INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The administration of inhaled nitric oxide (NO) has been used to treat pulmonary hypertension (1) and the acute respiratory distress syndrome (ARDS) (4). At present, NO is administered in the clinical setting to intubated patients via expensive and cumbersome devices that require ancillary personnel to monitor and maintain. Recently, compounds that release NO, the NONOates (7), have been administered into the alveolar space of animals to selectively reduce pulmonary hypertension (8). NONOates may be aerosolized, have a prolonged residence in the lung, and may have half-lives of approximately 20 h at physiologic pH (8). Therefore, on first consideration, NONOate administration may be an attractive alternative method of NO administration in the clinical arena.

While the attenuation of pulmonary hypertension is beneficial, other lung functions may be adversely affected by NONOate administration. In particular, the administration of NO donors has adversely affected lung cell functions in vitro which include surfactant synthesis (12), ATP generation (12), and vectorial Na⁺ transport by alveolar type II (ATII) cells grown to confluent monolayers (13, 14). The ability to actively transport fluid from the alveolar space is a crucial lung function, and impairment of alveolar fluid clearance (AFC) has been associated with increased mortality in the setting of ARDS (15) and increased lung water in rats exposed to hyperoxia (16). The determination of effects of pulmonary NONOate administration on AFC in vivo is consequently of significant clinical interest.

The present study tested the hypothesis that administration of the NO donor DETANONOate into the alveolar space of rabbits would decrease AFC. While DETANONOate administration over a 30-fold dose range did not decrease total AFC, the amiloride-sensitive fraction of AFC was decreased. This decrease in the amiloride-sensitive fraction of AFC was accompanied by an increase in the amiloride-insensitive fraction not attributable to DETANONOate-mediated degradation of amiloride or by modulation of inhibitory function of amiloride in vitro.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The study was approved by the Animal Review Committee of the University of Alabama at Birmingham. All animals received humane care in compliance with the "Principles of Laboratory Care" formulated by the National Society for Medical Research and with the Guide for the Care and Use of Laboratory Animals prepared by the National Research Council and the National Academy Press (revised October 1996, Washington, DC).

Surgical Protocol

Male, New Zealand White rabbits (Myrtle's Rabbits, Thompson Station, TN), weighing 1.8 to 2.8 kg (n = 102) were anesthetized with 10 mg · kg¹ ketamine (Parke-Davis, Morris Plains, NJ) via a marginal ear vein. Anesthesia was maintained with 60 µg · kg¹ · h¹ fentanyl intravenously (Elkins-Sinn, Inc., Cherry Hill, NJ) and 3.0 mg · kg¹ · h¹ droperidol intravenously (American Reagent Laboratories, Shirley, NY). An anesthetic was administered intravenously as halogenated anesthetics may interfere with alveolar epithelial Na⁺ transport (17, 18). After tracheotomy, an endotracheal tube with 3.5 mm interior diameter and a modified 4-French Fogarty catheter (American Edwards Laboratory, Irvine, CA) were placed into the trachea. The Fogarty catheter was modified by removal of the balloon tip and placement of a yellow suture bootie (Oxboro-Medical International, Inc., Ham Lake, MN) that was punctured on its distal end. Placement of this catheter approximately 9 to 10 cm into the trachea resulted in an air-tight intubation of the right caudal lobe of the lung that was confirmed postmortem. Rabbits were ventilated with a Harvard Apparatus ventilator (Model 683; Millis, MA). The tidal volume and rate were adjusted to yield a peak inspiratory pressure between 19 to 23 cm H₂O measured from within the endotracheal tube and a Pa_CO2 of 32 to 45 mm Hg. Pancuronium bromide (Elkins-Sinn, Inc., Cherry Hill, NJ) 0.3 mg · kg¹ · h¹ was administered intravenously to ensure relaxed chest wall muscle tone. Arterial pressure was monitored by placement of a 22-gauge central ear artery catheter. All pressures were recorded on a Grass Model 7D polygraph (Grass Instruments, Quincy, MA). All rabbits received a maintenance infusion of lactated Ringer's solution at 20 ml · kg¹ · h¹, and esophageal temperatures were maintained at 38 to 39° C with a heating pad. A 15-min equilibration period followed completion of the surgical preparation. The pH, Pa_CO2, and Pa_O2 of arterial blood samples were determined after 15 min of equilibration and every hour thereafter throughout the experimental period. Blood gas samples were analyzed at 37° C using a blood gas analyzer (Model 1306; Instrumentation Laboratory, Lexington, MA).

Determination of AFC

After equilibration, 4 ml · kg¹ of a 5% fatty acid-free bovine serum albumin (BSA; Sigma Chemicals, St. Louis, MO) suspended in solutions with varying ionic compositions were instilled into the right caudal lobe over a 2-min period. The osmolality of the solutions instilled was between 280 and 290 mOsm/kg, with a pH of 6.6. The dead space in the modified Fogarty catheter was cleared by injection of 600 µl of 100% O₂. The fluid in the right caudal lobe was subsequently withdrawn in 1-ml increments 2 h after instillation with the last 500 µl of fluid collected for analysis. The 500-µl samples were centrifuged at 1,000 × g for 10 min to pellet cells and debris, and the protein concentration determined by a modification of a spectrophotometric method (19).

AFC, expressed as percent of total instilled volume (excluding the volume of albumin), was calculated from the following relationship, as described previously (16, 20, 21):

(1)

where the variables C_i and C_t are, respectively, the protein concentrations at time zero and at 2 h. Time zero was considered to be the end of the instillation. All rabbits were euthanized with 1 ml of a saturated KCl solution after the 2-h alveolar sample was obtained.

Generation of NO by DETANONOate In Vitro

Before in vivo experimentation, steady-state NO concentrations of 5% BSA in 0.9% NaCl solutions containing 3 µM, 300 µM, and 100 µM DETANONOate were determined using an isolated NO meter (World Precision Instruments, Inc., Sarasota, FL). The samples were maintained at 39° C and NO concentrations measured for 2 h.

Effects of DETANONOate on AFC after NaCl Instillation

Four groups of rabbits (n = 9 per group) were used to determine the effects of intratracheal 3 mM DETANONOate (Cayman Chemical, Ann Arbor, MI) instillation on AFC and its dependence on Na⁺ reabsorption. An initial 3-mM dose of DETANONOate was chosen to test our hypothesis, with dose-response relationships determined after the initial experiments. The determination of a dose-response relationship was necessary as the alveolar NO concentration may differ from that measured in vitro (e.g., NO uptake by macrophages, NO scavenging by catalase in the alveolar hypophase, etc.). In the first group (control group), NaCl containing 5% BSA (NaCl-BSA) was instilled as described previously. A second group of rabbits was instilled with NaCl-BSA containing 1 mM amiloride to block amiloride-sensitive Na⁺ channels which are responsible for the majority of epithelial Na⁺ transport (16, 18, 16, 21). A third group of rabbits was instilled with NaCl-BSA containing 3 mM DETANONOate. Lastly, the fourth group was instilled with NaCl-BSA containing 3 mM DETANONOate and 1 mM amiloride. The concentration of amiloride of alveolar samples was determined by measuring the fluorescence intensity at an excitation of 360 nm and emission of 415 nm (22). As data generated from these first four groups demonstrated that 3 mM DETANONOate increased the amiloride-insensitive fraction of AFC, an additional two groups (n = 9 per group) consisting of animals instilled with NaCl-BSA, amiloride 1 mM, and either 300 µM or 100 µM DETANONOate were generated to establish a relationship between AFC, alveolar amiloride concentration, and the dose of DETANONOate administered.

Effects of DETANONOate on Amiloride Stability and Inhibitory Function

To determine whether some of the observed in vivo DETANONOate-associated physiologic effects were due to degradation of the chemical stability or inhibitory function of amiloride, the following in vitro studies were performed.

Incubation of amiloride and DETANONOate. NaCl-BSA solutions containing 1 mM amiloride with (n = 6) or without (n = 6) 3 mM DETANONOate in closed sample tubes were placed into a 39° C water bath for 2 h, and the concentration of amiloride determined as mentioned previously (22).

Effects of DETANONOate on Xenopus oocyte amiloride-sensitive Na⁺ current. Injection of the chromosomal RNA (cRNA) of the recently cloned Na⁺ channel (rENaC) into Xenopus laevis oocytes resulted in Na⁺ currents which are inhibited by small concentrations of amiloride (10 µM) (24). Xenopus oocytes were injected with 4.3 ng of , , and -rENaC cRNAs as previously described (25). Two days later, current-voltage relationships across the oocytes were recorded as previously described (25). To test the hypothesis that preincubation to DETANONOate may decrease the ability of amiloride to inhibit Na⁺ transport, solutions consisting of either 1 mM amiloride in 0.9% NaCl or 1 mM amiloride and 300 µM DETANONOate in 0.9% NaCl solutions were diluted 100-fold and added to the bath solution of the oocytes.

Effects of Flufenamic Acid on DETANONOate-mediated Upregulation of Amiloride-insensitive AFC

As the aforementioned studies demonstrated that DETANONOate administration increased the amiloride-insensitive fraction of AFC, further experiments were designed to identify the ion channels responsible for the increase in amiloride-insensitive AFC. A nonspecific cation (NSC) channel has been effectively inhibited with 450 µM flufenamic acid (Sigma Chemicals, St. Louis, MO) (14). The greatest concentration of flufenamic acid that could be dissolved into NaCl- BSA solution was 500 µM. Consequently, rabbits anesthetized and instrumented as described previously were administered a NaCl-BSA solution containing 1 mM amiloride and 500 µM flufenamic acid in the presence (n = 8) or absence (n = 8) of 100 µM DETANONOate. An alveolar sample was obtained 2 h after instillation, and the 500-µl sample was centrifuged and processed as mentioned previously. AFC, expressed as percent of total instilled volume (excluding the volume of albumin), was calculated by changes in protein concentration and amiloride concentration was determined as described previously.

Effects of DETANONOate on AFC after Na⁺- or Cl-free Solution Instillation

In an effort to determine if DETANONOate-mediated changes of the amiloride-sensitive and amiloride-insensitive fractions were secondary to changes in transalveolar movement of Na⁺ or Cl, experiments were performed wherein either Na⁺- or Cl-free 5% BSA solutions were administered. Rabbits were anesthetized and instrumented as mentioned previously. The effects of DETANONOate on transalveolar Cl movement and AFC were determined by instilling a Cl-free 5% BSA solution in which NaCl was replaced with sodium methane sulfonate (Na⁺ MS-BSA, 6.6 pH, 280 to 290 mOsm) in the presence (n = 8) or absence (n = 8) of 100 µM DETANONOate. The effects of DETANONOate on transalveolar Na⁺ movement and AFC were determined by instilling a Na⁺-free 5% BSA solution in which NaCl was replaced with N-methyl-D-glucamine chloride (NMDG Cl-BSA 6.6 pH, 280 to 290 mOsm) in the presence (n = 8) or absence (n = 8) of 100 µM DETANONOate. An alveolar sample was obtained 2 h after instillation, and the 500-µl sample was centrifuged and processed as mentioned previously. AFC, expressed as percent of total instilled volume (excluding the volume of albumin), was calculated by changes in protein concentration as described previously. Sample Na⁺ and Cl concentrations were determined with a Synchron Lx 20 (Beckman Instruments, Inc., Schaumburg, IL).

Statistical Analyses

All variables are expressed as mean ± SD. All the following analyses were performed in accordance with common biostatistical principles (26). Analyses of the effects of DETANONOate and amiloride administration on AFC were conducted by one-way measures of analysis of variance (ANOVA). Analysis of the effects of DETANONOate on alveolar amiloride concentration in vivo was conducted by one-way ANOVA. The Student-Newman-Keuls test was used for post hoc comparisons between groups. Analysis of the effects of DETANONOate administration on AFC, Na⁺ concentration, Cl concentration, or amiloride concentration in the groups administered sodium-free, chloride-free, and amiloride-flufenamic-containing solutions was performed with Student's t test. Analyses of the effects of DETANONOate on amiloride concentration and function in vitro were performed with Student's t test. An alpha error of < 0.05 was considered significant. Analyses of the effect of DETANONOate and amiloride administration on hemodynamic and arterial blood gas variables were performed with repeated measures ANOVA, with Bonferroni correction of the alpha error to < 0.0125, accounting for multiple comparisons (four groups).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Generation of NO by DETANONOate In Vitro

Solutions composed of 5% BSA and 0.9% NaCl containing 3 mM, 300 µM, or 100 µM DETANONOate were determined to reach steady-state NO concentrations of approximately 30 µM, 3 µM, and 1 µM, respectively, after 15 min at 39° C. After reaching equilibrium, these NO concentrations persisted throughout the 2 h of observation.

Effects of DETANONOate on AFC after NaCl Instillation

Administration of 3 mM DETANONOate did not significantly decrease AFC compared with control group values (Figure 1A). Administration of 1 mM amiloride did significantly decrease AFC, but coadministration of DETANONOate resulted in AFC values significantly greater than the amiloride group, but not significantly different from the group administered DETANONOate. Consequently, the amiloride-sensitive fraction of AFC was significantly decreased by administration of DETANONOate (Figure 1B). Of interest, the alveolar amiloride concentration was significantly less in the group administered 3 mM DETANONOate with amiloride (0.48 ± 0.39 mM) compared with the group administered amiloride alone (0.91 ± 0.24 mM). Two additional groups were generated to establish a dose-response relationship between the concentration of DETANONOate administered and AFC and alveolar amiloride concentration. Rabbits administered 300 µM or 100 µM DETANONOate with amiloride had AFC values of 19.5 ± 13.5% and 23.0 ± 7.9%, respectively. There was no significant difference in AFC among the three groups administered DETANONOate with amiloride, but all three groups had significantly greater AFC values than the group administered amiloride alone. The alveolar amiloride concentrations of the groups administered 300 µM or 100 µM DETANONOate with amiloride were 0.86 ± 0.33 mM and 0.76 ± 0.21 mM, respectively. These alveolar amiloride concentration values were not significantly different from those of the group administered amiloride alone but were significantly greater than the values of the group administered 3 mM DETANONOate. Administration of DETANONOate at concentrations that did not significantly decrease alveolar amiloride concentration as compared with the amiloride group decreased the amiloride-sensitive fraction and increased the amiloride-insensitive fraction of AFC. Consequently, the lowest dosage of DETANONOate (100 µM) was used in all subsequent in vivo experimentation.

View larger version (27K): [in this window] [in a new window]   Figure 1.   Effects of DETANONOate on AFC after NaCl instillation. (Panel A) The control group (white bar, left) 2 h AFC values were not significantly different from the group administered 3 mM DETANONOate (white bar, right). Animals administered 1 mM amiloride (black bar, left) had AFC values significantly (*p < 0.05) less than the control group. Animals administered amiloride alone had significantly (*p < 0.05) lower AFC values compared with the group administered 1 mM amiloride with 3 mM DETANONOate (black bar, right). (Panel B) The alveolar-sensitive fraction of AFC (calculated from the difference in group means in the presence and absence of amiloride) was lower in animals administered DETANONOate.

Effects of DETANONOate on Amiloride Stability and Inhibitory Function

Incubation of amiloride and DETANONOate. Incubation of 1 mM amiloride with 3 mM DETANONOate at 39° C for 2 h did not significantly change the concentration of amiloride. Samples containing amiloride with DETANONOate had 98 ± 3% of the amiloride present as did samples containing amiloride alone.

Effects of DETANONOate on Xenopus oocyte amiloride-sensitive Na⁺ current. Xenopus oocytes expressing -rENaC had large inward currents that were almost completely inhibited by amiloride (Figure 2A). Preexposure of amiloride to 300 µM DETANONOate for 2 h at 39° C did not significantly decrease amiloride-mediated inhibition of inward currents (Figure 2B).

View larger version (18K): [in this window] [in a new window]   Figure 2.   Current-voltage relationships of Xenopus oocytes after amiloride administration. Current-voltage relationships of Xenopus oocytes were significantly decreased by 10 µM amiloride ( panels A and B). Preincubation of amiloride with 300 µM DETANONOate ( panel B) did not significantly change the amiloride-mediated decrease in oocyte current-voltage relationships.

Effects of Flufenamic Acid on DETANONOate-mediated Upregulation of Amiloride-insensitive AFC

Administration of DETANONOate significantly increased AFC after coadministration of amiloride and flufenamic acid (Figure 3). There was no significant difference in alveolar amiloride concentration between the groups administered amiloride and flufenamic acid in the presence or absence of 100 µM DETANONOate (0.71 ± 0.21 mM and 0.79 ± 0.09 mM, respectively). Lastly, the AFC values of the group administered both amiloride and flufenamic acid were not significantly different from the amiloride group (Figure 1A).

View larger version (12K): [in this window] [in a new window]   Figure 3.   Effects of flufenamic acid on DETANONOate-mediated upregulation of amiloride-insensitive AFC. Rabbits administered 1 mM amiloride with 500 µM flufenamic acid had significantly (*p < 0.05) greater 2 h AFC values in the presence of 100 µM DETANONOate (black bar) compared with animals not administered DETANONOate (white bars).

Effects of DETANONOate on AFC after Na⁺ or Cl-free Solution Instillation

To determine if DETANONOate increases the amiloride- insensitive fraction of AFC by modifying transepithelial transport of either Na⁺ or Cl, rabbits were administered either Na⁺-free (NMDG Cl) or Cl-free (Na⁺ MS) BSA solutions in the presence or absence of 100 µM DETANONOate (Figure 4). Administration of DETANONOate did not significantly modify AFC in the presence of either NMDG Cl or Na⁺ MS solutions (Figure 4A). The influx of Na⁺ into the alveolar space in animals administered NMDG Cl was not significantly influenced by DETANONOate (Figure 4B). Similarly, movement of Cl into the alveolar space in rabbits administered Na⁺ MS was not significantly influenced by DETANONOate (Figure 4C). Lastly, the AFC values of the groups administered either NMDG Cl or Na⁺ MS were significantly less than that of the control group (Figure 1A).

View larger version (20K): [in this window] [in a new window]   Figure 4.   Effects of DETANONOate on AFC after Na+ or Cl -free solution instillation. (Panel A) The AFC values of rabbits administered either Na+- or Cl-free solutions were not significantly changed by the presence (black bars) or absence (white bars) of 100 µM DETANONOate. (Panel B) The alveolar concentration of Na+ was not significantly changed by the presence or absence of 100 µM DETANONOate. (Panel C ) The alveolar concentration of Cl was not significantly changed by the presence or absence of 100 µM DETANONOate.

Hemodynamics and Arterial Blood Gas Analysis

The hemodynamic and arterial blood gas data obtained from the first six groups (e.g., control, 3 mM DETANONOate) are displayed in Tables 1 and 2, respectively. There were no significant differences in mean arterial blood pressure, heart rate, peak inspiratory pressure, arterial pH, or arterial Pa_CO2 between the groups. However, the control group had a significantly greater Pa_O2 than the groups administered 3 mM DETANONOate, amiloride, and 3 mM DETANONOate with amiloride at 2 h. There were no significant differences in hemodynamic and arterial blood gas data between the groups administered the three different doses of DETANONOate with amiloride. There were no significant differences in hemodynamic and arterial blood gas data between the subsequent groups (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The administration of NONOates has been advocated to reduce pulmonary artery hypertension (8). However, this benefit may have been offset by the potential risk of decreased alveolar epithelial Na⁺ transport-dependent AFC based on in vitro studies with monolayers of rat pulmonary epithelial cells (13, 14). For example, if inhaled NO were to be administered to a patient with pulmonary hypertension and alveolar flooding secondary to a pathologic state (e.g., ARDS, congestive heart failure), inhaled NO therapy could prevent the removal of alveolar fluid. This could be disastrous, given that impairment of AFC has been associated with increased mortality (15). The results of the present short-term study demonstrated that in vivo administration of 3 mM DETANONOate did not adversely affect total AFC in the rabbit, a sharp contrast from what was predicted from in vitro investigations (13, 14). In the present study, administration of DETANONOate instead appeared to concomitantly decrease the amiloride-sensitive fraction and increase the amiloride-insensitive fraction of AFC.

The mechanisms by which NO decreases epithelial Na⁺ transport appear to be both species and cell line specific. One mechanism by which NO decreased epithelial Na⁺ transport in mouse renal epithelium was by increasing intracellular cyclic guanosine monophosphate (cGMP) (27). However, Guo and coworkers determined that NO decreased both apical amiloride-sensitive Na⁺ transport and basolateral Na⁺-K⁺-ATPase activity via a cGMP-independent mechanism in rat alveolar type II monolayers (13). Oxidation and/or nitrosylation of key thiols in the apical ion channels or basolateral Na⁺-K⁺-ATPase by NO may be the underlying mechanism by which amiloride-sensitive Na⁺ transport is decreased. In support of this concept, dithiothreitol reversed the NO-mediated decrease in single-channel activity of a Ca²⁺-activated, 25 pS cation channel in rat brown fat cells (28). In the present study DETANONOate administration appeared to decrease apical amiloride-sensitive Na⁺ transport, but most likely not basolateral Na⁺-K⁺-ATPase activity, as total AFC was unaffected by DETANONOate. Of interest, NO has been demonstrated to decrease epithelial Cl transport in rat renal epithelium (29). A concurrent decrease in active Cl secretion (and iso-osmolar fluid entry) into the alveolar space epithelium could account for the unchanged total AFC and increase in the amiloride-insensitive fraction of AFC documented in the present study. However, the Cl substitution studies we performed did not demonstrate a difference in alveolar Cl transport or total AFC in the presence or absence of DETANONOate. Consequently, DETANONOate decreased the amiloride-sensitive fraction of AFC in our model.

In addition to decreasing the amiloride-sensitive fraction of AFC, another novel finding of the present study was that DETANONOate increased the amiloride-insensitive fraction of AFC through more than one mechanism. First, at the highest dose (3 mM) of DETANONOate, an increase in elimination of amiloride from the alveolar space in vivo was likely a mechanism that contributes to the increase in the amiloride-insensitive fraction of AFC. The observations in vitro demonstrating that the incubation of amiloride in the presence of 3 mM DETANONOate did not significantly degrade amiloride support the concept that the rabbit increased its elimination of amiloride through a DETANONOate-mediated mechanism. In further support of this concept, the oocyte experiments demonstrated that preexposure of amiloride to DETANONOate did not functionally impair amiloride-mediated decreases in whole cell current-voltage relationships. Consequently, the mechanism by which lower concentrations (300 and 100 µM) of DETANONOate increased the amiloride-insensitive fraction of AFC could not be determined.

In an effort to determine how DETANONOate increased the amiloride-insensitive fraction of AFC, ion substitution studies were performed. Recruitment of a typically quiescent Na⁺ channel or perhaps a decrease in Cl secretion could account for the phenomena observed in the amiloride experiments. However, the Na⁺ and Cl substitution studies did not demonstrate that DETANONOate administration significantly modified the movement of Na⁺ or Cl into or out of the alveolar space, nor did DETANONOate administration change AFC. It is of interest that the AFC values observed after either Na⁺ or Cl substitution (20 to 28%) were similar to one another, and significantly lower than the AFC value of the control group (43%). This decrease in AFC after either Na⁺ or Cl substitution is likely related to the time required for the substituted cation or anion to diffuse into the alveolar space in sufficient concentrations to allow normal iso-osmolar clearance. These data suggest strongly that both Na⁺ and Cl are necessary for AFC to occur in the rabbit. Our findings also demonstrate that quantification of the Na⁺-independent fraction AFC with ion substitution studies is likely most valid in the short term, as the substituted ion enters the alveolar space in quantities sufficient to resume iso-osmotic fluid clearance. Lastly, we were unable to demonstrate that DETANONOate activated a NSC channel, as either the channel may not exist in vivo in the rabbit, an insufficient concentration of flufenamic acid was administered, or flufenamic acid could not inhibit the channel/pathway activated by DETANONOate. In summary, DETANONOate administration (3 mM to 100 µM) may increase the amiloride-insensitive AFC by (1) increasing the clearance of amiloride from the alveolar space, and (2) activating a transcellular ion channel/pathway that functions when significant inhibition of transcellular Na⁺ transport is present (e.g., amiloride administration).

In addition to modulating AFC, DETANONOate likely increased regional blood flow to the lung wherein the BSA solutions were administered. Arterial blood gas analyses demonstrated that rabbits administered 3 mM DETANONOate had significantly less Pa_O2 than control group rabbits. NO released from DETANONOate may have decreased regional hypoxic vasoconstriction, resulting in an increase in blood flow to the nonventilated lung and consequent decrease in Pa_O2. Also of interest, the group administered amiloride had a significantly lower Pa_O2 compared with the control group. As amiloride is a Na⁺ channel blocker, it is likely that amiloride also may have decreased regional hypoxic vasoconstriction. Consequently, because there is both a DETANONOate and amiloride effect present, we were unable to determine what dose of DETANONOate is required to significantly increase regional lung blood flow. Whereas the present study was designed to test hypotheses involving AFC, we plan future studies to determine the dose of DETANONOate required to increase regional lung blood flow.

While our novel in vivo results serve as a starting point, several important questions remain to be answered before advocating the clinical administration of NONOates for the treatment of pulmonary hypertension. By design, the experiments of the present study were performed over a short period of time. Long-term exposure (e.g., days) to NONOates could result in significant changes in epithelial ion transport or nitric oxide synthase activity. Of further interest, the method of delivery and dose of drug could change the phenomena we observed. We administered DETANONOate in a solution that was slowly absorbed by pulmonary epithelium. If administered in a solution known to be rapidly cleared from the alveolar space (e.g., 0.9% NaCl), the tissue concentration of DETANONOate would likely be significantly greater than in our experiments. The formation of toxic intermediates (e.g., nitrogen dioxide, peroxynitrite) could be enhanced by excessive tissue concentrations of NONOates. Further in vivo pharmacokinetic and pharmacodynamic investigation remains to be performed before utilizing NONOates in the clinical arena.

In addition to the aforementioned findings of the present study, additional observations not central to the hypotheses tested are of interest. First, the AFC values of the control group at 2 h (approximately 40%) is greater than that observed in our previous study (approximately 27%) using the same rabbit model (21). However, the amiloride-sensitive fraction of AFC is similar between the two studies (approximately 67 to 75%). The AFC values of animals administered Cl-free solutions were also significantly greater than in the previous study (21). The primary difference between the two studies is that we previously used an isoflurane anesthetic whereas the present study involved the administration of an intravenous anesthetic. Halogenated anesthetics (e.g., halothane, isoflurane) interfere with alveolar epithelial Na⁺ transport (16, 17) in the rat. Consequently, the greater AFC values observed in the present study may be secondary to the anesthetic administered, and we plan to test this hypothesis in future investigations. A second finding of interest is the similarity between alveolar amiloride concentrations previously (21) found after 2 h (approximately 0.7 to 0.8 mM) and those observed after 2 h in the present study (approximately 0.7 to 0.9 mM).

Our findings are significantly different from those reported by O'Brodovich and coworkers (30), as they found that a 100-µM amiloride solution administered to two rabbits resulted in an alveolar amiloride concentration of 50 and 5 µM 2 h later. One explanation for the difference in the rate of change of alveolar amiloride concentration may be the difference in vehicle used to administer amiloride. O'Brodovich and coworkers (30) administered amiloride in saline, a fluid predicted to be rapidly cleared from the alveolar space of rabbits (23). Consequently, the majority of amiloride administered would be in contact with the epithelial surface, be significantly more concentrated, and have maximal exposure to clearance mechanisms. On the other hand, we administered amiloride in a 5% BSA solution known to bind amiloride (22) and to be cleared less quickly from the alveolar space than crystalloid solutions (23). Consequently, more amiloride is bound to albumin and kept in solution, not in contact with the alveolar epithelium in our model. This may explain why in order to have physiologically measurable effects, alveolar amiloride concentrations in our system must exceed 0.6 mM (21). The determination of the concentration of amiloride presented to Na⁺ channels on the alveolar epithelial surface is difficult, as it is dependent on the alveolar amiloride concentration, the dissociation kinetics of amiloride and albumin, the rate of elimination of alveolar fluid, and the rate of elimination of amiloride from the alveolar space. In summary, both anesthetic effects on AFC and alveolar amiloride kinetics will be subjects of future investigations.

In conclusion, the administration of DETANONOate did not significantly decrease total AFC in rabbits. Instead, DETANONOate administration (3 mM to 100 µM) resulted in a concomitant decrease in the amiloride-sensitive and increase in the amiloride-insensitive fraction of AFC. Further, the decrease in the amiloride-sensitive fraction of AFC is likely due to NO-mediated inhibition of amiloride-sensitive Na⁺ transport, as administration of DETANONOate did not decrease Cl transport (or basal fluid secretion) into the alveolar space. Future chronic in vivo investigations are required to determine the most favorable method of delivery and dose of NONOate administered that effectively decreases pulmonary hypertension. If these investigations demonstrate no significant toxicity (e.g., decreased total AFC), the administration of NONOates may serve as a more convenient and equally efficacious method of NO delivery to patients with pulmonary hypertension.