Flow cytometric detection of endothelial progenitor cells (EPC) in acute coronary syndrome

Introduction

It has been believed that vascular endothelial cells, damaged for a while or lost, could be regenerated only by adjacent endothelial cells [1]. However, this belief has changed with the identification of endothelial progenitor cells (EPC) in the peripheral circulation, as well as by demonstrating their ability to differentiate into mature endothelial cells and exhibiting their role in vascular repair [2], [3].

Endothelial progenitor cells (EPCs) are bone marrow-derived cells that can differentiate into mature endothelial cells [4]. They mobilize during tissue ischemia and vascular injury and pass into circulation from the bone marrow (BM) as hemangioblasts and enable the reintegration of damaged endothelium [5]. Hypoxia, hematopoietic cytokines and granulocyte-macrophage colony-stimulating factor (GM-GSF) induce EPC release [6], [7], [8]. It is known that EPCs are at a place between stem cells and endothelial cells in terms of maturation [8]. EPCs are potential biomarkers of neovascularization, cardiac regeneration, vessel repair and aid in tumor growth, or in monitoring response to treatment [7]. There is correlation between decreased circulating EPC count and decreased endothelial repair capacity, which may lead to the development of cardiovascular disease [8], [9]. Cardiovascular risk factors such as age, diabetes mellitus (DM), hyperlipidemia and smoking unfavorably influence the number and functions of circulating EPCs [10], [11], [12], [13]. The importance of this correlation in the development of atherosclerosis is also underlined. It has been demonstrated that EPC count is negatively correlated with endothelial dysfunction and cardiovascular events [14]. Today, it is being emphasized that EPCs can be considered as an independent risk factor for cardiovascular diseases [8], [11], [12].

Flow cytometry (FCM) is accepted as the gold standard for quantification of these cells, which are less in number in the circulation [13]. The advantages of FCM is the ability to perform rapid, automated and repeatable multiparameter analysis at the level of single cell [8], [13], [14]. The flow cytometry protocol is based on a mononuclear cell analysis for size, nuclear complexity and binding of specific antibodies conjugated to given fluorochromes [8]. Minimal antigenic profile in order to define the EPCs should include at least one stemness/immaturity (CD34 and/or CD133) marker and at least one vascular endothelial growth factor receptor 2 (VEGFR-2) marker [8], [13]. Its immunophenotype overlaps with hematopoietic stem cell (HSC) and circulating endothelial cell (CEC) [13], [15]. Numerous studies have been conducted since 2000s, but monoclonal antibody (moAb) combination and study procedure, which would accurately identify immunophenotype of EPCs, has not been standardized yet [16], [17]. The aim of the present study is to investigate moAb combination that will accurately determine the circulating EPC count using FCM, as well as to investigate whether there is a difference between EPC counts of the patients with ST-elevation myocardial infarction (STEMI) and the patients with unstable angina pectoris (USAP), who have been admitted to the coronary intensive care unit for acute coronary syndrome (ACS).

Materials and Methods

Flow cytometry

A cocktail of fluorochrome-labeled moAb is used to identify EPCs, allowing highly specific definition of EPCs. Peripheral blood was evacuated into a 4-mL vacutainer (BD Vacutainer^®, CA, USA) tube as an anticoagulant and processed for 2 h after collection. Nonspecific antibody binding was blocked using 20 μL Fc-receptor (FcR) blocking reagent and 200 μL of mouse serum (Sigma-Aldrich, St. Louis, MO, USA) for 20 min at room temperature before staining with conjugated antibodies. Five-color staining was performed for quantification of EPCs could be done accurately. MoAbs used in the panel included CD45 PC7 (Beckman Coulter (BC), CA, USA), CD31PE (BC, CA, USA), CD34 (BC, CA, USA), ECD, CD309 (VEGFR-2/KDR) APC (Miltenyi Biotec, Germany), and nuclear staining syto16 FITC (Dako, Germany). [Anticoagulated venous blood sample was dispensed in 200 μL aliquots into two polypropylene test-tubes (BD, USA), which were 12×75 mm in size.] After mixing gently, the tubes were incubated with suitable fluorochrome-conjugated MoAbs at a concentration recommended by the manufacturer for 15 min at room temperature in the dark Stained whole blood samples were subjected to red blood cell lysis with 2 mL of lysing solution (Optilyse, BC, USA), and then shaken and incubated for 15 min at room temperature in the darkness. Prepared samples were stored at room temperature in the dark and then analyzed within 1 h. Nucleated cells derived from whole blood specimens were evaluated using a FC 500 (BC, USA) device with identical set up parameters between samples. A total of 1×10⁶ cells were counted for each sample. Results of the analyses were considered informative when adequate number of cells (approximately 1000 events) was collected in the EPCs enumeration gates. FCM data was analyzed using BC Cell Quest software program. Gating strategies and whole blood analysis were as following: non-viable cells, platelets, debris and non-specific binding in the CD45/ (side scatter) SSC dot plot were excluded from the analysis by syto16^® nuclear staining and isotype control and consecutive gating. Syto16^®+, CD31+ nucleated cells were retrieved from the CD45 dimmer/negative region. EPCs with CD34+ KDR+ immunophenotype were determined among these cells (Figure 1). The absolute EPCs count was derived from absolute white blood cell count obtained in a hematology analyzer (LH 750, BC, USA), and the percentage of EPCs was determined by FCM using the following formula: percentage of EPCs×white blood cell (WBC) count/100 [18]. Before each run, the flow cytometer was cleaned according to the manufacturer’s instructions. Passage between the samples was prevented using FC500 device. A range of internal quality assurance procedures including daily compensation (Flow-set, BC, USA) of optical alignment (Flow-Check, BC, USA) of the flow cytometer and the use of fluidic stability was applied. We The accuracy and precision of cell count were tested using international quality controls purchased from United Kingdom National External Quality Assessment Scheme (UK NEQASLI, Sheffield, UK) (z score range –2.0 to 2.0)

Figure 1:

EPC detection by flow cytometry, gating strategy.

(A) Gate1: FSC vs. SSC to exclude debris. (B) Gate2: SSC vs. CD45 to exclude hematopoietic cells. (C) Gate3: Syto16 (DNA marker) vs. CD31 to exclude platelets (red). (D) CD34+KDR+ EPCs (blue).

Results

Demographic characteristics of 40 healthy controls and 50 acute coronary syndrome cases, which have been investigated in the present study, are demonstrated in Table 1. The mean age was 40.3±4.4 years in the healthy control group and 56.9±10.3 years in the ACS group. While circulating EPC count was 11.33 (7.89–15.25) cells/μL in the healthy control group, it was found to be 3.80 (0.70–7.65) cells/μL in the ACS group.

Regarding EPC count in the STEMI and USAP groups of ACS patients, it was found higher in the USAP group as compared to the STEMI group (6.89±6.25 vs. 3.82±2.75, p=0.011). The factors that might influence EPC count in each disease group are demonstrated in Table 2. Among these variables, hsCRP was significantly lower in the USAP group (p=0.011), while HDL was significantly lower in the STEMI group (p=0.035). The groups were similar in terms of other factors. No statistically significant difference was determined between the groups in terms of age, presence of DM or HT, smoking, serum low-density lipoprotein (LDL), and statin use. Fibrinogen and high-sensitive C-reactive protein (hsCRP) values as well were compared between the groups. The comparison revealed no difference between the groups in terms of fibrinogen (p=0.571), whereas hsCRP was found higher in the STEMI group as compared to the USAP group (p=0.011).

Discussion

In the present study, we used a method that identifies the EPCs by excluding HSC, CEC, thrombocyte and EMP-another nucleus-free cell-, which may be confused with EPCs, in one tube by means of gating from CD45-negative and dim+ cells by 5-color FCM and found EPC count significantly lower in the STEMI and USAP patients as compared to the healthy subjects (p<0.001). Comparison between two patient groups revealed lower EPC count in the STEMI group versus USAP group (p=0.011). A standardized method of FCM for identification and quantification of circulating EPCs has not been developed yet. Therefore, different studies report different results. Accurate quantification of EPC is of great importance for its usage in clinical practice. Pre-analytic factors, different moAbs, and different sample preparation, marking and analysis techniques also influence the outcomes. There is no standardized protocol or gating strategy, which is operator-dependent [14], [19], [20], [21], [22].

Not a single EPC-specific antigen has been defined until today. The antigen profile overlaps with CEC and HSC [8]. In the study panel, we used CD45, CD34, CD31 and VEGFR-2 (KDR) to distinguish EPCs from CEC and HSC. HSC expresses CD34 and sometimes weak KDR, but these cells are CD45-positive. However, EPCs are CD45 dim-positive or CD45-negative; therefore they can be easily distinguished. Some researchers exclude all leukocytes stained positive with CD45 in the EPC analysis [16], [17]; nevertheless, most of CD34+ cells are CD45-dim and the number of CD45-CD34+ and CD45-CD133+ cells is very low in the peripheral circulation [8]. In the present study, we performed the analysis by gating from CD45-negative and dim+ cells. CEC has KDR expression and sometimes weak CD34 expression and, through this aspect, may be confused with EPC. However, CEC contamination is so low in the CD34+KDR+ EPC population that could be ignored [13]. Some researchers emphasized that CD133, which is another stem cell antigen, enhances the specificity of EPCs as CD133 is not found on the surface of CEC [7]. CD133 is in vitro and in vivo downregulated during development of endothelial cell line. In addition, CD133+ EPCs are early EPCs and are 20 times lesser in the peripheral circulation and have also lower proliferation capacity than CD34+KDR+ late EPCs. They are poorly reliable for the quantification of EPCs [13], this is why we did not use CD133 in the present study. Fadini et al. [23] and Schmidt et al. demonstrated that clinical correlation of CD34+KDR+ EPCs is better than that of CD133+KDR+ and CD34+CD133+KDR+ cells and recommended it as the best EPC phenotype to be analyzed [13], [24]. Lanuti et al. determined no CD34+CD133+KDR+ cell in the peripheral circulation. Different from the other studies, we used syto16^® DNA stain as was recommended by Mancuso et al. and excluded endothelial micro particles (EMP) and platelets, which are nucleus-free cells found in the CD45-negative region and may show antigenic overlap with EPCs [22]. Subsequently, we performed the analysis by gating from Syto16+CD31+ cells (Figure 1). Again, in the present study, we exposed the flow cytometer to long clean before each acquisition to prevent potential contamination. We counted at least 1×10⁶ cells for each sample to enhance the sensitivity.

Although these two subgroups have not been compared in the studies conducted in ACS patients, individual evaluation of each group revealed increased EPC count in such patients as compared to healthy controls [25], [26], [27]. Contrary to these studies, we determined decreased EPC count in both patient groups as compared to the healthy controls. Healthy controls’ being at younger age and selected very carefully might have played a role in this result. Furthermore, the time of phlebotomy in ACS patients might have had a role. Earlier studies stated that EPC peaks on the 7th day of myocardial infarction [28], [29]. In the present study, we performed phlebotomy at the time of hospital visit before starting any treatment or procedure (the first 24 h at the latest). We found EPC count higher in the USAP group, which is one of the two subgroups of acute coronary syndrome, as compared to the STEMI group. This might have resulted from low number of atherosclerotic plaques in the USAP patients [25], [26], [27]. Among inflammatory markers, fibrinogen and high-sensitive C-reactive protein (hs-CRP) values were also compared between the subgroups of ACS for atherosclerosis, which is an inflammatory process. While there was no difference between the groups in terms of fibrinogen values, hs-CRP value was significantly higher in the STEMI group as compared to the USAP group. Hs-CRP is one of the important predictors of coronary artery disease and this may be explained by higher atherosclerotic burden and expectation of more inflammatory process in acute phase in STEMI patients as compared to USAP patients indicating that atherosclerotic burden is lower in the USAP group versus the other groups. This may explain why EPC count is higher in the USAP group.

It was demonstrated that cardiovascular risk factors such as HT, DM, hyperlipidemia and smoking decrease EPC count independently from the development of coronary artery disease [10], [11], [12], [21].

Comparing other independent variables that influence EPC count, no statistically significant difference was determined between the two groups in terms of age, HT, DM, smoking, statin use and serum LDL levels [26], [30].

The limitations of the present study are as following: magnetic beads could be used instead of Kröbling formula in the dual platform while calculating absolute EPC count. The number of healthy controls could be higher.

In the present study, primarily the subjects who have been selected in accordance with the exclusion criteria depending on the patients’ detailed anamnesis were enrolled in the healthy control group. The significance of the present study is that the analysis can be performed simply and without consuming too much time by excluding the platelet, debris, EMPs and HSC using a single tube. In the present study, we identified the circulating EPC count in healthy subjects and the patients with ACS using multicolor FCM. We investigated whether there is a difference between STEMI and USAP subgroups of ACS patients in terms of EPC count and found the EPC count lower in both disease groups as compared to the healthy controls. First the method must be standardized and then harmonization studies are required to use this parameter as a prognostic factor in cardiovascular diseases.