Atherosclerosis impairs endothelium-dependent vascular relaxation to acetylcholine and
thrombin in primates.
P C Freiman, G G Mitchell, D D Heistad, M L Armstrong and D G Harrison
Circ Res. 1986;58:783-789
doi: 10.1161/01.RES.58.6.783
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1986 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/58/6/783
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation Research is online at:
http://circres.ahajournals.org//subscriptions/
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
783
Atherosclerosis Impairs Endothelium-Dependent
Vascular Relaxation to Acetylcholine and Thrombin
in Primates
Paul C. Freiman, Gordon G. Mitchell, Donald D. Heistad, Mark L. Armstrong, and
David G. Harrison
From the Cardiovascular Center and Department of Internal Medicine, University of Iowa College of Medicine, Iowa City, Iowa
SUMMARY. To test the hypothesis that atherosclerosis impairs endothelium-dependent vascular
relaxation, we examined the effect of the endothelium-dependent vasodilators acetylcholine and
thrombin and the endothehum-independent vasodilator nitroglycerin on iliac arteries from normal
cynomolgus monkeys and cynomolgus monkeys with diet-induced atherosclerosis. Rings of iliac
artery were suspended in organ chambers at their optimal length for generating tension. After
preconstriction with prostaglandin F2,,, cumulative concentration-response curves to acetylcholine,
thrombin, and nitroglycerin were examined The presence of endothehum was confirmed in each
vessel by scanning electron microscopy. Atherosclerotic vessels showed morpholigic evidence of
moderate to severe atherosclerosis. Acetylcholine produced a maximal relaxation of 65 ± 10% in
the normal group and 27 ± 10% in atherosclerotic vessels (P < 0 05). Thrombin (10 0 U/ml)
produced relaxation of 39 ± 9% in the normal group and 13 ± 7% in atherosclerotic iliac arteries
(P < 0.05). Nitroglycerin relaxed both normal and atherosclerotic blood vessels to an equal extentmaximal relaxation was 92 ± 4% in normal vessels and 98 ± 2% in atherosclerotic vessels To
determine if hypercholesterolemia alone produces an abnormality in endothelium-dependent
relaxation, we performed two additional studies First, because veins are exposed to hypercholesterolemia, but do not develop atherosclerosis, we studied relaxation responses to acetylcholine
and thrombin in veins from normal monkeys and monkeys with diet-induced atherosclerosis.
Veins from normal and atherosclerotic monkeys relaxed to a similar extent upon exposure to the
endothelium-dependent vasodilators acetylcholine and thrombin Second, we studied relaxation
responses to acetylcholine, thrombin, and nitroglycerin in left circumflex coronary arteries from
normal dogs and dogs fed a hypercholesterolemic diet for 4-5 weeks when serum cholesterol
levels were elevated (serum cholesterol 442 ± 14 mg/dl), but before the onset of atherosclerosis.
The endothelium-dependent vasodilators acetylcholine and thrombin produced equivalent degrees of relaxation in artenes removed from normal and hypercholesterolemic dogs These studies
demonstrate that atherosclerosis impairs endothelium-dependent relaxation in primate iliac arteries, and that this impairment is not due to a generalized defect in the endothelium caused by
hypercholesterolemia, but requires the presence of atherosclerosis (Che Res 58: 783-789, 1986)
CLINICAL and experimental observations suggest
that atherosclerosis alters vascular reactivity. Coronary vessels with atherosclerosis are more susceptible to spasm induced by ergonovine (Schroeder et
al., 1978; Waters et al, 1983) and may be predisposed to spontaneous coronary vasospasm (Schroeder et al., 1977; Maseri et al., 1978; Waters et al,
1983). Recent studies have demonstrated enhanced
vasoconstrictor responses to serotonin and norepinephrine in the hindlimb of atherosclerotic monkeys
(Heistad et al, 1984). Hypersensitivity of atherosclerotic rabbit aortas to ergonovine has also been demonstrated (Henry and Yokoyama, 1980). Enhanced
constrictor responses to serotonin and histamine
have been observed in coronary arteries following
This manuscript from the University of Iowa was sent to Robert
M Berne, Consulting Editor, for reviexv by expert referees, for editorial
decision, and final disposition
focal denudation of endothelium and hypercholesterolemia (Shimowawa et al., 1983; Kawachi et al.,
1984).
The vasodilator actions of many agents, including
acetylcholine and thrombin, depend upon release of
endothelium-derived relaxing factor (Furchgott and
Zawadski, 1980a, 1980b; Furchgott 1981; DeMey et
al., 1982; Ku, 1982; Furchgott, 1984). In addition to
promoting vasohdation, the endothelium attenuates
vasoconstrictor effects of several agents such as
platelets, serotonin, and norepinephrine in vitro
(Cohen et al., 1983; Cocks and Angus, 1983) and in
vivo (Brum et al., 1984; Lamping et al., 1985). Recently, it has been reported that the endothelium
produces a vasoconstrictor factor in response to
hypoxia (Rubanyi and Vanhoutte, 1985) and in
spontaneously hypertensive rats (Luscher and Vanhoutte, 1985).
Atherosclerosis thus might contribute to increased
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
Circulation Research/Vo/. 58, No. 6, June 1986
784
sensitivity to vasoconstrictor stimuli by producing a
functional defect in the endothelium. Hypercholesterolemia causes morphological (Trillo and Pritchard, 1979; Ingerman-Wojensky et al., 1983) and
perhaps functional (Henriksen et al., 1979) alterations of the endothelium. Furthermore, intimal
thickening in atherosclerosis might cause a barrier
to diffusion of endothelium-dependent relaxing factor (EDRF) from endothelium to vascular media.
We tested the hypothesis that atherosclerosis
impairs responses to endothelium-mediated vasodilators. We examined effects of two endotheliumdependent vasodilators, acetylcholine and thrombin, and an endothelium-independent vasodilator,
nitroglycerine, on iliac arteries from normal and
atherosclerotic cynomolgus monkeys. This experimental primate model of diet-induced atherosclerosis closely resembles human atherosclerosis (Wissler and Vesselinovitch, 1974; Armstrong et al.,
1980).
To separate effects of hypercholesterolemia from
effects of atherosclerosis, we performed two additional studies. First, we studied relaxation responses
in veins from monkeys with diet-induced atherosclerosis. Veins as well as arteries are exposed to
hypercholesterolemia, but veins do not develop atherosclerotic lesions. Second, we examined the effect
of short-term hypercholesterolemia on the endothelium-dependent relaxation responses to acetylcholine and thrombin in circumflex arteries from dogs,
an animal model that is relatively resistant to the
development of atherosclerosis. Using these two
approaches, we sought to determine whether alterations of endothelium-dependent relaxation in atherosclerotic arteries are due to a diffuse abnormality
of endothehal function produced by hypercholesterolemia, or if the presence of atherosclerosis is required.
Methods
Source of Normal and Atherosclerotic Blood Vessels
Atherosclerosis was induced in a group of cynomolgus
monkeys (n = 12) by feeding them an atherogenic diet
(semisynthehc diet containing 40% fat and 0 74% cholesterol) for 18 months. A control group of monkeys (n — 13)
was maintained on commercial laboratory chow (Purina
Monkey Chow, Ralston Purina Company) The mean
serum cholesterol at the end of 18 months was 622 ± 56
mg/dl in the atherosclerotic group vs 99 ± 7 mg/dl in the
control group (P < 0 005)
On the day of the study, normal or atherosclerotic
monkeys were sedated with ketamine (15 mg/kg lm) and
anesthetized with chloralose (100 mg/kg IV) The iliac
artenes were isolated and excised
Isolated Vascular Ring Preparation
Vessels were cut into 5-mm ring segments and were
suspended in a vertically oriented organ bath in 25 ml
of Kreb's buffer (composition in HIM: NaCl, 118.3, KG,
4 7; CaCl2, 2 5, MgSO4, 1.2; KH2PO4, 1 2; NaHCO3, 25,
EDTACa, 0.026; glucose, 11.1; pH 7.40) aerated with
95% O2, 5% CO2, and maintained at 37°C All studies
were performed in the presence of propranolol (10~7M).
Tension was recorded with a linear force transducer (Grass
FTO3c) on an oscillographic recorder.
In some expenments, the endothelium was removed
from the segment by rubbing the intimal surface with the
tip of a closed hemostat.
Over 2 hours, the resting tension of the vascular ring
was gradually increased until the optimal tension for
generating force dunng isometric contraction was reached.
At each tension, the vessel was exposed to KC1 (100 HIM),
and the tension generated was recorded After each KG
dose, the baths were washed with fresh buffer. The resting
tension was increased until additional doses of KG failed
to increase further the constrictor response. The vessels
were left at this optimal resting tension throughout the
remainder of the .study.
Determining Preconstricting Dose of Prostaglandin F2a
(PGF2a)
To study vasodilator responses, we preconstncted the
vascular rings with PGF2n. To establish the concentration
of PGF2o that would give a submaximal constriction, we
determined a complete PGF2(t concentration-response relationship for each vessel. A dose of PGF2n that produced
30%-50% of the maximal constriction was used in subsequent expenments to preconstrict the vessel before the
vasodilator drugs were added
Drug Preparation
Drugs used in the study were prostaglandin F2n (Tris
salt, Sigma), acetylcholine (Sigma), and nitroglycenn
(American Critical Care) Bovine thrombin was generously
supplied by Dr. Whyte Owen. All drug dilutions were
prepared with distilled water. Drugs were diluted so that
less than 0 1 ml was added for each dose
Protocols
We examined cumulative relaxation responses to acetylcholine (10~9 to 10- 4 M), thrombin (0.1, 1.0, 10.0 U/ml),
and nitroglycenn (10~10 to 10~6M). Before each concentration-response curve, vessels were preconstncted with
PGF2o to an ED30 to ED50. Between each concentrationresponse curve, the vessels were washed at least three
times with fresh buffer and were allowed to reequilibrate
for at least 30 minutes.
Studies in Veins
In seven normal and five atherosclerotic monkeys, segments of jugular veins were removed and studied in a
fashion identical to that used for iliac arteries. Responses
to acetylcholine (10~9 to 10"" M), thrombin (0.1 U/ml, 1 0
U/ml, and 10.0 U/ml) and nitroglycenn (10~10 to 10~6M)
were examined, each after preconstnction with PGF2n.
Studies in Dogs
Mongrel dogs of either sex (n = 5), weighing 19-30 kg,
were fed a high-cholesterol diet deficient in essential free
fatty acids consisting of 5% cholesterol, total fat 21% (wt/
wt) (Ehrhart Atherogenic Test Diet, Teklad).
At the end of 4-5 weeks, serum cholesterol was 442 ±
14 (n = 5). Eight mongrel dogs (serum cholesterol 85 ±
12) were used as controls. On the day of the study, dogs
were anesthetized with sodium pentobarbital (30 mg/kg,
iv), the chest was opened by left thoracotomy, and the
pericardium was opened. The heart was electrically arrested and removed. The left circumflex coronary artery
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
Freiman et al /Endothelium-Dependent Relaxation in Atherosclerosis
was dissected free and excised. Vascular rings were prepared and studied in a manner identical to that used for
monkey iliac arteries.
Quantification of Endothelium
All vascular rings were examined by scanning electron
microscopy. At the completion of the experiment, each
vascular segment was immersed in fixative (2 5% glutaraldehyde in 0 1 M sodium cacodylate buffer, pH 7.2) for
approximately 5 minutes, while still mounted on the isolated ring apparatus. The vessel was then removed from
the apparatus and maintained in fixative at 4°C for at
least 24 hours. After fixation, the vessel was cut longitudinally to expose the intimal surface, mounted, and prepared for scanning electron microscopy Each vessel segment was examined at 600X magnification, and a visual
estimate of the percent surface area covered by endothelium was made Despite efforts to minimize damage to
the endothelium, some loss of endothelium did occur
Only those vessels with more than 30% endothelial coverage were included m the subsequent data analysis for
acetylcholine and thrombin. The average endothelial coverage determined by this method was 46 ± 4% in normal
monkeys and 63 ± 5% in the atherosclerotic monkeys.
To confirm the accuracy of our visual estimate of endothelial coverage as determined by scanning electron
microscopy, six normal and six atherosclerotic vessels were
chosen at random and examined by transmission electron
microscopy at 2800X. From each vessel, 60-80 photographs of the intimal surface were obtained from randomly chosen sites. The area covered by endothelial cells
and the total surface were planimetered with a linear
digitizer (MOP-3, Zeiss). The area covered by endothelium
was expressed as the amount of intimal surface covered
by endothelium divided by the total intimal surface. By
this method, the normal vessels had 53 ± 8% endothelial
coverage, and atherosclerotic vessels were covered with
endothelium over 51 ± 3% of their surface (P > 0.5).
785
relaxation response was absent after removal of the endothelium. The ED50 was calculated as the concentration
of acetylcholine that produced half-maximal relaxation.
Student's f-test for unpaired comparisons was used to
compare concentration-response curves at each drug concentration. Data are expressed as the mean ± SEM. The
level of confidence chosen for statistical significance was
P<0.05.
Results
Baseline Characteristics
The optimal resting tensions were not different in
normal and atherosclerotic vessels, 5.5 ± 0.4 g and
5.9 ± 0.6 g, respectively (P > 0.50). The response to
KC1 (100 m\d) was greater in the normal vessels, 4.7
± 0.6 g vs. 2.1 ± 0.4 g in the atherosclerotic vessels
(P < 0 05). Similarly, the peak response to PGF2a
was greater in the normal vessels, 8.7 ± 0.6 g,
compared to the atherosclerotic group, 4.7 ± 0.7 g
(P < 0.05).
Responses to Acetylcholine
Responses to acetylcholine in atherosclerotic and
normal vessels are shown in Figures 1 and 2. In all
normal vessels with endothelium, acetylcholine produced concentration-dependent relaxation. In normal vessels denuded of endothelium, the response
to acetylcholine was markedly attenuated, averaging
12% relaxation. Only one of four denuded vessels
exhibited any response to acetylcholine. In contrast
to the normal vessels, only seven of 12 atherosclerotic vessels with endothelium present relaxed to
acetylcholine. The average total relaxation to acetyl-
Assessment of Atherosclerosis
After fixation, a 1-mm section was cut from the end of
each monkey iliac artery vascular ring and stained with
hematoxylin and eosin or hematoxyhn and orcein. Gross
and histological examination showed moderate to severe
atherosclerosis in the iliac arteries of monkeys fed an
atherogenic diet for 18 months All vessels in the atherosclerotic group had histological evidence of atherosclerosis. Atherosclerotic lesions were not present in iliac arteries
from normal monkeys.
Data Analysis and Statistics
Relaxation responses to acetylcholine and nitroglycenn
were expressed as the percent relaxation from the amount
of preconstriction produced by PGF2,,. Thrombin, when
administered to a preconstncted normal vessel, typically
caused a biphasic response, initial relaxation followed by
a return to the preconstricted tension Based on preliminary observations from our laboratory and by others (Ku,
1982; DeMey et al, 1982), only the relaxation portion of
the thrombin response is endothelium dependent Thus,
for thrombin, we compared the relaxation responses in
normal and atherosclerotic monkeys This response was
recorded as the cumulative relaxation response at each
dose, expressed as a percentage of the initial preconstncted
tension The jugular veins relaxed to lower concentrations
of acetylcholine, but often began to contract at higher
concentrations. In preliminary experiments, we found the
Normal Monkey
Atherosclerotic Monkey
0 5 giam
PGF20C
FIGURE 1. Response to acetylcholine in preconstricted normal and
atherosclerotic monkey iliac arteries Vessels were preconstricted with
prostaglandm F2a and then were subjected to serial increasing doses
of acetylcholine The responses seen m normal {above) and atherosclerotic (below) iliac arteries are shown Doses are expressed as the
logarithm of the (mal acetylcholine concentration following each dose
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
786
Circulation Research/Vol. 58, No. 6, June 1986
n
^>
20
*\
Percent
Relaxation
40
60
nn
1 I
*
•p<0 05
80
1
1
1
1
1
1
1
1
1
*
*
i
i
FIGURE 2. Relaxation responses to acetylcholine
m normal (O) and atherosclerotic (0) monkey
iliac arteries Cumulative doses of acetylcholine
are shown along the abscissa, and percent relaxation is shown along the ordmate At each
dose, acetylcholine produced less relaxation in
atherosclerotic iliac arteries than in normal
vessels. *P < 0.05 compared to normal arteries
-log ^Acetylcholine]
choline in all atherosclerotic vessels was less than
half that observed in the normal vessels. The ED50
for relaxation to acetylcholine could not be calculated for the atherosclerotic vessles that failed to
respond; however, in the remaining atherosclerotic
vessels, the ED50 was not different from that of the
normal vessels.
Normal Monkey
T 5 grams
Responses to Thrombin
Typical responses to thrombin are shown in Figure
3. In normal vessels, thrombin produced initial vascular relaxation, often followed by either vasoconstriction or return to the baseline preconstricted tension. In atherosclerotic vessels, the relaxation response was markedly diminished, while contraction
was preserved. The relaxation response was greater
in normal vessels than atherosclerotic vessels (Table
1). Thrombin did not cause relaxation in normal
vessels denuded of endothelium (n = 4).
Responses to Nitroglycerin
Figure 4 shows the cumulative relaxation responses to nitroglycerin observed in normal and
atherosclerotic vessels. The maximal relaxation response and ED50 for nitroglycerin were similar in
normal and atherosclerotic vessels (Table 1).
Atherosclerotic Monkey
1 gram
FIGURE 3. Response to thrombin m preconstricted normal and atherosclerotic monkey iliac arteries Following preconstnctwn with
prostaglandin F2m serial increasing doses of thrombin were added In
normal iliac arteries (above), thrombin caused a biphasic response,
initial relaxation followed by a return to the preconstricted tension
In atherosclerotic vessels (below), the relaxation response was markedly decreased, while the constrictor response remained
Responses to Acetylcholine, Thrombin, and
Nitroglycerin in Jugular Veins of Normal and
Atherosclerotic Monkeys
Relaxation responses to acetylcholine, thrombin,
and nitroglycerin in jugular veins removed from
atherosclerotic and normal animals are summarized
in Table 2. These three agents produced similar
degrees of relaxation in veins from atherosclerotic
and normal monkeys.
Responses to Acetylcholine, Thrombin, and
Nitroglycerin in Circumflex Coronary Arteries
from Normal and Hypercholesterolemic Dogs
The endothelium-dependent agonists acetylcholine and thrombin and the endothelium-independent agonist nitroglycerin produced similar degrees
of relaxation in circumflex coronary arteries from
normal and hypercholesterolemic dogs (Table 2).
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
Freiman et a/./Endothelium-Dependent Relaxation in Atherosclerosis
787
TABLE 1
Vascular Relaxation of Normal and Atherosclerotic Iliac Arteries
Acetylcholine
Nitroglycenn
Thrombm
%]Relaxation
Max%
relaxation
4±1X10"7M
65
Normal monkeys
(n = 13)
±10
Normal monkeys
(» = 10)
27
±10*
Atherosclerotic
monkeys
(" = 12)
0 1 10 10 0
U/ml U/ml U/ml
ED50
9 ± 3 x 10"7 M | Atherosclerotic
monkeys
(« = 9)
13
31
39
±5
±8
±9
4
±3*
12
13 Atherosclerotic
±6* ±7*
monkeys
(« = 9)
Max%
relaxation
ED 5 0
92
2 x 1 x 10"8 M
Normal monkeys
(n = 9)
±4
98
±1
5 ± 2 X 10~8M
Results are expressed as mean ± SE
* P < 0 05 compared to normal monkeys
f Calculation based only on vessels that responded to acetylcholine. % Relaxation = percent relaxation from preconstncted tension,
Max% relaxation = maximal percent relaxation
Discussion
The major finding in this study is that vascular
relaxation to endothelium-dependent vasodilator
stimuli is impaired in atherosclerotic blood vessels.
This defect in relaxation was observed in response
to endothelium-dependent vasodilator agonists,
acetylcholine and thrombin, but not to the endothehum-independent vasodilator nitroglycerin. Furthermore, the defect in relaxation observed in atherosclerotic blood vessels apparently was not produced by a diffuse alteration in the endothelium
caused by hypercholesterolemia, since veins from
atherosclerotic monkeys and coronary arteries from
hypercholesterolemic dogs relaxed normally in response to endothelium-dependent agonists.
All vessels used in this study were examined by
scanning electron microscopy to confirm the presence of endothelium. Despite efforts to minimize
endothelial injury, some degree of endothehal denudation occurred in virtually all vessels during the
course of an experiment. To assess the effect of
partial denudation on the response to vasodilators,
we measured the percentage of the surface area
covered by endothelium. Our results showed that
the intimal surface covered by endothelium was
similar in normal and atherosclerotic vessels. Thus,
impairment of endothelium-dependent vascular relaxation in atherosclerotic vessels could not be ascribed to greater denudation of endothelium.
Vessels from hypercholesterolemic monkeys with
atherosclerotic lesions may show impaired relaxation in response to endothelium-dependent
vasodilators for several reasons. First, hypercholesterolemia itself may contribute to the abnormality.
Abundant evidence suggests that elevated serum
cholesterol can produce morphological alterations of
the endothelium (Trillo and Prichard, 1979; Ingerman-Wojensky et al., 1983). Thus, in damaged endothelium, synthesis of endothelium-derived relaxing factors (EDRF) may be impaired. Second, the
atherosclerotic process may produce a barrier between endothelium and vascular smooth muscle so
that EDRF, although present, is unable to reach the
site of its action. This barrier might take several
forms. Because EDRF has an extremely short half-
o -
FIGURE 4. Relaxation responses to nitroglycerin
in normal (O) and atherosclerotic (9) monkey
iliac arteries Increasing concentrations of nitroglycerin are shown along the abscissa, and
percent relaxation is shown along the ordmale
At each concentration of nitroglycerin, normal
and atherosclerotic monkey iliac arteries relaxed
to the same extent
Percent
Relaxation
80
-
100
-log [Nitroglycerin]
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
788
Circulation Research/Vol. 58, No. 6, June 1986
TABLE 2
Effect of Hypercholesterolemia on Vascular Relaxation of Jugular Veins in Monkeys and of Circumflex Coronary Arteries
Thrombin
Acetylchohne
Nitroglycenn
%: Relaxation
Max%
relaxation
Jugular vein
Normal monkey (« = 9)
Atherosclerotic
monkey (M =
6)
Coronary artery
Normal dog
(n = 8)
Hypercholesterolemic
dog (n = 5)
01
10 0
10
U/ml U/ml U/ml
Max%
relaxation
Normal monkey
38
48
56 Normal monkey
(n = 7)
± 10 ± 10 ± 10
(n = 6)
Atherosclerotic
26
40
48 Atherosclerotic
monkey (n =
±6* ±5* ±7* monkey (n =
±0
98
EDso
8
39 2 ± 1 x 10" M
± 11
36 2 ± 1 X 10"8 M"
± 14*
±0
95
±5f
7 ± 3 x 10~8
Normal dog
(n = 8)
4 ± 1 X 10' M
±O5X
±2f
6)
5)
100
100
ED50
38
±13
1 ± 0 7 x 10~8 M$ Hypercholes51
terolemic dog ±16*
(« = 5)
91
±6
93
±7*
92 Normal dog
±6
(» = 8)
99 Hypercholes±1*
terolemic dog
(n = 5)
100
±0
100
±0*
1±O3X1O"8M
1 ± 0 3 X 10~8 M*
Data are expressed as mean ± SE. % Relaxation = percent relaxation from preconstncted tension Max% relaxation = maximal percent
relaxation * P > 0 5, f P > 0 2, J P > 0 1 Hypercholesterolemic vs normal vessels
life (Griffith et al., 1984; Rubanyi et al., 1985),
intimal thickening and increased diffusion distance
might lead to diminution of the EDRF-mediated
response. Alternatively, increases in lipids in the
intima might adsorb lipophilic substances and impair diffusion of EDRF to the vascular smooth muscle. Finally, cellular elements in the advanced atherosclerotic plaque might lead to increased degradation of EDRF. A third factor in alteration of vascular responses in atherosclerosis might be that
changes in membrane lipid composition and fluidity
alter affinity of vascular muscle for EDRF or alter
EDRF release (Lurie et al., 1985). This possibility is
unlikely because responses were not impaired in
hypercholesterolemic monkey veins or canine arteries. Fourth, the endothelium in atherosclerotic vessels might produce a vasoconstrictor agent (DeMey
and Vanhoutte, 1982). Recent work in hypertensive
rats has shown that endothelium is capable of producing a constrictor substance (Luscher and Vanhoutte, 1985; Rubanyi and Vanhoutte, 1985). Atherosclerosis may increase synthesis of an endothelium-derived vasoconstrictor.
Previous investigations have demonstrated that
vasoconstrictor responses to serotonin and platelets
are enhanced in vessels denuded of endothelium
(Cohen et al., 1983; Cocks and Angus, 1983). These
findings suggest that the endothelium attenuates
responses to certain vasoconstrictor agents. This observation was extended by the recent observation
that endothelial denudation and hypercholesterolemia can be used to produce coronary artery spasm
in miniature swine and dogs (Shimowawa et al,
1983; Kawachi et al., 1984). Recent preliminary reports by Habib et al. (1984) and Herman et al. (1985)
indicate that, after hypercholesterolemia induces intimal changes, endothelium-dependent vascular re-
laxation is impaired in rabbit aorta. This finding
suggests that hypercholesterolemia and/or atherosclerosis impair endothelium-dependent vascular responses. Our results in monkey veins and hypercholesterolemic dog arteries suggest that the defect in
endothelial function is produced by atherosclerosis
and not by hypercholesterolemia per se.
Recently, a preliminary study by Bossaller et al.
(1985) showed that responses to acetylcholine were
abnormal in cholesterol-fed rabbits, whereas responses to A23187 were normal. These data suggest
a defect in the muscarinic receptor rather than an
inability of the endothelium to produce the relaxing
factor. In our studies, we found abnormal responses
to both acetylcholine and thrombin. Thus, in this
primate model of atherosclerosis, abnormalities of
endothelium-dependent relaxation are not solely related to an alteration of the muscarinic receptor, but
involve either multiple receptors or some aspect of
the underlying effector mechanism.
Our findings may explain in part the clinical observation that patients with coronary atherosclerosis
are susceptible to coronary spasm and are more
sensitive to the vasoconstrictor effects of ergonovine
(Schroeder et al., 1977; Maseri et al, 1978; Waters
et al., 1983). We speculate that a defect in endothelium-dependent vasodilation may be important in
the pathogenesis of vascular spasm, and in clinical
disorders in which vasospasm is associated with
atherosclerosis.
We gratefully acknowledge the technical assistance of Knsten
Orgren, Tony Abboud, and Nancy Gagnon The authors also thank
Marlene Blakley for secretarial assistance.
Published as an abstract in Fed Proc. 44:523, 1985
Dr. Freimaii is a recipient of NHLB1 NRSA (HL07176) Dr Harrison is a recipient of an NHLBI Clinical Investigator Award
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013
Freiman et al /Endothelium-Dependent Relaxation in Atherosclerosis
(HL01046) Supported by American Heart Association Crant-m-Aid
831069, N1H Grants HL27633, HL20827, and HL 14388, lschemic
SCOR HL32295, and Atherosclerosis SCOR HL 14230
Address for reprints David C. Harrison, MD, Cardiovascular
Division, Department of Internal Medicine, University of Iowa Hospitals, Iowa City, loioa 52242
Received October 30, 1985, accepted for publication February 14,
1986
References
Armstrong ML, Tnllo A, Pnchard RW (1980) Naturally occurring
and experimentally induced atherosclerosis in nonhuman primates In The Use of Nonhuman Primates in Cardiovascular
Disease, pp 58-101
Bossaller C, Habib JB, Wells S, Henry PD (1985) Dissociation
between muscannic and Ca-ionophore induced endothehumdependent relaxation in atherosclerotic rabbit aorta and human
coronary artery (abstr). Circulation 72 [suppl III): 137
Brum JM, Sufan Q, Lane G, Bove AA (1984) Increased vasoconstrictor activity of proximal coronary arteries with endothelial
damage in intact dogs Circulation 70: 1066-1073
Cocks TM, Angus AR (1983) Endothehum dependent relaxation
of coronary arteries by nonadrenahne and serotonin Nature
305: 627-630
Cohen RA, Shepherd JT, Vanhoutte PM (1983) Inhibitory role of
the endothehum in the response of isolated coronary arteries
to platelets Science 221: 273-274
DeMey JG, Vanhoutte PM (1980) Endothehum and relaxation of
isolated canine arteries (abstr) Pharmacologist 22: 282
DeMey JG, Vanhoutte PM (1982) Heterogenous behavior of the
canine arterial and venous wall Circ Res 51: 439-447
DeMey JG, Claeys M, Vanhoutte PM (1982) Endothehum dependent inhibitory effects of acetylchohne, adenosine tnphosphate, thrombin and arachidonic acid in the canine femoral
artery J Pharmacol Exp Ther 222: 166-173
Furchgott RF (1981) The requirement for endothelial cells in the
relaxation of arteries by acetylchohne and some other vasodilators Trends Pharmacol Sci 2: 173-176
Furchgott RF (1983) Role of endothehum in responses of vascular
smooth muscle Circ Res 53: 557-573
Furchgott RF, Zawadski JV (1980a) Acetylchohne relaxes artenal
smooth muscle by releasing a relaxing substance from endothelial cells (abstr) Fed Proc 39: 581
Furchgott RF, Zawadski JV (1980b) The obligatory role of endothelial cells in the relaxation of artenal smooth muscle by
acetylchohne Nature 288: 373-376
Griffith TM, Edwards DH, Lewis MJ, Newby AC, Henderson AH
(1984) The nature of endothelium-denved vascular relaxant
factor Nature 308: 645-647
Habib JB, Wells SL, Williams CL, Henry PD (1984) Atherosclerosis impairs endothehum-dependent arterial relaxation (abstr)
Circulation 70 [suppl II): 123
Heistad DD, Armstrong ML, Marcus ML, Piegors DJ, Mark AL
(1984) Augmented responses to vasconstnetor simuh in hypercholesterolemic and atherosclerotic monkeys Circ Res 54: 711 —
718
789
Hennksen T, Evenson SA, Carlander B (1979) Injury to human
endothelial cells in culture induced by low density hpoproteins.
Scand J Clin Lab Invest 39: 361-368
Henry PD, Yokoyama M (1980) Supersensitivity of atherosclerotic
rabbit aorta to ergonovine mediated by a serotonergic mechanism J Clin Invest 66: 306-313
Herman AG, Coene MC, Jordaens F, VanHore C, Zonnekeyn L,
Verbeuren TJ (1985) Endothehum-dependent relaxations are
modified in isolated arteries from hypercholesterolemic rabbits
(abstr) 50th Jubilee Congress of the Hungarian Physiological
Society, July 1-7, 1985, p 94
Ingerman-Wojensky CM, Sedar AW, Nissenbaum M, Silver MJ,
Klurfield DM, Kntchevsky D (1983) Early morphological
changes in the endothehum of a peripheral artery of rabbits
fed an atherogenic diet Exp Mol Pathol 38: 48-60
Kawachi Y, Tomoike H, Maruoka Y, Kikuchi Y, Araki H, Ishn Y,
Tanaka K, Nakamura M (1984) Selective hypercontraction
caused by ergonovine in the canine coronary artery under
conditions of induced atherosclerosis Circulation 69: 441-450
Ku D (1982) Coronary vascular reactivity after acute myocardial
ischemia. Science 218: 576-578
Lamping KG, Marcus ML, Dole WF (1985) Removal of the
endothehum potentiates canine large coronary artery constrictor responses to 5-hydroxytryptamine in vivo. Circ Res 57: 4654
Lune KG, Chin JH, Hoffman BB (1985) Decreased membrane
fluidity and /3-adrenergic responsiveness in atherosclerotic
quail Am J Physiol 249: H380-H385
Luscher TF, Vanhoutte PM (1985) Endothehum-dependent contractions to acetylchohne in the aorta of spontaneously hypertensive rats Abstract presented at the annual meeting of the
Council for High Blood Pressure Research, September
Masen A, L'Abbate A, Barold G, Chierchia S, Marzilh M, Ballestra
AM, Sevenj S, Parodi O, Biagini A, Distante A, Pesula A (1978)
Coronary vasospasm as a possible cause of myocardial infarction, a conclusion derived from the study of preinfarction
angina N Engl J Med 299: 1271-1277
Rubanyi GM, Vanhoutte PM (1985) Hypoxia releases a vasoconstrictor substance from the canine vascular endothehum J
Physiol (Lond) 364: 45-56
Schroeder JS, Bolen JL, Quint RA, Clark DA, Hayden WG,
Higgins CB, Wexler L (1977) Provocation of coronary spasm
with ergonovine maleate Am J Cardiol 40: 487-491
Shimowawa H, Tomoike H, Nabeyama S, Yamamoto H, Araki
H, Nakamura M (1983) Coronary artery spasm induced in
atherosclerotic miniature swine Science 221: 560-562
Tnllo AA, Pnchard RW (1979) Early endothehal" changes in
expenmental pnmate atherosclerosis,. Lab Invest 41: 294-302
Waters DD, Szlachcic J, Boran R, Miller DD, Dauwe F, Theroux
P (1983) Comparative sensitivity of exercise, cold pressor and
ergonovine testing in provoking attacks of vanant angina in
patients with active disease Circulation 67: 310-315
Wissler RW, Vessehnovitch D (1974) Difference between human
and animal atherosclerosis In Atherosclerosis III, edited by G
Schettler, A Weizel New York, Spnnger-Verlag, pp 319-325
INDEX TERMS Endothehum • Atherosclerosis • Vasodilation •
Acetylchohne • Thrombin
Downloaded from http://circres.ahajournals.org/ by guest on February 21, 2013