SEARCH

SEARCH BY CITATION

Keywords:

  • Atherosclerosis;
  • Endothelial progenitors;
  • Neoangiogenesis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

Endothelial progenitor cells (EPCs) are under investigation due to their association with vascular injury. In response to chemotactic stimuli they are mobilized from bone-marrow and nonbone marrow sites, they migrate, adhere and home to the injured vessel. Numerous molecular and cellular pathways participate and converge to the EPCs mediated vascular repair. However, the exact phenotypic properties, modes of functions and effects in vascular diseases and particularly in atherosclerosis are under investigation. EPCs represent a heterogeneous group of cells in different stages of differentiation, from hematopoietic bone marrow progenitors to mature endothelial cells that participate in adult vascular repair under ischemic or apoptotic stimuli. This review aims to provide an integrative view of EPC-mediated vascular repair.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

New vessel formation is a crucial process in both normal and pathological conditions. Two different mechanisms are responsible for vascularization: vasculogenesis and angiogenesis. Vasculogenesis involves the differentiation of endothelial precursor cells, also known as angioblasts, into endothelial cells (ECs) and their de novo organization to a primitive vascular blood plexus. Angiogenesis is defined as the growth of new capillaries from preexisting blood vessels. Vascular development plays a decisive role during prenatal development and is downregulated postnatally.

During the prenatal period, the circulation of the extra-embryonic yolk sac is generated by vasculogenesis followed by formation of primitive cardiovascular system and vascularization of mesoderm and endoderm derived organs. The dogma that after birth new blood vessels develop by angiogenesis was challenged about a decade ago. Artificial vascular grafts are re-endothelialized by both sprouting vessels and blood cells (fallout endothelialization) [1]. This phenomenon has been proposed to be mediated by cells exhibiting similar antigenic profile [2] with the mononuclear cells that Asahara et al. injected in the hindlimb ischemia models [3]. These cells share common antigenic profile with embryonic angioblasts [4]. Early experimental studies demonstrated that endothelial progenitor cells (EPCs) from bone marrow are involved in normal (placenta, corpus luteum), pathological (tumor growth), and pathophysiological (wound healing, regeneration of ischaemic tissue) neovascularization [5]. Many years after these observations, several studies pointed out that different cell populations existing in the postnatal organism (such as mesnchymal cells [6], adipose cells [7], smooth muscle cells [8], umbilical cord cells [9], monocytes [10] are capable of de novo formation of capillary structures by migration and proliferation.

Since the initial experiments, a heterogeneous population of hematopoietic progenitors identified by the expression of CD34 (CD for cluster of differentiation) and Vascular Endothelial Growing Factor Receptor 2 (VEGF2) transmembranic proteins, has been suggested to trigger adult vasculogenesis. This population is characterized by some fundamental properties: differentiation of EC both ex vivo and in vivo, improvement of perfusion and recovery of limb ischemia (in animal models) [11] and myocardial infarction [12]. Recently, it was shown that these cells act as potent inhibitors of atherosclerotic plaque progression and as biomarkers of vascular wealth both in healthy individuals and patients [13].

However, in some reports, the BM contribution to endothelium has been estimated to be very low, ranging from undetectable [14] to 4.9%[15]. Recently, Purhonen et al. in a parabiotic system demonstrated that no BM-derived precursors contribute to vascular endothelium and that cancer growth does not require BM-derived endothelial progenitors implying that endothelial differentiation is not a typical in vivo function of normal BM-derived stem cells, while it seems be an extremely rare event [16]. These results are in concordance to previous experiments by Shinde Patil et al., demonstrating that lin-/c-kit+/sca-1+ progenitor cells contribute to significant recruitment to carcinomas in vivo, while they do not appear to participate functionally in tumor neovascularization [17]. These results illustrate substantial differences between different tumor and hind limb ischemia models with respect to postnatal vasculogenesis and raise important questions: Do EPCs represent a well-defined population or are they solely an in vitro phenomenon a cell culture phenomenon? Which cell populations do actually contribute to vascular reparation? Most importantly, how do they exert their effects and which molecular and cellular procedures mediate them? What is their role in atherosclerosis (where vasculogenesis and angiogenesis exert minor contribution)? The current review article summarizes the major points of neoangiogenesis and discusses the crucial role of EPCs in vascular repair (schematically represented on Figure 1).

image

Figure 1. In humans, cardiovascular risk factors, such as hyperlipidemia, hypertension, diabetes, smoking, physical inactivity and aging, impair number and function of EPCs while factors demonstrated to be protective against atherosclerosis have been correlated with increased numbers of EPCs, such as high-density lipoprotein, statins, and angiotensin II inhibitors Chemokines inducing mobilization interfere with the interactions between stem cells and bone marrow stromal cells and subsequently allow stem cells to disengage the bone marrow, and to pass through the sinusoidal endothelium to enter the blood stream. Substantial data support the notion the SDF-1/CXCR-4 pathway exerts a crucial role in modulating the mobilization of proangiogenic hematopoietic cells from bone marrow. SDF-1, after binding to CXCR4, causes mobilization of calcium, decrease of cyclic AMP within the cells, and activation of multiple signal transduction pathways, including PI3K, phospholipase C/protein kinase C, and MAP kinases ERK1/2. The interaction of mobilized EPCs with the injured vascular wall requires adhesion molecules. The adhesion procedure continues with the substitution of the loose, selectin mediated connections of EPCs to the endothelial monolayer, by firm connections with ECM molecules. ECM substances provide cells with a structural, chemical and mechanical substrate that is essential for normal development and responses to pathophysiological signals. The last step in EPC recruitment is extravasation and transendothelial migration. As soon as EPC home injured vessel walls they differentiate into mature endothelial cells under the influence of local growth factor and chemokine production, or exert indirect paracrine effects. Sdf-1; stromal cell derived factor-1, cxcr4; cysteine-amino acid X-cysteine receptor-4, VEGF; Vascular Endothelium Growth Factor, MMP-9; Matrix Metalloproteinase-9, ROS; Reactive oxygen species, skitL; soluble kit (stem cell factor) ligand, PSGL-1; P selectin glycoprotein ligand-1, VCAM-1; vascular cell adhesion molecule-1, ICAM-1; intercellular adhesion molecule-1, VLA-4; Very Late Antigen-4, LFA-1; Lymphocyte function-associated antigen-1, MAC-1; macrophage adhesion molecule-1, PECAM-1; Platelet-endothelial cell adhesion molecule-1.

Download figure to PowerPoint

Origin and Multiple Identities of Cells Mediating Vascular Repair

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

It has been proposed that during embryonic development haematopoietic cells arise from a mesodermal progenitor with both endothelial and haematopoietic potential, called the haemangioblast [18]. The haemangioblast can be isolated from differentiated mouse embryonic stem cells based on Kdr (previously known as Flk-1) expression and generates a blast colony containing haematopoietic and ECs after 4 days of culture [18]. The Flk-1+ population generates haematopoietic cells through the formation of a haemogenic endothelium intermediate characterized as Tie2hi/c-kit+/CD41- [19]. HSCs and progenitors that populate the fetal liver and adult bone marrow originate from the differentiated endothelium that resides in the functional vasculature of the murine conceptus [19,20].

Ingram et al. defined the differentiation stages from the EPCs to mature ECs, based on their potential to form clonogenic cell clusters [21]. An important characteristic of endothelial progenitors is the ability to form tubular lumenized vessel-like structures after networking of cell clusters via cell branches [22].

Not worth mentioning that it is difficult to differentiate immature EPC from primitive hematopoietic stem cells (HSC) as those cells share common surface, that is, AC133, CD34 or VEGFR-2 and there is no exclusive EPC marker. Both EPC and mature EC express similar endothelial-specific markers, including VEGFR-2, Tie-1, Tie-2, and VE-cadherin. Identification of different cell subsets is further complicated by the fact that hematopoietic cells (HC) express markers similar to those found on EC, such as CD34, PECAM, Tie-1, Tie-2, ephrin and VEGFR-1, transcription factors, such as SCL/tal-1 and AML1, vWF. AC133 and CD45 are expressed only on primitive HC, but have not been found on EC. Additionally, CD45 is expressed only during differentiation into hematopoietic precursors while it is not expressed on endothelial lineage cells. CD 146 appears to be expressed almost ubiquitously on EC but is not found on HC [23].

Early experimental results reported that a 3-day culture of CD34+ cells in fibronectin coated plates produces functional ECs characterized by expression of endothelial lineage markers (CD31, Tie, VEGFR2) and uptake of acetylated low density lipoproteins (acLDL) [24]. However, CD34 is not a stem cell marker and KDR is not a specific mature EC marker. An early HSC marker, CD133, has been adopted as an additional marker to indicate “real” EPCs. The CD133 marker is expressed on hematopoietic stem and progenitor cells isolated from peripheral blood, human bone marrow and fetal liver. CD133 is a reliable marker in terms of defining and tracking human angioblast-like EPCs and distinguishing these from mature endothelial or monocytic cells [25]. Therefore, the combined phenotype of CD34+/CD133+/VEGFR2+ is the current definition for EPCs [26]. However, Case et al. recently reported that human umbilical cord blood cells or adult CD34+VEGFR2+CD133+ cells in fact represent an enriched population of CD45+ haematopoietic precursors, but they do not contribute to the formation of ECs in vitro[27]. However, the homologous murine c-kit+/sca-1+/lin- population is reported to repopulate bone marrow in totally irradiated mice and differentiate ex vivo[28]. Other candidate populations, such as CD34+CD45+CD146+ cells previously reported to represent EPCs, have been investigated for potential differentiation into ECs [29]. Nevertheless, isolating EPCs solely by their phenotype is challenging since different sorts of endothelial colonies can be grown in culture.

Early EPCs originate as early as 3 days after plating. They tend to display limited proliferative capacity and disappear after 2 weeks of culture. Cells surviving after 2–3 weeks are named as late EPCs. They show a more typical morphology of ECs, have a higher proliferative potency [30,31]. Both early and late EPCs bind Lectin and uptake acLDL, but late EPCs express higher density of endothelial lineage markers and have higher vasculogenetic potency than early EPCs. It has been proposed that early EPCs display features of myeloid cells and that their modest ability to stimulate vascular regeneration occurs mainly through the secretion of growth factors, rather than through physically integrating into the nascent vasculature [32,33]. Interestingly, early EPCs are currently contemplated as trans-differentiated cells of the monocyte lineage. CD14+ cells give rise to Colony Forming Units (CFUs) exhibiting endothelial phenotype, but lacking gene expression consistent with angiogenesis-related processes [34,35]. CFUs are previously reported to originate from a population of CD45+/CD14+/CD115+ monocytes without direct angiogenic properties [36] and a population of CD3+/CXCR4+ cells which release angiogenic factors [37]. In contrast, only the CD34+CD45- cell fraction generates outgrowth cells, while CD34+CD45+ haematopoietic cells differentiate into early outgrowth cells through a CD14+ monocytic pathway [38].

Importantly, there is evidence that simultaneous infusion of (in vitro generated) early and late outgrowth EPCs act synergistically during vascular repair. Late outgrowth cells appear to contribute structurally to vasculogenesis, whereas early outgowth cells do not directly contribute to the vasculogenesis, but rather act in an indirect paracrine manner by locally secreting angiogenic substances that promote structural healing by resident ECs and incorporated EPCs [39]. In conclusion, cultured EPCs of early or late outgrowth have little in common with cells counted by flow cytometry or sorted by magnetic beads. CD34+ cells sorted by magnetic beads include late outgrowth EPCs and exhibit similar angiogenic properties.

Further debate has been brought up by the demonstration of EPCs isolated from tissues other than bone marrow. CD34+/CD31+/Tie+/cadherin+ cells were isolated from small vessels of muscle, adipose and dermal tissue. Due to their angiogenic properties they could be acknowledged as tissue specific EPCs [40,41]. Furthermore, previous studies have proposed the existence of nonhematopoietic, nonmyeloid, nonendothelial tissue specific c-kit+/CD45-/CD146- cells which are mobilized in models of limb ischemia [42]. Adipose tissue derived stromal cell (ADSCs) clusters have also been studied to form branched tube-like structures in culture, which were strongly positive for CD34 and CD31, and lose their ability to undergo adipocyte differentiation [43]. Additionally, these cells have been reported to enhance ischemia-induced angiogenesis [44] and promote wound healing [45]. However, isolation and culture methods of ADSCs are a major challenge. These data make even more complicated the establishment of a discrete population of EPCs, as it seems that in response to angiogenic stimuli various sources of progenitor cells contribute to postnatal vasculogenesis.

Mobilization of Vascular Progenitors by Chemotactic Signals

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

The microenvironment or bone-marrow niche determines the mobilization of bone-marrow derived endothelial progenitors. The pathways leading to the mobilization of nonbone marrow endothelial progenitors are not entirely elucidated. Chemokines interfere with the interactions between stem cells and bone marrow stromal cells and allow stem cells to disengage from the bone marrow and pass through the sinusoidal endothelium to enter the blood stream.

The SDF-1/CXCR-4 pathway exerts crucial role in modulating the mobilization of proangiogenic hematopoietic cells from bone marrow. The chemokine stromal cell derived factor-1 (SDF-1, also known as CXCL12) is a G-protein coupled receptor, constitutively expressed and inducible chemokine that regulates multiple physiological processes. It is expressed in bone marrow, liver, spleen and lung. The cognate receptor for SDF-1, CXCR4, is widely and constitutively expressed by hematopoietic and ECs. The SDF-1/CXCR-4 pathway is a prerequisite of embryonic vasculogenesis as demonstrated by the vascular abnormalities that CXCR4−/- or SDF-1−/- mice exhibit [46]. An alternate CXCR7 receptor of SDF-1 is not associated with cell mobilization but only with increased adhesion and survival of progenitor cells [47].

SDF-1, after binding to CXCR4, induces mobilization of calcium, decrease of cyclic AMP within the cells, and activation of multiple signal transduction pathways, including PI3K, phospholipase C/protein kinase C, and MAP kinases ERK1/2. Both PI3K/Akt and MAPK/ERK signal transduction pathways have been found to mediate the cell migration induced by chemokines or cytokines in different cell types [48]. Recently, studies in EPCs have demonstrated that statins [49], estrogen [50], and erythropoietin (EPO) [51] could improve migration, proliferation, and prevent senescence or apoptosis of EPCs. In these procedures the activation of the PI3K/Akt signal transduction pathway exerts important role.

Chemokine production at sites in demand of rapid vascularization is crucial for recruiting vascular progenitors. SDF-1 is found expressed in ECs and pericytes of hypoxic and injured tissues, platelets, tumor microenvironment and EPCs and appears to increase in states of endothelial dysfunction and vascular injury. In other words, apoptotic and ischemic conditions such as myocardial infarction and limb ischemia trigger expression of chemokines and cytokines [52]. Furthermore, induction of SDF-1 expression seems to have a major role in initiating revascularization of the ischemic injured tissues [53]. As regards SDF-1 expression in ischemic sites it seems to be directly correlated with the severity of ischemia [54]. Hypoxia-inducible factor-1 [55] and integrin-linked kinase [56] promote SDF-1 expression in hypoxic situations. SDF-1 is documented to regulate a variety of cellular functions of EPCs such as cell migration, proliferation, and survival. Inhibition of the SDF-1/CXCR4 axis by AMD3100 partially blocks the homing of progenitor/stem cells to the ischemic tissues probably due to attenuation of migration, adhesion and invasion properties of EPCs and impairs blood flow recover in ischemic hind-limb [57]. In contrast, the S1P, a CXCR4 agonist enhances neovascularization [58]. However, other studies have reported that AMD3100 attenuates neointima formation implying that bone-marrow derived progenitors contribute to neointima formation [59]. Therefore, it seems that CXCR4 mobilizes different types of bone marrow-derived cells which susequently contribute to neointima formation by differentiating into smooth muscle-like and to endothelial lining by differentiating into ECs.

Other studies involve additional chemokines in EPC recruitment. CXC-chemokine IL-8 is an inflammatory molecule able to stimulate angiogenesis. Its cellular receptors CXCR1 and CXCR2 contribute to homing of endothelial progenitors to ischemic tissues. Myocardial infarction induces a local increase of the expression of IL-8 [60]. In addition, neutralizing anti-IL-8 antibodies and antibodies against the IL-8 receptors, CXCR1 or CXCR2, reduce neovascularization in sites of arterial injury [61]. Exhibit MIP-1 and interleukins exhibit same angiogenic properties [62]. The later may upregulate high mobility group box-1 (HMGB-1) nuclear protein. This is released extracellularly and recruits EPCs upon ischemic and apoptotic stimuli [63].

Growth factors constitute another major source of chemotactic molecules that can recruit bone-marrow derived endothelial progenitors. VEGF, angiopoietins, and basic Fibloblast Growth Factor (bFGF) are major angiogenic mitogens.

VEGFs encompass a family of structurally related proteins that include placental-derived growth factor, VEGF-A, VEGF-B, VEGF-C (or VEGF-2), VEGF-D, VEGF-E, and placental growth factor (PIGF). The downstream signals of VEGFs in the vascular endothelium are mediated by 3 tyrosine kinase signaling receptors (VEGF receptor [VEGFR]-1, -2, and -3) [64]. VEGFR-1 and -2 receptors are necessary for the VEGF-mediated mobilization and recruitment of EPCs [65].

Intracellularly, the activation of the small GTPases of the Rho family is centrally involved in regulating EC migration in response to activation of VEGFR-2 probably via the PI3K/Akt-PKB pathway [66]. NO is a major regulator of angiogenesis and is rapidly produced by eNOS following its activation downstream of the VEGFR-2/PI3K/Akt-PKB axis in ECs activated by VEGF [67]. VEGF can contribute to activate eNOS and NO production by dissociating eNOS from caveolin-1 [68]. VEGF and other mobilizing stimuli might activate the eNOS within the bone marrow stromal cells. This stimulates and/or maintains matrix metalloproteinase-9 (MMP-9) activity resulting in cleavage soluble kit ligand (sKitL) from membrane-bound kit ligand to mobilize c-kit+ stem cells into the circulation [69]. The role of eNOS was demonstrated in mouse models of eNOS deficiency in the bone marrow microenvironment that exhibit impaired mobilization of stem and progenitor cells from the bone marrow [70]. In addition to NO reactive oxygen species (ROS) also promote angiogenesis and endothelial progenitors’ recruitment due to activation of NADPH oxidase by VEGF [71].

Interestingly, organ-specific expression of VEGF is sufficient to mobilize and recruit hematopoietic cells from the bone marrow to the blood, but retention of the proangiogenic subpopulation of hematopoietic cells in peripheral organs requires SDF-1. Thus, VEGF recruits heterogenous populations of bone derived precursors and induces SDF-1 expression in mural cells around blood vessels. SDF-1 and VEGF-A demonstrate similar properties and one is activated by each other. However, only SDF-1 determines angio-competent cells [72].

Angiopoietins (Ang) 1–4 are proangiogenic growth factors that specifically activate ECs in a paracrine manner. They exert their gunction via binding to the tyrosine kinase receptor Tie-2, whereas the ligand for the related Tie-1 receptor remains to be identified. Ang1 is an activator of Tie-2 and Ang2 is known to antagonize the binding of Ang1 to Tie-2 [73]. Angiopoietin-1 stimulates angiogenesis under both physiological and pathological conditions while Ang2 has been implicated as being a natural angiostatic factor. In the presence of VEGF, Ang2 induces EC migration, proliferation, and sprouting, as well as an increase in the diameter of capillaries [74]. In contrast, in the absence of VEGF, Ang2 does not promote cell migration; rather, it induces apoptosis of ECs and regression of blood vessel. Ang1 alone and Ang2 in the presence of VEGF are thus positive regulators of angiogenesis [75].

Adhesion and Transendothelial Migration of EPCs

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

The interaction of EPCs with the injured vascular wall requires expression of adhesion molecules. EPCs have to attach and roll to vessel ECs in order to get activated. Rolling of ECs is mediated by selectins. Three related adhesion molecules E-selectin, L-selectin, and P-selectin constitute the selectin family. These molecules bind sialyl-Lewis-X-like carbohydrate ligand/sialomucin-like surface molecule (P selectin glycoprotein ligand-1, PSGL-1) complexes. E-selectin is predominantly expressed on ECs while P-selectin is expressed on platelets as well. As initially reported in E-selectin deficient mice, E-selectin administration promotes adhesion molecule expression on endothelial surface cells, enhances mobilization and adhesion of EPCs, improves blood flow after limb ischemia and upregulates chemokine (interleukin-8) expression [76,77].

Intriguingly, the EPO-producing human hepatocellular carcinoma (Eph) receptor/ephrin pathway is recently shown to upregulate PSGL-1 and E, P-selectin expression [78]. Eph receptors and ephrins are membrane proteins that are classified into 2 broad subclasses, A and B. Eph receptors belong to the family of receptor tyrosine kinases and they autophosphorylate upon binding to their cognate ephrin ligands. Interaction between Eph receptors and ephrins requires cell–cell contacts as both molecules are anchored to the plasma membrane. The resulting signals propagate bidirectionally into both the Eph receptor-expressing cells (forward signaling) and the ephrin-expressing cells (reverse signaling). During the early stages of vascular development, Ephrin-B (EphB4) is expressed in venous endothelium whereas ephrin-B2 is expressed in arterial endothelium. Studies in mouse embryos showed that these 2 proteins are essential for embryonic heart development and angiogenesis [79]. The role of ephrins in postnatal angiogenesis has been recently described. EphB4 stimulates EPC rolling via selectins, but not EPC mobilization via VEGF and SDF-1/CXCR-4 pathways [78].

The process of adhesion continues with the substitution of the loose, selectin mediated connections of EPCs to the endothelial monolayer, by firm connections with extracellular matrix (ECM) molecules. ECM substances provide cells with a structural, chemical and mechanical substrate that is essential for normal development and responses to pathophysiological signals. EPCs and ECs adhere to ECM to migrate either dependently (chemotaxis supported by collagen I) or independently (by fibronectin) of chemoattractants [80]. It is worth noting here that an increase in maturation attenuates the migratory capacity of endothelial precursors.

Glycoprotein transmembrane receptors (integrins) are the primary link between ECM ligands and cytoskeletal structures.

Data have shown that integrin receptors may exert diverse actions on EPC migration, depending on the local milieu. During vasculogenesis a5b1 integrins are involved in vascular development [81] while avb3 are crucial for angioblast migration [82]. In murine models of limb ischemia both murine and human endothelial precursors (Lin/sca+) demonstrated migration mediated by a4b1 and ab2 (LFA-1) integrins [83].

MMPs contribute to both mobilization (e.g., the MMP-9-mediated release of kit ligand [69] and the sdf-1/cxcr4 cleavage by the MMP-2 [84]) and migration. In response to chemotactic signals, EPCs express highly cathepsin L and MMPs. Particularly cathepsin L is crucial for EPCs transendothelial migration (TEM) as shown in cathepsin L knockout mice [85]. EPCs further induce bone marrow progenitor mobilization through the MMP-mediated release of urokinase-type plasminogen activator (u-PA) which increases EPC migration [86].

The last step in EPC recruitment is extravasation and TEM. It seems that this process is mediated by the Platelet-EC adhesion molecule-1 (PECAM-1/CD31). This is expressed on the surface of EC and leukocytes and exerts an important role in endothelial-leukocyte and endothelial-endothelial cell–cell interactions. It has been previously reported that anti-PECAM-1 antibody blocks of these interactions and inhibits TEM of leukocytes and angiogenesis [87,88]. CD99 [89], Jam-A [90], and –C [91] molecules might also affect TEM. An underlying mechanism of these processes could be the attenuation of cadherin mediated cell–cell interaction.

As soon as EPC home injured vessel walls they differentiate into mature ECs under the influence of local growth factor and chemokine production. Shear stress also plays important role into their homing and maturation since it activates the PI3K/Akt downstream pathway [91] leading to eNOS upregulation, expression of EC markers, histone acetyltransferase (HDAC) 3 acetylation and subsequentl upregulation of p53 and p21 tumor suppressor genes [92]. Commitment to endothelial lineage is sharply diminished when HDAC activity is inhibited, mainly because of down-regulation of HoxA9 expression. Experiments using HoxA9 –/–mice elucidated a defect in circulating EPCs and an impaired postnatal neovascularization capacity after the induction of ischaemia.

Interaction of EPCs with Aspects of Atherosclerosis (Table 1)

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References
Table 1.  Studies of EPCS effects on atherosclerotic plaque.
StudyExperimental modelDoseTime of deliveryTherapeutic results
Zoll et al. [125]ApoE–/– RAG2–/– mice5×105 EPCs and SPCs1 or 12 weeksNo significant alterations in plaque composition with EPC infusion
   No significant effect on atherosclerosis
Wassmann et al. [121]ApoE –/– mice2×107 MNCs3 consecutive daysIncreased vascular eNOS activity
   Improved endothelium dependent vasodilation
George et al. [124]ApoE –/– mice106 BM cells or spleen cell-derived EPCs2 weekly intervals (3 times)Increased atherosclerotic lesion size in both groups,
   Decreased plaque stability in mice injected spleen cell-derived EPCs
Silvestre et al. [123]ApoE –/– mice106 BM-derived MNCsAt ischemia inductionBM-MNC transplantation without ischemia does not affect atherosclerotic plaque size
   BM-MNC transplantation with ischemia increases lesion size
Rauscher et al. [122]ApoE –/– mice106 BM-derived progenitor cellsEvery two weeksPrevention of atherosclerosis progression
Hasegawa et al. [109]Hyperlipidemic rabbits10 mg/kg/day G-CSFFor 7 daysReduced stenosis score
Haghigat et al. [110]ApoE –/– mice10 mg/kg/day G-CSF5 days/week total of 20 dosesIncreased plaque burden

Recent evidence suggests that levels of EPCs in healthy individuals may be a surrogate biologic marker for vascular function and cumulative cardiovascular risk. In addition, reduced levels of circulating EPCs may predict independently early subclinical atherosclerosis, atherosclerotic disease progression, occurrence of cardiovascular events, death from cardiovascular disorders [24,93] and prognosis after ischemic stroke [94]. However, the exact role of EPCs in atherosclerosis has not been fully determined yet.

The early response to vascular injury (adherent platelets, fibrin, thrombin and protease activated receptor-1), and mechanical removal of endothelial monolayer recruit bone marrow-derived progenitor cells to arterial thrombi in vitro and in vivo which in turn mediates the regeneration of ECs [95,96]. Thus, EPCs may provide an endogenous repair mechanism to counteract ongoing risk factor-induced endothelial dysfunction and injury. In humans, cardiovascular risk factors, such as hyperlipidemia, hypertension, diabetes, smoking, physical inactivity, and aging [97–100], impair number and function of EPCs, while factors demonstrated to be protective against atherosclerosis have been correlated with increased numbers of EPCs, such as high-density lipoprotein, statins, and angiotensin II inhibitors [101–104]. Functional studies in atherosclerotic apolipoprotein E (ApoE) deficient mice indicate that progressive progenitor cell deficits may contribute to the development of atherosclerosis and that EPC administration can improve endothelium-dependent vasodilation [121].

Mobilization of EPCs rendered controversial results. Injection of recombinant human G-CSF increased the number of circulating MNCs that express EC lineage markers, facilitates reendothelialization and inhibits neointima development following balloon angioplasty or intravascular radiation [105,106].

Similar inhibitory effects on neointima formation were reported after erythropoietin infusion [107].

However, recent evidence indicates that G-CSF treatment does not attenuate neointimal hyperplasia and restenosis formation in a canine model of renal arterial balloon injury [108]. Similarly, the effect of G-CSF administration in the context of atherosclerosis is also controversial. Although there are some reports indicating G-CSF administration prevents the progression of atherosclerosis in animals and improves the clinical signs and symptoms of patients with intractable atherosclerotic peripheral artery disease [109]. Data from other groups have provided controversial conclusions. Haghighat et al. [110] showed (in ApoE deficient mice model of atherosclerosis) that not only short-term administration of G-CSF or GMCSF failed to demonstrate any beneficial therapeutic effect, but also both resulted in a worsening of atherosclerosis. Interestingly, an increase in adventitial vascularity was found, suggesting a mechanistic role for vasa vasorum neovascularization. Hence, the effect of EPC mobilization by G-CSF or GM-CSF in vascular injury and atherosclerosis should be further investigated. VEGF-A administration in ApoE –/– mice has also induced plaque expansion. Plausible explanations accounting for these ambiguous results could be the upregulation of proinflammatory cytokines and recruitment of leukocytes at the injured vessel. In addition, a recent study by Zhang et al. [111] demonstrated that both early and late outgrowth EPCs release proinflammatory adhesion molecules, and chemokines ex vivo. These effects were modulated by statin treatment. The beneficial effects of statins have been previously established in in vivo studies of EPC mobilization and neointima formation inhibition by statin administration [112,113]. Whether some of the reported negative results, are due to secretion of proinflammatory mediators by EPCs, or the infusion of unfractioned bone marrow populations including large numbers of monocytes, requires further investigation.

Regarding direct bone-marrow cell infusion, the results still remain controversial. According to Foteinos et al., endothelial turnover and repair by EPCs are, at least in part, derived from bone marrow during development of atherosclerosis in ApoE–/– mice [114]. Infusion of different EPC populations and ex vivo expanded EPCs contributes to reendothelialization and inhibits intimal hyperplasia in both murine and rabbit model of vascular injury [113,115–119]. However, a recent study failed to validate previous results [120].

The impact of bone marrow-derived cells on atherosclerotic plaque varies according to cell type, cell number, animal model, and existence of hind-limb ischemia (as shown on Table 1). Short-term intravenous transfusion of spleen-derived MNCs improves endothelium-dependent vasodilation in atherosclerotic ApoE –/– mice. Transfusion of either in vitro-differentiated Dil-Ac-LDL/lectin-positive EPCs, CD11b-positive (monocyte marker), CD45R-positive (B-cell marker), or Sca-1-positive (stem cell marker) MNC subpopulations significantly improved endothelium-dependent vasodilation, although these treatments were not as effective as transfusion of total MNCs. Either CD11b-, CD45R-, or Sca-1- MNCs cells resulted in significant attenuation of endothelium-dependent vasodilation as compared with unfractioned MNCs [121].

Chronic treatment with bone marrow-derived EPCs from young nonatherosclerotic ApoE –/– mice prevented atherosclerosis progression in ApoE–/– recipients despite persistent hypercholesterolemia [122]. However, Silvestre et al. reported that transplantation of bone marrow-derived MNCs in ApoE knockout mice with hindlimb ischemia promotes further atherosclerotic plaque progression in an ischemic setting [123]. Similar effects were reported by George et al. [124]. Administration of total bone marrow cells or spleen cell-derived EPCs results in an increase of atherosclerotic lesion size in the ApoE knockout mouse, whereas EPC transfer reduces markers associated with plaque stability. Interestingly, a recent study reported that chronic human EPC injection had no effect on atherosclerosis development or progression in iApoE–/– RAG–/–, but chronic smooth muscle progenitor cells injection limited plaque development and promoted changes in plaque composition towards a stable phenotype in mice [125]. It seems that bone marrow derived cells could exert a variety of effects in an atherosclerotic milieu. Definitely, their beneficial effects are not favored by the local proinflammatory/prothrombotic environment of advanced atherosclerotic disease. Whether EPC could prevent early stage lesions or ameliorate endothelial dysfunction needs further clarification.

Conclusive Remarks

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References

Conclusively, the major points in respect to EPC mediated vascular recovery, requiring further clarification or revision, are:

EPC Phenotype

Different surface marker combinations used in flow cytometry studies, yielded statistically significant results in various animal models of vascular injury and postnatal vasculogenesis. Since common used markers (CD34, VEGFR2, CD133, CD31) and the uptake of acetylated LDLs and lectins have been previously used to stain hematopoietic lineage cells, the populations isolated probably comprise heterogeneous cell subpopulations some of which are hematopoietic progenitors and monocytes. These cells indirectly promote vasculogenesis by secreting regulatory cytokines that promote vessel homeostasis and repair by local cells, including local vessel wall ECs [126,127]. These haematopoietic-derived cells may contribute to vascular repair and homeostasis in an indirect manner providing an explanation for the association between different cell subsets and improved cardiovascular outcomes particularly in experimental studies.

The scientific foundation for using the variable surface marker combinations and CD34+ subsets remain elusive. Moreover, the use of these diverse combinations to define a singular entity (the CEPC), makes the significance of flow cytometric studies difficult to interpret, creates obstacles to the direct comparison of data between laboratories, and may result in discrepancies in the interpretations of study results among different laboratories. Therefore, investigators should strongly consider that any “putative” CEPC, whatever its phenotype, should be carefully assessed by validating its postnatal endothelial differentiation capacity in vitro and in vivo.

One of the strategies that have been used to directly evaluate cell differentiation into ECs in vivo has been the use of transgenic mice that express a fluorescent marker (e.g., green fluorescent protein (GFP)) only in cells expressing an endothelial specific gene, such as Tie-2 [128]. However, these approaches have yielded contradictory results, probably because expression of Tie-2 is not entirely restricted to the endothelial lineage, and is also expressed in haematopoietic cells and monocytes, that also migrate to sites of vascular repair [129].

When choosing the most appropriate EPC phenotype for a given study, one fundamental point is that EPCs are very rare in the circulation, in normal conditions. Identification of cells with one of the above-mentioned phenotypes by flow cytometry is a rare event. In addition, definition of EPCs by flow cytometry implies a conceptual abstraction, because a presumed function is attributed to a relatively simple antigenic phenotype. The rarity of circulating EPCs imposes the use of a very limited number of surface antigens and renders the identification of a cell population with complex function almost impossible.

The antigenic phenotype of EPCs includes both progenitor (CD34, CD133, sca-1, c-kit+) and endothelial markers (VEGFR2 or KDR). Total CD34+ cells should be considered as generic progenitor cells (mostly haematopoietic) rather than EPCs, because a minority of these circulating cells expresses endothelial lineage antigens. It has been criticized that the CD34+KDR+ phenotype may overlap in part with that of mature ECs, because CD34 is known to be expressed also on some microvascular ECs. Unlike CD34, CD133 is never expressed on mature ECs and, therefore, CD133+KDR+ cells may better correspond to EPCs. Unluckily, CD133 is expressed on more immature cells than CD34 and, for this reason CD133+KDR+ cells are rarer than CD34+KDR+ cells in the circulation. The CD34+CD133+VEGFR2+ subset could be used as a restrictive EPC phenotype, but these cells are very rare in the circulation and include haematopoietic rather than endothelial progenitors. The addition of negative selection of CD45+ cells (leukocyte antigen) would increase the purity of the population but it would approximate the lower detection threshold of flow cytometry and increase substantially its variability.

The qualitative evaluation of EPCs requires isolation from peripheral blood, spleen or bone marrow and expansion in culture. Two major types of EPCs can be distinguished phenotypically in a culture of total mononuclear cells grown in endothelial medium [130]. Cells originating as early as 3 days after plating and organizing in clusters are called early EPCs: they display limited proliferative capacity and disappear after 2 weeks of culture. Cells surviving after 2–3 weeks were called late EPCs: they show a more typical morphology of ECs, tend to form a confluent cobblestone layer and have a higher proliferative potential. Both early and late EPCs are acLDL+Lectin+, but late EPCs express higher density of endothelial lineage markers and have a higher vasculogenetic potential than early EPCs. Early EPCs are considered as a manifestation of the extreme plasticity of the monocyte/macrophage lineage which, in certain culture conditions, can assume an endothelial-like phenotype. Their contribution to vascular repair is the secretion of growth factors and proinflammatory cytokines.

In conclusion, the recommended approach would be (i) presortment of cells according to EPC markers (CD34/VEGFR2, CD133/VEGFR2 in human, c-kit/sca-1) before culturing them. CD34+CD133+KDR+ cells include haematopoietic rather than endothelial progenitors, and that true EPCs are not derived from CD133+ or CD34+CD45+ haematopoietic precursor cells, but from CD34+CD45- cells. Currently, the CD34+KDR+ cells represent the best compromise of EPC phenotype in terms of detection accuracy, biological meaning and clinical usefulness. (ii) Evaluation of the culture originating cells with a wide panel (CD11, CD45, VE-Cadherin, CD146, CD31, CD105) of antigens to illustrate their true endothelial lineage and discriminate them from other cell lineages, especially the haematopoietic lineage. (iii) functional studies by means of proliferative capacity or tube forming capacity, iv)use of confocal microscopy for the assessement of in vivo and in vitro differentiation and engraftment of EPCs.

Vascular Repair Endothelial Cell Regeneration or Paracrine Effects?

Endothelial cell in the neointima could be at least in part derived from EPCs. Griese et al., demonstrated that LacZ-transduced EPC transplantation leads to about 60% reendothelialization of balloon-injured rabbit carotid arteries as early as 4 days after transplantation and LacZ-positive cells co-staining with CD31 are seen lining the lumen of injured vessel [117]. Furthermore, it has also been documented that in ApoE –/– mice, intravenous transfusion of spleen-derived MNCs improved endothelium-dependent vasodilation [105]. Finally it has been reported that endothelial turnover and repair by EPCs are, at least in part, derived from bone marrow during development of atherosclerosis in ApoE –/– mice [115].

EPCs have been shown to exert multiple paracrine effects: (a) Early EPCs exert a strong paracrine mitogenic effect on mature ECs in part via IL-8 secretion, (b) enhances the angiogenic response to hind-limb ischemia through release of growth factors [131] but not transdifferentiation into ECs [132], (c) sustained up-regulation of recipient cytokines and growth factors (while cell derived cytokines decrease rapidly after infusion) [133], (d) vasoprotection by increasing prostacyclin PGI-2 production and intracellular concentration of cAMP [134].

However, EPCs can also contribute to atherogenesis by secreting proinflammatory cytokines, inducing smooth muscle cell recruitment to the plaque, and increasing adventitia neovascularization (and intraplaque hemorrhage risk). Finally, it has been demonstrated that platelets could induce differentiation of CD34+ progenitor cells into foam cells [135]. These proinflammatory-mediated proatherogenic effects of EPCs could be attenuated by pretreatment with statins [136] or by administration of late EPCs only. Conclusively, there are many unresolved questions regarding EPC therapy in atherosclerosis and we are not close to their clinical application.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Origin and Multiple Identities of Cells Mediating Vascular Repair
  5. Mobilization of Vascular Progenitors by Chemotactic Signals
  6. Adhesion and Transendothelial Migration of EPCs
  7. Interaction of EPCs with Aspects of Atherosclerosis ()
  8. Conclusive Remarks
  9. Acknowledgments
  10. Conflict of Interest
  11. References