Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Human Phosphatidylethanolamine-binding Protein Facilitates Heterotrimeric G Protein-dependent Signaling

2001; Elsevier BV; Volume: 276; Issue: 43 Linguagem: Inglês

10.1074/jbc.m106991200

ISSN

1083-351X

Autores

Thomas Kroslak, Thomas Koch, Evelyn Kahl, Volker Höllt,

Tópico(s)

Receptor Mechanisms and Signaling

Resumo

In this study we report that human phosphatidylethanolamine-binding protein (hPBP) facilitates heterotrimeric G protein-coupled signaling. In Xenopus laevis oocytes, coexpression of hPBP with human µ opioid receptor, human δ opioid receptor, or human somatostatin receptor 2 evoked an agonist-induced increase in potassium conductance of G protein-activated inwardly rectifying potassium channels. This activation of heterotrimeric G protein signaling in oocytes could also be elicited by injection of bacterially overexpressed and purified hPBP. Stimulatory effect was pertussis toxin-sensitive and present even in the absence of coexpressed receptors. Additionally, an increase in G protein-mediated inhibition of adenylate cyclase activity, measured by the inhibition of forskolin-mediated cAMP accumulation, could be detected in HEK293 and NIH3T3 cells after expression of hPBP and inXenopus oocytes after injection of hPBP. As [35S]guanosine 5′-3-O-(thio)triphosphate (GTPγS) binding to membranes prepared from hPBP-expressing cells was significantly elevated and recombinant hPBP dose-dependently stimulated [35S]GTPγS binding to native membranes, the results presented provide strong evidence that hPBP-induced effects are G protein-dependent. These data suggest a novel function of hPBP in regulating G protein and G protein-coupled receptor signalingin vivo. In this study we report that human phosphatidylethanolamine-binding protein (hPBP) facilitates heterotrimeric G protein-coupled signaling. In Xenopus laevis oocytes, coexpression of hPBP with human µ opioid receptor, human δ opioid receptor, or human somatostatin receptor 2 evoked an agonist-induced increase in potassium conductance of G protein-activated inwardly rectifying potassium channels. This activation of heterotrimeric G protein signaling in oocytes could also be elicited by injection of bacterially overexpressed and purified hPBP. Stimulatory effect was pertussis toxin-sensitive and present even in the absence of coexpressed receptors. Additionally, an increase in G protein-mediated inhibition of adenylate cyclase activity, measured by the inhibition of forskolin-mediated cAMP accumulation, could be detected in HEK293 and NIH3T3 cells after expression of hPBP and inXenopus oocytes after injection of hPBP. As [35S]guanosine 5′-3-O-(thio)triphosphate (GTPγS) binding to membranes prepared from hPBP-expressing cells was significantly elevated and recombinant hPBP dose-dependently stimulated [35S]GTPγS binding to native membranes, the results presented provide strong evidence that hPBP-induced effects are G protein-dependent. These data suggest a novel function of hPBP in regulating G protein and G protein-coupled receptor signalingin vivo. phosphatidylethanolamine-binding protein [d-Ala2,d-Leu5]enkephalin [d-Ala2,N-Me-Phe4,Gly-ol5]enkephalin G protein-coupled receptor guanosine 5′-O-(thio)triphosphate cloned human δ opioid receptor, hMOR, cloned human µ opioid receptor cloned human phosphatidylethanolamine-binding protein (also called RKIP) cloned human somatostatin receptor (type 2) cloned rat µ opioid receptor cloned rat G protein-coupled inwardly rectifying potassium channels two-electrode voltage clamp N,N,N',N'-tetramethylethylenediamine pertussis toxin Phosphatidylethanolamine-binding proteins (PBPs)1 comprise a family of polypeptides found to be present in various plant and animal organisms like Arabidopsis (1Ohshima S. Murata M. Sakamoto W. Ogura Y. Motoyoshi F. Mol. Gen. Genet. 1997; 254: 186-194Crossref PubMed Scopus (148) Google Scholar, 2Pnueli L. Carmel-Goren L. Hareven D. Gutfinger T. Alvarez J. Ganal M. Zamir D. Lifschitz E. Development. 1998; 125: 1979-1989PubMed Google Scholar, 3Amaya I. Ratcliffe O.J. Bradley D.J. Plant Cell. 1999; 11: 1405-1418Crossref PubMed Scopus (119) Google Scholar), tomato (1Ohshima S. Murata M. Sakamoto W. Ogura Y. Motoyoshi F. Mol. Gen. Genet. 1997; 254: 186-194Crossref PubMed Scopus (148) Google Scholar, 2Pnueli L. Carmel-Goren L. Hareven D. Gutfinger T. Alvarez J. Ganal M. Zamir D. Lifschitz E. Development. 1998; 125: 1979-1989PubMed Google Scholar, 3Amaya I. Ratcliffe O.J. Bradley D.J. Plant Cell. 1999; 11: 1405-1418Crossref PubMed Scopus (119) Google Scholar), tobacco (1Ohshima S. Murata M. Sakamoto W. Ogura Y. Motoyoshi F. Mol. Gen. Genet. 1997; 254: 186-194Crossref PubMed Scopus (148) Google Scholar, 2Pnueli L. Carmel-Goren L. Hareven D. Gutfinger T. Alvarez J. Ganal M. Zamir D. Lifschitz E. Development. 1998; 125: 1979-1989PubMed Google Scholar, 3Amaya I. Ratcliffe O.J. Bradley D.J. Plant Cell. 1999; 11: 1405-1418Crossref PubMed Scopus (119) Google Scholar),Saccharomyces cerevisiae (4Robinson L.C. Tatchell K. Mol. Gen. Genet. 1991; 230: 241-250Crossref PubMed Scopus (52) Google Scholar), Onchocerca volvulus(5Erttmann K.D. Gallin M.Y. Gene (Amst.). 1996; 174: 203-207Crossref PubMed Scopus (12) Google Scholar), Caenorhabditis elegans (6Wilson R. Ainscough R. Anderson K. Baynes C. Berks M. Bonfield J. Burton J. Connell M. Copsey T. Cooper J. et al.Nature. 1994; 368: 32-38Crossref PubMed Scopus (1438) Google Scholar), and Drosophila melanogaster (7Adams M.D. Celniker S.E. Holt R.A. Evans C.A. Gocayne J.D. Amanatides P.G. Scherer S.E. Li P.W. Hoskins R.A. Galle R.F. George R.A. et al.Science. 2000; 287: 2185-2195Crossref PubMed Scopus (4854) Google Scholar), as well as in mammals (human, rat, mouse, and bovine) (8Seddiqi N. Segretain D. Bucquoy S. Pineau C. Jegou B. Jolles P. Schoentgen F. Experientia. 1996; 52: 101-110Crossref PubMed Scopus (21) Google Scholar, 9Bernier I. Jolles P. Biochim. Biophys. Acta. 1984; 790: 174-181Crossref PubMed Scopus (144) Google Scholar). Although a variety of biological roles have previously been described, the cellular and molecular function of species-specific PBPs remains unclear. The plant PBP orthologues TERMINAL FLOWER 1, SELF-PRUNING, and CENTRORADIALIS were shown to be involved in the regulation of flowering signaling and meristem growth (1Ohshima S. Murata M. Sakamoto W. Ogura Y. Motoyoshi F. Mol. Gen. Genet. 1997; 254: 186-194Crossref PubMed Scopus (148) Google Scholar, 2Pnueli L. Carmel-Goren L. Hareven D. Gutfinger T. Alvarez J. Ganal M. Zamir D. Lifschitz E. Development. 1998; 125: 1979-1989PubMed Google Scholar, 3Amaya I. Ratcliffe O.J. Bradley D.J. Plant Cell. 1999; 11: 1405-1418Crossref PubMed Scopus (119) Google Scholar). In nematodes PBPs were found to be part of secreted cell-surface proteins and proposed to function as a protective mechanism against host immunological response (10Gems D. Ferguson C.J. Robertson B.D. Nieves R. Page A.P. Blaxter M.L. Maizels R.M. J. Biol. Chem. 1995; 270: 18517-18522Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar). TheDrosophila PBP homologues were decribed as putative odorant-binding effector molecules, expressed in different subsets of olfactory hairs (11Pikielny C.W. Hasan G. Rouyer F. Rosbash M. Neuron. 1994; 12: 35-49Abstract Full Text PDF PubMed Scopus (309) Google Scholar). The human and rat brain PBPs act as precursor proteins for the hippocampal cholinergic neurostimulating peptide. This undecapeptide is released of the N-terminal region of PBP (12Tohdoh N. Tojo S. Agui H. Ojika K. Brain Res. Mol. Brain Res. 1995; 30: 381-384Crossref PubMed Scopus (65) Google Scholar) and stimulates the acetylcholine synthesis and the cholinergic activity in rat medial septal nuclei (13Ojika K. Mitake S. Kamiya T. Kosuge N. Taiji M. Brain Res. Dev. Brain Res. 1994; 79: 1-9Crossref PubMed Scopus (44) Google Scholar). Although PBP expression could be detected in all mammalian tissues tested (14Katada E. Mitake S. Matsukawa N. Otsuka Y. Tsugu Y. Fujimori O. Ojika K. Histochem. Cell Biol. 1996; 105: 43-51Crossref PubMed Scopus (22) Google Scholar, 15Frayne J. Ingram C. Love S. Hall L. Cell Tissue Res. 1999; 298: 415-423Crossref PubMed Scopus (60) Google Scholar), particularly high levels were found in spermatids (16Frayne J. McMillen A. Love S. Hall L. Mol. Reprod. Dev. 1998; 49: 454-460Crossref PubMed Scopus (37) Google Scholar), brain oligodendrocytes, Purkinje cells, and specific cortical and hippocampal neuronal cell layers (15Frayne J. Ingram C. Love S. Hall L. Cell Tissue Res. 1999; 298: 415-423Crossref PubMed Scopus (60) Google Scholar,17Moore C. Perry A.C. Love S. Hall L. Brain Res. Mol. Brain Res. 1996; 37: 74-78Crossref PubMed Scopus (34) Google Scholar). Some studies support the hypothesis that PBPs are involved in cell-signaling machinery. The S. cerevisiae PBP orthologue TFS1 (for TWENTY-FIVE SUPPRESSOR1) was demonstrated to act as a dose-dependent suppressor of CDC25 mutations (4Robinson L.C. Tatchell K. Mol. Gen. Genet. 1991; 230: 241-250Crossref PubMed Scopus (52) Google Scholar). CDC25 acts as a GTP/GTP exchange factor for Ras and is implicated in Ras activation (18Innocenti M. Zippel R. Brambilla R. Sturani E. FEBS Lett. 1999; 460: 357-362Crossref PubMed Scopus (41) Google Scholar, 19Overbeck A.F. Brtva T.R. Cox A.D. Graham S.M. Huff S.Y. Khosravi-Far R. Quilliam L.A. Solski P.A. Der C.J. Mol. Reprod. Dev. 1995; 42: 468-476Crossref PubMed Scopus (64) Google Scholar, 20Zippel R. Orecchia S. Sturani E. Martegani E. Oncogene. 1996; 12: 2697-2703PubMed Google Scholar). In contrast human RKIP (for Raf kinaseinhibitor protein), which is identical to hPBP, has been characterized as an inhibitor of Ras-mediated signaling cascade and AP1- dependent transcription (21Yeung K. Seitz T. Li S. Janosch P. McFerran B. Kaiser C. Fee F. Katsanakis K.D. Rose D.W. Mischak H. Sedivy J.M. Kolch W. Nature. 1999; 401: 173-177Crossref PubMed Scopus (763) Google Scholar). Analysis of the family of PBPs revealed high structural homology between PBPs from different species (22Banfield M.J. Barker J.J. Perry A.C. Brady R.L. Structure. 1998; 6: 1245-1254Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar, 23Banfield M.J. Brady R.L. J. Mol. Biol. 2000; 297: 1159-1170Crossref PubMed Scopus (127) Google Scholar, 24Serre L. Vallee B. Bureaud N. Schoentgen F. Zelwer C. Structure. 1998; 6: 1255-1265Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar). Mammalian PBPs consist of 187 amino acids and show a significant degree in sequence similarity (12Tohdoh N. Tojo S. Agui H. Ojika K. Brain Res. Mol. Brain Res. 1995; 30: 381-384Crossref PubMed Scopus (65) Google Scholar). The central region of these proteins constituted by residues 60–126 appears to be best conserved and contains a region believed to play an essential role in PBP function and to be responsible for binding to G proteins (25Schoentgen F. Jolles P. FEBS Lett. 1995; 369: 22-26Crossref PubMed Scopus (90) Google Scholar). In fact, bovine PBP was shown to associate with small GTP-binding proteins but failed to bind GTPs themselves (26Bucquoy S. Jolles P. Schoentgen F. Eur. J. Biochem. 1994; 225: 1203-1210Crossref PubMed Scopus (41) Google Scholar). To date, known data about PBPs implicate an involvement in multiple signaling mechanisms (22Banfield M.J. Barker J.J. Perry A.C. Brady R.L. Structure. 1998; 6: 1245-1254Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar, 23Banfield M.J. Brady R.L. J. Mol. Biol. 2000; 297: 1159-1170Crossref PubMed Scopus (127) Google Scholar, 25Schoentgen F. Jolles P. FEBS Lett. 1995; 369: 22-26Crossref PubMed Scopus (90) Google Scholar). Interestingly, the rat PBP homologue could be isolated by morphine affinity chromatography from rat brain membranes, suggesting an association with the G protein-coupled µ opioid receptor (MOR) (27Grandy D.K. Hanneman E. Bunzow J. Shih M. Machida C.A. Bidlack J.M. Civelli O. Mol. Endocrinol. 1990; 4: 1370-1376Crossref PubMed Scopus (82) Google Scholar). The postulated association of PBP with MOR prompted us to test whether hPBP is involved in receptor-mediated heterotrimeric G protein signaling. In the present work we investigated the influence of human PBP on G protein signaling mediated by hMOR and other G protein-coupled receptors (GPCRs) (human δ opioid receptor, hDOR; human somatostatin receptor, hSSTR2) in Xenopus laevis as well as in HEK293 and NIH3T3 cells. [35S]GTPγS (specific activity, 1300 Ci/mmol) was obtained from PerkinElmer Life Sciences. GTPγS, GDP, HEPES, isopropyl-1-thio-β-d-galactopyranoside, ampicillin, kanamycin, puromycin, polyethylenimine, acrylamide, bisacrylamide, TEMED, and ammonium peroxodisulfate were from Sigma (Deisenhofen, Germany). Nitrocellulose filters (NC45) were from Serva (Heidelberg, Germany). DAMGO, DADLE, octreotide, and naloxone were from Tocris (Biotrend, Germany). Morphine was from Synopharm (Barsbuettel, Germany). PTX and rat recombinant Gαi1 was from Calbiochem-Novabiochem GmbH (Schwalbach, Germany). Materials required for gel electrophoresis were from Bio-Rad (Muenchen, Germany). FemaleX. laevis were from Nasco (Fort Atkinson, WI). Two-electrode voltage clamp (TEVC) amplifier (Turbo TEC-05), breakout box, and perfusion system were from Polder electronics (Tamm, Germany). Buffers and salts were from Sigma and Merck; cAMP radioimmunoassay kit was fromAmersham Pharmacia Biotech (Braunschweig, Germany). Glass capillaries (Kwik-fil™ TW 150-F-3) for microinjection in oocytes and reference electrodes (Flexref) for TEVC were from World Precision Instruments (Sarasota, FL). Borosilicate glass capillaries for TEVC were from Hilgenberg (Malsfeld, Germany), and clamp electrodes were produced using a DMZ-Universal Puller (Zeitz Instruments, Germany). Micromanipulator (DC3001R) and electronic micromanipulator controller (MS 314) were from World Precision Instruments. Geneticin (G-418), agarose, bactotryptone, yeast extract, and agar were from Life Technologies, Inc. mMessage mMachine® kit for synthesis of capped RNA transcripts was from Ambion. cDNA for hPBP (cloned in pT7/T3) was kindly provided by Prof. Rommelspacher (Benjamin Franklin University, Berlin, Germany), and hPBP cDNA was subcloned into pcDNA3.1(−) (pcDNA/hPBP) for stable expression in cells. rMOR1 cDNA was a gift from Prof. Lei Yu, and cDNAs for rMOR1 and rKir3.4 (subcloned in pcDNA3.1; Invitrogen) were described previously (28Koch T. Kroslak T. Mayer P. Raulf E. Hollt V. J. Neurochem. 1997; 69: 1767-1770Crossref PubMed Scopus (97) Google Scholar, 29Koch T. Kroslak T. Averbeck M. Mayer P. Schroder H. Raulf E. Hollt V. Mol. Pharmacol. 2000; 58: 328-334Crossref PubMed Scopus (63) Google Scholar). For coexpression experiments, cDNA of rMOR1 was subcloned into expression plasmid pEAK 10 (Edge BioSystems) (pEAK10/rMOR1). H. A. Lester (Pasadena, CA) kindly provided rKir3.1 expression plasmid. hDOR expression plasmid was kindly provided by P. Mayer (Magdeburg, Germany), and hSSTR2 expression plasmid was from Novartis (Basel, Switzerland). HIS·BIND® Quick 900 cartridges, bacterial strain BCL1(DE3), and inducible expression plasmid pET28a(+) were from Novagen (Schwalbach, Germany). Preparation and culture of oocytes and cRNA preparation have been described previously (28Koch T. Kroslak T. Mayer P. Raulf E. Hollt V. J. Neurochem. 1997; 69: 1767-1770Crossref PubMed Scopus (97) Google Scholar). Oocytes were injected with 50 nl containing 0.5 ng each of rKir3.1 and rKir3.4 cRNA, and 1 ng of receptor cRNA, 5 ng of hPBP cRNA, or both receptor and hPBP cRNA. Oocytes were incubated for 2–4 days at 18 °C in ND96 (96 mm NaCl, 2.0 mm KCl, 1.8 mmCaCl2, 1 mm MgCl2, 5 mmHEPES) supplemented with 5% fetal bovine serum, 2.5 mmsodium pyruvate, 100 units/ml penicillin, and 100 µg/ml streptomycin. To generate capped mRNA for injection, mMessage mMachine® capped RNA transcription kit (Ambion) was used. 50 nl (10 ng) of hPBP were injected per oocyte. The method was performed as described previously (28Koch T. Kroslak T. Mayer P. Raulf E. Hollt V. J. Neurochem. 1997; 69: 1767-1770Crossref PubMed Scopus (97) Google Scholar, 29Koch T. Kroslak T. Averbeck M. Mayer P. Schroder H. Raulf E. Hollt V. Mol. Pharmacol. 2000; 58: 328-334Crossref PubMed Scopus (63) Google Scholar). Data were analyzed with Eggworks software (Polder Electronics, Tamm, Germany). Clamped oocytes were superfused with either ND96, with a moderate potassium solution (K16: 82 mm NaCl, 16 mm KCl, 1.8 mmCaCl2, 1 mm MgCl2) with K16 containing receptor-specific agonist (1 µm) or K16/BaCl2 solution (K16 with 300 µmBaCl2) (see scheme in Fig. 1). PTX experiments were performed by incubating oocytes with 1 µg/ml pertussis toxin for ∼18 h. Endogenous Ca2+ release-activated Cl−current (ICl1), which has an outwardly rectifying steady state current-voltage relationship (30Hartzell H.C. J. Gen. Physiol. 1996; 108: 157-175Crossref PubMed Scopus (87) Google Scholar, 31Machaca K. Hartzell H.C. J. Biol. Chem. 1999; 274: 4824-4831Abstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar) was measured in ND96 at positive membrane potentials. Oocytes were clamped at −80 mV and stepped to +40 mV for 500 ms every 3 s to monitor ICl1. Increase in current ICl1 after injection of either CaCl2, GTPγS, or hPBP (50 nl/oocyte each) was displayed as ΔICl1 (see Fig. 6).Figure 6Injection of hPBP protein oocytes stimulates Ca2+ release-activated Cl− conductance.Oocytes Ca2+release-activated Cl− conductance was measured as outwardly rectifying current at positive membrane potential (+40 mV). A, typical current traces corresponding to the −80 mV/+40 mV pulse combination after mock injection (1) or injection of 60 pmol of CaCl2(2), 10 ng of His-hPBP (3), or 50 pmol of GTPγS (4). Voltage protocol is shown at the top (see also "Experimental Procedures"). B, change in chloride current (ΔICl1) after injection (data: mean ± S.E.,n = 5–8 oocytes; *, p ≤ 0.05; **,p ≤ 0.01, unpaired two-tailed ttest).View Large Image Figure ViewerDownload Hi-res image Download (PPT) For generation of stable NIH3T3 cells expressing hPBP or HEK293 cells expressing rMOR1 or rMOR1 and hPBP, cells were transfected according to protocol described previously (29Koch T. Kroslak T. Averbeck M. Mayer P. Schroder H. Raulf E. Hollt V. Mol. Pharmacol. 2000; 58: 328-334Crossref PubMed Scopus (63) Google Scholar). Selection of stable clones was performed by addition of either 500 µg/ml Geneticin (pcDNA3.1/hPBP) or 1 µg/ml puromycin (pEAK10/rMOR1). For coexpression of rMOR1 and hPBP in HEK293 cells, stable rMOR1-expressing clones were transfected with pcDNA3.1/hPBP and cells selected with Geneticin and puromycin. Sequences of cloned inserts were confirmed by DNA sequencing. Control NIH3T3 cells were transfected with pcDNA3.1(−) (without cloned insert). [35S]GTPγS binding was performed as described previously (29Koch T. Kroslak T. Averbeck M. Mayer P. Schroder H. Raulf E. Hollt V. Mol. Pharmacol. 2000; 58: 328-334Crossref PubMed Scopus (63) Google Scholar). Stock [35S]GTPγS was diluted 10-fold in 50 mmTris, pH 7.4, 10 mm EDTA, aliquoted, and stored at −80 °C until use. For each assay, 28 µg of membrane protein were incubated in 20 mm HEPES, pH 7.4, 10 mmMgCl2, 100 mm NaCl, 1 mm EDTA, 3 µm GDP, 0.05 nm [35S]GTPγS. Nonspecific binding was determined in the presence of 10 µm GTPγS. After 30 min at either 4 or 20 °C, membranes mixtures were washed three times with ice-cold 50 µm Tris HCl, pH 7.4, on GF/B filters using a Inotech cell harvester. Bound radioactivity was determined by scintillation counting. For hPBP dose-response curves, 20 µg of membrane protein from HEK293 cells were incubated with increasing amounts of recombinant hPBP (0.01–1000 ng) in 50 µl at 4 °C for 90 min. Data were analyzed using GraphPad Prism (sigmoidal dose-response curve fit analysis; GraphPad Software Inc., San Diego, CA). [35S]GTPγS binding to soluble Gαi1 was performed according to Ref. 32Carty D.J. Iyengar R. Methods Enzymol. 1994; 237: 38-44Crossref PubMed Scopus (26) Google Scholar at 30 °C in 50 µl of HEPES buffer (pH 7.9, 25 mm MgCl2, 100 mm NaCl, 1 mm EDTA, 1 mm dithiothreitol) supplemented with 12.5 ng of Gαi1 or with Gαi1 and 100 ng of His-hPBP. Reaction was stopped by filtration on nitrocellulose filters at indicated time points and filters washed five times with ice-cold stop buffer (25 mm Tris-HCl, 100 mmNaCl, 25 mm MgCl2, pH 7.5). Bound radioactivity was determined by scintillation counting. 1.5 × 105cells were seeded in 22-mm 12-well dishes with Dulbecco's modified Eagle's medium Nut-F12 medium containing 10% fetal calf serum. On the day of assay, media were removed from the individual wells and were replaced with 0.5 ml of serum-free medium containing 25 µm forskolin or a combination of forskolin (25 µm) and DAMGO (1 µm). The cells were then incubated at 37 °C for 15 min. The reaction was terminated by removing the medium and sonicating the cells in 1 ml of ice-cold HCl/EtOH (1 volume of 1 n HCl/100 volumes of EtOH). After centrifugation, the supernatant was evaporated, the residue dissolved in TE buffer and cAMP content measured. X. laevis oocytes were incubated in ND96 containing 20 µmisobutylmethylxanthine for 20 min and then injected with His-hPBP and incubated for additional 20 min. Buffer was rapidly aspirated, and oocytes were sonicated and assayed for cAMP content as described above. ANdeI-EcoRI DNA fragment, containing the hPBP reading frame, was generated from pT7/T3(hPBP) by polymerase chain reaction mutagenesis and subcloned into pET28a(+). Bacterial strain BCL1(DE3) was transformed with pET28a(+)(hPBP). Cloned sequence was proofed by sequencing. 100 ml of LB medium supplemented with glucose (1% final concentration) were inoculated and grown to anA 595 nm of 0.6. Cells were induced with 1 mm isopropyl-1-thio-β-d-galactopyranoside and harvested 4 h after induction by centrifugation according to the Novagen protocol. Cell pellets were resuspended in ice-cold HIS·BIND® buffer (500 mm NaCl, 25 mm HEPES, 5 mm imidazole, pH 7.9) and sonicated. Purification of His-tagged hPBP with His·BIND® Quick 900 cartridges was performed according to Novagen's protocol and elution buffer (500 mm NaCL, 25 mm HEPES, 1 m imidazole) was used to elute hPBP. A 4-ml sample was further dialyzed five times in 2 liters of hPBP storage buffer (100 mm, 25 mm HEPES, pH 7.9). Protein concentration was determined, and aliquoted stocks (200 ng/µl) were frozen (−70 °C) in storage buffer. As reported previously, superfusion of oocytes expressing G protein-coupled inwardly rectifying potassium channels (rKir3.1; rKir3.4) with potassium solution led to an increase in potassium conductance, that could be blocked by BaCl2 (28Koch T. Kroslak T. Mayer P. Raulf E. Hollt V. J. Neurochem. 1997; 69: 1767-1770Crossref PubMed Scopus (97) Google Scholar, 29Koch T. Kroslak T. Averbeck M. Mayer P. Schroder H. Raulf E. Hollt V. Mol. Pharmacol. 2000; 58: 328-334Crossref PubMed Scopus (63) Google Scholar). This basal channel-mediated current (IKir-basal) could be further enhanced (IKir-activated) by application of a specific opioid agonist (e.g. DAMGO or morphine) to oocytes coexpressing human µ opioid receptor (hMOR) (see scheme in Fig.1). In this study we tested influence of hPBP on basal and agonist-induced Kir currents. Coexpression of hPBP, hMOR and Kir channels resulted in a significant enhancement of both IKir-basal (∼1.3-fold) and IKir-activated (∼1.6-fold) compared with oocytes expressing hMOR and Kir channels (Fig.2A). In oocytes coexpressing hPBP and Kir channels, only no effect on IKir-basal could be detected (data not shown). In addition, the EC50 values were not found to be significantly different for morphine-treated hMOR- and hMOR/PBP-expressing oocytes (∼52 and ∼63 nm, respectively) (Fig. 2 B). To evaluate whether the stimulatory effect on the Kir channel activity is restricted to the presence of hMOR, we coexpressed hPBP with either hDOR or hSSTR2. Similar to hMOR, the agonist-induced current IKir-activated in both hSSTR2- and hDOR-expressing oocytes was significantly increased after stimulation with a specific agonist (hSSTR2: octreotide, 1 µm; hDOR: DADLE, 1 µm) when hPBP was coexpressed (Fig. 2 C). In case of hSSTR2, receptor-mediated current (IKir-activated) was stimulated less efficiently (∼1.2-fold) than hDOR-mediated IKir-activated (∼2-fold). Although the extent of stimulation differed, these data indicate that the effect of hPBP on G protein-dependent coupling of receptors to Kiris not only restricted to the presence of the µ opioid receptor. To test whether the influence of hPBP on Kir-mediated conductance is directly mediated by the presence of hPBP and not due to long term expression-induced intracellular changes, purified histidine-tagged hPBP was injected inX. laevis oocytes, expressing either Kirchannels and hMOR or Kir channels only. Control oocytes were injected with heat-inactivated hPBP. Changes in IKir-basal and/or IKir-activated were measured 1–2 min after injection. PBP injection in hMOR-expressing oocytes resulted in an significant enhancement of both IKir-basaland IKir-activated as described previously in coexpression experiments. In contrast to the hPBP coexpression experiments, injection of His-hPBP was more potent to stimulate IKir-basal (1.6-fold) than IKir-activated(1.3-fold) (Fig. 3A). In addition, hPBP injection could directly stimulate basal Kirconductance (IKir-basal) ∼2.2-fold, even in the absence of a coexpressed receptor (Fig. 3 B). This effect could be blocked by PTX (60% inhibition). A comparable inhibition could be detected in PTX-treated DAMGO-stimulated oocytes expressing hMOR (85% inhibition), indicating that hPBP could stimulate Kirchannels and G protein coupling of hMOR to Kir channels in a direct and acute Gi/o protein-dependent fashion (Fig. 3 C). In order to evaluate the potency of hPBP to influence other G protein-coupled systems (e.g. adenylate cyclase), cAMP levels of oocytes injected with hPBP were compared with that of mock-injected ones. As cAMP levels varied in X. laevis oocytes isolated from different frogs (1.3–2.9 pmol of cAMP), data from independent experiments were normalized. Injection of hPBP led to an ∼38% reduction of the intracellular cAMP level (Fig.3 D). The results obtained suggest that hPBP injection may influence adenylate cyclase activity. In order to specify hPBP-mediated cAMP reduction, forskolin-induced stimulation of adenylate cyclase was studied in HEK293 and NIH3T3 cells overexpressing hPBP, rMOR1, or both hPBP and rMOR1. Forskolin stimulation of HEK293 cells stably expressing rat µ opioid receptor (rMOR1) and hPBP resulted in a significant reduction in cAMP level (55% of control) compared with control HEK293 cells expressing rMOR1 only (Fig.4A). Basal cAMP content was not found to be significantly different. In addition, coexpression of hPBP and rMOR1 also led to a significantly enhanced DAMGO (1 µm)-mediated reduction of forskolin-induced cAMP level compared with rMOR1-expressing cells (4.5 and 14% for rMOR1/hPBP- and rMOR1-expressing cells, respectively). Thus, as shown for G protein-coupled Kir conductance in Xenopusoocytes, expression of hPBP in HEK293 significantly amplified receptor-mediated G protein-coupled signaling of rMOR1, indicating a stimulatory effect of hPBP on opioid receptor-mediated inhibition of adenylate cyclase. Moreover, in mouse NIH3T3 cells stably expressing hPBP (Fig. 4 B), stimulation of adenylate cyclase with forskolin led to an significant decrease of cAMP levels (∼60%) compared with pcDNA3.1-transfected control cells, indicating a direct involvement of hPBP in reduction of adenylate cyclase-mediated cAMP formation. Basal cAMP levels were not significantly changed in both cell lines. PBPs were shown to be associated with cellular membranes and therefore could be implicated in G protein-dependent signaling in a membrane-dependent fashion. To prove the hypothesis of an involvement of hPBP in GTP-binding of G proteins, we used [35S]GTPγS-binding assay. We tested [35S]GTPγS-binding on membrane preparations of either hPBP-expressing or mock-transfected NIH3T3 cells. Binding was performed at two different temperatures (4 and 20 °C). As shown in Fig.5A, basal [35S]GTPγS binding (performed at 4 °C) to membranes of hPBP-expressing cells showed a significant ∼2.1-fold increase in specific [35S]GTPγS binding (208.8%) compared with control cells, indicating a stimulatory effect of hPBP on GTP binding to cellular membranes. A weaker but still significant increase of only ∼1.2-fold (120.2%) could be measured at room temperature (20 °C). Unspecific binding was not altered by hPBP at given temperatures (data not shown). This stimulatory effect on [35S]GTPγS binding could be also detected on membranes of untransfected HEK293 cell if recombinant His-hPBP was used for stimulation. His-hPBP dose-dependently stimulated specific [35S]GTPγS binding with an apparent EC50 of 3.6 nm (Fig.5 B). Stimulation of membrane-bound G proteins is obviously due to binding to heterotrimeric Gi proteins, as hPBP facilitated GTP binding to soluble rat Gαi1 in vitro in a time-dependent manner (Fig.5 C). Gq-coupled receptors have been shown to stimulate a Ca2+-activated chloride current inXenopus oocytes (33Quick M.W. Simon M.I. Davidson N. Lester H.A. Aragay A.M. J. Biol. Chem. 1994; 269: 30164-30172Abstract Full Text PDF PubMed Google Scholar). Increases in cytosolic Ca2+ levels result from phospholipase C/inositol trisphosphate-dependent Ca2+ release from internal stores and Ca2+ influx from the extracellular medium. To test whether hPBP could also stimulate PLC-dependent pathways, we tested effects of hPBP injection in oocytes on the Ca2+ release-activated outwardly rectifying Cl− current (ICl1) (Fig.6, A and B). GTPγS injection as well as CaCl2 injection (60 pmol) resulted in a significant stimulation of ICl1 (maximum induced change in current ΔICl1: GTPγS, ∼1100 nA; CaCl2, ∼296 nA). The CaCl2-mediated effect could be mimicked by injection of 10 ng/oocyte hPBP (ΔICl1 ∼ 210 nA), indicating that hPBP leads to an release of Ca2+ from internal stores. In previous studies PBPs were shown to affect regulation of cellular signaling and growth. They were found to associate with G proteins but failed to bind GTP themselves (26Bucquoy S. Jolles P. Schoentgen F. Eur. J. Biochem. 1994; 225: 1203-1210Crossref PubMed Scopus (41) Google Scholar). In the present study we demonstrate the involvement of hPBP in heterotrimeric G protein-dependent signaling. According to oocyte studies, hPBP acted like an enhancer of G protein-coupled receptor-mediated activation of inwardly rectifying Kir3 channels. The Kir-mediated potassium conductance is stimulated directly by free β-γ subunits of heterotrimeric G proteins (34Reuveny E. Slesinger P.A. Inglese J. Morales J.M. Iniguez-Lluhi J.A. Lefkowitz R.J. Bourne H.R. Jan Y.N. Jan L.Y. Nature. 1994; 370: 143-146Crossref PubMed Scopus (423) Google Scholar, 35Lei Q. Jones M.B. Talley E.M. Schrier A.D. McIntire W.E. Garrison J.C. Bayliss D.A. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 9771-9776Crossref PubMed Scopus (74) Google Scholar). Therefore, the observed effect of an enhanced conductance in the presence of hPBP might be explained by an increased liberation of free β-γ subunits from heterotrimeric G proteins, resulting from a direct interaction of hPBP with heterotrimeric G proteins themselves. This hypothesis is supported by the observation that stimulation was not restricted to the presence and stimulation of a specific GPCR and that direct hPBP injection resulted in a comparable PTX-sensitive stimulation even in the absence of coexpressed receptors. As hPBP mimicked the effect of an activated Gi/Go protein, it could be assumed that hPBP activates heterotrimeric G proteins. Such an activity would further explain the ability of hPBP to reduce cAMP levels in oocytes as well as HEK293 and NIH3T3 cells, since adenylate cyclase activity is inhibited by activated GTP-bound Gi/Go α subunits. In addition hPBP also seems to be involved in other G protein-dependent pathways, as hPBP was able to elicit a stimulatory effect on the Ca2+ release-activated Cl− current in oocytes comparable to that seen after CaCl2 injection. Although PLC-mediated Ca2+release in oocytes is Go-dependent, it is important to note that receptors that couple to Gq in native tissue regularly couple to Xenopus Go. In agreement with this finding, we could show that the presence of hPBP stimulates [35S]GTPγS binding to membranes in a dose-dependent manner, indicating that hPBP led to an activation of membrane-bound G proteins. PBP is a major protein component (∼18%) in detergent extracts and plasma membrane preparations of rat epididymal spermatozoa (36Jones R. Hall L. Biochim. Biophys. Acta. 1991; 1080: 78-82Crossref PubMed Scopus (50) Google Scholar). Analysis of overall protein and mRNA expression in different rat tissues (15Frayne J. Ingram C. Love S. Hall L. Cell Tissue Res. 1999; 298: 415-423Crossref PubMed Scopus (60) Google Scholar) displayed comparable amounts of hPBP in epididymis, liver, adrenal gland, and brain. PBP is predominantly expressed at highest levels in distinct cell types and specific brain regions (37Taiji M. Tohdoh N. Ojika K. J. Neurosci. Res. 1996; 45: 202-215Crossref PubMed Scopus (20) Google Scholar) including neuronal cell layers. The high efficiency (EC50 values ∼3.6 nm; ∼8.3 ng (hPBP)/20 µg (membrane protein)) of hPBP mediated [35S]GTPγS binding to membranes strikingly indicate that hPBP would play a role in membrane-associated heterotrimeric G protein-coupled signaling in this cell region. In recent years, several proteins regulating G protein activity have been described. GDP dissociation inhibitors were shown to act as negative modulators of G protein activity by inhibiting GDP release from the G protein (38De Vries L. Fischer T. Tronchere H. Brothers G.M. Strockbine B. Siderovski D.P. Farquhar M.G. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 14364-14369Crossref PubMed Scopus (134) Google Scholar, 39Natochin M. Lester B. Peterson Y.K. Bernard M.L. Lanier S.M. Artemyev N.O. J. Biol. Chem. 2000; 275: 40981-40985Abstract Full Text Full Text PDF PubMed Scopus (94) Google Scholar, 40Olofsson B. Cell. Signal. 1999; 11: 545-554Crossref PubMed Scopus (412) Google Scholar). GTPase-activating proteins accelerate GTPase activity on G proteins, thus leading to a GDP-bound inactive G protein (41Watson N. Linder M.E. Druey K.M. Kehrl J.H. Blumer K.J. Nature. 1996; 383: 172-175Crossref PubMed Scopus (479) Google Scholar, 42Settleman J. Albright C.F. Foster L.C. Weinberg R.A. Nature. 1992; 359: 153-154Crossref PubMed Scopus (253) Google Scholar). In contrast, GTP exchange factors mediate nucleotide exchange of GDP, resulting in an active GTP-bound G protein (19Overbeck A.F. Brtva T.R. Cox A.D. Graham S.M. Huff S.Y. Khosravi-Far R. Quilliam L.A. Solski P.A. Der C.J. Mol. Reprod. Dev. 1995; 42: 468-476Crossref PubMed Scopus (64) Google Scholar, 43Cismowski M.J. Ma C. Ribas C. Xie X. Spruyt M. Lizano J.S. Lanier S.M. Duzic E. J. Biol. Chem. 2000; 275: 23421-23424Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). Although the mechanism of the activation of heterotrimeric G proteins by hPBP is unclear, our observations make it reasonable to speculate that hPBP activates heterotrimeric G protein coupling by elevating GDP/GTP exchange on Gα subunits, as described for GTP exchange factors. This is in good agreement with our data showing that hPBP stimulates GTP binding to soluble Gαi1 subunit in vitro. Surprisingly, comparison between injection and expression experiments in oocytes further revealed that injection of recombinant hPBP more potently stimulated basal Kir conductance than receptor-mediated Kir conductance; moreover, in the absence of coexpressed receptors, hPBP injection resulted in a GPCR-like PTX-sensitive stimulation. These results suggest that, under given circumstances, injected hPBP lacking eukaryotic posttranslational modifications represents a functional activated form of the protein and that intracellular activity of expressed hPBP is regulated in the presence and after stimulation of given GPCRs. In the absence of such receptors, hPBP coexpression in oocytes had no effect on Kir channels (data not shown). For an interaction with the membrane-bound heterotrimeric G protein, it is necessary that hPBP is located near the plasma membrane. As PBPs were mostly found to be cytosolic, regulation of both the hPBP potency to bind to phospholipids like phosphatidylethanolamine and phosphatidylcholine (26Bucquoy S. Jolles P. Schoentgen F. Eur. J. Biochem. 1994; 225: 1203-1210Crossref PubMed Scopus (41) Google Scholar) and the targeting of PBPs to the cell membrane could lead to a modulation of the activity of heterotrimeric G proteins. Consistent with this model, Yeung and co-workers (21Yeung K. Seitz T. Li S. Janosch P. McFerran B. Kaiser C. Fee F. Katsanakis K.D. Rose D.W. Mischak H. Sedivy J.M. Kolch W. Nature. 1999; 401: 173-177Crossref PubMed Scopus (763) Google Scholar, 44Yeung K. Janosch P. McFerran B. Rose D.W. Mischak H. Sedivy J.M. Kolch W. Mol. Cell. Biol. 2000; 20: 3079-3085Crossref PubMed Scopus (325) Google Scholar) reported that both Raf1 and mitogen-activated protein/extracellular signal-regulated kinase kinase bind to a highly conserved region of hPBP described as phosphatidylethanolamine-binding domain (see Fig.7; accession no. PS01220) and that targeting of hPBP to the cell membrane occurred after mitogenic stimulation (21Yeung K. Seitz T. Li S. Janosch P. McFerran B. Kaiser C. Fee F. Katsanakis K.D. Rose D.W. Mischak H. Sedivy J.M. Kolch W. Nature. 1999; 401: 173-177Crossref PubMed Scopus (763) Google Scholar, 44Yeung K. Janosch P. McFerran B. Rose D.W. Mischak H. Sedivy J.M. Kolch W. Mol. Cell. Biol. 2000; 20: 3079-3085Crossref PubMed Scopus (325) Google Scholar). In this context it is interesting to note that a second highly conserved region (proline 112 to tyrosine 125) described as a putative nucleotide binding domain within hPBP overlaps with an amino acid region (valine 107 to leucine 123) that shares a significant homology (94%) to a sequence pattern also known as G protein-coupled receptor signature (Fig. 7) (22Banfield M.J. Barker J.J. Perry A.C. Brady R.L. Structure. 1998; 6: 1245-1254Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar, 25Schoentgen F. Jolles P. FEBS Lett. 1995; 369: 22-26Crossref PubMed Scopus (90) Google Scholar). The 17-amino acid consensus sequence signature (Prosite accession no. PS00237, Fig. 7 C) is found in the second intracellular loop of GPCRs and contains a conserved acidic-arginine-aromatic triplet (arginine conserved to 100%) shown to be important for interaction and coupling of GPCRs to heterotrimeric G proteins (45Seibold A. Dagarag M. Birnbaumer M. Receptors Channels. 1998; 5: 375-385PubMed Google Scholar, 46Zhu S.Z. Wang S.Z. Hu J. el-Fakahany E.E. Mol. Pharmacol. 1994; 45: 517-523PubMed Google Scholar). As shown by amino acid sequence analysis/alignment (Fig. 7 B), this signature as well as the arginine (signature position 13) is also highly conserved within specific PBPs. Although a proline present in PBP sequences is not found in the consensus G protein-coupled receptor signature (signature position 6), its high conservation within most animal and plant PBPs (data not shown) indicates an essential role in PBP function. This suggests that the mechanism of hPBP-mediated effects could be closely related to that elicited by GPCRs. The nature of this mechanism is currently under investigation. The identification of hPBP as a novel modulator involved in heterotrimeric G protein signaling could offer new insights into PBP function and regulation of receptor-dependent heterotrimeric G protein-coupled signaling. We thank I. Schwarz for excellent technical assistance.

Referência(s)