J Biol Chem, Vol. 274, Issue 51, 36409-36414, December 17, 1999

Novel Physiological Function of Sphingomyelin in Plasma
INHIBITION OF LIPID PEROXIDATION IN LOW DENSITY LIPOPROTEINS^*

Papasani V. Subbaiah, Veedamali S. Subramanian, and Kewei Wang

From the Departments of Medicine and Biochemistry, Rush Medical College, Chicago, Illinois 60612

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Although sphingomyelin (SPH) is a major constituent of all lipoproteins, its physiological function in plasma is not known. In this study, we tested the hypothesis that SPH inhibits lipid peroxidation in low density lipoproteins (LDL) because of its effects on surface fluidity and packing density and that the relative resistance of the buoyant LDL to oxidation, compared with the dense LDL, is partly due to their higher SPH content. Depletion of SPH by treatment with SPHase resulted in shortened lag times and increased rates of oxidation in both LDL subfractions, as measured by the conjugated diene formation in the presence of Cu²⁺. Oxidation of LDL by soybean lipoxygenase was similarly stimulated by the degradation of SPH. Oxidation-induced fluorescence decay of diphenylhexatriene-labeled phosphatidylcholine (PC), equilibrated with LDL-PC, was accelerated significantly by the enzymatic depletion of SPH from the lipoprotein. Oxidation of 16:0-18:2 PC in the proteoliposomes was inhibited progressively by the incorporation of increasing amounts of egg SPH into the liposomes. Treatment of SPH-containing proteoliposomes with SPHase reversed the effect of SPH, showing that the presence of intact SPH is necessary for the inhibition of oxidation. Although the incorporation of SPH into the same liposome as the PC (intrinsic SPH) protected the PC against oxidation, the addition of SPH liposomes to PC liposomes (extrinsic SPH) was not effective. Oxidation of 16:0-18:2 PC in liposomes was also inhibited by the incorporation of dipalmitoyl-PC, but not by free cholesterol. These results suggest that SPH acts as a physiological inhibitor of lipoprotein oxidation, possibly by modifying the fluidity of the phospholipid monolayer and thereby inhibiting the lateral propagation of the lipid peroxy radicals.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Oxidative modification of the low density lipoproteins (LDL)¹ is believed to play a critical role in the infiltration of monocytes and in the formation of lipid-laden foam cells found in the atherosclerotic lesion (1-3). Several studies showed that the small dense LDL fractions are more susceptible to in vitro oxidation than the large buoyant LDL fractions (4-6). Patients with a higher percentage of dense LDL in their plasma (Pattern B) are known to be at an increased risk for atherosclerosis (7, 8). The possible mechanisms for the increased oxidative susceptibility of small dense LDL fractions is not well understood. Although a higher concentration of polyunsaturated fatty acids was reported in the dense LDL, this parameter did not correlate with the difference in the oxidation rates between dense and buoyant LDLs (4). Tribble et al. (6) found significantly lower levels of ubiquinol-10 and -tocopherol in the dense LDL, compared with the buoyant LDL. However, the variation in these antioxidants could account for less than 35% of the variation in oxidation rates (6). Furthermore, the antioxidant status of LDL is in general not correlated with the lag phase of oxidation (9). On the basis of the oxidizability of selectively localized probes in LDL particles, Tribble et al. (10) proposed that the greater susceptibility of small dense LDL is due to differences in the surface lipid composition rather than the core lipid or protein composition. In an earlier study (5), they found an inverse correlation between the free cholesterol (FC) content of the LDL particle and its rate of oxidation and suggested that FC might protect against peroxidation of the surface lipids by altering the fluidity of the monolayer. However, the effect of FC on the oxidative susceptibility of LDL or defined liposomes was not experimentally tested.

In our previous studies on the lipid composition of LDL subfractions, we found a positive correlation between the size of the LDL particle and its sphingomyelin (SPH)/phosphatidylcholine (PC) ratio (11). Because SPH is generally colocalized with FC in various cell membranes and lipoproteins (12, 13), we investigated the possibility that the observed correlation between the FC concentration and resistance to oxidation is actually related to the SPH content rather than the FC content of the lipoprotein. The basis for this hypothesis is that SPH is known to decrease the fluidity of the monolayer and that the fluidity of the monolayer plays a significant role in the rate of lipid oxidation (14, 15). It has also been shown that SPH inhibits the oxidation of lipoprotein FC by cholesterol oxidase by increasing the packing density of the surface monolayer (16). The results presented here show that altering the SPH content of the isolated LDL significantly affects its oxidizability in the presence of Cu²⁺ or lipoxygenase. Furthermore, we show that the incorporation of SPH into synthetic liposomes containing unsaturated PC greatly inhibits the oxidation of PC and that this effect can be reversed by treatment with sphingomyelinase (SPHase). We propose that SPH inhibits the rate of peroxidation of PC in the surface of the lipoproteins by retarding the lateral propagation of the lipid-free radicals because of its membrane-rigidifying properties. This inhibition may be physiologically significant because of the high concentrations of SPH found in certain lipoprotein fractions and in the interstitial fluid.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Egg SPH, 16:0-18:2 PC, 16:0-16:0 PC, SLO (Type IV, 440,000 units/mg), and Staphylococcus aureus SPHase (297 units/mg) were all obtained from Sigma. DPH-PC was purchased from Molecular Probes Inc. (Eugene, OR). Apoprotein A-I was prepared from delipidated HDL₃ as described before (17).

LDL Subfractions-- Fresh plasma, prepared from non-fasting normal volunteers, was purchased from a commercial source (United Blood Center, Chicago). LDL subfractions were isolated by sequential centrifugation of separate aliquots of plasma in the presence of 1 mM EDTA, as described by Tribble et al. (6). The buoyant LDL was obtained between the densities of 1.026 and 1.032 g/ml, whereas the dense LDL was obtained between the densities of 1.040 and 1.054 g/ml. All densities were adjusted using solid KBr. Whole LDL was prepared by centrifugation between the densities of 1.019 and 1.060 g/ml. The lipoproteins were collected by tube-slicing and were dialyzed against Tris-NaCl buffer, pH 7.4, containing 1 mM EDTA and stored in the dark at 4 °C. Before the lipoproteins were used for oxidation studies, they were dialyzed against the same buffer without EDTA.

Liposomes-- Proteoliposomes containing 16:0-18:2 PC and apoA-I were prepared by the cholate-dialysis procedure (18), with the apoprotein/PC ratio at 1:300. Where indicated, egg SPH, 16:0-16:0 PC, or FC was also included. The concentration of 16:0-18:2 PC was kept at 0.164 µmol/ml in all samples. All liposome preparations were dialyzed against Tris-NaCl buffer, pH 7.4, and stored in the dark at 4 °C until used. All samples were used for oxidation within 1 week of preparation. Where indicated, the proteoliposomes were treated with S. aureus SPHase in the presence of 0.8 mM MnCl₂, and again dialyzed, to remove the metal ions.

Oxidation-- LDL subfractions were oxidized in the presence of 5 µM CuSO₄ at 37 °C in thermostated cuvettes in a spectrophotometer (Shimadzu), and the formation of conjugated dienes (CD) was measured by continuous monitoring of the absorbance at 234 nm. The data were imported and plotted in a spreadsheet program (Microsoft Excel), and the slopes were calculated in the linear region. The lag times were determined manually from the intercepts of the lines of lag phase and propagative phase, as described (4). The oxidation in the presence of SLO was performed using 500 units/ml of SLO at 37 °C, and the CD formation was monitored as described above.

Oxidation of DPH-PC incorporated into LDL or liposomes was performed as described by Hofer et al. (19). DPH-PC was added as ethanol solution to LDL or the liposomes to give a molar ratio of DPH-PC/substrate PC of 1:100. Samples were incubated in the dark at 37 °C for 12 h in a water bath for the equilibration of the label. The oxidation of DPH-PC was initiated by the addition of 10 µM CuSO₄, and the decay of fluorescence was monitored in a spectrofluorometer at 37 °C (excitation wavelength, 360 nm, and emission wavelength, 430 nm). Labeled LDL was depleted of SPH by treatment with S. aureus SPHase (0.8 unit of SPHase/3 mg of LDL protein) for 4 h at 37 °C, in the presence of 0.8 mM MnCl₂. The samples were dialyzed against Tris-NaCl buffer, pH 7.4 (without EDTA), to remove the metal ions. Control samples without SPHase were incubated and dialyzed under the same conditions.

Analytical Procedures-- Lipid phosphorus was determined by the modified Bartlett procedure (20), after the separation of the phospholipids by silica gel TLC (11). Protein was estimated by the modified Lowry procedure (21).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Oxidation of Buoyant and Dense LDL Subfractions: Effect of SPH Depletion-- Fig. 1 shows the formation of conjugated dienes (CD) in the buoyant and dense LDL subfractions in the presence of 5 µM Cu²⁺. The lag time for the dense LDL was considerably lower (188 min) compared with the buoyant LDL (216 min), in agreement with previous studies. Treatment with SPHase significantly shortened the lag time in both buoyant and dense LDL fractions, in a dose-dependent manner, showing that the hydrolysis of SPH results in increased susceptibility of LDL to oxidation. However, the differences between the two LDL fractions remained even after the SPHase treatment, indicating that other factors such as the antioxidant content also contribute to the differences in susceptibility. The slope for the CD formation increased by 9% in buoyant LDL and by 38% in the dense LDL after treatment with 0.26 unit of SPHase. SPHase alone did not have any effect on the oxidation of LDL in the absence of added Cu²⁺, showing that the effect of SPHase is not due to the presence of divalent metal ions in the enzyme preparation. Analysis of phospholipids after treatment with 0.26 unit of the enzyme showed that the SPH/PC ratio decreased from 0.395 to 0.108 in buoyant LDL and from 0.265 to 0.081 in dense LDL. Although previous studies (22, 23) reported that LDL gets aggregated after SPHase treatment, we did not find any aggregation under these conditions (low concentrations of LDL and metal ions).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effect of SPH depletion on the oxidizability of LDL subfractions. The buoyant and dense LDL subfractions (200 µg of protein/ml) were dialyzed against EDTA-free buffer and incubated with 5 µM CuSO4 at 37 °C in a spectrophotometer cuvette, and the absorbance at 234 nm was recorded at 10-min intervals. Where indicated, S. aureus SPHase (0.13 or 0.26 unit/ml) was added to each cuvette before the start of the oxidation. Control tubes containing only LDL and SPHase (no CuSO4) were included to determine the effect of any metal ions in the enzyme preparation. The slopes and lag times were calculated for each curve as described in the text. There was no aggregation of LDL under these conditions.

Oxidation of LDL with Soybean Lipoxygenase (SLO)-- Because the lipoxygenase-mediated oxidation is believed to play a major role in the formation of modified LDL in vivo (24), we tested the effect of SPH depletion of LDL on the oxidation by SLO. Whole LDL fraction (d 1.019-1.060 g/ml, 0.3 mg of cholesterol) was first passed through a PD-10 column (Amersham Pharmacia Biotech) to remove EDTA and then treated with 0.48 unit of SPHase for 2 h at 37 °C. The sample was then incubated with 500 units/ml of SLO in a spectrophotometer cuvette at 37 °C, and the increase in A₂₃₄ was recorded. During the oxidation with SLO, 5 µM EDTA was included in all cuvettes to inhibit any oxidation due to metal ions in the SPHase preparation. As shown in Fig. 2, SLO caused an increase in A₂₃₄ in intact LDL, with a shorter lag time than observed with Cu²⁺-mediated oxidation. When LDL was pretreated with SPHase, the rate of oxidation was stimulated significantly, the slope being 45% higher than with intact LDL. The lag period was shortened by about 30 min after treatment with SPHase. Samples containing only SPHase exhibited a modest increase in A₂₃₄, but this increase could not account for the effect of SPHase in the SLO-mediated oxidation. Analysis of phospholipid composition showed that the SPHase treatment reduced the SPH/PC ratio from 0.38 to 0.21, showing that the enzyme was active, although no metal ions were added, presumably because of the presence of some metal ions in the enzyme preparation. There was no aggregation of LDL under these conditions. These results thus show that SPH inhibits the oxidation of LDL under physiological conditions.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Oxidation of LDL by SLO: effect of SPH depletion. Whole LDL preparation (d 1.019-1.60 g/ml) (300 µg of protein) was first passed through a PD-10 column (Amersham Pharmacia Biotech) to remove EDTA and then treated with 0.48 unit of bacterial SPHase for 2 h at 37 °C. The sample (and the untreated control) was then incubated with 500 units/ml of SLO at 37 °C in a spectrophotometer cuvette, and the absorbance was recorded at 234 nm. All cuvettes contained 5 µM EDTA to minimize the oxidation due to contaminating metal ions in the SPHase preparation.

Oxidation of DPH-PC Incorporated into LDL-- The fluorescence decay of DPH-PC equilibrated with lipoprotein PC has been used as a sensitive marker of the oxidative susceptibility of the lipoproteins (19). This fluorescent phospholipid is primarily incorporated into the surface lipids of LDL, and therefore its degradation may be used as a specific reporter of surface lipid oxidation. DPH-PC was incorporated into normal LDL at a concentration of 1 nmol/100 nmol of LDL phospholipid, and its fluorescence was measured at 37 °C in the presence of 5 µM Cu²⁺. As shown in Fig. 3, there was a linear decrease in fluorescence after an initial lag period in the control LDL. When LDL was first treated with SPHase, and then oxidized, the lag period was significantly decreased (860 s versus 1830 s for control), and the slope was increased by 20%. These results further support the hypothesis that SPH has a protective effect on the oxidation of unsaturated phospholipids in LDL.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Oxidation of DPH-PC incorporated into LDL: effect of SPH depletion. DPH-PC was added as ethanol solution to whole LDL to give a molar ratio of DPH-PC to LDL-PC of 1:100 and incubated at 37 °C for 12 h in the dark to equilibrate the label with endogenous PC. An aliquot of the labeled LDL was treated with SPHase (0.5 units/ml for 4 h) in the presence of 0.8 mM MnCl2, and the sample was dialyzed against Tris-NaCl buffer, pH 7.4, to remove the metal ions. The control and the SPHase-treated LDLs were then oxidized in the presence of 5 µM Cu2+, and the fluorescence was monitored in a spectrofluorometer (SLM-Aminco) with the excitation and emission wavelengths set at 360 nm and 430 nm, respectively.

Effect of SPH on the Oxidation of PC in the Liposomes-- To directly determine whether SPH can protect the unsaturated PC against peroxidation, we incorporated varying amounts of SPH into proteoliposomes containing apoA-I and 16:0-18:2 PC (at a molar ratio of 1:300) (18) and tested the rate of lipid oxidation in the presence of 10 µM Cu²⁺. As shown in Fig. 4, there was a gradual increase in lag time and decrease in the slope, with increasing ratio of SPH/PC in the proteoliposomes up to a ratio of 0.75. However, at the highest SPH/PC ratio tested (1.0), the oxidation was higher than at 0.75, probably due to a phase separation, which is known to occur as the SPH/PC ratio is increased or when the temperature is decreased (12). The concentration of PC was kept the same in all the samples, and therefore the decrease in slope is not due to a decrease in the oxidizable lipid. These results show that SPH does protect unsaturated PC against oxidation in a defined system.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Effect of SPH on the oxidation of 16:0-18:2 PC incorporated into proteoliposomes. Proteoliposomes containing apoA-I and 16:0-18:2 PC at the molar ratio of 1:300 and varying amounts of egg SPH were prepared by the cholate dialysis (18). The concentration of 16:0-18:2 PC was the same (0.164 mM) in all the samples. Oxidation of the samples was carried out at 37 °C in the presence of 10 µM CuSO4, and the absorbance at 234 nm was recorded. The calculated slopes and lag times are shown in the lower panel.

Inclusion of SPH in the liposomes similarly protected PC against oxidation by SLO in a dose-dependent manner. Furthermore, the fluorescence decay of DPH-PC incorporated into the liposomes in the presence of either Cu²⁺ or the free radical generator, 2,2'-azo-bis-(2-amidinopropane)dihydrochloride, was protected by SPH in the liposomes (results not shown).

Reversal of SPH Effect in Proteoliposomes by the Action of SPHase-- To exclude the possibility that the protection by SPH is due to its nonspecific effects on the structure of the liposomes, we treated the proteoliposomes containing SPH (with a SPH/PC ratio of 0.75) with SPHase and studied the effect on the Cu²⁺-mediated oxidation. As shown in Fig. 5, the hydrolysis of SPH significantly increased the oxidation of PC, although the rate of oxidation did not reach that of pure PC alone. Treatment of proteoliposomes containing PC alone with SPHase did not have any effects on oxidation. These results show that the presence of intact SPH is necessary for the maximal protection. Although the product of SPHase reaction (ceramide) remains with the liposomes, it does not appear to protect PC against oxidation.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Effect of SPH depletion in proteoliposomes. The proteoliposomes containing either PC alone (control) or SPH/PC ratio of 0.75 were treated with SPHase (0.5 unit of enzyme/ml, in the presence of 0.8 mM MnCl2) and then dialyzed to remove the metal ions. The untreated proteoliposomes were also dialyzed under the same conditions. All samples were oxidized in the presence of 20 µM Cu2+, and the CD formation was monitored at 234 nm.

Effect of Extrinsic versus Intrinsic SPH-- To determine the possible mechanism of inhibition of PC oxidation, we tested the effect of extrinsic and intrinsic SPH on the oxidation of PC in defined liposomes. If the protection by SPH is dependent upon its presence in close proximity to PC, the extrinsic SPH should not be protective. On the other hand, if SPH acts on the oxidizing agent, it should protect equally well whether it is in the same liposome or not. As shown in Fig. 6, there was a stimulation of PC oxidation, rather than an inhibition, by the addition of SPH-AI liposomes to the PC-AI liposomes (extrinsic SPH). However, as shown above, the presence of SPH in the same proteoliposome as the PC (intrinsic SPH) inhibited the oxidation. These results show that the effect of SPH is related to its physical interaction with the PC in the bilayer rather than to its other possible effects such as the chelation of metal ions or the scavenging of the peroxy radicals. This mechanism is further supported by the effect of dipalmitoyl-PC on the oxidation of 16:0-18:2 PC. When 25 mol % dipalmitoyl-PC was incorporated into the proteoliposomes containing 16:0-18:2 PC, the rate of oxidation (slope) of the latter by Cu²⁺ was decreased by 38%, as compared with a 24% decrease in the presence of 25 mol % SPH (Fig. 7). The lag period in the presence of dipalmitoyl-PC was 100 min, compared with 65 min in the presence of SPH and 50 min with 16:0-18:2 PC alone. The presence of 25 mol % FC, however, had no significant effect on either the lag period or slope.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   Effect of intrinsic versus extrinsic SPH. Proteoliposomes containing 16:0-18:2 PC and apoA-I or egg SPH and apoA-I were prepared by cholate dialysis at the molar ratio of phospholipid/apoA-I at 300:1. Proteoliposomes containing both PC and SPH in equal concentrations were also prepared. In the extrinsic assay, equal volumes of the PC liposomes and SPH liposomes were present in the reaction mixture, whereas in the intrinsic assay the PC-SPH proteoliposome were present (1:1 molar ratio). All samples were oxidized in the presence of 20 µM Cu2+ at 37 °C.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Effect of disaturated PC and FC on the oxidation of unsaturated PC. Proteoliposomes containing PC alone, PC + 25 mol % egg SPH, PC + 25 mol % FC, and PC + 25 mol % dipalmitoyl-PC (Di16:0 PC) were prepared by cholate dialysis and oxidized in the presence of 20 µM Cu2+.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

SPH is the most abundant phospholipid in the lipoproteins next to PC, comprising up to 30% of the total phospholipids in certain lipoproteins (11). Its concentration is further increased in apoprotein E deficiency (25) and in aging and atherosclerosis (12). Although its importance as a structural component of lipoproteins is well recognized, its possible involvement in the metabolism of the lipoproteins has not received much attention because it is relatively inert in the plasma and there is no plasma enzyme for which it is a substrate. We have previously demonstrated that SPH is a strong inhibitor of cholesterol esterification by lecithin:cholesterol acyltransferase (26) and suggested that, because of its structural similarities to PC, it may act as a competitive inhibitor of lecithin:cholesterol acyltransferase as well as various other phospholipases. Recent studies by Arimoto et al. (27) and Lobo and Wilton (28) show that SPH also inhibits the activity of lipoprotein lipase, suggesting that it may have broader physiological function as an inhibitor of PC hydrolysis, thus helping maintain the structural integrity of lipoprotein particle. A similar role has been proposed for membrane SPH in the protection of membrane PC from the action of cellular phospholipase A (29). The results presented here demonstrate another possible mechanism by which SPH preserves the integrity of lipoprotein structure, namely the protection of surface PC against oxidation, which is also known to lead to extensive degradation of PC (1).

The importance of oxidative modification of lipoproteins in the development of atherosclerosis has been studied extensively in recent years (3). Several studies showed that oxidized LDL not only exerts chemotactic and cytotoxic effects but is also taken up by macrophages rapidly through an unregulated pathway, leading to the formation of foam cells (1-3). Furthermore, oxidation of HDL lipids renders this lipoprotein less efficient in the promotion of cholesterol efflux (30). Although some studies reported that the antioxidant protection of LDL, as indicated by the increase in the lag time of CD formation, is not directly correlated with the atherogenic risk (31), increasing the resistance of LDL to oxidation is generally accepted to be beneficial. While previous studies showed that large buoyant LDL are more resistant to oxidation than the small dense LDL, the possible mechanism for this is not established. Based on the correlation between the FC content and the oxidizability, Tribble et al. (5) suggested that the FC content of LDL particles may be an important determinant of oxidation. However the effect of FC on lipoprotein oxidation has not been experimentally demonstrated. Our previous studies showed that the SPH/PC ratios are also positively correlated with the size of LDL particles (11) and therefore the susceptibility to oxidation is also inversely correlated with the SPH content. The results presented here show that the high concentration SPH may contribute in part to the resistance of the large buoyant LDL particles to oxidation.

The mechanism by which SPH inhibits PC oxidation is yet to be established. An examination of the structure of SPH reveals the presence of no easily exchangeable hydrogen atoms, and therefore, it cannot act as a chain-breaking reagent. It does not chelate metal ions and thus cannot prevent metal ion-induced amplification of the hydroperoxy radicals. However one property of SPH that may contribute to its antioxidant function is its ability to decrease the fluidity of a membrane or monolayer because of its saturated hydrocarbon chains (12). Studies by Wiseman and colleagues (14) showed that tamoxifen and other steroid molecules have "membrane-rigidifying" effects that correlate strongly with their antioxidant properties. SPH may belong to this class of antioxidants, which act by retarding the propagation of oxidation reaction, rather than by preventing the initial formation of lipid-free radicals. The fact that only the intrinsic SPH protects the PC while the externally added SPH cannot supports this hypothesis. Furthermore, treatment of the lipoproteins or liposomes with SPHase abolished the effect of SPH, suggesting the requirement of an intact SPH molecule. Ceramide, the product of SPHase action, is more hydrophobic than SPH and may be localized deeper in the PC monolayer than SPH and therefore may not block the contact of adjacent PC molecules. Thus, the propagation of the free radical reaction along the PC monolayer may be impeded physically by the presence of SPH but not by ceramide. Similar inhibition was found when dipalmitoyl-PC was incorporated into 16:0-18:2 PC liposomes, showing that the presence of a long chain saturated phospholipid molecule prevents the oxidation of unsaturated PC in the same monolayer. However, the presence of FC at a similar concentration did not have any effect. Unlike some of the well known antioxidants that scavenge the free radicals, the amount of SPH needed to block the oxidation by the physical modification of the monolayer would be expected to be relatively high. The physiological concentration of SPH and hence the SPH/PC ratio in certain lipoproteins is indeed quite high (0.406 in large LDL) (11), and at such SPH/PC ratios the PC oxidation is significantly inhibited in the defined systems (Fig. 4). In contrast to FC, which can be oxidized (32, 33) and thus can contribute to the lipoprotein modification, plasma SPH is very resistant to oxidation under physiological conditions because its acyl groups are predominantly saturated or monounsaturated (34). It may thus insulate the unsaturated phospholipids of the lipoproteins against oxidizing agents in the plasma as well as in the subendothelial space. The oxidation of cholesterol by cholesterol oxidase has also been shown to be inhibited by SPH in vitro (16). It is of interest to note that the lipoproteins in the interstitial fluid have high SPH/PC ratios compared with plasma lipoproteins (35), and this may be physiologically important in the protection of unsaturated phospholipids against oxidation in this compartment. Studies by Tabas and colleagues (22) show that the arterial SPHase may be involved in the aggregation of LDL particles, leading to an increased uptake by macrophages. The results presented here suggest another possible mechanism by which arterial SPHase may influence the atherogenicity of LDL, because the depletion of LDL SPH by this enzyme would increase the susceptibility of LDL to the oxidation in the subendothelial space and convert it into a more atherogenic particle.

	FOOTNOTES

^* This research was supported by National Institutes of Health Grant HL-52597 and by funds from the Max Baer Heart Fund of the Fraternal Order of Eagles.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Section of Endocrinology and Metabolism, Rush Medical College, 1653 W. Congress Parkway, Chicago, IL 60612. Tel.: 312-455-2439; Fax: 312-455-9814; E-mail: psubbaia@rush.edu.

	ABBREVIATIONS

The abbreviations used are: LDL, low density lipoproteins; HDL, high density lipoproteins; CD, conjugated dienes; DPH, diphenylhexatriene; FC, free cholesterol; PC, phosphatidylcholine; SLO, soybean lipoxygenase; SPH, sphingomyelin; SPHase, sphingomyelinase.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES