At birth, associated with the rise in oxygen tension, the pulmonary arteries (PA) dilate and the ductus arteriosus (DA) constricts. Both PA and DA constrict with vasoconstrictors and dilate with vasodilators. They respond in a contrary manner only to changes in oxygen tension. We hypothesized that the effects of changes in oxygen are mediated by changes in redox status. Consequently, we tested whether a reducing agent, DTT, and an oxidizing agent, dithionitrobenzoic acid (DTNB), would have opposite effects on a major oxygen signaling pathway in the PA and DA smooth muscle cells (SMCs), the sequence of change in potassium current (IK), membrane potential (Em), cytosolic calcium, and vessel tone. Under normoxic conditions, DTT constricted adult and fetal resistance PA rings, whereas in DA rings DTT acted as a potent vasodilator. In normoxia, voltage-clamp measurements showed inhibition of IK by DTT in PASMCs and, in contrast, activation in DASMCs. Consequently, DTT depolarized fetal and adult PASMCs and hyperpolarized DASMCs. [Ca2+]i was increased by DTT in fetal and adult PASMCs and decreased in DASMCs. Under hypoxic conditions, DTNB constricted DA rings and caused vasodilatation in fetal PA rings. DTNB inhibited IK and depolarized the cell membrane in DASMCs. In contrast, activation of IK and hyperpolarization was seen in PASMCs. Thus the same redox signal can elicit opposite effects on IK, Em, cytosolic calcium, and vascular tone in resistance PA and the DA. These observations support the concept that redox changes could signal the opposite effects of oxygen in the PA and DA.
- potassium channels
- resting membrane potential
- pulmonary vasoconstriction
change in oxygen (o2) tension is a critical regulatory factor in the pulmonary vasculature and ductus arteriosus (DA) during the perinatal period. In the developing fetus, O2 tension is low and the pulmonary arteries (PA) are constricted, while the DA is dilated and acts as a right-to-left shunt pathway to divert blood directly into the systemic circulation (13). Birth results in a dramatic change in O2 tension. The PA dilate rapidly as a result of the increase in alveolar O2 at the onset of lung ventilation, while in contrast, the DA constricts. In the adult pulmonary vasculature, the resistance arteries, similar to those of the newborn, are dilated under normoxia and constrict in response to a fall in O2 tension (10). The cellular mechanisms of the diametrically opposite responses to changes in O2 tension in these tissues (DA and PA) are not completely characterized. The smooth muscle cells (SMCs) of both vascular tissues have O2-sensitive, 4-aminopyridine (4-AP)-sensitive, potassium (K+) channels, which appear to control resting membrane potential (Em) and vascular tone (2, 19, 25, 32, 38). In the DASMCs, normoxia inhibits one or more of these voltage-sensitive K+ (Kv) channels to initiate vasoconstriction, whereas in the PASMCs normoxia activates the same or other Kv channels (14, 27, 28, 32, 40). A rise in intracellular Ca2+ is usually a prerequisite for vasoconstriction. This rise might occur through several pathways, an influx of extracellular Ca2+ through voltage-dependent Ca2+ channels or nonselective cation channels, secondary influx through store-operated Ca2+ channels, or direct release from intracellular stores.
It has been suggested that changes in the redox status of the cytoplasm of the PASMC to a more reduced state might induce vasoconstriction in response to hypoxia (5). There are several reports that the activity or gating of K+ channels can be redox modulated (1, 17, 27, 29, 31). Our hypothesis is that the change in O2 can alter the redox status in the cytosol of the SMCs and that this change may then influence the activity of some K+ channels. This study examined the effect of a known reducing agent, dithiothreitol (DTT), and an oxidizing agent, dithionitrobenzoicacid (DTNB), on K+ channel activity, Em, change in intracellular calcium and vascular ring tension in the DA and PA to determine whether the same cytosolic redox change had the same or opposite effects in the two vessels.
MATERIALS AND METHODS
All animal studies were approved by the Institutional Animal Care and Use Committee of the Minneapolis VA Medical Center and conform to current National Institutes of Health and American Physiological Society guidelines for the use and care of laboratory animals.
Cytochrome c and DTNB reduction assay. An in vitro assay system was used to test the ability of DTT to reduce the electron acceptor ferric cytochrome c and the disulfide DTNB. Reaction tubes contained 50 mmol/l of Tris buffer (pH 7.4) and 23 μmol/l of cytochrome c, with varying concentrations of the redox agent (0.5–5 mmol/l) DTT for cytochrome c reduction assay and (5.4–54 μmol/l) DTT for DTNB (8 μmol/l) reduction assay, respectively. For the cytochrome c reduction assay, baseline absorbance was read at 550 nm, and then the DTT was added. The reaction was followed at 550 nm for 3 min with absorbance readings being made every 30 s. For the DTNB reduction assay, baseline absorbance was read at 412 nm, and then the DTT was added. The reaction was followed at 412 nm for 1.5 min with absorbance readings being made every 30 s.
All reactions were carried out at 25°C.
Measurement of changes in tone in isolated adult and fetal PA and DA rings. Rings were isolated and mounted in 2-ml baths. Adult rat resistance PA, adult rabbit resistance PA, and fetal rabbit resistance PA and DA rings were equilibrated in Earle's bicarbonate buffered medium under normoxia (21% O2/5% CO2/balance N2; pH, 7.4; Po2, 120–150 Torr) or under hypoxia (0% O2/5% CO2/balance N2; pH, 7.4; Po2, 20–30 Torr) at 37°C for 60 min at tensions of 900, 500, 600, and 400 mg, respectively. The endothelium was left intact. Arteries were preconstricted with 1 μmol/l of phenylephrine (PE) in rat PA and fetal rabbit PA and DA to provide a modest degree of pretone. In adult rabbit PA rings, the response to PE declined rapidly; consequently, we used 1 μmol/l of U-46619 (a stable thromboxane A2 analog). For the purposes of data analysis, the level of preconstriction was taken to be zero. DTT and DTNB from a stock solution were added directly to the bath. O2 tension was recorded continuously with an M1-730 oxygen electrode (Microelectrode, Bedford, NH).
Cell isolation. Adult rat, adult rabbit, and fetal rabbit PA, as well as fetal rabbit DA, SMCs were freshly dissociated for patch-clamp and Ca2+ imaging studies every day. PA and DA cells were isolated by a method adapted from that described previously (2, 19, 23, 32).
Several digestions were done each day to ensure cell viability. Gentle trituration produced a suspension of single cells, which was then aliquoted into a perfusion chamber on the stage of an inverted microscope (Diaphot 200, Nikon). After a brief period to allow partial adherence to the bottom of the recording chamber, cells were perfused via gravity with an extracellular solution (see Solutions and drugs) at a rate of 2 ml/min.
Electrophysiology. Whole cell recordings were performed with the amphotericin-perforated patch-clamp technique (26). Voltage- and current-clamp measurements were performed as previously described (19). Patch pipettes were pulled from glass tubes (PG 150T, Warner Instruments). The pipettes were fire polished directly before the experiments and had a resistance of 2–3MΩ when filled with a pipette solution. The patch-clamp amplifiers were Axopatch 200A and B (Axon Instruments, Foster City, CA) in all voltage- and current-clamp experiments. Offset potentials were nulled directly before formation of a seal. Whole cell capacitance and series resistance were corrected (usually 80%). No leak subtraction was made. Leakage current was monitored using hyperpolarizing steps (-30 mV) from the holding potential, a procedure that did not activate ion channels but allowed measurement of passive membrane properties and leak during the experiments. Cells expressing holding current at -70 mV of >20 pA before or during the recordings were discarded. The effective corner frequency of the low-pass filter was 1 kHz. The frequency of digitization was at least twice that of the filter. For resting Em experiments, cells were held in current clamp at their resting Em (without current injection). Series resistance and leak were checked at the beginning and end of each Em experiment to eliminate artifactual changes in potential.
The data were stored and analyzed with commercially available pCLAMP 8.0 software (Axon Instruments).
All experiments were performed at 30°C and in low light intensity because of the light sensitivity of amphotericin B.
Pulse protocols and analysis. SMCs were voltage clamped at a holding potential of -70 mV. The standard protocol used to obtain current-voltage relationships consisted of 300-ms voltage-clamp pulses applied in 20-mV steps between -70 and +50 mV. We obtained steady-state current-voltage relationships by measuring the current at the end of the voltage-clamp pulse in control and after application of extracellular solution containing DTT or DTNB and plotting this against the test potential. Currents were normalized relative to each cell's control current at +50 mV.
Measurement of intracellular Ca2+. [Ca2+] was measured by dual-excitation ratiometric imaging, using fura 2 (12). Freshly dispersed cells were transferred to the experimental chamber (Molecular Probes, Eugene, OR) and incubated in Ca2+-free extracellular solution with 0.1 μmol/l of fura 2-AM and 0.8 μmol/l of Pluronic acid for 15 min at room temperature (19). The plates were then washed with extracellular solution containing 2.0 mmol/l of Ca2+ and incubated at room temperature for a further 15 min. Plates were rewashed and placed on the stage of an inverted microscope and perfused with a warmed experimental solution (30°C). This loading method allowed low concentrations of fura 2 to be quickly introduced into the cells without the potential effects on cell morphology that may occur from long exposures to high concentrations. Background fluorescence was recorded from each dish of cells and subtracted before calculation of the 340- to 380-nm ratio. Changes in intracellular calcium concentration ([Ca2+]i) were measured with a cooled charge-coupled device camera (Hamamatsu) with MetaFluor image capture and analysis software (Universal Imaging, West Chester, PA). Individual exposure times were adjusted so that similar gray values were used for both wavelengths. Measurements were made every 5 s, and [Ca2+]i was calculated according to the method of Grynkiewicz et al. (12). Dissociation constants of 220 and 245 nmol/l were calculated from in vitro calibration for PASMCs and DASMCs, respectively. We determined maximal and minimal ratio values at the end of each experiment by first treating the cells with 1 μmol/l of ionomycin (maximal ratio) and then chelating all free Ca2+ with 10 mmol/l of EGTA (minimal ratio). Any cells not responding to ionomycin were discarded, as were cells showing significant photobleaching. Peak increases in [Ca2+]i were measured during each intervention, and data are given as averaged peak values.
Solutions and drugs. The extracellular or experimental solution contained (in mmol/l) 115 NaCl, 5.4 KCl, 1 MgCl2, 2.0 CaCl2, 25 NaHCO3, 10 HEPES, and 10 glucose (pH 7.4 with NaOH). The standard intracellular pipette solution contained (in mmol/l) 145 KCl, 1 MgCl2, 1 ATP, 0.1 EGTA, 10 HEPES, and 240 μg/ml amphotericin B (pH was adjusted to 7.2 by KOH).
Experimental solutions were equilibrated with 21% O2, 5% CO2, and 74% N2 or 0% O2, 5% CO2, and balance N2. These procedures produced Po2 values in the cell chamber of 140–160 Torr under normoxic and 20–30 Torr under hypoxic conditions. Pco2 was 36–42 Torr, and pH was 7.37–7.42 under these conditions.
Fura 2-AM was obtained from Molecular Probes. All other compounds were purchased from Sigma (St. Louis, MO). All drugs were dissolved in Earl's or in experimental solution, with the exception of nifedipine and DTNB, which were dissolved in ethanol and DMSO, respectively. Stock solutions of 1 mol/l, 3 mol/l, and 100 mmol/l were made for DTT, DTNB, and nifedipine, respectively. The pH of solutions containing drugs was tested and corrected to eliminate potential pH-induced effects. Stock solutions of nifedipine in ethanol were diluted at least 1:10,000. At this concentration the vehicle alone had no effect on the baseline levels of Ca2+. The stock solution of DTNB in DMSO was diluted at least 1:1,000. At this concentration the vehicle alone had no effect on K+ current or resting Em. The vehicle effect of DMSO on isolated PA and DA rings is shown in Fig. 8.
Statistical analysis. Numerical values are given as means ± SE of n cells. In all figures the SE is indicated when it exceeds the symbol size. Student's unpaired and paired t-tests were used to compare Em recordings and changes in DA and PA vascular tone from preconstriction to the level attained after DTT and DTNB, respectively. Intergroup differences in K+ current, intracellular Ca2+ levels, and vascular tone were assessed by a factorial analysis of variance with post hoc analysis with Fisher's least significant difference test. P values < 0.05 were considered significant.
Assay. DTT reduced ferric cytochrome c at physiological pH in a dose-dependent manner (Fig. 1A). DTT reduced the disulfide DTNB (Fig. 1B) in proportion to the dose applied. These results show that DTT and DTNB act as reducing and oxidizing agents, respectively.
Electrophysiology. The effect of the externally applied membrane-permeable reducing agent DTT and oxidizing agent DTNB on whole cell outward K+ current was tested by the amphotericin-perforated patch-clamp technique in DA and fetal as well as adult PA SMCs.
K+ currents recorded from rabbit DASMCs were signifi-cantly increased after application of DTT. Representative traces recorded from a holding potential of -70 mV in incremental depolarizing steps to +50 mV under normoxic conditions are shown in Fig. 2A (left). When the DASMCs were superfused with an external solution containing 3 mmol/l of DTT, a time-dependent increase in K+ current was observed, plateauing at 8 min (277 ± 71% increase from control at +50 mV, n = 4; Fig. 2B). The oxidizing agent DTNB significantly decreased whole cell K+ current in fetal rabbit DASMCs under hypoxic conditions (Fig. 2C). The averaged current-voltage relationships for K+ current after an 8-min application of 1 mmol/l of DTNB are shown in Fig. 2D (28 ± 6% decrease from control at +50 mV, n = 4).
Figure 3A shows the effect of externally applied 3 mmol/l of DTT on K+ current, recorded from fetal rabbit PASMCs of resistance vessels under normoxic conditions. The change in current induced by DTT application usually reached steady state within 8 min. The block of K+ current caused by 3 mmol/l of DTT was down to 67 ± 4% of control (at + 50 mV, n = 5; Fig. 3B). The oxidizing agent DTNB significantly increased whole cell K+ current of fetal PASMCs under hypoxic conditions (Fig. 3C). The averaged current-voltage relationships for K+ current in control and after 8-min application of 1 mmol/l of DTNB are shown in Fig. 3D (41 ± 11% increase from control at +50 mV, n = 4). K+ current of adult rabbit PASMCs from resistance vessels was also inhibited by 3 mmol/l of DTT (down to 81 ± 3% of control at +50 mV, n = 4; not shown). DTT (under normoxic conditions) and DTNB (in hypoxia) significantly modulated K+ current in DASMCs and fetal PASMCs also at more negative Em (at -10 mV) as shown in Fig. 4A. The time course of the effect of DTT and DTNB on K+ current and resting Em of SMCs was also examined. Figure 4B demonstrates the time-dependent effect of DTT and DTNB on K+ current and resting Em of PASMCs under the same conditions as indicated above.
In rabbit DASMCs, fetal rabbit PASMCs, and adult rabbit PASMCs, the average resting Em value, measured by current-clamp technique under normoxic conditions, was -12.9 ± 1.6, -49.3 ± 1.8, and -51.3 ± 1.1 mV, respectively. Superfusion of DASMCs with experimental solution containing 3 mmol/l of DTT showed a time-dependent membrane hyperpolarization, which would be consistent with the opening of K+ channels (-14.6 ± 2.1 mV, n = 5; Fig. 5A). The hyperpolarization began after ∼5–8 min (Fig. 4B). This is a similar time course to the increase in K+ current that was recorded in voltage clamp (Fig. 4B). Bath application of 3 mmol/l DTT depolarized fetal as well as adult rabbit PASMCs resting Em after ∼8 min, 11.9 ± 4.5 mV (n = 5) and 6.7 ± 2.6 mV (n = 4), respectively (Fig. 5A). Under hypoxic conditions, the average resting Em value in rabbit DASMCs and fetal rabbit PASMCs was -32.0 ± 1.2 and -32.8 ± 1.1 mV, respectively. DTNB significantly depolarized DASMCs resting Em, 13.3 ± 1.5 mV (n = 4, Fig. 5B). Superfusion of PASMCs with an experimental solution containing 1 mmol/l of DTNB showed a time-dependent membrane hyperpolarization of -9.5 ± 1.5 mV (n = 4, Fig. 5B). Changes in Em recovered either partially or completely after returning to control solution.
Measurement of [Ca2+]i and role of Ca2+ channels. DTT caused a decrease in [Ca2+]i in rabbit DASMCs. Average resting [Ca2+] under normoxia was calculated to be 171 ± 6 nmol/l (n = 33). DTT at 3 mmol/l decreased [Ca2+] by 54 ± 12 nmol/l (n = 18, Fig. 6). In fetal rabbit PASMCs, adult rabbit PASMCs, and adult rat PASMCs, the average resting [Ca2+] under normoxia was calculated to be 98 ± 2 nmol/l (n = 39), 84 ± 4 nmol/l (n = 24), and 91 ± 2 nmol/l (n = 16), respectively. DTT significantly increased the [Ca2+]i of PASMCs (Fig. 6). The calculated values of the changes were 335 ± 62 nmol/l (n = 20) in fetal rabbit PASMCs, 76 ± 9 nmol/l (n = 15) in adult rabbit PASMCs, and 454 ± 82 nmol/l (n = 15) in adult rat PASMCs. At 3 mmol/l of DTT, pretreatment of the fetal PA smooth muscle cells for 2 min with 10 μmol/l of nifedipine to block the influx of extracellular Ca2+ through L-type Ca2+ channels significantly reduced the elevation in [Ca2+]i in response to DTT (Fig. 6).
Measurement of changes in tension in isolated DA and PA rings. To test the hypothesis that the reducing agent DTT dilates the normoxic DA and constricts PA, we examined the effect of 3 mmol/l of DTT on tension in isolated fetal rabbit DA, fetal as well as adult rabbit PA, and adult rat PA rings. DTT consistently dilated PE-constricted DA rings under normoxic conditions. Compared with the effect of DTT on DA rings, this agent at the same concentration produced vasoconstriction in PE- or U-46619-constricted PA rings (Fig. 7). Diphenyleneiodonium (DPI; 4 μmol/l) completely blocked the effect of DTT on adult rat PA rings under normoxic conditions (n = 8, not shown). This concentration of DPI has been reported to block hypoxic pulmonary vasoconstriction. Under hypoxic conditions, DTT slightly dilated DA rings (-83.9 ± 22.8 mg, n = 13), which was significantly less compared with the DTT response on DA rings under normoxia (-378.1 ± 62.8 mg, n = 14; P = 0.0002). The effect of DTT on fetal or adult PA rings under hypoxic conditions was not significant (n = 14 and n = 10, respectively; not shown).
The effect of the oxidizing agent DTNB on PE-constricted DA and fetal PA rings was tested under hypoxic conditions. Figure 8 shows the mean data for modulation of vascular tone in endothelium-intact rings by 1 mmol/l of DTNB. DTNB constricted DA rings and dilated fetal PA rings consistently. Under normoxic conditions DTNB caused a slight vasoconstriction of DA rings (64.9 ± 23.5 mg, n = 7). The effect was significantly different from the effect of DTNB in hypoxia (376.8 ± 75.8 mg, n = 10; P = 0.0045), but not from the effect of vehicle (DMSO) alone. DTNB had no significant effect on fetal or adult PA rings under normoxic conditions (n = 11 and n = 5, respectively; not shown).
In general, the responses of PAs and systemic arteries (including DA) to constrictor agents are the same as they are to dilator agents. The dramatic difference between these vessels is that the response to an increase in O2 tension is opposite. Any single intervention that can mimic this disparity might cast light on the mechanism of O2 sensing. Our findings indicate that the same redox signal can elicit opposite effects on whole cell K+ current, Em, cytosolic calcium, and vascular tone in the resistance PAs (both fetal and adult) and in the DA.
Although the effector mechanism of the change in vascular tone in response to an alteration in O2 tension is fairly well characterized, little is known about how the “sensor” detects changes in O2 tension. It has been demonstrated that hypoxic vasoconstriction of PASMCs is mediated, at least in part, by the inhibition of K+ channels leading to cell depolarization, calcium influx, and myocyte contraction (2, 19, 20, 25, 35, 38). In the DASMCs, normoxia-induced constriction is also associated with K+ channel inhibition, membrane depolarization, and an influx of calcium (27, 30, 32). Consequently, it seems likely that the O2-sensitive K+ channels in the PA and DASMCs play an important role in O2 signaling, although it is also possible that calcium release from the sarcoplasmic reticulum (SR) may be involved. The mechanism by which the K+ channels and SR sense these changes in O2 is controversial. Hypoxic signaling may be related to 1) a change to a more reduced cellular redox status, secondary to decreased NAD(P)H oxidase- and/or mitochondria-dependent oxygen radical formation, or 2) a decrease in the ratios of the cytosolic redox couples [NAD/NADH, NADP/NADPH, reduced glutathione/oxidized glutathione (GSH/GSSG)], secondary to slowing of mitochondrial electron transfer.
Reactive oxygen species (ROS) including superoxide anion (), hydroxyl radical (·OH), and hydrogen peroxide (H2O2) are formed in proportion to the level of cellular O2. Acute hypoxia is thought to decrease production and tissue concentrations of ROS (3, 4, 11, 18, 21, 27), although this is still disputed (9, 15, 16, 34). Reeve et al. (27) also showed in DA rings that the increase in ROS recorded when switching from hypoxia to normoxia could be completely reversed by incubation of the tissue with the H2O2 metabolizing enzyme catalase. This finding indicates that the majority of the increase in ROS is due to endogenous increases in levels of H2O2 or a subsequent ROS, such as the ·OH. The source of this increase is not clear but could be the result of mitochondrial production or the activity of NADPH oxidase. Patch-clamp experiments demonstrate that catalase results in activation of the K+ current in DASMCs and consequently hyperpolarizes these cells. Superoxide dismutase, given intracellularly, had no significant effect on the K+ current of DASMCs, suggesting that a change in the level of endogenous H2O2 affects the K+ channels when O2 tension changes, rather than or ·OH. H2O2 might act directly or through changes in redox couples.
Many studies have demonstrated a direct effect of ROS on voltage-gated K+ channels. Chen et al. (8) found that inactivation of the Shaker K+ channels is accelerated by H2O2. Such inactivation would indicate a potential mechanism for O2-induced depolarization in DASMC. In the human ether-a-gogo-related gene channel (which encodes the cardiac repolarizing potassium current), both activation and inactivation are accelerated by H2O2 (6). In lung neuroepithelial body (NEB) cells and the H-146 cell line, the O2-sensitive K+ currents are potentiated by H2O2 or by activation of the NADPH oxidase with phorbol esters (18, 33). These experiments suggest that increased production of H2O2 by NADPH oxidase might participate in the sensing of an increase in O2 tension by NEB. Contrary to these observations that H2O2 production is decreased in hypoxia, Weissmann et al. (36, 37) suggest that in the isolated perfused rabbit lung ROS production is increased by hypoxia. Consistent with these data, Waypa et al. (34) report that H2O2 constricts the pulmonary circulation during normoxia. This discrepancy may relate to differences in preparation methods or to experimental techniques used. In addition, questions arise concerning the precise subcellular locations of ROS production. Regardless of the direction of the change in level of ROS, the redox couples within the cytosol could be affected and modify channel activity, or, alternatively, the ROS could directly alter ion channel activity.
Instead of the ROS being the prime signaling mechanism involved in O2 sensing, it is possible that hypoxic inhibition of mitochondrial electron transport causes the cytosolic redox couples to be more reduced and that this redox shift alters K+ channel gating. Hypoxia increases the ratios of the reduced to the oxidized forms of the cytosolic redox pairs, such as NADH/NAD, NADPH/NADP, and/or glutathione (GSH/GSSG) (1, 7), and thus shifts the cells to a more reduced state. Exogenous reducing agents mimic the effect of hypoxia on several types of O2-sensitive K+ channels. GSH causes an inhibition in the whole cell K+ current and a significant membrane depolarization of cultured PASMCs. In the presence of GSH in these cells, hypoxia has no further effect on K+ current or resting Em, suggesting that both hypoxia and GSH block the same K+ channels (39). GSSG, on the other hand, increases whole cell K+ current (35). In freshly isolated PASMCs, coenzyme Q10 and duroquinone inhibit whole cell K+ current and depolarize the resting Em (29). Park et al. (22) showed that 5 mmol/l of DTT partially blocks whole cell K+ current in PASMCs and accelerates the inactivation kinetics but does not affect steady-state activation and inactivation. Contrary to these findings, in the DA the reducing agent N-mercaptopropionylglycine both increases K+ current and dilates normoxia-constricted rings (27). These data suggest that changes in the redox status of the cytoplasm might alter the gating of K+ channels.
It has been suggested that, in the PASMCs, the first effect of hypoxia might be the release of intracellular calcium, which could in turn lead to inhibition of K+ channels and influx of external calcium (24). If this hypothesis is extrapolated to the DASMCs, normoxic contraction would also have to be initiated by the release of intracellular calcium. However, it has been demonstrated that the normoxic contraction of the DA can be completely prevented by inhibition of calcium influx by either lanthanum or nisoldipine (32). It is still possible that intracellular release of calcium occurs sufficiently to act as a signal but not enough to cause contraction by itself.
The present study demonstrates the opposite modulation of K+ channel activity by the same redox changes in the cytosol of SMCs. The reducing agent DTT decreases K+ current, causes membrane depolarization and pulmonary vasoconstriction under normoxic conditions in PA, and, on the contrary, enhances the K+ current, causes hyperpolarization and vasodilatation in DA, and thus mimics hypoxia in both tissues. We also show an increase of [Ca2+]i after application of DTT in PASMCs and a decrease in DASMCs. The response to DTT in fetal rabbit PASMCs was significantly inhibited by pretreatment of the cells with 10 μmol/l of nifedipine, suggesting that entry of extracellular Ca2+ through voltage-dependent Ca2+ channels is a major source of DTT-induced increase in [Ca2+]i and DTT-induced vasoconstriction. The oxidizing agent DTNB increases K+ current and causes membrane hyperpolarization and vasodilatation under hypoxic conditions in PA. In DA, a decrease of K+ current, membrane depolarization, and vasoconstriction could be demonstrated, and thus DTNB mimics normoxia in both tissues. Because DTT and DTNB are redox-active chemicals that alter thiol groups on proteins, the redox sites could involve many of the sulfhydryl residues present in the voltage-gated K+ channels in PA and DASMCs. It is possible that the actions of these sulfhydryl-specific redox agents are distinct from the redox couples, such as NADP/NADPH and GSSG/GSH, or from the physiological effects of changes in O2 tension. However, the ability of these redox agents to elicit exactly opposite responses in the DA and PA is consistent with a role for redox changes in the contrary vascular responses to hypoxia and normoxia. One can speculate that the key to our understanding of the mechanism of O2 sensing may lie in redox control of the gating of K+ channels and perhaps also of the release of calcium from intracellular stores.
Andrea Olschewski is supported by the Deutsche Forschungsgemeinschaft (Ol 127/1-1). E. K. Weir is supported by VA merit review funding and National Heart, Lung, and Blood Institute Grant RO1-HL-65322-01A1.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society