Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

It has been shown that the inducible form of nitric oxide (NO) synthase (NOS2) is expressed constitutively in normal human airway epithelium (1). Recent evidence suggests that this constitutive expression of NOS2 messenger RNA (mRNA) and protein are decreased in airway epithelial cells that are lacking either expression or function of the cystic fibrosis (CF) transmembrane conductance regulator (CFTR) (2, 3). The potential consequences of the loss of epithelial NOS2 expression include diminished antibacterial defenses and altered transepithelial ion transport regulation (4, 5). It is unclear, however, why NOS2 expression is reduced in CF epithelial cells, especially given the inflammatory nature of CF. An issue that needs to be addressed in relation to decreased NOS2 expression in CF airway epithelium is whether NOS2 expression is altered by the presence of active CFTR.

CFTR has been shown to influence a variety of cellular functions both directly and indirectly. Primarily, CFTR has been shown to regulate transepithelial ion transport by influencing the activity of other ion channels such as the epithelial sodium channel (ENaC) (6, 7), the outwardly rectifying chloride channel (ORCC) (8), and the inwardly rectifying potassium channel (ROMK2) (9). Proposed mechanisms for CFTR-dependent regulation of these channels include direct interaction between CFTR and ENaC (6), direct interaction of the nucleotide binding fold-1 of CFTR and ROMK2 (9), and the facilitation of adenosine triphosphate (ATP)-dependent stimulation of ORCC activity by CFTR-mediated ATP release (8).

Because NO production has been shown to effectively influence both transepithelial chloride and sodium transport in several systems (4, 5, 10), it is plausible that CFTR conveys some regulation to NOS2 production to further regulate basal ion transport properties. To begin identifying mechanisms involved in reduced NOS2 expression in CF airway epithelia despite chronic inflammation, it must be determined first whether NOS2 expression is altered by the presence of CFTR activity. The goal of this paper is to exploit two model systems of CF to determine the influence of CFTR function and expression on NOS2 production. First, a human tracheal epithelial cell line (9/HTEo-) that has been stably transfected to overexpress the CFTR regulatory (R) domain, resulting in blocked CFTR-mediated chloride transport (11), was studied to directly measure NOS2 mRNA levels and NO production. Second, an in vivo model consisting of mice deficient in CFTR expression (cftr /), wild-type mice (cftr +/+), and cftr / mice that have had human CFTR introduced to the intestinal epithelium using the fatty acid binding protein (FABP) promoter (FABP-hcftr) (12) was utilized to examine the effects of differential CFTR expression on NOS2 production in separate tissues of the same animal.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

9/HTEo- cells overexpressing the CFTR R domain (pCEP-RF) and mock-transfected 9/HTEo- cells (pCEP2) were a generous gift from the lab of Dr. Pamela B. Davis, Case Western Reserve University, Cleveland, OH. Cells were cared for as previously described (11). Briefly, 9/HTEo- cells were grown at 37°C in 95% O₂/5% CO₂ to confluency in Dulbecco's modified Eagle's medium with 10% fetal bovine serum, L-glutamine (2.5 mM), and hygromycin (150 µg/ml) on Vitrogen-coated, 30-mm tissue culture plates.

Mice

Mice lacking CFTR expression (CFTR^tm1Unc) and FABP-hcftr mice (CFTR^tm1Unc-TgN[FABPCFTR]#Jaw) were obtained from Jackson Laboratories (Bar Harbor, ME). The cftr +/+ mice were siblings of cftr / mice. All mice used were between 10 and 12 wk of age. CF mice were fed a liquid diet as described by Eckman and colleagues (13). Mice were cared for in accordance with Case Western Reserve University IACUC guidelines by the CF Animal Core Facility.

Reverse Transcriptase/Polymerase Chain Reaction Analysis of NOS2 mRNA Expression

Total RNA from 9/HTEo- pCEP2 and pCEP-RF cell lines was isolated by phenol guanidine-isothiocyanate extraction using TRIzol Reagent (Life Technologies, Inc., Gaithersburg, MD) according to manufacturer's protocol. RNA was stored at 80°C until used. Total RNA was reverse transcribed to complementary DNA (cDNA) at 42°C for 1 h. In a 20-µl volume, the reverse transcriptase (RT) reaction included 1 mM of each deoxynucleotide triphosphate (dNTP), 18 mM Tris-HCl, 45 mM KCl, 0.1 mg/ml bovine serum albumin, 40 units ribonuclease inhibitor, 100 pmol random hexamers, 2.5 mM MgCl₂, 200 units Moloney murine leukemia virus (M-MLV) RT, and 0.3 µg total RNA. cDNA was amplified by polymerase chain reaction (PCR) in a 50-µl volume containing 0.5 mM of each dNTP, 20 mM Tris-HCl, 50 mM KCl, 1.2 µM of each NOS2 primer or 200 nM of each glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primer, 3 mM MgCl₂, 2.5 units of Taq polymerase, and 4 µl cDNA. All reagents were obtained from Life Technologies, Inc. PCR primer sequences used were as follows: human NOS2 forward primer: 5'-TCTGTGCCTTTGCTCATGAC-3', human NOS2 reverse primer: 5'-CATGGTGAACACGTTCT TGG-3' (14); GAPDH forward primer: 5'-CCATGGAGAAGGCTGGGG-3', and GAPDH reverse primer: 5'-CAAAGTTGTCATGGATGA-3'. The cDNA was denatured for 2 min at 95°C followed by 40 cycles of amplification. Each cycle consisted of denaturation at 95°C for 35 s, annealing at 58°C for 35 s, and primer extension at 72°C for 45 s. A 30-cycle extension for 5 min at 72°C followed the 40-cycle program with each cycle consisting of denaturation at 95°C for 1 min, annealing at 62°C for 1 min, and primer extension at 72°C for 90 s. A final extension for 5 min at 72°C followed the 30-cycle program. A control reaction using GAPDH primers was performed for each cDNA sample using the respective conditions described previously. Semiquantification of RT-PCR products was accomplished by densitometry using Kodak Digital Science 1D software. RT-PCR reactions were performed on each sample to amplify both NOS2 and GAPDH message. Quantification of NOS2 message is reported as relative band intensity compared with GAPDH for each sample.

Direct NO Measurement from 9/HTEo- Cells

For studies of NO production, 9/HTEo- cells (pCEP2 and pCEP-RF) were treated with lipopolysaccharide (LPS) (5 µg/ml) from Pseudomonas aeruginosa (Calbiochem, San Diego, CA) which was added to the media 4 to 5 h before assay. NO assays were performed with an Iso-NO NO meter utilizing the 2 mM Iso-NOP NO sensor (World Precision Instruments, Sarasota, FL). Measurements were taken at 37°C by placing both the NO sensor and samples in an incubator. Data were recorded on a MacLab/4e data acquisition system from Advanced Instruments (Norwood, MD). The Iso-NO meter was calibrated according to manufacturer's instructions using the chemical method of NO production.

Nasal Transepithelial Potential Difference and Short-Circuit Current Measurements

Nasal transepithelial potential difference (TEPD) measurements and short-circuit current (I_sc) analysis of transepithelial ion transport were performed as previously described (15).

NOS2-Specific Immunostaining

Nasal epithelium and ileum were excised from cftr /, cftr +/+, and FABP-hcftr mice, paraffin blocked and sectioned as previously described (3). Sections were deparafinized, solubilized in ice-cold methanol for 5 min, and placed in 2% goat immunoglobulin (Ig) in phosphate-buffered saline (PBS) for 2 h. Antibody against mouse and human NOS2 was obtained from Calbiochem (La Jolla, CA) and incubated with the samples at 4°C overnight at a dilution of 1:300 in PBS. Samples were washed four times in PBS for 10 min per wash. Goat antirabbit IgG conjugated to alkaline phosphatase was diluted 1:200 in PBS and incubated with samples for 2 h at 37°C. Samples were washed as before and stained for 20 min in Vector Red from Novacastra Laboratories Ltd. (Newcastle, UK) according to manufacturer's instructions and slides were counterstained with hematoxylin. Images were attained using ImagePro imaging software.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

NOS2 mRNA Expression in 9/HTEo- Cells Overexpressing the CFTR R Domain

To examine the correlation between CFTR function and production of NOS2 mRNA in epithelial cells, a 9/HTEo- cell line model of CF was used. It has previously been shown that overexpression of the CFTR regulatory domain in 9/HTEo- cells blocks basal and cyclic adenosine monophosphate (cAMP)-stimulated chloride secretion through the inhibition of CFTR activity (11, 19). This model system allowed us to determine the effects of lost CFTR function on NOS2 activity and expression in epithelial cells by comparing the R domain expressing 9/HTEo- cells (pCEP-RF) with a mock-transfected control cell line (pCEP2). NOS2 mRNA was essentially undetectable by RT-PCR in pCEP-RF cells, whereas very strong product was obtained with the control pCEP2 cells in both the presence and absence of LPS (5 µg/ml) treatment (Figure 1A). The amount of NOS2 RT-PCR product relative to GAPDH RT-PCR product in each of the cell lines was quantified by measuring band intensity using densitometry (Figure 1B). All samples were treated with LPS to maximize the opportunity for NOS2 mRNA production. With GAPDH intensity set at 1.0, relative NOS2 mRNA product was measured at 0.8 ± 0.2 relative units (n = 14) in pCEP2 9/HTEo- cells, whereas relative band intensity was measured at only 0.1 ± 0.1 relative units (n = 15; P = 0.003) in pCEP-RF 9/HTEo- cells. Although it still remains unclear whether the decreased expression of NOS2 message is due to a lack of transcription or decreased mRNA stability, it is clear that the CF-related phenomenon of decreased epithelial NOS2 protein production originates at the RNA level. These data are consistent with previous findings by Meng and associates (2).

View larger version (15K): [in this window] [in a new window]   Figure 1.   Reduced NOS2 expression and NO production in 9/HTEo- cells that lack CFTR activity. (A) RT-PCR of NOS2 message in 9/HTEo- cells that exhibit CF phenotype (pCEP-RF) due to overexpression of the CFTR R domain and mock-transfected controls (pCEP2). Separate RT-PCR reactions specific for GAPDH message were performed on the same RNA samples as controls for RNA integrity. Cells treated with LPS (5 µg/ml) were treated 4 to 5 h before RNA isolation. (B) Quantification of NOS2 mRNA production relative to GAPDH mRNA in pCEP2 and pCEP-RF 9/HTEo- cells. Band intensities of RT-PCR products were determined by densitometry and compared with band intensities for GAPDH-specific RT-PCR reaction products from the same RNA samples. GAPDH band intensities are set at 1.0 and intensities of NOS2 RT-PCR products are reported as relative band intensity compared with GAPDH. All samples were treated with LPS (5 µg/ml) as described above. The number of experiments (n) is given in parentheses above each bar. *P = 0.003 compared with pCEP-RF NOS2 RT-PCR product relative intensity as determined by t test. (C) NO production was measured using an Iso-NOP NO meter 4 to 5 h after treatment with LPS (5 µg/ml). Concentrations of NO were determined by comparing voltage values against a standard curve that was developed using the chemical method of NO production described by the manufacturer. Values are reported as fold increase in NO production in samples compared with NO production in pCEP2 9/HTEo- cells not treated with LPS. Average values for each cell type are shown by solid bars. The number of experiments (n) is 14 for each cell type. *P < 0.0001 compared with pCEP2 NO value by t test.

NO production was measured in pCEP2 cells and pCEP-RF cells in response to LPS. It was found that pCEP2 cells produce a 4.0 ± 0.4-fold increase in NO production (n = 14); P < 0.00001 compared with basal NO production in pCEP2 cells not treated with LPS (Figure 1C). However, pCEP-RF cells exhibited only a 0.7 ± 0.1-fold increase in NO production (n = 14) under these conditions. Although the RT-PCR data show that NOS2 message is present before stimulation with LPS, actual NO production requires LPS treatment. These data support our hypothesis that the presence of functional CFTR influences NOS2 expression and NO production in response to inflammatory stimuli.

Electrical Characterization of cftr +/+, cftr /, and FABP-hcftr Mice

Responses to chloride-free Ringers and to the adenylate cyclase agonist forskolin (FSK) were measured by nasal TEPD in cftr +/+, cftr /, and FABP-hcftr mice to determine the level of functional CFTR activity in the nasal epithelium of these mice. The nasal TEPD in cftr +/+ mice hyperpolarized 10.8 ± 2.1 mV (n = 5) in response to chloride-free Ringers and further hyperpolarized 3.7 ± 3.3 mV (n = 5) in response to FSK, indicating normal CFTR activity (16). Nasal TEPD values, however, depolarized in cftr / mice in response both to chloride-free Ringers by 4.4 ± 1.2 mV (n = 4) and to FSK by 0.7 ± 1.8 mV (n = 4), consistent with a lack of functional CFTR. The nasal TEPD values of FABP-hcftr mice, which have human CFTR reintroduced into the intestinal epithelium, resemble those of cftr / mice. FABP-hcftr nasal TEPD values further depolarized 3.4 ± 1.5 mV (n= 4) in response to chloride-free Ringers and 2.4 ± 2.8 mV (n = 4) in response to FSK, suggesting a lack of significant levels of functional CFTR in the nasal epithelium (Figure 2).

View larger version (23K): [in this window] [in a new window]   Figure 2.   Comparison of nasal TEPD values in cftr +/+, cftr /, and FABP-hcftr mice. Nasal TEPD traces were averaged and are presented as changes in TEPD in response to chloride-free Ringers and the addition of FSK (10 µM). The number of traces averaged (n) is shown in the key in parentheses. Amiloride (100 µM) was present throughout the entire trace in each experiment. Error bars represent standard error of the mean (SEM).

To assess the level of CFTR activity in the intestines, I_sc values were measured in response to FSK in cftr +/+, cftr /, and FABP-hcftr mice in sections of ileum (Figure 3). The cftr +/+ mice showed a change in I_sc of 37.1 ± 4.0 µA/cm² (n = 3) in response to FSK, whereas cftr / mice had only a I_sc of 1.6 ± 0.5 µA/cm² (n = 3), P = 0.02 due to FSK. Unlike the nasal epithelium, however, measurements of transepithelial chloride transport across the ileum of FABP-hcftr mice showed complete correction of electrical responses to FSK. The FABP-hcftr mice showed a I_sc of 31.7 ± 10.8 µA/cm² (n = 4) when FSK was added. These data demonstrate a differential correction of CFTR- dependent electrical responses between the corrected intestinal epithelium and the uncorrected nasal epithelium in the FABP-hcftr mice.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Comparison of Isc measurements across excised sections of ileum from cftr +/+, cftr /, and FABP-hcftr mice. Raw traces of Isc measurements across excised ileum for (A) cftr +/+, (B) cftr /, and (C) FABP-hcftr mice. (D) Averaged changes in Isc in response to FSK (10 µM). Values represent averages of two traces from each animal. The number of animals tested is shown in parentheses above each bar. Error bars represent SEM. *P = 0.02 compared with Isc of cftr +/+ mice determined by Duncan's multiple range test.

NOS2-Specific Immunostaining in Ileum and Nasal Epithelium of cftr +/+, cftr /, and FABP-hcftr Mice

Ileum and nasal epithelium were excised from cftr +/+, cftr /, and FABP-hcftr mice, fixed, and stained for expression of NOS2. Both tissues have previously been shown to constitutively express NOS2 in normal mice. These data were confirmed in cftr +/+ mice where strong staining for NOS2 was seen in both the ileum and nasal epithelium (Figure 4). Consistent with previous findings (2, 3), neither tissue expressed NOS2 in cftr / mice. In FABP-hcftr mice, where intestinal expression of human CFTR had been introduced, NOS2 expression was seen only in the ileum and not in the nasal epithelium. These data demonstrate that epithelial NOS2 expression is strictly dependent on the presence of functional CFTR. Thus we have shown in separate tissues of the same animal differential expression of NOS2 influenced by the presence of CFTR activity.

View larger version (104K): [in this window] [in a new window]   Figure 4.   NOS2 expression in nasal and intestinal epithelium from cftr +/+, cftr /, and FABP-hcftr mice. Excised nasal epithelium (A-C ) and ileum (D-F ) were immunostained for NOS2 expression. NOS2 expression in both nasal and ileal epithelium were compared in cftr +/+ (A and D), cftr / (B and E), and FABP-hcftr (C and F ) mice. Results are representative of multiple sections taken from three separate cftr +/+ mice, three cftr / mice, and four FABP-hcftr mice. NOS2 expression was visualized with Vector Red ABC stain and sections were counter- stained with hematoxylin. Sections are magnified ×200.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The loss of NO production by airway epithelium has potential consequences that may be directly involved in the progression of CF pathogenesis. Antimicrobial defense mechanisms may be compromised (14, 20, 21), as well as the loss of an important regulatory mechanism associated with the control of transepithelial ion transport (4, 5, 22). It has previously been reported that exhaled NO levels are reduced in patients with CF, particularly from the upper airways (25- 27). Although various mechanisms may contribute to the lack of exhaled NO in CF, it has been demonstrated that expression of immunoreactive NOS2 protein as well as NOS2 mRNA are diminished in CF airway epithelial cells in both mouse and human samples (2, 3).

The goal of this paper was to determine whether NOS2 expression in epithelial cells is directly influenced by CFTR expression and function. We have been able to show in 9/HTEo- cultured human trachea epithelial cells that NOS2 mRNA expression and overall NO production are reduced when CFTR activity is inhibited by overexpressing the CFTR R domain. These cells have previously been shown to lack cAMP-mediated chloride transport (11) and to display characteristic CF phenotypes (28). Inasmuch as the 9/HTEo- cells express CFTR message but CFTR activity is blocked by excess R domain, the results indicate that NOS2 expression is regulated by the presence of active CFTR. The presence of CFTR protein without active chloride transport is insufficient to positively regulate NOS2 expression.

These studies were extended to a mouse model in which the effects of CFTR function on NOS2 expression could be directly compared in different epithelial tissues in the same animal. The cftr +/+ mice exhibit clear epithelial NOS2 expression in both the ileum and in nasal epithelium, whereas cftr / mice lack NOS2 expression in both of these tissues. In FABP-hcftr mice, which express human CFTR in intestinal epithelium but not in airway epithelium, NOS2 expression is present in the ileum but absent in the nasal epithelium. The presence of CFTR activity was confirmed in the ileum of FABP-hcftr by measuring changes in I_sc in response to FSK. The absence of functional CFTR in nasal epithelium was confirmed by measuring changes in TEPD in response to chloride-free Ringers and FSK in FABP-hcftr mice. The FABP-hcftr mice are a unique model system because they have corrected intestinal symptoms associated with CF mice, but retain CF nasal TEPD characteristics as well as other phenotypes associated with CF mice, such as white teeth (29).

Why NOS2 expression is reduced in CF becomes a primary question, especially in an inflammatory disease that would normally produce conditions favorable to the increased expression of NOS2. Although an exact mechanism needs to be clarified, this work shows that CFTR function is an important influence in the regulation of NOS2 expression and NO production. However, it is clear from other published reports that CFTR is not the only determining influence in the regulation of NOS2 expression in epithelial cells. It has been shown that constitutive intestinal NOS2 expression in the mouse is restricted to the ileum (30). CFTR function, however, has been shown throughout the intestinal tract (17). What other factors influence NOS2 expression, why different epithelial tissues differ in constitutive NOS2 expression, and how these characteristics are related to CFTR function are areas that need further exploration. Similar influence of CFTR on gene expression has recently been reported with respect to the production of the cytokine regulated on activation, normal T cells expressed and secreted (RANTES). Schwiebert and coworkers have demonstrated that the production of RANTES by both airway and pancreatic epithelial cells is dependent on the presence of functional CFTR (31). This finding is significant because both RANTES and NOS2 are similarly regulated in that the expression of both are driven primarily by nuclear factor-B and the signal tranducer and activator of transcription 1. These similarities suggest that there may be a change in cell signaling events that alter the expression of these proteins in the absence of functional CFTR. The hyperexpression of various cytokines by CF airway cells in response to infection also indicate that proinflammatory signaling pathways are intrinsically altered in CF (32, 33).

This manuscript shows that epithelial NOS2 expression is influenced at some level by the presence of active CFTR, and it demonstrates that the reintroduction of functional CFTR into a CF tissue is capable of reversing secondary cellular effects associated with diminished CFTR expression or activity. The FABP-hcftr mice represent an exciting model system in which secondary effects and changes in cell signaling resulting from lost CFTR function can be explored systematically. The examination of how a loss of CFTR function alters inflammatory signaling pathways will illuminate mechanisms responsible for the largely unexplained, multifaceted symptoms associated with CF airway disease. Identification of these pathways and subsequent intervention may lead to powerful new therapeutic options for the treatment of CF.