UBC Faculty Research and Publications

Gene expression analysis in asthma using a targeted multiplex array Pascoe, Christopher D; Obeidat, Ma’en; Arsenault, Bryna A; Nie, Yunlong; Warner, Stephanie; Stefanowicz, Dorota; Wadsworth, Samuel J; Hirota, Jeremy A; Jasemine Yang, S.; Dorscheid, Delbert R; Carlsten, Chris; Hackett, Tillie L; Seow, Chun Y; Paré, Peter D Dec 11, 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12890_2017_Article_545.pdf [ 1.24MB ]
JSON: 52383-1.0361789.json
JSON-LD: 52383-1.0361789-ld.json
RDF/XML (Pretty): 52383-1.0361789-rdf.xml
RDF/JSON: 52383-1.0361789-rdf.json
Turtle: 52383-1.0361789-turtle.txt
N-Triples: 52383-1.0361789-rdf-ntriples.txt
Original Record: 52383-1.0361789-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessGene expression analysis in asthma using atargeted multiplex arrayChristopher D. Pascoe1,4,9*, Ma’en Obeidat1,4, Bryna A. Arsenault1,4, Yunlong Nie1,4, Stephanie Warner1,4,Dorota Stefanowicz1,4, Samuel J. Wadsworth1,4, Jeremy A. Hirota8, S. Jasemine Yang1,4, Delbert R. Dorscheid1,4,Chris Carlsten1,2,3,4,5, Tillie L. Hackett1,4,6, Chun Y. Seow1,4,7 and Peter D. Paré1,2,4AbstractBackground: Gene expression changes in the structural cells of the airways are thought to play a role in thedevelopment of asthma and airway hyperresponsiveness. This includes changes to smooth muscle contractilemachinery and epithelial barrier integrity genes. We used a targeted gene expression arrays to identify changes inthe expression and co-expression of genes important in asthma pathology.Methods: RNA was isolated from the airways of donor lungs from 12 patients with asthma (8 fatal) and 12 non-asthmatics controls and analyzed using a multiplexed, hypothesis-directed platform to detect differences in geneexpression. Genes were grouped according to their role in airway dysfunction: airway smooth muscle contraction,cytoskeleton structure and regulation, epithelial barrier function, innate and adaptive immunity, fibrosis andremodeling, and epigenetics.Results: Differential gene expression and gene co-expression analyses were used to identify disease associatedchanges in the airways of asthmatics. There was significantly decreased abundance of integrin beta 6 and Ras-Related C3 Botulinum Toxin Substrate 1 (RAC1) in the airways of asthmatics, genes which are known to play animportant role in barrier function. Significantly elevated levels of Collagen Type 1 Alpha 1 (COL1A1) and COL3A1which have been shown to modulate cell proliferation and inflammation, were found in asthmatic airways.Additionally, we identified patterns of differentially co-expressed genes related to pathways involved in virusrecognition and regulation of interferon production. 7 of 8 pairs of differentially co-expressed genes werefound to contain CCCTC-binding factor (CTCF) motifs in their upstream promoters.Conclusions: Changes in the abundance of genes involved in cell-cell and cell-matrix interactions could playan important role in regulating inflammation and remodeling in asthma. Additionally, our results suggest thatalterations to the binding site of the transcriptional regulator CTCF could drive changes in gene expression inasthmatic airways. Several asthma susceptibility loci are known to contain CTCF motifs and so understandingthe role of this transcription factor may expand our understanding of asthma pathophysiology and therapeutic options.Keywords: Asthma, Co-expression, Nanostring, Extracellular matrix, CTCF, Smooth muscle, Remodeling, Epithelium,Targeted expression* Correspondence: cpascoe@chrim.ca1UBC Institute for Heart Lung Health, St. Paul’s Hospital, 1081 Burrard St,Vancouver, BC, Canada4University of British Columbia Centre for Heart Lung Innovation, St. Paul’sHospital, 1081 Burrard St, Vancouver, BC, CanadaFull list of author information is available at the end of the article© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 DOI 10.1186/s12890-017-0545-9BackgroundAsthma is a chronic inflammatory disease of the airways,characterized by symptoms of breathlessness, wheezingand cough, associated with variable airflow limitation andairway hyperresponsiveness (AHR). Asthma is also charac-terized by airway remodeling which includes goblet cellmetaplasia, epithelial damage, subepithelial fibrosis, base-ment membrane thickening, and increased airway smoothmuscle (ASM) mass [1]. The pathogenesis of asthma is be-lieved to involve an interaction between the innate andadaptive immune systems [2], and phenotypic changeswithin the epithelial-mesenchymal trophic unit [3]. Gen-etic and genomic analyses have been used to discover themolecular mechanisms underlying these phenotypicchanges. Large genome-wide association studies (GWAS)have reproducibly identified single nucleotide polymor-phisms (SNPs) in or near genes predominantly expressedin the airway epithelium and immune cells as susceptibil-ity factors for asthma. The specific genes include gasder-min B (GSDMB) [4], interleukin 33 (IL33), and thymicstromal lymphopoietin (TSLP) [5]. Another candidategene ADAM33, expressed in ASM, has been shown to beassociated with asthma in linkage analysis [6]. Genetic var-iants associated with susceptibility for asthma may exerttheir effect by altering gene expression levels; indeed manyof the SNPs associated with asthma and AHR have beenshown to be expression quantitative trait loci (eQTL) inlung tissue, epithelial and blood cells, and altered proteinexpression of some of these genes has been found in cellsand tissue from asthmatic individuals [7].In this hypothesis driven study we used a reproduciblemultiplexed technology (Nanostring®) to quantify the ex-pression of 334 genes potentially involved in phenotypicchanges in asthmatic airways. This technology is highlysensitive, reproducible, and is suitable for archived tissuespecimens as it is insensitive to RNA degradation [8].We hypothesized that changes in the expression patternsof genes involved in ASM contraction, the cytoskeleton,epithelial barrier function, innate/adaptive immunity, fi-brosis and remodeling, and epigenetics would be presentin the airway tissue of asthmatics compared to non-asthmatics. Our results suggest that alterations in the ex-pression of genes involved in cell-cell and cell-matrix in-teractions may contribute to the pathogenesis of asthma,particularly severe asthma. The identification of alteredgene co-expression networks may identify changes intranscriptional regulation that could be pathogenic andmissed with commonly used analyses for differentialexpression.MethodsAdditional description of methods is provided in the on-line supplement.Subject selection and RNA isolationHuman lungs were donated with consent from the IIAMand used with approval from the University of BritishColumbia and St. Paul’s Hospital ethics committee.Diagnosis of asthma was determined through patientmedical history and asthma medication usage as deter-mined by family interview. Non-asthma donor deathswere primarily due to head trauma while 8 of the 12 do-nors with asthma died during exacerbations of theirasthma. The other four donors with asthma died due toother, accidental causes (eg. head trauma). Subject demo-graphics can be seen in Table 1 with full subject character-istics found in Additional file 1: Table S1. After surgicalremoval the lungs were flushed with Custodiol HTK solu-tion (Odyssey Pharmaceuticals, East Hanover, NJ, USA)and transported on ice by plane. The time between har-vesting and arrival at the University of British Columbiawas 15–20 h. Tissues from the lungs have been used inprevious studies [9–11]. Inflated frozen lungs were proc-essed into tissue cores for sectioning on a crytostat. Atotal of twenty 10 μm thick sections per core were cut andstored at −80 °C until RNA was isolated. Sections 1, 5, 10,15 and 20 were stained with hematoxylin and eosin (H&E)for morphometric measurements. For the remaining 15sections, airways and a small amount of surrounding par-enchyma were macroscopically dissected using a scalpel(Fisher Scientific® No. 11) for RNA isolation. Samples wereonly used if the airways seen on the first section were con-tinuous for the 20 sequential sections. Large vessels wereavoided. Sample airway is seen in Fig. 1. RNA was isolatedusing the Qiagen® RNeasy Mini Kit according to manufac-turers protocol.Characterization of airway dimensionsMeasurements of ASM, epithelial, collagen, and total wallarea in addition to basement membrane thickness weremeasured to quantify the degree of airway remodeling.The five sections stained with H&E (Fig. 2) from each corewere digitally scanned and airway wall compartmentswere quantified using the Aperio® system (Leica Biosys-tems, Germany). ASM area, epithelial area and total air-way wall area were quantified using a point countingmethod, where a grid of 4000 points was overlaid ontoeach airway of interest and the points falling on the areaof interest were counted (Image Pro Plus ®, Media Cyber-netics, Maryland). The measurements for all airways werenormalized to the internal perimeter (Pi) of the airwayand averaged across all the sections for each subject. ThePi was measured by tracing along the luminal side of theepithelium. For the measurement of basement membranethickness and collagen area, two additional sections werecut from each core (same as used for RNA isolation andother measurements) to stain with Masson’s trichrome(basement membrane) and Picrosirius red (collagen). ToPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 2 of 14quantify the basement membrane thickness, a randomseries of line segments was placed over each image andthe thickness of the basement membrane was measured atthe points where any line segment crossed the basementmembrane. All thickness measurements were made per-pendicular to the epithelium. A minimum of 40 measure-ments were made for each airway. For the measurementof collagen content, Picrosirius red stained slides were vi-sualized under polarized light where the Picrosirius redstain shows birefringence. The collagen appears red on ablack background and the amount of collagen was quanti-fied using color segmentation in Image Pro Plus®. All mea-surements were carried out in a blinded manner.Gene expression analysisExpression of mRNA for the 334 candidate genes and 12housekeeping genes was measured with the Nanostring®system using a custom codeset panel. The most stablehousekeeping genes were selected by measuring 12 com-mon housekeeping genes. By comparing the % Coefficientof variation (%CV) across the 12 housekeeping genes wewere able to determine that the 5 most stable genes for datanormalization were: RNA Polymerase R2A (POLR2A),TATA box binding protein (TBP), Ribosomal Protein L19(RPL19), Guanine nucleotide-binding protein subunit beta-2-like 1 (GNB2L1), and β-Glucuronidase (GUSB). These 5genes were also selected because they spanned a range ofcounts from low (average 198 counts for GUSB) to high(average 22,486 counts for RPL19). Data were normalizedin accordance with Nanostring® guidelines. See Additionalfile 1: Table S2 in the online supplement for a complete listof genes in the panel.Candidate gene selectionWe selected the candidate genes based on a priorihypotheses and grouped them based on their functionand potential role in the pathogenesis of asthma and/or AHR. These are: ASM contraction and relaxation,structure and regulation of the cytoskeleton, epithelialbarrier function, innate and adaptive immunity,fibrosis and remodeling and epigenetics. The rationalefor the choice of groups of genes is provided belowand the list of the genes by category is in Additionalfile 1: Table S21) The contraction and relaxation of ASM and itsregulation; MYH11, MLCK, SM-22 and actin havebeen previously examined in the context of asthma[12]. Many of the genes in the contractile machinerygroup were taken from the Kyoto Encyclopedia ofGenes and Genomes (KEGG) pathway for vascularsmooth muscle contraction. Additionally, work bySieck et al. [13] led to the selection of CD38 andother calcium handling proteins that have beenshown to be expressed in cultured ASM cells. Geneswithin this pathway are involved in either contrac-tion or relaxation of smooth muscle and so have thepotential to play a role in AHR in asthma.2) The structure and regulation of the cytoskeleton;genes within the cytoskeletal group were selectedbased on previous work by Gunst et al. [14]. Genesin this group have been shown to be important intransmitting ASM force to the external environmentat adherens junctions, in maintaining the actinfilament lattice or or regulating ASM stiffnessindependent of force generation [15].3) Epithelial barrier function; a number ofobservations suggest that the airway epithelium isdisrupted in asthma and that this may in part resultfrom altered repair mechanisms [16]. Disruptedfeatures include detachment of columnar ciliatedcells, the presence of epithelial cell aggregates(Creola bodies) in sputum, decreased expression ofepithelial cell-cell junction proteins (E-cadherin, ZO-Table 1 Patient demographics. Ages not significantly differentNon-asthmatic (n = 12) Asthmatic (n = 12)Median Age (Range) 21 (4–63) 17.5 (8–36)Male Sex – # (%) 6 (50) 7 (58.3)Average Weight - kg (±SEM) 75.7 (7.5) 68.8 (6.4)Inhaled Corticosteroids – #(%)0 (0) 6 (50)Smoking - # (%) 2 (16.7) 4 (33.3)End of life steroids - # (%) 4 (33.3) 8 (66.7)Fatal Asthma - # (%) 8 (66.7)Fig. 1 Section of a frozen human lung core cut at 10 μm thickness.Blue arrows indicate airways. Red arrow indicates a blood vessel. Theblack lines show the outline of the tissue taken for isolation of totalairway RNA. Scale bar = 5 mmPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 3 of 141, protocadherin-1) and increased expression of epi-thelial repair markers (TGF-β, EGFR and CD44),mucins, and altered expression of repair-associatedfucosylated glycoproteins [17]. A defective epithelialbarrier may have important consequences in asthmaas it is thought to lead to increased accessibility ofallergens to immune and structural cells within themucosal and submucosal spaces.4) Innate/Adaptive immunity; innate immunereceptors and related mediators have been implicatedin the pathogenesis of asthma (e.g. IL-33, TSLP, andST2) [18]. Furthermore, it has been demonstrated thatantiviral immune responses are compromised in airwayepithelial cells from asthmatics [19]. Intrinsicdifferences in innate immune responses in airwayepithelial cells may therefore contribute to diseasedevelopment and exacerbations in response toenvironmentalexposures including allergens, viruses, and air pollution[20]. We therefore determined the expression patternsof all Toll-like receptors (TLRs), Nod-like receptors(NLRs), and Rig-like receptors (RLRs) and relatedmediators to provide a comprehensive screen of thesecandidate genes in asthmatic and non-asthmatic airwaywall samples.5) Fibrosis and remodeling; the role of the myocardinpathway in proliferation of ASM cells [21] led to theselection of genes in this pathway. In addition, weinterrogated members of the Notch family as thesegenes are integral to airway development anddifferentiation [22] and may play important roles inasthma. ECM proteins are altered in the airways ofasthmatics and as such, we included a number ofgenes that code for ECM components thought to beinvolved in the remodeling of asthmatic airways. Invivo work has shown a role for matrixmetalloproteinases (MMPs) in the development ofairway inflammation and hyperresponsiveness [23].6) Epigenetics; as a first line of contact with theexternal environment, the airway epithelium is anattractive target for epigenetic research. Alterationsin DNA methylation and histone modifications havebeen reported in the airway epithelium of asthmaticsubjects [24], however many of the mediatorsinvolved have not been studied. We targeted thehistone acetyltransferases KAT2A, CREBBP, andEP300 as they are responsible for acetylating lysine18 on histone 3 which is up regulated in asthmaticepithelial cells [24]. We also focused on AURKA,PRMT5, SUV39H1, and HDAC10 which have beenidentified to be potentially involved in thepathogenesis of asthma based on preliminary datafrom an array analysis of epigenetic modifyingenzymes (58). A number of genes chosen for thisstudy were previously found to be differentiallymethylated in preliminary (PTK7, BCL3, DNMT3b,and PTPRO) [25] and final (CRIP1, STAT5A,FGFR1, S100A2, ITGA2, EGR4, EID1, and IGSF4C)analyses of DNA methylation in asthmatic airwayepithelial cells [26].Additional genes were added to the list given their re-producible association in asthma GWAS and observa-tion that the SNP’s in these genes act as eQTLs.Data and statistical analysesFinal Nanostring results were filtered to keep only genesthat had an average count of at least 30. NormalizedmRNA expression values were compared between asth-matic and non-asthmatic subjects using a linear modelwith a negative binomial distribution controlling for age,sex, and inhaled corticosteroid use. Differential gene ex-pression data are presented as volcano plots as well as in asummary table showing the top differentially regulatedgenes. The level of expression of each transcript is notcompletely independent since there was strong co-expression between the 344 genes. To account for this, weemployed the Matrix Spectral Decomposition analysis ofNyholt and Li et al. [27, 28] to identify the effective num-ber of independent genes. This lead to the multiple com-parison correction shown in Table 2 that is based on aneffective n of 31. Adjusted p-values will be listed as p.adjand genes with a nominal unadjusted p < 0.05 will belisted as p.unadj.a bFig. 2 Comparison of non-asthmatic airway (a) and asthmatic airway (b). Airways are stained using hematoxylin and eosin (H&E) to highlight remodelingchanges within the airways. Scale bar is 200 μmPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 4 of 14In addition to the analysis of differential gene expres-sion, we performed an analysis of differential co-expression in our data set using the analysis packageCoXpress for R. Differential co-expression analysis iden-tifies pairs of genes that are differentially co-expressedi.e. have opposite correlation patterns in cases vs. con-trols or show correlations in one condition only. Genesthat were differentially co-expressed were entered intoWebGestalt for pathway enrichment analysis and arepresented in table form. Network analyst was used tounderstand the protein-protein interaction network[PMID: 25,950,236} of all nominally significant genes(p.unadj < 0.05). Data are plotted using GraphPad ver-sion 5.04 (La Jolla California USA).ResultsAirway characteristicsThe total number of airways analyzed was 52 in asth-matic and 53 in non-asthmatic subjects; with an averageof 4.3 airways per subject (p > 0.05, asthmatic vs. non-asthmatic). The average internal perimeter (Pi) of asth-matic subjects was 5.1 ± 1.5 mm (geometric mean 4.0 ±1.7 mm) and in non-asthmatic subjects was 5.3 ±1.5 mm (geometric mean 4.9 ± 1.3 mm) (p > 0.05 forboth arithmetic and geometric mean). Airway wall areaper unit length of Pi was significantly greater in asth-matics (0.22 ± 0.024 mm2/mm) than in non-asthmatics(0.13 ± 0.019 mm2/mm, p < 0.01). There was also an in-crease in the ASM area per unit Pi (0.018 ± 0.0024 mm2/mm vs. 0.011 ± 0.0015 mm2/mm, p < 0.05) and in base-ment membrane thickness (6.9 ± 0.81 mm vs. 3.9 ±0.73 mm, p < 0.01) in asthmatics versus non-asthmatics.There was no significant difference in the area of epithe-lium or collagen per unit Pi between donor groups (p >0.05). Side by side comparison of asthmatic and non-asthmatic airway can be seen in Fig. 2.Differential gene expression analysisGene expression changes in all genes are summarized inFig. 3 with the candidate gene hypothesis categoriesplotted in Fig. 4. In total there were 51 genes differen-tially expressed based on a threshold p-value of p < 0.05and three genes that were significant after p-value cor-rection (Table 2, p.adj). In brief, there were three geneswhose significance reached the adjusted p-value cutoff,Collagen Type 1 Alpha 1 (COL1A1), COL3A1, and integ-rin beta 6 (ITGB6). The gene for COL1A1 was the mostsignificantly up-regulated gene (1.83-fold increase, p.adj= 0.01) and integrin beta 6 (ITGB6) was the most signifi-cantly down-regulated gene (1.29-fold decrease, p.adj =0.002) (Additional file 1: Table S2). COL1A1 expressionwas positively associated with the amount of collagen inthe airway in asthmatics and non-asthmatics (Collagen/Table 2 Significant differentially expressed genes after p-valuecorrectionGenenameSymbol FoldchangeCounts(Asthma)Counts(Non-Asthma)Adjustedp-value(p.adj)IntegrinBeta 6ITGB6 −1.51 367.4 ± 27.9 475.3 ± 46.4 0.002CollagenType 1Alpha 1COL1A1 1.92 1344.4 ± 222.0 735.8 ± 172.1 0.01CollagenType 3Alpha 1COL3A1 1.84 5324.3 ± 1517.9 3321.1 ± 321.3 0.03-2 -1 0 1 2-5-4-3-2-10Log2 fold changelog10 p-valueFig. 3 Volcano plot summarizing the results of the gene expression analysis. Dotted vertical line indicates fold difference of zero. Fold differences greaterthan zero (positive) indicate increased gene expressed in asthmatics compared to non-asthmatics. Dotted horizontal line indicates significance at nominalp-value of 0.05Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 5 of 14acegbdfhFig. 4 (See legend on next page.)Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 6 of 14Pi, R2 = 0.2221 and 0.2182 for asthma and non-asthmarespectively, p < 0.01) and the thickness of the basementmembrane in both groups combined (R2 = 0.1313, p =0.01). COL3A1 expression was positively associated withASM/Pi (R2 = 0.2089, p = 0.02) and Collagen/Pi (R2 =0.2083, p = 0.03) in asthma. ITGB6 expression was nega-tively associated with basement membrane thickness inasthma (R2 = 0.1801, p = 0.04) and negatively associatedwith Epithelial area/Pi in both groups combined (R2 =0.1158, p = 0.02). There was an association betweenITGB6 expression and Collagen/Pi in asthma that didnot quite reach significance. (R2 = 0.144, p = 0.06). Ineach hypothesis group there were a number of differen-tially expressed genes that did not reach significanceafter p-value adjustment, these included: contractile ap-paratus structure – Smoothelin (1.40 fold increase,p.unadj = 0.01); regulation of contraction – CD38 (1.66fold decrease, p.unadj =0.003); cytoskeletal structure –ITGB6 (see above); cytoskeletal regulation – RAC1 (1.60fold decrease, p.unadj =0.002); epigenetic regulation –PRMT5 (1.52 fold decrease, p.unadj =0.03); epithelialfunction – LAMC2 (1.59 fold decrease, p.unadj =0.006);fibrosis and remodeling – COL1A1 (see above); innateand adaptive immunity – PTGFR (2.18 fold decrease,p.unadj =0.004). The counts, p-values, and adjusted p-values for all significant genes can be seen in Additionalfile 1: Table S2. Of the 12 genes identified in GWAS orlinkage analysis, only ADAM33 (1.56-fold increase,p.unadj =0.0057) was up-regulated. There were no path-ways significantly enriched in the differentially up ordown-regulated genes.Using Network Analyst, we identified a minimumprotein-protein interaction network and key nodes fromour nominally differentially expressed genes (Fig. 5).Green nodes indicate down-regulated genes, red nodesindicate up-regulated nodes (both relative to non-asthmatics), and grey nodes indicate first order interac-tions. This network highlights a number of key nodes inour data set including: Mitogen-Activated Protein Kinase1 (MAPK1, degree = 24), c-FOS (degree = 22, andCalmodulin 3 (CALM3, degree = 21). Using this net-work, we were able to identify pathways significantly as-sociated with up and down-regulated nodes (Table 3).These included key pathways in collagen degradationand remodeling and alterations to cell-cell communication.Differential co-expression analysisWe identified groups of genes that were differentiallyco-expressed between asthmatics and non-asthmatics. Inthis analysis, genes are clustered together based on howtheir expression values correlate with each other. Theanalysis was performed twice with the comparison groupbeing the non-asthmatics or asthmatics in the differentanalyses. Clusters of genes that are significantly co-expressed in one condition and not in the other are saidto be differentially co-expressed. Figure 5 shows an ex-ample; expression of genes in cluster 11 changes fromsubject to subject in asthmatics (left) and non-asthmatics (right). Each line represents one gene in thegroup. Genes in cluster 11 follow a similar pattern of ex-pression in asthmatics (p < 0.001) but not in non-asthmatics (p = 0.32). The rest of these figures can beseen in Additional file 1: Fig. S3 and S4.In non-asthmatics, there were 3 groups of genes thatwere differentially co-expressed compared to asthmatics.These clusters (10, 35, and 53) had an average co-expression correlation coefficient of 0.772 (p < 0.0001) innon-asthmatics and 0.223 (p > 0.05) in asthmatics. Thegenes in these groupings are summarized in Additionalfile 1: Table S3. We performed pathways analysis on eachgroup of genes with Webgestalt. These results are shownin Table 4. There were no pathways that were significantly(See figure on previous page.)Fig. 4 Volcano plots summarizing the results of gene expression for each hypothesis group. Genes involved in (a) Contractile regulation, (b)Structure of the contractile apparatus, (c) Cytoskeletal regulation, (d) Structure of the cytoskeleton, (e) Epigenetic control, (f) Epithelial function, (g)Fibrosis and remodeling, (h) Innate and adaptive immunity. Some genes fall into more than one category and so are plotted in applicablecategories. Dotted vertical line indicates fold differences of zero, dotted horizontal lines indicates significance at nominal p-value of 0.05−4−2024Asthmagene expression723871877233717370167239723772327188721872987236Cluster 11−4−2024Non−Asthmagene expression717972347220721972727184701772807185701872597281Cluster 11Fig. 5 Example of co-expression plots. Each line represents one gene in the cluster. Subjects are indicated along the x-axis, log expression valueson the y-axis. P-values are for the significance of the co-expression in each groupPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 7 of 14enriched for genes differentially co-expressed in non-asthmatics. Further analysis identified specific gene pairsfrom cluster 10 that were both significantly positively cor-related in non-asthmatics and significantly negatively cor-related in asthmatics (Table 5). These included: chitinase3-like 1(CHI3L1) and GSDMB (R = 0.760 and R = −0.707);CHI3L1 and histone deacetylase 10 (HDAC10) (R = 0.731and R = −0.676); HDAC10 and thymocyte antigen 1(THY1 or CD90) (R = 0.727 and R = −0.596); and indolea-mine 2,3-dioxygenase 1 (IDO1) and nuclear factor of acti-vated T-cells, cytoplasmic 2 (NFATC2) (R = 0.659 and R =−0.604), in non-asthmatics and asthmatics respectively.All showed significant but opposite direction of correl-ation (p < 0.05 Fig. 6).In asthmatic samples there were 6 clusters of genesthat were found to be differentially co-expressed. Theseclusters had an average correlation coefficient of 0.728(p < 0.0001) in asthmatics and 0.169 (p > 0.05) in non-asthmatics. The genes in these 6 clusters are summa-rized in Additional file 1: Table S3. Each of the clusterswas also analyzed with Webgestalt (Table 4). In brief,asthmatic co-expressed genes were significantly enrichedin pathways for cytoplasmic virus pattern recognitionsignaling (p = 3.0 × 10−4), positive, and negative regula-tion of type 1 interferon production (p = 3.9 × 10−3, andp = 2.0 × 10−3 respectively). Within the cluster of genesthat were differentially co-expressed, there were 4 pairsTable 3 Pathways significantly associated with protein-proteininteraction nodesPathway name Hits/Total FDRUp-Regulated PathwaysDegradation of Collagen 4/61 0.003Extracellular Matrix Organization 5/157 0.003Degradation of Extracellular Matrix 4/77 0.004Assembly of Collagen Fibrills andOther Multimeric Structures3/54 0.04Collagen Biosynthesis andModifying Enzymes3/62 0.04Down-regulated pathwaysIntegrin Cell Surface Interactions 6/85 0.0007Signal Transduction 17/1690 0.004Cell-Cell Communication 6/143 0.004Platelet Activation, Signaling,and Aggregation7/220 0.004TGF-beta Receptor Signaling in EMT 3/17 0.009Hemostatis 9/511 0.01Signaling by TGF-beta Receptor Complex 4/70 0.02Sema4D Induced Cell Migration andGrowth-Cone Collapse3/29 0.03Sema4D in Semaphorin Signaling 3/34 0.04Fc-gamma Receptor Dependent Phagocytosis 4/86 0.04Table 4 Pathways enriched in differentially co-expressed genes.p-value comes from using list of 334 genes as backgroundCoexpressed in Non-asthmaticsCluster##GenesinCluster# GenesFrom Clusterin PathwayBiological Process p-value10 25 3 Regulation of Cell-CellAdhesion Involvedin Gastrulation2.0 × 10−113 Regulation of MulticellularOrganismal Development2.0 × 10−18 Regulation ofCell Adhesion2.0 × 10−153 6 4 Cell Migration 4.0 × 10−14 Locomotion 5.7 × 10−14 Localization of Cell 4.0 × 10−1Coexpressed inAsthmatics16 6 3 Cytoplasmic PatternRecognition ReceptorSignaling Pathway inResponse to Virus3.0 × 10−43 Positive Regulation ofType 1 InterferonProduction3.9 × 10−33 Negative Regulation ofType 1 InterferonProduction2.0 × 10−320 6 4 Activation ofMAPK Activity6.9 × 10−24 Peptidyl-TyrosinePhosphorylation7.9 × 10−23 Eicosanoind BiosyntheticPathway7.9 × 10−211 9 3 Odontogenesis 4.2 × 10−117 7 4 Response to Bacterium 3.9 × 10−1Bolded pathways highlight those that reached statistical significance in the co-expression data setTable 5 Pairs of differentially co-expressed genesGene 1 Gene 2 Asthma R p-value Non-Asthma R p-valueLAMB2 MMP9 0.727 0.007 −0.686 0.013BPIFA1 DDX58 0.814 0.001 −0.646 0.023BMP2 NFATC2 0.707 0.010 −0.660 0.020GNAS NFATC2 0.665 0.018 −0.798 0.002CHI3L1 GSDMB −0.707 0.010 0.760 0.004CHI3L1 HDAC10 −0.676 0.016 0.731 0.007HDAC10 THY1 −0.596 0.041 0.727 0.007IDO1 NFATC2 −0.604 0.037 0.659 0.0198Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 8 of 14acegbdfhFig. 6 (See legend on next page.)Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 9 of 14of genes whose expression were positively and signifi-cantly correlated in asthmatics but negatively and signifi-cantly correlated in non-asthmatics (Table 5). In brief:laminin B2 (LAMB2) and matrix metallopeptidase 9(MMP9) (R = 0.727 and R = −0.686); BPI fold containingfamily A (BPFIA1) and DEAD box polypeptide 58(DDX58 or RIG-1, retinoic acid inducible gene 1 protein)(R = 0.814 and R = −0.646); bone morphogenetic protein2 (BMP2) and NFATC2 (R = 0.707 and R = −0.660); andG-protein alpha stimulating (GNAS) and NFATC2 (R =0.665 and R = −0.798), in asthmatics and non-asthmaticrespectively. All were significantly correlated but in op-posite directions (p < 0.05, Fig. 6).All eight pairs of differentially co-expressed genes (Table5) were entered into the ENCODE ChIP-SEQ significancetool. All but one of the genes (IDO1) from the pairs wasregulated by the transcriptional repressor CCCTC-bindingfactor or CTCF. Since only 47% of the candidate genes havea CTCF binding site this represents a significant enrich-ment (p = 0.0035, Chi-squared test).DiscussionIn this study the Nanostring® platform was used to quan-tify the airway expression of candidate genes hypothe-sized to be important in the pathophysiology of asthma.mRNA was obtained from the airways of asthmatic(mainly fatal asthmatics) and non-asthmatic donor lungsand measures of airway remodeling were made on thesampled airways. Candidate genes were grouped into thefollowing categories: ASM contraction, the cytoskeleton,epithelial barrier function, innate/adaptive immunity, fi-brosis/remodeling, and epigenetics. 51 genes (15%) werenominally differentially expressed (p.unadj <0.05) inasthmatic airway tissue and included many genes im-portant in cell-cell and cell-matrix interactions(COL1A1, COL3A1, ITGB6, LAMC2, RAC1). Of the 51genes differentially expressed based on a nominal p-value, only 3 were significant following multiple com-parison correction (ITGB6, COL1A1, COL3A1).Cell-cell junctions are altered in the airway epitheliumof asthmatics [29] and this may result in greater perme-ability of the epithelial layer and ultimately hypersensi-tivity of the ASM to agonist challenge [30]. The mostsignificant differentially expressed gene in the data setwas ITGB6. ITGB6 rapidly accumulates following injuryto the epithelial layer and is considered to be importantfor normal wound healing [31]. In the mouse loss ofITGB6 causes an increase in the number of B-cells andT-cells around the airways, an increase in IL-4 produc-tion, and airway hyperresponsivenss in naïve mice with-out allergen challenge [31]. Additionally, influenza virushas been shown to interact with ITGB6 to cause epithe-lial cell death and collagen deposition in a TGF-ßdependent manner [32]. Another gene with lower abun-dance in asthmatics that did not meet the adjusted p-value cut off was RAC1. RAC1 has been shown to beimportant in the formation of tight junctions in anEGFR dependent manner [33], specifically by regulatingtight junction protein 1 (or zona occluden 1) [34] whichwas also lower in abundance in the asthmatic samples.Changes in these genes could also affect epithelial-mesenchymal transition (EMT) [35], cytoskeletal stabil-ity, actin filament assembly/disassembly, cell stiffnessand/or cell migration [36]. In addition, a protein-proteininteraction network highlighted that down-regulatedgenes were enriched in pathways for cell-cell communi-cation and integrin cell surface interactions.The most significantly up-regulated gene in asthmaticsamples was COL1A1 which codes for the alpha chain intype 1 collagen. Collagen 1 is the major type of collagenin basement membrane and a major protein found in re-modeled airways. Collagen 1 is important in airway re-modeling [37], in particular thickening of thesubepithelial space which is associated with worseningof asthma symptoms [38]. Collagen type 3 was also sig-nificantly elevated in asthmatic subjects and is also sig-nificantly elevated in the basement membrane ofasthmatic subjects [39]. Type 1 and 3 collagen have bothbeen associated with worsening lung function in a horsemodel of asthma [40]. Beyond the ability of collagens toaffect distensibility of the airways, collagen 1 has beenshown to stimulate ASM to produce MMP1 [41] andproliferate, in conjunction with FAK [42]. Collagen 1and 3 expression has also been shown to be unaffectedby corticosteroid usage in severe asthmatics [43] andcollagen 1 and 3 contribute to the loss of the anti-mitotic effect of corticosteroids [44]. Enrichment ofpathways involved in collagen remodeling was seen inour network analysis and highlights the importance ofunderstanding fibrosis as it relates to fiber production,degradation, and organization and how this impacts nor-mal cell function.ADAM33 has been described in candidate gene studiesto be associated with asthma [6] and was significantlyelevated in asthmatic airways. ADAM33 has been im-plicated in smooth muscle development, cell-cell(See figure on previous page.)Fig. 6 Individual correlation plots for the pairs of differentially co-expressed genes in asthmatics and non-asthmatics. a LAMB2 vs. MMP9, (b) BPIFA1 vs.DDX58, (c) BMP2 vs. NFACT2, (d) GNAS vs. NFACT2, (e) CHI3L1 vs. GSDMB, (F) CHI3L1 vs. HDAC10, (G) HDAC10 vs. THY1, (F) IDO1 vs. NFATC2. Each pointrepresents a sample and the values on the axes are gene counts. Lines of best fit are plotted with the 95% confidence interval (grey shaded area) foreach correlation. Each of the correlations is significant (p < 0.05) and the direction indicates a positive or negative correlationPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 10 of 14connections, and cell proliferation and differentiation[45]. Over-expression of ADAM33 in asthmatic air-ways has been described [46] and may be an import-ant determinant of disease progression. There isevidence that ADAM33 can stimulate angiogenesis exvivo and in vivo and by this mechanisms may con-tribute to airway remodeling [47]. Furthermore,ADAM33 family member TACE/ADAM17 can medi-ate release of TNF-α and fracktalkine (or CX3CL1)from the cell membrane [48, 49] and ADAM9 maymediate the release of growth factor HB-EGF [50]. IfADAM33 has a similar capacity for cytokine orgrowth factor cleavage this could make it a majorcontributor to airway remodeling in asthma.Other genes in the 5 most differentially up or downregulated genes that were nominaly significant include:Cyclic ADP Ribose (CD38), Interleukin 13 ReceptorAlpha 1 (IL13RA1), Prostaglandin F Receptor (PTGFR),Heat Shock Protein Beta 1 (HSPB1), and Interferon In-duced with Helicase C Domain 1 (IFIH1). CD38 is aprotein that generates the second messenger cADPR tocause calcium release. Recent work has explored the roleof CD38 in asthma and has suggested that increasedCD38 expression causes hypercontractility in ASM cellsfrom asthmatics [51] although in our samples we saw noCD38 staining in the ASM layer (Additional file 1: Fig.S1 and S2). Additionally, CD38 deficient mice have re-duced AHR following ovalbumin challenge [52]. De-creased IL13RA1 expression is surprising in the contextof asthma but this could be due to a compensatory re-sponse to continued eosinophilia and IL13/IL4 exposurein the lung [53]. Prostaglandin’s can be both pro and antiinflammatory but there is little research on the role ofprostaglandin F in the context of asthma. HSPB1 (orheat shock protein 27) is a chaperone protein that hasbeen implicated in cellular differentiation, apoptosis, andsmooth muscle contraction [54]. Up-regulation of thegene could contribute to ASM hypercontractility inasthma but further work investigating the phosphoryl-ation state and activity of HSP27 in asthma is needed toanswer this question. IFIH1, also known as MDA5, is aDEAD box double stranded (ds) RNA helicases that candetect intracellular viral dsRNA and lead to the produc-tion of interferons [55]. MDA5 and TLR3 signaling havebeen shown to be deficient in bronchial epithelial cellsfrom asthmatic subjects [56] and this could be respon-sible for the defective epithelial release of interferon Iand III in response to rhino virus infection [57, 58].Co-expression of genes does not imply interaction be-tween their proteins but instead may suggest similaritiesin their regulation by transcription factors or epigeneticmechanisms [59]. Co-expression analyses can revealchanges in the regulation of gene expression [60] andhave been used to identify epigenetic changes that affectgene co-expression in cancer [61]. In our study, genesthat were differentially co-expressed between asthmaticsand non-asthmatics were significantly enriched for path-ways involved in virus recognition and regulation ofinterferon production (Table 4). The genes enriched inthese pathways were from cluster 16 and were RIG-I(DDX58), RIG-1-like receptor 3 (DHX58), and interferoninduced with helicase C domain 1 (IFIH1). This findingsuggests that a central molecular mechanism may regu-late diverse antiviral immune molecules in response toviral infections that may trigger asthma exacerbationsand/or pathogenesis [62]. IFIH1, as discussed earlier, wasalso differentially expressed.One of the most intriguing results was the identifica-tion of a single transcriptional repressor, CTCF, thatcontrols the expression of all but one of the differentiallyco-expressed pairs of genes. CTCF influences gene ex-pression through chromatin modifications [63] resultingin insulation of the target regions [64]. CTCF is anarchitectural protein that mediates inter- and intra-chromosomal interactions at distant genomic sites, andregulates three-dimensional genome architecture [63].There are examples of CTCF silencing one gene whileactivating another [63]. Specific to asthma, differentialexpression at the ZPBP2/GSDMB/ORMDL3 locus wasidentified resulting from allele-specific chromatin re-modeling mediated by CTCF [65]. A SNP in ZPBP2 cre-ated a CTCF binding site resulting in increasedexpression of ZPBP2 but diminished expression ofGSDMB and ORMDL3 [65]. Additionally, CTCF ishighly sensitive to DNA methylation at CTCF bindingsites [63]; changes to the methylome can have direct ef-fects on the regulation of CTCF target genes. CTCFcould play a crucial role in controlling the many geneexpression changes observed in the airways of asth-matics and is worthy of more intense research that is be-yond the scope of this paper.There are several limitations of this study. Firstly, theuse of whole airway RNA rather than RNA from specificcell types precludes us from conclusively identifying thesite of gene expression. Secondly, the majority of theasthmatic patients were fatal asthmatics and experiencedhypoxia and treatment with steroids during their fatal at-tack which can affect mRNA expression in tissues takenfor research purposes. We addressed this by controllingfor steroid use in our analysis of differential gene expres-sion. Additionally, the use of more severe asthmaticsmay mean that these results are not generalizable toasthmatics as a whole. But considering severe asthmaticpopulations have the most hospital visits and are most atrisk for exacerbations, we believe our results provide sig-nificant insight into the genes that are altered in fataldisease. Finally, the relatively small sample size limitsour ability to detect differences in gene expression lessPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 11 of 14than ~1.5 fold on average, although this also means thatthe changes we see are likely to be real. Procurement ofdonor lungs is time consuming and costly so increasingthe number of patients for this study was not feasible,however future studies in asthmatic biopsies or cell cul-ture experiments could confirm these results with theability for much larger sample size. A limited number ofdonors with non-fatal asthma (n = 4) means we were un-able to test for differences between these two groups ofdonors (fatal vs. non-fatal).ConclusionThis study identifies changes in the expression and co-expression of genes thought to be important in asthmaand AHR. Specifically, we identified altered abundanceof genes involved in cell-cell and cell-matrix connectionsas well as those involved in the immune response andcell homeostasis. We also identified changes in the co-expression of genes involved in virus recognition andinterferon production. The transcription factor CTCFcould be an important regulator of the asthmatic pheno-type and warrants further investigation. Future workshould focus on elucidating the potential mechanismsbehind altered CTCF binding as it relates to asthmapathophysiology.Additional fileAdditional file 1: Supplementary Methods – Methods describing selectionof house keeping genes and immunohistochemical staining procedure.Supplementary Tables – Tables containing clinical demographics forsubjects, average counts, fold change, and p-value for all genes studied, andall differentially co-expressed genes. Supplementary Figures and Legends –Figures showing sample immunohistochemical staining for proteins ofsignificantly altered genes, co-expression plots. (DOCX 35 kb)Abbreviations%CV: Percent coefficient of variation; ADAM17: ADAM metallopeptidasedomain 17; ADAM33: ADAM metallopeptidase domain 33; ADAM9: ADAMmetallopeptidase domain 9; AHR: Airway hyperresponsiveness; ASM: Airwaysmooth muscle; AURKA: Aurora kinase A; BCL3: B-Cell CLL/lymphoma 3;BMP2: Bone morphogenetic protein 2; BPIFA1: BPI fold containing family Amember 1; cADPR: Cyclic ADP Ribose; CD38: Cluster of differentiation 38;CD44: Cluster of differentiation 44; CHI3L1: Chitinase 3-like; COL1A1: CollagenType 1 Alpha 1; COL3A1: Collagen Type 3 Alpha 1; CREBBP: CREB bindingprotein; CRIP1: Cysteine rich protein 1; CTCF: CCCTC-binding factor;CX3CL1: Fracktalkine; DDX58: DExD/H-box helicase 58; DNA: Deoxyribonuleicacid; DNMT3b: DNA methyltransferase 3 Beta; ECM: Extracellular matrix;EGFR: Epidermal growth factor receptor; EGR4: Early growth response 4;EID1: EP300 interacting inhibitor of differentiation 1; EMT: Epithelial-mesenchymal transition; EP300: E1A binding protein P300; eQTL: Expressionquantitative trait loci; FAK: Focal adhesion kinase; FGFR1: Fibroblast growthfactor receptor 1; GNAS: Adenylate cyclase-stimulating G alpha protein;GNB2L1: Guanine nucleotide-binding protein subunit Beta-2-Like 1;GSDMB: Gasdermin B; GUSB: β-Glucuronidase; GWAS: Genome wideassociation study; H&E: Hemoatoxylin and eosin; HB-EGF: Heparin-bindingEGF-like growth factor; HDAC10: Histone deacetylase 10; HSPB1: Heat shockprotein 27; IDO1: Indoleamine 2,3-dioxygenase 1; IFIH1: Interferon inducedwith helicase C domain 1; IGSF4C: Immunoglobulin superfamily member 4C;IIAM: International Institute of the Advancement of Medicine; IL-13: Interleukin-13; IL33: Interleukin-33; IL4: Interleukin-4; ITGA2: Integrinsubunit Alpha 2; ITGB6: Integrin subunit Beta 6; KAT2A: Lysineacetyltransferase 2A; KEGG: Kyoto encyclopedia of genes and genomes;LAMB2: Laminin subunit Beta 2; LAMC2: Laminin subunit Gamma 2;MDA5: Melanoma differentiation-associated protein 5; MLCK: Myosin lightchain kinase; MMP1: Matrix metallopeptidase 1; MMP9: Matrixmetallopeptidase 9; MYH11: Myosin heavy chain 11; NFATC2: Nuclear factorof activated T-cells, cytoplasmic 2; NLR: Nod-like receptor; ORMDL3: ORMDLsphingolipid biosynthesis regulator 3; Pi: Internal perimeter of the airway;POLR2A: RNA polymerase R2A; PRMT5: Protein arginine methyltransferase 5;PTGFR: Prostaglandin F receptor; PTK7: Protein tyrosine kinase 7;PTPRO: Protein tyrosine phosphatase, receptor type O; RAC1: Ras-related C3botulinum toxin substrate 1; RIG-1: Retinoic acid inducible gene 1; RLR: Rig-like receptor; RNA: Ribonucleic acid; RPL19: Ribosomal protein L19;S100A2: S100 calcium binding protein A2; SEM: Standard error of the mean;SM-22: Smooth muscle protein 22-alpha/transgelin; SNP: Single nucleotidepolymorphism; ST-2: Suppression of tumorigenicity 2; STAT5A: Signaltransducer and activator of transcription 5A; SUV39H1: Suppressor ofvariegation 3–9 homolog 1; TACE: Tumor necrosis factor-α-converting en-zyme; TBP: TATA box binding protein; TGF-β: Transforming growth factorBeta 1; THY1: Thymocyte antigen 1; TLR: Toll-like receptors; TNF- α: Tumornecoris factor alpha; TSLP: Thymic stromal lymphopoietin; ZO-1: Zonaoccludens-1; ZPBP2: Zona pellucida binding protein 2AcknowledgementsLu Wang PhD for her assistance in collection of histological images.FundingResearch was funded by the Canadian Institute for Health Research (CIHR)and no role in the design, collection, analysis, or interpretation of data andresults, and did not contribute to the writing of the manuscript.Availability of data and materialsComplete gene expression data is available in Additional file 1: Table S2 inthe online supplement. Data is also available from corresponding authorupon request.Authors’ contributionsCDP, PDP, and CYS contributed to the experimental design and setup. CDP,PDP, CYS, SW, DS, SW, JAH, SJY, DRD, CC, and TLH all contributed to thedesign of the gene expression panel. CDP, MO, YN, PDP, and CYScontributed to data analysis. TLH collected and inflated lung samples. CDPand BAA collected data for remodeling and protein expression. CDP, PDP,and MO drafted manuscript. CDP, PDP, MO, CC, TLH, JAH, DRD, SW, DS, andSW edited and revised manuscript. All authors have read and approved thefinal version of this manuscript.Ethics approval and consent to participateHuman lungs were donated with consent from the IIAM and used withapproval from the University of British Columbia and St. Paul’s Hospital ethicscommittee.Consent for publicationNot applicable.Competing interestsThe authors declare that they have no competing interests.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1UBC Institute for Heart Lung Health, St. Paul’s Hospital, 1081 Burrard St,Vancouver, BC, Canada. 2UBC Department of Medicine, Division ofRespirology, University of British Columbia, Vancouver, BC, Canada. 3UBCChan-Yeung Centre for Occupational and Environmental Respiratory Disease,Gordon & Leslie Diamond Health Care Centre, Vancouver General Hospital,2775 Laurel Street, 7th floor, Vancouver, BC, Canada. 4University of BritishColumbia Centre for Heart Lung Innovation, St. Paul’s Hospital, 1081 BurrardSt, Vancouver, BC, Canada. 5UBC School of Population and Public Health,University of British Columbia, Vancouver, BC, Canada. 6UBC Department ofPascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 12 of 14Anesthesiology, Pharmacology and Therapeutics, University of BritishColumbia, Vancouver, BC, Canada. 7UBC Department of Pathology andLaboratory Medicine, University of British Columbia, Vancouver, BC, Canada.8Division of Respirology, Department of Medicine, McMaster University,Hamilton, ON, Canada. 9Children’s Hospital Research Institute of Manitoba,513-715 McDermot Avenue, Winnipeg, MB R3E 3P4, Canada.Received: 2 June 2017 Accepted: 30 November 2017References1. James AL, Elliot JG, Jones RL, Carroll ML, Mauad T, Bai TR, et al. Airwaysmooth muscle hypertrophy and hyperplasia in asthma. Am J Respir CritCare Med. 2012;185:1058–64.2. Barnes PJ. Inhaled Glucocorticoids for asthma. N Engl J Med. 1995;332:868–75.3. Holgate ST, Holloway J, Wilson S, Bucchieri F, Puddicombe S, Davies DE.Epithelial–Mesenchymal communication in the pathogenesis of chronicasthma. Proc Am Thorac Soc. 2004;1:93–8.4. Tulah AS, Holloway JW, Sayers I. Defining the contribution of SNPsidentified in asthma GWAS to clinical variables in asthmatic children.BMC Med Genet. 2013;14:100.5. Torgerson DG, Ampleford EJ, Chiu GY, Gauderman WJ, Gignoux CR, GravesPE, et al. Meta-analysis of genome-wide association studies of asthma inethnically diverse north American populations. Nat Genet. 2011;43:887–92.6. van Eerdewegh P, Little RD, Dupuis J, del Mastro RG, Falls K, Simon J,et al. Association of the ADAM33 gene with asthma and bronchialhyperresponsiveness. Nature. 2002;418:426.7. Hao K, Bosse Y, Nickle DC, Pare PD, Postma DS, Laviolette M, et al.Lung eQTLs to help reveal the molecular underpinnings of asthma.PLoS Genet. 2012;8:e1003029.8. Veldman-Jones MH, Brant R, Rooney C, Geh C, Emery H, Harbron CG, etal. Evaluating robustness and sensitivity of the NanoString technologiesnCounter platform to enable multiplexed gene expression analysis ofclinical samples. Cancer Res. 2015;75:2587–93.9. Chin LY, Bosse Y, Pascoe CD, Hackett TL, Seow CY, Pare PD. Mechanicalproperties of asthmatic airway smooth muscle. Eur Respir J. 2012;40:45–54.10. Hackett T-L, Warner SM, Stefanowicz D, Shaheen F, Pechkovsky DV, MurrayLA, et al. Induction of epithelial-mesenchymal transition in primary airwayepithelial cells from patients with asthma by transforming growth factor-beta1. Am J Respir Crit Care Med. 2009;180:122–33.11. Hirota JA, Hirota SA, Warner SM, Stefanowicz D, Shaheen F, Beck PL, et al.The airway epithelium nucleotide-binding domain and leucine-rich repeatprotein 3 inflammasome is activated by urban particulate matter. J AllergyClin Immunol. 2012;129:1116–25. e612. Woodruff PG, Dolganov GM, Ferrando RE, Donnelly S, Hays SR, SolbergOD, et al. Hyperplasia of smooth muscle in mild to moderate asthmawithout changes in cell size or gene expression. Am J Respir Crit CareMed. 2004;169:1001–6.13. Sieck GC, White TA, Thompson MA, Pabelick CM, Wylam ME, PrakashYS. Regulation of store-operated Ca2+ entry by CD38 in humanairway smooth muscle. Am J Physiol - Lung Cell Mol Physiol. 2008;294:L378–85.14. Gunst SJ, Zhang W. Actin cytoskeletal dynamics in smooth muscle: a newparadigm for the regulation of smooth muscle contraction. Am J Physiol -Cell Physiol. 2008;295:C576–87.15. Lan B, Wang L, Zhang J, Pascoe CD, Norris BA, Liu JC-Y, et al. Rho-kinasemediated cytoskeletal stiffness in skinned smooth muscle. J Appl PhysiolBethesda Md. 1985;2013 https://doi.org/10.1152/japplphysiol.00654.2013.16. Allahverdian S, Harada N, Singhera GK, Knight DA, Dorscheid DR. Secretionof IL-13 by airway epithelial cells enhances epithelial repair via HB-EGF. AmJ Respir Cell Mol Biol. 2008;38:153–60.17. Dorscheid DR, Wojcik KR, Yule K, White SR. Role of cell surface glycosylationin mediating repair of human airway epithelial cell monolayers. Am JPhysiol Lung Cell Mol Physiol. 2001;281:L982–92.18. Moffatt MF, Gut IG, Demenais F, Strachan DP, Bouzigon E, Heath S, et al. Alarge-scale, consortium-based genomewide association study of asthma. NEngl J Med. 2010;363:1211–21.19. Xu J, Singhera GK, Dorscheid DR. Expression of surfactant protein D inairways of asthmatics and interleukin-13 modulation of surfactant protein Din human models of airway epithelium. Respir Res. 2015;16:26.20. Hirota JA, Gold MJ, Hiebert PR, Parkinson LG, Wee T, Smith D, et al. Thenucleotide-binding domain, leucine-rich repeat protein 3 inflammasome/IL-1 receptor I axis mediates innate, but not adaptive, immune responses afterexposure to particulate matter under 10 μm. Am J Respir Cell Mol Biol.2015;52:96–105.21. Wang D-Z, Olson EN. Control of smooth muscle development by themyocardin family of transcriptional coactivators. Curr Opin Genet Dev. 2004;14:558.22. Tsao P-N, Vasconcelos M, Izvolsky KI, Qian J, Lu J, Cardoso WV. Notchsignaling controls the balance of ciliated and secretory cell fates indeveloping airways. Dev Camb Engl. 2009;136:2297–307.23. Kumagai K, Ohno I, Okada S, Ohkawara Y, Suzuki K, Shinya T, et al. Inhibitionof matrix metalloproteinases prevents allergen-induced airway inflammationin a murine model of asthma. J Immunol Baltim Md 1950. 1999;162:4212–9.24. Stefanowicz D, Lee JY, Lee K, Shaheen F, Koo H-K, Booth S, et al. ElevatedH3K18 acetylation in airway epithelial cells of asthmatic subjects. Respir Res.2015;16:95.25. Stefanowicz D. Profiling the epigenetic landscape of the airwayepithelium in asthma. 2014. https://circle.ubc.ca/handle/2429/47071.Accessed 2 Oct 2015.26. Stefanowicz D, Hackett T-L, Garmaroudi FS, Günther OP, Neumann S,Sutanto EN, et al. DNA methylation profiles of airway epithelial cells andPBMCs from healthy, atopic and asthmatic children. PLoS One. 2012;7:e44213.27. Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other.Am J Hum Genet. 2004;74:765–9.28. Li J, Ji L. Adjusting multiple testing in multilocus analyses using theeigenvalues of a correlation matrix. Heredity. 2005;95:221–7.29. de Boer WI, Sharma HS, Baelemans SMI, Hoogsteden HC, LambrechtBN, Braunstahl GJ. Altered expression of epithelial junctional proteinsin atopic asthma: possible role in inflammation. Can J PhysiolPharmacol. 2008;86:105–12.30. Barnes PJ, Cuss FM, Palmer JB. The effect of airway epithelium onsmooth muscle contractility in bovine trachea. Br J Pharmacol. 1985;86:685–91.31. Huang XZ, Wu JF, Cass D, Erle DJ, Corry D, Young SG, et al. Inactivation ofthe integrin beta 6 subunit gene reveals a role of epithelial integrins inregulating inflammation in the lung and skin. J Cell Biol. 1996;133:921–8.32. Jolly L, Stavrou A, Vanderstoken G, Meliopoulos VA, Habgood A, Tatler AL,et al. Influenza promotes collagen deposition via αvβ6 Integrin-mediatedtransforming growth factor β activation. J Biol Chem. 2014;289:35246–63.33. Terakado M, Gon Y, Sekiyama A, Takeshita I, Kozu Y, Matsumoto K,et al. The Rac1/JNK pathway is critical for EGFR-dependent barrierformation in human airway epithelial cells. Am J Physiol Lung Cell MolPhysiol. 2011;300:L56–63.34. Jou T-S, Schneeberger EE, Nelson WJ. Structural and functionalregulation of tight junctions by RhoA and Rac1 small GTPases. J CellBiol. 1998;142:101–15.35. Hackett T-L. Epithelial-mesenchymal transition in the pathophysiologyof airway remodelling in asthma. Curr Opin Allergy Clin Immunol.2012;12:53–9.36. Parsons JT, Horwitz AR, Schwartz MA. Cell adhesion: integrating cytoskeletaldynamics and cellular tension. Nat Rev Mol Cell Biol. 2010;11:633–43.37. Liesker JJ, Hacken NHT, Zeinstra-Smith M, Rutgers SR, Postma DS,Timens W. Reticular basement membrane in asthma and COPD: similarthickness, yet different composition. Int J Chron Obstruct Pulmon Dis.2009;4:127–35.38. Chetta A, Foresi A, Del Donno M, Bertorelli G, Pesci A, Olivieri D. Airwaysremodeling is a distinctive feature of asthma and is related to severity ofdisease. Chest. 1997;111:852–7.39. Wilson JW, Li X. The measurement of reticular basement membraneand submucosal collagen in the asthmatic airway. Clin Exp Allergy.1997;27:363–71.40. Setlakwe EL, Lemos KR, Lavoie-Lamoureux A, Duguay J-D, Lavoie J-P. Airwaycollagen and elastic fiber content correlates with lung function in equineheaves. Am J Physiol - Lung Cell Mol Physiol. 2014;307:L252–60.41. Rogers NK, Clements D, Dongre A, Harrison TW, Shaw D, Johnson SR. Extra-cellular matrix proteins induce matrix Metalloproteinase-1 (MMP-1) activityand increase airway smooth muscle contraction in asthma. PLoS One. 2014;9:e90565.Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 13 of 1442. Dekkers BGJ, Spanjer AIR, van der Schuyt RD, Kuik WJ, Zaagsma J, Meurs H.Focal adhesion Kinase regulates collagen I–induced airway smooth musclephenotype switching. J Pharmacol Exp Ther. 2013;346:86–95.43. Chakir J, Shannon J, Molet S, Fukakusa M, Elias J, Laviolette M, et al. Airwayremodeling-associated mediators in moderate to severe asthma: effect ofsteroids on TGF-β, IL-11, IL-17, and type I and type III collagen expression. JAllergy Clin Immunol. 2003;111:1293–8.44. Bonacci JV, Schuliga M, Harris T, Stewart AG. Collagen impairsglucocorticoid actions in airway smooth muscle through integrin signalling.Br J Pharmacol. 2006;149:365–73.45. Shapiro SD, Owen CA. ADAM-33 surfaces as an asthma gene. N Engl J Med.2002;347:936–8.46. Foley SC, Mogas AK, Olivenstein R, Fiset PO, Chakir J, Bourbeau J, et al.Increased expression of ADAM33 and ADAM8 with disease progression inasthma. J Allergy Clin Immunol. 2007;119:863–71.47. Puxeddu I, Pang YY, Harvey A, Haitchi HM, Nicholas B, Yoshisue H, et al. Thesoluble form of a disintegrin and metalloprotease 33 promotesangiogenesis: implications for airway remodeling in asthma. J Allergy ClinImmunol. 2008;121:1400–6. 1406.e1–448. Black RA. Tumor necrosis factor-α converting enzyme. Int J Biochem CellBiol. 2002;34:1–5.49. Garton KJ, Gough PJ, Blobel CP, Murphy G, Greaves DR, Dempsey PJ, et al.Tumor necrosis factor-α-converting enzyme (ADAM17) mediates thecleavage and shedding of Fractalkine (CX3CL1). J Biol Chem. 2001;276:37993–8001.50. Izumi Y, Hirata M, Hasuwa H, Iwamoto R, Umata T, Miyado K, et al. Ametalloprotease-disintegrin, MDC9/meltrin-gamma/ADAM9 and PKCdeltaare involved in TPA-induced ectodomain shedding of membrane-anchoredheparin-binding EGF-like growth factor. EMBO J. 1998;17:7260–72.51. Jude JA, Dileepan M, Panettieri RA, Walseth TF, Kannan MS. Altered CD38/cyclic ADP-ribose Signaling contributes to the asthmatic phenotype. JAllergy. 2012;2012:289468.52. Deshpande DA, White TA, Guedes AGP, Milla C, Walseth TF, Lund FE, et al.Altered airway responsiveness in CD38-deficient mice. Am J Respir Cell MolBiol. 2005;32:149–56.53. Myrtek D, Knoll M, Matthiesen T, Krause S, Lohrmann J, Schillinger D, et al.Expression of interleukin-13 receptor alpha 1-subunit on peripheral bloodeosinophils is regulated by cytokines. Immunology. 2004;112:597–604.54. Salinthone S, Tyagi M, Gerthoffer WT. Small heat shock proteins in smoothmuscle. Pharmacol Ther. 2008;119:44–54.55. Reikine S, Nguyen JB, Modis Y. Pattern recognition and Signalingmechanisms of RIG-I and MDA5. Front Immunol. 2014;5 https://doi.org/10.3389/fimmu.2014.00342.56. Parsons KS, Hsu AC, Wark PAB. TLR3 and MDA5 signalling, although notexpression, is impaired in asthmatic epithelial cells in response to rhinovirusinfection. Clin Exp Allergy J Br Soc Allergy Clin Immunol. 2014;44:91–101.57. Wark PAB, Johnston SL, Bucchieri F, Powell R, Puddicombe S, Laza-Stanca V,et al. Asthmatic bronchial epithelial cells have a deficient innate immuneresponse to infection with rhinovirus. J Exp Med. 2005;201:937–47.58. Contoli M, Message SD, Laza-Stanca V, Edwards MR, Wark PAB, Bartlett NW,et al. Role of deficient type III interferon-lambda production in asthmaexacerbations. Nat Med. 2006;12:1023–6.59. Hsu C-L, Juan H-F, Huang H-C. Functional analysis and characterization ofdifferential Coexpression networks. Sci Rep. 2015;5:13295.60. de la Fuente A. From “differential expression” to “differential networking” –identification of dysfunctional regulatory networks in diseases. TrendsGenet. 2010;26:326–33.61. Ii M, Yamamoto H, Taniguchi H, Adachi Y, Nakazawa M, Ohashi H, et al. Co-expression of laminin β3 and γ2 chains and epigenetic inactivation oflaminin α3 chain in gastric cancer. Int J Oncol. 2011;39:593–9.62. Porritt RA, Hertzog PJ. Dynamic control of type I IFN signalling by anintegrated network of negative regulators. Trends Immunol. 2015;36:150–60.63. Ong C-T, Corces VG. CTCF: an architectural protein bridging genometopology and function. Nat Rev Genet. 2014;15:234–46.64. Mishiro T, Ishihara K, Hino S, Tsutsumi S, Aburatani H, Shirahige K, et al.Architectural roles of multiple chromatin insulators at the humanapolipoprotein gene cluster. EMBO J. 2009;28:1234–45.65. Verlaan DJ, Berlivet S, Hunninghake GM, Madore A-M, Larivière M,Moussette S, et al. Allele-specific chromatin Remodeling in the ZPBP2/GSDMB/ORMDL3 locus associated with the risk of asthma and autoimmunedisease. Am J Hum Genet. 2009;85:377–93.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Pascoe et al. BMC Pulmonary Medicine  (2017) 17:189 Page 14 of 14


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items