Cellular proliferation in atherosclerosis and hypertension

Progress in

Cardiovascular VOL. XXVI,

Cellular

MARCH/APRIL

Proliferation

in Atherosclerosis

Stephen

M. Schwartz

A

THEROSCLEROSIS and hypertension, the two major forms of vascular disease, share a number of features. The frequency of both diseases increases with age, and the severity of disease represents a continuum from no disease to severe illness. The risk of atherosclerosis is greatly increased in individuals with hypertension. The comparison extends beyond the level of natural history and epidemiology, to the pathology of these diseases. Both diseases are characterized by smooth muscle proliferation. The central cellular feature of atherosclerotic lesions is proliferation of smooth muscle cells in the arterial intima of larger arteries. As these smooth muscle lesions enlarge, lipid accumulates, thrombosis occurs, the lumen is narrowed, and patients die of infarction. Smooth muscle proliferation is also characteristic of hypertensive vascular disease. Smooth muscle proliferation or hypertrophy of smaller arteries results in increased wall mass and a narrowed lumen. The small vessel change is thought by some to be etiologic of the increased peripheral resistance that causes high blood pressure. This common role of accumulation of smooth muscle cells suggests that control of cell proliferation may be critical to both diseases. CELL PROLIFERATION

IN ATHEROSCLEROSIS

Virchow’ recognized the presence of cell proliferation in atherosclerosis over a century ago. His view of the cellular events as a reaction to the accumulation of toxic materials in the vessel wall was largely neglected while research was directed at the equally important issue of lipid accumulation in lesions. This began to change in the latter part of the 1960s and early 1970s. Experimental evidence in animals, and observaProgress

in Cardiovascular

Diseases

NO. 5

Diseases,

Vol. XXVI,

No. 5 (March/April),

1984

and Hypertension

and Russell Ross

tions in humans, supported the idea that the initial step in lesion formation, prior to an increase in intimal lipids, was the formation of focal masses of proliferated smooth muscle cells in the intima.24 These observations began to lead to a possible understanding of mechanism. Ross et al discovered that platelets contain a growth factor able to stimulate cell proliferation in culture and potentially present at sites of endothelial denudation in the injured vessel wa11.536These studies in culture and in animals occurred along with detailed histologic studies of lesion formation in the coronary arteries in humans, showing a sequence from minor changes in vessels of the very young to diffuse and focal proliferation of cells in the intima of slightly older vessels, with characteristic lipid accumulation in lesions at yet an older age.’ The result has been a new emphasis on cell proliferation, to the point where it is reasonable to ask whether the question of the origin of atherosclerotic lesions will be answered once we understand how and why smooth muscle cells accumulate. CELL PROLIFERATION

IN HYPERTENSION

The history of our understanding of the role of smooth muscle cell proliferation in hypertension is less clear. Arteriolar hypertrophy in hypertension was described as early as 186g8 and the concept that these thickened vessel walls might

From the Department of Pathology, University of Washington, Seattle, WA. Address reprint requests to S. M. Schwartz, MD, PhD. Department of Pathology SM-30, University of Washington, Seattle, WA 98195. 0 I984 by Grune & Stratton, Inc. 0033~620/84/2605~001$02.00/0 1984

355

356

increase peripheral resistance was offered by Ewald’ in 1877. These morphological observations meant little until physiologists began to understand the central role of smooth muscle contractility in controlling resistance to flow through small arteries. Only a small proportion of cases of hypertension is readily explained on the basis of classical renal, adrenal, or sympathetic neural mechanisms,” thus abnormal function of the arterial smooth muscle cell must play a central role. The most obvious is that hypertension results from a general change in smooth muscle contractility or responsiveness to vasoactive stimuli. A variety of complex hypotheses and observations in a wide range of forms of hypertension suggest the possibility that inborn errors in membrane metabolism account for altered smooth muscle function.“-” The appeal of these “functional” concepts of smooth muscle contribution to vascular resistance is their yield of exciting scientific insights into basic cellular functions. If such insights are correct, there is good reason to hope that abnormal cell functions can be treated by drugs that we already understand: membrane agents, channel blockers, receptor analogues, and so on. In comparison, a hypothesis based on structural arguments is less immediately appealing. This is particularly true because a very small, fixed change in the ratio of wall thickness to internal radius can produce a massive change in pressure regulation. Folkow” estimates that a 5% change in the average internal radius of resistance vessels is capable of producing a 25% increase in peripheral resistance. Thus, it may be difficult even to measure the small morphologic changes required to produce a large physiologic effect. Nonetheless, there is excellent evidence that structural changes occur in hypertension and account for a substantial portion of the increase in pressure.” We do know that the resistance to flow through peripheral vessels of humans and animals with hypertension is increased even when the vessel wall is totally relaxed. This has been interpreted as evidence that chronic hypertension is associated with vessel wall narrowing, presumably resulting from a thickening of the vessel wa11.‘“*‘6-‘8 From physiologic data, Folkow estimates that the average internal radius of the vessels of a spontaneously

SCHWARTZ

AND

ROSS

hypertensive rat is narrowed by 6%-7%, while the vessel wall is about 35% thicker than in a normal rat.” These average values for the whole vessel bed agree with morphometric studies showing a lo%-15% increased ratio of wall thickness to internal radius.‘9-22 The basis of this increase in vessel wall mass is unknown. Available data, however, suggest that the change involves all elements of the vessel wall, including an increase in smooth muscle cell mass.” This idea of an increase of cell mass is clearly somewhat different from the concept of an increase in cell number as seen in atherosclerosis. In hypertension, the principal measures of mass have been physiologic studies showing an increase in the total contractile mass of the vessel wall,” and morphometric studies showing an increase in the volume of smooth muscle cytoplasm. ‘6-22Recent studies from one of our laboratories suggest that these phenomena, that is, increased mass in hypertension, also depend on cell replication. These studies began as an attempt to quantify the size of smooth muscle cells isolated from vessel walls. As a control for the intactness of his preparations, Owens et al determined the amount of DNA isolated per cell. This value was within normal limits for diploid cells when the isolated cells came from vessel walls of normotensive rats. With cells from hypertensive arteries, however, DNA content appeared 30% in excess over the normal diploid value. This surprising finding was accounted for by measuring the DNA content of each individual cell. When this was done, the hypertensive vessel turned out to contain as many as 50% of cells with a tetraploid or higher DNA content.23*24Moreover, when the DNA content of the wall was corrected for ploidy, there appeared to be no actual increase in cell number, contrary to earlier conclusions based on simple estimates of DNA content.25 This work, all done in the aorta and carotid arteries, showed that the increase in mass is associated with endoreplication, at least in large vessel smooth muscle cells. Since polyploidy represents a permanent change in a cell, it is possible to propose that some forms of chronic hypertension may be secondary to an increase in DNA content following acute hypertensive injury with secondary chronic increases in protein synthesis in the vessel wall. The result would be a common end-stage form of chronic hyper-

CELL

357

PROLIFERATION

tension, apparently idiopathic, but reflecting several different initiating mechanisms. ENDOTHELIAL DENUDATION ATHEROSCLEROSIS

IN

The proposition that the endothelial cell is critical in these vascular responses to injury has grown out of current concepts of the role of thrombosis in stimulating formation of atherosclerotic lesions. The potential importance of endothelial denudation and platelet contribution to atherogenesis was originally highlighted by Rokitansky26 and later by Duguid,*’ who postulated that platelet accretion contributed to the development of the atherosclerotic mass. Clearly, organized thrombi represent an important component of the increase in lesion size. This, however, is different from the notion that platelets may provide an important initiating role in the proliferative response via the platelet-derived growth factor as suggested in the response to injury hypothesis of atherogenesis.6 When the endothelium of a large artery is removed by abrasion, platelets adhere, aggregate, and degranulate. Subsequent to this, smooth muscle cells of the vessel wall respond by migrating across the internal elastic lamina into the intima, and there occurs accelerated replication of both intimal and medial smooth muscle cells. These events result in a characteristic smooth muscle hyperplasia. ROSS,~ Bjorkerud and Bondjers,*’ Stemerman et a1,29Moore et a1,30 and others have proposed that this process is an accelerated model for the events leading to the cellular accumulation characteristic of atherosclerotic plaques.2+32 Evidence in favor of a role for platelets in atherosclerosis came originally from a number of investigators who noted similarities between organization of mural thrombi and evolution of atherosclerotic plaques. In 1974, however, Ross e! al5 noted that serum prepared in the absence of platelets, plasma-derived serum, would not support growth of cultured smooth muscle cells. These cells would grow if platelet releasate was added to the plasmaderived serum. This work led to the description of a platelet-derived growth factor and substantial evidence that this material represents the principal mitogen present in serum. At the same time, in vivo evidence has supported the posit that the response to endothelial denudation is mediated

via platelet interaction with the vessel wall. Friedman et a133and Moore et a13’ were able to inhibit the smooth muscle cell response by making rabbits thrombocytopenic. Interpretation of these experiments is problematic, since the manipulations required to create thrombocytopenia (radiation plus antiplatelet serum) were quite severe. Clowes and Karnovsky3’ were unable to inhibit the smooth muscle response using drugs that interfere with platelet function. It was not clear, however, that the doses used were adequate to achieve the desired levels in vivo. While they did find inhibition of intimal hyperplasia when heparin was infused, recent studies from the same laboratory have shown that heparin has a direct inhibitory effect on growth of smooth muscle cells.34 In summary, these studies demonstrated that platelet-derived growth factor (PDGF) is present at sites of injury in vivo and is able to initiate DNA synthesis in smooth muscle cells in culture.5.35-38 The factor PDGF, is a highly ~characterized polypeptide localized normally in the platelet alpha-granule.39 Immunocytochemical studies have shown the release of platelet alpha-granule contents at sites of endothelial denudation.40 Abolition of smooth muscle proliferation occurs when animals are depleted of platelets,33 or when platelet adherence and release is deficient.41 Thus, there is little doubt that thrombosis, and therefore endothelial denudation, can stimulate proliferation of vascular smooth muscle cells. There is some evidence for endothelial cell loss as an early step in atherosclerosis. Risk factors increase endothelial turnover. Acute hypertension causes an increased uptake of thymidine in rat aortic endothelium.42 Similarly, in the early stages of experimental hypercholesterolemia in the rabbit there is a report of increased thymidine labeling of the endothelium.4’ This increase in DNA synthesis may precede the development of the lesions, since in chronic experimental models for both hypertension and hyperlipemia the replication rate appears to be norma1.43-45 The localization of normal cell turnover is also consistent with a possible etiologic role in atherosclerosis. Cells labeled with tritiated thymidine appear in clusters at branching points, both in normal and hypertensive animals.42*46v47 This clustering may be related to increased cell turnover at sites of hemodynamic strain in these

SCHWARTZ

358

regions.48 Under experimental conditions, endothelial cells can be detached by shearing forces in excess of 200 dynes/cm’ from perfused vessel walls.48 Such forces can occur in vivo, and if single cells or small patches of cells are removed, the endothelium has the capacity to reestablish continuity rapidly.4v In the normal rat such small areas of desquamation are covered by migration of surrounding cells, without proliferation, in a few hours. In contrast, areas of extensive denudation may never heal,50 possibly because the endothelium in vivo eventually runs out of replicative capacity much as it does in culture.5’~52 It is conceivable that prolonged normal turnover at single sites, or changes in turnover seen in hypertension and hyperlipemia may ultimately lead to a loss of endothelial continuity. These data show that endothelial cells turn over and therefore desquamate. The hypothesis of platelet interaction, however, requires endothelial denudation. The literature is somewhat confusing on this subject. This may be due in part to the species under study, the time course of lesion formation, and the methods used to detect

Fig 1. Scanning electron micrograph of the surface been hypercholesterolemic (approximately 500 mg/dL surface of the lesion has retracted, exposing several thrombus consisting of numerous adherent degranulated

AND

ROSS

injury. Many square meters of the endothelial surfaces of rabbits, rats, monkeys, and other species have been examined by scanning electron microscopy. The most extensive studies, however, have been of rabbits. These studies, in normal and hyperlipemic animals, have shown a continuous layer of endothelial cells covering the surface of normal arteries, even in the early stages of lesion formation.53-s7 Since we know that thrombotic material appears in lesions as they progress, this lack of evidence for discontinuity in normal animals indicates that sampling may be a problem, that denudation does not occur in the unmanipulated animal, that denuded areas are too small to be detected by the scanning electron microscope, that regeneration is sufficiently rapid that denudation is only a transient affair, or that denudation occurs during the progression of the disease rather than in its most initial stages. Again it is important to consider the period of time elapsed after induction of hypercholesterolemia and the regions of the arterial tree that were sampled. Recently Faggiotto et al *’ have studied the effects of

of a fatty streak in the right iliac artery of a pigtail plasma cholesterol) for five months. The endothelium underlying lipid-laden macrophages, one of which platelets (X 1 ,ooO).

monkey that had covering the is covered by a

CELL

PROLIFERATION

chronic hypercholesterolemia (plasma cholesterol 300-800 mg/dL) in pigtail monkeys over a period of 12 months. They were unable to detect any endothelial denudation for the first four months-even at these high levels of plasma cholesterol. By five months endothelial denudation, platelet adherence and aggregation were observed over fatty streaks at sites in the iliac bifurcation and in the leg arteries (Fig 1). The denudation occurred over accumulations of fatfilled macrophages in the intima. Denudation higher in the arterial tree was not observed until 10-12 months on the diet; therefore, species, duration, and site are all important variables. It is also important to consider the possibility that brief, repeated exposures of the subendothelium might lead to repeated episodes of release of PDGF at a focal site without the appearance of denuded areas detectable by current methods. The endothelium can recover an area one cell wide within about three hours!9 Turnover studies in the same species imply a rate of cell loss of approximately 1 x lo-’ to 1 x 1O-3 per day.59*60 These values may be combined to estimate the total average area of denudation present at one time: (these assumptions produce maximal estimates for the exposed area). (1) E = ART; where E = denuded area = (m*/cell). (2) E = Approximately 0.125 m’/cell; A = area of each cell, approximately 10’ m*, R = rate of turnover, approximately 1O-3 days,-’ T = time required to replace each cell, approximately three hours. (3) D = E / (C/2) = 0.001, with D = width of denuded space between cells and C = circumference of an endothelial cell, approximately 200 m. An average separation of only 0.001 would be difficult to detect at the usual level of resolution available by scanning electron microscopy. Even assuming a focal rate of cell turnover of 1% (lo-’ days-‘), the value for D is less than the diameter of a platelet. In an attempt to resolve this issue, Reidy et al reexamined the effects of endotoxin on the rat. Endotoxemia provides a well-documented form of desquamating endothelial injury in this species.6’q62The studies of Reidy et al, however, show that turnover increases without formation of denuded areas. The presence of cells loosely attached to the monolayer implies that this lack of denudation is a result of the coordinated

359

undermining of detaching cells by adjacent viable endothelial cells. At least in this particular case, this appears to represent a form of nondenuding desquamation.63 Another aspect of endothelial denudation and platelet adherence at such sites concerns whether the platelet interaction is a single event, or whether multiple events occur with repeated denudation. This distinction is significant since we know that small areas of denudation heal within a few days following a narrow “scratch” of the endothelium. These wounds at least, in a rabbit or rat, do not lead to smooth muscle proliferation.49,64~65 Moreover, an attempt to use antiplatelet drugs to intervene in the response to a single episode of deendothelialization was unsuccessful.3’ These data in vivo may have an analogy in studies of smooth muscle cells in culture. Freshly isolated cells proliferate poorly even in the presence of platelet-released growth factors.66 In one report, the responsive phase is prolonged if cells are maintained in plasmaderived serum, that is in normal culture medium lacking growth factors.@j Finally, we need to consider the possibility that multiple denuding events may alter the thrombogenicity of the exposed surfaces as shown in recent studies of the aortic surface following a second removal of the endothelium with the balloon catheter.67 A sequence of many injurious events may be necessary before a lesion begins to develop. For example, the severity of lesion formation after mechanical trauma is exacerbated by repeated trauma or by lipid feeding. 68This may be related to the marked differences in the thrombotic and coagulative events seen when a ballooned vessel is traumatized for a second time.69 In summary, more attention needs to be given to interactions occurring among the components of the lesion, its cells, connective tissue, and lipid content, as the lesion develops. The best available evidence suggests that detection of endothelial denudation does not occur in the normal vessel or in the early stage of change induced by experimental conditions. We need to point out, however, that this does not say that the endothelium maintains continuity once lesions have begun to develop. Thrombosis, as part of lesion development, is well documented and apparent at early stages of hypercholesterolemia in the monkey.

360

SCHWARTZ

Endothelial denudation may have a critical role in lesion progression from a relatively benign state to a more serious lesion. This, in turn, implies that the ultimate loss of the ability of the endothelium to maintain continuity is a crucial event in the natural history of atherosclerotic lesions. ENDOTHELIAL

DENUDATION

,.:,., ‘(.‘,..;

Endothelial Injury

.,..! :;:. ‘.:,:f.’

, : ; . . . :‘.::.: .::: . . ..,

AND

ROSS

. . . :‘:.:<.:.;;,.:.~, ..‘;.:.;‘..‘.: .,.‘.:

,.

w~~::.‘~~...~~:zG%+

IN HYPERTENSION

Endothelial denudation is a major event in severe acute hypertension. As discussed below, it is known that the endothelium loses continuity, that platelets adhere to the injured vessel wall, and that arterial smooth muscle proliferation occurs following these injuries. The analogy to current concepts of atherosclerosis should be quite clear. Changes during acute hypertension could, at least in part, be responsible for the increases in cell number or mass observed in chronic hypertension. There is abundant evidence for smooth-muscle hyperplasia in response to acute experimental hypertension in both small arteries and the aorta.” Observations of this response suggest that smooth muscle hyperplasia may result from one or both of two distinct processes. The first, which we associate with the hypothesis of a platelet effect on cell growth, is a response to trauma occurring at sites of severe plasma exudation;‘7~70*7’ the second is a more direct physiologic response of the smooth muscle cells to increased wall tension.2s*72*73Our interest in the first of these hypotheses stems from resemblance of the response of these small vessels following hypertensive injury to the response of the aorta to mechanical denudation (Fig 2). Studies directed toward the role of platelets in the hypertensive microvasculature are lacking. There is, however, evidence that hypertensive injury includes loss of endothelial continuity, exposure of the subendothelium, platelet adherence, and proliferation of smooth muscle cells. A number of investigators have shown that the endothelium of small arteries and arterioles breaks down during the acute phase of hypertension.7”86 This occurs in dilated segments that alternate with constricted segments (Figs 3 and 4). The latter do not show evidence of loss of endothelial integrity. 83,85The most extensive evidence of injury comes from studies of the exudation of plasma solutes using isotope-labeled plasma proteins, fluorescein-labeled plasma pro-

arteriolanclerosis

It

fibroystreak hyalne arteriolarsclerosis

!:A:j:i Plasma solutes S Collagen w Elastin

Small-vessel arteriosclerosis, or arteriolarscleFig 2. rosis, has many features in common with large-vessel atherosclerosis. This diagram outlines mechanisms whereby both lesions might originate from a common source, endotheliel injury, which leads to entry of serum factors that stimulate replication of smooth muscle cells. In the lerge vessel, the result is accumulation of smooth muscle cells in the intims and formation of an atherosclerotic plaque. In small vessels, the result is hypertrophy, hyperplasia, and fibrosis of the vascular media.

teins, colloidal carbon or horseradish peroxidase More direct evidence for as tracers. 75*76~78,7g*82-84,86 loss of endothelial continuity, with exposure of the subendothelium and deposition of platelets on the exposed subendothelium, has come from studies using transmission electron microscopy.83-8’ The proliferative response of the smooth muscle cells of the arterial-wall following acute hypertensive injury is well documented in both rabbits and rats. Using tritiated-thymidine ( [3H]TdR) autoradiography or scintillation spectrometry, various investigators have demonstrated increased cell replication following DOCA salt-induced hypertension,87 aortic constriction, or renal hypertension.” The studies

CELL

PROLIFERATION

Fig 3. Photograph of an arteriole of the intestinal wall after 50 minutes of angiotensin II (Peninsula Laboratories, Belmont, CA) (5 micrograms per five-minute intervals) induced acute hypertension. Vessels were fixed in situ in 2.5% glutaraldehyde and 1% formaldehyde in phosphate buffer. Note the bead-like structure of the arteriole in which constricted segments alternate with dialated segments. Areas like these are characterized by endothelial denudation as shown by scanning electron microscopy and permeability studies’-.“’ (x 105).

by Bevan et al are particularly relevant because they represent the most detailed longitudinal studies of the cell kinetics of smooth muscle cells in the hypertensive anima1.70388Bevan used rabbits made hypertensive by constriction of the aorta just proximal to the superior mesenteric artery. Detailed quantitative data are given for elastic and large muscular arteries. These show, in the hypertensive vessels only, an increase in replication with a peak response within 14 days. Subsequently, replication falls to values close to, but slightly elevated above normal. No comparable response was seen in the normotensive vessels below the constriction. While Bevan did not gather quantitative data on the response in small vessels, she suggests, on the basis of inspection of the autoradiograms, that the time sequence of

361

events was similar. Crane and Dutta,87 Fernandez and Crane,” and Bevan” describe a maximal response at the sites of most marked exudation of the plasma. The proliferative response included fibroblasts and other cells adjacent to the site of injury. It is reasonable to guess that these sites represent regions of endothelial injury, and that products of thrombosis, including PDGF, are available at these sites. The response of nonsmooth muscle cells near the site of injury supports the role of some humoral factor, rather than a simple physiologic response of smooth muscle cells to increased tension. It is also interesting to note that Fernandez and Crane found a two- to threefold increase in replication of renal tubular epithelium in the hypertensive kidney following constriction of the aorta between the two renal arteries.” The hyperplasia was not due to a circulating factor, since no increase was seen in the normotensive kidney. The only more direct evidence implicating platelets in a proliferative response of small vessel smooth muscle cells comes from a preliminary report recently published by Potvliege and Bourgain.*’ These authors found morphological evidence of smooth muscle hypertrophy following focal injury to the vessel wall by electric shock. The smooth muscle changes were exacerbated when the platelet response was increased by administration of ADP, but could be prevented when the animals were made thrombocytopenic by a crude preparation of antiplatelet serum. These studies imply that the extent of injury and the time course of endothelial regeneration are important in determining the sequence of events after endothelial injury in large vessels and may also be critical to the outcome of hypertensive injury in small vessels. The extent of injury in hypertension is unknown. Available data, however, suggest that endothelial denudation is foca1.76,84It is also unclear whether the denudation results from cell loss or separation of cells at cell junctions due to cell contraction or stretching the endothelium beyond its tensile limits. There are only anecdotal, nonquantitative data to suggest that areas with increased insudation of plasma proteins show a greater elevation of smooth muscle cell replication than do areas where the only change is an increase in wall tension.70*90As for endothelial repair, it is possible that endothelial injury in small vessels may

362

Fig 4. Scanning electron micrograph by intravenous injections of angiotensin exposed subendothelium (x3.910).

SCHWARTZ

of the luminal II and a single

surface endothelial

persist much longer than in our large vessel model, particularly if, as suggested by Olsen, the underlying elastic lamina is disrupted in the injured, dilated areas.*’ Available morphological studies provide no information about the time course of repair of endothelial injury in small vessels. Cell kinetic data, however, suggest that the endothelial response to acute hypertension may go on for several weeks. The relevant data for small vessels come from Bevan’s study of animals made hypertensive by partial constriction of the aorta proximal to the superior mesenteric artery. She found no change in cell replication in vessels in normotensive areas. Hypertensive vessels showed a marked increase of endothelial cells, as well as smooth muscle cell replication during the first two weeks after surgery, with a subsequent return to normal levels.70 These data for small vessel endothelium were not quantitative. More quantitative data from Schwartz and Benditt4’ as well as data of Kunz et a1,47show that there is also a dramatic increase in aortic endothelial cell replication during acute Goldblatt hypertension. The basal rate of repli-

of a rat mesenteric cell has been

AND

ROSS

artery. The animal was made hypertensive lost. Note that platelets are adhering to the

cation in the endothelium of adult rats is extremely low, representing only 0.1%4.3% of the endothelial cells per day. Longitudinal data in rats at different ages indicated that this number most probably represents the basal rate of cell turnover.” The frequency of replicating endothelial cells, however, was increased tenfold in acute Goldblatt hypertension. We have now extended these studies to examine endothelial cell replication in chronically hypertensive rats. These studies have used two systems of hypertension, the Goldblatt rat and the spontaneously hypertensive rat developed by Okomoto. We found no evidence of a significant increase in replication in the Goldblatt rat with chronic hypertension, as compared to its control rat. Similarly, we found no evidence of a difference in endothelial cell replication when SHR rats were compared to normotensive WKy rats. However, when the SHR rats were put on an antihypertensive regimen, we found that the treated rats showed a decrease in cell replication. These results are perplexing. On the basis of the .Goldblatt experiment, we might suggest that there is

CELL

PROLIFERATION

363

no elevation of endothelial cell replication in chronic hypertension. This would be consistent with the comparison between SHR and WKy, the parental strain of SHR. The data in the treated SHR, however, suggest either that the antihypertensive drugs have some specific effect on replication in the hypertensive rat or that the WKy constitutes a poor genetic control for the SHR.92 In the latter case, chronic hypertension may be associated with chronic elevation in endothelial turnover.44.92

GROWTH

INTERACTIONS THROMBOSIS

WITHOUT

Until now, the emphasis of this review has been on stimulation of smooth muscle cell growth by loss of endothelium. We need to consider the possibility that endothelial cells may themselves directly regulate smooth muscle cell proliferation. There is no direct evidence for this in vivo. Recently, however, Cherry et a193reported that endothelial cells can release an, as yet, undefined vasodilator in response to acetylcholine. Indeed this, rather than a direct effect on smooth muscle cells, seems to be the major effect of acetylcholine on vascular smooth muscle. This opens up a new area: the endocrinology of factors or mediators released by one vascular wall cell and affecting the other.94-96 This concept extends to recent studies of growth interactions. Chamley-Campbell et a19’ found that a subpopulation of freshly isolated smooth muscle cells are not responsive to growth factors. This nonproliferative state was prolonged by cocultivation with endothelial cells. Similarly, Castellot et a134reported that endotheha1 cellconditioned medium could inhibit the normal ability of serum to stimulate replication of smooth muscle cells despite the presence of PDGF. In contrast, we have explored the effects of endothelial cells on smooth muscle cells under conditions more like the intact vessel wall, that is, in the absence of PDGF. Under these conditions, endothelial cells synthesize a polypeptide mitogen, endothelial cell-derived growth factor (ECDGF).98 ECDGF stimulates smooth muscle cells in the absence of other mitogens. Available biochemical data imply that ECDGF is a distinct polypeptide from PDGF. The endothelial cell factor has a molecular mass of about 20,000

daltons and an acidic, rather than a basic, charge. Before deciding that ECDGF and PDGF are different, however, we need to note that the peptides have been studied from different species (bovine versus primate).99 Furthermore, recent results suggest that a portion of the mitogenic material secreted by endothelial cells may be able to compete for the PDGF receptor, and that endothelial cells may secrete more than one mitogen.99 The existence of a smooth muscle cell growth factor derived from endothelial cells clearly presents a teleologic paradox. We have no evidence that ECDGF exists in vivo, and no evidence from in vitro studies for conditions that could modulate ECDGF synthesis in a manner consistent with a normal or pathologic function in vivo. In considering possible roles for this material, however, the possibility occurs that an endothelial cell factor could play a role in vascular development. In all forms of angiogenesis, endothelial growth or differentiation precedes the appearance of smooth muscle cells. To test this possibility, endothelial cells were added to an embryonic vascular bed. When bovine endothelial cells were placed on chick chorioallantoic membranes, however, there was no evidence that these foreign cells induced smooth muscle differentiation. There was, however, a dramatic angiogenic response. This response could also be produced by partially purified ECDGF.lm The relevance of angiogenesis to smooth muscle growth in arteries remains unclear. Other abnormal functions of endothelial cells might be important in smooth muscle proliferation. The availability of endothelial cells in culture has led to rapid progress in our understanding of the things endothelial cells can do. These include antithrombotic and anticoagulant functions likely to be important in interactions of the vessel wall with platelets. There is, however, no evidence that the intact endothelium ever becomes thrombogenic.“’ In contrast, we do have evidence of a role for the interaction between endothelial cells and leukocytes in both atherosclerosis and hypertension. Monocyte adhesion to the endothelium during the early phase of hyperlipemia has been known for 20 years. Recently this phenomenon has achieved a recrudescence of interest because of evidence that monocytes accumulate lipid and can become

364

SCHWARTZ

foam cells,“’ and the observation that these cells, like the platelet, can release a growth factor. ‘03-‘05Similarly, there are data, although less extensive, for an interaction of endothelium with leukocytes in some forms of hypertension.lo6 The nature of the changes responsible for these interactions in leukocytes or endothelium have begun to be explored, using in vitro model systems~107-109

INTRINSIC

CHANGES IN SMOOTH PROLIFERATION

MUSCLE

Before proceeding, we should consider the possibility that abnormal smooth muscle proliferation in atherosclerosis results from events intrinsic to smooth muscle itself. Thomas et al”’ suggested that preformed, intimal accumulations may be the ground for development of lesions. The evidence for this comes from our knowledge of vascular development. The aorta of the newborn animal of all species contains few, scattered smooth muscle cells in the intima.“’ With age, the normal intima shows an increase in numbers of intimal cells, as well as an increase in extracellular connective tissue.“‘,“’ These are diffuse changes that occur in humans regardless of the presence of atherosclerosis, and are even found in animal species that do not develop atherosclerosis. Spontaneous accumulation of cells in the intima suggests either that embryologically derived intimal cells proliferate in situ, or that there is a spontaneous migration from the media into the intima as the animal ages. It remains to be seen whether these cells become irreversibly trapped within the internal elastic lamella. Thomas et al”’ noted an impressive correlation of the distribution of lesions in swine aorta to the cells present in the intima prior to the onset of a high-fat diet. Benditt and Benditt also observed that atherosclerotic lesions begin as singular focal masses, and suggested that these masses were similar to benign tumors of other tissues. They supported this hypothesis with evidence that human atherosclerotic lesions of females appear to arise from single cells.1’3 This was based on the fact that cells in females always have one active and one inactive X-chromosome. Females, therefore, are mosaic for sex-linked heterozygotic traits. Lesions, however, displayed only one or the other

AND

ROSS

sex-linked marker. Since the only other known monoclonal growths are neoplasms, Benditt and Benditt proposed that atherosclerotic lesions begin by neoplastic transformation. PLATELETS

AND PDGF

As already noted, platelets have been shown to contain a potent mitogen, PDGF. This material represents one of the principal mitogens in serum, and accounts for the observation of a lack of growth-promoting activity in plasma.45 Platelet interactions with the artery wall have also been demonstrated to be critical events in the development of lesions of experimentally induced atherosclerosis in a number of model systems including: (1) indwelling catheter in the rabbit,‘14 (2) balloon catheter in rabbit and rat, 33~“5,‘16 (3) hypercholesterolemia with von Willebrand disease,“’ (4) hypercholesterolemia in the nonhuman primate,58*“7 and (5) humans with coronary bypass surgery.“’ The last set of data is particularly relevant because it represents the first direct data for a role of platelets in atherogenesis in our own species. Brown et al”* examined a series of patients with coronary bypass surgery. Quantitative analysis of serial cineangiograms of the bypassed segments of the coronary arteries were made. These studies used a computer program that permits quantitative analysis of computer reconstructions of the bypasses. It has been demonstrated that approximately 35% of all bypasses stenose due to the formation of a new proliferative intimal lesion of atherosclerosis that occurs at the proximal site of the anastomosis with the coronary artery. Brown studied three groups of patients, two of which received antiplatelet therapy (either aspirin alone, or aspirin-dipyridamole) daily, compared with a third group that received a placebo. Quantitative analysis of the double-blind data demonstrated that a statistically significant proportion of the patients receiving antiplatelet therapy had a reduced incidence of new lesion formation. This represents one of the few studies in man clearly implicating a role for platelets, and therefore for PDGF in new lesion formation. An additional series of studies by Smith et al’l9 and Faggiotto and Ross (authors’ unpublished observations) have shown that PDGF can adhere

CELL

365

PROLIFERATION

tenaciously to connective tissue matrix as well as plastic, and retain its activity for cells such as smooth muscle. This suggests that if PDGF is deposited in tissues locally, it might be available to stimulate cells for prolonged periods of time. Such a mechanism might account for a need for repeated injuries, as discussed above. LIFE AND DEATH

OF THE ENDOTHELIAL

CELL

We have already pointed out that endothelial cells, like other living cells, undergo turnover. The significance of that fact for arterial wall function, however, has only begun to be studied. Morphological studies can demonstrate injury, but are difficult to quantify. Most measures of cell injury are at best subjective. Hansson and Schwartz approached this subject by looking for evidence of loss of integrity of the endothelial cell membrane in vivo.‘20*‘2’ Autogenous IgG can be demonstrated in about 1% of the cells in the aorta of the rat by immunoperoxidase or by immunofluorescence techniques. The frequency of IgGpositive cells is an inverse estimate of the viability of the cell population. It is important to emphasize that the IgG is autogenous and is found free in the cytoplasm. This implies that the cell membrane must have broken down some time before perfusion fixation. More than 85% of these cells also contain Ca++ deposits, as indicated by accumulation of the calcium probe, chlortetracycline. Since we cannot easily follow the fate of these cells in vivo, Hansson and Schwartz confirmed the phenomenon in cell culture studies using bovine aortic endothelial cells and bovine serum. Direct time-lapse video microscopic observations show a prolonged process in intracellular movement, gradual fragmentation, and undergrowth by surrounding normal cells.‘20’2’ The functional significance of this phenomenon is not obvious. It is likely that sites of cell death in the endothelium will have altered permeability. This may be important in the accumulation of plasma solutes, particularly lipoproteins, at focal sites. We have no clear evidence that these dying cells can stimulate smooth muscle growth in vivo. In vitro, however, cytoplasm released from lysed endothelial cells is able to stimulate smooth muscle proliferation.‘** The released mitogenic material is tightly adherent to

the culture dish, raising the possibility that similar material released at sites of cell death in vivo might stimulate or facilitate growth of smooth muscle cells.‘22 It is reasonable to wonder what other factors may be released from cells as they die. SMOOTH MUSCLE REPLICATION IN HYPERTENSION

Evidence for endothelial denudation in the early stages of hypertension is conclusive and replication of smooth muscle cells is well documented during this phase of the disease.86~88~‘23 Furthermore, in chronic hypertension, increased smooth muscle cell DNA content correlates with changes in blood pressure. Bevan found that the change in DNA was seen only above the constriction following coarctation of the rabbit aorta. The bulk of the proliferative response in hypertension shown in her studies as well as others appeared to be in the media. The stimulus for this response is not clear. There is, as noted above, increased endothelial replication in the early phases of the response. These changes have not been associated with any evidence for denudation in the aorta.60,92*‘2”‘26Endothelial denudation of small vessels is, however, a prominent feature of acute hypertension.843’27 The increased endothelial replication, moreover, correlates in time with the increase in smooth muscle replication, since both cell types are quiescent in chronically hypertensive animals. One intriguing factor in the proliferation of smooth muscle cells in hypertension is the role of the sympathetic nervous system. There are consistent reports that mass of vascular tissue and, in some reports, DNA content is decreased by sympathectomy of normotensive animals or by treatment of hypertensive animals with drugs that block the sympathetic nervous system.‘28*‘29 It may also be worth noting that ablation of the sympathetic nervous system prevents development of one form of hypertension-the spontaneous hypertension seen in SHR rats. It is possible that this is due in part to a requirement for sympathetic innervation during normal development of the peripheral vascular smooth muscle mass involved in regulation of vascular tone.13’ Regardless of the mechanism involved in the control of smooth muscle cell proliferation, it

366

now seems apparent that the type of proliferation seen in hypertension is quite different from that seen in atherosclerosis. Pathologists have frequently stated that hypertension is associated with proliferation of vessel wall cells in humans. One laboratory has even claimed a highly significant quantitative correlation between morphological assessment of vessel change and level of blood pressure.16 This association is clearly hindered by the difficult morphometric problem of determining cell numbers per unit volume. Cells have complex sizes, shapes, and constitutions. Number per unit area may therefore be a misleading estimate of nuclei per unit volume. Alternative methods exist, including serial sectioning”’ or step sectioning. These methods are generally too laborious for application to complex vascular beds. Somewhat simpler methods were introduced by Olivetti et al.13’ They found no increase in the number of cells in the aortas of hypertensive rats. This presented a paradox, since, as noted above, several investigators had shown an increase in cell replication in the early phase of hypertension, as well as an increase in DNA content in the last phases. Our studies resolved this paradox, at least for the aorta, by showing that cells in the vessel wall can be tetraploid or have even higher levels of ploidy. The frequency of hyperploid cells was greatly increased in hypertensive rats, accounting for the apparent paradox.23*24 The observation that smooth muscle cells can exist in more than one resting growth state implies that there are two forms of replication in the normal vessel wall: replication and endoreplication. These responses differ in the diseases they are associated with, and in the distribution of the replicated cells. Furthermore, the onset of hyperploidy in normal rats and in humans is agerelated.24*‘32 It is interesting to speculate about the similarity of this increase in frequency of tetraploid cells, to the change in replicative life span of smooth muscle cells in culture. Smooth muscle cells from man, like cultured fibroblasts, will only replicate a finite number of times.‘33 The total number of replications in vitro, called the replicative lifespan, declines greatly up to adulthood, with a slower decline at later stages of life. Martin et al”’ have suggested that a decline in the ability of most vessel wall cells to replicate may result in the escape of other cells from

SCHWARTZ

AND

ROSS

normal growth control, resulting in atherosclerotic lesions. SMOOTH

MUSCLE REPLICATION ATHEROSCLEROSIS

IN

Atherosclerosis is a disease of the intima. Smooth muscle cell accumulations do not occur in the media. This is startling, since in the normal wall the intima is relatively hypocellular, whereas the media is densely packed with smooth muscle cells. Thus, regardless of how or why proliferation occurs, an adequate hypothesis must explain the localization of lesions. There is evidence that PDGF can play an important role in this process. Grotendorst et a1’34 found that PDGF, but not other mitogens, was chemotactic for smooth muscle cells. It is worth considering the possibility that this is a critical step in the initial phases of lesion formation. This raises several intriguing possibilities. Once in the intima, cells may not be able to migrate back into the media, but may be less subject to the growth controls found in the densely populated media. Unfortunately, there is relatively little quantitative data on migration. The direct evidence for migration comes from experimental arterial injury. An extraordinary variety of forms of trauma have been used to injure the vessel wall. “2~“6*‘3s~‘36Regardless of the form of injury, the end result is accumulation of smooth muscle cells in the intima. For example, in our studies of endothelial regeneration in the aorta of the rat and monkey, a balloon catheter was passed over the intima, removing the endothelium by abrasion.13’ In both species, the result is smooth muscle cell accumulation in the intima. The normal aortic intima of the adult rat, however, contains few, if any, smooth mucle cells, and none were observed in the immediate period following passage of the catheter. Despite this, lesions were found. Thus it seems likely that an early step in lesion formation was migration of smooth muscle cells from the media. More quantitative studies by Hassler’38 and Webster et a1’39 using tritiated thymidine incorporation and autoradiography indicate that cells in intimal thickenings arrive there by migration from the tunica media, rather than by simple replication of intima1 cells. The same conclusion has been reached on the basis of morphologic evidence following a variety of forms of intimal injury. Evidence for

CELL

367

PROLIFERATION

migration in cholesterol-induced injury comes from studies of the cell kinetics of lesion formation using [3H]TdR and autoradiography.’ These studies demonstrated that lesion formation following cholesterol feeding involves the recruitment of an increased number of intimal and medial smooth muscle cells into the growth fraction. Again, despite this increased mitotic activity, lesion formation occurs only in the intima. Thus any cells produced in the media must either die or migrate into the intima. If those processes of migration and replication are random, one would expect them to originate in a mixed polyclonal cell population. The result should be a polyclonal phenotype. As we have discussed, the evidence in humans seems to say that lesions start with only one or a few cells. The central issue, then, is whether smooth muscle proliferation is a random process. For example, if we suppose that lesions begin with trapping of widely scattered cells in the intima, then each individual lesion would be likely to begin with one, or at the most, a small number of cells. Individual lesions arising in this manner would be expected to be oligoclonal or monoclonal in origin, regardless of the mechanism of proliferation. This interpretation is consistent with published data on experimental atherosclerosis. Thomas et a114’ reasoned that lesions formed by proliferation of single cells would require a high number of generations to make a lesion. When cells are labeled with [3H]TdR before cholesterol feeding, there should be a great dilution of the number of grains per nucleus detectable over intimal cells in the lesions. Thomas et a114’ studied this in swine. The amount of grain-count dilution was consistent with only a small number of replications. They concluded that cholesterolinduced lesions originate by a few divisions of several cells, rather than by a large number of divisions of single cells. In other words, cholesterol lesions in animals appeared to be polyclonal. The apparent inconsistency with human data can be resolved if we posit that monoclonal accumulations exist in the intima prior to the effects of cholesterol. Cholesterol could then act by accelerating the proliferation of cells already present before cholesterol feeding. This sequence seems reasonable since, as stated above, accumulation of cells in the intima of all animals is a normal age-related change, and in humans the

distribution of lesions correlates with the location of diffuse intimal thickening.14’ In summary, the intimal cell accumulation of atherosclerosis may involve two quite separate processes, proliferation and migration. Diffuse intimal thickening may occur by a quite different etiology from focal lesions and may, in turn, provide an initial step before lesions can occur. SUMMARY

We have tried to compare the proliferative responses seen in two vascular diseases: atherosclerosis and hypertension. Both diseases involve endothelial injury and proliferation, but our knowledge of this phenomenon is just beginning to emerge. In atherosclerosis the best evidence is that denudation does not occur in the normal young animal. Man, however, ages over a much longer time than our usual animal models, and the study of denudation during the chronic progression of atherosclerotic lesions remains to be done. We need to consider the possibility that repetitive, small lesions may occur at sites of endothelial turnover. We also need to know more about the possible role of nondenuding injuries, including death of endothelial cells in situ and the apparent increased stickiness of endothelial cells and monocytes during the early stages of hypercholesterolemia. The role of endothelial injury in hypertension also needs more study. We know that extensive denudation and thrombosis occur in small vessels subjected to high blood pressure. It is highly probable that release of PDGF occurs at these sites, possibly accounting for the characteristic hyperplasia seen in malignant hypertension. Whether this process is related to the more subtle changes in vessel wall mass seen in chronic hypertension remains unknown. Finally, there are remarkable differences in the proliferative behavior of the smooth muscle cells themselves in these two diseases. Hypertensive vascular disease is, in large part, a disease of the media. Atherosclerosis is characterized by intimal hyperplasia. Injury results in migration of smooth muscle cells from the media and cell division in the intima. It is possible to identify chemotactic factors using putative atherosclerosis risk factors or normal components of serum. This has already been done for one component of lesion formation, PDGF,‘34 and there is a report

368

of a monocyte chemotactic factor released by smooth muscle cells.‘43 Factors released by other components of lesions may be of considerable interest. In contrast, changes in hypertension occur within a more orderly preservation of vessel wall structure. The wall thickens, but this occurs by increased synthesis of cell mass in the media. The cells themselves do not even divide, but they undergo a form of amitotic replication of their DNA. This results in an increase in the protein synthetic apparatus of the vessel wall without any change in cell number and, presumably, with preservation of mechanical and electrochemical connections between cells. Many open questions exist about this observation. We do not know the

SCHWARTZ

AND

ROSS

distribution of polyploid cells in the microvasculature or the fate of polyploid cells in animals undergoing therapy. These are critical issues if we are to consider the possibility that endoreplication is an etiologic event in increasing resistance to flow. We also need to understand why the same cell undergoes true replication in response to one set of stimuli, that is, trauma or action of the PDGF, but endoreplicates in hypertension. Finally, we should return to our opening paragraph. The similarities and relationships between the two major vascular diseases are striking. The search for understanding of common principles underlying the response of all vessels to injury may lead to valuable new directions.

REFERENCES 1. Virchow R von: Phlogose und Thrombose im Gefssscalcium binding by the red blood cell membrane in spontaSystem. Gesammelte Abhandlungen zur wissenschaftlichen neously hypertensive rats and in essential hypertension. Medizin. Frankfurt, Meidinger, 1956 Pflugers Arch 379:191, 1979 2. Ross R, Glomset J: The pathogenesis of atherosclerosis. 15. Hermsmeyer K, Walton S: Specificity of altered elecN Engl J Med 295:369-420,1976 trogenesis of membrane potential in hypertension. Circ Res 3. Parker F, Odland GF: A correlative histochemical, 40:153, 1977 biochemical and electron microscopic study of experimental 16. Tracy RE: Quantitative measures of the severity of atherosclerosis in the rabbit aorta with special reference to hypertensive nephrosclerosis. Am J Epidemiol91:25, 1970 the myo-intimal cell. Am J Path01 48: 197, 1966 17. Wiener J, Loud AV, Giacomelli F, et al: Morphomet4. Haust MD, Moore RH, Movat HZ: The role of the ric analysis of hypertension-induced hypertrophy of rat thosmooth muscle cells in the fibrogenesis of arteriosclerosis. racic aorta. Am J Pathol 88:619, 1977 Am J Pathol 37:377, 1960 18. Halpern W, Mulvany MJ, Warshaw DW: Mechanical 5. Ross R, Glomset J, Kariya B, et al: A platelet-derived properties of smooth muscle cells in the walls of arterial serum factor that stimulates the proliferation of arterial resistance vessels.J Physiol 275:85, 1978 smooth muscle cells in vitro. Proc Nat1 Acad Sci USA 19. Furuyama M: Histometrical investigations of arteries 7 I : 1207, I974 in reference to arterial hypertension. Tohoku J Exp Med 6. Ross R: Atherosclerosis: A problem of the biology of 76:388, 1962 arterial wall cells and their interactions with blood compo20. Short D: Morphology of the intestinal arterioles in nents. Arteriosclerosis 1:293, 1981 chronic human hypertension. Br Heart J 28:184, 1966 7. Velican C, Velican D: The precursors of coronary 21. Mulvany MJ, Halpern, W: Contractile properties of atherosclerotic plaques in subjects up to 40 years old. Atherosmall arterial resistance vessels in spontaneously hypertensclerosis 37:33, 1980 sive and normotensive rats. Circ Res 41:19, 1977 8. Johnson G: On certain points in the anatomy and 22. Mulvany MJ, Hansen PK, Aalkjaer C: Direct evipathology of Bright’s disease of the kidney. II. On the dence that the greater contractility of resistance vessels in influence of the minute blood-vessels upon the circulation. spontaneously hypertensive rats is associated with a narrowed Trans R Med Chir Sot 51:57, 1868 lumen, a thickened media and an increased number of 9. Ewald CA: Ueber die Verlfnderungen kleiner GeRsse smooth muscle cell layers. Circ Res 43:854, 1978 bei Morbus Brightii und die darauf beziiglichen Theorien. 23. Owens GK, Rabinovitch PS, Schwartz SM: Smooth Virchows Arch 71:453, 1877 muscle cell hypertrophy versus hyperplasia in hypertension IO. Folkow B: Physiologic aspects of primary hyperten(polyploidy/spontaneously hypertensive rat/cell cycle). Proc sion. Physiol Rev 62:347, 1982 Natl Acad Sci USA 78:7759,1982 I I. Bohr DF: What makes the pressure go up? A Hypothesis. Hypertension 3:II-160, 1981 24. Owens GK, Schwartz SM: Alterations in vascular 12. Canessa M, Adragua N, Solomon HS, et al: Increased smooth muscle mass in the spontaneously hypertensive rat: sodium-lithium countertransport in red cells of patients with Role of cellular hypertrophy, hyperploidy and hyperplasia. essential hypertension. N Engl J Med 302:772, 1980 Circ Res 51:280, 1982 13. Woods JW, Falk RJ, Pittman AW, et al: Increased 25. Wolinsky H: Long-term effects of hypertension on the red-cell sodium-lithium countertransport in normotensive rat aortic wall and their relation to concurrent aging changes. Morphological and chemical studies. Circ Res 30:301, 1972 sons of hypertensive parents. N Engl J Med 306:523, 1982 14. Postnov YV, Orlov SN, Pokudin NI: Decrease of 26. Rokitansky C: Organs of the respiration and circula-

CELL

PROLIFERATION

tion, in Day CT, trans. A Manual of Pathological Anatomy, ~014. London, Sydenham Society, 1855 27. Duguid JB: Thrombosis as a factor in the pathogenesis of aortic atherosclerosis. J Pathol Bacterial 6057, 1948 28. Bjarkerud S, Bonjers G: Arterial repair and atherosclerosis after mechanical injury. Atherosclerosis 18:235, 1973 29. Stemerman MB, Spaet TH, Pitlick F, et al: Intimal healing: The pattern of re-cndothelialization and intimal thickening. Am J Pathol87:125, 1977 30. Moore S, Friedman RJ, Singal DP, et al: Inhibition of injury-induced thromboatherosclerotic lesions by antiplatelet serum in rabbits. Thromb Haemost 35:70, 1976 31. Clowes AW, Karnovsky MJ: Failure of certain antiplatelet drugs to affect myointimal thickening following arterial endothelial injury in the rat. Lab Invest 36:452, 1977 32. Mustard JF, Packham MA: The role of blood and platelets in atherosclerosis and the complications of atherosclerosis. Thromb Diath Haemorrh 33:444, 1975 33. Friedman RJ, Stemerman MB, Wenz B, et al: The effect of thrombocytopenia on experimental arteriosclerotic lesion formation in rabbits. J Clin Invest 60:1191, 1977 34. Castellot J, Addonizio M, Rosenberg R, et al: Cultured endothelial cells produce a heparin-like inhibitor of smooth muscle cell growth. J Cell Biol90:372, 1981 35. Ross R, Vogel A: The platelet-derived growth factor. Cell 14:203, 1978 36. Heldin C-H, Westermark B, Wasteson A: Plateletderived growth factor: purification and partial characterization. Proc Nat1 Acad Sci USA 76:3722, 1979 37. Duel TF, Huang JS, Proffitt RT, et al: Human platelet-derived growth factor (PDGF): Purification of PDGF-I and PDGF-II and separation of their reduced subunits. Proc Nat1 Acad Sci USA 78:7314, 1981 38. Antoniades HN, Scher CD, Stiles CD: Purification of human platelet-derived growth factor. Proc Nat1 Acad Sci USA 76:1809,1979 39. Witte L, Kaplan K, Nossel H, et al: Studies of the release from human platelets of growth factor from cultured human arterial smooth muscle cells. Circ Res 28:198, 1980 (abstr) 40. Goldberg ID, Stemerman MB, Handin RI: Vascular permeation of platelet factor IV after endothelial injury. Science 209:6 10, 1980 41. Fuster V, Bowie EJW, Lewis JC, et al: Resistance to atherosclerosis in pigs with von Willebrand’s disease. Spantaneous and high cholesterol diet-induced arteriosclerosis. J Clin Invest 61:722, 1978 42. Schwartz SM, Benditt EP: Aortic endothelial cell replication. I. ElTects of age and hypertension in the rat. Circ Res 41:248, 1977 43. Florentin RA, Nam SC, Lee KT, et al: Increased ‘H-thymidine incorporation into endothelial cells of swine fed cholesterol for three days. Exp Mel Path01 10:250, 1969 44. Schwartz SM, Standaert DM: Effect of chronic hypertension and antihypertensive therapy on endothelial cell replication in the spontaneously hypertensive rat. Lab Invest 47:510, 1982 45. Vogel A, Ross R, Raines E: Role of serum components in density-dependent inhibition of growth of cells in culture.

369

Platelet-derived growth factor is the major serum determinant of saturation density. J Cell BioI85:377, 1980 46. Wright HP: Mitosis patterns in aortic endothelium. Atherosclerosis 15:93, 1972 47. Kunz J, Schreiter B, Schubert B, et al: Experimentelle Untersuchungen iiber die Regeneration der Aortenendothelzellen. Automatische und visuelle Auswertung von Autoradiogrammen. Acta Histochem 61:53, 1978 48. Fry DL: Acute vascular endothelial changes associated with increased blood velocity gradients. Circ Res 22:165, 1968 49. Reidy MA, Schwartz SM: Endothelial regeneration. III. Time course of intimal changes after small defined injury to rat aortic endothelium. Lab Invest 44:301, 1981 50. Reidy MA, Clowes AW, Schwartz SM: Endothelial regeneration V. Inhibition of endothelial regrowth in arteries of rat and rabbit. Lab Invest 49:569-575, 1983 51. Schwartz SM: Selection and characterization of bovine arotic endothelial cells. In Vitro 14:966, 1978 52. Levine EM, Mueller SN: Cultured vascular endothelial cells as a model system for the study of cellular senescence. Int Rev Cytol 10:67, 1979 53. Davies PF, Bowyer DE: Scanning electron microscopy: Arterial endothelial integrity after fixation at physiological pressure. Atherosclerosis 21:463, 1975 54. Clark JM, Glagov S: Luminal surface of distended arteries by scanning electron microscopy: Elimination of contigrational and technical artefacts. Br J Exp Pathol 57:129, 1976 55. Bylock A, Bjiirkerud S, Brattsand R, et al: Endothelial structure in rabbits with moderate hypercholesterolaemia. Acta Path01 Microbial Stand [A] 85:671, 1977 56. Reidy MA, Schwartz SM: Developments in the study of endothelial cells by scanning electron microscopy. Artery 8:236, 1980 57. Gerrity RG: The role of the monocyte in atherogenesis. I. Transition of blood-borne monocytes into foam cells in fatty lesions. Am J Pathol 103:181, 1981 58. Faggiotto A, Ross R, Harker LA: Early arterial changes in the hypercholesterolemic nonhuman primate. Circulation 66:11-225, 1982 59. Schwartz SM, Benditt EP: Cell replication in the aortic endothelium: A new method for study of the problem. Lab Invest 28:699, 1973 60. Schwartz SM, Bend&t EP: Clustering of replicating cells in aortic endothelium. Proc Nat1 Acad Sci USA 73:65 1, 1976 61. Gaynor E: Increased mitotic activity in rabbit endothelium after endotoxin. An autoradiographic study. Lab Invest 24:318, 1971 62. Reidy MA, Bowyer DE: Scanning electron microscopy. Morphology of aortic endothelium following injury by endotoxin and during subsequent repair. Atherosclerosis 261319, 1977 63. Reidy MA, Schwartz SM: Endothelial regeneration IV. The effect of endotoxin on arterial endothelium. Lab Invest 48:25, 1983 64. Hirsch EZ, Robertson AL: Selective acute arterial endothelial injury and repair. I. Methodology and surface characteristics. Atherosclerosis 28:271, 1977 65. Ramzi MM, Walker, LN, Bowyer DE: Marrow superli-

370 cial injury to rabbit aortic endothelium-the healing process as observed by scanning electron microscopy. Atherosclerosis 43:233, 1982 66. Campbell GR, Charnley-Campbell JH: Invited Review: The cellular pathobiology of atherosclerosis. Pathology 13:423, 1981 67. Groves HM, Kinlough-Rathbone RL, Richardson M, et al: Platelet interaction with damaged rabbit aorta. Lab Invest 40: 194, 1979 68. Minick CR, Stemerman MB, Insull W Jr: Role of endothelium and hypercholesterolemia in intimal thickening and lipid accumulation. Am J Pathol95:131, 1979 69. Adelman B, Stemerman MB, Handin RI: Interaction of platelets and fibrin with injured rabbit aortic neointima. Arteriosclerosis 3:141, 1983 70. Bevan RD: An autoradiographic and pathological study of cellular proliferation in rabbit arteries correlated with an increase in arterial pressure. Blood Vessels 13:100, 1976 71. Fernandez D, Crane WAJ: New cell formation in rats with accelerated hypertension due to partial arotic constriction. J Pathol 100:307, 1970 72. Wolinsky H: Response of the rat aortic media to hypertension. Circ Res 16507, 1970 73. Leung DYM, Glagov S, Matthews MB: Cyclic stretching stimulates synthesis of matrix components by arterial smooth muscle cells in vitro. Science 191:475, 1976 74. Esterly JA, Glagov S: Altered permeability of the renal artery of the hypertensive rat: An electron microscopic study. Am J Pathol43:6 19, I963 75. Ooneda G, Sato H, Matsuyama K, et al: Autoradiographic studies on the morphogenesis of arterial fibrinoid degeneration using “‘I labeled plasma protein and “‘I labeled fibrinogen. Gumma J Med Sci 12:26, 1963 76. Giese J: Acute hypertensive vascular disease: 2. Studies on vascular reaction and permeability changes by means of vital microscopy and colloidal tracer technique. Acta Pathol Microbial Stand 62:497, 1964 77. Ooneda G, Ooyama Y, Matsuyama K, et al: Electron microscopic studies on the morphogenesis of fibrinoid degeneration in the mesenteric arteries of hypertensive rats. Angiology l6:8, 1975 78. Olsen F: Penetration of circulating fluorescent proteins into walls of arterioles and venules in rats with intermittent acute angiotensin-hypertension. Acta Pathol Microbial Stand 74:325, 1968 79. Jellinek H, Nagy Z, Hllttner I, et al: Investigations of the permeability changes of the vascular wall in experimental malignant hypertension by means of a colloidal iron preparation. Br J Exp Pathol 50:13, 1969 80. Wiener J, Lattes RG, Meltzer BG, et al: The cellular pathology of experimental hypertension. IV. Evidence for increased vascular permeability. Am J Pathol54:187, 1969 81. Olsen F: Arteriolar permeability and destruction of elastic membrane in hypertension. Acta Pathol Microbial Stand 751527, 1969 82. Htlttner I, More RH, Rona G: Fine structural evidence of specific mechanisms for increased endothelial permeability in experimental hypertension. Am J Pathol61:395, 1970 83. Goldby FS, Beilin LJ: Relationship between arterial

SCHWARTZ

AND

ROSS

pressure and the permeability of arterioles to carbon particles in acute hypertension in the rat. Cardiovasc Res 6:384, 1972 84. Goldby FS, Beilin LJ: How an acute rise in arterial pressure damages arterioles: Electron microscopic changes during angiotensin infusion. Cardiovasc Res 6:569, 1972 85. Wiener J, Giacomelli F: The cellular pathology of experimental hypertension. VII. Structure and permeability of the mesenteric vasculature in angiotensin induced hypertension. Am J Pathol72:221, 1973 86. Giese J: Renin, angiotensin and hypertensive vascular damage: A review. Am J Med 55:3 15, 1973 87. Crane WAJ, Dutta LP: The utilisation of tritiated thymidine for deoxyribonucleic acid synthesis by the lesions of experimental hypertension in rats. J Pathol Bacterial 86:83, 1963 88. Bevan RD, van Marthens E, Bevan J: Hyperplasia of vascular smooth muscle in experimental hypertension in the rabbit. Circ Res 38:58, 1976 89. Potvliege PR, Bourgain RH: Thrombosis induced in vivo in the mesenteric artery of normal and thrombocytopenic rats, an electron microscopic study of the early artery wall reaction. Br J Exp Pathol58:670, 1977 90. Bevan JA, Bevan RD. Chang PC, et al: Analysis of changes in reactivity of rabbit arteries and veins two weeks after induction of hypertension by coarctation of the abdominal aorta. Circ Res 37:183, 1975 91. Reidy MA, Schwartz SM: Mechanism of endothelial repair after injury. Fed Proc 30: I 109, 1980 (abstr) 92. Schwartz SM, Standaert DM: Endothelial cell turnover in rats: Effects of angiotensin, age and chronic hypertension, in Gotto AM Jr, Paoletti R (eds): Atherosclerosis Reviews. New York, Raven Press, 1982, p 109 93. Cherry PD, Furchgott RF, Zawadzky JV: The indirect nature of brandykinin relaxation of isolated arteries: Endothelial dependent and independent components. Fed Proc 40:689, 1981 (abstr) 94. Chand N, Altura BM: Endothelial cells and the relaxation of smooth muscle cells: Possible relevance of lipoxygenases and their significance in vascular disease. Microcirculation 1:297, 1981 95. Davies PF, Kerr C: Co-cultivation of vascular endothelial and smooth muscle cells using microcarrier techniques. Exp Cell Res 141:455, 1982 96. Alhencgelas F, Tsai S, Callahan K, et al: Stimulation of prostaglandin formation by vasoactive mediators in cultured human endothelial cells. Prostaglandins 24:723, 1982 97. Charnley-Campbell J, Campbell GR, Ross R: The smooth muscle cell in culture. Physiol Rev 59:l, 1979 98. Gajdusek CM, DiCorleto P, Ross R, et al: An endothelial cell-derived growth factor. J Cell Biol85:467, 1980 99. DiCorleto PE, Gajdusek CM, Schwartz SM, et al: Biochemical properties of the endothelium-derived growth factor: Comparison to other growth factors. J Cell Physiol ll4:339, 1983 100. Harris-Hooker SA, Gajdusek CM, Wight TN, et al: Neovascular responses induced by cultured aortic endothelial cells. J Cell Physiol 114:302, 1983 101. Ross R, Schwartz SM: Platelet-blood vessel interactions, in Fisher AB, Fishman AP (eds): Handbooks of respiration: Pulmonary circulation and non-respiratory func-

CELL

PROLIFERATION

tions, vol 4. Bethesda, MD, The American Physiological Society, in press 102. Goldstein JL, Ho YK, Basu SK, et al: Binding site on macrophages that mediates uptake and degradation of acetylated low density lipoprotein producing massive cholesterol deposition. Prcc Natl Acad Sci USA 76:333, 1979 103. Liebovich SJ, Ross R: A macrophage-dependent factor that stimulates the proliferation of fibroblasts in vitro. Am J Pathol84501, 1976 104. Glenn KC, Ross R: Human monocyte-derived growth factor(s) for mesenchymal cells: Activation of secretion by endotoxin and conconavalin A. Cell 25603, 1981 105. Martin BM, Gimbrone MA Jr, Unanue ER, et al: Stimulation of nonlymphoid mesenchymal cell proliferation by a macrophage-derived growth factor. J Immunol 126:1510, 1981 106. Still WJ, Dennison S: The arterial endothelium of the hypertensive rat. Arch Pathol97:337, 1974 107. Taylor RF, Price TH, Schwartz SM, et al: Endothelial monolayers on micropore filters: A new assay system for the study of neutrophilendothelial cell interactions. J Clin Invest 67:584, 1981 108. MacGregor RR, Macarak EJ, Kefalides NA: Comparative adherence of granulocytes to endothelial monolayers and nylon fibers. J Clin Invest 61:697, 1978 109. Hoover RL, Briggs RT, Karnovsky MJ: The adhesive interaction between polymorphonuclear leukocytes and endothelial cells in vitro. Cell 14:423, 1978 I IO. Thomas WA, Scott RF, Florentin RA, et al: Population dynamics of arterial cells during atherogenesis. XI. Slowdown in multiplication and death rates of lesion smooth muscle cells in swine during period 105-165 days after balloon endothelial cell desquamation followed by a hyperlipidemic diet. Exp Mol Pathol 35: 153, I98 1 11 I. French JE: Atherosclerosis in relation to the structure and function of the arterial intima, with special reference to the endothelium. Int Rev Exp Path01 5:253, 1966 112. Velican D, Velican C: Accelerated atherosclerosis in subjects with some minor deviations from the common type of distribution of human coronary arteries. Atherosclerosis 40:309, 1981 113. Benditt EP, Benditt JM: Evidence for a monoclonal origin of human atherosclerotic plaques. Proc Nat] Acad Sci USA 70:1753, 1973 114. Moore S: Injury mechanisms in atherogenesis, in Moore S (ed): Vascular injury and atherosclerosis. New York, Marcel Dekker, 198 1, p I3 I 115. Stemcrman MB, Weinstein R, Rowe JW, et al: Vascular smooth muscle cell growth kinetics in vivo in aged rats. Proc Natl Acad Sci USA 79:3863, 1982 116. Friedman SM, Nakashima M, Mar MA: Morphological assessment of vasconstriction and vascular hypertrophy in sustained hypertension in the rat. Microvasc Res 3:416, 1971 117. Harker LA, Ross R, Slichter SJ, et al: Homocystineinduced arteriosclerosis: The role of endothelial cell injury and platelet response in its genesis. J Clin Invest 58:731, 1976 118. Brown BG, Cukingnan RA, Gocde L, et al: Improved graft patency with antiplatelet drugs in patients treated for 1

371

year following coronary bypass surgery. Am J Cardiol 47:494, 1981 (abstr) 119. Smith JC, Singh JP, Lilliguist DS, et al: Growth factors inherent to cell substrate are mitogenically active in situ. Nature 296: 154, 1982 120. Hansson GK, Schwartz SM: Endothelial dysfunction without cell loss, in Cryer A (ed): Biochemical interactions at the endothelium. London, Elsevier/North Holland Biomedical Press, 1983, p 343 121. Hansson GK, Schwartz SM: Evidence for cell death in the vascular endothelium in vivo and in vitro. Am J Path01 1983, in press 122. Gajdusek CM, Schwartz SM: Comparison of intracellular and extracellular endothelial-derived mitogens. J Cell Biol 199, 1982 (abstr) 123. Giacomelli F, Wiener J: Regional variation in the permeability of rat thoracic aorta. Am J Pathol 75:513, 1974 124. Majack R, Bhalla RC: Ultrastructural characteristics of endothelial permeability pathways in chronic hypertension. Hypertension 3:586, 1981 125. Htlttner I, Costabella PM, De Chastonay C, et al: Volume, surface, and junctions of rat aortic endothelium during experimental hypertension. A morphometric and freeze fracture study. Lab Invest 46:489, 1982 126. Limas C, Westrum B, Limas CJ: The evolution of vascular changes in the spontaneously hypertensive rat. Am J Path01 98:357, 1980 127. Reidy MA, Schwartz, SM: A technique to investigate surface morphology and endothelial cell replication of small arteries. A study in hypertensive rats. Microvasc Res 24:158, 1982 128. Rorive GL, Carlier PJ, Foidart JM: Hyperplasia of rat arteries smooth muscle cells associated with development and reversal of renal hypertension. Clin Sci 59:335, 1980 129. Sen S, Tarazi RC, Bumpus FM: Cardiac hypertrophy and antihypertensive therapy. Cardiovasc Res 11:427, 1977 130. Lais LT, Brody MJ: Vasoconstrictor hyper-responsiveness: An early pathogenic mechanism in the spontaneously hypertensive rat. Eur J Pharmaco147:177, 1978 13 1. Olivetti G, Anversa P, Melissari M, et al: Morphometry of medial hypertrophy in the rat thoracic aorta. Lab Invest 42:559, 1980 132. Barrett TB, Sampson P, Owens GK, et al: Polyploid nuclei in human artery wall smooth muscle cells. Proc Nat] Acad Sci USA 80:882, 1982 133. Bierman EL: The effect of donor age on the in vitro life span of cultured human arterial smooth muscle cells. In Vitro 14:951, 1978 134. Grotendorst GR, Seppa H, Kleinman HK, et al: Attachment of smooth muscle cells to collagen and their migration toward platelet derived growth factor. Proc Nat1 Acad Sci USA 78:3669, 1981 135. Hoff HF, Heideman CL, Gotto AM, et al: Apolipoprotein B retention in the grossly normal and atherosclerotic human aorta. Circ Res 41:684, 1977 136. Hoff HF, Heideman CL, Gaubatz JW, et al: Correlation of apolipoprotein B retention with the structure of atherosclerotic plaques from human aortas. Lab Invest 38:560, 1978

372 137. Schwartz SM, Stemerman MB, Benditt EP: The aortic intima. II. Repair of the aortic lining after mechanical denudation. Am J Path01 81:15, 1975 138. Hassler 0: The origin of the cells constituting arterial intimal thickening. Lab Invest 22:286, 1970 139. Webster WS, Bishop SP, Geer JC: Experimental aortic intimal thickening. I. Morphology and source of intimal cells. Am J Path01 76:245, 1974 140. Thomas WA, Reiner JM, Florentin RA, et al: Population dynamics of arterial cells in atherogenesis, in Moore S (ed): Vascular injury and atherosclerosis, in the series The

SCHWARR

AND

ROSS

Biochemistry of Disease. New York, Marcel Dekker, 1981, p 111 141. Thomas WA, Florentin RA, Reiner, JM, et al: Alterations in population dynamics of arterial smooth muscle cells during atherogenesis. IV. Evidence. for polyclonal origin in hypercholesterolemic diet-induced atherosclerotic lesions in young swine. Exp Mol Path01 24:244, 1976 142. Massman J, Oestreich S: Comparative histology and morphometrical studies into the relevance of intimal thickening to coronary sclerosis. Atherosclerosis 24:451, 1976 143. Sprague EA, Kelley JL, Schwartz CJ: Growth structure and function of baboon aortic smooth muscle cells in culture. Exp Mol Path01 37:48, 1982