Mechanisms

DONALD

E3. ZILVERSMIT,

of Cholesterol

PhD’

Ithaca, New York

From the Division of Nutritional Sciences and Section of Biochemistry, Molecular and Cell Biology, Division of Biological Sciences, Cornell University, Ithaca, N. Y. That part of the work reported here which was done in our laboratory was supported in part by funds provided by Public Health Research Grant HL 10933 from the National Heart and Lung Institute, U. S. Public Health Service and in part by funds provided through the State University of New York. Career Investigator of the American Heart Association. Address for reprints: Donald 6. Zilversmit. PhD. Division of Nutritional Sciences, 202 Savage Hall, Ithaca. N. Y. 14853.

Accumulation

in the Arterial Wall

The rate of cholesterol accumulation is a function of three separate processes: the transfer of lipid or lipoprotein from blood plasma to the artery, the binding and sequestering of lipids in the arterial wall and the solubiltzation and removal of lipid from the artery. These processes have been studied with lipids or lipoproteins labeled with radioisotopes by autoradiographic and quantitative chemical procedures. More recently immunochemical procedures have been applied to this problem. Studies have been performed with intact animals, isolated organs and cell cultures. In addition, homogenates have been used successfully to study intraarterial transformations of lipids, (for example, cholesterol esterification). Although epidemiologic and clinical studies, as well as animal experiments, have provided evidence that the concentration of plasma low or very low density lipoproteins parallels the rate of atherogenesis, the nature of the causal chain linking plasma lipoproteins to atherosclerosis Is as yet unclear. A possible link between plasma lipoproteins, arterial liproprotein lipase and atherogenesis has been postulated.

The purpose of my contribution to this Symposium is not to review the many aspects of cholesterol transport and metabolism and their relation to atherogenesis; several monographs and reviews are available on those topics.l-lo Instead, I shall limit myself to those aspects of cholesterol and lipid transport that appear to have the most direct bearing on atherogenesis. Even in this attempt I shall not try to be exhaustive in the review of literature but select for presentation those papers that have most closely paralleled or complemented the research in my laboratory. The accumulation of cholesterol and its esters in the arteries of human subjects and experimental animals may be regarded as the consequence of three separate processes: those facilitating or impeding the entry of cholesterol into the arterial wall; those required to bind, precipitate or segregate the arterial cholesterol into an insoluble form; and those mediating the release of cholesterol from the wall either as a balancing element in the process of atherogenesis or as the main factor responsible for regression of cholesterol-containing lesions. Most of the work available to date is concerned with the entry of cholesterol and thus of necessity most of this paper will deal with that subject. In addition, the release of cholesterol from arteries and from isolated cells, and the role of cholesteryl ester and low density and very low density lipoproteins in atherogenesis will be discussed.

l

Uptake of Labeled Cholesterol Several review articlesil-l4 have commented on studies performed with carbon-14 or tritium labeled cholesterol in animals and man. Be-

April 1975

The American Journal of CARDIOLOGY

Volume 35

559

CHOLESTEROL

ACCUMULATION

IN ARTERIAL

WALL-ZILVERSMI

fore enlarging on these comments and including data from more recent studies, I will discuss some difficulties in the execution and interpretation of such experiments. Pitfalls in execution periments: The labeled

and interpretation

rived from plasma cholesterol or was in rapid equilibrium with plasma cholesterollz~ze On the other hand, in the normal aorta, local synthesis of cholesterol may well take place. Dayton,21 for example, observed that about one third of the cholesterol in the abdominal aorta of normal chickens is derived from synthesis in situ. Field et a1.22 concluded from their studies in atherosclerotic patients that a major portion of arterial cholesterol is derived from synthesis by the artery. In similar studies Chobanian and Hollander2s and Gould et a1.24 concluded that arterial cholesterol was subject to turnover. Both of these conclusions are subject to the previously mentioned reservations about the interpretation of tracer experiments.

of ex-

cholesterol may be administered as an oral or an intravenous dose. The latter can be prepared by solubilizing the labeled cholesterol in plasma or in saline solution by the use of a nonionic detergent, l5 but the possible effect of the detergent on lipid transport and metabolism has to be considered. If cholesterol in an alcoholic solution without surfactant is injected directly, or after mixing with a salt solution, the labeled cholesterol is present in particulate form which is swept from the bloodstream by Kupffer cells in the liver and by macrophages in other tissues.16 In each instance the investigator needs to ask himself whether or not the labeled material is taken up by the aortic wall in a physiologic manner. Phagocytic uptake of particulate cholesterol by aortic endothelium, for example, would give misleading results. In our own experiments we have preferred to administer the cholesterol label as part of the food supply by dissolving the labeled cholesterol in a small amount of oil that is then thoroughly mixed with a portion of the diet. A second point of consideration is the manner of calculating the results of the uptake study. The labeled cholesterol found in the arterial wall at a given time may or may not represent the total amount of label taken up because any losses of label from the wall would influence the total label found in the wall. Consequently one should design the study so that loss of label is minimal or so that a correction for the loss can be made. A correction would not be difficult if the arterial cholesterol pool were homogeneously labeled, but this is usually not the case.r7m1s Thus, loss of label is particularly large in a long-term experiment after a single intravenous dose of label, when the specific activity of plasma cholesterol is continually decreasing while the specific activity of cholesterol in the artery is increasing. Later, the specific activity of the superficial cholesterol layers in the artery may well exceed that of plasma cholesterol, so that a net loss of label from the wall occurs. Under these conditions the calculated rate of uptake of cholesterol by the artery may be a gross underestimate of the actual uptake. In our experiments we have attempted to minimize this error in two ways: (1) by feeding the labeled cholesterol with the diet in divided doses, thus generating a rising plasma cholesterol specific activity curve, and (2) by terminating the experiment no later than 48 hours after the beginning of dosage. The rate of plasma cholesterol uptake by arterial wall is then calculated from the arterial radioactive cholesterol divided bv the area under the plasma cholesterol specific activity curve.l’)

In studies on experimental animals it is easier to overcome some of these objections, and the results of earlier studies have been critically examined by Zilversmit12 and Dayton and Hashimoto.14 In our studies17 as well as those by Dayton and Hashimoto,25 Jensen,26 and Day et a1.,27 uptake of labeled cholesterol by the arterial wall in vivo appeared to be roughly similar to that observed in vitro. Inactivation of enzymes by metabolic inhibitors, or by boiling strips of aorta prior to incubation with labeled cholesterol preparations, appeared to have little effect on cholesterol uptake. 17s25On the other hand, Jensen28 concluded that uptake of cholesterol by rabbit aorta was related to glycolytic activity of that organ. Hemodynamic tion and uptake:

April 1975

The American

Journal of CARDIOLOGY

Volume

and cholesterol

deposi-

In addition to metabolic activity, the effect of blood pressure on cholesterol deposition in the arterial wall has been subject to both speculation and experimental investigation. According to the filtration theory of atherogenesis,2g one would expect to find a large effect of hemodynamic factors on the uptake of labeled cholesterol by the arterial wall. Campbell et al.“O found little or no effect of hypertension on the uptake of labeled plasma cholesterol by atheromatous arteries of intact cholesterol-fed rabbits, but some increase in cholesterol concentration in the atheromatous arteries of the hypertensive animals was seen. Caro and Nerems1,a2 studied a variety of hemodynamic factors in perfused carotid arteries of the dog but found that the rate-controlling processes governing the uptake of plasma cholesterol by the arterial wall were related primarily to transport within the wall rather than to diffusion of lipid or lipoprotein through a boundary layer between the wall and the bulk of blood plasma. Other evidence that factors within the arterial wall determine the local uptake of plasma cholesterol has been provided by Somer and Schwartz,ig who found that areas of pig aorta that showed greater permeability to Evans blue also showed increased uptake of labeled plasma cholesterol. It is possible, of course, that some hemodynamic factor was originally at the root of the increased focal permeabilities but evidence supporting a direct role of the integrity of the arterial wall in cholesterol uptake has sprung from many other studies. Among the recent studies in this area the investigations of Bondjers and

Source of arterial wall cholesterol: From our studies12,20 on the source of cholesterol in the arterial wall of the cholesterol-fed rabbit, it was evident that most of the arterial cholesterol was either initially de-

560

factors

35

CHOLESTEROL ACCUMULATION

Bj6rkerud33-35 clearly show that in otherwise normal rabbit arteries, cholesterol uptake in areas of endothelial injury is 10 to 15 times greater than in areas in which the endothelium is intact. Free cholesterol in arteries and plasma equilibrated very rapidly in the injured areas compared with areas that had not been injured or in which repair had taken place. These findings demonstrate that intact endothelium provides a barrier against the free exchange of cholesterol between plasma and the arterial wa11.s6 Similar conclusions have been reached for the permeability of the arterial wall toward plasma proteins.s7 Release

of Cholesterol from Arteries Isolated Cells

and

Pitfalls in interpretation of experiments: If there are difficulties in the study of cholesterol uptake by the arterial wall, there are even greater difficulties in the interpretation of experiments designed to show, by means of labeling experiments, that net efflux of cholesterol is possible under appropriate conditions. Several types of difficulties exist: (1) Arteries used for in vivo or in vitro experiments are seldom uniformly labeled. Loss of labeled cholesterol from a surface layer with a much greater cholesterol specific activity than the rest of the lesion cannot easily be translated into an efflux rate. (2) Even if the artery were uniformly labeled by prolonged feeding of the label before the efflux experiment is begun, the depletion of label from the arterial surface during an in vitro release experiment would not allow the simple calculation of release rates, which assumes a homogeneously labeled pool. (3) In vitro efflux experiments have been performed in which loss of labeled cholesterol from both the intimal and adventitial surface has been measured although the loss from the adventitial tissue may or may not be relevant to regression of atheromas. (4) Loss of labeled cholesterol from the arterial wall in vivo, as described by Lofland and Clarkson in the pigeon, is complicated by the presence of labeled cholesterol in plasma during the regression period. In these latter experiments, for example, labeled cholesterol was fed to Carneau pigeons for 30 days, after which time the label was removed from the diet, but nonlabeled cholesterol remained as part of the diet. The efflux of label was calculated from the half-time of disappearance of radioactive cholesterol from normal areas, fatty streaks and plaques. From the half-times and amounts of cholesterol present in each area, an efflux of arterial cholesterol was calculated, but this calculation is valid only if no influx of label occurs while the halflife is measured. In other words, if the arterial cholesterol pool is labeled with a pulse label the half-life of the die-away curve gives a measure of efflux; but if the plasma continues to contribute label, as was the case in the pigeon experiments, the experimentally determined half-life is much longer than the true half-life and the efflux of arterial cholesterol would be underestimated.

IN ARTERIAL WALL-ZILVERSMIT

Results of in vitro experiments on cholesterol release: Notwithstanding the limitations of in vitro experiments for the measurement of cholesterol release, they seem to offer greater possibilities for control of essential variables than those performed in intact animals or in man. Dayton and Hashimoto,14 Newman and Zilversmit17 and Day et a1.27 have shown that both free and esterifed cholesterol can be released from arterial strips and that metabolic activity appears to play little or no role in this release. Bell et a1.3g demonstrated that pigeon aortas perfused with serum release labeled cholesterol into the medium, and that atheromatous aortas release about twice as much cholesterol as do normal arteries. Rutenberg and Soloff40 have measured the release of cholesterol from human arteries into normal and heat-inactivated serum. They speculate that 1ecithin:cholesterol acyltransferase may be involved in the transfer of free cholesterol from artery to serum and they have presented preliminary observations41 showing that patients with myocardial infarcts have reduced activity of this enzyme. Cholesterol uptake and release by isolated cells: Finally, isolated cell systems have been used to study the release of cholesterol into the medium. Rothblat and Kritchevskti2 have worked with tissue culture cells and have also measured the effect of added fatty acids, phospholipids and delipidized serum proteins on the uptake of cholesterol by cells in culture.43 Bailey44 has measured both uptake and release of cholesterol by cell cultures and has demonstrated that of various types of delipidized serum tested in the incubation media, rabbit serum was the least effective in promoting release of labeled cholesterol from cells. Heating serum sufficiently to inactivate 1ecithin:cholesterol acyltransferase did not alter the ability of serum to increase cholesterol release. Nilsson and Zilversmit45 used rat liver and spleen macrophages to study the release of cholesterol from cells that had ingested preparations of colloidal cholesterol. Cells incubated in buffers released little or no labeled cholesterol. In the presence of whole serum or serum lipoproteins, label was released into the medium; the release was not affected by cycloheximide, sodium fluoride, potassium cyanide or by linoleic acid or lysolecithin in physiologic concentrations. Incubation of the cholesterol-loaded cells with serum in which the free cholesterol had been depleted by previous exposure to 1ecithin:cholesterol acyltransferase activity did not increase the release of labeled cholesterol above the level obtained with the same amount of fresh serum. It would appear from these experiments and-from those of Werb and Cohn46 on mouse peritoneal macrophages that free cholesterol equilibrates readily between cell membranes and serum lipoprotein. In the state of equilibrium the free cholesterol moves to the same extent from cells to lipoproteins as in the reverse direction. In a steady state the movement of cholesterol, even though it proceeds in both directions, may predominate from cell to lipoprotein or in the reverse direction depending on the concentration April 1975

The American Journal of CARDIOLOGY

Volume 35

561

CHOLESTEROL

ACCUMULATION

IN ARTERIAL WALL-ZILVERSMIT

gradient of cholesterol and of cholesterol-binding sites existing at a given moment. Thus, uptake of free cholesterol by cells may occur when serum lipoproteins are relatively overloaded with unesterified cholesterol or other lipids that compete for the same binding sites. Release of free cholesterol probably occurs under the reverse condition. It is therefore important to have a better understanding of lipid-lipid and lipid-protein interactions and the effect of these interactions on the potential for cholesterol flux into and out of cells. The previous considerations have dealt mostly with the movement of unesterified cholesterol. The situation for esterified cholesterol may be quite different. The cholesteryl esters are much less polar and are thought to be present in the hydrophobic interiors of lipoproteins and segregated from the cell membrane into oil droplets of microscopic size. I shall therefore take a separate look at cholesteryl ester localization and formation in the arterial wall. Cholesteryl

Esters and the Arterial

Lesion

The nonhomogeneity of lipids in the arterial lesions is nowhere better demonstrated than in the study of esterified cholesterol. Lang and Insu1147 have described the presence of 0.5 to 5 pm oil droplets in aortic fatty streaks. The droplets floated when a homogenate was centrifuged; they showed a temperature-dependent birefringence when examined under the polarizing microscope; and they were composed nearly exclusively of esterified cholesterol with a preponderance of the oleate ester. As much as 70 percent of the esterified cholesterol in fatty streaks was found in floating droplets, whereas most of the free cholesterol and phospholipid was associated with tissue fragments that sediment upon centrifugation. Differences in cholesteryl esters in different areas of the lesion have been reported. Cholesteryl ester associated with perifibrous lipid showed a much higher linoleate content than that of fat-filled cells, which were relatively richer in cholesteryl oleate.48-50 The similarity in fatty acid composition of cholesteryl esters found in association with fibrous tissue48 and that of plasma low density lipoprotein, and the observation that cholesteryl esters of very low and low density lipoproteins readily transfer to delipidated elastin,51 has stimulated speculation that these esters originated in serum and might represent an index of the extent of lipoprotein infiltration (see later). However, the existence of cholesterol esterifying and hydrolyzing enzyme systems in the arterial wall suggests that the fatty acids esterified to cholesterol are subject to change within the wall. Radioassay of fatty acid uptake by cholesteryl ester: Cholesteryl ester synthesized from 14C-acetate by the rabbit artery in vivo was shown to contain nearly all of its label in the fatty acid portion of the molecule.52 When aortic strips from cholesterol-fed rabbits were incubated with 3H-cholesterol and 14Cpalmitate, they incorporated the palmitate readily into esterified cholesterol but showed little incorpo-

562

April 1975

The American Journal of CARDIOLOGY

ration of the labeled cholesterol.s” Uptake of labeled fatty acid by the cholesteryl ester fraction of human and animal arteries has been demonstrated by several workers.ss-“” Incorporation of labeled cholesterol into esterified cholesterol appears to be slower, but St. Clair and Lofland56 were able to demonstrate appreciable conversion of labeled cholesterol to its esters in organ cultures of pigeon aortas. It is difficult to estimate from these experiments the quantitative significance of in situ cholesterol esterification compared with that of cholesteryl ester influx from plasma.12 Relation to atherogenic process: Increased incorporation of labeled oleate into cholesteryl ester of pigeon aortas could be demonstrated in cell free preparations as early as 2 weeks and in rabbits as early as 3 days after initiation of cholesterol feeding.57,58 At this time little or no cholesterol had accumulated in the arterial wall. It is evident therefore that fatty acid incorporation into cholesteryl ester is closely linked with the atherogenic process and may well be one of the mechanisms whereby cholesteryl ester droplets accumulate in the arterial wall. This hypothesis is supported by the observation that during regression of lesions the incorporation of oleate into cholesteryl ester is markedly reduced.5g Hashimoto and Dayton,6o on the other hand, were unable to find a relation between the cholesterol-esterifying activity of aortas and the degree of resistance of several species to atherosclerosis, but they and their coworkers61 did observe greater conversion of oleoyl coenzyme A to cholesteryl ester than of palmityl- or linoleyl coenzyme A. Since cholesteryl oleate is the predominant cholesteryl ester of atheromatous lesions it is tempting to assume that esterification of cholesterol in the arterial wall is an integral rather than an incidental part of atherogenesis. Cholesteryl ester hydrolase and atherosclerosis: Two enzyme systems appear to be capable of incorporating labeled fatty acids and cholesterol into arterial cholesteryl ester: one with a pH optimum of 7.5 and requiring adenosine triphosphate and coenzyme A, the other with a pH optimum of 5 and no such cofactor requirements.62 The enzymatic hydrolysis of cholesteryl esters by arterial tissue has also been demonstrated.63*64 Labeled cholesteryl esters added to homogenates of rat or monkey aortas undergo hydrolysis but the activity of the cholesteryl ester hydrolase does not. appear to be related to atherogenesis.‘j4 The interpretation is somewhat difficult, however, because of the vastly different quantities of endogenous cholesteryl esters present in normal and atherosclerotic aortas.57 Cholesteryl ester hydrolase is found in the lysosomal fraction of arterial smooth muscle cells.s5 In the atheromatous aorta of cholesterol-fed rabbits much of the excess cholesteryl ester is found in low density lysosomes that appear to be deficient in this enzyme.66 It is interesting to speculate that hydrolysis of cholesteryl ester is a prerequisite before the sterol can be removed from the cell, and evidence for such a removal mechanism

Volume 35

CHOLESTEROL

exists for peritoneal macrophages.s7 An observation lending credibility to such an interpretation is that species resistant to atherosclerosis such as dogs and rats have a higher concentration of cholesteryl ester hydrolase in the aorta than do susceptible species such as rabbits, swine and guinea pigs.68 In summary, the significance of cholesteryl ester accumulation in the aorta is not at all clear. Does the esterification of free cholesterol by the arterial wall contribute significantly to the accumulation of esterified cholesterol in the diseased artery, or is the main route for cholesteryl ester accumulation an infiltration of esters from plasma? If the former is the case, is the conversion of free to esterified clholesterol a protective mechanism of storing a harmful sterol in a less harmful state,6g or is the accumulation of cholesteryl ester in the hydrophobic interior of the atheromatous lesion the beginning of irreversible fibrotic reactions that ultimately lead to the clinical consequences of vascular disease? Answering these questions is the aim of much current research. Role of Low Density Lipoproteins Atherogenesis

in

Arterial uptake and transmural distribution of low density lipoprotein: The high serum cholesterol concentrations associated with severe atherosclerosis are usually seen in patients with elevated levels of low density (beta) lipoproteins. The first investigation on the uptake of serum low density lipoprotein by the aorta was performed in the dog,l” an animal that has relatively little low density lipoprotein in its blood and is not very susceptible to atherosclerosis. Uptake of low density lipoprotein appe,ared to decrease along the length of the aorta. An in vitro comparison of low density lipoprotein uptake by arteries from different species showed that cockerel and rabbit aortas took up more low density lipoprotein than rat aorta, and that the aortas of cholesterol-fed rabbits took up more than those from rabbits on their normal diet.71 However, ilt is difficult to relate these findings to the relative susceptibilities of these animals to atheromatosis, since the incubated aortic disks had been stripped of their adventitia, and low density lipoprotein uptake probably took place both at the intimal and at the injured medialadventitial surfaces. Scott and Hurley72,73 have attempted to measure the rate of low density lipoprotein uptake in patients with terminal disease. They found that albumin as well as low density lipoprotein labeled with radioactive iodine entered the arterial wall primarily from the intimal side. This finding is at least in partial accord with results of similar studies in cholesterol-fed rabbits; in arteries with severe lesions uptake of labeled albumin and of labeled cholesterol proceeded primarily from the intimal side.18 In beginning lesions and in normal arteries the entry of albumin appeared to occur from the adventitial side.‘74 An interesting but unexplained observation is that the in vitro uptake of iodinated low density lipoprotein and

ACCUMULATION

IN ARTERIAL

WALL-ZILVERSMIT

albumin is reduced by the presence of clofibrate in the medium.75 Similarly, the uptake of injected low density lipoprotein appeared to be reduced in arteries of cholesterol-fed rabbits previously treated with clofibrate.75 In this instance the reduced serum cholesterol concentration could have accounted for the observed smaller low density lipoprotein uptake. The studies of Adams et a1.18 and Virag et a1.76 both point to differences in the pattern and quantities of protein and cholesterol uptake by the arterial wall. Adams et al. compared the transmural distribution of labeled albumin and labeled cholesterol in the aortas of the normal and cholesterol-fed rabbits, whereas .Virag et al. studied rabbit beta lipoproteins that were labeled in the protein and cholesterol moieties. Whereas Adams et al. stressed the difference in transmural distribution of the labeled protein and cholesterol, Virag et al. pointed to the larger uptake (expressed as plasma clearance) of cholesterol than of the protein portion of the beta lipoprotein molecule. This finding should not be surprising in view of previous studies which showed that at least the labeled free cholesterol of plasma exchanges readily with that of atheromatous aortas.12J4 Thus, the uptake process could well consist of lipoprotein uptake plus an additional apparent uptake of free cholesterol by an isotopic exchange reaction. In our laboratory we have examined the in vivo uptake of labeled unesterified and esterified cholesterol by the aorta of normal rabbits of different ages.77 The normal aorta, even in older animals, appeared to be all but impermeable to esterified but not to unesterified cholesterol, even though the uptake of the latter was lower by at least an order of magnitude than that observed by Virag et a1.76 If in the normal artery low density lipoprotein were taken up as an intact molecule, one would expect that esterified cholesterol, which is embedded in the core of this lipoprotein, should also enter the arterial wall. It may therefore be tentatively concluded that the normal arterial intima presents a structural or metabolic barrier to the uptake of low density lipoprotein, a suggestion made earlier on the basis of in vitro studies.36 Interpretation and significance of uptake studies of low density lipoprotein: The uptake of low density lipoprotein by the atheromatous aorta in experimental animals and the presence of low density lipoprotein in diseased arteries78-80 must therefore be interpreted with caution. Although it is tempting to conclude that the uptake of low density lipoprotein by the artery is the cause of atherogenesis, the findings are as easily interpreted as the result of the disease process. A breakdown of a metabolic or a structural barrier in the normal artery would allow these large molecules easier access to binding sites inside the intima or media. Several studies have suggested that mucopolysaccharidess1-83 or elastinsl could bind low density lipoproteins or cholesterol. Investigations aimed at elucidating the sequential relations among arterial disease, arterial permeability and the pres-

April 1975

The American Journal of CARDIOLOGY

Volume 35

563

CHOLESTEROL

ACCUMULATION

IN ARTERIAL WALL-ZILVERSMIT

ence of elevated levels of low density lipoproteins in serum and arterial wall and their interaction with structural components are urgently needed to clear up the present state of confusion. Role of Very Low Density Lipoproteins Chylomicrons in Atherogenesis

and

Clinical and epidemiologic studies84-86 as well as investigations on the genetic basis of lipoprotein abnormalitiess7ys8 have generated increasing evidence that atherosclerosis may be related to elevated triglyceride levels as well as to elevated cholesterol concentrations in serum. Until recently, however, it was difficult to establish a mechanism whereby high concentrations of serum triglycerides could influence the development of arterial lesions. The meticulous studies of Insull and Bartsch8g showed that atheromas in coronary arteries or in the aorta do not contain excessive amounts of triglyceride. A mechanism that reconciles the reported absence of excessive triglyceride in arterial lesions with a functional involvement of hypertriglyceridemia in atherogenesis has recently been suggested.g0 According to this view the lipolytic degradation of triglyceride-rich particles, chylomicrons and very low density lipoproteins, takes place in close association with vascular endothelium or other components of the arterial intimal surface. The presence of sulfated mucopolysaccharides in the arterial intima has been documented in numerous investigations reviewed by Berenson et a1.g1 Some of these mucopolysaccharides interact strongly with chylomicrons and very low and low density lipoproteins and also activate lipoprotein lipase found in adipose tissue,s2 heart muscleg2 and probably also in the arterial wa11,s3-g7 although the lipase in the latter organ has not been fully characterized. Adipose tissue lipase appears to be activated by a factor contained in blood platelets.g8 It is possible that local activation of arterial lipase by platelets adhering to focally injured endothelium may accelerate the degradation of triglycerides in sites that will ultimately have atheromatous lesions. The metabolism of triglyceride-rich lipoproteins that may be involved in atherogenesis might proceed somewhat as follows: Chylomicrons and very low density lipoprotein of endogenous origin are bound to the vascular endothelium by sulfated mucopolysaccharides. Triglycerides are hydrolyzed by a locally

present lipase, the activity of which may be regulated by the peptide composition of the substrateggJOO as well as by a platelet factor.g8 The size of the adsorbed particle is reduced and its composition is changed. Phospholipids and unesterifed cholesterol, which are present primarily in the surface coat of the particulate lipids,iciJos are transferred to the higher density lipoproteins circulating in the bloodstream. A similar transfer might also account for the loss of the smaller C-peptides from the larger complexes.lo3J04 What remains are the undigestible remnants of the original particles, composed primarily of cholesteryl ester and apolipoprotein B. These remnants may be partially released into the bloodstream and partially taken up by endothelial or smooth muscle cells in the arterial intima by a phagocytic or pinocytotic mechanism. The presence of protease, lipase and cholesteryl ester hydrolase activity in arterial lysosomes has been demonstrated,65Js6 and a further degradation of the remnant may take place by means of lysosomal enzymes. It is likely that the foam cells seen in atheromatous lesions represent the end product of a series of reactions. These may begin with the selective binding of low density lipoproteins or triglyceride-rich particles at the intimal surface. They may be followed by the ingestion of these particles, or their partially degraded remnants, by a pinocytotic mechanism and a second stage of digestion by hydrolases derived from lysosomes that coalesce with the pinocytotic vesicles. The final undigestible residues of cholesterol, both free and ester possibly combined with mucopolysaccharides, accumulate in membrane-bound vacuoles as seen in foam cell lesions. Finally, cell death occurs and the lipid-rich content of these cells is spilled into the extracellular spaces. There it remains as a mass of amorphous and liquid crystalline material, metabolically inert and not accessible to cellular or humoral mechanisms that might otherwise resolubilize the lipid deposits. That resolubilization of cholesterol deposits is possible is seen in the reversibility of a cholesterol fatty liver and on a smaller scale in the resolubilization of particulate cholesterol taken up by Kupffer cells in the liver. It is obvious that research on cholesterol and lipoprotein transport both into and out of cells and the influence of diet and drugs on this process might eventually provide a rational basis for the prevention and treatment of atherosclerosis.

References 1. Evolution of the Atherosclerotic Plaque (Jones RJ, ed). Chica2. 3.

4. 5.

564

go, University of Chicago Press, 1963 Adams CWM: Vascular Histochemistry. Chicago, Year Book Medical Publishers, 1967 Atherosclerosis: Pathology, Physiology, Aetiology, Diagnosis and Clinical Management (Schettler FG, Boyd GG, ed). New York, Elsevier, 1969 Friedman M: Pathogenesis of Coronary Artery Disease. New York, McGraw-Hill, 1969 Pollak OJ: Monographs on Atherosclerosis, Vol 1. Tissue Cul-

April 1975

The American Journal of CARDIOLOGY

Volume 35

tures. New York, S Karger, 1969 6. Atherosclerosis, Proceedings of the Second international Symposium (Jones RJ, ed). New York, Springer-Verlag, 1970 7. The Artery and the Process of Arteriosclerosis:Pathogenesis. (Wolf S, ed). New York, Plenum Press, 1971 8. Geer JC, Haust MD: Monographs on Atherosclerosis, Vol 2. Smooth Cells in Atherosclerosis. New York, S Karger, 1972 9. Atherosclerosis and Coronary Heart Disease(Likoff W, Segal BL, lnsull W, ed). New York, Grune & Stratton, 1972 10. Wissler RW, Geer JC: The Pathogenesis of Atherosclerosis.

CHOLESTEROL ACCUMULATION

Baltimore, Williams & Wilkins, 1972 11. Hollander W: Recent advances in experimental and molecular pathology. Influx, synthesis and transport of arterial lipoproteins in atherosclerosis. Exp Mol Pathol 7:248-258, 1967 12. Zllversrnit DB: Cholesterol flux in the atherosclerotic plaque. Ann NY Acad Sci 149:710-724, 1968 13. Zilversmli DB: Metabolism of arterial lipids. In Ref 6, p 35-41 14. Dayton S, Hashlmoto S: Recent advances in molecular pathology: cholesterol flux and metabolism in arterial tissue and in atheromata. Exp Mol Pathol 13:253-268, 1970 15. Whereat AF, Staple E: The preparation of serum lipoproteins labeled with radioactive cholesterol. Arch Biochem Biophys 90:224-228. 1960 16. Niiseon A, Zilversmft DB: Fate of intravenously administered particulate and lipoprotein cholesterol in the rat. J Lipid Res 13:32-38, 1972 17. Newman HAI, Zliversmit DB: Uptake and release of cholesterol by rabbit atheromatous lesions. Circ Res 18:293-302, 1966 18. Adams CWM, Virag S. Morgan RS, et al: Dissociation of [ 3H] cholesterol and 1251-labelledplasma protein influx in normal and atheromatous rabbit aorta. J Atheroscler Res 8:679696, 1968 19. Somer JB, Schwartz CJ: Focal 3H-cholesterol uptake in the pig aorta, Atherosclerosis 13:293-304, 1971 20. Newman HAI, Zilversmlt DB: Quantitative aspects of cholesterol flux in rabbit atheromatous lesions. J Biol Chem 237: 2078-2084, 1962 2 1. Dayton S: Turnover of cholesterol in the artery walls of normal chickens. Circ Res 7:468-475, 1959 22. Field H Jr, Swell L, Schools PE Jr, et al: Dynamic aspects of cholesterol metabolism in different areas of the aorta and other tissues in man and their relationship to atherosclerosis. Circulation 22:547-558. 1960 23. Chobanlan AV, Hollander W: Body cholesterol metabolism in man. I. The equilibration of serum and tissue cholesterol. J Clin Invest 41:1732-1737, 1962 24. Gould RG, Wisaler RW, Jones RJ: The dynamics of lipid deposition in arteries. In Ref 1. p 205-214 25. Dayton S, Hashlmoto S: Movement of labeled cholesterol between plasma lipoprotein and normal arterial wall across the intimal surface. Circ Res 19:1041-1049, 1966 26. Jensen J: The kinetics of the in vitro cholesterol uptake at the endothelial cell surface of the rabbit aorta. Biochim Biophys Acta 1351544-556, 1967 27. Day AJ, Wahlqvist ML, Campbell DJ: Differential uptake of cholesterol and of different cholesterol esters by atherosclerotic intima in vivo and in vitro. Atherosclerosis 11:301-320, 1970 28. Jensen J: On the relationship between metabolic activity and cholesterol uptake by intima-media of the rabbit aorta. Biochim Biophys Acta 183:204-214, 1969 29. Page IH: The Lewis A. Connor Memorial Lecture. Atherosclerosis, an introduction. Circulation lO:l-27, 1954 30. Campbell DJ, Day AJ, Skinner SL, et al: The effect of hypertension on the accumulation of lipids and the uptake of [3H]cholesterol by the aorta of normal-fed and cholesterol-fed rabbits. Atherosclerosis 18301-319, 1973 31. Caro CG, Nerem RM: Transport of ‘%-4-cholesterol between serum and wall in the perfused dog common carotid artery. Circ Res 32: 187-205, 1973 32. Caro CG: Transport of material between blood and wall in arteries. In, Atherogenesis: Initiating Factors (Ciba Foundation Symposium 12). New York, Elsevier, 1973, p 127-149 33. Bondjers G, Bj6rkerud S: Cholesterol accumulation and content in regions with defined endothelial integrity in the normal rabbit aorta. Atherosclerosis 17:71-83, 1973 34. Bondjers G. Bjorkerud S: Arterial repair and atherosclerosis after mechanical injury. Part 3. Cholesterol accumulation and removal in morphologically defined regions of aortic atherosclerotic lesions in the rabbit. Atherosclerosis 17:85-94, 1973 35. Bondjers G, Bjorkerud S: Arterial repair and atherosclerosis after mechanical injury. Part 4. Uptake and composition of cholesteryl ester in morphologically defined regions of atherosclerotic lesions. Atherosclerosis 15:273-284, 1972

IN ARTERIAL WALL-ZILVERSMIT

36. Zilversmlt DB, Newman HAI: Does a metabolic barrier to circulating cholesterol protect the arterial wall? Circulation 33:7, 1966 37. Fry DL: Localizing factors in arteriosclerosis. In Ref 9, p 85104 38. Lofland HB, Clarkson TB: The bidirectional transfer of cholesterol in normal aorta, fatty streaks and atheromatous plaques, Proc Sot Exp Biol Med 133:1-8. 1970 39. Bell FP, Lofland HB Jr, Stokes NA: Cholesterol flux in vitro in aortas of cholesterol-fed and non-cholesterol-fed pigeons. Atherosclerosis 11:235-246, 1970 40. Rutenberg HL, Soloff LA: Possible mechanism of egress of free cholesterol from the arterial wall. Nature 230:123-125, 1971 41. Rutenberg HL, Stern AG, Soloff LA, et al: In vitro serum cholesterol esterification in coronary artery disease. Am Heart J 841348-358, 1972 42. Rothblat GH, Krftchevsky D: The excretion of free and ester cholesterol by tissue culture cells: studies with L5178Y and Lcells. Biochim Biophys Acta 144:423-429, 1967 43. Rothbiat OH, Buchko MK, Kritchevsky D: Cholesterol uptake by L5178Y tissue culture cells: studies with delipidized serum. Biochim Biophys Acta 164:327-338, 1968 44. Bailey JM: Regulation of cell cholesterol content. In Ref 32, p 63-88 45. Nllsson A, Zilversmit DE: Release of phagocytosed cholesterol by liver macrophages and spleen cells. Biochim Biophys Acta 260:479-491, 1972 46. Werb Z, Cohn ZA: Cholesterol metabolism in the macrophage. I. The regulation of cholesterol exchange. J Exp Med 134: 1545-1569, 1971 47. Lang PD, lnsull W Jr: Lipid droplets in atherosclerotic fatty streaks of human aorta. J Clin Invest 49:1479-1488, 1970 48. Smith EB, Slater RS, Chu PK: The lipids in raised fatty and fibrous lesions in human aorta. A comparison of the changes at different stages of development. J Atheroscler Res 8:399419, 1968 49. Smith EB, Slater RS: The microdissection of large atherosclerotic plaques to give morphologically and topographically defined fractions for analysis. Part 1. The lipids in the isolated fractions. Atherosclerosis 15:37-56, 1972 50. Kunnert B, Krug H: The composition of cholesterol esters in fatty streaks and atherosclerotic plaques of the human aorta. Atherosclerosis 13:93-101, 1971 5 1. Kramsch DM, Hollander W: The interaction of serum and arterial lipoproteins with elastin of the arterial intima and its role in the lipid accumulation in atherosclerotic plaques. J Clin Invest 52:236-247, 1973 52. Newman HAI. Gray GW, Zilversmit DB: Cholesterol ester formation in aortas of cholesterol-fed rabbits. J Atheroscler Res 8~745-754, 1968 53. Day AJ, Wahlqvfst ML: Uptake and metabolism of “C-labeled oleic acid by atherosclerotic lesions in rabbit aorta. Circ Res 23:779-788, 1968 54. Wahlqvist ML, Day AJ, Tume RK: Incorporation of oleic acid into lipid by foam cells in human atherosclerotic lesions. Circ Res 24:123-130, 1969 55. Dayton S, Hashimoto S: Origin of cholesteryl oleate and other esterified lipids of rabbit atheroma. Atherosclerosis 12:37 l381, 1970 56. St. Clair RW, Lofland HB Jr: Uptake and esterification of exogenous cholesterol by organ cultures of normal and atherosclerotic pigeon aorta. Proc Sot Exp Biol Med 138:632-637, 1971 57. St. Clair RW, Lofland HB, Clarkson TB: Influence of duration of cholesterol feeding on esterification of fatty acids by cellfree preparation of pigeon aorta. Circ Res 27:213-225, 1970 58. Day AJ, Proudlock JW: Changes in aortic cholesterol-esterifying activity in rabbits fed cholesterol for three days. Atherosclerosis 19:253-258, 1974 59. St. Clalr RW, Clarkson TB, Lofland HB: Effects of regression of atherosclerotic lesions on the content and esterification of cholesterol by cell-free preparations of pigeon aorta. Circ Res 31:664-671, 1972

April 1975

The American Journal of CARDIOLOGY

Volume 35

565

CHOLESTEROL ACCUMULATION

IN ARTERIAL WALL-ZILVERSMIT

60. Hashfmoto S, Dayton S: Cholesterol-esterifying activity of aortas from atherosclerosis-resistant and atherosclerosis-susceptible species. Proc Sot Exp Biol Med 14589-92, 1974 61 Hashimoto S, Dayton S, Alfin-Slater RB, et al: Characteristics of the cholesterol-esterifying activity in normal and atherosclerotic rabbit aortas. Circ Res 34:176-183, 1974 62. Proudlock JW, Day AJ: Cholesterol esterifying enzymes of atherosclerotic rabbit intima. Biochim Biophys Acta 260:716723, 1972 63. Kothari HV, Miller BF, Krffchevsky b: Aortic cholesterol esterase: characteristics of normal rat and rabbit enzyme. Biochim Biophys Acta 2961446-454, 1973 64. Brecher P, Kessler M, Clifford C, et al: Cholesterol ester hydrolysis in aortic tissue. Biochim Biophys Acta 316:386-394, 1973 65. Peters TJ, Muller M, de Duve C: Lysosomes of the arterial wall. I. Isolation and subcellular fractionation of cells from normal rabbit aorta. J Exp Med 136:1117-l 139, 1972 66. Peters TJ, Takano T, de Duve C: Subcellular fractionation studies on the cells isolated from normal and atherosclerotic aorta. In Ref 32, p 197-214 67. Werb 2, Cohn ZA: Cholesterol metabolism in the macrophage. Ill. Ingestion and intracellular fate of cholesterol and cholesterol esters. J Exp Med 13521-44. 1972 68. Bonner MJ, Miller BF, Kothari HV: Lysosomal enzymes in aortas of species susceptible and resistant to atherosclerosis. Proc Sot Exp Biol Med 139:1359-1362, 1972 69. Abdulla YH, Adams CWM, Morgan RS: Connective-tissue reactions to implantation of purified sterol, sterol esters, phosphoglycerides, glycerides and free fatty acids. J Pathol Bacteriol 94:63-71, 1967 70. Duncan LE Jr, Buck K, Lynch A: Lipoprotein movement through canine aortic wall. Science 142:972-973, 1963 71. Vfrag S, Pozsonyi T, Denes R, et al: Uptake of 1251-labeledplipoprotein by the aortas of animals differently susceptible to cholesterol-induced atherosclerosis. J Atheroscler Res 8: 859-860, 1968 72. Scott PJ, Hurley PJ: Low-density lipoprotein accumulation in aortic and coronary artery walls. Isr J Med Sci 5:631-634, 1969 73. Scott PJ, Hurley PJ: The distribution of radio-iodinated serum albumin and low-density lipoprotein in tissues and the arterial wall. Atherosclerosis 11:77-103, 1970 74. Adams CWM, Morgan RS, Bayliss OB: The differential entry of [ i251]albumin into mildly and severely atheromatous rabbit aortas. Atherosclerosis 11:119-124, 1970 75. Vi&g S, Denes R, Pozsonyi T, et al: Effect of miscleron (clofibrate) on the labelled lipoprotein uptake of the aorta in normal and cholesterol pretreated rabbits. Ther Hung 20:62-64, 1972 76. Virag S, Denes R, Pozaonyi T: A study into transport characteristics of labeled @-lipoproteids passing through the aortic wall in rabbits with experimental atherosclerosis. Kardiologiia 12:61-65, 1972 77. Zilversmit DB, Hughes LB: Incorporation in vivo of labeled plasma cholesterol into aortas of young and old rabbits. Atherosclerosis 18:141-152, 1973 78. Smith EB, Slater RS: Relationship between low-density lipoprotein in aortic intima and serum-lipid levels. Lancet 1:463469. 1972 79. Smith EB, Slater RS, Hunter JA: Quantitative studies on fibrinogen and low-density lipoprotein in human aortic intima. Atherosclerosis 18:479-487, 1973 80. Smith EB, Slater RS: Relationship between plasma lipids and arterial tissue lipids. Nutr Metabol 15:17-26, 1973 81. Virag Sh, Denesh R, Porhonji T: Interrelation between the content of acid mucopolysaccharides in the aortic wall and the degree of absorption of beta-lipoproteins. Biull Eksp Biol Med 70:37-39, 1970 82. Srinivasan SR, Dolan P, Radhakriahnamurfhy B, et al: Isolation of lipoprotein-acid mucopolysaccharide complexes from

566

April 1975

The American Journal of CARDIOLOGY

Volume 35

83

84. 85.

86.

87.

88.

89.

90.

91.

92.

93.

94. 95.

96. 97.

98. 99.

100.

101. 102. 103.

104.

fatty streaks of human aortas. Atherosclerosis 16:95-104, 1972 Srinivasan SR, Dolan P, Radhakrishnamurfhy B, et al: Isolation of lipoprotein-acid mucopolysaccharide complexes from fatty streaks of human aorta. Prep Biochem 2:83-91, 1972 Albrink MJ, Meigs JW, Man EB: Serum lipids, hypertension and coronary artery disease. Am J Med 31:4-23. 1961 Heinle RA, Levy RI, Fredrickson DS, et al: Lipid and carbohydrate abnormalities in patients with angiographically documented coronary artery disease. Am J Cardiol 24:178-186, 1969 Carlson LA, Bottiger LE: lschaemic heart disease in relation to fasting values of plasma triglycerides and cholesterol. Lancet 1:885-868, 1972 Goldstein JL, Hazzard WR, Schrott HG, et al: Hyperlipidemia in coronary heart disease I. Lipid levels in 500 survivors of myocardial infarction. J Clin Invest 52: 1533- 1543. 1973 Goldstein JL, Schrott HG, Hazzard WR, et al: Hyperlipidemia in coronary heart disease II. Genetic analysis of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia. J Clin Invest 52: 1544- 1568, 1973 lnsull W Jr, Barfsch GE: Cholesterol, triglyceride and phospholipid content of intima, media, and atherosclerotic fatty streak in human thoracic aorta. J Clin Invest 45513-523, 1966 Zllversmif DB: A proposal linking atherogenesis to the interaction of endothelial lipoprotein lipase with triglyceride-rich lipoproteins. Circ Res 33~633-638, 1973 Berenson GS, Radhakrishnamurthy B, Dalferes ER Jr, et al: Carbohydrate macromolecules and atherosclerosis. Hum Pathol 2:57-79, 1971 Robinson DS: Function of the plasma triglycerides in fatty acid transport. In, Comprehensive Biochemistry, Vol 18. Lipid Metabolism (Florkin M, Stotz EH, ed). New York, Elsevier. 1970, p 51-116 Patelski J, Waligora Z, Szulc S: Demonstration and some properties of the phospholipase A, lipase and cholesterol esterase from the aortic wall. J Atheroscler Res 7:453-461, 1967 Zemplenyi T: Enzyme Biochemistry of the Arterial Wall. LloydLuke, London, 1968, p 224 Patelaki J, Bowyer DE, Howard AN, et al: Changes in phospholipase A, lipase and cholesterol esterase activity in the aorta in experimental atherosclerosis in the rabbit and rat. J Atheroscler Res 8:221-228, 1968 Kirk JE: Enzymes of the Arterial Wall. New York, Academic Press, 1969, p 222-232 Patelski J, Bowyer DE, Howard AN, et al: Modification of enzyme activities in experimental atherosclerosis in the rabbit. Atherosclerosis 12:41-53, 1970 Song S-Y, Garffnkel AS, Schotz MC: Activation of lipoprotein lipase by blood platelets. Submitted for publication Have1 RJ, Shore VG, Shore B, et al: Role of specific glycopeptides of human serum lipoproteins in the activation of lipoprotein lipase. Circ Res 27:595-600, 1970 Brown WV, Baginsky ML: Inhibition of lipoprotein lipase by an apoprotein of human very low density lipoprotein. Biochem Biophys Res Commun 46:375-382, 1972 Zilversmit DB: The surface coat of chylomicrons: lipid chemistry. J Lipid Res 9:180-186, 1968 Salpeter MM, Zilversmlt DB: The surface coat of chylomicrons: electron microscopy. J Lipid Res 9: 187-192, 1968 Eisenberg S, Bllheimer DW, Levy RI: The metabolism of very low density lipoprotein proteins. II. Studies on the transfer of apoproteins between plasma lipoproteins. Biochim Biophys Acta 280:94-104, 1972 Eisenberg S, Bilhefmer DW, Levy RI, et al: On the metabolic conversion of human plasma very low density lipoprotein to low density lipoprotein. Biochim Biophys Acta 326:361-377. 1973