Real-time Analysis of Ternary Complex on Particles
2004; Elsevier BV; Volume: 279; Issue: 14 Linguagem: Inglês
10.1074/jbc.m310306200
ISSN1083-351X
AutoresPeter C. Simons, Sean M. Biggs, Anna Waller, Terry D. Foutz, Daniel F. Cimino, Qing Guo, Richard R. Neubig, Wei‐Jen Tang, Eric R. Prossnitz, Larry A. Sklar,
Tópico(s)Advanced Biosensing Techniques and Applications
ResumoWe developed a novel and generalized approach to investigate G protein-coupled receptor molecular assemblies. We solubilized a fusion protein consisting of the β2-adrenergic receptor and green fluorescent protein (GFP) for bead-based flow cytometric analysis. β2-Adrenergic receptor GFP bound to dihydroalprenolol-conjugated beads, providing a Kd for the fusion protein and, in competition with β2-adrenergic receptor ligands, Kd values for agonists and antagonists. Beads displaying chelated nickel bound purified hexahistidine-tagged G protein heterotrimers and, subsequently, the binary complex of agonist with β2-adrenergic receptor GFP. The dose-response curves of ternary complex formation revealed maximal assembly for ligands previously classified as full agonists and reduced assembly for ligands previously classified as partial agonists. Guanosine 5′-3-O-(thio)triphosphate-induced dissociation rates of the ternary complex were the same for full and partial agonists. Soluble G protein, competing with ternary complexes on beads provided an affinity estimate of agonist-receptor complexes to G protein. When performed simultaneously, the two assemblies discriminated between agonist, antagonist or inactive molecule in a manner appropriate for high throughput, small volume drug discovery. The assemblies can be further generalized to other G protein coupled receptor protein-protein interactions. We developed a novel and generalized approach to investigate G protein-coupled receptor molecular assemblies. We solubilized a fusion protein consisting of the β2-adrenergic receptor and green fluorescent protein (GFP) for bead-based flow cytometric analysis. β2-Adrenergic receptor GFP bound to dihydroalprenolol-conjugated beads, providing a Kd for the fusion protein and, in competition with β2-adrenergic receptor ligands, Kd values for agonists and antagonists. Beads displaying chelated nickel bound purified hexahistidine-tagged G protein heterotrimers and, subsequently, the binary complex of agonist with β2-adrenergic receptor GFP. The dose-response curves of ternary complex formation revealed maximal assembly for ligands previously classified as full agonists and reduced assembly for ligands previously classified as partial agonists. Guanosine 5′-3-O-(thio)triphosphate-induced dissociation rates of the ternary complex were the same for full and partial agonists. Soluble G protein, competing with ternary complexes on beads provided an affinity estimate of agonist-receptor complexes to G protein. When performed simultaneously, the two assemblies discriminated between agonist, antagonist or inactive molecule in a manner appropriate for high throughput, small volume drug discovery. The assemblies can be further generalized to other G protein coupled receptor protein-protein interactions. G protein-coupled receptors (GPCRs) 1The abbreviations used are: GPCR, G protein-coupled receptor; β2AR, β2-adrenergic receptor; GFP, green fluorescent protein; FPR-GFP, formyl peptide receptor; GTPγS, guanosine 5′-3-O-(thio)triphosphate; ARG, agonist-receptor-G protein ternary complex; DHA, dihydroalprenolol; LR, ligand-receptor complex. transmit extracellular signals into cells via intracellular G protein heterotrimers (1Bockaert J. Pin J.P. EMBO J. 1999; 18: 1723-1729Google Scholar). Of currently marketed drugs, >30% modulate GPCRs (2Wise A. Gearing K. Rees S. Drug Discov. Today. 2002; 7: 235-246Google Scholar). Only 10% of the 367 human endogenous ligand GPCRs are targeted by current drugs, leaving many future targets. Novel modes of defining the activity of ligands that bind to GPCRs could contribute to the treatment of human disease. The response to epinephrine or adrenaline is a prototypic GPCR action. Equilibrium binding studies in frog erythrocyte membranes demonstrated homogeneous binding for antagonists, whereas agonists exhibited two states of agonist affinity (3Kent R.S. De Lean A. Lefkowitz R.J. Mol. Pharmacol. 1980; 17: 14-23Google Scholar). The ternary complex model of agonist, receptor, and G protein accounts for the ternary complex exhibiting a higher agonist affinity than the binary complex (4DeLean A. Stadel J.M. Lefkowitz R.J. J. Biol. Chem. 1980; 255: 7108-7117Google Scholar). Adenylyl cyclase assays define the intrinsic activity, or efficacy, for each compound. Receptors in the high affinity state range from 50% for agonists of the lowest intrinsic activity to 95% for full agonists; the percent correlated roughly with the intrinsic adenylyl cyclase activity of the agonist. The functional consequences of cellular ternary complex formation include the rapid binding of GTP to the Gα subunit, release of the receptor and the Gβγ dimer, and exposure of new Gα and Gβγ surfaces to interact with effectors such as adenylyl cyclase (5Neer E.J. Cell. 1995; 80: 249-257Google Scholar). Ternary complex formation for a series of agonists is expected to correlate with adenylyl cyclase activities. More detailed ternary complex formulations take into account the idea that receptors can exist in different activity states (6Christopoulos A. Kenakin T. Pharmacol. Rev. 2002; 54: 323-374Google Scholar). We have previously studied the formyl peptide receptor and its numerous fluorescent ligands. The solubilized receptor forms a high agonist affinity complex with G proteins and arrestins (7Bennett T.A. Key T.A. Gurevich V.V. Neubig R. Prossnitz E.R. Sklar L.A. J. Biol. Chem. 2001; 276: 22453-22460Google Scholar, 8Bennett T.A. Foutz T.D. Gurevich V.V. Sklar L.A. Prossnitz E.R. J. Biol. Chem. 2001; 276: 49195-49203Google Scholar, 9Key T.A. Bennett T.A. Foutz T.D. Gurevich V.V. Sklar L.A. Prossnitz E.R. J. Biol. Chem. 2001; 276: 49204-49212Google Scholar). Beads derivatized with chelated nickel bind hexahistidine-tagged G protein heterotrimers, and as shown by flow cytometry, form ternary complex with formyl peptide receptor (FPR) constructs on G protein beads. The constructs included wild type receptor detected with fluorescent ligand, receptor-Gα fusion protein detected with fluorescent ligand, and receptor-GFP fusion protein detected with nonfluorescent ligand (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). The latter has the potential of being generalized. We have now extended the approach to the β2-adrenergic receptor (β2AR-GFP) using a β2AR-GFP fusion protein. We derivatized beads to discriminate GPCR assemblies sensitive to full and partial agonists and antagonists. Beads displayed dihydroalprenolol (DHA beads), following earlier work (11Caron M.G. Srinivasan Y. Pitha J. Kociolek K. Lefkowitz R.J. J. Biol. Chem. 1979; 254: 2923-2927Google Scholar). DHA beads bound detergent-solubilized β2AR-GFP with Kd ∼ 3.4 nm. Competition with agonists gave Kd values for these ligands similar to published values. Beads displaying Gαsβ1γ2 hexahistidine-tagged proteins bound agonist-β2AR-GFP binary complexes in a ternary complex. For full and partial agonists, the maximal amount of ternary complex correlated with agonist efficacy. Interestingly, ternary complexes formed by both full and partial agonists were disassembled by GTPγS at the same rate, suggesting that partial agonism depends upon ternary complex assembly rather than disassembly. The affinity of β2AR-GFP for G protein was measured by competition of soluble G protein, creating a fairly complete description of the assemblies. Moreover, these well characterized assemblies can be assayed simultaneously in a manner consistent with small volume, real time, high throughput discrimination of agonists and antagonists in a primary screen, and resolution of partial agonists in a secondary dose-response screen. Reagents and Cell Culture—All reagents were from Sigma and were of analytical quality unless otherwise noted; plasticware was from VWR. The β2-adrenergic receptor-GFP construct in pSFFV.Neo was obtained using the polymerase chain reaction, resulting in a fusion protein containing an extra AGANGAAA sequence between the final amino acid of the β-adrenergic receptor and the first amino acid of the GFP. The U937 cells were maintained, selected for high expression, expanded in spinner flasks, and frozen in aliquots as previously described (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). Membrane Preparation and Solubilization—Preparation of crude, post-nuclear membrane aliquots, and solubilization of the membrane aliquots have been described (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). A typical solubilized membrane aliquot contained the membrane proteins from 108 cells, 100–200 nm β2AR-GFP, about 500 nm Gαs and 5 mg/ml total protein in 0.25 ml of 30 mm HEPES hemisodium salt, pH 7.5, 100 mm KCl, 20 mm NaCl, 1 mm MgCl2 (HPSM) with 1% dodecyl maltoside. The quantification of β2AR-GFP by fluorescence is not absolute, as each harvest of cells may have a different percent of the protein enzymatically converted to the fully fluorescent form; a sample of hexahistidine-tagged GFP (kindly supplied by John Nolan, Los Alamos, NM) gave a molar fluorescence (quantum yield) equal to 0.7 of the molar fluorescence of carboxyfluorescein in our fluorimeter (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). Two determinations of the active formyl peptide receptor in a preparation gave a value of 50% active receptor compared with the amount given by GFP fluorescence, and we have assumed 50% active βAR to GFP fluorescence in all calculations in this report. Synthesis of DHA Beads—We have previously described the synthesis of the Ni2+ beads used in this report (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). The synthesis of the DHA beads was based on the successful affinity chromatography material developed earlier (11Caron M.G. Srinivasan Y. Pitha J. Kociolek K. Lefkowitz R.J. J. Biol. Chem. 1979; 254: 2923-2927Google Scholar). Both start with the epoxy-activation of Superdex Peptide beads, a cross-linked agarose/dextran matrix with an exclusion limit of 7000 daltons, which were extruded from a packed column purchased from Amersham Biosciences. One settled volume of the epoxy-activated beads was mixed with 1 volume of 0.2 m dithiothreitol in 0.2 m NaHCO3 for 4 h at 37 °C, then rinsed on a coarse sintered glass filter five times with water. One settled milliliter of these sulfhydryl-activated beads was stirred with 1 ml of water and 40 mg of (-)-alprenolol, and 10 μl of 10% ammonium persulfate was added every 10 min at 90 °C for 2 h (we note that this can be done at 25 °C (11Caron M.G. Srinivasan Y. Pitha J. Kociolek K. Lefkowitz R.J. J. Biol. Chem. 1979; 254: 2923-2927Google Scholar), but we obtained lower substitution), with added water to keep the volume constant. The derivatized beads were washed twice with water, once with 50% ethanol, five times with ethanol, once with 50% ethanol, once with water, and twice with HPSM. The beads were stored as a 50% slurry in HPSM with 0.02% NaN3 and 0.01% dodecyl maltoside at 4 °C. Binding Assay of β2AR-GFP to DHA Beads—Typically, 2 μl of the 50% suspension of DHA beads (∼3.5 × 105 beads/μl) was treated with 200 μl of HPSM containing 0.1% dodecyl maltoside and 0.1% bovine serum albumin at 4 °C for 30–60 min to reduce nonspecific binding. The beads were centrifuged at 1400 × gmax for 20 s, the buffer was removed, and the beads were resuspended in 50 μl of HPSM containing 0.1% dodecyl maltoside. This provided 25 aliquots of 24,000 beads for 25 binding assays. A 10-μl binding assay generally consisted of 2 μl of solublilized receptor preparation, 2 μl of a ligand at some concentration, and 4 μl of HPSM containing 0.1% dodecyl maltoside, which was mixed in a 96-well plate with a V bottom (Costar) by pipetting and allowed to equilibrate for 5 min. Then 2 μl of the treated DHA bead suspension was added and mixed by pipetting followed by orbital mixing for 2 h at 4–7 °C. Nonspecific binding was determined by the inclusion of 1 mm alprenolol. The wells were brought to 200 μl with HPSM containing 0.1% dodecyl maltoside, and their contents were transferred to 12 × 75-mm tubes immediately before flow cytometric analysis of the fluorescence on the beads. Conversion of the fluorescence measured to bound β2AR-GFP was made with calibration beads (Clontech). Kinetic binding data of β2AR-GFP binding to DHA beads were analyzed via Scientist (MicroMath™, Salt Lake City, UT). A single site binding model was utilized to fit the data, and the forward binding rate constant, kf, was evaluated with the reverse binding rate constant, kr, constrained by the equilibrium dissociation constant, Kd = kr/kf, at 3.4 nm (from experimental data). The concentration of DHA was assumed to be 0.4 nm in each 10-μl binding assay based on 100,000 binding sites per bead and 24,000 beads per assay. Agonist-Receptor-G Protein (ARG) Assay on G Protein-coated Beads—Coating of the Ni2+ beads with heterotrimeric G proteins has been described before in detail (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). Briefly, 25 pmol of the desired α subunit and 25 pmol of hexahistidine-tagged β1γ2 were mixed with 2.5 μl of a 50% slurry of dextran chelate nickel beads (2.5 × 105 beads/μl) and 190 μl of HPSM containing 0.1% dodecyl maltoside and 1 mm dithiothreitol, then kept in suspension by rocking at 4–7 °C for 1 h. The beads were then centrifuged for 30 s at 1500 × gmax, the supernatant was removed, and the beads were resuspended in 50 μl of the buffer. This provided 25 aliquots of 24,000 G protein-coated beads for 25 ARG assembly assays (some beads were lost to surfaces). A 10-μl ARG assembly assay generally consisted of 2 μl of solubilized receptor preparation, 2 μl of desired ligand, 4 μl of HPSM containing 0.1% dodecyl maltoside, and 2 μl of G protein-coated beads. Nonspecific fluorescence was defined in the presence of 0.1 mm GTPγS. These suspensions were mixed on an orbital mixer for 2 h, brought to 200 μl with HPSM containing 0.1% dodecyl maltoside, transferred to a 12 × 75-mm tube, and immediately analyzed by flow cytometry for bead fluorescence. For kinetic data tubes were removed from the cytometer after 20 s, 2 μl of 10-2m GTPγS was added and mixed, and the tubes were returned to the cytometer for measurement of the bead fluorescence. The cytometer data were converted to alphanumeric form and binned into 1-s intervals using the FACSQuery program, 2Available free from Bruce Edwards, [email protected] which provides a series of mean channel fluorescence values in an Excel file. These data were then analyzed using Prism (Graphpad Software). For multiplex analysis, the Ni2+ beads were given a red “address label” for dual bead experiments (Ni2+ and DHA beads in one well) by reacting 10 μl of a 50% slurry of dextran chelate nickel beads with 10 μl of 10-4m NHS-Texas Red™ in Me2SO (Molecular Probes) in 80 μl of phosphate-buffered saline for 10 min at 22 °C, then washing with 900 μl of 50% ethanol, twice with 100% ethanol, once with 50% ethanol, and three times with HPSM. Both types of beads were stored at 4 °C in HPSM with 0.01% dodecyl maltoside and 0.02% sodium azide for at least six months and were stable to at least one snap-freeze at -80 °C. One settled milliliter of beads (∼5 × 108 beads) could be used for ∼20,000 assays of 24,000 beads each. Binding of β2AR-GFP to DHA Beads—Based upon earlier work demonstrating affinity chromatography of β2AR (11Caron M.G. Srinivasan Y. Pitha J. Kociolek K. Lefkowitz R.J. J. Biol. Chem. 1979; 254: 2923-2927Google Scholar), we derivatized Superdex Peptide beads to display dihydroalprenolol on the end of an 18-atom linker (Fig. 1). These DHA beads are built on a matrix of cross-linked agarose/dextran 13 μm in diameter with a 7000-dalton exclusion pore size, which restricts proteins to the surface. We also produced a β2AR-GFP fusion protein and expected that after solubilization it would bind to the DHA on the surface of the beads as shown schematically in Fig. 2A, making the beads fluorescent. Membranes from U937 cells that expressed β2AR-GFP were isolated and frozen, then aliquots of the membranes were thawed and solubilized (see “Experimental Procedures”) to produce soluble β2AR-GFP in a background of other membrane proteins. Binding assays were conducted (see “Experimental Procedures”) to test the specificity of the proposed interaction (Fig. 3A). In the presence of 50 nm β2AR-GFP, ∼60,000 GFP molecules were bound per bead, whereas when the receptor was blocked with 1 mm alprenolol, only 10,000 GFP molecules were bound per bead. Specific binding is the difference between these 2 bars, or 50,000 β2AR-GFP/bead. When a fusion protein consisting of the formyl peptide receptor and GFP (FPR-GFP) was used instead of β2AR-GFP, the binding was low in both the absence and presence of alprenolol, as expected. Underivatized beads, less hydrophobic, bound even less “background” fluorescence than the derivatized beads, as expected.Fig. 2Schematic diagrams of the two positive receptor-bead interactions used in this report. The receptor-GFP fusion protein is depicted as a snakeview receptor with an oval GFP on its C terminus. A, the DHA beads will bind to the β2AR-GFP unless a ligand has occupied the binding site previously. B, Ni2+ beads were first converted to G protein beads by incubation with hexahistidine-tagged G protein heterotrimers, as described under “Experimental Procedures.” The β2AR-GFP will bind these G protein beads only when occupied by an agonist, here shown as isoproterenol (ISO).View Large Image Figure ViewerDownload (PPT)Fig. 3Characterization of the DHA bead binding assay. Binding assays were conducted as described under “Experimental Procedures.” A, specificity of the DHA bead interaction with the β2AR-GFP fusion protein, with 1 mm alprenolol (ALP) included in assays represented with striped bars. The first two assays were standard, using 50 nm β2AR-GFP; the second two assays used FPR-GFP instead of β2AR-GFP, and the third pair of assays used underivatized beads in place of DHA beads. B, binding curves were obtained by varying the amount of β2AR-GFP as shown in the absence (filled symbols) or presence (open symbols) of 1 mm alprenolol for the times indicated on the side. C, time course of β2AR-GFP binding to DHA beads; data are replotted from panel B.View Large Image Figure ViewerDownload (PPT) We then tested this interaction as a function of time of incubation of β2AR-GFP with the DHA beads using 0.3–30 nm β2AR-GFP (Fig. 3B). 1, 2, and 4 h of binding resulted in binding curves that gave a Kd = 3.4 ± 0.4 nm, which agrees with values obtained from membrane preparations (12Mukherjee C. Caron M.G. Mullikin D. Lefkowitz R.J. Mol. Pharmacol. 1976; 12: 16-31Google Scholar) and dodecyl maltoside-solubilized preparations (13Gether U. Lin S. Kobilka B.K. J. Biol. Chem. 1995; 270: 28268-28275Google Scholar). Because the reaction was time-dependent, we chose2hasa standard time. With 100,000 β2AR-GFP molecules on the surface of 24,000 beads in 10 μl, only 0.4 nm β2AR-GFP is on the beads compared with the tens of nanomolar concentrations of β2AR-GFP in solution. Analysis of the kinetic data of β2AR-GFP binding to DHA beads (Fig. 3C) resulted in values for the forward binding rate constant, kf = 2.9 ± 1 × 103m-1s-1, and for the dissociation rate constant, kr = 8.6 ± 4 × 10-6 s-1. We note that kf is ∼2 orders of magnitude lower than reported for the binding of similar size antibody Fab fragments to epitopes on cells at 4 °C (14Nolan J.P. Chambers J.D. Sklar L.A. Robinson J.P. Babcock G.F. Phagocyte Function: A Guide for Research and Clinical Evaluation. Wiley-Liss, Inc., New York1998: 19-46Google Scholar). The active (-) stereoisomer of alprenolol was used in constructing the present DHA beads, in contrast to the racemic mixture used in the original affinity medium (11Caron M.G. Srinivasan Y. Pitha J. Kociolek K. Lefkowitz R.J. J. Biol. Chem. 1979; 254: 2923-2927Google Scholar). The elution of 40–60% of the receptors from the original affinity medium may have resulted from receptors that were bound to the more weakly binding (+) stereoisomer. ARG Ternary Complex Assembly on G Protein Beads—The ARG assembly (Fig. 2B) assay was based on earlier work (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). Here, assemblies were formed in 10 μl volumes to maximize concentrations of A, R, and G protein-coated beads, then diluted for immediate flow cytometric determination of the bead fluorescence. In Fig. 4A we demonstrate that this ARG assembly requires a cognate set of agonist, receptor, and G protein-coated beads. The binding of 40 nm β2AR-GFP to G protein beads in the presence of saturating isoproterenol (1 mm) gave a total fluorescence of 80,000 β2AR-GFP/bead. When 0.1 mm GTPγS, a non-hyrolyzable GTP analog, was added in this assembly, the fluorescence was reduced to a background of about 10,000 β2AR-GFP/bead, which demonstrates that a G protein α subunit is necessary for specific fluorescence. Specific binding was defined as the difference between these two values and is referred to as ARG/bead. In the absence of the agonist isoproterenol, only background binding was observed, demonstrating the necessity for a correct ligand for the assembly. When GFP was fused to the formyl peptide receptor, background fluorescence was again obtained, demonstrating the necessity of the correct receptor for specific fluorescence. Finally, G protein beads were assembled with αi3 subunits instead of the cognate αs subunits, and these beads also gave background fluorescence (the control assembly with cognate FPR-GFP showed specific binding, indicating active G protein beads; data not shown). Thus, the cognate agonist, receptor, and G protein were all necessary to obtain the specific fluorescence, or ARG assembly. We then varied the concentration of β2AR-GFP in the presence of two selected agonists, at saturating concentrations, in the standard ARG assembly assay, as shown in Fig. 4B. It can be seen that the GFP/bead increases linearly with the concentration of β2AR-GFP for both agonists, with the full agonist, isoproterenol, showing the greatest slope. These data are consistent with a low affinity between agonist-bound β2AR-GFP and the G beads; a Kd of 0.2–0.4 μm has been found for the interaction of agonist-bound formyl peptide receptor, and Gαi3, for example (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar), and would be consistent with the present data (see also Fig. 4E). The full agonists epinephrine and norepinephrine were indistinguishable from isoproterenol, and the partial agonist dobutamine gave a line with ⅔ the slope of salbutamol (data not shown). These data suggested the presence of ligand specific conformations of binary AR complexes, with full agonists giving a maximal slope and partial agonists giving a lower slope. The amount of G protein applied to the beads before washing and introduction to the ARG assembly assay was varied next. Polyacrylamide gel electrophoresis showed that for the standard application, 1 pmol of G protein/24,000 beads, more than 80% of the applied protein bound to the beads and stayed on when the beads were resuspended (data not shown). In Fig. 4C it is shown that the amount of ARG assembly on the beads was a saturable function of the G protein applied to the beads. We believe that this represents saturation of the surface of the beads with G protein that is in the correct orientation to allow binding of the subsequently added partners, not an EC50 for ARG formation, which will be discussed later. Our standard assembly assay protocol, thus, results in 75% saturation of the surface of the G beads using 1 pmol of G protein heterotrimers per assay. The time of assembly was varied with saturating amounts of agonists in Fig. 4D, and as with the DHA bead binding, ARG assembly continued increasing past 3 h for isoproterenol and salbutamol, whereas for the weak partial agonist dobutamine, ARG assembly was maximal by 1 h. Similar results were obtained in three other experiments in which different receptor preparations and concentrations were used; in one case, the curve for salbutamol was maximal at 2 h. The amount of ARG formed with isoproterenol was always greater than that formed with salbutamol, which was always greater than that formed with dobutamine. To compare results between experiments, the time of assembly was standardized to 2 h. Thus, our standard assembly assay was 75% saturated with respect to G protein coverage on the bead, was linear with respect to β2AR-GFP past 60 nm β2AR-GFP, depended on the specific agonist used for assembly, and would increase with time past the standard 2 h if allowed to do so for isoproterenol and would usually increase past 2 h for salbutamol. Competition between G protein on the G beads and soluble G protein was used to estimate the affinity of agonist-bound β2AR-GFP for G protein in Fig. 4E. The large amount of soluble G protein used in this experiment precluded multiple determinations, but the data are consistent with Kd = 0.1 μm, similar to the 0.3–1 μm Kd of the agonist-bound formyl peptide receptor and Gαi3β1γ2 (10Simons P.C. Shi M. Foutz T. Cimino D.F. Lewis J. Buranda T. Lim W.K. Neubig R.R. McIntire W.E. Garrison J. Prossnitz E. Sklar L.A. Mol. Pharmacol. 2003; 64: 1227-1238Google Scholar). Standard ARG assemblies were made, then diluted for kinetic determination of bead fluorescence by flow cytometry, as shown in Fig. 4F. The open squares represent a sample in which the bead fluorescence was followed uninterrupted for 2 min to determine the dissociation due to dilution alone, and it is clear that there was only minimal loss of ARG over this time frame. The closed squares show the bead fluorescence when 0.1 mm GTPγS was added manually to a parallel assembly at about 25 s, and data collection was resumed at about 30 s to follow the disassembly of the ARG. A substantial loss of fluorescence occurred in the first 5 s after GTPγS addition followed by a gradual loss for the rest of the data collection. The plus symbols represent the fluorescence of an assembly that had been conducted in the presence of GTPγS and constitute background fluorescence for the experiment. Determinations of Kd for LR Dissociation and EC50 for ARG Formation—LR formation on DHA beads was competed by the addition of selected ligands, which allowed us to measure the Kd values as detailed under “Experimental Procedures” (since the beads act as a sensor, with [bound receptor] << [free receptor] < [L], the IC50 values equal Kd values to within experimental error for all the agonists). Fig. 5A shows the competition curves for these determinations, which are consistent with a single population of noninteracting binding sites, as expected. The Kd values are compared with previously reported Kd values obtained with membrane preparations (15Green S.A. Cole G. Jacinto M. Innis M. Liggett S.B. J. Biol. Chem. 1993; 268: 23116-23121Google Scholar) in Table I. The Kd for alprenolol in dodecylmaltoside solution has been reported to be 2.9 nm (13Gether U. Lin S. Kobilka B.K. J. Biol. Chem. 1995; 270: 28268-28275Google Scholar); in a separate experiment in which the concentration of receptor was reduced to 3 nm, we obtained a Kd of 1.8 nm. The Kd values for the rest of the ligands agreed with the previously reported Kd values obtained with membrane preparations to within a factor of three.Table IComparison of the Kd values for LR formation for selected ligands using DHA beads in the second column to their EC50 values for ARG assembly using G protein beads in the third column (potency) and to their values for ARG assembly using G protein beads in the fourth column The data were obtained from three experiments as shown in Fig. 5. The data in parentheses are from earlier work (15Green S.A. Cole G. Jacinto M. Innis M. Liggett S.B. J. Biol. Chem. 1993; 268: 23116-23121Google Scholar). ND, not determined.LigandLR KdARG EC50ARG assemblynmnmAlprenolol1.8 (0.1aValue is 5—20-fold lower than most other reported values for mammalian β2AR (28—31))NDNDIsoproterenol220 (68)180100%Epinephrine680 (370)28090%Norepinephrine19,000 (10,000)19,00090%Salbutamol2,300 (ND)1,20030%Dobutamine2,400 (2,300)2,60010%a Value is 5—20-fold lower than most other reported values for mammalian β2AR (28U'Prichard D.C. Bylund D.B. Snyder S.H. J. Biol. Chem. 1978; 253: 5090-5102Google Scholar—31Leysen J.E. Gommeren W. Eens A. de Chaffoy d.C. Stoof J.C. Janssen P.A. J. Pharmacol. Exp. Ther. 1988; 247: 661-670Google Scholar) Open table in a new tab ARG assembly on G beads was measured as a function of the concentration of selected agonists, and Fig. 5B displays these results. The curves were consistent with the assembly acting as a simple
Referência(s)