Histological and immunofluorescent studies on the evolution of the human atheromatous plaque

Differences in the distribution of versican, decorin, and biglycan in atherosclerotic human coronary arteries

The distributions of versican, biglycan, and decorin have been examined in segments of normal and atherosclerotic human coronary arteries using antibodies directed against the core proteins of these macromolecules. Versican immunostaining was prominent throughout the extracellular matrix (ECM) in regions of the vessels that contained abundant smooth-muscle cells, such as in diffuse intimal thickenings, fibrous caps, and in zones of loose, myxoid connective tissue. Versican also was present in smooth-muscle-rich thrombi and at borders of the lipid-rich cores of advanced atherosclerotic lesions. Biglycan immunostaining was observed in diffuse intimal thickenings, fibrous caps, and myxoid areas, but, unlike versican, it was abundant in the lipid-rich core of advanced plaques. However, biglycan immunostaining was absent in smooth-muscle cell-enriched thrombi. Decorin immunostaining paralleled biglycan immunostaining except that it was conspicuously absent in the myxoid areas of the plaque and markedly reduced in diffuse intimal thickenings. Both biglycan and decorin immunostaining were consistently associated with some of the microvessels in the thrombi and in advanced atherosclerotic plaques. Taken together, these results indicate that specific proteoglycans distribute to topographically defined regions of normal and atherosclerotic human coronary arteries and that these different distributions may indicate a diversity of functions in normal and pathologic processes of the arterial wall.

IMMUNOHISTOCHEMICAL STUDY OF HUMAN XANTHOMA

Specimens of human tuberous xanthomas obtained from two hyper-beta-lipoproteinemic patients were prepared with HE and Sudan III stain and by direct immunofluorescence with fluorescein isothiocyanate labeled antihuman beta-lipoprotein rabbit serum. In some histiocytes and foam cells in early stages, lipids were stained in granular patterns and specific fluorescence was observed. In some foam cell nests, specific fluorescence was weak or absent, and orange-colored lipid autofluorescence was present. The foam cells in such nests were probably late foam cells and corresponded to the large cytoplasmic foam cells stained densely and amorphously for lipids. It was concluded that at early stages of xanthoma formation, serum lipoproteins were incorporated into dermal histiocytes and early foam cells. The apoproteins of the incorporated lipoproteins were probably degraded and lipid residues accumulated in the cells. Specimens of tuberous xanthoma covered by plane xanthoma showed lipids and bright specific fluorescence at the basement membrane zone. A faint fluorescence was observed in the upper dermis and even in parts of the epidermal intercellular spaces.

Ultrastructural localization of adiponectin protein in vasculature of normal and atherosclerotic mice

Adiponectin, adipose-specific secretory protein, abundantly circulates in bloodstream and its concentration is around 1000-fold higher than that of other cytokines and hormones. Hypoadiponectinemia is a risk factor for atherosclerosis. There is little or no information on ultrastructural localization of adiponectin in the vasculature. Herein we investigated the localization of vascular adiponectin in the aorta using the immunoelectron microscopic technique. In wild-type (WT) mice, adiponectin was mainly detected on the luminal surface membrane of endothelial cells (ECs) and also found intracellularly in the endocytic vesicles of ECs. In the atherosclerotic lesions of apolipoprotein E-knockout (ApoE-KO) mice, adiponectin was detected in ECs, on the cell surface membrane of synthetic smooth muscle cells, and on the surface of monocytes adherent to ECs. Changes in adiponectin localization within the wall of the aorta may provide novel insight into the pathogenesis of atherosclerosis.

Family study of high density lipoprotein cholesterol and the relation to age and sex. The Tromso Heart Study

A family study of serum high density lipoprotein (HDL) cholesterol and total serum cholesterol concentration has been undertaken, and the relation to age, sex, cigarette smoking, physical activity and familial occurrence of myocardial infarction (MI) was examined. HDL cholesterol was determined in 251 females and 194 males and total serum cholesterol in 677 females and 657 males, all aged 0-49 years. With respect to HDL cholesterol, significant sex differences were observed both in absolute level and in age-related change. A negative correlation between HDL cholesterol and total serum cholesterol was observed in all age groups except females aged 0-19 years, supporting the hypothesis of HDL as a ""clearing'' lipoprotein. HDL cholesterol showed a positive correlation only in pairs of first-degree relatives involving the mother and in sib-sib paris of the same sex. On the other hand, for serum cholesterol a positive correlation was found among all family members, although significantly higher between first-degree relatives than between spouses. No relation was found between cigarette smoking, physical activity of familial occurrence of MI and the HDL cholesterol or total serum cholesterol concentrations. In accordance with the ""HDL hypothesis'', the present finding could partly explain the higher incidence of ischaemic heart disease (IHD) in males than in females, and partly also the high risk which is transmitted from women with IHD to their first-degree relatives.

Varied low density lipoprotein binding property of proteoglycans synthesized by vascular smooth muscle cells cultured on extracellular matrix

Earlier we showed that the extracellular matrix (ECM) secreted by vascular cells modulated proteoglycan synthesis by vascular smooth muscle cells in culture and altered the proteoglycan characteristics. In this study, we tested the hypothesis that these ECM-mediated alterations increased the affinity of the proteoglycans for plasma low density lipoprotein (LDL). Newly synthesized proteoglycans were isolated from smooth muscle cells cultured on the ECMs secreted by vascular endothelial cells, smooth muscle cells, or THP-1 macrophages and their binding affinity for LDL determined. Proteoglycans from all cultures contained sub-fractions that bound LDL with low and high affinity. However, compared with the cells plated on the endothelial cell ECM, the cells plated on the smooth muscle cell ECM and macrophage ECM synthesized significantly more high affinity proteoglycans. Removal of collagen, elastin, and chondroitin sulfates from the smooth muscle cell ECM and chondroitin sulfates from the macrophage ECM increased the production of high affinity proteoglycans by 15-22%. However, neutralization of fibronectin from both ECMs decreased the high affinity proteoglycans by 20%. Removal of matrix-bound growth factors had no effect on the synthesis of high affinity proteoglycans. Compared with the low affinity proteoglycans, the high affinity proteoglycans were larger, more sulfated and contained higher proportions of chondritin sulfate, dermatan sulfate, and N-sulfated heparan sulfate chains. These results suggest that the ECM-mediated alterations in vascular smooth muscle cell proteoglycans may lead to increased deposition of LDL in the arterial wall.

Myeloperoxidase and hypochlorite, but not copper ions, oxidize heparin-bound LDL particles and release them from heparin

A key factor in atherosclerosis is the retention of low density lipoprotein (LDL) in the extracellular matrix of the arterial intima, where it binds to the negatively charged glycosaminoglycan chains of proteoglycans. Oxidation may lead to modification of the lysine residues of apolipoprotein B-100 of LDL, which normally mediate the binding of LDL to glycosaminoglycans. Here, we studied whether various modes of oxidation can release LDL from heparin, a glycosaminoglycan with a strong negative charge, in vitro. We found that copper ions were unable to oxidize heparin-bound LDL particles because of their redox inactivation by the glycosaminoglycans. In contrast, myeloperoxidase and hypochlorite, a product of myeloperoxidase, were able to oxidize heparin-bound LDL, and this oxidation led to the release of the oxidized particles from heparin. When the released LDL particles were compared with the residual heparin-bound LDL particles, the released particles were more electronegative and contained more modified lysine residues than did the particles that remained bound. Because human atherosclerotic lesions contain catalytically active myeloperoxidase and (lipo)proteins modified by hypochlorite, the results suggest that myeloperoxidase-secreting monocytes/macrophages in the arterial intima can oxidize and extract LDL from the extracellular matrix with ensuing uptake by the macrophages of the oxidized and released LDL, with eventual formation of foam cells.

Expression and localization of the proteoglycan decorin during the progression of cholesterol induced atherosclerosis in Japanese quail: implications for interaction with collagen type I and lipoproteins

The temporal and spatial distribution, and relative levels of the proteoglycan decorin and collagen type I were examined during the progression of atherosclerosis in the dorsal aortas of Japanese quail selected for cholesterol induced atherosclerosis. The quail were placed on either a control or 0.5% added cholesterol diet at approximately 16 weeks of age. Dorsal aortas were collected at 1- or 2-week intervals over a 15-week period after initiating cholesterol feeding. Biochemical analysis for decorin and collagen type I showed that both increased in the cholesterol-fed birds compared to control-fed birds beginning at 9 weeks and continued through the duration of the study. Through immunohistochemical staining for decorin and collagen type I, the spatial localization of decorin and collagen type I in control and less severe plaques in cholesterol-fed birds was most prominent in the arterial adventitia. However, in severe atherosclerotic plaques, decorin was localized in foam cell regions and collagen type I was found surrounding the foam cell regions where decorin accumulated. These results demonstrated a localization of decorin in the core of the atherosclerotic plaque foam cell region with collagen type I being located on the plaque surface.

Association of apo B lipoproteins with arterial proteoglycans: pathological significance and molecular basis

Retention of apo B-100 lipoproteins, low density lipoprotein (LDL) and probably lipoprotein(a), Lp(a), by intima proteoglycans (PGs) appears to increase the residence time needed for their structural, hydrolytic and oxidative modifications. If the rate of LDL entry exceeds the tissue capacity to eliminate the modified products, this process may be a contributor to atherogenesis and lesion advancement. LDL binds to PGs of the intima, by association of specific positive segments of the apo B-100 with the negatively-charged glycosaminoglycans (GAGs) made of chondroitin sulfate (CS), dermatan sulfate (DS) and probably heparan sulfate (HS). Small, dense LDL has a higher affinity for CS-PGs than large buoyant particles, probably because they expose more of the segments binding the GAGs than larger LDL. PGs cause irreversible structural alterations of LDL that potentiate hydrolytic and oxidative modifications. These alterations also increase LDL uptake by macrophages and smooth muscle cells. These in vitro data suggest that part of the atherogenicity of LDL may depend on its tendency to form complexes with arterial PGs in vivo. Ex vivo results support this hypothesis. Subjects with coronary heart disease have LDL with significantly higher affinity for arterial PGs. This is also a characteristic of subjects with the atherogenic lipoprotein phenotype, with high levels of small, dense LDL. The LDL-PG affinity, however can be modified by dietary or pharmacological interventions that change the composition and size of LDL. Lesion-prone intima contain PGs with a high affinity for LDL. Increased LDL entrapment at these sites may be a key step in a cyclic atherogenic process.

Identification of the principal proteoglycan-binding site in LDL. A single-point mutation in apo-B100 severely affects proteoglycan interaction without affecting LDL receptor binding

The subendothelial retention of LDLs through their interaction with proteoglycans has been proposed to be a key process in the pathogenesis of atherosclerosis. In vitro studies have identified eight clusters of basic amino acids in delipidated apo-B100, the protein moiety of LDL, that bind the negatively charged proteoglycans. To determine which of these sites is functional on the surface of LDL particles, we analyzed the proteoglycan-binding activity of recombinant human LDL isolated from transgenic mice. Substitution of neutral amino acids for the basic amino acids residues in site B (residues 3359-3369) abolished both the receptor-binding and the proteoglycan-binding activities of the recombinant LDL. Chemical modification of the remaining basic residues caused only a marginal further reduction in proteoglycan binding, indicating that site B is the primary proteoglycan-binding site of LDL. Although site B was essential for normal receptor-binding and proteoglycan-binding activities, these activities could be separated in recombinant LDL containing single-point mutation. Recombinant LDL with a K3363E mutation, in which a glutamic acid had been inserted into the basic cluster RKR in site B, had normal receptor binding but interacted defectively with proteoglycans; in contrast, another mutant LDL, R3500Q, displayed defective receptor binding but interacted normally with proteoglycans. LDL with normal receptor-binding activity but with severely impaired proteoglycan binding will be a unique resource for analyzing the importance of LDL- proteoglycan interaction in atherogenesis. If the subendothelial retention of LDL by proteoglycans is the initial event in early atherosclerosis, then LDL with defective proteoglycan binding may have little or no atherogenic potential.

Effects of contraceptive estrogen and progestin on the atherogenic potential of plasma LDLs in cynomolgus monkeys

This study was designed to determine the effect of oral contraceptive treatment (estrogen and progestin), alone or in combination, on LDL composition and atherogenic potential in cynomolgus monkeys fed an atherogenic diet. Groups (n = 8 each) of monkeys were untreated (control) or treated with ethinyl estradiol (EE), levonorgestrel (LNG); or triphasic oral contraceptive (EE+LNG) for 1.5 years before plasma LDLs were isolated for characterization. Total plasma cholesterol concentrations were unaffected by the treatments. LDL particle size (measured as LDL molecular weight, g/mumol) was significantly smaller, in the EE (4.61 +/- 0.09) and EE+LNG (4.43 +/- 0.09) treatment groups compared with the control (4.99 +/- 0.09) or LNG (5.29 +/- 0.17) groups and contained fewer molecules of free and esterified cholesterol. Both the EE and EE+LNG groups had significantly less cholesterol and apolipoprotein B distributed in the d = 1.015 to 1.025 g/mL subfraction and correspondingly more in the d = 1.025 to 1.035 g/mL subfraction of LDL compared with the control and LNG groups. The apolipoprotein E content (molecules/particle) of LDL was significantly less in the EE (0.35 +/- 0.1) and EE+LNG (0.28 +/- 0.1) groups compared with the control (0.86 +/- 0.2) and LNG (0.99 +/- 0.2) groups, and this trend was apparent in all three LDL subfractions. The atherogenic potential of LDL was tested using an in vitro binding assay to arterial proteoglycans. Twice as much LDL bound to arterial proteoglycans in the LNG group (11.3 +/- 1.8% of total LDL cholesterol in the incubation) compared with the control (6.4 +/- 1.9%), EE (5.5 +/- 1.5%), or EE+LNG (5.2 +/- 1.2%) groups. We conclude that EE and EE+LNG treatment alters the composition of LDL toward a less atherogenic particle that is smaller and more dense, contains less cholesterol and less apolipoprotein E, and is less reactive with arterial proteoglycans compared with LNG treatment. The inclusion of EE in the triphasic oral contraceptive treatment was sufficient to negate the potentially atherogenic effects of LNG on LDL composition.

Cellular consequences of the association of apoB lipoproteins with proteoglycans. Potential contribution to atherogenesis

Many of the discussed results come from empirical experiments performed with in vitro models whose relevance to the complex environment of the intima is limited. However, they are consistent with the line of reasoning that intima PGs interact specifically with apoB lipoproteins and contribute to their retention. This could provide the residence time and the initial alterations of the lipoproteins that favor their further modifications by oxidative processes and hydrolytic enzymes. Products of such modifications, and the modified particles, may be stimuli for changes in the functionality of endothelium, smooth muscle cells, and macrophages. The focal synthesis of PGs with high affinity for apoB lipoproteins could make the phenomena chronic. Clinical and laboratory studies indicate that dense LDL, poor in surface polar lipids, is associated with an atherogenic phenotype. Particles with these properties may contribute to the disease via its high affinity for arterial PGs. This affinity can be modulated by diet, lifestyle, and lipid-lowering drugs.

The proteoglycan decorin links low density lipoproteins with collagen type I

Decorin is a small dermatan sulfate-rich proteoglycan which binds to collagen type I in vitro and in vivo. In atherosclerotic lesions the contents of low density lipoprotein (LDL), decorin, and collagen type I are increased, and ultrastructural studies have suggested an association between LDL and collagen in the lesions. To study interactions between LDL, decorin, and collagen type I, we used solid phase systems in which LDL was coupled to a Sepharose column, or in which LDL, decorin, or collagen type I was attached to microtiter wells. The interaction between LDL and decorin in the fluid phase was evaluated using a gel mobility shift assay. We found that LDL binds to decorin by ionic interactions. After treatment with chondroitinase ABC, decorin did not bind to LDL, showing that the glycosaminoglycan side chain of decorin is essential for LDL binding. Acetylated and cyclohexanedione-treated LDL did not bind to decorin, demonstrating that both lysine and arginine residues of apoB-100 are necessary for the interaction. When collagen type I was attached to the microtiter plates, only insignificant amounts of LDL bound to the collagen. However, if decorin was first allowed to bind to the collagen, binding of LDL to the decorin-collagen complexes was over 10-fold higher than to collagen alone. Thus, decorin can link LDL with collagen type I in vitro, which suggests a novel mechanism for retention of LDL in collagen-rich areas of atherosclerotic lesions.

Residence time of low-density lipoprotein in the normal and atherosclerotic rabbit aorta

Previous results from this laboratory found that the arterial low-density lipoprotein (LDL) residence time in lesion-prone aortic sites was longer in hyperlipidemic rabbits before lesion formation than in the corresponding sites in normolipidemic rabbits. The calculation of residence time in the previous study assumed that the arterial wall was homogeneous; the present study reexamines the issue using a method that does not require such an assumption. The concentration of radiolabeled arterial LDL was measured in New Zealand White rabbits killed at several different times (0.5 to 72 hours) after injection of labeled LDL. Using a stochastic analysis, arterial LDL residence time was calculated from the pooled labeled arterial LDL measurements from these rabbits. In these studies, the arterial LDL residence times in normolipidemic and hyperlipidemic rabbits before lesion formation were similar in both the lesion-prone and -resistant sites. However, immediately upon development of early fatty streak lesions, the arterial LDL residence time increased dramatically. After only 16 days of cholesterol feeding, the residence time was 10 times longer in the lesioned aortic arch compared with similar tissue from normolipidemic rabbits (4 to 45 hours). After 21 days of cholesterol feeding, the residence time of LDL in the lesioned aortic arch increased to > 25-fold that of normolipidemic tissue. Similar results were observed in the lesioned tissue of the abdominal branchings. This early retention of LDL suggests that significant changes are taking place within the arterial wall during this critical stage of early lesion development.

Effects of hormone replacement modalities on low density lipoprotein composition and distribution in ovariectomized cynomolgus monkeys

This study was designed to determine the effect of several hormone replacement therapies on LDL size, density, heterogeneity, and composition in surgically postmenopausal cynomolgus monkeys fed an atherogenic diet. Groups (n = 5 each) of ovariectomized cynomolgus monkeys were untreated (control), or treated with conjugated equine estrogens, medroxyprogesterone acetate (progesterone), combined estrogen-progesterone, or tamoxifen for 9 weeks. There were no differences among treatment groups in total plasma, LDL, or HDL cholesterol or triglyceride concentrations. Plasma LDL were isolated by ultracentrifugation and size exclusion chromatography and subfractionated by density gradient centrifugation for subsequent chemical analysis. Estrogen treatment was associated with significantly smaller (measured as LDL molecular weight, 3.9 +/- 0.2 g/mu mol) and denser plasma LDL (1.034 g/ml peak density) compared with control (4.5 +/- 0.1 g/mu mol; 1.030 g/ml peak density) or progesterone-treated animals (4.6 +/- 0.2; 1.029 g/ml peak density). LDL from the estrogen group were relatively enriched in protein and triglyceride and poor in cholesteryl ester and apolipoprotein F (apoE) compared to the control group. Triglyceride enrichment with estrogen treatment occurred predominantly in the lighter, larger LDL subfractions (d = 1.015-1.025 g/ml), which were reduced in concentration (26 +/- 10 mg cholesterol/dl) compared to control (61 +/- 19 mg/dl) or progesterone treated animals (67 +/- 16 mg/dl). Combined estrogen-progesterone or tamoxifen treatment resulted in changes in LDL that followed the same trend as those observed with estrogen treatment. We conclude that short-term estrogen treatment of ovariectomized cynomolgus monkeys results in changes in plasma LDL size, density, and composition while having no apparent effect on overall plasma lipid concentrations.

Alteration of glycosaminoglycans induced by cadmium in cultured vascular smooth muscle cells

Alteration of glycosaminoglycans (GAGs) in cultured bovine aortic smooth muscle cells after exposure to cadmium was investigated. It was revealed that cadmium increased the accumulation of GAGs metabolically labeled with [3H]glucosamine but decreased that with [35S]sulfate in the cell fraction, the cell surface fraction and the medium fraction. This suggested that cadmium stimulated the biosynthesis of GAGs but inhibited their sulfation in the cells. A similar alteration was observed in cadmium-treated human aortic smooth muscle cell layer. Of tested cations including cadmium, bismuth, cobalt, copper, lead, manganese, nickel and zinc, only cadmium stimulated [3H]glucosamine incorporation, with a strong inhibition of the [35S]sulfate incorporation in the bovine cells. Characterization of bovine smooth muscle GAGs showed that the cadmium-induced increase in the [3H]glucosamine incorporation was mainly observed in heparan sulfate; the inhibition of the [35S]sulfate incorporation occurred non-selectively. Cadmium accumulated in bovine vascular smooth muscle cells in a dose-dependent manner with an increase in the leakage of lactate dehydrogenase into the medium. The present data suggest that vascular smooth muscle cells respond to the cytotoxicity of cadmium and promote the GAG synthesis with a reduction of their sulfation. It is postulated that this response may be a defensive one to the damage of the vascular tissue caused by cadmium but would be a component of the metal-induced atherosclerosis.

Dietary polyunsaturated fat decreases interaction between low density lipoproteins and arterial proteoglycans

Polyunsaturated dietary fat (n-3 and n-6) results in less atherosclerosis in monkeys compared to lard (Parks, J.S., Kaduck-Sawyer, J., Bullock, B.C., and Rudel, L.L., Arteriosclerosis 10, 1102-1112; Rudel, L.L., Parks, J.S., Johnson, F.L., and Babiak, J., J. Lipid Res. 27, 465-474, 1986). We hypothesized that this was due, in part, to a decreased reactivity of low density lipoproteins (LDL) with arterial proteoglycans (PG). To test this hypothesis, cynomolgus monkeys were fed diets containing lard, safflower oil (n-6 polyunsaturated; Poly), menhanden fish oil (FO), or oleic acid-rich safflower oil (oleinate; Mono) for 14 mon, and plasma LDL were isolated and characterized. Several properties of LDL thought to be important in the interaction of LDL with arterial PG were measured including LDL particle size, chemical composition, sialic acid content, density distribution, apolipoprotein E (apoE) content and cholesteryl ester transition temperature. Plasma LDL cholesterol concentrations (mg/dL) after 14 mon of diet consumption averaged (mean +/- SEM): FO (366 +/- 45), Lard (352 +/- 27), Poly (279 +/- 24), and Mono (230 +/- 43). The composition of LDL was similar among diet groups except that FO LDL were relatively depleted of cholesteryl ester and enriched in protein and were smaller in size. LDL sialic acid content was similar among diet groups (4.5-5.0 micrograms/mg LDL protein). The LDL apoE/B molar ratio, a measure of the apoE content per LDL particle averaged: Mono (3.0 +/- 1.0), Poly (2.0 +/- 0.1), Lard (1.8 +/- 0.5), and FO (1.0 +/- 0.2).(ABSTRACT TRUNCATED AT 250 WORDS)

Evidence for loss of apo B from LDL in human atherosclerotic lesions: extracellular cholesteryl ester lipid particles lacking apo B

Previous studies have demonstrated the accumulation of low density lipoprotein (LDL) in the extracellular spaces of the intima of normal and atherosclerotic human vessels. In this study we have assessed the degree of colocalization in vessels of apolipoprotein B (apo B), the major protein of LDL, with cholesteryl ester, the predominant lipid of LDL. Apo B was detected immunohistochemically and cholesteryl ester was detected after its enzymatic hydrolysis and staining with the fluorescent probe, filipin. Most normal intima showed apo B staining without associated cholesteryl ester staining. This result would be expected with LDL having intact apo B; intact apo B interferes with hydrolysis and filipin staining of LDL cholesteryl ester. Fatty streaks and fibrous plagues showed regions of congruent apo B and cholesteryl ester staining in the extracellular space, suggesting fragmentation of apo B without loss of its immunoreactivity. Still other areas of lesions showed cholesteryl ester staining in the extracellular space without apo B staining. This staining pattern suggests loss of apo B from LDL leaving only the cholesteryl ester-rich core of LDL. Progressive loss of apo B from LDL can explain the patterns of apo B and cholesteryl ester colocalization that occur in vessel wall intima. The distribution of these patterns in normal and atherosclerotic lesions suggests that loss of apo B from the cholesteryl ester core of LDL is associated with lesion development.

Relative significance of endothelium and internal elastic lamina in regulating the entry of macromolecules into arteries in vivo

A role for the internal elastic lamina (IEL), which separates the intima and media of an artery wall, as a restrictive barrier to macromolecular movement has been suggested in atherosclerotic lesion development or restenosis during angioplasty. The permeability coefficient of the IEL, however, has never been quantified in unperturbed vessels in vivo. Using a newly developed technique, we measured the concentration distributions in both intima and media of cationic (pI approximately 8.5) and anionic (pI approximately 6.3) isozymes of the 44-kD macromolecule horseradish peroxidase (HRP). Two mathematical models of arterial wall transport differing in their resolution of the intima were required to simulate the concentration distribution data and to estimate the parameters of interest. Optimal estimates of the permeability coefficients of the endothelium (PE) and IEL (PIEL) to HRP were determined by the best least-squares fit of the two models to experimental data. These estimates (anionic: PE = 0.050 +/- 0.021 microns/min, PIEL = 0.146 +/- 0.082 microns/min, n = 8; cationic: PE = 0.034 +/- 0.018 microns/min, PIEL = 0.110 +/- 0.047 microns/min, n = 8) indicate that the IEL is responsible for approximately 25% (anionic, 26 +/- 9%; cationic, 25 +/- 13%) of the resistance to HRP transport from the blood into the arterial media. Although both parameters were less for the cationic preparation, the differences were not significant, and the relative role of the IEL was similar for both molecules.(ABSTRACT TRUNCATED AT 250 WORDS)

Recent observations on the role of hemostatic determinants in the development of the atherothrombotic plaque

Recent evidence suggests that hemostatic determinants play a major role in the evolution of the atherothrombotic plaque. Platelets can serve as cholesterol donors for macrophages, thereby facilitating foam cell formation. Lipoprotein(a) inhibits fibrinolysis and may also contribute to atherogenesis by serving as a ligand for the scavenger receptor. By complexing with fibrin(ogen) in atheromatous lesions, lipoprotein(a) attenuates clearance of this protein, promoting atherogenesis and vascular dysfunction. These observations suggest that thrombotic determinants are critical for the development of the atheromatous plaque, and may guide the appropriate selection of potential therapeutic options in the future.

Lipid lowering and plaque regression. New insights into prevention of plaque disruption and clinical events in coronary disease

The consensus of evidence from angiographic trials demonstrates both coronary artery and clinical benefits from lowering of lipids by a variety of regimens. The findings of reduced arterial disease progression and increased regression have been convincing but, at best, modest in their magnitude. For example, among those treated intensively in FATS, the mean improvement in proximal stenosis severity per patient was < 1% stenosis, and only 12% of all lesions showed convincing regression. In view of these modest arterial benefits, the associated reductions in cardiovascular events have been surprisingly great. For example, coronary events were reduced 75% in FATS; this was entirely a result of a 93% reduction in the likelihood that a mildly or moderately diseased arterial segment would experience substantial progression to a severe lesion at the time of a clinical event. We believe that the magnitude of the clinical benefit is best explained in terms of this observation, according to the following lines of reasoning. Clinical events most commonly spring from lesions that are initially of mild or moderate severity and then abruptly undergo a disruptive transformation to a severe culprit lesion. The process of plaque fissuring, leading to plaque disruption and thrombosis, triggers most clinical coronary events. Fissuring is predicted by a large accumulation of core lipid in the plaque and by a high density of lipid-laden macrophages in its thinned fibrous cap. Lesions with these characteristics constitute only 10-20% of the overall lesion population but account for 80-90% of the acute clinical events. In the experimental setting, normalization of an atherogenic lipid profile substantially decreases the number of lipid-laden intimal macrophages (foam cells) and depletes cholesterol from the core lipid pool. In the clinical setting, intensive lipid lowering virtually halts the progression of mild and moderate lesions to clinical events. Thus, the reduction in clinical events observed in these trials appears to be best explained by the relation of the lipid and foam cell content of the plaque to its likelihood of fissuring and by the effects of lipid-lowering therapy on these "high-risk" features of plaque morphology. The composite of data presented here supports the hypothesis that lipid-lowering therapy selectively depletes (regresses) that relatively small but dangerous subgroup of fatty lesions containing a large lipid core and dense clusters of intimal macrophages. By doing so, these lesions are effectively stabilized and clinical event rate is accordingly decreased.

Tumor necrosis factor alpha-induced alteration of glycosaminoglycans in cultured vascular smooth-muscle cells

We studied alteration of glycosaminoglycans (GAGs) induced by recombinant human tumor necrosis factor alpha (rhTNF alpha) in vascular smooth-muscle cells from bovine aorta in a culture system. It was found that rhTNF alpha at 10 ng/ml and below significantly increased the incorporation of [35S]sulfate (35S) but conversely decreased that of [3H]glucosamine (3H) into GAGs in the trypsinate fraction of the cell layer after a 24-h incubation. These results suggested that rhTNF alpha reduced the formation and/or the anchorage of sugar chains in the cell layer but enhanced their sulfation in whole GAG synthesis by the cells. In results, the ratio of 35S to 3H in the GAGs was markedly increased. This increase occurred after 24 h and longer when the cells were treated with 1.0 ng/ml rhTNF alpha. The TNF alpha-induced alteration of the incorporation of both 35S and 3H was completely blocked by anti-rhTNF alpha antibody. Other cytokines including recombinant human interleukin-1 beta and -6, and platelet-derived growth factor failed to alter the ratio of 35S to 3H in the GAGs of the trypsinate fraction of the cell layer. In cultured vascular endothelial cells from bovine aorta, however, rhTNF alpha at 1.0 ng/ml significantly decreased the incorporation of both 35S and 3H into GAGs of both the trypsinate fraction and the medium; the ratio of 35S to 3H was not changed. Characterization of GAGs in vascular smooth muscle cell trypsinate fraction revealed that rhTNF alpha at 10 ng/ml induced (i) no change of the incorporation of 3H in the hyaluronate fraction, (ii) a marked increase in the incorporation of 35S and no change of that of 3H in chondroitin sulfates (A plus C) fraction, (iii) a significant decrease in the incorporation of both 35S and 3H in the heparan sulfate fraction, and (iv) no change of the incorporation of 35S and a marked decrease in that of 3H in the dermatan sulfate fraction. In the medium, rhTNF alpha also induced various changes of GAGs. It was therefore concluded that TNF alpha may have a capacity of inducing a qualitative change of vascular smooth-muscle cell GAGs, which may be involved in the vascular pathology such as atherosclerosis.

Atherosclerosis in a rhesus monkey with genetic hypercholesterolemia and elevated plasma Lp(a)

A female rhesus monkey with a marked elevation of total plasma cholesterol LDL and Lp(a) while on a low-fat, low-cholesterol diet, died at 22 years of age. Her spontaneous hypercholesterolemia was related to a genetically determined LDL receptor deficiency (Scanu, A.M. et al., J. Lipid Res., 29 (1988) 1671). Autopsy revealed grossly visible multifocal to diffuse raised yellow plaques predominantly in the aorta and, to a lesser extent, in the coronary arteries. Microscopically, the plaques in the aorta and in the coronary arteries showed heavy lipid deposition. Some had features seen in advanced human atherosclerotic plaques, including a fibrous cap and a necrotic core. Immunohistochemical staining showed a co-localization of apo(a) with apo B in lesion sites, a pattern seen frequently in advanced human atherosclerotic plaques. Evidence of fibrinogen/fibrin in the plaque areas was also seen, but was not co-localised with either Lp(a) or apo B. This monkey developed progressive atherosclerosis which was not induced by diet, but rather was dependent on the LDL receptor deficiency with a possible contribution by the elevated plasma levels of Lp(a).

Inhibition by serum components of oxidation and collagen-binding of low-density lipoprotein

Low-density lipoprotein (LDL) is oxidized by cellular and noncellular mechanisms, both leading to an increased binding to collagen. We have investigated the effect of serum on lipid peroxidation, apoprotein oxidation and the binding of oxidized apoprotein to collagen. During noncellular oxidation, lipoprotein-deficient serum strongly inhibited all three processes. The serum fraction of M(r) > 100,000 was equally inhibitory; this effect was not due to alpha 1 or gamma globulins, alpha 2 macroglobulins, haptoglobins or ceruloplasmin. The serum fraction of M(r) 30,000-100,000 stimulated the binding of oxidized apoprotein but the albumin in this fraction inhibited lipid peroxidation and apoprotein oxidation. Serum ultrafiltrate (M(r) < 1000) inhibited lipid and protein oxidation, and binding; the inhibitory effect was abolished by deionization which removed histidine. The effects of lipoprotein-deficient serum and its fractions on cellular oxidation were similar but weaker than those on noncellular oxidation, HDL inhibited noncellular oxidation as well as binding of oxidized apoprotein. VLDL also inhibited oxidation; this could not be accounted for by its content of apo B. If present in vivo, these inhibitory effects would completely suppress both cellular and noncellular oxidation of LDL and its subsequent binding to collagen.

Autoradiographic analysis of the distribution of 125I-tyramine-cellobiose-LDL in atherosclerotic lesions of the WHHL rabbit

It is well established that plasma lipoproteins enter the artery wall and play a role in the atherogenic process. However, it is still unclear where within developing atherosclerotic lesions lipoproteins accumulate and which arterial cells participate in the metabolism of these lipoproteins. For this reason, light and electron microscopic autoradiograms were prepared from sections of lesioned aortas of Watanabe heritable hyperlipidemic (WHHL) rabbits 44 hours after injection of 125I-tyramine cellobiose-low density lipoprotein (TC-LDL). After uptake of 125I-TC-LDL and intracellular degradation of the LDL protein, the nondegradable TC ligand remains trapped and thus demarcates the cells participating in the degradation of LDL. Results of other studies indicate that 48 hours after injection into WHHL rabbits, about one half of the 125I label present in lesions represents accumulated degradation products while the remaining 125I label is present as intact 125I-TC-LDL. The distribution of autoradiographic silver grains was analyzed at low resolution in fatty streaks, transitional lesions, and advanced atheroma. In all cases, the majority of silver grains were associated with superficially located subendothelial macrophage-derived foam cells. In more advanced lesions, labeling was predominant in foam cells situated within the lateral margins of the lesions. Morphometric quantification of the distribution of silver grains in electron photomicrographs of fatty streaks from two young WHHL rabbits strongly supported the data obtained at the light microscopic level. In early fatty streaks from the aortic arch and the thoracic and abdominal aortas, subendothelial macrophage-derived foam cells contained a high proportion of the silver grains (40-60% of the total) and accounted for between 30% and 40% of the lesion volume. In contrast, smooth muscle cells in the lesions contained only 7-10% of the total silver grains and accounted for approximately 20% of the lesion volume. Endothelial cells contained the most silver grains on a per-unit-volume basis by occupying only 1-2% of the lesion volume. However, the endothelium contained less than 5% of the total grains in lesions. The remaining silver grains (25-45%) were associated with the extracellular matrix, which constituted between 40% and 50% of the lesion volume. These data indicate that in the WHHL rabbit, subendothelial macrophage-derived foam cells avidly accumulate and metabolize LDL despite having few functional LDL receptors.

Stimulation of human arterial smooth muscle cell chondroitin sulfate proteoglycan synthesis by transforming growth factor-beta

Human platelet-derived transforming growth factor-beta (TGF-beta) is a cell-type specific promotor of proteoglycan synthesis in human adult arterial cells. Cultured human adult arterial smooth muscle cells synthesized chondroitin sulfate, dermatan sulfate, and heparan sulfate proteoglycans, and the percent composition of these three proteoglycan subclasses varied to some extent from cell strain to cell strain. However, TGF-beta consistently stimulated the synthesis of chondroitin sulfate proteoglycan. Both chondroitin 4- and chondroitin 6-sulfate were stimulated by TGF-beta to the same extent. TGF-beta had no stimulatory effect on either class of [35S]sulfate-labeled proteoglycans which appeared in an approximately 1:1 and 2:1 ratio of heparan sulfate to dermatan sulfate of the medium and cell layers, respectively, of arterial endothelial cells. Human adult arterial endothelial cells synthesized little or no chondroitin sulfate proteoglycan. Pulsechase labeling revealed that the appearance of smooth muscle cell proteoglycans into the medium over a 36-h period equaled the disappearance of labeled proteoglycans from the cell layer, independent of TGF-beta. Inhibitors of RNA synthesis blocked TGF-beta-stimulated proteoglycan synthesis in the smooth muscle cells. The incorporation of [35S]methionine into chondroitin sulfate proteoglycan core proteins was stimulated by TGF-beta. Taken together, the results presented indicate that TGF-beta stimulates chondroitin sulfate proteoglycan synthesis in human adult arterial smooth muscle cells by promoting the core protein synthesis.

Visualization and quantification of transmural concentration profiles of macromolecules across the arterial wall

Transport parameters that describe a macromolecule entering the arterial wall from plasma can be obtained from concentration profiles of the labeled macromolecule entering the tissue. A new technique has been developed for measuring such concentration profiles, which offers spatial resolution superior to methods that measure profiles of radiolabeled macromolecules by serially sectioning tissue in planes parallel to the endothelium. In addition, this new method preserves cellular organization and tissue structure and permits measurement of concentration profiles underlying focal endothelial injuries or vascular lesions. The technique quantifies the concentration of a protein by measuring associated peroxidase activity. Although the present study was performed using horseradish peroxidase (HRP), the same principles can be applied to other macromolecules linked to HRP or microperoxidase. The colored reaction product of HRP was detected in transverse aortic sections using an image processing system. In the present study, profiles obtained by this new method were validated by comparison with HRP concentration profiles in rat aortas obtained by a serial slicing technique using radiolabeled HRP. We used the technique to measure high-resolution HRP concentration profiles in the intima and media of normal animals. These concentration profiles suggest that the internal elastic lamina acts as a major barrier to transport of macromolecules across the wall of the normal rat aorta. The new method should allow concentration profiles for macromolecules to be quantified in tissue surrounding vessels in the microcirculation, within the thickened intima of large vessels, and across coronary artery walls.

Visualization of apo B, fibrinogen/fibrin, and fibronectin in the intima of normal human aorta and large arteries and during atherosclerosis

Apolipoprotein B (apo B), fibrinogen/fibrin, blood platelets, factor VIII-related antigen of the blood coagulation system, and smooth muscle cells (SMC) were identified in the intima of normal and atherosclerotic human aorta and large arteries by the indirect immunofluorescence technique. Fibrinogen/fibrin was revealed by a monoclonal antibody (monAb) against the C-terminal region of human fibrinogen A alpha-chain. Fibronectin was visualized by monAb to the cellular form and against an epitope shared by different fibronectin subunit variants. In normal intima, fatty streaks, small amounts of fibrinogen/fibrin together with large amounts of apo B were observed. Fibronectin detected by two types of monAb was not found in extracellular matrix (ECM), whereas cellular fibronectin encircled SMC. According to the data obtained, fibrinogen/fibrin accumulates in plaques as a result of intramural thrombus incorporation, blood insudation, intramural haemorrhage, and in or around cells, apparently macrophages.

Distribution of oxidation specific lipid-protein adducts and apolipoprotein B in atherosclerotic lesions of varying severity from WHHL rabbits

Antisera and monoclonal antibodies generated against autologous malondialdehyde-conjugated low density lipoprotein (MDA-LDL), 4-hydroxynonenal conjugated LDL (4-HNE-LDL), and the protein fragments of apoprotein B resulting from the copper oxidation of LDL, as well as antibodies against apoprotein B, were used to immunostain atherosclerotic lesions of varying severity from Watanabe heritable hyperlipemic rabbits. In macrophage-rich fatty streaks and transitional lesions, all of the antibodies recognizing oxidation specific epitopes exhibited predominantly cell-associated staining in particulate and annular patterns. This is in contrast to the limited, extracellular, diffuse staining obtained with the antibodies recognizing apoprotein B. In more advanced lesions containing areas with reduced numbers of cells, there was increased extracellular, diffuse staining with the antibodies against oxidation specific epitopes and co-localization with apoprotein B. In addition, there were annular staining patterns associated with the necrotic core and increased staining of intimal and medial smooth muscle cells. We interpret these data as suggesting that in areas of lesions rich in macrophages, LDL is oxidized and taken up by the cells. In more advanced lesions that are relatively devoid of macrophages, both native and oxidized LDL, as well as oxidation products released from dead and decaying cells, are trapped in the matrix, out of reach of those cells capable of accumulating oxidized LDL.

A competitive assay of lipoprotein: proteoglycan interaction using a 96-well microtitration plate

A method for the microassay in vitro of lipoprotein: proteoglycan interactions is described. The wells of a plastic 96-well microtitration plate are coated with low density lipoprotein. A limiting quantity of biotin-conjugated proteoglycan is allowed to bind to each coated well, and the amount of the latter retained in wells is estimated spectrophotometrically through subsequent binding of alkaline phosphatase-conjugated avidin. Many of the incubation parameters (e.g., time, pH, salt concentration, divalent cations), which influence the extent of binding of biotin-conjugated proteoglycan, have been studied and optimized. The effect upon binding of introducing different levels of proteoglycans or lipoproteins at the interaction step can be measured readily. Thus, the orders of increasing relative binding affinities were found to be high density lipoprotein less than Lipoprotein (a) less than low density lipoprotein; rat chondrosarcoma proteoglycan less than bovine nasal cartilage proteoglycan less than human aorta proteoglycan; chondroitin 4-sulfate less than chondroitin 6-sulfate less than dermatan sulfate for lipoproteins, proteoglycans, and glycosaminoglycans, respectively.

Distribution of endogenous albumin across the rat aortic wall as revealed by quantitative immunocytochemistry

Endogenous albumin was revealed over thin sections of rat aortic wall, with high resolution and specificity, by applying the protein A-gold immunocytochemical technique. Gold particles, revealing albumin antigenic sites, were observed over plasmalemmal vesicles in endothelial cells and over the interstitial space throughout the thickness of the aortic wall. The distribution of the labeling in the interstitial space varied from region to region and was associated with the collagen fibers, following the orientation of the bundles. The morphometric evaluation of this labeling demonstrated a first peak in labeling intensity in the intima followed by a steep decrease with low levels in the media, and an increasing gradient towards the adventitia. In the subendothelium, a moderate labeling was observed at the base of the endothelial cells of both aortic and capillary endothelia, followed by a decreasing gradient. Ratios between the labeling density in the intima as well as in the adventitia and that in the capillary lumen (plasma albumin) revealed different concentrations of albumin in these compartments. Endogenous albumin, under steady-state conditions, is thus unevenly distributed over the interstitial spaces across the rat aortic wall, and appears associated along the collagen fibers.

LDL accumulation in the grossly normal human iliac bifurcation and common iliac arteries

We had previously used an electrophoretic transfer procedure to determine the topographic distribution of low density lipoprotein (LDL) accumulation in the aortic intima of normolipemic swine. In this present study we have employed a similar procedure to assess whether LDL-rich sites consistently demonstrate increased intimal thickening at the iliac bifurcation and common iliac arteries. The topographic distribution of LDL-rich sites was determined in the aortas of six subjects ranging in age from 16 to 36 years, by transferring LDL by electrophoresis from the tissue into an agarose gel containing anti-LDL, and then staining the immunofixed LDL in the gel for lipid. LDL-rich sites were found in all but two of these cases. On the basis of control studies establishing the level of nonspecific staining, we determined that the cutoff between LDL-rich and LDL-poor zones was 37 mg apoB protein/mm2 intimal surface area. Intimal thickening was found to be threefold greater in LDL-rich than in LDL-poor regions. These results confirm and extend earlier immunohistochemical studies suggesting a preferential accumulation of LDL at sites of intimal thickening in human arteries.

Apolipoprotein B in human aortic biopsies in relation to serum lipids and lipoproteins

A total of 46 patients, aged 39-71 years (mean 57.7), were studied. Forty-eight percent of the patients were hyperlipidemic and 63% had earlier suffered a myocardial infarction. Biopsies from aorta were obtained during coronary bypass surgery. Apo B was extracted from the intima by incubation of the tissue in buffer, followed by collagenase digestion. Intimal apo B was quantified in an immunoradiometric assay. There were significant correlations between total or collagenase-extractable apo B and serum cholesterol (rs = 0.39, P less than 0.01), serum triglycerides (rs = 0.33, P less than 0.05), LDL cholesterol (rs = 0.33, P less than 0.05) and serum apo B (rs = 0.37, P less than 0.05). The correlations were strongest for the collagenase-extractable apo B, while no correlations were observed for the buffer-extractable intimal apo B. No significant correlations were found between intimal apo B and serum HDL, apo A-I, smoking habits, history of hypertension or sustained myocardial infarction. Follow-up data were available for 42 of the patients, with a mean follow-up period of 35.1 months. The patients were classified according to symptoms of angina pectoris at the time of follow-up. There were significantly lower levels of serum apo A-I in the patients with poorer clinical prognosis. In a linear multiple stepwise regression analysis, apo A-I and serum LDL were significantly and independently related to clinical prognosis (R2 = 0.31).

Effects of cigarette smoke, carbon monoxide and nicotine on the uptake of fibrinogen by the canine arterial wall

An animal model has been used to investigate the effects of cigarette smoke, carbon monoxide and nicotine on the uptake of 125I-labelled fibrinogen by the arterial wall. The uptake of fibrinogen in the smoking group (5.5 +/- 2.8 counts.cm-2 x 10(-4)) and carbon monoxide group (6.1 +/- 2.7 counts.cm-2 x 10(-4)) was greater than the uptake in the control group (4.5 +/- 1.4 counts.cm-2 x 10(-4)) but this difference was not significant. In the nicotine group, there was a highly significant increase in wall uptake of 125I-fibrinogen (9.1 +/- 2.1 counts.cm-2 x 10(-4)) (P less than or equal to 0.001). These results suggest that nicotine, a major constituent of cigarette smoke, increases the retention of 125I-fibrinogen by the arterial wall and that this might be one mechanism by which cigarette smoking exerts its atherogenic effect.

Soluble complex formation between low-density lipoprotein and glycosaminoglycans. A 2H and 31P-NMR, and quasi-elastic light scattering study

Soluble complex formation between LDL and heparin (HEP) and chondroitin sulfate (CS) has been studied by 2H- and 31P-NMR and light scattering. The 2H-NMR linewidths of [2H]HEP and [2H]C4S increase substantially upon binding to LDL, with the [2H]HEP linewidths broader at low glycosaminoglycan (GAG)/low density lipoprotein (LDL) ratios. Preliminary analysis of the bound C2H3 group correlation times suggests that the observed linewidths are determined by the complex size, and that both [2H]GAGs have similar motions when bound to LDL. The 31P-NMR data demonstrate that large LDL-HEP complexes (diameter approx. 50 nm) are formed only over a narrow range of HEP concentrations, whereas the size of LDL-CS complexes increases continuously over the range of CS concentrations studied, reaching values of 32-35 nm for both C4S and C6S. At the lower protein concentrations studied by light scattering (less than or equal to 1 mg/ml), the same trends are observed, although the mean diameters are less than those estimated by 31P-NMR. Soluble complex formation was unaffected by the presence of 2 mM Ca2+. Dilution studies demonstrate that complex size varies with protein concentration. The binding of GAGs to LDL was also examined by HEP-CS competition studies. HEP has the higher affinity while no differences in binding could be detected between C4S and C6S.

Human aortic fibrolipid lesions. Immunochemical localization of apolipoprotein B and apolipoprotein A

Comparative localization of apolipoproteins (apo) B and A in small, raised fibrolipid lesions was performed to determine whether low density lipoproteins (LDL) and high density lipoproteins (HDL) may be involved in the formation of the earliest identifiable lipid-rich core regions found in atherosclerotic fibrous plaques. Apo B was observed associated with the superficial layer of foam cells in collagenous areas of the lesion cap and within the lipid-rich core region. The lipid-rich core region was well-circumscribed by an intense band of apo B staining along both luminal and medial aspects. Apo A staining was confined to the noncellular elements of the fibrolipid lesion. Collagenous areas were speckled with fine, punctate granules of immunoreactive apo A. A homogeneous, granular apo A staining pattern was characteristic throughout the core region. With these data from a relatively early lesion in the development of the atherosclerotic fibrous plaque, we conclude that: 1) both LDL and HDL can accumulate in the lipid-rich core region since no preferential staining for apo B over apo A was observed in this region, and 2) the paucity of cell-associated apo A staining in small fibrolipid lesions suggests that HDL interacts with the arterial wall differently than does HDL.

Interaction of native and cell-modified low density lipoprotein with collagen gel

We have examined the binding of native and cell-modified low density lipoprotein (LDL) to gels of Type I collagen. Diffusion of native 125I-LDL into the collagen gel was slow, reaching equilibrium after 24 to 48 hours, while L-3H-glucose, a low molecular weight marker, equilibrated in 6 hours. Binding of 125I-LDL was measured at 48 hours as the amount associated with the collagen after extensive washing. Binding was saturable with an increasing concentration of LDL. Prior incubation with cell-free culture medium resulted in modest, but progressive, increases in electrophoretic mobility and binding to collagen. Incubation with cells produced a marked increase in electrophoretic mobility and a 5- to 10-fold increase in collagen binding; the presence of butylated hydroxytoluene during incubation prevented both effects. These changes in LDL were induced by porcine aortic endothelial cells, smooth muscle cells, human skin fibroblasts, and a variety of cell lines, as well as by acetylation. There was a curvilinear relationship between the amount of LDL protein bound and the net negative charge of the LDL; increasing net charge was associated with progressively greater increases in binding. These results suggest a potential role for collagen in trapping lipid in the extracellular matrix of arterial intima by slowing the diffusion of and by binding LDL. The data also demonstrate that binding of LDL to collagen is enhanced by modifications that increase its net negative charge.

Influx and cellular degradation of low density lipoproteins in rabbit aorta determined in an in vitro perfusion system

The accumulation of 125I-low density lipoprotein (LDL) into normal and atherosclerotic arterial tissue and cellular uptake in arterial cells were studied in an in vitro perfusion system for rabbit aorta. The accumulation of 125I-LDL in normal tissue could be fitted to an inverse exponential function with an initial influx rate of 1.39 nl/mg wet weight/hour and an equilibration volume of about 2% of the tissue volume. The influx rate into atherosclerotic plaques was about 10 times faster and the equilibration volume, 50 times higher. In atherosclerotic tissue there was a steep concentration gradient between the plaque and the underlying media. The accumulation of 125I-LDL in the media under plaque and in normal tissue adjacent to plaques was similar to that seen in normal tissue. For studies of cellular uptake of LDL a trace label, 125I-tyramine-cellobiose (TC), was used. Normal or atherosclerotic rabbit aorta was perfused in vitro with medium containing 125I-TC-LDL. After perfusion the tissue was digested and the cells were isolated by density gradient centrifugation. Two main cell fractions with characteristics of smooth muscle cells and foam cells, respectively, were obtained. A 70-fold higher uptake was seen in the foam cells. In conclusion, these studies suggest a higher influx rate into atherosclerotic plaques, as well as a high LDL concentration in the plaque, compared with normal tissue or underlying media. We suggest that most of the cellular uptake of LDL in the arterial wall is caused by the foam cells.

Artery wall derived proteoglycan-plasma lipoprotein interaction: lipoprotein binding properties of extracted proteoglycans

Artery proteoglycan-lipoprotein binding characteristics were determined using intact, high molecular weight chondroitin sulfate proteoglycans (CS-PG) isolated from grossly appearing normal aortas of atherosclerosis susceptible WC-2 pigeons and plasma lipoproteins from normolipemic, randomly bred White Carneau pigeons. Optimum formation of particulate proteoglycan-lipoprotein complexes occurred in 5 mM Tris, 6 mM KCl, 4 mM CaCl2, 1 mM MgSO4, pH 7.2. The binding of CS-PG was specific for low density lipoprotein (LDL) and not high density lipoprotein (HDL). The relative importance of the intact monomeric structure of the PG was suggested in studies where glycosaminoglycan chains isolated from the PG monomer possessed less than 1% of the binding reactivity of the intact PG. The core protein prepared from the CS-PG monomer formed no measurable particulate complex.

Monoclonal antibodies recognizing lipid-laden cells and extracellular regions with lipid-deposits in atherosclerotic aorta

Monoclonal antibodies against lipid-laden cells and against extracellular regions of lipid-deposits in atherosclerotic aorta were prepared. Mice were immunized with a delipidated homogenate of atherosclerotic aorta of Watanabe-heritable hyperlipidemic rabbits. Hybridomas were obtained by fusion and cultured in hypoxanthine, aminopterin and thymidine selection medium. Specific antibodies were selected by indirect immunohistochemical staining of frozen sections of atherosclerotic aorta. Nine clones that produced antibodies that stained the atherosclerotic intima exclusively were selected and cloned by limiting dilution. Finally two clones (FCR1a/201F, FCR1b/904B) producing antibodies specific to lipid-laden cells and one clone (EMR1a/212D) producing an antibody specific to regions with lipid deposits in the extracellular matrix were established. These monoclonal antibodies may help in understanding how lipids accumulate in atherosclerosis.

Platelets and foam cells in the evolution of atherosclerosis. Histological and immunohistological studies of human lesions

Sixty-seven coronary, aortic and other lesions (39 subjects) ranging from minor to advanced disease were studied histologically and by immunohistology for antigens of platelets, beta-lipoprotein (LpB) and fibrinogen. Collections of lipid-filled foam cells were found in nearly every lesion mainly in the thickened intima, sometimes associated with macrophages. Foam cells were also present elsewhere including, in advanced lesions, necrotic zones filled with cholesterol needles. Then they too were necrotic. Platelets were observed in many lesions as small free clusters and/or phagocytosed in foam cells and sometimes also in macrophages. Cellular and extracellular reactions for platelet derivatives and for LpB were very frequent. Reactions for fibrinogen were only extracellular. Foam cell cytoplasm reacted for platelet derivatives and for LpB whilst macrophages, endothelium and certain other cells sometimes reacted for platelet derivatives but rarely for LpB. The studies indicate that: Foam cells originate from macrophages mainly in the intima and many migrate elsewhere. Their formation and their uptake of LpB requires cellular acquisition of a platelet factor or factors. Acquisition of platelet factor(s) is by phagocytosis of intact platelets and/or uptake of extracellular soluble derivatives from disrupted platelets. These processes begin early, are likely to be continuous and are significant for the evolution of atherosclerosis including the necrotic foci of advanced lesions.

Atherosclerosis in vascular grafts for peripheral vascular disease. Part 2. Synthetic arterial prostheses

Thirty-nine synthetic (32 Dacron and 7 Teflon) arterial prostheses (from 38 patients with peripheral arterial disease) removed after periods between 2 months and 18 years, were examined by histology and immuno-histology. The grafts were initially permeated by thrombus containing platelet antigens and this became organised and converted to granulation, and then to fibrous, tissue. The newly-formed tissue contained 'foreign-body' giant-cells in contact with the plastic prosthesis and showed evidence of permeation by plasma proteins. In grafts of over 2 years duration, this reactive tissue no longer contained platelet antigens but invariably revealed bound lipid, identifiable as apolipoprotein-B-containing lipoproteins (LpB), and fibrinogen-related antigens (FRA), in a distribution resembling that seen in atherosclerotic arteries. LpB and FRA were also seen in organised, or partially organised, mural thrombi in older grafts. The oldest grafts additionally showed stenosis, calcification or aneurysm formation. Lipid deposition increases with the age of grafts; is independent of the nature of the plastic fibre used or its mode of fabrication; and sometimes contributes to graft failure. Immuno-histology indicates that this is an insudative process indistinguishable from 'true' atherosclerosis which occurs in graft-linings of prostheses of long duration and in old mural thrombi in grafts and that the lipid in these lesions derives from plasma LpB rather than from platelets. This source for the lipid suggests that the insudative and thrombogenic theories of atherogenesis can be reconciled.

Interaction of low density lipoproteins with arterial proteoglycans. The role of charge and sialic acid content

The interaction of low density lipoproteins (LDL) with chondroitin sulfate-rich arterial proteoglycans appears to be initiated by coulombic interactions that lead to insoluble complexes. Once formed, large LDL aggregates are held together by non-polar associations. The irreversible formation of LDL proteoglycans aggregates was evaluated for different LDL preparations by definition of an avidity coefficient (Ar) using a Langmuir isotherm. LDL from different subjects, when tested against the same lipoprotein-complexing proteoglycan (LCP), gave Ar values ranging from 1-9 X 10(6) L/M. High avidity values were associated to lipoproteins with apparent isoelectric points above 6.5. These lipoproteins show low sialic acids content. The content of N-acetyl and N-acetyl,O-acetyl sialic acid was found inversely correlated with the avidity coefficient for the arterial LCP. Reductions of 42% on the LDL sialic acid content, by neuraminidase treatment, induced a 10-fold increment in their avidity for the lipoprotein complexing proteoglycan. The results indicate that at low ionic strength and physiological Ca2+-concentration and pH, the surface charge of LDL is an important modulator of the interaction with the arterial proteoglycan. Sialic acid, perhaps because of its exposure at the LDL surface, plays a determinant role in the in vitro association of LDL with the polyanionic proteoglycans. It is possible that in the intima-media the sialic residues of LDL and its balance of surface charges will control part of the interactions with the proteoglycans of the extracellular matrix.