent to which the rate of cholesterol esterification is directly linked to the rate of triglyceride formation and secretion in the rat liver. MATERIALS AND METHODS [4-'4C]Cholesterol, DL-[2-3H]mevalono1actone, DL- [2-'4C]mevalonic acid (dibenzylethylenediamine salt) and [l-14c]acetate were obtained from Radiochemical Centre (Amersham, England). Unlabelled mevalonolactone was obtained from Sigma. The mevalonolactones were hydrolyzed with a slight excess of 0.1-M NaOH at 37 "C for 1 h before use. The dibenzylethylenediamine salt of the DL-[~-'~C]- mevalonic acid was dissolved in alkaline buffer and washed with diethylether before use. In the experiments with added radioactive cholesterol, the [14C]cholesterol was bound to serum lipoproteins by adsorbing it to a filter paper, which was

2 338 Cholesterol Esterification in Rat Hepatocytes then incubated with serum [2]. The serum had been heated at 56 "C for 30 min to inactivate the lecithin cholesterol-acyl transferase activity. Fatty acids were bound to defatted [3] and dialyzed bovine serum albumin and the concentration was determined as described earlier (for reference see [4]). Male white Sprague-Dawley rats weighing g were obtained from Anticimex AB (Stockholm, Sweden). They were kept on a controlled feeding schedule with access to a standard pellet diet between 8 and 11 a.m. and the light was on between 6 a.m. and 6 p.m. The experiments were started between noon and 1 p.m. Rat liver parenchymal cells were prepared essentially according to Berry and Friend [5] as described earlier [4] except that the perfusion medium was similar to the one of Ingebretsen and Wagle [6]. The medium contained 0.02 % collagenase (Worthington CLS, U per mg) and 1.5% defatted [3] and dialyzed bovine serum albumin in Ca2+-free Hanks' solution buffered with 10 mm phosphate. 1.5 mm CaC1, was added after 15 min and the perfusion was then continued for another 15 min. Each cell preparation was checked by phasecontrast microscopy and by vital staining with trypan blue % of the cells excluded the stain. The cells were incubated in a total volume of 1.5 ml Hanks' solution [7], buffered with 25 mm NaHCO, and 19.4 mm N-2-hydroxy-ethyl-piperazine-N'-2- ethane-sulfonic acid. The NaCl concentration was 5.6 g per liter. The medium contained amino acids in concentrations similar to those used by East et al. [8], which have been found to give optimal lipoprotein secretion in the hepatocyte suspensions (R. Sundler, unpublished observation). The incubations were performed under an atmosphere of 5 % CO, and 95 % 0,. They were interrupted by cooling on ice, and cells and media were separated by low speed centrifugation as described earlier [9]. Lipids were extracted from the cell pellets and the supernatants with chloroform/methanol (1 : l), equilibrated with water, and washed twice with methanol/ water/chloroform (48 : 47 : 3). Radioactivity of cholesterol and cholesterol esters was determined after separation by thin-layer chromatography on silica gel H in light petroleum/diethyl ether/acetic acid (80 : 20 : 1). The cholesterol spot was scraped into the counting vials. The cholesterol ester spot was eluted with chloroform/methanol (2 : l), evaporated with nitrogen, and hydrolyzed [ 101. After extraction with light petroleum the cholesterol was isolated by thinlayer chromatography as above and scraped directly into the counting vials. In the experiments where mevalonate was isolated, the incubations were interrupted with 1 N KOH and 100 ymol mevalonate carrier, and the mevalonate was lactonized with an ; [Mevalonate] (mm) Fig. 1. Incorporation oj 3H20 and [2-'4C]mevalonate into cholesterol at different concentrations of mevalonate mg cell protein was incubated with 1.5 mci 3H20, 0.3 pci [2-'4C]mevalonate and increasing concentrations of unlabelled mevalonate for 60min. The figure shows incorporation of 3H20 (ed) and ['4C]mevalonate (A-A) into cholesterol, expressed as ng-atoms 3H and ng-atoms [2-I4C]- mevalonate carbon excess of sulfuric acid [tl]. After addition of an excess of ammonium sulfate the mevalonolactone was extracted with ether [12]. The extract was taken to dryness, dissolved in water, washed twice with light petroleum, treated with Dowex 50 and chromatographed on a Dowex 1 column [11].The mevalonactone was further purified by thin-layer chromatography on silica gel H in acetone/benzene (1 : 1) [12], and was eluted from the silica gel with acetone. Radioactivity was measured by liquid scintillation counting using a Packard Tricarb liquid scintillator and a dioxanebased scintillation fluid. Protein was determined according to Lowry et a/. [13]. RESULTS Effect of Mevalonate on the Rate of Cholesterol Biosynthesis When hepatocytes were incubated with [1-14C]- acetate the incorporation of I4C into cholesterol was negligible at mevalonate concentrations above 25 mm, indicating that cholesterol was formed almost entirely from added mevalonate. The addition of mevalonate, however, did not significantly decrease the incorporation of,h20 into cholesterol. Instead a constant relation between the incorporation of 3H20 and mevalonate carbon was established, and it could be calculated that ng-atom was incorporated into cholesterol per carbon atom during the conversion of mevalonate to cholesterol (Fig. 1, Table 1). In an earlier study Brunengraber et al. [14] found that with glucose as the major carbon precursor the

3 A. Nilsson 339 Table 1. Relation between incorporation of exogenous mevalonate and 3Hz0 into cholesterol of rat hepatocytes mg cell protein was incubated for 60 min with increasing concentrations of [2-'4C]mevalonate and 1.5 mci 3H,0. The values from each experiment are mean values of at least 4 incubations at mevalonate concentrations between 25.0 and 51.5 mm. The incorporation of 5 ng-atoms 14C at C-2 of mevalonate corresponds to the biosynthesis of one nmol cholesterol, since one out of six C-2 atoms is eliminated by demethylation. Amounts of 3H or I4C are expressed as ng-atoms Expt 3H/14C in 3H 3H cholesterol nmol cholesterol ng-atom cholestersynthesized from ol carbon added mevalonate mean =0.670 Table 2. Eflect of mevalonate on the relative incorporation of 3H20 and [I4C]acetate into mevalonate, cholesterol and,fatty acids 22.0 and 11.7 mg cell protein was incubated for 2 h with 0.25 pci [14C]acetate (3.3 mm) and 1.5 mci 3H20 under the conditions given in the text, with and without 17.9mM mevalonate present. The values are means of duplicate incubations. Without exogenous mevalonate present no significant radioactivity was found in mevalonate Expt Mevalonate 3H/14C ratio in concentration cholesterol mevalonate fatty acids mm perfused rat liver incorporation 0.76 ng-atom 3H from 3H20 per carbon atom into cholesterol. Combined with the present data this means that significantly less 3H20 is incorporated per newly synthesized cholesterol molecule with mevalonate instead of glucose as the carbon precursor. It is therefore likely that when cholesterol synthesis from glucose is estimated by measuring the incorporation of 3H20, most of the incorporation of 3H occurs at the stage before mevalonate. This means that in the present experiments addition of mevalonate may stimulate the cholesterol synthesis significantly although it did not cause much change in the incorporation of 3H20 into cholesterol (Fig. 1). For instance, if the incorporation of 3H20 is equal before and after addition of 25 mm mevalonate to the medium, but the number of atoms 3H incorporated per carbon atom is decreased from 0.76 to 0.12 the cholesterol biosynthesis must be increased 0.76/0.12 = 6.3 times. However, the relation between the incorporation of 3H20 and of acetyl carbon into cholesterol may well be different from the one observed by Brunengraber et al. [14], if the acetyl-coa that is utilized for cholesterol synthesis is derived not only from glucose, but also from amino acids and long-chain fatty acids. Experiments were therefore performed to estimate how much of the incorporation of 3H into cholesterol that occurred at the stage before mevalonate under our conditions. By adding cold mevalonate when hepatocytes were incubated with ['4C]acetate and 3H20, radioactive mevalonate could be trapped and its 3H/'4C-ratio compared to that of cholesterol formed in the absence of added mevalonate (Table 2). It was found that significant incorporation of 3H into mevalonate occurred. Indeed the 3H/'4C-ratio of the trapped mevalonate was even somewhat higher than that of cholesterol formed in the absence of exogenous mevalonate (Table 2), despite the fact that no I4C may be incorporated during the conversion of mevalonate to cholesterol. The most likely interpretation of these data is that the major portion of the incorporation of 3H into cholesterol occurs already during the formation of mevalonate. Furthermore the amounts of mevalonate that were necessary for efficient trapping may cause metabolic changes that affect the balance between the incorporation of 3H,0 and ['4C]acetate into mevalonate, since they were also found to increase the 3H/'4C-ratio of newly synthesized fatty acids (Table 2). Effect of Mevalonate on Cholesterol-Ester Formation When the cells were incubated with a tracer amount of radioactive mevalonate on average % of the radioactive cholesterol that had been formed during a 2-4 h incubation period was esterified (Table 3). A significant proportion of the radioactive cholesterol ester radioactivity had been secreted into the medium, and this secretion could be inhibited to about 80 % by cycloheximide (Table 4), which is in line with earlier findings that the suspended cells are able to form and secrete lipoproteins [9]. When increasing levels of mevalonate were added, the proportion of newly synthesized cholesterol that was esterified could be increased about five-fold (Table 3, Fig. 3). In two experiments the addition of mevalonate was also shown to cause an increased esterification of ['"CIcholesterol, that was taken up by the cells from added

4 340 Cholesterol Esterification in Rat Hepatocytes Table 3. Effect of mevalonate, oleic acid and cycloheximide on the esterification of newly synthesized cholesterol mg cell protein was incubated with pci [1,2-3H]- or [2-'4C]-mevalonate under the conditions described in Methods Addition Incuba- Radioactive Number Increase tion cholesterol of experitime as ester ments 0' h /a -fold k mevalonate k k mm mevalonate f f mm mevalonate k k mm oleic acid k & mm oleic acid f mm cycloheximide mm f 0.05 cycloheximide f Table 4. Effect of cycloheximide on the release of radioactive cholesterol and cholesterol-esters from hepatocytes mg cell protein was incubated with pci [3H]- or [14C]mevalonate under the conditions described in the text. After incubation the cells were sedimented by centrifugation at 150 x g for 5 min, and the cell pellet and supernatant analyzed separately. In the four experiments, where incubations were performed both with and without cycloheximide, the inhibition of cholesterol-ester secretion was 79.1 f 3.4% after 2 h and 85.5 f 1.6% after 4 h. The release of unesterified cholesterol was inhibited by 25.9 f 4.5 after 2 h and f 6.8% after 4 h. Results are radioactivity as a percentage of that in the medium, expressed as mean f S.E.M. Incuba- Number Cyclo- Radioactivity in tion of experi- heximide time ments cholesterol cholesterolesters h mm % f f f f f f 0.4 serum lipoproteins presumably by exchange with the non-esterified cholesterol of the hepatocytes (Fig. 2). The increased proportion of radioactivity in cholesterol ester at high concentrations of mevalonate [Mevalonate] (mm) Fig. 2. Effect of increasing concentrations of mevalonate on the esterification of added lipoprotein cholesterol mg cell protein was incubated for 4 h in the presence of 16.7% serum that had been labelled in vitro with [14C]cholesterol. The figure shows the percentage of the radioactivity of the cell pellet that was in cholesterol esters. Values are means of duplicate incubations. In another experiment the percentage radioactivity in cholesterol ester increased from 1.7 to 4.0% when the concentration of mevalonate was increased over the same range - 14,--" - t 13 m [Oleic acid] (rnm) Fig. 3. Effect of oleic acid on the esterification of newly synthesized cholesterol mg cell protein was incubated for 120min with 1 yci [1,2-3H]mevalonate added either as a tracer amount or as a 3.34 mm solution. Increasing concentrations of oleic acid were added at constant (1%) albumin concentration. The figure shows one of two similar experiments. The values are means of duplicate incubations. ( O d ) Percentage as cholesterol-ester without addition of mevalonate mass ; (A-A) percentage as cholesterolester with 3.34 mm mevalonate present was seen even when cycloheximide was present (Fig. 4) but much less of the radioactive cholesterolester was secreted into the medium (Table 4, Fig. 4). The effect of cycloheximide on the release of radioactive non-esterified cholesterol in the presence of serum was also tested, but the drug was found to have little or no effect on this process (Fig. 5).

5 A. Nilsson o 1-15 r'o well above the normal levels in rat plasma. It is therefore possible that the stimulation of cholesterol esterification at this fatty acid concentration is due to metabolic changes that do not occur under physiological conditions. I I I I I I 1' [Mevalcnate] (mm) Fig. 4. Effect of cycloheximide on the esterification of newly synthesized cholesterol at different mevalonate concentrations mg cell protein was incubated for 2 h with increasing concentrations of [2-'4C]mevalonate in the presence (A) and absence (0) of cycloheximide (0.035 mm). The figure shows the percentage of radioactive cholesterol that was esterified (----), and the percentage of the radioactive cholesterol that was secreted into the medium (-). One of two similar experiments is shown. The values are means of duplicate incubations w c c 3 - s? Incubation time (rnin) Fig. 5. Release of nonesterified cholesterol from hepatocytes in thepresence of serum. Cells containing 125 pg nonesterified cholesterol were incubated for different periods of time with 1 mm [2-'4C]mevalonate in the presence of 33 % serum (125 pg nonesterified cholesterol) under conditions described in the text. Values are means of duplicate incubations. (M) 14C in nonesterified cell cholesterol; (A-A) I4C in nonesterified cholesterol of the medium. (A) Without cycloheximide; (B) with mm cycloheximide Effect of Oleic Acid on Cholesterol Ester Formation The addition of oleic acid bound to albumin had less effect on the proportion of radioactive cholesterol that was esterified. At fatty acid concentrations below 1 mm little or no effect on the cholesterol esterification was noted (Fig. 3). With 2 mm oleic acid bound to 1 % albumin the esterification of cholesterol formed from radioactive mevalonate was increased on average by 72% (Table 3). This fatty acid concentration is DISCUSSION The present study is an attempt to examine how the rate of cholesterol ester formation in rat hepatocytes is influenced by the availability of the two substrates, cholesterol and oleic acid. A major difficulty in such a type of study is the question how the amount of cholesterol available for esterification should be varied. Although the cholesterol content of the cells could possibly be increased by adding for instance phospholipid-cholesterol emulsions with a high content of cholesterol, little is known about the intracellular distribution of cholesterol taken up by such a process and how fast it enters the esterifiable pool. It was considered more attractive to increase the supply of newly synthesized cholesterol, since the cholesterol formation as well as the enzyme likely to catalyze the formation of very low density lipoprotein cholesterol esters, namely the acyl-coa : cholesterol acyl transferase [15,16], are located to the endoplasmic reticulum, which has normally a very low cholesterol content. Significant stimulation of the cholesterol synthesis could be achieved by adding mevalonate to the cell suspensions, which is in line with the concept that the conversion of hydroxy-methylglutaryl-coenzyme A to mevalonate is a rate-limiting step in the conversion of acetyl-coa to cholesterol (for references see [17, 181). The degree of stimulation could not be calculated exactly, since the precise relation between the incorporation of 3Hz0 and of acetyl carbon into cholesterol in the absence of added mevalonate is not known. This relation may vary somewhat with the source of the acetyl carbon, since the finding that the incorporation of 3H per carbon atom derived from added mevalonate was 6.3-times lower than that reported earlier for the incorporation of 3H per atom of glucose carbon [14] suggests that the overwhelming portion of the 3H-incorporation into cholesterol normally occurs at the stages before mevalonate. This is likely to be true even under our incubation conditions, since the 3H/'4C-ratio of trapped mevalonate was similar to or even slightly higher than that of cholesterol formed in the absence of exogenous mevalonate (Table 2). Then an approximate calculation using the factor of Brunengraber et al. [14] indicates that the cholesterol synthesis could be stimulated 4-7 times by adding mevalonate.

6 342 A. Nilsson : Cholesterol Esterification in Rat Hepatocytes When the cholesterol synthesis was forced by the mevalonate addition an increased proportion of the newly synthesized cholesterol was esterified. This means either that newly synthesized cholesterol replaces other cholesterol molecules in the pool that participates in the esterification reactions or that a true increase in the rate of cholesterol ester formation occurs. The first alternative seems less likely, since a rapid distribution of newly synthesized cholesterol to different cell compartments is likely to occur. This is illustrated in Fig. 5 where it is seen that during a linear rate of incorporation of [14C]mevalonate into cholesterol the non-esterified cholesterol of surrounding plasma lipoproteins reached a certain specific activity only one hour later than the non-esterified cell cholesterol. This means that a rapid transfer of radioactive cholesterol to the plasma membranes, where it participates in exchange reactions with surrounding serum lipoproteins, must occur. The finding that radioactive cholesterol entering the cells presumably by the same exchange process was also esterified to an increasing extent after mevalonate addition provides further evidence that a true increase in the rate of cholesterol ester formation occurred (Fig. 3). Neither the cholesterol biosynthesis from mevalonate nor the esterification of newly synthesized cholesterol, or the increase in cholesterol esterification after mevalonate addition were inhibited by cycloheximide (Fig. 4 and 5). The increased cholesterol esterification is therefore not likely to depend on increased synthesis of cholesterol esterifying enzymes. The secretion of radioactive cholesterol ester into the medium, however, was strongly inhibited, which supports the earlier finding [9] that the suspended hepatocytes are secreting lipoproteins. In contrast the rate of transfer of non-esterified radioactive cholesterol to surrounding serum lipoproteins was unaffected by cycloheximide, indicating that neither the intracellular transfer to the plasma membrane nor the exchange process itself depend on an active protein synthesis (Fig. 5). The addition of high concentrations of albuminbound oleic acid to the medium increased the proportion of newly synthesized cholesterol that was esterified. The increase, however, was much less than the increase that could be induced by addition of mevalonate, and was noticeable only at an unphysiologically high oleic acid concentration (Fig. 2, Table 3). Still under similar conditions the addition of albuminbound fatty acids is known to cause a several-fold increase in the rate of triglyceride formation as estimat- ed by the incorporation of added fatty acids [I91 and glycerol [20]. It is therefore suggested that the rate of cholesterol ester formation in the rat liver is not regulated entirely by the rate at which fatty acids are taken up and esterified by the liver. Instead the data suggest, that the amount of cholesterol available for esterification is a major factor governing the rate of hepatic cholesterol ester formation in the rat. If this is so the rate of cholesterol ester formation and secretion would depend directly on short-term fluctuations in both the supply of cholesterol to the esterifiable pool and the rate at which cholesterol of this pool is utilized for other pathways as bile acid formation and net transfer to other membrane structures. Mrs Hildegun Lundberg and Miss Gertrud Olson provided skillful technical assistance. The work was supported by grants from the Swedish Medical Research Council (grant no. 03X-3969), the Medical Faculty, University of Lund, the Swedish Nutrition foundation and Albert Pahlssons stifielse. REFERENCES 1. Roheim, P. S., Haft, D. S., Gidez, L. J., White, A. & Eder, H. A. (1973) J. Clin. Invest. 42, Nilsson, A. & Zilversmit, D. B. (1972) J. Lipid Res. 13, Chen, R. F. (1967) J. Biol. Chem. 242, Nilsson, A,, Sundler, R. & Akesson, B. (1973) Eur. J. Biochem. 39, Berry, M. N. & Friend, D. S. (1969) J. Cell Biol. 43, Ingebretsen, W. R. & Wagle, S. R. (1972) Biochem. Biophys. Res. Commun. 47, Hanks, J. H. & Wallace, R. E. (1949) Proc. Soc. Exp. Biol. Med. 71, East, A. G., Louis, L. N. & Hoffenberg, R. (1973) Exp. Cell Res. 76, Sundler, R., Akesson, B. & Nilsson, A. (1973) Biochem. Biophys. Res. Commun. 55, Abell, L. L., Levy, B. B., Brodie, B. B. & Kendall, F. E. (1952) J. Biol. Chem. 195, White, L. W. & Rudney, H. (1970) Biochemistry, 9, Shapiro, D. J., Imblum, R. L. & Rodwell, V. M. (1969) Anal. Biochem. 31, Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, Brunengraber, H., Sabine, J. R., Boutry, M. & Lowenstein, J. M. (1972) Arch. Biochem. Biophys. 150, Goodman, D. S., Deykin, D. & Shiratori, T. (1964) J. Biol. Chem. 239, Stokke, K. T. & Norum, K. R. (1970) Biochim. Biophys. Acta, 210, Bortz, W. M. (1973) Metabolism, 22, McNamara, D. J. & Rodwell, V. W. (1973) Advan. Enzy mol. 38, Ontko, J. A. (1972) J. Biol. Chem. 247, Sundler, R., Akesson, B. & Nilsson, A. (1974) J. Bid Chem. 249, Nilsson, Medicinska-Kemiska Institutionen 4, Box 750, Solvegatan 39, S Lund, Sweden