Free cholesterol transfer from human lower-density lipoproteins (d < 1.063) to lipoprotein-deficient serum and high-density lipoproteins

Free Cholesterol Transfer From Human Lower-Density Lipoproteins (d < 1.063) to Lipoprotein-Deficient Serum and High-Density Lipoproteins Esther Velazquez,

Agustin

Montes,

and Juan Miguel Ruiz-Albusac

The in vitro transfer of free cholesterol (FC) between human serum lipoproteins in the absence of lecithin:cholesterol acyltransferase (LCAT) activity has been examined. The results show that the amount of FC that the high-density lipoprotein (HDL) and lipoprotein-deficient serum (LDS) fractions were able to capture from low-density lipoproteins (LDL) and very-low-density lipoproteins (VLDL) was proportional to the amount of FC present in d < 1.063 lipoproteins. The presence of HDL increased this transfer markedly. These results indicate that, in the absence of LCAT activity, FC can transfer from lower-density lipoproteins to higher-density serum fractions, and this transfer might increase under hypercholesterolemic conditions. The possible importance of this phenomena in regard to the exchange of FC between serum lipoproteins and tissue cells is discussed. @ 1990 by W. B. Saunders Company.

T

HE AQUEOUS SOLUBILITY of unesterified or free cholesterol (FC), although small, is sufficient for simple Fick diffusion to account for a major part of the observed flux between different lipoprotein particles or between lipoproteins and cell membranes.‘,* According to these observations, it is generally assumed that the transfer of FC between lipoproteins is a spontaneous process that is not determined by any plasma factor, and that the net movement of cholesterol depends mainly on the chemica1 potential of cholesterol in the donor and recipient surfaces3 Therefore, the net transfer of FC from very-low-density lipoproteins (VLDL) and low-density lipoproteins (LDL) to high-density lipoproteins (HDL) is favored by the circulating enzyme 1ecithin:cholesterol acyltransferase (LCAT), which decreases the concentration of FC in HDL surface.4 However, circulating HDL is a mixture of heterogeneous particles in terms of lipid and apolipoprotein composition,’ and modifying the apolipoprotein composition can change the lipid transporting capacity of the particles.6 The present study gives evidence of the transfer of FC from d < 1.063 lipoproteins to HDL and lipoprotein-deficient serum (LDS) in the absence of LCAT activity. The study of exchange or transfer among plasma components has always been handicapped by the fact that during the time taken in ultracentrifugation, the FC became equilibrated among lipoprotein fractions.7 Accordingly, in our experiments we studied the transfer of FC from VLDL and LDL to HDL and LDS, separating the donor and acceptor paticles by precipitation of the lower-density lipoproteins with heparin and MnCl,.8 MATERIALS AND METHODS Isolation

of Lipoprotein

Fractions

VLDL + LDL(d < 1.063g/mL),HDL(1.063 1.tl g/mL) were isolated from normal human serum by sequential ultracentrifugal flotation in potassium bromide.’ Lipoprotein fractions and LDS were dialyzed extensively against 50 mmol/L Tris, 150 mmol/L NaCI, 0.01% EDTA, pH 7.4 buffer, with 0.02% NaN,. Lipoprotein purity was assessed by electrophoresis on cellulose acetate,” and the fractions were stored at 4OC with thymerosal(O.1 mg/mL)” under N, atmosphere until use. Lipoprotein

Labeling

VLDL + LDL were labeled with (I-2(n))-‘H-cholesterol (Amersham, Buckinghamshire, England; specific activity, 44 Ci/mmol). The radioactive cholesterol was purified by thin-layer chromatograMetabolism, Vol39, No 12 (December), 1990: pp 1263-l 266

phy (TLC) and added to the lipoprotein fraction dissolved in acetone; after this, the solvent was evaporated under N,.” The mixture was incubated at 37“C for 4 hours. Before labeling, the lipoproteins were preincubated with 5,5’-dithiobis(2nitrobenzoic acid) (DTNB) (0.28 mmol/L) for 4 hours to inhibit LCAT activity.” At the end of labeling, more than 98% of the incorporated radioactivity was in the form of FC, as checked by TLC. Measurement

of Cholesterol

and Protein

The concentrations of FC and total cholesterol in the lipoprotein fractions were determined by a commercial enzymatic method (Wako Pure Chemicals, Osaka, Japan).14 To measure free or total cholesterol after heparin-MnCl, precipitation, the supernatant was extracted in hexane, an aliquot was evaporated under N, and the enzymatic reagent added. Protein was measured by the Lowry method.15 FC Transfer Labeled or unlabeled VLDL + LDL were incubated at 37°C in Tris buffer pH 7.4, with bovine serum albumin (BSA) (Fraction V, Sigma Chemical, St Louis, MO) or LDS at various protein concentrations, with or without unlabeled HDL, in the presence of DTNB (final volume, 1mL). At time zero and intervals thereafter, the tubes were removed from the incubation bath, chilled, and 0.1 mL of heparin-manganese reagent added (final concentrations, 0.08% heparin and 24 mmol/L MnCl,).* After mixing, the tubes were maintained at room temperature for 20 minutes and then centrifuged at 2,700 rpm for 20 minutes. When necessary, before the addition of the precipitating reagent, they were diluted with buffer to a final protein concentration of 2 g/dL. An aliquot of each supernatant was removed and extracted in hexane to measure FC or total cholesterol. When radioactive samples were used, aliquots of the supernatant were counted with toluene-Triton X-100 scintillation fluid to determine cholesterol radioactivity. TLC showed that at the end of the incubation period greater than 99% of the radioactivity was in the form of FC.

From the Department of Biochemistry, Faculty of Medicine. Universidad Complutense. Madrid; and the Department of Physiology and Pharmacology, Universidad de Alcalb de Henares. Madrid, Spain. Address reprint requests to Juan Miguel Ruiz-Albusac, PhD, Department of Biochemistry, Faculty of Medicine, Universidad Complutense. Madrid, Spain. a 1990 by W.B. Saunders Company. 00260495/90/3912-0008$03.00/0 1263

VELAZQUEZ, MONTES, AND RUIZ-ALBUSAC

1264

RESULTS

When whole fresh human serum diluted to a protein concentration of 2 g/dL was incubated at 37“C with DTNB for 4 hours, and the lower-density lipoproteins were precipitated with heparin and MnCl, (Table l), it was found that the FC in the supernatant increased from 37.9 to 72.5 pg. No changes in the amount of esterified cholesterol in the supernatant were detected during incubation (98.0 + 5.2 at t = 0 v 96.8 + 4.1 pg at t = 4 hours). The excess of FC found in the supernatant represented 27% of the FC present in serum VLDL + LDL. This seems to indicate that, in the absence of LCAT activity, FC was transferred from the VLDL + LDL to the nonprecipitable fraction of serum. To confirm this point, VLDL + LDL (isolated by ultracentifugation at d < 1.063) was added to whole serum and incubated as before (Table 1). The amount of FC recovered in the supernatant was proportional to the amount of FC present in the form of VLDL + LDL at all the concentrations assayed. To know whether (VLDL + LDL) - FC was taken up by the HDL or by the LDS, we incubated isolated VLDL + LDL with or without HDL in the presence of LDS or BSA (2 g/dL of protein) (Fig 1). When VLDL + LDL was mixed with BSA, the FC in the supernatant did not change with incubation time, but it increased in the presence of LDS; after 4 hours incubation, the FC present in the supernatant represented 9% of the FC originally in VLDL -t LDL. When HDL was added, the FC recovered in the supernatant again did not change with incubation with BSA, but it increased with LDS, up to 18% of the FC in VLDL + LDL at 4 hours. These results seem to indicate that when VLDL + LDL was incubated with LDS, the FC was transferred to a heparinMnCl, nonprecipitable fraction, and the presence of HDL increased this effect. The FC present in the supernatant at zero time in the absence of HDL could be due to nonprecipitated VLDL + LDL.’ The marked increase in FC with HDL may be similar to the “initial burst” phenomenon described by Lange et alI6 between plasma and red blood cells. To show that FC exchange between VLDL + LDL and HDL was taking place during incubation, labeled VLDL + LDL with radioactive FC was incubated with HDL in the presence of BSA or LDS (Fig 2). After 4 hours incubation, isotopic equilibrium was reached, but while with BSA the transfer of radioactive cholesterol stopped at 7% (the expected value for a [VLDL + LDL]:HDL ratio of 12:1), the

Fig 1. Absolute changes in the amount of FC in the supernatant during incubation of VLDL + LDL with BSA or LDS in the presence or absence of HDL; d < 1.063 lipoprotein (273 Ag of FC) was incubated at 37°C in Tris buffer with BSA or LDS (final protein concentration 2 g/dL), in the absence or presence of HDL (22 Ag of FC). Aliquots were taken at different times and precipitated with MnCI, and heperin to measure FC in the supernatant. The values represent the absolute changes + SD in FC content (fig) of the supernatant in relation to the original content of HDL. Each point represents the mean of duplicates of four different experiments. Statistical comparisons are with 0 hours. lP < .Ol, l*P < .005, l**p< .OOl.

radioactivity transferred with LDS was 26%, corresponding to a net transfer of 19%. When 3H-(VLDL + LDL) was incubated with increasing amounts of LDS with or without HDL (Fig 3), the amount of radioactivity present in the supernatant was found to increase with the amount of LDS, but not with BSA, and it was higher at all protein concentrations when HDL was added. In b LDS

26 1

Table 1. Net Transfer of FC in Human Serum FC in Supsmatantfpg)

MDL + LDL Added (pg of FC)

0 Hours

4 Hours

Increaseof FC in Supsmatant PB

% of MDL + LDL

0

37.9

k 3.4

72.5

f 6.7

34.60

134

39.0

k 3.3

110.9

k 8.5

71.90

27.65

261

40.5

k 3.8

147.0

f

106.50

26.96

13.1

27.46

NOTE. Fresh human serum was diluted to a protein concentration of 2 g/dL (total FC, 16.4 k 0.4 mg/dL). One-milliliter aliquots of the diluted serum were incubated at 37’C without d < 1.063 heparin-manganese, (VLDL + LDL) -

for 4 hours in presence of DTNB, with or

lipoproteins (VLDL + LDL). After precipitation with the FC in the supernatant was measured. The

FC in 1 mL of the diluted serum was 126 + 3.8 pg.

Data are mean f SD of four different experiments in duplicate.

0

I TIME

OF

3 2 INCUBATION fh)

4

Fig 2. ‘H-cholesterol transfer from VLDL + LDL to the supernatant. Radiolabeled d < 1.063 lipoproteins (273 pg of FC) were incubated with HDL (22 pg of FC) in Tris buffer with BSA or LDS (protein concentration of 2 g/dLl at 37°C. Aliquots taken at different times were precipitated with heparin + MnCI, and the radioactivity in the supernatant measured. Each point represents the mean of duplicates of two diierent experiments.

1265

CHOLESTEROL TRANSFER IN PLASMA

LDS+HDL

LDS

BSA

I

2

4

6

8 PROTEIN

IO (g/dl)

Fig 3. Effect of the amount of LDS on the transfer of FC. Radiolabeled d < 1 .D63 lipoproteins (273 pg of FC) were incubated in Tris buffer with increasing amounts of ESA or LDS for 4 hours at 37°C. in the absence or presence of HDL (22 pg of FC), and the amount of radioactivity present in the supernatant after heparin + MnCI, precipitation was measured. The values represent net transfer of ‘H-cholesterol, ie, the measured value after subtracting the calculated amount of radioactivity transferred by exchange. Each point represents the mean of duplicates of two different experiments.

pig plasma was incubated for 24 hours at 37°C in the absence of LCAT, the absolute amount of FC decreased up to 25% in LDL, whereas it increased in HDL up to 42%. They also found that cholesteryl esters decreased in HDL between 4% and 9%. We also have found a small decrease, although not significant, in HDL cholesteryl esters. The results obtained after incubating isolated VLDL + LDL with HDL, show that the presence of LDS was necessary for FC transfer from the lower-density lipoproteins to HDL because no net FC transfer took place with albumin (Figs 1 and 2). In fact, by itself, LDS facilitated the transfer of FC to the nonprecipitable fraction. The capacity of LDS to accept cholesterol from cell membranes has been well documented.‘8.‘9 The mechanism whereby LDS induces the efflux of tissue cholesterol could be related to its content of apo A-I,*” apo A-IV,2’ or very-high-density lipoprotein. The role of apo A-I in the transfer of FC from cells has been stressed by Fielding and Fielding,** who suggest that the movement of cholesterol from cells into the plasma is greatly facilitated by apo A-I-containing lipoproteins. Recently, they have reported that a significant part of the cell-derived cholesterol is transferred specifically to a pre-@migrating lipoprotein A-I species and that this cholesterol subsequently appears in a particle of greater apparent molecular weight.” It is tempting to speculate that in the presence of LDS the FC from VLDL + LDL form some kind of recombinant with apo A-I similar to the one formed by cell cholesterol.

Table 2. Net Transfer Nonprecipitable

fact, at the highest concentration of LDS assayed (10 g/dL), up to 40% of the FC was transferred to the supernatant. In the following experiment, the effect of changing the concentrations of lipoproteins (VLDL + LDL and HDL) on the transfer of FC stimulated by 2 g/dL of LDS was studied with three different protocols: (1) ‘H-(VLDL + LDL) were incubated in increasing concentrations with a fixed amount of HDL; (2) a fixed amount of ‘H-(VLDL + LDL) was incubated with increasing concentrations of HDL; and (3) the concentration of both lipoproteins were increased to keep a fixed FC ratio of 12: 1. In all cases, isotopic equilibrium was reached after 4 hours’ incubation. At this time, the VLDL + LDL fraction was precipitated and the radioactivity in the supernatant measured. After subtracting the radioactivity transfer due to FC exchange, the net FC transfer was calculated. As is shown in Table 2, the net transfer of FC found at all lipoprotein ratios was a constant percent (- 20% to 22%) of the FC present in the donor particle (VLDL + LDL), and was not influenced by the amount of HDL present. DISCUSSION

Herein, we report that when fresh human serum was incubated at 37OC in the absence of LCAT activity, part of the FC of the lipoprotein fraction that was precipitable with heparin + MnCl, (VLDL + LDL) was transferred to the nonprecipitable fraction (HDL + LDS). These results agree with a recent report by Knipping et al,” who found that when

of FC From VLDL + LDL to the

Fraction at Different

Lipoprotein

Ratios in the

Presence of LDS Increment

of FC in the

Supmatant

After

4 Hours’

FC in Incubated

Incubation at 37°C

Lipoproteins (fig)

.__.

% 0‘ VLDL + LDL

HDL

Pg

IVLDL + LDLI -

FC

A. 100

32

20.70

20.7

200

32

44.09

22.0

300

32

70.46

23.5

400

32

95.15

23.8

500

32

120.60

21.9

B. 200

40

47.36

23.7

200

64

44.03

22.0

200

00

50.69

25.3

200

120

41.83

20.9

200

160

45.00

22.5

C. 100

8

18.99

19.0

200

16

44.47

22.2

300

24

62.64

20.9

400

32

95.22

23.8

500

40

119.00

23.8

NOTE. Labeled VLDL + LDL and unlabeled HDL were incubated with LDS (2 g/dL) at 37°C.

After precipitation with heparin-manganese

chloride, the radioactivity in the supernatant was measured. The data represent net transfer of FC (see text). Each value is the mean of triplicate incubations.

1266

VELAZQUEZ, MONTES, AND RUIZ-ALBUSAC

On the other hand, in our experiments, the transfer of FC to nonprecipitable fractions was markedly increased by the presence of HDL (Figs 1 and 3). The HDL fraction in human serum is heterogeneous, composed of several discrete subpopulations of particles with distinct density, size and composition.5,6 The interconversions of HDL particles of different apo A-I content can change the lipid transporting capacity of these particles.24 Moreover, an HDL conversion factor that modulates the particle size of HDL by converting a homogeneous population into new populations of particles, has been identified.25*26It is possible that these particles increase not only the reactivity with LCAT,” but also the FC transporting capacity of HDL. Another potentially important finding in our results is that the amount of FC transferred is always a constant percent of the FC content of VLDL + LDL at a fixed concentration of LDS (Tables 1 and 2). It has been described that the transfer of FC between lipoproteins is driven by the chemical potential of this molecule in both lipoproteins.3 Under our in vitro

conditions, the transfer of FC from VLDL + LDL to HDL could be the result of the increase in the capacity of HDL to accept FC from VLDL + LDL caused by the presence of LDS; this transfer could be going on until equilibrium between the chemical potential of FC between donor and acceptor particles is reached; this happens when VLDL + LDL lose a fixed percent of their FC for each amount of LDS (Table 2, Fig 3). Consequently, the flux of FC from VLDL and LDL to HDL should rise under hypercholesterolemic situations, not only because of the different FC/phospholipid ratio as stated by Fielding,’ but also by the amount of the lower-density lipoprotein pool. As the cholesterol fluxes from cell membranes and plasma are a function of the FC potential between both structures,2.3 these situations may decrease the capacity of plasma or HDL to accept cholesterol from cell membranes, or even the direction of the transfer of FC from cell membranes to plasma could be reversed, thereby facilitating the deposition of FC in the tissues.

REFERENCES

1. Lund-Katz S, Hammerschlag B, Phillips MC: Kinetics and

mechanism of free cholesterol exchange between human serum highand low-density lipoproteins. Biochemistry 21:2964-2969, 1982 2. Phillips MC, Jhonson WJ, Rothblat GH: Mechanism and consequences of cellular cholesterol exchange and transfer. Biochim Biophys Acta 906~223-276, 1987 3. Fielding CJ: The origin and properties of free cholesterol potential gradients in plasma, and their relation to atherogenesis. J Lipid Res 25:1624-1628, 1984 4. Fielding PE, Fielding CJ, Have1 RJ, et al: Cholesterol net transport, esterification and transfer in human hyperlipidemic plasma. J Clin Invest 7 1:449-460, 1983 5. Eisenberg S: High density lipoprotein metabolism. J Lipid Res 25:1017-1058, 1984 6. Cheung MC, Segrest JP, Albers J, et al: Characterization of high density lipoprotein subspecies: Structural studies by single vertical spin ultracentrifugation and inmunoaffinity chromatography. J Lipid Res 28:913-929, 1987 7. Schwartz C, Vlahcevic ZR, Berman M, et al: Central role of high density lipoprotein in plasma free cholesterol metabolism. J Clin Invest 70:105-l 16, 1982 8. Ruiz-Albusac JM, Velazquez E, Montes A: Differential precipitation of isolated human plasma lipoproteins with heparin and manganese chloride. Clin Chem 34:240-243, 1988 9. Have1 RI, Eder HA, Bragdon JA: The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest 34:1345-1353,1955 10. Magnani HN, Howard AN: A quantitative method for blood lipoproteins using cellulose acetate electrophoresis. J Clin Path01 24:837-845, 1971 11. Hatch F, Lees R: Practical methods for plasma lipoprotein analysis. Adv Lipid Res 6: l-68, 1968 12. Chajek T, Fielding CL Isolation and characterization of human serum cholesteryl ester transfer protein. Proc Nat1 Acad Sci USA 75:3445-3449, 1978 13. Stokke KT, Norum KR: Determination of 1ecithin:cholesterol acyltransfer in human blood plasma. Stand J Clin Lab Invest 27~21-27, 1971

14. Allain CC, Poon LS, Chan CSG, et al: Enzymatic determination of total serum cholesterol. Clin Chem 20:470-475, 1974 15. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the Folin phenol reagent. J Biol Chem. 193:265-275,195l 16. Lange Y, Molinaro AL, Chantey TR, et al: On the mechanism of transfer of cholesterol between human erythrocytes and plasma. J Biol Chem 258:6920-6926, 1983 17. Knipping G, Birchbauer A, Steyrer E, et al: Action of lecithin-cholesterol acyltransferase on low-density lipoproteins in native pig plasma. Biochemistry 26:7945-7953, 1987 18. Rothblat GH, Barriberger M, Phillips MC: Reverse Cholesterol Transport, in Albcrs JJ, Segrest JP (eds): Methods in Enzymology, ~01129. San Diego, CA, Academic, 1986, pp 628-664 19. Kano M, Koiumi J, Jadhav A, et al: Plasma exchange and low density lipoprotein apheresis in Watanabe heritable hyperlipidemic rabbits. Arteriosclerosis 7:256-261, 1987 20. Oram JF, Albers JJ, Cheung MC, et al: The effects of subfractions of high density lipoprotein on cholesterol efflux from cultured fibroblast. J Biol Chem 256:8348-8356, 1981 2 1. Ghiselli G, Krishnan S, Beige1 Y, et al: Plasma metabolism of apolipoprotein A-IV in humans. J Lipid Res 27:813-827, 1986 22. Fielding CJ, Fielding PE: Evidence for a lipoprotein carrier in human plasma catalyzing sterol efflux from cultured fibroblasts and its relationship to 1ecithin:cholesterol acyltransferase. Proc Nat1 AcadSciUSA78:3911-3914,198l 23. Castro GR, Fielding CJ: Early incorporation of cell-derived cholesterol into pre-beta-migrating high-density lipoprotein. Biochemistry 27~25-29, 1988 24. Nichols AV, Gong EL, Blanche PJ, et al: Pathways in the formation of human plasma high density lipoprotein subpopulations containing apolipoprotein A-I without apolipoprotein A-II. J Lipid Res 28:719-732, 1987 25. Rye A, Barter PJ: Changes in the size and density of human high-density lipoproteins promoted by a plasma-conversion factor. Biochim Biophys Acta 875:429-438, 1986 26. Barter PJ, Rajaram OV, Chang LBF, et al: Isolation of a high-density-lipoprotein conversion factor from human plasma. Biochem J 254:179-184, 1988