Novel Large Apolipoprotein E-Containing Lipoproteins of Density 1.006–1.060 g/ml in Human Cerebrospinal Fluid

Abstract: Although the critical role of apolipoprotein E (apoE) allelic variation in Alzheimer's disease and in the outcome of CNS injury is now recognized, the functions of apoE in the CNS remain obscure, particularly with regard to lipid metabolism. We used density gradient ultracentrifugation to identify apoE-containing lipoproteins in human CSF. CSF apoE lipoproteins, previously identified only in the 1.063–1.21 g/ml density range, were also demonstrated in the 1.006–1.060 g/ml density range. Plasma lipoproteins in this density range include low-density lipoprotein and high-density lipoprotein (HDL) subfraction 1 (HDL₁). The novel CSF apoE lipoproteins are designated HDL₁. No immunoreactive apolipoprotein A-I (apo A-I) or B could be identified in the CSF HDL₁ fractions. Large lipoproteins 18.3 ± 6.6 nm in diameter (mean ± SD) in the HDL₁ density range were demonstrated by electron microscopy. Following fast protein liquid chromatography of CSF at physiologic ionic strength, apoE was demonstrated in particles of average size greater than particles containing apoA-I. The largest lipoproteins separated by this technique contained apoE without apoA-I. Thus, the presence of large apoE-containing lipoproteins was confirmed without ultracentrifugation. Interconversion between the more abundant smaller apoE-HDL subfractions 2 and 3 and the novel larger apoE-HDL₁ is postulated to mediate a role in cholesterol redistribution in brain.     

The distribution and partial characterization of the serum apolipoproteins in the guinea pig.

1. Very-low-density (VLD), low-density (LD) and high-density (HD) lipoproteins were isolated by sequential ultracentrifugation from the serum of male guinea pigs fed on a diet containing 3--4% fat. The apoproteins of these lipoproteins (apo-VLD, apo-LD and apo-HD lipoproteins) were studied after delipidation with organic solvents or extraction with tetramethylurea. 2. The major apolipoprotein of LD lipoprotein isolated by gel filtration was found to closely resemble apolipoprotein B of human serum in its chemical and physical properties. Electrophoresis in sodium dodecyl sulphate-polyacrylamide gel showed that this apoprotein consisted of a number of polypeptides. 3. Tetramethylurea precipitated an apoprotein from guinea-pig serum lipoproteins that is probably the apolipoprotein B-like component. This apoprotein accounted for about 80% of the apo-LD lipoprotein, about 55% of the apo-VLD lipoprotein and about 50% of the apo-HD lipoprotein. 4. The distribution of apolipoproteins soluble in tetramethylurea was determined by densitometric scanning of stained polyacrylamide disc gels. 5. A glycine-rich component of high electrophoretic mobility (band I) and a triplet of soluble apolipoproteins (bands II-IV) were present in both VLD and LD lipoprotein classes. These components constituted a higher proportion of the tetramethylurea-soluble apoproteins of VLD lipoprotein (60--80%) than of LD lipoprotein (40--55%). 6. Small amounts (10--15%) of a component of intermediate mobility, which contained traces of half-cystine, were also present in both VLD and LD lipoproteins. 7. A group of soluble components of basic character (bands VI-X), present as minor components of VLD lipoprotein (10--20%), constituted a major proportion (30--45%) of the soluble apoproteins of LD lipoprotein. Two of these apoproteins were rich in lysine, and two of lower electrophoretic mobility were rich in arginine. 8. The pattern of tetramethylurea-soluble apoproteins in HD lipoprotein was distinguished by the presence of two polypeptides of low electrophoretic mobility as its predominant components. One of these components, band VI, resembled the A-I apolipoprotein of man in both its amino acid profile and in its electrophoretic mobility. The second major component, band VI-B, was rich in lysine and resembled the C-I apolipoprotein of man in amino acid composition. 9. The soluble components of bands I and IX were analogous in physicochemical properties to the R-X1 and R-X2 (high-arginine polypeptide) peptides of human serum lipoproteins respectively.     

Distribution of apolipoprotein E among lipoprotein fractions in the lactating cow

Apolipoprotein (apo) E plays a key role in regulating plasma levels of lipoproteins. We investigated the serum apoE concentrations in cows during different lactating stages by ELISA. To confirm the distribution of apoE in lipoprotein fractions, cow plasma was separated by gel filtration, ultracentrifugation and agarose gel electrophoresis. The apoE concentrations during early, mid- and late lactating stages in cows were significantly higher than that during the non-lactating stage. In lactating plasma, apoE eluted in high-density lipoprotein (HDL) fractions separated by gel filtration increased. The portion of this apoE in plasma was 49%. However, when lactating plasma was separated by ultracentrifugation, less then 5% apoE was recovered in the HDL fraction, and more apoE was recovered in the non-lipoprotein fraction (d>1.21 g/ml, 46%). In agarose gel electrophoresis, plasma apoE was found in β-migrating lipoprotein, but it was not present in α-migrating lipoprotein. To purify apoE-containing particles, the HDL fraction separated by gel filtration was pooled and the fraction retained on Heparin–Sepharose chromatography collected. Cholesterol was absent from this fraction. These results suggest that apoE-containing particles, which increased during the lactating stage, were not associated with HDL particles, and that lipid-free forms were included in cow plasma.     

Microdetermination of cholesterol in serum lipoproteins.

A two-step procedure for the microdetermination of cholesterol in serum lipoproteins is compared with cholesterol quantitation after density gradient ultracentrifugation. Serum lipoproteins from 10 mul of serum are separated by electrophoresis on agarose and visualized by precipitation with dextran sulfate--CaCl2. The lipoprotein bands are cut off from the plates, the agarose slices are hydrolyzed by gas-liquid chromatography. The comparison between the two procedures reveals satisfactory correlations for beta-and pre-beta-lipoproteins and total serum. There is excellent recovery of cholesterol in fractionated lipoproteins.     

Tissue and whole-body protein synthesis in immature Zucker rats and their relationship to protein deposition.

Lipoprotein(a) was purified by agarose-gel chromatography from human plasma from which lipoproteins of Sf greater than 0 had been removed either by sequential or by density-gradient ultracentrifugation. After delipidation, the apoprotein B of this lipoprotein was analysed by sodium dodecyl sulphate/polyacrylamide-gel electrophoresis. It could not be distinguished from the apoprotein B of low-density lipoproteins (rho 1.019-1.063 g/ml). A significant increase in the concentration of apoprotein B in plasma from which the Sf greater than 0 lipoproteins had been removed was observed in six subjects 4 h after a fatty meal.     

Agarose--starch gel electrophoresis of rat serum lipoproteins

Rat serum lipoproteins were separated into at least four fractions by agarose-starch gel electrophoresis. The system used was discontinuous in that glycine and sodium barbitone buffer was used in the reservoirs and Tris buffer was used for the gels. The four major bands could be related to the pattern obtained by ultracentrifugation. The high density lipoproteins consisted of at least two poorly resolved bands and were not separated from albumin. The vertical gel apparatus was further modified to accept 0.4 ml of rat plasma, which was prestained with Sudan black. After electrophoresis the different lipoprotein bands could conveniently be cut out and the lipid phosphorus determined. The addition of Sudan black B decreased the recovery of the low and high density lipoproteins by 5-9%. However, the recovery of phospholipids was reproducible (80 +/- 2%) and the high density lipoproteins contained over two-thirds of the plasma lipid phosphorus.     

Remodeling of apolipoprotein E-containing spherical reconstituted high density lipoproteins by phospholipid transfer protein

Phospholipid transfer protein (PLTP) transfers phospholipids between HDL and other lipoproteins in plasma. It also remodels spherical, apolipoprotein A-I (apoA-I)-containing HDL into large and small particles in a process involving the dissociation of lipid-free/lipid-poor apoA-I. ApoE is another apolipoprotein that is mostly associated with large, spherical HDL that do not contain apoA-I. Three isoforms of apoE have been identified in human plasma: apoE2, apoE3, and apoE4. This study investigates the remodeling of spherical apoE-containing HDL by PLTP and the ability of PLTP to transfer phospholipids between apoE-containing HDL and phospholipid vesicles. Spherical reconstituted high density lipoproteins (rHDL) containing apoA-I [(A-I)rHDL], apoE2 [(E2)rHDL], apoE3 [(E3)rHDL], or apoE4 [(E4)rHDL] as the sole apolipoprotein were prepared by incubating discoidal rHDL with low density lipoproteins and lecithin:cholesterol acyltransferase. PLTP remodeled the spherical, apoE-containing rHDL into large and small particles without the dissociation of apoE. The PLTP-mediated remodeling of apoE-containing rHDL was more extensive than that of (A-I)rHDL. PLTP transferred phospholipids from small unilamellar vesicles to apoE-containing rHDL in an isoform-dependent manner, but at a rate slower than that for spherical (A-I)rHDL. It is concluded that apoE enhances the capacity of PLTP to remodel HDL but reduces the ability of HDL to participate in PLTP-mediated phospholipid transfers.     

A micro-enzymatic method to measure cholesterol and triglyceride in lipoprotein subfractions separated by density gradient ultracentrifugation from 200 microliters of plasma or serum

A micro-enzymatic method was developed to measure total cholesterol (CHOL) and triglyceride (TG) in lipoproteins and their subfractions separated by density gradient ultracentrifugation. This method had a detection limit and sensitivity below 2 mg/dl and accuracy (bias to reference sera) and imprecision (coefficient of variation) of less than 3% between 2 and 30 mg/dl for both CHOL and TG. In addition, the method was in good agreement with standardized Abell-Kendall CHOL (r = 0.98) and enzymatic TG (r = 0.99) methods. Lipoproteins from 200 microliters of plasma or serum were separated by either equilibrium (EQ)- or rate zonal (RZ)-density gradient ultracentrifugation and the resulting fractions were analyzed for CHOL and TG by the micro-enzymatic method. Lipoprotein measurements by these micro-enzymatic/density gradient methods were highly correlated with standardized Lipid Research Clinic (LRC) procedures and preparative ultracentrifugation. The EQ-density gradient procedure also allowed determination of CHOL and TG in LDL and HDL subfractions within any desired density interval. These methods will facilitate the measurements and study of lipoproteins and their subfractions especially in infants, children, the elderly, and small animals. In addition, the micro-enzymatic method may be adapted to other modes of lipoprotein separation such as liquid chromatography, electrophoresis, and precipitation. CHOL or TG determinations could be made on approximately 500 density gradient fractions per hour.     

Analysis of rat serum apolipoproteins by isoelectric focusing. I. Studies on the middle molecular weight subunits.

Analytical isoelectric focusing (IEF) has been applied to the study of the apolipoprotein components of rat serum high density and very low density lipoproteins. The apolipoproteins were separated on 7.5% polyacrylamide gels containing 6.8% urea, with a pH gradient of 4-6. The middle molecular weight range apolipoproteins were identified on IEF gels by the use of apolipoproteins purified by electrophoresis on gels containing sodium dodecyl sulfate (SDS). The A-1 protein focused as 4 to 5 bands from pH 5.46 to 5.82; the A-IV protein and the arginine-rich protein each focused as 4 to 6 bands from pH 5.31 to 5.46. The low molecular weight proteins focused from pH. 4.43 to 4.83 and are the subject of a separate communication. Comparisons of the IEF method with SDS gel electrophoresis, polyacrylamide gel electrophoresis in urea, and Sephadex chromatography are also reported. Additional studies were also carried out that tend to rule out carbamylation or incomplete unfolding of the proteins in the presence of urea as the causes of the observed heterogeneity.     

Preferential interactions between ApoE-containing lipoproteins and Aβ revealed by a detection method that combines size exclusion chromatography with non-reducing gel-shift

The association between apolipoprotein E (apoE) and amyloid-β peptide (Aβ) may significantly impact the function of both proteins, thus affecting the etiology of Alzheimer's disease (AD). However, apoE/Aβ interactions remain fundamentally defined by the stringency of the detection method. Here we use size exclusion chromatography (SEC) as a non-stringent approach to the detection of apoE/Aβ interactions in solution, specifically apoE and both endogenous and exogenous Aβ from plasma, CSF and astrocyte conditioned media. By SEC analysis, Aβ association with plasma and CNS lipoproteins is apoE-dependent. While endogenous Aβ elutes to specific human plasma lipoproteins distinct from those containing apoE, it is the apoE-containing lipoproteins that absorb excess amounts of exogenous Aβ40. In human CSF, apoE, endogenous Aβ and phospholipid elute in an almost identical profile, as do apoE, exogenous Aβ and phospholipid from astrocyte conditioned media. Combining SEC fractionation with subsequent analysis for SDS-stable apoE/Aβ complex reveals that apoE-containing astrocyte lipoproteins exhibit the most robust interactions with Aβ. Thus, standardization of the methods for detecting apoE/Aβ complex is necessary to determine its functional significance in the neuropathology characteristic of AD. Importantly, a systematic understanding of the role of apoE-containing plasma and CNS lipoproteins in Aβ homeostasis could potentially contribute to identifying a plasma biomarker currently over-looked because it has multiple components.     

Heterogeneity of lipoprotein B

Combined very low and low density lipoproteins were derived from human plasma by polyanion precipitation and the low density lipoprotein fraction (density 1.027–1.050 g/ml) was isolated by sequential ultracentrifugation. When this fraction was applied to Sepharose column chromatography, three lipoproteins were eluted. The first and third peaks were minor components while the second peak represented the bulk of LDL. Further chromatographic and electrophoretic studies indicated that the component representing the second peak was heterogeneous. This component was subsequently delipidated at pH 4 in a quaternary biphasic solvent system. The apoproteins remained soluble after delipidation and were treated with various deaggregating agents. On column isoelectric focusing in the presence of 4 M urea the apoproteins banded as broad overlapping peaks between pH 3 and 7. When hexanol was added to the system, distinct apoprotein subfractions were resolved.     

Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions

Procedures are described for the isolation of lipoproteins from human serum by precipitation with polyanions and divalent cations. A mixture of low and very low density lipoproteins can be prepared without ultracentrifugation by precipitation with heparin and either MnCl(2) alone or MgCl(2) plus sucrose. In both cases the precipitation is reversible, selective, and complete. The highly concentrated isolated lipoproteins are free of other plasma proteins as judged by immunological and electrophoretic methods. The low density and very low density lipoproteins can then be separated from each other by ultracentrifugation. The advantage of the method is that large amounts of lipoproteins can be prepared with only a single preparative ultracentrifugation. Polyanions other than heparin may also be used; when the precipitation of the low and very low density lipoproteins is achieved with dextran sulfate and MnCl(2), or sodium phosphotungstate and MgCl(2), the high density lipoproteins can subsequently be precipitated by increasing the concentrations of the reagents. These lipoproteins, containing small amounts of protein contaminants, are further purified by ultracentrifugation at d 1.22. With a single preparative ultracentrifugation, immunologically pure high density lipoproteins can be isolated from large volumes of serum.     

Qualitative and quantitative alterations of bovine serum lipoproteins with ageing

1. Plasma lipoproteins from six calves at 8, 43 and 118 days old, six heifers and six cows were separated by density gradient ultracentrifugation. For each sample lipoproteins bands were visualized by prestained control and characterized by electrophoretic, chemical and morphological analysis. 2. Two resolved bands were detected in the low density lipoprotein fraction (LDL). At an early stage of development, LDLI and LDLII were present with almost equal concentration. With ageing, LDLII became the major fraction of LDL lipoproteins. 3. HDL were isolated as a single band distributed over the range 1.064-1.166 mg/ml in young calf and 1.050-1.152 mg/ml in adult. This progressive decrease of density limits with ageing, associated with a decrease of protein content and an increase of phospholipids and cholesteryl esters content, was consistent with higher HDL particle diameters in adult. 4. With ageing, free cholesterol/esterified cholesterol ratio decreased in LDL fractions and increased in HDL fractions.     

Studies on the protein composition of human serum very low density lipoproteins: demonstration of the beta 2-glycoprotein-I.

Human serum VLDL isolated by polyanion precipitation and ultracentrifugation have been delipidated with ethanal/diethyl ether. By electrophoresis in 10% polyacrylamide gels containing 8M urea, we found a protein which comigrated with apolipoprotein E. This protein was purified by column chromatography and turned out to be identical with beta 2-glycoprotein-I, the serum factor which is necessary for the precipitation of triglyceride-rich lipoproteins with sodium decyl sulfate or sodium dodecyl sulfate. Upon analytical isoelectric focusing, beta 2-glycoprotein-I gave four major bands in the pH region 5.7--6.6. All four bands gave an immunochemical reaction of identity with a monospecific antiserum. From its unique amino acid composition we conclude that beta 2-glycoprotein-I is distinct from all apolipoproteins described previously in the literature.     

Preparative isoelectric focusing of human serum very low-density and low-density lipoproteins.

Ten percent glycerol prevented the usual precipitation of human serum very low-density lipoproteins (VLDL) and low-density lipoproteins (LDL) at their isoelectric points during their preparative isoelectric focusing (IEF), IEF separated VLDL and LDL into two major fractions. The observed optical density peaks are not artifacts caused by binding of Ampholines to VLDL or LDL since no radioactivity accumulated in the fractions containing VLDL or LDL during IEF in the presence of [¹⁴C]Ampholine, and gel filtration completely separated the lipoproteins from [¹⁴C]Ampholine. These results suggest that IEF may separate subspecies of VLDL and LDL under suitable experimental conditions.     

Apolipoproteins as the basis for heterogeneity in high-density lipoprotein2 and high-density lipoprotein3. Studies by isoelectric focusing on agarose films

A method is described for the isoelectric focusing (IEF) of lipoproteins on thin films of agarose. Within a pH gradient of 4.60-5.30 both high-density lipoproteins 2 and 3 (HDL2 and HDL3) are resolved into more than 10 fractions which could be stained either for protein or for lipids. The isoelectric focusing patterns for HDL2 and HDL3 are similar although HDL2 appears richer in the more alkaline bands. Narrow film strips from the IEF separation of HDL2 and HDL3 were interfaced with various agarose plates containing antisera against apolipoproteins apoAI, apoAII and apoCIII either alone or in combination, to provide two-dimensional IEF immunoelectrophoresis patterns. This technique demonstrated that apoAI and apoAII were present throughout the IEF gel for both subclasses of HDL. It also provided evidence for the existence of lipoproteins containing both apoAI and apoAII and other lipoproteins present in the alkaline region of the gel which contained apoAI but no apoAII. ApoCIII was found mostly in acidic lipoproteins and was not distributed identically in HDL2 and HDL3. The lipoproteins separated by IEF on agarose were also analysed by two-dimensional IEF-SDS electrophoresis and the individual apolipoproteins were identified by reaction with antibodies to apolipoproteins AI, AII, CI, CII, CIII, D, and E. This technique confirmed that in IEF of HDL, apoAI extended throughout the spectrum of lipoproteins whereas apoE was only present in alkaline lipoproteins and apoD was only present in acidic lipoproteins. IEF on agarose of either HDL2 or HDL3 allowed us to collect eight different fractions, which have the same pI in either lipoprotein class. The apolipoprotein composition of each isolated band was analysed by electroimmuno-assays for apolipoproteins AI, AII, CI, CII, CIII, D, and E and the results expressed as the ratio of the measured apolipoprotein to measured apoAI. In both HDL2 and HDL3, acidic lipoprotein fractions were enriched in apoAII, apoCIII and apoD. ApoCII and apoCII were not similarly distributed in HDL2 and HDL3 subfractions whereas the apoCI distribution was similar in both classes. Noteworthy in all experiments was the difference in the distributions of apoCI, apoCII, and apoCIII in HDL2 and HDL3, which indicated that the existence of a lipoprotein containing simultaneously CI, CII and CIII can only account for a small fraction of these apolipoproteins. Therefore these experiments substantiate the theory of the protein basis of HDL heterogeneity and suggest that the majority of apolipoproteins are present in complexes which upon IEF result in lipoprotein fractions of identical pI for both HDL2 or HDL3.(ABSTRACT TRUNCATED AT 400 WORDS)     

A rapid micromethod for apolipoprotein E phenotyping directly in serum

A new method for the apolipoprotein E phenotyping has been developed. The method is based on isoelectric focusing of either delipidated or guanidine-HC1-treated serum or plasma in a horizontal slab gel system followed by immunoblotting using either polyclonal or monoclonal anti-apolipoprotein E antibodies as first antibody. Apolipoprotein E phenotyping with this method in 200 serum samples that had been stored at -20 degrees C for more than one year gave exactly the same results as obtained with the conventional method based on isoelectric focusing of delipidated very low density lipoproteins isolated from fresh serum followed by protein staining. Compared with the conventional method, the present method is less laborious because ultracentrifugation to isolate VLDL is not needed; it is suitable for large scale screening purposes; it needs only a few microliters of serum or plasma, and can easily be performed with samples with low concentrations of apolipoprotein E.     

Preferential interactions between ApoE-containing lipoproteins and Aβ revealed by a detection method that combines size exclusion chromatography with non-reducing gel-shift

The association between apolipoprotein E (apoE) and amyloid-β peptide (Aβ) may significantly impact the function of both proteins, thus affecting the etiology of Alzheimer's disease (AD). However, apoE/Aβ interactions remain fundamentally defined by the stringency of the detection method. Here we use size exclusion chromatography (SEC) as a non-stringent approach to the detection of apoE/Aβ interactions in solution, specifically apoE and both endogenous and exogenous Aβ from plasma, CSF and astrocyte conditioned media. By SEC analysis, Aβ association with plasma and CNS lipoproteins is apoE-dependent. While endogenous Aβ elutes to specific human plasma lipoproteins distinct from those containing apoE, it is the apoE-containing lipoproteins that absorb excess amounts of exogenous Aβ40. In human CSF, apoE, endogenous Aβ and phospholipid elute in an almost identical profile, as do apoE, exogenous Aβ and phospholipid from astrocyte conditioned media. Combining SEC fractionation with subsequent analysis for SDS-stable apoE/Aβ complex reveals that apoE-containing astrocyte lipoproteins exhibit the most robust interactions with Aβ. Thus, standardization of the methods for detecting apoE/Aβ complex is necessary to determine its functional significance in the neuropathology characteristic of AD. Importantly, a systematic understanding of the role of apoE-containing plasma and CNS lipoproteins in Aβ homeostasis could potentially contribute to identifying a plasma biomarker currently over-looked because it has multiple components.     

Polypeptides from serum low-density lipoproteins of pigs (Sus domesticus).

Low-density lipoproteins floating between densities 1-006 and 1-063 g cm-3 were isolated by centrifugation of blood serum obtained from 24-h fasted pigs (Sus domesticus). This lipoprotein fraction contained two components with Sf 1-063 values of 3-4 and 2-3 at 20 degrees C when examined by analytical ultracentrifugation. Delipidation of the lipoprotein yielded 15% recovery of soluble protein. Chromatography on Sephadex G100 in 8 M urea of these delipidation products yielded three fractions of different sizes which were present in both native and succinylated apoproteins. These fractions from the succinylated apolipoproteins were further characterized. A polypeptide fraction comprising 70% of the total protein had an apparent molecular weight of 34000 and contained greater amounts of amino acids with hydrophobic side chains than did the second fraction of apparent molecular weight 22000 which contained 15% of the protein. The third fraction of apparent molecular weight 12500 contained 15% of the protein.