Department of Cardiology, University Hospital Benjamin Franklin, Free University of Berlin, D-12200 Berlin, Germany
Correspondence: Dr. Christian Seligmann, Medizinische Klinik II mit Poliklinik, Friedrich-Alexander-Universität Erlangen-Nürnberg, Östliche Stadtmauerstr. 29, D-91054 Erlangen, Germany. E-mail: cseligmann{at}talknet.de
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: contractile function • leukocytes • oxygen free radicals • guinea pig
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Most studies suggesting major involvement of PMNs in cardiac reperfusion injury have focused on hearts exposed to a no-flow ischemia [14 , 15 ]. However, this interaction is still not well understood, including whether PMNs also have deleterious effects on myocardium after the relatively mild stimulus of a low-flow ischemia. This is an important question, because acute coronary syndromes often do not occur as a result of complete occlusion of coronary vessels but in the course of subtotally stenosed coronary arteries. In a clinical setting, myocardial stunning in patients presenting with angina pectoris has already been demonstrated [16 ]. The few existing experimental studies using a low-flow ischemia in this context applied an artificial "cocktail" of substances which are known for their potential to stimulate PMNs but are not present in an in vivo setting [17 , 18 ]. The cell activator thrombin, however, is known for its potential to stimulate PMNs on one hand [19 ] and is released in significant concentrations in reperfusion on the other hand [20 ].
The aim of the study reported here was, therefore, to investigate whether thrombin-stimulated PMNs induce myocardial dysfunction also under conditions of low-flow ischemia. Moreover, we investigated whether PMNs induce myocardial dysfunction when administered at different times during ischemia and reperfusion; i.e., low-flow ischemia and late reperfusion.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The constituents for the Krebs-Henseleit buffer, phosphate-buffered saline (PBS), and Tyrode‘s solution were obtained from Merck (Darmstadt, Germany). Superoxide dismutase (SOD) and thrombin were obtained from Sigma (St. Louis, MO). Poly-hydroxyl-ethyl-starch was obtained from Fresenius (Bad Homburg, Germany).
Isolation of PMNs
Experiments were performed in conformance with National
Institutes of Health guidelines [21
].
To obtain homologous PMNs, male guinea pigs weighing 600–1,000 g each were anesthetized by intramuscular administration of xylazine hydrochloride (125 µL of 2% solution/500 g of body weight) and ketamine hydrochloride (1.5 mL of 5% solution/500 g of body weight), after which samples of whole blood were obtained by cannulation of the right carotid arteries. About 30 mL of blood were collected in three 10-mL polypropylene syringes each containing 200 µL of ethylenediaminetetraacetic acid solution (end concentration, 0.1%) for anticoagulation.
The blood was spun at 350 g (15 min), plasma was rejected,
and the remaining hematocrit was mixed with poly-hydroxyl-ethyl-starch.
This solution was sedimented at 70 g (10 min). The
nonsedimented part of this solution was spun at 350 g (10
min), and the pellet formed was washed with PBS. After that, density
centrifugation was performed (350 g for 25 min) by using
Percoll® solution (density, 1.082 g/mL). The remaining cells were
rinsed with distilled water at 4°C in order to eliminate
erythrocytes. The suspension of cells in distilled water was mixed 1:1
with NaCl solution (1.8%) to restore isotonicity of the suspension.
The obtained PMNs were washed twice with PBS, resuspended in 1 mL of
Tyrode’s solution, counted in triplicate 50-µL aliquots (with a
Coulter Counter®), and adjusted to a value of 
10,000 cells/µL by
adding further Tyrode‘s solution. Purity (> 95%) and viability
(>95%) of the cell preparation were routinely controlled by light
microscopy (using a Pappenheim stain and the trypan blue exclusion
test). Cells obtained by this method have been found to respond
normally in cell adhesion tests.
Heart preparation
Hearts of male guinea pigs (weight, 200–300 g), different from
the animals used for blood sampling, were isolated as described before
[22
]. In short, the animals were stunned by neck
dislocation, incision of the right carotid artery was performed, and
the thorax was opened. Hearts were arrested by superfusion with cold
(4°C) isotonic saline, and the aortas were cannulated. Perfusion was
started 30 s after cardiac arrest with a modified Krebs-Henseleit
buffer at 37.5°C, pH 7.4, at constant pressure (60–80 mm Hg) via the
aorta, and hearts started beating spontaneously with an average
frequency of 230 ± 10 beats/min. Further on, the veins entering
the left and right atrium were ligated to ensure that the effluent from
the coronary sinus emerged via the pulmonary artery for cell sampling
and to prevent perfusion buffer leakage via the left atrium in working
heart mode. The perfusion pressure was continuously monitored with a
pressure transducer. Then the left atrium was incised, and an
additional cannula was tightly inserted. When buffer flow was initiated
through this cannula, the left atrium and ventricle could be filled in
diastole, enabling hearts to perform pressure-volume work against a
preset aortic pressure. Pre- and afterloads were adjusted to and
maintained constant at 14 mm Hg and 64 mm Hg, respectively.
Experimental protocol
The experimental protocols employed using working heart
preparations are outlined in Figure 1
. All hearts were allowed to stabilize at constant pre- and
afterload for 20 min (preischemic work phase). After a 20-min working
phase, they were then subjected to a 30-min period of low-flow ischemia
(1 mL/min, 37°C), followed by 5 min of reperfusion at constant
coronary flow (5 mL/min) and 10 min at constant pressure (60–80 mm
Hg), conducted in a nonworking mode. Thereafter, pressure-volume work
was resumed, and external heart work (EHW) was determined 20 min later.
 View larger version (15K): [in a new window] |
Figure 1. Experimental protocols. A 1-min bolus of polymorphonuclear granulocytes
was applied during either ischemia (protocol a), early reperfusion
(protocol b), or late reperfusion (protocol c). Thrombin (0.3 U/mL
perfusion buffer) was given throughout the whole period of ischemia and
reperfusion. In further series of experiments, superoxide dismutase
(SOD) was added in a concentration of 3 U/mL of perfusion buffer. In
control experiments, neither cells nor SOD was applied.
|
At the times indicated in Figure 1
, homologous PMNs were introduced
into the coronary system of hearts via the aortic feed line as
standardized boluses (1,000 PMNs/µL of perfusion buffer; 60-s
duration) in the presence of thrombin (0.3 U/mL of perfusion buffer) by
using an infusion pump (Infors, Basel, Switzerland). Accordingly,
during low-flow ischemia a total of 1 x 106 PMNs
and in reperfusion a total of 5 x 106 PMNs were
infused into the hearts. These relatively low numbers of neutrophils,
as compared with the physiologic concentration, which is about fivefold
as high, were chosen in order to minimize the risk of capillary
plugging and thereby induce loss of myocardial function. Moreover,
coronary perfusion pressure was quantified to exclude occurrence of
significant capillary plugging. In recent studies comparing the effects
of microspheres (5 and 10 µm in diameter) and PMNs on coronary
perfusion pressure of isolated and constant-volume (5-mL/min) perfused
guinea pig hearts exposed to a 15-min ischemia, we could demonstrate
that coronary perfusion pressure is a very sensitive parameter in order
to exclude significant capillary plugging [23
]. The cell
activator thrombin was administered throughout the whole ischemia and
reperfusion phase in a concentration of 0.3 U/mL of perfusion buffer.
Coronary effluent from the pulmonary artery was collected for 3 min,
starting from the beginning of PMN application. Only insignificant
numbers of cells emerged at times, later than this. The
coronary-effluent sample was weighted (
volume), and the PMN
concentration was quantified with a Coulter Counter® (triplicate
determination in 50-µL aliquots). The product of volume and
neutrophil count yielded "PMN output." Immediately before cell
application, the cell concentration in the stock cell solution was
determined (PMN input). The percentage of PMNs remaining in the
coronary bed was then calculated as 100 - ([PMN output/PMN
input] x 100) = PMN retention (%).
In additional sets of experiments, the radical scavenger SOD was added in a concentration of 3 U/mL of perfusion buffer from 10 min before until 10 min after PMN application.
Hearts with application of thrombin (0.3 U/mL perfusion buffer) but without application of PMNs served as controls in both sets of experiments.
Statistical methods
Each group consisted of five hearts unless otherwise stated. The
results are given as mean values ± standard errors of the means.
Statistical analysis was performed with one-way analysis of variance.
Whenever a significant effect was obtained with one-way analysis of
variance, multiple-comparison tests between the groups were performed
by using Student-Newman-Keul‘s test. Differences between groups were
considered significant at P < 0.05.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (20K): [in a new window] |
Figure 2. Recovery of external heart work after sole polymorphonuclear
granulocyte (PMN) or thrombin application at different times during
ischemia and reperfusion. In controls, hearts were exposed to ischemia
and reperfusion without PMN or thrombin application. Columns are means;
error bars represent standard errors of the means; n =
5 each; P < 0.05.
|
 View larger version (22K): [in a new window] |
Figure 3. Recovery of external heart work. Comparison of sole polymorphonuclear
granulocyte (PMN) application with combined PMN and superoxide
dismutase (SOD) application at different times during ischemia and
reperfusion. In controls, hearts were exposed to ischemia and
reperfusion in the presence of thrombin but without application of
PMNs. Columns are means; error bars represent standard errors of the
means; n = 5 each. * means significant differences
between time-matched experiments with or without coapplication of SOD;
P < 0.05.
|
 View larger version (16K): [in a new window] |
Figure 4. Intracoronary polymorphonuclear granulocyte (PMN) retention (percentage
of PMN applied) at different times of ischemia and reperfusion in the
absence or presence of superoxide dismutase (SOD). In controls, hearts
were perfused with normal flow without being exposed to ischemia and
reperfusion but with coapplication of PMNs and thrombin. Columns are
means; error bars represent standard errors of the means;
n = 5 each. * means significant differences between
control and the other experiments, respectively; P <
0.05.
|
We also examined whether SOD itself has any effect on heart function by administering increasing concentrations (1–6 U/mL of perfusion buffer) of SOD in working hearts not exposed to ischemia or reperfusion and without PMN application (data not shown). This sole administration of SOD did not have a significant effect on EHW in any concentrations used. Impairment of heart function was not a result of capillary plugging, which would have led to a significant increase of perfusion pressure as a consequence. Such an increase was not observed in experiments with application of PMNs.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to no-flow ischemia, PMNs continue to pass coronary vessels during a low-flow ischemia. Low shear forces during low-flow ischemia could thereby lead to an increased number of PMNs retained in the heart. This hypothesis is confirmed by our results, showing highest PMN retention during low-flow ischemia.
Unfortunately, in the few existing studies with hearts exposed to a low-flow ischemia in the presence of PMNs, an artificial "cocktail" consisting of N-formyl-L-methionyl-L-leucyl-L-phenylalanine, H2O2, and thrombin was used to induce at least some degree of myocardial dysfunction, under these conditions [17 , 18 ]. It is interesting, in hearts used in our experiments and exposed to a low-flow ischemia, that PMNs significantly impaired myocardial function without application of such a cocktail. After sole application of thrombin throughout the whole ischemia-reperfusion phase in a concentration of 0.3 U/mL of perfusion buffer, a significant cardiodepressive effect of PMNs was observed. In contrast to these other studies, thrombin was present not only during ischemia but also in reperfusion in our experiments. Because we know from clinical studies that thrombin is released in high concentrations during reperfusion, our experiments resemble conditions found in an in vivo situation of ischemia and reperfusion [12 , 13 ].
To investigate a possible time dependency of PMN-mediated deleterious effects, cells were administered at different times of the ischemia-reperfusion period, i.e., ischemia and early or late reperfusion. A significant impairment of myocardial function was observed at all three times investigated. The degree of myocardial dysfunction did not vary significantly between the three times of PMN application. This is an interesting observation because PMN-induced injury of the heart mediated by free radicals is believed to take place very early in reperfusion.
Moreover, we demonstrated that application of PMNs during ischemia could induce a reduction of myocardial function too, possibly because either PMNs induce a myocardial injury already during ischemia or injury occurs in reperfusion via PMNs retained in coronary circulation during ischemia. The finding that additional administration of SOD significantly improved REHW led to the conclusion that reactive oxygen species play an important role in the induction of a myocardial dysfunction also during phases of ischemia and late reperfusion. Given the fact that myocardial pump failure induced by PMN application during ischemia could be prevented by the presence of SOD only during this ischemia, a radical-mediated effect before reperfusion is probable.
One important point of our study is that PMN concentrations were significantly lower than were those seen in vivo. Coronary perfusion pressure, which continuously was monitored throughout the experiments, did not increase significantly, thereby excluding a myocardial dysfunction caused by capillary plugging.
In conclusion, our results suggest an important deleterious role of PMNs also under conditions of a low-flow ischemia and reperfusion of isolated, working guinea pig hearts. These deleterious effects are not only restricted to early reperfusion but can also be demonstrated during ischemia and in later phases of reperfusion. Cardiac compromise during ischemia and in late reperfusion is mediated by increased intracoronary PMN retention and emission of free oxygen species, like in early reperfusion. Therefore, there seems to be a basis for the application of anti-PMN strategies also in acute coronary syndromes.
Received September 26, 2000; revised January 4, 2001; accepted January 5, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
