OUP user menu

★ Editor's choice ★

Circulating platelet–progenitor cell coaggregate formation is increased in patients with acute coronary syndromes and augments recruitment of CD34+ cells in the ischaemic microcirculation

Konstantinos Stellos , Boris Bigalke , Oliver Borst , Florian Pfaff , Alexandra Elskamp , Saskia Sachsenmaier , Ruben Zachmann , Kimon Stamatelopoulos , Tanja Schönberger , Tobias Geisler , Harald Langer , Meinrad Gawaz
DOI: http://dx.doi.org/10.1093/eurheartj/eht131 2548-2556 First published online: 17 April 2013

Abstract

Aims The aim of the present study was to evaluate the levels of platelet interaction with circulating CD34+ cells in patients with stable angina pectoris (SAP) and acute coronary syndromes (ACS) and to study the functional consequence of coaggregates formation in vitro and in vivo.

Methods and results Platelet binding to circulating progenitor cells was defined by the presence of the platelet-specific marker glycoprotein Ib (CD42b) on the surface of CD34+ cells using flow cytometry. The percentage of CD34+/CD42b+ cell coaggregates was increased in patients with ACS (n = 162), and especially in patients with ST-elevation myocardial infarction (STEMI) (n = 44), compared with patients with SAP (n = 116; P < 0.001). In the ANCOVA analysis, platelet/CD34+ cell coaggregates were independently increased in ACS after adjustment for possible confounders. In a subgroup of our cohort, we also evaluated the levels of CD34+/CD133+/CD42b+ cell coaggregates, which were also significantly increased in ACS, and especially in STEMI (P < 0.05). Platelet/CD34+ cell coaggregates formation correlated with platelet activation (P = 0.001). In a prospective pilot study of patients with AMI (n = 40) using cardiac MRI, patients with increased baseline platelet/CD34+ cell coaggregates presented with a less myocardial infarct size and better left ventricular function at a 3-month follow-up compared with patients with lower coaggregates (P < 0.05 for all). The adhesion of platelet/CD34+ cell coaggregates onto the extracellular matrix and to endothelial monolayer was enhanced compared with CD34+ under high shear rates in vitro (P < 0.05) and within the microcirculation in mice after ischaemia/reperfusion injury as assessed by intravital microscopy (P < 0.05).

Conclusions These findings imply that circulating platelet/CD34+ cell coaggregate levels are increased in ACS, especially in STEMI, which may be a novel mechanism of domiciliation of CD34+ progenitor cells to the injured microvasculature after acute myocardial infarction.

  • Platelets
  • Progenitor cells
  • Acute coronary syndrome
  • Microcirculation

See page 2501 for the editorial comment on this article (doi:10.1093/eurheartj/eht198)

Introduction

Increasing evidence suggests that circulating CD34+ progenitor cells contribute to cardiovascular homoeostasis in a paracrinic way or through differentiation into endothelial cells and smooth muscle cells in vivo.1,2 In patients with acute myocardial infarction (AMI), CD34+ progenitor cells are increased in the peripheral blood and are rapidly recruited to the myocardium mediating a protective effect of ischaemic preconditioning.35 Trafficking and peripheral recruitment of circulating progenitor cells is a multi-step process that includes chemoattraction, migration, adhesion, and tissue invasion.2 Domiciliation of circulating CD34+ progenitor cells is still not adequately elucidated in patients with acute coronary syndrome (ACS).

It has been recently reported that platelets interact with progenitor cells in vitro and in vivo playing an important role in their chemotaxis,6 adhesion on vascular wall,711 and differentiation to endothelial cells.7,11 Progenitor cells poorly adhere to the exposed extracellular matrix proteins, such as collagen, due to lack of the respective receptors, but they are tethered to platelet surface P-selectin via P-selectin glycoprotein ligand (PSGL)-1 binding.7 We have recently reported that platelet adhesion represents a critical step for the targeting of progenitor cells to sites of endothelial disruption.8,1113 In addition to directing the migration and adherence of bone marrow-derived cells to sites of vascular injury, platelets may also induce the differentiation of progenitor cells into mature endothelial cells.7,10,11

Platelet-derived stromal cell-derived factor-1 (SDF-1), the dominant chemokine for bone marrow-derived progenitor cells, is enhanced in ACS and platelet-SDF-1 levels correlated with the number of circulating CD34+ cells in patients with coronary artery disease.12,14 Platelet-SDF-1 and number of CD34+ cells are associated with haemodynamic function in patients with AMI.15

Whether platelets form coaggregates with circulating CD34+ progenitor cells in patients with ACSs and augment peripheral recruitment within the ischaemic microcirculation has not been evaluated so far.

Subjects and methods

Detailed description of patients and methods as well as of material used is presented on Supplementary material online, Methods and Materials.

Patients

A total of 278 consecutive patients were recruited consecutively. one hundred and sixteen patients with suspected or known coronary artery disease with typical symptoms for stable angina (SAP) were referred to our hospital for coronary angiography according to the ACC/AHA guidelines for coronary angiography.16 Patients with SAP had either typical angina on exertion and/or a pathological exercise test and were negative for markers of myocardial ischaemia (troponin I, creatinine kinase). One hundred and sixty-two patients presented in our emergency room with ACS were immediately proceeded to percutaneous coronary intervention. Among patients with ACS, 28 patients presented with unstable angina pectoris (UAP), 90 with non-ST-elevation myocardial infarction (NSTEMI) and 44 with ST-elevation myocardial infarction (STEMI). Coronary interventions were performed within 6 h after admission to the hospital in patients with ACS. Patients with SAP were investigated on schedule. Blood samples for biochemical parameters including myocardial necrosis markers and C-reactive protein as well as for flow cytometry were obtained on admission. The institutional ethical committee of the University of Tübingen approved the study and all subjects gave their written informed consent.

We also recruited n = 40 consecutive patients with acute myocardial infarction presenting to the emergency department of the University Hospital of Tübingen, Germany, as previously described.17 The baseline characteristics of the ERMIS cohort are presented in our previous publication.17 All patients underwent coronary angiography and primary coronary intervention. All patients underwent 48 h and 3 months after the myocardial infarction cardiac MRI examinations at 1.5T (Magneton Avanto, Siemens Medical Solutions, Erlangen, Germany) for volumetric assessment and LGE scar imaging using standard methods.18 Image analysis was based on the 17-segment model according to the recommendations of the ESC/AHA as previously described.18,19

Peripheral blood mononuclear cell isolation and flow cytometry

Mononuclear cells were isolated according to standard protocols, as previously described12,20 and data were analysed with the use of the CellQuest software (FACSCalibur, Becton Dickinson, Heidelberg, Germany). Units of all measured components are absolute cell counts obtained after the measurement of 250 000 events in the lymphocyte gate.

Whole blood platelet flow cytometry

Platelets obtained from 170 consecutive patients were studied for the surface expression of P-selectin (CD62P), SDF-1, and GPIb (CD42b) by flow cytometric analysis, as previously described.12,21 Conjugated monoclonal antibodies were used to measure platelet-SDF-1 with a two-colour flow cytometry in patients' whole blood, as previously described.12 CD42b-PE served as control antibody to identify the platelet population in the whole blood. Specific monoclonal antibody binding was expressed as a mean fluorescence intensity (MFI) and was used as a quantitative measurement of platelet proteins surface expression.12,21

Isolation and culture of platelets, human CD34+ cells, and human arterial endothelial cells

Human platelets were isolated as previously described.7,11 Human CD34+ cells were isolated from either human umbilical cord blood or bone marrow and cultured as previously described.7,10 Human arterial endothelial cells (HAECs) were isolated and passaged according to techniques previously described.22

Adhesion under dynamic conditions (flow chamber) and colony-forming unit assay

Evaluation of CD34+ cell adhesion to collagen and to cultured endothelial monolayer was performed as previously described.7,11,13 Analysis of the effect of platelet- CD34+ adhesion on the formation of endothelial progenitor cell colonies was performed, as previously described.11,13

Intestinal ischaemia/reperfusion model in mice and intravital microscopy

Fluorescent platelet- CD34+ coaggregates were infused before intestinal ischaemia/reperfusion injury and visualized in the post-ischaemic microcirculation by intravital fluorescence microscopy, as previously described.11,13

Data presentation and statistical analysis

Data from patients are presented as median ± inter-quartile range (IQR), unless otherwise stated. In vitro data are presented as mean ± SD. All tests were two-tailed and statistical significance was considered for P-values <0.05. All statistical analyses were performed using SPSS version 21 for windows (Chicago, IL, USA).

Results

Circulating platelet/CD34+ cell coaggregates levels are increased in patients with acute coronary syndromes and especially with ST-elevation myocardial infarction

The demographic details of our cohort based on the clinical presentation of patients are given on Table 1. The number of CD34+ progenitor cells in patients with CAD was measured in the lymphocyte gate, as depicted in Figure 1A and described in ‘Methods’. A significant increase in the number of CD34+ progenitor cells in patients with ACS (P = 0.011), especially in patients with STEMI (P = 0.005), was observed in comparison with patients with SAP, which was independent of cardiovascular risk factors and medication (data not shown). Platelet binding to circulating progenitor cells was defined by the presence of the platelet-specific marker glycoprotein Ib (CD42b) on the surface of CD34+ cells using flow cytometry (Figure 1B). The formation of platelet-CD34+ cell coaggregates was further verified by immunofluorescence microscopy (Figure 1C). The percentage of CD34+/CD42b+ cell coaggregates of the total CD34+ cells was defined as platelet/CD34+ cell coaggregates. Platelet/CD34+ cell coaggregates were increased in patients with ACS compared with SAP or to healthy subjects (P < 0.001 for both; Figure 1D). No significant difference was observed between SAP patients and healthy subjects (Figure 1D). The highest levels of platelet/CD34+ cell coaggregates were observed in patients with STEMI [SAP vs. UAP vs. NSTEMI vs. STEMI: median (IQR): 7.42 (8.19) vs. 16.45 (14.3) vs. 16.28 (15.23) vs. 25.43 (28.2); P < 0.001 Kruskal–Wallis test: Figure 1E].

View this table:
Table 1

Baseline patients' characteristics of the study population

CharacteristicsTotal (n = 278)SAP (n = 116)UAP (n = 28)NSTEMI (n = 90)STEMI (n = 44)P-value
Age, years (mean ± SD)66.5 ± 11.866.9 ± 8.969.3 ± 8.766.6 ± 14.263.6 ± 14.40.234
Female, n (%)61 (21.9)17 (14.7)8 (28.6)25 (27.8)11 (25)0.075
CVRFs, n (%)
 Arterial Hypertension228 (82)101 (87.1)24 (85.7)74 (82.2)29 (65.9)0.025
 BMI (mean ± SD)28 ± 4.228.6 ± 4.327.9 ± 427.5 ± 3.927.1 ± 4.70.142
 Diabetes mellitus type 297 (34.9)46 (39.7)7 (25)33 (36.7)11 (25)0.264
 Family history of CAD62 (22.3)28 (24.1)10 (35.7)15 (16.7)9 (20.5)0.226
 Hyperlipidaemia195 (70.1)95 (81.9)18 (64.3)56 (62.2)26 (59.1)0.039
 Smoking116 (41.7)50 (43.1)11 (39.3)35 (38.9)20 (45.5)0.940
Coronary angiography
 TIMI flow score 08 (2.9)2 (1.7)1 (3.6)3 (3.3)2 (4.5)0.829
 TIMI flow score 149 (17.6)25 (21.6)5 (21.4)10 (11.1)9 (20.5)
 TIMI flow score 2110 (39.6)42 (36.2)12 (42.9)40 (44.4)16 (36.4)
 TIMI flow score 3111 (39.9)47 (40.5)10 (35.7)37 (41.1)17 (38.6)
 Calcium score13.4 ± 613.9 ± 5.814.6 ± 6.813.4 ± 6.111.3 ± 60.056
 EF (laevocardiography)51.9 ± 11.953.3 ± 12.154.1 ± 9.951.2 ± 12.348 ± 11.10.094
Medication, n (%)
 Aspirin171 (61.5)90 (77.6)21(75)41 (45.6)19 (43.2)<0.001
 Clopidogrel87 (31.3)47 (40.5)10 (35.7)20 (22.2)10 (22.7)0.037
 Statins135 (48.6)74 (63.8)14 (50)31 (34.4)16 (36.4)<0.001
 ACE-inhibitors136 (48.9)70 (60.3)15 (53.6)36 (40)15 (34.1)0.011
 AT1 receptor blocker38 (13.7)19 (16.4)4 (14.3)9 (10)6 (13.6)0.698
 Aldosterone antagonist7 (2.5)4 (3.4)0 (0)3 (3.3)0 (0)0.481
 GPIIb/IIIa antagonist42 (15.1)8 (6.9)4 (14.3)14 (15.6)16 (36.4)<0.001
 Beta-blocker177 (63.7)91 (78.4)18 (64.3)45 (50)23 (52.3)0.003
 Diuretics97 (34.9)39 (33.6)9 (32.1)36 (40)13 (29.5)0.522
 Vitamin K antagonist17 (6.1)7 (6)2 (7.1)7 (7.8)1 (2.3)0.639
Platelet activation and reactivity (n = 170, mean ± SD)
 CD62P expression15.4 ± 9.712.5 ± 7.913.3 ± 5.917.1 ± 9.918.5 ± 11.90.007
 SDF-1/CXCL12 (MFI)67.2 ± 8943.8 ± 69.949.1 ± 53.783.8 ± 11488.5 ± 660.025
Figure 1

Circulating platelet–progenitor cell coaggregates are increased in patients with acute coronary syndromes. (A) Representative forward scatter and side scatter dot plot of peripheral blood mononuclear cells. Progenitor cells were counted using the lymphocyte gate as shown. (B) Representative CD42b-PE (platelet marker) and CD34-FITC (progenitor cell marker) flow cytometry dot plot in a patient with stable angina pectoris or acute coronary syndrome, respectively. The upper right area depicts the platelet-CD34+ coaggregates, while the lower right shows the CD34+ progenitor cells, which are negative for the platelet marker CD42b. The upper and lower areas together show the total CD34+ progenitor cells. (C) Representative immunofluorescence photo of coaggregates formation. (D) Percentage of platelet/CD34+ cell coaggregates of total CD34+ cells in healthy subjects (n = 15), patients with stable angina pectoris (n = 116) and acute coronary syndromes (n = 162; ns: not significant; ***P < 0.001 compared with stable angina pectoris or healthy subjects). (E) Levels of platelet/CD34+ cell coaggregates according to the clinical presentation of the patients are depicted (ns, not significant; *P < 0.05, **P < 0.01, ***P < 0.001).

Next, we evaluated the association of platelet/CD34+ cell coaggregates with cardiovascular risk factors, baseline, and maximal troponin I (Tn-I) and creatine kinase (CK), C-reactive protein, cardiovascular therapy including antiplatelet therapy, and platelet activation measured by the surface expression of P-selectin on platelets and total number of circulating CD34+ cells. All significant correlations are presented in the Table 2. Platelet/CD34+ cell coaggregates levels were associated with maximal Tn-I (r = 0.221, P = 0.029) and C-reactive protein (r = 0.235, P = 0.021) independent of platelet activation (expression of platelet-bound P-selectin), antiplatelet therapy (aspirin and clopidogrel treatment), and number of total CD34+ cells. Of interest, there was a positive correlation with platelet activation and an inverse correlation with antiplatelet medication. We further assessed the impact of antiplatelet therapy on the platelet/CD34+ cell coaggregates levels in our cohort. Patients under aspirin treatment or under dual therapy (aspirin and clopidogrel) presented with significantly lower platelet/CD34+ cell coaggregates levels (no antiplatelet therapy vs. aspirin treatment vs. dual antiplatelet therapy: median (IQR): 16.04 (17.42) vs. 11.74 (12.98) vs. 10.72 (15.79), Kruskal–Wallis, P = 0.016; Mann–Whitney U test, P = 0.015 for both aspirin therapy or dual therapy vs. no antiplatelet therapy; P = 0.543 for aspirin treatment vs. dual antiplatelet therapy; Figure 2A). Nevertheless, in the subsequent ANCOVA analysis with all parameters that were found to be significantly associated with platelet/CD34+ cell coaggregates levels as covariates, the stepwise increase of platelet/CD34+ cell coaggregates in ACS (according to clinical presentation) was independent of cardiovascular risk factors, CK, Tn-I, C-reactive protein, medication, platelet activation (P-selectin expression), and number of circulating CD34+ cells (Table 2). In a subgroup of 91 patients, we further characterized the interaction of circulating CD34+/CD133+ progenitor cells with platelets. In a similar manner, platelet/CD34+/CD133+ cell coaggregates levels are stepwise significantly increased in patients with ACS depending on the clinical presentation (UAP, NSTEMI, STEMI) compared with patients with SAP (Figure 2B).

View this table:
Table 2

Spearman's rho correlation analysis and analysis of covariance for %platelet/CD34+ coaggregates and other covariates in CAD

ParametersSpearman's correlationANCOVA
rhoP-valueEffect sizeP-value
Cardiovascular risk factors
 Female sex0.1570.0080.110.738
 Diabetes mellitus type 2−0.1380.0222.930.088
 History of myocardial infarction−0.1890.0020.380.539
Biochemical parameters
 Baseline creatine kinase (CK)0.1570.0160.50.479
 Baseline troponin I (Tn-I)0.309<0.0011.750.188
 Maximal CK0.293<0.0010.0030.954
 Maximal Tn-I0.326<0.0010.560.456
 C-reactive protein0.20.0020.010.909
Cardiovascular therapy
 Antiplatelet therapy−0.1650.0070.0140.907
 Beta-blocker−0.1970.0010.220.641
 ACE-inhibitor−0.1270.0380.170.677
Flow cytometry parameters
 Platelet P-selectin (CD62P)0.283<0.0010.0850.77
 Platelet-SDF-1/CXCL120.357<0.0011.060.305
 Circulating CD34+ cell number−0.0240.6980.0530.818
Clinical presentation (Group SAP vs. UAP vs. NSTEMI vs. STEMI)0.523<0.00132.52<0.001
Figure 2

Circulating platelet–progenitor cell coaggregates and antiplatelet therapy and platelet activation (A) Platelet/CD34+ cell coaggregates in patients under aspirin treatment or dual therapy (aspirin and clopidogrel) compared with patients with no antiplatelet therapy. *P < 0.05 (B) In a subpopulation of the study (n = 91), we also measured the platelet binding on circulating CD34+/CD133+ progenitor cells. In a similar manner percentage of platelet/CD34+/CD133+ cell coaggregates are increased in patients with ST-elevation myocardial infarction when compared with patients with stable angina pectoris (*P < 0.05, **P < 0.01, ***P < 0.001). (C and D) In a subpopulation of the study (n = 170), we simultaneously investigated the relationship of platelet activation with the formation of coaggregates. The percentage of platelet/CD34+ coaggregates correlate with the surface expression of platelet-bound P-selectin and stromal cell-derived factor-1.

Moreover, we investigated if there is any significant difference regarding platelet/CD34+ cell coaggregates levels among patients with TIMI flow 0 vs. TIMI flow 1 vs. TIMI flow 2 vs. TIMI flow 3. Performing a Kruskal–Wallis test we observed a trend of higher platelet/CD34+ cell coaggregates levels in patients with no reflow (TIMI score 0; n = 8) compared with TIMI flow 1 (n = 48), 2 (n = 114), and 3 (n = 108) [% CD34+/CD42b+ cell coaggregates: TIMI-0 vs. TIMI-1 vs. TIMI-2 vs. TIMI-3: median (IQR): 19.62 (45.78) vs. 11.31 (13.88) vs.12.56 (14.69) vs. 13.25 (77.74), P = 0.08].

We recruited 16 ACS patients whom we followed up for 24 h and 5 ACS patients whom we followed up for 48–72 h. We observed that the mean value of platelet/CD34+ cell coaggregates levels is increased 24 h after myocardial infarction compared with baseline levels, although this difference did not reach significance due to probably the high SD and the small number of patients recruited (mean ± SD: Day 0 vs. Day 1: 20.5 ± 11.6% vs. 27 ± 22 (Wilcoxon signed-rank test: P = 0.332). After 48–72 h, there was practically no difference with baseline levels of platelet/CD34+ cell coaggregates (mean ± SD: Day 0 vs. Day 2–3: 19 ± 9.9% vs. 19.2 ± 21.6; Wilcoxon signed-rank test: P = 0.979).

Formation of platelet/CD34+ cell coaggregates correlates with the surface expression of platelet P-selectin and stromal cell-derived factor-1 in patients with coronary artery disease

Expression of both platelet-bound P-selectin and SDF-1/CXCL12 was significantly increased in patients with STEMI compared with patients with SAP, which was studied in a subcohort of n = 170 consecutive patients with CAD and presented at the Table 1. The formation of platelet/CD34+ cell coaggregates correlated with both platelet-bound P-selectin (Pearson correlation test after logarithmic transformation of the data: CD62P: r = 0.242, P = 0.001; SDF-1: r = 0.339, P < 0.001; Figure 2C and D, respectively).

Baseline circulating platelet/CD34+ cell coaggregates levels in patients with acute myocardial infarction and myocardial contractility and infarct size at baseline after a 3-month follow-up period using cardiac MRI (Evaluation of Regeneration after Myocardial Infarction Study—ERMIS pilot study)

We have previously reported using cardiac MRI at the pilot prospective ERMIS study that patients with initially increased number of CD34+ progenitor cells (>median value) showed a significant amelioration of stroke volume at the 3-month follow-up compared with patients with a decreased number of CD34+ cells.17 To assess differences in changes of left ventricular ejection fraction, stroke volume, and myocardial infarct size from baseline to the 3-month follow-up among tertiles of baseline coaggregates (platelet/CD34+ cell coaggregates and platelet/CD34+/CD133+ cell coaggregates), a Wilcoxon-matched pairs samples signed-rank test was used. As depicted at Table 3, patients belonging into the third tertile, but not in the first tertile, of platelet/CD34+ cell coaggregates presented a significantly decreased myocardial infarct size and better left ventricular ejection fraction and stroke volume after 3 months compared with baseline (P < 0.05 for all three parameters was observed only at the third tertile). Patients belonging into the third tertile of platelet/CD34+/CD133+ cell coaggregates presented with similar decrease in a myocardial infarct size, but the amelioration of LVEF and stroke volume was also significant at the first tertile.

View this table:
Table 3

Wilcoxon signed-rank test of the endpoints of the study (left ventricular ejection fraction, stroke volume and infarct size at baseline vs. 3-month follow-up using cardiac MRI) according to tertiles of CD34+ cells, platelet/CD34+ coaggregates and platelet CD34+/CD133+ cell coaggregates in patients with acute myocardial infarction (ERMIS pilot study cohort)

EndpointsFirst tertileSecond tertileThird tertile
Time 0 vs. FUP-valueTime 0 vs. FUP-valueTime 0 vs. FUP-value
% CD34+/CD42b+ cell coaggregates
 LVEF, mean ± SD52.1 ± 9.6 vs. 56.2 ± 5.70.0650.1 ± 10.1 vs. 53 ± 10.70.06551.8 ± 9.8 vs. 56 ± 8.60.011
 Stroke volume, mean ± SD72.8 ± 14.4 vs. 87.4 ± 17.10.07171.4 ± 16.7 vs. 75.8 ± 180.27284.4 ± 18.1 vs. 92.6 ± 23.30.037
 Infarct size, mean ± SD12.4 ± 13.8 vs. 10 ± 10.70.28616.6 ± 10.9 vs. 12.7 ± 9.70.03413.4 ± 10.4 vs. 8 ± 7.10.032
% CD34+/CD133+/ CD42b+ cell coaggregates
 LVEF, mean ± SD50 ± 9.4 vs. 55 ± 60.01251.3 ± 10.7 vs. 55.3 ± 9.50.03152.9 ± 10.2 vs. 56.7 ± 9.60.041
 Stroke volume, mean ± SD69.6 ± 18.2 vs. 85.7 ± 16.40.00781.7 ± 12.7 vs. 78 ± 16.20.37479.3 ± 18.2 vs. 95.4 ± 21.50.01
 Infarct size, mean ± SD11.1 ± 10.2 vs. 10.3 ± 8.70.38614.8 ± 16.2 vs. 10.6 ± 11.10.10916.3 ± 9.4 vs. 8.5 ± 7.90.003

Enhanced adhesion of platelet/CD34+ cell coaggregates and formation of endothelial colonies on extracellular matrix proteins

To evaluate the adhesive properties of platelet/CD34+ cell coaggregates, isolated CD34+ cells were pre-incubated with a physiological concentration of platelets (200 000/μL) and platelet/CD34+ coaggregates were quantified by flow cytometry (Figure 3A). Adhesion of platelet/CD34+ coaggregates onto immobilized collagen was significantly enhanced compared with platelet-free CD34+ cells under static (Figure 3B) and dynamic conditions (under high shear stress: CD34+ cells vs. coaggregates: mean ± SD: 13.5 ± 2.7 vs. 91.1 ± 23.3, P < 0.001; Figure 3C). Formation of endothelial colonies derived from platelet-CD34+ coaggregates was significantly increased compared with CD34+ cells in the absence of platelets (CD34+ cells vs. coaggregates: mean ± SD: 1.6 ± 2.8 vs. 9.7 ± 1.5; P < 0.001; Figure 3D).

Figure 3

Formation of platelet-CD34+ coaggregates facilitate the adhesion of CD34+ cells on collagen and their further differentiation to endothelial progenitor cells. (A) Formation of platelet-CD34+ coaggregates in vitro. Progenitor cells were pre-incubated with a physiological concentration of platelets (200 000/µL) and subsequently stained for the platelet marker CD42b and the progenitor cell marker CD34. Isolated CD34+ progenitor cells are negative for CD42b (left panel; ‘platelet-free’ CD34+ cells) while after pre-incubation of CD34+ progenitor cells with isolated platelets most of the progenitor cells became positive for the platelet marker CD42b (right panel). (B) Adhesion of platelet-CD34+ coaggregates on collagen under high shear stress in vitro. ADP-activated platelets were incubated under static conditions with human CD34+ cells in different ratios from 20:1 to 800:1 and the subsequently formed coaggregates were perfused over collagen. The mean number and SD of three-independent experiments of each group of adherent CD34+ cells over collagen are shown per high powerfield. Based on these results, the 400:1 ratio was chosen for all next described in vitro adhesion experiments unless otherwise mentioned. *P ≤ 0.05 vs. sole CD34+ cells over collagen. (C) CD34+ progenitor cells were pre-incubated with washed platelets in a 1:400 ratio. Isolated ‘platelet-free’ CD34+ cells or platelet-CD34+ cell coaggregates were perfused over immobilized collagen cover slips. The mean and SD of three-independent flow chamber experiments are shown. *P ≤ 0.05 vs. sole CD34+ cells over collagen. (D) Role of platelet-CD34+ cell coaggregate formation in the differentiation of CD34+ cells to late outgrowth endothelial progenitor cells in vitro. CD34+ cells or platelet-CD34+ coaggregates were cultivated on immobilized collagen, as described in ‘Methods’. Cultivation of CD34+ cells on immobilized fibronectin cover slips was used as a positive control for differentiation of CD34+ cells. Late outgrowth endothelial progenitor cell colony-forming units were counted between Days 5 and 10. Pre-incubation of CD34+ cells with platelets in a ratio of 100 platelets: 1 CD34+ cell resulted in an increased differentiation of CD34+ cells over collagen to late outgrowth endothelial progenitor cells. The mean and SD of three-independent experiments are shown. *P ≤ 0.05 vs. sole CD34+ cells over collagen.

Enhanced adhesion of platelet/CD34+ cell coaggregates over endothelial cells and in the microcirculation of mice after ischaemia/reperfusion injury

To evaluate the adhesion of platelet–progenitor cell coaggregates on endothelial cells in vitro, perfusion experiments were conducted over an endothelial monolayer of primary human endothelial cells under high shear stress of 2000−s. The adhesion of platelet-CD34+ coaggregates was substantially enhanced on both resting and TNF-α/INF-γ-activated HAECs under flow conditions (CD34+ cells vs. platelet/CD34+ cell coaggregates: mean ± SD: resting haECs: 1.8 ± 0.2 vs. 5.7 ± 0.6, P < 0.05; activated haECs: 18.7 ± 6.4 vs. 140 ± 45.6; P < 0.001; Figure 4A).

Figure 4

Platelet–progenitor cell coaggregates exhibit enhanced adhesive properties on human arterial endothelial cells under high shear stress in vitro and within the microcirculation after ischaemia/reperfusion injury in vivo. (A) To evaluate the adhesion of platelet-CD34+ coaggregates in vitro, we pre-incubated isolated human CD34+ cells with washed platelets as described in ‘Methods’. Human arterial endothelial cells were cultured until confluency. Isolated CD34+ cells or platelet-CD34+ coaggregates were perfused over these cover slips under high shear stress of 2000−s. The mean and SD of three-independent flow chamber experiments are shown. *P ≤ 0.05 vs. CD34+ cells. (B) Coaggregates–microvasculature interactions were investigated in the microcirculation of the small intestine of mice after ischaemia/reperfusion injury with the help of intravital fluorescence microscopy. Pre-incubation of CD34+ cells with platelets in a ratio of 100 platelets:1 CD34+ cell resulted in an increased adhesion of progenitor cells in the microvasculature. The mean and SD of adherent CD34+ progenitor cells and congregates 5, 10, and 30 min after ischaemia/reperfusion (n = 3–4 animals per group) in venules.

Adhesion of platelet-CD34+ coaggregates was investigated in vivo in the microcirculation of the small intestine of mice after ischaemia/reperfusion injury using intravital fluorescence microscopy, as previously described.8,11,13 Enhanced adhesion of platelet/CD34+ coaggregates occurs during reperfusion in the microcirculation of ischaemic intestine compared with control CD34+ cells in the absence of platelets (after 5 min reperfusion: CD34+ cells vs. coaggregates: mean ± SD: 2.9 ± 5.1 vs. 27.4 ± 7.4; P < 0.05; Figure 4B).

Discussion

The major findings of the present study are: (i) The levels of circulating platelet/CD34+ cell coaggregates are enhanced in patients with ACSs and especially in patients with STEMI independent of other confounders. (ii) The levels of circulating platelet/CD34+ cell coaggregates are associated with systemic platelet activation and antiplatelet therapy. (iii) Patients with increased baseline platelet/CD34+ cell coaggregates presented with less myocardial infarct size at the 3-month follow-up compared with patients with lower coaggregates. (iv) Platelet/CD34+ cell coaggregates show enhanced adhesion to collagen and endothelial cells in vitro and in vivo. Our findings imply that enhanced platelet activation in patients with STEMI result in an increased formation of circulating platelet/CD34+ cell coaggregates. An increase of platelet interaction with circulating CD34+ progenitor cells augments the adhesive properties and facilitates the recruitment of these cells to the vascular wall after ischaemia/reperfusion injury. Moreover, in a pilot study patients with increased baseline platelet/CD34+ cell coaggregates presented with a less myocardial infarct size at the 3-month follow-up compared with patients with lower coaggregates, which encourages the realization of further larger prospective studies to sufficiently address the clinical role and prognostic value of circulating platelet/CD34+ cell coaggregates in patients with acute myocardial infarction.

CD34+ progenitor cells are mobilized from bone marrow to peripheral circulation after myocardial ischaemia23 and play an important role in cardiovascular homoeostasis.1,24,25 In numerous studies, it has been shown that the levels of CD34+ or CD133+ cells predict cardiovascular event-free survival or all cause survival.20,2628 Moreover, transplantation of CD34+ or CD133+ cells exhibits increased potency and safety for therapeutic neovascularization, and functional regenerative recovery after myocardial infarction in vivo2934 or in patients with acute myocardial infarction or heart failure.3032,3541 Interestingly, CD133+ progenitor cell subpopulation is reported to be functionally more potent than CD34+ cells with respect to homing and vascular repair,42 which is in accordance to our findings reporting a higher coaggregates formation in CD34+/CD133+ cells than in CD34+ cells in patients with STEMI. Taken together, CD34+ and CD133+ progenitor cells are not only known as haematopoietic progenitor cells, but are also established as important cardiovascular cells taking a critical role in cardiovascular homoeostasis and wound healing. In the present study, we show that platelets adhere to CD34+ progenitor cells in the circulation in patients with ACS and especially with STEMI. We show that, in contrast to CD34+ cells that were not exposed to platelets, adhesion of platelet-CD34+ coaggregates is substantially enhanced on collagen and vascular lesions in vitro and in vivo. The presented ∼10-fold increase in adhesion of coaggregates in vitro and 6-fold increase in vivo compared with platelet-free CDC34+ cells shall be interpreted with caution since this situation (no coaggregates vs. 95% coaggregates) does not totally reflect the biological phenomenon observed in patients with ACS. Nevertheless, the experimental findings clearly demonstrate that platelet/CD34+ cell coaggregates present with increased adhesive properties. In a similar manner, platelet microparticles enhance the vasoregenerative potential of angiogenic early outgrowth cells after vascular injury in vivo.43 In line with our present findings, platelet interaction with human CD34+ cells induces kinase-insert domain-containing receptor (KDR) translocation from an endosomal compartment to the cell-surface within 15 min supporting vascular repair.44 Therefore, platelet interaction with circulating progenitor cells in patients with acute myocardial infarction may critically regulate the CD34+ or CD34+/CD133+ cell-mediated cardiovascular effects facilitating their adhesion at sites of vascular injury.

Activated platelets release potent chemokines including SDF-1, the major chemoattractant for CXCR4-positive progenitor cells.6,8 Platelet-derived SDF-1 promotes chemotaxis, adhesion, and migration of CD34+ progenitor cells towards vascular and myocardial injury.7,11,45 Further, platelets stimulate differentiation of CD34+ progenitor cells into mature endothelial cells.7,13 In humans, platelet activation and release of platelet-derived SDF-1 correlate with the number of circulating CD34+ progenitor cells in patients with ACS12 are associated with improved left ventricular function after acute myocardial infarction.15 At the present study, we report a positive correlation between platelet activation and SDF-1 expression and platelet–progenitor cell coaggregates formation, which may reflect to the molecular mechanisms involved in platelet binding onto the surface of CD34+ cells. For instance, platelet-derived SDF-1 and P-selectin may bind to CXCR4 and PSGL-1, respectively, on the surface of circulating CD34+ cells, as previously reported in vitro.911

In conclusion, our present data indicate that platelet interaction with mobilized progenitor cells is increased in patients with STEMI and that this interaction may be a novel mechanism of domiciliation of CD34+ progenitor cells to the injured microvasculature after acute myocardial infarction. The functional relevance of coaggregates formation between CD34+ cells and platelets is only partially addressed in our pilot prospective study and further larger prospective studies are essential to adequately address this important point. Moreover, further studies are needed to reproduce our novel findings and evaluate the platelet interaction with other subpopulations of CD34+ cells, including CD34+/KDR+/CD45dim endothelial progenitor cells, which are reported to be the best compromise in terms of sensitivity, specificity and reliability to quantify EPCs in the clinical setting.46 Identification of the underlying molecular mechanisms of platelet–progenitor cell interaction may open new therapeutic strategies for cardiovascular medicine.22,47

Funding

The study was supported by the Klinische Forschergruppe KFO 274 ‘Platelets- Molecular Mechanisms and Translational Implications’. The study was supported in part by the Tuebingen Platelet Investigative Consortium (TuePIC) and by the German Cardiac Society to K.S.

Conflict of interest: none declared.

Footnotes

  • Present address: Department of Cardiology, J.W. Goethe University Frankfurt am Main, Frankfurt am Main, Germany.

References

View Abstract