Spurious elevation of serum high-density lipoprotein cholesterol in patients with cholestatic liver diseases.

Strikingly discrepant values were obtained by two commercial precipitating reagents for serum HDL cholesterol determination in three patients with cholestatic liver diseases (two patients with primary biliary cirrhosis and one patient with chronic hepatitis). An abnormal alpha 2-migrating lipoprotein (slow alpha-lipoprotein) was observed in agarose gel electrophoresis for each serum. The slow alpha-lipoprotein was partly recovered in the supernatant by precipitation with polyethylene glycol, and was completely precipitated with a polyanion-containing reagent, which well explains the discrepancy. The slow alpha-lipoprotein isolated from one of the cases is notable for (1) having an intermediate particle size between normal LDL and normal HDL, (2) containing apo E as the major apolipoprotein, and (3) being enriched with cholesterol (esterified and free) and phospholipid. Cholesterol accumulation was also found in another HDL subclass, alpha 1-migrating HDL. A severe impediment in the clearance of cholesterol-loaded HDL particles from plasma was implied. Electrophoresis of serum lipoproteins and/or the measurement of serum apo E concentrations are necessary to avoid an erroneous estimation of HDL cholesterol in patients with hepatobiliary diseases.     

Identification and characterization of rat serum lipoprotein subclasses. Isolation by chromatography on agarose columns and sequential immunoprecipitation

Lipoproteins, present in serum of chow-fed rats, were fractionated according to size by chromatography of serum on 6% agarose columns. The distributions of apolipoprotein (apo) A-I, E, and A-IV within the high density lipoprotein (HDL) size range (i.e., lipoprotein complexes smaller than low density lipoproteins) showed the existence of lipoprotein subclasses with different size and chemical composition. Sequential immunoprecipitations were performed on these fractions obtained by agarose column chromatography, using specific antisera against apoA-I, apoE, and apoA-IV. The resulting precipitates and supernatants were analyzed for cholesteryl esters, unesterified cholesterol, phospholipids, triglycerides, and specific lipoproteins. The following conclusions were drawn from these experiments. Sixty-three +/- 3% of apoE in the total HDL size range is present on a large particle (mol wt 750,000). This lipoprotein contains apoE as its sole protein constituent and is called LpE. Thirty-nine +/- 4% of the cholesterol found in the HDL size range is present in this fraction. The cholesterol:phospholipid ratio is 1:1.1. Sixty-nine +/- 8% of apoA-I in the total HDL size range is present on a smaller particle (mol wt 250,000). This apoA-I-HDL has apoA-I as its major protein component and possibly contains minor amounts of C apoproteins and A-II, but neither apoE nor apoA-IV. It contains 39 +/- 8% of the total cholesterol found in the HDL size range and the cholesterol:phospholipid ratio is 1:1.6.(ABSTRACT TRUNCATED AT 250 WORDS)     

Association of antisense oligonucleotides with lipoproteins prolongs the plasma half-life and modifies the tissue distribution.

In order to direct antisense oligonucleotides to specific tissues or cell types in vivo, we are exploring the possibility to utilize lipoproteins as transport vehicles. A 16-mer oligonucleotide (ODN) was derivatized at the 5' prime through a 32P phosphate spacer with cholesterol, yielding a 32P-labeled amphiphatic cholesteryl-oligonucleotide (cholODN). Incubation of cholODN with low-density lipoprotein (LDL) for 2 hr at 37 degrees C resulted in the formation of a cholODN-LDL complex that migrates as a single peak on agarose gel electrophoresis. The cholODN was found to bind quantitatively to both high-density lipoproteins (HDL) and LDL, but not to albumin. Stable oligonucleotide-LDL particles with up to 50 molecules of cholODN per LDL particle could be obtained. In contrast, the control ODN did not show affinity for plasma lipoproteins. Upon injection into rats, cholODN became rapidly associated with plasma lipoproteins while control ODNs were recovered in the lipoprotein deficient serum fraction. The plasma half-life of cholODN (9-11 min) is considerably prolonged as compared with the control ODN (t1/2 less than 1 min). The cholODN-LDL was at least 5 min stable against degradation by rat plasma nucleases. It is concluded that derivatization of antisense oligonucleotides with cholesterol profoundly modifies their in vivo fate and opens possibilities for efficient and specific receptor-dependent targeting, mediated by lipoproteins coupled with specific recognition markers to various hepatic cell types.     

The kinetics of binding of serum lipoproteins by immobilized heparin.

A model arterial system of heparin immobilized on an agarose gel was used to study the amount and kinetics of binding of porcine serum lipoproteins to heparin. Binding occurred to lipoproteins in the density range 1.006 less than d less than 1.062, but there was no binding with high density lipoprotein. A theoretical model of the kinetic experiments was formulated and used to demonstrate that the rate of the binding reaction could be considered instantaneous relative to the rate of transport of lipoproteins. Extrapolation of these results to arterial levels of glycosaminoglycans and lipoprotein indicate that complexes of lipoprotein and the glycosaminoglycans could account for much of the cholesterol entrapment in atherosclerotic lesions.     

Characterization of subfractions of triglyceride-rich lipoproteins separated by gel chromatography from blood plasma of normolipemic and hyperlipemic humans

As judged from measurements of the diameters of particles fixed with osmium tetroxide and shadowed with platinum, gel chromatography on 2% agarose has been shown to be an effective quantitative method for separating triglyceride-rich lipoproteins according to particle size. Particles in the size range of chylomicrons, uncontaminated by lipoproteins smaller than about 700 A or by other serum proteins, emerged in the void volume of the column, and very low density lipoproteins with diameters between 400 and 700 A were separated into fractions with average standard deviation of 71 A from the mean. Systematic comparison of the relationship between diameter and chemical composition of fractions obtained from subjects with various hyperlipoproteinemic disorders demonstrated a precise correlation consistent with a spherical model for these lipoproteins in which phospholipids, free cholesterol, and protein occupy a surface monolayer with an invariant thickness of 21.5 A surrounding a liquid core of triglycerides and cholesteryl esters. The chemical composition of very low density lipoproteins of given particle size in most recognized types of hyperlipemia was similar to that of normolipemic subjects, but particles in the size range of chylomicrons sometimes had higher contents of cholesteryl esters and free cholesterol. Results obtained in subjects with dysbetalipoproteinemia were consistent with the presence of three populations of particles. Two of these, with mean diameters of about 850 and 350 A, had unusually high cholesteryl ester content and reduced triglyceride content and may represent "remnants" of the metabolism of structurally normal chylomicrons and very low density lipoproteins, respectively. The third, a heterogeneous group with intermediate range of particle size and pre-beta mobility, may represent a population of very low density lipoproteins with relatively normal composition.     

Transfer of excess cholesterol from human red blood cells to plasma in vitro

This study examined the ability of plasma and plasma fractions from normolipidaemic subjects and plasma from a patient with homozygous familial high density lipoprotein deficiency (Tangier disease) to promote loss of excess cholesterol from red blood cells in vitro. Isolated high density lipoproteins were the most potent plasma fraction for removing excess cellular cholesterol. Lipoprotein-deficient plasma and human serum albumin, but not very low density lipoproteins and low density lipoproteins, also removed excess cholesterol from the red blood cells. The near absence of high density lipoproteins in plasma from the patient with Tangier disease did not result in an abnormally low rate of cholesterol loss from the enriched red blood cells. These results suggest that normal levels of high density lipoproteins are not vital for the removal of excess cholesterol from red blood cells by plasma.     

Stimulation of cholesterol esterification in hepatic microsomes by lipoproteins from normal and hypercholesterolemic rabbit serum.

Incubation of plasma lipoproteins with rabbit hepatic microsomes enriched the microsomes with free cholesterol and stimulated cholesterol esterification. The rate of cholesterol esterification correlated well (r = 0.96) with the concentration of microsomal free cholesterol. Lipoproteins from normal and hypercholesterolemic serum varied in their propensity to stimulate cholesterol esterification. Among the normal lipoproteins, low density lipoproteins was more stimulatory than either high density lipoproteins or intermediate density lipoproteins. However, the intermediate density lipoproteins fraction from hypercholesterolemic serum was consistently more stimulatory than any of the normal lipoproteins. The augmentation of cholesterol content, when microsomes were exposed to mixed hyperlipidemic lipoproteins, was proportionately much greater than augementation of phospholipid or protein concentration.     

Kinetics and mechanism of phosphatidylcholine and cholesterol exchange between chylomicrons and high-density lipoproteins.

The exchange of phosphatidylcholine and unesterified cholesterol between rat mesenteric lymph chylomicrons and human high-density lipoproteins was studied in vitro by incubation of radiolabelled chylomicrons (with [N-methyl-14C]phosphatidylcholine and [7(n)-3H]cholesterol) with unlabelled high-density lipoproteins. The kinetic analysis was based on the extent of radioisotope exchange, which was determined by the proportion of label appearing in the high-density lipoprotein elution peak after rapid fractionation on analytical agarose columns. Under our experimental conditions, no net transfer of either phosphatidylcholine or cholesterol is observed. The kinetics of exchange of both phosphatidylcholine and cholesterol are biphasic. Over the first 30 min a maximum of 25% of the phosphatidylcholine and 33% of the cholesterol in chylomicrons exchanges rapidly into the high-density-lipoprotein fraction. Thereafter both lipids continue to exchange for up to 3 h at a much lower rate. For the rapid exchange process the calculated exchange rates for phosphatidylcholine and cholesterol are proportional to the concentrations of both chylomicrons and high-density lipoproteins. The second-order rate constants are (10.5 +/- 0.5) X 10(-5) microM-1 X min-1 for phosphatidylcholine and (32.1 +/- 4.5) X 10(-5) microM-1 X min-1 for cholesterol. The kinetics of the exchange process thus suggest that a significant proportion of both phosphatidylcholine and unesterified cholesterol is rapidly exchangeable between these lipoproteins, and that this exchange is mediated by a 'bimolecular', or collisional, mechanism.     

Serum lipoproteins as cholesterol donors to mycoplasma membranes.

Mycoplasma hominis andAcholeplasma laidlawii were grown in media in which a fraction of human serum lipoproteins provided the sole source of cholesterol. Increasing levels of very low density lipoproteins had an inhibitory effect on the growth of the organisms. Low and high density lipoproteins in all concentrations proved to be excellent sources of cholesterol. Both organisms were able to limit the amount of cholesterol taken up and to preferentially incorporate free cholesterol despite an excess of esterified cholesterol in the medium. When similar levels of free cholesterol were provided by low density or high density lipoproteins, the organisms incorporated from 20–45% more cholesterol from the former. This preference for cholesterol from low density lipoproteins partially supports the theory that the low density lipoproteins act as a donor while the high density lipoproteins are a scavenger of cholesterol.     

GPI-specific phospholipase D associates with an apoA-I- and apoA-IV-containing complex

Glycosylphosphatidylinositol-specific phospholipase D (GPI-PLD) is abundant in serum and associates with high density lipoproteins (HDL). We have characterized the distribution of GPI-PLD among lipoproteins in human plasma. Apolipoprotein (apo)-specific lipoproteins containing apoB (Lp[B]), apoA-I and A-II (Lp[A-I, A-II]), or apoA-I only (Lp[A-I]) were isolated using dextran sulfate and immunoaffinity chromatography. In six human plasma samples with HDL cholesterol ranging from 39 to 129 mg/dl, 79 +/- 14% (mean +/- SD) of the total plasma GPI-PLD activity was associated with Lp[A-I], 9 +/- 12% with Lp[A-I, A-II], and 1 +/- 1% with Lp[B]; and 11 +/- 10% was present in plasma devoid of these lipoproteins. Further characterization of the GPI-PLD-containing lipoproteins by gel-filtration chromatography and nondenaturing polyacrylamide and agarose gel electrophoresis revealed that these apoA-I-containing particles/complexes were small (8 nm) and migrated with pre-beta particles on agarose electrophoresis. Immunoprecipitation of GPI-PLD with a monoclonal antibody to GPI-PLD co-precipitated apoA-I and apoA-IV but little or no apoA-II, apoC-II, apoC-III, apoD, or apoE. In vitro, apoA-I but not apoA-IV or bovine serum albumin interacted directly with GPI-PLD, but did not stimulate GPI-PLD-mediated cleavage of a cell surface GPI-anchored protein. Thus, the majority of plasma GPI-PLD appears to be specifically associated with a small, discrete, and minor fraction of lipoproteins containing apoA-I and apoA-IV. -- Deeg, M. A., E. L. Bierman, and M. C. Cheung. GPI-specific phospholipase D associates with an apoA-I- and apoA-IV-containing complex. J. Lipid Res. 2001. 42: 442--451.     

Analysis method for lipoproteins by high-performance liquid chromatography with sulfopropyl-ligand column and magnesium ion-containing eluents

We have developed a new analysis method for lipoproteins in serum by high-performance liquid chromatography using a sulfopropyl-ligand column with eluents containing magnesium nitrate. The magnesium ion anchors lipoproteins to the ligands on the column gel. Lipoproteins are eluted from the column with a magnesium nitrate concentration gradient and detected by postcolumn reaction using a reagent containing cholesterol esterase and cholesterol oxidase. High-density lipoprotein, low-density lipoprotein, and very-low-density lipoprotein were eluted in order from the column. The within-assay and between-assay coefficients of variation for cholesterol concentration in lipoproteins were 1.1-3.7 and 1.3-5.8%, respectively. The correlation coefficients between the values of total cholesterol, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol obtained by the new method and those obtained by an enzymatic method using an automated chemical analyzer were 0.940, 0.979, and 0.909, respectively. The new method was successfully applied to the analysis of plasma lipoproteins of patients with hyperlipidemia.     

Lipid composition and cholesterol esterification in serum lipoprotein fraction of the horse, Equus caballus.

1. Changes in lipid components of lipoproteins during incubation of horse serum at 37 degrees C were investigated. In non-incubated serum, cholesterol and lecithin existed predominantly in alpha-lipoprotein or in high-density lipoprotein (HDL). Lysolecithin was mainly associated with the fraction with density above 1.21. 2. When serum was separated into alpha- and beta-lipoproteins by the heparin precipitation method after 1 hr incubation, the decrease in alpha-lipoprotein free cholesterol and lecithin was about four times that in beta-lipoprotein counterparts. 3. When serum lipoproteins were separated by ultracentrifugation, the decrease in each lipoprotein free cholesterol was closely paralleled with that in lecithin. 4. HDL appeared to be a preferential substrate for the lecithin: cholesterol acyltransferase reaction. 5. Disc electrophoretic patterns indicated significant differences in the composition of horse serum lipoproteins from those of human and rat.     

Diffusible Fraction of Serum Cholesterol and Atherogenesis

CHOLESTEROL is found in the blood as a structural component of lipoproteins concerned with the transport of other lipids¹. The high resolution nuclear magnetic resonance spectra of high density serum lipoproteins are similar to that observed when lipids are dissolved in organic solvents, or dispersed in water by bile salts or detergents, or in sonicated form. The lipid component in lipoproteins is therefore probably in an extremely fluid condition². If human serum is mixed with paraffin oil, some of the cholesterol diffuses into the oil without affecting the ultraviolet absorption spectrum of serum proteins. This procedure avoids any protein denaturing action used for cholesterol extraction^3–5. It therefore seems that serum cholesterol has two fractions, one strongly bound by lipoprotein structures and the other loosely bound and diffusible in an oil phase. In this article I designate the loosely bound fraction “diffusible”.     

Lipid peroxidation--the factor promoting cholesterol accumulation in cells in atherogenesis

The cholesterol transfer between human erythrocytes and main classes of serum lipoproteins (LP) from healthy donors and artery-coronary disease patients was studied (artery-coronary disease is the main manifestation of atherosclerosis). It is shown that low-density lipoproteins (LDL) are capable of transporting cholesterol to erythrocytes, which lack the specific receptors for LDL. The cell cholesterol content in comparison with erythrocytes incubated without LDL was increased by 11.4%. The effect was even higher in case of LDL, isolated from serum of artery-coronary subjects (the cell cholesterol content was increased by 33.8%). High-density lipoproteins (HDL) accept cholesterol from cell membranes. However, cholesterol-accepting properties of HDL from artery-coronary disease patients were suppressed as compared with normal HDL. Both discovered events must promote the cholesterol accumulation in cell membranes in atherosclerosis. As it is shown by the spin probe method, lipid peroxidation (LPO) causes the disturbance of the structural organization of LP and as the consequence of that--the increase of LDL cholesterol-donating ability and the decrease of HDL cholesterol-accepting ability. The greater LDL are oxidized, the more cholesterol they transport to erythrocytes during incubation. The greater is the level of HDL peroxidation, the stronger their cholesterol-accepting function is suppressed. These results suggest that LPO can play an important role in LP modification, the disturbance of their interaction with cell surface and the cholesterol accumulation in cells in atherosclerosis.     

Activation of lipoprotein lipase by lipoprotein fractions of human serum

Triglycerides in fat emulsions are hydrolyzed by lipoprotein lipase only when they are "activated" by serum lipoproteins. The contribution of different lipoprotein fractions to hydrolysis of triglycerides in soybean oil emulsion was assessed by determining the quantity of lipoprotein fraction required to give half-maximal hydrolysis. Most of the activator property of whole serum from normolipidemic, postabsorptive subjects was in high density lipoproteins. Low density lipoproteins and serum from which all lipoprotein classes were removed had little or no activity. Also, little activator was present in guinea pig serum or in very low density poor serum from an individual with lecithin:cholesterol acyltransferase deficiency, both of which are deficient in high density lipoproteins. Human very low density lipoproteins are potent activators and are much more active than predicted from their content of high density lipoprotein-protein. Per unit weight of protein, very low density lipoproteins had 13 times the activity of high density lipoproteins. These observations suggest that one or more of the major apoproteins of very low density lipoproteins, present as a minor constituent of high density lipoproteins, may be required for the activation process.     

Removal of apoprotein-B-containing lipoproteins by plasmapheresis using immobilized phosphorylcholine-binding protein affinity adsorbent

Rat serum phosphorylcholine binding protein was earlier shown to bind lipoproteins containing apoproteins B and E from human very low and low density lipoproteins. The present studies were undertaken to show the effectiveness of rat serum phosphorylcholine-binding protein immobilized on Sepharose affinity column to remove apoprotein-B-containing lipoproteins from normal and hypercholesterolemic rabbit plasma, when used in a plasmapheresis system. The maximum in vitro binding of very low and low density lipoproteins from hypercholesterolemic rabbit plasma to the affinity adsorbent was Ca2+ dependent, and the cholesterol bound to the column at the optimum calcium concentration (2.5 mM) was 21% of the total plasma cholesterol applied. The in vivo binding of total cholesterol from normal and hypercholesterolemic rabbit plasma during plasmapheresis ranged from 0.22 to 7.7%. Total mass of cholesterol bound ranged from 3.86 and 27.52 mg at plasma cholesterol concentrations 13.8 and 282 mg/dL, respectively. Most (greater than 95%) of the bound cholesterol was associated with very low and low density lipoproteins. These studies show the ability of immobilized rat serum phosphorylcholine-binding protein to lower the atherogenic apoprotein-B-containing lipoproteins from plasma of hypercholesterolemic rabbits.     

In vitro incorporation of radiolabeled cholesteryl esters into high and low density lipoproteins

We have developed and validated a method for in vitro incorporation of radiolabeled cholesteryl esters into low density (LDL) and high density lipoproteins (HDL). Radiolabeled cholesteryl esters dissolved in absolute ethanol were mixed with LDL or HDL in the presence of lipoprotein-deficient serum (LPDS) as a source of core lipid transfer activity. The efficiency of incorporation was dependent on: a) the core lipid transfer activity and quantity of LPDS, b) the mass of added radiolabeled cholesteryl esters, c) the length of incubation, and d) the amount of acceptor lipoprotein cholesterol. The tracer incorporation was documented by repeat density gradient ultracentrifugation, agarose gel electrophoresis, and precipitation with heparin-MnCl2. The radiolabeling conditions did not affect the following properties of the lipoproteins: 1) chemical composition, 2) electrophoretic mobility on agarose gels, 3) hydrated density, 4) distribution of apoproteins on SDS gels, 5) plasma clearance rates, and 6) immunoprecipitability of HDL apoproteins A-I and A-II. Rat HDL containing radiolabeled cholesteryl esters incorporated in vitro had plasma disappearance rates identical to HDL radiolabeled in vivo.     

Lipid transfer between human serum high density lipoproteins and egg yolk lipoproteins in incubation mixtures

Ultracentrifugally isolated human serum high density lipoproteins of d 1.063-1.21 (HDL) were incubated with egg yolk lipoproteins of d < 1.006 for up to 24 hr at various concentrations. Transfer of HDL cholesterol esters to egg yolk lipoproteins occurred simultaneously with transfer of glycerides from egg yolk lipoproteins to HDL. These observations show that exchange of lipids can take place between lipoproteins in the absence of other serum proteins and enzymes. The mole ratios of HDL cholesterol esters to glycerides approached an integral value of 1 : 1 during the course of the incubation. These results suggest that lipid components form complexes within the HDL structure.     

Lipoprotein-mediated efflux of radiolabeled cholesterol from cells does not indicate net removal of cellular cholesterol mass

The efflux of cholesterol from human skin fibroblasts was determined using radioisotope techniques and mass measurements. When the cells were labeled with [14C]- or [3H]-cholesterol and then incubated with very low density, low density, or high density lipoproteins or with serum, 20 to 30% of the label was released into the medium in 20 h. However, when the cellular cholesterol content was determined after incubation with various lipoproteins under identical conditions, only the heavier subfraction of high density lipoproteins (HDL3) caused a significant decrease in cellular cholesterol. This net removal of cholesterol can be observed in the cells without overloading them with cholesterol, by incubation with low density lipoproteins. Time studies indicated that at least 24 h of incubation is required to detect significant removal of cellular cholesterol. These experiments show that methods using the release of labeled cholesterol from cultured cells to determine net cholesterol removal mediated by high density lipoprotein, although currently used by many investigators, can lead to erroneous conclusions when employed without the measurement of cholesterol mass.     

Density profile and cholesterol concentration of serum lipoproteins in experimental animals and human subjects on hypercholesterolaemic diets

The density profile of Sudan black stained serum lipoproteins was studied in human subjects and various animal species on diets supplemented with cholesterol. In the animals studied (rabbits, calves, mice, chickens, rats and guinea-pigs), the feeding of cholesterol resulted in an elevation of serum cholesterol levels together with marked changes in the density profile and the cholesterol concentration of the serum lipoproteins. Large differences between animal species in their response to dietary cholesterol were found. In a human subject, an increased concentration of serum cholesterol due to the consumption of a diet supplemented with six egg yolks per day was reflected in an elevated level of LDL cholesterol, while changes in the density profile of stained serum lipoproteins were not observed. In subjects with familial type III and type IV hyperlipoproteinaemia, marked differences in the density profile of lipoproteins were found when compared with that of normolipoproteinaemic subjects. The density profile of stained lipoproteins in the type III patients was remarkably similar to that in cholesterol-fed chickens and lean Zucker rats.