Full text of DOI:10.1007/s11745-001-0742-4

ARTICLES
Characterization of Simple and Reproducible Vascular
Stenosis Model in Hypercholesterolemic Hamsters
Hiroyuki Matsunoa,*, Osamu Kozawaa, Masayuki Niwaa, Akira Abeb,
Yoshiharu Takiguchic, and Toshihiko Uematsua
aDepartment of Pharmacology, Gifu University School of Medicine, Gifu 500, Japan, bDepartment of Clinical Chemistry,
Gifu College of Medical Technology, Seki 501-3822, Japan, and cDepartment of Clinical Pharmacology,
Graduate School of Pharmaceutical Sciences, The University of Tokushima, Tokushima 770, Japan
ABSTRACT: The importance of low-density lipoprotein (LDL) in shown that LDL stimulates endothelial cells to produce a vari-
the etiology of atherosclerosis is well recognized. We have estab-
lished a reproducible stenosis model in hypercholesterolemic
hamsters, and the process of arterial stenosis by thrombus or
neointima was studied and compared with that in normal ham-
sters. The level of plasma LDL was 4.6 times higher in hamsters
fed a high-cholesterol diet than in hamsters fed normal food. En-
dothelial injury in right common carotid arteries was induced
using a modified catheter. Arterial blood flow was monitored con-
tinuously using a Doppler flow probe. Arterial patency after the
initiation of injury in high-cholesterol hamsters was significantly
changed as compared with that of normal hamsters. Neointima
ety of growth factors that induce the migration and prolifera-
tion of granulocytes, macrophages, and endothelial cells (2,3).
These studies raise the possibility that increases in LDL con-
tent stimulate the production of factors that play an important
role in the proliferation of cells and matrix elements found in
both atherosclerosis and vascular neointima formation.
On the other hand, both platelets and LDL are intimately in-
volved in the pathogenesis of atherosclerosis (2). Platelet func-
tion is directly influenced by lipoproteins, and platelets from
patients with hypercholesterolemia display enhanced platelet
was observed 2 wk after the vascular injury. The neointimal area reactivity (4). Each platelet processes specific high-affinity
of high-cholesterol hamsters was significantly larger than that of
normal hamsters. To characterize the stenosis in hypercholes-
terolemic hamsters, we measured platelet aggregation, thrombin
time, activated partial thromboplastin time, and proliferating
smooth muscle cells (SMC) in vitro and in vivo. The half-maximal
inhibitory concentration value for platelet aggregation induced by
thrombin or collagen, the DNA synthesis stimulated by platelet-
derived growth factor (PDGF)-BB, and 5-bromo-2-deoxy-uridine
labeling indices (proliferating index of SMC in vivo) in high-cho-
lesterol hamsters were each significantly higher than the compara-
ble value from normal hamsters. However, specific binding of
binding sites for LDL, ranging from 1000 to 8000 copies per
platelet with a dissociation constant (KD) value between 40 and
100 nmol/L (5,6). Moreover, the LDL-induced sensitization is
accompanied by decreased angular movement in platelet mem-
branes, possibly caused by increased cholesterol transfer (7),
and platelets enriched with cholesterol show increased arachi-
donic acid release and thromboxane B2 formation (8).
Until now, rabbits fed a high-cholesterol diet or Watanabe
rabbits (9) have been the main animals used for the investiga-
tion of hypercholesterolemia or atherosclerotic lesions of vas-
PDGF-BB in SMC was not different between the two types of ham- cular diseases. Other small animals that can be used for this
sters. Furthermore, we investigated the inhibitory effects of probu-
col or losartan on neointima formation using this model. Probu-
col, but not losartan, significantly reduced the neointimal area in
hypercholesterolemic hamsters. These findings indicated that high
levels of plasma LDL strongly contributed to the development of
thrombus and neointima formation via both up-regulation of
platelet aggregation and the enhancement of SMC proliferation.
This stenosis model may be useful for the investigation of hyperc-
holesterolemia-associated cardiovascular diseases.
type of study are exogenously hypercholesterolemic rats and
hamsters. In rats, a few months are required to establish
hypercholesterolemia after the start of a high-cholesterol diet
(10,11). On the contrary, only a few weeks are necessary for
hamsters to be hypercholesterolemic (11). Recently, we es-
tablished a simple and reproducible vascular stenosis model
in hamsters (12). We investigated the inhibitory effects of an-
tiplatelet compounds (13–16), angiotensin-converting en-
zyme inhibitors (17), and an angiotensin II receptor antago-
nist (18) using this model. Vascular stenosis often accompa-
Paper no. L8606 in Lipids 36, 453–460 (May 2001).
Hypercholesterolemia increases the risk of developing cardio- nies injury to endothelial tissues. Progression of stenosis
vascular diseases, and low density lipoprotein (LDL) plays an occurs by thrombus formation during the acute phase or
important role in atherosclerosis (1). Recent studies have neointima formation during the chronic phase after endothe-
lial injury. Our previous observations clearly indicated that
*To whom correspondence should be addressed at Department of Pharma-
the inhibition of both platelet activation and smooth muscle
cell (SMC) proliferation or migration strongly reduced the de-
velopment of vascular stenosis after endothelial injury (13,
18). On the other hand, the likelihood of vascular occlusion
in patients is enhanced by the additive effect of other risk fac-
tors, such as diabetes and hypercholesterolemia. Therefore,
in the present study, this model was applied to examine the
effect of plasma LDL levels on platelets and SMC in order to
cology, Gifu University School of Medicine, Tsukasa-machi 40, Gifu 500,
Japan. E-mail: leuven@cc.gifu-u.ac.jp
Abbreviations: aPTT, activated partial thromboplastin time; BrdU, 5-bromo-
2-deoxy-uridine; DAB, diaminobenzidine; DMEM, Dulbecco’s modified Ea-
gle’s medium; FCS, fetal calf serum; HDL, high density lipoprotein; IA, inti-
mal area; IC50, half-maximally inhibitory concentration; IELA, internal elas-
tic lamina area; LA, lumen area; LDL, low density lipoprotein; PDGF,
platelet-derived growth factor; PRP, platelet rich plasma; SEM, scanning
electron microscopy; SMC, smooth muscle cell; TEM, transmission electron
microscopy; TT, thrombin time.
Copyright © 2001 by AOCS Press
453
Lipids, Vol. 36, no. 5 (2001)

454
H. MATSUNO ET AL.
define easily determined, clinically significant measures of
vascular stenosis progression.
Measurement of serum cholesterol. Blood samples (0.3
mL) from hamsters fed either a regular or a high-cholesterol
diet were collected in each time period via the jugular vein
after the animals had been anesthetized with ether. The sam-
ples were treated with 3.15% sodium citrate. High density
MATERIALS AND METHODS
Animals. Male syrian hamsters (SLC, Sizuoka, Japan) weigh- lipoprotein cholesterol (HDL) and LDL cholesterol were de-
ing 60–80 g were selected and fed either a standard chow termined by enzymatic assays of blood samples taken at 0
(RC4; Oriental Yeast Co., Ltd., Japan) or chow supplemented (before the start of diet), 2, 4, and 6 wk after the start of feed-
with 0.5% (w/w) cholesterol. An operation for vascular injury ing regimens These measurement were performed with Nipro
using a modified catheter was carried out 4 wk after the start of cholesterol determination kits (Nipro Co. Ltd., Osaka, Japan).
the diet regimen. The body weight of hamsters fed a high-cho-
Platelet aggregation. At the end of the experiments, 4.0 mL
lesterol diet was not statistically significantly different from of blood was collected from every hamster in each group by
that of hamsters fed a normal diet (body weight: 120–130 g). heart puncture into sodium citrate (3.15%, final concentration)
All experiments were performed in accordance with institu- and centrifuged for 10 min at 155 × g to obtain platelet-rich
tional guidelines governing animal experimentation.
plasma (PRP). Platelets were counted and adjusted to 4 × 108
Reagents. Collagen for platelet aggregation was obtained cells/mL (final concentration), and then platelet aggregation was
from Nycomed Arzneimittela GmbH (Munich, Germany). induced by collagen (1.0–10.0 µg/mL) or ADP (0.5–5.0 µM)
[Methyl-3H] thymidine and [125I] platelet-derived growth fac- using PRP. Washed hamster platelets were prepared as described
tor (PDGF)-BB were purchased from Amersham Japan previously (19). Washed platelet aggregation was induced by
(Tokyo, Japan). Probucol and losartan are a kind gift from thrombin (0.0001– 0.3 unit/mL). Platelet aggregation was fol-
Otuka Co. Ltd. (Tokushima, Japan) and Banyu Co. Ltd. lowed using an aggregometer (Aggrecorder II, DA-3220; Ky-
(Tokyo, Japan), respectively. The other chemicals were ob- otodaiichi-Chemical, Kyoto, Japan) at 37°C with a stirring speed
tained from Sigma (St. Louis, MO).
of 800 rpm. All measurements were performed in triplicate.
Induction of vascular stenosis and quantitation of neointima
Ex vivo anticoagulant studies. After the separation of PRP
formation. Four weeks after the start of either the regular or the for platelet aggregation, the remaining blood samples were
high-cholesterol dietary regimen, hamsters were anesthetized further centrifuged for 10 min at 1550 × g to obtain platelet-
by intraperitonial injection of 50 mg/kg sodium pentobarbital. poor plasma (PPP). The activated partial thromboplastin time
In brief, the right common carotid artery and the right femoral (aPTT) and thrombin time (TT) were determined using stan-
artery were exposed, and an arterial injury to the right common dard clinical laboratory procedures.
carotid artery using a 2FG catheter (Portex Ltd., Kent, United
Electron microscopic observation. In separate experiments,
Kingdom) with a roughened tip was performed according to several samples taken from hamsters fed a high-cholesterol or
the previously described technique (12). Another catheter (i.d. normal diet were used for histological observations by means
= 0.5 mm, o.d. = 0.8 mm, polyethylene sp3, Natume Co. Ltd., of scanning or transmission electron microscopy. Hamsters
Tokyo, Japan) was connected to the right femoral artery for were anesthetized and perfused transcardially with saline 2 wk
monitoring blood pressure and pulse rate using a pressure after the initiation of vascular injury. The common carotid
transducer (AP601G; Nihon Koden, Tokyo, Japan) during ex- artery was excised and fixed with 2.0% glutaraldehyde in 50
periments on day 0 (initiation of vascular injury). Blood flow mM sodium phosphate buffer for 30 min. Each segment was
in the carotid artery was continuously monitored for 90 min on cut open longitudinally to allow visual inspection for scanning
day 0 using a Doppler flow probe (model PDV-20; Crystal electron microscopy (SEM) or cross-sectioned for transmis-
Biotech Co. Ltd., Tokyo, Japan) positioned proximally to the sion electron microscopy (TEM) as described (12).
injured area of the carotid artery. Our previous histological ob-
Proliferation index of SMC in vivo. Proliferating SMC
servations revealed that a platelet-rich thrombus was obviously were identified by in vivo DNA labeling with the thymidine
established when the blood flow was zero (12). After the re- analog 5-bromo-2-deoxy-uridine (BrdU) (12). BrdU (50
covery from anesthesia, animals were kept in individual cages. mg/kg) was injected subcutaneously 1, 8, 16, and 24 h prior to
Three, 5, 7, and 14 d after vascular injury, hamsters (n = 7 each) removal of the carotid artery. Following carotid artery removal
in each group were anesthetized and perfused transcardially 1, 3, 5, 7, or 14 d after injury from both control and choles-
with saline. The common carotid artery was excised, divided terol-supplemented hamsters (n = 4, each time point), frozen
into several sections, and frozen. The tissue samples were cross sections were prepared from these arteries. BrdU-posi-
cross-sectioned, and stained with hematoxylin (Sigma Chemi- tive cells were detected with a murine monoclonal antibody
cal Co.) to determine the intimal area. Part of these samples (Sigma), followed by treatment with goat antimouse im-
were used for the stain of lipids using oil red. The total areas of munoglobulin-antibodies conjugated to peroxidase. The com-
the internal elastic lamina (IELA) and lumen (LA) were mea- plexes were then stained with diaminobenzidine (DAB). The
sured using a computerized image graphic analysis system. The BrdU labeling index was calculated using the following for-
measurement was performed in triplicate for each sample. The mula: (nuclei stained positive by DAB)/(total nuclei stained
average of three intimal area (IA = IELA − LA) determinations by hematoxylin) × 100. Animals were sacrificed by an over-
was then expressed as a percentage of IELA.
dose of sodium pentobarbital at the end of the experiment.
Lipids, Vol. 36, no. 5 (2001)

LDL PROMOTE VASCULAR STENOSIS
455
Cell culture and measurement of DNA synthesis. Vascular Fourteen days after vascular injury, the common carotid
SMC were isolated from the thoracic aorta of hamsters fed ei- artery of hamsters in each group was treated as mentioned in
ther a normal or supplemented diet. The cells were cultured the paragraph on quantitation of neointima formation. At the
over several passages using the method of Ross (20). SMC end of observation period, blood pressure was monitored for
were grown to confluency in 7500-mm2 culture flasks in Dul- 10 min. A catheter (i.d. = 0.5 mm, o.d. = 0.8 mm, polyethyl-
becco’s modified Eagle’s medium [DMEM; Gibco BRL, Grand ene sp3, Natsume Co. Ltd.) connected to a pressure trans-
Island, NY; 5% fetal calf serum (FCS), 100 µg/mL strepto- ducer (AP601G; Nihon Koden) was inserted into the left
mycin, 100 U/mL penicillin, 4 µmol/L glutamine] at 37°C femoral artery for 10 min. After the measurement of blood
under humidified 5% CO2/95% air. The cells were then treated pressure, a blood sample was taken for platelet aggregation
with 0.25% trypsin in phosphate-buffered saline (pH = 7.4), and measurements of coagulation factors (TT and aPTT).
washed, and counted. A total of 8 × 104 isolated cells were cul-
Statistics. All data are presented as the mean SEM. The
tured as above. Cell culture medium was replaced every 2 d. statistical significance of the data was determined by analysis
On day 8, cells were detached with trypsin and counted. The of variance followed by the Student-Newman-Keuls test.
cultured cells were stimulated with 5% FCS or various doses
of angiotensin II (10−6–10−8 M), thrombin (0.001–0.03
RESULTS
units/mL), vasopressin (10−6–10−8 M), endothelin I (10−6–10−8
M), or PDGF-BB (1.0–100 ng/mL) in 1 mL of DMEM at 37°C Alteration of plasma cholesterol levels. The mean plasma LDL
for 24 h. Six hours before harvest, the cells were pulse-labeled level in all animals before beginning a specific dietary regimen
with [methyl-3H]thymidine (0.5 µCi/dish). Incubation was ter- was <0.5 g/L. After 4 wk of a high-cholesterol diet, the LDL
minated by adding 1 mL of 10% trichoroacetic acid, and ra- level had increased to ~2 to 2.5 g/L and remained elevated to
dioactivity in the acid-insoluble materials was determined by the end of the observation period. In contrast, there was no sig-
using a Beckman LS-6000IC liquid scintillation spectrometer.
nificant difference in plasma HDL levels between hamsters fed
[
125I]PDGF binding. Cultured cells were subjected to a normal and cholesterol-supplemented diets (Table 1).
binding assay that was essentially as described (21). Briefly,
Acute thrombus formation and vascular patency. Vascular
the cells were treated at 37ºC in 10 mM HEPES-buffered Dul- patency is shown in Table 2. Blood flow in the carotid artery
becco’s Eagle’s medium (1.0 mL, pH = 7.4; Gibco BRL). The in normal hamsters (n = 28) was interrupted in 6.4 0.6 min
cells were incubated with or without a 1,000-fold molar ex- after the initiation of vascular injury by a catheter. After the
cess of nonradioactive PDGF-BB for 20 min at 37ºC. At the blood flow was zero, spontaneous cyclic reflow and reocclu-
end of the incubation, the cells were thoroughly washed with sion were observed in 21 hamsters during the observation pe-
cold phosphate-buffered saline and solubilized using 0.1% riod of 90 min. The other arteries were not reperfused during
sodium dodecyl sulfate (1.0 mL). The radioactivity of the the observation period. In the group of high-cholesterol ham-
lysate was then determined using a Wallac 1480 WIZARD 3′″ sters, the time to occlusion was significantly shortened to 4.8
automatic gamma counter (Turku, Finland).
The effect of probucol or losartan on neointima. To further reocclusion and reflow during the observation period (n = 28).
define this stenosis model in pharmacological experiments, Neointima formation in response to endothelial injury. All
0.4 min and only four arteries showed spontaneous cyclic
we investigated the effect of probucol or losartan using this hamsters developed concentric intimal lesions in response to
model. Hamsters fed a high-cholesterol diet were divided into endothelial denudation by a catheter. The ratios of time-de-
seven groups, a control group (n = 10), three groups treated pendent vascular stenosis by neointima formation are shown
with probucol (twice a day, p.o.) at doses of 30.0, 60.0, or in Figure 1. In hamsters fed a high-cholesterol diet, the extent
120.0 mg/kg per day (n = 6 each), and three groups treated of vascular stenosis was significantly greater compared with
with losartan (twice a day, p.o.) at doses of 2.0, 6.0, or 20.0 that of hamsters fed a normal diet.
mg/kg per day (n = 6 each). Oral administration of each com-
Platelet aggregation and hemostasis analysis. The half-
pound was started 2 h before the initiation of endothelial in- maximally inhibitory concentration (IC50) values for platelets
jury by a modified catheter and continued for the next 2 wk. induced to aggregate by collagen, ADP, and thrombin in ham-
TABLE 1
Plasma Cholesterol Levels (g/L) in Hamstersa
Diet group
Normal group
LDL
HDL
LDL
HDL
Pretreatment
2 wk
0.50 0.07
1.12 0.41
1.92 0.31*
2.34 0.22*
2.33 0.29*
0.49 0.08
0.46 0.09
0.45 0.06
0.50 0.07
0.55 0.07
0.47 0.08
0.49 0.06
0.49 0.06
0.49 0.02
0.52 0.07
0.50 0.06
0.48 0.05
0.50 0.07
0.49 0.09
0.48 0.07
3 wk
4 wk
6 wk
aHamsters (4 wk old) were allotted to a high-cholesterol diet group or to the normal diet group. Pre-
treatment means before beginning a dietary regimen; each week means the time after the start of a
particular regimen. Vascular injury in each hamster occurred at 4 wk. *P < 0.05 vs. control (normal
hamsters in each time course).
Lipids, Vol. 36, no. 5 (2001)

456
H. MATSUNO ET AL.
TABLE 2
TABLE 3
Time to Occlusion and Vascular Patency Status in Hamstersa
Platelet Aggregation and Hemostatic Analysis of Hamstersa
Normal hamsters
High-cholesterol hamstersb
Normal hamster
High-cholesterol hamsters
Time to occlusion
(min)
PP
CR
PO
ADP
6.6 0.9 µM
7.8 1.1 µg/mL
0.061 0.1 unit/mL
15.4 1.1 s
4.9 1.9 µM
3.8 0.2 µg/mL*
0.002 0.0004 unit/mL**
13.6 2.3 s
6.4 0.8
4.8 0.4*
Collagen
Thrombin
TT
0
21
7
0
4**
24**
aPTT
58.2 6.8 s
52.2 4.8 s
aVascular patency was judged at the end of the observation period. Carotid
arterial patency was expressed as persistent occlusion (PO) when no reper-
fusion was observed at all, as cyclic flow reduction (CR) when the arterial
reflow alternately showed stops and flows, and persistent patency (PP) when
the arterial flow was maintained until the end of observation period. Data
correspond to the number of arteries in each group (n = 28 animals in each
group).
aThe values of the half-maximally inhibitory concentration for platelet ag-
gregation induced by ADP, collagen (PRP), or thrombin (washed platelets) in
hamsters fed a normal diet or a high-cholesterol diet. aPTT, activated partial
thromboplastin time; TT, thrombin time. Data represent mean SEM. *P <
0.05; **P < 0.01.
b*P < 0.05; **P < 0.01. Time to occlusion data represented mean SEM.
Proliferation of SMC in vivo. Figure 4 shows the percent-
age of proliferating SMC on days 1, 3, 5, 7, and 14 after vas-
cular injury. The increased level of plasma cholesterol caused
a significant increase in SMC proliferation, measured on days
1, 3, 5, and 7. These differences represented increases of 21.4,
27.3, 11.1, and 14.8% in proliferation index, respectively.
Proliferation index in vitro. The amount of DNA synthesis
induced by various agonists is shown in Figure 5A. DNA syn-
thesis is markedly increased in high-cholesterol hamsters
when SMC are stimulated by FCS or PDGF. Dose-dependent
alterations of DNA synthesis induced by PDGF are shown in
Figure 5B. DNA synthesis in response to PDGF-BB stimula-
tion of SMC in hamsters fed a high-cholesterol diet is signifi-
cantly higher than in hamsters fed a normal diet. When SMC
were stimulated by the other stimulation factors (thrombin,
vasopressin, endothelin I, or angiotensin II), DNA synthesis
was not changed in SMC from either group of hamsters.
sters fed a normal or a high-cholesterol diet are shown in
Table 3. When platelets were induced by thrombin or colla-
gen, the IC50 values in high-cholesterol hamsters decreased
compared with the values for hamsters fed a normal diet. In
terms of hemostatic values, no significant difference was ob-
served between normal and high-cholesterol hamsters. How-
ever, these values in high-cholesterol hamsters were slightly
shortened as compared with those from normal ones.
Histological observation. According to observations ob-
tained through TEM, foam cells and extracellular lipids in
neointima were clearly observed in hamsters fed a high-cho-
lesterol diet (Fig. 2). When neointima formation was estab-
lished 2 wk after vascular injury, the vascular surface in in-
jured area was not smooth; however, repaired endothelial
cells were completely covered with neointima. Blood ele-
ments were not observed on vascular surfaces in hamsters fed
a normal diet (Fig. 3A). On the other hand, platelets were lo-
cally adhered on neointimal vascular surfaces in hamsters fed
a high-cholesterol diet 2 wk after vascular injury even if en-
dothelial cells were recovered with the vascular surface in the
injured area (Figs. 3B,3C).
[
125I]PDGF binding in SMC. The binding affinities in both
groups of hamsters were unchanged (Fig. 6).
Effects of probucol or losartan on neointima. Figure 7
shows the inhibitory effects of probucol or losartan on neoin-
FIG. 1. The development of neointima in hamsters fed a normal or a
high-cholesterol diet shown as percentage of luminal stenosis. Vascular
injury occurred 4 wk after the start of regular (open circles) or high-cho-
lesterol diet (closed circles), and neointima was measured 3, 5, 7, or 14
d after the vascular injury. *P < 0.05; **P < 0.01. The error bars repre-
sent SEM.
FIG. 2. A transmission electron micrograph of neointima formation 14
d after vascular injury in hamsters fed a high-cholesterol diet. Internal
elastic lamina (IEL) can be clearly observed (small arrows) and lipid par-
ticles (large arrow) and foam cells (asterisk) are observed in media (dur-
ing IEL) and newly formed intima. The scale bar indicates 200 µm.
Lipids, Vol. 36, no. 5 (2001)

LDL PROMOTE VASCULAR STENOSIS
457
FIG. 4. Smooth muscle cell proliferation (n = 4, each time point) mea-
sured as the 5-bromo-2-deoxy-uridine (BrdU) index (%) following vas-
cular injury in hamsters fed a normal (ꢀ) or a high-cholesterol diet (ꢀ).
*P < 0.05; **P < 0.01 vs. control (normal hamsters). The error bars rep-
resent SEM.
FIG. 5. The amount of DNA synthesis induced by various agonists (A)
and dose-response line plot of DNA synthesis induced by platelet-de-
rived growth factor (PDGF)-BB (B). (A) The cultured cells from either
normally fed hamsters (open bars) or high-cholesterol-fed hamsters
FIG. 3. Scanning electron micrographs from hamsters fed a normal or a (hatched bars) were stimulated with 5% fetal calf serum (FCS), 10−6
M
high-cholesterol diet 14 d after the initiation of vascular injury. (A)
angiotensin II (AngII), 0.03 unit/mL thrombin (TR), 10−6 M vasopressin
Neointima formation on injured area. The luminal surface is entirely (VP), 10−6 M endothelin I (ET), or 30 ng/mL PDGF-BB (PDGF) in 1 mL
covered with newly formed endothelial cells, which are morphologi- of Dulbecco’s modified Eagle’s medium. Alteration of DNA synthesis in
cally different from native ones and irregularly oriented. Blood elements high-cholesterol hamsters shows a ratio vs. normal hamsters. (B) Dose-
are not observed in this area. (B) Locally activated platelets (a window) response line plot: DNA synthesis induced by PDGF-BB. Each plot pre-
and monocytes (arrows) are adhered on irregularly oriented vascular sents data from hamsters with normal (open circle) or high-cholesterol
surface. (C) A high magnification of a window of panel B. Adherent
platelets consist of microthrombus formation.
diets (closed circles). Each point represents the mean of duplicate cul-
tures.
Lipids, Vol. 36, no. 5 (2001)

458
H. MATSUNO ET AL.
FIG. 6. Specific binding of PDGF-BB using explanted vascular smooth
muscle cells from hamsters fed a normal or a high-cholesterol diet. Each
plot presents data from hamsters with normal (open circle) or high-cho-
lesterol diets (closed circles). Each point represents the mean of dupli-
cate cultures. For abbreviation see Figure 5.
tima formation. The treatment with probucol at a dose of 120
mg/kg/d significantly reduced the neointimal area in hyper-
cholesterolemic hamsters. However, losartan slightly de-
creased the neointimal area at the highest dose. Photomicro-
graphs of typical neointima formation 2 wk after arterial in-
jury of hamsters either treated with probucol at a dose of 120
mg/kg/d or not, are shown in Figure 8. Formation of neoin-
tima including a lot of lipids was observed in hamsters not
treated with probucol. On the contrary, neointimal areas and
points of lipid accumulation were clearly reduced by treatment
with probucol. Platelet aggregation and coagulation factors
were not different in hamsters fed a high-cholesterol diet com-
pared with those fed a normal diet. Blood pressure was slightly
decreased when the highest dose of losartan was used to treat
hamsters.
FIG. 8. Photomicrographs show neointima formation in damaged
carotid arteries of hypercholesterolemic hamsters either treated with
120.0 mg/kg/d probucol (B) or not treated (A). Lipids were stained bright
red. The tissue was counterstained with hematoxylin. A packed neoin-
tima was observed in both samples, however, the stenosis area in tissue
from animals treated with probucol was diminished and lipids in steno-
sis clearly decreased.
DISCUSSION
This study demonstrates that high plasma LDL levels in ham-
sters fed a high-cholesterol diet are associated with marked in-
creases in the development of thrombus and neointima forma-
tion, as compared to normally fed hamsters, in injured carotid
arteries, which we have established and used to demonstrate
the inhibitory effects of several kinds of compounds on vascu-
lar stenosis (12–18). Furthermore, these phenomena are mainly
characterized by the enhancement of platelet activation induced
by thrombin and of SMC proliferation induced by PDGF-BB.
A high arterial LDL level is an important factor in the de-
velopment of vascular stenosis. Several molecular mechanisms
have been suggested to account for the correlation of elevated
LDL serum levels with the development of chronic vascular
FIG. 7. This graph shows dose-dependent inhibition of neointima forma-
tion in hamsters fed a high-cholesterol diet by treatment with probucol
or losartan. The compound was administered orally twice a day from 2
h before until 14 d after carotid artery injury; control (n = 12) and treated
animals (n = 6) were analyzed at day 14. The ratio between neointimal
area and the area within the internal elastic lamina was determined on
three cross sections of carotid arteries. **P < 0.01 vs. control. The error
bars represent SEM.
Lipids, Vol. 36, no. 5 (2001)

LDL PROMOTE VASCULAR STENOSIS
459
diseases (22,23). In our experiments, the plasma LDL level of (32), or when platelets adhere to sites of injured blood vessels,
hamsters fed a high cholesterol diet (2.24 0.32 g/L) 2 wk after as seen in our findings using SEM. Therefore, up-regulation of
the start of the diet regime was 4.6 times higher than that of sensitivity in platelets in high-cholesterol hamsters also plays a
hamsters fed a normal diet (0.49 0.02 g/L), but the HDL role in the enhancement of neointima formation. Indeed, our
plasma level was not significantly different. These levels are previous findings indicated that platelets play a significant role
similar to those found in pathologic conditions, such as famil- in the development of neointima formation since antiplatelet
ial combined hyperlipidemia (24), nephrotic syndrome (25) compounds improved SMC proliferation during the acute
(LDL = 1.6 to 2.0 g/L), and especially homozygous familial phase of vascular injury (13,14).
hypercholesterolemia (LDL = 3.0 to 5.5 g/L) (26,27).
To define the utility of this model in pharmacological exper-
Vascular stenosis was related mainly to thrombus develop- iments, we used losartan and probucol to treat hamsters fed a
ment in the acute phase or to neointima formation in the high-cholesterol diet, because probucol inhibited aortic choles-
chronic phase after endothelial injury. In the present experi- terol accumulation (33) and reduced neointima formation (34)
ments, the development of thrombus formation of high-choles- and losartan also reduced neointima formation (35). In these
terol hamsters in vivo was clearly accelerated, because the time experiments, probucol significantly reduced the neointimal
to occlusion and vascular patency in high-cholesterol hamsters area and attenuated lipid formation in vascular SMC. Indeed,
were significantly changed as compared with those of normal foam cell lesions in hypercholesterolemic hamsters were de-
hamsters. This phenomenon was mainly attributable to up-reg- creased by treatment with probucol (33). On the contrary, losar-
ulation of platelet activation since IC50 platelet aggregation val- tan did not show significant effects on neointima formation
ues significantly decreased in high-cholesterol hamsters, but even though our previous study showed that losartan signifi-
coagulation factors were not different in both hamsters. Our cantly reduced neointima formation in normal hamsters (18).
previous data indicated that platelets play a major role in the These findings indicated that probucol greatly affects the de-
development of thombus formation since activated platelets ad- velopment of neointima formation promoted by hypercholes-
here to injured vascular surfaces in the days immediately after terolemia, but not the inhibition of angiotensin II by losartan.
injury (13,14). Moreover, LDL may induce a state of hypersen-
sitivity in the platelets that contributes to the high rate of throm- such responses to the development of neointima formation in
bosis formation observed in patients (7,28). hypercholesterolemia may be modeled as follows: First, after
Finally, we speculate that the physiological importance of
Our findings also showed that neointima formation in in- endothelial injury, activated platelets adhere to and aggregate
jured carotid arteries of hamsters fed a high cholesterol diet was at damaged endothelial surfaces. Next, these platelets are sen-
significantly greater than in hamsters fed a normal diet. The sitized by high levels of plasma LDL. The production of lo-
number of proliferating SMC in vivo measured at each time cally secreted PDGF by activated platelets increases prolifer-
point after vascular injury in hamsters fed a high-cholesterol ation of SMC in the region of the injury. The continuous pro-
diet was also significantly higher compared with hamsters fed duction of PDGF may play a very significant role in the
a normal diet. LDL particles, trapped in the extracellular ma- amplified development of neointima in hypercholesterolemia.
trix of the vessel wall, are well known to undergo substantial This is especially relevant since the other known stimulants
structural and chemical modification in response to many kinds (angiotensin II, vasopressin, or endothelin) did not increase
of stimuli (29). In our histological observations using TEM, DNA synthesis in SMC from high-cholesterol hamsters. This
lipid particles can be found in vascular neointima formation. model is a sensitive one for pharmacological experiments.
Furthermore, to define the findings of increased neointima le-
In conclusion, high levels of plasma LDL lead to intimal
sion in high-cholesterol hamsters, we studied the DNA synthe- hyperplasia via up-regulation of platelet aggregation and en-
sis using explanted SMC induced by various agonists. These hancement of proliferating SMC. Further research is neces-
results indicated that PDGF plays a key role in arterial stenosis sary to determine whether this LDL-induced hyperplasia oc-
by neointima after endothelial injury. When PDGF-BB were curs in humans with elevated LDL levels. If so, antiplatelet
applied to SMC from hamsters fed a high-cholesterol diet, and antithrombin drugs could be part of a new supportive
DNA synthesis increased markedly compared with synthesis therapeutic concept in the treatment of hypercholesterolemia-
in hamsters fed a normal diet. LDL is capable of increasing associated cardiovascular diseases. This model could be a
PDGF production and expression of PDGF receptors in human useful tool for investigation of this field since hamsters are
vascular SMC (30). Therefore, we performed binding experi- smaller and permit a reduction in the amount of drug required.
ments using 135I-labeled PDGF-BB on SMC from hamsters fed
a high-cholesterol diet or a normal diet. The results indicated
there were no differences in binding assay in SMC between
hamster populations. According to these findings, we specu-
Wentworth, D. (1986) Serum Cholesterol, Blood Pressure, and
REFERENCES
1. Martin, M.J., Hulley, S.B., Browner, W.S., Kuller, L.H., and
lated that the responsibility of post-PDGF receptors, such as
intracellular signaling, could play an important role in the en-
hancement of neointima formation in hypercholesterolemia.
Moreover, PDGF is an α granule component (31) that is re-
leased during the platelet reaction induced by collagen or ADP
Mortality: Implications from a Cohort of 361,662 Men, Lancet
2, 933–936.
2. Ross, R. (1993) The Pathogenesis of Atherosclerosis: A Per-
spective for the 1990s, Nature 362, 801–809.
3. Lupu, F., Heim, D., Bachmann, F., and Kruithof, E.K. (1994)
Expression of LDL Receptor-Related Protein/Alpha 2-
Lipids, Vol. 36, no. 5 (2001)

460
H. MATSUNO ET AL.
Macroglobulin Receptor in Human Normal and Atherosclerotic
Arteries, Arterioscler. Thromb. 14, 1438–1444.
Plays an Inhibitory Role in Platelet Activation, FEBS Lett. 429,
327–329.
4. Surya, I.I., and Akkerman, J.W. (1993) The Influence of 20. Ross, R., Glomset, J., Kariya, B., and Harker, L. (1974) A
Lipoproteins on Blood Platelets, Am. Heart J. 125, 272–275.
5. Malle, E., Ibovnik, A., Steinmetz, A., Kostner, G.M., and Satt-
ler, W. (1994) Identification of Glycoprotein IIb as the Lipopro-
Platelet-Dependent Serum Factor That Stimulates the Prolifera-
tion of Arterial Smooth Muscle Cells in vitro, Proc. Natl. Acad.
Sci. USA 71, 1207–1210.
tein(a)-Binding Protein on Platelets. Lipoprotein(a) Binding Is 21. Bronnegard, M., Werner, S., and Gustafsson, J.A. (1986) Pri-
Independent of an Arginyl-glycyl-aspartate Tripeptide Located
in Apolipoprotein(a), Arterioscler. Thromb. 14, 345–352.
mary Cortisol Resistance Associated with a Thermolabile Glu-
cocorticoid Receptor in a Patient with Fatigue as the Only
Symptom, J. Clin. Invest. 78, 1270–1278.
6. Pedreno, J., de Castellarnau, C., Cullare, C., Ortin, R., Sanchez,
J.L., Llopart, R., and Gonzalez-Sastre, F. (1994) Platelet LDL 22. Seed, M., Hoppichler, F., Reaveley, D., McCarthy, S., Thomp-
Receptor Recognizes with the Same Apparent Affinity Both Ox-
idized and Native LDL. Evidence That the Receptor-Ligand
Complexes Are Not Internalized, Arterioscler. Thromb. 14,
401–408.
son, G.R., Boerwinkle, E., and Utermann, G. (1990) Relation of
Serum Lipoprotein(a) Concentration and Apolipoprotein(a) Phe-
notype to Coronary Heart Disease in Patients with Familial Hy-
percholesterolemia, N. Engl. J. Med. 323, 1773–1784.
7. Ardlie, N.G., Selley, M.L., and Simons, L.A. (1989) Platelet Ac- 23. Flesch, M., Sachinidis, A., Ko, Y.D., Kraft, K., and Vetter, H.
tivation by Oxidatively Modified Low Density Lipoproteins,
Atherosclerosis 76, 117–124.
8. Stuart, M.J., Gross, S.J., Elrad, H., and Graeber, J.E. (1982) Ef- 24. Cummings, M.H., Watts, G.F., Umpleby, M., Hennessy, T.R.,
(1994) Plasma Lipids and Lipoproteins and Essential Hyperten-
sion, J. Clin. Invest. 72, 944–950.
fects of Acetylsalicylic-Acid Ingestion on Maternal and Neona-
tal Hemostasis, N. Engl. J. Med. 307, 909–912.
9. Rosenfeld, M.E., Tsukada, T., Gown, A.M., and Ross, R. (1987)
Fatty Streak Initiation in Watanabe Heritable Hyperlipemic and
Quiney, J.R., and Sonksen, P.H. (1995) Increased Hepatic Se-
cretion of Very-Low-Density-Lipoprotein Apolipoprotein B-
100 in Heterozygous Familial Hypercholesterolaemia: A Stable
Isotope, Atherosclerosis 113, 79–89.
Comparably Hypercholesterolemic Fat-Fed Rabbits, Arte- 25. Aguilar-Salinas, C.A., Barrett, P.H., Kelber, J., Delmez, J., and
riosclerosis 7, 9–23.
Schonfeld, G. (1995) Physiologic Mechanisms of Action of Lo-
vastatin in Nephrotic Syndrome, J. Lipid Res. 36, 188–199.
10. Adan, Y., Shibata, K., Sato, M., Ikeda, I., and Imaizumi, K.
(1999) Effects of Docosahexaenoic and Eicosapentaenoic Acid 26. Lasuncion, M.A., Teruel, J.L., Alvarez, J.J., Carrero, P., Ortuno,
on Lipid Metabolism, Eicosanoid Production, Platelet Aggrega-
tion and Atherosclerosis in Hypercholesterolemic Rats, Biosci.,
Biotechnol. Biochem. 63, 111–119.
J., and Gomez-Coronado, D. (1993) Changes in Lipoprotein(a),
LDL-Cholesterol and Apolipoprotein B in Homozygous Famil-
ial Hypercholesterolaemic Patients Treated with Dextran Sul-
fate LDL-Apheresis, Eur. J. Clin. Invest. 23, 819–826.
11. Murakami, S., Nitanai, I., Uchida, S., Kondo-Ohta, Y., Asami,
Y., Kondo, K., Sato, M., Kawashima, A., Hara, H., Tomisawa, 27. Kroon, A.A., Swinkels, D.W., van Dongen, P.W., and Stalen-
K., et al. (1999) Up-regulation of Low Density Lipoprotein Re-
ceptor by a Novel Isobenzofranone Derivative, MD-700, Ather-
osclerosis 146, 281–290.
hoef, A.F. (1994) Pregnancy in a Patient with Homozygous Fa-
milial Hypercholesterolemia Treated with Long-Term Low-
Density Lipoprotein Apheresis, Metabolism 43, 1164–1170.
12. Matsuno, H., Stassen, J.M., Hoylaerts, M.F., Vermylen, J., and 28. Nofer, J.R., Tepel, M., Kehrel, B., Wierwille, S., Walter, M.,
Deckmyn, H. (1995) Fast and Reproducible Vascular Neointima
Formation in the Hamster Carotid Artery: Effects of Trapidil
and Captopril, Thromb. Haemost. 74, 1591–1596.
Seedorf, U., Zidek, W., and Assmann, G. (1997) Low-Density
Lipoproteins Inhibit the Na+/H+ Antiport in Human Platelets. A
Novel Mechanism Enhancing Platelet Activity in Hypercholes-
terolemia, Circulation 95, 1347–1358.
13. Matsuno, H., Stassen, J.M., Vermylen, J., and Deckmyn, H.
(1994) Inhibition of Integrin Function by a Cyclic RGD-Con- 29. Witztum, J.L., and Steinberg, D. (1991) Role of Oxidized Low
taining Peptide Prevents Neointima Formation, Circulation 90,
2203–2206.
Density Lipoprotein in Atherogenesis, J. Clin. Invest. 88,
1785–1792.
14. Matsuno, H., Kozawa, O., Niwa, M., and Uematsu, T. (1997) 30. Sachinidis, A., Mengden, T., Locher, R., Brunner, C., and Vet-
Inhibition of von Willebrand Factor Binding to Platelet GP Ib
by a Fractionated Aurintricarboxylic Acid Prevents Restenosis
After Vascular Injury in Hamster Carotid Artery, Circulation
96, 1299–1304.
ter, W. (1990) Novel Cellular Activities for Low Density
Lipoprotein in Vascular Smooth Muscle Cells, Hypertension 15,
704–711.
31. Kaplan, K.L., Broekman, M.J., Chernoff, A., Lesznik, G.R., and
Drillings, M. (1979) Platelet Alpha-Granule Proteins: Studies
on Release and Subcellular Localization, Blood 53, 604–618.
15. Matsuno, H., Kozawa, O., Niwa, M., Tanabe, K., Ichimaru, K.,
Takiguchi, Y., Yokota, M., Hayashi, H., and Uematsu, T. (1998)
Multiple Inhibition of Platelet Activation by Aurintricarboxylic 32. Witte, L.D., Kaplan, K.L., Nossel, H.L., Lages, B.A., Weiss,
Acid Prevents Vascular Stenosis After Endothelial Injury in
Hamster Carotid Artery, Thromb. Haemost. 79, 865–871.
H.J., and Goodman, D.S. (1978) Studies of the Release from
Human Platelets of the Growth Factor for Cultured Human Ar-
terial Smooth Muscle Cells, Circ. Res. 42, 402–409.
16. Matsuno, H., Kozawa, O., Niwa, M., Ito, T., Tanabe, K.,
Nishida, M., Hayashi, H., and Uematsu, T. (1998) Effect of 33. Baumann, D.S., Doblas, M., Schonfeld, G., Sicard, G.A., and
GR144053, a Fibrinogen-Receptor Antagonist, on Thrombus
Formation and Vascular Patency After Thrombolysis by tPA in
the Injured Carotid Artery of the Hamster, J. Cardiovasc. Phar-
macol. 32, 191–197.
Daugherty, A. (1994) Probucol Reduces the Cellularity of Aor-
tic Intimal Thickening at Anastomotic Regions Adjacent to
Prosthetic Grafts in Cholesterol-Fed Rabbits, Arterioscler.
Thromb. 14, 162–167.
17. Matsuno, H., Stassen, J.M., Moons, L., Vermylen, J., and Hoy- 34. Stiko-Rahm, A., Hultgardh-Nilsson, A., Regnstrom, J., Ham-
laerts, M.F. (1996) Neointima Formation in Injured Hamster
Carotid Artery Is Effectively Prevented by the Combination
G4120 and Quinapril, Thromb. Haemost. 76, 263–269.
sten, A., and Nilsson, J. (1992) Native and Oxidized LDL En-
hances Production of PDGF AA and the Surface Expression of
PDGF Receptors in Cultured Human Smooth Muscle Cells, Ar-
terioscler. Thromb. 12, 1099–1109.
18. Matsuno, H., Kozawa, O., Niwa, M., Kaida, T., Hayashi, H., and
Uematsu, T. (1997) GR144053, a Fibrinogen Receptor Antago- 35. Wilson, D.P., Saward, L., Zahradka, P., and Cheung, P.K.
nist, Enhances the Suppression of Neointima Formation by
Losartan, an Angiotensin II Receptor Antagonist, in the Injured
Carotid Artery of Hamster, Br. J. Pharmacol. 122, 1099–1104.
19. Matsuno, H., Kozawa, O., Niwa, M., Usui, A., Ito, H., Uematsu,
T., and Kato, K. (1998) A Heat Shock-Related Protein, p20,
(1999) Angiotensin II Receptor Antagonists Prevent Neointimal
Proliferation in a Porcine Coronary Artery Organ Culture
Model, Cardiovasc. Res. 42, 761–772.
[Received August 18, 2000, and in revised form February 28, 2001;
revision accepted April 17, 2001]
Lipids, Vol. 36, no. 5 (2001)