According to the Global Initiative for asthma (GINA), approximately 300 million people, regardless of age and ethnicity,

suffer from asthma. The pathophysiology and immunopathology of this serious condition is complex; new clinical and experimental in vitro , ex vivo andin vivo findings continue to expand the scope of this complexity. A major issue with asthma is its heterogeneity and the multiplicity of different and overlapping phenotypes. Indeed, identifying and defining asthma phenotypes remain major challenges facing the three constituencies of clinics, experimental laboratories and industry. Cardinal features of asthma include airway inflammation, reversible airflow obstruction, airway hyperresponsiveness (AHR) and tissue remodeling. Asthma immunopathology is dominated by a T helper (Th) cell response skewed towards a Th2 cytokine/chemokine pattern that also involves airway smooth muscle (ASM) cells and tissue remodeling. Pathogenesis is orchestrated by a complex interplay between several immune and inflammatory cells and their secreted products. The objective of this article is to review recent advances in the pathophysiology and immunopathology of asthma and to highlight new players within its immune and inflammatory components. In addition, the article will examine recent developments in asthma therapy aimed at improving the clinical management of this disease. Immune cells Dendritic cells Dendritic cells (DCs) are professional and efficient antigen-presenting cells (APCs) that can prime naive CD4+ T cells towards Th1, Th2, Treg or Th17 phenotypic cell patterns [1-4] . The lung is equipped with an elaborate network of DCs that can be found throughout the conducting airways, lung stroma, lung vasculature and bronchial lymph nodes [1] . DCs play an important role in determining the initiation and preservation of allergic immune responses[5] . They recognize inhaled antigens through expression of biologically conserved pattern-recognition receptors, including Toll-like receptors (TLRs), nucleotide-binding oligomerization domain (NOD)-like receptors, and C-type lectin receptors that recognize specific allergenic and pathogenic motifs [6] . In addition, lung DCs express numerous receptors for inflammatory mediators that are released upon tissue damage. By exerting direct and indirect sensing mechanisms for danger in the airways, DCs are at the nexus of innate and adaptive immunity in the lung [7] .

One of the most important roles of DCs in the airway is to sample antigens from the airway lumen by forming dendritic extensions between epithelial cells where they uptake and transport inhaled allergens [5] . Uptake of allergen is facilitated by IgE bound to high-affinity receptors on DCs [8] . Processing of allergens by cathepsin S and the selection of peptides loaded together with and presented by MHC class II molecules is fundamental for DC antigen presentation to T cells via its receptor (T-cell receptor [TCR]), sensitization of appropriate T-cell subpopulations and generation of subsequent immune responses [9] . Allergens are often associated with TLR agonists; DC stimulation through TLRs induces NF-κB activation and subsequent DC activation. The latter involves the production of chemokines and cytokines, including thymic stromal lymphopoietin (TSLP) [10] and IL-33 [11] . Although numerous reports have suggested that once antigen is taken up by lung DCs, the result is their migration in a CC-chemokine receptor 7 (CCR7) - and CCR8- (and to a lesser extent CXCR4- )-dependent manner [12] to the draining mediastinal lymph nodes [13,14] , it remains unclear precisely where and which DC subset is involved in sampling the inhaled antigen from the airways [15] . Engagement of parallel costimulatory molecules plays an important role in shaping the nature of the immune response; either CD80 or CD86 leads to sensitization, while lack of or inefficient engagement of these molecules leads to T-cell anergy [16] . In experimental asthma models and, unlike conventional lung-derived DCs, plasmacytoid DCs (pDCs), do not present antigens in an immunogenic manner to CD4+ T cells [5] . The precise mechanism for the depletion of pDCs that results in Th2 cell sensitization is still unresolved. However, data suggest that pDCs suppress the potential of lung-derived myeloid DCs to generate effector T cells directly [17] . Moreover, both in vitro and in vivo studies have shown that pDCs can induce Treg development, possibly in an inducible T-cell costimulator ligand (ICOSL)dependent manner [5,7,18] . Once induced, Tregs can also induce the production of the tryptophan catabolizing enzyme, indoleamine 2,3 dioxygenase (IDO), through reverse signaling in pDCs. The downstream tryptophan catabolites by IDO might exert an anti-inflammatory and/or tolerant effect in the lungs, by inhibiting T-cell activation [19] (see later in the 'Eosinophils' section). Th1/Th2 cells Endotoxin stimulates macrophages/APCs to produce IL-12, which in turn triggers the development of antigen-specific Th1 cells and inhibits the development of Th2 cells; thus, endotoxin levels inversely correlate with asthma [20] . Most epidemiological studies support the principles of the hygiene

hypothesis associated with an increasing prevalence of allergic diseases as theorized in the Th1/Th2 paradigm [21] . Immune regulation in T-cell homeostasis, first introduced in 1986, described mouse T-cell expression of distinct cytokine patterns. This hypothesis was later shown to also apply to human immunity, especially the classical Th1/Th2 balance thought to be critical in immune regulation and health [22] . The incidence of Th1-mediated autoimmune diseases, however, is known to have increased in the last half century in parallel with the increase of Th2-mediated allergic diseases [23,24] . As such, the classic Th1/Th2 paradigm cannot explain this imbalance. Relatively recent studies have introduced the concept of Tregs in immune suppression of Th1/Th2 interactions. Tregs The concept of Treg suppression of Th cell responses has led to intensive investigation into the potential of Tregs to prevent immune responses regulated by various subsets of CD4+ cells. Since Tregs inhibit the number and effector function of both Th1 and Th2 cells, the incidence of Th1- and Th2-mediated diseases are thought to be related to an insufficient development or recruitment of Treg cells [25] and are altered in asthma [26,27] . Among Treg subsets, naturally occurring Treg cells (nTregs) were first identified in mice and have since been well investigated in humans. The nTreg cells originate in the thymus, express a repertoire of CD4+ CD25+ (mouse) or CD4+ CD25high (human) cells, along with the transcription factor forkhead box protein P3 (Foxp3), and play a major role in modulating the activity of self-reactive cells by inducing the destruction of autoreactive T cells mainly via cell-to-cell contact-dependent mechanisms [28] . The newly identified IL-12-related cytokine, IL-35, was described to be produced by Foxp3+ Tregs in mice [29] ; however, IL-35 expression is not constitutive in human Tregs [30] . Other Treg subsets bearing suppressive capacity are designated adaptive Tregs or inducible Tregs, and acquire antigen specificity after birth. The mechanisms involved in Treg function operate through Foxp3 [31] as well as secretion of IL-10 (Tr1 subset) [32] and/or TGF-[beta] (Th3 subset) [33] . IL-10 and TGF-[beta] are the molecules that were shown to have suppressive roles in asthma through inducing Treg-mediated regulation [34] . IL-10 also induces T-cell anergy, possibly through blockade of T-cell costimulatory pathways [35,36] , promotes IgG4 and inhibits IgE class switching [33] . A recent study showed that OX40 expression on Tregs binds to OX40L on mast cells (MCs) and inhibits their degranulation [37] . Moreover, IDO has given rise to a possible immunosuppressive effector mechanism of Tregs. Tregs were shown to induce high levels of functional IDO through CTLA4 signaling on Tregs via B7-1 and B7-2 on DCs [34,38] . Indeed, increased ratios of a number of proinflammatory (TNF, TSLP and IL-6) and Th1/Th2 cytokines versus regulatory cytokines (IL-

10 and TGF-[beta]) in severe asthma [10,22,39] were shown to induce a balance shift towards a chronic inflammatory state. Th17 cells The recent surge in interest in Th17 cells has characterized their ability to elaborate IL-17, IL-22, IL-17F and CCL20. IL-6, in the presence of TGF-[beta] rather than IL-23, was identified as a crucial combination for the differentiation of Th17 [40,41] . Furthermore, retinoic acid receptor-related organ receptor (ROR)gt was found to be a key transcription factor for Th17 differentiation [42] . Recent studies showed these cells to be associated with the development and pathophysiology of chronic allergic diseases. A role for Th17 cells in the development of psoriasis, allergic contact dermatitis and atopic dermatitis has been described; however, a specific role for Th17 cells in asthma remains controversial. Several studies reported the presence and function of IL-17 in asthma, supporting the possible involvement of Th17 cells in asthma pathophysiology [43-45] . It is important to note that aside from CD4+ Th17 cells, several other cell types including CD8+ T cells, natural killer (NK) cells, gd T cells and NKT cells produce IL-17. In addition, il17 mRNA has also been found in eosinophils and neutrophils. Thus, while IL-17 may play a role in asthma, it does not necessarily indicate Th17 cell involvement. Several studies suggest that IL-17 production correlates more closely with neutrophilic but not eosinophilic asthma [46,47] . In support of this notion, studies comparing epicutaneous versus intraperitoneal sensitization of mice with ovalbumin [48] suggested that the route of sensitization may be important in directing Th17 versus Th2 responses in allergic lung inflammation. Notably, Th17 cells were absent in murine lungs with eosinophilic non-neutrophilic asthma. With the discovery of Th17 and Treg subsets, we have a better grip on understanding the new balancing square paradigm (Figure 1), which has replaced the classical two-sided (Th1 vs Th2) balance model to explain the pathogenesis of complex immune diseases. Natural killer & natural killer T cells Natural killer cell function is altered towards a Th2 cytokine-producing phenotype [49,50] , and current data suggest a dual role for NK cells in the pathogenesis of asthma. NK cells can promote allergic airway inflammation during sensitization and ongoing inflammation [51,52] . Conversely, stimulation of NK cells towards an IFN-γ-secreting phenotype may reduce allergic airway pathology [53,54] .

Natural killer T cells, a newly described small population of lymphocytes that express markers of both T cells and NK cells, have been proposed to contribute to the development of asthma. Extensive animal model studies of asthma induced by allergen, viruses, bacteria and environmental factors suggest that NKT cells function both in concert with Th2 cells and independently to cause airway hyperreactivity [55-59] . In clinically relevant human studies, NKT cells have been found in the lungs of patients with severe, poorly controlled asthma [60-64] . Inflammatory cells Eosinophils Eosinophils are multifunctional, pleiotropic granulocytes that play a prominent role in the pathogenesis of asthma [65,66] . In uncontrolled asthma, elevated numbers of these cells are found in airways, sputum and bronchoalveolar lavage (BAL) fluid, bone marrow, blood and tissues [67,68] . Treatment of asthma with corticosteroids (CSs) results in a dramatic reduction in sputum and tissue eosinophil numbers that parallels clinical improvements [69] . Thus, eosinophils may be fundamental to airway dysfunction in asthma [70] . Upon the release of prostaglandin (PG) D2 , cysteinyl leukotrienes (CysLT), cytokines and chemokines from asthmatic airways, eosinophils are recruited from the bone marrow as CD34+ precursors that egress under eotaxins 1-3 (CCL11, CCL24 and CCL26) influence to reach the airway wall. Thus, granulocyte-macrophage colony-stimulating factor [GM-CSF], IL-3, IL-5 and eotaxins 1-3 are crucial for eosinophil growth and differentiation from precursor cells; IL-5 is critical in eosinophil maturation and terminal differentiation. Recruitment into the airways is under the influence of eotaxins 1-3 [71,72] . Recent advances in eosinophil research have pointed to three major areas (Figure 2) where this cell type may play a role in the pathophysiology of asthma. The first relates to eosinophil potential as an active effector cell in its capacity to secrete a wide array of granule-stored cationic and other proteins, as well as reactive oxygen species four-to-ten-times more than neutrophils [73] , de novo synthesized lipid mediators (e.g., CysLT, PG, thromboxanes [TX] and platelet-activating factor [PAF]) and matrix metalloproteinases (MMPs) [66,70,74,75] . This assumed function continues to plague the eosinophil with uncertainty. Recent studies on anti-IL-5 monoclonal antibody (mAb) therapy that targeted eosinophilic asthma subjects have provided convincing evidence for an effector role of this cell type at least in recurrent asthma exacerbations [76,77] . In addition, there are experimental animal model studies that have provided further support for eosinophil effector function in allergic airway disease, including the PHIL eosinophil-deficient mouse model [78] . The

second role assigned to the eosinophil relates to its contribution to airway tissue remodeling through its proclivity to generate a wide range of cytokines and growth factors that contribute to mucogenic responses, fibrogenesis, collagen deposition, extracellular matrix (ECM) expansion, angiogenesis and smooth muscle hyperplasia [79-83] . Eosinophils are the source of several fibrogenic and growth factors, including TGF-[alpha], TGF-[beta], FGF-2, VEGF, MMP-9, IL1[beta], IL-13 and IL-17 [70] . This was further confirmed in genetically modified mouse models [84] . The third (and in our view, the most exciting) new perspective on the role of eosinophils beyond their effector and tissue modulator roles relates to their likely immunoregulatory function in the context of the immunopathogenesis of asthma through their interaction with T cells. Eosinophils have the ability to influence lymphocyte function directly [85] . In murine models of asthma, eosinophils transmigrate to and from lymphoid tissues and present antigens to naive T cells [86,87] (at a lower efficiency than, for example, DCs and other APCs). There are four aspects that relate to the likely role of eosinophils in immune regulation. The first is that eosinophils home naturally to the infant thymus in the absence of any identifiable 'danger signal' [88,89] . Whether eosinophils are involved during early immune ontogeny, which has been suggested by studies on murine thymus eosinophils shown to be active participants in MHC class I-restricted deletion of autoreactive T cells during the early neonatal period [90] , remains unclear. We have recently studied this question in the infant human thymus and showed significant positive correlations between eosinophil numbers, immunoregulatory molecules and Th2 markers [91] . Second, eosinophils synthesize, store and release over 30 cytokines, chemokines, interferons and growth factors that can potentially exert regulatory effects within the immune milieu in lymphoid tissue in situ [92] . These eosinophil-derived cytokines can influence T-cell selection by DCs directly and the choice between tolerance and immune stimulation. Specific eosinophil recruitment to lymphoid tissues places this cell type in a position to exert immunomodulatory effects on T cells through their cytokines and other products. Third, we believe that the induction by IFN-γ of the rate-limiting enzyme, IDO, in the oxidative catabolism of tryptophan is a significant mechanism by which the proliferation of a subset of Th-cell population is modulated [93] . Lymphoid tissue eosinophils may also induce T-cell apoptosis through the synthesis and release of IFN-γ as a consequence of the ligation of CD28 on their surface, leading to the induction of IDO [94] . Recent studies from our laboratory showed that IDO was constitutively expressed in eosinophils and catabolized tryptophan to kynurenines (KYN), which in turn induced Th1, but not Th2, apoptosis [95] . Thus, eosinophil-derived KYNs may be important regulators of Th2 bias (and also DCs) via IDO. The fourth piece of evidence is that eosinophils have the capacity to influence T-cell regulation and

its damaging inflammatory sequelae in a mouse allergic airway limitation model. This study clearly demonstrated that eosinophils are indispensible in the recruitment and differentiation of Th2 cells [96] . All this has led to the Local Immunity And/or Remodeling/Repair (LIAR) hypothesis, which proposed that tissue accumulation of eosinophils may be associated with their role as regulators of local immunity and/or remodeling and repair in health and disease [97] . Neutrophils Cellular inflammation of the airways is a characteristic feature of asthma pathogenesis. Although eosinophilic airway inflammation is recognized as an important feature of asthma immunopathology, evidence supports a prominent role for neutrophils in severe chronic asthma and sudden severe asthma attacks [98] . Studies have suggested that a significant percentage of asthma is associated with atopic, IgE-dependent and eosinophilic responses [99] . However, recent increasing evidence suggests that a percentage of asthmatic subjects are characterized by neutrophilic inflammation [98] . In these noneosinophilic asthmatics, airway neutrophilia appears to be enhanced; the extent of neutrophilia was shown to correlate with asthma severity [100-103] . In sudden-onset fatal asthma, neutrophilic inflammation is largely present in the absence of eosinophils; indeed, neutrophil numbers are highly elevated in status asthmaticus [104,105] , indicating that severe and fatal asthma may be more of a neutrophil- rather than an eosinophil-mediated phenomenon. Neutrophils have also been shown to be prominent in steroid-resistant asthma [106] , which may relate to the observation that CSs induce eosinophil but not neutrophil apoptosis [107] . A distinct subpopulation of asthmatic subjects were shown by sputum analysis to be neutrophilic; these subjects responded poorly to CS treatment as measured by decreased improvement in FEV1 and methacholine (MCh) responsiveness [108] . In fact, CSs have also been shown to downregulate the expression of a number of eosinophil-associated chemokines, including eotaxins, monocyte chemotactic protein (MCP)-3 (CCL7) and MCP-4 (CCL13), while enhancing the expression of neutrophilic chemokines, including IL-8 (CXCL8), IP-10 (CXCL10) and MCP-2 (CCL8). It was suggested that the inability of CSs to block the expression of IL-8 may contribute to persistent moderate-to-severe asthma associated with airway neutrophilia in spite of CSs treatment [109] . In addition, neutrophils are steroid insensitive, and neutrophil apoptosis is inhibited by CSs [110,111] . Taken together, these data suggest a role for neutrophils in mediating severe and steroid-resistant asthma [112-114] . Recent studies have linked neutrophilic responses in asthma with Th17, IL-17, asthma severity and chronicity. Severe asthmatics show elevated levels of IL-

17A [115-118] and increased IL-17A and IL-17F levels are positively correlated with disease severity, indicating an important role for IL-17A and IL-17F in severe asthma [45,119] . Elevated IL-17A levels also correlate with increased neutrophilic inflammation [116] and with increased AHR [120] . These data proposed IL-17A as an important mediator in the pathogenesis of severe neutrophilic asthma. The cellular source of IL-17A in asthma remains unclear, although a recent study has shown the presence of a significant number of IL17A-producing CD4+ cells in asthmatic tissues [121] . Neutrophils produce numerous mediators that can contribute to early and late asthma pathophysiology, including metalloproteinases, elastase, lactoferrin, myeloperoxidase, adhesion molecules, lipid mediators, reactive oxygen species, eosinophilic cationic protein and IL-8 [122] . Mast cells Mast cells are derived from hematopoietic progenitor cells; since they do not usually circulate in mature form, their differentiation and maturation occur locally [123,124] . MCs are key effector cells in IgE-associated immune responses, including asthma [125] , and are present within the airway epithelium and submucosa [126] , as well as deeper in the airway wall (Figure 1) [127] . Upon inhaled antigen-induced high-affinity IgE receptor (FcεRI) crosslinking and activation, MCs release preformed secretory granule mediator complex, including histamine, tryptase and other proteases. Activation is also accompanied by the rapid synthesis of lipid mediators such as CysLT, dihydroxy LTs and PGD2 , as well as cytokines and chemokines including IL-4, IL-13, IL12, IL-1, IL-18 and TNF. MCs also elaborate heparin and newly formed eicosanoids including PGD2 , TXA2 and LTC4 /LTD4 [128,129] . This array of inflammatory mediators acts sequentially and/or synergistically on target tissues and elicits smooth muscle contraction, increased microvascular permeability and inflammatory cell infiltration. A member of the EGF family, amphiregulin (AREG), thought to be related to airway remodeling in asthma (see later in the section entitled 'Structural cells and airway remodeling'), was reported to be secreted by FcεRI-activated MCs. AREG expression was not inhibitable by CS treatment [130] . TSLP is also produced and stored in FcεRIactivated MCs in the presence of IL-4 [131] . IL-33 has been suggested to promote human MC survival, adhesion and cytokine elaboration [132] . MCs not only counteract Treg suppression, through IL-6 and OX40/OX40L axis, towards a Th17 cell-type differentiation [133], but also express IL-17A [134] . In addition to FcεRI-mediated activation, MCs are shown to contribute to the development of asthma via an FcγR-dependent pathway [135] .

Recently, MCs were shown to release histamine in response to PAF stimulation through the PAF receptor (PAF-R) [136] . Although MCs are best known for their role in IgE-associated exacerbation of allergic disorders, they can also exert regulatory immunomodulatory effects [137] . Their protective function has been suggested since MCs were shown to be essential in Foxp3+ Treg-dependent peripheral tolerance following IL-9 signaling [138] . Furthermore, MC-derived IL-10 is a recognized modulator of allergic inflammation [139] . Basophils Basophils are often recruited to the site of allergic inflammation, even though they represent less than 1% of peripheral blood leukocytes [140] . Basophils express high levels of FcεRI [141,142] . They are activated through FcεRI elaborate cytokines such as IL-4, IL-13 and IL-3 and degranulate, releasing many pharmacologically active mediators such as histamine and eicosanoids [141,143] that contribute to Th2-type inflammation. Similar pathways can also be induced by IgG1 crosslinking on MCs that can lead to a cell-specific basophil-mediated anaphylaxis [144,145] . However, recent data have revealed a role for basophils not only in effector responses, but also in the initiation of the adaptive immune response, by boosting the memory immune response [144] . Basophils are the main source of IL-4 and IL-6 in the spleen and bone marrow of murine allergic models [144] . Since depletion of basophils leads to a reduction in IgG1 and IgG2a levels in both MC-sufficient and MC-deficient mice, these cytokines may be important in boosting antibody production during the memory phase [144] . This critical role of basophils in IgE-mediated allergic inflammation was confirmed using mice that were depleted of basophils by the mAb Ba103, which specifically targets CD200 receptor 3 (CD200R3; a CD200 receptor-like glycoprotein) [146] . Indeed, basophils are no longer considered redundant 'circulating MCs' [147] but may be important initiators of immune reactions, as well as effectors in IgE-mediated allergic inflammation [146,148] . Structural cells & airway remodeling Airway remodeling is one of the four main features of asthma, which include a chronic inflammatory response in the airways associated with reversible airway obstruction, and airway hyperreactivity. Airway remodeling occurs throughout the airways and involves measurable changes in epithelial cell properties, including shedding, goblet cell hyperplasia, increased mucus secretion, increased blood vessel number and area, increased and changed deposition of ECM, and proliferation and differentiation of mesenchymal cells (myofibroblasts and

smooth muscle cells) (Figure 1). Although airway remodeling even occurs in young children [149,150] , this is thought to be a consequence of repeated tissue injury and persistent inflammation [151,152] . Cytokines (TGF-[beta]) [153] , Th2 cytokines (IL-5, IL-9 and IL-13), chemokines (NF-κB-regulated chemokines) [154,155] , and growth factors (VEGF [156] and AREG [157] ) released from inflammatory and structural cells in the airway also play a critical role in the development and persistence of remodeling. Epithelial cells The airway epithelium is the primary mucosal barrier, which plays a central role in maintaining airway homoeostasis and asthma pathogenesis. Indeed, the barrier function of airway epithelium is impaired in asthma [158,159] . While 'epithelial desquamation' was considered a classic pathological feature in asthma death [160] , the significance of the epithelium was elucidated relatively recently [161] . Tight junctions (TJs) are characteristically located at the apicolateral borders of adjacent epithelial cells. In addition to their key role in maintaining barrier function of the epithelium, TJs have been implicated in asthma pathogenesis. TJs consist of a series of interacting proteins and receptors including zona occuldens (ZO) 1-3, occludin and claudins 1-5, as well as adhesion proteins, which decrease epithelial resistance and promote junction disassembly [162] . Epithelial cells are also important as producers of many factors that enhance and/or regulate inflammation. Endothelin (ET) family members are potent bronchoconstrictors and have a number of other important biological properties in asthma pathophysiology. Expression of immunoreactive ET is increased in asthmatic subjects receiving only [beta]2 agonists, but not CS therapy [163] . In addition, epithelial TLR2-6 and NOD1-2 expression is important in innate immune system function and production of TNF, IL-1[beta], IL-6, GM-CSF, IL8[164] and other chemokines. In addition, TSLP, an indispensable cytokine in the development of allergic diseases (see later in the section entitled 'Thymic stromal lymphopoietin'), was recently shown to be produced by primary airway epithelial cells following treatment with cytokines [165] , poly I:C [166] and/or medications [167] . Mouse studies indicate that not only do IL-25 and activin A play a role in driving airway hyperreactivity and remodeling [168] , but also IL-33, a member of the IL-1 family (see later in the section entitled 'IL-33 [IL-1 family]'). The latter was found to be elevated in BAL and epithelial cells from asthmatic patients [169] . Airway epithelium may also generate nitric oxide (NO) in exhaled breath, which may be a useful marker for asthma exacerbation, especially in children (see later in the section on endothelial cells).

Creola bodies (CrBs) are conglomerates of epithelial cells, present specifically in the sputum of asthmatic patients. CrBs are clinically useful in estimating airway hyperresponsiveness through correlation with levels of eosinophil cationic protein in the sputum [170] . Epithelial cell proliferation contributes to the thickened lamina reticularis in severe asthma [171] . There is also evidence that epithelial injury and repair are abnormal in asthmatic airways. Studies show that the expression of EGF receptor (EGFR or ErbB1) and the upstream tyrosine kinase are increased in asthmatic bronchial epithelium, but are CS insensitive [172,173] . Recently, Enomoto et al. showed that two of the EGFR ligands, EGF and AREG, enhance proliferation and increase mucin production from epithelial cells in vitro. The importance of these factors in the pathogenesis of chronic asthma was further supported by in vivo /ex vivo data from asthmatic patients [157] . During airway inflammation, the number of goblet cells within the airway epithelium and the amount of intralumenal mucus are markedly increased (525%) and correlate significantly [174,175] . Normal airway mucus stays on the epithelial surface and provides an important innate immune function by detoxifying noxious molecules and by trapping and removing pathogens and particulates from the airway by mucociliary clearance. Among more than 20 identified mucins, 12 have been shown to be expressed in the airway epithelium [176] . Among two structurally and functionally distinct classes of mucin, the gel-forming secreted mucins, MUC5AC and MUC5B, are most prominent in the airways [175] . The CysLT, LTD4 , stimulates mucus secretion and impairs mucociliary clearance [177] . To support this, LT modifiers, including LT receptor antagonists (LTRAs), inhibit MUC2 gene expression in cultured human airway epithelial cells [178] . Epithelial cells become migratory in response to growth factors, such as EGF or TGF-[beta], by undergoing an epithelial-to-mesenchymal transition (EMT), characterized by downregulation of TJs and increased expression of MMPs and ECM components (see later in the section on ECM proteins). In other lung diseases, EMT has been attributed to pathological changes associated with tissue fibrosis [179] ; however, markers such as [alpha]-smooth muscle actin or vimentin are not expressed in asthmatic airway epithelium [180] , and there is currently no evidence that epithelial cells invade the basement membrane and contribute to the mesenchymal compartment in asthma [173] . Furthermore, in cultures of epithelial cells from children, the asthmatic airway epithelium exhibits dysregulated repair responses to injury[181] , showing that fibronectin is an essential component of the provisional ECM [182] . Ultimately, all of these factors operate together and interact to produce functional changes in the epithelium,

which lead to the structural changes characteristic of the asthmatic airway epithelium. Airway smooth muscle cells Fatal asthma has revealed two major causes for reversible airway obstruction, namely: mucus hypersecretion and smooth muscle spasms [183] . As a result, beyond hematopoietic cells, ASM is considered among the major tissues responsible for bronchial asthma. ASM cells are key effector cells in chronic asthma and an important source of proinflammatory cytokines, chemokines and other mediators [184] . ASM hyperplasia and hypertrophy are crucial factors in airway remodeling [185,186] , described mainly in severe, CS-refractory asthma. Thus, it is likely that the ASM milieu contributes to asthma airway obstruction chronicity and hyperresponsiveness [149] . A study describing MC infiltration of ASM in asthma supported the speculation that the interactions between ASM and infiltrating MCs may be a key element in the development of asthma hyperreactivity [187] . Although some strong opposing views remain, data support the view that MCs may be present within the ASM compartment and also in the bronchial mucosae of allergic but not nonallergic asthmatics [188] . Bronchoconstriction, a cardinal feature in asthma, may limit the increase in resistance to gas flow caused by ASM shortening and/or congestion as well as airway wall edema. Many endogenous paracrine mediators, putatively involved in asthma and bronchial hyperresponsiveness, have both bronchomotor and vascular effects [184] . Like epithelial cells, ASM cells produce TSLP but in higher quantities compared with epithelial cells [167] , in addition to other proinflammatory cytokines. Very recently, increased expression of IL-33 was shown in human asthma, along with TNF expression, seen mainly in severe asthma phenotypes. IL-33 expression levels in ASM cells were upregulated by TNF and IFN-γ, but not by any Th2 cytokines [189] . Moreover, a disintegrin and metalloprotease (ADAM) 33 protein (Figure 1), the first asthma susceptibility gene to be identified by positional cloning, was shown to be present in mesenchymal progenitor cells in developing lungs and in ASM [190] . These studies indicate that ASM cells undergo functional changes that contribute to airway remodeling; the latter continues to be a major therapeutic target in asthma (aside from inflammation). Abnormalities in ASM that surround the airways and regulate lumen diameter may contribute to asthma pathogenesis [191] . In nonfatal disease, airway wall thickening is shown to be markedly increased in small membranous and mid-sized airways, in contrast to fatal asthma where thickening is seen more widely throughout the bronchial

tree [192] . A report showed a three- to fourfold increase in muscle volume in asthmatic airways compared with normal subjects [160] . Increase in ASM mass occurs due to several factors, including ASM cell proliferation (hyperplasia) induced by inflammatory mediators (CysLT, TX and histamine) [193-195] , cytokines [196] and growth factors [192,197] . Other factors include repeated episodes of bronchospasm or reduced inhibitory control, resulting in myogenic activity and ASM cell hypertrophy [198,199] , depending on the severity of asthma [187,192] . Morphological changes can be determined using markers including smooth muscle myosin heavy chain, calponin, [alpha]-actin, desmin and collagen I [200] . These changes also cause the ASM cells to lose contractility and flexibility. There is a correlation between asthma severity and the magnitude of ASM thickening, consistent with a recent study showing that ASM thickening is more significant in fatal compared with non-fatal asthma [191] . Fibroblasts Fibroblasts are the most common cell types of the connective tissue in all animals. They synthesize ECM and collagen and play a critical role in tissue remodeling. In addition to ASM hypertrophy and hyperplasia, subepithelial fibrosis is another important feature of airway remodeling. In asthmatics, fibrosis is detected below the basement membrane (lamina reticularis) due to increased deposition of ECM, particularly collagen type I and III, fibronectin and proteoglycans [201] . This fibrosis is consequence of fibroblast activation, through upregulation of a number of cytokines (TGF-[beta] and IL-11) and growth factors [151] . TGF-[beta] is a major fibrogenic cytokine [202] , the increased expression of which significantly correlates with markers of remodeling that include lamina reticularis thickness and increased fibroblast numbers, in both asthma and chronic bronchitis [203] . TGF-[beta] plays separate important roles in regulating inflammation, cell growth and differentiation, including ECM metabolism (Figure 1) [204] . In asthmatic airways, increased levels of TGF-[beta]1 correlate predominantly with increased numbers of fibroblasts, ASM, eosinophils, macrophages, airway ECM and epithelial cells [205-207] . Recently, osteopontin, an ECM protein, was also shown to contribute to fibrosis by regulating cell-ECM interactions and TGF-[beta]1 modulation [208] . Interestingly, remodeling was also suggested to be beneficial in airway diseases in that the airway wall thickening may protect against bronchoconstriction [209] . Collagen deposition in the subepithelial matrix can inhibit narrowing by making the airway wall stiffer, thereby creating an additional load on the ASM [210] .

Experimental models of allergen-induced asthma suggested that on the one hand airway responsiveness may increase after airway inflammation, but, on the other, may decrease with airway fibronectin and collagen deposition[207,209] . Endothelial cells The literature on angiogenesis and microvascular remodeling in human airway disease is relatively limited, although evidence suggests changes in airway microvasculature are a feature of asthma. It is also recognized that the airway mucosa in fatal asthma is edematous and contains dilated, congested blood vessels [211] . Nitric oxide is a short-lived free radical gas involved in diverse physiological and pathological processes and catalyzed by three different isoforms of NO synthase (NOS; neuronal, inducible and endothelial). Endothelial NOS (eNOS) is a constitutively expressed latent enzyme and requires a higher concentration of Ca2+ for its enzymatic activity, indicating that endothelial cells are important in airway NO. NO production from endothelial cells is stimulated by a variety of mechanical forces, including shear stress and cyclic strain, as well as humoral factors ranging from growth factors to peptide hormones, including acetylcholine, VEGF, bradykinin, estrogen, hydrogen peroxide and angiotensin II. In bronchial blood vessel vasodilatation and/or microvascular leakage, eNOS plays a crucial role [212] . In addition, NO released from the endothelium modulates other processes including platelet aggregation, platelet and leukocyte adhesion to the endothelium, and ET-1 generation [213] . Increased levels of VEGF in the airway have been implicated in the pathogenesis of asthma [214] . VEGF increases microvascular permeability, leading to plasma protein leakage into the extravascular space, and subsequent mucosal edema and narrow airway diameters, which possibly amplifies the ASM contraction effect [211,215] . Blood vessels in asthmatic airways are in a hypervascularized, destabilized state, which contributes to upregulation of airway microvascular permeability [216] . To date, numerous studies have shown that VEGF is critically involved in microvascular alterations in asthma [214,215,217] . In a murine model of bronchial asthma, lung-specific and inducible overexpression of VEGF was shown to enhance not only angiogenesis but also Th2-driven airway inflammation and hyperresponsiveness[218] . Moreover, subcytotoxic concentrations of major basic protein (MBP) are shown to possess proangiogenic effects in vitro and in vivo , and enhance the promitogenic effect of VEGF, but they do not affect VEGF release [219] . In humans, inhaled CSs (ICSs) reduced airway VEGF production effectively, thereby affecting submucosal vascularity [220,221] . The process of angiogenesis is known to be regulated by a

complex interaction of activators and inhibitors, and inappropriate production of these factors may cause aberrant angiogenesis. A recent report clearly showed the importance of the CXCR3 chemokine, I-TAC (CXCL11), which inhibited IL-8mediated angiogenesis [222] . Recently, Matsuda et al. have shown that inflammation (only in the presence of TNF) mimics in vitro conditions, whereby IL-4/13 and IFN-γ reciprocally regulated angiogenesis. This occurred through autocrine synthesis of CXCR2 and CXCR3, respectively, and induced ICAM-1 and VCAM-1 expression [223] . It should be noted that unlike MBP, CXCR2-induced angiogenesis is independent of VEGF. Furthermore, unlike VEGF, CS therapies offer little hope for improving TNF/CXCR2 chemokineassociated angiogenesis. This is due to their minor effect on TNF-mediated pulmonary microvascular alterations, including CXCR2 chemokine-mediated angiogenesis, CXCR3 chemokine-mediated avascularity and inflammatory cell recruitment [224] . Recently, endothelial cells were reported to express IDO and generate KYN catabolites with the capacity to modulate adaptive immunity [225] . Although the study was on sepsis, systemic inflammation was shown to induce tryptophan catabolism and KYN generation by endothelial cells. Remarkably, the metabolism of arginine to NO is functionally related to tryptophan catabolism via IDO. In this study, KYN also activated cAMP pathways, which are important in smooth muscle cell function in asthma [225] . In a recent study, our laboratory reported that IDO may also be an important regulator of tolerogenic responses to allergen [226] ; however, studies are underway to determine the role of NO in this process. Cytokines, chemokines & growth factors The literature is replete with recent reviews and articles documenting the numerous players in the context of cytokines and mediators involved in asthma pathogenesis [227,228] . IL-17 family IL-17A (IL-17) is a potent proinflammatory cytokine produced by activated Th17 cells. In addition to IL-17A, five other cytokines identified so far are recognized as a new IL-17 family that includes IL-17B, IL-17C, IL-17D, IL-17E (IL-25) and IL-17F [229] . IL-17A increase in asthmatic patients has been described and its level is correlated with that of airway hypersensitivity [116,230] , which implies a

contribution to asthma pathogenesis. IL-17A levels were profoundly elevated in the sputum of severe asthmatic patients with increased neutrophilia [116] . Among the IL-17 family members, IL-17F has the highest amino acid sequence homology to IL-17A [231] and is expressed in the airway of asthmatics; its expression was shown to be correlated with disease severity. IL-17F is derived from several cell types - not only from Th17 cells, but also MCs and basophils, and shows a wide tissue distribution, including the lung [231] . Both IL-17A and IL-17F have been reported to bind to the same receptor complexes, IL-17RA and IL-17RC [232] . Indeed, both IL-17A and IL-17F failed to induce chemokine expression in either il17ra -/ or il17rc / cells [233,234] . The expressions of IL-17RA and IL-17RC are different; IL-17RA is highly expressed in lymphoid tissues, whereas IL-17RC is mainly expressed in nonhematopoietic tissues [234] . Furthermore, IL-17A and IL-17F are secreted as both homodimers and heterodimers [235,236] , and recent studies showed that the IL-17A-IL-17F heterodimer is more potent than the IL-17F homodimer but less potent than the IL-17A homodimer in inducing chemokine expression [230] . IL-17A stimulates airway epithelial cells, ASM cells and fibroblasts to produce various proinflammatory mediators such as IL-6, IL-8, IL-11, Groa (CXCL1), LARC (CCL20), granulocyte colony-stimulating factor, MUC5AC, MUC5B, [beta]-defensin-2 and ICAM-1[41,43] . These mediators contribute less to the development of Th2 inflammation, but recruit neutrophils, which are particularly found in severe and CS-insensitive asthma. Thymic stromal lymphopoietin Increasing evidence suggests that this cytokine may play an important role in the pathogenesis of allergic diseases, including bronchial asthma [10,237] . TSLP was originally isolated from a mouse thymic stromal cell line and considered a growth factor for lymphocytes [167] . The most clinically relevant role of TSLP is mediated by DCs through the induction of OX40L expression on DCs [10,238] . Recently, it was found that TSLP is produced by epithelial cells [166] , smooth muscle cells [239] , fibroblasts [240] and MCs [131] . While epithelial cells produce TSLP in high concentrations following stimulation with IL-4 and dsRNA [166] , maximum induction of TSLP was seen in mesenchymal-type cells when stimulated with IL-4, TNF and cAMP-elevating reagents [167] . TSLP expression is increased in asthmatic airways[241] . Furthermore, transgenic mice overexpressing TSLP in the lung develop severe airway inflammation [242] . Several cellular targets of TSLP (i.e., expressing

TSLP receptors) have been identified, including DCs, lymphocytes and granulocytes [228] . Naive CD4+ T cells develop into Th2 cells upon antigen presentation from TSLP-primed DCs [10,238] . TSLP has a wide range of celltargeted effects, which include inducing CD8 + T-cell cytotoxicity and OX40L expression, generating B-1 cells from precursors, decreasing IL-12 and IL-23 p40 production in and from DCs, increasing IL-13 production from NKT cells and MCs, and tissue recruitment of eosinophils [228] . In fact, under TSLP-primed conditions, Th2 cells additionally release TNF; indeed, these cells are now referred to as 'inflammatory Th2 cells'[25] . Thus, TSLP may have the capacity to initiate allergic inflammation independently. IL-33 (IL-1 family) IL-33 (also known as IL-1F11) is a recently identified member of the IL-1 family of cytokines that mediates its biological effects via the receptor ST2 (also known as IL1RL1, DER4, Fit-1 or T1) [243] . Over the past decade, various studies have established the ST2 receptor as a selective marker of both murine and human Th2 cells [244] . In addition to Th2 cells, recent studies demonstrated that ST2 is also expressed on MCs [245,246] , eosinophils [247,248] , basophils [249] and lung tissue [250] , but not on Th1 cells or neutrophils. IL-33 drives these cells potently to produce proinflammatory Th2 cytokines and chemokines by an NF-κBindependent mitogen-activated protein kinase (MAPK)-dependent pathway [244] . In addition, circulating CD34+ hematopoietic progenitor cells are reported to express ST2 and respond to IL-33 by rapidly releasing high levels of Th2 cytokines [251] . Furthermore, IL-33 directly activates eosinophils, basophils and DCs [252] , and its expression by ASM cells [189] and epithelial cells [169] is shown to be increased in asthma. Interestingly, a genome-wide association scan analysis showed that a singlenucleotide polymorphism in st2/il1rl1 correlated most strongly with asthma in a collection of ten different populations (7996 cases and 44,890 controls) [253] . A polymorphism in il33 that showed a suggestive association with the circulating eosinophil count was also significantly associated with atopic asthma [253] . These findings further support the pathophysiological relevance of the IL-33/ST2 pathway to asthma. Thus, IL-33 appears to have a potential role in Th2 responses and may be crucially involved in the development and/or exacerbation of chronic allergic inflammatory diseases, including atopic asthma. Interestingly, a number of IL-1 family members are produced in response to TLR signaling [254] . Unlike IL-2 and other cytokines, most members of the IL-1 family do not directly affect lymphocyte function. However, these cytokines are purported to amplify the danger message in the context of the innate immune

system and serve as a crucial link in the transition between innate and appropriate adaptive immune responses [255] . Very recently, IL-4-independent IL-9 production, shown to have multiple effects in the development and maintenance of allergic inflammation and airway remodeling) by CD4+ T cells [256] , was shown to be triggered following TGF-[beta] interaction with IL-1 family members, IL-1[alpha], IL-1[beta], IL-18 and IL-33 [257] . ECM proteins Airway remodeling is characterized by subepithelial deposition of ECM molecules, including collagens and proteoglycans, which correlates with increased fibroblast/myofibroblast numbers. Connective tissue cells elaborate ECM macromolecules that fill the extracellular spaces of airway walls. The composition of secreted ECM depends on cell types, their state of differentiation and their metabolic status. ECM molecules include fibrous proteins (collagen, osteopontin and elastin), structural or adhesive proteins (fibronectin and laminin) expressed during morphogenesis and tissue repair. These proteins are modified proteoglycans, such as heparan sulfate, chondroitin sulfate, keratan sulfate or non-proteoglycan polysaccharide, and glycosaminoglycans, including hyaluronic acid or hyaluronan (HA). HA is normally found in small concentrations in adult tissues and its concentration is increased during wound healing. HA facilitates cell migration and proliferation during injury and repair [258] and its levels are increased in asthma BAL in association with asthma severity [259] . While collagen, a major ECM protein, exhibits hyperplasia in some asthmatics, its fibers may be irregularly deposited [205] .The exact potential of collagen as a remodeling factor in asthma requires further investigation. Asthmatic subepithelial tissue contains significantly more collagen I and III than normal controls [207] . TGF-[beta] is equally important in regulating synthesis and deposition of collagen in the airways. The latter has greater functional implications than the increase in lamina reticularis thickness. Elastin is a cross-linked protein that confers elasticity (recoil) on tissues and is indispensable in tissues that bend, twist and stretch reversibly. Subepithelial elastic fibers in asthmatic airways are fragmented and, in deeper layers, are often found to be twisted and thickened [260]. The abnormality of the subepithelial elastic fibers is a fatal asthma phenotype [198] . Chitinases Chitin, a [beta]1,4-linked polymer of N -acetylglucosamine, is the main component of the cell walls of fungi, arthropod exoskeletons (house dust mite),

crustaceans, insects and worms. Chitin is degraded in nature by chitinases and [beta]-N -1-4-acetylhexosaminidases (chitobiases). Chitinases are members of the glycoside hydrolase family 18, which retain glycoside hydrolases found in various organisms ranging from bacteria to humans [261] . They also include enzymatically inactive chitinase-like proteins (CLPs), and other digestive enzymes that break down glycosidic bonds in chitin. In humans, acidic mammalian chitinase (AMCase), chitotriosidase, oviductin, human cartilage glycoprotein-39kd (HcGP-39)/YKL-40 and YKL-39 have been described [262] . Asthma has been linked to enhanced chitinase expression levels [262,263] . This may explain the nature of interaction with common chitin-rich airborne allergens (dust mites and mold spores) and suggests a link between allergy and chitinous helminthic parasites with implications for the hygiene hypothesis [263,264] . CLPs are upregulated in both human and murine models of allergy and asthma [265] . For instance, AMCase is an enzyme that is acid stable, with an optimum pH of 2.0, and has chitinase bioactivity; this is upregulated in asthmatic airways in response to IL-4 and IL-13 [266] , suggesting that AMCase may exacerbate allergic reactions. Levels of YKL-40 are also reported to be increased in the blood and lungs of asthmatics in association with disease severity. Animal studies using BRP-39 (homologue of YKL-40 in mouse)-null mutant mice demonstrated that BRP-39 is required for optimal allergen sensitization and Th2 inflammation [267] . However, overexpression of chitinase alone is not a sufficient stimulus to drive asthma, but increased chitinases and CLP production may be a consequence of asthma [265] . Allosamidin, a strong inhibitor of family 18 chitinases, attenuated asthmatic responses induced by IL-13 without affecting PGE synthesis in an ovalbumin-induced asthma model [268] . However, nonspecific inhibition of chitinase activity may result in greater susceptibility to fungi, parasitic worms and allergy against inhaled chitin [269,270] . This field is still in its infancy but has strong potential to reveal exciting new insights. Current treatments for asthma Corticosteroids Introduced in the 1970s, CSs have revolutionized the field of asthma therapeutics as the most effective therapeutic strategy to control asthma-associated airway inflammation and, to a lesser degree, AHR, depending on clinical variability of the patients. By binding to ubiquitously expressed cytoplasmic CS receptors, CSs cause repression of inflammatory genes and increased transcription of antiinflammatory genes. CSs even decrease the inflammatory cell numbers in the sputum, especially eosinophils, suppress inflammatory immune responses and reduce proinflammatory cytokine/chemokine expression, including MIP-1[alpha]

(CCL3), IL-1[beta], IL-4, IL-5, IL-13, GM-CSF, TNF and RANTES (CCL5) [185] . Even though different categories of CSs are available, such as topical/inhaled and systemic (oral), considering the inflammatory condition in asthma, ICS became the first choice of treatment in patients with persistent asthma, according to the GINA guidelines [271] . Over the years, several randomized clinical trials (RCTs) have demonstrated the clinical efficacy and safety of ICS treatment. This class of agents was extremely effective in alleviating asthma scores, improving pulmonary functions, reversing airway mucosal thickening (remodeling), reducing goblet cell hyperplasia and decreasing AHR in moderate-to-severe asthmatics [271] . Similarly, sustained ICS treatment in certain studies demonstrated a marked reduction in asthma exacerbations combined with decreased hospitalizations, which suggests the efficacy and success of ICS therapy in asthma, especially in eosinophilic asthma [272,273] . Current evidence suggests the possibility of reducing eosinophilic exacerbations in asthma by controlling eosinophilia using ICS; indeed, there is a poor correlation between sputum eosinophil numbers, disease symptoms and airway function [274] . Therefore, comprehensive measurement of airway inflammation using methods such as quantitative sputum cell counts is essential to guide asthma treatment and determine minimum CS dosage required for controlling asthma exacerbations, especially in adults [275,276] . ICS treatment benefits have also been shown in children, where studies showed improved MCh airway challenge and decreased requirement for rescue medication in children treated with budenoside 200 g/day [277] . Ironically, long-term treatment using high doses of ICS (800-1600 g budenoside or equivalent) can cause certain local side effects (e.g., oral candidiasis, cough, pharyngitis, hoarseness and dysphonia) and systemic side effects (eg., growth retardation in children, osteoporosis and increased susceptibility to diseases such as glaucoma, cataract, adrenal gland suppression and diabetes) [271] . CSinsensitivity is another important limitation with ICS therapy in asthma, where 510% of severe asthma patients do not show any improvement in disease symptoms even following maximum ICS dosage, which suggests the need to develop alternative treatment strategies for this type of asthma [278] . Concerted efforts towards effective and safe ICS lead to the development of drugs such as ciclesonide and nonsteroidal selective glucocorticoid receptor activators like AL438 and ZK-216438, which are in different phases of clinical trials [279] . Long-acting [beta]2-agonists

The use of long-acting [beta]2-agonists (LABAs; e.g., salmeterol or formoterol) in combination with low- or medium-dose ICS is highly recommended as an addon therapy for alleviating both asthma-associated airway inflammation and bronchoconstriction, specifically in persistent asthmatics [271,280] . LABA administration causes b2-receptor activation in ASM cells, resulting in smooth muscle relaxation and induction of rapid bronchodilatory responses [281] . However, LABAs alone do not have any anti-inflammatory properties or improve asthma scores [282] . Some meta-analysis studies even reported enhanced sputum eosinophil counts combined with increased hospitalization incidents, lifethreatening disease exacerbations and asthma-related mortality with LABAs alone, which reinforced asthma treatment guidelines that recommend the use of LABAs only in combination with ICS [283] . Formoterol And Corticosteroids Establishing Therapy (FACET) and other RCTs confirmed the efficacy of LABA/ICS combination therapy, where budenoside/formoterol combination in moderate-to-severe asthmatics resulted in improved asthma symptoms, reduced ICS dosage and improved long-term disease control by restoring pulmonary functions, decreasing sputum counts and reducing asthma exacerbations [284] . However, dosage requirements and longterm risks of LABA/ICS treatment pertaining to asthma-related morbidity and mortality need to be evaluated further [277] . Recent advancements in LABAs include the development of drugs such as indacaterol and carmoterol, which are capable of acting for even longer than the conventional LABAs [279] . Recent advances in asthma treatment Despite major advances in our understanding of asthma pathophysiology and immunopathology, it is very disappointing that the field of asthma, whether clinically, pharmacologically or at the basic science level, has made little or no serious advances towards novel therapies. An exception to this (and beyond the classic combination of LABA/ICS) has been the development of LTRAs. This places an enormous burden of responsibility on researchers in asthma as well as on the pharmaceutical industry to promote and accelerate funding for studies aimed at developing novel and efficacious asthma therapies. It is agreed that asthma is a complex disease and as such no single therapy targeting only one of the myriad of factors involved is likely to succeed. Complex diseases like asthma require complex combination therapies to reduce the clinical and economic burden. In addition, asthma phenotypes are complex and at the moment difficult to pin down, so attempts are being made to narrow down these phenotypes using various genomic, proteomic and metabolomic approaches [285-293]. The ultimate aim is to design therapies that are targeted to the individual patient rather than the current 'shotgun' approaches that have proven to be limited. This failure has

contributed to serious non-compliance (adherence) with current treatments due to lack of efficacy in a significant percentage of patients. Leukotriene receptor antagonists Cysteinyl LTs including LTC4 , LTD4 , and LTE4 are endogenous arachidonic acid derivatives that potently modulate asthma-associated bronchoconstriction and airway remodeling induced by eosinophils and MCs. The ability of LTRAs (e.g., montelukast [Singulair , Merck and Co., Inc., QC, Canada], zafirlukast [Accolate , AstraZeneca, ON, Canada] and pranlukast) to inhibit the production of CysLTs, block LTD4 receptors and function independently of short-acting [beta]2 agonists in alleviating bronchoconstriction and airway inflammation renders them interesting alternatives as add-on therapies to ICS in the treatment of moderate-to-severe asthma [294] . Placebo-controlled Phase III RCTs demonstrated that LTRAs are moderately effective in controlling asthma; clinical responses apparently differed based on patient variability [295] . Studies on exercise-induced bronchoconstriction or allergen-induced asthma demonstrated a 30-70% reduction in exercise-induced bronchoconstriction responses with orally administered LTRAs [296] . Subsequent clinical studies demonstrated reduced bronchoconstriction, 15% improvement in pulmonary function, decreased exacerbations and reduced CS dosage when LTRAs were used as an add-on therapy [297,271] . Studies also demonstrated a mild bronchodilatory effect, which suggests the use of LTRAs as a second-choice treatment in children with mild asthma [298] or patients with exercise- [295] and aspirin-induced bronchoconstriction [299] . However, evidence for LTRA efficacy in asthma remains inconclusive as certain clinical studies have shown that although LTRAs can reduce airway eosinophilia [300] , the effect is weaker compared with ICS [301] . In addition, various RCTs have demonstrated a better efficacy when LABAs were used as an add-on treatment instead of LTRAs [302,303] . Interestingly, treatment with LTRAs is suspected in the pathogenesis of Churg-Strauss syndrome [304] , suggesting a need for additional studies on LTRA mechanism and function. Immunological therapeutic approaches IgE, IgA & IgG Allergy is defined as a hypersensitivity reaction mediated by humoral or cellular immunological mechanisms, where IgE is the predominant antibody involved in allergic (atopic) reactions involving degranulation of inflammatory cells and

inflammatory mediator release [305,306] . In fact, serum IgE levels in atopic individuals are up to ten-times the levels in nonatopic subjects [305] . IgE receptors, also known as Fce receptors, are of two types: FcεRI, the high-affinity IgE receptor that is only found on MCs and basophils; and the lowaffinity FcεRII, also known as CD23, found constitutively on human B cells, but expressed on eosinophils, monocytes, macrophages and platelets following stimulation with IL-4. In airway inflammatory diseases, local B-cell class-switch recombination and IgE synthesis mediate activation of airway MCs and eosinophils respectively, following antigen exposure [307,308] . Class-switch recombination requires two signals [309] : T-cell cytokines (including IL-4, IL-10, IL-13 and TGF-[beta]), and B-cell activation that occurs via interaction with members of the TNF superfamily. These include CD40 B cell-activating factor of the TNF family (BAFF; also known as BLyS) and a proliferation-inducing ligand (APRIL) [310] . BAFF and APRIL are important in B-cell survival, proliferation, maturation through engagement of transmembrane activator and calcium-modulator and cytophilin ligand interactor (TACI) and B-cell maturation antigen (BCMA) receptors on B cells. BAFF and APRIL were recently discovered to play a role in human B cell immunoglobulin isotype switching [311,312] , and may, therefore, be worthy of pursuit in atopic asthma therapy. IL-4 and IL-13 are known to induce IgE production from B cells [313] . BAFF levels were found to be elevated in airways after allergen challenge in allergic subjects [307-309] , Interestingly, BAFF levels at allergen-challenged sites had a high correlation with IL-13 and IL-6 levels [309] , suggesting an important role of BAFF production in regulating local antigen-specific IgE production by allergen [309] . Because increased IgE is an important feature in atopy, treatment with omalizumab (humanized anti-IgE mAb) appears to result in recognition and neutralization of existing IgE, thereby preventing IgE from binding to MCs. However, omalizumab is effective for certain types of asthmatics [314] (see later in the section 'Anti-IgE'). In addition to IgE, allergen and bacterial specific IgA and IgG are also measureable in atopic asthmatic sputum [315] . IgA or secretory IgA (sIgA) contained in mucins including those collected via BAL plays an important role in both mucosal immunity and regulating inflammation [316] . In atopic asthmatics, IgA levels in BAL fluid are higher than controls [317] . Interestingly, IgA/sIgA levels and eosinophil cationic protein (a marker for eosinophil degranulation) showed significant correlations between BAL fluid and sputum from atopic asthmatic patients [318] . Thus, increased IgA levels and priming for IgA-induced eosinophil stimulation may further contribute to asthma inflammation.

IgA and IgG4 were described recently as early allergen-specific immunotherapy (SIT) markers [319] , where induction of peripheral T-cell tolerance is crucial and is initiated through autocrine actions of IL-10 and/or TGF-[beta] produced by the antigen-specific Treg [320] . Tr1 cells, a subset of Treg cells, are shown to be induced upon antigen exposure under certain tolerogenic conditions characterized by the production of immunosuppressive cytokines, mainly IL-10 [321] , which counter-regulates antigen-specific IgE and IgG4 synthesis [322] . IL-10 is a potent suppressor of both total and allergen-specific IgE, while it skews T-cell antigenspecific responses from an IgE- to IgA- and IgG4-dominated phenotype accompanied by increased TGF-[beta] and IL-10 levels from allergen-specific T cells, respectively[322,323] . A change in the IgG4/IgE ratio occurs within weeks as a result of an early increase in IgG4 [324] . Furthermore, IgG4 does not induce complement cascade activation but inhibits immune complex formation by other isotypes [325] . Therefore, IgG4 antibodies can be viewed as an initial marker when allergen is introduced in allergen SIT. Sublingual immunotherapy (SLIT) is another currently available/applicable allergen SIT, although the mechanism of SLIT action is not yet well understood [326] . Animal experiments, however, reveal that mucosal vaccinations with antigens are potential candidates for mucosal tolerance induction [327] , where Tregs seem to provide help for IgA and IgG4 production via TGF-[beta] and IL-10 secretion from Th3 and Tr1, respectively. Therefore, further investigations are warranted to study the role of Tregs and TGF-[beta] in allergic subjects treated with SLIT [327] . In addition, IDO appears to play a role in maintaining mucosal tolerance in a murine model of airway sensitization [226] . Intravenous immunoglobulin Chronic airway inflammation in asthma can contribute to irreversible airway structural changes, which may not respond to currently available therapies [149] . Over the past 20 years, administration of exogenous intravenous immunoglobulin (IVIg) has become an important therapy in clinical medicine [328] . The potent anti-inflammatory properties of IVIg have been recently exploited in clinical trials involving severe chronic asthmatic patients, where IVIg decreased total serum IgE levels, inhibited lymphocyte activation, reduced IL-2 and IL-4 production in vivo , suppressed cytokine-dependent lymphocyte proliferation in vitro and lessened immediate skin test reactivity to allergens [329] . Even in severe, CS-refractory asthmatic patients, IVIg increased lymphocyte sensitivity to CS effects, remarkably reduced oral/topical CS dose and asthma symptoms, improved peak expiratory flow rates (PEFR) and decreased total serum IgE levels [330] . The effects of IVIg were also attributed to increased glucocorticoid receptor binding affinity in vitro [328] .

Anti-IgE Discovered in 1967, IgE is an obvious therapeutic target for treating atopic conditions, including asthma. Over a decade ago, a humanized anti-IgE mAb was successfully developed (omalizumab or Xolair [Novartis Pharmaceuticals, NJ, USA]) as a treatment for chronic allergic respiratory diseases [331] . Omalizumab (rhuMAb-E25) binds to the Cε3 epitope on IgE with a 10 10 M affinity, thus preventing FcεRI- and FcεRII-IgE interactions, and thereby abrogating cross-linking and degranulation [332] . Studies indicate that a single subcutaneous anti-IgE injection reduces unbound circulating serum IgE concentration (84-99%) within approximately 60 min and lasts for 4-6 weeks. Engagement of serum IgE with omalizumab causes a rapid reduction in the cell surface expression of FcεRI, receptor engulfment and degradation, leading to desensitization and decreased histamine release from MCs and basophils [333] . Recent studies suggest that omalizumab inhibits the allergen presentation capacity of DCs, promotes eosinophil apoptosis and downregulates the induction of the proinflammatory Th2 cytokines IL-4 and IL13 [334] , but was shown to induce serious anaphylactic responses in only 0.1% of treated patients [335] . Meta-analysis data from several randomized double-blind, placebo-controlled multicenter clinical trials have demonstrated the safety and efficacy of omalizumab in asthmatics, including children. Subjects demonstrated a marked improvement in asthma symptoms and pulmonary functions, reduced airway exacerbations, decreased AHR, enhanced morning PEFR and reduction in CS dosage, resulting in improved overall quality of life [336] . However, in none of these studies did omalizumab show any effect on bronchoconstriction and MChinduced AHR. Clinical studies in nonatopic patients demonstrated that omalizumab treatment reduces nasal irritability and causes a marked decrease in antihistamine requirement [337] . Omalizumab may be a promising therapeutic strategy in moderate-to-severe persistent allergic asthmatics who do not respond well to other therapies. Further evaluation of the antibody is, however, required to demonstrate its long-term side effects and determine whether it is suitable for treating pediatric as well as nonatopic subjects [338] . Anti-IL-5

Increased BAL eosinophil numbers correlate with increased IL-5 levels and AHR [339] . Therefore, blocking IL-5 activity using the humanized anti-IL-5 mAb, mepolizumab, may be a potentially effective therapeutic strategy for asthma [340] . Pilot studies on mepolizumab have demonstrated the lack of significant improvement in clinical outcomes in asthmatics [341] , although some of these studies suffered from improper designs [342] . However, Flood-Page et al. showed that mepolizumab depleted less than 50% of bronchial tissue and bone marrow eosinophils, while significantly diminishing blood and BAL fluid eosinophils in treated subjects [343] . Two recent studies targeted asthma subjects using mepolizumab only in patients with greater than 3% sputum eosinophilia [77] ; these provided convincing evidence for a reduction of recurrent asthma exacerbations accompanied by an improvement in eosinophil counts, asthma symptoms and FEV1 values. Mepolizumab is safe and effective and there are no predisposing factors or serious risks to adverse reactions [77,343,344] . Further careful assessment is needed to determine whether mepolizumab would be an important adjunct therapy in eosinophilic asthma. Anti-TNF TNF is an important Th1 proinflammatory cytokine involved in regulating multiple pathological and inflammatory conditions including rheumatoid arthritis (RA) and inflammatory bowel disease (IBD). In asthma, elevated TNF levels result in activation and elaboration of proinflammatory cytokines/chemokines (IL-5, IL-8 [CXCL8], IL-13, eotaxins [CCL11, CCL24 and CCL26], MCP-1 [CCL2], RANTES [CCL5], granulocyte colony-stimulating factor and GM-CSF), inflammatory cell recruitment, upregulation of ICAM-1, proliferation and differentiation of myofibroblasts, increased ECM protein secretion, mucins, neuropeptides and ET-1, resulting in bronchial hyperplasia and airway remodeling. Randomized controlled trials using US FDA-approved TNF inhibitors (e.g., infliximab, etanercept, adalimumab, certolizumab pegol and golimumab) in moderate-to-severe asthmatics demonstrated the efficacy of anti-TNF mAbs in effectively reducing PEFR variations, but they were ineffective in improving lung function and disease exacerbations [345-348] . Interestingly, the golimumab study was discontinued after 24 weeks of treatment due to serious adverse reactions that included malignancies and infections with life-threatening pathogens [349]. Other immunotherapies under investigation

In recent years, humanized mAbs have been developed against different candidate targets of asthma, some of which are undergoing Phase II and III clinical trials such as daclizumab, a humanized mAb against CD25/IL-2 receptor[alpha]. Early Phase I and II clinical studies on daclizumab in mild-to-severe asthmatics demonstrated encouraging results in controlling disease symptoms, including reduced exacerbation intervals, decreased blood eosinophil and serum IgE counts, and improved pulmonary function associated with decreased use of ICS. However, further studies are required before daclizumab can be used in future asthma therapy [350] . IL-9 is another interesting target thought to be involved in MC activation and mucus production in asthma. Early Phase I clinical studies using Medi-528 (an injectable humanized anti-IL-9 mAb) demonstrated improved clinical symptoms in patients without causing adverse side reactions. Further evaluation of the antibody using randomized multicenter studies is essential to determine its therapeutic efficacy [351] . IL-4 antagonists Th2 polarization and induction of IL-4 is central to airway inflammation and asthma pathogenesis [352] . Clinical studies testing IL-4 antagonists such as Nuvance[trademark] (Immunex, Amgen, CA, USA), an inhaled preparation of soluble human IL-4R, and pascolizumab (a humanized anti-IL-4 antibody) showed that while Nuvance significantly reduced CS dosage and FEV1 values, other treatment strategies failed to reduce asthma symptoms [353] . Similar results were obtained in trials with newly developed anti-IL-4 antagonists, such as IL-4 double mutein (R121D/Y124D, a human IL-4R antagonist) and human anti-IL4R mAb [354] . Toll-like receptor 9 agonists Synthetic Toll-like receptor (TLR)9 agonists, CpG oligodeoxynucleotides, have demonstrated substantial potential as vaccine adjuvants and as Th1 immunomodulators in the treatment of infectious and allergic diseases in animal models and early human clinical trials [355] . Murine models of chronic lung inflammation have demonstrated specific Th1 response induction combined with reduced bronchial inflammation and hyperresponsiveness when immunized with an allergen-CpG conjugate. Similar results were obtained in Phase I clinical trials using Tolamba[trademark] (Dynavax Technologies Corporation, CA, USA), a synthetic CpG motif covalently linked with a ragweed allergen Amb a 1. Patients administered with Tolamba experienced a significant reduction in allergenspecific IgE responses along with reduced nasal scores and increased Amb a 1-

specific IgG levels in the serum when compared with placebo control. However, subsequent large-scale clinical trials failed to demonstrate any clinical efficacy, although there was a considerable increase in IFN-γ mRNA expression [356] . Although the potential therapeutic contributions of TLR9 agonists are enormous, the safety and efficacy of these compounds in humans remain to be determined. Expert commentary Asthma is a serious and clinically heterogeneous disease condition; asthmatics do not uniformly present with the same clinical features or pathophysiological inflammatory profiles. For improved asthma management, a major goal is the identification of a specific and clinically sustainable classification of disease phenotypes. Asthma pathophysiology is complex, and research into the intricacies of such complexity has provided more questions than answers. Nevertheless, researchers in this field are determined to achieve the major longterm goal of identifying novel and effective therapeutic strategies based on a clear understanding of the multifarious elements involved in this disease condition. Owing to the heterogeneity that plagues asthma, therapies will probably address and be based on individual clinical asthma profiles. This implies acceleration and intensification of studies into asthma phenotypes using combinations of various approaches including genomic, transcriptomic, proteomic and metabolomic analyses. Whether the paradigm that complex diseases require complex therapies is true is yet to be tested. There is a plethora of research findings emerging all the time that further add to the complexity of asthma pathophysiology; this should not divert attention from determining convergence of the diverse elements involved towards development of novel and efficacious therapies. The increased interest and recent discoveries in asthma research, in both the clinical and basic science realms, will accelerate the processes leading to the achievement of the overall objective, namely, to better manage asthma and contribute to the reduction of its health and economic impact and burden worldwide. Five-year view A major setback in asthma therapy thus far has been the attempt to target either a single inflammatory cell type or a particular cell type, process or mediator product. This approach, we believe, would not yield significant success in asthma remission. Thus, one disease, one cell type and one molecule is doomed to fail as a viable strategy for the identification of clinical and therapeutic resolution in asthma. Reductionist approaches have, hitherto, been helpful in revealing the potential function of individual cell types and molecules. However, researchers in

this complex field can no longer afford to pursue similar approaches if the goal is to treat a disease characterized by complicated cellular and cytokine dyscrasia. The challenge over the next 5 years is to welcome and embrace the process of learning about and eventually targeting biological systems. Moreover, there is a great need to incorporate the utilization of advances in bioinformatics to make sense of the extraordinary amount of data that will emerge from multiple asthma phenotypes as well as tissue and organ microenvironments, and individualized patterns of response with current and future therapies. Finally, in terms of developing new therapeutic strategies, there may be merit in carefully identifying and targeting certain functional pathways that may be common to all cells involved in asthma. Our own bias leans towards intracellular and molecular pathways regulating mediator release in immune and inflammatory cellular infiltrates that may serve as a unifying strategy towards a more effective antiinflammatory therapy. Key issues * This article has reviewed a number of new pathways, cell types and mediators that have been the subject of intense study over the past decade. These elements may shed more light on the interactions between the immune and inflammatory components of asthma. * The classical two-sided (Th1 vs Th2) balance model is gradually being replaced by the new balancing square paradigm with the discovery of Th17 and Treg subsets to explain the pathogenesis of complex immune diseases such as asthma. * The Th17 and IL-17 family are a fascinating new addition to the asthma lexicon in terms of their significant correlation with disease severity and chronicity, especially in corticosteroid-resistant asthma. * Eosinophils are suggested to exert three main functions, namely, as effector, tissue modifier and immune regulator cells. The latter role is linked to two discoveries. The first is their constitutive expression of biologically active indoleamine 2,3 dioxygenase involved in tryptophan oxidative catabolism and the generation of kynurenines that suppress Th1, but not Th2, responses. The second is their indispensible role in the recruitment and differentiation of Th2 cells in murine models of asthma. Furthermore, the role of basophils is now thought to go beyond IgE-dependent anaphylaxis where they may have immune regulatory functions via their capacity to generate high concentrations of immune-active IL-4 and IL-6.

* Airway structural (mesenchymal) cells are now considered integral to the regulation of the immune and inflammatory reactions in asthma through the release of large concentrations of a range of cytokines and chemokines. * IL-33 (a member of the IL-1 family) has been shown to be important in mast cell, epithelial and endothelial cell, as well as airway smooth muscle cell biology and activity in asthma. IL-33 has broad effects within the context of inflammation, especially in Th2-type responses. * An increasing number of interesting new proteins in asthma include thymus stromal lymphopoietin that has been shown to be a potentially indispensible element of Th2 inflammation due to its capacity to induce allergic reactions on its own. In addition, chitinases were shown to play an exacerbating role in allergic-type inflammation in combination with Th2-type products, including IL4, IL-13 and thymus stromal lymphopoietin. * The last two decades have seen a number of advances in asthma therapy, especially leukotriene receptor antagonists. There are, however, a number of other therapies currently being tested clinically, including various anti-cytokines, chemokines and immunoglobulins that deserve attention. Many of these studies are hampered by asthma heterogeneity and a lack of accurate and reproducible disease phenotyping. CAPTION(S): Figure 1. Asthma pathophysiology of early- and late-phase responses. Upon presentation of allergens by APCs, naive T cells (ThP) undergo differentiation. Onset of asthma may be determined by the balance between proinflammatory T-cell subsets (Th17 and Th2) and Treg subsets. In patients with chronic asthma, proinflammatory T-cell subsets that generate TNF, TSLP and other cytokines are upregulated. This inflammatory response leads to infiltration by inflammatory cells to the airways. The epithelium and mesenchyme are also important sources of various inflammatory molecules, including TSLP, IL-33, IL-25 and ADAM33, as well as other cytokines/chemokines and mucins. Many of these events associated with late phase responses are involved in the induction of airway remodeling in asthmatics. There are, however, some regulatory factors; several Treg subsets regulate Th2 inflammation by producing cytokines that induce B cells to synthesize and elaborate IgA and IgG4, instead of IgE. Furthermore, IDOmediated tryptophan catabolism by eosinophils and DCs may be important in the generation of kynurenines, with the potential to regulate Th2 responses.

ADAM: A disintegrin and metalloprotease; APC: Antigen-presenting cell; CysLT: Cysteinyl leukotrienes; DC: Dendritic cell; ECP: Eosinophil cationic protein; EPO: Eosinophil peroxidase; FEV1 : Forced expiratory volume in 1 s; IDO: Indoleamine 2,3 dioxygenase; KYN: Kynurenines; NO: Nitric oxide; PAF: Platelet-activating factor; PG: Prostaglandin; TSLP: Thymic stromal lymphopoietin. Figure 2. Eosinophil functions. Three major functions have been proposed in relation to a wide range of mediators, cytokines and chemokines that are either synthesized de novo or stored in eosinophils (listed in boxes). The first is an effector function that may potentially contribute to tissue damage, cell activation and airway hyperresponsiveness. The second is tissue remodeling involving effects leading to the deposition of collagen, extracellular matrix and mucus hypersecretion. The third is immune regulation, potentially through the action of an array of cytokines and chemokines generated by eosinophils as well as tryptophan catabolism via IDO and the subsequent presence of kynurenines that target Th1 cells for net apoptosis and Th2 for net proliferation. ECP: Eosinophil cationic protein; EPO: Eosinophil peroxidase; GM-CSF: Granulocyte-macrophage colony-stimulating factor; IDO: Indoleamine 2,3 dioxygenase; LTC4: Leukotriene C4; MBP: Major basic protein; MMP: Matrix metalloproteinase; ROS: Reactive oxygen species.