Characteristics of ten charge-differing subfractions isolated from human native low-density lipoproteins (LDL). No evidence of peroxidative modifications.

Native plasma low-density lipoproteins (LDL) were fractionated into ten subfractions with increasingly negative charges (LDL-1, the least electronegative, to LDL-10) using an anion-exchange column coupled to a fast protein-liquid chromatography system. Prior to fractionation, contaminating Lp(a) and apo A-I-containing lipoproteins were removed from LDL preparations by immunoaffinity chromatography. No significant difference in thiobarbituric acid-reactive substances, vitamin E or free aminogroup was found among subfractions, and no peptide with a higher molecular weight than apo B was observed on SDS-PAGE. We observed a gradual increase in cholesterol esters and a concomitant decrease in triglycerides from LDL-1 to LDL-7, and a reverse tendency from LDL-8 to LDL-10 (P < 0.01). Free cholesterol increased linearly from LDL-1 to LDL-10 (P < 0.01). LDL-1 to -3 had a homogeneous density profile, while other more electronegative subfractions showed a bimodal distribution with a second, minor peak of slightly higher density. A gradual increase in apolipoprotein C-III content related to LDL electronegativity was observed (P < 0.001). Apolipoprotein E content was also increased in the last two subfractions (P < 0.01). LDL subfractions displayed a similar binding fate on human fibroblasts, with the exception of the most electronegative subfractions [LDL-(9 + 10)], which bound more actively to apo B/E receptors (P < 0.05). This study shows that charge heterogeneity of native LDL is not related to lipid peroxidation or derivatization of free aminogroups of apolipoprotein B. In contrast, the enrichment of LDL in apolipoproteins other than apo B may explain, in part, the difference in their particle charge.

Enhanced modifications of low-density lipoproteins (LDL) by endothelial cells from smokers: a possible mechanism of smoking-related atherosclerosis.

OBJECTIVE: The aim of the study was to investigate LDL modifications by cultured human umbilical vein endothelial cells (HUVEC) from women smokers and non-smokers. METHODS: Modifications of LDL by HUVEC were studied by determining the values of thiobarbituric acid-reactive substances (TBARS) and the percentage of the most electronegative oxidized LDL fraction (fraction C) by using an ion-exchange chromatographic method based on fast protein liquid chromatography (FPLC). We also studied the cellular production of superoxide anion, the effect of various inhibitors and cysteine, and determined total intracellular glutathione content and cell growth. RESULTS: LDL exposed to HUVEC from smokers for 48 h showed significantly greater modifications than LDL exposed to HUVEC from non-smokers, as assessed by TBARS determination (19.4 +/- 1.2, mean +/- s.e.m., n = 20 versus 15.4 +/- 0.7 nmol/mg LDL, n = 19; P < 0.01) and by FPLC (percentage of fraction C: 39 +/- 7, n = 29 versus 14 +/- 3, n = 34; P < 0.001). Moreover, HUVEC from smokers produced significantly more superoxide anion than those from non-smokers (0.46 +/- 0.13 nmol/10(5) cell/min, n = 9 versus 0.22 +/- 0.05, n = 10; P < 0.05). Superoxide production, like cell-induced modification of LDL, was strongly dependent on the presence of cysteine in the medium. Furthermore, HUVEC from smokers had a significantly (P < 0.05) higher total intracellular glutathione content than those from non-smokers (39.9 +/- 3.1 nmol/mg, n = 9 versus 31.8 +/- 2.2, n = 7). Finally, HUVEC from smokers and non-smokers showed similar growth at 48 h. CONCLUSION: HUVEC from smokers converted significantly more LDL into an atherogenic form than HUVEC from non-smokers, a phenomenon that was not due to altered cell growth. HUVEC-mediated LDL modifications were strongly thiol-dependent, as both LDL modifications and superoxide anion production were inhibited in cysteine-free medium. Stimulation of cystine uptake by HUVEC, reflected by the enhanced total glutathione content, could account for the enhanced superoxide anion production. All these observations may be relevant to the pathophysiology of smoking-related cardiovascular disease.

Relationship between cell density and prolidase activity in human skin fibroblasts: effects of ascorbate and fructose.

To evaluate the influence of cell density on the activity of fibroblast prolidase (EC 3.4.13.9), we determined this activity in sparse and dense cultures. We also investigated, the effects of different concentrations of beta-D (-) fructose and L (+) ascorbate, which both increased cell density at confluency. For a fructose concentration of 25 mM, we observed that in the absence of glucose, intracellular total proteins increased 1.5-fold and prolidase specific activity, 1.8-fold. For ascorbate, a broad optimum concentration was found (range 0.01 - 0.50 mM). Addition to cultures of 0.1 mM ascorbate increased total proteins 1.4-fold, and doubled prolidase activity. This investigation was prompted by our previous results [J. Metab. Dis. 1983, 6, 27-31], confirmed here, and suggesting that increased prolidase activity at confluency was due to a rise in cell density.

Copper- and malondialdehyde-induced modification of high density lipoprotein and parallel loss of lecithin cholesterol acyltransferase activation.

Incubation of high density lipoproteins (HDL) with 0.1-10 microM copper ions resulted in a decrease in tryptophan residues and a moderate diminution of lysine residues. Polymerization of apolipoprotein AI (apo A-I) was only observed for the highest concentration of Cu2+. A dose-dependent loss in lecithin cholesterol acyl-transferase (LCAT) activity was noted. Following incubation with 10 mM malondialdehyde, the physicochemical properties of HDL were more pronouncedly affected, in terms of lipid peroxidation products, relative electrophoretic mobility and percentages of intact tryptophan and lysine residues. Polymerization of apo A-I occurred after 40 min incubation, and a time-dependent loss of LCAT activation was noted. Since the deficiency in LCAT activation was observed in relatively mild conditions, when no perturbation of the physico-chemical properties of the particle could be shown, the determination of LCAT activity appears to be a sensitive test for HDL discrete modification.

Determination of prolinase activity in plasma. Application to liver disease and its relation with prolidase activity.

We describe prolinase (EC 3.4.13.8) activity in human plasma for the first time. Optimum activity was obtained with prolylvaline as substrate and 0.02 mmol/L manganese concentration at pH 9.0. Moreover, preincubation with manganese was not required, contrary to prolidase (EC 3.4.13.9) activity. The mean value observed in 106 subjects without liver and renal disorders was 16 U/L +/- 14 (2 SD). We determined this plasma enzyme activity in patients with acute hepatitis and chronic liver disease. Plasma prolinase activity was strongly dependent upon cytolysis because of the high activity in liver and the low activity in plasma. Of 24 patients with chronic liver disease (4 chronic hepatitis and 20 cirrhosis) and without cytolysis, prolinase activity was slightly increased in only three patients, whereas prolidase activity was increased in 13. This could be due to a difference in the activation of these two enzymes in liver during the fibrotic process.

The effect of high density lipoprotein subfractions on endothelial eicosanoid secretion.

Epidemiological, clinical, and experimental studies have demonstrated that high density lipoproteins (HDL) are protective against atherosclerosis. However, the respective influence of two main HDL subfractions (HDL2 and HDL3) on atherosclerosis process is not yet clear. The present study was designed to determine, which HDL subfraction was antiatherogenic in terms of eicosanoid release by human umbilical vein endothelial cells (HUVEC). Endothelial cells were incubated for 4 hours with HDL2 or HDL3 and prostaglandins 6-keto-PGF1alpha, thromboxane B2 and prostaglandin E2 were measured by RIA in culture supernatant. HDL2 has a dose dependent stimulatory effect on 6-keto-PGF1alpha release without stimulatory effect on thromboxane B2 secretion. The 6-keto-PGF1alpha/thromboxane B2 ratio increased progressively from 1.65 to 4.65 for 0.39 to 6.25 mg HDL protein/ml. The pattern of prostanoid secretion under influence of HDL3 showed a predominant response in 6-keto-PGF1alpha and TxB2 release. As regards PGE2, both HDL subfractions stimulated considerably secretion of this prostanoid in a dose dependent manner. In terms of PGI2/TxA2 balance the better antiatherogenic effect was observed with HDL2 subfraction.

Oxidative modification of low-density lipoprotein by the human hepatoma cell line HepG2.

The human hepatoblastoma cell line HepG2 is a liver model commonly used for lipid metabolism studies. Numerous cell types have been found to oxidize low-density lipoprotein (LDL) but, to our knowledge, the effects of HepG2 cells on LDL have not been investigated. We found that LDL is modified by HepG2 cells through a peroxidative mechanism, as judged by an increase in TBARS content (which was prevented in the presence of the antioxidants vitamin E, 2,6-di-tertbutyl-cresol and probucol), increased degradation by J774 macrophages, decreased internalization by MRC5 fibroblasts, and aggregation of apo B. Aspirin and allopurinol, which inhibit cyclooxygenase and xanthine-oxidase activities, respectively, had no effect on HepG2-induced LDL modification, and neither did catalase, which dismutates hydrogen peroxide; or mannitol, which scavenges hydroxyl radicals. In contrast, superoxide dismutase, a superoxide anion scavenger, and glutamate and threonine, which alter cellular cystine uptake, prevented LDL modifications, as did the removal of cysteine/cystine from the culture medium. Oxidation of LDL by HepG2 cells might thus involve superoxide anion production and/or thiol metabolism.

Effect of long preincubation on the two forms of human erythrocyte prolidase.

The effect of prolonged preincubation for 24 h at 37 degrees C in the presence of 1 mmol/l manganese at pH 7.8 was investigated on the two forms of human erythrocyte prolidase after their separation by DEAE-Sephadex chromatography. Prolidase I activity, which was eluted in chromatographic fractions of low ionic strength of about 160 mmol/l NaCl, increased after preincubation and this activation was maintained throughout the subsequent steps of chromatofocusing, chromatography on Sephacryl S-200 in tandem with Blue-Sepharose, and Phenyl-Sepharose chromatography. It disappeared after the last step consisting of a second Blue-Sepharose chromatography. When the protein environment was restored, isolated prolidase I was reactivated by preincubation, and as a result, gly-pro procollagen dipeptide became the best substrate. Gly-pro-hyp procollagen tripeptide was also observed to have a strong activator effect. The activity of prolidase II, which was eluted in high ionic strength fractions of about 260 mmol/l NaCl during DEAE-Sephadex chromatography, diminished markedly after preincubation and was very low against the gly-pro substrate. These results seem to indicate that prolidase I is much more active than prolidase II in the intracellular degradation of procollagen.

Optimal conditions for prolidase assay by proline colorimetric determination: application to iminodipeptiduria.

Prolidase assay was reinvestigated by determining proline, using Chinard's method. Although several authors had previously tested this colorimetric reaction, accurate details regarding enzyme activity were not available. The need for greater sensitivity led to the introduction of several modifications: dialysis was eliminated and the substrate concentration and incubation time were changed. In addition, the reaction mixture was preincubated with Mn2+ for 24 h in order to triple prolidase activity. Color development followed at 90 degrees C, because of partial glycylproline hydrolysis at higher temperatures. The effect of several divalent cations on prolidase activity were tested with and without Mn2+. This modified assay was applied to erythrocytes, plasma and skin fibroblasts from a female patient with iminodipeptiduria.

Plasma prolidase and prolinase activity in alcoholic liver disease.

Prolidase (EC 3.4.13.9) and prolinase (EC 3.4.13.8) activity was measured in the plasma of 53 patients with alcoholic liver disease. Plasma prolinase activity was not correlated with histological characteristics in liver biopsies. In contrast, prolidase activity rose significantly (p less than 0.02) in cirrhotic patients with alcoholic hepatitis in comparison with those without alcoholic hepatitis. It also showed a significant positive correlation with ASAT activity (r = 0.505, p less than 0.001) and with the ASAT/ALAT ratio (r = 0.452, p less than 0.001). Plasma prolidase activity did not allow the differentiation of patients with reversible fibrosis from those with cirrhosis. The interest of this new marker is discussed.

Cell density affects prolidase and prolinase activity and intracellular amino acid levels in cultured human cells.

Prolidase (EC 3.4.13.9) and prolinase (EC 3.4.13.8) and intracellular amino acid levels in cultured human cells increased when cell density rose. Firstly, two normal fibroblast strains were continuously cultured for 21 days and these parameters were measured on days 3, 7, 10, 14, 17 and 21 after plating. Prolidase, prolinase and amino acid levels varied considerably depending on the duration of culture and growth rate. Secondly, we studied the action of different cadmium and cobalt concentrations on prolidase activity. These two effectors altered this enzyme activity, but secondarily to modifying cell density. Thirdly, prolidase activity was investigated in 8 control amniotic cell strains, with a view to prenatal diagnosis of inherited prolidase deficiency, and we noted the same cell density interference. Due to the large variations related to cell density, we recommend specifying the number of cells per unit surface, and avoiding the term 'cells at confluency' which is unduly vague.

Reactivity of lecithin-cholesterol acyl transferase (LCAT) towards glycated high-density lipoproteins (HDL).

Hyperglycaemia in diabetic patients results in non-enzymatic glycation of plasma proteins, including lipoproteins such as high-density lipoproteins (HDL). We studied the effects of in vitro HDL glycation on the activity of lecithin-cholesterol acyl transferase (LCAT), a key enzyme in HDL plasma metabolism. LCAT was prepared from non-diabetic subjects and HDL by sequential density ultracentrifugation (in the density range of 1.063-1.21 g/ml) from both diabetic and non-diabetic patients. HDL from non-diabetic patients were glycated in vitro by incubating lipoproteins with 100 mmol/l glucose for various times at 37 degrees C with sodium cyanoborohydride as reducing agent. Glycation of HDL protein was quantified by measuring the percentage of derived amino acid residues using the TNBS assay. Kinetic parameters of LCAT were first determined using native HDL from non-diabetic patients and in vitro glycated HDL. With native HDL, Km and Vmax were 51.1 +/- 4.2 mumol/l (n = 8) and 12.9 +/- 2.4 nmol/ml/h (n = 8), respectively. Enzyme reactivity, calculated as the Vmax/Km ratio, was 0.25 +/- 0.04 h-1 (n = 8). In the case of moderate glycation (derived residues < 30%; n = 19) a significant increase in both Km (18.2 +/- 3.4%; mean +/- S.D.) and Vmax (9.3 +/- 2.4%) was observed. In contrast, with a high level of glycation (derived residues > 30%; n = 8), both parameters fell (Km, 25 +/- 6.3%; Vmax, 34.1 +/- 3.3%). In addition, whatever the level of glycation, enzyme reactivity was lower in the presence of in vitro glycated HDL. This decrease in LCAT reactivity was not due to a peroxidative process nor to an alteration of the protein and lipid composition of in vitro glycated HDL. It could, however, be explained by glycation of lysine residues in apolipoprotein A-I, which is the most potent activator of LCAT. In a second series of experiments, native diabetic HDL preparations were used as LCAT substrate. No alteration in Km values was observed, but there was a significant decrease in both Vmax (28%) and enzyme reactivity (32%). This difference in Km and Vmax alterations between native diabetic HDL and in vitro glycated HDL with low levels of glycation might be explained by the impact of physiological modifications, other than glycation, which could differently affect the chemicophysical properties of HDL in diabetic patients.(ABSTRACT TRUNCATED AT 400 WORDS)

Analysis of dipeptides in urine by gas chromatography/mass spectrometry: implications for collagen breakdown in iminodipeptiduria following a study of the dipeptides by electron impact and chemical ionization.

Dipeptides in the urine of a patient suffering from dermatological purpura, associated with iminodipeptiduria , were determined by gas chromatography/mass spectrometry. The dipeptides were identified as R-proline and R'-hydroxyproline where R is any one of the residues, glycyl, alanyl, valyl, leucyl, isoleucyl, seryl, aspartyl, glutamyl, prolyl, phenylalanyl and R' is alanyl, valyl, leucyl or isoleucyl, seryl, prolyl, glutamyl, phenylalanyl. The predominance of proline- and hydroxyproline-containing dipeptides and the percentage distributions of the other amino acid residues, R and R', strongly implicate an abnormality of collagen metabolism. Structural assignments are confidently based on (a) gas chromatographic retention times, (b) electron impact mass spectra and automatic comparison with reference to spectra stored in a specialized library, (c) chemical ionization mass spectra with isobutane and methanol as reactant gases and (d) the use of deuteriated acetic anhydride as a derivatizing agent.

A method to screen for the antioxidant effect of compounds on low-density lipoprotein (LDL): illustration with flavonoids.

We used a recently described anion-exchange chromatographic method (Vedie et al. J Lipid Res 1991;32:1359) to study the protective effect of potential inhibitors of low-density lipoprotein (LDL) oxidation mediated by cupric ion. By way of an example, we studied eight flavonoids (flavone, 3-hydroxyflavone, chrysin, galangin, fisetin, morin, quercetin, and myricetin) as well as three non-flavonoid antioxidants, butylated hydroxytoluene (BHT), probucol, and vitamin C, as reference compounds. Each compound was tested at various concentrations (1-100 microM). For flavonoid concentrations of 10 microM, an index was calculated as the (LDL control-flavonoid)/(LDL control-probucol) ratio, in which each term is expressed as the percentage of the most electronegative LDL fraction (fraction E). If the index is positive, the flavonoid inhibits LDL oxidation. A value > 1 (3-hydroxyflavone and galangin) means greater activity than probucol, whereas a value < 1 means lower activity (fisetin). If the index is around 0 (flavone and chrysin), the flavonoid is inactive. Finally, a negative value reflects possible prooxidant activity (morin, quercetin, and myricetin). Our results show that this chromatographic method can be applied to screening new pharmacological agents for activity against LDL oxidation.

Prolidase and prolidase deficiency.

Prolidase deficiency seems to be a rather rare metabolic disorder. However, many new cases can be detected because screening is easy to perform and enzymatic confirmation allows the differentiation from other iminodipeptidurias . Clinical symptoms are briefly reviewed, while biological considerations and prolidase properties are exhaustively described. Methods for investigating urinary iminodipeptides are given with results. Moreover, several collagen modifications observed in this disorder led us to formulate a hypothesis for their mechanism. Genetic considerations and treatment attempts are discussed.

[Cryoglobulinemia and hepatitis C virus]. / Cryoglobulinémie et virus de l'hépatite C.

Cryoglobulins are immunoglobulins that precipitate at low temperatures. They are easy to characterize by using sensitive techniques such as immunofixation and immunoblotting. Recently, hepatitis C virus (HCV) has been found to be associated with essential mixed cryoglobulinemia. The different frequencies reported might be explained by heterogeneity in patient subpopulations and advances in serologic tests. In addition to commercial kits to detect HCV antibodies, polymerase chain reaction (PCR) of HCV RNA can now be applied to serum or cryoprecipitates. The pathogenic role of hepatitis C virus is well known, but the precise mechanism remains to be determined.

[Biology of the endothelial cell and atherogenesis]. / Biologie de la cellule endothéliale et athérogenèse.

Through their specific biological properties, vascular endothelial cells play a major role in maintaining the homeostasis of the cardiovascular system. The vascular endothelium participates actively in coagulation and fibrinolysis, contributes to regulating vascular tone, is involved in inflammation and immunological responses, produces several different stromal components, plays a crucial role in angiogenesis and wound-healing, and interacts with plasma lipoproteins. These physiological functions of endothelial cells are triggered by different endothelium derived mediators and are regulated by numerous environmental factors that can markedly modulate the functional state of these cells by affecting their biosynthetic capabilities, giving them different phenotypes in native, activated and injured states. Excessive endothelial activation leads to changes in endothelial cell gene expression, leading to what is referred to as a dysfunctional state. In this non-adaptative functional state, endothelial cells lose the ability to adjust, within the physiological range, some of their constitutive functions and express newly induced molecules, some of which act as proatherosclerotic factors. Activated endothelial cells thus participate actively in the pathogenesis of atherosclerosis.

[Modified LDL and atherosclerosis. Nature of modifications. Physicochemical and biological properties]. / LDL modifiées et athérosclérose. Nature des modifications. Propriétés physicochimiques et biologiques.

The role of plasma LDL in atherogenesis is now well established. Cholesteryl ester accumulation within macrophages leads to foam cell formation, an early atherosclerotic process. In vitro, foam cell formation scarcely ever occurs in the presence of native LDL. Modification of these lipoproteins is necessary for their binding to macrophage scavenger receptors. In vitro modifications reported have involved chemical reactions, physical mechanisms, enzymatic reactions, cellular interactions and association with macromolecules. In vivo, they can occur by glycation or desialylation, smoking or by hemodynamic interactions. Alterations of the physicochemical properties of LDL are induced by oxidation and include an increase in their density and their net electronegative charge, changes in lecithin composition, polyunsaturated fatty acid peroxidation of lipids and apolipoprotein B100 degradation. Apolipoprotein B100 fragmentation leads to an impairment of uptake through LDL receptors, while uptake through macrophage scavenger receptors is enhanced. Modified LDL have also other particularities such as cytotoxicity, chemotactism for circulating monocytes, inhibition of resident macrophage mobility, vasoconstriction, perturbation of the arachidonic acid cascade, involvement in haemostasis and immune mechanisms. Hypotheses concerning the role of modified LDL, in particular oxidized LDL, in atherogenesis open new therapeutic prospects.

[Charge heterogeneity of low density lipoproteins]. / Hétérogénéité de charge des lipoprotéines de basse densité (LDL).

Traditionally, low-density lipoprotein (LDL) are separated with respect to their size, density and apolipoprotein composition. Fractionation of LDL according to their electrical charge is also interesting as modified LDL have been implicated in the onset of atherosclerosis. This review discusses possible mechanisms underlying charge heterogeneity of human plasma LDL, such as oxidation, glycation, conjugation with aldehydes, carbamylation and changes in sialic acid content and protein composition.

Monitoring of heart allograft rejection by simultaneous measurement of serum beta 2-microglobulin and urinary neopterin.

In patients with heart transplant, the combined determination of serum beta 2-microglobulin and urinary neopterin, as rejection marker, prevented the interference by renal function and cyclosporin therapy. Unfortunately, the simultaneous measurement of these two parameters cannot distinguish between a rejection episode and the early stage of viral infection.