Lipoprotein (a) in the arterial wall

U. Beisiegel, A. Niendorf, K. Wolf, T. Reblin and M. Rath

EUROPEAN HEART JOURNAL (1990) 11 (SUPPL. E), 174-183

Introduction

Lp(a) was first demonstrated by Berg and his associates (1). They described Lp(a) as a human lipoprotein polymorphism. Characterization of Lp(a) revealed that it is very similar to LDL in lipid composition and presence of apo B-100. In contrast to LDL it contains an additional glycoprotein, designated apo (a) (2,3). Apo (a) is linked to apo B by disulfide bridges.

Lp(a) is found in all human beings, but in variable amounts, which are genetically determined. Size isoforms have been described for apo (a) (4,5) which were categorized by Utermann et al.(6) into five homozygous phenotypes and the respective double-band heterozygous forms. A highly significant association of the phenotypes with the serum level of Lp(a) could be demonstrated (7,8). Recently, striking homology between human apo (a) and plasminogen was demonstrated in both amino acid (9,10) and c-DNA (11) sequences. This unexpected homology between an apolipoprotein and an important protein of the coagulation system led Goldstein and Brown to speculate that "these dramatic findings may provide the long-sought link between lipoproteins and the clotting system" (12).

In a series of epidemiological studies, a positive correlation of high serum Lp(a) levels with CHD has been demonstrated (13-15). The pathomechanism by which Lp(a) contributes to the development of atherosclerotic plaques has not yet been studied extensively. As early as 1958, human arterial wall was analyzed for its lipoprotein content (16). This and alter studies mainly concentrated on LDL in the vessel tissue. LDL was measured in arterial wall and compared with serum lipid levels (17) and apo B was quantitated in normal intima compared with fibrous plaques (18). Moreover LDL and lipoprotein-like particles were extracted from human aorta and analyzed by physical, chemical and immunological means (19,20).

The aim of the present study was to analyze human arterial wall tissue for the presence of Lp(a). In biochemical studies we measured Lp(a) in arterial wall and extracted lipoproteins from the tissue (21). With immunohistochemistry we tried to localize apo (a) in different areas of arterial wall tissue (22) and compared the pattern with the deposition apo B and fibrin in consecutive tissue sections (22,23).

Materials and method

Fasting blood samples were taken from the patients after admission to the hospital one or two days before the bypass operation. It was not always possible to obtain fasting samples from blood donors, thus the difference in triglycerides and VLDL might be even more distinct in this group.

Cholesterol and triglycerides, as well as HDL cholesterol was determined with enzymatic test kits from Boehringer Mannheim. VLDL cholesterol concentration was calculated from the triglyceride levels measured (triglycerides: 5) and LDL cholesterol was calculated by the Friedewald formula (LDL cholesterol = total cholesterol Ð VDL cholesterol Ð HDL cholesterol).

Apo B was determined by turbidometric method from Behring Diagnostika (Marburg, F.R.G.) using standards and control sera from the same company. Lp(a) was measured by radioimmunoassay from Pharmacia/LKB (Freiburg, F.R.G.) using standards and control sera from Pharmacia/LKB.

We used as tissue samples the biopsies routinely taken during the aortocoronary bypass operation where the vein graft is attached to the ascending aorta. The biopsies were around 10-30 mg and by histological screening they showed different grades of intimal thickening than control tissue. No severe plaque areas or complicated lesions were examined. Venous samples were taken from the vena sephena magna, which served as the bypass graft.

Table 1.

Biochemical analysis of the tissue samples has been described elsewhere[21]. Apo(a) and apo B were determined by a special ELISA system, which allowed differentiation of apo B linked to apo (a) and free apo B. Standard immunohistochemical procedures were followed (22).

Anti-apo(a) monoclonal antibody was prepared in our laboratory (21) and polyclonal anti-apo B antiserum was raised in a rabbit. For detection, the ABC-method (avidin-biotin-complex) was used.

For morphometric quantitation, performed by two independent investigators, a staining score was deduced from the formula : [staining score = (percentage of stained area with regard to the lesion area/100) x (intensity of stained area) x (percentage of lesion with regard to total section area/100)] This gave a value between 0 and 12.

Western blot analyses of the tissue samples were performed with 6% SDS-polyacrylamide gel electrophoresis. From the aorta both the intima and media were dissected and solubilized together (without the adventitia). Following electrophoresis, the proteins were transferred to nitrocellulose. Apo (a) was detected with monoclonal antibodies on the nitrocellulose and the bands stained with a second antibody conjugated to horseradish peroxidase. [sitemap] Table 2

Results

To confirm the described evidence that Lp(a) serum levels above 25 mg dl-1 are more frequent in patients with CHD than in normal controls we compared 300 healthy blood donors with 235 patients who underwent coronary bypass operations (Rath et al., manuscript in preparation). For both groups, the various lipoprotein parameters are shown in Table 1. All lipid parameters except HDL (P<0.0045) were highly significantly different in the two groups (P<0.0001. Lp(a) is not normally distributed, but shows a highly skewed distribution. Therefore, in addition to the mean, the percentage of persons with Lp(a) >25 mg dl-1 is given for both groups. The means differ with P<0.0004 and in the CHD group 1.6 times more persons have Lp(a) > 25 mg dl-1 (45% vs 27% in the controls).

Table2.

From 107 of the 235 bypass patients, we were able to obtain an aortic biopsy taken routinely during operation. In these tissue samples, we measured lipids, total protein and apoproteins as described by Rath et al. (21). Table 2 shows the comparison of tissue values for patients with serum Lp(a) levels>and< 25mg dl-1. A clear difference was seen in apo(a) and apo(a)-linked apo B according to serum Lp(a) level, while free apo B, cholesterol, triglycerides and total protein showed no significant difference.

In a smaller group of patients (n=32) we obtained venous samples in addition to aortic biopsies. Comparing the two different types of vessels the following results were obtained : in the venous tissue lower values for cholesterol (1.4 compared with 4.3 mg.g-1WW) and apo (a) (2.4 compared with 11.1m g.g-1WW) were found. These values are comparable with the data on normal saphenous veins from Cushing et al. (24). They describe apo (a) in the venous tissue with <2 ng.mg-1WW and apo B with 3.3 ng.mg-1WW and for apo B 70 ng.mg-1WW, demonstrating a net accumulation of these apoproteins in the veins from the time of their grafting in the arterial bed (24).

Comparing areas of the arterial wall with no visible plaque with various stages of atherosclerotic plaques, we observed a more than twofold increase using semquantitative morphological measurements (Fig 10) (22). Biochemical data from autopsy tissue confirm this accumulation of Lp(a) in plaque areas. In a small number of cases, we analyzed pot-mortem arterial wall with plaque areas of <50% or >50%. The means differed by 14% (38m g.g-1WW versus 44m g.g-1WW).

Figure 1.

We tried to determine, by immunohistochemical methods, where in the arterial wall the Lp(a) accumulates. Slices from post-mortem aortic and coronary tissue from 74 patients (aged 1-98 years) were stained with monoclonal anti-apo(a) (22). Fig. 2a demonstrates apo (a) in the thickened intima of an early plaque formation. In Fig. 2b, a fibrous plaque is shown with a massive accumulation of apo (a). In both examples the apo(a) is located extracellularly in the intima, which is more clearly demonstrated in Fig. 3, where differential interference contrast microscopy was used. The higher magnification shows that apo(a) is extracellularly associated with fibrous structures; no intracellular staining can be seen. In few cases, however, we could demonstrate apo(a) intracellularly in foam cells. It can also be shown by biochemical means that apo (a) accumulates in the intima rather than in other layers of the vessel wall (Fig.4). In several aortic tissue samples the intima and media were dissected and separately analyzed. Two apo(a) bands are only detectable in the intima. They represent the two apo(a) isoforms which we had also determined in the patientsÕ serum. It is important to note that we found apo (a) as intact protein even in autopsy tissue; no significant proteolytic degradation seemed to affect this high seemed to affect this high molecular weight protein.

Figure 2.

From autopsy tissue, adjacent slices of the aorta were stained with monoclonal anti-apo(a) or polyclonal anti-apo B to investigate a possible co-localization of the two protein components of Lp(a). In all cases we found a strict co-localization of apo(a) and apo B (22). A representative example is given in Fig. 5. We found only very few areas where apo B alone was stained.

Considering the results of SDS-PAGE and co-localization studies, we wanted to determine whether Lp(a)-like lipoproteins could be extracted from the arterial wall. We had to use post-mortem tissue for this experiment since we did not obtain enough fresh tissue. Nevertheless, we were able to isolate particles within the density range of Lp(a) which contained lipids as well as apo(a) and apo B. In addition, we detected apo B in the density range of LDL and a comparable amount of apo (a) at the bottom of the gradient (data not shown) (21). Our conclusion from these data are that Lp(a) can indeed be extracted from arterial wall. We also assume that the apo(a) at the bottom of the gradient was split from the particles which appear at LDL density in post-mortem samples. Further extraction experiments with fresh tissue should clarify this.

In our latest studies, we used adjacent slices of pot-mortem aortic tissue to analyze further the areas in which Lp(a) is detected in the intima. With anti-fibrinogen antibodies (also recognizing fibrin and other breakdown products of fibrinogen) we looked for a possib le co-localization of apo(a) and fibrin deposition. Fibrin and apo(a) were found in the same areas of the intima (Fig.6), and this co-localization indicate that Lp(a) might be associated with fibrin in the arterial wall (Wolf, K. et al. manuscript in preparation).

Figure 3.

Figure 4.

Discussion

Before the main discussion it should be mentioned that we found, in contrast to most earlier studies, no highly significant difference in HDL cholesterol levels between the two groups investigated (blood donors and bypass patients). This disparity cannot be explained by the sex distribution in the two groups since they were comparable (percentage females = 21% in the controls and 24% in the CHD group). The other lipoprotein parameters we found increased in the CHD group, which is in good agreement with most former reports.

Figure 5.

Studies over the past 30 years on lipoproteins and the pathobiology of the arterial wall have concentrated on LDL as an important factor in plaque development. LDL contributes to the foam cell formation mainly in its oxidized form. Today it is necessary, besides LDL and oxidized LDL, to consider LP (a) in studies on lipoproteins in arterial wall.

An accumulation of the two protein constituents of Lp(a), apo (a) and apo B, in the intima of the arterial wall was demonstrated by biochemical quantification, and Western blotting demonstrated that apo (a) is in the intima of the arterial wall and is still intact as a high molecular weight band. Immunohistochemical studies revealed a mainly extracellular co-localization of both proteins. These experiments indicate an accumulation of the intact Lp(a) particle in the intima; Lp(a)-like lipoproteins could also be extracted from arterial wall tissue.

Figure 6.

We found very few areas where apo B alone was stained by immunohistochemistry. This could either mean that there is no accumulation of intact LDL in the analyzed samples or the LDL accumulates in exactly the same areas as Lp(a). The measurement of free apo B in the fresh samples in addition to a comparable amount of apo(a) and apo(a)-linked apo B would indicated some deposit of intact LDL-apo B. In this connection it has, however, to be questioned whether in former experiments where lipoprotein were extracted from arterial wall, the apo(a) might have been overlooked and Lp(a) had been isolated in addition to or instead of LDL.

It can be concluded from these data that most of the Lp(a) is not taken up by macrophages and degraded intracellularly, as is LDL. Lp(a) rather seems to invade the arterial wall very early in plaque development and becomes bound to extracellular structures, where it accumulates over the years without being degraded. A possible mechanism for this accumulation is the interaction of Lp(a) with glycosaminoglycans, as has been described by Dahlen et al. (25) and Bihari-Varga et al. (26) Salonen et al. (27) recently described apo(a) binding to fibronectin, a major substance in the basal membrane of the arterial wall.

Another line of thought on the pathophysiology of Lp(a) was introduced by the demonstration of homology with plasminogen. Competitively, apo(a) might bind to fibrin and thereby inhibit the action of plasminogen. Recently, evidence for such a mechanism was reported (28). If such inhibition occurs in vivo, fibrinolysis could be affected and the thrombi might remain on the endothelial surface. Our study supports this hypothesis by the co-localization of apo(a) and fibrin in the arterial wall. This might be due to the thrombi remaining on the endothelial surfaces since in presence of Lp(a) the fibrin and the apo(a) will eventually become part of the plaque and present the described co-localization pattern.

Following the same path, it was demonstrated (29,30) that Lp(a) can bind to the plasminogen receptor of endothelial cells. Lp(a) possibly enters the intima via this pathway. Moreover, apo(a) is reported to have proteolytic activity, as does plasminogen, first demonstrated versus fibronectin (27).

These data show that Lp(a) contributes to plaque development by extracellular accumulation and only a minor amount is taken up in foam cells. This indicates that the role Lp(a) plays in plaque formation differs from that of LDL. More careful studies are necessary to evaluate the present data and finally elucidate the role of Lp(a) in atherosclerosis.

References

1.Berg K. A new serum type system in man Ð the Lp system. Acta Pathologica 1963; 59: 369-82

2.Ehnholm C, Garoff H, Renkonen O, Simons K. Protein and carbohydrate composition of Lp(a) lipoprotein from human plasma. Biochemistry 1972; 11: 3229-32

3.Utermann G, Weber W. Protein composition of Lp(a) lipoprotein from human plasma. FEBS Letters 1983; 154: 357-61

4.Fless GM, Rolih CA, Scanu AM. Heterogeneity of human plasma lipoprotein(a). Journal of Biological Chemistry 1984; 259: 11470-11478.

5.Gaubatz JW, Chari MV, Nava ML, Guyton JR, Morrisett JD. Isolation and characterization of two major apoproteins in human lipoprotein (a). Journal of Lipid Research 1987; 28: 69-79

6.Utermann G, Menzel HJ, Kraft HG, Duba HC, Kemmler HG, Seitz C. Lp(a) glycoprotein pheno-types. Journal of Clinical Investigation 1987; 80: 458-65.

7.Utermann G, Kraft HG, Menzel HJ, Hopferwieser T, Seitz C. Genetics of the quantitative Lp(a) lipoprotein trait. Human Genetics 1988; 78: 41-6.

8.Utermann G, Duba C, Menzel HJ. Genetics of the quantitative Lp(a) lipoprotein trait. II. Human Genetics 1988; 78: 47-50.

9.Eaton DL, Fless GM, Kohr WJ et al. Partial amino acid sequence of apolipoprotein (a) shows that it is homologous to plasminogen. Proceedings of the National Academy of Sciences 1987; 84: 3224-8

10.Kratzin H, Armstrong VW, Niehaus M, Hilschmann N, Seidel, D. Structural relationship of an apolipoprotein (a) phenotype (570 kDa) to plasminogen: homologous Kringle domains are linked by carbohydrate-rich regions. Biolog. Chem. Hoppe-Seyler 1987; 368: 1533-44.

11.McLean JW, Tomlinson JE, Kuang W et al. CDNA sequence of human apolipoprotein(a) is homologous to plasminogen. Nature 1987; 300: 132-7.

12.Brown MS, Goldstein JL. Teaching old dogmas new tricks. Nature 1987; 330: 113-4

13.Schriewer H, Assmann G, Sandkamp M. The relationship of lipoprotein (a) [Lp(a)] to risk factors of coronary heart disease. Journal of Clinical Chemistry and Clinical Biochemistry 1984; 22: 591-6

14.Dahlen GH, Guyton JR, Attar M, Farmer JA, Kautz JA, Gotto AM. Association of levels of lipoprotein Lp(a), plasma lipids, and other lipoproteins with coronary artery disease documented by angiography. Circulation 1986; 74: 758-65

15.Armstrong VW, Cremer P, Eberle E et al. The association between serum Lp(a) concentrations and angiographically assessed coronary atherosclerosis. Dependence of serum LDL levels. Atherosclerosis 1986; 62: 249-57.

16.Ott H, Lohss F, Gergely J. Der Nachweis von Serum lipoproteinen in der Aortenintima. Klinische Wochschrift 1958; 8:383-4

17.Smith EB, Slater RS. Relationship between low-density lipoprotein in aortic intima and serum-lipid levels. Lancet 1972; 1: 463-9

18.Hoff HF, Heidemann CL, Gaubatz JW, Titus JL, Gotto AM. Quantitation of Apo B in human aortic fatty streaks. Atherosclerosis 1978; 30: 263-72

19.Hoff HF, Bradley WA, Heidemann CL, Gaubatz JW, Karagas MD, Gotto AM. Characterization of low-density lipoprotein-like particles in the human aorta from grossly normal and atherosclerotic lesions. Biochimica et Biophysica Acta 1979; 573: 361-74

20.Smith EB, Slater R. The chemical and immunological assay of low-density lipoproteins extracted from human aortic intima. Atherosclerosis 1970; 11: 417-38

21.Rath M, Niendorf A, Reblin T, Dietel M, Krebber Hj, Beisiegel U. Detection and quantification of lipoprotein (a) in the arterial wall of 107 coronary bypass patients. Atherosclerosis 1989; 9:579-92

22.Niendorf A, Rath M, Wolf K et al. Morphological detection and quantification of lipoprotein (a)-deposition in atheromatous lesions of human aorta and coronary arteries. VirchowÕs Archiv A, 1990; in press.

23.Wolf K, Rath M, Niendorf A, Beisiegel U, Dietel M. Morhological co-localization of Apo(a) and fibrin (ogen) in human coronary atheroma. Circulation 1989; 80 (Suppl II): 522 (abstr 2078).

24.Cushing GL, Gaubatz JW, Nava ML et al. Quantitation and localization of apolipoproteins(a) and B in coronary artery bypass vein grafts resected at re-operation. Atherosclerosis 1989; 9: 593-603.

25.Dahlen G, Ericson C, Berg K. In vitro studies on the interaction of isolated Lp(a) lipoprotein and other serum lipoproteins with glycosaminoglycans. Clinical Genetics 1978; 14: 36-42

26.Bihari-Varga M, Gruber E, Rotheneger M, Zechner R, Kostner GM. Interaction of lipoprotein(a) and low density lipoprotein with glycosaminoglycans from human aorta. Atherosclerosis 1988; 8: 851-7.

27.Salonen EM, Jauhiainen M, Valeri A, Ehnholm C. Lipoprotein(a) has serine protease activity that can cleave fibronectin. In: X International Symposium on Drugs affecting metabolism, 1989; Abstract Book: 24 (308).

28.Harpel PC, Gordon BR, Parker TS. Plasmin catalyzes binding of lipoprotein(a) to immobolized fibrinogen and fibrin. Proceedings of the National Academy of Siences USA 1989;86: 3847-51.

29.Miles LA, Fless GM. Levin EG, Scanu AM, Plow EF. A potential basis for the thrombotic risk associated with lipoprotein(a). Nature 1989; 339: 301-2.

30.Hajjar KA, Gavish D, Breslow JL, Nachmann RL. Lipoprotein (a) modulation of endothelial cell surface fibrinolysis and its potential role in atherosclerosis. Nature 1989; 339: 303-5