A Flavoprotein Mechanism Appears to Prevent an Oxygen-Dependent Inhibition of cGMP-Associated Nitric Oxide–Elicited Relaxation of Bovine Coronary Arteries
Abstract—The redox state of the heme of soluble guanylate cyclase (sGC) may regulate the sensitivity of vascular tissue to nitric oxide (NO). In this study, diphenyliodonium (DPI) is used as an inhibitor of flavoprotein oxidoreductases to examine their potential role in the expression of NO-elicited cGMP-associated arterial relaxation and sGC stimulation. The relaxation of endothelium-removed bovine coronary arteries (BCAs) precontracted with 30 mmol/L KCl to the NO donor S-nitroso-N-acetyl-penicillamine (SNAP) or to NO is markedly suppressed by 10 μmol/L DPI under an atmosphere of 21% O2 (5% CO2). In contrast, DPI has minimal effects on the relaxation to SNAP under 95% N2 (5% CO2). If BCAs are treated with DPI under 21% O2 and then exposed to the hemoprotein reductant sodium dithionite (1 mmol/L) under N2, there is a partial reversal of the inhibitory effects of DPI compared with BCAs that were not treated with dithionite. DPI did not inhibit relaxation elicited by 8-bromo-cGMP or forskolin. Increases in tissue cGMP levels stimulated by SNAP were eliminated by pretreatment of BCAs with DPI under 21% O2 but not under N2. Activation of sGC by SNAP in BCA homogenate was also eliminated when vessels were pretreated with 10 μmol/L DPI under 21% O2, but DPI did not have an inhibitory effect when directly added to the assay of sGC activity. These observations are consistent with a flavoprotein-dependent oxidoreductase functioning to prevent the expression of a novel O2-dependent process from oxidizing the heme on sGC and inhibiting NO-elicited cGMP-mediated BCA relaxation.
Nitric oxide (NO) stimulates the soluble form of guanylate cyclase (sGC) through binding its Fe2+ heme1 2 3 ; whereas, if the heme on guanylate cyclase is in its Fe3+ form, it is not readily activated by NO.4 5 6 Relaxation to NO is also attenuated by increases in the levels of endogenous superoxide anion when the activity of copper/zinc–superoxide dismutase is inhibited.7 8 In endothelium-removed bovine coronary arteries (BCAs), the main source of superoxide anion production appears to be an NADH oxidase that is inhibited by the flavoprotein probe diphenyliodonium (DPI).9 Previous studies in bovine pulmonary arteries examining the effects of modulation of cytosolic NAD(H) have provided evidence that superoxide anion derived from NADH oxidase has the ability to inhibit responses to NO when the activity of copper/zinc–superoxide dismutase is depressed.10 While studying the effects of DPI on coronary vasodilators, it was observed that this probe had some unexpected inhibitory actions on relaxation to NO generating agents, which suggested the DPI probe had additional interactions with systems influencing the response of BCAs to NO.
Diphenyliodonium and diphenyleneiodonium are agents that appear to inactivate flavoproteins at their flavin sites during electron-transfer reactions catalyzed by these proteins.11 These iodonium compounds seem to inhibit several important flavoprotein-containing systems including NAD(P)H oxidases, NO synthase, cytochrome P450 reductase, mitochondrial electron transport, xanthine oxidase, and other oxidoreductases.11 12 13 Diphenyleneiodonium has also been reported to inhibit K+ and Ca2+ channel currents.14 Although some of these systems are of potential importance in the control of vascular force generation and responses to vasoactive stimuli, none of the known actions of DPI would be expected to cause a selective inhibitory effect on cGMP-mediated relaxation to exogenous sources of NO that was seen in isolated endothelium-removed BCAs. In the present study, we examined aspects of the mechanism of action of DPI on relaxation of BCA to the NO donor S-nitroso-N-acetyl-penicillamine (SNAP) either under normoxic or severely hypoxic conditions and measured tissue cGMP levels and sGC activity to better establish the origins of the novel observed effects of DPI.
Materials and Methods
The sources of most of the reagents used for the present studies have been previously described.9 10 8-bromo-cGMP (8-Br-cGMP) was from Sigma, and SNAP was synthesized by methods previously published.15
Measurement of Changes in Force in BCAs
The preparation of isolated endothelium-removed BCA rings (left circumflex) and techniques used for measurement of changes in isometric force in Krebs bicarbonate buffer were adapted from previously described methods.8 9 The vessels were initially incubated in either the absence or presence of 10 μmol/L DPI for 30 minutes under the atmosphere of 21% O2 (5% CO2, balance N2) or 95% N2 (5% CO2, Po2=8 to 10 torr). After the incubation, the vessels were contracted with 30 mmol/L KCl (K+) and exposed to vasorelaxant agents in either the absence or continued presence of DPI under the conditions described in the Results. In experiments in which NO gas was used, a 10 or 100 ppm NO gas mixture (balance N2) was delivered in a manner that produced an ≈2.5 or 25 nmol/L steady-state buffer concentration of NO,16 respectively. In some experiments, arteries were treated with sodium dithionite (Baker Chemicals) for 30 minutes under 21% O2 or N2 after exposure to DPI under 21% O2. Studies involving reoxygenation of BCAs were avoided, because this treatment causes an increase in superoxide anion–derived hydrogen peroxide (H2O2), which stimulates the activity of sGC.9 Relaxation was expressed by the percentage of change of steady-state level of contraction.
Determination of cGMP Levels in BCAs
BCA rings were prepared and treated as described for the force measurement studies. The rings were immediately frozen in liquid N2 at the time point before the application of 30 mmol/L K+ (basal) or SNAP (30 mmol/L K+) or 1 minute after administration of each concentration of SNAP15 after the arteries had been incubated in either the absence or presence of 10 μmol/L DPI for 30 minutes under the atmosphere of either 21% O2 or N2. The frozen rings were pulverized in a mortar; 1 mL/g tissue of phosphate buffer containing 10% trichloroacetic acid was added to the samples, and cGMP was extracted as previously described.15 Levels of cGMP in this sample (50 μL) were measured by the enzyme immunoassay method,10 and the results were expressed as pmol cGMP/g tissue.
Determination of sGC Activity in BCA Homogenate
Arterial segments weighing ≈350 mg were prepared and treated similarly to the force measurement studies but without measuring force. After incubation in either the absence or presence of 10 μmol/L DPI for 30 minutes under the atmosphere of 21% O2, arteries were immediately frozen in liquid N2. Pulverized, frozen tissue samples were suspended in 1:3 wt/vol MOPS-sucrose buffer, pH 7.4, and then homogenized using a glass Teflon homogenizer at 4°C to 8°C. Homogenates (0.1 mL) were assayed for sGC assay activity (0.2 mL final volume) using 10 minutes of air-equilibrated incubations at 37°C using a previously published protocol10 with 0.1 mmol/L GTP and 2 mmol/L MgCl2. Activities of sGC in pmol cGMP per min/mg protein were based on measurements of cGMP by enzyme immunoassay and Bradford protein assays.17
Results were expressed as mean±SE, with n equal to the number of animals used. Comparisons between groups were made with an ANOVA and a Student’s t test with a Bonferroni correction for multiple comparisons. A P<0.05 was used to determine statistical significance.
Effects of DPI on BCA Relaxation to SNAP
In BCAs, the relaxation to SNAP is essentially eliminated by 10 μmol/L 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ),18 an agent that appears to inhibit the stimulation of sGC by oxidizing its NO-Fe2+ heme complex to the Fe3+ form.5 As shown in Figure 1A⇓, 10 μmol/L DPI markedly inhibited the relaxation of 30 mmol/L K+ precontracted BCAs to 0.1 μmol/L and 10 μmol/L SNAP when examined under the atmosphere of 21% O2. Figure 1B⇓ shows relaxation of BCAs to SNAP under the atmosphere of N2. In contrast to the effects of DPI under 21% O2, DPI had only minimal inhibitory effects on the dose-dependent relaxation of 30 mmol/L K+ precontracted BCAs to SNAP under N2. To exclude the possibility of an interaction of DPI with SNAP, these agents were incubated together for 15 minutes at 37°C under 21% O2 and then the mixture was added to the tissue bath in a manner that produces a final concentration of 1 μmol/L SNAP and 10 μmol/L DPI. SNAP-induced relaxation under 21% O2 was not significantly affected by preincubation of SNAP with DPI (control, 90.8±4.2%; DPI, 91.3±2.9% relaxation, n=4).
Effects of DPI on BCA Relaxation to NO
NO gas was also used as a relaxant agent to exclude the possibility of an interaction of DPI with the processes that release NO from SNAP. NO caused dose-dependent relaxation of 30 mmol/L K+ precontracted BCAs, and 10 μmol/L DPI significantly inhibited relaxation to 2.5 nmol/L and 25 nmol/L NO, as shown in Figure 2⇓.
Effects of Dithionite on the Inhibitory Effect of DPI
Sodium dithionite (1 mmol/L), an agent that reduces the oxidized form of hemoproteins, was used to examine whether it could reverse the inhibitory effects of DPI on SNAP-induced relaxation of BCAs. As shown in Figure 3A⇓, 1 mmol/L dithionite itself had no effect on SNAP-elicited relaxation and it did not alter the inhibitory effect of DPI when the vessels were incubated with dithionite under 21% O2 after DPI pretreatment under 21% O2. In contrast, dithionite partially reversed the inhibitory effects of DPI on SNAP-elicited relaxation when the vessels were incubated with dithionite under N2 after DPI pretreatment under 21% O2 compared with BCAs that were not treated with dithionite (Figure 3B⇓). The inhibitory effect of DPI was not significantly altered by BCA incubation under N2 (in the absence of dithionite) after DPI pretreatment under 21% O2 compared with BCAs that were incubated under 21% O2 without exposure to N2 or dithionite.
Effects of DPI on BCA Relaxation to a cGMP Analogue or Forskolin
As shown in Figure 4A⇓, the cell-permeable cGMP analogue 8-Br-cGMP caused dose-dependent relaxation of 30 mmol/L K+ precontracted BCAs. This relaxation was not significantly altered by 10 μmol/L DPI. A small enhancement of relaxation was observed only at the highest concentration of 8-Br-cGMP examined (300 μmol/L). Figure 4B⇓ shows dose-dependent relaxation of BCAs to the adenylate cyclase stimulant forskolin, and this relaxation was also not affected by DPI.
Effects of DPI on Tissue Levels of cGMP
The data in Figure 5⇓ summarize measurements of tissue levels of cGMP in BCAs. Basal cGMP level without any treatment was 61.1±14.8 pmol/g (n=7; control, 21% O2), and exposure of BCAs to 30 mmol/L K+ for 15 minutes did not affect the tissue levels of cGMP. A 30-minute DPI (10 μmol/L) pretreatment itself had no effect on cGMP levels under either 21% O2 or N2. SNAP (1 μmol/L and 10 μmol/L) markedly increased tissue cGMP levels in control tissues, and this increase was inhibited by 10 μmol/L DPI when treated under 21% O2, whereas no inhibition was observed under N2.
Effects of DPI on the Stimulation of sGC Activity by SNAP in BCA Homogenate
Figure 6⇓ shows sGC activity in BCA homogenates. Basal sGC activity was 86.0±29.4 pmol cGMP per min/mg protein (n=25), and 10 μmol/L SNAP increased the sGC activity by 227% (n=25) in the absence of DPI pretreatment. Pretreatment of BCAs with 10 μmol/L DPI itself had no significant effect on sGC activity, but SNAP failed to increase sGC activity in homogenates derived from DPI pretreated tissue. To examine whether the presence of DPI affects sGC activation by SNAP in assays of sGC activity, 10 μmol/L DPI was directly added to the homogenate derived from control tissues. Exposure to 10 μmol/L SNAP increased sGC activity by 233% (control, 63.3±8.8 and SNAP, 147.5±15.5 pmol cGMP per min/mg protein, n=4), and this activation was not affected by the presence of DPI (145.0±11.9 pmol cGMP per min/mg protein, n=4).
The effects of DPI observed in the present study suggest that a flavoprotein-containing system contributes to maintaining the sensitivity of BCA relaxation to NO under aerobic conditions. Because DPI inhibited BCA relaxation to SNAP (and NO) under a normoxic atmosphere but not under hypoxia, an O2-dependent process is contributing to the actions of DPI. Although DPI has often been used as an agent that inhibits superoxide production via NAD(P)H oxidases, this action of DPI would be expected to enhance the stimulation of sGC and relaxation of BCAs to NO by lowering the levels of endogenous superoxide. Because it has previously been shown9 that 10 μmol/L DPI does not appear to regulate force generation in BCAs under normoxia, it does not appear to be altering the activity of sGC through a process such as stimulating its activity with H2O2 that is derived from endogenously produced superoxide. Thus, the results of this study suggest that superoxide and H2O2 have little effect, if any, on BCA relaxation under the experimental conditions examined. Based on the reversal of the O2-dependent inhibitory effects of DPI by dithionite, these responses may originate from a potentially novel O2-dependent mechanism of inhibition of sGC stimulation by NO, which is shown in the model in Figure 7⇓. In this model, the loss of responsiveness of BCAs to NO presumably originates from an oxidation of the heme on sGC, and a flavoprotein-dependent process is hypothesized to maintain the heme of sGC in its ferrous form, which is essential for the actions of NO on this system.
Interpretation of the effects of the flavoprotein inhibitor DPI requires consideration of the fact that it has several potential sites of action that could influence the control of vascular contractile function. As mentioned above, it is unlikely that the actions of DPI result from inhibition of the generation of superoxide by flavoprotein systems such as NAD(P)H oxidases. Because the results of the present study indicate that DPI inhibits relaxation to NO in endothelium-removed BCAs, it is not functioning through inhibiting the generation of endogenous NO by the NO synthase reaction or by preventing the generation of NO from SNAP. DPI does not appear to be a direct oxidant of the heme of sGC, because it did not attenuate the relaxation to NO under a N2 atmosphere and the addition of DPI to assays of sGC activity in BCA homogenates did not alter the stimulation of cGMP production by SNAP. Because BCAs treated with DPI under aerobic conditions showed a loss of NO-elicited increases in tissue levels of cGMP and stimulation of homogenate sGC activity, without causing alterations in relaxation to 8-Br-cGMP, the treatment of BCAs with DPI appears to cause a loss of the ability of NO to stimulate sGC activity and it does not seem to have an action on the processes involved in the mechanism through which cGMP mediates vascular relaxation.
The observed restoration by the hemoprotein reductant dithionite of the loss of responsiveness to NO-elicited relaxation of BCAs pretreated with DPI under aerobic conditions is consistent with the heme of sGC undergoing oxidation to its ferric form as a result of the DPI treatment. The data in Figure 3⇑ indicate that hypoxia itself does not restore the loss of relaxation to SNAP caused by the initial DPI pretreatment under aerobic conditions and that dithionite does not appear to influence the response of BCAs to the NO donor SNAP unless it is first inhibited by DPI. It is well established that the heme of sGC does not rapidly autooxidize to its ferric form on incubation.1 3 4 5 6 Thus, the observation in the present study that exposure of the BCA homogenate to DPI during the assay of sGC does not alter the ability of SNAP to activate the production of cGMP suggests that the treatment of BCAs with DPI permits the detection of a tonically active O2-dependent process that causes the loss of responsiveness to NO through a hypothesized heme oxidation mechanism, which remains to be identified. Because of the low levels of sGC present in BCA, there is a lack of alternative approaches available to detect the hypothesized changes in the redox state of the heme of sGC that are caused by the DPI treatment. The novel O2-dependent process in BCAs that causes a loss of responsiveness to NO presumably functions through converting the ferrous heme of sGC to the ferric form, which does not readily bind NO. Detection of this process seems to be prevented by a DPI-inhibitable flavoprotein oxidoreductase, which appears to maintain the heme of sGC in the ferrous form, which is essential for the action of NO.
Observations made in the present study are consistent with the redox status of the heme on sGC controlling the sensitivity of BCA relaxation to NO. The data suggest that the heme of sGC is normally maintained in its Fe2+ oxidation state, which is presumably involved in the stimulation of cGMP-mediated relaxation to NO, through the actions of a flavoprotein-containing oxidoreductase. Although these observations do not prove the heme of sGC was oxidized, recent experiments19 have detected evidence that a flavoprotein-containing NADPH oxidoreductase appears to restore the sensitivity of sGC to NO in bovine pulmonary artery preparations that have been exposed to the hemoprotein oxidant ferricyanide and ODQ, an agent that oxidizes the NO heme of sGC to its ferric form. This other study also contains evidence that treatment of the pulmonary arteries with a dose of DPI that inhibits the NADPH-dependent reductase under an atmosphere of 21% O2 does not result in an attenuation of the relaxation to NO, suggesting that the process that is hypothesized to oxidize the heme of sGC appears to be much more active in bovine coronary arteries. The actions of DPI and dithionite observed in the present study suggest that when DPI inhibits the function of the flavoprotein oxidoreductase that reduces the heme of sGC, evidence for a novel O2-dependent process that impairs the sensitivity of sGC to NO can be detected. Although the actual mechanism of this O2-dependent impairment of NO stimulation of sGC is not yet understood, it may involve a tonically active O2-dependent process that oxidizes the heme of sGC to its Fe3+ form. Thus, an additional mechanism potentially controlling the sensitivity of vascular preparations to actions of NO may involve a balance between the systems that oxidize and reduce the heme of sGC. It is likely that the function of these systems may be altered by pathophysiological processes that modify vascular function.
This work was supported by United States Public Health Services Grants HL31069 and HL43023 from the National Heart, Lung, and Blood Institute. Dr Iesaki was partly supported by a Research Fellowship from the Uehara Memorial Foundation.
Presented in part at the 71st Scientific Sessions of the American Heart Association, Dallas, Tex, November 8–11, 1998, and published in abstract form (Circulation. 1998;98:I-666).
- Received May 21, 1999.
- Accepted September 13, 1999.
- © 1999 American Heart Association, Inc.
Craven PA, DeRubertis FR. Restoration of the responsiveness of purified guanylate cyclase to nitrosoguanidine, nitric oxide, and related activators by heme and hemoproteins. Evidence for involvement of the paramagnetic nitrosyl-heme complex in enzyme activation. J Biol Chem. 1978;253:8433–8443.
Wolin MS, Wood KS, Ignarro LJ. Guanylate cyclase from bovine lung: a kinetic analysis of the regulation of the purified soluble enzyme by protoporphyrin IX, heme and nitrosyl-heme. J Biol Chem. 1982;257: 13312–13320.
Schrammel A, Behrends S, Schmidt K, Koesling D, Mayer B. Characterization of 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one as a heme-site inhibitor of nitric oxide-sensitive guanyl cyclase. Mol Pharmacol. 1996;50:1–5.
Cherry PD, Omar HA, Farrell KA, Stuart JS, Wolin MS. Superoxide anion inhibits cGMP-associated bovine pulmonary arterial relaxation. Am J Physiol. 1990;259:H1056–H1062.
Omar HA, Cherry PD, Mortelliti MP, Burke-Wolin T, Wolin MS. Inhibition of coronary artery superoxide dismutase attenuates endothelium-dependent and independent nitrovasodilator relaxation. Circ Res. 1991;69:601–608.
Mohazzab-H KM, Fayngersh RP, Kaminski PM, Wolin MS. Oxygen-elicited responses in calf coronary arteries: role of H2O2 production via NADH-derived superoxide. Am J Physiol. 1996;270:H1044–H1053.
Gupte S, Rupawalla T, Mohazzab-H KM, Wolin MS. Regulation of NO-elicited pulmonary artery relaxation and guanylate cyclase activation by NADH oxidase and SOD. Am J Physiol. 1999;276:H1535–H1542.
O’Donnell VB, Smith GCM, Jones OTG. Involvement of phenyl radicals in iodonium compound inhibition of flavoprotein enzymes. Mol Pharmacol. 1994;46:778–785.
Stuehr DJ, Fasehun OA, Kwon NS, Gross SS, Gonzalez JA, Levi R, Nathan CF. Inhibition of macrophage and endothelial cell nitric oxide synthase by diphenyleneiodonium and its analogs. FASEB J. 1991;5:98–103.
Weir EK, Wyatt CN, Reeve HL, Huang J, Archer SL, Peers C. Diphenyleneiodonium inhibits both potassium and calcium channels in isolated pulmonary artery smooth muscle cells. J Appl Physiol. 1994;76:2611–2615.
Ignarro LJ, Lippton H, Edwards JC, Baricos WH, Hyman AL, Kadowitz PJ, Gruetter CA. Mechanism of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprusside and nitric oxide: evidence for the involvement of S-nitrosothiols as active intermediates. J Pharmacol Exp Ther. 1981;218:739–749.
Iesaki T, Gupte SA, Kaminski PM, Wolin MS. Inhibition of guanylate cyclase stimulation by NO and bovine arterial relaxation to peroxynitrite and H2O2. Am J Physiol. 1999;277:H978–H985.
Gupte S, Rupawalla T, Phillibert D, Wolin MS. NADPH and heme redox appear to modulate pulmonary artery relaxation and guanylate cyclase activation by NO. Am J Physiol. In press.