Supplementary Appendix

Transcription

1 Supplementary Appendix This appendix has been provided by the authors to give readers additional information about their work. Supplement to: Grainge CL, Lau LCK, Ward JA, et al. Effect of bronchoconstriction on airway remodeling in asthma. N Engl J Med 2011;364:

2 Supplementary appendix to; The influence of bronchoconstriction on airway remodelling in asthma Christopher L Grainge PhD MRCP, Laurie CK Lau PhD, Jonathon A Ward BSc, Valdeep Dulay BM BSc, Gemma Lahiff BSc, Susan Wilson PhD, Stephen Holgate DM DSc, Donna E Davies PhD, Peter H Howarth DM FRCP. Southampton University School of Medicine Division of Infection, Inflammation and Immunity; NIHR Respiratory Biomedical research Unit and the Wellcome Trust Clinical research Facility. Corresponding author; Dr Peter Howarth DM FRCP - 1 -

3 Table of Contents SECTION 1 ADDITIONAL METHODOLOGICAL INFORMATION Dates of study Spirometry Bronchoscopy Bronchial biopsies Bronchoalveolar lavage BAL processing BAL cell counts Processing of tissue into glycol methacrylate Sectioning of GMA embedded tissues Staining procedure for GMA embedded tissue Periodic acid-schiff staining Image analysis Additional methodological information references 9 SECTION 2 ADDITIONAL RESULTS Table 1: Group symptom scores for week before and week of challenges Table 2: Change in cytospin differential cell counts in BAL before and after repeated inhalation challenges Table 3: Change in mucosal cell counts in endobronchial biopsies before and after repeated inhalation challenges Figure 1. Eosinophils as a percentage of total cells recovered at bronchoalveolar lavage before and after repeated inhaled challenges Figure 2. Eosinophil cationic protein (ECP) measured in bronchoalveolar lavage fluid before and after repeated inhaled challenges Figure 3. Immunoexpression of transforming growth factor beta (TGFβ) expressed as percentage of total epithelial area in bronchial biopsies before and after repeated inhaled challenges Figure 4. Ki-67 positive cells per millimetre of epithelium in bronchial biopsies before and after repeated inhaled challenges Figure 5. Subepithelial basement membrane (collagen band) thickness (µm) in bronchial biopsies before and after repeated inhaled challenges Figure 6. Percentage of epithelium staining positive with periodic acid - Schiff (PAS) reagent in bronchial biopsies before and after repeated inhaled challenges

4 SECTION 3 SUPPLEMENTAL DISCUSSION POINTS Use of atopic subjects for study Timing of the second bronchoscopy The use of BAL rather than sputum to assess allergen-induced airway eosinophil recruitment The relevance of methacholine to asthma and its comparison to allergen Statistical handling of remodelling outcome data Supplemental discussion references

5 Section 1 information Additional methodological 1.1 Dates of study The study received ethical approval on 2/4/2008, recruited its first volunteer on 2/8/2008 and the last volunteer completed the study on 11/17/2009. Data acquisition and analysis was completed on 6/30/ Spirometry Spirometry was performed with a dry bellows spirometer (Vitalograph, UK) and the best of at least three successive readings within 100 ml of each other was recorded as the FEV Bronchoscopy Fibreoptic bronchoscopy was performed according to British Thoracic Society (BTS) guidelines (1) and the local departmental standard operating procedure in the Wellcome Trust Clinical Research Facility. Samples were randomly taken from either the right or the left lung at the initial bronchoscopy, and the opposite side at the second bronchoscopy. 1.4 Bronchial biopsies Biopsies were taken with disposable alligator forceps (Bard, Ref , size: 1.8mm) (KeyMed (Medical & Industrial Equipment) Ltd., OLYMPUS Group Company, Southend-on-Sea, UK), after application of local anaesthetic, from 3rd and 4th airway carinae of the upper, middle / lingula and lower lobes (2-4 biopsies) and further processed for glycol methacrylate (GMA) embedding and immunohistochemistry

6 1.5 Bronchoalveolar lavage Bronchoalveolar lavage (BAL) was performed by wedging the bronchoscope in a segmental bronchus and installing 100ml (5 x 20ml aliquots) pre-warmed (37 C) normal saline into the segments of the upper lobes and then, after a 10 second dwell time for each 20 ml aliquot, recollecting the fluid by suction (approximately 40 60ml) into a bronchial lavage fluid trap. 1.6 BAL processing On removal from the subject, BAL fluid was filtered using a 100µm nylon filter (BD Falcon cell strainer, Marathon Lab. Supplies. London, UK) and then centrifuged at 1300G for 10 mins at 4 C. The supernatant was removed and aliquoted prior to storage at -80 C for later analysis. The cells were resuspended in PBS and cytocentrifuge slides (Thermo Shandon Ltd, Runcorn, UK) prepared and stored at -80 C for later analysis. 1.7 BAL cell counts BAL total cell counts were performed using a Neubauer hemocytometer and the trypan blue exclusion method. Differential cell counts were performed manually on cytocentrifuge slides stained with rapid Romanowsky stain (Raymond Lamb Ltd, Eastbourne, UK). Differential cell counts were obtained from a 400 cell count. 1.8 Processing of tissue into glycol methacrylate Method adapted from Britten et al 1993 (2). On removal of biopsy from the subject, the biopsy was initially rapidly assessed for adequate size, and then placed immediately into ice cold acetone (Fisher Scientific, Loughborough, UK) containing 2mM phenyl methyl sulphonyl fluoride (Sigma, Poole, UK) and 20mM iodoacetamide (Sigma, Poole, UK) and fixed overnight at -20 C

7 The fixative was replaced with dry acetone (as above) at room temperature for 15 minutes, then replaced with methyl benzoate (Fisher Scientific, Loughborough, UK) at room temperature for a further 15 minutes. The biopsy was then infiltrated with processing solution comprised of 5% methyl benzoate in glycol methacrylate (GMA solution A) (Polysciences Inc., Warrington, USA) at 4 C for 6 hours with the processing solution being changed every 2 hours. Following this, the biopsies were placed in freshly prepared embedding solution comprising 10ml GMA solution A, 70mg benzoyl peroxide (Polysciences Inc., Warrington, USA) and 250µl GMA solution B (Polysciences Inc., Warrington, USA) in Taab flat bottomed capsules (Taab, Aldermaston, UK cat no 0094) at 4 C for 48 hours. Biopsies were then stored until required in an airtight box containing silica gel at -20 C. 1.9 Sectioning of GMA embedded tissues Following rough trimming, GMA embedded biopsies were cut using a Leica Jung supercut 2065 glass knife microtome (Leica, Milton Keynes, UK) with section thickness 2µm, floated on reverse osmosis water (ROW) containing 1% ammonia, and then transferred to 0.01% poly-l-lysine (PLL) (Sigma-aldrich, Poole, UK) coated glass slides (Knittel Glaser, Baunschweig, Germany) for further processing. Slides were stored at -20 C for up to 2 weeks prior to use Staining procedure for GMA embedded tissue Biopsy sections on PLL coated slides were initially incubated with 0.1% sodium azide (Fisher Scientific, Loughborough,UK) and 0.3% hydrogen peroxide (Sigma-Aldrich, Poole, UK) in ROW for 30 mins to inhibit endogenous peroxidases. The slides were then washed with tris buffered saline (TBS) for 3 x 5 minutes, prior to the addition of blocking medium (Dulbecco s modified eagles medium (DMEM) with 20% foetal bovine serum (FBS) and bovine serum albumin (BSA)) for 30 minutes. Slides were then drained and primary - 6 -

8 antibodies applied at appropriate dilutions (as determined by titration) under coverslips overnight at room temperature. Slides were then washed with TBS for 3 x 5 mins, drained and biotinylated second stage antibodies applied at appropriate dilutions for 2 hours at room temperature. Slides were washed again with TBS for 3 x 5 mins, drained, and streptavidin biotin-peroxidase complexes (stabc-hrp complex, Dako, Stockport, UK) applied for 2 hours at room temperature. Following TBS wash (3 x 5 mins) either 3-amino, 9-ethylycarbazole (AEC) (AEC substrate pack, Launch diagnostics, Longfield, UK) or diaminobenzidine (DAB) (liquid DAB substrate pack, Launch diagnostics, Longfield, UK) substrates were applied for 20 or 10 minutes respectively at room temperature as required. Slides were rinsed in TBS and then running water for 5 minutes, prior to counterstaining with Mayer s haematoxylin (90 seconds) and a further running water rinse. Finally sections were sealed with aqueous mounting medium (AbD Serotec, Kidlington, UK) incubated at 80 C for 30 minutes and allowed to cool prior to coverslipping using Pertex (Surgipath, Peterborough, UK). All antibody dilutions were established by titration, and absence of non specific staining established by isotype controls. For each staining run isotype controls and slides processed in the absence of the primary antibody were run as controls to ensure the specificity of the immunostaining Periodic acid-schiff staining Initial experiments were performed to optimise staining with periodic acid-schiff (PAS) on GMA embedded biopsies. Biopsy sections on PLL coated slides were incubated with periodic acid for 10 minutes at room temperature, then washed well with several changes of ROW. Sections were then incubated with Schiff s reagent (Sigma, Poole, UK) for 20 minutes at room temperature, and then washed in running tap water for 10 minutes. Slides were then counterstained in - 7 -

9 Meyer s haematoxylin for 30 seconds and washed again in running tap water for 5 minutes. Sections were then coverslipped using Pertex as above Image analysis Staining was assessed using computer assisted image analysis (Zeiss KS400 image analysis system, Zeiss, Welyn Garden City, UK). Biopsies were examined manually and areas of epithelium and submucosa identified morphologically. The length of the epithelium and the area of the submucosa were calculated using computer assisted image analysis. The percentage of the epithelium that stained positive for TGFβ was calculated by image analysis thresholding of positive staining and exclusion of all areas of the biopsy not of interest. This method was not reliable with the colouration of the PAS stain and therefore positive areas of PAS staining were demarcated on enlarged digital images, and the areas calculated as a percentage of the total epithelial area. The thickness of the basement membrane and lamina reticularis was calculated beneath epithelium that was cut perpendicular and of full height and longitudinally orientated. In brief, an image of the lamina reticularis was captured from the Collagen III stained section and delineated. An Euclidean distance trans- formation was applied; this identified the central line within the lamina reticularis. A gray level coding was then assigned to each pixel between the central line and the outer boundary, relative to the distance (ie, the higher the gray level, the greater the distance [thickness]). An appropriate scaling for the magnification used was then applied, and the distance in micrometers was calculated. Cell counts were performed by counting positively stained cells manually; they were then expressed as cells per mm length of epithelium or per square mm of submucosa (the length of epithelium or area of submucosa calculated as above). A minimum of two biopsy sections separated by at least 30µm were examined for each immunohistochemical stain in each patient both before and after inhalation challenge, mean results from this repeated sampling were used for - 8 -

10 further calculations. Samples from all four exposure groups were intermingled during processing and analysis, which was performed blinded Additional methodological information references 1. British Thoracic Society Bronchoscopy Guidelines Committee, a Subcommittee of Standards of Care Committee of British Thoracic Society. British Thoracic Society guidelines on diagnostic flexible bronchoscopy. Thorax (2001) vol. 56 Suppl 1 pp. i Britten et al. Immunohistochemistry on resin sections: a comparison of resin embedding techniques for small mucosal biopsies. Biotechnic & histochemistry (1993) vol. 68 (5) pp

11 Section 2 Additional results 2.1 Table 1: Group symptom scores for week before and week of challenges Week before repeated challenges Mean symptom scores Allergen (n=12) 0.7 (1.2) 3.0 (2.0) Methacholine (n=12) 1.4 (1.8) 2.7 (1.6) Saline (n=12) 0.4 (0.4) 0.7 (1.1) Albuterol (n=12) 1.2 (1.2) 1.1 (1.3) p value between groups Week of repeated challenges Symptom scores in arbitrary units, all score are positive. Minimum score is zero, maximum 30. Values are means with SD in parentheses. p value between groups calculated using the Kruskall Wallis Test

12 2.2 Table 2: Change in cytospin differential cell counts in BAL before and after repeated inhalation challenges. Allergen 3.1 ( ) Methacholine -0.3 ( ) Saline -0.3 ( ) Albuterol / methacholine Change in percentage of cells recovered at bronchoalveolar lavage before and after repeated inhaled challenges. Eosinophils Neutrophils Macrophages Lymphocytes -0.3 ( ) 0.5 ( ) 0.00 ( ) 0.63 ( ) 2.1 ( ) ( ) 1.1 ( ) 0.6 ( ) -0.8 ( ) p value ( ) -0.5 ( ) 0.0 ( ) 0.00 ( ) n=12 for all groups. Values are medians with IQR in parentheses. Between group differences calculated using Kruskall Wallis test

13 2.3 Table 3: Change in mucosal cell counts in endobronchial biopsies before and after repeated inhalation challenges. Allergen 9.2 ( ) Methacholine 2.8 ( ) Saline 0.00 ( ) Albuterol / methacholine p value between groups Change in cells per mm 2 of submucosa before and after repeated inhaled challenge. Eosinophils Mast cells Macrophages 0.07 ( ) 5.9 ( ) 15.2 ( ) 3.5 ( ) -3.1 ( ) ( ) 2.0 ( ) 0.0 ( ) 2.0 ( ) n=12 for all groups. Values are median with IQR in parentheses. Between group differences calculated using Kruskall Wallis test

14 2.4 Figure 1. Eosinophils as a percentage of total cells recovered at bronchoalveolar lavage before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

15 2.5 Figure 2. Eosinophil cationic protein (ECP) measured in bronchoalveolar lavage fluid before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

16 2.6 Figure 3. Immunoexpression of transforming growth factor beta (TGFβ) expressed as percentage of total epithelial area in bronchial biopsies before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

17 2.7 Figure 4. Ki-67 positive cells per millimetre of epithelium in bronchial biopsies before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

18 2.8 Figure 5. Subepithelial basement membrane (collagen band) thickness (µm) in bronchial biopsies before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

19 2.9 Figure 6. Percentage of epithelium staining positive with periodic acid - Schiff (PAS) reagent in bronchial biopsies before and after repeated inhaled challenges. Panel A - Allergen challenges, B - methacholine, C - albuterol followed by methacholine and D - saline. Bars show mean. A B C D

20 Section 3 Supplemental discussion points 3.1 Use of atopic subjects for study Whilst there are many phenotypes of asthma, atopic asthmatics were selected to participate in this study as it is recognised that allergen challenge in these individuals induces an eosinophil influx within the airways. Eosinophilic inflammation of the airways, with an increase in activated and degranulated eosinophils, is, however, a characteristic feature of both atopic and non-atopic asthma in adults and children (1,2). The aetiology of the airway eosinophil recruitment is not known in many instances of non-atopic asthma so it is not possible to model this. As it was necessary to have a challenge that induced airway eosinophil recruitment as part of this study atopic asthmatics sensitive to house dust mite allergen were chosen to participate. Furthermore airway allergen inhalation challenge is well validated model in atopic asthma. It is of interest to note that although the phenotypes of atopic and non-atopic asthma are highly distinct, immunological similarities have lead to uncertainty over whether they are distinct immunopathological entities. In addition to the common airway eosinophilia, bronchial biopsy studies have suggested that both variants are also characterised by Th2 cells secreting IL-4, IL-13 and IL-5, the presence of CC chemokines and FCεR1+ cells, and local IgE synthesis (3-5). The significance of IgE in non-allergic asthma thus remains unclear. Humbert et al, who found evidence of local IgE production and abnormal macrophage activation in biopsy specimens from non-atopic asthmatic adults, postulated the existence of an ongoing immune response against an as-yet unidentified antigen (4). Considerations about the atopic and non-atopic nature of the asthmatics is however adjacent to the study, as the purpose was to assess whether bronchoconstriction per se can be a factor that induces airway remodelling in the absence of induced airway eosinophil recruitment. Bronchoconstriction is a

21 feature of both atopic and non-atopic asthma and as such the atopic status is not a determinant factor 3.2 Timing of the second bronchoscopy The timing of the second bronchoscopy to 4 days was chosen as, if the allergen-induced response was dependent on eosinophil influx and downstream events, as a consequence of this, it would provide sufficient time for these events to have arisen. Also 4 days was selected as a time point as any changes that had arisen, if they had been transient, being evident only 24 hrs post challenge but not evident at this more prolonged time point, they may not have such pathophysiological relevance to disease chronicity as those that were persistent. The choice of 4 days was thus a balance reflecting these considerations. 3.3 The use of BAL rather than sputum to assess allergen-induced airway eosinophil recruitment BAL was chosen on 2 accounts. Firstly bronchoscopy had to be undertaken to obtain the airway tissue samples to assess airway remodelling so undertaking a bronchoalveolar lavage at the same bronchoscopy provided a ready means to sample the airway lumen. Bronchoalveolar lavage is a well established and validated method of evaluating airway luminal inflammation. However, potentially more importantly, this method was selected in advance of induced sputum as to collect sputum it is necessary to cough. Cough contracts the airways and there are studies reported that indicate that cough can induce airway remodelling (6-8). Thus using induced sputum as an outcome variable would be a significant confounding factor in the interpretation of the results

22 3.4 The relevance of methacholine to asthma and its comparison to allergen Methacholine acts on acetyl choline receptors on airway smooth muscle to induce bronchoconstriction. The intracellular events that give rise to bronchoconstriction are similar to that induced by other agonists. The nature of the bronchoconstriction with methacholine is thus similar to that arising during the early bronchoconstrictor response to allergen. Released agonists such as histamine and leukotrienes contribute to this immediate airway response. There are 2 components of the airway response to allergen, the immediate and the late bronchoconstrictor response. The study was designed to match the immediate bronchoconstrictor response to allergen and methacholine. The immediate response to allergen is inhibited by beta-agonists (9, 10), suggestive of the predominance of smooth muscle constriction as the basis for this rapidly evolving bronchoconstriction. The late reaction is not however, modified to the same extent once alteration in baseline lung function is taken into account. Additionally the associated allergen-induced eosinophil response is unaffected by beta agonist pre-treatment. Thus the late asthmatic reaction has a different profile to the immediate response and other factors, such as airway wall oedema and airway secretions, may be addition contributors to the reduced lung function that features in the late asthmatic reaction to allergen. 3.5 Statistical handling of remodelling outcome data No correction for multiple comparisons has been made with respect to the 4 outcome markers of epithelial repair and airway remodelling. An ANOVA has been used to compare the means across treatment groups for each outcome rather than perform multiple pair wise tests. The p-values presented for the ANOVA s directly show the strength of evidence. Controlling for Type I error

23 across outcomes using a Bonferroni correction will be overly conservative and will increase the Type II error. Given the small sample size and the objective of this exploratory randomized controlled trial, our preferred option was not to focus overly on an artificial statistically significant value but to maintain Type II error and present the original p-value for which the more conservative reader can make their own adjustment (11). 3.6 Supplemental discussion references 1. Gaga M, P.Lambrou, N. Papageorgiou, N.G. Koulouris, E. Kosmas, S. Fragakis, C. Sofios, A. Rasidakis, J. Jordanoglou Eosinophils are a feature of upper and lower airway pathology in non-atopic asthma, irrespective of the presence of rhinitis. Clin Exp Allergy. 30: Snijders D, Agostini S, Bertuola F, Panizzolo C, Baraldo S, Turato G, Faggian D, Plebani M, Saetta M, Barbato A. Markers of eosinophilic and neutrophilic inflammation in bronchoalveolar lavage of asthmatic and atopic children. Allergy Aug;65(8): Novak N. T. Bieber Allergic and non-allergic forms of atopic diseases J Allergy Clin Immunology. 112: Humbert M., G. Menz, S. Ying, C. Corrigan, D. Robinson, S. Durham, A. B. Kay The immunopathology of extrinsic (atopic) and intrinsic (non-atopic) asthma: more similarities than differences. Immunology Today. 20: Ying S., M. Humbert, Q. Meng, R. Pfister, G. Menz, H. Gould, A.B. Kay, S. Durham Local expression of ε germline gene transcripts and RNA for the ε heavy chain of IgE in the bronchial mucosa in atopic and non atopic asthma. J Allergy Clin Immunol. 107: Xie S, Macedo P, Hew M, Nassenstein C, Lee KY, Chung KF. Expression of transforming growth factor-beta (TGF-beta) in chronic idiopathic cough. Respir Res May 22;10:

24 7. Matsumoto H, Niimi A, Tabuena RP, Takemura M, Ueda T, Yamaguchi M, Matsuoka H, Jinnai M, Chin K, Mishima M. Airway wall thickening in patients with cough variant asthma and nonasthmatic chronic cough. Chest Apr;131(4): Niimi A, Torrego A, Nicholson AG, Cosio BG, Oates TB, Chung KF. Nature of airway inflammation and remodelling in chronic cough. J Allergy Clin Immunol, 2005; 116: Howarth PH, Durham SR, Lee TH, Kay AB, Church MK, Holgate ST. Influence of albuterol, cromolyn sodium and ipratropium bromide on the airway and circulating mediator responses to allergen bronchial provocation in asthma. Am Rev Respir Dis Nov;132(5): Weersink EJ, Aalbers R, Koëter GH, Kauffman HF, De Monchy JG, Postma DS Partial inhibition of the early and late asthmatic response by a single dose of salmeterol. Am J Respir Crit Care Med Nov;150(5 Pt 1): ) 11. Proschan MA, Waclawiw MA Practical guidelines for Multiplicity adjustments in Clinical Trials Controlled Clin. Trials 2000; 21: