Accumulation of native and methylated low density lipoproteins by healing rabbit arterial wall

FDG-PET can distinguish inflamed from non-inflamed plaque in an animal model of atherosclerosis

The presence of activated macrophages is an important predictor of atherosclerotic plaque rupture. In this study, our aim was to determine the accuracy of (18)F- fluorodeoxyglucose (FDG) microPET imaging for quantifying aortic wall macrophage content in a rabbit model of atherosclerosis. Rabbits were divided into a control group and two groups post aortic balloon injury: 6 months high-cholesterol diet (HC); and 3 months HC followed by 3 months low-cholesterol diet plus statin (LCS). In vivo and ex vivo microPET, ex vivo well counting and histological quantification of the atherosclerotic aortas were performed for all groups. Macrophage density was greater in the HC group than the LCS group (5.1 +/- 1.4% vs. 0.6 +/- 0.7%, P < 0.001) with a trend towards greater macrophage density in LCS compared to controls (P = 0.08). There was a strong correlation across all groups between macrophage density and standardized uptake value (SUV) derived from ex vivo microPET (r = 0.95, P < 0.001) and well counting (r = 0.96, P < 0.001). Ex vivo FDG SUV was significantly different between the three groups (P < 0.001). However, the correlation between in vivo microPET FDG SUV and macrophage density was insignificant (r = 0.16, P = 0.57) with no statistical differences in FDG SUV seen between the three groups. This study confirms that in an animal model of inflamed and non-inflamed atherosclerosis, significant differences in FDG SUV allow differentiation of highly inflamed atherosclerotic aortas from those stabilized by statin therapy and low cholesterol diet and controls.

Intravascular detection of inflamed atherosclerotic plaques using a fluorescent photosensitizer targeted to the scavenger receptor

Inflammation plays an important role in the pathophysiology of atherosclerotic disease. We have previously shown that the targeted photosensitizer chlorin (e(6)) conjugated with maleylated albumin (MA-ce6) is taken up by macrophages via the scavenger receptor with high selectivity. In a rabbit model of inflamed plaque in New Zealand white rabbits via balloon injury of the aorto-iliac arteries and high cholesterol diet we showed that the targeted conjugate showed specificity towards plaques compared to free ce6. We now show that an intravascular fiber-based spectrofluorimeter advanced along the -iliac vessel through blood detects 24-fold higher fluorescence in atherosclerotic vessels compared to control rabbits (p < 0.001 ANOVA). Within the same animals, signal derived from the injured iliac artery was 16-fold higher than the contralateral uninjured iliac (p < 0.001). Arteries were removed and selective accumulation of MA-ce6 in plaques was confirmed using: (1) surface spectrofluorimetry, (2) fluorescence extraction of ce6 from aortic segments, and (3) confocal microscopy. Immunohistochemical analysis of the specimens showed a significant correlation between MA-ce6 uptake and RAM-11 macrophage staining (R = 0.83, p < 0.001) and an inverse correlation between MA-ce6 uptake and smooth muscle cell staining (R = -0.74, p < 0.001). MA-ce6 may function as a molecular imaging agent to detect and/or photodynamically treat inflamed plaques.

Atherin: a newly identified, lesion-specific, LDL-binding protein in human atherosclerosis

Prolonged retention of LDL in focal, atherosclerosis-prone areas of arteries is a primary event in atherogenesis. To determine whether unrecognized LDL-binding proteins participate in this process, we generated a cDNA expression library from deendothelialized rabbit aorta, a model for early atherosclerosis that shows striking focal LDL retention in healing lesions. Library screening identified a previously unknown, highly conserved, 56kDa LDL-binding protein that we call atherin. Confocal microscopy of human arteries shows that atherin is present only in atherosclerotic lesions, not in normal intima. Within lesions, atherin is found both in the extracellular compartment and within foam cells. Essentially all extracellular atherin, as well as atherin within foam cells, co-localizes with LDL across the entire spectrum of human disease, from early lesions to advanced plaques. Our results suggest that focal arterial LDL accumulation may be initiated and maintained by binding between LDL and atherin, and that atherin may play a central role in atherogenesis by immobilizing LDL in the arterial wall.

Noninvasive in vivo measurement of vascular inflammation with F-18 fluorodeoxyglucose positron emission tomography

BACKGROUND

Fluorine 18 fluorodeoxyglucose (FDG) has been shown to accumulate in inflamed tissues. However, it is not known whether vascular inflammation can be measured noninvasively. The aim of this study was to test the hypothesis that vascular inflammation can be measured noninvasively by use of positron emission tomography (PET) with FDG.

METHODS AND RESULTS

Inflamed atherosclerotic lesions were induced in 9 male New Zealand white rabbits via balloon injury of the aortoiliac arterial segment and exposure to a high cholesterol diet. Ten rabbits fed standard chow served as controls. Three to six months after balloon injury, the rabbits were injected with FDG (1 mCi/kg), after which aortic uptake of FDG was assessed (3 hours after injection). Biodistribution of FDG activity within aortic segments was obtained by use of standard well gamma counting. FDG uptake was also determined noninvasively in a subset of 6 live atherosclerotic rabbits and 5 normal rabbits, via PET imaging and measurement of standardized uptake values over the abdominal aorta. Plaque macrophage density and smooth muscle cell density were determined by planimetric analysis of RAM-11 and smooth muscle actin staining, respectively. Biodistribution of FDG within nontarget organs was similar between atherosclerotic and control rabbits. However, well counter measurements of FDG uptake were significantly higher within atherosclerotic aortas compared with control aortas (P < .001). Within the upper abdominal aorta of the atherosclerotic group (area of greatest plaque formation), there was an approximately 19-fold increase in FDG uptake compared with controls (108.9 +/- 55.6 percent injected dose [%ID]/g x 10(3) vs 5.7 +/- 1.2 %ID/g x 10(3) [mean +/- SEM], P < .001). In parallel with these findings, FDG uptake, as determined by PET, was higher in atherosclerotic aortas (standardized uptake value for atherosclerotic aortas vs control aortas, 0.68 +/- 0.06 vs 0.13 +/- 0.01; P < .001). Moreover, macrophage density, assessed histologically, correlated with noninvasive (PET) measurements of FDG uptake (r = 0.93, P < .0001). In contrast to this finding, FDG uptake did not correlate with either aortic wall thickness or smooth muscle cell staining of the specimens.

CONCLUSION

These data show that FDG accumulates in macrophage-rich atherosclerotic plaques and demonstrate that vascular macrophage activity can be quantified noninvasively with FDG-PET. As such, measurement of vascular FDG uptake with PET holds promise for the noninvasive characterization of vascular inflammation.

Reversible and irreversible non-internalized LDL and methyl LDL accumulation by human fibroblasts

In previous in vivo animal studies, we showed that low density lipoprotein (LDL) accumulated irreversibly at the edges of healing arterial lesions rather than being internalized and degraded. To see if similar LDL accumulation occurs in vitro, fibroblasts from normal and homozygous familial hypercholesterolemic (FH) subjects were incubated at 37 degrees C with 125I-LDL and 125I-methyl LDL; the latter is not recognized by any known LDL receptor. Normal fibroblast accumulation of LDL and methyl LDL (5 microg/ml) plateaued within 1 h at 200 and 100 ng/mg, respectively. With FH cells, both LDL and methyl LDL accumulation plateaued at 100 ng/mg. Lipoprotein accumulation by both cell types rose steeply at concentrations up to 15-25 microg/ml, and less so at higher concentrations. Except for degradation of LDL by normal cells, degradation was minimal, which indicated that much of the lipoprotein accumulation was unaccompanied by internalization. The accumulation of both lipoproteins by both cell types was greater at 37 degrees C than at 4 degrees C, and was inhibited between 43 and 75% by homologous unlabeled lipoprotein. To see if any accumulation was irreversible, cells were incubated with radiolabeled lipoproteins for 3 h (pulse), then with homologous unlabeled lipoproteins for up to 20 h (chase). About 50% of intact radiolabeled lipoprotein rapidly dissociated from cells into the medium in the first 4 h of the chase period. In contrast, between 4 and 20 h, most of the remaining intact LDL and methyl LDL appeared to be irreversibly bound, because it was released at a rate of only 0-1%/h. Thus, we conclude that, under the conditions studied, both reversible and irreversible non-internalized LDL binding play a major role in LDL accumulation by cultured cells.

Preparation and biodistribution of 99m technetium labelled oxidized LDL in man

Radiolabelled autologous low density lipoprotein (LDL) has previously been used to study in vivo distribution and metabolism of native-LDL. Non-invasive imaging of atherosclerotic lesions using 99mTc-LDL was shown to be feasible in animal models and patients but the clinical utility remains to be assessed. Since recent reports suggest that oxidized LDL may play a major role in the pathogenesis of atherosclerosis, we developed a technique to oxidize autologous LDL and compared the biodistribution of oxidized-LDL with that of native-LDL in man. In addition, we evaluated the uptake in vivo of oxidized- and native-LDL by atherosclerotic plaques. LDL, obtained from human plasma was treated with various combinations of copper ions and H2O2 to induce oxidative modification by increasing the content of lipid peroxidation products and electrophoretic mobility. When LDL (0.3 mg/ml) was incubated with 100 microM Cu2+ and 500 microM H2O2 oxidation occurred rapidly within 1 h, and was labelled with 99mTc efficiently as native LDL. In vivo distribution studies revealed a faster plasma clearance of oxidized-LDL compared to native-LDL, and a higher uptake by the reticuloendothelial system. Tomographic scintigraphy of the neck in patients suffering from transient ischemic attacks, revealed accumulation of radiolabelled LDL preparations in the carotid artery affected by atherosclerotic lesions. We developed a technique to rapidly oxidize LDL using copper and H2O2. Biodistribution data demonstrate that oxidized-LDL is rapidly cleared from circulation, is taken up mostly by organs rich in macrophages, and can be detected at the level of carotid plaques.

External imaging of atherosclerosis in rabbits using an 123I-labeled synthetic peptide fragment

The oligopeptide fragment of apolipoprotein B, SP-4, has demonstrated pronounced uptake in the healing edges of balloon-injured rabbit aortic endothelium. To assess 123I-labeled SP-4 for identification of atherosclerotic plaques by gamma camera imaging, 14 Watanabe heritable hyperlipidemic (WHHL) and 5 normal rabbits were imaged 5 minutes and 12 and 24 hours after intravenous injection of 123I-SP-4. In addition, two WHHL and two normal rabbits were injected with 125I-SP-4 for autoradiography. Twelve of the 14 WHHL, but none of the normal, rabbits had visually apparent focal radioiodine accumulation in the region of the aorta. Focus-to-lung and focus-to-heart count ratios were 2.4 +/- 1.3 and 1.0 +/- 0.4, respectively. Five of the visually positive WHHL rabbits were reimaged 4 and 8 weeks later with 123I-NaI and 123I-SP-2 (an apo E peptide), respectively, as negative controls. Perceptible, but faint, aortic localization of 123I-NaI and of 123I-SP-2 was seen in only one animal each. The distributions of atherosclerotic lesions on photographs of the opened WHHL aortas and of film blackening on 125I-SP-4 autoradiograms were identical. In contrast, the two normal rabbit aortas did not exhibit plaques on photographs or film blackening on autoradiograms. Thus, in an animal model closely simulating human atherosclerotic disease, SP-4 localizes specifically in aortic atherosclerotic lesions.

Surface-induced conformational switching in amphiphilic peptide segments of apolipoproteins B and E and model peptides

The conformational and surface-binding properties of a synthetic peptide corresponding to Tyr-apolipoprotein B-100(1000-1016) amide, SP-4, which was previously shown to mimic the focal accumulation pattern of LDL on the healing de-endothelialized rabbit aorta [Shih et al. (1990) Proc. Natl. Acad. Sci. USA 87, 1436-1440], have been investigated. SP-4 behaves as an amphiphilic alpha-helical peptide at the air-water interface and bound to siliconized quartz slides. However, its N alpha-acetylated analogue formed beta-sheet structures at the air-water interface. Nonhomologous peptide models of SP-4 also exhibited mixed alpha-helical and beta-sheet surface-binding behavior. Peptides corresponding to the cationic apolipoprotein (apo) B/E receptor binding regions of apoE (SP-2) and apoB (SP-11) were also studied. SP-2 behaved as an amphiphilic alpha helix, but, surprisingly, SP-11 formed surface-induced beta-sheets. These results demonstrate that all of the peptides studied have surface-binding properties, and suggest further that either alpha-helical or beta-sheet peptide structures may determine the binding of LDL to the arterial wall or the apoB/E receptor.

Long-term nicotine exposure increases aortic endothelial cell death and enhances transendothelial macromolecular transport in rats

Repeated endothelial cell injury has been suggested as an initiating factor in atherogenesis. Dying or dead endothelial cells have been shown to make significant contributions to the local enhancement of transendothelial macromolecular transport. Since cigarette smoking is one of the major risk factors for atherosclerosis, we examined the hypothesis that smoking accelerates atherogenesis by increasing the frequency of endothelial cell death and hence transendothelial macromolecular transport. Sixteen male Sprague-Dawley rats were given nicotine at a weight-adjusted dose of 5 mg/kg body wt per day in their drinking water over a period of 6 weeks. A group of 16 age-matched male Sprague-Dawley rats not exposed to nicotine and maintained over the same time period served as the control group. In en face preparations of thoracic aorta, immunoglobulin G-containing dying or dead endothelial cells were identified by the indirect immunoperoxidase method, and endothelial leakage to Evans blue-albumin (EBA) complexes (5 minutes after intravenous injection) was visualized by fluorescence microscopy. The results showed that in nicotine-treated rats, 51% of dead endothelial cells were associated with EBA leakage, which was responsible for 57% of total EBA leaky foci. Both the frequency of endothelial cell death (0.94 +/- 0.11% versus 0.40 +/- 0.04%, p < 0.0001 by two-tailed, unpaired Student's t test) and the number density of EBA leaky foci (6.45 +/- 1.23/mm2 versus 3.30 +/- 0.49/mm2, p < 0.05 by two-tailed, unpaired t test) were significantly greater in nicotine-treated rats than in control rats.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of methods for incorporating a radioiodinated residualizing cholesteryl ester analog into low density lipoprotein

Two different methods were evaluated for incorporating [125I]cholesteryl iopanoate ([125I]CI), a non-hydrolyzable cholesteryl ester analog, into LDL. The first procedure was an organic solvent delipidation-reconstitution procedure (R[125I-CI]LDL) while the second involved incubation of detergent (Tween-20)-solubilized [125I]CI with whole plasma (D[125I-CI]LDL). R[125I-CI]LDL behaved similar to native LDL in vitro, but was markedly different in vivo, apparently due to a heterogeneity in particle size. D[125I-CI]LDL, however, was metabolized normally both in vitro and in vivo. These results, combined with the residualizing nature of [125I]CI, demonstrate that D[125I-CI]LDL is appropriate for tracing LDL uptake in vivo.

Time course of 125I-labeled LDL accumulation in the healing, balloon-deendothelialized rabbit aorta

We previously showed by qualitative en face autoradiography that after 24 hours of circulation, 125I-labeled low density lipoprotein (LDL) injected in tracer amounts accumulated focally at the edges of regenerating endothelial islands in the balloon catheter-deendothelialized aorta of the normocholesterolemic rabbit. In the present study with the same animal model, we have used quantitative autoradiography to examine 125I-LDL accumulation in the healing aorta as a function of LDL circulation time from 2.5 to 40 hours. The results demonstrated that 125I-LDL accumulation in the healing aorta occurred in two kinetically and biochemically distinct compartments, one of which was in equilibrium with plasma and one of which sequestered LDL. LDL accumulation in the still-deendothelialized aorta (DEA) was diffuse and only moderately intense on autoradiography. It peaked 4 hours after injection; over the following 36 hours the disappearance of 125I-LDL from DEA paralleled the disappearance of 125I-LDL from plasma. In contrast, accumulation of 125I-LDL at the edges of regenerating endothelial islands was focal and intense. LDL accumulation in this compartment also peaked 4 hours after injection but remained elevated even at 40 hours, despite falling plasma levels of LDL. At 24 hours, edge LDL accumulation per unit area was more than five times greater than DEA accumulation. The data indicate that LDL accumulation in specific compartments of the functionally modified arterial wall occurs independently of either acute or chronic hypercholesterolemia. The contrast between labile LDL accumulation in DEA and persistent accumulation at the edges of healing aortic islands indicates that LDL accumulation in the two areas must involve different processes within the arterial wall itself.

Low density lipoprotein uptake by an endothelial-smooth muscle cell bilayer

To study the interaction of endothelial and smooth muscle cells, and the means by which such interaction may affect lipid permeability of the arterial wall, cell bilayers were established by use of a transwell culture system. After confluent growth of both cell types had been achieved, iodine 125 bound to low-density lipoprotein (10 ng protein/ml) was added to the media of the upper well. After a 3-hour incubation period, the iodine 125-bound low-density lipoprotein content of the upper and lower media demonstrated an impedance to lipoprotein movement across the endothelial cell monolayer as compared to the bare porous polycarbonate filter of the transwell (p less than 10(-6)). The presence of smooth muscle cells in the bottom well significantly enhanced the permeability of the endothelial cell layer (p less than 10(-60). This effect remained unchanged over a 9-day time course. Membrane binding and cellular uptake of low-density lipoprotein by endothelial cells was not altered by smooth muscle cells, indicating that this change in permeability could not be easily attributed to changes in receptor-mediated transport or transcytosis. Membrane binding (p less than 0.02) and cellular uptake (p less than 10(-6)) of low-density lipoprotein by smooth muscle cells in the bilayer, when adjusted for counts available in the smooth muscle cell media, were both reduced in the early incubation period as compared to isolated smooth muscle cells. The disproportionate reduction in uptake as compared to binding would suggest that this was not entirely a receptor-dependent process.(ABSTRACT TRUNCATED AT 250 WORDS)

Focal accumulation of an apolipoprotein B-based synthetic oligopeptide in the healing rabbit arterial wall

The functions of surface-accessible domains of apolipoprotein (apo) B, the protein moiety of low density lipoprotein (LDL), are unknown, aside from the LDL receptor-binding domain, which lies toward the carboxyl-terminal end of apoB. Since LDL accumulation in arterial lesions does not depend on recognition of LDLs by a cell-surface receptor, we synthesized an oligopeptide with the sequence of the trypsin-accessible domain of apoB that lies closest to the amino-terminal end of the protein and compared its biological activity to that of another synthetic oligopeptide with the sequence of the heparin- and apoB/apoE receptor-binding domains of apoE. (Tyrosine was added at the amino-terminal end of each peptide to facilitate radiolabeling.) The 18-amino acid apoB-based peptide included residues 1000-1016 of apoB, for which no function has been previously described. In radioautographs, the 125I-labeled peptide accumulated focally at the healing edges of regenerating endothelial islands in the balloon-catheter deendothelialized rabbit aorta. In contrast, the 21-residue apoE-based peptide, which included residues 129-148 of apoE, accumulated diffusely and uniformly throughout the deendothelialized areas of the aorta. The data show that focal binding of the apoB-based peptide can delineate arterial lesions and suggest that this arterial wall-binding domain of apoB mediates accumulation of LDLs in arterial lesions.

Biodistribution of 131I-radiolabelled plasma low density lipoprotein in hyperlipidaemic vervet monkeys

Low density lipoprotein (LDL) labelled with 131I has been administered to 6 Vervets 2 of which were high responders to an atherogenic Western diet in terms of plasma cholesterol, 2 were low responders and 2 were fed a high carbohydrate control diet. The ratio of hepatic to cardiac activity was recorded for up to 10 days after administration of the labelled LDL. Liver activity had a longer biological half life in the high-responders and this can be interpreted in terms of a variation of hepatic metabolism of LDL, with direct relevance to the human situation.

The arterial barrier to lipoprotein influx in the hypercholesterolemic rabbit. 2. Long-term studies in deendothelialized and reendothelialized aortas

These studies consider whether a mild aortic injury that does not increase cholesteryl ester influx during the first few days promotes atheromatosis in the hypercholesterolemic rabbit. The cholesteryl ester influx in uninjured, deendothelialized, and reendothelialized aorta was also measured in order to account for the different cholesteryl ester contents in these areas. By 32-33 days after localized aortic injury which was made after 5-7 days of cholesterol feeding, uninjured (control) areas of the thoracic aortas had accumulated 48 micrograms/cm2 each of esterified and nonesterified cholesterol due to continued cholesterol feeding. However, the previously injured deendothelialized and reendothelialized areas of the aortas had accumulated 6 and 10 times as much esterified cholesterol, and 2 and 5 times as much nonesterified cholesterol, respectively, as the adjacent uninjured areas. Esterified cholesterol influx was low during the second day after injury but increased with time so that 30-31 days later the esterified cholesterol influx in deendothelialized and reendothelialized aorta was respectively 44 and 7 times as great as the 0.1 microgram/h/cm2 in the adjacent uninjured aorta. However, expressed per mg aortic cholesteryl ester, cholesteryl ester influx in reendothelialized aorta was no greater than in noninjured aorta. These studies suggest that an injury that does not initially increase cholesteryl ester influx can promote atheromatosis and that the higher rate of cholesteryl ester influx that develops with time in reendothelialized areas is closely related to its increased cholesteryl ester content.

The arterial barrier to lipoprotein influx in the hypercholesterolemic rabbit. 1. Studies during the first two days after mild aortic injury

These studies examine the hypothesis that removal of aortic endothelium eliminates a barrier to lipoprotein cholesterol influx. The aortas of rabbits fed a cholesterol-rich diet from 7 to 16 days before balloon injury were studied 1 or 2 days after deendothelialization of specific areas of the aorta. By this design the aortic sterol content was near normal on the day of injury, but areas of injured and noninjured aorta were exposed to identical levels of elevated plasma cholesterol. Measuring the arterial [3H]- and [14C] cholesterol fractions accumulated during 2 different intervals after dosage in the same animals permitted calculation of total influx and fractional loss of aortic cholesterol. During the first 2 days after deendothelialization, total (unidirectional) cholesteryl ester influx in deendothelialized aorta was similar to that in adjacent uninjured aorta, but total influx of nonesterified cholesterol was increased. The calculated increase in influx of nonesterified cholesterol was not a result of an increase in hydrolysis of entered cholesteryl ester but probably represents increased exchange of labeled cholesterol between artery and plasma. These results suggest that subendothelial layers of the aorta of short-term cholesterol-fed rabbits function as significant barriers to lipoprotein influx and that processes other than increased permeation by lipoproteins initiate injury-induced lesions.

Actin stress fiber content of regenerated endothelial cells correlates with intramural retention of intermediate plus low density lipoproteins in rat aorta after balloon injury

The rat aortic model of endothelial injury (balloon catheter induced) has been used to establish whether changes in protein intramural penetration in specific areas of the injured aorta were accompanied by phenotypic modifications of the regenerated endothelial cells covering these particular regions. Iodinated lipoproteins (IDL/LDL fraction) and albumin were used as tracers to localize protein permeability and retention in the aorta. Lipoproteins, but not albumin, were retained in the thickened areas covered with regenerated endothelium (i.e., 60 days after balloon induced injury). Neither lipoproteins nor albumin were retained in the other aortic areas studied, including the intimal thickening of de-endothelialized areas (15 days after injury). The relative volume of cytoplasmic stress fibers was significantly increased in regenerated endothelium covering thickened areas as compared with the other regions of the injured or normal aorta. The accumulation of lipids usually observed in atherosclerotic lesions, compatible with the trapping of lipoproteins by the matrix component of the intimal thickening, may be related to modulated features of endothelial cells regenerated over thickened areas of the aorta.

Imaging human atherosclerosis with 99mTc-labeled low density lipoproteins

The feasibility of localizing human atherosclerotic plaques by gamma scintillation camera external imaging with technetium-99m-labeled low density lipoproteins (99mTc-LDL) was tested in 17 patients who had atherosclerosis. Imaging demonstrated focal accumulation of radiolabel consistent with 99mTc-LDL sequestration by plaques in the carotid, iliac, or femoral vessels of four patients 8 to 21 hours after intravenous injection of the radiopharmaceutical. Focal accumulation of 99mTc-LDL also appeared in the location of coronary lesions in four patients, but this accumulation could not be distinguished with certainty from residual blood pool radioactivity. When carotid endarterectomy specimens from six patients who received 99mTc-LDL 1 day before endarterectomy were examined, the specimens had focal accumulations of radiolabel, with two to four times greater radioactivity in some regions of each specimen than in others; this occurred whether or not the lesions were detected on the gamma camera images. Lesion composition may have determined whether accumulation was quantitatively sufficient to produce an external image. Histologically, the imaged carotid specimen had abundant foam cells and macrophages and poorly organized intramural blood consistent with a plaque hemorrhage; in contrast, nonimaged endarterectomy specimens were mature, fibrocalcific plaques. We conclude that: 1) 99mTc-LDL did accumulate in human atherosclerotic plaques; 2) in some patients, the accumulation of 99mTc-LDL was sufficient for detection by gamma camera imaging; 3) the amount of LDL that accumulated appeared to depend on lesion composition; and 4) the design of new radiopharmaceuticals with reduced residual blood pool activity relative to plaque accumulation should lead to improved external imaging of atherosclerosis.