Developmental Biology of Lung Stem Cells

Introduction

Insights into the biology and function of stem cells have revealed their therapeutic potential in the treatment of numerous diseases, including diseases of the lung. ^, Multiple studies to date have shown protective effects of mesenchymal stromal cells (MSCs) or endothelial progenitor cells (EPCs) derived from bone marrow (BM) or umbilical cord (UC) in animal models of neonatal lung diseases. The discovery of resident stem/progenitor cells in adult tissues led to the development of a new field of investigation into the role of resident stem/progenitor cells in lung development, injury, and repair. ^{, ,}

Stem cells are identified by (1) their ability to self-renew (remain in an undifferentiated state) and (2) their potency (ability to produce differentiated cells). According to their ability to produce differentiated somatic cells, stem cells are characterized in four categories: (1) totipotent cells of zygote and morula, which can give rise to all embryonic and extraembryonic tissues; (2) pluripotent cells of the blastula capable of forming any cell type in the developing embryo; (3) multipotent cells, which are often found in adult organisms (e.g., hematopoietic stem cells) and can give rise to several cell types; and (4) unipotent cells, which can only differentiate into one cell type (e.g., alveolar epithelial type 2 cell [AECII] differentiating into type 1 cell [AECI]). Somatic stem cells (also called adult stem cells ) are typically multipotent or unipotent. While progenitor cells can also give rise to one specific type of cells, they are thought to lack the capacity to self-renew. Therefore they are typically distinguished from the population termed stem cells .

Due to their low turnover, the lungs are relatively quiescent organs. This is particularly true in more distal alveolar regions, as cellular turnover time decreases along the proximal-distal axis. Lung somatic stem cells are therefore activated through their microenvironment, especially following tissue damage, and participate in tissue repair and remodeling. Injurious stimuli, however, can also cause substantial damage to resident stem cells. Such impacted cells may then undergo an unusual differentiation process, followed by the proliferation of abnormal cell populations and even tumorigenesis. In fact, the hypothesis that the proliferation of stem cells can drive carcinogenesis (so-called cancer stem cell hypothesis) has been widely accepted as one of the underlying principles of cancer research and therapy. Alternatively, alterations to stem cells can result in increased or decreased production of a particular differentiated cell population leading to changes in tissue composition. Such mechanisms contribute, for example, to lung fibrosis in bronchiolitis obliterans syndrome following lung transplantation or loss of tissue in chronic obstructive pulmonary disease, respectively. Thus, impairment of somatic lung stem cells and progenitors and disturbance in their environmental niche may contribute to several pathologies in the developing lung.

To date, multiple populations of stem cells within the pulmonary endothelium, epithelium, and mesenchyme have been described. Putative epithelial stem or progenitor cells, such as basal cells of proximal airways, secretory cells of conducting airways, and bronchial alveolar stem cells (BASCs) have been extensively studied. However, relatively little is known by comparison about mesenchymal and endothelial stem cells of distal alveolar regions ( Fig. 74.1 ).

Endothelial Progenitor Cells

The pulmonary vascular tree is formed by two processes of vessel formation: vasculogenesis (de novo vessel formation) and angiogenesis (formation of new vessels from pre-existing vessels). ^, The relative contribution and timing of the two processes during individual stages of lung development remain unclear, and two main hypotheses are currently considered: (1) proximal/distal angiogenesis model and (2) proximal angiogenesis/distal vasculogenesis model. The formation of the vascular tree during the pseudoglandular and canalicular stages is followed by the process of microvascular maturation. During the saccular stage, capillaries are organized in bilayers within the relatively thick intersaccular space. Progressive thinning of intersaccular (interalveolar) walls during the alveolar stage then leads to the remodeling of the double capillary layer and the formation of the single capillary layer.

In a broader sense, EPCs represent a very heterogeneous and rare population of cells in peripheral and cord blood. Typically, EPCs have been described as cells able to differentiate into endothelial cells and form new blood vessels. Expression of CD31 and vascular endothelial growth factor receptor 2 (VEGFR), as well as acetylated low-density lipoprotein (AcLDL) uptake and lectin binding, are often used as means to identify circulating EPCs in vitro. A proposed nomenclature differentiates between myeloid angiogenic cells (MACs) and endothelial colony-forming cells (ECFCs), representing two separate cell types promoting vascular repair by two distinct mechanisms of action. MACs are characterized as CD45 ⁺ /CD14 ⁺ /CD31 ⁺ and CD146 ⁻ /CD133 ⁻ /Tie2 ⁻ cultured cells derived from peripheral blood mononuclear cells. While MACs do not have the capacity to become endothelial cells, they promote angiogenesis through paracrine mechanisms. ECFCs, also known as late outgrowth endothelial cells, possess potent intrinsic angiogenic capacity and act as a source of angiogenic paracrine factors. They are characterized by (1) their high proliferative potential, (2) the ability to form secondary and tertiary colonies in vitro, and (3) the de novo blood vessels in vivo. The presence of resident EPCs has been demonstrated in rodent and human lungs.

An ever-increasing number of studies reinforce the notion that blood vessel formation actively drives pulmonary growth and constitutes a critical aspect of alveolarization. The hypothesis that angiogenesis drives alveolarization is for the most part founded on two experimental and clinical observations: (1) defects of vascular organization and production of angiogenic factors in patients with underdeveloped lungs suffering from bronchopulmonary dysplasia (BPD), as well as in animal models of BPD, ^, and (2) studies of angiogenesis inhibition in animal models of lung development. This was best demonstrated by studies where the crucial angiogenic factor, vascular endothelial growth factor (VEGF), was pharmacologically or genetically inhibited. ^, Disruption of VEGF resulted in arrested alveolarization in the lungs of developing rats and in emphysema-like enlargement of airspaces in adult rats, ^, indicating its importance in development and maintenance of the alveolar architecture. Conversely, enhancing angiogenesis during the critical developmental period had partially rescued stunted alveolar formation in experimental models of BPD. ^, These observations suggest the existence of EPC populations in the distal pulmonary region.

EPCs in the developing lung are poorly understood ( Table 74.1 ; see Fig. 74.1 ). Multiple reports of resident EPCs in developing mouse, rat, and human lung have been published over the past decade; however, these studies are often limited to in vitro characterization of the cells, and little is known about their location and active role in the developing lung.

In general, pulmonary EPCs population comprises populations of (1) lung resident (LR) and (2) circulating vascular progenitors expressing characteristic cell-surface markers, able to proliferate and form colonies in vitro and to form de novo vessels in vitro and in vivo . ^, Unfortunately, no cell surface markers have yet been established to distinguish between the resident and circulating population, making it particularly challenging to determine their individual functions in tissue development and repair. ^,

Most authors identify EPCs as CD31 ⁺ , CD34 ⁺ , CD144 ⁺ , and CD45 ⁻ cells. ^, Initial isolation of the cells, however, often relies on a single surface marker, CD31, with subsequent selection for EPCs during the in vitro culture. This practice is rooted in the notion that only cells with “stem-cell-like” characteristics will survive, proliferate, and form colonies in culture. One study characterized such cells isolated from human fetal lungs (17 to 20 weeks of gestation) and developing rat lungs (postnatal day [P]14). In addition to CD31, cells surviving in culture expressed CD105 (endoglin), CD144 (VE-cadherin), von Willebrand factor (vWF), and VEGFR2 and formed colonies and tube formations in vitro. Furthermore, when transplanted subcutaneously to mice in form of cell/matrix plugs, these cells formed de novo vessels, demonstrating their progenitor capacity.

To better understand the nature of EPCs in their study, Alvarez and colleagues performed separate assessments of pulmonary artery endothelial cells (PAECs) and pulmonary microvascular endothelial cells (PMVECs) in adult rats. While PAECs were isolated directly by manually scraping the intima of pulmonary arteries, PMVECs were isolated by bulk digestions of distal pulmonary regions. Interestingly, while the majority of PAECs were fully differentiated, three-fourths of PMVECs proliferated in culture, with 50% giving rise to large colonies. These cells, dubbed by authors as resident microvascular endothelial progenitor cells (RMEPCs), expressed CD31, CD34, CD144, CD309, CD105, VEGFR2, vWF, and endothelial nitric oxide synthase (eNOS), while lacking the expression of CD45. Similarly to other studies, these cells formed de novo vessels following transplantation in vivo .

In addition to vascular endothelial progenitors, in their study, Schniederman and colleagues identified a subpopulation of progenitors of lymphatic endothelium. Murine lung CD31 ⁺ microvascular endothelial cells were cultured and assessed for colony formation and expression of endothelial and progenitor-specific cell antigens. All cells were CD31, CD34, CD105, VEGFR2, and CD144 positive and CD45, CD41, and CD117 negative. However, in addition, a considerable portion of cells also expressed Lyve1 , Prox1, and Vegfr3 , indicating their commitment to lymphatic endothelium. In fact, when these cells were transplanted in Matrigel plugs to recipient mice, de novo formation of both blood and lymphatic vessels was observed.

The pivotal role of EPCs during lung development is further supported by studies of arrested lung development, particularly BPD, where both alveolar and vascular development are halted. A decrease in the number of both circulating and resident EPCs, likely MACs, was observed in a murine model of BPD. Furthermore, ECFCs isolated from lungs of diseased rats exhibited reduced cell growth, formed less colonies, and had less complex tubular networks in vitro .

More recently, several novel EPC-specific markers were proposed. BST1(CD157) was described as an endothelial stem cell marker in multiple mouse organs, including the lung. BST1 ⁺ /CD31 ⁺ /CD45 ⁻ cells constituted approximately 5.5% of all lung cells in mice and were localized within large vessels. However, these cells were absent from capillaries. Another proposed population of vascular endothelial stem cells, PROCR ⁺ cells, was identified in multiple organs, including the mammary gland—the only other organ that undergoes the process of alveolarization. PROCR ⁺ cells showed bipotent character and gave rise de novo to both endothelial cells and pericytes. However, no further studies on the nature of these cells in the developing lung exist to date. Additional markers of pulmonary EPCs, c-KIT, and FOXF1 were proposed by Ren and colleagues. A population of c-KIT ⁺ cells could be detected as early as E16.5 and represented approximately 50% of lung endothelial cells until birth (P1). By 2 weeks of life (P14), the number of c-KIT ⁺ endothelial cells decreased by 20%; this decrease was more severe in animals with hyperoxia-induced arrest in alveolarization. Furthermore, the numbers of both c-KIT ⁺ and FOXF1 ⁺ cells were lower in the lungs of BPD patients. Room-air housed transgenic mice with endothelial cells-specific deletion of c-Kit or Foxf1 displayed alveolar simplification. Similarly, FOXF1 haplo insufficient (Foxf1 ^+/− ) developed spontaneous alveolar hypoplasia and exhibited worse alveolar structure after exposure to hyperoxia, when compared with wild-type mice. Alveolar simplification in both wild-type and Foxf1 ^+/− hyperoxia-exposed animals could be reversed by treatment with lung-derived c-KIT ⁺ endothelial cells. These findings are in agreement with previous work by the same authors, demonstrating that FOXF1 is required for the formation of embryonic vasculature, endothelial proliferation, and VEGF signaling.

A growing body of evidence points to the existence of resident pulmonary EPCs and their contribution to normal lung development and repair. New technologies will enable a better characterization of these cells to improve our understanding of normal lung development and disease. This, in turn, may lead to novel therapies to either protect resident EPCs from injury or provide exogeneous EPC for lung regeneration.