Articles |
From the Institute for Clinical Chemistry, University Hospital Großhadern, Munich, FRG (J.T., D.T., D.S.); the Department of Experimental Animal Research, University Hospital Göttingen, Göttingen, FRG (K.N.); Central Institute for Laboratory Animal Breeding, Hannover, FRG (K.R.); and the Institute for Laboratory Animal Science, University Hospital, Aachen, FRG (R.K.).
Correspondence to Joachim Thiery, Institute for Clinical Chemistry, University Hospital Großhadern, Marchioninistr 15, D-81366 Munich, FRG.
| Abstract |
|---|
|
|
|---|
Key Words: cholesterol-fed rabbit atherosclerotic response hypercholesterolemia protein polymorphisms breeding
| Introduction |
|---|
|
|
|---|
In addition to environmental influences, this variability is considered to be genetically controlled. In several prospective studies, the role of family history in early-onset coronary artery disease has been established as an independent coronary risk factor.13 17 18 19 20 The basis of such variability is thought to lie in inherited risk factors such as the concentration of plasma lipoprotein(a),13 the polymorphism of apolipoprotein E,16 and/or the biological response of the arterial wall in the presence of a certain plasma cholesterol concentration.15 In addition, oxidation of lipoproteins by endothelial cells and monocyte-derived macrophages appears to enhance their atherogenic activity and may contribute to the phenotypic variation of coronary artery disease.21
Given its multifactorial nature, the mode of inheritance of susceptibility or resistance to atherosclerosis is probably polygenic, thus rendering it difficult to study in humans. In animals, a genetically determined high and low atherosclerotic response to hypercholesterolemia was described in pigeons22 and in some strains of inbred mice.23 In the hypercholesterolemic rabbit, the classic animal model for the induction of arterial lesions,24 variability in the atherosclerotic response of the aorta to hypercholesterolemia was observed.25 26 However, except for a rabbit colony resistant to diet-induced hypercholesterolemia,27 the rare finding of low atherosclerotic response in cholesterol-fed rabbits has not been studied in detail.
In this report we describe the development of two rabbit strains, HAR (high atherosclerotic response) and LAR (low atherosclerotic response), and their propensities to develop atherosclerosis in the aorta despite similar levels of diet-induced hypercholesterolemia.
| Methods |
|---|
|
|
|---|
Breeding and Stock Selection of HAR and LAR Rabbits
Sixty-two pairs of New Zealand White rabbits (parent generation,
F0) were obtained from 13 breeders. After random mating,
the male rabbits were fed the cholesterol-supplemented
diet for 84 days. The offspring were divided into two groups according
to the atherosclerotic response of their cholesterol-fed
fathers. Progeny of male rabbits that exhibited an atherosclerotic
response of more than 70% aortic lesion coverage were designated as
HAR rabbits, while progeny of male rabbits with less than 30% aortic
lesions were designated as LAR rabbits (Fig 1
). The
rabbits were mated within the HAR and LAR colonies. Thereafter, the
males were again examined for development of
atherosclerosis by feeding 0.5 g
cholesterol per day for 84 days. After determination of the
extent of the aortic lesions, the breeding and cholesterol
feeding were performed in the same manner for four subsequent
generations.
|
To establish two strains of rabbits with low and high response, respectively, we selected in the 4th generation 9 nonrelated pairs of HAR rabbits and 10 nonrelated pairs of LAR rabbits. Because of limited husbandry capacity, only 10 to 15 pairs of rabbits per line and generation could be housed at one time. A computer program was developed to reduce the disadvantageous effects of inbreeding. The parentage of the animals was stored, and by iteration calculation the computer searched for the pairing partner with the lowest expected inbreeding coefficient. Rabbits with the least number of common progenitors were used for further mating, thereby minimizing the increase in homozygotes in each strain.28 29 30
To increase the separation of both strains in the subsequent generations, LAR rabbits were fed the cholesterol diet for 112 days, whereas HAR rabbits were fed the cholesterol diet for 84 days only. Offspring of male rabbits with an atherosclerotic response of the aorta between 30% and 70% were not used for further breeding. This selection and mating of offspring according to the atherosclerotic response of their fathers was continued for the subsequent 5 generations. In the 10th generation, the extent of atherosclerosis development was examined by feeding both strains the cholesterol diet for 112 days.
Planimetry of the Aorta and Classification of the
Atherosclerotic Response
The rabbits were injected with the anticoagulant heparin (500
IU/kg) and given an overdosage of sodium pentobarbital (100 mg/kg). The
aorta from the aortic valve to the iliac bifurcation was removed, freed
of adventitial tissue, opened longitudinally, pinned flat, and fixed in
4% PBS-buffered formalin. The fixed aortas were photographed, stained
with Sudan III, and photographed again.31 The individual
photographs were enlarged to twice the original size of the specimen.
The micrograph was affixed to a digitizing pad, and the area of the
aortic surface was traced (MOP-Videoplan Kontron). An atherosclerotic
lesion was defined as any flat or raised area with a defined border.
The percentage of lesion coverage was calculated as the sum of the
total lesion area divided by the total aortic area. The percentage of
lesion coverage was measured separately in the aortic arch and the
thoracic and abdominal aorta. Aortas from rabbits in the 5th, 6th, and
7th generations were divided longitudinally. One part of the aorta was
stained as described and the other part was used for
cholesterol determination.
Arterial Concentration of
Cholesterol
Longitudinal segments of the total aorta derived from HAR and
LAR rabbits were assayed for their total cholesterol
content. The adventitia was stripped, and the aortic tissue was frozen
in liquid nitrogen and pulverized by grinding. The wet weight of the
tissue was determined, and 14C-cholesterol
oleate (0.53 µCi/mL) was added as an internal standard (Amersham).
The lipids were then extracted in chloroform/methanol (2:1) by the
procedure of Folch et al.32 The chloroform extracts were
washed with 0.034% MgSO4 and dried under nitrogen. The
lipid residue was redissolved in 0.1 mL propanol and 0.4 mL methanol,
and total cholesterol was measured using a standard
enzymatic technique (Boehringer-Mannheim). The recovery of
14C radioactivity as determined in a scintillation counter
was 70% to 80%. The values were corrected for the internal standard
and expressed as milligrams per gram of tissue wet weight.
Plasma Lipoproteins and Biochemical Determinations
Plasma cholesterol and triglycerides
were determined at baseline, after 21 and 42 days of
cholesterol feeding, and at the time the animals were
killed. Blood was drawn in the nonfasting state with a 20-gauge Teflon
catheter introduced into the central artery of the ear (Vasofix, B.
Braun) into tubes containing disodium EDTA to a final concentration of
1 mg/mL. The plasma lipoprotein cholesterol concentrations
were measured on individual samples obtained at baseline and at
necropsy. VLDL (d<1.006 g/mL) and IDL+LDL
(d=1.006 to 1.063 g/mL) were isolated from rabbit plasma by
sequential ultracentrifugation in a fixed-angle rotor
in a Beckmann TL-100 ultracentrifuge.33 34 HDL
cholesterol was measured after precipitation with
phosphotungstate in the 1.006-g/mL infranatant by standard enzymatic
procedures (Boehringer-Mannheim). Total
cholesterol, cholesterol ester,
triglycerides, and phospholipids were measured by standard
enzymatic techniques (Boehringer-Mannheim). The distribution of
lipids and protein in the plasma lipoprotein fractions (chylomicrons,
VLDL, IDL, LDL, and HDL) was determined in pooled plasma samples of 5
HAR and 5 LAR rabbits before the start of the diet and at the time the
animals were killed according to the sequential
ultracentrifugation technique of Havel et
al33 in a Beckmann L-55 ultracentrifuge. The
protein concentration of the lipoprotein fractions was determined
according to Lowry et al.35 Clinical chemical evaluations
of plasma enzymes and substrates were performed at baseline and at the
time the animals were killed with standard methods.36
Genetic Markers in Plasma and Erythrocytes
Plasma and erythrocytes from 59 rabbits with high
atherosclerotic response and 61 rabbits with low atherosclerotic
response were tested for inherited protein polymorphisms in order
to reveal the degree of genetic diversity between the two rabbit lines
(HAR and LAR) at the protein level.
Blood was stabilized in EDTA obtained from the ear artery as described. Plasma and red cells were separated by centrifugation at 4°C for 10 minutes at 4000g, and the samples were stored at -80°C until analysis. Before electrophoresis was performed, the erythrocytes were diluted 1:2 (vol/vol) with bidistilled water and vortexed. After centrifugation for 10 minutes at 10 000g at 4°C, the supernatants were used for electrophoresis.
Electrophoresis was performed using both 2-mm starch gels (13% hydrolyzed starch) (STAGE) and ultrathin (0.1-mm) polyacrylamide gels (PAGE). The latter were run in a multiphor II chamber (LKB). For STAGE, the equipment used was as described by Geldermann.37 The gel buffer consisted of 0.1 mol/L Tris/citric acid, pH 8.8, and the electrode buffer was composed of 0.2 mol/L boric acid/lithium hydroxide, pH 8.6.38 The running time was about 3 hours under a field strength of 6 V/cm. The systems were cooled to 4°C by circulating water. Staining was carried out using 1% amido black for the unspecific proteins, while the substrate/coenzyme/dye reaction was used to demonstrate the specific enzyme phenotypes. All chemicals were obtained from Sigma. The phenotypes were classified according to the migration rate of the respective electrophoretic bands. The esterase typing was based on methods of Van Zutphen39 and Schiff and Stormont.40 According to the monogenetic inheritance of the polymorphic proteins and the presence of two different alleles at each of the responsible loci, the respective allele frequencies were derived from the phenotype frequencies using the formula a2+2ab+b2, where a and b are the allele frequencies of a given locus.41
Statistical Analysis
Probability values were calculated by Student's t
test.
| Results |
|---|
|
|
|---|
|
Plasma Lipid Concentration and Lipoprotein Composition in Normal
and Cholesterol-Fed HAR and LAR Rabbits
The plasma cholesterol concentrations of
cholesterol-fed HAR and LAR rabbits showed a 13-fold
increase after 21 days and a 21-fold increase after 84 days on the
cholesterol-enriched diet. No further increase in plasma
cholesterol was seen when the feeding period was extended
to 112 days. The development of diet-induced
hypercholesterolemia was similar in both rabbit
strains. Before induction of hypercholesterolemia, the
plasma cholesterol concentration (mean±SD) was 52±23
mg/dL for the HAR rabbits and 57±25 mg/dL for the LAR rabbits. At the
time the animals were killed, the plasma concentrations in the HAR and
LAR rabbits were 1241±489 mg/dL and 1370±473 mg/dL, respectively (Fig 3
). The total exposure to
hypercholesterolemia was assessed by plotting all the
values of plasma cholesterol (Fig 3
) as a function of time,
and the area under the curve was measured as 39.2 cm2 for
the HAR rabbits and 51.1 cm2 for the LAR rabbits. VLDL
cholesterol increased significantly from 12±11 mg/dL to
742±383 mg/dL in the HAR rabbits and from 17±15 mg/dL to 817±377
mg/dL in the LAR rabbits. There was a comparable effect of
cholesterol feeding on the plasma IDL and LDL levels. The
IDL+LDL cholesterol concentration increased from 23±16 to
453±163 mg/dL in the HAR rabbits and from 20±16 to 506±198 mg/dL in
the LAR rabbits. HDL cholesterol concentrations showed no
significant changes during the cholesterol feeding period
(Table 1
).
|
|
The percentage of cholesterol, triglycerides,
phospholipids, and protein in the various lipoprotein fractions derived
from the plasma of HAR and LAR rabbits before and after
cholesterol diet are presented in Table 2
. We did observe the expected enhancement of
cholesterol in the chylomicron, VLDL, IDL, and LDL density
fraction, but there were no significant differences in the lipid and
protein composition of lipoproteins between the two rabbit strains.
|
Extent of Atherosclerotic Involvement in
Cholesterol-Fed HAR and LAR Rabbits
A representative photograph of the macroscopic
appearance of atherosclerotic lesions covering the aorta of HAR and LAR
rabbits after cholesterol feeding for 112 days (10th
generation) is shown in Fig 4
. After Sudan staining of
the fixed, opened aortas, it could be clearly demonstrated that aortas
from LAR rabbits exhibited a significantly lower percentage of surface
covered with lesions in the thoracic and abdominal part as compared
with the same regions of the aorta from HAR rabbits. The heritability
of the low atherosclerotic response of the aorta to
cholesterol feeding in LAR rabbits was consistent
for the next generations (Fig 5a
). The extent of aortic
atherosclerosis in the LAR rabbit generations decreased
significantly from 57±25% (n=62, parental generation) to a mean of
27±17% in the 5th to 9th generations (n=82) and to 14±8% in the
10th generation (n=11). In contrast, HAR rabbits exhibited a mean
atherosclerotic response of the aorta of 58±21% in the 5th to 9th
generations (n=80), which was not significantly different from the
findings in the parental generation (Fig 5b
). However, when HAR rabbits
of the 10th generation (n=12) were fed the cholesterol diet
for the same time as LAR rabbits (112 days), an increase of
atherosclerosis of the aortic area to 85±24% was
seen, which was significantly different from the low atherosclerotic
response in the LAR rabbits (P<.001). In the ascending
aorta and aortic arch, atherosclerotic lesions were found in both HAR
and LAR strains. HAR rabbits developed atherosclerotic lesions also in
the thoracic and abdominal aorta, whereas in LAR rabbits these regions
showed little or no atherosclerotic response (Fig 6
).
These macroscopic findings were supported by the results of
cholesterol determinations in the aorta derived from
cholesterol-fed HAR and LAR rabbits (Table 3
).
|
|
|
|
For logistic reasons, the responsiveness of the rabbit aorta was studied for 9 generations only in male rabbits. However, female HAR and LAR rabbits of the 10th generation were also fed the cholesterol diet for 112 days. We found the same characteristic differences of the inherited atherosclerosis responsiveness between female HAR and LAR rabbits as we did in the cholesterol-fed male rabbits (data not shown). Therefore, a sex-linked susceptibility of LAR and HAR rabbits can be excluded.
Biochemical Markers for the Selection of the HAR and LAR
Rabbit Strains
Fifty-nine animals of the HAR rabbit strain and 61 animals of the
LAR rabbit strain were investigated for biochemical markers to
differentiate between the two rabbit strains. Protein
phenotypes of 20 different gene loci were typed by applying
electrophoretic separation techniques. Six of the electrophoretically
tested proteins were shown to be polymorphic, expressing genetic
variants that are controlled by two alleles each: an unspecific
postalbumin protein (PA) and two esterases (EST-2, EST-4) from
plasma and the 6-phosphogluconate dehydrogenase (PGD) and two
additional esterases (ES-1, ES-2) from erythrocytes. The calculated
allele frequencies of the corresponding loci are given in Table 4
for both rabbit lines. The alleles of the six
markers segregate in both rabbit strains with significant differences
at the Es-1 and the Pgd loci.
|
| Discussion |
|---|
|
|
|---|
In humans heterozygous for familial hypercholesterolemia and presenting high serum LDL cholesterol concentrations, the severity of the disease in terms of cardiovascular symptoms varies and is not necessarily correlated with the level of cholesterol in plasma.14 16 42 Even in patients with homozygous LDL receptor deficiency, death from coronary atherosclerosis can occur at any age between 5 and 30 years.43 44 45 The extent to which the familial occurrence of coronary heart disease is due to genetic mechanisms was recently assessed in twin and adoption studies.20 46
Schwenke and Carew47 48 analyzed the focal increase of arterial LDL concentration and the LDL permeability in several sites of the aorta in cholesterol-fed rabbits. Whereas permeability to LDL did not increase in any aortic site during 16 days of cholesterol feeding, they found an increased LDL retention and diminished fractional rates of LDL degradation in precisely those sites in the rabbit aorta that are most prone to early atherosclerotic lesions. These changes occurred before significant accumulation of subendothelial macrophage foam cells were detectable. It can be speculated that in LAR rabbits with low atherosclerotic response, LDL and ß-VLDL are degraded in the arterial wall to a higher extent than in HAR rabbits.
There are two other experimental animal models, namely pigeons and inbred mice, with a genetically determined atherosclerotic response to hypercholesterolemia. White Carneau pigeons develop aortic atherosclerosis naturally and at an accelerated rate with cholesterol feeding, whereas Show Racer pigeons are resistant to atherosclerosis, but no differences have been found in the levels of traditional risk factors.22 Yancey and St Clair49 showed that cholesteryl ester clearance from cholesteryl esterloaded pigeon macrophages in the presence of HDL/phosphatidylcholine is significantly less than in macrophages derived from White Carneau pigeons than in Show Racer pigeons.
The possible underlying causes for the different atherosclerotic response in inbred mice strains have also been studied.50 In the murine animal model of genetically determined atherosclerosis, A/J inbred mice show resistance to diet-induced atherosclerosis, whereas C57BL/6J inbred mice developed fatty streak lesions in the region of their aortic sinus valves when fed an atherogenic diet.23 51 Several structural genes for lipoproteins, apolipoproteins, and lipoprotein lipase have recently been identified that may play a role in the determination of the murine susceptibility to atherosclerosis,50 52 but there are no comparable data available in rabbits.
Our strains of rabbits with low and high atherosclerotic response must be differentiated from a rabbit colony described by Overturf et al,27 which is known to exhibit unusual resistance to induction of hypercholesterolemia by dietary cholesterol. These rabbits showed a normocholesterolemic response when they were fed a 0.1% cholesterolsupplemented diet for 7 months and were classified as "cholesterol-resistant" or "resistant rabbits." An enhanced LDL catabolic rate in skin fibroblasts and an increased bile acid secretion were observed in these cholesterol-resistant rabbits. There was no consistent difference between the plasma concentrations and the composition of lipoproteins between the cholesterol-resistant and normal rabbits consuming regular chow.27 53 We found comparable results in the plasma lipoprotein pattern of the HAR and LAR rabbits maintained on a normal diet. No significant differences in the plasma lipoprotein concentrations or their composition were observed between strains. There was a trend in the composition of all of the lipoproteins such that there is more cholesterol relative to protein, phospholipids, or triacylglycerol in the HAR rabbits than in the LAR rabbits. However, these small variations are unlikely to account for the observed differences in the extent of atherosclerosis between the HAR and LAR rabbit strains.
In contrast to the findings in cholesterol-resistant rabbits, LAR rabbits fed a cholesterol-supplemented diet developed severe hypercholesterolemia. Receptor-mediated degradation of LDL by skin fibroblasts derived from LAR and HAR rabbits was normal (data not shown). Unlike the cholesterol resistant rabbit colony,27 our principal finding was that the diet-induced hypercholesterolemia was comparable among both the LAR and the HAR rabbits, but there was a significantly different phenotypic atherosclerotic response in the two strains.
| Acknowledgments |
|---|
Received March 1, 1995; accepted May 5, 1995.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Teupser, R. Burkhardt, W. Wilfert, I. Haffner, K. Nebendahl, and J. Thiery Identification of Macrophage Arginase I as a New Candidate Gene of Atherosclerosis Resistance Arterioscler. Thromb. Vasc. Biol., February 1, 2006; 26(2): 365 - 371. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Teupser, M. Tan, A. D. Persky, and J. L. Breslow Atherosclerosis quantitative trait loci are sex- and lineage-dependent in an intercross of C57BL/6 and FVB/N low-density lipoprotein receptor-/- mice PNAS, January 3, 2006; 103(1): 123 - 128. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Ruschitzka, U. Moehrlen, T. Quaschning, M. Lachat, G. Noll, S. Shaw, Z. Yang, D. Teupser, T. Subkowski, M. I. Turina, et al. Tissue Endothelin-Converting Enzyme Activity Correlates With Cardiovascular Risk Factors in Coronary Artery Disease Circulation, September 5, 2000; 102(10): 1086 - 1092. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Teupser, O. Stein, R. Burkhardt, K. Nebendahl, Y. Stein, and J. Thiery Scavenger Receptor Activity Is Increased in Macrophages From Rabbits With Low Atherosclerotic Response: : Studies in Normocholesterolemic High and Low Atherosclerotic Response Rabbits Arterioscler. Thromb. Vasc. Biol., May 1, 1999; 19(5): 1299 - 1305. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |