INTRODUCTION

Top Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

NITRIC OXIDE (NO) is a powerful signaling agent shown to influence several cellular mechanisms. In this study, we explored the role of NO in regulating the balance between sodium absorption and chloride secretion across airway epithelial cells. The regulation of ion transport by NO in the respiratory system gains significance in light of recent evidence that NO is constitutively expressed by airway epithelial cells (11) and that NO production is decreased in exhaled air from the upper airways of cystic fibrosis (CF) patients (2, 6, 9, 25). Additionally, we have evidence that NO production is reduced in the airways of CF mice through a loss of expression of the inducible form of NO synthase (iNOS) in airway epithelium (19, 26). These mice carrying various CF-related mutations exhibit reduced NO production in response to lipopolysaccharide (LPS), reduced bactericidal activity in the lungs, and altered cGMP production in response to NO-donating compounds. Because NO has a wide range of effects in different systems, it is possible that the lack of NO production in CF airways contributes to CF-related dysfunction in epithelial ion transport. Aberrant ion transport properties that are characteristic of CF include a lack of cAMP-stimulated chloride secretion, a lack of basal gradient-driven chloride secretion as measured by the nasal potential difference assay, and an increase in the rate of amiloride-sensitive sodium absorption (21-23). Although the loss of cAMP-stimulated chloride transport has been established to be directly related to the loss of CF transmembrane conductance regulator (CFTR) activity (1), the mechanisms underlying the loss of basal chloride transport and increased sodium absorption are not clearly delineated. We set out to explore how a loss of NO production by airway epithelial cells would influence these two areas of ion transport regulation and to determine whether decreased NO contributes to CF-related ion transport abnormalities.

Chloride transport relevant to disease, whether basal or activated, in airway epithelia likely involves the activity of other chloride channels in addition to CFTR (5). Kamosinska et al. (16) have shown that endogenous NO stimulates a CFTR-independent chloride conductance in A549 human lung epithelial cells through a cGMP-dependent mechanism via the activation of soluble guanylate cyclase activity by NO. These data suggest that a loss of NO production in CF airway epithelia may result in the loss of a secondary chloride secretory response, further compounding problems caused by the lack of CFTR-mediated chloride secretion.

In addition to the stimulation of CFTR-independent chloride secretory pathways, NO has been shown to downregulate sodium reabsorption. Although the exact mechanisms of inhibition are not well understood, both Na⁺-K⁺-ATPase from the porcine cerebral cortex (31) and the Na⁺/H⁺ exchanger in rabbit proximal tubules (29) have been shown to be inhibited by NO-dependent mechanisms. However, NO has been reported to stimulate Na⁺-K⁺-ATPase activity in human corpus cavernosum smooth muscle, indicating that NO-dependent regulatory mechanisms are not well understood and that there may be tissue-specific differences in these regulatory pathways (13). NO is known to stimulate the production of cGMP through the activation of soluble guanylate cyclases, and cGMP is a well-studied signaling agent that has been shown to reduce the rate of amiloride-sensitive sodium absorption in several systems (30). As an example of this regulatory system, atrial natriuretic peptide stimulates cGMP formation via activation of guanylate cyclase A and causes a decrease in amiloride-sensitive sodium absorption in renal inner medullary collecting duct cells (24). We have demonstrated identical regulation of sodium transport using C-type natriuretic peptide (CNP) in nasal and tracheal epithelium of mice (17, 18), although a similar report looking at nasal epithelial cells obtained from scrapings from human subjects found no effect of CNP on either chloride or sodium transport (8). In this study, we explored whether NO may be a vital component in the normal regulation of basal sodium and chloride transport across epithelial cells and the possibility that a loss of this regulatory system in CF may further complicate ion transport abnormalities associated with a loss of CFTR activity.

	MATERIALS AND METHODS

Top Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

Measurement of mouse nasal transepithelial potential difference values. Mouse nasal transepithelial potential difference (TEPD) was measured by the protocols of Grubb et al. (10) and Kelley et al. (20). Briefly, mice were anesthetized with 10 µl/g body wt of (in mg/ml) 0.4 acepromazine, 11 ketamine, and 2 xylazine in PBS. PE-10 tubing drawn out to approximately one-half of its original diameter was inserted 2-3 mm into the nostril of the mouse. Solutions were perfused at room temperature with a Razel A-99 syringe pump (Razel Scientific Instruments, Stamford, CT) at a rate of ~7 µl/min. A series of valves was used to change solutions, with a delay time of ~45 s between solution change and solution contact with the nasal epithelium. Ringer solutions consisted of chloride-replete HEPES-buffered Ringer containing (in mM) 10 HEPES, pH 7.4, 138 NaCl, 5 KCl, 2.5 Na₂HPO₄, 1.8 CaCl₂, and 1.0 MgSO₄, and chloride-free HEPES-buffered Ringer containing (in mM) 10 HEPES, pH 7.4, 138 sodium gluconate, 5 potassium gluconate, 2.5 Na₂HPO₄, 3.6 hemicalcium gluconate, and 1.0 MgSO₄. All chemicals are from Sigma (St. Louis, MO).

iNOS-specific immunostaining. Nasal epithelia were excised from mice intraperitoneally injected with either dexamethasone (45 mg/kg body wt) or PBS, paraffin blocked, and sectioned. Sections were deparaffinized, solubilized in ice-cold methanol for 5 min, and placed in 2% goat immunoglobulin in PBS for 2 h. Antibody against mouse and human iNOS 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 4 times in PBS for 10 min/wash. Goat anti-rabbit 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 (Newcastle, UK) according to the manufacturer's instructions, and slides were counterstained with hematoxylin. Quantitation of staining was performed with ImagePro imaging software.

Direct NO measurement from 9/HTEo cells. 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, 2.5 mM L-glutamine, and 150 µg/ml hygromycin on Vitrogen-coated 30-mm tissue culture plates. For studies of NO production, cells were treated with either dexamethasone (1 mM) or vehicle (PBS) for 16 h before assay, and LPS (5 µg/ml) from Pseudomonas aeruginosa (Sigma) was added to the medium of all samples 5-7 h before assay. NO assays were performed with an Iso-NO nitric oxide 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. The Iso-NO meter was calibrated according to manufacturer's instructions using the chemical method of NO production.

Mice. Pure C57BL/6J mice and mice lacking iNOS expression [NOS2(/)] were obtained from Jackson Laboratories. Mice homozygous for the F508 CFTR mutation and C57BL/6 and 129/Sv mice used to breed second-generation mice of mixed backgrounds (B6129F2) were obtained from Dr. Kirk Thomas (University of Utah School of Medicine). CF mice were fed a liquid diet as described by Eckman et al. (7). Mice were cared for in accordance with Case Western Reserve University Institutional Animal Care and Use Committee guidelines.

	RESULTS

Top Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

Effect of iNOS inhibition on basal sodium and chloride transport. To assess the effects of iNOS inhibition on sodium transport, the iNOS-selective inhibitor S-methylisothiourea (SMT; 100 µM) was perfused into the nasal passage of mice in the absence of amiloride, and the effect on TEPD was measured (3, 14, 27). The initial decline in TEPD magnitude in the absence of amiloride seen in Fig. 1 is consistent with a previous report describing murine nasal TEPD assays (10). A plateau value was reached before the addition of SMT. The addition of SMT results in an ~5-mV hyperpolarization of lumen negative TEPD (Fig. 1). The addition of amiloride (100 µM) abolishes the effect of SMT and results in an immediate depolarization of TEPD. Similarly, pretreatment with either amiloride or the cGMP analog 8-bromo-cGMP (8-BrcGMP; 100 µM) prevents the effects of SMT on nasal TEPD, showing that changes in TEPD mediated by SMT are amiloride sensitive and a result of cGMP-dependent pathways. These data are consistent with the notion that constitutively produced NO mediates a downregulatory effect on amiloride-sensitive sodium absorption through the stimulation of guanylate cyclase activity and cGMP production.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effect of S-methylisothiourea (SMT) on nasal transepithelial potential difference (TEPD) in C57BL/6J mice. Averaged traces are shown, with values at 15-s intervals. Time 0, point at which SMT was added to perfusion. Traces shown consist of addition of SMT followed by amiloride (100 µM; ), SMT in presence of amiloride (), and SMT in presence of 8-bromo-cGMP (100 µM; ). Values are means ± SE; n = 3 mice/experiment.

To assess whether iNOS-generated NO has a role in regulating basal chloride transport as well as sodium transport, we assayed C57BL/6J mice by the nasal TEPD method. C57BL/6J mice have consistently large changes in TEPD (>5-mV hyperpolarization of lumen negative TEPD) in response to perfusion into the nasal passage of chloride-free Ringer solution that has gluconate-containing salts substituted for chloride-containing salts. This perfusion in association with the nasal TEPD assay increases the electrochemical driving force for secretion of chloride into the lumen from nasal epithelial cells and can be used as an approximation of basal activation of chloride transport pathways. These same mice were reassayed after a minimum 48-h recovery period with SMT (100 µM) added to the perfusion solutions. In the presence of SMT, hyperpolarization of lumen negative TEPD in response to chloride-free Ringer solution is reduced from 9.0 ± 1.5 to 2.8 ± 0.3 mV (n = 4 mice; P = 0.009; Fig. 2A) 2.75 min after the switch to chloride-free Ringer solution, representing a 69% reduction in response. In B6129F2 mice that consistently lack a significant response to chloride-free Ringer solution (<5-mV hyperpolarization of lumen negative TEPD), the NOS substrate L-arginine was added to the perfusion solution. When mice were treated with L-arginine (100 µM), changes in nasal TEPD increased from 1.3 ± 0.5 to 6.9 ± 1.1 mV (n = 10 mice; P = 0.0004; Fig. 2B). These data demonstrate that basal NO production is a vital part of the regulatory mechanisms controlling chloride transport, although whether this NO-mediated chloride transport is CFTR dependent still needs to be examined.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Nitric oxide (NO)-mediated hyperpolarization of nasal TEPD in response to chloride-free Ringer solution. A: C57BL/6J mice assayed in presence and absence of inducible NO synthase (iNOS)-selective inhibitor SMT (n = 7/experiment). TEPD value changes (TEPD) were 5.9 ± 1.3 and 3.3 ± 1.3 mV for untreated and SMT-treated mice, respectively. B: B6129F2 mice assayed in presence and absence of NOS substrate L-arginine (L-Arg; 100 µM; n = 10/experiment). TEPD values were 5.0 ± 1.4 and 3.3 ± 0.8 mV for untreated and L-Arg-treated mice, respectively. Averaged traces are shown, with values at 15-s intervals. Values are means ± SE. Time 0, point at which perfusion was changed to chloride-free Ringer solution. Perfusion was changed to chloride-free Ringer solution when plateau value was reached in chloride-replete Ringer solution containing amiloride. Amiloride (100 µM) was present in all perfusion solutions. Mice were exposed to either SMT or L-Arg for at least 2 min before being switched to chloride-free Ringer solution.

Effect of dexamethasone on transepithelial ion transport. It has been shown in several systems that glucocorticoid treatment downregulates iNOS expression and NO production (4, 32, 33). It was our goal to determine whether dexamethasone would mimic the effects of the iNOS-selective inhibitor SMT in modulating both chloride and sodium transport. Sodium transport was indirectly monitored by measuring baseline TEPD values in mice before, during, and after intraperitoneal injection of dexamethasone (45 mg/kg body wt). Baseline values before treatment were 6.2 ± 0.8 mV (n = 5 mice) and increased to 11.5 ± 1.0 mV (n = 5 mice) with dexamethasone (Fig. 3A). Baseline TEPD was depolarized by amiloride to a plateau value of 4.9 ± 1.2 mV before treatment but only to 8.2 ± 1.3 mV (n = 9 mice; P = 0.05) after treatment with dexamethasone (Fig. 3B). Although the percentage of baseline TEPD sensitive to amiloride increased from 21 to 29% in the presence of dexamethasone, these data indicate that the majority of the dexamethasone-mediated increase in baseline TEPD is due to an amiloride-insensitive pathway.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Baseline TEPD of C57BL/6J mice before () and after () intraperitoneal injection of dexamethasone (DEX; 45 mg/kg body wt). Mice were given 3 injections over 2 days, with final assay performed 2 h after last injection. A: bars give average ± SE for 5 readings; n = 5 mice. * P = 0.006 by t-test. B: effect of DEX on amiloride-sensitive TEPD (n = 9 mice for each; P = 0.05).

To examine the effects of dexamethasone on chloride transport, the response to chloride-free Ringer solution was tested before and after intraperitoneal injection of dexamethasone. Mice before dexamethasone treatment had a 9.4 ± 0.7-mV hyperpolarization of lumen negative potential in response to chloride-free Ringer solution. After dexamethasone treatment, the response to chloride-free Ringer solution was reduced to a 2.8 ± 1.7-mV hyperpolarization (n = 9 mice; P = 0.007; Fig. 4). If NO production is indeed responsible for the stimulation of basal chloride secretion, it is likely that cGMP is the messenger that stimulates this transport. To test the role of cGMP in the stimulation of basal chloride transport, we examined the effects of the general guanylate cyclase inhibitor methylene blue on the response to chloride-free Ringer solution. Before treatment with methylene blue (100 µM), mice demonstrated a 13.6 ± 1.8-mV hyperpolarization of lumen negative TEPD in response to chloride-free Ringer solution. In the presence of methylene blue, these same mice exhibited only a 2.6 ± 0.7-mV hyperpolarization of TEPD (n = 5; Fig. 5).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Effect of DEX (45 mg/kg body wt) on chloride-free responses in C57BL/6J mice. Averaged traces are shown, with values at 15-s intervals. Mice were given 3 injections over 2 days, with final assay performed 2 h after last injection. Time 0, point at which perfusion was changed to chloride-free Ringer solution. Perfusion was changed to chloride-free Ringer solution when plateau value was reached in chloride-replete Ringer solution containing amiloride. Amiloride (100 µM) was present in all perfusion solutions. Same mice were tested in presence and absence of DEX; n = 9/experiment. TEPD values were 2.0 ± 1.2 and 4.2 ± 1.3 mV for untreated and DEX-treated mice, respectively. Values are means ± SE. Negative numbers for TEPD indicate hyperpolarization of lumen negative TEPD.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Effect of general guanylate cyclase inhibitor methylene blue (MethBlue; 100 µM) on chloride-free responses in C57BL/6J mice. Averaged traces are shown, with values at 30-s intervals. Time 0, point at which perfusion was changed to chloride-free Ringer solution. Perfusion was changed to chloride-free Ringer solution when plateau value was reached in chloride-replete Ringer solution containing amiloride. Amiloride (100 µM) was present in all perfusion solutions. Mice were exposed to MethBlue for at least 2 min before being switched to chloride-free Ringer solution. Same mice were tested in presence and absence of MethBlue; n = 5/experiment. TEPD values were 3.9 ± 0.6 and 8.9 ± 1.1 mV for untreated and MethBlue-treated mice, respectively. Values are means ± SE. Negative numbers for TEPD indicate hyperpolarization of lumen negative TEPD.

Effect of dexamethasone on iNOS expression and NO production. It has been previously reported that dexamethasone prevents the expression of iNOS (4, 32, 33). To determine whether dexamethasone treatment directly effects iNOS expression in murine airway epithelia, we performed iNOS-specific immunostaining. Nasal sections from mice treated with either dexamethasone (45 mg/kg body wt) or vehicle were obtained and stained for immunoreactive iNOS. Mice treated with vehicle showed strong staining for iNOS in airway epithelial cells, whereas dexamethasone-treated mice demonstrated limited iNOS expression (Fig. 6). These data are consistent with previously reported results and with the effects of dexamethasone on transepithelial ion transport compared with the effects of the iNOS-selective inhibitor SMT.

View larger version (104K): [in this window] [in a new window]   Fig. 6.   Immunohistochemical staining for iNOS. A and B: fluorescent iNOS-specific staining and bright-field view, respectively, of nasal section from C57BL/6J mouse treated with DEX (45 mg/kg body wt). C and D: fluorescent iNOS-specific staining and bright-field view, respectively, of nasal section from mouse treated with PBS.

The effects of dexamethasone on the production of NO were also tested with the human tracheal epithelial cell line 9/HTEo. 9/HTEo cells stimulated with LPS from P. aeruginosa produced NO levels measured to be 967.7 ± 101.6 nM (n = 4 mice), whereas cells stimulated with LPS in the presence of dexamethasone (1 mM) produced only 100.0 ± 28.9 nM (n = 4 mice, P = 0.003; Fig. 7). For comparison, 9/HTEo cells not stimulated with LPS had an NO level of 132.3 ± 11.7 nM (n = 3 mice; data not shown). These data demonstrate that treatment with dexamethasone reduces LPS-stimulated NO production to background levels.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Effect of DEX (1 mM) on NO production in 9/HTEo cells in response to treatment with lipopolysaccharide (LPS; 5 µg/ml). DEX was added 8 h before assay, and LPS was added 5-6 h before assay. NO meter was calibrated with chemical method of NO production as described by manufacturer's instructions. Values are means ± SE; n = 4 mice/condition tested. * P = 0.003 by t-test.

Ion transport characteristics of NOS2(/) mice. Through inhibition of enzymatic activity with the iNOS-selective inhibitor SMT or through inhibition of protein expression with dexamethasone, we have shown that influencing iNOS activity or production has regulatory effects on transepithelial ion transport. Therefore, mice lacking the expression of iNOS should exhibit these same changes in ion transport regulation. We compared the response to chloride-free Ringer solution using the nasal TEPD assay in C57BL/6J mice and NOS2(/) mice that have been backcrossed at least seven generations to the C57BL/6J background. Wild-type C57BL/6J mice (n = 8) possessed a large 12.1 ± 1.7-mV hyperpolarization of lumen negative potential in response to perfusion with chloride-free Ringer solution. The NOS2(/) mice, however, responded to this solution change with only a 1.8 ± 1.1-mV hyperpolarization (n = 8; P = 0.0004; Fig. 8A). These data are consistent with our data showing that inhibition of iNOS activity with either SMT or dexamethasone reduces the chloride-free response. As a comparison with other low-responder mice as shown in Fig. 3, the NOS2(/) mice were reassayed with L-arginine (100 µM) in the perfusion. Unlike the B6129F2 mice that possessed a low chloride-free response but responded to exogenously added L-arginine, the NOS2(/) mice still lacked a response to this agonist. These data suggest that the response induced by L-arginine in the B6129F2 mice is indeed due to the stimulation of NO production through iNOS. Although there is a striking decrease in the chloride-free response in the NOS2(/) mice compared with the wild-type C57BL/6J mice, both the NOS2(/) and the wild-type C57BL/6J mice have significant responses to forskolin, suggesting that there is normal CFTR activity (data not shown).

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Comparison of bioelectric characteristics between C57BL/6J mice and mice lacking iNOS expression [NOS2(/)]. A: responses to chloride-free Ringer solution from C57BL/6J mice (n = 8), NOS2(/) mice (n = 8), and NOS2(/) mice with L-Arg (100 µM; n = 8). Time 0, point at which perfusion was changed to chloride-free Ringer solution. Perfusion was changed to chloride-free Ringer solution when plateau value was reached in chloride-replete Ringer solution containing amiloride. Amiloride (100 µM) was present in all perfusion solutions. Mice were exposed to L-Arg for at least 2 min before being switched to chloride-free Ringer solution. TEPD values (in mV) were 8.6 ± 1.4 for NOS2(/) mice, 5.0 ± 0.4 for C57BL/6J mice, 4.4 ± 0.8 for NOS2(/) mice treated with L-Arg, and 11.4 ± 1.7 for L-Arg-treated cystic fibrosis mice. B: baseline TEPD values from B6129F2, C57BL/6J, and NOS2(/) mice (n = 8/strain). Values are means ± SE.

Baseline TEPD was not significantly different between the two groups of mice. NOS2(/) mice had a baseline potential value of 11.4 ± 1.1 mV (n = 8) compared with 11.1 ± 0.9 mV for C57BL/6J mice (n = 8; Fig. 8B). However, the baselines for the C57BL/6J strain appear to be already elevated relative to baselines of B6129F2 mice with mixed backgrounds (C57BL/6 and 129/Sv from the University of Utah). Mice (n = 8) with a mixed background averaged only a 7.8 ± 0.7-mV baseline. Any increase in baseline TEPD that would possibly occur due to a lack of iNOS expression may be masked by the elevated baseline of the background strain. Baselines measured in the presence of amiloride, however, did show an increased amiloride-insensitive TEPD in the NOS2(/) mice. C57BL/6J mice have a baseline of 5.0 ± 0.4 mV (n = 9) in the presence of amiloride compared with 8.6 ± 1.4 mV (n = 9; P = 0.02) without amiloride (data not shown). This magnitude of increased amiloride-insensitive TEPD is identical to that induced with dexamethasone, suggesting that a lack of iNOS expression somehow influences amiloride-insensitive TEPD by a currently unknown mechanism. The influence of strain background on basal ion transport parameters is an area that needs further examination.

	DISCUSSION

Top Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

It has been postulated that the response to chloride-free Ringer solution represents basal chloride transport and that this response is dependent on CFTR activity (23). However, the exact regulatory mechanisms that govern the basal chloride-secretory response are still unknown. In this study, we examined the role of endogenous NO production in regulating basal chloride transport in murine nasal epithelia as well as the effects NO has on the regulation of amiloride-sensitive sodium absorption.

NO has been shown to be a key regulatory agent in controlling the basal chloride current in A549 airway alveolar epithelial cells through a non-CFTR-mediated pathway. Kamosinska et al. (16) demonstrated that constitutive production of NO regulates tonic transepithelial chloride transport through the activation of soluble guanylate cyclases. Given recent results suggesting that NO production is diminished in CF airway epithelial cells, we wanted to examine whether a lack of NO production in vivo would effect basal chloride transport. We examined the effects of iNOS-selective inhibition on the chloride-free response in C57BL/6J mice. We previously reported that mice of mixed 129/Sv and C57BL/6 backgrounds had differential responses to chloride-free Ringer solution (35). Mice with a black coat color failed to respond to chloride-free Ringer solution, whereas agouti mice from the colony had a robust response. It was determined that mice with the black coat color that did not possess a chloride-free response had predominantly C57BL/6 alleles over a 4-cM region distal to agouti (35). To identify a possible C57BL/6 allele that regulated chloride-free response, pure C57BL/6J and 129/SvJ strains were studied. Surprisingly, pure C57BL/6J mice possessed a large chloride-free response, and pure 129/SvJ mice lacked this response (34). It is unclear what factors may be responsible for this difference, although heterogeneity of the strains used in the original study may account for these observed results. We have continued to use the pure C57BL/6J mice for these pharmacological studies because they have exhibited very reproducible responses to chloride-free Ringer solution. In these mice, the chloride-free response appears to be regulated by an NO-dependent pathway because pharmacological inhibition of iNOS activity significantly blocks this mode of chloride transport.

Similar results are obtained when C57BL/6J mice are treated with the glucocorticoid dexamethasone. Dexamethasone has been shown to downregulate the production of iNOS (4, 32, 33). We have substantiated this property of dexamethasone in our system by showing reduced staining of immunoreactive iNOS in lung sections from dexamethasone-treated mice. Although ion transport properties of lower airway and nasal epithelium may vary, this experiment substantiates the in vivo loss of iNOS expression in response to dexamethasone treatment. To further verify that the effects of dexamethasone on iNOS expression are consistent in airway epithelium, NO production from the human trachea epithelial cell line 9/HTEo cells treated with LPS was examined in the presence and absence of dexamethasone. Dexamethasone-treated cells had significantly reduced NO production compared with control cells. Notably, Robbins et al. (28) have shown that iNOS is expressed in the mouse lung epithelial cell line LA-4 in response to various cytokines and that its expression is blocked by dexamethasone. In this study, we compared the effects of dexamethasone on ion transport regulation in murine nasal epithelium with the effects of agents that directly influence the NO and cGMP pathways. Although experiments with dexamethasone are difficult to interpret because of the pleiotropic effects of corticosteroids, the inhibition of the chloride-free response is consistent with the known ability of dexamethasone to inhibit NO production. Additionally, glucocorticoids have been shown to be quite effective in downregulating iNOS production and thus may represent a natural regulator of iNOS expression and the subsequent downstream effects such as ion transport regulation.

Finally, we examined NOS2(/) mice that have been backcrossed onto a C57BL/6J background for a minimum of seven generations and compared their chloride-free responses with those of wild-type C57BL/6J mice. Consistent with the previous studies with SMT and dexamethasone, the NOS2(/) mice have a significantly reduced chloride-free response compared with the robust responses from C57BL/6J mice. These mice, however, have normal CFTR activity as indicated by responses to forskolin (data not shown). These data suggest that CFTR activity is an important component of the chloride-free response but is not the major contributor to basal chloride transport. As a comparison to the B6129F2 mice, chloride-free responses in NOS2(/) mice were tested in the presence of L-arginine. Unlike the B6129F2 mice, L-arginine had no effect on chloride transport in the NOS2(/) mice, demonstrating that constitutively expressed iNOS is an important factor in transepithelial chloride transport.

These studies clearly show the involvement of NO-dependent signaling pathways in regulating the chloride-free response. These methods are less conclusive, however, in understanding the role of NO in regulating basal sodium transport. Acute inhibition of iNOS activity by the addition of SMT clearly indicates an increase in amiloride-sensitive sodium transport. We have previously shown that cGMP plays an important function in regulating amiloride-sensitive sodium transport in airway epithelial cells (18), which is consistent with other reports showing a similar role for cGMP in renal and vascular systems (24, 30). The effects seen with SMT are expected given the stimulation of guanylate cyclase activity by NO. Also as expected on the basis of results with SMT, mice treated with dexamethasone demonstrated an increased hyperpolarization of baseline nasal TEPD. However, this increase does not appear to be solely the result of an increase in amiloride-sensitive sodium transport. This finding may be an artifact associated with the many systems that are influenced by dexamethasone and may not reflect strict NO regulation of sodium transport. Likewise, NOS2(/) mice have baseline nasal TEPD values that are identical to those in the C57BL/6J mice. However, both the wild-type C57BL/6J and NOS2(/) mice have elevated baseline values compared with the B6129F2 mice and compared with other strains tested (data not shown). Therefore, any elevation in baseline TEPD as a result of a lack of iNOS expression in the NOS2(/) mice may be masked by a strain variation that leads to elevated baselines in the C57BL/6J mice. Despite the ambiguity of the effects of dexamethasone and the deletion of iNOS expression in the NOS2(/) mice on baseline TEPD, our conclusion that NO plays a vital role in the downregulation of transepithelial sodium reabsorption is supported by two recent reports that show that NO inhibits amiloride-sensitive sodium absorption in rat type II alveolar epithelial cells (12, 15).

Given recent observations that iNOS levels are reduced in CF airway epithelium (19, 26), a clearer understanding of how NO influences various aspects of lung physiology is needed. With evidence that a lack of NO will further compound electrolyte transport problems associated with CF, perhaps therapeutic options that restore these disrupted pathways can be identified and used to augment existing attempts to correct CFTR function.