AJP - Lung Email Content Delivery
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Lung Cell Mol Physiol 292: L1248-L1256, 2007. First published January 12, 2007; doi:10.1152/ajplung.00290.2006
1040-0605/07 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figure and Tables
Right arrow All Versions of this Article:
292/5/L1248    most recent
00290.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maunders, H.
Right arrow Articles by Richter, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maunders, H.
Right arrow Articles by Richter, A.

Human bronchial epithelial cell transcriptome: gene expression changes following acute exposure to whole cigarette smoke in vitro

Heather Maunders,1 Sudhanshu Patwardhan,1 Jeremy Phillips,2 Aaron Clack,1 and Audrey Richter2

1Advanced Technologies Cambridge, Cambridge; and 2Group R&D, British American Tobacco, Southampton, England, United Kingdom

Submitted 31 July 2006 ; accepted in final form 3 January 2007


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cigarette smoke is a complex mixture of more than 4,000 constituents. Its effects on cell biology are poorly understood, partly because whole smoke exposure in vitro is technically challenging. To investigate the effects of smoke on cell signaling and function, a three-dimensional air-liquid interface model of tracheobronchial epithelium, grown from primary human lung epithelial cells, was exposed to air or whole mainstream cigarette smoke for 1 h in a purpose-designed chamber. Gene expression profiles were then determined at 1, 6, and 24 h postexposure using Affymetrix HGU133-2 Plus microarrays. Cells from three different donors were used in the study, and the experiment was performed in triplicate for each donor. Genes significantly regulated by smoke, compared with the air control, in all experiments were determined. Genes exhibiting differential expression were assigned to functional categories and mapped to signaling pathways. Effects were observed on many cellular processes including xenobiotic metabolism, oxidant/antioxidant balance, and DNA damage and repair. Notably, there was marked downregulation of the transforming growth factor-beta pathway, which has not been previously reported. This study provides important data on the acute effects of whole cigarette smoke on mucociliary epithelium and may be used to gain a greater understanding of smoke toxicity.

bronchial epithelium; microarray

ALTHOUGH ASSOCIATIONS BETWEEN cigarette smoking and diseases such as lung cancer and chronic obstructive pulmonary disease (COPD) are well documented, surprising little is known about the mechanistic basis of smoking-related disease at the cellular level. This is due, in part, to the fact that cigarette smoke is a complex and dynamic mixture of more than 4,000 individual chemical constituents (4). Cigarette smoke has been shown to have multiple effects on gene expression in the human airways: microarray studies of bronchial epithelial cells obtained from the airway of smokers and nonsmokers by bronchial brushing have indicated that cigarette smoke induces, primarily, the expression of xenobiotic-metabolizing and redox-regulating genes but also point to effects on tumor suppressor genes, oncogenes, and genes involved in the regulation of inflammation (14, 32).

The gene expression changes observed in the bronchial epithelium of healthy smokers may be modulated by multiple parameters such as lifestyle, genetic variability, and subclinical disease including mild inflammation. To dissect out the direct effects of smoke and to begin to understand the cellular basis of smoke toxicity, it is necessary to perform in vitro studies. But, due to the technical challenges posed by whole smoke exposure, aqueous extracts of smoke (consisting of a buffered aqueous solution through which smoke has been bubbled) or total particulate matter (also known as cigarette smoke condensate) are generally used as surrogates for whole cigarette smoke. Although such studies have increased our knowledge of smoke toxicity and have pointed to oxidative stress as a major contributor to cellular damage (20), both preparations contain only a subset of total smoke constituents, and it is difficult to equate levels of exposure in vitro to dosimetry in smokers.

The selection of a cell line for a bronchial epithelial model is controversial. The most common model, the A549 tumor cell line of type II alveolar cell origin, for instance, fails to form the intracellular tight junctions found in normal alveolar epithelium. Whereas the practical properties of transformed or tumor cell lines make them useful for mechanistic studies, air-liquid interface cultures of primary human bronchial epithelial cells are generally considered to be the best available model. They also provide an opportunity to investigate the effects of whole smoke, which can be applied to the apical surface of the cultures using specialized smoke generation and exposure equipment (2, 6, 26). For example, using this, model Beisswenger et al. (6) found that whole cigarette smoke induced a proinflammatory response in differentiated human bronchial epithelial cells, as demonstrated by the upregulation of IL-6 and IL-8, and that the response was mediated by the transcription factor NF-{kappa}B.

In this study, we have constructed a model of human bronchial epithelial exposure to cigarette smoke using mucociliary, air-liquid interface cultures and have profiled global changes in gene expression using high-density microarrays. Differentiated cultures of low-passage bronchial epithelial cells, with similar morphology to human tracheobronchial epithelium, were prepared using methods developed by Gray et al. (13). These were treated with mainstream cigarette smoke in a purpose-designed exposure system (26), and changes in gene expression up to 24 h postexposure were profiled using whole genome Affymetrix microarrays. The data provide a comprehensive picture of the cell's transcriptomic response to acute cigarette smoke exposure and represents progress in the understanding of cigarette smoke toxicity at the cellular level.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell culture. Primary human bronchial epithelial cells from nonsmokers were purchased from Cambrex (Walkersville, MD). The protocol for cell differentiation was adapted from Gray et al. (13). Cells at passage 3 were seeded onto cell culture inserts (Transwell-Clear, 6.5-mm diameter, 0.4-µM pore size; Corning) at 7.5 x 104 cells per insert in a volume of 105 µl of bronchial epithelial growth medium (BEGM, Cambrex). The following day, the apical BEGM was removed such that the cells were at an air-liquid interface. The basal medium was replaced with 300 µl of 50% bronchial epithelial basal medium (BEBM, Cambrex) in DMEM (vol:vol) containing 0.4% (vol:vol) bovine pituitary extract, 5 µg/ml insulin, 75 ng/ml hydrocortisone, 10 µg/ml transferrin, 6.5 ng/ml thyroxine, 0.5 µg/ml epinephrine, 0.5 ng/ml epidermal growth factor, 15 ng/ml retinoic acid, and GA-1000. The medium was replaced daily (Monday through Friday) for the following 23 days, during which time the cultures differentiated to form a mucociliary epithelium, as observed by light microscopy.

Experimental design. Cell cultures from three human donors (Table 1) were exposed to diluted mainstream cigarette smoke or filtered air for 1 h. RNA extractions were made either immediately after exposure or following a 6- or 24-h recovery period, during which cultures were incubated at 37°C, 5% CO2. A "no treatment" (NT) control received neither smoke nor filtered air; these cultures remained in the incubator before RNA extraction. Four replicate inserts were used for each treatment condition/time point, and cell lysates from these were pooled before RNA extraction. To perform robust statistical analyses, three independent replicate experiments were performed on each of the three donors (using commercial human cell lines purchased from Cambrex), using a separate stock of cells for each experiment.


View this table:
[in this window]
[in a new window]

  		Table 1. Details of the three primary human lung epithelial cell donors

 
Whole smoke exposure. The cell cultures were exposed to mainstream cigarette smoke from 2R4F University of Kentucky reference cigarettes using a purpose-designed exposure system (26). Smoke was drawn from the cigarettes under the International Organization for Standardization standard conditions (35-ml puff drawn over 2 s every 1 min) and diluted in filtered air using an RM20s smoke engine (Borgwaldt Technik, Hamburg, Germany). Diluted smoke (1/50 smoke:air, vol/vol) was continually delivered to exposure chambers (UK patent number WO 03/100417 A1) (Fig. 1) containing the culture inserts for a period of 1 h. During this time, smoke from ~10 cigarettes was delivered to each chamber, and, in the absence of cells, the total deposition of particulates on the base of the cell culture inserts was determined to be 1.84 µg/cm2. In comparison, analysis of data from a human cigarette smoke inhalation study, using a radioactive tracer incorporated into a cigarette, reported regional particulate deposition efficiency from the equivalent of a single cigarette in the extrathoracic, bronchial/bronchiolar, and pulmonary (respiratory bronchioles and alveoli) regions of the lung to be 5%:31%:64% for males and 13%:31%:56% for females (22). This equates to deposition levels in the human lung of: extrathoracic, females 0.74 µg/cm2, males 1.94 µg/cm2; bronchial/bronchiolar, males and females 0.81 µg/cm2 (J. McAughey, personal communication). These calculations assume a 10-mg delivery cigarette with an average of 70% total particulate retention. Surface areas of 470 and 2,690 cm2 were used for the extrathoracic and bronchial/bronchiolar lung regions (1). Air control cells were exposed to filtered air alone. Following smoke treatment, cells for the 6- and 24-h time points were transferred to 24-well trays with 300 µl of basal serum-free medium (Ultraculture, Cambrex), returned to the cell culture incubator, and maintained at 37°C. Immediately after smoke exposure, or at the selected time points, cultures were solubilized in 150 µl of TRIzol (Invitrogen, Paisley, Scotland) per insert, and samples from replicate inserts were pooled.


Figure 1
View larger version (52K):
[in this window]
[in a new window]

  		Fig. 1. A: a smoke exposure chamber. B: the mechanistic characteristics of the exposure chamber.

 
Neutral red cytotoxicity assay. The neutral red uptake protocol was based on guidelines set by the National Institutes of Health (25). Briefly, the medium was removed from the inserts and replaced with neutral red solution (0.05 g/l in Ultraculture). The cultures were incubated with neutral red solution for 3 h at 37°C to allow for active uptake of the dye into viable cells. The inserts were then washed thoroughly with PBS to ensure complete removal of unincorporated dye. The neutral red was eluted from the cells by placing 250 µl of distilled water containing 50% ethanol and 1% acetic acid (vol:vol) into each insert. Aliquots of the eluates (100 µl) were read on a microplate spectrophotometer at 540 nm using a reference filter of 620 nm. Blank inserts were included in the study to establish background levels of dye that were adsorbed onto the insert matrix in the absence of cells. These background measurements were subsequently subtracted from those of the treated and untreated control cultures. Neutral red uptake from each treatment was compared with that of the air control to calculate the percentage cell viability relative to the control.

Determination of transepithelial electrical resistance. At the 24-h time point, 100 µl of Ultraculture was added apically to each insert, and the transepithelial electrical resistance (TEER) was measured using a handheld voltmeter (World Precision Instruments, Aston, UK) to determine the integrity of the epithelial cultures.

Immunocytochemistry. Representative cultures of cells were trypsinized, and cytospin slides were prepared. The cells were stained for the mucin MUC5AC to identify the goblet cells using an indirect immunostaining technique and the murine monoclonal antibody 45M1 (Neomarkers, Fremont, CA) at 2 µg/ml. The percentage of goblet cells in the population was determined by counting the number of positively stained cells per hundred in a total of ~400–500 cells counted.

RNA isolation and preparation. Total RNA was isolated using TRIzol according to the manufacturer's instructions, followed by purification using a Qiagen RNeasy Mini Kit (Qiagen, West Sussex, UK). The integrity of the RNA was assessed using a 2100 Bioanalyzer (Agilent, Palo Alto, CA). cRNA probes were prepared from 1 µg of total RNA using an Affymetrix One Cycle cDNA synthesis kit. cDNA was hybridized to the HGU133-2 plus whole genome GeneChip containing 66,577 transcripts.

Microarray data acquisition and processing. The quality of expression data from the chips was assessed using Affymetrix Microarray Suite (version 5.0). Chips were visually inspected for artifacts that may have been related to hybridization, washing, or scanning followed by a technical parameter screen using Affymetrix software Quality Reporter. Noise, average background signal, average signal intensity, scaling factor, percent of genes detected, bioB, bioC, bioD, Cre, and 3'/5' signal ratio of housekeeping genes (GAPDH and actin) were evaluated to identify parameters falling outside the quality acceptance criteria (30). Chips were allocated a pass or flag for further inspection according to this evaluation. Microarray data have been submitted to the European Bioinformatics Institute ArrayExpress repository (http://www.ebi.ac.uk/arrayexpress, acc. no. E-TABM-127).

GeneChip data scaled to a target intensity of 100 using the Affymetrix MAS 5.0 algorithm were also assessed in GeneSpring 7.2 (Agilent) using hierarchical clustering based on Spearman (nonparametric) correlations to group arrays with similar expression profiles and validate those chips flagged for further inspection. All 90 microarrays from samples in this study passed the quality acceptance criteria, and data were used for expression analysis. (Supplementary data for this article is available online at the AJP-Lung web site.)

Quantitative PCR. One microgram of total RNA was reverse transcribed into cDNA using random hexamers as a template. Real-time PCR reactions were carried out on an iCycler (BioRad, Hemel Hempstead, Herts, UK) using iQ Supermix, and each sample was tested in triplicate. Threshold values for target genes were normalized to 18S, and quantitative PCR (Q-PCR) data were calculated using the {Delta}{Delta}CT method (19). Primers and double-dye probes for CYP1A1 and 18S ribosomal RNA were purchased from Primer Design (Southampton, Hants, UK). All other primer/probe sets were purchased from Applied Biosystems (Foster City, CA).

Statistical analysis. Raw unscaled data were preprocessed in GeneSpring 7.2 (Agilent) using the Robust Multichip Average, a method for normalizing and summarizing probe-level intensity measurements from Affymetrix GeneChips (7, 17). Starting with the probe-level data from a set of GeneChips, the perfect-match values were background corrected, normalized, and finally summarized, resulting in a set of expression measures. Per gene normalization was applied to these values, dividing the data by the median of the expression level for the gene across all samples (16).

Genes showing no change in global expression were removed from further analyses. A t-test, assuming unequal variance, P value cutoff of <0.05 with Benjamini Hochberg multiple testing corrections for false discovery rate, was applied to the remaining 24,310 genes. Those genes showing significant differential expression between the nine (3 donors x 3 replicate assays) air control samples and the nine smoke treatment samples at each of the 1-, 6-, and 24-h time points were determined. To reduce loss of subtle changes in gene expression, a low-stringency filter of 1.5-fold up- or downregulation was applied. The final number of differentially expressed genes is shown in Table 2.


View this table:
[in this window]
[in a new window]

  		Table 2. The number of genes significantly regulated in response to smoke compared with the air control

 
Principal component analysis on conditions was carried out for those genes showing expression changes using GeneSpring 7.2 (Agilent). This approach was used to reduce data dimensionality and visualize global gene expression patterns. The time- and donor-specific effects were also captured using hierarchical clustering with standard correlation (12).

Functional analyses. Many bioinformatics tools are now available online to map the genes from gene lists to Gene Ontology (GO) categories they represent and then calculate the statistics for whether these categories are overrepresented in the gene lists compared with their occurrence on the chip. The GOstat tool (http://gostat.wehi.edu.au/cgi-bin/goStat.pl) (5) was used to calculate the overrepresentation statistics for GO categories. Lists of differentially expressed genes from the 1/50 dose and air treatment comparisons were used as input into the tool. Each gene list was loaded in the Group IDs box, and the goa_human GO gene association database for Affymetrix HG-U133 plus 2.0 was selected as the reference. The minimal length of GO paths was 3; the GO super category selected was "Biological Process," with a prefilter of P value less than 0.1 for the output. The top 30 categories in the output (supplementary data) were manually inspected for overlap of member genes, and a representative set of nonredundant categories was used for biological interpretation.

To determine significantly regulated pathways, lists of differentially expressed genes between the 1/50 dose and air control at each time point (Table 2) were mapped to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (http://www.genome.jp/kegg/). The lists were uploaded to the Database for Annotation, Visualization, and Integrated Discovery (DAVID) website (http://david.niaid.nih.gov/david/ease.htm) where DAVID beta 2.1 functional annotation tool identified KEGG pathways showing enrichment for genes in the lists and the probability of that pathway being significantly altered. Fisher exact test (maximum probability 0.1) was used to determine enrichment probability. Output charts were obtained for the distribution of significantly expressed genes in those KEGG pathways (supplementary data).


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The differentiated cultures of primary human bronchial epithelial cells used in this study had a similar morphology to human tracheobronchial epithelium (13) (Fig. 2). For whole smoke exposure, cells were obtained from three donors (Table 1), and experiments were performed on each donor on three separate occasions. To ensure relative consistency of the cultures before smoke exposure, the number of goblet cells in representative cultures, which can be identified by immunocytochemical staining, was determined by using an antibody to the mucin MUC5AC. The percentage of goblet cells in the cultures was as follows: donor A, 5 ± 1; donor B, 8 ± 6; donor C, 8 ± 6 (means ± SD of combined data from 3 independent experiments).


Figure 2
View larger version (101K):
[in this window]
[in a new window]

  		Fig. 2. A and B: scanning and transmission electron micrographs, respectively, of differentiated human lung epithelial cell cultures. These consist of a complex columnar epithelium containing basal cells, goblet cells with apical microvilli, and ciliated cells (arrows in B).

 
The cultures were exposed to filtered air or mainstream smoke diluted in filtered air. An NT control was also included in each experiment to determine the effect of the exposure process. Previous neutral red cell viability assays had established that the ID50 of whole cigarette smoke in this model system was ~1/30 smoke:air (vol/vol), and negative effects on cell viability were observed at doses greater than 1/50 at 24 h postexposure (data not shown). By comparison, the TEER of the cultures, which is a measure of the integrity of the epithelial tight junctions and permeability, was more sensitive to smoke, and reductions in TEER were observed at doses greater than 1/100. Therefore, a dose of 1/50 smoke:air (minimally toxic, expected reductions in TEER but not cell viability) was chosen for this study.

The cultures were exposed to 1/50 dose of whole smoke or air continually for 1 h. The consistency of the effect of smoke exposure across experiments was monitored at the 24-h time point by measuring cell viability through the uptake of neutral red dye and TEER. This was carried out on a parallel set of exposed cultures that had been treated in the same manner as the cultures for microarray analysis. No significant difference was observed in cell viability between the smoke-treated air control and NT control cultures, as expected. However, for TEER data, the mean values of measurements from three donors and three assays for 1/50 whole smoke was 901 {Omega}/cm2 SD ± 824, air control was 3,203 {Omega}/cm2 SD ± 1,160, and NT control was 3,250 {Omega}/cm2 SD ± 1,090, which highlighted a difference between treatments. ANOVA for this data showed a significant difference (P < 0.00001) between the groups, and Tukey's simultaneous confidence intervals (significant at a 5% level) defined the smoke-treated cultures as different from the NT control and the air control. This suggested that the reduction in TEER following smoke treatment was not due to a decrease in cell viability but to a breakdown of the intracellular tight junctions and an increase in epithelial permeability.

Gene expression analysis was performed for the 1-h (immediately after exposure), the 6-h, and the 24-h time points. In this study, the NT control was not transferred to the exposure chamber at any point but served as a reference to ensure that the exposure procedure alone (in the absence of smoke) was not having a significant effect on global gene expression.

Principal component analysis on conditions demonstrated that cigarette smoke exposure induced time-specific responses in global gene expression, with 63.7% of the variance lying within the first three principal components. Principal components 1, 2, and 3 identified a clear separation of the smoke-treated samples at 1, 6, and 24 h from the air and NT controls (Fig. 3). The three donors showed minimal variation within their time point group, as highlighted by hierarchical clustering of significantly expressed genes (supplementary data). There was no separation of the NT control and air controls at the 1-, 6-, and 24-h time points, confirming that the exposure procedure itself had minimal effects on gene expression. The data from the NT control was not used in any further analyses.


Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 3. Principal component analysis of global gene expression. The data points for the triplicate air controls and no treatment (NT) control at 1 and 24 h form a single cluster, whereas the data points for 1/50 (smoke in air) cigarette smoke treatment at time points of 1, 6, and 24 h form 3 distinct clusters, which are separate from the large cluster of controls highlighting the time-dependant effect of smoke on gene expression. Color bar at right: C, air control; NT, no treatment control; 1/50, 1/50 smoke in air.

 
Microarray data from the smoke-treated cultures were then analyzed to identify genes showing differential expression from the air control for each corresponding time point (supplementary data). Some of the largest fold changes in expression, as described in Table 3, were for genes involved in oxidative metabolism, response to stimuli or cell cycle regulation and the response to DNA damage, such as heat shock protein 70 kDa (HSP70B), cytochrome P-450s CYP1A1, CYP1B1 (both of which are involved in bioactivation of polyaromatic hydrocarbons) and CYP26A1, hemoxygenase 1 (HMOX1), thioredoxin reductase 1 (TXNRD1), and DNA-damage-inducible transcript 3 (DDIT3). When further functional analyses were applied to the smoke response using GOstat, which ascribes a biological process to the expressed genes, there were distinct significantly overrepresented categories of genes for each time point. A summary of selected categories for each time point is shown in Table 4 and 5.


View this table:
[in this window]
[in a new window]

  		Table 3. Selected genes showing some of the largest fold changes in expression from the air control in response to smoke

 

View this table:
[in this window]
[in a new window]

  		Table 4. Top overrepresented categories of genes downregulated by 1/50 smoke compared with the air exposure as determined by GOstat

 

View this table:
[in this window]
[in a new window]

  		Table 5. Top overrepresented categories of genes upregulated by 1/50 smoke compared with the air exposure as determined by GOstat

 
At 1 h, 91% of 4,038 differentially expressed genes were downregulated compared with the air control. Those involved in transcription, protein modification, cell cycle, and negative regulation of physiological process were significantly overrepresented (Table 4). Upregulated genes were overrepresented in the RNA processing, protein biosynthesis, and nucleic acid metabolism categories. The relatively high number of downregulated genes may signify a shut down of many cellular processes immediately following the toxic insult. Pathways significantly downregulated include many controlling cell junctions, signaling pathways, and the cell cycle. Oxidative phosphorylation necessary for cellular antioxidant function was one of the three upregulated pathways at this time point (supplementary data).

At 6 h, the majority of differentially expressed genes were downregulated compared with the air control (74%), but there were significant changes in expression pointing to a renewal of cellular activity (Tables 4 and 5). The responses to unfolded protein and stress were the most significantly overrepresented upregulated categories and represent an attempt to recover from exposure to cigarette smoke. It suggests the targeting of damaged, unfolded proteins for ubiquitinization and proteasomal degradation (36). A third of genes upregulated in the cell cycle category were negative regulators of cell cycle progression indicating that cell cycle suppression was occurring and possibly enabling DNA repair or preceding apoptosis. Despite an upregulation of protein kinase activity, the downregulation of genes in protein amino acid phosphorylation, regulation of cellular processes, protein modification, and DNA-dependent transcription categories at this time point suggested that many cellular functions were still in stasis while damage repair processes were carried out. The pathways regulated (supplementary data) largely substantiate the findings from the GOstat categories with negative regulation of pathways controlling cell junctions, signaling pathways, and cell cycle. Upregulated pathways include MAPK signaling, cell cycle, and DNA polymerase required for transcription.

At 24 h, 53% of 5,355 significantly differentially expressed genes were upregulated compared with the air control. Again, genes involved in regulation of progression through the cell cycle were significantly upregulated, almost a third of which were known to induce cell cycle arrest or negatively regulate proliferation. However, there were fewer DNA damage response genes at this time point. The upregulation of genes that initiate apoptosis suggested that excessive and irreparable DNA damage had occurred in some cells. Furthermore, the significant downregulation of the Wnt receptor signaling pathway, which regulates development and proliferation through beta-catenin, supported the suppression of proliferation in favor of apoptosis. But the continued upregulation of response to unfolded protein category suggested that repair processes were still ongoing. The overrepresentation of genes in the RNA processing category, a vital genetic activity required for translational initiation and active protein synthesis, point to the initiation of new protein biosynthesis. Pathways regulated at this time point indicate that the downregulation of the pathways controlling cellular junctions had returned to normal in all but the adherens junctions and that regulation of signaling pathways was confined to the downregulation of Wnt signaling and phosphatidylinositol signaling (involved in maintaining focal adhesion). Downregulation of several metabolic pathways at 24 h supported the suppression of proliferation in favor of apoptosis, although the cell viability data were inconsistent with smoke-induced cell death, and the highly significant upregulation of the proteasome pathway highlighted controlled protein degradation.

Pathway mapping also suggested marked downregulation of the transforming growth factor-beta (TGF-beta) pathway (GoGoFig. 6). This was confirmed by Q-PCR analysis of samples from 7/9 of the original experiments (insufficient RNA was available from the 2 remaining experiments). Q-PCR data were calculated using the {Delta}{Delta}CT method, and the level of gene expression for each target in 6-h samples from the smoke-treated cells was expressed in arbitrary units relative to that in the corresponding 6-h air control samples that were given a value of 1. CYP1A1 was included as a control. The data from the smoke-treated and air control samples were compared using a one-sample t-test. As shown in Table 6, the downregulation of genes in the TGF-beta pathway and upregulation of CYP1A1 was confirmed by PCR with the exception of SMAD3, where the change relative to the air control was not significant.


Figure 4
View larger version (43K):
[in this window]
[in a new window]

  		Fig. 4. Summary of the response of human bronchial epithelial cells at 1-, 6-, and 24-h time points, to a 1/50 dose of whole smoke. Blue boxes, KEGG pathways regulated (supplementary data); yellow boxes, Gene Ontology biological processes significantly regulated (GOstat); green boxes, the cellular responses to smoke.

 

Figure 5
View larger version (38K):
[in this window]
[in a new window]

  		Fig. 5. Summary of the effects of 1/50 smoke on the MAPK pathway derived from the 24-h time point data.

 

Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 6. Summary of the effects of 1/50 smoke on the transforming growth factor-beta (TGF-beta) pathway at the 6-h time point.

 

View this table:
[in this window]
[in a new window]

  		Table 6. Quantitative PCR determination of gene expression at 6 h for selected genes regulating the TGF-beta pathway

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this study, a 1/50 dose of cigarette smoke had significant effects on epithelial gene expression. In the early phase of the smoke response, marked reductions in gene expression suggested a shut down of many cellular processes. In addition, there was marked downregulation of many genes involved in the formation of tight junctions such as occludin, tight junction protein 1 (TJP1 or ZO1), and claudin-1 (CLDN1). Tight junctions, the most apical intercellular epithelial junctions, selectively regulate passive electrolyte transport across the paracellular space. Evidence suggests that tight junctions are also involved in basic cellular processes like the regulation of cell growth and differentiation (3, 21) and the recovery from toxic injury (11). Downregulation of tight junction proteins was consistent with the reduction in TEER observed at 24 h, for example, as a direct relationship between the amount of expressed occludin and barrier TEER has been previously established (15).

Cigarette smoking is reported to be the major risk factor for COPD, and oxidative stress is implicated in disease development (9, 28). The redox state of cells plays an important part in the regulation and potentiation of the inflammatory response. Evidence suggests that respiratory epithelial cells respond to cigarette smoke by enhancing the activity of the glutathione system and downregulating catalases that play a critical role in the prevention of damage by lung toxicants (27). The glutathione pathway is modulated by the negative feedback exerted by glutathione on glutaminecystiene synthase and the glutathione redox system (14). The antioxidant response was evident in this study, at one or more time points, as demonstrated by the regulation of superoxide dismutase, glutathione metabolism genes such as glutathione reductase (GSR) and glutathione S-transferase (GSTA2), and the increased expression of redox balance genes TXN, TXNRD1, NAD(P)H-dehydrogenase quinone 2 (NQO2). Hackett et al. (14) reported that 16 antioxidant genes were upregulated in airway epithelium from healthy smokers compared with nonsmokers, supporting the hypothesis that the pathogenesis of COPD is associated with oxidative stress resulting from exposure to constituents of smoke and/or oxidative species produced by inflammatory cells in the respiratory tract. Of those 16 genes, this study shows that 7 were similarly upregulated by smoke alone in vitro.

Cellular DNA is subject to many chemical alterations such as breakages, mismatches, and cross-links, which, in many cases, can be corrected by endogenous repair processes. Interestingly, the tumor suppressor gene and transcription factor p53, which is activated by DNA double strand breaks and arrests the cell cycle to enable DNA repair, is downregulated by smoke at all time points in this study. However, the p53-inducible genes, growth arrest and DNA damage-45 B (GADD45-B), also known to inhibit proliferation (38) and have involvement in nucleotide excision repair (31), were highly upregulated at 6 and 24 h. Several DNA repair genes were upregulated by smoke, including the nuclear and mitochondrial DNA polymerases POLB and POLG. The data suggested that DNA repair was taking place at 6 and 24 h, possibly aided by suppression of cell cycle progression and the induction of GADD45 by DNA-damaging agents. There appeared to be an initiation of an apoptotic response indicative of irreparable damage in some cells; however, cell viability was not reduced at 24 h compared with the air control. Therefore, either the apoptotic response did not go to completion or cell death occurred later than 24 h postexposure. Although proapoptotic signaling was evident, there were no apparent effects on genes in the caspase cascade or downregulation of cytochrome c and no expression of apoptotic peptidase activating factor 1 (APAF1) and caspase-9 (CASP9), required for apoptosome formation, at any time point. Key features of the smoke response are summarized in Fig. 4.

The gene lists for the 1/50 dose (all time points and normalized to the air control) were mapped to signaling pathways using GeneSpring 7.2 and DAVID. Consistent with the reported effects of smoke exposure on stress signaling (27), the MAPK pathway had a large number of significantly regulated genes. This was substantiated by the GOstat overrepresented categories at 6 and 24 h that included significant upregulation of protein kinase regulation genes. The activation of the MAPK pathways, through environmental stress stimuli, reactive oxygen species, cytokines, and chemokines, leads to the activation of transcription factors involved in chromatin remodeling, the promotion of the proinflammatory response, and the regulation of genes involved in cell cycle and apoptosis. All three major MAPK pathways had a number of genes significantly differentially regulated in response to smoke at 24 h (Fig. 5). The p38 pathway and the c-jun NH2-terminal kinase pathway (JNK) activated by inflammatory cytokines and oxidative stress are associated with induction of apoptosis. c-Jun, which was upregulated fivefold at 6 and 24 h, is known to be significantly regulated in the lungs of smokers and is a component of the activator protein 1 (AP1) transcription complex, responsible for the control of many cytokine genes and essential for cellular proliferation and differentiation (29). Jun D is a regulatory protein for the transcription of antioxidant genes such as ferritin H, upregulated at 24 h, which may confer protection from oxidative stress (37). The ERKs are typically activated by transmembrane growth factor receptors with tyrosine kinase activity. c-Fos, which was highly upregulated by smoke at 6 and 24 h, can dimerize with members of the Jun family to form the AP1 complex. This pathway has been associated with cell survival and proliferation and linked with COPD through upregulation of ligands for the epidermal growth factor receptor, including amphiregulin (AR) and heparin binding epidermal growth factor-like growth factor (HB-EGF). In vitro, 1/50 smoke upregulated AR and HB-EGF at one or more time points. The ERK pathway has also been shown to regulate expression of the interstitial collagenase matrix metalloproteinase 1 (MMP-1), the upregulation of which is implicated as an underlying mechanism in the development of emphysema in smokers (23). MMP-1 was highly upregulated by smoke at 6 and 24 h in vitro.

Interestingly, pathway mapping also demonstrated the almost complete downregulation of the TGF-beta pathway (Fig. 6), which was confirmed by Q-PCR. TGF-beta is a potent mediator of fibrosis as it stimulates the production of extracellular matrix proteins, such as collagen, while inhibiting matrix-degrading enzymes including MMPs. It is regulated by a redox-sensitive mechanism (18), and some evidence suggests that the TGF-beta pathway is upregulated in the lungs of smokers and contributes to fibrosis (34). For example, Springer et al. (33) have shown that inhibitors of the TGF-beta pathway (SMAD-6 and SMAD-7) are downregulated in lung epithelial biopsies from COPD patients. Therefore, the general inhibitory effects on TGF-beta signaling observed for the first time in this study would suggest that either the primary source of excess TGF-beta in the lungs of COPD patients is not the epithelium or that regulation of this pathway is not a direct effect of smoke exposure but a consequence of secondary pathological changes in the lung. It is also notable that TGF-beta is considered to act as a tumor suppressor in the early stages of carcinogenesis (35), suggesting that the observed in vitro response to smoke may be relevant to the development of lung tumors.

Spira et al. (32) have published a comprehensive expression profiling study of lung epithelial brushings from smokers, former smokers, and never-smokers that employed Affymetrix microarrays. This identified 97 genes differentially expressed between smokers and never-smokers. The majority of genes significantly expressed in smokers, compared with never-smokers, were involved in the regulation of oxidative stress, glutathione, and xenobiotic metabolism. In contrast, those significantly downregulated were putative tumor suppressor genes and genes regulating inflammation. The characterization of the in vitro acute response to cigarette smoke, also using Affymetrix arrays, is complementary to that of Spira et al. (32). There were some similarities with the results from the clinical study (supplementary data) such as the upregulation of xenobiotic metabolism and redox/antioxidant-related genes. However, there was no significant downregulation of putative tumor suppressor genes with the exception of growth arrest specific 6 (GAS6) and chemokine CXC motif ligand 1 (CXCL1). Yoneda et al. (39) profiled the effects of smoke exposure over time on the HBE1 human bronchial epithelial cell line using a high-density DNA microarray. Their data generated a functional profile with similarities to the data from this study including an early response to DNA damage and repair, regulation of MAPK, and glutathione metabolism involved in maintaining redox balance.

In summary, this is the first study to profile the whole genome response to mainstream cigarette smoke in an air-liquid interface model using mucociliary cultures of human lung epithelial cells. The experimental design of three independent experiments for each of three cell donors generated a data set suitable for robust statistical analysis, and bioinformatics tools enabled the categorization of responsive genes into functional categories followed by pathway analysis. Many direct effects of smoke revealed in this study are consistent with previous reports of in vivo and in vitro smoke toxicity studies, such as increased epithelial permeability, activation of antioxidant responses, and MAPK pathways (8, 14, 24). The observed inhibitory effect of smoke on the expression of genes involved in cellular adhesion provides a possible mechanism for smoke-induced epithelial permeability and may have implications for the development of diseases such as lung cancer (10). But, in addition, the data have revealed potential inhibitory effects of smoke on the TGF-beta signaling pathway.


   ACKNOWLEDGMENTS
 
We thank Linsey Haswell and Stacy Fiebelkorn for technical assistance with in vitro work, Sugnet Gardner and Martin Ward for expert advice on statistical analysis of data, and John McAughey for particulate deposition data.


   FOOTNOTES
 

Address for reprint requests and other correspondence: S. Patwardhan, Advanced Technologies Cambridge, 210 Cambridge Science Park, Milton Road, Cambridge, CB4 OWA, United Kingdom (e-mail: Sudhanshu.Patwardhan{at}atcbiotech.com)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. International Commission on Radiological Protection. Human respiratory tract model for radiological protection. A report of a Task Group of the International Commission on Radiological Protection. Oxford: Pergamon, 1994.
  2. Aufderheide M, Knebel JW, Ritter D. An improved in vitro model for testing the pulmonary toxicity of complex mixtures such as cigarette smoke. Exp Toxicol Pathol 55: 51–57, 2003.[CrossRef][ISI][Medline]
  3. Balda MS, Matter K. Tight junctions. J Cell Sci 111: 541–547, 1998.[Abstract]
  4. Baum SL, Anderson IGM, Baker RR, Murphy DM, Rowlands CC. Electron spin resonance and spin trap investigation of free radicals in cigarette smoke: development of a quantification procedure. Anal Chim Acta 481: 1–13, 2003.[CrossRef][ISI]
  5. Beissbarth T, Speed TP. GOstat: find statistically overrepresented Gene Ontologies within a group of genes. Bioinformatics 20: 1464–1465, 2004.[Abstract/Free Full Text]
  6. Beisswenger C, Platz J, Seifart C, Vogelmeier C, Bals R. Exposure of differentiated airway epithelial cells to volatile smoke in vitro. Respiration 71: 402–409, 2004.[CrossRef][ISI][Medline]
  7. Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics 19: 185–193, 2003.[Abstract/Free Full Text]
  8. Boucher RC, Johnson J, Inoue S, Hulbert W, Hogg JC. The effect of cigarette smoke on the permeability of guinea pig airways. Lab Invest 43: 94–100, 1980.[ISI]
  9. Comhair SA, Erzurum SC. Antioxidant responses to oxidant-mediated lung diseases. Am J Physiol Lung Cell Mol Physiol 283: L246–L255, 2002.[Abstract/Free Full Text]
  10. Dasari V, Gallup M, Lemjabbar H, Maltseva I, McNamara N. Epithelial-mesenchymal transition in lung cancer: is tobacco the "smoking gun"? Am J Respir Cell Mol Biol 35: 3–9, 2006.[Free Full Text]
  11. Denker BM, Nigam SK. Molecular structure and assembly of the tight junction. Am J Physiol Renal Physiol 274: F1–F9, 1998.[Abstract/Free Full Text]
  12. Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci USA 95: 14863–14868, 1998.[Abstract/Free Full Text]
  13. 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 14: 104–112, 1996.[Abstract]
  14. Hackett NR, Heguy A, Harvey BG, O'Connor TP, Luettich K, Flieder DB, Kaplan R, Crystal RG. Variability of antioxidant-related gene expression in the airway epithelium of cigarette smokers. Am J Respir Cell Mol Biol 29: 331–343, 2003.[Abstract/Free Full Text]
  15. Hirase T, Staddon JM, Saitou M, Ando-Akatsuka Y, Itoh M, Furuse M, Fujimoto K, Tsukita S, Rubin LL. Occludin as a possible determinant of tight junction permeability in endothelial cells. J Cell Sci 110: 1603–1613, 1997.[Abstract]
  16. Irizarry RA, Bolstad BM, Collin F, Cope LM, Hobbs B, Speed TP. Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res 31: e15, 2003.[Abstract/Free Full Text]
  17. 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 4: 249–264, 2003.[Abstract]
  18. Jardine H, MacNee W, Donaldson K, Rahman I. Molecular mechanism of transforming growth factor (TGF)-beta1-induced glutathione depletion in alveolar epithelial cells. Involvement of AP-1/ARE and Fra-1. J Biol Chem 277: 21158–21166, 2002.[Abstract/Free Full Text]
  19. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2[-Delta Delta C(T)] method. Methods 25: 402–408, 2001.[CrossRef][ISI][Medline]
  20. Marwick JA, Kirkham P, Gilmour PS, Donaldson K, MacNee W, Rahman I. Cigarette smoke-induced oxidative stress and TGF-beta1 increase p21waf1/cip1 expression in alveolar epithelial cells. Ann NY Acad Sci 973: 278–283, 2002.[Abstract/Free Full Text]
  21. Matter K, Aijaz S, Tsapara A, Balda MS. Mammalian tight junctions in the regulation of epithelial differentiation and proliferation. Curr Opin Cell Biol 17: 453–458, 2005.[CrossRef][ISI][Medline]
  22. McAughey JJ, Pritchard JN, Black A, Hoare GC, Knight DA. Tar retention and regional lung deposition in male and female smokers switching to products with lower tar yields, edited by Swann C and Froggatt Sir P. AEA Technology report AEAT-0159, 65–66, 1991. The Royal Society of Medicine Press. Tobacco Products Research Trust, 1982–1996.
  23. Mercer BA, Kolesnikova N, Sonett J, D'Armiento J. Extracellular regulated kinase/mitogen activated protein kinase is up-regulated in pulmonary emphysema and mediates matrix metalloproteinase-1 induction by cigarette smoke. J Biol Chem 279: 17690–17696, 2004.[Abstract/Free Full Text]
  24. Mossman BT, Lounsbury KM, Reddy SP. Oxidants and signaling by mitogen-activated protein kinases in lung epithelium. Am J Respir Cell Mol Biol 34: 666–669, 2006.[Abstract/Free Full Text]
  25. National Institutes of Health. Guidance document on using in vitro data to estimate in vivo starting doses for acute toxicity, 2001.
  26. Phillips J, Kluss B, Richter A, Massey E. Exposure of bronchial epithelial cells to whole cigarette smoke: assessment of cellular responses. Altern Lab Anim 33: 239–248, 2005.[Medline]
  27. Rahman I. Redox signaling in the lungs. Antioxid Redox Signal 7: 1–5, 2005.[CrossRef][ISI][Medline]
  28. Rahman I. The role of oxidative stress in the pathogenesis of COPD: implications for therapy. Treat Respir Med 4: 175–200, 2005.[CrossRef][Medline]
  29. Rahman I, MacNee W. Role of transcription factors in inflammatory lung diseases. Thorax 53: 601–612, 1998.[Free Full Text]
  30. Ryan MM, Huffaker SJ, Webster MJ, Wayland M, Freeman T, Bahn S. Application and optimization of microarray technologies for human postmortem brain studies. Biol Psychiatry 55: 329–336, 2004.[CrossRef][ISI][Medline]
  31. Smith ML, Ford JM, Hollander MC, Bortnick RA, Amundson SA, Seo YR, Deng CX, Hanawalt PC, Fornace AJ Jr. p53-mediated DNA repair responses to UV radiation: studies of mouse cells lacking p53, p21, and/or gadd45 genes. Mol Cell Biol 20: 3705–3714, 2000.[Abstract/Free Full Text]
  32. Spira A, Beane J, Shah V, Liu G, Schembri F, Yang X, Palma J, Brody JS. Effects of cigarette smoke on the human airway epithelial cell transcriptome. Proc Natl Acad Sci USA 101: 10143–10148, 2004.[Abstract/Free Full Text]
  33. Springer J, Scholz FR, Peiser C, Groneberg DA, Fischer A. SMAD-signaling in chronic obstructive pulmonary disease: transcriptional down-regulation of inhibitory SMAD 6 and 7 by cigarette smoke. Biol Chem 385: 649–653, 2004.[CrossRef][ISI][Medline]
  34. Takizawa H, Tanaka M, Takami K, Ohtoshi T, Ito K, Satoh M, Okada Y, Yamasawa F, Nakahara K, Umeda A. Increased expression of transforming growth factor-beta1 in small airway epithelium from tobacco smokers and patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 163: 1476–1483, 2001.[Abstract/Free Full Text]
  35. Tang B, Vu M, Booker T, Santner SJ, Miller FR, Anver MR, Wakefield LM. TGF-beta switches from tumor suppressor to prometastatic factor in a model of breast cancer progression. J Clin Invest 112: 1116–1124, 2003.[CrossRef][ISI][Medline]
  36. Travers KJ, Patil CK, Wodicka L, Lockhart DJ, Weissman JS, Walter P. Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell 101: 249–258, 2000.[CrossRef][ISI][Medline]
  37. Wilkinson J, Di X, Schonig K, Buss JL, Kock ND, Cline JM, Saunders TL, Bujard H, Torti SV, Torti FM. Tissue-specific expression of ferritin H regulates cellular iron homoeostasis in vivo. Biochem J 395: 501–507, 2006.[CrossRef][ISI][Medline]
  38. Yin F, Bruemmer D, Blaschke F, Hsueh WA, Law RE, Herle AJ. Signaling pathways involved in induction of GADD45 gene expression and apoptosis by troglitazone in human MCF-7 breast carcinoma cells. Oncogene 23: 4614–4623, 2004.[CrossRef][ISI][Medline]
  39. Yoneda K, Peck K, Chang MM, Chmiel K, Sher YP, Chen J, Yang PC, Chen Y, Wu R. Development of high-density DNA microarray membrane for profiling smoke- and hydrogen peroxide-induced genes in a human bronchial epithelial cell line. Am J Respir Crit Care Med 164: S85–S89, 2001.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Eur Respir JHome page
K. F. Chung and I. M. Adcock
Multifaceted mechanisms in COPD: inflammation, immunity, and tissue repair and destruction
Eur. Respir. J., June 1, 2008; 31(6): 1334 - 1356.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figure and Tables
Right arrow All Versions of this Article:
292/5/L1248    most recent
00290.2006v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maunders, H.
Right arrow Articles by Richter, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maunders, H.
Right arrow Articles by Richter, A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2007 by the American Physiological Society.