|
|
, M.D.
Background The number of type 2 helper CD4+ T cells is increased in the airways of persons with asthma. Whether the majority of these cells are class II major-histocompatibility-complex–restricted cells or are among the recently identified CD1d-restricted invariant natural killer T cells is a matter of controversy. We studied the frequency of invariant natural killer T cells in the airways of subjects with mild or moderately severe asthma to investigate the possibility of an association between the number of invariant natural killer T cells in the airway and disease severity. We also studied whether an increased number of these cells is a feature of chronic obstructive pulmonary disease (COPD).
Methods We enumerated invariant natural killer T cells by flow cytometry with the use of CD1d tetramers loaded with
Results Fewer than 2% of the T cells obtained from all subjects on airway biopsy, bronchoalveolar lavage, and sputum induction were invariant natural killer T cells, with no significant differences among the three groups of subjects. No expression of messenger RNA for the invariant natural killer T-cell–receptor domains V
Conclusions Invariant natural killer T cells are found in low numbers in the airways of subjects with asthma, subjects with COPD, and controls.
-galactosylceramide and antibodies specific to the invariant natural killer T-cell receptor in samples of bronchoalveolar-lavage fluid, induced sputum, and bronchial-biopsy specimens obtained from subjects with mild or moderately severe asthma, subjects with COPD, and healthy control subjects. Real-time polymerase-chain-reaction analysis was performed on bronchoalveolar-lavage cells for evidence of gene expression of the invariant natural killer T-cell receptor. 24 and V
11 was detected in bronchoalveolar-lavage cells from subjects with asthma.
-galactosylceramide, a potent stimulus for invariant natural killer T cells, produce Th2 cytokines, implying a pivotal role in allergic airway inflammation.4 There have been two other reports of elevated levels of invariant natural killer T cells in the airways of persons with asthma,5,6 although both of them reported lower cell counts than those reported by Akbari et al. More recently, the findings of Akbari et al. have been challenged by Thomas et al.,7 who found low numbers of invariant natural killer T cells (0.4 to 2.1%) in persons with asthma not treated with corticosteroids.
We further investigated the role of invariant natural killer T cells in airway disease to determine whether these cells are not only a feature of moderately severe asthma requiring treatment with inhaled corticosteroids, as reported by Akbari et al.,4 but are also abundant in mild asthma and chronic obstructive pulmonary disease (COPD). We obtained samples from the airways of subjects with asthma, subjects with COPD, and healthy controls, using bronchoalveolar lavage, sputum induction, and bronchial biopsy. We examined the specimens for invariant natural killer T cells using a combination of flow cytometry and real-time polymerase-chain-reaction (PCR) analysis of gene expression for the V24 and V11 domains of the T-cell receptor that characterize invariant natural killer T cells.8,9,10
Methods
Study Population
The study was approved by the ethics committee of the Southampton University Hospitals Trust, and all subjects provided written informed consent. Twenty-four subjects with asthma (8 with mild asthma not treated with inhaled corticosteroids, and 16 with moderately severe asthma treated with inhaled corticosteroids), 10 subjects with COPD meeting established diagnostic criteria,11,12 and 12 controls were studied (Table E1 in the Supplementary Appendix, available with the full text of this article at www.nejm.org). All subjects with asthma had had stable symptoms within at least 6 weeks before enrollment. Of these, 22 subjects had atopic asthma, as shown by a positive skin-prick test for a panel of common allergens, and all had bronchial hyperresponsiveness, as demonstrated by the provocative concentration of histamine causing a 20% fall in the forced expiratory volume in 1 second of less than 8 mg per milliliter or peak flow variability of more than 20%. Five of the 10 subjects with COPD were studied during an infectious exacerbation to determine whether there was any contribution by infection to the involvement of invariant natural killer T cells, given the well-established role of these T cells in innate immunity.8 All subjects with COPD had disease ranging from stage 0 to stage 3, according to the criteria of the Global Initiative for Chronic Obstructive Lung Disease,12 and were either current or former heavy smokers; none had a history of asthma or other lung diseases. Exacerbation was defined on clinical and radiologic grounds, according to the guidelines of the British Thoracic Society.13
Collection and Preparation of the Samples
Sputum induction was performed according to the recommendations of the European Respiratory Society, and the samples were immediately processed with the use of dithioerythritol to separate cells from the fluid phase of sputum.14 Cell aliquots were immediately processed for flow-cytometric analysis.
Of the 24 subjects with asthma, 18 (5 with mild asthma and 13 with moderately severe asthma) underwent fiberoptic bronchoscopy for either bronchoalveolar lavage or bronchial biopsy, or both, in accordance with the recommendations of the American Thoracic Society.15 Cell aliquots from bronchoalveolar-lavage fluid were either subjected to analysis by flow cytometry or placed into lysis buffer for subsequent real-time PCR analysis. Specimens obtained on bronchial biopsy were immediately dispersed with the use of collagenase I (Sigma) reconstituted in RPMI 1640 without L-glutamine (Cambrex) at 1 mg per milliliter for a 1-hour period at 37°C (98.6°F).
Phenotypic Analysis
Cell-surface expression of the invariant natural killer T-cell receptor was detected on flow cytometry with the use of the seven-color FACSAria cell sorter (BD Biosciences) by applying the following monoclonal antibodies in combination with anti-CD3 and anti-CD4 monoclonal antibodies (BD Biosciences): R-phycoerythrin–conjugated anti-V24 monoclonal antibody, fluorescein isothiocyanate (FITC)–conjugated anti-V11 monoclonal antibody (Immunotech), R-phycoerythrin–conjugated CD1d tetramers loaded with -galactosylceramide and R-phycoerythrin–conjugated 6B11 monoclonal antibody directed against the complementarity-determining–region 3 of the V24-J18 T-cell receptor of invariant natural killer T cells (BD Biosciences).
Nonspecific binding of antibodies to Fc
receptors was blocked with 2 mg per milliliter of polyclonal human IgG (Sigma). Initially, cell doublets, which could contain adherent cells (e.g., macrophages) and dead cells and thus could bind antibodies nonspecifically, were excluded on gating with the use of pulse width and area (Figure 1, and Fig. E1 in the Supplementary Appendix). Dead cells were further excluded on the basis of their staining with the DNA-binding agent propidium iodide (Figure 1, and Fig. E2 in the Supplementary Appendix). Large cells and debris were excluded in the forward- and side-scatter plot (Figure 1). CD3+ T cells were gated positively on the basis of their staining with anti-CD3 antibody conjugated to allophycocyanin, a fluorescent dye that emits light at a wavelength beyond the range of autofluorescent cells (macrophages). As expected, these CD3+ cells were found in a tight group of cells with low granularity (black dots, or cells) (Figure 1, and Fig. E1 in the Supplementary Appendix). Finally, CD3+CD4+ T cells were gated on the basis of their staining with anti-CD4 antibody. (The gating strategy is described in detail in Fig. E1 through E6 in the Supplementary Appendix.)
|
We used oligonucleotide probes for real-time PCR of the genes encoding V24 (TRAV10, according to the International ImMunoGeneTics [IMGT] information system nomenclature) and V11 (TRBV25-1) as well as probes to detect expression of the constant chain of the T-cell receptor (TRBC2 gene) (PrimerDesign). The primers were aligned with other variable regions to confirm that there was no significant homology with any other variable region (for primer detection kits, see the Supplementary Appendix).
Results
Bronchoalveolar-Lavage Analysis
The initial flow-cytometric analyses for invariant natural killer T cells were performed on samples of bronchoalveolar-lavage fluid obtained from subjects with asthma, permitting direct comparisons with previous studies of invariant natural killer T cells in asthma4 and sarcoidosis.16 The analysis of samples obtained from 11 subjects with asthma (3 with mild asthma and 8 with moderately severe asthma) using the V24 and V11 monoclonal antibodies, which detected all invariant natural killer T cells,8,9,10 showed that invariant natural killer T cells constituted less than 2% of the CD3+ cells and less than 1.5% of the CD3+CD4+ cells (Table 1). Similar counts were obtained when we used CD1d tetramers loaded with -galactosylceramide17 or the 6B11 monoclonal antibody (Figure 1 and Figure 2 and Table 2). Routine total and differential cell counts of bronchoalveolar-lavage cells are shown in Table 3.
|
|
|
|
|
The next analysis for invariant natural killer T cells was performed on airway cells obtained by sputum induction,14 which permitted us to study a larger number of subjects than did the analysis of bronchoalveolar-lavage fluid. This method samples predominantly proximal airways, the sites of significant inflammation in asthma, which could, conceivably, be different from the distal airways sampled by bronchoalveolar lavage. We also used induced sputum to study controls and subjects with stable COPD and COPD during an infectious exacerbation. In preliminary experiments (Fig. E9 in the Supplementary Appendix), the reducing agent dithioerythritol, which is used to solubilize sputum and thereby make its analysis possible, was shown not to affect the expression of the invariant natural killer T-cell receptor on human invariant natural killer T-cell clones.
Between 0.2 and 14.3% of the sputum cells were CD3+, of which 27 to 91% were CD4+. As in the bronchoalveolar-lavage analysis, very few invariant natural killer T cells were identified in samples from subjects with mild asthma (0 to 0.1% of all CD3+CD4+ cells) and those with moderately severe asthma (0 to 1.3% of all CD3+CD4+ cells) (Figure 2 and Table 1). Similarly low counts of invariant natural killer T cells were found in sputum from controls and subjects with COPD (Table 1). Cell counts in sputum from subjects with mild asthma and those with moderately severe asthma were grouped for statistical comparison (with the use of the Mann–Whitney U test) with cell counts in subjects with COPD and controls. No significant differences were found between the groups (P>0.30 for the comparison between subjects with asthma and controls, and P>0.40 for the comparison between subjects with stable or exacerbated COPD and controls).
Bronchial-Biopsy Analysis
Because cells in mucosal tissue could differ from luminal cells, the final flow-cytometric analysis of invariant natural killer T cells was conducted on cells from bronchial-biopsy specimens with the use of enzymatic dispersion with collagenase. We used flow cytometry on dispersed bronchial biopsy-tissue cells to make possible a direct comparison with cells obtained on bronchoalveolar lavage and sputum induction. In addition, flow cytometry permitted the elimination of nonspecifically stained monocytes and macrophages. Because the CD4 epitope on T cells is cleaved during treatment with collagenase,18 the analysis was restricted to CD3+ T cells. In preliminary experiments (Fig. E9 in the Supplementary Appendix), collagenase treatment was shown not to affect the detection of the V24 and V11 T-cell receptors on invariant natural killer T-cell clones. Between 0.2 and 7.6% of all cells obtained on biopsy were CD3+ T cells. Consistent with the findings in samples of bronchoalveolar-lavage cells and sputum, counts of invariant natural killer T cells ranged from 0 to 1.7% of CD3+ T cells (Figure 3 and Table 1).
Expression of Messenger RNA for T-Cell Receptors V24 and V11
To corroborate the results obtained on flow cytometry with the use of a different method, we performed quantitative (real-time) PCR to detect the messenger RNA (mRNA) for the genes of the T-cell–receptor family, V24 (TRAV10) and V11 (TRVB25-1), on bronchoalveolar-lavage cells obtained from eight subjects with asthma and sputum cells obtained from four subjects with asthma, four with COPD, and seven controls. Amplified expression values for mRNA were normalized against signals obtained for the constant chain of the T-cell receptor encoding mRNA. The signal used to normalize the results was lower in the bronchoalveolar-lavage cells than in the invariant natural killer T-cell clones; titration experiments with the use of 10-fold serial dilution of the natural killer T-cell clones showed that the PCR detection kits for the T-cell receptors V24 and V11 and the constant chain were extremely sensitive, detecting as few as 10 cells (Fig. E10 in the Supplementary Appendix).
Strong signals for both the constant chain of the T-cell receptor and the mRNA of the individual invariant natural killer T-cell–receptor domains V24 and V11 were measured in positive control experiments with the use of complementary DNA of invariant natural killer T-cell clones as a template. In contrast, bronchoalveolar-lavage cells from subjects with asthma did not express mRNA for either the V24 or V11 gene, even though mRNA for the constant chain of the T-cell receptor was readily detected. This finding confirmed the presence of T cells in the samples, but it also ruled out the possibility that invariant natural killer T cells were numerous in bronchoalveolar-lavage fluid from subjects with asthma (Figure 3). Similar findings were observed with the use of sputum cells (Table E2 in the Supplementary Appendix).
In further experiments, real-time PCR for mRNA expression of the invariant natural killer T-cell receptor (V11) relative to the expression of the constant chain of the T-cell receptor was performed on peripheral blood T cells that had been "spiked" with invariant natural killer T-cell clones so that the samples contained 100%, 60%, and 2% of invariant natural killer T cells. This method generated a "standard curve" for the expression of the V11 gene (TRVB25-1) relative to the gene encoding the constant chain of the T-cell receptor. The result established that the proportion of T cells expressing T-cell receptor V11-encoding mRNA was less than 2% in all samples obtained by bronchoalveolar lavage and sputum induction (Figure 4).
|
Unlike some previous reports, our data show that invariant natural killer T cells are a minority cell population in human lungs of subjects with asthma, constituting less than 2% of the T cells in both the airway lumen (in samples of sputum and bronchoalveolar-lavage fluid) and mucosa (in samples obtained on bronchial biopsy). Moreover, the frequency of invariant natural killer T cells in samples obtained from subjects with asthma did not differ significantly from the frequency in samples obtained from healthy controls or from subjects with stable COPD or COPD during exacerbation.
Invariant natural killer T cells are characterized on the basis of their dual staining for the T-cell–receptor domains V24 and V11, positive staining with the use of CD1d tetramers loaded with -galactosylceramide, or both.8,9,10 In this study, we used both criteria in addition to using the 6B11 antibody directed against the invariant V24-J18 region of the invariant natural killer T-cell receptor. Our findings of low counts of invariant natural killer T cells were supported by our inability to detect mRNA for either T-cell–receptor gene V24 or V11 in samples of bronchoalveolar-lavage fluid or sputum from subjects with asthma (Figure 4, and Table E2 in the Supplementary Appendix). In addition, we were unable to demonstrate the production of Th2 cytokines by bronchoalveolar-lavage cells from subjects with asthma stimulated with -galactosylceramide (Supplementary Appendix), the specific stimulus for invariant natural killer T cells. In support of our findings, Thomas et al. found similarly low numbers of invariant natural killer T cells in subjects with asthma not receiving corticosteroids.7 Our review of two studies claiming that the numbers of invariant natural killer T cells were elevated in asthma showed that the strict criteria for detecting invariant natural killer T cells were not met in these studies.5,19 Although another study6 suggested that the numbers of invariant natural killer T cells were slightly elevated in children with severe asthma, the numbers were within the same range as those in our study.
Flow-cytometric analysis of airway T cells is fraught with difficulty because of large numbers of autofluorescent macrophages and cell debris.20,21 For this reason, we have excluded dead cells and cell doublets (Figure 1). Also, we used allophycocyanin as the fluorescent dye to avoid autofluorescence. Alveolar macrophages can nonspecifically bind both antibodies and CD1d tetramers loaded with -galactosylceramide used to identify invariant natural killer T cells, giving rise to further false positive results on staining. We show that the major fraction of all CD4+V24+ cells are identifiable in the nonlymphocyte portion of the forward- and side-scatter plot (Fig. E4 in the Supplementary Appendix) (a finding that can be explained by the presence of CD4 on macrophages) if these cells are not excluded by their typical forward- and side-scatter properties. In contrast, when all measures are used to avoid nonspecific staining and when combining anti-CD3 and anti-CD4 antibodies (Figure 1, and Fig. E4 in the Supplementary Appendix), CD4+ cells are restricted to the typical lymphocyte cluster in the forward- and side-scatter plot. Similar problems related to nonspecific binding are encountered with both anti-CD3 antibodies and CD1d tetramers loaded with -galactosylceramide (Fig. E5 and E6 in the Supplementary Appendix). Thomas et al. have proposed that the use of the 6B11 antibody, which recognizes the invariant complementarity-determining–region 3 loop of the invariant natural killer T-cell receptor, without using anti-V24 or anti-V11 antibodies is inappropriate.7 Therefore, we used a more comprehensive panel of antibodies, including 6B11 monoclonal antibody, anti-V24 monoclonal antibody, and anti-V11 monoclonal antibody, as well as CD1d tetramers loaded with the invariant natural killer T ligand -galactosylceramide, and all these consistently showed low counts of invariant natural killer T cells in all analyzed samples.
In summary, in contrast to the report by Akbari et al.,4 our study found very few invariant natural killer T cells in the airways of subjects with either mild or moderately severe asthma, subjects with stable COPD or COPD during an exacerbation, and healthy controls. This finding strongly questions the significance of invariant natural killer T cells in the pathogenesis of asthma. Our data suggest that therapeutic strategies in asthma should therefore continue to be focused on class II MHC–restricted T cells rather than on invariant natural killer T cells.
The study was funded by the University of Southampton.
Mr. Seumois reports receiving grants from the Société Belge de Pneumologie and the Union Chimique Belge Institute for Allergy; Dr. Gadola, consulting fees from Pfizer, lecture fees from Mepha Pharma and Robapharm, and grants from Novartis, Essex, Abbott, and Avidex; Dr. Friedmann, fees for serving on the international safety review panel of Astellas, consulting fees from Schering–Plough, and a grant from GlaxoSmithKline; and Dr. Djukanovi
We thank the staff of the Wellcome Trust Clinical Research Facility at Southampton General Hospital, where recruitment, sputum induction, and bronchoscopy were conducted, for their administrative and technical support; Dr. Karl Staples for reviewing the manuscript; and Dr. Laurie Lau for performing the enzyme-linked immunosorbent assay.
, consulting fees from Kyowa Pharmaceuticals, Trinity–Chiesi, and Shionogi, lecture fees from Novartis and AstraZeneca, and grants from GlaxoSmithKline and AstraZeneca. Dr. Powell is the founder of PrimerDesign, which made the primers for the real-time PCR analyses performed in the study, and Dr. Djukanovi is one of the three founders of Synairgen. No other potential conflict of interest relevant to this article was reported.
Source Information
From the Division of Infection, Inflammation, and Repair, University of Southampton, Southampton General Hospital (P.V., G.S., C.P., G.A., D.S., P.S.F., R.D.); and PrimerDesign, Roger Brooke Laboratory, University of Southampton, Southampton General Hospital (R.M.P.) — both in Southampton, United Kingdom; and University of Bern, Inselspital, Bern, Switzerland (S.D.G.).
Dr. Vijayanand and Mr. Seumois contributed equally to this article.
Address reprint requests to Dr. Vijayanand at the Inflammatory Cell Biology Group, Division of Infection, Inflammation, and Repair, Level F, South Block, Mail Point 810, Southampton General Hospital, Tremona Rd., Southampton SO16 6YD, United Kingdom, or at vijay{at}soton.ac.uk.
References
R, Wilson JW, Lai CK, Holgate ST, Howarth PH. The safety aspects of fiberoptic bronchoscopy, bronchoalveolar lavage, and endobronchial biopsy in asthma. Am Rev Respir Dis 1991;143:772-777. [ISI][Medline]
| |||||||||||||||||||||||||||||||||||||||||||||||||||
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
