Increased expression of immunoreactive thymic stromal lymphopoietin in patients with severe asthma

Article Outline

• Abstract
• Methods
  • Subjects
    • Leicester cohort
    • Belfast cohort
  • Fiberoptic bronchoscopy
  • Immunohistochemistry
  • Assessment and quantification of immunohistochemical staining
  • Gene expression analyses
• Results
  • Demographic data
  • Inflammatory cell infiltration (Leicester cohort)
  • TSLP immunoreactivity is increased in both the airway epithelium and lamina propria in patients with severe asthma
  • The TSLP–dendritic cell–OX40L axis is not evident in human asthmatic airway tissue
  • IL-4 and IL-13 expression in human asthmatic airway tissue
  • Relationships between TSLP and TH2 cytokine immunoreactivity
  • TSLP gene expression relative to eosinophilic inflammation and TH2 gene signature in biopsy tissue
• Discussion
• Acknowledgment
• Methods
  • Subjects
    • Leicester cohort
    • Belfast cohort
  • Fiberoptic bronchoscopy
  • Immunohistochemistry
  • Assessment and quantification of immunohistochemical staining
  • Gene expression analyses
• Results
  • Inflammatory cell infiltration (Leicester cohort)
• References
• References
• Copyright

Background

Thymic stromal lymphopoietin (TSLP) is a cytokine implicated in the pathophysiology of asthma through 2 distinct pathways: a TSLP–OX40 ligand (OX40L)–T cell axis and a TSLP–mast cell axis. Whether these pathways are active in human asthma is unknown.

Objective

We sought to investigate whether mucosal TSLP protein expression relates to asthma severity and distinct immunologic pathways.

Methods

In healthy subjects and patients with mild-to-severe asthma, we immunostained bronchial biopsy specimens for TSLP, OX40, OX40L, T_H2 cytokines, and inflammatory cell markers. We examined gene expression using RNA microarrays and quantitative RT-PCR.

Results

There was considerable heterogeneity in the levels of TSLP, IL-13, and IL-4 immunostaining across the cohort of asthmatic patients examined. Overall, TSLP protein expression was significantly increased in airway epithelium and lamina propria of asthmatic patients, particularly in patients with severe asthma. TSLP immunostaining in both compartments correlated with the severity of airflow obstruction. The majority of leukocytes expressing IL-13 were possibly nuocytes. Accounting for intersubject variability, the 55% of asthmatic patients with increased IL-13 immunostaining in the lamina propria also had increased IL-4 and TSLP expression. This was further substantiated by significant correlations between TSLP gene expression, a T_H2 gene expression signature, and eosinophilic inflammation in bronchial biopsy specimens. Immunostaining for OX40, OX40L, and CD83 was sparse, with no difference between asthmatic patients and healthy subjects.

Conclusion

TSLP expression is increased in a subset of patients with severe asthma in spite of high-dose inhaled or oral corticosteroid therapy. Targeting TSLP might only be efficacious in the subset of asthma characterized by increased TSLP expression and T_H2 inflammation.

Key words: Thymic stromal lymphopoietin, asthma, mast cell, OX40, IL-13, T_H2, airway remodeling, nuocyte

Abbreviations used: FVC, Forced vital capacity, GMA, Glycol methacrylate, HSI, Hue saturation and intensity, OX40L, OX40 ligand, qPCR, Quantitative real-time PCR, TSLP, Thymic stromal lymphopoietin

Discuss this article on the JACI Journal Club blog: www.jaci-online.blogspot.com.

Asthma is an important cause of morbidity and mortality. Approximately 10% of asthmatic patients have poorly controlled disease using current treatments; in consequence, novel approaches to treatment are required urgently.¹

The pathophysiology of severe asthma is poorly understood, and its study has been complicated by the recognition that asthma is a highly heterogeneous disease both clinically² and at the molecular level.³ For many years, research has focused on the T_H2 cell hypothesis, whereby it is proposed that T cell–derived T_H2 cytokines, such as IL-4, IL-5, and IL-13, promote airway mucosal eosinophilia and disordered airway function.⁴ These cytokines can contribute to disease pathogenesis in a proportion of patients, but a recent study of gene expression from asthmatic tissue suggests that up to 50% of patients do not exhibit evidence of T_H2-driven inflammation.³

Thymic stromal lymphopoietin (TSLP) is a cytokine released from airway epithelial cells in response to bacterial peptidoglycan, Toll-like receptor 3 stimulation, inflammatory cytokines, and IL-13.⁵, ⁶ Its overexpression in mice leads to asthma-type T_H2-driven airway changes, whereas its inhibition attenuates the airway changes seen in murine models of asthma.⁷ The current paradigm for the role of TSLP is that it activates a dendritic cell–T_H2 T-cell axis through an interaction between OX40 ligand (OX40L) and OX40.⁸, ⁹, ¹⁰ OX40L-OX40 interactions also contribute to the maintenance of ongoing T_H2 memory.¹¹

However, T cell–independent pathways also contribute to TSLP-dependent allergic-type airway inflammation. In particular, the direct interaction of TSLP with mast cells is highly relevant. TSLP derived from human epithelial cells induces the release of numerous cytokines and chemokines from human mast cells, including IL-13.⁶ Furthermore, in a murine model of rhinitis, mast cell activation was required for the increased TSLP production by airway epithelium.¹² Human mast cells are also able to produce TSLP.¹³ Taken together, these studies suggest the presence of 2 distinct TSLP-driven pathways that might drive asthma pathophysiology: a TSLP–OX40L–T-cell axis and a TSLP–mast cell axis.

Whether these pathways are operating in human asthma is unknown. Increased expression of TSLP mRNA has been reported in both the airway epithelium and lamina propria in a small group of asthmatic patients with disease of varying severity, although expression in relation to severity was not presented.¹⁴ A further study demonstrated increased concentrations of TSLP in bronchoalveolar lavage fluid from 13 patients with moderate-to-severe asthma compared with that seen in healthy control subjects.¹⁵ Whether the increased expression of TSLP mRNA and bronchoalveolar lavage protein extends to mucosal protein expression is not known. However, one study demonstrated increased numbers of mast cells expressing immunoreactive TSLP in the airway mucosa of patients with predominantly mild-to-moderate asthma.¹³

Therefore there are few data regarding TSLP protein expression in asthmatic airways and in particular how TSLP expression relates to downstream effector pathways, disease severity, and phenotype. The aim of our study was to examine the expression and cellular provenance of immunoreactive TSLP and the downstream cellular networks associated with TSLP overexpression in subjects with well-characterized asthma of varying severity.

Back to Article Outline

Methods 

Subjects 

Asthmatic patients (n = 36) and healthy volunteers (n = 12) were recruited for the study of both gene expression and immunohistochemistry. Full details and the assessments undertaken are described in the Methods section in this article’s Online Repository at www.jacionline.org. Asthma severity was defined by the “British guideline on the management of asthma” treatment steps (see the Methods section in this article’s Online Repository).¹⁶ Of the 16 patients with severe asthma at steps 4 and 5, 13 met the American Thoracic Society criteria for refractory asthma.¹⁷

For examination of airway gene expression and inflammatory cell infiltration, an additional cohort of subjects from Belfast was also studied. Participants with refractory asthma were recruited from the Belfast City Hospital, where systematic evaluation protocols ensure patients have well-characterized refractory asthma.¹⁸ Of the 9 patients at steps 4 and 5, 7 met the American Thoracic Society criteria for refractory asthma.¹⁷ Further details regarding patients with mild-to-moderate asthma are provided in the Methods section in this article’s Online Repository.

Demographic details for each cohort are shown in Tables E1 and E2 in this article’s Online Repository at www.jacionline.org. The study was approved by the research ethics committees of both institutions. Written informed consent was gained from all participants before their involvement.

Fiberoptic bronchoscopy 

Subjects underwent bronchoscopy conducted according to British Thoracic Society guidelines.¹⁹ Bronchial mucosal biopsy specimens were taken from the right middle lobe and lower lobe carinae, fixed in acetone, and embedded in glycol methacrylate (GMA), as described previously.²⁰ Biopsy specimens were also placed immediately in RNA preservative (RNAlater; Ambion, Austin, Tex) and submitted for microarray and quantitative real-time PCR (qPCR) analysis.

Immunohistochemistry 

GMA-embedded tissue was cut and immunostained, as described previously.²⁰ Primary antibodies were used against the following antigens: TSLP, mast cell tryptase, CD1a, CD83, IL-4, IL-13, CD68, CD3, CD4, CD8, neutrophil elastase, eosinophil major basic protein, OX40, and OX40L. Isotype controls were also performed. Full details of the antibodies used, including isotype, concentration, and source, are provided in the Methods section in this article’s Online Repository. Tonsil tissue obtained from Scottish Biomedical (Glasgow, United Kingdom) was embedded in GMA and used as a positive control.

Assessment and quantification of immunohistochemical staining 

Epithelial, submucosal, and smooth muscle bundle areas in sections were identified and measured by using a computer analysis system (analysis docu; Olympus, Hamburg, Germany). Numbers of positively stained nucleated cells in each compartment were counted blind. Cells staining in sequential sections were colocalized by using computer analysis.

Two methods were used for quantitative assessment of TSLP expression in the epithelium. First, a thresholding technique was developed based on the hue saturation and intensity (HSI) of TSLP staining. Further methodological details are provided in the Methods section in this article’s Online Repository. We also used a semiquantitative score of staining (0, none; 1, weak intermittent; 2, strong intermittent or weak generalized; and 3, strong generalized). There was a good correlation between the 2 methods (r_s = 0.837, P < .0001).

Prism 5 software (GraphPad Software, Inc, La Jolla, Calif) was used for statistical analysis. Group comparisons were made with the Kruskal-Wallis test; nonparametric data comparisons between groups were made with the Mann-Whitney U test, and unpaired t tests were used for parametric data. The spearman rank correlation test was used for correlations. A P value of less than .05 was considered statistically significant.

Gene expression analyses 

RNA was isolated from homogenized bronchial biopsy specimens, and qPCR was performed, as described previously.²¹ TaqMan Gene Expression Assays (Applied Biosystems, Foster City, Calif) were purchased and conducted per the manufacturer’s instructions for TSLP (id: Hs00263639_m1). Further details are provided in the Methods section in this article’s Online Repository. All gene expression analyses and plotting were performed with the R Project software package, version 2.10.1 (refer to http://www.R-project.org). The T_H2 score was calculated by using a generalized procedure²¹ and is described in detail in the Methods section in this article’s Online Repository.

Back to Article Outline

Results 

Demographic data 

The demographic data of the asthmatic patients from the Leicester and Belfast cohorts are shown in Tables E1 and E2.

Inflammatory cell infiltration (Leicester cohort) 

Numbers of epithelial and lamina propria mast cells, macrophages, eosinophils, neutrophils, and T cells are shown in Table E3 and discussed further in the Results section in this article’s Online Repository at www.jacionline.org.

TSLP immunoreactivity is increased in both the airway epithelium and lamina propria in patients with severe asthma 

Epithelial tissue for analysis was available from 11 healthy subjects and 5 patients with mild, 8 patients with moderate, and 14 patients with severe asthma. Lamina propria tissue was available for analysis from 12 healthy subjects and 6 patients with mild, 11 patients with moderate, and 15 patients with severe asthma. TSLP immunostaining in control human tonsil tissue showed the typical epithelial distribution described previously (see Fig E1, A, in this article’s Online Repository at www.jacionline.org).²² TSLP immunoreactivity was evident in both the airway epithelium and lamina propria of both healthy subjects and asthmatic patients, and this was inhibited by preincubating the TSLP antibody with recombinant human TSLP (Fig 1, A-D). TSLP immunostaining in the airway epithelium was localized predominantly to epithelial cells, although occasional mast cells, macrophages, and neutrophils were identified through colocalization in sequential sections. TSLP epithelial expression was significantly increased in asthmatic patients as a whole compared with that seen in healthy control subjects (P = .0005) whether measured semiquantitatively or based on the threshold intensity of staining. Subgroup analysis demonstrated significant increases in the epithelia of both patients with mild and patients with severe asthma compared with that seen in healthy control subjects by using the threshold assessment and in all asthmatic groups when scored semiquantitatively (Fig 1, E and F).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  TSLP immunostaining is increased in asthmatic patients. A and B, TSLP immunostaining in the airway epithelium from a patient with asthma (Fig 1, A) and inhibition by recombinant TSLP (Fig 1, B). C and D, TSLP immunostaining in the lamina propria from an asthmatic subject (Fig 1, C) and inhibition by recombinant TSLP (Fig 1, D). E and F, Extent of TSLP immunostaining in airway epithelium assessed based on threshold measurements and scoring, respectively. ∗P < .05, ∗∗P = .0001, and ∗∗∗P = .0014 compared with healthy control subjects. G, Extent of TSLP immunostaining in the lamina propria. ∗P = .0058 compared with healthy control subjects. Immunohistology is shown at ×400 magnification.

Numerous cells staining for TSLP were evident in the bronchial lamina propria (Fig 1, C). The number of TSLP⁺ cells in the lamina propria was increased in the asthma group as a whole (P = .043), but interestingly, this was accounted for by a significant increase only in the group of patients with severe asthma (P = .0058 compared with healthy control subjects; Fig 1, G). Both endothelial cell and mononuclear cell staining in the lamina propria was evident (Fig 1, C). Colocalization of cells in serial 2-μm sections from a subset of 6 patients with severe asthma showed that approximately 5% of TSLP⁺ cells in the lamina propria were CD68⁺ macrophages and 20% were tryptase-positive mast cells (see Fig E2 in this article’s Online Repository at www.jacionline.org). Endothelial cells identified based on morphology and location accounted for 42.0% ± 5.0% of TSLP⁺ cells. Neutrophils accounted for only occasional TSLP⁺ cells. There were significant inverse correlations between TSLP counts in both the asthmatic bronchial lamina propria and epithelium with the FEV₁/forced vital capacity (FVC) ratio (r_s = −0.53, P = .002 and r_s = −0.40, P = .037, respectively). There was also a significant correlation between lamina propria TSLP count and reliever use per week (r_s = 0.40, P = .045). No other significant correlations between TSLP staining and clinical parameters were evident.

Appropriate antibody isotype control sections were negative for TSLP immunostaining (see Fig E1, B and C).

The TSLP–dendritic cell–OX40L axis is not evident in human asthmatic airway tissue 

For validation of CD83, OX40, and OX40L immunostaining, human tonsil tissue was first examined and shown to exhibit immunoreactivity for each of these markers (see Fig E1, D-F). In the airways immunostaining for each of these antigens was sparse, and data from all asthma groups are presented together (see Table E4 in this article’s Online Repository at www.jacionline.org). There was a significant increase in the number of CD1a⁺ dendritic cells in the lamina propria in the asthma group as a whole (see Fig E1, G, and Table E4). There were no significant differences for CD83, OX40, or OX40L.

IL-4 and IL-13 expression in human asthmatic airway tissue 

Because of consumption of tissue, numbers of donors available for studying these 2 cytokines were reduced, particularly in the mild asthma group (see figure legends for details). The extent of IL-4 immunostaining in the airway epithelium was similar in all groups (see Fig E3, A, in this article’s Online Repository at www.jacionline.org) but was increased in the lamina propria in patients with mild asthma but not those with severe asthma, as described previously (see Fig E3, B).²³, ²⁴

IL-13⁺ cells were evident in both the airway epithelium and lamina propria, and the staining was inhibited by preincubation of the antibody with recombinant human IL-13 (Fig 2, A-C). Within the epithelium, both epithelial cells and nonepithelial cells were stained (Fig 2, D-F). The epithelial cell staining was discrete, intracellular, and located predominantly in the differentiated pseudocolumnar cells (Fig 2, A, C, and D). There was no difference in the extent of epithelial cell IL-13 immunoreactivity between asthmatic patients and healthy subjects (see Fig E3, C). The nonepithelial cells expressing IL-13 within the epithelium and lamina propria were mononuclear, with a distinct cell-surface staining pattern accompanied by punctuate intracellular staining (Fig 2, D and F). There was an increased number of IL-13⁺ cells that were classified as nonepithelial cells within the airway epithelium in the patients with severe asthma compared with that seen in the healthy control subjects (P = .033; Fig 2, G), but there were no differences across the study groups with respect to lamina propria IL-13 expression (Fig 2, H). Colocalization of cell markers in sequential 2-μm sections demonstrated that IL-13 was localized to less than 2% of mast cells and T cells and that these accounted for less than 4% of the IL-13⁺ nonepithelial cells. No IL-13 was colocalized to eosinophils. However, 70.2% ± 12.4% of the IL-13⁺ nonepithelial cells in the airway epithelium and 36.4% ± 2.9% of the IL-13⁺ cells in the lamina propria were CD45⁺ cells (Fig 2, E and F), suggesting they might represent the human equivalent of the recently described murine nuocyte.²⁵

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  IL-13 expression in the airway epithelium and lamina propria. A, C, D, and F, IL-13 immunoreactivity in the airway epithelium and lamina propria from 4 patients with severe asthma. In Fig 2, D, arrows demonstrate IL-13⁺ pseudocolumnar epithelial cells, and arrowheads highlight IL-13⁺ nonepithelial cells. B, An immunoabsorption control with recombinant human IL-13 in a section adjacent to Fig 2, A. E and F, Sequential sections demonstrating colocalization of IL-13 to CD45⁺ cells (arrows; ×400 magnification). G, The density of IL-13⁺ nonepithelial cells within the airway epithelium. ∗P = .033. H, The density of IL-13⁺ cells within the airway lamina propria.

Relationships between TSLP and T_H2 cytokine immunoreactivity 

Taken continuously, staining patterns for IL-4, IL-13, and TSLP did not show any significant intercorrelations (data not shown). However, emerging evidence suggests that there might be discrete molecular and pathophysiological phenotypes of asthma. Examination of IL-4 and IL-13 immunostaining in the lamina propria (see Fig E3, B, and Fig 2, H) revealed that these cytokines were detectable in only a subset of asthmatic patients in the study. Similarly, increased levels of TSLP staining in the lamina propria were observed only in a subset of patients with severe asthma. Therefore taken categorically as defined by the presence or absence of IL-13⁺ cells in the lamina propria, we find that the IL-13⁺ subjects show significant enhancement of both IL-4 (Fig 3, A) and TSLP (Fig 3, B) immunoreactivity in the lamina propria, suggesting that in terms of TSLP and T_H2 cytokine protein expression in bronchial tissue, there might be discrete “T_H2-high” and “T_H2-low” phenotypes of asthma.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  Relationships between IL-13, IL-4, and TSLP expression in the lamina propria in asthmatic patients. IL-13 immunoreactivity in the lamina propria was scored as absent or present; these categorical descriptions of IL-13 immunoreactivity were compared against IL-4 (A) and TSLP (B) immunoreactivity in the lamina propria. Gray symbols, Mild asthma; white symbols, moderate asthma; black symbols, severe asthma. ∗P = .014 and ∗∗P = .022.

TSLP gene expression relative to eosinophilic inflammation and T_H2 gene signature in biopsy tissue 

To determine whether the expression of TSLP is related to independent molecular metrics of T_H2 inflammation, we performed gene expression microarray analyses of bronchial biopsy tissue. We assigned scores to individual subjects on the basis of a T_H2 inflammation gene signature, as we have previously described for a separate cohort of bronchial biopsy specimens from asthmatic patients.²¹ There was a continuum of T_H2 signature intensity across the subjects in this cohort (Fig 4, A), although it was somewhat attenuated relative to the previous study, likely because of the variable effects of inhaled corticosteroid treatment on T_H2 cytokine–induced gene expression in asthmatic airways.³ We performed qPCR to precisely measure the level of TSLP expression in the same biopsy samples. TSLP gene expression was significantly correlated with lamina propria TSLP protein expression (r = 0.43, P = .026, n = 27) and with the magnitude of the T_H2 signature and lamina propria eosinophil counts (Fig 4, B-D).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  Bronchial biopsy specimen gene expression of TSLP, T_H2 signature score, and eosinophil count are intercorrelated. A, A quantitative score for T_H2 signature score was calculated from gene expression microarrays (see the Methods section in this article’s Online Repository). Relative expression is represented by means of heat map, where patient samples (columns) are organized left to right by increasing the T_H2 signature score, as indicated by the adjacent bar plot (top). Genes (rows) are organized by principal component analysis loadings as indicated by adjacent dot plots (left). B-D, All pairwise comparisons of bronchial biopsy TSLP expression, T_H2 signature scores, and eosinophil counts are represented by scatterplots (Spearman rank, all comparisons P < .05).

Back to Article Outline

Discussion 

We have shown that expression of TSLP protein is upregulated in the airways of patients with asthma and that this persists in a cohort of patients with severe asthma despite high-dose corticosteroid treatment at steps 4 and 5 of the British asthma treatment guidelines. TSLP expression in the airway epithelium was increased across the spectrum of disease severity, although to a lesser extent in patients with relatively well-controlled moderate asthma. However, in the airway lamina propria, increased expression was confined to the group with severe disease. These findings suggest TSLP might be a key mediator of severe chronic asthma and that its anatomic site of expression might be related to asthma severity.

Our study extends previous work demonstrating increased TSLP mRNA expression in asthmatic airway mucosa and increased TSLP protein expression in the bronchoalveolar lavage fluid of 13 patients with moderate-to-severe asthma. However, our findings extend our understanding of the potential roles of TSLP in asthmatic patients through the coexamination of downstream TSLP-dependent cellular networks. Animal models propose that TSLP contributes to the immunopathology of asthma through the activation of the OX40L/OX40 pathway, leading to the generation of T_H2 T cells. There has therefore been great interest in targeting the OX40L/OX40 pathway in asthmatic patients.²⁶ However, despite clear OX40 and OX40L staining in human tonsil control tissue, staining for these cellular markers in the airways was notably sparse, and we could not find any evidence for increased OX40L or OX40 expression in asthmatic patients. Another study suggested that OX40 and OX40L expression was increased in patients with mild asthma compared with that seen in healthy subjects and patients with severe asthma, but the numbers of cells present were very small.²⁷ In parallel, we found no significant increase in the number of dendritic cells displaying the activation marker CD83 in asthmatic patients. Our data therefore suggest that OX40L and OX40 are unlikely to play significant roles in persistent airway inflammation in the bronchial mucosa. However, given the current understanding of the dynamics of dendritic cell–mediated T-cell stimulation, it remains possible that OX40L/OX40 interactions take place in secondary lymphoid organs, such as mediastinal lymph nodes, which could not be sampled in this study.

TSLP also activates mast cell IL-13 production,⁶ and in turn, mast cell activation induces epithelial TSLP production.¹² Importantly, in a murine model of allergic rhinitis, mast cell–dependent regulation of epithelial TSLP secretion was a key determinant of disease expression.¹² The presence of activated mast cells within and in close proximity to the bronchial epithelium in asthmatic patients²³, ²⁸ suggests that mast cells might regulate TSLP activity very closely in patients with this disease. Conversely, epithelial TSLP production in response to diverse innate stimuli might regulate mast cell IL-13 production. This would be in keeping with recent hypotheses that a key factor driving chronic asthma is the activation of innate immune pathways.⁴

Mast cells are also a source of TSLP.¹³, ¹⁵ We have confirmed this and found that 5% of mast cells in patients with severe asthma express TSLP protein, which accounted for 20% of TSLP⁺ cells in the lamina propria. These findings are in broad agreement with the data of Ying et al¹⁵ with respect to TSLP mRNA expression but in marked disagreement with the data of Okayama et al,¹³ who studied TSLP protein. Ying et al¹⁵ also demonstrated TSLP mRNA expression in endothelial cells, neutrophils, CD68⁺ macrophages, and CD3⁺ T cells. We identified endothelial cell TSLP immunoreactivity by means of morphology and have confirmed the presence of TSLP protein within CD68⁺ cells but have not investigated T-cell colocalization.

Ying et al¹⁴ found a significant inverse correlation between TSLP mRNA expression in both the airway epithelium and lamina propria with FEV₁ percent predicted. We did not find a significant correlation with FEV₁ percent predicted but did find significant inverse correlations between TSLP protein expression in both the airway epithelium and lamina propria with the FEV₁/FVC ratio. Because there is wide variability between subjects in predicted FEV₁,²⁹ the FEV₁/FVC ratio provides a more accurate measure of airflow obstruction. This finding suggests that in addition to promoting airway inflammation, TSLP overproduction might play an important role in the remodeling of asthmatic airways and the subsequent development of fixed airflow obstruction.

In keeping with previous work in asthmatic patients, IL-4 expression was suppressed in patients with severe asthma compared with that seen in patients with mild asthma.²⁴ Interestingly, IL-13 protein was detected in the airway epithelium in patients with severe asthma in cells that were not epithelial cells but was also detected in columnar epithelial cells. This latter finding is novel and was not described in previous studies of IL-13 expression in asthmatic patients using a different anti–IL-13 antibody.³⁰, ³¹ Our current study used an in-house–generated IL-13 rabbit polyclonal antibody. Specificity of immunostaining was confirmed by using isotype controls and immunoabsorption with recombinant human IL-13, suggesting that the staining is unlikely to be artifactual. Two recent studies have described the expression of IL-13 mRNA and secretion of IL-13 protein in submerged monolayers of primary human airway epithelial cells, suggesting that they might produce this cytokine.³², ³³ An alternate explanation for our immunostaining is that IL-13 within airway epithelial cells represents IL-13 that has been taken up rather than synthesized. The cellular distribution of IL-13 in our current study was different than that described previously³¹ in that most nonepithelial IL-13⁺ cells in the airway epithelium and lamina propria were not mast cells, T cells, or eosinophils. However, the majority were CD45⁺ cells, suggesting that they are the human equivalent of the recently described CD45⁺ murine nuocyte.²⁵ Nuocytes are lineage-negative innate immune cells that provide an early source of IL-13 during murine infection with parasites. These cells might therefore contribute to the pathophysiology of chronic severe asthma and, through their presence within both the airway epithelium and lamina propria, contribute to the upregulation of TSLP expression.

Increased IL-13 expression in the nonepithelial cells within the epithelial compartment in patients with severe asthma supports the previous observation that IL-13 protein concentrations in induced sputum are increased in a subgroup of patients with severe refractory asthma.³¹ However, it is again evident that there is a subgroup of patients with severe asthma in whom IL-13 expression persists and a group in whom it is suppressed to levels similar to those seen in healthy subjects. This has important implications for studies examining the efficacy of anti–IL-13 therapy in patients with severe asthma. In the IL-13–low group, it is unlikely that IL-13 will be driving their disease, and anti–IL-13 therapy might be ineffective. Therefore it might be necessary to target this therapy to patients in whom there is evidence of ongoing IL-13 expression.

We observed considerable heterogeneity in the levels of TSLP, IL-13, and IL-4 staining across the cohort of asthmatic patients examined, which is consistent with previous reports demonstrating subsets of asthmatic patients with variable levels of T_H2 inflammation.³ In general, subjects with increased IL-13 immunostaining in the lamina propria also had increased IL-4 and TSLP immunostaining. This observation is further substantiated by a significant correlation between TSLP gene expression, a T_H2 gene expression signature, and eosinophilic inflammation in bronchial biopsy specimens. Taken together with the established roles of TSLP in mediating both adaptive and innate IL-13–driven inflammation, these findings suggest that TSLP is indeed associated with the “T_H2-high” subphenotype of asthma.

In summary, our study demonstrates increased expression of TSLP in asthmatic patients, particularly in patients with severe disease, and this correlates with the severity of airflow obstruction. We cannot find evidence to support the presence of a TSLP–OX40L–dendritic cell axis in ongoing disease, but our results are compatible with the presence of a TSLP–mast cell and TSLP-nuocyte axis. The finding of increased TSLP expression in patients with severe asthma suggests that TSLP is an attractive target for the development of novel antiasthma therapies but that similar to other cytokine targets, such as IL-5 and IL-13, targeting TSLP might only be efficacious in the subset of asthma characterized by increased TSLP expression and T_H2 inflammation. Our data suggest that targeting TSLP in patients with severe asthma has the potential to inhibit both airway inflammation and remodeling and might therefore offer a new approach with which to address the unmet clinical need in this group of patients.

Back to Article Outline

We thank Dr J. Bennett, Professor C. Brightling, and Dr T. J. Warke for assistance with bronchoscopy.

Back to Article Outline

Methods 

Subjects 

For the study of both gene expression and immunohistochemistry, asthmatic patients (n = 36) and healthy volunteers (n = 12) were recruited from respiratory clinics, including the Leicester Difficult Asthma Clinic; from staff at Glenfield Hospital; and from the general population through advertisement (Leicester cohort). Patients with asthma provided a suggestive history and had objective evidence of variable airflow obstruction, as indicated by 1 or more of the following: (1) methacholine airway hyperresponsiveness (PC₂₀FEV₁, <8 mg/mL), (2) greater than 15% improvement in FEV₁ 10 minutes after 200 μg of inhaled salbutamol, and (3) peak expiratory flow (>20% maximum within-day amplitude from twice-daily peak expiratory flow measurements over a period of 14 days). Subjects underwent spirometry; allergen skin prick tests for Dermatophagoides pteronyssinus, dog, cat, grass pollen, and Aspergillus fumigatus; a methacholine inhalation test using the tidal breathing method; and sputum induction with incremental concentrations of nebulized hypertonic saline (ie, 3%, 4%, and 5%, each for 5 minutes).^E1 Patients with asthma also kept a diary card for 2 weeks before bronchoscopy, recording daytime and nighttime symptoms, daily short-acting β₂-agonist use, and twice-daily peak expiratory flow. Asthma severity was defined based on “British guideline on the management of asthma” treatment steps (mild = step 1, β₂-agonist only; moderate = steps 2 and 3, inhaled corticosteroid ≤800 mg of beclomethasone equivalent per day ± long-acting β₂-agonist; and severe = steps 4 and 5).^E2 Of the 16 patients with severe asthma at steps 4 and 5, 13 met the American Thoracic Society criteria for refractory asthma.^E3

For examination of airway gene expression and inflammatory cell infiltration, an additional cohort of subjects from Belfast was also studied. Participants with refractory asthma were recruited from the Belfast City Hospital, where systematic evaluation protocols ensure patients have well-characterized refractory asthma.^E4 All participants had persisting symptoms despite treatment at steps 4 or 5 of the Global Initiative for Asthma guidelines. In addition, mild-to-moderate asthmatic and healthy participants were recruited either from hospital clinic or by advertisement. Subjects with asthma had a clinical diagnosis of asthma with a current history of recurrent wheezing and documented response to asthma medication. Healthy volunteers had no history of asthma or persistent respiratory symptoms and normal lung function.

In both cohorts all participants were current nonsmokers with no upper or lower respiratory tract infections in the 6 weeks before bronchoscopy. All asthmatic participants’ symptoms were clinically stable on their usual medication at the time of bronchoscopy. Demographic details for each cohort are shown in Tables E1 and E2.

The study was approved by the Research Ethics Committee of both institutions (Leicester: Leicestershire, Northamptonshire, & Rutland Research Ethics Committee reference 04/Q2502/74; Belfast: Office of Research and Ethics Committee of Northern Ireland reference 06/NIR02/114). Written informed consent was gained from all participants before their involvement.

Fiberoptic bronchoscopy 

Subjects underwent bronchoscopy conducted according to the British Thoracic Society guidelines.^E5 Bronchial mucosal biopsy specimens were taken from the right middle lobe and lower lobe carinae, fixed in acetone, and embedded in GMA, as described previously.^E6 Biopsy specimens were also placed immediately in RNA preservative (RNAlater, Ambion) and submitted for microarray and qPCR analysis.

Immunohistochemistry 

GMA-embedded tissue was cut and stained as described previously.^E6 The following primary antibodies were used (murine mAbs unless otherwise stated): sheep polyclonal anti-TSLP (catalogue no. AF1398, 7.5 μg/mL; R&D Systems [Minneapolis, Minn], Abingdon, United Kingdom), anti–mast cell tryptase clone AA1 (1:1000 dilution; Dako UK, Ely, United Kingdom), anti-CD1a clone 010 (26 μg/mL, Dako), anti-CD83 clone HB15e (1 μg/mL; AbD Serotec, Oxford, United Kingdom), anti–IL-4 clone 3H4 (12.5 μg/mL; a gift from Dr Chris Heusser, Novartis, Switzerland), rabbit polyclonal anti–IL-13 (5 μg/mL, in-house generated; Genentech, San Francisco, Calif), anti-CD68 clone PG-M1 (Dako), anti-CD3 clone UCHT1 (3 μg/mL; BD Biosciences, Oxford, United Kingdom), anti-CD4 clone MT310 (10 μg/mL, Dako), anti-CD8 clone DK25 (0.5 μg/mL, Dako), anti–neutrophil elastase clone NP57 (0.1 μg/mL, Dako), anti–eosinophil major basic protein clone BMK-13 (0.4 μg/mL; Monosan, Uden, The Netherlands), anti-OX40 clone ACT35 (5 μg/mL, BD Biosciences), anti-OX40L clone 159403 (10 μg/mL), and appropriate isotype controls (R&D, Dako, and BD Biosciences). The sections were counterstained with Mayer hematoxylin. Tonsil tissue obtained from Scottish Biomedical was embedded in GMA and used as a positive control.

Assessment and quantification of immunohistochemical staining 

Epithelial, submucosal, and smooth muscle bundle areas in sections were identified and measured by using a computer analysis system (analysis docu). Numbers of positively stained nucleated cells in each compartment were counted and expressed per square millimeter. All counts were performed blind to clinical characteristics. Areas of less than 0.1 mm² were considered insufficient to quantify. Cells staining in sequential sections were colocalized by using the computer analysis system.

Two methods were used for quantitative assessment of TSLP expression in the epithelium. First, a thresholding technique was developed based on the HSI of TSLP staining. The HSI color system was defined by a scale of 0 to 255 for HSI. Sections were acquired at ×20 magnification and white balance corrected, and pixels of representative TSLP staining selected in the epithelium were measured. A minimum of 2 noncontiguous sections were measured for each patient. The median of the lower and upper limit of the HSI was then defined from 10 validation patients. The HSI was then selected as the final threshold to produce a threshold that appropriately captured highly saturated red light. All biopsy specimens were subsequently measured by using this final threshold. The mean percentage area of TSLP staining in 10 areas in the epithelium from 2 sections was taken as the final measurement. We also tested the final threshold measurement using comparison with a semiquantitative score of staining (0, none; 1, weak intermittent; 2, strong intermittent or weak generalized; and 3, strong generalized). We found there was a good correlation between the 2 methods (r_s = 0.837, P < .0001).

Prism 5 software was used for statistical analysis of immunohistochemical staining. Group comparisons were made by using the Kruskal-Wallis test; nonparametric data comparisons between groups were made with the Mann-Whitney U test, and unpaired t tests were used for parametric data. The spearman rank correlation test was used for correlations. A P value of less than .05 was considered statistically significant.

Gene expression analyses 

RNA was isolated from homogenized bronchial biopsy specimens, and qPCR was performed as described previously.^E7 TaqMan Gene Expression Assays (Applied Biosystems, Foster City, Calif) were purchased and conducted per the manufacturer’s instructions for TSLP (id: Hs00263639_m1). RNA was amplified (Ambion) for Agilent (Santa Clara, Calif) 2-color Whole Human Genome 4x44k gene expression microarray analysis. Universal Human Reference RNA (Stratagene, La Jolla, Calif) was used for the reference channel. Probe intensities were transformed as log₂ ratios of test and reference channels calculated by using the Agilent Feature Extraction software, protocol GE2-v5_95 (Agilent). Flagged outliers were not included in any subsequent analyses. All gene expression analyses and plotting were performed by using the R Project software package, version 2.10.1 (refer to http://www.R-project.org).

The T_H2 score was calculated by using a generalized procedure.^E7 Thirty-eight signature genes were selected based on array features corresponding to T_H2 signature genes with Entrez gene annotation that were previously observed to be upregulated in T_H2-high subjects were selected. If individual Entrez genes mapped to multiple probes, the probe with the greatest interquartile range was retained, and any others were removed. Gene expression values were mean centered. Missing T_H2 signature gene set values (5/2622 [0.19%]) were substituted by using mean replacement. Principal component analysis was conducted on these expression values, and PC1 (which retained 39.5% of the variance of the signature gene set) was used as the T_H2 score.

qPCR values were calculated by using the ΔΔ method and are expressed as log₂-transformed values.

Back to Article Outline

Results 

Inflammatory cell infiltration (Leicester cohort) 

The number of epithelial and lamina propria tryptase-positive mast cells, CD68⁺ macrophages, major basic protein–positive eosinophils, elastase-positive neutrophils, and T cells are shown in Table E3. Of note, mast cell numbers were increased significantly in the airway epithelium in patients with mild asthma, which is consistent with previous reports describing intraepithelial mast cells associated with T_H2 inflammation in patients with mild asthma.^E8., ^E9. They were also increased in the airway smooth muscle in asthmatic patients (median, 4.0; interquartile range, 0.8-9.0) compared with those seen in healthy control subjects (median, 0.0; interquartile range, 0-1.6; P = .029), but there was no difference across asthma severity. Eosinophil counts were increased in the airway lamina propria of asthmatic patients compared with those seen in healthy subjects, as accounted for by significant increases in the mild group (Table E3). However, 4 patients in each of the groups with moderate and severe asthma had lamina propria eosinophil counts higher than those in the mild asthma group, suggesting the presence of steroid-refractory eosinophilic inflammation. Lamina propria and epithelial eosinophil counts correlated significantly within the asthmatic population (r_s = 0.45, P = .018), but consistent with previous reports,^E10 neither correlated with sputum eosinophil counts. CD68⁺ macrophage density was increased in the airway epithelium of patients with mild asthma compared with that seen in healthy control subjects but not between asthmatic patients and healthy subjects as a whole (Table E3).

In keeping with previous studies of asthma,^E8 there was no significant difference in the number of epithelial or lamina propria CD3⁺ or CD4⁺ T cells in asthmatic patients compared with that seen in healthy subjects (Table E3). There was a significant increase in the number of CD8⁺ T cells in the lamina propria of patients with mild asthma, which was of uncertain clinical significance (Table E3).

• 
  • View Large Image
  • Download to PowerPoint

• Fig E1. 

  A and B, Positive control immunostaining for TSLP in human tonsil tissue (Fig E1, A) and negative immunostaining in tonsil tissue with sheep IgG isotype control (Fig E1, B; 7.5 μg/mL). C, Negative immunostaining with sheep IgG isotype control in a bronchial biopsy specimen from a patient with severe asthma. D-F, Immunostaining in tonsil tissue for CD83, OX40, and OX40L, respectively. G, Immunostaining for CD1a in normal bronchial epithelium and lamina propria. H, Immunostaining for CD83 in normal bronchial lamina propria.

• 
  • View Large Image
  • Download to PowerPoint

• Fig E2. 

  Colocalization of TSLP to mast cells and macrophages within the lamina propria. A, Two sequential 2-μm sections demonstrating colocalization of mast cell tryptase and TSLP (arrows). Lines demonstrate local landmarks for reference. B, The percentage of macrophages and mast cells expressing TSLP and vice versa in a subset of 6 patients with severe asthma. Immunohistology is shown at ×400 magnification.

• 
  • View Large Image
  • Download to PowerPoint

• Fig E3. 

  A and B, The density of IL-4⁺ cells in the bronchial epithelium and lamina propria, respectively. C, The density of IL-13⁺ epithelial cells within the airway epithelium.

Table E1. Demographic data of the Leicester asthmatic subjects according to disease severity

Significant values are shown in boldfaced text.

BTS, British Thoracic Society; NA, not applicable; PEF, peak expiratory flow.

Table E2. Demographic data of the Belfast asthmatic subjects according to disease severity

BTS, British Thoracic Society; NA, not applicable; ND, not done.

Table E3. Inflammatory cell counts in the airway epithelium and lamina propria in patients from the Leicester cohort

	Healthy control subjects	Patients with mild asthma	Patients with moderate asthma	Patients with severe asthma
Epithelium: cells/mm2
Tryptase	2.8 (0-7.1)	31.7∗ (16.7-40.1)	9.2 (4.1-18.1)	3.2 (1.3-10.4)
MBP	0.0 (0-10.0)	13.5 (4.4-18.8)	1.7 (0.0-5.7)	3.2 (0.0-14.0)
CD68	0.0 (0-6.7)	10.0∗ (6.3-23.1)	2.4 (0.0-4.8)	15.3 (0.0-31.4)
Neutrophil elastase	2.8 (0-16.7)	0.0 (0.0-4.7)	3.3 (0.0-9.2)	5.6 (2.8-10.8)
CD3	33.3 (11.1-70.2)	60.0 (16.3-132.7)	31.1 (1.3-100.1)	44.7 (8.2-99.4)
CD4	3.3 (2.8-13.9)	10.0 (1.5-32.7)	5.0 (0.0-21.2)	9.8 (0.0-43.3)
CD8	14.7 (8.3-76.7)	30.2 (13.2-103.8)	9.2 (0.0-52.6)	29.1 (2.3-59.1)
Lamina propria: cells/mm2
Tryptase	14.5 (10.6-19.0)	12.5 (6.8-26.9)	15.1 (8.9-31.4)	8.4 (5.4-15.6)
MBP	2.4 (1.4-5.6)	11.2∗ (9.8-12.7)	4.4 (2.7-17.5)	5.3 (1.3-24.4)
CD68	4.0 (2.3-5.1)	4.7 (3.6-6.5)	2.7 (1.4-7.2)	4.5 (1.4-11.2)
Neutrophil elastase	9.4 (3.7-20.2)	3.4 (1.1-8.6)	8.4 (3.0-17.1)	8.9 (1.9-15.7)
CD3	31.0 (12.0-45.8)	41.0 (32.2-49.6)	19.9 (9.5-29.3)	30.0 (18.4-43.2)
CD4	9.7 (5.4-14.6)	11.5 (8.9-21.6)	7.22 (2.1-11.9)	11.7 (5.1-15.7)
CD8	20.3 (10.9-31.0)	30.2∗ (26.8-37.1)	11.3 (6.1-18.6)	13.3 (7.4-26.1)

Values are presented as medians (interquartile ranges).

MBP, Major basic protein.

Table E4. Counts for cells expressing CD1a, CD83, OX40, and OX40L in the airway epithelium and airway lamina propria in patients from the Leicester cohort

	Healthy control subjects	Asthmatic patients
Epithelial cells/mm2
CD1a	1.3 (0.0-7.4)	0.0 (0.0-0.0)
CD83	0.0 (0.0-2.8)	2.1 (0.0-6.0)
OX40	0.0 (0.0-0.0)	0.0 (0.0-0.0)
OX40L	0.0 (0.0-0.0)	0.0 (0.0-1.0)
Lamina propria cells/mm2
CD1a	0.0 (0.0-1.2)	0.6∗ (0.0-1.2)
CD83	1.2 (0.6-2.2)	1.8 (0.8-3.1)
OX40	0.0 (0.0-0.8)	0.6 (0.4-0.7)
OX40L	0.1 (0.0-0.6)	0.4 (0.0-0.9)

Values are presented as medians (interquartile ranges).

Back to Article Outline

References 

1. Wenzel S. Severe asthma in adults. Am J Respir Crit Care Med. 2005;172:149–160
   • View In Article
   • CrossRef
2. Haldar P, Pavord ID, Shaw DE, Berry MA, Thomas M, Brightling CE, et al. Cluster analysis and clinical asthma phenotypes. Am J Respir Crit Care Med. 2008;178:218–224
   • View In Article
   • CrossRef
3. Woodruff PG, Modrek B, Choy DF, Jia G, Abbas AR, Ellwanger A, et al. T-helper type 2-driven inflammation defines major subphenotypes of asthma. Am J Respir Crit Care Med. 2009;180:388–395
   • View In Article
   • CrossRef
4. Anderson GP. Endotyping asthma: new insights into key pathogenic mechanisms in a complex, heterogeneous disease. Lancet. 2008;372:1107–1119
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (217 KB)
   • CrossRef
5. Kato A, Favoreto S, Avila PC, Schleimer RP. TLR3- and Th2 cytokine-dependent production of thymic stromal lymphopoietin in human airway epithelial cells. J Immunol. 2007;179:1080–1087
   • View In Article
6. Allakhverdi Z, Comeau MR, Jessup HK, Yoon BR, Brewer A, Chartier S, et al. Thymic stromal lymphopoietin is released by human epithelial cells in response to microbes, trauma, or inflammation and potently activates mast cells. J Exp Med. 2007;204:253–258
   • View In Article
   • MEDLINE
   • CrossRef
7. Zhou B, Comeau MR, De ST, Liggitt HD, Dahl ME, Lewis DB, et al. Thymic stromal lymphopoietin as a key initiator of allergic airway inflammation in mice. Nat Immunol. 2005;6:1047–1053
   • View In Article
   • MEDLINE
   • CrossRef
8. Reche PA, Soumelis V, Gorman DM, Clifford T, Mr Liu, Travis M, et al. Human thymic stromal lymphopoietin preferentially stimulates myeloid cells. J Immunol. 2001;167:336–343
   • View In Article
   • MEDLINE
9. Ito T, Wang YH, Duramad O, Hori T, Delespesse GJ, Watanabe N, et al. TSLP-activated dendritic cells induce an inflammatory T helper type 2 cell response through OX40 ligand. J Exp Med. 2005;202:1213–1223
   • View In Article
   • MEDLINE
   • CrossRef
10. Seshasayee D, Lee WP, Zhou M, Shu J, Suto E, Zhang J, et al. In vivo blockade of OX40 ligand inhibits thymic stromal lymphopoietin driven atopic inflammation. J Clin Invest. 2007;117:3868–3878
    • View In Article
    • CrossRef
11. Wang YH, Ito T, Wang YH, Homey B, Watanabe N, Martin R, et al. Maintenance and polarization of human TH2 central memory T cells by thymic stromal lymphopoietin-activated dendritic cells. Immunity. 2006;24:827–838
    • View In Article
    • MEDLINE
    • CrossRef
12. Miyata M, Hatsushika K, Ando T, Shimokawa N, Ohnuma Y, Katoh R, et al. Mast cell regulation of epithelial TSLP expression plays an important role in the development of allergic rhinitis. Eur J Immunol. 2008;38:1487–1492
    • View In Article
    • CrossRef
13. Okayama Y, Okumura S, Sagara H, Yuki K, Sasaki T, Watanabe N, et al. FcepsilonRI-mediated thymic stromal lymphopoietin production by interleukin-4-primed human mast cells. Eur Respir J. 2009;34:425–435
    • View In Article
    • CrossRef
14. Ying S, O’Connor B, Ratoff J, Meng Q, Mallett K, Cousins D, et al. Thymic stromal lymphopoietin expression is increased in asthmatic airways and correlates with expression of Th2-attracting chemokines and disease severity. J Immunol. 2005;174:8183–8190
    • View In Article
    • MEDLINE
15. Ying S, O’Connor B, Ratoff J, Meng Q, Fang C, Cousins D, et al. Expression and cellular provenance of thymic stromal lymphopoietin and chemokines in patients with severe asthma and chronic obstructive pulmonary disease. J Immunol. 2008;181:2790–2798
    • View In Article
16. British guideline on the management of asthma. Thorax. 2008;63:iv1–iv121
    • View In Article
    • CrossRef
17. Proceedings of the ATS Workshop on Refractory Asthma. Current understanding, recommendations, and unanswered questions. Am J Respir Crit Care Med. 2000;162:2341–2351
    • View In Article
18. Heaney LG, Conway E, Kelly C, Johnston BT, English C, Stevenson M, et al. Predictors of therapy resistant asthma: outcome of a systematic evaluation protocol. Thorax. 2003;58:561–566
    • View In Article
    • MEDLINE
    • CrossRef
19. British Thoracic Society guidelines on diagnostic flexible bronchoscopy. Thorax. 2001;56:i1–i21
    • View In Article
20. Bradding P, Feather IH, Howarth PH, Mueller R, Roberts JA, Britten K, et al. Interleukin 4 is localized to and released by human mast cells. J Exp Med. 1992;176:1381–1386
    • View In Article
    • MEDLINE
    • CrossRef
21. Choy DF, Modrek B, Abbas AR, Kummerfeld S, Clark HF, Wu LC, et al. Gene expression patterns of Th2 inflammation and intercellular communication in asthmatic airways. J Immunol. 2011;186:1861–1869
    • View In Article
    • CrossRef
22. Soumelis V, Reche PA, Kanzler H, Yuan W, Edward G, Homey B, et al. Human epithelial cells trigger dendritic cell mediated allergic inflammation by producing TSLP. Nat Immunol. 2002;3:673–680
    • View In Article
    • MEDLINE
23. Bradding P, Roberts JA, Britten KM, Montefort S, Djukanovic R, Mueller R, et al. Interleukin-4, -5, and -6 and tumor necrosis factor-alpha in normal and asthmatic airways: evidence for the human mast cell as a source of these cytokines. Am J Respir Cell Mol Biol. 1994;10:471–480
    • View In Article
    • MEDLINE
24. Vrugt B, Wilson S, Underwood J, Bron A, de BR, Bradding P, et al. Mucosal inflammation in severe glucocorticoid-dependent asthma. Eur Respir J. 1999;13:1245–1252
    • View In Article
    • MEDLINE
    • CrossRef
25. Neill DR, Wong SH, Bellosi A, Flynn RJ, Daly M, Langford TK, et al. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature. 2010;464:1367–1370
    • View In Article
    • CrossRef
26. Rochman Y, Leonard WJ. Thymic stromal lymphopoietin: a new cytokine in asthma. Curr Opin Pharmacol. 2008;8:249–254
    • View In Article
    • CrossRef
27. Siddiqui S, Mistry V, Doe C, Stinson S, Foster M, Brightling C. Airway wall expression of OX40/OX40L and interleukin-4 in asthma. Chest. 2010;137:797–804
    • View In Article
    • CrossRef
28. Dougherty RH, Sidhu SS, Raman K, Solon M, Solberg OD, Caughey GH, et al. Accumulation of intraepithelial mast cells with a unique protease phenotype in T(H)2-high asthma. J Allergy Clin Immunol. 2010;125:1046–1053
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (1512 KB)
    • CrossRef
29. Falaschetti E, Laiho J, Primatesta P, Purdon S. Prediction equations for normal and low lung function from the Health Survey for England. Eur Respir J. 2004;23:456–463
    • View In Article
    • MEDLINE
    • CrossRef
30. Berry MA, Parker D, Neale N, Woodman L, Morgan A, Monk P, et al. Sputum and bronchial submucosal IL-13 expression in asthma and eosinophilic bronchitis. J Allergy Clin Immunol. 2004;114:1106–1109
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (148 KB)
    • CrossRef
31. Saha SK, Berry MA, Parker D, Siddiqui S, Morgan A, May R, et al. Increased sputum and bronchial biopsy IL-13 expression in severe asthma. J Allergy Clin Immunol. 2008;121:685–691
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (410 KB)
    • CrossRef
32. Allahverdian S, Harada N, Singhera GK, Knight DA, Dorscheid DR. Secretion of IL-13 by airway epithelial cells enhances epithelial repair via HB-EGF. Am J Respir Cell Mol Biol. 2008;38:153–160
    • View In Article
    • CrossRef
33. Semlali A, Jacques E, Koussih L, Gounni AS, Chakir J. Thymic stromal lymphopoietin-induced human asthmatic airway epithelial cell proliferation through an IL-13-dependent pathway. J Allergy Clin Immunol. 2010;125:844–850
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (741 KB)
    • CrossRef

Back to Article Outline

References 

1. Brightling CE, Ward R, Woltmann G, Bradding P, Sheller JR, Dworski R, et al. Induced sputum inflammatory mediator concentrations in eosinophilic bronchitis and asthma. Am J Respir Crit Care Med. 2000;162:878–882
2. British guideline on the management of asthma. Thorax. 2008;63:iv1–iv121
3. Proceedings of the ATS Workshop on Refractory Asthma. Current understanding, recommendations, and unanswered questions. Am J Respir Crit Care Med. 2000;162:2341–2351
4. Heaney LG, Conway E, Kelly C, Johnston BT, English C, Stevenson M, et al. Predictors of therapy resistant asthma: outcome of a systematic evaluation protocol. Thorax. 2003;58:561–566
5. British Thoracic Society guidelines on diagnostic flexible bronchoscopy. Thorax. 2001;56:i1–i21
6. Bradding P, Feather IH, Howarth PH, Mueller R, Roberts JA, Britten K, et al. Interleukin 4 is localized to and released by human mast cells. J Exp Med. 1992;176:1381–1386
7. Choy DF, Modrek B, Abbas AR, Kummerfeld S, Clark HF, Wu LC, et al. Gene expression patterns of Th2 inflammation and intercellular communication in asthmatic airways. J Immunol. 2011;186:1861–1869
8. Bradding P, Roberts JA, Britten KM, Montefort S, Djukanovic R, Mueller R, et al. Interleukin-4, -5, and -6 and tumor necrosis factor-alpha in normal and asthmatic airways: evidence for the human mast cell as a source of these cytokines. Am J Respir Cell Mol Biol. 1994;10:471–480
9. Dougherty RH, Sidhu SS, Raman K, Solon M, Solberg OD, Caughey GH, et al. Accumulation of intraepithelial mast cells with a unique protease phenotype in T(H)2-high asthma. J Allergy Clin Immunol. 2010;125:1046–1053
10. Lemiere C, Ernst P, Olivenstein R, Yamauchi Y, Govindaraju K, Ludwig MS, et al. Airway inflammation assessed by invasive and noninvasive means in severe asthma: eosinophilic and noneosinophilic phenotypes. J Allergy Clin Immunol. 2006;118:1033–1039