Changes in transcriptome of native nasal epithelium expressing F508del-CFTR and intersecting data from comparable studies
© Clarke et al.; licensee BioMed Central Ltd. 2013
Received: 6 September 2012
Accepted: 7 March 2013
Published: 28 March 2013
Microarray studies related to cystic fibrosis (CF) airway gene expression have gone some way in clarifying the complex molecular background of CF lung diseases, but have made little progress in defining a robust “molecular signature” associated with mutant CFTR expression. Disparate methodological and statistical analyses complicate comparisons between independent studies of the CF transcriptome, and although each study may be valid in isolation, the conclusions reached differ widely.
We carried out a small-scale whole genome microarray study of gene expression in human native nasal epithelial cells from F508del-CFTR homozygotes in comparison to non-CF controls. We performed superficial comparisons with other microarray datasets in an attempt to identify a subset of regulated genes that could act as a signature of F508del-CFTR expression in native airway tissue samples.
Among the alterations detected in CF, up-regulation of genes involved in cell proliferation, and down-regulation of cilia genes were the most notable. Other changes involved gene expression changes in calcium and membrane pathways, inflammation, defence response, wound healing and the involvement of estrogen signalling. Comparison of our data set with previously published studies allowed us to assess the consistency of independent microarray data sets, and shed light on the limitations of such snapshot studies in measuring a system as subtle and dynamic as the transcriptome. Comparison of in-vivo studies nevertheless yielded a small molecular CF signature worthy of future investigation.
Despite the variability among the independent studies, the current CF transcriptome meta-analysis identified subsets of differentially expressed genes in native airway tissues which provide both interesting clues to CF pathogenesis and a possible CF biomarker.
Cystic Fibrosis (CF) is a clinically complex disease  caused primarily by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene , which encodes a chloride (Cl-) channel that plays a fundamental role in ion and fluid transport across epithelial surfaces . The CF phenotype depends greatly on what combination of mutant CFTR alleles is present out of the more than 1,900 currently listed (http://www.genet.sickkids.on.ca/StatisticsPage.html) . F508del-CFTR , which accounts for up to 90% of CF alleles , is associated with a severe clinical phenotype, but even F508del-homozygous CF patients display much phenotypic heterogeneity . Although such heterogeneity can partly be explained by genetic modifiers [8–12] or environmental factors [13, 14], it is desirable to determine how F508del-CFTR specifically affects global gene expression, in order to clarify how a dynamic network of interactions surrounding CFTR at the cellular level  is perturbed in the most widespread form of CF.
Several CF transcriptomics studies have employed microarrays to measure differences in global gene expression caused by the F508del mutation in isogenic bronchial cells  (in this case the CFTR genotype was F508del/W1282X), primary cultures of tracheal and bronchial cells , native nasal epithelial and bronchial cells [18, 19] and immortalized foetal tracheal cell lines . Two of these studies [16, 20] used technical replicates of the same source material, thus avoiding the problem of individual variation present in studies using biological replicates, but also reducing their interest as general models of F508del-CFTR related gene expression. Studies on native tissues have reported differential expression of genes involved in a variety of cellular processes relevant to CF, such as airway defence and mitochondrial function , or inflammation and cellular movement . In contrast, similar work in primary cultures of epithelial cells from CF patients, led to the conclusion that F508del-CFTR had a minimal effect on global gene expression , suggesting that the differences found in native cells were secondary. Studies focusing on expression differences associated with severity of CF phenotype  and others on expression patterns of nasal vs. bronchial epithelium  also produced widely differing patterns of global gene expression. A recent meta-analysis of four independent microarray studies  concluded that very few individual genes were among the highest regulated in more than two of the four studies, and that there was little evidence associating induction of pro-inflammatory pathways with the presence of F508del-CFTR.
Herein, we present the results of a small-scale microarray study of differential gene expression in human native nasal epithelial cells from five F508del-homozygous CF patients vs. five control individuals. Data analysis using the Rank Products (RP) method resulted in a list of differentially expressed genes, many of which are functionally relevant, given our knowledge of CF. Some others, although not previously connected to CF, fall into enriched gene ontology (GO) groups relevant to CF, including cell proliferation, calcium binding, plasma membrane and cilium. A comparison of our data with gene lists obtained in five similar studies led us to conclude that the genes shared between independent gene lists did not constitute a robust molecular signature, although many of the genes shared were highly relevant to CF. However, reanalysis of the results of a recent study of CF related gene expression in bronchial and nasal epithelium  followed by direct comparison with our own dataset led to the identification of a small subset of regulated genes as a putative gene signature characteristic of the CF airway.
Materials and methods
Participant selection and nasal respiratory epithelial cell collection
Identity of nasal epithelial cell samples used in microarray analysis
% inflamm. cells
FVC (% of expected)
FEV1 (% of expected)
CF (F508del homozygous)
1.70 L (88.7%)
1.39 L (84.6%)
4.16 L (90.6%)
3.36 L (88.5%)
2.65 L (60.7%)
1.43 L (39.7%)
0.60 L (38.3%)
0.60 L (43.8%)
3.00 L (78.5%)
2.28 L (70.7%)
A cytospin centrifuge (Shandon, Thermo Scientific, USA) was used to prepare formaldehyde-fixed respiratory epithelial cell samples for standard May-Grünwald-Giemsa (MGG) staining, as previously described . Identity of samples was obscured with randomly numbered labels for blind counting. Slides were then evaluated twice using light microscopy and digital photography of 5 high power fields of view per sample, and cells were categorized as epithelial or inflammatory. Results were expressed as the percentage of total cells (Table 1). Samples with more than 10% inflammatory cells were excluded from the study. The data obtained agreed with results from previous studies, in which approximately 5-10% of cells from nasal brushing have been shown to be of inflammatory origin regardless of CF status [18, 24].
RNA isolation, target synthesis and hybridization to AffymetrixGeneChips
Total RNA was extracted using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Concentration and purity was determined by spectrophotometry (Nanodrop) and integrity (RIN > 7.0: mean RIN, CF - 8.2; Control - 8.4) was confirmed using an Agilent 2100 Bioanalyzer with a RNA 6000 Nano Assay (Agilent Technologies, Palo Alto, CA). Five CF and five control RNA samples were chosen for the final microarray hybridization. RNA was processed for use on Affymetrix (Santa Clara, CA, USA) GeneChip HsAirwaya520108F Arrays, which were custom-designed to determine gene expression in the human airway epithelium , according to the manufacturer’s Two-Cycle Target Labelling Assay. Briefly, 90 ng of total RNA containing spiked in Poly-A RNA controls (GeneChip Expression GeneChip Eukaryotic Poly-A RNA Control Kit; Affymetrix) were used in a reverse transcription reaction (Two-Cycle DNA synthesis kit; Affymetrix) to generate first-strand cDNA. After second-strand synthesis, double-stranded cDNA was used in an in vitro transcription (IVT) reaction to generate cRNA (MEGAscript T7 kit; Ambion, Austin, TX). 600 ng of the cRNA obtained was used for a second round of cDNA and cRNA synthesis, resulting in biotinylated cRNA (GeneChip Expression 3′-Amplification Reagents for IVT-Labeling; Affymetrix). Size distribution of the cRNA and fragmented cRNA, respectively, were assessed using an Agilent 2100 Bioanalyzer with a RNA 6000 Nano Assay. 15 μg of fragmented cRNA was used in a 300-μl hybridization containing added hybridization controls. A final volume of 200 μl was hybridized on arrays for 16 h at 45°C. Standard post hybridization wash and double-stain protocols (EukGE-WS2v5) were used on an Affymetrix GeneChip Fluidics Station 400. Arrays were scanned on an Affymetrix GeneChip scanner 3000.
Microarray data analysis
Genechip expression data were quantile normalized in RMA Express , following examination of QC parameters (GAPDH ratios, log2PM distributions and RLE/NUSE plots). Normalized values were then analysed using the Rank Products method (Bioconductor Package RankProd). Rank Products (RP) is a non-parametric method used to detect genes that are consistently highly ranked (strongly up-regulated/down-regulated between two conditions), particularly in experiments with a small number of replicates where it has been shown to generate accurate results [27, 28]. The null hypothesis assumes that the order of all genes is random, thus the RPs are compared with the RPs for 1000 random permutations, with the same number of replicates and genes as the real experiment in order to correct for the multiple testing problem inherent in microarray experiments. To assign a significance level, the associated p-value and the false discovery rate (FDR) are included in the output alongside the genes that are detected by using certain criteria. This method has been used in various application domains, including proteomics, metabolomics, statistical meta-analysis, and general feature selection [29–31]. The gene list thus ranked according to the RP statistic can be further organised according to p value and FDR. A dendrogram showing how the individual array experiments clustered was also generated in BRB Arraytools, using centred correlation and average linkage. A MIAME-compliant microarray data submission  was made to the Gene Expression Omnibus (https://www.ncbi.nlm.nih.gov/geo/, accession number GSE40445).
Pathway and GO analysis
A reduced RP-ranked gene list was produced from the analysed microarray data by using a p-value cut-off of 0.0001 followed by removal of any gene with FDR > 0.05 and addition of a detection call filter (>20% present in all samples). This resulted in a list comprising 133 up-regulated and 255 down-regulated probesets (CF/Control: Additional file 1). This list was then submitted to the DAVID functional annotation tool [34, 35] (http://david.abcc.ncifcrf.gov/). This software generates a list of GO terms (classified as Biological Process, Cellular Compartment or Molecular Function) found to be enriched (ie, non-randomly distributed) in the submitted gene list. The same list was also used in pathway discovery using the GeneGo Metacore® platform (http://thomsonreuters.com/products_services/science/systems-biology/). Gene Set Enrichment Analysis  was used to assess the distributions in our data set of regulated genes (used as gene sets) from five comparable studies [16–20], and gene sets relevant to the previous analysis from the molecular signature database of the GSEA website (http://www.broadinstitute.org/gsea/msigdb/index.jsp). GSEA software provides an enrichment score (ES) and a p value to assess whether a given gene set is preferentially associated with one end of a data set, meaning that expression of the genes in a given gene set is associated with one of the phenotype groups under study.
Quantitative real time PCR
Primers used in qRT-PCR amplification
Official gene symbol
Primer sequences (5′to 3′)
ACTB (Ref. Gene)
GAPDH (Ref. Gene)
GJA1 (Connexin 43)
Comparison with published gene lists
Summary of independent microarray experiments compared in the present study
Clarke (this paper)
Native nasal epithelium (brushings)
5 CF (F508del homoz.) vs. 5 controls
Affymetrix Custom HsAirwaya520108F
Virella-Lowell et al., 2004 1
Isogenic bronchial cells (IB3-1 and S9)
F508del/W1282X vs. WT-CFTR corrected: 3 technical replicates each
Zabner et al., 2005 1
Primary tracheal and bronchial cell cultures
10 CF (F508del homoz.) vs. 10 controls
Wright et al., 2006 1,3
Native nasal epithelium (brushings)
4 CF (F508del homoz.) vs. 12 controls
Verhaeghe et al., 2007 1
Fetal tracheal cells (CFT-2 and NT-1)
F508del homoz. vs. WT-CFTR: 3 technical replicates each
Ogilvie et al., 2011 2
Native bronchial (and nasal) epithelium (brushings)
F508del homoz. (in most cases) vs. controls: (8 vs. 15 bronchial; 20 vs. 16 nasal)
Illumina HumanRef-8 v1 Expression BeadChips
Reanalysis of published dataset
For a more in-depth comparison between our gene list and the in vivo data from Ogilvie et al. (2011) , we reanalyzed their data (raw files available at http://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-360). The Bioconductor Lumi package (http://www.bioconductor.org/packages/2.11/bioc/html/lumi.html) was used for quality control and normalization. Following removal of two outlier samples, data from 78 Illumina HumanRef-8 v1 Expression BeadChips representing 20 CF nasal cell samples, 16 control nasal cell samples, 8 CF bronchial cell samples and 15 control bronchial cell samples (with some samples represented by 2–3 technical replicates) was quantile normalized and the normalized values subjected to RP analysis (CF-vs-control nasal and CF-vs-control bronchial), followed by detection call filtering (present at p < 0.01 in more than 10% of samples compared). Extended genelists were chosen using variable cutoffs as in exploratory gene array analysis , resulting in lists of 616 up- and 303 down-regulated genes in nasal epithelium and 441 up and 510 down-regulated genes in bronchial epithelium, including 69% and 70% of the originally published bronchial gene lists  (for gene lists resulting from new analysis see Additional file 3). For comparison of our dataset with the reanalysis of Ogilvie et al. (2011) , we lowered our fold change cutoffs and thereby extended our lists to comparable length, while still maintaining a false positive cutoff of pfp < 0.05 (for extended gene list from this study see Additional file 4). The final reanalyzed gene lists collapsed to single gene symbols and used for comparison are shown in Additional file 5. Regulated genes identified as shared were submitted to DAVID for GO analysis and GeneMania (http://www.genemania.org/) was used to generate a gene association network.
GO term enrichment and pathway analysis
Functional enrichment from DAVID analysis
GOTERM_BP_FAT (Biological process)
GO:0008285 negative regulation of cell proliferation
GO:0006953 acute-phase response
GO:0008202 steroid metabolic process
GO:0001960 negative regulation of cytokine-mediated signaling pathway
GO:0042127 regulation of cell proliferation
GO:0010038 response to metal ion
GOTERM_ CC_FAT (Cell compartment)
GO:0030286 dynein complex
GO:0005615 extracellular space
GO:0044421 extracellular region part
GO:0042995 cell projection
GO:0005875 microtubule associated complex
GO:0044430 cytoskeletal part
GO:0035085 cilium axoneme
GO:0005576 extracellular region
GOTERM_MF_FAT (Molecular Function)
GO:0003777 microtubule motor activity
GO:0005509 calcium ion binding
The full list of 388 regulated probesets was also submitted to the GeneGo Metacore tool (http://thomsonreuters.com/products_services/science/systems-biology/), which comprises an integrated knowledge database and software suite for pathway analysis of gene lists. As well as providing a curated literature-based pathway analysis tool, Metacore generates lists of GO terms (Processes, Molecular Functions or Localizations) found to be enriched in the submitted gene list, which were similar to those obtained using DAVID. The 3 most enriched “GO Processes” were “acute phase response” (p = 1.19E-05, n = 7 genes), “response to glucocorticoid stimulus” (p = 1.86E-05, n = 12 genes) and “cell projection organization” (p = 3.92E-05, n = 21 genes). The 3 most enriched “GO Localizations” were “microtubule” (p = 4.12E-06, n = 15 genes), “microtubule cytoskeleton” (p = 5.50E-06, n = 22 genes) and “cilium” (p = 2.46E-05, n = 10 genes). Of the “GO Molecular Functions”, the most enriched was “calcium ion binding” (p = 1.35E-04, n = 26 genes).
Using Metacore to construct networks based on the best documented relationships between proteins encoded by list genes, there were found to be few direct interactions between proteins encoded by list genes, and only a limited number of known CFTR interacting proteins (namely up-regulated channel proteins CLCA2 and AQP9). Metacore found the gene list to be enriched in targets for several important transcription factors, including SP1 (45 genes, p = 3.64E-113), c-MYC (21 genes, p = 5.75E-52), ESR1 (20 genes, p = 1.86E-49) and NF-κB (20 genes, p = 1.86E-49). NF-κB has been implicated as a mediator of IL-8 inflammatory signalling in CF , but (along with SP-1 and c-MYC) NF-κB itself was not found to be regulated in our data set, although significant enrichment of the NF-κB pathway was detected by GSEA (see below).
Relative qRT-PCR validation of microarray data
Genes for which differential expression was reanalysed by qRT-PCR in independent nasal cell samples
p (t test)
Upregulated in CF
Gap junction protein alpha 1 or Connexin 43
Insulin-like growth factor binding protein 3
N-myc downstream regulated gene 1
Transmembrane protein 45A or DERP7
Downregulated in CF
Secretoglobin family 1A member 1 or Uteroglobin
Sperm associated antigen 6
Glyceraldehyde 3-phosphate dehydrogenase
Comparison with other studies
Percentages of genes in common among differentially expressed genes from six microarray studies of CF related gene expression
(N = 300)
(N = 300)
(N = 300)
(N = 300)
(N = 115)
(N = 117)
(N = 300)
(N = 300)
(N = 300)
(N = 300)
(N = 110)
(N = 220)
(N = 300)
(N = 300)
(N = 300)
(N = 300)
(N = 115)
(N = 117)
(N = 300)
(N = 300)
(N = 300)
(N = 300)
(N = 110)
(N = 220)
Differentially expressed genes common to two or more of six comparable studies of CF related gene expression
Regulated genes shared with current study (Clarke)
Zabner et al., 2005
UP: ACAA2, CDKN2B.
Wright et al., 2006
UP: C9orf3, KRT14.
DOWN: CYP24A1, HLA-DQA1, SAA4.
Virella-Lowell et al., 2004
UP: CAV1, CCNE2.
DOWN: CLGN, ENO2, EPB41L3, GPX3, TIMP4.
Verhaeghe et al., 2007
UP: BCL2A1, G0S2, IL1B, MMP1, RGS2.
DOWN: CKB, CRIP1, CYP24A1, DNALI1, FHL1, GSTT1, IGFBP2.
Ogilvie et al., 2011
UP: BCL2A1, G0S2, IL1B, IL1R2, LCP2, NDRG1, RGS2, RNF149, TCN1.
DOWN: PROS1, SCGB1A1, SPAG8.
All shared genes (six studies combined)
UP (n = 75): ACAA2, AGL, AKR1C1, ANXA8L2, BCHE, BLOC1S1, BTBD3, C9ORF3, CAPG, CAV1, CCL20, CCNE2, CD24, CD83, CDKN2B, CLGN, CSF3, DDX3Y, ELK3, FOLR1, FOLR3, FOXG1, GCA, GFPT2, HCLS1, HIST1H1C, HMOX1, HPCAL1, HSPB11, IFIT1, IFIT3, IFITM1, IL1R2, IL7R, ISG15, KCTD12, KRT14, KRT81, LCP2, LITAF, LYPD1, MLF1, MMP1, MRPL28, MX2, NCF1, NDRG1, NET1, PLAU, PLAUR, PLTP, PRSS3, PSG9, PTPN13, RAC2, RAGE, RNF149, RPA3, SEMA3B, SERPINA3, SERPINF1, SLITRK5, SOD2, SULT1A3, TCF15, TCIRG1, TCN1, TXNIP, BCL2A1 , G0S2 , IL1B , NCF2, PLAT, PTGS2, RGS2 .
DOWN (n = 114): ACTA2, ADAR, ALDH1A1, ANKRD1, ASNS, ASS1, BEX4, BST2, BTG1, C5ORF13, CALD1, CAP1, CCL20, CD164, CFB, CGREF1, CKB, CLGN, COL8A1, COL9A3, CRIP1, CSTA, CXCR4, CYP51A1, DDB2, DDIT4, DNALI1, DSP, DSTN, DYNLT1, EDNRA, EFEMP1, EIF4A2, ENO2, EPB41L3, EPS8, F3, FBLN5, FCGBP, FHL1, GABRP, GCH1, GINS1, GPNMB, GPR1, GPX3, GSTT1, GZMB, HCP5, HES1, HLA-B, HLA-DQA1, HLA-DRA, HLA-DRB1, HLA-F, HLA-G, HMGCS1, HSPB1, HTRA1, ID1, IFI16, IFITM1, IFRD1, IGFBP7, IL32, KCNN4, KIT, KLRK1, KRT15, LCN2, LGALS3BP, LOX, MGAM, MSC, NID2, NPR3, NTS, PGD, PNMA2, PPP1R3C, PROS1, RASGRP1, RND3, RUNX3, SAA4, SC5DL, SCGB1A1, SERPINB3, SERPINB4, SGK1, SLC2A3P1, SNAPC1, SPAG8, STAC, STAT4, TES, TFPI, THBD, TIMP4, TPBG, TRIB2, TWIST1, VDAC1, ZNF643, CTSC, CYP24A1 , IFI27, IGFBP2 , IGFBP3, NAMPT, PRSS23, SEL1L3, TMSB4X, TRIM22.
Functional enrichment analysis of genes shared between two or three studies
GO:0006952 defence response
GO:0042060 wound healing
GO:0042127 regulation of cell proliferation
GO:0006955 immune response
GO:0006928 cell motion
GO:0005576 extracellular region
GO:0044421 extracellular region part
GO:0005615 extracellular space
GO:0000267 cell fraction
GO:0005625 soluble fraction
GO:0004857 enzyme inhibitor activity
GO:0032395 MHC class II receptor activity
GO:0004866 endopeptidase inhibitor activity
GO:0030414 peptidase inhibitor activity
GO:0005520 insulin-like growth factor binding
hsa05332: Graft-versus-host disease
hsa04940: Type I diabetes mellitus
hsa05330: Allograft rejection
hsa04610: Complement and coagulation cascades
hsa05416: Viral myocarditis
Gene set enrichment analysis
Reanalysis of independent dataset, and comparison with our data
Small molecular signature for native CF airway epithelial cells
arachidonate 5-lipoxygenase-activating protein; #241
BCL2-related protein A1; #597
chemokine (C-X-C motif) receptor 4; #7852
Fc fragment of IgG, low affinity IIIa, receptor (CD16a); #2214
FBJ murine osteosarcoma viral oncogene homolog; #2353
G0/G1switch 2; #50486
glucosaminyl (N-acetyl) transferase 3, mucin type; #9245
histone cluster 1, H1c; #3006
histone cluster 1, H2bk; #85236
interleukin 1, beta; #3553
interleukin 1 receptor, type II; #7850
lipopolysaccharide-induced TNF factor; #9516
oncostatin M; #5008
prostaglandin-endoperoxide synthase 2; #5743
regulator of G-protein signaling 2, 24 kDa; #5997
S100 calcium binding protein A8; #6279
S100 calcium binding protein A9; #6280
serpin peptidase inhibitor, clade A, member 3; #12
transcobalamin I (vitamin B12 bp, R binder family); #6947
tropomyosin 4; #7171
carbohydrate (N-acet.gal.am. 4–0) sulfotransferase 9; #83539
dynein, axonemal, light intermediate chain 1; #7802
monoamine oxidase B; #4129
NEL-like 2 (chicken); #4753
protein phosphatase 1, regulatory subunit 16A; #84988
protein S (alpha); #5627
prostaglandin F receptor (FP); #5737
MOK protein kinase; #5891
The present study of global gene expression in nasal epithelial cell samples from CF patients and healthy controls yields a snapshot of the CF transcriptome providing interesting insights into the consequences of CFTR dysfunction. Our primary aim was the identification of a CF molecular signature – a robust set of genes with potential utility as diagnostic markers or as targets for future therapeutic strategies. The approach we used – applying the same statistical method to both newer and older data, allowed us to propose such a signature, while also shedding light on the limitations of such snapshot studies in measuring a system as subtle and dynamic as the transcriptome.
Many of the individual genes within our microarray gene list are of known functional significance in CF pathophysiology, but for a better understanding of the cellular processes and pathways altered in CF epithelium we utilised Gene Ontology (GO) term enrichment in the whole lists of 133 up-regulated and 255 down-regulated genes. The most highly enriched GO terms were “negative regulation of cell proliferation” (biological process), mainly composed of up-regulated genes, “cilium” (cell compartment), and “microtubule motor activity” (molecular function) which were both composed of only down-regulated genes. “Extracellular space” (cell compartment) and “calcium ion binding” (molecular function) also accounted for a significant number of regulated genes in both directions (see Table 4). Taken together, these systemic alterations of gene expression might imply that CFTR dysfunction causes: 1) ER stress and alteration of calcium signalling, plausibly to activate alternative chloride channels, 2) disturbances in the normal processes of epithelial cell differentiation and extracellular signalling, and 3) a reduction in ciliogenesis or cilia activity.
Proliferation and inflammation
In airway epithelium, the proliferating cell population is likely to be composed of the basal-like, rather than the ciliated epithelial cells , and in fact the two populations can be seen as extremes of a proliferation-differentiation continuum. Of the 14 genes classified as negative regulators of cell proliferation (anti-proliferative) GO group, 12 (86%) were up-regulated in CF, suggesting that, overall, proliferation is reduced in CF. However, GSEA analysis showed significant enrichment of proliferative genes in general, and sets of both positive and negative regulators of proliferation in CF (see Figure 6). In similar situations elucidation of the function of each individual regulated protein builds up a picture of a complex network of apparently contradictory processes. Examples found here to be up-regulated in CF include ADM (adrenomedullin), a vasodilator which promotes alveolar development and repair , and which is speculated to have a protective effect in the immuno-inflammatory process of asthma , implying that it may respond to airway injury in CF. EREG (epiregulin), also up-regulated in CF, is a member of the epidermal growth factor (EGF) family generally associated with enhanced proliferation, but which in ciliated human airway epithelial cells can act via ERBB2 binding to maintain their differentiated phenotype . GJA1 (Connexin 43), also up-regulated in CF, suppresses cell proliferation via maintenance of cell-cell communication, possibly via an association with CAV1 , which has an important role in maintenance of airway ECM integrity via inhibition of the TGF beta-induced fibrosis , and may also play an important role in modulating the immune response to P. aeruginosa infection through the formation of CFTR-expressing lipid rafts . Other genes in this group are inflammatory cytokines (IL1B) or potentiate cell proliferation (FGFBP1, IGFBP3), and upon inspection there are few bona fide inhibitors of proliferation (OSM, CDKN2B). Taken together, the up-regulated genes belonging to the “negative regulation of Cell Proliferation” GO group may therefore represent a complex reaction to the effects of hyper-inflammation and tissue injury which are hallmarks of the CF airway. Given, however, that this was measured in the nasal epithelium, which is free of much of the pathology inherent in the lower CF airway, it can also be argued that these patterns, including the hyperinflammatory component, form part of a primary response to CFTR dysfunction. It is nevertheless probable that while some of these genes are deregulated as a direct consequence of absence of CFTR, others are secondary, i.e., actively involved in balancing the negative effects of the former.
AREG (amphiregulin), identified here as up-regulated in CF, is a binding partner of EREG and a ligand of EGFR which was a representative of “extracellular space”, another highly enriched GO group. AREG is present in the sputum of CF patients , and is involved in both proliferation and inflammation in human airways [49, 50]. Its expression in CF airway and blood neutrophils , might be suggestive that a component of the gene expression observed in nasal cell samples analysed here was derived from the 5-10% of inflammatory cells present (see Table 1). However, neutrophil genes in general were not over-represented in our gene lists: only 4 genes stringently identified as regulated in CF airway neutrophils  were present among our up-regulated genes (CXCR2, CXCL9, CXCL10, and ADM, which is supposedly down-regulated in CF neutrophils ), and none among our down-regulated genes. Most of our CF gene expression profile can therefore be associated to the nasal epithelial cell population, and the presence on our list of several genes associated with inflammation lends support to the idea that CFTR dysfunction on its own can stimulate inflammatory signalling to some extent.
The cilium is an organelle directly affected by CF pathophysiology, given its role in mucus clearance and the physical barrier to such clearance in CF. Given that the CF nasal epithelium is not burdened with abnormally thick mucus to the same extent as the CF lung, the down-regulation of cilia genes in this tissue suggests a primary disruption in CFTR related signalling rather than a secondary response related to abnormal mucus. Of the 10 down-regulated genes in this GO group, 5 are axonemal components (DNAH9, DNAH12, DNAI1, DNAI2 and DNAAF1/LRRC50) and 4 of the others are clearly involved in cilium or sperm flagellum function. Furthermore, other down-regulated genes which were not flagged in this GO term by DAVID can clearly be assigned to this group, increasing its significance (DNAH6, DNALI1, DNAAF3/ LOC352909 and TEKT1). Down-regulation in CF was broadly confirmed for two of these genes (SPAG6 and TEKT1) by rtQ-PCR in independent nasal cell samples (see Figure 3).
Suppression of cilia gene expression as a primary consequence of CFTR dysfunction might complicate the already compromised mucociliary clearance that is a hallmark of CF. Furthermore, many of these genes are also relevant to spermatogenesis or sperm motility via their role in the flagellum, and might therefore be relevant to cases of CF-associated male infertility (non-CBAVD [51–53]). Interestingly, GJA1 (CX43), an up-regulated gene in CF, has a functional link to cilia, given that in epithelial cells of nasal mucosa, only functional gap junctions of GJA1 are expressed , and it is through these junctions that the intracellular calcium wave that controls the beating of cilia is communicated. Taken together, our expression data suggest that CFTR dysfunction might predispose the airway to suboptimal cilia function, thereby compounding the CF phenotype.
Estrogen receptor targets
Several of our list genes are targets of the transcription factor ESR1 (see Figure 2), and targets for ESR1 were found to be significantly enriched in our CF samples by GSEA (see Figure 6). The presence of CF up/down-regulated genes in a network provides clues not only on how expression is affected by CFTR dysfunction, but also the opposite, e.g., how systemic alterations in circulating estrogen over the course of the female menstrual cycle, might bring about differential gene expression profiles, which help to explain subtle differences in lung function in male and female CF patients [40, 55].
Our data on changes in gene expression in proliferation pathways, calcium, membrane and cilia biology can all be related back to the defect in CFTR processing, and can potentially be characterized within a model of a perturbed CFTR interactome . The involvement of estrogenic signalling, however, introduces an “external modifier” providing a feasible explanation for some of the variability within a heterogeneous group of subjects. Although functional data are outside the scope of the present study, the ESR1 network identified here constitutes a source for clues as to how there may be crosstalk between different mechanisms during dysregulation of gene expression in CF. A sharp and sustained rise in circulating estrogen and its presumably positive effect on expression of several genes involved in regulating the transition between proliferative and differentiating cellular phenotypes (MMP1, ADM, AREG, GJA1, RUNX2: all found here to be up-regulated in CF, see Figure 2) might be a key factor in determining the equilibrium between these two states in the female CF airway epithelium.
Comparison with other studies
The present study provides a momentary snapshot of Cystic Fibrosis-related gene expression in native nasal epithelial cell samples from CF patients compared to controls. We compared our gene-lists with those from other studies in an attempt to quantify the similarity between data sets. These comparative data enabled us to gauge which of the six studies are more alike in terms of the numbers of up-regulated and down-regulated genes they have in common (see Table 6), showing, for example, that our data have more up-regulated genes in common with a bronchial cell dataset  and more down-regulated genes in common with the study using immortalized foetal tracheal cells . These results are set against a background of similar numbers of genes whose direction of expression is inverted between studies (see Figure 4). GSEA data point to partial inversion of gene regulation between our study and one also using native nasal epithelial cells  (see Figure 5D,F), but numbers of regulated genes also show inverted expression between several other studies (see Table 6). It is possible to speculate that the direction of CF marker gene regulation might not be as important as the fact of their deregulation in CF, and that the appearance of pathophysiologically relevant genes at different extremes of distinct studies might simply reflect fluctuations in what is a cyclical process of infection, inflammation, and airway repair, but in reality, the presence of biological replicates should cancel out any such effects. Three pathways previously suggested to be characteristic of CF-related gene expression , were found to be enriched to varying degrees in our data set (see Figure 6), including significant enrichment of the NF-kB pathway as previously noted in foetal tracheal cell lines , and used as evidence of intrinsic hyper-inflammation in CF. The fact that this is observed in nasal epithelium, along with over-expression of several genes related to the regulation of the inflammation does lend support to the presence of an intrinsic hyperinflammatory response associated with F508del-CFTR expression without, however, clarifying its origin . The shared down-regulation of the antigen presentation pathway suggested by Hampton & Stanton  was also seen to some extent, and these data help to characterize our data set as belonging to the same group as the other CF data sets. Taken together (see Table 7) the 189 genes which share similar expression between 2 or 3 studies are enriched for functional categories (eg, defence response, wound healing, regulation of cell proliferation: see Table 8) which succinctly sum up the processes involved in CF, and whose expression might well prove to be a reliable marker of CF. However, for a more feasible molecular signature of CF, we decided to reanalyse the raw microarray data of Ogilvie et al. , using the RP method as a way of seeing “further down” their data sets for both nasal epithelium, in which they only identified a handful of significantly regulated genes, and bronchial epithelium, which was the tissue which shared more up-regulated genes with our dataset in the preliminary comparison.
Use of the RP statistical method to detect even incremental differential gene expression with a high level of significance allowed us to produce gene lists of comparable sizes for both nasal and bronchial epithelial cell samples from the Ogilvie et al.  study. Our analysis identified a large number of regulated genes shared by bronchial and nasal epithelium, partly contradicting the authors’ claim that nasal epithelium is not a good surrogate for the CF airway , but also supporting that claim given the greatly reduced fold change of expression shown by these genes in nasal cells (see Additional file 3). Comparing the reanalysed gene lists with our own nasal cell data (see Figure 7) identified 30 genes regulated in all three lists (see Table 9). These genes represent a small putative molecular signature for F508del-CFTR expression in airway epithelial cells. A significant number of these genes are involved in inflammation, defence, and responses to wounding, and a number of them are involved in extracellular signalling and calcium ion binding (see Figure 7C). Construction of an association network (see Figure 7D) shows that all of these genes have been linked by co-expression or co-localization in other studies, or have proven functional relationships with each other and with other genes known to be regulated in CF. The most connected gene in this network is IL1B, an important mediator of the inflammatory response and a known modifier of CF lung disease .
In summary, our small-scale microarray study of the CF nasal epithelial transcriptome has generated a list of differentially expressed genes which mostly suggest defects in gene regulation networks related to cell proliferation and cilia biology. Comparison of our data set with previously published studies allowed us to assess the consistency of independent microarray data sets, thus revealing the limitations of such snapshot studies in measuring a system as subtle and dynamic as the transcriptome and suggesting that a molecular signature for CF is likely to be of elevated plasticity. Nevertheless, similarities in pathway and Gene Ontology enrichment between our data set and shared genes from other data sets do give evidence for common gene expression components with elevated functional significance to both primary and secondary cellular responses to F508del-CFTR. The novelty of our approach lies in the new perspectives enabled by the application of new statistical analyses to both new and old data sets, and underlines the importance of public data repositories for high throughput data. This has allowed us to identify a small molecular signature characterizing F508del-CFTR expression in both nasal and bronchial native airway epithelium which we believe is worthy of further investigation. Future studies may be able to refine this signature and test its value as a predictive tool for discriminating between CF and healthy tissue samples. Comparing the signature genes with functional genomics data may also help to clarify cellular responses to CFTR dysfunction in the airway epithelium.
Congenital bilateral absence of vas deferens
CF transmembrane conductance regulator
Database for annotation, visualization and integrated discovery
False discovery rate
Gene set enrichment analysis
Rank products (analysis).
This work was financed by national funds through the FCT – Fundação para a Ciência e a Tecnologia - under the following projects: PTDC/SAU-GMG/122299/2010, PIC/IC/83103/2007, and PEst-OE/BIA/UI4046/2011 (BioFIG). LS was supported by the following FCT grants: PTDC/MAT/118335/2010 and PEst-OE/MAT/UI0006/2011 (CEAUL). We thank Dr. Paul McCray for supplying us with HsAirwaya520108F microarrays and for many helpful discussions, Dr. Pilar Azevedo (Hospital Santa Maria, Lisboa) for access to patients for control nasal cell samples, and Dr. Jörg Becker (Instituto Gulbenkian de Ciência microarray facility, Oeiras) for array hybridization and scanning. We also thank Dr. Tom Hampton for providing the re-analysed gene lists from his publication, Dr. Jerry Wright for discussions concerning his data, and Dr. Chris Boyd, Dr. Rob Kitchen and Dr. Varrie Ogilvie for answering questions about their data set. Finally we thank two anonymous reviewers for helping to improve the manuscript.
- Hodson M: Cystic fibrosis: the disease. Monaldi Arch Chest Dis. 1993, 48 (6): 647-652.PubMedGoogle Scholar
- Riordan J, Rommens J, Kerem B, Alon N, Rozmahel R, Grzelczak Z, Zielenski J, Lok S, Plavsic N, Chou J: Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science. 1989, 245 (4922): 1066-1073. 10.1126/science.2475911.PubMedView ArticleGoogle Scholar
- Frizzell R: Functions of the cystic fibrosis transmembrane conductance regulator protein. Am J Respir Crit Care Med. 1995, 151 (3 Pt 2): S54-S58.PubMedView ArticleGoogle Scholar
- Rowntree R, Harris A: The phenotypic consequences of CFTR mutations. Ann Hum Genet. 2003, 67 (Pt 5): 471-485.PubMedView ArticleGoogle Scholar
- Kerem B, Rommens J, Buchanan J, Markiewicz D, Cox T, Chakravarti A, Buchwald M, Tsui L: Identification of the cystic fibrosis gene: genetic analysis. Science. 1989, 245 (4922): 1073-1080. 10.1126/science.2570460.PubMedView ArticleGoogle Scholar
- Bobadilla J, Macek MJ, Fine J, Farrell P: Cystic fibrosis: a worldwide analysis of CFTR mutations–correlation with incidence data and application to screening. Hum Mutat. 2002, 19 (6): 575-606. 10.1002/humu.10041.PubMedView ArticleGoogle Scholar
- Kerem E, Corey M, Kerem B, Rommens J, Markiewicz D, Levison H, Tsui L, Durie P: The relation between genotype and phenotype in cystic fibrosis–analysis of the most common mutation (delta F508). N Engl J Med. 1990, 323 (22): 1517-1522. 10.1056/NEJM199011293232203.PubMedView ArticleGoogle Scholar
- Sun L, Rommens JM, Corvol H, Li W, Li X, Chiang TA, Lin F, Dorfman R, Busson PF, Parekh RV: Multiple apical plasma membrane constituents are associated with susceptibility to meconium ileus in individuals with cystic fibrosis. Nat Genet. 2012, 44 (5): 562-569. 10.1038/ng.2221.PubMedPubMed CentralView ArticleGoogle Scholar
- Henderson LB, Doshi VK, Blackman SM, Naughton KM, Pace RG, Moskovitz J, Knowles MR, Durie PR, Drumm ML, Cutting GR: Variation in MSRA modifies risk of neonatal intestinal obstruction in cystic fibrosis. PLoS Genet. 2012, 8 (3): e1002580-10.1371/journal.pgen.1002580.PubMedPubMed CentralView ArticleGoogle Scholar
- Guo X, Pace RG, Stonebraker JR, Commander CW, Dang AT, Drumm ML, Harris A, Zou F, Swallow DM, Wright FA: Mucin variable number tandem repeat polymorphisms and severity of cystic fibrosis lung disease: significant association with MUC5AC. PLoS One. 2011, 6 (10): e25452-10.1371/journal.pone.0025452.PubMedPubMed CentralView ArticleGoogle Scholar
- Bezzerri V, d’Adamo P, Rimessi A, Lanzara C, Crovella S, Nicolis E, Tamanini A, Athanasakis E, Tebon M, Bisoffi G: Phospholipase C-β3 is a key modulator of IL-8 expression in cystic fibrosis bronchial epithelial cells. J Immunol. 2011, 186 (8): 4946-4958. 10.4049/jimmunol.1003535.PubMedPubMed CentralView ArticleGoogle Scholar
- Gu Y, Harley IT, Henderson LB, Aronow BJ, Vietor I, Huber LA, Harley JB, Kilpatrick JR, Langefeld CD, Williams AH: Identification of IFRD1 as a modifier gene for cystic fibrosis lung disease. Nature. 2009, 458 (7241): 1039-1042. 10.1038/nature07811.PubMedPubMed CentralView ArticleGoogle Scholar
- Collaco JM, Cutting GR: Update on gene modifiers in cystic fibrosis. Curr Opin Pulm Med. 2008, 14 (6): 559-566. 10.1097/MCP.0b013e3283121cdc.PubMedPubMed CentralView ArticleGoogle Scholar
- Clunes LA, Davies CM, Coakley RD, Aleksandrov AA, Henderson AG, Zeman KL, Worthington EN, Gentzsch M, Kreda SM, Cholon D: Cigarette smoke exposure induces CFTR internalization and insolubility, leading to airway surface liquid dehydration. FASEB J. 2012, 26 (2): 533-545. 10.1096/fj.11-192377.PubMedPubMed CentralView ArticleGoogle Scholar
- Wang X, Venable J, LaPointe P, Hutt D, Koulov A, Coppinger J, Gurkan C, Kellner W, Matteson J, Plutner H: Hsp90 cochaperone Aha1 downregulation rescues misfolding of CFTR in cystic fibrosis. Cell. 2006, 127 (4): 803-815. 10.1016/j.cell.2006.09.043.PubMedView ArticleGoogle Scholar
- Virella-Lowell I, Herlihy J, Liu B, Lopez C, Cruz P, Muller C, Baker H, Flotte T: Effects of CFTR, interleukin-10, and Pseudomonas aeruginosa on gene expression profiles in a CF bronchial epithelial cell Line. Mol Ther. 2004, 10 (3): 562-573. 10.1016/j.ymthe.2004.06.215.PubMedView ArticleGoogle Scholar
- Zabner J, Scheetz TE, Almabrazi HG, Casavant TL, Huang J, Keshavjee S, McCray PB: CFTR DeltaF508 mutation has minimal effect on the gene expression profile of differentiated human airway epithelia. Am J Physiol Lung Cell Mol Physiol. 2005, 289 (4): L545-L553. 10.1152/ajplung.00065.2005.PubMedView ArticleGoogle Scholar
- Wright JM, Merlo CA, Reynolds JB, Zeitlin PL, Garcia JG, Guggino WB, Boyle MP: Respiratory epithelial gene expression in patients with mild and severe cystic fibrosis lung disease. Am J Respir Cell Mol Biol. 2006, 35 (3): 327-336. 10.1165/rcmb.2005-0359OC.PubMedPubMed CentralView ArticleGoogle Scholar
- Ogilvie V, Passmore M, Hyndman L, Jones L, Stevenson B, Wilson A, Davidson H, Kitchen RR, Gray RD, Shah P: Differential global gene expression in cystic fibrosis nasal and bronchial epithelium. Genomics. 2011, 98 (5): 327-336. 10.1016/j.ygeno.2011.06.008.PubMedView ArticleGoogle Scholar
- Verhaeghe C, Remouchamps C, Hennuy B, Vanderplasschen A, Chariot A, Tabruyn S, Oury C, Bours V: Role of IKK and ERK pathways in intrinsic inflammation of cystic fibrosis airways. Biochem Pharmacol. 2007, 73 (12): 1982-1994. 10.1016/j.bcp.2007.03.019.PubMedView ArticleGoogle Scholar
- Hampton T, Stanton B: A novel approach to analyze gene expression data demonstrates that the DeltaF508 mutation in CFTR downregulates the antigen presentation pathway. Am J Physiol Lung Cell Mol Physiol. 2010, 298 (4): L473-L482. 10.1152/ajplung.00379.2009.PubMedPubMed CentralView ArticleGoogle Scholar
- Beck S, Penque D, Garcia S, Gomes A, Farinha C, Mata L, Gulbenkian S, Gil-Ferreira K, Duarte A, Pacheco P: Cystic fibrosis patients with the 3272–26A– > G mutation have mild disease, leaky alternative mRNA splicing, and CFTR protein at the cell membrane. Hum Mutat. 1999, 14 (2): 133-144. 10.1002/(SICI)1098-1004(1999)14:2<133::AID-HUMU5>3.0.CO;2-T.PubMedView ArticleGoogle Scholar
- Harris CM, Mendes F, Dragomir A, Doull IJ, Carvalho-Oliveira I, Bebok Z, Clancy JP, Eubanks V, Sorscher EJ, Roomans GM: Assessment of CFTR localisation in native airway epithelial cells obtained by nasal brushing. J Cyst Fibros. 2004, 3 (Suppl 2): 43-48.PubMedView ArticleGoogle Scholar
- Penque D, Mendes F, Beck S, Farinha C, Pacheco P, Nogueira P, Lavinha J, Malhó R, Amaral M: Cystic fibrosis F508del patients have apically localized CFTR in a reduced number of airway cells. Lab Invest. 2000, 80 (6): 857-868. 10.1038/labinvest.3780090.PubMedView ArticleGoogle Scholar
- Scheetz T, Zabner J, Welsh M, Coco J, Eyestone MF, Bonaldo M, Kucaba T, Casavant T, Soares M, McCray PJ: Large-scale gene discovery in human airway epithelia reveals novel transcripts. Physiol Genomics. 2004, 17 (1): 69-77. 10.1152/physiolgenomics.00188.2003.PubMedView ArticleGoogle Scholar
- Irizarry R, Hobbs B, Collin F, Beazer-Barclay Y, Antonellis K, Scherf U, Speed T: Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics. 2003, 4 (2): 249-264. 10.1093/biostatistics/4.2.249.PubMedView ArticleGoogle Scholar
- Breitling R, Armengaud P, Amtmann A, Herzyk P: Rank products: a simple, yet powerful, new method to detect differentially regulated genes in replicated microarray experiments. FEBS Lett. 2004, 573 (1–3): 83-92.PubMedView ArticleGoogle Scholar
- Breitling R, Herzyk P: Rank-based methods as a non-parametric alternative of the T-statistic for the analysis of biological microarray data. J Bioinform Comput Biol. 2005, 3 (5): 1171-1189. 10.1142/S0219720005001442.PubMedView ArticleGoogle Scholar
- Hong F, Breitling R, McEntee C, Wittner B, Nemhauser J, Chory J: RankProd: a bioconductor package for detecting differentially expressed genes in meta-analysis. Bioinformatics. 2006, 22 (22): 2825-2827. 10.1093/bioinformatics/btl476.PubMedView ArticleGoogle Scholar
- Smit S, van Breemen M, Hoefsloot H, Smilde A, Aerts J, de Koster C: Assessing the statistical validity of proteomics based biomarkers. Anal Chim Acta. 2007, 592 (2): 210-217. 10.1016/j.aca.2007.04.043.PubMedView ArticleGoogle Scholar
- Figueiredo A, Fortes A, Ferreira S, Sebastiana M, Choi Y, Sousa L, Acioli-Santos B, Pessoa F, Verpoorte R, Pais M: Transcriptional and metabolic profiling of grape (Vitis vinifera L.) leaves unravel possible innate resistance against pathogenic fungi. J Exp Bot. 2008, 59 (12): 3371-3381. 10.1093/jxb/ern187.PubMedView ArticleGoogle Scholar
- Simon R, Lam A, Li MC, Ngan M, Menenzes S, Zhao Y: Analysis of gene expression data using BRB-ArrayTools. Cancer Inform. 2007, 3: 11-17.PubMedPubMed CentralGoogle Scholar
- Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, Stoeckert C, Aach J, Ansorge W, Ball C, Causton H: Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat Genet. 2001, 29 (4): 365-371. 10.1038/ng1201-365.PubMedView ArticleGoogle Scholar
- Dennis GJ, Sherman B, Hosack D, Yang J, Gao W, Lane H, Lempicki R: DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003, 4 (5): P3-10.1186/gb-2003-4-5-p3.PubMedView ArticleGoogle Scholar
- Huang DW, Sherman BT, Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009, 4 (1): 44-57.View ArticleGoogle Scholar
- Subramanian A, Tamayo P, Mootha VK, Mukherjee S, Ebert BL, Gillette MA, Paulovich A, Pomeroy SL, Golub TR, Lander ES: Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA. 2005, 102 (43): 15545-15550. 10.1073/pnas.0506580102.PubMedPubMed CentralView ArticleGoogle Scholar
- Spandidos A, Wang X, Wang H, Seed B: PrimerBank: a resource of human and mouse PCR primer pairs for gene expression detection and quantification. Nucleic Acids Res. 2010, 38 (Database issue): D792-D799. doi:D792.PubMedPubMed CentralView ArticleGoogle Scholar
- Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001, 29 (9): e45-10.1093/nar/29.9.e45.PubMedPubMed CentralView ArticleGoogle Scholar
- Vij N, Mazur S, Zeitlin PL: CFTR is a negative regulator of NFkappaB mediated innate immune response. PLoS One. 2009, 4 (2): e4664-10.1371/journal.pone.0004664.PubMedPubMed CentralView ArticleGoogle Scholar
- Coakley R, Sun H, Clunes L, Rasmussen J, Stackhouse J, Okada S, Fricks I, Young S, Tarran R: 17beta-Estradiol inhibits Ca2 + −dependent homeostasis of airway surface liquid volume in human cystic fibrosis airway epithelia. J Clin Invest. 2008, 118 (12): 4025-4035.PubMedPubMed CentralGoogle Scholar
- Voynow J, Fischer B, Roberts B, Proia A: Basal-like cells constitute the proliferating cell population in cystic fibrosis airways. Am J Respir Crit Care Med. 2005, 172 (8): 1013-1018. 10.1164/rccm.200410-1398OC.PubMedView ArticleGoogle Scholar
- Vadivel A, Abozaid S, van Haaften T, Sawicka M, Eaton F, Chen M, Thébaud B: Adrenomedullin promotes lung angiogenesis, alveolar development, and repair. Am J Respir Cell Mol Biol. 2010, 43 (2): 152-160. 10.1165/rcmb.2009-0004OC.PubMedView ArticleGoogle Scholar
- Kucukosmanoglu E, Keskin O, Karcin M, Cekmen M, Balat A: Plasma adrenomedullin levels in children with asthma: Any relation with atopic dermatitis?. Allergol Immunopathol (Madr). 2011, 40 (4): 215-219.View ArticleGoogle Scholar
- Vermeer PD, Panko L, Karp P, Lee JH, Zabner J: Differentiation of human airway epithelia is dependent on erbB2. Am J Physiol Lung Cell Mol Physiol. 2006, 291 (2): L175-L180. 10.1152/ajplung.00547.2005.PubMedView ArticleGoogle Scholar
- Langlois S, Cowan K, Shao Q, Cowan B, Laird D: The tumor-suppressive function of Connexin43 in keratinocytes is mediated in part via interaction with caveolin-1. Cancer Res. 2010, 70 (10): 4222-4232. 10.1158/0008-5472.CAN-09-3281.PubMedView ArticleGoogle Scholar
- Le Saux O, Teeters K, Miyasato S, Choi J, Nakamatsu G, Richardson JA, Starcher B, Davis EC, Tam EK, Jourdan-Le Saux C: The role of caveolin-1 in pulmonary matrix remodeling and mechanical properties. Am J Physiol Lung Cell Mol Physiol. 2008, 295 (6): L1007-L1017. 10.1152/ajplung.90207.2008.PubMedPubMed CentralView ArticleGoogle Scholar
- Gadjeva M, Paradis-Bleau C, Priebe G, Fichorova R, Pier G: Caveolin-1 modifies the immunity to Pseudomonas aeruginosa. J Immunol. 2010, 184 (1): 296-302. 10.4049/jimmunol.0900604.PubMedPubMed CentralView ArticleGoogle Scholar
- Adib-Conquy M, Pedron T, Petit-Bertron A, Tabary O, Corvol H, Jacquot J, Clément A, Cavaillon J: Neutrophils in cystic fibrosis display a distinct gene expression pattern. Mol Med. 2008, 14 (1–2): 36-44.PubMedPubMed CentralGoogle Scholar
- Wang SW, Oh CK, Cho SH, Hu G, Martin R, Demissie-Sanders S, Li K, Moyle M, Yao Z: Amphiregulin expression in human mast cells and its effect on the primary human lung fibroblasts. J Allergy Clin Immunol. 2005, 115 (2): 287-294. 10.1016/j.jaci.2004.11.037.PubMedView ArticleGoogle Scholar
- Chokki M, Eguchi H, Hamamura I, Mitsuhashi H, Kamimura T: Human airway trypsin-like protease induces amphiregulin release through a mechanism involving protease-activated receptor-2-mediated ERK activation and TNF alpha-converting enzyme activity in airway epithelial cells. FEBS J. 2005, 272 (24): 6387-6399. 10.1111/j.1742-4658.2005.05035.x.PubMedView ArticleGoogle Scholar
- Tomaiuolo R, Fausto M, Elce A, Strina I, Ranieri A, Amato F, Castaldo G, De Placido G, Alviggi C: Enhanced frequency of CFTR gene variants in couples who are candidates for assisted reproductive technology treatment. Clin Chem Lab Med. 2011, 49 (8): 1289-1293.PubMedView ArticleGoogle Scholar
- Xu HM, Li HG, Xu LG, Zhang JR, Chen WY, Shi QX: The decline of fertility in male uremic patients is correlated with low expression of the cystic fibrosis transmembrane conductance regulator protein (CFTR) in human sperm. Hum Reprod. 2012, 27 (2): 340-348. 10.1093/humrep/der403.PubMedView ArticleGoogle Scholar
- Smith H: Fertility in men with cystic fibrosis assessment, investigations and management. Paediatr Respir Rev. 2010, 11 (2): 80-83. 10.1016/j.prrv.2010.01.003.PubMedView ArticleGoogle Scholar
- Yeh TH, Su MC, Hsu CJ, Chen YH, Lee SY: Epithelial cells of nasal mucosa express functional gap junctions of connexin 43. Acta Otolaryngol. 2003, 123 (2): 314-320. 10.1080/0036554021000028104.PubMedView ArticleGoogle Scholar
- Johannesson M, Lúdvíksdóttir D, Janson C: Lung function changes in relation to menstrual cycle in females with cystic fibrosis. Respir Med. 2000, 94 (11): 1043-1046. 10.1053/rmed.2000.0891.PubMedView ArticleGoogle Scholar
- Machen TE: Innate immune response in CF airway epithelia: hyperinflammatory?. Am J Physiol Cell Physiol. 2006, 291 (2): C218-C230. 10.1152/ajpcell.00605.2005.PubMedView ArticleGoogle Scholar
- Labenski H, Hedtfeld S, Becker T, Tümmler B, Stanke F: Initial interrogation, confirmation and fine mapping of modifying genes: STAT3, IL1B and IFNGR1 determine cystic fibrosis disease manifestation. Eur J Hum Genet. 2011, 19 (12): 1281-1288. 10.1038/ejhg.2011.129.PubMedPubMed CentralView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.