SEARCH

SEARCH BY CITATION

Keywords:

  • endothelial cells;
  • rare events;
  • CEC;
  • endothelial progenitor

Abstract

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

The finding of angiogenic and vasculogenic cells in the peripheral circulation may have profound effects on the course of a variety of diseases ranging from cancer to cardiovascular disease. These cells are ascribed to be endothelial in nature and are generally referred to as circulating endothelial cells if mature or as endothelial progenitor cells if immature. Different approaches have been used to detect these cells, including in vitro culture, magnetic bead isolation, and flow cytometry. We review flow cytometric methods for the detection and enumeration of these cells and provide technical suggestions to promote the accurate enumeration of circulating endothelial cells and endothelial progenitor cells. Published 2005 Wiley-Liss, Inc.

Angiogenesis and neovascularization are increasingly recognized as having an important role in a wide array of diseases. In diseases such as cancer, sickle cell disease, vasculitides, and pulmonary hypertension, these processes are undesired and may be associated with a poor prognosis. Angiogenesis has shown to be crucial for tumor growth, and novel anti-angiogenic therapies are currently being tested as means of slowing or preventing growth of tumors (1–3). In other diseases, such as coronary artery disease, neovascularization (or revascularization) is a desired event to restore blood flow that putatively correlates with patient recovery or improved cardiac function (4–6). Recent studies in animal models and humans have provided data suggesting that cells contributing to angiogenesis and vascular remodeling may be found in the peripheral circulation. These data indicate that mature and immature endothelial cells are present in the peripheral blood of normal individuals and in patients with different diseases. It is important to determine whether the numbers of angiogenic cells in the circulation change in different disease states, whether these changes correlate with prognosis, and whether altering the number of circulating angiogenic cells might affect disease progression. To reliably address these issues, techniques must be developed that provide accurate and reproducible enumeration of circulating endothelial cells (CECs) or endothelial progenitor cells (EPCs).

CECs and EPCs are extremely rare events in normal peripheral blood, representing somewhere between 0.01% and 0.0001% of peripheral mononuclear cells. Different cell surface markers have been used to detect these cells and consensus immunophenotypes have not yet been established. Therefore, the reliable enumeration of CECs and EPCs remains a technical challenge. This review examines the evidence for circulating angiogenic cells, the potential significance of these cells in certain diseases, and the approaches used to detect and quantify these cells.

BACKGROUND

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

The interest in circulating angiogenic cells dates back at least three decades, with several publications in the early 1970s describing the existence of CECs in various diseases (7–9). Most early studies of CECs relied solely on morphologic identification of cell type. Immunofluorescence staining for von Willebrand factor was another early method used for identifying endothelial cells. Although these methods could be used to identify CECs, the rare frequency of these cells made these techniques too cumbersome and insensitive for widespread application. Flow cytometry is clearly a method well suited for detection and quantitation of CECs, but the development of cytometric assays was constrained by the lack of reasonably specific monoclonal antibodies for this task. The use of flow cytometry to detect CECs became practical with the introduction of the S-endo-1 monoclonal antibody (10). S-endo-1, also known as P1H12 and CD146, is a key marker in the identification of endothelial cells and is particularly useful when used in combination with other monoclonal antibodies.

The interest in EPCs as distinct angiogenic cells from CECs occurred later and arose in part from the observations that embryonic endothelial progenitors derive endothelial and hematopoietic cells (11, 12). However, as discussed in more detail below, the identification of EPCs is complicated by the presence of CD34 on mature endothelial cells and other hematopoietic stem cells (13, 14). Asahara et al. were the first to isolate putative EPCs from human peripheral blood (15). They demonstrated that CD34+ cells could differentiate into endothelial cells in vitro and, in isolation, most expressed vascular endothelial growth factor receptor 2 (VEGFR2). The discovery of CD133 (AC133), an antigen that identifies primitive stem cells, provided a means to delineate mature (CECs) from immature (EPCs) endothelial cell types that was not possible with high precision using CD34 alone (14, 16).

RARE EVENT ANALYSIS BY FLOW CYTOMETRY

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

CECs and circulating EPCs represent between 0.01% and 0.0001% of mononuclear cells. Accurate detection and enumeration of cells occurring at such low frequencies require procedures not routinely used when analyzing relatively common cells such as circulating CD4+ T helper cells. For accurate detection of rare events, background “noise” must be substantially less than the frequency of the sought events. Thus, for EPCs and CECs, the cytometric assay would need to be greater than 99.999% free of background noise.

Background noise is generally due to a variety of factors that yield fluorescent signals, or at least signals perceived by the cytometer to be fluorescent, in the absence of antibodies or fluorescent probes specific to the cells of interest. Several layers of controls are necessary to evaluate background and establish the veracity of specific staining in rare event analysis. Fluorochrome-matched isotype controls, currently out of favor for common assays such as CD4 enumeration, are fairly crucial in rare event analysis, where they provide a good estimate of nonspecific binding of antibodies to cells. Even with freshly drawn peripheral blood, it is not unusual to observe nonspecific binding of isotype controls representing from 0.1% to 0.5% of cells analyzed. In most clinical assays, this level of staining does not significantly affect reportable data. However, in staining for CECs and EPCs, this represents a background higher than the anticipated frequency of specific events.

Binding of isotype controls to cells may be the result of dead or dying cells, cells bearing Fc receptors, or several other factors. The methods most commonly used to minimize this type of nonspecific binding include (a) inclusion of a real-time viability stain (such as propidium iodide or 7-aminoactinomycin D [7-AAD]) in each tube, (b) preincubation of the cells with a “blocking” serum, and (c) use of a “dump” channel to exclude from analysis cells of known immunophenotypes that are not of interest. Although different laboratories use different blocking reagents, it is our experience that whole heat-inactivated sera from the same species used to create the reagent monoclonal antibodies (e.g., mouse serum if conjugated mouse anti-human CDx is to be used) are the most appropriate and efficient blockers. The dump channel is one of the fluorescent channels (FL1, FL2, etc.) that is set aside for exclusionary gating. A dump channel may contain more than one parameter or marker for exclusion. For example, in the analysis of EPCs, a dump channel may include markers for dead cells (7-AAD) and leukocytes (peridinin chlorophyll protein CD45, CD3, CD19, and CD33).

Another source of background in rare event analysis is “carryover” of cells in the cytometer from previous experiments. Before rare event analysis on any cytometer, extensive cleaning and washing is recommended to remove residual cells and particles. The extent and nature of this cleaning will differ with the make and model of cytometer.

In rare event analysis, large numbers of cells must be counted to obtain statistically meaningful numbers of rare cells. It is likely in some instances that 1 or 2 million cells will need to be analyzed. This creates huge listmode data files, whose precise sizes are also determined by the number of parameters collected, the resolution used, and the number of cells collected. Although file sizes are not nearly as limited by computing capability as in previous years, enormous files can be problematic for some computers and software programs. To decrease file size and increase ease of file management, rare events are often collected using live gating when events of no interest are not saved or only a portion of these events is saved.

DETECTION OF CECs

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

CECs were detected decades ago, less rigorously, by using morphology and rather nonspecific markers. The advent of more specific markers and multiparametric flow cytometry made direct assessment of CECs in the peripheral blood a feasible, albeit difficult, undertaking. Nonetheless, many investigators are currently detecting and quantifying CECs using other methods such as magnetic bead separation with CD146 antibodies (17–20). Studies using magnetic bead separation of CECs rarely provide the precise purity of the separated cells, a fact not surprising because CD146 is also expressed on activated T cells (Fig. 1). Thus, this CD146-based magnetic bead separation is best accompanied by a second step (generally fluorescence microscopy) to confirm that all separated cells are of endothelial origin and to provide a more precise quantitation than that achieved solely by counting of elutriated cells. Manual fluorescence microscopy alone, using antibodies against von Willebrand factor (21) or gp96 (22), has also been used to enumerate CECs in peripheral blood. Across different methodologies, data from published studies (17–21, 23–27) are varied in the number of CECs that are present in the peripheral blood of normal individuals (Table 1), although most investigations have detected 1 to 100 CECs/ml in normal peripheral blood (17, 19, 20, 26–28).

thumbnail image

Figure 1. Gating strategy for CECs. Upper panel shows gating region around mononuclear cells. Bottom panel shows CECs staining positive for CD146 and negative for CD3. Positivity for CD146 and CD3 represent activated T lymphocytes.

Download figure to PowerPoint

Table 1. Reported Levels of CECs in Normal Human Peripheral Blood*
StudyMethodMarkers/OtherReported range of CECs
  • *

    CECs, circulating endothelial cells; KDR, kinase insert domain receptor (VEGFR2); UEA-1, Ulex European Agglutin 1; vWF, von Willebrand factor.

George et al. (10)Flow cytometryS-endo-1 (CD146)>100/ml
Solovey et al. (27)Immunohistochemistry, light microscopyP1H12 (CD146)2.6 ± 1.6/ml
Mancuso et al. (25)Flow cytometryCD45, P1H12 (CD146), CD31, 34, 105, 1067,900/ml (“resting” CECs)
   1,200/ml (“activated” CECs)
Mutunga et al. (26)ImmunocytochemistryvWF1.9 ± 0.5/ml
  KDR0.7 ± 0.3/ml
Bull et al. (20)Magnetic bead separation, fluorescence microscopyP1H12 (CD146), vWF, KDR3.5 ± 1.3/ml
O'Sullivan (51)ImmunohistochemistryvWF71 ± 13/ml (fit subjects)
   41 ± 5/ml (unfit subjects)
Nakatani et al. (17)Magnetic bead separation, fluorescence microscopyP1H12 (CD146)3.2 ± 0.4/ml
Woywodt et al. (19)Magnetic bead separation, fluorescence microscopyCD146, 3, 8, 14, vWF, CD31, UEA-17 ± 5/ml
Del Papa et al. (28)Flow cytometryCD45, 34, P1H12 (CD146)77/ml

Flow cytometric enumeration of CECs is far from a standardized procedure and several approaches have been published (10, 25, 29). Different antigens suitable for flow cytometric detection are associated with endothelial cells (Table 2). Because expression of these antigens may be variable on endothelial cells and may involve other cell lines, multiparametric flow cytometry is best used to detect endothelial cells and discriminate them from cells with overlapping expression of antigens. For example, CD146 expression on activated T cells can be distinguished from CD146 on endothelial cells by costaining with CD45 or CD3 (or both). These antigens are present on T cells but not on CECs. CD34 expression on endothelial cells presents an interesting problem because this antigen is also found on hematopoietic stem cells. Because both are present as rare events in the peripheral circulation, staining for CD34 alone is insufficient to distinguish among hematopoietic stem cells, CECs, and EPCs. By simultaneously staining for CD31 and CD146, which are present on CECs but not on EPCs or hematopoietic stem cells, CD34+ CECs may be better identified. In addition, CD133 will help to identify EPCs because it is not present on CECs or any mature endothelial cell.

Table 2. Cell Surface Antigens Present on Endothelial Cells*
CD/antigen nameOther names/cloneExpression by other cells
  • *

    ICAM-1, intracellular cell adhesion molecule 1; KDR, kinase insert domain receptor (VEGFR2); PECAM-1, platelet endothelium cell adhesion molecule 1; VCAM-1, vascular cell adhesion molecule 1; VE, vascular endothelium; VEGFR2, vascular endothelial growth factor receptor 2.

CD31PECAM-1Platelets, monocytes, neutrophils, T cell subsets
CD34 Hematopoietic stem cells
CD62eE-selectinActivated skin/synovial endothelium
CD54ICAM-1Activated B and T lymphocytes, monocytes
CD105EndoglinActivated monocytes, tissue macrophages, erythroid marrow precursors
CD106VCAM-1Activated endothelium, stromal cells
CD141ThrombomodulinKeratinocytes, platelets, monocytes, neutrophils
CD144VE-cadherinEndothelium adherens junction
CD146P1H12, S-endo-1Activated T lymphocytes
CD202bTie-2Hematopoietic stem cells
VEGFR-2KDR, Flk-1Hematopoietic stem cells

Although there are various protocols for detection of CECs, consensus seems to be building toward the use of CD146 (P1H12, S-endo-1), CD45, and a viability stain such as propidium iodide or 7-AAD, perhaps in combination with additional endothelial markers such as CD31, as an appropriate means of identifying and enumerating CECs. Redundant endothelial markers increase the accuracy of endothelial cell detection, whereas exclusionary gating of CD45+ cells and dead cells by the viability stain decreases the incidence of false-positive cells due to staining of lymphocytes (for CD146) or nonspecific antibody binding by dead cells.

DETECTION OF ACTIVATED CECs

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

As with many other cells types, endothelial cells may be resting or activated. Functionally, activated endothelial cells are characterized by the production of nitric oxide, prostacyclin, and toxic oxygen radicals and by procoagulant activity and increased leukocyte adhesion and phagocytosis. In addition, stimulated cells secrete numerous chemokines and cytokines. Phenotypically, the description of activated endothelial cells is more diverse. Anti-nitrotyrosine antibodies, which detect “footprints” of reactive nitric oxide, have been used in immunohistochemistry to detect activated CECs (30). Activated endothelial cells also express increased adhesion molecules such as CD54 (intracellular adhesion molecule-1), CD62e (E-selectin), CD106 (vascular cell adhesion molecule-1), and markers of procoagulant activity such as CD142 (tissue factor). In one study, activated CECs were defined as those CECs that were positive for CD105 (endoglin) or CD106; resting and activated CECs were increased in patients with lymphoma and breast cancer compared with healthy controls and decreased to “normal” levels after therapy (25). The clinical significance of distinguishing between these subsets has yet to be determined. Because enumeration of CECs remains a difficult and nonstandardized test, attempts to subset these cells may be premature.

Analysis of rare events, such as CECs in the peripheral blood, is challenging. Subsetting these cells with activation markers is even more complex. Further, subset analysis of CECs requires large volumes of blood to ensure sufficient cells are acquired to achieve statistical significance. Thus, most studies to date have not focused on the issue of resting versus activated CECs. However, the ultimate clinical utility of CEC detection may rely on accurate detection of CEC subsets (e.g., activated vs. resting or live vs. dead) rather than a gross quantitation of all CECs.

ONGOING CLINICAL INVESTIGATIONS OF CECs

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

Using a variety of the methods described above, CECs have been recently studied in several systemic illnesses. In general, these studies are designed to correlate the number of CECs, or activated CECs, with vascular endothelial injury, disease progression, or response to therapy. Table 3 lists some recently published investigations. As might be expected, all the diseases studied involve vascular damage, neovascularization, or revascularization. In general, these studies found an increase in CECs with any type of vascular perturbation such as that seen in vasculitis, pulmonary hypertension, and septic shock (18, 20, 26). CEC enumeration has also been described as a marker of response to therapy in Mediterranean spotted fever, lymphoma, and breast cancer (23, 25). The quantitation of CECs is evolving as a novel marker to distinguish between quiescent and active disease states, in addition to various phenotypes of a given disorder as described in patients with sickle cell anemia, thalassemia, Kawasaki's disease, and various cancers (17, 21, 24, 27). In addition, the characterization of CECs as resting or activated may also serve as a surrogate biological marker in establishing the stage of disease (25, 28).

Table 3. Recent Clinical Studies of CECs*
DiseaseReferenceResults
  • *

    ANCA, antibodies to neutrophil cytoplasmic antigens; CECs, circulating endothelial cells.

Mediterranean spotted feverGeorge et al. (23)Increased in patients with severe or malignant Mediterranean spotted fever, decreased during treatment and recovery
Sickle cell anemiaSolovey et al. (27)Increased in patients with acute crisis (P < 0.001) and in patients without acute crisis (P = 0.002 vs. normals and P = 0.019 vs. patients with acute events)
Septic shockMutunga et al. (26)Increased in patients with sepsis and septic shock (P < 0.0001)
Lymphoma/breast cancerMancuso et al. (25)Increased in patients with newly diagnosed breast cancer and lymphoma (P < 0.0008 for resting and activated CECs); resting and activated CECs were similar to those in controls after complete remission after chemotherapy
ThalassemiaButthep et al. (21)Increased in thalassemic patients; numbers in nondisease forms were lower than in disease forms but higher than normal
ANCA-associated small vessel vasculitisWoywodt et al. (18)Increased CECs in patients with active vasculitis (P < 0.0001)
Kawasaki's diseaseNakatani et al. (17)Increased in Kawasaki's disease (P < 0.05) and in patients with known coronary artery lesions than in those without (P < 0.05)
Pulmonary hypertensionBull et al. (20)Increased in patients with primary pulmonary hypertension and those with secondary pulmonary hypertension (P < 0.001)
Renal transplantationWoywodt et al. (19)Increased in renal transplant recipients: acute vascular rejection (P < 0.02), all transplant patients regardless of biopsy results (P < 0.001)
CancerBeerepoot et al. (24)Increased in cancer patients (various types) with progressive disease (P < 0.0001); patients with stable disease had CEC numbers similar to those of controls (P = 0.69)

DETECTION OF EPCs

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

Detection and quantitation of circulating EPCs are complex tasks from technical and biological viewpoints. Not only must EPCs be demonstrated to have angiogenic potential, they must also be shown to be distinct from other angiogenic cells such as mature CECs. The extreme rarity of these cells and the lack of EPC-specific marker present technical challenges in these assays.

EPCs in peripheral blood were first described by Asahara et al. (15) in 1997. In this study, CD34 magnetic bead separation was used to obtain an enriched population of CD34+ stem cells. These cells were demonstrated to contain cells that, after culture in vitro, displayed morphology, surface markers, and immunohistochemical staining associated with endothelial cells. Although this work was seminal in demonstrating the existence of EPCs in peripheral blood, no effort was made to quantitate these cells. Because CD34 may stain CECs and EPCs and both may have angiogenic potential, delineation of these two populations was not possible in this study. The observation that SP cells (hematopoietic stem cells identified by unique staining with Hoechst 33342 dye) can give rise to endothelial cells strongly supports the hypothesis that hematopoietic cells and endothelial cells may develop from a common progenitor (31).

Attempts to quantitate EPCs began to appear in the literature within a few years after the initial description of these cells (32). These efforts centered on the use of antibodies to VEGFR2 (such as KDR) in combination with CD34 to identify EPCs. Because neither is specific for EPCs alone or together, it is unclear whether the data accurately reflect the number of true EPCs in the circulation. The quantitative nature of these assays is also complicated by the fact that some investigators culture mononuclear cells on fibronectin-coated plates before surface marker analysis. It is uncertain how this pre-enrichment step affects EPC yield and enumeration. Thus, precise quantitation by this method is problematic.

The advent of CD133 (AC133) for use in flow cytometry provided a means for detecting primitive stem cells in the circulation without the use of CD34. Gehling et al. demonstrated that purified CD133 positive stem cells have the capacity to differentiate into endothelial cells (33). Recent studies have demonstrated that CD133+ and VEGFR2+ cells in the circulation have functional properties of EPCs (14, 33, 34). Expression of CD34 on these cells is variable. CD45 expression on these cells has been reported by various groups to be positive or negative (15, 34, 35). Dim expression of CD45 by these cells is the likely cause of this confusion; hence, the use of CD45 as a gating reagent for EPCs is not recommended. An exclusionary gate consisting of CD19, CD3, and CD33 may be used in lieu of CD45 (Fig. 2).

thumbnail image

Figure 2. Gating strategy for EPCs. A broad gate is set around mononuclear cells. Top: A dump channel consisting of allophycocyanin CD3, CD19, and CD33 is used to exclude these cells from further analysis. Bottom: Mononuclear cells not in the dump channel are then examined for coexpression of CD133 and VEGFR2 (KDR).

Download figure to PowerPoint

As an alternative to flow cytometric detection and quantitation of EPCs, a multistep manual technique has been used (36). This approach relies on density gradient separation of mononuclear cells followed by culture on fibronectin. EPCs are then enumerated by manually counting cells that are doubly stained with Di-LDL, and Ulex europaeus I lectin. The interlaboratory reproducibility of this method may be suboptimal because this approach involves multiple steps and subjective evaluation of staining. This approach is not recommended.

Several reports have indicated that at least some EPCs have monocytoid features or are derived from monocytes (37, 38). Although these observations are divergent from the main body of literature pertaining to EPCs, they nonetheless appear to be reproducible. The relation between monocytes and EPCs is complex; there are clearly EPCs, which are nonmonocytic, but there also appears to be monocytes that have functional characteristics of CEC and EPC. Investigators have recently reviewed the overlapping phenotypes among cells with angiogenic potential, including monocytoid cells (39). Two distinct groups of EPCs seem to have been described in various studies: one with monocytoid features and the other expressing CD133 and VEGFR2. There may be several cell types that can differentiate, or transdifferentiate, into endothelial cells (or endothelial-like cells), perhaps with slightly different functions in vivo or in response to different stimuli. Another recent study demonstrated that the vast majority of EPCs express CD14 and originate from CD14+ peripheral blood mononuclear cells (37). It is likely that the term “endothelial progenitor cell” is used too broadly and better distinction must be made between stem cells with angiogenic potential and monocytoid cells with similar potential.

ONGOING CLINICAL INVESTIGATIONS OF EPCs

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

The majority of current clinical EPC research focuses on the role of these cells in cardiovascular disease. Recent evidence has suggested that EPCs derived from bone marrow and blood are mobilized to sites of tissue ischemia or infarction and vascular injury to assist in the repair process (40–42). Granulocyte colony-stimulating factor is being used to mobilize and increase EPCs in circulation (43). Recently, therapy with granulocyte colony-stimulating factor and intracoronary infusion of peripheral blood stem cells was shown to increase cardiac function in patients with myocardial infarction (44). Thus, efforts to augment mobilization of EPCs and promote neovascularization at sites of vascular injury have become one of the foremost areas of research in clinical therapy for ischemic heart disease. Pharmacologic agents such as statins and erythropoietin have been demonstrated to enhance EPC mobilization and are promising therapies from this standpoint in cardiovascular disease (36, 45). Another potential novel therapeutic agent is Ginkgo biloba extract, which has been shown to promote EPC proliferative, migratory, adhesive, and in vitro vasculogenesis capacity (46). Recent studies have also shown that EPCs can be successfully transplanted in patients with myocardial ischemia and provide potential benefit (47, 48). Gene therapy combined with EPCs is currently being explored as a therapeutic option in various diseases such as pulmonary hypertension and hemophilia A (49, 50).

CONCLUSIONS AND RECOMMENDATIONS

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED

Identification and enumeration of angiogenic cells present in the circulation remains a difficult and nonstandardized method. The lack of markers truly specific for endothelial cells dictates that marker combinations must be used to best identify CECs and EPCs. However, even this concept is not universally accepted because many studies still use one-parameter magnetic bead selection to quantitate these cells. It is little wonder that there is such a wide variety in the numbers of circulating CECs and EPCs reported in the literature.

Thus, the question is: What is the optimal method for quantitation of CECs and/or EPCs? Unfortunately, the probable truth is that the optimal method has yet to be determined. Many recent studies have consisted of anecdotal observations, without efforts to rigorously correlate the immunophenotype of these cells with cell function. Without consensus “target values” (i.e., number of CECs or EPCs in normal subjects) to hit when establishing these assays, how would one know if the assay is properly working? The ultimate answer is to do the correlative functional or genomic studies on cells identified to establish the veracity of the immunophenotyping.

There are general recommendations that can be made for performing these rare event assays, even if precise consensus protocols cannot be agreed upon.

  • 1
    Thoroughly clean the cytometer before data acquisition. Even trace amounts of cellular debris or, worse yet, residual cells from a previous tube have the potential to contaminate the sample of interest. A contamination rate in the neighborhood of 1 per 100,000 events may be a substantial problem. Long cleaning (≥30 min) with sequential tubes of different cleaning solutions may be desirable before data acquisition of specimens for CEC or EPC enumeration.
  • 2
    The use of blocking serum to inhibit nonspecific binding or specific binding via Fc receptors is highly recommended. This generally consists of preincubating the specimen to be stained with serum (we use mouse serum because we stain with mouse antibodies) for at least 15 min before staining. Any decrease in nonspecific staining gained by this method is extremely useful.
  • 3
    Use of a real-time viability stain such as propidium iodide or 7-AAD is crucial for these studies. Dead cells can be a major source of nonspecific staining by monoclonal antibodies. Identifying the dead cells and excluding them from analysis by gating will largely eliminate this problem and improve the resolution of the assay.
  • 4
    The establishment of a dump channel to exclude cells not of interest from analysis is extremely useful. As Sherlock Holmes stated: Eliminate all other factors, and the one which remains must be the truth. Thus, if cells of known, but undesired, immunophenotype are eliminated from analysis, cells of less definitive, but desired, immunophenotype should become more prominent and easier to detect.
  • 5
    Collection of a large number of events is necessary to identify adequate numbers of the rare event population. In CEC and EPC assays, at least 500,000 to 1 million listmode events should be collected. Larger numbers are even more preferable, and live gating can be use if it has been established that all desired cells fall within the gate.

For detecting and enumerating CECs, a multicolor approach is key because no markers are entirely specific for these cells. The general recommendations listed above include a dump channel and a viability stain that should be used in immunophenotyping CECs. For the former, the use of CD3, CD20, CD33 (or CD15), CD64, and CD16 in combination are useful. CD146 and CD31 are useful as endothelial cell markers separately or in combination. CD146 (clone P1H12) is the most widely used for this purpose. Reasonable values for normal individuals seem to be in the range of 1 to 100 CECs/ml.

Protocols for immunophenotyping of EPCs should include the recommendations listed above. The use of CD34 in these assays is probably not necessary and in some instances may actually be confusing. CD133 and antibodies to VEGFR2 are crucial and must be included.

Given the current high level of interest in circulating angiogenic cells, it is likely that consensus protocols will start to emerge in the next few years. Whether these involve the development of new markers for these cells or finding optimal combinations of existing markers is uncertain. Either way, the rigors of rare event analysis will require attention to the recommendations listed above. Because CECs and EPCs can be identified only through the use of multiple markers and because both are extremely scarce in the peripheral circulation, flow cytometry will likely become the preferred method of identifying and enumerating these cells if reproducible and consistent results can be obtained.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. BACKGROUND
  4. RARE EVENT ANALYSIS BY FLOW CYTOMETRY
  5. DETECTION OF CECs
  6. DETECTION OF ACTIVATED CECs
  7. ONGOING CLINICAL INVESTIGATIONS OF CECs
  8. DETECTION OF EPCs
  9. ONGOING CLINICAL INVESTIGATIONS OF EPCs
  10. CONCLUSIONS AND RECOMMENDATIONS
  11. LITERATURE CITED