ELSEVIER
CardiovascularResearch32 (1996) 294-305
Influence of the terminal complement-complex on reperfusion injury,
no-reflow and arrhythmias: a comparison between C6-competent and
Cb-deficient rabbits
w. It0 a,* , H.J. Sch’tifer b, S. Bhakdi d, R. Klask a, S. Hansen a, S. Schaarschmidt b,
J. Schafer a, F. Hugo ‘, T. Hamdoch a, D. Mathey a
a Department of Cardiology, University Hospital Eppendorj Hamburg, Germany
b Department of Pathology Uniuersie Hospital Eppendoti Hamburg, Germany
’ Institute of Medical Microbiology, Unkersity of Giessen, Giessen, German)
d Institute of Medical Microbiology, Unicersity of Mainz, Mainz, Germany
Received 28 July 1995;accepted29 February 1996
Abstract
Objective: The complement system has been suggested to play a role in reperfusion injury which may result from an enhanced
destruction of myocardial tissue or from an impairment of reflow. We investigated the influence of the C5b-9 complement complex on
infarct size, reflow and arrhythmogenesis. Methods: Twenty-eight C6-competent rabbits and 18 rabbits with congenital C6 deficiency
were subjected to either 30 min or 2 h of coronary artery occlusion followed by reperfusion. C6 deficiency was confirmed by the
complementtitration test and immunohistology. The triphenyl tetrazolium chloride method was used to delineate infarct size. Reflow into
infarcted areas was evaluated histologically after an in vivo injection of propidium iodide which served as an early fluorescence
microscopic marker of damagedmyocardium subjectedto reflow. Continuous ECG monitoring allowed the recording of arrhythmias.
Results: After 30 min of coronary artery occlusion infarct size was significantly smaller in C6-deficient rabbits (5.0 + 2% of the risk
region) as comparedto C6-competentrabbits (28.4 + 8.5%, P = 0.0371). The extent of reflow into damagedmyocardium was nearly the
same in both animal groups at this time (38 k 9 vs. 39 + 7% of the risk region). After 2 h of coronary artery occlusion, infarct size was
not different between both animal groups, but the extent of reflow into damaged myocardium was significantly smaller in C6-competent
rabbits than in C6-deficient rabbits (25 + 4 vs. 40 + 4%: P = 0.0185). Two of the 18 C6-deficient rabbits had ventricular arrhythmias
(Lawn II-IV), none of which was fatal. Eleven of the 28 C6-competent animals had major ventricular arrhythmmias which were fatal in 6
rabbits. Conclusions: These results suggest that the lytic C5b-9 complementcomplex leads to reperfusion injury in the early phase(30
min) of ischaemia, resulting in a larger infarct. After 2 h of ischaemia, complement activation enhancesthe no-reflow phenomenonbut
does not affect infarct size. Finally, the C6 status seems to influence the susceptibility
to ventricular arrhythmias after coronary artery
occlusion, independent of reperfusion.
Keywords: Complement system; Myocardial infarction; Reperfusion;Myocardial infarct size; Arrhythmias; Rabbit, anesthetized
1. Introduction
The role of the complement system in the pathogenesis
of myocardial ischaemic tissue damage has been investigated in several studies [l-7]. We recently observed that
’ Corresponding author. Max-Planck-Institut for Physiological and
Clinical Research, Dept. of Experimental Cardiology, Benekestrasse2,
D-61231 Bad Nauheim, Germany. Tel.: f49 6032 705410; fax:
+49 6032 705 419.
terminal C5b-9 complement accumulates rather late in
myocardial infarction without reperfusion but rather early
if the ischaemic myocardium is reperfused [8]. Therefore
C5b-9 complement is more likely to contribute to reperfusion injury than to ischaemic tissue damage in the non-reperfused heart. This contention is in line with other studies
which suggest that complement activation plays an important role in myocardial reperfusion injury [9--l I]. Smith et
Time for primary
0008-6363/96/$15.00 Copyright 0 1996 Published by Elsevier ScienceB.V. All rights reserved
PU SOOOS-6363(96)00082-X
review 21 days
W. Ito et al. /Cardiovascular
al. were able to demonstrate that complement inhibition
with soluble complement receptor type 1 given before
coronary occlusion reduces myocardial infarct size after
reperfusion in rats [9]. Soluble complement receptor type 1
suppresses formation of both C5a and lytic C5b-9 complement complex [12]. Whereas C5a is a strong chemotactic
and activating factor for neutrophils, C5b-9 complexes
generated in cell membranes are cytotoxic [13,14]. The
question whether C5a or C5b-9 alone or in conjunction
contributed to reperfusion injury remained unanswered in
the study of Smith et al. [9]. Homeister et al., however,
were able to demonstrate in an isolated heart preparation
that complement activation depressed myocardial function
and increased coronary perfusion pressure 1151. These
effects were not mediated by the anaphylatoxins but were
dependent on C8 complement, suggesting that C5b-9 was
responsible for these changes [15]. Neutrophils, the main
target of the anaphylatoxins, however, were not present in
the perfusate and the hearts were not subjected to classical
ischemia and reperfusion [15]. For these reasons the question whether C5b-9 complement alone contributes significantly to reperfusion injury was not conclusively answered
in the study of Homeister et al. In the present study the
influence of C5b-9 on reperfusion injury was therefore
studied here in an in vivo model of myocardial infarction
and reperfusion.
We first intended to determine histologically the precise
localization of C5b-9 in relation to myocardial tissue damage and reflow after shorter and longer periods of coronary
artery occlusion followed by reperfusion. Infarct size of
congenitally C6-deficient rabbits which were not capable
of forming C5b-9 was then compared with that of normal
C6-competent rabbits.
C5b-9 complement may influence infarct size after
reperfusion either directly by increased destruction of myocardial tissue or indirectly by impairing reflow. The
increase in coronary perfusion pressure by complement
activation observed by Homeister et al. indirectly suggested that the impairment of reflow might be an important
mechanism [15]. We therefore also studied the areas of
reflow. The same group was able to demonstrate that
myocardial tissue accumulated sodium and calcium and
lost potassium as a result of complement activation [15].
This was not surprising since lytic complement affects cell
membrane permeability. An attempt was therefore made to
demonstrate whether C5b-9 complement complex influences arrythmogenesis after myocardial infarction and
reperfusion.
2. Methods
The present study was performed with permission of the
Amt fur Gesundheit und Veterinarmedizin of the Freie und
Hansestadt Hamburg (No. F I 56/90). It conforms with
the Guide ,for the Care and Use of Laboratory Animals
Research 32 (1996) 294-305
295
published by the US National Institutes of Health (NIH
Publication No. 85-23, revised 1985).
2.1. Experimental protocol
Twenty-eight C6-competent and 18 congenitally C6-deficient rabbits of the same race were included in the study.
C6-deficient rabbits were originally a gift of Drs. K. and
U. Rother (Institute for Immunology, University of Heidelberg, Germany). A breeding colony was established at the
Institute of Medical Microbiology and Hygiene, University
of Mainz. The animals used in the experiments were
usually pairs of homozygotes and heterozygotes from the
same litter. C6-deficiency arises from a single gene defect
and is not associated with other abnormalities. C6 competence or deficiency remained unknown during the experiment. The determination of infarct size and area of reflow
were also performed without knowledge of the C6 status.
The latter was determined from blood samples by the
complement titration test described below. The rabbits
were randomly assigned to 30 min or 2 h of coronary
artery occlusion followed by 1 or 3 h of reperfusion.
The animals were sedated by an intramuscular injection
of ketamine HCl (4-8 mg per kilogram bodyweight),
xylazine (8-9 mg per kilogram body weight) and received
atropine (0.1 mg per kilogram bodyweight). Anaesthesia
was induced by a continuous intravenous infusion of
pentobarbital (12 mg per kilogram bodyweight per hour).
Electrodes were placed on the extremities to allow continuous ECG monitoring throughout the procedure. All ventricular ectopies were recorded and classified according to
the grading system of Lown [ 161. The animals then underwent tracheotomy and were artificially ventilated with a
constant tidal volume of 25 ml, a respiratory rate of
50/min and an oxygen flow of 1 l/min ensuring maximal
oxygenation in all animals studied. A median thoracotomy
was performed and the heart was suspended in a pericardial cradle before a large marginal branch of the circumflex coronary artery was ligated. This coronary artery is a
large reproducible vessel in rabbits, which have no left
anterior descending coronary artery. The ligature (4.0 prolene) was felt-armed to prevent artery damage and was
maintained using a tourniquet and a clamp.
Reperfusion was allowed to occur by releasing the
clamp and retrieving the tourniquet. Fifteen minutes before
the end of the reperfusion period propidium iodide (2 mg
in 2 ml NaCl 0.9%; Sigma Chemicals GmbH) was injected
into the left atrium to permit detection of the area of
reflow. The dye entered the damaged cells and led to
intense fluorescent staining of the nuclei. At the end of the
reperfusion period the ligature was reinstituted and 3 ml of
a 1% solution of patent blue violet (Sigma Chemicals
GmbH) were slowly injected into the left atrium to delineate the ‘area at risk’ [ 171. An overdose of potassium
chloride was used to kill the animal.
W. Ito et al. /Cardiovascular
296
Research 32 (19961 294-305
Fig. 1. Impregnation of cardiac vessel walls with CSb-9 (small arrows) in a C6-competent rabbit. X 370.
2.2. Analysis of infarct size
Analysis of infarct size was performed by methods
previously described in the literature [17-201 and without
knowledge of the C6 status [8]. The excised heart of each
rabbit was rinsed in 0.9% saline solution, weighed and cut
in 2 mm slices parallel to the atrioventricular groove. The
slices were photographed with a Kodak Ektachrome film
to define the ‘area at risk’, which was the area not stained
by patent blue violet. The slices were then immersed in 1%
triphenyl tetrazolium chloride (TTC) in 0.2 ml Tris buffer
at pH 7.8, incubated for 10 min at 37°C and rephotographed [18]. Viable myocardium was stained bright red
while the region of the infarcted myocardium was demar-
Fig. 2. Intense cytoplasmic positivity of infarcted myocardial cells (small arrows) in comparison with negative immunocytochemical staining of C5b-9 in
normal myocardium (large arrows). C&competent rabbit after 2 h of coronary artery occlusion followed by 3 h of reperfusion. x 150.
W. Ito et al. / Cardiocascular
Research
32 (19961 294-305
291
Fig. 3. Lack of C5b-9 deposition in vessel walls (small arrows) of a C6-deficient rabbit. X 370.
cated by the absence of TTC staining. An enlarged tracing
(X 10) was then made from each slide and the ‘area at
risk’ and the area of infarction determined by planimetry.
2.3. Area of refzow
The area of reflow into damaged myocardium was
determined following a protocol recently described by
Schafer et al. [21,22]. This protocol was developed on the
basis of a suggestion of Bhakdi who previously had used
propidium iodide to study cell membrane permeability
[23]. Propidium iodide is only able to penetrate damaged
cell membranes. After intercalation with DNA it leads to a
bright red fluorescence [23]. Its presence indicates that
reperfusion has reached the cellular level. Other fluorescent agents have previously been employed to determine
Fig. 4. Negativity of C5b-9 staining in myocardial infarction in C6-deficient rabbits after 2 h of coronary artery occlusion followed by 3 h of reperfusion.
x 150.
298
W. Ito et al./ Cardiocascular Research 32 (19961 294-305
the distribution of arterial blood flow [24]. From the
myocardial slices, biopsies from the area at risk as delineated by patent blue violet staining were taken and quickly
frozen in n-hexane cooled to -6°C by a mixture of dry
ice in 100% ethanol and were kept at - 196°C. Cross-sections (10 km thick) were obtained and evaluated under the
microscope using fluorescent light and a filter (BP 455500, FT 510, LP 528). The reflow into damaged myocardium (i.e., the area of fluorescent staining) was related
to the whole cross-section and expressed in intervals of
10%. This analysis was performed by an independent
observer who was not aware of the origin of the biopsy,
the duration of occlusion and the C6 status of the animal.
The reflow into damaged myocardium expressed as a
percentage of the area at risk was calculated from the
results of the two biopsies.
Six biopsies were analysed by three independent observers. Their evaluations did not differ. Evaluations of the
same biopsy by the same observer also did not differ when
obtained several months apart.
2.4. Immunohistological
studies
Adjacent cross-sections from the same material were
fixed for 5 min in cooled acetone (4”(Z), rinsed in phosphate-buffered saline (PBS) and incubated for 2 h in a 1:20
dilution of a specific polyclonal sheep antibody against
rabbit C5b-9 developed by Hugo and Bhakdi and used in a
previous study [8]. After rinsing in PBS the sections were
treated with the second antibody (biotinylated rabbit antisheep IgG) by employing a Kit (Vectastain (RI) using a
biotin avidin (ABC) bridging technique and peroxidase as
a marker enzyme. Peroxidase activity was stained by 3,3’diaminobenzidine. All processing steps were done according to the instructions of the Vectastain (R) kit.
2.5. Histological comparison of reperfused damaged myocardium and C5b-9 accumulation
CS-9 accumulation in comparison with the reperfuseddamaged myocardium was assessed in adjacent cross-sections. In some cases plain histological sections were photographed under fluorescent light and subsequently subjected
to immunohistological staining. This allowed direct comparison of C5b-9 accumulation and reperfused damaged
myocardium in the same cross-section.
2.6. Complement titration
Serum was collected from each animal and total
hemolytic activity tested by incubating 1 volume of serum
with 1 volume of a 2% suspension of sheep erythrocytes
coated with polyclonal antibodies (Ambozeptor 6000;
Behring, Marburg, Germany) for 30 min at 37°C. C6-deficient sera were totally devoid of hemolytic activity [25].
Fig. 5. Extent of propidium iodide fluorescence within the area at risk
from the subendocardial (small arrows) to the epicardial layer (large
arrows) after 30 min of coronary artery occlusion followed by reperfusion. (a) In a C6-competent rabbit. Note the non-fluorescent zone in the
mid-myocardium. Collage, X6. (b) In a C6-deficient rabbit. Note the
pervading fluorescence. Collage, X 5.
2.7. Statistical analysis
All data are presented as mean + SEM. Intergroup comparison of infarct size and reflow was performed by exact
W. Ito et al. / Cardiocascular
permutation. Probability values of 0.05 or less were required for assumption of statistical significance.
3. Results
Forty-six rabbits were included in the experiment: 28
were C6-competent, 18 were C6-deficient. Eight of the 28
C6-competent animals died before completion of the experimental protocol, 6 from arrhythmias and 2 of other
causes.
Of the remaining 20 C6-competent animals 8 were
submitted to 30 min of coronary artery occlusion followed
by 1 h of reperfusion in 3 animals and 3 h of reperfusion
in 5 animals. Twelve C6-competentrabbits were submitted
to 2 h of coronary artery occlusion followed by 1 h of
reperfusion in 6 animals and 3 h of reperfusion in the 6
remaining animals. By definition, the complement titration
test was always positive in C6-competent rabbits. C5b-9
depositions as revealed by immunohistological staining
were always clearly visible in infarcted tissue and in vessel
walls in these animals (Figs. 1 and 2).
Nine of the 18 C6-deficient rabbits were subjectedto 30
min of coronary artery occlusion followed by 1 h of
reperfusion in 4 animals and 3 h of reperfusion in 5
animals. The remaining 9 C6-deficient rabbits underwent 2
h of coronary artery occlusion followed by 1 h of reperfusion in 4 animals and 3 h of reperfusion in 5 animals.
C6-deficient rabbits had a negative complement titration
test and were devoid of immunohistochemically detectable
C5b-9 accumulation (Figs. 3 and 4).
3.1. Extent and localization of the repe&sed damaged
myocardium within the area at risk as revealed histologically by propidium iodide staining (Fig. 5a,b and Fig. 6)
After 30 min of coronary artery occlusion followed by
reperfusion the reperfused areas of the damaged my-
Research
32 (1996)
294-305
299
ocardium as revealed by propidium iodide staining extended from the subendocardiumto the mid-myocardium
and often extended to the epicardium. In most casesthere
was a small non-fluorescent zone in the mid-myocardium.
This histological appearancerevealed only slight differences between C6-competent and C6-deficient animals
(Fig. 5a,b).
After 2 h of coronary artery occlusion followed by
reperfusion propidium iodide fluorescence was visible
subendocardially and epicardially in almost all animals
studied. There was a large non-fluorescent area in the
mid-myocardium which was much more pronounced in
C6-competentanimals (Fig. 6).
3.2. Localization of C’s,., complement accumulation in
relation to the reperfused damaged myocardium (Fig.
7a,b,c)
After 30 min of coronary artery occlusion followed by
reperfusion C5b-9 complement accumulated mainly at the
border with the non-fluorescent area in the infarcted midmyocardium as described above. After 2 h of coronary
artery occlusion followed by reperfusion C5b-9 staining
was visible throughout the infarcted myocardium.
3.3. Granulocytic infiltration of infarcted tissue in C6competent and C6-deJicient animals
Infarcted areas presented with a beginning slight and
inhomogeneous granulocytic infiltration without differences between the C6-deficient and C6-competent groups
(data not shown).
3.4. C6 status and infarct size (Fig. 8)
Only animals that were subjectedto 3 h of reperfusion
were considered in the final analysis of infarct size to
Fig. 6. Extent of propidium iodide fluorescence within the area at risk from the subendocardial (small arrows) to the epicardial layer (large arrows) after 2
h coronary artery occlusion followed by reperfusion in a C6-competent rabbit. Note the large non-fluorescent zone in the mid-myocardium. Collage, X 6.
300
W. Ito et al. / Cardiocuscular
allow sufficient time for complete infarct size delineation
by TTC staining [27].
After 30 min of coronary artery occlusion followed by 3
h of reperfusion infarct size (TIC negative area) expressed
as a percentage of the ‘area at risk’ (patent blue-violet
negative area) was significantly smaller in C6-deficient
Research
32 (19961 294-305
rabbits (5.0 k 2%) than in C6-competent rabbits (28.4 IfI
8.5, P = 0.0371).
After 2 h of coronary artery occlusion followed by 3 h
of reperfusion infarct size did not differ between C6-deficient and C6-competent animals.
As expected, infarct size increased with the time of
Fig. 7. (a) Middle section of Fig. 5 at higher magnification. Collage, X 25. (b) Same cross-section as Fig. 7a stained immunohistologically for C5b-9. Note
the po!sitivity of C5b-9 staining at the border with the non-fluorescent area. Collage, X 25. (c) Area from same slide (large arrow) at a higher magnific :ation
(Magn itication X 370).
301
W. Ito et al. / Cardiocascular Research 32 (19Y612Y4-305
infarct size
(“A of ” area at risk”)
I=“-0371
r p=NSi
-
60
NS
N=5
-r
I
1
.
2h occlusion
30 min occlusion
Fig. 8. Infarct size in C6-competent rabbits (solid bars) and C6-deficient rabbits (open bars) after 30 min and after 2 h of coronary artery occlusion
followed by 3 h of reperfusion. Note the significantly smaller infarct size in C6-deficient animals after 30 min of coronary artery occlusion.
coronary artery occlusion, suggesting that after 30 min of
coronary artery occlusion the infarct was not complete.
3.5. 05 status and area of reflow (Fig. 9)
For the evaluation of the area of reflow, differences in
the duration of reperfusion (l-3 h) were ignored because
they have no influence on the area of reflow and C5b-9
complement accumulation [8,21,22].
Three animals had to be excluded because the amount
of material obtained was not sufficient for evaluation.
After 30 min of coronary artery occlusion followed by
reperfusion the area of reflow (positive propidium iodide
area of reflow
(% of “area at risk”)
T
-
P=NS
‘-1
fluorescence) expressed as a percentage of the ‘area at
risk’ was nearly the same in both animal groups (38 -t 9%
in C&deficient rabbits vs. 39 + 7% in C6-competent rabbits; P = 0.5722, n.s.>.
After 2 h of coronary artery occlusion followed by
either 1 or 3 h of reperfusion the area of reflow was
significantly smaller in C6-competent rabbits (25 & 4) than
in C6-deficient animals (40 + 4; P = 0.0185).
Thus, in contrast to infarct size, a statistically significant difference in reflow between C6-deficient and C6competent animals could only be detected after 2 h of
coronary artery occlusion followed by reperfusion.
r
p=O.O185
1
T
N=9
N=8
I
30
min occludon
Fig. 9. Area of reflow in C6-competent rabbits (solid bars) and C6-deficient rabbits (open bars) after 30 min and after 2 h of coronary artery occlusion
followed by 3 h of reperfusion. Note the significantly larger area of reflow in C6-deficient animals after 2 h of coronary artery occlusion.
W. Ito et al./ Cardiocascular Research 32 (1994) 294-305
302
3.6. Correlation
lOa,b)
In contrast to C&competent animals whose reflow declined with the time of coronary artery occlusion, in
C6-deficient animals reflow into damaged myocardium
was nearly the same after 30 min and 2 h of coronary
artery occlusion followed by reperfusion in C6-deficient
animals.
6o
between reflow and infarct size (Fig.
There was no correlation between reflow and infarct
size either in C6-competent or in C6-deficient animals
after shorter and longer coronary artery occlusions.
ia)
4
50 -::
?:
x 40 -I
%
z
a?30-.E
4IA
1 zo-5
4
4
0
10 --
4
0
4
P
O-
I
I-?+0
14-20
0
I
L
2\-30
3Y40
Reflow
80
in % ot area
I
I
41-150
t
5 1.6'0
61!70
71-8
at risk
lb)
0
70
4
4
Y 6o
*
-C
B
m 50
0
0
6
z
s
40
c
4
A
z 30
;
2
+
20
4
0
41-50
I
51-60
0
4
10
0 -ll-10
11.20
21.30
Reflow
in % ot area
31-40
at risk
Fig. 10. Absence of any correlation between reflow and infarct size after coronary artery occlusion followed by reperfusion. (open symbols: C6-deficient
animals; closed symbols: G-competent animals). (a) 30 min of coronary artery occlusion. (b) 2 h of coronary artery occlusion.
W. Ito et al./Cardiovascular
303
Research 32 (1996) 294-305
no arrhythmias
no arrhythmias
Fig. 1 1, Arrhythmia profile in C6-competent rabbits (top figure) and C6-deficient rabbits (bottom figure) during myocardial ischemia and reperfusion. The
incidence of severe arrhythmias was much higher in Cb-competent animals.
3.7. Arrhythmia profile (Fig. 11)
3.8. Timing of arrhythmias (Table 1)
Seven of the 18 C6-deficient animals had arrhythmias.
All of them occurred during coronary artery occlusion.
Five animals suffered from minor arrhythmias (Lawn I-11).
The remaining 2 animals experienced major arrhythmias
(Lawn III-IV); none of these was fatal.
Of the 28 C6-competent rabbits 16 experienced arrhythmias: 13 occurred during coronary artery occlusion, 3 after
reperfusion was initiated; 5 animals had minor arrythmias
and 11 animals had major arrhythmias. In 6 cases these
arrhythmias resulted in the death of the animal.
Light arrythmias occurred shortly after ligation and
sometimes continued throughout coronary artery occlusion.
There was no difference in the incidence between C6-competent and C6-deficient rabbits. Severe arrythmias in C6deficient rabbits also occurred shortly after ligation of the
coronary artery. In C6-competent rabbits, however, severe
arrhythmias more often occurred later after coronary artery
occlusion and then tended to be fatal.
Thus C6-deficient animals suffered from fewer and less
severe arrhythmias than C6-competent rabbits. Most of the
Table I
Timing of arrhythmias
30 min-1 h
occlusion
1 h-2 h
occlusion
4
C
-
D
-
C
-
D
-
C
-
D
-
1
1
2
-
3
-
3
-
-
-
2
-.
3
-
1
-
Timing of arrhythmias
O-2 h
occlusion
Cb-status
C
D
C
D
No. of
(Lawn
No. of
(Lawn
No. of
4
1
1
2
1
-
-
animals experiencing only light arrhythmias
I-II)
animals experiencing severe arrhythmias
I&IV)
animals experiencing lethal arrhythmias
O-30 min
occlusion
reperfusion
Note that more C6-competent rabbits experienced severe and lethal arrhythmias more than 30 min after coronary artery occlusion. C = C6-competent; D
= C6-deficient.
304
W. Ito et al. / Cardiovascular
arrhythmias occurred after coronary artery occlusion. Severe arrhythmias with fatal outcome only occurred in
C6competent rabbits. The onset of these fatal arrythmias
was later after coronary artery occlusion. Only 3 C6-competent animals presented with reperfusion arrhythmias.
4. Discussion
The major aim of the present investigation was to
determine whether the terminal complement sequence (i.e.,
formation of C5b-9) might additionally contribute to reperfusion tissue damage. This question appeared significant in
view of the fact that C5b-9 complement complexes accumulate during myocardial infarction in humans and in
rabbits, and that CSb-9 deposition appears to occur during
the reperfusion phase [8]. We elected to perform experiments using C6-deficient rabbits because these animals
totally lack C6; thus, complement activation leads to normal production of C5a but not to production of C5b-9. In
most experiments, paired homozygote and heterozygote
siblings from the same litter were employed. The general
morbidity and proneness to infection were not higher in
C6-deficient rabbits. Additional macroscopic and histological examinations of other organs did not reveal any differences between Cti-competent and C6-deficient rabbits. Experiments were performed in a blinded fashion and myocardial infarct sizes and reperfusion areas were documented without knowledge of the C6 status.
Our results suggest that cytotoxic C5b-9 complex contributes to reperfusion injury after short coronary artery
occlusions. We were able to exclude that this was merely
an effect of patchy infarction in the C6-deficient group
using propidium iodide as a histological marker of cell
damage and reperfusion. In both groups there were large
areas of propidium iodide fluorescence often interrupted
by a central zone of absent fluorescence which corresponds to a central zone of no-reflow previously described
[24]. C5b-9 complement mainly accumulated at the border
of this zone.
After 2 h of coronary artery occlusion the infarct sizes,
however, did not differ between C6-competent and C6-deficient animals. Other groups have shown that the infarct is
complete in the rabbit heart at this time [26-281. Our
results therefore confirm that C5b-9 complement rather
contributes to reperfusion injury than to ischaemic tissue
damage in the non-reperfused heart. The small infarct sizes
observed in our study are probably due to the shrinkage of
tissue in the TTC solution which others have also observed
in their experiments (Schaper, personal communication).
Since the area at risk is delineated before TTC staining,
this leads to an underestimation of infarct size in percentage terms. This, however, is a systematic error applicable
to all groups studied and still permits comparison of infarct
Research
32 (19961294-305
sizes. Propidium idodide staining at the end of reperfusion
revealed that C5b-9 contributes to the ‘no-reflow’ phenomenon after 2 h of coronary artery occlusion when no
difference in infarct size was observed. Our results also
show that there is no correlation between infarct sizes and
reflow. This is in accordance with the observations of
Darsee and Kloner and Fishbein et al. who have shown
that any area of microvascular injury resulting from reperfusion lies within the area of the infarct and therefore is
not a causative factor in increasing infarct size [29,30].
Propidum iodide staining did not render an explanation for
the smaller infarct sizes of C6-deficient animals after short
coronary artery occlusions. This fluorophore, however,
only stains the reperfused damaged myocardium. After 2 h
of coronary artery occlusion when infarction is complete,
propidium iodide labeling therefore renders a very good
estimate of the area of reflow. After 30 min of coronary
artery occlusion, however, we cannot exclude the possibility that the propidium-iodide-negative areas contain nonperfused as well as viable areas even though we had
already seen epicardial propidium iodide staining. This
may have blunted a possible effect of C5b-9 on the
‘no-reflow’ phenomenon after short coronary artery occlusions
In spite of the rapid accumulation of C5b-9 during
reperfusion this does not lead to a greater incidence of
reperfusion arrhythmias which we only saw in 3 cases. The
ability to form C5b-9 surprisingly seems to influence
arrhythmogenesis after coronary artery occlusion. Severe
arrhythmias were mainly observed after 30 min of coronary artery occlusion in C6-competent rabbits. These
‘late-occurring’ arrhythmias tended to be fatal. At this time
complement accumulates in small amounts at the border
zone of the infarcted tissue, as our group has previously
described 181. Border zones have been implicated to play
an important role in the generation of a re-entry mechanism which is thought to be responsible for arrhythmogenesis after myocardial infarction [31,32]. The ability of
nucleated cells to withstand an attack of lytic complement
might also lead to transient alterations of membrane potentials of myocytes at the border zone [14,33]. Thus, the
higher incidence of arrhythmias observed in C6-competent
rabbits might be due to facilitated automaticity and reentry.
Acknowledgements
We like to thank Prof. Dr. J. Berger, Institute of
Mathematics and Computer Science in Medicine, University Hospital Eppendorf, Hamburg, for his statistical advice. This study was supported by grants from the Deutsche
Forschungsgemeinschaft (SFB 3 11, Mainz) and the Verband der Chemischen Industrie.
W. Ito et al./CardioL,ascular
References
[l] Hill JH, Ward PA. The phlogistic role of C3 leukotactic fragments
in myocardial infarcts of rats. J Exp Med 1971;133:885-900.
[2] Pinckard RN, Olsen MS, Kelly RE, et al. Antibody-independent
activation of human Cl after interaction with heart subcellular
membranes. J Immunol 1973;110:1376-1382.
[3] Pinckard RN, O’Rourke RA, Crawford MH, et al. Complement
localisation and mediation of ischemic injury to baboon myocardium. J Clin Invest 1980;66:1050-1056.
[4] McManus LM, Kolb WP, Crawford MH, O’Rourke RA, Grover FL,
Pinckard RN. Complement localisation in ischemic baboon myocardium. Lab Invest 1983;48:436-447.
[5] Rossen RD, Swain JL, Michael LH, Weakley S. Gianninnini E,
Entman ML. Selective accumulation of the first component of
complement and leukocytes in ischemic canine heart muscle. A
possible initiator of an extra myocardial mechanism of ischemic
injury. Circ Res 1985;57:119-130.
[6] Schafer HJ, Mathey D. Hugo F, Bhakdi S. Deposition of the
terminal C5b-9 complement complex in infarcted areas of human
myocardium. J Immunol 1986; 137: 1945- 1949.
[7] Crawford MH, Grover FL, Kolb WP, et al. Complement and neutrophil activation in the pathogenesis of ischemic myocardial injury.
Circulation 1988;78:1449-1458.
[8] Mathey D. Schafer J. Schafer HJ, et al. Early accumulation of the
terminal complement-complex in the ischaemic myocardium after
reperfusion. Eur Heart .I 1994;15(3):418-423.
[9] Smith EF, Griswold DE, Egan JW, et al. Reduction of myocardial
reperfusion injury with human soluble complement receptor type 1
(BRL 55730) Eur J Pharmacol 1993;236(3):477-481.
[lo] Shandelya SM, Kuppusamy P, Weisfeld ML, Zweier JL. Evaluation
of the role of polymorphonuclear leukocytes of contractile function
in myocardial reperfusion injury. Evidence for plasma-mediated
leukocyte activation. Circulation 1993;87(2):536-546,
[l 11 Dreyer WJ, Michael LH. Nguyen T, et al. Kinetics of C5a release in
cardiac lymph of dogs experiencing coronary artery ischemia-reperfusion injury. Circ Res 1992;71(6):1518-1524.
[12] Ross GD. Complement receptor type 1. Curr Top Microbial Immunol 1992;178:31-44.
[ 131 Miiller-Eberhard HJ. Molecular organization and function of the
complement system. Annu Rev Biochem 1988;57:321-347.
[14] Bhakdi S, Hugo F, Tranum-Jensen J. Functions and relevance of the
terminal complement sequence. Blut 1990;60:309-318.
[15] Homeister WH, Satoh P, Lucchesi BR. Effects of complement
activation in the isolated heart. Role of the terminal complement
components. Circ Res 1992;71:303-319.
[16] Lown B, Wolf M. Grading system for ventricular ectopy. Circulation
1971;44:130-142.
[17] Forman MB, Bingham S, Kopelman HA, et al. Reduction of infarct
size with intracoronary perfluorochemical in a canine preparation of
reperfusion. Circulation 1985;71:1060-1068.
Research
32 (1996) 294-305
305
[18] Fishbein MC, Meerbaum S, Rit J, et al. Early phase acute myocardial infarct size quantification: validitation of the triphenyl tetrazolium chloride tissue enzyme staining technique. Am Heart J
1981;101:593-600.
[19] Lie JT, Pairolero PC, Holley KE, et al. Macroscopic enzyme mapping verification of large, homogeneous experimental myocardial
infarcts of predictable size and location in dogs. J Thor Cardiovasc
Surg 1975;69(41:599-605,
[20] Boor PJ, Reynolds ES. A simple planimetric method for determination of left ventricular mass and necrotic myocardial mass in postmortem hearts. Am J Clin Path01 1977;68:387-392.
[21] Schafer H, Joachim HC, Rodewald W, Schaarschmidt S, Mathey D.
Tierexperimentelle
Untersuchungen zur friihen Markierung von
Myokardinfarkten durch C5b-9 bzw. Propidiumjodid-Einflub
von
IschPmie und Reperfusionsdauer.
Verh Dtsch Ges Path01
1992:76:277.
[22] Schafer H, et al. Submitted for publication.
[23] Bhakdi S. Greulich S, Muhly M, et al. Potent leukocidal action of
Escherichia
co/i hemolysin mediated by permaeabilization of target
cell membranes. J Exp Med 1989;169:737-754.
[24] Kloner RA, Ganote CE, Jennings RB. The ‘no-reflow’ phenomenon
after temporary coronary occlusion in the dog. J Clin Invest
1974(61;54:1496- 1508.
[25] Newell S, German JC, Bell E, Atkinson JP. Haemolytic and antigenie measurements of complement. J Lab Clin Med 1982:100:437.
[26] Connelly C, Vogel WM, Hernandez YM, Apstein CS. Movement of
necrotic wavefront after coronary artery occlusion in rabbit. Am J
Physiol 1982;243:H682-H690.
[27] Shirato C. Miura T, Ooiwa H, Toyofuku T, Wilborn WH, Downey
JM. Tetrazolium artifactually indicates superoxide dismutase-induced salvage in reperfused rabbit heart. J Mol Cell Cardiol
1989:21:1187-1193.
1281 Miura T. Downey JM, Ooiwa H, et al. Progression of myocardial
infarction in a collateral flow deficient species. Jpn Heart J
1989;30(5):695-708.
1291 Darsee JR, Kloner RA. The no-reflow phenomenon: a time-limiting
factor for reperfusion after coronary occlusion‘? Am J Cardiol
1980;61:452-458.
[30] Fishbein MC, Y-Rit J, Lando U, Kanmatsuse K. Mercier JC, Ganz
W. The relationship of vascular injury and myocardial hemorrhage
to necrosis after reperfusion. Circulation 1980;62: 1274- 1279.
[31] Manning AS, Hearse DJ. Repetfusion-induced arrhythmias: mechanisms and prevention. J Mol Cell Cardiol 1984; 16:497-5 18.
[32] Wilensky RL, Tranum-Jensen J, Coronel R, Wilde AAM. Fiolet JW,
Janse MJ. The subendocardial border zone during acute ischemia of
the rabbit heart: an electrophysiologic, metabolic. and morphologic
correlative study. Circulation 1986;74:1 137-1146.
[33] Morgan BP. Complement membrane attack on nucleated cells: resistance, recovery and non-lethal effects, Biochem J 1989:264(l): 1-14.