Endothelial progenitor cells: unexpected disclosures.

The formation of coronary collateral vessels constitutes a compensatory adaptation of the heart to ischemia. This vascular response increases blood supply to the myocardium reducing the area of hypoxic damage and, ultimately, infarct size. The recruitment of collateral vessels within the coronary circulation is favored by multiple episodes of ischemia, which provide the stimuli necessary for the growth of vascular structures and increased blood flow. The degree of collateralization significantly influences the prognosis of an ischemic episode, which is more severe with diabetes, metabolic syndrome, or in the elderly, the patient population mostly affected by coronary artery disease.1 Originally, angiogenesis, ie, sprouting of vessels from the preexisting circulation, was considered the exclusive mechanism of vascular remodeling and growth. The identification of circulating endothelial progenitor cells (EPCs) has modified this view.2 Migration and homing of EPCs to ischemic regions leads to de novo formation of vascular structures. In analogy with vessel development in the embryo, this process has been called vasculogenesis. Also, circulating EPCs participate in reendothelization of damaged vessels, together with resident vascular progenitor cells.3 EPCs represent a heterogeneous cell population of multiple origins and distinct phenotypes, which are able to give rise to functionally competent endothelial cells.4 In the hierarchy established in the hematopoietic system, progenitors identify cells with lower differentiation potential than stem cells (SCs). However, EPCs possess degrees of stemness, which include self-renewal, clonogenicity, and differentiation capacity.4 EPCs represent various subsets of progenitor cells that have distinct phenotypes but share the ability to differentiate into mature endothelial cells. During embryogenesis, hematopoietic cells and endothelial cells develop temporally and spatially in close association. Blood cells are surrounded by layers of endothelial cells suggesting a joint origin of these lineages from a common precursor, the embryonic hemangioblast.4 The adult hemangioblast resides in the bone marrow (BM), and hematopoietic stem cells (HSCs) are considered the major source of circulating EPCs. Moreover, EPCs can derive from cells of the myeloid/monocytic lineage and from BM mesenchymal stromal cells. The recognition of vascular progenitors in the vessel wall, adipose tissue, brain, and heart suggests that EPCs are released from different sites into the peripheral circulation.4,5 In the absence of tissue-related markers, it is impossible to establish the relative contribution of different organs to the circulating pool of EPCs. It cannot be excluded, however, that EPCs are stored in the bone marrow and, when the need arises, take residence in distant organs (Figure 1). A specific phenotype of EPCs has not yet emerged. In spite of the identification of tissue-resident immature endothelial cells, the prevailing view remains that HSCs are the ancestors of EPCs. And the presence of EPCs in organs other than the bone marrow has been interpreted as the result of trafficking of cells from the BM through the bloodstream or preservation of HSC-derived embryonic remnants.4,5 Accordingly, markers of EPCs have been searched for among HSC epitopes. The historic antigen of EPCs, CD34, was identified in 1997 and this work promoted the study of circulating angioblasts.2 Subsequently, other markers were detected including CD133, an immature HSC epitope, and CD14, an indicator of monocyte lineage. The coexpression of the VEGF receptor-2, flk1, with CD34 and CD14 is a fundamental attribute of EPCs.4,5 In this issue of Circulation Research, Romagnani and collaborators6 report on a novel population of human peripheral blood mononuclear cells (PBMCs). PBMCs were identified as a subset with a uniform, bright presence of CD14 and flk1 accompanied by low expression of CD34. This epitope was undetectable by FACS and apparent only with a highly sensitive technique. This approach consisted of a combination of immunofluorescence and immunomagnetism. Liposomes were loaded with fluorescein and magnetic beads, whereas CD34 antibodies were conjugated to the liposomes. By this technique, staining intensity increased 100to 1000-fold, whereas background fluorescence was unchanged.6 Because of magnetic labeling, cells with low-level antigens were isolated by magnetic sorting, and CD14CD34 cells were identified in the peripheral blood and BM. Contrary to expectation, these cells were clonogenic and multipotent generating, in addition to mature endothelial cells, osteoblasts, adipocytes, and neural cells. Growth and differentiation of CD14CD34 cells required the presence of VEGF. Thus, CD14CD34 cells possess the criteria of SCs and give rise to cells from different germ layers comprising the mesoderm and ectoderm. The level of enrichment of CD14CD34 cells, 95%, indicates that commitment to adipocytes, osteoblasts and neural cells occurred by transdifferentiation. The possibility that fusion between CD14CD34 cells and CD14CD34 cells took place is unlikely. CD14CD34 cells constituted only The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association. From the Cardiovascular Research Institute, Department of Medicine, New York Medical College, Valhalla. Correspondence to Dr Annarosa Leri, Cardiovascular Research Institute, Vosburgh Pavilion, New York Medical College, Valhalla, NY 10595. E-mail [email protected] (Circ Res. 2005;97:299-301.) © 2005 American Heart Association, Inc.