Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Numerous studies have demonstrated that the relaxation of vascular smooth muscle by nitric oxide (NO), an endothelium-derived relaxing factor, is mediated primarily by cGMP, a soluble guanylyl cyclase (sGC)-derived second messenger. Recent evidence suggests that the ratios of NADP⁺/NADPH and NAD⁺/NADH control NO-induced activation of sGC and the resulting vascular relaxation (Gupte et al., 1999a,b; Iesaki et al., 1999). For example, NADPH-dependent oxidoreductase is important in maintaining the heme of sGC in its NO-binding Fe²⁺ oxidation state, and reduction of NADPH production by pentose phosphate pathway (PPP) inhibitors, such as 6-aminonicotinamide (6-AN) and epiandrosterone (EPI), impairs NO-elicited relaxation of bovine pulmonary artery (PA) (Gupte et al., 1999b). Since endogenous NO attenuates acute hypoxic pulmonary vasoconstriction (HPV) (Hasunuma et al., 1991), it could be postulated that reduction of NADPH by inhibition of PPP would impair NO-sGC-related signaling mechanisms and potentiate HPV.

On the other hand, inhibition of PPP would also be expected to increase vascular smooth muscle levels of NADP⁺ and oxidized glutathione (GSSG), which might inhibit contraction by stimulating K⁺ channels and hyperpolarizing the plasma membrane (Lee et al., 1994; Weir and Archer, 1995). If this occurs in the pulmonary vasculature, the resulting membrane hyperpolarization would be expected to inhibit HPV, which is critically dependent on depolarization and activation of L-type Ca²⁺ channels. Therefore, to test whether inhibition of NADPH production would either potentiate or attenuate HPV (Fig. 1) we investigated the effects of EPI and 6-AN on pressor responses to acute hypoxia in isolated perfused rat lungs. The results of perfused lungs showed attenuation of HPV, additional studies were done with these two PPP inhibitors and another, dehydroepiandrosterone (DHEA), to identify the mechanism of vasodilation.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Two possible pathways through which inhibition of PPP may potentiate or inhibit HPV.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

The Institutional Animal Use Committee of Juntendo University School of Medicine (Tokyo, Japan) and the University of Colorado approved all protocols and surgical procedures, which were in accordance with National Institutes of Health and American Physiological Society guidelines.

Reagents. 6-AN, EPI, DHEA, angiotensin II (A-II), tetraethylammonium chloride (TEA), 4-aminopyridine (4-AP), glybenclamide (GLY), charybdotoxin (CTX), apamin (APA), nitro-L-arginine (L-NNA), and other salts were purchased from Sigma-Aldrich (St. Louis, MO). The stock solutions of steroids were made either in ethanol (Sigma-Aldrich) or dimethyl sulfoxide (DMSO; Sigma-Aldrich), and final 1:1000 dilutions in buffered physiological salt solution were used in the study. 1H-[1,2,4]Oxadiazolo[4,3-a]quinoxalin-1-one (ODQ; Cayman Chemical, Ann Arbor, MI) was dissolved in DMSO.

Animals. Experiments were performed with adult male Sprague-Dawley rats (250-350 g). The rats were housed at the ambient barometric pressure. All rats were exposed to a 12-h light/dark cycle and allowed free access to standard rat food and water.

Determination of Pulmonary Arterial Pressure of Isolated Perfused Rat Lungs. Isolated lungs were prepared as previously described (Oka et al., 1993) with minor modifications. Isolated lungs were ventilated with a humid mixture of 21% O₂/5% CO₂/74% N₂ at 60 breaths/min, with an inspiratory pressure of 9 cm of H₂O and an end-expiratory pressure of 2.5 cm of H₂O. The perfusate was a physiological salt solution (PSS) containing 116.3 mM NaCl, 5.4 mM KCl, 0.83 mM MgSO₄, 19.0 mM NaHCO₃, 1.04 mM NaH₂PO₄, 1.8 mM CaCl₂·2H₂O, and 5.5 mM of D-glucose (Earle's balanced salt solution; Sigma-Aldrich). Ficoll (4 g/100 ml, type 70; Pharmacia, Uppsala, Sweden) was included as a colloid and meclofenamate (3.1 µM) was added to inhibit cyclooxygenase and production of vasodilator prostaglandins. The lungs were perfused with a recirculated volume of 30 ml at a constant flow of 0.04 ml/g of b.wt./min. After 20 min of equilibration, the lungs were challenged four times with alternating arterial injection of 0.05 µg A-II and 5-min periods of hypoxic ventilation (0% O₂/5% CO₂/95% N₂). 6-AN, EPI, or vehicle (ethanol + DMSO) was added to the perfusate 15 min prior to the fourth injection of A-II. The maximal increase in perfusion pressure over baseline in response to a given stimulus was measured as the pressor response.

Determination of NADPH Levels in Isolated Lungs. The levels of NAD(P)H in isolated lungs were determined by HPLC after slight modification of previously published methods (Pinder et al., 1971; Jones, 1981). Briefly, isolated normoxia-ventilated lungs were treated with 500 µM 6-AN, 300 µM EPI, or vehicle for 30 min and then freeze-clamped in liquid nitrogen. The frozen tissues were homogenized in an extraction medium consisting of 2 ml of 0.02 N NaOH containing 0.5 mM cysteine at 0°C. The extracts were then heated at 60°C for 10 min and neutralized with 0.2 to 0.4 ml of 0.25 M glycylglycine buffer, pH 7.6. The neutralized extracts were centrifuged at 10,000g for 10 min, the supernatants were passed through 0.45-µm Millipore filters, and the filtered solutions were used for measurement of NAD(P)H by HPLC. NAD(P)H was eluted on a reverse-phase HPLC column (4.6 × 250 mm; Bondapak C₁₈; Shiseido, Tokyo, Japan) at 40°C by the CMA HPLC system (Tokyo, Japan) with slight modifications to a previously reported (Jones, 1981) buffer system consisting of 100 mM potassium phosphate, pH 6.0 (buffer A), and 100 mM potassium phosphate, pH 6.0, containing 5% methanol (buffer B). The column was eluted with 100% buffer A from 0 to 8.5 min, 80% buffer A plus 20% buffer B from 8.5 to 14.5 min, and 100% buffer B from 14.5 to 40 min. The flow rate was 1.0 ml/min, and the fluorescence was monitored at 260 nm (excitation) and 430 nm (emission). NADPH standards were used to calibrate the HPLC. Internal standards containing 2 nM NADPH were used to verify the quantitative recovery of the extraction procedure and HPLC retention time in the presence and absence of tissue samples.

Determination of Changes in Force of Isolated Rat PA and Aorta. PA (main branch) and aortic rings (thoracic) were prepared as previously described (Oka et al., 1993) with minor modifications. The rings were mounted on wire hooks attached to force displacement transducers (TB 611T; Nihon Kohden, Tokyo, Japan) for measurement of changes in the isometric force. Resting passive force was adjusted to a previously determined optimum (determined by maximum response to 80 mM KCl: 750 mg for PA and 1,000 mg for aorta), and vessels were equilibrated for 1 h in muscle baths containing Earle's balanced salt solution, gassed with 21% O₂/5% CO₂/74% N₂. In some cases endothelium-denuded PA rings were prepared by gently rubbing the intima with a roughened steel rod. In all cases endothelium-denuded rings had <20% relaxation by acetylcholine (10 µM) of phenylenephrine (0.5 µM) precontraction. Concentration-response curves to the PPP inhibitors (EPI, 6-AN, and DHEA) were determined with and without pretreatment by various K⁺ channel blockers in rings precontracted with 30 mM KCl (in some PA rings with 80 mM KCl). TEA (10 or 100 mM), 4-AP (2 or 10 mM), CTX (100 nM), APA (100 nM), or GLY (10 µM) was added to muscle baths 30 min before addition of PPP inhibitors. In some PA rings, dilator effect of PPP inhibitors was examined in the presence of either L-NNA (200 µM) or ODQ (50 µM) instead of K⁺ channel blockers. All subsequent values for vasodilation are expressed as percentage relaxation of the precontracted force.

Measurement of K_v Current (I_K(v)) in PASMC. Single rat PASMCs were freshly dissociated each day as described previously (Li et al., 1999). They were placed in a perfusion chamber on the stage of an Olympus CK-2 inverted microscope (Tokyo, Japan) for ~15 min to allow adherence to the bottom of the chamber and then were continuously superfused with PSS. To eliminate the effect of Ca²⁺ influx on I_K(v), during current recording we changed the superfusate to Ca²⁺-free PSS. Patch-clamp electrodes (2-4 M) were pulled from borosilicate glass capillaries (World Precision Instruments, New Haven, CT) using Flaming/Brown pipette puller model P-80 (Sutter Instruments, Novato, CA) and fire polished. The internal solution contained 125 mM KCl, 5 mM Na₂ATP, 0.5 mM NaGTP, 5 mM MgCl₂, 10 mM HEPES, and 10 mM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (Sigma-Aldrich), pH 7.2 adjusted with KOH. After whole cell configuration was achieved, resting membrane potential was first measured using current-clamp mode. For voltage-clamp experiments cell membrane potential was held at 80 mV. Whole cell current-voltage (I-V) relationship was acquired in voltage-clamp mode using a List ECP-7 amplifier (List Medical Electronics, Darmstadt, Germany). Pulses of 400 ms from 70 to +60 mV in 10-mV increments were applied. Current recordings were filtered at 3 kHz, digitized with a Digidata 1200 interface (Axon Instruments, Union City, CA), and stored in a Dell Dimension computer (Dell, Round Rock, TX) for off-line analysis. pCLAMP software version 7.0 (Axon Instruments) was used for data acquisition and analysis. All experiments were performed at room temperature (22-24°C).

Statistical Analysis. Values are means ± S.E.M. Comparisons between groups were made with Student's t test, analysis of variance (ANOVA) with Scheffé's post hoc test for multiple comparisons, or repeated measure ANOVA. Differences were considered significant at p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effects of Epiandrosterone and 6-Aminonicotinamide on the Pressor Response to Hypoxia. Representative traces of the effects of EPI (Fig. 2, top) and 6-AN (Fig. 2, bottom) on the pressor responses to A-II and hypoxia are shown. In nine isolated lungs, injection of A-II (0.05 µg) and hypoxic ventilation (0% O₂) increased perfusion pressure by 2.6 ± 0.4 and 7.4 ± 1.7 mm Hg (the third responses), respectively. Addition of EPI (300 µM) to the perfusate suppressed the pressor responses to A-II (0.9 ± 0.3 mm Hg, p < 0.05; n = 4) and hypoxia (0.9 ± 0.3 mm Hg, p < 0.05; n = 4). In another group of lungs (n = 3), addition of 6-AN (500 µM) inhibited pressor response to hypoxia (from 6.2 ± 0.5 to 2.3 ± 0.5 mm Hg, p < 0.05) but not to A-II (from 2.0 ± 0.5 to 2.5 ± 0.3 mm Hg).

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Representative traces of the effects of EPI and 6-AN on hypoxic pulmonary vasoconstriction in isolated perfused rat lungs. Third and fourth responses to A-II (0.05 µg) and hypoxia (0% O2) are shown. After the third cycle, the lungs were pretreated with EPI (300 µM, n = 4; top trace) or 6-AN (500 µM, n = 3; bottom trace) and then the lungs were challenged again with A-II and hypoxia.

Effects of 6-Aminonicotinamide or Epiandrosterone on NADPH Levels in Isolated Lungs. It is well documented that 6-AN and EPI inhibit glucose-6-phosphate dehydrogenase (Kohler et al., 1970; Gordon et al., 1995). To determine whether inhibition of PPP decreased lung NADPH levels, we treated the perfused lungs with 6-AN (500 µM) and EPI (300 µM) for 30 min after the initial 20 min of stabilization. Both drugs significantly reduced lung tissue levels of NADPH (Table 1).

Effects of 6-Aminonicotinamide, Epiandrosterone, and Dehydroepiandrosterone on Isolated PA and Aortic Rings Precontracted with 30 mM KCl. We determined concentration-response curves for the PPP inhibitors in PA and aortic rings precontracted with 30 mM KCl. Responses to 30 mM KCl were not different among the ring groups (0.61 ± 0.04, 0.68 ± 0.05, and 0.69 ± 0.05 g for PA intact, PA denuded, and aortic rings, respectively). Neither ethanol nor DMSO had effects on the precontraction induced by 30 mM KCl. As shown in Fig. 3, all three inhibitors of PPP elicited relaxation of PA and aortic rings in a dose-dependent and endothelium-independent manner. Since 6-AN caused greater relaxation in the aortic than in the PA ring, the VSMC of aorta may be more sensitive to 6-AN than that of PA. Neither L-NNA (Fig. 3) nor ODQ (n = 2-3, data not shown) inhibited the relaxation of PA rings induced by 6-AN, EPI, and DHEA.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Concentration-response curves to three structurally different PPP inhibitors, 6-aminonicotinamide (panel A), epiandrosterone (panel B), and dehydroepiandrosterone (panel C) in endothelium-intact (PA:Control, ) and -denuded (PA: E(), ), and nitro-L-arginine (200 µM) pretreated endothelium-intact pulmonary artery (PA:L-NNA, ), and aortic rings (AO, ) are shown. Number of rings is indicated in parentheses. , p < 0.05 versus PA:Control by repeated measure ANOVA.

Determination of Type of K⁺ Channels Involved in the PA Relaxation Elicited by PPP Inhibitors. To determine whether K⁺ channels are involved and which K⁺ channels are responsible for the PA relaxation elicited by PPP inhibitors, we studied the effects of high concentrations of extracellular K⁺ and various K⁺ channel blockers on the PA relaxation. As demonstrated in Fig. 4, responses to 6-AN, EPI, and DHEA were significantly reduced in PA rings precontracted with 80 mM KCl compared with those precontracted with 30 mM KCl. TEA (100 mM), a nonspecific K⁺ channel blocker, also caused rightward shifts of concentration-relaxation curves for the PPP inhibitors (Fig. 4). However, lower concentrations of TEA (10 mM) caused only slight inhibition of the PPP inhibitor-elicited PA relaxation, and the difference did not reach statistical significance except for 6-AN (n = 2-3, data not shown). Pretreatment of PA rings with specific inhibitors of calcium-sensitive K⁺ (K_Ca) channels, CTX (100 nM) (Nelson and Quayle, 1995; Seitz et al., 1999) or APA (100 nM) (Wang et al., 1999), or with a specific inhibitor of ATP-dependent K⁺ (K_ATP) channels, GLY (10 µM) (Nelson and Quayle, 1995), did not suppress the relaxation elicited by the PPP inhibitors (Fig. 5). Furthermore, we also tested the combined effect of CTX (100 nM) and APA (100 nM) on DHEA relaxation and found it did not affect the relaxation (n = 3, data not shown). In contrast, a K_v channel antagonist, 4-AP (10 mM), caused rightward shifts of concentration-relaxation curves for PPP inhibitors. Similar to the effect of TEA, however, lower concentrations of 4-AP (2 mM) did not significantly reduce the PPP inhibitor-induced relaxation except for 6-AN (n = 2-4, data not shown). Force generated by the various K⁺ channel antagonists or L-NNA in the presence of 30 mM KCl is given in Table 2.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Pretreatment of PA rings with 80 mM KCl () or with TEA (; 100 mM) significantly reduced the relaxations elicited by 6-aminonicotinamide (panel A), epiandrosterone (panel B), and dehydroepiandrosterone (panel C). , control; PA rings precontracted with 30 mM KCl. Number of rings is indicated in parentheses. , p < 0.05 versus control by repeated measure ANOVA.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Pretreatment of PA rings with 4-AP (; 10 mM) but neither CTX (100 nM, ), APA (100 nM, ), nor GLY (10 µM, ) significantly reduced the relaxation elicited by 6-aminonicotinamide (panel A), epiandrosterone (panel B), or dehydroepiandrosterone (panel C). Effects of APA and GLY were not examined for dehydroepiandrosterone. , control; PA rings precontracted with 30 mM KCl. Number of rings is indicated in parentheses. , p < 0.05 versus control by repeated measure ANOVA.

Effects of 6-AN on K_v Current (I_K(v)) in PASMC. I_K(v) was measured in the presence of 10 mM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid and 5 mM ATP to eliminate the other predominant outward currents I_K(Ca) and I_K(ATP). Membrane potential was held at 80 mV and a family of depolarizing pulses was applied from 70 mV to +60 mV in 10 mV increments. Figure 6, A and B, shows representative traces of I_K(v) before and after application of 6-AN, respectively. The I-V relationship is shown in Fig. 6C. The I_K(v) was evoked at membrane potentials higher than 30 mV, and it increased with the depolarizing pulses. Application of 6-AN (1 mM) caused a rapid increase in I_K(v) in a voltage-dependent manner, but it did not change the threshold. At +60 mV, 6-AN increased the I_K(v) by 18% (from 2290 ± 539 to 2702 ± 618 pA, n = 6; p < 0.05). This increase was abolished by pretreatment with 4-AP (5 mM), verifying that the current 6-AN increased was I_K(v) (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Representative traces of IK(v) before (panel A) and after (panel B) application of 6-AN (1 mM). Panel C shows current-voltage (I-V) relationship generated from PASMC treated with () and without () 6-AN (1 mM). The I-V curve indicates that 6-AN increased the IK(v) in a voltage-dependent manner.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The major novel findings of this study are: 1) inhibition of PPP by 6-AN or EPI-attenuated HPV in isolated perfused rat lungs, and the PPP inhibitors (6-AN, EPI, and DHEA) elicited endothelium-independent relaxation of KCl (30 mM) precontracted rat PA and aortic rings; 2) the relaxation induced by PPP inhibitors was reduced either by a nonspecific K⁺ channel inhibitor, TEA, or a K_v channel inhibitor, 4-AP, but not by either GLY, CTX, or APA; and 3) 6-AN increased I_K(v) in PA smooth muscle cell (SMC). Although it has been suggested that NADP⁺/NADPH may modulate the activity of K⁺ channels, and it has separately been shown that DHEA and EPI elicit vasodilation, we believe this is the first report that inhibition of a major source of NADPH in vascular SMC actually induces vasodilation in PA and aorta, which appears to be mediated, at least partly, by opening of K_v channels.

It has been reported that 6-AN, a competitive inhibitor of glucose-6-phosphate dehydrogenase (a rate limiting enzyme of PPP) (Kohler et al.,1970), as well as DHEA and EPI, two noncompetitive inhibitors of glucose-6-phosphate dehydrogenase (Gordon et al., 1995), inhibit PPP in red blood cells, kidneys, and PA (Rostand and Work, 1985; Grossman et al., 1995; Gupte et al., 1999b). Because in this study at least two structurally different PPP inhibitors (6-AN and the steroid hormones, EPI and DHEA) caused decreases in tissue NADPH level and vasodilation, it seems likely that a decrease of NADPH and an increase in NADP⁺/NADPH ratio were at least partly responsible for the vasodilation induced by the PPP inhibitors.

Previous studies have demonstrated that DHEA inhibits HPV and opens K⁺ channels of pulmonary VSMC in a cAMP- and cGMP-independent manner (Farrukh et al., 1998; Peng et al., 1999). Another study has shown that EPI relaxes rabbit coronary artery in an endothelium-, receptor-, cAMP-, cGMP-, prostaglandin-, and NO-independent manner (Yue et al., 1995). In our study, since meclofenamate was included in the perfusate of isolated lungs and endothelium removal, L-NNA, or ODQ had no effects on PPP inhibitor-induced PA relaxation, it is apparent that vasodilation was prostaglandin, endothelium, and NO-cGMP pathway-independent.

Since recent evidence indicates that oxidizing agents such as diamide, NAD⁺, and GSSG open K⁺ channels of isolated vascular SMCs by binding to the channel protein or by regulating the redox state of a cysteine residue that regulates the "ball and chain" mechanism occluding the internal mouth of the ion channel (Reeve et al., 1995; Weir and Archer, 1995), we speculated that the mechanism responsible for the PPP inhibitor-induced relaxation might involve the opening of K⁺ channels. This speculation was supported by our pharmacological findings that the relaxation was reduced either by higher concentrations of extracellular K⁺ (80 mM KCl) or by TEA, a nonspecific inhibitor of K⁺ channels. It is noteworthy, however, that even after nearly complete K⁺ channel blockade by a high concentration of TEA (100 mM), there was still considerable relaxation especially by EPI and DHEA. In addition, we also found that EPI and DHEA (M. Oka, unpublished observation) but not 6-AN inhibited pressor response to A-II. It has been reported that the pressor response to A-II is less susceptible to Ca²⁺ channel blockers than that to acute hypoxia in isolated perfused rat lungs (McMurtry et al., 1976), and that the A-II-induced increase in intracellular Ca²⁺ concentration is mainly due to Ca²⁺ release from a thapsigargin-sensitive intracellular store in PASMC (Guibert et al., 1996). Thus, intracellular Ca²⁺ release rather than Ca²⁺ influx could be a major player in this transient response. These results, taken together, suggest that although opening of K⁺ channels is one common mechanism of PA relaxation elicited by PPP inhibitors, a majority of the relaxation induced by EPI and DHEA may be mediated by other mechanism(s) including modulation of Ca²⁺ mobilization and direct inhibition of L-type Ca²⁺ channels (Chiamvimonvat et al., 1995).

Free radicals generated in vascular SMC and changes in the ratio of cellular reducing cofactors, including NAD⁺/NADH, NADP⁺/NADPH, and GSH/GSSG, are thought to regulate the activity of K_Ca, K_ATP, and K_v channels and thereby modulate membrane potential (Nelson and Quayle, 1995; Weir and Archer, 1995; Michelakis et al., 1997). To identify which K⁺ channel(s) is involved in the PPP inhibitor-induced relaxation, we tested the effects of specific K⁺ channel inhibitors, including 4-AP (K_v), CTX (big conductance K_Ca), APA (small conductance K_Ca), and GLY (K_ATP), on the relaxation in PA rings. We found that 4-AP caused a significant rightward shift of concentration-response curves for 6-AN, EPI, and DHEA as effectively as TEA, suggesting K_v channels are involved in the relaxant effect of PPP inhibitors. Although 4-AP is the most selective inhibitor of K_v channels in VSMC (Nelson and Quayle, 1995; Michelakis et al., 1997) and has been used to separate K_v currents from K_Ca currents, high concentrations of 4-AP may inhibit both the K_v and K_Ca channels. However, since pretreatment of PA with CTX and APA failed to reduce the relaxation elicited by PPP inhibiting agents, opening of VSMC K_Ca channels did not appear to mediate the relaxation observed in this study. Furthermore, regulation of K_ATP currents by changes in the ratio of reducing cofactor did not appear to mediate the relaxation of PA, because dilatation of PA elicited by PPP-inhibiting agents was not inhibited by GLY. These results indicate that the PPP inhibitor-induced PA relaxation is mediated at least partly by opening of K_v channels. This pharmacological finding was supported by our electrophysiological observation that 6-AN increased I_K(v) in PASMCs.

It has previously been reported that DHEA causes hyperpolarization of normal ferret and chronic hypoxia-exposed human PASMCs through opening of K_Ca channels (Farrukh et al., 1998; Peng et al., 1999). It is not clear why these findings differ from ours that K_v channels are involved in PPP inhibitor-induced rat PA relaxation. It is, however, reported that both K_v and K_Ca channels can be gated by redox changes (Weir and Archer, 1995). Indeed, several studies have shown that oxidizing agents activate both K_v and K_Ca channels of isolated PASMC (Lee et al., 1994; Park et al., 1995, 1997). Thus, it seems likely that the redox change induced by inhibition of PPP can potentially open both K_v and K_Ca channels, and that the difference between the previous reports and ours may reflect the differences in species examined and/or experimental conditions.

In addition to an increase in the oxidizing potentials caused by the inhibition of PPP, reduction of NADPH levels in vascular SMC may deplete the cytosolic GSH pool and elevate H₂O₂ generation. Generation of NADPH-oxidoreductase-derived H₂O₂ opens the K_v3.3 channel of pulmonary neuroepithelial bodies by regulating the redox state of a cysteine residue that regulates the ball and chain mechanism occluding the internal mouth of the ion channel (Wang et al., 1996). Opening of K_v3.3 channel by elevated H₂O₂ subsequent to the reduction of NADPH and GSH may be another possible explanation for decreasing tone of KCl precontracted PA. The data in the present study, therefore, suggest that 4-AP-sensitive K_v currents, possibly including K_v3.3, contributes to relaxation of PA elicited by 6-AN, EPI, and DHEA. Further studies are required to identify the exact K_v channel involved in the relaxation elicited by the PPP inhibitors.

In summary, the results of this study have demonstrated that three different PPP inhibitors, EPI, DHEA, and 6-AN, cause potent pulmonary and systemic artery dilation through modulation of NADP⁺/NADPH levels, and that activation of the 4-AP sensitive K_v currents but not K_Ca channels appears to be at least partly responsible for the dilation.