• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. May 1, 2007; 104(18): 7682–7687.
Published online Apr 23, 2007. doi:  10.1073/pnas.0611448104
PMCID: PMC1863461

A monomeric G protein-coupled receptor isolated in a high-density lipoprotein particle efficiently activates its G protein


G protein-coupled receptors (GPCRs) respond to a diverse array of ligands, mediating cellular responses to hormones and neurotransmitters, as well as the senses of smell and taste. The structures of the GPCR rhodopsin and several G proteins have been determined by x-ray crystallography, yet the organization of the signaling complex between GPCRs and G proteins is poorly understood. The observations that some GPCRs are obligate heterodimers, and that many GPCRs form both homo- and heterodimers, has led to speculation that GPCR dimers may be required for efficient activation of G proteins. However, technical limitations have precluded a definitive analysis of G protein coupling to monomeric GPCRs in a biochemically defined and membrane-bound system. Here we demonstrate that a prototypical GPCR, the β2-adrenergic receptor (β2AR), can be incorporated into a reconstituted high-density lipoprotein (rHDL) phospholipid bilayer particle together with the stimulatory heterotrimeric G protein, Gs. Single-molecule fluorescence imaging and FRET analysis demonstrate that a single β2AR is incorporated per rHDL particle. The monomeric β2AR efficiently activates Gs and displays GTP-sensitive allosteric ligand-binding properties. These data suggest that a monomeric receptor in a lipid bilayer is the minimal functional unit necessary for signaling, and that the cooperativity of agonist binding is due to G protein association with a receptor monomer and not receptor oligomerization.

Keywords: β2-adrenergic receptor, single-molecule spectroscopy, oligomerization

Oligomerization appears as a common theme for numerous integral membrane proteins. Although the role of protein oligomerization is clear for some proteins such as ion channels (13) and receptor tyrosine kinases (4), the contribution of oligomerization to G protein-coupled receptor (GPCR) function has become a topic of debate (57). The most compelling case exists for the GABAB receptor (810), where heterodimerization is required for both plasma membrane targeting and G protein activation. Although plasma membrane targeting has been attributed to oligomerization for some Class A GPCRs (11, 12), a more comprehensive functional significance has yet to be discovered. Considering that many GPCRs form physical oligomeric interactions (1315), including the provocative demonstration that rhodopsin exists as arrays of dimers (5, 16), it seems plausible that GPCRs may function optimally as oligomers. Functional studies on the GABAB receptor (810), as well as biophysical and biochemical evidence from Class A receptors including rhodopsin (6, 17) and the leukotriene B4 receptor (18), suggest that a pentameric complex, consisting of a GPCR dimer and a G protein heterotrimer, is required for efficient G protein activation.

However, the fundamental question as to whether a monomeric GPCR is capable of coupling to a G protein in a membrane environment arises from these studies. Efficient coupling should be reflected by receptor-mediated allosteric changes in the G protein structure that result in nucleotide exchange and mutual G protein-dependent effects on receptor affinity for agonists. To address this, we rely on a unique approach for isolating GPCR monomers within high-density lipoproteins (HDL). HDL, composed of a dimer of apolipoprotein A-I (apoA-I) surrounding a planar bilayer of ≈160 phospholipids, is easily reconstituted in vitro [reconstituted HDL (rHDL)] (19, 20). EM of these particles highlight the disk-shaped structure (≈10–12 nm in diameter and thickness of ≈40 Å, the same thickness of a plasma membrane; Fig. 1). This reconstitution system has been elegantly adapted by the Sligar laboratory to incorporate various membrane proteins into the phospholipid bilayer, including bacteriorhodopsin (21), cytochrome P450 (22) and the β2-adrenergic receptor (β2AR) (23). In this report, we use an HDL-based reconstitution system to incorporate purified β2AR and its cognate G protein, Gs. We demonstrate that β2AR reconstituted in HDL (β2AR·rHDL) is monomeric and fully functional by virtue of its capacity to support both high-affinity agonist binding and rapid agonist-mediated nucleotide exchange.

Fig. 1.
Depiction of rHDL particles. (a) Transmission electron micrograph of negatively stained rHDL. Well defined 10-nm rHDL particles are clearly visible. (Scale bar, 100 nm.) (b) Molecular model illustrating rHDL composed of a dimer of apoA-I proteins wrapped ...


Reconstitution of β2AR into HDL.

A variety of conditions were examined for incorporating purified cyan fluorescent protein (CFP) fused-β2AR fusion protein (CFP-β2AR) or WT-β2AR (β2AR) into HDL disks. By reconstituting CFP-β2AR into a mixture of lipids [palmitoyl-oleoyl-phosphatidylcholine (POPC) and palmitoyl-oleoyl-phosphatidylglycerol] and apoA-I, we obtained >98% recovery of [3H]dihydroalprenolol ([3H]DHAP)-binding activity (Fig. 2a). CFP-β2AR·rHDL exhibits binding affinities for antagonists and agonists with Ki values that are more consistent with those observed for β2AR in biological membranes rather than in detergent micelles, as demonstrated in Fig. 2b and Fig. 5 c and d. β2AR and fluorophore-labeled receptors (see below) exhibits similar properties (data not shown).

Fig. 2.
Functional reconstitution of β2AR into rHDL. (a) β2AR requires apoA-I and lipids to survive detergent removal with high efficiency. Equal amounts of DDM-solubilized β2AR were included in the rHDL reconstitution assay with or without ...
Fig. 5.
Monomeric β2AR incorporated in rHDL particles couples efficiently to G proteins. (a) ISO-induced [35S]GTPγS binding to monomeric-β2AR·Gs·rHDL particles. β2AR·Gs·rHDL particles were preincubated ...

Single-Molecule Spectroscopy and FRET Confirm That β2AR in rHDL Is Monomeric.

The inner diameter of an HDL particle is estimated to be ≈85 Å (24); therefore, we predict that a maximum of two receptors can possibly fit within an rHDL [the receptor diameter is ≈40 Å, as determined from the rhodopsin crystal structure (25)]. Moreover the conditions used for the initial stage of the β2AR reconstitution (i.e., vast excess of apoA-I:receptor, ratio 100:1, or HDL:receptor of 50:1) favor the incorporation of a single receptor per rHDL.

To make a more definitive assessment of the β2AR:rHDL stoichiometry, we used total internal reflection fluorescence (TIRF) microscopy to image single molecules of fluorescently tagged β2AR·rHDL (Cy3 or Cy5). The degree of Cy3 and Cy5 label colocalization is related to the fraction of rHDL particles containing two or more receptors, and the fraction of monomeric receptors.

We have determined labeling conditions such that >99% of β2ARs are labeled with either Cy3 or Cy5 in detergent micelles [see supporting information (SI) Fig. 6]. In addition, to confirm that the receptor preparations are not oligomeric in detergent micelles before labeling and reconstitution, we used a crosslinking approach with bifunctional amine-reactive reagents (see SI Fig. 7). Although extensive cross-linking occurred when Cy3-β2AR was reconstituted in phospholipid vesicles, very little intermolecular cross-linking was observed in detergent micelles. These data suggest that receptors are not proximal enough to be crosslinked in detergent micelles using either 11- or 22-Å crosslinkers and are presumably monomeric. The homogeneity and monodispersion are important properties of these preparations that are relevant to TIRF imaging and FRET analysis below.

Preparations of Cy3- and Cy5-β2AR were reconstituted into rHDL and resolved by size-exclusion chromatography (SEC). Both Cy3- and Cy5-β2AR·rHDL elute as a uniform absorbance peak (Fig. 2c) with a Stokes' diameter of 11 nm (as determined by protein standards), slightly larger than the diameter for empty discs alone (≈10.5 nm; not shown). This small difference, likely because of the extramembrane loops and termini of the receptor, demonstrates that incorporation of β2AR does not perturb the structure or stability of the rHDL particle. Fractions representing the chromatogram peak were isolated and used for TIRF and FRET analysis (marked with an asterisk Fig. 2 c and Inset). These fractions represent both the peak in fluorescence intensity and also [3H]DHAP-binding activity (illustrated for Cy3-β2AR·rHDL).

TIRF imaging of either Cy3-β2AR·rHDL (Fig. 3a) or Cy5-β2AR·rHDL (Fig. 3b) reveals discrete monodispersed fluorescent particles. Subtle variability in intensity is due to incorporation of multiple fluorophores per molecule of β2AR (see step photobleaching in SI Fig. 6). When equal amounts of Cy3-β2AR·rHDL and Cy5-β2AR·rHDL were mixed together and imaged, only 1.8 ± 0.6% displayed colocalization (average ± SEM of 10 images, total of 1,916 molecules; Fig. 3 c and f). Strikingly, when equal amounts of Cy3-β2AR and Cy5-β2AR receptors were mixed before reconstitution in HDL, only 2.3 ± 0.6% of Cy3- and Cy5-labeled receptors colocalized (10 images containing a total of 2,022 receptors; Fig. 3 d and f). The degree of colocalization of these fluorophores does not differ from mixtures of Cy3-β2AR·rHDL and Cy5-β2AR·rHDL (Fig. 3f) or from TIRF imaging of Cy3-β2AR or Cy5-β2AR in detergent micelles (data not shown). Both percentages are highly statistically different from a positive control for colocalization (double-labeled Cy3-Cy5-β2AR, 44 ± 2% of which colocalized, i.e., fluoresced in both the Cy3 and Cy5 channels; Fig. 3 e and f). These data strongly support the notion that, under the conditions used for reconstitution, each HDL particle contains only a single monomeric β2AR molecule.

Fig. 3.
TIRF of Cy3- and Cy5-labeled β2AR in rHDL reveals that the vast majority of β2AR are monomeric. Conditions for the reconstitutions illustrated are: (a) Cy3-β2AR·rHDL and (b) Cy5-β2AR·rHDL particles alone, ...

Single-molecule experiments were confirmed by ensemble FRET measurements to assess the relative physical distance between Cy3- and Cy5-labeled receptors incorporated into rHDL. Because the inner diameter of rHDL (≈85 Å) is close to the Förster distance for Cy3 and Cy5 (≈60 Å), FRET should only occur from Cy3- to Cy5-labeled receptors within the same rHDL particle. However, no FRET was observed from rHDL particles formed from a mixture of Cy3-β2AR and Cy5-β2AR under conditions described above (Fig. 4b) and remained unchanged when resolubilized in 1% dodecylmaltoside (DDM) and comparable to mixtures of Cy3-β2AR and Cy5-β2AR in detergent micelles (Fig. 4c). In contrast, FRET was observed when Cy3-β2AR and Cy5-β2AR were reconstituted in phospholipid vesicles (i.e., in the absence of apoA-I; Fig. 4a), suggesting that these receptors are oligomeric. The addition of 1% DDM to the vesicles markedly reduced the FRET (Fig. 4a). These data are consistent with crosslinking studies with bifunctional amine crosslinkers in vesicles (see SI Fig. 7). Taken together, these data demonstrate that rHDL particles represent a unique experimental system for studying monomeric GPCRs in a phospholipid environment.

Fig. 4.
FRET measurements confirm monomeric β2AR in rHDL. Cy3- and Cy5-labeled β2AR preparations were reconstituted under different conditions, and normalized spectra were analyzed for the presence of FRET, as indicated by increased acceptor (Cy5) ...

Monomeric β2AR in rHDL Functionally Couples to G Proteins.

Monomeric β2ARs in rHDL couple efficiently to the purified stimulatory heterotrimeric G protein Gs, as shown in Fig. 5(see Materials and Methods). Isoproterenol (ISO) promotes rapid guanine nucleotide exchange on Gs reconstituted into CFP-β2AR·rHDL particles (50-fmol receptor; Fig. 5a). [35S]GTPγS binding appears biphasic with a Bmax of ≈60 ± 10 fmol. In contrast, [35S]GTPγS binding in the presence of timolol (an inverse agonist) occurred in a slow but saturable manner within 15 min and with an estimated Bmax of 33.5 fmol. The ISO effect can be reversed by coincubation with propranolol (an antagonist with weak inverse agonist activity; Fig. 5b). The difference between these binding conditions yields the ISO-β2AR-specific [35S]GTPγS-binding component of 26.5 fmol and represents ≈45 ± 7% of the total [35S]GTPγS binding. The ISO-specific stimulated [35S]GTPγS binding yields a final R:G ratio of 50 fmol:26.5 fmol or ≈1:0.53, suggesting that up to 53% of the β2AR·rHDL particles may contain a single G protein. Although it is possible that multiple G proteins may be reconstituted per receptor-containing particle, steric crowding due to the physical dimensions of the Gsαβγ heterotrimer (>80 Å from tip to tip) suggests this is unlikely to occur.

When fit to a two-phase exponential curve, [35S]GTPγS-binding data also reveal an initial burst of binding with a halftime of 3.9 ± 0.9 s, followed by a slower rate with a halftime of 4.3 ± 1.3 min. The initial burst, not observed in the presence of timolol, is consistent with [35S]GTPγS binding to empty Gs (devoid of GDP) and represents a precoupled G protein–receptor complex (2628). The slower phase is likely due to the inability of inverse agonists to completely inhibit basal receptor activation (29) and may slightly underestimate the ISO-stimulated component of [35S]GTPγS binding. It should also be noted that, although high amounts of G protein are required in the initial reconstitution (R:G ratio of 1:50), very low amounts of G protein actually incorporate into the receptor-containing particles. The vast majority of G proteins aggregate and precipitate because of the absence of detergent, a requirement to keep the acylated (Gsα) and prenylated (Gγ) subunits of the heterotrimeric G protein in solution. We have quantified the proportion of G protein that partitions into β2AR·rHDL particles using immunoaffinity column chromatography (anti-FLAG Sepharose) taking advantage of the N-terminal Flag epitope on β2AR. Only 5.8% of the total initial Gs added to the reconstitution copurified with β2AR·rHDL particles (see SI Text). The remaining GTPγS-binding activity was not detected in the FLAG column flowthrough or in any of the washes and was likely held up in the column. This is a typical characteristic of a large protein aggregate that is incapable of interacting with the β2AR·rHDL particles on the column and most likely incapable of interacting with β2AR·rHDL particles in the [35S]GTPγS- and [3H]DHAP-binding assays. Furthermore, failure to detect this aggregate in the ISO-stimulated [35S]GTPγS-binding assay is probably because the assay conditions are better-suited for receptor-catalyzed nucleotide binding (i.e., 100 nM [35S]GTPγS and 2 mM MgCl2) vs. those used to detect total [35S]GTPγS binding (10 μM [35S]GTPγS and 50 mM MgCl2) (see SI Text).

Monomeric β2AR also undergoes allosteric modulation of agonist binding by G proteins. Analysis of ISO inhibition of [3H]DHAP binding to these preparations reveals a classic biphasic competition curve with an observed Khigh of 8.0 ± 2.4 nM and a Klow of 0.36 ± 0.11 μM (n = 5) (Fig. 5c), in good agreement with reported values (30). The functional uncoupling of Gs by GTPγS yielded a monophasic inhibition curve with a Ki of 0.13 ± 0.09 μM (n = 2), very similar to the Klow value. The high-affinity ISO sites (Khigh) represented 57 ± 4% (n = 5) of the total [3H]DHAP binding, remarkably close to the proportion of receptors that contain G proteins (53% from [35S]GTPγS binding). The fraction of high-affinity ISO sites appears to depend on the presence of G proteins in the HDL particle (Fig. 5d). Simply stated, the more receptor particles occupied by G proteins, the larger fraction of high-affinity agonist sites are observed. In fact, increasing G protein concentrations with a fixed monomeric β2AR·rHDL concentration (i.e., increasing the initial R:G ratio from 1:50 to 1:200) increases the fractional high-affinity ISO states to levels approaching 90% (Fig. 5d). Accordingly, coincubation of a 1:200 R:G preparation with 10 μM GTPγS completely uncouples the G protein and yields a monophasic low-affinity ISO site indistinguishable from inhibition curve obtained in the absence of G proteins (Fig. 5d).


One of the earliest reports providing evidence that GPCRs may exist as dimers in the plasma membrane was published in 1996 (31) and was met with some criticism by investigators in the field. However, since that study, which used the β2AR as a model system, there have been a growing number of reports documenting both homo- and heterodimerization of GPCRs using a variety of techniques (1315, 3234). Consequently, the existence of GPCR dimers is now widely accepted. Studies documenting the requirement for heterodimerization in GABAB receptor signaling as well as studies providing evidence that dimerization is required for efficient export of GPCRs from the endoplasmic reticulum have led to speculation that a dimer is likely to be the functional unit for GPCR signal transduction (32). However, rigorous analysis of the role of dimers in G protein activation has previously not been possible because of the inability to physically isolate and characterize the function of a monomeric GPCR embedded in its membrane milieu.

In this study, we incorporate purified β2AR into rHDL particles along with purified Gs heterotrimer. We demonstrate that the detergent-solubilized and purified β2AR is monomeric before reconstitution, and that the rHDL particles contain only a single β2AR. Our study shows that the minimal functional unit is a monomer by virtue of the fact that agonist treatment induces rapid guanine-nucleotide exchange in G proteins. Moreover, G proteins directly induce the monomeric receptor to adopt a conformation that binds agonists with high affinity (nanomolar) compared with conditions where G proteins are absent or uncoupled (micromolar). By increasing the proportion of β2ARs occupied by G proteins, we observe an increase in the proportion of high-affinity agonist-binding sites. These data represent direct support of the original “ternary complex” coined by DeLean et al. ≈30 yr ago (35) and refutes the contributions of receptor oligomers toward high-affinity agonist binding.

The data presented here do not refute that receptors exist as oligomers in membranes, but rather the data suggest that oligomerization may play a minor role in G protein activation. In fact, the data remain consistent with a pentameric receptor–G protein complex model (R:R:Gα;Gβ:Gγ), as has been proposed for the GABABR, mGluR, and rhodopsin (6, 36, 37). In this model, only one of the two receptors within the dimer is capable of coupling to the G proteins.

Cumulatively, these data resolve a highly controversial question at the core of GPCR signal transduction. Although the β2AR is only one of several hundred GPCRs in the genome, it is likely that the observations reported here can be generalized to at least the class A or rhodopsin-like receptors. Members of this family encompass most of the hormone receptors that are targeted by clinically useful therapeutics. Thus, elucidating their minimal functional oligomeric state is critical for understanding their mechanism of action.

Materials and Methods

G protein viruses encoding Gαs, his6-Gβ1, and Gγ2 were generously provided by Alfred G. Gilman (University of Texas Southwestern, Dallas, TX). Expired human serum was generously donated by Bert La Du (University of Michigan, Ann Arbor). All lipids were purchased from Avanti Polar Lipids (Alabaster, AL). DDM was obtained from Dojindo Molecular Technologies (Gaithersburg, MD). Sodium cholate was purchased from Sigma (St. Louis, MO). (±)-Alprenolol, S(−)-timolol, (±)-propranolol, and (−)-ISO were obtained from Sigma. [3H]Dihydroalprenolol and [35S]GTPγS were obtained from Perkin-Elmer (Foster City, CA). Cy3 and Cy5 maleimide were purchased from GE Healthcare (Piscataway, NJ). All other reagents were of analytical grade.

Expression and Purification of β2AR.

Baculoviruses were generated by using transfer vectors (pFastBac; Invitrogen, Carlsbad, CA) that encoding an N-terminally 10xHis-tagged fusion protein between monomeric, enhanced CFP (ECFP) (Clontech) and the β2AR. 10xHis-mECFP-β2AR (mCBAR) was expressed in HighFive cells (Invitrogen), detergent extracted (1% DDM) and purified by immobilized metal affinity chromatography (IMAC) (Talon; Clontech) and anion-exchange and gel-filtration chromatography. Nonfluorescent encoded β2AR was purified as described (38). Additional details of the cloning method and purification may be obtained in SI Text.

In Vitro Reconstitution of rHDL.

WT human apoA-I was purified from human serum by a protocol adapted from Venter et al. (39) and described in detail in SI Text. HDL were reconstituted in vitro according to a protocol adapted from Jonas (20). Briefly, dimyristoyl phoshatidyl-choline and POPC were used alone or as a mixture of POPC and palmitoyl-oleoyl-phosphatidylglycerol in combination (3:2 molar ratio), to mimic the zwitterionic environment of a cell membrane (40). An rHDL reconstitution consisted of the following: 24 mM detergent (cholate or DDM), 8 mM lipid, and 100 μM apoA-I. Lipids were solubilized with 20 mM Hepes, pH 8/100 mM NaCl/1 mM EDTA plus 50 mM detergent. Purified apoA-I was added (at least 10-fold excess) to receptor preparations diluted in solubilized lipids. After an incubation of 1–2 h at the melting temperature of the lipid combination, samples were subjected to BioBeads (BioRad, Hercules, CA) to remove detergents, resulting in the formation of rHDL particles. Samples were stored on ice until used. If necessary, β2AR·rHDL particles from receptor-free rHDL were subsequently purified by M1-anti-FLAG immunoaffinity chromatography. Purified β2AR·rHDL particles were eluted with 1 mM EDTA plus 200 μg/ml FLAG peptide and stored on ice until further use.

Negative Staining and Transmission EM of rHDL Particles.

rHDL samples were placed on a carbon-coated copper grid and stained with 1% phosphotungstic acid, pH 6.5. Samples were imaged in a Philips (Eindhoven, The Netherlands) CM-100 transmission EM operating at 60 kV. See SI Text for additional details.

Analytical SEC.

Analytical SEC was performed on a HR10/30 column (GE Healthcare, Pittsburgh, PA) Superdex 200 (Amersham) in a buffer containing 20 mM Hepes, pH 8, 100 mM NaCl, and 1 mM EDTA using the BioLogic DuoFlow system (BioRad) at 4°C. Fractions from the column (200 μl) were collected in a 96-well plate for further analysis. Cy3 fluorescence (λex = 544 nm, λem = 595 nm) was analyzed in a Victor fluorescence plate reader (Perkin-Elmer, Waltham, MA).

Cy3 and Cy5 Labeling of β2AR.

Purified β2AR was incubated with 20-fold molar excess of maleimide-conjugated Cy3 and/or Cy5 (GE Healthcare) for 1 h on ice in detergent micelles. Receptors were subsequently treated with 2 mM iodoacetamide for 30 min to alkylate any reactive cysteines, followed by quenching with 2 mM cysteine. Free dye and unreacted iodoacetamide were removed by gel filtration (G-50). Labeled receptors were concentrated to 8 μM by using a Centricon 30 (Millipore, Billerica, MA) and reconstituted into rHDL as described above. Labeled receptors were stored on ice until further use.

Single-Molecule Imaging.

Single-molecule imaging was performed on an in-house custom-designed TIRF microscope based on a Nikon (Melville, NY) TE2000-U inverted microscope using a standard through-the-objective configuration (41). A 532-nm green diode-pumped frequency-doubled Nd:YAG laser (for Cy3 excitation, Compass 215M; Coherent, Santa Clara, CA) and a 638-nm red diode laser (for Cy5 excitation, RCL-638–025; Crystalaser, Reno, NV) are used as the excitation sources. To image a sample, 500 μl of ≈10 pM β2AR·rHDL was added to the chamber to allow nonspecific adsorption to the glass surface. After a 5-min incubation, the protein solution was pipetted off and immediately replaced by 500 μl of PBS (Invitrogen) to stop adsorption of additional molecules from solution. Images were acquired with WinView (Roper Scientific, Tucson, AZ) and analyzed with an in-house custom-designed program (see SI Text). The percent of colocalized molecules was calculated as: 2 × (no. of colocalized spots)/(no. of Cy3 spots + no. of Cy5 spots), assuming that a colocalized spot contains two receptors. The percent of colabeled molecules for the dual-labeled Cy3-Cy5-β2AR sample was calculated as: (no. of colocalized spots)/(no. of Cy3 spots + no. of Cy5 spots − no. of colocalized spots).

Ensemble FRET Spectra.

Steady-state FRET measurements were determined on a Spex FluoroMax-3 spectrofluorometer (Horiba Jobin Yvon, Edison, NJ) with photon-counting mode at 25°C. Spectra were collected as follows: Cy3 emission (λex = 525 nm, λem = 535–751 nm), Cy5 emission (λex = 625 nm, λem = 635–751 nm).

Saturation Radioligand-Binding Assays.

Binding reactions were prepared in 100-μl volumes in 96-well plates. Samples were incubated with various concentrations of [3H]DHAP (0.1–46 nM) in 50 mM Tris, pH 8/150 mM NaCl (TBS) (or TBS with 1% DDM or cholate, for detergent-solubilized binding). Nonspecific binding was determined in the presence of 20 μM propranolol. These data were fit to a one-site binding model by using Prism 4.0 (GraphPad, San Diego, CA) to determine Kd and Bmax.

G Protein Reconstitution and Agonist Competition Assays.

The Gs heterotrimer (Gαs, his6-β1, γ2) was expressed in Sf9 cells and purified as described (42). Purified Gs was reconstituted into preformed, pure β2AR·rHDL particles at varying R:G initial ratios (1:20–1:200). Concentrated Gs stocks were added such that the CHAPS was diluted at least 700-fold to reduce its concentration well below the critical micellar concentration. Agonist competition assays were performed under similar conditions as used in the saturation-binding assays with [3H]DHAP (2 nM), coincubated with various concentrations of ISO (ranging from 10−12 to 10−4 M) with or without 10 μM GTPγS. Samples were incubated for 30 min at 30°C and then filtered on glass-fiber plates as above. Normalized data were fit to a two-site competition binding model by using Prism (GraphPad). Further details are available in SI Text.

GTPγS-Binding Assay.

Purified Gs was added to preformed β2AR·rHDL complexes as above. Agonist-stimulated GTPγS-binding assays were performed as described by Asano et al. (43). Briefly, particles were preincubated with ligand (1 μM ISO, 10 μM inverse agonist timolol, or 1 μM ISO plus 3 mM propranolol) for 5 min at 30°C. The reactions were initiated by the addition of 100 nM [35S]GTPγS and terminated by addition of a 1,000-fold excess of ice-cold GTPγS and 10 mM MgCl2. Free [35S]GTPγS was removed by rapid filtration. See SI Text.

Supplementary Material

Supporting Information:


We thank Richard Neubig, John Tesmer, and Valerie Tesmer (University of Michigan) for critical discussion and Dorothy Sorenson and the University of Michigan Microscopy and Image Analysis Laboratory for assistance with EM. This work was supported by the Stanford Medical Scientist Training Program (M.P.B.), the Lundbeck Foundation (S.G.F.R.), National Science Foundation Grant BES-0508531 (to R.N.Z.), National Institute of Neurological Disorders and Stroke Grant NS28471 (to B.K.), the Mather Charitable Foundation (B.K.), National Institute of General Medical Sciences Grant GM068603 (to R.K.S.), Michigan Diabetes Research and Training Center Grant NIDDK P60DK-20572 (to R.K.S.), and the University of Michigan Biological Sciences Scholars Program (to R.K.S.).


G protein-coupled receptor
β2-adrenergic receptor
high-density lipoprotein
reconstituted HDL
apolipoprotein A-I
cyan fluorescent protein
total internal reflection fluorescence
size-exclusion chromatography


The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0611448104/DC1.


1. Catterall WA. Annu Rev Biochem. 1995;64:493–531. [PubMed]
2. Clapham DE, Runnels LW, Strubing C. Nat Rev Neurosci. 2001;2:387–396. [PubMed]
3. Dingledine R, Borges K, Bowie D, Traynelis SF. Pharmacol Rev. 1999;51:7–61. [PubMed]
4. Schlessinger J. Cell. 2000;103:211–225. [PubMed]
5. Chabre M, le Maire M. Biochemistry. 2005;44:9395–9403. [PubMed]
6. Fotiadis D, Jastrzebska B, Philippsen A, Muller DJ, Palczewski K, Engel A. Curr Opin Struct Biol. 2006;16:252–259. [PubMed]
7. James JR, Oliveira MI, Carmo AM, Iaboni A, Davis SJ. Nat Methods. 2006;3:1001–1006. [PubMed]
8. White JH, Wise A, Main MJ, Green A, Fraser NJ, Disney GH, Barnes AA, Emson P, Foord SM, Marshall FH. Nature. 1998;396:679–682. [PubMed]
9. Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W, Beck P, Mosbacher J, Bischoff S, Kulik A, Shigemoto R, et al. Nature. 1998;396:683–687. [PubMed]
10. Jones KA, Borowsky B, Tamm JA, Craig DA, Durkin MM, Dai M, Yao WJ, Johnson M, Gunwaldsen C, Huang LY, et al. Nature. 1998;396:674–679. [PubMed]
11. Hague C, Uberti MA, Chen Z, Hall RA, Minneman KP. J Biol Chem. 2004;279:15541–15549. [PubMed]
12. Uberti MA, Hague C, Oller H, Minneman KP, Hall RA. J Pharmacol Exp Ther. 2005;313:16–23. [PubMed]
13. Angers S, Salahpour A, Joly E, Hilairet S, Chelsky D, Dennis M, Bouvier M. Proc Natl Acad Sci USA. 2000;97:3684–3689. [PMC free article] [PubMed]
14. Overton MC, Blumer KJ. Curr Biol. 2000;10:341–344. [PubMed]
15. Guo W, Shi L, Javitch JA. J Biol Chem. 2003;278:4385–4388. [PubMed]
16. Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A, Palczewski K. Nature. 2003;421:127–128. [PubMed]
17. El-Asmar L, Springael JY, Ballet S, Andrieu EU, Vassart G, Parmentier M. Mol Pharmacol. 2005;67:460–469. [PubMed]
18. Baneres JL, Martin A, Hullot P, Girard JP, Rossi JC, Parello J. J Mol Biol. 2003;329:801–814. [PubMed]
19. Matz CE, Jonas A. J Biol Chem. 1982;257:4535–4540. [PubMed]
20. Jonas A. Methods Enzymol. 1986;128:553–582. [PubMed]
21. Bayburt TH, Sligar SG. Protein Sci. 2003;12:2476–2481. [PMC free article] [PubMed]
22. Bayburt TH, Sligar SG. Proc Natl Acad Sci USA. 2002;99:6725–6730. [PMC free article] [PubMed]
23. Leitz AJ, Bayburt TH, Barnakov AN, Springer BA, Sligar SG. BioTechniques. 2006;40:601–602. 604, 606, passim. [PubMed]
24. Segrest JP, Jones MK, Klon AE, Sheldahl CJ, Hellinger M, De Loof H, Harvey SC. J Biol Chem. 1999;274:31755–31758. [PubMed]
25. Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, Stenkamp RE, et al. Science. 2000;289:739–745. [PubMed]
26. Neubig RR, Gantzos RD, Thomsen WJ. Biochemistry. 1988;27:2374–2384. [PubMed]
27. Mukhopadhyay S, Ross EM. Proc Natl Acad Sci USA. 1999;96:9539–9544. [PMC free article] [PubMed]
28. Biddlecome GH, Berstein G, Ross EM. J Biol Chem. 1996;271:7999–8007. [PubMed]
29. Seifert R, Wenzel-Seifert K, Lee TW, Gether U, Sanders-Bush E, Kobilka BK. J Biol Chem. 1998;273:5109–5116. [PubMed]
30. Kent RS, De Lean A, Lefkowitz RJ. Mol Pharmacol. 1980;17:14–23. [PubMed]
31. Hebert TE, Moffett S, Morello JP, Loisel TP, Bichet DG, Barret C, Bouvier M. J Biol Chem. 1996;271:16384–16392. [PubMed]
32. Galvez T, Duthey B, Kniazeff J, Blahos J, Rovelli G, Bettler B, Prezeau L, Pin JP. EMBO J. 2001;20:2152–2159. [PMC free article] [PubMed]
33. Lavoie C, Mercier JF, Salahpour A, Umapathy D, Breit A, Villeneuve LR, Zhu WZ, Xiao RP, Lakatta EG, Bouvier M, et al. J Biol Chem. 2002;277:35402–35410. [PubMed]
34. Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A, Palczewski K. Nature. 2003;421:127–128. [PubMed]
35. De Lean A, Stadel JM, Lefkowitz RJ. J Biol Chem. 1980;255:7108–7117. [PubMed]
36. Goudet C, Kniazeff J, Hlavackova V, Malhaire F, Maurel D, Acher F, Blahos J, Prezeau L, Pin JP. J Biol Chem. 2005;280:24380–24385. [PMC free article] [PubMed]
37. Baneres JL, Parello J. J Mol Biol. 2003;329:815–829. [PubMed]
38. Devanathan S, Yao Z, Salamon Z, Kobilka B, Tollin G. Biochemistry. 2004;43:3280–3288. [PubMed]
39. Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA, et al. Science. 2001;291:1304–1351. [PubMed]
40. Cerione RA, Ross EM. Methods Enzymol. 1991;195:329–342. [PubMed]
41. Huang B, Perroud TD, Zare RN. ChemPhysChem. 2004;5:1523–1531. [PubMed]
42. Kozasa T, Gilman AG. J Biol Chem. 1995;270:1734–1741. [PubMed]
43. Asano T, Pedersen SE, Scott CW, Ross EM. Biochemistry. 1984;23:5460–5467. [PubMed]
44. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, Ferrin TE. J Comput Chem. 2004;25:1605–1612. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...