Download by Angiotensin II - Circulation Research

1185
Cardiac Myocyte Necrosis Induced
by Angiotensin II
Lip-Bun Tan, Jorge E. Jalil, Ruth Pick, Joseph S. Janicki, and Karl T. Weber
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Although the role of angiotensin II (Ang II) in the pathogenesis and progression of the failing
heart is uncertain, previous reports have suggested that myocyte injury may be a component in
this process. In this study, we investigated this possibility in more detail. Cardiotoxic effects of
nonacutely hypertensive doses of Ang II were examined in 90 rats, including those receiving an
angiotensin infusion (200 ng/min i.p.) and those with renovascular hypertension, where
endogenous stimulation of Ang II occurred. Myocyte injury and wound healing resulting from
these treatments were evaluated by 1) immunofluorescence after in vivo monoclonal antibody
labeling of myosin to detect abnormal sarcolemmal permeability, 2) [3H] thymidine incorporation into DNA, to detect fibroblast proliferation, and 3) light microscopic evidence of
myocytolysis and subsequent scar formation. We found that exogenous Ang II produced
multifocal antimyosin labeling of cardiac myocytes and myocytolysis, which were maximal on
days 1-2 of the infusion. Subsequently, DNA synthesis rates were increased, with fibroblast
proliferation reaching peak levels on day 2 (Ang H-treated rats, 90.0±t18.6 cpm/,ug DNA;
control rats, 11.4±2.3 cpm/,ug DNA; p<0.05); microscopic scarring was found on day 14 and
represented 0.12±0.02% of the myocardium. Concurrent treatment with both propranolol (30
mg/kg/day s.c.) and phenoxybenzamine (5 mg/kg/day i.m.) did not attenuate Ang II-induced
antimyosin labeling. Increased endogenous Ang II, resulting from renal ischemia after
abdominal aortic constriction, produced both antimyosin labeling and increased rates of DNA
synthesis like that observed with Ang H infusion. Both myocyte injury and fibroplasia were
prevented with captopril (65 mg/day p.o.), but this protective effect was not seen with reserpine
pretreatment. Infrarenal aortic banding without renal ischemia, on the other hand, produced
hypertension without necrosis. We conclude that pathophysiological levels of endogenous as
well as low-dose exogenous Ang II were associated with altered sarcolemmal permeability and
myocytolysis with subsequent fibroblast proliferation and scar formation. Myocyte injury was
unrelated to the hypertensive or enhanced adrenergic effects of Ang II or to hypertension per
se. Captopril was effective in preventing myocyte injury in renovascular hypertension. The
mechanism(s) responsible for Ang HI-induced necrosis will require further study. (Circulation
Research 1991;69:1185-1195)
T he neuroendocrine system is widely recognized to play a dominant role in the pathophysiology of congestive heart failure.1-3
Both the adrenergic and renin-angiotensin systems
are involved. The commonly acknowledged role of
From the Cardiovascular Institute (R.P.), Michael Reese Hospital and Medical Center, and The University of Chicago, Pritzker
School of Medicine, Chicago, Ill., the Cardiology Department
(L.-B.T.), Killingbeck Hospital, Leeds, UK, the Dept. de Enfermedades Cardiovasculares (J.E.J.), Catholic University of Chile,
Santiago, Chile, and the Division of Cardiology (J.S.J., K.T.W.),
University of Missouri, Columbia, Mo.
Supported in part by National Heart, Lung, and Blood Institute
grant R01-HL-31701. L.-B.T. was a recipient of the Travelling
Fellowship from the Medical Research Council, UK.
Address for reprints: Dr. Karl T. Weber, Division of Cardiology
(MA432), University of Missouri, One Hospital Drive, Columbia,
MO 65212.
Received November 1, 1988; accepted June 7, 1991.
these systems in heart failure is in triggering mechanisms that compensate for the reduction in systemic
blood flow and perfusion pressure.4 Serum levels of
norepinephrine5-7 and angiotensin II (Ang 11)8.9 are
each elevated in heart failure at rest and rise even
further with physical activity.'0"11 Both neurohumoral
agents are also known to raise myocardial protein
synthesis and cause myocyte hypertrophy.12-16
Less well known are the roles played by the
adrenergic and renin-angiotensin systems in the
pathogenesis and progression of the failing heart
itself. Pharmacological doses17-21 or pathophysiolog-
ical levels22'23 of catecholamines are widely known to
produce cardiac myocyte necrosis, although the
mechanism of cell death is still unclear.24 Like catecholamines, Ang II infused intravenously in doses
that induce acute systemic hypertension has been
1186
Circulation Research Vol 69, No 5 November 1991
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
shown to cause cardiac myocyte necrosis.25,26 However, it is unknown whether smaller exogenous doses
of Ang II that do not induce acute hypertension or
elevated endogenous levels of Ang II have a similar
toxic effect on the myocardium. The issue of whether
Ang II mediates cardiac myocyte necrosis is especially relevant in view of the fact that this condition
may be eminently treatable by angiotensin converting
enzyme inhibitors.
It was our aim, therefore, to determine whether
exogenous levels of Ang II, which did not precipitate
acute hypertension, were related to the appearance
of myocyte necrosis. We sought to further determine
whether increased endogenous levels of Ang II,
developing in response to renal ischemia, might
likewise induce necrosis (vis-a-vis nonrenovascular
hypertension). Myocyte injury was assessed directly
by in vivo labeling with monoclonal antibodies specific to cardiac myosin,27 a specific and sensitive
means of detecting abnormal sarcolemmal permeability in individual myocytes,2128 and standard light
microscopic examination of fixed tissue. The woundhealing response, specifically fibroblast proliferation
and subsequent replacement fibrosis, was also examined to indirectly confirm the presence and extent of
myocyte necrosis.
We found that abnormal sarcolemmal permeability
and myocytolysis occurred with pathophysiological
levels of Ang II and were followed by fibroplasia and
scarring. Unlike pretreatment with pharmacological
agents that would alternate the catecholamine response to Ang II, captopril prevented both myocyte
necrosis and wound healing in renovascular hypertension, whereas myocyte necrosis and wound healing were not observed with infrarenal aorta banding.
Materials and Methods
Animal Population
Experiments were performed on 90 male SpragueDawley rats (190-300 g) maintained on a diet of
standard chow (Purina No. 5001, Ralston Purina Co.,
St. Louis, Mo.) containing 174 meq/kg sodium and
282 meq/kg potassium. The rats were allowed to
acclimatize to the housing, with free access to food
and tap water for 7 days before entry into the study.
Ang II Infusion
Thirty-one rats were anesthetized with 35-50
mg/kg i.p. methohexital sodium. Each then underwent an intraperitoneal implantation of an Alzet
osmotic minipump (model 2002, Alza Corp., Palo
Alto, Calif.) containing Ang II (CIBA-GEIGY, West
Caldwell, N.J.) dissolved in 0.O1N acetic acid. The
concentration of Ang II was adjusted so that 200
ng/min was delivered at the pump rate of 0.5 ,ul/hr.
This dose was selected on the basis of preliminary
studies that indicated that it did not produce an
immediate elevation in arterial pressure in contrast
to the acute hypertensive effects seen with the intra-
venous infusion of the same dose. Rats received Ang
II for 1-14 days before they were killed.
A separate group of rats (n=6), which were untreated or which received an intraperitoneal implant
of a minipump containing only the vehicle (0.O1N
acetic acid), served as the controls. A third set of rats
(n=7) received injections of propranolol (10 mg/kg
s.c. three times daily) and phenoxybenzamine (5
mg/kg i.m. once daily) before and during intraperitoneal Ang II infusion. In preliminary experiments,
this dose of propranolol maximally suppressed both
the hypotensive and chronotropic responses to 1
mg/kg s.c. isoproterenol. It also prevented myocyte
necrosis and myocardial hypertrophy induced by this
dose of isoproterenol.21 The dose of phenoxybenzamine (t112=24 hours) was in excess of the 1 mg/kg
required to maximally reduce the pressure response
to 400 ng norepinephrine.29
To minimize weight loss, potassium supplements
(0.3% in drinking water) were given after the operation to rats receiving Ang ll.30 Systolic blood pressures, using standard tail-cuff manometry,31 and body
weights were measured at 1-2-day intervals. Control
rats and those receiving propranolol and phenoxybenzamine were killed 1 and 2 days after insertion of
the pumps.
Animals were killed with methohexital (35-50
mg/kg i.p.), the thorax was opened, and the heart was
quickly excised and placed in ice-cold saline. The
great vessels and the atria were carefully trimmed
away, and the ventricles were divided into right and
left (plus septum) ventricles and weighed.
Abdominal Aorta Banding With and Without
Renal Ischemia
To stimulate increased endogenous production of
Ang II, we used a model of renovascular hypertension. Since unilateral renal ischemia in a two-kidney
animal is accompanied by systemic hypertension, any
observation regarding cell necrosis might be interpreted as secondary to either increased circulating
Ang II, hypertension, or both. In an attempt to
distinguish these factors, five groups of rats were
used. These groups were constructed in the following
manner after laparotomy following anesthesia with
35-50 mg/kg i.p. methohexital: In the first group
(n=20), abdominal aorta constriction to a diameter
of 0.8 mm together with constriction of the right
renal artery was performed. This model of renovascular hypertension has been used widely in our
laboratory and is known to lead to progressive atrophy of the right kidney and hypertrophy of the
contralateral nonischemic kidney.32 A second group
(n=5) was similarly operated, except in this group
the banding procedure did not involve the renal
arteries, and neither kidney appeared ischemic. In a
third group (n = 10) of rats, the abdominal aorta and
right renal artery were constricted as above, and
endogenous catecholamine stores were depleted with
a course of reserpine33 before inducing renovascular
hypertension. The pretreatment protocol consisted
Tan et al Ang II Cardiotoxicity
of reserpine (2 mg/kg i.p. on day 1 and 1 mg/kg i.p.
on day 2); aortic constriction with right renal stenosis
was performed on day 3; and 0.2 mg/kg/day reserpine was given on postoperative days. A fourth group
(n=8) with abdominal aorta and right renal artery
constriction was pretreated for 2 days with captopril,
an angiotensin converting enzyme inhibitor, in the
drinking water (2 g/l), and this was continued after
the operation. The captopril dosage was equivalent
to 65 mg/day based on the mean consumption of 32
ml water/day. A fifth group of rats (n =5), which was
surgically treated as described above except the
ligature was not tightened, served as a set of sham-
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
operated controls.
All rats were killed 1-7 days after abdominal aorta
constriction. At this time, the rats were anesthetized
with methohexital (35-50 mg/kg i.p.) and artificially
ventilated via tracheostomy. A carotid artery was
cannulated, and arterial pressure was measured. To
approximate the pressure of conscious animals, the
reading was taken when arterial pressure was stable
for a level of anesthesia that was light but sufficient to
prevent spontaneous physical motion by the rat. The
rat was given further anesthesia before thoracotomy
was performed. After thoracotomy, the heart was
rapidly excised and placed in ice-cold saline. The
great vessels and the atria were removed, and the
ventricles were divided and weighed.
Determination of Myocyte Injury
Evaluation of myocyte injury resulting from Ang II
administration was accomplished by two techniques.
The first involved in vivo labeling of myocytes with
abnormal sarcolemmal permeability using monoclonal anticardiac myosin.21,28 Each rat received 1 mg
immunoglobulin G fraction of monoclonal antibody
CCM-52 intraperitoneally 24 hours before death.
The antibody used in this study, CCM-52, has been
shown to be specific for cardiac myosin.27 Antibody
administered intraperitoneally appears in the serum
in less than an hour, reaches a maximal serum -titer by
3 hours, and remains well above a 1:100 dilution titer
necessary for immunofluorescent staining for more
than a week (authors' unpublished data). After removal of the heart, 2-mm-thick transverse sections of
the left ventricle were quick-frozen in isopentane
cooled to the freezing point (- 160°C) in liquid
nitrogen. After cryostat sectioning of the heart, the
sections were stained with fluorescein isothiocyanate-conjugated rabbit anti-mouse immunoglobulin
G (Cappel Laboratories, Cochranville, Pa.) to localize myocyte binding antimyosin monoclonal antibody.
Immunofluorescent staining was evaluated with a
microscope (Carl Zeiss, Inc., Thornwood, N.Y.)
equipped with epifluorescent optics optimized for
fluorescein isothiocyanate fluorescence.
The second technique used to assess myocyte
injury was based on light microscopic examination of
formalin-fixed left ventricular tissue sections stained
with hematoxylin and eosin. Specifically, we examined a coronal section of the ventricle (free wall and
1187
septum) for evidence of myocytolysis, based on a
disruption of the internal cellular architecture. The
entire left ventricular myocardium was scanned in a
clockwise fashion beginning at the endocardial surface of the septum. Whenever an area of myocytolysis
was found, its location and approximate size with
respect to the starting point were noted on a standardized diagram of a left ventricular cross section.
Determination of Wound Healing
DNA synthesis in heart tissue of each experimental group was analyzed by estimation of total incorporation of [3H]thymidine into DNA. [methyl3H]Thymidine (TRK.686, specific activity >50 Ci/
mmol, Amersham, Arlington Heights, Ill.) was
administered by intraperitoneal injection (1 ,Ci/g
body wt) 4 hours before death. Thymidine incorporation into DNA was determined as described by
Benjamin et al.21
Immediately after the heart was excised and
weighed, the right and left ventricles were finely
minced with scissors and homogenized with a Polytron (Brinkmann Instruments, Inc., Westbury, N.Y.)
in 2 ml ice-cold phosphate-buffered saline. Two
milliliters of cold 10% trichloroacetic acid was added
to the homogenate to precipitate DNA and proteins.
The tissue homogenates were centrifuged and
washed three times with 3 ml cold 5% trichloroacetic
acid and then with a final rinse of 3 ml of 2% sodium
acetate in 90% ethanol to remove unincorporated
[3H]thymidine. After centrifugation, the supernatant
was discarded, and the precipitates were resuspended in 2 ml of 5% trichloroacetic acid. DNA was
extracted by heating the samples at 90°C for 30
minutes. After centrifugation, aliquots of the supernatant were taken for determination of tritium radioactivity by scintillation counting and of DNA content
by the method of Burton.34 DNA standards were
prepared from calf thymus DNA, and the concentrations of the standard were determined by its absorbance at 260 nm. Stimulation of DNA synthesis was
determined from the specific radioactivity (cpm/,g
DNA) in each tissue sample.
Subsequent scarring of the myocardium was examined using a collagen-specific stain, sirius red F3BA.
Here coronal sections of the left ventricle were
examined at day 14 of the Ang II infusion and after
2 weeks of renovascular or nonrenovascular hypertension. Similar analysis of myocardium was performed in control rats.
The slide was placed in a projecting microscope, and
the image was projected onto a digitizing pad. The
entire left ventricular cross section was scanned (magnification, x40) with the investigator (R.P.) blinded as
to the source of the tissue. Whenever an area of
replacement fibrosis, as represented by the presence of
confluent fibrillar collagen in regions formerly occupied
by myocytes, was found, its contour was traced, and the
area within the contour was calculated. The entire area
of muscle was also obtained, and the collagen volume
fraction of replacement fibrosis was calculated as the
1188
Circulation Research Vol 69, No 5 November 1991
sum of all sampled areas containing lesions divided by
the sum of all areas with and without lesions. The
density of fibrotic areas was expressed as number of
foci per square millimeter.
Statistical Analysis
The grouped results are expressed as mean±SEM
and were compared using the Kruskal-Wallis analysis
of variance (BMDP, Los Angeles). When a significant intergroup difference was found, then pairwise
comparisons were performed by calculating confidence intervals for the rank differences. Statistical
significance was set at p<0.05.
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Results
Cardiac Myocyte Necrosis After Ang II Infusion
Antimyosin labeling of cardiac myocytes was observed within 24 hours of intraperitoneal infusion of
Ang II. A scattered multifocal distribution of labeled
myocytes was found throughout the myocardium (see
Figure 1), with some tendency toward clustering of
antibody-labeled myocytes around the perivascular
regions of intramyocardial arteries and arterioles. In
contrast to antibody labeling after isoproterenol
treatment,2' labeled myocytes resulting from Ang II
infusion were not preferentially distributed in the
subendocardial region. The number of labeled myocytes peaked between days 1 and 2 and declined
thereafter. At no time was there any myocyte labeling
detected in rats treated with minipumps containing
the O.O1N acetic acid vehicle.
Myocytolysis was evident on light microscopic
examination on day 2 of the Ang II infusion; it was
not observed in control rats. Myocyte necrosis was
patchy and distributed throughout the myocardium
without a discernible pattern of distribution. Occasionally, large numbers of myocytes were involved
(see Figure 2, top panel). In most cases, however,
myocytolysis was confined to small numbers of cells
and accompanied by fibroblast proliferation (see
Figure 2, bottom panel).
Effect of Adrenergic Hormone Antagonists
To rule out the possibility that myocyte necrosis in
the heart occurred secondary to enhanced adrenergic
hormone release facilitated by Ang II, we used
adrenergic hormone antagonists to block this effect.
The prior and concurrent treatment with a- and
,B-adrenergic inhibitors did not attenuate the necrotic
effects of infused Ang II. The pattern and extent of
myocyte necrosis with Ang II and the combination of
propranolol and phenoxybenzamine were similar to
those seen with Ang II alone. It should be noted that
the same level of propranolol applied in a prior
study21 completely blocked the necrogenic effects of 1
mg/kg isoproterenol. Thus, Ang II-mediated myocyte necrosis observed in this study did not result
from secondary adrenergic hormone activity.
Cardiac Myocyte Necrosis After Abdominal
Aorta Banding
Since cardiac myocyte necrosis was detected after
intraperitoneal Ang II infusion, we sought to determine whether Ang II arising from stimulation of the
renin-angiotensin system after renal ischemia might
generate a similar response. Thus, we examined
hearts of rats treated with monoclonal antimyosin at
1-7 days after surgical induction of renovascular
hypertension. To evaluate whether the myocardial
response resulted from surgical manipulation, aortic
constriction, or renin production after renal ischemia, myocyte injury was compared in sham-operated and abdominal aorta-banded rats with and
without right renal artery constriction.
At the time of the study, rats with banding of the
abdominal aorta including constriction of the right
renal artery had an ischemic-appearing right kidney;
the left kidney appeared well perfused. In rats without renal ischemia, both kidneys appeared well perfused. In rats with renovascular hypertension, antimyosin-labeled necrotic myocytes were found
throughout the myocardium. Labeling was most numerous on postoperative day 1, and the pattern and
distribution of labeling were similar in appearance to
those seen with Ang Il-induced necrosis. In rats
without renal ischemia, no myocyte necrosis was
detected. Prior depletion of endogenous catecholamine stores with reserpine pretreatment did not
attenuate the cardiac myocyte necrosis associated
with renal artery constriction.
In renovascular hypertension, light microscopic evidence of myocytolysis was seen to involve individual
myocytes in a diffuse, random distribution throughout
the myocardium. This was not the case in nonrenovascular hypertension, where myocytes appeared intact.
Effect of Captopril
Further evidence that Ang II was the active factor in
generation of myocyte injury was provided by evaluation of the effect of captopril on myocyte necrosis in
rats with abdominal aorta constriction. When rats having abdominal aorta constriction with renal ischemia
were treated with captopril, antibody-labeled cardiac
myocytes were not seen on serial sections.
Increased Cardiac DNA Synthesis
A direct response to cardiac myocyte injury is
cardiac fibroblast proliferation, which may be detected with great sensitivity by the increase in cardiac
DNA synthesis rates. Thus, an independent confirmation of myocyte injury is provided by assessment of
rates of tritiated thymidine incorporation. We observed that the rate of ['H]thymidine incorporation
closely paralleled the appearance of necrotic myocytes in the heart after Ang II infusion. The rate of
['H]thymidine incorporation in the left ventricle increased slightly on the first day of Ang II infusion,
reached a peak on day 2, and declined rapidly
thereafter (Figure 3).
Tan et al Ang II Cardiotoxicity
1189
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
FIGURE 1. Photomicrographs showing angiotensin II-induced cardiac myocyte necrosis. The necrotic cells were labeled by in vivo
administration of monoclonal antimyosin and postmortem staining of tissue sections with fluorescein isothiocyanate-conjugated
rabbit anti-mouse immunoglobulin G. Permeable necrotic cells admit the monoclonal antibody and are thus stained with the
fluorescein isothiocyanate-conjugated secondary antibody and appear as the bright cells in the photographs. Magnification,
x148.75 for top panel and x374 for bottom panel.
The time course of stimulation of DNA synthesis
in renal ischemia was similar to that seen with Ang II
infusion, being maximal on postoperative day 2. The
peak rate of DNA synthesis on day 2 was also of the
same order of magnitude as that seen with Ang II
infusion (Figure 4). In rats with aortic banding
without right renal artery constriction, the DNA
synthesis rate was not different from that of controls
(19.5+±5.5 cpm/gg for banded rats versus 11.4±2.3
cpm/gg for controls). In rats with abdominal aorta
1190
Circulation Research Vol 69, No 5 November 1991
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
FIGURE 2. Light microscopic evidence of myocytolysis observed on day 2 of angiotensin II infusion together with inflammatory
cell and fibroblast infiltration in the region of cell injury. Occasionally, large numbers of myocytes were involved (top panel). In
most cases, however, myocytolysis was confined to small numbers of cells and was accompanied by mononuclear cell infiltrates and
fibroblast proliferation (bottom panel). Stain, hematoxylin and eosin; magnification, x36 for top and bottom panels and x 90 for
insert to bottom panel.
Tan et al Ang II Cardiotoxicity
Rates of DNA Synthesis during AII Infusion
1191
251
20
t)o
N.
8
0.
04
15
l40
w
W
k-
w
¢
10
LL
ii
U
We
0.4
C)
5
a
1
0
2
4
6
8
DAYS
10
12
14
FIGURE 3. Graph showing the specific radioactivity of
[3Hlthymidine incorporation into DNA produced by angiotenDownloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
sin II (AII) infusion for 1-14 days (day O=sham-operated
rats). Rats were pulse-labeled for 4 hours with [3H]thymidine
on the days indicated. After death by methohexital, DNA was
extracted, and the specific radioactivity (cpm/pg DNA) was
determined. Elevated specific radioactivities indicate increased
cell proliferation. *p<0.05 compared with day 0.
constriction and right renal ischemia, which were also
treated with captopril, the rate of DNA synthesis was
not elevated and was similar to that observed in
sham-operated rats (Figure 4).
Replacement Fibrosis
In picrosirius red-prepared sections, punctate foci of
microscopic scarring were evident throughout the left
ventricular myocardium. The number of scars per
square millimeter found in hearts receiving the Ang II
infusion for 14 days was 0.5+±0.1, and the average size
of the scars was 2,950±500 gm2. As shown in Figure 5,
the majority of scars were <4,500 ,um2; hence, the
reparative fibrosis occurred in only 0.12+0.02% of the
myocardium and is in keeping with necrosis of myocytes. No scarring was found in control hearts.
Effects of Ang II and Abdominal Aorta Constriction
on Blood Pressure
Intraperitoneal Ang II infusion did not cause a
significant rise in arterial pressure during the first 3
o0
625
1500
2600
3300
4500
5300
6800
SCAR SIZE (pm2)
FIGURE 5. Bar graph showing frequency distribution of
microscopic scars found scattered throughout the myocardium
in rats receiving intraperitoneal angiotensin II infusion for 14
days. The number and size of these scars were detected by the
picrosirius red technique.
days of infusion. Thereafter, a gradual rise in systolic
pressure was observed, which reached a peak on the
10th day (Figure 6), when systolic arterial pressure
exceeded 200 mm Hg.
With surgical constriction of the abdominal aorta,
the systolic carotid artery pressure immediately before death was 187+10 mm Hg in rats with right
renal artery constriction and 173 + 10 mm Hg in rats
without right renal constriction as compared with
138 +3 mm Hg obtained in the sham-operated controls. The rate of increase in blood pressure was not
determined after abdominal aorta constriction, since
tail-cuff pressures could not be taken after this
procedure.
Pretreatment with captopril in the abdominal
aorta constriction with renal ischemia model prevented the elevation in systolic blood pressure
(119±4 mm Hg) as compared with the untreated
group (187 mm Hg).
Effects of Ang II and Abdominal Aorta Constriction
on Myocardial Hypertrophy
The combined ventricular weights normalized to
body weights were increased by 10-12% during Ang
Systolic BP during AII infusion
150-
100
*
200
R
0..
U
to
50.
0
M
0.
u)
150-
U)
0o
SHA AAE +RI
RI R+C
FIGURE 4. Bar graph showing DNA specific radioactivity on
day 2 in rats treated with angiotensin II (A II), aortic banding
with (+) and without (-) renal ischemia (RI), and renal
ischemia with concurrent captopril (RI+C). Values are
mean+±SEM. *p<0.05 vs. sham-operated rats (SHAM).
100-
.
0
2
4
6 8
DAYS
10
12
14
FIGURE 6. Graph showing mean+±SEM systolic arterial
blood pressure (BP) during 14 days of angiotensin II (AII)
infusion. *p<0.05 vs. day 0.
1192
Circulation Research Vol 69, No 5 November 1991
II infusion relative to controls, reaching a peak at day
14 on continued infusion. The ventricular weight
normalized to body weight increased from 2.98+±0.05
g/kg in controls to 3.41±0.13 g/kg after 2 weeks of
sustained Ang II infusion. A similar observation was
noted in the rats with abdominal aorta banding, with
normalized ventricular weights rising from 2.98± 0.05
g/kg in control rats to 3.76±0.20 g/kg on day 7 after
the operation. The ventricular weight/body weight
ratios were not increased after renal ischemia in the
group treated with captopril (2.63±0.06 g/kg).
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Discussion
Association ofAng II With the Genesis of Cardiac
Myocyte Necrosis
Using in vivo labeling with monoclonal antimyosin,
we have shown that both Ang II infusion and endogenous Ang II, arising as a consequence of renal
ischemia, are associated with the development of
cardiac myocyte necrosis. This necrosis does not
appear to be mediated by either adrenergic activation or systemic hypertension, which are both known
to result from Ang II. Furthermore, when augmented
synthesis of Ang II by concurrent captopril treatment
in the renovascular hypertension model was prevented, myocyte injury and wound healing were not
seen.
The development of cardiac myocyte injury in the
setting of renal ischemia is consistent with earlier
reports.35-37 Infusion of ischemic renal extracts was
shown to produce myocyte injury; however, this injury was absent with removal of the ischemic kidney.37 Whether this effect could be attributed to
endogenous angiotensin has not been established.38
Our experiments provide new evidence for the role of
endogenous Ang IL in the genesis of cardiac myocyte
necrosis in renovascular hypertension and in congestive heart failure, where endogenous circulating Ang
II achieves levels similar to those seen in this study
(see below).
Stimulation of DNA synthesis and scar tissue
formation serve as an independent confirmation of
myocyte necrosis and are indicative of the woundhealing process that involves fibroblast proliferation
and collagen replacement of cell loss. Both Ang II
infusion and renovascular hypertension resulted in
enhanced myocardial DNA synthesis and replacement fibrosis. The rate of increase and scarring
closely paralleled the pattern of necrosis observed
with antimyosin labeling.
The fact that concurrent treatment with captopril (by
preventing excess Ang II synthesis) abolished myocyte
necrosis and the subsequent increase in DNA synthesis
rates provides further evidence that elevated endogenous Ang II was responsible for abnormal sarcolemmal
permeability and myocytolysis. The cardioprotective
effect of captopril is of clinical importance, in that the
progression of cardiac dysfunction secondary to myocyte loss induced by elevation of endogenous Ang II
levels can be effectively treated by angiotensin converting enzyme inhibition.
Ang II-Mediated Necrosis Secondary to Hypertension
Previous studies have shown that large doses of
Ang II can cause focal myocardial injury.26,39 It is
currently believed that this injury occurs in close
conjunction with the hypertension induced by Ang
11.25 Using a model of renovascular hypertension,
Bhan et a135 observed foci of myocardial damage in
the subendocardium of the left ventricle; these foci
were thought to be consistent with ischemic or anoxic
or reflow injury. They also observed vascular lesions
in epicardial and intramyocardial arteries that were
related to the severity of hypertension in the case of
epicardial vessels and to the vasoconstrictive action
of Ang II in the case of intramural vessels.
Bhan et al40 subsequently extended their study and
showed that when the hypertensive effects of 1,700
ng/kg/min i.v. Ang II were counteracted by concurrent phentolamine treatment, Ang II produced vascular injury confined to intramyocardial coronary
arteries, with minimal evidence of myocyte injury.
This suggests that irreversible myocardial injury is
dependent on the hypertensive and vasoconstrictive
effects of Ang II. Our study is the first to demonstrate
that Ang II given in a manner that does not induce
acute hypertension is involved in the generation of
cardiac myocyte necrosis, suggesting that the mechanisms producing necrosis are independent of the
hypertensive effects of Ang II. In fact, as arterial
pressure gradually rose with the sustained infusion of
Ang II, no new injury could be detected in our study.
Additional corroborative evidence supporting the
dissociation between necrosis and hypertension is
provided by evidence that abdominal aorta constriction applied in a manner that did not produce renal
ischemia results in hypertension but not in myocyte
necrosis. It is not clear why myocyte necrosis in the
absence of hypertension was detected in our study,
whereas it appears not to have been detected in the
study of Bhan et al.40
Plasma Ang II Levels
The dose range of Ang II infusion used in our
experiment (0.6-1.0 ,ug/kg/min) was only marginally
smaller than the ones used in other studies (0.9-1.8
j.tg/kg/min).26,39 However, we used an intraperitoneal rather than intravenous route of administration.
Plasma Ang II levels of 70.6±6.4 pg/ml were determined after 24 hours of intraperitoneal infusion. This
was similar to the levels of plasma Ang II seen at 24
hours in the rats with abdominal aorta banding and
renal ischemia (56.7±21.9 pg/ml). These levels are
well within the pathophysiological range reported in
cases of essential hypertension (147-189 pg/ml),41
renal hypertension (154+±93 pg/mI),42 and severe
heart failure (28-155 pg/ml).43,44 Our normal control
rats had levels of 9.5±3.5 pg/ml.
Intravenous infusion of Ang II at 200 ng/min
produced plasma levels in excess of 400 pg/ml and an
Tan et al Ang II Cardiotoxicity
acute hypertensive response, whereas intraperitoneal
infusion at the same dose produced lower plasma
levels and no acute hypertensive response. The differences may be due to the inactivation of the
intraperitoneal Ang II by angiotensinase in the liver,
lungs, and kidneys.43 Whether Ang II metabolites are
responsible for the observed lesions cannot as yet be
ascertained and would form an essential part of
future studies to elucidate the mechanism(s) of Ang
Il-induced cardiomyocyte necrosis.
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Relation Between Necrotic and Vasculotoxic Effects
of Ang II
Epicardial coronary artery constriction has been
observed with intracoronary infusion of Ang II.45
Whether the mechanism of necrosis involved vasculotoxic or vasoconstrictive effects of Ang II on intramyocardial coronary arteries has not been resolved. Giacomelli et a125 showed that foci of
myocardial injury produced by intravenous infusion
of 1,700 ng/kg/min Ang II were observed in areas
that have no consistent topographic relation to these
intramural arteries. They speculated that the myocardial damage was secondary to ischemic or anoxic
injury, with or without reflow damage. Bhan et al35,40
have suggested a similar relation between the vasculotoxic and myocyte necrotic effects of Ang II. Although these studies suggest a vasculotoxic effect,
they do not conclusively demonstrate that the myotoxic effect of Ang II is secondary to vascular effects.
Ideally, to demonstrate that the myocardial lesions
are secondary to coronary vascular lesions, one needs
to show that myocyte necrosis resides in regions
distal to the vascular lesions. The correlative method
used in the study of Bhan et al,40 who showed
increased myocardial lesions in hearts with increased
vascular lesions, cannot lead to any firm conclusion
about the spatial relation of the lesions.
Permeability of systemic and intramural arteries is
abnormally increased by Ang 11.25,46-48 Thus, myocytes adjacent to arterioles may be subject to a
cascade of events, including the release of vasoactive
amines,49 that leads to sarcolemmal disruption and
ultimately cell death. Further studies are required to
elucidate the relation between augmented vessel
permeability, release of vasoactive amines, and cell
death due to low-dose Ang II infusion.
Adrenergic Activation as a Mechanism
Several lines of evidence suggest that angiotensininduced necrosis is not mediated by sympathetic activation or catecholamine release. In contrast to the
pattern of necrosis produced after intraperitoneal infusion of 200 ng/min Ang II, isoproterenol characteristically produces subendomyocardial necrosis.1821 In addition, doses of propranolol and phenoxybenzamine
that were effective in inhibiting the hemodynamic response to high levels of circulating catecholamines and
that totally blocked isoproterenol-induced myocardial
injury did not block angiotensin-induced myocyte necrosis. Further evidence that Ang TI-mediated cardio-
1193
myocyte necrosis was not due to secondary catecholamine release is provided by the observation that the
same extent of necrosis occurred after reserpine depletion of catecholamine stores before renal artery constriction, as occurred in the absence of reserpine treatment. Thus, Ang II appeared to produce cardiac
myocyte necrosis independent of the cardiotoxicity that
might result from sympathetic activation or facilitated
release of catecholamines.
Role of Ang II Receptors
Whether the cardiotoxic effects of Ang II are
mediated through myocardial receptors is uncertain.
It is notable that Ang II receptors have not been
demonstrated at the same level in the adult rat heart
as have been found in other species (e.g., rabbit,
guinea pig, and bovine myocardium; T.B. Rodgers,
personal communication). The explanation is unclear, especially in view of the fact that the same
techniques have demonstrated a high density of
angiotensin receptors in neonatal rat cardiac myocytes.50 Therefore, we cannot suggest that cardiac
myocyte necrosis resulting from Ang II infusion or
abdominal aorta constriction is mediated by Ang II
binding to cardiac myocyte receptors. Robertson and
Khairallah51 reported binding of Ang II in nuclei of
rat cardiac muscle. Whether nuclear or other hitherto unknown types of binding play a role in mediating Ang II effects in adult rat myocardium is unclear.
That the adult rat heart is responsive to Ang II was
demonstrated by Xiang et al,52 who showed that Ang
II added to the perfusate in Langendorff preparations increased contractile force in adult rat hearts.
Conclusion
We have demonstrated that Ang II arising either
from endogenous or exogenous sources is associated
with the appearance of abnormal sarcolemmal permeability, myocytolysis, fibroblast proliferation, and subsequent replacement fibrosis. It is not clear whether Ang
II acts directly through myocyte receptors to produce
this effect nor whether a metabolite of Ang II might be
the active factor in this process. Our data do appear to
exclude secondary adrenergic stimulation and arterial
hypertension as being responsible for this effect. The
mechanism by which Ang IL leads to myocyte necrosis
remains to be determined.
Acknowledgments
We thank Smith Kline & French for the supply of
phenoxybenzamine and Dr. Ron White of E.R. Squibb
& Sons for conducting the plasma Ang II assays.
References
1. Laragh JH: Hormones in the pathogenesis of congestive heart
failure: Vasopressin, aldosterone, angiotensin II. Circulation
1962;25:1015-1023
2. Harris P: Evolution and the cardiac patient. Cardiovasc Res
1983;17:437-445
3. Francis GS: Neurohumoral mechanisms involved in congestive
heart failure. Am I Cardiol 1985;55:15A-21A
1194
Circulation Research Vol 69, No 5 November 1991
4. Harris P: Congestive cardiac failure: Central role of the
arterial blood pressure. Br Heart J 1987;58:190-203
5. Cohn JN, Levine TB, Olivary MT, Garberg V, Lura D,
Francis GS, Simon AB, Rector T: Plasma norepinephrine as
a guide to prognosis in patients with chronic congestive heart
failure. N Engl J Med 1984;311:819-823
6. Hasking GJ, Esler MD, Jennings GL, Burton D, Johns JA,
Korner PI: Norepinephrine spillover to plasma in patients
with congestive heart failure: Evidence of increased overall
and cardiorenal sympathetic nervous activity. Circulation 1986;
73:615-621
7. Leimbach WN, Wallin G, Victor RG, Aylward PE, Sundlof G,
Mark AL: Direct evidence from intraneural recordings for
increased central sympathetic outflow in patients with heart
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
failure. Circulation 1986;73:913-919
8. Cleland JGF, Dargie HJ, Hodsman GP, Ball SG, Robertson
JIS, Morton JJ, East BW, Robertson I, Murray GD, Gillen G:
Captopril in heart failure: A double blind controlled trial. Br
Heart J 1984;52:530-535
9. Cleland JGF, Dargie HJ, Robertson I, Robertson JIS, East
BW: Total body electrolyte composition in patients with heart
failure: A comparison with normal subjects and patients with
untreated hypertension. Br Heart J 1987;58:230-238
10. Chidsey CA, Harrison DC, Braunwald E: Augmentation of the
plasma norepinephrine response to exercise in patients with
congestive heart failure. N Engl J Med 1962;267:650-654
11. Francis GS: Hemodynamic and neurohumoral responses to
dynamic exercise: Normal subjects versus patients with heart
disease. Circulation 1987;76(suppl VI):VI-11-VI-17
12. Re RN: Cellular mechanisms of growth in cardiovascular
tissue. Am J Cardiol 1987;60:1041-1091
13. Upsher ME, Khairallah PA: ,B-Adrenergic receptors in rat
myocardium during the development and reversal of hypertrophy and following chronic infusions of angiotensin II and
epinephrine. Arch Int Pharmacodyn 1985;274:65-79
14. Khairallah PA, Kanabus J: Angiotensin and myocardial protein synthesis, in Tarazi RC, Dunbar JB (eds): Perspectives in
Cardiovascular Research. New York, Raven Press, Publishers,
1983, vol 8, pp 337-347
15. Chappel CI, Rona G, Balazs T, Gaudry R: Comparison of
cardiotoxic actions of certain sympathomimetic amines. Can J
Biochem Physiol 1959;37:35-42
16. Rosenblum I, Wohl A, Stein AA: Studies of cardiac necrosis:
I. Production of cardiac lesions with sympathomimetic amines.
Toxicol Appl Pharmacol 1965;7:1-8
17. Todd GL, Cullan GE, Cullan GM: Isoproterenol-induced
myocardial necrosis and membrane permeability alterations in
the isolated perfused rabbit heart. Exp Mol Pathol 1980;33:
43-54
18. Collins P, Billings CG, Barer GR, Daly JJ, Jolly A: Quantitation of isoprenaline-induced changes in the ventricular myocardium. Cardiovasc Res 1975;9:797-806
19. Reichenbach DD, Benditt EP: Catecholamines and cardiomyopathy: The pathogenesis and potential importance of myofibrillar degeneration. Hum Pathol 1970;I:125-150
20. Bloom S, Cancilla PA: Myocytolysis and mitochodrial calcification in rat myocardium after low doses of isoproterenol. Am
JPathol 1969;54:373-391
21. Benjamin IJ, Jalil JE, Tan LB, Cho K, Weber KT, Clark WA:
Isoproterenol-induced myocardial fibrosis in relation to myocyte necrosis. Circ Res 1989;65:657-670
22. Alpert LI, Pai SH, Zak FG, Wertheimer S: Cardiomyopathy
associated with pheochromocytoma. Arch Pathol 1972;93:
544-548
23. Cho T, Tanimura A, Saito Y: Catecholamine-induced cardiopathy accompanied with pheochromocytoma. Acta Pathol Jpn
1987;37:123-132
24. Rona G: Catecholamine cardiotoxicity. J Mol Cell Cardiol
1985;17:291-306
25. Giacomelli F, Anversa P, Wiener J: Effect of angiotensininduced hypertension on rat coronary arteries and myocardium. Am J Pathol 1976;84:111-138
26. Gavras H, Brown JJ, Lever AF, Macadam RF, Robertson JIS:
Acute renal failure, tubular necrosis, and myocardial infarc-
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
tion induced in the rabbit by intravenous angiotensin II.
Lancet 1971;2:19-22
Clark WA, Chizzonite RA, Everett AW, Rabinowitz M, Zak R:
Species correlations between cardiac isomyosins: A comparison
of electrophoretic and immunological properties. J Biol Chem
1982;257:5449-5454
Nolan AC, Clark WA, Karwoski T, Zak R: Patterns of cellular
injury in myocardial ischemia determined by monoclonal
antimyosin. Proc Natl Acad Sci U S A 1983;80:6046-6050
Paller MS, Linas SL: Role of All, a-adrenergic system, and
arginine vasopressin on arterial pressure in rat. Am J Physiol
1984;246:H25-H30
Garwitz ET, Jones AW: Aldosterone infusion into the rat and
dose-dependent changes in blood pressure and arterial ionic
transport. Hypertension 1982;4:374-381
Lovenberg W: Techniques for the measurement of blood
pressure. Hypertension 1987;9(suppl I):I-5-I-6
Weber KT, Janicki JS: Angiotensin and the remodeling of the
myocardium. Br J Clin Pharmacol 1989;28:141S-150S
Antonaccio MJ, Smith CB: Effects of chronic pretreatment
with small doses of reserpine upon adrenergic nerve function.
J Pharmacol Exp Ther 1974;188:654-667
Burton K: A study of the conditions and mechanism of the
diphenylamine reaction for colorimetric estimation of deoxyribonucleic acid. Biochem J 1956;62:315-323
Bhan RD, Giacomelli F, Wiener J: Ultrastructure of coronary
arteries and myocardium in experimental hypertension. Exp
Mol Pathol 1978;29:66-81
Winternitz MC, Mylon E, Waters LL, Katzenstein R: Studies
on the relation of the kidney to cardiovascular disease. Yale J
Biol Med 1940;12:623-679
Muirhead EE: Renal tissue and extracts vs cardiovascular
injury. Arch Pathol 1963;76:613-619
Fernandez LA, Downing E: Cardiomyopathy produced in rats
with acute renal hypertension. J Lab Clin Med 1980;95:
159-167
Gavras H, Kremer D, Brown JJ, Gray B, Lever AF, MacAdam
RF, Medina A, Morton JJ, Robertson JIS: Angiotensin- and
norepinephrine-induced myocardial lesions: Experimental
and clinical studies in rabbits and man. Am Heart J 1975;89:
321-332
Bhan RD, Giacomelli F, Wiener J: Adrenergic blockade in
angiotensin-induced hypertension: Effect on rat coronary
arteries and myocardium. Am J Pathol 1982;108:60-71
Dusterdieck G, McElwee G: Estimation of angiotensin II
concentration in human plasma by radioimmunoassay: Some
applications to physiological and clinical states. Eur J Clin
Invest 1971;2:32-38
Catt EJ, Cain MD, Coghlan JP, Zimmet PZ, Cran R, Best JB:
Metabolism and blood levels of angiotensin II in normal
subjects, renal disease and essential hypertension. Circ Res
1970;26/27(suppl II):II-177-II-193
Staroukine M, Devriendt J, Decoodt P, Verniory A: Relationship between plasma epinephrine, norepinephrine, dopamine
and angiotensin II concentrations, renin activity, hemodynamic state and prognosis in acute heart failure. Acta Cardiol
1984;39:131-138
Kuroda T, Shida H: Angiotensin II-induced myocardial damage with a special reference to low cardiac output syndrome.
Jpn Heart J 1983;24:235-243
Cohen MV, Kirk ES: Differential response of large and small
arteries to nitroglycerin and angiotensin. Circ Res 1973;33:
445-453
Robertson AL, Khairallah PA: Effects of angiotensin II and
some analogues on vascular permeability in the rabbit. Circ
Res 1972;31:923-931
Giese J: Renin, angiotensin and hypertensive vascular damage: A review. Am J Med 1973;55:315-332
Wiener J, Giacomelli F: The cellular pathology of experimental hypertension: VII. Structure and permeability of the mesenteric vasculature in angiotensin-induced hypertension. Am J
Cardiol 1973;72:221-240
Tan et al Ang II Cardiotoxicity
49. Fantone JC, Schier D, Weingarten B: Inhibition of vascular
permeability changes in rats by captopril. J Clin Invest 1982;
69:1207-1211
50. Rogers TB, Gaa ST, Allen IS: Identification and characterization
of functional AII receptors on cultured heart myocytes. J Pharmacol Exp Ther 1986;236:438-444
51. Robertson AL, Khairallah PA: Angiotensin II: Rapid localization in nuclei of smooth and cardiac muscle. Science 1971;
172:1138-1139
1195
52. Xiang J, Linz W, Becker H, Ganten D, Lang RE, Scholkens B,
Unger T: Effects of converting enzyme inhibitors ramipril and
enalapril on peptide action and sympathetic neurotransmission in the isolated heart. Eur J Pharmacol 1985;113:215-223
KEY WORDS * monoclonal antimyosin *
renovascular
hypertension * captopril * myocardial fibrosis
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Cardiac myocyte necrosis induced by angiotensin II.
L B Tan, J E Jalil, R Pick, J S Janicki and K T Weber
Downloaded from http://circres.ahajournals.org/ by guest on May 4, 2017
Circ Res. 1991;69:1185-1195
doi: 10.1161/01.RES.69.5.1185
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1991 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/69/5/1185
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/