This Article Abstract Full Text (PDF) Online Supplement All Versions of this Article: 2006-0466OCv1 37/2/169    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ross, A. J. Articles by Devlin, R. B. Search for Related Content PubMed PubMed Citation Articles by Ross, A. J. Articles by Devlin, R. B.

Transcriptional Profiling of Mucociliary Differentiation in Human Airway Epithelial Cells

Clinical Research Branch, Human Studies Division, National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park; and Center for Environmental Medicine, Asthma, and Lung Biology, University of North Carolina, Chapel Hill, North Carolina

Correspondence and requests for reprints should be addressed to Robert B. Devlin, U.S. Environmental Protection Agency, Clinical Research Branch, Human Studies Division, 104 Mason Farm Road, Chapel Hill, NC 27599–7315. E-mail: devlin.robert{at}epa.gov

	Abstract

Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References

 
When cultured at an air–liquid interface (ALI) in the appropriate medium, primary human airway epithelial cells form a polarized, pseudostratified epithelium composed of ciliated and mucus-secreting cells. This culture system provides a useful tool for the in vitro study of airway epithelial biology and differentiation. We have performed microarray analysis on ALI cultures of human bronchial epithelial cells (HBECs) grown over a 28-d period to identify genes involved in mucociliary differentiation. We identified over 2,000 genes that displayed statistically significant 2-fold or greater changes in expression during the time course. Of the genes showing the largest increases, many are involved in processes associated with airway epithelial biology, such as cell adhesion, immunity, transport, and cilia formation; however, many novel genes were also identified. We compared our results with data from proteomic analyses of the ciliary axoneme and identified candidate genes that may have roles in cilia formation or function. Gene networks were generated using Ingenuity Pathways Analysis (Ingenuity Systems, Redwood City, CA) to identify signaling pathways involved in mucociliary cell differentiation or function. Networks containing genes involved in TGF-, WNT/-catenin, and epidermal growth factor receptor (EGFR) pathways were identified, suggesting potential roles for these families in airway epithelia. Microarray results were validated by real-time RT-PCR for a number of representative genes. This work has provided extensive information about gene expression changes during differentiation of airway epithelial cells, and will be a useful resource for researchers interested in respiratory function, pathology, and toxicology.

Key Words: bronchial epithelium • differentiation • cilia • microarrays

CLINICAL RELEVANCE Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References   These microarray analyses give new insight into the genetics of airway epithelial cell differentiation and provide a catalog of gene expression that will be a valuable resource for scientists interested in airway biology, pathology, and toxicology.

 
The airway epithelium provides a critical interface between the body and the external environment and acts as a protective barrier against inhaled toxins, pathogens, and particles. The epithelium is a pseudostratified layer consisting of basal cells, secretory cells, and columnar ciliated cells. It exerts its protective effects through a number of distinct defense mechanisms: by providing a physical barrier; through the secretion of factors that mediate immunity, inflammation, and antioxidant defense; and through mucociliary clearance. Mucociliary clearance involves the movement of inhaled particles that are trapped by mucus through the coordinated beating of cilia. Defects in mucociliary clearance are associated with a wide range of respiratory disorders, including cystic fibrosis, asthma, and chronic obstructive pulmonary disease (COPD). Thus, it is important to understand the molecular mechanisms regulating mucociliary cell differentiation and function.

Multiple culture systems have been developed for the study of airway epithelial cells. Primary cells isolated from the airway can be grown in standard submersed culture, but these cultures lack many of the features of the in vivo airway epithelium. However, under the appropriate conditions, formation of a ciliated, mucus-producing, pseudocolumnar epithelium can be induced in primary cultures of nasal, tracheal, or bronchial epithelial cells from different mammalian species (1–3). These studies demonstrated that mucociliary differentiation of primary cells in culture requires a combination of retinoic acid (RA), a collagen gel substratum, and culture at an air–liquid interface (ALI). In the absence of RA, the cells form a stratified squamous epithelium, and mucin secretion is dramatically decreased (1, 2). The presence of collagen gel does have some effects on secretory cell differentiation, but appears to be most critical for ciliated cell differentiation (4). Finally, culture of the cells at an ALI is required for ciliated cell differentiation, as submersion of epithelial cells blocks ciliogenesis (5).

The ALI culture system provides a unique opportunity to examine the factors and pathways that regulate mucociliary differentiation. Previous studies have examined morphologic and molecular changes that occur during ALI culture of human bronchial epithelial cells (HBECs). We have expanded upon these studies by performing a global analysis of gene expression changes during mucociliary differentiation of HBECs. Through these microarray analyses, we have identified more than 2,000 genes that display statistically significant increases or decreases during this differentiation process. Using gene ontology databases, we have characterized major biological processes that occur during mucociliary differentiation. In addition, we have identified genes with candidate roles in cilia formation or function by comparing our results with those from proteomic studies of the mammalian axoneme. Finally, we have identified signaling pathways that may have roles in regulating airway epithelial differentiation and/or function. These studies provided valuable new insight into the genetic regulation of differentiation of the bronchial epithelium, and identified many novel genes with no previously known functions in this tissue.

	MATERIALS AND METHODS

Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References

 
Cell Culture
Primary human bronchial epithelial cells were obtained from three different healthy, nonsmoking adult donors. The protocol and consent form were approved by the University of North Carolina School of Medicine Committee on the Protection of the Rights of Human Subjects. Cells were obtained by cytological brushing at bronchoscopy and expanded to passage three in bronchial epithelial growth medium (BEGM; Clonetics, Cambrex Corp., East Rutherford, NJ). Cells were plated on vitrogen-coated filter supports inserted into 12-well culture plates and maintained as described previously (6). Five hundred nanomolars of RA were added to culture medium after cells reached 100% confluence to promote differentiation. ALI culture was initiated 48 h later by removing the apical medium, and basolateral media containing 100 nM RA was used for the remainder of the culture period. The cells were maintained in ALI culture for 28 or 29 d and the medium was changed every 48 h (Days 2, 4, 6, 8, etc.). Cells were harvested at 11 different time points during the culture period: Days 0, 1, 2, 4, 8, 10, 12, 14, 17, 21, and 28. Media was not changed immediately before harvesting; for cells harvested on Days 2, 4, 8, 10, 12, 14, and 28, the last media change occurred 48 h earlier. For cells harvested on Days 1, 17, and 21, the media was changed 24 h earlier. Day 0 refers to the time point just before the removal of media to initiate ALI culture. All time points were performed in triplicate (one sample from each of the three donor lines), except for Days 2, 12, and 17, which were performed in duplicate due to a reduced cell number obtained from one of the donor lines.

Histology
Cells were fixed for 20 min at room temperature in 4% paraformaldehyde, washed with PBS, then transferred to 70% ethanol for paraffin embedding. Samples were sectioned at 5 µm and deparaffinized in xylene, followed by incubating in a series of graded alcohols to rehydrate. For general histology, sections were stained using standard protocols for either hematoxylin and eosin (H&E) or Alcian blue/periodic acid/Schiff's (PAS) reagents. For immunostaining, sections were placed in a humid box, blocked with Power Block (BioGenex, San Ramon, CA) for 30 min at room temperature. Sections were incubated with mouse anti-Muc5AC monoclonal (Clone CLH2, 1 mg/ml stock; Chemicon, Temecula, CA) diluted 1:100 in Tris-buffered saline (TBS), overnight at 4°C. Sections were washed with TBS and incubated with Alexa 488 goat anti-mouse secondary antibodies (Molecular Probes, Invitrogen, Carlsbad, CA) for 1 h at room temperature. Sections were washed with TBS and blocked with Power Block for 30 min at room temperature. Sections were incubated with anti-acetylated -Tubulin (clone 6–11 B-1, 0.5 mg/ml stock; Zymed, San Francisco, CA), diluted 1:800 in TBS, for 1 h at 37°C, then washed with TBS. Sections were then incubated with Alexa 561 goat anti-mouse secondary antiboides (Molecular Probes) for 1 h at room temperature and mounted using Vectashield with DAPI (Vector Laboratories, Burlingame, CA). Images were captured using a Nikon C1 si confocal microscope (Nikon, Torrance, CA).

Microarrays
Total RNA was prepared for microarray analysis using the RNeasy kit (Qiagen, Valencia, CA), as per the manufacturer's protocol. Briefly, cells were lysed with 120 µl of RLT lysis buffer, with cells from three Transwells pooled for each time point. The lysate was frozen at –80°C until processed further. Lysates were spun through a QIAshredder spin column (Qiagen) at 10,000 rpm for 2 min to homogenize the samples. Lysates were added to RNeasy columns, and all incubation and washing steps were performed following the kit's instructions. The purified total RNA was stored at –80°C until all time points were collected. RNA was quantified spectrophotometrically and the quality of the RNA preps was confirmed using the Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA).

Ten micrograms of RNA was provided to Expression Analysis, Inc. (Durham, NC), where microarray data collection was performed using GeneChip Human Genome HU133 Plus 2.0 Arrays (Affymetrix, Santa Clara, CA). These arrays contain oligonucleotide probes for 47,000 transcripts and variants, including over 38,500 well-annotated genes. All probe sets represented on the GeneChip Human Genome U133 Set are identically replicated on the GeneChip Human Genome U133 Plus 2.0 Array. The sequences from which these probe sets were derived were selected from GenBank, dbEST, and RefSeq. The sequence clusters were created from the UniGene database (Build 133, April 20, 2001) and then refined by analysis and comparison with a number of other publicly available databases, including the Washington University EST trace repository and the University of California, Santa Cruz Golden-Path human genome database (April 2001 release). In addition, there are 9,921 new probe sets representing 6,500 genes, for which the sequences were selected from GenBank, dbEST, and RefSeq. These sequence clusters were created from the UniGene database (Build 159, January 25, 2003) and refined by analysis and comparison with a number of other publicly available databases, including the Washington University EST trace repository and the NCBI human genome assembly (Build 31).

Double-stranded cDNA was synthesized from total RNA using a primer that is comprised of an oligo(dT) segment and a T7 promoter sequence, using an RNase H deletion mutation of MMLV reverse transcriptase (Invitrogen). Biotin-labeled cRNA was synthesized using either the ENZO BioArray HighYield RNA Transcript Labeling Kit (Enzo Biochem, New York, NY; T7) or Affymetrix GeneChip IVT Labeling Kit. Fragmented cRNA was hybridized to the microarray according to the Affymetrix GeneChip Expression Analysis Technical Manual (November 2004). All subsequent post-hybridization steps were performed according to the GeneChip Technical Manual.

Data Analysis
Affymetrix .cel files were processed at Expression Analysis using the REDI (REduction of Invariant Probes) Analysis to remove data from poorly performing probes (for description, see http://www.expressionanalysis.com/redi_analysis.html) in combination with Robust Multichip Average (RMA) analysis, which performs background correction, probe-level normalization, and probe-set summary. Both raw and the RMA-processed data have been submitted to the NCBI Gene Expression Omnibus repository (GEO) and can be accessed at http://www.ncbi.nlm.nih.gov/geo/ (series accession number GSE5264; sample accession numbers GSM119354 through GSM119383). Further data analysis was performed using GeneSpring 7 software (Agilent). The cross gene error model was used to estimate the precision of the expression intensity for genes. Deviation values calculated with this model were then used to calculate t test P values. The data were filtered on confidence, requiring P < 0.005 for more than half (6 out of 10) of the time points for each series. All data were normalized to Day 0, so normalized intensity is the fold change relative to this time point. Data were filtered using different fold change cutoffs. Differentially expressed genes were analyzed by one-way ANOVA, using a stringent statistical cutoff of P < 0.005 with the Benjamini and Hochberg false discovery rate test. K-means clustering was performed on some of the gene lists, using 100 iterations and a distance measure based on the Pearson Correlation. Gene network analysis was performed using Ingenuity Systems Pathway Analysis web-based software application. Detailed information about this program and the IPA database can be found at (http://www.ingenuity.com/products/pathways_analysis.html).

Real Time RT-PCR
Relative gene expression in human airway epithelial cells was quantified using real-time quantitative PCR. Samples of the RNA that had been isolated for the microarray analysis were reverse transcribed to generate cDNA. Primer/probe sets were obtained as Taqman pre-developed assay reagents (concentrated and pre-optimized mix of primers and FAM-labeled Taqman probe) from Applied Biosystems (University Park, IL). Assay IDs for each reagent set used were: MUC1 (Hs00159356_m1), MUC5AC (Hs01365601_m1), MUC5B (Hs00861588_m1), MUC15 (Hs00377336_m1), CAPS (Hs00362033_g1), IFT57 (Hs00215973_m1), RABL5 (Hs00224367_m1), ROPN1L (Hs00230481_m1), TUBA3 (Hs00362387_m1), AREG (Hs00155832_m1), BMP4 (Hs00370078_m1), BMP7 (Hs00233477_m1), TGFB1 (Hs99999918_m1), WNT4 (Hs00229142_m1). Quantitative fluorogenic amplification of cDNA was performed using the ABI Prism 7500 Sequence Detection System (Applied Biosystems), primer/probe sets of interest, and TaqMan Universal PCR Master Mix (Applied Biosysytems). The relative abundance of mRNA levels was determined from standard curves generated from a serially diluted standard pool of cDNA prepared from cultured human airway epithelial cells. The relative abundance of -actin mRNA was used to normalize levels of the mRNAs of interest.

	RESULTS

Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References

 
Mucociliary Differentiation of Cultured HBECs
Many of the morphologic and physiologic features of HBEC differentiation in ALI culture, including cilia formation, mucus production, and transepithelial resistance, have already been well characterized (3, 7, 8). We performed a simple histologic analysis to establish the timing of mucociliary differentiation in our cultures for comparison with data from the expression profiling. At five different time points during culture, cells were fixed, sectioned, and stained with H&E to examine general morphology or with Alcian blue/PAS reagents, which stain acid mucins blue and neutral mucins pink. Immunostaining was also performed to examine expression of markers of cilia formation (acetylated -tubulin) and mucus production (the mucin protein, MUC5AC).

For approximately the first week of ALI culture (Day 0 and Day 6), the cells formed a stratified squamous epithelial layer (Figures 1A–1F). By Day 14, the cells were less flattened in appearance, but no other dramatic morphologic changes were apparent (Figures 1D and 1E). However, while cilia could not be discerned on cells at this stage, concentrations of -tubulin were observed on the surface of individual cells (Figure 1I, yellow arrow), suggesting that ciliogenesis had initiated and protein constituents of cilia were being synthesized and transported. By Day 21, many of the cells had a columnar morphology and were arranged in a pseudostratified layer (Figure 1J). At both Day 21 and Day 29, the cells had hairlike extensions on their apical surface, suggestive of cilia (Figures 1J and 1M). The presence of cilia was confirmed by the immunostaining, which showed extensions on the cell surfaces that were positively labeled for -tubulin (Figures 1L and 1O, red staining, indicated by yellow arrows).

View larger version (59K): [in this window] [in a new window]   Figure 1. Mucociliary differentiation of primary HBECs during ALI culture. Cells were collected at Day 0 (A–C), Day 6 (D–F), Day 14 (G–I), Day 21 (J–L), and Day 29 (M–O) of culture, fixed, and sectioned. Sections were stained with H&E (left column), Alcian blue/PAS reagent (middle column), or immunostained to label acetylated -tubulin (red) and MUC5AC (green) with DAPI nuclear staining (blue, right column). (J, M) Ciliated cells are indicated with black arrows and basal cells are denoted with black arrowheads. (K, N) Individual cells with high levels of PAS and Alcian blue staining are indicated with green arrows. (I, L, O) Concentrations of acetylated -tubulin staining were observed on the surface of ciliated cells (yellow arrows) and high levels of mucin staining were also observed on the surface of individual secretory cells (white arrowheads).

Global Gene Expression Profiling of Differentiating HBECs
To examine changes in gene expression during differentiation of airway epithelial cells, RNA was isolated from HBECs harvested at 11 different time points from Days 0–28 of ALI culture. The majority of time points were performed in triplicate, using an RNA sample from each of the three different donor cell lines (Days 2, 12, and 17 were performed in duplicate due to a lower yield from one donor line). A transcriptional analysis of the cultured HBECs was performed using the Affymetrix Human Genome U133 Plus 2.0 array, which represents more than 47,000 transcripts (http://www.affymetrix.com/products/arrays/specific/hgu133plus.affx). The data was processed using the RMA algorithm, which performs a background correction, a normalization step, and a probe-level summary. This method was selected because it has been demonstrated to have higher precision, particularly for low expression values, and higher specificity and sensitivity than many of the other commonly used methods (9). 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 GSE5264. The original .CEL files from the experiment are also available from the GEO website for investigators interested in performing their own analyses.

Log-log plots of the expression data were examined to determine the consistency between replicates as a quality control measure. Figure 2A shows two representative plots comparing replicates from Day 0 and Day 28. The tight clustering of the data around the central line indicates strong correlation between replicates. All replicate combinations were examined in this manner, and all showed similar correlation, suggesting that there was strong similarity between all of the three donor cell lines (data not shown). Normalized data were filtered based on fold changes relative to Day 0 and analyzed with statistical tests that are described in more detail in MATERIALS AND METHODS. The data were first filtered to identify probe sets that display at least a 2-fold increase or decrease in expression relative to Day 0 of culture for one or more time points of the culture period. A total of 1,538 probe sets representing 1,257 different genes or expressed sequence tags (ESTs) displayed increases of 2-fold or greater, while 1,309 probe sets representing 957 genes/ESTs that displayed 2-fold or larger decreases were identified (Figure 2B). A breakdown of the number of probe sets showing 2-fold changes for each stage relative to Day 0 is given in Figure 2C. The number of genes with significant changes in expression increased gradually between Day 1 and Day 10, and then leveled off, consistent with the trends displayed in the line graphs (Figure 2B). This is interesting, as the appearance of cilia and increased mucus production did not occur until after Day 14 in our cultures (Figure 1), suggesting that transcriptional regulation of genes involved in these processes occurs much earlier in the culture period. A complete list of all the genes displaying 2-fold changes during the culture period is available in the online supplemental.

View larger version (43K): [in this window] [in a new window]   Figure 2. (A) Representative log-log plots of normalized expression data. Plots of log (base 2) intensity values for replicates from Day 0 (left) and Day 28 (middle). The tight clustering of the data indicates strong consistency between replicates. Log-log plot comparing expression data from individual samples at different stages, such as Day 0 and Day 28 (right), shows the distribution of genes with differential expression. The central red line indicates no difference in expression between samples, and the green lines indicate 2-fold changes. (B) Genes displaying statistically significant 2-fold changes during ALI culture of HBECs. Line graphs show temporal expression patterns for genes that displayed at least a 2-fold increase (left) or 2-fold decrease (right) for one or more time points during the culture period. Normalized intensity equals the fold change relative to Day 0 of culture. (C) Summary of the number of probe sets that displayed a 2-fold increase or decrease for each day of culture as compared with Day 0.

View larger version (57K): [in this window] [in a new window]   Figure 3. K-means clusters for genes that displayed 2-fold changes during the ALI culture time course. Line graphs demonstrate the temporal changes for all the probe sets in each cluster. The column on the right lists the highest scoring Biological Process terms obtained for each cluster, from gene ontology classification using DAVID. The number of probe sets for each term is given, as well as the associated P value for each term.

The three clusters containing down-regulated genes also represented very distinct biological processes (Figure 3). Cluster 3 included genes that undergo an early increase between Day 1 and Day 4 of culture, followed by a drop in expression, whereas cluster 4 included genes that are down-regulated gradually over the culture period. Interestingly, the highest scoring category for both of these clusters is "ectoderm development." The genes in this category include keratinocyte markers such as small proline rich proteins (SPRR1A, SPRR1B) and sciellin (SCEL), as well as several different keratins (KRT6B, KRT16, KRTHA1, KRTHA4) and laminins (LAMA3, LAMB3, LAMG2). While the description "ectoderm development" obviously does not refer to the airway epithelium, this categorization does suggest that the squamous, undifferentiated HBECs may share characteristics with keratinocytes at the molecular level. Other biological processes for genes in clusters 3 and 4 were more distinct. Several of the highest-scoring categories for cluster 3 involved defense or wound healing responses. Genes in this cluster had an interesting temporal expression profile, with an initial, short increase in expression followed by a gradual decline. The initial transition to ALI culture may be stressful to the cells, and might explain the transient up-regulation of genes related to healing and defense. The categories for cluster 4 included a broad range of processes such as alcohol or carbohydrate metabolism, regulation of the cell cycle and cell death, and cell signaling.

By contrast, cluster 5 genes displayed a very distinct temporal pattern and represented a more restricted set of processes. Genes in this cluster underwent an initial sharp increase from Day 0 to Day 1, followed by a gradual decrease until approximately Day 8. A second peak in expression occurred later in the culture period, between Days 14 and 21 (Figure 3). The top three biological process categories for this cluster were all related to cell cycle. The genes in these categories included cyclins (CCNA2, CCNB1, CCNB2, CCNF), kinesins (KIF11, KIF2C, KIF23), checkpoint homolog 1 (CHK1), cyclin-dependent kinase inhibitor 3 (CDKN3), S-phase kinase-associated protein (SKP2), protein regulator of cytokinesis (PRC1), and kinetochore-associated 2 (KNTC2). Many of the remaining categories for this cluster were related to lipid, steroid, or fatty acid metabolism. The significance of this is not clear, but these genes could also be related to cell division, as lipids have numerous signaling roles in regulation of cell proliferation, and lipid biosynthesis may be required for aspects of cell growth and cytokinesis. These changes suggest there may be different periods of proliferative activity over the culture time course. As both basal cells and nonciliated secretory cells of the bronchial airways have proliferative capacity in vivo, either or both of these cell populations may have mitotic activity in culture as well (10, 11). However, further analyses are required to determine the extent and specificity of cell proliferation in this culture system.

Genes with the Largest Changes in Expression
As the list of genes generated using a 2-fold filter was too long for careful scrutiny of individual genes, we examined the genes with the largest changes in expression level during the culture time course. Genes that displayed 10-fold change for at least one time point relative to Day 0 are listed in Table 1. Several genes in this list have previously been shown to be upregulated during ALI culture of HBECs, providing support for our results. For example, FOXJ1 encodes a transcription factor required for ciliogenesis that regulates the organization and localization of the basal bodies, and expression of FOXJ1 mRNA has been shown to increase between Day 7 and Day 14 of ALI culture (12, 13). In our analysis, FOXJ1 transcripts increased by close to 10-fold by Day 10 of culture. Several experiments have shown that PLUNC is one of the major secreted proteins in ALI cultures of primary human airway epithelial cells, and we observed an very large increase in PLUNC expression (14). C9orf24 encodes ciliated bronchial epithelium 1 (CBE1), which is not expressed in undifferentiated HBECs but is strongly induced by Day 7 of ALI culture (13). Similarly, we identified an increase of > 35-fold for C9orf24. We also detected a large increase in the mucin gene, MUC5B, and numerous other experiments have previously shown this gene to be up-regulated during ALI culture of HBECs (8, 15). Finally, our results corroborate a recently published SAGE (serial analysis of gene expression) analysis, which identified several of the same genes as being highly abundant in the bronchial epithelium, including MSMB, CD74, TUBA3, MUC5B, SCGB3A1, and CGI-38 (16).

Of the remaining genes that were not as easily categorized, several have known roles associated with airway epithelial biology or cell differentiation. Large increases were observed for the salivary protein statherin (STATH) and SCGB3A1, a member of the secretoglobin family that is expressed by secretory cells of the airway and is down-regulated in non–small cell lung carcinomas (18). Cancer susceptibility 1 (CASC1) is a locus associated with predisposition to lung tumor development, and also displayed a large increase (19). Finally, ID2 encodes an inhibitor of bHLH transcription factors that is RA-responsive and involved in regulation of cell differentiation and proliferation in a range of tissues (20).

The genes with the largest decreases in expression were also divided into several distinct functional categories (Table 1). As previously discussed, several genes that are markers of keratinocytes were down-regulated, including small proline-rich proteins and keratins. Multiple proteinases showed large decreases, including members of the kallikrein enzyme family (KLK6, KLK7, KLK8). Factors with established roles in the stimulation of cell growth and proliferation (PGF, EREG) as well as molecules that have roles in cell adhesion (LOXL2, LGALS1, ASAM) also showed very large decreases during the culture period (Table 1). There were a number of genes that we were not able to categorize by function, but that did share some interesting characteristics. S100A7, also called psoriasin 1, is a calcium-binding protein that is up-regulated in psoriatic skin and in different types of squamous cell carcinomas (SCCs) (21). SERPINE1, also known as plasminogen activator inhibitor 1 (PAI-1), is a regulator of the fibrinolytic system, and is up-regulated in some squamous cell carcinomas (22). ECM1 is expressed in normal skin, but is similarly increased in SCCs (23). Finally, both the carbonic anhydrase, CA9, and xanthine dehydrogenase (XDH) show increased expression or activity in certain types of SCCs (24, 25). Whereas all of these genes have very different functions, it is particularly striking that so many of them are associated with either normal keratinocyte development or SCC.

Examination of Mucin Gene Expression during HBEC Differentiation
As two of the primary features of HBEC differentiation are mucus production and cilia formation, we decided to examine genes involved in these processes in more detail. First, we looked at expression data for genes encoding mucins. Mucins are large, glycosylated proteins that are a major constituent of mucus, and are important for innate immune defense of the airway epithelium (26). A number of mucin genes and proteins, including MUC2, MUC3, MUC4, MUC5AC, MUC6, MUC7, and MUC8, have previously been shown to be expressed by cultured human airway epithelial cells (8, 15, 27). Thus, we expected that numerous mucin genes would display significant increases in our analysis. Interestingly, however, the only mucin gene that showed a large increase in our experiment was MUC5B. Smaller, but statistically significant, increases were also detected for genes encoding the membrane-bound mucins MUC1 and MUC15 (Table 1, Figure 4). However, no significant changes were detected for the other mucin genes known to be expressed by HBECs. It was particularly striking that we did not detect a significant increase in MUC5AC, which is one of the two major secreted mucins (the other being MUC5B) and has been shown to be up-regulated in a differentiation-dependent manner in cultured bronchial cells (8).

View larger version (33K): [in this window] [in a new window]   Figure 4. Examination of mucin expression in differentiating HBECs. Bar graphs indicate fold changes in expression of four mucin genes during ALI culture as determined by the microarray analysis (light gray bars) and by real-time RT-PCR (dark gray bars). Fold changes for the microarray data and the PCR data for MUC1 and MUC15 are relative to Day 0. Error bars indicate SEM. For MUC5B and MUC5AC the data were normalized to the values for Day 2, which was the first stage at which a significant signal was detected. Microarray results include data for the Affymetrix following probe sets: MUC1 (213693_s_at, 211695_x_at, 207847_s_at), MUC15 (227238_at), MUC5AC (217187_at, 217182_at, 214385_s_at, 214303_x_at), and MUC5B (222268_x_at, 213432_at).

Identification of Genes with Potential Roles in Cilia Formation or Function
In the airway, cilia are involved in the removal of inhaled particles and pathogens, and ciliary function is critical for respiratory health (28). The cilia of the airway epithelium have a 9+2 microtubule organization and are referred to as "motile" cilia, which are also found in the oviduct, testis, and brain. Abnormalities in cilia formation (primary ciliary dyskinesia) can cause clinical defects related to respiratory function, reproduction, and development (28). Thus, identification of genes that regulate ciliogenesis and encode components of the ciliary axoneme is of great importance.

The ALI cell culture system provides a unique opportunity to examine the molecular changes that occur during ciliogenesis in HBECs. However, the gene ontology analyses we performed did not identify many cilia-associated genes, primarily because there are very few human genes in the ontogeny databases that have been classified in this manner. Much of the information we do have regarding the molecular structure of the ciliary and flagellar axonemes has come from proteomics studies in a range of organisms. Recently, a proteomic analysis was performed on human cilia, using axonemes purified from primary ALI cultures of HBECs (29). Through a combination of two-dimensional Page and liquid chromatography/tandem mass spectrometry approaches, researchers were able to obtain sequence data for over 1,400 peptides representing more than 200 proteins found in purified axonemes (29). To determine which of the genes that displayed increases in our microarray analysis are involved in cilia formation, we compared our results with these published proteomics data.

A list of 301 genes represented on the Affymetrix array was generated using the protein accession numbers and names from the proteomics analysis (29). This list is longer than the total number of proteins identified in the analysis (214) because, for several of the proteins, such as -tubulin, -tubulin, or actin, there are multiple genes that could encode the described protein. For these situations, all candidate genes were included in the comparison. The Venn diagram in Figure 5A shows the overlap between the list of genes that displayed 2-fold increases in our microarray analysis and the list of genes encoding proteins from the proteomic analysis. The overlap of these two lists contained 49 probe sets representing 37 different genes. Figure 5A shows the temporal expression profiles for these 49 probe sets. Strikingly, the majority of these genes had very similar profiles, with a dramatic increase in expression that occurred between Days 2 and 8 of culture. This supports the idea that these genes have related functions and may be transcriptionally regulated by a shared mechanism. Of note, these genes all were included in group 2 in the clustering analysis (Figure 3), which contained many genes involved in "cytoskeleton organization and biogenesis," as previously discussed.

View larger version (38K): [in this window] [in a new window]   Figure 5. Identification of candidate cilia genes. (A) Venn diagram (left) showing the overlap between the list of genes displaying statistically significant 2-fold increases during the culture experiment and the list containing genes encoding proteins identified in cilia proteomic analyses. Line graph (right) showing microarray expression data for the 49 probe sets contained in both lists. (B) Validation of the expression data for five candidate cilia genes by real-time RT-PCR. Bar graphs indicate fold changes in expression of four mucin genes during ALI culture as determined by the microarray analysis (light gray bars) and by real-time RT-PCR (dark gray bars). All data are fold changes relative to Day 0 except the PCR data for ROPN1L, which was normalized to Day 4, the earliest stage with a detectable signal. Error bars indicate SEM. Microarray results include data for the following Affymetrix probe sets: CAPS (231729_s_at, 231728_at, 226424_at), IFT57 (222520_s_at, 222519_s_at, 218100_s_at), RABL5 (222742_s_at), ROPNL1 (223609_at) and TUBA3 (209118_s_at).

Calcium homeostasis is important for many aspects of ciliary and flagellar function. Several genes encoding proteins that bind calcium ions or are involved in calcium transport were identified in addition to CETN2. C10orf63 encodes enkurin, which is expressed by sperm and physically interacts with Ca⁺⁺ channels (35). Calcyphosine (CAPS) is a cyclic AMP-regulated protein that binds calcium via multiple EF-hand domains (36). Two other genes encoding proteins with EF-hand domains were identified: EFHC1 and EFHC2. EFHC1 is a calcium ion-binding protein that is abundant in sperm flagella and tracheal cilia and a homolog of the Chlamydomonas axonemal protein, Rib72 (37). Less is known about EFHC2, but its strong similarity to EFHC1 suggests that it may have related functions. Finally, RAB proteins are members of the RAS family of GTP-binding proteins that have a variety of roles, including regulation of membrane-associated intracellular transport. Two member of this family were identified (RABL4 and RABL5), and while specific functions for these genes in cilia are not known, related small GTPases have been demonstrated to undergo ciliary transport and be required for normal ciliary function (38).

Confirmation of the microarray results was performed by real-time PCR, and several genes not known to be expressed by ALI cultures of HBECs and representing different functional categories were examined (Figure 5B). The PCR results confirmed large increases in expression of CAPS, IFT57, RABL5, ROPN1L, and TUBA3. Similar to the microarray results, the PCR analysis also demonstrated a sharp increase in expression occurring between Day 4 and Day 10, supporting the idea that this is the critical period for transcriptional regulation of genes involved in ciliogenesis.

Identification of Signaling Pathways with Candidate Roles in HBEC Differentiation
To identify candidate signaling pathways and gene networks that may regulate differentiation of HBECs, we used Ingenuity Pathway Analysis (IPA) to evaluate relationships between genes showing at least a 2-fold increase or decrease. It is important to clarify that IPA-generated networks are not the same as canonical biological pathways, but rather connect different proteins and genes based upon a wide range of interactions reported in the scientific literature. All the genes that showed a 2-fold or larger increase or decrease for at least one stage relative to Day 0 were included in the analysis. Of the 2,847 genes entered, the IPA database contained information for 1,049. With this list of genes, 70 networks were formed, of which 13 had the highest possible statistical score. We examined these networks to specifically identify ones containing signaling factors with known roles in cell differentiation. Three networks were identified: one including members of the TGF- signaling pathway (Figure 6A), one containing genes involved in the WNT/-catenin pathway (Figure 6B), and a third network centered on the epidermal growth factor receptor (Figure 6C). The remaining networks were centered around common or ubiquitous transcription factors with very broad functions, making interpretation difficult. Verification and investigation of genes in other networks will be a future focus.

View larger version (58K): [in this window] [in a new window]   Figure 6. Identification of candidate signaling pathways involved in mucociliary differentiation of cultured HBECs. (A) IPA gene network (left) displaying potential interactions between genes in TGF- family signaling pathways. Solid lines denote direct interactions, while dotted lines represent indirect interactions between the genes. Genes colored green displayed 2-fold or larger decreases during the time course, while genes colored red displayed 2-fold or larger increases. Fold change expression values by each gene name are for Day 28 of culture relative to Day 0. Line graph (right) showing the temporal changes in expression for genes in the canonical TGF- pathway, as determined from the microarray analysis. (B) IPA gene network (left) and graph (right) of genes related to WNT/-catenin signaling that displayed 2-fold or greater changes during ALI culture. (C) IPA network (left) and graph (right) of genes related to EGFR signaling that displayed significant changes during culture. Additional information about the genes and the indicated interactions can be found at www.ingenuity.com.

In addition to the genes belonging to the canonical signaling pathways, a number of potential genetic targets of these pathways were included in the networks. Many of the down-regulated genes in the TGFB1 network interact with either actin or microtubules, including TPM1, TPM2, TPM3, TPM4, PDLIM7, CALD1, SMTN, MAP4, and KIF3C. Several collagen genes (COL1A1, COL4A2, COL8A1) were also down-regulated. This suggests that the TGFB1 pathway may be involved in the regulation of cell morphology, motility, and cell adhesion in these cells. The observation that TGFB1 and all of these potential target genes are down-regulated during the culture period may also suggest that the pathway primarily functions in the undifferentiated, squamous epithelial cells. By contrast, BMP4 and BMP7 are up-regulated during culture, and therefore may play an active role in the process of mucociliary differentiation. CTNNB1 and the other components of canonical WNT signaling, as well as most of the candidate target genes in the network, all increase during mucociliary differentiation of HBECs (Figure 6B). Genes having reported interactions with CTNNB1 (SLC26A2, ADRB1, LMO2, ALDH1A1, etc.) fell into a range of functional categories, so it is difficult to extrapolate what biological processes may be regulated by this pathway. Finally, the majority of genes that interact with EGFR decreased during the culture period, as did EGFR and its ligands (Figure 6C). Genes in this network included members of the serpine family of serine proteinase inhibitors (SERPINE1, SERPINE2, SERPINA3) and many genes associated with the cytoskeleton and/or the extracellular matrix (KRT16, COL5A2, LAMA4, LAMC2, ITGA5, ITGB1, and FN1). This raises the possibility that the EGFR pathway is involved in regulating aspects of cell structure, morphology, and remodeling. Further work is needed to determine if these gene interactions are relevant in the bronchial epithelium, as many of the interactions in the Ingenuity database are based on data from other tissues and organisms.

Validation of the microarray results was performed by real-time PCR, examining expression of genes from each of the IPA networks that encode secreted signaling factors (Figure 7). PCR confirmed the up-regulation of BMP4, BMP7, and WNT4 during culture, as well as the down-regulation of TGFB1 and AREG that was observed in the microarray analysis. These results, in combination with the fact that many genes also belonging to related signaling pathways also displayed significant changes, strongly suggest that these genes have roles in HBEC function or differentiation.

View larger version (39K): [in this window] [in a new window]   Figure 7. Validation of microarray results for genes encoding ligands in TGF-, WNT/-catenin, and EGFR signaling pathways. Bar graphs indicate fold changes in expression of four mucin genes during ALI culture as determined by the microarray analysis (light gray bars) and by real-time RT-PCR (dark gray bars). Fold changes for the microarray data and the PCR data are relative to Day 0. Error bars indicate SEM. Microarray results include data for the following Affymetrix probe sets: BMP4 (211518_s_at), BMP7 (209591_s_at, 209590_at), TGFB1 (203085_s_at), WNT4 (230751_at, 208606_s_at), and AREG (205239_at).

	DISCUSSION

Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References

We identified a large number of genes that display significant increases or decreases during differentiation of bronchial epithelial cells, many with established functions in the respiratory epithelium and others without known roles in this tissue. Using gene ontology classifications, we characterized a number of biological processes that occur during HBEC differentiation (Figure 3). As discussed previously, the airway epithelium plays a protective role by providing a physical barrier, through the production of a number of immune and inflammatory mediators, and via mucociliary clearance. Importantly, genes related to all three of these functional aspects of the epithelium showed significant changes. Genes involved in cell adhesion were also highly represented in lists of genes showing 2-fold or greater increases. By contrast, many of the down-regulated genes are markers of keratinocytes or other squamous epithelia (Table 1, Figure 3).

We were especially interested in examining expression of genes related to mucus production or cilia formation, as those are two of the major structural/functional features of HBECs cultured at an ALI. Mucins are the major protein component of mucus, and we identified increases in expression of three mucin genes: MUC5B, which encodes a secreted mucin; and MUC1 and MUC15, which encode membrane-tethered proteins. PCR analyses validated these findings, but also detected an increase in expression of MUC5AC, which was not observed in the microarray experiment. There are numerous other mucin genes that are known to be expressed by primary cultures of HBECs that also were not identified by the microarray analysis. Thus, there may be probe sets for some of the mucin genes on these arrays that have problems with sensitivity or specificity and may not useful for examining expression of these genes. Also important to note is that mucin proteins are synthesized and stored until a stimulus for secretion is received, suggesting that studies only looking at RNA levels of mucin genes may not provide a complete perspective. Thus, it is recommended that researchers who wish to thoroughly and accurately examine expression of mucins in the airway epithelium use multiple analysis methods.

By comparing our gene expression data to results from proteomic studies of the mammalian axoneme, we identified candidate genes that may be involved in cilia formation or function. This included several genes with known or predicted roles in this cellular structure, but also many novel or uncharacterized genes. Importantly, all the genes shared a similar kinetic expression profile, supporting the idea that they have related functions. Interestingly, though, we did not find statistically significant increases in the genes encoding the majority of the proteins identified in the proteomics analysis. There are a number of potential explanations for this. Some of the genes may have shown increases in expression, but not large enough to pass the 2-fold filter. In addition, some ciliary proteins may be regulated in a post-transcriptional manner, so we might not detect changes at the level of gene expression. Also, as mentioned by the authors of the proteomics study, different cellular contaminants are likely associated with the purified axonemes, resulting in identification of proteins that are not truly axonemal components. Thus, this analysis was particularly useful in that it provides support for the idea that these 37 genes do have important functions in the ciliary axoneme.

We identified members of several signaling pathways that are transcriptionally regulated during mucociliary cell differentiation, by generating gene networks. Numerous members of the TGF-/BMP superfamily displayed increases or decreases during the culture period, as did several inhibitors or downstream targets of these pathways. TGFB1 was down-regulated during the culture time course, which is consistent with the roles that have been demonstrated for this factor in other studies. TGFB1 inhibits proliferation of airway epithelial cells in culture and induces terminal squamous differentiation (39–41). It has also been demonstrated that TGFB1 induces reorganization of the cytoskeleton and causes shape changes and cell adhesion alterations in epithelial tracheal cells in culture (39). Thus it is not surprising that many of the potential genetic targets of TGFB1 identified in the network analysis (Figure 5C) are associated with the cytoskeleton or extracellular matrix. When taken together, these findings suggest that TGFB1 could have important roles in the early proliferation block and squamous phenotype of HBECs, but its down-regulation may be required for the cells to differentiate into a ciliated, columnar epithelium.

Bone morphogenetic proteins (BMPs) are involved in differentiation of a number of different cell types and tissues. Expression of both BMP4 and BMP7 increased during the culture time course, while BMP2 was down-regulated (Figure 6A). All three of these genes have previously been shown to be expressed by cultured bronchial epithelial cells, although their precise function in these cells is not known (42). However, roles for BMP4 in lung morphogenesis, specifically in proximal-distal patterning of the lung epithelium, have been described (43). In addition, two other members of the TGF- superfamily, INHBA and INHBB, which encode subunits of activins and inhibin, both showed significant changes during the time course. In summary, there are many complex changes in expression of TGF- family signaling molecules during the differentiation of these cells, and further analyses will be required to dissect the specific roles of these many different factors.

Network analysis also identified a large number of genes that are involved in WNT/-catenin signaling (Figure 5). WNTs are secreted ligands that bind to members of the frizzled family of transmembrane receptors. -catenin is an intracellular downstream mediator of this signal, although it also has functions independent of the WNT pathway. We observed an increase in -catenin expression during mucociliary differentiation of HBECs, which is consistent with other published data that suggest a role for this molecule in the airway. Expression of -catenin was increased in the bronchial epithelial after naphthalene injury in mice and ectopic expression of an activated form of -catenin in airway epithelium of mice caused goblet cell hyperplasia (44, 45). In addition, conditional deletion of the -catenin gene in the mouse lung resulted in a disruption of proximal and distal epithelial cell fate specification (46). Thus, -catenin clearly has important roles in the airway epithelium, but its precise functions are still not well understood.

WNT4 was the only WNT gene that displayed a significant increase during the culture experiment, but there are currently no described roles for WNT4 signaling during normal lung development. However, WNT4 has well-established roles in the regulation of cell differentiation in a range of other tissues, and has also been implicated in the development of lung cancer (47). Multiple receptors for WNTs and secreted inhibitors of WNT signaling also displayed significant changes during the time course (Figure 5B). Combined with the observation that intracellular components of the canonical WNT/-catenin pathway (AXIN2, FRAT1) also increase during differentiation, these results strongly suggest a potential role for this pathway in HBECs.

Finally, the EGFR and three of its ligands were also down-regulated during the differentiation time course. The EGFR pathway has been extensively studied in cells of the airway, and has been implicated in numerous processes in this tissue, including mucin production, cell proliferation, and inflammation (48, 49). EGFR is also involved in the response to epithelial injury, as its expression is elevated in areas of structural damage in the bronchial epithelium (50). In addition, all three of the EGFR ligands identified, AREG, EREG, and HBEGF, have been shown to be expressed by cells of the airway and are induced in chronic airway disease or in response to different types of pollutants or toxins (51, 52).

Interestingly, several of the genes included in the EGFR network (Figure 6C) are also highly expressed by squamous epithelia, including KRT16, LAMA4, LAMC2, SERPINE1, and SERPINE2 (53–57). As previously discussed, our analyses identified many down-regulated genes that are markers of keratinocytes or squamous cell carcinoma (Figure 3, Table 2). This is consistent with previous studies demonstrating cells cultured in the absence of RA undergo squamous differentiation, suggesting an important role for RA in regulating squamous versus columnar cell fates (1, 2). Additional examination of the list of genes displaying 2-fold decreases identified even more genes expressed by squamous epithelia. These included keratin 4 (KRT4), laminin 3 (LAMB3), cadherin 13 (CDH13), numerous genes that encode proteins specifically associated with the cornified cell envelope of the epidermis (CST6, DSG2, DSG3, PPL, SCEL, SPRR1A, SPRR1B, SPRR2B, SPRR3), and basonuclin (BNC1), a zinc finger protein expressed by the basal layer of the epidermis (53, 57–59). Two other down-regulated genes identified by the microarray analysis, transglutamase I and involucrin, are expressed by squamous epithelia and are also established markers of squamous metaplasia in the lung in vivo (60, 61). Squamous cell metaplasia is a common alteration of the tracheobronchial epithelium that can occur as a result of airway injury. These lesions are associated with various pathological conditions such as asthma or COPD, and may be pre-neoplastic. Upon injury, ciliated epithelial cells of the airway transform to a squamous phenotype, spread beneath injured cells, proliferate, and then differentiate from squamous to cuboidal to ciliated, columnar cells (44). This is strikingly similar to the morphologic changes observed in our ALI cultures of primary bronchial epithelial cells (Figure 1). The observation that so many genes associated with normal squamous epithelia, squamous cell carcinomas, or EGFR signaling are initially expressed, and then down-regulated, suggests that there may be molecular parallels between these cultures and injured epithelia or neoplastic lesions in vivo. Importantly, while we described the cells at Day 0 as being "undifferentiated," they may not truly be representative of undifferentiated airway epithelial cells in vivo, as they had been extensively manipulated, undergone a period of submerged culture on a plastic substrate, and were exposed to RA before removal of media. Instead, our results may indicate that cells at the initial stages of ALI culture may more closely represent damaged epithelium, and that this culture system could be useful for the study of airway wound healing and repair and/or carcinogenesis. In addition, some of the genes identified in this study may be useful as diagnostic markers for the identification and characterization of neoplastic lesions in vivo.

In summary, the identification of genes displaying significant changes in expression during ALI culture has provided valuable insight into the major biological process that occur during differentiation of HBECs. The specific classes of genes that are up- and down-regulated correlate well with the transformation of these cultures from a proliferative, squamous cell layer into a pseudostratified, ciliated epithelium that has many of the properties of the in vivo airway epithelium. These studies have also identified many individual genes that may have important roles in airway epithelial differentiation or function and warrant further study. This catalog of gene expression changes can serve as an important reference tool and a framework for more in-depth analyses of airway epithelial function.

	Footnotes

 
This report has been reviewed by the National Health and Environmental Effects Research Laboratory, United States Environmental Protection Agency and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

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.2006-0466OC on April 5, 2007

Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form December 19, 2006

Accepted in final form March 23, 2007

	References

Top Abstract CLINICAL RELEVANCE MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Whitcutt MJ, Adler KB, Wu R. A biphasic chamber system for maintaining polarity of differentiation of cultured respiratory tract epithelial cells. In Vitro Cell Dev Biol 1988;24:420–428.[Medline]
2. Kaartinen L, Nettesheim P, Adler KB, Randell SH. Rat tracheal epithelial cell differentiation in vitro. In Vitro Cell Dev Biol Anim 1993;29A:481–492.[CrossRef]
3. Gray TE, Guzman K, Davis CW, Abdullah LH, Nettesheim P. Mucociliary differentiation of serially passaged normal human tracheobronchial epithelial cells. Am J Respir Cell Mol Biol 1996;14:104–112.[Abstract]
4. Davenport EA, Nettesheim P. Regulation of mucociliary differentiation of rat tracheal epithelial cells by type i collagen gel substratum. Am J Respir Cell Mol Biol 1996;14:19–26.[Abstract]
5. Ostrowski LE, Nettesheim P. Inhibition of ciliated cell differentiation by fluid submersion. Exp Lung Res 1995;21:957–970.[Medline]
6. Turi JL, Jaspers I, Dailey LA, Madden MC, Brighton LE, Carter JD, Nozik-Grayck E, Piantadosi CA, Ghio AJ. Oxidative stress activates anion exchange protein 2 and ap-1 in airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 2002;283:L791–L798.[Abstract/Free Full Text]
7. de Jong PM, van Sterkenburg MA, Hesseling SC, Kempenaar JA, Mulder AA, Mommaas AM, Dijkman JH, Ponec M. Ciliogenesis in human bronchial epithelial cells cultured at the air-liquid interface. Am J Respir Cell Mol Biol 1994;10:271–277.[Abstract]
8. Bernacki SH, Nelson AL, Abdullah L, Sheehan JK, Harris A, Davis CW, Randell SH. Mucin gene expression during differentiation of human airway epithelia in vitro: Muc4 and muc5b are strongly induced. Am J Respir Cell Mol Biol 1999;20:595–604.[Abstract/Free Full Text]
9. 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]
10. Boers JE, Ambergen AW, Thunnissen FB. Number and proliferation of basal and parabasal cells in normal human airway epithelium. Am J Respir Crit Care Med 1998;157:2000–2006.[Abstract/Free Full Text]
11. Evans MJ, Shami SG, Cabral-Anderson LJ, Dekker NP. Role of nonciliated cells in renewal of the bronchial epithelium of rats exposed to no2. Am J Pathol 1986;123:126–133.[Abstract]
12. Gomperts BN, Gong-Cooper X, Hackett BP. Foxj1 regulates basal body anchoring to the cytoskeleton of ciliated pulmonary epithelial cells. J Cell Sci 2004;117:1329–1337.[Abstract/Free Full Text]
13. Yoshisue H, Puddicombe SM, Wilson SJ, Haitchi HM, Powell RM, Wilson DI, Pandit A, Berger AE, Davies DE, Holgate ST, et al. Characterization of ciliated bronchial epithelium 1, a ciliated cell-associated gene induced during mucociliary differentiation. Am J Respir Cell Mol Biol 2004;31:491–500.[Abstract/Free Full Text]
14. Campos MA, Abreu AR, Nlend MC, Cobas MA, Conner GE, Whitney PL. Purification and characterization of plunc from human tracheobronchial secretions. Am J Respir Cell Mol Biol 2004;30:184–192.[Abstract/Free Full Text]
15. Holmen JM, Karlsson NG, Abdullah LH, Randell SH, Sheehan JK, Hansson GC, Davis CW. Mucins and their o-glycans from human bronchial epithelial cell cultures. Am J Physiol Lung Cell Mol Physiol 2004;287:L824–L834.[Abstract/Free Full Text]
16. Lonergan KM, Chari R, Deleeuw RJ, Shadeo A, Chi B, Tsao MS, Jones S, Marra M, Ling V, Ng R, et al. Identification of novel lung genes in bronchial epithelium by serial analysis of gene expression. Am J Respir Cell Mol Biol 2006;35:651–661.[Abstract/Free Full Text]
17. Castell JV, Donato MT, Gomez-Lechon MJ. Metabolism and bioactivation of toxicants in the lung. The in vitro cellular approach. Exp Toxicol Pathol 2005;57:189–204.[CrossRef][Medline]
18. Reynolds SD, Reynolds PR, Pryhuber GS, Finder JD, Stripp BR. Secretoglobins scgb3a1 and scgb3a2 define secretory cell subsets in mouse and human airways. Am J Respir Crit Care Med 2002;166:1498–1509.[Abstract/Free Full Text]
19. Zhang Z, Futamura M, Vikis HG, Wang M, Li J, Wang Y, Guan KL, You M. Positional cloning of the major quantitative trait locus underlying lung tumor susceptibility in mice. Proc Natl Acad Sci USA 2003;100:12642–12647.[Abstract/Free Full Text]
20. Nigten J, Breems-de Ridder MC, Erpelinck-Verschueren CA, Nikoloski G, van der Reijden BA, van Wageningen S, van Hennik PB, de Witte T, Lowenberg B, Jansen JH. Id1 and id2 are retinoic acid responsive genes and induce a g0/g1 accumulation in acute promyelocytic leukemia cells. Leukemia 2005;19:799–805.[CrossRef][Medline]
21. Madsen P, Rasmussen HH, Leffers H, Honore B, Dejgaard K, Olsen E, Kiil J, Walbum E, Andersen AH, Basse B, et al. Molecular cloning, occurrence, and expression of a novel partially secreted protein "Psoriasin" That is highly up-regulated in psoriatic skin. J Invest Dermatol 1991;97:701–712.[CrossRef][Medline]
22. Hundsdorfer B, Zeilhofer HF, Bock KP, Dettmar P, Schmitt M, Kolk A, Pautke C, Horch HH. Tumour-associated urokinase-type plasminogen activator (upa) and its inhibitor pai-1 in normal and neoplastic tissues of patients with squamous cell cancer of the oral cavity - clinical relevance and prognostic value. J Craniomaxillofac Surg 2005;33:191–196.[Medline]
23. Wang L, Yu J, Ni J, Xu XM, Wang J, Ning H, Pei XF, Chen J, Yang S, Underhill CB, et al. Extracellular matrix protein 1 (ecm1) is over-expressed in malignant epithelial tumors. Cancer Lett 2003;200:57–67.[CrossRef][Medline]
24. Hoogsteen IJ, Marres HA, Wijffels KI, Rijken PF, Peters JP, van den Hoogen FJ, Oosterwijk E, van der Kogel AJ, Kaanders JH. Colocalization of carbonic anhydrase 9 expression and cell proliferation in human head and neck squamous cell carcinoma. Clin Cancer Res 2005;11:97–106.[Abstract/Free Full Text]
25. Durak I, Isik AC, Canbolat O, Akyol O, Kavutcu M. Adenosine deaminase, 5' nucleotidase, xanthine oxidase, superoxide dismutase, and catalase activities in cancerous and noncancerous human laryngeal tissues. Free Radic Biol Med 1993;15:681–684.[CrossRef][Medline]
26. Rose MC, Voynow JA. Respiratory tract mucin genes and mucin glycoproteins in health and disease. Physiol Rev 2006;86:245–278.[Abstract/Free Full Text]
27. Koo JS, Yoon JH, Gray T, Norford D, Jetten AM, Nettesheim P. Restoration of the mucous phenotype by retinoic acid in retinoid-deficient human bronchial cell cultures: changes in mucin gene expression. Am J Respir Cell Mol Biol 1999;20:43–52.[Abstract/Free Full Text]
28. Eley L, Yates LM, Goodship JA. Cilia and disease. Curr Opin Genet Dev 2005;15:308–314.[CrossRef][Medline]
29. Ostrowski LE, Blackburn K, Radde KM, Moyer MB, Schlatzer DM, Moseley A, Boucher RC. A proteomic analysis of human cilia: Identification of novel components. Mol Cell Proteomics 2002;1:451–465.[Abstract/Free Full Text]
30. Pennarun G, Escudier E, Chapelin C, Bridoux AM, Cacheux V, Roger G, Clement A, Goossens M, Amselem S, Duriez B. Loss-of-function mutations in a human gene related to chlamydomonas reinhardtii dynein ic78 result in primary ciliary dyskinesia. Am J Hum Genet 1999;65:1508–1519.[CrossRef][Medline]
31. Fliegauf M, Olbrich H, Horvath J, Wildhaber JH, Zariwala MA, Kennedy M, Knowles MR, Omran H. Mislocalization of dnah5 and dnah9 in respiratory cells from patients with primary ciliary dyskinesia. Am J Respir Crit Care Med 2005;171:1343–1349.[Abstract/Free Full Text]
32. Tanaka H, Iguchi N, Toyama Y, Kitamura K, Takahashi T, Kaseda K, Maekawa M, Nishimune Y. Mice deficient in the axonemal protein tektin-t exhibit male infertility and immotile-cilium syndrome due to impaired inner arm dynein function. Mol Cell Biol 2004;24:7958–7964.[Abstract/Free Full Text]
33. Laoukili J, Perret E, Middendorp S, Houcine O, Guennou C, Marano F, Bornens M, Tournier F. Differential expression and cellular distribution of centrin isoforms during human ciliated cell differentiation in vitro. J Cell Sci 2000;113:1355–1364.[Abstract]
34. Sapiro R, Kostetskii I, Olds-Clarke P, Gerton GL, Radice GL, Strauss IJ. Male infertility, impaired sperm motility, and hydrocephalus in mice deficient in sperm-associated antigen 6. Mol Cell Biol 2002;22:6298–6305.[Abstract/Free Full Text]
35. Sutton KA, Jungnickel MK, Wang Y, Cullen K, Lambert S, Florman HM. Enkurin is a novel calmodulin and trpc channel binding protein in sperm. Dev Biol 2004;274:426–435.[CrossRef][Medline]
36. Lefort A, Lecocq R, Libert F, Lamy F, Swillens S, Vassart G, Dumont JE. Cloning and sequencing of a calcium-binding protein regulated by cyclic amp in the thyroid. EMBO J 1989;8:111–116.[Medline]
37. Ikeda T, Ikeda K, Enomoto M, Park MK, Hirono M, Kamiya R. The mouse ortholog of efhc1 implicated in juvenile myoclonic epilepsy is an axonemal protein widely conserved among organisms with motile cilia and flagella. FEBS Lett 2005;579:819–822.[CrossRef][Medline]
38. Fan Y, Esmail MA, Ansley SJ, Blacque OE, Boroevich K, Ross AJ, Moore SJ, Badano JL, May-Simera H, Compton DS, et al. Mutations in a member of the ras superfamily of small gtp-binding proteins causes bardet-biedl syndrome. Nat Genet 2004;36:989–993.[CrossRef][Medline]
39. Boland S, Boisvieux-Ulrich E, Houcine O, Baeza-Squiban A, Pouchelet M, Schoevaert D, Marano F. Tgf beta 1 promotes actin cytoskeleton reorganization and migratory phenotype in epithelial tracheal cells in primary culture. J Cell Sci 1996;109:2207–2219.[Abstract]
40. Masui T, Wakefield LM, Lechner JF, LaVeck MA, Sporn MB, Harris CC. Type beta transforming growth factor is the primary differentiation-inducing serum factor for normal human bronchial epithelial cells. Proc Natl Acad Sci USA 1986;83:2438–2442.[Abstract/Free Full Text]
41. Jetten AM, Shirley JE, Stoner G. Regulation of proliferation and differentiation of respiratory tract epithelial cells by tgf beta. Exp Cell Res 1986;167:539–549.[CrossRef][Medline]
42. Rosendahl A, Pardali E, Speletas M, Ten Dijke P, Heldin CH, Sideras P. Activation of bone morphogenetic protein/smad signaling in bronchial epithelial cells during airway inflammation. Am J Respir Cell Mol Biol 2002;27:160–169.[Abstract/Free Full Text]
43. Warburton D, Schwarz M, Tefft D, Flores-Delgado G, Anderson KD, Cardoso WV. The molecular basis of lung morphogenesis. Mech Dev 2000;92:55–81.[CrossRef][Medline]
44. Park KS, Wells JM, Zorn AM, Wert SE, Laubach VE, Fernandez LG, Whitsett JA. Transdifferentiation of ciliated cells during repair of the respiratory epithelium. Am J Respir Cell Mol Biol 2006;34:151–157.[Abstract/Free Full Text]
45. Mucenski ML, Nation JM, Thitoff AR, Besnard V, Xu Y, Wert SE, Harada N, Taketo MM, Stahlman MT, Whitsett JA. Beta-catenin regulates differentiation of respiratory epithelial cells in vivo. Am J Physiol Lung Cell Mol Physiol 2005;289:L971–L979.[Abstract/Free Full Text]
46. Mucenski ML, Wert SE, Nation JM, Loudy DE, Huelsken J, Birchmeier W, Morrisey EE, Whitsett JA. Beta-catenin is required for specification of proximal/distal cell fate during lung morphogenesis. J Biol Chem 2003;278:40231–40238.[Abstract/Free Full Text]
47. Garnis C, Campbell J, Davies JJ, Macaulay C, Lam S, Lam WL. Involvement of multiple developmental genes on chromosome 1p in lung tumorigenesis. Hum Mol Genet 2005;14:475–482.[Abstract/Free Full Text]
48. 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]
49. Kim HJ, Nel AE. The role of phase ii antioxidant enzymes in protecting memory t cells from spontaneous apoptosis in young and old mice. J Immunol 2005;175:2948–2959.[Abstract/Free Full Text]
50. White SR, Dorscheid DR, Rabe KF, Wojcik KR, Hamann KJ. Role of very late adhesion integrins in mediating repair of human airway epithelial cell monolayers after mechanical injury. Am J Respir Cell Mol Biol 1999;20:787–796.[Abstract/Free Full Text]
51. Richter A, O'Donnell RA, Powell RM, Sanders MW, Holgate ST, Djukanovic R, Davies DE. Autocrine ligands for the epidermal growth factor receptor mediate interleukin-8 release from bronchial epithelial cells in response to cigarette smoke. Am J Respir Cell Mol Biol 2002;27:85–90.[Abstract/Free Full Text]
52. 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;272:6387–6399.[CrossRef][Medline]
53. Rosenberg M, Chaudhury A, Shows TB, Le Beau MM, Fuchs E. A group of type i keratin genes on human chromosome 17: Characterization and expression. Mol Cell Biol 1988;8:722–736.[Abstract/Free Full Text]
54. Matsuura H, Momota Y, Murata K, Matsushima H, Suzuki N, Nomizu M, Shinkai H, Utani A. Localization of the laminin alpha4 chain in the skin and identification of a heparin-dependent cell adhesion site within the laminin alpha4 chain c-terminal lg4 module. J Invest Dermatol 2004;122:614–620.[CrossRef][Medline]
55. Meng X, Klement JF, Leperi DA, Birk DE, Sasaki T, Timpl R, Uitto J, Pulkkinen L. Targeted inactivation of murine laminin gamma2-chain gene recapitulates human junctional epidermolysis bullosa. J Invest Dermatol 2003;121:720–731.[CrossRef][Medline]
56. Smith SL, Watson SG, Ratschiller D, Gugger M, Betticher DC, Heighway J. Maspin—the most commonly-expressed gene of the 18q21.3 serpin cluster in lung cancer—is strongly expressed in preneoplastic bronchial lesions. Oncogene 2003;22:8677–8687.[CrossRef][Medline]
57. Steinert PM, Marekov LN. Initiation of assembly of the cell envelope barrier structure of stratified squamous epithelia. Mol Biol Cell 1999;10:4247–4261.[Abstract/Free Full Text]
58. Leube RE, Bader BL, Bosch FX, Zimbelmann R, Achtstaetter T, Franke WW. Molecular characterization and expression of the stratification-related cytokeratins 4 and 15. J Cell Biol 1988;106:1249–1261.[Abstract/Free Full Text]
59. Takeuchi T, Liang SB, Matsuyoshi N, Zhou S, Miyachi Y, Sonobe H, Ohtsuki Y. Loss of t-cadherin (cdh13, h-cadherin) expression in cutaneous squamous cell carcinoma. Lab Invest 2002;82:1023–1029.[Medline]
60. Said JW, Nash G, Sassoon AF, Shintaku IP, Banks-Schlegel S. Involucrin in lung tumors: a specific marker for squamous differentiation. Lab Invest 1983;49:563–568.[Medline]
61. Martinet N, Bonnard L, Regnault V, Picard E, Burke L, Siat J, Grosdidier G, Martinet Y, Vignaud JM. In vivo transglutaminase type 1 expression in normal lung, preinvasive bronchial lesions, and lung cancer. Am J Respir Cell Mol Biol 2003;28:428–435.[Abstract/Free Full Text]

This article has been cited by other articles:

				J.-A. Park and D. J. Tschumperlin Chronic Intermittent Mechanical Stress Increases MUC5AC Protein Expression Am. J. Respir. Cell Mol. Biol., October 1, 2009; 41(4): 459 - 466. [Abstract] [Full Text] [PDF]

				L. El Zein, A. Ait-Lounis, L. Morle, J. Thomas, B. Chhin, N. Spassky, W. Reith, and B. Durand RFX3 governs growth and beating efficiency of motile cilia in mouse and controls the expression of genes involved in human ciliopathies J. Cell Sci., September 1, 2009; 122(17): 3180 - 3189. [Abstract] [Full Text] [PDF]

				A. J. Fischer, K. L. Goss, T. E. Scheetz, C. L. Wohlford-Lenane, J. M. Snyder, and P. B. McCray Jr. Differential Gene Expression in Human Conducting Airway Surface Epithelia and Submucosal Glands Am. J. Respir. Cell Mol. Biol., February 1, 2009; 40(2): 189 - 199. [Abstract] [Full Text] [PDF]

				A. Kubo, A. Yuba-Kubo, S. Tsukita, S. Tsukita, and M. Amagai Sentan: A Novel Specific Component of the Apical Structure of Vertebrate Motile Cilia Mol. Biol. Cell, December 1, 2008; 19(12): 5338 - 5346. [Abstract] [Full Text] [PDF]

				C. Coraux, J. Roux, T. Jolly, and P. Birembaut Epithelial Cell-Extracellular Matrix Interactions and Stem Cells in Airway Epithelial Regeneration Proceedings of the ATS, August 15, 2008; 5(6): 689 - 694. [Abstract] [Full Text] [PDF]

				K. Pawlowski Uncharacterized/hypothetical proteins in biomedical 'omics' experiments: is novelty being swept under the carpet? Brief Funct Genomic Proteomic, July 19, 2008; (2008) eln033v1. [Abstract] [Full Text] [PDF]

				B. Chhin, J. T. Pham, L. El Zein, K. Kaiser, O. Merrot, and P. Bouvagnet Identification of transcripts overexpressed during airway epithelium differentiation Eur. Respir. J., July 1, 2008; 32(1): 121 - 128. [Abstract] [Full Text] [PDF]

				F. A. Barnes, L. Bingle, and C. D. Bingle Pulmonary Genomics, Proteomics, and PLUNCs Am. J. Respir. Cell Mol. Biol., April 1, 2008; 38(4): 377 - 379. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Online Supplement All Versions of this Article: 2006-0466OCv1 37/2/169    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ross, A. J. Articles by Devlin, R. B. Search for Related Content PubMed PubMed Citation Articles by Ross, A. J. Articles by Devlin, R. B.