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. 1997;17:786-796

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 Goulinet, S.
Right arrow Articles by Chapman, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Goulinet, S.
Right arrow Articles by Chapman, M. J.
(Arteriosclerosis, Thrombosis, and Vascular Biology. 1997;17:786-796.)
© 1997 American Heart Association, Inc.

Articles

Plasma LDL and HDL Subspecies Are Heterogenous in Particle Content of Tocopherols and Oxygenated and Hydrocarbon Carotenoids

Relevance to Oxidative Resistance and Atherogenesis

Sylvie Goulinet; ; M. John Chapman

From the Institut National de la Santé et de la Recherche Médicale (INSERM), Unité 321, Unité de Recherches sur les Lipoprotéines et Athérogénèse, Pavillon Benjamin Delessert, Hôpital de la Pitié, Paris, France.

Correspondence to Mlle S. Goulinet and Dr M.J. Chapman, INSERM U.321, Pavillon Benjamin Delessert, Hôpital de la Pitié, 75651 Paris Cedex 13, France.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Abstract Epidemiological data indicate that dietary tocopherols and carotenoids can exert cardioprotective effects, which may be mediated by their antioxidant actions. The oxidative modification of LDL underlies the atherogenicity of these cholesterol-rich particles. The resistance of LDL to oxidation is influenced by several endogenous factors, among which the content of tocopherols and carotenoids is prominent. Of the exogenous factors, HDL inhibits oxidation of LDL via several mechanisms. In view of the paucity of data on the distribution of diverse tocopherol and carotenoid components among the apoB- and apoA-I–containing lipoproteins of human plasma, we evaluated the quantitative and qualitative features of the LDL and HDL particle subspecies in normolipidemic subjects. The bulk of tocopherols and hydrocarbon carotenoids (lycopene, {alpha}- and ß-carotene) was transported in LDL (45% and 76%, respectively), in contrast to the oxygenated carotenoids (lutein/zeaxanthin, canthaxanthin, and ß-cryptoxanthin), which were equally distributed between LDL and HDL. {alpha}-Tocopherol content was independently associated with lipid core size (cholesteryl ester and triglyceride) in VLDL, LDL, and HDL (P<.005); by contrast, the particle content of the oxygenated carotenoids lutein/zeaxanthin and canthaxanthin was strongly related to that of phospholipids. A significant and progressive decrease in the molar content of {alpha}- and {gamma}-tocopherols was found with increase in density from light to dense LDL subspecies (LDL1 to LDL5); a similar trend was observed in HDL subspecies. Furthermore, particle contents of lutein/zeaxanthin, ß-cryptoxanthin, ß-carotene, and lycopene were markedly reduced in small, dense LDL (LDL5, d=1.050 to 1.065 g/mL). We conclude that diminished contents in such carotenoids as well as in tocopherols could underlie not only the diminished oxidative resistance of small, dense LDL but also reduced tissue targeting of antioxidants in subjects with a dense LDL phenotype.


Key Words: reversed-phase HPLC • dense LDL • isopycnic density–gradient ultracentrifugation • lipid core • lutein/zeaxanthin


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
There is now substantial evidence that the free radical–mediated oxidation of cholesterol-rich LDL plays a key role in the formation of atherosclerotic lesions in humans.1 2 3 Indeed, such lesions are frequently characterized not only by the presence of cholesterol-rich macrophage foam cells but also by a large lipid core containing cholesterol crystals, oxygenated derivatives of both free and esterified cholesterol, and multiple species of oxidized fatty acids in both free and esterified forms.4 5 6 7 Moreover, advanced atherosclerotic plaques contain significant amounts of both water-soluble and lipophilic antioxidants, such as ascorbate and {alpha}-tocopherol, in their reduced, antioxidant-active forms, in addition to detectable amounts of the HCCs lycopene and ß-carotene.6

The accumulation of cholesterol at lesion sites in the arterial wall appears to result primarily from the specific binding and preferential retention of LDL particles by components of the extracellular matrix.8 9 Recent findings suggest that retention of LDL in the arterial intima exposes these particles to oxidative stress, involving the action of reactive oxygen species and transition metal ions.10 11 Indeed, several cell types present in arterial tissue, including endothelial cells, monocyte-macrophages, T lymphocytes, and fibroblasts, can facilitate LDL oxidation in vitro.1 11 The initial phase of the oxidation of LDL involves the concomitant destruction of its lipophilic antioxidant content and formation of conjugated dienes and lipid hydroperoxides, among which CE hydroperoxides predominate.3 12 13 More advanced stages of lipid peroxidation ensue, with formation of several species of oxysterols,14 the loss of polyunsaturated fatty acids, and the decomposition of lipid hydroperoxides to aldehydes and other reactive products.1 3 15 The reaction of these breakdown products with amino acid residues such as lysine in apoB100 of LDL results in covalent modification, with increase in net negative charge.1 16 Such alterations underlie formation of electronegative, oxidized LDL, which is readily internalized by macrophage scavenger receptors, thereby leading to foam cell formation.1

The degree to which LDL may resist oxidative stress is clearly of paramount importance in atherogenesis and is determined by multiple factors, which are mutually interactive and fall into two groups; those that are exogenous and those that are endogenous to the lipoprotein particle. Of the exogenous factors, oxygen tension and the presence of metal cations such as copper and iron, soluble antioxidants such as vitamin C, HDL particles, lipoxygenases, and oxidative enzymes are of major impact.1 3 11 17 18 19 20 Among the former factors, HDL not only inhibits the copper-induced oxidation of LDL in vitro but may also constitute the primary vehicle for transport and clearance of plasma lipid hydroperoxides, thereby preventing their accumulation in LDL.17 21 The principal endogenous factors include particle size, the number and relative affinities of binding sites for metal cations, the degree of unsaturation of fatty acids in lipid esters (mainly CEs and PLs), lipid hydroperoxide content, antioxidant content and efficacy, and the molecular organization and packing of the lipoprotein core and surface constituents.3 12 13 15 18 22 23 24 25 26

The structural and metabolic heterogeneity of both LDL and HDL particles is now well established and is closely related to their impact on atherogenesis.27 28 Indeed, distinct LDL particle subspecies differ significantly in their susceptibility to oxidation in vitro, small, dense LDL displaying a diminished resistance to copper-mediated oxidation.29 30 31 32 In this context, it is especially relevant that the relationship between antioxidant content, in particular that of tocopherols and carotenoids, and the oxidative resistance of LDL subspecies remains indeterminate; for example, the {alpha}-tocopherol content of dense LDL in normolipidemic subjects has been variously described as similar or significantly reduced in comparison with that of buoyant LDL, while fragmentary data for carotenoids suggest that the content of ß-carotene is similar in LDL subfractions.29 30 31 32 As for LDL, the structural heterogeneity of HDL is equally reflected in the functionality of these particles. Thus, the capacity of HDL to inhibit LDL oxidation in vitro is differentially distributed among HDL particle subpopulations and involves several mechanisms, including transition metal binding, paraoxonase activity, and free radical trapping.17 33 34 Moreover, antioxidants such as tocopherols and carotenoids appear to be differentially distributed between apoE-rich and apoE-poor HDL subfractions.35

Until recently, the presence of ubiquinol-10, tocopherols, and carotenoids in plasma lipoproteins has been considered to endow them with antioxidant activity, consistent with their potential to quench lipid peroxyl radicals.3 12 15 18 23 32 In this respect, however, carotenoids are distinct from tocopherols in their capacity to trap not only lipid peroxyl radicals but also singlet oxygen species.36 37 Indeed, carotenoids such as zeaxanthin, canthaxanthin, lutein, and ß-carotene can exert superior antioxidant activity to {alpha}-tocopherol in biological systems.38 By contrast, it is now evident that certain lipophilic antioxidants may act as prooxidants, as demonstrated for {alpha}-tocopherol under conditions of low fluxes of aqueous peroxyl radicals.20 Clearly, then, the relationship between the combined lipoprotein particle content of tocopherols and carotenoids and the nature of the free radical environment constitute key elements in determining whether these lipophilic compounds promote particle oxidation or act as protective antioxidants.

To evaluate the potential contributions of lipophilic antioxidants to the oxidative resistance of lipoprotein particles, we have investigated the quantitative and qualitative distribution of the major molecular forms of tocopherols and carotenoids in defined subpopulations of LDL and HDL particles isolated from normolipidemic plasma. A rigorous approach to the fractionation of lipoprotein subspecies and to the separation and quantification of tocopherols and carotenoids was applied, with the aim of minimizing artifacts due to oxidative changes. Our results reveal a significant and progressive decrease in the molar content of {alpha}- and {gamma}-tocopherol in both LDL and HDL subspecies, with diminution in particle size and lipid core from light to dense particles. Moreover, the particle contents of specific OXCs and HCCs, notably lutein/zeaxanthin, ß-cryptoxanthin, ß-carotene, and lycopene, were markedly reduced in small, dense LDL, thereby contributing to the diminished oxidative resistance of these particles at both normal and reduced oxygen tensions.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Biological Samples
Plasma
Blood was collected from healthy, fasting normolipidemic male subjects into sterile evacuated tubes (Vacutainer) containing K3EDTA (Ref 367655, Becton Dickinson). Plasma was then immediately isolated by centrifugation (1200g for 20 minutes, Jouan model CR4 11) at 4°C and treated as described below. An aliquot (200 µL) was kept in the dark and maintained at -20°C under argon with 0.025% (wt/vol) BHT (Farmitalia Carlo Erba SpA) until HPLC analysis.

Lipoprotein Subfractions
Lipoproteins were subfractionated on the basis of their hydrated density by gradient ultracentrifugation as described earlier.39 On completion of ultracentrifugation, fractions of 0.8 mL were collected successively downward from the meniscus of the tube with a Gilson precision pipette. However, the two first fractions were removed in a volume of 0.4 mL with a narrow-bore Pasteur pipette to permit a more satisfactory separation of VLDL, which tended to adhere to the tube walls. Each nondialyzed lipoprotein subfraction was immediately frozen in aliquots containing 300 µg of total cholesterol, with the exception of VHDL, and the bottom fractions were aliquotted in a maximal volume of 2 mL. All fractions were stored under the same conditions as the plasma samples.

Chemical Analyses
The analytical methods employed for determination of the protein and lipid contents (FC, CE, TG, and PL) of whole plasma and of isolated lipoprotein subfractions were as described earlier.40

Standard Solutions of Lipid-Soluble Tocopherols and Carotenoids
Solutions were prepared in indirect light in inactinic glassware at room temperature. Solvents used were HPLC or analytical-grade quality and were degassed before use.

(+)-{alpha}-Tocopherol ({approx}95%, in the form of an oil), (+)-{alpha}-tocopherol acetate (oil form), lycopene (90% to 95%, crystalline form), {alpha}-carotene (type V, crystalline and substantially free of ß-carotene), and ß-carotene (type IV, crystalline and substantially free of {alpha}-carotene) were purchased from Sigma-Aldrich Chimie. Retinol (all trans, crystalline) and d-{gamma}-tocopherol (crystalline) were obtained from ICN Pharmaceuticals France. Lutein, zeaxanthin, ß-cryptoxanthin, and canthaxanthin were the kind gift of Hoffmann La Roche (Basel, Switzerland). All compounds were stored under argon at -20°C.

Individual stock solutions were prepared fresh, except for the isomers of tocopherols, which are stable for 3 months at -20°C. The retinol standard was prepared in absolute ethanol ({approx}0.3 mg/mL) and the following tocopherols in 95% to 96% ethanol: {alpha}-tocopherol ({approx}10 mg/mL), {gamma}-tocopherol ({approx}5 mg/mL), and {alpha}-tocopherol acetate ({approx}10 mg/mL). All carotenoids were dissolved in chloroform ({approx}0.3 mg/mL). The concentration of each solution was determined by spectrophotometry (Kontron, model Uvikon 930) using the appropriate absorption coefficients in the literature.41 42 Concentrations were corrected after analysis by HPLC to evaluate the purity and possible presence of degradation products.

Working combined standard solutions, corresponding to accurate dilutions of the stock standards in ethanol containing 0.025% BHT, were used immediately for preparation of standard curves in a plasma matrix. Final concentrations of each analyte were in the range of 0.1 to 10 µg/mL. {alpha}-Tocopherol acetate was used as internal standard to correct the recovery of the analytes after the extraction procedure (20 µg/mL in ethanol containing 0.025% BHT).

HPLC Analysis
Sample Extraction
Samples were extracted in duplicate, on ice and with degassed solvents. Sample volume was adjusted to 1 mL with physiological saline buffer. The internal standard (100 µL) was added followed by 900 µL ethanol containing 0.025% BHT. Tubes were vortexed vigorously for 1 minute (IKA-Vibrax-VXR, OSI). n-Heptane (1.5 mL) was then added, vortexed for 5 minutes, and centrifuged at 4°C for 5 minutes at 2000 rpm. One milliliter was transferred from the n-heptane phase to conical vials (1.1 CTVG, Chromacol) and evaporated under argon, and the extraction was repeated once with 0.75 mL n-heptane. The supernatant (0.8 mL) was combined with the precedent extract and evaporated to dryness. The residue was reconstituted in 40 µL of mobile phase containing 20% tetrahydrofuran stabilized with 0.025% BHT. To facilitate dissolution, the extract was sonicated for 1 minute.

Chromatographic Conditions
The chromatographic apparatus was a Beckman system "Gold" HPLC and consisted of a 126 binary pump, a 168 programmable diode array detector, and a 507 automated sample injector. Extract (10 µL) was separated on an ODS Hypersil column (C18, 5 µm, 200*2.1 mm; Hewlett Packard) using a mobile phase of acetonitrile:dichloromethane:methanol with 25 mmol/L ammonium acetate and containing 0.025% triethylamine (vol/vol/vol, 89.5:10.0:0.5) at a flow rate of 0.4 mL/min. The mobile phase was purged with helium gas for about 30 minutes before use, and column temperature was maintained at 40°C. Detection was performed simultaneously at 326 nm for retinol, 292 nm to measure tocopherol isomers, and 450 nm to detect carotenoids. Peaks were identified by comparing the retention time with those of the corresponding standards and by their characteristic spectrum in the UV or visible region.

Calibration Curves
Standard curves were prepared each month in a plasma matrix by the addition of aliquots of working combined standards as suggested by Kaplan et al.41 All additions were performed in triplicate. Standard curves were constructed by plotting the peak area ratios (analyte/internal standard) versus the amount of added analyte, using at least four concentrations for each analyte. The average slope of the three curves was used for the determination of the concentration in assay samples. When the correlation coefficient was <.99 for one of these calibration curves, this curve was excluded.

Recovery
Analytical recovery was measured as suggested by Bui43 with three groups of samples.

CV in the Assay
The within-day and between-day variation in this method was estimated by repetition of extractions and analyses of the same plasma pool and involved 5 injections over a period of 5 hours and 18 injections over a period of 5 months, respectively. The between-duplicate variation was calculated from 45 samples analyzed in duplicate using the formula for the technical error.39

Sample Stability
One extraction of a plasma sample was analyzed with 35 injections over a total period of 9 hours of analysis as described above.

Validation of the Method
For validation of this methodology, we employed a standard reference material (SRM 968a, fat soluble vitamins in human serum) commercialized by NIST, Gaithersburg, Md).

Statistical Methods
All results are presented as mean±SD. The differences between the means of the content of analytes among the lipoprotein subspecies were analyzed with ANOVA. If the F statistic was significant (P<.05), the Bonferroni-Dunn test was used to determine the differences between the subspecies. The same model was used to analyze the influence of the core size (CE+TG), or more polar lipids (PL, FC), on analytes among the lipoprotein fractions. A multiple linear regression was applied to test the existence of a relationship between the molar content of analyte and the molar content of lipid in particle subspecies; ANOVA was used to determine significant variations. Statistical significance was taken as values of P<.05. Statistical analyses were performed with a Statview 4.05 program (Abacus Concept, Berkeley, Calif). The InStat program (GraphPad software, San Diego, Calif) was used to test the trend between the molar content of analytes and the LDL and HDL subspecies, respectively.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Characteristics of the Analytical System
The present HPLC procedure allowed a clear and rapid separation of the major lipid-soluble vitamins, ie, retinol, {gamma}- and {alpha}-tocopherol, and the major carotenoids, ie, xanthophylls (lutein, zeaxanthin, canthaxanthin, ß-cryptoxanthin), lycopene, and {alpha}- and ß-carotene, over a time period of <20 minutes. Fig 1Down shows a representative chromatogram of the HPLC analysis of an extract of human plasma; similar separations were obtained with extracts of lipoprotein subfractions (data not shown). It is noteworthy that a high degree of resolution was achieved for {alpha}- and ß-carotene, although the two major xanthophylls coeluted. Indeed, it was observed that lutein and zeaxanthin could be separated only on columns with polymeric phases.44 The shoulder that we occasionally detected in the lycopene peak appears to correspond to coelution of the cis and trans forms of this substance, as indicated by the presence of a second peak at {approx}350 nm on the absorption spectrum. Our absorption detection was of insufficient sensitivity to allow reliable quantitation of the low concentrations of ubiquinol-10 in lipoprotein extracts.




View larger version (28K):
[in this window]
[in a new window]
  		Figure 1. Representative chromatograms for the HPLC separation of liposoluble retinol, tocopherols, and carotenoids in an extracted plasma sample at indicated wavelengths. The numbers adjacent to each peak indicate the respective retention times. For details of the chromatographic conditions and mobile phase, see the "Methods" section. Roh indicates retinol; {gamma}T, {gamma}-tocopherol; {alpha}T, {alpha}-tocopherol; {alpha}T-ac, {alpha}-tocopherol acetate; IS, internal standard; L/T, lutein/zeaxanthin; Ca, canthaxanthin; ßCr, ß-cryptoxanthin; Ly, lycopene; {alpha}c, {alpha}-carotene; and ßc, ß-carotene.

Recovery and Stability of the Analytes
In view of both their lability and hydrophobicity, several members of the carotenoid family are difficult to recover completely from lipid extracts; equally, loss of such substances may occur as a result of precipitation on the HPLC column. We therefore estimated the recovery of each analyte by adding known amounts of both analyte and internal standard to a pool of plasma and extracting both the spiked and unspiked samples as described in the "Methods" section. We employed the extraction procedure proposed by Elinder and Walldius,45 but with the exception that we introduced a second n-heptane extraction; in this way we recovered at least 91% of {alpha}- and ß-carotene and 96% to 103% of the other analytes (data not shown).

The stability of carotenoids in chlorinated solvents and their dissolution constitute critical steps. The presence of triethylamine and ammonium acetate together,42 in addition to tetrahydrofuran in our sample phase, avoided loss of carotenoids over a period of 9 hours of analysis, with excellent stability of other compounds (CV 2.5±0.8 except for {gamma}-tocopherol, for which the CV was 8.5%). Thereafter, the concentration of lycopene and carotenes started to decline.

Validity, Linearity, and Sensitivity of the Method
The CVs for our procedure are comparable to those found in the literature.41 42 43 45 46 47 Our within-day precision assay (expressed as a percentage of the CV) was superior to our between-day precision for retinol, the tocopherols, lutein, zeaxanthin, canthaxanthin, and lycopene (range of 0.8% to 4.7% versus 4.1% to 11.3%); by contrast, the inverse was true for ß-cryptoxanthin and {alpha}- and ß-carotene (range of 6.4% to 8.3% versus 7.0% to 9.1%). With the exception of {alpha}-tocopherol, for which the between-duplicate assay precision was 5.9%, this parameter was entirely satisfactory for the other compounds (<4%). We observed a linear response over at least an eightfold concentration range for the calibration curves for all analytes with correlation coefficients >.99 (data not shown). The sensitivity of the method, based on a detection limit defined with a signal-to-noise ratio of 10:1, was as follows: {approx}15 ng for {gamma}- and {alpha}-tocopherols and {approx}0.5 ng for other substances.

Validation of the Method
In view of the critical importance of the standardization of the method, we employed special standard reference materials from NIST. Our results reveal good agreement with the values for all components, with the exception of {alpha}-carotene and lycopene, which were inferior to NIST values by an average of 37% and 25%, respectively (data not shown).

Plasma Levels of Retinol, Tocopherols, and Carotenoids
Plasma concentrations of retinol, {alpha}- and {gamma}-tocopherol, lycopene, {alpha}- and ß-carotene, lutein/zeaxanthin, canthaxanthin, and ß-cryptoxanthin are presented in Table 1Down and are entirely consistent with those reported elsewhere for normolipidemic subjects receiving nonsupplemented diets.46 48 49 50 Nonetheless, the wide interindividual variation in plasma {alpha}- and ß-carotene levels (up to sevenfold) probably reflects differences in dietary consumption. It is noteworthy that the lipid and apolipoprotein profile in our subjects is comparable to that described earlier51 ; mean values, expressed in milligrams per deciliter were as follows: 186.6±24.4 for total cholesterol, 87.8±30.5 for TG, 15.0±20.0 for lipoprotein(a), 120.3±23.5 for LDL cholesterol, 47.2±11.5 for HDL cholesterol, 93.4±21.5 for apoB, and 143.4±26.8 for apoA-I (n=11).


View this table:
[in this window]
[in a new window]
  		Table 1. Plasma Concentrations of Retinol, Tocopherols, and Carotenoids in Normolipidemic Subjects

Distribution of Retinol, Tocopherols, and Carotenoids Among Major Density Classes of Lipoproteins and the Ultracentrifugal Residue
The quantitative and qualitative distribution of retinol, tocopherols, and carotenoids in the major density classes of plasma lipoproteins isolated by single-step density-gradient ultracentrifugation, and in the ultracentrifugal bottom residue, is shown in Table 2Down. The recovery of these substances from plasma was >92% for tocopherols, retinol, canthaxanthin, and ß-cryptoxanthin, and >95% for the remaining compounds; recovery of {alpha}- and ß-carotene was complete (101.6±11.7% and 102.5±10.5%). All of the lipophilic substances were distributed among the major lipoprotein classes, with the exception of retinol, which was exclusively located in VHDL (d=1.140 to 1.191 g/mL; 3%) and in the ultracentrifugal bottom residue (d>1.191 g/mL; 97%), thereby reflecting its high affinity for the retinol binding protein.


View this table:
[in this window]
[in a new window]
  		Table 2. Characteristics of Major Plasma Lipoprotein Fractions Isolated by Density-Gradient Ultracentrifugation and Distribution of Retinol, Tocopherols, and Carotenoids

Low-density lipoproteins accounted for the transport of the bulk (488 µg/dL; 45%) of lipophilic tocopherols and carotenoids, with lesser amounts in HDL (301 µg/dL; 27.8%) and VLDL (156 µg/dL; 14.4%). By contrast, VHDL and the ultracentrifugal residue accounted for only 5.1% and 7.7%, respectively, of the total plasma content of lipophilic tocopherols and carotenoids.

We next calculated the distribution of tocopherols and carotenoids among the major lipoprotein density classes in terms of the number of molecules present within a single lipoprotein particle (Table 2Up). Such calculations could not be made for VHDL (nor for the bottom fraction), given the marked structural and compositional heterogeneity of lipoprotein particles distributed over the density range from 1.140 to 1.191 g/mL.39 On this basis, a marked and progressive decrease in the particle content of tocopherols and carotenoids was seen with increase in density. This concentration gradient is well illustrated by {alpha}-tocopherol, which was present at high levels in VLDL (mean, 39 mol per particle), moderate levels in LDL (8.5 mol per particle), and in amounts representing 1 mol {alpha}-tocopherol per two particles of HDL. Indeed, not only the molecular species of tocopherols but also those of the carotenoids were present in amounts corresponding to a stoichiometry of significantly less than one molecule per HDL particle; in fact, each carotenoid component was on average present at a level of approximately one molecule for 100 HDL particles, with the exception of {alpha}-carotene and canthaxanthin, in which case 2 to 4 HDL particles per 1000 contained one such molecule.

Among the two major classes of apoB-containing lipoproteins, ie, VLDL and LDL, two distinct distribution patterns of carotenoid components were observed (Table 2Up). Thus, particle content of the hydrocarbon subgroup ({alpha}- and ß-carotene and lycopene) was similar or superior in LDL compared with VLDL particles; indeed, the {alpha}-carotene content of LDL was significantly greater than that in VLDL (P<.05). By contrast, particle content of OXCs diminished almost threefold from VLDL to LDL, approximately one OXC molecule being present in one to two VLDL particles, while the content in LDL was approximately one per four LDL particles (P<.0005).

The question as to whether the relative distribution of tocopherols and carotenoids among the major lipoprotein classes was associated with the size of the neutral lipid core (mainly TG and CE) of each lipoprotein class was then evaluated (Table 2Up). Calculation of the ratio of the total number of molecules of neutral lipids to the number of molecules of each analyte in each lipoprotein density class revealed minor but significant variations in this ratio for both {gamma}- and {alpha}-tocopherol between apoB-containing lipoproteins and particles containing apoA-I; the ratio for {alpha}-tocopherol showed a maximal 1.3-fold range, from 212:1 to 162:1 in VLDL and HDL, respectively (P<.005), while {gamma}-tocopherol content relative to CE+TG varied over a 1.5-fold range from LDL (3519:1) to HDL (2287:1) (P<.05). These findings were indicative of a strong and relatively invariant association between the size of the hydrophobic lipid core and tocopherol content; it is relevant in this context that the neutral lipid core content of VLDL, LDL, and HDL decreased markedly from 8200 to 1700 and 90 molecules per lipoprotein particle, respectively. This hypothesis was tested by a multiple linear regression analysis of the relationship between the molar content of tocopherols and that of each of the lipid components TG, CE, FC, and PL. The analysis revealed that {alpha}-tocopherol content was significantly and independently associated with both TG (P<.0005) and CE content (P<.005) in VLDL, LDL, and HDL considered together. In contrast to {alpha}-tocopherol, the content of {gamma}-tocopherol was not significantly associated with that of any of the major lipid components but rather was dependent on the structure of the lipoprotein particle itself (ANOVA, P<.0001). By contrast, no relationship between the content of carotenoid components and lipid core size and composition was detectable.

Evaluation of a potential association between the various tocopherol and carotenoid components and the molar content of the polar lipids and sterols, ie, PL and FC, failed to reveal any pronounced relationship, with the exception of the OXCs lutein/zeaxanthin and canthaxanthin, in which case the molar ratio of PL to carotenoid varied over a narrow range (1.1- to 1.4-fold) among VLDL, LDL, and HDL (data not shown). This relationship was confirmed by ANOVA, which showed that no significant differences existed between these ratios in individual lipoprotein density classes.

Distribution of Lipid-Soluble Tocopherols and Carotenoids in ApoB- and ApoA-I–Containing Lipoprotein Subspecies
The distribution of the lipid-soluble tocopherols and carotenoids among the apoB-containing and apoA-I–containing lipoprotein subspecies, defined on the basis of their hydrated density and physicochemical characteristics as described earlier39 was then determined.

The profile of the total mass of tocopherols and carotenoids among the five LDL subspecies closely paralleled that of the chemical mass of each particle subspecies in plasma (data not shown). Quantitatively, LDL3 and LDL4 subspecies predominated (25% to 45% of total LDL mass). The dense subspecies, LDL5, accounted for 16% to 20% of the transport of each tocopherol and carotenoid component among LDL subspecies (Table 3Down).


View this table:
[in this window]
[in a new window]
  		Table 3. Distribution of Lipid-Soluble Tocopherols and Carotenoids in Human LDL Subspecies

The respective molar contents of {alpha}- and {gamma}-tocopherol in the five LDL subspecies were significantly different (P<.0001 and P<.005 for {alpha}- and {gamma}-tocopherol, respectively; Fig 2Down). Furthermore, Bonferroni-Dunn analysis was applied to the mean molar content of each lipid-soluble vitamin and carotenoid per particle and revealed that the dense LDL5 subfraction (d=1.051 to 1.065 g/mL) contained significantly less {alpha}-tocopherol not only relative to the lightest subfraction (LDL1, P<.001) but also to the LDL2 (P<.001) and LDL3 subspecies (P<.0005). Thus, dense LDL particles contained approximately half of the molar content of both tocopherols compared with the light LDL1 subspecies; indeed, the extreme ranges in {alpha}-tocopherol in LDL1 compared with LDL5 were {approx}11 to 16 and {approx}5 to 8 molecules per particle, respectively. Moreover, a progressive and significant reduction (linear trend test; P<.0001) in both {alpha}- and {gamma}-tocopherol content was observed with increase in the density of LDL subspecies. Similarly, the molar contents of the OXCs diminished across the density range from light to dense LDL; more specifically, the concentrations of lutein/zeaxanthin per particle were distinct between individual LDL subspecies (P<.05) and decreased progressively with increase in density (linear trend test; P<.001). While {alpha}-carotene differed but little between LDL subspecies, both ß-carotene and lutein showed a bell-shaped profile, with a peak in LDL3 and a minimum in dense LDL. It is noteworthy that none of the molar contents of the carotenoid components attained a stoichiometry of one molecule per LDL particle subspecies, ranging from a minimum for {alpha}-carotene ({approx}1:30 particles) to a maximum for ß-carotene (1:3 particles for LDL3).





View larger version (50K):
[in this window]
[in a new window]
  		Figure 2. Distribution of liposoluble tocopherols and carotenoids in LDL subspecies separated by density-gradient ultracentrifugation. Data are expressed as the number of molecules of each analyte per LDL particle. A, {alpha}-Tocopherol ({circ}) and {gamma}-tocopherol ({bullet}); B, lycopene, {alpha}-carotene, and ß-carotene; and C, lutein/zeaxanthin, canthaxanthin ({square}), and ß-cryptoxanthin. The vertical bars for each data point represent a single-sided value for SD. *Fractions statistically different from LDL1 ({gamma}-tocopherol: LDL4 vs LDL1 P<.005, LDL5 vs LDL1 P<.0005; {alpha}-tocopherol: LDL2, LDL3, LDL4, and LDL5 vs LDL1 P<.001; lutein/zeaxanthin: LDL5 vs LDL1 P<.005). **Fractions statistically different from LDL2 ({alpha}-tocopherol: LDL3, LDL4, and LDL5 vs LDL2 P<.001). ***Fractions statistically different from LDL3 ({alpha}-tocopherol: LDL5 vs LDL3 P<.0005).

The apoA-I–containing HDL was subfractionated into four subspecies, HDL2-1, HDL2-2, HDL3-1, and HDL3-2, on the basis of their hydrated density.40 On a quantitative basis, the HDL3-1 subfraction was the most abundant, representing 35.6% of total HDL mass and accounting for 34.7% of the total amount of analytes transported by HDL (Table 4Down); by contrast, lesser amounts (17.8%, 26.1%, and 21.4%) were transported by HDL2-1, HDL2-2, and HDL3-2, respectively. We noted equally that tocopherols and OXCs were primarily distributed among HDL3 subspecies (55% to 64% of total in HDL), while HCCs were preferentially associated with HDL2 (58% to 61% of total in HDL) (Table 4Down).


View this table:
[in this window]
[in a new window]
  		Table 4. Distribution of Lipid-Soluble Tocopherols and Carotenoids in Human HDL Subspecies

The molar content of each tocopherol and carotenoid component per HDL particle varied significantly over a wide range among the respective HDL subspecies but was consistently superior in HDL2 (Fig 3Down); for example, {alpha}-tocopherol content varied over a 30-fold range from HDL2-1 ({approx}4 mol per 5 particles) to HDL3-2 ({approx}1 mol per 35 particles), while molar content of lycopene varied over a 7-fold range ({approx}1 mol in 58 HDL2 particles to {approx}1 mol per 400 of HDL3). For all components, there was a significant linear tendency (P<.001) for the molar particle content of each analyte to decrease with increase in density from HDL2-1 to HDL3-2; this tendency was equally reflected in the evolution of the mass ratio of total analyte per milligram lipoprotein mass (Table 4Up).





View larger version (50K):
[in this window]
[in a new window]
  		Figure 3. Distribution of liposoluble tocopherols and carotenoids in HDL subspecies separated by density-gradient ultracentrifugation. Data are expressed as the number of molecules of each analyte per HDL particle. A, {alpha}-Tocopherol ({circ}), {gamma}-tocopherol ({bullet}); B, lycopene, {alpha}-carotene, and ß-carotene; and C, lutein/zeaxanthin, canthaxanthin ({square}), and ß-cryptoxanthin. The vertical bars for each data point represent a single-sided value for SD. *Fractions statistically different from HDL2-1 ({gamma}-tocopherol: HDL3-1 and HDL3-2 vs HDL2-1 P<.0001; {alpha}-tocopherol: HDL3-1 and HDL3-2 vs HDL2-1 P<.0001; lycopene: HDL2-2, HDL3-1, and HDL3-2 vs HDL2-1 P<.001; {alpha}-carotene: HDL2-2, HDL3-1, and HDL3-2 vs HDL2-1 P<.001; ß-carotene: HDL2-2, HDL3-1, and HDL3-2 vs HDL2-1 P<.001; canthaxanthin: HDL3-1 and HDL3-2 vs HDL2-1 P<.005; and ß-cryptoxanthin: HDL3-1 and HDL3-2 vs HDL2-1 P<.0001). **Fractions statistically different from HDL2-2 ({gamma}-tocopherol: HDL3-1 and HDL3-2 vs HDL2-2 P<.0001; {alpha}-tocopherol: HDL3-1 and HDL3-2 vs HDL2-2 P<.0001; lycopene: HDL3-2 vs HDL2-2 P<.01; lutein/zeaxanthin: HDL3-1 and HDL3-2 vs HDL2-2 P<.005; canthaxanthin: HDL3-1 and HDL3-2 vs HDL2-2 P<.0005; and ß-cryptoxanthin: HDL3-1 and HDL3-2 vs HDL2-2 P<.0001).


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
The relationship between antioxidant status and cardiovascular disease remains a subject of debate.52 53 The cardioprotective action of lipophilic antioxidants appears to be mediated in part by their capacity to increase the oxidative resistance of LDL but equally by their direct effects on cellular activities.36 37 38 52 54 55 As the plasma transport form of lipophilic antioxidants is directly related to their potential antiatherogenic effects, and as data on the respective contributions of apoB- and of apoA-I–containing lipoprotein particles to antioxidant transport are fragmentary, we describe the qualitative and quantitative distribution of tocopherols and of the polar and apolar carotenoids among the principal particle subspecies of LDL and HDL in normolipidemic plasma. Our analyses were performed with a highly sensitive, resolutive, and reproducible HPLC method characterized by maximal recoveries and rigorous standardization; losses of antioxidants due to artifactual oxidation were restricted to a minimum both during lipoprotein isolation and analyte extraction. Our data reveal that small, dense subspecies of LDL (LDL5, d=1.051 to 1.065 g/mL) contain diminished amounts of {alpha}- and {gamma}-tocopherol, ß-carotene, lycopene, ß-cryptoxanthin, and lutein/zeaxanthin compared with particle subspecies of light (LDL1 and LDL2, d=1.018 to 1.030 g/mL) and intermediate density (LDL3, d=1.030 to 1.040 g/mL) LDL. Furthermore, small, dense subspecies of HDL3 (HDL3-1 and HDL3-2, d=1.098 to 1.141 g/mL) were significantly depleted not only in tocopherols but also in all HCC and OXC components. These findings have major implications not only for the behavior of these particles under oxidative stress but also for the tissue targeting and molecular actions of both tocopherols and carotenoids at the cellular level.

Comparison of the percent distribution of tocopherols and carotenoids between VLDL, LDL, HDL, and the ultracentrifugal residue (d>1.19 g/mL) (Table 2Up) with fragmentary data in the literature revealed an overall consistency,38 50 56 57 58 despite major losses that may have occurred during sample preparation in some earlier studies. The distinct distribution patterns of HCCs versus their OXC components are, however, noteworthy, the former showing selectivity for LDL (76% of total lipoprotein-associated HCCs), while the latter were equally distributed between LDL and HDL (44% and 39% of total, respectively). The cosegregation of lutein and ß-cryptoxanthin in LDL and HDL contrasts with the differential distribution observed earlier.50 Furthermore, and again in contrast to the latter study, we detected significant amounts of all tocopherol and carotenoid components both in VHDL and in the d>1.19 g/mL bottom fraction (Table 2Up).

When data were expressed on a per-particle basis, the progressive decrease in {alpha}-tocopherol content with increase in density and decrease in particle size from VLDL to LDL to HDL reported by Romanchik et al50 was confirmed (Table 2Up). Nonetheless, quantitative differences were evident, the latter authors observing mean values of 145, 12, and 0.7 mol of {alpha}-tocopherol per particle in VLDL, LDL, and HDL, respectively, compared with 39, 8.5, and 0.5 mol per particle in the present study. The marked discrepancy concerning VLDL is due to the application by these investigators50 of a value for molecular weight some 12-fold greater than that applied for normolipidemic VLDL in the present study ({approx}12.5x107 versus 10.2x106, respectively); the latter value was obtained by computation from the mean particle size (330 Å) and the composition determined earlier by us.39 The present values for {alpha}-tocopherol content in LDL and HDL are entirely comparable to those reported by others (8.5 mol/mol LDL3 20 21 ). Only in VLDL did {gamma}-tocopherol attain a level superior to 1 mol per particle; levels in LDL and HDL corresponded to 1 mol per 2 mol LDL and 1 mol per 23 mol HDL, respectively. In a similar manner to the tocopherols, particle contents of both the HCCs and their OXC counterparts diminished progressively, with increase in density and decrease in particle size. In terms of total particle carotenoid content, our findings resemble those of others for the total LDL fraction ({approx}3 mol carotenoid per 4 particles3 50 59 and for HDL ({approx}1 mol per 45 particles).50

Examination of the potential relationships between particle content of tocopherols and carotenoids and lipoprotein structure and composition revealed that only {alpha}-tocopherol content was significantly and independently associated with neutral lipid core size in all lipoprotein classes. Indeed, {approx}200 mol neutral lipid are associated with solubilization of 1 mol {alpha}-tocopherol in VLDL, LDL, and HDL. These data suggest that the incorporation of {alpha}-tocopherol in VLDL, LDL, and HDL is primarily influenced by neutral lipid core size. By contrast, incorporation of OXCs appears principally to reflect polar PL content, which would be consistent with its preferential localization to the polar surface monolayer of these particles.

The present finding that the content of individual carotenoid components in LDL and HDL was insufficient for a single particle to contain one molecule of each immediately implied heterogeneity in their distribution between subpopulations of both lipoproteins. Analyses of the tocopherol and OXC and HCC components in five defined LDL subfractions isolated by isopycnic density–gradient ultracentrifugation revealed distinct distribution profiles when expressed on a molar basis (Fig 2Up). For the purposes of this discussion, we shall consider LDL1 and LDL2 as light LDL (d=1.018 to 1.030 g/mL), LDL3 as intermediate LDL (d=1.030 to 1.040 g/mL), and LDL4 and LDL5 as dense LDL (d=1.040 to 1.065 g/mL). Particle content of {alpha}- and {gamma}-tocopherol diminished significantly with increase in density and decrease in particle size from light to dense LDL.

To date, contradictory findings have been reported regarding the tocopherol content of LDL subfractions; such contradictions appear to arise in part from expression of the data relative to lipid or protein content rather than on a mole per particle basis and equally from the use of contrasting definitions for the density ranges of light and dense LDL.29 30 31 32 (It is noteworthy that the dense LDL fraction studied by de Graaf et al29 was of d=1.040 to 1.045 g/mL, that studied by Tribble et al30 of d=1.046 g/mL, and that studied by Tribble et al32 of d=1.040 to 1.054 g/mL.) Although quite distinct in absolute terms, our present results are not, however, inconsistent with those of Tribble et al,32 who found that the {alpha}-tocopherol content of dense (d=1.040 to 1.054 g/mL) LDL was less than that of the buoyant (d=1.026 to 1.032 g/mL) subfraction (6.8 versus. 2.4 mol/mol LDL in buoyant and dense LDL, respectively32 ). The bulk of {alpha}- and {gamma}-tocopherol transport in the total LDL fraction was accounted for by the most abundant subfractions in the normolipidemic plasmas studied, ie, LDL3 and LDL4 ({approx}25% and 45%, respectively); this was equally the case for both HCCs and OXCs ({approx}26% and {approx}46%, respectively). It is, however, noteworthy that dense LDL (LDL4 and LDL5) predominated in the 11 normolipidemic subjects studied, in contrast to our earlier reports, in which intermediate LDL (LDL3) was consistently the most abundant subspecies.31 51 60 The LDL particle content of ß-carotene and lycopene was distinguished by a bell-shaped distribution as a function of density, the maximum value residing in intermediate LDL (LDL3) (Fig 2BUp). Dense LDL typically displayed the lowest particle content of both ß-carotene and lycopene among LDL subspecies, differences between LDL3 and LDL5 ranging from {approx}35% to 40%; marked variation occurred between subjects. Like the tocopherols, values for the OXC content of LDL subfractions (Fig 2CUp) decreased progressively from light to dense LDL. Among these components, the particle content of lutein/zeaxanthin was significantly lower in dense LDL (P<.05).

The above findings are relevant to both the plasma transport and tissue fate of tocopherols and carotenoids and to the oxidative resistance and atherogenic potential of LDL. Thus, on the one hand, intermediate LDL (LDL3) possesses optimal binding affinity for the cellular LDL receptor, thereby assuring efficient uptake of a major fraction of tocopherols and carotenoids by target tissues.60 On the other hand, the diminished content not only of tocopherols but also of carotenoids in atherogenic dense LDL (LDL5, d=1.051 to 1.065 g/mL) could constitute a major factor in their reduced resistance to oxidative stress.29 31 32 In this context, the markedly reduced amounts of lutein/zeaxanthin in dense LDL could be of special impact on its oxidative behavior; indeed, Thurnham37 has shown that lutein is >10 times as potent on a molar basis as {alpha}-tocopherol in prolonging the lag phase of LDL during copper-mediated oxidation. These data are of special relevance to the free radical and oxygen environment of both normal and atherosclerotic arterial wall, as the low O2 tensions prevalent in mammalian tissues (<20 torr) favor the radical-trapping, antioxidant activities of both OXCs and HCCs.38 Moreover, not only are carotenoids efficient in directly decreasing lipid peroxidation, but they are also able to modulate endogenous levels of other antioxidants, at least in biological membranes and homogeneous lipid solutions.38 Indeed, synergistic protective action has been observed between some carotenoids and {alpha}-tocopherols in such experimental systems.38 On the basis of present evidence, then, the diminished carotenoid content of dense LDL could contribute significantly to the reduced oxidative resistance of these particles, and especially at low tissue O2 tension, an effect that could be exacerbated by the absence of their sparing action on {alpha}-tocopherol. Such effects may underlie the elevated cardiovascular risk associated with increased plasma levels of dense LDL.27 Furthermore, individuals displaying an atherogenic LDL profile dominated by dense LDL may deliver antioxidants less efficiently to cells, given the low binding affinity of such particles for the cellular LDL receptor.61 In such subjects, the protective effects of antioxidants on cell-mediated oxidation of LDL54 55 may be markedly diminished. Finally, in terms of the heterogeneity of the chemical composition of LDL subspecies, it should be emphasized that only a small proportion of LDL particles will contain carotenoids. For example, in intermediate LDL, 1 in 3 particles will contain 1 mol ß-carotene, 1 in 5 lycopene, 1 in 10 lutein and ß-cryptoxanthin, and 1 in 20 {alpha}-carotene and canthaxanthin. The question as to the frequency with which such components might occur within the same LDL particle will be the subject of further studies, which may provide insight into the question of possible associations in the transport and metabolism of these substances, especially as their rates of exchange and transfer between plasma lipoproteins appear to be particularly slow.50

Although HDL particles exert protective action on the copper-mediated oxidation of LDL,34 HDL lipids nonetheless readily undergo oxidative modification.17 18 21 The oxidative behavior of HDL particles depends, at least in part, on their antioxidant content. Absolute particle contents of antioxidants were uniformly maximal in the lightest, largest particle subspecies (HDL2-1) and decreased markedly across the density spectrum to attain minima in the densest HDL3 subclass (HDL3-2) (Fig 3Up). The highest concentration gradients were seen for tocopherols and HCCs (ß-carotene and lycopene), which diminished fivefold and eightfold, respectively, across the density range from HDL2-1 to HDL3-2. By contrast, OXC content decreased twofold to threefold over this same range. Only {alpha}- and {gamma}-tocopherol attained levels approaching one molecule per HDL particle in the lightest subfraction, HDL2-1. HDL particle contents of the carotenoids were substantially lower in this same subfraction, representing 1 mol ß-carotene in 40 HDL2-1 particles, 1 mol lycopene, lutein, and ß-cryptoxanthin in {approx}70 particles, and {approx}1 mol canthaxanthin and {alpha}-carotene in 200 HDL2-1 particles. These analytical findings are consistent with the observation of Bowry et al21 that no antioxidant-dependent lag period was seen during in vitro oxidation of HDL and also extend the findings of Brown and Fragoso35 that heterogeneity in tocopherol and carotenoid content exists in HDL subfractions enriched or depleted in apoE. The relevance of these data to the antiatherogenic action of HDL particle subspecies at the cellular level and to their role in the plasma transport and clearance of lipid hydroperoxides remains indeterminate, however.21

In conclusion, the present data provoke the working hypothesis that the diminished contents of specific OXC and HCC components in dense LDL may be implicated in its proatherogenic action in the arterial wall and lead us to speculate that the enrichment of dense LDL in key carotenoid components with elevated antioxidant activity,37 such as lutein, may represent a new and novel therapeutic approach in coronary heart disease patients.


*    Selected Abbreviations and Acronyms
 
apo = apolipoprotein
CE = cholesteryl ester
CV = coefficient of variation
FC = free cholesterol
HCC = hydrocarbon carotenoid
HPLC = high-performance liquid chromatography
NIST = National Institute of Standards and Technology
OXC = oxygenated carotenoid
PL = phospholipid
TG = triglyceride


*    Acknowledgments
 
These studies were supported by Lesieur Alimentaire (Contrat de Valorisation INSERM-Industrie No. 91017), by the Ministère de la Recherche et de la Technologie (MRT No. 91.C.0941), by INSERM, and by the Faculté de Medicine Pitié-Salpétrière. Lutein, zeaxanthin, ß-cryptoxanthin, and canthaxanthin were the kind gift of Hoffmann La Roche, Basel, Switzerland. We are indebted to Drs E. Bruckert and P. Giral for critical advice in statistical analysis and stimulating discussions; it is a pleasure to acknowledge the continued support and interest of P. Boulos and Drs P. Serog and F. Zwobada. The typescript was kindly prepared by V. Soulier.

Received April 25, 1996; accepted July 22, 1996.


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

  1. Steinberg S, Parthasarathy D, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low density lipoproteins that increase its atherogenicity. N Engl J Med. 1989;4:915-924.
  2. Yla Hertualla S, Palinski W, Rosenfeld ME, Parthasarathy S, Carew TE, Butler S, Witztum JL, Steinberg D. Evidence for the presence of oxidatively modified low density lipoprotein in atherosclerotic lesions of rabbit and man. J Clin Invest. 1989;84:1086-1095.
  3. 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]
  4. Brooks CJW, Harland WA, Steel G. Squalene, 26-hydroxycholesterol and 7-ketocholesterol in human atheromatous plaques. Biochim Biophys Acta. 1966;125:620-622.[Medline] [Order article via Infotrieve]
  5. Carpenter KLH, Taylor SE, Ballantine JA, Fussel B, Halliwell B, Mitchinson MJ. Lipids and oxidised lipids in human atheroma and normal aorta. Biochim Biophys Acta. 1993;1167:121-130.[Medline] [Order article via Infotrieve]
  6. Suarna C, Dean RT, May J, Stocker R. Human atherosclerotic plaque contains both oxidized lipids and relatively large amounts of {alpha}-tocopherol and ascorbate. Arterioscler Thromb Vasc Biol. 1995;15:1616-1621.[Abstract/Free Full Text]
  7. Davies MJ, Richardson PD, Woolf N, Katz DR, Mann J. Risk of thrombosis in human atherosclerotic plaques: role of extracellular lipid, macrophage and smooth muscle content. Br Heart J. 1993;69:377-381.[Abstract/Free Full Text]
  8. Schwenke DC, Carew TE. Initiation of atherosclerotic lesion in cholesterol-fed rabbits, II: selective retention of LDL vs selective increase in LDL permeability in susceptible sites of arteries. Arteriosclerosis. 1989;9:908-918.[Abstract/Free Full Text]
  9. Camejo G, Hurt-Camejo E, Olsson U, Bondjers G. Proteoglycans and lipoproteins in atherosclerosis. Curr Opin Lipidol. 1993;4:385-391.
  10. Smith CA, Mitchinson MJ, Aruoma OI, Halliwell B. Stimulation of lipid peroxidation and hydroxyl radical generation by the contents of human atherosclerotic lesions. Biochem J. 1992;286:901-905.
  11. Halliwell B. Mechanisms of low-density lipoprotein oxidation. Curr Opin Lipidol. 1993;4:382-384.
  12. Frei B, Gaziano JM. Content of antioxidants, preformed lipid hydroperoxides, and cholesterol as predictors of the susceptibility of human LDL to metal ion-dependent and ion-independent oxidation. J Lipid Res. 1993;34:2135-2145.[Abstract]
  13. Sattler W, Mohr D, Stocker R. Rapid isolation of lipoproteins and assessment of their peroxidation by high-performance liquid chromatography with postcolumn chemiluminescence. Methods Enzymol. 1994;233:469-489.[Medline] [Order article via Infotrieve]
  14. Dzeletovic S, Babiker A, Lund E, Diczfalusy U. Time course of oxysterol formation during in vitro oxidation of low-density lipoprotein. Chem Phys Lipids. 1995;78:119-128.[Medline] [Order article via Infotrieve]
  15. Esterbauer H, Jurgens G, Quehenberger O, Koller E. Autooxidation 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]
  16. Steinbrecher UP. Oxidation of human low-density lipoprotein results in derivatization of lysine residues of apolipoprotein B by lipid peroxide decomposition products. J Biol Chem. 1987;262:3603-3608.[Abstract/Free Full Text]
  17. Parthasarathy S, Barnett J, Fong LG. High-density lipoproteins inhibit the oxidative modification of low-density lipoproteins. Biochim Biophys Acta. 1990;1044:275-283.[Medline] [Order article via Infotrieve]
  18. Parthasarathy S, Santanam N. Mechanisms of oxidation, antioxidants, and atherosclerosis. Curr Opin Lipidol. 1994;5:371-375.[Medline] [Order article via Infotrieve]
  19. Jialal I, Grundy SM. Preservation of the endogenous antioxidants in low-density lipoproteins by ascorbate but not probucol during oxidative modification. J Clin Invest. 1991;87:597-601.
  20. Bowry VW, Ingold KU, Stocker R. Vitamin-E in human low-density lipoprotein: when and how this antioxidant becomes a pro-oxidant. Biochem J. 1992:288:341-344.
  21. Bowry VW, Stanley KK, Stocker R. High density lipoprotein is the major carrier of lipid hydroperoxides in human blood plasma from fasting donors. Proc Natl Acad Sci U S A. 1992;89:10316-10320.[Abstract/Free Full Text]
  22. Thomas MJ, Thornburg T, Manning J, Hooper K, Rudel LL. Fatty acid composition of low-density lipoprotein influences its susceptibility to autoxidation. Biochemistry. 1994;33:1828-1834.[Medline] [Order article via Infotrieve]
  23. Kontush A, Hubner C, Finckh B, Kohlschutter A, Beisiegel U. Low density lipoprotein oxidizability by copper correlates to its initial ubiquinol-10 and polyunsaturated fatty acid content. FEBS Lett. 1994;341:69-73.[Medline] [Order article via Infotrieve]
  24. Gieseg SP, Esterbauer H. Low density lipoprotein is saturable by pro-oxidant copper. FEBS Lett. 1994;343:188-194.[Medline] [Order article via Infotrieve]
  25. Schuster B, Prassl R, Nigon F, Chapman MJ, Laggner P. Core lipid structure is a major determinant of the oxidative resistance of low density lipoprotein. Proc Natl Acad Sci U S A. 1995;92:2509-2513.[Abstract/Free Full Text]
  26. Tribble DL, Krauss RM, Lansberg MG, Thiel PM, van den Berg JJM. Greater oxidative susceptibility of the surface monolayer in small dense LDL may contribute to differences in copper-induced oxidation among LDL density subfractions. J Lipid Res. 1995;36:662-671.[Abstract]
  27. Krauss RM. Heterogeneity of plasma low-density lipoproteins and atherosclerosis risk. Curr Opin Lipidol. 1994;5:339-349.[Medline] [Order article via Infotrieve]
  28. Assmann G, Von Eckardstein A, Funke H. High density lipoproteins, reverse cholesterol transport of cholesterol, and coronary artery disease. Circulation. 1993;87:28-34.
  29. de Graaf J, Hak-Lemmers HLM, Hectors MPC, Demacker PNM, Hendricks JCM, Stalenhoef AFH. Enhanced susceptibility to in vitro oxidation of the dense low density lipoprotein subfraction in healthy subjects. Arteriosclerosis. 1991;11:298-306.[Abstract/Free Full Text]
  30. Tribble DL, Holl LG, Wood PD, Krauss RM. Variations in oxidative susceptibility among six low density lipoprotein subfractions of differing density and particle size. Atherosclerosis. 1992;93:189-199.[Medline] [Order article via Infotrieve]
  31. Dejager S, Bruckert E, Chapman MJ. Dense low-density lipoprotein subspecies with diminished oxidative resistance predominate in combined hyperlipidemia. J Lipid Res. 1993;34:295-308.[Abstract]
  32. Tribble DL, van den Berg JJM, Motchnik PA, Ames BN, Lewis DM, Chait A, Krauss RM. Oxidative susceptibility of low density lipoprotein subfractions is related to their ubiquinol-10 and alpha-tocopherol content. Proc Natl Acad Sci U S A. 1994;91:1183-1187.[Abstract/Free Full Text]
  33. Mackness MI, Arrol S, Durrington PN. Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein. FEBS Lett. 1991;286:152-154.[Medline] [Order article via Infotrieve]
  34. Kunitake SJ, Jarvis MR, Hamilton RL, Kane JP. Binding of transition metals by apolipoprotein AI–containing plasma lipoproteins: inhibition of oxidation of low density lipoproteins. Proc Natl Acad Sci U S A. 1992;89:6993-6997.[Abstract/Free Full Text]
  35. Brown AJ, Fragoso YD. Tocopherols and carotenes are differently distributed in subfractions of high-density lipoprotein. Biochim Biophys Acta. 1994;1210:373-376.[Medline] [Order article via Infotrieve]
  36. Liebler DC. Carotenoids in human health: antioxidant reactions of carotenoids. Ann N Y Acad Sci. 1993;691:20-31.[Medline] [Order article via Infotrieve]
  37. Thurnham DI. Carotenoids: functions and fallacies. Proc Nutr Soc. 1994;53:77-87.[Medline] [Order article via Infotrieve]
  38. Krinsky NI. Actions of carotenoids in biological systems. Annu Rev Nutr. 1993;13:561-587.[Medline] [Order article via Infotrieve]
  39. Chapman MJ, Goldstein S, Lagrange D, Laplaud MP. A density gradient ultracentrifugal procedure for the isolation of the major lipoprotein classes from human serum. J Lipid Res. 1981;22:339-358.[Abstract]
  40. Goulinet S, Chapman MJ. Plasma lipoproteins in the Golden Syrian hamster (Mesocricetus auratus): heterogeneity of apoB- and apoAI-containing particles. J Lipid Res. 1993;34:943-959.[Abstract]
  41. Kaplan LA, Miller JA, Stein EA. Simultaneous measurement of serum retinol, tocopherols, carotenes, and carotenoids by high performance liquid chromatography. J Clin Lab Anal. 1987;1:147-152.
  42. Epler KS, Ziegler RG, Craft NE. Liquid chromatographic method for the determination of carotenoids, retinoids and tocopherols in human serum and in food. J Chromatogr. 1993;619:37-48.[Medline] [Order article via Infotrieve]
  43. Bui MH. Simple determination of retinol, alpha-tocopherol and carotenoids (lutein, all-trans-lycopene, alpha-carotene and beta-carotene) in human plasma by isocratic liquid chromatography. J Chromatogr B Biomed Appl. 1994;654:129-133.[Medline] [Order article via Infotrieve]
  44. Epler KS, Sander LC, Ziegler RG, Wise SA, Craft NE. Evaluation of reversed-phase liquid chromatographic columns for recovery and selectivity of selected carotenoids. J Chromatogr. 1992;595:89-101.[Medline] [Order article via Infotrieve]
  45. Elinder LS, Walldius G. Simultaneous measurement of serum probucol and lipid-soluble antioxidants. J Lipid Res. 1992;33:131-137.[Abstract]
  46. Stacewicz-Sapuntzakis M, Bower PE, Kinkendall JW, Burgess M. Simultaneous determination of serum retinol and various carotenoids: their distribution in middle-aged men and women. J Micronutr Anal. 1987;3:27-45.
  47. Zaman Z, Fielden P, Frost PG. Simultaneous determination of vitamin-A and vitamin-E and carotenoids in plasma by reversed-phase HPLC in elderly and younger subjects. Clin Chem. 1993;39:2229-2234.[Abstract]
  48. Rougereau A, Person O, Rougereau G. Retinol, ß-carotene, and {alpha}-carotene status in a French population of healthy subjects. Int J Vitam Nutr Res. 1987;57:31-35.[Medline] [Order article via Infotrieve]
  49. Piironen V, Varo P, Syvajoa EL, Salminen K, Koivistanen PHA. High performance liquid chromatographic determination of tocopherols and tocotrienols and its application to diets and plasma in Finnish men, II: application. Int J Vitam Nutr Res. 1984;54:41-46.[Medline] [Order article via Infotrieve]
  50. Romanchik JE, Morel DW, Harrison EH.