Atherosclerosis, 87 (1991) 211-220 ‘a 1991 Elsevier Scientific Publishers ADONIS 0021915091000997
ATHERO
211 Ireland,
Ltd. 0021-9150/91/$03.50
04622
Formation of a lipid gradient across the human aortic wall during ageing and the development of atherosclerosis Lawrence T. McGrath * and Robert J. Elliott Division of Biochemistry
School of Biology and Biochemistry, The Queen’s University of Belfast, Belfast BT9 7BL, Northern Ireland (U.K.) (Received 17 July, 1990) (Revised received 26 November, 1990) (Accepted 19 December, 1990)
Summary
The smooth muscle cell invasion and macrophage stimulation within the intima during prolonged exposure to high blood levels of cholesterol esters contribute to increased production of connective tissue matrix. The thickened intima in turn immobilising more LDL derived lipid from the plasma. With damage to the internal elastic lamellae, from essential hypertension, the absorbed lipid can move down a concentration gradient into the medial tissue. This model was supported by our laboratory finding of a lipid gradient across the aorta wall. The gradient commenced shortly after completion of body growth, when the transmedial gradient became detectable. The slope of the gradient progressively increased during ageing. Association of the lipid medial gradient with the degree of atherosclerotic involvement suggested that the gradient influenced the development of intimal lesions. Accumulation of lipid within the medial tissue may then reduce the inward lipid transfer rate from the intima, promoting increased intimal retention and cause the formation of atherosclerotic plaques from the fat saturated intima.
Key words: Ageing; Aorta; Atherosclerosis; LDL
Cholesterol ester; Connective tissue matrix; Elastic lamina;
Introduction
Atherosclerosis has been considered an example of a disease with a common initiating source,
Present address and correspondence: Dr. L.T. McGrath, Department of Therapeutics and Pharmacology, The Queen’s University of Belfast, Belfast BT9 7BL, Northern Ireland, U.K. Tel. (0232) 245133 ext. 2203; Fax (0232)438346.
the amount of cholesterol and saturated fat ingested [l]. The level of cholesterol circulating in the plasma, mainly in the form of beta-lipoproteins, has been implicated in the atherosclerotic process from 40 epidemiological studies throughout the world [2]. The role of plasma cholesterol in the progression of atherosclerosis has been linked to the composition of atherosclerotic plaques, the cores of which contain up to
212 65% (by dry weight) of lipid, the majority composed of cholesterol and its esters [3]. The hallmark of atherosclerotic plaques is their complexity and the many different processes that can lead to their formation such as: endothelial injury, lipid uptake by macrophages, lipid infiltration, disintegration of lipid filled (foam) cells, extracellular lipid deposition, elastin destruction, smooth muscle cell proliferation, collagen synthesis, proteoglycan accumulation, calcium deposition, necrosis, and plaque thrombosis [4]. The effects of experimentally induced hyperlipidaemia on the lipid composition of arterial tissue has been examined using laboratory animals [5]. These short term studies are unlikely to reflect the patterns found in the much longer development time of atherosclerosis in humans. The flux of lipid entering and leaving the tissue, even when taken as a single variable, must be considered within a tissue which itself is changing metabolically and structurally during ageing [6]. The accumulating lipid would be expected to progressively alter both the rate and the nature of this ageing pattern. If to this already complex process is added the unpredictable occurrence, duration and effects of secondary risk factors, a multitude of pathological variations can be generated [7]. A chronic disease process, such as atherosclerosis, with it’s long development period restricts investigation to human tissue. This requirement has imposed limitations on experimental design and availability of test materials. The present study has utilised aortas obtained during the post mortem examination of patients, aged 17 to 97 years, who had died in hospital. The inference has had to be made that this sample of aortas presented a uniform pattern of lipid progression, such as could be found in the general population during ageing. Physically subdividing the aortic wall into anatomically defined zones has allowed us to make a more detailed examination of the rate of lipid accumulation during ageing. The dynamics of lipid uptake and saturation levels in different zones of the human aorta could then be determined, Materials and methods Aorta collection and dissection Anatomically defined segments
of descending
thoracic aorta, were collected during post-mortem examination from patients who had died in hospital from non-infectious diseases. Samples were wrapped in damp muslin and, if not processed immediately, stored at - 20” C. Frozen samples, when required for analysis, were thawed overnight at 4 o C. The adventitia and any adhering adipose tissue were removed. The segment of aorta selected for analysis was the region that contained the first six intercostal branches (numbered from the aortic arch end). The sample was opened along its frontal longitudinal plane to expose the intimal surface, which was washed with cold 0.9% (w/v) NaCl to remove any adhering blood products. The intercostal holes were located down either side of the centre axis of the opened specimen. These intercostal branches were used as reference points during subsequent tissue mapping and sampling. The flat opened segment of aorta was placed, intimal surface upwards, on a dissection block and the areas covered by each lesion type were mapped onto a clear plastic sheet [8]. Lesion types were then measured as a percentage of the total surface area, graded according to the WHO classification system [9]. The degree of tissue involvement with atherosclerotic lesions, the atherosclerotic index (AI), was expressed as the cumulative percentage of the total surface area covered by atheroma, fibrous plaques and complicated lesions. Macroscopically involved and uninvolved plaque areas of aorta were dissected from the sample. A section of tissue, approximately 1 cm square, was selected to the left of the third intercostal branch. When uninvolved tissue was sampled the area dissected did not come closer than 3 mm from the boundaries of adjacent plaques. If the tissue was extensively ulcerated, smaller lesion free areas were pooled to produce a viable sample for analysis. Quality control of these samples prepared for lipid extraction were screened by histological procedures (see below). The intimal tissue was removed by microdissection and the media was then subdivided into three layers of equal thickness. The dissection of the media into subdivisions was aided by its inherent laminar structure. Although more than three layers could have been obtained the present protocol was found to be practical, quick and reliable. A description of the sampling and analysis procedures developed
213 for this study has been reported [lo]. The medial layers, when separated, were designated Mi (inner media), Mm (middle media), and MO (outer media), the former being that layer closest to the intima. Individual samples were weighed prior to lipid extraction providing a further check that the three medial samples were of similar thickness. Microscopic examination
Small guide sections, cut vertical to the lumen surface of the sampled tissue, were processed through formaldehyde fixation and paraffin embedding to produce 5 pm thick serial sections. Where a number of small samples were pooled, histology was performed on each. Sections were stained by haematoxylin-eosin and by Weigert’s elastin (Hart’s modification) techniques. The prepared tissue sections were examined to confirm; (a) that samples, from lesion free areas had an intact intimal surface (the endothelial layer was present and extended lipid vacules were absent from within the tissue); (b) that intimal removal from the inner medial sample was complete and had not removed the internal elastic lamina. The haematoxylin-eosin treated sections were most appropriate for evaluation of the intimal material and the elastin stained sections were used to confirm the retention of the internal elastic lamina by the inner medial samples. Approximately 15% of samples had to be excluded from the study by failing to comply with the requirements described above. The tissue sections prepared for quality control were subsequently used to measure intimal and medial thickness. An Orthoplan microscope, fitted with a micrometer eyepiece was used at an overall magnification of x 100. Each tissue layer was measured in triplicate and the mean thickness determined. The thickness data (see Results) relates to fixed, paraffin embedded tissue, adopting the processing protocols used by previous investigators [11,12].
acetone, then 3 times with chloroform/methanol (2 : 1, v/v), and finally with acetone. Extractions were carried out at 4°C under nitrogen in the dark. The five lipid extracts from each tissue sample were pooled and stored at - 20” C under nitrogen to await analysis. The delipidated tissue was vacuum dried to a constant weight and recorded. Total cholesterol was analysed by the method of Watson [14], and unesterified cholesterol by the method of Sale et al. [15]. Esterified cholesterol was calculated from the total and unesterified cholesterol. Triacylglycerols were determined as described by Wahlefield [16], and phospholipid by the method of Anderson and Davis [17]. Expression of results
The total lipid content of each layer of tissue was calculated as the sum of esterified cholesterol, unesterified cholesterol, triacylglycerol and phospholipid and expressed as ug lipid/mg dry defatted tissue (Table 1). Individual lipid components were expressed as a percentage of the total lipid content (Table 2). The presence of inorganic material did not significantly affect the data, except with grossly atherosclerotic arteries [18]. Assay performance analysis produced day to day variation for total cholesterol of 6.18, unesterified cholesterol 6.78, triacylglycerol 5.9% and phospholipid 5.8% [19]. Statistical significance between 20 year age groups for each tissue layer was analysed. As normal distribution for each group could not be assumed non-parametric methods were used [20]. Analyses between age groups were carried out using Wilcoxon’s rank sum test for independent groups. Analyses within age groups were carried out using Wilcoxon’s signed rank test for matched groups (Table 1). Results Morphometric analysis
Lipid extraction and analysis
Lipids were extracted from the intact tissue samples by a modification of the procedure of Folch [13]. Samples were placed into 5-ml glassstoppered tubes and extracted successively (20 ml per g wet tissue weight) for 24 h periods, with
Intimal thickness was found to progressively increase during ageing, from 20 pm in the 2nd decade to 300 pm by the 10th decade. There was a strong correlation between intimal thickness and age (r = 0.762). The medial layer thickness from the same aortae displayed large variations between
214 samples of similar age, the correlation was r = 0.256 (Fig. 1).
with age 16001
/
Lipid content and composition during ageing
The lipid content and composition was examined after grouping the samples into five 20-y age groups (Table 1). The total lipid content of the intima progressively increased with age until the ninth decade (r = 0.724). Medial tissue lipid also increased with ageing. The lipid content and ageing in the inner, middle and outer media were significantly correlated (r = 0.703,0.754 and 0.747, respectively). The lipid content of the three medial layers was highly correlated with the degree of atherosclerotic involvement of the aorta (r = 0.740, 0.526 and 0.523 for inner, middle and outer media). The predominant lesion type, as was to be expected, changed with age. There was poor correlation (r = -C0.200) between the lipid content of the three medial layers and any particular lesion type. It is also somewhat difficult to avoid the conclusion that a ‘complicated lesion’ was not at some stage of its development an atheroma and/or fibrous plaque. The use of the AI was an attempt to avoid this complication. The lipid in the medial layers was also compared with the overlying intimal thickness (r = 0.509, 0.663 and 0.698 for inner, middle and outer media). Medial lipid can also be expressed as a percentage of intimal lipid content from the data in Table 1. In the youngest age group (c 20 y) all three medial tissue zones contained approximately 50% of the intimal tissue concentration. With increasing age the inner medial tissue lipid, as a percentage of age matched intimal lipid content, increased gradually to reach approximately 75% of the intimal value, as found in the oldest age group samples (81-97 y). In contrast the deepest medial zone (MO) where, although the quantity of lipid accumulated with age increased, the lipid as a percentage of the intimal tissue steadily decreased with age (from 55% in the < 20 y group to 34% in the 81-97 y age group). As medial lipid increased it was associated with a rise in the relative proportion of esterified cholesterol and as a consequence with dilution of the phospholipid content (Table 2). The ratio of unesterified cholesterol to phospholipid decreased with increasing lipid content.
,
,
,
,
.1600
lY /’ ,
;
a t
B
1
.u 5
000-
.o P
80 years) in our sample had a lipid profile similar to that which was present in the middle medial zone of the 61-80 year old samples. This middle medial age group in turn matched the inner medial zone lipid values found in the 41-60 year age group. Plasma derived LDL can be transported through the endothelium by selective receptor mediated transport systems [32]. A number of workers have proposed that non-specific fluid endocytosis can account for all of the LDL transport into the artery wall [33-351. Srinivasan [36] has demonstrated that more than 70% of LDL enters the aorta by a receptor independent pathway. Intimal and medial cellular modification of the LDL can occur before its partial return to the plasma or transfer inwards towards the media [37]. The findings in this study suggest that interaction
218 of insuded lipid with the aortic matrix was followed by increasing retention. The lipid that accumulated was correlated with age, r = 0.703 (P < O.OOl), and to a lesser degree with atherosclerosis, r =‘0.608 (P -c 0.001). Curmi [38] has shown the establishment of a LDL gradient across the aorta wall in rabbits. This gradient was highly dependant upon pressure. The gradient only extended into the inner half of the media. In short term studies in squirrel monkeys Tompkins [39] has shown that a steep LDL gradient is established across the intima but does not extend into the media. Smith and Staples [40] demonstrated that penetration of LDL into the media depended upon fragmentation of the internal elastic lamina. During ageing, with associated hypertension, partial loss of the integrity of the internal elastic lamina would occur [12]. The lamina has been proposed as an effective inward barrier to LDL migration [41]. Where fragmentation of the internal lamina had occurred lipid entry into the medial tissue would then be possible [42]. Smith and Staples [43] have suggested that large molecules, such as LDL, would be retained within the intima by both molecular sieving and an intact internal elastic lamina. In young adults the LDL content of medial tissue immediately adjacent to the elastic lamina was found to be 0.3% of that found in the intima [12]. Where the elastic lamina had been extensively fragmented, as in older tissue, there was a 25fold increase of non-membrane lipid in the medial tissue [40]. Kramsch and Hollander [44] found that LDL interacted with the aortic elastin, resulting in the formation of an elastinlipid complex. The residual protein component of the LDL, became immobilised to form a fibrin type residue. The gradients of total lipid, free and esterified cholesterol found in the present study were consistent with lipid derived from LDL that had migrated from the intima. The establishment of the transmedial lipid gradient may well precede the development of an atherosclerotic intimal lesion. Accumulation of lipid in the medial tissue would be expected to significantly reduce the rate of lipid transfer from the intima into the media, eventually causing lipid saturation of the intima. Local areas of the intimal tissue matrix could then form lipid plaque lesions.
Atheroma formation and associated lesions represent the outcomes of a long and complex interaction between the blood derived lipid and the arterial tissue. The present work has attempted to resolve part of this process. We have found that the extracellular matrix of the arterial tissue accumulated large quantities of cholesterol esters. The thin intima readily became saturated, promoting the formation of fat filled plaques. Diffusion of cholesterol esters also occurred inwards to reach even the deepest layers of medial tissue. The results presented here have revealed the formation of a lipid gradient across the full width of the largest artery of the body, the conclusion of long term exposure of the aorta to high levels of circulating lipids. Interaction between the circulating milieu and the artery wall has its effects not just upon the intima and adjacent media but penetrates into the deepest medial layers of the thoracic aorta.
Acknowledgements We would like to thank Dr. J.D. Biggart and Dr. G.W. Moore for the samples of aortic tissue obtained during post-mortem examinations and Mr. J. Orchin for producing the paraffin embedded tissue sections. We also wish to express our thanks to Professor J. Shepherd for a useful discussion of this work, prior to the preparation of this paper.
References Kuller, L.H., Epidemiology of coronary heart disease. In: E.G. Perkins and W.J. Visek (Eds.), Dietary Fats and Health, Champaign, IL, American Oil Chemists’ Society, 1983, p. 466. Intersociety Commission for Heart Disease Resources Atherosclerosis Group, Optimal resources for primary prevention of atherosclerosis (Abstract). Circulation, 70 (1984) 157A. Panganamala, R.V., Geer, J.C., Sharma, H.M. and Comwell, D.G., The gross and histologic appearance and the lipid composition of normal intima and lesions from human coronary arteries and aorta. Atherosclerosis, 20 (1974) 93. Constantinides, P., Arteries and arterioles. In: Ultrastructural Pathobiology, Elsevier, New York, 1984, p. 78. Reitman, J.S., Mahley, R.W. and Fry, D.L., Yucatan minia-
219
6
7 8
9 10
11 12 13 14 15
16
17
1X
19
20 21
22 23
ture swine as a model for diet-induced atherosclerosis. Atherosclerosis, 43 (1982) 119. Merrilees, M.J., Campbell, J.H., Spanidis, E. and Campbell, G.R., Glycosaminoglycans synthesis by smooth muscle cells of differing phenotype and their response to endothelial cell conditioned medium. Atherosclerosis, 81 (1990) 245. McGill, H.C., The pathogenesis of atherosclerosis. Clin. Chem., 34 (1988) B33. Mitchell, J.R.A. and Cranston, WI., A simple method for the quantitative assessment of aortic disease. J. Atheroscler. Res., 5 (1965) 135. WHO, Classification of atherosclerotic lesions. WHO Tech. Rep. Ser., 143 (1958) 1. McGrath, L.T., Biggart, D.J. and Elliott, R.J., Lipid composition in the intimal and medial zones of human aorta. Biochem. Sot. Trans., 15 (1987) 254. Cook. T.A. and Yates, P.O., A critical survey of techniques for arterial mensuration. J. Pathol., 108 (1972) 119. Semba. I., A histometrical analysis of age changes in the human lingual artery. Arch.Oral Biol., 34(7) (1989) 483. Radin. N.S., Preparation of lipid extracts. Methods Enzymol.. 14 (1969) 245. Watson, D., A simple method for the determination of serum cholesterol. Clin. Chim. Acta, 5 (1960) 637. Sale, F.O., Marchesini, S.. Fishman, P.H. and Berra, B., A sensitive enzymatic assay for determination of cholesterol in lipid extracts. Anal. Biochem., 142 (1984) 347. Wahlefield, A.W., Triglycerides. Determination after enzymatic hydrolysis. In: H.U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, Vol. 4, Verlag-Chemie, Weinheim and Academic Press, New York, 1974, p.1831. Anderson, R.L. and Davis, S., An organic phosphorous assay which avoids the use of hazardous perchloric acid. Clin. Chim. Acta, 121 (1982) 111. Yla-Herttuala, S., Sumuvori, H., Karkola, K., Mottonen, M. and Nikkari, T., Comparison of different ways of presenting the results of biochemical analyses of human carotid arteries. Atherosclerosis, 61 (1986) 77. McGrath, L.T. and Elliott, R.J., Lipid analysis and fatty acid profiles of individual arterial atherosclerotic plaques. Anal. Biochem., 187 (1990) 273. Bland, M., An Introduction to Medical Statistics, Oxford University Press, Oxford, 1989, p.216. Smith, E.B., Evans, P.H. and Downham, M.D., Lipid in the aortic intima. The correlation of morphological and chemical characteristics. J. Atheroscler. Res., 7 (1967) 171. Smith, E.B., Molecular interactions in human atherossclerotic plaques. Am. J. Pathol., 86 (1977) 665. Milch, R.A., Matrix properties of the aging arterial wall. In: W.G. Analyan and D.C. Sabiston (Eds.), Monographs in the Surgical Sciences. Vol. 2, Williams and Wilkins Co..
Baltimore, MD, 1965. p. 261. 24 Ross, R., The pathogenesis of atherosclerosis-an update. New Engl. J. Med., 314 (1986) 488. 25 Lundberg, B.. Chemical composition and physical state of lipid deposits in atherosclerosis. Atherosclerosis, 56 (1985) 93. 26 Smith.
E.B.. The relationship
between
plasma
and
tissue
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
lipids in human atherosclerosis. Adv. Lipid Res., 12 (1974) 1. Smith, E.B., The influence of age and atherosclerosis on the chemistry of aortic intima. J. Atheroscler. Res., 5 (1965) 224. Katz, S.S., Shipley, G.G. and Small, D.M., Physical chemistry of the lipids of human atherosclerotic lesions. Demonstration of a lesion intermediate between fatty streaks and advanced plaques. J. Clin. Invest.. 58 (1976) 200. Katz, S.S., The lipids of grossly normal human aortic intima from birth to old age. J. Biol. Chem.. 256 (1981) 12275. Chao, F.. Amende, L.M., Blanchette-Mackie, E.J.. Skarlatos, S.I., Gamble, W., Resau, J.H., Mersner. W.T. and Kruth, H.S., Unesterified cholesterol-rich lipid particles in atherosclerotic lesions of human and rabbit aortas. Am. J. Pathol., 131 (1988) 73. McGrath, L.T. and Elliott, R.J., Lipid distribution in the media of the human thoracic aorta. B&hem. Sot. Trans., 16 (1988) 164. Vasile, E., Simionescu, M. and Simionescu. N., Visualisation of the binding, endocytosis and transcytosis of low density lipoprotein in the arterial endothelium in s~ru. J. Cell Biol., 96 (1983) 1677. Robertson, A.L. and Khairallah, P.A., Arterial endothelial permeability and vascular disease. The “Trap Door” effect. Exp. Mol. Pathol.. 18 (1973) 241. Steinberg, D., Pittman, R.C. and Carew, T.E., Mechanisms involved in the uptake and degradation of low density lipoprotein by the artery wall in viva. Ann. N. Y. Acad. Sci., 454 (1985) 195. Wiklund. 0.. Carew, T.E. and Steinberg, D.. Role of the low density lipoprotein receptor in penetration of low density lipoprotein into the rabbit aortic wall. Arteriosclerosis, 5 (1985) 135. Srinivasan, S.R., Vijayagopal, P., Dalferes, E., Abbate, B., Radhakrishnamurthy, B. and Berensen. G., Low density lipoprotein retention by aortic tissue. Atherosclerosis, 62 (1986) 201. Hoff, H.F. and O’Neil, J., Extracts of human atherosclerotic lesions can modify low density lipoproteins leading to enhanced uptake by macrophages. Atherosclerosis, 70 (1988) 29. Curmi, P.A. and Tedgui, A., Effects of transmural pressure on the transport and distribution of low density lipoproteins in the arterial wall. C. R. Acad. Sci.. 308(6) (1989) 149. Tomkins. R.G., Yarmush, M.L.. Schnitzer. J.J., Colton, C.K.. Smith, K.A. and Stemerman. M.B., Low density lipoprotein transport in blood vessel walls of squirrel monkeys. Am. J. Physiol.. 257 (1989) H452. Smith, E.B. and Staples, E.M.. Distribution of plasma proteins across the human aortic wall. Atherosclerosis. 37 (1980) 579. Smith, E.B.. Plasma macromolecules in Interstitial fluid from normal and atherosclerotic human aorta. Monogr. Atheroscler., 14 (1986) 179. Adams, C.W.M. and Bayliss. O.B.. The relationship be-
220 tween diffuse intimal thickening, medial enzyme failure and intimal lipid deposition in various human arteries. J. Atheroscler. Res., 10 (1969) 327. 43 Smith, E.B. and Staples, E.M., Plasma protein concentrations in interstitial fluids from human aortas. Proc. R. Sot. Lond., B217 (1982) 59.
44 Kramsch, D.M. and Hollander, W., The interaction of serum arterial lipoproteins with elastin of the arterial intima and its role in the lipid accumulation in atherosclerotic plaques. J. Clin. Invest., 52 (1973) 236.
ATHERO
211 Ireland,
Ltd. 0021-9150/91/$03.50
04622
Formation of a lipid gradient across the human aortic wall during ageing and the development of atherosclerosis Lawrence T. McGrath * and Robert J. Elliott Division of Biochemistry
School of Biology and Biochemistry, The Queen’s University of Belfast, Belfast BT9 7BL, Northern Ireland (U.K.) (Received 17 July, 1990) (Revised received 26 November, 1990) (Accepted 19 December, 1990)
Summary
The smooth muscle cell invasion and macrophage stimulation within the intima during prolonged exposure to high blood levels of cholesterol esters contribute to increased production of connective tissue matrix. The thickened intima in turn immobilising more LDL derived lipid from the plasma. With damage to the internal elastic lamellae, from essential hypertension, the absorbed lipid can move down a concentration gradient into the medial tissue. This model was supported by our laboratory finding of a lipid gradient across the aorta wall. The gradient commenced shortly after completion of body growth, when the transmedial gradient became detectable. The slope of the gradient progressively increased during ageing. Association of the lipid medial gradient with the degree of atherosclerotic involvement suggested that the gradient influenced the development of intimal lesions. Accumulation of lipid within the medial tissue may then reduce the inward lipid transfer rate from the intima, promoting increased intimal retention and cause the formation of atherosclerotic plaques from the fat saturated intima.
Key words: Ageing; Aorta; Atherosclerosis; LDL
Cholesterol ester; Connective tissue matrix; Elastic lamina;
Introduction
Atherosclerosis has been considered an example of a disease with a common initiating source,
Present address and correspondence: Dr. L.T. McGrath, Department of Therapeutics and Pharmacology, The Queen’s University of Belfast, Belfast BT9 7BL, Northern Ireland, U.K. Tel. (0232) 245133 ext. 2203; Fax (0232)438346.
the amount of cholesterol and saturated fat ingested [l]. The level of cholesterol circulating in the plasma, mainly in the form of beta-lipoproteins, has been implicated in the atherosclerotic process from 40 epidemiological studies throughout the world [2]. The role of plasma cholesterol in the progression of atherosclerosis has been linked to the composition of atherosclerotic plaques, the cores of which contain up to
212 65% (by dry weight) of lipid, the majority composed of cholesterol and its esters [3]. The hallmark of atherosclerotic plaques is their complexity and the many different processes that can lead to their formation such as: endothelial injury, lipid uptake by macrophages, lipid infiltration, disintegration of lipid filled (foam) cells, extracellular lipid deposition, elastin destruction, smooth muscle cell proliferation, collagen synthesis, proteoglycan accumulation, calcium deposition, necrosis, and plaque thrombosis [4]. The effects of experimentally induced hyperlipidaemia on the lipid composition of arterial tissue has been examined using laboratory animals [5]. These short term studies are unlikely to reflect the patterns found in the much longer development time of atherosclerosis in humans. The flux of lipid entering and leaving the tissue, even when taken as a single variable, must be considered within a tissue which itself is changing metabolically and structurally during ageing [6]. The accumulating lipid would be expected to progressively alter both the rate and the nature of this ageing pattern. If to this already complex process is added the unpredictable occurrence, duration and effects of secondary risk factors, a multitude of pathological variations can be generated [7]. A chronic disease process, such as atherosclerosis, with it’s long development period restricts investigation to human tissue. This requirement has imposed limitations on experimental design and availability of test materials. The present study has utilised aortas obtained during the post mortem examination of patients, aged 17 to 97 years, who had died in hospital. The inference has had to be made that this sample of aortas presented a uniform pattern of lipid progression, such as could be found in the general population during ageing. Physically subdividing the aortic wall into anatomically defined zones has allowed us to make a more detailed examination of the rate of lipid accumulation during ageing. The dynamics of lipid uptake and saturation levels in different zones of the human aorta could then be determined, Materials and methods Aorta collection and dissection Anatomically defined segments
of descending
thoracic aorta, were collected during post-mortem examination from patients who had died in hospital from non-infectious diseases. Samples were wrapped in damp muslin and, if not processed immediately, stored at - 20” C. Frozen samples, when required for analysis, were thawed overnight at 4 o C. The adventitia and any adhering adipose tissue were removed. The segment of aorta selected for analysis was the region that contained the first six intercostal branches (numbered from the aortic arch end). The sample was opened along its frontal longitudinal plane to expose the intimal surface, which was washed with cold 0.9% (w/v) NaCl to remove any adhering blood products. The intercostal holes were located down either side of the centre axis of the opened specimen. These intercostal branches were used as reference points during subsequent tissue mapping and sampling. The flat opened segment of aorta was placed, intimal surface upwards, on a dissection block and the areas covered by each lesion type were mapped onto a clear plastic sheet [8]. Lesion types were then measured as a percentage of the total surface area, graded according to the WHO classification system [9]. The degree of tissue involvement with atherosclerotic lesions, the atherosclerotic index (AI), was expressed as the cumulative percentage of the total surface area covered by atheroma, fibrous plaques and complicated lesions. Macroscopically involved and uninvolved plaque areas of aorta were dissected from the sample. A section of tissue, approximately 1 cm square, was selected to the left of the third intercostal branch. When uninvolved tissue was sampled the area dissected did not come closer than 3 mm from the boundaries of adjacent plaques. If the tissue was extensively ulcerated, smaller lesion free areas were pooled to produce a viable sample for analysis. Quality control of these samples prepared for lipid extraction were screened by histological procedures (see below). The intimal tissue was removed by microdissection and the media was then subdivided into three layers of equal thickness. The dissection of the media into subdivisions was aided by its inherent laminar structure. Although more than three layers could have been obtained the present protocol was found to be practical, quick and reliable. A description of the sampling and analysis procedures developed
213 for this study has been reported [lo]. The medial layers, when separated, were designated Mi (inner media), Mm (middle media), and MO (outer media), the former being that layer closest to the intima. Individual samples were weighed prior to lipid extraction providing a further check that the three medial samples were of similar thickness. Microscopic examination
Small guide sections, cut vertical to the lumen surface of the sampled tissue, were processed through formaldehyde fixation and paraffin embedding to produce 5 pm thick serial sections. Where a number of small samples were pooled, histology was performed on each. Sections were stained by haematoxylin-eosin and by Weigert’s elastin (Hart’s modification) techniques. The prepared tissue sections were examined to confirm; (a) that samples, from lesion free areas had an intact intimal surface (the endothelial layer was present and extended lipid vacules were absent from within the tissue); (b) that intimal removal from the inner medial sample was complete and had not removed the internal elastic lamina. The haematoxylin-eosin treated sections were most appropriate for evaluation of the intimal material and the elastin stained sections were used to confirm the retention of the internal elastic lamina by the inner medial samples. Approximately 15% of samples had to be excluded from the study by failing to comply with the requirements described above. The tissue sections prepared for quality control were subsequently used to measure intimal and medial thickness. An Orthoplan microscope, fitted with a micrometer eyepiece was used at an overall magnification of x 100. Each tissue layer was measured in triplicate and the mean thickness determined. The thickness data (see Results) relates to fixed, paraffin embedded tissue, adopting the processing protocols used by previous investigators [11,12].
acetone, then 3 times with chloroform/methanol (2 : 1, v/v), and finally with acetone. Extractions were carried out at 4°C under nitrogen in the dark. The five lipid extracts from each tissue sample were pooled and stored at - 20” C under nitrogen to await analysis. The delipidated tissue was vacuum dried to a constant weight and recorded. Total cholesterol was analysed by the method of Watson [14], and unesterified cholesterol by the method of Sale et al. [15]. Esterified cholesterol was calculated from the total and unesterified cholesterol. Triacylglycerols were determined as described by Wahlefield [16], and phospholipid by the method of Anderson and Davis [17]. Expression of results
The total lipid content of each layer of tissue was calculated as the sum of esterified cholesterol, unesterified cholesterol, triacylglycerol and phospholipid and expressed as ug lipid/mg dry defatted tissue (Table 1). Individual lipid components were expressed as a percentage of the total lipid content (Table 2). The presence of inorganic material did not significantly affect the data, except with grossly atherosclerotic arteries [18]. Assay performance analysis produced day to day variation for total cholesterol of 6.18, unesterified cholesterol 6.78, triacylglycerol 5.9% and phospholipid 5.8% [19]. Statistical significance between 20 year age groups for each tissue layer was analysed. As normal distribution for each group could not be assumed non-parametric methods were used [20]. Analyses between age groups were carried out using Wilcoxon’s rank sum test for independent groups. Analyses within age groups were carried out using Wilcoxon’s signed rank test for matched groups (Table 1). Results Morphometric analysis
Lipid extraction and analysis
Lipids were extracted from the intact tissue samples by a modification of the procedure of Folch [13]. Samples were placed into 5-ml glassstoppered tubes and extracted successively (20 ml per g wet tissue weight) for 24 h periods, with
Intimal thickness was found to progressively increase during ageing, from 20 pm in the 2nd decade to 300 pm by the 10th decade. There was a strong correlation between intimal thickness and age (r = 0.762). The medial layer thickness from the same aortae displayed large variations between
214 samples of similar age, the correlation was r = 0.256 (Fig. 1).
with age 16001
/
Lipid content and composition during ageing
The lipid content and composition was examined after grouping the samples into five 20-y age groups (Table 1). The total lipid content of the intima progressively increased with age until the ninth decade (r = 0.724). Medial tissue lipid also increased with ageing. The lipid content and ageing in the inner, middle and outer media were significantly correlated (r = 0.703,0.754 and 0.747, respectively). The lipid content of the three medial layers was highly correlated with the degree of atherosclerotic involvement of the aorta (r = 0.740, 0.526 and 0.523 for inner, middle and outer media). The predominant lesion type, as was to be expected, changed with age. There was poor correlation (r = -C0.200) between the lipid content of the three medial layers and any particular lesion type. It is also somewhat difficult to avoid the conclusion that a ‘complicated lesion’ was not at some stage of its development an atheroma and/or fibrous plaque. The use of the AI was an attempt to avoid this complication. The lipid in the medial layers was also compared with the overlying intimal thickness (r = 0.509, 0.663 and 0.698 for inner, middle and outer media). Medial lipid can also be expressed as a percentage of intimal lipid content from the data in Table 1. In the youngest age group (c 20 y) all three medial tissue zones contained approximately 50% of the intimal tissue concentration. With increasing age the inner medial tissue lipid, as a percentage of age matched intimal lipid content, increased gradually to reach approximately 75% of the intimal value, as found in the oldest age group samples (81-97 y). In contrast the deepest medial zone (MO) where, although the quantity of lipid accumulated with age increased, the lipid as a percentage of the intimal tissue steadily decreased with age (from 55% in the < 20 y group to 34% in the 81-97 y age group). As medial lipid increased it was associated with a rise in the relative proportion of esterified cholesterol and as a consequence with dilution of the phospholipid content (Table 2). The ratio of unesterified cholesterol to phospholipid decreased with increasing lipid content.
,
,
,
,
.1600
lY /’ ,
;
a t
B
1
.u 5
000-
.o P
80 years) in our sample had a lipid profile similar to that which was present in the middle medial zone of the 61-80 year old samples. This middle medial age group in turn matched the inner medial zone lipid values found in the 41-60 year age group. Plasma derived LDL can be transported through the endothelium by selective receptor mediated transport systems [32]. A number of workers have proposed that non-specific fluid endocytosis can account for all of the LDL transport into the artery wall [33-351. Srinivasan [36] has demonstrated that more than 70% of LDL enters the aorta by a receptor independent pathway. Intimal and medial cellular modification of the LDL can occur before its partial return to the plasma or transfer inwards towards the media [37]. The findings in this study suggest that interaction
218 of insuded lipid with the aortic matrix was followed by increasing retention. The lipid that accumulated was correlated with age, r = 0.703 (P < O.OOl), and to a lesser degree with atherosclerosis, r =‘0.608 (P -c 0.001). Curmi [38] has shown the establishment of a LDL gradient across the aorta wall in rabbits. This gradient was highly dependant upon pressure. The gradient only extended into the inner half of the media. In short term studies in squirrel monkeys Tompkins [39] has shown that a steep LDL gradient is established across the intima but does not extend into the media. Smith and Staples [40] demonstrated that penetration of LDL into the media depended upon fragmentation of the internal elastic lamina. During ageing, with associated hypertension, partial loss of the integrity of the internal elastic lamina would occur [12]. The lamina has been proposed as an effective inward barrier to LDL migration [41]. Where fragmentation of the internal lamina had occurred lipid entry into the medial tissue would then be possible [42]. Smith and Staples [43] have suggested that large molecules, such as LDL, would be retained within the intima by both molecular sieving and an intact internal elastic lamina. In young adults the LDL content of medial tissue immediately adjacent to the elastic lamina was found to be 0.3% of that found in the intima [12]. Where the elastic lamina had been extensively fragmented, as in older tissue, there was a 25fold increase of non-membrane lipid in the medial tissue [40]. Kramsch and Hollander [44] found that LDL interacted with the aortic elastin, resulting in the formation of an elastinlipid complex. The residual protein component of the LDL, became immobilised to form a fibrin type residue. The gradients of total lipid, free and esterified cholesterol found in the present study were consistent with lipid derived from LDL that had migrated from the intima. The establishment of the transmedial lipid gradient may well precede the development of an atherosclerotic intimal lesion. Accumulation of lipid in the medial tissue would be expected to significantly reduce the rate of lipid transfer from the intima into the media, eventually causing lipid saturation of the intima. Local areas of the intimal tissue matrix could then form lipid plaque lesions.
Atheroma formation and associated lesions represent the outcomes of a long and complex interaction between the blood derived lipid and the arterial tissue. The present work has attempted to resolve part of this process. We have found that the extracellular matrix of the arterial tissue accumulated large quantities of cholesterol esters. The thin intima readily became saturated, promoting the formation of fat filled plaques. Diffusion of cholesterol esters also occurred inwards to reach even the deepest layers of medial tissue. The results presented here have revealed the formation of a lipid gradient across the full width of the largest artery of the body, the conclusion of long term exposure of the aorta to high levels of circulating lipids. Interaction between the circulating milieu and the artery wall has its effects not just upon the intima and adjacent media but penetrates into the deepest medial layers of the thoracic aorta.
Acknowledgements We would like to thank Dr. J.D. Biggart and Dr. G.W. Moore for the samples of aortic tissue obtained during post-mortem examinations and Mr. J. Orchin for producing the paraffin embedded tissue sections. We also wish to express our thanks to Professor J. Shepherd for a useful discussion of this work, prior to the preparation of this paper.
References Kuller, L.H., Epidemiology of coronary heart disease. In: E.G. Perkins and W.J. Visek (Eds.), Dietary Fats and Health, Champaign, IL, American Oil Chemists’ Society, 1983, p. 466. Intersociety Commission for Heart Disease Resources Atherosclerosis Group, Optimal resources for primary prevention of atherosclerosis (Abstract). Circulation, 70 (1984) 157A. Panganamala, R.V., Geer, J.C., Sharma, H.M. and Comwell, D.G., The gross and histologic appearance and the lipid composition of normal intima and lesions from human coronary arteries and aorta. Atherosclerosis, 20 (1974) 93. Constantinides, P., Arteries and arterioles. In: Ultrastructural Pathobiology, Elsevier, New York, 1984, p. 78. Reitman, J.S., Mahley, R.W. and Fry, D.L., Yucatan minia-
219
6
7 8
9 10
11 12 13 14 15
16
17
1X
19
20 21
22 23
ture swine as a model for diet-induced atherosclerosis. Atherosclerosis, 43 (1982) 119. Merrilees, M.J., Campbell, J.H., Spanidis, E. and Campbell, G.R., Glycosaminoglycans synthesis by smooth muscle cells of differing phenotype and their response to endothelial cell conditioned medium. Atherosclerosis, 81 (1990) 245. McGill, H.C., The pathogenesis of atherosclerosis. Clin. Chem., 34 (1988) B33. Mitchell, J.R.A. and Cranston, WI., A simple method for the quantitative assessment of aortic disease. J. Atheroscler. Res., 5 (1965) 135. WHO, Classification of atherosclerotic lesions. WHO Tech. Rep. Ser., 143 (1958) 1. McGrath, L.T., Biggart, D.J. and Elliott, R.J., Lipid composition in the intimal and medial zones of human aorta. Biochem. Sot. Trans., 15 (1987) 254. Cook. T.A. and Yates, P.O., A critical survey of techniques for arterial mensuration. J. Pathol., 108 (1972) 119. Semba. I., A histometrical analysis of age changes in the human lingual artery. Arch.Oral Biol., 34(7) (1989) 483. Radin. N.S., Preparation of lipid extracts. Methods Enzymol.. 14 (1969) 245. Watson, D., A simple method for the determination of serum cholesterol. Clin. Chim. Acta, 5 (1960) 637. Sale, F.O., Marchesini, S.. Fishman, P.H. and Berra, B., A sensitive enzymatic assay for determination of cholesterol in lipid extracts. Anal. Biochem., 142 (1984) 347. Wahlefield, A.W., Triglycerides. Determination after enzymatic hydrolysis. In: H.U. Bergmeyer (Ed.), Methods of Enzymatic Analysis, Vol. 4, Verlag-Chemie, Weinheim and Academic Press, New York, 1974, p.1831. Anderson, R.L. and Davis, S., An organic phosphorous assay which avoids the use of hazardous perchloric acid. Clin. Chim. Acta, 121 (1982) 111. Yla-Herttuala, S., Sumuvori, H., Karkola, K., Mottonen, M. and Nikkari, T., Comparison of different ways of presenting the results of biochemical analyses of human carotid arteries. Atherosclerosis, 61 (1986) 77. McGrath, L.T. and Elliott, R.J., Lipid analysis and fatty acid profiles of individual arterial atherosclerotic plaques. Anal. Biochem., 187 (1990) 273. Bland, M., An Introduction to Medical Statistics, Oxford University Press, Oxford, 1989, p.216. Smith, E.B., Evans, P.H. and Downham, M.D., Lipid in the aortic intima. The correlation of morphological and chemical characteristics. J. Atheroscler. Res., 7 (1967) 171. Smith, E.B., Molecular interactions in human atherossclerotic plaques. Am. J. Pathol., 86 (1977) 665. Milch, R.A., Matrix properties of the aging arterial wall. In: W.G. Analyan and D.C. Sabiston (Eds.), Monographs in the Surgical Sciences. Vol. 2, Williams and Wilkins Co..
Baltimore, MD, 1965. p. 261. 24 Ross, R., The pathogenesis of atherosclerosis-an update. New Engl. J. Med., 314 (1986) 488. 25 Lundberg, B.. Chemical composition and physical state of lipid deposits in atherosclerosis. Atherosclerosis, 56 (1985) 93. 26 Smith.
E.B.. The relationship
between
plasma
and
tissue
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
lipids in human atherosclerosis. Adv. Lipid Res., 12 (1974) 1. Smith, E.B., The influence of age and atherosclerosis on the chemistry of aortic intima. J. Atheroscler. Res., 5 (1965) 224. Katz, S.S., Shipley, G.G. and Small, D.M., Physical chemistry of the lipids of human atherosclerotic lesions. Demonstration of a lesion intermediate between fatty streaks and advanced plaques. J. Clin. Invest.. 58 (1976) 200. Katz, S.S., The lipids of grossly normal human aortic intima from birth to old age. J. Biol. Chem.. 256 (1981) 12275. Chao, F.. Amende, L.M., Blanchette-Mackie, E.J.. Skarlatos, S.I., Gamble, W., Resau, J.H., Mersner. W.T. and Kruth, H.S., Unesterified cholesterol-rich lipid particles in atherosclerotic lesions of human and rabbit aortas. Am. J. Pathol., 131 (1988) 73. McGrath, L.T. and Elliott, R.J., Lipid distribution in the media of the human thoracic aorta. B&hem. Sot. Trans., 16 (1988) 164. Vasile, E., Simionescu, M. and Simionescu. N., Visualisation of the binding, endocytosis and transcytosis of low density lipoprotein in the arterial endothelium in s~ru. J. Cell Biol., 96 (1983) 1677. Robertson, A.L. and Khairallah, P.A., Arterial endothelial permeability and vascular disease. The “Trap Door” effect. Exp. Mol. Pathol.. 18 (1973) 241. Steinberg, D., Pittman, R.C. and Carew, T.E., Mechanisms involved in the uptake and degradation of low density lipoprotein by the artery wall in viva. Ann. N. Y. Acad. Sci., 454 (1985) 195. Wiklund. 0.. Carew, T.E. and Steinberg, D.. Role of the low density lipoprotein receptor in penetration of low density lipoprotein into the rabbit aortic wall. Arteriosclerosis, 5 (1985) 135. Srinivasan, S.R., Vijayagopal, P., Dalferes, E., Abbate, B., Radhakrishnamurthy, B. and Berensen. G., Low density lipoprotein retention by aortic tissue. Atherosclerosis, 62 (1986) 201. Hoff, H.F. and O’Neil, J., Extracts of human atherosclerotic lesions can modify low density lipoproteins leading to enhanced uptake by macrophages. Atherosclerosis, 70 (1988) 29. Curmi, P.A. and Tedgui, A., Effects of transmural pressure on the transport and distribution of low density lipoproteins in the arterial wall. C. R. Acad. Sci.. 308(6) (1989) 149. Tomkins. R.G., Yarmush, M.L.. Schnitzer. J.J., Colton, C.K.. Smith, K.A. and Stemerman. M.B., Low density lipoprotein transport in blood vessel walls of squirrel monkeys. Am. J. Physiol.. 257 (1989) H452. Smith, E.B. and Staples, E.M.. Distribution of plasma proteins across the human aortic wall. Atherosclerosis. 37 (1980) 579. Smith, E.B.. Plasma macromolecules in Interstitial fluid from normal and atherosclerotic human aorta. Monogr. Atheroscler., 14 (1986) 179. Adams, C.W.M. and Bayliss. O.B.. The relationship be-
220 tween diffuse intimal thickening, medial enzyme failure and intimal lipid deposition in various human arteries. J. Atheroscler. Res., 10 (1969) 327. 43 Smith, E.B. and Staples, E.M., Plasma protein concentrations in interstitial fluids from human aortas. Proc. R. Sot. Lond., B217 (1982) 59.
44 Kramsch, D.M. and Hollander, W., The interaction of serum arterial lipoproteins with elastin of the arterial intima and its role in the lipid accumulation in atherosclerotic plaques. J. Clin. Invest., 52 (1973) 236.
