Online Repository
Asthma inflammatory phenotypes show differential microRNA expression in sputum
Tania Maes, PhD1*, Francisco Avila Cobos, Msc2, Florence Schleich, MD, PhD3, Valentina Sorbello, PhD 4, Monique Henket, MLT3, Katleen De Preter, PhD2, Ken R. Bracke, PhD1, Griet Conickx, Pharm1, Claire Mesnil, PhD5,Jo Vandesompele, PhD2, Lies Lahousse, PhD1, Fabrice Bureau, PhD5, Pieter Mestdagh, PhD2, Guy F. Joos, MD, PhD1, Fabio L.M. Ricciardolo, MD, PhD4, Guy G. Brusselle, MD PhD1, Renaud Louis, MD, PhD3
1 Department of Respiratory Medicine, Ghent University Hospital, Belgium
2Center for Medical Genetics, Ghent University, Belgium
3Department of Respiratory Medicine, CHU Sart-TilmanB35, Liege, Belgium
4Department of Clinical and Biological Sciences, University of Torino, Italy
5Department of Cellular and Molecular Immunology, GIGA-Research Centre and Faculty of Veterinary Medicine, University of Liege, Belgium
* Corresponding author
Running title: miRNAs in asthma
Corresponding author: Maes Tania
Department of Respiratory Medicine, Ghent University Hospital
Blok B, room 420, De Pintelaan 185, 9000 Ghent, Belgium
Telephone: 0032 9 2406725
Fax: 0032 9 2402341
Tania.Maes@UGent
List of abbreviations
ACQ: Asthma Control Questionnaire
APC: Adenomatous polyposis coli
BAL: bronchoalveolar lavage
BMI: Body mass index
CBA: cytometric bead array
C/EBP: CCAAT-enhancer-binding protein
FeNO: Fraction of exhaled nitric oxide
FEV1: Forced expiratory volume in 1s
FVC: Forced vital capacity
ICS: Inhaled corticosteroids
IL: Interleukin
IgE: Immunoglobulin E
JAK-STAT: Janus kinase-Signal transducers and activators of transcription
KEGG: Kyoto encyclopedia of genes and genomes
LNA: Locked nucleic acid
miRNA: microRNA
MAPK: Mitogen-activated protein kinase
NF-κB: Nuclear factor kappa-light chain enhancer of activated B-cells
NOD: nucleotide-binding oligomerization domain
PBS: phosphate buffered saline
qPCR: Quantitative real-time polymerase chain reaction
RT: reverse transcriptase
SD: standard deviation
WNT: Wingless tail
METHODS
Sputum induction
Prior to sputum induction, subjects inhaled 400 µg salbutamol using a metered-dose inhaler (+spacer). Sputum was induced using an ultrasonic nebulizer (ultra-Neb 2000, Devilbiss; output set at 0.9 ml/min). Subjects inhaled hypertonic saline (NaCl 5%) when FEV1 post salbutamol was 65% predicted and isotonic saline (NaCl 0.9%) when FEV1 was <65% predicted (11). The aerosol was inhaled for three consecutive periods of 5 min. FEV1 was monitored every 5 minutes for safety reasons, and when FEV1 dropped to 80% of the post-bronchodilator values, the induction procedure was stopped. The whole sputum was weighted and three volumes of PBS were added. After homogenizing by manual agitation for 30 sec and centrifugation (800g) for 10 min at 4°C, the cell pellet and supernatant were separated. Cells were treated with Sputolysin 0.1% (Calbiochem, Germany), washed with PBS and resuspended in 1 ml. Total cell counts, % squamous cells and cell viability (trypan blue staining) were determined with a manual haemocytometer. Sputum cell differentials were determined by counting 500 cells non squamous cells on Cytospin samples that were stained with RAPI-DIFF II stain (Atom Scientific, Manchester, United Kingdom). The samples had a squamous cell count between 0 and 30%. Eosinophilic asthma was defined as asthmatics who had 3% eosinophils and <76% neutrophils in sputum; neutrophilic asthma as asthmatics with 76% neutrophils and <3% eosinophils in sputum and paucigranulocytic asthma as asthmatics with <3% eosinophils and <76% neutrophils in sputum(12,13). Asthmatics with a mixed granulocytic phenotype (76% neutrophils and 3% eosinophils in sputum) were excluded.
miRNA detection in sputum supernatant
Total RNA from cell-free sputum supernatant (150 µl) was extracted using the miRNeasy Micro kit, without carrier RNA, according to the manufacturer’s guidelines (Qiagen, Hilden, Germany). miRNA profiling on 80 ng RNA was performed at Biogazelle (Zwijnaarde, Belgium) using high-throughput stem-loop RT-qPCR miRNA expression profiling (14). Briefly, Megaplex RT stem-loop primer pool (Life Technologies) was used to reverse transcribe total RNA, enabling miRNA specific cDNA synthesis of 755 human miRNAs and small RNA controls. Increased detection sensitivity was obtained by a pre-amplification (14-cycle PCR reaction) of the Megaplex RT product using a miRNA specific forward primer and a universal reverse primer. The preamplified miRNA cDNA was subsequently used for a qPCR reaction with miRNA specific hydrolysis probes and primers (Life Technologies). All reactions were performed in TaqMan arrays using the ViiA7 instrument (Life technologies) using the gene maximization strategy. Data quality was confirmed by the small replicate variability of an endogenous small RNA control (U6). There was some variability in the number of detected miRNAs for each sample, but the mean Cq values fell within the expected range for all samples (data not shown). The detection cut-off was put at a Cq value of 32. The mean raw Cq values per group are shown in Table E5. Data normalization was performed in qbase+ software (Biogazelle, using the modified global mean normalization procedure(15). The normalized expression data (calibrated normalized relative quantities, CNRQ) are log10-transformed. To investigate differential miRNA expression between the groups of the screening cohort, only those miRNAs that were detectable (Cq<32) in at least 80% of the samples of one of the groups were further analyzed. The miRNAs with undetectable expression levels were imputed by substracting one log2 value from the minimum expression of this miRNA across the samples of the cohort. Differential miRNA expression between 2 test groups was calculated in the data analysis software R3.1.2. with the limma-package. An FDR adjusted P value <0.05 (False Discovery Rate method) was considered significant. Graphic visualization of miRNA expression levels was done using box plot presentations of the CNRQ, according to asthma severity (healthy vs. mild asthma vs. severe asthma) or inflammatory profile (healthy vs. eosinophilic asthma vs. neutrophilic asthma). Heat maps were generated using the cluster method Ward D2, distance Manhattan.
In the validation study, the expression of the identified miRNAs from the screening study (hsa-miR-629-3p, hsa-miR-223-3p andhsa-miR-142-3p) was quantified with specific TaqMan Micro Assays (Life Technologies). To increase detection sensitivity, a custom RT and preamplification pool was created (12-cyle PCR, according to the manufacturer’s guidelines). The preamplified miRNA cDNA was subsequently used for a qPCR reaction with miRNA specific hydrolysis probes and primers (Life Technologies). Validation reactions were performed on a LightCycler480 instrument (Roche Diagnostics) in a 96 well plate with a total reaction volume of 20µl (1x Taqman microassay, according to Life Technology guidelines). The data were normalized in qbase+ using two stably expressed miRNAs (hsa-miR-20a and hsa-miR-30b) identified from the screening cohort. The mean Cq values (SD) on a sample pool of the validation cohort were for hsa-miR-20a: 23.440.09; hsa-miR-30b: 27.770.14; hsa-miR-629-3p: 29.900.29; hsa-miR-223-3p: 19.870.31 and hsa-miR-142-3p: 26.040.11. The normalized expression data are log10-transformed prior to data analysis.
In situ hybridization for miRNAs
Cryosections of bronchial biopsies were fixed with formalin, treated with proteinase K followed by hybridization with miRCury LNA detection probes (5’ and 3’ DIG, Exiqon, Denmark) according to the manufacturer’s guidelines. Detection was performed with anti-digoxigenin, labeled with alkaline phosphatase, using nitro-blue tetrazolium and 5-bromo-4-chloro-3’indolyphosphate (NBT-BCIP) as substrate. Sections were counterstained with nuclear fast red. Probes used were homo sapiens hsa-miR-629-3p, hsa-miR-223-3p and scrambled control (see Figure E1 online repository, for in situ with scrambled negative controls). The small RNA U6 was used a positive control.
Neutrophil elastase staining
Cryosections of bronchial biopsies were stained with mouse anti-human neutrophil elastase (DAKO, M0752). Detection was performed with a goat anti-mouse poly-alkaline-phosphatase (Klinipath Immunologic, DPVM-55AP), with as substrate new fushin (DAKO, K0625). Counterstaining was done with Mayer’s haematoxyline (Sigma, MHS80).
Flow cytometry and miRNA analysis on sorted sputum cells
All staining procedures were performed in PBS without Ca2+ or Mg2+ (BioWhittaker, Lonza) containing 5mM EDTA and 1% BSA. Sputum cells were labeled with a combination of mouse anti-human CD45 (V450, clone HI30), mouse anti-human CD14 (allophycocyanin-conjugated clone M5E2) and mouse anti-human CD16 (PE, clone B73.1). All antibodies were from BD Pharmingen (San Diego, CA). Neutrophils were defined as CD45+, CD14-, CD16+; monocytes as CD45+, CD14-, SSClow and macrophages as CD45+, CD14-, SSChigh based on Brooks et al (1) and Figure E3 for gating strategy). Live gate selection for neutrophils and monocytes was done using 7AAD. Cells (10000 per cell type) were sorted to >92% purity on a FACS ARIA II (BD) and captured in Qiazol followed by RNA extraction with the miRNeasy micro kit (Qiagen). The miRNA expression in sorted sputum cells was analysed using Qiagen reagents (miScript primer assays and recommended snord primer assays for normalization, miScript II reverse transcriptase kit and miScript SybR Green PCR kit, without preamplification step). With this approach, the mean Cq values on a pooled sample were for snord68: 19.670.10; snord96A: 20.420.03; snord95: 19.000.04; hsa-miR-223-3p: 16.160.04 and hsa-miR-142-3p: 17.680.03). The miRNA expression data were normalized to snord 68, 96A and 95 and log10-transformed prior to analysis.
In silico analysis into the functional role of the miRNAs
All experimentally validated targets of hsa-miR-142-3p, hsa-miR-223-3p and hsa-miR-629-3p were retrieved using miRWalk 2.0 (129 genes) (2). Using public available gene expression data from bronchial epithelial cells (3), BAL macrophages (4) and BAL neutrophils (5) (available at Gene Expression Omnibus (GEO.NCBI) under accession numbers GSE63142, GSE7368 and GSE2322, respectively) we identified 110 out of the list of 129 genes as being specifically expressed in these cell types and picked these for further analysis (Figure E4 Online Repository).
For GSE63142, the data were already normalized by cyclic-LOESS (6). Bronchial epithelial cells (BECs) from healthy controls (n = 27), moderate asthmatics (n = 72) and severe asthmatics (n=56) were considered as three independent groups. According to the methods section from the associated manuscript (2), low-variance probes were filtered out, and remaining ones were considered as expressed. For GSE7368 (4), the data were RMA background corrected and quantile normalized using the “affy” library (7). Again, low-variance probes were filtered out. Corticosteroid-resistant (CR, n=3) and corticosteroid-sensitive (CS, n=3) were considered as two independent groups. For GSE2322 (5), the data were already pre-processed using GC Robust Multi-array Average (GCRMA). Genes called as present in at least 70% of the samples (BAL neutrophils) were considered as expressed. This way, the initial set of 129 genes was reduced 110 genes being expressed in at least one of the six different groups considered (Figure 4E Online repository).
Finally, within all KEGG pathway genesets (c2.cp.kegg.v5.0.symbols.gmt; 186 different gene sets) available at MsigDB v5.0 (Broad Institute. Released in March 2015), we selected those gene sets that contained at least 1 of the 110 potential target genes (Figure E5 online repository and Figure 5). Gene sets were ordered according to the ratio of the number of target genes/ total gene set size.
All analyses were performed using R statistical software (R version 3.1.1).
Transfection of human bronchial epithelial cells (HBEC) with miRNA mimics
Normal human bronchial epithelial cells (CC-2540, Lonza) were cultured in BEGM medium (CC-3170, Lonza) according to the manufacturers’ guidelines. At passage 2, 150000 cells were transfected with mirVana 30nM miRNA mimics and controls (Ambion, Life Technologies) in 6-well plates using lipofectamine (RNAiMAX, Invitrogen)(N=4 for each condition). Mimics for hsa-miR-223-3p, hsa-miR-142-3p, hsa-miR-629-3p, a lipofectamine control, negative scrambled control (30nM), and miR-1 positive control (30nM) were included. Culture medium was refreshed after 12 hours. At 48 hours after transfection, culture supernatans was collected and cells were lysed in Qiazol, followed by RNA extraction with the miRNeasy mini kit (Qiagen).
cDNA of HBECs was prepared with the miScript II reverse transcriptase kit (Qiagen) or iScriptAdvanced cDNA synthesis kit (Biorad). RT-PCR starting from 10 ng cDNA was performed using Taqman gene expression assays (Life Technologies) or Biorad PrimePCR assays. The mRNA gene expression was normalized to 3 reference genes (RPL13A, HPRT and GAPDH) with qbase+ software (Biogazelle, followed by data analysis.
Successful transfection was shown by increased expression of the respective miRNAs (using miScript primer assays) as well as by demonstrating a reduced mRNA expression of PTK9 and IL-8 (2 target genes of the positive control miR-1) in the samples transfected with miR-1 mimic (data not shown).
RESULTS:
Characteristics of the Screening Cohort
Table E3 Online Repository shows demographic, functional characteristics and sputum characteristics of the 36 subjects from the screening cohort, based on inflammatory phenotype. The screening population consisted of 10 healthy controls, 16 subjects with eosinophilic asthma and 8 subjects with neutrophilic asthma. Both the neutrophilic and eosinophilic asthmatics had a significantly reduced FEV1 and FEV1/FVC compared to healthy controls. The eosinophilic asthmatics were more frequently atopic and had higher FeNOcompared to healthy controls and neutrophilic asthmatics. Individuals with neutrophilic asthma were older than healthy controls and eosinophilic asthmatics.
Characteristics of the ValidationCohort
Table E6and E7Online Repository shows demographic, functional characteristics and sputum characteristics of the Validation cohort of 60 individuals, including 10 healthy controls and 50 asthma patients (29 mild-to-moderate and 21 severe) based on disease severity or inflammatory phenotype, respectively. There were no significant differences in gender, atopy, smoking status and pack-years when groups were compared according to either disease severity or inflammatory profile.
Severe asthmatics had significantly lower FEV1, FVC, FEV1/FVC and significantly higher ACQ scores compared to mild-to-moderate asthmatics and healthy subjects..Within the mild-to-moderate asthmatics group, 65.5 % was ICS naive, while 34.5 % were treated with low-to-moderate doses of ICS. The percentage of eosinophils in sputum was higher in the mild-to moderate and severe asthma groups compared to healthy subjects.
Within the asthmatics, 29/50 had an eosinophilic (3% sputum eosinophils and <76% neutrophils) and 21/50 had a neutrophilic phenotype (76% sputum neutrophils and <3% eosinophils). FEV1 and FEV1/FVC did not differ between subjects with eosinophilic and neutrophilic asthma, but was reduced in both groups compared to healthy subjects.
Screening Cohort / Healthy / Mild to moderate asthma / Severe asthmaN / 10 / 17 / 9
Demographic characteristics
Age ( median years, IQR) / 40 (28-55) / 58 (51-66) / 50 (46-63)
Gender (M/F) / 3/7 / 5/12 / 2/7
BMI (mean SD) / 24.53.5 / 27.15.5 / 27.66.5
Atopy (n, % ) / 4 (40%) / 10 (59%) / 6 (67%)
Smoking status (NS/Ex)1 / 8/2 / 13/4 / 4/5
Pack-year (median, IQR) / 0 (0-2.5) / 0 (0-6.3) / 5 (0-30.3)
Functional characteristics and therapy
FEV1 (meanSD)(%) / 98.217.8 / 97.713.8 / 53.812.3**,§§
FEV1 (meanSD)(L) / 3.21.0 / 2.70.8 / 1.50.5**,§§
FVC (meanSD) (%) / 103.414.7 / 109.810.9 / 72.811.4**,§§
FVC (meanSD)(L) / 3.91.1 / 3.61.0 / 2.40.8**,§
FEV1/FVC (%)(meanSD) / 80.16.3 / 74.15.3* / 61.19.8**,§§
Reversibility (%)(meanSD) / - / 7.17.6 / 17.428.1
ACQ (median, IQR) / ND / 0.7 (0.3-1.1) / 2.8 (1.9-3.9) §§
FeNO (median ppb, IQR) / 18.8 (15.9-22.8) / 32.3 (23.8-50.0) / 22.5 (16.5-72.9)
ICS therapy (Yes/no) / 0/10 / 8/9* / 9/0**,§
ICS dose (median, IQR) / 0 / 0 (0-800)* / 2000 (1600-4000)**, §§
LABA (Yes/no) / 0/10 / 5/12 / 9/0**, §§
SABA (Yes/no) / 0/10 / 7/10* / 6/3*
Sputum characteristics
Inflammatory phenotype2:
eosinophilic/neutrophilic/ paucigranulocytic / 0/0/10 / 11/4/2** / 5/4/0**
Sputum weight (g) / 4.8 (2.0-7.2) / 3.4 (2.5-5.4) / 2.4 (1.7-4.9)
Sputum cell count (x106/g) / 1.4 (0.5-2.2) / 1.2 (0.7-2.2) / 3.7 (1.1-6.6)
% Viability sputum cells / 65.2 (52.8-81.6) / 68.0 (59.5-83.5) / 68.0 (48.0-84.9)
% macrophages / 62.7 (34.1-82.6) / 21.2 (16.6-57.8) / 12.0 (3.5-20.3)**
% neutrophils / 22.5 (1.3-62.8) / 39.0 (15.5-71.0) / 72.0 (43.6-93.8)
% eosinophils / 0.0 (0.0-0.0) / 7.8 (0-26.9)* / 8.5 (1.2-19.7)*
% lymphocytes / 0.9 (0.4-1.9) / 1.0 (0.5-3.8) / 2.0 (0.9-3.3)
% epithelial cells / 20.5 (12.3-31.7) / 13.8 (7.8-19.2) / 13 (4.5-26.5)
Table E1 Online Repository : Demographic, functional and inflammatory characteristics of the Screening Cohort according to disease severity (Full Table)
Abbreviations: IQR: interquartile range; yrs: years; M: male; F: female; BMI: body mass index; SD: standard deviation; FEV1: forced expiratory volume in 1s (before bronchodilator); L: litre; FVC: forced vital capacity; ACQ: asthma Controle Questionnaire; ICS: inhaled corticosteroids
1NS: never-smoker, Ex: ex-smoker since at least 6 months
2Sputum eosinophilic inflammation:3% eosinophilsand <76% neutrophils; neutrophilic inflammation: 76% neutrophilsand <3% eosinophils, paucigranulocytic inflammation: eosinophils 3% and neutrophils 76%
Data are expressed distribution (Yes/No), mean SD or median with IQR; Comparisons were done with Pearson Chi-square tests or by Kruskall Wallis (KW)-test, followed by Mann-Whitney U+bonferroni.*: P<0.05 vs healthy; **:P<0.001 vs healthy ; §: P<0.05 vs mild-to moderate asthma; §§:P<0.001 vs mild-to-moderate asthma
Differential expression analysis in Screening Cohort based on severitySevere asthma compared to healthy subjects
miRNA ID / Log10Fold change / P.Value / adj.P.Val (FDR)
hsa-miR-629-3p / 0.93 / 0.0001 / 0.0115
hsa-miR-223-3p / 0.84 / 0.0001 / 0.0115
hsa-miR-142-3p / 0.74 / 0.0003 / 0.0201
hsa-miR-596 / 0.90 / 0.0056 / 0.2379
hsa-miR-566 / 0.64 / 0.0077 / 0.2646
hsa-miR-939-5p / 0.88 / 0.0100 / 0.2717
hsa-miR-541-5p / -2.30 / 0.0111 / 0.2717
hsa-miR-636 / 0.52 / 0.0265 / 0.5033
hsa-miR-548c-3p / -0.48 / 0.0261 / 0.5033
hsa-.miR-1275 / 0.56 / 0.0347 / 0.5940
hsa-miR-520c-3p / 0.29 / 0.0455 / 0.7065
hsa-miR-1233-3p / 0.74 / 0.0644 / 0.7610
hsa-miR-302a-3p / 0.47 / 0.0809 / 0.7610
hsa-miR-381-3p / 0.45 / 0.0777 / 0.7610
hsa-miR-661 / 0.42 / 0.0931 / 0.7610
hsa-miR-29b-3p / 0.40 / 0.0812 / 0.7610
hsa-miR-103a-3p / 0.40 / 0.0911 / 0.7610
hsa-miR-1208 / 0.37 / 0.0773 / 0.7610
hsa-miR-222-3p / 0.29 / 0.0797 / 0.7610
hsa-miR-638 / 0.28 / 0.0739 / 0.7610
Severe asthma compared to mild-to-moderate asthma
miRNA ID / Log10Fold change / P.Value / adj.P.Val (FDR)
hsa-miR-629-3p / 0.64 / 0.0023 / 0.3950
hsa-miR-211-5p / 0.97 / 0.0150 / 0.5772
hsa-miR-1233-3p / 0.75 / 0.0390 / 0.5772
hsa-miR-636 / 0.56 / 0.0083 / 0.5772
hsa-miR-204-5p / 0.48 / 0.0246 / 0.5772
hsa-miR-9-5p / 0.48 / 0.0208 / 0.5772
hsa-miR-381-3p / 0.47 / 0.0405 / 0.5772
hsa-miR-202-3p / 0.42 / 0.0267 / 0.5572
hsa-miR-142-3p / 0.37 / 0.0369 / 0.5572
hsa-miR-155-5p / 0.35 / 0.0186 / 0.5572
hsa-miR-222-3p / 0.31 / 0.0400 / 0.5572
hsa-miR-106a-5p / -0.54 / 0.0314 / 0.5572
hsa-miR-135b-5p / 0.34 / 0.0489 / 0.6358
hsa-miR-523-3p / 0.33 / 0.0521 / 0.6358
hsa-miR-596 / 0.53 / 0.0611 / 0.6836
hsa-miR-223-3p / 0.33 / 0.0640 / 0.6836
hsa-miR-1208 / 0.34 / 0.0743 / 0.7059
hsa-miR-541-5p / -1.44 / 0.0707 / 0.7059
hsa-miR-520c-3p / 0.23 / 0.0786 / 0.7072
hsa-miR-1275 / 0.39 / 0.0921 / 0.7862
Mild-to-moderate asthma compared to healthy subjects
miRNA ID / Log10Fold change / P.Value / adj.P.Val (FDR)
hsa-miR-223-3p / 0.51 / 0.0048 / 0.6106
hsa-miR-9-5p / -0.55 / 0.0071 / 0.6106
hsa-miR-454-3p / 0.49 / 0.0309 / 0.9927
hsa-miR-1255b-5p / 0.45 / 0.0694 / 0.9927
hsa-miR-939-5p / 0.44 / 0.1272 / 0.9927
hsa-miR-1291 / 0.39 / 0.1429 / 0.9927
hsa-miR-142-3p / 0.38 / 0.0274 / 0.9927
hsa-mir-206 / 0.37 / 0.0607 / 0.9927
hsa-mir-596 / 0.37 / 0.1709 / 0.9972
hsa-mir-618 / 0.32 / 0.3769 / 0.9972
hsa-mir-566 / 0.32 / 0.1111 / 0.9927
hsa-mir-7201 / 0.31 / 0.0274 / 0.9927
hsa-mir-509-5p / 0.30 / 0.2207 / 0.9927
hsa-mir-629-3p / 0.29 / 0.1360 / 0.9927
hsa-mir-1290 / 0.27 / 0.1084 / 0.9927
hsa-mir-572 / 0.25 / 0.2971 / 0.9927
hsa-mir-1274B1 / 0.25 / 0.2127 / 0.9927
hsa-mir-886-3p1 / 0.24 / 0.4503 / 0.9927
hsa-mir-1260a / 0.24 / 0.0795 / 0.9927
hsa-mir-548d-5p / 0.22 / 0.2853 / 0.9927
Table E2: Screening Cohort: Differential miRNA expression analysis based on severity
Differential miRNA expression was investigated for 171 miRNAs. The Top 20 miRNAs for each comparison ordered according to smallest adjusted P value and highest fold change in expression are included in the table. An adjusted P value (FDR) <0.05 was considered significant. The miRNAs in bold and shaded in grey are significantly differentially expressed.
1No longer a miRNA according to miRBase version2014
Screening Cohort / Healthy / Eosinophilic asthma / Neutrophilic asthmaN / 10 / 16 / 8
Demographic characteristics
Age ( median years, IQR) / 40 (28-55) / 55 (49-64) / 65 (57-69)*
Gender (M/F) / 3/7 / 3/13 / 3/5
BMI (mean SD) / 24.53.5 / 28.66.3 / 24.64.3
Atopy (n, % ) / 4(40%) / 13 (81%)* / 2 (33%)
Smoking status (NS/Ex)1 / 8/2 / 12/4 / 3/5
Pack-year (median, IQR) / 0 (0-2.5) / 0 (0-3.8) / 18 (0-33.9)
Functional characteristics and therapy
Distribution mild/severe asthma / - / 11/5 / 4/4
FEV1 (meanSD)(%) / 98.217.8 / 82.125.8 / 78.325.9
FEV1 (meanSD)(L) / 3.21.0 / 2.20.8* / 2.00.9*
FVC (meanSD) (%) / 103.414.7 / 96.822.0 / 93.621.1
FVC (meanSD)(L) / 3.91.1 / 3.11.0 / 2.91.1
FEV1/FVC (%)(meanSD) / 80.16.2 / 69.89.5** / 66.99.0*
Reversibility (%)(meanSD) / - / 15.022.3 / 5.216.1
ACQ (median, IQR) / - / 1.1 (0.3-2.4) / 1.3 (0.9-2.1)
FeNO (median ppb, IQR) / 18.8 (15.9-22.8) / 40.3 (25.8-85.6) * / 19.5 (14.1-27.6)
ICS therapy (Yes/no) / 0/10 / 12/4* / 5/3*
ICS dose (median, IQR) / 0 (0-0) / 900 (50-1600)* / 1000 (0-2000)
LABA (Yes/no) / 0/10 / 10/6* / 4/4*
SABA (Yes/no) / 0/10 / 6/10* / 5/3*
Sputum characteristics
Sputum weight (g) / 4.8 (2.0-7.2) / 3.0 (1.9-3.8) / 4.9 (3.2-7.7)
Sputum cell count (x106/g) / 1.4 (0.5-2.2) / 1.5 (0.7-3.4) / 1.0 (0.6-6.7)
% Viability / 65.2 (52.8-81.6) / 62.0 (57.5-73.3) / 84.9 (57.8-89.3)
% macrophages / 62.7 (34.1-82.6) / 22.4 (17.1-40.6)* / 3.5 (2.1-11.8)**,
% neutrophils / 22.5 (1.3-62.8) / 37.1 (24.5-50.8) / 93.8 (82.4-96) **,
% eosinophils / 0 (0-0) / 20.7 (8.6-37.3)** / 0 (0-1.9)
% lymphocytes / 0.9 (0.4-1.9) / 2.5 (0.9-3.6) / 0.9 (0.1-1.8)
% epithelial cells / 20.5 (12.3-31.7) / 14.5 (7.6-21.3) / 7.5 (4.2-21.3)
Table E3 Online repository: Demographic, functional and inflammatory characteristics of the Screening Cohort according to inflammatory phenotype(Full Table)