Endothelial progenitor cell homing in human myocardium in patients with coronary artery disease

Endothelial progenitor cells (EPCs) are mobilized from bone marrow into peripheral blood, contributing to the revascularization of ischemic areas, to endothelial repair and to the physiological maintenance of vascularization. EPC mobilization and homing have been primarily linked to ischemia and inflammation presence [1]. EPCs are directly correlated with endothelial function and inversely correlated with cardiovascular risk factors and atherosclerosis progression [2]. EPC levels have also been correlated to prognosis evaluation in cardiovascular disease [3]. Regarding EPC correlation with coronary artery disease (CAD) presence and severity, an inverse relationship with CAD severity, independent of traditional risk factors, was demonstrated by EPC colony counting [4], while a high number of EPCs associated with CAD and correlated with stenosis severity were shown by flow cytometry [5]. Recently a new protocol, adapted from the standardized ISHAGE protocol for hematopoietic stem cells, has been developed for EPC level evaluation to enable comparison of clinical and laboratory data [6]. While the presence of circulating EPCs has been widely evaluated in different diseases, few studies tried to evaluate the presence of EPCs in human vital myocardium. The aim of our study was to investigate EPC levels both in peripheral blood and in myocardium in the same patients at the same time, evaluating the correlation with CAD presence. In both samples we quantified CD34+KDR+ cells, a phenotype that has been frequently used to define EPCs in clinical cardiovascular studies [7]. 36 consecutive patients admitted to the Cardiac, Thoracic and Vascular Department either for valve replacement surgery or ascending aorta substitution (n = 14, Group A) or for coronary artery bypass grafting surgery (n = 22, Group B) were enrolled for the study. Exclusion criteria were recent (< 2 weeks) acute myocardial infarction, history of renal, hepatic, hematologic-coagulative disorders, acute–chronic inflammatory diseases, malignancies, recent infections, infectious or autoimmune diseases, or major bleeding requiring blood transfusion. All patients were evaluated by carotid and peripheral artery Eco-color Doppler to exclude presence of poly-distrectual atherosclerosis. All patients received therapy according to the current AHA Guidelines and gave their written informed consent to be enrolled into the study protocol. Group A patients (4 males, 10 females, age 75.1 ± 7.0 years) had non-ischemic heart disease with no evidence of CAD presence, while Group B patients (13 males, 9 females, age 76.2 ± 4.5 years) comprised mono- (9 out of 22), bi- (7 out of 22), or tri- (6 out of 22) diseased coronary vessels (CAD presence, evaluated as a stenosis > 75%). EPC (CD34+KDR+) levels were assessed before the intervention by flow cytometry on whole blood (circulating EPCs) and by immunohistochemistry on a right atrial appendage segment collected during cardioplegia induction (tissue EPCs). For circulating EPCs, whole blood was incubated with monoclonal antibodies against CD45, KDR, and CD34 (all from BD Pharmingen, San Jose, CA, US). The ISHAGE sequential strategy was used and CD45dimCD34+ cells were quantified for KDR. A minimum of 100 CD34+ events were collected. Results were expressed as cells/ml of blood. For tissue EPCs, tissues were fixed in formalin and embedded in paraffin. 3-μm sections were processed and incubated with monoclonal antibodies (Anti-CD34, clone QBEnd10, Ventana Medical Systems/Roche, Basel, CH, and anti-KDR, clone 55B11, Cell Signaling Technology Inc., Danvers, MA, US) at the manufacturer's recommended dilutions. After incubation with opportune enzymes and chromogen substrates (AEC, red, with HRP polymer anti-mouse IgG, for CD34; BCIP/NBT, purple/blue, with AP polymer anti-rabbit IgG for KDR, Polink DS-MR Hu B1 kit, Golden Bridge International, WA, US), EPCs were quantified by counting double positive cells. The normality of distributions of EPC numbers was verified using a one-sample Kolmogorov–Smirnov test. EPC levels were then compared with Student's t-test corrected with Fisher's exact probability test. A p value < 0.05 was considered statistically significant. All data are presented as mean ± standard deviation. In myocardial tissue, EPCs were primarily located inside the endothelium or the interstitium at epicardiac level (Fig. 1). As shown in Fig. 2, a significant increase of tissue EPCs (p < 0.001), accompanied by a significant reduction of circulating EPCs (p < 0.01) was observed in Group B patients, characterized by CAD presence, as compared to Group A patients (tissue EPCs: Group A 0.218 ± 0.052 vs. Group B 0.533 ± 0.211 EPCs/mm2; circulating EPCs: Group A 87.5 ± 16.6 vs. Group B 57.4 ± 6.9 EPCs/ml). Our data show an opposite effect of CAD presence on circulating and tissue EPCs. The presence of CAD disease and the consequent chronic ischemia could represent a trigger to increase EPC recruitment through mobilization from bone marrow and homing in myocardium, supporting the hypothesis of EPC involvement in the reparative mechanisms of ischemic myocardium. A recent study observed EPC decrease both in the circulation and in the graft of heart transplanted patients with microvasculopathy [8]. However, biopsy specimens and blood samples for EPC counts were not simultaneous. Further studies in a larger population of patients are required to support our hypothesis as well as to prospectively define the importance of increased levels of EPCs in myocardiu