|
|
Background The evolutionarily conserved 18-glycosyl-hydrolase family contains true chitinases and chitinase-like proteins that lack enzymatic activity. Acidic mammalian chitinase has recently been associated with animal models of asthma. The related chitinase-like protein, YKL-40 (also called human cartilage glycoprotein 39 [HCgp-39] and chitinase 3–like 1), can be readily measured in the serum. However, its relationship to asthma has not been evaluated.
Methods We quantified serum YKL-40 levels in three cohorts of patients with asthma — one recruited from the patient population at Yale University, one from the University of Paris, and one from the University of Wisconsin — as well as in controls from the surrounding communities. In the Paris cohort, immunohistochemical analysis and morphometric quantitation were used to evaluate the locus of expression of YKL-40 in the lung. The clinical characteristics of the patients with high serum or lung YKL-40 levels were also evaluated.
Results Serum YKL-40 levels were significantly elevated in patients with asthma as compared with controls. In the Paris cohort, lung YKL-40 levels were elevated and were correlated with circulating YKL-40 levels (r=0.55, P<0.001) and with airway remodeling (measured as the thickness of the subepithelial basement membrane) (r=0.51, P=0.003). In all three cohorts, serum YKL-40 levels correlated positively with the severity of asthma and inversely with the forced expiratory volume in 1 second. Patients with elevated levels of YKL-40 had significantly more frequent rescue-inhaler use, greater oral corticosteroid use, and a greater rate of hospitalization than patients with lower levels.
Conclusions YKL-40 is found in increased quantities in the serum and lungs in a subgroup of patients with asthma, in whom expression of chitinase in both compartments correlates with the severity of asthma. The recovery of YKL-40 from these patients indicates either a causative or a sentinel role for this molecule in asthma.
We hypothesized that the level of expression of YKL-40 would be increased in patients with asthma and would correlate with the severity of the asthma. To test this hypothesis, we quantitated the levels of circulating YKL-40 in three cohorts of patients with asthma and controls and compared these findings with the severity of asthma, levels of expression of YKL-40 in the airway, and measures of airway remodeling.
Methods
Study Design and Subjects
We performed a cross-sectional analysis of serum samples from an established cohort of patients with asthma from Yale University, in New Haven, Connecticut, plus controls. On the basis of the results of this exploratory analysis, we analyzed sets of serum samples from a cohort of patients with asthma from the University of Paris, plus controls, and from a cohort of patients with asthma from the University of Wisconsin in Madison, plus controls. In addition, the levels of expression of YKL-40 in the airway and their relationship to serum YKL-40 levels was investigated in the Paris cohort. The methods of recruitment of patients with asthma and controls from the surrounding community were similar at each of the three centers. Each center had its own inclusion and exclusion criteria for controls and patients with asthma and its own criteria for the severity of asthma, based on established guidelines (see the Supplementary Appendix, available with the full text of this article at www.nejm.org).13
In the Yale and Wisconsin cohorts, patients with asthma were stratified according to the severity of asthma as defined by medication requirements to achieve asthma control (the use of a rescue inhaler 
2 days per week). The case definition for mild asthma was asthma requiring the use of 200 µg or less of fluticasone per day, or the equivalent, to achieve asthma control. Moderate asthma was defined as that requiring the use of 200 to 800 µg of fluticasone per day, or the equivalent, to achieve control. In the Paris cohort, the case definition for mild asthma was asthma requiring exclusive treatment with short-acting inhaled 
2-agonists only when needed for asthma symptoms. The definition of moderate asthma was asthma requiring treatment with 250 to 500 µg of fluticasone per day, or the equivalent, in addition to long-acting 2-agonists to achieve asthma control. In contrast to the Yale and Wisconsin cohorts, for the Paris cohort, the evaluation of pulmonary function was also used in the case definition for asthma: patients with mild or moderate disease had a prebronchodilator forced expiratory volume in 1 second (FEV1) of 80% or more of the predicted value and a ratio of FEV1 to forced vital capacity (FVC) of 0.70 or more; patients with severe asthma had a prebronchodilator FEV1 of less than 70% of the predicted value and a ratio of FEV1 to FVC of less than 0.70. In all three cohorts, patients with severe asthma fulfilled the definition of severe asthma adopted by the American Thoracic Society Workshop on Refractory Asthma.14
Our studies were approved by the human investigation committees at all three institutions. All subjects gave written informed consent.
Measurement of Serum YKL-40 and Serum IgE Levels
Measurement of serum YKL-40 and serum IgE levels was performed in duplicate with the use of commercially available enzyme-linked immunosorbent assay (ELISA) kits (for YKL-40, Quidel; for IgE, Pharmacia or Dade-Behring); median values are presented. The minimum detection limit of the YKL-40 assay is 20 ng per milliliter. We evaluated the specificity of the YKL-40 ELISA by confirming that the capture and detection antibodies lacked cross-reactivity to other human chitinases, including acidic mammalian chitinase, YKL-39, and chitotriosidase.
Fiber-Optic Bronchoscopy and Sample Collection
At the University of Paris, bronchoscopy was performed to obtain specimens for biochemical and immunohistochemical analysis. The bronchoscopy was performed on the same day as phlebotomy, according to the guidelines of the American Thoracic Society.15,16 The participating patients with asthma had not had an exacerbation within 2 months before bronchoscopy. Bronchoalveolar lavage was performed with the use of 50 ml of sterile saline, and the samples were immediately centrifuged (at 150xg for 10 minutes at 4°C). Cytospin preparations from cell pellets were fixed in acetone for 10 minutes at room temperature. Four bronchial-biopsy specimens were taken from the bifurcations of the right middle or lower lobes, immediately immersed in Tissue-Tek compound (Sakura Finetek), snap-frozen in liquid nitrogen, and kept at –80°C until they were processed.
Immunohistochemical and Morphometric Analyses
Serial 5-µm sections of bronchial-biopsy specimens were mounted on glass slides coated with 
-methacryloxypropyltrimethoxysilane (Sigma) and fixed in acetone for 10 minutes at room temperature. Immunohistochemical analysis was performed as previously described.16 The primary antibody was rabbit anti–human YKL-40 antibody (1 µg per milliliter) (MedImmune). Control antibodies were rabbit IgG (DakoCytomation) and anti–YKL-40 antibody coincubated with 1 µg of recombinant human acidic mammalian chitinase and chitotriosidase per milliliter. The secondary antibody was biotin-conjugated antirabbit antibody. Detection was accomplished with the use of the avidine–alkaline phosphatase complex (Vector), fast red (DakoCytomation), and a light nuclear Mayer's hematoxylin counterstain.
The number of YKL-40–positive cells was counted per cubic millimeter of bronchial epithelium and subepithelium in tissue sections. The total biopsy area (in cubic millimeters) and thickness of the subepithelial basement membrane were determined by means of morphometry and computer-assisted image analysis (Microvision Instruments).16 The areas of biopsy were similar among patients with mild asthma, those with moderate asthma, those with severe asthma, and controls (median, 0.433 to 0.496 mm2; P=0.40 by the Kruskal–Wallis test). Immunohistochemical measurements were made at a magnification of 400, and the total biopsy area and thickness of the subepithelial basement membrane were assessed at a magnification of 60.
Statistical Analysis
Analyses were performed with the use of SPSS software, version 6.1.3, or MedCalc. We compared the characteristics of subjects using the Kruskal–Wallis test and Pearson's chi-square test. Simple associations were assessed with the use of Spearman's rank-correlation analysis. YKL-40 levels were not normally distributed and therefore were logarithmetically transformed. The assumption of normality of the log-transformed values was verified by means of the Kolmogorov–Smirnov test, and the values were compared among the study groups with the use of the independent-samples t-test and analysis of covariance with a test for linear trend. The YKL-40 levels are expressed as medians and interquartile ranges. Post hoc comparisons of pairwise differences between patients with mild asthma, those with moderate asthma, and those with severe asthma were evaluated with significance tests corrected with the use of Tukey's test. Multivariable linear regression analysis was performed to determine the associations of variables with YKL-40 in the entire study sample (patients with asthma and controls). The coefficient of variation representing the intrasubject variability was calculated as the square root of the mean squared coefficient of variation for each subject's measurements over the study period.
Results
Characteristics of the Patients
In the Yale cohort, there were significant differences between controls and patients with asthma in factors associated with asthma; for example, patients with asthma had a higher mean body-mass index (BMI) (P=0.01), a history of atopy (P=0.001), and elevated IgE levels (P=0.001) (Table 1).17 More black and Latino patients had severe asthma as compared with mild or moderate asthma. As compared with patients with mild or moderate asthma, those with severe asthma had a history of more hospitalizations, intubations, use of rescue medication, and periods of tapering of oral corticosteroids; a longer duration of asthma; and more severely compromised pulmonary function.
|
Serum YKL-40 Levels
In the Yale cohort, the serum YKL-40 levels in patients with asthma were higher than those in controls (median, 69.7 ng per milliliter [interquartile range, 40.0 to 107.1] vs. 58.3 ng per milliliter [40.0 to 73.3]; P=0.02) (Figure 1A). Among 40 patients with stable asthma and controls who had a total of 114 repeat measurements during the 4-year study period (20 subjects had two measurements, 10 had three measurements, 8 had four measurements, 1 had five measurements, and 1 had seven measurements), the mean coefficient of variation per subject was 37% (data not shown).
|
Immunohistochemical Analysis of YKL-40 in the Lung
The expression of YKL-40 in bronchial-biopsy specimens was evaluated in the Paris cohort (Figure 2). These studies showed that there were few YKL-40–expressing cells in controls, with significantly increased numbers in patients with asthma (median, 3.1 YKL-40–positive cells per cubic millimeter [interquartile range, 2.1 to 7.4] vs. 16.2 per cubic millimeter [9.1 to 30.2]) (P=0.005). YKL-40 staining was seen in subepithelial cells from the majority of patients with asthma (Figure 2). Among patients with severe asthma, the number of subepithelial cells with YKL-40 staining was increased, and staining of the bronchial epithelium was also evident (Figure 2D and 2E). In cytospin preparations of bronchoalveolar-lavage specimens from the patients with severe asthma, YKL-40 was also noted in the cytoplasm of macrophages and neutrophils (Figure 2F). In patients with asthma, lung YKL-40 levels correlated with serum YKL-40 levels (r=0.55, P<0.001) (Figure 3). No correlation was found between the number of YKL-40–positive cells in biopsy specimens and the number of inflammatory cells (macrophages, eosinophils, lymphocytes, or neutrophils) in bronchoalveolar-lavage specimens (data not shown).
|
|
On the basis of the findings described above, the relationship between YKL-40 levels in the airway and the severity of asthma was evaluated (Figure 4). These studies showed that, in the Paris cohort, the median number of tissue cells expressing YKL-40 correlated positively with the severity of asthma, with 3.1 cells per cubic millimeter (interquartile range, 2.1 to 7.4) in controls, 13.7 per cubic millimeter (3.0 to 28.3) in patients with mild asthma, 11.3 (7.9 to 17.5) in those with moderate asthma, and 23.1 per cubic millimeter (15.2 to 54.1) in those with severe asthma (P for trend=0.003) (Figure 4A).
|
The circulating YKL-40 levels and the severity of asthma were also evaluated among patients with asthma in all three cohorts. In the Yale cohort, the median YKL-40 levels were 49.1 ng per milliliter (interquartile range, 36.7 to 94.2) among patients with mild asthma, 68.4 ng per milliliter (38.0 to 88.0) among those with moderate asthma, and 77.0 ng per milliliter (44.6 to 158.4) among those with severe asthma (P for trend=0.003) (Figure 4C). This association between the severity of asthma and the circulating YKL-40 levels was also evident in the Paris cohort, with a median of 45.5 ng per milliliter (interquartile range, 24.5 to 78.5) among patients with mild asthma, 41.0 ng per milliliter (25.0 to 67.0) among those with moderate asthma, and 94.0 ng per milliliter (72.0 to 181.5) among those with severe asthma (P for trend=0.007) (Figure 4D), and in the Wisconsin cohort, with 37.5 ng per milliliter (25.1 to 60.9), 62.5 ng per milliliter (42.8 to 80.0), and 64.1 ng per milliliter (45.8 to 82.2), respectively (P for trend=0.048) (Figure 4E). Serum YKL-40 levels also correlated inversely with the FEV1 in all three cohorts (Yale: r=–0.22, P=0.01; Paris: r=–0.21, P=0.005; and Wisconsin: r=–0.33, P=0.009). Thus, the circulating YKL-40 levels correlated with the severity of asthma, the thickness of the subepithelial basement membrane, and pulmonary function in these cohorts.
Characteristics of Patients with High YKL-40 Levels
To further investigate high circulating YKL-40 levels, we performed a post hoc analysis of the data for correlations between serum YKL-40 levels and asthma characteristics in the Yale cohort (Table 2). The YKL-40 levels correlated positively with a number of clinical measurements associated with severe asthma, including the number of oral corticosteroid tapering periods in the previous year, the dose of oral corticosteroids, and the frequency of rescue-inhaler use, and correlated negatively with the percentage of the predicted FEV1. YKL-40 levels did not correlate with the age at the onset of asthma, the duration of asthma, the number of intubations or hospitalizations, the serum IgE level, or the inhaled-corticosteroid dose.
|
A pairwise examination of differences among the three asthma-severity groups revealed that patients with severe asthma had significantly higher YKL-40 levels than either those with mild asthma or those with moderate asthma (P=0.006 and (P=0.004, respectively). Furthermore, patients with asthma who had serum YKL-40 levels above the 50th percentile (60.94 ng per milliliter) had more impaired lung function and required higher doses of corticosteroids than those with YKL-40 levels below this threshold (P=0.04 and P=0.007, respectively) (data not shown).
Discussion
Our data from three independently recruited populations of patients with asthma show that circulating levels of YKL-40 are increased in patients with asthma, as compared with healthy persons, and correlate with the severity of disease. In addition, the serum YKL-40 levels correlate positively with the level of expression of YKL-40 in the airway, the thickness of the subepithelial basement membrane, and clinical indexes of disease severity and correlate inversely with lung function. Although these findings do not allow us to determine whether YKL-40 is a link in the chain of pathogenesis of more severe asthma or a sentinel molecule indicating more severe disease, they support the hypothesis that YKL-40 plays an important role in determining the pathobiologic characteristics of severe asthma.
Multivariable analysis showed that the association between severe asthma and elevated serum YKL-40 levels persisted even after adjustments were made for other variables associated with the severity of asthma, including race or ethnic group, sex, and BMI. Although this finding suggests that elevated circulating YKL-40 levels are a general feature of patients with severe asthma, the association was largely due to increases in the serum YKL-40 levels in the subpopulation of patients with severe asthma. One interpretation of this finding is that severe asthma is a heterogeneous disorder and that elevated YKL-40 levels indicate a pathobiologically distinct form of the disease.
Our studies showed that the association between YKL-40 and asthma could be generalized across populations of patients with asthma of varying severity, as defined with the use of various criteria, and during treatment with their usual therapeutic regimens. This allowed for the evaluation of multiple cohorts of patients with asthma under "real world" conditions. However, as a result of this design, the studies were cross-sectional, the populations were not strictly controlled in terms of medication use, an exacerbation study was not conducted, and biopsy specimens were available for only one cohort. These limitations need to be kept in mind in interpreting our findings.
Chitin, a polymer of N-acetylglucosamine, is the second most abundant polysaccharide in nature. It is an essential part of the exoskeleton of crustaceans and insects, the walls of fungi, and the microfilarial sheath of nematodes.2,3,18,19,20 In these settings, chitin confers protection against the harsh conditions in the environment of the organism. Although mammals do not produce chitin, studies have shown that they do express true chitinases, including acidic mammalian chitinase and chitotriosidase.1,21,22,23 These studies also identified the chitinase-like protein YKL-40, which binds chitin but lacks chitinase activity.12 Although the biologic characteristics of YKL-40 and other chitinase-like proteins remain incompletely understood, a role in inflammation and tissue remodeling has been suggested by the findings that YKL-40 and its murine homolog (breast regression protein 39) regulate stromal-cell proliferation, differentiation, and survival and that elevated circulating YKL-40 levels are present in patients with meningitis, pneumonia, rheumatoid arthritis, osteoarthritis, breast or lung cancer, and hepatic fibrosis.10,24,25 Our studies add to the understanding of the biologic characteristics of chitinase-like proteins by defining important relationships between circulating YKL-40 levels and severe asthma.
In conclusion, we have shown that YKL-40 is found in increased quantities in the circulation and lungs of a subpopulation of patients with asthma, in whom the protein levels correlate positively with the severity of disease and the thickness of the subepithelial basement membrane and inversely with lung function. Although our data suggest that YKL-40 either participates in the pathogenesis of asthma or is a biomarker of severity, prospective studies will be required to determine the setting in which serum YKL-40 levels increase in relation to the development of severe disease, whether YKL-40 levels are stable or increase during exacerbations of asthma, and the potential role of serum YKL-40 levels in the management of asthma and asthma research. Our data do, however, indicate that the biologic characteristics of YKL-40 should be incorporated into the understanding of the pathobiologic characteristics of asthma.
Supported in part by grants from MedImmune and from the National Institutes of Health (R01-HL-081639).
Dr. Chupp reports receiving speaker's fees from GlaxoSmithKline, Merck, AstraZeneca, and Novartis and grant support from Boehringer Ingelheim, Wyeth, Centocor, Pfizer, MedImmune, and Genentech. Dr. Jarjour reports receiving consulting fees from GlaxoSmithKline, Genentech, Asthmatix, and Sanofi-Aventis; lecture fees from Schering-Plough; and grant support from GlaxoSmithKline, Dey, Sanofi-Aventis, and Chiesi. Dr. Shim reports receiving grant support from Merck. Drs. Reed, Coyle, and Keiner report being employees of MedImmune. Dr. Cullen reports receiving grant support from Alcoa. Drs. Pretolani and Elias report receiving grant support from MedImmune, which is a company investigating YKL-40 as a therapeutic target in asthma. A U.S. patent (no. 7,214,373) entitled "Methods, compositions and kits relating to chitinases and chitinase-like molecules and inflammatory disease" was issued on May 8, 2007, and is exclusively licensed to MedImmune (now owned by AstraZeneca); Dr. Elias is one of the inventors. Dr. Elias also reports receiving consulting fees from Eli Lilly, Pfizer, Promedior, MedImmune, Roche, and Millenium. No other potential conflict of interest relevant to this article was reported.
We thank Karen D'Angelo for her support in the preparation of an earlier version of this manuscript and all the subjects who made this study possible.
Source Information
From the Yale University School of Medicine, New Haven, CT (G.L.C., C.G.L., Y.M.S., C.T.H., S.H., J.D.D., M.C., J.A.E.); the University of Wisconsin School of Medicine, Madison (N.J.); MedImmune, Gaithersburg, MD (J.R., A.J.C., P.K.); and INSERM Unité 700, Université Paris 7 (M.G., M.A., M.P.), and Centre Hospitalier Universitaire Bichat Claude Bernard (M.-C.D., M.A.) — both in Paris.
Drs. Pretolani and Elias contributed equally to this article.
Address reprint requests to Dr. Chupp or Dr. Elias at the Section of Pulmonary and Critical Care Medicine, Yale University School of Medicine, P.O. Box 208057, 300 Cedar St., TAC 441, New Haven, CT 06520, or at geoffrey.chupp{at}yale.edu or jack.elias{at}yale.edu.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Chitinase-like Protein and Asthma
Kuepper M., Bratke K., Virchow J. C., Chupp G., Elias J. A.
Extract |
Full Text |
PDF
N Engl J Med 2008;
358:1073-1075, Mar 6, 2008.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
