- Email: [email protected]
Cholesterol metabolism in hypercholesterolemia-resistant rabbits
Atherosclerosis, 87 (1991) 169-181 e 1991 Elsevier Scientific Publishers
169 Ireland,
Ltd. 0021-9150/91/$03.50
A DONIS 002191509100095N
ATHERO
04615
Cholesterol metabolism in hypercholesterolemia-resistant
rabbits
David S. Loose-Mitchell ‘, Julie A. Poorman ‘, Sheryl A. Smith ‘, Merrill L. Overturf ‘, Joel D. Morrisett 2, Antonio M. Gotto, Jr. 2 and Maurizio R. Soma ’ ’Departments of Pharmacology and Medicine, University of Texas Health Sciences Cenrer at Houston, Houston TX 77030 (U.S.A.), and -’Departments of Medicine and Biochemistry, Baylor College of Medicine and The Methodist Hospital, Houston, TX, 77030 (U.S.A.) (Revised,
(Received 12 June, 1990) received 7 September and 3 December, (Accepted 4 December, 1990)
1990)
Normal rabbits typically respond to a diet high in cholesterol with a large increase in the concentration of plasma cholesterol. We have previously described the breeding and partial characterization of a variant rabbit which does not respond to a high cholesterol diet with changes in plasma cholesterol concentration. In the present report we have characterized three components involved in cholesterol homeostasis: the B/E (LDL) receptor, 3-hydroxy-3-methylglutaryl coenzyme A reductase activity (HMG-CoA reductase, EC 1.1.1.34) and acyl-coenzyme A : cholesterol acyltransferase activity (ACAT, EC 2.3.1.26) in the livers of the hypercholesterolemia-resistant rabbits. Using normal cholesterol-fed rabbit [‘251]/3-VLDL as a ligand, liver membranes prepared from resistant rabbits fed a low-cholesterol diet had 70% higher binding capacity than membranes from normal rabbits fed the same diet. Similar experiments demonstrated that the resistant rabbits had a 240% higher B/E receptor binding capacity compared to normal animals when liver membranes were prepared from animals fed a 0.25% cholesterol-enriched diet. No difference in the binding affinity of [‘251]/3-VLDL was detected in membranes prepared from normal or resistant animals. When fed a low-cholesterol diet, the resistant rabbits had approximately 2-fold higher hepatic HMG-CoA reductase activity (97.4 + 3.5 pmol product/mg/min in resistant animals compared to 45 + 1.1 pmol product/min/mg in normal animals). The difference was exaggerated in animals fed the 0.25% cholesterol-enriched diet, 73.3 + 5.5 vs 2.4 + 0.56 pmol product/min/mg for resistant and normal membranes respectively. The basal activity of ACAT in hepatic membranes was significantly lower in the resistant rabbits compared to normal rabbits (138 + 11 vs 268 + 19 pmol cholesteryl ester/min/mg in
Correspondence to: Dr. D.S. Loose-Mitchell, Department of Pharmacology, University of Texas Health Sciences Center, 6431 Fannin, Houston, TX 77030, U.S.A. Abbreviations: HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; ACAT, acyl-coenzyme A : cholesterol acyltrans-
ferase; LDL, low density lipoprotein; /?-VLDL, j3migrating very low density lipoprotein; B/E receptor, apoprotein B and apoprotein E receptor; NADP, B-nicotinamide adenine dinucleotide phosphate; DTT, dithiothreitol; BSA, bovine serum albumin; TLC, thin layer chromatography.
170 resistant and normal rabbits respectively); when fed a 0.25% cholesterol-enriched diet, the enzyme was induced 6-fold in normal animals but was increased only 2-fold in the resistant animal. These biochemical data suggested that the resistant rabbit maintained low intracellular cholesterol even when fed a cholesterol-enriched diet. Direct measurement of cellular cholesterol and cholesteryl esters demonstrated that the concentration of these lipids was significantly lower in the resistant animal than in normal animals with the largest differences found in the cytoplasmic rather than the membrane compartment. These studies demonstrate that the resistant rabbit manifests several quantitative differences in cholesterol metabolism and in the regulation of cholesterol metabolism; but these studies do not directly explain the underlying cause of the resistance to hypercholesterolemia in the resistant rabbit.
Key words: Hypercholesterolemia; Atherosclerosis; Cholesterol metabolism; reductase; Acyl-CoA : cholesterol acyltransfersase; (Rabbits)
Introduction We have developed and partially characterized a colony of New Zealand white rabbits which are strikingly resistant to the hypercholesterolemia which typically results from a high-cholesterol diet [l]. A 0.1% cholesterol enriched diet fed to normal rabbits results in plasma cholesterol levels of approximately 330 mg% after 8 weeks. Under the same dietary conditions the resistant rabbits maintain plasma cholesterol of less than 50 mg% *. The phenotype of the resistant rabbit is inheritable; however, we have not yet identified the precise mode of this inheritance. The phenomenon of resistance to hypercholesterolemia following dietary cholesterol challenge has been described in rabbits [2-41, rats [5], pigeons [6], and nonhuman primates [7,8]. Well controlled studies in humans have also documented variable resistance to hypercholesterolemia in response to dietary cholesterol [9-141. In all species which have been investigated, wide variations have been observed in plasma cholesterol levels following cholesterol challenge. This variation extends from individuals with little or no increase in plasma cholesterol to individuals with severe hy-
* The resistant phenotype is defined as an elevation in plasma cholesterol concentration greater than or equal to 1.5 SD below the mean plasma cholesterol concentration of normal rabbits at both 4 and 8 weeks on a 0.1% cholesterol-enriched diet.
LDL receptor;
HMG-CoA
percholesterolemia. In the human population, differences in response to cholesterol challenge have been noted in the same individual [14] tested at multiple times, but significant differences in response between different individuals has also been documented [13]. While genetic factors have been implicated in determining part of this variable cholesterol response, the mechanism(s) by which certain individuals maintain a low plasma cholesterol concentration following cholesterol ingestion is not clear in any species. It has been suggested that compensatory reductions in cholesterol synthesis [15], increased biliary cholesterol excretion [16], or differences in the ability of gut bacteria to conjugate bile acids [13] may contribute to the variability in responsiveness. The resistant rabbit colony we have developed may provide a model for the “resistant”, or hyporesponsive individuals within the human population. In this study we have investigated the biochemical response of several enzymes that play critical roles in cholesterol metabolism. These studies have focussed on the in vivo characteristics of hepatic proteins since the liver is the major site of cholesterol metabolism in normal mammals [17]. Activities of the enzymes 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) and acyl-coenzyme A : cholesterol acyltransferase (ACAT), and the lipoprotein binding capacity of the B/E receptor have all been found to differ significantly in hepatic tissue from resistant rabbits maintained on a low-cholesterol
171 diet compared to levels of these activities in normal rabbits. In addition, the expected biochemical responses in the activities of these proteins to a cholesterol-enriched diet were blunted in resistant rabbits as compared to the normal rabbits. These results suggest that the resistant animal may maintain reduced intracellular cholesterol concentrations in both basal and cholesterol-fed states. Direct measurement of cytosolic cholesterol in the resistant rabbits has confirmed that even on a high-cholesterol diet these animals maintained very low cellular cholesterol and cholesteryl ester concentrations.
Materials and methods Materials
3-Hydroxy-3-methyl[3-‘4C]glutaryl coenzyme A (52 Ci/mol) and [l-‘4C]oleoyl coenzyme A (52 Ci/mol) were purchased from Amersham (Arlington Heights, IL). [la,2a(n)-3H]Cholesterol (50 Ci/mol), [5-3H]mevalonolactone (21 Ci/mmol), and Na”‘I (17 Ci/mg) were purchased from New England Nuclear (DuPont Co., Boston, MA). Silica (Si 250) plates were purchased from Baker (Phillipsburg, NJ). [ 3H]Cholesteryl oleate was prepared from [3H]cholesterol and oleic anhydride [18] and purified by TLC. All solvents for chromatography were analytical grade or better and were purchased from Baker. Unless specifically noted all other reagents were purchased from Sigma Chemical Co. (St. Louis, MO) and were the highest quality available.
cholesterol-enriched diet used in the current studies consisted of standard chow supplemented with 0.25 g cholesterol (USP) and 2.0 ml corn oil per 100 g chow. Preliminary experiments established that the resistant rabbits lacked a hypercholesterolemic response when fed this cholesterol-enriched diet while normal rabbits developed a pronounced and rapid hypercholesterolemia. The cholesterol and corn oil were dissolved in chloroform (ACS) and mixed with chow pellets which had been freed of fines by sieving [19]. Saturated pellets were left in a well ventilated fume hood until the odor of chloroform was not detectable. Rabbits were fed 100 g chow per day until they were four months old; thereafter the ration was increased to 150 g/day. Baseline cholesterol concentration in the normal animals was 29.5 mg/dl (fed low-cholesterol chow) and after 8 weeks on the 0.25% cholesterol-enriched diet, plasma cholesterol in the normal animals increased to an average of 1052 mg/dl (range, 768-1335). Resistant animals used in this portion of the study had average baseline cholesterol concentrations of 21 mg/dl which increased to 51 mg/dl (range, 33-68) 8 weeks after being fed the 0.25% cholesterol-enriched diet. Serum cholesterol in all normal animals fed the low-cholesterol diet was 35.9 + 4.3 mg/dl and 16.6 + 3.3 mg/dl in all resistant animals fed the same diet. Breeding and phenotypic characterization of the resistant rabbit has been described previously [I]. Enzyme
assays
Preparation Animals
All rabbits were caged individually and maintained with normal 12 hour light/dark cycles. All rabbits were males between 24 and 30 months of age and resistant rabbits were from either the 4th or 6th generation within the resistant pedigree. The standard laboratory rabbit chow (Ralston Purina, St. Louis, MO) contained approximately 60 pg cholesterol per g chow. The 0.1% cholesterol-enriched diet used for phenotypic analysis has been described previously [l] and all animals were phenotyped prior to study. The 0.25%
of hepatic microsomes
Microsomes for enzymatic studies were isolated by centrifugation essentially as described by Erickson et al. [18]. In brief, liver sections were isolated from anesthetized animals killed between 9 and 11 a.m. One gram of tissue was homogenized in 5 volumes (w/v) of 0.25 M sucrose, 1 mM Na,EDTA (pH 7.4). The homogenate was then centrifuged at 4°C for 15 min at 9000 x g. After removal of the floating fatty layer, the supernatant was removed and ultracentrifuged at 4’C for 45 min at 134 000 x g. The resulting pellet was resuspended in 2 ml of the same solution. This suspension was ultracentrifuged at 4°C a second time
172 and the final pellet resuspended in 1 ml 0.2 M K,HPO, (pH 7.6) and then stored at -80°C.
A CA T assays
Samples of liver microsomes (25-200 pg) were assayed [20] in 225 ~10.2 M K,HPO, containing 5 mg/ml BSA and, except in experiments examining the effects of substrate concentration, 67.68 PM oleoyl CoA (7.68 PM [‘4C]oleoyl CoA, plus 60 PM unlabelled oleoyl CoA). All experiments employed a minimum of two microsomal protein concentrations for activity analysis. Microsomes were preincubated at 37°C in the phosphate-BSA buffer for 5 min before addition of the substrate. Substrate was then added and the reaction mixture was incubated at 37°C for 5 min and the reaction was terminated by the addition of 5 ml chloroform/methanol (2 : 1). In some experiments exogenous cholesterol was added to the ACAT assay to assess the effects of cholesterol availability essentially as described by Erickson and Fielding [21]. Cholesterol was added as cholesterol/ phospholecithin unilammellar liposomes (1 : 6 molar ratio) prepared exactly as described [22]. Liposomes were incubated with microsomes 30 min at 37°C prior to assay. Approximately 10000 dpm of [3H]cholesteryl oleate were added to each sample in order to estimate recovery. Following the incubation period, 1.05 ml pH 3.0 water was added to the reaction mixture, samples were mixed thoroughly and left at 4°C overnight. The mixture was then centrifuged at 2000 rpm for 10 min and the upper phase was removed and discarded. The lower phase was washed twice with 1.2 ml fresh upper phase (chloroform/ methanol/ water, 3 : 48 : 47). The addition of less than 1 ml methanol resulted in solubilization of the remaining interface before the organic phase was dried in air. The residue was resuspended in hexane, spotted on a TLC plate and chromatographed a distance of 10 cm in hexane/ethyl acetate (9 : 1). Following autoradiography, the band corresponding to cholesteryl oleate was isolated and counted in a dual label scintillation counter. Recovery of cholesteryl oleate was 41 k 0.04% for all samples assayed (n = 44) and recovery from samples in an individual experiment varied by no more than 11%.
HMG-CoA
reductase assays
Samples of liver microsomes (25-250 pg) were assayed in a buffer containing 0.1 M K,HPO, (pH 7.5), 20 mM glucose 6-phosphate, 2.5 mM NADP, 0.7 units glucose 6-phosphate dehydrogenase, 5 mM DTT, and, except in experiments where substrate concentration was varied, 87 PM HMG-CoA (12.5 PM [‘4C]HMG-CoA plus 74.5 PM unlabelled HMG-CoA) in a total volume of 200 ~1. At least two protein concentrations were used to measure enzyme activity. Microsomes were preincubated in the reaction buffer at 37°C for 10 min before addition of the HMG-CoA and then incubated for 20 min at 37°C with moderate shaking. The reaction was stopped by the addition of 30 ~15 N HCl, and 10 000 dpm [ 3H]mevalonolactone standard were added to measure recovery. The reaction solution was then incubated at 37°C for 30 min to allow the formation of the lactone of mevalonate. After addition of 5 volumes of diethyl ether, the solution was mixed thoroughly and left at 4°C overnight. The upper phase was transferred to a second tube and the lower phase was extracted 5 times with 5 volumes of ether. The combined upper phases were dried and the residue was resuspended in acetone. The incubates were spotted on a TLC plate and chromatographed for 15 cm in acetone/benzene (1: 1). Following autoradiography, the band corresponding to mevalonolactone was isolated and counted in a dual label scintillation counter. Recovery of labelled mevalonolactone was 51 f 0.03% for all samples assayed (n = 37) and recovery from samples in an individual experiment varied by less than 13%. p- VLDL binding assays
Preparation of membrane fractions for lipoprotein binding studies was done as described by Kovanen et al. [23,24]. Liver tissue was homogenized in a solution containing 50 mM Tris-HCl (pH 8.0) 0.15 M NaCl and 1 mM Na,EDTA on ice. The homogenate was centrifuged at 9000 x g and the pellet discarded. The supernatant was ultracentrifuged at 134000 X g for 45 minutes and the pellet was washed in the same buffer and pelleted by ultracentrifugation. The final pellet was resuspended in a solution of 50 mM NaCl, 1
173 mM CaCl,, and 20 mM Tris-HCl frozen until use. Preparation
of [“‘I]
(pH
8.0) and
/3- VLDL
P-VLDL (d < 1.006 g/ml) was isolated [25] from a normal New Zealand white rabbit maintained on a 2% cholesterol diet. Purified /3-VLDL was iodinated according to Bilheimer [26] to an average specific activity of 500 cpm/ng protein. Binding assays
B/E receptor binding assays were performed as described by Kovanen et al. 123,241 with some modifications. Rabbit liver membranes were thawed, passed through a 25-gauge hypodermic needle and sonicated for 45 s with a Branson Sonifier at a power setting of 3. Binding assays were performed in 110 ~1 of a buffer containing 50 mM Tris-HCl, 25 mM NaCl, 1 mM CaCl,, and 20 mg/ml bovine serum albumin. /3-VLDL binding resistant to EDTA was assessed by incubating membranes as described above except a buffer which contained 2 mM Na,EDTA rather than 1 mM CaCl, was used. Membranes were incubated with 25-2000 ng of [‘251]j3-VLDL for 60 min at 4°C. Non-specific binding was assessed by incubating parallel samples with a 50-fold excess of unlabelled P-VLDL. Bound ligand was separated from free by layering 75 ~1 of the binding reaction on top of 100 ~1 100% fetal bovine serum in a 225 ~1 nitrocellulose tube. Samples were then centrifuged at 100000 X g for 30 min at 4°C in a Beckman Airfuge. The tubes were then decanted and the bottom was cut from the tube and the pellet was counted. Localization
of intracellular
cholesterol
Liver sections, 0.5 g wet weight, were prepared from animals maintained on a 0.25% cholesterolenriched diet. Tissue was homogenized in 3.0 ml 25 mM phosphate buffer (pH 7.5) with a Teflon grinder. Particulate matter and cytosol were separated by ultracentrifugation at 104000 X g for 1 h. The pellet was resuspended in a volume of homogenization buffer equal to the volume of the supernatant. One ml each of the supernatant and the pellet was mixed with 20 volumes of chloroform/methanol (2 : 1) and 4.2 ml pH 3.0 water
and then stored overnight at 4°C. The extraction was repeated and the organic phases were combined. The combined lower phase was washed twice with fresh upper phase (chloroform/ methanol/ water, 3 : 48 : 47). The interface was discharged by the addition of less than 1 ml methanol before the solution was dried under nitrogen. The residue was resuspended in hexane, spotted on a TLC plate, and chromatographed for 12 cm in acetic acid/acetone/petroleum ether (1 : 25 : 75). The lipids were visualized in iodine vapors, and the bands corresponding to cholesterol and cholesteryl esters were eluted from the silica taken to dryness in 2 : 1 cholorform/methanol, and redissolved in isopropanol. Total cellular cholesterol was measured using the cholesterol oxidase assay (Boehringer Mannheim Biochemicals, Indianapolis, IN) as described by the manufacturer. The isopropanol is required to re-dissolve cholesteryl esters prior to assay. Recovery of both lipids was 75-80% and was assessed using 14Clabelled internal standards. Results The B/E receptor binding capacity and affi,$ty of liver membranes were assessed using Ilabelled rabbit P-VLDL as ligand. Under the conditions used, binding was found to be linear from 50 to 300 pg membrane protein and binding equilibrium was obtained within 30 min at 4°C (data not shown). In typical binding experiments, 150 pg membrane protein were incubated with [‘251]P-VLDL for 60 min. A Scatchard analysis combined from the data in 5-7 individual binding experiments from liver membranes prepared from normal and resistant rabbits fed a normal chow (low cholesterol) diet is illustrated in Fig. 1. The binding capacity of membranes from resistant rabbits was 70% higher than that of normal rabbits with no apparent difference in receptor affinity (Kd = 6.78 pg/ml in normals and 5.21 pgg/ml in resistant liver membranes). An EDTA-resistant binding site which is relatively non-specific for lipoprotein particles has been described in the liver of rabbit [27] and pig [14]. To assess the contribution made by this binding site in our studies, binding experiments were conducted in 2 mM EDTA. Under these conditions, normal rab-
Fig. 1. Scatchard analysis of rabbit (‘*sI]/3-VLDL binding to hepatic membranes prepared from normal (0) and resistant animals (0) on a normal (low cholesterol) diet. Liver membranes, 150 ng protein/sample, were incubated with 40-2500 ng [‘*‘1]/3-VLDL in 110 pl for 60 min at 4°C. Non-specific binding was assessed in parallel samples by including a 50-fold excess of unlabelled b-VLDL. Non-specific binding was approximately 28%. The data presented are the mean of 5-7 separate binding experiments from 3 normal and 3 resistant rabbits. Serum cholesterol concentration in the normal animals was 35.9 f 4.3 and 16.6 f 4.3 in the resistant animals, respectively.
bit hepatic membranes bound 0.123 ng [‘251]pVLDL/pg protein and liver membranes from the resistant rabbits bound 0.14 ng [1251]/3-VLDL/pg protein (at 10 pg/ml [‘2sI]j%VLDL). Subtraction of the contribution of the EDTA-resistant binding site increases the difference between the two phenotypes. Ca2+-dependent binding in membranes from resistant rabbits was 95% higher than in normal rabbits compared to 70% higher when total binding is analyzed. The results of binding experiments to liver membranes from rabbits fed a 0.25% cholesterol diet for two months were analyzed in a similar manner and the results are presented in Fig. 2. As expected, the high-cholesterol diet resulted in a 50% decrease in the number of hepatic B/E receptors in the normal animals (0.400 5 0.020 vs 0.205 + 0.070 ng [i2?]j3-VLDL/pg membrane protein off and on diet, respectively, P < 0.02). However, the resistant rabbits fed the high cholesterol diet had a much smaller reduction in the number of B/E receptors, and the change was not statistically significant (0.68 * 0.150 vs 0.50 + 0.10, P > 0.05). The combined effects of (i) higher numbers of B/E receptors under basal (low-cholesterol diet)
and (ii) a smaller percent reduction in receptor numbers following cholesterol challenge, resulted in a 2.4-fold higher B/E binding capacity in the cholesterol-fed resistant rabbits compared to cholesterol-fed normal rabbits. Subtraction of the contribution of the EDTA-resistant binding, which is not regulated by plasma cholesterol concentration [27], from these data again increased the difference observed between the two phenotypes; in this case Ca2+-dependent binding was 4.4-fold higher in the resistant rabbit versus 2.4-fold higher for total binding. In order to assess the capacity of the resistant rabbit liver to synthesize cholesterol we examined the activity of HMG-CoA reductase, the ratelimiting enzyme in cholesterol biosynthesis [28]. Liver membranes were prepared from the livers of normal and resistant rabbits and then assayed for HMG-CoA reductase activity by measuring the formation of mevalonic acid from 3-hydroxy-3methyl[‘4C]glutaryl coenzyme A. Conditions for the assay were optimized using membranes from rabbits fed the low-cholesterol diet (i.e., with induced levels of HMG-CoA reductase activity). HMG-CoA reductase activity as a function of protein concentration is illustrated in Fig. 3. The assay was linear to approximately 100 pg mem-
F
-
+ NORMAL
-
+
J
RESISTANT
Fig. 2. Summary of the effect of dietary cholesterol feeding on [‘251]P-VLDL binding in liver membranes from normal and resistant rabbits. Normal and resistant animals (3 in each group) were fed standard rabbit chow (-) or rabbit chow supplemented with 0.25% cholesterol (+) (2 animals in each group) for 8 weeks prior to experimentation. Membranes were prepared and analyzed as in Fig. 1. Scatchard analysis of the binding data was used to determine the B/E receptor binding capacity (N,,,,,) in each experiment. Results of 5 separate experiments (+ SEM) on each group are presented.
175
0
25
50
75
MICROSOMAL
100
125
PROTEIN
(pg)
150
175
Fig. 3. HMG-CoA reductase activity in hepatic membranes from normal (0) and resistant (0) rabbits on a low cholesterol diet as a function of protein concentration. The indicated mass of membrane protein was incubated under standard conditions in a final volume of 200 pl. Membranes were incubated with [‘4C]HMG-CoA (87 PM) for 20 min. After acidification of the reaction mixture to lactonize the mevalonate, samples were fractionated by TLC and the products were scraped from the plates and then counted in a liquid scintillation counter. Results presented are the mean and standard error of 4-6 determinations from samples prepared from 3 normal and 3 resistant rabbits.
brane protein, hence in all subsequent studies we employed 50 and 75 pg protein. These data were obtained using an incubation time of 20 min which was found to be optimum for this preparation (data not shown). When the animals were fed the low-cholesterol diet, membranes from resistant rabbits had approximately twice the hepatic HMG-CoA reductase activity compared to the normal rabbits (97 pmol mevalonolactone/min/ mg protein vs 45 pm01 mevalonolactone/min/mg protein, P < 0.02). This difference in enzyme activity could not be attributed to changes in the K, of the enzyme for the substrate HMG-CoA. Liver membranes were incubated with various concentrations of [‘4C]HMG-CoA (6.25-150 PM) and the results of this substrate-dependence analysis are illustrated in Fig. 4. Double reciprocal plot analysis of the data presented in Fig. 4 indicates that the K, of HMG-CoA was 21 PM in normal rabbits and 34 PM in the resistant animals; a difference which was not statistically significant. A series of studies was then undertaken to assess the effects of dietary cholesterol on the activity of HMG-CoA reductase in the liver of normal and resistant rabbits and the results are summarized in Fig. 5. After 2 months on an 0.25% cholesterol
HMG COENZYME
A (PM)
Fig. 4. Substrate dependence of HMG-CoA reductase in liver membranes from normal (0) and resistant (0) rabbits on a low-cholesterol diet. Membranes were prepared from 3 resistant and 3 normal animals and incubated with various concentrations of HMG-CoA. Each point represents the mean activity of determinations at 50 and 75 yg membrane protein from 2 separate experiments.
diet, HMG-CoA reductase activity was nearly abolished (reduced 91%) in the normal rabbits. In the resistant rabbits, HMG-CoA reductase activity was reduced only 25% by dietary cholesterol. These results are similar to those obtained for the B/E receptor binding capacity (Fig. 2) in that dietary cholesterol was less effective in reducing HMGCoA reductase activity in the resistant rabbits than in normal rabbits. In addition to the B/E receptor binding and ^,
125,
+ NORMAL
+ RESISTANT
Fig. 5. Effect of dietary cholesterol on HMG-CoA reductase activity in normal and resistant rabbits. Animals were fed standard rabbit chow (- , 3 animals in each group) or standard chow supplemented with 0.25% cholesterol (+, 2 animals in each group) for 8 weeks. Membranes were prepared and HMG-CoA reductase activity was assayed at 50 and 75 pg membrane protein. Results are the mean and standard error of 6-8 determinations.
176 HMG-CoA reductase activity, both of which are inhibited by dietary cholesterol in normal animals, ACAT, an enzyme normally induced by exogenous cholesterol was also studied. The assay for ACAT activity was optimized using liver membranes of animals fed the 0.25% cholesterol diet. The assay employed measures the rate of incorporation of [‘4C]oleoyl-CoA into “C-labelled cholesteryl oleate using endogenous cholesterol as substrate; the reaction products were resolved by TLC. Conditions were optimized with respect to oleate/BSA ratio, and incubation time and the assay was found to be linear for 2-10 min (data not shown). Fig. 6 illustrates ACAT activity as a function of membrane protein. The assay was linear up to approximately 75 pg protein and in all subsequent assays 50 and 75 pg protein were used to quantitate enzyme activity. Membranes from normal animals had about a six-fold higher ACAT activity when fed the high-cholesterol diet compared to membranes from resistant animals fed the same diet. The effect of substrate concentration on ACAT activity in rabbit hepatic membranes is illustrated in Fig. 7. These data indicate that the difference in ACAT activity between the normal and resistant rabbit membranes is not attributable to changes in the K, of the enzyme. Linear analysis of these data indicate that the apparent K, of ACAT for oleoyl-CoA was 22 PM for both the
.
. 0
20
40 OLEOYL
60
60 COENZYME
100
1
A (@.d)
Fig. 7. Effect of varying oleoyl CoA concentration on the activity of ACAT in rabbit liver membranes. Membranes were prepared from normal (0) or resistant (0) rabbits which had been maintained for 8 weeks on a 0.25% cholesterol diet. Membrane protein, 50 and 75 pg/assay, was incubated with the indicated concentrations of oleoyl CoA, and ACAT activity was determined as described in Fig. 6.
normal and resistant rabbits. The effect of manipulating dietary cholesterol on ACAT activity in the resistant and normal rabbit is summarized in Fig. 8. While the normal animals respond to an increase in dietary cholesterol with approximately a 6-fold increase in the activity of the enzyme, ACAT activity in the resistant animals increases by less than 2-fold. Several investigators have reported that ACAT activity is regulated, in part, by the availability of substrate [29-311. To determine if cholesterol substrate pools were af-
1
120 100 60 60
0
25
50
75
MICROSOMAL
100
125
PROTEIN
(pg)
150
175
Fig. 6. ACAT activity in liver membranes prepared from normal (0) and resistant (0) rabbits after 8 weeks on a 0.25% cholesterol diet. Hepatic membranes were prepared and ACAT activity was assessed by determining the rate of incorporation of [‘4C]oleoyl CoA into [t4C]cholesteryl oleate. The indicated mass of membrane protein was incubated under standard conditions in a final volume of 225 pl. Endogenous cholesterol was used as the fatty acid acceptor.
I
_ NORMAL
+
-
+
-J
RESISTANT
Fig. 8. Effect of a low cholesterol (- , 3 animals in each group) or a 0.25% cholesterol supplemented diet ( + , 2 animals in each group) on ACAT activity in liver membranes from normal and resistant rabbits. ACAT activity was assessed at 50 and 75 pg membrane protein from animals that had been maintained on the indicated diet for 8 weeks. Results are the mean and standard error of 6-8 determinations.
fecting ACAT activities, exogenous cholesterol (220 pg) as cholesterol/phospholecithin unilammelar liposomes, was added to liver membranes prepared from normal and resistant rabbits fed the 0.25% cholesterol-enriched diet. The exogenous cholesterol was pre-incubated with the membranes 30 min prior to assay. Addition of 20 pg cholesterol to the rnicrosomes increased ACAT activity from 298 f. 20 to 330 + 27 in samples from resistant rabbits and from 1388 f 60 to 1671 k 111 in samples from normal animals. These slight increases suggest that cholesterol pool size does not account for the differences in ACAT activity between normal and resistant rabbits. The above studies suggest that the resistant animals were refractory to cholesterol-mediated changes in the activity of several biochemical pathways involved in cholesterol metabolism. One interpretation of these results is that the resistant animals maintained a low intracellular cholesterol level, even in the presence of high dietary cholesterol. We therefore sought to measure directly the intracellular cholesterol content in liver samples from normal and resistant rabbits. Pairs of normal and resistant animals were fed an 0.25% cholesterol enriched diet for 2 months. Samples of liver were removed, homogenized and fractionated into cytoplasmic and particulate fractions. The floating lipid layer was combined into the cytoplasmic fraction. Total lipid extracts were fractionated by TLC and analyzed for total cholesterol mass using the cholesterol oxidase assay and the data are shown in Fig. 9B. The combined sterol content in the liver of resistant rabbits was much lower than in livers from normal animals. The free cholesterol content of the supematant in the resistant animals was 12% of the mass of cholesterol in the normal rabbits; similarily the cholesteryl ester mass in the cytosol of the resistant rabbits was 2% of the mass found in the normal animals. This latter result is consistent with our earlier observation of reduced ACAT activity in liver membranes from the resistant animal fed a highcholesterol diet. The mass of lipid in the particulate fraction was also reduced in the resistant animals, but to a lesser extent than found in the supernatant fraction as illustrated in Fig. 9B. The extracts from the particulate fraction in resistant rabbits contained 65 and 62% of the mass of free
0 0
12
3
4
5
6
7
6
6
1011
FRACTION
Fig. 9. Determination of cellular cholesterol mass in normal and resistant rabbits. Livers were removed, and 0.5-g (wet weight) samples were homogenized and then centrifuged at 104ooO~ g. Cytosolic (9A) and pellet (9B) components were extracted twice with chloroform/methanol (2 : 1) and the lipids were then fractionated by TLC and assayed for cholesterol by the cholesterol oxidase technique. The migration of cholesterol and cholesteryl ester (as cholesteryl oleate) standards is indicated. Samples from 2 animals in each group were pooled before fractionation, and the data are expressed as pg/fraction with the equivalent of 100 mg liver wet weight analyzed from normal ( 0) and resistant (0) rabbits.
cholesterol compared
and cholesteryl esters, to normal rabbits.
respectively,
as
The present studies have focussed on defining the phenotype of three relevant biochemical markers involved in cholesterol homeostasis in a colony of hypercholesterolemia-resistant rabbits. While several accounts of variable response to cholesterol feeding have been reported, no data exist on the B/E receptor binding characteristics,
178 cholesterol synthesis or cholesterol esterification in a defined population of hypo- verus hyperresponders. Two fundamentally different mechanism(s) could be responsible for the biochemical differences we have demonstrated in ACAT activity, HMG-CoA reductase activity and B/E receptor binding capacity in the resistant rabbits. The observed differences could reflect primary differences in the expression or regulation of these proteins which must have an underlying genetic component. Alternatively, these differences may reflect a secondary response due to the lower plasma and intracellular cholesterol concentration maintained by the resistant rabbit. Several independent lines of evidence suggest that the observed biochemical differences reflect, at least in part, inherent genetic differences in the resistant animals. The lipoprotein binding capacity of the B/E receptor as measured using homologous pVLDL as the ligand, was consistently higher in the resistant rabbits than in normal rabbits. During consumption of the low-cholesterol diet, the resistant animals had approximately 70% higher B/E receptor binding than did normal animals. The basal serum cholesterol concentration in the resistant rabbits fed a low cholesterol diet was slightly lower than the normal animals fed the same diet (16.6 + 3.3 vs. 35.9 &-4.3 mg/dl, respectively). Thus it could be considered that the lower plasma cholesterol concentration in the resistant rabbits could have contributed to the higher numbers of B/E receptors in the resistant animals. However, other studies have shown that the normal and resistant rabbits fed the low-cholesterol chow have comparable plasma cholesterol concentrations. That is, normal rabbits had plasma cholesterol levels within a range of 12-70 mg/dl (mean = 44 + 4 mg/dl, n = 139) [1,32,33]; resistant rabbits had a similar range (8-76 mg/dl, mean = 37 f 15 mg/dl, n = 70) and mean value. Accordingly, the rabbits used in the present study had values within these ranges. In contrast, when cholesterol ingestion was increased, the plasma cholesterol responses between the two groups of rabbits were sharply disparate. This would indicate that there may be fundamental receptor differences which are independent of small differences in cholesterol levels when fed a low-
cholesterol diet. These differences could thus constitute primary mechanisms for the regulation of plasma cholesterol levels during periods of increased cholesterol ingestion. Secondly, more direct evidence for an intrinsic difference in B/E receptor expression is provided by other studies which showed that circulating peripheral mononuclear cells from the resistant rabbits had approximately 50% higher B/E receptor binding capacity than cells from normal animals even after 48 h of culture ex vivo under identical conditions [l]. In addition, the [‘251]LDL binding capacity of cultured skin-cell fibroblasts from resistant rabbits was 50% greater than the [‘251]LDL binding capacity found in normal rabbit fibroblasts [34]. Taken together these data strongly suggest that the differences in binding capacity in the resistant animals under conditions of this study were independent of plasma cholesterol levels. It therefore appears that the quantitative differences in B/E receptor binding between normal and resistant rabbits was not due to modulation by high dietary cholesterol but represent a constituitive condition. The increased number of B/E receptors may have contributed to the lower plasma cholesterol levels in the resistant rabbits. The resistant rabbits did not repress hepatic B/E receptor binding to the same degree as normal animals and the difference in total binding capacity increased to 240% when the animals were fed the high cholesterol diet. Significantly, we did not observe any difference in the binding affinity of the hepatic B/E receptor between normal and resistant animals. The P-VLDL labelled in the binding studies included both apo E and B. Apo E binds to the B/E receptors with an affinity which is about lo-fold higher than apo B [35]. We have not yet assessed the binding characteristics of the resistant hepatic membranes using radiolabelled apo B as ligand. However, in the studies with mononuclear cells [l], and in skin-cell fibroblasts [34] apo B (as LDL) was the sole ligand and gave results similar to those reported in the present studies. This indicates that the increased binding capacity of the B/E receptor in resistant rabbits occurs with both high and low affinity lipoprotein binding. B/E receptor binding studies in CaZf-free media, combined with the additional studies in mononuclear cells [l] which do not contain the EDTA-resistant
179 binding site, indicates that the observed differences in P-VLDL binding in the resistant rabbits are not due to the non-specific receptor. The activity of HMG-CoA reductase was higher in the resistant rabbits than in normal rabbits, when fed either the low- or high-cholesterol diet. This was not attributable to an alteration in the K, of the enzyme for substrate. One mechanism which has been reported to contribute to variability in cholesterol responsiveness in humans is a greater reduction in cholesterol biosynthesis in hyporesponsive individuals [15]; but this is clearly not the case in the resistant rabbits in the present study. Resistant rabbits fed the high-cholesterol diet were refractory to the expected inhibition of HMG-CoA reductase activity. The decrease in enzymic activity was about 90% in normal animals whereas the reduction in the resistant animals was only 25%. These findings are consistent with the results which showed that the expected decrease in the number of B/E receptors in response to the high-cholesterol diet was also blunted in the resistant rabbit. That similar differences in HMGCoA reductase activity have been observed in fibroblasts cultured from the resistant and normal rabbits suggests that the increased enzyme activity in the resistant animals is not due simply to differences in serum cholesterol concentrations [34]. Whereas HMG-CoA reductase activity was elevated in the resistant animals, ACAT activity was significantly reduced as compared to normal rabbits and this difference was maintained when the animals were fed either the low or cholesterolenriched diets. This may be due to the opposing regulation of these two enzymes by cholesterol, and reflect a fundamental difference in cholesterol metabolism in the resistant rabbits. In any case, the resistant animal does not simply remove plasma cholesterol by esterification with subsequent hepatic storage. Fed a high-cholesterol diet, the normal animals had 6-fold higher ACAT activity compared to resistant animals and no difference in the K, of the enzyme between normal and resistant rabbits was detected. All three of the biochemical markers we have investigated (B/E receptor binding, HMG-CoA reductase and ACAT activities) suggested that the resistant animal has a lower basal level of intracellular cholesterol. This is reflected in higher
P-VLDL receptor binding capacity, higher HMGCoA reductase activity, and reduced ACAT activity as compared to normal animals with comparable levels of plasma cholesterol. The differences in binding and enzyme activity are exaggerated when the animals are placed on a high-cholesterol diet. The resistant rabbits had smaller changes in B/E receptors, HMG-CoA reductase activity, or ACAT activity than normal animals in response to a cholesterol challenge. These findings suggest that the resistant animals maintained a low intracellular cholesterol concentration as well as a low plasma cholesterol concentration even when challenged with a cholesterol-enriched diet. Direct measurement of intracellular cholesterol supports this conclusion. In the cholesterol-fed state, the concentration of intracellular cholesterol in the cytosolic fraction was very low in the resistant rabbit (approximately 10% of the levels found in normal rabbits) measured as total cholesterol mass. Preliminary sterol balance studies have demonstrated that the resistant animals do not have a cholesterol absorption defect. Cholesterol absorption in normal rabbits was 22 & 3 mg/kg and in resistant rabbits it was 18.3 f 2.35 mg/kg [36]. Some of the difference in cholesterol levels can be attributed to differences in cholesterol excretion. When fed a 0.10% cholesterol diet, the resistant rabbits excreted approximately twice as much deoxycholic acid per diem compared to normal rabbits with no significant difference in the excretion of neutral sterols or lithocholic acid [36]. It is not yet possible to conclude that the increase in deoxycholic acid excretion accounts for all of the ingested cholesterol. The most abundant lipoprotein in the resistant animal fed a high cholesterol diet is HDL, whereas for the normal animal it is fi-VLDL [l]. However, the higher concentration of HDL in the resistant rabbit does not account for all of the ingested cholesterol. Several possibilities are under investigation regarding the metabolic fate of dietary cholesterol but these must be considered as speculative at this time. The resistant rabbit may compartmentalize cholesterol in a manner that reduces its effectiveness as a biochemical regulator (e.g., the ability to cause a reduction in the B/E receptor). In addition, perhaps the resistant animal is able to efficiently degrade cholesterol by an undefined mechanism.
180
These several considerations investigated.
are currently
being
Acknowledgements
This research was supported by NIH grants HL-38284 to M.L.O.; HL-27341 to J.D.M and DK-38965 to D.S.L.-M. J.P. is a Pharmaceutical Manufacturer’s Association Foundation Advanced Predoctoral fellow. References 1 Overturf, M.L., Smith, S.A., Hewett-Emmett, D., LooseMitchell, D.S., Soma, M.R., Gotto, A.M., Jr. and Morrisett, J.D., Development and partial metabolic characterization of a dietary cholesterol-resistant colony of rabbits, J. Lipid Res., 30 (1989) 263. 2 Fillios, L.C. and Mann, G.V., The importance of sex in the variability of cholesteremic response of rabbits fed cholesterol, Circ. Res., 4 (1956) 406. 3 Roberts, D.C.K., West, C.E., Redgrave, T.G. and Smith, J.B., Plasma cholesterol in normal and cholesterol-fed rabbits. Its variation and heritability, Atherosclerosis, 19 (1974) 369. 4 Van Zutphen, L.F.M., Den Bieman, M.G.C.W. Hulsman, W.C. and Fox, R.R., Genetic and physiological aspects of cholesterol accumulation in hyperresponding and hyporesponding rabbits, Lab. Animals, 15 (1981) 61. 5 Takeuchi, N., lto, M. and Yamamura, Y., Cholesterol metabolism in rats sensitive to high cholesterol diet, Adv. Exp. Med. Biol., 67 (1976) 267. 6 Clarkson, T.B. and Lofland, H.B., Effects of cholesterol-fat diets on pigeons susceptible and resistant to atherosclerosis, Circ. Res., 9 (1961) 106. 7 Clarkson, T.B., Lofland, H.B., Bullock, B.C. and Goodman, H.O., Genetic control of plasma cholesterol, Arch. Pathol., 92 (1971) 37. 8 Clarkson, T.B., Kaplan, J.R. and Adams, M.R., The role of individual differences in lipoprotein, artery wall, gender and behavioral responses in the development of atherosclerosis, Ann. N.Y. Acad. Sci., 454 (1985) 28. 9 Cormor, W.E., Hodges, R.E. and Bleiler, R.E., The serum lipids in men receiving high cholesterol and cholesterol-free diets, J. Clin. Invest., 40 (1961) 894. 10 Ginsberg, H., Le, M.A., Mays, C., Gibson, J. and Brown, W.V., Lipoprotein metabolism in nonresponders to increased dietary cholesterol, Arteriosclerosis, 1 (1981) 463. 11 Jacobs, D.R., Anderson, J.T., Hannan, P., Keys, A. and Blackbum, H., Variability in individual serum cholesterol response to change in diet, Arteriosclerosis, 3 (1983) 349. 12 McNamara, D.J., Kolb, R., Parker, T.S., Batwin, H., Samuel, P., Brown, C.D. and Ahrens, E.H., Heterogeneity of cholesterol homeostasis in man, J. Clin. Invest, 79 (1987) 1729.
13 Katan, M.B. and Beynen, A.C., Characteristics of human hypo- and hyperresponders to dietary cholesterol, Am. J. Epidemiol., 125 (1987) 387. 14 Bachorik, P.S., Franklin, F.A., Virgil, D.G. and Kwiterovich, P.O., Jr., High-affinity uptake and degradation of apolipoprotein E free high-density lipoprotein and low-density lipoprotein in cultured porcine hepatocytes, Biochemistry, 21 (1982) 5675. 15 Quintao, E., Grundy, S.M. and Ahrens, E.H., Jr., Effects of dietary cholesterol on the regulation of total body cholesterol in man, J. Lipid Res., 12 (1971) 233. 16 Nestel, P.J. and Poyser, A., Changes in cholesterol synthesis and excretion when cholesterol intake is increased, Metabolism, 25 (1976) 1591. 17 Brown, M.S. and Goldstein, J.L., Lipoprotein receptors in the liver, J. Clin. Invest., 72 (1983) 743. 18 Erickson, S.K., Shrewsbury, M.A., Brooks, C. and Meyer, D.J., Rat liver acyl-coenzyme A : cholesterol acetyltransferase: its regulation in vivo and some of its properties in vitro, J. Lipid Res., 21 (1980) 930. 19 Kritchevsky, D., Tepper, S.A. and Langan, J., Cholesterol vehicle in experimental atherosclerosis. Part IV. Influence of heated fat and fatty acids, J. Atherscler. Res., 2 (1962) 115. 20 Helgerud, P., Saarem, K. and Norum, K.R., AcylCoA : cholesterol acyl transferase in human small intestine: its activity and some properties of the enzymic reaction, J. Lipid Res., 22 (1981) 271. 21 Erickson, SK. and Fielding, P.E., Parameters of cholesterol metabolism in the human hepatoma cell line, Hep-G2, J. Lipid Res., 27 (1986) 875. 22 Brecher, P., Chobanian, J., Small, D.M. and Chobanian, A.V., The use of phospholipid vesicles for in vitro studies of cholesteryl ester hydrolysis, J. Lipid Res., 17 (1976) 239. P.T., Brown, M.S., Basu, S.K., Bilheimer, D.W. 23 Kovanen, and Goldstein, J.L., Saturation and supression of hepatic lipoprotein receptors: a mechanism for the hypercholesterolemia of cholesterol-fed rabbits, Proc. Natl. Acad. Sci. USA, 78 (1981) 1396. 24 Kovanen, P.T., Bilheimer, D.W., Goldstein, J.L., Jaramillo, J.J. and Brown, M.S., Regulatory role for hepatic low density lipoprotein receptors in vivo in the dog, Proc. Natl. Acad. Sci. USA, 78 (1981) 1194. J.D., Gaubatz, J.W., Tarver, A.P., Allen, J.K., 25 Morrisett, Pownall, H.J., Laggner, P. and Hamilton, J.A., Thermotropic properties and molecular dynamics of cholesteryl ester rich very low density lipoproteins: Effect of hydrophobic core on polar surface, Biochemistry 23 (1984) 5343. 26 Bilheimer, D.W., Eisenberg, S. and Levy, RI., The metabolism of very low density lipoprotein proteins. I. Preliminary in vitro and in vivo observations, B&him. Biophys. Acta, 260 (1972) 212. 27 Kita, T., Brown, M.S., Watanabe, Y. and Goldstein, J.L., Deficiency of low density lipoprotein receptors in liver and adrenal gland of the WHHL rabbit, an animal model of familial hypercholesterolemia, Proc. Natl. Acad. Sci. USA, 78 (1981) 2268. 28 Siperstein, M.D. and Fagan, V.M., Feedback control of
181
29
30
31
32
mevalonate synthesis by dietary cholesterol, J. Biol. Chem.. 241 (1966) 602. Rothblatt, G.H., Nauftulin, M. and Arbrogast, L.Y., Stimulation of acyl-CoA : cholesterol acyltransferase activity by hyperlipemic serum proteins, Proc. Sot. Exp. Biol. Med., 155 (1977) 501. Bilheimer, J.T., Tavani, D. and Nes, W.R., Effect of a dispersion of cholesterol in Triton WR-1339 on acyl CoA : cholesterol acyltransferase in rat liver microsomes. Anal. Biochem., 111 (1981) 501. Gilles, P.J., Rathgeb, K.A., Perri, M.A. and Robison, C.S.. Regulation of acyl CoA : cholesterol acyltransferase activity in normal and atherosclerotic rabbit aortas: role of a cholesterol substrate pool, Exp. Mol. Pathol., 44 (1986) 329. Overturf, M.. Sybers, S.. Schaper, J. and Teagtmeyer, H., Hypertension and atherosclerosis in cholesterol-fed rabbits. Part 1. Mild, two-kidney, one-clip Goldlatt hypertension treated with enalopril, Atherosclerosis, 59 (1986) 283.
33 Overturf, M., Sybers, H., Schaper, J. and Taegtmeyer, H., Hypertension and atherosclerosis in cholesterol-fed rabbits. Part 2. One kidney, one clip Goldblatt hypertension treated with nifedipine, Atherosclerosis, 66 (1987) 63. 34 Soma, M.R., Morrisett. J.D., Gotto Jr.. A.M., LooseMitchell, D.S., Poorman, J.A., Smith, S.A. and Overturf, M.L.. Cholesterol metabolism in fibroblasts from rabbits resistant to dietary-induced hypercholesterolemia, J. Lipid Res., 31 (1990) 985. 35 Mahley. R.W., Innerarity, T.L.. Rail, S.C., Jr. and Weisgraber, K.H.. Plasma lipoproteins: apolipoprotein structure and function, J. Lipid Res., 25 (1984) 1277. 36 Overturf. M.L., Smith, S.A., Gotto. A.M.. Jr., Morrisett. J.D., Tewson. T., Poorman, J. and Loose-Mitchell, D.S., Dietary cholesterol absorption, and sterol and bile acid excretion in hypercholesterolemia-resistant white rabbits, J. Lipid Res., (1990) in Press.
169 Ireland,
Ltd. 0021-9150/91/$03.50
A DONIS 002191509100095N
ATHERO
04615
Cholesterol metabolism in hypercholesterolemia-resistant
rabbits
David S. Loose-Mitchell ‘, Julie A. Poorman ‘, Sheryl A. Smith ‘, Merrill L. Overturf ‘, Joel D. Morrisett 2, Antonio M. Gotto, Jr. 2 and Maurizio R. Soma ’ ’Departments of Pharmacology and Medicine, University of Texas Health Sciences Cenrer at Houston, Houston TX 77030 (U.S.A.), and -’Departments of Medicine and Biochemistry, Baylor College of Medicine and The Methodist Hospital, Houston, TX, 77030 (U.S.A.) (Revised,
(Received 12 June, 1990) received 7 September and 3 December, (Accepted 4 December, 1990)
1990)
Normal rabbits typically respond to a diet high in cholesterol with a large increase in the concentration of plasma cholesterol. We have previously described the breeding and partial characterization of a variant rabbit which does not respond to a high cholesterol diet with changes in plasma cholesterol concentration. In the present report we have characterized three components involved in cholesterol homeostasis: the B/E (LDL) receptor, 3-hydroxy-3-methylglutaryl coenzyme A reductase activity (HMG-CoA reductase, EC 1.1.1.34) and acyl-coenzyme A : cholesterol acyltransferase activity (ACAT, EC 2.3.1.26) in the livers of the hypercholesterolemia-resistant rabbits. Using normal cholesterol-fed rabbit [‘251]/3-VLDL as a ligand, liver membranes prepared from resistant rabbits fed a low-cholesterol diet had 70% higher binding capacity than membranes from normal rabbits fed the same diet. Similar experiments demonstrated that the resistant rabbits had a 240% higher B/E receptor binding capacity compared to normal animals when liver membranes were prepared from animals fed a 0.25% cholesterol-enriched diet. No difference in the binding affinity of [‘251]/3-VLDL was detected in membranes prepared from normal or resistant animals. When fed a low-cholesterol diet, the resistant rabbits had approximately 2-fold higher hepatic HMG-CoA reductase activity (97.4 + 3.5 pmol product/mg/min in resistant animals compared to 45 + 1.1 pmol product/min/mg in normal animals). The difference was exaggerated in animals fed the 0.25% cholesterol-enriched diet, 73.3 + 5.5 vs 2.4 + 0.56 pmol product/min/mg for resistant and normal membranes respectively. The basal activity of ACAT in hepatic membranes was significantly lower in the resistant rabbits compared to normal rabbits (138 + 11 vs 268 + 19 pmol cholesteryl ester/min/mg in
Correspondence to: Dr. D.S. Loose-Mitchell, Department of Pharmacology, University of Texas Health Sciences Center, 6431 Fannin, Houston, TX 77030, U.S.A. Abbreviations: HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; ACAT, acyl-coenzyme A : cholesterol acyltrans-
ferase; LDL, low density lipoprotein; /?-VLDL, j3migrating very low density lipoprotein; B/E receptor, apoprotein B and apoprotein E receptor; NADP, B-nicotinamide adenine dinucleotide phosphate; DTT, dithiothreitol; BSA, bovine serum albumin; TLC, thin layer chromatography.
170 resistant and normal rabbits respectively); when fed a 0.25% cholesterol-enriched diet, the enzyme was induced 6-fold in normal animals but was increased only 2-fold in the resistant animal. These biochemical data suggested that the resistant rabbit maintained low intracellular cholesterol even when fed a cholesterol-enriched diet. Direct measurement of cellular cholesterol and cholesteryl esters demonstrated that the concentration of these lipids was significantly lower in the resistant animal than in normal animals with the largest differences found in the cytoplasmic rather than the membrane compartment. These studies demonstrate that the resistant rabbit manifests several quantitative differences in cholesterol metabolism and in the regulation of cholesterol metabolism; but these studies do not directly explain the underlying cause of the resistance to hypercholesterolemia in the resistant rabbit.
Key words: Hypercholesterolemia; Atherosclerosis; Cholesterol metabolism; reductase; Acyl-CoA : cholesterol acyltransfersase; (Rabbits)
Introduction We have developed and partially characterized a colony of New Zealand white rabbits which are strikingly resistant to the hypercholesterolemia which typically results from a high-cholesterol diet [l]. A 0.1% cholesterol enriched diet fed to normal rabbits results in plasma cholesterol levels of approximately 330 mg% after 8 weeks. Under the same dietary conditions the resistant rabbits maintain plasma cholesterol of less than 50 mg% *. The phenotype of the resistant rabbit is inheritable; however, we have not yet identified the precise mode of this inheritance. The phenomenon of resistance to hypercholesterolemia following dietary cholesterol challenge has been described in rabbits [2-41, rats [5], pigeons [6], and nonhuman primates [7,8]. Well controlled studies in humans have also documented variable resistance to hypercholesterolemia in response to dietary cholesterol [9-141. In all species which have been investigated, wide variations have been observed in plasma cholesterol levels following cholesterol challenge. This variation extends from individuals with little or no increase in plasma cholesterol to individuals with severe hy-
* The resistant phenotype is defined as an elevation in plasma cholesterol concentration greater than or equal to 1.5 SD below the mean plasma cholesterol concentration of normal rabbits at both 4 and 8 weeks on a 0.1% cholesterol-enriched diet.
LDL receptor;
HMG-CoA
percholesterolemia. In the human population, differences in response to cholesterol challenge have been noted in the same individual [14] tested at multiple times, but significant differences in response between different individuals has also been documented [13]. While genetic factors have been implicated in determining part of this variable cholesterol response, the mechanism(s) by which certain individuals maintain a low plasma cholesterol concentration following cholesterol ingestion is not clear in any species. It has been suggested that compensatory reductions in cholesterol synthesis [15], increased biliary cholesterol excretion [16], or differences in the ability of gut bacteria to conjugate bile acids [13] may contribute to the variability in responsiveness. The resistant rabbit colony we have developed may provide a model for the “resistant”, or hyporesponsive individuals within the human population. In this study we have investigated the biochemical response of several enzymes that play critical roles in cholesterol metabolism. These studies have focussed on the in vivo characteristics of hepatic proteins since the liver is the major site of cholesterol metabolism in normal mammals [17]. Activities of the enzymes 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase) and acyl-coenzyme A : cholesterol acyltransferase (ACAT), and the lipoprotein binding capacity of the B/E receptor have all been found to differ significantly in hepatic tissue from resistant rabbits maintained on a low-cholesterol
171 diet compared to levels of these activities in normal rabbits. In addition, the expected biochemical responses in the activities of these proteins to a cholesterol-enriched diet were blunted in resistant rabbits as compared to the normal rabbits. These results suggest that the resistant animal may maintain reduced intracellular cholesterol concentrations in both basal and cholesterol-fed states. Direct measurement of cytosolic cholesterol in the resistant rabbits has confirmed that even on a high-cholesterol diet these animals maintained very low cellular cholesterol and cholesteryl ester concentrations.
Materials and methods Materials
3-Hydroxy-3-methyl[3-‘4C]glutaryl coenzyme A (52 Ci/mol) and [l-‘4C]oleoyl coenzyme A (52 Ci/mol) were purchased from Amersham (Arlington Heights, IL). [la,2a(n)-3H]Cholesterol (50 Ci/mol), [5-3H]mevalonolactone (21 Ci/mmol), and Na”‘I (17 Ci/mg) were purchased from New England Nuclear (DuPont Co., Boston, MA). Silica (Si 250) plates were purchased from Baker (Phillipsburg, NJ). [ 3H]Cholesteryl oleate was prepared from [3H]cholesterol and oleic anhydride [18] and purified by TLC. All solvents for chromatography were analytical grade or better and were purchased from Baker. Unless specifically noted all other reagents were purchased from Sigma Chemical Co. (St. Louis, MO) and were the highest quality available.
cholesterol-enriched diet used in the current studies consisted of standard chow supplemented with 0.25 g cholesterol (USP) and 2.0 ml corn oil per 100 g chow. Preliminary experiments established that the resistant rabbits lacked a hypercholesterolemic response when fed this cholesterol-enriched diet while normal rabbits developed a pronounced and rapid hypercholesterolemia. The cholesterol and corn oil were dissolved in chloroform (ACS) and mixed with chow pellets which had been freed of fines by sieving [19]. Saturated pellets were left in a well ventilated fume hood until the odor of chloroform was not detectable. Rabbits were fed 100 g chow per day until they were four months old; thereafter the ration was increased to 150 g/day. Baseline cholesterol concentration in the normal animals was 29.5 mg/dl (fed low-cholesterol chow) and after 8 weeks on the 0.25% cholesterol-enriched diet, plasma cholesterol in the normal animals increased to an average of 1052 mg/dl (range, 768-1335). Resistant animals used in this portion of the study had average baseline cholesterol concentrations of 21 mg/dl which increased to 51 mg/dl (range, 33-68) 8 weeks after being fed the 0.25% cholesterol-enriched diet. Serum cholesterol in all normal animals fed the low-cholesterol diet was 35.9 + 4.3 mg/dl and 16.6 + 3.3 mg/dl in all resistant animals fed the same diet. Breeding and phenotypic characterization of the resistant rabbit has been described previously [I]. Enzyme
assays
Preparation Animals
All rabbits were caged individually and maintained with normal 12 hour light/dark cycles. All rabbits were males between 24 and 30 months of age and resistant rabbits were from either the 4th or 6th generation within the resistant pedigree. The standard laboratory rabbit chow (Ralston Purina, St. Louis, MO) contained approximately 60 pg cholesterol per g chow. The 0.1% cholesterol-enriched diet used for phenotypic analysis has been described previously [l] and all animals were phenotyped prior to study. The 0.25%
of hepatic microsomes
Microsomes for enzymatic studies were isolated by centrifugation essentially as described by Erickson et al. [18]. In brief, liver sections were isolated from anesthetized animals killed between 9 and 11 a.m. One gram of tissue was homogenized in 5 volumes (w/v) of 0.25 M sucrose, 1 mM Na,EDTA (pH 7.4). The homogenate was then centrifuged at 4°C for 15 min at 9000 x g. After removal of the floating fatty layer, the supernatant was removed and ultracentrifuged at 4’C for 45 min at 134 000 x g. The resulting pellet was resuspended in 2 ml of the same solution. This suspension was ultracentrifuged at 4°C a second time
172 and the final pellet resuspended in 1 ml 0.2 M K,HPO, (pH 7.6) and then stored at -80°C.
A CA T assays
Samples of liver microsomes (25-200 pg) were assayed [20] in 225 ~10.2 M K,HPO, containing 5 mg/ml BSA and, except in experiments examining the effects of substrate concentration, 67.68 PM oleoyl CoA (7.68 PM [‘4C]oleoyl CoA, plus 60 PM unlabelled oleoyl CoA). All experiments employed a minimum of two microsomal protein concentrations for activity analysis. Microsomes were preincubated at 37°C in the phosphate-BSA buffer for 5 min before addition of the substrate. Substrate was then added and the reaction mixture was incubated at 37°C for 5 min and the reaction was terminated by the addition of 5 ml chloroform/methanol (2 : 1). In some experiments exogenous cholesterol was added to the ACAT assay to assess the effects of cholesterol availability essentially as described by Erickson and Fielding [21]. Cholesterol was added as cholesterol/ phospholecithin unilammellar liposomes (1 : 6 molar ratio) prepared exactly as described [22]. Liposomes were incubated with microsomes 30 min at 37°C prior to assay. Approximately 10000 dpm of [3H]cholesteryl oleate were added to each sample in order to estimate recovery. Following the incubation period, 1.05 ml pH 3.0 water was added to the reaction mixture, samples were mixed thoroughly and left at 4°C overnight. The mixture was then centrifuged at 2000 rpm for 10 min and the upper phase was removed and discarded. The lower phase was washed twice with 1.2 ml fresh upper phase (chloroform/ methanol/ water, 3 : 48 : 47). The addition of less than 1 ml methanol resulted in solubilization of the remaining interface before the organic phase was dried in air. The residue was resuspended in hexane, spotted on a TLC plate and chromatographed a distance of 10 cm in hexane/ethyl acetate (9 : 1). Following autoradiography, the band corresponding to cholesteryl oleate was isolated and counted in a dual label scintillation counter. Recovery of cholesteryl oleate was 41 k 0.04% for all samples assayed (n = 44) and recovery from samples in an individual experiment varied by no more than 11%.
HMG-CoA
reductase assays
Samples of liver microsomes (25-250 pg) were assayed in a buffer containing 0.1 M K,HPO, (pH 7.5), 20 mM glucose 6-phosphate, 2.5 mM NADP, 0.7 units glucose 6-phosphate dehydrogenase, 5 mM DTT, and, except in experiments where substrate concentration was varied, 87 PM HMG-CoA (12.5 PM [‘4C]HMG-CoA plus 74.5 PM unlabelled HMG-CoA) in a total volume of 200 ~1. At least two protein concentrations were used to measure enzyme activity. Microsomes were preincubated in the reaction buffer at 37°C for 10 min before addition of the HMG-CoA and then incubated for 20 min at 37°C with moderate shaking. The reaction was stopped by the addition of 30 ~15 N HCl, and 10 000 dpm [ 3H]mevalonolactone standard were added to measure recovery. The reaction solution was then incubated at 37°C for 30 min to allow the formation of the lactone of mevalonate. After addition of 5 volumes of diethyl ether, the solution was mixed thoroughly and left at 4°C overnight. The upper phase was transferred to a second tube and the lower phase was extracted 5 times with 5 volumes of ether. The combined upper phases were dried and the residue was resuspended in acetone. The incubates were spotted on a TLC plate and chromatographed for 15 cm in acetone/benzene (1: 1). Following autoradiography, the band corresponding to mevalonolactone was isolated and counted in a dual label scintillation counter. Recovery of labelled mevalonolactone was 51 f 0.03% for all samples assayed (n = 37) and recovery from samples in an individual experiment varied by less than 13%. p- VLDL binding assays
Preparation of membrane fractions for lipoprotein binding studies was done as described by Kovanen et al. [23,24]. Liver tissue was homogenized in a solution containing 50 mM Tris-HCl (pH 8.0) 0.15 M NaCl and 1 mM Na,EDTA on ice. The homogenate was centrifuged at 9000 x g and the pellet discarded. The supernatant was ultracentrifuged at 134000 X g for 45 minutes and the pellet was washed in the same buffer and pelleted by ultracentrifugation. The final pellet was resuspended in a solution of 50 mM NaCl, 1
173 mM CaCl,, and 20 mM Tris-HCl frozen until use. Preparation
of [“‘I]
(pH
8.0) and
/3- VLDL
P-VLDL (d < 1.006 g/ml) was isolated [25] from a normal New Zealand white rabbit maintained on a 2% cholesterol diet. Purified /3-VLDL was iodinated according to Bilheimer [26] to an average specific activity of 500 cpm/ng protein. Binding assays
B/E receptor binding assays were performed as described by Kovanen et al. 123,241 with some modifications. Rabbit liver membranes were thawed, passed through a 25-gauge hypodermic needle and sonicated for 45 s with a Branson Sonifier at a power setting of 3. Binding assays were performed in 110 ~1 of a buffer containing 50 mM Tris-HCl, 25 mM NaCl, 1 mM CaCl,, and 20 mg/ml bovine serum albumin. /3-VLDL binding resistant to EDTA was assessed by incubating membranes as described above except a buffer which contained 2 mM Na,EDTA rather than 1 mM CaCl, was used. Membranes were incubated with 25-2000 ng of [‘251]j3-VLDL for 60 min at 4°C. Non-specific binding was assessed by incubating parallel samples with a 50-fold excess of unlabelled P-VLDL. Bound ligand was separated from free by layering 75 ~1 of the binding reaction on top of 100 ~1 100% fetal bovine serum in a 225 ~1 nitrocellulose tube. Samples were then centrifuged at 100000 X g for 30 min at 4°C in a Beckman Airfuge. The tubes were then decanted and the bottom was cut from the tube and the pellet was counted. Localization
of intracellular
cholesterol
Liver sections, 0.5 g wet weight, were prepared from animals maintained on a 0.25% cholesterolenriched diet. Tissue was homogenized in 3.0 ml 25 mM phosphate buffer (pH 7.5) with a Teflon grinder. Particulate matter and cytosol were separated by ultracentrifugation at 104000 X g for 1 h. The pellet was resuspended in a volume of homogenization buffer equal to the volume of the supernatant. One ml each of the supernatant and the pellet was mixed with 20 volumes of chloroform/methanol (2 : 1) and 4.2 ml pH 3.0 water
and then stored overnight at 4°C. The extraction was repeated and the organic phases were combined. The combined lower phase was washed twice with fresh upper phase (chloroform/ methanol/ water, 3 : 48 : 47). The interface was discharged by the addition of less than 1 ml methanol before the solution was dried under nitrogen. The residue was resuspended in hexane, spotted on a TLC plate, and chromatographed for 12 cm in acetic acid/acetone/petroleum ether (1 : 25 : 75). The lipids were visualized in iodine vapors, and the bands corresponding to cholesterol and cholesteryl esters were eluted from the silica taken to dryness in 2 : 1 cholorform/methanol, and redissolved in isopropanol. Total cellular cholesterol was measured using the cholesterol oxidase assay (Boehringer Mannheim Biochemicals, Indianapolis, IN) as described by the manufacturer. The isopropanol is required to re-dissolve cholesteryl esters prior to assay. Recovery of both lipids was 75-80% and was assessed using 14Clabelled internal standards. Results The B/E receptor binding capacity and affi,$ty of liver membranes were assessed using Ilabelled rabbit P-VLDL as ligand. Under the conditions used, binding was found to be linear from 50 to 300 pg membrane protein and binding equilibrium was obtained within 30 min at 4°C (data not shown). In typical binding experiments, 150 pg membrane protein were incubated with [‘251]P-VLDL for 60 min. A Scatchard analysis combined from the data in 5-7 individual binding experiments from liver membranes prepared from normal and resistant rabbits fed a normal chow (low cholesterol) diet is illustrated in Fig. 1. The binding capacity of membranes from resistant rabbits was 70% higher than that of normal rabbits with no apparent difference in receptor affinity (Kd = 6.78 pg/ml in normals and 5.21 pgg/ml in resistant liver membranes). An EDTA-resistant binding site which is relatively non-specific for lipoprotein particles has been described in the liver of rabbit [27] and pig [14]. To assess the contribution made by this binding site in our studies, binding experiments were conducted in 2 mM EDTA. Under these conditions, normal rab-
Fig. 1. Scatchard analysis of rabbit (‘*sI]/3-VLDL binding to hepatic membranes prepared from normal (0) and resistant animals (0) on a normal (low cholesterol) diet. Liver membranes, 150 ng protein/sample, were incubated with 40-2500 ng [‘*‘1]/3-VLDL in 110 pl for 60 min at 4°C. Non-specific binding was assessed in parallel samples by including a 50-fold excess of unlabelled b-VLDL. Non-specific binding was approximately 28%. The data presented are the mean of 5-7 separate binding experiments from 3 normal and 3 resistant rabbits. Serum cholesterol concentration in the normal animals was 35.9 f 4.3 and 16.6 f 4.3 in the resistant animals, respectively.
bit hepatic membranes bound 0.123 ng [‘251]pVLDL/pg protein and liver membranes from the resistant rabbits bound 0.14 ng [1251]/3-VLDL/pg protein (at 10 pg/ml [‘2sI]j%VLDL). Subtraction of the contribution of the EDTA-resistant binding site increases the difference between the two phenotypes. Ca2+-dependent binding in membranes from resistant rabbits was 95% higher than in normal rabbits compared to 70% higher when total binding is analyzed. The results of binding experiments to liver membranes from rabbits fed a 0.25% cholesterol diet for two months were analyzed in a similar manner and the results are presented in Fig. 2. As expected, the high-cholesterol diet resulted in a 50% decrease in the number of hepatic B/E receptors in the normal animals (0.400 5 0.020 vs 0.205 + 0.070 ng [i2?]j3-VLDL/pg membrane protein off and on diet, respectively, P < 0.02). However, the resistant rabbits fed the high cholesterol diet had a much smaller reduction in the number of B/E receptors, and the change was not statistically significant (0.68 * 0.150 vs 0.50 + 0.10, P > 0.05). The combined effects of (i) higher numbers of B/E receptors under basal (low-cholesterol diet)
and (ii) a smaller percent reduction in receptor numbers following cholesterol challenge, resulted in a 2.4-fold higher B/E binding capacity in the cholesterol-fed resistant rabbits compared to cholesterol-fed normal rabbits. Subtraction of the contribution of the EDTA-resistant binding, which is not regulated by plasma cholesterol concentration [27], from these data again increased the difference observed between the two phenotypes; in this case Ca2+-dependent binding was 4.4-fold higher in the resistant rabbit versus 2.4-fold higher for total binding. In order to assess the capacity of the resistant rabbit liver to synthesize cholesterol we examined the activity of HMG-CoA reductase, the ratelimiting enzyme in cholesterol biosynthesis [28]. Liver membranes were prepared from the livers of normal and resistant rabbits and then assayed for HMG-CoA reductase activity by measuring the formation of mevalonic acid from 3-hydroxy-3methyl[‘4C]glutaryl coenzyme A. Conditions for the assay were optimized using membranes from rabbits fed the low-cholesterol diet (i.e., with induced levels of HMG-CoA reductase activity). HMG-CoA reductase activity as a function of protein concentration is illustrated in Fig. 3. The assay was linear to approximately 100 pg mem-
F
-
+ NORMAL
-
+
J
RESISTANT
Fig. 2. Summary of the effect of dietary cholesterol feeding on [‘251]P-VLDL binding in liver membranes from normal and resistant rabbits. Normal and resistant animals (3 in each group) were fed standard rabbit chow (-) or rabbit chow supplemented with 0.25% cholesterol (+) (2 animals in each group) for 8 weeks prior to experimentation. Membranes were prepared and analyzed as in Fig. 1. Scatchard analysis of the binding data was used to determine the B/E receptor binding capacity (N,,,,,) in each experiment. Results of 5 separate experiments (+ SEM) on each group are presented.
175
0
25
50
75
MICROSOMAL
100
125
PROTEIN
(pg)
150
175
Fig. 3. HMG-CoA reductase activity in hepatic membranes from normal (0) and resistant (0) rabbits on a low cholesterol diet as a function of protein concentration. The indicated mass of membrane protein was incubated under standard conditions in a final volume of 200 pl. Membranes were incubated with [‘4C]HMG-CoA (87 PM) for 20 min. After acidification of the reaction mixture to lactonize the mevalonate, samples were fractionated by TLC and the products were scraped from the plates and then counted in a liquid scintillation counter. Results presented are the mean and standard error of 4-6 determinations from samples prepared from 3 normal and 3 resistant rabbits.
brane protein, hence in all subsequent studies we employed 50 and 75 pg protein. These data were obtained using an incubation time of 20 min which was found to be optimum for this preparation (data not shown). When the animals were fed the low-cholesterol diet, membranes from resistant rabbits had approximately twice the hepatic HMG-CoA reductase activity compared to the normal rabbits (97 pmol mevalonolactone/min/ mg protein vs 45 pm01 mevalonolactone/min/mg protein, P < 0.02). This difference in enzyme activity could not be attributed to changes in the K, of the enzyme for the substrate HMG-CoA. Liver membranes were incubated with various concentrations of [‘4C]HMG-CoA (6.25-150 PM) and the results of this substrate-dependence analysis are illustrated in Fig. 4. Double reciprocal plot analysis of the data presented in Fig. 4 indicates that the K, of HMG-CoA was 21 PM in normal rabbits and 34 PM in the resistant animals; a difference which was not statistically significant. A series of studies was then undertaken to assess the effects of dietary cholesterol on the activity of HMG-CoA reductase in the liver of normal and resistant rabbits and the results are summarized in Fig. 5. After 2 months on an 0.25% cholesterol
HMG COENZYME
A (PM)
Fig. 4. Substrate dependence of HMG-CoA reductase in liver membranes from normal (0) and resistant (0) rabbits on a low-cholesterol diet. Membranes were prepared from 3 resistant and 3 normal animals and incubated with various concentrations of HMG-CoA. Each point represents the mean activity of determinations at 50 and 75 yg membrane protein from 2 separate experiments.
diet, HMG-CoA reductase activity was nearly abolished (reduced 91%) in the normal rabbits. In the resistant rabbits, HMG-CoA reductase activity was reduced only 25% by dietary cholesterol. These results are similar to those obtained for the B/E receptor binding capacity (Fig. 2) in that dietary cholesterol was less effective in reducing HMGCoA reductase activity in the resistant rabbits than in normal rabbits. In addition to the B/E receptor binding and ^,
125,
+ NORMAL
+ RESISTANT
Fig. 5. Effect of dietary cholesterol on HMG-CoA reductase activity in normal and resistant rabbits. Animals were fed standard rabbit chow (- , 3 animals in each group) or standard chow supplemented with 0.25% cholesterol (+, 2 animals in each group) for 8 weeks. Membranes were prepared and HMG-CoA reductase activity was assayed at 50 and 75 pg membrane protein. Results are the mean and standard error of 6-8 determinations.
176 HMG-CoA reductase activity, both of which are inhibited by dietary cholesterol in normal animals, ACAT, an enzyme normally induced by exogenous cholesterol was also studied. The assay for ACAT activity was optimized using liver membranes of animals fed the 0.25% cholesterol diet. The assay employed measures the rate of incorporation of [‘4C]oleoyl-CoA into “C-labelled cholesteryl oleate using endogenous cholesterol as substrate; the reaction products were resolved by TLC. Conditions were optimized with respect to oleate/BSA ratio, and incubation time and the assay was found to be linear for 2-10 min (data not shown). Fig. 6 illustrates ACAT activity as a function of membrane protein. The assay was linear up to approximately 75 pg protein and in all subsequent assays 50 and 75 pg protein were used to quantitate enzyme activity. Membranes from normal animals had about a six-fold higher ACAT activity when fed the high-cholesterol diet compared to membranes from resistant animals fed the same diet. The effect of substrate concentration on ACAT activity in rabbit hepatic membranes is illustrated in Fig. 7. These data indicate that the difference in ACAT activity between the normal and resistant rabbit membranes is not attributable to changes in the K, of the enzyme. Linear analysis of these data indicate that the apparent K, of ACAT for oleoyl-CoA was 22 PM for both the
.
. 0
20
40 OLEOYL
60
60 COENZYME
100
1
A (@.d)
Fig. 7. Effect of varying oleoyl CoA concentration on the activity of ACAT in rabbit liver membranes. Membranes were prepared from normal (0) or resistant (0) rabbits which had been maintained for 8 weeks on a 0.25% cholesterol diet. Membrane protein, 50 and 75 pg/assay, was incubated with the indicated concentrations of oleoyl CoA, and ACAT activity was determined as described in Fig. 6.
normal and resistant rabbits. The effect of manipulating dietary cholesterol on ACAT activity in the resistant and normal rabbit is summarized in Fig. 8. While the normal animals respond to an increase in dietary cholesterol with approximately a 6-fold increase in the activity of the enzyme, ACAT activity in the resistant animals increases by less than 2-fold. Several investigators have reported that ACAT activity is regulated, in part, by the availability of substrate [29-311. To determine if cholesterol substrate pools were af-
1
120 100 60 60
0
25
50
75
MICROSOMAL
100
125
PROTEIN
(pg)
150
175
Fig. 6. ACAT activity in liver membranes prepared from normal (0) and resistant (0) rabbits after 8 weeks on a 0.25% cholesterol diet. Hepatic membranes were prepared and ACAT activity was assessed by determining the rate of incorporation of [‘4C]oleoyl CoA into [t4C]cholesteryl oleate. The indicated mass of membrane protein was incubated under standard conditions in a final volume of 225 pl. Endogenous cholesterol was used as the fatty acid acceptor.
I
_ NORMAL
+
-
+
-J
RESISTANT
Fig. 8. Effect of a low cholesterol (- , 3 animals in each group) or a 0.25% cholesterol supplemented diet ( + , 2 animals in each group) on ACAT activity in liver membranes from normal and resistant rabbits. ACAT activity was assessed at 50 and 75 pg membrane protein from animals that had been maintained on the indicated diet for 8 weeks. Results are the mean and standard error of 6-8 determinations.
fecting ACAT activities, exogenous cholesterol (220 pg) as cholesterol/phospholecithin unilammelar liposomes, was added to liver membranes prepared from normal and resistant rabbits fed the 0.25% cholesterol-enriched diet. The exogenous cholesterol was pre-incubated with the membranes 30 min prior to assay. Addition of 20 pg cholesterol to the rnicrosomes increased ACAT activity from 298 f. 20 to 330 + 27 in samples from resistant rabbits and from 1388 f 60 to 1671 k 111 in samples from normal animals. These slight increases suggest that cholesterol pool size does not account for the differences in ACAT activity between normal and resistant rabbits. The above studies suggest that the resistant animals were refractory to cholesterol-mediated changes in the activity of several biochemical pathways involved in cholesterol metabolism. One interpretation of these results is that the resistant animals maintained a low intracellular cholesterol level, even in the presence of high dietary cholesterol. We therefore sought to measure directly the intracellular cholesterol content in liver samples from normal and resistant rabbits. Pairs of normal and resistant animals were fed an 0.25% cholesterol enriched diet for 2 months. Samples of liver were removed, homogenized and fractionated into cytoplasmic and particulate fractions. The floating lipid layer was combined into the cytoplasmic fraction. Total lipid extracts were fractionated by TLC and analyzed for total cholesterol mass using the cholesterol oxidase assay and the data are shown in Fig. 9B. The combined sterol content in the liver of resistant rabbits was much lower than in livers from normal animals. The free cholesterol content of the supematant in the resistant animals was 12% of the mass of cholesterol in the normal rabbits; similarily the cholesteryl ester mass in the cytosol of the resistant rabbits was 2% of the mass found in the normal animals. This latter result is consistent with our earlier observation of reduced ACAT activity in liver membranes from the resistant animal fed a highcholesterol diet. The mass of lipid in the particulate fraction was also reduced in the resistant animals, but to a lesser extent than found in the supernatant fraction as illustrated in Fig. 9B. The extracts from the particulate fraction in resistant rabbits contained 65 and 62% of the mass of free
0 0
12
3
4
5
6
7
6
6
1011
FRACTION
Fig. 9. Determination of cellular cholesterol mass in normal and resistant rabbits. Livers were removed, and 0.5-g (wet weight) samples were homogenized and then centrifuged at 104ooO~ g. Cytosolic (9A) and pellet (9B) components were extracted twice with chloroform/methanol (2 : 1) and the lipids were then fractionated by TLC and assayed for cholesterol by the cholesterol oxidase technique. The migration of cholesterol and cholesteryl ester (as cholesteryl oleate) standards is indicated. Samples from 2 animals in each group were pooled before fractionation, and the data are expressed as pg/fraction with the equivalent of 100 mg liver wet weight analyzed from normal ( 0) and resistant (0) rabbits.
cholesterol compared
and cholesteryl esters, to normal rabbits.
respectively,
as
The present studies have focussed on defining the phenotype of three relevant biochemical markers involved in cholesterol homeostasis in a colony of hypercholesterolemia-resistant rabbits. While several accounts of variable response to cholesterol feeding have been reported, no data exist on the B/E receptor binding characteristics,
178 cholesterol synthesis or cholesterol esterification in a defined population of hypo- verus hyperresponders. Two fundamentally different mechanism(s) could be responsible for the biochemical differences we have demonstrated in ACAT activity, HMG-CoA reductase activity and B/E receptor binding capacity in the resistant rabbits. The observed differences could reflect primary differences in the expression or regulation of these proteins which must have an underlying genetic component. Alternatively, these differences may reflect a secondary response due to the lower plasma and intracellular cholesterol concentration maintained by the resistant rabbit. Several independent lines of evidence suggest that the observed biochemical differences reflect, at least in part, inherent genetic differences in the resistant animals. The lipoprotein binding capacity of the B/E receptor as measured using homologous pVLDL as the ligand, was consistently higher in the resistant rabbits than in normal rabbits. During consumption of the low-cholesterol diet, the resistant animals had approximately 70% higher B/E receptor binding than did normal animals. The basal serum cholesterol concentration in the resistant rabbits fed a low cholesterol diet was slightly lower than the normal animals fed the same diet (16.6 + 3.3 vs. 35.9 &-4.3 mg/dl, respectively). Thus it could be considered that the lower plasma cholesterol concentration in the resistant rabbits could have contributed to the higher numbers of B/E receptors in the resistant animals. However, other studies have shown that the normal and resistant rabbits fed the low-cholesterol chow have comparable plasma cholesterol concentrations. That is, normal rabbits had plasma cholesterol levels within a range of 12-70 mg/dl (mean = 44 + 4 mg/dl, n = 139) [1,32,33]; resistant rabbits had a similar range (8-76 mg/dl, mean = 37 f 15 mg/dl, n = 70) and mean value. Accordingly, the rabbits used in the present study had values within these ranges. In contrast, when cholesterol ingestion was increased, the plasma cholesterol responses between the two groups of rabbits were sharply disparate. This would indicate that there may be fundamental receptor differences which are independent of small differences in cholesterol levels when fed a low-
cholesterol diet. These differences could thus constitute primary mechanisms for the regulation of plasma cholesterol levels during periods of increased cholesterol ingestion. Secondly, more direct evidence for an intrinsic difference in B/E receptor expression is provided by other studies which showed that circulating peripheral mononuclear cells from the resistant rabbits had approximately 50% higher B/E receptor binding capacity than cells from normal animals even after 48 h of culture ex vivo under identical conditions [l]. In addition, the [‘251]LDL binding capacity of cultured skin-cell fibroblasts from resistant rabbits was 50% greater than the [‘251]LDL binding capacity found in normal rabbit fibroblasts [34]. Taken together these data strongly suggest that the differences in binding capacity in the resistant animals under conditions of this study were independent of plasma cholesterol levels. It therefore appears that the quantitative differences in B/E receptor binding between normal and resistant rabbits was not due to modulation by high dietary cholesterol but represent a constituitive condition. The increased number of B/E receptors may have contributed to the lower plasma cholesterol levels in the resistant rabbits. The resistant rabbits did not repress hepatic B/E receptor binding to the same degree as normal animals and the difference in total binding capacity increased to 240% when the animals were fed the high cholesterol diet. Significantly, we did not observe any difference in the binding affinity of the hepatic B/E receptor between normal and resistant animals. The P-VLDL labelled in the binding studies included both apo E and B. Apo E binds to the B/E receptors with an affinity which is about lo-fold higher than apo B [35]. We have not yet assessed the binding characteristics of the resistant hepatic membranes using radiolabelled apo B as ligand. However, in the studies with mononuclear cells [l], and in skin-cell fibroblasts [34] apo B (as LDL) was the sole ligand and gave results similar to those reported in the present studies. This indicates that the increased binding capacity of the B/E receptor in resistant rabbits occurs with both high and low affinity lipoprotein binding. B/E receptor binding studies in CaZf-free media, combined with the additional studies in mononuclear cells [l] which do not contain the EDTA-resistant
179 binding site, indicates that the observed differences in P-VLDL binding in the resistant rabbits are not due to the non-specific receptor. The activity of HMG-CoA reductase was higher in the resistant rabbits than in normal rabbits, when fed either the low- or high-cholesterol diet. This was not attributable to an alteration in the K, of the enzyme for substrate. One mechanism which has been reported to contribute to variability in cholesterol responsiveness in humans is a greater reduction in cholesterol biosynthesis in hyporesponsive individuals [15]; but this is clearly not the case in the resistant rabbits in the present study. Resistant rabbits fed the high-cholesterol diet were refractory to the expected inhibition of HMG-CoA reductase activity. The decrease in enzymic activity was about 90% in normal animals whereas the reduction in the resistant animals was only 25%. These findings are consistent with the results which showed that the expected decrease in the number of B/E receptors in response to the high-cholesterol diet was also blunted in the resistant rabbit. That similar differences in HMGCoA reductase activity have been observed in fibroblasts cultured from the resistant and normal rabbits suggests that the increased enzyme activity in the resistant animals is not due simply to differences in serum cholesterol concentrations [34]. Whereas HMG-CoA reductase activity was elevated in the resistant animals, ACAT activity was significantly reduced as compared to normal rabbits and this difference was maintained when the animals were fed either the low or cholesterolenriched diets. This may be due to the opposing regulation of these two enzymes by cholesterol, and reflect a fundamental difference in cholesterol metabolism in the resistant rabbits. In any case, the resistant animal does not simply remove plasma cholesterol by esterification with subsequent hepatic storage. Fed a high-cholesterol diet, the normal animals had 6-fold higher ACAT activity compared to resistant animals and no difference in the K, of the enzyme between normal and resistant rabbits was detected. All three of the biochemical markers we have investigated (B/E receptor binding, HMG-CoA reductase and ACAT activities) suggested that the resistant animal has a lower basal level of intracellular cholesterol. This is reflected in higher
P-VLDL receptor binding capacity, higher HMGCoA reductase activity, and reduced ACAT activity as compared to normal animals with comparable levels of plasma cholesterol. The differences in binding and enzyme activity are exaggerated when the animals are placed on a high-cholesterol diet. The resistant rabbits had smaller changes in B/E receptors, HMG-CoA reductase activity, or ACAT activity than normal animals in response to a cholesterol challenge. These findings suggest that the resistant animals maintained a low intracellular cholesterol concentration as well as a low plasma cholesterol concentration even when challenged with a cholesterol-enriched diet. Direct measurement of intracellular cholesterol supports this conclusion. In the cholesterol-fed state, the concentration of intracellular cholesterol in the cytosolic fraction was very low in the resistant rabbit (approximately 10% of the levels found in normal rabbits) measured as total cholesterol mass. Preliminary sterol balance studies have demonstrated that the resistant animals do not have a cholesterol absorption defect. Cholesterol absorption in normal rabbits was 22 & 3 mg/kg and in resistant rabbits it was 18.3 f 2.35 mg/kg [36]. Some of the difference in cholesterol levels can be attributed to differences in cholesterol excretion. When fed a 0.10% cholesterol diet, the resistant rabbits excreted approximately twice as much deoxycholic acid per diem compared to normal rabbits with no significant difference in the excretion of neutral sterols or lithocholic acid [36]. It is not yet possible to conclude that the increase in deoxycholic acid excretion accounts for all of the ingested cholesterol. The most abundant lipoprotein in the resistant animal fed a high cholesterol diet is HDL, whereas for the normal animal it is fi-VLDL [l]. However, the higher concentration of HDL in the resistant rabbit does not account for all of the ingested cholesterol. Several possibilities are under investigation regarding the metabolic fate of dietary cholesterol but these must be considered as speculative at this time. The resistant rabbit may compartmentalize cholesterol in a manner that reduces its effectiveness as a biochemical regulator (e.g., the ability to cause a reduction in the B/E receptor). In addition, perhaps the resistant animal is able to efficiently degrade cholesterol by an undefined mechanism.
180
These several considerations investigated.
are currently
being
Acknowledgements
This research was supported by NIH grants HL-38284 to M.L.O.; HL-27341 to J.D.M and DK-38965 to D.S.L.-M. J.P. is a Pharmaceutical Manufacturer’s Association Foundation Advanced Predoctoral fellow. References 1 Overturf, M.L., Smith, S.A., Hewett-Emmett, D., LooseMitchell, D.S., Soma, M.R., Gotto, A.M., Jr. and Morrisett, J.D., Development and partial metabolic characterization of a dietary cholesterol-resistant colony of rabbits, J. Lipid Res., 30 (1989) 263. 2 Fillios, L.C. and Mann, G.V., The importance of sex in the variability of cholesteremic response of rabbits fed cholesterol, Circ. Res., 4 (1956) 406. 3 Roberts, D.C.K., West, C.E., Redgrave, T.G. and Smith, J.B., Plasma cholesterol in normal and cholesterol-fed rabbits. Its variation and heritability, Atherosclerosis, 19 (1974) 369. 4 Van Zutphen, L.F.M., Den Bieman, M.G.C.W. Hulsman, W.C. and Fox, R.R., Genetic and physiological aspects of cholesterol accumulation in hyperresponding and hyporesponding rabbits, Lab. Animals, 15 (1981) 61. 5 Takeuchi, N., lto, M. and Yamamura, Y., Cholesterol metabolism in rats sensitive to high cholesterol diet, Adv. Exp. Med. Biol., 67 (1976) 267. 6 Clarkson, T.B. and Lofland, H.B., Effects of cholesterol-fat diets on pigeons susceptible and resistant to atherosclerosis, Circ. Res., 9 (1961) 106. 7 Clarkson, T.B., Lofland, H.B., Bullock, B.C. and Goodman, H.O., Genetic control of plasma cholesterol, Arch. Pathol., 92 (1971) 37. 8 Clarkson, T.B., Kaplan, J.R. and Adams, M.R., The role of individual differences in lipoprotein, artery wall, gender and behavioral responses in the development of atherosclerosis, Ann. N.Y. Acad. Sci., 454 (1985) 28. 9 Cormor, W.E., Hodges, R.E. and Bleiler, R.E., The serum lipids in men receiving high cholesterol and cholesterol-free diets, J. Clin. Invest., 40 (1961) 894. 10 Ginsberg, H., Le, M.A., Mays, C., Gibson, J. and Brown, W.V., Lipoprotein metabolism in nonresponders to increased dietary cholesterol, Arteriosclerosis, 1 (1981) 463. 11 Jacobs, D.R., Anderson, J.T., Hannan, P., Keys, A. and Blackbum, H., Variability in individual serum cholesterol response to change in diet, Arteriosclerosis, 3 (1983) 349. 12 McNamara, D.J., Kolb, R., Parker, T.S., Batwin, H., Samuel, P., Brown, C.D. and Ahrens, E.H., Heterogeneity of cholesterol homeostasis in man, J. Clin. Invest, 79 (1987) 1729.
13 Katan, M.B. and Beynen, A.C., Characteristics of human hypo- and hyperresponders to dietary cholesterol, Am. J. Epidemiol., 125 (1987) 387. 14 Bachorik, P.S., Franklin, F.A., Virgil, D.G. and Kwiterovich, P.O., Jr., High-affinity uptake and degradation of apolipoprotein E free high-density lipoprotein and low-density lipoprotein in cultured porcine hepatocytes, Biochemistry, 21 (1982) 5675. 15 Quintao, E., Grundy, S.M. and Ahrens, E.H., Jr., Effects of dietary cholesterol on the regulation of total body cholesterol in man, J. Lipid Res., 12 (1971) 233. 16 Nestel, P.J. and Poyser, A., Changes in cholesterol synthesis and excretion when cholesterol intake is increased, Metabolism, 25 (1976) 1591. 17 Brown, M.S. and Goldstein, J.L., Lipoprotein receptors in the liver, J. Clin. Invest., 72 (1983) 743. 18 Erickson, S.K., Shrewsbury, M.A., Brooks, C. and Meyer, D.J., Rat liver acyl-coenzyme A : cholesterol acetyltransferase: its regulation in vivo and some of its properties in vitro, J. Lipid Res., 21 (1980) 930. 19 Kritchevsky, D., Tepper, S.A. and Langan, J., Cholesterol vehicle in experimental atherosclerosis. Part IV. Influence of heated fat and fatty acids, J. Atherscler. Res., 2 (1962) 115. 20 Helgerud, P., Saarem, K. and Norum, K.R., AcylCoA : cholesterol acyl transferase in human small intestine: its activity and some properties of the enzymic reaction, J. Lipid Res., 22 (1981) 271. 21 Erickson, SK. and Fielding, P.E., Parameters of cholesterol metabolism in the human hepatoma cell line, Hep-G2, J. Lipid Res., 27 (1986) 875. 22 Brecher, P., Chobanian, J., Small, D.M. and Chobanian, A.V., The use of phospholipid vesicles for in vitro studies of cholesteryl ester hydrolysis, J. Lipid Res., 17 (1976) 239. P.T., Brown, M.S., Basu, S.K., Bilheimer, D.W. 23 Kovanen, and Goldstein, J.L., Saturation and supression of hepatic lipoprotein receptors: a mechanism for the hypercholesterolemia of cholesterol-fed rabbits, Proc. Natl. Acad. Sci. USA, 78 (1981) 1396. 24 Kovanen, P.T., Bilheimer, D.W., Goldstein, J.L., Jaramillo, J.J. and Brown, M.S., Regulatory role for hepatic low density lipoprotein receptors in vivo in the dog, Proc. Natl. Acad. Sci. USA, 78 (1981) 1194. J.D., Gaubatz, J.W., Tarver, A.P., Allen, J.K., 25 Morrisett, Pownall, H.J., Laggner, P. and Hamilton, J.A., Thermotropic properties and molecular dynamics of cholesteryl ester rich very low density lipoproteins: Effect of hydrophobic core on polar surface, Biochemistry 23 (1984) 5343. 26 Bilheimer, D.W., Eisenberg, S. and Levy, RI., The metabolism of very low density lipoprotein proteins. I. Preliminary in vitro and in vivo observations, B&him. Biophys. Acta, 260 (1972) 212. 27 Kita, T., Brown, M.S., Watanabe, Y. and Goldstein, J.L., Deficiency of low density lipoprotein receptors in liver and adrenal gland of the WHHL rabbit, an animal model of familial hypercholesterolemia, Proc. Natl. Acad. Sci. USA, 78 (1981) 2268. 28 Siperstein, M.D. and Fagan, V.M., Feedback control of
181
29
30
31
32
mevalonate synthesis by dietary cholesterol, J. Biol. Chem.. 241 (1966) 602. Rothblatt, G.H., Nauftulin, M. and Arbrogast, L.Y., Stimulation of acyl-CoA : cholesterol acyltransferase activity by hyperlipemic serum proteins, Proc. Sot. Exp. Biol. Med., 155 (1977) 501. Bilheimer, J.T., Tavani, D. and Nes, W.R., Effect of a dispersion of cholesterol in Triton WR-1339 on acyl CoA : cholesterol acyltransferase in rat liver microsomes. Anal. Biochem., 111 (1981) 501. Gilles, P.J., Rathgeb, K.A., Perri, M.A. and Robison, C.S.. Regulation of acyl CoA : cholesterol acyltransferase activity in normal and atherosclerotic rabbit aortas: role of a cholesterol substrate pool, Exp. Mol. Pathol., 44 (1986) 329. Overturf, M.. Sybers, S.. Schaper, J. and Teagtmeyer, H., Hypertension and atherosclerosis in cholesterol-fed rabbits. Part 1. Mild, two-kidney, one-clip Goldlatt hypertension treated with enalopril, Atherosclerosis, 59 (1986) 283.
33 Overturf, M., Sybers, H., Schaper, J. and Taegtmeyer, H., Hypertension and atherosclerosis in cholesterol-fed rabbits. Part 2. One kidney, one clip Goldblatt hypertension treated with nifedipine, Atherosclerosis, 66 (1987) 63. 34 Soma, M.R., Morrisett. J.D., Gotto Jr.. A.M., LooseMitchell, D.S., Poorman, J.A., Smith, S.A. and Overturf, M.L.. Cholesterol metabolism in fibroblasts from rabbits resistant to dietary-induced hypercholesterolemia, J. Lipid Res., 31 (1990) 985. 35 Mahley. R.W., Innerarity, T.L.. Rail, S.C., Jr. and Weisgraber, K.H.. Plasma lipoproteins: apolipoprotein structure and function, J. Lipid Res., 25 (1984) 1277. 36 Overturf. M.L., Smith, S.A., Gotto. A.M.. Jr., Morrisett. J.D., Tewson. T., Poorman, J. and Loose-Mitchell, D.S., Dietary cholesterol absorption, and sterol and bile acid excretion in hypercholesterolemia-resistant white rabbits, J. Lipid Res., (1990) in Press.