Title: Inhibition of Xanthine Oxidase and Xanthine Dehydrogenase by Nitric Oxide
Abstract: Xanthine oxidase (XO) and xanthine dehydrogenase (XDH) were inactivated by incubation with nitric oxide under anaerobic conditions in the presence of xanthine or allopurinol. The inactivation was not pronounced in the absence of an electron donor, indicating that only the reduced enzyme form was inactivated by nitric oxide. The second-order rate constant of the reaction between reduced XO and nitric oxide was determined to be 14.8 ± 1.4m−1 s−1 at 25 °C. The inactivated enzymes lacked xanthine-dichlorophenolindophenol activity, and the oxypurinol-bound form of XO was partly protected from the inactivation. The absorption spectrum of the inactivated enzyme was not markedly different from that of the normal enzyme. The flavin and iron-sulfur centers of inactivated XO were reduced by dithionite and reoxidized readily with oxygen, and inactivated XDH retained electron transfer activities from NADH to electron acceptors, consistent with the conclusion that the flavin and iron-sulfur centers of the inactivated enzyme both remained intact. Inactivated XO reduced with 6-methylpurine showed no "very rapid" spectra, indicating that the molybdopterin moiety was damaged. Furthermore, inactivated XO reduced by dithionite showed the same slow Mo(V) spectrum as that derived from the desulfo-type enzyme. On the other hand, inactivated XO reduced by dithionite exhibited the same signals for iron-sulfur centers as the normal enzyme. Inactivated XO recovered its activity in the presence of a sulfide-generating system. It is concluded that nitric oxide reacts with an essential sulfur of the reduced molybdenum center of XO and XDH to produce desulfo-type inactive enzymes. Xanthine oxidase (XO) and xanthine dehydrogenase (XDH) were inactivated by incubation with nitric oxide under anaerobic conditions in the presence of xanthine or allopurinol. The inactivation was not pronounced in the absence of an electron donor, indicating that only the reduced enzyme form was inactivated by nitric oxide. The second-order rate constant of the reaction between reduced XO and nitric oxide was determined to be 14.8 ± 1.4m−1 s−1 at 25 °C. The inactivated enzymes lacked xanthine-dichlorophenolindophenol activity, and the oxypurinol-bound form of XO was partly protected from the inactivation. The absorption spectrum of the inactivated enzyme was not markedly different from that of the normal enzyme. The flavin and iron-sulfur centers of inactivated XO were reduced by dithionite and reoxidized readily with oxygen, and inactivated XDH retained electron transfer activities from NADH to electron acceptors, consistent with the conclusion that the flavin and iron-sulfur centers of the inactivated enzyme both remained intact. Inactivated XO reduced with 6-methylpurine showed no "very rapid" spectra, indicating that the molybdopterin moiety was damaged. Furthermore, inactivated XO reduced by dithionite showed the same slow Mo(V) spectrum as that derived from the desulfo-type enzyme. On the other hand, inactivated XO reduced by dithionite exhibited the same signals for iron-sulfur centers as the normal enzyme. Inactivated XO recovered its activity in the presence of a sulfide-generating system. It is concluded that nitric oxide reacts with an essential sulfur of the reduced molybdenum center of XO and XDH to produce desulfo-type inactive enzymes. xanthine oxidase xanthine dehydrogenase nitric oxide activity/flavin ratio 2,6-dichlorophenolindophenol 2-(cyclohexylamino)ethanesulfonic acid N,N-bis(2-hydroxyethyl)glycine Mammalian xanthine oxidase (XO1; EC 1.1.3.22) and xanthine dehydrogenase (XDH; EC 1.1.1.204), which are alternative forms of the same gene product (for a recent review, see Ref. 1Hille R. Nishino T. FASEB J. 1995; 9: 995-1003Crossref PubMed Scopus (383) Google Scholar), are complex flavoproteins composed of two identical subunits ofM r 145,000; each subunit contains one molybdenum center, two non-identical Fe2S2-type iron-sulfur centers, and one FAD center. The enzymes catalyze oxidation of xanthine to uric acid with concomitant reduction of NAD+or molecular oxygen. The oxidative hydroxylation of xanthine to uric acid takes place at the molybdenum center, and reducing equivalents thus introduced into enzymes are transferred rapidly via intramolecular electron transfer to FAD, where physiological oxidation occurs (2Olson J.S. Ballou D.P. Palmer G. Massey V. J. Biol. Chem. 1974; 249: 4363-4382Abstract Full Text PDF PubMed Google Scholar). Both enzymes can reduce molecular oxygen to superoxide and hydrogen peroxide (1Hille R. Nishino T. FASEB J. 1995; 9: 995-1003Crossref PubMed Scopus (383) Google Scholar), but XDH is characterized by high reactivity toward NAD+, but low reactivity toward O2, whereas XO has high reactivity toward O2, but negligible reactivity toward NAD+ (2Olson J.S. Ballou D.P. Palmer G. Massey V. J. Biol. Chem. 1974; 249: 4363-4382Abstract Full Text PDF PubMed Google Scholar). Mammalian XDH can be converted to XO either by the oxidation of sulfhydryl groups or by limited proteolysis (3Corte E.D. Stirpe F. Biochem. J. 1968; 108: 349-351Crossref PubMed Scopus (92) Google Scholar, 4Corte E.D. Stirpe F. Biochem. J. 1972; 126: 739-745Crossref PubMed Scopus (351) Google Scholar). XO is thought to be one of the key enzymes for cellular injury by superoxide and related active oxygen species (5Granger D.N. Rutili G. McCord J.M. Gastroenterology. 1981; 81: 22-29Abstract Full Text PDF PubMed Google Scholar). In particular, much attention has been paid to XO in connection with the pathogenesis of ischemia/reperfusion injury since it has been proposed that superoxide/H2O2 production by XO is enhanced by the accelerated conversion of XDH to XO (6McCord J.M. N. Engl. J. Med. 1985; 312: 159-163Crossref PubMed Scopus (4999) Google Scholar), the accumulation of ATP degradation products (i.e. hypoxanthine and xanthine) that are substrates for XO (7Saugstad O.D. Pediatr. Res. 1988; 23: 143-150Crossref PubMed Scopus (324) Google Scholar), and the up-regulation of XDH/XO mRNA (8Hassoun P.M. Yu F.S. Shedd A.L. Zulueta J.J. Thannickal V.J. Lanzillo J.J. Fanburg B.L. Am. J. Physiol. 1994; 266: L163-L171PubMed Google Scholar). However, the role of XDH/XO in the pathogenesis of such injury is still controversial (9Nishino T. J. Biochem. (Tokyo). 1994; 116: 1-6Crossref PubMed Scopus (191) Google Scholar). Nitric oxide (NO) is now recognized as a multifunctional molecule (10Gross S.S. Wolin M.S. Annu. Rev. Physiol. 1995; 57: 737-769Crossref PubMed Scopus (821) Google Scholar,11Garthwaite J. Boulton C.L. Annu. Rev. Physiol. 1995; 57: 683-706Crossref PubMed Scopus (1542) Google Scholar), one function of which is to inactivate biologically important enzymes such as aconitase (12Gardner P.R. Costantino G. Szabo C. Salzman A.L. J. Biol. Chem. 1997; 272: 25071-25076Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar), ribonucleotide reductase (13Lepoivre M. Flaman J.M. Henry Y. J. Biol. Chem. 1992; 267: 22994-23000Abstract Full Text PDF PubMed Google Scholar), glutathione peroxidase (14Asahi M. Fujii J. Suzuki K. Seo H.G. Kuzuya T. Hori M. Tada M. Fujii S. Taniguchi N. J. Biol. Chem. 1995; 270: 21035-21039Crossref PubMed Scopus (288) Google Scholar), cytochrome c oxidase (15Cleeter M.W. Cooper J.M. Darley Usmar V.M. Moncada S. Schapira A.H. FEBS Lett. 1994; 345: 50-54Crossref PubMed Scopus (1158) Google Scholar), and NADPH oxidase (16Fujii H. Ichimori K. Hoshiai K. Nakazawa H. J. Biol. Chem. 1997; 272: 32773-32778Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar, 17Clancy R.M. Leszczynska Piziak J. Abramson S.B. J. Clin. Invest. 1992; 90: 1116-1121Crossref PubMed Scopus (645) Google Scholar). Although inhibitory actions of NO are attributed to its reaction with heme or non-heme iron or copper or toS-nitrosylation or sulfhydryl oxidation (18Stamler J.S. Jaraki O. Osborne J. Simon D.I. Keaney J. Vita J. Singel D. Valeri C.R. Loscalzo J. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 7674-7677Crossref PubMed Scopus (1133) Google Scholar), the precise mechanisms remain to be established (12Gardner P.R. Costantino G. Szabo C. Salzman A.L. J. Biol. Chem. 1997; 272: 25071-25076Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar). Considering that NO generation also increases under conditions where superoxide generation increases, it is necessary to clarify whether NO inhibits superoxide-generating systems. We recently investigated two major sources of superoxide. We identified the site affected by NO in neutrophil NADPH oxidase (16Fujii H. Ichimori K. Hoshiai K. Nakazawa H. J. Biol. Chem. 1997; 272: 32773-32778Abstract Full Text Full Text PDF PubMed Scopus (121) Google Scholar) and showed that NO inhibits XO during enzyme turnover under cell-free conditions (19Fukahori M. Ichimori K. Ishida H. Nakagawa H. Okino H. Free Radical Res. 1994; 21: 203-212Crossref PubMed Scopus (71) Google Scholar). However, the exact mechanism and kinetics of the reaction between NO and XO remained to be examined. Subsequent reports demonstrated the NO-induced inactivation of XDH and XO in interferon-γ-stimulated macrophages (20Rinaldo J.E. Clark M. Parinello J. Shepherd V.L. Am. J. Respir. Cell Mol. Biol. 1994; 11: 625-630Crossref PubMed Scopus (28) Google Scholar) and endothelial cells using exogenously and endogenously produced NO (21Hassoun P.M. Yu F.S. Zulueta J.J. White A.C. Lanzillo J.J. Am. J. Physiol. 1995; 268: L809-L817PubMed Google Scholar, 22Cote C.G. Yu F.S. Zulueta J.J. Vosatka R.J. Hassoun P.M. Am. J. Physiol. 1996; 271: L869-L874Crossref PubMed Google Scholar). Direct binding of NO to the enzyme iron-sulfur moiety or to its sulfhydryl groups was postulated since NO did not significantly affect the mRNA expression of XDH in these cellular systems (20Rinaldo J.E. Clark M. Parinello J. Shepherd V.L. Am. J. Respir. Cell Mol. Biol. 1994; 11: 625-630Crossref PubMed Scopus (28) Google Scholar,22Cote C.G. Yu F.S. Zulueta J.J. Vosatka R.J. Hassoun P.M. Am. J. Physiol. 1996; 271: L869-L874Crossref PubMed Google Scholar) and may modify XO directly at a post-translational level. The exact mechanism of the inhibition still remains to be determined because of the limitations of the cellular system. In this study, we examined the kinetics of the reaction between XO and NO and identified the site responsible for the inactivation. The inhibitory kinetics of XDH were also examined to obtain basic data on the interaction of NO and the superoxide-generating system in the event of reduced oxygen supply, such as ischemia. XO was isolated from fresh bovine milk and purified according to the method reported previously (23Nishino T. Nishino T. Tsushima K. FEBS Lett. 1981; 131: 369-372Crossref PubMed Scopus (70) Google Scholar). The activity/flavin ratio (AFR) of the prepared XO was >190 (fully active enzyme has a value of 210) (23Nishino T. Nishino T. Tsushima K. FEBS Lett. 1981; 131: 369-372Crossref PubMed Scopus (70) Google Scholar, 24Massey V. Komai H. Palmer G. Elion G.B. J. Biol. Chem. 1970; 245: 2837-2844Abstract Full Text PDF PubMed Google Scholar, 25Massey V. Edmondson D. J. Biol. Chem. 1970; 245: 6595-6598Abstract Full Text PDF PubMed Google Scholar). The desulfoenzyme was prepared by incubating the enzyme with 10 mm KCN for 2 h at 25 °C, followed by gel filtration to remove KCN as described by Massey and Edmondson (25Massey V. Edmondson D. J. Biol. Chem. 1970; 245: 6595-6598Abstract Full Text PDF PubMed Google Scholar). Bovine milk XDH, prepared essentially according to the method of Nakamura and Yamazaki (26Nakamura M. Yamazaki I. J. Biochem. (Tokyo). 1982; 92: 1279-1286Crossref PubMed Scopus (57) Google Scholar) without folate affinity chromatography, has an AFR of >100 and a dehydrogenase/oxidase activity ratio of >7. The dehydrogenase/oxidase activity ratio was determined as the ratio of the absorbance change at 295 nm under aerobic conditions in the presence of NAD+ to that in the absence of NAD+. Desulfo-XDH was prepared using the same procedure as described for XO, except that 5 mmdithiothreitol was added to the incubation mixture to avoid conversion to XO during preparation. Xanthine, allopurinol, and Hepes were obtained from Sigma. A stock solution of oxymyoglobin (∼0.7 mm) was prepared in 200 mm Hepes buffer (pH 7.0) by reducing metmyoglobin (horse heart, Sigma) with sodium dithionite, followed by gel filtration on a Sephadex G-25 column (medium) (27Wittenberg J.B. Wittenberg B.A. Methods Enzymol. 1981; 76: 29-42Crossref PubMed Scopus (72) Google Scholar). The solution was stored frozen at −80 °C until use. Spectrophotometric measurements were conducted on a Hitachi U-3200 spectrophotometer equipped with a temperature-controlled circulator. The XO concentration was determined spectrophotometrically using a molar absorption coefficient of 37,800 m−1 cm−1 at 450 nm (28Massey V. Brumby P.E. Komai H. J. Biol. Chem. 1969; 244: 1682-1691Abstract Full Text PDF PubMed Google Scholar). The concentration of nitric oxide in reaction mixtures was determined spectrophotometrically using oxymyoglobin. The principle of this procedure is the same as that of the oxyhemoglobin method reported previously (29Murphy M.E. Noack E. Methods Enzymol. 1994; 233: 240-250Crossref PubMed Scopus (347) Google Scholar). The standard oxymyoglobin concentration was determined spectrophotometrically using a molar absorption coefficient of 10,700m−1 cm−1 at 540 nm after conversion to cyanometmyoglobin (30Riggs A. Methods Enzymol. 1981; 76: 5-29Crossref PubMed Scopus (244) Google Scholar, 31Rothgeb T.M. Gurd F.R. Methods Enzymol. 1978; 52: 473-486Crossref PubMed Scopus (53) Google Scholar). Nitric oxide solution was mixed with 5–10 μm oxymyoglobin in 200 mmHepes buffer (pH 7.4), and the increase in absorbance at 406 nm due to metmyoglobin formation was monitored. The standard curve for the assay was obtained by oxidation of oxymyoglobin with titrated amounts of potassium ferricyanide. The oxidation was complete at 15 min at room temperature. All activity measurements were performed at 25 °C. For XO, xanthine-O2 activity was measured spectrophotometrically in terms of the absorbance change at 295 nm (ε = 9500 m−1 cm−1) in 50 mm sodium pyrophosphate buffer (pH 8.5) containing 0.2 mm EDTA and 0.15 mm xanthine under air-saturated conditions. The electron transfer activity from xanthine or allopurinol (0.15 mm) to 2,6-dichlorophenolindophenol (DCPIP; 50 μm) was determined spectrophotometrically by monitoring the absorbance of DCPIP at 600 nm (ε = 16,100m−1 cm−1) in 50 mmpotassium phosphate buffer (pH 7.8) under air. For XDH, all activity measurements were conducted in 50 mmsodium pyrophosphate buffer (pH 8.5) containing 0.2 mmEDTA. Xanthine-O2 activity was measured as described above for XO. Xanthine-NAD+ activity was measured with 0.15 mm xanthine and 0.5 mm NAD+. For the measurements of electron transfer activities between xanthine and DCPIP, NADH and DCPIP, NADH and ferricyanide, and NADH and methylene blue, 0.15 mm xanthine or NADH and 50 μmelectron acceptor (DCPIP, potassium ferricyanide, or methylene blue) were used. Each solution below was prepared with 0.2 m Hepes buffer containing 1 mm EDTA (pH 7.0). Anaerobic XO or XDH (40–80 μm) samples were prepared in an all-glass apparatus by sequential evacuation and re-equilibration with oxygen-free argon. Oxygen-free argon was prepared by passing commercially obtained pure argon through a column of Oxyout (Osaka Sanso). The solution of an electron donor such as xanthine, allopurinol, or NADH was sealed in a glass vial tube with a rubber cap and made anaerobic by bubbling oxygen-free argon. Saturated nitric oxide solution was prepared by bubbling NO gas, which was passed through 5 m KOH to remove NO2, into the anaerobic buffer. To prepare nitric oxide-treated enzymes, XO and XDH were first reduced by mixing enzyme solution into the electron donor solution anaerobically. The final concentrations of XO, XDH, and xanthine/allopurinol were 4–15, 10, and 150–300 μm, respectively. Then, saturated NO solution was added so that the final concentration became 250–500 μm. The mixture was drawn into a gas-tight syringe, incubated for 30 min at 25 °C, and filtered through a small column of Sephadex G-25. To check the effects of electron acceptors such as methylene blue (50 μm) and potassium ferricyanide (50 μm) during NO treatments, xanthine and the electron acceptor were mixed and made anaerobic before being mixed with enzyme solution. To examine the kinetics of the reaction between XDH/XO and nitric oxide, various amounts of saturated NO solution were added to enzyme solutions that had been pre-reduced anaerobically with xanthine or NADH (0.15 mm). The final concentrations of XO and XDH were 4 and 10 μm, respectively. The mixture was vigorously mixed and immediately drawn into a gas-tight syringe to avoid escape of NO into the gas phase and to maintain a constant NO concentration. The NO concentration of the reaction mixture in the syringe was immediately determined in duplicate by the oxyhemoglobin method. At various time points, 10-μl aliquots of the mixture were used to measure xanthine-O2 activity for XO and xanthine-NAD+reductase activity for XDH. XO (AFR > 200) or XDH (AFR = 109.00) in 0.2 m Hepes buffer (pH 7.4) containing 1 mm EDTA and 1 mm NADH was mixed with saturated NO solution, which was prepared by bubbling NO gas into the anaerobic buffer, under air at 25 °C. The enzyme concentration and initial NO concentration of the mixture were 5 and 470 μm, respectively. Control experiments were conducted under the same conditions by mixing with anaerobic 0.2 mmHepes buffer (pH 7.4) containing 1 mm EDTA instead of saturated NO solution. After having been incubated under air for 20 min, the mixture was gel-filtered, and aliquots were used for the determination of xanthine-oxygen or xanthine-NAD+ reductase activity. EPR measurements were conducted on a Jeol JES-FE2XG spectrometer. Sample temperature was controlled by a variable temperature controller above 77 K and by a continuous flow cryostat system (CT-470-esr-1, Research and Manufacturing Co., Inc.) below 77 K. The samples used to measure the Mo(V) signals of native XO, NO-treated XO, and desulfo-XO were prepared by reducing the different forms of XO based on reported procedures (32Malthouse J.P. George G.N. Lowe D.J. Bray R.C. Biochem. J. 1981; 199: 629-637Crossref PubMed Scopus (35) Google Scholar), and these EPR spectra were acquired at 123 K. To obtain the "very rapid" signal, enzymes (15 μm) in 20 mm Ches buffer (pH 10.2) were anaerobically reduced with 6-methylpurine (5 mm) for 1 min, and 0.5 ml of the solution was frozen 20 s after having been reoxidized by bubbling O2 through the solution. To obtain the "slow" or "rapid type 1" signal, enzymes (15 μm) in 20 mm Bicine buffer (pH 8.2) were made anaerobic by bubbling argon gas for 20 min, and 0.5 ml of the solution was frozen in an EPR sample tube 1 min after the addition of excess sodium dithionite. The EPR spectra of iron-sulfur centers (33Palmer G. Massey V. J. Biol. Chem. 1969; 244: 2614-2620Abstract Full Text PDF PubMed Google Scholar) were obtained using the same sample as prepared for the slow signal and measured at 22 K. The conditions of the EPR measurements were as follows: microwave frequency, 9.2 GHz; response, 0.3 s; microwave power, 10 milliwatts; field modulation width, 0.2 millitesla; and sweep time, 2 or 4 min. The magnetic field was calibrated with the signal of an external Mn2+ (in MgO) standard. At this microwave frequency, the apparent g values of the third and fourth Mn2+ signals are 2.033 and 1.981, respectively. EPR spectra were acquired through an A/D converter board and processed with an IBM-PC compatible computer. Both desulfo-XO (AFR = 0) and NO-treated XO (AFR = 4.6) (4.6 μm) were reactivated using sodium thiosulfate and rhodanese as previously reported (34Nishino T. Usami C. Tsushima K. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 1826-1829Crossref PubMed Scopus (33) Google Scholar). NO completely inactivated XO when XO was reduced with xanthine or allopurinol under anaerobic conditions prior to the NO treatment, although no marked decrease in the activity was observed when XO was not pre-reduced with xanthine or allopurinol, as shown in Table I. This is consistent with the previous report that XO was inactivated by NO during enzyme turnover of the xanthine oxidase reaction (19Fukahori M. Ichimori K. Ishida H. Nakagawa H. Okino H. Free Radical Res. 1994; 21: 203-212Crossref PubMed Scopus (71) Google Scholar). The fact that the activity of NO-treated XO did not recover upon gel filtration with Sephadex G-25 indicates that the inhibition is an irreversible process. In the presence of an electron acceptor such as methylene blue (50 μm) or potassium ferricyanide (50 μm), XO was not markedly inhibited by NO even in the presence of xanthine (data not shown), in accordance with the result that only reduced XO can be inhibited by NO. As NO-treated XO lost both xanthine-O2 and xanthine-DCPIP activities, it is likely that NO reacted with the molybdenum center since DCPIP can directly accept an electron from reduced molybdenum (35Ikegami T. Nishino T. Arch. Biochem. Biophys. 1986; 247: 254-260Crossref PubMed Scopus (42) Google Scholar). It should be noted that the extent of inhibition of XO pre-reduced with allopurinol is lower than that of XO pre-reduced with xanthine, although NO can also inhibit XO pre-reduced with allopurinol. This result is consistent with the conclusion that NO attacks XO at the molybdenum center since a small amount of oxypurinol, the oxidative product of allopurinol, might partially form a tight-binding complex with the reduced molybdenum (24Massey V. Komai H. Palmer G. Elion G.B. J. Biol. Chem. 1970; 245: 2837-2844Abstract Full Text PDF PubMed Google Scholar) to protect the enzyme from inactivation by NO.Table IInactivation of XO by NO under anaerobic conditionsConditionsActivityXanthine-O2Xanthine-DCPIPAllopurinol-DCPIP%Control100100100+NO77NDaND, not determined.ND+XanthinebXO was reduced by these substrates (150 μm) before the addition of NO under anaerobic conditions as described under "Experimental Procedures." + NO03.85+AllopurinolbXO was reduced by these substrates (150 μm) before the addition of NO under anaerobic conditions as described under "Experimental Procedures."+ NO0cActivity determination was performed without reactivation with ferricyanide.2324XO was treated anaerobically with 250 μm NO for 30 min at 25 °C, followed by gel filtration as described under "Experimental Procedures." The activities (measured in duplicate), are based on the rates of electron transfer from xanthine or allopurinol to O2or DCPIP and are expressed as mean percentage of the control (before treatment).a ND, not determined.b XO was reduced by these substrates (150 μm) before the addition of NO under anaerobic conditions as described under "Experimental Procedures."c Activity determination was performed without reactivation with ferricyanide. Open table in a new tab XO was treated anaerobically with 250 μm NO for 30 min at 25 °C, followed by gel filtration as described under "Experimental Procedures." The activities (measured in duplicate), are based on the rates of electron transfer from xanthine or allopurinol to O2or DCPIP and are expressed as mean percentage of the control (before treatment). To clarify further the characteristics of the reaction between XO and NO, the reaction kinetics were determined. XO (4 μm) was pre-reduced with xanthine (150 μm) and treated with various concentrations of nitric oxide anaerobically at 25 °C, and the time course of the xanthine-oxygen reductase activity was determined. As shown in Fig.1 A, the XO activity was inhibited dose-dependently by NO. We presumed that XO reacts with NO according to second-order kinetics as follows (Equations1 and 2), XO+NO→kXO*+productsEquation 1 d[XO]/dt=−k[XO][NO]Equation 2 where XO denotes the active form, XO* denotes the inactive form, and k is the second-order rate constant for the reaction. When [XO] ≪ [NO], k′ = k[NO] can be assumed to be constant, and the reaction should follow pseudo first-order kinetics. Since the time courses of enzyme inhibition depicted in Fig. 1 A showed a single exponential form, a pseudo first-order rate constant (k′) could be determined at various NO concentrations. Fig. 1 B shows the dependence ofk′ on nitric oxide concentration. As expected, k′ was proportional to nitric oxide concentration, indicating that the reaction of XO with nitric oxide follows second-order kinetics as formulated in Equations 1 and 2. The second-order rate constant (k) obtained from the slope in Fig. 1 B is 14.8 ± 1.4 m−1 s−1(25 °C). The initial spectrum obtained from the fully oxidized form of NO-inactivated XO is very similar to that of the native enzyme (Fig.2). Since the absorption in the visible region is mainly attributable to the flavin and iron-sulfur centers, it is suggested that these cofactors remain intact. Upon addition of xanthine to the inactivated enzyme under anaerobic conditions, negligible immediate bleaching of the visible absorbance at around 450 nm was observed, but gradual bleaching was observed after prolonged incubation; only <20% reduction occurred even after 4 h. Such a biphasic reduction mode is very similar to that of the desulfo-type inactive enzyme (25Massey V. Edmondson D. J. Biol. Chem. 1970; 245: 6595-6598Abstract Full Text PDF PubMed Google Scholar). The second phase of reduction might be due to slow reduction of the inactive enzyme by a small amount of the active enzyme remaining in the sample. Although bleaching of the visible absorbance at around 450 nm of the NO-inactivated enzyme was very slow, it occurred readily upon addition of dithionite as in the case of the native enzyme. As shown in Fig. 2, the flavin and iron-sulfur centers in NO-treated XO can be reduced with dithionite. Moreover, the fully reduced inactive enzyme was reoxidized readily when exposed to air, and the spectrum returned to that of the initial oxidized form (data not shown), indicating that the redox function of the flavin and iron-sulfur centers in NO-treated XO was preserved. Fig.3 shows the EPR spectra of iron-sulfur centers of dithionite-reduced XO before and after treatment with NO. It has been shown that the fully reduced active enzyme exhibits EPR signals due to two distinct Fe2S2 centers (33Palmer G. Massey V. J. Biol. Chem. 1969; 244: 2614-2620Abstract Full Text PDF PubMed Google Scholar). Native and NO-treated XO showed essentially the same signals due to the iron-sulfur centers (gav = 1.95 and 2.01, respectively), and there was no indication that NO interacted with the iron-sulfur center of NO-treated XO. These spectroscopic observations suggested that NO reacted with neither flavins nor iron-sulfur centers, but rather with the molybdenum center.Figure 3EPR spectra of iron-sulfur centers of XO reduced with dithionite. The EPR spectra of dithionite-reduced iron-sulfur centers were obtained at 22 K for native XO (trace a) and NO-treated XO (trace b) (15 μm) in 20 mm Bicine buffer (pH 8.2) under the conditions described under "Experimental Procedures." mT, milliteslas.View Large Image Figure ViewerDownload (PPT) All of the above results indicate that the inactivation of XO by nitric oxide occurs at the molybdenum center. To confirm this conclusion and to clarify what kind of change occurs at the molybdenum center, the EPR spectrum of NO-treated XO was measured and compared with those of the native and desulfo-type enzymes. First, the EPR spectrum of NO-treated XO was observed without treatment with reducing reagents. Although XO in the resting state, which has Mo(VI) and two antiferromagnetically coupled Fe(III)2S2 clusters, is EPR-silent, it can be paramagnetic if the NO adduct of XO forms at either iron center to produce Fe(II)-nitrosyl heme complexes. However, no paramagnetic signal other than a weak signal due to a non-heme iron impurity at g = 4.3 was observed (data not shown). Thus, no NO adduct was observable in the oxidized state of NO-treated XO. Fig.4 shows the EPR spectra of native and NO-treated XO reduced under conditions where the very rapid signal can be obtained. As shown in Fig. 4 (trace a), native XO showed a Mo(V) very rapid signal that has large g anisotropy when reduced with 6-methylpurine for a short time as reported previously (32Malthouse J.P. George G.N. Lowe D.J. Bray R.C. Biochem. J. 1981; 199: 629-637Crossref PubMed Scopus (35) Google Scholar). The species that gives the very rapid signal has a very short lifetime if xanthine is used as a substrate and corresponds to a transient intermediate Mo(V)-substrate complex in enzyme turnover (36Bennett B. Benson N. McEwan A.G. Bray R.C. Biochem. Soc. Trans. 1994; 22 (suppl.): 285sCrossref Scopus (6) Google Scholar). Thus, the observation of the very rapid signal provides evidence of reaction between molybdopterin and the substrate. However, as shown in Fig.4 B, no EPR signal was observed for the NO-treated enzyme under the same conditions, indicating that molybdopterin in NO-treated XO has no ability to interact with purine derivatives and that NO damages the molybdopterin. Fig. 5 shows Mo(V) EPR spectra of native XO (trace a), NO-treated XO (trace b), and CN-treated desulfo-XO (trace c) (15 μm) in 20 mm frozen Bicine buffer (pH 8.2) reduced with dithionite for 1 min. The native enzyme exhibits a rapid type 1 signal (32Malthouse J.P. George G.N. Lowe D.J. Bray R.C. Biochem. J. 1981; 199: 629-637Crossref PubMed Scopus (35) Google Scholar), as shown in Fig. 5 (trace a). On the other hand, NO-treated XO showed a different EPR spectrum (Fig. 5, trace b). The spectrum showed no evidence of an extra hyperfine interaction of the molybdenum center with NO nitrogen, which should split the signal into three lines. Thus, the molybdopterin in NO-treated XO had not been changed into its NO adduct. Furthermore, NO-treated XO showed a spectrum substantially identical to that of the CN-treated desulfoenzyme (Fig. 5, trace c), having similar apparent g values to those reported previously (32Malthouse J.P. George G.N. Lowe D.J. Bray R.C. Biochem. J. 1981; 199: 629-637Crossref PubMed Scopus (35) Google Scholar). It is known that reduced desulfo-type xanthine oxidase, which is inactive because an essential sulfur atom coordinated to the molybdenum is replaced by an oxygen atom, shows the slow signal due to Mo(V), which is stable under anaerobic conditions (32Malthouse J.P. George G.N. Lowe D.J. Bray R.C. Biochem. J. 1981; 199: 629-637Crossref PubMed Scopus (35) Google Scholar). The Mo(V) in reduced NO-treated XO was also stable under anaerobic conditions. Thus, it is suggested that nitric oxide reacts with an essential sulfur in molybdopterin and removes it. It has been demonstrated that desulfo-XO undergoes reactivation by a sulfide-generating system containing rhodanese, thiosulfate, and sulfhydryl reagent up to half its maximal activity (34Nishino T. Usami C. Tsushima K. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 1826-1829Crossref PubMed Scopus (33) Google Scholar). To determine whether NO-treated XO can recover its activity by incubation with a sulfide-generating system, the change in the activity of NO-treated XO was determined over 3 h, and the time course was compared with that of the CN-treated desulfoenzyme (Fig.6). In the presence of sulfide, NO-treated XO was also reactivated, and its time course was almost the same as that of CN-treated desulfo-XO. Thus, it was confirmed that NO-treated XO has the same reactivity to sulfide as CN-treated desulfo-XO. The recovery of NO-treated XO was slightly low compared with that of the CN-treated desulfoenzyme, presumably because NO treatment caused minor inactivation via a route other than conversion to desulfo-XO. Since XDH has the same molybdopterin as XO, XDH is also expected to be inhibited by NO. As shown in Fig. 7, XDH (10 μm) reduced with xanthine (0.15 mm) under anaerobic conditions was inhibited by NO in the same manner as XO, whereas the oxidized enzyme was not markedly inhibited. Without xanthine, XDH was not inhibited by NO either. Moreover, XDH reduced with excess NADH was also inhibited by NO. Since the NADH-DCPIP, NADH-ferricyanide, and NADH-methylene blue activities of NO-treated XDH were almost equivalent to those of the native enzyme, as also found for CN-treated XDH (data not shown), it is concluded that FAD remained intact. Thus, it is concluded that reduced XDH is inhibited by NO in the same manner as XO. In the presence of enough NADH, XDH is kept in the reduced form even under aerobic conditions. So, it is possible that XDH is inhibited by NO in air. In the presence of 1 mm NADH, XO (5 μm) or XDH (5 μm) was mixed with saturated NO solution so that the initial NO concentration was 470 μm, and the catalytic activities were measured after 20 min. With NO treatment in air, XO activity was not markedly decreased (93.8 ± 2.8% of untreated XO), whereas XDH activity was almost completely inhibited (13 ± 0.8% of untreated XDH). Thus, as long as enough NADH is present, XDH is susceptible to the attack of NO even under aerobic conditions, where XO is not inhibited. XO contains multiple redox centers that are essential for enzyme activity and are plausible targets of NO attack. We examined the effect of NO on each redox center using enzyme kinetics and spectroscopic studies. Since NO forms an iron-nitrosyl adduct with iron-sulfur complexes (12Gardner P.R. Costantino G. Szabo C. Salzman A.L. J. Biol. Chem. 1997; 272: 25071-25076Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar), as was demonstrated in aconitase (37Henry Y. Lepoivre M. Drapier J.C. Ducrocq C. Boucher J.L. Guissani A. FASEB J. 1993; 7: 1124-1134Crossref PubMed Scopus (352) Google Scholar), we first examined the iron-sulfur centers by means of absorption spectroscopy and EPR. In NO-treated XO, the iron-sulfur centers were redox-active, and EPR proved that their environment remained unchanged; no nitrosyl-iron-sulfur complex was observed. Since aconitase has an Fe4S4-type cluster with at least one iron coordination with a solvent molecule, it may be more susceptible to NO attack than the Fe2S2-type clusters with complete cysteinyl ligation in XO. This is consistent with the fact that most of the reports on the NO sensitivity of iron-sulfur proteins refer to Fe4S4 clusters (12Gardner P.R. Costantino G. Szabo C. Salzman A.L. J. Biol. Chem. 1997; 272: 25071-25076Abstract Full Text Full Text PDF PubMed Scopus (210) Google Scholar, 37Henry Y. Lepoivre M. Drapier J.C. Ducrocq C. Boucher J.L. Guissani A. FASEB J. 1993; 7: 1124-1134Crossref PubMed Scopus (352) Google Scholar), except for a report on nitrosyl-iron complex formation from Fe2S2 ferredoxins (38Drapier J.C. Pellat C. Henry Y. J. Biol. Chem. 1991; 266: 10162-10167Abstract Full Text PDF PubMed Google Scholar). Since it seemed that FAD in NO-treated XO was intact, the molybdenum center was considered the next most likely candidate for NO attack. The lack of electron transfer activity from xanthine to DCPIP and the very slow reduction of NO-treated XO by xanthine suggested that the molybdenum center was damaged. The EPR measurements indicated that NO reacts with an essential sulfur atom coordinated to the molybdenum center to afford a desulfo-type enzyme. This was confirmed by the fact that the NO-inactivated enzyme was reactivated by the sulfide-generating system, like CN-treated desulfo-XO. Since NO can react only with the reduced enzyme, i.e. ≥Mo(IV)–SH, not with the oxidized form, Mo(VI)=S (1Hille R. Nishino T. FASEB J. 1995; 9: 995-1003Crossref PubMed Scopus (383) Google Scholar), it seems that sulfhydryl coordinated to molybdenum is more susceptible to NO attack than Mo=S. Although the chemical mechanism of conversion of the sulfo to desulfo form by NO is not clear, probably cationic NO might attack the reduced sulfide to release SNO. It is interesting to note that the desulfo form of XO exists in a significant amount in rat liver cells (35Ikegami T. Nishino T. Arch. Biochem. Biophys. 1986; 247: 254-260Crossref PubMed Scopus (42) Google Scholar). The fact that NO reacts only with reduced XO explains why XO is not markedly inhibited by NO exposure under aerobic conditions. The moderate inhibition during enzyme turnover may reflect a comparatively low steady-state level of the reduced enzyme. However, since XO operates under a much lower oxygen tension in vivo, close to 10 mm Hg (39Staub N.C. Berne R.M. Levy M.N. Physiology. 4th Ed. Mosby, New York1998: 561-571Google Scholar), this restriction does not reduce the potential importance of this NO inhibition. Furthermore, the reactivity of NO toward XO in a reducing environment rather reinforces the importance of this inhibition under hypoxic conditions, where more reducing substrates such as hypoxanthine, xanthine, and NADH might be accumulated. Our finding that, under aerobic conditions, only XDH was inactivated in the presence of a reducing substrate, NADH, suggests that XDH, but not XO, may be inactivated by NO under conditions such as post-ischemic reperfusion, where oxygen is resupplied. Although XDH can produce a large amount of O·̄2 without NAD+ during xanthine-O2 turnover, this pathway should be inhibited since an excess amount of NAD+ is present in the normal cell (40Nishino T. Nishino T. Schopfer L.M. Massey V. J. Biol. Chem. 1989; 264: 2518-2527Abstract Full Text PDF PubMed Google Scholar). Therefore, the conversion of XDH to XO seems to be necessary to explain the increased production of O·̄2 in so-called superoxide-induced injury during ischemia/reperfusion. However, it is still controversial whether conversion from XDH to XO occurs under post-ischemic conditions as postulated by McCord (6McCord J.M. N. Engl. J. Med. 1985; 312: 159-163Crossref PubMed Scopus (4999) Google Scholar). The O·̄2generation, however, was suggested to increase when the apparent ratio of XDH to XO changes. That is, as XDH is inhibited by the accumulated NADH during ischemia, accumulated hypoxanthine is utilized more by XO than would be the case under normal conditions, and increased superoxide production may occur upon reperfusion even though conversion from XDH to XO does not take place (41Nishino T. Nakanishi S. Okamoto K. Mizushima J. Hori H. Iwasaki T. Nishino T. Ichimori K. Nakazawa H. Biochem. Soc. Trans. 1997; 25: 783-786Crossref PubMed Scopus (51) Google Scholar, 42Nishino T. Tamura I. Adv. Exp. Med. Biol. 1991; 309: 327-337Crossref Scopus (46) Google Scholar). The increased O·̄2generation can be alternatively explained in terms of NADH oxidation by XDH (42Nishino T. Tamura I. Adv. Exp. Med. Biol. 1991; 309: 327-337Crossref Scopus (46) Google Scholar, 43Harrison R. Biochem. Soc. Trans. 1997; 25: 786-791Crossref PubMed Scopus (41) Google Scholar). The present results suggest another mechanism for change in the relative contributions of XO and XDH to O·̄2 generation if selective inactivation of XDH by NO occurs under post-ischemic conditions prior to the formation enough O·̄2. If once O·̄2 is generated in a large amount, which can react with NO with diffusion-limited rates to give peroxynitrite, decreased NO and formed peroxynitrite may make the situation worse for the tissue. Thus, although the generation of NO upon reperfusion was suggested to down-regulate O·̄2 generation through inhibition of XO by NO (21Hassoun P.M. Yu F.S. Zulueta J.J. White A.C. Lanzillo J.J. Am. J. Physiol. 1995; 268: L809-L817PubMed Google Scholar), it is clear from the present experiments that the idea that NO serves as an antioxidant effector by suppressing XO activity in actual cellular systems is an oversimplification. It should also be noted that O·̄2 formation due to NADH oxidation, in which the molybdenum center does not participate, cannot be inhibited by NO. We thank Dr. Vincent Massey (University of Michigan) for valuable discussions.