This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Navab, M. Articles by Fogelman, A. M. Search for Related Content PubMed PubMed Citation Articles by Navab, M. Articles by Fogelman, A. M.

(Arteriosclerosis, Thrombosis, and Vascular Biology. 1996;16:831-842.)
© 1996 American Heart Association, Inc.

Articles

The Yin and Yang of Oxidation in the Development of the Fatty Streak

A Review Based on the 1994 George Lyman Duff Memorial Lecture

Mohamad Navab; Judith A. Berliner; Andrew D. Watson; Susan Y. Hama; Mary C. Territo; Aldons J. Lusis; Diana M. Shih; Brian J. Van Lenten; Joy S. Frank; Linda L. Demer; Peter A. Edwards; Alan M. Fogelman

the Atherosclerosis Research Unit and the Departments of Medicine, Pathology (J.A.B.), Physiology (J.S.F., L.L.D.), and Biological Chemistry (P.A.E.), School of Medicine; and the Department of Microbiology and Molecular Genetics, College of Letters and Sciences (A.J.L., D.M.S.), University of California at Los Angeles.

Correspondence to Alan M. Fogelman, MD, Department of Medicine, UCLA School of Medicine, Los Angeles, CA 90095-1736.

	Abstract

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Recent data support the hypothesis that the fatty streak develops in response to specific phospholipids contained in LDL that become trapped in the artery wall and become oxidized as a result of exposure to the oxidative waste of the artery wall cells. The antioxidants present within both LDL and the microenvironments in which LDL is trapped function to prevent the formation of these biologically active, oxidized lipids. Enzymes associated with LDL and HDL (eg, platelet activating factor acetylhydrolase) or with HDL alone (eg, paraoxonase) destroy these biologically active lipids. The regulation and expression of these enzymes are determined genetically and are also significantly modified by environmental influences, including the acute-phase response or an atherogenic diet. The balance of these multiple factors leads to an induction or suppression of the inflammatory response in the artery wall and determines the clinical course.

Key Words: lipid oxide • inflammatory reaction

	Introduction

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
On November 14, 1994, one of us (Dr Fogelman) had the privilege of presenting the George Lyman Duff Memorial Lecture at the 67th Scientific Sessions of the American Heart Association in Dallas, Tex. This review is a summary of the research and ideas presented in that lecture updated to include more recent work relating to the topics discussed in the lecture. The basic premise to be presented here is that the fatty streak develops as a response to specific lipids that are carried into the artery wall with LDL and that subsequently become oxidized as a result of exposure to the oxidative waste of the artery wall cells (the Yang). The opposing (calming) forces (the Yin) include the antioxidants present both within the LDL and within the microenvironments where the LDL is trapped. These antioxidants function to prevent the formation of the oxidized lipids. Other opposing (calming) forces include specific enzymes contained within LDL or associated with HDL that inactivate these biologically active lipids. The balance between the Yin and the Yang determines the response of the artery wall.

	Trapping of LDL in the Artery Wall

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
The earliest event in the development of the fatty streak is the transport of LDL into the artery wall. This is a concentration-dependent process that does not require receptor-mediated endocytosis.^{1 2} The late Tom Carew and his colleague Dawn Schwenke^{3 4} demonstrated that for any given concentration of lipoprotein in the plasma, lipoprotein retention in the artery wall was more important than the rate of transport into the artery wall. Kruth⁵ and the Simionescus and their colleagues⁶ were among the first to report changes in the lipoproteins that were retained in the artery wall. With state-of-the-art ultrastructural techniques, Frank and Fogelman⁷ extended these findings and demonstrated in exquisite detail the three-dimensional cage-work of fibers and fibrils secreted by the cells of the artery wall into the subendothelial space. Nivelstein-Post and coworkers⁸ demonstrated that when LDL was injected into a normal rabbit, it rapidly crossed the intact endothelium and became trapped in this three-dimensional cage-work. Subsequently, these same researchers demonstrated⁹ that LDL associates with the collagen fibers in the vicinity of the cross-connecting fibrils. The intimate association of LDL with the extracellular matrix of the subendothelial space described by Camejo and colleagues¹⁰ explains why higher concentrations of apoB are present in the artery wall than in the plasma.^{11 12 13}

	Initial Oxidation of LDL

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Oxidative modification of the trapped lipoproteins was originally proposed in the early 1980s.^{14 15 16 17 18} Proof that supported this proposal was published in the late 1980s.^{19 20 21 22} Witztum, Steinberg, and Parthasarathy^{23 24 25 26} and Chisolm²⁷ emphasized that the cells of the artery wall are constantly secreting oxidative waste products into their membranes and into the subendothelial space. Parthasarathy²⁵ has emphasized that the ability of the cells of the artery wall to oxidize LDL is directly related to their ability to "seed" the LDL with reactive oxygen species. Navab and colleagues²⁸ demonstrated in vitro, using human artery wall cell cocultures, that these cells were capable of creating microenvironments into which their oxidative waste could be secreted. At the same time, these microenvironments excluded aqueous antioxidants and allowed trapped LDL to undergo mild oxidation.²⁸ However, if the cells in the coculture were first treated with lipid-soluble (but not water-soluble) antioxidants, the oxidation of LDL was prevented. Adding antioxidants to preformed mildly oxidized LDL did not alter its biological activity. Thus, it was proposed that the antioxidants prevented the seeding of LDL and the formation of biologically active mildly oxidized LDL. In contrast, once the active lipids were formed in LDL, antioxidants did not prevent or reverse their biological activities. A number of studies have subsequently indicated that oral administration of lipid-soluble vitamins to humans attenuates the in vitro oxidation of LDL isolated from their plasma.^{29 30 31}

	Biological Consequences of Mildly Oxidized LDL

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Berliner and colleagues³² demonstrated that exposure of endothelial cells to the polar lipid fraction isolated from mildly oxidized LDL induced monocytes but not neutrophils to adhere to the cells. The adhesion of the monocytes was induced by the induction in the endothelial cells of P-selectin³³ and the monocyte-activating proteins MCP-1,³⁴ M-CSF,³⁵ and GRO.³⁶ Thus, all three stages (tethering, activation, and attachment)³⁷ of monocyte binding were induced by the oxidized lipids. In artery wall cocultures, the monocytes not only adhered to the endothelial cells but migrated along the MCP-1 gradient into the subendothelial space.^{28 38} In vitro, the formation of MCP-1 appeared to be a density-dependent property of differentiating monocytes.³⁹ Liao and colleagues⁴⁰ provided evidence that the mildly oxidized LDL was also biologically active in vivo, and Neiken and colleagues⁴¹ reported that the highest expression of MCP-1 was in areas of the arterial lesion where monocyte density was greatest. In the subendothelial space, the monocytes differentiated into macrophages under the influence of M-CSF.³⁵ The mRNA levels for these proteins were induced by the oxidized lipids as a result of increased transcription, mediated by the activation of an NFB-like transcription factor,^{42 43} and mRNA stabilization⁴² (Fig 1). The oxidized lipids also induced high levels of intracellular cAMP via a G protein-mediated mechanism.^{43 44} These high levels of cAMP decreased the expression of the neutrophil-binding receptor ELAM-1.^{43 45} Partly as a result of the decreased expression of ELAM-1, neutrophils did not bind. Another consequence of the oxidized lipids was the induction in endothelial cells of tissue factor.⁴⁶

View larger version (196K): [in this window] [in a new window]   Figure 1. Induction of monocyte binding and migration into the subendothelial space of a human artery wall cell coculture. A, Depiction of the trapping of LDL in the extracellular matrix (ECM), the seeding of LDL with the waste products of the artery wall cells (oxidants), and the ability of HDL to block the production of the biologically active mildly oxidized LDL (MM-LDL), which contains the biologically active oxidized lipid that induces the endothelial cells to express a binding molecule for monocytes (X-CAM) and causes both the smooth muscle cells and the endothelial cells to express a potent monocyte chemoattractant (MCP-1), which leads to monocyte chemotaxis into the subendothelial space of the coculture. B, Induction of the NFB-like transcription factor and the induction of mRNA for the MCP-1 in the absence (-) and presence (+) of added LDL. C, Photomicrograph demonstrating that the activation of the NFB-like transcription factor and induction of MCP-1 lead to monocyte binding and migration into the subendothelial space, where the monocytes can be counted as a sensitive bioassay for the production of MCP-1. Mono indicates monocytes; EC, endothelial monolayer; COL, collagen and extracellular matrix; SMC, smooth muscle cells. (C is reproduced from The Journal of Clinical Investigation by copyright permission of The American Society for Clinical Investigation: from Navab M, Hough GP, Stevenson LW, Drinkwater DC, Laks H, Fogelman AM. J Clin Invest. 1988;82:1859, Fig 2.)

The series of events initiated by mildly oxidized LDL would result in the migration of monocytes into the subendothelial space and their conversion into macrophages. This would in turn be expected to enrich the microenvironment in reactive oxygen species that could convert the trapped, mildly oxidized LDL (recognized by the LDL receptor) into highly oxidized LDL (recognized by the scavenger and/or oxidized LDL receptors). As a result, foam cells would be expected to form. The subsequent recruitment of more monocytes, proliferation and retention of monocyte-macrophages, elaboration of growth factors, secretion of other potent biological factors by the monocyte-macrophages, and eventually death of the foam cells would be expected to result in progression of the lesion.⁴⁷

The role of iron in promoting LDL oxidation was first elucidated by Hessler et al.⁴⁸ While the role of extracellular iron in promoting LDL oxidation has long been known, the role of free intracellular iron in promoting lipid oxidation within the cells of the artery wall has received less attention. Such lipid oxidation products could be transferred from the cells of the artery wall to LDL, thus, seeding the LDL and leading to the production of biologically active mildly oxidized LDL. Van Lenten and colleagues⁴⁹ demonstrated that increasing the intracellular content of free iron in artery wall cocultures led to both enhanced formation of lipid hydroperoxides within the cells and enhanced production of mildly oxidized LDL. The latter, in turn, induced monocyte transmigration into the subendothelial space of the cocultures. Additionally, Van Lenten and colleagues⁴⁹ found that the mildly oxidized LDL itself was capable of inducing increased levels of free iron within the artery wall cells. The result was to amplify the production of the mildly oxidized LDL in the cocultures. The increased levels of free iron that were induced in the artery wall cells by the mildly oxidized LDL resulted in a compensatory induction within the cells of both mRNA and protein for the heavy chain of apoferritin. These experiments suggested that the regulation of intracellular free iron levels in artery wall cells can be perturbed by the lipids in mildly oxidized LDL. Furthermore, apoferritin synthesis within these cells may be a compensatory mechanism to protect the cells against the resulting increases in intracellular free iron and the generation of further intracellular lipid hydroperoxides.

	Defining the Biologically Active Lipids in Mildly Oxidized LDL and the Role of HDL in Inactivating These Lipids

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Navab and colleagues⁵⁰ reported that an anti-inflammatory compound, leumedin, inhibited artery wall cell cocultures from producing biologically active mildly oxidized LDL without inhibiting the formation of conjugated dienes or TBARS in the LDL. These experiments suggested that a specific oxidized lipid(s) within the mildly oxidized LDL was responsible for its biological activity and that these biologically active lipids constituted such a small fraction of the oxidized lipids in the LDL that their destruction or neutralization was not detectable by changes in LDL conjugated diene or TBARS content.

In pursuing the identity of the biologically active oxidized lipid(s), Watson and colleagues⁵¹ found that after LDL was incubated with artery wall cells in coculture it contained phosphatidylcholine that had less 203-nm UV absorbance than the corresponding native LDL. Sphingomyelin appeared unaltered by similar incubations. Simultaneously, conjugated dienes (measured as absorbance at 235 nm) increased in the phospholipid fraction after LDL had been incubated with the artery wall cells. Under the HPLC conditions used by Watson and colleagues, phosphatidylcholine eluted as a bimodal peak. The first of these peaks eluted with the same retention time as the single peak produced by authentic PAPC. The second peak eluted with the same retention time as the single peak produced by authentic 1-palmitoyl-2-linoleoyl-sn-glycero-3-phosphorylcholine. In 12 of 12 experiments, it was found that the arachidonic acid-containing phosphatidylcholine associated with the first peak was diminished after the LDL was treated with oxidative enzymes or was incubated with the artery wall cocultures.⁵¹ The oxidized phospholipid fractions, isolated by HPLC from LDL after incubation with the artery wall cocultures, demonstrated biological activity in vitro (ie, these oxidized phospholipids induced cultured endothelial cells to bind more monocytes). This biological activity of the oxidized phospholipids isolated by HPLC from LDL after incubation with artery wall cocultures was mimicked by incubating the endothelial cells with oxidized authentic PAPC. Biological activity was also detected in oxidized phospholipids extracted from the LDL that eluted near phosphatidylethanolamine. The oxidized phospholipids recovered from mildly oxidized LDL reacted with dinitrophenylhydrazine, indicating the presence of aldehydes that presumably resulted from ß-scission of polyunsaturated fatty acids esterified in the phospholipids. Stremler and colleagues^{52 53} had previously reported that an oxidized derivative of phosphatidylcholine with a five-carbon aldehyde at the sn-2 position was a good substrate for the PAF acetylhydrolase isolated from human plasma. Moreover, Stafforini and colleagues⁵⁴ had reported that the PAF acetylhydrolase, isolated from human plasma, prevented the oxidative modification of LDL in vitro that was normally induced by metal ions.

Since the amount of biologically active phospholipids in the coculture-produced mildly oxidized LDL was a small fraction of the total phospholipids, Watson and colleagues⁵¹ enriched for oxidized phospholipids by taking advantage of the observations by Deigner et al⁵⁵ that incubation of oxidized LDL with albumin can remove oxidized lipids from the LDL. Watson and colleagues⁵¹ incubated the biologically active coculture-produced mildly oxidized LDL with defatted albumin and then separated the LDL and albumin. After separation, the LDL no longer induced monocyte binding to the endothelial cells. Before incubation, the defatted albumin was unable to induce monocyte binding to the endothelial cells. However, after incubation the albumin induced the endothelial cells to bind monocytes to the same degree as did the coculture-produced mildly oxidized LDL before incubation with the defatted albumin. Thus, the biological activity of the mildly oxidized LDL had been transferred to the defatted albumin. The albumin fraction containing the biological activity was subjected to lipid extraction, and the phospholipids were purified by solid-phase extraction chromatography. These polar phospholipids were resuspended in phosphate-buffered saline and either treated with purified PAF acetylhydrolase or buffer without enzyme. The phospholipids were then reextracted and tested for their ability to induce monocyte binding to endothelial cells. The phospholipids that were extracted from the albumin after incubation with the mildly oxidized LDL were capable of inducing monocyte-binding activity equal to that of the intact mildly oxidized LDL. Treatment of these biologically active phospholipids with purified PAF acetylhydrolase completely abolished their ability to induce binding activity. HPLC analysis revealed that concomitant with the loss in biological activity, the albumin-bound phospholipids, which were seen as a single peak before treatment with the purified PAF acetylhydrolase, were no longer detectable after treatment with the enzyme.

In other experiments, Watson and colleagues⁵¹ demonstrated that pretreatment of LDL with a serine esterase inhibitor that inactivates PAF acetylhydrolase (DFP) resulted in a small (28% to 38%) but reproducible and statistically significant increase in biological activity of the mildly oxidized LDL recovered from the artery wall cocultures.

HDL has been known to be capable of preventing LDL oxidation for some time.^{28 48 56 57 58} Watson and colleagues⁵¹ found that when LDL was isolated from a first set of cocultures and then incubated with native HDL before being presented to a second set of cocultures, there was 80% decrease in the biological activity of the LDL. However, if the HDL was first treated with DFP before incubation with the coculture-produced mildly oxidized LDL, there was no loss in biological activity when the mildly oxidized LDL was presented to a second set of cocultures. Reconstitution of the DFP-treated HDL with purified PAF acetylhydrolase restored the ability of the DFP-treated HDL to destroy the biological activity of the mildly oxidized LDL. In contrast, addition of DFP-treated purified PAF acetylhydrolase to LDL was ineffective in preventing LDL oxidation by the cocultures or in restoring the protective effect of HDL that was previously inactivated by DFP, indicating that the protection of PAF acetylhydrolase derived from its enzymatic activity. Others^{54 59 60} have shown that oxidized lipids destroy PAF acetylhydrolase activity. Thus, oxidized lipids might be expected to mimic the effects of DFP treatment of LDL. Therefore, Watson and colleagues⁵¹ proposed that HDL either (1) removes biologically active oxidized phospholipids from mildly oxidized LDL and PAF acetylhydrolase in the HDL subsequently hydrolyzes them into lysophosphatidylcholine and fatty acid fragmentation products or (2) replenishes the mildly oxidized LDL with active PAF acetylhydrolase, which then hydrolyzes and inactivates the bioactive phospholipids within the mildly oxidized LDL particle, and the products of this hydrolysis are then transferred to HDL.

Watson and colleagues⁶¹ went on to further define the biologically active lipids in mildly oxidized LDL using a strategy of testing for biological activity before and after treatment with HDL-associated enzymes that destroy the biological activity of these lipids and correlating the presence or absence of biological activity with the presence or absence of specific oxidized lipids. In addition to PAF acetylhydrolase, which inhibits apoB modification and conjugated diene formation but has no effect on the formation of TBARS,⁵⁴ HDL also contains an enzyme, paraoxonase, that has been shown by Mackness and colleagues^{62 63 64 65} to be capable of inhibiting the production of lipoperoxides and TBARS in LDL subjected to metal ion oxidation in vitro in the absence of cells. To determine the relative contributions of these two HDL-associated enzymes to destroy the biologically active lipids in mildly oxidized LDL, Watson and colleagues⁶¹ treated HDL with either PMSF (which destroys PAF acetylhydrolase activity but has no effect on paraoxonase activity) or EDTA (which destroys paraoxonase activity but has no effect on PAF acetylhydrolase activity) and compared these individual treatments to HDL treated with both PMSF and EDTA. Inhibiting PAF acetylhydrolase activity in HDL without inhibiting paraoxonase activity partially reduced the ability of HDL to protect against the formation of biologically active mildly oxidized LDL in the cocultures. Similarly, inhibiting paraoxonase activity without inhibiting PAF acetylhydrolase activity only partially reduced the ability of HDL to protect against the formation of biologically active mildly oxidized LDL in the cocultures. However, when both PAF acetylhydrolase activity and paraoxonase activity in HDL were inhibited, the resulting HDL was completely unable to protect against the formation of biologically active mildly oxidized LDL in the cocultures. These experiments indicated that both PAF acetylhydrolase and paraoxonase contribute to the ability of HDL to destroy the biological activity of coculture-produced mildly oxidized LDL.

Having previously shown⁵¹ that incubation of LDL with the cells of the artery wall resulted in increased levels of phospholipid conjugated dienes (measured as UV absorbance at 235 nm), Watson and colleagues⁶¹ went on to show that some oxidized phospholipids also contained an increase in UV absorbance at 270 nm. The increase in 270-nm absorbance was thought to be due to the formation of carbonyl groups or conjugated trienes associated with polyunsaturated fatty acids in the LDL phospholipids. Treatment of coculture-produced mildly oxidized LDL with paraoxonase had little effect on the absorbance of the extracted phospholipids at 235 nm (conjugated dienes). In contrast, paraoxonase treatment of the mildly oxidized LDL greatly reduced the 270-nm-absorbing material (conjugated trienes) associated with the phospholipids. Since incubation with artery wall cell cocultures decreased the abundance of arachidonic acid-containing phospholipids in LDL,⁵¹ Watson and colleagues⁶¹ studied the oxidation of authentic PAPC and found that the oxidized phospholipids that resulted mimicked the ability of those extracted from coculture-produced mildly oxidized LDL to induce endothelial cells to bind monocytes. Unoxidized authentic PAPC did not absorb at either 235 or 270 nm and did not induce endothelial cells to bind monocytes.⁶¹ In contrast, oxidized authentic PAPC absorbed at both 235 and 270 nm and induced endothelial cells to bind monocytes.⁶¹ HPLC fractions were collected under sterile conditions from authentic PAPC that had been oxidized. The lipids collected in these fractions were tested individually for their ability to induce endothelial cells to bind monocytes. The fractions that contained residual unoxidized PAPC (no absorbance at either 235 or 270 nm) did not induce monocyte binding. However, two fractions containing 270-nm-absorbing material dramatically induced monocyte binding to the endothelial cells. Analysis by FAB-MS of a fraction containing unoxidized PAPC that was biologically inactive revealed that (as predicted) the predominant ion was seen at m/z 783. FAB-MS of fractions containing 270-nm-absorbing material that induced monocyte binding produced a mixture of ions with both lower and higher masses than unoxidized PAPC. Those with the higher mass ranged from m/z 815 to 879, with each of these ions being separated by about 16 D (the molecular mass of an oxygen atom). These ions were weak or nonexistent in fractions that did not have biological activity and were not detectable in fractions from unoxidized PAPC. The ions produced from the oxidized PAPC that had a greater mass than that of unoxidized PAPC were destroyed by treatment with purified paraoxonase, whereas the signal of the ion produced by unoxidized PAPC was not altered by paraoxonase treatment. In other experiments, Watson and colleagues⁶¹ extracted the lipids from coculture-produced mildly oxidized LDL and native LDL obtained from the same donor and analyzed these lipids by preparative HPLC. Three fractions were collected that encompassed the region of biological activity, with the second fraction containing 80% of the biological activity. Each fraction was injected sequentially into an electrospray mass spectrometer, and reconstructed selected ion monitoring was conducted to determine the relative abundance of ions with m/z 831 and 847. The signal from each fraction (1 through 3) in native LDL was compared with the respective fraction in the coculture-produced mildly oxidized LDL. The ion with m/z 831 showed a distribution that paralleled the biological activity in the three fractions. Phospholipid-bound F₂-isoprostanes (m/z 831) have been shown to be present in oxidized LDL⁶⁶ and have been shown to possess biological activity on smooth muscle cells.^{67 68} Watson and colleagues⁶¹ concluded that the substrate for paraoxonase was an arachidonic acid-containing phospholipid to which three to four oxygens have been added in an orientation that alters the spectral absorbance of the molecule. They suggested that the presence of the 270-nm absorbance may be the result of the formation of carbonyl groups or the addition of hydroxyl or hydroperoxyl groups at the 5 and 12 carbons of arachidonic acid, forming a conjugated triene structure characteristic of leukotrienes. Any of these compounds may decompose to form short-chain acyl residues at the sn-2 position (eg, 5-oxovalerate), which have been shown to have biological activity⁶⁹ and are substrates for PAF acetylhydrolase.^{52 53} They⁶¹ also noted that it is possible that phospholipids containing both oxygenated and cyclized arachidonic acid, as well as fragmentation products of arachidonic acid (eg, 5-oxovalerate), could represent the biologically active lipids in mildly oxidized LDL. Fig 2, taken from the paper by Watson and colleagues,⁶¹ represents a hypothetical model for the formation of biologically active phospholipids and for the mechanisms by which HDL destroys them. The investigators⁶¹ reported that there was considerable variation in the ratios of PAF acetylhydrolase activity to paraoxonase activity among the 20 different HDL preparations that were studied. It was concluded⁶¹ that the protective effect of HDL may not be dependent on the absolute levels of HDL cholesterol in the blood but rather may be dependent on the abundance of HDL particles that contain protective enzymes relative to the concentration of mildly oxidized LDL in the artery wall. It was postulated⁶¹ that this ratio is determined by both genetic and environmental factors. Genetic determinants were postulated to include plasma levels and isoforms of paraoxonase and PAF acetylhydrolase, while environmental influences were postulated to include factors that augment the accumulation of LDL in the subendothelial space [eg, increased levels of LDL and lipoprotein(a)], as well as factors that increase oxidative stress, such as smoking and diabetes.

View larger version (41K): [in this window] [in a new window]   Figure 2. Hypothetical model for the mechanism by which HDL destroys biologically active lipids in mildly oxidized LDL (MM-LDL). Reactive oxygen species may be formed in the artery wall (A) in areas sequestered from plasma antioxidants by a variety of possible mechanisms, some of which may involve cyclooxygenase, lipoxygenase, inducible nitric oxide synthase, cytochrome P450, mitochondrial respiration, or free iron (B). These oxygen radicals may then seed LDL within the subendothelial space by oxidation of phospholipids in LDL (C). This process involves abstraction of hydrogen atoms from bis-allylic methylenes (D). Addition of molecular oxygen at multiple sites may generate multioxygenated phospholipids (E), which may be substrates for paraoxonase in HDL (F). In cases in which paraoxonase concentrations are low or depleted or in which lipid peroxide levels are excessive, oxidized phospholipids may undergo oxidative fragmentation (G) to form molecules (H) that evoke the characteristic inflammatory responses in endothelial cells induced by MM-LDL. These oxidatively fragmented phospholipids may then be substrates for the second line of defense, PAF acetylhydrolase (PAF-AH) (I). PAF-AH hydrolyzes these biologically active lipids into molecules that do not evoke the characteristic inflammatory responses in endothelial cells induced by MM-LDL. Further oxidative decomposition of lipids in LDL, including cholesteryl esters, leads to the deposition of highly oxidized LDL (J) present in the necrotic core of advanced atherosclerotic lesions. EC indicates endothelial cells; IEL, internal elastic lamina; SMC, smooth muscle cells; NOS, nitric oxide synthase; and ·OOR, lipoperoxides. (Reproduced from The Journal of Clinical Investigation by copyright permission of The American Society for Clinical Investigation: from Watson AD, Berliner JA, Hama SY, La Du BN, Faull KF, Fogelman AM, Navab M. J Clin Invest. 1995;96:2889, Fig 8.)

	HDL: Chameleon-Like Lipoprotein

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
The data reviewed above indicate that HDL in the basal state contains enzymes that can destroy oxidized lipids that mediate a chronic inflammatory response. In this sense, HDL in the basal state is anti-inflammatory. Van Lenten and colleagues⁷⁰ studied the alterations in HDL that were induced by the acute-phase response. The acute-phase response is a systemic reaction to infectious and noninfectious tissue-destructive processes. The hepatic synthesis of a number of plasma proteins commonly referred to as acute-phase reactants is among the most dramatic changes that occur during the acute-phase response.⁷¹ Two of the acute-phase reactants are known to interact with lipoproteins: CRP⁷² and SAA.⁷³ CRP binds to lipoproteins containing apoB, and SAA is found in the circulation almost exclusively associated with HDL. SAA includes a family of proteins that appears in plasma after a variety of stimuli, including surgery, myocardial infarction, and infection, and in chronic inflammatory conditions such as arthritis.⁷⁴ In response to such inflammatory stimuli, SAA expression can be increased by up to 1000-fold as a result of increased gene transcription.⁷⁵ Another acute-phase reactant, ceruloplasmin, ordinarily carries about 95% of plasma copper.⁷⁶ Twenty years ago it was proposed that ceruloplasmin was an independent risk factor in coronary heart disease, based on elevated ceruloplasmin levels in patients with arteriosclerosis.⁷⁷ Ehrenwald and colleagues⁷⁸ subsequently demonstrated that nondegraded ceruloplasmin can mediate LDL oxidation. Additionally, it was shown by others that at an acidic pH, similar to what might be expected in an area of inflammation, LDL oxidation by ceruloplasmin was enhanced.⁷⁹

Using the artery wall cocultures, Van Lenten and colleagues⁷⁰ found that HDL taken from the same subjects before and after surgery (ie, before and during an acute phase) behaved very differently. Before cardiac surgery, HDL prevented the mild oxidation of LDL by the cocultures and hence prevented the induction of the proteins that mediate monocyte transmigration into the subendothelial space of the cocultures. In contrast, the same concentrations of HDL from the same patients taken during the acute-phase response that follows cardiac surgery were not as effective in preventing lipid hydroperoxide formation in cocultures with added LDL. Moreover, the HDL taken during the acute phase actually enhanced LDL-induced monocyte migration into the subendothelial space of the cocultures. Van Lenten and colleagues⁷⁰ demonstrated that the enhanced monocyte transmigration was mediated by increased levels of MCP-1 in the cocultures containing both normal LDL and acute-phase HDL.

Since ceruloplasmin is a known acute-phase reactant and was reported to stimulate LDL oxidation in vitro,⁷⁸ Van Lenten and colleagues⁷⁰ performed Western blot analysis for ceruloplasmin on the HDL taken from the same subjects before and after cardiac surgery. No ceruloplasmin was detected in the HDL taken before surgery, but after surgery the HDL contained substantial quantities of ceruloplasmin. In parallel studies in rabbits performed before and after the injection of croton oil (a standard technique for inducing an acute-phase response in rabbits), no ceruloplasmin was detected in HDL taken from rabbits before injection, but by 48 hours after injection, ceruloplasmin was clearly present in the HDL and persisted for at least 72 hours. Interestingly, no ceruloplasmin was detected before or after injection in either VLDL or LDL. To further define the role of ceruloplasmin, Van Lenten and colleagues⁷⁰ incubated purified ceruloplasmin with HDL in vitro and then reisolated the HDL. The reisolated HDL contained ceruloplasmin and enhanced the LDL-induced increase in MCP-1 production by the artery wall cocultures in contrast to the sham-treated HDL that prevented LDL induction of MCP-1 in the cocultures. These experiments indicated that ceruloplasmin can associate with HDL during an acute-phase reaction in vivo in humans and rabbits and contribute to the conversion of HDL from an anti-inflammatory molecule in the basal state to a proinflammatory molecule in the acute phase.

Van Lenten and colleagues⁷⁰ also measured paraoxonase and PAF acetylhydrolase activities associated with HDL before and during the acute-phase response. They found that after cardiac surgery there was more than a fourfold reduction in HDL-associated paraoxonase activity and an eightfold reduction in PAF acetylhydrolase activity. Incubation in vitro of the acute-phase HDL with purified paraoxonase or purified PAF acetylhydrolase followed by reisolation of the HDL restored the activities of these two enzymes. These experiments suggested that paraoxonase and PAF acetylhydrolase may have been displaced from the HDL during the acute-phase response and that incubation with the purified enzymes restored the proteins and activities. Consistent with this hypothesis was the finding by Van Lenten and colleagues⁷⁰ that apoA-I levels were reduced by 58% in the acute-phase HDL, and the acute-phase HDL was greatly enriched in the acute-phase reactant SAA. To test the hypothesis, Van Lenten and colleagues⁷⁰ incubated native HDL with purified SAA in vitro. After incubation, the HDL was found to have lost 87% of its apoA-I content, 91% of its paraoxonase activity, and 88% of its PAF acetylhydrolase activity (it should be noted that paraoxonase is known to be tightly bound to apoA-I in HDL⁸⁰ ). When HDL, enriched with SAA in vitro, was added, together with LDL, to the artery wall cocultures, there was a stimulation in the LDL-induced monocyte transmigration into the subendothelial space of the coculture. A similar stimulation was observed with HDL taken after cardiac surgery. Therefore, Van Lenten and colleagues⁷⁰ incubated HDL, taken either after cardiac surgery or after in vitro enrichment with SAA, with purified paraoxonase or PAF acetylhydrolase and subsequently added the reisolated HDL to the cocultures containing LDL. The enzyme-repleted HDL prevented LDL-induced monocyte transmigration into the subendothelial space of the coculture.

To further test this hypothesis, Van Lenten and colleagues⁷⁰ returned to the rabbit model of the acute-phase response. After injection of croton oil, they saw a marked increase in serum SAA levels that reached a maximum on day 3 after injection, at which time the SAA levels had increased by 60-fold. The SAA levels began to decline 4 days after injection. The increase and decrease of SAA levels were exactly paralleled by a decrease and increase in the levels of apoA-I associated with HDL. Control rabbits injected with saline did not show any of these changes. Exactly paralleling the fall in apoA-I levels on days 1 to 3 after croton oil injection were decreases in paraoxonase and PAF acetylhydrolase activities. On day 4 after injection of the croton oil, as SAA levels declined and apoA-I levels increased, there was a parallel increase in paraoxonase and PAF acetylhydrolase activities. HDL isolated from rabbits before croton oil injection was very effective in preventing the monocyte transmigration into the subendothelial space of the cocultures that was induced by LDL oxidation. In contrast, after injection of the croton oil and exactly paralleling the increase in SAA levels and the decrease in apoA-I levels, together with the decrease in paraoxonase and PAF acetylhydrolase activities, the rabbit HDL progressively lost (from days 1 to 3 after injection) the ability to protect against the LDL-induced monocyte transmigration. On day 4 after the injection of croton oil, as SAA levels in HDL declined and as apoA-I levels, as well as paraoxonase and PAF acetylhydrolase activities, increased, the rabbit HDL showed a significantly increased ability to protect against the LDL-induced monocyte transmigration.

Taken together, these experiments support the concept that unlike LDL, HDL is chameleon-like, changing its colors (apoproteins and associated enzymes) as the landscape changes (going from the basal state to the acute-phase response and back to the basal state). If HDL protection is largely due to its ability to inhibit or destroy the biologically active lipids in mildly oxidized LDL, the changes in HDL induced by the acute-phase response could result in an increase in the local modification (oxidation) of LDL in the artery wall. Consequently, monocyte infiltration at such sites may increase during an acute-phase response. The major site of monocyte entry into lipid-rich plaques that are prone to rupture is at the shoulder region of lesions,⁸¹ and the major determinant of plaque rupture appears to be the intensity of monocyte infiltration at such sites.^{82 83 84 85} Liuzzo and colleagues⁸⁶ have reported that the concentration of CRP and SAA exceeded the 90th percentile of normal distribution in nearly two thirds of patients with unstable angina and in three fourths of those patients with acute myocardial infarction. These authors demonstrated that the elevated CRP and SAA predicted a poor outcome and may reflect an important inflammatory component in the pathogenesis of acute coronary syndromes. Van Lenten and colleagues⁷⁰ concluded that a marked reduction of paraoxonase and PAF acetylhydrolase and an increase in ceruloplasmin in HDL during the acute phase, with a consequent loss of the protective effect of acute-phase HDL against LDL oxidation, coupled with the resulting monocyte-endothelial interaction, may be contributing factors to the inflammatory reaction in patients with acute coronary syndromes.

	Genetic Control of Both the Yin and the Yang

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Liao and colleagues⁸⁷ reasoned that if oxidized lipids were responsible for the induction of a set of genes whose protein products induced the chronic inflammatory response characteristic of the fatty streak, these genes might be induced in any tissue that accumulated the oxidized lipids. Therefore, they⁸⁷ used the livers of mice as a convenient tissue that allowed for detailed genetic studies. C57BL/6 mice fed an atherogenic diet readily develop fatty streaks in their aortas. In contrast, C3H/HeJ mice do not develop fatty streaks on the same atherogenic diet despite the fact that the two strains develop similar plasma levels of the atherogenic apoB-containing lipoproteins.⁸⁸ On the atherogenic diet, both strains of mice accumulated substantial amounts of total lipid, cholesterol, and triglycerides in their livers, and the levels of these lipids did not differ significantly between the two strains.⁸⁷ However, the fatty streak-susceptible strain (C57BL/6) accumulated significantly more oxidized lipid than did the fatty streak-resistant strain (C3H/HeJ).⁸⁷ The activation of an NFB-like transcription factor was associated with these higher levels of oxidized lipids in the livers of the fatty streak-susceptible mice (C57BL/6), and no activation of the NFB-like transcription factor was seen in the livers of the fatty streak-resistant mice (C3H/HeJ).⁸⁷ Moreover, the transcription of genes such as JE, the mouse homologue of MCP-1, and SAA that contain functional NFB binding sites in their promoters was increased in the fatty streak-sensitive strain (C57BL/6) but not in the fatty streak-resistant strain (C3H/HeJ).⁸⁷ Similarly, the oxidative stress-responsive gene, heme oxygenase, showed increased expression in the fatty streak-susceptible strain (C57BL/6) but not in the fatty streak-resistant strain (C3H/HeJ).⁸⁷

Van Lenten and colleagues⁴⁹ reported that the fatty streak-resistant mice (C3H/HeJ) responded to the atherogenic diet with higher levels of apoferritin and lower intracellular concentrations of free iron than did the fatty streak-susceptible mice (C57BL/6). Additionally, ferritin repressor protein mRNA was not significantly suppressed after 15 weeks on the atherogenic diet in female C57BL/6 mice, which exhibit the most extensive fatty streak formation in their aortas.⁴⁹ In contrast, ferritin repressor protein mRNA was significantly reduced in the fatty streak-resistant mice (C3H/HeJ).⁴⁹

Liao and colleagues⁸⁷ found that injection of mildly oxidized LDL into the tail veins of the fatty streak-susceptible mice (C57BL/6) induced the same set of genes in the livers of these mice as was induced by feeding the atherogenic diet. Berliner and colleagues⁸⁹ demonstrated that injection of the mildly oxidized LDL into the tail veins of these mice also induced monocyte-endothelial interactions in the aorta, as assessed by scanning electron microscopy.

Thus, on an atherogenic diet, the C57BL/6 mice were more susceptible to the accumulation of oxidized lipids in their livers and demonstrated a greater expression of inflammatory genes in their livers and were more susceptible to fatty streak formation in their aortas than were the C3H/HeJ mice. This finding indicated an association between these phenomena but did not prove that the same genes were involved in the susceptibility to inflammatory gene expression in the liver and in the development of fatty streaks in the aorta. To test this possibility directly, Liao and colleagues used recombinant inbred strains derived from the parental strains.⁹⁰ They found that the recombinant inbred strains resulting from the breeding of the parental strains C57BL/6 and C3H/HeJ demonstrated great variance in the expression of inflammatory genes in their livers and the development of aortic fatty streak lesions on the atherogenic diet. These investigators reasoned that if the level of oxidized lipids and/or the activation of the NFB-like transcription factor and the expression of the inflammatory genes cosegregated together as the two genomes were randomly distributed by the cross-breeding of the parental strains, it would be strong evidence that these parameters were genetically linked. The analyses of Liao and colleagues⁹⁰ revealed that indeed there was cosegregation of inflammatory gene expression, activation of an NFB-like transcription factor, and the level of oxidized lipids in the livers of the recombinant inbred strains. There was a positive and statistically significant correlation between the level of expression of inflammatory genes, activation of an NFB-like transcription factor, and the level of oxidized lipids in the livers of the recombinant inbred strains.⁹⁰

Liao and colleagues⁹⁰ then reasoned that if the same genes were involved in the susceptibility to inflammatory gene expression in the liver and in the development of fatty streaks in the aorta, then the increase in aortic lesions and the activation of a hepatic NFB-like transcription factor and the expression of hepatic inflammatory genes on the atherogenic diet should cosegregate among the recombinant inbred strains. The data from Liao and colleagues⁹⁰ indicated that this was indeed the case. There was a positive and statistically significant correlation between the development of fatty streaks in the aorta and the activation of a hepatic NFB-like transcription factor and the expression of hepatic inflammatory genes on the atherogenic diet. These investigators concluded that the data strongly suggested that a major gene contributing to aortic lesion development in this mouse model, previously termed Ath-1 by Paigen and colleagues,⁹¹ may control either the accumulation (as a result of increased formation and/or decreased destruction) of specific oxidized phospholipids in the tissues or the cellular responses to such oxidized phospholipids.

Shih and colleagues⁹² found that HDL isolated from either the fatty streak-sensitive (C57BL/6) or the fatty streak-resistant (C3H/HeJ) mice fed a chow diet, prevented the conversion of native human LDL by human artery wall cell cocultures to biologically active mildly oxidized LDL. When HDL was isolated from these mouse strains after 12 or 15 weeks on an atherogenic diet, the HDL from the fatty streak-resistant mice (C3H/HeJ) continued to be protective against LDL modification by the artery wall cell cocultures. In contrast, the HDL isolated from the fatty streak-sensitive mice (C57BL/6) was no longer protective against LDL modification by the artery wall cell cocultures. Indeed, HDL isolated from C57BL/6 mice fed the atherogenic diet potentiated the biological activity of the mildly oxidized LDL.⁹² Shih and colleagues⁹² reported that paraoxonase activity increased slightly in HDL taken from the fatty streak-resistant (C3H/HeJ) mice on the atherogenic diet compared with paraoxonase activity in HDL taken from the same mice on the chow diet. In contrast, paraoxonase activity in HDL taken from the fatty streak-sensitive (C57BL/6) mice on the atherogenic diet was significantly less than the paraoxonase activity in HDL taken from the same mice on the chow diet. PAF acetylhydrolase activity did not differ significantly in the two mouse strains on either the chow or the atherogenic diet.

Shih and colleagues⁹² cloned the mouse paraoxonase cDNA and found that it had 79.1% and 81.7% identity with the human serum paraoxonase nucleotide and amino acid sequences, respectively. The mouse serum paraoxonase cDNA was used to survey mouse tissues, and it was found that gene expression was restricted to the liver.⁹² On measuring hepatic paraoxonase mRNA levels, it was found that there was a slight increase in mRNA for paraoxonase in livers of the fatty streak-resistant (C3H/HeJ) mice on the atherogenic diet compared with the chow diet.⁹² In contrast, mRNA levels for paraoxonase declined significantly and progressively in the livers of the fatty streak-sensitive (C57BL/6) mice on the atherogenic diet.⁹² Shih and colleagues⁹² then extended their studies to measure hepatic paraoxonase mRNA levels in the recombinant inbred strains derived from the parental C57BL/6 and C3H/HeJ strains that had been studied by Liao and colleagues.⁹⁰ Liao and colleagues⁹⁰ had found a positive and statistically significant correlation between the development of fatty streaks in the aorta and the activation of a hepatic NFB-like transcription factor and the expression of hepatic inflammatory genes on the atherogenic diet among the recombinant inbred strains. Shih and colleagues⁹² also found a significant correlation but the relationship was exactly the inverse of what Liao and colleagues⁹⁰ had found. There was a negative and statistically significant correlation between the development of fatty streaks in the aorta and the level of hepatic paraoxonase mRNA.⁹² Thus, an animal that had low levels of hepatic paraoxonase mRNA had extensive aortic lesions and conversely an animal with high levels of hepatic paraoxonase mRNA had little or no aortic lesions. In addition, there was a significant inverse relationship between hepatic paraoxonase mRNA levels and both the activation of a hepatic NFB-like transcription factor and the expression of hepatic inflammatory genes on the atherogenic diet.⁹²

A positive correlation has been shown to exist in human populations between plasma serum paraoxonase mass and activity levels and HDL levels.^{80 93 94} Also, two recent studies have suggested that polymorphisms of the paraoxonase gene may be associated with coronary heart disease.^{95 96 97} Therefore, Shih and colleagues⁹² examined the paraoxonase mRNA levels and HDL levels in the recombinant inbred strains of mice. They found a highly significant correlation, suggesting that either the synthesis of serum paraoxonase is coordinately regulated with HDL levels or that paraoxonase expression influences HDL levels.

Shih and colleagues⁹² then mapped the mouse serum paraoxonase gene by linkage analysis in an interspecific backcross between Mus spretus and C57BL/6, with C57BL/6 as the recurrent parent. Restriction fragment length variants between the parental strains were identified by Southern hybridization analysis, allowing the segregation of paraoxonase gene alleles to be examined in genetic crosses. The segregation patterns of the serum paraoxonase gene locus were compared with the segregation patterns of more than 350 typed genetic markers that spanned most of the mouse genome among the backcross animals. Linkage was observed with several chromosome 6 markers, including D6Mit1 and D6Mit48. The data for the linkage between the serum paraoxonase gene and the nearest markers indicated a region of mouse chromosome 6 that is syntenic with a region of human chromosome 7 to which the paraoxonase gene has been mapped.⁹⁸ In the recombinant inbred strains created by the crossbreeding of the fatty streak-susceptible C57BL/6 and the fatty streak-resistant C3H/HeJ parental strains, the serum paraoxonase gene locus on chromosome 6 exhibited no evidence of linkage with paraoxonase mRNA levels or with the other traits examined (aortic lesions, the activation of an hepatic NFB-like transcription factor, or the expression of hepatic inflammatory genes on the atherogenic diet). Therefore, the striking difference between the fatty streak-susceptible (C57BL/6) mice and the fatty streak-resistant (C3H/HeJ) mice in paraoxonase expression on the atherogenic diet is not due to genetic variations in the serum paraoxonase gene. Rather, the difference appears to be due to genetic variations in pathways that act to control paraoxonase mRNA levels in trans.⁹²

	Summary

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
The fatty streak develops as a response to specific phospholipids that are carried into the artery wall with LDL and that subsequently become oxidized as a result of exposure to the oxidative waste of the artery wall cells (the Yang). The opposing and calming forces include the antioxidants present within the LDL and in the microenvironments where the LDL is trapped that function to prevent the formation of the oxidized lipids. Other opposing forces include the enzymes (eg, PAF acetylhydrolase) contained in LDL or associated with HDL (eg, paraoxonase and PAF acetylhydrolase) that destroy these biologically active lipids. Together, these opposing forces constitute the Yin. The elements of the Yin and the Yang are inherited, but the environment can modify these forces (eg, in the acute-phase response, the Yin gives way to the Yang as paraoxonase is displaced from HDL, and ceruloplasmin associates with HDL). The balance between the Yin and the Yang determines the response of the artery wall and ultimately the clinical course.

	Studying the Yin and the Yang

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 
Many of the observations on the nature of the Yin and the Yang that have been described above have emanated from the Atherosclerosis Research Unit at UCLA. This unit was made possible by a grant from the American Heart Association, Greater Los Angeles Affiliate, in 1980. Subsequently, this unit has been sustained with some critical institutional support but mainly by investigator-initiated peer-reviewed grants from the National Heart, Lung, and Blood Institute. Two of the founders of the Atherosclerosis Research Unit (Dr Fogelman and Dr Edwards) trained under Prof George J. Popjak (see References 99-108), who still works at the bench in our unit. Professor Popjak has contributed significantly to the work of our group, as indicated in the acknowledgments to many of the papers described above.^{50 51 61 70} Perhaps the most important message of this review is that a real understanding of disease processes takes years, indeed professional lifetimes, and the fostering of an environment that will allow knowledge and commitment to scientific excellence to pass from one generation to the next is ultimately as important as the specific discoveries that result from the research in any given period.

	Selected Abbreviations and Acronyms

 

apo = apolipoprotein CRP = C-reactive protein DFP = diisopropyl fluorophosphate ELAM-1 = endothelial leukocyte adhesion molecule-1 FAB-MS = fast atom bombardment/mass spectrometry HPLC = high-performance liquid chromatography MCP-1 = monocyte chemoattractant protein-1 M-CSF = macrophage-colony stimulating factor NFB = nuclear factor-B PAF = platelet-activating factor PAPC = 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphorylcholine SAA = serum amyloid A protein TBARS = thiobarbituric acid-reactive substances

	Acknowledgments

 
This work was supported in part by the US Public Health Services grants HL 30568 and RR-865, the Laubisch and MK Grey Funds, and the Cigarette and Tobacco Surtax Fund of the State of California through the Tobacco-Related Disease Research Program at the University of California. We thank all the members of the Atherosclerosis Research Unit for their many and continuing contributions. We are especially grateful to Cynthia Harper and Alan Wagner for providing the core support for many of the studies described in this review.

Received January 31, 1996; revision received February 27, 1996;

	References

Top Abstract Introduction Trapping of LDL in... Initial Oxidation of LDL Biological Consequences of... Defining the Biologically Active... HDL: Chameleon-Like Lipoprotein Genetic Control of Both... Summary Studying the Yin and... References

 

1. Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witztum JL. Beyond cholesterol: modifications of low density lipoprotein that increase its atherogenicity. N Engl J Med. 1989;320:915-924.[Medline] [Order article via Infotrieve]
2. Young SG, Parthasarathy S. Why are low density lipoproteins atherogenic? West J Med. 1994;160:153-164.[Medline] [Order article via Infotrieve]
3. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits, I: focal increases in arterial LDL concentrations precede development of fatty streak lesions. Arteriosclerosis. 1989;9:895-907.[Abstract/Free Full Text]
4. Schwenke DC, Carew TE. Initiation of atherosclerotic lesions in cholesterol-fed rabbits, II: selective retention of LDL vs selective increases in LDL permeability in susceptible sites of arteries. Arteriosclerosis. 1989;9:908-913.[Abstract/Free Full Text]
5. Kruth HS. Subendothelial accumulation of unesterified cholesterol: an early event in the atherosclerotic lesion development. Atherosclerosis. 1985;57:337-341.[Medline] [Order article via Infotrieve]
6. Simionescu N, Vasile E, Lupu F, Popescu G, Simionescu M. Prelesional events in atherogenesis: accumulation of extracellular cholesterol-rich liposomes in the arterial intima and cardiac valves of the hyperlipidemic rabbit. Am J Pathol. 1986;123:1109-1125.
7. Frank JS, Fogelman AM. Ultrastructure of the intima in WHHL and cholesterol-fed rabbit aortas prepared by ultra-rapid freezing and freeze-etching. J Lipid Res. 1989;30:967-978.[Abstract]
8. Nievelstein PFEM, Fogelman AM, Mottino G, Frank JS. Lipid accumulation in rabbit aortic intima 2 hours after bolus infusion of low-density lipoprotein: a deep-etch and immunolocalization study of rapidly frozen tissue. Arterioscler Thromb. 1991;11:1795-1805.[Abstract/Free Full Text]
9. Nievelstein-Post P, Mottino G, Fogelman A, Frank J. An ultrastructural study of lipoprotein accumulation in cardiac valves of the rabbit. Arterioscler Thromb. 1994;14:1151-1161.[Abstract/Free Full Text]
10. Camejo G, Olafsson F, Lopez P, Carisson P, Bondjers G. Identification of apoB-100 segments mediating the interaction of low-density lipoproteins with arterial proteoglycans. Arteriosclerosis. 1988;8:368-377.[Abstract/Free Full Text]
11. Smith EB. The relationship between plasma and tissue lipids in human atherosclerosis. Adv Lipid Res. 1974;12:1-49.[Medline] [Order article via Infotrieve]
12. Hoff HF, Heideman CL, Gaubatz JW, Gotto AM Jr, Erickson EE, Jackson RL. Quantification of apolipoprotein B in grossly normal human aorta. Circ Res. 1977;40:56-64.[Abstract/Free Full Text]
13. Hoff HF, Heideman CL, Gotto AM Jr, Gaubatz JW. Apolipoprotein B retention in the grossly normal and atherosclerotic human aorta. Circ Res. 1977;41:684-690.[Abstract/Free Full Text]
14. Fogelman AM, Shechter I, Seager J, Hokom M, Child JS, Edwards PA. Malondialdehyde alteration of low density lipoproteins leads to cholesteryl ester accumulation in human monocyte-macrophages. Proc Natl Acad Sci U S A. 1980;77:2214-2218.[Abstract/Free Full Text]
15. Henricksen T, Mahoney EM, Steinberg D. Enhanced macrophage degradation of low density lipoprotein previously incubated with cultured endothelial cells: recognition by receptor for acetylated low density lipoproteins. Proc Natl Acad Sci U S A. 1981;78:6499-6503.[Abstract/Free Full Text]
16. Morel DW, DiCorleto PE, Chisolm GM. Endothelial and smooth muscle cells alter low-density lipoprotein in vitro by free radical oxidation. Arteriosclerosis. 1984;4:357-364.[Abstract/Free Full Text]
17. Steinbrecher U, Parthasarathy S, Leake DS, Witztum JL, Steinberg D. Modification of low density lipoprotein by endothelial cells involves lipid peroxidation and degradation of low density lipoprotein phospholipids. Proc Natl Acad Sci U S A. 1984;83:3883-3887.
18. Heinecke JW, Rosen H, Chait A. Iron and copper promote modification of LDL by human arterial smooth muscle cells. J Clin Invest. 1984;74:1890-1894.
19. Haberland ME, Fong D, Cheng L. Malondialdehyde altered protein occurs in atheroma of WHHL rabbits. Science. 1988;241:215-218.[Abstract/Free Full Text]
20. Palinski W, Rosenfeld ME, Yla-Herttuala S, Gurtner GC, Socher SS, Butler SW, Parthasarathy S, Carew TE, Steinberg D, Witztum JL. LDL undergoes oxidative modification in vivo. Proc Natl Acad Sci U S A. 1989;86:1372-1380.[Abstract/Free Full Text]
21. Yla-Herttuala S, Palinski W, Rosenfeld S, 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.
22. Haberland ME, Steinbrecher UP. Modified low density lipoproteins: diversity and biologic relevance in atherosclerosis. In: Lusis AJ, Rotter JI, Sparkes RS, eds. Basel, Switzerland: S Karger; Molecular Genetics of Coronary Artery Disease. 1992:35-61.
23. Witztum JL, Steinberg D. Role of oxidized low density lipoprotein in atherogenesis. J Clin Invest. 1991;88:1785-1792.
24. Parthasarathy S. Modified Lipoproteins in the Pathogenesis of Atherosclerosis. Austin, Tex: RG Landes Co; 1994:91-119.
25. Parthasarathy S. Mechanism(s) of cell-mediated oxidation of low density lipoprotein. Free Radicals in the Environment, Medicine and Toxicology. Nohl H, Esterbauer H, Evans CR, eds. London, UK: Richelieu Press; 1994:163-179.
26. Witztum JL. The oxidation hypothesis of atherosclerosis. Lancet. 1994;344:793-795.[Medline] [Order article via Infotrieve]
27. Chisolm GM. Antioxidants and atherosclerosis: a current assessment. Clin Cardiol. 1991;14:125-130.[Medline] [Order article via Infotrieve]
28. Navab M, Imes SS, H