Oxidation of liposomal cholesterol and its effect on phospholipid peroxidation.

Lipid peroxidation is believed to play an important role in the pathogenesis of many diseases. Much research has therefore been devoted to peroxidation of different lipids in biomembranes and in model systems (liposomes) of different compositions. Yet, in spite of the relative simplicity of the liposomes, the existing literature is insufficient to reach definite conclusions regarding basic questions including the susceptibility of cholesterol to oxidation, its effect on the peroxidation of polyunsaturated phospholipids such as palmitoyllinoleoylphosphatidylcholine (PLPC) and how cholesterol influences the effect of water-soluble antioxidants such as urate on the peroxidation. The aim of the present study was to clarify these issues. Its major findings are that: (i) AAPH-induced peroxidation of cholesterol is slow and independent of the peroxidation of PLPC. In turn, AAPH-induced peroxidation of PLPC is not affected by cholesterol, independent of the presence of urate in the system. (ii) Cholesterol is not susceptible to copper-induced oxidation, but its inclusion in PLPC liposomes affects the peroxidation of PLPC, slowing down the initial stage of oxidation but promoting later stages. (iii) Addition of urate accelerates copper-induced peroxidation of PLPC in the absence of cholesterol, whereas in cholesterol-containing liposomes it inhibits PLPC oxidation. We attribute the complexity of the observed kinetics to the known cholesterol-induced rigidization of liquid crystalline bilayers.

The effect of cholesterol on the solubilization of phosphatidylcholine bilayers by the non-ionic surfactant Triton X-100.

Most of the studies on the solubilization of model membranes by Triton X-100 (TR) involve one lipid. The aim of the present study was to evaluate the effect of the addition of cholesterol on the solubilization of bilayers made of palmitoyloleoylphosphatidylcholine (POPC) or dipalmitoylphosphatidylcholine (DPPC). Detailed investigation of the kinetics of solubilization of the cholesterol-containing bilayers by TR at different temperatures reveals that: (i) At 4 degrees C, solubilization of both systems is relatively slow. Hence, in order to prevent misleading conclusions from turbidity measurements it is important to monitor the solubilization after steady-state values of optical density (OD) are reached. (ii) Studies of the temperature-induced changes of the aggregates present in mixtures of TR, POPC and cholesterol indicate that the state of aggregation at all temperatures (including 4 degrees C) represents equilibrium. By contrast, for DPPC/cholesterol/TR mixtures "kinetic traps" may occur not only at 4 degrees C but at higher temperatures as well (e.g. 37 degrees C). (iii) The presence of cholesterol in POPC bilayers makes the bilayers more resistant to solubilization at low temperatures, especially at 4 degrees C. As a consequence, the temperature dependence of the TR concentration required for complete solubilization (Dt(sol)) is no longer a monotonically increasing function (as for POPC bilayers) but a bell-shaped function, with a minimum at about 25 degrees C. Inclusion of cholesterol in DPPC bilayers makes the bilayers more resistant to solubilization at all temperatures except 4 degrees C. In this system, we observe a bell-shaped dependence of Dt(sol) on temperature, with a minimum at 37 degrees C. (iv) Both the rate of vesicle size growth and the rate of the solubilization of POPC vesicles are not affected by the inclusion of cholesterol in the bilayers. Similarly, cholesterol did not affect significantly the rate of size growth of DPPC bilayers at all temperatures, but reduced the rate of solubilization at 4 degrees C.

Temperature-dependence of the solubilization of dipalmitoylphosphatidylcholine (DPPC) by the non-ionic surfactant Triton X-100, kinetic and structural aspects.

Most of the studies on the solubilization of model membranes conducted thus far involved model membranes made of liquid-crystalline phospholipids. Relatively little is known on the influence of temperature and of the phase of the lipid bilayers on their solubilization by detergents. The aim of the present study was to gain knowledge about the temperature and phase dependence of the solubilization of phospholipid bilayers by the non-ionic detergent Triton X-100 (TR). Detailed investigation of the kinetics of the solubilization of dipalmitoylphosphatidylcholine (DPPC), as well as of palmitoyloleoylphosphatidylcholine (POPC) by TR at different temperatures reveals that: (i) solubilization of DPPC is a relatively slow process, especially below Tm. This means that in order to prevent misleading conclusions it is important to monitor the solubilization after a steady state is established. (ii) Both the steady state structure and size of DPPC/TR aggregates and the kinetics of solubilization depend on temperature. (iii) The TR concentration required for solubilization of POPC bilayers is an increasing function of temperature, although no phase change of bilayers occurs in the studied temperature range. (iv) Detailed studies of the temperature-induced changes of the aggregates present in DPPC/TR or POPC/TR mixtures suggest that the state of aggregation at any temperature above 23 degrees C represents equilibrium. By contrast, for DPPC/TR mixtures at 4 degrees C all the processes are very slow, which complicates the interpretation of results obtained through the common practice of studying "rafts" by investigating detergent-resistant membranes.

Carcinoid tumor mistaken for persistent neurogenic bowel symptoms in a patient with paraplegia: a case report.

Neurogenic bowel in spinal cord injury (SCI) can present with constipation and diarrhea as ongoing problems. Usually, these manifestations are adequately controlled with modification in the bowel program. When these symptoms persist, other causes should be considered. This case report describes a jejunal carcinoid tumor with colonic extension that was diagnosed in a paraplegic patient with persistent constipation and diarrhea. A 39-year-old man sustained a T1 paraplegia with neurogenic bowel and bladder dysfunction from a gunshot wound. His bowels were initially managed adequately with digital disimpaction. Over the next 8 years, he had intermittent constipation that was managed with the addition of various suppositories. He then developed progressively worsening constipation, and other gastrointestinal (GI) symptoms. Although his symptoms initially resolved with medical management, the constipation worsened. Upper endoscopy revealed a submucosal bulge in the duodenal bulb. A month later, gallstones were found on renal ultrasound performed to evaluate recurrent urinary tract infections. He underwent cholecystectomy, but his GI symptoms persisted over the next several months. Repeat upper endoscopy subsequently revealed an ulcerated tumor at the duodenojejunal flexure. An upper-GI scan with small bowel follow through showed a proximal jejunal mass. The patient underwent laparotomy with resection of the mass. Final pathologic diagnosis was malignant carcinoid tumor. This case shows the importance of entertaining other clinical entities in patients with SCI when constipation and diarrhea persist despite adequate management.

Inhibition of phospholipase A(2) as a therapeutic target.

The hydrolysis of cell membrane phospholipids by phospholipase A(2) (PLA(2)) leads to the production of numerous lipid mediators of diverse pathological conditions, mainly inflammatory diseases. These include lysophospholipids and their derivatives, and arachidonic acid and its derivatives (the eicosanoids). Both these groups of mediators are produced predominantly by the secretory PLA(2)s (sPLA(2)s) which hydrolyze the phospholipids of the cell surface membrane. Protection of cell membrane from these 'inflammatory enzymes' can therefore be used for the treatment of inflammatory processes. A prototype of cell-impermeable PLA(2) inhibitors, which protect the cell membrane from different sPLA(2)s without affecting vital phospholipid metabolism, is presented and discussed in the present review.

Kinetic analysis of copper-induced peroxidation of HDL, autoaccelerated and tocopherol-mediated peroxidation.

Comparison of the kinetic profiles of copper-induced peroxidation of HDL and LDL at different copper concentrations reveals that under all the studied experimental conditions HDL is more susceptible to oxidation than LDL. The mechanism responsible for HDL oxidation is a complex function of the copper/HDL ratio and of the tocopherol content of the HDL. At high copper concentrations, the kinetic profiles were similar to those observed for LDL oxidation, namely, relatively rapid accumulation of oxidation products, via an autoaccelerated, noninhibited mechanism, was preceded by an initial "lag phase." Under these conditions, the maximal peroxidation rate (V(max)) of HDL and LDL depended similarly on the molar ratio of bound copper/lipoprotein. Analysis of this dependency in terms of the binding characteristics of copper to lipoprotein, yielded similar dissociation constant (K = 10(-6) M) but different maximal binding capacities for the two lipoproteins (8 Cu(+2)/HDL as compared to 17 Cu(+2)/LDL). Given the size difference between HDL and LDL, these results imply that the maximal surface density of bound copper is at least 2-fold higher for HDL than for LDL. This difference may be responsible for the higher susceptibility of HDL to copper-induced oxidation in the presence of high copper concentrations. At relatively low copper concentrations, the kinetic profile of HDL oxidation was biphasic, similar to but more pronounced than the biphasic kinetics observed for the oxidation of LDL lipids at the same concentration of copper. Our results are consistent with the hypothesis that the first phase of rapid oxidation occurs via a tocopherol-mediated-peroxidation (TMP) mechanism. Accordingly, enrichment of HDL with tocopherol resulted in enhanced accumulation of hydroperoxides during the first phase of copper-induced oxidation. Notably, the maximal accumulation during the first phase decreased upon increasing the ratio of bound copper/HDL. This behavior can be predicted theoretically for peroxidation via a TMP mechanism, in opposition to autoaccelerated peroxidation. The possible pathophysiological significance of these findings is discussed.

Interaction of hyaluronic acid-linked phosphatidylethanolamine (HyPE) with LDL and its effect on the susceptibility of LDL lipids to oxidation.

The amphiphilic polysaccharide hyaluronic acid-linked phosphatidylethanolamine (HyPE), synthesized by covalently binding dipalmitoyl-phosphatidylethanolamine (DPPE) to short chain hyaluronic acid (mol. wt. approximately = 30 000), interacts with low-density lipoproteins (LDL), to form a 'sugar-decoration' of the LDL surface. This results in an increase in the apparent size of the LDL particles, as studied by photon correlation spectroscopy, and in broadening of the 1H NMR signals of the LDL's phospholipids. Experiments conducted with fluorescently-labeled HyPE indicate that the interaction of HyPE with LDL involves incorporation of the hydrocarbon chains of this amphiphilic polysaccharide into the outer monolayer of the LDL. This interaction also inhibits the copper-induced oxidation of the LDL polyunsaturated fatty acids, avoiding oxidation altogether when the concentration of HyPE is higher than a tenth of the concentration of the LDL's phospholipids. This can not be attributed to competitive binding of copper by HyPE. We propose that the protection of LDL lipids against copper-induced oxidation is due to formation of a sugar network around the LDL.

Lipid oxidation in unfractionated serum and plasma.

In an attempt to develop an assay for the susceptibility of plasma lipids to oxidation, we have studied the kinetics of copper-induced oxidation in diluted serum and plasma prepared with different anticoagulants (heparin, citrate and EDTA) by monitoring the absorbance of oxidation-products at several wavelengths. These studies revealed the complex and interrelated effects of the water-soluble antioxidant ascorbic acid, citrate and chloride ions on the kinetics of copper-induced oxidation of plasma lipids. Specifically, the onset of oxidation induced by copper-citrate chelates is only slightly affected by chloride ions and is accelerated upon increasing the copper concentration. By contrast, in the absence of citrate, the lag preceding oxidation in diluted serum or plasma (but not the maximal rate of oxidation) depends markedly on the chloride concentration in the diluting medium. In the absence of Cl-, the lag preceding oxidation is a decreasing saturable function of copper concentration, whereas in a normal phosphate-buffered saline solution (PBS), the lag shows a biphasic dependence on copper concentration such that at copper concentrations above 10-30 microM (depending on the extent of plasma dilution), increasing the concentration of copper results in prolongation of the lag. This dependence of copper-induced oxidation on the concentration of copper is not observed for dialyzed serum unless ascorbic acid is added. Our interpretation of these results is that water-soluble reductants and chloride ions act synergistically to stabilize Cu+, on the expense of Cu2+. Quenching of free radicals by Cu+ may be responsible for the prolongation of the lag at high copper concentrations, with no reduction of the maximal rate of oxidation. In spite of the complex dependencies described above, spectrophotometric monitoring of the kinetics of oxidation of plasma lipids, under 'optimized conditions' (50-fold diluted serum, in PBS containing 720 microM sodium citrate and 100 microM copper), agrees with independent measurements of the consumption of polyunsaturated fatty acids. Hence, the spectroscopic method may become useful for evaluation of the susceptibility of plasma lipids to oxidation. This possibility, however, has yet to be elucidated through investigations of the correlation between the susceptibility of serum lipids to copper-induced oxidation in vitro and clinical factors of significance.

LDL-associated phospholipase A does not protect LDL against lipid peroxidation in vitro.

The irreversible proteinase inhibitor Pefabloc (4-[2-aminoethyl] benzenesulfonyl fluoride) inactivates LDL-catalyzed hydrolysis of the short-chain fluorescent phospholipid C6-NBD-PC (1-acyl-2-(N-4-nitrobenzo-2-oxa-1,3-diazole)-aminocaproyl phosphatidylcholine). The dose-dependence of this inactivation is similar to that obtained previously for the inhibitory effect of Pefabloc on the hydrolysis of platelet activating factor (PAF) by the LDL-associated PAF acetylhydrolase (PAF-AH), in agreement with the notion that the hydrolysis of C6-NBD-PC and PAF is catalyzed by the same enzyme (LDL-associated phospholipase A; LDL-PLA). This conclusion is also supported by the finding that hydrolysis of C6-NBD-PC by LDL becomes inactivated by LDL oxidation only at late stages of the oxidation, similar to the effect of oxidation on the hydrolysis of PAF by the LDL-associated PAF-AH. Under conditions of complete inactivation of this enzyme towards C6-NBD-PC, the kinetics of lipid peroxidation, induced either by copper ions or by the free radical generator AAPH at varying doses of the prooxidant, was similar to that observed when the PLA was active (i.e., in the absence of Pefabloc). Hence, LDL-associated PLA (PAF-AH) does not protect LDL lipids from peroxidation. Similar results were obtained with fractionated LDL in albumin-containing buffer and for non-fractionated serum, in which copper-induced peroxidation was also not influenced by inactivation of the enzyme responsible for hydrolysis of C6-NBD-PC. Phospholipolysis of short chain phospholipids by LDL-PLA may still play a protective role against the toxic effects of oxidized phospholipids by reducing their internalization into cells (Schmitt et al. 1995).

Kinetic analysis of copper-induced peroxidation of LDL.

We have employed our recently developed spectroscopic method of continuous monitoring of lipid oxidation to study the formation and decomposition of hydroperoxides in the time course of LDL oxidation. The results show satisfactory agreement with simulated time courses based on the following assumptions: (a) Both the rates of formation and decomposition of hydroperoxides depend on the ratio of bound copper to LDL as computed under the assumption that each LDL particle has 17 equivalent copper binding sites characterized by a dissociation constant K = 1 microM. (b) Peroxidation is initiated by copper-catalyzed decomposition of hydroperoxides (LOOH) into peroxy radicals (LOO.) and other products, including dienals. Under these assumptions, the rate of accumulation of LOOH can be computed from the equation (equation in text). The agreement between the simulated and experimentally-observed kinetics supports the assumptions used for simulations. The close agreement between the values of lipid oxidizability (kp/square root 2kt) obtained for LDL (0.035 (Ms)[-1/2]) and previously published data on the oxidizability of linoleates (0.02-0.11 (Ms)[-1/2]) lends further support for these assumptions.

The effect of albumin on copper-induced LDL oxidation.

In an attempt to gain deeper understanding of the mechanism or mechanisms responsible for the protective effect of serum albumin against Cu(2+)-induced peroxidation of low density lipoprotein (LDL), we have examined the influence of the concentrations of bovine serum albumin (BSA), Cu2+ and LDL on the kinetics of peroxidation. Since the common method of monitoring the oxidation by continuous recording of the absorbance of conjugated dienes at 234 nm cannot be used at high BSA-concentrations because of the intensive absorption of BSA, we have monitored the time-dependent increase of absorbance at 245 nm. At this wavelength, conjugated dienes absorb intensely, whereas the background absorbance of BSA is low. Using this method, as well as the TBARS assay for determination of malondialdehyde, over a large range of BSA concentrations, we show that in many cases the influence of BSA on the kinetics of oxidation can be compensated for by increasing the concentration of copper. This reconciles the apparent contradiction between previously published data. Detailed studies of the kinetic profiles obtained under different conditions indicate that binding of Cu2+ to albumin plays the major role in its protective effect while other mechanisms contribute much less than copper binding. This conclusion is consistent with the less pronounced effect of BSA on the oxidation induced by the free radical generator AAPH. It is also shown that the copper-albumin complex is capable of inducing LDL oxidation, although the kinetics of the latter process is very different from that of copper-induced oxidation. Nevertheless, when compared to copper induced oxidation at similar concentration of the oxidation-promotor, the kinetics of oxidation induced by copper-albumin complex is very different and is consistent with a tocopherol mediated peroxidation, characteristic under low radical flux. Similar kinetics was observed for copper-induced oxidation only at much lower copper concentrations.

Copper-induced lipid oxidation in unfractionated plasma: the lag preceding oxidation as a measure of oxidation-resistance.

In spite of the pathological significance of oxidation of plasma lipids, no method is currently available for the evaluation of the susceptibility of these lipids to oxidation in unfractionated plasma. Here we demonstrate that copper-induced oxidation of diluted plasma, in the presence of citrate can be monitored continuously by recording the absorbance at 245 nm. The kinetics of accumulation of oxidation products in unfractionated plasma is a sum of lipid oxidation products obtained in low and high density lipoproteins isolated from the same plasma. The kinetic profiles are reproducible and can be performed with plasma samples even after prolonged storage at 4 degrees C (up to two months) or after freezing and thawing of the plasma. Being simple and reproducible, yet correlating with the oxidizability of low and high density lipoproteins, this method can be used to evaluate the "oxidation resistance" of plasma lipids and thus serve as a standard index of the susceptibility of the plasma lipids of patients to oxidation-inflicted pathologies, including atherosclerosis.

Oxidation of low density lipoprotein upon sequential exposure to copper ions.

Copper-induced LDL oxidation is characterized by an 'induction phase' (lag phase) during which the endogenous antioxidants are consumed, followed by a 'propagation phase' in which the LDL-associated polyunsaturated fatty acids are oxidized. Oxidation products may play an important role in the propagation of the oxidative process in the arterial intima as they increase the permeability of the damaged endothelium to various plasma components, including LDL. We therefore found it of interest to investigate the kinetics of LDL oxidation in vitro under conditions where LDL is sequentially exposed to Cu(2+)-induced oxidation. The results of our studies demonstrate that when native LDL is exposed to copper oxidation in a medium containing oxidized LDL, oxidation of the added LDL may be almost instantaneous. Furthermore, even when native LDL is added to 'oxidizing LDL' towards the end of the lag phase or during the propagation phase it becomes oxidized after a very short lag. This oxidation process, occurring in spite of the possible protective effect of the antioxidants present in the newly added LDL, indicates that although antioxidants prolong the latency period by preventing the formation of active free radicals, when such radicals are present in the system, oxidation propagates. These results lend strong support to the generally accepted paradigm regarding the mechanism of propagation of lipid oxidation. In view of the effect of oxidation products on the permeability of the endothelium, the observed shortening of the lag period may result in a vicious cycle, independent of the LDL-associated antioxidants, leading to continuing oxidation and foam cell formation.

Re-evaluation of the structure of low density lipoproteins.

Low density lipoprotein (LDL) is an established atherogenic factor. Much effort has therefore been devoted to elucidation of its structure, yielding the generally accepted model according to which the neutral lipids (cholesterol ester and triglycerides) form a lipid core emulsified by phospholipids, cholesterol and the amphipathic Apolipoprotein B. Yet, the detailed structure of LDL is not clear. The present work was carried out with the aim of re-evaluating the LDL structure using the minimal number of assumptions: in view of the previously noted surface deficit (lack of sufficient PL and cholesterol to cover the surface of the lipid core) we have assumed that polar head groups are not covered by apo B. Other than that, we have 'allowed' Apo B to penetrate into the PL monolayers and the lipidic core and to pertrude into the solution (be elevated above the PL head group level). We have also 'allowed' neutral lipid penetration into the monolayer and variation of the thickness of the phospholipid monolayers within reasonable boundaries. Based on the established values of relevant constants (molecular weights and volumes, densities and surface areas) we have computed the radius of the particle, the penetration of Apo B into lipidic milieus and the fraction of the surface area covered by Apo B as functions of the LDL composition, the monolayer thickness and the 'elevation' of Apo B above this monolayer. These computations show that at least 40% of the LDL surface must be covered by protein and that the protein penetrates, on the average, only about a half of the PL monolayer. Thus it is not very likely to penetrate into the lipid core. These general features are preserved in the smaller LDL particles of hypertriglyceridemic patients. Assuming that no PL head group is covered by Apo B, the previously described immobilization of 20% of the phospholipids is likely to result from the interaction of Apo B with neighboring PL. According to our computations this can be regarded consistent with the previously proposed arrangement of the apo B as a '3-4 domain structure' or a long string configuration but inconsistent with 'one domain' or 'twenty domain' structures.

Long-term desipramine or chlorpromazine treatment and rat brain N-methyl-D-aspartate receptor.

The effects of tricyclic antidepressants and antipsychotic drugs on the NMDA receptor in rat brain were investigated. Both types of drugs, in the micromolar concentration range and with Hill coefficients greater than unity, inhibited the binding of [3H]TCP to rat brain membranes. Tricyclic antidepressants containing secondary amine on the aliphatic side chain were more potent than the corresponding tertiary amine derivatives. Antipsychotic drugs containing large side chains were much less active than chlorpromazine. Neither tricyclic antidepressants nor antipsychotic drugs had an inhibitory effect on [3H]CPP binding to the glutamate site of the NMDA complex. Chronic administration of desipramine or chlorpromazine did not change the binding parameters of [3H]TCP to rat cerebral cortex membranes. It is possible that these psychoactive drugs bind to a site adjacent to the [3H]TCP binding site within the ionic channel of the NMDA receptor complex.