Vascular Biology |
From Internal Medicine I (C.K., P.H., R.W., D.H., P.B.), Klinikum Grosshadern, the Institute for Surgical Research (H.H.), and the Department of Physiology (B.F.B.), Ludwig-Maximilians-University of Munich, Munich, Germany.
Correspondence to Christian Kupatt, MD, Internal Medicine I, Klinikum Grosshadern, Marchioninistrasse 15, 81377 Munich, Germany. E-mail c.kupatt{at}lrz.uni-muenchen.de
| Abstract |
|---|
|
|
|---|
Key Words: abciximab polymorphonuclear neutrophils platelets myocardial stunning microcirculation fibrinogen
| Introduction |
|---|
|
|
|---|
2b/ß3, CD41/CD61), have
been introduced, eg, a Fab fragment of the chimeric monoclonal antibody
7E3 (c7E3Fab, Abciximab [Centocor]) against human
GPIIb/IIIa.1 2 3 Besides improving coronary flow
reserve, c7E3Fab protects from myocardial stunning in humans, which is
expressed as regional wall shortening of the reperfused
myocardium.4 Factors beyond inhibition of
platelet aggregation might contribute to the improvement in flow
and myocardial function, eg, inhibition of
platelet-endothelium or platelet-leukocyte
interactions. Fibrinogen competition studies with monoclonal antibodies
suggest that MAC-1 (
Mß2, CD11b/CD18) is a major, if not
unique, fibrinogen receptor on the polymorphonuclear neutrophil
(PMN) surface.5 6 Interestingly, MAC-1 expression
decreases after application of c7E3Fab in patients undergoing
percutaneous transluminal coronary
angioplasty.7 These observations suggest that attenuation
of fibrinogen-mediated PMN-platelet interactions might contribute
to the beneficial effects of c7E3Fab on myocardial infarction,
myocardial stunning,4 and clinical
restenosis.1 Experimentally, coinfusion of platelets, PMNs, and plasma in isolated hearts exacerbated myocardial dysfunction after ischemia and reperfusion, an effect involving adhesion molecules of the selectin class.8 Generally, however, selectins mediate a first contact between cell membranes, inducing deceleration of cell movement (rolling).9 For firm adhesion, subsequent interaction of adhesion molecules, eg, GPIIb/IIIa on platelets and MAC-1 on PMNs with their common ligand, fibrinogen, has to prevail.6
Although the clinical and experimental evidence is compelling, PMN-platelet interaction inside the heart has not yet been studied. Therefore, in the present study, we investigated whether platelets interact with PMNs via fibrinogen binding to its receptors during postischemic reperfusion. We used ex vivo microscopy to analyze whether c7E3Fab and LPM19c, a blocking antibody against the fibrinogen-binding site of MAC-1,10 could reduce PMN-platelet coaggregate formation in the coronary effluent and in epicardial microvessels. Moreover, the influence of both fibrinogen receptor blockade strategies on external heart work before and after an ischemic period was studied. The study was conducted to reveal whether the therapeutic benefit of GPIIb/IIIa blockade implicates, beyond inhibition of platelet aggregation, a functionally relevant reduction of heterotypic PMN-platelet coaggregate formation.
| Methods |
|---|
|
|
|---|
Preparation of Human PMNs and Platelets
Because c7E3Fab was raised against the activated human
platelet epitope GPIIb/IIIa,11 human PMNs and
platelets were used throughout the study. Human PMNs were isolated
from peripheral venous blood of healthy volunteers by
magnetic bead separation, as described elsewhere.12 In
brief, purification of neutrophils from 10 mL of human blood (0.1%
EDTA) was achieved by magnetic separation of cells previously labeled
with a CD15 antibody carrying iron microbeads (Miltenyi Biotech). Cells
were analyzed by flow cytometry (FACScan, Becton Dickinson) and
proved to be 99% PMNs. Cells were resuspended in Tyrodes solution
(pH 7.40), counted, and diluted in 3 mL Tyrodes solution;
where indicated, they were incubated with c7E3Fab (16.6 µg/mL),
LPM19c (10 µg/mL), or control antibody MCA928 (10 µg/mL) for 10
minutes.
Platelets were isolated from the platelet-rich plasma of the same blood samples described above by centrifugation (2000g, 5 minutes), discarding the supernatant, and washing them twice in PBS (pelleting was performed by centrifugation at 2000g, 5 minutes). From the resulting pellet, 400x106 platelets were taken, diluted in 3 mL of Tyrodes solution, and, where indicated, incubated with c7E3Fab (16.6 µg/mL) or F2883 (16.6 µg/mL) for 10 minutes.
For heterotypic aggregate analysis, PMNs were labeled with rhodamine 6G and platelets were labeled with BCECF-AM for 10 minutes. This labeling specifically increases PMN fluorescence (from 13 to 265 fluorescence units [FU]; mean fluorescence) and platelet fluorescence (from 10 to 1355 FU; mean fluorescence) without activating either cell compartment.
Preparation of Isolated Hearts
The care of the animals and all experimental procedures
conformed with the Guide for the Care and Use of Laboratory
Animals. Hearts were isolated and perfused as previously
described.13 In brief, male guinea pigs (250 to
350 g) were anesthetized (20 mg ketamine) and
decapitated. The aorta was quickly cannulated, and hearts were
retrogradely perfused with a modified Krebs-Henseleit buffer gassed
with 94.5% O2 and 5.5%
CO2 (37°C, pH 7.40±0.05).
The working heart preparation was established as previously described.13 The caval and azygos veins were ligated. Pulmonary venous entry was used for a canula that, in the working mode of the heart, provided orthograde access for the perfusate. In the perfusion apparatus, it was possible to switch between nonworking (Langendorff) mode and working heart mode. External heart work was calculated as the sum of pressure-volume work (developed aortic pressure multiplied by cardiac output) and acceleration work. For microscopic evaluation (Langendorff preparation), caval, azygos, and pulmonary veins were ligated. Retrograde perfusion was maintained throughout the experiment and during the in situ microscopic period.
Flow Cytometry
FITC- or PE-labeled isotype controls of either
fluorescence with a nonbinding, fluorescence-labeled
antibody (anti-mouse IgG1, MCA928F, or MCA928PE) served to determine
nonspecific binding and background fluorescence in a flow
cytometer (FACScan, Becton Dickinson). For isolated PMNs and
platelets, flow cytometric analysis was performed before
and after postischemic coronary passage. CD11b and
GPIIb/IIIa (CD41) were detected by Serotec antibodies MAC551 and
MCA467, respectively, except that after LPM19c incubation, CD11b was
detected by the PE-labeled LPM19c antibody (Dako), because MCA551 did
not recognize the LPM19c binding.
In fluorescence microscopy experiments, detection of the interaction of rhodamine 6Glabeled PMNs and BCECF-AMlabeled platelets14 in the coronary effluent (see below) was performed by flow cytometry. Data analysis of 10 000 events was performed with LYSIS II software on a Hewlett-Packard system.
Fluorescence Microscopy
For the purpose of in situ microscopy, hearts were
arrested in cold cardioplegic solution (28 mmol/L potassium) after
ischemia, 5 minutes of reperfusion, and infusion of PMNs and
platelets for 3 minutes (Figure 1
).
Thereafter hearts were placed on a specially designed microscopic stage
holder, with the surface of the left ventricle exposed for microscopic
observation (Ploemopak, Leitz) with a 10x objective (L10, 0.22
numerical aperture; Leitz). Images were generated by a
charge-coupled-device camera (COHU 4400, Prospective Measurements). The
distribution of platelets that were labeled with BCECF-AM was
analyzed by epi-illumination with an H130 mercury light source
and a I2,3 filter block (Leitz). The
distribution of PMNs was studied under a rhodamine 6Gselective
N2 filter block (Leitz).
|
This method allowed for quantitative analysis of coaggregate retention of PMNs and platelets in epicardial regions of the myocardium (0.5- to 1.5-mm depth). Moreover, PMN-platelet interaction was detected in cell clusters containing both rhodamine 6G (PMN-specific) and BCECF-AM (platelet-specific) fluorescence. The resolution would not allow for detection of single platelets on PMNs, thus revealing heterotypic coaggregates of 2 or more platelets on 1 or more PMNs.
External Heart Work
Saline-perfused, isolated hearts performing pressure-volume work
at 12 cm H2O preload and 60 mm Hg afterload
(20 minutes) were subjected to ischemia (37°C, 15 minutes)
and reperfused in Langen-dorff mode (20 minutes). After 2 minutes of
reperfusion, which allowed for fibrinogen saturation of the perfusion
system, PMNs and platelets were simultaneously infused
with fibrinogen for 3 minutes (with or without previously incubated
antibody). After 20 minutes of reperfusion, external heart work was
initiated for a second time (20-minute duration). The ratio of external
heart work achieved postischemically divided by the
preischemic level was defined as recovery of external heart
work, given as a percentage of the preischemic value.
Statistical Methods
The results are given as mean±SEM. Statistical analysis
was performed with 1-way ANOVA. Whenever a significant effect was
obtained with ANOVA, we performed multiple comparison tests between the
groups by using the Student Newman Keuls procedure (SPSS statistical
software package). Differences between groups were considered
significant for P<0.05.
| Results |
|---|
|
|
|---|
Fibrinogen Receptor Expression on Platelets and PMNs
Pharmacological studies revealed that c7E3Fab dose-dependently
competed with MCA467, a GPIIb/IIIa-recognizing antibody, and with
fibrinogen (Figure 2A
). On the other
hand, LPM19c dose-dependently bound to its epitope, MAC-1. Both c7E3Fab
and LPM19c competed with PMN fibrinogen binding (Figure 2B
).
|
GPIIb/IIIa expression on platelets before and after
coronary passage did not change significantly (238±54 FU
before versus 262±52 FU after coronary passage). Incubation of
platelets with c7E3Fab reduced GPIIb/IIIa detection with the MCA
467 to 12±5fU before and 10±4 FU after coronary passage
(Figure 3A
). Concomitantly,
P-selectin detection on platelets (145±16 FU before and 117±13 FU
after coronary passage) decreased after c7E3Fab incubation
(54±11 FU before and 45±11 FU after coronary passage; data
not shown). Concerning neutrophils, although no significant difference
was found in MAC-1 detection by MCA551 with or without c7E3Fab
treatment, an increase of PMN MAC-1 detection during
postischemic coronary passage was reversed by
c7E3Fab (Figure 3B
). Application of LPM19c reduced available
MAC-1, as detected by a PE-labeled LPM19c antibody, from 36.5±3.1 to
3.2±1.7 FU (Figure 3b
), whereas platelets, which were not
incubated with LPM19c before coronary passage, were left
unchanged (Figure 3a
).
|
PMN-Platelet Interaction in Coronary Effluent
After Ischemia
In addition to its inhibitory effects on adhesion
molecule expression, the potential of c7E3Fab to reduce coaggregate
formation was more directly assessed by labeling whole PMNs and
platelets with 2 different fluorescent dyes, rhodamine 6G
and BCECF-AM, respectively. Thereafter, PMNs and platelets were
infused with or without fibrinogen in a postischemic guinea
pig heart.
Compared with nonischemic control hearts, which released
16±2% of the PMNs occupied with platelets, ischemia (15
minutes, 37°C) enhanced PMN-platelet interaction, the extent of
which depended on the presence of fibrinogen (49±13% versus 30±15%
with or without fibrinogen, respectively; Figure 4
). Inhibition of GPIIb/IIIa (c7E3Fab) or
MAC-1 (LPM19c) reduced PMN-platelet interaction to control levels
(17±2% and 14±3%, respectively), whereas control antibodies had no
effect (data not shown). Moreover, the number of
platelets interacting with each PMN was analyzed, expressed
as the mean of platelet-specific fluorescence on PMNs. The
results paralleled the percentage of coaggregates found in the PMN
population (Figure 4
).
|
Retention of PMN-Platelet Aggregates in the Reperfused
Heart
Using double-fluorescence videomicroscopy ex vivo, we
studied the retention of PMNs, platelet aggregates, and heterotypic
PMN-platelet aggregates in the epicardial microcirculation.
Examples of this method are given in Figure 5
, showing PMN retention (rhodamine
fluorescence) (Figure 5A
) and platelet aggregation
(BCECF-AM fluorescence) (Figure 5B
) at the same site, whereas
c7E3Fab inhibited the interaction (Figure 5C
and D).
Quantitative analysis revealed that postischemic
retention of PMN-platelet aggregates in the presence of fibrinogen
increased from 1.7±0.2/cm2 (nonischemic
or postischemic controls) to
5.0±0.7/cm2 (Figure 6
). This PMN-platelet
coaggregation was blocked by the addition of c7E3Fab and LPM19c (Figure 6
).
|
|
Analysis of the homotypic platelet aggregates revealed a
similar impact of ischemia and reperfusion, fibrinogen, or
c7E3Fab treatment on the distribution in the epicardial
microcirculation (Figure 6
). The latter reduced the platelet
aggregate retention from 21±3/cm2 to
7±1/cm2. LPM19c had no effect exceeding the use
of an unspecific antibody (13±2/cm2).
Myocardial Function
At a constant preload of 12 cm H2O and an
afterload of 60 mm Hg, isolated guinea pig hearts performed an
external heart work of 377±37 mJ ·
min-1 · g-1. After
ischemia and reperfusion without further intervention (Figure 7
), heart work decreased to 313±34
mJ · min-1 ·
g-1, indicating a postischemic
recovery of 83±6%. Simultaneous addition of washed PMNs,
platelets, and fibrinogen during early reperfusion (2 to 5 minutes)
further decreased the postischemic external heart work to
46±4% of the preischemic level, an effect reversed by
addition of c7E3Fab (74±11%) or LPM19c (71±6%). These changes in
recovery were not related to changes in heart rate. Thus,
postischemic recovery of external heart work was sensitive
to the application of a PMN-platelet bolus in the presence of
fibrinogen, a phenomenon inhibited by the coapplication of c7E3Fab or
LPM19c.
|
| Discussion |
|---|
|
|
|---|
Although the leukocyte contribution to myocardial reperfusion injury
has been demonstrated13 15 16 17 18 19 and
discussed20 before, only have recent studies revealed an
interaction between the leukocyte and platelet compartments as
effectors of reperfusion injury.8 21 22 In particular, the
study of Lefer and coworkers8 suggested that blockade of
P-selectin and its ligands, eg, PSGL-1, reduced
PMN-plateletdependent myocardial reperfusion injury. PMN
interaction with activated platelets resembles
PMNendothelial cell interaction,23
wherein a first, reversible contact of both cell types is mediated
preferentially by selectins. This process is followed by the highly
selective, firm adhesion24 only if integrin activation is
achieved by other substances, eg, platelet-activating
factor6 or chemokines,25 among others. Firm
adhesion of PMNs on platelets preferentially involves the integrins
MAC-1 (aMß2, PMNs) and GPIIb/IIIa (aIIbß3, platelets) and their
mutual ligand fibrinogen.6 Accordingly, in our study, the
blockade of its receptors by c7E3Fab or LPM19c specifically reduced
fibrinogen-dependent PMN-platelet coaggregate formation (Figure 4
) and retention (Figure 6
) in reperfused hearts.
However, other PMN-platelet interactions may have contributed to
the observed effects. Interestingly, platelet inhibition through
c7E3Fab extends to P-selectin reduction, since P-selectin expression
was reduced by two thirds after c7E3Fab incubation. Therefore, the
initial contact between platelets and PMNs is also reduced by this
intervention. However, P-selectin inhibition is exceeded by the
reduction of PMN-platelet coaggregation achieved by c7E3Fab and
particularly LPM19c, which was not added to the platelet
compartment, suggesting an essential role of the fibrinogen receptor
interaction. On the other hand, coaggregate formation was still
observed without fibrinogen, possibly due to additional firm
adhesion-molecule interactions, eg, CD11a/CD18 (PMNs) and intercellular
adhesion molecule-2 (platelets).
By virtue of the chosen model, interaction of human PMN MAC-1 with
endothelial intercellular adhesion molecule-1 was
excluded (data not shown). Therefore, the protective effect of c7E3Fab
and LPM19c on myocardial function (Figure 7
) depends on a
reduction of coaggregate formation (Figure 6
) and nonspecific
PMN plugging in capillaries. Although the latter effect was more
pronounced in the presence of LPM19c (data not shown), functional
improvement with both interventions was similar (Figure 7
).
Taken together, formation of a fibrinogen bridge between PMNs and
platelets appears to be a crucial step for postischemic
functional detriment in our model. Moreover, because LPM19c was
incubated only with PMNs and not with the platelet compartment,
inhibition of fibrinogen binding on PMNs appears sufficient for
cardioprotection in this model. Fibrinogen-dependent interaction of
PMNs and platelets itself signals cell activation through tyrosine
kinases.26 27 28 In addition, coaggregation decreases the
distance for platelet-derived activation factors, eg,
platelet-activating factor6 or
interleukin-1,29 to their receptors on PMNs and may
subsequently induce PMN activation.30
Beyond platelet GPIIb/IIIa inhibition, did c7E3Fab interfere with
MAC-1 on PMNs? Although c7E3Fab was not directly competing with the
MAC-1 detection antibody used (Figure 3B
), it interfered with
MAC-1 upregulation (Figure 3B
, cf Reference 7 )
and PMN-fibrinogen binding (Figure 2B
). A functional interaction
of the original clone 7E3 or c7E3Fab with purified leukocytes has been
found31 32 33 34 (for a review, see Reference
35 ), although no direct binding of FITC-labeled
c7E3Fab to purified PMNs could be detected.7 Moreover, in
a homologous model of postischemic PMN-platelet
infusion in rat hearts, an
4-fold decrease of PMN retention was
described,36 which in such a model implies an
MAC-1/intercellular adhesion molecule-1 interaction.37
Therefore, the functional blockade of MAC-1 appears to be an effect of
c7E3Fab application, in addition to platelet GPIIb/IIIa
blockade.
Clinically, the effectiveness of GPIIb/IIIa blockade is confirmed by a reduction in the incidence of death, myocardial infarction, or revascularization procedures after acute coronary syndromes.1 2 3 38 An association between restenosis and PMN-platelet interactions has been suggested before.39 Interestingly, the occurrence of restenosis, 1 clinical end point reduced by c7E3Fab treatment, is experimentally also attenuated by an MAC-1 antibody.40 In addition, c7E3Fab application improves regional myocardial function after acute myocardial infarction and primary percutaneous transluminal coronary angioplasty with stent implantation.4 The functional impact of c7E3Fab might involve the reduced formation of platelet-PMN coaggregates.7 41 Therefore, it is tempting to speculate that the beneficial effect of adjuvant c7E3Fab therapy on myocardial reperfusion injury in patients has an anti-PMN component.
In conclusion, we have shown that c7E3Fab reduces the formation of PMN-platelet coaggregates concomitantly with the retention of such aggregates in the postischemic myocardium, the latter phenomenon inducing a distinct impairment of postischemic myocardial function. Further studies are needed to characterize the contribution of PMNs to postischemic microthrombosis. A strict separation of inflammatory and thrombotic events may give way to a more interactive paradigm of both processes, since c7E3Fab appears as adjuvant coronary therapy aimed at both events.
| Acknowledgments |
|---|
Received February 3, 2000; accepted February 18, 2000.
| References |
|---|
|
|
|---|
2ß3 and
stimulated by platelet-activating factor. J Clin
Invest. 1997;100:20852093.[Medline]
[Order article via Infotrieve]
Mß2 integrin I domain recognized by anti-
IIbß3 monoclonal
antibody 7E3 and its involvement in leukocyte adherence. J
Biol Chem. 1998;273:2037220377.
Mß2) and thereby results in a functional blockade in vitro and in
vivo. Circulation. 1999;18(suppl I):I-333. Abstract.
Vß3 and
activated
Mß2 receptors: with a review of
platelet-leukocyte interactions. Thromb Haemost. 1999;82:326336.[Medline]
[Order article via Infotrieve]
This article has been cited by other articles:
![]() |
A. R. Pries, H. Habazettl, G. Ambrosio, P. R. Hansen, J. C. Kaski, V. Schachinger, H. Tillmanns, G. Vassalli, I. Tritto, M. Weis, et al. A review of methods for assessment of coronary microvascular disease in both clinical and experimental settings Cardiovasc Res, June 25, 2008; (2008) cvn136v2. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Xu, Y. Huo, M.-C. Toufektsian, S. I. Ramos, Y. Ma, A. D. Tejani, B. A. French, and Z. Yang Activated platelets contribute importantly to myocardial reperfusion injury Am J Physiol Heart Circ Physiol, February 1, 2006; 290(2): H692 - H699. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Herrmann Peri-procedural myocardial injury: 2005 update Eur. Heart J., December 1, 2005; 26(23): 2493 - 2519. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ferenc and F.-J. Neumann Efficacy of primary PCI: the microvessel perspective Eur. Heart J. Suppl., October 1, 2005; 7(suppl_I): I4 - I9. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. S. Ritter, K. M. Stempel, B. M. Coull, and P. F. McDonagh Leukocyte-Platelet Aggregates in Rat Peripheral Blood After Ischemic Stroke and Reperfusion Biol Res Nurs, April 1, 2005; 6(4): 281 - 288. [Abstract] [PDF] |
||||
![]() |
M. Gawaz Role of platelets in coronary thrombosis and reperfusion of ischemic myocardium Cardiovasc Res, February 15, 2004; 61(3): 498 - 511. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Kupatt, R. Wichels, J. Horstkotte, F. Krombach, H. Habazettl, and P. Boekstegers Molecular mechanisms of platelet-mediated leukocyte recruitment during myocardial reperfusion J. Leukoc. Biol., September 1, 2002; 72(3): 455 - 461. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Palmerini, M. A. Nedelman, L. E. Scudder, M. T. Nakada, R. E. Jordan, S. Smyth, R. E. Gordon, J. T. Fallon, and B. S. Coller Effects of abciximab on the acute pathology of blood vessels after arterial stenting in nonhuman primates J. Am. Coll. Cardiol., July 17, 2002; 40(2): 360 - 366. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Barrabes, D. Garcia-Dorado, M. Mirabet, R.-M. Lidon, B. Soriano, M. Ruiz-Meana, P. Pizcueta, J. Blanco, Y. Puigfel, and J. Soler-Soler Lack of effect of glycoprotein IIb/IIIa blockade on myocardial platelet or polymorphonuclear leukocyte accumulation and on infarct size after transient coronary occlusion in pigs J. Am. Coll. Cardiol., January 2, 2002; 39(1): 157 - 165. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |