help button home button
AJRCMB
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on September 27, 2007, doi:10.1165/rcmb.2007-0278OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
2007-0278OCv1
38/3/293    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vroling, A. B.
Right arrow Articles by van Drunen, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vroling, A. B.
Right arrow Articles by van Drunen, C. M.
American Journal of Respiratory Cell and Molecular Biology. Vol. 38, pp. 293-299, 2008
© 2008 American Thoracic Society
DOI: 10.1165/rcmb.2007-0278OC

Primary Nasal Epithelium Exposed to House Dust Mite Extract Shows Activated Expression in Allergic Individuals

Aram B. Vroling1, Martijs J. Jonker2, Silvia Luiten1, Timo M. Breit2, Wytske J. Fokkens1 and Cornelis M. van Drunen1

1 Department of Otorhinolaryngology, Academic Medical Center, Amsterdam; and 2 Integrative Bioinformatics Unit, Swammerdam Institute for Life Sciences, Faculty of Science, University of Amsterdam, Amsterdam, The Netherlands

Correspondence and requests for reprints should be addressed to Aram B. Vroling, MSc., Room L3-106, Meibergdreef 9, 1100 DD Amsterdam, The Netherlands. E-mail a.b.vroling{at}amc.uva.nl


   Abstract
Top
 Abstract
CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Nasal epithelial cells form the outermost protective layer against environmental factors. However, this defense is not just physical; it has been shown that epithelial cells respond by the production of inflammatory mediators that may affect local immune responses. In this research we set out to characterize potential differences between the responses of nasal epithelium from healthy and allergic individuals to house dust mite (HDM) allergen. These differences will help us to define local mechanisms that could contribute to allergic disease expression. Epithelial cells were cultured from nasal biopsies taken from five healthy and five allergic individuals. These cultures were exposed for 24 hours to culture medium containing HDM allergen, or to culture medium alone. Isolated RNA was used for microarray analysis. Gene-ontology of the response in healthy epithelium revealed mainly up-regulation of chemokines, growth factors, and structural proteins. Moreover, we saw increased expression of two transcription factors (NF-{kappa}B and AP-1) and their regulatory members. The expression pattern of epithelium from allergic individuals in the absence of the HDM stimulus suggests that it is already in an activated state. Most striking is that, while the already activated NF-{kappa}B regulatory pathway remained unchanged in allergic epithelium, the AP-1 pathway is down-regulated upon exposure to HDM allergen; this is contrary to what we see in healthy epithelium. Clear differences in the expression pattern exist between epithelial cells isolated from healthy and allergic individuals at baseline and between their responses to allergen exposure; these differences may contribute to the inflammatory response.

Key Words: allergy • epithelial response • innate immunity • local tissue response



   CLINICAL RELEVANCE
Top
 Abstract
 CLINICAL RELEVANCE
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Response of epithelial cells to allergen has not before been studied on a genome expression level. Characterization of the responses will give new insight to the pathophysiology, and open roads to new treatments.

 
The mucosal layer in the nose is constantly exposed to viruses, bacteria, parasites, and harmless allergens. It is crucial that a correct immune response is initiated to all these environmental factors. When harmless allergens are mistaken for dangerous pathogens the immune system will mount an unwanted inflammatory response to the allergens, resulting in allergic inflammation. An important player in the initiation of immune responses is the antigen-presenting cell that resides in the mucosal tissue; the dendritic cell (DC). In recent years it has become increasingly clear that the peripheral DC initiates the immune response within an active local tissue environment, and that epithelial cells can play a role in this initiation process. Epithelial cells are more than a physical barrier and are themselves able to detect and respond to environmental signals. Epithelial cells can produce mediators that affect recruitment of immunocompetent cells to the local tissue and help create a micro-environment in which these cells function (1, 2).

In relation to the initiation of the immune response, it is very interesting that epithelial cells produce mediators that can influence DCs. An example of such a mediator is CCL20/MIP3{alpha}, the chemokine for CCR6-positive Langerhans cells (LCs), which is produced by bronchial epithelial cells after a variety of stimuli (3, 4). Not only recruitment is affected by epithelial mediators, but epithelial expressed granulocyte macrophage–colony-stimulating factor and transforming growth factor-β guide the differentiation of myeloid DCs and LCs, respectively, from their precursors (5, 6). Activation of tissue resident LCs is partly dependent on locally produced TNF-{alpha} and IL-1 (7, 8), and recently thymic stromal lymphopoietin produced by epithelial cells was shown to be important in the activation of DC-mediated allergic inflammation (9–11). Currently, there are just a few players for which the effect on DC function has been documented. Even less is known about potential differences in mediators produced by epithelial cells from healthy or allergic individuals, or whether any of these differences contribute to the expression or development of allergic disease.

In previous experiments we have investigated the epithelial response of a bronchial epithelial cell line, H292, to house dust mite (HDM) allergen. There we could show that epithelial cells display a broad and diverse expression of genes in response to exposure to allergen, and that a substantial number of these regulated genes have a function in cell communication based on their gene ontology classification. Moreover, we identified a potential regulatory network centered around TNF-{alpha} and NF-{kappa}B (12).

In this research we wanted to expand on these observations and investigate the response induced by HDM allergen in primary epithelium from healthy individuals and from individuals allergic to HDM. To this end, cultures of primary epithelial cells obtained from nasal biopsies were exposed to HDM extract diluted in saline, or to saline solution alone. RNA from this experiment was used for microarray analysis, and the resulting expression profiles were subjected to bioinformatical, biostatistical, and interaction network analysis. Characterization of potential differences in the expression pattern at baseline or in response to allergen exposure will provide valuable insight into the role of the epithelium in the allergic response. Identification of the mechanism by which nasal epithelium influences the allergic response can lead to development of new therapies that target the epithelial cells.

The complete dataset from the time course is accessible through the NCBI Gene Expression Omnibus repository (GEO; http://www.ncbi.nlm.nih.gov/geo/), series accession number GSE9150.


   MATERIALS AND METHODS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Patient Characteristics
This study was reviewed and approved by the medical ethical committee of the Amsterdam Medical Center, and all participants read and signed an informed consent. Five allergic volunteers (aged 19–55 years) and five healthy nonsmoking volunteers (aged 21–32 years) were selected based on skin prick test for HDM and other common allergens, and on a nasal allergen provocation to asses their response. Only monotypically HDM-allergic and nonallergic volunteers were included. Allergic individuals had refrained from using any medication for their allergy in the 4 weeks before the visit when biopsies were taken. Biopsies were taken from the lower edge of the inferior turbinate, 1 and 2 cm from the anterior end, using Fokkens' forceps with a cup diameter of 2.5 mm. Local anesthesia was achieved by application of adrenalin and cocaine under the inferior turbinate without touching the biopsy site, for a period of 10 minutes.

Primary Epithelial Cell Culture
Primary cells were obtained by digesting nasal biopsies of volunteers with 0.5 mg/ml collagenase 4 (Worthington Biochemical Corp., Lakewood, NJ) for 1 hour in Hanks' balanced salt solution (HBSS; Sigma-Aldrich, Zwijndrecht, The Netherlands). Subsequently cells were washed with HBSS and resuspended in bovine epithelial growth medium (Lonza Clonetics, Breda, The Netherlands) and seeded in two wells of a 6-well plate. Culture medium was replaced every other day. Cells were grown in fully humidified air containing 5% CO2 at 37°C.

HDM Extract and Exposure Experiment
HDM extract containing biologically nonrelevant trace amounts of LPS was kindly provided by Prof. Dr. M. L. Kapsenberg (AMC, The Netherlands) as a lyophilized powder. It was dissolved in PBS, then dialyzed against PBS and diluted to a stock concentration of 8 µg/µl. Cells were cultured for 2 weeks to 80% confluence and were subsequently pre-incubated with HBSS for 8 hours before exposure to HDM. Pre-incubation medium was removed and cells were then exposed to HBSS containing a previously (12) determined optimal concentration of house dust mite extract (2 µg/ml) or with HBSS alone (control condition) for 24 hours. Supernatants were removed and stored for further analysis; cells were used for RNA extraction.

RNA Extraction
Total RNA from each sample was extracted using Trizol (Life Technologies, Inc., Gaitersburg, MD) according to manufacturer's protocol, followed by purification by nucleospin RNA II (Machery-Nagel, Düren, Germany). The RNA concentration was measured on the nanodrop ND-1000 (NanoDrop Technologies inc., Wilmington, DE) and RNA quality was checked on the Agilent 2100 bio-analyzer (Agilent Technologies, Palo Alto, CA).

Microarray Affymetrix u133 Plus 2
Human Genome U133 Plus 2.0 Genechip Array (Affymetrix inc., Santa Clara, CA) representing 47,000 transcripts, including 38,500 well-characterized genes, was used in the analysis of HDM-induced genes. Technical handling of microarray experiments was performed at the MicroArray Department (MAD) of the University of Amsterdam (Amsterdam, The Netherlands), a fully licensed microarray technology center for Affymetrix Genechip platforms and official Dutch Affymetrix Service Provider. In short, biotin-labeled cRNA samples were prepared as described in the Affymetrix expression analysis technical manual (Affymetrix) using 3 µg of purified total RNA as template for the reaction. For this the One-Cycle cDNA Synthesis Kit (Affymetrix) was used. The Array images were acquired using a GeneChip Scanner 3,000 (Affymetrix) and analyzed with Affymetrix GeneChip Operating Software (Affymetrix).

Microarray Data Analysis and Statistics
The quality of the images was checked by visual inspection, and all raw data passed the quality criteria for average background, scale factors, percentage present calls, 3'/5' ratios GAPDH, 3'/5' ratios β-actin, hybridization spike-in controls, and poly-A spike-in controls. The data also passed a set of quality control checks provided by the affy, affyPLM, and affyQCreport packages from Bioconductor (http://www.bioconductor.org/). Expression values were calculated using the robust multi-array average (RMA) algorithm (13), and statistically analyzed for differential gene expression using ANOVA (MAANOVA package, version 0.98.8 [14]). The permutation based Fs test was used for hypothesis testing (15), and all P values were adjusted for false discovery rate correction (16). To quantify the effect of HDM extract on gene expression, pairwise statistical tests were performed to analyze: (1) the effect of HDM on epithelial cells from healthy individuals separately, (2) the effect of HDM on epithelial cells from allergic individuals separately, and (3) the difference in response to HDM in allergic compared with healthy individuals.

Ontology, Cluster, and Network Analysis
Gene ontology was done using the online gene ontology program GOstat (http://gostat.wehi.edu.au/), with which we used curated datasets and subsets to investigate the overrepresentation of geneontology groups (17). Cluster analysis on all significantly affected genes was done by transforming the means of the expression values for a gene in the four groups (healthy or allergic after control exposure or after HDM exposure) to Z-scores and using unsupervised K-means clustering based on correlation; for this we used Spotfire DecisionSite Functional Genomics. Network analysis was performed using Pathway Architect software (Stratagene, La Jolla, CA); here we used the curated dataset to build a regulation interaction network. We overlaid the colors used in our K-means cluster analysis to clarify the relative expression of the genes in our network.

Quantitative PCR
Quantitative PCR was used to validate the differential expression of selected genes. Isolated RNA from control-treated and HDM-treated cells was used to synthesize cDNA using the MBI Fermentas first-strand cDNA synthesis kit (Fermentas GmbH, St. Leon-Rot, Germany). PCR was performed on Bio-Rad iCycler (Bio-Rad, Veenendaal, The Netherlands). mRNA-specific TaqMan gene expression assays for ACTB, ATF3, CCL20, CNFN, EGR1, FLG, GAPDH, GATA3, GCSF, GROA, IL1RA, IL12B, IL13-R{alpha}2, IL8, KRT4, MCP1, MIP1B, PLAUR, SPINK7, TNFA, TNFR1, and TNF-IP3 were ordered form from Applied Biosystems (Nieuwerkerk a/d IJssel, The Netherlands). We performed all PCR assays three times, on all samples. Fold changes were calculated using the {delta}{delta}ct method.


   RESULTS
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Validation of the Microarray Results
After incubation with HDM extract, we identified 555 probe sets that were statistically differentially expressed in primary epithelium from healthy individuals and 301 probe sets that were statistically differentially expressed in primary epithelium from allergic patients. This collection of probe sets was first curated by, when possible, discarding the less specific x_at probe sets and the splice variant specific s_at probe sets. For analysis we further selected only those genes for which the expression level of their probe sets change by more than 1.5-fold (see Table E1 in the online supplement). In healthy epithelium, the original 555 probe sets correspond to 209 uniquely annotated genes and 19 unannotated and/or hypothetical sequences in our curated dataset. Of these genes, 206 were up-regulated and 22 were down-regulated. In allergic epithelium the original 301 probe sets correspond to 87 uniquely annotated genes and 12 unannotated and/or hypothetical sequences in our curated dataset. Here 62 genes showed increased and 37 decreased expression.

The results of this microarray experiment were validated by independent real-time PCR on the same starting material used for the microarray analysis. A selection of 20 genes that had revealed an increased, decreased, or unchanged expression level in either the healthy or allergic curated gene set was used for this validation experiment (see MATERIALS AND METHODS). After normalization for three household genes (GAPDH, β-actin, and β2-microglobulin), the relative change in expression of these genes measured by PCR was directly comparable to the relative change obtained from our microarray experiment, both for the healthy epithelium (r = 0.621, P = 0.006) and for the allergic epithelium (r = 0.735, P < 0.001) (data are shown in Table E2).

Global Analysis Reveals an Activated State in Primary Allergic Epithelium
Within our curated dataset, the expression of only a few genes (9/311) shows identical statistically significant changes upon HDM exposure in healthy and allergic epithelium. The expression of the other genes either change in healthy or in allergic epithelium, and in many cases shows a differential response in both groups (Figure 1). The absence of an overlap in the response could be due to a differential change in the expression profile by the HDM extract, to intrinsic differences between healthy and allergic epithelium at baseline, or a combination of both. To investigate the relative contribution of these factors, we first used principal components analysis (PCA) on the whole data collection of micro-array chips. Most of the variation observed for the chips (57%, Figure 2) lies in the difference between healthy and allergic epithelium (PCA1). What was unexpected was that the next largest contribution (12%) comes from the variability of the individuals within the healthy epithelium group (PCA2), with only a very limited variation in the allergic epithelium group. Exposure to HDM extract only has the third largest contribution (7%) to the observed variation between all the genes in the healthy and allergic groups.


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		Figure 1. Gene expressions in different groups. This scatter plot shows the fold changes for the genes in our curated dataset as calculated in the healthy control group (x axis), and in the allergic group (y axis).

 

Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		Figure 2. Principal components analysis. (Left panel) The relative contribution to variance for the first 10 principal components. (Right panel) A scatter plot of all 20 arrays and their individual relative contribution to principal component 1 (PC1) and principal component 2 (PC2). Array IDs are indicated in the graph.

 
The PCA showed that not only differences in response to allergen, but also differences at baseline between healthy and allergic epithelium contribute to variation; therefore, we used K-means clustering on the expression levels in the four groups. This allows us to compare groups of genes that share specific expression patterns. The 311 genes that change their expression in healthy and/or allergic epithelium can be effectively separated into 12 clusters (Figure 3). Baseline expression of a given gene in 5 out of 12 clusters is substantially higher in allergic epithelium than in healthy epithelium. Together these clusters (numbers 7, 9, 10, 11, and 12) represent 74% of all regulated genes. The K-mean cluster analysis shows that for some of these clusters expression after HDM exposure is high and remains largely unchanged in the allergic group, and is up-regulated to a certain degree in the healthy group. For cluster 9 (8 genes) the expression level in healthy epithelium after HDM exposure is even higher than in the allergic group, in cluster 11 (46 genes) the expression level is similar, while in cluster 12 (129 genes) the expression level in the healthy group does not quite reach the expression level of the allergic group. Evidently, a substantial group of genes in the epithelium from allergic individuals already displays an activated state at baseline. Interestingly, some clusters (numbers 3, 4, 7, and 10) show changes in the allergic group, whereas the expression in the healthy group remains unchanged, while two genes (cluster 5) even show an opposite response to allergen exposure.


Figure 3
View larger version (38K):
[in this window]
[in a new window]

  		Figure 3. K-means clusters. Here all genes from curated dataset (see RESULTS) were transformed to Z-scores, and were organized in 12 clusters using K-means clustering. Table E1 gives all 311 genes and their respective cluster number. AB, allergic at baseline; AH, allergic after HDM exposure; CB, healthy control at baseline; CH, healthy control after HDM exposure.

 
The Activated State in Allergic Epithelium Is Reflected by Only a Few Ontology Classes
With only a limited number of expression profiles describing the effect of HDM exposure in healthy and allergic epithelium, we wondered whether these patterns could be linked to specific functions of these genes.

Ontology analysis for genes that change their expression due to exposure to HDM extract in the primary epithelium from healthy individuals shows that these genes are principally involved in cell-to-cell contact. Cell communication (GO:7154) is by far the biggest group (66 genes) that is significantly overrepresented in the curated healthy data set (P = 2.4 x 10–5). Related ontology groups are also overrepresented: signal transduction (GO:7165, P = 4.6 x 10–4), receptor binding (GO:5102, P = 2.8 x 10–5), extra-cellular space (GO:5615, P = 8.9 x 10–5), and transcription factor activity (GO:3700, P = 2.9 x 10–6). These changes seem to be a consequence of a general response of healthy epithelium to environmental factors as the ontology group response to external stimulus (GO:9605, P = 4.4 x 10–7) and the related groups response to wounding (GO:9611, P = 5.8 x 10–6), response to abiotic factors (GO:9628, P = 1 x 10–4), and response to stress (GO:6950, P = 2.4 x 10–5) are all significantly overrepresented in healthy gene set. Other processes that are affected are cell proliferation (GO:8283, P = 3.3 x 10–9), (negative regulation) of cell death (GO:43069 P = 3.5 x 10–4 and GO:8219 P = 1.9 x 10–4), and intracellular junctions (GO:5911, P = 3 x 10–4). A similar picture emerges when we combine the K-mean cluster analysis with the gene ontology class analysis. As clusters 9, 11, and 12 represent 58% of all genes that behave differently between healthy and allergic epithelium with expression in healthy epithelium going up by HDM exposure and expression remaining high in allergic epithelium, the activated status of allergic epithelium is best described by genes involved in cell communication and cell proliferation.

Although HDM extract also affects the expression pattern in the primary epithelium of allergic individuals, these genes do not seem to represent a specific gene ontology class. The only class that is significantly overrepresented is protease inhibitor activity (GO:0030414, P = 8.9 x 10–3), but of this group only a few genes are statistically altered in allergic epithelium (SPINK5, SPINK6, SPINK7, SERPINB3, C3, and WFDC5).

Involvement of the NF-{kappa}B and AP-1 Transcription Factor Complexes in the Differential Gene Expression of Healthy and Allergic Epithelium
Now we found cell communication and cell proliferation activated in epithelial cells of allergic individuals, we wanted to find regulators that could be responsible for this activated state. To find these regulators, we used regulation network analysis. In the network (Figure 4) we saw the expected players involved in cell communication: inflammatory markers (IL-1{alpha}, IL-1β, and TNF-{alpha}), chemokines (CXCL8/IL-8, CXCL1/GRO-{alpha}, CXCL2/GRO-β, and CXCL10/IP-10), growth factors (EREG, AREG, and HBEGF), and receptors (TLR3, PLAUR, PTGER4, and IL-7R). More interestingly, this regulation seems to be mediated by proteins from the transcription factor complexes NF-{kappa}B (NFKB1, NFKB2, NFKBIA, and NFKBIZ), and AP-1 (FOS, JUN, JUNB, FOSL1, and ATF3). When we use a color code to identify the different gene clusters from Figure 3 in the regulation network analysis we see that these transcription factor complexes also share a similar expression pattern. The activated state we identified above is also reflected at the transcription factor level, with the NF-{kappa}B and the noncanonical AP-1 transcription factors (JUNB and FOSL1) mapping to cluster 11. Interestingly, the canonical AP-1 transcription factors (FOS and JUN) behave differently. FOS and ATF3 are expressed at similar levels at baseline in healthy and allergic epithelium (cluster 8), and expression goes up in healthy epithelium, but goes down in allergic epithelium. For JUN the baseline expression is higher in allergic epithelium (cluster 9), but again expression in allergic epithelium is down-regulated and is up-regulated in healthy epithelium upon allergen exposure.


Figure 4
View larger version (90K):
[in this window]
[in a new window]

  		Figure 4. Regulation interaction network. Here genes within our dataset that are known to have a regulatory effect on each other are presented in their cellular location, with arrows indicating the interactions. Genes are colored by the K-means cluster to which they were assigned.

 

   DISCUSSION
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The first important observation coming from our data is that for a selective group of genes the allergic epithelium already is in an activated state. Baseline expression levels for these genes are higher in allergic individuals than corresponding levels in healthy individuals. Moreover, for these genes, allergen exposure in healthy individuals leads to an increase in their expression levels, while the levels in allergic individuals remain largely unchanged. In our analysis of which processes are affected by the allergic status, we find the genes belonging to the ontology classis cell communication and proliferation. In cell communication we find chemokines (IL-8, GRO-{alpha}, GRO-β, and IP-10) and cytokines (IL1-{alpha}, IL1β, IL1{varepsilon}, and TNF-{alpha}) which have a known and well-documented functions in the recruitment and activation of cells of the immune system. In cell communication we also find genes for growth factors (CTGF, HBEGF, AREG, EREG, and FGF5) and intercellular junction (TJP-1, -2, CLDN-1, -4, and OCLN). The latter groups are known to be involved in the repair of damaged epithelium that, in this case, could be the consequence of cleavage of intracellular junction proteins by the protease activity contained within the HDM allergen extract. Where a substantial number of the regulated genes in epithelium from healthy individuals can be categorized in well-defined gene ontology classes, this does not hold true for the regulated genes in allergic epithelium. Only a single class of protease inhibitors is significantly enriched. Most likely this reflects a protective mechanism in allergic epithelium to counteract the activity of the protease activities contained within the allergen extract or those released by tissue resident mast cells (tryptase, chymase). Although relatively little attention has been given to the role of proteases in health and disease, it is important to note that mutations in one of the protease inhibitors SPINK5, as well as the protease ADAM33, have been associated to asthma (18, 19). Other proteases like the matrix metalloproteinases (MMP) have a crucial role in the local tissue remodeling associated with asthma (20, 21).

In this research we set out to characterize the response of epithelial cells; to do so we chose to culture cells taken from a biopsy for 2 weeks, after which the cell cultures contain only epithelial cells, thus eliminating any contamination of gene expression by other cell types. One of our concerns was that culturing the cells would alter the baseline expression, or the ability to respond, in a way that would prevent us from detecting differences between healthy and allergic individuals. Given that we detect so many well-established mediators that have been confirmed both in vitro and in vivo gives us confidence that the differences in the expression profiles are likely to reflect to situation in situ. Another deliberate choice in our model was to stimulate the cells for 24 hours and then investigate the mRNA profile. We wanted to mimic the continuous exposure to which the nasal epithelium is subject in vivo. As a consequence we were able to determine the relevant expression patterns of the NF-{kappa}B and AP-1 transcription factor families. However, our data do not reveal how this state is achieved after allergen exposure. For this purpose, earlier time points after induction should be investigated, and this could well have consequence for the overall outcome of our experiment, with other genes and/or regulatory pathways playing a more prominent role.

There is little overlap in the response to HDM extract in primary epithelium from healthy and allergic individuals. As described above, a substantial group of genes that are up-regulated in healthy individuals are already expressed at high levels in allergic epithelium even without exposure to allergen. To expand on this observation it would seem that the induced expression of these "activated" genes is part of the normal response in healthy primary epithelium. How is this activated status in allergic epithelium induced or maintained? Both healthy and allergic individuals are constantly exposed to HDM allergen in every day life. Epithelial cells of healthy individuals do express PAR2 (protease activated receptor 2), though at lower levels than in allergic individuals, so in vivo activation of epithelial cells by the proteases contained within HDM extract before isolation could occur in both (22). However, in vitro experiments with the epithelial cell line H292 have shown that the effect of PAR2 activation normally is transient, with expression levels of the responding genes returning to baseline levels 72 hours after the initial exposure to HDM extract. Therefore it does not seem likely that the activated state we see at baseline can be directly maintained during the 2 weeks of culturing that precedes our HDM induction experiment (not without a positive feedback loop).

In previous work we have suggested that TNF-{alpha}, induced by HDM exposure, could be a central player both in mediating the HDM effect and in restimulating epithelial cells. No matter whether we assume a direct effect of HDM or an indirect effect through an HDM-induced feedback loop, the outcome is different for healthy or allergic epithelium. This would lead us to conclude that the allergic status could be aggravated in allergic individuals because of an inability to shut this response down. This would explain the high level of correspondence between genes activated in healthy epithelium and those already activated in allergic epithelium. Alternatively, the "activated" status could be not related to the in vivo action of HDM on epithelium directly, but to the indirect effect that allergen-induced mediators from other tissue resident cells like mast cells have on the epithelium. However, it is not entirely clear how this would explain the correspondence between the HDM-induced genes in our in vitro experiment and the "activated" state in allergic epithelium.

When we assume a differential regulation of the HDM response in epithelium of healthy and allergic individuals, this seems also reflected in the response of two important transcription factor complexes. The transcription factor NF-{kappa}B is a family of (hetero)dimeric proteins that has been linked to inflammation in many cell types, and is known to regulate (and to be regulated by) many different mediators (23, 24). That we find genes belonging to the NF-{kappa}B family up-regulated in response to allergen in epithelium from healthy individuals is perhaps not surprising, but given the complex interactions between the family members it is hard to firmly establish NF-{kappa}B as the responsible factor for the observed induction of genes (25). The NF-{kappa}B proteins NFKB1/p105 and NFKB2/p100 are the cytoplasmic precursors of, respectively, the nuclear factors p50 and p52 (26, 27). Whereas homodimers of p50 or p52 are linked to transcriptional repression, the formation of heterodimers with the NF-{kappa}B Rel-family members is linked to transcriptional activation (28, 29). In our microarray experiments RELB is expressed at a considerably higher level in allergic compared with healthy epithelium and the expression levels are unaffected by allergen exposure. Interestingly, it has been described that knockout mice for NFKB1 show a reduced stress response with lower IL-6 and COX-2/PTGS2 levels, suggesting that at least in the overall outcome NFKB1 is required for a correct response to environmental signals (30). Given that the expression of these genes remains high when allergic epithelial cells are cultured for 2 weeks in the absence of HDM suggests that NF-{kappa}B in allergic individuals could be (partly) responsible for maintaining the "activated" state. Although these observations could well explain the "activated" state in allergic epithelium or the induction of genes in healthy epithelium, it fails to account for how this "activated" state is maintained or why this does not occur in epithelium from healthy individuals. Further experiments should investigate whether epigenetic modification of key regulators differs between healthy and allergic epithelium, or the presence of an activating auto-feedback loop in allergic epithelium could be responsible for the maintenance of the allergic phenotype in tissue culture. Also, one might wonder whether the observed effect can be observed for other allergens and whether allergen-specific PAMP receptors could be involved in the response of epithelial cells to allergens.

What is interesting is that epithelial cells also show differential expression for some of the AP-1 family members. In these family members it is strange that the noncanonical members FOSL1 and JUND show the same pattern as NF-{kappa}B, whereas the FOS, and JUN, the genes for c-FOS and c-JUN protein, do show an increase in healthy individuals, but are actually down-regulated in allergic individuals. If this expression could be linked to a protective mechanism against the effects of NF-{kappa}B activation, then this could explain some of the differences we see between healthy and allergic individuals. It has been described that in Drosophila AP-1 is required to down-regulate NF-{kappa}B target genes by interfering with promoter binding (31). If this mechanism also applies in humans, then the up-regulation of FOS/JUN in healthy epithelium may contribute to the down-regulation of NF-{kappa}B–regulated genes, and the absence of this response in allergic epithelium may contribute to the maintenance of the activated state seen in allergic epithelium. Interestingly, the FOS-related AP-1 family member ATF3 displays an expression profile similar to that of the FOS gene itself. Identified as a stress factor, this bZIP protein in its homodimeric form acts as transcriptional repressor, but when it heterodimerizes with members of the JUN family it acts as a transcriptional activator (32).

Airway epithelium is becoming more widely accepted as an active player in the response to allergens; however, it was unknown if the response to allergen exposure would differ between allergic and healthy individuals. We have shown that not only the response differs, but also that the expression at baseline is different between healthy and allergic individuals. Linking our transcriptional observations to functional consequences is hard. Foremost, it is not clear how differences in the transcription level translate into effects on the protein level. This is further compounded by the formation of different homo- and hetero-dimers that each can have their distinct effects, by the functional interaction of the NF-{kappa}B and AP-1 transcription factors at promoter sites, and even by the formation of hetero-dimers between NF-{kappa}B and AP-1 family members. Despite all these considerations, the allergic epithelium seems an important target in the treatment, and possibly prevention, of allergic disease. Reducing the "activated" state in allergic epithelium may have direct consequence for the expression of allergic complaints, for instance by reducing the influx of effecter cells, and understanding how and why the response in allergic epithelium differs from that in healthy epithelium may help in preventing the initiation of the allergic response.


   Footnotes
 
This research has been funded by the Department of Otorhinolaryngology, Academic Medical Center, Amsterdam, The Netherlands.

This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org.

Originally Published in Press as DOI: 10.1165/rcmb.2007-0278OC on September 27, 2007

Conflict of Interest Statement: A.B.V. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. M.J.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. S.L. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. T.M.B. is head of an academic microarray core facility. He also acts as service provider for for-profit parties, and is involved in two government-funded projects that have about 25% matching funding from various companies such as Organon, Eli Lilly, GSM, Biodetection Systems, and Notox. W.J.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. C.M.v.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form July 23, 2007

Accepted in final form August 26, 2007


   References
Top
 Abstract
 CLINICAL RELEVANCE
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Asokananthan N, Graham PT, Stewart DJ, Bakker AJ, Eidne KA, Thompson PJ, Stewart GA. House dust mite allergens induce proinflammatory cytokines from respiratory epithelial cells: the cysteine protease allergen, Der p 1, activates protease-activated receptor (PAR)-2 and inactivates PAR-1. J Immunol 2002;169:4572.[Abstract/Free Full Text]
  2. King C, Brennan S, Thompson PJ, Stewart GA. Dust mite proteolytic allergens induce cytokine release from cultured airway epithelium. J Immunol 1998;161:3645–3651.[Abstract/Free Full Text]
  3. Reibman J, Hsu Y, Chen LC, Bleck B, Gordon T. Airway epithelial cells release MIP-3alpha/CCL20 in response to cytokines and ambient particulate matter. Am J Respir Cell Mol Biol 2003;28:648–654.[Abstract/Free Full Text]
  4. Pichavant M, Charbonnier AS, Taront S, Brichet A, Wallaert B, Pestel J, Tonnel AB, Gosset P. Asthmatic bronchial epithelium activated by the proteolytic allergen Der p 1 increases selective dendritic cell recruitment. J Allergy Clin Immunol 2005;115:771–778.[CrossRef][Medline]
  5. Riedl E, Stockl J, Majdic O, Scheinecker C, Rappersberger K, Knapp W, Strobl H. Functional involvement of E-cadherin in TGF-Beta 1-induced cell cluster formation of in vitro developing human Langerhans-type dendritic cells. J Immunol 2000;165:1381–1386.[Abstract/Free Full Text]
  6. Pieri L, Domenici L, Romagnoli P. Langerhans cells differentiation: a three-act play. Ital J Anat Embryol 2001;106:47–69.[Medline]
  7. Cumberbatch M, Dearman RJ, Kimber I. Langerhans cells require signals from both tumour necrosis factor-alpha and interleukin-1 beta for migration. Immunology 1997;92:388–395.[CrossRef][Medline]
  8. Chang CH, Furue M, Tamaki K. B7–1 expression of Langerhans cells is up-regulated by proinflammatory cytokines, and is down-regulated by interferon-gamma or by interleukin-10. Eur J Immunol 1995;25:394–398.[Medline]
  9. Soumelis V, Reche PA, Kanzler H, Yuan W, Edward G, Homey B, Gilliet M, Ho S, Antonenko S, Lauerma A, et al. Human epithelial cells trigger dendritic cell mediated allergic inflammation by producing TSLP. Nat Immunol 2002;3:673–680.[Medline]
  10. Gilliet M, Soumelis V, Watanabe N, Hanabuchi S, Antonenko S, Waal-Malefyt R, Liu YJ. Human dendritic cells activated by TSLP and CD40L induce proallergic cytotoxic T cells. J Exp Med 2003;197:1059–1063.[Abstract/Free Full Text]
  11. Ying S, O'Connor B, Ratoff J, Meng Q, Mallett K, Cousins D, Robinson D, Zhang G, Zhao J, Lee TH, et al. Thymic stromal lymphopoietin expression is increased in asthmatic airways and correlates with expression of Th2-attracting chemokines and disease severity. J Immunol 2005;174:8183–8190.[Abstract/Free Full Text]
  12. Vroling AB, Duinsbergen D, Fokkens WJ, van Drunen CM. Allergen induced gene expression of airway epithelial cells shows a possible role for TNF-alpha. Allergy 2007.
  13. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, Speed TP. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 2003;4:249–264.[Abstract]
  14. Cui X, Churchill GA. Statistical tests for differential expression in CDNA microarray experiments. Genome Biol 2003;4:210.[CrossRef][Medline]
  15. Cui X, Hwang JTG, Qiu J, Blades NJ, Churchill GA. Improved statistical tests for differential gene expression by shrinking variance components estimates. Biostatistics 2005;6:59–75.[Abstract]
  16. Hochberg Y, Benjamini Y. More powerful procedures for multiple significance testing. Stat Med 1990;9:811–818.[Medline]
  17. Beissbarth T, Speed TP. GOstat: find statistically overrepresented gene ontologies within a group of genes. Bioinformatics 2004;20:1464–1465.[Abstract/Free Full Text]
  18. Moffatt MF. SPINK5: a gene for atopic dermatitis and asthma. Clin Exp Allergy 2004;34:325–327.[CrossRef][Medline]
  19. Shapiro SD, Owen CA. ADAM-33 surfaces as an asthma gene. N Engl J Med 2002;347:936–938.[Free Full Text]
  20. Holgate ST, Lackie PM, Davies DE, Roche WR, Walls AF. The bronchial epithelium as a key regulator of airway inflammation and remodelling in asthma. Clin Exp Allergy 1999;29:90–95.[CrossRef][Medline]
  21. Kumagai K, Ohno I, Okada S, Ohkawara Y, Suzuki K, Shinya T, Nagase H, Iwata K, Shirato K. Inhibition of matrix metalloproteinases prevents allergen-induced airway inflammation in a murine model of asthma. J Immunol 1999;162:4212–4219.[Abstract/Free Full Text]
  22. Dinh QT, Cryer A, Trevisani M, Dinh S, Wu S, Cifuentes LB, Feleszko WK, Williams A, Geppetti P, Fan Chung K, et al. Gene and protein expression of protease-activated receptor 2 in structural and inflammatory cells in the nasal mucosa in seasonal allergic rhinitis. Clin Exp Allergy 2006;36:1039–1048.[CrossRef][Medline]
  23. Cheng DS, Han W, Chen SM, Sherrill TP, Chont M, Park GY, Sheller JR, Polosukhin VV, Christman JW, Yull FE, et al. Airway epithelium controls lung inflammation and injury through the NF-{Kappa}B pathway. J Immunol 2007;178:6504–6513.[Abstract/Free Full Text]
  24. Lee HC, Ziegler SF. Inducible expression of the proallergic cytokine thymic stromal lymphopoietin in airway epithelial cells is controlled by NF{Kappa}B. Proc Natl Acad Sci USA 2007;104:914–919.[Abstract/Free Full Text]
  25. Macfarlane SR, Sloss CM, Cameron P, Kanke T, McKenzie RC, Plevin R. The role of intracellular Ca2+ in the regulation of proteinase-activated receptor-2 mediated nuclear factor kappa B signalling in keratinocytes. Br J Pharmacol 2005;145:535–544.[CrossRef][Medline]
  26. Heron E, Deloukas P, van Loon AP. The complete exon-intron structure of the 156-kb human gene NFKB1, which encodes the p105 and p50 proteins of transcription factors NF-Kappa B and I Kappa B-Gamma: implications for NF-Kappa B-mediated signal transduction. Genomics 1995;30:493–505.[CrossRef][Medline]
  27. Heusch M, Lin L, Geleziunas R, Greene WC. The generation of Nfkb2 P52: mechanism and efficiency. Oncogene 1999;18:6201–6208.[CrossRef][Medline]
  28. Lin R, Gewert D, Hiscott J. Differential transcriptional activation in vitro by NF-KappaB/Rel proteins. J Biol Chem 1995;270:3123–3131.[Abstract/Free Full Text]
  29. Driessler F, Venstrom K, Sabat R, Asadullah K, Schottelius AJ. Molecular mechanisms of interleukin-10-mediated inhibition of NF-KappaB activity: a role for P50. Clin Exp Immunol 2004;135:64–73.[CrossRef][Medline]
  30. Waterfield MR, Zhang M, Norman LP, Sun SC. NF-KappaB1/P105 regulates lipopolysaccharide-stimulated MAP kinase signaling by governing the stability and function of the Tpl2 kinase. Mol Cell 2003;11:685–694.[CrossRef][Medline]
  31. Kim T, Yoon J, Cho H, Lee WB, Kim J, Song YH, Kim SN, Yoon JH, Kim-Ha J, Kim YJ. Downregulation of lipopolysaccharide response in drosophila by negative crosstalk between the AP1 and NF-KappaB signaling modules. Nat Immunol 2005;6:211–218.[CrossRef][Medline]
  32. Hai T, Wolfgang CD, Marsee DK, Allen AE, Sivaprasad U. ATF3 and stress responses. Gene Expr 1999;7:321–335.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
2007-0278OCv1
38/3/293    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vroling, A. B.
Right arrow Articles by van Drunen, C. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vroling, A. B.
Right arrow Articles by van Drunen, C. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Crit. Care Med.
Copyright © 2008 American Thoracic Society.