Neovascular capacity of endothelial progenitor cells in the adult pulmonary circulation

pulmonary postnatal neovascularization or formation of de novo vessels in the adult lung has been controversial. Most of the argument has been focused on findings in the systemic circulation, which exhibits diverse degrees of angiogenesis (vessel formation from preexisting vessels) and arteriogenesis (collateral development proximal to a site of arterial occlusion). With the advent of stem cell plasticity, neovascularization was demonstrated to occur postnatally in skeletal and cardiac muscle, peripheral vasculature, adipose tissue, and the endometrium and during pathological conditions such as expansion of solid tumors. Furthermore, remodeling of the tunica intima and media is observed in the systemic circulation during atherosclerosis. Interestingly, parallel angiogenic responses are not frequently observed in the pulmonary circulation leading early investigators to theorize that the pulmonary circulation was quiescent, lacking the capacity to generate de novo vessels or undergo vascular remodeling. Moreover, during focal tissue hypoxia in the setting of idiopathic pulmonary arterial hypertension (IPAH), PAH secondary to fibrosis, or chronic obstructive pulmonary disease (COPD), the native pulmonary vasculature experiences a series of perturbations, which results in the loss, instead of the formation, of vessels.

In 1847, Virchow (17) observed that under regional ischemia in the human lung, the systemic circulation restored circulation to nonperfused areas. From early studies, it was clear that the ischemic lung depends on the adjacent systemic circulation to provide necessary collateral circulation. More recently, neoangiogenesis by systemic vessels in the pulmonary artery vaso vasorum has been shown to occur in response to a variety of physiological and pathological stimuli (6). Despite the compelling support indicating that neovascularization of the pulmonary circulation occurs by extension of the systemic circulation, there are many studies supporting mechanisms that do not involve contribution of the systemic vasculature. Patients with scleroderma-induced vasculitis (4), pulmonary venoocclusive disease (15), idiopathic interstitial pneumonia (14), and acute respiratory distress syndrome (ARDS) survivors are important examples of lung neoangiogenesis that occur independent of the contribution from the systemic circulation. Moreover, experimental models of pneumonectomy in rats (9), caloric restriction in mice (13), intermittent levels of chronic hypoxia in rats (8), and lung tumorigenesis in dogs (11) further support that pulmonary neoangiogenesis arises in a systemic vasculature-independent manner. Although it is clear that neoangiogenesis and vascular remodeling develops within the pulmonary circulation, the origin of the cellular constituents that facilitate or contribute to this process is inconclusive. Recent evidence implicates several progenitors with the potential to engineer a proangiogenic environment or induce de novo vessel formation, including resident lung vascular cells (1, 10, 21), circulating bone marrow-derived cells such as endothelial progenitors (cEPCs; Ref. 16), and mesenchymal progenitors (fibrocytes; Ref. 7).

Consistent with the concept that vascular progenitor cells arise from the bone marrow, pioneering work by Asahara et al. (2) resolved that isolated cells from the mononuclear fraction of the peripheral blood regulated neoangiogenesis in an ischemic experimental model. The cells identified by Asahara et al. (2) have been further characterized by their expression of the membrane proteins CD34, VEGFR2 (Flk-1), and CD133. In the lung, vascular progenitor cells are hypothesized to be recruited from the bone marrow or to exist within the tunica intima of blood vessels and capillaries (1), within the tunica adventitia of larger vessels (21), and within the tissue parenchyma (10). In agreement with the presence of circulating progenitor cells in the lung, patients with PAH demonstrate higher numbers of circulating CD34/CD133^pos cells compared with healthy controls. The isolated CD34/CD133^pos cells revealed a phenotype with enhanced ability to form angiogenic networks in vitro and displayed hyperproliferative, apoptosis-resistant behavior ex vivo (3, 12). In separate studies, increased numbers of CD133^pos cells were found within the pulmonary vasculature in humans with PAH and in vessels of a BMPR2^R899X transgene murine model of PAH (19). The level of circulating mesenchymal progenitor CD45-/collagen I-positive fibrocytes has been shown in PAH to correlate with a poor prognosis, indicating that circulating vascular and mesenchymal progenitors played a role in the pathogenesis of pulmonary vascular diseases.

Disruption of pulmonary vascular flow such as that occurring during chronic pulmonary thromboembolism (CPT) has germane seminal contributions to the emergence of neoangiogenesis and remodeling within the occlusive lesion. Lung specimens obtained from patients with CPT reveal that organized thromboembolic lesions remodel into fibrotic tissue, instead of self-resolving, reminiscent of plexiform lesions (5). Interestingly, the fibrotic lesions often exhibit mild inflammatory infiltrates and display recanalized or vascularized lumens (5). These observations are important considering that in the lung, cellular inflammatory constituents regulate neoangiogenesis and neointimal remodeling. In a mouse model of chronic pulmonary artery ligation, lymphocytes were restrictive, whereas monocytes/macrophages promoted neovascularization (18). Based on the evidence found in the literature, it is plausible that the neoangiogenesis and remodeling processes occurring in the CPT lesions may be impacted by the contribution of vascular progenitors that arise from the circulation and from resident vascular cell populations, with their activity coordinated by cellular inflammatory mediators.

In this issue of AJP-Lung, Yao et al. (20) provide further evidence that CD133/CD34 and procollagen I/α-smooth muscle actin-expressing cells in combination with CD45-positive inflammatory cells play a role in neovascularization and remodeling of thromboembolic occluded vessels. To assess a role for putative progenitors in the development of CPT, the authors examined the location of CD133/Flk-1, CD34/Flk-1, CD45, or procollagen I/α-smooth muscle actin-positive cells in the endarterectomized tissues as well as the vascular wall tissues adjacent to the thromboemboli from patients with CPT. Through this detailed characterization, the authors demonstrate that, although rare, subpopulations of CD133- or CD34-positive cells were specifically localized to the distal region of the vessel wall where recanalization/neovascularization was present. In contrast, procollagen I-/α-smooth muscle actin-positive cells were localized in the vessel wall proximal to the thromboemboli. Collectively, the data suggest that progenitors expressing CD133 and collagen I-/α-smooth muscle actin-expressing fibrocytes play a role in remodeling CPT lesions. In the current issue, Yao et al. (20) have broadened our knowledge of cellular constituents involved in the pathogenesis of PAH with regard to CPT patients. The study by Yao et al. (20) offers evidence of the complex assortment of cellular relationships that take place in established pulmonary vascular lesions. Although it was clear that neovascularization occurred, the heterogeneous mosaic of cells within the fibrous tissue opens a door for a new series of experiments addressing their individual roles in the process. It is now important to recognize that rigorous characterization and isolation of those cells is needed to understand their basic biology. It is evident that the interplay of recruited circulating inflammatory cells and circulating vascular and mesenchymal progenitors with resident progenitor cells is critical in the generation of de novo vessels. The observations by Yao et al. (20) allow the field to merge our current understanding of how both circulating and resident progenitor cells contribute either to the development of pathogenic processes or, as appears in the case of CPT, to respond to preexisting noxious events within the pulmonary circulation (e.g., contributing to repair).

Several aspects of progenitor biology in the context of vascular remodeling need to be considered. For instance, it is important to define the signals that orchestrate recruitment and activity of the specific progenitors involved. In the lung, the chemoattractant mechanisms for inflammatory cells are known; however, such observations are missing for circulating vascular and mesenchymal progenitor cells. Since circulating CD133-/CD34-/Flk-1-positive cells are incriminated in neoangiogenesis but may lack the ability to directly form de novo vessels, it is intriguing to know the function of these cells. Considering that three different resident progenitor cells with a neoangiogenic capacity have been revealed, including those existing in the intima of conduit and microvascular pulmonary circulation (1), in the adventitia of the vessel wall (21), and within the lung parenchyma (10), it becomes fundamental to understand which phenotype(s) are directly incriminated in CPT-resulting neovascularization. The study by Yao et al. (20) has created an opportunity to resolve the molecular signatures that result from the signals to and from the outlined circulating and resident bona fide endothelial progenitor cells.