In vitro regulation of cholesterol metabolism by low density lipoproteins in skin fibroblasts from hypo- and hyperresponding squirrel monkeys

458

Biochimica et Biophysics @ Elsevier/North-Holland

Acta,

3X7 (1977)

Biomedical

45X-471

Press

BBA 57006

IN VITRO REGULATION OF CHOLESTEROL METABOLISM BY LOW DENSITY LIPOPROTEINS IN SKIN FIBROBLASTS FROM HYPO- AND HYPERRESPONDING SQUIRREL MONKEYS

LEE S. GUERTLER

and RICHARD

W. St. CLAIR

The Arteriosclerosis Research Center, Department Medicine, Winston-Salem, N.C. 27103 (U.S.A.) (Received

December

22nd,

ofpathology,

Bowman

Gray School

of

1976)

Summary When squirrel monkeys (Suimiri sciureus) are fed diets containing cholesterol, some individuals (hyperresponders) become hypercholesterolemic, while others (hyporesponders) are able to maintain nearly normal plasma cholesterol concentrations. Skin fibroblasts were grown from three hyperresponder and three hyporesponder squirrel monkeys, previously characterized on the basis of their plasma cholesterol response to two cholesterol-containing diets and the phenotype of their parents. The rates of cholesterol synthesis and esterification were determined in the cultured fibroblasts incubated with low density lipoproteins isolated from normocholesterolemic squirrel monkeys or hypercholesterolemic rhesus monkeys. Both lipoprotein preparations influenced the metabolic parameters measured in a similar manner in cells from both hypo- and hyperresponder animals. Exposure of skin fibroblasts to low density lipoproteins resulted in a stimulation of cholesterol esterification and a suppression of cholesterol synthesis in cells from both hypo- and hyperresponder animals. When incubated with increasing concentrations of low density lipoprotein cholesterol, up to 50 pg/ml, fibroblasts from both hypo- and hyperresponding animals responded with a similar maximum percentage suppression of sterol synthesis. Thus, hyperresponsiveness to dietary cholesterol in squirrel monkeys, although a heritable characteristic, is not associated with an inability of low density lipoprotein to suppress cholesterol synthesis or stimulate cholesterol esterification as occurs in familial hypercholesterolemia in man.

Introduction The mechanisms and man are poorly

controlling understood.

plasma cholesterol concentrations in animals It is apparent, however, that genetic factors

459

play a major role in influencing the concentration of plasma lipoproteins [1,2]. Genetically, the simplest of these disorders is familial hypercholesterolemia. Studied extensively by Brown and Goldstein [ 31, this is the only one in which the molecular defect is known. As a result of an autosomal dominant defect, the plasma membranes of extrahepatic cells such as skin fibroblasts lack functionally normal low density lipoprotein receptor sites. Familial hypercholesterolemia accounts for only a small proportion of cases of hypercholesterolemia in human beings, with more complex genetic disorders, such as familial combined hyperlipemia and polygenic hypercholesterolemia being more prevalent [ 121. Regardless of the genetic defect, all of these disorders result in hyperlipoproteinemia. Thus, there appear to be a number of biochemical mechanisms that can result in hypercholesterolemia. Intake of dietary cholesterol represents another means of producing hypercholesterolemia, with some individuals being particularly responsive to dietary cholesterol. This phenomenon of hyper- and hyporesponsiveness to dietary cholesterol is seen in a number of animal species including squirrel monkeys [4], rhesus monkeys [ 5,6], African green monkeys [7], rabbits [ES], and pigeons [9] . In squirrel monkeys [4], and presumably in other species as well, the individuality to response of dietary cholesterol is under genetic control. The biochemical expression of this genetic control is poorly understood, and although differences in bile acid excretion [lO,ll] and cholesterol absorption [ 5,111 between hypo- and hyperresponders have been described, it is difficult to know whether these changes represent the underlying biochemical mechanisms responsible for individuality in response to dietary cholesterol. The purpose of this study was to determine whether differences in cholesterol metabolism exist in skin fibroblasts from hypo- or hyperresponsive squirrel monkeys when exposed in culture to low density lipoproteins. Methods Classification of animals as hypo- and hyperresponders. 24 juvenile squirrel monkeys (Saimiri sciureus), the offspring of known hypo- or hyperresponders, were fed a diet of Purina Special Monkey Chow-25 (Ralston-Purina Co.) until 18 months of age. They were then fed a semipurified diet (Table I, Diet A) containing 1.056 mg cholesterol/cal. This same diet had been used to classify their parents as hypo- or hyperresponders. After the animals had consumed Diet A for 6 months, an adequate period of time for them to have reached a maximum plasma cholesterol response [ 10,121, plasma cholesterol concentrations were determined weekly for three consecutive weeks and the mean of these three determinations was used to classify the animals. All animals then were fed monkey chow for 6 months, then a natural products diet (Diet B) containing 0.66 mg cholesterol/cal for an additional 6 months, with three consecutive weekly plasma cholesterol concentrations determined at the end of this period, exactly as for Diet A. Blood was drawn into tubes containing 1 mg disodium ethylenediamine tetraacetic acid/ml, and plasma was separated by centrifugation. Total plasma cholesterol concentrations were determined using the AutoAnalyzer II methodology of Rush et al. [13]. The three monkeys (2237, 2241, 2231) with the highest mean plasma cholesterol concentrations for both

360

Diets A and B were selected as hyperresponders for these studies, while the three with the lowest mean plasma cholesterol concentrations (2225, 2227, 2301) were selected as hyporesponders (Table II). In some experiments we also used cells obtained from two older siblings identified as hypo- and hyperresponders in a similar manner. These were animals 1414 and 1422. Initiation and maintenance of skin fibroblasts in culture. Skin biopsies (1 X 2 cm) were obtained from the abdomen under aspetic conditions. The tissue was cut into 1Omm cubes and approx. 20-30 of these explants were placed in each of three Falcon tissue culture flasks (75 cm’). Initial outgrowth of epithelial cells was seen after 2-3 days, with fibroblasts appearing after 3-4 days. Primary cell cultures were grown to confluency in 2-3 weeks in a humid atmosphere of 5% CO2 and 95% air at 37°C. Each tissue culture flask contained 20 ml of culture medium consisting of Eagles minimum essential medium (,4utoPow), supplemented with twice the normal concentration of vitamins (Eagle’s vitamins), 10% fetal calf serum, 200 mM L-glutamine, 1 mg a-u-(+)-glucose/ml, 100 I.U. penicillin/ml and 100 pg streptomycin/ml. All tissue culture supplies were obtained from Flow Laboratories. Cells were utilized after 4-10 passages in culture with a 1 : 3 split for each passage. For all experiments, cells from stock 75-cm’ flasks were dissociated with 0.05% trypsin/0.02% disodium ethylenediamine tetraacetic acid and were placed at a concentration of approx. 2 . 10’ cells into 100-mm tissue culture dishes containing 10 ml of culture medium. The culture medium was changed every 3 days and cells were used after approx. 7 days while they were still in the log growth phase and contained 100-200 pg cell protein per dish. Isolation and analysis of lipoproteins. Lipoproteins were isolated from 21 squirrel monkeys consuming monkey chow and having a mean plasma cholesterol concentration of 135 mg/dl and from five rhesus monkeys (Macaca mulatta) with a mean plasma cholesterol concentration of 615 mg/dl and consuming a hypercholesterolemic diet (Diet A, Table I). ;‘ill animals were fasted for 16 h; blood was collected in tubes containing 1 mg disodium ethylenediamine tetraacetic acid/ml and plasma was separated by centrifugation. An aliquot of plasma was taken for total plasma cholesterol concentration and low density lipoproteins were isolated using the combined ultracentrifugal-agarose column chromatographic method of Rude1 et al. [ 141. Low density lipoproteins isolated by this technique have a flotation density of 1.019-1.063 g/cm3. Isolated lipoprotein fractions were subsequently concentrated for use in culture by dialyzing them for 2 h at 4°C against 1 1 of a solution of 30% sucrose and 0.10% disodium ethylenediamine tetraacetic acid, followed by 12 h against a second liter of the same solution. The sucrose was removed by dialysis against two l-l changes of phosphate-buffered saline and two l-l changes of Eagle’s minimum essential medium. The isolated low density lipoproteins were sterilized through a 0.45~pm Millipore filter (Millipore Filter Co.) and were added to culture medium containing lipid-deficient serum 1151 at the specific cholesterol concentrations indicated in the legends of the individual tables and figures. Preparation of cholesterol/lecithin liposomes. Liposomes containing 2 mol of cholesterol per mol egg lecithin were prepared by dissolving 390 mg of twice recrystallized cholesterol and 350 mg egg lecithin in 10 ml chloroform. The

461 chloroform was evaporated under N, and 125 ml of 0.9% NaCl solution was added. The mixture was sonicated (Sonifier Cell Disrupter, No. W185, Heat Systems-Ultrasonics, Inc.) until it passed through a 0.45~pm Millipore filter. After centrifugation to remove titanium particles, the liposomes were dialyzed against 1 1 of Eagle’s minimum essential medium for 24 h. The liposomes were sterilized by filtration through a 0.45 pm filter before use in culture. Preparation of substrates. Sodium [ 1-14C] acetate (New England Nuclear) was dissolved in 0.9% NaCl solution to a final concentration of 1.7 pmol and 5 PCi of sodium [1-‘“Cl acetate per 100 ~1. This solution was sterilized by filtration through a 0.45 pm Millipore filter, and when added in 5 ml of culture medium, the final acetate concentration was 0.34 PM. The [ l-l4 C] oleic acid substrate was prepared exactly as described previously [15]. When 50 1.11were added to 5 ml of culture medium it equaled a final oleate concentration of 0.16 mM. The oleate to albumin molar ratio of the substrate was 6 : 1.

Measurement of total cholesterol and total fatty acid synthesis acetate. Fibroblasts in the log phase of growth were maintained

from [1-‘“Cl-

with culture medium containing lipid-deficient serum (2.5 mg protein/ml) for 24 h in order to deplete them of lipid prior to addition of the specific lipoprotein to be tested. At this time, the culture medium containing lipid-deficient serum was replaced with 5 ml of culture medium containing lipid-deficient serum (2.5 mg protein/ml) as well as the specific lipoprotein to be tested and incubated for an additional 24 h. The cells were freed from the dishes with trypsin-disodium ethylenediamine tetraacetic acid, transferred to 12 ml centrifuge tubes, and the cells were washed twice with phosphate-buffered saline. 1 ml of deionized water was added and the cells were disrupted by sonication for 5 s at 55 W utilizing a Heat Systems Model W185 Sonifier with a micro tip. Protein was determined on 0.2 ml of the sonicated cell suspension by the method of Lowry et al. [16] using bovine albumin as the standard. To the remaining 0.8 ml were added 3.5 ml redistilled ethanol containing [1,2-3H,] cholesterol as an internal standard, 1 mg of non-radioactive cholesterol as a carrier, 0.5 ml of 10 M NaOH and 0.2 ml deionized water. The samples were saponified at 70°C for 2 h and the non-saponifiable lipids were extracted three times with a few ml each of redistilled Skellysolve B (Skelly Oil Co.). The 3-ohydroxy sterols were isolated from the Skellysolve B extract as the digitonide, by the method of Sperry and Webb [17], and in one experiment the procedure of Schwenk and Werthessen [18] was utilized for the subsequent purification of cholesterol as the dibromide. Following extraction of the non-saponifiable lipids, the remaining aqueous phase was acidified with HCl and fatty acids were extracted with three washes of Skellysolve B. The cholesterol digitonide was dissolved in 1 ml of warm methanol and added to a vial containing 10 ml of scintillation fluid composed of 6 g 2,5-diphenyloxazole per 1 of toluene. Fatty acids were dissolved in the above scintillation fluid without methanol. Radioactivity was counted in a Beckman LS-230 liquid scintillation counter to a 2-sigma error of <2%. Quenching was corrected for by an external standard channels ratio method and cholesterol synthesis was corrected for recovery of the [ 1 ,2-3H2] cholesterol internal standard. Measurement of cholesterol esterification from [I-’ 4C] oleic acid. After

462

incubation of cells with culture medium containing lipid-deficient serum for 24 h, exactly as described for the [1-‘4C] acetate studies, culture medium containing lipiddeficient serum and the lipoproteins to be tested was added for 6 h, followed by addition of the [ I-l41 oleic acid substrate for 4 h. After the cells were isolated and disrupted by sonication as previously described, the lipids were extracted [19] and separated by thin-layer chromatography on silica gel G, using Skellysolve B, ethyl ether and acetic acid (146 : 50 : 4, v/v) as the running solvent. The fractions corresponding to authentic standards of cholesterol and cholesteryl oleate were scraped directly into scintillation vials and counted for radioactivity in scintillation fluid containing 6 g 2,5-diphenyloxazole, 100 ml Bio-Solv (Beckman) and 30 ml water per 1 of toluene. Results for the esterification of [l-14C] oleic acid to cholesterol were corrected for recovery of the [ 1,2-“H,] cholesterol internal standard. Results The squirrel monkeys selected for this study were identified as hypo- or hyperresponders according to their plasma cholesterol response to two different cholesterol-containing diets, and by virtue of their parental history. The compositions of the diets used for classification are shown in Table I and the results of this classification are shown in Table II. All animals, except 1422 and 1414, were selected from a group of 24 animals, all offspring born in our primate colony. Most of the parents of these animals were classified previously as hypo- (HO) or hyperresponders (HP) with Diet A. These 24 animals were fed Diets A and B as described in Methods and ranked according to their plasma cholesterol response. The animal with the lowest plasma cholesterol concentration was ranked number 1, while the highest was number 24. As shown in Table II, the animals selected for these studies represented the extremes of the rank order of plasma cholesterol response. This was true when they were challenged with either diets A or B, indicating that hypo- and hyperresponsiveness to dietary cholesterol is not a pecularity of a single diet [ 4,111. The average plasma cholesterol concentration of hyperresponder animals selected for this study and consuming 1 mg cholesterol/cal (diet A) was 638 mg/dl, while hyporesponders averaged 241 mg/dl. When information on the phenotype of the parents was available, it was found to be the same as that of the offspring, emphasizing the major genetic component of hypo- and hyperresponsiveness to dietary cholesterol 141. Initially, we investigated the growth characteristics of skin fibroblasts from a representative hypo- and hyperresponsive animal. Results are shown in Fig. 1. As can be seen, there were no obvious differences in the rate of growth of skin fibroblasts from the two phenotypes. In addition, the number of cells and the amount of cell protein were closely parallel to one another. Therefore, we have expressed the subsequent metabolic data on the basis of cell protein as an expression of the number of cells in culture. All metabolic experiments typically were initiated when cultures were approx. 80% confluent, while still in the log phase of growth, and containing approx. 100-200 pg protein/dish. For studies of cholesterol synthesis, we used [ 1-‘4C] acetate as substrate. We

463 TABLE

I

COMPOSITION

OF DIETS G/100 g

Ingredient

Protein

(g)

Cholestxrol

Calories

Carbohydrate

Fat (9)

(mg)

(K)

Diet A 25 20 7.3 30 13 2

Lard Wheat flour Applesauce Non-dat dry milk solids Casein. U.S.P. U.S.P. XIV salts mixture Complete vitamin mix (devoid of vitamin D) Vitamin D-3 * Cholesterol * *

2.10 0.01 10.68 13

90 10 100

Total

_

8.8 _ _

_

100

25.79

25.21

36.23

474.96

22.9

4.59

6.60

46.44 0.25

387.9

2.2 25.1

11.19

46.69

453.9

500 502

220

66 .o

’ 2.5 I.U. per g of diet. 2 Equivalent to 1.056 mg cholesterol/al. ’ Analysis of Purina Monkey chow-25 provided by the Ralston Purina Co. The remaining monkey chow not shown in this table represents nondigestible constituents. 4 Equivalent to 0.48 mg cholesterol/cal diet. TABLE

2 _ _ _

52

_ 2.2 -

0.5

***

225 78.2 5.82 105.14

17 1.43 15.6

2.2

Total Diet B Purina special Monkey chow-25 Dried egg yolk ’

25 0.02 0.007 0.003 _ _

220

16.07 g of

II

DIETARY CLASSIFICATION FIBROBLASTS WERE GROWN

AND

PARENTAL

HISTORY

OF

MONKEYS

FROM

WHICH

SKIN

The dietary rank order is a comparison of the mean response of an animal’s plasma cholesterol concentration obtained from three weekly determinations while consuming Diets A and B. The animal with a rank order of 1 had the lowest mean plasma cholesterol concentration and the animal with a rank order of 24 the highest mean plasma cholesterol concentration out of the 24 animals tested. Animals 1422 and 1414 were older than the other animals selected for this study, but were also classified according to their plasma cholesterol response to Diet A. as well as their being offspring of known hype- (HO) or hyperrewonder (HP) parents. Parents were classified as hype- and hyperresponders utilizing Diet A. We have indicated those sires and dams in which the dietary classification was unknown (UN). S‘.ZX

Age (Years)

Cholestwo1 Baseline (mg/dl)

Diet A Cholesterol

Diet B Rank order

(mg/dl) Hyperrespcnders 2237 3 2241 3 2231 3 1422 5 Hyporespnders 2225 3 2227 3 2301 3 1414 5

(HP) M F M F

Cholestero1

Rank order

Dam and classification

Sire and classification

1507 1553 409 409

1148 1148 747 1717

(mgldl)

159 148 143

983 474 466 631

24 20 19 -

237 178 186 _

24 18 21

116 136 152 _

218 224 264 259

1 2 4

110 155 134

1 6 2

UN HP HP HP

HP HP HP HP

(HO) M F M F

1418 HO 419 HO UN 434 HO

1039 HO 1039 HO UN 1039 HO

464

Days of Culture Fig. 1. Growth curves for skin fibroblasts from hype- and hyperresponding squirrel monkeys. 1 . 10’ cells each from animals, 1414 fhyporesponder) and 1422 (hyperresponderf, were transferred to 60 X 15 mm culture dishes on day 0. Viable cells (based on trypan blue exclusion) were counted each C&Y for 10 days using a hemocytometer. Cell protein was also measured beginning on day 3. Each point represents the results from a single dish of cells.

initially isolated the cholesterol synthesized from acetate both by thin-layer chromatography and by digitonin precipitation. Of the total sterol radioactivity isolated by chromatography, only approx. 40% was precipitable with digitonin, while there was virtually 100% recovery of the authentic cholesterol added as carrier. Consequently, considerable quantities of non-saponifiable radioactive material from [l-‘“Cl acetate ~o~hromato~aphed with free cholesterol, but apparently were not 3-/3-hydroxysterols. Thus, we have used digitonin precipitation, rather than thin-layer chromatography, to isolate newly synthesized 3-phydroxysterols. Further purification, as the dibromide of the sterols isolated with digitonin, indicated that only about 2% of the 3-/3-hydroxysterols precipitated by the digitonin was actually cholesterol. The remainder presumably represents sterols synthesized from acetate, but prior to conversion to cholesterol. Similar results have been reported by others [20,21]. In subsequent experiments, when we refer to cholesterol synthesis from [ 1-14C] acetate, we recognize that this actually represents total sterol synthesis. Before comparing sterol synthesis in skin fibroblasts from hypo- and hyperresponding animals, we determined the concentration of substrates required for saturation kinetics for both cholesterol synthesis and cholesterol esterification. Skin fibroblasts from a representative hypo- and hyperresponsive animal were grown to approx. 80% confluency and incubated with lipid-deficient, serumcontaining culture medium for 24 h. The [ 1-‘4C] acetate substrate was added to the culture medium at the concentrations and times shown in Fig. 2. Acetate was no longer rate limiting for cholesterol synthesis above a concentration of approx. 0.17 mM (Fig. 2, Panel A). The rate of cholesterol synthesis in cells grown previously on lipiddeficient serum appeared greater for cells from the hyporesponder than from the hyperresponder. The synthesis of cholesterol was linear up to at least 24 h of incubation with [l-14C] acetate (Fig. 2, Panel B). In subsequent metabolic studies involving cholesterol synthesis, we have used

465

0---o M

1414 HO 1422 HP

I I ,017 ,068

I

I

17

.34

Acetate

Concentration

(mM

4

8 12 Incubation

5 ‘; 0

I

24 time (h)

0.016

0.064 Oleate

0.16 Concentration

0.32 (mM)

I 2 3 4 Incubation time (h)

Fig. 2. Effect of substrate concentration (A) and incubation time (B) on sterol synthesis from [ 1-14C] acetate by skin fibroblasts from hype- (HO) and hyperresponder (HP animals. Experiment A: Ceils were incubated for 24 h with culture medium containing lipid-deficient serum in order to maximize cholesterol synthesis. [1-‘4C1 Acetate was added at the indicated concentrations and the cells were incubated for an additional 24 h. Experiment B: C&s were incubated for 24 h with culture medium containing lipid-deficient serum, then [l-14~lacetate (0.32 mM) was added and the cells were incubated for the indicated times. Each point represents the mean of duplicate determinations. Fig. 3. Effect of substrate concentration (A) and incubation time (B) on esterification of [ 1-14C] oleate to cholesterol in skin fibroblasts from hypo- (HO) and hyperresponding (HP) animals. Ceils were incubated for 24 h with culture medium containing lipid-deficient serum and then for 6 h with culture medium containing whole normocholesterolemic squirrel monkey serum (174 mg/dl) at a final cholesterol concentration of 58 pg/ml. In Experiment A. the cells were incubated for an additional 4 h with the [l-‘4Cloleate substrate at the concentrations indicated. In Experiment B, the cells were incubated with the [I-‘“C]oleate substrate for an additional l-4 h at an oleate concentration of 0.16 mM. Each value represents the result of a single determination.

an incubation time of 24 h and an acetate concentration of 0.34 mM. Fig. 3 shows the influence of oleate concentration and incubation time on the rate of cholesterol esterification in skin fibroblasts from a representative hypo- and hyperresponder. Above a concentration of approx. 0.16 mM, the concentration of oleate was no longer rate-limiting (Fig. 3, Panel A). This concentration is similar to that shown by others for skin fibroblasts from human beings [ 221. The rate of esterification of oleate to cholesterol was linear up to 4 h of incubation (Fig. 3, Panel B). In subsequent studies in which cholesterol esterification was measured, we used an incubation time of 4 h and an oleate concentration of 0.16 mM. The results of addition of either squirrel or rhesus monkey low density lipo-

SQUIRREL

MONKEY

LDL

RHESUS

MONKEY LDL l 2237 HP ~2241 HP 02301 HO 0 1414HO l 2227 HO

s+-A-$ I

2

~CJ Cholesterol

/ml

Culture

Medium

Fig. 4. The effect of low density lipoprotein (LDL) cholesterol from squirrel or rhesus monkeys on the incorporation of [l-‘4Clacetate into 3-(3-hydroxy sterols and fatty acids in skin fibroblasts from hypo(HO) and hyperresponding (HP) squirrel monkeys. F‘ibroblasts were grown to approx. 80% confluency at which time culture medium containing lipiddeficient serum was added. After 24 h of incubation, culture medium containinglow density lipoprotein was added and the cells were incubated for 24 h. [ 1-‘4C1 Acetate substrate was added and the cells were incubated for an additional 24 h in order to measure sterol synthesis. In all incubations, acetate was added at a final concentration of 0.34 mM. In the experiments in which squirrel monkey low density lipoprotein was used (Panels A. B, and C), the specific activity of the [1-‘4] acetate substrate was 1.11 Ci/mol. In those experiments in which rhesus monkey low density lipoprotein was used (Panels D, E, and F), the specific activity of the [l-‘4Clacetate substrate was 3.235.81 Ci/mol. Each point in Panels A, B. and C represents the results from a single dish of cells with the exception of those cells on lipid-dcficlrnt strum (0 cholesterol concentration) whew each point is the mean of triplicate determinations. Each point in Panels D and F represents the mean of duplicate dishes, while the results shown in Panel E are the mean of the following number of determinations: No. 2237 (12), No. 2301 (9). NO. 2241 (2). No. 1414 (2). LUS. lipid-deficient serum.

protein to skin fibroblasts from hypo- and hyperresponder squirrel monkeys on cholesterol synthesis are shown in Fig. 4. For these studies, cells were incubated for 24 h with culture medium containing lipid-deficient serum, and for an additional 24 h with low density lipoprotein at the cholesterol concentrations indicated. Cholesterol synthesis was measured by incubating the cells for 24 h with [1-14C] acetate. The influence of squirrel monkey low density lipoprotein on sterol synthesis by skin fibroblasts is shown in Fig. 4, (Panels A, B, and C). Even at the lowest concentration of low density lipoprotein cholesterol used (1.25 pg/ml), there

467

appeared to be maximum suppression of sterol synthesis to levels 20-40% of controls. There was no further decrease in the rate of sterol synthesis with increasing cholesterol concentrations up to 9 pg/ml of culture medium. The degree of suppression of cholesterol synthesis was similar for cells from hypoand hyperresponders. We have used the rate of fatty acid synthesis as an independent indication of cellular activity. Consequently, if the added lipoprotein was altering the rate of sterol synthesis simply as a result of a general reduction in cellular metabolism, we might expect to see a similar reduction in fatty acid synthesis. This is clearly not the case, however, as can be seen from the results in Fig. 4, Panel C. In contrast to the marked suppression of cholesterol synthesis with addition of low density lipoprotein, there was little consistent change in fatty acid synthesis upon addition of the lipoprotein. These results are consistent with those reported by others [ 231. Due to the relatively small size of squirrel monkeys, it was difficult to obtain

o---o

A

150

o----o

\

\

\

\

pg Fig.

5.

Effect

Cholesterol/mlCultureMedium of

non-lipoprotein

hyperresponding were

incubated

somes and

(2

:

(HP) with

1 molar

incubated

with

for

was

at a final

tions

with

represents

the

culture

monkeys. medium

were

cells

a further

24

for

24

h.

The

of

determination.

the

on All

stem1

cells

containing

added

concentration

exception

a single

cholesterol

squirrel

ratio)

incubated added

2301 HO 2237 HP 1414 HO

u

h.

to

the

The

specific of

0.34

[1-14C1

of

point

serum

lipid-deficient

serum

[ l-14Cl

80%

for

24

from

was

added,

then

value

substrate the

for

and

(HO)

and

the

cells

h. Cholesterol/lecithin cholesterol

acetate

hype-

confluency,

indicated

represents

control

swum.

fibroblasts

approx.

at the

substrate

the

Each

in skin to

medium

acetate

activity

lipiddeficient I,DS.

grown

lipid-deficient

culture

mM.

synthesis

were

mean 2237

and was

lipo-

concentrations, the

3.29

cells

were

Ci/mol

and

of duplicate (hyperresponder)

determinawhich

468

the needed quantities of lipoprotein; thus, we evaluated the results of the addition of low density lipoprotein obtained from a much larger non-human primate species, the rhesus monkey. Results are shown in Fig. 4, Panels D, E, and F. There was little difference between rhesus and squirrel monkey low density lipoprotein with respect to their ability to suppress cholesterol synt,hesis when added to skin fihroblasts from hypo- and hyperresponding squirrel monkeys at similar cholesterol concentrations. In addition, we observed no obvious toxic effects of the rhesus monkey low densit,y lipoproteiI1 on squirrel monkey cells. The fact that there was little influence of rhesus monkey low density lipoprotein on fatty acid synthesis is consistent with this conclusion (Fig. 4, Panel F). Addition of cholesterol as cholesterol/phospholipid liposomes also produced a suppression of cholesterol synthesis (Fig. 5). Again, this did not appear to be the result of a non-specific effect on general cellular metabolism as there was relatively little influence on fatty acid synthesis. We also wanted to determine if suppression of cholesterol synthesis was accompanied by a stimulation in cholesterol esterification, as has been described for a number of other cell lines. For these studies, cells were incubated for 24 h with culture medium containing lipid-deficient serum, then for 6 h with rhesus monkey low density li~~oprot,eins. Cholesterol esterifi~ation was measured after incubation of the cells for an additional 4 h with [ 1 -“C] oleate. As can be seen in Table 111, addition of the lipoprotein stimulated cholesterol esterification up to 4.8-fold. Although only one hype- and one hyperresponder cell line were tested, there appeared to be no defect in the ability of either of these cell lines to stimulate cholesterol csterification in response to low density lipoprot,ein.

TABLE

III

EI’F‘ECT OF FIBROBLASTS

LOW DENSITY FROM IIYPO-

LlPOPKOTI:INS ON CHOLESTl
F:STEKIl~‘ICATION RIONKI~YS

IN SKIN

All cells were grown to approx. 80% conflurncy in 100-mm dishes at which time culture medium containing lipid-deficient serum was added. Thr cells were incubated for 24 h, rhesus monkey low density lipoprotein was added in the indicated concentrations, and the cells wew incubated for 6 h. [ I-“%1 Oleate (5.81 Ciimol, 0.16 mM) was added as substrate, and the 1~11s were incubatrd for au additional 4 h. Each point represents the mean * S.K. of fire determinations. The valuw in parcnthrsw represent the relativc increase over thch lipid-deficient serum control values (indicated as a low density lipoprotein concentrdtion of 0 ,ug!ml). Cells from hvl,rrrrsr,(,~ltlt~rs aw rlt~si~natc~d Ill’ and ttrow from tlvl,rlr~,sr,o,lclc‘rs IfO. f’t~sults are cxrtresseti as rhttlrstery1[ I- “C14t.‘,tv lorlncri (nrl~r>l!lnrpr
0

1

5

10

0.23 1 0.02 (I .2(j)

-

2237

HP

0.18

+ 0.01

2237

HP

0.81

0.21

0.42 f 0.02 (2.:33) _

0.27

-

23t1t HO

1.73

’

3.84 ’ 0.34 (4.80) 3.00 ’ 0.61 (1.73)

469

Discussion The results of this study further substantiate the marked individuality of control of plasma cholesterol concentrations in squirrel monkeys and the significant genetic component of this control. An understanding of the mechanisms of hypo- and hyperresponsiveness to dietary cholesterol can have major significance to man since the known genetic mechanisms of hypercholesterolemia, such as familial hyper~holesterolemia, mixed, and polygenic hyperchoiesterolemia account for only a small percentage of the variability in plasma cholesterol concentration in the human population [ 1,24,25]. The present study was designed specifically to determine whether differences exist in cholesterol synthesis and esterification in skin fibroblasts from hypoand hyperresponding squirrel monkeys when the cells are incubated with low density lipoprotein_ Cholesterol synthesis and esterification are both metabolic processes, the control of which is tied directly to the functioning of the low density lipoprotein binding sites [3]. Thus, a failure of skin fibroblasts to respond to low concentrations of low density lipoprotein by reduction of the rate of cholesterol synthesis or stimulation in cholesterol esterification would suggest an abnormality or deficiency of low density lipoprotein receptors [3,26]. In these studies we used low density lipoprotein isolated from either normocholesterolemic squirrel monkeys or hypercholesterolemic rhesus monkeys. Both lipoprotein preparations influenced the metabolic parameters measured in a similar manner in cells from both hypo- and hyperresponsive animals. This was of considerable practical significance since large quantities of squirrel monkey lipoproteins are difficult to obtain. The fact that the plasma cholesterol response to dietary cholesterol has a major genetic component [lo], resulting in increased concentrations of low density lipoproteins in the blood [5,27], and an increase in the animal’s susceptibility to atherosclerosis [4], initially suggested to us that the metabolic defect might be similar to that of familial hypercholesterolemia in man. It is clear from the results of this study, however, that this is not the case. In all experiments, low density lipoprotein effectively suppressed cholesterol synthesis in cells from animals of both phenotypes. Maximum suppression of sterol synthesis was attained at low density lipoprotein cholesterol concentrations of 2-5 pg/ml and was similar for cells from both hypo- and hyperresponsive animals. These data, combined with the fact that fibroblasts from both hypoand hyperresponder animals have the same maximum percentage suppression of sterol synthesis with increasing low density lipoprotein concentrations up to 50 pg/ml (Fig. 4, Panel E), suggest that their plasma membrane receptors are binding the lipoprotein with equal affinity. If one assumes that these skin fibroblasts metabolize low density lipoproteins as described by Brown and Goldstein [3], for human skin fibroblasts, then hyperresponsiveness to dietary cholesterol in this species does not appear to result from a deficiency of low density Iipoprotein receptors. This assumption seems reasonable since exposure of skin fibroblasts from both hypo- and hyperresponder animals to low density lipoprotein resulted in both a stimulation of cholesterol esterification and a suppression of cholesterol synthesis. Both of these metabolic changes are specific

470

and appear not to be due to non-specific changes in cellular metabolism as seen by the lack of influence of the lipoprotein on fatty acid synthesis. There also appeared to be little difference in cells from hypo- and hyperresponding animals in the suppression of cholesterol synthesis when the presumed low density lipoprotein binding sites were circumvented using cholesterol-containing liposomes rather than lipoprotein. Similar results of suppression of sterol synthesis with non-lipoprotein cholesterol have been reported by others [ 281. We also considered the possibility that, in terms of their binding site characteristics, the hyperresponsive animals may resemble more closely the familial hypercholesterolemic receptor-defective state than the receptor-negative condition. It has been demonstrated, however, that fibroblasts from receptor-defective human beings [29] have a lo-fold higher than normal activity of the ratelimiting enzyme of sterol synthesis (3-hydroxy-3-methylglutaryl-CoA reductase; mevalonate: NADP’ oxidoreductase (CoA-acylating), EC 1.1.1.34) at concentrations of low density lipoprotein cholesterol as low as 50 pg/ml. As shown in Fig. 4, Panel E, suppression of sterol synthesis, a function shown to vary directly with the activity of 3-hydroxy-3-methylglutaryl-CoA reductase [ 3,301, is similar in fibroblasts from hypo- and hyperresponsive monkeys at all concentrations tested. This suggests that the hyperresponsive animals are not receptor defective. Thus, the results from this study suggest that the phenomenon of hyperresponsiveness to dietary cholesterol in squirrel monkeys is not the result of similar mechanisms to those described for familial hypercholesterolemia in man. Acknowledgements The authors gratefully acknowledge the technical assistance of Ms. Molly Leight and Ms. Grayce Greene, as well as the aid of Mrs. Brenda Warner and Mrs. Norma Lofland in the preparation of this manuscript. This work was supported by the National Heart, Lung, and Blood Institute (SCOR) grant HL-14164 and was in partial fulfillment for the master’s degree (L.S.G.) in Comparative and Experimental Pathology from the Bowman Gray School of Medicine of Wake Forest IJniversity. References 1

Brown.

2

Goldstein. Invest.

M.S. 52.

Brown,

4

Clarkson,

5

Eggen,

D.A.

Eggen.

D.A.

pp.

Goldstein.

J.L.

Schrott,

(1975)

H.D.,

Adv.

Hazard,

Int.

Med.

20,

273-296

Bierman,

W.R..

E.L.

and

Motulsky,

H.O.

(1971)

A.G.

(1973)

.J.

M.S.

and

T.B.,

Golstrin.

.J.L.

Lofland.

(1976)

.J. Lipid

(1976)

(1976)

H.B.,

in

Science

Bullock, Res.

17,

Primates

in

B.C.

191. and

150-154

Goodman,

Arch.

Pathol.

92.

37-45

663473 Medicine

(Goldsmith,

E.I.

and

Moor-.Jankowski,

J..

eds.),

Vol.

267-299

7

Parks,

8

Roberts.

9

Wagner,

10

Lofland,

11

.Jones,

.J.S.

(1976)

Master’s

D.C.K. W.D.

and and

H.B., D.C..

12

Middleton,

13

Rush,

14

Rudel,

Clin.

1544-1568

3

6

and .J.L..

R.L., L.L.,

Lofland. Leon. Lee,

C.E.

Clarkson,

Clarkson,

C.C.,

Thesis,

West,

T.B. T.B.,

H.B.,

Clarkson. L. and .J.A.,

Wake (1974) (1974)

St.

T.B..

Morris,

Proc.

Clair,

Clarkson, Turrell,

Forest Lipids R.W.

T.B.

Lofland, .J. (1971) M.D.

and

University 9,

495-497

Sot. and

and H.B.

St. and

Exp. Clair.

(1974)

Med.

N.D.M. K.W.

Prichard,

in Advances E‘elts,.J.M.

Biol.

Lehner,

145, (1972)

(1975) R.W.

1050-1057 .J. Lipid

.J. Food (1967)

in Automated Biochem..J.

Kes.

Sci.

i\rch.

40, Pathol.

Analysis. 139.

13,

89-95

39-47

2-7

Technicon

83,

145-153

9.

15

St.

16

Lowry.

O.H..

Rosenbrough,

17

Sperry,

W.M.

and Webb,

18

Schwenk.

19

Bligh,

20

Clair,

R.W.,

Smith,

B.P.

and

and

Goldstein,

Dyer,

J.L.,

Dana,

Can.

Faust,

Circ.

and Arch.

40,

R.J.

166-173

(1951)

J. Biol.

Chem.

193.

265-275

187.97-106 Biochem.

J. Biochem.

J.R..

Kes.

Randall,

Chem.

(1952)

(1959)

S.E.,

(1977)

A.L.

J. Biol.

N.T.

W.J.

L.L.

Farr.

M. (1950)

E. and Werthessen.

E.G.

Wood,

N.J.,

Beaudet.

Biophys.

Physiol. A.L.

40,

334

37.911-917

and

Brown,

M.S.

(1975)

J. Biol.

Chem.

250.

8487-8795 21

Assman,

G.,

22

Kayden,

H.J..

Brown,

23

Goldstein.

24

Brown,

.J.L.. M.S.

Dana, P.O.,

Kwiterovich.

Jr.,

Brown,

M.S.,

Faust,

Rudel.

L.L.

28

Brown,

29

Goldstein,

S.E.

Goldstein.

25

.J.. cd%),

and

L. and

and

26 27

G.B.

Hatam.

and

Vol. M.S.,

Mahley.

and

and

Lofland,

N.G.

Brown,

J.L.

D.S.

(1976)

Biochemistry Biochemistry

(1974)

Proc.

Goldstein,

H.B.

(1975) (1976)

M.S.

(1976)

Fredrickson,

J.R.

R.W.

Beratis,

Med.

and

Levy,

.J.L.

(1975)

in Primates

Proc.

Natl.

Genet. R.I.

14.

39964002

15,

521-528

Acad.

Sci.

U.S.

(1974)

J. Clin.

J. Clin.

Invest.

in Medicine

55,

Invest.

Goldstein,

53,

1237-1249

(Goldsmith,

E.I.

and

Moor-Jankowski,

9, PP. 224-266 Dana,

.J.L., J.L.

42884292

783-793

S.E.

Dana.

and S.E..

Goldstein,.J.L. Brunschede.

(1973) G.Y.

and

Proc. Brown,

Natl. MS.

Acad. (1975)

Sci.

U.S.

Prac.

70, Natl.

1092-1096 30

71.

1. 103-119

and

Brown,

M.S.

(1973)

Proc.

NatI.

Acad.

Sci.

U.S.

70.

2804-2808

2162-2166 Acad.

Sci.

U.S.

72,