Characterization of human high density lipoproteins by zonal ultracentrifugation

Denaturation of human plasma high-density lipoproteins by urea studied by apolipoprotein A-I dissociation

We studied the mechanism of HDL denaturation with concomitant apoA-I dissociation with HDL preparations from 48 patients with a wide range of plasma HDL-C and evaluated the contribution of lipid-free apoA-I into cholesterol efflux from macrophage, in particular, mediated by cholesterol transporter ABCA1. We prepared HDL by precipitation of apoB-containing lipoproteins by polyethylene glycol and used the chaotropic agent urea to denature HDL preparations. Apo-I dissociation from urea-treated HDL was assessed by the increase of preβ-band fraction with agarose gel electrophoresis followed by electro transfer and immunodetection and by the increase of ABCA1-mediated efflux of fluorescent analogue BODIPY-Cholesterol from RAW 264.7 macrophages. The HDL denaturation is governed by a single transition to fully dissociated apoA-I and the transition cooperativity decreases with increasing HDL-C. The apoA-I release depends on phospholipid concentration of HDL preparation and HDL compositional and structural heterogeneity and is well described by apolipoprotein partition between aqueous and lipid phases. Dissociated apoA-I determines the increase of ABCA1-mediated efflux of BODIPY-Cholesterol from RAW 264.7 macrophages to patient HDL. The increase in apoA-I dissociation is associated with the increase of ABCA1 gene transcript in peripheral blood mononuclear cells from patients. The low level of plasma HDL particles may be compensated by their increased potency for apoA-I release, thus suggesting apoA-I dissociation as a new HDL functional property.

Volumetric determination of apolipoprotein stoichiometry of circulating HDL subspecies

Although HDL is inversely correlated with coronary heart disease, elevated HDL-cholesterol is not always protective. Additionally, HDL has biological functions that transcend any antiatherogenic role: shotgun proteomics show that HDL particles contain 84 proteins (latest count), many correlating with antioxidant and anti-inflammatory properties of HDL. ApoA-I has been suggested to serve as a platform for the assembly of these protein components on HDL with specific functions - the HDL proteome. However, the stoichiometry of apoA-I in HDL subspecies is poorly understood. Here we use a combination of immunoaffinity chromatography data and volumetric analysis to evaluate the size and stoichiometry of LpA-I and LpA-I,A-II particles. We conclude that there are three major LpA-I subspecies: two major particles, HDL[4] in the HDL3 size range (d = 85.0 ± 1.2 Å) and HDL[7] in the HDL2 size range (d = 108.5 ± 3.8 Å) with apoA-I stoichiometries of 3 and 4, respectively, and a small minor particle, HDL[1] (d = 73.8 ± 2.1Å) with an apoA-I stoichiometry of 2. Additionally, we conclude that the molar ratio of apolipoprotein to surface lipid is significantly higher in circulating HDL subspecies than in reconstituted spherical HDL particles, presumably reflecting a lack of phospholipid transfer protein in reconstitution protocols.

The relationship between high density lipoprotein subclass profile and apolipoprotein concentrations

The HDL fraction in human plasma is heterogeneous in terms of size, shape, composition, and surface charge. The HDL subclasses contents were quantified by 2-dimensional non-denaturing gel electrophoresis, immunoblotting, and image analysis. This research review systematically analyzed the relationship between the contents of HDL subclasses and the concentrations and ratios of the 5 major plasma apolipoproteins (apo). As the concentration of apoA-I increases, the contents of all HDL subclasses increase significantly. The most significant association was observed between large-sized HDL2b contents and apoA-I. ApoA-II played a dual function in the contents of HDL subclasses, and both small-sized HDL3b and HDL3a and large-sized HDL2b tended to increase with apoA-II concentration. An increase in the concentrations of apoC-II, C-III, and B-100 resulted in higher levels of small-sized HDL particles and lower levels of large-sized HDL particles. Plasma apoB- 100, apoC-II, and apoC-III appear to play a coordinated role in assembly of HDL particles and the determination of their contents. Higher concentrations of apoA-I could inhibit the reduction in content of large-sized HDL2b effected by apoB-100, C-II, and C-III. The preβ1-HDL contents increased significantly and those of HDL2b declined progressively with an increased apoB-100/apoA-I or a decreased apoC-III/apoC-II ratio. In summary, each apo has distinct but interrelated roles in HDL particle generation and metabolism. ApoA-I and apoC-II concentrations are independent determinants of HDL subtypes in circulation and apoA-I levels might be a more powerful factor to influence HDL subclasses distribution. Moreover, apoB- 100/apoA-I ratio could reliably and sensitively reflect the HDL subclass profile.

High density lipoprotein subfractions: isolation, composition, and their duplicitous role in oxidation

The plasma HDLs represent a major class of cholesterol-transporting lipoprotein that can be divided into two distinct subfractions, HDL(2) and HDL(3), by ultracentrifugation. Existing methods for the subfractionation of HDL requires lengthy ultracentrifugations, making them unappealing for large-scale studies. We describe a method that subfractionates HDL from plasma in only 6 h, representing a substantial decrease in total isolation time. The subfractions so isolated were assessed for a variety of lipid and protein components, in addition to their susceptibility to oxidation, both alone and in combination with VLDL and LDL. We report for the first time a prooxidant role for HDL during VLDL oxidation, in which HDL donates preformed hydroperoxides to VLDL in a cholesteryl ester transfer protein (CETP)-dependent process. Examination of the participation of HDL in LDL oxidation has reinforced its classic role as a potent antioxidant. Furthermore, we have also implicated the second major HDL-associated enzyme, LCAT, in these processes, whereby it acts as a potent prooxidant during VLDL oxidation but as an antioxidant during LDL oxidation. Thus, we have identified a potentially duplicitous role for HDL in the pathogenesis of atherosclerosis, attributable to both CETP and LCAT.

Alterations of high density lipoprotein subclasses in obese subjects

The object of this study was to investigate the characteristics of lipid metabolism in obese subjects, with particular emphasis on the alteration of HDL subclass contents and distributions. A population of 581 Chinese individuals was divided into four groups (25 underweight subjects, 288 of desirable weight, 187 overweight, and 45 obese) according to body mass index (BMI). Apoprotein A-I (apoA-I) contents of plasma HDL sub-classes were determined by 2-D gel electrophoresis associated with an immunodetection method. The concentrations of TG and the apoA-I content of pre-beta 1-HDL were significantly higher (P < 0.01 and P < 0.01, respectively), but the levels of HDL cholesterol, and the apoA-I contents of HDL2a and HDL2b were significantly lower (P < 0.01, P < 0.05, and P < 0.01, respectively) in obese subjects than in subjects having a desirable weight. Moreover, with the elevation of BMI, small-sized pre-beta 1-HDL increased gradually and significantly, whereas large-sized HDL2b decreased gradually and significantly. Meanwhile, the variations in HDL subclass distribution were more obvious with the elevation of TG levels in obese as well as overweight subjects. In addition, Pearson correlation analysis revealed that BMI and TG levels were positively correlated with pre-beta 1-HDL but negatively correlated with HDL2b. Multiple regression analysis also showed that TG concentrations were associated independently and positively with high pre-beta 1-HDL and independently and negatively with low HDL2b in obese and overweight subjects. The HDL particle size was smaller in obese and overweight subjects. The shift to smaller size was more obvious with the elevation of BMI and TG, especially TG levels. These observations, in turn, indicated that HDL maturation might be abnormal, and reverse cholesterol transport might be impaired.

Alterations of high-density lipoprotein subclasses in endogenous hypertriglyceridemia

BACKGROUND

To investigate the alterations of high-density lipoprotein (HDL) subclasses in endogenous hypertriglyceridemic subjects.

METHODS

Apolipoprotein A-I contents of plasma HDL subclasses were quantitated by 2-dimensional gel electrophoresis in 236 normolipidemic subjects (including 146 males and 90 females) and 176 endogenous hypertriglyceridemic subjects (including 103 males and 73 females).

RESULTS

Apolipoprotein A-I contents of small-sized pre-beta1-HDL and HDL3a were significantly higher (P < .01 and P < .01, respectively), but those of large-sized HDL2a and HDL2b were significantly lower (P < .01 and P < .01, respectively) in hypertriglyceridemic subjects versus normolipidemic subjects. Moreover, with the elevation of triglyceride levels, small-sized pre-beta1-HDL and HDL3a increased successively; however, large-sized HDL2a and HDL2b decreased successively. Males had significantly higher apolipoprotein A-I contents of small-sized pre-beta1-HDL and HDL3b (P < .05 and P < .05, respectively), but lower contents of large-sized HDL2b (P < .01) than females in both normolipidemic and hypertriglyceridemic subjects.

CONCLUSIONS

The particle size of HDL shifted toward smaller size in hypertriglyceridemic subjects, especially in male subjects. Of note, the shift was more obvious with the elevation of triglyceride levels. The changes mentioned above indicate that HDL maturation might be abnormal and reverse cholesterol transport might be weakened.

Cholesterol and phospholipid efflux from cultured cells

The removal of phospholipids and cholesterol from tissues is the major mechanism mediating the initial assembly of high density lipoproteins (HDL), as well as being the main reason HDL are thought to protect against atherosclerosis. Investigations of the mechanisms of HDL assembly and testing of novel HDL-raising agents typically involve assays to determine phospholipid and/or cholesterol removal or "efflux" from cultured cells. The purpose of this chapter is to describe experimental protocols that can be used in the determination of cholesterol and phospholipid efflux from cultured cells by HDL apolipoproteins for the formation of new HDL particles, and the testing of novel HDL-raising therapies in vitro. A protocol is also provided for determining the size and nature of HDL particles formed in cell-conditioned medium using two-dimensional gel electrophoresis.

Relationship between plasma lipid concentrations and HDL subclasses

BACKGROUND

It is generally accepted that different high-density lipoprotein (HDL) subclasses have distinct but interrelated metabolic functions. HDL is known to directly influence the atherogenic process and changes in HDL subclasses distribution may be related to the incidence and prevalence of atherosclerosis.

METHODS

Apo-AI contents(mg/l) of plasma HDL subclasses were determined by 2-dimensional gel electrophoresis coupled with immunodetection for apo-AI. Four hundred forty-two Chinese adults subjects aged 33 to 78 years were assigned to different groups according to the third Report of NCEP (ATP III) guidelines. The subjects were first divided into 2 groups, normal and high TG, then further classified by plasma TC, HDL-C and LDL-C concentrations. The subjects were also divided into TC desirable and TC high groups.

RESULTS

Apo-A contents of prebeta(1)-HDL were higher while HDL(2b) were lower in high TG subjects vs. the corresponding normal TG subjects according to plasma TC and LDL-C concentrations. With the increase of plasma TC concentrations, apo-AI contents of prebeta(1)-HDL were significantly higher in high TC subgroup vs. TC desirable subgroup in normal TG subjects. With the decrease of HDL-C concentrations, apo-AI contents of HDL(2b) tended to decrease in normal TG subjects. And, with the increases of LDL-C concentration, in normal TG subjects, apo-AI contents of prebeta(1)-HDL and HDL(3b) were significantly higher and those of HDL(2b) were significantly lower in very high LDL-C subgroup vs. LDL-C optimal subgroup. On the other hand, apo-AI contents of prebeta(1)-HDL and HDL(3a) were significantly higher, while HDL(2a) and HDL(2b) were significantly lower in high TG and very high TG subgroup vs. normal TG subgroup within either TC desirable or TC high subjects. In a multivariate linear regression model, TG and TC concentrations were all associated independently and positively with high prebeta(1)-HDL; however, HDL-C were inversely associated with high prebeta(1)-HDL. And TG and TC concentrations were all associated independently and negatively with low HDL(2b), but HDL-C and apo-AI were positively associated with low HDL(2b).

CONCLUSIONS

With the increase of plasma TG, TC, LDL-C or the decrease of plasma HDL-C concentrations, there was a general shift toward smaller-sized HDL, which, in turn, indicates that reverse cholesterol transport might be weakened and HDL maturation might be abnormal. Plasma TG concentration is a more important factor than TC concentration on the changes of HDL subclass distribution. Moreover, when TG is normal and HDL-C decreased, large-size HDL particles tended to decrease.

Alterations of HDL subclasses in hyperlipidemia

BACKGROUND

It is generally accepted that different high-density lipoprotein (HDL) subclasses have distinct but interrelated metabolic functions. HDL is known to directly influence the atherogenic process and changes in HDL subclasses distribution may be related to the incidence and prevalence of atherosclerosis.

METHOD

The relative apolipoprotein (apo)A-I contents (% apoA-I) of plasma HDL subclasses were determined by two-dimensional gel electrophoresis coupled with immunodetection for apoA-I, in 39 hypercholesterolemic (HTC) subjects, 97 hypertriglyceridemic (HTG) subjects and 32 mixed hyperlipidemic (MHL) subjects, and 124 normolipidemic subjects.

RESULTS

The relative apoA-I contents of prebeta(1)-HDL, prebeta(2)-HDL, HDL(3c), HDL(3b) and HDL(3a) significantly increased while HDL(2a) and HDL(2b) significantly decreased in hyperlipidemic subjects. In HTC subjects of hyperlipidemia, the concentrations of prebeta(1)-HDL were significantly lower and HDL(2b) concentrations were significantly higher than in HTG and MHL subjects. In HTG subjects, the concentrations of HDL(3a) were significantly higher and the concentrations of HDL(2b) were lower than in HTC and MHL subjects. In total hyperlipidemic subjects, plasma triglyceride (TG) concentrations showed positive correlation with prebeta(1)-HDL, prebeta(2)-HDL, HDL(3b) and HDL(3a) and negative correlation with HDL(2a) and HDL(2b). The total cholesterol (TC) concentrations showed positive correlation with the relative apoA-I contents of prebeta(1)-HDL and HDL(3b), whereas the HDL-C concentrations showed negative correlation with the relative apoA-I contents of prebeta(1)-HDL and HDL(3a) and positive correlation with those of HDL(2a) and HDL(2b). The relative apoA-I contents of prebeta(1)-HDL, prebeta(2)-HDL, HDL(3b), and HDL(3a) were positively correlated whereas those of HDL(2a) and HDL(2b) were negatively correlated with TG/HDL-C ratio.

CONCLUSION

The particle size of HDL in hyperlipidemic subjects shifted towards smaller sizes, which, in turn, indicates that the maturation of HDL may be abnormal in hyperlipidemic subjects.

Human apolipoprotein A-II inhibits the formation of pre-beta high density lipoproteins

The role of human apolipoprotein A-II (apoA-II) in the remodeling of human high density lipoproteins (HDL) was investigated during incubation of native and reduced-carboxamidomethylated (RCM) HDL3 with a lipoprotein-depleted plasma fraction (LPDP) in the presence of triglyceride-rich particles (TGRP) isolated from Intralipid. Reduction-carboxamidomethylation of HDL3 entirely converts the disulfide-linked apoA-II dimers into monomers, without affecting the structure, composition and particle size distribution of HDL3. Following incubation with LPDP and TGRP, unmodified HDL3 are mainly converted into large, HDL2 particles (diameter: 9.90 +/- 0.07 nm), enriched in triglycerides and depleted of cholesteryl esters. RCM-HDL3 are converted into both large HDL2 (9.86 +/- 0.07 nm) and small (7.53 +/- 0.06 nm) HDL3. The small products are protein-rich and cholesterol-poor, and consist of two different particles: a component with pre-beta mobility, containing only apoA-I, and a component with alpha mobility, containing both apoA-I and apoA-II. Kinetic studies suggest that a two-step process is involved in the formation of small, pre beta-HDL3, by which changes in lipid composition cause alterations in lipoprotein structure/stability, favoring the dissociation of apolipoproteins and reduction of particle size. These findings indicate that apolipoprotein structure is a major determinant of HDL remodeling, apoA-II potentially counteracting the anti-atherogenic properties of apoA-I by inhibiting the formation of small, pre-beta-migrating HDL.

Postprandial changes in high-density lipoprotein composition and subfraction distribution are not altered in patients with insulin-dependent diabetes mellitus

A detailed analysis of postprandial changes in the size, density, composition, and relative proportion of the major high-density lipoprotein (HDL) subfractions, HDL2 and HDL3, was performed in seven normolipidemic patients with insulin-dependent diabetes mellitus (IDDM) in moderate glycemic control and seven age-, sex-, and weight-matched healthy nondiabetic controls. IDDM subjects received an overnight insulin infusion to maintain euglycemia, with an incremental increase in the insulin infusion rate at the time of the test meal (containing 60 g fat/m2). Samples for detailed analysis of HDL by gradient density ultracentrifugation and nondenaturing gradient gel electrophoresis (GGE) were collected at 0, 4, Br, and 12 hours after the test meal. The composition of HDL, HDL2, and HDL3 was significantly altered in the postprandial state in IDDM subjects and controls with an increase in triglyceride content at 4 to 8 hours and a reciprocal decrease in cholesteryl ester, reflecting exchange of lipid constituents of HDL with triglyceride (TG)-rich lipoproteins. In addition, the phospholipid content of the particles increased at 8 hours after the meal. Peak density of HDL2 and HDL3 decreased slightly at 4 to 8 hours, reaching significance only in controls at 8 hours (P < .05), whereas the mean radius size of these subfractions did not change significantly. In controls and IDDM subjects, the ratio of HDL3 to HDL2 at 8 to 12 hours increased significantly (P < .005). Significant differences in the composition, size, density, or subfraction distribution of HDL between subjects with IDDM and controls were not observed following ingestion of the lipid-rich meal. We conclude from these data that in patients with IDDM in moderate glycemic control, there do not appear to be any significant gross abnormalities in postprandial HDL metabolism with respect to the size, density, or compositional changes of HDL particles.

PAF-degrading acetylhydrolase is preferentially associated with dense LDL and VHDL-1 in human plasma. Catalytic characteristics and relation to the monocyte-derived enzyme

In human plasma, platelet activating factor (PAF)-degrading acetylhydrolase (acetylhydrolase) is principally transported in association with LDLs and HDLs; this enzyme hydrolyzes PAF and short-chain forms of oxidized phosphatidylcholine, transforming them into lyso-PAF and lysophosphatidylcholine, respectively. We have examined the distribution, catalytic characteristics, and transfer of acetylhydrolase activity among plasma lipoprotein subspecies separated by isopycnic density gradient ultracentrifugation; the possibility that the plasma enzyme may be partially derived from adherent monocytes has also been evaluated. In normolipidemic subjects with Lp(a) levels < 0.1 mg/mL, acetylhydrolase was associated preferentially with small, dense LDL particles (LDL-5; d = 1.050 to 1.063 g/mL) and with the very-high-density lipoprotein-1 subfraction (VHDL-1; d = 1.156 to 1.179 g/mL), representing 23.9 +/- 1.7% and 20.6 +/- 3.2%, respectively, of total plasma activity. The apparent Km values for PAF of the enzyme associated with such lipoproteins were 89.7 +/- 23.4 and 34.8 +/- 4.5 mumol/L for LDL-5 and VHDL-1, respectively: indeed, the Km value for LDL-5 was some 10-fold higher than that of the light LDL-1, LDL-2, and LDL-3 subspecies, whereas the Km of VHDL-1 was some twofold greater than those of the HDL-2 and HDL-3 subspecies. Furthermore, when expressed on the basis of unit plasma volume, the Vmax of the acetylhydrolase associated with LDL-5 was some 150-fold greater than that in LDL-1 (d = 1.019 to 1.023 g/mL). No significant differences in the pH dependence of enzyme activity or in sensitivity to protease inactivation, sulfydryl reagents, the serine protease inhibitor Pefabloc, or the PAF antagonist CV 3988 could be detected between apo B-containing and apo A-I-containing lipoprotein particle subspecies. Incubation of LDL-1 (Km = 8.4 +/- 2.6 mumol/L) and LDL-2 (d = 1.023 to 1.029 g/mL; Km = 8.4 +/- 3.3 mumol/L) subspecies with LDL-5, in which acetylhydrolase had been inactivated by pretreatment with Pefabloc, demonstrated preferential transfer of acetylhydrolase to LDL-5. Acetylhydrolase transferred to LDL-5 from the light LDL subspecies exhibited a Km of 9.4 +/- 2.2 mumol/L, a value characteristic of the particle donors. Finally, acetylhydrolase (Km = 23.4 +/- 7.6 mumol/L) released by adherent human monocytes in culture was found to bind preferentially to small, dense LDL subspecies upon incubation of Pefabloc-inactivated plasma with monocyte supernatant.(ABSTRACT TRUNCATED AT 400 WORDS)

Cardiovascular risk factors and alcohol consumption in France and Northern Ireland

The levels of different lipid, lipoprotein and haemostatic variables were assessed in 615 control subjects of the ECTIM Study, defined by five groups of alcohol consumption: non-drinkers, 0 < or = 15 g/day, 15 < . < or = 36 g/day, 36 < . < or = 66 g/day and > 66 g/day. After adjustment for age, body mass index, cigarette consumption and country, alcohol consumption was associated with an increase in HDL-cholesterol (0.47 +/- 0.02 to 0.59 +/- 0.01 g/l in non-drinkers and > 66 g/day consumers, mean +/- S.E.M., P < 0.0001), apolipoproteins A-I and A-II (1.37 +/- 0.03 to 1.60 +/- 0.03 g/l and 0.32 +/- 0.01 to 0.41 +/- 0.01 g/l, respectively, P < 0.0001), LpA-I, LpA-I:A-II (0.46 +/- 0.01 to 0.50 +/- 0.01 g/l and 0.75 +/- 0.02 to 0.91 +/- 0.02 g/l, respectively, P < 0.001) and PAi-1 activity (134 +/- 11 to 177 +/- 11 U/ml, P < 0.001). Conversely, no increases were found for total and LDL-cholesterol, triglycerides, apolipoproteins B and C-III, LpE:B, LpC-III:B, fibrinogen and factor VII. Hence, among the lipid and haemostatic variables studied, only HDL parameters and PAi-1 activity were increased by alcohol consumption.

Plasma factors affecting the in vitro conversion of high-density lipoproteins labeled with a non-transferable marker

We studied the in vitro conversion of HDL3 labeled with a radioiodinated diacyl lipid associating peptide (diLAP). DiLAP was previously shown to be nontransferable, which permitted its' use as a reliable marker of HDL particles. DiLAP-labeled HDL3 was incubated for 23 h at 37 degrees C in human or rat plasma or in reconstituted media containing delipidated plasma and/or lipoproteins and/or partially purified CETP. At the end of the incubations, the samples were adjusted to a density of 1.125 g/ml and ultracentrifuged. The two resulting fractions containing HDL2 and HDL3, respectively, were analyzed by gradient gel electrophoresis. Depending upon experimental conditions, diLAP-labeled HDL3 was converted into HDL2b- and/or small HDL3c-like particles. LCAT inhibition and to a lesser extent CETP promoted the formation of small HDL3c. Reactivation of LCAT led to the disappearance of small HDL3c. No HDL3c formed from HDL2 even in the absence of LCAT activity. When the incubations were performed in the presence of 100 mM thimerosal, which inhibited PLTP but not CETP activity, the conversion of diLAP-labeled HDL3 into HDL2 was almost completely blocked. Collective consideration of these data indicates that the formation of small HDL is moderately facilitated by CETP; that small HDL are converted to larger HDL species by LCAT and that the transformation of HDL3 into HDL2 is a process which largely depends upon PLTP activity.

Purification and characterization of recombinant human apolipoprotein A-II expressed in Escherichia coli

We have expressed recombinant human apolipoprotein A-II (apoA-II) in Escherichia coli, as a fusion protein with Schistosoma japonicum glutathione-S-transferase (GST). The GST-AII fusion protein was recovered by affinity chromatography using glutathione as a ligand. After thrombin cleavage and removal of the GST carrier, recombinant apoA-II was obtained in a highly purified form and was exclusively composed of dimeric apoA-II. Kinetics of association to dimyristoylglycerophosphocholine (Myr2GroPCho) vesicles showed that recombinant apoA-II exhibited the same pattern of association as human plasma apoA-II. Electron microscopic analysis of the complexes showed a typical pattern of rouleaux, characteristic of stacked discs, with a diameter similar to that determined by gradient-gel electrophoresis. Circular dichroism measurements showed that the alpha-helical content of both plasma and recombinant apoA-II increased similarly when the proteins associated with Myr2GroPCho vesicles, at the expense of a random-coil structure. Lipid-bound apoA-II consisted of 70-72% alpha helices, suggesting the presence of three 18-residue alpha helices/apoA-II monomer. Cross-linking experiments indicated that Myr2GroPCho complexes contained two molecules dimeric apoA-II/vesicle. Recombinant apoA-II was as efficient as plasma apoA-II in associating with HDL subclasses, and in displacing apoA-I from dipalmitoylglycerophosphocholine/cholesterol/apoA-I complexes, most likely due to its highly ordered secondary structure when associated with Myr2GroPCho vesicles. These findings demonstrate that recombinant apoA-II exhibits the same structural and functional properties as human plasma apoA-II. Thus, the expression system utilized is appropriate to produce mutagenized forms to further structure/function analysis.

Anticoagulant activity of tissue factor pathway inhibitor in human plasma is preferentially associated with dense subspecies of LDL and HDL and with Lp(a)

Human plasma contains a multivalent, Kunitz-type proteinase inhibitor termed tissue factor pathway inhibitor (TFPI), which specifically inhibits the action of the factor VII(a)-tissue factor complex in coagulation. In the present study, we examined the distribution and anticoagulant activity of TFPI among plasma lipoprotein subspecies separated by isopycnic density gradient ultracentrifugation; this procedure permitted the simultaneous fractionation of the major lipoprotein classes (very-low-density lipoprotein [VLDL], intermediate-density lipoprotein [IDL], low-density lipoprotein [LDL], high-density lipoprotein [HDL] 2 and 3, and very-high-density lipoprotein [VHDL]). Studies of eight normolipidemic subjects revealed two major lipoprotein carriers of TFPI activity: dense LDL subspecies (d = 1.039 to 1.063 g/mL) and both dense HDL particles and VHDL (d = 1.133 to 1.190 g/mL), representing 33.8% and 35.9%, respectively, of the total lipoprotein-associated TFPI activity in plasma. TFPI activity was also associated with lipoprotein(a) (Lp[a]), whose density distribution (d = 1.044 to 1.100 g/mL) overlapped that of LDL and HDL2; such association was related to Lp(a)'s particle size and phenotype. VLDL, IDL, and LDL1 through LDL3 (d = 1.019 to 1.039 g/mL), HDL2 (d = 1.063 to 1.100 g/mL), and light subfractions of HDL3 (d = 1.100 to 1.167 g/mL) conveyed only 1.8%, 10%, and 18.5%, respectively, of lipoprotein-associated TFPI activity. Such anticoagulant activity was dependent on the presence of TFPI protein. The dense subspecies of HDL3 (d = 1.133 to 1.167 g/mL) with which TFPI was preferentially associated were small, displayed a cholesteryl ester to protein ratio of approximately 0.2, and were deficient in phospholipid (13.6% to 18.3%). HDL subspecies of d = 1.110 to 1.167 g/mL mainly contained the higher relative molecular mass form of TFPI of 41 kD (a form that is known to be covalently associated with apolipoprotein [apo] A-II) and minor bands of the 35- and 52-kD forms. The second major localization of TFPI was within the hydrated density range of small, dense LDL particles (d = 1.033 to 1.063 g/mL), which in comparison with light LDL (d = 1.019 to 1.033 g/mL) exhibited a markedly lower proportion of triglyceride and enrichment in cholesteryl ester.(ABSTRACT TRUNCATED AT 400 WORDS)

High-density lipoprotein subfractions

High-density lipoprotein (HDL) consists of a heterogeneous group of particles defined either by size or by apolipoprotein content. Subfractions of HDL appear to have distinct but interrelated metabolic functions, including facilitation of cholesteryl ester transfer to low- and very-low-density lipoproteins, modulation of triglyceride-rich particle catabolism, and, possibly, removal of cholesterol from peripheral tissues. Like HDL cholesterol, HDL subfractions are widely affected by a variety of factors. Subfractions also are markers for epidemiologic risk for coronary artery disease. Because they provide information about the physiologic processes of cholesterol metabolism, HDL subfractions are emerging as an increasingly important tool in the study of the relationship between lipids and cardiovascular disease.

Identification of a distinct human high-density lipoprotein subspecies defined by a lipoprotein-associated protein, K-45. Identity of K-45 with paraoxonase

In an attempt to provide immunological tools for subfractionation of high-density lipoproteins (HDL), monoclonal antibodies were raised against HDL complexes. Two clones identified a peptide, provisionally named K-45 (pI 4.5-4.9; molecular mass 45 kDa, range 42-48 kDa), whose plasma distribution and lipoprotein association were fully characterised. Gel filtration localised the peptide to the HDL region of human plasma where it co-eluted with apolipoprotein (apo) A-I, the structural protein of HDL. Complementary studies employing immunoabsorption with anti-(apo A-I) antibodies removed 90% of K-45 from plasma: conversely, anti-(apo A-II) antibodies eliminated only 10% of K-45. Immunoaffinity chromatography on an anti-(K-45) column revealed that the peptide was present in a distinct HDL subsepecies containing three major proteins: K-45, apo A-I and clusterin or apo J. The lipoprotein nature of the bound fraction was indicated by electron microscopy (diameter 9.6 +/- 3.3 nm) and quantification of lipids, the latter showing an unusually high triacyglycerol concentration. Plasma concentrations of K-45 were positively correlated with apo A-I and HDL-cholesterol and negatively correlated with apo B and total cholesterol. Thus, the peptide appears to be linked, directly or indirectly, to processes which give rise to an anti-atherogenic lipid profile. After completion of the present studies, an N-terminal sequence identical to that of K-45 was reported in recently isolated cDNA clones. These clones encode paraoxonase.

High density lipoprotein metabolism in the horse (Equus caballus)

1. Apolipoprotein A-I dependent lecithin:cholesterol acyl transferase (LCAT) activity was identified in equine lipoprotein deficient plasma (LPDP). 2. LCAT activity showed no breed or sex variation, and was unaltered postprandially. 3. There was no significant cholesteryl ester transfer activity in equine LPDP. 4. Hydrophobic interaction chromatography on phenyl sepharose failed to unmask transfer activity or identify an inhibitor of cholesteryl ester transfer. 5. In 12 Shetland ponies, plasma high density lipoprotein (HDL) concentrations were positively correlated with those of triglyceride, but not with the activities of LCAT, lipoprotein lipase or hepatic lipase.

Protein transfer between A-I-containing lipoprotein subpopulations: evidence of non-transferable A-I in particles with A-II

Transfer of apolipoproteins (apo) between the two subpopulations of apo A-I-containing lipoproteins in human plasma: those with A-II [Lp(AI w AII)] and those without [Lp(AI w/o AII)], were studied by observing the transfer of 125I-apo from a radiolabeled subpopulation to an unlabeled subpopulation in vitro. When Lp(AI w AII) was directly radioiodinated, 50.3 +/- 7.4 and 19.5 +/- 7.7% (n = 6) of the total radioactivity was associated with A-I and A-II, respectively. In radioiodinated Lp(AI w/o AII), 71.5 +/- 6.8% (n = 6) of the total radioactivity was A-I-associated. Time-course studies showed that, while some radiolabeled proteins transferred from one population of HDL particles to another within minutes, at least several hours were necessary for transfer to approach equilibrium. Incubation of the subpopulations at equal A-I mass resulted in the transfer of 51.8 +/- 5.0% (n = 4) of total radioactivity from [125I]Lp(AI w/o AII) to Lp(AI w AII) at 37 degrees C in 24 h. The specific activity (S.A.) of A-I in the two subpopulations after incubation was nearly identical. Under similar incubation conditions, only 13.4 +/- 4.6% (n = 4) of total radioactivity was transferred from [125I]Lp(AI w AII) to Lp(AI w/o AII). The S.A. of A-I after incubation was 2-fold higher in particles with A-II than in particles without A-II. These phenomena were also observed with iodinated high-density lipoproteins (HDL) isolated by ultracentrifugation and subsequently subfractionated by immunoaffinity chromatography. However, when Lp(AI w AII) radiolabeled by in vitro exchange with free [125I]A-I was incubated with unlabeled Lp(AI w/o AII), the S.A. of A-I in particles with and without A-II differed by only 18% after incubation. These data are consistent with the following: (1) in both populations of HDL particles, some radiolabeled proteins transferred rapidly (minutes or less), while others transferred slowly (hours); (2) when Lp(AI w AII) and Lp(AI w/o AII) were directly iodinated, all labeled A-I in particles without A-II were transferable, but some labeled AI in particles with A-II were not; (3) when Lp(AI w AII) were labeled by in vitro exchange with [125I]A-I, considerably more labeled A-I were transferable. These observations suggest the presence of non-transferable A-I in Lp(AI w AII).

Regulation of PGI2 activity by serum proteins: serum albumin but not high density lipoprotein is the PGI2 binding and stabilizing protein in human blood

Although previous studies have shown that serum albumin binds PGI2 and protects it from rapid degradation, it remains debatable whether it is physiologically important due to its low binding affinity for PGI2. We were intrigued by the observations of Yui et al. (J. Clin. Invest. 82 (1988) 803-807) which suggested that apo A-I of the high density lipoprotein (HDL) is the "serum PGI2 stabilizing factor". To clarify this, we carried out experiments to determine the binding kinetics and parameters of HDL and albumin purified from normal pooled human serum. Despite the use of multiple binding assays, we could not detect any binding activity in HDL2, HDL3 or nascent HDL preparations, nor could we demonstrate any PGI2 protecting activity by these molecules. By contrast, purified albumin exhibited essentially identical binding parameters as the native serum from which the albumin was purified. The binding activity of various albumin preparations was not due to the contamination of apo A-I. Computer simulation analysis also failed to provide evidence to support the notion that HDL bound and prolonged PGI2 activity. To determine whether physiological concentrations of albumin influence PGI2 binding to platelet receptors, we measured PGI2 binding to platelet membrane in the absence and presence of albumin. Albumin at 40 mg/ml increased the KD of PGI2 binding to the receptors by 2-3 fold. These findings indicate that albumin plays a major role in protecting PGI2 activity and regulating its availability for platelet PGI2 receptors.

Cholesterol efflux from macrophages mediated by high-density lipoprotein subfractions, which differ principally in apolipoprotein A-I and apolipoprotein A-II ratios

High-density lipoprotein (HDL) was fractionated by preparative isoelectric focussing into six distinct subpopulations. The major difference between the subfractions was in the molar ratio of apolipoprotein A-I to apolipoprotein A-II, ranging from 2.1 to 0.5. The least acidic particles had little apolipoprotein A-II, were larger and contained the most lipid. The efflux capacity of the HDL subfractions was tested with mouse peritoneal macrophages and a mouse macrophage cell line (P388D1), either fed with acetylated low-density lipoprotein or free cholesterol. All the HDL subfractions were equally able to efflux cholesterol. The efflux was concentration dependant and linear for the first 6 h. The HDL subfractions bound with high affinity (Kd = 6.7-7.9 micrograms/ml) at 4 degrees C to the cell surface of P388D1 cells (211,000-359,000 sites/cell). Ligand blotting showed that all the HDL subfractions bound to membrane polypeptides at 60, 100, and 210 kDa. These HDL binding proteins may represent HDL receptors. In summary HDL particles, which differed principally in ratio of apolipoprotein A-I to apolipoprotein A-II behaved in a similar manner for both cholesterol efflux and cell surface binding.

Cholesterol-induced alteration of apolipoprotein A-I conformation in reassembled high density lipoprotein

Recent reports have shown that apolipoprotein A-I (apo A-I), the major protein of high density lipoprotein (HDL) may exist in different conformational states. We studied the effects of apolipoprotein A-II and/or cholesterol on the conformation of apo A-I in reassembled HDL. Analysis of tryptophan fluorescence quenching in the presence of iodine suggested that cholesterol increased the number of apo A-I tryptophan residues accessible to the aqueous phase, but decreased their mean degree of hydration. These observations cannot be totally explained on the basis of the effect of cholesterol on phospholipid viscosity as determined by fluorescence anisotropy of diphenyl hexatriene. We did not observe any effect of apo A-II on the conformation of apo A-I.

Genetic and environmental contributions to cholesterol and its subfractions in 11-year-old twins. The Medical College of Virginia Twin Study

We conducted a cross-sectional analysis of the genetic and environmental contributions to the variance of lipoprotein cholesterol and its subfractions in children during early adolescence. Univariate path analysis was used to determine the relative contributions of genes, individual environment, and family environment to these measures in 233 11-year-old Caucasian twin pairs. For high density lipoprotein, high density lipoprotein2, low density lipoprotein, very low density lipoprotein, and triglycerides, a model that incorporated genes and individual environmental variation but not common environment was sufficient to explain the variation. Different magnitudes of genetic effects were seen for total cholesterol in boys and girls. High density lipoprotein3 showed different magnitudes by sex for genetic and individual environmental effect. Intermediate density lipoprotein was the only cholesterol subfraction in which shared, or common, environment was found to make a statistically significant contribution to the variation.

Reverse cholesterol transport: physiology and pharmacology

Reverse cholesterol transport identifies a series of metabolic events resulting in the transport of cholesterol from peripheral tissues to the liver and plays a major role in maintaining cholesterol homeostasis in the body. High density lipoproteins (HDL) are the vehicle of cholesterol in this reverse transport, a function believed to explain the inverse correlation between plasma HDL levels and atherosclerosis. An attempt to stimulate, by the use of drugs, this transport process seems to be of great promise in the prevention and treatment of arterial disease. Only few drugs are now known that can modify the activity of the various factors involved in the process. Clofibrate reduces cholesterol esterification, but the newer fibric acids are generally ineffective as anion-exchange resins. Probucol directly increases the activity and mass of cholesteryl ester transfer protein, thus possibly improving the physiological process of cholesterol removal from tissues. The few available data on the effects of drugs on reverse cholesterol transport should stimulate the search for new agents specifically stimulating this antiatherogenic process.

Effect of chronic ethanol feeding on high density lipoprotein subfractions in rats

We have reported previously that chronic alcohol consumption in the rat produced elevated total serum high density lipoprotein (HDL) fraction, but HDL particles of the alcohol-fed rat were deficient in apolipoprotein (apo) E. In that report, serum HDL particles were prepared by successive ultracentrifugation method and there were concerns that the apo E deficiency in HDL particles was artificially produced by centrifugal forces. In the present report, apo Al affinity column chromatography was used instead of successive ultracentrifugation and it likewise yielded HDL particles from alcohol-fed rats that exhibited lower apo E: apo Al ratio than HDL from control rats (0.185 +/- 0.016 vs. 0.303 +/- 0.017, respectively). When the total serum lipoprotein fraction (d less than 1.21) was analyzed by high performance liquid chromatography (HPLC), both HDL and VLDL peaks were higher in alcohol-fed rats than controls. The size of apo E deficient HDL particles from alcohol-fed rats determined by HPLC did not differ from that of normal HDL particles. When HDL (1.063 less than d less than 1.21) was subfractionated into HDL2 (1.063 less than d less than 1.125) and HDL3 (1.125 less than d less than 1.21), only HDL2 of alcohol-fed rats showed lowered apo E: apo Al ratio when compared with same HDL subfraction of control animals. Therefore, the molecular structure of only HDL2 (but not HDL3) was affected by alcohol-feeding. Another HDL subpopulation which is enriched with apo E, i.e. HDL1 (1.054 less than d less than 1.063), was also prepared.(ABSTRACT TRUNCATED AT 250 WORDS)

Lipoprotein and hepatic lipase activity and high-density lipoprotein subclasses after cardiac transplantation

Atherosclerosis is the leading obstacle to long-term survival in cardiac transplant patients. Increases in plasma triglycerides and lipoprotein cholesterol levels occur after transplantation that may contribute to transplant atherosclerosis. The etiology of this increase is unclear. We investigated the interaction of immunosuppressive medications with plasma triglycerides, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, the HDL subclasses HDL2 and HDL3 cholesterol, and hepatic and lipoprotein lipase activity in 72 consecutive cardiac transplant patients compared to 51 healthy control subjects. In the transplantation group, greater concentrations of plasma triglyceride (80%, p less than 0.001), LDL cholesterol (16%, p less than 0.005) and hepatic lipase activity (100%, p less than 0.001) were noted, whereas lipoprotein lipase activity was noted to be significantly lower (124%, p less than 0.001). No difference was detected in HDL, HDL2, or HDL3 cholesterol. Cyclosporine dose was significantly associated with hepatic lipase activity (r = 0.33, p less than 0.02) and inversely associated with lipoprotein lipase activity (r = -0.28, p less than 0.05). Lipoprotein lipase activity after transplantation correlated inversely with triglycerides (r = -0.36, p less than 0.002) and positively with HDL cholesterol (r = 0.23, p less than 0.05) and HDL2 cholesterol (r = 0.29, p less than 0.05). Hepatic lipase activity correlated inversely with LDL cholesterol (r = -0.21, p less than 0.08). In multiple regression analysis, cyclosporine dose was the major source of variation in hepatic lipase activity.

Distribution of HDL2 and HDL3 in a random population of healthy French males and females--evaluation by a two-step precipitation procedure

A rapid and reliable method for the determination of HDL2 and HDL3 cholesterol is described. The Apo B containing fractions (VLDL, IDL, LDL) were precipitated by addition of dextran sulfate (Mr 500,000) to 2 mmol/l final concentration followed by MgCl2 to a final concentration of 0.05 mol/l. The supernatant, was brought to 6 mmol/l dextran sulfate and 0.250 mol/l MgCl2 to precipitate HDL2. Cholesterol determination on total serum and both supernatants yielded the concentrations of Apo B-associated cholesterol, total HDL, HDL2 and HDL3 cholesterol. The application of this technique to a random population of healthy French people gave HDL-cholesterol values of 1.35 and 1.54 mmol.l-1, respectively, in 93 males and 95 females (p less than 0.001). All of the difference was attributable to HDL2 (0.43 vs 0.65 mmol.l-1, p less than 0.001) while HDL3 were almost identical at 0.92 and 0.91 nmol.l-1. These values are in good agreement with previously reported figures for French individuals, but markedly higher in males than values reported from North America.

The effect of triphasic oral contraceptives on plasma lipids and lipoproteins

To determine the effect of triphasic oral contraceptives on plasma lipid transport, 150 nonsmoking women with normolipidemia, ages 18 to 35 years, were randomly assigned to receive one of three contraceptive formulations: (1) ethinyl estradiol, 30, 40, and 30 micrograms/day, each for 6, 5, and 10 days per menstrual cycle, and levonorgestrel, 50, 75, and 125 micrograms/day, each for 6, 5, and 10 days; (2) ethinyl estradiol, 35 micrograms/day for 21 days, and phased norethindrone, 500, 750, and 1000 micrograms/day each for 7 consecutive days; and (3) ethinyl estradiol, 35 micrograms/day for 21 consecutive days, and norethindrone, 500, 1000, and 500 micrograms/day for 7, 9, and 5 days, respectively. A control group consisting of 49 women taking a nonhormonal form of contraception was also included. After 6 months of oral contraceptive treatment, significant increases in plasma triglyceride (28% to 52%) and plasma apolipoprotein B levels (20% to 23%) were observed in each treatment group. The changes in total plasma cholesterol (3% to 10%) and low-density lipoprotein cholesterol values (0% to 11%) were less striking. Changes in total high-density lipoprotein cholesterol levels were statistically insignificant (-2% to -4%); however, high-density lipoprotein2 cholesterol levels decreased by 29% to 33% and high-density lipoprotein3 cholesterol levels increased by 20% to 23%. Concomitantly, plasma apoliporprotein A-1 values increased by 5% to 12%. No consistent significant differences among analyses were observed between and of the groups receiving different oral contraceptives for 6 months.

Quantitation and localization of apolipoproteins [a] and B in coronary artery bypass vein grafts resected at re-operation

Lp[a] is a lipoprotein whose plasma concentration is highly correlated with cardiovascular disease. Its protein moiety, apoLp[a], consists of two large polypeptides, apo[a] and apo B. The possible contribution of Lp[a] to atherosclerosis in saphenous vein aortocoronary bypass grafts was studied in a population of patients undergoing coronary re-bypass surgery. The vein graft tissue levels of apoLp[a] were compared with graft duration, gross and light microscopic pathology, as well as plasma levels of apoLp[a]. The localization pattern of apo[a] and apo B in vein graft tissue was determined. In addition, the plasma levels of cholesterol, triglycerides, apoproteins (apo) A-I, A-II, and E were measured. In a representative subpopulation of 17 patients with a mean age of 63 years from whom grafts with a mean duration of 112 months were resected, the mean total plasma cholesterol level was 221 mg/dl, the mean high density lipoprotein cholesterol level was 31 mg/dl, and the mean plasma triglyceride level was 228 mg/dl. In normal saphenous veins, the level of apoLp[a] was below measurable limits (less than 2 ng/mg), and the level of apo B was very low (3.3 ng/mg). In resected grafts, the mean tissue level of apoLp[a] was 32 ng/mg, and that of apo B was 70 ng/mg, demonstrating the net accumulation of these apoproteins in veins from the time of their grafting into the arterial bed. The apoLp[a]/apo B ratio was determined in 77 tissue segments from 59 grafts (28 patients) and was found to be 0.313. This tissue ratio was significantly higher (p = 0.02) than the plasma apoLp[a]/apo B ratio from these patients, which was 0.132. Immunostaining showed co-localization of apo[a] and apo B in the neointima of grafts. The most abundant pathologic features observed in resected grafts were proliferated intima (43/52), thrombus (28/52), and atherosclerotic core regions (21/52). The level of tissue apo B correlated well with the abundance of core regions (r = 0.501), whereas the level of tissue apoLp[a] did not correlate as well with this feature (r = 0.233). Although apo[a] and apo B are almost absent from normal saphenous vein, these apoproteins (and presumably the lipoproteins Lp[a] and low density lipoprotein) accumulate in bypass vein grafts. The data support the view that these lipoproteins play a significant role in vein graft atherosclerosis.