AJP - Lung Ad Instruments
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Lung Cell Mol Physiol 289: L545-L553, 2005. First published June 3, 2005; doi:10.1152/ajplung.00065.2005
1040-0605/05 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
289/4/L545    most recent
00065.2005v1
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 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 Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zabner, J.
Right arrow Articles by McCray, P. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zabner, J.
Right arrow Articles by McCray, P. B., Jr.

CFTR {Delta}F508 mutation has minimal effect on the gene expression profile of differentiated human airway epithelia

Joseph Zabner,1 Todd E. Scheetz,3 Hakeem G. Almabrazi,3 Thomas L. Casavant,3 Jian Huang,3 Shaf Keshavjee,4 and Paul B. McCray, Jr.2

Departments of 1Internal Medicine and 2Pediatrics, University of Iowa Roy J. and Lucille A. Carver College of Medicine, 3The University of Iowa Center for Bioinformatics and Computational Biology, University of Iowa, Iowa City, Iowa; and 4Toronto Lung Transplant Program, University of Toronto, Toronto, Ontario, Canada

Submitted 8 February 2005 ; accepted in final form 2 June 2005


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cystic fibrosis (CF) is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR), an epithelial chloride channel regulated by phosphorylation. Most of the disease-associated morbidity is the consequence of chronic lung infection with progressive tissue destruction. As an approach to investigate the cellular effects of CFTR mutations, we used large-scale microarray hybridization to contrast the gene expression profiles of well-differentiated primary cultures of human CF and non-CF airway epithelia grown under resting culture conditions. We surveyed the expression profiles for 10 non-CF and 10 {Delta}F508 homozygote samples. Of the 22,283 genes represented on the Affymetrix U133A GeneChip, we found evidence of significant changes in expression in 24 genes by two-sample t-test (P < 0.00001). A second, three-filter method of comparative analysis found no significant differences between the groups. The levels of CFTR mRNA were comparable in both groups. There were no significant differences in the gene expression patterns between male and female CF specimens. There were 18 genes with significant increases and 6 genes with decreases in CF relative to non-CF samples. Although the function of many of the differentially expressed genes is unknown, one transcript that was elevated in CF, the KCl cotransporter (KCC4), is a candidate for further study. Overall, the results indicate that CFTR dysfunction has little direct impact on airway epithelial gene expression in samples grown under these conditions.

cystic fibrosis transmembrane conductance regulator; airway epithelium

CYSTIC FIBROSIS (CF) is the consequence of mutations in the cystic fibrosis transmembrane conductance regulator (CFTR), an epithelial chloride channel regulated by phosphorylation (45). Although more than 900 disease-associated mutations are known, the most common CFTR mutation consists of a three base pair deletion resulting in the loss of a phenylalanine residue at position 508 in the protein ({Delta}F508) (21). The resultant mutant CFTR protein is misfolded, unstable, and degraded (43). Patients with the {Delta}F508 mutation typically develop defects in pulmonary host defense that lead to chronic bacterial infection and inflammation (32). This progressive lung disease accounts for most of the CF-associated morbidity and mortality.

Mutations in CFTR have many well-documented effects on the intracellular trafficking and/or function of the protein and its chloride channel activity in epithelia (46). In addition, CFTR mutations are associated with a number of other phenotypic changes in epithelia, including altered interactions with other ion channels (9, 13), modifications of airway surface liquid volume/composition (25, 40, 48, 54) and antimicrobial activity, mucus sulfation (5), and cytokine release (23). These diverse patterns of alterations in cellular function in CF epithelia may represent either the primary consequences of CFTR mutations or the end result of a series of secondary downstream events. Furthermore, it is widely believed that other modifier genes influence the variability of clinical CF phenotypes, even in patients homozygous for the {Delta}F508 mutation (1, 33). The identification of secondary effects that CFTR mutations have on the expression of other genes in disease-affected tissues may improve our understanding of CF disease pathogenesis and have therapeutic implications.

The availability of a broad genomics-based analytical approach to investigate differential gene expression in human airway epithelia offers a means to uncover the molecular basis for the variety of phenotypic changes associated with mutations in CFTR. The focus of the current study is to bring the power of functional genomics to bear on these issues. To date no such studies have been performed in primary cultures of human airway epithelia with a focus on CF. We performed experiments contrasting the gene expression profiles of 22,283 genes from well-differentiated primary cultures of human airway epithelia of CF and non-CF donors using high-density microarray hybridization. This approach allows study of gene expression profiles in a model of the tissue primarily affected by the disease yet avoids some of the difficulties in obtaining appropriate in vivo samples due to the influences of chronic infection and inflammation and other nonepithelial cell types. Because the {Delta}F508 mutation accounts for ~70% of CF alleles, we focused the analysis on CF epithelia homozygous for this mutation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cell culture. Airway epithelial cells were isolated from tracheae and bronchi of donor lungs. Cells were seeded onto collagen-coated, semipermeable membranes (0.6-cm2 Millicell-HA; Millipore, Bedford, MA) and grown at an air-liquid interface as previously described (22, 37, 55). Twenty-four to forty-eight hours after seeding, the airway cells form a confluent culture with electrically tight junctions. On the third day after seeding, scanning electron microscopy reveals a mostly flat, undifferentiated sheet of cells. Between days 3 and 14, the epithelial cells differentiate with a predominantly ciliated phenotype (20). Epithelial cells were cultured in a 1:1 mixture of Dulbecco’s modified Eagle’s medium and Ham’s F-12 medium supplemented with 2% Ultroser G (BioSepra; Villeneuve, La Garenne, France) and 100 mU/ml penicillin, 100 µg/ml streptomycin, 10 µg/ml gentamicin, 25 µg/ml colimycin, and 75 µg/ml ceftazidime, 25 µg/ml imipenem, 25 µg/ml cilastin, and 2 µg/ml fluconazole. Basolateral culture medium was changed every 2–4 days. Representative samples from all epithelia preparations were evaluated for morphology by scanning electron microscopy to document the development of a ciliated apical surface. The bioelectric properties of each preparation were also characterized (see below). All specimens were genotyped for CFTR mutations. All CF specimens used in this study were homozygous for the {Delta}F508 mutation. Samples used in the analysis were all well differentiated and in culture for >4 wk (Table 1). Samples were collected from organ donors and from CF lung transplant recipients with approval from the University of Iowa Institutional Review Board.


View this table:
[in this window]
[in a new window]
  		Table 1. Profile of CF and non-CF samples

 
Bioelectric properties. For measurement of transepithelial electrical properties, epithelia were mounted in Ussing chambers and studied as previously described (36, 53). Epithelia were bathed in symmetrical solutions containing (in mM): 135 NaCl, 2.4 K2HPO4, 0.6 KH2PO4, 1.2 CaCl2, 1.2 MgCl2, 10 dextrose, and 5 HEPES, at pH 7.2, 37°C and gassed with 100% O2. IscAmil is the decrease in short-circuit current after apical addition of 10 µM amiloride. cAMP-stimulated Isc is the increase in current after basolateral addition of cAMP agonists [10 µM forskolin plus 100 µM 3-isobutyl 1-methylxanthine (IBMX)]. Bumetanide-sensitive Isc (IscBumet) is the decrease in current after basolateral addition of 100 µM bumetanide to epithelia studied in the presence of apical 10 µM amiloride and basolateral cAMP agonists (10 µM forskolin plus 100 µM IBMX); IscBumet is a measure of the transepithelial Cl– transport pathway that includes CFTR.

Processing of RNA preparation and microarray hybridization. Total RNA was extracted in TRIzol reagent (Invitrogen, Carlsbad, CA) according to manufacturer’s recommendations. After a 30-min treatment with 1 U/10 µg RNA with DNase I (Invitrogen), an additional RNA cleanup step was performed using the Qiagen (Chatsworth, CA) RNeasy total RNA isolation kit. The RNA was processed for use on the expression arrays using the manufacturer’s recommended protocols as previously described. Gene expression analysis was used to identify molecular markers of uterine receptivity and embryo implantation (30). Briefly, 10 µg of each pooled total RNA preparation were used to generate cDNA using the Superscript Choice System (Life Technologies). First-strand synthesis was performed using a T7-(dT)24 primer (Sigma-Genosys, Woodlands, TX). The resulting cDNA was used to synthesize biotin-labeled cRNA via in vitro transcription using the ENZO BioArray HighYield RNA transcript labeling kit (Affymetrix). The cRNA was fragmented in fragmentation buffer [40 mM Tris (pH 8.1), 100 mM potassium acetate, and 30 mM magnesium acetate, final concentration] by heating to 94°C for 35 min. The quality of each RNA preparation was assessed by analysis with a Test3 array (Affymetrix) to ensure that the preparations meet Affymetrix’s recommended criteria for use on their expression arrays. Each cRNA preparation (15 µg) was used to inoculate human U133A GeneChip expression arrays (Affymetrix), and the hybridization, stain, scan, and analysis were performed per recommended protocols.

Quantitative RT-PCR comparison. Total RNA was isolated from airway epithelia cultures from each of five (different donors to the ones used in the microarray experiments) non-CF donor samples and from each of six different CF donor samples using RNA STAT-60 Reagent (Tel-Test, Friendswood, TX). RNA was used to quantify the relative difference in quantity of selected mRNAs between non-CF and CF airway epithelia. mRNA specific primers and probes (Integrated DNA Technologies, Coralville, IA) labeled at the 5'-end with 6-FAM and the 3'-end with TAMRA were constructed for each of the genes in interest see Table 2. mRNA expression was normalized to Eukaryotic 18S rRNA Endogenous Control VIC/MGB Probe, Primer Limited (Applied Biosystems, Foster City, CA). Quantitative RT-PCR was performed using TaqMan One-Step RT-PCR Master Mix Reagents Kit (Applied Biosystems). Quantitative RT-PCR was performed on an ABI Prism 7000 Sequence Detecting System with data being analyzed on ABI Prism 7000 SDS Software.


View this table:
[in this window]
[in a new window]
  		Table 2. Minimum detectable fold change

 
Analysis of microarray data. The Affymetrix U133A contain probe sets for ~20,000 human genes, including most of the well-annotated or well-defined gene targets described to date. After hybridization of the cRNA target, the chips were analyzed with Affymetrix’s Microarray suite (MAS) version 5.0 software. A global scaling adjustment was applied with a target intensity of 1,500 to take into account the inherent differences between the chips and their hybridization efficiencies. The global scaling adjustment permitted comparison between arrays. The output of the MAS software provided a qualitative assessment "increase," "marginal increase," "no change," "decrease," and "marginal decrease" and the numerical values used to make the call (see Statistical Algorithms Description Document, http://www.affymetrix.com/support/technical/whitepapers.affx). Absolute analysis of individual chips was also performed to obtain absolute levels of expression for each gene correlated to a confidence interval.

Statistical analysis. Two independent methods of analysis were undertaken to look for differences between non-CF and CF samples.

Method 1. Filtering method of Yechoor et al. We adapted a three-filter method of comparative analysis previously described by Yechoor et al. (52). In this analysis the means and SDs of the expression intensities of each gene for the 10 samples in each group were calculated (non-CF and CF). After this calculation, three separate independent filters of significance were applied serially to obtain a list of those candidate genes that had a significant change between the control and the CF group. The first filter excluded all genes that had a mean expression value that was below the sum of the average background and the average standard difference threshold (SDT; equal to four times the scaled noise) in both the non-CF and CF groups. Genes passing this first filter were subjected to a second significance filter, which removed those genes for which the difference between the means of the CF and non-CF groups was less than twice the sum of the SDs of both groups. Finally, the third filter removed those genes that had an absolute difference between the means of the non-CF and the CF groups that was less than the average SDT. Genes that passed all three filters were then labeled as being significantly changed between the non-CF and the CF groups.

Method 2. Two-sample t-test. The second method of analysis used was the two-sample t-test. To stabilize the variances, the logarithm (with base 2) of the expression values was used in the analysis. The two-sample t-statistic with unequal variances was used to screen genes with significant differential expression. The Welch’s correction for degrees of freedom was used to compute the P values (44). The computation was carried out using the statistical package R (19) and the function t.test. The details of this function are given in the help file in R [by using help(t.test)]. The results of the two-sample t-test are presented graphically in Fig. 3A (see description in Large-scale gene expression profiles of CF and non-CF epithelia). Such a graph, termed a "volcano plot," has proven to be informative method of displaying microarray analysis results (6).



View larger version (40K):
[in this window]
[in a new window]
  		Fig. 3. Large-scale gene expression analysis of non-CF vs. CF epithelia. A: volcano plot showing the distribution of all data sets from 10 non-CF and 10 CF epithelia as described in Methods. Vertical green lines denote a 2-fold difference between the groups. B: 18 genes with significantly increased expression in CF (top right quadrant of volcano plot from A). C: 6 genes with significantly decreased expression in non-CF (top left quadrant of volcano plot from A).

 
Considerations in design of microarray experiments. In the design of experiments to compare the expression profiles of CF and non-CF airway epithelia, several theoretical and practical considerations influenced the final experimental design. These issues included 1) choice of microarray platform: we selected the Affymetrix U133A arrays for these studies because they contain probe sets for most well-characterized genes in a commercially available format. However, this array set may not represent all genes expressed in airway epithelia, and it may contain genes not expressed in airway epithelia. 2) Variability of gene expression: we had no knowledge of the baseline variability between epithelia obtained from different CF and non-CF donors. For these reasons it was not possible to make power calculations and use these to design expression profiling experiments. 3) Availability of tissue: growing airway epithelia from 20 different donors, including 10 specimens homozygous for {Delta}F508 CFTR was ambitious, and a larger sample size was not practical. 4) Paired vs. unpaired experimental design: to date, a gene transfer vector that can correct the {Delta}F508 CFTR transcript in 100% of the cells is not available. 5) Baseline expression levels and magnitude of differences in expression: the expression levels of some genes, particularly those that encode channels such as CFTR, are in very low abundance in airway epithelia (41); the CFTR mRNA signal was called absent by the Affymetrix Statistical Package, thus we included low abundance expression in our analysis.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Sample demographics. Non-CF human airway epithelial cells were obtained from lungs of organ and tissue donors whose lungs were not used for transplantation. CF epithelial cells were obtained from CF lungs at the time of transplantation. All CF samples were homozygous for the {Delta}F508 mutation. Table 1 presents the sample demographics for the CF and non-CF epithelia. On average, the ages of the donors were older for the non-CF samples. All samples were cultured for a similar number of days before microarray analysis and the gender ratios were similar between the groups. The RNA isolation was performed on the same day.

Bioelectric properties of non-CF and CF epithelia used in microarray analysis. Figure 1 presents the basal bioelectric properties of representative samples from all epithelia used in microarray analysis. All CF epithelia demonstrated the characteristic absence of cAMP-activated Cl secretion. These results contrast with the non-CF epithelia, which responded to cAMP agonists with an increase in Isc, and this current was completely blocked by 10–4 M bumetanide. Bumetanide inhibits the predominant mechanism for basolateral chloride uptake, Na-K-2Cl cotransport, in these cells. These results confirm that the samples in each group manifest the expected physiological features and that the main difference between the groups is the lack of CFTR chloride channel function in CF airway epithelia.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Bioelectric properties of non-cystic fibrosis (CF) and cystic fibrosis (CF) epithelia used in microarray analysis. Baseline bioelectric properties were measured from sample epithelia used in the study. CF epithelia demonstrate the characteristic absence of cAMP-activated Cl– secretion. Isc, short-circuit current; Amil, amiloride; Bum, bumetanide. Mean n = 2–3/donor.

 
Comparison of expression profiles for CFTR mRNA in the two sample groups. The {Delta}F508 mutation results in a misfolded CFTR that fails to be displayed on the cell surface. It has been previously shown that the mRNA levels for CFTR in the airway brushings are one to two copies per cell in both CF and non-CF (41). These results were obtained using semiquantitative RT-PCR and have not been replicated by other methods. We compared the CFTR expression profiles from our CF and non-CF epithelia. As shown in Fig. 2, there were no significant differences in CFTR mRNA abundance between CF and non-CF epithelia (probes 205043_at and 217026_at). Moreover, the overall of abundance of CFTR mRNA levels in both cell types was extremely low. For the probe 205043_at, the mean expression value in the CF sample was 9.06 (SD 0.72), and the mean expression value in the normal sample was 9.17 (SD 0.67). The two-sample t-statistic was –0.35, and P = 0.73. For the probe 217026_at, and the mean expression value in the CF sample was 9.10 (SD 1.40), the mean expression value in the normal sample was 8.55 (SD 0.89). The two-sample t-statistic was 1.01, and P = 0.31. We also calculated the nonparametric Wilcoxon test statistics for these two CFTR probes. For probe 205043_at, P from the Wilcoxon statistic was 0.58; for probe 217026_at, P from the Wilcoxon statistic was 0.40. Thus both the t-test and the nonparametric Wilcoxon test indicated no significant change in expression for the two CFTR probes.



View larger version (11K):
[in this window]
[in a new window]
  		Fig. 2. Comparison of expression profiles for CFTR mRNA in non-CF and CF epithelia. Data shown are for probe set 205043. Means ± SE; n = 10.

 
Large-scale gene expression profiles of CF and non-CF epithelia. The first method of analysis applied to the array data was the sequential three-filter comparative analysis of Yechoor and colleagues (52). Of the 22,283 genes represented on the Affymetrix U133A oligonucleotide array, 22,183 genes passed the first significance filter, i.e., had a minimum expression value that was greater than the sum of the average background plus the average SDT (four times the noise). After application of the second significance filter, i.e., requiring a difference between non-CF and CF of greater than two times the sum of the SDs, no genes demonstrated significance in CF epithelia.

Figure 3 presents results from the second method of analysis. In this figure each gene/probe is plotted with the negative log10 P sample values shown on the y-axis and the differences between the average log2-transformed expression values of the CF sample and the average log2-transformed expression values of the normal sample plotted on the x-axis. There were 6,067 genes whose average expression values in the CF samples were greater than those in the non-CF samples by 1 log2 unit, corresponding to a twofold increase. There were 2,029 genes whose average expression values in the CF samples were less than those in the non-CF samples by 1 log2 unit, corresponding to twofold decrease. These genes are shown in purple in Fig. 3A. The two vertical green lines indicate the positive and negative 1-log2 thresholds, respectively. However, based on the t-statistics, there were only 18 genes whose P values were <10–5 and whose fold changes were >2 (i.e., 0.00001, indicated in red). These included nine genes of unknown function, and nine others: glycine dehydrogenase (GenBank NM_000170), PEX10 (GenBank NM_002617), integrin {alpha}10-subunit (GenBank NM_003637), KCl cotransporter (KCC, GenBank NM_005072), long-chain acyl-CoA synthetase 5 (GenBank AF129166), protein serine kinase (GenBank AJ272212), heparan sulfate (glucosamine) 3-O-sulfotransferase 2 (GenBank NM_006043), copine VII (GenBank NM_014427), and a "novel interleukin receptor" (GenBank AF269133). There were also six genes with P values <10–5 and whose fold changes are >2 (indicated in red). These included four genes of unknown function, the transferrin receptor (GenBank BC001188), and topoisomerase I (GenBank J03250). In addition, there were four genes with P value <10-5 but less than twofold change (indicated in blue).

Here the threshold value 10–5 for the P values is much more stringent than the usual 0.05 value. Because there are 22,238 Affymetrix probe sets on the array, it is important to correct for multiple comparisons in controlling the "false positive" rate (i.e., the type 1 error rate). One standard approach is the Bonferroni method, which adjusts for the overall type 1 error rate by evenly distributing it among the individual comparisons. For example, if the overall type 1 error is specified at 0.05, then the threshold for significance at each individual comparison is 0.05/22,283 = 0.0000024. However, the Bonferroni method is based on the assumption that the expression levels of the probes are statistically independent. This assumption is not satisfied in our experiment, because the expression levels of the probes in the same functional class are correlated. Thus the Bonferroni correction is much too conservative. We note that there are other more elaborate multiple-comparison methods, but they also tend to be too conservative for dependent data (47). Therefore, we use the method of controlling the false discovery rate (FDR) of Benjamini and Hochberg (4). This method seeks to control the FDR among the significant genes at a given threshold value. It rigorously deals with the multiple-comparison problem but in the mean time is less conservative than the Bonferroni method, so that truly interesting findings will not be missed. Based on this method, we seek to control the FDR to be <1% (0.01). That is, among the genes found to be significant, <1% of them that are false positive. Some calculation shows that the corresponding threshold for the individual P values is ~0.00001. Here the FDR is calculated as follows: there are 28 genes found to be significant at 0.00001 level. The expected false positive is 0.00001 x 22,238 = 0.22. Thus the FDR is 0.22/28 = 0.8% < 1%.

Requiring the FDR <1% is a quite stringent criterion, but nonetheless less conservative than the Bonferroni correction. Figure 3, B and C, shows the individual data sets for the genes significantly increased or decreased in CF epithelia. All the data sets for these comparisons are available in web-based supplemental materials at http://ajplung.physiology.org/cgi/content/full/00065.2005/DC1.

Epithelial gene expression profiles from male and female CF samples. A number of studies have shown that females with CF have significantly higher mortality than males from age 1 to 20 yr, resulting in an ~4-yr difference in median survival age (31). The reasons underlying these differences are not known. We contrasted the gene expression profiles for the five female and five male CF epithelial samples by methods identical to those described above for Fig. 3. As shown in Fig. 4, gender had very little influence on the results. There were no genes that were significantly increased in either group, using the P value cut-off of 0.00001. Using the same criteria, the analysis identified two genes that were lower in males. These were the c-myc binding protein (probe 203360_s_at, GenBank NM_012333) and a second unknown gene (probe 203640_at, GenBank NM_018615).



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 4. Comparison of gene expression profiles for airway epithelia from 5 female and 5 male CF airway epithelial samples. Data presented in a volcano plot as described in MATERIALS AND METHODS.

 
Quantitative RT-PCR. Using quantitative RT-PCR, we compared the expression profiles of airway epithelia obtained from six independent CF and six non-CF donors. We focused on the genes identified by microarray to be different between CF and non-CF epithelia. Data was obtained from 21 out of the 24 genes differentially expressed. Three of the genes were below the limit of detection of the assay. The difference between the expressions of these genes was not statistically different when a cut-off of 0.003 was used (Bonferroni correction). Nevertheless we compared the trend in differences between the results of the microarray and RT-PCR. Figure 5A shows the genes where the ratio of non-CF over CF expression by microarray and quantitative RT-PCR was concordant. Figure 5B shows the genes where the ratio of non-CF over CF expression by microarray and quantitative RT-PCR were not in agreement.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 5. Correlation of difference in CF vs. non-CF gene expression profiles analyzed by quantitative RT-PCR. The ratio of CF to non-CF mRNA levels for the candidate genes was analyzed by quantitative RT-PCR (open bars) done on 5–6 different donors. Results are compared with the ratio in the microarray data (gray bars). A: genes in which the 2 techniques were in concordance; B: genes with discordant results.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Recent advances in microarray technology allow us to investigate the transcriptional regulation of thousands of genes in different tissues or cells and study responses to a diverse range of cellular processes including development, genetic mutations that cause loss or gain of function, cancer, environmental exposures, therapeutic interventions, and others. As with other research technologies, the complexity of the models used may correlate directly with the relevance and inversely with the fidelity and significance of the results obtained using gene expression profiles.

CF is a multisystem disease; however, infection in the airways is the most important clinical problem, as it leads to most of the morbidity and mortality. Initial infections are caused by a variety of organisms and often can be eradicated by antibiotics (12, 29, 45). As the disease progresses, Pseudomonas aeruginosa establishes permanent colonization in the airways of most patients, and the bacteria persist in the airways for life (29). The combination of chronic infection and exuberant neutrophilic inflammation leads to bronchiectasis, permanent lung damage, and death. Thus discovering the transcriptional consequences of the loss of CFTR function on airway epithelial cells could lead to a better understanding of CF pathophysiology and the development of novel therapeutic interventions. To date, the best models available for expression profiling in cells that lack CFTR include immortalized airway cell lines expressing wild-type vs. mutated forms of CFTR (16, 18, 51), primary airway epithelia from CF and non-CF donors, polarized primary airway cells grown at the air liquid interface (15, 17, 20, 50), or epithelial biopsies or lung tissue obtained from either humans or mice with targeted disruption of the CFTR gene.

Airway epithelia cell lines have several advantages: they are readily available, culture conditions are highly reproducible, it is easy to obtain large amounts of RNA, the lack of CFTR can be transiently or stably corrected, and others. However, there are several limitations to such cell lines. Permanently transformed epithelia with either wild-type or mutant CFTR usually do not form polarized sheets with characteristic transepithelial voltage and resistance (2). Similarly transformed cells do not assume the morphology of airway epithelia, and thus they may not express all relevant genes that may be affected by altered CFTR function. Finally, even though transformed cells are often generated by similar methods, the process of immortalization usually results in genomic instability that may differentially affect each resultant cell line. Thus, although the results are highly reproducible and useful to test specific hypothesis, their relevance needs to be confirmed in a more complex system.

Evidence from studies in several airway cell lines indicates that dysregulation of airway inflammation in patients with CF may manifest as inflammation independent of infection and/or inflammation disproportionately increased or prolonged relative to the level of proinflammatory stimuli. Although these results are highly reproducible, few reports have compared inflammatory responses to different stimuli using multiple CF cell-line models. Using microarray technology in a matched pair of cell lines expressing wild-type (IB3a) (11) and mutant {Delta}F508/W1282X CFTR (IB3-1) (34), Srivastava et al. (39) found that a significant number of genes involved in the IL-8 pathway were altered in the absence of wild-type CFTR expression. These data have led to identification of targets for novel pharmacological therapies and may offer a simple model to experimentally test the change in gene expression profile by current pharmacologic therapies. Of interest, recent work by Aldallal et al. (2) and Becker and colleagues (3) have shown that although increased IL-8 expression is observed in some CF airway epithelial cell models, many CF cells do not exhibit altered regulation of these important inflammatory genes.

Multiple murine models of CF have been generated by targeted disruption of the CFTR gene (10, 27, 33, 38, 56). Surprisingly, although most mice develop significant intestinal pathology reminiscent of meconium ileus, lung pathology has been observed by only a few investigators (7, 8). Spontaneous onset of pulmonary abnormalities has been reported in congenic C57BL/6 mice (33). Xu et al. (49) recently studied the gene expression profile of whole lungs from mixed FVB/N, C57BL/6 background mice homozygous for a null mutation in CFTR and expressing normal human CFTR in the intestinal epithelium under control of the intestinal fatty acid-binding protein gene promoter. In marked contrast to the level of complexity of gene expression profile experiments in the cell lines described above, they collected RNA from whole lungs containing the airway and alveolar epithelia, vascular, nerves, interstitial and connective tissue, smooth muscle, and hematopoietic cells. They reported that the lack of CFTR expression in murine lungs resulted in increased expression of genes in the pathways of IL-1 and CCAAT enhancer-binding protein (CEBP). However, these studies cannot differentiate between responses that are secondary to changes in the expression and function of CFTR in the epithelial cells or in nonepithelial cells. Moreover, since the airway epithelium represents only a small portion of the lung mass, it is possible that the gene expression profile changes detected arise mainly from nonepithelial cells and are only indirectly related to the lack of CFTR function in epithelia. Thus more studies are required to understand the transcriptional consequences of a lack of CFTR expression in the airway epithelium in vivo.

The present study offers an intermediate level of complexity that allowed us to compare the expression profiles of non-CF airway epithelia to that of airway epithelia obtained from {Delta}F508 homozygote donors. Using this model we tested the effect of a lack of CFTR on epithelial cells grown at the air-liquid interface, in isolation from the rest of the lung milieu (20). Whereas the results of the gene expression profiles are exclusively from airway epithelial cells, we must cautiously interpret the data since the epithelia are cultured in a context that may differ in vivo between CF and non-CF cells.

We used two methods to calculate statistically significant differences in gene expression. The first is the two-sample t-test combined with controlling the FDR to correct for multiple comparisons. For an expected 1% FDR with ~20,000 genes we established a P < 10–5 as a cut-off for significance in individual comparisons. This degree of stringency was applied to lessen the number of false positive hits. Although this method may be somewhat restrictive, it is still subject to false positive differences. The volcano plot representation of these data (Figs. 3 and 4) provides the advantages of allowing simultaneous visual comparison of the actual differences in gene expression profiles and the statistical significance of these data. Interestingly, we found very few genes that were significantly differentially expressed in CF vs. non-CF epithelia based on the cut-off of P < 10–5. Most of these transcripts have unknown function. Some genes identified such as the novel interleukin receptor (28) and the KCC (14) are candidates for further investigation using biological interventions. However, the discordant result when an independent set of epithelia was studied by quantitative RT-PCR limits our enthusiasm for novel interleukin receptor. Other genes, like the transferrin receptor, may be due to experimental variation or may lead to novel hypotheses.

The KCC are members of the gene family of electroneutral cation-chloride cotransporters, which include the thiazide-sensitive Na+-Cl– cotransporter and bumetanide-sensitive Na+-K+-2Cl– cotransporter. There are four KCC homologs, all but KCC2 expressed in the lung (26). KCC4 is expressed in muscle, brain, lung, heart, and kidney (42). Activation of KCCs by cAMP agonist has been implicated in activation of Cl– transport across the basolateral side of alveolar epithelium (24).

We used a quite stringent cut off for selecting "significant" genes, even after adjusting for multiple comparisons based on the method of controlling FDR. Therefore, there may be genes of interest that are not in our list of significant genes. Clearly, relaxation of the cut-off for the P values will allow generation of a longer list of significant genes. Indeed, some of these genes may deserve further investigation. However, we note that no matter what reasonable cut-off for the P values we use, there will not be a significant difference between the groups for the two CFTR probes (probes 205043_at and 217026_at). A second statistical method used was a sequential method. This method has been used successfully to compare the expression profiles of unpaired complex models of diabetes (52). Interestingly, using this method we failed to identify a single differentially expressed gene between CF and non-CF airway epithelia.

We were surprised with the small number of genes that were differentially regulated in the CF epithelia compared with the non-CF. Moreover, as expected, even under controlled culture conditions there was great variability in the gene expression profiles within individuals from the same group. This variability may explain some of the broad variations in CF lung disease phenotype in individuals with identical CFTR mutations. Also, the distribution of (nonsignificant) differentially expressed genes is highly asymmetrical (6,067 genes more than twofold higher in CF vs. only 2,029 genes more than twofold higher in non-CF samples), whereas this is hard to explain it may reflect the difference in variability between the groups. Moreover, the variability within groups suggests that comparisons of gene expression profiles from lung human specimens will require a significant number of specimens to avoid false positive hits. Based on the variability we calculated the minimum detectable fold changes at various power levels when comparing the gene expression levels between paired airway epithelia. Calculations are based on the paired t-test using a two-sided test at various significance levels and assuming a SD estimate of SD 0.5 (35). We calculated the minimum detectable differences, then the fold changes are 2 to the power of the differences, because we work on the log2 transformation of the original data. With the type 1 error rate of 0.00001 and with ~22,000 human genes to be profiles, the expected number of false discoveries is controlled to be <1. For unpaired experiments the same table can be used by multiplying the number by 4.

In contrast with what was expected from the studies in CF mice that predicted an increased expression of CFTR mRNA due to increased transcription factor CEBP{delta} (49), we found that CFTR mRNA levels in CF and non-CF epithelia were similar and low. This is in agreement with earlier studies that showed that epithelial cells recovered by fiber-optic bronchoscopy contained CFTR mRNA transcripts at approximately one to two copies per cell in both normal and CF individuals (41).

In summary we find that under resting culture conditions CF and non-CF epithelial cells manifest small differences in their gene expression profiles. The results of this study form the foundation for future studies to evaluate the effects of the environment on CF and non-CF epithelia. Such studies could contrast the responses to bacteria, bacterial products, cytokines, or other CFTR mutations. Finally, complementary studies in simpler and more complex models will likely be required to fully understand the functional transcriptional consequences of a lack of CFTR function.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by the Cystic Fibrosis Foundation (McCray00V0, McCray 03FG0) and the National Heart, Lung, and Blood Institute (NHLBI) Shared Microarray Grant 5R01HL-072288-03. We also acknowledge the support of the Cell Culture Core and Cell Morphology Core, partially supported by the Cystic Fibrosis Foundation, NHLBI Program Project Grant HL-51670, and the Center for Gene Therapy for Cystic Fibrosis (National Institute of Diabetes and Digestive and Kidney Diseases Grant P30 DK-54759).


   ACKNOWLEDGMENTS
 
We are grateful for the technical contributions of Geri Traver and Thomas Recker and to Phil Karp, Janice Launspach, and Tami Nesselhauf for culturing the human epithelial cells. Finally, thanks to Jamie Kesselring for excellent assistance.

T. E. Scheetz was partially supported under a Career Development Award from Research to Prevent Blindness.


   FOOTNOTES
 

Address for reprint requests and other correspondence: J. Zabner, Univ. of Iowa Roy J. and Lucille A. Carver College of Medicine, 440 EMRB, Iowa City, IA 52242 (e-mail: joseph-zabner{at}uiowa.edu)

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
 GRANTS
 REFERENCES
 

  1. Accurso FJ and Sontag MK. Seeking modifier genes in cystic fibrosis. Am J Respir Crit Care Med 167: 289–290, 2003.[Free Full Text]
  2. Aldallal N, McNaughton EE, Manzel LJ, Richards AM, Zabner J, Ferkol TW, and Look DC. Inflammatory response in airway epithelial cells isolated from patients with cystic fibrosis. Am J Respir Crit Care Med 166: 1248–1256, 2002.[Abstract/Free Full Text]
  3. Becker MN, Sauer MS, Muhlebach MS, Hirsh AJ, Wu Q, Verghese MW, and Randell SH. Cytokine secretion by cystic fibrosis airway epithelial cells. Am J Respir Crit Care Med 169: 645–653, 2004.[Abstract/Free Full Text]
  4. Benjamini Y and Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J Royal Stat Soc Ser B 57: 289–300, 1995.
  5. Cheng PW, Boat TF, Cranfill K, Yankaskas JR, and Boucher RC. Increased sulfation of glycoconjugates by cultured nasal epithelial cells from patients with cystic fibrosis. J Clin Invest 84: 68–72, 1989.[Web of Science][Medline]
  6. Churchill GA and Oliver B. Sex, flies and microarrays. Nat Genet 29: 355–356, 2001.[CrossRef][Web of Science][Medline]
  7. Coleman FT, Mueschenborn S, Meluleni G, Ray C, Carey VJ, Vargas SO, Cannon CL, Ausubel FM, and Pier GB. Hypersusceptibility of cystic fibrosis mice to chronic Pseudomonas aeruginosa oropharyngeal colonization and lung infection. Proc Natl Acad Sci USA 100: 1949–1954, 2003.[Abstract/Free Full Text]
  8. Davidson DJ, Dorin JR, McLachlan G, Ranaldi V, Lamb D, Doherty C, Govan J, and Porteous DJ. Lung disease in the cystic fibrosis mouse exposed to bacterial pathogens. Nat Genet 9: 351–357, 1995.[CrossRef][Web of Science][Medline]
  9. Donaldson SH, Poligone EG, and Stutts MJ. CFTR regulation of ENaC. Meth Mol Med 70: 343–364, 2002.
  10. Dorin JR, Dickinson P, Alton EW, Smith SN, Geddes DM, Stevenson BJ, Kimber WL, Fleming S, Clarke AR, Hooper ML, Anderson L, Beddington RSP, and Porteous DJ. Cystic fibrosis in the mouse by targeted insertional mutagenesis. Nature 359: 211–215, 1992.[CrossRef][Medline]
  11. Flotte TR, Afione SA, Solow R, Drumm ML, Markakis D, Guggino WB, Zeitlin PL, and Carter BJ. Expression of the cystic fibrosis transmembrane conductance regulator from a novel adeno-associated virus promoter. J Biol Chem 268: 3781–3790, 1993.[Abstract/Free Full Text]
  12. Frederiksen B, Koch C, and Hoiby N. Antibiotic treatment of initial colonization with Pseudomonas aeruginosa postpones chronic infection and prevents deterioration of pulmonary function in cystic fibrosis. Pediatr Pulmonol 23: 330–335, 1997.[CrossRef][Web of Science][Medline]
  13. Gabriel SE, Clarke LL, Boucher RC, and Stutts MJ. CFTR and outward rectifying chloride channels are distinct proteins with a regulatory relationship. Nature 363: 263–268, 1993.[CrossRef][Medline]
  14. Gillen CM and Forbush B III. Functional interaction of the K-Cl cotransporter (KCC1) with the Na-K-Cl cotransporter in HEK-293 cells. Am J Physiol Cell Physiol 276: C328–C336, 1999.[Abstract/Free Full Text]
  15. Gruenert DC, Basbaum CB, and Widdicombe JH. Long-term culture of normal and cystic fibrosis epithelial cells grown under serum-free conditions. In Vitro Cell Dev Biol 26: 411–418, 1990.
  16. Gruenert DC, Finkbeiner WE, and Widdicombe JH. Culture and transformation of human airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 268: L347–L360, 1995.[Abstract/Free Full Text]
  17. Harris CC, Lechner JF, Yoakum GH, Amstad P, Korba BE, Gabrielson E, Grafstrom R, Shamsuddin A, and Trump BF. In vitro studies of human lung carcinogenesis. Carcinog Compr Surv 9: 257–269, 1985.[Medline]
  18. Hopfer U, Jacobberger JW, Gruenert DC, Eckert RL, Jat PS, and Whitsett JA. Immortalization of epithelial cells. Am J Physiol Cell Physiol 270: C1–C11, 1996.[Abstract/Free Full Text]
  19. Ihaka R and Gentleman RR. A language for data analysis and graphics. J Comp Graph Stat 5: 299–314, 1996.[CrossRef]
  20. Karp PH, Moninger T, Weber SP, Nesselhauf TS, Launspach J, Zabner J, and Welsh MJ. An in vitro model of differentiated human airway epithelia: methods and evaluation of primary cultures. In: Epithelial Cell Culture Protocols, edited by Wise C. Totowa, NJ: Humana, 2002, p. 115–137.
  21. Kerem B, Rommens JM, Buchanan JA, Markiewicz D, Cox TK, Chakravarti A, Buchwald M, and Tsui LC. Identification of the cystic fibrosis gene: genetic analysis. Science 245: 1073–1080, 1989.[Abstract/Free Full Text]
  22. Kondo M, Finkbeiner WE, and Widdicombe JH. Simple technique for culture of highly differentiated cells from dog tracheal epithelium. Am J Physiol Lung Cell Mol Physiol 261: L106–L117, 1991.[Abstract/Free Full Text]
  23. Kube D, Sontich U, Fletcher D, and Davis PB. Proinflammatory cytokine responses to P. aeruginosa infection in human airway epithelial cell lines. Am J Physiol Lung Cell Mol Physiol 280: L493–L502, 2001.[Abstract/Free Full Text]
  24. Lee SY, Maniak PJ, Rhodes R, Ingbar DH, and O’Grady SM. Basolateral Cl– transport is stimulated by terbutaline in adult rat alveolar epithelial cells. J Membr Biol 191: 133–139, 2003.[CrossRef][Web of Science][Medline]
  25. Matsui H, Grubb BR, Tarran R, Randell SH, Gatzy JT, Davis CW, and Boucher RC. Evidence for periciliary liquid layer depletion, not abnormal ion composition, in the pathogenesis of cystic fibrosis airways disease. Cell 95: 1005–1015, 1998.[CrossRef][Web of Science][Medline]
  26. Mount DB, Mercado A, Song L, Xu J, George AL Jr, Delpire E, and Gamba G. Cloning and characterization of KCC3 and KCC4, new members of the cation-chloride cotransporter gene family. J Biol Chem 274: 16355–16362, 1999.[Abstract/Free Full Text]
  27. O’Neal WK, Hasty P, McCray PB Jr, Casey B, Rivera-Perez J, Welsh MJ, Beaudet AL, and Bradley A. A severe phenotype in mice with a duplication of exon 3 in the cystic fibrosis locus. Hum Mol Genet 2: 1561–1569, 1993.[Abstract/Free Full Text]
  28. Parrish-Novak J, Foster DC, Holly RD, and Clegg CH. Interleukin-21 and the IL-21 receptor: novel effectors of NK and T cell responses. J Leukoc Biol 72: 856–863, 2002.[Abstract/Free Full Text]
  29. Rajan S and Saiman L. Pulmonary infections in patients with cystic fibrosis. Semin Respir Infect 17: 47–56, 2002.[CrossRef][Medline]
  30. Reese J, Das SK, Paria BC, Lim H, Song H, Matsumoto H, Knudtson KL, DuBois RN, and Dey SK. Global gene expression analysis to identify molecular markers of uterine receptivity and embryo implantation. J Biol Chem 276: 44137–44145, 2001.[Abstract/Free Full Text]
  31. Rosenfeld M, Davis R, FitzSimmons S, Pepe M, and Ramsey B. Gender gap in cystic fibrosis mortality. Am J Epidemiol 145: 794–803, 1997.[Abstract/Free Full Text]
  32. Rosenfeld M, Gibson RL, McNamara S, Emerson J, Burns JL, Castile R, Hiatt P, McCoy K, Wilson CB, Inglis A, Smith A, Martin TR, and Ramsey BW. Early pulmonary infection, inflammation, and clinical outcomes in infants with cystic fibrosis. Pediatr Pulmonol 32: 356–366, 2001.[CrossRef][Web of Science][Medline]
  33. Rozmahel R, Wilschanski M, Matin A, Plyte S, Oliver M, Auerbach W, Moore A, Forstner J, Durie P, Nadeau J, Bear C, and Tsui LC. Modulation of disease severity in cystic fibrosis transmembrane conductance regulator deficient mice by a secondary genetic factor. Nat Genet 12: 280–287, 1996.[CrossRef][Web of Science][Medline]
  34. Rubenstein RC, Egan ME, and Zeitlin PL. In vitro pharmacologic restoration of CFTR-mediated chloride transport with sodium 4-phenylbutyrate in cystic fibrosis epithelial cells containing delta F508-CFTR. J Clin Invest 100: 2457–2465, 1997.[Web of Science][Medline]
  35. Simon R, Radmacher MD, and Dobbin K. Design of studies using DNA microarrays. Genet Epidemiol 23: 21–36, 2002.[CrossRef][Web of Science][Medline]
  36. Smith JJ, Karp PH, and Welsh MJ. Defective fluid transport by cystic fibrosis airway epithelia. J Clin Invest 93: 1307–1311, 1994.[Web of Science][Medline]
  37. Smith JJ, Travis SM, Greenberg EP, and Welsh MJ. Cystic fibrosis airway epithelia fail to kill bacteria because of abnormal airway surface fluid. Cell 85: 229–236, 1996.[CrossRef][Web of Science][Medline]
  38. Snouwaert JN, Brigman KK, Latour AM, Malouf NN, Boucher RC, Smithies O, and Koller BH. An animal model for cystic fibrosis made by gene targeting. Science 257: 1083–1088, 1992.[Abstract/Free Full Text]
  39. Srivastava M, Eidelman O, and Pollard HB. Pharmacogenomics of the cystic fibrosis transmembrane conductance regulator (CFTR) and the cystic fibrosis drug CPX using genome microarray analysis. Mol Med 5: 753–767, 1999.[Web of Science][Medline]
  40. Tate S, MacGregor G, Davis M, Innes JA, and Greening AP. Airways in cystic fibrosis are acidified: detection by exhaled breath condensate. Thorax 57: 926–929, 2002.[Abstract/Free Full Text]
  41. Trapnell BC, Chu CS, Paakko PK, Banks TC, Yoshimura K, Ferrans VJ, Chernick MS, and Crystal RG. Expression of the cystic fibrosis transmembrane conductance regulator gene in the respiratory tract of normal individuals and individuals with cystic fibrosis. Proc Natl Acad Sci USA 88: 6565–6569, 1991.[Abstract/Free Full Text]
  42. Velazquez H and Silva T. Cloning and localization of KCC4 in rabbit kidney: expression in distal convoluted tubule. Am J Physiol Renal Physiol 285: F49–F58, 2003.[Abstract/Free Full Text]
  43. Ward CL and Kopito RR. Intracellular turnover of cystic fibrosis transmembrane conductance regulator. Inefficient processing and rapid degradation of wild-type and mutant proteins. J Biol Chem 269: 25710–25718, 1994.[Abstract/Free Full Text]
  44. Welch BL. ’Student’ and small sample theory. J Am Stat Assoc 53: 777–788, 1958.[CrossRef][Web of Science]
  45. Welsh MJ, Ramsey BW, Accurso F, and Cutting GR. Cystic fibrosis. In: The Metabolic and Molecular Bases of Inherited Disease (8th ed.), edited by Scriver CR, Beaudet AL, Sly WS, Valle D, Childs B, and Vogelstein B. New York: McGraw-Hill, 2001.
  46. Welsh MJ and Smith AE. Molecular mechanisms of CFTR chloride channel dysfunction in cystic fibrosis. Cell 73: 1251–1254, 1993.[CrossRef][Web of Science][Medline]
  47. Westfall PH and Young SS. Resampling-based Multiple Testing: Examples and Methods for p-Value Adjustment (Wiley series in probability and mathematical statistics). New York: Wiley-Interscience, 1993.
  48. Widdicombe JH. Regulation of the depth and composition of airway surface liquid. J Anat 201: 313–318, 2002.[CrossRef][Web of Science][Medline]
  49. Xu Y, Clark JC, Aronow BJ, Dey CR, Liu C, Wooldridge JL, and Whitsett JA. Transcriptional adaptation to cystic fibrosis transmembrane conductance regulator deficiency. J Biol Chem 278: 7674–7682, 2003.[Abstract/Free Full Text]
  50. Yamaya M, Finkbeiner WE, Chun SY, and Widdicombe JH. Differentiated structure and function of cultures from human tracheal epithelium. Am J Physiol Lung Cell Mol Physiol 262: L713–L724, 1992.[Abstract/Free Full Text]
  51. Yankaskas JR, Haizlip JE, Conrad M, Koval D, Lazarowski E, Paradiso AM, Rinehart CAJ, Sarkadi B, Schlegel R, and Boucher RC. Papilloma virus immortalized tracheal epithelial cells retain a well-differentiated phenotype. Am J Physiol Cell Physiol 264: C1219–C1230, 1993.[Abstract/Free Full Text]
  52. Yechoor VK, Patti ME, Saccone R, and Kahn CR. Coordinated patterns of gene expression for substrate and energy metabolism in skeletal muscle of diabetic mice. Proc Natl Acad Sci USA 99: 10587–10592, 2002.[Abstract/Free Full Text]
  53. Zabner J, Ramsey BW, Meeker DP, Aitken MI, Balfour RP, Gibson RL, Launspach J, Moscicki RA, Richards SM, Standaert TA, Williams-Warren J, Wadsworth SC, Smith AE, and Welsh MJ. Repeat administration of an adenovirus vector encoding cystic fibrosis transmembrane conductance regulator to the nasal epithelium of patients with cystic fibrosis. J Clin Invest 97: 1504–1511, 1996.[Web of Science][Medline]
  54. Zabner J, Smith JJ, Karp PH, Widdicombe JH, and Welsh MJ. Loss of CFTR chloride channels alters salt absorption by cystic fibrosis airway epithelia in vitro. Mol Cell 2: 397–403, 1998.[CrossRef][Web of Science][Medline]
  55. Zabner J, Zeiher BG, Friedman E, and Welsh MJ. Adenovirus-mediated gene transfer to ciliated airway epithelia requires prolonged incubation time. J Virol 70: 6994–7003, 1996.[Abstract/Free Full Text]
  56. Zeiher BG, Eichwald E, Zabner J, Smith JJ, Puga AP, McCray PBJ, Capecchi MR, Welsh MJ, and Thomas KR. A mouse model for the delta F508 allele of cystic fibrosis. J Clin Invest 96: 2051–2064, 1995.[Web of Science][Medline]



This article has been cited by other articles:


Home page
 ScienceHome page
A. S. Shah, Y. Ben-Shahar, T. O. Moninger, J. N. Kline, and M. J. Welsh
Motile Cilia of Human Airway Epithelia Are Chemosensory
Science, August 28, 2009; 325(5944): 1131 - 1134.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
H. Hamai, F. Keyserman, L. M. Quittell, and T. S. Worgall
Defective CFTR increases synthesis and mass of sphingolipids that modulate membrane composition and lipid signaling
J. Lipid Res., June 1, 2009; 50(6): 1101 - 1108.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
O. Granio, C. Norez, K. J. D. Ashbourne Excoffon, P. H. Karp, M. Lusky, F. Becq, P. Boulanger, J. Zabner, and S.-S. Hong
Cellular Localization and Activity of Ad-Delivered GFP-CFTR in Airway Epithelial and Tracheal Cells
Am. J. Respir. Cell Mol. Biol., December 1, 2007; 37(6): 631 - 639.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
K. Hybiske, Z. Fu, C. Schwarzer, J. Tseng, J. Do, N. Huang, and T. E. Machen
Effects of cystic fibrosis transmembrane conductance regulator and {Delta}F508CFTR on inflammatory response, ER stress, and Ca2+ of airway epithelia
Am J Physiol Lung Cell Mol Physiol, November 1, 2007; 293(5): L1250 - L1260.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
D. A. Stoltz, E. A. Ozer, C. J. Ng, J. M. Yu, S. T. Reddy, A. J. Lusis, N. Bourquard, M. R. Parsek, J. Zabner, and D. M. Shih
Paraoxonase-2 deficiency enhances Pseudomonas aeruginosa quorum sensing in murine tracheal epithelia
Am J Physiol Lung Cell Mol Physiol, April 1, 2007; 292(4): L852 - L860.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
J. M. Wright, C. A. Merlo, J. B. Reynolds, P. L. Zeitlin, J. G. N. Garcia, W. B. Guggino, and M. P. Boyle
Respiratory Epithelial Gene Expression in Patients with Mild and Severe Cystic Fibrosis Lung Disease
Am. J. Respir. Cell Mol. Biol., September 1, 2006; 35(3): 327 - 336.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
289/4/L545    most recent
00065.2005v1
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 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 Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zabner, J.
Right arrow Articles by McCray, P. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zabner, J.
Right arrow Articles by McCray, P. B., Jr.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2005 by the American Physiological Society.