Oxidized low density lipoprotein (oxLDL) induces apoptosis in vascular cells. To elucidate the mechanisms involved in this apoptosis, we studied the apoptosisinducing activity in lipid fractions of oxLDL and the roles of two common mechanisms, ceramide generation and the activation of caspases, in apoptosis in human umbilical vein endothelial cells treated with oxLDL. We also studied the effects of antioxidants and cholesterol. oxLDL induced endothelial apoptosis in a time-and dosedependent fashion. Apoptosis-inducing activity was recovered in the neutral lipid fraction of oxLDL. Various oxysterols in this fraction induced endothelial apoptosis. Neither the phospholipid fraction nor its component lysophosphatidylcholine induced apoptosis. ox-LDL induced ceramide accumulation temporarily at 15 min in a dose-dependent fashion. Two inhibitors of acid sphinogomyelinase inhibited both the increase in ceramide and the apoptosis induced by oxLDL. Furthermore, a membrane-permeable ceramide (C 2 -ceramide) induced endothelial apoptosis. These findings demonstrated that ceramide generation by acid sphingomyelinase is indispensable for the endothelial apoptosis induced by oxLDL. Inhibitors of both caspase-1 and caspase-3 inhibited the apoptosis, suggesting that oxLDL induced apoptosis by activating these cysteine proteases. The antioxidants butylated hydroxytoluene and superoxide dismutase but not catalase inhibited the apoptosis induced by oxLDL or 25-hydroxycholesterol. This suggests not only that superoxide plays an important role but also that a critical interaction between oxLDL and the cell takes place on the outer surface of the membrane, because superoxide dismutase is not membrane-permeable. Exogenous cholesterol also inhibited the apoptosis. Our study demonstrated that neutral lipids in oxLDL induce endothelial apoptosis by activating membrane sphingomyelinase in a superoxidedependent manner, as well as by activating caspases.

Glycosylation is a major pathway for posttranslational modification of tissue protein and begins with nonenzymatic addition of carbohydrate to the primary amino groups. Excessive glycation of tissue protein has been implicated in the pathogenesis of diabetes and ageing. While glycation of aminophospholipids has also been postulated, glycated aminophospholipids have not been isolated. Using normal phase HPLC with on-line electrospray mass spectrometry we found glycated ethanolamine phospholipids to make up 10-16% of the total phosphatidylethanolamine (PE) of the red blood cells and plasma of the diabetic subjects. The corresponding values for glycated PE of control subjects were 1-2%.Key words: Glucosylated aminophospholipid; Glucose; Phosphatidylethanolamine; Phosphatidylserine; Electrospray; Thin-layer chromatography; Liquid chromatography, mass spectrometry; Normal phase HPLC bovine brain phosphatidylinositol (PI) and sphingomyelin (SPH) were obtained from Sigma Chemical Co., St. Louis, MO. All chemicals were of reagent grade quality, while the solvents were of chromatographic purity and were obtained from local suppliers. The purity of the reference compounds was ascertained by thin-layer chromatography (TLC) [3,7]. Isolation of phospholipids from bloodBlood was obtained from six diabetic patients and six non-diabetic donors. The diabetics were selected for elevated blood glucose levels indicated by their content of glycosylated hemoglobin (9-15%). EDTA blood was centrifuged (2300xg for 10 min) in a swinging bucket rotor to separate the plasma from the red cells. The cells were washed three times with five volumes of phosphate buffered saline (150 mM NaC1, 50 mM sodium phosphate, pH 8.0) and centrifuged (2300xg for 10 min). The red blood cell phospholipids were extracted according to Rose and Oaklander [8]. The plasma phospholipids were extracted with chloroform-methanol 2:1 modified from Folch et al. [9]. Glucosylated PE could be stored at -20°C in neutral chloroform methanol for several days without decomposition. The Schiff base dissociated in dilute acetic acid.

Synthetic cholesteryl 5-oxovalerate and 9-oxononanoate were used as reference standards for the isolation and identification of cholesteryl ester core aldehydes from tert-butyl hydroperoxide/Fe++ oxidation of synthetic and natural cholesteryl esters. The core aldehydes were recovered from the peroxidation products by thin-layer chromatography as the free aldehydes or the 2,4-dinitrophenylhydrazones and were identified, respectively, by gas-liquid chromatography (GLC) and by GLC combined with mass spectrometry (GC/MS) or by reverse-phase high-performance liquid chromatography (HPLC) and by HPLC with MS (LC/MS). The core aldehydes produced by peroxidation of cholesteryl linoleate were identified as mainly 9-oxononanoates of cholesterol and oxycholesterols, with smaller amounts of the 8-oxooctenoates, 10-oxodecenoates, 11-oxoundecenoates and 12-oxododecenoates. Peroxidation of cholesteryl arachidonate yielded 5-oxovalerates of cholesterol and the oxycholesterols as the main products with smaller amounts of the 4-oxobutyrates, 6-oxohexenoates, 7-oxoheptenoates, 8-oxooctenoates, 9-oxononenoates, 9-oxononadienoates and 10-oxodecadienotes. The oxycholesterols resulting from the peroxidation of the steroid ring were identified as mainly 7-keto-, 7 alpha-hydroxy- and 7 beta-hydroxy-cholesterols and 5 alpha,6 alpha- and 5 beta,6 beta-epoxy-cholestanols. Cholesteryl palmitate and oleate did not yield core aldehydes in the present peroxidation system. In these esters, the sterol and linoleic acid moieties appeared to be oxygenated at about the same rate, while the arachidonic acid moiety reacted more rapidly than did the sterol moiety.

No abstract

We identified and quantified the hydroperoxides, hydroxides, epoxides, isoprostanes, and core aldehydes of the major phospholipids as the main components of the oxophospholipids (a total of 5-25 pmol/micromol phosphatidylcholine) in a comparative study of human atheroma from selected stages of lesion development. The developmental stages examined included fatty streak, fibrous plaque, necrotic core, and calcified tissue. The lipid analyses were performed by normal-phase HPLC with on-line electrospray MS using conventional total lipid extracts. There was great variability in the proportions of the various oxidation products and a lack of a general trend. Specifically, the early oxidation products (hydroperoxides and epoxides) of the glycerophosphocholines were found at the advanced stages of the plaques in nearly the same relative abundance as the more advanced oxidation products (core aldehydes and acids). The anticipated linear accumulation of the more stable oxidation products with progressive development of the atherosclerotic plaque was not apparent. It is therefore suggested that lipid infiltration and/or local peroxidation is a continuous process characterized by the formation and destruction of both early and advanced products of lipid oxidation at all times. The process of lipid deposition appears to have been subject to both enzymatic and chemical modification of the normal tissue lipids. Clearly, the appearance of new and disproportionate old lipid species excludes randomness in any accumulation of oxidized LDL lipids in atheroma.