• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Curr Opin Drug Discov Devel. Author manuscript; available in PMC Feb 28, 2006.
Published in final edited form as:
Curr Opin Drug Discov Devel. Sep 2001; 4(5): 561–574.
PMCID: PMC1383658
NIHMSID: NIHMS7994

G protein-coupled receptor drug discovery: Implications from the crystal structure of rhodopsin

Abstract

G protein-coupled receptors (GPCRs) are a functionally diverse group of membrane proteins that play a critical role in signal transduction. Because of the lack of a high-resolution structure, the heptahelical transmembrane bundle within the N-terminal extracellular and C-terminal intracellular region of these receptors has initially been modeled based on the high-resolution structure of bacterial retinal-binding protein, bacteriorhodopsin. However, the low-resolution structure of rhodopsin, a prototypical GPCR, revealed that there is a minor relationship between GPCRs and bacteriorhodopsins. The high-resolution crystal structure of the rhodopsin ground state and further refinements of the model provide the first structural information about the entire organization of the polypeptide chain and post-translational moieties. These studies provide a structural template for Family 1 GPCRs that has the potential to significantly improve structure-based approaches to GPCR drug discovery.

Keywords: Agonist, antagonist, drug discovery and design, G protein, G protein-coupled receptors (GPCRs), rhodopsin, signal transduction
Abbreviations: β2R β2-adrenergic receptor, EPI epinephrine, FVS fuzzy virtual screening, GPCR G protein-coupled receptors SAR structure-activity relationship, SCAM substituted Cys accessibility method, TM transmembrane

Introduction

The plasma membrane of mammalian cells separates the extracellular and intracellular environments; however, the vast array of information between the cells and their environment must be transmitted across the cell surface. Proteins embedded in the membrane bilayer mostly accomplish this function. Membrane proteins represent a large and versatile group of protein sensors that are involved in nearly all physiological processes in vertebrates, and are thus primary targets for drug discovery. More than 40% of the total sales of available drugs are aimed at a single class of membrane proteins termed G protein-coupled receptors (GPCRs) [1,2●]. In all organisms, from the most primitive to mammals, GPCRs are present in multiple molecular forms; for example, 5% of the Caenorhabditus elegans genome encodes GPCRs [3]. Approximately 400 non-sensory and thousands of sensory GPCRs have been identified in humans, comprising one of the largest families of proteins encoded by our genome [4]. The ligands are known for many of the GPCRs, but there is still a long list of 'orphan receptors' that need to be paired with physiologically relevant substances [5]. The classification of novel GPCR subtypes could be aided by diagnostic 'fingerprinting' [6].

The advent of powerful structural techniques to elucidate the molecular mechanisms of drug action is driving discovery of novel lead compounds towards structural-based approaches. The structure of a therapeutic target of interest is quickly becoming the centerpiece unto which molecular information is integrated and analyzed in order to further guide the drug discovery process [2●].

The difficulty in elucidating the structure of membrane proteins, including GPCRs, at atomic resolution continues to be a daunting task, and implies that structure-based approaches to GPCR drug discovery will rely on a limited number of available structures. In the absence of a structure for the GPCR of interest, structure-based drug discovery relies on molecular models generated computationally. The favored approach to generate these models is homology modeling, where the structure of a target protein is modeled using the structure of a related protein as a template [710]. The recent elucidation of rhodopsin [11,12●●,13●,14●●], thus enables the application of homology modeling techniques to generate reasonable models for homologous GPCRs. In addition, the pioneering study solving the first crystal structure of a GPCR [14●●] has generated a reasonable expectation that other structures may soon become available.

The aim of this review is to evaluate the extent to which the structure of rhodopsin enables a more powerful approach for modeling homologous GPCRs suitable for drug discovery applications. We will briefly describe relationships between three families of GPCRs, the potential benefits and problems of molecular modeling based on the high-resolution structure of rhodopsin, changes in our thinking about identification of lead compounds, and further work in the identification of conformational changes that these receptors undergo upon binding of agonists and antagonists. We conclude this review with a section dedicated to ascertaining which additional structures and pharmacological data appear to be necessary to support further progress structure-based approaches to GPCR drug discovery.

Rhodopsin and the GPCR family

GPCRs are integral membrane proteins composed of seven membrane-spanning segments that transduce an extracellular signal by coupling to G proteins on the cytoplasmic side of the cell [1518]. These receptors may also modulate the intracellular processes independent of G proteins [1921]. The membrane-spanning segments have been widely predicted to fold as a seven α-helical bundle, as observed for rhodopsin, and this topology of GPCRs is commonly represented in two-dimensional (2D) diagrams (Figure 1). On the basis of the conservation pattern of the primary amino acid sequence, three families of GPCRs have been described (for review see [22●]). Family 1, the most numerous of the three, is defined by receptors whose sequences are homologous to rhodopsin. Family 2 encompasses GPCRs homologous in sequence to the secreting receptor, and Family 3 includes GPCRs homologous in sequence to the GABA receptor. Beyond the seven transmembrane (TM) topology and in the absence of significant sequence homology, a similar three-dimensional (3D) structure cannot be automatically assumed. There are no identifiable sequence homologies between these three families. The structure of rhodopsin may serve as a template for Family 1, although its use as a template for Family 2 and Family 3 requires additional confirmations. For example, the analysis of the physicochemical properties for secretin-like GPCRs [23] identified similar patterns as previously found for rhodopsin-like GPCRs [24]. Furthermore, crosslinking experiments between TM3 and TM6 of the parathyroid hormone receptor that mimic similar studies on rhodopsin and the β2-adrenergic receptor (β2R) gave similar phenotypes inactivating the receptors [24]. These findings suggest that the tertiary fold of rhodopsin might be conserved through evolution and also extends to the secretin family.

Figure 1
Helical net representation of the transmembrane domain of rhodopsin, showing the most conserved residues across Family 1 GPCRs.

The overall homology within Family 1 is relatively low (in many cases < 35% compared to rhodopsin), although the presence of seven TM domains and highly conserved motifs within these domains clearly identifies the evolutionary relationship among these proteins since related domains are conserved more strongly than the amino acid sequence [25]. These characteristic motifs, highlighted in Figure 1, are used to define a general numbering scheme that applies to all rhodopsin-like GPCRs: each residue is identified by the TM number (1 to 7), the most conserved residue within each TM is assigned the number '50', and the positions of all other residues are numbered relative to the most conserved amino acid in each TM segment [26]. For example, a conserved Asn, N1.50, often preceded by G1.49 and with Val at position 1.53, characterizes TM1. A conserved Asp, D2.50, preceded by L2.56 and often A2.57, are present in TM2. The most conserved motif in TM3 is an Arg residue, R3.50, at the cytoplasmic boundary preceded by an acidic residue E/D3.49 and often followed by Y3.51 and in the next turn by a branched amino acid I/V3.54. This (D/E)RYxx(V/I) motif has been referred to as the 'Arg-Cage' (containing 'DRY' region) for its role in maintaining the receptor in its inactive conformation [27,Ballesteros et al, personal communication]. TM3 is further characterized by a conserved L3.43 and a Cys at position 3.25 that are crosslinked to a Cys within the TM4-TM5 extracellular loop. The conserved Trp, W4.50 in the middle of the helical TM segment characterizes TM4. The most conserved residue in the TM5 is Y5.58, although the highly conserved P5.50, often associated with F5.47, is better suited as an easily identifiable reference amino acid. TM6 is one of the most conserved TM segments, whose signature is the set of residues F6.44xxC6.47W6.48xP6.50, all except P6.50, which defines a continuous area on the helical surface of TM6 (Figure 1). The cytoplasmic boundary of TM6 contains highly conserved basic residues (R/K) at positions 6.32 and 6.35, as well as an often-conserved acid residue at 6.30. The seventh TM is characterized by the NPxxY motif where the Pro residue is easily recognized and thus becomes P7.50. There is a short helical segment following TM7 that resides at the boundary between the TM domain and the cytoplasm, referred to as the H8 helix. A conserved F8.50, a palmitoylated Cys at its C-terminus, and a solvent exposed cluster of non-conserved Arg and Lys characterize this TM8 α-helix.

The 3D position of these conserved residues in the structure of rhodopsin is shown in Figure 2. Because these diverse GPCRs recognize an extraordinary number of agonists, and couple to different G-proteins, the role of conserved residues is either structural, or related to the molecular mechanism of receptor activation. Some of these conserved residues cluster in conserved 3D motifs are referred to as 'functional microdomains' (Figure 2) [27], because they define microdomains within the receptor structure responsible for the functional integrity of the receptor. The presence of these functional microdomains is further supported by the consistency in the functional phenotype of equivalent mutations of conserved residues in different receptors. The assumption of a common structural fold among Family 1 of the GPCRs thus enables homology-modeling approaches based on the crystal structure of rhodopsin.

Figure 2
Conserved residues in rhodopsin 3D structure: Functional microdomains.

Rhodopsin structure as a template for Family 1 GPCRs

Several recent reviews critically address the use of the rhodopsin structure as a template to model related Family 1 GPCRs [2●,28●●,29,30]. These reviews highlight the consistency of the rhodopsin structure with prior structural inferences derived from biochemical data and structure-function studies in related GPCRs, validating the experimental techniques and their respective inferences in the absence of a high-resolution structure. The rhodopsin structure is consistent with the tertiary interactions proposed for different GPCRs based on Cys crosslinking, double revertant mutants, engineered His2-Zn2+-binding sites, spin labeling and ligand-receptor interactions [28●●,29]. This analysis is restricted to interactions pertaining to the inactive state of rhodopsin, as the structure of the active state is yet unknown (see below).

Beyond the structural studies, the most comprehensive methods used to characterize the structure of GPCRs focus on the accessibility and orientation of each residue within the TM domain. These methods rely on the substitution of continuous, one at a time stretches of residues by Cys whose accessibility to several sulfhydryl-specific chemical probes can be characterized experimentally. The laboratories of Hubbell and Khorana have utilized this method extensively to study the accessibility and orientation of spin-labeled probes attached to these Cys residues in rhodopsin and bacteriorhodopsin [3147,48●●]. The results of these studies pertain mostly to the orientations and conformational rearrangements occurring at the cytoplasmic side of rhodopsin upon activation. The laboratory of Javitch has also utilized this method, termed substituted Cys accessibility method (SCAM), to study the accessibility to hydrophilic, ligand-mimetic probes on the human dopamine D2 receptor (for a review see [28●●]). The goal of these studies was to identify the residues accessible within the receptor ligand-binding site crevice, and thus lead to more pertinent drug discovery applications. The consistency of the accessibility patterns derived from both sets of data, based on rhodopsin studies and the D2 receptor, further support the assumption of a common fold for these receptors. However, an important exception is the inconsistency found for the TM4 helix, where a significant portion of the patch of accessible residues identified by SCAM on the D2 receptor is facing the lipid phase in the rhodopsin structure. The patch of residues accessible at the binding site of the D2 receptor includes the highly conserved Trp in TM4, W4.50, which was expected to be facing the protein interior based on its high degree of sequence conservation. Surprisingly, the conserved W4.50 is exposed to the lipid phase in the rhodopsin structure, further confounding the rationalization of the SCAM data and the elucidation of its functional and/or structural role [49]. Although the highly conserved Tyr in TM5, Y5.58, is also exposed to the lipid milieu in the rhodopsin, it is still accessible by SCAM techniques. It might be that these residues become reoriented towards the protein interior upon activation, or that they form part of the dimerization interface of these receptors.

The characterization of the binding site crevice of the D2 receptor based on SCAM studies on the seven TM segments of the D2 receptor has been recently reviewed in the context of the structure of rhodopsin [28●●]. The residues identified as accessible to the binding site crevice by a chemical probe with the physicochemical and volume properties of a small molecule ligand extend deep into the TM domain reaching the cytoplasmic boundaries. This finding is inconsistent with identified ligand-receptor interaction sites as well as the preserved binding pocket between rhodopsin and other GPCRs, and thus suggests the dynamic nature of the protein, rather than the availability of these residues for specific drug-receptor interactions. The analysis presented for the relative orientation of amino acids residues within the accessible area for each TM segment allows the identification of divergent motifs that may be responsible for local structural deviations from the rhodopsin structure. The role of Pro-kink motifs and Ser/Thr/Cys residues in modulating the conformation of TM helices suggests that these motifs may function as either flexible hinges, or supporting alternative conformations of these TM helices. The implications for localized structural divergence among GPCRs is addressed in this review, with specific illustrations of the degree of conformational heterogeneity to be expected among different GPCRs. Surprisingly, the high potential for conformational variability among GPCRs is coupled with the realization that GPCR structures may have evolved to mimic a common fold through convergent 'structural mimicry' molecular mechanisms [28●●]. This hypothesis is based on the observation that among the alternative helical conformations supported by divergent motifs between rhodopsin and the D2 receptor; the conformations that satisfy the accessibility data derived from the D2 receptor are those that mimic the conformation found in the rhodopsin structure. A good example of structural mimicry is the bend observed in TM2 of rhodopsin at the level of the GGxTT; the helical turn is mimicked in the D2 receptor by the kink induced by Pro2.59. This Pro residue, conserved among neurotransmitter GPCRs, yet absent in opsins, represents an alternative molecular mechanism to achieve an otherwise convergent bent.

Taking into account these considerations, Visiers and colleagues have presented a comprehensive review on the methodologies available to computationally model the structure of GPCRs based on the rhodopsin structure [29]. This review also addresses the methodologies available for computational probing of these molecular models to derive functional inferences, such as ligand binding and the ensuing conformational changes. The necessary interplay between molecular models and experimental probing to synergistically derive functional conclusions is also addressed in some recent reviews [2●,30].

Cytoplasmic and extracellular loops of rhodopsin could not be used to infer the structure of similar domains in other GPCRs because they are highly divergent among GPCRs in terms of their sequence, length and functional roles (Figure 3). For several GPCRs, the extracellular loops and N-terminus are probably involved in the recognition of peptide ligands such as chemokines, and therefore de novo modeling of these loops may be necessary [2●,29]. Although most of the free energy of binding of these peptides comes from interactions within the extracellular loops, activation of the receptors probably involves direct interactions between a fragment of the peptide ligand and the TM domain. In addition, small molecule drugs are still likely to bind within the TM domain, and thus rhodopsin offers an appropriate structural template to explore the potential drug-receptor interactions. Two structural motifs outside the TM domain may be preserved in some GPCRs. The short helical segment following TM7, termed helix 8 (H8), is predicted for most GPCRs [29,50], and identified by NMR on non-rhodopsin GPCRs [51]. However, the absence of this segment in the short C-terminus of the human GnRH receptor [52] questions the requirement of the H8 for the structural and functional integrity of GPCRs. Another common motif might be the short β-strain adjacent to the conserved disulfide bridge between C3.25 and a Cys in the loop between TM4 and TM5. The disulfide link is known to be conserved and present in most GPCRs, but the TM4-TM5 loop is different in length and can form several structural elements different from the β-strain. The realization from the rhodopsin structure that a short β-sheet surrounding the Cys in the TM4-TM5 loop forms the top of the binding site has, however, prompted a re-evaluation of this issue for adrenergic receptors [53].

Figure 3
Divergent structures for the extracellular segment of GPCRs: Crystal structures from different GPCRs.

Modeling ligand-receptor complexes based on the rhodopsin template

The availability of reasonable structural models for the binding site of GPCRs based on the rhodopsin template enables the computational exploration of potential ligand-receptor interactions at the atomic level. There are two main applications of this structural approach to drug discovery: lead identification and lead optimization. For lead identification, the generation of 3D receptor pharmacophores can guide the selection of compound libraries for screening purposes. Docking the structure of a lead compound onto a receptor model enables the exploration of the potential effects of chemical modifications of the compound, and thus facilitates the lead optimization process. In this approach, the TM domain of the receptor model and its binding pocket is explored for the most salient physicochemical properties (see [2] for a review on available methodologies). For example, an aromatic microdomain between TM2, TM3 and TM7 of the dopamine D2 receptor identified by molecular modeling and confirmed experimentally [54], was later found to be responsible for D2/D4 receptor selectivity among several D4-selective ligands [55]. Generating a 3D pharmacophore for the receptor of interest enables the identification of key residues and physicochemical properties that are likely the molecular determinants for ligand recognition. Structure-activity relationships (SAR) among available ligands for a given receptor, if available, provide equivalent pharmacophores at the ligand level. The parallel development of ligand-based and receptor-based 3D pharmacophores, developed independently from each other, enables a self-consistency check search for spatial complementarities between them based on favorable physicochemical interactions. Development of a 3D pharmacophore, based on a computational model of a GPCR derived from the rhodopsin structure, should take into consideration the intrinsic inaccuracy embedded in these models. This inaccuracy results from the intrinsic flexibility of proteins, the potential for significant deviations in the helical conformation among different GPCRs as compared to rhodopsin (see [28●●] for a review), and the different rotamer conformations that amino acid side chains may adopt within the binding site of a modeled GPCR. Given this range of uncertainty in the resulting model, the goal at this stage is to develop a fuzzy pharmacophore that highlights the key molecular determinants for ligand recognition, to be revised and optimized based on ongoing experimental data. Modeling and energetic probing of ligand-receptor complexes at the atomic level at this stage are quite speculative in nature, and are utilized only to generate novel and experimentally testable hypotheses to guide the drug discovery process. The study of ligand-receptor interactions is an iterative process where, at every stage of a given drug discovery project, models are refined based on available data and used to continuously derive new inferences.

Lead identification: Compound libraries biased by a GPCR 3D pharmacophore

Development of a 3D receptor pharmacophore can be used to guide the selection and/or design of compound libraries with physicochemical properties that match the pharmacophore of the target receptor. Some companies are able to significantly enhance their hit rate in screening compound libraries by first performing virtual screens. In virtual screening approaches, potential compounds are docked into the binding site and a docking score is determined computationally. Those compounds with favorable docking scores are then selected for screening the target of interest. Although these approaches are most successful when the structure of a protein target is known at atomic resolution, the availability of the structure of rhodopsin will enable better resolution models of GPCRs to support virtual screening approaches. However, given the uncertainties in these models described above and addressed in detail in a recent review [28●●], the structure of rhodopsin would only support low stringent virtual screening approaches ('fuzzy virtual screening' (FVS)) that take into consideration these inaccuracies which might be more appropriate at this stage.

There is significant interest in biasing compound libraries for a given class of therapeutic targets such as GPCRs. The rationale of these approaches is the commonalities in drug delivery (extracellular), overall size of the binding site and potentially shared physicochemical determinants within receptor subclasses. The structure of rhodopsin offers the possibility of guiding the chemical design by providing a structural template for the binding site of nearly all members of Family 1. Notably, a recent comparison between the retinal-binding site in the rhodopsin structure and other GPCRs uncovered a surprising similarity in the key contact residues defining the binding pocket [28●●]. The residues in the TM domain of rhodopsin in direct contact with 11-cis-retinal comprise the most important ligand-binding residues identified for other GPCRs (Figure 4).

Figure 4
The residues in the TM domain of rhodopsin in direct contact with retinal comprise the most important ligand binding residues identified for other GPCRs.

Lead optimization: Guidelines from 3D ligand-receptor interactions

There are two recent reviews that address the methodologies available to model the interactions between GPCRs and their ligands [2●,29]. The availability of an appropriate high-resolution structural template to model the binding site for homologous Family 1 GPCRs represents a significant advance for GPCR drug discovery. To the extent that the 3D fold is significantly maintained in other receptors, these molecular models might be reasonable structural templates at 3 to 5 Å resolution for the GPCR target of interest. To decide which approaches at this level of resolution are most suitable to model ligand-receptor interactions on a structural template, two-steps are proposed. Firstly, the ligand-receptor interactions need to be identified so as to guide docking of the ligand into the receptor pocket. Secondly, the model can be used to probe specific interactions at the atomic level starting at a 3 to 5 Å level of resolution.

The best approach to identify ligand-receptor interactions relies on SAR, supported by experimental data. In most cases, experimental data is restricted to the compound chemical structures and its pharmacological characterization of the wild-type receptor. Unfortunately, this type of experimental data does not provide any specific ligand-receptor interactions, and molecular models of ligand-receptor complexes are based on computational docking approaches and theoretical SAR. The reliance on computational calculations to drive our understanding of the ligand-receptor interactions is subjected to the intrinsic uncertainty of these models. The binding site and ligand-receptor interactions derived from models based on rhodopsin might be significantly different as a consequence of localized structural divergence [28●●]. Furthermore, energetic calculations should be sensitive to the complex and multiphasic environment of a lipid bilayer embedded receptor [56]. Given the variability in ligand-receptor models the structure-based approaches to GPCR drug discovery would require experimental validation, and the goal would be to use computational calculations to probe what is structurally feasible, while experimental data guides the actual decision-making process. Thus, the focus shifts towards defining the experimental data necessary to support structure-based drug discovery applications based on the rhodopsin structure.

Current approaches for the identification of specific ligand-receptor interaction sites for a ligand and receptor pair of interest are based on correlating the effects of minor modifications of the ligand with mutations of the receptor. Strader pioneered this approach in a landmark study investigating the interaction of epinephrine derivatives with the β2R [57]. Rationalizing the correlations in the affinity constants observed when the 3- or 4-hydroxyls of epinephrine were removed, compared with removing S5.43 or S5.46 on the receptor, the authors have proposed hydrogen bonding interactions of the 3-OH with S5.43, and the 4-OH with S5.46. These interactions are shown in Figure 5A together with other identified interaction sites for epinephrine on the β2R. Though this 2D exchange method to identify ligand-receptor interactions was pioneered at a pharmaceutical company, the time and effort necessary to obtain this kind of data has restricted its application. Although in the past this method has been prohibitory slow for drug discovery applications, novel technologies are quickly enabling its utilization in a high-throughput mode, making it suitable for industrial applications. As these technologies become available, they should provide reliable biochemically derived molecular models of ligand-receptor complexes that enable a powerful structure-based approach to drug discovery.

Figure 5
Epinephrine and β2 -adrenergic receptor.

As noted by Muller [2●], a structure-based approach to GPCR drug discovery in the absence of the real structures requires a multi-disciplinary approach, where molecular models represent a structural context to efficiently integrate experimental data and inferences derived from molecular biology, biophysics, bioinformatics, pharmacology and organic chemistry methods. Although not always achievable, the success of a synergistic effect among these disciplines is highly dependent on the experimental design. Synergy is best achieved when mutations are structurally interpretable, structural hypotheses are experimentally testable, ligands are well characterized pharmacologically, and the necessary chemical modifications of the ligands are available. The extent to which these conditions are met define the quality of the information derived to guide the lead compound optimization process.

Modeling ligand-receptor complexes, as provided by the high resolution structure of rhodopsin, enables the study of specific interactions at higher resolution. An improvement in the structural template from the 7 to 15 Å resolution structure of rhodopsin derived by the laboratory of Schertler [58], to the recent 2.8 Å resolution structure of rhodopsin [12,14] implies approximately a 5 Å resolution in the model of the binding site of a GPCR target of interest. This leads us to the question: what is the next step in understanding the molecular basis of drug-receptor interactions at 5 Å resolution? The improvement in the accuracy shifts the focus from the identification of specific ligand-receptor contacts to the definition of the specific conformations and spatial orientation in which these contacts take place. The insightful studies of epinephrine (EPI) binding to the β2R can be used as an illustrative case.

Based on the seven TM helical arrangement derived from rhodopsin and extensive mutagenesis data, a set of interactions have been proposed between EPI and the hamster β2R (Figure 5A). An example is the described interactions between the catecholamine hydroxyls of the ligand with Ser residues in TM5 [57]. A recent re-evaluation of these interactions on the human β2R also identified S5.42 as participating in these interactions [59]. The resulting proposal is that the 3-OH of catecholamines interacts with both S5.42 and S5.43 of the receptor, and that the 4-OH interacts with S5.46 of the receptor through hydrogen bonding. In order to translate these proposed interactions into 3D structural models of EPI docked into the receptor's binding site, there are several possibilities available. Each OH moiety, in either the ligand or the receptor, can act as a single hydrogen bond donor and/or as a double hydrogen bond acceptor, and each hydrogen bonding interaction between them is subject to certain geometric requirements. Furthermore, modeling these hydrogen bonding interactions is also subject to the conformational degrees of freedom in each moiety. Serine residues in an α-helix conformation can exist in either of three different rotamer conformations, defined by the first dihedral angle of the residue centered around the Cα-Cβ bond termed chi1 (Figure 5B). The different rotamer conformations are centered on 180° (trans), -60° (gauge+, g+) and +60° (gauge−, g−), and each span a 120° arc for a total of 360°. The most stable side chain conformation for Ser residues in α-helices is the g+ rotamer, representing 51% of all high resolution PDB structures [60], followed by trans at 30% and g- at 17%. The actual ratio between Ser in trans versus g- conformations in α-helices of known protein structures is reversed in membrane proteins (15 versus 38 %) relative to soluble proteins (30 versus 17%), which means that all three rotamer configurations should be explored. We should, therefore, explore how, at the atomic level, two OH moieties from EPI hydrogen bond to either 2 (2-OH) or 1 (4-OH) hydroxy moieties from Ser residues in the same helical turn of the receptor, which can be either hydrogen bonding donor (1) or acceptor (2), where each of the three Ser can adopt either of the trans, g+ or g-conformations of rotamers. Figure 5B illustrates the significant reorientation of the phenyl-(OH)2 moiety of EPI depending on whether we model EPI hydrogen bonding to S5.43 and S5.46 based on the original published data [57], or if we include the interaction with S5.42 identified recently [59]. Figure 5C illustrates the significant reorientation of the phenyl- moiety of EPI, depending on the particular modeling (OH)2 choice, when we only change the rotamer of S5.46 from g+ to g-. Different spatial reorientations of EPI will result in distinct positioning of chemical substituents of the phenyl ring within the binding site of the receptor. The identification of the set of key ligand-receptor interactions involved (Figure 5), and the choice of a particular rotamer conformation for these Ser, as well as the other parameters involved, would result in different interactions of the potential substituents, and thus different guidelines for lead optimization. Furthermore, another important degree of conformational heterogeneity results from potential interactions between the side chain of a Ser residue in a α-helix and the helix backbone. This side chain-backbone interaction of Ser residues has the potential to induce a significant bend in the α-helical structure, especially in the g-rotamer conformation [60]. The actual modeling process is even more complicated because the spatial positioning of these three Ser residues is also determined by the conserved Pro-kink in TM5 in the turn S5.46-P5.50. Pro-kinks represent flexible hinges within α-helices, inducing a significant bend in the helix from 0 to 40°, with an average of 26° [61]. The degree of potential conformational heterogeneity for the conserved Pro-kink in TM5 of GPCRs is addressed in a recent review [28●●].

The analysis above illustrates the level of complexity that needs to be addressed to efficiently guide the lead optimization process based on ligand-receptor models, and highlights the next level of questions we need to address given the resolution that the rhodopsin structure enables for these models. Clearly, several important issues need to be resolved to translate identified ligand-receptor interactions from mutagenesis experiments or other inferences into reasonable 3D models suitable to guide the lead optimization process. While computational methods are becoming increasingly powerful to explore this level of detail (revised in [29]), these issues cannot be addressed until experimental validation at this level of detail is obtained. This requires the availability of 3D structures of both wild-type and mutant receptor with a variety of ligands.

The active state of GPCRs: Inferences from the rhodopsin structure

Although GPCR function involves a number of functional states [22●,62,63], and several proteins beyond G proteins are involved such as arrestins and G protein-receptor kinases [21,6466], we will focus on the minimal set of functional states necessary to support a structure-based approach to GPCR drug discovery. Structural templates are required for both antagonist and agonist binding, which mean the inactive and active states of a GPCR. There are a number of discrete functional states for rhodopsin, including some intermediates of the activated state. The correspondence of functional states between rhodopsin and other GPCRs has been reviewed recently in terms of energy [13●]. The ternary complex model of receptor activation is commonly used to describe the GPCR activation cycle, shown for ligand-activated GPCRs (Scheme 1, left) and rhodopsin (Scheme 1, right).

Scheme 1
Ligand-activated and rhodopsin GPCRs.

In the absence of ligand (L), ligand-activated receptors exist in equilibrium between inactive (R) and active (R*) states, and the active state (R*) has high affinity for the G protein (R*G). The basal level of G-protein activation is low because GPCRs are predominantly in an inactive conformation, the state which does not couple and activate G proteins. Ligand binding to a receptor (Scheme 1, lower left) establishes a new equilibrium among inactive and active states resulting in enhanced (agonist) or decreased (inverse agonist) levels of activation. The conformational change of the receptor is driven by the ligand binding, providing as much as ~ 46 (for KD = 10 nM) to ~ 63 kJ/mol (for KD = 10 pM). Agonists have a higher affinity for the activated coupled receptor (LR*G), while inverse agonists have a higher affinity for the inactive receptor (LR). Ligands that minimally affect the ligand-free equilibrium are called neutral antagonists. Traditional antagonists can thus be either inverse agonists or neutral antagonists. Under most conditions studied, including agonist binding, the active receptor (R* or LR*) without G protein is a transient, unstable state that quickly decays back to the inactive state or is stabilized by G protein-binding [67]. This interconversion between these conformational states of GPCRs will hinder attempts to crystallize the active state.

Rhodopsin is a special GPCR in that the ligand is covalently linked to apoprotein holding the receptor in the inactive conformation [13●]. According to the GPCR activation cycle (Scheme 1, lower right), the structure of rhodopsin corresponds to the ligand-bound inactive state of a GPCR (LR), where the ligand behaves as a strong inverse agonist. Light catalyzes the conversion of 11-cis-retinal (LcR) into all-trans-retinal (LtR), resulting in a number of short-lived intermediate states that quickly decay (ms) into the Meta I (MI) state. The spontaneous MI to Meta II (MII) conversion, in face of the positive enthalpy, indicates that entropy must drive this conversion. This observation suggests increased overall disorder of the MII state, an observation such as this is consistent with the idea that formation of the active state is merely a release of constraints of the 'floppy' cytoplasmic surface including helix 6 (Figure 6). MII has high affinity for the G protein transducin, which leads to the G protein-bound MII state responsible for signaling [13].

Figure 6
Similar conformational changes upon receptor activation in rhodopsin and other GPCRs: TM3-TM6.

In terms of the activation cycle of GPCRs, the MII state corresponds to LR*G, and the pre-MII states correspond to the intermediate LR* (Scheme 1, lower right). Because of the energy delivered by light, the transformation from rhodopsin to MII is an irreversible process far away from equilibrium. Another difference in the rhodopsin activation cycle relative to GPCRs activated by diffusible ligands is that the covalent bond is irreversible, and consequently there is no equilibrium between bound and unbound states. However, the active MII state of rhodopsin progresses into MIII (not shown), where the covalent bond joining retinal to the receptor is hydrolyzed and the ligand is released resulting in the unbound receptor called opsin (Scheme 1, arrow). The resulting opsin (Scheme 1, top right) is like a standard GPCR in the absence of ligand, shifting from inactive to active states (R, R*, R*G) in equilibrium maintaining a low basal level of activation. Opsin is converted to rhodopsin when newly synthesized 11-cis-retinal couples to TM Lys296 [68].

The correspondence in functional states does not address several other intermediate states identified for rhodopsin. Similarly, a number of distinct activation states have been recently reported for several GPCRs activated by diffusible ligands. β2Rs undergo different conformational changes upon binding of partial agonists from these, induced by full agonists [50,69●,70], the resulting state (LR*G) may not be the determinant of agonist efficacy [71]. Receptor activation involves a series of concerted conformational changes, with a number of intermediate steps that may not be easily detectable experimentally. Spectroscopy studies of rhodopsin have indeed characterized a number of identifiable structural intermediates in its activation mechanism, reviewed recently [13●,72], and the same should be expected for other GPCRs as experimental tools enable such studies. However, the presence of several distinct inactive and active conformational states of a given receptor does not invalidate the notion of inactive versus active receptor states. Initially, a distinction needs to be made between thermodynamically stable functional states under physiological conditions and intermediate steps trapped under non-physiological conditions. The later are useful in understanding the receptor activation mechanism, but do not vitiate the concept of inactive versus active states. Next, in physiologically relevant conditions, there are also a number of distinct active and inactive sub-states of conformations observed for a given receptor. This observation is intrinsic to the relationship between functional states and conformational states depending on the specific ligand, G protein, or other modulators factors. In energetic terms, the active and inactive states can be considered as two different global minima, each represented by a number of sub-states or local minima. In structural terms, this means that there is a set of characteristic conformational changes that define and differentiate the inactive from the active form of the receptor, likely represented by changes within the conserved functional microdomains. An example would be the movement of TM6 away from TM3 at the cytoplasmic side that has been consistently shown in different GPCRs, yet the degree and stability of this movement may be slightly different for different receptors and different agonists within the same receptor (Figure 6). These characteristic conformational changes will be accompanied by other structural rearrangements that would differ, depending on the interaction of a given receptors with specific ligands, G proteins, or other conformational modulators.

Given the correspondence among functional states between rhodopsin and GPCRs activated by diffusible ligands, it needs to be determined whether the activated state of rhodopsin, MII, is a suitable structural template of the active state of these GPCRs. As discussed in the Introduction, rhodopsin shares with Family 1 a set of three conserved functional microdomains, whose functional role is to support the conformational changes responsible for receptor activation. Conformational changes upon receptor activation identified for rhodopsin or other GPCRs are, so far, consistent with each other. The movement of TM6 away from TM3 identified initially in rhodopsin [36,48●●] has been validated in the β2R [24,73●●, Ballesteros et al, personal communication] and in the M2 and M3 muscarinic receptors [74] (Figure 6). The functional phenotype of mutations of these conserved residues in rhodopsin are often consistent with the results obtained for other members of Family 1, further suggesting that these conserved residues performed equivalent roles in a commonly shared activation mechanism. Whenever the equivalent mutation in rhodopsin and another GPCR lead to the functional phenotype in one but not in the other, it is most likely because the altered interactions differ minimally in the energy between wild-type and mutant states, which does not reach the threshold required for experimental detection. Thus it seems likely that the structure of the activated state of rhodopsin will resemble the active state of other Family 1 GPCRs, and serve as a suitable template to model the agonist-bound conformation of these other receptors. In the absence of a known structure, current attempts to develop molecular models of the active state of a GPCR are based on the inactive rhodopsin structure as a starting point, implementing computationally the set of proposed conformational changes that lead to receptor activation (reviewed in [29]).

Conclusions and future directions: GPCR structures beyond rhodopsin

Which additional GPCR structures are necessary to support drug discovery efforts on this class of proteins is an open question. The crystal structure of the first GPCR [14●●] has generated an expectation that other GPCR structures may soon become available. However, given the difficulty of elucidating the structures of membrane proteins such as GPCRs and the track record of GPCR crystallography, the expectation should be limited to a few novel structures.

Inactive states of GPCRs

The structural information will be frequently generated by homology modeling techniques, and thus the question is: which GPCR structural templates will be most useful? Homology modeling approaches require a significant degree of sequence homology; the structure of those receptors without a suitable structural template would thus be the most attractive targets. Rhodopsin seems to be a suitable template for most of Family 1, therefore, significant efforts should be devoted to elucidate the structure of a representative from Family 2 and Family 3. As drug discovery is centered on drug recognition, the structure of different receptors bound to a variety of ligands would be highly desirable. The structure of the receptor in the absence of a ligand will also be required to understand the conformational changes necessary for ligand binding. The structure of opsin, the unbound form of rhodopsin, which behaves functionally as a typical uncomplexed GPCR (Scheme 1), seems highly desirable. These include structures for Family 1 other than rhodopsin, which would provide better templates for other closely related GPCRs. Emphasis should be on the structures of different receptors subclasses with different types of ligands such as neurotransmitters, small peptides, small folded proteins such as chemokines, glyco-hormone receptors and lipid-like ligand receptors. In fact, the difficulty in crystallizing GPCRs suggests that we should learn as much as possible from available structures or variations of the same theme, which for now means rhodopsin.

Active states of GPCRs

Most GPCR drug discovery projects are focused on Family 1, because it is by far the largest of the three families. The structure of rhodopsin is a suitable template for the inactive state of most Family 1 GPCRs, and the wealth of experimental data indicates that the structure of the active state of rhodopsin is also a suitable template for the active state of other members of Family 1 [2●,28●●,29]. A major goal of future crystallography efforts is to elucidate the structure of rhodopsin in its active state. The availability of structures for the inactive and active state of the same representative of Family 1 GPCRs would comprise the minimal set of structures necessary to understand GPCR function in regards to drug discovery efforts. While a structural template of the inactive state enables structure-based approaches to discover new antagonist drugs, a structural template of the active state would enable structure-based approaches to discover new agonist drugs. The combination of both structural templates would also empower novel approaches to understanding the molecular determinants of agonistic versus antagonistic actions of related molecules. Therefore, the next goal would be to crystallize inactive and active state conformations bound to chemically related agonist and antagonists. The observation of a common set of functional microdomains spanning TM domain, whose functional role would be to support a conserved receptor activation mechanism, suggests that such understanding may arise from the crystal structure of activated rhodopsin. Furthermore, the hypothesis that the agonist triggering mechanism may rely on unleashing the constraints imposed upon the aromatic cluster motif centered on TM6 [29], consistent with the structure, SAR studies and spectroscopic data on rhodopsin indicates that the active state structure of rhodopsin may shed light on the triggering mechanisms by which agonists activate other GPCRs. Because elucidating molecular mechanisms often requires structural information on the intermediates, the structure of variant receptors with activating or inactivating phenotypes would be informative.

Beyond the essential functional states to characterize structurally GPCRs defined by the inactive and active states of the same receptor, there are other states that should shed light on GPCR function. Of special interest are the complexes between GPCRs and proteins that mediate and/or modulate their function, such as G proteins and arrestins. These complexes may indeed provide the best tools to stabilize and crystallize the active state of GPCRs, which has been proposed to be intrinsically dynamic [73●●,75] and unstable [67]. Recently, the complex between the FPR receptor and G proteins or arrestin has been proposed to stabilize the receptor in the activated state [76]. Interestingly, functional fusion proteins were made between β2R and G protein receptor kinase 2 (GRK2) [77]. There is considerable interest in the possibility that GPCR homo- and hetero-dimerization may affect the structural and pharmacological properties of these receptors [78●,79]. Although the current structure of rhodopsin does not provide a structural template for these receptor-receptor complexes, the elucidation of the structure of physiologically relevant GPCR dimers would also be attractive.

The structure of the receptor-G protein complex would also be useful in identifying their interacting surfaces. The sequence and structures of both Family 1 and G proteins seem to be conserved through evolution, and the structure of a representative receptor-G protein pair would also qualify as a suitable template for other receptor-G protein complexes. The first structure of this complex would enable powerful protein engineering approaches to modulate the recognition of GPCRs by G proteins, which would then support the development of sensitive screening assays. Structural elucidation of variant receptors with altered functional phenotypes would provide insight into the mechanism by which receptors catalyze the exchange of GDP to GTP in the G protein. Understanding the molecular basis of receptor-G protein recognition would give us insight into perplexing observation that receptors without detectable sequence homology are nonetheless capable of activating similar G proteins. Other modulatory proteins or modifications of the receptors would also be candidates for structural elucidation, such as the phosphorylated state of the receptor to characterize the structure of the desensitized state of a GPCR.

In summary, it is conceivable that in the near future there will be several GPCRs for which a crystallization protocol has been established that enable the systematic crystallization of several variant receptors. This scenario will enable a focus on information-rich structures, based on the elucidation of multiple, related structures of the same receptor complexed with multiple and related ligands and mutant receptors. This series of structures will enable us to address the challenging issues that explore localized conformational rearrangements which currently hinder molecular approaches to GPCR ligand recognition, resulting in an effective structure-based drug discovery platform.

Acknowledgments

We would like to thank Drs Ning Li, Harel Weinstein, Irache Visiers and Jonathan Javitch for helpful discussions, Dr Slawomir Filipek for his help with the preparation of Figure 3, and J Preston Van Hooser, Connie Arthur and Naomi Wilson for their help during the preparation of the manuscript. This research was supported by NIH grants AI48517 (JB) and EY08061 (KP), a grant from Research to Prevent Blindness Inc (RPB) to the Department of Ophthalmology at the University of Washington, and the EK Bishop Foundation. KP is a recipient of an RPB Senior Investigator Award.

References

●● of outstanding interest

● of special interest

1. Sautel M, Milligan G. Molecular manipulation of G-protein-coupled receptors: A new avenue into drug discovery. Curr Med Chem. 2000;7:889–896. [PubMed]
2. Muller G. Towards 3D structures of G protein-coupled receptors: A multidisciplinary approach. Curr Med Chem. 2000;7:861–888.This review summarizes the recent results on GPCR structural studies. [PubMed]
3. Bargmann CI. Neurobiology of the Caenorhabditis elegans genome. Science. 1998;282:2028–2033. [PubMed]
4. Larhammar D, Blomqvist AG, Wahlestedt C. The receptor revolution - multiplicity of G-protein-coupled receptors. Drug Des Discov. 1993;9:179–188. [PubMed]
5. Howard AD, McAllister G, Feighner SD, Liu Q, Nargund RP, Van der Ploeg LH, Patchett AA. Orphan G-protein-coupled receptors and natural ligand discovery. Trends Pharmacol Sci. 2001;22:132–140. [PubMed]
6. Attwood TK. A compendium of specific motifs for diagnosing GPCR subtypes. Trends Pharmacol Sci. 2001;22 :162–165. [PubMed]
7. Sali A. 100,000 protein structures for the biologist. Nature Struct Biol. 1998;5:1029–1032. [PubMed]
8. Marti-Renom MA, Stuart AC, Fiser A, Sanchez R, Melo F, Sali A. Comparative protein structure modeling of genes and genomes. Annu Rev Biophys Biomol Struct. 2000;29:291–325. [PubMed]
9. Al-Lazikani B, Jung J, Xiang Z, Honig B. Protein structure prediction. Curr Opin Chem Biol. 2001;5:51–56. [PubMed]
10. Mason JS, Good AC, Martin EJ. 3-D pharmacophores in drug discovery. Curr Pharm Des. 2001;7:567–597. [PubMed]
11. Okada T, Palczewski K: Crystal structure of rhodopsin: Implication for vision and beyond.Curr Opin Struct Biol (2001) in press. [PubMed]
12. Teller DC, Okada T, Behnke CA, Palczewski K, Stenkamp RE. Advances in determination of a high-resolution three-dimensional structure of rhodopsin, a model of G-protein-coupled receptors (GPCRs) Biochemistry. 2001;40:7761–7772. ●● The paper described the most current information on the refined structure of rhodopsin, with the newest accession number to the PDB file. [PMC free article] [PubMed]
13. Okada T, Ernst OP, Palczewski K, Hofmann KP. Activation of rhodopsin: New insights from structural and biochemical studies. Trends Biochem Sci. 2001;26:318–324.Insightful review of rhodopsin activation. [PubMed]
14. Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, Stenkamp RE, Yamamoto M, Miyano M. Crystal structure of rhodopsin: A G protein-coupled receptor. Science. 2000;289:739–745. ●● This study describes the first high-resolution structure of rhodopsin. [PubMed]
15. Polans A, Baehr W, Palczewski K. Turned on by Ca2+! The physiology and pathology of Ca2+-binding proteins in the retina. Trends Neurosci. 1996;19:547–554. [PubMed]
16. Hamm HE. The many faces of G protein signaling. J Biol Chem. 1998;273:669–672. [PubMed]
17. Gilman AG. Nobel Lecture. G proteins and regulation of adenylyl cyclase. Biosci Rep. 1995;15:65–97. [PubMed]
18. Rodbell M. Nobel lecture. Signal transduction: Evolution of an idea. Biosci Rep. 1995;15:117–133. [PubMed]
19. Liebmann C, Bohmer FD. Signal transduction pathways of G protein-coupled receptors and their cross-talk with receptor tyrosine kinases: Lessons from bradykinin signaling. Curr Med Chem. 2000;7:911–943. [PubMed]
20. Brzostowski JA, Kimmel AR. Signaling at zero G: G-protein-independent functions for 7-TM receptors. Trends Biochem Sci. 2001;26:291–297. [PubMed]
21. Hall RA, Premont RT, Lefkowitz RJ. Heptahelical receptor signaling: Beyond the G protein paradigm. J Cell Biol. 1999;145:927–932. [PMC free article] [PubMed]
22. Gether U. Uncovering molecular mechanisms involved in activation of G protein-coupled receptors. Endocr Rev. 2000;21:90–113.Informative review on GPCR structure-function. [PubMed]
23. Frimurer TM, Bywater RP. Structure of the integral membrane domain of the GLP1 receptor. Proteins. 1999;35 :375–386. [PubMed]
24. Sheikh SP, Vilardarga JP, Baranski TJ, Lichtarge O, Iiri T, Meng EC, Nissenson RA, Bourne HR. Similar structures and shared switch mechanisms of the β2-adrenoceptor and the parathyroid hormone receptor. Zn(II) bridges between helices III and VI block activation. J Biol Chem. 1999;274:17033–17041. [PubMed]
25. Weber IT. Evaluation of homology modeling of HIV protease. Proteins. 1990;7:172–184. [PubMed]
26. Ballesteros JA, Weinstein H. Integrated methods for the construction of three dimensional models and computational probing of structure-function relations in G-protein coupled receptors. Methods Neurosci. 1995;25:366–428.
27. Ballesteros J, Kitanovic S, Guarnieri F, Davies P, Fromme BJ, Konvicka K, Chi L, Millar RP, Davidson JS, Weinstein H, Sealfon SC. Functional microdomains in G-protein-coupled receptors. The conserved arginine-cage motif in the gonadotropin-releasing hormone receptor. J Biol Chem. 1998;273:10445–10453. [PubMed]
28. Ballesteros JA, Shi L, Javitch JA. Structural mimicry in G-protein-coupled receptors: Implications of the high-resolution structure of rhodopsin for structure-function analysis of rhodopsin-like receptors. Mol Pharmacol. 2001;60:1–19. ●● A comprehensive review on GPCR activation. [PubMed]
29. Visiers I, Weinstein H, Ballesteros J: Methods for the prediction and molecular modeling of membrane proteins: Application to G protein coupled receptors.Methods Enzymol (2001) in press.
30. Gershengorn MC, Osman R. Minireview: Insights into G protein-coupled receptor function using molecular models. Endocrinology. 2001;142:2–10. [PubMed]
31. Cai K, Klein-Seetharaman J, Hwa J, Hubbell WL, Khorana HG. Structure and function in rhodopsin: Effects of disulfide cross-links in the cytoplasmic face of rhodopsin on transducin activation and phosphorylation by rhodopsin kinase. Biochemistry. 1999;38:12893–12898. [PubMed]
32. Altenbach C, Klein-Seetharaman J, Hwa J, Khorana HG, Hubbell WL. Structural features and light-dependent changes in the sequence 59–75 connecting helices I and II in rhodopsin: A site-directed spin-labeling study. Biochemistry. 1999;38:7945–7949. [PubMed]
33. Klein-Seetharaman J, Hwa J, Cai K, Altenbach C, Hubbell WL, Khorana HG. Single-cysteine substitution mutants at amino acid positions 55–75, the sequence connecting the cytoplasmic ends of helices I and II in rhodopsin: Reactivity of the sulfhydryl groups and their derivatives identifies a tertiary structure that changes upon light-activation. Biochemistry. 1999;38:7938–7944. [PubMed]
34. Altenbach C, Cai K, Khorana HG, Hubbell WL. Structural features and light-dependent changes in the sequence 306–322 extending from helix VII to the palmitoylation sites in rhodopsin: A site-directed spin-labeling study. Biochemistry. 1999;38:7931–7937. [PubMed]
35. Cai K, Klein-Seetharaman J, Farrens D, Zhang C, Altenbach C, Hubbell WL, Khorana HG. Single-cysteine substitution mutants at amino acid positions 306–321 in rhodopsin, the sequence between the cytoplasmic end of helix VII and the palmitoylation sites: Sulfhydryl reactivity and transducin activation reveal a tertiary structure. Biochemistry. 1999;38 :7925–7930. [PubMed]
36. Dunham TD, Farrens DL. Conformational changes in rhodopsin. Movement of helix f detected by site-specific chemical labeling and fluorescence spectroscopy. J Biol Chem. 1999;274:1683–1690. [PubMed]
37. Kim JM, Altenbach C, Thurmond RL, Khorana HG, Hubbell WL. Structure and function in rhodopsin: Rhodopsin mutants with a neutral amino acid at E134 have a partially activated conformation in the dark state. Proc Natl Acad Sci USA. 1997;94:14273–14278. [PMC free article] [PubMed]
38. Yang K, Farrens DL, Altenbach C, Farahbakhsh ZT, Hubbell WL, Khorana HG. Structure and function in rhodopsin. Cysteines 65 and 316 are in proximity in a rhodopsin mutant as indicated by disulfide formation and interactions between attached spin labels. Biochemistry. 1996;35:14040–14046. [PubMed]
39. Cai K, Langen R, Hubbell WL, Khorana HG. Structure and function in rhodopsin: Topology of the C-terminal polypeptide chain in relation to the cytoplasmic loops. Proc Natl Acad Sci USA. 1997;94:14267–14272. [PMC free article] [PubMed]
40. Greenhalgh DA, Altenbach C, Hubbell WL, Khorana HG. Locations of Arg-82, Asp-85, and Asp-96 in helix C of bacteriorhodopsin relative to the aqueous boundaries. Proc Natl Acad Sci USA. 1991;88:8626–8630. [PMC free article] [PubMed]
41. Resek JF, Farahbakhsh ZT, Hubbell WL, Khorana HG. Formation of the meta II photointermediate is accompanied by conformational changes in the cytoplasmic surface of rhodopsin. Biochemistry. 1993;32 :12025–12032. [PubMed]
42. Altenbach C, Greenhalgh DA, Khorana HG, Hubbell WL. A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: Application to spin-labeled mutants of bacteriorhodopsin. Proc Natl Acad Sci USA. 1994;91:1667–1671. [PMC free article] [PubMed]
43. Steinhoff HJ, Mollaaghababa R, Altenbach C, Hideg K, Krebs M, Khorana HG, Hubbell WL. Time-resolved detection of structural changes during the photocycle of spin-labeled bacteriorhodopsin. Science. 1994;266:105–107. [PubMed]
44. Farahbakhsh ZT, Ridge KD, Khorana HG, Hubbell WL. Mapping light-dependent structural changes in the cytoplasmic loop connecting helices C and D in rhodopsin: A site-directed spin labeling study. Biochemistry. 1995;34:8812–8819. [PubMed]
45. Steinhoff HJ, Mollaaghababa R, Altenbach C, Khorana HG, Hubbell WL. Site directed spin labeling studies of structure and dynamics in bacteriorhodopsin. Biophys Chem. 1995;56 :89–94. [PubMed]
46. Yang K, Farrens DL, Hubbell WL, Khorana HG. Structure and function in rhodopsin. Single cysteine substitution mutants in the cytoplasmic interhelical E-F loop region show position- specific effects in transducin activation. Biochemistry. 1996;35:12464–12469. [PubMed]
47. Altenbach C, Yang K, Farrens DL, Farahbakhsh ZT, Khorana HG, Hubbell WL. Structural features and light-dependent changes in the cytoplasmic interhelical E-F loop region of rhodopsin: A site-directed spin-labeling study. Biochemistry. 1996;35:12470–12478. [PubMed]
48. Farrens DL, Altenbach C, Yang K, Hubbell WL, Khorana HG. Requirement of rigid-body motion of transmembrane helices for light activation of rhodopsin. Science. 1996;274:768–770. ●● This study suggests the importance of a helical movement for the activation of rhodopsin. [PubMed]
49. Javitch JA, Shi L, Simpson MM, Chen J, Chiappa V, Visiers I, Weinstein H, Ballesteros JA. The fourth transmembrane segment of the dopamine D2 receptor: Accessibility in the binding-site crevice and position in the transmembrane bundle. Biochemistry. 2000;39:12190–12199. [PubMed]
50. Gether U, Lin S, Ghanouni P, Ballesteros JA, Weinstein H, Kobilka BK. Agonists induce conformational changes in transmembrane domains III and VI of the β2 adrenoceptor. EMBO J. 1997;16:6737–6747. [PMC free article] [PubMed]
51. Jung H, Windhaber R, Palm D, Schnackerz KD. Conformation of a β-adrenoceptor-derived signal transducing peptide as inferred by circular dichroism and 1H NMR spectroscopy. Biochemistry. 1996;35:6399–6405. [PubMed]
52. Chi L, Zhou W, Prikhozhan A, Flanagan C, Davidson JS, Golembo M, Illing N, Millar RP, Sealfon SC. Cloning and characterization of the human GnRH receptor. Mol Cell Endocrinol. 1993;91:R1–R6. [PubMed]
53. Zhao MM, Hwa J, Perez DM. Identification of critical extracellular loop residues involved in α1-adrenergic receptor subtype-selective antagonist binding. Mol Pharmacol. 1996;50:1118–1126. [PubMed]
54. Javitch JA, Ballesteros JA, Chen J, Chiappa V, Simpson MM. Electrostatic and aromatic microdomains within the binding-site crevice of the D2 receptor: Contributions of the second membrane-spanning segment. Biochemistry. 1999;38:7961–7968. [PubMed]
55. Simpson MM, Ballesteros JA, Chiappa V, Chen J, Suehiro M, Hartman DS, Godel T, Snyder LA, Sakmar TP, Javitch JA. Dopamine D4/D2 receptor selectivity is determined by a divergent aromatic microdomain contained within the second, third, and seventh membrane-spanning segments. Mol Pharmacol. 1999;56:1116–1126. [PubMed]
56. Isele J, Sakmar TP, Siebert F. Rhodopsin activation affects the environment of specific neighboring phospholipids: An FTIR spectroscopic study. Biophys J. 2000;79:3063–3071. [PMC free article] [PubMed]
57. Strader CD, Candelore MR, Hill WS, Sigal IS, Dixon RA. Identification of two serine residues involved in agonist activation of the β-adrenergic receptor. J Biol Chem. 1989;264:13572–13578. [PubMed]
58. Baldwin JM, Schertler GF, Unger VM. An α-carbon template for the transmembrane helices in the rhodopsin family of G-protein-coupled receptors. J Mol Biol. 1997;272:144–164. [PubMed]
59. Liapakis G, Ballesteros JA, Papachristou S, Chan WC, Chen X, Javitch JA. The forgotten serine. A critical role for Ser-203542 in ligands binding to and activation of the β2-adrenergic receptor. J Biol Chem. 2000;275:37779–37788. [PubMed]
60. Ballesteros JA, Deupi X, Olivella M, Haaksma EE, Pardo L. Serine and threonine residues bend α-helices in the chi(1) = g(-) conformation. Biophys J. 2000;79:2754–2760. [PMC free article] [PubMed]
61. Sansom MS, Weinstein H. Hinges, swivels and switches: The role of prolines in signaling via transmembrane α-helices. Trends Pharmacol Sci. 2000;21:445–451. [PubMed]
62. Ferguson SS. Evolving concepts in G protein-coupled receptor endocytosis: The role in receptor desensitization and signaling. Pharmacol Rev. 2001;53:1–24. [PubMed]
63. Sexton PM, Albiston A, Morfis M, Tilakaratne N. Receptor activity modifying proteins. Cell Signal. 2001;13:73–83. [PubMed]
64. Lefkowitz RJ. The superfamily of heptahelical receptors. Nature Cell Biol. 2000;2:E133–E136. [PubMed]
65. Palczewski K, Benovic JL. G-protein-coupled receptor kinases. Trends Biochem Sci. 1991;16:387–391. [PubMed]
66. Palczewski K. Structure and functions of arrestins. Protein Sci. 1994;3:1355–1361. [PMC free article] [PubMed]
67. Gether U, Ballesteros JA, Seifert R, Sanders-Bush E, Weinstein H, Kobilka BK. Structural instability of a constitutively active G protein-coupled receptor. Agonist-independent activation due to conformational flexibility. J Biol Chem. 1997;272:2587–2590. [PubMed]
68. McBee JK, Palczewski K, Baehr W, Pepperberg DR. Confronting complexity: The interlink of phototransduction and retinoid metabolism in the vertebrate retina. Prog Retin Eye Res. 2001;20:469–529. [PubMed]
69. Ghanouni P, Steenhuis JJ, Farrens DL, Kobilka BK. Agonist-induced conformational changes in the G-protein-coupling domain of the β2 adrenergic receptor. Proc Natl Acad Sci USA. 2001;98:5997–6002.This study indicate a general mechanism for GPCR activation for the β2-adrenergic receptor and rhodopsin. [PMC free article] [PubMed]
70. Ghanouni P, Gryczynski Z, Steenhuis JJ, Lee TW, Farrens DL, Lakowicz JR, Kobilka BK: Functionally different agonists induce distinct conformations in the G protein coupling domain of the β2 adrenergic receptor.J Biol Chem (2001) 276 in press. [PubMed]
71. Seifert R, Wenzel-Seifert K, Gether U, Kobilka BK. Functional differences between full and partial agonists: Evidence for ligand-specific receptor conformations. J Pharmacol Exp Ther. 2001;297:1218–1226. [PubMed]
72. Shichida Y, Imai H. Visual pigment: G-protein-coupled receptor for light signals. Cell Mol Life Sci. 1998;54:1299–1315. [PubMed]
73. Jensen AD, Guarnieri F, Rasmussen SG, Asmar F, Ballesteros JA, Gether U. Agonist-induced conformational changes at the cytoplasmic side of transmembrane segment 6 in the β2 adrenergic receptor mapped by site-selective fluorescent labeling. J Biol Chem. 2001;276:9279–9290. ●● This study demonstrate structural changes of a GPCR that reflect an agonist-promoted movement of the cytoplasmic part of TM 6 away from the receptor core and upwards toward the membrane bilayer. [PubMed]
74. Zeng FY, Hopp A, Soldner A, Wess J. Use of a disulfide cross-linking strategy to study muscarinic receptor structure and mechanisms of activation. J Biol Chem. 1999;274:16629–16640. [PubMed]
75. Hubbell WL, Cafiso DS, Altenbach C. Identifying conformational changes with site-directed spin labeling. Nature Struct Biol. 2000;7:735–739. [PubMed]
76. Bennett TA, Key TA, Gurevich VV, Neubig R, Prossnitz ER, Sklar LA. Real-time analysis of G protein coupled receptor reconstitution in a solubilized system. J Biol Chem. 2001;276:22453–22460. [PubMed]
77. Shiina T, Arai K, Tanabe S, Yoshida N, Haga T, Nagao T, Kurose H: Clathrin-box in G protein-coupled Receptor Kinase 2.J Biol Chem (2001) 276 in press. [PubMed]
78. Bouvier M. Oligomerization of G-protein-coupled transmitter receptors. Nature Rev Neurosci. 2001;2:274–286.Comprehensive review on the dimerization of GPCR. [PubMed]
79. George SR, Fan T, Xie Z, Tse R, Tam V, Varghese G, O'Dowd BF. Oligomerization of μ- and δ-opioid receptors. Generation of novel functional properties. J Biol Chem. 2000;275:26128–26135. [PubMed]
80. West AP, Jr, Llamas LL, Snow PM, Benzer S, Bjorkman PJ. Crystal structure of the ectodomain of Methuselah, a Drosophila G protein-coupled receptor associated with extended lifespan. Proc Natl Acad Sci USA. 2001;98:3744–3749. [PMC free article] [PubMed]
81. Kunishima N, Shimada Y, Tsuji Y, Sato T, Yamamoto M, Kumasaka T, Nakanishi S, Jingami H, Morikawa K. Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature. 2000;407:971–977.Structure of the extracellular domain of the glutamate receptor that binds the ligand. [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...