327

Atherosclerosis, 31 (1978) 327-333 @ Elsevier/North-Holland Scientific Publishers, Ltd.

CHOLESTEROL ESTER EXCHANGE AND LOW DENSITY LIPOPROTEINS FACTOR

BETWEEN HUMAN PLASMA HIGH MEDIATED BY A PLASMA PROTEIN

A. SNIDERMAN, B. TENG, C. VEZINA and Y.L. MARCEL Cardiovascular Montreal,

Research

affiliated

Unit, Royal

to the University

Victoria

Hospital

of Montreal,

and Clinical

Montreal,

Que.

Research

Institute

of

(Canada)

(Received 16 May, 1978) (Revised, received 4 August, 1978) (Accepted 7 August, 1978)

Although unesterified cholesterol and phospholipids exchange freely, a protein factor from the d > 1.25 g/ml plasma fraction was found to be necessary for cholesterol esters to transfer from HDL to LDL. This transfer was reversible, time-dependent and a function of the concentration of the d > 1.25 fraction, but independent of lecithin : cholesterol acyltransferase reaction. The transfer represented an equilibration of molecules, but no net mass transfer of cholesterol esters could be demonstrated from HDL to LDL.

Key words:

Cholesterol ester exchange - Cholesterol protein L Low density lipoprotein

metabolism

-Higher

density

lipo-

Introduction Rapid exchange of unesterified cholesterol and phospholipids between lipoproteins can readily be demonstrated during in vivo incubations [1,2]. Similar evidence for exchange of cholesterol ester was not obtained and it has been

This research was supported by grants from the Medical Research Council of Canada (MA-5480 and MT-4011) and from the Quebec Heart Foundation. Abbreviations: HDL. high density lipoprotein. LDL. low density lipoprotein. VLDL, very low density lipoproteins, LCAT, lecithirxcholesterol acyltransferase, DTNB. dithionitrobenzoic acid, EDTA, ethylene diamine tetracetate.

328

concluded that cholesterol ester and triglyceride, perhaps because of their hydrophobicity or location within the lipoprotein core, do not participate in this phenomenon [3]. At the same time, contradictory evidence has been presented: the work of Nichols et al. [4] and that of Akanuma and Glomset [5] indicated a net transfer of cholesterol ester between lipoproteins, probably in exchange for triglyceride. Zilversmit and colleagues [6] have recently shown no net transfer but exchange of cholesterol ester between LDL and VLDL from hypercholesterolemic rabbits. An important finding of this work was that such exchange occurred only in the presence of a protein found in the d > 1.25 infranate of normal rabbit plasma. VLDL in this case, however, was subsequently shown to be a chylomicron remnant. The present study reexamines the possibility of cholesterol ester exchange between normal human HDL, and LDL. Methods

(1) Isolation of lipoproteins Samples of fetal cord blood or blood from normolipemic adults were collected in tubes containing EDTA (1 mg/ml). Plasma was separated by centrifugation at 2,500 Xg X 30 min. From these samples, LDL and HDL were isolated by preparative ultracentrifugation [ 71. LDL was isolated between d 1.019-1.050 g/ml using a 50 Ti rotor in a Beckman L5-65 ultracentrifuge for 18 h ‘at 105,000 Xg. HDL3 (d 1.125-1.21) and HDL (d 1.065-1.21) were washed by an additional centrifugation at d 1.21 g/ml. After isolation, each lipoprotein fraction was dialysed against 0.9% saline EDTA at pH 7.4.

(2) La belling of lipoproteins The d 1.21-1.25 g/ml fraction, after dialysis against 10 mM phosphate buffer pH 7.4, 10 mM fl-mercaptoethanol, was used as a crude LCAT preparation. The [ 14C]cholesterol-albumin suspension was prepared as described by Stokke and Norum [ 81. A typical incubation mixture contained: 5 ml HDL or HDL3 (100 mg protein), 5 ml d 1.21--1.25, and 1.5 ml [ 14C]cholesterol in 5% albumin (12 X lo6 cpm), which was incubated at 37°C for 20 h. The density was then adjusted to 1.21 g/ml and HDL or HDL3 reisolated as previously described. LDL cholesterol was labelled by a similar technique; in which the HDL3 was replaced by the fraction d 1.019--1.21 g/ml. The partition of the [ 14C]cholesterol label between unesterified and esterified cholesterol was determined by extraction of the labelled lipoprotein with chloroform : methanol. Unesterified and esterified cholesterol were then separated on polysilicic acid impregnated glass fiber sheets (Gelman Instrument Co., Ann Arbor, MI) using hexane : diethyl ether : glacial acetic acid (45 : 5 : 0.5). Radioactivity in the two cholesterol fractions was then determined by liquid scintillation counting in a toluene-based scintillation mixture. A representative HDL obtained by this procedure had specific activities of 210 and 380 cpm/pg cholesterol in the cholesterol ester and unesterified cholesterol fractions, respectively. The mean percentage of labelled cholesterol esterified in the HDL preparations was 74 f 17%, with a range of 59-93.

329

(3) Incubation Either cord blood or adult LDL was used for the incubations. Except as noted, the results are those using cord blood LDL. Typically, in a final volume of 1 ml, LDL (30-50 pg total cholesterol) was incubated with HDL or HDL, (30-50 pg total cholesterol) in the presence of varied amounts of d > 1.25 g/ml fraction for up to 6 h at 37°C. Following this, LDL was precipitated by addition of heparin-manganese chloride [9]. The supemate was removed by Pasteur pipette and the precipitate washed twice with distilled water. The precipitate was then redissolved in saline d = 1.063 g/ml and LDL recovered as the d = 1.063 supernate by preparative ultracentrifugation as previously described. Recovery was determined by measurement of LDL B protein by radial immunodiffusion [lo] before incubation and after re-isolation by ultracentrifugation. In all experiments at least 94% of LDL was recovered. Unesterified and esterified cholesterol fractions were separated by preparative thin layer chromatography on silica gel G with hexane : diethyl ether : acetic acid (90 : 10 : 1) as a developing solvent. Cholesterol concentration was measured by gas-liquid chromatography after appropriate hydrolysis [ll] or by colorimetric reaction [ 121 after saponification [ 131. Results The initial experiments determined the conditions necessary for isotopic cholesterol ester transfer from HDL3 to LDL. When labelled HDL3 (50 pg total cholesterol) were incubated with LDL (50 pg total cholesterol) in the presence of the fraction d > 1.25 g/ml (9 mg protein) more than 10% of the labelled cholesterol ester was recovered in the LDL within 4 h (Fig. 1, right panel). When the fraction d > 1.25 g/ml was omitted from the incubation mixture, no transfer of labelled cholesterol ester was observed (Fig. 1, left panel). The factor present in the d > 1.25 infranate heated at 60°C for 10 min was inactive in promoting the transfer of labelled cholesterol ester (Fig. 1, center panel). Similar results were obtained with either HDL3 or HDL, but most experiments were carried out with HDL+ The transfer of [ 14C]cholesterol ester from HDLJ to LDL was a function of time (Fig. 2) and occurred with either cord blood or normal adult LDL matched for cholesterol ester concentration (Fig. 2). The relative amount of cholesterol ester transferred from HDL3 to LDL was a function of the protein concentration of the d > 1.25 g/ml fraction which was added to the incubation mixture (Fig. 3). Isotopic transfer does not distinguish between net transfer, i.e. absolute change in mass, and an exchange reaction. In order to define the nature of the observed process, the absolute concentration of unesterified and esterified cholesterol were measured in both HDL3 and LDL, before and after a 4 h incubation (Table 1). The absolute concentration of cholesterol esters in HDL and LDL were unchanged during incubation in the presence of the d > 1.25 g/ml fraction measured either by gas-liquid chromatography or calorimetry. In contrast, there was a net decrease in unesterified cholesterol in LDL accompanied by a similar increase in HDL, indicating a net mass transfer of this molecule between the two lipoprotein fractions. The percentage of cholesterol esterified

330 HDL3 LDL

12

HDL3 LDL d >I.25 I (60’~

NaCl

+-

HDL3 LDL d>1.251

IOmmn~

I

1

0

0

4

TIME (h

)

Fig. 1. Only when d > 1.25 infranate (right panel) is added to the incubation does radioactive transfer of cholesterol ester occur. If omitted (left panel) or heated (middle) then transfer is absent. The vertical bars represent the standard deviations (n = 5).

in the total incubation mixture (HDL + LDL) increased by 2.5% in experiment A, and 1.5% in experiment B. These increments are well within the experimental error and are probably not related to residual LCAT activity. Finally, the corollary experiment, that is, the transfer of 14C-labelled cholesterol ester from LDL to HDL3 demonstrated that cholesterol esters equilibrate reversibly between these lipoproteins but only in the presence of the d > 1.25 fraction (Fig. 4). Although, the HDL fractions used were washed, at d 1.21 g/ml, a possible contamination with LCAT remained which could conceivably influence the

3

??

0

Adult LDL Cord blood

LDL

“0

x2 5 u I

0

4

2 Tame

(h

I

6

2

1

Fig. 2. The amount of labelled cholesterol time for both adult and cord blood LDL.

10

4

d> ester transferred

from

Fig. 3. The amount of HDL-labelled cholesterol ester transferred amount of d > 1.25 infranate protein added to the incubation.

HDL3

l-25

to LDL

to LDL

16

mg

20

protein

increases

is a direct

linearly

function

with

of the

331 TABLE

1

UNESTERIFIED AND ESTERIFIED CHOLESTEROL IN LDL CUBATION IN THE PRESENCE OF d = 1.25 INFRANATE

AND

HDL

BEFORE

AND

AFTER

IN-

A and B are two separate experiments, each represents the mean + SD of 6 determinations where cholesterol was measured by gas-liquid chromatography in A and by the colorimetrlc method in B. Lipoprotein

analysed HDL3 (inc. time 0 h)

LDL (inc. time 0 h)

A. UC

a (wz)

41 f 7.2 92.9 t 19.2

cEb B. UC (pg) CE

65 202

f 4.2 + 20.6

16.8 145.9

f f

3.3 9.5

51 205

f 4 r 10

HDL3 (inc. tie

LDL (inc. time 4 h) 28.5 98.6 35 195

4 h)

f 2.0 + 12

32.3 148.6

+ 5.5 f 5.9

t 6.1 +_ 6.7

70 208

+3 +9

a UC = unesterified cholesterol. b CE = cholesterol ester.

exchange process. To eliminate this possibility, the exchange experiments between HDL3 and LDL described above were duplicated in the presence of DTNB (1.2 mM). At this concentration of DTNB, LCAT activity is inhibited by 95% [8], but the lack of effect of this inhibitor demonstrate8 that the enzyme does not play a direct role in the exchange of cholesterol esters (Fig. 5). Partial purification of the protein factor was obtained first by precipitation at 42% saturation of ammonium sulfate and second by isoelectric focusing on

d>1.251

NaCI

I__jllln I_: 1

I

0

4

0 TIME

4

(h j

Fig. 4. Labelled LDL is incubated with unlabelled HDL3 on the left in the presence of d > 1.25 infranate; on the right in its absence. Transfer of radioactive cholesterol ester from LDL occurs only with d > 1.25 and the radioactivity appearing in HDL is balanced by that leaving LDL. The vertical bars represent the standard deviations (n = 5).

332

lncubotcon

tme

(h )

Fig. 5. Effect of LCAT inhibition on the transfer mM) has no significant effect on this transfer.

of cholesterol

ester from

HDL3

to LDL.

DTNB

(1.2

Sephadex G-75 according to Radola [ 141. The fractions were eluted with 0.5 M Tris-HCl pH 8.0, dialysed against 0.9% NaCl, and assayed for cholesterol ester transfer between HDL and LDL. The fractions focused between pH 6.1 to 6.4 promoted cholesterol ester transfer: 220 pg protein transferred as much labelled cholesterol ester (9%) as 9 mg of protein of d > 1.25, which represents a purification of about 40-fold. Discussion The results of the present study indicate that cholesterol ester can exchange between lipoproteins. Specifically, it has been shown that radioactive cholesterol ester transfers from HDL to LDL. That this process represents exchange rather than net transfer has been shown both by studies on reverse radioactive transfer and by direct determination of cholesterol ester mass before and after incubation. These results are not inconsistent with the previous observations responsible for the view that cholesterol ester was not exchangeable since the process is dependent on the presence of a protein in the d > 1.25 infranate. The present results differ from those of Nichols [4] who found a mass transfer of HDL cholesterol ester in return for VLDL or LDL triglyceride. However, a prolonged incubation time of up to 16 h was required raising concern about stability of the lipoproteins. No necessity for a transfer protein was demonstrated. Akanuma and Glomset [ 51 did not demonstrate mass transfer directly, but inferred such a process was necessary to account for differences in fatty acid specific activity of the cholesterol esters of HDL and VLDL. The present results do, however, conform closely to those reported for cholesterol ester exchange in hypercholesterolemic rabbit LDL and VLDL in the presence of a factor from normal rabbit plasma [6], which have now been extended to demonstrate cholesterol ester exchange between normal human lipoproteins. In our experiments, the plasma factor responsible for this transfer was found to be inactivated by heating at 60°C for 10 min, acondition commonly used for inactivation of LCAT [ 161. Such an inactivation could be an argument for the participation of LCAT in this transfer. However, we believe to have demonstrated satisfactorily that this is not the case. First, in the mass transfer experi-

333

ment (Table l), only minimal increases of cholesterol ester were noted (2.5 and 1.5%/4 h). Second, the addition of DTNB (1.2 mM) which inhibits LCAT activity by about 95% [8], did not affect the rate of exchange of cholesterol esters (Fig. 5). That the cholesterol ester exchange factor was not heat-labile in the rabbit plasma, while that of human plasma was heat-labile under similar conditions, remains to be clarified. For the present process to be exchange rather than net transfer does not deprive it of potential physiological significance. In man, the cholesterol esters of LDL and HDL are essentially similar in composition [ 151. However, it has been argued that plasma cholesterol esters arise by the action of LCAT on HDL [16]. Cholesterol ester exchange, rather than net transfer of cholesterol ester for triglyceride, would satisfactorily account for their similar composition. It must be emphasized that no net mass transfer of cholesterol esters could be demonstrated between HDL or HDL3 and LDL even in the most favourable conditions, that is: by incubation of an HDL3 rich in cholesterol esters and poor in unesterified cholesterol (HDL3 after incubation with LCAT) with an LDL poor in cholesterol esters and with low cholesterol ester to apo-B ratio (cord blood as compared to adult LDL). Under these conditions, there is a mass transfer of unesterified cholesterol for the stabilization of these lipoproteins, but there is no mass transfer of cholesterol esters in spite of their molecular equilibration. It necessarily follows that LDL cholesterol esters must be incorporated in this lipoprotein at the time of its synthesis or incorporated in its precursor the VLDL. Alternatively, since these in vitro experiments represent a simiplification of in vivo conditions, we may have adopted a model too limited to allow for cholesterol ester net transfer. References 1 Bell, F.P.. Transfer of cholesterol between serum lipoproteins, isolated membranes and intact tissues, EXP. Mol. Path., 19 (1973) 293. 2 Kunkel, H.G. and Bearn. A.G., Phospholipid studies of different lipoproteins employing P 32, Proc. Sot. EXP. Biol. Med.. 86 (1954) 887. 3 Jackson, R.L., Morrisett, J.D. and Gotto, A.M., Lipoprotein structure and metabolism, Physiol Rev., 56 (1976) 259. 4 Nichols, A.V. and Smith, L., Effect of very low density lipoproteins on lipid transfer in incubated serum. J. Lipid. Res., 6 (1965) 206. 5 Akanuma, Y. and Glomset, J., In vitro incorporation of cholesterol 14C into very low density lipoprotein cholesterol esters, J. Lipid Res., 9 (1968) 620. 6 Zilversmit. D.B., Hughes, L.B. and Balmer, J.. Stimulation of cholesterol ester exchange by lipoprotein-free rabbit plasma, Biochim. Biophys. Acta. 409 (1975) 393. 7 Havel, R.J., Eder, H.A. and Bragdon, J.H., The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum, J. Clin. Invest., 34 (1965) 1345. 8 Stokke, K.T. and Norum. K.R.. Determination of lecithin : cholesterol acyltransferase in human blood plasma, &and. J. Lab. Invest.. 27 (1971) 21. 9 Burstein, M., Scholnick. H.R. and Morfin, R., Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions, J. Lipid Res., 11 (1970) 583. of B-protein of low density lipoprotein 10 Sniderman. A.D., Teng. B. and Jerry, M., Determination directly in plasma, J. Lipid Res., 16 (1975) 465. 11 Marcel, Y.L. and Vezina, C.. A method for the determination of the initial rate of reaction of lecithin : cholesterol acyltransferase in human plasma. Biochim. Biophys. Acta, 306 (1973) 497. 12 Zak. B.. Moss, N., Boyle, A.J. and Zlutkis, A., Reactions of certain unsaturated steroids with acid iron reagent, Anal. Chem., 26 (1954) 767. 13 Abell, L.L., Levy, B.B., Brodie, B.B. and Kendall, F.E.. A simplified method for the estimation of total cholesterol in serum and determination of its specificity, J. Biol. Chem., 195 (1952) 357. 14 Radola, B.J., Analytical and preparative isoelectric focusing in gel stabilized layer. Ann. N.Y. Acad. Sci., 209 (1973) 127. 15 Goodman, D.S. and Shiratori, T., Fatty acid composition of human plasma IiPoDrotein fractions, J. Lipid Res., 5 (1964) 307. 16 Glomset. J.A.. The plasma 1ecithin:cholesterol acyltransferase reaction, J. Lipid Res.. 9 (1968) 155.

Cholesterol ester exchange between human plasma high and low density lipoproteins mediated by a plasma protein factor.

327 Atherosclerosis, 31 (1978) 327-333 @ Elsevier/North-Holland Scientific Publishers, Ltd. CHOLESTEROL ESTER EXCHANGE AND LOW DENSITY LIPOPROTEINS...
540KB Sizes 0 Downloads 0 Views