Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M310306200 on January 15, 2004

J. Biol. Chem., Vol. 279, Issue 14, 13514-13521, April 2, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/14/13514    most recent
M310306200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simons, P. C.
Right arrow Articles by Sklar, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simons, P. C.
Right arrow Articles by Sklar, L. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Real-time Analysis of Ternary Complex on Particles

DIRECT EVIDENCE FOR PARTIAL AGONISM AT THE AGONIST-RECEPTOR-G PROTEIN COMPLEX ASSEMBLY STEP OF SIGNAL TRANSDUCTION*

Peter C. Simons{ddagger}, Sean M. Biggs{ddagger}, Anna Waller{ddagger}, Terry Foutz{ddagger}, Daniel F. Cimino§, Qing Guo¶, Richard R. Neubig||, Wei-Jen Tang¶, Eric R. Prossnitz§, and Larry A. Sklar{ddagger}**

From the {ddagger}Cancer Center, Department of Pathology and §Department of Cell Biology and Physiology, University of New Mexico Health Sciences Center, Albuquerque, New Mexico 87131, Ben-May Cancer Research Institute, University of Chicago, Chicago, Illinois, 60637, and ||University of Michigan School of Medicine, Ann Arbor, Michigan 48109

Received for publication, September 17, 2003 , and in revised form, December 22, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
We developed a novel and generalized approach to investigate G protein-coupled receptor molecular assemblies. We solubilized a fusion protein consisting of the {beta}2-adrenergic receptor and green fluorescent protein (GFP) for bead-based flow cytometric analysis. {beta}2-Adrenergic receptor GFP bound to dihydroalprenolol-conjugated beads, providing a Kd for the fusion protein and, in competition with {beta}2-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 {beta}2-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.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
G protein-coupled receptors (GPCRs)1 transmit extracellular signals into cells via intracellular G protein heterotrimers (1). Of currently marketed drugs, >30% modulate GPCRs (2). 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 (3). The ternary complex model of agonist, receptor, and G protein accounts for the ternary complex exhibiting a higher agonist affinity than the binary complex (4). 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 G{alpha} subunit, release of the receptor and the G{beta}{gamma} dimer, and exposure of new G{alpha} and G{beta}{gamma} surfaces to interact with effectors such as adenylyl cyclase (5). 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 (6).

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 (7–9). 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-G{alpha} fusion protein detected with fluorescent ligand, and receptor-GFP fusion protein detected with nonfluorescent ligand (10). The latter has the potential of being generalized.

We have now extended the approach to the {beta}2-adrenergic receptor ({beta}2AR-GFP) using a {beta}2AR-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 (11). DHA beads bound detergent-solubilized {beta}2AR-GFP with Kd ~ 3.4 nM. Competition with agonists gave Kd values for these ligands similar to published values. Beads displaying G{alpha}s{beta}1{gamma}2 hexahistidine-tagged proteins bound agonist-{beta}2AR-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 GTP{gamma}S at the same rate, suggesting that partial agonism depends upon ternary complex assembly rather than disassembly. The affinity of {beta}2AR-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.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Reagents and Cell Culture—All reagents were from Sigma and were of analytical quality unless otherwise noted; plasticware was from VWR. The {beta}2-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 {beta}-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 (10).

Membrane Preparation and Solubilization—Preparation of crude, post-nuclear membrane aliquots, and solubilization of the membrane aliquots have been described (10). A typical solubilized membrane aliquot contained the membrane proteins from 108 cells, 100–200 nM {beta}2AR-GFP, about 500 nM G{alpha}s 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 {beta}2AR-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 (10). 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 {beta}AR 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 (10). The synthesis of the DHA beads was based on the successful affinity chromatography material developed earlier (11). 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 (11), 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 {beta}2AR-GFP to DHA Beads—Typically, 2 µl of the 50% suspension of DHA beads (~3.5 x 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 x 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 x 75-mm tubes immediately before flow cytometric analysis of the fluorescence on the beads. Conversion of the fluorescence measured to bound {beta}2AR-GFP was made with calibration beads (Clontech).

Kinetic binding data of {beta}2AR-GFP binding to DHA beads were analyzed via Scientist (MicroMathTM, 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 (10). Briefly, 25 pmol of the desired {alpha} subunit and 25 pmol of hexahistidine-tagged {beta}1{gamma}2 were mixed with 2.5 µl of a 50% slurry of dextran chelate nickel beads (2.5 x 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 x 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 GTP{gamma}S. 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 x 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 10-2 M GTP{gamma}S 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,2 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 10-4 M NHS-Texas RedTM 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 x 108 beads) could be used for ~20,000 assays of 24,000 beads each.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Binding of {beta}2AR-GFP to DHA Beads—Based upon earlier work demonstrating affinity chromatography of {beta}2AR (11), 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 {beta}2AR-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 {beta}2AR-GFP were isolated and frozen, then aliquots of the membranes were thawed and solubilized (see "Experimental Procedures") to produce soluble {beta}2AR-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 {beta}2AR-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 {beta}2AR-GFP/bead. When a fusion protein consisting of the formyl peptide receptor and GFP (FPR-GFP) was used instead of {beta}2AR-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.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 1.
Outline of the syntheses of DHA beads and Ni2+ beads. Both syntheses begin with the epoxidation of agarose/dextran beads as shown at the top. Details are given under "Experimental Procedures." DTT, dithiothreitol.

 



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 2.
Schematic 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 {beta}2AR-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 {beta}2AR-GFP will bind these G protein beads only when occupied by an agonist, here shown as isoproterenol (ISO).

 



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 3.
Characterization of the DHA bead binding assay. Binding assays were conducted as described under "Experimental Procedures." A, specificity of the DHA bead interaction with the {beta}2AR-GFP fusion protein, with 1 mM alprenolol (ALP) included in assays represented with striped bars. The first two assays were standard, using 50 nM {beta}2AR-GFP; the second two assays used FPR-GFP instead of {beta}2AR-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 {beta}2AR-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 {beta}2AR-GFP binding to DHA beads; data are replotted from panel B.

 
We then tested this interaction as a function of time of incubation of {beta}2AR-GFP with the DHA beads using 0.3–30 nM {beta}2AR-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 (12) and dodecyl maltoside-solubilized preparations (13). Because the reaction was time-dependent, we chose2hasa standard time. With 100,000 {beta}2AR-GFP molecules on the surface of 24,000 beads in 10 µl, only 0.4 nM {beta}2AR-GFP is on the beads compared with the tens of nanomolar concentrations of {beta}2AR-GFP in solution. Analysis of the kinetic data of {beta}2AR-GFP binding to DHA beads (Fig. 3C) resulted in values for the forward binding rate constant, kf = 2.9 ± 1 x 103 M-1s-1, and for the dissociation rate constant, kr = 8.6 ± 4 x 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 (14). 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 (11). 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 (10). 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 {beta}2AR-GFP to G protein beads in the presence of saturating isoproterenol (1 mM) gave a total fluorescence of 80,000 {beta}2AR-GFP/bead. When 0.1 mM GTP{gamma}S, a non-hyrolyzable GTP analog, was added in this assembly, the fluorescence was reduced to a background of about 10,000 {beta}2AR-GFP/bead, which demonstrates that a G protein {alpha} 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 {alpha}i3 subunits instead of the cognate {alpha}s 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.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 4.
Characterization of the ARG assembly on G beads. A, ARG assembly requires cognate agonist, receptor, and G protein. Assemblies on G protein-coated Ni2+ beads were conducted as described under "Experimental Procedures," except as noted, using 40 nM {beta}2AR-GFP or FPR-GFP, 1 mM isoproterenol (ISO) as the agonist, cognate G{alpha}s{beta}1{gamma}2 (s) or noncognate G{alpha}i3{beta}1{gamma}2 (i) heterotrimer, and 0.1 mM GTP{gamma}S, as indicated. B, the ligand used and the amount of {beta}2AR-GFP used in the standard assembly were varied as shown using 1 mM isoproterenol or salbutamol (SAL) for agonists and 0.1 mM GTP{gamma}S where indicated. C, the amount of G protein used to coat the Ni2+ beads was varied before conducting the standard ARG assembly. D, the time of assembly was varied as shown using 40 nM {beta}2AR-GFP and 1 mM isoproterenol, salbutamol, or dobutamine (DOB). E, soluble G{alpha}s{beta}1{gamma}2 heterotrimer was added to standard assays to compete with the bead-borne G{alpha}s{beta}1{gamma}2; 0.1 mM GTP{gamma}S was added as indicated. F, kinetic dissociation of ARG by the manual addition of 0.1 mM GTP{gamma}S. Open squares show the standard assembly with the fluorescence of the beads followed uninterrupted. The plus symbols show the assembly in which GTP{gamma}S was present during the entire assembly. The filled squares show a standard assembly in which the bead fluorescence was followed for 20 s, the tube was removed from the cytometer, 0.1 mM GTP{gamma}S was added manually and mixed at 25 s, the tube was returned to the cytometer, and bead fluorescence was followed for the remaining time.

 
We then varied the concentration of {beta}2AR-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 {beta}2AR-GFP for both agonists, with the full agonist, isoproterenol, showing the greatest slope. These data are consistent with a low affinity between agonist-bound {beta}2AR-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 G{alpha}i3, for example (10), 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 2/3 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 {beta}2AR-GFP past 60 nM {beta}2AR-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 {beta}2AR-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 G{alpha}i3{beta}1{gamma}2 (10).

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 GTP{gamma}S 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 GTP{gamma}S 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 GTP{gamma}S 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 (15) in Table I. The Kd for alprenolol in dodecylmaltoside solution has been reported to be 2.9 nM (13); 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.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 5.
Determination of Kd values for LR formation and EC50 values for ARG assembly (potency) for selected ligands. A, DHA beads were used in the standard binding assay with 9 nM {beta}2AR-GFP. The ligands were allowed to block the receptor for 5 min before the DHA beads were added and mixed for 2 h. B, G protein-coated beads were used in the standard ARG assembly assay with 20 nM {beta}2AR-GFP and agonists at the concentrations shown. ALP, alprenolol; ISO, isoproterenol; EPI, epinephrine; NE, norepinephrine; DOB, dobutamine; L, ligand; SAL, salbutanol.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Comparison 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 (15). ND, not determined.

 
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 dose-response with respect to the concentration of each ligand. EC50 values obtained from these data, or potencies of the agonists, are reported in Table I. Although precise potencies depend on the specific G protein heterotrimers in a given preparation, the rank order of potency for these ligands agrees with membrane data (12). We noticed also the appearance of partial agonism, in which the curves for salbutamol and dobutamine each climbed to ARG/bead levels that were 30 and 10 percent, respectively, of the level reached by the full agonist isoproterenol. Although precise efficacies depend on the specific G protein heterotrimers in a preparation, the rank order of ARG formation of these ligands agrees with membrane efficacy data (12).

Kinetics of ARG Disassembly by GTP{gamma}S—Ternary complexes were assembled as above using the full agonist isoproterenol and the partial agonists salbutamol and dobutamine. These were analyzed by flow cytometry as described in the legend to Fig. 6. The rate of disassembly of each ternary complex was followed after the addition of GTP{gamma}S. All three ligands gave the same kinetic rate for this process, 0.09 s-1, within experimental error. Mechanistically, this rate-determining step of ARG disassembly is therefore independent of the affinity of the ligand. Based on previous results with the formyl peptide receptor (10), it most likely arises from the dissociation of AR from G{alpha}:GTP{gamma}S rather than dissociation of G{alpha} from G{beta}{gamma}.



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 6.
Full and partial agonist ARG complexes are activated by GTP{gamma}S at the same rate. ARG assemblies were made as described under "Experimental Procedures," then diluted and applied to a flow cytometer for measurement of bead fluorescence. At 20 s, each tube was removed from the cytometer, 0.1 mM GTP{gamma}S was added and mixed, and the tube was returned to the cytometer again. The first dot in each panel is the average of the first 20 s of data and is positioned in the figure where the GTP{gamma}S was mixed with the sample. The lines are fits to a single exponential decay. ISO, isoproterenol; SAL, salbutamol; DOB, dobutamine; MCF, mean channel fluorescence.

 
Simultaneous Determination of Agonist and Antagonist— Results with the DHA beads and the G protein beads suggested that one could obtain data from a single 10-µl mixture of both beads in one well sharing the same receptor and ligand, which would determine whether the ligand was an agonist, an antagonist, or inactive. To this end, a sample of Ni2+ beads was reacted with activated Texas RedTM (see "Experimental Procedures"), then coated with G proteins using the standard protocol. We verified that labeling the beads did not affect their behavior (data not shown). These labeled beads allowed us to observe molecular assemblies in duplex form, with DHA beads incubated in the same well as the Texas RedTM-colored G protein beads. In this case the flow cytometer separated the green fluorescence of the red beads from the green fluorescence of the colorless beads. The results for DHA beads in duplex form were similar to the results in standard form (data not shown), and the results for the Texas RedTM-colored G protein beads in duplex form were the same as G protein beads incubated in separate wells. This duplex data is shown in the form in which the data would be obtained in a screen for active ligands (Fig. 7). The inactive compound allows {beta}2AR-GFP binding to DHA beads but not ARG formation; the agonist prevents {beta}2AR-GFP binding to DHA beads and allows ARG formation, and an antagonist blocks {beta}2AR-GFP binding to DHA beads and does not promote ARG formation.



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 7.
Simultaneous determination of agonist, antagonist, and inactive compounds using duplex flow cytometry. G protein beads colored with Texas RedTM (G-beads) and DHA beads were mixed with 20 nM {beta}2AR-GFP in the absence of ligand, the presence of 1 mM isoproterenol, or the presence of 1 mM ALP. The cytometer determined the green fluorescence of the uncolored beads separately from the green fluorescence of the red-colored beads for each well. Data from the uncolored DHA beads are represented as striped bars; data from the colored G protein beads in the same wells are represented as filled bars. MCF, mean channel fluorescence.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Solubilization of GPCRs and Cytometric Display—Both the FPR-GFP and {beta}2AR-GFP have now been successfully displayed in solubilized form on particles for flow cytometric analysis. Although three different constructs of the FPR have been detected (both by GFP and fluorescent ligands), the most general approach to GPCR display is likely to be based on GFP fusion proteins. Although solubilization has proven to be a key element in characterizing GPCR assemblies (7–9), the retention of GPCR binding activity has proven elusive and largely a matter of trial and error. Particle display, an analog of the methods used in routine protein purification, leads to the possibility of a general solubilization approach, in which different detergents are screened to solubilize receptor-GFP fusion proteins, and the retention of activity is assessed using the criterion of specific assembly on the particles by flow cytometry.

Ligand Beads—To display the {beta}2AR ligand, DHA, we used an affinity chromatography support (11). The {beta}2-adrenergic receptor-GFP construct bound saturably to DHA beads with Kd = 3.4 nM, which agrees with data from membrane preparations (12) and solubilized preparations (13). The binding of {beta}2AR-GFP to the DHA beads was specific for the cognate receptor, was blocked with cognate ligand, and was insensitive to GTP{gamma}S. {beta}2AR-GFP binding to DHA beads was slower than the binding of antibody Fab fragments of comparable size to a cell epitope. This may have been because a Fab fragment binds its epitope in a lock and key manner, whereas a receptor binds its ligand with induced conformational changes. The rank order of potency of selected {beta}2-adrenergic agonists was the same as reported previously (15). The derived Kd values were within a factor of three of those shown earlier, showing that the receptor preserves its natural ligand interaction. The Kd value for alprenolol in competition was 1.8 nM, similar to the Kd for DHA on the bead of 3.5 nM.

G Protein Beads—To display G proteins we used beads bearing chelated nickel (10), which bound purified hexahistidine-tagged G protein heterotrimers (G{alpha}s{beta}1{gamma}2) and recognized {beta}2AR. The assembly specificity appears to have been governed by the {alpha} subunit, whereas the {beta}1{gamma}2 may not have been the optimal choice (16). The Kd for ARG assembly was determined to be about 0.1 µM G protein by competition with free G heterotrimer in the presence of saturating isoproterenol. The time of assembly, far longer than assembly in a cell, was not unexpected based on receptors and G protein diluted many times compared with their state in membranes. It is noteworthy that, taken together, ligand beads and G protein beads with GPCR-GFPs provide essentially the same level of detail normally available through GPCR radioligand binding analysis.

Partial Agonists and Their Assays—The physiological response to an adrenergic ligand depends on the types and levels of expression of receptor, G protein subunits, adenylate cyclase isozymes, and cyclic nucleotide-dependent enzymes in various tissues plus the adrenergic tone of the organism. Our results are analogous to classical studies of radioligand binding to {beta}2AR in membranes (3), which showed two distinct binding states for agonists, one of low affinity (AR) and one of high affinity (ARG). The maximal effect of an agonist for adenylyl cyclase activity (efficacy) was compared with that for the full agonist isoproterenol, which was set to 100%, whereas a partial agonist produced an efficacy of less than 100%. In membrane binding assays isoproterenol induced a high amount of ARG (80% ARG form) and adenylyl cyclase activity (defined as 100%). The percent of R in the ARG form was found to correlate with the efficacy of an agonist. Agonists ranged from a low of 58% ARG form with an efficacy of 8% (soterenol) to a high of 93% ARG form with an efficacy of 110% (hydroxybenzylisoproterenol) (3). Our present approach with solubilized components provides a more direct measurement of ARG assembly and can be compared with the adenylyl cyclase efficacies obtained by others. Using 20 nM {beta}2AR-GFP, saturating concentrations of the partial agonists dobutamine and salbutamol produced 10 and 30% ARG assembly in our assay compared with the ARG assembly produced by isoproterenol (defined as 100% ARG assembly). Measured adenylyl cyclase efficacies were 30 and 80% for dobutamine and salbutamol, respectively (13). Our results with the chosen ligands, thus, mirror membrane data in three ways; binding affinities correlate well, EC50 values for ARG formation, which contain different G proteins in different studies, show the same rank order of potency, and partial agonists, which could be masked by different levels of R to G protein, display the same rank order of ARG assembly as found for adenylyl cyclase efficacy.

(Eq. 1)

Mechanism of Partial Agonism—Under conditions of agonist saturation receptors exist predominantly either as AR or ARG complexes, and the ternary complex steps are represented simplistically as Equation 1. The overall rate of activation of G protein depends upon both the rate of assembly of ARG and the rate of activation with saturating GTP. In our hands, the kinetics of ARG assembly depend on the nature of the agonist used, whereas the kinetics of GTP{gamma}S-induced disassembly of AR from ARG do not. Our data do not directly address the rate of G protein activation nor have we resolved the association and dissociation rates of the ternary complex. The assembly data suggest that AR complexes formed by partial agonists have a lower affinity for G than AR complexes formed by full agonists; the approach to ternary complex equilibrium is faster for the partial agonist dobutamine than for isoproterenol, and there is a lower amount of ternary complex formed. Our data are consistent with evidence for agonist-induced conformational states in the {beta}2AR (13, 17) and with preliminary modeling using the ternary complex model. Our results may be comparable to membrane systems, because it is difficult to imagine that the underlying rate constants would change between membrane and soluble systems while conserving affinity, potency, and maximal agonist effects.

Our results agree with a previous study of the muscarinic acetylcholine receptor, which showed indirectly that the reduced efficacy of a partial agonist was the result of a decrease in affinity of the agonist-receptor complex for G protein (18). Our results do not directly address another previous study of the {beta}2AR, which showed indirectly that there were different rates of heterotrimer activation for partial agonists compared with full agonists. That study used the long splice variant of {alpha}s, compared with the short splice variant used here (19).

Although likely to reflect insufficient G protein activation, partial agonism in cells could be masked by spare receptors (20) that allow saturation of G protein activation without full receptor occupancy. Thus, molecular tests for partial agonists are physiologically significant. Partial {beta}2-adrenoreceptor agonists have been useful in the treatment of asthma, where there is substantial evidence that use of high dose formulations of full agonists, taken with inhalers to relax airway smooth muscle, resulted in epidemics of mortality (21). Intrinsic efficacies of agonists have been difficult to measure because both the concentration dependence and the maximal effect of an agonist depends on the density of receptors in tissues (20). The high density of adrenoreceptors in airway smooth muscle results in maximal relaxation despite salbutamol partial agonism, whereas the lower receptor density in nontarget tissues limits cardiac and metabolic side effects (22). A second possible benefit of partial agonists is that they induce less desensitization than full agonists, thus allowing a patient to use the inhaler many times without diminished effect (23, 24).

The ARG assembly assay, at high doses of ligands, is able to predict partial as well as full agonists, since it is performed under conditions of receptor excess in which the G protein beads serve essentially as sensors. The components of the assay cost less than 2 cents except for the {beta}2AR-GFP, for which the fetal calf serum costs about 5 cents per assay, and purified G proteins. The receptor-GFP construct retains all physiological binding properties that we have measured.

Discrimination of Agonists and Antagonists—Multiplexing of flow cytometric data gives multiple determinations from one sample well, thus having each constituent determination performed under identical conditions. We used both types of beads described in this report together in a duplex (the simplest form of multiplex) assay to simultaneously determine whether a test compound was an agonist, antagonist, or neither. It is noteworthy that a partial agonist could also be determined if a full agonist response were already in the data set or by a doseresponse analysis of ARG assembly in a secondary screen. Our data only use a single G protein heterotrimer, but different heterotrimers could be put onto beads with different addresses as a suspension array (25), enabling one to scan various classes of G proteins simultaneously for interaction. This has been done recently for the 5-hydroxytryptamine receptor using a more complex antibody capture procedure (26). Multiplexed analysis using HyperCytTM, an automated system capable of sampling up to 100 samples of about 1 µl from multiwell plates per minute (27), provides the potential for simultaneously discriminating agonists and antagonists for high throughput flow cytometric drug discovery. HyperCytTM could also make possible high throughput screening of detergents giving solubilization of active GPCR-GFPs and could be applied to numerous other GPCR or other molecular assemblies.


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grants GM60799 and EB00264 (to L. A. S.), HL-476417 (to R. R. N.), and GM53459 (to W.-J. Tang). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

** To whom correspondence should be addressed. Tel.: 505-272-6892; Fax: 505-272-6995; E-mail: lsklar{at}salud.unm.edu.

1 The abbreviations used are: GPCR, G protein-coupled receptor; {beta}2AR, {beta}2-adrenergic receptor; GFP, green fluorescent protein; FPR-GFP, formyl peptide receptor; GTP{gamma}S, guanosine 5'-3-O-(thio)triphosphate; ARG, agonist-receptor-G protein ternary complex; DHA, dihydroalprenolol; LR, ligand-receptor complex. Back

2 Available free from Bruce Edwards, bedwards{at}salud.unm.edu. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Bockaert, J., and Pin, J. P. (1999) EMBO J. 18, 1723-1729[CrossRef][Medline] [Order article via Infotrieve]
  2. Wise, A., Gearing, K., and Rees, S. (2002) Drug Discov. Today 7, 235-246[CrossRef][Medline] [Order article via Infotrieve]
  3. Kent, R. S., De Lean, A., and Lefkowitz, R. J. (1980) Mol. Pharmacol. 17, 14-23[Abstract/Free Full Text]
  4. DeLean, A., Stadel, J. M., and Lefkowitz, R. J. (1980) J. Biol. Chem 255, 7108-7117[Abstract/Free Full Text]
  5. Neer, E. J. (1995) Cell 80, 249-257[CrossRef][Medline] [Order article via Infotrieve]
  6. Christopoulos, A., and Kenakin, T. (2002) Pharmacol. Rev. 54, 323-374[Abstract/Free Full Text]
  7. Bennett, T. A., Key, T. A., Gurevich, V. V., Neubig, R., Prossnitz, E. R., and Sklar, L. A. (2001) J. Biol. Chem. 276, 22453-22460[Abstract/Free Full Text]
  8. Bennett, T. A., Foutz, T. D., Gurevich, V. V., Sklar, L. A., and Prossnitz, E. R. (2001) J. Biol. Chem. 276, 49195-49203[Abstract/Free Full Text]
  9. Key, T. A., Bennett, T. A., Foutz, T. D., Gurevich, V. V., Sklar, L. A., and Prossnitz, E. R. (2001) J. Biol. Chem. 276, 49204-49212[Abstract/Free Full Text]
  10. Simons, 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., and Sklar, L. A. (2003) Mol. Pharmacol. 64, 1227-1238[Abstract/Free Full Text]
  11. Caron, M. G., Srinivasan, Y., Pitha, J., Kociolek, K., and Lefkowitz, R. J. (1979) J. Biol. Chem. 254, 2923-2927[Free Full Text]
  12. Mukherjee, C., Caron, M. G., Mullikin, D., and Lefkowitz, R. J. (1976) Mol. Pharmacol. 12, 16-31[Abstract/Free Full Text]
  13. Gether, U., Lin, S., and Kobilka, B. K. (1995) J. Biol. Chem. 270, 28268-28275[Abstract/Free Full Text]
  14. Nolan, J. P., Chambers, J. D., and Sklar, L. A. (1998) in Phagocyte Function: A Guide for Research and Clinical Evaluation (Robinson, J. P., and Babcock, G. F., eds) pp. 19-46, Wiley-Liss, Inc., New York
  15. Green, S. A., Cole, G., Jacinto, M., Innis, M., and Liggett, S. B. (1993) J. Biol. Chem. 268, 23116-23121[Abstract/Free Full Text]
  16. Robillard, L., Ethier, N., Lachance, M., and Hebert, T. E. (2000) Cell. Signal. 12, 673-682[CrossRef][Medline] [Order article via Infotrieve]
  17. Ghanouni, P., Gryczynski, Z., Steenhuis, J. J., Lee, T. W., Farrens, D. L., Lakowicz, J. R., and Kobilka, B. K. (2001) J. Biol. Chem. 276, 24433-24436[Abstract/Free Full Text]
  18. Tota, M. R., and Schimerlik, M. I. (1990) Mol. Pharmacol. 37, 996-1004[Abstract]
  19. Krumins, A. M., and Barber, R. (1997) Mol. Pharmacol. 52, 144-154[Abstract/Free Full Text]
  20. Kenakin, T. (2002) Nat. Rev. Drug Discov. 1, 103-110[CrossRef][Medline] [Order article via Infotrieve]
  21. Barrett, T. E., and Strom, B. L. (1995) Am. J. Respir. Crit. Care Med.151, 574-577[Medline] [Order article via Infotrieve]
  22. Bremner, P., Siebers, R., Crane, J., Beasley, R., and Burgess, C. (1996) Chest 109, 957-962[Medline] [Order article via Infotrieve]
  23. Clark, R. B., Knoll, B. J., and Barber, R. (1999) Trends Pharmacol. Sci. 20, 279-286[CrossRef][Medline] [Order article via Infotrieve]
  24. January, B., Seibold, A., Whaley, B., Hipkin, R. W., Lin, D., Schonbrunn, A., Barber, R., and Clark, R. B. (1997) J. Biol. Chem. 272, 23871-23879[Abstract/Free Full Text]
  25. Nolan, J. P., and Sklar, L. A. (1998) Nat. Biotechnol. 16, 633-638[CrossRef][Medline] [Order article via Infotrieve]
  26. Cussac, D., Newman-Tancredi, A., Duqueyroix, D., Pasteau, V., and Millan, M. J. (2002) Mol. Pharmacol. 62, 578-589[Abstract/Free Full Text]
  27. Waller, A., Simons, P., Prossnitz, E. R., Edwards, B. S., and Sklar, L. A. (2003) Comb. Chem. High Throughput Screen 6, 389-397[Medline] [Order article via Infotrieve]
  28. U'Prichard, D. C., Bylund, D. B., and Snyder, S. H. (1978) J. Biol. Chem. 253, 5090-5102[Abstract/Free Full Text]
  29. Bylund, D. B. (1978) Brain Res. 152, 391-395[CrossRef][Medline] [Order article via Infotrieve]
  30. Staehelin, M., Simons, P., Jaeggi, K., and Wigger, N. (1983) J. Biol. Chem. 258, 3496-3502[Abstract/Free Full Text]
  31. Leysen, J. E., Gommeren, W., Eens, A., de Chaffoy, d. C., Stoof, J. C., and Janssen, P. A. (1988) J. Pharmacol. Exp. Ther. 247, 661-670[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Brief Funct Genomic ProteomicHome page
J. P. Nolan and L. Yang
The flow of cytometry into systems biology
Brief Funct Genomic Proteomic, July 4, 2007; (2007) elm011v1.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/14/13514    most recent
M310306200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Simons, P. C.
Right arrow Articles by Sklar, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Simons, P. C.
Right arrow Articles by Sklar, L. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement