Oxidized Human Neuroglobin Acts as a Heterotrimeric Gα Protein Guanine Nucleotide Dissociation Inhibitor
2003; Elsevier BV; Volume: 278; Issue: 38 Linguagem: Inglês
10.1074/jbc.m305519200
ISSN1083-351X
AutoresKeisuke Wakasugi, Tomomi Nakano, Isao Morishima,
Tópico(s)Heme Oxygenase-1 and Carbon Monoxide
ResumoNeuroglobin (Ngb) is a newly discovered vertebrate heme protein that is expressed in the brain and can reversibly bind oxygen. It has been reported that Ngb expression levels increase in response to oxygen deprivation and that it protects neurons from hypoxia in vitro and in vivo. However, the mechanism of this neuroprotection remains unclear. In the present study, we tried to clarify the neuroprotective role of Ngb under oxidative stress in vitro. By surface plasmon resonance, we found that ferric Ngb, which is generated spontaneously as a result of the rapid autoxidation, binds exclusively to the GDP-bound form of the α subunit of heterotrimeric G protein (Gαi). In GDP dissociation assays or guanosine 5′-O-(3-thio)triphosphate binding assays, ferric Ngb behaved as a guanine nucleotide dissociation inhibitor (GDI), inhibiting the rate of exchange of GDP for GTP. The interaction of GDP-bound Gαi with ferric Ngb will liberate Gβγ, leading to protection against neuronal death. In contrast, ferrous ligand-bound Ngb under normoxia did not have GDI activities. Taken together, we propose that human Ngb may be a novel oxidative stress-responsive sensor for signal transduction in the brain. Neuroglobin (Ngb) is a newly discovered vertebrate heme protein that is expressed in the brain and can reversibly bind oxygen. It has been reported that Ngb expression levels increase in response to oxygen deprivation and that it protects neurons from hypoxia in vitro and in vivo. However, the mechanism of this neuroprotection remains unclear. In the present study, we tried to clarify the neuroprotective role of Ngb under oxidative stress in vitro. By surface plasmon resonance, we found that ferric Ngb, which is generated spontaneously as a result of the rapid autoxidation, binds exclusively to the GDP-bound form of the α subunit of heterotrimeric G protein (Gαi). In GDP dissociation assays or guanosine 5′-O-(3-thio)triphosphate binding assays, ferric Ngb behaved as a guanine nucleotide dissociation inhibitor (GDI), inhibiting the rate of exchange of GDP for GTP. The interaction of GDP-bound Gαi with ferric Ngb will liberate Gβγ, leading to protection against neuronal death. In contrast, ferrous ligand-bound Ngb under normoxia did not have GDI activities. Taken together, we propose that human Ngb may be a novel oxidative stress-responsive sensor for signal transduction in the brain. Neuroglobin (Ngb) 1The abbreviations used are: Ngb, neuroglobin; ferrous-O2 Ngb, ferrous oxygen-bound Ngb; ferrous-CO Ngb, ferrous carbon monoxide-bound Ngb; G protein, guanine nucleotide-binding protein; Hb, hemoglobin; nsHb, nonsymbiotic plant Hb; Mb, myoglobin; GRK, G protein-coupled receptor kinase; RGS, regulator of G protein signaling; GPCR, G protein-coupled receptor; GEF, guanine nucleotide exchange factor; GDI, guanine nucleotide dissociation inhibitor; GAP, GTPase-activating protein; DTT, dithiothreitol; SPR, surface plasmon resonance; RU, resonance units; GTPγS, guanosine 5′-O-(3-thio)triphosphate; Pcp2, Purkinje cell protein-2; GPR, G protein regulatory; GoLoco, Gαi/o-Loco interaction; GAIP, Gα-interacting protein; aa, amino acids. is a recently discovered globin found in the vertebrate brain that has a high affinity for oxygen (1Burmester T. Weich B. Reinhardt S. Hankeln T. Nature. 2000; 407: 520-523Crossref PubMed Scopus (905) Google Scholar, 2Awenius C. Hankeln T. Burmester T. Biochem. Biophys. Res. Commun. 2001; 287: 418-421Crossref PubMed Scopus (86) Google Scholar, 3Zhang C. Wang C. Deng M. Li L. Wang H. Fan M. Xu W. Meng F. Qian L. He F. Biochem. Biophys. Res. Commun. 2002; 290: 1411-1419Crossref PubMed Scopus (81) Google Scholar). Globins are iron porphyrin complex (heme)-containing proteins that bind reversibly to oxygen and as such play an important role in respiratory function. They have been found in many taxa including bacteria, fungi, plants, and animals (4Hardison R.C. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 5675-5679Crossref PubMed Scopus (290) Google Scholar). The two major globins that have been described in vertebrates are hemoglobin and myoglobin. Hemoglobin (Hb), which consists of four subunits that cooperatively bind oxygen, is present in red blood cells where it is responsible for transporting oxygen from the lungs to the tissues (5Dickerson R.E. Geis I. Hemoglobin: Structure, Function, Evolution and Pathology. Benjamin/Cummings, Menlo Park, CA1983Google Scholar). Myoglobin (Mb) is a monomeric intracellular globin that stores oxygen in muscle tissue and facilitates its diffusion from the periphery of the cell to mitochondria, which use it during oxidative phosphorylation (6Wittenberg B.A. Wittenberg J.B. Annu. Rev. Physiol. 1989; 51: 857-878Crossref PubMed Scopus (410) Google Scholar). Although Ngb shares only 21–25% sequence identity with vertebrate Hb and Mb, it conserves the key amino acid residues that are required for Hb and Mb function (1Burmester T. Weich B. Reinhardt S. Hankeln T. Nature. 2000; 407: 520-523Crossref PubMed Scopus (905) Google Scholar). Like Hb and Mb, Ngb can reversibly bind oxygen (1Burmester T. Weich B. Reinhardt S. Hankeln T. Nature. 2000; 407: 520-523Crossref PubMed Scopus (905) Google Scholar, 7Dewilde S. Kiger L. Burmester T. Hankeln T. Baudin-Creuza V. Aerts T. Marden M.C. Caubergs R. Moens L. J. Biol. Chem. 2001; 276: 38949-38955Abstract Full Text Full Text PDF PubMed Scopus (413) Google Scholar, 8Trent III, J.T. Watts R.A. Hargrove M.S. J. Biol. Chem. 2001; 276: 30106-30110Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar). The iron atom in the heme prosthetic group of each globin normally exists in either the ferrous (Fe2+) or ferric (Fe3+) state. In the absence of exogenous ligands, the ferric and ferrous forms of Ngb are hexacoordinated with the endogenous protein ligands, distal histidine and proximal histidine (7Dewilde S. Kiger L. Burmester T. Hankeln T. Baudin-Creuza V. Aerts T. Marden M.C. Caubergs R. Moens L. J. Biol. Chem. 2001; 276: 38949-38955Abstract Full Text Full Text PDF PubMed Scopus (413) Google Scholar) (Fig. 1). Oxygen (O2) or carbon monoxide (CO) can displace the distal histidine of ferrous Ngb to produce ferrous oxygen-bound Ngb (ferrous-O2 Ngb) or ferrous carbon monoxide-bound Ngb (ferrous-CO Ngb) (7Dewilde S. Kiger L. Burmester T. Hankeln T. Baudin-Creuza V. Aerts T. Marden M.C. Caubergs R. Moens L. J. Biol. Chem. 2001; 276: 38949-38955Abstract Full Text Full Text PDF PubMed Scopus (413) Google Scholar). On the other hand, Hb and Mb are normally hexacoordinated in the ferric state, with a water molecule coordinated to iron and pentacoordinated in the ferrous form, leaving the sixth position empty and available for the binding of exogenous ligands such as O2 and CO. The mammalian brain accounts for up to 20% of the total oxygen consumption, even though it constitutes only 2% of total body weight, and it is the most sensitive organ to the effects of tissue hypoxia (9Ereciñska M. Silver I.A. Respir. Physiol. 2001; 128: 263-276Crossref PubMed Scopus (447) Google Scholar). Ngb is widely expressed in the cerebral cortex, hippocampus (CA1, CA2, CA3, and CA4, especially in the pyramidal layer), thalamus, hypothalamus, and cerebellum (1Burmester T. Weich B. Reinhardt S. Hankeln T. Nature. 2000; 407: 520-523Crossref PubMed Scopus (905) Google Scholar, 3Zhang C. Wang C. Deng M. Li L. Wang H. Fan M. Xu W. Meng F. Qian L. He F. Biochem. Biophys. Res. Commun. 2002; 290: 1411-1419Crossref PubMed Scopus (81) Google Scholar, 10Reuss S. Saaler-Reinhardt S. Weich B. Wystub S. Reuss M.H. Burmester T. Hankeln T. Neuroscience. 2002; 115: 645-656Crossref PubMed Scopus (166) Google Scholar) of the rat brain. Recently, it has been suggested that Ngb plays a role in the neuronal response to hypoxia and ischemia (11Sun Y. Jin K. Mao X.O. Zhu Y. Greenberg D.A. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 15306-15311Crossref PubMed Scopus (456) Google Scholar, 12Sun Y. Jin K. Peel A. Mao X.O. Xie L. Greenberg D.A. Natl. Acad. Sci. U. S. A. 2003; 100: 3497-3500Crossref PubMed Scopus (355) Google Scholar). Ngb expression was reported to increase in response to neuronal hypoxia in vitro and focal cerebral ischemia in vivo (11Sun Y. Jin K. Mao X.O. Zhu Y. Greenberg D.A. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 15306-15311Crossref PubMed Scopus (456) Google Scholar, 12Sun Y. Jin K. Peel A. Mao X.O. Xie L. Greenberg D.A. Natl. Acad. Sci. U. S. A. 2003; 100: 3497-3500Crossref PubMed Scopus (355) Google Scholar). Neuronal survival following hypoxia was reduced by inhibiting Ngb expression with an antisense oligodeoxynucleotide and was enhanced by Ngb overexpression, supporting the notion that Ngb protects neurons from hypoxicischemic insults (11Sun Y. Jin K. Mao X.O. Zhu Y. Greenberg D.A. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 15306-15311Crossref PubMed Scopus (456) Google Scholar). Moreover, Ngb protected the brain from experimental stroke in vivo (12Sun Y. Jin K. Peel A. Mao X.O. Xie L. Greenberg D.A. Natl. Acad. Sci. U. S. A. 2003; 100: 3497-3500Crossref PubMed Scopus (355) Google Scholar). A possible mechanism by which Ngb protects these neurons is by functioning as an O2 carrier, facilitating the diffusion of O2 to the mitochondria within cells that are engaging in active aerobic metabolism, in a manner similar to the way Mb acts in muscle cells. However, Ngb has been estimated to comprise less than 0.01% of the total protein content in the brain (1Burmester T. Weich B. Reinhardt S. Hankeln T. Nature. 2000; 407: 520-523Crossref PubMed Scopus (905) Google Scholar). The low concentration (in the micromolar range) of Ngb in brain tissue perhaps argues against its role in storing and carrying significant amounts of O2. On the other hand, local concentrations of Ngb may reach sufficiently high levels to allow it to regulate local oxygen consumption (10Reuss S. Saaler-Reinhardt S. Weich B. Wystub S. Reuss M.H. Burmester T. Hankeln T. Neuroscience. 2002; 115: 645-656Crossref PubMed Scopus (166) Google Scholar, 13Schmidt M. Giessl A. Laufs T. Hankeln T. Wolfrum U. Burmester T. J. Biol. Chem. 2003; 278: 1932-1935Abstract Full Text Full Text PDF PubMed Scopus (284) Google Scholar). Finally, although debatable, the affinity of Ngb for oxygen may be so high as to prevent its release under physiological conditions (7Dewilde S. Kiger L. Burmester T. Hankeln T. Baudin-Creuza V. Aerts T. Marden M.C. Caubergs R. Moens L. J. Biol. Chem. 2001; 276: 38949-38955Abstract Full Text Full Text PDF PubMed Scopus (413) Google Scholar, 8Trent III, J.T. Watts R.A. Hargrove M.S. J. Biol. Chem. 2001; 276: 30106-30110Abstract Full Text Full Text PDF PubMed Scopus (224) Google Scholar). Thus, the mechanism by which Ngb affords neuroprotection under oxidative stress conditions such as ischemia and reperfusion remains unclear. The objective of this study was to investigate whether Ngb has novel functions that are related to neuroprotective roles under oxidative stress. On line BLAST searches were performed via the website of the National Center for Biotechnology Information (Conserved Domain Database, www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi; www.ncbi.nlm.nih.gov/BLAST). These analyses revealed that human Ngb has 25–35% amino acid sequence homology with regulators of G protein signaling (RGS) and RGS domains of G protein-coupled receptor kinases (GRK) (Fig. 1). The protein that most closely resembles Ngb (24% amino acid identity) is GRK4 (Fig. 1). RGS and GRK proteins are modulators of heterotrimeric G proteins (14Pitcher J.A. Freedman N.J. Lefkowitz R.J. Annu. Rev. Biochem. 1998; 67: 653-692Crossref PubMed Scopus (1068) Google Scholar, 15Penn R.B. Pronin A.N. Benovic J.L. Trends Cardiovasc. Med. 2000; 10: 81-89Crossref PubMed Scopus (188) Google Scholar, 16Zheng B. De Vries L. Farquhar M.G. Trends Biochem. Sci. 1999; 24: 411-414Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar, 17Wieland T. Chen C.-K. Naunyn-Schmiedeberg's Arch. Pharmacol. 1999; 360: 14-26Crossref PubMed Scopus (62) Google Scholar). Heterotrimeric G proteins (G proteins) consist of an α subunit (Gα) with GTPase activity and a βγ dimer (Gβγ) and belong to a family of proteins, whose signal transduction function depends on the binding of guanine nucleotides (18Gilman A.G. Annu. Rev. Biochem. 1987; 56: 615-649Crossref PubMed Scopus (4714) Google Scholar, 19Bourne H.R. Sanders D.A. McCormick F. Nature. 1991; 349: 117-127Crossref PubMed Scopus (2690) Google Scholar, 20Simon M.I. Strathmann M.P. Gautam N. Science. 1991; 252: 802-808Crossref PubMed Scopus (1585) Google Scholar, 21Hepler J.R. Gilman A.G. Trends Biochem. Sci. 1992; 17: 383-387Abstract Full Text PDF PubMed Scopus (924) Google Scholar, 22Sprang S.R. Annu. Rev. Biochem. 1997; 66: 639-678Crossref PubMed Scopus (888) Google Scholar, 23Vetter I.R. Wittinghofer A. Science. 2001; 294: 1299-1304Crossref PubMed Scopus (1384) Google Scholar). Ligand- or signal-activated G protein-coupled receptors (GPCRs) induce GDP release from a Gα subunit, which is followed by the binding of GTP. Binding of GTP to Gα “turns on” the system and causes conformational changes that result in dissociation of the GTP-bound Gα from both the receptor and Gβγ. The GTP-bound Gα and Gβγ can then regulate the activity of different effector molecules, such as adenylyl cyclase, phospholipase Cβ, and ion channels. Signal transduction is “turned off” by the intrinsic GTPase activity of the Gα protein, which hydrolyzes the bound GTP to GDP, inducing the reassociation of GDP-bound Gα with Gβγ. The on/off G protein ratio can be regulated by three groups of protein modulators: guanine nucleotide exchange factors (GEFs), which stimulate GDP dissociation and subsequent GTP binding; guanine nucleotide dissociation inhibitors (GDIs), which inhibit GDP dissociation; and GTPase-activating proteins (GAPs), which enhance GTP hydrolysis (18Gilman A.G. Annu. Rev. Biochem. 1987; 56: 615-649Crossref PubMed Scopus (4714) Google Scholar, 19Bourne H.R. Sanders D.A. McCormick F. Nature. 1991; 349: 117-127Crossref PubMed Scopus (2690) Google Scholar, 20Simon M.I. Strathmann M.P. Gautam N. Science. 1991; 252: 802-808Crossref PubMed Scopus (1585) Google Scholar, 21Hepler J.R. Gilman A.G. Trends Biochem. Sci. 1992; 17: 383-387Abstract Full Text PDF PubMed Scopus (924) Google Scholar, 22Sprang S.R. Annu. Rev. Biochem. 1997; 66: 639-678Crossref PubMed Scopus (888) Google Scholar, 23Vetter I.R. Wittinghofer A. Science. 2001; 294: 1299-1304Crossref PubMed Scopus (1384) Google Scholar). GPCRs play a role as functional analogues of GEFs (18Gilman A.G. Annu. Rev. Biochem. 1987; 56: 615-649Crossref PubMed Scopus (4714) Google Scholar, 20Simon M.I. Strathmann M.P. Gautam N. Science. 1991; 252: 802-808Crossref PubMed Scopus (1585) Google Scholar, 21Hepler J.R. Gilman A.G. Trends Biochem. Sci. 1992; 17: 383-387Abstract Full Text PDF PubMed Scopus (924) Google Scholar). GRKs phosphorylate agonist-activated forms of GPCRs to induce homologous desentsitization of signaling pathways (14Pitcher J.A. Freedman N.J. Lefkowitz R.J. Annu. Rev. Biochem. 1998; 67: 653-692Crossref PubMed Scopus (1068) Google Scholar, 15Penn R.B. Pronin A.N. Benovic J.L. Trends Cardiovasc. Med. 2000; 10: 81-89Crossref PubMed Scopus (188) Google Scholar). RGS proteins act as GAPs for Gαi or Gαo and play a role in desensitization (16Zheng B. De Vries L. Farquhar M.G. Trends Biochem. Sci. 1999; 24: 411-414Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar, 17Wieland T. Chen C.-K. Naunyn-Schmiedeberg's Arch. Pharmacol. 1999; 360: 14-26Crossref PubMed Scopus (62) Google Scholar). In the present study, we examined the possibility of interaction of Ngb with Gα by surface plasmon resonance (SPR) measurements. Ferric Ngb interacted exclusively with Gαi in their GDP-bound forms. In GDP dissociation assays or GTPγS binding assays, ferric Ngb exhibited GDI activity, inhibiting the rate of exchange of GDP for GTP by Gαi. Since ferrous ligand-bound Ngb under normoxia did not have GDI activities, human Ngb may function as a novel oxidative stress-responsive sensor for signal transduction in the brain. Samples—Rat myristoylated Gα subunits (Gαi1, Gαi2, Gαi3, and Gαo; Calbiochem) and bovine Gβγ (Calbiochem) were used. [35S]GTPγS (>1000 Ci/mmol), [8-3H]GDP (10–15 Ci/mmol), and [α-32P]GTP (∼3000 Ci/mmol) were purchased from Amersham Biosciences (Buckinghamshire, England). Preparation of Proteins—Amplification of human Ngb cDNA was performed by PCR using human universal Quick-clone cDNA (Clontech, Palo Alto, CA). Human Ngb cDNA was cloned into plasmid PET20b (Novagen, Madison, WI) and was sequenced using an ABI 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA). Overexpression of human Ngb was induced in Escherichia coli strain BL 21 (DE 3) (Novagen) by treatment with isopropyl β-d-thiogalactopyranoside for 4 h. Purification of Ngb without His6-tag was carried out as follows (24Kriegl J.M. Bhattacharyya A.J. Nienhaus K. Deng P. Minkow O. Nienhaus G.U. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 7992-7997Crossref PubMed Scopus (153) Google Scholar). Soluble cell extract was loaded onto a DEAE-Sepharose anion-exchange column equilibrated with 20 mm Tris-HCl, pH 8.0. Ngb was eluted from the column with buffer containing 75 mm NaCl and was further purified by passage through a Sephacryl S-200 HR gel filtration column. Human Ngb mutants, including a COOH-terminal tag of six histidine residues (His6-tag), were purified on nickel affinity columns (His·Bind® resin; Novagen) from the supernatant of lysed cells using the protocol provided by Novagen. Mass spectrometric measurements of purified Ngbs were performed using matrix-assisted laser desorption ionization time-of-flight mass spectrometry (PerSpective Biosystems Voyager™ DE PRO-S.D., Applied Biosystems), and Edman degradation was carried out on purified Ngbs using a G1005A protein sequencing system (Hewlett-Packard, Palo Alto, CA) at Takara Biomedicals, Inc. to determine their NH2-terminal sequences. Ferric Ngbs were incubated with 2 mm dithiothreitol (DTT) for 2 h, and then DTT was removed by chromatography using a PD-10 column (Amersham Biosciences). Ferrous-CO Ngbs were generated after addition of sodium dithionite and CO gas to the DTT-treated ferric Ngb followed by gel filtration. Site-directed Mutagenesis—A QuikChange™ site-directed mutagenesis system (Stratagene, La Jolla, CA) was used to alter cysteine residues (amino acid residues 46, 55, and/or 120) in human Ngb. The point mutations were confirmed by DNA sequencing using BigDye terminator cycle sequencing FS (Applied Biosystems) and an ABI 3100 genetic analyzer (Applied Biosystems). SPR Experiments—SPR measurements were performed on a BIAcore® X Instrument (Biacore, Uppsala, Sweden). Rat myristoylated Gα subunit (Gαi1, Gαi2, Gαi3, or Gαo) was immobilized on the surface of a CM5 sensor chip using an amine coupling kit (Biacore) according to the instructions of the manufacturer. Activation of the carboxymethylated dextran in the CM5 sensor chip was carried out by mixing equal volumes of 400 mm N-ethyl-N′-(3-dimethylaminopropyl)carbodiimide in water and 100 mm hydrochloride/N-hydroxysuccinimide in water, and injecting the mixture into the instrument at 10 μl/min for 7 min. This was followed by the injection of 5 μg/μl Gα protein dissolved in 10 mm acetate buffer, pH 4.5, over the activated surface of the sensor chip for 7 min at a flow rate of 10 μl/min. The unreacted sites of the sensor chip were masked by the injection for 7 min of 1 m ethanolamine, pH 8.5. After immobilization, nonspecifically bound protein was removed by washing with running buffer (10 mm Hepes, 150 mm NaCl, 0.005% Tween 20, pH 7.4) until the value of the resonance units (RU) became nearly constant. All binding experiments were performed at 25 °C at a flow rate of 5 μl/min. Ferric or ferrous-CO Ngb in the running buffer was injected for 5 min, during which association occurred. Dissociation then took place in the running buffer over the next 10 min. The BIAcore response was expressed in relative RU, i.e. the difference in response between flow cell with immobilized protein and the control flow channel. 1000 RU corresponded to 1 ng/mm2 of bound ligand. For binding analyses in the presence of guanine nucleotides, the running buffer containing 5 mm MgSO4 and either 500 μm GDP, or 500 μm GDP plus 500 μm AlCl3 and 10 mm NaF was loaded for 60 min to allow binding of guanine nucleotide to immobilized Gα, after which the samples were injected. After each binding cycle, the sensor chip was regenerated with 5 μl of 0.05% SDS in the running buffer and was washed with running buffer for 5–10 min prior to the next injection. Experimental curves (sensorgrams) were analyzed by means of the BIAevaluation 3.1 software package using the model A + B ⇔ AB to estimate the association and dissociation rate constants ka and kd . GTPγS Binding Assays—100 nm Gαi1 or Gαo was incubated for 3 min at 25 °C in buffer A (20 mm Tris-HCl, 100 mm NaCl, and 10 mm MgSO4 at pH 8.0) with 10 μm GDP in the absence or presence of Ngb (5 μm). Binding assays were initiated with additions of 50 nm [35S]GTPγS (>1000 Ci/mmol). Aliquots (10 μl) were withdrawn from the binding mixtures and were passed through nitrocellulose filters (0.45 μm) (Millipore, Bedford, MA). The filters were then washed three times with 1 ml of ice-cold buffer A and were counted in a liquid scintillation counter (LSC-6100; Aloka, Tokyo, Japan). The apparent rate constant (k app) values for the binding reactions were calculated by fitting the data to the following equation: GTPγS binding (%) = 100% × (1–e– kt). GDP Dissociation Assays—Gαi1 complexed with [3H]GDP (0.3 μm) was prepared by incubating 0.3 μm Gαi1 with 2 μm [3H]GDP in buffer A for 1.5 h at 25 °C. Excess unlabeled GTP or GDP (200 μm) was added to monitor dissociation of [3H]GDP from Gαi1 in the absence or presence of Ngb (5 μm). Aliquots were withdrawn at the indicated times and were passed through nitrocellulose filters (0.45 μm) (Millipore, Bedford, MA). The filters were then washed three times with 1 ml of ice-cold buffer A and were counted in a liquid scintillation counter (LSC-6100; Aloka). As for preparation of Gαi1 complexed with [α-32P]GDP (1 μm), 1 μm Gαi1 and 2 μm [α-32P]GTP were incubated in buffer A for 1.5 h at 25 °C. Experiments using Gαi1 complexed with [α-32P]GDP were also performed, as described above. SPR Detection of Human Ferric Ngb Binding to Gα—Proteins that interacted with human Ngb were sought by SPR experimentation. SPR is a powerful tool for real time measurement of direct protein-protein interactions that do not require labeling of the proteins. We covalently coupled rat Gαi1, which is highly expressed in the brain (18Gilman A.G. Annu. Rev. Biochem. 1987; 56: 615-649Crossref PubMed Scopus (4714) Google Scholar), to a sensor chip. As positive and negative controls we used Gβγ and Mb, respectively, and confirmed that Gβγ interacts with GDP-bound Gαi1, but Mb does not bind to Gαi1 by SPR (see Supplemental Material). Then we characterized the interaction between Ngb and Gαi1 by SPR. A representative sensorgram in Fig. 2A shows that the resonance response reflecting Gαi1-ferric Ngb interaction occurred in an analyte concentration-dependent manner. In the association phase (0∼300 s), the intensity of SPR increased, indicating that ferric Ngb bound to Gαi1 specifically, while in the dissociation phase (300∼900 s), the intensity of SPR decreased, indicating that ferric Ngb dissociated from the immobilized Gαi1. Binding parameters for the interaction of ferric Ngb with Gαi1 were determined to be as follows: association rate constant, ka = 5.0 × 102m–1 s–1; dissociation rate constant, kd = 3.0 × 10–4 s–1; and equilibrium dissociation constant, Kd = kd /ka = 6.0 × 102 nm. No significant resonance signals were obtained from sensor chip surfaces that did not have attached ligands (data not shown), indicating an absence of nonspecific interactions between the sensor chip surfaces and analytes. Next we investigated the possibility of interaction of ferrous-O2 Ngb with Gαi. Since ferrous-O2 Ngb is unstable and is converted into ferric Ngb very rapidly due to its autoxidation (7Dewilde S. Kiger L. Burmester T. Hankeln T. Baudin-Creuza V. Aerts T. Marden M.C. Caubergs R. Moens L. J. Biol. Chem. 2001; 276: 38949-38955Abstract Full Text Full Text PDF PubMed Scopus (413) Google Scholar), stable ferrous-CO Ngb was used for SPR experiments. As shown in Fig. 2B, the binding affinity of ferrous-CO Ngb to Gαi1 was significantly low (Kd > 1 mm) as compared with that of ferric Ngb. Moreover, further SPR measurements clarified that human ferric Ngb binds to Gαi2, Gαi3, and Gαo (K d = 5.8 × 102, 5.5 × 102, and 6.1 × 102 nm, respectively), whereas ferrous-CO Ngb does not bind them (Kd > 1 mm). Ferric Ngb Interacts Exclusively with the GDP-bound Form of Gα—Next we investigated guanine nucleotide dependence of the binding of Gαi1 to Ngb by SPR measurements. As shown in Fig. 2C, ferric Ngb bound to Gαi1, even in the presence of Mg2+ and GDP. The binding parameters (ka = 1.1 × 103m–1 s–1, k d = 6.8 × 10–4 s–1, Kd = 6.0 × 102 nm) were almost the same as those seen in the absence of Mg2+ and GDP. Aluminum tetrafluoride (ALF4-), together with Mg +, can interact with Gαi1-bound GDP and mimic GTP and thereby activate Gαi1 (21Hepler J.R. Gilman A.G. Trends Biochem. Sci. 1992; 17: 383-387Abstract Full Text PDF PubMed Scopus (924) Google Scholar, 22Sprang S.R. Annu. Rev. Biochem. 1997; 66: 639-678Crossref PubMed Scopus (888) Google Scholar). In the presence of Mg2+, GDP, and ALF4-, ferric Ngb did not bind to the activated Gαi1 (Fig. 2C). Therefore, human ferric Ngb clearly interacts exclusively with the inactive (GDP-bound) form of Gαi1. Effects of Ngb on GTPγS Binding to Gα—Since our SPR data suggested that human ferric Ngb interacts with GDP-bound Gαi1 but does not interact with activated GTP-bound Gαi1, we hypothesized that Ngb may function as a GEF or GDI for Gαi1. To determine whether ferric Ngb functions as a GEF or a GDI, we performed GTPγS (a nonhydrolyzable analog of GTP) binding experiments. Increased GTPγS binding to Gαi1 would imply ferric Ngb is a GEF, whereas decreased binding would imply that ferric Ngb is a GDI. As shown in Fig. 3A, Gαi1 bound GTPγS due to spontaneous guanine nucleotide exchange (k app = 0.081 min–1). In the presence of ferric Ngb, the rate of GTPγS binding to Gαi1 was reduced 6.2-fold (k app = 0.013 min–1) (Fig. 3A), implying that ferric Ngb functions as a GDI for Gαi. On the other hand, ferrous-CO Ngb had no effect on the GTPγS binding (k app = 0.078 min–1) (Fig. 3A). Moreover, ferric Ngb inhibited the rate of GTPγS binding to Gαo by 10-fold; in contrast ferrous-CO Ngb had no effect (k app = 0.040, 0.004, and 0.038 min–1 in the absence and presence of ferric and ferrous-CO Ngb, respectively) (Fig. 3B). These results imply that ferric Ngb is a GDI for Gαo as well as Gαi. Ferric Ngb Acts as a GDI—We then addressed the mechanism by which human ferric Ngb inhibited GTPγS binding to Gαi and Gαo. The inhibition of GTPγS binding to Gαi/o by ferric Ngb may reflect a reduction in the rate of nucleotide exchange. To examine the effects of ferric Ngb on the release of GDP from Gαi1, we measured the rates of GDP dissociation in the absence or presence of ferric Ngb. In the presence of an excess amount of unlabeled GTP, [3H]GDP release from [3H]GDP-bound Gαi1 was inhibited by ferric Ngb (Fig. 4A). The inhibition of GDP dissociation by ferric Ngb suggests that ferric Ngb diminished the rates of spontaneous GTPγS binding to Gαi and Gαo by blocking the GDP release. In other words, ferric Ngb functions as a GDI for Gαi. The most representative GDIs for heterotrimeric G proteins share conserved sequence repeats named the G protein regulatory (GPR) (25Takesono A. Cismowski M.J. Ribas C. Bernard M. Chung P. Hazard III, S. Duzic E. Lanier S.M. J. Biol. Chem. 1999; 274: 33202-33205Abstract Full Text Full Text PDF PubMed Scopus (230) Google Scholar) or GoLoco motifs (26Siderovski D.P. Diversé-Pierluissi M.A. De Vries L. Trends Biochem. Sci. 1999; 24: 340-341Abstract Full Text Full Text PDF PubMed Scopus (148) Google Scholar). One of the family, Purkinje cell protein-2 (Pcp2), can modulate GDP binding to Gαo and Gαi (27Luo Y. Denker B.M. J. Biol. Chem. 1999; 274: 10685-10688Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar, 28Natochin M. Gasimov K.G. Artemyev N.O. Biochemistry. 2001; 40: 5322-5328Crossref PubMed Scopus (79) Google Scholar). In the presence of excess unlabeled GTP, Pcp2 preferentially interacts with the GDP-bound conformation of Gα and serves exclusively as a GDI as does human ferric Ngb (28Natochin M. Gasimov K.G. Artemyev N.O. Biochemistry. 2001; 40: 5322-5328Crossref PubMed Scopus (79) Google Scholar, 29Natochin M. Gasimov K.G. Artemyev N.O. Biochemistry. 2002; 41: 258-265Crossref PubMed Scopus (22) Google Scholar). On the other hand, in the presence of excess unlabeled GDP, Pcp2 was reported to stimulate GDP release from Gαo (27Luo Y. Denker B.M. J. Biol. Chem. 1999; 274: 10685-10688Abstract Full Text Full Text PDF PubMed Scopus (78) Google Scholar). To further characterize properties of Ngb as a GDI, we performed GDP dissociation assays in the presence of excess GDP. As shown in Fig. 4B, ferric Ngb stimulated [3H]GDP release from [3H]GDP-bound Gαi1 in the presence of an excess amount of unlabeled GDP, suggesting that the mechanism of ferric Ngb as a GDI is similar to that of Pcp2. Experiments using [α-32P]GDP instead of [3H]GDP also supported these results (data not shown). Functional Analyses of Ngb with an Intra- or Intermolecular Disulfide Bond—Cysteine (Cys) residues are particularly sensitive to oxidation by almost all forms of reactive oxygen species during ischemia and reperfusion (30Eaton P. Byers H.L. Leeds N. Ward M.A. Shattock M.J. J. Biol. Chem. 2002; 2
Referência(s)