Binding of modified high density lipoproteins to endothelial cells: relation with cellular cholesterol efflux?

High density lipoprotein oxidation: in vitro susceptibility and potential in vivo consequences

Elevated levels of plasma high density lipoprotein (HDL) are strongly predictive of protection against atherosclerotic vascular disease. HDL particles likely have several beneficial actions in vivo, including the initiation of reverse cholesterol transport. The apparent importance of oxidative modification of low density lipoprotein in atherogenesis raises the question of how oxidative modification of HDL might affect its cardioprotective actions. HDL is readily oxidized using numerous models of lipoprotein oxidation. In vitro evidence suggests oxidation might impair some protective actions, but actually enhance other mechanisms induced by HDL that prevent the accumulation of cholesterol in the artery wall. This article reviews the current literature concerning the relative oxidizability of HDL, the structural changes induced in HDL by oxidation in vitro, and the potential consequences of oxidative modification on the protective actions of HDL in vivo.

Oxidation of high density lipoproteins: characterization and effects on cholesterol efflux from J774 macrophages

Oxidative modification of high density lipoproteins (HDL) may alter their capacity to mediate cellular cholesterol efflux. We studied the kinetics of copper-mediated oxidation of HDL and cholesterol efflux mediated by unmodified and oxidized HDL (oxHDL). Oxidation was measured by increases in absorbance at 234 nm (delta A234), production of thiobarbituric acid reactive substances (TBARS) and loss of trinitrobenzene sulfonic acid reactivity. Oxidation was dependent on copper concentration and showed a lag phase and propagation phase. Efflux of cholesterol from J774 macrophages measured by appearance of cellular [3H]cholesterol in the medium was lower by 16% after 4 h and 36% after 24 h with oxHDL compared to HDL. OxHDL-mediated efflux was also lower by 27% to 37% at lipoprotein concentrations of 10 to 200 micrograms protein/ml. Cholesterol efflux correlated negatively with TBARS production (r = -0.97, P < 0.003) and delta A234 (r = -0.77, P < 0.080). There was no difference in efflux mediated by apoproteins prepared from HDL and oxHDL. Efflux measured by change in cholesterol mass in medium was 78% lower with oxHDL. Inhibition of oxidation with butylated hydroxytoluene maintained the capacity of HDL to stimulate efflux. These results suggest that oxidation of HDL may impair its protective role against atherosclerosis.

Uptake of apolipoprotein E-rich and apolipoprotein E-poor subfractions of high-density lipoprotein by liver membranes and HepG2 cells

Apolipoprotein (apo) E plays an important role in mediating high-density lipoprotein (HDL) cholesterol transport and uptake by the liver. Evidence for and against the existence of conventional liver receptors for HDL containing apoE have been reported, although the selective uptake of the cholesterol moiety of HDL has been demonstrated. The present study investigated the hepatic uptake of subfractions of HDL separated on the basis of their apoE content. Rabbit HDL and its apoE-rich and apoE-poor subfractions, separated by heparin-Sepharose affinity chromatography, were labelled in their apoprotein moieties with [14C]sucrose and in their cholesteryl ester moiety with 3H. No binding of either subfraction to rabbit liver membranes could be detected. With cultured HepG2 cells, however, there was a high uptake of 3H but a very low uptake of 14C from both HDL subfractions, demonstrating that selective uptake was operating. Addition of unlabelled apoE-poor HDL inhibited the uptake of both labels from the two subfractions to the same extent. These studies, which differed from previously reported investigations by employing native homologous HDL subfractions of known apolipoprotein composition, demonstrated that apoE is not directly involved in the selective uptake of HDL cholesterol by the liver. In the absence of specific binding sites on liver membranes, it is suggested that an alternative mechanism might exist for the clearance of HDL cholesterol from the plasma.

Involvement of the Ca(2+)-dependent phosphorylation of a 20 kDa protein in the proliferative effect of high-density lipoproteins (subclass 3) on the adenocarcinoma cell line A549

Previous studies from our laboratory demonstrated that high-density lipoproteins (subclass 3; HDL3) bind to sites specific for apolipoprotein AI on the human adenocarcinoma cell line A549 and that HDL3 binding promotes a mitogenic effect [Favre, Tazi, Le Gaillard, Bennis, Hachem and Soula (1993) J. Lipid Res. 34, 1093-1106]. In the present study, we have examined the cell proteins that showed modified phosphorylation after binding of HDL3 to specific sites, and the roles of Ca2+ and protein kinase C. Native HDL3 (but not tetranitromethane-modified HDL3) and Ca2+ ionophore A23187 strongly enhanced the phosphorylation of a 20 kDa protein (x 3.6) and, to a lower extent, the phosphorylation of 24 and 28 kDa proteins (x 2.2 and 2.6 respectively). The two effectors were equally able to stimulate cell growth. Down-regulation of protein kinase C by a 24 h incubation of cells with phorbol myristate acetate prevented the effects of HDL3 on the phosphorylation of 24 and 28 kDa proteins. However, the extent of phosphorylation of the 20 kDa protein was not affected. In contrast, activation of protein kinase C by a short incubation with phorbol myristate acetate resulted in inhibition of proliferation and an increase in 24 and 28 kDa (but not 20 kDa) protein phosphorylation. These results suggest that HDL3 putative receptors exert their proliferative effect on A549 cells through activation of a Ca(2+)-dependent protein kinase. This kinase activity is not modulated by phorbol ester and thus may be a calmodulin kinase or an isoenzyme of protein kinase C that is independent of phorbol ester. It allows a subsequent 20 kDa protein to be phosphorylated.

Omega-3 fatty acids in smooth muscle cell phospholipids increase membrane cholesterol efflux

The aim of our work was to determine whether fatty acid modifications in smooth muscle cell phospholipids affect cholesterol efflux and desorption. [3H]Cholesterol was used to label cholesterol pools in the whole cell or selectively in the plasma membrane. Cells were incubated for 12 h in order to increase oleate, linoleate, arachidonate, eicosapentaenoic acid (EPA) or docosahexaenoic acid (DHA) in phospholipids. Cholesterol efflux was monitored using native or tetranitromethane modified high-density lipoprotein3 (HDL3). When all cholesterol pools were labeled, the efflux from cells treated with different fatty acids were not different. Plasma membrane cholesterol efflux remained unchanged after oleate, linoleate or arachidonate treatments, but was markedly increased after EPA and DHA enrichment, both with native HDL3 and with tetranitromethane-high-density lipoprotein. These results suggest that the positive effects of n-3 fatty acid consumption on the atherosclerotic process could be linked in part to an increase in plasma membrane cholesterol efflux from vascular smooth muscle cells.

Structural relation between HDL-binding proteins in porcine liver

We have found strong evidence for a relation between three high-density lipoprotein (HDL)-binding proteins of 90, 110, and 180 kDa in porcine liver that were detected by ligand blotting. Because HDL-binding proteins with identical molecular masses were detected in human liver, all subsequent experiments were performed with porcine liver proteins. An antiserum raised against a highly purified preparation of the 90-kDa HDL-binding protein, designated 90-PC, showed cross-immunoreactivity with the 110- and 180-kDa HDL-binding proteins. Purified protein preparations of the 90-, 110-, and 180-kDa HDL-binding proteins were obtained and analyzed by polyacrylamide gel electrophoresis with sodium dodecyl sulfate. Under nonreducing conditions these preparations showed protein bands with the expected molecular masses. Reduction of these preparations resulted in protein bands of 90 kDa. Ligand blotting experiments with 125I-HDL showed protein bands of 90, 110, and 180 kDa under nonreducing conditions and a 90-kDa protein band in all three preparations under reducing conditions. Immunoblotting experiments with 90-PC antiserum resulted in a similar pattern. The three protein preparations were then subjected to cyanogen bromide cleavage and the resulting peptides separated on gel. Immunoblotting with the 90-PC antibody revealed a pattern of protein bands that was remarkably similar in all three protein preparations. Immunohistochemical localization studies with the 90-PC antibody showed that the HDL-binding proteins were present both at the borders of the sinusoids as well as within the hepatocellular plates.(ABSTRACT TRUNCATED AT 250 WORDS)