Direct Interaction of the NifL Regulatory Protein with the GlnK Signal Transducer Enables the Azotobacter vinelandiiNifL-NifA Regulatory System to Respond to Conditions Replete for Nitrogen
2002; Elsevier BV; Volume: 277; Issue: 18 Linguagem: Inglês
10.1074/jbc.m112262200
ISSN1083-351X
AutoresRichard Little, Victoria Colombo, Andrew Leech, Ray Dixon,
Tópico(s)Legume Nitrogen Fixing Symbiosis
ResumoThe Azotobacter vinelandii NifL-NifA regulatory system integrates metabolic signals for redox, energy, and nitrogen status to fine tune regulation of the synthesis of molybdenum nitrogenase. The NifL protein utilizes discrete mechanisms to perceive these signals leading to the formation of a protein-protein complex, which inhibits NifA activity. Whereas redox signaling is mediated via a flavin-containing PAS domain in the N-terminal region of NifL, the nitrogen status is sensed via interaction with PII-like signal transduction proteins and small molecular weight effectors. The nonuridylylated form of the PII-like protein encoded by A. vinelandii glnK (Av GlnK) stimulates NifL to inhibit transcriptional activation by NifA in vitro. Here we demonstrate that the nonmodified form of Av GlnK directly interacts with A. vinelandii NifL and that this interaction is dependent on Mg2+, ATP, and 2-oxoglutarate. Differences were observed in the regulation of the Av GlnK-NifL interaction by 2-oxoglutarate compared with the role of this effector in modulating the interaction of enteric PII-like proteins with their receptors. We also report that the interaction between Av GlnK and NifL is abolished by site-directed substitution of a single amino acid in the T-loop region of Av GlnK and that uridylylation of the conserved tyrosine residue in the T-loop inhibits the interaction. No association was detected between Av GlnK and the N-terminal region of NifL and our results demonstrate that Av GlnK directly interacts with the C-terminal histidine protein kinase-like domain. The Azotobacter vinelandii NifL-NifA regulatory system integrates metabolic signals for redox, energy, and nitrogen status to fine tune regulation of the synthesis of molybdenum nitrogenase. The NifL protein utilizes discrete mechanisms to perceive these signals leading to the formation of a protein-protein complex, which inhibits NifA activity. Whereas redox signaling is mediated via a flavin-containing PAS domain in the N-terminal region of NifL, the nitrogen status is sensed via interaction with PII-like signal transduction proteins and small molecular weight effectors. The nonuridylylated form of the PII-like protein encoded by A. vinelandii glnK (Av GlnK) stimulates NifL to inhibit transcriptional activation by NifA in vitro. Here we demonstrate that the nonmodified form of Av GlnK directly interacts with A. vinelandii NifL and that this interaction is dependent on Mg2+, ATP, and 2-oxoglutarate. Differences were observed in the regulation of the Av GlnK-NifL interaction by 2-oxoglutarate compared with the role of this effector in modulating the interaction of enteric PII-like proteins with their receptors. We also report that the interaction between Av GlnK and NifL is abolished by site-directed substitution of a single amino acid in the T-loop region of Av GlnK and that uridylylation of the conserved tyrosine residue in the T-loop inhibits the interaction. No association was detected between Av GlnK and the N-terminal region of NifL and our results demonstrate that Av GlnK directly interacts with the C-terminal histidine protein kinase-like domain. adenylyltransferase, product of glnE PII-like signal transduction protein, product of A. vinelandii glnK PII signal transduction protein, product of E. coli glnB integration host factor isothermal titration calorimetry nitrogen regulator II or NRII, product of E. coli ntrB surface plasmon resonance uridylyltransferase/uridylyl-removing enzyme, product ofglnD The NifL-NifA regulatory system in Azotobacter vinelandii controls transcription of nitrogen fixation (nif) genes in response to redox, carbon, and nitrogen status. The nif-specific transcriptional activator, NifA, activates transcription by σ54 (σN)-RNA polymerase holoenzyme at nif promoters under conditions appropriate for nitrogen fixation, and the regulatory protein NifL controls the transcriptional activation functions of NifA in response to environmental cues (1Dixon R. Arch. Microbiol. 1998; 169: 371-380Crossref PubMed Scopus (109) Google Scholar). The NifL protein utilizes discrete mechanisms to perceive these signals (2Söderbäck E. Reyes-Ramirez F. Eydmann T Austin S. Hill S. Dixon R. Mol. Microbiol. 1998; 28: 179-192Crossref PubMed Scopus (72) Google Scholar), leading to the formation of a protein-protein complex that inhibits NifA activity (3Money T. Jones T. Dixon R. Austin S. J. Bacteriol. 1999; 181: 4461-4468Crossref PubMed Google Scholar, 4Money T. Barrett J. Dixon R. Austin S. J. Bacteriol. 2001; 183: 1359-1368Crossref PubMed Scopus (28) Google Scholar). The binding of adenosine nucleotides to NifL plays a key role in transducing environmental signals to form the inhibitory protein complex (3Money T. Jones T. Dixon R. Austin S. J. Bacteriol. 1999; 181: 4461-4468Crossref PubMed Google Scholar, 5Eydmann T. Söderbäck E. Jones T. Hill S. Austin S. Dixon R. J. Bacteriol. 1995; 177: 1186-1195Crossref PubMed Google Scholar). NifL is a flavoprotein that senses redox status via an N-terminal FAD-containing PAS domain (6Hill S. Austin S. Eydmann T. Jones T. Dixon R. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 2143-2148Crossref PubMed Scopus (137) Google Scholar, 7Macheroux P. Hill S. Austin S. Eydmann T. Jones T. Kim S.-O. Poole R. Dixon R. Biochem. J. 1998; 332: 413-419Crossref PubMed Scopus (58) Google Scholar, 8Zhulin I.B. Taylor B.L. Dixon R. Trends Biochem. Sci. 1997; 22: 331-333Abstract Full Text PDF PubMed Scopus (344) Google Scholar). The mechanism whereby the NifL-NifA system perceives the nitrogen status is less well understood although recent evidence implicates direct interaction with PII-like signal transduction proteins (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). PII-like proteins are among the most widely distributed signal transduction proteins in nature and serve to integrate signals of carbon and nitrogen status to regulate nitrogen metabolism (10Arcondéguy T. Jack R. Merrick M. Microbiol. Mol. Biol. Rev. 2001; 65: 80-105Crossref PubMed Scopus (352) Google Scholar, 11Ninfa A.J. Jiang P. Atkinson M.R. Peliska J.A. Curr. Top. Cell Regul. 2000; 36: 31-75Crossref PubMed Scopus (79) Google Scholar, 12Ninfa A. Atkinson M. Trends Microbiol. 2000; 8: 172-179Abstract Full Text Full Text PDF PubMed Scopus (219) Google Scholar). Many proteobacteria encode two paralogous PII signal-transducing proteins, termed PII and GlnK, and representatives of this protein family have been most thoroughly characterized in the enteric bacteria. The Escherichia coli PII-like proteins are covalently modified in response to nitrogen limitation by the uridylyltransferase/uridylyl-removing enzyme (UTase/UR) encoded byglnD (13Jiang P. Peliska J.A. Ninfa A.J. Biochemistry. 1998; 37: 12782-12794Crossref PubMed Scopus (169) Google Scholar). This enzyme catalyzes the noncooperative uridylylation of PII with up to three UMP groups covalently attached per trimer. Glutamine is the primary signal for the fixed nitrogen status in enteric bacteria (14Ikeda T.P. Shauger A.E. Kustu S. J. Mol. Biol. 1996; 259: 589-607Crossref PubMed Scopus (154) Google Scholar, 15Schmitz R.A. Curr. Microbiol. 2000; 41: 357-362Crossref PubMed Scopus (21) Google Scholar), and this effector binds to UTase/UR to stimulate the uridylyl-removing activity, thus de-uridylylating PII under nitrogen-sufficient conditions. Conversely, under conditions of nitrogen limitation when the concentration of glutamine is low, the UTase/UR uridylylates PII (10Arcondéguy T. Jack R. Merrick M. Microbiol. Mol. Biol. Rev. 2001; 65: 80-105Crossref PubMed Scopus (352) Google Scholar, 11Ninfa A.J. Jiang P. Atkinson M.R. Peliska J.A. Curr. Top. Cell Regul. 2000; 36: 31-75Crossref PubMed Scopus (79) Google Scholar, 12Ninfa A. Atkinson M. Trends Microbiol. 2000; 8: 172-179Abstract Full Text Full Text PDF PubMed Scopus (219) Google Scholar). Covalent modification of the PII-like proteins influences their ability to interact with different receptors to control nitrogen assimilation, including adenylyltransferase (ATase)1 (16Jaggi R. van Heeswijk W.C. Westerhoff H.V. Ollis D.L. Vasudevan S.G. EMBO J. 1997; 16: 5562-5571Crossref PubMed Scopus (82) Google Scholar, 17Jiang P. Peliska J.A. Ninfa A.J. Biochemistry. 1998; 37: 12802-12810Crossref PubMed Scopus (123) Google Scholar) and NtrB (NRII) (18Jiang P. Peliska J.A. Ninfa A.J. Biochemistry. 1998; 37: 12795-12801Crossref PubMed Scopus (91) Google Scholar). The Escherichia coli PII protein (Ec PII) activates the phosphatase activity and inhibits the kinase activity of NtrB (NRII) (19Atkinson M. Kamberov E.S. Weiss R.L. Ninfa A.J. J. Biol. Chem. 1994; 269: 28288-28293Abstract Full Text PDF PubMed Google Scholar, 20Jiang P. Ninfa A.J. J. Bacteriol. 1999; 181: 1906-1911Crossref PubMed Google Scholar). PII also stimulates ATase to adenylylate glutamine synthetase. In contrast, PII-UMP inhibits the activity of ATase to stimulate de-adenylylation of glutamine synthetase but does not interact with NtrB, thus enabling the latter to phosphorylate NtrC (nitrogen regulatory protein NRI) (10Arcondéguy T. Jack R. Merrick M. Microbiol. Mol. Biol. Rev. 2001; 65: 80-105Crossref PubMed Scopus (352) Google Scholar, 11Ninfa A.J. Jiang P. Atkinson M.R. Peliska J.A. Curr. Top. Cell Regul. 2000; 36: 31-75Crossref PubMed Scopus (79) Google Scholar, 12Ninfa A. Atkinson M. Trends Microbiol. 2000; 8: 172-179Abstract Full Text Full Text PDF PubMed Scopus (219) Google Scholar). In addition to control via covalent modification, the activity of the PII paralogues is also modulated by the synergistic binding of the low molecular weight effectors, ATP and 2-oxoglutarate (21Kamberov E. Atkinson M. Ninfa A. J. Biol. Chem. 1995; 270: 17797-17807Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar, 22Atkinson M.R. Ninfa A.J. Mol. Microbiol. 1999; 32: 301-313Crossref PubMed Scopus (83) Google Scholar). The interaction between PII with the ATase and NtrB is allosterically regulated by the binding of 2-oxoglutarate to PII such that binding of a single molecule of 2-oxoglutarate enhances the interaction of PII with these effectors, whereas anti-cooperative binding of more than one 2-oxoglutarate molecule per PII trimer diminishes the interaction with NtrB and ATase (17Jiang P. Peliska J.A. Ninfa A.J. Biochemistry. 1998; 37: 12802-12810Crossref PubMed Scopus (123) Google Scholar, 18Jiang P. Peliska J.A. Ninfa A.J. Biochemistry. 1998; 37: 12795-12801Crossref PubMed Scopus (91) Google Scholar). A. vinelandii apparently encodes only a single PII-like protein (called Av GlnK or formerly Av PII), which is expressed from the glnK amtB operon (23Meletzus D. Rudnick P. Doetsch N. Green A. Kennedy C. J. Bacteriol. 1998; 180: 3260-3264Crossref PubMed Google Scholar). This protein appears to be essential for cell survival, as knock-out mutations in theglnK gene are apparently lethal. Av GlnK is subject to covalent modification by a homologue of E. coli UTase/UR encoded by A. vinelandii glnD (24Rudnick P. Colnaghi R. Green A. Kennedy C. Newton W. Biological Nitrogen Fixation in the 21st Century. Kluwer Academic Publishers, Dordrecht, The Netherlands1998: 123-124Google Scholar, 25Colnaghi R. He P. Rudnick L. Green A. Yan D. Larson E. Kennedy C. Microbiology. 2001; 147: 1267-1276Crossref PubMed Scopus (19) Google Scholar). The UTase/UR appears to be involved in the regulation of nitrogen fixation in A. vinelandii because mutations in glnD give rise to a Nif phenotype that can be suppressed by secondary mutations innifL (25Colnaghi R. He P. Rudnick L. Green A. Yan D. Larson E. Kennedy C. Microbiology. 2001; 147: 1267-1276Crossref PubMed Scopus (19) Google Scholar, 26Contreras A. Drummond M. Bali A. Blanco G. Garcia E. Bush G. Kennedy C. Merrick M. J. Bacteriol. 1991; 173: 7741-7749Crossref PubMed Google Scholar). One possible interpretation of this result is that the uridylylation status of Av GlnK mediates regulation of nitrogen fixation via interaction with the NifL-NifA system. We have recently demonstrated with purified components that Av GlnK, but not Av GlnK-UMP, stimulates the ability of NifL to inhibit transcriptional activation by NifA in the presence of 2-oxoglutarate and ATP (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). The inhibitory activity of NifL was also stimulated by Ec PII, consistent with our observation that PII-like proteins are required for NifL to inhibit NifA under conditions of nitrogen excess when the A. vinelandii nifL-nifA operon is expressed in E. coli (27Reyes-Ramirez F. Little R. Dixon R. J. Bacteriol. 2001; 183: 3076-3082Crossref PubMed Scopus (47) Google Scholar). The role of 2-oxoglutarate in regulating the NifL-NifA system is further complicated by our observation that the activities of these proteins are modulated by 2-oxoglutarate in the absence of Av GlnK (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). Thus, 2-oxoglutarate controls NifL-NifA activity directly, in addition to its potential role in controlling the activity of Av GlnK. These observations suggest a model in which Av GlnK signals the nitrogen status by direct interaction with the NifL-NifA system under conditions of nitrogen excess and that inhibition of NifA by NifL is relieved by elevated levels of 2-oxoglutarate when Av GlnK is uridylylated under conditions of nitrogen limitation (Fig.1). To elucidate the mechanism of nitrogen sensing by the A. vinelandii NifL-NifA system, it is necessary to determine which protein component(s) interact with Av GlnK and to analyze the effectors required for this interaction. Potentially, Av GlnK could interact with either NifL or NifA, or both of these components, to modulate transcriptional activation. In this report we have investigated, with purified components, the interaction of Av GlnK with NifL and NifA. We observe a specific interaction between NifL and Av GlnK that is dependent on the presence of both 2-oxoglutarate and ATP. The interaction was not detectable with a mutant form of GlnK that is defective in stimulating inhibition of NifA activity by NifL in vitro. Furthermore, our data indicate that the C-terminal nucleotide-binding domain of NifL is sufficient for the interaction with Av GlnK. Plasmid pTJ42 expressing native A. vinelandii NifA with a C-terminal hexahistidine tag (NifA6-His) was prepared from plasmid pDB737 (28Austin S. Buck M. Cannon W. Eydmann T. Dixon R. J. Bacteriol. 1994; 176: 3460-3465Crossref PubMed Scopus (65) Google Scholar). An EcoRI-BamHI fragment from pDB737 containing the 3′ end of NifA was cloned into the vector pTE103 to give plasmid pTJ39 containing a single BglII site within theEcoRI-BamHI region. A double-stranded oligonucleotide with engineered BglII and BamHI sticky ends and encoding a hexahistidine tag and stop codon, respectively, was cloned into pTJ39 yielding plasmid pTJ41. TheEcoRI-BamHI fragment from pTJ41 was subsequently cloned into pDB737 to give plasmid pTJ42. Plasmid pVCO1, expressingA. vinelandii GlnD with a N-terminal histidine tag, was obtained by cloning a NdeI-BamHI fragment from pYZ5 (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar) into the expression vector pET28(a)+ (Novagen). Ec PII was expressed from plasmid pBOP1 in E. coli strain RB9060 (ΔglnB) as described in Ref. 29Kamberov E.S. Atkinson M.R. Feng J. Chandran P. Ninfa A.J. Cell. Mol. Biol. Res. 1994; 40: 175-191PubMed Google Scholar. All other constructs for protein expression have been previously described (2Söderbäck E. Reyes-Ramirez F. Eydmann T Austin S. Hill S. Dixon R. Mol. Microbiol. 1998; 28: 179-192Crossref PubMed Scopus (72) Google Scholar, 6Hill S. Austin S. Eydmann T. Jones T. Dixon R. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 2143-2148Crossref PubMed Scopus (137) Google Scholar, 9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). Mutagenesis was carried out with the QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer's instructions. Mutagenic primers were 5′-AAGGGCCATACCTGCCTGTACCGGGG-3′ (E44C) and 5′-GTACCGGGGTGCGTGCTACGTAGTCGAC-3′ (E50C). All constructs were sequenced commercially (MWG Biotech) over their entire length to ensure that only the desired mutations were introduced. Plasmids overexpressing Av GlnK mutants were derived from plasmid pYZ1 described previously (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). Tagged proteins used in this work were purified by metal chelate affinity chromatography. Two 1-ml HiTrap Chelating HP columns (Amersham Biosciences) were connected in series and equilibrated with 20 mm Tris-HCl, 50 mmKSCN, 5 mm imidazole, 250 mm NaCl, pH 8.0. Purification was carried out using Biocad Sprint Perfusion chromatography (PerkinElmer Biosystems) using a linear gradient from 0 to 750 mm imidazole over a 40-ml elution volume. Av GlnK, Ec PII, σ54, and IHF were purified as described previously (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). Molar protein concentrations were calculated on the basis that NifL6-His purifies as a tetramer and the truncated NifL polypeptides NifL-(1–284)6-His, NifL-(147–519)6-His, and NifL-(360–519)6-Hisare tetrameric, dimeric, and monomeric, respectively (2Söderbäck E. Reyes-Ramirez F. Eydmann T Austin S. Hill S. Dixon R. Mol. Microbiol. 1998; 28: 179-192Crossref PubMed Scopus (72) Google Scholar). The molar concentrations of NifA and Av GlnK were calculated on the basis that these proteins are a dimer and trimer, respectively. NifA-promoted catalysis of open promoter complexes by σ54-RNA polymerase was used to assay NifA activity and its inhibition by NifL as described previously (5Eydmann T. Söderbäck E. Jones T. Hill S. Austin S. Dixon R. J. Bacteriol. 1995; 177: 1186-1195Crossref PubMed Google Scholar, 9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar, 30Barrett J. Ray P. Sobczyk A. Little R. Dixon R. Mol. Microbiol. 2001; 39: 480-494Crossref PubMed Scopus (38) Google Scholar). Linearized template DNA was provided by digesting plasmid pNH8 with EcoRI and BamHI to yield a 260-bp fragment, containing the Klebsiella pneumoniae nifH promoter and upstream activator sequences, which was 3′ end-labeled with [α-32P]dGTP at the BamHI site. Reactions (final volume of 15 μl) were carried out in TAP buffer (50 mm Tris acetate, 100 mm potassium acetate, 8 mm magnesium acetate, 3.5% polyethylene glycol 8000, 1 mm dithiothreitol, pH 7.9) and contained 5 nmtemplate DNA, 3.4 μg/ml denatured salmon sperm DNA, 125 nm core RNA polymerase, 200 nmσ54, 100 nm IHF, and 3.5 mm ATP. GTP (final concentration 500 μm) was present prior to the heparin challenge to allow formation of initiated complexes, which are more stable than open promoter complexes (5Eydmann T. Söderbäck E. Jones T. Hill S. Austin S. Dixon R. J. Bacteriol. 1995; 177: 1186-1195Crossref PubMed Google Scholar). In some cases an ATP regenerating system was provided by adding creatine kinase (20 units/ml) and creatine phosphate (12 mm). The reaction components (including Av GlnK at concentrations indicated in Fig. 2) were preincubated for 2 min at 30 °C, and reactions were then initiated by the addition of either NifA alone or NifA plus NifL. After 20 min of incubation, reactions were mixed with 3 μl of dye mix containing 50% glycerol, 0.05% bromphenol blue, 0.1% xylene cyanol, and 2 μg of heparin and immediately loaded onto a 4% (w/v) polyacrylamide gel (acrylamide:bisacrylamide ratio 80:1) in 25 mm Tris, 400 mm glycine, pH 8.6, which had been prerun at 180 V at room temperature down to a constant power of 2 watts. Gels were run for 2.5–3 h at 100 V. Gels were dried down, and the percentage of radioactivity in open complexes quantitated with the Fujix BAS1000 phosphorimager. Ni-NTA-magnetic agarose beads (Qiagen) were equilibrated in buffer containing 10 mm HEPES, 150 mm NaCl, 25 mm MgCl2, and 20 mm imidazole, pH 7.5. NifL polypeptides, at a concentration of 0.3 μm, were immobilized by preincubation in a total volume of 500 μl containing 50 μl of the magnetic bead suspension. After 30 min, the beads were washed in the above buffer and Av GlnK was added to a final concentration of 1 μm in a volume of 500 μl. Following an additional 60-min incubation, the beads were washed, the buffer subsequently removed, and elution performed with HEPES buffer as above but containing 500 mm imidazole. Aliquots of each sample were analyzed by SDS-polyacrylamide gel electrophoresis. Surface plasmon resonance experiments were performed using a BIAcore X biosensor system (BIAcore AB). Hexahistidine-tagged derivatives of NifL and NifA were immobilized on nickel-NTA biosensor chip surfaces. Nickel was first bound to the NTA surface through injection of 20-μl volumes of 500 μm nickel chloride. Proteins for immobilization were introduced to the protein-chip surface at a concentration of 40 nm. Experiments were performed at 25 °C in buffer containing 10 mm HEPES, 150 mm NaCl, 25 mm MgCl2, pH 7.4, at a flow rate of 20 μl/min. Most experiments were carried out in the presence of a control protein in the reference flow cell, as stated in the figure legends. In all cases it was ensured that the binding response of the control protein upon immobilization was at least equal to that of the sample protein. Experiments were performed in a VP-ITC isothermal titration calorimeter (MicroCal, Inc.) at 28 °C in a cell volume of 1.35 ml. Binding of 2-oxoglutarate to PII-like proteins was measured by titrating 2 mm ligand from a 250-μl injection syringe into the sample cell, which was stirred at 300 rpm. 5- or 10-μl injections were made to the stoichiometric excess given in Fig. 6. Buffer conditions were 10 mm HEPES, 50 mm NaCl, 25 mmMgCl2, 1 mm ATP, pH 7.0. The PII proteins were dialyzed overnight prior to ITC, and protein concentration was determined by the Bradford method, with bovine serum albumin as the standard. The heat change for the dilution of the ligand in the absence of protein was measured for each experiment and was subtracted from the measured heat change of ligand binding to protein. Data analysis was performed with the Origin program, provided by MicroCal. The uridylylation method was based on previously published methods (22Atkinson M.R. Ninfa A.J. Mol. Microbiol. 1999; 32: 301-313Crossref PubMed Scopus (83) Google Scholar). The reaction mixture contained 100 mm Tris-HCl, 25 mm MgCl2, 100 mm KCl, 1 mm dithiothreitol, 0.3 mg/ml bovine serum albumin, 0.5 mm ATP, 0.5 mm UTP, 500 μm 2-oxoglutarate, pH 7.5. Av GlnK was present at a concentration of 15 μm and GlnD6-His at a concentration of 1 μm. Following incubation at 30 °C, aliquots of the reaction were removed at time intervals to monitor uridylylation. Glycerol and KCl were added to 10% (v/v) and 350 mm final concentrations, respectively, and the samples were heated to 60 °C for 15 min to inactivate UTase/UR. Samples were added to an equal volume of native sample buffer (125 mmTris-HCl, 20% glycerol, 0.05% bromphenol blue, pH 8.0) and analyzed by nondenaturing polyacrylamide gel electrophoresis. Limited proteolysis was performed at 20 °C in Tris acetate buffer comprising 50 mm Tris acetate, 100 mm potassium acetate, 8 mmmagnesium acetate, 1 mm dithiothreitol, 1 mmATP, 500 μm 2-oxoglutarate, pH 7.0. NifL and GlnK were combined in a final reaction volume of 130 μl and preincubated for 10 min prior to initiation of digestion. A trypsin:NifL weight ratio of 1:200 was used. 15-μl samples were removed at the time intervals indicated in the figure legend to tubes containing a 2-fold weight excess of soybean trypsin-chymotrypsin inhibitor. To these samples was added 15 μl of gel loading buffer (125 mm Tris-HCl, 4% sodium dodecyl sulfate, 20% glycerol, 10% β-mercaptoethanol, 0.05% bromphenol blue, pH 8.6). Samples were heated at 100 °C for 4 min prior to electrophoretic separation. To provide controls for the protein-protein interaction experiments, we performed site-directed mutagenesis of the surface exposed T-loop region of Av GlnK, which contains the tyrosine residue that is the target site for uridylylation and, by analogy with the well characterized E. coli PII-like proteins, is likely to play a major role in the interaction with receptors (31Jiang P. Zucker P. Atkinson M.R. Kamberov E.S. Tirasophon W. Chandran P. Schefke B.R. Ninfa A.J. J. Bacteriol. 1997; 179: 4342-4353Crossref PubMed Scopus (85) Google Scholar, 32Jiang P. Zucker P. Ninfa A.J. J. Bacteriol. 1997; 179: 4354-4360Crossref PubMed Google Scholar, 33Jaggi R. Ybarlucea W. Cheah E. Carr P.D. Edwards K.J. Ollis D.L. Vasesuvan S.G. FEBS Lett. 1996; 391: 223-228Crossref PubMed Scopus (41) Google Scholar, 34Arcondéguy T. Lawson D. Merrick M. J. Biol. Chem. 2000; 275: 38452-38456Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar). The T-loop sequence of the Av GlnK protein (residues 37–55) differs from Ec PII in only a single amino acid at position 52. Engineered derivatives of Ec PII with single cysteine substitutions at glutamate 44 and glutamate 50 have been used previously to probe interactions between Ec PII and NtrB (NRII) with heterobifunctional cross-linking reagents, which label thiol groups (35Pioszak A.A. Jiang P. Ninfa A.J. Biochemistry. 2000; 39: 13450-13461Crossref PubMed Scopus (64) Google Scholar). Although these mutations do not perturb the PII-NtrB interaction, mutations in the T-loop frequently allow discrimination between receptors (31Jiang P. Zucker P. Atkinson M.R. Kamberov E.S. Tirasophon W. Chandran P. Schefke B.R. Ninfa A.J. J. Bacteriol. 1997; 179: 4342-4353Crossref PubMed Scopus (85) Google Scholar). We prepared the equivalent E44C and E50C mutations in Av GlnK and purified the mutant proteins following overexpression in E. coli (Fig.2A). Both mutant proteins behaved similarly to the native protein on purification. However, in contrast to the native protein, neither of the mutant proteins could be uridylylated by purified A. vinelandii UTase/UR (data not shown), indicating that they are either defective in the interaction with the UTase or are defective as substrates in the catalytic step. The ability of the mutant proteins to activate the inhibitory function of NifL was assayed by measuring the influence of NifL on transcriptional activation by NifA at the nifH promoterin vitro. These assays quantitate the formation of heparin-resistant open promoter complexes by σ54-RNA polymerase holoenzyme catalyzed by NifA in the presence of nucleoside triphosphates and IHF (5Eydmann T. Söderbäck E. Jones T. Hill S. Austin S. Dixon R. J. Bacteriol. 1995; 177: 1186-1195Crossref PubMed Google Scholar). To simplify the assays, we use a truncated form of NifL, NifL-(147–519)6-His, which lacks the redox response but retains the response to nitrogen status in vivo(2Söderbäck E. Reyes-Ramirez F. Eydmann T Austin S. Hill S. Dixon R. Mol. Microbiol. 1998; 28: 179-192Crossref PubMed Scopus (72) Google Scholar) and the ability to respond to Av GlnK in vitro (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar). This truncated protein thus allows reactions to be performed under aerobic conditions without activating the redox response of NifL. NifL-(147–519)6-His inhibits NifA activity in the presence of ATP, but this inhibition is relieved by the addition of 2-oxoglutarate. However, the addition of native Av GlnK enhances the inhibitory activity of NifL-(147–519)6-His when both ATP and 2-oxoglutarate are present (9Little R. Reyes-Ramirez F. Zhang Y. van Heeswijk W.C. Dixon R. EMBO J. 2000; 19: 6041-6050Crossref PubMed Scopus (85) Google Scholar) (Fig. 2B). Whereas Av GlnK E50C, like native Av GlnK, possesses this activity, the mutant Av GlnK E44C protein exerted little or no influence on inhibition of NifA activity by NifL (Fig. 2B). Neither the native nor the mutant proteins had any influence on NifA activity in the absence of NifL (data not shown). It would therefore appear that the GlnK E44C mutant is substantially defective in activating the inhibitory function of NifL, whereas the Av GlnK E50C mutant is not defective in this function. A "pull-down" assay using Ni-NTA-magnetic agarose beads was employed to probe for interactions between Av GlnK and immobilized NifL and NifA in the presence and absence of effectors. Purified derivatives of NifL and NifA bearing a hexahistidine tag at the C terminus were immobilized on the Ni-NTA bead surface, and the beads were then incubated in the presence of Av GlnK. A wash phase was carried out to remove adventitiously bound protein, and elution of the His-tagged proteins was subsequently performed with 500 mmimidazole. By analogy with the well characterized interactions between Ec PII-like proteins and their receptors, we suspected that the binding of both 2-oxoglutarate and ATP to Av GlnK would be required to detect any interaction. Accordingly, no interaction was resolved in the absence of ATP or 2-oxoglutarate, or in the presence of only a single effector (Fig. 3, lanes 2–4). In the presence of both ATP (3.5 mm) and 2-oxoglutarate (2 mm), Av GlnK co-eluted with the native NifL6-His protein (Fig. 3, lanes 5 and6) but not with the NifA6-His protein (Fig. 3,lane 9). The Av GlnK E44C mutant protein defective in stimulation of NifL inhibition in vitro did not co-elute with NifL (Fig. 3, lane 7). In contrast, the GlnK E50
Referência(s)