Strikingly Different Angiogenic Properties of Endothelial Progenitor Cell SubpopulationsInsights From a Novel Human Angiogenesis Assay
Daniel P. Sieveking, BSc*, ,
Andrew Buckle, PhD*,
David S. Celermajer, MBBS, PhD, DSc, and
Martin K.C. Ng, MBBS, PhD*,,,*
* Heart Research Institute, Sydney, Australia
Department of Medicine, University of Sydney, Sydney, Australia
Royal Prince Alfred Hospital, Sydney, Australia.
Manuscript received June 27, 2007;
revised manuscript received August 15, 2007,
accepted September 10, 2007.
* Reprint requests and correspondence: Dr. Martin K. C. Ng, Department of Cardiology, Royal Prince Alfred Hospital, Missenden Road, Camperdown, Sydney, NSW 2050, Australia. (Email: mkcng{at}med.usyd.edu.au).
 |
Abstract
|
|---|
Objectives: An endothelial cell (EC)-specific angiogenesis assay was developed to functionally characterize angiogenic properties of 2 distinct putative endothelial progenitor cells (EPCs): early EPCs and late outgrowth endothelial cells (OECs).
Background: Endothelial progenitor cells promote revascularization of ischemic tissue. However, the nature of different EPCs and their role in angiogenesis remains debated.
Methods: Tubulogenesis was assessed by immunohistochemistry in co-cultures of differentiated ECs (including human umbilical vein, coronary artery, and microvascular ECs) or non-ECs with monolayers of human fibroblasts (MRC5). Using adaptations of the co-culture assay, early EPCs and OECs, isolated from peripheral blood mononuclear cells, were assessed by 3-dimensional immunofluorescence microscopy for their capacity for: 1) independent tubulogenesis; 2) incorporation into pre-existing vascular networks; and 3) paracrine angiogenic effects using transwell cultures.
Results: Branched interconnecting EC-specific tubules formed with all differentiated ECs after 72 h. Proangiogenic and antiangiogenic agents modulated tubulogenesis appropriately (vascular endothelial growth factor 10 ng: +142 ± 13%, 1 µM anti-vascular endothelial growth factor: –44 ± 7% vs. control, p < 0.001). In contrast, early EPCs, along with nonendothelial cell types, failed to independently form tubules or incorporate into differentiated EC tubules. Nevertheless, early EPCs indirectly augmented tubulogenesis by differentiated ECs even when physically separated by transwells (+115 ± 4% vs. control; p < 0.001). By contrast, OECs independently formed tubules and incorporated into differentiated EC tubules but exerted no significant paracrine angiogenic effects.
Conclusions: A novel EC-specific tubulogenesis assay highlights strikingly different angiogenic properties of different EPCs: late OECs directly participate in tubulogenesis, whereas early EPCs augment angiogenesis in a paracrine fashion, with implications for optimizing cell therapies for neovascularization.
|
Abbreviations and Acronyms
| | ANOVA = analysis of variance | | Dil-AcLDL = 1,1'–dioctadecyl-3,3,3',3-tetramethyl-indocarbocyanine perchlorate labeled acetylated low-density lipoprotein | | EC = endothelial cell | | eNOS = endothelial nitric oxide synthase | | EPC = endothelial progenitor cell | | MNC = mononuclear cell | | OEC = late-outgrowth endothelial cell | | VEGF = vascular endothelial growth factor |
|
The endothelium plays a critical role in cardiovascular homeostasis, including the regulation of vascular growth and neovascularization in response to ischemia (1). Mature endothelial cells (ECs), however, have limited regenerative capacity. The description 10 years ago of putative endothelial progenitor cells (EPCs) that could mobilize from bone marrow to participate in neovascularization at sites of ischemia was enthusiastically greeted because of the potential they hold for cardiovascular regeneration (2). However, to date, pre-clinical and clinical studies evaluating the therapeutic potential of EPCs have reported conflicting results, generating controversy (3). In pre-clinical studies, contradictory findings have been reported with respect to the nature of EPCs and their role in vascular biology (4). Moreover, recent randomized clinical trials evaluating efficacy of delivery of heterogenous bone marrow cell preparations have yielded inconsistent data, with some studies reporting small but significant improvement in cardiac function (5), whereas others report largely neutral results (6,7).
Key issues within the field of EPC research include a lack of uniform cellular definitions, as well as inadequate functional characterization of the role of putative EPCs in angiogenesis (3). Typically, circulating EPCs are identified using flow cytometry by selecting cells expressing CD34, CD133, and/or vascular endothelial growth factor (VEGF) receptor 2 (2,8). However, hematopoietic stem cell populations also express these markers; therefore, flow cytometric EPC identification is affected by hematopoietic contamination. Alternatively, EPCs are identified by culture of peripheral blood-derived mononuclear cells in EC-specific media. Interestingly, culture of these mononuclear cells reveals 2 distinct populations of cells with angiogenic capabilities that have been classified according to the time at which they appear in culture: early EPCs and late-outgrowth endothelial cells (OECs) (2,9). Early EPCs, which appear in culture after 4 to 7 days, are similar to those originally described by Asahara et al. (2) and have been used for therapeutic studies (10,11). The OECs appear much later in culture, after 14 to 21 days, and form cobblestone-shaped colonies with high rates of proliferation (9). These 2 different populations of cells, despite being phenotypically and morphologically distinct, have nonetheless been classified as EPCs because they both express endothelial markers.
The specific contribution of putative EPCs to neovascularization remains debated. Early preclinical studies reported that EPCs contributed significantly (as much as 50%) to new vessel formation (11). However, others have reported that EPCs do not directly incorporate into new vessels but promote angiogenesis by indirect mechanisms (4). Furthermore, difficulties arise from the fact that current in vitro assays of angiogenesis, typically incorporating extracellular matrix protein gels such as Matrigel (BD Biosciences, Bedford, Massachusetts), are not adapted to accurately assess the role of putative endothelial progenitors, because Matrigel-based assays induce tubulogenesis in a wide variety of cells (e.g., kidney cells) that have no role in vascular network formation in vivo (12). We have therefore sought to develop an endothelial-cell-specific in vitro angiogenesis assay that would allow systematic functional characterization of the role of putative EPCs in vascular network formation, including their capacity for: 1) tubulogenesis de novo; 2) incorporation into pre-existing vascular networks; and 3) ability to augment angiogenesis in a paracrine fashion.
|
Methods
|
|---|
Adult peripheral blood samples.
Human peripheral blood samples were obtained from healthy volunteers by venepuncture and collected into ethylenediaminetetraacetic acid (EDTA) tubes. All samples were obtained with informed consent and in accordance with the Declaration of Helsinki.
Cell culture.
The human fetal lung fibroblasts (MRC5) were obtained from American Type Culture Collection and cultured per supplier’s instructions. Human umbilical ECs were freshly isolated from umbilical cords as described previously by us (13). Human coronary artery ECs (Cell Applications, Inc., San Diego, California), and human microvascular ECs (Lonza Walkersville Inc., Walkersville, Maryland) were obtained and grown in specialized medium per the manufacturer’s instructions as previously described (14).
EPCs.
The EPCs were isolated from adult peripheral blood and cultured in accordance with previously published methods (9). Briefly, mononuclear cells (MNCs) were isolated using Lymphoprep (Axis-Shield, Oslo, Norway), a density gradient separation, and seeded into 6-well tissue culture plates pre-coated with human fibronectin (BD Biosciences) at 1 x 107 cells/ml to 1.5 x 107 cells/ml in endothelial cell growth medium-2 (EGM-2) media (Lonza Walkersville Inc.). After 24-h culture, nonadherent cells were discarded. Cells were then cultured for 4 days to 7 days to obtain early EPCs (2). To obtain OECs, MNCs were plated out as above on tissue culture plates pre-coated with type I collagen I (BD Biosciences) and grown 14 days to 21 days as described previously (9). For each independent experiment, blood from a different volunteer was used, with early EPCs and OECs for each experiment being isolated from the same volunteer.
Immunophenotyping and uptake of acetylated low-density lipoprotein (AcLDL).
Isolated EPCs were immunophenotyped via fluorescence microscopy or flow cytometric analysis (Beckman Coulter, FC500, Fullerton, California) with phycoerythrin-conjugated antibodies against human CD31, CD34, CD146, CD14, CD45 (BD Pharmingen, San Diego, California), KDR (RnD Systems, Minneapolis, Minnesota). For isotype control for BD antibodies we used PE-conjugated IgG1,  (BD Pharmingen). For isotype control for KDR we used PE-conjugated IgG1 (RnD Systems). Cells were also assessed for the ability to ingest AcLDL (4 µg/ml) (Invitrogen, Carlsbad, California) and to bind fluorescein isothiocyanate–Ulex europaeus agglutinin lectin (10 µg/ml, Sigma, St. Louis, Missouri).
Co-culture angiogenesis assay.
Fibroblasts and ECs were harvested using trypsin/EDTA (0.25%, Thermo Trace, Milford, Massachusetts). Fibroblasts were seeded in tissue culture plates or chamber slides at a concentration of 1 x 105 cells/ml. Once confluent, suspensions of ECs were added to the monolayers at a concentration of 1 x 104 cells /ml. Co-cultures were then grown for at least 72 h in EGM-2 medium. The effects of proangiogenic and antiangiogenic stimuli were assessed by addition of 10 ng VEGF (Sigma), 1 µg/ml anti-VEGF (RnD), or 100 nM Suramin (Sigma), respectively.
EPC angiogenesis assay.
For assessment of EPCs, cells were harvested using trypsin/EDTA and added in differing concentrations to monolayers of fibroblasts either with or without differentiated ECs. The EPCs were added to fibroblast monolayers at a concentration of 1 x 104 cells/ml. Alternatively, EPCs were added to co-cultures of fibroblasts and differentiated ECs, as above, at concentrations of 1 x 104 or 5 x 104 cells/ml. When resolution of cellular relationships was sought, EPCs were pre-tagged with 1,1'–dioctadecyl-3,3,3',3-tetramethyl-indocarbocyanine perchlorate labeled AcLDL (Dil-AcLDL), as outlined above, or with 1 µM CellTracker 5-(((4-chloromethyl)benzoyl)amine) tetramethyl rhodamine (Molecular Probes, Carlsbad, California), a nontransferable fluorescent dye.
Transwell assay for paracrine angiogenic effects.
To further assess the potential paracrine effect of these cells, EPCs were added in differing concentrations to transwell chambers. Briefly, transwell chambers (Costar, Cambridge, Massachusetts) equipped with a 12-mm-diameter polyester membrane (pore size 0.4 µm) were used. The EPCs were plated in the top chamber, and fibroblast/EC co-cultures were established in the lower chamber. This system allowed cells to be physically separated from each other but enabled molecules to communicate freely.
Immunohistochemistry.
At the end of the assay, the multicellular co-cultures were fixed in ice-cold ethanol for 30 min. Co-cultures were then washed in phosphate-buffered saline and incubated with mouse anti-human CD31 or anti-von Willebrand Factor antibodies (BD Pharmingen) at 37°C for 30 to 45 min. Cells were then washed and incubated with horseradish-peroxidase-conjugated sheep antimouse antibody (Amersham, Piscataway, New Jersey) at 37°C for 30 min to 45 min. After washing, cells were stained using diaminobenzidine (Vectorlabs, Burlingame, California). Images were captured using a Nikon Diaphot-TMD (Nikon, Tokyo, Japan) inverted light microscope equipped with a phase-contrast condenser (Phase contrast-2 ELWD 0.3, Nikon) and a digital camera (D100, Nikon).
Fluorescence microscopy.
Cells were analyzed after immunofluorescent staining as described above. Cells were counterstained for 4 , 6-diamidino-2-phenylindole and mounted using VectaMount (Vectorlabs). Fluorescence was examined using a Zeiss AxioImager M1 (Carl Zeiss, Thornwood, New York). Images were acquired with the manufacturer’s software, and 3-dimensional (3D) deconvolution was applied to z-stacked images.
Image analysis.
Analysis on light microscopic images was performed using ImageJ (National Institutes of Health, Bethesda, Maryland) in accordance with previously published methods (15). Briefly, at least 15 random microscopic fields in each well (triplicate wells per treatment) were captured. Color images were then converted to 8-bit grayscale, then binarized and threshold-adjusted to obtain optimal contrast of tubules with the background. The tubule area was then calculated according to the number of pixels.
Statistical analysis.
Data are expressed as mean ± SEM. Groups were compared by a 1-way analysis of variance (ANOVA) with post-hoc analyses for pairwise comparisons (Newman-Keuls multiple comparison). Statistical significance was inferred at a 2-sided value of p < 0.05. GraphPad Prism version 4.00 (Graphpad Software, San Diego, California) for Windows was used for statistical analysis.
|
Results
|
|---|
A novel EC-specific human angiogenesis assay.
We investigated co-culture of human ECs, including human umbilical ECs, human coronary artery ECs, and human microvascular ECs, with pre-formed monolayers of human fibroblasts as a means of developing an in vitro angiogenesis assay specific for ECs. After 72 h, we observed the formation of branched, interconnecting networks of EC tubules (Fig. 1). Tubule formation was observable by 72 h and was maximal after 14 days. We observed branching vascular networks for all differentiated ECs tested. Immunostaining for CD31 and Ulex europaeus lectin revealed that only ECs and not fibroblasts formed tubes in this assay (Fig. 1B). Given the non-EC specificity of the Matrigel tubulogenesis assay, we assessed tubule formation in this co-culture assay in a variety of non-EC types including smooth muscle cells, hepatocytes, monocytes, and cervical cells. Unlike Matrigel-based assays, no tubulogenesis was observed in any co-cultures featuring non-ECs, indicating that the co-culture assay facilitates tubulogenesis only by cells capable of developing an EC phenotype (Table 1).
View larger version (72K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1 EC–Specific Human Angiogenesis Assay
(A) Light photomicrograph (x20 original magnification) of endothelial cells (ECs) forming tubular structures (arrows) when co-cultured with MRC5 fibroblasts. (B) Immunofluorescent micrograph (x20 original magnification) of EC tubules, stained for 1,1'–dioctadecyl-3,3,3',3-tetramethyl-indocarbocyanine perchlorate labeled acetylated low-density lipoprotein (Dil-AcLDL) (red), Ulex europaeus agglutinin lectin (UEA-1) (green), and counterstained with 4, 6-diamidino-2-phenylindole (DAPI) (blue). (C) Light photomicrograph (x10 original magnification) ECs stained for CD31 (brown) (arrows) failed to form tubules when cultured on smooth muscle cells.
|
|
Interestingly, EC tubulogenesis was not supported by fibroblast-conditioned media alone nor by separation of intercellular contact between ECs and fibroblasts using transwell chambers, indicating that direct EC-fibroblast contact is necessary for tubulogenesis in this model. When vascular smooth muscle cells were tested as an alternative feeder layer to the MRC5 fibroblasts, we also failed to observe EC tubulogenesis (Fig. 1C).
Responsiveness to proangiogenic and antiangiogenic stimuli and assay reproducibility.
Having shown endothelial specificity, we sought to assess the responsiveness of our tubulogenesis assay to angiogenic factors. Classical proangiogenic and antiangiogenic stimuli, namely VEGF, anti-VEGF, and suramin respectively, were added to our co-cultures. Addition of 10 ng/ml VEGF increased tubule formation (142 ± 13% vs. control as 100%, p < 0.0001 by ANOVA, p < 0.001 for VEGF vs. control), whereas addition of antiangiogenic stimuli 1 µM anti-VEGF mAb and 100 µM suramin decreased tubule formation (44 ± 7% and 45 ± 4% vs. control as 100%, respectively, p < 0.001 for both vs. control) (Fig. 2). These findings show that EC tubulogenesis in our model is responsive to conventional angiogenic stimuli.
View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2 Angiogenic Responsiveness of Angiogenesis Assay
(A) Representative binarized photomicrographs and (B) quantification of tubulogenesis in response to treatment. Data expressed as mean ± SEM. ***p < 0.001 versus control. VEGF = vascular endothelial growth factor.
|
|
To assess the reproducibility of the assay, blinded image analysis was performed on 3 independent triplicate experiments covering 400 photomicrographs. The blinded intraobserver and interobserver errors were 4.8% and 5.7%, respectively.
Role of EPC subpopulations in angiogenesis.
Having established an angiogenesis assay specific for differentiated ECs, we used this assay to systematically evaluate the angiogenic capacity of early EPCs and OECs. Consistent with previous studies, endothelial antigens including CD31, CD146, and VEGF receptor 2 were expressed by both early EPCs (34.5 ± 8%, 1 ± 0.5%, and 82.9 ± 4% respectively) and OECs (98.5 ± 0.4%, 97.1 ± 1%, and 42.4 ± 10% respectively). These 2 populations could be distinguished by CD14 and CD45 expression, with early EPCs showing high expression of these markers (97.3 ± 2% and 98.65 ± 0.9%, respectively) whereas OECs did not express either maker (0.75 ± 0.2% and 0.3 ± 0.2%, respectively).
Late-outgrowth ECs but not early EPCs form vascular networks de novo.
The de novo tubulogenic capacity of EPCs was determined by co-culture of either early EPCs or OECs with fibroblasts. When cultured with fibroblasts alone, early EPCs failed to form interconnecting vascular networks (Fig. 3A). Indeed, in quadruplicate experiments using cells from the blood of 5 independent donors, early EPCs failed to form vascular networks, even when culture conditions were extended to 14 days.
View larger version (38K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3 Late-OECs Not Early EPCs Form Tubules De Novo
Representative immunofluorescent micrographs of (A) early EPCs and (B) OECs were co-cultured with MRC5 fibroblasts and stained as in Figure 1B. EPC = endothelial progenitor cell; OEC = late-outgrowth endothelial cell.
|
|
Strikingly, in contrast to early EPCs, after 72 h we consistently observed the formation of branched interconnecting vascular networks by OECs (Fig. 3B). In similar fashion to differentiated ECs, tubulogenesis continued after 72 h and was maximal after 14 days.
Late-outgrowth ECs but not early EPCs incorporate into vascular networks.
Having shown differences in the capacity for independent tubulogenesis by different EPCs, we then assessed whether EPCs would incorporate into developing vascular networks formed by differentiated ECs. To do this, pre-tagged EPCs were added to co-cultures of differentiated ECs and fibroblasts. In quadruplicate experiments using cells from 5 independent donors, covering at least 500 random microscopic fields, we failed to observe incorporation of early EPCs into differentiated EC tubules (Fig. 4). Where we observed co-localization of EPCs with tubules, we sought to resolve intercellular relationships by use of 3D z-stacking microscopy (Figs. 4B and 4C). In all cases of early EPC co-localization, no incorporation into an endothelial tubular structure was observed. Representative immunofluorescent micrographs are shown that depict an early EPC in 1 focal plane (Fig. 4B), overlying a tubule structure formed by a differentiated EC in a different focal plane (Fig. 4C). These intercellular relationships are also confirmed by 3D rendered microscopic video (Online Video 1).
View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4 Early EPCs Do Not Incorporate Into Vascular Networks
(A) Early EPCs were pre-tagged with CellTracker 5-(((4-Chloromethyl)benzoyl)amine)tetramethyl rhodamine (red), co-cultured with differentiated ECs, then co-cultures were stained with UEA-1, fluorescein isothiocyanate (FITC) (green), and 4, 6-diamidino-2-phenylindole (blue). (B and C) Three-dimensional z-stacking micrograph showing a tagged early EPC and differentiated EC tubule in separate focal planes (white lines above and right of the images represent position in the z axis). Abbreviations as in Figures 1 and 3.
|
|
In contrast to early EPCs, we observed direct incorporation of OECs into vascular networks formed by ECs by 72 h (Figs. 5A and 5B). Three-dimensional z-stacking microscopy confirmed direct incorporation of OECs into vascular networks formed by differentiated ECs. Figure 5C shows a representative micrograph of an OEC incorporated into a network of surrounding ECs in the same focal plane. This is further confirmed using 3D rendered microscopic video (Online Video 2). Where late EPCs were not associated with differentiated EC networks, independent de novo network formation was observed.
View larger version (63K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5 OECs Incorporate Into Vascular Networks
Co-cultures were stained according to Figure 4. Pre-tagged OECs directly incorporated into developing vascular networks at (A) x20 magnification and (B) x40 magnification. (C) Three-dimensional z-stacking micrograph shows 2 tagged OECs directly incorporated into surrounding differentiated EC tubules in the same focal plane (white lines above and right of the images represent position in the z axis). Abbreviations as in Figures 1 and 3.
|
|
Early EPCs but not late-outgrowth ECs augment angiogenesis in a paracrine fashion.
Early EPCs failed to form tubules de novo or to incorporate into tubules formed by differentiated ECs. Despite this, addition of early EPCs to co-cultures of ECs and fibroblasts augmented tubulogenesis by differentiated ECs in a concentration-dependent manner, suggesting a paracrine effect (124 ± 4% and 155 ± 5% increase vs. control as 100% for EPCs at 1 x 104 cells/ml and 5 x 104 cells/ml respectively, p < 0.0001 by ANOVA, p < 0.001 for each concentration of early EPCs vs. control) (Fig. 6). To confirm a paracrine effect, a transwell culture technique was used, enabling physical separation of EPCs from ECs and fibroblast co-cultures (Fig. 7A). Even when placed in transwell chambers, early EPCs augmented angiogenesis in a concentration-dependent manner, consistent with a paracrine effect (103 ± 3% and 115 ± 4% increase vs. control as 100% for EPCs at 1 x 104 cells/ml and 5 x 104 cells/ml respectively, p = 0.01 by ANOVA) (Fig. 7B).
View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6 Early EPCs Augment Angiogenesis
Early endothelial progenitor cells (EPCs) augmented angiogenesis by differentiated endothelial cells (ECs) in a dose-dependent fashion. Data expressed as the mean ± SEM. ***p < 0.001 for each concentration of early EPCs versus control. CTRL = control.
|
|
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7 Early EPCs and Not OECs Augment Angiogenesis in a Paracrine Fashion
(A) A transwell apparatus was used to assess the paracrine effects of different EPCs. (B) Early EPCs, still exerted a dose-dependent proangiogenic effect when placed in transwells (TW) (p = 0.01 by analysis of variance). (C) The OECs did not exert proangiogenic effects when placed in transwells (p = 0.3 by analysis of variance). Data expressed as the mean ± SEM. *p < 0.05 versus control. CTRL = control; other abbreviations as in Figure 3.
|
|
In contrast to early EPCs, OECs failed to augment tubulogenesis when co-cultured using the transwell technique described above (107 ± 6% and 96 ± 6% vs. control as 100% for EPCs at 1 x 104 cells/ml and 5 x 104 cells/ml respectively, p = 0.3 by ANOVA) (Fig. 7C). These data suggest that OECs do not exert a significant paracrine effect on tubulogenesis by differentiated ECs.
|
Discussion
|
|---|
The present investigation reports a novel human angiogenesis assay that allows assessment of EC-specific tubulogenesis, thereby allowing systematic functional characterization of the role of different putative EPCs in angiogenesis. Using this assay, we report that different populations of EPCs have divergent angiogenic capabilities (Table 2). The salient findings of this study are that early EPCs do not form new vessels de novo, and do not incorporate into newly forming vessels but rather promote angiogenesis in a paracrine fashion. In contrast, we show that OECs facilitate angiogenesis by direct participation in new vessel formation but exert no significant paracrine angiogenic effects. Our results may shed light on divergent pre-clinical and clinical study results using different preparations of mixed cell populations. Moreover, our findings may have implications for refining and optimizing cell therapies.
The identity, origin, and function of EPCs remain debated. Currently, there exists a phenotypic rather than functional definition of an EPC, focusing on the expression of certain hematopoietic antigens including CD133 and CD34, and a raft of endothelial antigens including CD31, CD105, CD144, CD146, and VEGF receptors, as well as the uptake of Dil-AcLDL, the binding of certain plant lectins, and/or colony formation (2,8). A recent study of various bone marrow-derived cells showed EPC-specific expression of endothelial nitric oxide synthase (eNOS), suggesting that this may be a means by which to identify endothelial progenitors (16). However, numerous studies have highlighted significant overlap in the expression of the abovementioned antigens, including eNOS, by monocytes and macrophages (17–19). Therefore, it seems that assessment of cell surface antigens or eNOS expression is insufficient to identify a putative EPC. Some functional assessment is necessary.
The differentiated fate of any putative endothelial progenitor cell is necessarily an EC. Therefore, in the phenotypic characterization of EPCs, it is necessary to compare the functional behavior of a putative progenitor with that of a differentiated EC. The ECs are derived from hemangioblasts. They either form or incorporate into tubular structures through the processes of vasculogenesis or angiogenesis, respectively (20). Hence, it may be argued that de novo tubulogenesis and incorporation into established vascular networks are important functional criteria by which putative EPCs should be assessed.
Currently the most extensively used in vitro method for assessing tubulogenesis involves growing cells on Matrigel, which facilitates nonspecific tubulogenesis by many non-EC types, including kidney cells, epithelial cells, fibroblasts, monocytes, and macrophages (12). This nonspecificity makes the angiogenic relevance of any tubules observed questionable. Indeed conflicting data exist regarding the ability of different EPC populations to form tubules or incorporate into vascular networks in Matrigel (16,19). Furthermore, the honeycomb-like tubular structures induced by Matrigel are not morphologically representative of vascular networks in vivo (21). The advantages of using the in vitro model outlined here are that tubulogenesis is specific for ECs and more closely resembles capillaries found in vivo, allowing more functionally relevant assessment of EPCs.
Here, using an EC-specific angiogenesis model, we show via 3D microscopy that early EPCs do not directly participate in tubulogenesis. Despite this, early EPCs strongly augmented tubule formation by differentiated ECs, even when physically separated from ECs using transwell culture systems, consistent with a paracrine angiogenic effect. Our findings suggest that early EPCs are not true progenitors of ECs but a monocytic cell capable of indirectly facilitating angiogenesis in a paracrine fashion. Consistent with our findings, Yoder et al. (22) have reported that early EPCs are cells of low proliferative potential that are hematopoietic in origin and differentiate into macrophages rather than ECs in culture. Previous conflicting reports regarding the tubulogenic capacity of early EPCs are likely attributable to the nonspecific tubulogenic effects of Matrigel, which has been shown to induce tubulogenesis by monocytes (18). Our finding that early EPCs augment angiogenesis in a paracrine fashion is concordant with findings of several others who have reported a facilitator role for bone marrow-derived progenitors (4). This ancillary function has been proposed to occur via secretion of paracrine factors such as VEGF, hepatocyte growth factor, granulocyte-colony stimulating factor and interleukin 8 (23).
Originally described by Lin et al. (9) in a study of gender-mismatched bone marrow transplant recipients, OECs are rare circulating cells with high proliferative potential (being capable of more than 100-fold expansion ex vivo) that are clonally distinct from early EPCs (22). Although OECs can be distinguished from early EPCs by their lack of expression of the hematopoietic-specific cell surface antigens CD14 and CD45 (22), their origin is debated. Moreover, markers for identifying the more primitive precursors of both EPC types are largely unknown. The OECs express similar cell surface markers to mature ECs but can be distinguished from mature ECs in vitro by their delayed appearance in culture (approximately 2 weeks), increased telomerase activity, resistance to apoptosis, and a much higher rate proliferation, consistent with differentiating progeny from adult stem cells (9,23). Unlike early EPCs, the role of OECs in angiogenesis has been much less studied. In this study, we found that OECs, in marked contrast to early EPCs, directly participated in angiogenesis, both by de novo tubulogenesis and direct incorporation into pre-formed EC tubules. These functional observations suggest that OECs are true EC progenitors from a neovascularization perspective. Interestingly, OECs did not seem to exert a significant paracrine angiogenic effect in that they do not augment tubulogenesis by ECs when physically separated from ECs in culture. However, we were unable to exclude paracrine effects when OECs are in direct contact with ECs, because OECs directly incorporate into mature EC networks, making discrimination of paracrine effects in this scenario very difficult.
Implications for the optimization of cell therapies.
Conflicting data from studies using different cell preparations highlight the need for more thorough characterization of different putative progenitor populations. Our current study suggests that distinct therapeutic roles may exist for different EPCs. Because early EPCs seem to principally promote angiogenesis in a paracrine manner, their potential in therapeutic angiogenesis may be better harnessed by facilitating local delivery angiogenic factors with or without cells. In this respect, more research is needed to identify and characterize pro-angiogenic factors secreted by these cells. Because they directly participate in endothelial tubulogenesis, OECs may provide the building blocks for neovascularization. Moreover, because OECs are highly proliferative, they are potentially well suited for ex vivo expansion for subsequent therapeutic delivery. Nevertheless, there are challenges in harnessing OECs for cell therapy. Their rarity (0.017 per 106 MNCs) (22) makes isolation challenging, whereas their delayed outgrowth from culture may limit their application in the context of recent ischemia if there proves to be a limited time window for clinical benefit. Optimization of OEC isolation/culture is required before these cells may be more appropriately investigated for clinical therapies.
|
Conclusions
|
|---|
We report a novel human EC-specific angiogenesis assay that, unlike previous assays, permits detailed functional characterization of the role of 2 putative EPCs, early EPCs and OECs, in angiogenesis. We found that early EPCs do not directly participate in but rather augment angiogenesis in a paracrine fashion. In contrast, OECs form tubules independently and incorporate into developing vascular networks but exert no paracrine angiogenic effects, suggesting that OECs and not early EPCs are true progenitors of ECs. The different angiogenic properties of early EPCs and OECs suggest that these cells may have distinct therapeutic roles and that different strategies may be required to optimally harness their respective potential for therapeutic neovascularization.
|
Appendix
|
|---|
For accompanying videos, please see the online version of this article.
|
Footnotes
|
|---|
Supported by a grant from the National Health and Medical Research Council (No. 457534).
|
References
|
|---|
1. Celermajer DS. Endothelial dysfunction: does it matter? Is it reversible? J Am Coll Cardiol 1997;30:325-333.[Abstract]2. Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis Science 1997;275:964-967.[Abstract/Free Full Text] 3. Shantsila E, Watson T, Lip GY. Endothelial progenitor cells in cardiovascular disorders J Am Coll Cardiol 2007;49:741-752.[Abstract/Free Full Text] 4. Ziegelhoeffer T, Fernandez B, Kostin S, et al. Bone marrow-derived cells do not incorporate into the adult growing vasculature Circ Res 2004;94:230-238.[Abstract/Free Full Text] 5. Schachinger V, Erbs S, Elsasser A, et al. Intracoronary bone marrow-derived progenitor cells in acute myocardial infarction N Engl J Med 2006;355:1210-1221.[Abstract/Free Full Text] 6. Janssens S, Dubois C, Bogaert J, et al. Autologous bone marrow-derived stem-cell transfer in patients with ST-segment elevation myocardial infarction: double-blind, randomised controlled trial Lancet 2006;367:113-121.[CrossRef][Web of Science][Medline] 7. Lunde K, Solheim S, Aakhus S, et al. Intracoronary injection of mononuclear bone marrow cells in acute myocardial infarction N Engl J Med 2006;355:1199-1209.[Abstract/Free Full Text] 8. Peichev M, Naiyer AJ, Pereira D, et al. Expression of VEGFR-2 and AC133 by circulating human CD34 (+) cells identifies a population of functional endothelial precursors Blood 2000;95:952-958.[Abstract/Free Full Text] 9. Lin Y, Weisdorf DJ, Solovey A, Hebbel RP. Origins of circulating endothelial cells and endothelial outgrowth from blood J Clin Invest 2000;105:71-77.[Web of Science][Medline] 10. Assmus B, Schachinger V, Teupe C, et al. Transplantation of Progenitor Cells and Regeneration Enhancement in Acute Myocardial Infarction (TOPCARE-AMI) Circulation 2002;106:3009-3017.[Abstract/Free Full Text] 11. Kalka C, Masuda H, Takahashi T, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization Proc Natl Acad Sci U S A 2000;97:3422-3427.[Abstract/Free Full Text] 12. Taub M, Wang Y, Szczesny TM, Kleinman HK. Epidermal growth factor or transforming growth factor alpha is required for kidney tubulogenesis in Matrigel cultures in serum-free medium Proc Natl Acad Sci U S A 1990;87:4002-4006.[Abstract/Free Full Text] 13. Ng MK, Nakhla S, Baoutina A, Jessup W, Handelsman DJ, Celermajer DS. Dehydroepiandrosterone, an adrenal androgen, increases human foam cell formation: a potentially pro-atherogenic effect J Am Coll Cardiol 2003;42:1967-1974.[Abstract/Free Full Text] 14. Ng MK, Wu J, Chang E, et al. A central role for nicotinic cholinergic regulation of growth factor-induced endothelial cell migration Arterioscler Thromb Vasc Biol 2007;27:106-112.[Abstract/Free Full Text] 15. Donovan D, Brown NJ, Bishop ET, Lewis CE. Comparison of three in vitro human ‘angiogenesis’ assays with capillaries formed in vivo Angiogenesis 2001;4:113-121.[CrossRef][Medline] 16. Loomans CJ, Wan H, de Crom R, et al. Angiogenic murine endothelial progenitor cells are derived from a myeloid bone marrow fraction and can be identified by endothelial NO synthase expression Arterioscler Thromb Vasc Biol 2006;26:1760-1767.[Abstract/Free Full Text] 17. Rohde E, Malischnik C, Thaler D, et al. Blood monocytes mimic endothelial progenitor cells Stem Cells 2006;24:357-367.[Abstract/Free Full Text] 18. Schmeisser A, Garlichs CD, Zhang H, et al. Monocytes coexpress endothelial and macrophagocytic lineage markers and form cord-like structures in Matrigel under angiogenic conditions Cardiovasc Res 2001;49:671-680.[Abstract/Free Full Text] 19. Zhang SJ, Zhang H, Wei YJ, et al. Adult endothelial progenitor cells from human peripheral blood maintain monocyte/macrophage function throughout in vitro culture Cell Res 2006;16:577-584.[CrossRef][Web of Science][Medline] 20. Carmeliet P. Angiogenesis in life, disease and medicine Nature 2005;438:932-936.[CrossRef][Medline] 21. Montanez E, Casaroli-Marano RP, Vilaro S, Pagan R. Comparative study of tube assembly in three-dimensional collagen matrix and on Matrigel coats Angiogenesis 2002;5:167-172.[CrossRef][Medline] 22. Yoder MC, Mead LE, Prater D, et al. Redefining endothelial progenitor cells via clonal analysis and hematopoietic stem/progenitor cell principals Blood 2007;109:1801-1809.[Abstract/Free Full Text] 23. Yoon CH, Hur J, Park KW, et al. Synergistic neovascularization by mixed transplantation of early endothelial progenitor cells and late outgrowth endothelial cells: the role of angiogenic cytokines and matrix metalloproteinases Circulation 2005;112:1618-1627.[Abstract/Free Full Text]
Related Articles
-
New Insights on Endothelial Progenitor Cell Subpopulations and Their Angiogenic Properties
- Eduard Shantsila, Timothy Watson, Hung-Fat Tse, and Gregory Y.H. Lip
J. Am. Coll. Cardiol. 2008 51: 669-671.
[Full Text]
[PDF]
-
Inside This Issue of JACC
J. Am. Coll. Cardiol. 2008 51: A25-A26.
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
D. P Sieveking and M. K. Ng
Cell therapies for therapeutic angiogenesis: back to the bench
Vascular Medicine,
May 1, 2009;
14(2):
153 - 166.
[Abstract]
[PDF]
|
|
|
|
|
|
|
Y. Zhang, D. A. Ingram, M. P. Murphy, M. R. Saadatzadeh, L. E. Mead, D. N. Prater, and J. Rehman
Release of proinflammatory mediators and expression of proinflammatory adhesion molecules by endothelial progenitor cells
Am J Physiol Heart Circ Physiol,
May 1, 2009;
296(5):
H1675 - H1682.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. M. Leone, M. Valgimigli, M. B. Giannico, V. Zaccone, M. Perfetti, D. D'Amario, A. G. Rebuzzi, and F. Crea
From bone marrow to the arterial wall: the ongoing tale of endothelial progenitor cells
Eur. Heart J.,
April 2, 2009;
30(8):
890 - 899.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. S. Bonder, W. Y. Sun, T. Matthews, C. Cassano, X. Li, H. S. Ramshaw, S. M. Pitson, A. F. Lopez, P. T. Coates, R. L. Proia, et al.
Sphingosine kinase regulates the rate of endothelial progenitor cell differentiation
Blood,
February 26, 2009;
113(9):
2108 - 2117.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. A. Silva, E.-S. Kim, H. J. Kong, and D. J. Mooney
Material-based deployment enhances efficacy of endothelial progenitor cells
PNAS,
September 23, 2008;
105(38):
14347 - 14352.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. K. Hirschi, D. A. Ingram, and M. C. Yoder
Assessing Identity, Phenotype, and Fate of Endothelial Progenitor Cells
Arterioscler. Thromb. Vasc. Biol.,
September 1, 2008;
28(9):
1584 - 1595.
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. J. Povsic and P. J. Goldschmidt-Clermont
Review: Endothelial progenitor cells: markers of vascular reparative capacity
Therapeutic Advances in Cardiovascular Disease,
June 1, 2008;
2(3):
199 - 213.
[Abstract]
[PDF]
|
|
|
|
|
|
|
D. S. Celermajer and M. K.C. Ng
Where There's Smoke...
J. Am. Coll. Cardiol.,
May 6, 2008;
51(18):
1772 - 1774.
[Full Text]
[PDF]
|
|
|
|
|
|
|
E. Shantsila, T. Watson, H.-F. Tse, and G. Y.H. Lip
New insights on endothelial progenitor cell subpopulations and their angiogenic properties.
J. Am. Coll. Cardiol.,
February 12, 2008;
51(6):
669 - 671.
[Full Text]
[PDF]
|
|
|
|
Top
Abstract
Methods