In vivo studies of HDL assembly and metabolism using adenovirus-mediated transfer of ApoA-I mutants in ApoA-I-deficient mice.

We have used adenovirus-mediated gene transfer in apoA-I-deficient (A-I-/-) mice to probe the in vivo assembly and metabolism of HDL using apoA-I variants, focusing primarily on the role of the C-terminal 32 amino acids (helices 9-10). Lipid, lipoprotein, and apoA-I analyses showed that plasma levels of apoA-I and HDL of the mutants were 40-88% lower than that of wild type (WT) human apoA-I despite comparable levels of expression in the liver. WT apoA-I and mutant 1 (P165A, E172A) formed spherical particles with the size and density of HDL2 and HDL3. Mutant 2 (E234A, E235A, K238A, K239A) generated spherical particles with density between HDL2 and HDL3. Mutant 3 (L211V, L214V, L218V, L219V) and mutant 4 (L222K, F225K, F229K), which have substitutions of hydrophobic residues in the C-terminus, generated discoidal HDL particles indicating a defect in their conversion to mature spherical HDL. Significant amounts of mutant 4 and mutant 5 (truncated at residue 219) were found in the lipid poor fractions after ultracentrifugation of the plasma (18 and 35%, respectively, of total apoA-I). These findings suggest that hydrophobic residues in and/or between helices 9 and 10 are important for the maturation of HDL in vivo.

SAA-only HDL formed during the acute phase response in apoA-I+/+ and apoA-I-/- mice.

Serum amyloid A (SAA) is an acute phase protein of unknown function that is involved in systemic amyloidosis and may also be involved in atherogenesis. The precise role of SAA in these processes has not been established. SAA circulates in plasma bound to high density lipoprotein-3 (HDL3). The pathway for the production of SAA-containing HDL is not known. To test whether apolipoprotein (apo)A-I-HDL is required in the production of SAA-HDL, we analyzed the lipopolysaccharide (LPS)-induced changes in apoA-I+/+ and apoA-I-/- mice. In apoA-I+/+ mice, after injection of LPS, remodeling of HDL occurred: total cholesterol increased and apoA-I decreased slightly and shifted to lighter density. Dense (density of HDL3) but large (size of HDL2 ) SAA-containing particles were formed. Upon fast phase liquid chromatography fractionation of plasma, >90% of SAA eluted with HDL that was enriched in cholesterol and phospholipid and shifted "leftward" to larger particles. Non-denaturing immunoprecipitation with anti-mouse apoA-I precipitated all of the apoA-I but not all of the SAA, confirming the presence of SAA-HDL devoid of apoA-I. In the apoA-I-/- mice, which normally have very low plasma lipid levels, LPS injection resulted in significantly increased total and HDL cholesterol. Greater than 90% of the SAA was lipid associated and was found on dense but large, spherical HDL particles essentially devoid of other apolipoproteins.We conclude that serum amyloid A (SAA) is able to sequester lipid, forming dense but large HDL particles with or without apoA-I or other apolipoproteins. The capacity to isolate lipoprotein particles containing SAA as the predominant or only apolipoprotein provides an important system to further explore the biological function of SAA.

Nascent astrocyte particles differ from lipoproteins in CSF.

Little is known about lipid transport and metabolism in the brain. As a further step toward understanding the origin and function of CNS lipoproteins, we have characterized by size and density fractionation lipoprotein particles from human CSF and primary cultures of rat astrocytes. The fractions were analyzed for esterified and free cholesterol, triglyceride, phospholipid, albumin, and apolipoproteins (apo) E, AI, AII, and J. As determined by lipid and apolipoprotein profiles, gel electrophoresis, and electron microscopy, nascent astrocyte particles contain little core lipid, are primarily discoidal in shape, and contain apoE and apoJ. In contrast, CSF lipoproteins are the size and density of plasma high-density lipoprotein, contain the core lipid, esterified cholesterol, and are spherical. CSF lipoproteins were heterogeneous in apolipoprotein content with apoE, the most abundant apolipoprotein, localized to the largest particles, apoAI and apoAII localized to progressively smaller particles, and apoJ distributed relatively evenly across particle size. There was substantial loss of protein from both CSF and astrocyte particles after density centrifugation compared with gel-filtration chromatography. The differences between lipoproteins secreted by astrocytes and present in CSF suggest that in addition to delivery of their constituents to cells, lipoprotein particles secreted within the brain by astrocytes may have the potential to participate in cholesterol clearance, developing a core of esterified cholesterol before reaching the CSF. Study of the functional properties of both astrocyte-secreted and CSF lipoproteins isolated by techniques that preserve native particle structure may also provide insight into the function of apoE in the pathophysiology of specific neurological diseases such as Alzheimer's disease.

Association of human, rat, and rabbit apolipoprotein E with beta-amyloid.

In humans, apolipoprotein E (apoE) has three major isoforms, E2 (Cys112, Cys158), E3 (Cys112, Arg158), and E4 (Arg112, Arg158). While epsilon4 is a genetic risk factor for Alzheimer's disease (AD), epsilon2 may protect against late-onset AD. Using native preparations of apoE from conditioned tissue culture media or plasma lipoproteins, we have previously shown that when equivalent amounts of apoE3 or E4 were incubated with beta-amyloid (A beta), apoE3 formed 20 times as much SDS-stable complex with the peptide as apoE4. This preferential binding of A beta to apoE3 was abolished when apoE was purified by a process which includes delipidation and denaturation. Here we expand these observations to include A beta binding to lipoprotein-associated and purified apoE2. Lipoproteins isolated from the plasma of individuals homozygous for either epsilon2 or epsilon3 were incubated with A beta(1-40). SDS-stable complex formation was analyzed by a non-reducing gel shift assay, followed by immunoblotting with either A beta or apoE antibodies. ApoE2:A beta complex formation was comparable to apoE3:A beta in both native and purified preparations of apoE. In addition, lipoprotein-associated rat apoE (Arg112, Arg158), like human apoE4, did not form complex with A beta, while lipoprotein-associated rabbit apoE (Cys112, Arg158) did bind the peptide. These binding studies provide one possible explanation for protective effects of both apoE2 and E3 against the development of Alzheimer's disease.

HDL content and composition in acute phase response in three species: triglyceride enrichment of HDL a factor in its decrease.

High density lipoprotein (HDL) levels decrease during the acute phase response (APR). We have used the APR model of rabbit, baboon, and mouse to study the factors that influence HDL level. In the baboons and rabbits there was massive hypertriglyceridemia, triglyceride enrichment of HDL (60-80% of core lipids), decreases of HDL-cholesterol and apolipoprotein (apo)A-I (to 10% of baseline), and increases of apoA-I in the non-lipoprotein bottom fraction suggesting dissociation of apoA-I from the particles. Detailed analyses of serum amyloid A (SAA)-rich HDL done in the rabbit revealed large, triglyceride-enriched (> 60% of core lipids) particles containing > 95% SAA. These particles had a high surface to core ratio (13.4 +/- 1.94, control = 3.0 +/- 0.12) and a very high protein (79.71 +/- 5.25 weight %, control = 37.2 +/- 0.43) proportion, large (r = 5.95 nm) when examined by non-denaturing gradient electrophoresis but small when examined by electron microscopy (r = 4.2 nm). In the mouse there was no hypertriglyceridemia, no triglyceride enrichment of HDL, no decrease of HDL cholesterol. ApoA-I decreased to about 61.4% of baseline but did not increase in the bottom fraction although large but dense SAA-enriched HDL particles were also produced. These results suggest that hypertriglyceridemia, triglyceride-enrichment of HDL, and dissociation of apoA-I from the particles, possibly by displacement of apoA-I by SAA, are important factors in the decline of HDL during the APR. Whether differences in triglyceride metabolism account for the differences in the HDL response in the species studied requires further experimentation.

Heparin-induced lipolysis in hypertriglyceridemic subjects results in the formation of atypical HDL particles.

This study reports on the characterization of high density lipoprotein (HDL) in normotriglyceridemic and hypertriglyceridemic (HTG) subjects, after a fat meal and heparin-induced release of lipases. Samples for detailed analysis of HDL by density gradient ultracentrifugation and nondenaturing gradient gel electrophoresis were collected at 0 h and 5 h after the meal and 15 min after the administration of heparin. The normotriglyceridemic subjects were subdivided into two groups: those who remained normotriglyceridemic 5 h after the meal (NTG) or those who were hypertriglyceridemic at this time point (NTG-HTG). At the outset of the study, mean triglyceride levels were significantly higher (P < 0.001) and HDL cholesterol levels lower (P < 0.02) in the HTG group. The HDL particles in this group were enriched with triglyceride (P < 0.001). Serum triglyceride levels rose in all three groups after the fat meal and this was associated with further triglyceride enrichment of the HDL particles. In all groups, rapid lipolysis induced by heparin caused a significant decrease in plasma triglycerides and increase in free fatty acid levels, these changes being greatest in the HTG group. HDL density profiles of the study groups prior to the administration of heparin demonstrated two distinct peaks at density 1.09 g/ml (HDL2) and 1.13 g/ml (HDL3). However, after the administration of heparin to the HTG group, only a single peak in the HDL profile was evident that was located at the density region corresponding to HDL2 (1.09 g/ml). Upon gradient gel electrophoresis of this peak, there was an increased number (P < 0.005 vs. NTG) of small particles (< 4.37 nm) whose size was similar to the size range normally associated with HDL3b and HDL3c. Similar changes in HDL density and size after the administration of heparin were observed in the NTG-HTG group who were also hypertriglyceridemic postprandially. By contrast, the density gradient profiles and sizes of the HDL particles did not change after the administration of heparin to NTG subjects. Thus, the activation of lipolysis in HTG subjects leads to the generation of atypical HDL particles that are small but of reduced density. Rapid clearance of such particles could account for the inverse relationship between triglyceride and HDL cholesterol in this population subgroup.

Postprandial triglyceride response in type 1 (insulin-dependent) diabetes mellitus is not altered by short-term deterioration in glycaemic control or level of postprandial insulin replacement.

UNLABELLED: The effect of deteriorating glycaemic control on the lipoprotein responses to the ingestion of a high fat meal was investigated in seven normolipidaemic Type 1 (insulin-dependent) diabetic patients and the results were compared with corresponding responses in seven normolipidaemic control subjects. In addition, the importance of insulin in regulating the postprandial lipoprotein responses was examined by comparing the results obtained from the diabetic patients maintained on a basal infusion of insulin throughout the study with those obtained when a step-up, step-down insulin infusion was administered following the meal. Vitamin A was added to the test meal in all subjects to trace the metabolism of the chylomicron (Sf greater than 1000) and non-chylomicron (Sf less than 1000) fractions in the postprandial period. No differences in fasting and postprandial triglyceride levels nor in the concentration of the chylomicron and non-chylomicron fractions were observed between diabetic and control subjects. In the diabetic patients short-term (two-week) deterioration in glycaemic control did not have any adverse influence on the basal and postprandial lipid responses. However, while the amount of insulin administered after the meal in the diabetic patients did not have any effect on the postprandial triglyceride or chylomicron responses, the concentration of non-esterified fatty acids was significantly higher (p less than 0.0005) when only a basal infusion of insulin was administered. IN CONCLUSION: 1) Short-term deterioration in glycaemic control does not adversely affect lipoprotein concentrations in Type 1 diabetes. 2) Non-esterified fatty acids appear to be a more sensitive index of insulinization post-prandially than triglycerides.