ABSTRACT
Markedly increased numbers of goblet cells in the conducting airways is a pathophysiological feature of asthma and chronic bronchitis, two common severe respiratory conditions (1,2). Chronic bronchitis is a clinical definition based upon long-standing sputum production with the implication that this is associated with mucus hypersecretion in the respiratory tract. Hypersecretion, chronic bronchiolitis (small airways disease), and emphysema (alveolar destruction) comprise chronic obstructive pulmonary disease (COPD). The relative contribution of each component to disease progression varies between patients. Airway submucosal gland hypertrophy is also a feature of asthma and COPD. However, hypersecretion associated with gland hypertrophy is likely to be in the more proximal airways, where cough aids mucociliary clearance to expel excess mucus. Consequently, gland hypertrophy has received only scant attention in experimental models of airway hypersecretion (3). The present chapter considers goblet cell hyperplasia in asthma and COPD and examines data from animal models of these
two respiratory conditions in terms of induction mechanisms of hyperplasia and of changes in mucin (MUC) gene expression. AIRWAY MUCUS
Under normal circumstances, a thin film of viscoelastic liquid protects the epithelial surface of the airways. The liquid is often referred to as “ mucus” and is a dilute and complex aqueous solution of electrolytes, mucous glycoproteins (termed mucins), proteoglycans, enzymes, antienzymes, oxidants, antioxidants, bacterial products, antibacterial agents, lipids, cellular mediators, and plasma-derived proteins and mediators (4). The liquid forms an upper gel layer, for entrapment of inhaled particles, and a lower sol layer in which cilia beat. Intimate interaction between mucus and beating cilia facilitates removal of inhaled particles from the lung, a process termed mucociliary clearance. The efficiency of mucociliary clearance is directly related to the elasticity and inversely proportional to the viscosity of the gel layer and is dependent upon the depth of the sol layer. Chronic increases in volume and viscosity of the mucus layer impair clearance and precipitate hypersecretory conditions of the airways, for example, asthma and chronic bronchitis. RESPIRATORY MUCINS IN HEALTH AND DISEASE
The viscoelastic properties of airway mucus are attributed largely to high molecular weight mucins that are secreted by epithelial goblet cells and submucosal gland mucous cells. Other cells in the airways that may be secretory, although not necessarily of mucin, are the epithelial serous cell and its equivalent in the glands, the ciliated cell and the Clara cell. In airway hypersecretory conditions, goblet cell hyperplasia is associated with reductions in serous, ciliated, and Clara cells. Mucins comprise a peptide backbone, termed apomucin, to which multiple oligosaccharide side chains are bound (5). Apomucins are encoded by several genes (see next section) and are expressed in goblet cells and submucosal glands.Goblet cells and submucosal glands may produce different mucins. Postmortem analysis of the trachea from a single individual indicated that the product of mucin gene 5AC (MUC5AC) was a goblet cell mucin rather than a gland mucin (6). In contrast, MUC5B was found mainly in glands, but also in goblet cells, and there was a suggestion that there were populations of gland cells containing different glycoforms of MUC5B (7). It is vital to determine whether or not this differential expression of mucins between goblet cells and glands is retained after analysis of further samples because it has implications for categorizing different diseases, for developing appropriate disease models, and for rationale design of therapeutic drugs (see below).MUC5AC is the major gel-forming species in pooled secretions from young
healthy nonsmokers aspirated into tracheal cannulae at minor dental surgery under general anaesthesia (8-10). These aspirates are the most “ normal” respiratory secretions described to date. Analysis of the viscid secretions collected postmortem from an asthmatic patient demonstrates both MUC5AC (9) and a low charge glycoform of MUC5B (11). MUC5B is also a major mucin species in sputum from a patient with chronic bronchitis (7). In contrast, although MUC5AC is present in chronic bronchitic sputum (10), there is a suggestion that it may be only a minor component of the secretions (9). Thus, although requiring confirmation in more samples, these preliminary observations indicate an intrinsic difference between the secretions from normal, asthmatic, and chronic bronchitic subjects, with MUC5AC more characteristic of normal and asthma and MUC5B more characteristic of chronic bronchitis. It also indicates that normal and asthmatic airway secretions are derived more from goblet cells, with COPD secretions derived more from glands (see above). If supported by further data, the above suggestions imply that specific MUC profiles may need to be generated in experimental models of asthma or chronic bronchitis if they are to have relevance to disease. Interestingly, MUC2 is not found in secretions from “ normals” or bron-chitics (6). However, in sputum from patients with cystic fibrosis (CF), although MUC5AC and MUC5B were the major mucin species found, small amounts of MUC2 were also present (12). This observation suggests a further MUC profile for a specific hypersecretory condition. HUMAN RESPIRATORY MUCIN GENES IN HEALTH AND DISEASE
Currently, ten human mucin genes are recognized: MUC1-4, MUC5AC, MUC5B, and MUC6-9 (Fig. 1). Using currently available antisera, the MUC5AC and MUC5B gene products appear to be the major gel-forming moieties in airway secretions from healthy individuals and patients with asthma or chronic bronchitis (see previous section). A greater range of antisera is required to more precisely define the MUC content of secretions. Generation of molecular probes for MUC gene mRNA has superseded that of probes for MUC gene products, leading to information on the distribution of MUC genes in the airways.The expression of the MUC2 gene in the airways is thought to depend upon the induction of mucus overproduction. For example, transcription of MUC2 is upregulated by Pseudomonas aeruginosa lipopolysaccharide, present in the lungs of CF patients (13), and by TNF-a in human airways in vitro (14). A separate study comparing expression of the MUC1, 2, and 5AC genes in noninflamed nasal epithelial cells from CF, allergic rhinitis, and control subjects found that in all groups the mRNA levels of MUC5AC were greater than those for MUC1 and 2. The expression of MUC2 in all subjects remained constant, whereas the ratio of MUC5AC to MUC2 gene expression decreased significantly in CF cells
when compared to normals (15). In contrast, levels of MUC5AC mRNA increase in cultured human airway epithelial cells exposed to airway bronchoalveolar lavage (BAL) fluid from asthmatic patients or allergen-challenged allergic dogs(16). Further work in vitro and in vivo in mice suggested the involvement of the Th2 cell cytokine IL-9, present in BAL and upregulated in asthma (17). This cytokine is thought to trigger a signal cascade, which in turn activates MUC5AC transcription by activating epithelial IL-9 receptors normally present on the airway epithelium.There is increasing evidence to suggest that inflammatory mediators increase the expression of mucin genes. However, little is known about the molecular mechanisms involved in this upregulation. The inflammatory mediator neutro
phil elastase, known to stimulate mucus secretion acutely, increases the mRNA and protein expression of MUC5 AC in A549 human carcinoma cells and primary respiratory epithelial cells in culture (18). In A549 cells the increase was due to an increased stability of the mRNA. TNF-a also increases mucin biosynthesis by increasing MUC5AC mRNA half-life (19), indicating that postranscriptional regulation of mucins is an important mechanism controlling mRNA levels in disease states. MUCIN GENES IN LABORATORY ANIMALS As a result of the molecular cloning of human mucin genes, the identification of corresponding animal MUC genes, such as those of the rat and mouse, is now an active area of research. The cloning of animal genes allows for easier manipulation, permitting elucidation of their regulation, function, and abnormalities in experimental models of respiratory disease. Rodents are frequently used in these models due to their small size and, apart from a lack of submucosal glands, the similarity of their airway surface epithelia to that of human airways.For the rat, homologs of human MUC 1-4 and 5 AC have been cloned (20-26). These rat mucin genes show a high degree of sequence similarity to their human counterparts. For example, the rat cDNA homolog of human MUC5AC has a sequence similarity of 73% at the amino acid level and 71% at the nucleotide level (25). The size and tissue distribution of rMUC5AC mRNA was found to be consistent with that of human MUC5AC. However, mRNA levels were found to be much lower in rat airways but became strongly expressed during mucous differentiation (25). The tissue distribution of rMUC2 is also similar to that of its human homolog in that it is strongly expressed in the small intestine and is dependent upon mucus hypersecretion (27).At present, the only murine MUC homologs reported are those for MUC1, 2, 3, and 5AC (28-31). The murine mucins have not yet been extensively studied. However, murine MUC5AC is mainly confined to the stomach, with no expression in the trachea or lung, as for rMUC5AC (30). The expression of both murine MUC2 and MUC3 is absent in the airways of pathogen free mice. MUC2 is confined to the intestinal goblet cells (29), whereas MUC3 is mainly present in the caecum and small intestine (31). GOBLET CELL HYPERPLASIA IN ASTHMA AND COPD One of the cardinal pathophysiological features of chronic bronchitis is goblet cell hyperplasia (32). The hyperplasia is in larger airways, whereas metaplasia occurs in the bronchioles in which goblet cells are normally scarce or absent. This observation has been repeated by others and is confirmed by semi-quantitative analysis (33). Goblet cell hyperplasia has also been noted in patients dying of asthma (34). Quantitative analysis demonstrates marked
goblet cell hyperplasia throughout the lower airways of patients dying of acute severe asthma but not of chronic asthma (35). These observations indicate that, at least for the latter group of patients, disproportionate goblet cell hyperplasia is associated with fatal attacks of asthma. From the above, goblet cell hyperplasia/metaplasia appear to be features of asthma and chronic bronchitis. However, there are exceptions. Goblet cell hyperplasia is not obvious in all patients with asthma or chronic bronchitis (36), and in a small group of Japanese patients with COPD quantitation did not demonstrate hyperplasia compared with controls (37). GOBLET CELL HYPERPLASIA IN EXPERIMENTAL ANIMALS
Goblet cell hyperplasia and metaplasia can be readily induced in a variety of laboratory animals by a variety of experimental procedures (38,39) (Table 1). Most of the “ early” (i.e., late 1960s, 1970s, and 1980s) models were primarily of chronic bronchitis, with goblet cell hyperplasia induced by inhaled gases, principally cigarette tobacco smoke, sulfur dioxide, nitrogen dioxide, or ozone. Elastase was also frequently used, although it tended to generate emphysematous changes rather than hypersecretory changes. The molecular and cellular mechanisms underlying the development of goblet cell hyperplasia in these COPD models were for the most part unexplored. More recently, models of allergy and asthma have been described, with mechanisms being delineated. These models are principally based on animals sensitized systemi-cally to ovalbumin and subsequently challenged with inhaled ovalbumin, usually repeatedly every few days for a number of weeks. The allergy/asthma models demonstrate goblet cell hyperplasia associated with recruitment of inflammatory cells (40). Th2 cells may orchestrate the inflammatory response, including hypersecretion (41), in concert with the cytokines IL-4 and IL-13 (42-44). The latter cytokines probably both act through the a chain of the IL-4 receptor (45). Tumor necrosis factor a (TNF-a) alone has little effect on goblet cell number but potentiates goblet cell hyperplasia induced by platelet activating factor (PAF) (46). TNF-a also upregulates receptors for epidermal growth factor (EGRF-R) leading to associated goblet cell hyperplasia (47). The hyperplasia is prevented by an EGF-R tyrosine kinase inhibitor (BIBX 1522) in both allergic rats (47) and rats with agarose plugs instilled into their bronchi (48). MUCIN GENE EXPRESSION IN RESPIRATORY DISEASE MODELS It is thought that the pathophysiology of mucus hypersecretion involves abnormalities that lead to an increase in the production of mucin. These abnormalities
may occur at the level of transcription or stabilization of mucin RNA transcripts or at the level of translation or protein stabilization. To help determine the association of mucin gene expression with specific airway diseases, mucus hypersecretion can be induced experimentally by the exposure of the respiratory tract to a variety of agents (Table 1).Jany et al. (27) were the first to report that induced hypersecretion may be controlled in part at the level of mucin mRNA. Sprague-Dawley rats infected with Sendai virus and exposed subacutely to S02 to induce experimental chronic bronchitis had detectable levels of MUC2 mRNA in their airways, unlike control
rats that were pathogen-free. Rats exposed to acrolein, an aldehyde found in high concentrations in tobacco smoke and that is absorbed in the upper respiratory tract, exhibit a pulmonary pathology similar to that of human bronchitis (49). The resulting mucus hypersecretion was in part due to an increase in the steady-state mRNA levels of MUC5AC in the trachea and lung and was accompanied by an increase in MUC5AC protein. In contrast, expression of MUC2 was not significantly different in the trachea or lung.Models of allergy/asthma are currently receiving considerable attention. For example, brown Norway rats sensitized systemically to ovalbumin and the airways subsequently challenged by inhaled ovalbumin exhibit the bronchial hyperresponsiveness and extensive airway remodeling characteristic of asthma (50). There is also evidence of goblet cell hyperplasia (Fig. 2). We found that part of the remodeling includes marked increases in expression of mRNA for MUC5AC (Fig. 3), with associated decreases in MUC1 (51). Similarly, mice sensitized and
subsequently challenged with ovalbumin, or treated with the cytokine IL-13, demonstrate induction of MUC5AC mRNA and protein expression in the lungs (44). The ovalbumin murine model has pathophysiological changes similar to those seen in human allergic asthma, including goblet cell metaplasia (40), and IL-13, a Th2 cytokine, has been found to have a critical role in the expression of murine asthma (52). In addition to MUC5AC, the expression of MUC1, 2, and 3 mRNA was also investigated. Neither IL-13 nor ovalbumin treatment altered the levels of MUC1 or induced the expression of MUC2 or 3, which supports the concept that differential regulation of mucin genes occurs in the respiratory tract.
