Introduction

Top Abstract Introduction Methods Results Discussion References

AA, a 20-carbon polyunsaturated fatty acid, is normally found esterified to cell membrane glycerophospholipids. In response to several stimuli, AA can be released from these cellular pools by phospholipases and can serve as a precursor to several biologically active compounds. The three major enzymatic pathways for AA oxidation are the cycloxygenase pathway, the lipoxygenase pathway and the cytochrome P450 monoxygenase pathway (Capdevila et al., 1992, 1995; Fitzpatrick and Murphy, 1989; McGiff, 1991). In the cycloxygenase pathway, PGH synthases generate PGH₂, which can be further metabolized into other PG, thromboxane and prostacyclin. The lipoxygenases generate HPETEs, which are converted to leukotrienes and HETEs. Cytochrome P450 monoxygenases generate four regioisomeric epoxyeicosatrienoic acids (5,6-, 8,9-, 11,12- and 14,15-EET), several mid-chain cis,trans-conjugated dienols (HETEs), and /-1 alcohols of AA (19-OH-AA and 20-OH-AA) (Capdevila et al., 1992, 1995; Fitzpatrick and Murphy, 1989; McGiff, 1991). EETs are further metabolized by epoxide hydrolases to four regioisomeric dihydroxyeicosatrienoic acids (5,6-, 8,9-, 11,12- and 14,15-DHETs) (Zeldin et al., 1993). Both EETs and DHETs have been shown to influence a variety of biological processes including control of vascular (Carrol et al., 1987; Gebremedhin et al., 1992; Katoh et al., 1991) and airway (Zeldin et al., 1995) smooth muscle tone, regulation of pituitary/hypothalamic and pancreatic peptide hormone release (Cashman et al., 1987; Falck et al., 1983; Snyder et al., 1983), inhibition of platelet aggregation (Fitzpatrick et al., 1986; Malcolm and Fitzpatrick, 1992), and modulation of fluid and electrolyte transport (Harris et al., 1990; Hu and Kim, 1993; Jacobson et al., 1984).

Much of what is known about the effects of P450-derived AA metabolites on ion transport is based on studies in the kidney. For example, 5,6-EET caused an increase in cytosolic calcium (Ca⁺⁺) concentration via verapamil-sensitive channels and inhibited mucosal-submucosal sodium (Na⁺) flux in proximal tubular cells (Madhun et al., 1991), and inhibited Na⁺ reabsorption and potassium (K⁺) secretion in the cortical collecting duct (Jacobson et al., 1984). 14,15-EET activated Na⁺/H⁺ exchange and potentiated vasopressin-induced changes in cytosolic Ca⁺⁺ in glomerular mesangial cells (Force et al., 1991; Harris et al., 1990). In addition, several studies have demonstrated that the EETs are active in modulating ion transport in extrarenal tissues. For example, 14,15-EET decreased calcium uptake in vascular smooth muscle and platelets (Kutsky et al., 1983; Malcolm and Fitzpatrick, 1992), 5,6-EET and 11,12-EET increased intracellular Ca⁺⁺ in cardiac myocytes (Moffat et al., 1993), and all of the EETs activated a Ca⁺⁺-dependent K⁺ channel in portal vein and coronary artery smooth muscle cells (Hu and Kim, 1993). Importantly, although AA and prostaglandins have been shown to affect ion transport in tracheal tissue (Eling et al., 1986; Hwang et al., 1990; Mochizuki et al., 1992; Van Scott et al., 1990), the role of P450-derived eicosanoids on airway epithelial cell electrical parameters has not been investigated.

Cytochrome P450 monooxygenases are present in lung tissue of different species including rat, rabbit, guinea pig and humans (Gonzalez et al., 1992; Knickle and Bend, 1994; Nhamburo et al., 1989; Zeldin et al., 1995, 1996). Previous immunohistochemical studies have demonstrated that nonciliated bronchiolar (Clara) cells contain the highest levels of P450 proteins (Domin et al., 1986; Serabjit-Singh et al., 1980, 1988; Strum et al., 1990). Recently, our laboratory has found that CYP2J subfamily P450s are: 1) highly expressed in human and rat lung tissue; 2) primarily present in ciliated and nonciliated epithelia and 3) able to metabolize AA to EETs (Wu et al., 1996; Zeldin et al., 1996). Furthermore, both EETs and DHETs are present in vivo in lung homogenates and in bronchoalveolar lavage fluid (Zeldin et al., 1995, 1996). Given the potent effects of EETs on electrolyte transport in extrapulmonary tissues, we hypothesized that these P450-derived AA metabolites were active in controlling ion transport in the lung. Thus, the purpose of this study was to: 1) evaluate the effects of these lung-derived EETs and DHETs on tracheal electrophysiological properties; 2) investigate for regio- and stereoselectivity of such effects and 3) begin to understand the mechanisms involved.

	Methods

Top Abstract Introduction Methods Results Discussion References

Materials. All studies used male Fisher 344 rats, 8 to 16 wk old, fed ad libitum with NIH-31 rodent food, and allowed free access to water. Animals were handled under an Institutional Animal Care and Use Committee approved protocol. All chemicals were purchased from Sigma Chemical Co. (St. Louis, MO), except sodium pentobarbital (Anpro Pharmaceutical, Arcadia, CA), amiloride (Merck, Sharp & Dome, West Point, PA), adenosine triphosphate (Boehringer Mannheim, Penzberg, Germany), AA (Nu-Chek-Prep, Inc., Elysan, MN), [1-¹⁴C]AA (Du Pont-NEN, Wilmington, DE), rat tail collagen (Collaborative Biomedical Products, Bedford, MA), and Vitrogen 100 (Celtrix Labs., Santa Clara, CA). Polyethylene tubing (PE-190) and nylon cell strainers were purchased from Becton Dickinson (Parsippany, NJ). Six well plates were obtained from Costar Co. (Cambridge, MA).

Experiments using segments of IET. Animals were killed by i.p. injection of sodium pentobarbital, and the tracheas were dissected and placed in 37°C DMEM. The distal two-thirds of the tracheas were separated, longitudinally opened on the ventral side and mounted in modified Ussing chambers (Ussing and Zerahn, 1951). The lucite chambers exposed 7.1 mm² tracheal surfaces to separate mucosal (10 ml) and submucosal (10 ml) reservoirs containing 37°C gassed (95% O₂, 5% CO₂) KBR solution (in mM: Na⁺:140, Cl:120, K⁺:5.2, Mg⁺⁺:1.2, Ca⁺⁺:1.2, HCO₃:25, HPO₄:2.2, H₂PO₄:0.4, Glucose:5.1). The transepithelial potential difference was measured with a voltage current clamp (VCC 600, Physiologic Instruments, San Diego, CA), connected to the mucosal and submucosal solutions by calomel half cells and 3M KCl-4% agar bridges, with tips closely opposed to the tissue sample. Constant current pulses (3 µA, 500 msec length, 10-sec interval) were passed through the samples by Ag-AgCl electrodes and KBR-4% agar bridges using a pulse generator. A strip chart recorder (model 3021, Yokogawa Electric Corporation, Yamanashi-Ken, Japan) was used for measurements Vt. Rt was calculated from the change in voltage induced by the constant current (42 µA/cm²) pulse. Isc was calculated by dividing the open circuit (spontaneous) voltage by the calculated tissue resistance. After the establishment of a steady-state, as indicated by maintenance of stable Vt readings for at least 10 min, five 10-µl aliquots of vehicle (100% ethanol) or one 10-µl vehicle aliquot followed by four 10-µl aliquots of increasing concentrations of one of the test compounds dissolved in vehicle were added to the mucosal reservoir at 3-min intervals [final concentration of ethanol 0.1 to 0.5% (v/v), final concentration of test compounds 10⁹ M to 10⁶ M]. Test compounds included AA, 5,6-EET methyl ester, 8,9-, 11,12- and 14,15-EETs, 5,6-, 8,9-, 11,12- and 14,15-DHETs. After the addition of each compound, standard tissue responses were evaluated using mucosal additions of amiloride (final concentration 10⁴ M), isoproterenol (final concentration 10⁵ M) and ATP (final concentration 10⁴ M) at 3-min intervals. Variations in the three parameters (Vt, Rt, Isc) were calculated using the following values as reference: for vehicle, baseline values were used; for the test compounds, post-vehicle values were used; for amiloride, the value after the last addition of test compound was used; for isoproterenol, the plateau value after addition of amiloride was used; for ATP, the plateau value after addition of isoproterenol was used. Control experiments using [1-¹⁴C]EETs and [1-¹⁴C]DHETs demonstrated that test compounds remained fully soluble in the bathing solution for the duration of the experiment.

Experiments using isolated RTE cells. Permeable CMS were created over polycarbonate plastic disks with 3-mm central orifices as previously described (Yankaskas et al., 1985). Plain disks were sterilized and cleaned by sequential immersion in 70% ethanol and 30% bleach solutions. Coating was then performed using rat tail collagen and Vitrogen 100. Coated disks were kept in Ham's F-12 media, containing 50 U/ml of penicillin and 50 µg/ml of streptomycin (F12/P + S) for at least 24 hr before cells were seeded. RTE cells were obtained by pronase dissociation as previously described (Wu et al., 1982). Briefly, animals were killed by CO₂ asphyxiation. After sternotomy, tracheas were isolated from the surrounding tissues and a small incision was performed at the cricoid membrane. Polyethylene tubing with a flanged end was used to cannulate the tracheas and was secured in place with a 2-0 silk suture. Tracheas were excised, filled with 0.1% Protease XIV/0.001% DNase solution and left in F12/P + S at 4°C overnight. RTE cells were flushed out with F12/P + S containing 5% FBS and filtered through a 70-µm nylon cell strainer to produce a single cell suspension. The suspension was centrifuged at 1500 rpm for 15 min and cells were resuspended in 1 ml of F12/P + S/10% FBS. Cells were counted and viability was assessed by trypan blue dye exclusion. Cells were accepted for further experiments only if viability was 85% or more. CMS disks were placed in six-well plates containing 1 ml of F12/P + S/10% FBS per well and the RTE cells were plated at a density of 50,000 cells per disk. After 24 hr the initial media was removed and a previously described formulation of DMEM:Ham's F12 with retinoic acid (final concentration 5 × 10⁸ M) (Kaartinen et al., 1993) was added the submucosal (2 ml) and mucosal (1 ml) sides. Cell culture media were thereafter changed every other day and cells were inspected daily under phase-contrast microscopy. Confluence was usually achieved in 5 to 7 days. Between days 8 and 14, RTE cell cultures were mounted in modified Ussing chambers and used in three separate sets of experiments. The chamber set-up was the same in all groups and was similar to the IET experiments, except that the current pulse was 1 µA (current density of 14 µA/cm²). To determine whether the AA epoxide effects on IETs were due to effects on the epithelia, 11,12-EET, the most active compound identified in the IET experiments, was added in increasing doses (10⁹ to 10⁶ M final concentration) to cultured RTE cells. The standard drugs (amiloride, isoproterenol and ATP) were then added to the mucosal bathing solution at concentrations used in the IET experiments. Time control studies recorded data for 12 min, after which the standard drugs were added. Vehicle control studies recorded data after addition of vehicle alone to the bathing solution, followed by additions of the standard drugs.

6

9

6

4

6

6

4

4

6

Synthesis and purification of EETs and DHETs. Racemic EETs were prepared as previously described (Corey et al., 1980; Falck and Manna, 1982). [1-¹⁴C]EETs were synthesized from [1-¹⁴C]AA (55-57 µCi/µmol) by nonselective epoxidation (Falck et al., 1990). The enantiomers of 11,12-EET were prepared by total asymmetric synthesis according to published procedures and by chiral phase high-performance liquid chromatography separation of racemic 11,12-EET (Hammonds et al., 1989; Mosset et al., 1986; Moustakis et al., 1986). DHETs and [1-¹⁴C]DHETs were prepared by chemical hydration of individual EETs as described (Zeldin et al., 1993). Synthetic EETs, DHETs, and AA were purified by reverse-phase high-performance liquid chromatography before use (Capdevila et al., 1990).

Statistical methods. Data from experiments with stable baseline voltage of more than 1 mV (127 of 151 preparations) were analyzed. All data were expressed as the mean ± S.E.M. Groups were compared using the t test for independent means; paired t tests were used when necessary.

	Results

Top Abstract Introduction Methods Results Discussion References

Intact Rat Tracheas

The intact rat tracheal studies were designed to evaluate the effects of P450-derived eicosanoids on airway epithelial cell electrophysiological parameters and to determine which compound(s) to investigate in the cultured rat tracheal epithelial cell experiments.

Baseline bioelectric properties and responses to standard drugs. Baseline values for Vt, Rt and Isc in intact rat tracheas (n = 19) were 3.58 ± 0.36 mV, 49.0 ± 2.3 Ohm.cm² and 71.6 ± 5.3 µA/cm², respectively (table 1). Addition of amiloride, isoproterenol and ATP resulted in significant changes in these parameters as illustrated in figure 1 and table 1. Amiloride resulted in a significant reduction in Vt and Isc, and a small but significant increase in Rt (table 1). The effects of amiloride were consistent with inhibition of mucosal Na⁺ conductance (Willumsen and Boucher, 1991). Isoproterenol induced a moderate increase in Vt and Isc with a slight reduction in Rt (table 1; fig. 1). These effects were persistent and were consistent with an increase in mucosal Cl conductance via a cAMP-dependent channel (Boucher, 1994). ATP induced a marked but transient increase in Vt and Isc with a reduction in Rt (table 1; fig. 1). These effects were consistent with a further increase in mucosal Cl conductance due to activation of purinergic receptors (Knowles et al., 1991). Addition of ethanol to the mucosal solution to a final concentration of 0.5% (v/v) did not significantly change Vt but resulted in a small increase in Rt and a parallel reduction in Isc (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Representative experimental tracing depicting rat intact tracheal response to standard drugs. Transepithelial voltage (Vt) was recorded over time, and intermittent pulses (3 µA) were applied at 10-sec intervals. Amiloride (Amil, 104 M final concentration), isoproterenol (ISO, 105 M final concentration), and adenosine triphosphate (ATP, 104 M final concentration), were added to the mucosal side at 3-min intervals.

Effects of eicosanoids. Compared to vehicle, 11,12-EET caused a significant reduction in Vt and Isc (fig. 2), without additional changes in Rt. The electrophysiologic effects of 11,12-EET were dose-dependent (Vt = 0.15 ± 0.02, 0.26 ± 0.05 and 0.38 ± 0.08 mV, at 10⁸, 10⁷ and 10⁶ M, respectively, P < .05; Isc = 8.18 ± 3.04, 11.95 ± 3.22, and 16.95 ± 5.42 µA/cm², at 10⁸, 10⁷ and 10⁶ M, respectively, P < .05). Addition of 11,12-DHET was associated with a smaller but significant reduction in Vt and Isc that reached statistical significance at concentrations of 10⁶ M (Vt = 0.39 ± 0.12 mV; Isc = 10.42 ± 2.21 µA/cm² at 10⁶ M, P < .05). Addition of AA, 5,6-EET methyl ester, 8,9-EET, 14,15-EET, 5,6-DHET, 8,9-DHET and 14,15-DHET did not elicit consistent or significant changes in Vt, Rt or Isc compared to vehicle alone. After addition of eicosanoids, tissues appropriately responded to amiloride, isoproterenol and ATP additions, indicating tissue integrity and responsiveness (fig. 2).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Representative experimental tracing depicting rat intact tracheal response to 11,12-EET and standard drugs. Transepithelial voltage (Vt) was recorded over time, and intermittent pulses (3 µA) were applied at 10-sec intervals. After an initial addition of ethanol, increasing concentrations of 11,12-EET were added to the mucosal side at 3-min intervals, followed by addition of standard drugs as described in figure 1.

Cultured Rat Trachea Epithelial Cells

The results obtained with intact rat tracheas were further investigated using a cultured rat tracheal epithelial cell preparation.

Baseline bioelectric properties and responses to standard drugs. Baseline values for Vt, Rt and Isc in cultured tracheal epithelial cells (n = 8) were 19.9 ± 8.6 mV, 148.4 ± 32.1 Ohm.cm² and 122.0 ± 26.5 µA/cm², respectively (table 1). Values of Vt and Rt were higher in cultured epithelial cell sheets compared to intact tracheal preparations, presumably because of edge damage in the intact tracheal experiments. Addition of amiloride, isoproterenol and ATP caused changes in Vt, Rt and Isc that were similar in magnitude and polarity to the intact excised tracheal experiments (table 1; fig. 3), thus indicating that the responses observed in intact tracheas were primarily due to effects on epithelial cells. No significant changes in Vt, Rt and Isc occurred over time (table 2). Addition of ethanol to the mucosal side of cell inserts, to a final concentration of 0.08% (v/v) caused a progressive decrease in Vt that reached statistical significance only for the highest concentration (table 2). As was the case for intact tracheas, ethanol caused a significant increase in Rt and a parallel but statistically insignificant reduction in Isc (table 2).

Dose response effects of 11,12-EET. Mucosal addition of 11,12-EET reproduced the effects observed in intact trachea studies (fig. 3; table 2). Thus, compared to vehicle alone, 11,12-EET caused a significant and dose-dependent decrease in Vt, a small but statistically insignificant reduction in Isc, and no additional changes in Rt (table 2). The changes in Vt were observed at concentrations as low as 10⁹ M (table 2).

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Representative experimental tracing depicting rat tracheal epithelial cell culture response to 11,12-EET and standard drugs. Transepithelial voltage (Vt) was recorded over time, and intermittent pulses (1 µA) were applied at 10 sec intervals. Increasing concentrations of 11,12-EET were added to the mucosal side, followed by addition of standard drugs, as described in figure 1.

11,12-EET enantiomer studies. To determine if the effects of 11,12-EET on tracheal epithelial cell electrophysiology were stereoselective, we exposed cultured cells to either 11(R),12(S)-EET or 11(S),12(R)-EET. Mucosal addition of 11(R),12(S)-EET (final concentration 10⁶ M) induced marked reductions in Vt similar to the effects observed after addition of racemic 11,12-EET (fig. 4). In contrast, no significant effects in Vt were noted after the addition of 11(S),12(R)-EET (final concentration 10⁶ M) (fig. 4). The effects of 11,12-EET, 11(R),12(S)-EET and 11(S),12(R)-EET on Isc were similar in magnitude and direction to their effects on Vt.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   11,12-EET enantiomer studies in rat tracheal epithelial cell cultures. Graph depicts the variation in transepithelial voltage, measured after a single mucosal infusion of either ethanol (vehicle), racemic 11,12-EET, 11(R),12(S)-EET or 11(S),12(R)-EET. Asterisk indicates P < .05 compared to time and vehicle control groups.

11,12-EET mechanistic studies. Mucosal addition of racemic 11,12-EET (final concentration 10⁶ M) caused a significant reduction compared to both time and vehicle controls (fig. 5). Pretreatment of the tissue with the Na⁺ channel blocker amiloride did not significantly alter the effects of 11,12-EET on Vt (fig. 5). Similarly, perfusion of the mucosal chamber with sodium free media also failed to modify the 11,12-EET-induced effects on Vt (fig. 5). In contrast, perfusion of the submucosal chamber with bumetanide (final concentration 10⁴ M) abolished the 11,12-EET-induced changes in Vt (fig. 5). The effects of 11,12-EET in the presence of amiloride, sodium free media and bumetinide on Isc were similar in magnitude and direction to their effects on Vt. Of note, bumetanide itself caused a reduction in Isc and Vt and an increase in Rt. Furthermore, as expected, the effects of isoproterenol and ATP were attenuated after exposure of cells to bumetanide (data not shown). Taken together, these results suggest that the effects of 11,12-EET may be mediated through inhibition of a chloride conducting pathway.

View larger version (46K): [in this window] [in a new window]   Fig. 5.   11,12-EET mechanistic studies in rat tracheal epithelial cell cultures. Graph depicts variation in transepithelial voltage measured after a single mucosal infusion of either ethanol (vehicle) or 11,12-EET. 11,12-EET was used alone or after mucosal pretreatment with amiloride (104 M final concentration), during exposure of the mucosal side to a sodium free solution, and after submucosal pretreatment with bumetanide (104 M final concentration). Asterisk indicates P < .05 compared to time and vehicle groups.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Ion transport in the respiratory epithelium is important to control the volume and composition of the epithelial surface liquid layer (Boucher, 1994; Widdicombe et al., 1991). Na⁺ ions move continuously from the airway lumen to the pulmonary interstitium through epithelial cells, although Cl ions follow, most probably through a paracellular route (Willumsen et al., 1989). Na⁺ is actively secreted from the submucosal side of the cell membrane via the Na⁺/K⁺-ATPase, thus creating a transcellular electrochemical gradient that allows Na⁺ entry via amiloride-sensitive Na⁺ channels (Willumsen and Boucher, 1991). During basal conditions in human cells, there is little transcellular movement of Cl except for a small Cl influx through the mucosal membrane, primarily due to a slight electrochemical imbalance. Less is known about the control of Cl ion transport in non-human respiratory epithelium. In the rabbit, Na⁺ channel blockade with amiloride causes mucosal membrane hyperpolarization (and therefore reduction in the Vt). Under these circumstances, mucosal Cl secretion can be demonstrated (Van Scott et al., 1989). Several mechanisms have been postulated to account for the mucosal Cl conductance but at least two channels have been identified: 1) CFTR, the cystic fibrosis transmembrane regulator, that is activated by increases in intracellular cAMP (e.g., by exposure to sympathetic ₂-agonist agents) and 2) calcium sensitive channels that are activated by intracellular surges of Ca⁺⁺ (e.g., after exposure to agents such as histamine and bradykinin). Recently, mucosal exposure to triphosphate nucleotides such as ATP and UTP has also been demonstrated to stimulate Cl secretion through membrane triphosphate nucleotide (P2Y2) receptors. It is not clear if these receptors activate Ca⁺⁺-sensitive channels or act through a distinct Cl channel (Knowles et al., 1991).

The P450-derived eicosanoids are known to modulate ion transport in the kidney, cornea, heart and vascular smooth muscle (Force et al., 1991; Harris et al., 1990; Hu and Kim, 1993; Jacobson et al., 1984; Kutsky et al., 1983; Madhun et al., 1991; Malcolm and Fitzpatrick, 1992; Masferrer et al., 1990; Moffat et al., 1993; Roman and Harder, 1993). Among the effects observed were: 1) increase in Ca⁺⁺ permeability mediated through verapamil-sensitive and voltage-sensitive channels (Madhum et al., 1991; Snyder et al., 1983); 2) stimulation or inhibition of Na⁺/K⁺-ATPase (Escalante et al., 1988; Masferrer et al., 1990); 3) activation of Na⁺/H⁺ exchange (Harris et al., 1990) and 4) activation or inhibition of Ca⁺⁺-dependent K⁺ channel (Hu and Kim, 1993; Masferrer et al., 1990). These data suggest that the P450-derived AA metabolites might also affect epithelial electrolyte transport in the airway. We therefore examined the effects of synthetic, HPLC purified EETs, DHETs and AA on transepithelial electrical parameters. In intact tracheas, of the compounds tested, only 11,12-EET and its corresponding diol, 11,12-DHET, caused significant, dose-dependent reductions in Vt and Isc. Changes in electrical parameters induced by the other eicosanoids were small, inconsistent and statistically insignificant; however, we cannot rule out the possibility that these other eicosanoids may have small but significant effects at higher concentrations, particularly if tested in larger numbers of animals. The effects of 11,12-EET were rapid and occurred at physiologically relevant concentrations (10⁸-10⁶ M). Furthermore, these effects of 11,12-EET were reproduced in cultured tracheal epithelial cells indicating that the responses observed in intact tracheas were primarily epithelial mediated. The 11,12-EET inhibition of Vt is smaller in magnitude than the stimulatory effects of isoproterenol and ATP (figs. 2 and 3) suggesting that P450-derived AA metabolites may be less important regulators of tracheal transepithelial ion transport. To our knowledge, this is the first demonstration of a physiological effect of AA epoxygenase metabolites on the airway epithelium.

Previous work has shown constitutive expression of a number of P450 AA monooxygenases in the lung including members of the CYP1A, CYP2B, CYP2E and CYP2J subfamilies (Domin et al., 1986; Knickle and Bend, 1994; Zeldin et al., 1995, 1996). Immunohistochemistry and in situ hybridization have demonstrated that these hemoproteins are highly expressed in Clara cells and present in much lower levels in ciliated epithelial cells and goblet cells (Domin et al., 1986; Serabjit-Singh et al., 1980, 1988; Strum et al., 1990). Recently, our laboratory has shown that members of the CYP2J subfamily are highly expressed in both ciliated and nonciliated epithelial cells throughout the airway from trachea to bronchioles (Zeldin et al., 1996). We describe electrophysiological effects of P450-derived AA metabolites in rat trachea (which has both ciliated and nonciliated epithelial cells) and in cultured rat tracheal epithelial cells. Future studies should attempt to localize the effects of 11,12-EET to specific cell types within the airway.

Isc has been demonstrated by radioisotope ion flow studies to correlate well with net electrogenic transepithelial ion transport (Boucher, 1994). Therefore, one would predict based on our studies, that 11,12-EET caused a reduction in net electrogenic transepithelial ion transport. However, Isc measurements are not able to identify specific ion currents. Therefore, studies were devised to elucidate the mechanisms of action of 11,12-EET-induced changes in Isc. We presumed that a reduction in Isc may be caused by one or more of the following: 1) reduction in Na⁺ inflow; 2) decrease in Cl secretion and/or 3) general metabolic inhibition. To evaluate effects on Na⁺ flux, we first blocked mucosal Na⁺ channels with amiloride and demonstrated that this manipulation did not alter the electrical effects induced by 11,12-EET. Next we removed Na⁺ from the mucosal bathing solution and showed that eliminating mucosal inflow of Na⁺ had no effect on 11,12-EET-induced electrical changes. Based on these results, we concluded that 11,12-EET did not affect mucosal amiloride-sensitive Na⁺ channels and was not likely involved in Na⁺ transcellular ion flow. To study potential effects of 11,12-EET on Cl secretion, cells were exposed to bumetanide, a submucosal membrane Na⁺/K⁺/Cl cotransport inhibitor, to reduce intracellular Cl availability for mucosal secretion. Bumetanide alone caused a significant reduction in Isc and loss in Vt, suggesting that, at difference to humans (Boucher, 1994) there was a basal secretion of Cl in rat tracheal epithelial cells that is impaired by bumetanide. Importantly, pretreatment of cells with bumetanide significantly reduced the effects of 11,12-EET on Vt and Isc, thus suggesting that the mechanism of action of 11,12-EET may involve inhibition of a Cl conductive pathway. Previous work has demonstrated that, at higher doses than those used in our study, AA caused inhibition of mucosal Cl secretion in the dog and human airway (Hwang et al., 1990; Mochizuki et al., 1992). Our findings suggest that the active compound may actually be a P450-derived metabolite of AA. In this regard, Kersting et al. (1993a, b) have demonstrated that the P450 epoxygenase inhibitor, ketoconazole, increases mucosal Cl transport in gallbladder epithelial cells and activates Cl conductance in cultured cystic fibrosis cells. Chloride conductance in airway epithelium depends on the stimulus and multiple Cl channels probably exist in airway epithelial cells (Knowles et al., 1991). The inhibitory effects of 11,12-EET on airway epithelial electrophysiological parameters may be small in magnitude, in part because only one of several Cl channels are affected. Future studies should focus on characterizing the effects of 11,12-EET on specific Cl channels.

The most striking finding in our study was that the effects of 11,12-EET on airway epithelial electrophysiology were highly enantioselective. Thus, although 11(R),12(S)-EET was active in reducing Vt and Isc, the opposite enantiomer 11(S),12(R)-EET was inactive. In light of these findings, the possibility that 11,12-EET causes a generalized nonspecific metabolic inhibition seems less likely. The high degree of enantioselectivity of the 11,12-EET effects, together with the fact that these actions occur in the nanomolar concentration range, suggest the presence of specific receptors for the EETs (Wong et al., 1993). It is interesting that biologically active 11(R),12(S)-EET is also the least abundant rat lung enantiomer (Zeldin et al., 1996) suggesting that availability of the active compound may be a limiting factor. Other biological actions of the EETs, including cycloxygenase inhibition and renal vascular effects, have been shown to be stereoselective (Fitzpatrick et al., 1986; Katoh et al., 1991).

There are technical limitations to using Ussing chamber studies of small tissues. When tissues are clamped in these chambers, edge damage may occur and can lead to a pathway for ion and current leakage. As the diameter of the orifice decreases, this becomes more of a problem because the circumferential region of edge damage increases disproportionately to the exposed surface area. The opening of the chamber in our study is small (~7 mm²) by necessity because rat tracheas have a relatively small circumference. Thus, it is possible that edge damage may have occurred in our intact tracheal experiments. However, accurate data can and have been obtained with small tissues using careful techniques. For example, Grubb et al. (1994) reported a transepithelial resistance of 32 to 33 ohm.cm² in intact mouse tracheas using techniques analogous to ours. Importantly, our results with the intact tracheas were confirmed in cultured cell preparations, where the experimental preparation eliminates the possibility of edge damage.

In summary, we provide evidence that some P450-derived AA metabolites cause significant and dose-dependent effects on airway epithelial cell electrophysiology. The effects appear to be limited to one of the EET regioisomers (11,12-EET) and are highly enantioselective for 11(R),12(S)-EET, the least abundant rat lung enantiomer. Furthermore, the EET-induced effects may be mediated through inhibition of a Cl conductive pathway. We conclude, based on these data, that the AA epoxygenase metabolites are involved in the control of lung fluid/electrolyte transport. We speculate that altered local concentration of these bioactive eicosanoids may contribute to lung dysfunction.