Cystic Fibrosis and Other Respiratory Diseases of Impaired Mucus Clearance

Abstract

Exposed to a diverse array of potentially noxious agents, the respiratory tract is protected by a highly developed innate defense system. Physiologically regulated epithelial ion and water transport coordinated with mucin secretion, beating cilia, and cough results in continuous flow of fluid and mucus over airway surfaces toward the larynx. This cleansing action is the initial and perhaps most quantitatively important innate defense mechanism. Repeated lung infections and eventual respiratory insufficiency characteristic of human cystic fibrosis (CF) and primary ciliary dyskinesia (PCD) illustrate the consequences of impaired mucus clearance. Altered mucus clearance likely contributes to the initiation, progression, and chronicity of other airway diseases characterized by inflammation and mucous secretory cell hyper/metaplasia that afflict millions worldwide, including chronic obstructive pulmonary disease (COPD). This review concisely discusses the pathophysiology of human diseases characterized by genetic defects that impair mucus clearance. It then explores animal models in which components of the mucus clearance system have been disrupted. These models firmly establish the importance of mucus clearance for respiratory health, and will help elucidate disease mechanisms and therapeutic strategies in CF, PCD and COPD.

Introduction

Respiratory tract diseases such as asthma, chronic obstructive pulmonary disease (COPD) and lung cancer typically result from complex interactions between the environment and host genetics. Constantly exposed to the environment, the airways and lungs have evolved to efficiently deliver inhaled air to the alveoli, while protecting gas exchange structures from potentially harmful airborne chemicals, particles, and pathogens. A layer of fluid and mucus, propelled by cilia beat and cough, continuously flows over conducting airway surfaces towards the pharynx. Mucus clearance is a critical, physiologically regulated, protective function of the airways and lungs.

The genetic defects in cystic fibrosis (CF) and primary ciliary dyskinesia (PCD) impair mucus clearance, and study of these diseases has provided key insights regarding the pathobiology that occurs when constant protection of the respiratory tract by effective mucus clearance is lost. In this review, we will concisely discuss the structural components and physiology of mucus clearance and the pathophysiology of CF and PCD in humans. We will then summarize animal models in which genes involved in the mucus clearance system have been disrupted or modified to model the human diseases.

Respiratory Tract Physical Structure, Innate and Acquired Host Defense

Integrated airway and lung defense is the product of a complex network, acting at distinct levels of organization (Figure 1). The nasal passages, nasopharynx and pharynx constitute the upper airways and the trachea to ~6th generation bronchi are the proximal, large lower airways. Portions of the nasal cavity and all the large airways are normally lined by a pseudostratified mucociliary epithelium consisting principally of basal, intermediate, ciliated and mucous secretory (goblet) cells. The proximal portion of the small, distal airways (bronchioles) is also lined by a columnar pseudostratified epithelium, which transitions in more distal airways to a shorter, columnar to cuboidal epithelium in which Clara cells replace mucous secretory cells.

Finally, the alveolar region is composed mainly of thin type I, and cuboidal type II alveolar epithelial cells. Thus, a continuous epithelial layer forms a vital physical barrier between the outside environment and the underlying lung parenchyma and circulatory system. Just as the relative abundance of cell types changes along the respiratory tract axis, each segment has specialized functions that are key to host defense. The epithelial physical barrier, in combination with secreted antimicrobial factors, broad specificity soluble binding proteins (opsonins), mucus, mucociliary clearance, cough, phagocytic and natural killer cells constitutes the innate immune system (Martin and Frevert, 2005; Zaas and Schwartz, 2005). This system is constantly on guard and requires no previous exposure to elicit a protective response.

To eliminate noxious stimuli, enhanced mucus and fluid secretion, cilia beat, antimicrobials, cytokines, chemokines and inflammatory cells are induced when pathogen associated patterns or other chemical motifs are recognized by innate immune system receptors. The cell and cytokine/chemokine milieu induced by the innate immune system is critical for the involvement and development of the more specialized adaptive immune response; for example, specific memory and effector T and B cells. Mucous secretory and ciliated cells play key roles in mucociliary clearance, which in conjunction with cough, provides mucus clearance, perhaps the quantitatively most important airway innate defense mechanism (Knowles and Boucher, 2002; Randell and Boucher, 2006).

Mucus, Mucous Secretory Cells, and Mucins

Mucus is the collective term for mucins, ions, water and other substances normally present on mucosal surfaces. By virtue of highly glycosylated secreted mucins, lipids, and soluble proteins, mucus binds and entraps a broad array of inhaled particles. Since increased airway secretions characterize obstructive lung diseases, including asthma, COPD and CF, particular attention has been paid to the regulation of mucous secretory cell number, mucin gene expression and glycoprotein production, and the secretory pathways that mediate mucin release (Thornton and Sheehan, 2004; Rose and Voynow, 2006; Williams et al., 2006). Beyond obstruction of airflow, excessive mucus that fails to clear may form plaques and plugs that can serve as a nidus for infection by opportunistic pathogens, especially facultative anaerobic bacteria such as P. aeruginosa, able to adapt to relatively hypoxic conditions.

Often, single cell, free-living bacteria (planktonic) evolve into alginate-producing colonies or biofilms that are much more resistant to attack from both neutrophils and antibiotics, thus becoming difficult to eradicate (Worlitzsch et al., 2002; Matsui et al., 2005). The robust inflammatory response triggered by recurrent, or chronic infection can lead to severe remodeling of the airways perpetuating the vicious cycle initiated by mucus accumulation.

Ciliated Cells

At least 8 categories of cilia, or cilia derived organelles, with different functions have been described in the human body (Afzelius, 2004). In the respiratory tract, ciliated cells have ~200 cilia per cell, each with a diameter of 250 nm and a length of ~6 μm. There are an estimated 10⁹ cilia/cm² of upper and large airway surface. Cilia in the upper and large airways are denser and longer than in the bronchioles. Each respiratory cilia axoneme has a typical 9 + 2 microtubule structure (9 outer doublets and 2 central single microtubules) with characteristic inner and outer dynein arms, nexin links, and radial spoke apparatus. Each axoneme inserts into the ciliated cell at a specialized structure that terminates in a cytoplasmic basal body that resembles the centriole of the mitotic apparatus (Figure 2). It is estimated that >200 unique proteins constitute the cilia and basal body structures (Fliegauf and Omran, 2006). In normal lungs, cilia beat in coordinated, directional, metachronal waves at a frequency estimated to be ~10–20 Hz, and mucus and the embedded particles are constantly transported towards the pharynx to be swallowed or expectorated, thereby promoting airway sterility (Meeks and Bush, 2000).

The Peri-Ciliary Layer (PCL), Glands, and Cough

Together with cilia and mucus, an adequate peri-ciliary layer (PCL) is required for effective mucociliary clearance. The PCL lines the airway surface and provides a low viscosity milieu in which cilia can freely beat to propel overlying mucus. The existence of the PCL has been demonstrated in vivo by osmium/perfluorocarbon fixation techniques followed by transmission electron microscopy (Sims et al., 1991) or in vitro by confocal microscopy (Matsui et al., 1998; Tarran and Boucher, 2002) (Figure 3). The ability of the airway epithelium to regulate PCL volume to maintain an ~7 μm-thick layer required for cilia to beat and effectively clear mucus is critical for lung innate defense (Matsui et al., 1998; Tarran et al., 2001, 2005).

As discussed in the legend of Figure 1, recent studies suggest that PCL volume homeostasis represents the coordinated balancing of Cl⁻ secretion and Na⁺ absorption and is sensitive to changes in the concentration of soluble mediators (nucleotides and nucleosides) present in the PCL (Lazarowski et al., 2004; Tarran, 2004, 2006a, 2006b). Upper and proximal airways also contain submucosal glands that contribute to airway secretions and amplify the ability of airway epithelial cells to produce antimicrobial factors and mucus (Wine and Joo, 2004; Inglis and Wilson, 2005). Finally, cough is an important innate defense mechanism that prevents pulmonary aspiration, promotes ciliary activity and clears airway debris (Chang, 2006; McCool, 2006). The combination of mucociliary and cough clearance constitutes “mucus clearance.”

Cystic Fibrosis (CF)

In the following section, we highlight the seminal characteristics of CF. For a more a comprehensive treatise on the history, genetics, diagnosis, pathogenesis, microbiology, treatment, or controversies in this intensely studied disorder, the reader is directed to an excellent recent textbook chapter (Boucher, 2005) and review articles (Davis, 2006; Rowe et al., 2005). CF is the most common life-threatening genetic disease of Caucasians, affecting ~1/2500 and ~1/17,000 live births in the White and Black populations of the United States, respectively. It has been recognized as a distinct disease entity since the 1930s. It is an autosomal recessive, monogenic disorder resulting from mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which was discovered in 1989 (Online Mendelian Inheritance in Man (OMIM) # 602421, National Center for Biotechnology Information).

The CFTR protein is a cAMP-activated transmembrane anion channel expressed at a variety of epithelial surfaces, where it is thought to principally conduct Cl⁻ ions. However, CFTR also coordinately regulates the function of other epithelial ion channels located in the apical membrane. There are over 1400 different recognized CFTR mutations but the majority, from ~80% to >95%, depending on the patient ethnicity, can be detected in a screen of 97 of the most common mutations. CFTR mutations include deletions (the deletion of 3 base pairs resulting in the loss of phenylalanine at position 508, known as ΔF508, is the most common CF mutation), missense, frameshift, nonsense, and also mutations causing alterations in RNA splicing.

Acting through a variety of molecular and cellular mechanisms, severe mutations result in an almost total absence of functional CFTR protein at the cell surface. Milder mutations resulting in partial CFTR function may produce abnormalities in other organs but respiratory disease may be mild or even absent. CF is a multi-system disorder with a childhood onset. The defining feature is abnormal ion and water transport across a variety of epithelial surfaces, which leads to the characteristic salty sweat and elevated nasal and rectal trans-epithelial electrical potential differences (PD) that can be measured for diagnostic purposes. Relatively dehydrated, thickened secretions and mucus accumulation/obstruction of several tubular duct structures are characteristic of CF.

Significant morbidities include obstructive azoospermia in males due to in utero blockage, subsequent fibrotic obliteration and total loss of the vas deferens, epididymis and seminal vesicles. In the female, the uterine cervix can be inappropriately obstructed with thickened mucus. Other morbidities are principally associated with the gastrointestinal tract. Meconium ileus in the newborn is highly suggestive of CF and recurrent bouts of intestinal obstruction are common. Gastrointestinal changes and frequent coughing increases the frequency of rectal prolapse. Biliary tract obstruction variably results in liver dysfunction and fibrosis. Pancreatic involvement is highly prevalent and obstruction begins in utero, which serves as the basis for the blood trypsinogen neonatal screening test.

Exocrine pancreatic insufficiency and duct blockage decrease digestive enzyme secretion into the intestine and can result in significant nutritional deficiencies and steatorrhea if not properly treated. Ultimately, postobstructive atrophy results in complete replacement of the exocrine pancreas by fibrosis and fatty tissue, with eventual disruption of the islets, which can cause insulin dependent hyperglycemia. Progressive pancreatic lesions are illustrated in Figure 4.

Neonatal respiratory distress is not specifically associated with CF and, except for dilated mucus gland ducts, the lungs are apparently normal at birth. Postnatal respiratory tract lesions are, by far, the leading cause of morbidity and mortality in CF. In the upper airways, rhinitis and sinusitis due to infection are highly prevalent and development of nasal polyposis is common. Recurrent bacterial bronchopneumonia, often initially due to H. influenzae or S. aureus and then P. aeruginosa, is frequent in newborns and young children with CF. Evidence for lung infection with P. aeruginosa occurs in ~60% of CF individuals by age 3, which is likely to be an underestimate (Burns et al., 2001; West et al., 2002). Repeated and ultimately chronic infection with mucoid P. aeruginosa, or other characteristic Gram-negative bacteria, including Burkholderia sp. (previously known as Pseudomonas cepacia), S. maltophilia, A. xylosoxidans, as well as variable involvement of Aspergillus and nontuberculous mycobacteria, leads to progressive bronchiectasis, bronchiolectasis and respiratory insufficiency that ultimately causes respiratory failure and death of >95% of CF patients, currently with a median survival age of ~37 years.

The sequence of events predisposing to airway infection in CF has been debated through the years, but clarified more recently. Generalized immune deficiency and specific abnormalities in acquired immunity are highly unlikely, since systemic infection is not characteristic of CF. In fact, sepsis due to P. aeruginosa, even after decades of lung infection, is rare, presumably due to effective humoral immunity. More likely, the disease represents a failure of local airway defense. It has been suggested that bacteria adhere more readily to CF airway epithelial cells due to enhanced expression of the cell surface ganglioside asialoGM1, promoting infection (Saiman and Prince, 1993). Paradoxically, it has also been proposed that CFTR serves as a bacterial receptor and that its absence leads to failure to internalize and kill bacteria (Pier, 2000). Defects in anti-microbial activity of airway fluid (Ganz, 2002) or in neutrophil phagocytosis (Berger et al., 1989) likely occur in the inflamed CF airway environment but are unlikely to be the primary defect. The presence of an intrinsic hyper-inflammatory phenotype in CF cells, even in the absence of infection, is debated (Machen, 2006). A consensus now exists that respiratory tract pathophysiology in CF principally results from the inability to secrete Cl⁻ and regulate Na⁺ absorption, which causes relative dehydration of the airway surface, depleting the PCL and causing accumulation of hyper-viscous mucus that cannot be cleared by mucociliary clearance or cough (Matsui et al., 1997).

As discussed previously and as illustrated in Figure 5, mucus plaques and plugs serve as a nidus for intra-luminal infection. Interestingly, hypoxic microenvironments, which are exploited by characteristic CF pathogens, develop in macroscopic mucus accumulations, even within ventilated airways (Worlitzsch et al., 2002). Bacteria resident within the thickened luminal mucus may evade chemical antimicrobial factors and phagocytes (Matsui et al., 2005). Complex bacterial evolution and host adaptation occurs in the chronically infected airway, which is likely unique in CF due to the constant and severe degree of mucus dehydration and impaired mucus clearance. Bacterial colonies exhibiting biofilm-like properties may develop, which are difficult or impossible to eradicate. The continuous presence of bacteria and the accompanying intense inflammation ultimately remodel the airway wall, causing the ubiquitous mucous secretory cell hyperplasia and metaplasia, submucosal gland enlargement, hypertrophy of the bronchial circulation, ectasis of bronchi and bronchioles, and variable parenchymal cyst formation, sometimes progressing to cavitary disease, with adjacent fibrosis and pleural involvement.

Primary Ciliary Dyskinesia (PCD)

PCD illustrates the importance of normally beating cilia to maintain healthy airways. In recent years, there has been impressive progress to elucidate the role of cilia in diverse biological processes. It has been demonstrated that motile, rotating, 9 + 0 cilia at the embryonic node function to break bilateral symmetry in the developing mammalian embryo and establish sidedness of the body plan (Hirokawa et al., 2006). A role for immotile 9 + 0 cilia as sensing antennae and cell signaling platforms in various cells has been suggested by recent studies (Davenport and Yoder, 2005; Scholey and Anderson, 2006; Singla and Reiter, 2006). Forms of polycystic kidney disease and retinitis pigmentosa are associated with gene products expressed in renal 9 + 0 monocilia and in the connecting cilium of retinal photoreceptor cells, respectively (Eley et al., 2005). As in other aspects of cilia biology, there have been recent strides in our understanding of PCD. In the United States, the Genetic Diseases of Mucociliary Clearance Consortium 〈http://rarediseasesnetwork.epi.usf.edu/gdmcc/index.htm〉, a clinical research network sponsored by the National Institutes of Health Office of Rare Diseases, includes PCD as a disease target. Furthermore, a patient information and advocacy group has been established 〈http://www.pcdfoundation.org〉. In this section we concisely discuss the consequences of defects in motile 9 + 2 cilia, focusing on the respiratory tract, and refer the reader to excellent recent articles and reviews for more in-depth discussions and clinical information (Meeks and Bush, 2000; Afzelius, 2004; Geremek and Witt, 2004; Noone et al., 2004; Van’s Gravesande and Omran, 2005; Badano et al., 2006; Stannard and O’Callaghan, 2006).

PCD is less common than CF, with an estimated incidence of 1 in 15,000 to 30,000 live births (OMIM #242650). It is a genetically heterogeneous disorder that is usually autosomal recessive. Due to multiple potential defective genes, wide variability in symptoms, and similarity to acquired abnormalities of cilia due to infection and inflammation, the true incidence of PCD may be greater. PCD is more frequent in certain isolated populations and in consanguineous families. In the 1930s, 4 cases of individuals with the triad of situs inversus, chronic rhinosinusitis, and bronchiectasis were reported by Kartagener (1936) and in the 1970s it was recognized that this syndrome was correlated with impaired motility of respiratory tract cilia and spermatozoa in affected males (Afzelius, 1976).

We now know that Kartagener Syndrome is a subset (~50%) of PCD, because the abnormalities affecting respiratory cilia also impair the motile 9 + 0 cilia in the embryonic node, resulting in the randomization of sidedness during embryonic development. Randomization of laterality is illustrated by the presence of situs solitus (normal) and situs inversus totalis (mirror image of the thoracic and abdominal organs) in monozygous twins, both of which have PCD (Noone et al., 1999). Like CF, PCD usually presents as an autosomal recessive disorder in affected families, but X-linked cases of retinitis pigmentosa with respiratory tract symptoms identical to PCD have been reported, and rarely, other forms of inheritance.

The respiratory tract pathophysiolgy of PCD is attributed to abnormal function of motile 9 + 2 cilia. The functional changes can vary from subtle differences in beat frequency, direction, strength or coordination (dysmotility) to complete immotility and, rarely, complete absence of cilia (aplasia). Direct microscopic examination of cilia beating patterns in nasal and/or bronchial brushing or curettage specimens is a useful clinical test for PCD, although some PCD patients may demonstrate subtle changes. Acquired abnormalities due to infection, allergic inflammation, pollution or smoking, which may impair cilia motility (Calderon-Garciduenas et al., 2001), may be present in non-PCD patients. The morphology of cilia in routine histological sections can be suggestive but is not usually diagnostic (Figure 6).

Ultrastructural abnormalities can be confirmed by cross-section transmission electron microscopic (TEM) analysis of the cilia axoneme, but again, acquired abnormalities may result in false positives. The structural changes identifiable by TEM fall into distinct categories, including absence of inner or outer dynein arms, radial spoke or microtubule defects, or other rare alterations. However, dysmotile cilia from patients with PCD may appear normal by TEM. Extended in vitro culturing of respiratory epithelial cells at an air-liquid interface, enabling maintenance of the ciliated cell phenotype, removes confounding in vivo influences, thereby eliminating secondary abnormalities and likely improving the specificity of the functional and TEM tests (Jorissen et al., 2000).

For unknown reasons, nasal nitric oxide (NO) production is typically very low in PCD patients (Noone et al., 2004). While this feature can be highly useful, it is not totally specific and there may be overlap in nasal NO levels between PCD and CF or other allergic or infectious rhinosinusal diseases. Due to the wide range of symptom severity and the potential non-specificity of clinical tests, the diagnosis of PCD can be difficult.

As noted before, it is estimated that >200 unique proteins constitute the cilia and basal body apparatus (Fliegauf and Omran, 2006) and mutations in any of these genes, or even in non-cilia genes responsible for cilia assembly or maintenance may cause PCD. The model di-flagellated unicellular organism Chlamydomonas reinhardtii, is amenable to genetic manipulation and analysis of structural-functional correlation, and is a very useful tool to study homologous genes and proteins present in mammalian cilia (Silflow and Lefebvre, 2001).

Genome-wide linkage studies in human PCD have not been highly useful to elucidate specific mutations, but instead confirm genetic heterogeneity, identifying several potential loci in different families. A large proportion of PCD is characterized by absent or reduced outer dynein arms and the study of C. reinhardtii has identified mutations causing similar structural and functional defects. This observation pointed towards human orthologs of C. reinhardtii genes as candidate genes for PCD. Mutations in DNAI1 and DNAH5, the human orthologs of the C. reinhardtii IC78 and dynein γ-heavy chain genes, respectively, have been identified as causing PCD (Noone et al., 2002; Olbrich et al., 2002; Fliegauf et al., 2005). It has also been reported that the gene DNAH11, which is homologous to the gene mutated in the Ird mouse model of situs inversus (Supp et al., 1999), is mutated in a patient with situs inversus totalis and immotile cilia (Bartoloni et al., 2002), but this patient is also homozygous for ΔF508 CFTR, and it is unclear whether the individual truly had PCD or had acquired cilia defects due to CF and a concomitant situs abnormality, or possibly, the DNAH11 mutation caused situs inversus but not PCD (Van’s Gravesande and Omran, 2005).

Recent more detailed studies of PCD families have better defined the spectrum and frequency of DNAI1 and DNAH5 gene mutations (Hornef et al., 2006; Zariwala et al., 2006). These studies are important because they suggest the future applicability of novel clinical tests for PCD. First, antibody detectable absence or abnormal localization of mutant cilia proteins may enable novel tests on nasal or bronchial cells (Fliegauf and Omran, 2006). Second, the combination of DNAI1 and DNAH5 mutations likely define ~40% of all known cases of PCD, potentially ushering in an era of genetic testing for this underappreciated and potentially difficult to diagnose disease (Bush and Ferkol, 2006). Accurate and timely diagnosis is key, since aggressive and appropriate therapy may delay the ultimate development of severe lung disease.

The pathogenesis of respiratory tract disease in PCD has been generally thought of as being similar to CF, but somewhat milder in general, and with unique features. There is an interesting association of respiratory distress in term neonates with PCD but not CF, apparently due to a failure to rapidly and fully transition to air breathing at birth. The mechanism is not well understood but may involve a role for cilia action in removal of fetal lung liquid. Chronic infectious rhinosinusitis is prominent in both PCD and CF, but childhood otitis media is much more common in PCD, and patients frequently seek initial treatment at the otolaryngology clinic.

Incidence of lung infection, offending organisms, development of bronchiectasis and longitudinal declines in lung function are similar to CF but appear to be delayed, and serious lung disease tends to develop later in life. The care of PCD patients is generally modeled on CF treatment paradigms. A factor possibly explaining the “PCD vs. CF delay” is that ion transport mechanisms regulating airway surface hydration are preserved in PCD. Thus, airway secretions are less dehydrated than in CF, and the PCL does not become as severely reduced, still allowing for effective cough clearance. Although the progression of airway disease may be delayed, the physiologic impairment due to progressive mucus accumulation, infection and lung pathology in PCD can become severe and similar to end stage CF (Figure 7), and require lung transplantation as a final therapeutic option.

There are unique non-pulmonary manifestations of PCD that are uncommon in CF. Male infertility is common to both CF and PCD. However, men with PCD are not azoospermic, but exhibit sperm tail dysmotility. As noted above, situs inversus totalis, namely a mirror image of the thoracic and abdominal organs, affects ~50% of PCD patients and is not seen in CF. A significant percentage of PCD patients, on the order of 7%, appear to have heterotaxy, also known as situs ambiguous, which represents a range of abnormalities between situs solitus (normal) and situs inversus totalis that may include isolated or paired reversals, midline organ distribution, aspleenia or polyspleenia. About half of the individuals with PCD and heterotaxy also have congenital heart defects (Personal communication, Dr. Michael R. Knowles, The University of North Carolina). If substantiated, it will be informative to explore the mechanistic links between genes causing both PCD and congenital heart disease.

Hydrocephalus is a rare finding, inconsistent even within affected families, but is definitely associated with PCD. Hydrocephalus frequently occurs in mouse models of PCD (see below). The exact mechanism is unknown but ependymal cells with motile 9 + 2 cilia lining the ventricles may participate in cerebrospinal fluid flow, especially at narrow passages, which may become damaged, obstructed and ultimately fibrotic in the absence of effective cilia beat.

Role of Impaired Mucus Clearance in COPD?

COPD is a leading cause of disability and death worldwide. Like CF and PCD, submucosal gland enlargement, mucous secretory cell hyperplasia in the large airways and metaplasia in the small airways and sputum production are common features of COPD (Szilasi et al., 2006). Thickening of the bronchiolar wall due to fibrosis, an increased volume of epithelial cells and the presence of luminal inflammatory mucus exudates are correlated with the severity of COPD (Hogg et al., 2004). Toxic particles and gases, the main cause of COPD, inhibit cilia function and Cl⁻ secretion (Sisson et al., 1994; Kreindler et al., 2005; Cantin et al., 2006), and emphysema and loss of small airways tethers may distort and/or promote collapse of the airway wall, all of which would decrease the efficiency of mucociliary and cough clearance. Bacterial infection and exacerbation by viral illness are prominent features of COPD, and this sequence is also important in CF and PCD. Thus, reduced efficiency of mucus clearance likely plays an important role in the pathogenesis of COPD. A more detailed comparison of mucus clearance in CF and COPD is available in a recent review (Randell and Boucher, 2006).

Animal Models of CF

The cloning of the human CFTR gene pointed the CF research community towards generation of an animal model, in which the gene could be either deleted or mutated, to enable study of the pathophysiology and potential treatments. To date, both knockout mice and those harboring specific mutations in murine Cftr, notably ΔF508, have been created. A brief summary of CF mouse genotype and phenotype, adapted from the European working group on CFTR expression web site 〈http://pen2.igc.gulbenkian.pt/cftr/vr/f/scholte mouse models table.pdf〉, is given in Table 1. Detailed reviews of the pathophysiological features of CF mice have been published (Grubb and Boucher, 1999; Guilbault et al., 2006). Briefly, a consistent phenotype is exhibited in the gastrointestinal tract of CF mice. Untreated mice typically die shortly after birth due to meconium ileus or after weaning from intestinal obstruction/rupture caused by impaired luminal fluid secretion, which correlates with the absence of functional CFTR protein in the gut epithelium and resembles the human CF intestinal phenotype. The incidence of death due to intestinal obstruction can be reduced by a liquid diet or by supplementing the drinking water with a laxative (Colyte).

An unexpected finding was the lack of an overt pulmonary phenotype. All of the CF mice generated exhibited inconsistent or minor pathological alterations in the lungs. In Cftr ^tm1Unc (knockout) mice, the nasal septal epithelium exhibited hyperplastic and hypertrophic mucous secretory cells (Snouwaert et al., 1992) in conjunction with reduced PCL volume, in comparison to wild-type littermates (Tarran et al., 2001). When Cftr ^tm1Unc mice were backcrossed onto the C57BL/6 strain, patchy alveolar distension, interstitial thickening, early mild neutrophilic infiltrate at day 30, an increase in monocytes at 3 months, and fibrosis were reported (Kent et al., 1997; Durie et al., 2004;). G551D CF mice exhibited inspissated eosinophilic material in the lumen of the pharyngeal submucosal glands (Delaney et al., 1996).

A number of hypotheses have been formulated to explain minimal pulmonary pathology in CF mice. Except for the proximal trachea, murine airways are devoid of submucosal glands (Pack et al., 1981), which may be an important difference from humans, but is not consistent with the presence of disease in gland-free human distal bronchioles. As discussed previously, human CF is characterized by both the absence of cAMP-stimulated, CFTR-mediated Cl⁻ secretion and Na⁺ hyperabsorption, due to unregulated ENaC activity in the upper and lower airways. Short circuit current (I_sc), as measured using Ussing chambers, is an index of ion transport across the epithelium.

In excised mouse nasal epithelium virtually all of the basal I_sc is inhibitable by amiloride (ENaC-mediated) and in Cftr ^tm1Unc mice amiloride-sensitive I_sc is elevated indicating sodium hyperabsorption (Clarke et al., 1992). This mimics the higher nasal PD found in CF humans, which is also thought to be due to up-regulation of ENaC activity. However, in the mouse tracheal and bronchial epithelium, the amiloride-sensitive I_sc accounts for only part of the basal I_sc. Tracheas from both wild-type and Cftr ^tm1Unc mice respond with significant Cl⁻ secretion when stimulated with forskolin, typically used to activate adenyl cyclase, increase cellular cAMP and activate CFTR. However, in freshly excised mouse tracheas, forskolin also elevates intracellular Ca²⁺, activating an alternative (non-CFTR,) Cl⁻ channel (Grubb et al., 1994).

Therefore, CFTR seems to have a small role in the mouse lower airways where it is replaced by prominent activity of alternative, Ca²⁺-activated Cl⁻ channels. Thus, minimal spontaneous lung pathology in CF mice is most likely due to regional, species-specific differences in ion transport between mice and humans. There is an active research effort to create CF in larger animals that may be physiologically more similar to humans, including ferrets, pigs and sheep (Scholte et al., 2004; Li et al., 2006).

Scnn1b Transgenic Mice

Based on the hypothesis that hydration of the airway surface represents the balance between Cl⁻ secretion and Na⁺ absorption, mice that hyper-absorb Na⁺ in the lower airways were created by overexpressing the β subunit of ENaC (encoded by the Scnn1b gene) using the airway-specific Clara cell secretory protein (CCSP) promoter (Mall et al., 2004). The tracheas of neonatal and adult Scnn1b-transgenic mice exhibited elevated basal and amiloride-sensitive I_sc, compared to nontransgenic littermates. Accordingly, tracheal and bronchial PCL height was significantly reduced, the percent solids content of lower airway mucus was significantly increased and mucus transport, measured in vivo, was significantly reduced in comparison to wild-type littermates.

Scnn1b overexpression did not have adverse effects on fetal survival, but Scnn1b mice showed a significant postnatal mortality that began in the first days of life and continued up to 4 weeks, with an overall mortality of ~50%. Scnn1b mice exhibited postnatal mucus accumulation and obstruction in the large and small airways associated with goblet cell metaplasia and neutrophil infiltration (Figure 8). The neutrophil-attracting chemokines macrophage inflammatory protein-2 (MIP-2) and KC (the two murine analogues of human IL-8) were significantly increased. Lungs were free of cultureable bacteria, indicating that the neutrophilic inflammation occurred in the absence of active infection. Interestingly, the upregulation of MIP-2 and KC was not a direct effect of Scnn1b overexpression in the epithelial cells, since their secretion was not increased in epithelial cultures derived from Scnn1b airways.

It was proposed that inflammation occurs because environmental inflammatory stimuli accumulate in the lungs of Scnn1b mice due to poor mucus clearance. Finally, H. influenzae or P. aeruginosa instilled into the trachea persisted longer in Scnn1b mice, indicating impaired airway clearance. In summary, the Scnn1b mice exhibit airway Na⁺ hyper-absorption, a hallmark of human CF, and recapitulate many features of the human disease, including mucus accumulation and neutrophilic airway inflammation. This mouse is thus a useful in vivo tool for investigating pathophysiologic mechanisms relevant to CF and other diseases.

Animal Models of PCD

A phenotype similar to PCD occurs naturally in both dogs and pigs, which appears identical to that seen in humans, except that hydrocephalus is a prominent feature (Edwards et al., 1989; Roperto et al., 1993). The genetic basis of canine and porcine PCD-like airway disease remains unknown. Numerous mouse models with left-right symmetry defects have been identified and are the subject of various reviews (Ibanez-Tallon et al., 2003; Eley et al., 2005), but due to the lack of respiratory tract symptoms these are not classified as PCD models (Geremek and Witt, 2004). Table 2 summarizes the mouse models with known genetic defects affecting 9 + 2 motile cilia function.

Mouse axonemal dynein heavy chain 5, encoded by the Mdnah5 gene, is the homologue of human DNAH5. Mdnah5-deficient mice exhibit early mortality (~4–5 weeks), hydrocephalus, respiratory cilia immotility due to the complete lack of outer dynein arms, randomization of sidedness, and a PCD-like respiratory phenotype characterized by early development of chronic rhinitis (Ibanez-Tallon et al., 2002). Targeted mutation of the mouse homologue of human DNAI1, Mdhc7, which encodes for an inner dynein arm heavy chain results in reduced sperm motility and male infertility. These mice show a 50% reduction of the respiratory ciliary beat frequency but apparently this does not cause airway disease (Neesen et al., 2001; Vernon et al., 2005; Woolley et al., 2005). Other candidate genes for PCD include transcription factors implicated in ciliated cell differentiation. Hfh4 (also known as Foxj1) is a transcription factor that regulates dynein gene expression and Hfh4-deficient mice completely lack epithelial cell cilia and may be a model for cilia aplasia. These mice exhibit hydrocephalus and randomization of sidedness, but still have monocilia in the embryonic nodal pole (Chen et al., 1998; Brody et al., 2000). However, mutations in the FOXJ1 gene have not been found in PCD patients, generally excluding it as a candidate gene for PCD (Maiti et al., 2000).

It was reported that deletion of DNA polymerase λ gene (Poll) in mice produced inner dynein arm defects in ependymal and respiratory cilia and a phenotype characteristic of PCD, including hydrocephalus, situs inversus, chronic suppurative sinusitis, and male sterility (Kobayashi et al., 2002). However, it was difficult to reconcile a mutation in a DNA polymerase with an axonemal defect and there was evidence that mice with deletion of the sequence encoding for the catalytic domain of Poll were viable and fertile (Bertocci et al., 2002). Thus, the targeting construct used to generate Poll−/− mice was examined in further detail.

The region of chromosome 19 that was deleted to generate the Poll−/− mouse also included the first exon and the start codon of a newly identified gene, Dpcd, which is transcribed from the opposite strand relative to Poll. Dpcd is predicted to encode for a 23 KDa protein of unknown function. Since Dpcd mRNA is expressed in airway epithelial cells and its expression is upregulated during ciliogenesis, it has been suggested that mutation in Dpcd can cause PCD, although thus far no mutations in the human homolog of Dpcd have been found in 51 unrelated PCD patients (Zariwala et al., 2004).

Collectively, the existing animal models of PCD are characterized by a high incidence of hydrocephalus and perinatal mortality. These pathologic features hamper the establishment of models suitable for long-term studies of chronic infection characteristic of the human pulmonary PCD phenotype. Nevertheless, mouse models where a homolog of a candidate human gene has been targeted can confirm the likely role of that gene in PCD, and novel animal models may also reveal important differences in the physiology of the mucociliary clearance system between mice and humans. It is possible to genetically manipulate mice to alter candidate PCD genes only in the trachea and lungs, and future development of such mice may avoid hydrocephalus, enabling more in-depth study of the lower respiratory tract.

Summary and Conclusion

Mucus clearance, the continuous flow of fluid and mucus over airway surfaces towards the larynx, is a vital protective, innate defense mechanism. Failure of this system, due to genetic defects in specific components, occurs in CF and PCD and results in repeated lung infections and eventual respiratory insufficiency. These diseases illustrate key principles underlying efficient mucus clearance and the pathophysiologic consequences of its impairment. Altered mucus clearance likely plays a role in several common acquired respiratory tract diseases, including COPD. A greater understanding of the basic science of mucus clearance and new animal models will promote advances in the diagnosis and treatment of this important class of diseases.