Clinical & Experimental Allergy

Pathogenesis of Asthma




While asthma is considered an inflammatory disorder of the conducting airways, it is becoming increasingly apparent that the disease is heterogeneous with respect to immunopathology, clinical phenotypes, response to therapies, and natural history. Once considered purely an allergic disorder dominated by Th2-type lymphocytes, IgE, mast cells, eosinophils, macrophages, and cytokines, the disease also involves local epithelial, mesenchymal, vascular and neurologic events that are involved in directing the Th2 phenotype to the lung and through aberrant injury-repair mechanisms to remodeling of the airway wall. Structural cells provide the necessary “soil” upon which the “seeds” of the inflammatory response are able to take root and maintain a chronic phenotype and upon which are superimposed acute and subacute episodes usually driven by environmental factors such as exposure to allergens, microorganisms, pollutants or caused by inadequate antiinflammatory treatment. Greater consideration of additional immunologic and inflammatory pathways are revealing new ways of intervening in the prevention and treatment of the disease. Thus increased focus on environmental factors beyond allergic exposure (such as virus infection, air pollution, and diet) are identifying targets in structural as well as immune and inflammatory cells at which to direct new interventions.


Asthma is a disorder of the conducting airways, which contract too much and too easily spontaneously and in response to a wide range of exogenous and endogenous stimuli. This airway hyperresponsiveness is accompanied by enhanced sensory irritability of the airways and increased mucus secretion. The different clinical expressions of asthma involve varying environmental factors that interact with the airways to cause acute and chronic inflammation, and the varying contributions of smooth muscle contraction, edema and remodeling of the formed elements of the airways. Heterogeneity of asthma also relates to the different response to therapies. Asthma is considered a good example of gene–environment interactions, although no single gene or environmental factor accounts for the disease. This chapter focuses on the pathophysiologic events underlying the inflammatory and remodeling response.


Airway inflammation in asthma is a multicellular process involving mainly eosinophils, neutrophils, CD4+T lymphocytes and mast cells, with eosinophilic infiltration being the most striking feature (Kay 2005) (Fig. 78.1). The inflammatory process is largely restricted to the conducting airways but as the disease becomes more severe and chronic the inflammatory infiltrate spreads both proximately and distally to include the small airways and in some cases adjacent alveoli (Kraft et al. 1999). The inflammatory response in the small airways appears to be predominantly outside the airway smooth muscle (Fig. 78.1), whereas in the large airways inflammation of the submucosa dominates (Haley et al. 1998). This Th2 type of inflammation is common to chronic allergic inflammatory responses at multiple tissue sites and indeed is seen at these sites in patients with asthma who frequently express comorbidities such as chronic rhinitis, sinusitis, atopic dermatitis, and food allergy (Kay 2001). There continues to be debate about the specific contribution of individual cell types in these other diseases, but at least in asthma with an allergic component a common theme is now emerging. Other chapters have provided considerable detail of the potential individual roles of these cells and their mediators in the allergic tissue response, so only brief mention is made of them here with respect to their relevance to human asthma.

Figure 781.

 Cross-section of a small asthmatic airway showing abundant inflammation both inside and outside the smooth muscle and remodeling as evidenced by the deposition of new matrix beneath the epithelium and in all layers of the airway. (Holgate & Polosa 2006).


A fundamental feature of asthma associated with allergic sensitization is the ability of the airway to recognize common environmental allergens and to generate a Th2 cytokine response to them. Recognizing that in excess of 40% of the Western population is atopic (i.e., have elevated IgE to common environmental allergens), only about 7% express their atopy in the form of asthma (Beasley et al. 2001). Therefore, a crucial question to ask is what mechanisms account for the specific expression of atopy in the conducting airways and why some patients despite being highly atopic have no evidence of asthma? One explanation is the way that the immune response to allergens is regulated at the surface of the airways.

A fundamental feature of allergen sensitization is the uptake and processing of inhaled allergens by dendritic cells situated in the airway epithelium and submucosa and which extend their processes to the airway surface (von Garnier et al. 2005; Hammad & Lambrecht 2006). Uptake of allergen is enhanced by IgE bound to high-affinity receptors on dendritic cells that facilitate allergen internalization (Kitamura et al. 2007). Once inside the dendritic cell, processing of allergens by cathepsin S and the subsequent selection of peptides loaded onto and presented by HLA molecules (MHC class II) is fundamental to the ability of these cells to serve as antigen-presenting cells to T lymphocytes (Riese & Chapman 2000). Once the dendritic cell has engaged allergen, it receives signals to migrate to local lymphoid collections where antigen presentation takes place. Its specific chemokine receptors, including CCR7 and its ligands CCL19 and CCL21 (and to a lesser extent CXCR4 and its ligand CXCL12), are involved in this chemotactic migration to enable contact with naive T cells (Humrich et al. 2006; Pease & Williams 2006) (see Chapter 8). Presentation of a selected antigen peptide to the T-cell receptor initiates sensitization and the subsequent immune response to the specific allergen (Smit & Lukacs 2006). The nature of this immune response depends on whether engagement of selective costimulatory molecules occurs in parallel. For efficient antigen-dependent T-cell activation engagement of either CD80 (B7.1) or CD86 (B7.2) on the dendritic cells with CD28 on T cells leads to sensitization, whereas lack of, or inefficient, engagement of these costimulatory molecules may lead to anergy (Larche et al. 1998; van Rijt et al. 2004). An alternative method of preventing sensitization and rendering T cells anergic is engagement of a second costimulatory molecule, cytotoxic T-lymphocyte antigen (CTLA)4, which has a higher affinity than either CD80 or CD86 for CD28 and can therefore prevent CD80/CD86 costimulation (Jaffar et al. 1999a, b). This is the basis of the successful clinical application of the CTLA4–immunoglobulin fusion protein abatacept, used as an immunomodulatory agent in such diseases as rheumatoid arthritis and in an animal model of allergen-induced airway inflammation (Weinblatt et al. 2006).

In more severe asthma the relative importance of CD28 signaling in supporting the inflammatory response is reduced (Lordan et al. 2001). Under these circumstances other costimulatory pathways are thought to be engaged in T-cell activation, including ICOS and its multiple possible ligands B7-RP1, PD-L1 (B7-H1), PD-L2 (B7-DC), B7-H3 and B7-H4 (B7-S1) (Greenwald et al. 2005) as well as OX40 (CD134) and its ligand OX40L (CD252) (Salek-Ardakani et al. 2003; Burgess et al. 2005).

The capacity of dendritic cells to generate interleukin (IL)-12 determines the balance between Th1 and Th2 responses, IL-12 polarizing T-cell differentation in favor of a Th1 response (Kuipers et al. 2004). However, while IL-12 is able to counteract Th2 sensitization, it is also able to contribute to maximal expression of allergic airway disease post sensitization (Meyts et al. 2006). Once sensitized, T cells not only migrate back to the airways to the site of antigen presentation under the influence of the chemokines CCL11, CCL24, CCL26, CCL7, CCL13, CCL17 and CCL22 (which interact with their reciprocal receptors including CCR3, CCR4, CCR5, CCR6, CCR7 and CCR8) (Garcia et al. 2005; Kallinich et al. 2005), but these cells also become potent producers of a range of cytokines, the majority of which are expressed on the long arm of chromosome 5, namely IL-3, IL-4, IL-5, IL-6, IL-9, IL-13 and granulocyte–macrophage colony-stimulating factor (GM-CSF) (Kay 2006; Ryu et al. 2006). IL-1β produced by macrophages, monocytes, dendritic cells, and smooth muscle and epithelial cells in large amounts (Schmitz et al. 2003; Dragon et al. 2006) and IL-2 produced by T cells further enhance antigen-induced T-cell proliferation and maturation (Anderson 2002).

There is now persuasive evidence that at least in mild to moderate asthma Th2-type cells dominate the T-cell repertoire in the airways (Anderson 2002). Through cytokine production, they have the capacity to recruit secondary effector cells such as macrophages, basophils and eosinophils into the inflammatory zone where these cells become primed and subsequently activated for mediator secretion (Fig. 78.2). There has also been recent interest in the potential role of IL-4- and IL-13-secreting, CCR9+ natural killer (NK) T cells in orchestrating the inflammatory response in chronic asthma (Sen et al. 2005; Akbari et al. 2006; Umetsu & Dekruyff 2006) (see Chapter 3) (Pham-Thi et al. 2006; Thomas et al. 2006), although initial findings of their primacy has been challenged (Ho 2007; Vijayanand et al. 2007). Overall, it is the Th2-type T cell bearing the CCR4 chemokine receptor that is the cell which dominates the allergic immune response and may be the cell most probably responsible for contributing to the ongoing chronic inflammatory response. Indeed, asthma severity is reported to be associated with an increase in CCR4+ T cells (Ishida et al. 2006). Thus, inhibitors of CCR4 (e.g., the antibody-dependent cell cytotoxic monoclonal antibody KM2760) could be highly effective in the treatment of asthma by inactivating or removing CCR4+ Th2 cells (Ishida et al. 2006).

Figure 782.

 Schematic representation of the inflammatory cascade in allergic asthma. (Bradding et al. 2006).

While Th2-type T cells may be important in the pathogenesis of mild to moderate asthma, as the disease becomes more severe and chronic Th1-type T cells are recruited that have the capacity to secrete tumor necrosis factor (TNF)-α and interferon (IFN)-γ (Truyen et al. 2006). This more complicated T-cell profile may help to explain the aggressive and tissue-damaging aspects of the immune response in more severe disease. Although these Th1-type cells, as well as CD8+ T cells, have been incriminated in both more severe asthma (Hamzaoui et al. 2005) and during asthma exacerbations (especially following virus infection) (O'Sullivan 2005), the precise mechanisms whereby they achieve this is still largely unknown. Although the T lymphocyte has been given primacy with regard to the orchestration of the inflammatory response in asthma, there have been no studies using selective T-cell-depleting strategies with either monoclonal antibodies or drugs in mild to moderate asthma (Kon & Kay 1999; Bharadwaj & Agrawal 2004). However, in a small proof of concept study, a single infusion of anti-CD3 improved lung function in patients with severe disease (Kon et al. 1998). The immunosuppressant approach to asthma treatment has not been pursued largely due to difficulties in developing calcineurin inhibitors such as cyclosporin A and tacrolimus in an inhaled form and problems with side effects.


The mast cell has long been associated with asthma. The early asthmatic reaction following inhaled allergen provocation is mast-cell dependent and drugs such as sodium cromoglycate and nedocromil sodium are believed to mediate their effects by inhibiting mast cell mediator secretion (Holgate 1996). For many years it was thought that the mast cells present in the airway epithelium and submucosa were fundamental to the contribution that these cells make to asthma (Shahana et al. 2005), but recent studies have indicated that mast cells deeper in the airway wall are also important. While undoubtedly mucosal-type mast cells (tryptase positive, chymase negative) under the control of T lymphocytes (specifically IL-3, IL-4 and IL-9) are highly responsive to inhaled allergens (and possibly other stimuli such as hypertonicity) in causing bronchoconstriction, there has been recent interest in mast cells present deeper in the airway wall and in the more peripheral airways as being more fundamental to some of the chronic inflammatory responses in asthma (Bradding et al. 2006). Of particular interest has been the discovery that in chronic asthma (but apparently not in eosinophilic bronchitis), mast cells are markedly increased in association with airway smooth muscle in both the large and the small airways (Brightling et al. 2002). It is at this site that mast cells interact with airway smooth muscle through the action of autacoid mediators such as leukotriene (LT)D4, prostaglandin (PG)D2 and histamine, but also contributing to fibrogenesis and an increase in smooth muscle as part of the “remodeling” response (Kaur et al. 2006; Plante et al. 2006). Although many factors are involved in the regulation of mast cell function in the submucosa and deeper in the airway wall, mast cells in the airway smooth muscle differ from mucosal mast cells in the being of the connective tissue-type (tryptase positive, chymase positive, and carboxypeptidase positive) and in being more dependent on stem cell factor (SCF) for their survival. SCF (c-kit ligand) is produced by the epithelium, smooth muscle and fibroblasts and is upregulated in asthma (Al Muhsen et al. 2004; Plante et al. 2006). In addition, CXCL8 and CXCL10 produced by airway smooth muscle itself are not only important in the recruitment of mast cells into this compartment by interacting with their receptors CXCR3 and CXCR2 respectively, but also in their priming for enhanced mediator secretion (Woodman et al. 2006; Scott & Bradding 2005). In the reverse direction, mast cells secrete CCL19 that, through its receptor CCR7, stimulates airway smooth muscle migration and probably contributes to the smooth muscle hyperplasia (Kaur et al. 2006). Thus, airway smooth muscle is partly dependent on mast cells for its survival and enhanced contractility, whereas mast cells are dependent on smooth muscle factors for their survival and activation.

On activation, irrespective of their subtype, mast cells release preformed granule-associated mediators such as histamine, tryptase and other proteases, heparin and some cytokines, as well as newly formed eicosanoids that include PGD2, thromboxane (TX)A2, and the cysteinyl leukotreines (LTC4 and LTD4) (Bradding et al. 2006). These mediators are potent smooth muscle contractile agents and also increase microvascular permeability. Both PGD2 and LTD4 interact with cell-surface receptors on eosinophils, macrophages, basophils and mast cells where they serve as chemoattractant as well as priming agents (Ogawa & Calhoun 2006). Thus, cysteinyl leukotriene antagonists such as montelukast and pranlukast are not only able to block the acute effects of leukotrienes on the formed elements of the airway, but also exert some antiinflammatory action.

The mast cell is also well endowed with neutral proteases (Reed & Kita 2004). One that has attracted particular interest is tryptase, a tetrameric neutral protease that has a preferential action on G protein-coupled protease-activated receptor type 2 (PAR2), which is present on many cell types including epithelial cells, fibroblasts, smooth muscle cells, endothelial cells, and on a number of inflammatory cells (Cairns 2005; Hallgren & Pejler 2006). The exact role of the PAR2 receptor in asthma has yet to be clearly defined, but there is increasing evidence that its activation is involved in mesenchymal cell proliferation and airway wall remodeling (Berger et al. 2001). Mast cells and eosinophils are also an important source of the zinc-dependent matrix metalloproteinases (MMP)-3 and MMP-9, which through their interaction with matrix proteins and proteogylcans have also been incriminated in airway wall remodeling (Dahlen et al. 1999; Wenzel et al. 2003).

Activation of mast cells, particularly through the highaffinity IgE receptor (FCɛRI), leads to the release of certain cytokines that are packaged within mast cell granules, including TNF-α, IL-4 and IL-5 (Bradding et al. 1995; Wilson et al. 2000), but also induces transcription of these and other cytokines and chemokines that are then secreted over a period of up to 72 hours following cell perturbation (Okayama et al. 1995, 2003). These cytokines and chemokines undoubtedly contribute to the ongoing inflammatory response in asthma and may be partly responsible for the allergen-induced late-phase inflammatory response characteristic of allergen provocation. Blockade of IgE using the monoclonal antibody omalizumab leads to marked attenuation of both the early and late phase allergen-induced bronchoconstrictor and skin inflammatory responses, linking both to mast cell activation (Fahy et al. 1997; Ong et al. 2005). However, the precise mechanisms through which late-phase bronchoconstriction and the associated inflammatory cell influx with neutrophils and eosinophils lead to bronchoconstriction is not fully established, although its partial attenuation with antileukotriene drugs suggests at least some role for cysteinyl leukotrienes (Leigh et al. 2002).


A very prominent cell in the inflammation of allergic asthma is the eosinophil leukocyte, which is present not only in the airway wall but in uncontrolled asthma is also found in large numbers in the sputum and bronchoalveolar lavage fluid (Kay 2005; Lemiere et al. 2006). These cells are in large part initially recruited from the bone marrow as CD34 precursors, following the release of PGD2, cysteinyl leukotrienes, cytokines and chemokines from the asthmatic airway. The developing eosinophils then pass from the circulation via the microvascular compartment into the airway wall. There has been much research on the mechanisms involved in this recruitment, including the role of specific chemokines and adhesion molecules (reviewed in detail elsewhere in this book). Suffice to say that IL-3 and GM-CSF and eotaxins 1–3 are crucial to the early derivation of eosinophils from CD34+ bone marrow precursor cells, with IL-5 being responsible for their maturation and recruitment into the airways (Robinson et al. 1999; Sehmi et al. 2003) (Fig. 78.3). Eosinophils are a rich source of granule basic proteins, such as major basic protein, eosinophil peroxidase, and eosinophil cationic protein, and also have the capacity to generate eicosanoids such as prostacyclin (PGI2) and cysteinyl leukotrienes and release potentially tissue-damaging superoxide and a range of cytokines and chemokines (Kariyawasam & Robinson 2006). The dramatic reduction in sputum and tissue eosinophils that occurs on treatment of asthma with inhaled or oral corticosteroids associated with clinical improvement has led to the idea that eosinophils are fundamental to airway dysfunction in asthma and are the principal target for this drug class (Djukanovic et al. 1992, 1997).

Figure 783.

 Th2 cytokines involved in the maturation and recruitment of eosinophils from CD34+ precursors in the bone marrow and airway wall. Note the prominent contribution of IL-5. See text for definition of abbreviations.

Undoubtedly, eosinophils play an important role in the allergic inflammatory response, but recently their primacy in the inflammatory milieux of asthma has been challenged (Flood-Page et al. 2003a). Following allergen challenge in moderately severe asthma, the administration of a humanized blocking IgG monoclonal antibody directed to IL-5 resulted in a dramatic reduction (>80%) in circulating and sputum eosinophils. However, this was not accompanied by any inhibitory effect on the allergen-induced late-phase reaction in the airways or skin or acquired airway hyperreactivity (Leckie et al. 2000; Phipps et al. 2004). In a large clinical study three injections of anti-IL-5 given 2 weeks apart also had a dramatic effect on circulating and sputum eosinophils but paradoxically did not affect any of the clinical outcome measures of asthma including baseline lung function (Fig. 78.4). These studies questioned either the primacy given to the eosinophil as the dominant inflammatory cell of asthma or the effectiveness of anti-IL-5 to deplete eosinophils. Thus, in a series of subsequent bronchial biopsy studies, anti-IL-5 antibody therapy only reduced airway eosinophils by about 50% and it has been suggested that the remaining 50% of eosinophils could account for ongoing asthma symptomology (Flood-Page et al. 2003a) (Fig. 78.4). Anti-IL-5 monoclonal antibody also induced bone marrow eosinophil maturation arrest and decreased CD34+ eosinophil progenitors in the bronchial mucosa of atopic asthmatics (Flood-Page et al. 2003b). Further studies have demonstrated that eosinophil depletion using this approach was able to modify certain matrix proteins in the subepithelial basement membrane such as tenascin C, lumican and procollagen 3 (Flood-Page et al. 2003b) (Fig. 78.4). Based on these findings, it has been suggested that the eosinophil may be playing a more important role in airway remodeling than has hitherto been recognized (Kay et al. 2004). Persistence of some eosinophils in asthmatic airway tissue in the presence of IL-5 blockade might be explained by some loss of the IL-5 receptor from such cells when eosinophils are recruited into the airways (Liu et al. 2002). A positive contribution of eosinophils to tissue remodeling is supported by their capacity to generate transforming growth factor (TGF)-β1 and support fibroblast proliferation, collagen synthesis and myofibroblast maturation (Williams & Jose 2000).

Figure 784.

 Effect of intravenous injections of anti-IL-5 monoclonal antibody (mepolizumab) on (a) circulating eosinophils in over 100 moderately severe asthmatic patients; (b) comparative effect on eosinophils at different sites; (c) immunoreactivity to tenascin C; (d) tenascin C quantified by image analysis. (Reproduced with permission from (b & c) Food-Page et al. 2003b (a & d) Food-Page et al. 2007).

Attraction of eosinophils to the site of inflammation is largely mediated through eotaxins 1, 2 and 3, macrophage chemotactic factor (MCP)-3, MCP-4, and RANTES (Williams & Jose 2000). These chemokines interact with the CCR3 and, to a lesser degree, CCR5 receptors on eosinophils and, as with other chemokines, not only provoke their directed migration but also their priming. Therefore, it is somewhat disappointing that despite showing activity in mouse models (Das et al. 2006), CCR3 receptor antagonists have so far proven disappointing when tested in the clinic.

There is also considerable debate about how eosinophils are cleared from the airways in asthma. At one time it was thought that programmed cell death, mediated by withdrawal of specific cytokines such as IL-5 and GM-CSF, resulted in accelerated programmed cell death (Meagher et al. 1996) and uptake by macrophages and epithelial cells (Sexton et al. 2004). Similarly, it was believed that an important component of corticosteroid action was through acceleration of this apoptotic response (Meagher et al. 1996). However, detailed examination of asthmatic tissue has demonstrated little evidence of eosinophil apoptosis; rather eosinophils display an unusual form of cytolysis that leaves their eosinophil granules intact within the airway wall even though the cell membrane and cytoplasm have disappeared (Uller et al. 2004). Corticosteroids have also been shown to increase clearance of airway eosinophils through egression into the airway lumen as well as by inhibiting chemokine production (Uller et al. 2006). In an animal model, activation of apoptosis through the Fas receptor has been shown to magnify rather than resolve asthmatic type inflammation (Uller et al. 2005). Thus, while much has been learnt about eosinophils in asthma, there still remain many unanswered questions. The role of neutrophils is discussed later in this chapter.


Monocytes are able to differentiate into macrophages and dendritic cells in the presence of GM-CSF (Gajewska et al. 2003), the latter requiring IL-4 (Webb et al. 2007). In chronic asthma both monocytes and macrophages are prominent cells in the airway mucosa and undoubtedly play an important role in disease pathogenesis. While these cells are an important source of cysteinyl leukotrienes, reactive oxygen and a variety of lysosomal enzymes, their precise role in mediating tissue damage and contributing to the overall airway pathology of asthma is largely unknown. In corticosteroid-refractory asthma, monocytes and macrophages are thought to play an increasingly important role and may well account for the ongoing chronic inflammation associated with their preferential infiltration into the airway wall in patients with longstanding corticosteroid-resistant disease (Sher et al. 1994; Loke et al. 2006). It is clear that more attention focused on these cells is required, especially how they fit into the overall Th2 paradigm and whether they represent a separate component of the immune and inflammatory response which has not yet been adequately explored.


Although basophils have largely been thought of as circulating IgE-triggered inflammatory cells, in certain types of immune response they do accumulate in tissues. The discovery of unique basophil-specific markers such as basogranulin (Mochizuki et al. 2003; Agis et al. 2006) has enabled their identification in the airways of subjects with asthma (Macfarlane et al. 2000; Kepley et al. 2001). However, at this time, it is not clear what their precise role is in either acute or chronic disease, although it is known that they share many of their recruitment mechanisms with eosinophils and are likely to be accompaniments of eosinophil infiltration (Gangur et al. 2003).


Although the majority of asthma is associated with atopy and has its onset in early childhood across all ages and especially after the age of 40 years, there are forms of asthma that appear to be independent of atopy (Humbert 2000). These nonallergic forms of the disease have been subject to careful comparative investigation but so far no clear pathogenic pathways have been identified (Humbert et al. 1999; Corrigan 2004). There is some evidence to suggest that local IgE mechanisms may be involved with the detection of IgE isotype switching in airway biopsies, but the clinical significance of this has yet to be determined (Ying et al. 2001; Jayaratnam et al. 2005). The immunopathology of late-onset nonallergic asthma appears to be very similar to that of allergic asthma, although there have been some differences reported in the relative proportion of the various inflammatory cells present (Ying et al. 1997; Ying et al. 1999; Strek 2006). It is important to recognize that some forms of late-onset asthma have an occupational cause due to non-IgE-dependent sensitization to chemicals in the workplace (Malo 2005; Slavin 2005). In addition to immunologic sensitization to chemicals, both intrinsic asthma (Nahm et al. 2002) and asthma caused by diisocyanate exposure are associated with antibodies directed to epithelial components (Ye et al. 2006a). Other autoantibodies in asthma include those directed to heat-shock protein (HSP)-70 (Yang et al. 2005), CD28 (Neuber et al. 2006), and α-enolase (Nahm et al. 2006). Whether such autoantibodies are truly pathogenetic or related more to tissue damage and inflammation remains to be established (Ye et al. 2006b). Whatever the mechanisms, this type of asthma is likely to be highly heterogeneous, exhibiting overlap with chronic obstructive pulmonary disease. It is clearly an area where much more careful phenotyping is required as well as the application of novel cell and molecular methodology to determine underlying mechanisms.


The airway epithelium, while playing an important role as a physical barrier, is now known to be fundamental to asthma pathogenesis. Bronchial biopsies in anything but the mildest forms of asthma show areas of epithelial metaplasia and damage, thickening of the subepithelial basal lamina, increased number of myofibroblasts, and other evidence of airway remodeling such as hypertrophy and hyperplasia of airway smooth muscle, mucous gland hyperplasia, angiogenesis and an altered deposition and composition of extracellular matrix proteins and proteoglycans (Knight & Holgate 2003; Holgate et al. 2004). While these pathologic features have been commonly reported in asthma deaths and in bronchial biopsies from patients with asthma of varying severity, more recently similar findings have been found in the airways of children in relation to the onset of asthma (Fedorov et al. 2005; Saglani et al. 2005; Barbato et al. 2006). Evidence of epithelial damage with upregulation of epidermal growth factor receptors (EGFRs) and features of impaired proliferation, e.g., reduced expression of proliferative markers such as Ki67 and proliferating cell nuclear antigen (PCNA) and upregulation of the cell cyclin inhibitor, nuclear p21wat, suggest that, as in adult asthma (Puddicombe et al. 2003), the epithelium is chronically injured and unable to repair properly (Bucchieri et al. 2002; Kicic et al. 2006) (Fig. 78.5). One important feature of the asthmatic epithelium is its capacity to defend itself against oxidant injury (Bucchieri et al. 2002; Comhair et al. 2005), a feature that may partly explain why asthmatic subjects are so sensitive to oxidant pollutants such as ozone, environmental tobacco smoke and ambient air particulates (Truong-Tran et al. 2003).

Figure 785.

 Schematic diagram of epithelial injury and activation of subepithelial mesenchymal cells through the release of growth factors creating a chronic wound scenario. See text for definition of abbreviations.

Under normal circumstances the epithelium forms a highly regulated and almost impermeable barrier through the formation of tight junctions (Godfrey 1997). These protein complexes at the apex of the columnar cells (Mullin et al. 2005) comprise a series of proteins that includes claudins and transmembrane adhesion proteins that connect adjacent cells. Structural integrity is also maintained through cell-cell and cell–extracellular matrix interactions that includes involvement of E-cadherin, desmosomes and hemidesmosomes (Roche et al. 1993). Although at one time thought to be an artifact of sample handling (Ordonez et al. 2000), it is now clear in asthma that the epithelium is more fragile with easy loss of the columnar cells due to disruption of both tight junctions and desmosomal attachments (Montefort et al. 1992; Barbato et al. 2006; Shahana et al. 2006) (Fig. 78.6). This abnormality is also apparent in the epithelium of nasal polyps from asthmatics compared with those from nonasthmatic subjects (Shahana et al. 2005). Using differentiated epithelial cells in culture brushed from normal and asthmatic airways, it has been shown that the permeability (leakiness) of the asthmatic epithelium is greatly increased, leading to greater access of inhaled allergens, pollutants and other irritants to basal cells and the underlying airway tissue. This increase in epithelial permeability has also been observed in vivo using inhaled radiolabeled impermanent probes, such as technetium labeled DTPA (Ilowite et al. 1989). A reduction in the ability of the airway epithelium to exclude inhaled environmental agents may partly explain why certain atopic individuals go on to develop asthma whereas those with good barrier function do not. This loss of barrier function may reflect a broader abnormality in affecting other organs such as the skin, conjunctiva and gut that are foci for other atopic disorders (Hijazi et al. 2004; Liu et al. 2005; Hughes et al. 2006; Proksch et al. 2006).

Figure 786.

 Role of the epithelium in chronic asthma. (a) Extensive epithelial damage and mucus plugging of an airway from a patient who has died from asthma. (b) Progressive epithelial damage and disposition of matrix in asthma relative to disease severity. (c) Infiltration of the asthmatic epithelium with eosinophils associated with damage. (d) Molecular components of tight junctions that control intracellular permeability.

A recent development has been the discovery of a number of genes encoded on chromosome 1q, including filaggrin and the S100 proteins, that are involved in maintaining epithelial integrity in both the skin and the airways (Marshall et al. 2001) (Fig. 78.7). The association of genetic polymorphism of the pro-filaggrin gene on chromosome 1q13 with atopic dermatitis and asthma (Palmer et al. 2006; Ying et al. 2006) is of considerable interest and refocuses the possible origins of asthma more toward the epithelium and formed elements of the airways, rather than the immune response alone (Hudson 2006).

Figure 787.

 Molecules involved in regulating epithelial integrity. (a) Gene cluster on chromosome 1q encoding a series of genes involved in maintaining epithelial integrity, including filaggrin (FLG), involucrin (INV), loricin (LOR), and the S100 proteins. (b) Genomic architecture of the filaggrin gene. The arrows denote single nucleotide polymorphisms in exon 3 associated with loss of function. (c) Six family pedigrees with ichthyosis vulgaris (homozygotes shown as symbols and heterozygotes as symbols); note the strong association with atopic dermatitis (AD). (d) Immunohistochemical localization of filaggrin to superficial layer of the skin and its absence in a patient homozygous for the R501X genotype of ichthyosis vulgaris. (Reproduced from Palmer et al. 2006, with permission).

Another example of an environmental injury targeting the airway epithelium in asthma is the effect of respiratory virus infections. It has long been known that common cold viruses such as rhinovirus are associated with exacerbations of asthma in both children and adults (Johnston et al. 1995; Corne et al. 2002). Bronchial biopsy studies have shown that the airway epithelium is the preferential site for these viruses to enter the airway tissue (Papadopoulos et al. 2000). Infection of asthmatic epithelial cells compared with normal epithelial cells in vitro has revealed that the former lack the ability to generate IFN-β and IFN-λ, cytokines essential for eliminating viruses partly through induction of apoptosis (Wark et al. 2005; Contoli et al. 2006). In asthma the viruses continue to replicate until they kill the epithelial cells cytotoxically, leading to massive virus shedding and infection of adjacent cells, as well as the release of mediators from the damaged cells. By adding IFN-β back into the cell cultures, resistance to rhinovirus infection is restored (Wark et al. 2005), suggesting that this may be a new approach to the prevention and treatment of acute asthma exacerbations (Holgate 2005).

A second type of injury that targets the epithelium is air pollution, asthma worsening at times of air pollution episodes. A number of studies have shown that the antioxidant defenses mounted by the airway epithelium in asthma are markedly reduced and are associated with reductions in superoxide dismutase (Comhair et al. 2005) and glutathione peroxidase (Qujeq et al. 2003; Misso & Thompson 2005). In not being able to defend itself adequately against oxidant damage, the airway epithelium is damaged more easily (Rahman et al. 2006). The association of asthma with ozone and particle pollution episodes can be explained on this basis (Morrison et al. 2006).

A third example is the inflammatory and immune cascade initiated by activation of Toll-like receptors (TLRs). The recent discovery that a novel cytokine, thymic stromal lymphopoietin (TSLP), is generated by airway and skin epithelial cells following activation of TLR3, TLR6, TLR7, TLR8 and TLR9 and the capacity of this cytokine to generate Th2-polarized and mast cell activation responses places this cytokine in a unique position to orchestrate the airway inflammatory response of asthma (Ying et al. 2005; Liu 2006).

When damaged, the airway epithelium needs to repair but, as referred to above, the repair process is compromised in asthma. The airway epithelium enters into a chronic “wound scenario” with consequent production of a variety of cytokines and growth factors in an attempt to repair the “wound” (Fig. 78.8). One group of growth factors central to epithelial repair and its altered phenotype in asthma is epidermal growth factor (EGF) and related molecules (HB-EGF, amphiregulin) (Puddicombe et al. 2000) that, by interacting through their tyrosine kinase receptors, promote repair. EGFR stimulation of the damaged epithelium also generates a mucus-secretory phenotype (Polosa et al. 2002; Hamilton et al. 2005; Casalino-Matsuda et al. 2006) and an altered inflammatory response involving neutrophils (Hamilton et al. 2003), which are all characteristic of more chronic and severe asthma. Other epithelialderived factors include chemokines attracting neutrophils and other inflammatory cells, as well as a range of growth factors including platelet-derived growth factor (PDGF), fibroblast growth factor (FGF)-1, FGF-2 and TGF-β that are active on fibroblasts and smooth muscle (Fig. 78.5). The secretion of latent TGF-β and its activation by proteolytic enzymes are fundamental to the subsequent airway remodeling that accompanies all but the most mild asthma (Boxall et al. 2006; Howell & McAnulty 2006). TGF-β itself is able to further impair epithelial repair responses while at the same time promoting the differentiation of fibroblasts into myofibroblasts (Leung et al. 2006). Bronchial biopsies from asthma have revealed that myofibroblasts are present in increased numbers in the subepithelial and submucosal region of asthmatic patients and increase in proportion to disease chronicity and severity (Choe et al. 2006; Wicks et al. 2006). The precise role of these cells in airway wall remodeling has yet to be proven, but it is highly likely that they do play a key role, both with respect to making good the damage caused to the epithelium through the deposition of matrix in the lamina reticulosa of the basement membrane and also as a result of interactions with recruited inflammatory cells such as eosinophils and mast cells, thereby helping sustain the chronic inflammatory response (Fig. 78.9).

Figure 788.

  Evidence for epithelial injury and impaired repair in childhood asthma. Bronchial biopsies in asthma show increased thickening of the lamina reticularis (a), increased expression of the epidermal growth factor receptor (b), reduced biomarkers of cell proliferation and increased nuclear translocation of the cell cycle inhibitor p21 (c), reproduced with permission of Fedorov et al. Thorax 2005.

Figure 789.

 Schematic representation of the interactions between airway inflammation and remodeling in clinical asthma involving the epithelial–mesenchymal trophic unit. (Reproduced with permission of Holgate et al. 2004).

The conducting airways contain an epithelium that is stratified, the upper layer containing a mixture of ciliated, goblet and Clara cells. In chronic asthma the number of goblet cells that secrete viscous mucus increases, with a parallel reduction in ciliated cells. Since mucus production is fundamental to the pathogenesis of chronic asthma, this metaplastic change in the airway epithelium is of great importance, particularly since this has been shown to occur in the more peripheral airways which are normally devoid of goblet cells (Ordonez et al. 2001; Perez-Vilar 2006). The factors responsible for goblet cell metaplasia include the Th2 cytokines IL-4, IL-9 and IL-13 as well as TNF-α, which are all capable of directing differentiation to a mucus-secreting phenotype and interact by inducing EGFR-mediated goblet cell metaplasia through receptor transduction and induction of EGFR ligands such as HB-EGF and amphiregulin (Cohn 2006; Temann et al. 2007). In asthma it is mucin 5AC that dominates the mucin glycoproteins that are secreted and it is this mucin which is largely responsible for the unusual viscoelastic properties observed in asthmatic sputum that lead to difficulties in its expectoration (Morcillo & Cortijo 2006; Voynow et al. 2006). Recently IL-13 has been shown to regulate a chloride channel (GOB5; CLCA3), which is intimately involved in the regulation and secretion of mucin from goblet cells (Thai et al. 2005; Long et al. 2006). However, the identification of multiple forms of GOB5 and its precise role in asthma has yet to be fully understood, though it does represent a therapeutic target.

One interesting idea that may partly account for the onset of asthma in genetically susceptible children is the fact that the airway epithelium is fundamentally abnormal both in its response to environmental injury and its repair. A recent study (COAST) in children born of asthmatic and atopic parents has shown that those who develop more persistent wheezing at the age of 4 years are those who had frequent symptomatic virus infections in early infancy, particularly rhinovirus (Lemanske et al. 2005). Impaired innate immunity, possible due to a genetic defect in interferon production in response to virus infection, may initiate the chronic injury–repair cycle associated with the onset of chronic disease. Longitudinal cohort studies have also demonstrated that while atopy failed to predict persistent wheezing up to the age of 5 years, beyond this age those who persistently wheeze are the atopic children whereas those who lose their wheezing (“grow out of ” asthma) are the nonatopics (Illi et al. 2006) (Fig. 78.10). Together, these observations point to environmental factors operating independently at different stages during the evolution of asthma and again places the airway epithelium at the boundary where environmental interactions take place to generate the asthma phenotype.

Figure 7810.

 Influence of atopy on the persistence of asthma after 5 years of age. (Reproduced from Illi et al. 2006).


Although difficult to define functionally, from a structural standpoint there is ample evidence that, beyond airway inflammation, changes to the formed elements of the airway contribute significantly to the pathophysiology of asthma (James & Carroll 2000) (Figs 78.1 and 78.9). The most obvious change is in the airway smooth muscle, which not only increases in amount due to hypertrophy and hyperplasia, but also spreads both up and down the airways (Fig. 78.1). In asthma the increase in muscle and its altered function most likely underlies an important component of hyperresponsiveness that characterizes this disease. However, intensive studies have tried to identify cell and molecular abnormalities in the airway smooth muscle that may account for its abnormal behavior in this disease, but as yet there are no clear mechanisms that explain this (James 2005). What has emerged is that in chronic asthma the airways become thickened, not only due to an increase in airway smooth muscle, but also as a consequence of the laying down of new matrix proteins including collagen fibers, increased proliferation of microvessels along with vascular leakage and deposition of proteoglycans, with their ability to sequestrate water. In both children and adults, high-resolution computed tomography (HRCT) has revealed an association between airway wall thickness and disease chronicity and severity that can best be explained on the basis of remodeling (Bumbacea et al. 2004; Jain et al. 2005). Paradoxically, however, airway wall thickness is inversely correlated with airway hyperresponsiveness, suggesting that this thickening with deposition of matrix proteins may be a protective response against frequent smooth muscle contraction (Shaw et al. 2004). Thus in patients who are highly hyperreactive with brittle asthma, the airway remodeling response is minimal, whereas in those with chronic disease who exhibit some degree of fixed airflow obstruction, remodeling is more prominent (Paganin et al. 1996). Remodeling as a series of interacting processes is complex. For example, while immunostaining for proteoglycans such as biglycan, lumican, versican and decorin is increased in both moderate and severe asthma, with no differences in the amount present in the subepithelial layer, deposition of biglycan and lumican is significantly greater in the smooth muscle in moderate compared with severe asthma, suggesting a compensatory role (Pini et al. 2007).

One molecule that has emerged as involved in airway remodeling is ADAM33. This susceptibility gene for asthma was first identified by positional cloning (Van Eerdewegh et al. 2002) and has been replicated in a number of studies (Blakey et al. 2005). In being preferentially expressed in airway mesenchymal cells (fibroblasts and smooth muscle), it has been shown to be involved in the pathogenesis of airway hyperresponsiveness and decline in lung function over time (Jongepier et al. 2004; van Diemen et al. 2005). ADAM33 has metalloproteinase, fusagenic, adhesion and intracellular signaling activities and exist in at least six alternatively spliced isoforms (Van Eerdewegh et al. 2002). Although the full-length molecule (120 kDa) is expressed as a transmembrane protein, a soluble form (∼55 kDa) has also recently been identified whose levels increase in proportion to disease severity (Lee et al. 2006; Foley et al. 2007). Of the many potential biological actions that ADAM33 possesses, its proteolytic activity, which is also present in the soluble form, is likely to be important in generating growth factors that influence mesenchymal cell number and or maturation (Holgate et al. 2006).

Persistent airway inflammation may be an important factor that contributes to airway wall remodeling, including the secretion of mediators and growth factors such as TGF-β1 from eosinophils. Kariyawasam et al. (2007) have recently shown that, in those with both early and late asthmatic responses to allergen challenge, markers of airway wall remodeling such as tenascin, procollagens 1 and 3, HSP-47 and α-smooth muscle actin (marker of myofibroblasts) were all elevated beyond the 7-day time point when the inflammatory response had resolved. However, as discussed above, epithelial injury, impaired repair, and the secretion of growth factors from this structure may also contribute to the ongoing repair response characteristic of remodeling. Since these changes have been described in the absence of eosinophil infiltration, it may be that epithelial injury and remodeling is a necessary precursor to the onset of asthma, the altered microenvironment in the airway then providing the opportunity for inhaled environmental insults to gain access and generate the characteristic chronic inflammatory response involving atopy. The various growth factors involved in airway wall remodeling are similar to those involved in branching morphogenesis of the lung in the developing fetus (Bousquet et al. 2004; Holgate et al. 2006) and this has led to use of the term “epithelial mesenchymal trophic unit” to describe the interaction between the epithelium and the underlying mesenchymal cells in generating this altered structural response in chronic disease (Holgate et al. 2001; Knight et al. 2004) (Fig. 78.9). Regarding asthma not only as an inflammatory disease but also one in which there is abnormal signaling between the epithelium and mesenchymal cells has created an opportunity to search for novel therapeutics that act on this aspect of the disease. If epithelial injury and impaired repair are important as pathophysiologic mechanisms, then new treatments targeted to these pathways will represent a real advance. When applied to a damaged asthmatic epithelium in vitro, both EGF and keratinocyte growth factor are able to restore full barrier function as well reestablishing the protection of the epithelium against environmental injuries (Berlanga et al. 2002; Basuroy et al. 2006). Clinical trials are also currently in progress to investigate the effect of surfactant as a potential “barrier treatment” for chronic asthma, with the idea that by excluding environmental agents from penetrating the airway wall, continued aggravation of the underlying airway inflammation may be reduced (Babu et al. 2003).

The relative importance of airway inflammation and epithelial mesenchymal signaling in asthma pathogenesis has recently been highlighted by three studies that have demonstrated that inhaled corticosteroids, when administered to children born of asthmatic and atopic parents for 1–4 years, has no effect on the natural history of asthma (Bisgaard et al. 2006; Guilbert et al. 2006; Murray et al. 2006). Two previous trials of inhaled corticosteroids administered over prolonged periods to older children also failed to influence the course of the disease (Childhood Asthma Management Program Research Group 2000; Pauwels et al. 2003). Thus, if airway inflammation (and by implication Th2 driven-immunity) was the cause of the asthmatic process and the airway dysfunction that is associated with it, then its prevention at the time of disease onset should influence the natural cause, whereas in these trials it did not.


One aspect of airway remodeling that is frequently neglected is the role of the vasculature. In both adults and children recent biopsy studies have demonstrated a large increase in the number of microvessels present in the airways of patients with chronic asthma, with some evidence that the endothelium of these vessels is undergoing proliferation (Vrugt et al. 2000; Wilson & Robertson 2002; Barbato et al. 2006). Vascular remodeling has also been shown to occur in a chronic allergen exposure rat model (Tigani et al. 2006). Since microvascular leakage as well as recruitment of inflammatory cells is fundamental to asthma pathogenesis, a greater understanding of the cellular and molecular mechanisms involved in new vessel formation in the airways is important. The discovery of vascular endothelial growth factor (VEGF) in the damaged airway epithelium in asthma as an important mediator contributing to microvascular leak, proliferation, and vascular remodeling is being pursued with much interest (Chetta et al. 2005; Abdel-Rahman et al. 2006; Bhandari et al. 2006). An additional mediator that has recently been discovered in association with the T-cell-dependent component of the allergen-induced late asthmatic response is calcitonin generelated peptide, a potent vasodilator (Kay et al. 2007). It is likely that, as with fibroblasts and smooth muscle, this aspect of the remodeling response involves interaction of a wide range of different mediators including factors that act simultaneously on all three cell types (e.g., TGF-β).


An increase in neural networks in the asthmatic airway is also key to disease pathogenesis. Nerve growth factors or neurotrophins are clearly of importance in promoting the altered neural regulation that undoubtedly occurs in asthma (Nassenstein et al. 2006) but the precise mechanisms have yet to be defined. Nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF) as well as other members of this family are produced by epithelial cells and smooth muscle as well as a range of inflammatory cells including mast cells and eosinophils (Olgart et al. 2002). Circulating levels of NGF correlate with asthma severity (Bonini et al. 1996). The neurotrophins activate two receptor subtypes: TrkA (tropomyosin receptor kinase A) and p75 NTR (neurotrophin receptor) in the death receptor family (Frossard et al. 2004). The neurotrophins not only promote neural growth and maturation but are also important mast cell growth factors, enhance eosinophil survival, and augment human airway smooth muscle responsiveness (Kassel et al. 2001; Nassenstein et al. 2003; Frossard et al. 2005; Hahn et al. 2006) and therefore have the capacity to enhance asthmatic airway inflammation (Quarcoo et al. 2004). Bronchoconstriction provoked by irritants such as sulfur dioxide and ozone are also likely to operative via this abnormal neural network, with the release of acetylcholine and neuropeptides that have multiple effects on inflammatory cells, smooth muscle, blood vessels, and mucus-secreting cells. Although neuropeptide antagonists have so far been disappointing in asthma, the pleiotropic effects of the neurotrophins and their receptors make them more attractive therapeutic targets (de Vries et al. 2006; Watson et al. 2006).


Asthma is a heterogeneous disorder (Wenzel 2006). Although different types of asthma have longbeen recognized, such as that triggered by exposure to allergens, occupational chemicals, nonsteroidal antiinflammatory drugs (NSAIDs) and nonallergic “intrinsic” asthma, there has never been a cellular or molecular basis for these different manifestations of intermittent airflow obstruction other than exposure to a particular environmental stimulus. With the advent of fiberoptic bronchoscopy and the ability to biopsy different parts of the bronchial tree, noninvasive lung imaging such as HRCT and the development of therapeutics that target single cells or pathways, it is now clear that asthma is not a single disease but a wide range of different disorders that share some common phenotypes such as reversible airflow obstruction and associated symptoms. For the last 50 years we have only had corticosteroids, β2-adrenoceptor agonists and other bronchodilators (such as xanthines), and cromone-like drugs to treat asthma and this has led to the reinforcement of asthma as a “single disease.” Using noninvasive markers of airway inflammation suggests the presence of at least four distinct “phenotypes”: eosinophilic, neutrophilic, mixed inflammatory, and paucigranulocytic asthma (Wenzel 2006).

Eosinophilic asthma

Bronchial biopsy studies of patients with mild asthma have shown that eosinophilic airway inflammation is highly characteristic regardless of whether patients were atopic, nonatopic, aspirin-sensitive, or had occupational asthma (Brightling et al. 2003; Bochner & Busse 2005). A broad correlation between clinical asthma severity and the degree of airway eosinophilia has been recorded, especially when eosinophils appear to be activated. With the advent of sputum induction it appears that there is a complex relationship between eosinophilic inflammation and other markers of asthma, including lung function and airway hyperresponsiveness. In contrast eosinophilic airway inflammation appears to be much more closely related to the risk of severe asthma exacerbation (Green et al. 2002). This particular phenotype would be worth investigating for a potential therapeutic effect of anti-IL-5 therapy which, as referred to earlier, has failed to modify baseline asthma even though it has modifying effects on indices of remodeling.

Noneosinophilic asthma

The use of induced sputum as well as lavage and fiberoptic bronchoscopy has revealed that some patients with asthma have a sputum neutrophilia in the absence of eosinophils (Wenzel et al. 1999; Tsoumakidou et al. 2006). Other studies have noted neutrophilic inflammation in some patients with severe asthma and during virus-induced exacerbations (Wark et al. 2002). In addition, patients with severe asthma treated with oral (but not inhaled) corticosteroids also exhibit a predominantly neutrophilic airway inflammation and absence of eosinophils (Nguyen et al. 2005). Intense neutrophilic inflammation has also been reported in patients ventilated for acute severe asthma (Tonnel et al. 2001) and in those who died suddenly of asthma (Carroll et al. 1996). In general, asthma associated with neutrophils tends to be a more aggressive disease possibly with more tissue destruction and airway remodeling (Holgate & Polosa 2006). Intervention with the anti-TNF p75 fusion protein etanercept over a period of 10–12 weeks has shown clinical benefit in such patients, suggesting that as the disease becomes more chronic and severe the inflammatory phenotype changes from Th2 more toward a Th1 type (Kumar et al. 2006) (Fig. 78.11).

Figure 7811.

 Role of tumor necrosis factor (TNF)-α in severe asthma. (a) Relative expression of TNF-α mRNA in bronchial biopsies from patients with mild and severe corticosteroid-refractory asthma. (b) Immunohistochemical staining for TNF-α in severe asthmatic biopsy. (c) Effect of etanercept twice weekly for 12 weeks on methacholine airway responsiveness. (d) Effect of etanercept on asthma symptoms in severe asthma. (Reproduced with permission of Howarth et al. 2005).

Neutrophilic and mixed eosinophilic and neutrophilic asthma has also been associated with Mycoplasma pneumoniae and/or Chlamydia pneumoniae infection, with a beneficial response to macrolide antimicrobials (Johnston et al. 2006; Kraft & Hamid 2006). Whether the chronic bacterial infection is primary or secondary to the underlying asthmatic response in such patients is not known, but since asthmatic airway epithelial cells are known to be deficient in their ability to mount a primary interferon response following infection with common respiratory viruses such as rhinoviruses (Wark et al. 2005; Contoli et al. 2006), it is possible that defective innate immunity may be fundamental to the origins and progression of chronic disease. Indeed, longitudinal cohort studies have shown that children with early onset and persistent wheezing in the first 3–5 years of life and classified as severe childhood asthma tend to maintain this phenotype throughout life (Stein & Martinez 2004) and it is these individuals who appear to have increased susceptibility to virus infection early in life (Lemanske et al. 2005) and possibly are those who are also susceptible to chronic bacterial infection (Chaudhuri et al. 2005, 2006; Singh et al. 2007).

Tobacco smoking is also associated with a greater neutrophil component and, importantly, corticosteroid refractoriness in both the airways (Chaudhuri et al. 2003; Tomlinson et al. 2005) and systemically (Livingston et al. 2007). One possible explanation for this is the effect of smoking and oxidative stress in reducing histone deacetylase activity in the nuclear chromatin, thereby diminishing the opportunity for corticosteroids to access antiinflammatory genes (Adcock et al. 2005).

Paucigranulocytic asthma

Although sudden asthma death has been recorded in the absence of airway inflammation, this is highly unusual. Asthma in the absence of either neutrophils or eosinophils (paucigranulocytic asthma) has been described in which MMP-9 levels in sputum disease were normal (as opposed to elevated levels in patients with eosinophilic asthma; Simpson et al. 2005), suggesting that an abnormal epithelium or underlying mesenchyme and/or smooth muscle may itself lead to an asthma phenotype without the presence of obvious inflammation.

Refractory asthma as a distinct inflammatory phenotype

Much recent attention has focused on patients with chronic asthma taking high doses of inhaled and oral corticosteroids, long-acting β2-adrenoceptor agonists, and other antiasthma therapies and yet who remain symptomatic. This refractory asthmatic phenotype is associated with upregulation of the TNF-α pathway with increased expression of membranebound TNF-α, TNF-receptor 1, and TNF-α converting enzyme by peripheral blood monocytes (Berry et al. 2005). This is consistent with the finding of increased mucosal TNF-α expression in severe asthma compared with patients with mild disease (Howarth et al. 2005; Silvestri et al. 2006) and the identification of a phenotype that is responsive to anti-TNF therapy (Russo & Polosa 2005) (Fig. 78.8).

A range of abnormalities has been described that are said to be associated with corticosteroid-refractory asthma including defects in nuclear histone acetylation (Barnes 2006), overexpression of the β-isoform of the corticosteroid receptor serving a dominant negative function in the presence of corticosteroids (Lewis-Tuffin & Cidlowski 2006), a defect in vitamin D signaling (Xystrakis et al. 2006), and increased airway wall remodeling leading to a degree of fixed airflow obstruction (Bai & Knight 2005).


Recognizing that asthma is a complex disease involving varying components of airway inflammation, hyperresponsiveness, variable airflow obstruction, and a range of symptoms it has been possible to apply statistical methods in an attempt to dissect this complexity into subphenotypes within a population of patients with severe asthma (Wardlaw et al. 2005). Cluster analysis is one such statistical tool that seeks to organize information about variables so that heterogeneous groups of subjects can be classified into relatively homogeneous “clusters.”Haldar et al. (2005) have used this technique of cluster analysis to identify subphenotypes within a population of patients with severe disease. In an analysis of over 270 patients with severe asthma they identified four distinct phenotypes.

  • 1Patients with relatively well-controlled symptoms and minimal airway inflammation.
  • 2A group with early-onset atopic asthma with severe symptoms, persistent airway inflammation, and markedly variable airflow obstruction.
  • 3A group of mainly females who have late-onset asthma with marked symptoms but minimal eosinophilic inflammation, many of whom are obese.
  • 4A group with male predominance who have late-onset asthma characterized by persistent eosinophilic inflammation in the absence of symptoms.

It may be that these different subphenotypes have therapeutic implications but clearly much larger studies are needed to answer this question.


As we learn more about asthma subphenotypes and the immunopathology of asthma it is becoming increasingly apparent that the disorder is a spectrum of pathophysiologic processes that to some degree map onto the different therapeutic responses (Fig. 78.9). At the mildest end of the spectrum, i.e., allergic asthma associated with other allergic comorbidities such as rhinitis and atopic dermatitis, the disease is dominated by Th2-mediated inflammatory responses that are usually (but not always) responsive to corticosteroids. Also in this category is intermittent asthma associated with seasonal changes in aeroallergens, which in addition to being responsive to inhaled corticosteroids may also benefit from allergen-specific immunotherapy. In patients with anything more than mild disease evidence of altered epithelial–mesenchymal communication is found that might explain the more chronic nature of asthma as it evolves from childhood into adult disease and also its variable nature in response to a variety of environmental factors in addition to allergens such as air pollutants, infectious agents and certain drugs. Alterations in the epithelium or underlying mesenchyme leading to the generation of growth factors and cytokines that sustain inflammation can be thought of as generating a “fertile soil” in which the “seeds” of inflammation are able to take hold and persist (Davies et al. 2003; Holgate et al. 2004). The third type of asthma is one in which the epithelial–mesenchymal component becomes dominant, leading to progressive airflow obstruction, a component of which is irreversible with β2-adrenoceptor agonists and corticosteroids (James et al. 2005; Bergeron & Boulet 2006). It is in these patients that tissue damage and remodeling becomes prominent along with increased mucus secretion and the “plugging” of peripheral airways. Superimposed on these three broad pathophysiologic phenotypes are a range of environmental factors that individuals with asthma will respond to. Thus, asthma associated with NSAIDs is characterized by increased cysteinyl leukotriene production and in general a more beneficial response to antileukotriene therapy (Kumlin et al. 1992; Dahlen et al. 2002). Patients with severe allergic asthma in which IgE plays a predominant role in driving the disease are responsive to the IgE-blocking monoclonal antibody omalizumab, with attendant reductions in IgE-associated airway inflammation (Holgate et al. 2005). The more aggressive disease associated with increasing neutrophilic inflammation may respond to macrolide antimicrobials, but it is in this group that anti-TNF therapy may be beneficial. Additional factors that can influence the clinical expression of asthma include female hormone status (Beck 2001; Stanford et al. 2006), obesity (Appleton et al. 2006; McLachlan et al. 2006), infection with Aspergillus (bronchopulmonary allergic aspergillosis) (Gibson 2006), and in rare cases pulmonary vasculitis (Churg–Strauss syndrome) (Keogh & Specks 2006).


With increased knowledge asthma appears more and more complex in its pathophysiology, although genetic studies are beginning to reveal molecules and pathways that may underlie the origins of this disease (Holloway & Holgate 2004; Meurer et al. 2006). It is of particular interest that the majority of the novel genes identified through positional cloning that increase susceptibility to asthma are preferentially expressed in either the epithelium or the mesenchyme and smooth muscle rather than the immune or inflammatory pathways (Cookson 2004; Vendelin et al. 2005; Holgate et al. 2006) (Fig. 78.12). It is also of note that many of the molecules that appear to be involved in disordered epithelial–mesenchymal signaling in asthma are also utilized in fetal branching morphogenesis of the lung, suggesting that asthma at least in some of its manifestations has morphogenetic origins (Vendelin et al. 2005; Holgate et al. 2006). Babies whose airways are unable to adequately resist the early-life environmental insults such as environmental tobacco smoke, air pollutants, infections and allergens may be the children who go on to develop asthma, not because he or she has any fundamental abnormality in their adaptive immune system, but because their airways are inadequately equipped to defend themselves against environmental insults (Eder et al. 2006). Thus, in planning new therapeutic strategies it is interesting to speculate whether a novel approach to asthma might be to increase the intrinsic resistance of the airways to the environment rather than concentrating all our efforts in suppressing inflammation or manipulating the immune response.

Figure 7812.

 Asthma susceptibility genes expressed in the epithelium and mesenchyme.


The recognition that asthma is a highly heterogeneous disorder in terms of associated environmental factors, clinical expression, response to different therapies, and natural history has opened up the disease beyond allergic sensitization. This has important implications, moving us back to considering asthma as a disease whose origins lie in the airways. Human genetics linked to environmental epidemiology through epigenetics is revealing entirely novel mechanisms and approaches to treatment. The application of cell and molecular biology to different phenotypes of asthma at different points in their natural history will eventually reveal why atopy in some individuals evolves into asthma while in others it does not. An important aspect of this relates to factors that downregulate immune and inflammatory reactions such as regulatory T cells. These are considered in detail in Chapter 4.

Although at one time asthma was considered a relatively simple disease, treated with a limited range of drugs, it is now realized that all of these suppress rather than cure the disease. What is now needed is a concerted effort to understand why certain people develop asthma and others do not. It is only with this increased understanding that a real chance of cure and prevention will occur.