The Proceedings of the American Thoracic Society 1:125-135 (2004)
© 2004 The American Thoracic Society
Mucosal Immunity in Asthma and Chronic Obstructive Pulmonary Disease
A Role for Immunoglobulin A?
Charles Pilette,
Stephen R. Durham,
Jean-Pierre Vaerman and
Yves Sibille
Unit of Experimental Medicine, C. de Duve Institute of Cellular Pathology, University of Louvain, Brussels, Belgium; and Upper Respiratory Medicine, Imperial College London, London, United Kingdom
Correspondence and requests for reprints should be addressed to Charles Pilette, M.D., Ph.D., Unit of Microbiology, University of Louvain, 54 Avenue Hippocrate BP5490, B-1200 Brussels, Belgium. E-mail: charles_pilette{at}yahoo.co.uk
 |
ABSTRACT
|
|---|
Despite our knowledge on the role of IgA in mucosal homeostasis and host defense and clinical evidence suggesting deficient first-line defense mechanisms in chronic airway disorders, little is known regarding its role in asthma and chronic obstructive pulmonary disease (COPD). Studies suggest that the mucosal IgA response is impaired in COPD, and a deficient transport of IgA across the bronchial epithelium in COPD has been identified, possibly involving neutrophil proteinases, which may degrade the Ig receptor mediating this transepithelial routing. In contrast, the IgA response to allergens in patients with asthma may play a pathogenic role through eosinophil activation. Thus, secretory IgA can induce eosinophil degranulation in vitro, a feature in keeping with the correlations observed in vivo between airway IgA levels and eosinophil cationic protein during late asthmatic responses. Selective IgA deficiency is associated with an increased prevalence of atopy, and a protective role of IgA has been seen in murine models of asthma, delineating the complexity of the IgA system in the airway mucosa. Future studies will hopefully yield better knowledge of IgA biology and lung mucosal immunity and help to use more efficiently the mucosal route for immunotherapy or target specific genes in inflamed airways.
Key Words: immune tolerance • lung • mucosal defense • epithelium
The airway mucosa, which is continuously exposed to inhaled antigens and biotoxins, represents a major challenge for the immune system. This body surface of approximately 100 m2 has (like the gut, approximately 300 m2) to discriminate between noxious and harmless antigens, to achieve efficient protection against pathogens, and to avoid deleterious inflammation. Thus, defects in protective mechanisms can lead to tissue invasion and injury by pathogenic stimuli. Conversely, hypersensitivity reactions are characterized by exaggerated inflammatory immune responses to environmental antigens. This review concentrates on one of the first-line defense mechanisms of the airways, namely the secretory IgA (S-IgA) system, with a specific focus on the mucosal IgA response observed in chronic obstructive pulmonary disease (COPD) and asthma. The interactions between mucosal and systemic immunity are discussed, as well as the determinants of the delicate balance between antigen responsiveness and tolerance. Finally, mucosal active vaccination and passive immunotherapy, as well as mucosa-targeted gene therapy, which may constitute promising therapeutic approaches for chronic inflammatory diseases of the airways, will be considered.
|
FIRST-LINE DEFENSE MECHANISMS IN THE AIRWAYS
|
|---|
Whereas alveolar macrophages clear particles and antigens in the distal airways and lung alveoli (1), the first-line defense mechanisms of the upper (nose, pharynx, larynx) and lower (trachea, cartilaginous bronchi, membranous bronchioles) conducting airways rely mainly on S-IgA (Table 1). Secretory Igs have been selected through evolution to protect mucosal surfaces. They display unique properties (high valency for antigen binding and relative resistance to proteolysis by commensal microflora) to fulfill their role in the mucosae.
Resident cells of the airway wall can also participate in rapid innate responses to aggression, secreting within minutes numerous mediators with antiinfectious and/or antiinflammatory properties, including lysozyme, phospholipase A2, 
-defensins, mucins, and lectins (such as surfactant proteins and galectins), Clara cell protein, and secretory component (SC) (2). Moreover, the epithelial layer with its tight junctions constitutes a physical barrier, and epithelial cells are covered by the electronegatively charged glycocalix.
Studies in mice have shown that mucosal tissues are enriched in intraepithelial lymphocytes (IELs), especially in young individuals (3). These lymphocytes are T cells that usually express the 
T-cell receptor (TCR) and CD8 homodimer, in contrast to conventional T cells expressing ß TCR (complexed to CD3) and either the CD4 or CD8 ß coreceptor. It is thought that T cells contribute to early stages of immune responses; they recognize and kill infected epithelial cells that express major histocompatibility complex class I-like molecules as generic distress signals (4) and may thereby also act in first-line defense. Putative antiinfectious properties of IELs include direct cytolytic effects (they are rich in cytotoxic granules) and/or Th1 activity, activation of neutrophils and macrophages, and stimulation of the survival of epithelial cells via the production of epithelial growth factors. This is underscored by the observation that IEL defects in mice increase their susceptibility to invasive infections (5). In addition, recent studies suggest that T cells are also involved in the regulation of ß T-cell and B-cell adaptive responses. Thus, although the mechanisms remain to elucidate, it was shown that T cells may downregulate skin and systemic infiltration by ß T cells (6).
IgA may also provide a link between innate and adaptive immunity. Thus, IgA may be produced by nonconventional B cells (B1 cells) in a T-cell–independent fashion (7). The so-called polyclonal, polyreactive "natural" antibodies (8) produced through this process are probably crucial as innate first-line immunity before an adaptive immune response is elicited. These B1 cells are, however, preferentially attracted to the peritoneal cavity where macrophages produce a specific B lymphocyte-attracting chemokine (9). In contrast, conventional B cells (B2 cells) lead to IgA-producing plasma cells after stimulation through cognate interactions between antigen-presenting cells and T cells, in a specific, adaptive manner. In this regard, IgA is the main Ig isotype produced in secretions after mucosal immunization, including in the respiratory tract (10). Although IgA is not mandatory to develop mucosal tolerance to harmless antigens, different lines of evidence suggest that the IgA response is associated with the mechanisms leading to systemic unresponsiveness induced in mucosal sites.
|
TOLEROGENIC MUCOSAL RESPONSES
|
|---|
During the orchestration of immune responses, activation of T cells usually requires two signals: one from the TCR/CD3 complex recognizing a specific antigen presented in the context of the major histocompatibility complex on antigen-presenting cells and one from a costimulatory interaction such as that between CD28 and CD80 or CD86. Partial T-cell activation can generate regulatory subsets of CD4+ T cells, underlying the development of tolerance, that is, a state of nonresponsiveness or hyporesponsiveness to an antigen, which is mediated by some antigen-driven T cells that may induce tolerance in naive animal after passive transfer (11). This tolerance, which was described initially for proteins absorbed by the oral route but which is also effective in the respiratory tract (12), may develop through three main mechanisms: (1) clonal deletion of antigen-specific T cells (as observed after persistent activation), (2) inactivation of responsive T cells by insufficient costimulatory signals (anergy), or (3) inactivation through an active process (suppression) (13). Inaccessibility to the antigen, as achieved by first-line mechanisms, including mucosal IgA, leads to ignorance by the systemic immune system.
The peripheral mechanisms, which protect us from self-reactivity, also guard us from reacting to nontoxic inhaled particles. Although the respiratory tract is a nonabsorptive surface in contrast to the gut, both the airway and gut mucosae have a natural propensity to favor the development of tolerance. This is thought to be due to the functional characteristics of mucosal, as compared with systemic dendritic cells (DCs). Mucosal DCs preferentially induce the development of regulatory, suppressive, T cells (14) and production of IgA. These suppressive T cells typically display nonresponsiveness or low responsiveness to antigen for proliferation and cytokine production. It was shown in mice that myeloid-derived CD8– DCs, an immature phenotype producing interleukin (IL)-10, which preferentially skews T cell differentiation toward Th2/Th3, are recovered from the mucosae (respiratory tract and Peyer's patches), whereas lymphoid-derived CD8+ DCs recovered from the spleen preferentially orientate toward Th1-like responses (15). The presentation of Notch ligands on mucosal DCs in the respiratory tract exposed to allergens may also play a role in the preferential induction of regulatory T cells (16). In addition to the production of IL-10, which inhibits the costimulatory signal provided by CD28 activation (17), mucosal cells expressing CD8 (DCs or T cells) can delete reactive T cells by inducing apoptosis (18). Moreover, antigen uptake by nonprofessional mucosal antigen-presenting cells, such as major histocompatibility complex class II–expressing epithelial cells, might lead to tolerance by providing only partial T-cell activation because of their low expression of CD80/CD86 costimulatory signals (Table 2).
Oral tolerance could inhibit the development of Th2-mediated allergic diseases such as asthma. The mechanisms have been well characterized in a model of ovalbumin (OVA) allergy. A high dose of oral antigen usually leads to deletion and anergy, notably through downregulation of TCR and CD25 (IL-2R) expression, whereas a low-dose oral regimen stimulates active suppression. The CD4+ suppressor T-cell subset induced after low-dose antigen treatment shifts its cytokine profile to production of IL-10 (Tr1 cells) or transforming growth factor-ß (Th3 cells, also sometimes producing IL-4). Allergen-specific Th2 cells, as observed in atopic asthma, may be inhibited by high-dose and/or continuous oral administration of the antigen, whereas low-dose/less frequent administration might allow preferential suppression of Th1 responses (19). Interestingly, oral feeding of mice with OVA before systemic (subcutaneous or intraperitoneal) immunization may inhibit the development of asthma-like pathophysiology characterized by airway infiltration by CD4+ T cells and eosinophils, mucus overproduction, and bronchial hyperreactivity (20). The ability to induce tolerance decreases as the delay between feeding and sensitization decreases, and mucosal exposures in sensitized animals may further increase the IgE immune response (21), potentially limiting its therapeutic application in established allergic diseases. However, despite this prediction from animal models, immunotherapy for allergic rhinitis and/or asthma using mucosal routes of administration has proven safe and sometimes effective (discussed later here).
|
IMMUNOGENIC MUCOSAL RESPONSES
|
|---|
Once T cells are triggered by appropriate stimulatory signals provided by mucosal DCs (Table 2), they produce IL-2 and proliferate and differentiate toward the Th1 or Th2 phenotype depending notably on the nature of DC-derived cytokines, with IL-4 inducing Th2 while IL-12 induces Th1 polarization. Inductive sites where mucosal responses are elicited consist of lymphoid aggregates in close relationship with the epithelium. Inhaled allergens are first encountering the so-called mucosa-associated lymphoid tissue of the upper airway, the nasal-associated lymphoid tissue (NALT), which shares some features of the gut-associated lymphoid tissue and appears much more developed than in the bronchi-associated lymphoid tissue (22). In humans, the NALT consists of the lingual, palatine, and nasopharyngeal tonsils (the so-called Waldeyer's ring), with tubal tonsils and lateral pharyngeal bands as less prominent components, as well as of the lymphoepithelial structures (lymphoid aggregates and specialized microfold cells) within the nasal mucosa; the NALT drains toward the cervical lymph nodes. In contrast, a true bronchial-associated lymphoid tissue organization—with microfold cells and organized lymphoid follicles—is only observed in human bronchi under pathologic conditions, such as diffuse panbronchiolitis or rheumatoid arthritis (23), and therefore, the bronchial mucosa could rely primarily on peribronchial glands to elicit specific immunity. It is thought that the natural antigenic load (commensal flora) of the gut and upper respiratory tract underlies the development and organization of the corresponding mucosa-associated lymphoid tissue. After antigen priming, the migration of lymphocytes from a mucosal inductive site is oriented to the corresponding effector tissue by the so-called lymphocyte homing, the selectivity of the homing referring to the regional specialization of mucosal immunity (24, 25). However, although the gut and the bronchi may appear segregated in terms of generating secretory antibodies (26), the association of cranial, oral, and nasal lymphoid tissues in a same entity (sometimes called CONALT) explains the salivary immune response observed after nasal immunization (27), and oral antigen priming may protect against subsequent respiratory challenge.
When required, the mucosal immune system may rely on systemic immunity. This "informational relay" (28) consists of various pathways. First, resident DCs, after sampling antigen from the airway lumen, may migrate to regional lymph nodes to present the processed antigen to (systemic) T cells (28). Second, B cells can be activated by DCs bearing the intact antigen (29), mucosal homing of systemically activated B and T cells back to the aggressed body surface being driven by selective homing cell-surface molecules. Third, mucosal IELs may secrete chemokines (30) regulating cell trafficking, in particular for neutrophils, from the circulation to the mucosa. Fourth, surface epithelial cells activated by pathogenic, but not by nonpathogenic, bacteria secrete chemokines, including IL-8 and macrophage inflammatory protein-3 (31) underlying, respectively, the mucosal recruitment of neutrophils and macrophages. This cross-talk between mucosal and systemic immune systems may underlie three different possibilities of immune response elicited at a mucosal surface, namely (1) purely local, mucosal polymeric IgA (pIgA) response; (2) systemic cellular and antibody response (IgM, IgG, ± IgA), with or without local pIgA production; and (3) systemic unresponsiveness, with or without change in mucosal pIgA response.
|
THE IGA RESPONSE
|
|---|
Mucosal IgA production represents the highest synthesis rate (40–60 mg · kg–1 · day–1) among Igs, overcoming the production rate of all isotypes combined (32). It long remained unknown why the IgA plasma cell is the main mature mucosal plasma cell (33), with IgG-producing cells representing approximately 5 to 20% of Ig-secreting cells in the bronchi. IgA synthesized locally, that is, within the mucosal tissues, is mainly of polymeric (dimeric) isoform in contrast with serum IgA synthesized by bone marrow plasma cells, which is mostly monomeric. Although a pIgA response could be elicited by systemic immunization (34), the molecular form of IgA can be considered as an indicator of the site of induction of the immune response. Moreover, mucosal-derived pIgA is covalently linked to a small, 15-kD joining polypeptide called the J chain, secreted concomitantly by pIgA-producing cells (25). Furthermore, in secretions, pIgA is found mainly bound covalently to the SC, which represents the soluble, cleaved product of the epithelial transmembrane pIg receptor (pIgR).
In addition to J chain, which seems in some way related to homing of B cells in the mucosae (because most Ig-producing mucosal plasma cells, including IgG cells, are expressing this polypeptide), B and T lymphocytes committed to populate the intestinal mucosal surface express 4ß7 integrin, which recognizes mucosal addressin cellular adhesion molecule-1 (also called the mucosal homing receptor) on endothelial cells of gut venules (35). In contrast to the gut, 4ß7 and mucosal addressin cellular adhesion molecule-1 are only weakly expressed in the lung, and therefore, the molecular interactions underlying the recruitment of lymphocytes to the airways remain under investigation. Although other adhesion molecule pathways may be involved, such as leukocyte function-associated molecule-1/intercellular adhesion molecule-1 (LFA-1/ICAM-1) and very late antigen-4 (4ß1)/vascular cell adhesion molecule-1 (VLA-4 [4ß1]/VCAM-1), chemokine receptors might allow selective chemoattraction of B and T cells to the airways. However, only two populations of circulating lymphocytes have been identified to have a specific homing profile: the CCR4high CLA+ and 4ß7+ CCR9+ populations, which are known to migrate to the skin and the gut, respectively (36). It has been shown recently (37) that circulating IgA plasmablasts and IgA-secreting plasma cells in various intestinal and nonintestinal mucosal tissues express CCR10, a receptor for CCL28 produced by mucosal epithelial cells. There is a subpopulation of CCR4+ blood T cells, which is CLA– and 4ß7– (38), but bronchoalveolar lavage (BAL) T lymphocytes have been shown to express only low levels of CCR4, rendering them unable to respond to CCR4 ligands such as monocyte-derived chemokine (36). These lung T cells express high levels of CXCR3 and CCR5, but these receptors are present on the majority of tissue leukocytes, including in the skin and synovial fluid (39). Moreover, studies in vitro suggest that expression of chemokine receptors is phenotype and not tissue specific, with Th2 differentiation being associated with upregulation of CCR4 and CCR8, whereas Th1 cells express preferentially CXCR3 and CCR5 (40). Thus, the homing profile of lymphocytes committed to traffic to the airway and lung mucosa remains so far unknown.
Once B cells have been primed by mucosal or systemic T cells (Table 3), they recirculate and migrate to the primary mucosal site of antigen exposure (and elsewhere) where they undergo several functional changes relating to terminal B-cell differentiation, namely class switching, clonal proliferation, and somatic hypermutation. Activated B cells are characterized by class-switch DNA recombination from Cµ to downstream isotypes, a complex process tightly regulated by cytokines. Transforming growth factor-ß is the crucial cytokine inducing switching to IgA (41), whereas IL-10 as well as IL-2, IL-5, and IL-6 promote clonal proliferation of antigen-specific, IgA-committed B cells (42). These cytokines are potentially provided by resident cells of the airway wall, such as bronchial epithelial cells (43). With respect to allergic reactions, the preferential switch to IgE is induced by IL-4 and/or IL-13 (44), provided by mast cells or by infiltrating Th2 lymphocytes. The process of somatic hypermutation of variable region genes in mucosal plasma cells is twofold greater than in corresponding Ig-producing splenic plasma cells. This observation is probably relating to the high antigenic pressure of the mucosae and allows a higher "affinity maturation" of the mucosal antibody response as compared with systemic responses.
J chain–containing pIgA produced in the lamina propria must be transported across the epithelium to reach secretions. Although some IgA (particularly monomeric serum IgA) may diffuse passively, especially during inflammatory processes associated with extravasation of plasma proteins, most IgA found in secretions is transported actively across the bronchial epithelium via the pIgR-mediated transcellular routing. The pIgR, expressed on the basolateral pole of epithelial cells (and hepatocytes in rodents), binds pIgA and transports it toward the apical pole, where a cleavage releases the extracellular part of the receptor (called soluble SC) covalently linked to pIgA to generate S-IgA, and apical cleavage of unbound pIgR releases free SC into secretions (reviewed in 45). This receptor, expressed in the airways by glandular and ciliated bronchial epithelial cells (46), is upregulated in vitro by cytokines such as interferon-, IL-4, or tumor necrosis factor- (47, 48) through pathways that involve interferon- regulatory factor-1 and nuclear factor-
B. Phospholipase C–mediated signals, such as calcium increase and protein kinase C activation, may also stimulate the transport of pIgA via an increased transcytotic rate of the pIgR (49). Several factors regulate the pIgR transcytosis, which constitutes one of the most important transcellular protein–transport systems in the body, as reviewed recently (50).
|
FUNCTIONS OF IGA
|
|---|
It has been assumed that the primary role of IgA produced at mucosal surfaces is to neutralize bacteria (51) by interfering with their motility or by competing for epithelial adhesion sites. The so-called immune exclusion of microorganisms has been extended to viruses (52) and IgA-containing immune complexes potentially formed in the lamina propria. S-IgA and free SC, only of human origin, were shown recently to bind to the pneumococcal surface protein A (53) and that, in vitro, this binding could enable the pathogenic strain to enter through inverted transport into the subepithelial tissue (54). It remains, however, unclear whether this putative invasion process observed in vitro in the absence of IgA also occurs in vivo and is not epithelial cell line and pneumococcal strain specific (55).
It has been demonstrated that IgA may improve the viscoelastic properties of the airway secretions (56), an observation further implicating IgA in the first-line clearance mechanisms of inhaled particles. Interestingly, an opposite effect was observed for abnormally low or high concentrations of IgA, and a decreased mucociliary clearance was also observed in the presence of significant numbers of neutrophils (57).
IgA is also unique because of its restricted ability to use ancillary defense mechanisms, namely complement-dependent lysis and phagocytosis. Thus, IgA has a limited capacity to activate the complement system, in contrast with IgG and IgM; IgA immune complexes are able to induce only the alternate pathway of complement activation (58). Moreover, IgA can block competitively the IgG-mediated activation of the complement. Regulation of leukocyte activation by IgA occurs via the IgA-Fc receptor (FcR) (CD89), which is associated with the signaling FcR -chain homodimer, also associated with the high-affinity Fc
R and the TCR (59). Interestingly, IgA may inhibit the oxidative burst and tumor necrosis factor- release in activated monocytes (60) and the opsonizing effect of serum IgG on Haemophilus influenzae (61), further indicating that IgA may have antiinflammatory properties unlike the other Igs. However, IgA can also trigger phagocytosis, pathogen killing, and release of oxidants and proinflammatory mediators from phagocytes (62, 63). The outcome of the interactions between IgA and FcR-expressing leukocytes seems to depend on several factors, including the state of preactivation and the nature of the stimulus and cytokine milieu, as highlighted by studies of alveolar macrophages, which express a splicing variant of FcR. Alveolar macrophages from healthy subjects rapidly endocytose IgA and when incubated with pIgA or S-IgA exhibit a decreased response to bacterial endotoxin in terms of oxidative burst (64). In contrast, pIgA potentiates the effect of phorbol esters on the oxidative burst of alveolar macrophages. IgA-mediated regulation of both endotoxin- and phorbol ester–induced oxidative burst occurs through modulation of the mitogen-activated protein kinase pathway. In addition, pIgA or S-IgA increases the release of tumor necrosis factor- by alveolar macrophages via activation of nuclear factor-B, illustrating that IgA may display both stimulatory and inhibitory activities on innate immunity.
|
MUCOSAL IMMUNITY IN COPD
|
|---|
First-line Defense Mechanisms in COPD
The increased susceptibility of COPD patients to airway colonization and infections (65) is probably multifactorial. The bronchial mucosa from patients with COPD displays hyperplasia and metaplasia of mucus-secreting goblet cells in large and small airways, respectively. Although the consequences of such changes are not known, it is likely that this remodeling of the bronchial surface affects the first-line defense mechanisms. First, mucociliary clearance is impaired in COPD (66). This is probably mediated by infiltrating neutrophils, which once in the airway lumen, can release serine-proteinases, which while mediating increased mucus production (67), also induce reduced ciliary beating (68) and reduced viscoelastic properties (57). It has been suggested recently that the epithelial response to neutrophil elastase is linked to repair processes (69). Thus, neutrophil elastase may increase in bronchial epithelial cells the synthesis of MUC4, a mucin that can bind, via ErbB2, to the epithelial growth factor receptor. Interestingly, it has been shown recently that mice with elastase-induced emphysema died from sublethal inhalation of Streptococcus pneumoniae, because of insufficient mucosal response notably in terms of neutrophil recruitment to the airspaces, a feature associated with increased systemic tumor necrosis factor- response (70). Thus, although COPD airways are chronically infiltrated by neutrophils, the coordinated recruitment of leukocytes acutely required to the airway exposed to pathogens might be deficient. Another possibility is that the neutrophil influx in the alveoli is reduced in emphysema because of the destruction of the capillary bed.
Despite the absence of available data in COPD, different observations suggest that IELs may be involved in chronic inflammatory diseases of the mucosae. Thus, it has been shown that T cells potentially downregulate the inflammatory damage elicited by noninfectious agents (71), as supported by the increased susceptibility of IELs-deficient mice to develop inflammatory bowel disease (72), in addition to their increased susceptibility to invasive infections (5). Similarly, tolerance mechanisms naturally occurring in the airway mucosa may be deficient in COPD, although this remains to be assessed. Interestingly, IL-10–deficient mice are more susceptible to develop inflammatory bowel disease (73).
The Impaired Mucosal IgA Response in COPD
IgA has been assessed in the BAL fluid of smokers with COPD. However, the different methods used and different patient populations assessed render these studies difficult to compare. A decreased BAL concentration of IgA has been observed in some heavy smokers (74), whereas IgA was increased in serum from smokers with chronic bronchitis (75). A previous study also reported that IgA was increased in sputum from chronic bronchitics (76), but standard techniques used then to distinguish between the molecular forms of IgA indicated that this increase was not due to a local production of IgA. Similarly, the IgA level corrected for albumin (IgA/albumin ratio) appeared decreased in COPD patients, as well as in patients with bronchiectasis (77), suggesting an impairment of the local production of IgA within the COPD bronchial mucosa. In contrast, the systemic IgA response elicited in the bronchial mucosa seems preserved in COPD, as notably underscored by the higher prevalence of serum-specific IgA, but not IgG, antibodies to Chlamydia pneumoniae in COPD patients (78). In addition, the recurrent infections experienced by most COPD patients are not associated with deficiency in serum IgG subclasses (79), further indicating that systemic adaptive immunity is not significantly impaired in COPD.
In patients with severe COPD, it was observed that pIgR expression by bronchial epithelial cells was strongly decreased in both large bronchi and membranous bronchioles, as compared with patients with normal bronchial epithelium transplanted for primary pulmonary hypertension (80). In line with the decrease in pIgR, immunostaining for IgA was also reduced in the COPD bronchial epithelium (Figure 1). Interestingly, reduced expression of pIgR in large airways was correlated inversely with infiltration by neutrophils of peribronchial glands. These observations were in keeping with the previously described inverse relationship between locally produced IgA and inflammation (76).
View larger version (113K):
[in this window]
[in a new window]
|
Figure 1. Epithelial content of IgA in small airways from a control subject with primary pulmonary hypertension (A and B) and a patient with severe chronic obstructive pulmonary disease (C and D), as assessed by immunohistochemistry using goat antiserum against human chain, at x40 (A and C) and x400 (B and D) magnifications. Bronchiolar epithelium from control exhibits a consistent staining; in contrast, IgA immunostaining is strongly reduced in chronic obstructive pulmonary disease epithelium, in line with the decreased polymeric lgR expression (80).
|
|
Mechanisms of Impaired S-IgA Response in COPD
The hypothesis that inflammation could impair the bronchial transport of IgA was recently confirmed. Thus, both cell-surface pIgR and soluble SC are highly sensitive to neutrophil serine proteinases (mainly neutrophil elastase and proteinase-3) (81). This observation, which provides a molecular basis for the decreased pIgR content in inflamed COPD airways, was challenged when interactions between normal epithelial cells and neutrophils were considered. Thus, neutrophils could potentiate the transport of IgA across the normal bronchial epithelium, notably through epithelial cell activation of nuclear factor-B and p38 mitogen-activated protein kinase pathways (80). It may be hypothesized that epithelial cell reactivity might be different in COPD and does not allow pIgR upregulation in the presence of activated neutrophils. On the other hand, IgA can stimulate several functions of neutrophils, through interaction with FcR, which is highly expressed on these leukocytes (82), which can in turn degrade the epithelial pIgR. Nevertheless, IgA can enhance protective mechanisms such as the phagocytosis and oxidant-dependent killing of pathogens by neutrophils, as shown for Pseudomonas aeruginosa (83).
If neutrophils contribute to the decreased content in pIgR observed in bronchi from COPD patients, impaired production of S-IgA and SC may potentially enhance the recruitment of neutrophils through different mechanisms. First, limited first-line defense normally provided by S-IgA and SC might induce a more pronounced recruitment of neutrophils to the airway. Second, it was shown that free SC can inhibit the neutrophil chemotactic activity of IL-8 via the formation of inactive covalent complexes with this chemokine (84). Moreover, neutrophil elastase can degrade the different Ig classes (Figure 2), including IgA, although S-IgA is to some extent protected from cleavage into IgA monomers by its covalent association with SC (81). Thus, the identified impairment of the IgA system within the bronchial mucosa in COPD appears closely related to neutrophilic inflammation and may contribute to the disease pathogenesis, as supported by the relationship between reduced pIgR in small airways and lung function tests of airflow obstruction (80). Impaired bronchial production of S-IgA could favor airway bacterial colonization and infections, which may trigger the recruitment of neutrophils, as observed in COPD colonized with H. influenzae (85). Although neutrophil-derived elastase and proteinase-3 can degrade the pIgR/SC, both neutrophil and bacterial products can cleave IgA, leading to a vicious circle of impaired mucosal immunity and exaggerated systemic inflammatory response, potentially underlying the bronchial and lung damage associated with COPD.
View larger version (142K):
[in this window]
[in a new window]
|
Figure 2. Purified serum IgG, IgE, and IgM were incubated with neutrophil elastase at 37°C for 3 minutes, 1 hour, or 20 hours at 10:8 of substrate:enzyme molar ratio and were subjected to electrophoretic analysis in sodium dodecyl sulfate-polyacrylamide gradient (5 to 20%). * IgM has been reduced before gel analysis. In the last lane, molecular weight markers (kD) are shown. Monomeric immunoglobulins IgG and IgE were significantly cleaved (especially at 20 hours), whereas pentameric IgM appeared resistant to elastase-mediated proteolysis, as observed for both secretory and monomeric IgA (81).
|
|
|
MUCOSAL IMMUNITY IN ASTHMA
|
|---|
First-line Defense Mechanisms in Asthma
As in COPD, mucociliary clearance is impaired in asthma (86), both in large and small airways, and this defect is correlated with asthma severity in terms of airflow limitation. Mucociliary clearance decreases further during asthma attacks (87) and on allergen challenge (88). This defect in the mucociliary system is likely to contribute to the formation of mucus plugs representing one pathologic hallmark of the disease. Moreover, the shedding of bronchial epithelial cells, another classic feature of asthma, is likely to increase the permeability of the mucosal barrier to foreign particles, favoring thereby both invasion by pathogens and sensitization to allergens.
It has been hypothesized that early skewing toward Th1 response, driven by T cells, which are usually biased toward Th1 phenotype (with high production of interferon-), could accelerate the bias away from Th2 response and, therefore, reduce the susceptibility to allergy (3). Interestingly, after OVA sensitization (by the systemic route) followed by aerosol challenge, TCR-deficient mice exhibit decreased levels of BAL antigen-specific IgA and IgG and absence of serum OVA-specific IgE production, whereas the ß T-cell response could be evoked and driven toward a Th2 type (89). Thus, T cells may be necessary to support systemic IgE production and the mucosal immune response, notably via signals underlying B-cell homing or local Ig production. The mechanisms of cross-talk between ß and T cells remain, however, to be defined.
The IgA Response in Asthma
Numerous studies have indicated that mucosal and systemic production of IgA, specific to the allergen, occurs in patients with asthma and/or allergic rhinitis. The mucosal IgA response has been well studied in the nasal mucosa of patients with allergic rhinitis after allergen challenge (90). The increase in IgA levels in the nasal fluid appeared biphasic, with an early response (10 minutes) before normalization within 1 hour and a late response. Interestingly, the IgA/albumin ratio was decreased in the early phase while increased in the late phase, suggesting that the initial increase in IgA is due to plasma leakage and the late IgA response is relating to activation of its local production and epithelial transport. A specific IgA response has been documented in the nasal and bronchial mucosa from patients with atopic asthma and/or rhinitis sensitized to house dust mite (Dermatophagoides farinae) (91), grass pollen (92), or ragweed pollen (93). Similarly, an increase in IgM has been observed in BAL from subjects with asthma (94, 95), and both the correction for 2-macroglobulin and the association of IgM to SC (S-IgM) indicated that this increase was relating to a local production and not to transudation from peripheral blood (95). However, the results of studies, which have assessed SC in airway secretions from subjects with asthma, are more conflicting. Thus, although some authors (96) found seasonal increases in free SC in patients with hay fever or seasonal asthma, Van Vyve and colleagues (94) observed that BAL SC was decreased in asthmatics despite the increase in IgM.
In asthma, mucosal B cells preferentially undergo class-switch recombination to C. This process, under the control of IL-4 and/or IL-13 and leading to IgE production, occurs locally within the bronchial mucosa in subjects with asthma on allergen exposure (97). The mucosal production of IgE and the subsequent degranulation of mast cells on cross-linking of cell-surface IgE by the allergen, is a major pathway of the inflammatory response underlying asthma (98). In contrast, it remains unclear how and to what extent the mucosal IgA response, although well-documented in asthma, plays a role in the allergic reaction. The specific IgA response in asthma does not seem to represent an epiphenomenon of the IgE-mediated reaction, as it was shown in grass pollen-sensitive patients that the epitope specificity of IgA antibodies is different from that of IgE antibodies (92). Moreover, the IgA response has been linked to eosinophil activation, which represents also a key player in asthma (98). Thus, in ragweed-sensitive subjects with asthma challenged with the allergen, BAL levels of eosinophil cationic protein were strongly correlated with those of BAL and serum IgA (99), as well as with serum IgE. In contrast, healthy control subjects had only detectable ragweed-specific IgA and no IgE in their BAL. Patients with detectable levels of serum and BAL allergen-specific IgA before challenge did have significantly stronger late-phase responses after allergen exposure. The correlation of the mucosal specific IgA response with markers of eosinophil degranulation such as eosinophil cationic protein has been observed in several studies in subjects with asthma or patients with allergic rhinitis (90, 99, 100) as well as in chronic eosinophilic pneumonia (101), suggesting that IgA may be implicated in eosinophil activation and degranulation in vivo.
IgA and Eosinophils in Asthma
Several studies have shown that S-IgA is a potent stimulus for eosinophils and, among Igs, represents the main trigger of eosinophil degranulation. Blood eosinophils incubated with S-IgA in vitro release large amounts of eosinophil cationic protein, eosinophil peroxidase, eosinophil-derived neurotoxin, as well as IL-4 and IL-5 (102, 103). This stimulatory effect of IgA on eosinophils is mediated at least partly by the FcR, which is upregulated on eosinophils from allergic patients (104). Moreover, it was shown that eosinophils from atopic subjects do not need cytokine priming by IL-4 or IL-5 to bind IgA via FcR in vitro (105), in contrast to eosinophils from healthy donors. It was further suggested that eosinophils from atopics are primed in vivo to bind IgA, and this preactivation could occur through an inside-out signaling involving p38 and PI-3 kinase pathways (106). Activation of eosinophils by IgA occurs preferentially with its secretory form and is not—or is to a much lower extent—induced by serum IgA. It is thought that the preferential effect of S-IgA is relating to the possibility that eosinophils may express a specific receptor, of C-lectin type, for SC (107), as supported by the observation that eosinophils may be activated by free SC (108). Finally, IgA may participate in the regulation of the immune response through modulation of the cytokine profile. First, IgA-mediated activation of eosinophils leads to IL-4 and IL-5 production. Second, eosinophils release interferon- on CD28 activation, and it has been shown that S-IgA complexes can inhibit this effect, possibly through IL-10 upregulation (109). IgA may thus favor a Th2 profile through the modulation of the cytokine response of eosinophils.
In addition to eosinophils, basophils also degranulate on activation by S-IgA (110). This view of the IgA response as a pathogenic mechanism in asthma, acting along with IgE, is, however, challenged by several observations. First, selective IgA deficiency or delayed serum IgA production in childhood is a well-known risk factor for atopy (111). Second, unlike IgE, antigen-specific IgA is detected in BAL from healthy subjects (99). Third, intranasal treatment of mice with antigen-specific monoclonal IgA antibody prevented increases in bronchial hyperreactivity, tissue eosinophilia, and IL-4 and IL-5 production after allergen challenge (112), suggesting that neutralization of aeroallergens by IgA is a protective mechanism achieving a form of "molecular allergen avoidance." This may underlie the induction of tolerance by repeated high-dose immunotherapy, which induces in patients with allergic airway disease a preferential switching to IgA (113) (unpublished data) along with IgG4. Moreover, the mucosal treatment by specific IgA may increase the serum concentration of IgG2a, while failing to decrease total serum IgE and IgG (112), indicating that IgA may also regulate the systemic response toward a Th1 pattern.
|
LESSONS FROM IGA DEFICIENCIES
|
|---|
Selective IgA deficiency
A selective deficiency, defined by serum IgA of less than 0.05 mg · ml–1, is the most frequent human defect in humoral immunity, with a prevalence of 1/2,000 to 1/500. This defect, affecting mucosal and systemic plasma cells, is due to genetic defects in the mechanisms of class switching to IgA (25). Although these subjects are usually healthy, selective IgA deficiency has been associated with an increased prevalence of atopy, food sensitization, recurrent infections, and neoplastic and autoimmune disorders. Interestingly, IgA deficiency has also been associated with an increased risk to develop COPD (114), as observed for deficiency in IgG subclasses (115). It has been described that the association of IgA and 1-antitrypsin deficiencies leads to bronchiectasis, emphysema. and recurrent infections (116).
S-IgA–deficient mice
Transgenic mice deficient in pIgR expression were generated recently (117), which are characterized by absence of S-IgA and S-IgM, and appear relatively healthy. They exhibit, however, very high levels of serum dimeric IgA (dIgA), probably relating to both absence of transport of dIgA produced at mucosal surfaces and increased numbers of IgA-producing cells. Moreover, pIgR-deficient mice have in their serum increased levels of IgG, including specific antibodies to Escherichia coli, a feature consistent with an impaired mucosal barrier. In addition, increased albumin levels are also observed in secretions because of increased mucosal leak underlying plasma protein exudation. The susceptibility of pIgR-deficient mice to pathogens and allergens is under investigation. A recent study indicates that induction of oral tolerance to OVA seems unaffected in these mice, as well as a cytotoxic T-lymphocyte response to influenzae virus (118).
Similarly to pIgR-deficient mice, mice deficient in IgA (119) or in J chain expression (120) are healthy but display, respectively, increased numbers of mucosal activated B cells and impaired antitoxin protection after oral immunization with cholera toxin. Finally, mice deficient in activation-induced cytidine deaminase, an enzyme required for class-switch recombination (121) that is selectively expressed in germinal-center activated B cells, have features mimicking the hyper-IgM syndrome. They exhibit hypertrophy of isolated lymphoid follicles, accumulation of IgM+ plasma cells with lack of IgA+ and IgG+ cells, and an increase in the number of intestinal nonpathogenic anaerobic bacteria (122). These data support that IgA not only protects mucosae against pathogens but also prevents overstimulation of the systemic immune system by controlling the commensal flora.
|
THE MUCOSAL ROUTE TO TREAT INFLAMMATORY AIRWAY DISEASES
|
|---|
Passive Mucosal Immunization
Several experimental models have documented a protective effect of mucosal administration of IgA antibodies against various infectious agents, including influenza virus (123). This potential protection is relevant to asthma and COPD, which are characterized by episodes of exacerbation, sometimes due to bacterial or viral infections (see reviews by Sethi [124] and Wedzicha [125], respectively, in this issue). However, the degradation of IgA within the airway lumen in the presence of neutrophil enzymes, although limited by SC, could be a drawback to that approach. Recently, the application of recombinant anti-Streptococcus mutans S-IgA/IgG chimeric antibodies, produced in tobacco plant cells, provides long-lasting protection against colonization of the oral mucosa (126). One may also use recombinant S-IgA antibodies, that is, pIgA associated with both J chain and SC, from mammalian origin (to avoid differences in assembly and glycosylation of SC, as observed with plant cells), which can be produced in the same cell after appropriate transfections, and their specificity may be selected by cloning Ig-variable region genes. These S-IgA antibodies could achieve both protective effects against infections and possibly also against allergens (112).
Mucosal Vaccination
Active immunization at mucosal sites may under appropriate conditions elicit a specific mucosal IgA response. The mucosal response may confer an advantage to systemic vaccination, as suggested by the observation that protection against reinfection by various pathogens is better correlated with levels of specific S-IgA antibodies than with serum antibodies (127). This may be related to the protective response being induced in the target organ, where the pathogen will first encounter the immune system and/or to the higher affinity/avidity of mucosal antibodies. The nasal route has been studied most because NALT is much more developed than bronchi-associated lymphoid tissue under normal circumstances, intranasal administration is relatively easy, and usually, the antibody response can be assessed readily in the saliva since salivary glands are included in NALT (as referred by CONALT terminology). A protective mucosal IgA response elicited by intranasal vaccination has been documented for S. pneumoniae (S-IgA antibodies to capsular polysaccharides), influenzae, and adenoviruses (127, 128). A sustained pIgA response to capsular polysaccharides of S. pneumoniae has been associated with a better mucosal clearance of the bacteria (129). Despite this, only a few vaccines are available for human use via mucosal routes (polio OPV, influenza A; and not in routine: typhoid, cholera, and adenovirus). Limiting factors of the mucosal route for vaccination are related mainly to the difficulty to elicit a response to nonreplicating agents, probably because of their rapid clearance and/or degradation. Therefore, different strategies are developed, including subunit vaccines, synthetic peptides, DNA vaccines, recombinant vaccines, and liposome-complexed vaccines. Finally, recombinant "antigenized" (with the epitope from a pathogen) S-IgA molecules may be used as mucosal vaccines because S-IgA can bind to microfold cells, which then transport the antigenized molecule under the epithelium for immunogenic interaction with DCs and lymphocytes (130).
Besides vaccination against infectious pathogens, the mucosal route may also be used to induce tolerance to allergens or autoantigens. Different conditions of immunization are required to potentially lead to tolerance (Table 2), namely the use of T-cell–dependent soluble antigens (replicating, particulate antigens usually induce an immunogenic response) and the use of adjuvants promoting tolerance and usually also the IgA response, such as endotoxin, the B subunit of cholera toxin (128), or alum. Alternatively, the use of a mucosal route is by itself in favor of the development of tolerance. This form of immunotherapy implies the suppression of systemic immunity to a mucosally administered antigen. The oral route leading to induce immune tolerance has been mainly used to treat autoimmune disorders, such as rodent experimental autoimmune encephalitis, connective tissue diseases and type I diabetes mellitus, and recently, inflammatory bowel disease. In contrast, allergen immunotherapy is classically achieved through the systemic, parenteral (subcutaneous) route, whereas recent studies suggest that mucosal (sublingual, oral or nasal) immunotherapy could be efficient to treat patients with allergic airway disease.
Oral feeding of sensitized mice with house dust mite (Der p 1) allergen reduced the IgE and IgG response (131). A recent study using a dog model of OVA allergy showed that oral administration of the allergen induces reduced responses to ensuing subcutaneous, ocular, and airway challenges (132). Interestingly, induction of tolerance in this model was associated with a local, mucosal production of IL-10 and transforming growth factor-ß probably by macrophages. However, oral immunotherapy has had significant clinical benefit in only some human studies. Results of clinical trials confirm that the sublingual route (which does not represent a strictly mucosal route because of the high level of systemic absorption), using high dose of allergen, could represent an effective alternative to subcutaneous injection immunotherapy to treat patients with allergic rhinitis and/or asthma (133–136). Intranasal treatment of sensitized mice to birch pollen (Bet v 1) allergen decreased the IgE response and tissue eosinophilia (137). Moreover, it has been recently shown that intranasal immunotherapy of mice induces airway tolerance to OVA more efficiently as compared with intradermal treatment, whereas the systemic responsiveness is similarly modulated (138). In humans, however, the effects of vaccination through the airway route are more complex. Thus, low-dose inhaled allergen treatment decreases the late-phase response to a high-dose antigen challenge, despite an increased airway hyperreactivity and tissue eosinophilia prior to the challenge (139). There is epidemiologic evidence that high respiratory exposure to cat allergens in children may induce a form of immune tolerance, as achieved by allergen immunotherapy, through a so-called modified Th2 response characterized by IgG and IgG4 production (140). Interestingly, immune deviation toward IgG production (including blocking IgG antibodies) is one of the mechanisms by which subcutaneous immunotherapy acts (141). This protective IgG response observed in patients under allergen immunotherapy could be related to the induction of IL-10 (142), which can increase together with IL-4 IgG4 synthesis while inhibiting IgE (143). Interestingly, mature lung DCs, which stimulate the development of regulatory T cells, also produce IL-10 (144), and the cellular source of mucosal IL-10 induced during tolerogenic responses predominantly consists of a CD14+ monocyte/macrophage-like population (132, 142). Thus, although further clinical trials are required, immunologic findings support that the mucosal route to induce tolerance in allergic patients, including those with asthma, may represent an attractive alternative to systemic immunotherapy.
Polymeric Immunoglobulin Receptor-targeted Gene Therapy
Selective expression of the pIgR transporter in mucosal tissues may be used to target expression of normal copies of a gene, which is mutated or of a gene encoding a protein mediating protective activity. Thus, expression plasmids encoding the cystic fibrosis transmembrane conductance regulator (mutated gene in cystic fibrosis) complexed to anti-SC antibodies are specifically and efficiently incorporated into epithelial cells (145). Unfortunately, problems limiting this promising treatment are related to variable expression levels or the route of administration. Similarly, 1-antitrypsin complexed to anti-SC antibody fragment is also efficiently transported across epithelial cells (146) and could provide a strategy to target the delivery of this major antiproteinase into the bronchial lining fluid from patients with cystic fibrosis or COPD, exhibiting unopposed neutrophil elastase activity.
CONCLUSIONS AND PERSPECTIVES
IgA was initially thought to mediate the reaginic activity of the serum (147). After the identification of the factor mediating the skin-sensitizing reaginic activity as IgE (148), the role of IgA has been mainly confined to a "protective" immune protein acting as a scavenger in mucosal secretions. More recent studies have identified the mechanisms of IgA transepithelial transport and extended its role to various functions in mucosal immunity, including immunomodulatory properties on innate immunity and effector leukocytes such as eosinophils, neutrophils, and alveolar macrophages. It is expected that future studies will characterize the role of IgA in asthma and COPD more precisely and improve our knowledge regarding the mucosal route as a unique, discriminant, and efficient way to induce in humans protective immunity to pathogens and immune tolerance to allergens. These achievements will allow us to understand better the role of IgA and mucosal immunity in the airways, hopefully leading to new therapeutic strategies, including mucosal immunization and mucosa-targeted gene therapy for COPD and asthma and/or their exacerbations.
|
FOOTNOTES
|
|---|
Supported by the Fonds National de la Recherche Scientifique, Belgium (C.P.), and the European Respiratory Society, Lausanne, Switzerland (grant no. LTRF2002–037).
(Received in original form June 25, 2003; accepted in final form September 23, 2003)
|
REFERENCES
|
|---|
- Sibille Y, Reynolds HY. Macrophages and polymorphonuclear neutrophils in lung defense and injury. Am Rev Respir Dis 1990;141:471–501.[Medline]
- Phalipon A, Corthesy B. Novel functions of the polymeric Ig receptor: well beyond transport of immunoglobulins. Trends Immunol 2003;24:55–58.[CrossRef][Medline]
- Hayday AC, Roberts S, Ramsburg E. Cells and the regulation of mucosal immune responses. Am J Respir Crit Care Med 2000;162:S161–S163.[Abstract/Free Full Text]
- Janeway CA Jr, Jones B, Hayday A. Specificity and function of T cells bearing gamma delta receptors. Immunol Today 1988;9:73–76.[CrossRef][Medline]
- Hayday AC. Gammadelta cells: a right time and a right place for a conserved third way of protection. Annu Rev Immunol 2000;18:975–1026.[CrossRef][Medline]
- Shiohara T, Moriya N, Hayakawa J, Itohara S, Ishikawa H. Resistance to cutaneous graft-vs.-host disease is not induced in T cell receptor delta gene-mutant mice. J Exp Med 1996;183:1483–1489.[Abstract/Free Full Text]
- MacPherson G, Kushnir N, Wykes M. Dendritic cells, B cells and the regulation of antibody synthesis. Immunol Rev 1999;172:325–334.[CrossRef][Medline]
- Bouvet JP, Dighiero G. From natural polyreactive autoantibodies to a la carte monoreactive antibodies to infectious agents: is it a small world after all? Infect Immun 1998;66:1–4.[Free Full Text]
- Ansel KM, Harris RB, Cyster JG. CXCL13 is required for B1 cell homing, natural antibody production, and body cavity immunity. Immunity 2002;16:67–76.[CrossRef][Medline]
- Tomasi TB Jr. Mechanisms of immune regulation at mucosal surfaces. Rev Infect Dis 1983;5:S784–S792.
- Gershon RK, Kondo K. Infectious immunological tolerance. Immunology 1971;21:903–914.[Medline]
- Holt PG, Batty JE, Turner KJ. Inhibition of specific IgE responses in mice by pre-exposure to inhaled antigen. Immunology 1981;42:409–417.[Medline]
- von Herrath MG, Harrison LC. Regulatory lymphocytes: antigen-induced regulatory T cells in autoimmunity. Nat Rev Immunol 2003;3:223–232.[CrossRef][Medline]
- Faria AM, Weiner HL. Oral tolerance: mechanisms and therapeutic applications. Adv Immunol 1999;73:153–264.[Medline]
- Stumbles PA, Thomas JA, Pimm CL, Lee PT, Venaille TJ, Proksch S, Holt PG. Resting respiratory tract dendritic cells preferentially stimulate T helper cell type 2 (Th2) responses and require obligatory cytokine signals for induction of Th1 immunity. J Exp Med 1998;188:2019–2031.[Abstract/Free Full Text]
- Hoyne GF, Dallman MJ, Lamb JR. T-cell regulation of peripheral tolerance and immunity: the potential role for Notch signalling. Immunology 2000;100:281–288.[CrossRef][Medline]
- Akdis CA, Blaser K. Mechanisms of interleukin-10-mediated immune suppression. Immunology 2001;103:131–136.[CrossRef][Medline]
- Suss G, Shortman K. A subclass of dendritic cells kills CD4 T cells via Fas/Fas-ligand-induced apoptosis. J Exp Med 1996;183:1789–1796.[Abstract/Free Full Text]
- Melamed D, Fishman-Lovell J, Uni Z, Weiner HL, Friedman A. Peripheral tolerance of Th2 lymphocytes induced by continuous feeding of ovalbumin. Int Immunol 1996;8:717–724.[Abstract/Free Full Text]
- Russo M, Nahori MA, Lefort J, Gomes E, de Castro Keller A, Rodriguez D, Ribeiro OG, Adriouch S, Gallois V, de Faria AM, et al. Suppression of asthma-like responses in different mouse strains by oral tolerance. Am J Respir Cell Mol Biol 2001;24:518–526.[Abstract/Free Full Text]
- Holt PG, Leivers S. Tolerance induction via antigen inhalation: isotype specificity, stability, and involvement of suppressor T cells. Int Arch Allergy Appl Immunol 1982;67:155–160.[Medline]
- Pabst R. Is BALT a major component of the human lung immune system? Immunol Today 1992;13:119–122.[CrossRef][Medline]
- Sato A, Hayakawa H, Uchiyama H, Chida K. Cellular distribution of bronchus-associated lymphoid tissue in rheumatoid arthritis. Am J Respir Crit Care Med 1996;154:1903–1907.[Abstract]
- Butcher EC, Picker LJ. Lymphocyte homing and homeostasis. Science 1996;272:60–66.[Abstract]
- Brandtzaeg P, Farstad IN, Johansen FE, Morton HC, Norderhaug IN, Yamanaka T. The B-cell system of human mucosae and exocrine glands. Immunol Rev 1999;171:45–87.[CrossRef][Medline]
- Nadal D, Albini B, Chen CY, Schlapfer E, Bernstein JM, Ogra PL. Distribution and engraftment patterns of human tonsillar mononuclear cells and immunoglobulin-secreting cells in mice with severe combined immunodeficiency: role of the Epstein-Barr virus. Int Arch Allergy Appl Immunol 1991;95:341–351.[Medline]
- Csencsits KL, Jutila MA, Pascual DW. Mucosal addressin expression and binding-interactions with naive lymphocytes vary among the cranial, oral, and nasal-associated lymphoid tissues. Eur J Immunol 2002;32:3029–3039.[CrossRef][Medline]
- Hayday A, Viney JL. The ins and outs of body surface immunology. Science 2000;290:97–100.[Abstract/Free Full Text]
- Gordon J, Pound JD. Fortifying B cells with CD154: an engaging tale of many hues. Immunology 2000;100:269–280.[CrossRef][Medline]
- Boismenu R, Feng L, Xia YY, Chang JC, Havran WL. Chemokine expression by intraepithelial gamma delta T cells: implications for the recruitment of inflammatory cells to damaged epithelia. J Immunol 1996;157:985–992.[Abstract]
- Neish AS, Gewirtz AT, Zeng H, Young AN, Hobert ME, Karmali V, Rao AS, Madara JL. Prokaryotic regulation of epithelial responses by inhibition of IB- ubiquitination. Science 2000;289:1560–1563.[Abstract/Free Full Text]
- Conley ME, Delacroix DL. Intravascular and mucosal immunoglobulin A: two separate but related systems of immune defense? Ann Intern Med 1987;106:892–899.
TOP
ABSTRACT
FIRST-LINE DEFENSE MECHANISMS IN...
