Introduction

Cystic fibrosis (CF) is an autosomal recessive disease caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene that occurs with an incidence of approximately 1 in 3000 live births¹,². The CFTR protein is a chloride channel that is expressed in many tissues. In the lung, CFTR functions to regulate the volume of airway surface liquid³. Absence of functional CFTR protein results in thickened airway mucus that leads to chronic airway infection and, ultimately, severe bronchiectasis. Most CF patients die at an early age (30 years) as a result of respiratory failure. Introducing a vector coding for wild-type CFTR into CF airway cells has been demonstrated to correct the basic defect in chloride secretion, suggesting that gene therapy may be useful to treat individuals afflicted with this disease⁴,⁵.

To date, gene therapy trials for CF have been largely unsuccessful, primarily due to low efficiency of gene transfer to the airway epithelium⁶,⁷. However, several laboratories have recently reported strategies to increase the delivery of transgenes to airway epithelium. For example, the brief disruption of epithelial tight junctions allows viral vectors improved access to basolateral receptors, resulting in significantly greater gene transfer⁸,⁹,¹⁰. Others have reported that different serotypes of adeno-associated virus (AAV) can more efficiently penetrate the apical surface of well-differentiated epithelium¹¹. These studies suggest that the barriers to gene transfer to the airway epithelium may soon be overcome. However, it is likely that further improvements in vector design will be necessary before gene therapy for CF is successful.

One potential limitation of the gene transfer vectors commonly used for CF studies is a lack of specificity. Adenovirus, AAV, and pseudotyped retrovirus will infect and express high levels of CFTR in many cell types, which may have deleterious effects in cells that do not normally express CFTR¹²,¹³. An approach to improving the specificity of these vectors is through the inclusion in the vector of a cell- or tissue-specific promoter. For gene therapy of CF, we believe that a ciliated cell-specific promoter provides several advantages. First, a ciliated cell-specific promoter would direct CFTR expression to the ciliated cells at the apical surface of the airway epithelium. This would result in CFTR being expressed in direct contact with the airway surface liquid, where it is believed to be essential to regulate the volume of airway surface liquid. Second, CFTR would not be expressed in other cell types in the lung, including macrophages, fibroblasts, and basal cells. This feature would prevent the expression of CFTR in cells that normally do not express the protein and may also limit immune responses against the "foreign" wt CFTR protein¹⁴. Third, transgene expression from some of the viral promoters is transient, possibly due to promoter silencing¹⁵. In contrast, an endogenous human ciliated cell-specific promoter may be resistant to promoter silencing, resulting in long-term expression. Finally, while the potential risk of viral integration is unknown, it is possible that integration of a strong promoter may activate the expression of a critical cellular gene, leading to unrestricted cellular growth. Recent reports suggest that integration of a viral vector may have induced an unusual form of leukemia¹⁶. Ciliated cells, however, are considered to be "terminally differentiated" and may not be capable of proliferation¹⁷. Integration of a ciliated cell-specific promoter upstream of a proto-oncogene may therefore be of little consequence. Thus a ciliated cell-specific promoter may provide an additional safety advantage, relative to a promoter from a ubiquitously expressed gene.

To develop a ciliated cell-specific promoter for gene therapy of CF and other airway diseases, we have begun to identify and characterize the promoter regions of ciliated cell-specific genes. FOXJ1 (hepatocyte nuclear factor-3/forkhead homologue 4; HFH-4) is a member of the forkhead/winged helix family of transcription factors whose expression is tightly restricted to cells possessing motile cilia or flagella¹⁸,¹⁹,²⁰. In this study, we have investigated whether the promoter region of the FOXJ1 gene is sufficient to direct expression of a reporter gene to the ciliated cells in a transgenic mouse model.

Results

In preliminary studies, putative promoter regions from ciliated cell-specific genes were inserted into retroviral vectors and used to infect human bronchial epithelial cells in vitro. These experiments demonstrated that a 5' fragment of the human FOXJ1 gene was capable of directing expression of a reporter gene to the ciliated cells in cultures of well-differentiated human bronchial epithelial cells²⁴. To test the activity of this promoter fragment in vivo, we inserted a 1.0-kb fragment of the human FOXJ1 gene (GenBank Accession No. X99350), consisting of nucleotides -1008 to -80, into an expression cassette upstream of the EGFP gene (Fig. 1A). This fragment contains a CpG island (-643 to -877) and 137 bp of the 5' untranslated region²¹. The expression cassette also contained an intronic region containing splice donor and acceptor sites and the SV40 polyadenylation signal to improve processing.

Figure 1.

(A) Diagram of the FOXJ1 promoter construct used to produce transgenic animals. The promoter fragment consists of 1008 bp of the FOXJ1 genomic region (-80 to -1088 relative to the ATG) and includes the RNA transcription start site (-216) and a CpG island (-643 to -877). The construct also includes portions of two noncoding exons (shaded boxes, I and II) and the first intron from the mouse transthyretin (TTR) gene. The relative positions of the EGFP cDNA and the SV40 poly(A) site are indicated. The total size of the injected construct was 3.2 kb. (B) Southern analysis of FOXJ1/EGFP founder animals. DNA isolated from transgenic animals was probed for the presence of the EGFP transgene. Animals 45, 46, 49, and 85 were bred to establish four separate lines expressing EGFP from the FOXJ1 promoter.

Full figure and legend (111K)

Pronuclear injection of the expression cassette resulted in seven transgene-positive animals that successfully transmitted the transgene to their offspring (founder animals) (Fig. 1B). Preliminary examination of FOXJ1/EGFP-positive offspring from each of the seven lines by RT-PCR demonstrated expression of EGFP in tracheal RNA from four of the lines (data not shown). We chose these four lines (45, 46, 49, and 85) for further studies.

To characterize the pattern of expression of EGFP from the FOXJ1 promoter, we sacrificed transgene-positive and -negative littermates and examined frozen sections of various tissues by fluorescence microscopy. EGFP fluorescence was easily detected in the tracheal epithelium of FOXJ1/EGFP transgenic mice (Fig. 2A), while wild-type littermates demonstrated only weak background fluorescence under the same exposure conditions (Fig. 2K). The fluorescence appeared to be confined to a single layer of strongly positive cells around the lumen of the trachea. EGFP expression was also clearly detected in the large airways and in nasal epithelium (Figs. 2B and 2C); control tissues were negative (Figs. 2L and 2M). Tissues without cilia or flagella, including esophagus and skeletal muscle, exhibited only background fluorescence that was equivalent in the transgene-positive and transgene-negative animals (Figs. 2D, 2E, 2N, and 2O).

Figure 2.

EGFP fluorescence in frozen sections from FOXJ1/EGFP transgenic and wild-type mice. EGFP is clearly expressed in the ciliated epithelium of tracheal, lung, and nasal tissue from the transgenic animals (A, B, C), while sections from esophagus and muscle (D, E) and from wild-type animals (K, L, M, N, and O) show no EGFP fluorescence. (F, G, H, I, J, P, Q, R, S, and T) Adjacent sections stained with hematoxylin and eosin. All images were from line 45, except B, G, L, and Q, which were from line 49. Scale bars, 100 m.

Full figure and legend (428K)

In addition to the ciliated cells of the respiratory tract, FOXJ1 has previously been reported to be expressed in other tissues that contain axonemal structures, including the brain, testis, and oviduct¹⁸,¹⁹,²⁰. To compare the expression pattern of the FOXJ1/EGFP transgene to the previously reported pattern of endogenous FOXJ1 expression, we examined these tissues also for EGFP fluorescence. In sections of brain tissue from the FOXJ1/EGFP-positive animals, the ciliated ependymal cells lining the ventricles were clearly EGFP positive (Fig. 3A). Other areas of the brain and sections from control animals showed no detectable EGFP fluorescence. The oviduct and testis of the transgenic animals were also strongly positive (Figs. 3B and 3C), while in surrounding tissue and tissue from control animals fluorescence was negligible. These results demonstrate that the FOXJ1 promoter fragment is capable of targeting the expression of a transgene to the ciliated epithelium, and the expression pattern is very similar to that of the endogenous FOXJ1 gene.

Figure 3.

EGFP fluorescence in nonrespiratory tract tissues from FOXJ1/EGFP transgenic mice. (A) Ciliated ependymal cells lining the ventricles of the brain are strongly EGFP positive. Ciliated epithelium of the oviduct (B) and developing sperm in the testis (C) also express EGFP from the FOXJ1 promoter. Scale bars, 50 m. A and B, line 45; C line 85.

Full figure and legend (101K)

To determine if the expression of EGFP from the FOXJ1 promoter varied between the different transgenic lines, we analyzed positive animals from each of the four lines in more detail. We used Southern analysis to estimate the copy number of the transgene; two lines (49, 85) were estimated to contain between 1 and 10 copies of the transgene, while two lines (45, 46) were estimated to contain between 20 and 50 copies (not shown). Frozen sections of trachea and lung from each of the lines exhibited EGFP fluorescence of similar intensity in the ciliated epithelium, although animals from line 46 demonstrated a patchy pattern of fluorescence, with some strongly positive cells adjacent to negative cells. Esophagus and other adjacent tissue was routinely negative (Fig. 4A). To compare the levels of EGFP expression in the different lines more directly, we prepared protein extracts from isolated tracheas and analyzed them by Western blotting. Probing with an affinity-purified rabbit anti-GFP antibody produced a clear signal of the expected size from extracts of FOXJ1/EGFP transgenic mice, while wild-type animals produced no signal (Fig. 4B). We probed duplicate blots with an anti-IV tubulin antibody, a specific marker for cilia²⁵, to control for the percentage of ciliated cells in each extract. We analyzed three animals from each transgenic line and compared the levels of EGFP by densitometry after normalization to the signal for IV tubulin. While the results varied between experiments, the data suggest that the steady-state levels of EGFP in lines 45 and 85 (EGFP/tubulin ratios of 2.1 and 2.9) were greater than the levels in lines 46 and 49 (1.3 and 0.5). These results indicate that while the pattern of transgene expression from the FOXJ1 promoter fragment is relatively independent of integration site, the level of expression does not appear to correlate with the copy number of the transgene in these four lines.

Figure 4.

Analysis of FOXJ1-driven EGFP expression in four different transgenic lines. (A) Frozen sections were prepared from trachea, lung, and esophagus from each of the indicated lines. Images were then captured using the same exposure conditions. All four lines examined showed similar patterns and intensities of EGFP fluorescence. Scale bar, 50 m for all images. (B) Tracheal extracts were prepared from each of the four transgenic lines and wild-type littermate controls and analyzed by Western blotting. EGFP is present at varying levels in the transgenic lines and was not detected in the wild-type animals. Blots were reprobed with an antibody to IV tubulin as a control for the amount of ciliary protein loaded. The level of EGFP expressed appeared independent of the transgene copy number.

Full figure and legend (302K)

Figure 5.

Localization of EGFP expression by immunochemistry. Paraffin-embedded sections from FOXJ1/EGFP transgenic (A, B, C, D, E, G, I, and J) and wild-type (F, H, and K) animals were immunostained with the anti-GFP antibody, except that the section in C was probed with control rabbit IgG. EGFP was detected in ciliated cells of the trachea (A, B), conducting airways (D, E), and nasal epithelium (G, I, J) of the transgenic animals, while no specific staining was observed in the same tissues from wild-type animals. All animals were from line 85. Scale bars, A and C, 30 m; D, F, I, and K, 50 m; B, E, and J, 10 m.

Full figure and legend (462K)

Although the above experiments showed that in the FOXJ1/EGFP transgenic animals EGFP expression was restricted to the ciliated epithelium, the morphology of the frozen sections made localization at the cellular level difficult. To examine further the location of EGFP expression within the epithelium, we utilized two additional approaches. In the first approach, we used the affinity-purified rabbit anti-GFP antibody to detect EGFP expression in formaldehyde-fixed, paraffin-embedded sections. Ciliated cells in the trachea of FOXJ1/EGFP transgenic animals were specifically labeled with the anti-GFP antibody, while other cells in the epithelium were unlabeled (Figs. 5A and 5B). Ciliated cells from wild-type animals were not labeled with the anti-GFP antibody (not shown), and normal rabbit IgG at the same concentrations also did not label the ciliated cells (Fig. 5C). Similarly, immunostaining of lung sections demonstrated EGFP expression in ciliated cells throughout the airways (Figs. 5D and 5E), while other cell types and tissues from control animals were negative (Fig. 5F). In the nose and paranasal sinuses, EGFP was again specifically detected in the ciliated cells of the respiratory epithelium (Fig. 5G). This can be clearly seen in the transition region between respiratory and olfactory epithelium (Figs. 5I and 5J). While the majority of the olfactory epithelium was negative, we did observe staining of a small, round cell with a tube-like extension. Morphologically, these cells resemble regenerating neuroepithelial cells. Staining was particularly evident in the vomeronasal organ (data not shown). Tissue from control animals demonstrated only weak nonspecific background staining (Figs. 5H and 5K). Animals from all four lines exhibited similar staining, with the exception of line 46. Animals from line 46, while showing strong staining of almost all ciliated cells in nasal epithelium comparable to the other lines, showed few stained ciliated cells in the trachea and conducting airways. While the staining of individual ciliated cells was strong, the number of ciliated cells stained was clearly reduced, suggesting that this line was showing a variegated expression pattern²⁶.

In a second approach to localize EGFP expression at the cellular level, we embedded formaldehyde-fixed tissues in agarose and prepared thick sections using a Vibratome. This approach allowed the direct visualization of EGFP fluorescence, while providing improved ciliated cell morphology compared to frozen sections. We stained the cilia with an anti-IV tubulin antibody and viewed sections on a confocal microscope. As shown in Fig. 6, ciliated cells in the trachea, identified by the anti-IV tubulin antibody, clearly express EGFP, while the surrounding cell types are negative. Taken together, the above studies clearly demonstrate that elements within the FOXJ1 promoter fragment are capable of specifically directing expression of a transgene to the ciliated cells in vivo.

Figure 6.

Localization of EGFP expression in ciliated cells identified by IV tubulin immunostaining and confocal microscopy. Thick sections of a trachea from a FOXJ1/EGFP transgenic animal were immunostained with an anti-IV tubulin antibody (detected with Texas red secondary) and examined by fluorescence confocal microscopy. Two different images are shown (A, B). EGFP expression is clearly restricted to the ciliated cells. Animal is from line 49. Scale bar, 10 m.

Full figure and legend (167K)

Discussion

Cystic fibrosis patients suffer from chronic respiratory infections, which ultimately result in severe bronchiectasis and respiratory failure. Because most CF patients die of respiratory disease, the primary target for gene therapy is the lung. In the lung, CFTR is expressed at low levels in ciliated cells throughout the surface epithelium of the airways, and it has also been reported to be expressed at higher levels in the serous cells of the submucosal glands²⁷,²⁸. For gene therapy of CF, it would therefore be desirable to target expression of the wild-type CFTR to the ciliated cells of the airways and/or the serous cells of the submucosal glands. In a promising study by Zhang et al. it was recently demonstrated that respiratory syncytial virus specifically infects ciliated cells in well-differentiated cultures of human airway epithelial cells²⁹. However, most of the vectors currently being used for CF gene therapy studies have no specificity for these cell types. Further, many of the constructs used contain strong promoters from ubiquitously expressed genes, which will drive high levels of CFTR expression in most cell types. This may produce undesirable side effects, as high level expression of CFTR has been shown to have growth inhibitory effects on some cell lines¹²,¹³. Therefore the development of cell- or tissue-specific promoters would be useful to target the expression of CFTR to the appropriate cell type(s).

While the endogenous CFTR promoter might be considered a good candidate promoter for a gene therapy vector, a number of studies have demonstrated that the CFTR promoter is large and complex³⁰,³¹. Further, because the efficiency of gene transfer is likely to be low, it is possible that higher than normal levels of CFTR will be necessary for correction. Thus the low levels of CFTR expression from the endogenous promoter³² may not be sufficient to correct the disease phenotype. As an alternative, Suzuki et al. constructed a chimeric promoter consisting of a portion of the 5'-flanking region of the CFTR cDNA along with several cAMP response elements³³,³⁴. While this promoter was regulatable by cAMP, the tissue specificity was not evaluated. In another study, Zhou et al. were able to correct the intestinal defect in CF mice by expressing human CFTR in the intestinal epithelium under control of the fatty acid-binding protein promoter³⁵. This study demonstrated both the feasibility of gene therapy for CF and the successful use of a tissue-specific promoter. Another study utilized a 3.7-kb transcriptional element from the 5' region of the human surfactant protein C gene to direct expression of CFTR to the bronchiolar and alveolar epithelial cells of transgenic mice; no expression of CFTR was detected in the trachea or bronchial regions³⁶. Of particular relevance to the work reported here, Hu et al. have developed an expression cassette based on the promoter of the keratin 18 gene that is reported to direct expression of transgenes to the epithelium of many tissues that express CFTR. In one study, a K18 promoter construct was used to direct expression of -galactosidase in transgenic animals³⁷. Detailed examination of two lines that were selected for high level expression of -galactosidase demonstrated similar, but different, patterns of expression in the trachea and airways. In these animals, expression was observed in both the surface epithelium and the submucosal glands, sites of endogenous CFTR expression. More recently, the expression cassette has been modified to include 2.2 kb of 5' flanking sequence, the K18 promoter, the K18 first intron and enhancer, a translational enhancer from alfalfa mosaic virus, and the coding sequence of the transgene, followed by a 3' untranslated region containing intron 6, exon 7, the polyadenylation signal, and 272 bp of genomic sequence from the K18 gene³⁸,³⁹. This cassette, which contains approximately 4 kb of K18 sequence, was reported to have increased expression while maintaining specificity for epithelial cells. Thus, the K18-based expression cassette may be useful for targeting expression of CFTR to the epithelium. However, the large size of the expression cassette would prevent it from being incorporated into some vectors (e.g., retrovirus, AAV).

In this study, we have begun to characterize a ciliated cell-specific promoter that may be useful to provide targeted expression of CFTR. Analysis of FOXJ1/EGFP transgenic mice demonstrated that a small fragment of the human FOXJ1 promoter region is sufficient to target expression of a reporter gene to the ciliated epithelium. Expression of EGFP was easily detected in ciliated cells throughout the tracheal, bronchial, and nasal epithelium of transgenic animals, while basal and other cell types were negative. Detailed analyses of respiratory epithelium, using both immunohistochemistry and confocal microscopy, confirmed that EGFP expression in the transgenic mice was tightly restricted to the ciliated cells. By immunohistochemistry, we did observe staining of some cells in the olfactory epithelium of the mouse nose, especially in the vomeronasal organ. These cells may represent a particular subset or stage of development of olfactory epithelial cells. While it is clear that mature olfactory epithelial cells did not express detectable amounts of FOXJ1 in a previous study¹⁹, it is not clear if the expression of EGFP in these unidentified cells in the transgenic animals is aberrant or accurately represents the pattern of endogenous FOXJ1 expression. Further studies are under way to identify these cells. Other tissues that contain motile axonemes and have been previously shown to express FOXJ1, including brain, oviduct, and testis, also demonstrated strong EGFP fluorescence. The pattern of transgene expression from the FOXJ1 promoter therefore closely follows the pattern of endogenous FOXJ1 expression. This finding suggests that most, if not all, of the transcriptional control elements required for the tissue- and cell-specific expression of FOXJ1 are contained in our construct.

While many of the gene therapy trials for CF have been conducted using adenovirus, long-term correction of CF will likely require the stable integration of a CFTR expression cassette. Thus the ability of a promoter to maintain activity and specificity following integration at random sites throughout the genome will be important to the overall success of a vector. In this study, four FOXJ1/EGFP transgenic lines that expressed EGFP were identified. All four lines showed very similar patterns of EGFP expression by fluorescence microscopy and immunohistochemistry, with easily detectable levels of EGFP in ciliated cells of the tracheal and nasal epithelium, and three of the four lines showed strong signals in ciliated cells throughout the bronchial epithelium. Western blot analysis of tracheal extracts demonstrated the presence of EGFP protein in tracheal extracts from all four lines; however, the steady-state level of EGFP expression did not appear to correlate with copy number of the transgene in these studies. Because the EGFP protein has been reported to have a long half-life in vivo, these results may not reflect directly the level of transcription. However, the number of independent lines that expressed EGFP (4/7) and the similarities in the expression pattern between the four lines suggest that expression from the FOXJ1 promoter is not highly dependent on the site of integration.

The FOXJ1 promoter fragment characterized in these studies has several features that may make it useful for future gene therapy studies of CF and other airway diseases. First, the size of the fragment utilized, about 900 bp, is relatively small, especially compared to other tissue-specific promoters that have been proposed for gene therapy of CF (e.g., 3.7 kb for the SP-C promoter and 4 kb for the K18 promoter). This is an advantage because the CFTR cDNA is relatively large (4.6 kb) compared to the packaging size of some vectors. The small size of the FOXJ1 fragment utilized in these studies will allow the promoter to be used in most vectors, including retrovirus. Future studies that identify specific regulatory elements of the promoter may allow even further reduction in the size of the promoter by deleting unnecessary sequence.

Second, the FOXJ1 promoter fragment retains tissue and cell specificity in a transgenic model, suggesting that administration of a gene therapy vector to the lung would result in CFTR expression only in the ciliated cells lining the airway surface. This would place CFTR in direct contact with the airway surface liquid and, in addition, would eliminate expression of CFTR in other cell types, which may be detrimental. The FOXJ1 promoter will not drive expression of CFTR in submucosal glands, as will the K18 promoter. However, at present it is not known if CFTR expression is required in ciliated cells, serous cells, or both for gene therapy to be successful.

Third, the expression level of a transgene from the promoter was relatively robust, as EGFP was readily detected by Western blotting, immunohistochemistry, and direct fluorescence microscopy. Although it has been demonstrated that expression of wt CFTR in 5–10% of CF airway epithelial cells can correct the chloride secretion defect⁴⁰, the percentage of cells that must be corrected and the level of expression required to achieve a therapeutic effect are unknown. These two parameters are likely to be interdependent. In the FOXJ1/EGFP transgenic mice, the level of EGFP expression is probably higher than the endogenous level of CFTR expression, which may be as low as 1–3 copies per cell³². Additional experiments will be required to determine if the levels of expression from the FOXJ1 promoter are sufficient to correct the CF phenotype. However, if the level of CFTR expression from the FOXJ1 promoter proves to be insufficient for correction of the CF phenotype, due to a low level of gene transfer, additional elements could conceivably be added to the promoter to increase transcription. Alternatively, many of the genes coding for ciliated cell-specific proteins are expressed at very high levels during ciliated cell differentiation⁴¹, and their promoters may also be useful for gene therapy.

Finally, the ability to express CFTR specifically in a single cell type will provide useful information regarding the normal function of CFTR. Experiments in progress will determine if expression of CFTR in the ciliated cells is capable of correcting the ion transport abnormalities in the nasal epithelium of CF mice. These experiments will provide an increased understanding of CFTR function and lead to improved strategies for gene therapy of CF.

Materials and methods

Cloning of human FOXJ1 promoter

Approximately 4.6 kb of human genomic DNA containing the FOXJ1 promoter region was amplified from human genomic DNA using primers GGCTGGAGGATCGCTCCTGTGGAA and CGCTGAACCTGGCACCTGGTGGTA (based on Accession No. AC018665) and Pfu DNA polymerase (Stratagene, La Jolla, CA). The 1-kb promoter fragment used in these studies was amplified from the genomic fragment using forward primer TATGGATCCTGAGCCGAGCCGGGACTTAG and reverse primer TATGGATCCGTCCCCAGACTCCCGTTACACG, based on the sequence reported by Murphy et al.²¹. Both primers contained a BamHI linker site for subcloning. The identity of the cloned promoter fragment was confirmed by sequencing.

Generation of FOXJ1/EGFP transgenic mice

The backbone of the construct used for transgenic mice generation was a plasmid known as TG-1, which was developed in the UNC Animal Models Core Facility. This vector is a direct modification of vector pTTR1EXV3²², which was used to direct liver-specific gene transcription in transgenic mice. The vector contains two noncoding exons and a heterologous intron, elements that have been shown to increase expression in transgenic mice²³. For the development of the vector pTG1, the liver-specific promoter in pTTR1EXV3 was removed and replaced with a multiple cloning site, which allowed for the insertion of the FOXJ1 promoter fragment. This fragment is immediately followed by 10 bp upstream of the transcription start site, exon 1 (104 bp), intron 1 (947 bp), and 10 bp of exon 2 of the mouse transthyretin (TTR) gene. Addition of a second polylinker following the exon 2 spice acceptor site at the StuI site in pTTR1EXV3 allowed insertion of the EGFP cDNA. Further modifications of pTTR1EXV3 3' of the SV40 early region polyadenylation sequence generated unique restriction sites that allowed for the entire expression cassette (approximately 3170 bp) to be removed from plasmid sequence (pGEM) before nuclear injection. Thus, the transcription of EGFP in the construct utilized in this paper starts in the first exon of mouse TTR. The transcript extends through the first intron of mouse TTR, into the first few bases of TTR exon 2 and then into the EGFP cassette.

Transgenic mice were generated by standard procedures utilizing pronuclear injection in collaboration with the UNC Animal Models Core Facility. Injections were carried out on eggs of C3H C57Bl/6J F1 background. The mice were maintained on this mixed background throughout the experiments. Potential founder animals were screened by Southern blot analysis using an EGFP-specific fragment as probe. Animals shown to carry the transgene were bred and progeny screened for transmission of the transgene by PCR. Animals positive for the presence of the transgene were analyzed further for expression of EGFP. All animal protocols were approved by the Institutional Animal Care and Use Committee in accordance with established guidelines for animal care.

Analysis of EGFP fluorescence

FOXJ1/EGFP heterozygote transgenic mice (positive animals) and wild-type littermates (negative animals) were sacrificed by CO₂ asphyxia. Samples of seven organs (trachea, lung, liver, muscle, brain, nasal septum, and ovary/testis) were obtained and fixed in 4% PFA for 30 min at room temperature. The tissue samples were rinsed briefly in 1 PBS and rapidly frozen in Optimal Cutting Temperature Compound (Sakura Finetechnical Co., Tokyo, Japan) using solid CO₂. The frozen tissue samples were stored at -80°C until processed for evaluation of fluorescence. Thick sections (8 m) were prepared and mounted with glycerol vinyl alcohol aqueous mounting solution (Zymed Laboratories, Inc., San Francisco, CA). Slides were viewed under a Leitz DM IRB fluorescence microscope (Leica, Heidelberg, Germany) and images captured with a CCD camera (QImaging, Canada).

Western blotting

Tracheas were removed from animals sacrificed by CO₂ asphyxia and dissected free of all nontracheal tissue. The excised trachea was split lengthwise and ground in 2 sample buffer (125 mM Tris–HCl, pH 6.8, 20% glycerol, 4% SDS) using a mortar and pestle (pellet pestle motor; Kimble–Kontes Glass, Vineland, NJ). The suspension was boiled 5 min and centrifuged at 16,000g for 5 min at 4°C. The supernatant was collected and passed through a 20-gauge needle 10 times. The resulting solution was again centrifuged at 16,000g for 5 min at 4°C. The supernatant was collected and stored at -80°C. Protein concentration was determined using Pierce's BCA protein assay kit (Pierce Chemical, Rockford, IL) according to the manufacturer's protocol. Due to high SDS levels, all samples were diluted 1:8. Twenty-five micrograms of each sample was electrophoresed on a 4–12% Tris–Glycine gel (Invitrogen) and transferred to a nitrocellulose membrane (Amersham Pharmacia Biotech). The membrane was blocked in 5% nonfat dry milk in PBS-T (0.1% Tween 20 in PBS) for 1 h and incubated with primary antibody (affinity-purified rabbit polyclonal to GFP, 1:2000, Abcam Ltd., Cambridge, UK, or anti IV tubulin, 1:1500, BioGenex, San Ramon, CA) in 1% nonfat dried milk in PBS-T for 2 h at room temperature. The membrane was washed in wash buffer (0.1% nonfat dried milk in PBS-T) and the appropriate secondary antibody was applied at a concentration of 1:10,000 and incubated 1 h at room temperature. Development of the membrane was performed using the appropriate horseradish peroxidase-conjugated secondary antibody and ECL Plus detection reagent (Amersham Pharmacia Biotech) according to the manufacturer.

Localization of EGFP expression by immunostaining

Tissue samples obtained as above were fixed in 4% PFA for 6 h at 4°C and stored in 70% ethanol. For nasal sections, samples were decalcified by incubation in Formical-4 (Decal Chemical Corp., Congers, NY) for 24 h. Lungs were inflation fixed using a syringe and gentle pressure. Paraffin sections (5 m) were prepared by standard processing techniques. Slides were deparaffinized with xylene, rehydrated in an ethanol series, and incubated with 0.6% H₂O₂ in MeOH for 30 min to inactivate endogenous peroxidase activity. The slides were washed with PBS and a hydrophobic ring was drawn around the tissue section using a PAP pen (Sigma, St. Louis, MO). The sections were incubated in blocking solution consisting of 5% normal goat serum (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), 1% gelatin (Sigma), 1% bovine serum albumin (Sigma) dissolved in PBS-T for 1 h at room temperature. The affinity-purified rabbit anti-GFP antibody (Abcam Ltd.) was added at a dilution of 1:250 in blocking solution and allowed to incubate 1 h at room temperature. The sections were washed three times with a 1:3 dilution of PBS-T to blocking solution (diluent). The sections were incubated in Biotin–SP–AffinityPure goat anti-rabbit IgG (H+L) (Jackson ImmunoResearch Laboratories, Inc.) at a 1:2000 dilution for 1 h, washed, and incubated for 1 h with peroxidase–streptavidin (Jackson ImmunoResearch Laboratories, Inc.) at a dilution of 1:1000. Immunoreactivity was visualized with diaminobenzine (Fast 3,3'-Diaminobenzine; Sigma) for 30 min. Sections were counterstained with light green solution, dehydrated, and coverslipped with Permount. Images were captured as above.

Analysis of EGFP fluorescence by confocal microscopy

Tissues were obtained from transgenic animals, fixed in 4% PFA overnight at 4°C, and stored in 70% EtOH at room temperature until use. Tissues were embedded in 2% agarose and sliced using a Vibratome (Leica VT 1000S Vibratome) at a thickness between 40 and 70 m. These 40- to 70-m sections were stored protected from light in PBS at 4°C. For immunostaining, sections were placed into a 96-well plate and incubated in 100 l blocking solution for 2 h at room temperature. The anti-IV tubulin antibody (BioGenex) was added at a 1:500 dilution in blocking solution and incubated at 4°C for 18 h. The sections were washed with a 1:3 dilution of PBS-T to blocking solution and incubated with Texas red-labeled goat anti-mouse IgG (Jackson ImmunoResearch Laboratories, Inc.) at a dilution of 1:500 and incubated for 2 h completely protected from light. The sections were briefly washed with ddH₂O, mounted using Vecta-Shield Mounting Medium (Vector Laboratories, Inc., Burlingame, CA), and sealed with nail polish. Images were captured using a Leica 4D laser scanning confocal microscope (Leica) and a 100 oil immersion lens.

References

1. Welsh, M. J., Tsui, L.-C., Boat, T. F., and Beaudet, A. L. (1995). Cystic fibrosis. In: Scriver C. R., Beaudet A. L., Sly W. S., and Valle D. (Eds.), The Metabolic and Molecular Basis of Inherited Disease. McGraw–Hill: New York. pp. 3799–3876.
2. Davis, P. B., Drumm, M., and Konstan, M. W. (1996). Cystic fibrosis. Am. J. Respir. Crit. Care Med. 154: 1229–1256. | PubMed | ISI | ChemPort |
3. Boucher, R. C. (2002). An overview of the pathogenesis of cystic fibrosis lung disease. Adv. Drug Delivery Rev. 54: 1359–1371. | Article | ChemPort |
4. Rich, D. P., et al. (1990). Expression of cystic fibrosis transmembrane conductance regulator corrects defective chloride channel regulation in cystic fibrosis airway epithelial cells. Nature 347: 358–363. | Article | PubMed | ISI | ChemPort |
5. Drumm, M. L., et al. (1990). Correction of the cystic fibrosis defect in vitro by retrovirus-mediated gene transfer. Cell 62: 1227–1233. | Article | PubMed | ISI | ChemPort |
6. Boucher, R. C. (1999). Status of gene therapy for cystic fibrosis lung disease. J. Clin. Invest. 103: 441–445. | PubMed | ISI | ChemPort |
7. Wilson, J. M. (1995). Gene therapy for cystic fibrosis: Challenges and future directions. J. Clin. Invest. 96: 2547–2554. | PubMed | ISI | ChemPort |
8. Parsons, D. W., Grubb, B. R., Johnson, L. G., and Boucher, R. C. (1998). Enhanced in vivo airway gene transfer via transient modification of host barrier properties with a surface-active agent. Hum. Gene Ther. 9: 2661–2672. | PubMed | ChemPort |
9. Wang, G., et al. (2000). Increasing epithelial junction permeability enhances gene transfer to airway epithelia in vivo. Am. J. Respir. Cell Mol. Biol. 22: 129–138. | PubMed | ISI | ChemPort |
10. Coyne, C. B., Kelly, M. M., Boucher, R. C., and Johnson, L. G. (2000). Enhanced epithelial gene transfer by modulation of tight junctions with sodium caprate. Am. J. Respir. Cell Mol. Biol. 23: 602–609. | PubMed | ChemPort |
11. Zabner, J., et al. (2000). Adeno-associated virus type 5 (AAV5) but not AAV2 binds to the apical surfaces of airway epithelia and facilitates gene transfer. J. Virol. 74: 3852–3858. | Article | PubMed | ISI | ChemPort |
12. Stutts, M. J., et al. (1993). Functional consequences of heterologous expression of the cystic fibrosis transmembrane conductance regulator in fibroblasts. J. Biol. Chem. 268: 20653–20658. | PubMed | ChemPort |
13. Schiavi, S. C., et al. (1996). Biosynthetic and growth abnormalities are associated with high-level expression of CFTR in heterologous cells. Am. J. Physiol. 270: C341–C351. | PubMed | ISI | ChemPort |
14. Pastore, L., et al. (1999). Use of a liver-specific promoter reduces immune response to the transgene in adenoviral vectors. Hum. Gene Ther. 10: 1773–1781. | Article | PubMed | ChemPort |
15. Paterna, J. C., Moccetti, T., Mura, A., Feldon, J., and Bueler, H. (2000). Influence of promoter and WHV post-transcriptional regulatory element on AAV-mediated transgene expression in the rat brain. Gene Ther. 7: 1304–1311. | Article | PubMed | ChemPort |
16. Check, E. (2002). A tragic setback. Nature 420: 116–118. | Article | PubMed | ISI | ChemPort |
17. Randell, S. H. (1992). Progenitor–progeny relationships in airway epithelium. Chest 101: 11S–16S. | PubMed | ChemPort |
18. Lim, L., Zhou, H., and Costa, R. H. (1997). The winged helix transcription factor HFH-4 is expressed during choroid plexus epithelial development in the mouse embryo. Proc. Natl. Acad. Sci. USA 94: 3094–3099. | Article | PubMed | ChemPort |
19. Blatt, E. N., Yan, X. H., Wuerffel, M. K., Hamilos, D. L., and Brody, S. L. (1999). Forkhead transcription factor HFH-4 expression is temporally related to ciliogenesis. Am. J. Respir. Cell Mol. Biol. 21: 168–176. | PubMed | ChemPort |
20. Hackett, B. P., Brody, S. L., Liang, M., Zeitz, I. D., Bruns, L. A., and Gitlin, J. D. (1995). Primary structure of hepatocyte nuclear factor/forkhead homologue 4 and characterization of gene expression in the developing respiratory and reproductive epithelium. Proc. Natl. Acad. Sci. USA 92: 4249–4253. | Article | PubMed | ChemPort |
21. Murphy, D. B., Seemann, S., Wiese, S., Kirschner, R., Grzeschik, K. H., and Thies, U. (1997). The human hepatocyte nuclear factor 3/fork head gene FKHL13: Genomic structure and pattern of expression. Genomics 40: 462–469. | Article | PubMed | ChemPort |
22. Wu, H., Wade, M., Krall, L., Grisham, J., Xiong, Y., and Van Dyke, T. (1996). Targeted in vivo expression of the cyclin-dependent kinase inhibitor p21 halts hepatocyte cell-cycle progression, postnatal liver development and regeneration. Genes Dev. 10: 245–260. | PubMed | ISI | ChemPort |
23. Choi, T., Huang, M., Gorman, C., and Jaenisch, R. (1991). A generic intron increases gene expression in transgenic mice. Mol. Cell. Biol. 11: 3070–3074. | PubMed | ISI | ChemPort |
24. Ostrowski, L. E., Zakel, K., Hutchins, J. R., and O'Neal, W. (2002). Targeting expression of CFTR to the airway epithelium: Identification and characterization of a ciliated cell-specific promoter for use in gene therapy vectors. Pediatric Pulmonol 24 (Suppl.): 254.
25. Renthal, R., Schneider, B. G., Miller, M. M., and Luduena, R. F. (1993). Beta IV is the major beta-tubulin isotype in bovine cilia. Cell Motil. Cytoskeleton 25: 19–29. | Article | PubMed | ChemPort |
26. Alami, R., et al. (2000). Beta-globin YAC transgenes exhibit uniform expression levels but position effect variegation in mice. Hum. Mol. Genet. 9: 631–636. | Article | PubMed | ISI | ChemPort |
27. Puchelle, E., et al. (1992). Differential localization of the cystic fibrosis transmembrane conductance regulator in normal and cystic fibrosis airway epithelium. Am. J. Respir. Cell Mol. Biol. 7: 485–491. | PubMed | ISI | ChemPort |
28. Engelhardt, J. F., et al. (1992). Submucosal glands are the predominant site of CFTR expression in the human bronchus. Nat. Genet. 2: 240–248. | Article | PubMed | ISI | ChemPort |
29. Zhang, L., Peeples, M. E., Boucher, R. C., Collins, P. L., and Pickles, R. J. (2002). Respiratory syncytial virus infection of human airway epithelial cells is polarized, specific to ciliated cells, and without obvious cytopathology. J. Virol. 76: 5654–5666. | Article | PubMed | ChemPort |
30. Moulin, D. S., Manson, A. L., Nuthall, H. N., Smith, D. J., Huxley, C., and Harris, A. (1999). In vivo analysis of DNase I hypersensitive sites in the human CFTR gene. Mol. Med. 5: 211–223. | PubMed | ChemPort |
31. Mogayzel, P. J. Jr., and Ashlock, M. A. (2000). CFTR intron 1 increases luciferase expression driven by CFTR 5'-flanking DNA in a yeast artificial chromosome. Genomics 64: 211–215. | Article | PubMed | ChemPort |
32. Trapnell, B. C., et al. (1991). 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. | Article | PubMed | ChemPort |
33. Suzuki, M., Singh, R. N., and Crystal, R. G. (1996). Regulatable promoters for use in gene therapy applications: Modification of the 5'-flanking region of the CFTR gene with multiple cAMP response elements to support basal, low-level gene expression that can be upregulated by exogenous agents that raise intracellular levels of cAMP. Hum. Gene Ther. 7: 1883–1893. | PubMed | ISI | ChemPort |
34. Suzuki, M., Singh, R. N., and Crystal, R. G. (1997). Ability of a chimeric cAMP-responsive promoter to confer pharmacologic control of CFTR cDNA expression and cAMP-mediated Cl^- secretion. Gene Ther. 4: 1195–1201. | Article | PubMed | ChemPort |
35. Zhou, L., Dey, C. R., Wert, S. E., DuVall, M. D., Frizzell, R. A., and Whitsett, J. A. (1994). Correction of lethal intestinal defect in a mouse model of cystic fibrosis by human CFTR. Science 266: 1705–1708. | Article | PubMed | ISI | ChemPort |
36. Whitsett, J. A., et al. (1992). Human cystic fibrosis transmembrane conductance regulator directed to respiratory epithelial cells of transgenic mice. Nat. Genet. 2: 13–20. | Article | PubMed | ISI | ChemPort |
37. Chow, Y. H., et al. (2000). Targeting transgene expression to airway epithelia and submucosal glands, prominent sites of human CFTR expression. Mol. Ther. 2: 359–367. | Article | PubMed | ISI | ChemPort |
38. Koehler, D. R., et al. (2000). A human epithelium-specific vector optimized in rat pneumocytes for lung gene therapy. Pediatr. Res. 48: 184–190. | PubMed | ChemPort |
39. Koehler, D. R., et al. (2001). Targeting transgene expression for cystic fibrosis gene therapy. Mol. Ther. 4: 58–65. | Article | PubMed | ChemPort |
40. Johnson, L. G., Olsen, J. C., Sarkadi, B., Moore, K. L., Swanstrom, R., and Boucher, R. C. (1992). Efficiency of gene transfer for restoration of normal airway epithelial function in cystic fibrosis. Nat. Genet. 2: 21–25. | Article | PubMed | ISI | ChemPort |
41. Andrews, K. L., Nettesheim, P., Asai, D. J., and Ostrowski, L. E. (1996). Identification of seven rat axonemal dynein heavy chain genes: Expression during ciliated cell differentiation. Mol. Biol. Cell. 7: 71–79. | PubMed | ISI | ChemPort |

Acknowledgements

The authors thank Drs. J. Harkema and A.-S. LaMantia for helpful discussions of nasal epithelial cell morphology; Drs. M. Chua and T. Oliver of the Michael Hooker Microscopy Facility; Kim Burns, Tracy Eldred, and Elizabeth Andrews for excellent histology assistance; Miriam Kelly Vanhook and Jacyln Stonebraker for technical assistance; and Dr. R. C. Boucher for helpful discussions and support of this project. This work was funded in part by Cystic Fibrosis Foundation Grant S880 and National Institutes of Health Grant HL070199.