Transfer of Lipoprotein(a) and LDL Into Aortic Intima in Normal and in Cholesterol-Fed Rabbits
Abstract To study the relative atherogenic potential of lipoprotein(a) [Lp(a)], the transfer of Lp(a) and LDL into the arterial wall was compared in normal rabbits, cholesterol-fed rabbits, and normal rabbits in which the plasma concentration of Lp(a) before injection of labeled lipoproteins was increased by an intravenous mass injection of 45 mg Lp(a). Aorta was removed either 60 minutes or 180 minutes after intravenous injection of a mixed preparation of human 125I-Lp(a) and 131I-LDL; intimal clearance was calculated as radioactivity in aortic intima/inner media divided by the average concentration of the appropriate radioactivity in plasma and by the length of the exposure time. The intimal clearance of labeled Lp(a) and LDL in the aortic arch after 60 minutes of exposure was 87±9 and 47±7 nL · cm−2 · h−1 (n=9) in normal rabbits and 82±14 and 62±10 nL · cm−2 · h−1 (n=10) in cholesterol-fed rabbits; after 180 minutes of exposure, the intimal clearance of labeled Lp(a) and LDL was 62±14 and 84±21 nL · cm−2 · h−1 (n=6) and 30±6 and 47±12 nL · cm−2 · h−1 (n=4) in cholesterol-fed and Lp(a)-injected rabbits, respectively. Linear regression analysis showed positive associations between intimal clearance of the two lipoproteins in all four groups of rabbits in the aortic arch, the thoracic aorta, and the abdominal aorta. Aortic immunoreactivity of human apolipoprotein(a) was detected in the intima in association with fatty streak lesions, predominantly within the cytoplasm of foam cells. These results suggest that Lp(a) is transferred into the aortic intima by a mechanism similar to that for LDL and that Lp(a) can be taken up by intimal foam cells; however, Lp(a) and LDL may be metabolized differently upon entrance into the arterial wall.
Reprint requests to Dr Lars Bo Nielsen, Department of Clinical Biochemistry 3011, Rigshospitalet, Blegdamsvej 9, DK-2100 Copenhagen, Denmark.
- Received January 9, 1995.
- Accepted June 14, 1995.
Lp(a) was first discovered in 1963.1 Since then several case-control studies have demonstrated that high plasma levels of Lp(a) are associated with an increased risk for vascular disease.2 3 The mechanism for this association remains to be delineated.
The Lp(a) particle resembles the major atherogenic lipoprotein, LDL. Both lipoproteins contain the large apo B protein, but Lp(a) has an additional glycoprotein, apo(a), attached to the apo B by a single disulfide bridge.4 5 Apo(a) has an apparent molecular weight of 300 000 to 700 000. The smaller isoforms of apo(a) are associated with the highest levels of Lp(a) in plasma and the highest risk for atherosclerotic disease.6 7
Apo(a) contains protein sequences with striking homology to protein sequences of the fibrinolytic proenzyme plasminogen.8 Hence, Lp(a) could potentially be involved in ischemic heart disease via inhibition of fibrinolysis,9 10 11 but a direct promotion of atherosclerosis via the similarity between Lp(a) and LDL is also possible. Lp(a) has been identified in human arterial intima,12 and the relative concentration of apo-Lp(a) in vein grafts was 2.4-fold that of apo B compared with the plasma concentration of the two apolipoproteins.13 The latter finding may reflect that Lp(a) enters the vessel wall at faster rates than LDL or that Lp(a) after its entry into the vessel wall is retained in the subendothelial space to a higher degree than LDL.
In the present report, the transfer of Lp(a) from plasma into the arterial intima was studied; after intravenous injection of labeled Lp(a) and LDL, the aortic intimal clearance during 60 and 180 minutes of labeled Lp(a) is compared with that of labeled human LDL in normal, cholesterol-fed, and Lp(a)-injected rabbits. Also, the location of newly entered Lp(a) in the arterial wall has been studied by immunohistochemistry.
Fifty-five rabbits of the Danish Country strain (Statens Seruminstitut) weighing 3.2±0.1 kg were used. Eight rabbits were used for determination of plasma contamination of aortic intima/inner media, 19 for 60-minute intimal clearance experiments, 10 for 180-minute intimal clearance experiments, 5 for investigation of the time dependency of aortic Lp(a) accumulation, 5 for investigation of the time dependency of aortic LDL accumulation, and 8 for determination of FCRs of Lp(a) and LDL. Of these rabbits, 29 were fed essentially cholesterol-free standard rabbit chow (Altromin 2113) and were designated normal rabbits. The other 26 rabbits were fed cholesterol-enriched chow for 3 to 10 weeks before in vivo experiments commenced; cholesterol-enriched chow contained 1% cholesterol (USP cholesterol, Sigma Chemical Co), 5% corn oil (Oleum Maides BP80, Mecobenzon), and 94% standard rabbit chow. The experimental protocols were approved by the Danish government body supervising animal experiments, the Animal Experiment Inspectorate.
Isolation of Human Lp(a) and LDL
Lp(a) and LDL were each prepared from freshly obtained plasma containing Na2EDTA (1.2 mg/mL), chloramphenicol (80 μg/mL), gentamicin sulfate (80 μg/mL), benzamidin (10 μg/mL), and aprotinin (10 kallikrein units/mL) (all from Sigma). Plasma used for LDL isolation additionally contained ε-amino-n-caproic acid (2.6 mg/mL) (Sigma). The lipoprotein preservation cocktail was added to citrated blood samples before separation of plasma by low-speed centrifugation. Individual plasma Lp(a) concentrations ranged from 0.2 to 0.6 mg/mL. In five separations of Lp(a) and LDL, pooled plasma samples from four individuals were used, and in three separations, plasma samples from only one individual were used. The size of apo(a) varies between individuals and is inversely related to the plasma concentration of Lp(a).6 Since Lp(a) from individuals with high plasma levels was used, the present results may be valid only for smaller isoforms of Lp(a).
To isolate Lp(a), sequential ultracentrifugation was performed at 10°C at solvent densities of 1.05 and 1.12 g/mL, followed by an additional washing step at 1.12 g/mL. After equilibration of the 1.05<d<1.12–g/mL density fraction with PBS by either dialysis or PD-10 columns (Sephadex G-25, Pharmacia), Lp(a) was adsorbed onto a lysine-sepharose 4B column at 10°C (Pharmacia). The column was washed with PBS, and Lp(a) was eluted with ε-amino-n-caproic acid (10 mmol/L) as described by Karàdi et al14 ; in some cases Lp(a) was then concentrated by use of an Amicon cell (Amicon Division). Finally, the Lp(a) preparation was passed through a 0.45-μm filter (Millex GS Millipore SA). Except for one preparation, which had an Lp(a) concentration of <0.1 mg/mL and was used for determination of plasma contamination in three rabbits, the Lp(a) preparations had Lp(a) concentrations of 0.6 to 7.5 mg/mL. The total isolation procedure took from 1 to 2 weeks to complete, after which Lp(a) was iodinated and used immediately for in vivo experiments.
LDL was isolated by sequential ultracentrifugation at 4°C at solvent densities of 1.019 and 1.05 g/mL, followed by an additional washing step at 1.05 g/mL. LDL preparations were dialyzed against PBS containing Na2EDTA (0.1 mg/mL), chloramphenicol, and gentamicin sulfate as described, were filtered through 0.45-μm filters, and were kept at 4°C for a maximum of 5 days before iodination and experimental use.
Analytical and preparative fixed-density ultracentrifugations were performed in a 50.3- or a 55.2-Ti Beckmann rotor. Samples with densities of 1.019 and 1.063 g/mL were centrifuged for at least 1.73×108g · min (average) at 4°C or 10°C, and samples with densities of 1.05, 1.12, and 1.21 g/mL for at least 2.1×108g · min at 10°C. Plasma and dose material with added carrier plasma were adjusted to the desired density by addition of Na2EDTA containing NaBr solutions.
As judged from conventional rocket immunoelectrophoresis using appropriate antibodies and calibrators (Dako A/S), isolated Lp(a) and LDL contained <1% apo A1, plasminogen, and albumin. There was no cross-contamination between the two lipoproteins, as evaluated with a two-tier rocket immunoelectrophoresis (Fig 1⇓). Furthermore, Lp(a) in plasma and isolated Lp(a) had similar electrophoretic mobilities in agarose gel electrophoresis. This supports the idea that Lp(a) had not been significantly oxidized during isolation procedures.
Labeling of Lp(a) and LDL and Total Plasma Lipoproteins
Lp(a) and LDL were each iodinated 13 times15 16 : Lp(a) [2 mL, <0.2 to 14 mg Lp(a)] and LDL (0.25 to 0.35 mL, 5 mg protein) were each mixed with 0.4 mL glycine buffer (1 mol/L, pH 10) and 185 to 740 MBq 125I or 131I (Amersham) before iodine monochloride was added and the tube shaken gently. Lp(a) was labeled with 125I and LDL with 131I, except for one batch, in which the isotopes were reversed. This batch was used to study intimal clearance of labeled lipoproteins during 60 minutes in two normal and in two cholesterol-fed rabbits. The iodination efficiency was 46±8% for Lp(a) and 48±2% for LDL. Unbound iodine was removed, and the labeled lipoprotein preparations were equilibrated to PBS with Na2EDTA with PD-10 columns before 100 mg of rabbit albumin (Sigma) was added. In some experiments, labeled Lp(a) and labeled LDL were additionally dialyzed overnight at 4°C against PBS containing Na2EDTA before the in vivo experiments. The specific activity ranges were 0.15×109 to 0.69×109 cpm/mg LDL protein and 0.15×109 to 2.2×109 cpm/mg Lp(a); one batch of labeled Lp(a) used for plasma contamination determination in three rabbits had a specific activity of >7×1010 cpm/mg Lp(a).
Human total plasma lipoproteins were labeled with 3H-cholesteryl ester by the lecithin–cholesterol acyltransferase method as previously described.17 18 Briefly, human plasma was passed through a 0.22-μm filter (Millex GS, Millipore SA) into a rubber-sealed sterile glass container. After addition of [1α,2α(n)-3H]-cholesterol (Amersham), the plasma was incubated at 37°C for at least 48 hours. To reduce the content of labeled free cholesterol in the lipoproteins by exchange with unlabeled cholesterol in the red blood cells, the labeled plasma was finally incubated with red blood cells at 4°C.
Characterization of Labeled LDL and Lp(a)
In the labeled preparations, >97% of the radioactivity in both lipoproteins was precipitable with TCA, and 5.5±0.7% and 4.4±0.5% of the radioactivity in Lp(a) and LDL, respectively, was lipid-soluble, ie, extractable into chloroform-methanol (1:1, vol/vol).
Iodinated Lp(a) and LDL comigrated with their respective nonlabeled lipoproteins on two-tier immunoelectrophoresis and gradient gel electrophoresis as evaluated by autoradiography (Fig 1⇑). Notably, no labeled free apo(a) was detected on gradient gel electrophoresis and subsequent autoradiography of labeled Lp(a) (Fig 1⇑). Also, similarly prepared labeled Lp(a) and LDL comigrated with their nonlabeled lipoproteins on agarose gel electrophoresis, which supports the idea that labeling did not oxidize the lipoproteins (data not shown).
In the gradient gel electrophoresis analysis, samples of purified lipoproteins containing trace amounts of labeled lipoproteins and samples of purified apo(a) were each mixed with 40% glycerol with bromphenol blue 4:1 (vol/vol) before the 2.5% to 16% polyacrylamide gels (Isolab Inc) were loaded. Apo(a) was purified by M. Jauhiainen and C. Ehnholm, Helsinki, Finland, as described in References 19 and 2019 20 ; the apparent molecular weight of this apo(a) corresponded to the S1 isoform, ie, ≈512 kD.6 After preequilibration of the gels with running buffer (90 mmol/L Tris, 80 mmol/L boric acid, and 2.5 mmol/L Na2EDTA, pH 8.4), lipoproteins and apo(a) were subjected to electrophoresis at 10°C on a Pharmacia GE 2/4 apparatus for 15 minutes at 15 V (1 mA) and for 20 minutes at 70 V (10 to 15 mA) followed by 22 to 24 hours at 125 V (20 to 27 mA) for a total of 2750 to 3000 V · hours. The gels were stained with Coomassie blue (2 mg/mL) with acetic acid (5% to 7%) as combined fixative, solvent for coloring agent, and destainer. Stained gels were restored to the original size in 40% glycerol/5% methanol before they were dried.
In the two-tier rocket immunoelectrophoresis,21 three separate bands of 1.25% agarose were applied onto a GelBond film (FMC Bioproducts). The lower band of the gel was pure agarose, the middle band contained anti-Lp(a) (Dako A/S), and the upper band contained anti-apo B (Dako A/S). Wells were punched in the lower band of the gel and loaded with samples containing labeled lipoproteins before electrophoresis was conducted with a barbital buffer. After drying, rockets were visualized with Coomassie blue. To visualize labeled lipoproteins, dried gels were placed on x-ray films (Cronex, DuPont) for 1 to 3 days before the films were developed.
The two labeled lipoproteins could be separated by gel filtration chromatography (Fig 2⇓); Lp(a) was the larger particle. To estimate the relative distribution of radioactivity between apo(a) and apo B in Lp(a), a mixed dose of labeled Lp(a) and LDL was subjected to density-gradient ultracentrifugation (Fig 3⇓). Labeled Lp(a) and LDL were separated, and about 10% of the Lp(a) label was in the bottom protein fraction. After incubation with DTE, which dissociates apo(a) from apo B in the Lp(a) particle, some 25% of the total radioactivity was now in the bottom protein fraction, and the remaining radioactivity comigrated with labeled LDL; these results suggest that at least 15% of the radioactivity was in apo(a) and at least 75% was in apo B.
In fixed-density ultracentrifugation analysis of labeled LDL used for injections as well as in plasma 60 and 180 minutes after injection, >90% of the total radioactivity was in the 1.019<d<1.063–g/mL lipoprotein fraction. Only 3±1% (n=13) of the radioactivity in the labeled LDL preparations and <1.5% of the LDL radioactivity in plasma at 60 and 180 minutes was in the d>1.21–g/mL fraction. These values for labeled LDL are in accordance with previous results from our laboratory.22 In the labeled Lp(a) preparations, 84±2% (n=13) of the total radioactivity was in the 1.05<d<1.12–g/mL fraction, whereas 7.4±0.5% of the total radioactivity was in the d>1.21–g/mL fraction. In the plasma samples as well, some 10% of the total radioactivity was in the d>1.21–g/mL fraction. This radioactivity in the d>1.21–g/mL fraction in Lp(a) experiments is believed to reflect mainly an ultracentrifugation artifact rather than impurities in the labeled preparations, for the following reasons. First, in an attempt to purify labeled Lp(a) further, one batch was adjusted to a solvent density of 1.12 g/mL and ultracentrifuged as described. However, after a new ultracentrifugation of this “purified” Lp(a), some 10% of the radioactivity was still recovered in the d>1.21–g/mL fraction. Second, on gel electrophoresis of the d>1.21–g/mL fraction [containing about 10% of the total radioactivity in Lp(a)] using a 4% to 14% SDS polyacrylamide gel (SDS-PAGE) followed by autoradiography, the radioactivity did not migrate into the gel. Third, on gel filtration using a PD-10 column (Sephadex G-25), 65% of the radioactivity in the d>1.21–g/mL fraction did not pass the column. Fourth, on gradient gel electrophoresis and gel filtration of labeled Lp(a) used for injections as well as on gel filtration of labeled Lp(a) in plasma 180 minutes after injection (not shown), no disintegration products of Lp(a) were observed (Figs 1⇑ and 2⇑). Fifth, in other studies in our laboratory (data to be published elsewhere), the relative distribution of radioactivity between free apo(a) and Lp(a) in preparations of labeled Lp(a) was determined by two-tier rocket immunoelectrophoresis. In these analyses, labeled Lp(a) migrated first through an anti–apo B–containing and then into an anti-Lp(a)–containing agarose gel. Intact Lp(a) particles then precipitated in the anti-apo B gel, whereas free apo(a) precipitated in the anti-Lp(a) gel. In these analyses, on average 1% (n=2), 0% (n=12), and 3% (n=6) of the total radioactivity migrated as free apo(a) in labeled Lp(a) used for injections, in rabbit plasma 3 hours after intravenous injection, and in rabbit plasma 23 hours after intravenous injection, respectively. Taken together, these results suggest that the larger part of the radioactivity in the d>1.21–g/mL fraction are labeled aggregates formed during ultracentrifugation procedures rather than small disintegration products of labeled Lp(a), including free apo(a).
ε-Amino-n-caproic acid blocks lysine binding sites on Lp(a). The concentration of ε-amino-n-caproic acid was 14 μmol/L (estimated by amino acid analysis) after dialysis of the Lp(a) preparation against PBS with Na2EDTA (compared with 10 mmol/L in the elution buffer). Taking into consideration the dilution in plasma after intravenous injection of the labeled preparations, it therefore seems unlikely that ε-amino-n-caproic acid interfered significantly with the metabolism of Lp(a) in vivo.
Protocol for In Vivo Experiments
The aortic intimal clearance of labeled Lp(a) and LDL was determined simultaneously in 9 normal rabbits, 16 cholesterol-fed rabbits, and 4 otherwise normal rabbits with plasma levels of Lp(a) comparable to those in humans; just before injection of labeled lipoproteins, the latter 4 rabbits received an intravenous injection of human lipoproteins from the d<1.12–g/mL plasma fraction. This Lp(a)-rich preparation was isolated from human plasma by a single ultracentrifugation, followed by dialysis against PBS and filtration through a 0.22–μm filter. Each rabbit received 8 mL of the preparation, corresponding to 45 mg Lp(a) and 0.35 mmol cholesterol; 20% of this cholesterol was in d<1.019–g/mL lipoproteins, 63% in 1.019<d<1.050–g/mL lipoproteins, 9% in 1.050<d<1.063–g/mL lipoproteins, and 7% in 1.063<d<1.12–g/mL lipoproteins. Labeled Lp(a) and LDL used in these rabbits was obtained from the same plasma pool as the d<1.12–g/mL plasma lipoprotein fraction.
To measure intimal clearance, a mixed dose of Lp(a) and LDL labeled with 125I and 131I, respectively, was injected into an ear vein, followed by 2 mL of saline to rinse the syringe (the isotopes were reversed in 5 of the 29 rabbits). Each rabbit received an average of 0.23±0.06 mg labeled Lp(a)/kg body wt and 0.32±0.01 mg labeled LDL protein/kg body wt, corresponding to 2.1±0.3×108 cpm/kg body wt and 1.7±0.2×108 cpm/kg body wt, respectively. Subsequently, blood samples were drawn at regular intervals until the aorta was removed after 60 minutes (9 normal and 10 cholesterol-fed rabbits) or after 180 minutes [6 cholesterol-fed and 4 Lp(a)-injected rabbits]. In 8 normal and 8 cholesterol-fed rabbits exposed to labeled lipoproteins for 60 minutes, the spleen and a part of the liver were also removed.
To evaluate the plasma contamination of the aortic intima/inner media, 6 cholesterol-fed rabbits in the 180-minute experiments received an intravenous injection of [3H]cholesteryl ester–labeled human plasma (0.5±0.1×108 cpm/kg) 5 to 10 minutes before the aorta was removed. In addition, the plasma contamination of the aortic intima/inner media of iodinated Lp(a) and LDL was compared directly in 6 normal and 2 cholesterol-fed rabbits; the aorta was removed 5 to 10 minutes after injection of labeled Lp(a) and LDL.
The accumulation of labeled Lp(a) in aortic intima/inner media as a function of time was determined in 2 normal and 3 cholesterol-fed rabbits without atherosclerotic lesions. 131I-Lp(a) and 125I-Lp(a) were injected 180 minutes and 60 minutes, respectively, before removal of aorta. Data from these rabbits were compared with similar data on the time-dependent aortic accumulation of labeled human LDL obtained in 5 normal rabbits, which were also studied for a different purpose.
The FCR of biologically screened labeled Lp(a) and LDL was determined in 8 rabbits: At 60 minutes after injection of labeled Lp(a) and LDL into 4 normal and 4 cholesterol-fed rabbits, 2 to 4 mL of serum was obtained and injected intravenously into 8 other rabbits (5 normal and 3 cholesterol-fed). Blood samples were then taken regularly during the following 3 to 4 days for FCR calculations.
Rabbits were euthanatized by an injection of pentobarbital (40 to 80 mg/kg body wt IV). A needle was inserted into the left ventricle of the heart, and the systemic circulation was perfused with 500 to 1000 mL of saline (4°C) before the aorta. In 16 rabbits, the spleen and a part of the liver were also removed. The aorta was freed of adventitia immediately, and in 10 rabbits, an ≈0.5-cm-wide ring of the aortic arch (just above the obliterated ductus arteriosus) was taken for histological examination. The vessel was opened longitudinally, and the area of the intimal surface was outlined on graph paper before the aorta was divided into the aortic arch, the thoracic aorta, and the abdominal aorta at the level of the first intercostal branches and the coeliac axis, respectively. The intima/inner media in each of these parts was stripped from the outer media. In 11 cholesterol-fed rabbits, the wet weight of the intima/inner media represented 63±2%, 57±2%, and 57±2% of the total wet weight in the aortic arch, thoracic aorta, and abdominal aorta, respectively. In 8 normal rabbits, the corresponding values were 56±3%, 62±2%, and 53±2%, respectively. The intima/inner media was minced with scissors in 900 μL cold saline with 100 μL albumin solution (100 mg/mL) or 100 μL plasma was added to the tissues (to provide cholesteryl ester mass as a carrier for [3H]cholesteryl ester during thin-layer chromatography procedures), and proteins were precipitated with TCA at a final concentration of 15% (wt/vol) at 4°C. Aliquots of plasma samples (20 to 25 μL), diluted doses containing 25 to 250 nL of the initial dose, minced spleen, and liver tissue (100 to 200 mg) had 900 μL cold saline and 100 μL albumin solution added, followed by TCA precipitation. After mixing and centrifugation, total as well as TCA-soluble radioactivity (in the supernatant) was determined by counting samples for 42 to 60 minutes in a double-channel gamma counter (Selektronik or 1282 Compugamma, LKB, Wallac). Typical standard deviations for observed counting rates were <1% for both isotopes.
Lipids in the TCA precipitates of aortic intima/inner media and aliquots of plasma and doses were extracted for at least 24 hours with chloroform/methanol (2:1, vol/vol); after addition of methanol to form a 1:1 solution, precipitates were centrifuged and washed twice with chloroform/methanol before chloroform was added to the combined washes to reestablish the original 2:1 ratio. Subsequently, the extract was washed by the procedure of Folch et al,23 and cholesterol was determined by the Liebermann-Burchard method after saponification.24 The cholesterol content of tissues, to which had been added human plasma to provide cholesteryl ester for thin-layer chromatography, was corrected for this additional cholesterol by subtraction; the fraction of total cholesterol from carrier cholesterol was 8±2% in the aortic arch, 31±2% in the thoracic aorta, and 31±2% in the abdominal aorta. Esterified cholesterol was isolated from the extract by thin-layer chromatography,17 and the amount of 3H was determined by liquid scintillation counting in a Tricarb 2000 counter (Packard). Standard deviations for observed 3H counting rates were <2%.
Plasma and lipoprotein cholesterol concentrations were determined with the CHOD-PAP enzymatic method (Boehringer Mannheim). Lp(a) concentrations were determined by rocket immunoelectrophoresis or a turbidimetric test system; presented values represent total Lp(a) mass [Lp(a) calibrators, polyclonal rabbit antibodies, and the turbidimetric test system were all from Dako A/S]. LDL protein was estimated from the absorbance at 220 nm.25
Tissue rings from the aortic arch were fixed in buffered formalin (3.7%) and paraffin-embedded according to standard procedures. Three sections of each paraffin block were conventionally stained (hematoxylin-eosin,26 Masson-Goldner, and elastica27 ) by use of kits from Camon to define the degree of atherosclerosis. Subsequently, immunohistochemical staining was performed on the same sections with the avidin-biotin-complex technique as previously described.28 Human apo(a), rabbit macrophages, and rabbit smooth muscle cells were visualized by use of monoclonal mouse antibodies. Anti–human apo(a) was kindly provided by Dr U. Beisiegel (Department of Medicine, University of Hamburg, Germany) and has previously been shown not to cross-react with plasminogen. RAM11 (diluted 1:50) (Dako A/S) and HHF 35 (undiluted) (Enzo) antibodies were used to visualize macrophages and smooth muscle cells, respectively. Staining for albumin was performed with polyclonal goat antibodies (diluted 1:150 000) (Medac).
TCA-precipitable radioactivity in tissues, plasma, and doses was used in the calculations. The aim of the study was to compare the transfer of Lp(a) and LDL across the luminal surface of the arterial wall. Since plasma macromolecules probably enter aortic media from both the luminal and adventitial sides,29 30 only radioactivity in the intima/inner media was included in the calculations. The data presented on intimal clearance of labeled lipoproteins, I (nL · cm−2 · h−1), was calculated as the aortic radioactivity at the end of the experiment, A (cpm/cm2), divided by the mean plasma radioactivity concentration, cavg (cpm/nL), and the duration of the experiment, t (hours). Additionally, corrections for plasma contamination of arterial tissue were included in intimal clearance calculations in six cholesterol-fed rabbits by solving the equation where C is the plasma contamination (nL/cm2). Plasma contamination is calculated as the amount of [3H]cholesteryl ester in the arterial tissue divided by the [3H]cholesteryl ester plasma concentration at the end of the experiment. cend (cpm/nL) is the plasma concentration at the end of the experiment of the tracer used to measure intimal clearance.
To calculate FCR (hour−1), a double exponential curve-fitting plot of plasma radioactivity against time was used according to Matthews31 : where C1 and C2 are the constants and b1 and b2 are the slopes of the two exponentials. In the present experiments, the fractional catabolic rate of LDL was determined with biologically screened labeled human LDL as the tracer molecule; the values obtained were highly similar to values obtained for biologically screened rabbit LDL32 ; human LDL has previously been used to study LDL metabolism in rabbits.33 The calculation of FCR for biologically screened human Lp(a), however, may not be entirely correct if there is no endogenous Lp(a) already present in the rabbits31 ; one preliminary report, however, has indicated that rabbits may have Lp(a).34
The distribution volume (mL/kg) for labeled lipoproteins was calculated as the amount of injected radioactivity (cpm) divided by the initial plasma radioactivity concentration (cpm/mL) and the weight of the rabbit (kg).
Values are presented as mean±SEM. Statistical analysis of differences between values were performed by use of paired or nonpaired Student’s t tests. Linear regression analysis was evaluated by use of the Pearson correlation coefficient. All probability values are for two-tailed tests.
In Vivo Metabolism of Lp(a)
The volume of distribution was 37.7±1.8 and 38.5±1.8 mL/kg (n=33) for simultaneously injected labeled human Lp(a) and LDL, respectively. In rabbits used for determination of aortic intimal clearance, 24±3% of labeled Lp(a) and 12±2% of labeled LDL (n=29, P<.001 paired test) was removed from plasma during the first 60 minutes after injection of the lipoproteins. Similarly, when labeled lipoproteins were biologically screened and reinjected into eight other rabbits, 30±2% of Lp(a) and 10±4% of LDL was cleared from plasma during the first 60 minutes; there was no statistically significant difference between normal and cholesterol-fed rabbits. Sixty minutes after injection of labeled lipoproteins, significantly more Lp(a) label than LDL label was found in the liver and spleen in both normal and cholesterol-fed rabbits (Table 1⇓).
The FCR of biologically screened Lp(a) was significantly higher than that of LDL in both normal and cholesterol-fed rabbits (Fig 4⇓, Table 2⇓). There was no statistically significant association between the two lipoprotein FCRs as determined by linear regression analysis (data not shown). The mean FCR of Lp(a) was similar in normal and cholesterol-fed rabbits.
Simultaneous Determination of Aortic Intimal Clearance of Labeled Lp(a) and LDL
The aortic intimal clearance of labeled Lp(a) and LDL was determined simultaneously in rabbits exposed to labeled lipoproteins during either 60 minutes or 180 minutes. Linear regression analysis of the intimal clearance of labeled LDL as a predictor for intimal clearance of labeled Lp(a) showed positive associations between the two variables in normal and cholesterol-fed rabbits in the 60-minute experiments as well as in cholesterol-fed and otherwise normal but Lp(a)-injected rabbits in the 180-minute experiments (Fig 5⇓). In each of the four groups studied, the association between the intimal clearance of LDL and Lp(a) was statistically significant in at least one aortic segment.
In the 60-minute experiments, the average intimal clearance of labeled Lp(a) was significantly higher than that of labeled LDL in both normal and cholesterol-fed rabbits in all three aortic segments (Table 3⇓). In the 180-minute experiments, the intimal clearance of labeled Lp(a) tended to be lower than that of labeled LDL in the aortic arch in both cholesterol-fed and Lp(a)-injected rabbits (Table 3⇓). However, the differences were not statistically significant. The intimal clearance during 180 minutes was similar for labeled Lp(a) and LDL in the thoracic and abdominal aorta in both cholesterol-fed and Lp(a)-injected rabbits (Table 3⇓).
Impact of Plasma Lp(a) Concentration on Aortic Intimal Clearance of Labeled Lp(a)
The impact of differences in plasma Lp(a) concentrations on transfer of Lp(a) into the arterial wall was investigated: The plasma Lp(a) concentration in otherwise normal rabbits was increased to an average of 0.33 mg/mL by intravenous injection of d<1.12–g/mL human lipoproteins containing Lp(a) before determination of the intimal clearance of labeled Lp(a) and labeled LDL during 180 minutes.
Rabbits with high plasma Lp(a) levels had a similar intimal clearance of labeled Lp(a) in the thoracic and abdominal aorta compared with cholesterol-fed rabbits with only trace amounts of Lp(a) in plasma; the ratio of the intimal clearance of labeled Lp(a) to that of labeled LDL was similar in Lp(a)-injected and cholesterol-fed rabbits in both aortic sites (Table 3⇑, 180-minute experiments; ratios can be calculated).
In the aortic arch, the intimal clearance of labeled Lp(a) tended to be higher in cholesterol-fed rabbits compared with Lp(a)-injected rabbits (P=.12, nonpaired test). However, the ratio of the intimal clearance of labeled Lp(a) to that of labeled LDL was very similar in cholesterol-fed and Lp(a)-injected rabbits (0.74 versus 0.64).
Time Dependency of Aortic Accumulation of Labeled Lp(a) and LDL
The time dependency of aortic accumulation of labeled Lp(a) and LDL was investigated in rabbits without atherosclerosis: The accumulations of labeled lipoprotein during 60 minutes and 180 minutes were compared in the same animal. The time-dependent accumulation of labeled Lp(a) in the aortic arch (expressed in plasma equivalents as the aortic radioactivity divided by the time-averaged concentration of radioactivity in plasma) was curvilinear, with a concavity toward the x axis (Fig 6⇓). This suggests that some of the labeled Lp(a) label was removed from the intima/inner media between 60 and 180 minutes after intravenous injection. In contrast, the amount of labeled LDL that accumulated in the aortic arch of normal rabbits increased almost linearly for 3 hours (Fig 6⇓). In the thoracic and abdominal aorta, the time-dependent accumulation was compatible with a significant loss of label between 60 minutes and 180 minutes for both Lp(a) and LDL (data not shown).
Arterial Wall Contamination With Labeled Lipoproteins
The contamination of the arterial intima/inner media was determined simultaneously for Lp(a) and LDL (Table 4⇓). The contamination of labeled Lp(a) and LDL was similar in the aortic arch, the thoracic aorta, and the abdominal aorta in 6 normal and 2 cholesterol-fed rabbits. On linear regression analysis of LDL contamination as a determinant for Lp(a) contamination, there was a close association between the two variables in all three aortic segments in normal rabbits. The contamination of [3H]cholesteryl ester–labeled lipoproteins, determined in 6 cholesterol-fed rabbits in the 180-minute experiment, was similar to that of radiolabeled Lp(a) and LDL in all three aortic segments (Table 4⇓).
In the 6 cholesterol-fed rabbits, corrections for contamination reduced the intimal clearance of labeled Lp(a) and LDL by an average 17% and 14%, 47% and 48%, and 43% and 44% in the aortic arch, the thoracic aorta, and the abdominal aorta, respectively.
Immunohistochemical Detection of Lp(a) in the Arterial Wall
In the same rabbits as were used for intimal clearance measurements, immunohistochemistry showed human apo(a) predominantly in the intima at sites with fatty streak lesions. Apo(a) immunoreactivity was situated mainly in the core region, predominantly within the cytoplasm of macrophage-derived foam cells (Fig 7⇓). Aortas from 10 rabbits were studied; the staining pattern of apo(a) immunoreactivity was similar to the one shown in Fig 7⇓ in tissue specimens from 4 rabbits with fatty streak lesions, whereas apo(a) immunoreactivity was not detected in tissue specimens from 6 rabbits without atherosclerotic lesions.
The physiological role of Lp(a) is not known. So far, Lp(a) has been detected in humans, old-world monkeys, hedgehogs, and guinea pigs,35 36 37 and preliminary results also suggest the presence of an Lp(a)-like particle in rabbit plasma.34 Nevertheless, the use of the rabbit to study in vivo metabolism of human Lp(a) may have the limitation that the metabolism of the Lp(a) particle by the rabbit could be different from that in humans. However, the rabbit may still be useful to study the barrier function of the aortic endothelium with regard to the Lp(a) particle: Comparable studies in humans and rabbits suggest that the transfer into the arterial wall of other plasma lipoproteins such as HDL and LDL may be a passive process that is similar in the two species.17
The use of tracer molecules to study in vivo metabolism of lipoproteins critically depends on the ability of the labeled compound to mimic the metabolism of the native particle. Isolated Lp(a) can undergo self-degradation,38 and the homogeneity and integrity of labeled Lp(a) (and LDL) was therefore confirmed in vitro by gel electrophoresis and gel chromatography techniques. Furthermore, the similar initial removal in vivo of labeled lipoproteins from plasma before and after biological screening suggests that neither Lp(a) nor LDL had been exposed to significant damage during isolation and labeling procedures. Ultracentrifugation procedures have previously been reported to induce loss of Lp(a) immunoreactivity.39 Also, in the present study, fixed-density ultracentrifugation procedures presumably induced formation of aggregates of Lp(a) label (see “Methods”).
In Vivo Catabolism of Lp(a)
Studies on Lp(a) levels in subjects with familial hypercholesterolemia as well as simultaneous determination of the FCRs of labeled Lp(a) and LDL in humans are conflicting with regard to the role of the LDL receptor in the catabolism of Lp(a).40 41 42 43 44 In the present study, the mean FCR of labeled Lp(a) was twofold to threefold that of LDL, which could be caused by a smaller pool of Lp(a) in rabbits compared with the LDL pool. Since there was no association between the simultaneously determined FCRs of human Lp(a) and LDL and since the mean FCR of Lp(a) was similar in normal and in cholesterol-fed rabbits even though hepatic LDL receptors are downregulated by hypercholesterolemia in rabbits,45 the present data do not support the hypothesis that LDL receptors play a major role in the catabolism of Lp(a) in the rabbit. This finding is in accordance with recent observations in humans.46 The accumulation of Lp(a) radioactivity in the liver after 1 hour of exposure to labeled Lp(a) was similar to the amount of radioactivity removed from plasma, assuming a liver weight of 150 g. This finding is in accordance with the notion that the liver is the major organ for catabolism of Lp(a) in the rabbit.
Transfer of Lp(a) and LDL Into Aortic Intima
The intimal clearance of labeled lipoprotein represents the influx of labeled lipoprotein minus the loss of labeled lipoprotein from the intima due to efflux or cellular degradation. In the aortic arch of rabbits without atherosclerosis, the loss of labeled LDL from the intima was small compared with the total influx during 3 hours. This is in accordance with a previous report18 and implies that the intimal clearance of labeled LDL in the aortic arch intima during 3 hours, when expressed in plasma equivalents, reflects the LDL permeability of the arterial wall. In contrast, in the thoracic and abdominal aorta, the intimal clearance of labeled LDL may underestimate the true LDL permeability.
Comparison of the intimal clearance of labeled Lp(a) during 60 and 180 minutes in both cholesterol-fed and nonatherosclerotic rabbits, as well as direct studies of the time-dependent accumulation of labeled Lp(a) in the aortic intima/inner media of nonatherosclerotic rabbits, suggests that labeled Lp(a) was lost from all three aortic sites from 1 to 3 hours in both normal and cholesterol-fed rabbits. Therefore, the intimal clearance of labeled Lp(a) during 3 hours may to some extent underestimate the aortic permeability to Lp(a), even in the aortic arch, and the intimal clearance during 1 hour may be a more reliable measure of the Lp(a) permeability under the present experimental conditions. Whether the presumed loss of labeled Lp(a) from the intima during 180 minutes is due to efflux of intact Lp(a) particles or degradation of the protein moiety of Lp(a) is not known. It could be of importance that the major part of the Lp(a) preparations used in the 60-minute experiment and in the 60- versus 180-minute experiment was isolated in the laboratory of K. Christoffersen (Dako A/S) with an isolation time of 2 weeks, whereas all Lp(a) preparations used in the 180-minute experiment were isolated in our own laboratory (Department of Clinical Biochemistry, Rigshospitalet) with an isolation time of about 1 week. However, the isolation procedures were very similar, and in vitro characterizations of labeled Lp(a) showed no indication of damage to Lp(a) in any of the Lp(a) preparations. Therefore, it appears unlikely that the observed difference in intimal clearance of labeled Lp(a) during 60 minutes and 180 minutes can be explained by differences between batches of Lp(a).
A recent autoradiographic study from Kreuzer et al47 suggested that iodinated Lp(a) displays increased accumulation compared with iodinated LDL in the murine arterial wall. Such increased accumulation could reflect an increased entry or decreased efflux of Lp(a) compared with LDL. In the present experiments, the intimal clearance of labeled Lp(a) was higher than that of labeled LDL when the arterial wall was exposed to the labeled lipoproteins for 60 minutes. This could imply that the aortic permeability to Lp(a) is larger than to LDL and would support the previous findings in mice. Conversely, when the intimal clearance of LDL during 60 minutes was evaluated as a predictor of the intimal clearance of Lp(a), the slope of the regression lines was close to 1, whereas the y intercept was 12 to 60 nL/cm2 in the normal rabbits. This favors the idea that the aortic permeability is similar for the two lipoproteins but that Lp(a) may bind to the endothelium of normal rabbits to a larger extent than LDL. However, the plasma contamination of labeled Lp(a) and LDL was similar in normal rabbits. Finally, if Lp(a) is transferred into the arterial wall by mechanisms similar to those of LDL (as discussed below), one would expect that Lp(a), being larger than LDL, would enter the arterial wall at slower rates than LDL. Hence, the hypothesis that the aortic permeability to Lp(a) is larger than that to LDL should be confirmed in other studies, preferably in humans.
Some ultrastructural studies support the concept of a vesicular transport of macromolecules, including LDL, through the endothelial cells.48 Because Lp(a) binds to endothelial cells in vitro,9 it was speculated that Lp(a) may be transported across the endothelium at faster rates than LDL by transcytosis of Lp(a) bound to the plasma membrane. However, the aortic contamination of labeled Lp(a) and LDL after an exposure period of 5 to 10 minutes was almost equal for both particles. The difference between the average contamination of the two lipoprotein species was about 2 nL/cm2. This difference is quantitatively insignificant in comparison with the intimal clearance of the particles during both 60 and 180 minutes (30 to 85 nL · cm−2 · h−1 in the aortic arch). Therefore, the present data do not support the idea that preferential binding of Lp(a) to endothelial cells compared with LDL is a major determinant of transfer of Lp(a) into the arterial wall. These data obtained by tracer kinetic methods were supported by the immunohistochemical evaluations: In contrast to core regions of fatty streak lesions, no apo(a) was detected in relation to the endothelial surface.
The contamination of aortic intima/inner media reported here is three to seven times higher than in one previous study from our laboratory18 but similar to contamination values in another.22 This divergence between studies may result from differences between rabbits rather than from technical reasons: In the present study, the plasma contamination was similar when either iodinated Lp(a), iodinated LDL, or lipoprotein labeled in vitro with [3H]cholesteryl ester was used. Furthermore, similar high contamination values were previously obtained with lipoproteins labeled in vivo by oral administration of [3H]cholesterol to a cholesterol-fed rabbit.22 Importantly, however, since the contaminations of Lp(a) and LDL were similar, the association between intimal clearance of Lp(a) and LDL depicted in Fig 5⇑ remains whether or not the data are corrected for plasma contamination.
The transfer of LDL into the arterial wall is independent of LDL receptors,49 and the linear relationship between plasma LDL concentrations and influx of LDL into the arterial wall,50 as well as the relationship of lipoprotein influx and lipoprotein size for VLDL, IDL, LDL, and HDL,17 32 50 51 52 favors the notion of a passive, concentration- and size-dependent sieving of these lipoproteins from plasma into the arterial wall.
The present studies support the idea that transfer of Lp(a) and LDL into the arterial wall occurs by a similar mechanism in rabbits. Although not all correlations reached statistical significance, the simultaneous use of differently labeled LDL and Lp(a) made it possible to demonstrate positive associations between intimal clearance of the two labeled lipoproteins in all aortic segments in normal, cholesterol-fed, and Lp(a)-injected rabbits by either 60-minute or 180-minute exposure to labeled lipoproteins; in each group of rabbits, significant associations were observed in at least one aortic segment.
The intimal clearance of labeled Lp(a) during 180 minutes presumably results from the combination of influx and loss of labeled Lp(a) during this time period. Therefore, it is difficult to conclude whether the association between the intimal clearance of labeled Lp(a) and the intimal clearance of labeled LDL during 180 minutes reflects an association between transfer of labeled Lp(a) and LDL into the intima, loss of the two lipoproteins from the intima, or a combination of the two.
The intimal clearance of labeled Lp(a) (in nL · cm−2 · h−1) seems to be independent of plasma Lp(a) levels: In both the thoracic aorta and the abdominal aorta, the intimal clearance of labeled Lp(a) during 180 minutes was similar in cholesterol-fed rabbits with trace amounts of Lp(a) in plasma and in Lp(a)-injected rabbits with plasma Lp(a) concentrations similar to those in humans, ie, 0.3 mg/mL. Since the arterial wall permeability increases linearly with increasing lesion severity in the rabbit,50 the trend toward a higher intimal clearance of Lp(a) in the aortic arch of cholesterol-fed compared with the otherwise normal Lp(a)-injected rabbits may result from the presence of atherosclerosis in the aortic arch in the cholesterol-fed animals (all the cholesterol-fed rabbits had visible lesions in the aortic arch). At this stage, it should be noted that comparison of cholesterol-fed rabbits with otherwise normal Lp(a)-injected rabbits requires the assumption of a similar mechanism for transfer of Lp(a) into the arterial wall in normal and cholesterol-fed rabbits. Also, since labeled Lp(a) may be lost from the intima/inner media during 180 minutes of exposure, it cannot be entirely excluded that injection of Lp(a) mass reduces or increases the transfer of labeled Lp(a) into the arterial wall. Such effects, however, would need to have been balanced by similar increases or decreases in loss of labeled Lp(a) to obtain the present results. This possibility seems unlikely.
Altogether, the present kinetic studies favor the hypothesis that transfer of Lp(a) across the endothelial barrier of the rabbit aorta is a passive, concentration-dependent process similar to that of LDL. However, Lp(a) and LDL may exhibit different kinetics upon entrance into the intima/inner media.
The Potential Role of Lp(a) in Atherogenesis
The mechanism by which Lp(a) may promote the development of atherosclerosis remains to be elucidated. In humans, comparison of the flux of LDL cholesterol from plasma into the aortic intima (which can be calculated as the product of the plasma concentration of LDL and the aortic permeability to LDL) with the cholesterol content of the arterial wall suggests that only a smaller fraction of the cholesterol that enters the intima is actually deposited.17 If the transfer of Lp(a) into the arterial wall is similar to that of human LDL, the amount of Lp(a) cholesterol entering the arterial wall in subjects with high Lp(a) levels will be quantitatively significant in relation to the aortic cholesterol accumulation during the development of atherosclerosis if Lp(a) is trapped in the intima, although the latter notion cannot be supported by the present study.
In the present study, human apo(a) immunoreactivity was located mainly in intima with fatty streak lesions within the cytoplasm of foam cells. In human arterial tissue, apo(a) immunoreactivity has been detected predominantly in atherosclerotic lesions,12 28 and morphometric analysis suggested a preferential accumulation of apo(a) within the extracellular space in comparison with the cytoplasm of foam cells.28 However, those human arterial tissues had been exposed to Lp(a) for many years. The present study, on the other hand, examined the initial metabolism of Lp(a) in developing atherosclerotic lesions. The data suggest that Lp(a), upon entering the arterial intima, can be taken up by macrophages and thus contribute to foam cell formation. This notion is in accordance with an autoradiographic study in mice,47 in which autoradiographic grains, upon injection of 125I-Lp(a), predominated in the foam cell–rich shoulder regions in preference to the central portion of intermediate lesions.
In summary, the present in vivo study suggests that the transfer of Lp(a) into the arterial wall takes place by a mechanism similar to that of LDL, ie, by a passive plasma concentration-dependent transfer, and that Lp(a) can be taken up by foam cells after it has entered the arterial intima; however, the data gave some indications that Lp(a) and LDL may be metabolized differently upon entrance into the arterial wall.
Selected Abbreviations and Acronyms
|FCR||=||fractional catabolic rate|
This study was supported by the Danish Heart Foundation, the Danish Medical Research Council, and the Danish Foundations Novo’s Fond Komite, Overlæge Johan Boserup og Lise Boserups Legat, and Den Lægevidenskabelige Forskningsfond for Storkøbenhavn, Færøerne og Grønland. Thanks are extended to Kim Christophersen, MS, Dako A/S, Denmark, for fruitful discussions on procedures for isolation of Lp(a) and for measuring apo A1, albumin, and plasminogen in isolated Lp(a) and LDL preparations. Anders Johnsen, MS, PhD, Department of Clinical Biochemistry, Rigshospitalet, Copenhagen, kindly measured ε-amino-n-caproic acid. Apo(a) was generously prepared by C. Ehnholm and M. Jauhiainen, National Public Health Institute, Helsinki, Finland. Karen Rasmussen, Kurt S. Jensen, and Hanne Damm provided excellent technical assistance. Antibodies to Lp(a) and apo B were a gift from Dako A/S, Denmark.
Scanu AM, Fless GM. Lipoprotein(a): heterogeneity and biological relevance. J Clin Invest. 1990;85:1709-1715.
Utermann G. The mysteries of lipoprotein(a). Science. 1989;246:904-910.
Fless GM, Rolih CA, Scanu AM. Heterogeneity of human plasma lipoprotein (a): isolation and characterisation of the lipoprotein subspecies and their apoproteins. J Biol Chem. 1984;259:11470-11478.
Sandholzer C, Nilmani S, Kark JD, Rees A, Jaross W, Dieplinger H, Hoppichler F, Boerwinkle E, Utermann G. Apo(a) isoforms predict risk for coronary heart disease: a study in six populations. Arterioscler Thromb. 1992;12:1214-1226.
Mölgaard J, Klausen IC, Lassvik C, Færgeman O, Gerdes LU, Olsson AG. Significant association between low-molecular-weight apolipoprotein(a) isoforms and intermittent claudication. Arterioscler Thromb. 1992;12:895-901.
Eaton DL, Fless GM, Kohr WJ, Mclean JW, Xu Q, Miller CG, Lawn RM, Scanu AM. Partial amino acid sequence of apolipoprotein(a) shows that it is homologous to plasminogen. Proc Natl Acad Sci U S A. 1987;84:3224-3228.
Loscalzo J, Weinfeld M, Fless GM, Scanu AM. Lipoprotein(a), fibrin binding, and plasminogen activation. Arteriosclerosis. 1990;10:240-245.
Rath M, Niendorf A, Reblin T, Dietel M, Krebber H, Beisiegel U. Detection and quantification of lipoprotein(a) in the arterial wall of 107 coronary bypass patients. Arteriosclerosis. 1989;9:579-592.
Cushing GL, Gaubatz JW, Nava ML, Burdick BJ, Bocan TMA, Guyton JR, Weilbaecher D, DeBakey ME, Lawrie GM, Morrisett JD. Quantitation and localization of apolipoprotein [a] and B in coronary artery bypass vein grafts resected at reoperation. Arteriosclerosis. 1989;9:593-603.
Stender S, Hjelms E. In vivo transfer of cholesteryl ester from high and low density plasma lipoproteins into human aortic tissue. Arteriosclerosis. 1988;8:252-262.
Nielsen LB, Nordestgaard BG, Stender S, Kjeldsen K. Aortic permeability to LDL as a predictor of aortic cholesterol accumulation in cholesterol-fed rabbits. Arterioscler Thromb. 1992;12:1402-1409.
Jauhiainen M, Metso J, Koskinen P, Ehnholm C. Characterization of the enzyme activity of human plasma lipoprotein(a) using synthetic peptide substrates. Biochem J. 1991;274:491-496.
Nielsen LB, Stender S, Kjeldsen K, Nordestgaard BG. Effect of angiotensin II and enalapril on transfer of low density lipoprotein into aortic intima in rabbits. Circ Res. 1994;75:63-69.
Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem. 1957;226:497-509.
Zilversmit DB, Shea TM. Quantitation of apo B-48 and apo B-100 by gel scanning or radioiodination. J Lipid Res. 1989;30:1639-1646.
Sheehan DC, Hrapchak BB. Theory and Practice of Histotechnology. St Louis, Mo: The CV Mosby Co; 1980:137.
Niendorf A, Rath M, Wolf K, Peters S, Arps H, Beisiegel U, Dietel M. Morphological detection and quantification of lipoprotein(a) deposition in atheromatous lesions of human aorta and coronary arteries. Virchows Arch A Pathol Anat. 1990;417:105-111.
Bratzler RL, Chisolm GM, Colton CK, Smith KA, Zilversmit DB, Lees RS. The distribution of labeled albumin across the rabbit thoracic aorta in vivo. Circ Res. 1977;40:182-190.
Nordestgaard BG, Zilversmit DB. Comparison of arterial intimal clearances of LDL from diabetic and nondiabetic cholesterol-fed rabbits: differences in arterial intimal clearance explained by size differences. Arteriosclerosis. 1989;9:176-183.
Slater HR, Packard CJ, Bicker S, Shepard J. Effects of cholestyramine on receptor-mediated plasma clearance and tissue uptake of human low density lipoproteins in the rabbit. J Biol Chem. 1980;255:10210-10213.
Higgins JA, Munro L, Wilkinson J. The rabbit as a model for studies of lipoprotein(a) synthesis and assembly. Proc Second International Conference on Lipoprotein(a). New Orleans, La. 1992:31. Abstract.
Rath M, Pauling L. Immunological evidence for the accumulation of lipoprotein(a) in the atherosclerotic lesion of the hypoascorbemic guinea pig. Proc Natl Acad Sci U S A. 1990;87:9388-9390.
Laplaud PM, Beaubatie L, Rall SCJ, Luc G, Saboureau M. Lipoprotein[a] is the major apoB-containing lipoprotein in the plasma of a hibernator, the hedgehog (Erinaceus europaeus). J Lipid Res. 1988;29:1157-1170.
Krempler F, Kostner GM, Roscher A, Haslauer F, Bolzano K. Studies on the role of specific cell surface receptors in the removal of lipoprotein (a) in man. J Clin Invest. 1983;71:1431-1441.
Mbewu AD, Bhatnagar D, Durrington PN, Hunt L, Ishola M, Arrol S, Mackness M, Lockley P, Miller JP. Serum lipoprotein(a) in patients heterozygous for familial hypercholesterolemia, their relatives, and unrelated control populations. Arterioscler Thromb. 1991;11:940-946.
Kovanen PT, Brown MS, Basu SK, Bilheimer DW, Goldstein JL. Saturation and suppression of hepatic lipoprotein receptors: a mechanism for the hypercholesterolemia of cholesterol-fed rabbits. Proc Natl Acad Sci U S A. 1981;78:1396-1400.
Rader DJ, Mann WA, Cain W, Kraft HG, Usher D, Zech LA, Hoek JB, Davignon J, Lupien P, Grossmann M, Wilson JM, Brewer HBJ. The low density lipoprotein receptor is not required for normal catabolism of Lp(a) in humans. J Clin Invest. 1995;95:1403-1408.
Kreutzer J, Lloyd MB, Bok D, Fless GM, Scanu AM, Lusis AJ, Haberland ME. Lipoprotein(a) displays increased accumulation compared with low-density lipoprotein in the murine arterial wall. Chem Phys Lipids. 1994;67/68:173-190.
Vasile E, Simionescu M, Simionescu N. Visualization of the binding, endocytosis and transcytosis of low density lipoprotein in the arterial endothelium in situ. J Cell Biol. 1983;96:1677-1689.
Wiklund O, Carew TE, Steinberg D. Role of the low density lipoprotein receptor in penetration of low density lipoprotein into rabbit aortic wall. Arteriosclerosis. 1985;5:135-141.
Nordestgaard BG, Tybjærg-Hansen A, Lewis B. Influx in vivo of low density, intermediate density, and very low density lipoproteins into aortic intimas of genetically hyperlipidemic rabbits: roles of plasma concentration, extent of aortic lesion, and lipoprotein particle size as determinants. Arterioscler Thromb. 1992;12:6-18.
Stender S, Zilversmit DB. Transfer of plasma lipoprotein components and of plasma proteins into aortas of cholesterol-fed rabbits: molecular size as a determinant of plasma lipoprotein influx. Arteriosclerosis. 1981;1:38-49.
Nordestgaard BG, Zilversmit DB. Large lipoproteins are excluded from the arterial wall in diabetic cholesterol-fed rabbits. J Lipid Res. 1988;29:1491-1500.
Fless GM, ZumMallen ME, Scanu AM. Isolation of apolipoprotein(a) from lipoprotein(a). J Lipid Res. 1985;26:1224-1229.