Published ahead of print on January 31, 2008, doi:10.1165/rcmb.2007-0284OC
© 2008 American Thoracic Society DOI: 10.1165/rcmb.2007-0284OC Nasal Abnormalities in Cystic Fibrosis Mice Independent of Infection and Inflammation1 Department of Paediatric Respiratory Medicine, Royal Brompton Hospital, London, United Kingdom; 2 Department of Gene Therapy, and 3 Department of Lung Pathology, National Heart and Lung Institute, Imperial College, London, United Kingdom; and 4 Institute for Molecular Bioscience, University of Queensland, St. Lucia, Queensland, Australia Correspondence and requests for reprints should be addressed to Jane C. Davies, Department of Gene Therapy, Emmanuel Kaye Building, 1b Manresa Road, London SW3 6LR, UK. E-mail: j.c.davies{at}imperial.ac.uk
It is not known whether the progressive airway changes in cystic fibrosis (CF) are all secondary to infection and inflammation. The CF mouse nose shares electrophysiologic and cellular properties with human CF airway epithelium. In the present work, we tested the hypothesis that structural abnormalities in the nasal mucosa of CF mice develop independent of infection and inflammation. We performed nasal lavage and subsequent serial coronal section through the nasal tissue of adult CF (mutations CftrTgHm1G551D and Cftrtm1Unc-TgN(FABPCFTR)) and wild-type mice raised under normal housing conditions. Nasal tissue was also obtained from Day 17 embryos and newborn pups. Detailed histologic examination of the respiratory and olfactory epithelium within the nasal cavity was performed. Bacterial culture, cell count, and macrophage inflammatory protein-2 (MIP-2) concentration were assessed in nasal lavage fluid. Significantly thickened respiratory epithelium and increased mucous cell density was found in adult CF mice of both mutations compared with wild-type animals. In contrast, the olfactory epithelium was thinner, with a decreased cell density. Areas of lymphoid aggregates were found in CF mice but not in non-CF mice. There were no differences in bacterial growth, cell count, or MIP-2 concentrations. No genotype differences were observed in the embryonic or newborn periods. There are significant histologic changes in the nasal mucosa of adult CF mice, not associated with increased lumenal inflammation or bacterial content, and which are not present perinatally. These may be novel therapeutic targets.
Key Words: mice • cystic fibrosis • pathology • inflammation • cystic fibrosis transmembrane conductance regulator
Cystic fibrosis (CF) is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene (1). The airways are essentially normal at birth, but develop chronic bacterial infection, inflammation, airway obstruction, and subsequent bronchiectasis (2). At the time of death, there is extensive airway destruction, but little is known about the early structural airway wall changes (remodeling), or their relationship to infection and inflammation. This is important, because if remodeling arises spontaneously, as a direct result of CFTR dysfunction, this may offer a novel therapeutic target. A number of CF mouse models have been generated and assessed (3). However, CF mice show little sign of spontaneous pulmonary disease (4–8). Whereas Pseudomonas aeruginosa infection and heightened pulmonary inflammation can occur in young infants (2, 9), CF mice require direct instillation of bacteria to establish a similar picture (10–12). The characteristic electrophysiologic properties of human CF respiratory epithelium are not displayed in the lower airways of CF mice (3, 5). However, the CF mouse nose shares a similar electrophysiologic profile to the human CF nose and lower airways (5, 13). Moreover, its cell content is similar to the human nose (14, 15). Consequently, the mouse nose has been used in the assessment of new therapies (16). There has, however, been only limited investigation of the histology of this tissue in the CF mouse, with one report of an increased number of goblet cells within its epithelium (17). In contrast, the wild-type mouse nose has been well characterized (18), with squamous epithelium in the anterior part, followed by ciliated respiratory epithelium (RE) lining the mid-portion, and then a thick layer of neuroepithelial olfactory epithelium (OE) over the dorsoposterior regions of the nasal cavity. Previous studies have used serial sections of the mouse nose to investigate the pathophysiology of rhinosinusitis (19, 20). In the present study, we used a histologic method that preserved nasal anatomy and epithelial structure, and established a novel technique of nasal lavage, to look for abnormalities in the nasal epithelium and investigate their relationship with inflammation and infection within the nasal cavity. In the present work, we test the hypothesis that there are structural abnormalities in the nasal mucosa of CF mice independent of infection and inflammation. Some of the results of this study have been previously reported in the form of abstracts (21, 22).
See the online supplement for more details regarding methods.
Adult Animals
Processing of Tissue
Histologic Analysis
Nasal Lavage
Pup and Embryo Investigation
Statistical Analysis
Histologic Analysis in Adult Mice A coronal section through a wild-type head at 4,200 µm is shown in Figure 1. Heads from CF mice tended to be smaller then wild-type animals; the lateral sinus in G551D mice closed at a median depth of 5,000 µm compared with wild type at 5,800 µm (P < 0.001); and 5,000 µm in FABP (P = 0.053 versus wild type). Despite this, the starting points of RE and OE were not significantly different between CF and wild-type mice (further details are given in the online supplement).
Respiratory Epithelium RE was significantly thicker in both types of CF mice (G551D 14.8 µm, FABP 14.9 µm) compared with wild-type mice (10.0 µm; P < 0.01 for both), as shown in Figure 2. Mucous cell density was significantly greater in both types of CF mice (G551D 89.6 cells/mm, FABP 84.7 cells/mm) compared with wild type (54.0 cells/mm; P < 0.001 and P = 0.001, respectively), as shown in Figure 3. Collections of inflammatory cells under the RE, all with the appearance of lymphoid aggregates, were present in at least one section in 11 out of the 12 G551D mice and in 2 out of the 6 FABP mice, compared with none of the 12 wild-type mice (  2=20.8, P < 0.001). An example of an area of lymphoid aggregates is shown in Figure 4.
Olfactory Epithelium OE was significantly thinner in both types of CF mice (G551D 32.6 µm, FABP 39.0 µm) compared with wild-type (50.8 µm; P < 0.001 for both), as shown in Figure 5. OE cell density was also decreased in both types of CF mice (G551D 27.4 cells/1,000 µm2, FABP 28.6 cells/1,000 µm2) compared with wild-type (36.4 cells/1,000 µm2; P < 0.001 and P = 0.001, respectively), as shown in Figure 6. Further details of histologic analysis at each depth are given in the online supplement.
Nasal Lavage in Adult Mice The median cell count in nasal lavage was not significantly different between G551D (5,000 cells/ml, range 0–18,000) and wild-type mice (2,000 cells/ml, 0–20000, P = 0.11). Bacteria were cultured from each nasal lavage; coliforms, coagulase-negative Staphylococci, and Streptococcus faecalis predominated. Total concentrations of bacteria were not significantly different between mouse types. MIP-2 concentrations were not significantly different between G551D (9.8 pg/ml, 7.2–82.0) and wild-type (10.1 pg/ml, 7.3–15.5). There was no correlation between MIP-2, bacterial concentration, or cell count.
Embryo and Pup Investigation
We have demonstrated significant alterations to the mucosa of the nasal cavity of adult CF mice compared with wild-type animals. We report increased respiratory epithelial thickness, mucous cell hyperplasia, subepithelial accumulations of lymphocytes, and decreased thickness and density of the olfactory epithelium in CF mice. Yet there were no differences in nasal lavage cell counts, bacteria, or pro-inflammatory cytokine concentration between CF and wild-type mice, implying that these CF-related changes may be independent of infection and lumenal inflammation. The histologic changes were not present in pups or immediately pre-term embryos. Previous investigation of the nasal mucosa in CF mice has reported either no changes (24), dilatation of the submucosal glands (4, 6), or increases in the number of goblet cells (17). To our knowledge, neither the increased thickness of RE or the lymphoid aggregates seen in CF mice in the present study has not been reported previously. Increased RE thickness may be a marker of epithelial remodeling, and it could be similar to the epithelial metaplasia seen in human CF airway epithelium (25). Mucous cell hyperplasia is a feature of lower airway disease in human CF, and our data would suggest that this is not initially the result of infection and lumenal inflammation. Although lymphoid tissue is normally seen at the base of the septum posteriorly within the mouse nasal cavity, the lymphoid aggregates in the CF sections was more widely distributed. These lesions may be similar to those reported in the lower airways of FABP mice in a model of P. aeruginosa colonization, showing expanded bronchus-associated lymphoid tissue and patchy lymphoid aggregates (26). Further work is needed to elucidate the nature of these lymphoid aggregates in the present study, and their immunologic significance. Interestingly, in human CF airways taken at transplant or lobectomy, B lymphocytes formed aggregates of several hundreds of cells beneath the airway epithelium (27), and there is recent evidence that lymphocytes may be a factor orchestrating the subsequent dysregulated inflammatory response in CF (28). Murine OE has been extensively investigated as a model system to study development of the mammalian nervous system (29). It is divided into an apical layer of sustentacular cells, a middle layer of olfactory receptor neurons (ORNs), and a basal layer of both horizontal basal cells and "globose" basal cells (30). The OE, unlike most of the rest of the nervous system, has the capacity to regenerate ORNs throughout life. The globose basal cells are believed to be the progenitor cells, giving rise to further ORNs under a highly regulated process that includes the Mash1 and brain factor 1 transcription factors and expression of the gene Hes1 (29, 29, 30). Human OE cells have also been shown to exhibit multipotent stem cell properties (31). Our study extends previous observations on the altered nature of the OE in CF mice (32). These authors reported that murine OE appeared normal at birth, but by 4 to 6 months of age, sustentacular cells were dysmorphic, and there were dramatic reductions in olfactory receptor neurons. Unlike the findings in the present report, they did not report thinning of the OE; we cannot account for the discrepancy in the findings, since we studied adult mice of approximately the same age. The physiological counterparts of their anatomical observations were reduced odor-evoked responses, a lack of forskolin-stimulated chloride secretion, and an around 12-fold increase in amiloride sensitive sodium absorption, compared with wild-type mice. We have confirmed that the OE changes are postnatal, by studying OE in fetal CF mice for the first time, and shown that the postnatal changes are probably not related to infection or lumenal inflammation, which was not studied in the previous report. The explanation for the abnormal OE seen in both CF genotypes is unclear, but is conceivably related to either an underlying cellular developmental anomaly or an ongoing insult that inhibits the production of ORNs and hence thickness of the epithelium. CFTR expression in the wild-type nose is highest in the OE (33), which might support a relationship between the expression of CFTR and OE development. Whether loss of olfactory epithelium is a marker of neurogenic dysfunction, is due to a reduced supply of stem cells, is secondary to airway surface desiccation (as suggested by Grubb and coworkers, Ref. 32), or is a mere epiphenomenon, is unknown. To our knowledge, the appearance of OE in human patients with CF has not been reported. Sense of smell in CF has had limited investigation (34, 35), and there appears to be no inherent abnormality, although it is likely to be secondarily affected by sinus disease (36, 37). A description of olfactory epithelium in human autopsy tissue, or from nasal biopsies, might be an important initial study. We used a novel technique of nasal lavage to assess lumenal inflammation and bacterial content. There was no difference in lumenal inflammation or bacterial concentration on nasal lavage between mouse types. Because several types of bacteria were present at relatively high concentration, it is conceivable that more pathogenic bacteria (e.g., P. aeruginosa) remained undetected. We were only able to measure, because of volume limitations, the concentration of a single cytokine MIP-2, which was present in the majority at the lower limit of detection. While other inflammatory mediators may be involved in the development of these changes, MIP-2 (the murine homolog of IL-8) is a central pro-inflammatory cytokine. The volume of nasal lavage used is also likely to represent a large dilution of any fluid lining the nose. Unlike other investigations of the CF mouse nose (4, 6, 17, 24), this study used a histologic technique that allowed investigation of the nasal epithelium and preserved its orientation within the nasal cavity. Measurements were made at a number of depths within the nasal cavity, to overcome the fact that the CF heads appear to be smaller than wild-type heads. Significant differences between CF and wild-type mice were noted at all depths measured. This study also provides data on the starting points of the different cell types in the mouse nose, which is relevant to the measurements of murine nasal potential difference. OE started at a median depth of 3 mm, and with a degree of shrinkage the in vivo OE starting depth is likely to be somewhat deeper. This supports a study using an electrodeposition marking catheter which suggested that potential difference catheters should be inserted 2.5 mm into the mouse nose to contact RE rather than OE (38). Taken together, our results suggest that there may be early histologic and immunopathologic changes in the CF nasal mucosa that are not secondary to, but rather precede, overt infection and inflammation. These may be related to at least two possible initiating factors. First, CFTR has been suggested to function as part of a regulatory cascade in cellular differentiation (39), so that absence of CFTR may result in abnormal airway development. Increased epithelial cell proliferation has been found in human CF airway epithelium (40–42) with basal cells having heightened expression of epidermal growth factor (41). Reduced apoptosis has also been found in CF cell lines (43). CF airway epithelial cells have also been shown to be intrinsically pro-angiogenic, with increased expression of angiogenic factors such as VEGF-A, VEGF-C, bFGF, and PLGF at mRNA and protein level (44). In addition, a mouse model that overexpresses the epithelial sodium channel showed spontaneous lung disease somewhat similar to CF, in the absence of overt infection (45). Finally, recent data demonstrated both pre- and postnatal smooth muscle and neural alterations in the lower airways of CF mice (46). However, the finding of similar epithelium in CF and wild-type heads at or before birth in the current study makes a primary abnormality in cellular development less likely. Second, there may be inhaled pollutants or unidentified microorganisms that could have some postnatal effect on the nasal epithelium to explain the differences seen in the adult mice. In addition, CF epithelia may have an altered response to a given bacterial burden compared with wild-type cells. Altered interaction between P. aeruginosa and respiratory epithelial cells has been shown in both murine (47) and human cell culture models (48). It could therefore be argued that the presence of normal amounts of bacteria in the CF mouse nose may have lead to an abnormal epithelial response. The fact that we detected similar levels of MIP-2 in CF and wild-type animals, implying similar epithelial responses to the level of nasal bacteria, suggests that an abnormal response to the present level of nasal bacteria is not the likely explanation. We cannot, however, exclude a different response to infection in the CF mice in terms of the pro-inflammatory or growth factors that we were unable to measure. Due to a viral infection in the breeding colony, we used a different background strain for the embryo and pup investigation compared with the adult mice. Some cell types within the airway may vary with strain (49), and we accept that this is a weakness in the study. Further work should investigate the effect of CF genotype on nasal histology at different stages of pre- and postnatal development, using the same background strain, and both sterile and nonsterile environments. In conclusion, we have investigated a novel murine model and detected structural and immunopathologic abnormalities in the nasal cavity of adult CF mice, likely unrelated to lumenal inflammation or infection. These may represent novel and potentially important therapeutic targets in CF. Nasal histology may provide a model in which to investigate the pathophysiology of early structural airway changes in CF, and may also provide a novel assay for CFTR-related therapeutic interventions.
The authors thank Sara Escudio Garcia and Luci Somerton for managing the G551D breeding colony and for performing the tail tip genotype analysis; John Williams for the processing of the adult tissue; Yusheng Qiu for the processing of the embryo tissue; the staff of the Central Biosciences and NHLI animal facilities; Neil Madden for help with the microbiological analysis; and Abraham Jacob, Brian Faddis, and Richard Chole, Department of Otolaryngology, Washington University School of Medicine, St. Louis, for their assistance in developing the technique of nasal section.
This work was supported by the Cystic Fibrosis Trust. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1165/rcmb.2007-0284OC on January 31, 2008 Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 25, 2007 Accepted in final form November 25, 2007
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
CLINICAL RELEVANCE
