Microarray expression profiling identifies genes with altered expression in HDL-deficient mice.

Based on the assumption that severe alterations in the expression of genes known to be involved in high-density lipoprotein (HDL) metabolism may affect the expression of other genes, we screened an array of >5000 mouse expressed sequence tags for altered gene expression in the livers of two lines of mice with dramatic decreases in HDL plasma concentrations. Labeled cDNA from livers of apolipoprotein AI (apoAI)-knockout mice, scavenger receptor BI (SR-BI) transgenic mice, and control mice were cohybridized to microarrays. Two-sample t statistics were used to identify genes with altered expression levels in the knockout or transgenic mice compared with control mice. In the SR-BI group we found nine array elements representing at least five genes that were significantly altered on the basis of an adjusted P value < 0.05. In the apoAI-knockout group, eight array elements representing four genes were altered compared with the control group (adjusted P < 0.05). Several of the genes identified in the SR-BI transgenic suggest altered sterol metabolism and oxidative processes. These studies illustrate the use of multiple-testing methods for the identification of genes with altered expression in replicated microarray experiments.

Increased low density lipoprotein degradation in aorta of irradiated mice is inhibited by preenrichment of low density lipoprotein with alpha-tocopherol.

We previously reported that upper thoracic exposure to ionizing radiation (IR) accelerates fatty streak formation in C57BL/6 mice and that such effects are inhibited by overexpression of the antioxidant enzyme CuZn-superoxide dismutase (SOD). Notably, IR-accelerated lesion formation is strictly dependent on a high fat diet (i.e., atherogenic lipoproteins) but does not involve alterations in circulating lipid or lipoprotein levels. We thus proposed that IR promotes changes in the artery wall that enhance the deposition of lipoprotein lipids. To address this hypothesis, we examined the effects of IR on aortic accumulation and degradation of low density lipoproteins (LDL). Ten-week-old C57BL/6 mice were exposed to a single (8-Gy) dose of (60)Co radiation to the upper thoracic area or were sham irradiated (controls) and were then placed on the high fat diet. Five days postexposure, the mice received either (125)I-labeled LDL ((125)I-LDL) (which was used to measure intact LDL) or (125)I-labeled tyramine cellobiose ((125)I-TC)-LDL (which was used to measure both intact and cell-degraded LDL) via tail vein injection. On the basis of trichloroacetic acid (TCA)-precipitable counts in retroorbital blood samples, > or =95% of donor LDL was cleared within 24 h and there were no differences in time-averaged plasma concentrations of the two forms of LDL among irradiated and control mice. Aortic values increased markedly within the first hour and thereafter exhibited a slow increase up to 24 h. There were no differences between irradiated and control mice at 1 h, when values primarily reflected LDL entry, but a divergence was observed thereafter. At 24 h, (125)I-TC-associated counts were 1.8-fold higher in irradiated mice (P = 0.10). In contrast, (125)I-LDL-associated counts were 30% lower in irradiated mice (P< 0.05), suggesting that most of the retained (125)I-TC was associated with LDL degradation products. Consistent with the proposed involvement of oxidative or redox-regulated events, IR-induced LDL degradation was lower in SOD-transgenic than wild-type mice (P<0.05). The importance of LDL oxidation was suggested by observations that IR-induced LDL degradation was significantly reduced by preenriching LDL with alpha-tocopherol. On the basis of these results, we propose that IR elicits SOD-inhibitable changes in the artery wall that enhance LDL oxidation and degradation leading to the deposition of LDL-borne lipids. These studies provide additional support for the role of oxidation in lipoprotein lipid deposition and atherogenesis and suggest that IR promotes an arterial environment that stimulates this process in vivo.

Ionizing radiation accelerates aortic lesion formation in fat-fed mice via SOD-inhibitable processes.

Ionizing radiation promotes formation of reactive oxygen species, including the superoxide anion (O2-). To evaluate whether O2- or O2--mediated perturbations may contribute to the known atherogenic effects of radiation, we examined aortic lesion formation in irradiated C57BL/6 mice and evaluated the effects of CuZn-superoxide dismutase (CuZn-SOD) overexpression. Ten-week-old mice were exposed to a 2-, 4-, or 8-Gy dose of 250-keV x-rays to the upper thorax and then placed on a high-fat diet for 18 weeks. Based on quantitative lipid staining of serial sections of the proximal aorta, mean lesion area was increased with increasing radiation dose and was 3-fold greater in 8-Gy-irradiated than sham-irradiated mice (7800+/-2140 versus 2635+/-709 micrometer(2), P<0.05). These effects were absolutely dependent on a high-fat diet, which had to be introduced within 1 to 2 weeks of the radiation exposure, suggesting the early involvement of atherogenic lipoproteins that were elevated in response to the diet. The importance of radiation-induced oxidative stress was supported by the observation of a 2-fold lower mean lesion area in irradiated CuZn-SOD transgenic mice than in their irradiated, nontransgenic littermates (3026+/-1590 versus 6102+/-1834 micrometer(2), P<0.05). Lucigenin-enhanced chemiluminescence, used as an index of aortic O2- concentrations, was significantly elevated in the postradiation period, and this response was reduced in CuZn-SOD transgenics. On the basis of these results, we propose that radiation may be a useful tool for initiating oxidative or redox-regulated events that promote atherogenesis and for testing the antiatherogenic properties of antioxidants.

Fatty streak formation in fat-fed mice expressing human copper-zinc superoxide dismutase.

Studies in vitro have shown that copper-zinc superoxide dismutase (CuZn-SOD) inhibits a number of events putatively involved in atherogenesis, including cell-mediated oxidation of LDL. To investigate whether increased activity of CuZn-SOD reduces atherogenesis in vivo, we examined diet-induced fatty streak formation in CuZn-SOD transgenic mice (n = 24) as compared with their nontransgenic littermates (n = 28). Transgenic animals were originally created by introduction of an EcoRI-BamHI human genomic DNA fragment containing the CuZn-SOD gene and its regulatory elements into B6SJL zygotes. For the current studies, the transgene was bred for 12 generations into the atherosclerosis-susceptible C57BL/6 background. Animals were fed atherogenic diets (15% fat, 1.25% cholesterol, 0.5% Na cholate) starting at 100 weeks of age and extending for 18 weeks. At the end of the diet period, aortic SOD activity was two-fold higher in transgenics than nontransgenics (mean +/- SE: 46.7 +/- 5.8 versus 20.1 +/- 2.4 units/mg of protein, P < .001). Levels of protein-bound amino acid oxidation products (meta-, ortho-, and dityrosine) were either similar or lower in aorta and heart from transgenics as compared with nontransgenics, suggesting that amplification of CuZn-SOD activity above the normal complement had modest inhibitory effects on basal oxidative stress in these tissues. CuZn-SOD overexpression did not reduce the extent of lesion development as analyzed by quantitative lipid staining of serial sections of the proximal aorta; mean lesion areas (+/- SE) were 997 +/- 478 and 943 +/- 221 mu 2 in transgenics and nontransgenics, respectively. Notably, the range of values for lesion area was 2.2-fold greater in transgenics (0-8403 versus 0-3868 mu 2 in nontransgenics). Moreover, within this group, lesion area showed a significant positive correlation with SOD activity (r = .611, P < .03). These results do not support an antiatherogenic effect of Cu-Zn-SOD over expression and raise the possibility that high tissue SOD activity may potentiate atherogenesis in fat-fed atherosclerosis-susceptible mice [corrected].

Selective resistance of LDL core lipids to iron-mediated oxidation. Implications for the biological properties of iron-oxidized LDL.

Although the nature and consequences of oxidative changes in the chemical constituents of low density lipoproteins (LDLs) have been extensively examined, the physical dynamics of LDL oxidation and the influence of physical organization on the biological effects of oxidized LDLs have remained relatively unexplored. To address these issues, in the present studies we monitored surface- and core-specific peroxidative stress relative to temporal changes in conjugated dienes (CDs), particle charge (an index of oxidative protein modification), and LDL-macrophage interactions. Peroxidative stress in LDL surface and core compartments was evaluated with the site-specific, oxidation-labile fluorescent probes parinaric acid (PnA) and PnA cholesteryl ester (PnCE), respectively. When oxidation was initiated by Cu2+, oxidative loss of the core probe (PnCE) closely followed that of the surface probe (PnA), as indicated by the time to 50% probe depletion (t1/2; 15.5 +/- 7.8 and 30.4 +/- 12 minutes for PnA and PnCE, respectively). Both probes were more resistant in LDL exposed to Fe3+ (t1/2, 53.2 +/- 8.1 and 346.7 +/- 155.4 minutes), although core probe resistance was much greater with this oxidant (PnCE t1/2/PnA t1/2 5.8 vs 2.0 for Cu2+). Despite differences in the rate and extent of oxidative changes in Cu(2+)- versus Fe(3+)-exposed LDLs, PnCE loss occurred in close correspondence with CD formation and appeared to precede changes in particle charge under both conditions. Exposure of LDLs to hemin, a lipophilic Fe(3+)-containing porphyrin that becomes incorporated into the LDL particle, resulted in rapid loss of PnCE and simultaneous changes in particle, charge, even at concentrations that yielded increases in CDs and thiobarbituric acid-reactive substances similar to those obtained with free Fe3+. These results suggest that oxidation of the LDL hydrophobic core occurs in conjunction with accelerated formation of CDs and may be essential for LDL protein modification. In accordance with the known effects of oxidative protein modifications on LDL receptor recognition, exposure of LDLs to Cu2+ and hemin but not Fe3+ produced particles that were readily processed by macrophages. Thus, the physical site of oxidative injury appears to be a critical determinant of the chemical and biological properties of LDLs, particularly when oxidized by Fe3+.

Contrasting in vivo effects of murine and human apolipoprotein A-II. Role of monomer versus dimer.

The role of apolipoprotein A-II (apoA-II) in high density lipoprotein (HDL) structure and metabolism has been studied previously in transgenic mice overexpressing either human or murine apoA-II. These studies have shown differences between these two groups of transgenic animals in the levels of very low density, low density, and high density lipoproteins, in the HDL particle size distribution, and in the relationship between apoA-II levels and lipoprotein levels. To determine whether these differences are due to the fact that human apoA-II is dimeric and murine apoA-II monomeric, we have examined the effects of monomeric human apoA-II (hA-IImon) in transgenic mice. Site-directed mutagenesis (Cys6 -> Ser) was used to generate 15 transgenic founder lines of hA-IImon mice, that contained plasma hA-IImon concentrations over a 10-fold range (11 mg/dl to 185 mg/dl). The hA-IImon floated in the d < or = 1.21 g/ml fraction and migrated as an apoA-II monomer by nonreducing SDS-polyacrylamide gel electrophoresis. HDL levels were not correlated with hA-IImon levels (r = -0.26); HDL particle size and size distribution, as well as very low density and low density lipoprotein levels and sizes, were unchanged compared to nontransgenic control mice. These results suggest that differences between mice overexpressing human dimeric apoA-II and those overexpressing murine apoA-II are the result of sequence differences between these two apoA-II molecules and are not solely due to the fact that human apoA-II exists as a dimer.

HDL antioxidant effects as assessed using a nonexchangeable probe to monitor particle-specific peroxidative stress in LDL-HDL mixtures.

High density lipoproteins (HDL) have been reported to inhibit oxidation of low density lipoproteins (LDL) based in part on observations that oxidative changes occur more slowly in LDL-HDL mixtures than in LDL alone. In the current studies, we developed an approach to discern particle-specific oxidation kinetics within mixed particle systems using the oxidation-labile fluorescent probe parinaric acid cholesteryl ester (PnCE) and applied this to the study of HDL inhibition effects. PnCE was introduced into acceptor lipoproteins by cholesteryl ester transfer protein (CETP)-mediated transfer from donor microemulsions. Incubation of PnCE-containing LDL and HDL with non-probe-containing HDL and LDL, respectively, followed by measurement of reisolated fractions, indicated that PnCE does not transfer appreciably between lipoprotein fractions. Oxidative loss of lipoprotein-associated PnCE occurred essentially in tandem with changes in conjugated dienes, suggesting that PnCE loss reflects the course of peroxidation of endogenous lipoprotein lipids. Using PnCE to separately monitor LDL- and HDL-specific oxidation within LDL-HDL mixtures, we obtained direct evidence that HDL inhibits both Cu(2+)- and Fe(3+)-induced peroxidation of LDL-associated lipids. Notably, in the presence of Cu2+, loss of HDL-associated PnCE fluorescence also was inhibited in LDL-HDL co-incubations, suggesting that LDL exert an antioxidant effect under these conditions as well. Thus, results obtained using this new methodology are consistent with previously reported antioxidant effects of HDL, but indicate that the behavior of individual lipoprotein particles may be more complicated than can be predicted from the collective behavior of the lipoprotein mixture.

Expression of human lecithin-cholesterol acyltransferase in transgenic mice. Effect of human apolipoprotein AI and human apolipoprotein all on plasma lipoprotein cholesterol metabolism.

Human (Hu) lecithin-cholesterol acyltransferase (LCAT) is a key enzyme in the plasma metabolism of cholesterol. To assess the effects of increased plasma levels of LCAT, four lines of transgenic mice were created expressing a Hu LCAT gene driven by either its natural or the mouse albumin enhancer promoter. Plasma LCAT activity increased from 1.2- to 1.6-fold higher than that found in control mouse plasma. Lipid profiles, upon comparing Hu LCAT transgenics to control animals, revealed a 20 t0 60% increase in total and cholesteryl esters that were mainly present in HDL. The in vivo substrate specificity of Hu LCAT was assessed by creating animals expressing Hu apo AI + Hu LCAT (HuAI/ LCAT), Hu apo AI + Hu apo AII + Hu LCAT (HuAI/ AII/LCAT), and Hu apo AII + Hu LCAT (HuAII/LCAT). Plasma cholesterol was increased up to 4.2-fold in HuAI/ LCAT transgenic mice and twofold in the HuAI/AII/LCAT transgenic mice, compared with HuAI and HuAI/AII transgenic mice. HDL cholesteryl ester levels were increased more than twofold in both the HuAI/LCAT and HuAI/AII/LCAT mice compared with the HuAI, HuAI/AII, and HuLCAT animals. The HDL particles were predominantly larger in the HuAI/LCAT and the HuAI/AII/LCAT mice compared with those in HuAI, HuAII/LCAT, and HuLCAT animals. The increase in LCAT activity in the HuAI/LCAT and HuAI/AII/LCAT mice was associated with 62 and 27% reductions respectively, in the proportion of Hu apo AI in the pre beta-HDL fraction, when compared with HuAI and HuAI/AII transgenic mice. These data demonstrate that moderate increases in LCAT activity are associated with significant changes in lipoprotein cholesterol levels and that Hu LCAT has a significant preference for HDL containing Hu apo AI.

Sodium oleate-facilitated reassembly of apolipoprotein A-I with phosphatidylcholine.

The influence of sodium oleate (oleate) on complexing of apolipoprotein A-I (apo A-I) with egg yolk phosphatidylcholine (EYPC) was evaluated. Without the use of additional detergent such as sodium cholate, oleate facilitates formation of a single complex of unique stoichiometry, approx. 76:2:20, molar ratio EYPC/apo A-I/oleate, and mean size 7.4 nm with round to ellipsoidal morphology. Near complete reassembly of apo A-I into the complex occurs when the stoichiometry of the mixture approximates that of the complex itself. With increasing content of EYPC in the mixture, the same complex is formed but in decreasing yield; larger complexes are not formed. The rate of complex formation decreases with increase of EYPC in the mixture. Reduction of pH in the reassembly mixture from 8.0 to 5.4 results in a marked reduction in complex formation indicating that ionized oleate facilitates lipidation. Removal of oleate by interaction of the complex with fatty acid-free human serum albumin does not degrade the complex. Incorporation of increasing amounts of unesterified cholesterol into the EYPC-sonicate progressively inhibits oleate-facilitated complex formation. This study shows that oleate, a physiologically relevant lipolysis-derived product, facilitates reassembly of apo A-I with EYPC and promotes formation of a small lipid-poor particle similar to that observed in nascent HDL and during in vivo or in vitro lipolysis of triacylglycerol-rich lipoproteins in the presence of HDL.

Structural and functional properties of human and mouse apolipoprotein A-I.

Mouse and human plasma apolipoprotein A-I (apo A-Im and apo A-Ih, respectively) were investigated to compare their molecular properties in solution, their incorporation into palmitoyloleoylphosphatidylcholine-apo A-I (POPC-apo A-I) discoidal complexes; their structural stability in discoidal complexes and high-density lipoproteins (HDL), and their effect on structural rearrangement of discoidal complexes upon interaction with low-density lipoproteins (LDL). Unlike apo A-Ih, only minimal concentration-dependent self-association was observed for apo A-Im. While both apo A-Im and apo A-Ih formed discoidal complexes of distinct composition and size that reflected reassembly molar ratios of POPC/apo A-I, apo A-Im demonstrated specific deficiencies in formation of larger-sized complexes. Denaturation of both apo A-Im- or apo A-Ih-containing complexes and HDL with guanidine hydrochloride (GuHCl) indicated significantly reduced stabilization of apo A-Im by lipid in these particles. Interaction of apo A-Im- or apo A-Ih-containing discoidal complexes with human plasma LDL revealed a more extensive conversion of apo A-Im-complexes to smaller species. Mean hydrophobicities and mean hydrophobic moments of amphipathic helical segments in apo A-Im and apo A-Ih were compared; differences potentially contributing to differential lipid-binding properties between apo A-Im and apo A-Ih were identified. Our results demonstrate differences between apo A-Im and apo A-Ih that may contribute to the major changes in plasma HDL distribution and function observed in apo A-Ih transgenic mice.

Apolipoprotein-lipid association in oxidatively modified HDL and LDL.

We investigated the effect of Cu2+ catalyzed peroxidation on the status of tryptophan (Trp) in protein moieties in HDL and LDL together with its effect on apolipoprotein-lipid association. Incubation of HDL with Cu2+ resulted in a rapid decrease of Trp fluorescence intensity with time with a concomitant increase in Trp maximum emission wavelength (lambda max). LDL incubated with Cu2+ also showed a rapid decrease in Trp fluorescence intensity with time, with no associated increase in lambda max. The status of apo HDL and apo LDL was investigated after 4 h oxidation (4h-oxHDL and 4h-oxLDL respectively). With 4h-oxHDL, the shift in lambda max was not associated with protein dissociation but rather with protein crosslinking and formation of larger HDL species. Progressive increase in lambda max was observed in 4h-oxHDL with increase in guanidine hydrochloride (GuHCl) concentration; this was not due to protein dissociation. Although oxidation of LDL did not produce an increase in lambda max, a significant increase in wavelength was observed when 4h-oxLDL was exposed to increasing concentration of GuHCl. SDS-polyacrylamide gel electrophoresis and nondenaturing gradient gel electrophoresis of the 4h-oxLDL indicated formation of smaller molecular weight protein fragments that were still associated with LDL. Ultracentrifugation of oxidized LDL in the presence and absence of GuHCl showed no dissociated protein. In summary, these data indicate the following: (a) lipid peroxidation has a direct effect on Trp residues in both HDL and LDL, (b) oxidation of HDL is associated with conformational change in apo HDL, crosslinking and formation of larger particles, (c) oxidized HDL have a more stable apolipoprotein-lipid association than native HDL, (d) oxidation of LDL is associated with changes in apo B, that by fluorescence are apparent only in presence of GuHCl and results in fragmentation of apo B without dissociation of protein or change in particle size, and (e) stability of apolipoprotein-lipid association is comparable in oxidized and native LDL.

Protein composition determines the anti-atherogenic properties of HDL in transgenic mice.

High-density lipoprotein (HDL) contains two major proteins, apolipoprotein A-I (apoA-I) and apolipoprotein A-II (apoA-II), comprising about 70% and 20% of the total HDL protein mass, respectively. HDL exists in human plasma in two main forms, one containing apoA-I with apoA-II (AI/AII-HDL) and another containing apoA-I without apoA-II (AI-HDL). A strong inverse relationship exists between total plasma HDL concentration and atherosclerosis, but the results of studies examining the relationship between AI-HDL and AI/AII-HDL and atherosclerosis have been conflicting. To determine whether these two HDL populations have different effects on atherogenesis, human apoA-I (AI) and human apoA-I and apoA-II (AI/AII) transgenic mice were produced in an atherosclerosis-susceptible strain. Following an atherogenic diet, despite similar total cholesterol and HDL cholesterol concentrations, the area of atherogenic lesions in the AI/AII mice was 15-fold greater than in the AI animals. These studies show that the protein composition of HDL significantly affects its role in atherogenesis and that AI-HDL is more antiatherogenic than AI/AII-HDL.

Expression of human apolipoprotein A-II and its effect on high density lipoproteins in transgenic mice.

Apolipoproteins A-I and A-II comprise approximately 70 and 20%, respectively, of the total protein content of HDL. Evidence suggests that apoA-I plays a central role in determining the structure and plasma concentration of HDL, while the role of apoA-II is uncertain. To help define the function of apoA-II and determine what effect increasing its plasma concentration has on HDL, transgenic mice expressing human apoA-II and both human apoA-I and human apoA-II were produced. Human apoA-II mRNA is expressed exclusively in the livers of transgenic animals, and the protein exists as a dimer as it does in humans. High level expression of human apoA-II did not increase HDL concentrations or decrease plasma concentrations of murine apoA-I and apoA-II in contrast to what was observed in mice overexpressing human apoA-I. The primary effect of overexpressing human apoA-II was the appearance of small HDL particles composed exclusively of human apoA-II. HDL from mice transgenic for both human apoA-I and human apoA-II displayed a unique size distribution when compared with either apoA-I or apoA-II transgenic mice and contain particles with both these human apolipoproteins. These results in mice, indicating that human apoA-II participates in determining HDL size, parallel results from human studies.

Apolipoprotein-specific populations in high density lipoproteins of human cord blood.

High density lipoproteins (HDL) in human cord blood have previously been shown to exhibit particle size profiles distinctly different from those of adult HDL. The adult HDL profile is comprised of separate contributions from two major apolipoprotein-specific populations; one population contains both apolipoproteins AI and AII (HDL(AIwAII], while the other has apolipoprotein AI without AII (HDL(AIw/oAII]. The present studies establish that cord blood HDL are also comprised of HDL(AIwAII) and HDL(AIw/oAII) populations whose particle size profiles closely reflect cholesterol and HDL-cholesterol levels in cord blood. Compared with the adult, cord blood HDL(AIwAII) profiles generally show both a greater subspeciation within HDL2a and HDL3b/3c size intervals as well as relative reduction of material in the HDL3a interval. In the cord blood HDL(AIw/oAII) profile, HDL2b(AIw/oAII) particles also show subspeciation with a major component that is consistently larger than that normally observed in the adult (11.2 vs. 10.3 nm). As in the adult, the HDL3a(AIw/oAII) component is present but, unlike the adult, its relative amount is low; hence, its peak is usually not discernable in the cord blood total HDL profile. Our studies show that the larger-sized HDL2b(AIw/oAII) of cord blood are enriched in phospholipid which probably accounts for their increased size. The protein moiety of the larger-sized HDL2b(AIw/oAII) has a molecular weight equivalent to four apolipoprotein AI molecules per particle similar to the normal-sized adult subpopulation. Phospholipid enrichment of cord blood HDL(AIwAII) subpopulations within the HDL2a size interval was not observed. However, the protein moiety of cord blood HDL2a(AIwAII) is unusual in that it exhibits an apolipoprotein AI:AII molar ratio considerably lower (0.8:1 vs. 1.6:1) than that of adult. We suggest that the unique particle size distribution of cord blood total HDL is due in large part to: (a) a specific enrichment of phospholipid in HDL2b(AIw/oAII) species, producing particles larger than normal adult counterparts and (b) an elevated proportion of apoAII carried by the HDL(AIwAII) particles that may influence subspeciation in the HDL3a/b/c size interval.

Dissociation of high density lipoprotein precursors from apolipoprotein B-containing lipoproteins in the presence of unesterified fatty acids and a source of apolipoprotein A-I.

Incubation of low (LDL), intermediate (IDL), or very low density lipoproteins (VLDL) with palmitic acid and either high density lipoproteins (HDL), delipidated HDL, or purified apolipoprotein (apo) A-I resulted in the formation of lipoprotein particles with discoidal structure and mean particle diameters ranging from 146 to 254 A by electron microscopy. Discs produced from IDL or LDL averaged 26% protein, 42% phospholipid, 5% cholesteryl esters, 24% free cholesterol, and 3% triglycerides; preparations derived from VLDL contained up to 21% triglycerides. ApoA-I was the predominant protein present, with smaller amounts of apoA-II. Crosslinking studies of discs derived from LDL or IDL indicated the presence of four apoA-I molecules per particle, while those derived from large VLDL varied more in size and contained as many as six apoA-I molecules per particle. Incubation of discs derived from IDL or LDL with purified lecithin:cholesterol acyltransferase (LCAT), albumin, and a source of free cholesterol produced core-containing particles with size and composition similar to HDL2b. VLDL-derived discs behaved similarly, although the HDL products were somewhat larger and more variable in size. When discs were incubated with plasma d greater than 1.21 g/ml fraction rather than LCAT, core-containing particles in the size range of normal HDL2a and HDL3a were also produced. A variety of other purified free fatty acids were shown to promote disc formation. In addition, some mono and polyunsaturated fatty acids facilitated the formation of smaller, spherical particles in the size range of HDL3c. Both discoidal and small spherical apoA-I-containing lipoproteins were generated when native VLDL was incubated with lipoprotein lipase in the presence of delipidated HDL. We conclude that lipolysis product-mediated dissociation of lipid-apoA-I complexes from VLDL, IDL, or LDL may be a mechanism for formation of HDL subclasses during lipolysis, and that the availability of different lipids may influence the type of HDL-precursors formed by this mechanism.

Discoidal complexes containing apolipoprotein E and their transformation by lecithin-cholesterol acyltransferase.

The primary objectives of this study were to determine whether analogs to native discoidal apolipoprotein (apo)E-containing high-density lipoproteins (HDL) could be prepared in vitro, and if so, whether their conversion by lecithin-cholesterol acyltransferase (LCAT; EC 2.3.1.43) produced particles with properties comparable to those of core-containing, spherical, apoE-containing HDL in human plasma. Complexes composed of apoE and POPC, without and with incorporated unesterified cholesterol, were prepared by the cholate-dialysis technique. Gradient gel electrophoresis showed that these preparations contain discrete species both within (14-40 nm) and outside (10.8-14 nm) the size range of discoidal apoE-containing HDL reported in LCAT deficiency. The isolated complexes were discoidal particles whose size directly correlated with their POPC:apoE molar ratio: increasing this ratio resulted in an increase in larger complexes and a reduction in smaller ones. At all POPC:apoE molar ratios, size profiles included a major peak corresponding to a discoidal complex 14.4 nm long. Preparations with POPC:apoE molar ratios greater than 150:1 contained two distinct groups of complexes, also in the size range of discoidal apoE-containing HDL from patients with LCAT deficiency. Incorporation of unesterified cholesterol into preparations (molar ratio of 0.5:1, unesterified cholesterol:POPC) resulted in component profiles exhibiting a major peak corresponding to a discoidal complex 10.9 nm long. An increase of unesterified cholesterol and POPC (at the 0.5:1 molar ratio) in the initial mixture, increased the proportion of larger complexes in the profile. Incubation of isolated POPC-apoE discoidal complexes (mean sizes, 14.4 and 23.9 nm) with purified LCAT and a source of unesterified cholesterol converted the complexes to spherical, cholesteryl ester-containing products with mean diameters of 11.1 nm and 14.0 nm, corresponding to apoE-containing HDL found in normal plasma. Conversion of smaller cholesterol-containing discoidal complexes (mean size, 10.9 nm) under identical conditions resulted in spherical products 11.3, 13.3, and 14.7 nm across. The mean sizes of these conversion products compared favorably with those (mean diameter, 12.3 nm) of apoE-containing HDL of human plasma. This conversion of cholesterol-containing complexes is accompanied by a shift of some apoE to the LDL particle size interval. Our study indicates that apoE-containing complexes formed by the cholate-dialysis method include species similar to discoidal apoE-containing HDL and that incubation with LCAT converts most of them to spherical core-containing particles in the size range of plasma apoE-containing HDL. Plasma HDL particles containing apoE may arise in part from direct conversion of discoidal apoE-containing HDL by LCAT.

Conversion of apolipoprotein-specific high-density lipoprotein populations during incubation of human plasma.

Incubation studies were performed on plasma obtained from subjects selected for relatively low levels of high-density lipoprotein cholesterol (HDL-C) (no greater than 30 mg/dl) and particle size distributions enriched in the HDL3 subclass. Incubation (12 h, 37 degrees C) of plasma in the presence or absence of lecithin: cholesterol acyltransferase activity produces marked alteration in size profiles of both major apolipoprotein-specific HDL3 populations (HDL3(AI w AII), HDL3 species containing both apolipoprotein A-I and apolipoprotein A-II, and HDL3(AI w/o AII), HDL3 species containing apolipoprotein A-I) as isolated by immunoaffinity chromatography. In the presence or absence of lecithin: cholesterol acyltransferase activity, plasma incubation results in a shift of HDL3(AI w AII) species (initial mean sizes of major components, approx. 8.8 and 8.0 nm) predominantly to larger particles (mean size, 9.8 nm). A less prominent shift to smaller particles (mean size, 7.8 nm) accompanies the conversion to larger particles only when the enzyme is active. Combined shifts to larger (mean size, 9.8 nm) and smaller (mean size, 7.4 nm) particles are observed for HDL3(AI w/o AII) particles (mean size, 8.3 nm) also only in the presence of enzyme activity. However, in the absence of enzyme activity, HDL3(AI w/o AII) species, unlike the HDL3(AI w AII) species, are converted to smaller (mean size 7.4 nm) rather than to larger particles. Like native HDL2b(AI w/o AII) particles, the larger HDL3(AI w/o AII) conversion products exhibit a protein moiety with molecular weight equivalent to four apolipoprotein A-I molecules per particle; small HDL3(AI w/o AII) products are comprised predominantly of particles with two apolipoprotein A-I per particle. Incubation-induced conversion of HDL3 particles in the presence of lecithin: cholesterol acyltransferase activity is associated with increased binding of both apolipoprotein-specific HDL populations to low-density lipoproteins (LDL). The present studies indicate that, in the absence of lecithin: cholesterol acyltransferase activity, the two HDL3 populations follow different conversion pathways, possibly due to apolipoprotein-specific activities of lipid transfer protein or conversion protein in plasma. Our studies also suggest that lecithin: cholesterol acyltransferase activity may play a role in the origins of large HDL2b(AI w/o AII) species in human plasma by participating in the conversion of HDL3(AI w/o AII) particles, initially with three apolipoprotein A-I, to larger particles with four apolipoprotein A-I per particle.

Transformation of large discoidal complexes of apolipoprotein A-I and phosphatidylcholine by lecithin-cholesterol acyltransferase.

Using a cholate-dialysis recombination procedure, complexes of apolipoprotein A-I and synthetic phosphatidylcholine (1-palmitoyl-2-oleoylphosphatidylcholine (POPC) or dioleoylphosphatidylcholine (DOPC] were prepared in mixtures at a relatively high molar ratio of 150:1 phosphatidylcholine/apolipoprotein A-I. Particle size distribution analysis by gradient gel electrophoresis of the recombinant mixtures indicated the presence of a series of discrete complexes that included species migrating at RF values observed for discoidal particles in nascent high-density lipoproteins (HDL) in plasma of lecithin-cholesterol acyltransferase-deficient subjects. One of these complex species, designated complex class 6, formed with either phosphatidylcholine, was isolated by gel filtration and characterized at follows: discoidal shape (mean diameter 20.8 nm (POPC) and 19.0 nm (DOPC]; molar ratio, phosphatidylcholine/apolipoprotein A-I, 155:1 (POPC) and 130:1 (DOPC); and both containing 4 molecules of apolipoprotein A-I per particle. Incubation of class 6 complexes with lecithin-cholesterol acyltransferase (EC 2.3.1.43) and a source of unesterified cholesterol (low-density lipoprotein (LDL] was shown by electron microscopy to result in a progressive transformation of the discoidal particles (0 h) to deformable (2.5 h) and to spherical particles (24 h). The spherical particles (diameter 13.6 nm (POPC) and 12.5 nm (DOPC) exhibit sizes at the upper boundary of the interval defining the human plasma (HDL2b)gge (12.9-9.8 nm). The spherical particles contain a cholesteryl ester core that reaches a limiting molar ratio of approx. 50-55:1 cholesteryl ester/apolipoprotein A-I. The deformable particles assume a rectangular shape under negative staining and, relative to the 24-h spherical product, are enriched in phosphatidylcholine. Chemical crosslinking (by dimethyl suberimidate) of the isolated transformation products shows the 24-h spherical particle to contain predominantly 4 apolipoprotein A-I molecules; products produced after intermediate periods of time appear to contain species with 3 and 4 apolipoproteins per particle. Our in vitro studies indicate a potential pathway in the origins of large, apolipoprotein A-I-containing plasma HDL particles. The deformable species observed during transformation were similar in size and shape to particles observed in interstitial fluid.

Characterization of A-I-containing lipoproteins in subjects with A-I Milano variant.

The A-I Milano variant of apolipoprotein A-I (A-IM), by virtue of its Arg-173----Cys substitution, is capable of forming a disulfide bond with the 77-amino-acid apolipoprotein A-II polypeptide (A-IIS) as well as with itself to produce dimers, A-IM/A-IIS and A-IM/A-IM, respectively. A-I-containing lipoproteins (Lp): particles with A-II (Lp(A-I with A-11)) and particles without A-II (Lp(A-I without A-II)) in the plasma of two nonhyperlipidemic A-IM carriers were investigated to determine the effect of A-IM on these lipoproteins. Despite the existence of abnormal apolipoprotein dimers and the unusually low HDL cholesterol (17 and 14 mg/dl), A-I (67 and 75 mg/dl), and A-II (18 and 18 mg/dl) levels in the two carriers, the plasma A-I of the carriers was distributed between Lp(A-I with A-II) and Lp(A-I without A-II) in a proportion comparable to that observed in normals. As expected, A-IM/A-IIS mixed dimer was found in carrier Lp(A-I with A-II). However, A-IM/A-IM dimer was located almost exclusively in carrier Lp(A-I without A-II). Chemical (dimethylsuberimidate) crosslinking of the protein moieties of the major subpopulations of Lp(A-I with A-II) and Lp(A-I without A-II) of normal and A-IM carriers showed that Lp(A-I with A-II), which is located predominantly in the 7.8-9.7 nm interval ((HDL2a + 3a + 3b)gge), had an apparent protein molecular weight equivalent to two molecules of A-I and one to two molecules of A-II per particle. Most of the Lp(A-I without A-II) particles, located predominantly in the size intervals of 9.7-12.9 nm (designated (HDL2b)gge) and 8.2-8.8 nm (HDL3a)gge) had protein moieties exhibiting a molecular weight equivalence predominantly of four and three molecules of A-I, respectively. A small quantity of particles with apparent protein content of two molecules of A-I in the 7.2-8.2 nm interval ((HDL3b + 3c)gge) was also detected. These studies showed that in nonhyperlipidemic A-IM carriers, the occurrence of apolipoprotein dimers had not markedly affected the protein stoichiometry of Lp(A-I with A-II) and Lp(A-I without A-II).

Characterization of complexes of egg yolk phosphatidylcholine and apolipoprotein A-II prepared in the absence and presence of sodium cholate.

Complexes of apolipoprotein A-II and egg yolk phosphatidylcholine were prepared in mixtures of different composition in the absence and presence of sodium cholate. By gradient gel electrophoresis, complex preparations were polydisperse and particle size distributions were influenced by the composition of the reconstitution mixture. Complexes generally exhibited a discoidal morphology by electron microscopy, but showed increased formation of vesicular complexes at elevated levels of egg yolk PC in the mixtures. By chemical crosslinking, complexes formed in the absence of cholate were shown to consist primarily of discoidal species with three apolipoprotein A-II molecules per particle in the mixtures investigated; complexes formed in the presence of cholate included species ranging from three to five apolipoprotein A-II per particle. The number of apolipoprotein A-II per particle and the sizes of the complexes, prepared in cholate, increased with increase of egg yolk PC in the reconstitution mixture. Relative to the particle size distribution of discoidal complexes formed in the absence of cholate, those prepared in cholate showed a distribution shifted to larger particle sizes. Complexes of similar particle size distribution formed in the presence or absence of cholate showed similar physical-chemical properties. Discoidal complexes with the same number of apolipoprotein A-II per particle but of different size and composition were observed, suggesting the possibility of some conformational adaptation of apolipoprotein A-II leading to stabilization of egg yolk PC bilayers of different diameter. Properties of particle size distributions of discoidal complexes prepared in cholate of apolipoprotein A-II and egg yolk PC were compared with those of complexes of apolipoprotein A-I previously reported (Nichols, A.V., Gong, E.L., Blanche, P.J. and Forte, T.M. (1983) Biochim. Biophys. Acta 750, 353-364).