- Email: [email protected]
Hypertension-induced cerebral atherosclerosis in the cholesterol-fed rabbit
137
Atherosclerosis, 30 (1978) 137-145 @ Elsevier/North-Holland Scientific Publishers, Ltd.
HYPERTENSION-INDUCED CEREBRAL ATHEROSCLEROSIS IN THE CHOLESTEROL-FED RABBIT
TAKESHI KUROZUMI, KENZO TANAKA and YOSHIAKI YAE Department of Pathology and Central Laboratory, Fukuoka 812 (Japan)
Faculty of Medicine, Kyushu University,
(Received 6 December, 1977) (Revised, received 30 December, 1977) (Accepted 30 December, 1977)
Foam cell lesions were found in cholesterol-fed rabbits with induced hypertension, particularly in intimal cushions at branching sites, where permeability to horseradish peroxidase was enhanced. Permeability to horseradish peroxidase was enhanced at the edge of intimal cushions without foam cell accumulation. This finding suggests that permeability is increased before foam cell infiltration. No foam cell lesions were observed in the intima of cerebral arteries distant from branching sites, but insudation of plasma constituents here caused endothelial cells to separate from the subendothelial matrices. Foam cell lesions were absent from the cerebral arteries in normotensive cholesterol-fed rabbits. Key words:
Atherosclerosis - Cerebral arteries - Cholesterol - Endothelium sion - Intimal cushion -Permeability - Feroxidase
- Hyperten-
Introduction
Previously we found that the permeability pattern of cerebral arteries differed from that of the coronary arteries and aorta, in relation to the ultrastructural characteristics of the intima of these arteries [l-3]. The cerebral artery was resistant to cholesterol-induced atherosclerosis in rats and rabbits, presumably because of a barrier function by the intima. Hypertension is known to promote the evolution of atherosclerosis in the experimental animal [4-71. Autopsy studies indicate that hypertension accelerates cerebral atherosclerosis [ 8,9]. In this paper, the effect was studied of renovascular hypertension on structure and permeability of cerebral arteries in hypercholesteremic rabbits.
138
Materials and Methods Three groups of New Zealand albino male rabbits weighing 2.0-2.5 kg were used. The first group of 5 rabbits was maintained on stock diet (Oriental Co., Japan) for 12 weeks. The second group of 10 rabbits was fed the same diet with added cholesterol for 12 weeks. Sufficient cholesterol was rubbed into the diet to provide about 0.5 g cholesterol per animal per day. The left renal artery in the animals of these two groups was constricted with a silver clip on the second week of the experiment, followed by constriction of the right renal artery 2 weeks later. The third group of 4 rabbits was simply fed the diet containing 0.5 g cholesterol per day for 16 weeks. Two rabbits in the first group and 3 in the second died in the course of the experiment. Two rabbits in each of the first and third groups and 4 rabbits in the second group were injected with horseradish peroxidase (Sigma Chemical Co., Type II) 250 mg/kg body weight dissolved in 30 ml of isotonic saline, 15 min before killing. Tissue blocks were taken from the aorta and the medium- and small-sized intracranial arteries, that is, basilar, middle and anterior cerebral arteries and the circle of Willis including their branches. Tissue blocks were fixed with 3% glutaraldehyde buffered with 0.1 M cacodylate for 5-6 h at room temperature and then washed overnight at 4°C in 0.1 M cacodylate buffer. After post-fixation with 1% 0~0,. buffered with 0.1 M cacodylate for 1.5 h, they were then dehydrated in graded ethanol, treated with propylene oxide, and embedded in Epon 812. Tissue blocks from rabbits injected with horseradish peroxidase were incubated for 15 min at room temperature in 10 ml of 0.05 M Tris-HCl buffer (pH 7.6), containing 0.01% hydrogen peroxide and 5 mg of 3,3’-diaminobenzidine tetrahydrochloride (Merck, Germany), they were then fixed in buffered oso4. Semithin sections (1 pm) from all tissue blocks were stained with toluidine blue and were surveyed by light microscopy. Thin sections were then cut on a LKB ultrotome, stained either with uranyl acetate and lead citrate or lead citrate only. These were then examined with a JEM 100 C electron microscope. The serum cholesterol level was measured at the beginning and end of the experiment by Zak’s method [lo]. The systolic blood pressure was measured by inserting a tube connected with a manometer into the femoral artery at the beginning of the experiment and just before killing [ 111. ReSUltS
(1) Serum cholesterol and systolic blood pressure
Serum cholesterol levels and systolic blood pressure at the end of the experiments are shown in Table 1. Serum cholesterol at the beginning was about 50 mg/dl and systolic blood pressure was below 100 mm Hg. Blood pressure was elevated in the first and second groups and serum cholesterol level in the second and third groups. (2) Cerebral arteries of cholesterol-fed rabbits with induced hypertension Semithin sections stained with toluidine blue showed severe foam cell lesions
139 TABLE 1 FINAL SERUM CHOLESTEROL ._.~
LEVEL AND BLOOD PRESSURE IN EXPERIMENTAL
ANIMALS
Group I (Constriction of renal arteries and stock diet feeding for 12 weeks) Number of rabbit Cholesterol (mg/dl) Blood pressure (mm Hg)
1 59 143
2 45 110
3 50 186
Group II (Constriction of renal arteries and cholesterol feeding for 12 weeks) Number of rabbit Cholesterol (mg/dI) Blood pressure (mm Hg)
1 568 138
2 288 170
3 867 164
4 1125 160
3 1130
4 1327
5 243 140
6 1254 160
7 334 135
Group III (Cholesterol feeding for 16 weeks) Number of rabbit Cholesterol (mg/dl)
1 810
2 835
in the thoracic and abdominal aorta of the rabbits in this group. Small foam cell lesions were found in the cerebral arteries in 3 out of 7 cholesterol-fed rabbits with induced hypertension. The lesions were most prominent in intimal CU8hiOn8 at branching sites, where the peroxidase reaction product was present in the intima (Fig. 1, inset). Foam cell lesions were hardly ever observed in the intima distant from branching sites.
Fig. 1. Foam cell lesion at branching site of cerebral artery. Group II, hypertensive and cholesterol-fed. x2.000. Lipid vacuoles in endothellal and eubendothelial ceils. Reduplicated internal elastic lamina encloses -00th mwle celJs with cytoplasmic projections. Widened subendotheIial space conteins finely oracular material (arrow). Inset: peroxidase in thickened lntima (arrow). X280.
140
Lipid vacuoles were found in both endothelial and the subendothelial cells at branching sites (Fig. 1). The internal elastic lamina was duplicated, enclosing a small number of smooth muscle cells with cytoplasmic projections (Fig. 1). In these areas, a small amount of peroxidase reaction product was present in the endothelial cell vesicles and in the interendothelial and the subendothelial spaces (Fig. 2). The endothelial cells were partly separated from subendothelial structures at the edge of intimal cushions at branching sites with no definite foam cell infiltration (Fig. 3). This separation corresponded to the interendothelial junction, where a small amount of peroxidase reaction product was present (Fig. 3). The enzyme was also present in the endothelial cell vesicles and in the interendothelial space (Fig. 3). In some foam cell lesions, lipid vacuoles were also found in the smooth muscle cells of the subendothelial space and of the media. A small number of foam cell lesions were also observed at branching sites in small-sized arteries. These small-sized arteries showed marked medial necroses and villous projections from their endothelial cells (Fig. 4A). Dense granular material was present in the subendothelial space, and the internal elastic lamina was thin and disrupted (Fig. 4A). In these areas, the peroxidase reaction product was observed in the endothelial cell vesicles, in the subendothelial space and also in the interendothelial space (Fig. 4B). In some of the small-sized
Fig. 2. Foam cell lesion at branching site of cerebral artery. Group II, hypertensive and cholesterol-fed. x10,ooo. Some peroxidsse is present in the endotbelisl cell vesicles and iu tbe iuterandotbelial and subendotbeUal spaces.
Fig. 3. The marginal portion of a cerebral arterlel intimal cushion. Group II. es for Fig 2. X6.300. Pertlal separationof endothelial celle from subendotheliel structure at dte of interendothelial junction. Note peroxidase in the endothellal cell vesicles end in the lnterendothelial end subendotheliel spaces.
Fin. 4. SmaU-elzed crerebral ertew. Group II, hyperterudve aad cholesterol-fed A: the endotbelial mrface shows vlllous projections. Note lipid vaeuolee end umell myelin &urea In subendothelirl cell. Denee grenuler meterlal is eeen in the eulmndotheliel spece. Mediel necroeis k prominent. X2.000. B: some perosbiese le preeent in the endothelial cell vesicles end in both the eubendotheliel end Intetendotheliel spaces. X10.000.
142
arteries with marked medial necroses, blood mononuclear cells were seen attached to the endothelial surface or interdigitated with endothelial cells or migrating to the intima between endothelial cells (Fig. 5). The cerebral arterial intima distant from branching sites occasionally showed separation of endothelial cells from the subendothelial structure, and the attenuated endothelial projections showed points of attachment associated with the endothelial dense zone (Fig. 6). Dense fibrillary deposits were present in the widened subendothelial space, but no peroxidase reaction product was observed in either the subendothelial or interendothelial spaces (Fig. 6). (3) Cerebral arteries of normocholesteremic rabbits with induced hypertension No remarkable changes were found in the endothelial cells and the interendothelial space of medium-sized cerebral arteries, but degeneration and necroses were seen in medial smooth muscle cells (Fig. 6, inset). Endothelial cells had separated from subendothelial structures. Subendothelial cells were seldom observed in the intima except at branching sites. Peroxidase reaction product was seen in the endothelial cell vesicles, but not in the interendothelial spaces, even at branching.sites. Medial smooth muscle cell degeneration and necroses were observed more frequently in small-sized than in medium-sized arteries.
6. Small-eked cerebral artery. Group II, hypertceive end cholesterol-fed X4.000. Blood mononuclear cell (Ml) on endotbellal m&ace, interd<etiue with endothelial cells. Another mononuclear cell (M2) migrates between the endotbelial cells. Peroxidaee Is prerent in the interendothelial space (indicated with anows). Grauular mater&l ie deposited in ths’subendothe~ space, while necrotic debti of emooth muecle cello ie seen in the media. F&.
143
Fig. 6. The intima of a cerebral artery distant from a branching site. Group II. hypertensive and cholesterol-fed. X10.000. Prominent separation of endothelial cells from subendothelial structure. Attenuated endothelial projections reveal the points of attachment associated with dense zones. Granular deposits and fibrillar materials in the subendothelial space. Inset: Cerebral artery distant from a branching site. Group I. hypertensive and normocholesteremic. X5.300. Necrosis of the medial smooth muscle cells is seen, but the endothelial cells show no remarkable changes.
(4) Cerebral arteries in normotensive cholesterol-fed rabbits Semithin sections stained with toluidine blue showed severe foam cell lesions in the thoracic and abdominal aorta of the rabbits in this group. However, the intima in the cerebral arteries showed no foam cell lesions, and no changes were found in the arterial wall. Peroxidase reaction product was observed in the endothelial cell vesicles, but not in the interendothelial or subendothelial spaces. Discussion Atherosclerosis is known to be less severe in the human cerebral artery than in the aorta and coronary artery [8,12]. In addition, the cerebral artery has proved to be more resistant to cholesterol-induced atherosclerosis in the experimental animal, especially in the rat and rabbit [ 1,3]. In our previous study [ 11, a positive horseradish peroxidase reaction was observed in the endothelial cell vesicles and the interendothelial and subendothelial spaces of the coronary arteries and aorta of normally fed and cholesterol-fed rabbits. Cholesterol feeding enhanced the permeability of the intima of these arteries. In intracranial arteries, however, peroxidase was observed only
144
in the endothelial cell vesicles, and not in the interendothelial or subendothelial space. The intima of intracranial arteries showed neither foam cell lesions nor altered permeability even after cholesterol feeding [ 11. The endothelial cells of these arteries were closely applied to the internal elastic lamina; the endothelial cells showed only a small number of vesicles, and tight junctions were the rule in the interendothelial space. These differences provide structural evidence of different intimal permeability characteristics between the aorta, coronary and cerebral arteries and, furthermore, suggest that the cerebral artery is more resistant to cholesterol-induced atherosclerosis than the other vessels studied. As well as the experimental evidence of hypertension enhancing atherosclerosis, autopsy evidence suggests that hypertension accelerates cerebral atherosclerosis [ 8,9]. Likewise, experimental hypertension enhances cerebrovascular permeability [ 13-161. In the present experiments, foam cell lesions were seen at branching sites; they were particularly prominent in intimal cushions in the cerebral arteries of cholesterol-fed rabbits with induced hypertension. Although the significance of the intimal cushion in cerebral atherogenesis is controversial, the intimal cushion might be either an important precursor or an integral stage in the evolution of cerebral atherosclerosis [ 171. Subendothelial cells were seldom found in the intima of rabbit cerebral arteries except for intimal cushions. In the present study, the foam cell lesions derived from subendothelial cells in the intimal cushion showed enhanced permeability to peroxidase, even though the interendothelial space was not evidently opened. In addition, peroxidase activity was seen in the endothelial cell vesicles, and in the interendothelial and subendothelial spaces at the edge of intimal cushions where no definite foam cell lesions were present. This suggested enhanced permeability preceded the development of the foam cell lesion in the intimal cushion [ 18,191, but it was not clear which route predominantly contributed to such enhanced permeability, junctional or vesicular transport. Enhanced permeability through the interendothelial space was apparent at branching sites in small-sized cerebral arteries, but it was uncertain to what extent vesicular transport contributed to it. In the cerebral arterial intima distant from branching sites, endothelial cells separated from the internal elastic lamina in hypertensive animals with or without hypercholesteremia. Adhesion of the attenuated endothelial projections persisted where dense zones were present. The dense zones in the endothelial projections seemed to be the structural point of endothelial attachment [20]. Our present experimental results in the rabbit show that hypertension accentuates permeability in cerebral arteries and that, when hypertension is linked with hypercholesteremia, cerebral atherosclerosis results. References 1 Kurozumi, T.. Electron microscopic study on permeability of the aorta and basilar artery of the rabbit - With special reference to the changes of permeability by hypercholerteremia, EXP. Mol. Path., 23 (1976) 1. 2 Kurozumi. T.. Sumiyoshi. A. and Tanaka, K.. Permeability of coronary artery In rabbi@ Jap. J. Atheroder.. 1 (1974) 193. 3 Kurozumi. T.. Tanaka, K. and Yae, Y., Different permeability pattern of the aorta and organ arteries -With special reference to its significance in atherogenesia, Jap. J. Atheroscler.. 4 (1977) 387.
145 Hept&&all, RI-I.. Barkely, II. aud Porter, K.A.. The relative roles of blood cholesterol level and blood pressure level in the production of experimental aortIc atberoma in rabbits, Anglology. 9 (1958) 84. 5 Page, I.H. and McCubbin, J.W., Renal Hypertension. Year Book Medical Publisher, Chicago. Ill. 1968, 4
P. 312. 6 Hollander. W.. Madoff, I., Paddock, J. and Kirkpatrick, B., Aggravation of atherosclerosis by hyper tension In a subhuman primate model with coarction of the aorta, Clrc. Res.. SUPP~. II. 38 (1976) 63. 7 Bretherton, K.N.. Day. A.J. and Skinner, S.L., Hypertension-accelerated atherogenesis in cholesterol-
fed rabbits, Atherosclerosis. 27 (1977) 79. 8 Sadoshima, S., Tanaka, K.. Takeshim, M.. Hirota. Y.. Omae, T. and Katsuki. S.. Atherosclerosis and atherosclerotic disorders In the autopsy cases of Hisayama survey in the recent 10 years, Fukuoka Acta Med., 85 (1974) 701. 9 Kuller, L. and Reisler, D.M., An explanation for variations in distribution of stroke and arteriosclerotic heart disease among populations and racial groups. Amer. J. Epidemlol.. 93 (1971) 1. 10 Zak, B., Dickenman. R.C., white. E.G., Burnett, H. and Chemey. P.J.. A rapid method for the estimation of total and free cholesterol, Amer. J. Clin. Path., 24 (1954) 1307. 11 Wakerlln, G.E., Moss, W.G. and Kiely, J.P.. Effect of experimental renal hypertension on exPerimental thiouracil-cholesterol atherosclerosis in dogs, Circ. Res., 5 (1957) 426. 12 Moses, C.. Atherosclerosis Mechanisms as a Guide to Prevention, Lea and Febiger, Philadelphia, Pa., 1963, p. 74. 13 Eto, T.. Omaa, T. and Yamamoto. T.. An electron microscope study of hypertensive encephalopathy in the rat with renal hypertension. Arch. Histol. Jap.. 33 (1971) 133. 14 Giacomelll, R., Wiener, J. and Spiro. D., The cellular pathology of experimental hypertension, Part 5 (Increased permeability of cerebral arterial vessels), Amer. J. Path., 59 (1970) 133. 15 Nag, S.. Robertson. D.M. and Dinsdale. J.B., Cerebral corticpl changes in acute experimental hypertension - An ultrastructuraI study, Lab. Invest., 36 (1977) 150. 16 Hanseon. H.A.. Jahansson. B. and Blomstrand, C., Uh’aStNCbTd studies on cerebrovascular permeability in acute hypertension, Acta Neuropath. (Berl.), 32 (1976) 187. 17 Stehbens. W.E.. Pathology of the Cerebral Blood Vessels, Ch. 3 (Atherosclerosis.), 3rd edition, Mosby. New York, N.Y.. 1972. p. 98. 18 Wright. H.P., Evans, M. and Green, R.P.. Aortic endothelial mitosis and Evans blue uptake in cholesterol-fed subscorbutic guinea-pigs, Atherosclerosis, 21 (1975) 106. 19 Adams, C.W.M., Bay&s. O.B. and Morgan, R.S., Permeability in atherosclerosis - Fluorescence test in green light with trypan blue, Atherosclerosis, 27 (1977) 353. 20 Ts’ao. C.H. and Glagov, S.. Basal endothellal attachment - Tenacity rabbit aorta, Lab. Invest., 23 (1970) 510.
at cytoplasmic
dense zones in the
Atherosclerosis, 30 (1978) 137-145 @ Elsevier/North-Holland Scientific Publishers, Ltd.
HYPERTENSION-INDUCED CEREBRAL ATHEROSCLEROSIS IN THE CHOLESTEROL-FED RABBIT
TAKESHI KUROZUMI, KENZO TANAKA and YOSHIAKI YAE Department of Pathology and Central Laboratory, Fukuoka 812 (Japan)
Faculty of Medicine, Kyushu University,
(Received 6 December, 1977) (Revised, received 30 December, 1977) (Accepted 30 December, 1977)
Foam cell lesions were found in cholesterol-fed rabbits with induced hypertension, particularly in intimal cushions at branching sites, where permeability to horseradish peroxidase was enhanced. Permeability to horseradish peroxidase was enhanced at the edge of intimal cushions without foam cell accumulation. This finding suggests that permeability is increased before foam cell infiltration. No foam cell lesions were observed in the intima of cerebral arteries distant from branching sites, but insudation of plasma constituents here caused endothelial cells to separate from the subendothelial matrices. Foam cell lesions were absent from the cerebral arteries in normotensive cholesterol-fed rabbits. Key words:
Atherosclerosis - Cerebral arteries - Cholesterol - Endothelium sion - Intimal cushion -Permeability - Feroxidase
- Hyperten-
Introduction
Previously we found that the permeability pattern of cerebral arteries differed from that of the coronary arteries and aorta, in relation to the ultrastructural characteristics of the intima of these arteries [l-3]. The cerebral artery was resistant to cholesterol-induced atherosclerosis in rats and rabbits, presumably because of a barrier function by the intima. Hypertension is known to promote the evolution of atherosclerosis in the experimental animal [4-71. Autopsy studies indicate that hypertension accelerates cerebral atherosclerosis [ 8,9]. In this paper, the effect was studied of renovascular hypertension on structure and permeability of cerebral arteries in hypercholesteremic rabbits.
138
Materials and Methods Three groups of New Zealand albino male rabbits weighing 2.0-2.5 kg were used. The first group of 5 rabbits was maintained on stock diet (Oriental Co., Japan) for 12 weeks. The second group of 10 rabbits was fed the same diet with added cholesterol for 12 weeks. Sufficient cholesterol was rubbed into the diet to provide about 0.5 g cholesterol per animal per day. The left renal artery in the animals of these two groups was constricted with a silver clip on the second week of the experiment, followed by constriction of the right renal artery 2 weeks later. The third group of 4 rabbits was simply fed the diet containing 0.5 g cholesterol per day for 16 weeks. Two rabbits in the first group and 3 in the second died in the course of the experiment. Two rabbits in each of the first and third groups and 4 rabbits in the second group were injected with horseradish peroxidase (Sigma Chemical Co., Type II) 250 mg/kg body weight dissolved in 30 ml of isotonic saline, 15 min before killing. Tissue blocks were taken from the aorta and the medium- and small-sized intracranial arteries, that is, basilar, middle and anterior cerebral arteries and the circle of Willis including their branches. Tissue blocks were fixed with 3% glutaraldehyde buffered with 0.1 M cacodylate for 5-6 h at room temperature and then washed overnight at 4°C in 0.1 M cacodylate buffer. After post-fixation with 1% 0~0,. buffered with 0.1 M cacodylate for 1.5 h, they were then dehydrated in graded ethanol, treated with propylene oxide, and embedded in Epon 812. Tissue blocks from rabbits injected with horseradish peroxidase were incubated for 15 min at room temperature in 10 ml of 0.05 M Tris-HCl buffer (pH 7.6), containing 0.01% hydrogen peroxide and 5 mg of 3,3’-diaminobenzidine tetrahydrochloride (Merck, Germany), they were then fixed in buffered oso4. Semithin sections (1 pm) from all tissue blocks were stained with toluidine blue and were surveyed by light microscopy. Thin sections were then cut on a LKB ultrotome, stained either with uranyl acetate and lead citrate or lead citrate only. These were then examined with a JEM 100 C electron microscope. The serum cholesterol level was measured at the beginning and end of the experiment by Zak’s method [lo]. The systolic blood pressure was measured by inserting a tube connected with a manometer into the femoral artery at the beginning of the experiment and just before killing [ 111. ReSUltS
(1) Serum cholesterol and systolic blood pressure
Serum cholesterol levels and systolic blood pressure at the end of the experiments are shown in Table 1. Serum cholesterol at the beginning was about 50 mg/dl and systolic blood pressure was below 100 mm Hg. Blood pressure was elevated in the first and second groups and serum cholesterol level in the second and third groups. (2) Cerebral arteries of cholesterol-fed rabbits with induced hypertension Semithin sections stained with toluidine blue showed severe foam cell lesions
139 TABLE 1 FINAL SERUM CHOLESTEROL ._.~
LEVEL AND BLOOD PRESSURE IN EXPERIMENTAL
ANIMALS
Group I (Constriction of renal arteries and stock diet feeding for 12 weeks) Number of rabbit Cholesterol (mg/dl) Blood pressure (mm Hg)
1 59 143
2 45 110
3 50 186
Group II (Constriction of renal arteries and cholesterol feeding for 12 weeks) Number of rabbit Cholesterol (mg/dI) Blood pressure (mm Hg)
1 568 138
2 288 170
3 867 164
4 1125 160
3 1130
4 1327
5 243 140
6 1254 160
7 334 135
Group III (Cholesterol feeding for 16 weeks) Number of rabbit Cholesterol (mg/dl)
1 810
2 835
in the thoracic and abdominal aorta of the rabbits in this group. Small foam cell lesions were found in the cerebral arteries in 3 out of 7 cholesterol-fed rabbits with induced hypertension. The lesions were most prominent in intimal CU8hiOn8 at branching sites, where the peroxidase reaction product was present in the intima (Fig. 1, inset). Foam cell lesions were hardly ever observed in the intima distant from branching sites.
Fig. 1. Foam cell lesion at branching site of cerebral artery. Group II, hypertensive and cholesterol-fed. x2.000. Lipid vacuoles in endothellal and eubendothelial ceils. Reduplicated internal elastic lamina encloses -00th mwle celJs with cytoplasmic projections. Widened subendotheIial space conteins finely oracular material (arrow). Inset: peroxidase in thickened lntima (arrow). X280.
140
Lipid vacuoles were found in both endothelial and the subendothelial cells at branching sites (Fig. 1). The internal elastic lamina was duplicated, enclosing a small number of smooth muscle cells with cytoplasmic projections (Fig. 1). In these areas, a small amount of peroxidase reaction product was present in the endothelial cell vesicles and in the interendothelial and the subendothelial spaces (Fig. 2). The endothelial cells were partly separated from subendothelial structures at the edge of intimal cushions at branching sites with no definite foam cell infiltration (Fig. 3). This separation corresponded to the interendothelial junction, where a small amount of peroxidase reaction product was present (Fig. 3). The enzyme was also present in the endothelial cell vesicles and in the interendothelial space (Fig. 3). In some foam cell lesions, lipid vacuoles were also found in the smooth muscle cells of the subendothelial space and of the media. A small number of foam cell lesions were also observed at branching sites in small-sized arteries. These small-sized arteries showed marked medial necroses and villous projections from their endothelial cells (Fig. 4A). Dense granular material was present in the subendothelial space, and the internal elastic lamina was thin and disrupted (Fig. 4A). In these areas, the peroxidase reaction product was observed in the endothelial cell vesicles, in the subendothelial space and also in the interendothelial space (Fig. 4B). In some of the small-sized
Fig. 2. Foam cell lesion at branching site of cerebral artery. Group II, hypertensive and cholesterol-fed. x10,ooo. Some peroxidsse is present in the endotbelisl cell vesicles and iu tbe iuterandotbelial and subendotbeUal spaces.
Fig. 3. The marginal portion of a cerebral arterlel intimal cushion. Group II. es for Fig 2. X6.300. Pertlal separationof endothelial celle from subendotheliel structure at dte of interendothelial junction. Note peroxidase in the endothellal cell vesicles end in the lnterendothelial end subendotheliel spaces.
Fin. 4. SmaU-elzed crerebral ertew. Group II, hyperterudve aad cholesterol-fed A: the endotbelial mrface shows vlllous projections. Note lipid vaeuolee end umell myelin &urea In subendothelirl cell. Denee grenuler meterlal is eeen in the eulmndotheliel spece. Mediel necroeis k prominent. X2.000. B: some perosbiese le preeent in the endothelial cell vesicles end in both the eubendotheliel end Intetendotheliel spaces. X10.000.
142
arteries with marked medial necroses, blood mononuclear cells were seen attached to the endothelial surface or interdigitated with endothelial cells or migrating to the intima between endothelial cells (Fig. 5). The cerebral arterial intima distant from branching sites occasionally showed separation of endothelial cells from the subendothelial structure, and the attenuated endothelial projections showed points of attachment associated with the endothelial dense zone (Fig. 6). Dense fibrillary deposits were present in the widened subendothelial space, but no peroxidase reaction product was observed in either the subendothelial or interendothelial spaces (Fig. 6). (3) Cerebral arteries of normocholesteremic rabbits with induced hypertension No remarkable changes were found in the endothelial cells and the interendothelial space of medium-sized cerebral arteries, but degeneration and necroses were seen in medial smooth muscle cells (Fig. 6, inset). Endothelial cells had separated from subendothelial structures. Subendothelial cells were seldom observed in the intima except at branching sites. Peroxidase reaction product was seen in the endothelial cell vesicles, but not in the interendothelial spaces, even at branching.sites. Medial smooth muscle cell degeneration and necroses were observed more frequently in small-sized than in medium-sized arteries.
6. Small-eked cerebral artery. Group II, hypertceive end cholesterol-fed X4.000. Blood mononuclear cell (Ml) on endotbellal m&ace, interd<etiue with endothelial cells. Another mononuclear cell (M2) migrates between the endotbelial cells. Peroxidaee Is prerent in the interendothelial space (indicated with anows). Grauular mater&l ie deposited in ths’subendothe~ space, while necrotic debti of emooth muecle cello ie seen in the media. F&.
143
Fig. 6. The intima of a cerebral artery distant from a branching site. Group II. hypertensive and cholesterol-fed. X10.000. Prominent separation of endothelial cells from subendothelial structure. Attenuated endothelial projections reveal the points of attachment associated with dense zones. Granular deposits and fibrillar materials in the subendothelial space. Inset: Cerebral artery distant from a branching site. Group I. hypertensive and normocholesteremic. X5.300. Necrosis of the medial smooth muscle cells is seen, but the endothelial cells show no remarkable changes.
(4) Cerebral arteries in normotensive cholesterol-fed rabbits Semithin sections stained with toluidine blue showed severe foam cell lesions in the thoracic and abdominal aorta of the rabbits in this group. However, the intima in the cerebral arteries showed no foam cell lesions, and no changes were found in the arterial wall. Peroxidase reaction product was observed in the endothelial cell vesicles, but not in the interendothelial or subendothelial spaces. Discussion Atherosclerosis is known to be less severe in the human cerebral artery than in the aorta and coronary artery [8,12]. In addition, the cerebral artery has proved to be more resistant to cholesterol-induced atherosclerosis in the experimental animal, especially in the rat and rabbit [ 1,3]. In our previous study [ 11, a positive horseradish peroxidase reaction was observed in the endothelial cell vesicles and the interendothelial and subendothelial spaces of the coronary arteries and aorta of normally fed and cholesterol-fed rabbits. Cholesterol feeding enhanced the permeability of the intima of these arteries. In intracranial arteries, however, peroxidase was observed only
144
in the endothelial cell vesicles, and not in the interendothelial or subendothelial space. The intima of intracranial arteries showed neither foam cell lesions nor altered permeability even after cholesterol feeding [ 11. The endothelial cells of these arteries were closely applied to the internal elastic lamina; the endothelial cells showed only a small number of vesicles, and tight junctions were the rule in the interendothelial space. These differences provide structural evidence of different intimal permeability characteristics between the aorta, coronary and cerebral arteries and, furthermore, suggest that the cerebral artery is more resistant to cholesterol-induced atherosclerosis than the other vessels studied. As well as the experimental evidence of hypertension enhancing atherosclerosis, autopsy evidence suggests that hypertension accelerates cerebral atherosclerosis [ 8,9]. Likewise, experimental hypertension enhances cerebrovascular permeability [ 13-161. In the present experiments, foam cell lesions were seen at branching sites; they were particularly prominent in intimal cushions in the cerebral arteries of cholesterol-fed rabbits with induced hypertension. Although the significance of the intimal cushion in cerebral atherogenesis is controversial, the intimal cushion might be either an important precursor or an integral stage in the evolution of cerebral atherosclerosis [ 171. Subendothelial cells were seldom found in the intima of rabbit cerebral arteries except for intimal cushions. In the present study, the foam cell lesions derived from subendothelial cells in the intimal cushion showed enhanced permeability to peroxidase, even though the interendothelial space was not evidently opened. In addition, peroxidase activity was seen in the endothelial cell vesicles, and in the interendothelial and subendothelial spaces at the edge of intimal cushions where no definite foam cell lesions were present. This suggested enhanced permeability preceded the development of the foam cell lesion in the intimal cushion [ 18,191, but it was not clear which route predominantly contributed to such enhanced permeability, junctional or vesicular transport. Enhanced permeability through the interendothelial space was apparent at branching sites in small-sized cerebral arteries, but it was uncertain to what extent vesicular transport contributed to it. In the cerebral arterial intima distant from branching sites, endothelial cells separated from the internal elastic lamina in hypertensive animals with or without hypercholesteremia. Adhesion of the attenuated endothelial projections persisted where dense zones were present. The dense zones in the endothelial projections seemed to be the structural point of endothelial attachment [20]. Our present experimental results in the rabbit show that hypertension accentuates permeability in cerebral arteries and that, when hypertension is linked with hypercholesteremia, cerebral atherosclerosis results. References 1 Kurozumi, T.. Electron microscopic study on permeability of the aorta and basilar artery of the rabbit - With special reference to the changes of permeability by hypercholerteremia, EXP. Mol. Path., 23 (1976) 1. 2 Kurozumi. T.. Sumiyoshi. A. and Tanaka, K.. Permeability of coronary artery In rabbi@ Jap. J. Atheroder.. 1 (1974) 193. 3 Kurozumi. T.. Tanaka, K. and Yae, Y., Different permeability pattern of the aorta and organ arteries -With special reference to its significance in atherogenesia, Jap. J. Atheroscler.. 4 (1977) 387.
145 Hept&&all, RI-I.. Barkely, II. aud Porter, K.A.. The relative roles of blood cholesterol level and blood pressure level in the production of experimental aortIc atberoma in rabbits, Anglology. 9 (1958) 84. 5 Page, I.H. and McCubbin, J.W., Renal Hypertension. Year Book Medical Publisher, Chicago. Ill. 1968, 4
P. 312. 6 Hollander. W.. Madoff, I., Paddock, J. and Kirkpatrick, B., Aggravation of atherosclerosis by hyper tension In a subhuman primate model with coarction of the aorta, Clrc. Res.. SUPP~. II. 38 (1976) 63. 7 Bretherton, K.N.. Day. A.J. and Skinner, S.L., Hypertension-accelerated atherogenesis in cholesterol-
fed rabbits, Atherosclerosis. 27 (1977) 79. 8 Sadoshima, S., Tanaka, K.. Takeshim, M.. Hirota. Y.. Omae, T. and Katsuki. S.. Atherosclerosis and atherosclerotic disorders In the autopsy cases of Hisayama survey in the recent 10 years, Fukuoka Acta Med., 85 (1974) 701. 9 Kuller, L. and Reisler, D.M., An explanation for variations in distribution of stroke and arteriosclerotic heart disease among populations and racial groups. Amer. J. Epidemlol.. 93 (1971) 1. 10 Zak, B., Dickenman. R.C., white. E.G., Burnett, H. and Chemey. P.J.. A rapid method for the estimation of total and free cholesterol, Amer. J. Clin. Path., 24 (1954) 1307. 11 Wakerlln, G.E., Moss, W.G. and Kiely, J.P.. Effect of experimental renal hypertension on exPerimental thiouracil-cholesterol atherosclerosis in dogs, Circ. Res., 5 (1957) 426. 12 Moses, C.. Atherosclerosis Mechanisms as a Guide to Prevention, Lea and Febiger, Philadelphia, Pa., 1963, p. 74. 13 Eto, T.. Omaa, T. and Yamamoto. T.. An electron microscope study of hypertensive encephalopathy in the rat with renal hypertension. Arch. Histol. Jap.. 33 (1971) 133. 14 Giacomelll, R., Wiener, J. and Spiro. D., The cellular pathology of experimental hypertension, Part 5 (Increased permeability of cerebral arterial vessels), Amer. J. Path., 59 (1970) 133. 15 Nag, S.. Robertson. D.M. and Dinsdale. J.B., Cerebral corticpl changes in acute experimental hypertension - An ultrastructuraI study, Lab. Invest., 36 (1977) 150. 16 Hanseon. H.A.. Jahansson. B. and Blomstrand, C., Uh’aStNCbTd studies on cerebrovascular permeability in acute hypertension, Acta Neuropath. (Berl.), 32 (1976) 187. 17 Stehbens. W.E.. Pathology of the Cerebral Blood Vessels, Ch. 3 (Atherosclerosis.), 3rd edition, Mosby. New York, N.Y.. 1972. p. 98. 18 Wright. H.P., Evans, M. and Green, R.P.. Aortic endothelial mitosis and Evans blue uptake in cholesterol-fed subscorbutic guinea-pigs, Atherosclerosis, 21 (1975) 106. 19 Adams, C.W.M., Bay&s. O.B. and Morgan, R.S., Permeability in atherosclerosis - Fluorescence test in green light with trypan blue, Atherosclerosis, 27 (1977) 353. 20 Ts’ao. C.H. and Glagov, S.. Basal endothellal attachment - Tenacity rabbit aorta, Lab. Invest., 23 (1970) 510.
at cytoplasmic
dense zones in the