|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Unit of Industrial Toxicology and Occupational Medicine (Drs. de Burbure, Clippe, Dumont, and Bernard), Faculty of Medicine, Université Catholique de Louvain, Belgium; Allergy and Immunology Unit (Drs. Pignatti and Moscato), Fondazione Salvatore Maugeri, IRCCS, Pavia, Italy; Laboratorio di Tossicologia Industriale (Drs. Corradi and Mutti), Universita degli Studi di Parma, Parma, Italy; and Department of Internal Medicine (Dr. Malerba), University of Brescia, Spedali Civili, Brescia, Italy.
Correspondence to: Alfred Bernard, PhD, Unit of Industrial Toxicology and Occupational Medicine, Université Catholique de Louvain, Clos Chapelle-aux-Champs 30, bte 3054, B-1200 Brussels, Belgium; e-mail: bernard{at}toxi.ucl.ac.be
Abstract
Rationale: Uteroglobin-related protein 1 (UGRP1) and Clara cell protein (CC16), members of the secretoglobin family, increasingly appear to play a role in airway inflammatory response.
Objective: To explore levels of UGRP1 and CC16 in induced sputum of patients with asthma and rhinitis.
Methods: Induced-sputum samples of patients with asthma or rhinitis (n = 32 each; atopic asthma, n = 24; atopic rhinitis, n = 20) and from 19 nonsmoking nonatopic control subjects were analyzed for cytology and levels of UGRP1, CC16, and albumin.
Measurements and main results: Sputum UGRP1 increased in both asthma and rhinitis, most strikingly so in asthma, in which changes were most significant in atopic individuals. By contrast, sputum CC16 did not change significantly in either condition, although it was positively correlated with UGRP1 in patients and control subjects. Changes in sputum UGRP1 in atopic asthma were not linked to permeability changes reflected by increased albumin levels but correlated positively with sputum macrophages and negatively with eosinophils. The observed differences in UGRP1 and CC16 may be linked to different cell populations being responsible for their secretion; UGRP1 is mainly secreted in larger conducting airways, whereas CC16 is mainly secreted by the nasal and peripheral airways epithelium.
Conclusions: The increase in UGRP1 but not of CC16 in asthma and rhinitis suggests that UGRP1 may play a role in these inflammatory diseases.
Key Words: asthma Clara cell protein induced sputum rhinitis uteroglobin-related protein 1
Noninvasive methods for assessing the airway inflammatory response, which is central to the pathogenesis of asthma and rhinitis, have been developed in the last decade with the hope to reliably and accurately monitor patient progress and response to therapy. One such newly developed and evolving tool consists in the study of lung-specific proteins secreted into the airways lining fluid by both the conducting and the respiratory airway epithelial cells.1
The Clara cell protein (CC16), a small 16-kd protein secreted throughout the airways but mainly by the bronchiolar Clara cell (hence called CC16), has been studied in relation to asthma and rhinitis in particular because its gene is localized on the p12-q13 region of human chromosome 11 involved in inflammation regulation. CC16 expression is indeed decreased in asthma2 and in case of respiratory tract inflammation in both man3 and animals.45 A38G gene polymorphism has also been linked to lower serum CC16 levels and a higher risk of being asthmatic in some populations.67 A recent gene-profiling study8 in allergic rhinitis reported CC16 to have the most decreased antiinflammatory gene expression among > 44,927 genes. CC16-deficient mice show increased sensitivity to hyperoxia or ozone damage, exaggerated inflammatory responses,1 with increased pulmonary eosinophilia when antigen-sensitized and challenged.9 CC16 has important immunosuppressive and antiinflammatory properties, which include the inhibition of inflammatory mediators such as cytosolic phospholipase A2, interferon-
, tumor necrosis factor-
, as well as platelet-derived growth factor-induced chemotaxis of fetal lung fibroblasts.1 Measuring CC16 in BAL fluid, where it reflects epithelial function, and in serum, bringing useful information on the permeability of the alveolocapillary membrane, has therefore proven useful in assessing both acute and chronic airway damage.
Recently, another small protein belonging to the same family as CC16, the secretoglobin family,10 has been discovered: uteroglobin-related protein 1 (UGRP1) is a 17-kd, homodimeric lung-specific protein111213 also called high in normal 2 protein or secretoglobin SCGB3A2.12 As mouse UGRP1 has an overall amino-acid sequence identity of 25% to mouse CC16, including an area called antiflammin containing several lysine residues, it is supposed to share some of its antiinflammatory properties, although its precise functions are still unknown.11 UGRP1 appears highly specific of the airways, and our laboratory has been characterizing, sequencing, and evaluating it along other pneumoproteins as potential biomarker of lung damage since 1999 (GenBank deposit: uteroglobin-related protein 2: AF43683941; UGRP1: AF4395447).14 Expression of UGRP1 messenger RNA has been shown to decrease in mouse lung after induction of inflammation by in situ hybridization.11 In this study, levels of CC16 and UGRP1 were analyzed for the first time in induced-sputum samples of patients with either asthma or rhinitis with a view to investigating the differential diagnostic potential of the two pneumoproteins in these two conditions.
Materials and Methods
Population Studied
Patients gave signed informed consent to participate in the study, which was approved by the ethical committees of the Salvatore Maugeri Foundation IRCCS and the University of Brescia, and was conducted according to the Declaration of Helsinki. Induced-sputum specimens and lung function measures were obtained from 32 patients with intermittent to moderate-persistent bronchial asthma, 32 rhinitis patients, and 19 healthy volunteers (21 men and 11 women, 10 men and 22 women, and 10 men and 9 women, respectively). Patients came from Pavia (Allergy and Immunology Unit, Fondazione Salvatore Maugeri), and control subjects came from Brescia or Parma University (Internal Medicine and Industrial Toxicology, respectively). Atopy was defined as at least one positive skin-prick test result for common inhalant allergens (with positive histamine prick result). Bronchial asthma was diagnosed according to National Institutes of Health guidelines.15 Allergic rhinitis was diagnosed according to Allergic Rhinitis and Its Impact on Asthma16 guidelines. Patients with rhinitis symptoms and negative skin test results, after consultation with an ear, nose, and throat specialist, received the diagnosis of nonallergic rhinitis.
Sputum Induction and Processing
Sputum induction was obtained as previously described.17 Sputum processing was performed with 0.1% dithiothreitol according to international guidelines.1718 After separation from supernatant by centrifugation, cell count and viability by trypan blue exclusion were determined by light microscopy. Cytospins were stained (Diff-Quick; Dade Diagnostika GmbH; Unterscheiheim, Germany), analyzed for differential cell count by counting 500 nonsquamous cells minimum, and reported as percentages of total nonsquamous cells. Only sputum samples with < 30% squamous cells were considered acceptable.
Sample Analysis
Sputum supernatants were stored at 80°C until evaluation. Sputum CC16 and albumin were quantified by semiautomated nonisotopic latex immunoassay as previously described.1920 The procedure for quantifying UGRP1 was similar to the above, using a combination of two in-house polyclonal anti-UGRP1 antibodies. The between-run and within-run coefficients of variation of both immunoassays in biological fluids, including sputum, were between 5% and 10%. The effect of dithiothreitol treatment on the concentrations of UGRP1 and CC16 was assessed by adding dithiothreitol (0.1%) to six sputum samples with UGRP1 and CC16 concentrations ranging from 20 to 3,980 µg/L and 27 to 2,170 µg/L, respectively. The concentrations of UGRP1 in treated and untreated samples were highly correlated (Pearson r = 0.98, p < 0.001) with posttreatment values averaging 70% of pretreatment values. Concentrations of CC16 in treated and untreated samples were also highly correlated (Pearson r = 0.94, p < 0.001) with no systematic difference (posttreatment values averaged 102% of the pretreatment values). Additional details on the method for making UGRP1 measurements are provided in an on-line data supplement. Due to the small volumes of some samples, all biological parameters could not be determined in all subjects; exact numbers are indicated in the tables.
Lung Function Tests
Spirometry was performed according to European Respiratory Society guidelines21 with a computerized water-sealed spirometer (Biomedin; Padova, Italy). Measurements were expressed as percentage of predicted normal values based on race, age, sex, and height. Methacholine challenge testing was performed with a nebulizer (Mefar; Brescia, Italy) connected to a dosimeter as previously reported.22 The provocative dose of methacholine causing a 20% fall in FEV1 was considered positive at < 1,000 µg.
Statistical Analysis
Results are presented as mean ± SEM. For statistical analysis, UGRP1, CC16, albumin, and absolute macrophage or eosinophil counts in sputum were normalized by log transformation, and their results are presented as geometric means. Group means were compared by two-sided, unpaired Student t test or by analysis of variance followed by the Dunnett post hoc test. Associations between sputum UGRP1, CC16, albumin, and sputum cell counts were evaluated by simple regression analysis and calculating using the Pearson correlation coefficient. Influence of age, gender (male/female), diagnosis (asthma, 0/1; rhinitis, 0/1), smoking status (never-smoker/current smoker), and atopy (0/1) was studied by stepwise multiple regression (entry, p = 0.25; staying in model, p = 0.05). Three former smokers were grouped with never-smokers. All statistical analysis was performed using statistical software (StatView 5; SAS Institute; Cary, NC). Figures were drawn with StatView or GraphPad Prism (GraphPad Prism 3.03; GraphPad Software; San Diego, CA). Statistical significance was set at 0.05.
Results
The general characteristics, lung function, and sputum cytology of control subjects and rhinitis and asthmatic patients are given in Table 1 , including disease duration, atopy and smoking status, steroid treatment, methacholine sensitivity, sputum cytology, and some lung function tests. Whereas all 19 control subjects were nonsmokers and nonatopic, approximately 4 in 10 asthmatic and rhinitis patients were smokers, and two thirds were atopic. There were no significant differences in lung function between patients and control subjects, indicating overall good control of the disease. However, in comparison with rhinitis patients, asthmatics had significantly lower mean values of FEV1, Tiffeneau index, and vital capacity. Only one rhinitis patient (1 of 32 patients, 3.1%) was methacholine sensitive, compared to 12 of 32 asthmatics (37.5%). The proportions of asthmatics with a positive methacholine test result were similar between those receiving inhaled corticosteroids (7 of 19 patients, 36.8%; low dose, 1/6; medium dose, 4/9; and high dose, 1/6) and those who were not receiving steroid medication (5 of 13 patients, 38.5%). As expected, asthmatics had significantly raised sputum eosinophil numbers but reduced macrophages compared to control subjects and rhinitis patients. Eosinophils were also raised in rhinitis compared to control subjects, although much less dramatically so than in asthma. Epithelial cell numbers were increased in both groups of patients compared to control subjects.
|
|
|
Sputum UGRP1 was significantly correlated to CC16 in all three subject groups, whereas correlations of these two proteins with albumin were weak and even nonexistent in the case of asthma (Fig 3 ). In atopic asthmatics, sputum UGRP1 and CC16 correlated positively with macrophages and negatively with eosinophils expressed as total cell percentage, whereas albumin indicated opposite correlations (Fig 4 ). These correlations were confirmed when using absolute cell numbers per milligram and more pertinently per milliliter of sputum unit used for proteins. In particular, positive correlations with sputum macrophages were then also observed for the whole population (UGRP1 [n = 64]: R2 = 0.165, p = 0.0009 and R2 = 0.187, p = 0.0004; CC16 [n = 64]: R2 = 0.21, p = 0.0001 and R2 = 0.234, p < 0.0001 with macrophage numbers per milliliter and milligram of sputum, respectively). It is important to realize that correlations emerging between URGP1 and CC16 as well as between these proteins and cells counts were calculated over very wide ranges of values, requiring logarithmic scales to be displayed. Although statistically very significant, most of these correlations reflect associations that are relatively weak. UGRP1 concentration in sputum, for instance, can show variations of more than one order of magnitude for a given value of CC16 (Fig 3), while < 25% of UGRP1 or CC16 variations are related to changes in sputum eosinophils (Fig 4).
|
|
To our knowledge, this is the first report of UGRP1 protein being assayed in any human biological sample, and the first time CC16 is measured in induced sputum. Most interestingly, UGRP1 levels were significantly increased in both asthma and rhinitis compared to control subjects, the rise in atopic asthma being the most remarkable. CC16 levels, although correlated to UGRP1, were by contrast not observed to change significantly in either asthma or rhinitis. These increases in UGRP1 cannot be explained by epithelial permeability changes, known to occur in both asthma and rhinitis.23 UGRP1 levels were indeed only weakly or not at all correlated with sputum albumin. Furthermore, as UGRP1 was undetectable in serum samples of this study (as shown in the "Materials and Methods" section), transepithelial leakage from blood is unlikely to be the cause of the high sputum UGRP1 levels. The only plausible mechanism for increased UGRP1 in sputum appears therefore to be an increased synthesis and secretion of the protein in those areas contributing to sputum collection.
Our findings concerning CC16 in sputum were rather unexpected from several points of view. According to the literature, a decrease in CC16 could have been expected both in asthma and in rhinitis. Indeed, patients who had asthma for > 10 years have decreased serum CC16,24 in correlation with decreased populations of Clara cells in their airways.2 Furthermore, children with allergic rhinitis were recently shown to have decreased CC16 levels in nasal fluid25 and highly reduced CC16 gene expression.8 The lack of changes of sputum CC16 in asthma or rhinitis probably stems from the fact that induced sputum mainly originates from central airways, with little or no recovery from the peripheral respiratory system where CC16 is predominantly produced.26 By contrast, serum levels of CC16 mainly originate from distal and peripheral airways, which offer a vast surface area for transepithelial leakage into the bloodstream.1 In comparison with sputum, serum might therefore be a more reliable indicator of the amount of CC16 secreted in the lower respiratory tract. Even though sputum CC16 levels in atopic asthmatics were overall unchanged, they were found to correlate positively with macrophage and negatively with eosinophil percentages. The correlation with macrophages might simply reflect their role in the respiratory tract defense mechanisms, and maybe also tie up with the known role of macrophages in phagocytosis and recycling of surfactant and other lung secretory proteins.27 The negative correlation with eosinophils is likely linked to the antiinflammatory role of CC16.825
With regard to UGRP1, similarly to CC16, genetic analyses13 predicted lower UGRP1 levels in some asthmatics, following observations of reduced transcriptional activity linked to a G-A point mutation at 112 base-pair in the human UGRP1 gene promoter. However, a study28 in CC16-null mice demonstrated compensatory increases in UGRP1 and uteroglobin-related protein 2 messenger RNA expression. Assuming altered CC16 gene expression in asthma and rhinitis, as suggested by the above-mentioned studies,824 could therefore in part explain a compensatory rise in UGRP1 as observed in our studies. Individual genetic differences linked to known CC16 and UGRP1 gene polymorphism could further explain the differences observed, notably the significant rise in UGRP1 in atopic asthmatics only.2913 Interestingly, although weak and not specific of URGP1 (also found with CC16), the negative correlation in atopic asthmatics between UGRP1 and eosinophil percentage is consistent with the downregulation of UGRP1 expression observed in ovalbumin-challenged mice with high tissue eosinophilia linked to high interleukin-5 levels,30 as well as the very recent study by Chiba et al (2006, epub ahead of print)31 reporting that intranasal administration of adeno-UGRP1 markedly reduced the number of infiltrating inflammatory cells, particularly eosinophils. As with CC16, correlations of UGRP1 with sputum macrophages may be linked to their role in defense mechanisms, UGRP1 being a known ligand of the MARCO receptor (macrophage scavenger receptor with collagenous structure) of alveolar macrophages.32 Interestingly, although cell numbers were overall in agreement with induced-sputum reference values,3334 total cell numbers were increased in our asthmatic patients, whereas macrophage percentages were significantly reduced. Further studies would be required to fully comprehend the interactions between UGRP1 and macrophages, particularly in asthma, in which macrophage function alterations feature high in the literature.3536
Morphologic studies offer possible mechanistic clues for the observed differences between UGRP1 and CC16: human UGRP1-secreting cells are located not only primarily in the larger upper airways but in submucosal glands in particular, whereas CC16 secretion is predominant in the lower conducting airways although it also occurs in submucosal acinar and duct cells early in life1128 and in some goblet cells.37 Since asthma is known to lead to airway remodeling, not only with goblet cell hyperplasia but with hypertrophy and hyperplasia of the submucosal glands in particular,3839 the rise in UGRP1 observed in asthmatic patients could reflect this hypersecretory state and therefore become an indirect biomarker of airway remodeling. The significant rise in UGRP1 observed in rhinitis, albeit smaller than in asthma, corroborates recent reports that remodeling occurs not only in asthma but also in rhinitis and other conditions inducing chronic cough.4041 It would furthermore be worth investigating how CC16 and UGRP1 levels would respond to corticosteroid treatment recently shown to reverse airway remodeling in mice,42 as the small number of patients in each treatment category did not allow us to make conclusive observations.
Finally, it appears from our study in asthma and rhinitis that induced sputum, clearly less invasive than bronchial lavage, could prove a most helpful tool for monitoring inflammatory cells and antiinflammatory proteins such as UGRP1 and CC16. This could be practical for clinicians keen to monitor treatment effect on remodeling in individual patients.43 As this study is the first report of UGRP1 and CC16 levels in induced sputum, further studies conducted in larger numbers of atopic and nonatopic patients would be required to confirm the differential levels of these two new biomarkers in asthma and rhinitis.
Acknowledgements
The authors thank the staff members of the Internal Medicine and Industrial Toxicology Departments of Brescia and Parma Universities, respectively, for volunteering to provide the induced-sputum control samples used in this study.
Footnotes
Abbreviations: CC16 = Clara cell protein; UGRP1 = uteroglobin-related protein 1
None of the authors have any conflicts of interest to disclose.
Received for publication April 21, 2006. Accepted for publication July 1, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |