Donate Help Contact The AHA Sign In Home
American Heart Association
Arteriosclerosis, Thrombosis, and Vascular Biology
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Arteriosclerosis, Thrombosis, and Vascular Biology. 1996;16:784-793

This Article
Right arrow Abstract Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sevanian, A.
Right arrow Articles by Bittolo-Bon, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sevanian, A.
Right arrow Articles by Bittolo-Bon, G.
Right arrowPubmed/NCBI databases
*Compound via MeSH
*Substance via MeSH
Hazardous Substances DB
*ASCORBIC ACID
*SODIUM ASCORBATE
(Arteriosclerosis, Thrombosis, and Vascular Biology. 1996;16:784-793.)
© 1996 American Heart Association, Inc.

Articles

Contribution of an In Vivo Oxidized LDL to LDL Oxidation and Its Association With Dense LDL Subpopulations

Alex Sevanian; Juliana Hwang; Howard Hodis; Giuseppe Cazzolato; Pietro Avogaro1; Gabriele Bittolo-Bon

From the Department of Molecular Pharmacology and Toxicology (A.S., J.H., H.H.), School of Pharmacy, and the Atherosclerosis Research Unit (A.S., H.H.), Division of Cardiology, University of Southern California, Los Angeles, Calif, and the Centro Regionale Specializzato per l'Arteriosclerosi (G.C., G.B.-B.), Servizio di Diabetologia, Ospedale Riuniti, Venezia, Italy.

Correspondence to Alex Sevanian, University of Southern California, Department of Molecular Pharmacology and Toxicology, School of Pharmacy, 1985 Zonal Ave, PSC 612, Los Angeles, CA 90033.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Abstract Oxidative modification of LDL is thought to be a radical-mediated process involving lipid peroxides. The small dense LDL subpopulations are particularly susceptible to oxidation, and individuals with high proportions of dense LDL are at a greater risk for atherosclerosis. An oxidatively modified plasma LDL, referred to as LDL-, is found largely among the dense LDL fractions. LDL- and dense LDL particles also contain much greater amounts of lipid peroxides compared with total LDL or the more buoyant LDL fractions. The content of LDL- in dense LDL particles appears to be related to copper- or heme-induced oxidative susceptibility, which may be attributable to peroxide levels. The rate of lipid peroxidation during the antioxidant-protected phase (lag period) and the length of the antioxidant-protected phase (lag time) are correlated with the LDL- content of total LDL. Once LDL oxidation enters the propagation phase, there is no relationship to the initial LDL- content or total LDL lipid peroxide or vitamin E levels. Beyond a threshold LDL- content of {approx}2%, there is a significant increase in the oxidative susceptibility of nLDL particles (ie, purified LDL that is free of LDL-), and this susceptibility becomes more pronounced as the LDL- content increases. nLDL is resistant to copper- or heme-induced oxidation. The oxidative susceptibility is not influenced by vitamin E content in LDL but is strongly inhibited by ascorbic acid in the medium. Involvement of LDL--associated peroxides during the stimulated oxidation of LDL is suggested by the inhibition of nLDL oxidation when LDL- is treated with ebselen prior to its addition to nLDL. Populations of LDL enriched with LDL- appear to contain peroxides at levels approaching the threshold required for progressive radical propagation reactions. We postulate that elevated LDL- may constitute a pro-oxidant state that facilitates oxidative reactions in vascular components.


Key Words: lipid peroxidation • peroxides • ascorbic acid • heme


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Oxidative modification of LDL is widely regarded as a contributing factor to the development of atherosclerosis. Oxidation of LDL increases the cytotoxicity and atherogenicity of this particle.1 Similar forms of oxidized LDL are produced by cell-mediated as well as metal- and oxidant-catalyzed reactions.2 3 4 LDL oxidatively modified by these latter methods is considerably more cytotoxic than unmodified LDL1 and elicits a variety of vascular responses observed during atherogenesis. Elevated plasma levels of an oxidatively modified LDL have been observed in hypercholesterolemic animals5 and humans.6 This lipoprotein, designated as LDL-, is relatively enriched in free cholesterol, protein, triglycerides, and oxidized lipids and contains diminished levels of vitamin E.7 Analysis of human plasma samples indicates that LDL- levels vary from <1% to as high as 8% of total LDL, but there is little or no correlation between the percentage of LDL- and the content of total LDL, total cholesterol, or LDL-C plasma levels.6 However, the absolute amounts of LDL- are related to plasma LDL-C levels, indicating that LDL- concentrations can be much greater in hypercholesterolemic subjects.

Elevated LDL-C is regarded as a risk factor for atherosclerosis, and high levels of LDL-C appear to contribute to vascular injury.8 Enhanced production of oxidizing agents by vascular tissues or resident inflammatory cells may be responsible for the formation of oxidatively modified LDL found in vivo, eg, LDL-.1 9 Production of oxidized LDL is thought to be augmented by elevated LDL, which both stimulates oxyradical generation by vascular cells10 and serves as an enlarged target for oxidation. Although the origin of LDL- remains uncertain, species resembling this lipoprotein are produced after mild Cu2+-induced oxidation of freshly isolated LDL,6 are more susceptible to in vitro oxidation,11 and are also found in minimally modified LDL.12 13 LDLs are a heterogeneous population of particles that vary in size, density, electrical charge, and composition.14 15 16 17 One significant aspect of this variability relevant to atherosclerosis has been pointed out by the studies of Austin et al18 on the phenotypic patterns of LDL density and size distribution among individuals with and without coronary artery disease. With nondenaturing gradient gel electrophoresis, various LDL subgroups were identified on the basis of size19 20 21 and traditionally divided into three main phenotypes: pattern A, characterized by a predominance of large LDL particles; pattern B, with a predominance of small dense LDL; and an intermediate pattern.22 Individuals with lipoprotein profiles enriched in small dense LDL (pattern B) were found to be predisposed to coronary artery disease.16 18

The reduced lipid-to-protein ratio and the elevated levels of lipid peroxidation products in LDL- compared with normal LDL6 are apparent similarities between the small dense LDL fraction and LDL--enriched LDL. Small dense LDLs appear to be more susceptible to oxidation,23 which suggests that small dense LDLs have diminished antioxidant potential and may contain higher levels of lipid peroxidation products that would facilitate further lipoprotein oxidation. Investigation of these properties of LDL- and dense LDL fractions was conducted through a series of analyses of plasma samples from volunteers visiting the Atherosclerosis Research Unit at the University of Southern California. Measurements of plasma LDL and LDL- were performed in conjunction with quantitative analysis of LDL- distribution among LDL density subclasses. Concurrently, determinations of oxidative susceptibility of LDL- versus normal LDL or LDL containing known amounts of LDL- were made by using LDL preparations from these subjects.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Chemicals and Reagents
The following reagents were obtained from Sigma Chemical Co: EGTA, EDTA, hematin, ascorbic acid, and phosphate buffer. Buffers and media were obtained from GIBCO. DL-{alpha}-tocopherol (vitamin E) was generously supplied by the Henkel Corp. All organic solvents were HPLC grade and purchased from J.T. Baker Chemical Co, as was CuSO4. Ebselen (PZ51) was obtained from Ciba-Geigy. Linoleic acid hydroperoxide (13-hydroperoxy-9,11-octadecadienoic acid) was prepared from pure linoleic acid obtained from NuCheck Prep.

Isolation of LDL From Plasma
Venous blood was obtained from fasting adult human volunteers with total plasma cholesterol levels ranging from 160 to 210 mg/dL. Blood was collected into 10-mL evacuated tubes containing EDTA, and plasma was immediately separated by centrifugation at 1500g for 10 minutes at 4°C. LDL ({delta}=1.019 to 1.063 g/mL) was separated from freshly drawn plasma by preparative ultracentrifugation with a Beckman L8-55 ultracentrifuge equipped with an SW-41 rotor.5 In addition, LDL density subfractions were prepared by NaBr density gradient ultracentrifugation by layering 3.5 mL total LDL onto a NaBr solution consisting of 4.5 mL {delta}=1.055, 3.5 mL {delta}=1.040, 2.5 mL {delta}=1.024, and 2 mL {delta}=1.019 g/mL. Samples were run with controls layered with the LDL buffer only from which densities were determined gravimetrically after ultracentrifugation. After centrifugation, the following density fractions were collected: >1.050, 1.050 to 1.040, 1.040 to 1.030, 1.030 to 1.020, and <1.020 g/mL. Occasionally, the {delta}>1.050 fraction was collected as two subfractions consisting of {delta}=1.065 and {delta}=1.065 to 1.055 g/mL. Total LDL or LDL subfractions were dialyzed against argon-sparged 0.05 mol/L Tris-HCl buffer, pH 7.2, containing 10 µmol/L EDTA at 4°C. Cholesterol levels were measured enzymatically with a VP Super System instrument (Abbott) according to Lipid Research Clinic methodology.24 The isolated LDL was kept in argon-sparged EDTA-buffer at 4°C for no more than 24 hours before further processing or was used as described below.

Preparation of LDL-
Separation of LDL- from unmodified LDL was accomplished with anion-exchange HPLC (Perkin-Elmer Series 4 HPLC).5 The purity of the isolated LDL was checked by means of rocket gel electrophoresis using agarose gels prepared with 10% polyclonal antibodies to apoB-100, apoA-1, and lipoprotein(a)25 with commercially available antibodies for these apoproteins (Apo-Tek, Organon Teknika/Biotechnology Research Institute). The isolated LDL or LDL density subfractions were then injected into the HPLC at an adjusted concentration of 0.5 mg cholesterol/mL buffer. The eluent was monitored at 280 nm, and peaks corresponding to nLDL or LDL- were collected in 1-mL aliquots and used immediately. The amount of LDL protein was determined for each peak with the method of Lowry et al26 and was employed for peak area calibration, from which the amounts of nLDL and LDL- were routinely computed. Fractions were collected into tubes containing 50 µmol/L EDTA in 0.01 mol/L Tris-HCl buffer, pH 7.2, and those fractions containing LDL- were pooled and concentrated, and all salts were removed by centrifugation with Centricon 10,000 molecular-weight microconcentrators. This procedure causes minimal amounts of aggregation, and the product is indistinguishable from that produced by conventional dialysis. Samples were then diluted in PBS, and the protein content was determined. LDL samples were then suspended in PBS containing 10 µmol/L EDTA for determinations of oxidative susceptibility. In some cases, samples were extracted for determinations of total lipid conjugated diene levels as detailed below.

In Vitro Oxidation of LDL
Fresh preparations of LDL samples were mixed and incubated with 10 µmol/L CuSO4 (net concentration) for intervals of up to 5 hours. This provided a rapid and reproducible means for determining oxidation kinetics. Alternately, samples were oxidized by adding a hematin solution27 (final concentration, 2 µmol/L hematin). This was particularly useful for measurement of oxidation kinetics under conditions not affected by EDTA. The reactions were monitored continuously at 234 nm with a Beckman DU650 spectrophotometer and Beckman proprietary software in temperature-controlled 0.5-mL quartz cuvettes (diameter, 1.0 cm) containing 500 µg/mL LDL protein. The protein content was determined via the absorbance at 280 nm and adjusted to an optical density (OD) of 0.50 (where 100 µg protein corresponds to an optical density of 0.10) for all samples unless otherwise specified. Since the rate and extent of LDL oxidation varies among samples from different donors, the degree of oxidation was determined by continuously monitoring the increase in conjugated dienes (OD=234 nm) over the incubation interval as described by Esterbauer et al.28 This enabled measurement of the oxidative lag phase, the rate of peroxidation during the lag phase (considered as the antioxidant-protected phase), and the oxidation rate during the subsequent propagation phase (considered as the antioxidant-unprotected phase). Total peroxide levels in the LDL were estimated by using a molar extinction coefficient of 2.54x104 for lipid conjugated dienes measured at the end of the incubation period and checked against the amounts measured in lipid extracts as described below.

Additional oxidation studies involved mixing LDL- with nLDL in proportions ranging from 1% to 5% LDL- (wt/wt protein). These mixtures were then subjected to Cu2+- or hematin-induced oxidation, and the rates of oxidation were determined as described above. Inhibition of LDL oxidation was tested by subjecting samples to Cu2+-induced oxidation in the presence of 100 µmol/L ascorbic acid (freshly prepared in water). In some experiments, LDL- was preincubated with ebselen and glutathione (50 µmol/L and 3 mmol/L, respectively), a concentration of ebselen that is in stoichiometric excess to the peroxide content of the lipoproteins, for 15 minutes before addition of the LDL- to nLDL and incubation with CuSO4. Under these conditions ebselen reduces all lipoprotein peroxides.29 At the end of the incubation period and after spectrophotometric analysis of oxidation kinetics, the samples were collected and dialyzed with Centricon 10,000 molecular-weight microconcentrators. All preparations were then adjusted to a fixed protein concentration with the Biuret protein assay (Bio-Rad) for analysis of LDL- content by HPLC as described above.

Measurement of LDL Lipid Peroxidation
The progress of LDL oxidation was determined either directly, by monitoring formation of conjugated dienes in LDL suspensions via the absorbance change at 234 nm as described above, or by measurement of conjugated dienes at specified intervals on lipid extracts of LDL with second-derivative UV spectroscopy.30 For the latter measurements, samples of LDL containing 500 µg protein/mL were extracted with 6 mL chloroform-methanol (2:1) after various periods of incubation with the oxidizing system. The organic phase was collected and saved, the aqueous phase was reextracted with another 3 mL chloroform-methanol (2:1), and the organic phases were pooled. After evaporation of the solvent under a stream of nitrogen at room temperature, the lipid residue was redissolved in absolute ethanol, and the absorbance was monitored over the frequency range of 220 to 300 nm against an ethanol blank. The extent of peroxidation was estimated from the sum of the absorbance minima at 242 and 233 nm corresponding to the cis/trans and trans/trans diene conjugate isomers. Linoleic acid hydroperoxide31 was used to develop a calibration curve. All scans were taken with a Beckman DU650 spectrophotometer. The content of lipid peroxides was estimated from the total cis/trans and trans/trans conjugated dienes with the use of a molar extinction coefficient of 2.54x104. Alternately, the lipid peroxide content of LDL was measured by means of the oxidation of leucomethylene blue reagent as described by Auerbach et al.32 The microtiter assay was adapted so that 150 µL of the leucomethylene blue reagent was mixed with 40 µL LDL (at 2 mg/mL) and allowed to react for 1 hour at room temperature before measuring the absorbance at 650 nm. The amount of peroxide was determined from a calibration curve with authentic linoleic acid hydroperoxide as a standard.

Measurement of LDL Vitamin E Content
The content of vitamin E in the pooled nLDL and LDL- fractions and the LDL density subfractions was determined by extraction of samples according to the method of Bieri et al,33 with minor modifications. Aliquots of plasma (500 µL) were mixed for 20 seconds with 50 µL internal standard ({alpha}-tocopherol acetate) and 250 µL 0.025% butylated hydroxytoluene in ethanol. The lipid phase was extracted twice with successive additions of 500 µL hexane containing 0.025% butylated hydroxytoluene with agitation for 1 minute. After centrifugation, the upper phases were collected, pooled, and evaporated to dryness under nitrogen. The residue was dissolved in 200 µL ethanol, and 20-µL aliquots were injected into a Perkin-Elmer Series 4 chromatograph equipped with a 25x0.4-cm ODS-5S column (Bio-Rad Instruments). The column was eluted with acetonitrile/tetrahydrofuran/H2O (85.5:9:5.5, vol/vol/vol) at a flow rate of 0.9 mL/min for the first 15 minutes. The solvent composition was then changed to acetonitrile/tetrahydrofuran/H2O (80:15:5, vol/vol/vol) over 2 minutes, held at this composition for 10 minutes, and then returned to the original solvent composition for the final 2 minutes of the run. The eluent was monitored fluorometrically by using excitation and emission wavelengths of 290 and 340 nm, respectively. The areas of the peaks corresponding to {alpha}- and {gamma}-tocopherol were integrated with Axxichrom 747 chromatographic software. The vitamin E content, including {alpha}- and {gamma}-tocopherol, was determined by using established calibration curves by means of an internal standard method.

Statistical Analyses
The data are expressed as mean±SE. The results shown derive from four to six independent experiments, each measurement performed in duplicate. Data for each treatment group were subjected, as indicated, to ANOVA or Student's t test. Measurement of the trend across the groups according to the LDL- content was done by repeated measures ANOVA. Probability values <.06 are regarded as being significant.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
nLDL and LDL- Content in LDL Density Subfractions
Fig 1Down shows a typical series of chromatograms prepared from six LDL density subfractions. Although the amounts of LDL-C differed in these samples, the content of LDL- in panels C through F (corresponding to mean {delta}=1.040, 1.035, 1.025, and <1.020 g/mL, respectively) was negligible, whereas most of the LDL- was evident in density fractions shown in panels A and B (mean {delta}=1.055 and 1.045 g/mL). Small amounts of LDL- are seen in panels C and D, eluting as a peak at {approx}24 to 26 minutes, and a peak at {approx}38 to 39 minutes is also present in all samples. The identity of the latter peak has not been established. Fig 2Down presents results from six separate analyses of plasma LDL, each from different individuals, where the content of LDL- is shown as the percentage of total LDL protein in five distinct density fractions. Most of the LDL- is associated with particles having densities >=1.045 g/mL, with substantially less in the lower-density fractions. The protein-to–LDL-C ratio for the density fractions shown at mean {delta}=1.045 g/mL or less was similar and calculated to be approximately 1.8:1 (mg/mg). On the other hand, the protein-to-cholesterol ratio for the fraction with mean {delta}>1.050 mg/L was approximately 8:1. Total LDL- recovered from these subfractions was >90% of that isolated from total unfractionated LDL, based on protein or cholesterol determinations. The content of lipid peroxidation products, measured as conjugated dienes and indicated as ROOH, are also shown in Fig 2Down. Lipid peroxides were highest in the dense LDL fractions and lower in the buoyant fractions. Indeed, the content of total peroxides corresponded to the LDL- content in each subfraction. The levels of vitamin E were also found to be lowest in the most dense fractions (the concentration of vitamin E in each subfraction is indicated along the abscissa).



View larger version (34K):
[in this window]
[in a new window]
  		Figure 1. HPLC chromatograms were prepared from six LDL density subfractions ({delta}, A=1.055; B=1.045; C=1.040; D=1.030; E=1.025; and F<1.020 g/mL) that were isolated from a subject having a total of 3.8% LDL- and total LDL-C of 138 mg/dL. The LDL fractions were isolated and analyzed as described in "Methods." Whenever possible, equal amounts of LDL-C were injected for each HPLC analysis. Retention times for the nLDL and LDL- peaks are 10 to 12 minutes and 24 to 25 minutes, respectively. The absorbance shown on the y axis is represented in mV, using 1 Volt full scale, while monitoring the eluent at 280 nm.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 2. The contents of LDL- and lipid peroxides (ROOH), measured by the leucomethylene blue reaction, are shown for five LDL density fractions as determined from six subjects. Vitamin E content represents the sum of {alpha}- and {gamma}-tocopherols in LDL-C (vit. E/Cholest.). Results are mean±SE for six separate determinations. P<.05 for {delta}>1.050 and {delta}=<1.035, between {delta}=1.045 and {delta}=<1.035 for LDL- content, between {delta}>1.050 and {delta}<1.035 for ROOH content, and for vitamin E levels, between {delta}>1.050 and {delta}<1.035 and {delta}=1.045 and {delta}<1.020.

Table 1Down presents the oxidative susceptibility of the LDL subfractions represented by the oxidation lag times. In this case, Cu2+-induced oxidation was measured by using samples containing the same amount of LDL-C. Oxidation lag times were longest for the density subfractions {delta}<1.020 and {delta}>1.050 g/mL; shorter lag times were measured in densities {delta}=1.025 to 1.030 and 1.030 to 1.040 g/mL, and the shortest lag time was found in the subfraction {delta}=1.040 to 1.050 g/mL. Fig 3Down shows representative rocket gels for the LDL density subfractions described above. All fractions were positive for apoB-100, thus confirming the identity of LDL (Fig 3Down, top). In addition, none of the fractions were positive for apoA-I or apoA-II (middle and bottom, respectively), indicating that HDL lipoproteins were below minimum detectable levels and comprised an insignificant portion of the LDL studied. This was particularly important in the case of the LDL subfractions of {delta}>1.050 g/mL. As noted in "Methods," this fraction included densities as high as 1.07 g/mL, in which possible contamination by HDL may exist. There was no measurable lipoprotein(a) in any of the fractions (data not shown). These findings indicate that LDL- is largely associated with the dense LDL fraction.


View this table:
[in this window]
[in a new window]
  		Table 1. Oxidation Lag Times for LDL Density Subclasses



View larger version (110K):
[in this window]
[in a new window]
  		Figure 3. Three rocket electrophoresis gels are shown for the analysis of LDL subfractions prepared by density gradient centrifugation as described in "Methods." The gels were prepared by using 1% agarose in Tricine buffer containing 10% polyclonal antibodies and developed in a running buffer consisting of Tris 16.3 g/L, EDTA 1.0 g/L, H3BO3 4.47 g/L, and NaN3 2.5 g/L and run at 88 V/15 mA for 3 to 4 hours. A, ApoB-100; B, apoA-I; and C, apoA-II. Lanes 1 and 2, 5-µg protein aliquots from nLDL and LDL-, respectively; lanes 3-11, 5 µg LDL of the following density subfractions ({delta}=g/mL): 3, 1.065; 4, 1.065-1.055; 5, 1.055; 6, 1.045; 7, 1.040; 8, 1.035; 9, 1.030; 10, 1.025; and 11, <1.020.

Oxidative Susceptibility of LDL and LDL-
LDL- is depleted in vitamin E content and contains elevated amounts of lipid peroxidation products.7 Because this lipoprotein was also found to be largely associated with the dense LDL fraction, which is susceptible to oxidation,23 studies were conducted to further characterize the oxidative susceptibility of LDL-. After screening individuals for plasma LDL- levels, it became apparent that the LDL- content was related to the tendency of LDL to oxidize when incubated with CuSO4. A compilation of this data is presented in Fig 4Down, which compares the content of LDL- (expressed as a percentage of LDL protein) to the lag time for total LDL oxidation for each subject. The content of LDL- ranged from <1% to slightly greater than 6% of the LDL protein, with most individuals having an LDL- content ranging from 1% to 2.5%. A strong inverse correlation was found between LDL- levels and the oxidative lag time of the LDL preparation. Indeed, the data fit an exponential curve (r=.678) with an inflection separating two groups: one group had low levels of LDL- and long lag times (<1.25% LDL-), and the second had high LDL- levels and short lag times (>2% LDL-). This inflection was found within the samples ranging from {approx}1.5% to 2% LDL-. Beyond a 2% LDL-, a weak negative relationship was found between LDL- content and oxidative lag times. The relationship between short oxidative lag times and LDL- content is supported by the finding that isolated LDL- from these individuals had lag times that were much shorter than isolated nLDL. A comparison of oxidative susceptibility for nLDL versus LDL-, with either CuSO4 or hematin as catalyst, is shown in Fig 5Down. Regardless of the oxidation method used, the oxidative lag time for LDL- was less than any of the nLDL samples or the total LDL.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Bar graph shows relationship between LDL- content and oxidation lag time (using 10 µmol/L CuSO4) for LDL isolated from 35 subjects. Inset shows formula for exponential decay to which data were fit and the correlation coefficient for the fit of the data points to the exponential curve. A weak correlation was found between LDL- and the lag time for samples containing >2% LDL-.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 5. Bar graph shows oxidation lag times for nLDL and LDL- isolated by HPLC with 10 µmol/L CuSO4 and 2 µmol/L hematin as catalysts. Samples were continuously monitored at 234 nm; lag time was determined as described in "Methods." Results are mean±SE from five independent measurements on nLDL and LDL-.

A comparison of the kinetics of Cu2+-induced oxidation for LDL- and nLDL is provided in Fig 6Down. The tracings represent a typical kinetic profile for LDL oxidation monitored at 234 nm, wherein the oxidation rate is determined during the lag phase and the duration of the lag phase is estimated according to the method of Esterbauer et al,28 along with the rate of oxidation during the propagation phase. The determination of the lag rates, lag-phase duration, and rates of propagation (log rate) were made from these tracings. No appreciable increase in conjugated diene content was found for nLDL for periods of {approx}1 hour. Thereafter, a progressive increase in conjugated diene content took place that represented the transition from the lag phase to the propagation phase, which is clearly evident by 80 minutes. Under these conditions, LDL- oxidized immediately after CuSO4 was added, as shown by rapid accumulation of conjugated dienes during the presumptive lag phase followed by rapid propagation reactions that began by {approx}50 minutes. The susceptibility of LDL- to onset of oxidation prompted further determination of its contribution to overall LDL oxidation as suggested from the data in Fig 4Up.



View larger version (10K):
[in this window]
[in a new window]
  		Figure 6. Representative spectrophotometric tracings show the oxidation of 0.5 mg/mL each nLDL (—) and LDL- (– – – –) monitored at 234 nm after addition of 10 µmol/L CuSO4. Optical density (O.D.) was continuously monitored for 150 minutes while maintaining the samples at 25°C in PBS.

Contribution of LDL- to LDL Oxidation
The susceptibility of LDL- to Cu2+- or heme-induced oxidation may be based on a higher lipid peroxide content, which is known to facilitate LDL oxidation.34 LDL- contains greater amounts of lipid peroxidation products than nLDL and is labile to oxidation.7 11 Lipid peroxide content was determined from the levels of conjugated dienes by second-derivative UV spectral analysis of lipid extracts from LDL- and nLDL, which showed significant differences at the time these lipoproteins were isolated (Fig 7Down). Upon further oxidation with Cu2+ and reanalysis after an arbitrarily established 1-hour interval, large increases in lipid peroxidation products (designated as ROOH in Fig 7Down) were measured in LDL-, whereas small increases were found for pure nLDL. This difference in oxidation rates is suggested from the data in Fig 6Up. The content of peroxidation products accumulating in nLDL after oxidation, representing the end of the lag phase, which was usually {approx}1 hour, matched the levels in LDL- before oxidation ({approx}15 nmol ROOH/mg protein). This level of peroxide in freshly isolated LDL- may account for its nearly immediate oxidation after addition of Cu2+. The content of lipid peroxides was also measured by means of the leucomethylene blue reaction with similar results, ie, nLDL before oxidation contained 9.7±3.9 and LDL- contained 35.3±9.1 nmol peroxide/mg LDL protein. The small discrepancy between values obtained by measurement of conjugated dienes and the reaction with leucomethylene blue may be due to the presence of oxidants other than fatty acid hydroperoxides in LDL that can generate the colored methylene blue product. Measurement of vitamin E levels before and after oxidation (Fig 7Down) showed that there was a rapid loss for both nLDL and LDL-, even though nLDL contained substantially more vitamin E than LDL- before the onset of oxidation. It appears that oxidation of vitamin E by copper is rapid and essentially complete by the 1-hour time point selected for these analyses. The preferential oxidation of LDL vitamin E by CuSO4 has been described.35



View larger version (23K):
[in this window]
[in a new window]
  		Figure 7. Bar graph. Lipid peroxide (ROOH) levels were estimated from the conjugated diene absorbance of the total lipid extracts isolated from nLDL and LDL- and converted to molar levels of peroxides by using an extinction coefficient of 2.54x104 (representative of linoleic acid-13-hydroperoxide). Lipid peroxide content was measured before and after the samples were incubated in the presence of 10 µmol/L CuSO4 for 1 hour at 25°C. Results are mean±SE from five independent determinations with different nLDL and LDL- preparations. *P<.05 nLDL vs LDL- before oxidation. **P<.001 nLDL vs LDL- after oxidation. ***P<.080 nLDL before and after oxidation (all by two-tailed t test).

The findings described in Figs 4 through 7UpUpUpUp suggest that the oxidative susceptibility of LDL may depend on the content of LDL-, which provides catalytically sufficient amounts of peroxide to stimulate propagation of lipid peroxidation. This propagation can occur in nLDL as well as LDL- when the lipoproteins exist as a mixed population, which occurs in the small dense fraction of LDL ({delta}>1.055 g/mL). To determine whether this is the case, purified nLDL (containing <0.5% LDL-) was incubated in the presence of LDL-, in which the content of LDL- added ranged from 1% to 5%, ie, representing the range of LDL- found in the LDL samples from subjects. Whenever possible, LDL- and nLDL from the same subject were used; in cases in which the total plasma LDL of the subject was low, the LDL- from a number of subjects was pooled to provide enough material for study. The results show the effect of added LDL- on the measured oxidative lag phase as well as the rates of oxidation during the lag and propagation phases (Fig 8Down). A progressive decrease in the lag time was found with increasing amounts of LDL- added to nLDL. The large SE of this analysis, owing to the variability of LDL oxidation among the subjects, required statistical analyses that measured the trend across the groups according to the LDL- content. This was done by repeated measures ANOVA, which compares each successive addition of LDL- to nLDL. Table 2Down shows that a significant decrease (P<.06) occurs in the lag time between additions of 2% and 3% LDL- to nLDL. A significant difference is also found between the groups containing 3% and 5% added LDL-. Addition of LDL- to nLDL also increased the rates of oxidation during the lag phase (lag rate), in which a progressive increase in the lag rate of oxidation was found with increasing amounts of added LDL-. A difference was found with the addition of 5% LDL- compared with nLDL alone (P<.06), and a significant increase was found for the trend across successive LDL- additions (P<.06). By contrast, no effect of LDL- addition was found for the propagation phase (log rate) of oxidation. Statistical analysis of these data is also included in Table 2Down. Fig 9Down compares the oxidation kinetics for nLDL alone with that of nLDL to which was added LDL- in 1% increments up to a total of 5% LDL-. Fig 9Down presents the results on four separate LDL preparations from different subjects.



View larger version (50K):
[in this window]
[in a new window]
  		Figure 8. Bar graph shows effect of LDL- addition to nLDL as measured by oxidation lag time with 10 µmol/L CuSO4 as the catalyst (gray bars) and oxidation rates for LDL containing increasing proportions of LDL- relative to nLDL. Rates of Cu2+-induced oxidation are presented on the basis of the change in conjugated diene absorbance (OD234) per minute while continuously monitoring the LDL samples over 300 minutes at 25°C. Log rate (hatched bars) indicates the rate of conjugated diene accumulation after the lag phase (representing the propagation phase of LDL lipid peroxidation), and lag rate (black bars) indicates the rate of conjugated diene accumulation during the lag phase. Cuvettes containing 0.5 mg nLDL protein were mixed with LDL- added at the percent of total LDL levels indicated in the figure. Results are mean±SE from four independent experiments. Statistical analysis of the data with repeated measures ANOVA comparing successive additions of LDL- to nLDL is presented in Table 1Up.


View this table:
[in this window]
[in a new window]
  		Table 2. Statistical Analysis of Data from Fig 8Up Using Repeated Measures ANOVA



View larger version (14K):
[in this window]
[in a new window]
  		Figure 9. Typical oxidation profiles for nLDL to which LDL- was added in successive 1% increments, expressed as wt/wt protein. The optical density (OD) was continuously monitored over 120 minutes while maintaining the samples at 25°C in PBS. Samples consisted of 0.5 mg/mL nLDL alone (—) and nLDL to which LDL- was added at 1% (– – – –), 2% (. . . .), 3% (- - - -), 4% (—), and 5% (– . – .). Tracings are offset along the y axis to resolve kinetic profiles.

The lipid peroxidation evident in these experiments was not due solely to that formed in LDL- for three reasons. (1) The amount of LDL- in the samples was too low to account for the levels of conjugated dienes produced during the analysis. Thus, a 2% solution of LDL- (ie, 10 µg protein) subjected to oxidation in the presence of 10 µmol/L Cu2+ produces unmeasurable amounts of conjugated dienes, whereas substantial amounts of conjugated dienes accumulate when 2% LDL- is added to a sample containing 480 µg nLDL protein per sample. (2) Collection of samples after oxidation and analysis by anion-exchange HPLC revealed that the amounts of LDL- were greater than the amounts added at the beginning of the analysis, increasing to proportions as high as 30% of total LDL (data not shown). (3) Pretreatment of LDL- with the glutathione peroxidase analogue ebselen36 resulted in a marked increase in oxidation lag times (44±14.6 minutes for nLDL+5% LDL- versus 75±16.1 minutes for nLDL+5% LDL- pretreated with ebselen) and a decrease in oxidation lag rates (0.0125±0.0041 OD234 U/min for nLDL+5% LDL- versus 0.0082±0.0043 OD234 U/min for nLDL+5% LDL- pretreated with ebselen). These findings indicate that LDL- stimulates the oxidation of nLDL, and reduction of LDL--associated peroxides suppresses the pro-oxidant effect.

Ingold et al37 report that peroxidative processes in LDL are influenced by antioxidants outside the particle in the aqueous phase. Prominent among these antioxidants is ascorbic acid, which interacts with antioxidants in LDL, notably ubiquinone and vitamin E, and where recycling of the latter during LDL oxidation has been described.38 In the presence of 100 µmol/L ascorbic acid there was a marked reduction in the oxidation lag rates when nLDL was incubated in the presence of LDL- (Fig 10Down), even up to an LDL- content of 5%. A similar effect of ascorbic acid was found on the basis of the measured oxidation lag time. In the case of nLDL the lag time increased in the presence of 100 µmol/L ascorbate (67±8.2 versus >300 minutes for nLDL versus nLDL+ascorbate, respectively), and the lag times in the presence of 5% LDL- were also extended in the presence of 100 µmol/L ascorbate (34±10.8 versus >300 minutes for nLDL+5% LDL- versus nLDL+5% LDL-+ascorbate). The protective effect of ascorbic acid was not found when LDL- alone was subjected to Cu2+-induced oxidation, as lag times were 28.3±5.05 minutes for LDL- versus 29.9±4.45 minutes for LDL-+ascorbate.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 10. Bar graph shows effect of ascorbic acid on LDL oxidation rate during lag phase (lag rates). Samples were prepared similarly to those shown in Fig 9Up, ie, the LDL preparations contained increasing proportions of LDL- relative to nLDL. Rates of Cu2+-induced oxidation are presented on the basis of the change in conjugated diene absorbance (OD234) per minute while monitoring the samples at 25°C. Oxidation rates are shown for mixtures of nLDL and LDL- and pure nLDL and pure LDL- alone. Results are mean±SE from three independent experiments except for LDL- treated in the presence of ascorbate (indicated by asterisk), which is mean±SE of two determinations. aP<.05 nLDL+3% LDL- absence vs presence of ascorbate; bP<.01 nLDL+5% LDL- absence vs presence of ascorbate (all by paired Student's t test).


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
The oxidative modification of LDL- is determined, at least in part, by its content of lipid peroxidation products, which are substantially greater than those found in unmodified LDL. LDL- contains greater amounts of thiobarbituric acid–reacting products7 and cholesterol oxidation products5 and lower vitamin E content than nLDL,7 findings consistent with the higher levels of lipid conjugated dienes and lipid peroxides measured in this study. Analysis of the phenotypic pattern of LDL density subclasses was not performed in this study, nor were the levels of glycation or sialylation39 determined, which could be related to the density, electronegativity, extent of oxidative modification, and susceptibility to in vitro oxidation. On the other hand, an interesting relationship between LDL- content and the oxidative susceptibility of plasma LDL density was found. Analysis of LDL isolated from 35 different subjects showed a strong relationship (r=.678) between the proportion of LDL- in total LDL and the oxidative lag phase. Individuals with <2% LDL- in total plasma LDL showed a strong negative correlation (r=.843) between the Cu2+-induced oxidative lag phase and the amount of LDL-. With an LDL- content >2%, a weak relationship was found between LDL- levels and the lag phase. This may be due to large variations in the rates of oxidation when LDL- levels are high and/or that the very short lag phases prevented accurate delineation of oxidative susceptibility under the conditions employed. Nevertheless, samples containing higher amounts of added LDL- were consistently more susceptible to oxidation. Based on the content of lipid peroxidation products measured in this study, nLDL contains three- to eightfold less peroxide than LDL- per milligram protein, depending on the assay method. The level of peroxides in LDL- may be sufficient to exceed the threshold required for the catalytic propagation of lipid peroxidation in LDL- particles (Fig 7Up). Thus, the duration of the lag phase for LDL- is very short. This level of peroxide is similar to that formed at the end of the oxidation lag period and may account for the peroxide threshold needed to facilitate the propagation of peroxidation in nLDL, observed as the log phase of oxidation. Indeed, the initiation of Cu2+-dependent LDL oxidation is enhanced by trace amounts of hydroperoxides, which facilitate propagation reactions.34 40 41

LDL resistant to oxidation has low levels of hydroperoxides (<0.5 nmol/mg LDL), clearly below the levels found in the oxidation-resistant nLDL fraction described in the present study. Based on the present findings, the peroxide threshold for facile propagation of LDL lipid peroxidation appears to be in the range of 10 to 15 nmol/mg LDL. This range is comparable to the threshold reported for hematin-induced peroxidation of unsaturated phosphatidylcholine liposomes,27 ie, 15 to 20 nmol/mg lipid. Interestingly, 50 µmol/L H2O2 was required to facilitate rapid oxidation of LDL by hemin.42 We hypothesize that peroxides at this concentration range are sufficient to sustain catalytic reactions involving Cu2+ or heme compounds and propagate lipid peroxidation.43

The predisposition of dense LDL to oxidation may not only be attributable to the larger amounts of lipid peroxidation products but also may be affected by a lower antioxidant capacity. The antioxidant deficit may account for rapid rates of peroxide accumulation upon addition of catalysts such as Cu2+ or hematin. Our findings are in general agreement with those of Tribble et al23 and others44 45 46 47 in that the most susceptible particles were found among the denser LDL fractions, which are enriched in lipid peroxides. However, depending on the method used for ultracentrifugation and particularly the kind of density gradient employed, variations in subfraction distribution can be expected. Thus, the densest fraction ({delta}>1.050 g/mL) appeared in our study to be relatively resistant to oxidation even though it is enriched in LDL- and peroxides. Although a single measurement was made on this fraction due to limited amounts of material and since determinations were based on cholesterol rather than protein content (due to low lipid-to-protein ratios in the densest fraction), it is difficult to conclude which other factors may have contributed to the oxidation characteristics in this density subfraction. The kinetics of LDL oxidation indicate that oxidation proceeds at more rapid rates for LDL enriched in LDL- content, even during the lag period. This is evident from the data in Fig 8Up, which show that addition of LDL- to nLDL at 1% to 5% increments caused an incremental increase in the rate of conjugated diene formation during the lag period (lag rate). In contrast, the rate of conjugated diene formation measured during the propagation phase was unaffected by LDL- content. This effect can be explained by the conventional notion that the lag phase represents the antioxidant-protected period of LDL oxidation, while the log phase represents oxidation during the antioxidant-depleted period. The latter is likely influenced by the availability of oxidizable lipids and the kinetics of radical propagation reactions.

The pro-oxidant effect of LDL- is proposed as resulting from the relatively higher content of lipid peroxidation products that exist in this subpopulation of LDL particles. Circulating LDL particles, or more likely the LDL in the extravascular spaces (from which plasma LDL- may originate), are likely to undergo rapid oxidation when the content of LDL- is high. This oxidation appears to occur by interaction between LDL- and nLDL particles. Reactions involving peroxides originating from extrinsic sources have been proposed for LDL oxidation, as in the case of endothelial cell–mediated oxidation involving lipoxygenase acting on cell polyunsaturated fatty acids, the products of which may transfer to LDL.1 Evidence in support of "interparticle reactions" was indicated from experiments in which ascorbic acid produced a strong inhibition of LDL--mediated oxidation of nLDL. Protection of nLDL and samples consisting of nLDL and LDL- was found when ascorbic acid was present in the medium, but protection was not afforded to pure LDL-. The protective effect of ascorbic acid was recently described for the more oxidatively labile dense LDL subfractions,48 wherein differences between buoyant and dense LDL oxidation appear to be reversed in the presence of ascorbate. The findings indicate that ascorbic acid either facilitates the antioxidant activity in nLDL when challenged by Cu2+ in the presence of LDL-, prolonging the antioxidant activity of LDL-associated antioxidants such as vitamin E, or serves as an antioxidant/reductant in the aqueous phase (vide infra). Under these conditions, ascorbic acid could intercept radical propagating species, such as lipid peroxides, or derived radicals propagating from LDL- to nLDL. These two possibilities may be operating simultaneously since the pro-oxidant effect of peroxides on LDL oxidation has been shown to be inhibited by ascorbic acid.49 The inability to protect LDL- alone may be due to inadequate vitamin E and/or other antioxidants or to the levels of lipid peroxides, which may be sufficiently high to cause rapid propagation of lipid peroxidation upon Cu2+ addition. Direct reduction of lipid peroxides by ascorbate during the propagation phase is not described, although reaction with peroxyl radicals is thermodynamically possible. Ascorbic acid inhibits oxidation of LDL by Cu2+ or hemin.50 51 52 This antioxidant action may occur by chemical reduction of LDL-associated antioxidants, such as vitamin E, that in turn can interact with and reduce lipid peroxyl or derived radicals. Since depletion of vitamin E appears to be rapid after addition of CuSO4, it was difficult to determine the extent to which preservation of vitamin E serves to enhance the oxidative resistance of these LDL preparations.

The protective effect of ascorbate resembles that of selenoenzymes53 or ebselen.34 Ebselen functions as a mimetic for glutathione peroxidase and related selenoenzymes.36 In this capacity it reduces lipid hydroperoxides in all LDL lipids34 and prevents further lipid autoxidation. Ebselen effectively reduces LDL-associated peroxides if added in molar excess to the lipoprotein peroxide levels, where reaction with the peroxide is very rapid.52 A similar inhibitory effect is produced by phospholipid hydroperoxide glutathione peroxidase.41 The similar effects of ascorbate and ebselen suggest that facilitation of LDL oxidation by LDL- is peroxide dependent. The effect of ascorbic acid on Cu2+-induced LDL oxidation may also involve metal chelation. This chelation process likely functions by retaining Cu2+ in the aqueous phase or interacting with apoprotein-bound Cu2+, processes known to facilitate ascorbate oxidation and reduction of Cu2+.54 A reduction of LDL-- associated tocopheryl radicals leading to the elimination of the otherwise oxidizing radical species into the aqueous phase and recycling of vitamin E has also been described.37

According to our procedure for LDL isolation and density fractionation, LDL- is associated largely with particles with densities >1.040 g/mL. Small amounts were also detected in more buoyant fractions, but in many samples LDL- was essentially undetectable in the buoyant LDL, whereas it was always found in the dense fractions. Although the oxidative susceptibility of small dense LDL particles may be related to a lower content of {alpha}-tocopherol, other factors may play a more important role since considerable amounts of {alpha}-tocopherol are present in the dense LDL fraction and the {alpha}-tocopherol levels appear to be comparable among the subject groups studied thus far.23 55 The susceptibility of dense LDL can also be influenced by the nature of dietary fatty acid intake; oleic acid–rich diets confer significant protection to dense LDL, whereas linoleic acid–rich diets increase its oxidative susceptibility regardless of the {alpha}-tocopherol levels.44 A decreased vitamin E–to-protein ratio in denser LDL but a normal vitamin E–to-lipid ratio was taken as evidence that other factors, such as differences in fatty acid composition,45 may account for the increased oxidizability of dense LDL. However, others have reported that oxidative susceptibility based on the vitamin E–to–LDL mass could not be accounted for by enrichment in polyunsaturated fatty acid content.46 Tribble et al23 found a lower free cholesterol content in denser LDL subfractions and suggest that a deficit in this constituent may impart a susceptibility to oxidative modification. LDL enriched in LDL- appears in general to be more readily oxidizable, which suggests that a combination of factors, including antioxidant depletion, altered lipid composition or structure,28 37 and increased peroxide content and lipid-to-protein ratio, may contribute to oxidative susceptibility.

Elevated levels of LDL- in total LDL and in certain LDL subpopulations appear to contribute to oxidative susceptibility. It is plausible that the preponderance of lipid peroxides found in the LDL- fraction may serve as catalysts for enhanced oxidative susceptibility. Thus, in the presence of LDL- ranging from 5% to 15% of the total LDL, a sufficient peroxide threshold could exist to facilitate LDL oxidation. We postulate that the denser LDL represents an older population of particles, which, by having a longer circulating half-life, have also experienced more contact with oxidants. Under conditions in which antioxidant levels are low and reducing equivalents are locally depleted and/or antioxidant enzymes are compromised, the rapid oxidation of dense LDL and lipoproteins similar to LDL- could promote further oxidation of lipoprotein particles.


*    Selected Abbreviations and Acronyms
 
HPLC = high-performance liquid chromatography
LDL-C = LDL cholesterol
nLDL = unmodified normal LDL
PBS = phosphate-buffered saline


*    Acknowledgments
 
This work was supported by grants HL50350 and ES03466 from the National Institutes of Health, NATO International Collaborative grant CRG 920513, and the Wright Foundation Research Award from the University of Southern California, School of Medicine.


*    Footnotes
 
1 Dr Avogaro died June 11, 1995. Back

Received August 7, 1995; accepted February 6, 1996.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JI. Beyond cholesterol: modifications of low-density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915-924. [Medline] [Order article via Infotrieve]
  2. Esterbauer H, Jurgens G, Quehenberger O, Koller E. Autoxidation of human low density lipoprotein: loss of polyunsaturated fatty acids and vitamin E and generation of aldehydes. J Lipid Res. 1987;28:495-509. [Abstract]
  3. Hessler JR, Morel DW, Lewis LJ, Chisolm GM. Lipoprotein oxidation and lipoprotein-induced cytotoxicity. Arteriosclerosis. 1983;3:215-222. [Abstract/Free Full Text]
  4. Morel DW, DiCorleto PE, Chisolm GM. Endothelial and smooth muscle cells alter low-density lipoprotein in vitro by free radical oxidation. Arteriosclerosis. 1984;4:357-364. [Abstract/Free Full Text]
  5. Hodis HN, Kramsch DM, Avogaro P, Bittolo-Bon G, Cazzolato G, Hwang J, Peterson H, Sevanian A. Biochemical and cytotoxic characteristics of an in vitro circulating oxidized LDL. J Lipid Res. 1994;35:669-677. [Abstract]
  6. Cazzolato G, Avogaro P, Bittolo-Bon G. Characterization of a more electronegatively charged LDL subfraction by ion exchange HPLC. Free Radic Biol Med. 1991;11:247-253. [Medline] [Order article via Infotrieve]
  7. Avogaro P, Bittolo-Bon G, Cazzolato G. Presence of modified low-density lipoprotein in humans. Arteriosclerosis. 1988;8:79-87. [Abstract/Free Full Text]
  8. Tyroler HA. Review of lipid-lowering clinical trials in relation to observational epidemiologic studies. Circulation. 1987;76:515-522. [Abstract/Free Full Text]
  9. Savenkova MI, Mueller DM, Heinecke J. Tyrosyl radical generated by myeloperoxidase is a physiological catalyst for the initiation of lipid peroxidation in low density lipoprotein. J Biol Chem. 1994;269:20394-20400. [Abstract/Free Full Text]
  10. Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases endothelial superoxide anion production. J Clin Invest. 1993;91:2546-2551.
  11. Shimano H, Yamada N, Ishibashi S, Hiroshi H, Mori N, Gotoda T, Harada K, Akanuama Y, Murase T, Yazaki Y, Takaku F. Oxidation-labile subfraction of human plasma low density lipoprotein isolated by ion exchange chromatography. J Lipid Res. 1991;32:763-770. [Abstract]
  12. Berliner JA, Territo MC, Sevanian A, Ramin S, Kim J, Bamshad B, Esterson M, Fogelman AM. Minimally modified LDL stimulates monocyte-endothelial interactions. J Clin Invest. 1990;85:1260-1266.
  13. Avogaro P, Cazzolato G, Bittolo-Bon G. Some questions concerning a small, more electronegative LDL circulating in human plasma. Atherosclerosis. 1991;91:163-166. [Medline] [Order article via Infotrieve]
  14. Shen MS, Krauss RM, Lindgren FT, Forte TM. Heterogeneity of serum low density lipoproteins in normal human subjects. J Lipid Res. 1981;22:236-242. [Abstract]
  15. Nelson CA, Morris MD. The ultracentrifugal heterogeneity of low density lipoproteins in normal humans. Biochem Med. 1977;18:1-8. [Medline] [Order article via Infotrieve]
  16. Kirchenhausen TG, Fless G, Scanu AM. The ultracentrifugal heterogeneity of low density lipoproteins: experimental facts and interpretation: a minireview. Lipids. 1980;15:464-467. [Medline] [Order article via Infotrieve]
  17. Sundaram SG, Shakir KMM, Margolis S. Preparative isoelectric focusing of human very low-density and low density lipoproteins. Anal Biochem. 1978;88:425-429. [Medline] [Order article via Infotrieve]
  18. Austin MA, Breslow JL, Hennekens CH, Buring JE, Willette WC, Krauss RM. Low density lipoprotein subclass patterns and risk of myocardial infarction. JAMA. 1988;260:1917-1921. [Abstract]
  19. Krauss RM, Burke DJ. Identification of multiple subclasses of plasma low density lipoproteins in normal humans. J Lipid Res. 1982;23:97-104. [Abstract]
  20. Nichols AV, Krauss RM, Musliner TA. Nondenaturating polyacrylamide gradient gel electrophoresis. Methods Enzymol. 1986;128:417-426. [Medline] [Order article via Infotrieve]
  21. McNamara JR, Campos H, Ordovas JM, Peterson J, Wilson PWF, Schaefer EJ. Effect of gender, age, and lipid status on low-density lipoprotein subfraction distribution: results of the Framingham Offspring Study. Arteriosclerosis. 1987;7:483-497. [Abstract/Free Full Text]
  22. Austin MA, King MC, Vranizan KM, Newman B, Krauss RM. Inheritance of low-density lipoprotein subclass patterns: results of complex segregation analysis. Am J Hum Genet. 1988;43:838-846. [Medline] [Order article via Infotrieve]
  23. Tribble DL, Holl LG, Wood PD, Krauss RM. Variations in oxidative susceptibility among six low density lipoprotein subfractions differing in density and particle size. Atherosclerosis. 1992;93:189-199. [Medline] [Order article via Infotrieve]
  24. Lipid Research Clinics Program. The Manual of Laboratory Operations: Lipids and Lipoprotein Analysis. Bethesda, Md: National Institutes of Health; 1974. Dept of Health, Education, and Welfare publication NIH 75-628.
  25. Laurell CB. Electroimmunoassay. Scand J Clin Lab Invest. 1972;29(suppl 124):21-37.
  26. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265-275. [Free Full Text]
  27. Kim E, Sevanian A. Hematin and peroxide catalyzed peroxidation of phospholipid liposomes. Arch Biochem Biophys. 1991;288:324-330. [Medline] [Order article via Infotrieve]
  28. Esterbauer H, Gebicki J, Puhl H, Jurgens G. The role of lipid peroxidation and antioxidants in oxidative modification of LDL. Free Radic Biol Med. 1992;13:341-390. [Medline] [Order article via Infotrieve]
  29. Sattler W, Maiorino M, Stocker R. Reduction of HDL- and LDL-associated cholesteryl-ester and phospholipid hydroperoxides by phospholipid hydroperoxide glutathione peroxidase and ebselen (PZ51). Arch Biochem Biophys. 1994;309:214-221. [Medline] [Order article via Infotrieve]
  30. Corongiu FP, Banni S, Dessi MA. Conjugated dienes detected in tissue lipid extracts by second derivative spectroscopy. Free Radic Biol Med. 1989;7:183-186. [Medline] [Order article via Infotrieve]
  31. Pacifici EHK, McLeod LL, Peterson H, Sevanian A. Linoleic acid hydroperoxide-induced peroxidation of endothelial cell phospholipids and cytotoxicity. Free Radic Biol Med. 1994;17:285-295. [Medline] [Order article via Infotrieve]
  32. Auerbach BJ, Kiely JS, Cornicelli JA. A spectrophotometric microtiter-based assay for the detection of hydroperoxy derivatives of linoleic acid. Anal Biochem. 1992;201:375-380. [Medline] [Order article via Infotrieve]
  33. Bieri JG, Tolliver TJ, Catignani GL. Simultaneous determination of {alpha}-tocopherol and retinol in plasma or red cells by high pressure liquid chromatography. Am J Clin Nutr. 1979;32:2143-2149. [Abstract/Free Full Text]
  34. Thomas CE, Jackson RL. Lipid hydroperoxide involvement in copper-dependent and independent oxidation of low density lipoproteins. J Pharmacol Exp Ther. 1991;256:1182-1188. [Abstract/Free Full Text]
  35. Maiorino M, Zamburlini A, Roveri A, Ursini F. Copper-induced lipid peroxidation in liposomes, micelles, and LDL: which is the role of vitamin E? Free Radic Biol Med. 1995;18:67-74. [Medline] [Order article via Infotrieve]
  36. Muller A, Cadenas E, Graf P, Sies H. A novel biologically active seleno-organic compound, I: glutathione peroxidase-like activity in vitro and antioxidant capacity of PZ51 (ebselen). Biochem Pharmacol. 1984;33:3235-3239. [Medline] [Order article via Infotrieve]
  37. Ingold KU, Bowry VW, Stocker R, Walling C. Autoxidation of lipids and antioxidation by {alpha}-tocopherol and ubiquinol in homogeneous solution and in aqueous dispersions of lipids: unrecognized consequences of lipid particle size as exemplified by oxidation of human low density lipoprotein. Proc Natl Acad Sci U S A. 1993;90:45-49. [Abstract/Free Full Text]
  38. Kagan VE, Serbinova EA, Forte T, Scita A, Packer L. Recycling of vitamin E in human low density lipoproteins. J Lipid Res. 1992;33:385-397. [Abstract]
  39. Tertov VV, Sobenin IA, Gabbasov ZA, Popov EG, Jaakkola O, Solakivi T, Nikkari T, Smirnov VN, Orekhov AN. Multiple-modified desialylated low density lipoproteins that cause intracellular lipid accumulation: isolation, fractionation and characterization. Lab Invest. 1992;67:665-675. [Medline] [Order article via Infotrieve]
  40. O'Leary VJ, Darley-Usmar VM, Russel LJ, Stone D. Pro-oxidant effects of lipoxygenase-derived peroxides on the copper-initiated oxidation of low density lipoprotein. Biochem J. 1992;282:631-634.
  41. Thomas JP, Kalyanaraman B, Girotti AW. Involvement of preexisting lipid hydroperoxides in Cu2+-stimulated oxidation of low-density lipoprotein. Arch Biochem Biophys. 1994;315:244-254. [Medline] [Order article via Infotrieve]
  42. Belcher JD, Balla J, Balla G, Jacobs DR, Gross M, Jacob HS, Vercellotti GM. Vitamin E, LDL, and endothelium: brief oral vitamin supplementation prevents oxidized LDL-mediated vascular injury in vitro. Arterioscler Thromb. 1993;13:1779-1789. [Abstract/Free Full Text]
  43. Galaris D, Sevanian A, Cadenas E, Hochstein P. Ferrylmyoglobin-catalyzed linoleic acid peroxidation. Arch Biochem Biophys. 1990;281:163-169. [Medline]