Endothelial progenitor cells in health and atherosclerotic disease

Abstract

Cardiovascular disease is associated with damage of the endothelial monolayer leading to endothelial dysfunction and atherosclerosis. A growing body of evidence suggests that circulating endothelial progenitor cells play an important role in endothelial cell regeneration. In this review we discuss the evolving role of stem‐ and progenitor cells in the maintenance of the vascular wall focusing on new pathophysiological concepts of endothelial cell regeneration. We discuss new insights into vascular stem cell biology derived from experimental and clinical studies.

• Atherosclerosis
• bone marrow
• endothelial progenitor cells
• endothelial dysfunction
• endothelium

Introduction

Cardiovascular disease is the leading cause of death in the Western world. Cardiovascular risk factors, such as diabetes, smoking, hypercholesterolemia, arterial hypertension, and aging, damage the endothelial monolayer by inducing endothelial cell (EC) senescence and apoptosis leading to endothelial dysfunction. The increased permeability of the endothelial monolayer as well as changes of the surface and adhesion molecules result in the adhesion of thrombocytes and the invasion of inflammatory cells. The migration and proliferation of vascular smooth muscle cells (VSMC) leads to the formation of atherosclerotic plaques 1. The degree of EC apoptosis is reflected by the number of circulating endothelial microparticles (EMP). EMP numbers are increased in patients with generalized EC damage, such as thrombotic thrombocytopenic purpura (TTP) 2, diabetes mellitus 3, arterial hypertension 4, and in patients with acute coronary syndrome 5.

The restoration and reconstitution of the damaged endothelial monolayer may be a prerequisite for the prevention of atherosclerotic lesion formation and the development of restenosis after percutaneous coronary interventions 6. Until recently endothelial cell repair mechanisms were thought to be mediated by the adjacent EC. However, adult blood vessels regenerate only moderately in adults under physiological conditions. The half‐life of an adult endothelial cell has been reported to be 3.1 years 7. A growing body of evidence suggest that circulating, bone marrow‐derived endothelial progenitor cells (EPC) play an important role in EC regeneration 8,9.

In this review we focus on the evolving role of stem‐ and progenitor cells in the maintenance of the vascular wall. Besides new pathophysiological concepts of endothelial cell regeneration we will discuss new insights into vascular stem cell biology derived from experimental and clinical studies.

Key messages

• Endothelial progenitor cells (EPC) are characterized by the surface markers CD34, CD133, and vascular endothelial growth factor receptor‐2 (VEGFR2 or KDR in humans; flk‐1 in mice). CD133+/CD34‐ EPC have a higher vascular regeneration potential compared to CD133+/CD34+ EPC. Peripheral blood‐derived EPC form ‘late‐outgrowth colony‐forming units‐endothelial cells’ (CFU‐EC).

• The systemic application of healthy wild‐type EPC enhances reendothelialization after endothelial cell damage, improves endothelial dysfunction, and decreases atherosclerotic lesion progression.

• Negative impact of cardiovascular risk factors on progenitor cell function and stem cell mobilization may thwart therapeutic attempts using endogenously mobilized stem and progenitor cells.

• The number of circulating EPC is an important predictor of cardiovascular morbidity and mortality in patients with coronary artery disease.

• Endothelial cell damage leads to endothelial cell apoptosis and the release of membrane microparticles. Microparticle counts positively correlate with impairment of coronary endothelial function in humans. Further studies will need to elucidate the role of endothelial microparticles as a marker of endothelial cell apoptosis in patients with atherosclerotic disease.

Endothelial progenitor cells in health

Endothelial progenitor cell characteristics and postulated physiological role

The close regional and functional development of peripheral blood and vascular wall cells from the angioblast during embryonic development suggests the existence of a common origin, the putative hemangioblast. In 1997 Asahara and colleagues isolated an angioblast from human peripheral blood of adults which differentiate in vitro into EC and contributes to neoangiogenesis after tissue ischemia in vivo8,10. The so‐called endothelial progenitor cell (EPC) has been characterized by the surface markers CD34 and vascular endothelial growth factor receptor‐2 (VEGFR2 or KDR in humans; flk‐1 in mice). An immature subset of EPC express the surface marker CD133 11–13. The expression of CD34, VEGFR2 and CD133 is typically found on angioblasts and demonstrates the immature character of the cells. Additionally, several groups have tried to characterize peripheral blood‐derived endothelial progenies in vitro by their ability to take up acetylated low‐density lipoprotein (acLDL) and the concomitant lectin binding 10,14. The high proportion of cells derived from monocytes during such cell culture conditions has led to the definition of an additional ‘angiogenic’ cell population which may mimic progenitor characteristics 15,16. The ability of peripheral blood‐derived EPC to form ‘late‐outgrowth colony‐forming units‐endothelial cells’ (CFU‐EC) characterizes the ‘true’ EPC in vitro and is a marker for the clonogenic potential of progenitor cells 17. The phenotypic and functional characterization of circulating EPC remains unclear. Rafii and colleagues distinguish between bone marrow (BM)‐residing EPC and circulating EPC 13. In addition, it has been demonstrated that not only myelomonocytic cells 16 but also spleen‐derived mononuclear cells (MNC) and cord‐blood‐derived MNC contribute to the pool of progenies of the endothelial cell lineage. Various surface markers have been described on EPC and used for EPC characterization. Interestingly, different EPC developmental stages may have different functional properties. We have recently demonstrated that CD133+/CD34‐ EPC have a higher potential to regenerate the injured endothelium and have a higher angiogenic capacity compared to CD133+/CD34+ EPC (Figure 1) 18. In addition, EPC have been shown to differentiate in vitro not only in EC but also in VSMC 19. One may speculate that heterogeneity in cells probably reflects different developmental stages of EPC during the maturational process from the BM residual cell to the mature vascular wall cell. To give consideration to this heterogeneity, only circulating, adult, naïve CD34/KDR‐double‐positive or CD133‐positive cells measured by flow cytometry without further cell culture steps are termed endothelial progenitor cells (EPC) in this review.

Figure 1. CD133‐positive endothelial progenitor cells (EPC) characterized by the lack of CD34 represent a functionally active subpopulation of cells which are upregulated during ischemia (upper panel) and which have a significant better reendothelialization potential compared to CD133+/CD34+ EPC. Adapted from Friedrich et al. 18.

Endothelial progenitor cells in atherosclerotic disease

Endothelial progenitor cells in atherosclerosis

In healthy humans, the number of CFU‐EC in vitro is a predictor for endothelial function 20. In patients with coronary artery disease (CAD) the number of cardiovascular risk factors negatively correlates with progenitor cell counts 14. Diabetes, hyperlipidemia, hypertension and a positive family history of CAD as well as aging impairs progenitor cell number and function 21. In an experimental model of endothelial cell damage, systemic application of cultured progenitor cells enhances reendothelialization and diminishes neointima formation 22,23. In addition, the endogenous mobilization of EPC from the BM by statins increases number and function of progenitor cells, improves reendothelialization after EC damage, and diminishes restenosis. Interestingly, vasculoprotection is directly mediated by EPC. Estrogens increase EPC numbers in mice and humans, which contributes to repair mechanisms of the vascular wall (Figure 2). Furthermore, physical activity, which is known to reduce cardiovascular morbidity and mortality by mainly unknown mechanisms, increases the number and function of EPC in rodents and healthy humans. The observation that EPC directly influence lesion formation and progression comes from experimental models using progenitor cell transfusion. The systemic application of healthy wild‐type EPC in atherosclerotic ApoE‐knockout mice improves endothelial function (Wassmann et al., Circ Res, 2006) and inhibits atherosclerotic lesion progression independently of increased serum cholesterol levels 24. In another study by George et al. 25 these beneficial effects were not observed. In this study aortic sinus lesion size was significantly increased in mice receiving EPC compared with controls (34% increase in plaque area). Mice receiving EPC showed plaques with larger lipid cores, and thinner fibrous caps, and a higher number of infiltrating CD3 cells, suggesting an effect on plaque stability. An important drawback of the study is that intravenously transfused spleen‐derived cells were administered without splenectomy of the recipient animals. Our own studies reproducibly demonstrate that spleen‐derived cells have a strong tendency to home back into the organ of origin 26. However, the migration, proliferation and matrix synthesis of VSMC plays a pivotal role in atherogenesis. Various experimental studies have suggested the presence of a smooth muscle progenitor cell (SMP) 19,27. After gender‐mismatched BM transplantation and severe, experimentally induced media damage, a significant number of obliterating VSMC show a Y‐chromosome and are therefore derived from the BM donor 28. Complete endothelial cell denudation as well as induction of severe vasculopathy after heterotopic heart transplantation in genetically mismatched mice is associated with the invasion of BM‐derived VSMC. Similar results are obtained in animal models of severe atherosclerosis 27. Atherosclerotic plaques of hyperlipidemic apolipoprotein E deficient (ApoE knockout) mice display a significant number of BM‐derived VSMC 27.

Figure 2. EPC were assessed in sham‐operated (Sham), ovariectomized (Ovarex), and ovariectomized mice with concomitant estrogen replacement treatment (Ovarex+E) by flow cytometry analysis to quantify Sca‐1/VEGF‐R2 positive EPC. A: Quantitative analysis of circulating numbers of Sca1 and VEGF‐R2 positive cells in peripheral blood (mean±SEM, n = 8, *P<0.05). B: Quantitative analysis of Sca1/VEGF‐R2 positive cells in the bone marrow (mean±SEM, n = 8, *P<0.05). Adapted from Strehlow et al. 31. (EPC = endothelial progenitor cells; Sca‐1 = stem cell antigen‐1; VEGF‐R2 = vascular endothelial growth factor receptor‐2.)

The in vitro differentiation of hematopoietic stem cells into VSMC together with the isolation of cells from peripheral human blood which can be differentiated into VSMC suggest the existence of a BM‐derived SMP. Until today, the vast majority of studies have demonstrated that severe damage of the media is a prerequisite for the invasion of SMP 28 whereas after isolated EC damage without significant injury of the media no invasion of SMP has been noted. Potentially, a circulating angioblast with differentiation potential into EC or VSMC (depending on various stimuli) exists. This notion definitely requires further attention. First indications that progenitor cells have a higher plasticity than initially thought come from in vitro experiments 29 in which cocultivation of EPC together with cardiomyocytes resulted in the potential transdifferentiation of EPC into cardiomyocytes 29,30.

In conclusion: The vast majority of current observations demonstrate that exogenous application of healthy EPC regenerates the endothelial monolayer, improves endothelial dysfunction, and prevents the progression of atherosclerotic lesion formation at least in experimental models. However, the direct proof that endogenous stem and progenitor cells actively contribute to the rejuvenation of the endothelial monolayer, independently of a mobilizing therapy, is still lacking. According to the current knowledge one has to expect that the functional properties of EPC in cardiovascular disease are impaired and that regeneration by endogenous cells without further mobilization of cells is diminished or absent.

Mobilization and regulation of endogenous endothelial progenitor cells in cardiovascular disease

Ischemic tissue damage after myocardial infarction, coronary‐aortic bypass grafting, or severe burns lead to a significant mobilization of EPC 11 suggesting a biological role of progenitor cells in pathological conditions. In parallel to the 50‐fold increase in CD133‐ and VEGFR2‐positive cells in peripheral blood, VEGF plasma concentrations significantly rise. The experimental induction of hindlimb ischemia in mice increases circulating EPC and enhances neoangiogenesis 10. EPC mobilization, tissue regeneration and neoangiogenesis have been described after treatment with G‐SCF (granulocyte colony stimulating factor) and stem cell factor (SCF) 10, stromal‐derived factor‐1 (SDF‐1), angiopoietin, placenta growth factor (PlGF), erythropoietin, estrogens as well as hydroxymethylglutaryl‐coenzyme A (HMG‐CoA) reductase inhibitors and physical activity (Figure 3) 31–36. In contrast, impaired mobilization of EPC has been associated with older age, the presence of cardiovascular risk factors, and the presence of atherosclerotic disease 14 as discussed above. The molecular mechanisms for an effective mobilization of stem cells are poorly understood. Nitric oxide (NO) within the bone marrow plays a significant role in the complex mobilization of stem and progenitor cells from the quiescent bone marrow niche into peripheral blood 37. Furthermore, MMP‐9 and c‐kit play a pivotal role in stem cell mobilization 38. We have recently demonstrated that fibroblast growth factor‐2 (FGF‐2) is essential for estrogen‐mediated mobilization of EPC. FGF‐2‐/‐ mice show a severe defect in estrogen‐mediated progenitor cell mobilization, whereas reconstitution of FGF‐2‐/‐ mice with wild‐type bone marrow results in the complete restoration of the peripheral blood progenitor cell pool suggesting that a variety of factors influence mobilization of the endogenous pool of stem and progenitor cells (Fontaine et al., Am J Pathol, 2006). First clinical studies using granulocyte colony‐stimulating factor (G‐CSF) mobilization therapies in myocardial infarction have shown divergent but mainly disappointing results 39–41. Further studies will have to elucidate the complexity of stem cell mobilization in order to use therapeutically the endogenous cell pool. A major drawback in this context may be the fact that the negative impact of cardiovascular risk factors, not only on progenitor cell function but also on stem cell mobilization, may thwart therapeutic attempts.

Figure 3. Endogenous stem cell mobilization requires the active recruitment of stromal cell‐bound stem cells mediated by proteases to the vascularized zones of the bone marrow. Stem cells then interact with the bone marrow endothelial cells to mobilize activated stem and progenitor cells to the peripheral blood. Various agents have been shown to effectively influence stem cell mobilization. (5‐FU = 5‐fluorouracil; CXCR‐4 = chemokine (CXC motif) receptor‐4; eNOS = endothelial nitric oxide synthase; FGF‐2 = fibroblast growth factor‐2; G‐CSF = granulocyte colony‐stimulating factor; GM‐CSF = granulocyte‐macrophage colony stimulating factor; ICAM‐1 = intercellular adhesion molecule‐1; LFA‐1 = leukocyte functional antigen‐1; mAb = monoclonal antibody; mKitL = membranous kit ligand; MMP‐9 = matrix metalloproteinase‐9; NO = nitric oxide; PlGF = placenta growth factor; SCF = stem cell factor; SDF‐1 = stromal derived factor‐1; sKitL = soluble kit ligand; VCAM‐1 = vascular cell adhesion molecule‐1; VEGF = vascular endothelial growth factor; VLA‐4 = very late antigen‐4.)

Prognostic role of endothelial progenitor cells in patients with coronary atherosclerotic disease

In order to evaluate the prognostic value of circulating EPC and their potentially vasculoprotective role, we performed the Endothelial Progenitor Cells in Coronary Artery Disease (EPCAD) study in which the number of CD34+/KDR+ EPC was measured in 519 patients with angiographically documented CAD and correlated with cardiovascular outcomes. Primary end points included cardiovascular mortality, the occurrence of a first major cardiovascular event (myocardial infarction, hospitalization, revascularization, and cardiovascular death), revascularization, hospitalization, and all‐cause mortality after 12 months. The cumulative event‐free survival increased stepwise across tertiles of baseline EPC levels for cardiovascular mortality, first major cardiovascular event, revascularization, and hospitalization (Figure 4). After adjustment for vascular risk factors, drug therapy, and concomitant disease, increased EPC levels were independently associated with a lower risk of cardiovascular death (P = 0.001), first major cardiovascular event (P = 0.002), revascularization (P = 0.017), and hospitalization (P = 0.012). Similar results were obtained when using CD133‐positive EPC or the number of colony‐forming units endothelial cells (CFU‐EC) as a marker of the clonogenic potential of formerly circulating EPC 42.

Figure 4. Circulating endothelial progenitor cells are predictive for the occurrence of a first major cardiovascular event in patients with coronary artery disease. The risk for the combined end point is reduced in a step‐wise fashion with increasing endothelial progenitor cell (EPC) levels. Adapted from Werner et al. 42.

Endothelial cell apoptosis, microparticles and cardiovascular disease

The physiological functioning of the endothelial monolayer is a prerequisite for the prevention of atherosclerotic lesion development. Endothelial cell damage mediated by chemical or mechanical injury leads to endothelial cell apoptosis which is associated with conformational changes of the cell's plasma membrane leading to the release of membrane microparticles. Microparticle sizes range from 0.1 to 1.5 µm and derive from platelets, monocytes, erythrocytes, granulocytes, lymphocytes or endothelial cells. Microparticles express antigens derived from their mother cell and the negatively charged phosphatidylserine which is normally exclusively located in the inner, cytoplasmic membrane but becomes surface‐exposed after cell activation or apoptosis. Accordingly, endothelial cell‐derived microparticles (EMP) can be quantified in vivo by flow cytometry 43 using an endothelial cell‐related surface marker in combination with annexin V which specifically binds to phosphatidylserine (Figure 5). In all conditions of severe endothelial cell damage elevated EMP levels have been described (e.g. thrombotic thrombocytopenic purpura, diabetes 3, arterial hypertension 4, acute coronary syndromes 44, and myocardial infarction 45). Microparticles themselves have direct effects on endothelium‐dependent vasorelaxation in vitro. Microparticles derived from patients with acute coronary syndromes or preeclampsia directly impaired endothelial function in rat aortic rings or myometrial arteries 45,46. In humans, increased apoptotic microparticle counts positively correlate with impairment of coronary endothelial function 47. In a study investigating coronary endothelial function in 50 patients with CAD, multivariate analysis revealed that increased apoptotic microparticle counts predict severe endothelial dysfunction independent of classical risk factors such as hypertension, hypercholesterolemia, smoking, diabetes, age, and gender 47. In the context of human atherogenesis, it may be pivotal to evaluate the current status of regeneration and endothelial cell apoptosis in each individual. EPC and EMP may be potentially attractive markers for risk stratification in patients with atherosclerotic disease.

Figure 5. Endothelial cell apoptosis is associated with the release of small membrane vesicles, the so‐called endothelial microparticles (EMP). Microparticle size ranges from 0.1–1.5 µm and derive from platelets, monocytes, erythrocytes, granulocytes, lymphocytes or endothelial cells. They express antigens derived from their mother cell and the negatively charged phosphatidylserine, which is normally exclusively located in the inner, cytoplasmic membrane but becomes surface‐exposed after cell activation or apoptosis. EMP can be quantified in peripheral blood by annexin V binding in combination with an endothelial cell marker, e.g. CD31 or vascular endothelial (VE)‐cadherin. P = 0.002 severe versus control P = <0.005. Modified from Sabatier et al. and Preston et al. 3,4. * acute coronary syndrome.

Conclusion and perspectives

BM‐derived EPC play a crucial role in (neo)angiogenesis of ischemic tissue. In addition, EPC contribute to the rejuvenation of the endothelial monolayer after EC damage (Figure 6). The effective regeneration of the endothelial monolayer might be a prerequisite for the prevention of atherosclerotic lesion formation. The ability to measure EC apoptosis on the one hand, using EMP shed from apoptotic EC, and the determination of the regenerative potential on the other hand, using number and function of EPC, may give a detailed insight into vascular wall homeostasis. The effectiveness of therapeutic approaches, including life‐style modifications and vasculoprotective drug therapies, might be monitored using the EPC and EMP balance. In conditions of severe apoptosis, risk factor modification and/or inhibition of apoptosis may be superior compared to conditions in which impaired regeneration is responsible for lesion formation. Evidence that BM‐derived EPC contribute in a sufficient number to vascular wall regeneration comes from a computational model 48,49. The authors demonstrate that, in conditions of (oxidative) stress, localized disturbance of the integrity of the endothelial monolayer by the simulated age of 65 was prevented by progenitor cell homing. A homing rate of 5% per year was sufficient to significantly delay defects of the EC integrity. These computer simulations together with the experimental findings underline the multiple facets of progenitor cells in angiogenesis and vascular wall homeostasis. Further studies are urgently needed to clarify the role of progenitor cells in cardiovascular disease.

Figure 6. EPC contribute to the rejuvenation of the endothelial monolayer after endothelial cell damage. The effective regeneration of the endothelial monolayer might be a prerequisite for the prevention of atherosclerotic lesion formation. Cardiovascular risk factors negatively influence EPC number and function while the vast majority of cardioprotective agents mediate their action at least in part by positively influencing EPC. The pool of EPC consists of a heterogeneous population of cells which may interact in concert in the mediation of endothelial healing.(EPC = endothelial progenitor cells; G‐CSF = granulocyte colony‐stimulating factor; MNC = mononuclear cells; SCF = stem cell factor; VEGF = vascular endothelial growth factor.)