1081
Eosinophils From Hypereosinophilic Patients
Damage Endocardium of Isolated Feline
Heart Muscle Preparations
Ajay M. Shah, MB, MRCP, Dirk L. Brutsaert, MD, PhD, Ann L. Meulemans, DRs,
Luc J. Andries, DRs, and Monique Capron, MD
Persistent eosinophilia in humans is often associated with endocardial damage to the heart, but
causal relation has not been established. We investigated the effect of eosinophils and
eosinophil supernatants obtained from eight hypereosinophilic patients on the contractile
performance and endocardial morphology of isolated, electrically stimulated cat papillary
muscle preparations (n=16). All these eosinophil suspensions contained high proportions of
"hypodense" or "activated" cells. Eosinophils (5-15 106/10 ml organ bath) or eosinophil
culture supernatants (prepared by overnight incubation at 370 C) when added to papillary
muscles produced acute changes in contractile behavior of these muscles identical to the
previously reported effects of selective endocardial damage: a reduction in time to peak
isometric twitch tension causing a reduction in peak isometric tension but with no significant
reduction in rate of tension development or in maximum unloaded shortening velocity. All of
these muscle preparations showed severely damaged endocardium at scanning electron
microscopy. Addition of eosinophils from hypereosinophilic patients to muscles with selectively
damaged endocardium (by previous transient [1-second] exposure to 1% Triton X-100)
produced no further change in contractile performance. No significant change in contractile
performance or endocardial morphology of papillary muscles (n = 16) was observed after
addition of eosinophils (7.5-lOx 106) or neutrophils (8-5x 106) from normal subjects or of
cell-free culture medium. Thus, activated human eosinophils produce specific morphological
and functional changes suggestive of specific damage to endocardium of isolated feline cardiac
muscle. (Circulation 1990;81:1081-1088)
a
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
X
P ersistent peripheral blood hypereosinophilia is
associated with a high incidence of tissue
damage.1,2 The cardiac disease associated
with eosinophilia is typically characterized by damage
to the endocardium and adjacent myocardiumendomyocardial fibrosis.34 It has been suggested that
there is a causal link between the presence of eosinophilia and the occurrence of cardiac damage. As a
result of release of granule products, eosinophils
have a high cytotoxic potential against parasites but
also a strong cytotoxic potential against mammalian
target cells (including cultured human atrial
From the Department of Physiology, University of Antwerp,
Antwerp, Belgium.
A.M.S. was a visiting fellow from the Department of Cardiology,
University of Wales College of Medicine, Cardiff, UK, and was
supported by the British Heart Foundation.
Address for correspondence: D.L. Brutsaert, MD, PhD, Professor of Physiology and Medicine, Department of Physiology, University of Antwerp (RUCA), Groenenborgerlaan, 171, 2020 Antwerp, Belgium.
Received February 28, 1989; revision accepted November 22,
1989.
myocytes).5'6 So-called "hypodense" and "degranulated" eosinophils, which are believed to be "activated," and eosinophil granule proteins have been
found in high concentration in the peripheral blood
of patients with clinical cardiac damage and in close
association with these cardiac lesions.27'8 Tai et a19
have reported that eosinophil supernatants and purified eosinophil granule products impair the metabolic performance of isolated rat cardiomyocytes.
Direct investigation of the acute effects of eosinophils and their secretion products on intact multicellular heart preparations has not, to our knowledge, been reported.
Brutsaert and coworkers10-14 have recently
reported that the endocardium exerts a unique modulating influence on the contraction of isolated cardiac muscle. Selective damage to endocardium
results in very characteristic changes in contractile
twitch behavior: an earlier onset of isometric tension
decline with reduction in peak isometric tension but
a relatively minor reduction in the rate of tension
development and no reduction in maximum
unloaded shortening velocity.
1082
Circulation Vol 81, No 3, March 1990
TABLE 1. Characteristics of Hypereosinophilic Subjects
eosinophil
"Hypodense"
Purity of
eosinophil
count
(x 109/l)
3.23
0.95
eosinophils
(%)
81
32
preparation
(%)
70
96
8.44
99
88.5
Colitis, asthenia
1.52
59.6
66
Asthma, nasal polyps,
3.07
48.1
96
neuropathy
Urticaria
1.56
36.4
94
3.43
7.20
87.6
84.7
68
94
Blood
Patient
(initials)
1 M.R.
2 B.D.
Age
(yr)
32
38
Sex
F
F
Diagnosis
Hypereosinophilic syndrome
Taeniasis and trichinosis
3 J.B.
61
M
4 F.D.
63
M
5 M.G.
46
F
"Myeloproliferative"
hypereosinophilic syndrome
Postparasitic
hypereosinophilia
Churg-Strauss syndrome
6 M.T.
36
F
7 R.J.
8 B.B.
60
44
M
M
Atopic hypereosinophilic
syndrome
Filariasis
"Myeloproliferative"
hypereosinophilic syndrome
Clinical features
Neurological
Diarrhea, skin rash,
respiratory
Skin rash, myalgia, anemia
Pruritus
Endomyocardial fibrosis,
diarrhea
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
We have studied the acute effects of activated eosinophils and eosinophil supernatants obtained from
patients with hypereosinophilia on the contractile performance and endocardial morphological integrity of
isolated cat cardiac papillary muscle preparations. As
controls, the effects of eosinophils or neutrophils purified from normal subjects also were studied.
Methods
Subjects
Eosinophils and eosinophil supernatants were
obtained from eight patients with hypereosinophilia.
The clinical characteristics and hematologic data for
these patients are shown in Table 1. None of the
patients had received any treatment for at least 6
months before collection of blood for eosinophil
purification. Only one subject (patient 8) had overt
clinical evidence of cardiac disease. All patients had
eosinophil counts of more than 0.9x109/1 and evidence of eosinophil activation as assessed by the
presence of a high proportion of "hypodense"
eosinophils.615,16 Eosinophils and neutrophils were
also obtained from five normal subjects.
Isolation of Eosinophils and Neutrophils
Peripheral blood eosinophils and neutrophils
were isolated as described previously.15 Briefly,
leukocytes were recovered from heparinized venous
blood by dextran sedimentation of erythrocytes and
were washed in minimum essential medium (MEM;
Difco, Detroit, Michigan) containing 10% inactivated fetal calf serum (FCS). Cells suspended in
MEM-FCS at 5 x 107 cells/ml were then layered on
discontinuous metrizamide gradients (Nyegaard
Co., Oslo, Norway), and cell fractions were collected from each gradient and interface. Percent
purity and morphology of eosinophils or neutrophils was assessed using Giemsa-stained cytocentrifuge preparations and cell viability by trypan blue
dye exclusion. All suspensions used for eosinophil
experiments contained at least 65%, and usually
more than 80%, eosinophils; neutrophil suspensions contained at least 90% neutrophils.
Eosinophil Supematants
Purified eosinophils (5-10x 106/ml) in MEM-FCS
were incubated overnight at 370 C in a 5% CO2 in air
atmosphere. The supernatant was collected by centrifugation and stored at -20° C until use. Viability
of eosinophils after preparation of the supernatants,
evaluated by ethydium bromide, was very high. In
addition, it has been previously shown that there is
no alteration in protein expression when freshly
purified eosinophils are compared with overnightcultured cells.17 Both purification of eosinophils or
neutrophils and preparation of overnight culture
supernatants were performed in the same place.
Freshly purified eosinophils and culture supernatants
or neutrophils were transported on ice to perform
the experiments on physiology and electron microscopy (total transport time, approximately 1 hour).
Papillary Muscle Preparations
Right ventricular papillary muscles (n=35) from 15
pentobarbitone-anesthetized cats were mounted in
10-ml organ baths containing Krebs-Ringer solution
of (mM) NaCI 118, NaHCO3 24, KCI 4.7, KH2PO4
1.2, MgSO4 *7H20 1.2, glucose 4.5, CaCI2 *2H20
1.25, pH 7.4) at 350 C and bubbled with a 95% 02-5%
CO2 gas mixture. Muscles were attached to a lengthtension transducer12 and stimulated electrically at 0.2
Hz and a voltage approximately 10% above threshold
via two longitudinally arranged platinum electrodes.
The basic characteristics of these muscles and the
interventions studied are shown in Table 2.
The following contractile parameters were measured at lm,, (the muscle length at which active
tension development was maximal): total peak isometric twitch tension (TT), peak rate of isometric
tension development (+dT/dt), time to peak isomet-
Shah et al Activated Eosinophils Injure Mammalian Endocardium
TABLE 2. Basic Characteristics of Cat Papillary Muscles
+ dT/dt
TT
L[..,,
(mN/mm2/sec)
(mN/mm2)
n
(mm)
Intervention studied
"Activated" eosinophils
276.2+74.04
19 8.1+0.79 50.5+12.19
and supernatant
261.6+53.81
5 7.4+0.40 44.4±7.29
Normal eosinophils
5 6.8-+0.38 44.5±6.27
258.2+51.43
Normal neutrophils
198.5±50.15
6 7.3±0.78 36.0±6.00
MEM-FCS
tTT
(msec)
RT1/2
(msec)
Vmax
(Lm.x/sec)
1083
PS
(proportion L4)
0.14+0.009
1.32+0.120
0.13±0.004
1.28±0.130
0.14+0.009
1.37+0.215
0.11±0.012
1.17±0.176
peak rate of
twitch
tension;
+dT/dt,
total
peak
isometric
was
maximal;
development
TT,
tension
which
active
at
length
muscle
L.
isometric tension development; tlT, time to peak isometric twitch tension; RT1/2, time to half isometric relaxation; Vm,,,, maximum velocity
of unloaded shortening; and PS, peak isotonic shortening.
Values are mean±SEM at 350 C and 1.25 mM Ca2+.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
ric twitch tension (tTT), time to half isometric relaxation (RT/2), peak isotonic shortening (PS), and
maximum velocity of unloaded shortening (Vma,).
Results were normalized for muscle cross-sectional
area and lmax.12 Only preparations with resting tension less than 15% of total developed tension were
studied. Statistical analysis of changes in contractile
performance was by the Student's paired t test.
Experimental Protocol
After a 2-3-hour equilibration period, single doses
of eosinophil or neutrophil suspension (5-15 x 106 cells;
volume, <500 ,ul), eosinophil supernatant (obtained
from 5-15 x 106 cells) or cell-free culture medium (with
FCS) were added to organ baths containing a papillary
muscle preparation, and any changes in contractile
performance were measured after establishment of a
stable response. Thirty minutes after addition of solution, muscles were fixed in their working position for
scanning electron microscopy. Only one intervention
was studied per muscle preparation.
Electron Microscopy
Preparations were fixed in 1% glutaraldehyde in
0.15 M cacodylate buffer for at least 2 hours and
postfixed with 0.2% osmium tetroxide in cacodylate
buffer for 1 hour. They were then dehydrated in an
ascending series of acetone, critical point-dried in a
Balzers apparatus, mounted on aluminum stubs, and
coated with a 15-nm thick layer of gold. All specimens were examined in a Leitz AMR-1200 -B microscope at 15 kV by one of us (L.A.) who was unaware
of the experimental intervention tested. The morphology of the endocardium was assessed for integrity of endothelial cell membranes and nuclei, presence of endothelial intercellular spaces, and
exposure of and damage to the basement membrane
underlying the endocardial endothelium and then
was quantified by assigning a score of 0-4+ for each
of the above. A score of 0 meant no detectable
abnormality, and a score of 4+ meant severe abnormality or damage.
Results
Effects on Contractile Performance
Addition of eosinophils or eosinophil supernatants
from the hypereosinophilic patients to cat papillary
268.8+18.02 421.1+27.61
265.2+17.35 421.2+22.81
271.8+19.69 420.6+23.78
278.7+29.96 401.3±33.66
muscles (n = 16) resulted, usually within 15 minutes,
in a characteristic alteration of contractile twitch
(Figure 1). The tlT decreased by approximately 7%
(p<0.005), resulting in an earlier onset of isometric
twitch tension decline, and there was a 9% reduction
in TT (p<0.01). RT1/2 also decreased by 5%
(p<0.01), thus contributing to the reduced duration
of twitch. +dT/dt was minimally reduced, and there
was no reduction in Vmax. This characteristic pattern
of "negative inotropic" effect -identical to the effect
of selective endocardial damage12-was observed
after addition of eosinophils or eosinophil supernatant and at all doses tested between 5 and 15 x 106
cells. The mean changes in all measured contractile
F (nnN)
50
0
>
'
0 1.
10
gnnrnc4uVmb
TT
dT
dt
tTT
RT 1
2 Vmax PS
! ~~~~~~~~~~~~.
1.
% DECREASE
FIGURE 1. Top panel: Representative isometric twitches
recorded from a cat papillary muscle showing effect of activated
eosinophils (lOx106 cells). a: Baseline twitch. b: Steady-state
twitch after addition of eosinophils. Bottom panel: Pooled data
showing percent change in contractile parameters after addition
of eosinophils or eosinophil supematant from hypereosinophilic
patients. * *p <O. 01. Total peak isometric twitch tension (TI),
peak rate of isometric tension development (+dT/dt), time to
peak isometric twitch tension (tTT), time to half isometric
relaxation (RT½12), maximum velocity of unloaded shortening
(Vma,x), and peak isotonic shortening (PS).
1084
Circulation Vol 81, No 3, March 1990
TABLE 3. (Percent) Change in Contractile Parameters of Cat Papillary Muscles
PS
RT'/2
tTT
+dT/dt
Vmdx
TT
(Lma,/sec [x10-9])
(msec)
(msec)
(Lm,,/sec [X10-])
(mN/mm2/sec)
(mN/mm2)
Intervention
-0.4±0.1
0.5±1.5
21.5± 4.4
-17.5±4.3*
-8.0 +6.78
-4.2+1.36
"Activated" cosinophils
(-3.8±1.2)
and supernatant
(0.1+1.3)
(-3.8±1.8)
(-6.6+1.3) (-5.4+0.9)
(-8.6±1.5)
0.0±0.1
0.2+2.7
4.2+5.0
-2.4±1.1
4.5±4.31
0.8±0.51
Normal eosinophils
(0..1+0.4)
(0.6+2.8)
(1.1±1.0)
(-0.9+0.4)
(1.6±1.3)
(1.5±1.2)
0.1+0.1
3.8-3.6
-4.8+3.9
-3.2+3.4
1.8+2.35
0.7+0.43
Normal neutrophils
(1.8+1.0)
(-1.2+1.0)
(-1.2+1.2)
(1.81.1)
(-0.6±0.9)
(1.9±1.3)
0.1+0.1
-1.0+ 1.0
-0.5±1.7
0.2±1.9
4.5+4.31
0.1+0.14
MEM-FCS
(0.9±0.6)
(-0.6±0.6)
(-0.2±0.4)
(0.3±0.6)
(1.6±2.3)
(0.3+0.3)
Values are given as meantSEM of raw data, and percent changes are between brackets.
Lmx muscle length at which active tension development was maximal; TT, total peak isometric twitch tension; +dT/dt, peak rate of
isometric tension development; tTT, time to peak isometric twitch tension; RTI/2, time to half isometric relaxation; Vma,, maximum velocity
of unloaded shortening; and PS, peak isotonic shortening.
*p<0.01 cf. baseline values before intervention.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
parameters (pooled data from both groups) are
shown in Figure 1 and Table 3.
Addition of cosinophils (7.5 -lOx 106 cells) or neutrophils (8-15x 106 cells) obtained from normal subjects or of 500 gl MEM-FCS medium produced no
significant change in papillary muscle contractile
behavior (Table 3).
Eosinophils (7.5 x 106 cells per experiment) from
patients 7 and 8 were added to three cat papillary
muscle preparations where endocardium had already
been selectively damaged by transient (1-second)
exposure to 1% Triton X-100 (Sigma, St. Louis, Missouri) as described previously.12 In these preparations,
no significant further change in contractile perfor-
FIGURE 2. Scanning electron micrographs of cat papillary muscles. Intact endocardial endothelium is present in a control muscle
after addition of culture medium (A) and in muscles after treatment with normal neutrophils (B) or normal eosinophils (C and D).
The endothelial cells have intact cell membranes, intercellular borders (arrowheads), and small microvilli (A-D). Scale bars, 5 um.
Shah et al Activated Eosinophils Injure Mammalian Endocardium
1085
TABLE 4. Morphology of Endocardium After Addition of Eosinophils, Eosinophil Supernatants, Neutrophils, or
Cell-Free Culture Medium
Intervention
Activated eosinophils
Subject
Cells (x 106) used
per experiment (n)
Cell membrane
damage
15
15
10
15
5
10
10
7.5
++++
+++
1
1
2
2
3
3
5
8
Eosinophil supernatants
Subject
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
4
4
5
5
6
6
7
8
15
15
5
5
10
7.5
7.5
7.5
++
++
+++
++
Endocardial morphology
Intercellular
Basement membrane
spaces
exposure and damage
+
+
+
+
++
+++
+
+
++
++
++
+++
+++
+
+
+++
++
+
++
+
++
++
+++
++
+++
+++
+++
+
++
++
+
++
+
+++
++++
++
+++
++
++
++
+
+
Cell-free culture medium
0
0
+
0
0
0
0
0
0
+
+
0
0
+
+
+
0
0
0
+
Normal eosinophils
8.5
8.5
mance was observed during 30 minutes' exposure to
eosinophils (tTT increased by 1.3+2.43%, TT by
0.1±0.17%, and RT1/2 by 0.9+1.90%, and Vmax
decreased by 0.2±0.20% [mean±SD]).
Scanning Electron Microscopy
Normal endocardial endothelium of cat papillary
muscle preparations12 mounted for up to 8 hours
consists of a smooth continuous cell sheet of closely
apposed cells with distinct intercellular borders,
intact cell membranes, and a variable number of
small microvilli. Specimens treated with eosinophilfree culture media (n=6; Figure 2A), normal eosinophils (n=5; Figures 2C and 2D), or normal neutrophils (n=5; Figure 2B) had similar appearancespredominantly normal endocardial endothelium but
also the very occasional cell with a damaged cell
surface (Table 4).
Muscle preparations (n=16) exposed to eosinophils or eosinophil supernatants from hypereosinophilic patients had severely damaged endocardium
(Figure 3 and Table 4). Both eosinophils (Figures 3A
0
and 3B) and their supernatants (Figures 3C and 3D)
typically caused a heavily extracted appearance of the
endocardial cells. This extraction of the endocardial
cells ranged from multiple holes in the cell membrane to a "pepper-pot" appearance and a more
severe, nearly complete extraction that left only some
cytoplasmic fragments and nuclei. Despite the severe
damage to the endocardial cells, their cell shape was
usually unaffected, and intercellular borders were still
present (Figures 3A-3C). Zones where endocardial
cells were less extracted showed many abnormal intercellular spaces and areas where the underlying basement membrane was exposed, presumably due to
damage and detachment of endocardial cells. In these
areas, the basement membrane also showed evidence
of damage. The pattern of endocardial damage
observed was similar for preparations treated with
eosinophils or with eosinophil supernatants (Table 4).
Discussion
In this study, eosinophils and eosinophil culture
supernatants from hypereosinophilic patients pro-
Circulation Vol 81, No 3, March 1990
1086
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
E
C
~
~
~
,
..4.0
FIGURE 3. Representative scanning electron micrographs of cat papillary muscles treated with activated eosinophils (A and B)
or eosinophil supernatant (C and D). A: Endocardial cells with a severely extracted appearance. Parts of the cell membrane and
cytoplasm were removed but the cell shape was not affected; intracellular borders (arrowheads) were still present. Arrows indicate
extracted nuclei. B: Same muscle as in A. The centrally located endothelial cell has numerous holes in its cell membrane From
the neighboning cells, only the intercellular borders (arrowheads), some cytoplasmic fragments (<), and an extracted nucleus
(arrow) are left. C: Endothelial cells with an extracted appearance, showing a fine meshwork (<), intercellular borders
(arrowheads), and a nearly isolated nucleus (arrow). D: In addition to damaged cell membranes (<), endothelial cells frequently
had intercellular gaps (arrowheads). Scale bars, 5 unm.
duced characteristic acute changes in contractile
behavior in association with severe morphological
damage to endocardium of cat papillary muscle preparations. Eosinophils or neutrophils from normal
volunteers and cell-free culture medium produced no
significant functional or morphological changes. We
have also previously found that aggregating blood
platelets do not damage endocardium under these
conditions. 14
Damage to endocardium by eosinophils was probably mediated by eosinophil secretion products
because eosinophil supernatants had the same effects
as intact cells. The crude eosinophil supernatants
were negative for proteins by sodium dodecyl sulfate
(SDS)-polyacrylamide gel electrophoresis, and it
would be necessary to concentrate them before
attempting identification of the products responsible
for the endocardial damage by a more sensitive assay.
One might speculate that eosinophil cationic protein,
eosinophil peroxidase, or eosinophil-derived neuro-
toxin might be responsible for the damage (rather
than the eosinophil major basic protein, which is less
soluble) as is suggested by previous studies on single
cells.618 It has also been shown that eosinophil
cationic protein may mediate membrane damage by
forming transmembrane pores.'9 The characteristic
morphology of the damaged endocardial cell membranes is consistent with this possibility.
The possible mechanisms of release of these substances and of induction of cell damage were not
explored in this study. Because no stimulating agent
was used, it is likely that there was spontaneous
release of eosinophil products perhaps due to a high
state of activation of eosinophils: all eosinophil suspensions obtained from hypereosinophilic patients
contained high proportions of the hypodense eosinophils, which are believed to be activated.6 '5 The
possibility of cell damage mediated by antibodies
cytophilically bound to the eosinophil cell surface (as
shown for IgE20) should also be considered. However,
Shah et al Activated Eosinophils Injure Mammalian Endocardium
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
the damaging effects of eosinophil supernatants would
then have to be explained on the basis of spontaneous
antibody release into the supernatant. IgE molecules
can be experimentally eluted from the eosinophil FCE
receptor by acid pH treatment but not spontaneously.21
The changes in contractile behavior produced by
activated eosinophils and their supernatants are
identical to those resulting from selective damage to
endocardium of isolated cat papillary muscle by
gentle mechanical abrasion or by transient exposure
to the detergent Triton X-100,12 suggesting that these
eosinophil soluble mediators similarly induce selective damage to the endocardium without damaging
myocardium. Our data showing no significant reduction in +dT/dt or in Vmax also provide functional
evidence of myocardial integrity. Vma, is a measure of
the force-velocity relation22 of the muscle and is
believed to represent the life cycle of a single crossbridge between actin and myosin: it would be
expected to decrease if myocardium were damaged.
Similarly, the rate of tension development usually
decreases significantly when myocardium is damaged. Furthermore, addition of eosinophils to preparations where endocardium had already been selectively damaged produced no further acute change in
contractile performance. However, because the myocardium was not systematically examined by transmission electron microscopy of all specimens, no
further conclusions can be made regarding morphological integrity of the myocardium.
It is possible that endocardium is selectively damaged by activated eosinophils because it is more
susceptible to damage than the myocardium. Alternatively, the relatively short time of exposure (30
minutes) in this study might have restricted damage
to superficial areas of the papillary muscles. The
clinical pattern of eosinophil-associated cardiac disease in humans is also one of predominantly endocardial damage supporting the hypothesis that endocardium is more prone to damage than myocardium. An
interesting and relevant question is whether acute
endocardial damage in vivo produces acute changes
in contractile behavior of subjacent myocardium similar to the changes observed in this study and whether
these might cause clinically significant hemodynamic
disturbance. Although the magnitude of such an
effect would be small, significant hemodynamic
changes could occur as a result of increased nonuniformity of contractile behavior.23 However, the phenomenon of endocardial modulation of myocardial
contractile behavior in vivo is still under investigation. Furthermore, the mechanisms responsible for
the effect of endocardium in isolated cardiac muscle
are also speculative at present.24,25
The results of this study show that the endocardium of isolated cardiac papillary muscle is specifically damaged by mediators from activated eosinophils and results in a characteristic change in the
pattern of contraction and relaxation. The cat papillary muscle preparation may be a useful simple
and sensitive model to further investigate the
1087
mechanical effects and mechanisms of eosinophilinduced cardiac damage.
Acknowledgments
We are grateful to Dr. L. Prin and Dr. F. Ajana
for access to the hypereosinophilic patients and to
J.P. Papin for purification of cells and supernatant,
to Prof. Callebaut and Prof. Van Landuyt for
allowing use of the electronmicroscopes, to W.
Delnat for drawing the illustrations, and to M.
Michiels for typing the manuscript. A.M.S. is
indebted to Professor A.H. Henderson and Dr. M.J.
Lewis for making his visit possible.
References
1. Chusid MJ, Dale DC, West BC, Wolff SM: The hypereosinophilic syndrome: Analysis of fifteen cases with review of the
literature. Medicine (Baltimore) 1975;54:1
2. Spry CJ, Davies J, Tai PC, Olsen EG, Oakley CM, Goodwin
JF: Clinical features of fifteen patients with the hypereosinophilic syndrome. Q J Med 1983;52:1
3. Olsen EGJ, Spry CJF: Relation between eosinophilia and
endomyocardial disease. Prog Cardiovasc Dis 1985;27:241
4. Spry CJF: Eosinophils and endocardial fibrosis, in Kawai C,
Abelmann W (eds): Cardiomyopathy Update: I. Pathogenesis of
Myocarditis and Cardiomyopathy. Tokyo, University of Tokyo
Press, 1987, p 293
5. Parrillo JE, Fauci AS: Human eosinophils: Purification and
cytotoxic capability of eosinophils from patients with the
hypereosinophilic syndrome. Blood 1978;51:457
6. Gleich GJ, Adolphson CR: The eosinophil leukocyte: Structure and function. Adv Immunol 1986;39:177
7. Tai P-C, Spry CJF, Olsen EGJ, Ackerman SJ, Dunnette S,
Gleich GJ: Deposits of eosinophil granule proteins in cardiac
tissues of patients with eosinophilic endomyocardial disease.
Lancet 1987;1:643
8. Molina HA, Kierszenbaum F: Immunohistochemical detection of deposits of eosinophil-derived neurotoxin and eosinophil peroxidase in the myocardium of patients with Chagas'
disease. Immunology 1988;64:725
9. Tai P-C, Hayes DJ, Clark JB, Spry CJF: Toxic effects of human
eosinophil secretion products in isolated rat heart cells in
vitro. Biochem J 1982;204:75
10. Brutsaert DL, Meulemans AL, Sipido KR, Sys SU: The
endocardium modulates the performance of myocardium.
Arch Int Physiol Biochem 1987;95:4
11. Brutsaert DL, Meulemans AL, Sipido KR, Sys SU: Mechanical performance of the myocardium is modulated by the
endocardium. Biophys J 1987;51:464
12. Brutsaert DL, Meulemans AL, Sipido KR, Sys SU: Effects of
damaging the endocardial surface on the mechanical performance of isolated cardiac muscle. Circ Res 1988;62:358
13. Meulemans AL, Sipido KR, Sys SU, Brutsaert DL: Atriopeptin III induces early relaxation of isolated mammalian papillary muscle. Circ Res 1988;62:1171
14. Shah AM, Meulemans AL, Brutsaert DL: Myocardial inotropic responses to aggregating platelets and modulation by the
endocardium. Circulation 1989;79:1315
15. Prin L, Capron M, Tonnel A-B, Bletry 0, Capron A: Heterogeneity of human peripheral blood eosinophils: Variability in
cell density and cytotoxic ability in relation to the level and the
origin of hypereosinophilia. Int Arch Allergy Appl Immunol
1983;72:336
16. Prin L, Charon J, Capron M, Gosset P, Taelman H, Tonnel
AB, Capron A: Heterogeneity of human eosinophils: II.
Variability of respiratory burst activity related to cell density.
Clin Exp Immunol 1984;57:735
17. Gruart V, Balloul JM, Prin L, Capron M, Tomassini M,
Loiseau S, Capron A: Variations in protein expression related
to human eosinophil heterogeneity. J Immunol 1989;142:4416
1088
Circulation Vol 81, No 3, March 1990
18. Molina HA, Kierszenbaum F: Interaction of human eosinophils or neutrophils with Trypanosoma cruzi in vitro causes
bystander cardiac cell damage. Immunol 1989;66:289
19. Young JD, Peterson CGB, Venge P, Cohn ZA: Mechanism of
membrane damage mediated by human eosinophil cationic
protein. Nature 1986;321:613
20. Capron M, Kusnierz JP, Prin L, Spiegelberg HL, Ovlaque G,
Gosset P, Tonnel AB, Capron A: Cytophilic IgE on human
blood and tissue eosinophils: detection by flow microfluorometry and relation to serum IgE. J Immunol 1985;134:3013
21. Capron M, Capron A: The IgE receptor of human eosinophils.
Allergy Inflamm 1987;11:151
22. Brutsaert DL, Claes VA, Sonnenblick EH: Velocity of shortening of unloaded heart muscle and the length-tension relation. Circ Res 1971;29:63
23. Brutsaert DL: Nonuniformity: A physiologic modulator of
contraction and relaxation of the normal heart. J Am Coll
Cardiol 1987;9:341
24. Brutsaert DL, Meulemans AL: Transendothelial ionic
exchange underlies endocardial control of myocardial performance. Biophys J 1988;53:59
25. Brutsaert DL: The endocardium. Annu Rev Physiol 1989;
51:263
KEY WORDS * eosinophils * endocardial endothelium
endomyocardial fibrosis * hypereosinophilia
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Eosinophils from hypereosinophilic patients damage endocardium of isolated feline heart
muscle preparations.
A M Shah, D L Brutsaert, A L Meulemans, L J Andries and M Capron
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 1990;81:1081-1088
doi: 10.1161/01.CIR.81.3.1081
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1990 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/81/3/1081
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation 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 is online at:
http://circ.ahajournals.org//subscriptions/