Apoproteins of human serum high density lipoproteins. Isolation and characterization of the peptides of Sephadex fraction V from normal subjects and patients with abeta-lipoproteinemia.

1. Sephadex fraction V, obtained from human serum high density lipoprotein apoprotein (HDL apoprotein) of normal subjects and of patients with abetalipoproteinemia, was resolved by DEAE-cellulose ion exchange column chromatography into several fractions which were defined in terms of amino acid composition, NH2- and COOH-terminsls, sialic acid content, immunologic and electrophoretic properties, and in vitro activation of purified lipoprotein lipase from rat adipose tissue. 2. Fraction V of HDL apoprotein of both normal and abetalipoproteinemic subjects was found to contain polypeptides corresponding to apolipoproteins C-I, C-II, C-III-1, and C-III-2, which had been described previously in very low-density lipoproteins (VLDL). The content of apo C-III-1 in abetalipoproteinemia-HDL was very low, whereas the percentage, by weight, of apo C-I was about twice as high as that in the normal subjects studied. Furthermore, both normal and abetalipoproteinemia-HDL apoprotein contained a previously unreported peptide which had a molecular weight of about 7 000 and electrophoretic, chemical, and immunological properties distinct from those of the known C apolipoproteins. Of all of the peptides comprising fraction V, only apo C-II activated a purified preparation of rat adipose tissue lipoprotein lipase. This was the case for both normal and abetalipoproteinemic subjects.     

The interaction of lipolysis products of very low density lipoprotein with plasma high density lipoprotein (HDL): perfusate HDL with plasma HDL subfractions

The nature of the interaction of high density lipoproteins (HDL), formed during lipolysis of human very low density lipoprotein (VLDL) by perfused rat heart, with subfractions of human plasma HDL was investigated. Perfusate HDL, containing apoliproproteins (apo) E, C-II, and C-III but no apo A-I or A-II, was incubated with a subfraction of HDL (HDL-A) containing apo A-I and A-II, but devoid of apo C-II, C-III, and E. The products of the incubation were resolved by heparin-Sepharose or hydroxylapatite chromatography under conditions which allowed the resolution of the initial HDL-A and perfusate HDL. The fractions were analyzed for apolipoprotein content and lipid composition and assessed for particle size by electron microscopy. Following the incubation, the apo-E-containing lipoproteins were distinct from perfusate HDL since they contained apo A-I as a major component and apo C-II and C-III in reduced proportions. However, the HDL-A fraction contained apo C-II and C-III as major constituents. Associated with these changes in apolipoprotein composition, the apo-E-rich lipoproteins acquired cholesteryl ester from the HDL-A fraction and lost phospholipid to the HDL-A fraction. The HDL-A fraction maintained a low unesterified cholesterol/phospholipid molar ratio (0.23), while the apo-E-containing lipoproteins possessed a high ratio (0.75) characteristic of the perfusate HDL.(ABSTRACT TRUNCATED AT 250 WORDS)     

Activation and inhibition of lipoprotein lipase. Studies with artificial lipoproteins.

Human plasma very low density apolipoproteins C-I, C-II and C-III were recombined in vitro with triolein. The lipid-protein complexes were analyzed by ultracentrifugal flotation, agarose gel electrophoresis, immunoelectrophoresis and electron microscopy. Maximal protein/triolein ratios for apoprotein C-I, C-II, C-III-1 and C-III-2 were 50, 45, 95 and 55 microgram/mg, respectively. Electron micrographs exhibited spherical particles with diameters ranging from 200--2000 A comparable to native VLDL and chylomicrons. On agarose gel electrophoresis these complexes showed alpha-mobility. Kinetics of triolein hydrolysis by purified human plasma lipoprotein lipase were studied using these artificial lipoprotein substrates with different apoprotein/triolein ratios. The reaction followed the Michaelis-Menten equation. With increasing amounts of apo C-II, the apparent Km decreased from 0.60 to 0.11 mM. Incubation of the substrate with either rabbit anti-apo C-II gamma-globulins or digestion with trypsin prior to hydrolysis reversed this lowering effect on apparent Km. V was not altered significantly. Increasing amounts of apo C-I, apo C-III-1 or apo C-III-2 without apo C-II caused inhibition of triolein hydrolysis. In the presence of apo C-II, however, similar kinetic parameters were obtained as described above.     

Analysis of rat serum apolipoproteins by isoelectric focusing. II. Studies on the low molecular weight subunits.

The low molecular weight proteins of rat apo HDL and apo VLDL have been isolated and analyzed by the technique of isoelectric focusing. Sephadex fractions from apo HDL (HS-3) and apo VLDL (VS-3) that contain these proteins reveal three major bands with apparent isoelectric points of pH 4.50, 4.67, and 4.74, as well as three minor bands at pH 4.43, 4.57, and 4.61. In addition, apo HDL has a major band at pI of 4.83. DEAE-Cellulose chromatography was used to prepare purified fractions of these components that were characterized by N-terminal analyses and molecular weight determinantions by SDS gel electrophoresis. The major low molecular weight components of apo HDL were focused on a slab gel and the bands were identified as A-II (pI 4.83), C-II (pI 4.74), C-III-0 (pI 4.67), and C-III-3 (pI 4.50). Neuraminidase treatment of apo HDL, followed by isoelectric focusing, suggested that the other bands, which have not previously been reported, may be additional forms of the C-III protein, differing only in their content of sialic acid.     

Determination of human apolipoprotein C-II by electroimmunoassay. Studies on standardization and determination before and after physical training

1. Human VLDL and HDL were fractionated by sequential ultracentrifugation until free of contaminant plasma proteins. 2. Column chromatofocusing method was used to isolate apolipoprotein C-II from apoVLDL and apo HDL. C-apoprotein peak was rechromatofocused and the second peak was the apo C-II (pI 4.7, homogeneous band on SDS slab gel). 3. New Zealand white rabbits were immunized with apo C-II. Antiserum gave a single precipitate are of identity between whole serum, apoVLDL, apoHDL and apo C-II. 4. Apo C-II concentration was measured by electroimmunoassay method. During standardization 1% Triton X-100 improved the rocket shapes and contours. Total delipidation did not affect the assay system and so the antigenic determinants of apo C-II are all available to antiserum. The lowest concentration of apo C-II possible to determine with this method was 70 ng/sample well. 5. There was no difference between the apo C-II values before (39.8 +/- 7.1 mg/l, n = 19) and after (41.6 +/- 6.4 mg/l, n = 19) moderate physical training among normolipemic subjects. 6. Specific immunoprecipitation technique was also used to determine apo C-II content in standard pool serum.     

Hydrolysis of chylomicron triacylglycerol by endothelium-bound lipoprotein lipase. Effect of decreased apoprotein C-II/C-III ratio.

Chylomicrons with a decreased ratio of C-II/C-III apoproteins on their surface produced by the addition of apoproteins C-III-0 or C-III-3 to intact rat lymph chylomicrons. These chylomicrons inhibited the activity of soluble lipoprotein lipase in vitro, but had no effect on the activity of the endothelium-bound enzyme in the perfused heart.     

Lipoprotein composition of human suction-blister interstitial fluid

Interstitial fluid (IF) was obtained in 27 apparently healthy subjects (12 males, 15 females) by applying mild suction (200-250 mm Hg) on the skin either on the midvolar forearm or on the paraumbilical region of the abdomen. The IF concentrations of lipids and apolipoproteins (apo) were studied and compared with those of serum (S). The mean ratio between interstitial fluid and serum (IF/S ratio) varied from 0.14 for forearm apoE to 0.29 for apoA-II on the abdomen. This ratio was consistently lower for apoE, C-II, C-III, and B than for apoA-I and A-II, and significantly lower on the arm than on the abdomen for all apolipoproteins studied. The IF/S ratios showed marked variations among individuals. However, interstitial fluid apolipoprotein concentrations at different blister sites were highly correlated within each individual. Studies with agarose gel electrophoresis and density gradient ultracentrifugation revealed that large triglyceride-rich particles were virtually lacking in interstitial fluid and that the relation between the low density lipoproteins (LDL) and high density lipoproteins (HDL) was shifted towards a greater proportion of HDL. The lipoprotein distribution in the HDL range of interstitial fluid differed from that of serum showing one maximum at a density of about 1.070 g/ml (serum HDL2 about 1.090 g/ml) and one at a density of 1.130-1.140 g/ml (serum HDL3, 1.110-1.120 g/ml). The former subfraction contained most of the lipoprotein-bound apoE while the latter contained the major part of apoA-I and apoA-II. Studies of the lipoproteins of interstitial fluid may add to our understanding of the development of atherosclerosis and xanthomatosis and may also provide valuable information on the permeability of the capillary membrane in normo- and pathophysiological states.     

Tocopherols and carotenes are differently distributed in subfractions of high-density lipoprotein

An apolipoprotein (apo) E-rich and an apo E-poor fraction of high-density lipoprotein (HDL) were isolated from four healthy men by heparin-Sepharose affinity chromatography. On a cholesterol basis, the apo E-poor HDL fraction contained a third more α- and γ-tocopherol and about a third less α- and β-carotene than the apo E-rich HDL fraction. Plasma concentrations of HDL cholesterol were highly correlated with the contribution of the apo E-rich HDL subfraction to total HDL α-tocopherol (r = − 0.990, P < 0.001).     

Distribution of C apolipoproteins between the vascular and lymph compartments of the rat

This study has investigated the kinetics of transfer of C apolipoproteins between the vascular and lymph compartments of the rat. Very-low-density lipoprotein, labeled with [125I]apolipoprotein C, was injected intravenously into lymph duct-cannulated rats and the redistribution of radioactivity between lymph and plasma followed at frequent intervals for 3 h. Equilibration between the two compartments was rapid (10-15 min), and thereafter removal from both compartments continued at similar rates. Specific radioactivity determinations showed that lymph C-III-0, C-III-3, and C-III-2,1 apolipoproteins rapidly reached values identical to those of corresponding plasma C apolipoproteins and the interrelationship between the curves were consistent with precursor-product relationships in which all, or most, of the product (lymph apolipoprotein C-III) was derived from the precursor (plasma). However, the specific radioactivity curves for C-II peptide did not cross; the lower value for lymph C-II apolipoprotein suggests that, unlike C-III apolipoproteins, a substantial proportion (approx. 40%) of lymph C-II peptide is not derived from the plasma compartment. The most likely source of the unlabeled lymph apolipoprotein C-II is synthesis and secretion from the intestine.     

No severe bottleneck during human evolution: evidence from two apolipoprotein C-II deficiency alleles.

The DNA sequences of a Japanese and a Venezuelan apolipoprotein (apo) C-II deficiency allele, of a normal Japanese apo C-II gene, and of a chimpanzee apo C-II gene were amplified by PCR, and their nucleotide sequences were determined on multiple clones of the PCR products. The normal Japanese sequence is identical to--and the chimpanzee sequence differs by only three nucleotides from--a previously published normal Caucasian sequence. In contrast, the two human mutant sequences each differ from the normal apo C-II gene sequence by several nucleotides, including deletions. The data suggest that both mutant alleles arose greater than 500,000 years ago. It is shown that a defective allele can persist in a population for only a short time if a bottleneck occurs. Therefore, the antiquity of the two alleles suggests no severe bottleneck during human evolution. Moreover, the fact that one allele is from Japan and the other is from a Venezuelan Caucasian family is more consistent with the multiregional evolution model of modern human origins than with the complete replacement or "out of Africa" model.     

Effects of rosuvastatin on electronegative LDL as characterized by capillary isotachophoresis: the ROSARY Study

Electronegative LDL, a charge-modified LDL (cm-LDL) subfraction that is more negatively charged than normal LDL, has been shown to be inflammatory. We previously showed that pravastatin and simvastatin reduced the electronegative LDL subfraction, fast-migrating LDL (fLDL), as analyzed by capillary isotachophoresis (cITP). The present study examined the effects of rosuvastatin on the more electronegative LDL subfraction, very-fast-migrating LDL (vfLDL), and small, dense charge-modified LDL (sd-cm-LDL) subfractions. Patients with hypercholesterolemia or those who were being treated with statins (n = 81) were treated with or switched to 2.5 mg/d rosuvastatin for 3 months. Rosuvastatin treatment effectively reduced cITP cm-LDL subfractions of LDL (vfLDL and fLDL) or sdLDL (sd-vfLDL and sd-fLDL), which were closely related to each other but were different from the normal subfraction of LDL [slow-migrating LDL (sLDL)] or sdLDL (sd-sLDL) in their relation to the levels of remnant-like particle cholesterol (RLP-C), apolipoprotein (apo) C-II, and apoE. The percent changes in cm-LDL or sd-cm-LDL caused by rosuvastatin were correlated with those in the particle concentrations of LDL or sdLDL measured as LDL-apoB or sdLDL-apoB and the levels of HDL-C, RLP-C, apoC-II, and apoE. In conclusion, rosuvastatin effectively reduced both the vfLDL subfraction and sd-cm-LDL subfractions as analyzed by cITP.     

Effect of copper deficiency on the composition of three high-density lipoprotein subclasses as separated by heparin-affinity chromatography

Copper deficiency in rats produces a hypercholesterolemia with a marked increase in HDL fraction. This study investigated changes in the plasma distribution and composition of HDL subclasses as affected by copper deficiency. Plasma HDL were separated into the following three subclasses by heparin-affinity chromatography: HDL containing no apo E but high in apo A-I (HDL-E0); HDL with an intermediate level of apo E (HDL-E1); and HDL highly enriched in apo E but low in apo A-I (HDL-E2). The compositional analysis showed that the hypercholesterolemia observed in copper-deficient rats was due specifically to an increase in plasma cholesterol carried by HDL-E0. Copper deficiency did not alter the percent distribution of apo A-I in HDL-E0, but lowered the apo A-I content in HDL-E1 and HDL-E2, with an increase in apo E in these subclasses. The total plasma concentration of apo A-I was, however, significantly elevated in Cu-deficient rats, which was attributable to an increase in the total number of circulating HDL particles. No difference was noted between Cu-deficient and control groups in the distribution of free cholesterol or the ratio of free cholesterol to esterified cholesterol in any of the HDL subclasses. The present results and earlier observations suggest that copper deficiency may produce a defect in the plasma clearance or tissue uptake of the HDL subclass high in apo A-I but devoid of apo E (HDL-E0), which may be mediated by the specific apo A-I receptor or non-endocytotic transfer of HDL-E0 cholesterol to the liver. Such metabolic defects may partly explain the simultaneous increases in both plasma HDL cholesterol and apo A-I and altered cholesterol homeostasis observed in copper deficiency.     

Inhibitory effects of C apolipoproteins from rats and humans on the uptake of triglyceride-rich lipoproteins and their remnants by the perfused rat liver

Like rat C apolipoproteins, each of the C apolipoproteins from human blood plasma (C-I, C-II, C-III-1, and C-III-2) bound to small chylomicrons from mesenteric lymph of estradiol-treated rats and inhibited their uptake by the isolated perfused rat liver. This inhibitory effect of the C apolipoproteins was independent of apolipoprotein E, which is present only in trace amounts in these chylomicrons. Addition of rat apolipoprotein E to small chylomicrons from mesenteric lymph of normal rats did not displace C apolipoproteins and had no effect on the uptake of these particles by the perfused liver, indicating that an increased ratio of E apolipoproteins to C apolipoproteins on chylomicron particles, unaccompanied by depletion of the latter, may not promote recognition by the chylomicron remnant receptor. The hepatic uptake of remnants of rat hepatic very low density lipoproteins (VLDL) and small chylomicrons, which had been produced in functionally eviscerated rats, was also inhibited by addition of C apolipoproteins. These observations are consistent with the hypothesis that the addition of all of the C apolipoproteins to newly secreted chylomicrons and VLDL inhibits premature uptake of these particles by the liver and that depletion of all of these apolipoproteins from remnant particles facilitates their hepatic uptake. Remnants of chylomicrons and VLDL incubated with rat C apolipoproteins efficiently took up C-III apolipoproteins, but not apolipoprotein C-II (the activator protein for lipoprotein lipase). Preferential loss of apolipoprotein C-II during remnant formation may regulate the termination of triglyceride hydrolysis prior to complete removal of triglycerides from chylomicrons and VLDL.     

Rapid clearance of serum amyloid a from high-density lipoproteins

The serum amyloid A proteins (SAA) occur in plasma in six polymorphic forms that are associated with the high-density lipoproteins (HDL). We studied two of the SAA proteins, SAA₁ and SAA₄, which have the same amino- and carboxy-terminal residues but different solution properties and electrophoretic mobilities, to determine whether they are interconverted in plasma in vivo. They were radioiodinated in vitro, incorporated into HDL, and administered to cynomolgus monkeys. Both remained associated with HDL for at least 6 h, had similar plasma die-away curves, and retained their characteristic electrophoretic mobilities, suggesting they are not related as precursor and product. The plasma clearance of the most prevalent SAA species, SAA₄, was also simultaneously compared with the human A-I and C-III-2 apolipoproteins. Human apolipoprotein A-I decayed from plasma at a rate comparable to that of monkey HDL proteins. Apolipoprotein C-III-2 was cleared more rapidly and SAA₄ at an even greater rate. These findings suggest that SAA are either dissociated from HDL before clearance from plasma or that SAA are contained in an HDL subspecies with metabolic fate different from that of most HDL particles.     

Metabolic disorders of serum lipoproteins in endotoxin-poisoned mice: the role of high density lipoprotein (HDL) and triglyceride-rich lipoproteins

A study was performed to clarify the role of serum lipoproteins, especially high density lipoprotein (HDL) and triglyceride-rich lipoproteins in endotoxemic or endotoxin-poisoned animals. The level of HDL-cholesterol decreased markedly in mouse serum 18-24 hr postintoxication, while the amount of low density lipoprotein (LDL)-cholesterol in the sera of poisoned mice was about 175% of that of the controls. Serum lecithin-cholesterol acyltransferase activity in the poisoned mice decreased slightly for 3-6 hr after endotoxin injection, but became markedly increased at 18-24 hr as compared with that in the controls. The amount of serum very low density lipoprotein (VLDL) showed a marked increase in the poisoned mice 8-24 hr postintoxication. The HDL fraction in the electrophoretic patterns of serum was reduced according to the dose of endotoxin 18 hr postintoxication. The HDL fraction in mice injected with lead acetate plus endotoxin was markedly lower than that in the poisoned mice. When streptozotocin-diabetic mice were injected with endotoxin, the HDL fraction was higher than that in the endotoxin-poisoned mice. In endotoxin-poisoned mice a correlation was observed between the lipid peroxide and LDL levels in the serum. In disk electrophoretic patterns, the HDL fraction in mice given vitamin E-supplemented diet showed a higher level than that in mice given a normal diet. Lipoprotein lipase (LPL) activity in poisoned mice significantly decreased to 59% of the control value 18 hr postintoxication, but hepatic triglyceride lipase activity was only slightly increased in endotoxin-poisoned mice. In analysis of HDL apoprotein peptide in serum lipoprotein, the apo C-II peptide level was clearly lower in mouse serum 18 hr postintoxication than that in the controls. These results suggest that the decrease in LPL activity in endotoxin-poisoned mice may be closely related to a decrease in the apo C-II peptide level, and also that it plays an important part in HDL and triglyceride-rich lipoprotein metabolism in the poisoned mice.     

Discoidal complexes of A and C apolipoproteins with lipids and their reactions with lecithin: cholesterol acyltransferase

Micellar, discoidal complexes of human apolipoproteins A-I, A-II, C-I, C-II, C-III-1, and C-III-2 with egg phosphatidylcholine (egg-PC) and cholesterol were prepared by the cholate dialysis method. The complexes, isolated by gel filtration, had similar lipid and protein contents by weight, on the average: 1.77:0.083:1.0, egg-PC/cholesterol/apolipoprotein (w/w). The diameters of the discs, visualized by electron microscopy and estimated by gel filtration, ranged from 100 to 200 A. The alpha-helix content of the apolipoproteins in the complexes was from 50-72%, and their fluorescence properties indicated nonpolar, but quite varied environments for the tryptophan residues in the various complexes. Initial reactions of purified human lecithin: cholesterol acyltransferase with the complexes, adjusted to equal egg-PC concentrations, indicated that all the apolipoproteins activate the enzyme from 6-fold to 400-fold over control vesicles of egg-PC and cholesterol. In decreasing order of reactivity were the complexes with A-I, C-I, C-III-1, C-III-2, C-II, and A-II. These results indicate that aside from lipid-binding capacity and high amphipathic alpha-helix content, other structural features are required for optimal enzyme activation by apolipoproteins. Concentration and temperature dependence experiments gave similar apparent Km values, markedly different apparent Vmax, and very similar activation energies (about 19 kcal/mol), for the various complexes. These observations suggest that the rate-limiting enzymatic step of the reaction is common to all the complexes but that the activated enzyme levels differ from complex to complex. We propose that enzyme activation occurs upon binding to complexes via apolipoproteins. Addition of excess (5-fold) free apolipoprotein A-I or A-II to complexes resulted in the exchange of bound for free apolipoproteins and in loss of reactivity with the enzyme.     

Regulation of high density lipoprotein receptors in cultured macrophages: role of acyl-CoA:cholesterol acyltransferase

The interaction of human serum high density lipoproteins (HDL) with mouse peritoneal macrophages and human blood monocytes was studied. Saturation curves for binding of apolipoprotein E-free [125I]HDL3 showed at least two components: non-specific binding and specific binding that saturated at approximately 40 micrograms HDL protein/ml. Scatchard analysis of specific binding of apo E-free [125I]-HDL3 to cultured macrophages yielded linear plots indicative of a single class of specific binding sites. Pretreatment of [125I]HDL3 with various apolipoprotein antibodies (anti apo A-I, anti apo A-II, anti apo C-II, anti apo C-III and anti apo E) and preincubation of the cells with anti-idiotype antibodies against apo A-I and apo A-II prior to the HDL binding studies revealed apolipoprotein A-I as the ligand involved in specific binding of HDL. Cellular cholesterol accumulation via incubation with acetylated LDL led to an increase in HDL binding sites as well as an increase in the activity of the cytoplasmic cholesterol esterifying enzyme acyl-CoA:cholesterol acyltransferase (ACAT). Incubation of the cholesterol-loaded cells in the presence of various ACAT inhibitors (Sandoz 58.035, Octimibate-Nattermann, progesterone) revealed a time- and dose-dependent amplification in HDL binding and HDL-mediated cholesterol efflux. It is concluded that the homeostasis of cellular cholesterol in macrophages is regulated in part by the number of HDL binding sites and that ACAT inhibitors enhance HDL-mediated cholesterol efflux from peripheral cells.     

Activation of lipoprotein lipase by apolipoprotein C-II is modulated by the COOH terminal region of apolipoprotein C-III

The in vitro effect of apolipoprotein C-II (apo C-II) on the apolipoprotein C-III (apo C-III) induced activation of bovine milk lipoprotein lipase (LPL) was studied in vitro using a synthetic substrate. Apo C-III effectively inhibited, in a dose-dependent manner, the activation of lipoprotein lipase induced by apo C-II. A 3-fold molar apo C-III excess decreased the lipoprotein lipase activity by 25%. Thrombin cleavage of apo C-III produced two fragments: only fragment 41-79 retained the inhibitory activity and was equipotent to native apo C-III1 on a molar basis. Neither displacement of apo C-II from the substrate, as determined using 125I-labeled apo C-II, nor the charge carried by sialic residues of apo C-III, as demonstrated in experiments performed after neuraminidase treatment, accounted for this effect. I speculate that apo C-III may act by inhibiting the apo C-II-LPL interaction.     

Characterization and measurement of human apolipoprotein A-II by radioimmunoassay

The development of a radioimmunoassay for apolipoprotein A-II (apo A-II) is described. Initial studies revealed a lack of immunological identity between purified apo A-II used as the standard and serum or HDL. Extensive testing of different buffers, standards, antisera, tracers, utilization of a detergent, and heating of sera failed to resolve the problem. Gel filtration of iodinated and non-iodinated apo A-II on Sephadex G-100 columns showed that apo A-II, in dilute solution, elutes in a higher molecular zone than expected with a broad, assymetrical profile. The use of a subfraction of the tracer in the assay resulted in parallelism in the serum and standard dilution curves. The apo A-II assay was sensitive, specific, and reproducible. Apo A-II added to sera was fully recovered and delipidation did not affect the immunoreactivity of either serum or HDL. Apo A-II contributed approximately 20% to the protein mass of HDL. Comparison of these results with those obtained by radial immunodiffusion, and with previously reported data, indicates that the reactivity of apo A-II in its native and delipidated forms may be markedly influenced by different immunologic methodologies and their specific reagents. Caution should thus be shown at present in assigning absolute concentrations to apo A-II in serum or HDL.     

Speciated human high-density lipoprotein protein proximity profiles

It is expected that the attendant structural heterogeneity of human high-density lipoprotein (HDL) complexes is a determinant of its varied metabolic functions. To determine the structural heterogeneity of HDL, we determined major apolipoprotein stoichiometry profiles in human HDL. First, HDL was separated into two main populations, with and without apolipoprotein (apo) A-II, LpA-I and LpA-I/A-II, respectively. Each main population was further separated into six individual subfractions using size exclusion chromatography (SEC). Protein proximity profiles (PPPs) of major apolipoproteins in each individual subfraction was determined by optimally cross-linking apolipoproteins within individual particles with bis(sulfosuccinimidyl) suberate (BS(3)), a bifunctional cross-linker, followed by molecular mass determination by MALDI-MS. The PPPs of LpA-I subfractions indicated that the number of apoA-I molecules increased from two to three to four with an increase in the LpA-I particle size. On the other hand, the entire population of LpA-I/A-II demonstrated the presence of only two proximal apoA-I molecules per particle, while the number of apoA-II molecules varied from one dimeric apoA-II to two and then to three. For most of the PPPs described above, an additional population that contained a single molecule of apoC-III in addition to apoA-I and/or apoA-II was detected. Upon composition analyses of individual subpopulations, LpA-I/A-II exhibited comparable proportions for total protein (～58%), phospholipids (～21%), total cholesterol (～16%), triglycerides (～5%), and free cholesterol (～4%) across subfractions. LpA-I components, on the other hand, showed significant variability. This novel information about HDL subfractions will form a basis for an improved understanding of particle-specific functions of HDL.