Advertisement

Lipid Peroxidation Induces Cholesterol Domain Formation in Model Membranes*

  • Robert F. Jacob
    Correspondence
    To whom correspondence should be addressed: Elucida Research, 100 Cummings Ctr., Ste. 135L, P. O. Box 7100, Beverly, MA 01915. Tel.: 978-921-4194; Fax: 978-921-4195;
    Affiliations
    Elucida Research, Beverly, Massachusetts 01915-0091
    Search for articles by this author
  • R. Preston Mason
    Affiliations
    Elucida Research, Beverly, Massachusetts 01915-0091

    Cardiovascular Division, Department of Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts 02115
    Search for articles by this author
  • Author Footnotes
    * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:September 28, 2005DOI:https://doi.org/10.1074/jbc.M507587200
      Numerous reports have established that lipid peroxidation contributes to cell injury by altering the basic physical properties and structural organization of membrane components. Oxidative modification of polyunsaturated phospholipids has been shown, in particular, to alter the intermolecular packing, thermodynamic, and phase parameters of the membrane bilayer. In this study, the effects of oxidative stress on membrane phospholipid and sterol organization were measured using small angle x-ray diffraction approaches. Model membranes enriched in dilinoleoylphosphatidylcholine were prepared at various concentrations of cholesterol and subjected to lipid peroxidation at physiologic conditions. At cholesterol-to-phospholipid mole ratios (C/P) as low as 0.4, lipid peroxidation induced the formation of discrete, membrane-restricted cholesterol domains having a unit cell periodicity or d-space value of 34 Å. The formation of cholesterol domains correlated directly with lipid hydroperoxide levels and was inhibited by treatment with vitamin E. In the absence of oxidative stress, similar cholesterol domains were observed only at C/P ratios of 1.0 or higher. In addition to changes in sterol organization, lipid peroxidation also caused reproducible changes in overall membrane structure, including a 10 Å reduction in the width of the surrounding, sterol-poor membrane bilayer. These data provided direct evidence that lipid peroxidation alters the essential organization and structure of membrane lipids in a manner that may contribute to changes in membrane function during aging and oxidative stress-related disorders.
      Lipid peroxidation is a degenerative process that affects unsaturated membrane lipids under conditions of oxidative stress (
      • Girotti A.W.
      ). This complex process is believed to contribute to human aging and disease by disrupting the structural conformation, the packing of lipid components and, ultimately, the function of biological membranes. The polyunsaturated fatty acids of membrane phospholipids are particularly susceptible to peroxidation and undergo significant modifications, including the rearrangement or loss of double bonds and, in some cases, the reductive degradation of lipid acyl side chains (
      • Leibowitz M.E.
      • Johnson M.C.
      ,
      • Gardner H.W.
      ,
      • Buege J.A.
      • Aust S.D.
      ). Lipid hydroperoxides, prominent intermediates of peroxidative reactions, also accumulate in the bilayer and further contribute to changes in the structural organization and packing of membrane lipid components (
      • Girotti A.W.
      ). Many of the biophysical consequences of these structural modifications have been well characterized and include changes in membrane fluidity (
      • Petrescu A.D.
      • Gallegos A.M.
      • Okamura Y.
      • Strauss III, J.F.
      • Schroeder F.
      ,
      • Chatterjee S.N.
      • Agarwal S.
      ,
      • Borchman D.
      • Lamba O.P.
      • Salmassi S.
      • Lou M.
      • Yappert M.C.
      ), increased membrane permeability (
      • Chatterjee S.N.
      • Agarwal S.
      ,
      • Goldstein R.M.
      • Weissmann G.
      ,
      • Mandal T.K.
      • Chatterjee S.N.
      ,
      • Kunimoto M.
      • Inoue K.
      • Nojima S.
      ), alteration of membrane thermotropic phase properties (
      • Chia L.S.
      • Thompson J.E.
      • Moscarello M.A.
      ,
      • van Duijn G.
      • Verkleij A.J.
      • de Kruijff B.
      ,
      • Verma S.P.
      ), and changes in membrane protein activity (
      • Yukawa O.
      • Nagatsuka S.
      • Nakazawa T.
      ,
      • Dinis T.C.
      • Almeida L.M.
      • Madeira V.M.
      ,
      • Goel R.
      • Mishra O.P.
      • Razdan B.
      • Delivoria-Papadopoulos M.
      ,
      • Kourie J.I.
      ,
      • Mattson M.P.
      ,
      • Mattson M.P.
      • Pedersen W.A.
      • Duan W.
      • Culmsee C.
      • Camandola S.
      ,
      • Sevanian A.
      • Ursini F.
      ).
      Recent studies conducted in our laboratory have provided direct evidence for changes in the molecular organization of membrane lipid bilayers following exposure to oxidative stress. Using small angle x-ray diffraction approaches, we examined model membranes composed of dilinoleoylphosphatidylcholine (DLPC)
      The abbreviations used are:
      DLPC
      dilinoleoylphosphatidylcholine (1,2-dilinoleoyl-3-sn-phosphatidylcholine)
      POPC
      palmitoylphosphatidylcholine (1-palmitoyl,2-oleoyl-3-sn-phosphatidylcholine)
      C/P
      cholesterol-to-phospholipid
      2The abbreviations used are:DLPC
      dilinoleoylphosphatidylcholine (1,2-dilinoleoyl-3-sn-phosphatidylcholine)
      POPC
      palmitoylphosphatidylcholine (1-palmitoyl,2-oleoyl-3-sn-phosphatidylcholine)
      C/P
      cholesterol-to-phospholipid
      , before and after exposure to oxidative stress (
      • Mason R.P.
      • Walter M.F.
      • Mason P.E.
      ). Moderate levels of lipid peroxidation produced significant and reproducible alterations in the basic structure of the membrane phospholipid bilayer, including a marked reduction in membrane d-space (i.e. bilayer width plus surface hydration) and a decrease in intrabilayer headgroup separation. In addition, lipid peroxidation promoted an increase in the molecular volume associated with the unsaturated region of the hydrocarbon core and induced the apparent interdigitation of the phospholipid acyl chain terminal methyl segments (
      • Mason R.P.
      • Walter M.F.
      • Mason P.E.
      ). Similar changes in membrane bilayer structure have also been observed in membranes reconstituted from cardiac phospholipids (
      • Mason R.P.
      • Walter M.F.
      • Mason P.E.
      ) and in DLPC-enriched model membranes exposed to autoxidative forms of lipid peroxidation (
      • Mason R.P.
      • Jacob R.F.
      ).
      In this report, we extended this line of investigation by measuring the effects of lipid peroxidation on cholesterol organization and distribution in model membrane bilayers. Perturbation of membrane cholesterol content and structural organization has been implicated in the pathogenesis of a number of human diseases, many of which are also linked to conditions of increased oxidative stress. Numerous studies have shown that cholesterol directly modulates the physical properties of lipid bilayers, altering membrane responses to degenerative processes, including lipid peroxidation (
      • McLean L.R.
      • Hagaman K.A.
      ,
      • Leonard A.
      • Dufourc E.J.
      ,
      • Mowri H.
      • Nojima S.
      • Inoue K.
      ); however, little is known about the consequences of such processes on the membrane properties of cholesterol itself. In this study, small angle x-ray diffraction approaches were used to directly characterize the distribution of cholesterol in DLPC-enriched membranes exposed to various levels of peroxidative stress. The results of these experiments demonstrated that lipid peroxidation of membranes containing low levels of cholesterol (cholesterol-to-phospholipid mole ratios ranging between 0.4 and 0.8) promotes the formation of immiscible, membrane-restricted cholesterol domains. The formation of cholesterol domains was highly dependent on the extent of lipid peroxidation and was significantly inhibited by the addition of vitamin E. These data provided direct evidence for membrane lipid reorganization following oxidative insult and further implicated lipid peroxidation as an important contributor to mechanisms of aging and disease.

      EXPERIMENTAL PROCEDURES

      Materials—DLPC and 1-palmitoyl,2-oleoyl-3-sn-phosphatidylcholine (POPC) were obtained from Avanti Polar Lipids (Alabaster, AL) and stored in chloroform (25 mg/ml) at -80 °C until use. Cholesterol was also purchased from Avanti Polar Lipids and stored in chloroform (10 mg/ml) at -20 °C. CHOD-iodide color reagent (stock) was prepared according to a procedure modified from El-Saadani et al. (
      • El-Saadani M.
      • Esterbauer H.
      • El-Sayed M.
      • Goher M.
      • Nassar A.Y.
      • Jurgens G.
      ) and consisted of 0.2 m K2HPO4, 0.12 m KI, 0.15 mm NaN3, 10 μm ammonium molybdate, and 0.1 g/liter benzalkonium chloride. Prior to experimental use, the CHOD reagent was activated by adding 24 μm EDTA, 20 μm butylated hydroxytoluene, and 0.2% Triton X-100. Vitamin E ((±)-α-tocopherol; Sigma) was prepared in ethanol just prior to experimental use and added together with component lipids during membrane sample preparation.
      Preparation of Membrane Multilayer Samples—Membrane samples consisting of DLPC ± cholesterol, with cholesterol-to-phospholipid (C/P) mole ratios ranging from 0 to 1.0, were prepared as follows. Component lipids (in chloroform) were transferred to 13 × 100 mm test tubes and shell-dried under a steady stream of nitrogen gas while vortex mixing. Residual solvent was removed by drying for a minimum of 3 h under vacuum. After desiccation, each membrane sample was resuspended in diffraction buffer (0.5 mm HEPES, 154 mm NaCl, pH 7.3) to yield a final phospholipid concentration of 1.0 mg/ml. Multilamellar vesicles were then formed by vortex mixing for 3 min at ambient temperature (
      • Bangham A.D.
      • Standish M.M.
      • Watkins J.C.
      ). Membrane vesicles composed of POPC and cholesterol were also prepared using this protocol. Immediately after initial multilamellar vesicle preparation, aliquots of each membrane sample were taken for baseline (0 h) peroxidation and diffraction analyses, as described below.
      Lipid Peroxidation Analysis—All lipid membrane samples were subjected to time-dependent autoxidation by incubating at 37 °C in an uncovered, shaking water bath. Small aliquots of each sample were removed at 24-h intervals and combined with 1.0 ml of active CHOD-iodide color reagent. To ensure spectrophotometric readings within the optimum absorbance range, sample volumes taken for measurement of lipid peroxide formation were adjusted for length of peroxidation and ranged between 100 and 10 μl. Test samples were immediately covered with foil and incubated at room temperature for >4 h in the absence of light. Absorbances were then measured against a CHOD blank at 365 nm using a Beckman DU-640 spectrophotometer.
      The CHOD colorimetric assay is based on the oxidation of iodide (I-) by lipid hydroperoxides (LOOH) and proceeds according to the following reaction.
      LOOH+2H++3ILOH+H2O+I3REACTION1


      The quantity of triiodide anion (I3) liberated in this reaction is directly proportional to the amount of lipid hydroperoxides present in the membrane sample. The molar absorptivity value (ϵ) of I3 is 2.46 × 104m-1 cm-1 at 365 nm (
      • El-Saadani M.
      • Esterbauer H.
      • El-Sayed M.
      • Goher M.
      • Nassar A.Y.
      • Jurgens G.
      ).
      Preparation of Samples for X-ray Diffraction Analysis—Membrane samples were oriented for x-ray diffraction analysis as described previously (
      • Herbette L.G.
      • DeFoor P.
      • Fleischer S.
      • Pascolini D.
      • Blaise J.K.
      ). Briefly, aliquots containing 250 μg of multilamellar vesicles (based on phospholipid) were transferred to custom-designed Lucite sedimentation cells, each containing an aluminum foil substrate upon which to collect a single membrane pellet. Samples were then loaded into a Sorvall AH-629 swinging bucket ultracentrifuge rotor (DuPont) and centrifuged at 35,000 × g, 5 °C. Samples that were taken prior to initiation of peroxidation (baseline samples) were centrifuged for 90 min; samples that had undergone peroxidation beyond 24 h required overnight (∼16-18 h) centrifugation. Following orientation, the supernatants were aspirated, and the aluminum foil substrates, supporting the membrane pellets, were removed from the sedimentation cells and mounted onto curved glass slides. The samples were then placed in hermetically sealed brass canisters in which temperature and relative humidity were controlled during x-ray diffraction experiments. Potassium tartrate (K2C4H4O6 · ½H2O) was used to establish a relative humidity level of 74% in these experiments, and samples were incubated at this relative humidity at least 1 h prior to experimental analysis.
      Small Angle X-ray Diffraction Analysis—The oriented membrane fraction samples were aligned at grazing incidence with respect to a collimated, monochromatic x-ray beam produced by a Rigaku Rotaflex RU-200 high brilliance microfocus generator (Rigaku-MSC, The Woodlands, TX). Analytical x-rays are generated by electron bombardment of a rotating copper anode and are filtered through a thin nickel foil to provide monochromatic CuKα radiation (Kα1 and Kα2 unresolved; λ = 1.54 Å). Collimation of the x-ray beam was achieved using a single Franks mirror. Diffraction data were collected on a one-dimensional, position-sensitive electronic detector (Hecus x-ray Systems, Graz, Austria) using a sample-to-detector distance of 150 mm. In addition to direct calibration, crystalline cholesterol monohydrate was used to verify the calibration of the detector.
      This technique allows for precise measurement of the unit cell periodicity, or d-space, of the membrane lipid bilayer, which is the distance from the center of one lipid bilayer to the next, including surface hydration. The d-space for any given membrane multibilayer is calculated from Bragg's Law.
      hλ=2dsinθ
      (Eq. 1)


      where h is the diffraction order, λ is the wavelength of the x-ray radiation (1.54 Å), d is the membrane lipid bilayer unit cell periodicity, and θ is the Bragg angle equal to one-half the angle between the incident beam and scattered beam.

      RESULTS

      Effects of Cholesterol on Lipid Peroxidation—Model membranes composed of DLPC alone, DLPC + cholesterol (at a C/P mole ratio of 0.6), or POPC + cholesterol (at the same C/P mole ratio) were prepared and subjected to autoxidation by incubation at 37 °C for 72 h. Lipid peroxidation was monitored at 24-h intervals from baseline (0 peroxidation) using the spectrophotometric CHOD-iodide assay (Fig. 1). In the absence of cholesterol, DLPC peroxidation was observed to increase by ∼50% between each time point up to 48 h; the concentration of lipid hydroperoxides increased by 200% at 72 h. The addition of cholesterol to DLPC membranes resulted in lipid hydroperoxide levels at each time point that were consistently greater than those observed for DLPC alone, with a 45% increase over control at the 72-h time point. No significant changes in POPC lipid peroxide levels were observed between experimental time points.
      Figure thumbnail gr1
      FIGURE 1Cholesterol increased the time-dependent autoxidation of DLPC vesicles. Vesicles were prepared in the absence (○) or presence (▪) of cholesterol at a C/P mole ratio of 0.6. Statistically significant difference (p < 0.001; two-tailed Student's t test) was achieved between each sample time point and between DLPC-only and DLPC-cholesterol time points (mean ± S.D., n = 7). No significant changes in POPC oxidation (♦) were observed during the experimental time course.
      Peroxidation-induced Cholesterol Domain Formation—These peroxidation data would suggest, as has been shown for DLPC alone (
      • Mason R.P.
      • Walter M.F.
      • Mason P.E.
      ), that progressive autoxidation significantly alters membrane structure, and such changes would be particularly enhanced in the presence of cholesterol. This hypothesis was tested in this study using small angle x-ray diffraction approaches. Representative diffraction profiles generated from both DLPC-only (controls) and DLPC + cholesterol (0.6 C/P) membranes, at 0- and 72-h peroxidation time points, are shown in Fig. 2. Scattering data collected from the non-peroxidized control yielded four strong diffraction orders and a unit cell periodicity (d-space) of 48 Å (Fig. 2A). Following peroxidation, the fourth lipid diffraction order was lost, and the membrane d-space was reduced to 44 Å (Fig. 2B), consistent with a membrane disordering effect. The addition of cholesterol to DLPC bilayers at peroxidation baseline resulted in an increase in membrane periodicity to 51 Å (Fig. 2C). Following 72 h of peroxidation, membrane phospholipid periodicity decreased to 41 Å in the presence of cholesterol (Fig. 2D). The diffraction pattern generated from this sample also revealed the presence of prominent cholesterol monohydrate domains, as indicated by well defined Bragg peaks having a unit cell periodicity of 34 Å. These data demonstrate that lipid peroxidation induces significant changes in membrane structure, the most striking of which is the lateral phase separation of cholesterol into distinct membrane domains. This biphasic lipid separation effect was not observed in POPC membranes containing equimolar concentrations of cholesterol and exposed to the same level of oxidative stress (Fig. 2, E and F).
      Figure thumbnail gr2
      FIGURE 2Lipid peroxidation altered membrane structure and promoted the formation of immiscible cholesterol domains. Representative x-ray diffraction patterns were obtained from non-peroxidized DLPC membranes (A), DLPC membranes subjected to 72 h of peroxidation (B), non-peroxidized DLPC + cholesterol membranes (C), and DLPC + cholesterol membranes subjected to 72 h of peroxidation (D). Diffraction patterns were also obtained from POPC + cholesterol controls collected at 0- and 72-h time points (E and F, respectively). All membrane samples containing cholesterol (C-F) were prepared at a C/P mole ratio of 0.6. Data were collected on a one-dimensional, position-sensitive electronic detector at 20 °C and 74% relative humidity. In each panel, peaks labeled as 1, 2, 3, or 4 represent ordered scattering from the liquid crystalline phospholipid bilayer phase. Diffraction peaks labeled as 1′, 2′, and 3′ in panel D correspond to immiscible cholesterol domains (periodicity of 34 Å).
      Effects of Peroxidation and Cholesterol Content on Domain Formation—To establish the basic biophysical requirements for the formation of cholesterol domains as observed in these model systems, x-ray diffraction approaches were used to examine DLPC membrane vesicles containing cholesterol at C/P mole ratios ranging from 0 to 1.0 (Fig. 3). At peroxidation baseline, cholesterol domain formation was not observed at C/P mole ratios below 1.0 (Fig. 3A), suggesting that greater cholesterol content is required for phase separation to occur under normal conditions. Changes in bilayer packing and distribution of cholesterol are particularly well illustrated in Fig. 3A. As the C/P mole ratio increases, the third-order phospholipid diffraction peak, prominent in the absence of cholesterol (0 C/P), decreases until completely diminished at 0.8 C/P. This third-order intensity effect is consistent with an increased packing of cholesterol within the hydrocarbon core of the phospholipid bilayer (
      • Mason R.P.
      • Moisey D.M.
      • Shajenko L.
      ,
      • McIntosh T.J.
      ). A small third-order diffraction peak reappeared at 1.0 C/P, slightly overlapping with a larger, second-order diffraction peak associated with lateral cholesterol domains (17 Å). The changes at 1.0 C/P reflect a cholesterol “saturation” effect in which cholesterol is forced into separate lamellar domains as cholesterol packing between membrane phospholipids is reduced (
      • Tulenko T.N.
      • Chen M.
      • Mason P.E.
      • Mason R.P.
      ). Lipid peroxidation significantly altered the cholesterol packing effects observed in DLPC membranes at baseline. After 72 h of peroxidation, well defined diffraction peaks corresponding to cholesterol domains were apparent at C/P mole ratios as low as 0.4 (Fig. 3B). The blunt right-handed shoulder associated with the first-order phospholipid diffraction peak at 0.2 C/P indicates that moderate cholesterol domain formation may be induced even at this lower cholesterol level. Interestingly, maximal cholesterol peak intensities were obtained at 0.6 C/P. This set of experiments suggested that a C/P range of 0.4-0.8 was optimum for studying peroxidation-induced cholesterol domain formation in DLPC vesicles.
      Figure thumbnail gr3
      FIGURE 3Characterization of cholesterol domain formation in model membranes as a function of increasing cholesterol content and lipid peroxidation state. Diffraction profiles were generated from DLPC ± cholesterol (Δ C/P) membranes that were exposed to 0 h of peroxidation (baseline) (A) or 72 h of peroxidation (B). Diffraction peaks corresponding to cholesterol domains were obtained only at a C/P mole ratio of 1.0 in the absence of peroxidation (panel A). After extensive peroxidation, well defined cholesterol domain peaks were observed at all C/P mole ratios greater than 0.4 (panel B). Bragg peaks associated with cholesterol domains are marked by arrows in each panel.
      Time-dependent Changes in Peroxidation-induced Cholesterol Domain Structure—To further characterize the effects of peroxidation on cholesterol domain formation, two samples composed of DLPC + cholesterol at 0.6 C/P were separately prepared and examined at 24-h intervals over a 72-h lipid peroxidation time course (Fig. 4). For both samples, well defined diffraction peaks associated with cholesterol domains were identified as early as 24 h, and further peroxidation promoted the formation of stronger, more intense cholesterol diffraction peaks. Relative cholesterol peak intensities were derived from the integration of the 17 Å, second-order cholesterol peaks for each sample and plotted versus the length of peroxidation (Fig. 5). It can be noted that the intensities of the 34 Å cholesterol peaks also increased as a function of increasing exposure to oxidative insult (Fig. 4); however, since the first-order cholesterol peaks often overlap with the first- or second-order phospholipid diffraction peaks, the 17 Å peaks were better suited for quantitative analysis. Integration was performed using a subroutine written for Origin 6.0 (Microcal Software, Inc.). These data indicated that cholesterol peak intensity increased in a time-dependent manner, with a curvilinear function that is strikingly similar to the time-dependent changes in lipid peroxide formation (Fig. 5, inset).
      Figure thumbnail gr4
      FIGURE 4Extent of cholesterol domain formation in DLPC membrane vesicles was dependent on length of exposure to peroxidative insult. Two separate membrane samples (A and B, respectively) were analyzed at 24-h intervals beginning at baseline. Diffraction peaks corresponding to cholesterol domains (marked by arrows in each panel) were observed as early as 24 h of peroxidation, and increasing membrane exposure to lipid peroxidation was associated with concomitant increases in cholesterol peak intensities (). C/P mole ratio = 0.6.
      Figure thumbnail gr5
      FIGURE 5Effects of lipid peroxidation on relative cholesterol peak intensity. Peak intensity was based on the integration of the 17.0 Å Bragg peaks observed in the diffraction patterns obtained from Samples 1 and 2 (). Note that cholesterol peak intensity increased sharply at 72 h in parallel with increased lipid peroxide formation as illustrated in the inset (▪, Sample 1; ○, Sample 2).
      Concomitant with these pronounced changes in cholesterol structural organization, lipid peroxidation also induced significant changes in the structural organization of the cholesterol-poor phospholipid bilayer. As shown in Fig. 6, the membrane d-space of both samples decreased sharply with increasing length of peroxidation (72 h). For purposes of comparison, the membrane periodicity values obtained from cholesterol-deprived DLPC membranes are also plotted in Fig. 6. Up to 48 h, cholesterol appeared to increase relative membrane structural order as indicated by the higher d-space values associated with Samples 1 and 2 versus DLPC controls; however, this relationship was reversed at 72 h, concomitant with pronounced increases in lipid peroxide and cholesterol domain formation (Fig. 5).
      Figure thumbnail gr6
      FIGURE 6Effects of lipid peroxidation on model membrane bilayer unit cell periodicity (d-space). In the absence of cholesterol (♦), DLPC membrane d-space decreased by 9.0% over the 72-h peroxidation time course, consistent with a disordering effect induced by oxidative stress. Samples 1 and 2 (DLPC + 0.6 C/P) exhibited greater sensitivity to oxidative insult, with time-dependent changes in membrane d-space of 20.1 and 15.1%, respectively (▪, Sample 1; ○, Sample 2).
      Effects of Vitamin E on Cholesterol Domain Formation—The treatment of model membranes (DLPC + 0.6 C/P) with vitamin E (1:50 vitamin E to phospholipid mole ratio) attenuated the formation of cholesterol domains up to the 24-h peroxidation time point (Fig. 7). Over this same time course, vitamin E inhibited lipid peroxide formation by 43% (data not shown). In additional x-ray diffraction experiments, vitamin E was shown to have no inherent disruptive effects on the structural organization of cholesterol domains in DLPC membranes saturated with cholesterol (>1.0 C/P mole ratio). These data indicate that vitamin E, as a function of its antioxidant properties, can significantly delay peroxidation-induced alteration of membrane structure and cholesterol organization.
      Figure thumbnail gr7
      FIGURE 7Vitamin E attenuated peroxidation-induced cholesterol domain formation in model membranes. DLPC membranes containing cholesterol at 0.6 C/P mole ratio were treated with vehicle (A) or vitamin E (B) and exposed to oxidative stress for 24 h. X-ray diffraction patterns were collected on a one-dimensional, position-sensitive electronic detector at 20 °C and 74% relative humidity. Peaks labeled as 1, 2, 3, or 4 represent ordered scattering from the liquid crystalline phospholipid bilayer phase. Diffraction peaks labeled as 1′, 2′, and 3′ correspond to immiscible cholesterol domains (periodicity of 34 Å).

      DISCUSSION

      Lipid peroxidation is a free radical-mediated reaction that predominately affects polyunsaturated fatty acids in biological membranes. This reaction is typically initiated by abstraction of a hydrogen atom from a methylene group at the site of unsaturation, resulting in the formation of an alkyl radical that rapidly combines with molecular oxygen to form lipid peroxyl radicals. Lipid peroxides are highly unstable and readily react with neighboring fatty acids, setting off a chain reaction that propagates through the membrane unless checked by free radical scavengers, such as antioxidants (
      • Buege J.A.
      • Aust S.D.
      ,
      • Porter N.A.
      • Caldwell S.E.
      • Mills K.A.
      ). The rate of lipid peroxidation in biological membranes has been shown to be dependent on the physical state of constituent lipids. McLean and Hagaman (
      • McLean L.R.
      • Hagaman K.A.
      ) showed that the rate of lipid peroxidation of arachidonic acid in dimyristoylphosphatidylcholine liposomes was more rapid below the phase transition temperature (Tm) of the host lipid. Parallel anisotropy studies confirmed that the peroxidation rate was directly related to the motional order of the lipid bilayer. Peroxidation was shown to increase as a function of increasing membrane rigidity (
      • McLean L.R.
      • Hagaman K.A.
      ). Similar effects have been observed in model membranes prepared from distearoylphosphatidylcholine (
      • Mowri H.
      • Nojima S.
      • Inoue K.
      ) and dipalmitoylphosphatidylcholine (
      • Cervato G.
      • Viani P.
      • Masserini M.
      • Di Iorio C.
      • Cestaro B.
      ,
      • Viani P.
      • Cervato G.
      • Fiorilli A.
      • Rigamonti E.
      • Cestaro B.
      ).
      Cholesterol also influences the rate of lipid peroxidation in biological membranes. Based on its stereospecific interactions with component phospholipids, cholesterol has been shown to directly modulate membrane lipid order (
      • Yeagle P.L.
      ,
      • Franks N.P.
      ). Specifically, increasing membrane cholesterol content reduces the motional order of membrane lipids in the liquid-crystalline state and increases the motional order of membrane lipids in the gel state (
      • Leonard A.
      • Dufourc E.J.
      ,
      • De Kruijff B.
      • Cullis P.R.
      • Radda G.K.
      ,
      • Engelman D.M.
      • Rothman J.E.
      ). Consistent with these biophysical effects, cholesterol reduces the rate of lipid peroxidation below Tm and increases the rate of peroxidation above Tm (
      • McLean L.R.
      • Hagaman K.A.
      ,
      • Mowri H.
      • Nojima S.
      • Inoue K.
      ). Cholesterol has also been shown to increase the amount of lipid peroxidation in liposomes exposed to γ-radiation, which increases membrane rigidity (
      • Nakazawa T.
      • Nagatsuka S.
      • Yukawa O.
      ).
      In this study, cholesterol significantly increased lipid peroxidation in model membranes enriched in DLPC (Fig. 1). Since the Tm for DLPC is -53 °C, cholesterol would be expected to increase membrane rigidity at physiologic conditions, an effect also confirmed in this study by measurement of cholesterol-induced changes in membrane width (Figs. 2 and 6). These results are consistent with the idea that cholesterol increases the lateral packing of phospholipid acyl chains, facilitating the propagation of free radicals through the lipid bilayer.
      Diffraction data collected from DLPC-only (control) membranes showed that membrane d-space values gradually decreased as a function of increasing peroxidation, with a change of ∼4 Å observed at the 72-h time point (Fig. 6). This change in membrane periodicity is consistent with a lipid disordering effect, as shown previously in DLPC-enriched model membranes exposed to either Fe2+/ascorbate or autoxidative forms of lipid peroxidation (
      • Mason R.P.
      • Walter M.F.
      • Mason P.E.
      ,
      • Mason R.P.
      • Jacob R.F.
      ). In both systems, lipid peroxidation caused a dose-dependent decrease in overall membrane bilayer periodicity, a decrease in intrabilayer phospholipid headgroup separation, and an increase in hydrocarbon core molecular volume. These effects are indicative of changes in the intermolecular packing and thermodynamic properties of the phospholipid bilayer (
      • Mason R.P.
      • Trumbore M.W.
      • Pettegrew J.W.
      ,
      • Mason R.P.
      • Trumbore M.W.
      • Pettegrew J.W.
      ).
      Several unique changes in membrane structure were observed with the addition of cholesterol to DLPC membranes. Most importantly, lipid peroxidation of membranes containing relatively low sterol concentrations induced the lateral phase separation of cholesterol within the plane of the membrane bilayer. Cholesterol domain formation was only observed at a C/P mole ratio of 1.0 in non-peroxidized control samples, consistent with previous reports (
      • Tulenko T.N.
      • Chen M.
      • Mason P.E.
      • Mason R.P.
      ,
      • Ruocco M.J.
      • Shipley G.G.
      ); however, cholesterol domain formation was apparent at C/P mole ratios as low as 0.4, following 72 h of peroxidation (Fig. 3). At a C/P mole ratio of 0.6, cholesterol domains were observed following only 24 h of peroxidation, and cholesterol domain formation increased as a function of increasing length of peroxidation (Figs. 4 and 5). Interestingly, the cholesterol domains that formed in these membranes following 72 h of peroxidation yielded a third diffraction order (Fig. 4), indicating that these domains are extremely stable and well defined. Membranes deficient in polyunsaturated fatty acids (POPC model membranes) were resistant to lipid peroxidation (Fig. 1) and cholesterol domain formation (Fig. 2), confirming that the phase separation observed in DLPC membranes was directly dependent on peroxidative changes in the phospholipid environment. This effect was also confirmed by the treatment of DLPC membranes with vitamin E, which delayed the onset of domain formation through the 24-h time point, as compared with control (Fig. 7). These data provide the first direct evidence for peroxidation-induced formation of cholesterol domains in biological membranes.
      Cholesterol oxide formation was not specifically measured in these experiments; however, it is reasonable to expect that cholesterol itself may have undergone some degree of oxidative modification in these model membranes. Oxidized forms of cholesterol are known to alter the structural order and dynamic properties of lipids, with adverse effects on the function of biological membranes (
      • Schroeder F.
      • Jefferson J.R.
      • Kier A.B.
      • Knittel J.
      • Scallen T.J.
      • Wood W.G.
      • Hapala I.
      ,
      • Kutryk M.J.
      • Maddaford T.G.
      • Ramjiawan B.
      • Pierce G.N.
      ,
      • El Yandouzi E.H.
      • Le Grimellec C.
      ,
      • Wood W.G.
      • Igbavboa U.
      • Rao A.M.
      • Schroeder F.
      • Avdulov N.A.
      ,
      • Verhagen J.C.
      • ter Braake P.
      • Teunissen J.
      • van Ginkel G.
      • Sevanian A.
      ). The presence of cholesterol oxides may partially explain the sharp decrease in d-space that was observed in DLPC + cholesterol membranes exposed to 72 h of peroxidation (Fig. 6). Sequestration of cholesterol into domains would be expected to promote a decrease in d-space (
      • Tulenko T.N.
      • Chen M.
      • Mason P.E.
      • Mason R.P.
      ), but the significant decrease over controls (DLPC alone) suggests a disordering effect beyond that induced by cholesterol displacement. Apart from this putative disordering effect on the phospholipid bilayer, cholesterol oxides appeared to have a limited involvement in the principle structural effects measured in these experiments. This argument is largely supported by the observation that cholesterol domain formation was absent in POPC/cholesterol membranes, as compared with DLPC/cholesterol membranes, despite equal sterol concentrations and exposure to the same amount of oxidative insult (Figs. 1 and 2). Despite obvious differences in phospholipid makeup (
      • Sevanian A.
      • McLeod L.L.
      ), cholesterol oxides would be expected to form in both membrane preparations over the time course of these experiments. These control experiments indicated that cholesterol domain formation occurred primarily as a function of phospholipid oxidation. Other aspects of cholesterol oxidation and membrane structure are currently under investigation in our laboratory.
      The results of this study have broad implications for a number of disease processes in which lipid peroxidation and cholesterol-enrichment may have etiologic roles. In general, lipid peroxidation alters the structure and function of cell membranes, often with cytopathologic consequences (
      • Girotti A.W.
      ). This process has been shown, for example, to impair calcium pump activity in sarcoplasmic reticulum in association with changes in membrane fluidity and other physical properties (
      • Dinis T.C.
      • Almeida L.M.
      • Madeira V.M.
      ). Similar changes in Ca2+-ATPase activity have been observed in bovine lens epithelial microsomes under conditions of oxidative stress (
      • Ahuja R.P.
      • Borchman D.
      • Dean W.L.
      • Paterson C.A.
      • Zeng J.
      • Zhang Z.
      • Ferguson-Yankey S.
      • Yappert M.C.
      ). In guinea pig neuronal membranes, lipid peroxidation disrupted the function of N-methyl-d-aspartate receptors, significantly decreasing the binding affinity of the N-methyl-d-aspartate antagonist MK-801 (
      • Goel R.
      • Mishra O.P.
      • Razdan B.
      • Delivoria-Papadopoulos M.
      ). In rat brain synaptosomes, lipid peroxidation and aging were shown to similarly perturb Na+K+-ATPase activity as well as membrane structure and lipid composition (
      • Viani P.
      • Cervato G.
      • Fiorilli A.
      • Cestaro B.
      ).
      Abnormal accumulation of cholesterol also adversely affects membrane function. Increased membrane cholesterol content has been shown to alter the conformation and activity of various membrane-associated ion channels, including calcium (
      • Chang H.M.
      • Reitstetter R.
      • Mason R.P.
      • Gruener R.
      ,
      • Bialecki R.A.
      • Tulenko T.N.
      ,
      • Gleason M.M.
      • Medow M.S.
      • Tulenko T.N.
      ) and potassium channels (
      • Chang H.M.
      • Reitstetter R.
      • Mason R.P.
      • Gruener R.
      ,
      • Bolotina V.
      • Gericke M.
      • Bregestovski P.
      ) in various tissues. Membrane cholesterol enrichment also disrupts Na+K+-ATPase activity in erythrocytes (
      • Giraud F.
      • Claret M.
      • Bruckdorfer K.R.
      • Chailley B.
      ,
      • Yeagle P.L.
      ), endothelial cells (
      • Lau Y.T.
      ), renal cells (
      • Yeagle P.L.
      • Young J.
      • Rice D.
      ), and smooth muscle cells (
      • Broderick R.
      • Bialecki R.
      • Tulenko T.N.
      ,
      • Chen M.
      • Mason R.P.
      • Tulenko T.N.
      ).
      The etiologic association of lipid peroxidation with atherosclerosis is well established (
      • Steinberg D.
      ,
      • Berliner J.A.
      • Heinecke J.W.
      ,
      • Diaz M.N.
      • Frei B.
      • Vita J.A.
      • Keaney Jr., J.F.
      ,
      • Heinecke J.W.
      ,
      • Yla-Herttuala S.
      • Palinski W.
      • Rosenfeld M.E.
      • Parthasarathy S.
      • Carew T.E.
      • Butler S.
      • Witztum J.L.
      • Steinberg D.
      ). Oxidative modification of unsaturated fatty acids and apolipoproteins in low density lipoproteins results in increased low density lipoprotein uptake by vascular macrophages and the progressive formation of atherosclerotic plaques (
      • Steinberg D.
      • Parthasarathy S.
      • Carew T.E.
      • Khoo J.C.
      • Witztum J.L.
      ,
      • Steinbrecher U.P.
      • Lougheed M.
      • Kwan W.C.
      • Dirks M.
      ,
      • Esterbauer H.
      • Gebicki J.
      • Puhl H.
      • Jurgens G.
      ). Lipid peroxidation also has deleterious effects on numerous vascular wall constituents, including endothelial cell membranes (
      • Shatos M.A.
      • Doherty J.M.
      • Hoak J.C.
      ).
      Cholesterol accumulation in vascular cell membranes has also been implicated in the etiology of atherosclerosis (
      • Tulenko T.N.
      • Chen M.
      • Mason P.E.
      • Mason R.P.
      ,
      • Kellner-Weibel G.
      • Yancey P.G.
      • Jerome W.G.
      • Walser T.
      • Mason R.P.
      • Phillips M.C.
      • Rothblat G.H.
      ,
      • Small D.M.
      ). Vascular smooth muscle cells and macrophage foam cells, as models of atherosclerosis, have been used to demonstrate that elevated cholesterol levels promote the formation of membrane-restricted cholesterol domains, with biophysical characteristics similar to those observed in this study (
      • Tulenko T.N.
      • Chen M.
      • Mason P.E.
      • Mason R.P.
      ,
      • Kellner-Weibel G.
      • Yancey P.G.
      • Jerome W.G.
      • Walser T.
      • Mason R.P.
      • Phillips M.C.
      • Rothblat G.H.
      ). In cultured mouse peritoneal macrophage foam cells, excess free cholesterol promoted the formation of crystal structures that extended from the cell membranes with various morphologies, including plates, needles, and helices (
      • Kellner-Weibel G.
      • Yancey P.G.
      • Jerome W.G.
      • Walser T.
      • Mason R.P.
      • Phillips M.C.
      • Rothblat G.H.
      ). Cholesterol domains in the plasma membranes of these cells are believed to precede and contribute to the development of atherosclerotic crystals (
      • Mason R.P.
      • Jacob R.F.
      ,
      • Mason R.P.
      • Tulenko T.N.
      • Jacob R.F.
      ).
      Lipid peroxidation also plays a significant role in cataract development (
      • Bhuyan K.C.
      • Bhuyan D.K.
      • Podos S.M.
      ). Studies using human lens organ cultures in the presence of free radical-producing systems have shown formation of cataracts in vitro, with concomitant increases in indices of lipid peroxidation and changes in Na+K+-ATPase activity leading to ionic imbalance within the lens cells (
      • Garner M.H.
      • Spector A.
      ,
      • Garner W.H.
      • Garner M.H.
      • Spector A.
      ,
      • Spector A.
      ). Infrared spectroscopy approaches have been used to demonstrate that lipid peroxidation products, including lipid carbonyls and trans double bonds, accumulate linearly in the human lens as a function of age (
      • Borchman D.
      • Yappert M.C.
      ), and incubation of rat and human lenses with lipid peroxidation products in organ culture promotes the formation of lenticular opacities (
      • Horikawa Y.
      • Mibu H.
      • Hikida M.
      ,
      • Zigler Jr., J.S.
      • Bodaness R.S.
      • Gery I.
      • Kinoshita J.H.
      ). Borchman and Yappert (
      • Borchman D.
      • Yappert M.C.
      ) provided evidence suggesting that lipid peroxidation impacts even monounsaturated fatty acids in the lens membrane, resulting in an increased number of lipid oxidation products.
      Recent studies in our laboratory have provided evidence that the structural organization of the lens cell plasma membrane is altered in early-stage cataracts. Normal lens membranes have been shown to have a biphasic structural motif consisting of immiscible cholesterol monohydrate domains surrounded by a cholesterol-poor, liquid-crystalline bilayer (
      • Jacob R.F.
      • Cenedella R.J.
      • Mason R.P.
      ). Direct analysis of membranes isolated from early-stage cataractous lenses revealed that cholesterol domains were more pronounced and more stable than those of normal lens membranes. This was true despite a significantly lower (∼54%) C/P mole ratio (
      • Jacob R.F.
      • Cenedella R.J.
      • Mason R.P.
      ). It was speculated that these structural changes may be due to modification of the lens membrane by age-related insults, such as lipid peroxidation.
      In conclusion, the results of this study demonstrated that lipid peroxidation disrupts the biophysical properties of membranes, inducing significant changes in lipid structure and organization. Under conditions of low oxidative stress, cholesterol is randomly distributed throughout the lipid bilayer, where it contributes to the width of the membrane bilayer. Exposure to lipid peroxidation causes cholesterol monomers to self-associate, forming membrane-restricted domains and effecting an overall reduction in the bilayer width (Fig. 8). Cholesterol domain formation was shown to increase in a time-dependent manner and in parallel with lipid peroxide accumulation. These data have important implications for a number of oxidative stress-related disorders, including atherosclerosis and cataractogenesis.
      Figure thumbnail gr8
      FIGURE 8Schematic representation of the effects of lipid peroxidation on membrane structure and cholesterol organization. Based on data collected in this study, cholesterol at relatively low concentrations is fully miscible in normal or non-peroxidized membranes and contributes to overall bilayer width. Exposure to oxidative stress induces segregation of cholesterol into distinct domain regions (34 Å) within the liquid crystalline membrane bilayer. This transfer of cholesterol from the phospholipid bilayer into separate cholesterol domains is also marked by a decrease in overall membrane bilayer width.

      References

        • Girotti A.W.
        J. Lipid Res. 1998; 39: 1529-1542
        • Leibowitz M.E.
        • Johnson M.C.
        J. Lipid Res. 1971; 12: 662-670
        • Gardner H.W.
        J. Agric. Food Chem. 1975; 23: 129-136
        • Buege J.A.
        • Aust S.D.
        Methods Enzymol. 1978; 52: 302-310
        • Petrescu A.D.
        • Gallegos A.M.
        • Okamura Y.
        • Strauss III, J.F.
        • Schroeder F.
        J. Biol. Chem. 2001; 276: 36970-36982
        • Chatterjee S.N.
        • Agarwal S.
        Free Radic. Biol. Med. 1988; 4: 51-72
        • Borchman D.
        • Lamba O.P.
        • Salmassi S.
        • Lou M.
        • Yappert M.C.
        Lipids. 1992; 27: 261-265
        • Goldstein R.M.
        • Weissmann G.
        Biochem. Biophys. Res. Commun. 1977; 75: 604-609
        • Mandal T.K.
        • Chatterjee S.N.
        Radiat. Res. 1980; 83: 290-302
        • Kunimoto M.
        • Inoue K.
        • Nojima S.
        Biochim. Biophys. Acta. 1981; 646: 169-178
        • Chia L.S.
        • Thompson J.E.
        • Moscarello M.A.
        Biochim. Biophys. Acta. 1984; 775: 308-312
        • van Duijn G.
        • Verkleij A.J.
        • de Kruijff B.
        Biochemistry. 1984; 23: 4969-4977
        • Verma S.P.
        Radiat. Res. 1986; 107: 183-193
        • Yukawa O.
        • Nagatsuka S.
        • Nakazawa T.
        Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 1983; 43: 391-398
        • Dinis T.C.
        • Almeida L.M.
        • Madeira V.M.
        Arch. Biochem. Biophys. 1993; 301: 256-264
        • Goel R.
        • Mishra O.P.
        • Razdan B.
        • Delivoria-Papadopoulos M.
        Neurosci. Lett. 1993; 151: 219-223
        • Kourie J.I.
        Am. J. Physiol. 1998; 275: C1-C24
        • Mattson M.P.
        Trends Neurosci. 1998; 21: 53-57
        • Mattson M.P.
        • Pedersen W.A.
        • Duan W.
        • Culmsee C.
        • Camandola S.
        Ann. N. Y. Acad. Sci. 1999; 893: 154-175
        • Sevanian A.
        • Ursini F.
        Free Radic. Biol. Med. 2000; 29: 306-311
        • Mason R.P.
        • Walter M.F.
        • Mason P.E.
        Free Radic. Biol. Med. 1997; 23: 419-425
        • Mason R.P.
        • Jacob R.F.
        Methods Mol. Biol. 2002; 196: 69-78
        • McLean L.R.
        • Hagaman K.A.
        Free Radic. Biol. Med. 1992; 12: 113-119
        • Leonard A.
        • Dufourc E.J.
        Biochimie (Paris). 1991; 73: 1295-1302
        • Mowri H.
        • Nojima S.
        • Inoue K.
        J. Biochem. (Tokyo). 1984; 95: 551-558
        • El-Saadani M.
        • Esterbauer H.
        • El-Sayed M.
        • Goher M.
        • Nassar A.Y.
        • Jurgens G.
        J. Lipid Res. 1989; 30: 627-630
        • Bangham A.D.
        • Standish M.M.
        • Watkins J.C.
        J. Mol. Biol. 1965; 13: 238-252
        • Herbette L.G.
        • DeFoor P.
        • Fleischer S.
        • Pascolini D.
        • Blaise J.K.
        Biophys. J. 1985; 20: 245-272
        • Mason R.P.
        • Moisey D.M.
        • Shajenko L.
        Mol. Pharmacol. 1992; 41: 315-321
        • McIntosh T.J.
        Biochim. Biophys. Acta. 1978; 513: 43-58
        • Tulenko T.N.
        • Chen M.
        • Mason P.E.
        • Mason R.P.
        J. Lipid Res. 1998; 39: 947-956
        • Porter N.A.
        • Caldwell S.E.
        • Mills K.A.
        Lipids. 1995; 30: 277-290
        • Cervato G.
        • Viani P.
        • Masserini M.
        • Di Iorio C.
        • Cestaro B.
        Chem. Phys. Lipids. 1988; 49: 135-139
        • Viani P.
        • Cervato G.
        • Fiorilli A.
        • Rigamonti E.
        • Cestaro B.
        Chem. Phys. Lipids. 1990; 52: 49-55
        • Yeagle P.L.
        Biochim. Biophys. Acta. 1985; 822: 267-287
        • Franks N.P.
        J. Mol. Biol. 1976; 100: 345-358
        • De Kruijff B.
        • Cullis P.R.
        • Radda G.K.
        Biochim. Biophys. Acta. 1976; 436: 729-740
        • Engelman D.M.
        • Rothman J.E.
        J. Biol. Chem. 1972; 247: 3694-3697
        • Nakazawa T.
        • Nagatsuka S.
        • Yukawa O.
        Drugs Exp. Clin. Res. 1986; 12: 831-835
        • Mason R.P.
        • Trumbore M.W.
        • Pettegrew J.W.
        Neurobiol. Aging. 1995; 16: 531-539
        • Mason R.P.
        • Trumbore M.W.
        • Pettegrew J.W.
        Ann. N. Y. Acad. Sci. 1996; 777: 368-373
        • Ruocco M.J.
        • Shipley G.G.
        Biophys. J. 1984; 46: 695-707
        • Schroeder F.
        • Jefferson J.R.
        • Kier A.B.
        • Knittel J.
        • Scallen T.J.
        • Wood W.G.
        • Hapala I.
        Proc. Soc. Exp. Biol. Med. 1991; 196: 235-252
        • Kutryk M.J.
        • Maddaford T.G.
        • Ramjiawan B.
        • Pierce G.N.
        Circ. Res. 1991; 68: 18-26
        • El Yandouzi E.H.
        • Le Grimellec C.
        Biochemistry. 1993; 32: 2047-2052
        • Wood W.G.
        • Igbavboa U.
        • Rao A.M.
        • Schroeder F.
        • Avdulov N.A.
        Brain Res. 1995; 683: 36-42
        • Verhagen J.C.
        • ter Braake P.
        • Teunissen J.
        • van Ginkel G.
        • Sevanian A.
        J. Lipid Res. 1996; 37: 1488-1502
        • Sevanian A.
        • McLeod L.L.
        Lipids. 1987; 22: 627-636
        • Ahuja R.P.
        • Borchman D.
        • Dean W.L.
        • Paterson C.A.
        • Zeng J.
        • Zhang Z.
        • Ferguson-Yankey S.
        • Yappert M.C.
        Free Radic. Biol. Med. 1999; 27: 177-185
        • Viani P.
        • Cervato G.
        • Fiorilli A.
        • Cestaro B.
        J. Neurochem. 1991; 56: 253-258
        • Chang H.M.
        • Reitstetter R.
        • Mason R.P.
        • Gruener R.
        J. Membr. Biol. 1995; 143: 51-63
        • Bialecki R.A.
        • Tulenko T.N.
        Am. J. Physiol. 1989; 257: C306-C314
        • Gleason M.M.
        • Medow M.S.
        • Tulenko T.N.
        Circ. Res. 1991; 69: 216-227
        • Bolotina V.
        • Gericke M.
        • Bregestovski P.
        Proc. Biol. Sci. 1991; 244: 51-55
        • Giraud F.
        • Claret M.
        • Bruckdorfer K.R.
        • Chailley B.
        Biochim. Biophys. Acta. 1981; 647: 249-258
        • Yeagle P.L.
        Biochim. Biophys. Acta. 1983; 727: 39-44
        • Lau Y.T.
        Atherosclerosis. 1994; 110: 251-257
        • Yeagle P.L.
        • Young J.
        • Rice D.
        Biochemistry. 1988; 27: 6449-6452
        • Broderick R.
        • Bialecki R.
        • Tulenko T.N.
        Am. J. Physiol. 1989; 257: H170-H178
        • Chen M.
        • Mason R.P.
        • Tulenko T.N.
        Biochim. Biophys. Acta. 1995; 1272: 101-112
        • Steinberg D.
        Circulation. 1991; 84: 1420-1425
        • Berliner J.A.
        • Heinecke J.W.
        Free Radic. Biol. Med. 1996; 20: 707-727
        • Diaz M.N.
        • Frei B.
        • Vita J.A.
        • Keaney Jr., J.F.
        N. Engl. J. Med. 1997; 337: 408-416
        • Heinecke J.W.
        Atherosclerosis. 1998; 141: 1-15
        • Yla-Herttuala S.
        • Palinski W.
        • Rosenfeld M.E.
        • Parthasarathy S.
        • Carew T.E.
        • Butler S.
        • Witztum J.L.
        • Steinberg D.
        J. Clin. Investig. 1989; 84: 1086-1095
        • Steinberg D.
        • Parthasarathy S.
        • Carew T.E.
        • Khoo J.C.
        • Witztum J.L.
        N. Engl. J. Med. 1989; 320: 915-924
        • Steinbrecher U.P.
        • Lougheed M.
        • Kwan W.C.
        • Dirks M.
        J. Biol. Chem. 1989; 264: 15216-15223
        • Esterbauer H.
        • Gebicki J.
        • Puhl H.
        • Jurgens G.
        Free Radic. Biol. Med. 1992; 13: 341-390
        • Shatos M.A.
        • Doherty J.M.
        • Hoak J.C.
        Arterioscler. Thromb. Vasc. Biol. 1991; 11: 594-601
        • Kellner-Weibel G.
        • Yancey P.G.
        • Jerome W.G.
        • Walser T.
        • Mason R.P.
        • Phillips M.C.
        • Rothblat G.H.
        Arterioscler. Thromb. Vasc. Biol. 1999; 19: 1891-1898
        • Small D.M.
        Arterioscler. Thromb. Vasc. Biol. 1988; 8: 103-129
        • Mason R.P.
        • Jacob R.F.
        Circulation. 2003; 107: 2270-2273
        • Mason R.P.
        • Tulenko T.N.
        • Jacob R.F.
        Biochim. Biophys. Acta. 2003; 1610: 198-207
        • Bhuyan K.C.
        • Bhuyan D.K.
        • Podos S.M.
        Life Sci. 1986; 38: 1463-1471
        • Garner M.H.
        • Spector A.
        Exp. Eye Res. 1986; 42: 339-348
        • Garner W.H.
        • Garner M.H.
        • Spector A.
        Proc. Natl. Acad. Sci. 1983; 80: 2044-2048
        • Spector A.
        FASEB J. 1995; 9: 1173-1182
        • Borchman D.
        • Yappert M.C.
        Investig. Ophthalmol. Vis. Sci. 1998; 39: 1053-1058
        • Horikawa Y.
        • Mibu H.
        • Hikida M.
        J. Eye (Atarashii Ganka). 1990; 7: 561-563
        • Zigler Jr., J.S.
        • Bodaness R.S.
        • Gery I.
        • Kinoshita J.H.
        Arch. Biochem. Biophys. 1983; 225: 149-156
        • Jacob R.F.
        • Cenedella R.J.
        • Mason R.P.
        J. Biol. Chem. 1999; 274: 31613-31618
        • Jacob R.F.
        • Cenedella R.J.
        • Mason R.P.
        J. Biol. Chem. 2001; 276: 13573-13578