Clinical and molecular phenotyping of asthma and COPD
Boudewijn, Ilse Maria
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Boudewijn, I. M. (2019). Clinical and molecular phenotyping of asthma and COPD. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
8
Nasal gene expression changes
with inhaled corticosteroid
treatment in asthma
Ilse M. Boudewijn, Andy Lan, Alen Faiz, Claire A. Cox, Sharon Brouwer, Siebrig Schokker, Sebastiaan J. Vroegop, Martijn C. Nawijn,
Prescott G. Woodruff, Stephanie A. Christenson, Paul Hagedoorn, Henderik W. Frijlink, David F. Choy, Uilke Brouwer, Marissa Wisman, Dirkje S. Postma, James Fingleton, Richard Beasley,
Maarten van den Berge*, Victor Guryev*
*shared last author
Submitted
Chapter 8
ABSTRACT
Background
Inhaled corticosteroids (ICS) improve asthma control to varying extents. Gene expression profiling in bronchial tissue has revealed a profile related to ICS-response, but access to this tissue is difficult. We recently showed that nasal gene expression can serve as a proxy for the lower airways, thereby providing a more accessible sampling alternative.
Objective
To investigate the effects of ICS-treatment and ICS-withdrawal on gene expression in nasal epithelium.
Methods
In a meta-analysis, we analyzed changes in nasal gene expression in 2 asthma cohorts with 2-week HFA-beclomethasone treatment (200µg b.i.d, n=39) and 12-week budesonide (400µg b.i.d, n=28) and compared results with ICS-withdrawal. Finally, we related our findings to those in bronchial biopsies in 2 independent asthma cohorts treated with ICS (n=12 and n=20) and in corticosteroid-treated air-liquid-interface (ALI) cultures of human bronchial epithelial cells (n=6) using Gene Set Enrichment Analysis (GSEA).
Results
In nasal epithelium, 76 genes were up- and 59 downregulated with ICS-treatment (FDR meta-analysis<0.05; nominal p-value in both cohorts<0.05). Downregulated genes after treatment significantly overlapped with upregulated genes after ICS-withdrawal (FDRGSEA=0.03). ICS-induced nasal gene expression changes overlapped with corticosteroid-induced gene expression changes in bronchial biopsies (FDRGSEA<0.01 for up- and downregulated genes) and ALI-cultures (FDRGSEA<0.01 and FDRGSEA=0.07 for up- and downregulated genes, respectively).
Conclusion
We show that nasal gene expression is dynamic with ICS-treatment in asthma and reflects ICS-induced gene expression changes in the lower airways. Our findings open new avenues for using nasal gene expression to explore mechanisms underlying ICS-therapy or to apply as a biomarker guiding ICS-ICS-therapy.
8
INTRODUCTION
For decades, inhaled corticosteroids (ICS) have been positioned as the mainstay of asthma treatment. ICS reduce airway inflammation and improve airway hyperresponsiveness(1,2). Moreover, ICS improve asthma control and protect against exacerbations and mortality(3,4).
Asthma patients benefit to varying degrees from ICS treatment(5,6). Currently, there is a limited number of biomarkers available that can help clinicians predict individual ICS response or monitor ICS therapy. One example of a biomarker used for guiding ICS treatment is the level of sputum eosinophils in asthma(7). Although this method has been proven effective in preventing exacerbations, its implementation in routine daily practice is hardly feasible. This emphasizes that a biomarker should not only be accurate and reliable, but also easily applicable. In this respect, biomarkers based on nasal epithelial gene expression, which is easily accessible, offer an appealing alternative for treatment monitoring.
We have previously shown that nasal gene expression may have broad applicability in differentiating disease states and reflecting the changes occurring in the difficult-to-access lower airways(8,9). For instance, a nasal epithelial gene expression profile distinguished COPD patients from healthy controls and overlapped with COPD-associated bronchial epithelial gene expression changes(8). Moreover, we recently demonstrated in individuals without airway disease that smoking-induced gene expression changes overlap between nasal and bronchial samples(9). Further, we have found that ICS treatment affects the gene expression profile in the bronchial airways of COPD patients(10). In asthma, genome-wide gene expression profiling of bronchial epithelium before and after 1-week ICS treatment showed differential expression of 30 genes with a decrease in SERPINB2 expression being significantly associated with improvement in pulmonary function(11). Together, these findings led us to hypothesize that, next to bronchial epithelium, the nasal epithelium can potentially be used to study the effects of ICS, to monitor ICS treatment or to identify those asthma patients who will or will not benefit from ICS.
In the current study, we investigated whether nasal gene expression is affected by ICS treatment in asthma patients. We confirmed our results both in nasal brushes of patients after ICS-withdrawal and in bronchial biopsies and air-liquid-interface cultures after corticosteroid-treatment. Finally, we explored the relation between changes in nasal gene expression and improvement in forced expiratory volume in 1 second (FEV1) with ICS treatment.
METHODS
Study populations and designs
Data were collected from asthma patients participating in two cohorts: the OLiVIA study (Effects of Extrafine Particle HFA-Beclomethasone versus Coarse Particle Treatment in Smokers and Ex-smokers with Asthma; ClinicalTrials.gov number NCT01741285) and the NZRHS (New Zealand Respiratory Health Survey; Australian New Zealand Clinical Trials Registry number ACTRN12610000666022). Both studies were approved by the local ethics committees and all participants gave their written informed consent.
Figure 1. Flowchart of the A) OLiVIA-study and B) NZRHS-study. *participants involved in NZRHS had not
used ICS at least 90 days prior to the start of the trial; ICS = inhaled corticosteroids.
The OLiVIA study investigated the effects of extrafine beclomethasone (HFA-QVAR) versus non-extrafine beclomethasone (HFA-Clenil) and fluticasone (HFA-Flixotide) in smokers and ex-smokers with asthma in an open-label, three-way cross-over design(12). It included participants with a doctor’s diagnosis of asthma, aged 18-65 years who were
8
current- or ex-smokers (smoking cessation≥6 months) with >5 pack-years. Exclusioncriteria included treatment with oral steroids, an FEV1 ≤1.2L and an upper respiratory tract infection (≤4 weeks prior to inclusion) or asthma exacerbation (≤6 weeks prior to inclusion). In OLiVIA, patients on ICS at the screening visit underwent an ICS washout period (4-6 weeks) prior to initiation, after which they received study medication for 2 weeks (Figure 1)(12).
The NZRHS study recruited participants aged 18-75 years through a random sample of the electoral roll in New Zealand who were sent a questionnaire on respiratory symptoms(13). Participants reporting symptoms of wheeze and breathlessness were invited for detailed phenotyping, and a subset without ICS use in the past 90 days were then treated with dry-powder formulation inhaled budesonide 400 µg b.i.d for 12 weeks. For this analysis, we included NZRHS participants who had both a doctor’s diagnosis of asthma and nasal brush samples pre- and post-ICS therapy (Figure 1).
Nasal brushes
Nasal epithelial brushes were obtained by brushing the inferior turbinate before and after ICS treatment(8). From OLiVIA, we used brushes taken after treatment with QVAR. For one patient a nasal brush after QVAR was not available, instead we used a nasal brush after treatment with Clenil 400µg b.i.d.. In OLiVIA, participants who used ICS or other asthma medication prior to the study received an additional nasal brush at the screening visit (Figure 1), followed by a withdrawal period of 4-6 weeks for ICS, nasal corticosteroids, long-acting-β2-agonists, long-acting anticholinergic agents, theophylline, leukotriene antagonists and antihistamines (Figure 1A). In NZRHS, participants had not used ICS for at least 90 days prior to start of the trial (Figure 1B).
Measurements
In both studies, spirometry was performed at all visits according to international guidelines(14).
Sample preparation, RNA sequencing, quality control and alignment
- OLiVIA
Total RNA was extracted using the miRNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. We assessed RNA quantity and integrity with the Agilent 2100 BioAnalyzer (Agilent, Santa Clara, CA, USA). Library preparation was performed using the NEXTflex Rapid Directional RNA-Seq kit with poly(A) selection (Bioo Scientific Corporation, Austin, TX, USA) using the Caliper Sciclone NGS Workstation
(PerkinElmer, Waltham, MA, USA). Samples were single-end sequenced (1x75bp) using an Illumina NextSeq500 sequencer. Quality control of the RNA sequencing data was done with FastQC (version 0.11.5). Alignment to human reference genome build GRCh38 was performed using STAR aligner (version 2.5.3a)(15) and Ensembl gene annotation (release 88, http://www.ensembl.org). Quantification was performed using HTSeq (version 0.6.1p1)(16). Additionally, we confirmed concordance between reported gender and gender-associated gene expression (XIST and Y-chromosomal genes). We filtered out lowly expressed genes by removing those with a mean expression less than 1 fragment per million (FPM) across all samples. Finally, RNA-seq data were log-transformed and normalized using the voom-function in the limma package (version 3.30.13) of the R statistical software (version 3.3.2).
- NZRHS
NZRHS samples were processed in accordance with previously described methods (8,17). In short, total RNA was isolated using the miRNeasy kit (Qiagen, Hilden, Germany). Microarray hybridization to Affymetrix Human Gene 1.0 ST Arrays was performed (Affymetrix, Santa Clara, CA, USA), followed by scanning of the microarrays using Affymetrix GeneArray Scanner 30007G Plus (Affymetrix, Santa Clara, CA, USA). Normalization was done using the Robust Multichip Average algorithm of the affy package of the R statistical software.
Statistics
All analyses were performed using R statistical software version 3.3.2. We performed principal component analysis and relative log expression plots. We used limma package version 3.30.13 with a paired sample design, adjusting for age, gender, smoking status and RNA integrity number to assess ICS-induced gene expression changes in OLiVIA and NZRHS. To identify genes that are universally changed by ICS in both studies, we performed a meta-analysis on results from OLiVIA and NZRHS by combining weighted z-scores. P-values were adjusted for multiple testing using the Benjamini-Hochberg procedure (FDR)(18). Statistically significant genes were defined by an FDR<0.05 in the meta-analysis combined with a nominal p-value<0.05 and similar direction of effect in both studies. In addition, we assessed gene expression changes after withdrawal of ICS in OLiVIA and compared results with the genes identified in the meta-analysis using Gene Set Enrichment Analysis (GSEA) version 3.0(19). Finally, we investigated in a meta-analysis of OLiVIA and NZRHS whether ICS-induced gene expression changes were associated with ICS-induced changes in pre-bronchodilator FEV1 (mL), adjusting for age, gender, height and smoking status.
8
Comparison of induced changes in nasal gene expression with ICS-induced changes in bronchial gene expression
We compared ICS-induced nasal gene expression changes identified in the meta-analysis with ICS-induced bronchial gene expression changes in two independent asthma datasets using GSEA. In the first dataset, bronchial biopsies were taken at baseline and after 8-week treatment with budesonide 180µg b.i.d.(20). In the second dataset, bronchial biopsies were taken at baseline and after 1-week treatment with fluticasone 500µg b.i.d. (GEO accession GSE4302)(11).
Comparison of ICS-induced changes in nasal gene expression with corticosteroid-induced changes in gene expression in air-liquid-interface (ALI) cultures of primary human bronchial epithelial cells
Detailed methods are available in the online supplement. In short, bronchial brushes were obtained from 2 healthy participants, 2 participants with asthma and 2 participants with remission of asthma (i.e. no respiratory symptoms or asthma medication; characteristics are shown in Table E1) (ClinicalTrials.gov Identifier: NCT03141814). The samples were brought into culture directly after the bronchoscopy. Then, the cells were exposed to air in an ALI culture and were cultured for 28 days. On day 28, airway epithelial cells were quiesced with Bronchial Epithelial Cell Growth Medium (BEBM) for 24h and treated with BEBM with or without Fluticasone Propionate 10-8M for 24h. Next, RNA was extracted using the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. RNA quality was determined by Agilent 2100 BioAnalyzer (Agilent, Santa Clara, USA). Library preparation was conducted using the NEXTflex Rapid Directional RNA-Seq kit (Bioo Scientific Corporation, Austin, TX, USA). Library quality was determined by Agilent 2100 Bioanalyzer. Libraries were pooled and ribosomal RNA was removed using R-probes from the SMARTer Stranded Total RNA-Seq Kit (Takara Bio USA Inc, Mountain View, USA). Samples were single-end sequenced (1x75bp) using an Illumina NextSeq500 sequencer. Quality control was done with FastQC (version 0.11.5). Alignment to human reference genome build GRCh38 was performed using STAR aligner (version 2.5.3a)(15) and Ensembl gene annotation (release 88, http://www. ensembl.org). Quantification was performed using HTSeq (version 0.6.1p1)(16). Genes with a mean expression less than 1 FPM across all samples were removed. RNA-seq data were log-transformed and normalized using the voom-function in the limma package (version 3.30.13) of the R statistical software (version 3.3.2). RNA-seq data was analyzed with limma using a paired analysis correcting for age, gender, disease status and the first principal component. We compared results with our findings in nasal epithelium by exploring direct overlap and by performing GSEA.
RESULTS
Seventeen samples were excluded due to poor quality metrics. Paired samples were available at baseline and after ICS treatment in 39 asthmatics in OLiVIA and 28 in NZRHS. Additionally, paired samples at the screening visit and at baseline (after ICS withdrawal) were available in 34 asthmatics (OLiVIA). Characteristics of the study populations are presented in Table 1; OLiVIA consisted of ex-smokers and smokers only, while the majority of participants in NZRHS were never-smokers.
Table 1. Characteristics of the study populations
Nasal epithelium Bronchial epithelium ICS
treatment withdrawalICS treatmentICS OLiVIA
(n=39) NZRHS (n=28) OLiVIA (n=34) Dataset 1 (n=12) Dataset 2 (n=20)
Male, n (%) 18 (46) 16 (57) 16 (47) 7 (58) 9 (45) Age, years 44 (13) 45 (12) 45 (12) 27 (23-50)* 34 (27-41)* BMI, kg m-2 26.6 (5.4) 28.9 (6.6) 27.7 (5.2) 27.6 (24.1-30.1)* 28.4 (24.2-32.3)* Smoking Current, n (%) 20 (51) 2 (7) 14 (41) 0 (0) 0 (0) Ex, n (%) 19 (49) 10 (36) 20 (59) 0 (0) 0 (0) Never, n (%) 0 (0) 16 (57) 0 (0) 12 (100) 20 (100) Packyears‡ 15.0 (11.0-25.3)* 11.0 (2.6-15.9)* 15.9 (11.0-23.8)* -
-Age of onset DDA 20 (6-41)* 19 (8-30)* 22 (6-41)* n/a n/a
FEV1, % predicted 84 (14) 88 (14) 85 (16) 86 (13) 85 (12)
FEV1/FVC, % 68 (10) 75 (11) 68 (10) 72 (6) 70 (9)
ICS use at screening, n (%) 29 (74) 0 (0) 34 (100) n/a n/a
Data are presented as mean with standard deviation, unless stated otherwise: *Median with Interquartile range; ‡Only
calculated for current and ex-smokers; BMI = body mass index; DDA = Doctor’s diagnosis of asthma; FEV1 = forced expiratory
volume in 1 second; FVC = forced vital capacity; ICS = inhaled corticosteroids; n/a = data not available.
Meta-analysis OLiVIA and NZRHS
We identified 135 genes in our meta-analysis that significantly changed in expression with ICS treatment, 79 being up- and 56 downregulated (FDR meta-analysis 0.05 and nominal p<0.05 in both studies; Figure 2). Among upregulated genes were FKPB5 and
CD163, and among downregulated genes were TRADD and CST1. Table 2 shows the
top 10 up- and downregulated genes, table E2 in the online supplement lists all 135 differentially expressed genes.
8
Figure 2. Delta heatmap showing the change in expression of 135 differential expressed genes in nasal epithelium after ICS treatment (log2 transformed fold change). Patients are shown in columns and genes
are shown in rows. A blue rectangle reflects downregulation of a gene after ICS treatment in a subject while a red rectangle reflects upregulation of a gene after ICS treatment in a subject.
Table 2. Top 10 up- and downregulated genes in nasal epithelium after ICS treatment
Top-10 upregulated genes Top-10 downregulated genes meta z-score FDR meta z-score FDR
CHPT1 5.913 <0.001 ATP8B2 -5.737 <0.001 TP53I3 5.676 <0.001 FHL1 -4.541 0.005 PLPP1 5.010 0.002 TRADD -4.482 0.006 FKBP5 4.959 0.002 CST1 -4.323 0.009 CD163 4.930 0.002 TRPV4 -4.261 0.011 ATAD2 4.921 0.002 C2orf82 -4.199 0.012 SPTSSA 4.822 0.002 P4HA2 -4.183 0.012 VSIG4 4.775 0.002 NRXN3 -4.137 0.013 RIPK1 4.755 0.002 CARMIL2 -4.091 0.013 HOMER2 4.538 0.005 CACNB3 -4.069 0.013
meta z-score = z-score meta-analysis; FDR = false discovery rate
Comparison of gene expression changes after ICS treatment and after ICS withdrawal in OLiVIA
To confirm that genes changing with ICS treatment reverse after ICS withdrawal, we compared the 135 ICS-induced genes with gene expression changes after ICS withdrawal in OLiVIA by GSEA. Genes downregulated after ICS treatment (n=56) were significantly enriched among genes upregulated after ICS withdrawal (FDR<0.05; Figure 3A). Genes
upregulated after ICS treatment (n=79) were not significantly enriched among genes downregulated after ICS withdrawal (FDR 0.69; Figure 3B).
Figure 3. Gene set enrichment analysis showing that A) 56 genes downregulated in nasal epithelium after ICS treatment are significantly enriched among genes upregulated after withdrawal of ICS (FDR=0.03) and B) 79 genes upregulated in nasal epithelium after ICS treatment are not significantly enriched among genes downregulated after withdrawal of ICS (FDR=0.69). The colored bar represents
ranked t values of the association of change in gene expression in nasal epithelium with withdrawal of ICS: a red bar represents a positive association while a blue bar represents a negative association. Black lines each represent a differential expressed gene after ICS treatment identified in the meta-analysis. The height of the black lines reflects the running enrichment scores of the gene set enrichment analysis.
Comparison with ICS-induced changes in bronchial gene expression
We compared our findings in nasal epithelium with those in bronchial biopsies of 2 independent datasets of asthma patients treated with ICS. GSEA shows that genes upregulated after ICS treatment in nasal epithelium were significantly enriched among genes upregulated after ICS treatment in bronchial biopsies in both datasets (Figure 4A and 4C). Additionally, genes downregulated after ICS treatment in nasal epithelium were significantly enriched among genes downregulated after ICS treatment in bronchial biopsies in both datasets (Figure 4B and 4D). Core-enriched upregulated genes (genes that were upregulated in both nasal- and bronchial samples after ICS treatment) were among others FKBP5, CD163 and MUC1 (Table E3). Core-enriched downregulated genes (genes that were downregulated in both nasal- and bronchial samples after ICS treatment) were among others CST1 and NRXN3 (Table E3).
8
Figure 4. Gene set enrichment analysis showing that 79 genes upregulated in nasal epithelium after ICS treatment are significantly enriched among genes upregulated in bronchial biopsies of asthma patients after ICS treatment (A and C; FDR<0.001) and 56 genes downregulated in nasal epithelium after ICS treatment are significantly enriched among genes downregulated in bronchial biopsies of asthma patients after ICS treatment (B; FDR<0.001 and D; FDR=0.002). The colored bar represents ranked
t statistics values of the association of change in gene expression in bronchial biopsies with ICS treatment: a red bar represents a positive association while a blue bar represents a negative association. Black lines each represent a differential expressed gene after ICS treatment identified in the meta-analysis. The height of the black lines reflects the running enrichment scores of the gene set enrichment analysis.
Comparison with corticosteroid-induced changes in gene expression in ALI cultures of primary human bronchial epithelial cells
After filtering out lowly expressed genes in the ALI cultures, 64 out of 79 upregulated genes and 19 out of 56 downregulated genes were available for validation. Of these 64 genes that were upregulated by ICS in nasal epithelium, 9 (14%) were also significantly upregulated in corticosteroid-treated ALI cultured human bronchial epithelial cells, such as FKBP5, ATAD2 and SPTSSA (limma paired-analysis p-value<0.05, Table E4). Of the 19 genes downregulated by ICS, 6 (32%) were also significantly downregulated in the ALI cultures, such as TMEM63A and GDF11 (p-value<0.05, Table E4). Using GSEA, we confirmed that upregulated genes in nasal epithelium after ICS treatment significantly overlap with upregulated genes in the corticosteroid-treated ALI cultures (FDR<0.001, Figure 5A). For downregulated genes, we observed a trend toward significant overlap between downregulated genes in nasal epithelium and downregulated genes in corticosteroid-treated ALI cultures (FDR=0.07, Figure 5B).
Figure 5. Gene set enrichment analysis showing: A) significant enrichment of genes upregulated in nasal epithelium after ICS treatment among genes upregulated in corticosteroid-treated ALIs (FDR<0.001) and B) a trend towards enrichment of downregulated genes in nasal epithelium after ICS treatment among genes downregulated in corticosteroid-treated ALIs (FDR=0.07). The colored bar represents
ranked t values of the association of change in gene expression in ALIs after corticosteroid-treatment: a red bar represents a positive association while a blue bar represents a negative association. Black lines each represent a differential expressed gene after ICS treatment identified in the meta-analysis. The height of the black lines reflects the running enrichment scores of the gene set enrichment analysis.
8
Association of ICS-induced changes in nasal gene expression with changes in FEV1
We correlated the change in expression of the 135 differentially expressed genes in nasal epithelium with the change in FEV1 in a meta-analysis of OLiVIA and NZRHS. The change in expression of 7 genes was associated with change in FEV1 (p-value<0.05), but none of these genes showed a concordant p-value<0.05 in both OLiVIA and NZRHS (online supplement table E5).
DISCUSSION
We show that the nasal epithelium is a suitable site to study gene expression changes induced by ICS treatment in asthma patients. In a meta-analysis of 2 independent studies, we identified 135 genes that significantly changed in expression with ICS treatment, 79 being up- and 56 downregulated. Confirming the robustness of our findings, genes downregulated after ICS treatment were commonly upregulated after ICS withdrawal. In addition, ICS-induced gene expression changes in nasal epithelium considerably overlapped with ICS-induced gene expression changes in bronchial epithelium in 2 independent asthma cohorts as well as in corticosteroid-treated ALI-cultures of human bronchial epithelial cells.
We found 79 genes upregulated in nasal epithelium after ICS treatment. Among the top-10 upregulated genes were FKPB5 and CD163. The protein coded by FKBP5 is known to be upregulated by corticosteroids and functions as an inhibitor of the glucocorticosteroid receptor, thus playing a role in corticosteroid sensitivity(21). Woodruff et al showed that higher bronchial epithelial FKBP5 expression in asthma patients before ICS treatment was associated with less improvement in FEV1 after treatment with inhaled budesonide(11). Other studies investigating nasal and bronchial epithelium confirm our finding that
FKBP5 is upregulated in response to corticosteroids in healthy participants(22), and
those with asthma(11,23,24) and COPD(10). CD163 is a hemoglobin scavenger receptor exclusively expressed on monocytes and macrophages, and is a marker of alternatively activated M2 macrophages(25). Glucocorticosteroids are, next to the Th2-cytokines IL-4 and IL-13, potent stimuli for M2 macrophage activation which results in upregulation of
CD163(26,27). M2 macrophages are involved in allergic inflammation, wound healing
and downregulation of inflammation by releasing the anti-inflammatory cytokine IL-10(28). In a study investigating alveolar macrophages in bronchoalveolar lavage fluid (BALF), lower cell-surface expression of CD163 was observed in asthma patients compared to controls(29). Of interest, CD163-knockout mice sensitized and challenged with house dust mite exhibited more eosinophilic inflammation in BALF compared
to wild type controls(29). Together this suggests that a reduced activation of CD163 in macrophages might play a role in the pathophysiology of asthma, which can be reverted by corticosteroids.
A total of 56 genes were downregulated in nasal epithelium after ICS treatment. Among the top-10 downregulated genes were CST1 and TRADD. CST1 codes for the protein cystatin SN, a cysteine protease inhibitor, which is a member of the cystatin superfamily.
CST1 has been found to be upregulated in the nasal epithelium of patients with allergic
rhinitis compared to controls, suggesting a role in allergic airway inflammation(30,31). The protein coded by TRADD, Tumor Necrosis Factor (TNF) Receptor Type 1 Associated Death Domain Protein, is part of the TNF-α/NFκB signaling pathway. In a study by Lee
et al, treatment of human airway epithelial cells with the drug verproside disturbed
the interaction between the TNF-α receptor and TRADD, leading to decreased NFκB-activation and subsequently decreased MUC5AC production, which is one of the main components of mucus(32). The authors state that verproside is an interesting candidate drug for treatment of obstructive lung diseases. Taken together, the findings of Lee et al show that impaired TRADD functioning leads to decreased NFκB-activation. Corticosteroids are also well known to suppress NFκB-activity. Therefore, our findings are in agreement with Lee et al; we hypothesize that one of the mechanisms by which corticosteroids inhibit NFκB-activation, is by inhibiting TRADD expression.
Genes downregulated in nasal epithelium after ICS treatment, were significantly enriched among genes upregulated after ICS withdrawal. The finding that these genes change in opposite directions after ICS treatment and ICS withdrawal makes it more likely that they are truly affected by ICS. For genes upregulated in nasal epithelium after ICS treatment, there was not a significant enrichment amongst genes downregulated after ICS withdrawal in the GSEA analysis. The latter results might be explained by the fact that we compared genes affected by ICS treatment in the meta-analysis with genes affected by ICS withdrawal in a subset of the OLiVIA study (n=34, of which only 26 overlapped with the ICS treatment population), since not all participants in OLiVIA used ICS at the screening visit and data on ICS withdrawal was not available in NZRHS. Genes changed by ICS in nasal epithelium, strongly overlap with genes changed by ICS in bronchial epithelium in 2 independent cohorts. This is in line with two recent studies from our group, in which strong resemblance was shown between the nasal and bronchial gene expression profiles associated with COPD(8) and with smoking(9). The overlap between ICS-induced nasal and bronchial gene expression can be explained by multiple factors: 1) ICS are inhaled through the mouth but can be exhaled through
8
the nose, which induces direct exposure of the nasal epithelium to corticosteroids, 2)epithelial cells communicate with each other through excretion of cytokines and other mediating molecules, leading to a similar gene expression profile throughout the airways.
In addition, validation of our findings in corticosteroid-treated ALI cultures using GSEA showed a strong overlap between upregulated genes in nasal brushes after ICS treatment and upregulated genes in corticosteroid-treated ALIs. ALIs solely contain epithelial cells and no other cell types such as fibroblasts or inflammatory cells, which can be present in nasal brushes. Therefore, we confirmed for a subset of genes that the expression changes found in the nasal epithelium truly reflect expression changes of epithelial cells, and not a change in cell type composition induced by corticosteroids. GSEA did not show significant overlap of downregulated genes in nasal brushes with downregulated genes in ALI cultures. This might be due to the fact that many of the downregulated genes in nasal brushes were already lowly expressed in ALIs at baseline. A possible explanation for this finding is that these genes are not expressed by epithelial cells but by inflammatory cells that are also present in nasal brushes.
We did not find an association between ICS-induced changes in gene expression in the nasal epithelium and changes in FEV1. This contrasts with findings of Woodruff et al, who showed that a decrease in expression of SERPINB2 in the bronchial epithelium of asthma patients after ICS treatment was associated with improvement in FEV1(11). The discrepant findings can have multiple explanations. First, the nasal epithelium may not be a suitable site to reflect dynamic changes in lung function at a physiological level. Second, FEV1 might not be the appropriate pulmonary function measure to monitor ICS therapy. This is underscored by the observation that a considerable subset of our study population did not show improvement in FEV1, the median change in FEV1 being -40mL [IQR -125mL; +105mL] in OLiVIA and +40 mL [IQR -23mL; +193mL] in NZRHS. Third, in the study by Woodruff et al, participants were treated with fluticasone 500µg b.i.d., an ICS-dose twice as high as applied in OLiVIA and NZRHS, which might have led to more pronounced gene expression changes leading to significant associations with FEV1. A strength of our study is the comparison of changes in nasal gene expression after ICS treatment and ICS withdrawal, as well as the comparison of our findings with corticosteroid-induced bronchial gene expression changes in 2 independent asthma cohorts of bronchial biopsies as well as ALI cultures of human bronchial epithelial cells. The strong overlap found in these analyses strengthens the validity of our findings. A limitation of our study is that the study design and sampled populations (OLiVIA and
NZRHS) differ in several aspects. In OLiVIA, only smokers and ex-smokers with asthma were included and treatment duration was 2 weeks. In NZRHS, almost all participants were non-smokers and treatment duration was 3 months. We addressed these differences by performing a meta-analysis with the aim to identify genes universally changed by ICS, irrespective of smoking status or treatment duration. A drawback of this approach is that we might have missed genes that are affected by ICS in non-smokers only, or are only changed after a longer treatment period.
In summary, we show that nasal epithelial gene expression is dynamic, changes with ICS treatment in asthma patients and can be used as a proxy for the bronchial epithelium to investigate ICS-induced gene expression changes. Obtaining a nasal brush is a minimally invasive procedure, which takes no more than 10 minutes and is generally well-tolerated by patients. This opens the door for future applications of nasal gene expression in asthma, such as predicting response to therapy, monitoring therapy and developing novel treatments.
SUPPORT
The OLiVIA study was funded by TEVA pharmaceutical Industries Ltd. The NZRHS study was funded by the Health Research Council of New Zealand (grant no 10/174), Astra Zeneca Ltd and Genentech.
8
REFERENCES
(1) Childhood Asthma Management Program Research Group, Szefler S, Weiss S, Tonascia J, Adkinson NF, Bender B, et al. Long-term effects of budesonide or nedocromil in children with asthma. N Engl J Med. 2000;343(15):1054–63.
(2) van den Berge M, Kerstjens HA, Meijer RJ, de Reus DM, Koëter GH, Kauffman HF, et al. Corticosteroid-induced improvement in the PC20 of adenosine monophosphate is more closely associated with reduction in airway inflammation than improvement in the PC20 of methacholine. Am J Respir Crit Care Med. 2001 Oct 1;164(7):1127−32.
(3) Suissa S, Ernst P, Benayoun S, Baltzan M, Cai B. Low-dose inhaled corticosteroids and the prevention of death from asthma. N Engl J Med. 2000 Aug 3;343(5):332–6.
(4) Pauwels RA, Pedersen S, Busse WW, Tan WC, Chen Y-Z, Ohlsson S V, et al. Early intervention with budesonide in mild persistent asthma: a randomised, double-blind trial. Lancet. 2003;361(9363):1071– 6.
(5) Thomson NC, Spears M. The influence of smoking on the treatment response in patients with asthma. Curr Opin Allergy Clin Immunol. 2005 Feb;5(1):57–63.
(6) Chanez P, Wenzel SE, Anderson GP, Anto JM, Bel EH, Boulet LP, et al. Severe asthma in adults: What are the important questions? J Allergy Clin Immunol. 2007;119(6):1337–48.
(7) Green RH, Brightling CE, McKenna S, Hargadon B, Parker D, Bradding P, et al. Asthma exacerbations and sputum eosinophil counts: A randomised controlled trial. Lancet. 2002;360(9347):1715–21.
(8) Boudewijn IM, Faiz A, Steiling K, van der Wiel E, Telenga ED, Hoonhorst SJM, et al. Nasal gene expression differentiates COPD from controls and overlaps bronchial gene expression. Respir Res. 2017 Dec 21;18(1):213.
(9) Imkamp K, Berg M, Vermeulen CJ, Heijink IH, Guryev V, Kerstjens HA, et al. Nasal epithelium as a proxy for bronchial epithelium for smoking-induced gene expression and expression Quantitative Trait Loci. J Allergy Clin Immunol. 2018 Mar;pii: S0091-6749(18)30324-5.
(10) van den Berge M, Steiling K, Timens W, Hiemstra PS, Sterk PJ, Heijink IH, et al. Airway gene expression in COPD is dynamic with inhaled corticosteroid treatment and reflects biological pathways associated with disease activity. Thorax. 2014 Jan 7;69(1):14–23.
(11) Woodruff PG, Boushey H a, Dolganov GM, Barker CS, Yang YH, Donnelly S, et al. Genome-wide profiling identifies epithelial cell genes associated with asthma and with treatment response to corticosteroids. Proc Natl Acad Sci USA. 2007;104(40):15858–63.
(12) Cox CA, Boudewijn IM, Vroegop SJ, Schokker S, Lexmond AJ, Frijlink HW, et al. Extrafine compared to non-extrafine particle inhaled corticosteroids in smokers and ex-smokers with asthma. Respir Med. 2017;130:35–42.
(13) Fingleton J, Travers J, Williams M, Charles T, Bowles D, Strik R, et al. Treatment responsiveness of phenotypes of symptomatic airways obstruction in adults. J Allergy Clin Immunol. 2015;136(3):601–9. (14) Miller MR, Hankinson J, Brusasco V, Burgos F, Casaburi R, Coates A, et al. Standardisation of spirometry.
Vol. 26, European Respiratory Journal. 2005. p. 319–38.
(15) Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, et al. STAR: Ultrafast universal RNA-seq aligner. Bioinformatics. 2013;29(1):15–21.
(16) Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015 Jan 15;31(2):166–9.
(17) Steiling K, van den Berge M, Hijazi K, Florido R, Campbell J, Liu G, et al. A dynamic bronchial airway gene expression signature of chronic obstructive pulmonary disease and lung function impairment. Am J Respir Crit Care Med. 2013 May 1;187(9):933–42.
(18) Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. J R Stat Soc Ser B. 1995;57(1):289–300.
(19) Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci. 2005;102(43):15545–50.
(20) Bhakta NR, Christenson SA, Nerella S, Solberg OD, Nguyen CP, Choy DF, et al. IFN-stimulated gene expression, type 2 inflammation, and endoplasmic reticulum stress in asthma. Am J Respir Crit Care Med. 2018;197(3):313–24.
(21) Fries G, Gassen N, Rein T. The FKBP51 Glucocorticoid Receptor Co-Chaperone: Regulation, Function, and Implications in Health and Disease. Int J Mol Sci. 2017;18(12):2614.
(22) Leigh R, Mostafa MM, King EM, Rider CF, Shah S, Dumonceaux C, et al. An inhaled dose of budesonide induces genes involved in transcription and signaling in the human airways: enhancement of anti- and proinflammatory effector genes. Pharmacol Res Perspect. 2016;4(4):1–18.
(23) Kelly MM, King EM, Rider CF, Gwozd C, Holden NS, Eddleston J, et al. Corticosteroid-induced gene expression in allergen-challenged asthmatic subjects taking inhaled budesonide. Br J Pharmacol. 2012;165(6):1737–47.
(24) Fayon M, Lacoste-Rodrigues A, Barat P, Helbling JC, Nacka F, Berger P, et al. Nasal airway epithelial cell IL-6 and FKBP51 gene expression and steroid sensitivity in asthmatic children. PLoS One. 2017;12(5):7– 14.
(25) Rőszer T. Understanding the Mysterious M2 Macrophage through Activation Markers and Effector Mechanisms. Mediators Inflamm. 2015;2015:816460.
(26) Martinez FO, Gordon S. The M1 and M2 paradigm of macrophage activation: time for reassessment. F1000Prime Rep. 2014;6(March):1–13.
(27) Ehrchen J, Steinmüller L, Barczyk K, Tenbrock K, Nacken W, Eisenacher M, et al. Glucocorticoids induce differentiation of a specifically activated, anti-inflammatory subtype of human monocytes. Blood. 2007;109(3):1265−74.
(28) Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol. 2011 Oct 14;11(11):723–37.
(29) Dai C, Yao X, Gordon EM, Barochia A, Cuento RA, Kaler M, et al. A CCL24-dependent pathway augments eosinophilic airway inflammation in house dust mite-challenged Cd163 -/- mice. Mucosal Immunol. 2016;9(3):702–17.
(30) Imoto Y, Tokunaga T, Matsumoto Y, Hamada Y, Ono M, Yamada T, et al. Cystatin SN Upregulation in Patients with Seasonal Allergic Rhinitis. PLoS One. 2013;8(8).
(31) Giovannini-Chami L, Marcet B, Moreilhon C, Chevalier B, Illie MI, Lebrigand K, et al. Distinct epithelial gene expression phenotypes in childhood respiratory allergy. Eur Respir J. 2012;39(5):1197–205. (32) Lee SU, Sung MH, Ryu HW, Lee J, Kim HS, In HJ, et al. Verproside inhibits TNF-α-induced MUC5AC
expression through suppression of the TNF-α/NF-κB pathway in human airway epithelial cells. Cytokine. Elsevier Ltd; 2016;77:168–75.
8
SUPPLEMENTAL MATERIAL
Supplemental material is online available by scanning the following QR-code or using the URL: www.publicatie-online.nl/publicaties/i-boudewijn/chapter-8-supplemental/
