Post-translational Regulation of Human Indoleamine 2,3-Dioxygenase Activity by Nitric Oxide
2007; Elsevier BV; Volume: 282; Issue: 33 Linguagem: Inglês
10.1074/jbc.m700669200
ISSN1083-351X
AutoresShane R. Thomas, Andrew C. Terentis, Hong Cai, Osamu Takikawa, Aviva Levina, Peter A. Lay, Mohammed Freewan, Roland Stocker,
Tópico(s)Stress Responses and Cortisol
ResumoThe heme protein indoleamine 2,3-dioxygenase (IDO) is induced by the proinflammatory cytokine interferon-γ (IFNγ) and plays an important role in the immune response by catalyzing the oxidative degradation of l-tryptophan (Trp) that contributes to immune suppression and tolerance. Here we examined the mechanism by which nitric oxide (NO) inhibits human IDO activity. Exposure of IFNγ-stimulated human monocyte-derived macrophages (MDM) to NO donors had no material impact on IDO mRNA or protein expression, yet exposure of MDM or transfected COS-7 cells expressing active human IDO to NO donors resulted in reversible inhibition of IDO activity. NO also inhibited the activity of purified recombinant human IDO (rhIDO) in a reversible manner and this correlated with NO binding to the heme of rhIDO. Optical absorption and resonance Raman spectroscopy identified NO-inactivated rhIDO as a ferrous iron (FeII)-NO-Trp adduct. Stopped-flow kinetic studies revealed that NO reacted most rapidly with FeII rhIDO in the presence of Trp. These findings demonstrate that NO inhibits rhIDO activity reversibly by binding to the active site heme to trap the enzyme as an inactive nitrosyl-FeII enzyme adduct with Trp bound and O2 displaced. Reversible inhibition by NO may represent an important mechanism in controlling the immune regulatory actions of IDO. The heme protein indoleamine 2,3-dioxygenase (IDO) is induced by the proinflammatory cytokine interferon-γ (IFNγ) and plays an important role in the immune response by catalyzing the oxidative degradation of l-tryptophan (Trp) that contributes to immune suppression and tolerance. Here we examined the mechanism by which nitric oxide (NO) inhibits human IDO activity. Exposure of IFNγ-stimulated human monocyte-derived macrophages (MDM) to NO donors had no material impact on IDO mRNA or protein expression, yet exposure of MDM or transfected COS-7 cells expressing active human IDO to NO donors resulted in reversible inhibition of IDO activity. NO also inhibited the activity of purified recombinant human IDO (rhIDO) in a reversible manner and this correlated with NO binding to the heme of rhIDO. Optical absorption and resonance Raman spectroscopy identified NO-inactivated rhIDO as a ferrous iron (FeII)-NO-Trp adduct. Stopped-flow kinetic studies revealed that NO reacted most rapidly with FeII rhIDO in the presence of Trp. These findings demonstrate that NO inhibits rhIDO activity reversibly by binding to the active site heme to trap the enzyme as an inactive nitrosyl-FeII enzyme adduct with Trp bound and O2 displaced. Reversible inhibition by NO may represent an important mechanism in controlling the immune regulatory actions of IDO. Indoleamine 2,3-dioxygenase (IDO) 4The abbreviations used are: IDO, indoleamine 2,3-dioxygenase; DEANO, diethylamine NONOate; FeII, ferrous iron; FeIII, ferric iron; GSNO, glutathione nitric oxide adduct; HPLC, high pressure liquid chromatography; IFNγ, interferon-γ; l-Trp, l-tryptophan; MDM, monocyte-derived macrophage; NOS2, inducible isoform of nitric-oxide synthase; rhIDO, recombinant human IDO; RT, reverse transcription; SNAP, S-nitroso-N-acetylpenicillamine; TDO, tryptophan 2,3-dioxygenase. is an intracellular heme enzyme that catalyzes the initial and rate-limiting step of l-tryptophan (l-Trp) metabolism along the kynurenine pathway (reviewed in Refs. 1Taylor M.W. Feng G.S. FASEB J. 1991; 5: 2516-2522Crossref PubMed Scopus (933) Google Scholar, 2Thomas S.R. Stocker R. Redox Rep. 1999; 4: 199-220Crossref PubMed Scopus (171) Google Scholar, 3Mellor A.L. Munn D.H. Nat Rev. Immunol. 2004; 4: 762-774Crossref PubMed Scopus (1861) Google Scholar, 4Takikawa O. Biochem. Biophys. Res. Commun. 2005; 338: 12-19Crossref PubMed Scopus (267) Google Scholar). IDO catalyzes the oxidative cleavage of the pyrrole ring of l-Trp by insertion of molecular oxygen (O2) to generate N-formyl-kynurenine, which is hydrolyzed to kynurenine and formate (Fig. 1). Expression of IDO is induced at sites of inflammation in vivo principally by the cytokine interferon-γ (IFNγ). Such induction of IDO is considered part of the host innate immune response and is thought to prevent the growth of certain viruses, bacteria, intracellular pathogens, and tumor cells via local depletion of l-Trp, the least abundant of all essential amino acids (1Taylor M.W. Feng G.S. FASEB J. 1991; 5: 2516-2522Crossref PubMed Scopus (933) Google Scholar, 2Thomas S.R. Stocker R. Redox Rep. 1999; 4: 199-220Crossref PubMed Scopus (171) Google Scholar, 3Mellor A.L. Munn D.H. Nat Rev. Immunol. 2004; 4: 762-774Crossref PubMed Scopus (1861) Google Scholar). Recent in vivo studies have also established a role for IDO in immune regulation by promoting immune tolerance via suppression of local T cell responses under various physiological and pathophysiological conditions including mammalian pregnancy, tumor resistance, autoimmunity, chronic inflammation, and chronic infections (3Mellor A.L. Munn D.H. Nat Rev. Immunol. 2004; 4: 762-774Crossref PubMed Scopus (1861) Google Scholar). In light of the increasing awareness of the important roles of the enzyme it is critical to understand how IDO activity is regulated. In the majority of cells, IFNγ induces IDO expression at the transcriptional level. This requires the cooperative action of signal transducers and activators of transcription 1 and interferon regulatory factor 1 transcription factors (5Chon S.Y. Hassanain H.H. Gupta S.L. J. Biol. Chem. 1996; 271: 17247-17252Abstract Full Text Full Text PDF PubMed Scopus (145) Google Scholar, 6Konan K.V. Taylor M.W. J. Biol. Chem. 1996; 271: 19140-19145Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). A role for nuclear factor κB in IDO expression has been reported for other stimuli, including cytotoxic T lymphocyte-associated antigen 4 or lipopolysaccharide (7Muller A.J. DuHadaway J.B. Donover P.S. Sutanto-Ward E. Prendergast G.C. Nat. Med. 2005; 11: 312-319Crossref PubMed Scopus (914) Google Scholar, 8Orabona C. Belladonna M.L. Vacca C. Bianchi R. Fallarino F. Volpi C. Gizzi S. Fioretti M.C. Grohmann U. Puccetti P. J. Immunol. 2005; 174: 6582-6586Crossref PubMed Scopus (84) Google Scholar). IDO transcription may also be suppressed by certain proteins, including suppressor of cytokine signaling 3 and BAR adapter-encoding protein (7Muller A.J. DuHadaway J.B. Donover P.S. Sutanto-Ward E. Prendergast G.C. Nat. Med. 2005; 11: 312-319Crossref PubMed Scopus (914) Google Scholar, 9Grohmann U. Orabona C. Fallarino F. Vacca C. Calcinaro F. Falorni A. Candeloro P. Belladonna M.L. Bianchi R. Fioretti M.C. Puccetti P. Nat. Immunol. 2002; 3: 1097-1101Crossref PubMed Scopus (1010) Google Scholar). In addition to transcriptional regulation, emerging evidence indicates that IDO is also subject to post-translational control. For example, in IFNγ-stimulated human macrophages, IDO activity is controlled via the supply and insertion of heme, as well as via changes in the intracellular reduction and oxidation (redox) status (10Thomas S.R. Salahifar H. Mashima R. Hunt N.H. Richardson D.R. Stocker R. J. Immunol. 2001; 166: 6332-6340Crossref PubMed Scopus (106) Google Scholar). Also, in populations of dendritic cells expressing inactive protein, IDO activity can be evoked by certain cytokines (11Munn D.H. Sharma M.D. Lee J.R. Jhaver K.G. Johnson T.S. Keskin D.B. Marshall B. Chandler P. Antonia S.J. Burgess R. Slingluff C.L. Mellor Jr., A.L. Science. 2002; 297: 1867-1870Crossref PubMed Scopus (888) Google Scholar, 12Fallarino F. Vacca C. Orabona C. Belladonna M.L. Bianchi R. Marshall B. Keskin D.B. Mellor A.L. Fioretti M.C. Grohmann U. Puccetti P. Int. Immunol. 2002; 14: 65-68Crossref PubMed Scopus (219) Google Scholar). These studies suggest that post-translational control is important for regulating IDO activity in antigen-presenting cells, although the mechanisms involved remain largely unknown. Similar to the situation with IDO, proinflammatory cytokines, including IFNγ, also up-regulate the expression of inducible nitric-oxide synthase (NOS2), which produces large amounts of nitric oxide (NO) (13Bogdan C. Nat. Immunol. 2001; 2: 907-916Crossref PubMed Scopus (2606) Google Scholar). Also similar to IDO, NOS2 represents an important component of the innate immune response (13Bogdan C. Nat. Immunol. 2001; 2: 907-916Crossref PubMed Scopus (2606) Google Scholar), and there is evidence that the NOS2 and IDO pathways are interrelated (2Thomas S.R. Stocker R. Redox Rep. 1999; 4: 199-220Crossref PubMed Scopus (171) Google Scholar). One aspect of this proposed "interaction" is that the product(s) of one pathway regulates the enzyme of the other pathway. Specifically, several studies have reported NO to inhibit IDO activity (14Thomas S.R. Mohr D. Stocker R. J. Biol. Chem. 1994; 269: 14457-14464Abstract Full Text PDF PubMed Google Scholar, 15Alberati-Giani D. Malherbe P. Ricciardi-Castagnoli P. Kohler C. Denis-Donini S. Cesura A.M. J. Immunol. 1997; 159: 419-426PubMed Google Scholar, 16Hucke C. MacKenzie C.R. Adjogble K.D. Takikawa O. Daubener W. Infect. Immun. 2004; 72: 2723-2730Crossref PubMed Scopus (99) Google Scholar), although the underlying mechanism of this regulation remains unclear. The present study set out to understand how NO inhibits human IDO activity. Materials—Reagent peroxynitrite and the NO donors S-nitroso-N-acetylpenicillamine (SNAP), glutathione nitric oxide adduct (GSNO), diethylamine NONOate (DEANO), and spermine NONOate were obtained from Cayman Chemicals. Chemically pure grade NO gas was purchased from Airgas. Solutions of authentic NO were prepared as described previously (14Thomas S.R. Mohr D. Stocker R. J. Biol. Chem. 1994; 269: 14457-14464Abstract Full Text PDF PubMed Google Scholar). Recombinant human IFNγ was from R&D Systems. PD10 gel-filtration columns were obtained from Amersham Biosciences. Unless indicated otherwise, all other materials were purchased from Sigma-Aldrich and were of the highest purity available. Cell Culture and IDO Transfection—Monocytes were isolated from human blood buffy coats (Australian Red Cross Blood Bank) and matured into monocyte-derived macrophages (MDM) by 8–12 days of culture in RPMI 1640 medium supplemented with 10% pooled human serum (10Thomas S.R. Salahifar H. Mashima R. Hunt N.H. Richardson D.R. Stocker R. J. Immunol. 2001; 166: 6332-6340Crossref PubMed Scopus (106) Google Scholar). Upon maturation, MDM were treated with recombinant human IFNγ (500 units/ml) to induce IDO expression and activity. COS-7 cells (from ATCC) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum plus penicillin and streptomycin. For transient transfection, COS-7 cells were seeded in antibiotic-free medium in 6-well tissue culture plates and grown to 90% confluence. Cells were then cultured in serum-free Opti-MEM medium and transfected with pcDNA3 encoding full-length human IDO cDNA (1 μg/well) or pcDNA3 (empty vector control, 1 μg/well) using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. After overnight transfection, COS-7 cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and used for experiments. For all experiments, MDM or COS-7 cells were cultured at 37 °C in a humidified atmosphere of 5% CO2 and in culture media supplemented with 200 μm l-Trp. RT-PCR—RNA was extracted from cultured MDM with TRIzol reagent (Invitrogen). Reverse transcription of 1 μgof RNA was performed with a commercial kit (Invitrogen) according to the manufacturer's instructions. Conventional PCR conditions included a 10-min hot start at 95 °C followed by 35 cycles of denaturation (95 °C, 40 s), annealing (63 °C, 30 s), and extension (72 °C, 30 s). Real-time PCR reactions were performed in a total volume of 20 μl containing 4 μl of cDNA sample and 16 μl of SYBR Green PCR Master Mix (Quantace SensiMix) and specific primers (final concentration of 250 nm). Cycling conditions were 10 min hot start at 95 °C followed by 40 cycles of a denaturation step at 95 °C for 40 s and an annealing step at 60 °C for 60 s. Human IDO-specific sequences were amplified using a primer set described previously (17Zegarra-Moran O. Folli C. Manzari B. Ravazzolo R. Varesio L. Galietta L.J. J. Immunol. 2004; 173: 542-549Crossref PubMed Scopus (21) Google Scholar): fwd 5′-CAAAGGTCATGGAGATGTCC-3′, rev 5′-CCACCAATAGAGAGACCAGG-3′. This primer set produced a 240-base pair PCR product. Human β-actin was used as a housekeeping gene to normalize IDO transcript abundance. PCR primers for β-actin were: fwd 5′-CTGGAACGGTGAAGGTGACA-3′, rev 5′-CGGCCACATTGTGAACTTTG-3′. Real-time PCR was performed using a Corbett RotoGene 3000, and data were analyzed using Rotor Gene version 6.0 software (Corbett Research). Western Blotting and Cellular IDO Activity—Western blotting was performed as described previously using a mouse monoclonal antibody directed against human IDO (10Thomas S.R. Salahifar H. Mashima R. Hunt N.H. Richardson D.R. Stocker R. J. Immunol. 2001; 166: 6332-6340Crossref PubMed Scopus (106) Google Scholar). The presence of 3-nitrotyrosine in IDO was assessed by Western blotting using a mouse monoclonal anti-3-nitrotyrosine antibody (Clone 1A6, Upstate Biotechnology). Cellular IDO activity was assessed by measuring the extent to which l-Trp in the culture medium was converted to kynurenine. l-Trp, kynurenine and 3-hydroxyanthranilic acid were measured by HPLC as described previously (18Christen S. Thomas S.R. Garner B. Stocker R. J. Clin. Investig. 1994; 93: 2149-2158Crossref PubMed Scopus (85) Google Scholar, 19Sanni L.A. Thomas S.R. Tattam B.N. Moore D.E. Chaudhri G. Stocker R. Hunt N.H. Am. J. Pathol. 1998; 152: 611-619PubMed Google Scholar). Recombinant Human IDO—Recombinant human IDO (rhIDO) encoded by the pQE9-IDO plasmid vector was expressed in Escherichia coli as a fusion protein to a hexahistidyl tag and purified as described in detail (20Littlejohn T.K. Takikawa O. Skylas D. Jamie J.F. Walker M.J. Truscott R.J. Protein Expression Purif. 2000; 19: 22-29Crossref PubMed Scopus (75) Google Scholar, 21Terentis A.C. Thomas S.R. Takikawa O. Littlejohn T.K. Truscott R.J. Armstrong R.S. Yeh S.R. Stocker R. J. Biol. Chem. 2002; 277: 15788-15794Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar). Different batches of purified rhIDO used for the present studies appeared as a single major protein band at ∼42 kDa following SDS-PAGE and Coomassie Blue staining (not shown) and exhibited 404–280 nm absorption ratios of 1.5–1.8, with specific activities of ∼70–100 mol kynurenine formed/min/mol enzyme. IDO activity was determined routinely by the ascorbate/methylene blue assay (10Thomas S.R. Salahifar H. Mashima R. Hunt N.H. Richardson D.R. Stocker R. J. Immunol. 2001; 166: 6332-6340Crossref PubMed Scopus (106) Google Scholar, 14Thomas S.R. Mohr D. Stocker R. J. Biol. Chem. 1994; 269: 14457-14464Abstract Full Text PDF PubMed Google Scholar, 22Sono M. J. Biol. Chem. 1989; 264: 1616-1622Abstract Full Text PDF PubMed Google Scholar). UV-visible Spectroscopy—Optical absorption spectra of rhIDO (5–10 μm in heme) were recorded under anaerobic or aerobic conditions in 100 mm potassium phosphate buffer (pH 7.0) with 500 μm EDTA using a PerkinElmer Lambda EZ210 spectrophotometer. When added, the final concentration of l-Trp was 4 mm. When necessary, solutions were made anaerobic by purging with argon in septum-sealed quartz cuvettes (1-cm path length). Ferrous iron (FeII) rhIDO was formed by the addition of a molar excess of a buffered sodium dithionite solution. NO was added by direct addition of the purified gas using a gas-tight syringe. All experiments examining catalytically active rhIDO and its inhibition by NO were performed under normal atmospheric aerobic conditions using a Varian-Cary 300 spectrophotometer in quartz cuvettes (1-cm path length) containing rhIDO dissolved in the ascorbate/methylene blue assay buffer (50 mm KPO4 buffer (pH 7.0) with 500 μm EDTA, 50 μg of catalase, 400–800 μm l-Trp, 25 μm methylene blue, 10 mm ascorbate). For these experiments, NO was added either as a bolus from a 2 mm saturated NO gas solution or delivered by DEANO. Stopped-flow Kinetic Studies—Kinetic data for the reactions of NO with rhIDO were obtained using an Applied Photophysics SX-17 MV stopped-flow spectrophotometer at 22 °C. Solutions of dithionite-reduced FeII rhIDO (10 μm in heme) in the absence or presence of l-Trp (2 mm final concentration) were mixed rapidly with an anaerobic solution of NO gas (1 mm final concentration), and the initial rate of decay of the γ-heme Soret absorption intensity of FeII rhIDO was monitored at 428 nm. The concentration of NO was in large molar excess of rhIDO to ensure pseudo first-order conditions. The observed initial rate of reaction of NO with rhIDO (kobs) was calculated by fitting the stopped-flow kinetic data to a single exponential function. Resonance Raman Spectroscopy—Samples of rhIDO (∼100 μl, 20–30 μm in heme) were prepared in a septum-sealed, cylindrical quartz cell that was rotated at ∼1000 rpm. The sample cell was irradiated at 413.1 nm (∼3–5 milliwatts) using a mixed krypton/argon ion laser (Spectra Physics, Beamlok 2060). The spectral acquisition time was typically 1–5 min. The scattered light was collected at right angles to the incident beam and focused on the entrance slit (125 μm) of a 0.8-m spectrometer where it was dispersed by a 600 groove/mm grating and then detected by a liquid-N2-cooled charge-coupled device camera (Horiba-JY). Spectral calibration was performed against the lines of mercury and indene. NO Inhibits IDO Activity in Intact Cells—We reported previously that cultured human MDM do not contain measurable IDO enzyme activity, although treatment of these cells with IFNγ induces such activity in a time-dependent manner (10Thomas S.R. Salahifar H. Mashima R. Hunt N.H. Richardson D.R. Stocker R. J. Immunol. 2001; 166: 6332-6340Crossref PubMed Scopus (106) Google Scholar). Consistent with this finding, nonactivated human MDM did not contain detectable IDO mRNA or protein (supplemental Fig. S1A), whereas treatment of these cells with IFNγ resulted in the time-dependent induction of IDO as indicated by the parallel increases in IDO mRNA, protein, and activity, the latter reflected by increased l-Trp consumption and accumulation of kynurenine and 3-hydroxyanthranilic acid in the culture medium (supplemental Fig. S1B). We also reported previously that nontoxic concentrations of NO-generating compounds inhibit IDO activity in IFNγ-stimulated MDM (14Thomas S.R. Mohr D. Stocker R. J. Biol. Chem. 1994; 269: 14457-14464Abstract Full Text PDF PubMed Google Scholar). Inhibition of IDO activity by NO could result from inhibition at the level of transcription, activity, or promotion of IDO protein degradation (14Thomas S.R. Mohr D. Stocker R. J. Biol. Chem. 1994; 269: 14457-14464Abstract Full Text PDF PubMed Google Scholar, 15Alberati-Giani D. Malherbe P. Ricciardi-Castagnoli P. Kohler C. Denis-Donini S. Cesura A.M. J. Immunol. 1997; 159: 419-426PubMed Google Scholar, 16Hucke C. MacKenzie C.R. Adjogble K.D. Takikawa O. Daubener W. Infect. Immun. 2004; 72: 2723-2730Crossref PubMed Scopus (99) Google Scholar). To determine whether NO affected IDO expression in IFNγ-stimulated MDM, we tested the effect of adding the NO donor SNAP or GSNO concomitant with IFNγ and examining IDO mRNA and protein expression 18 h later. SNAP and GSNO when added to cells cultured in serum-supplemented medium produce NO with a half-life of 4–5 and 1–2 h, respectively (23Terwel D. Nieland L.J. Schutte B. Reutelingsperger C.P. Ramaekers F.C. Steinbusch H.W. Eur. J. Pharmacol. 2000; 400: 19-33Crossref PubMed Scopus (49) Google Scholar, 24Khan S. Kayahara M. Joashi U. Mazarakis N.D. Sarraf C. Edwards A.D. Hughes M.N. Mehmet H. J. Cell Sci. 1997; 110: 2315-2322Crossref PubMed Google Scholar, 25Hortelano S. Traves P.G. Zeini M. Alvarez A.M. Bosca L. J. Immunol. 2003; 171: 6059-6064Crossref PubMed Scopus (19) Google Scholar). We observed that neither SNAP (Fig. 2, A, B, and D) nor GSNO (Fig. 2, C and E) significantly impacted on IDO mRNA or protein expression when used at concentrations up to 500 μm. To investigate whether NO directly inhibited IDO activity, we first stimulated human MDM with IFNγ for 32 h to express functional IDO protein and then exposed the cells to increasing concentrations of NO donors for 4 h in fresh medium supplemented with l-Trp prior to assessing IDO protein and enzyme activity. The NO donors we employed were GSNO or Spermine NONOate, the latter donor releasing 2 mol of NO/mol of parent compound with a half-life of 40 min at pH 7 and 37 °C (26Maragos C.M. Morley D. Wink D.A. Dunams T.M. Saavedra J.E. Hoffman A. Bove A.A. Isaac L. Hrabie J.A. Keefer L.K. J. Med. Chem. 1991; 34: 3242-3247Crossref PubMed Scopus (701) Google Scholar). Fig. 3 shows that GSNO and spermine NONOate dose-dependently inhibited IDO activity, as indicated by the decrease in the extent to which MDM metabolized l-Trp to kynurenine. This inhibition was not due to altered IDO protein expression (Fig. 3). Similar findings were made with SNAP, which at 50 μm inhibited enzyme activity exhibited by IDO-expressing MDM by 75 ± 5% (n = 4) compared with control MDM after 4 h of incubation. We next tested whether inhibition of cellular IDO activity by NO donors is reversible. For this, we exposed IFNγ-primed MDM to GSNO and examined the capacity of these cells to metabolize l-Trp to kynurenine over the ensuing 8 h. We employed GSNO, which decays under cell culture conditions with a half-life of ∼1–2 h (24Khan S. Kayahara M. Joashi U. Mazarakis N.D. Sarraf C. Edwards A.D. Hughes M.N. Mehmet H. J. Cell Sci. 1997; 110: 2315-2322Crossref PubMed Google Scholar, 25Hortelano S. Traves P.G. Zeini M. Alvarez A.M. Bosca L. J. Immunol. 2003; 171: 6059-6064Crossref PubMed Scopus (19) Google Scholar), so that under the experimental conditions employed the majority of NO was anticipated to be liberated within the first 2–4 h. Indeed, measurement of the increase in nitrite afforded by GSNO as an index of NO generation showed that the donor produced NO for the initial 3 h only under the current experimental conditions (not shown). As shown in Fig. 4A, GSNO inhibited IDO activity relative to control cells over the initial 4 h of incubation. However, at later time points (4–8 h), when GSNO ceased releasing NO, the rate at which MDM metabolized l-Trp to kynurenine returned essentially to the corresponding control cell values. Further support for the implied reversible inhibition of IDO enzyme activity by NO was obtained with transfected COS-7 cells expressing active human IDO and exposed to GSNO for increasing periods of time. Thus, although GSNO substantially (i.e. by >75%) inhibited IDO activity over the initial 4 h, the extent of this inhibition decreased with increasing duration of incubation such that after 18 h IDO activity had returned to control cell levels (Fig. 4B). These changes in enzyme activity occurred independently of any significant changes in IDO protein expression by the transfected COS-7 cells (Fig. 4B). NO Inhibits rhIDO Activity—Results from the cellular experiments described above indicated that NO inhibited cellular IDO activity at the post-translational level. We next asked whether the observed NO-mediated inhibition of cellular IDO activity could be recapitulated with the purified protein in the presence of co-factors known to activate the isolated enzyme. For this, we exposed rhIDO to NO donors in the presence of ascorbate and methylene blue co-factors, which reduce the inactive ferric iron (FeIII) to the active FeII form of the enzyme. The NO donors employed were DEANO and spermine NONOate, which at pH 7 and 37 °C liberate 1.5 or 2.0 mol of NO/mol of parent molecule with half-lives of 2 and 40 min, respectively (26Maragos C.M. Morley D. Wink D.A. Dunams T.M. Saavedra J.E. Hoffman A. Bove A.A. Isaac L. Hrabie J.A. Keefer L.K. J. Med. Chem. 1991; 34: 3242-3247Crossref PubMed Scopus (701) Google Scholar). Both NO donors dose-dependently inhibited the activity of rhIDO, with DEANO the more efficacious, consistent with its shorter half-life of NO release (Fig. 5). For example, based on half-life after 15 min of incubation at 37 °C, 1 μm DEANO and Spermine NONOate produce 1.5 and 0.5 μm NO, which inhibited the activity of rhIDO (0.1 μm in heme) by ∼60 and ∼35%, respectively (Fig. 5). Spectroscopic Properties of NO-rhIDO Adducts—Reduction of the IDO heme from the FeIII to the FeII form facilitates binding of l-Trp and O2 to the active site (27Sono M. Taniguchi T. Watanabe Y. Hayaishi O. J. Biol. Chem. 1980; 255: 1339-1345Abstract Full Text PDF PubMed Google Scholar). As NO binding to heme can regulate the activity of various enzymes (28Thomas D.D. Miranda K.M. Colton C.A. Citrin D. Espey M.G. Wink D.A. Antioxid. Redox Signal. 2003; 5: 307-317Crossref PubMed Scopus (81) Google Scholar), we next characterized the spectral properties of FeIII and FeII rhIDO ± l-Trp and NO using UV-visible absorption spectroscopy. In the absence of l-Trp, native FeIII rhIDO exhibited a Soret band maximum at 404.5 nm, as well as α/β absorption band maxima at 501, ∼535, ∼570, and 632 nm (Table 1, supplemental Fig. S2, A and B). Addition of NO resulted in a large shift of the Soret band maximum from 404.5 to 416 nm and the appearance of two distinct bands in the α/β region at 532 and 566 nm (Table 1, supplemental Fig. S2, A and B). The subsequent addition of l-Trp induced a slight shift of the Soret band back to 415.5 nm but no changes to the α/β absorption band maxima. Reduction of FeIII to FeII rhIDO shifted the Soret band from 404.5 to 428 nm and resulted in the appearance of new peaks in the α/β region at ∼530 (shoulder) and 558 nm (Table 1, supplemental Fig. S2, C and D). The subsequent addition of l-Trp produced a slight blue-shift in the Soret band to 426 nm and in the α/β region to 557 nm. In contrast, the addition of NO to FeII-IDO caused a large blue-shift in the Soret band to 418 nm and the formation of new peaks at 545 and 574 nm in the α/β region. The addition of l-Trp further shifted the Soret band to 415 nm and altered the relative intensities of the absorption peaks at 546 and ∼573 nm (Table 1, supplemental Fig. S2, C and D). The results of this spectral analysis show that each individual rhIDO NO-heme adduct (i.e. FeIII or FeII ± l-Trp) has a unique, distinguishable, spectroscopic signature that allows for positive identification.TABLE 1Optical properties of ferric and ferrous NO complexes of rhIDOIDO adductγ-Soret bandaSpectra were obtained under anaerobic conditions at pH 7.0. The uncertainty in absorption band maxima is ± 0.2 nm.α/β bandsaSpectra were obtained under anaerobic conditions at pH 7.0. The uncertainty in absorption band maxima is ± 0.2 nm.nmnmFeIII404.5501, ∼535, ∼570, 632FeIII-Trp406538, ∼570, ∼632FeIII-NO416532, 566FeIII-Trp-NO415.5532, 566FeII428∼530, 558FeII-Trp426∼530, 557FeII-NO418545, 574FeII-Trp-NO415546, ∼573a Spectra were obtained under anaerobic conditions at pH 7.0. The uncertainty in absorption band maxima is ± 0.2 nm. Open table in a new tab Spectroscopic Properties of Active rhIDO—To examine how NO inhibits rhIDO, we first assessed the utility of optical absorption spectroscopy to study changes in the heme prosthetic group of rhIDO following activation of the enzyme by ascorbate plus methylene blue under aerobic conditions in the presence of l-Trp. Activation of rhIDO decreased the intensity and shifted the γ-Soret maximum to a longer wavelength (supplemental Fig. S3, A and B). In addition, a shoulder appeared that, upon subtraction of the resting rhIDO spectrum, exhibited a maximum of ∼426–429 nm (supplemental Fig. S3B, inset). We postulated that this maximum is characteristic for the active quaternary FeII species formed from the deoxy precursor, i.e. FeII-O2-Trp, which is likely to occur in resonance equilibrium with a FeIII-superoxo-Trp adduct (FeIII-O–2-Trp) (29Hirata F. Ohnishi T. Hayaishi O. J. Biol. Chem. 1977; 252: 4637-4642Abstract Full Text PDF PubMed Google Scholar). Also coinciding with activation of IDO was the time-dependent increase in absorbance at ∼325 nm, resulting from formation of N-formyl-kynurenine, the immediate reaction product of the IDO-catalyzed oxidation of l-Trp (supplemental Fig. S3A, arrow). This increase in absorbance at 325 nm correlated with the decrease in l-Trp and the accumulation of kynurenine as determined by HPLC (not shown). Together, these results indicate that the optical spectral changes observed as a result of activation of rhIDO directly reflect the changes occurring to both the heme environment and activity of the enzyme. Raman Characterization of Active rhIDO—Resonance Raman spectroscopy was employed to monitor changes to the structure, coordination, spin, and redox state of the heme active site as the result of activation by the addition of ascorbate and methylene blue. The top spectrum in Fig. 6 represents "resting" rhIDO under anaerobic conditions in the presence of methylene blue and l-Trp but the absence of ascorbate. As observed, this spectrum exhibited features almost identical to those of FeIII-IDO with l-Trp, which we reported previously (21Terentis A.C. Thomas S.R. Takikawa O. Littlejohn T.K. Truscott R.J. Armstrong R.S. Yeh S.R. Stocker R. J. Biol. Chem. 2002; 277: 15788-15794Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar). The core-size marker bands ν3 (1481/1500 cm–1) and ν2 (1561/1573 cm–1) are indicative of a six-coordinate, predominantly low-spin, heme iron. The addition of ascorbate under anaerobic conditions led to the partial reduction of heme, as evidenced by the shoulder appearing at ∼1353 cm–1 on the low frequency edge of the ν4 band and the appearance of a peak at 1468 cm–1 (Fig. 6, +
Referência(s)