Platelets and foam cells in the evolution of atherosclerosis

Arherosclerosis, 61 (1986) 107-115 Elsevier Scientific Publishers Ireland,

107 Ltd

ATH 03795

Platelets

and Foam Cells in the Evolution

Histological

and Immunohistological

of Atherosclerosis

Studies of Human Lesions

Simon Sevitt * Departmeni of Investigative Pathology, University of Birmmgham, Birmmgham (U.K.) (Received 23 January, 1986) (Revised, received 10 March, 1986) Accepted 15 March, 1986)

Summary Sixty-seven coronary, aortic and other lesions (39 subjects) ranging from minor to advanced disease were studied histologically and by immunohistology for antigens of platelets, P-lipoprotein (LpB) and fibrinogen. Collections of lipid-filled foam cells were found in nearly every lesion mainly in the thickened intima, sometimes associated with macrophages. Foam cells were also present elsewhere including, in advanced lesions, necrotic zones filled with cholesterol needles. Then they too were necrotic. Platelets were observed in many lesions as small free clusters and/or phagocytosed in foam cells and sometimes also in macrophages. Cellular and extracellular reactions for platelet derivatives and for LpB were very frequent. Reactions for fibrinogen were only extracellular. Foam cell cytoplasm reacted for platelet derivatives and for LpB whilst macrophages, endothelium and certain other cells sometimes reacted for platelet derivatives but rarely for LpB. The studies indicate that: (1) Foam cells originate from macrophages mainly in the intima and many migrate elsewhere. (2) Their formation and their uptake of LpB requires cellular acquisition of a platelet factor or factors. (3) Acquisition of platelet factor(s) is by phagocytosis of intact platelets and/or uptake of extracellular soluble derivatives from disrupted platelets. (4) These processes begin early, are likely to be continuous and are significant for the evolution of atherosclerosis including the necrotic foci of advanced lesions.

Key words:

Atherosclerosis - Lesion evolution - P-Lipoprotein Platelet derivatives

- Foam cells - Macrophages

~ Platelets

-

Introduction * Address for correspondence and reprints: Moseley, Birmingham B13 9XA, U.K. 0021-9150/86/$03.50

159 Green

0 1986 Elsevier Scientific

Road,

Publishers

Ireland,

In a study on atherogenesis arising from the organisation and reduction of pulmonary thromLtd

108

boemboli, Sevitt and Walton [l] demonstrated antigenic derivatives of platelets in the foam cells newly formed in the thickened intima, sometimes in other cells and in extracellular matrix. As these findings seemed potentially significant for atherogenesis, a pilot study on known lesions of atherosclerosis was done, and it also revealed platelet derivatives in their foam cells and sometimes elsewhere. The present investigation has confirmed and extended these findings which point to important roles by platelet derivatives and foam cells in the life history of atherosclerotic lesions. Cases, Materials and Methods Sixty-seven arterial lesions from 39 subject,s were studied, mostly injured or burned subjects who reached necropsy in the Birmingham Accident Hospital, but supplemented by selected specimens from current necropsies for H.M. Coroner. Most patients were elderly and the sexes about equally represented. Most had succumbed to various complications but some died of natural causes. The severity of lesions ranged from minor fatty aortic streaks (1 case, 3 specimens) and uncomplicated plaques in pulmonary arteries (1 case, 8 specimens) to advanced and/or complicated lesions in coronary arteries (26 subjects, 40 specimens), the aorta (5 cases, 8 specimens) carotid (2 cases, 2 specimens), cerebral (1 case, 3 specimens), iliac, popliteal and subclavian arteries (each 1 case, 1 specimen). Different stages including a few early lesions were found in some coronary vessels sectioned at different revels. Advanced lesions showed considerable concentric or eccentric fibrous thickening, deep necrotic zones, often narrowed lumens and some also had calcification. Thrombi were found in 12 coronary lesions and injury or burns, which can precipitate arterial including coronary thrombosis [2,3], may have been responsible for some of them. Coronary disease, including thrombi in 6 cases, was judged at least contributory to death in 12 subjects. Otherwise the lesions were incidental findings. Histology Cross-sectional samples of vessels were excised except from the aorta from which narrow strips

were cut. Two to 4 pieces up to 1 cm apart were taken from 22 coronary arteries and they were processed separately or in one block. Most coronary samples were removed after fixation of the heart for about 2 days in formol-saline, though many arteries had been opened for inspection before fixation. Mineralisation of some vessels necessitated acid-decalcification prior to sampling. Specimens were fixed in buffered formol-saline and processed to paraffin wax for embedding into blocks. Duplicates of some were also embedded in gelatine. Lesions from 3 current necropsies, that is the cases with aortic fatty streaks, primary disseminated pulmonary lesions and cerebral artery disease, were also examined as unfixed cryostat specimens from tissue snap-frozen in liquid nitrogen. Adjacent samples were fixed in cold alcohol and blocked in paraffin wax or polyethylene glycol. Paraffin sections were stained by standard methods to help identify fibrin, platelets, proteoelastica, smooth muscle and haemoglycans, siderin. Appropriate fixed and unfixed sections were stained for lipid using Oil red 0 and a counterstain. Immunohistology The antisera and methods of their use were as described previously [ 11. An immunoperoxidase (Imp) technique was employed on paraffin sections and in several cases also an immunofluorescent method on unfixed frozen sections. Both were indirect two-stage procedures. Appropriate negative and positive controls were included in each batch of tests. For negative controls, normal serum was substituted for the specific first-stage reagent. All antisera were polyclonal. Antisera To detect platelet antigens, 3 antibodies were employed, one routinely. This was a sheep antiserum against human P-thromboglobulin (/3-TG) (Scottish Antibody Production Unit). P-TG is a release product of platelet granules specific to platelets [4] and probably related to anti-heparin platelet factor 4 [5]. Its antiserum reacted with platelets but not fibrin, red cells or leucocytes in blood smears and sections of thrombi. The others,

109

a sheep anti-PF4 (SAPU) and a rabbit antiserum against intact human platelets prepared in Birmingham University Surgical Research Unit, were used to confirm and compare results obtained with the stronger reacting anti+TG serum. A section of a thrombus or embolus containing many platelets was the positive control. To detect the antigen of /3-lipoprotein (LpB) a rabbit antiserum against human low density lipoprotein isolated from freshly drawn blood was used [6]. This antibody is directed mainly against the protein antigen apolipoprotein B (apo B) and therefore reacts with it even in paraffin sections from which the lipid moiety had been extracted. A paraffin section of an aortic lesion known to react with this antiserum was the positive control. To detect fibrinogen, ImP tests utilized a rabbit antiserum to human fibrinogen (Dako). It reacted with fibrinogen and fibrin but not with platelets in blood smears and sections of thrombi. Results Platelets in lesions (Table 1) Platelets were observed histologically and/or by immunohistology in nearly two thirds of the lesions as small free clusters and/or phagocytosed in foam cells and sometimes also in macrophages. Most free collections were in thickened intima but some were elsewhere, including necrotic zones and

TABLE

their surrounding connective tissue. Some lay near small haemorrhages or capillaries. Phagocytosed platelets presented as apparently intact coarse granules larger than the pseudo-granules of foam cells and they reacted with anti-platelet antisera (Fig. 1). In some lesions they also reacted with anti-LpB antiserum. The latter had previously been found in platelets within foam cells related to atherogenesis from the reduction of pulmonary thrombo-emboli when platelet binding of cellular LpB was postulated [l]. The binding in vitro of plasma lipoproteins to human platelets [7] supports this view. Foam cells in lesions Special attention was paid to foam cells which were seen in the great majority of lesions including 35 of the 40 in coronary arteries and all those in other vessels. Limitations of histological sampling could account for their apparent absence in a few. Typically they presented as large plump or rounded cells (Fig. 2) but a few were relatively small and others oval or elongated. Frozen sections demonstrated numerous lipid drops of various sizes in their cytoplasm and also birefringent needle-shaped and plate-like crystals under polarized light, indicative of cholesterol and its esters. Appearances in paraffin sections depend on the sizes of the lipid spaces and on their surround-

1

PLATELETS

IN LESIONS

Numerals in brackets tosed in macrophages. Sites of lesions

OF ATHEROSCLEROSIS

refer to lesions

No. of lesions studied

with platelets

phagocy-

Lesions with platelets As free collections

Phagocytosed in foam cells

Totals

Coronary arteries

38

17

I4 (7)

22

Aorta Pulmonary arteries

4 1

2 1

2 I (I)

2 1

Other arteries

5

5

3

5

48

25

Totals

’ See text.

a 20 (8)

30

Fig. 1. let-like platelet antigen artery (P-TG),

Group of foam cells containing relatively coarse plategranules (arrow heads), reactive (dark stained) for the antigen. The cytoplasm also reacts for the platelet mainly in a pseudogranular pattern. From a coronary lesion. Immunoperoxidase (IMP)//3-thromboglobulin x 480.

Figs. 2-4. From a lesion of primary pulmonary atheroma with innumerable yellow plaques restricted to the intima. Fig. 2. Sheet of foam cells in the thickened intima with a histological pseudogranular cytoplasmic pattern. Picromallory, x 480. (Frozen sections revealed many lipid droplets and birefringent crystals of cholesterol and its esters in the cells). Fig. 3. Sheet of foam cells similar to those in Fig. 2. Their cytoplasm reacts strongly in a pseudo-granular pattern (dark stained) x 480. for the platelet antigen. IMP@TG, Fig. 4. Lipid droplets of foam cells in an intimal plaque. The bright fluorescent rims (arrow heads) around the droplets represent the specific reactivity of foam cell cytoplasm for the platelet antigen. The dull fluorescence of the droplets is non-specific. Unfixed frozen section. Immunofluorescence under ultra-violet light. P-TG, x 1500 approx.

ing stainable cytoplasm. When all the droplet spaces are very small, a finely stippled or pseudogranular pattern is found (Fig. 2) but when they are relatively large, the pattern is vesicular (Fig. 6 and 7). Most cells and collections showed one or other pattern, but the picture was mixed in some. In early lesions and uncomplicated plaques, foam cells were restricted to the intimal thickening and formed cell groups or sheets (Fig. 2) or groups of discrete cells (Fig. 5) many of which were near the lumen. Intimal location points to the bloodstream as the source of their ultimate precursors (monocytes) and of their lipid. In ad-

vanced lesions, foam cells were found near the lumen but also elsewhere including necrotic zones and their surrounding connective tissue. Most necrotic foci contained abnormal foam cells often lying between cholesterol crystals (Figs. 6 and 7). Many were large or very large necrotic cells with distended vesicles and without nuclei, and some were also fragmented. Unlike viable foam cells they stained avidly with basiphilic dyes like haematoxylin (Fig. 6) and toluidine blue, but like viable cells they reacted with anti-platelet antibodies (Fig. 7) and for LpB. Foam cells are migratory [8] and this explains

111

Immunohistology

Fig. 5. Intimal foam cells in an aortic stained) for LpB in a pseudo-granular x 480.

lesion reactive (dark pattern. IMP/LpB,

their presence in deeper parts of advanced and complicated lesions, including necrotic zones where all or many die. The frequent association of dead foam cells with free cholesterol crystals in these zones supports the view [9] that crystallization of cholesterol freed from the cells initiates the necrosis.

Foam cell reactions (Table 2) Fibrinogen was not detected in foam cells but their reactions for platelet and for LpB antigens were very frequent. The patterns of specific reactivity were similar to the pseudo-granular and vesicular patterns of histological staining (compare Figs. 2, 3 and 5 and Figs. 6 and 7). However, in the aortic lesion with only minor fatty streaks the specific reactions for LpB and platelet antigens affected the cell membranes and only irregular parts of the cytoplasm reacted. This may represent an early change heralding the common pseudo-granular and vesicular patterns, though further studies on fatty streaks are necessary to substantiate this. The reactions for P-TG in paraffin sections given by the immunoperoxidase method were confirmed by the immunofluorescent technique on unfixed frozen sections. Then the specific reactivity for P-TG was found to be confined to the narrow rims of cytoplasm surrounding each of the innumerable lipid droplets (Fig. 4). The high incidence of platelet derivatives in foam cells found in primary lesions of atherosclerosis is similar to that in thickenings secondary to pulmonary thromboemboli [l]. However, LpB in

Figs. 6 and 7. From necrotic zones in advanced coronary lesions. Fig. 6. Necrotic foam cells with strongly basophilic and vesiculated clefts. Some cells are fragmented. Haematoxylin and eosin, x480. Fig. 7. Several necrotic foam cells reactive for the platelet antigen IMP,+-TG, x 480.

cytoplasm within

in necrotic

necrotic

tissue containing

tissue containing

cholesterol

cholesterol

needle

needle clefts.

112

TABLE

2

FREQUENCIES OF REACTIONS ATHEROSCLEROSIS Lesions

FOR (a) PLATELET

ANTIGENS

AND

No. with reactions

for (a) platelet

antigens

Sites

No. tested

Foam cells

Coronary arteries

(a) 36 (b) 23

33 19

Aorta (fatty streaks)

(a) (b)

2 2

Other aortic lesions

(a) (b)

Pulmonary arteries Other arteries Totals

a

Macrophages .

(b) LpB BY VARIOUS

CELLS

IN LESIONS

OF

or (b) LpB by

Endothelium

Fibrocytes

Smooth muscle cells

15 1

4 0

3 1

2 0

2 2

0 0

0 0

0 0

2 2

4 4

4 4

2 0

1 0

0 0

2 1

(a) (b)

3 1

3 1

3 0

2 0

0 0

0 0

(a) (b)

5 2

5 2

1 0

3 1

0 0

3 0

(a) 50 (b) 32

41 28

21 1

10 1

3 1

9 3

a See text CASES, MATERIALS AND METHODS.

foam cells was more common in primary than in secondary lesions, possibly because cellular enzymes had destroyed the apo-B antigen in some of the latter [l]. Nevertheless, the results point to a functional connection between the acquisitions of LpB and platelet derivatives in foam cells.

Reactions by other cells (Table 2) Platelet antigens were detected in macrophages in about 40% of the lesions, in intimal endothelium and in smooth muscle cells of the intima or media in about 20%, and occasionally in fibrocytes. In contrast LpB was rarely or only occa-

Fig. 8. From the thickened intima in a coronary lesion. Collection of macrophages reactive (dark stained) for the platelet antigen, some diffusely, other in a granular manner. Occasional reactive foam cell also present. IMP/P-TG, X 480.

Fig. 9. From the thickened intima in a popliteal artery lesion. The endothelium of the lumen (above) and many elongated connective tissue cells in the intima react for the platelet antigen. IMP//S-Tg, X 480.

113

Fig. 10. From the thickened intima in an advanced coronary lesion showing extracellular reactivity (dark stained) of the connective tissue for the platelet antigen. Reactive macrophages and a reactive foam cell (right) present. The endothelium lining the lumen (above) is also reactive. IMP/fi-TG. x 480.

sionally observed in these cell types. Neither lymphocytes nor polymorphonuclear leucocytes reacted. Platelet antigen reactions were also given by fibro-endothelial cells lining and invading organised thrombi in two coronary lesions. The yellow-brown pigment ceroid was seen in some cells resembling macrophages and its colouration resembled that given by positive ImP reactions. It was distinguished by the identical pigmented appearances in Imp-negative control preparations and in stained and unstained sections. Cytoplasmic reactivity for LpB by cells other than foam cells was always diffuse. That for platelet derivatives was diffuse or granular or both in macrophages (Fig. 8) and fibrocytes (Fig. 9) and was diffuse in endothelial cells (Figs. 9 and 10). The diffuse pattern could be explained by endocytosis of the soluble contents discharged from disrupted platelets and the granular pattern by phagocytosis of initially intact platelets. Extracellular

reactions

Fibrinogen was detected in 5 of the 6 tested, including 2 early ones, and was entirely restricted to the thickened intima. of fibrin were often seen in advanced Clotting of insudated fibrinogen could them.

lesions almost Streaks lesions. explain

Extracellular LpB was found in 30 of the 32 lesions tested. Theintima was always affected but often other parts, including necrotic cholesterolbearing zones, were also involved. The regular intimal location of LpB and fibrinogen indicates that both stem from the insudated plasma. Extracellular derivatives of platelets were also demonstrated in the great majority (46 out of 50) of the lesions tested, including early ones. Again, the thickened intima or parts thereof were usually affected (Fig. 10) but other sites, including viable connective tissue matrix and non-viable cholesterol-bearing zones, were often involved. Extracellular distributions of LpB and of platelet antigens in individual lesions were rarely, if ever, identical. No special relationship between the distributions of platelet antigens and those of intimal smooth muscle was observed. Reactivity for LpB and platelet derivatives was also frequent in the adventitia of the vessels, features probably related to their lymphatic drainage. Discussion The presence of free platelets in lesions of atherosclerosis, their phagocytosis by foam cells and macrophages and uptake of their derivatives by these and other cells in the lesions do not seem to have been reported hitherto except in atherogenie lesions secondary to the organisation of pulmonary thromboemboli [l]. It was those observations which prompted the present study the results of which support and extend other evidence [lO,ll] and experimental studies [12-141 that platelets are important in atherogenesis. The frequent presence of extracellular fibrinogen and LpB reported by others [15-171 and probably derived from insudated plasma, has been confirmed. Extracellular platelet antigens had also been demonstrated previously mainly in advanced lesions where they were thought to be remnants of buried thrombi [18-201. However, in this study impregnation of connective tissue matrix by soluble products released from disrupted platelets is a more valid explanation. Potential sources of the free platelets likely to provide their products for uptake are lesion capillaries and the arterial blood, the latter through fissures involving the lumen or following endothelial desquamation. Release of

114

platelet contents might be explained by local formation of thrombin through tissue activation of plasma prothrombin, and by the ability of the contents of the lesions to accelerate blood coagulation [21].

Foam cells and the role of platelets Study of these cells has become important in order to decide whether they are innocent bystanders or important contributors to the lesions. The balance of evidence [22,23] supported by the present findings points to macrophages originating as monocytes infiltrating from the blood as their precursors. The macrophages take up insudated lipid, especially LpB, and transform into foam cells. This explains why the intima is their major location and site of formation even in advanced lesions with much fibrous thickening. Uptake of LpB by macrophages and foam cell formation might be regarded teleologically as a local defence mechanism against a mounting extracellular concentration of LpB, but if this is so, its power is limited. A functional connection between LpB and platelet derivatives in foam cell formation from macrophages has been demonstrated in vitro by Fogelman et al. [24]: uptake of native LpB by macrophages and their accumulation of cholesterol occurred when platelets were added to the suspension and underwent their release reaction, but not without their addition. It was concluded that one or more platelet factors had suitably altered native LpB to permit its uptake. That platelet derivatives are involved in cellular accumulation of LpB is supported by the present study, but their immunohistological confinement to the cytoplasmic rims of lipid droplets in foam cells implicates cell metabolism. Malondialdehyde, a product of the metabolism of platelet arachidonic acid, also permits in vitro uptake of native LpB by macrophages [24], but among the many products of platelets, others including platelet growth factor [25,26] and fi-thromboglobulin are also candidates. Platelets contain appreciable amounts of /3-TG which is also specific to them [4], and in support of its candidature are the strong immunohistological cellular reactions obtained with its antiserum, stronger than those obtained with the two other antisera used (see METHODS). On

the other hand, the reactivity of foam cells to all three anti-platelet sera employed might only indicate the simultaneous presence in the cells of malondialdehyde or other specifically active platelet product. Consequently one, or more than one, platelet factor might be implicated in the transformation of macrophages to foam cells. Further studies of lesions with antisera specific to malondialdehyde, platelet growth factor and other platelet products, if these become available, are indicated. Foam cells are not restricted to atherosclerosis. For example, similar cells were found in revascularizing necrotic bone marrow where they scavenge the lipid of dead fat cells [27]. It would be of interest to ascertain whether or not platelet derivatives are involved in that process. Migration of foam cells within lesions explains their presence in non-intimal locations including necrotic zones of advanced disease. Here they become enlarged, vesicular, liberate cholesterol crystals and often die. Thus, critical roles are played by foam cells in the evolution of atherosclerosis and the present findings substantiate and expand some of the views put forward by others [9,22,28]. In brief, the life history of foam cells is likely to be a key to lesion development. Transformed from macrophages with the aid of one or more platelet derivatives, foam cells form early on and continue to be formed as the lesions advance and become complicated. Many migrate from their origins to other locations, including sites where their liberation of cholesterol crystals initiates local tissue death. Continuation and repetition of these events helps to advance the lesions, though other processes like fibrous thickening, smooth muscle proliferation and invasion of blood vessels are also involved. Acknowledgements Thanks are due to Professor Kenneth Walton and to Dr. Rod Jones for their generous provision of Laboratory facilities, and to the former for helpful discussions. I am grateful to Mr. R. Hawker for providing one of the platelet antisera; to Mrs. Jean Fitzmaurice and Mr. Alan Wainwright for expert technical assistance and to Mrs. Valerie Stanley for secretarial help.

115

References 1 Sevitt, S. and Walton K.W., Atherosclerotic lesions arising from the reduction of pulmonary emboli, Atherosclerosis, 59 (1986) 173. 2 Sevitt. S., Coronary thrombosis following injury and burns, Medicine, Science and the Law, 13 (1973) 185. 3 Sevitt, S., Reactions to Injury and Burns, Heinemann, London, 1974, Ch. 12. 4 Moore, S.. Pepper D.S. and Cash, J.D., The isolation and characterization of platelet-specific globulin (/?-thromboglobulin) and the detection of anti-urokinase and antiplasmin release from thrombin-aggregated platelets, Biochim. Biophys. Acta, 371 (1975) 60. 5 Niewarowski, S., Proteins secreted by the platelet. Thromb. Haemost., 38 (1977) 924. 6 Walton, K.W. and Darke, S.J.. Immunological characteristics of the human low density lipoproteins, Immunochemistry, 1 (1964) 267. 7 Curtiss, L.K. and Plow, E.F., Interaction of plasma lipoproteins with human platelets, Blood, 64 (1984) 365. 8 Gerrity, R.G. and Goss, J.A. Chemotactic responses by monocyte-macrophages - A possible mechanism for their migration into atherosclerotic lesions. In: Atherosclerosis VI. Springer-Verlag, Berlin, 1983, pp. 455-461. 9 Small, D.M., Cellular mechanism for lipid deposition in atherosclerosis, N. Engl. J. Med., 297 (1977) 873, 924. 10 Mustard. J.F., Platelets, vessel wall and thrombosis. In: Haemostasis and Thrombosis (Proc. Serono Symposium, Vol. 15). Academic Press, New York, 1979, p. 77. 11 Murphy, E.A. and Mustard, J.F., Coagulation tests and platelet economy in atherosclerotic and control subjects. Circulation, 25 (1962) 114. 12 Moore, S., Friedman, R.J., Singal, D.P., Gauldie, J., Blackman, M.A. and Roberts, R.S., Inhibition of injury-induced thrombo-atherosclerotic lesions by anti-platelet serum in rabbits, Thromb. Haemost., 35 (1976) 70. 13 Friedman, R.J. Stemerman, M.B., Spaet, T.H., Moore, S. and Gauldie, J., The effect of thrombocytopenia on atherosclerotic plaque formation, Fed. Proc., 35 (1976) 207. 14 Ross, R. and Glomset, J., Response to injury and atherogenesis, Amer. J. Path., 86 (1977) 675. 15 Walton, K.W., Pathogenetic mechanisms in atherosclerosis, Amer. J. Cardiol., 35 (1975) 542. 16 Woolf. N., The pathogenesis of atherosclerosis, Recent Adv. Histopath., 10 (1978) 45.

17 Walton, K.W. and Williamson, N., Histological and immunofluorescent studies on the evolution of the human atherosclerotic plaque, J. Atheroscler. Res., 8 (1968) 599. 18 Carstairs, K.C., The identification of platelets and platelet antigens in histological sections, J. Path. Bact.. 90 (1965) 225. 19 Woolf, N. and Carstairs, K.C., Infiltration and thrombosis in atherogenesis - A study using immunofluorescent techniques, Amer. J. Path., 51 (1967) 373. 20 Hudson, J. and McCaughey, W.T.E., Mural thrombosis and atherogenesis in coronary arteries and aorta, Atherosclerosis. 19 (1974) 543. 21 Connor, W.E., Hoak, J.C. and Warner, E.P., The role of lipids in thrombosis. In: Koeler, Duckert and Streuli (Eds.), Pathogenesis and Treatment of Thrombo-Embohc Diseases, Schattauer-Verlag. Stuttgart, 1966, pp. 193-208. 22 Fowler, SD., Role of macrophage foam cells in arterial lesions. In: Atherosclerosis VI, Springer-Verlag, Berlin, 1983, pp. 452-456. 23 Schaffner, T., Taylor, K., Bartucci, E.J., Fisher-Dzoga, K., Beeson, J.H., Glagov, S. and Wissler, R.W., Arterial foam ceils with distinctive immunomorphological and histochemical features of macrophages, Amer. J. Path., 100 (1980) 57. 24 Fogelman, A.M., Shechter, I., Seager, J., Hokom, M., Child, J.S. and Edwards, P.A., Malondialdehyde alteration of low density lipoprotein leads to cholesterol ester accumulation in human monocyte-macrophages, Proc. Nat. Acad. Sci. (USA), 77 (4) (1980) 2214. 25 Day, A.J., Preet Singh, S., Hudson, K. and Majumder, S., Interaction of platelets, macrophages and lipoproteins in relationship to atherogenesis. In: Atherosclerosis VI, Springer-Verlag. Berlin, 1983, pp. 4488451. 26 Witte, L.D., Cornicelli, A., Fairbanks, K.P. and De Witt, S.G., Effects of PDGF and endothehal cell products on cellular lipid and lipoprotein metabolism in human fibroblasts. In: Atherosclerosis VI, Springer-Verlag. Berlin. 1983, pp. 416-419. 27 Sevitt, S., Bone Repair and Fracture Healing in Man. Churchill-Livingstone, Edinburgh, 1981. Ch. 8. 28 Stary. H.C., Macrophages in coronary artery and aortic intima in atherosclerotic lesions of children and young adults. In: Atherosclerosis VI, Springer-Verlag, Berlin, 1983, pp. 4622466.