Angiogenesis and lymphangiogenesis in bronchial asthma


  • Edited by: Hans-Uwe Simon

Gianni Marone, MD, FRCP, Department of Clinical Immunology and Allergy, and Center for Basic and Clinical Immunology Research (CISI), University of Naples Federico II, Via S. Pansini 5, 80131 Napoli, Italy.
Tel.: +39 (081) 7707492
Fax: +39 (081) 7462271


To cite this article: Detoraki A, Granata F, Staibano S, Rossi FW, Marone G, Genovese A. Angiogenesis and lymphangiogenesis in bronchial asthma. Allergy 2010; 65: 946–958.


Neovascularization plays a prominent role in inflammation and tissue remodeling in several chronic inflammatory disorders. Vessel number and size, vascular surface area and vascular leakage are all increased in biopsies from patients with asthma. High levels of VEGF and other angiogenic factors have been detected in tissues and biological samples of patients with asthma and correlate with disease activity and inversely with airway hyper-responsiveness. Inflammation in the lung stimulates the growth of new blood vessels and these contribute to the airway obstruction or airway hyper-responsiveness, or both. Effector cells of inflammation (human lung mast cells, basophils, eosinophils, macrophages, etc.) are major sources of a vast array of angiogenic and lymphangiogenic factors. Inhaled corticosteroids reduce vascularity and growth factor expression and might modulate bronchial vascular remodeling in asthma. Specific antagonists to VEGF and other angiogenic factors and their receptors might help to control chronic airway inflammation and vascular remodeling and offer a novel approach for the treatment of chronic inflammatory lung disorders.




chorioallantoic membrane


basic fibroblast growth factor-β


human lung mast cells


placental growth factor


vascular endothelial growth factor


vascular endothelial growth factor receptor

Basic concepts of angiogenesis and lymphangiogenesis

Angiogenesis is a term coined by John Hunter in 1787 to describe the growth of new vessels. This concept, however, has fascinated many scientists for centuries. For instance, Galen proposed that blood is locally regenerated by the body when its supplies are consumed (1), and Leonardo da Vinci speculated that the vasculature developed from the heart, like a tree from the seed. In 1628, William Harvey discovered that the heart pumps the blood through the arteries and that veins return the blood to the heart (1). In 1661, Marcello Malpighi identified the capillaries. At the same time, Caspar Aselius discovered the lymphatic vessels (1).

Blood vessels arise from endothelial precursors that share an origin with hematopoietic progenitors (1). These progenitors assemble into a primitive vascular labyrinth of small capillaries, in a process known as vasculogenesis. During the angiogenesis phase, the vascular plexus progressively expands by means of vessel sprouting and remodels into a highly organized vascular network of larger vessels ramifying into smaller ones. Nascent endothelial cell (EC) channels become covered by pericytes (PCs) and smooth muscle cells (SMCs), which provide strength and allow regulation of vessel perfusion; this is arteriogenesis.

The lymphatic system develops in parallel but secondarily to the blood vascular system through a process known as lymphangiogenesis (2). Blood vessels arose early in evolution whereas lymph vessels are present only in amphibians onwards (3). Specific markers for lymphatic endothelial cells are now available to identify and characterize the evolution of the lymphatic system in embryonic and adult lymphangiogenesis.

After birth, angiogenesis contributes to organ growth but during adult life, most blood vessels remain quiescent, and angiogenesis occurs only in the cycling ovary and in the placenta (1, 2). However, EC retain their remarkable ability to divide rapidly in response to physiological stimuli, such as hypoxia for blood vessels and inflammation for lymph vessels (1). Angiogenesis and lymphangiogenesis are reactivated during wound healing and repair. In certain disorders, these stimuli become excessive and the balance between stimulators and inhibitors shifts, resulting in a (lymph) angiogenic switch. The best known conditions in which this switch is seen are malignant and inflammatory disorders.

Proangiogenic and antiangiogenic factors

Over the past 20 years, genetic studies have provided insights into the key mechanisms and molecular players that regulate angiogenesis and lymphangiogenesis. Vascular endothelial growth factor (VEGF), previously known as vascular permeability factor (VPF) (4), is the most specific growth factor for vascular endothelium (5). Vascular endothelial growth factor is not a single protein but a small menagerie of several peptides (6). The VEGF family includes VEGF-A, VEGF-B, VEGF-C, VEGF-D, and placental growth factor (PlGF) (5, 7). VEGF-A and VEGF-B are key regulators of blood vessel growth, while VEGF-C and VEGF-D primarily regulate lymphangiogenesis (5). Some components of the VEGF family have differentially spliced forms that differ in their effects on angiogenesis. For example, human VEGF-A has at least six isoforms: 121, 145, 165, 183, 189, and 204 (5), VEGF-A165 being the most potent pro-angiogenic isoform (8). PlGF, expressed in placenta and certain tumors, has two major isoforms: PlGF-1 (PlGF131) and PlGF-2 (PlGF152) (9).

Vascular endothelial growth factors signal through three human members of the VEGF receptor (VEGFR) family: VEGFR-1, VEGFR-2, and VEGFR-3 (5). Vascular endothelial growth factor functions are also regulated through the production of an alternative mRNA variant of VEGFR-1, soluble VEGFR-1 (sVEGFR-1). In addition to the VEGF receptors, two other molecules, Neuropilin-1 (NRP-1) and Neuropilin-2 (NRP-2) have been identified as coreceptors for VEGF (10, 11). During development, NRP-1 is expressed by arterial (EC) (12), and NRP-2 is expressed by venous and lymphatic EC (13). NRP-1 boosts the affinity of VEGF-A165 for VEGFR-2 (11) and increases its phosphorylation, enhancing downstream signaling (14) (Fig. 1).

Figure 1.

 The family of vascular endothelial growth factors (VEGFs) and their receptors. This family of structurally related molecules includes VEGF-A, VEGF-B, VEGF-C, VEGF-D, and placental growth factor (PlGF). Vascular endothelial growth factor-A is the main mediator of angiogenesis. Several isoforms of VEGF-A activate tyrosine kinase receptors VEGFR-1 and VEGFR-2. However, VEGF-A signals mainly through VEGFR-2, which is expressed at high levels by blood endothelial cells (BEC). Vascular endothelial growth factor-B and PlGF specifically activate VEGFR-1. The role of VEGFR-1, also expressed in BEC, in angiogenesis is less clear. VEGF-C and VEGF-D activate VEGFR-3 and VEGFR-2. VEGFR-3 is largely restricted to lymphatic endothelial cells (LEC). Besides the three tyrosine kinase receptors, there are coreceptors for VEGFs such as neuropilins (NRPs). NRP-1 associates with VEGFR-1 and VEGFR-2 to bind VEGF-A165, VEGF-B, and PlGF. NRP-2 associates with VEGFR-3 to bind VEGF-C and VEGF-D to regulate lymphangiogenesis.

Other naturally occurring promoters of embryonic and postnatal neovascularization are angiopoietins (Ang) that interact with Tie-2 receptors. Angiopoietin-1 (Ang-1) promotes angiogenesis by establishing and maintaining vascular integrity and quiescence (15). Ang-1 binds the Tie-2 receptor and stabilizes nascent vessels by protecting the adult vasculature against plasma leakage induced by VEGF (16). In contrast, Ang-2 reduces vascular integrity by competing for the Tie-2 receptor. No ligand has been identified for Tie-1, currently considered an orphan receptor. Tie-2 mRNA protein is most abundant in the lung, which therefore appears to be uniquely dependent on Tie-2 signaling (16). While Ang-1 is widely expressed in normal adult tissues, Ang-2 is expressed mainly at sites of vascular remodeling such as chronic inflammation (17) (Fig. 2). Ang-2 increases in plasma and alveolar fluid in adults with acute lung injury and is a mediator of epithelial necrosis with an important role in hyperoxic lung injury and pulmonary edema (18). The Angs role in lymphangiogenesis are not clear. However, Ang-1 overexpression induces lymphatic vessel enlargement, sprouting, and proliferation (19).

Figure 2.

 Schematic representation of angiopoietins and their receptors. Angiopoietins (Angs) are structurally related endothelial growth factors, with similar binding affinity for the tyrosine kinase receptor Tie-2. Ang-1 acts as a Tie-2 agonist, promoting endothelial cell (EC) survival, migration and proliferation, platelet activating factor (PAF) synthesis, and inhibiting vascular permeability. Ang-2 is an endogenous Tie-2 antagonist thereby counteracting Ang-1 activities. In the mouse, Ang-3 mimics the functions of Ang-2, while Ang-4 resembles Ang-1. Besides on EC, Tie-2 is also found on human neutrophils, mouse and human mast cells, and mouse monocytes. Tie-1 is an orphan receptor, expressed on EC.

Angiogenin, normally present in the circulation, is one of the most potent tumor-derived angiogenic factors (20) and plays a role in a number of nonmalignant vasculoproliferative pathologies. It has been implicated as a mitogen for vascular EC, an immune modulator with suppressive effects on polymorphonuclear leukocytes, an activator of certain protease cascades, and an adhesion molecule. Angiogenin is produced by cells such as macrophages, EC, and peripheral blood lymphocytes (21).

Basic fibroblast growth factor (FGF-β) belongs to a group of heparin-binding growth factors that stimulate endothelial cell proliferation and migration in vitro and angiogenesis in vivo (22). FGF-β plays a significant role in inflammatory conditions, including wound healing (23) and pulmonary fibrosis (24), and its tissue distribution suggests a role in angiogenesis.

Several cytokines and chemokines are involved in angiogenesis. Interleukin 8 (IL-8), secreted by monocytes, macrophages, and mast cells, is a potent mediator of angiogenesis (25) by stimulating endothelial proliferation and inhibiting EC apoptosis (26). It also increases the EC mRNA expression of matrix metalloproteinases (MMPs). IL-8 has at least four distinct proangiogenic properties: it boosts EC proliferation, induces EC chemotaxis and survival, and activates proteases (27).

Interleukin 17 (IL-17) is produced mainly by a subset of activated CD4+ T cells (TH17) (28). Several homologous proteins with similar or different biological profiles (IL-17B, IL-17C, IL-17D, IL-17E, and IL-17F) are grouped in a cytokine family. IL-17 mediates angiogenesis in humans by stimulating EC migration and cord formation and regulating the production of a variety of proangiogenic factors (29, 30). Promotion of angiogenesis by IL-17 may result from enhancement of the action of the FGF-β, the hepatocyte growth factor (HGF) and VEGF (31).

Different inhibitors of angiogenesis have been identified. Thrombospondin (TSP), the first antiangiogenic factor, found in the 1990s, prevents VEGF-induced angiogenesis by directly binding to it and interfering with its binding to cell surface heparin sulfates (32). Endostatin, a strong endogenous inhibitor of angiogenesis (33), inhibits endothelial growth and migration and promotes apoptosis. It antagonizes the effects of VEGF (34).

Angiogenesis and lymphangiogenesis are a feature of airway remodeling in bronchial asthma

Bronchial asthma is a chronic inflammatory disease characterized by airflow obstruction that may be reversed spontaneously or in response to treatment. The disease also causes airway remodeling with structural changes including smooth muscle hypertrophy/hyperplasia, globlet cell hyperplasia, subepithelial fibrosis, and increased vascularity (35). Airway remodeling is a central feature of asthma, but vascular remodeling has been frequently neglected. However, in both adults and children recent biopsy studies have found a large increase in the number of microvessels in airways of patients with asthma, with evidence that the endothelium is proliferating (36–38). Vascular remodeling has also been shown in a chronic allergen exposure rat model (39). Several excellent reviews have analyzed the role of angiogenesis in chronic airway diseases (40–45).

Dunnill (46) empirically described vasodilation and congestion of the bronchial mucosa using postmortem specimens as a striking feature of fatal asthma. Isolated reports over the next three decades described vascular changes in asthmatic lung autopsies (47, 48). Li and Wilson (49) in examining bronchoscopy biopsies assessed vascularity in the airways of patients with atopic mild asthma and found a higher vessel density in patients with asthma than controls. These patients were not subject to hypercapnia, hypoxia or acidosis, factors that can be invoked in severe asthma to explain vascular differences in the airways. The authors found no correlation between clinical parameters (duration of asthma, medication, FEV1, or PC20) and either percent vascularity or number of vessels in asthmatic biopsies.

Hoshino et al. (50) found more vessels and a larger vascular area in patients with atopic, mild to moderate asthma than in control subjects (Fig. 3). These authors described an inverse correlation between the percentage of FEV1 predicted and/or airway responsiveness and the percentage vessel area. Similarly, Hashimoto et al. (51) confirmed the increased vascularity in the medium and small airways of patients with asthma. Again, these changes correlated with disease severity in mild and moderate asthma (51).

Figure 3.

 Bronchial vascularity in bronchial biopsy specimens from normal control biopsies (panel A) and patients with asthma (panel B). Bronchial biopsy specimens were stained with anti-collagen IV. Red open circles indicate vessels (Bar = 25 μm). Modified with permission from Hoshino M. et al. (J. Allergy Clin. Immunol. 2001, 107:295–301).

A morphometric quantitative analysis of biopsy specimens from patients with mild to severe allergic asthma showed that asthmatic bronchial lamina propria had more vessels, occupying a larger area than in patients without asthma (52). The vascular network consisted mainly of capillaries and venules with edemic walls, and only a few arterioles. Again, vascular density correlated with disease severity.

Vrught et al. (37) used immunochemistry to investigate the density of the submucosal vascular bed and the expression of adhesion molecules in endobronchial biopsies obtained from patients with severe, glucocorticoid-dependent asthma and patients with mild asthma compared to control subjects. They found that mucosal neovascularization was a feature of airway remodeling in severe asthma and was also associated with a higher density of vessels expressing intercellular adhesion molecule-1 (ICAM-1).

Hoshino et al. (53) explored the role of stromal cell-derived factor-1 (SDF-1) in angiogenesis in asthma. Immunohistochemical studies on bronchial biopsy specimens from asthmatic and control subjects indicated a positive correlation between vascularity and SFD-1+ cells, suggesting that the increased vascularity of asthmatic bronchial mucosa in patients with asthma might be related to the expression of SDF-1, which may play a role in airway remodeling through angiogenesis.

Using a high-magnification broncovideoscope, Tanaka et al. found that steroid-naïve patients with newly diagnosed asthma had greater vessel networks in the airway mucosa than controls. The increase in the vessel network in newly diagnosed asthma was similar to that of subjects with asthma who had used long-term inhaled corticosteroids (ICS) (54).

Airway vascularity has also been examined in children. Barbato et al. (36) analyzing bronchial biopsies found that vessel number, percentage of vessel area, and eosinophil number were increased in mild to moderate asthmatic atopic children, although increased airway vascularity and eosinophilia were found even in atopic children without asthma. Therefore, inflammatory and structural changes can occur early in the natural history of asthma, and some of these pathological lesions may be associated with atopy even in the absence of asthmatic symptoms.

These findings strongly suggest that airway vascularity is enhanced in lung biopsies from either children or adults and correlate with asthma severity. Thus, vascular changes can contribute to the airway remodeling as well as to the clinical manifestations of asthma.

Angiogenic and lymphangiogenic factors and their cellular sources in bronchial asthma

The initial demonstration of vascular remodeling in asthma prompted a number of researchers to investigate the angiogenic and lymphangiogenic factors involved and the cellular sources producing these factors in the human airways. Early studies reported mRNA expression of VEGF-A and its receptors, VEGFR-1 and VEGFR-2, by in situ hybridization in bronchial biopsies of patients with mild to moderate asthma (55). mRNA expression of both VEGF-A and its receptors was significantly up-regulated in patients with atopic asthma compared to controls. VEGF-A expression correlated with submucosal vascularity and was inversely correlated with FEV1 and airway hyper-responsiveness. A positive correlation was also reported between vascularity and VEGFR-1 and VEGFR-2 mRNA positive cells. In this study, CD34+ cells, eosinophils, and macrophages all expressed VEGF-A within the bronchial mucosa and CD34+ cells, macrophages, and T cells expressed both VEGF receptors. The same authors, using immunocytochemical staining, found more VEGF+ cells in the bronchial mucosa of asthmatic subjects than in controls (50). Cells expressing angiogenic factors such as VEGF-A, FGF-β, and angiogenin were significantly more numerous in asthmatic subjects than controls, providing the first evidence that these factors may be important in asthmatic airway angiogenesis. Again, CD34+ cells, eosinophils, and macrophages colocalized with angiogenic factors.

Feltis et al. (56) obtained quantitative measurements of VEGF-A, VEGFR-1, and VEGFR-2 in airway biopsies by performing immunohistochemistry and image analysis. They found that levels of VEGF were increased in patients with asthma and correlated with VEGFR-2, thereby suggesting that this receptor was mainly involved in VEGF-induced angiogenesis in asthma. This hypothesis was supported by the observation that in normal subjects, VEGF levels correlated with VEGFR-1 (56).

Lee et al. (57) reported significant increases in VEGF-A levels in sputum of patients with asthma. The levels were even higher in patients with acute asthma than controls and were significantly correlated with the numbers of neutrophils and eosinophils. Vascular endothelial growth factor-A and endostatin levels were also elevated in induced sputum of patients with asthma compared to controls (58), suggesting an imbalance between angiogenic and antiangiogenic factors in the asthmatic airways.

Kanazawa et al. (59) further investigated the role of angiopoietins in vascular remodeling in asthma. They reported higher levels of Ang-1, Ang-2, and VEGF-A in induced sputum of patients with asthma than controls, with an inverse correlation between Ang-1 concentration and vascular permeability index and a positive correlation between Ang-2 and that index in asthmatic subjects. The same group examined the roles of Ang-1 and Ang-2 in exercise-induced bronchoconstriction (EIB) in patients with asthma receiving ICS [beclomethasone diproprionate (BDP) 800 μg/day] (60). The concentration of Ang-1 in induced sputum was similar to the healthy group, whereas Ang-2 levels were significantly higher in patients with asthma and correlated significantly with the severity of EIB and the airway microvascular permeability index. A recent study confirmed that Ang-1 and Ang-2 levels in induced sputum were higher in patients with asthma than controls. Ang-2 levels were higher in patients with asthma who smoke than in patients with asthma who do not smoke. Moreover, the Ang-2 levels were positively correlated with airway vascular permeability index. Beclomethasone diproprionate therapy (800 μg/day) decreased Ang-1 levels in patients with asthma, and this effect was correlated with an improvement in FEV1 (61).

Vascular endothelial growth factor-A and angiogenin were also detected in sputum of asthmatic children during an acute attack and in higher levels than in healthy children (62). It has been suggested that both VEGF and angiogenin could contribute to airway hyper-responsiveness by inducing chronic airway remodeling in asthma. An inverse correlation was found between FEV1 percentage of predicted and both VEGF-A and angiogenin. That study found no correlation between serum IgE and sputum VEGF-A but there was a positive correlation between peripheral eosinophils and VEGF-A levels. Hossny et al. (63) measured VEGF-A in the induced sputum of asthmatic children having acute attacks of varying severity, and the levels were higher than during remission or in the control group. Again, patients’ VEGF-A expression did not correlate with IgE but was correlated with peripheral blood eosinophils.

Chetta et al. (64) reported that the expression of VEGF was upregulated in the bronchial mucosa of patients with mild to moderate asthma compared to control individuals. Vascular endothelial growth factor-A upregulation was also correlated with the number of vessels and mast cells, and with basement membrane thickness. Colocalization analysis suggested that mast cells are an important source of VEGF-A in these patients because there was a significant correlation between the number of VEGF-A+ cells and the number of mast cells. A more recent study from the same group indicated that the number of chymase-positive mast cells was significantly correlated with the vascular area and with the number of VEGF-A+ cells in the airways of patients with asthma (65).

These studies support the hypothesis that mast cells can produce VEGF-A in the airways, and this may contribute to the neovascularization in asthma. This is borne out by two independent studies showing that IgE-dependent activation of rodent mast cells in vitro induced the release of preformed VEGF-A and FGF-2 (66) and that supernatants of activated mast cells induced an angiogenic response in chick embryo chorioallantoic membranes (CAM) that was mainly because of these angiogenic factors (67). We recently reported that primary human lung mast cells (HLMC) constitutively produce other VEGFs in addition to VEGF-A, namely the angiogenic VEGF-B and the lymphangiogenic VEGF-C and VEGF-D (68). These are often present as preformed mediators in mast cells because immunostaining for the various VEGFs in resting HLMC detected VEGF-A (Fig. 4A), VEGF-C (Fig. 4B), and VEGF-D (Fig. 4C) in these cells. Interestingly, PGE2 and adenosine, two important mediators in asthma, induce the expression of both angiogenic (VEGF-A and VEGF-B) and lymphangiogenic (VEGF-C and VEGF-D) Vascular endothelial growth factors in HLMC and elicit the release of VEGF-A in HLMC supernatants which in turn induce an angiogenic response in the CAM system (68). These findings indicate that human mast cells have an intrinsic capacity to produce several VEGFs, suggesting that these cells might regulate both angiogenesis and lymphangiogenesis in asthma. However, mast cells appear to be not only a source of VEGFs in the airways but also a target for these angiogenic factors. In fact, HLMC express VEGFR-1 and VEGFR-2, which are the major receptors for the various VEGFs. In addition, we have demonstrated that different VEGFs (VEGF-A, VEGF-B, VEGF-C, VEGF-D, and PlGF-1) exert chemotactic effects on HLMC by engaging both receptors. Because there is evidence that mast cells accumulate at sites of angiogenesis (69), it is possible that VEGFs produced during allergic reactions by mast cells and other resident cells exert paracrine recruitment of mast cells in the airways, thereby sustaining both inflammatory and angiogenic processes.

Figure 4.

 Vascular endothelial growth factors in human lung mast cells (HLMC). Immunostaining of HLMC for VEGF-A (panel A), VEGF-C (panel B), and VEGF-D (panel C). Both membranous and cytoplasmic localization of the staining is appreciable for all VEGFs. Human lung mast cells were purified from the lung tissue as previously described (de Paulis et al.). Cytospins (105 cells per sample) were prepared by centrifugation (22°C, 500 g, 5 min) on microscope slides. The cells were fixed in 4% paraformaldehyde and nonspecific bonds were blocked by preincubation (22°C, 30 min) with nonimmune horse serum (Vector Laboratories, Burlingame, CA, USA). Immunostaining was performed by incubation (4°C, 18 h) with the primary Abs [A: rabbit polyclonal anti-VEGF-A (Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit polyclonal anti-VEGF-C (B: Santa Cruz Biotechnology), rabbit polyclonal anti-VEGF-D (C: Santa Cruz Biotechnology)], or irrelevant rabbit IgG Ab (data not shown). Cells were then incubated with HRP-conjugated swine anti-rabbit secondary Ab (Dako Cytomation, Glostrup, Denmark) (22°C, 1 h). The slides were mounted with coverslips using a synthetic mounting medium (Dako Cytomation). Human lung mast cells were observed under a Zeiss Axiovert 100 M microscope and images were recorded with ism 510 software (Zeiss, Thornwood, NY, USA).

Macrophages are also central to the pathogenesis of bronchial asthma (70). Activated macrophages induce neovascularization (71) and contribute to angiogenesis and lymphangiogenesis in inflammatory diseases and cancer (72, 73). Primary human macrophages purified from human lung parenchyma express angiogenic (VEGF-A and VEGF-B) and lymphangiogenic factors (VEGF-C and VEGF-D) (Granata F. et al., unpublished observation). Secretory phospholipases A2 (sPLA2s), proinflammatory mediators released in the airways of patients with pulmonary diseases including asthma (74), enhance the expression and release of VEGF-A and VEGF-C (Granata F. et al., unpublished observation).

Eosinophils may be involved in regulating angiogenesis in asthma (75). Eosinophils synthesize and store in their granules several proangiogenic mediators such as VEGF-A, FGF-2, TNF-α, GM-CSF, NGF, and IL-8. In addition, these cells promote EC proliferation in vitro and induce new-vessel formation in aortic rings and in the CAM (76). Feistritzer et al. (77) detected VEGFR-1 on human peripheral blood eosinophils and demonstrated that VEGF induces eosinophil migration and eosinophil cationic protein (ECP) release through VEGFR-1 activation.

Basophils, primary effector cells in allergic inflammation, produce several forms of VEGF-A (121, 165, 189) and VEGF-B (167, 186) and their secretory granules contain VEGF-A (78). IgE-mediated activation of these cells caused VEGF-A release, and supernatants of activated basophils induced an angiogenic response in vivo. VEGF-A165 has a chemotactic effect on basophils through the activation of VEGFR-2. Because these cells do not express VEGF-C and VEGF-D circulating basophils presumably play a role in neoangiogenesis, but not lymphangiogenesis.

Human lung epithelial cells may possibly be significant sources of VEGF-A165 as it has been detected in cell culture media and bronchoalveolar lavage fluids (BALF) at concentrations capable of stimulating EC growth. Hypoxia and TGF-β1 stimulated VEGF-A production in these cells (79).

Canine vascular smooth muscle (VSMC) expresses VEGFR-1, VEGFR-2, and NRP1 at the mRNA and protein level and respond to VEGF-A in vitro (80). Knox et al. (81) extended these results by showing that human airway smooth muscle cells of postmortem tracheas also express several splice variants of VEGF-A (121, 165, 189, and 206) and constitutively secrete VEGF-A protein. Bradykinin increases VEGF-A secretion suggesting a novel role for human airway muscle cells in bronchial mucosal angiogenesis in asthma. Simcock et al. (82) also showed that airway smooth muscle cells of patients with asthma promote in vitro angiogenesis via VEGF release.

Vascular endothelial growth factor stimulation enhanced the production of MMPs by human vascular smooth muscle cells (83). The levels of VEGF and MMP-9 in sputum were significantly increased in patients with stable asthma and even higher in patients with acute asthma compared with control subjects. The levels of VEGF correlated with those of MMP-9 (84). In addition, it has been shown that VEGF-A can induce fibronectin secretion by human airway muscle cells (85). Chetta et al. (64) demonstrated that VEGF-A expression is associated with subepithelial fibrosis in humans. These findings suggest that VEGF-A contributes to airway remodeling in patients with asthma not only by acting on angiogenesis, but also by affecting the extracellular matrix composition and stimulating fibrotic changes.

Thus, the angiogenic switch in asthmatic airways involves different angiogenic factors (VEGFs, angiopoietins, angiogenin) produced by several cellular sources in the inflamed lung, mostly circulating endothelial progenitor cells, mast cells, macrophages, basophils, eosinophils, and smooth muscle cells.

Angiogenesis and lymphangiogenesis in experimental asthma

The role of VEGFs in the pathogenesis of the asthmatic phenotype and the effector functions of VEGFs in the lung have not been defined. Even the functional significance of vascular changes of the airways in asthma is not altogether clear. Several groups have worked to develop a better understanding of angiogenesis and vascular remodeling in chronic airway inflammation using animal models. Mycoplasma pulmonis infection is a useful model for studying angiogenesis and lymphangiogenesis in chronic inflammatory airway disease (86). In this model, the microvasculature of the airway mucosa begins to change soon after infection, and angiogenesis and microvascular remodeling become lasting features of the disease (87).

Isocyanate chemicals including toluene diisocyanate (TDI) are the most common causes of occupational asthma. VEGF appears to be one of the major determinants of inflammation and airway hyper-responsiveness in a murine model of TDI-induced asthma (88).

Exaggerated TH2 inflammation and airway remodeling are cornerstones in the pathogenesis of asthma. The role of VEGF-A in TH2 cell-mediated inflammation was evaluated in lung-targeted VEGF-A165 transgenic mice (89). Through IL-13-dependent and independent pathways, VEGF-A levels comparable to those in human tissues and biological fluids induced an asthma-like phenotype with inflammation, vascular remodeling, edema, mucus metaplasia, subepithelial fibrosis, smooth muscle hyperplasia, and airway hyper-responsiveness. VEGF-A also enhanced respiratory antigen sensitization and TH2 inflammation and increased the number of activated dendritic cells. In antigen-induced inflammation, VEGF-A was produced by epithelial cells and mainly by TH2 rather than TH1 cells. In the airways, stimuli as the respiratory syncytial virus (RSV) and endotoxin can enhance antigen sensitization and TH2 inflammation by inducing VEGF, and this might explain how RSV infection early in life contributes to the development of asthma. The same study showed that remodeling can also be caused by innate immunity responses, not only by chronic TH2 inflammation, providing a potential explanation for the observation of airway remodeling in childhood asthma, well before the damage of chronic TH2 inflammation would be expected.

The possibility that viral infections, a frequent event associated to asthma exacerbations, may modulate angiogenesis has been evaluated by Psarras et al. (90). They demonstrated that human rhinovirus (HRV) infection of cultured human bronchial epithelial cells (HBEC) stimulated mRNA expression and release of VEGF. Moreover, supernatants of HBEC infected with HRV induced angiogenesis in vitro. Leigh et al. (91) demonstrated that HRV infection of HBEC is a major regulator of airway remodeling by generating mediators involved in this process (amphiregulin, activin A, and VEGF).

Angiogenesis is linked to the mobilization of bone marrow-derived endothelial progenitor cells (EPC). A study of circulating EPC and EPC-derived EC colony formation in individuals with stable asthma and healthy controls reported higher levels of EPC, which were extremely proliferative and had enhanced incorporation into tubes in a model of angiogenesis (92). In the same study, the allergen challenge mouse model of asthma (OVA) provided clear evidence that the mobilized EPC were selectively recruited into the allergic lung. This mobilization is an early and selective event during allergen challenge of sensitized animals and is followed by an increase in microvessel density that progresses over time with chronic allergen exposure. These data support the concept that EPC recruitment and an angiogenic switch are early events in allergic airway inflammation.

ADAM33, a disintegrin and metalloprotease, is an asthma susceptibility gene whose polymorphic variation has been linked to asthma and bronchial hyper-responsiveness. Puxeddu et al. (93) reported that the catalytic domain of ADAM33, but not its inactive mutant, caused rapid induction of endothelial cell differentiation in vitro (matrigel assay) and neovascularization ex vivo (human embryonic/fetal lung explants) and in vivo (CAM). They also demonstrated that TGF-β2 enhances release or shedding of a soluble form of ADAM33 (sADAM33) to promote angiogenesis (93). This is the first study to demonstrate that ADAM33 acts as a remodeling gene independently of airway inflammation. The ability of TGF-β2 to augment sADAM33, highlights the potential interplay between genetic and environmental factors in the pathogenesis of asthma.

In asthma and other inflammatory diseases edema, a cardinal sign of inflammation, occurs when plasma leakage from blood vessels exceeds the lymphatic vessels drainage capacity. Little is known about the status of lymphatic vessels, and the factors that drive lymphangiogenesis to compensate for increased leakage during airway inflammation. An elegant study by Baluk et al. (94) investigated the role of lymphangiogenesis in the mouse model of chronic airway inflammation caused by Mycoplasma pulmonis. They found not only angiogenesis but also extensive growth of lymphatic vessels in infected airways. Lymphangiogenesis appears to be driven by VEGF-C and VEGF-D produced by inflammatory cells that migrate into the airways. Airway dendritic cells, macrophages, neutrophils, and epithelial cells expressing the VEGFR-3 ligands VEGF-C or VEGF-D act as sources of lymphangiogenic factors in infected airways. Adenoviral delivery of either VEGF-C or VEGF-D evoked lymphangiogenesis without angiogenesis, whereas adenoviral VEGF-A had the opposite effect. Inhibition of growth factor signaling through VEGFR-3 completely prevented the growth of lymphatic vessels but had no apparent effect on blood vessel remodeling. Inhibition of lymphatic growth exacerbated mucosal edema and reduced the hypertrophy of draining lymph nodes. Thus, when lymphangiogenesis is impaired, airway inflammation may lead to bronchial lymphedema and airflow obstruction.

Pharmacologic and immunologic modulation of angiogenesis in bronchial asthma

Several studies have investigated the effects of pharmacologic intervention on airway vascular remodeling in patients with asthma. Corticosteroids are the most effective anti-asthma drugs (95). In patients with asthma, inhaled corticosteroids (ICS) act on the structural and functional changes in the airways partly by altering airway mucosal blood flow and partly by changing chronic inflammation within the bronchial wall (96). ICS have vasoconstrictive action that may involve nonadrenergic neurotransmission mechanisms or action on bronchial flow by reducing the NO level (97, 98) or both. ICS also downregulate several airway cytokines (99), reduce cell infiltration of the bronchial wall (100), and possibly reverse basement membrane thickening (101). Corticosteroids may affect bronchial microvascularity by inhibiting the production of proangiogenic cytokines/chemokines such as IL-8, GM-CSF, TNF-α, and MMPs (102). They may also inhibit the expression of VEGF (103). Finally, ICS may also affect airway vascularity through an inhibitory effect on immune cells expressing pro-angiogenic molecules (i.e. basophils, mast cells, eosinophils) (96).

Focusing on the effects of ICS on airway vascularity, Orsida et al. (104) examined endobronchial biopsy specimens from patients with asthma receiving beclomethasone diproprionate (BDP) (200–1500 μg/day). There was an inverse correlation between ICS dosage and vascularity, with higher doses having more effect on airway wall vascularity (above 800 μg of BDP). Hoshino et al. (105) found that high doses of BDP (800 μg/day) for 6 months reduced airway wall vascularity in patients with mild to moderate asthma. Similarly, in a randomized, double-blind, parallel-group study, Chetta et al. (96) assessed the effect of 6-week treatment with low (100 μg twice a day) or high dose (500 μg twice a day) of inhaled fluticasone diproprionate (FDP) on the vascular component of airway remodeling in patients with mild to moderately severe asthma: the number of vessels, the vascular area, and basement membrane thickness dropped only after high dose FDP.

Focusing on the effects of ICS on angiogenic factors in the airways, Asai et al. (106) showed that VEGF levels in the induced sputum of patients with asthma dropped after a 2-month treatment with BDP. Kanazawa et al. reported that FDP (400 μg/day) lowered VEGF and Ang-1 levels and improved the vascular permeability index, which was inversely correlated with the decrease in Ang-1 (59). Other studies have shown that corticosteroids reduce the VEGF/Ang-2 ratio in the induced sputum of patients with asthma (107) and reverse the imbalance between VEGF and endostatin levels (58). These findings support the hypothesis that ICS downregulate angiogenic remodeling in asthmatic airways by reducing VEGF and Ang activity.

Beta (β2)-agonists are bronchodilating drugs that interact with β2-adrenergic receptors expressed in smooth muscle cells and in bronchial and inflammatory cells. There is little evidence that treatment with long-lasting β2-agonists (LABA) affect bronchial microvasculature. Orsida et al. (108) reported that 3-month treatment with salmeterol (50 μg twice a day) in patients with asthma taking low-dose ICS had no adverse effects on airway vascularity.

Few studies have examined the effects of leukotriene receptor antagonists (LTRAs) on airway vascularity. Lee et al. (109) reported that montelukast and pranlukast affected vascular permeability by reducing VEGF expression in a mouse model of allergic asthma suggesting that these drugs regulate VEGF expression. Kanazawa et al. (110) found decreases in VEGF levels and the vascular permeability index in the induced sputum of steroid-untreated patients with asthma receiving pranlukast for 4 weeks. The same group reported decreases in VEGF and Ang-2 levels in the induced sputum of patients with asthma given montelukast therapy (59).

Lee et al. (111) examined the effect of immunostimulatory sequences of DNA (ISSs) on allergen-induced airway angiogenesis and expression of angiogenic cytokines in mice. Mice challenged with ovalbumin developed an increase in the peribronchial vascular area, with higher BALF levels of VEGF in peribronchial cells expressing VEGF. Treatment of mice with ISSs containing a cytosine phosphorothioate guanosine (CpG) motif reduced peribronchial angiogenesis and VEGF in the BALF and the levels of peribronchial cells expressing VEGF. Thus, ISS may prevent and reverse many of the structural features of airway remodeling in mice, including angiogenesis.

Increasing evidence indicates that inhibition of VEGF and of its receptors is a promising strategy to modulate tumor associated angiogenesis (112). However, the therapeutic potential of inhibiting VEGF and its receptors in asthma is still in its infancy. Some VEGFR inhibitors (SU5614, SU1498) reduce plasma extravasation and bronchial airway hyper-responsiveness by inhibiting the overexpression of VEGF in murine models of TDI-induced asthma (88). These inhibitors dramatically reversed all the symptoms examined, suggesting that VEGF is one of the major determinants of TDI-induced asthma and that its inhibition warrants consideration for anti-asthma therapy. One likely explanation for their effectiveness is that VEGF rapidly increases vascular permeability so that plasma proteins, including inflammatory mediators and cells, can leak into the extravascular space allowing the migration of inflammatory cells, including neutrophils, basophils, and eosinophils, into the airways.

Oxidative stress plays a critical role in airway inflammation, and reactive oxygen species (ROS) cause vascular leakage. Lee et al. (113) examined the effects of a prodrug of cysteine, L-2-oxothiazolidine-4-carboxylic acid (OTC), an anti-oxidant that reduces vascular permeability, in a murine model of asthma. OTC markedly reduced plasma extravasation and lowered VEGF levels in allergen-induced asthmatic lungs. The same study showed that OTC reduced the levels of hypoxia-inducible factor-1α (HIF-1α), a transcriptional activator of VEGF, in protein extracts of the lung, after ovalbumin inhalation. Therefore, OTC may influence vascular permeability by lowering VEGF expression.

Bhandari et al. (114) showed that VEGF induce pulmonary alterations by selectively stimulating nitric oxide (NO) production through activation of endothelial NO-synthase (eNOS) or inducible NO-synthase (iNOS). These findings suggest that eNOS and iNOS isoform-specific interventions might be used to control the pathologic vascular and extravascular manifestations of VEGF. This may be important for diseases such as asthma where NO production regulate tissue inflammation and viral replication and diseases in which iNOS activation generates peroxynitrite and other reactive oxidant species.

Several endogenous molecules specifically inhibit angiogenesis and could offer a novel approach against the bronchial microvascular changes in chronic inflammatory airway diseases, such as asthma. Endostatin is a potent naturally occurring anti-angiogenic protein (34, 35), and treatment with recombinant endostatin can prevent the development of asthma features in a mouse model (115). This class of agents merits further study as novel therapeutics for asthma.

MicroRNAs (miRNAs) are a novel regulatory class of noncoding, single-stranded RNAs with approximately 22 nucleotides identified in plants and animals (116). miRNAs repress protein expression at post-transcriptional level, mostly through base pairing to the 3′ untranslated region (UTR) of the target mRNA, leading to its degradation or reduced translation. Poliseno et al. (117) elegantly demonstrated the functional role of miRNAs in the control of angiogenesis. They identified miR-221 and miR-222 as key players of angiogenesis, showing that they influence expression of the c-kit receptor in human EC, hence the angiogenic properties of the c-kit ligand stem cell factor (SCF). Future miRNA studies on several aspects of angiogenesis in asthma should have substantial impact especially as clinically modified miRNAs (antagomirs) effectively alter the expression of various miRNA targets (118).

Closing thoughts

Several studies have investigated the role of vascular remodeling in asthma and have added new insights into the pathogenesis of chronic inflammatory airway diseases. Nevertheless, there are still several unanswered questions (Table 1).

Table 1.   Unanswered questions about angiogenesis and lymphangiogenesis in bronchial asthma
Are angiogenesis and lymphangiogenesis secondary to chronic inflammation and/or reparative processes, or an important early step in asthma?
What do other members of the VEGF family (e.g. VEGF-B, VEGF-C, VEGF-D) besides VEGF-A do in asthma?
Is the hyper-production of VEGF induced by local hypoxia an adaptive phenomenon or has it a pathogenic role in asthma?
What are the main immunologic and nonimmunologic stimuli (e.g. adenosine, PGE2, secreted phospholipases) that induce the release of angiogenic and lymphangiogenic factors from human inflammatory cells?
Could human inflammatory cells, under appropriate circumstances, produce anti-angiogenic factors?
What are the precise roles of pro- and anti-angiogenic chemokines synthesized by human inflammatory cells in asthma?
Several cytokines and chemokines (e.g. IL-8, IL-17) exert pro-angiogenic effects. What part does IL-17 play in angiogenesis in certain forms of asthma?
Are other angiogenic networks involved in asthma (e.g. Angs/Tie receptors, semaphorin/plexin, Delta/Notch)?
Vascular endothelial growth factors exert chemotactic effects on human basophils, mast cells, monocytes, etc. Do they have other proinflammatory effects?
What is the importance of lymphangiogenesis in asthma? What stimuli drive lymphangiogenesis during inflammation?
MicroRNAs (miRNAs) regulate gene expression on post-transcriptional level and specific miRNAs that regulate endothelial cell functions and angiogenesis have been described. Are specific miRNAs involved in the modulation of angiogenesis in asthma?

A number of studies examining lung biopsies and biological fluids in adults (37, 55, 106) and children (36) have provided evidence of early increases in the expression of VEGF and other angiogenic factors in the airways of patients with asthma. In some of these studies, VEGF expression and vascularity correlated with the severity of asthma (64, 119). These findings have been supported by the results of a number of studies using different rodent models of asthma (88, 89, 94) and data support the concept that an angiogenic switch is an early event during allergic airway inflammation. There are several possible sources of angiogenic factors in the inflamed lung: circulating EPCs, endothelial and lymphatic cells, monocyte/macrophages, dendritic cells, eosinophils, mast cells, basophils, neutrophils, epithelial, and stromal cells. A wide spectrum of physical (shear stress), chemical (adenosine, PGE2, etc.), or immunologic stimuli, local hypoxia, and viruses might be potent activators of the release of angiogenic factors (Fig. 5) (66, 90, 91, 120, 121).

Figure 5.

 Roles of immune cells in the modulation of angiogenesis and lymphangiogenesis in bronchial asthma. A number of genes and environmental factors (pollens, allergens, superallergens, viruses, bacteria, air pollution, etc.) can influence the outset and progression of asthma. An early event is damage to epithelial cells that induce either repair of the epithelium or chronic allergic inflammation. The latter event involves the production of several chemokines, cytokines, and growth factors by the damaged epithelium and activated immune cells [mast cells (MC), eosinophils (Eos), basophils (Baso), TH2 cell, and macrophages (Mø)]. There is increasing evidence that several immune cells produce angiogenic and lymphangiogenic factors that contribute to airway narrowing, chemotaxis of inflammatory cells, and tissue remodeling. Increased mast cell density of airway smooth muscle (mast cell myositis), increased basophil/mast cell releasability, myofibroblast activation, and airway hyper-responsiveness are also characteristics of asthma and their complex interplay contribute to the heterogeneous phenotypic expression of the disease.

The importance of lymphangiogenesis in asthma remains to be explored. We reported that HLMC and macrophages contain and express the two main lymphangiogenic factors VEGF-C and VEGF-D (68) and Granata F. et al. unpublished observation). Thus, primary effector cells of inflammation may be a source of lymphangiogenic factors. Correcting defective lymphangiogenesis may benefit patients with asthma and other inflammatory lymphatic networks after resolution of the inflammation would leave a drainage system in place for fluid and immune cells that might accelerate responses in subsequent inflammatory episodes.

What is the significance of VEGF expression in the bronchial vessels in asthma and how might VEGF contribute to these diseases? An increased bronchial vasculature would increase inflammatory cell trafficking and exudation of mediators, particularly if vascular permeability is altered. Increased vascular permeability and plasma leakage from newly formed vessels accumulate in the tissues around the airway contributing to airway wall thickness (94). Airway wall thickening would cause the airways to narrow further on stimulation with constrictors thereby contributing to bronchial hyper-responsiveness. The increased vasculature could also contribute to airway hyper-responsiveness by supporting the increase in airway smooth muscle mass, which is a feature of asthma.

Correction of altered angiogenesis/lymphangiogenesis may prove beneficial in the treatment of asthma and other chronic inflammatory airway diseases. Although the lowest dose of ICS required to treat airway vascular remodeling has yet to be established, it seems that only high doses are effective in asthma. CysLT receptor antagonists can alter vascular permeability by reducing angiogenic factor expression in the airways. Finally, agents that specifically inhibit various mediators (VEGFs, antibodies, etc.) and receptors controlling angiogenesis/lymphangiogenesis (VEGFR inhibitors, endostatins, ISS) may offer novel strategies for dealing with treatment of microvascular changes in asthma. The probable importance of angiogenesis/lymphangiogenesis in the pathophysiology and therapy of asthma makes the study of vascular remodeling in these disorders a priority for future research.