Reassembled Plasma Low Density Lipoproteins

THEJOURNAL OF BIOLOGICAL CHEMISTRK The American Society ofBiological Chemists, Inc.

Vol. 259, No. 10, Issue of May 25, pp. 6667-6673,1984 Pnnted in U.S.A.

(c)1984 by

Reassembled Plasma Low Density Lipoproteins PHOSPHOLIPID-CHOLESTEROLESTER-APOPROTEIN

B COMPLEXES*

1,

(Received for publication, October 31, 1983)

Geoffrey S. Ginsburg, Mary T. WalshS, Donald M. Small, and David Atkinsong From the Biophysics Institute, Departments of Medicine and Biochemistry, Boston University School of Medicine, Boston, Massachusetts 02118

Grants HL26335 and HL07291. A preliminary report of this work was presented at the American Heart Association’s 55th Scientific Sessions, November 16, 1982. Dallas, TX. Thecosts of publication of this article were defrayed in part by the payment of page charges. This articlemustthereforehe hereby marked“advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Samuel A. Levine Fellow from the American Heart Association, Greater Boston, Massachusetts Division (13-437-812) from 19821983. 5 Established Investigator of the American Heart Association. ’ The abbreviations used are: LDL, low density lipoprotein; apoB, apoprotein B; NaDC, sodium deoxycholate; standard buffer, 0.05 M sodium carbonate, 0.05 M sodium chloride, 0.02% sodium azide, pH 10; CO, cholesteryl oleate; EYPC, eggyolk phosphatidylcholine; DMPC, dimyristoyl phosphatidylcholine; PC, phosphatidylcholine; NaDodSO1, sodium dodecyl sulfate; AH, enthalpy of transition.

EXPERIMENTALPROCEDURES

Materials

All chemicals were standard reagent grade unless otherwise indicated. Sodium [’4C]deoxycholatewas purchased from Amersham (ArlingtonHeights,IL).SepharoseCL-4B and all chromatographic materials and columns were products of Pharmacia Fine Chemicals (Uppsala,Sweden). High molecular weight protein standards and reagents for polyacrylamide gel electrophoresis were electrophoresis purity grade and purchased from Bio-Rad (Richmond, CA). Methods Lipoprotein Isolation-Plasma was obtained from freshly drawn blood from normal human volunteers after an overnight fast. 0.01% disodium EDTA and 0.02% sodium azide were added to plasma. LDL

6667

Downloaded from www.jbc.org by guest, on July 14, 2011

Reassembled low densitylipoprotein(LDL) com- with a- model for the lipoprotein as being a microemulsion, plexes have been prepared by the interactionof lipid- -220 A in diameter, consistingof a neutral lipid core (cholesfree sodium deoxycholate-solubilizedapoprotein B terol ester (45 weight %), triglycerides (3 weight %)) surface(apoB) of native human LDL with preformed, 200 A stabilized by a monolayer of polar phospholipids (22 weight in diameter,microemulsions of cholesteryl oleate(CO), %) and cholesterol (10 weight %) and a single protein (apoB surface-stabilized by either eggyolkphosphatidyl(20 weight %)). The core-located lipids have been shown to choline(EYPC) or dimyristoylphosphatidylcholine undergo a liquid crystal4iquid transition nearbody temper(DMPC). Gel chromatography of PC/CO/apoB com- ature, the actual transition temperaturebeing determined by plexes shows co-elution of the complex at 43% PC, 43% CO, and 14% apoB. Negative stain electron microscopy both the diet-dependent fatty acyl chain composition of the shows the particles to be circular, homogeneous, and cholesterol esters in monkeys and the cholesterol ester/trimonkeys (1-3). Inanimal approximately 200 A in diameter. PC/CO/apoB com- glyceride ratioinhumansand plexes exhibit @-migration on agarose gels and show models, atherogenesis ispositively correlated with an ordered 3.0% so- physical state of the core lipids at body temperature (4-6). one high molecular weight protein band on dium dodecyl sulfate-polyacrylamide gels. Essential to understanding the atherogenicity of LDL is Differential scanning calorimetry and x-ray scatter-the elucidation of the preciselipid-lipid andlipid-protein ing show the lipids in the complexes to undergoa t least interactions which determinethe physical propertiesand two specific thermal transitions depending onlipid structure of theparticle, influence itsuptake by cellular composition, one associatedwith thecore-located cho- receptors, and direct its intracellular catabolism. To this end, lesterol esters similar to LDL and the protein-free microemulsions and the other from the phospholipid Kreiger et al. (7) have reported the reconstitutionof LDL and replacement of endogenous core lipids with retention of full forming the surface monolayer. In addition, particle disruption-protein unfolding/denaturation occur irre- biological activity (inuitro). However, to fully understand the role of each molecular component in LDL metabolism,a versibly at 80-85 “C. At 4 “C, the secondary structure of apoB on com- model system is needed in which not only can the cholesterol plexes of EYPC/CO/apoB is similar to that of native esters be systematically varied (with respect to degree of LDL. For complexes of DMPC/CO/apoB, the secondary unsaturation and chain length) but also the phospholipid and structure showsless a-helix which correlates with the apoB complement. difference in surfacelipid environment. We have recentlydescribeda method for the complete The reassembled complexes of PC/CO/apoB provide solubilization of apoB from native human LDL and the readefinedsysteminwhich the components may be varied systematically in order to study the molecular combination of apoB with phospholipid using sodium deoxycholate (8). We have also described the development of a organization, molecular interactions, and metabolism protein-free microemulsion system with the size and lipid of LDL. organization of LDL in which the phospholipid components and cholesterol ester components can be systematically varied (9). We now report the recombination of solubilized LDL Low density lipoprotein, the primarycholesterol transport apoB with phospholipid/cholesterol ester microemulsions to vehicle in theblood, has beenclearly implicated in the etiology yield reassembled LDL particles which exhibit many of the of atherosclerosis. Physical studies on LDL’ are consistent structural propertiesof native LDL. * This work was supported by United States Public Health Service

6668

Reassembled Plasma Low Density Lipoproteins cell was used. Protein concentrations ranged from 0.03-0.09 mg/ml. All samples which had been prepared in standardbuffer were dialyzed extensively against 0.005 M sodium tetraborate, pH 10, before recording CD spectra. All spectra reported are the average of four individual spectra on each of 4 to 5 different samples andhave been corrected for baseline contributions. Following calculation of the molar ellipticity, the percentage of n-helix was calculated and the percentage of random coil and &structure was estimated according to the methodof Greenfield & Fasman (18)utilizing the modified equation of Morrisett et al. (19). RESULTS

Preparation and Characterization of Microemulsion-ApoB Complexes-ApoB was interacted with the microemulsion particles at room temperature by placing 16 mg of either EYPC/CO or DMPC/COmicroemulsions into an open-ended dialysis bag (16 mg of total lipid, 8 mg of EYPC or DMPC plus 8 mg of CO in 2 ml of standard buffer)which was suspended in 6 liters of standard buffer. Two mg of NaDCsolubilized apoB (8) in a 2-ml volume was introduced to the microemulsions in the dialysis bag a t a very slow rate (0.5 ml/ h) by means of a peristaltic pump. Under these conditions, diffusion of NaDC across the dialysis membrane lowers the NaDC concentration well below its critical micellar concentration (20) upon entering the solutionallowing apoB tobind to the microemulsions. Initial experiments performed at pH 8.0 in 0.1 M KC1, 0.01 M Tris-HC1 buffer (the buffer utilized in the preparation of the microemulsions) were not successful since apoB did not bind to the microemulsions. Additional experiments in which the microemulsions were first dialyzed against 0.05 M sodium chloride, 0.05 M sodium carbonate a t pH 9.0 or at pH10 before the addition of NaDC-solubilized apoB, were performed. At these higher pH values, apoB interacted with the microemulsions. No disruption or degradation of the microemulsion was observed and the structural integrity of the large apoB molecule was maintained. The rate and extent of detergent removal was measured in an experiment utilizing [14C]sodiumdeoxycholate. Unlabeled DMPCand[3H]C0 were used inthepreparation of the microemulsions. Appearance of [I4C]NaDC in the dialysate confirmed the disassociation of the micellar detergent-protein complex while the solution in thedialysis bag remained optically clear,indicating that the protein had aggregated not and that the microemulsions were stable in the presence of small amounts of NaDC below its CMC. Two 6-liter changes of buffer were made over a 24-h period to complete the removal of NaDC. The amount of NaDC remaining in the dialysis bag, monitored using [14C]NaDC, was 3.3 pg of NaDC/mg of apoB. A typical gel chromatography elution profile of DMPC/ CO/apoB complexes prepared in this manner and eluted with standard buffer containing no detergent isshown in Fig. 1A. Phospholipid, cholesterol ester, and protein co-elute, indicative of a stable DMPC/CO/apoB complex. In the region of the elutionprofile where CO, DMPC, and apoBco-elute, both the CO/DMPC ratioat 1.05 mol/mol, which is comparable to the ratio in the starting microemulsions, and the apoB/DMPC ratio of 0.32 w/w are relatively constant. In a separate experiment utilizing unlabeled DMPC and ['HICO for the preparation of the microemulsions and [I4C]NaDC,[14C]NaDCwas below the limits of detection over the range of the elution profile where the above described ratios are constant. Fractions of a constant ratio were also pooled and concentrated to 1 mg of apoB/ml. [14C]NaDCwas also undetectable under these conditions. Negative stainelectron microscopy (Fig. 1B) shows the DMPC/CO/apoB comelex to havea circular morphology with a diameter of 219 k 24 A (mean S.D.) (Fig.IC) in agreement

*


was isolated by repetitive ultracentrifugation between salt densities of 1.025 and 1.050 g/ml by the addition of solid KBr (10). Isolated LDL was washedby ultracentrifugal flotation through an overlayering solution of d = 1.050 g/ml of KBr. All centrifugation was performed a t 55,000 rpm in a Beckman L8-70 ultracentrifuge in a 70 Ti rotor for 16 b at 4 "C. Purity of LDL from other lipoprotein fractions was verified by agarose electrophoresis (11) by staining with Oil red 0 and Coomassie brilliant blue R-250. Threepercent NaDodSO1polyacrylamide gels were used routinely to analyze preparations for degradation of apoB. Only LDLs showing one high molecular weight band a t M,= 366,000 (12) were used for this study. LDL Solubilization and Isolation of ApoB-LDL was dialyzed against 0.05 M sodium chloride and 0.05 M sodium carbonate, 0.02% sodiumazide, pH 10 (standard buffer).Based on a modified LDL solubilization method of Helenius & Simons (13), disruptionof LDL and solubilization of its molecular constituents were achieved with sodium deoxycholate as previously described (8). Following incubation, fractionation was carried out by gel filtration chromatography on Sepharose CL-4B withan elution buffer containing sodium deoxycholate. ApoB was recovered in -75% yield in a completely delipidated form. Protein-containing column fractions were pooled and concentrated to 2 mg of protein/ml by ultrafiltration and stored at 4 "C. Microemulsion Preparation and Isolation-Microemulsions were prepared by dispersing ["C]cholesteryl oleate and ['HH]phospholipid (egg yolk phosphatidylcholine or dimyristoyl phosphatidylcholine) at a 1:l molar ratio in0.1 M KCI, 0.01 M Tris-HCI, pH 8.0, and sonicating the mixture for 300 min under a nitrogen atmosphere. The internal temperature of the sonication vessel was kept at 51 "C (the crystal melting temperature of cholesteryl oleate). After sonication, the solution was fractionated by ultracentrifugation as previously described (9). AnalyticalMethods-Proteinwas quantitated using the Lowry protein method (14) utilizing the modified procedure of Markwell et al. (15). Lipid content was determined by liquid scintillation counting (9)orphosphorusassay(16). Gel filtrationchromatography was performed as described previously (8, 9). Electron Microscopy-Complexes were negatively stained with 2% sodium phosphotungstate, pH 7.4, on Formvar-coated copper grids. Electron micrographs were obtained with a Hitachi HU-11C electron microscope, calibrated with a grating replica (Pelco, Tustin, CA) at a magnification of approximately X 100,000. Gel Electrophoresis-NaDodS04-polyacrylamide gel electrophoresis was performed according to the method of Weber & Osborn (17) on 3.0 and 7.5% gels. Agarose gel electrophoresis on purified LDL and the reassembledcomplexeswasperformed by the method of Noble (11) with detection by staining with Coomassie brilliant blue R-250 and Oil red 0. Differential Scanning Calorimetry-These measurements were made in a Perkin-ElmerDSC-2(Perkin-Elmer, Norwalk, CT) at heating andcooling rates of 5 "C/min. Samples were concentrated by vacuum dialysis either immediately aftercentrifugation(LDL, EYPC/CO, DMPC/CO) or after pooling the appropriate fractions from the gel filtration column (EYPC/CO/apoB; DMPC/CO/apoB) and hermeticallysealed in 75-gl sample pans. The reference pan contained an equal mass of standard buffer. Sample masses were determined either by Lowry protein, liquid scintillation counting, or phosphorus assay (16). Enthalpies of transition (AH)were calculated from the areas under the peaks as measured by planimetry and related to the area of the crystal-liquid melting transition of an indium standard ( A H = 6.80 cal/g). X-ray Scattering-The x-ray scatteringcurves were recorded using a modified Luzzati-Baro x-ray scattering camerawith single mirror focusing collimation and used CuKa (X = 1.5418 A) radiation from a Jarrell Ash microfocus x-ray generator. Data were recorded with a Tennelec PSD 1100 (Tennelec, Oak Ridge, T N ) position-sensitive detector anda computer-based (Tracor Northern TN1710, WI) analysis system. Circular Dichroism-CD spectra of native LDL, EYPC/CO/apoB, and DMPC/CO/apoB complexes were recorded on a Cary 61 spectropolarimeter (courtesyof Dr. E. Simons, Dept.of Biochemistry, Boston University School of Medicine) calibrated withd-10-camphorsulfonic acid. The sample temperaturewas maintained by circulating ethylene glycol/water through the lamp compartment by means of a thermostatted refrigerator-heater bath. Temperature was measured to within 0.1 "C by means of a copper Constantan thermocouple positioned in contact with the CD cell in the lamp compartment. A 1-cm quartz

6669

Reassembled Plasma Law Density Lipoproteins

8

8

160 0

-:

140 120

A S

A

f

C

100

.4 c

.3

.60

7

40

.2

220

.1

€

EFFLUEHT

160 180

200

220 240

MAMETER

260 280

x

FIG.1. Gel chromatography and electron microscopy of L)MPC/CO/apoB complexes. A , gel filtration chromatography of DMPC/CO/apoR. The contentsof the dialysis bag were concentrated to2 ml by ultrafiltration and applied to a column (2.5 X 40 cm) of Sepharose CL-4B which was eluted downward with standard buffer. 0, protein, pg/ml; W, DMPC, pg/ml; A, CO,pg/ml; 0, CO/DMPC, mol/mol; 0, apoB/DMPC, w/w. R, electron microscopy of DMPC/CO/apoB from a fraction in the centralregion of co-eluting DMPC,-CO, and apoB (seeA ) negative staining was carried out using sodium phosphotungstate, pH 7.4, Ear = 500 A. C, size distribution histogram of DMPC/CO/apoB. Diameters of 200 randomly chosen circular particles were measured from representative fields of the electron micrographs. Average values were 219 f 24 A. with the diameter calculated from the partial specific volumes LDL and apoB of both DMPC/CO/apoB and EYPC/CO/ apoB complexes migrate identically when they were mixed (21, 22) and mass fractions of the protein and lipid compoandnents. co-electrophoresed same the on gel (Fig. 2, E and F). As shownin Fig. 2,3.0%NaDodS0,-polyacrylamide gels of NaDodS0,-polyacrylamide gels (7.5%) of similar column LDL (Fig. 2A), LDL dialyzed against standard buffer for 36 fractions performed up to 2 weeks after complex formation h as for the complexes (Fig. 2B), and both DMPC/CO/apoB show no degradationof apoB (i.e. all the protein that applied is gel remains at the top of the gel with no smaller and EYPC/CO/apoB complexes (Fig. 2, C and D)showed one to the co-migrating high molecular weight band (8).ApoB of native molecular weight bands appearing over the time course of the


40 60 80 100 1 2 0 140160180

R f x
0.001 the means are significantly different. LDL and EYPC/CO/apoB at 50 "C, at levels of significance 50.2 the means are not significantly different.

shown by otherinvestigators (26, 27), apoB ina neutral detergent environment undergoes a pH-dependent aggregation which may be prevented by titration toa higher pH. The pK, of 10 mM NaDC is approximately 6 and precipitation of the insoluble acid commences at pH 6.9 (20). Thus, in order


10 "C exhibit four maxima at similarly larger angular spacings, consistent with a particle size slightly smaller than LDL (46, 23). These results indicate that the structureof LDL and DMPC/CO/apoB complexes are similar having a spherical morphology and a radially organized "smectic-like'' arrangement of the core lipids. Studies on the Secondary Structureof ApoB-The secondary structure of apoB in the microemulsion-apoB complexes was investigated by circular dichroic spectroscopy and compared to apoB of LDL. Fig. 4A shows CD spectra of LDL, complexes of EYPC/CO/apoB and DMPC/CO/apoB in the far UV region which were recorded at 4 "C, a temperature at which the fatty acyl chains of the surface lipids of both LDL and EYPC/CO/apoB are in a disordered state and the fatty acyl chains of the surface lipid of DMPC/CO/apoB are in a n ordered state. At 4 "C, the core cholesterol esters of all the particles are in a smectic-like ordered state. Fig. 4B shows CD spectra at 50 "C, a temperature at which all the lipids of all the particles areabove their order-disorder transitions. The CD spectrum of LDL at 4 "C (Fig.4A)is characterized by a high overall ellipticity with negative minima at 217 nm ( = -12.20) and 222 nm ( [elp2,= -11.99) suggesting large amounts of a-helix and @-structure. The spectrum of EYPC/ CO/apoB is characterized bynegative minima at 217 nm = -8.30) and 222 nm ([e],,, = -8.68) suggesting that these particles containless a-helix and @-structure than LDL. The spectrum of DMPC/CO/apoB is characterizedby a negative minimum a t 222 nm ([e],,, = -4.94) suggesting a moderate amount of a-helix with little @-structure. TheCDspectrum of LDLat 50 "C (Fig. 4B) isagain characterized by a high overall ellipticity with negative minima at 217 nm = -11.48) and 222 nm ([e],,, = -11.55) suggesting again both large amounts of a-helix and @-structure. The spectrum of EYPC/CO/apoB complexes is similar to that of LDL, with negative minima at 217 nm ([e],li= -8.70) and 222 nm = -11.79), suggesting thatthese particles contain a similar amount of a-helix as LDL. However, thelower value of the ellipticitya t 217 nm suggests that complexes of EYPC/CO/apoBhave less @-structurethan LDL. The spectrum of DMPC/CO/apoB complexes is characterized by avery low overall ellipticitywith anegative minimum at 228 nm ([O],s = -5.20),suggestinga small amount of a-helix and a large amount of "unordered structure. The possibility of a contribution to theCD spectrum from the lipids of EYPC/CO or DMPC/CO microemulsionwas examined by recording spectra of the EYPC/CO or DMPC/ CO microemulsions alone over the wavelength region 200 to 250 nm at the appropriate temperatures andat lipid concentrations identicalwith those used for themicroemulsion/apoB measurements. For both EYPC/CO and DMPC/CO microemulsions, no contribution to the CD spectrum isobserved at the concentrations of lipid used for the studyover the wavelength region examined a t these temperatures.

6672

Reassembled Plasma Low Density Lipoproteins

At the median of the molecular weight range,the protein complement therefore comprises 595,000 daltons of the particle weight. Molecular mass determinations were carriedout by Dr. John S. Parks, Bowman Gray School of Medicine (28, 29).

For either LDL or EYPC/CO/apoB at 4 and 50 "C (Fig. 4, A and B ) , the CD spectra exhibit their minima at identical wavelengths with the only variation being the magnitude of the minima. For both LDLand EYPC/CO/apoB, the physical state of both the surface and core lipids are the same at each temperature (i.e. at 4 "C, the surface lipids are disordered and the core lipids are smectic-like; at 50 "C, all the lipids are disordered). For DMPC/CO/apoB, on the other hand,the CD spectrum at 4 "C is very different from the spectrum at 50 "C. The minima are shifted and their magnitudes are different. At 4 "C, the surface lipid of DMPC/CO/apoB is ordered and the core is smectic-like; at 50 "C, all the lipids are disordered. These variations in the CD spectra and the secondary structure of apoB onDMPC/CO/apoB complexes may be a reflection of the physical state of the lipids and may suggest that apoB is responsive to the physical state of its lipid environment. Reassembled LDL complexes of microemulsions and apoB provide a well defined model system in which to study the molecular interactions and structural organization of LDL, including the lipid-lipid interactions in the particle core, the lipid-lipid and lipid-protein interactions which determine the surface organization and protein conformation, and the interactions between the core and surface components. Preliminary experiments" have shown the ability of microemulsion/ apoB complexes to compete with human "'I-LDL for binding to the apoB/E receptor of cultured human fibroblasts. These reassembled LDL complexes should serve as an important model to study the delivery of isotopically labeled lipids with differing physical properties to cells in order to investigate the metabolic complexity of intracellular LDL catabolism and its relationship to positive cholesterol balance and atherogenesis. Acknowledgments-We wish to thank Drs. R. W . Mahley and T. L. Innerarity for performing the receptor bindingcompetition studies, Dr. John S. Parks for performing the molecular weight determinations, Eileen Fonferko and Ronald P. Coreyfor expert technical assistance, and Anne M. Gibbons for her expertise in preparing this manuscript. REFERENCES 1. Deckelbaum, R. J., Shipley, G. G., Small, D. M., Lees, R. S. & George, P. K. (1975) Science 1 9 0 , 392-394 2. Deckelbaum, R. J., Shipley, G. G. & Small, D. M. (1977) J. Biol. Chem. 252, 744-754 3. Tall, A. R., Small, D. M., Atkinson, D. & Rudel, L. L. (1978) J . Clin. Invest. 62,1354-1363 4. Atkinson, D., Deckelbaum, R. J., Small, D. M. & Shipley, G. G. (1977) Proc. Natl. Acad. Sci. U. S. A. 74, 1042-1046 5. Atkinson, D., Tall, A. R., Small, D. M. & Mahley, R. W . (1978) Biochemistry 17,3930-3933 6. Atkinson, D., Small, D. M. & Shipley, G. G. (1980) Ann. N . Y. Acad. Sci. 348, 284-298 7. Krieger, M., McPhaul, M. J., Goldstein, J. L. & Brown, M. S. (1979) J. Biol. Chem. 254, 3845-3853 8. Walsh, M. T. & Atkinson, D. (1983) Biochemistry 22,3170-3178 9. Ginsburg, G. S., Small, D. M. & Atkinson, D. (1982) J . Biol. Chem. 257,8216-8227 10. Havel, R. J., Eder, H. A. & Bragdon, J. H. (1955) J. Clin. Inuest. 34, 1345-1353 11. Noble, R. P. (1968) J. Lipid Res. 9, 693-700 12. Kane, J. P., Hardman, D. A. & Paulus, H. E. (1980) Proc. Natl. Acad. Sci. U. S. A. 77,2465-2469 13. Helenius, A. & Simons, K. (1971) Biochemistry 10,2542-2547 14. Lowry, 0.H.,Rosebrough, N. J., Farr,A. L. & Randall, R. J. (1951) J. Biol. Chem. 1 9 3 , 265-275 15. Markwell, M. A. K., Haas, S. M., Bieber, L. L. & Tolbert, N. E. (1978) Anal. Biochem. 8 7 , 206-210

'M. T. Walsh, T. L. Innerarity, G. S. Ginsburg,R. W. Mahley, and D. Atkinson, unpublished observations.


to favor the pH requirements of NaDC and the above cited observations about pH-dependentaggregation of apoB, either pH 9 or 10 was used for the reassembly. The complex, once formed and isolated, may be dialyzed to pH 7.4 or 8 without any apparent changes in chemistry or physical properties. The co-elution of PC, CO, and apoB on gel filtration chromatography is indicative of a stable PC/CO/apoB complex of stoichiometry: 43% PC, 43% CO, and 14% apoB. Negative stain electron microscopy shows the particles to b? circular (spherical), homogeneous, and approximately 200 A in diameter. The complexes exhibit /?-migration on agarose gels and one high molecular weight band on 3.0%NaDodS0,polyacrylamide gel electrophoresis which co-migrates with apoB of LDL on similar gels. A range of 3.9-4.6 X IOfifor the apparent molecular weight of the complex was determined by agarose gel column chromatography at 4 "C (28) and high performance gel filtration chromatography (29) using an lZRIlabeled LDL molecular weight standard.Z Thus, thePC/CO/ apoB complexes are stable and exhibit many of the chemical, compositional, and morphological properties of native LDL. Differential scanning calorimetryand x-ray scattering show the lipids in the complexes to undergo at least two specific thermal transitions depending on composition, one associated with the core-located cholesterol esters ( A H E 0.3 cal/g of cholesterol ester), similar to LDL and the protein-free microemulsions, and the otherin complexes with DMPC, from the phospholipid forming the surface monolayer ( A H 2.6 cal/g of phospholipid). In addition, particle disruption-protein unfolding/denaturation occur irreversibly at 80-85 "C. For EYPC/CO/apoB complexes, the transitionof the corelocated cholesterol esters occurs with a peak temperature of 38 "C and an enthalpy of 0.3 cal/g ofCO. Similarly, for DMPC/CO/apoB complexes, the transition of the core-located cholesterol ester occurs at a peak temperature of 42 "C with an enthalpy of 0.3 cal/g of CO. In addition, the orderdisorder transition of the surface phospholipid occurs at a peak temperature of 22 "C and an enthalpy of 2.6 cal/g of DMPC. In both EYPC/CO/apoB and DMPC/CO/apoB complexes, the thermal transitions are similar to those observed in their precursor protein-free microemulsion particles; however, the transition temperatures and enthalpies are significantly lower than thecorresponding transitions in the microemulsions. The lower lipid transition temperatures and enthalpies suggest that apoB either directly or indirectly influences the physical properties and order-disorder transitions of both the core and surface components. Thus, these studiessuggest that direct interactionsbetween the core and surface componentsperhaps occur innative lipoproteins. The protein at thesurface of native lipoproteins is thus likely to play a role in the interactions between the core and surface while providing additional stability to the "microemulsion." As shown in Fig. 4C, the secondary structure of apoB of native LDL is the same at 4 and 50 "C (30). The secondary structure of apoB on EYPC/CO/apoB complexes, however, does vary in response to temperature. A t 4 "C, the percentage of a-helix (27.17 f 1.27) is significantly different than that at 50 "C (37.93 +- 2.70% a-helix). The secondary structure of apoB on complexes of DMPC/ COjapoB is different from that of apoB of LDL or EYPC/ CO/apoB complexes at either 4 "C (20.35 f 0.74% a-helix, 0 f 5% @-sheet)or 50 "C (9.78 & 2.32 a-helix, 11 f 5% /?-sheet).

Reassembled Plasma Low Lipoproteins Density 16. McClare, C. W. F. (1971) Anal. Biochem. 39, 527-530 17. Weber, K. & Osborn, M. (1969) J. Bioi. Chem. 244,4406-4412 25. 18. Greenfield, N. & Fasman, G. D. (1969) Biochemistry 8, 41084116 26. 19. Morrisett, J. D., David, J. A. K., Pownall, H. & Gotto, A. M. (1973) Biochemistry 1 2 , 1290-1299 27. 20. Small, D. M. (1971) in The Bile Acids-Chamistry, Physiology and Metabolism (Nair, P. P. & Kritchevsky, D., eds) Vol. 1, Ch. 28. 7, pp. 247-354, Plenum PublishingNew Co., York 21. Lecuyer, J. & Dervichian, D. G. (1969) J. Mol. Bioi. 45.39-57 29. 22. Rand, R. P. & Luzzati, V. (1968) Biophys. J. 8, 125-13730. 23. Ginsburg, G. S.& Atkinson, D. (1983) Biophys. J. 4 1 , 182a 24. Walsh, M. T., Ginsburg, G. S., Small, D. M. & Atkinson, D.31.

6673

(1982) Circulotion 66-11, 100/3970 Ginsburg, G. S.,Small, D. M. & Atkinson, D. (1982) Fed. PFOC. 4 1,4392a Watt, R. M. & Reynolds, J. A. (1980) Biochemistry 19, 15931598 Watt, R. M. & Reynolds, J. A. (1981) Biochemistry 20, 38973901 Rudel, L. L., Pitts, L. L., I1 & Nelson, C. A. (1977) J . Lipid Res. 18,211-222 Carroll, R. M. & Rudel, L. L. (1983) J. Lipid Res. 24,200-207 Schefler, W. C. (1978) Statisticsforthe Biological Sciences, 2nd Ed, pp. 76-74, Addison-Wesley Publishing Co., Reading, MA Scanu, A. (1966) J. Lipid Res. 7,295-306
