G protein-coupled receptors: structure- and function-based drug discovery

As one of the most successful therapeutic target families, G protein-coupled receptors (GPCRs) have experienced a transformation from random ligand screening to knowledge-driven drug design. We are eye-witnessing tremendous progresses made recently in the understanding of their structure–function relationships that facilitated drug development at an unprecedented pace. This article intends to provide a comprehensive overview of this important field to a broader readership that shares some common interests in drug discovery.


INTRODUCTION
G protein-coupled receptors (GPCRs) represent the largest protein family encoded by the human genome. Located on the cell membrane, they transduce extracellular signals into key physiological effects. 1 Their endogenous ligands include odors, hormones, neurotransmitters, chemokines, etc., varying from photons, amines, carbohydrates, lipids, peptides to proteins. GPCRs have been implicated in a large number of diseases, such as type 2 diabetes mellitus (T2DM), obesity, depression, cancer, Alzheimer's disease, and many others. 2 Activated by external signals through coupling to different G proteins or arrestins, GPCRs elicit cyclic adenosine 3,5-monophosphate (cAMP) response, calcium mobilization, or phosphorylation of extracellular regulated protein kinases 1/2 (pERK1/2). 3 The seven-transmembrane protein property endows them easy to access, while the diversified downstream signaling pathways make them attractive for drug development. 4 The human GPCR family is divided into classes A (rhodopsin), B (secretin and adhesion), C (glutamate), and F (Frizzled) subfamilies according to their amino acid sequences (Fig. 1). Of the 826 human GPCRs, approximately 350 non-olfactory members are regarded as druggable and 165 of them are validated drug targets ( Fig. 1 and Table S1). [4][5][6] Latest statistical data indicate that 527 Food and Drug Administration (FDA)-approved drugs 4 and ∼60 drug candidates currently in clinical trials target GPCRs (Table S1). 5 Started with crystal structure determination and accelerated by cryo-electron microscopy (cryo-EM) technology, three-dimensional (3D) structural studies on a variety of GPCRs in complex with ligands, G proteins/arrestins, or both 7-10 (involving 455 structures from 82 different receptors) significantly deepened our knowledge of molecular mechanisms of signal transduction. Novel insights into ligand recognition and receptor activation are gained from inactive, transitional, active, and apo states, thereby offering new opportunities for structure-based drug design (SBDD). 11 Pharmacological parameters such as cAMP accumulation, calcium flux, ERK phosphorylation, arrestin recruitment, and G protein interaction, 12,13 are commonly used to evaluate ligand action and biased signaling. Ligand-binding kinetics and signaling timing render another dimension for interpreting signal bias profiles and link in vitro bioactivities with in vivo effects. 14 In this process, a series of biased and allosteric modulators were discovered by rational design, ligand screening, and pharmacological assessment leading to the identification of novel binding sites or action modes. 15,16 Apart from crystallography and cryo-EM, the striking advancement in GPCR biology is also attributable to the deployment of powerful technologies such as nuclear magnetic resonance (NMR), hydrogen-deuterium exchange (HDX), fluorescence resonance energy transfer, bioluminescence resonance energy transfer, surface plasmon resonance, single molecule fluorescence, CRISPR/Cas9, artificial intelligence, etc. This review systematically summarizes the latest information on this important drug target family to cover both basic and translational sciences in the context of drug discovery and development.

GPCR AS DRUG TARGET
Class A Class A GPCRs, the so called "rhodopsin-like family" consisting of 719 members, are divided into several subgroups: aminergic, peptide, protein, lipid, melatonin, nucleotide, steroid, alicarboxylic acid, sensory, and orphan. 17 They have a conventional transmembrane domain (TMD) that forms ligand-binding pocket and additional eight helices with a palmitoylated cysteine at the C terminal. 18,19 Given the wide range of their physiological functions, this class of receptors is the most targeted therapeutically among all other classes. By manually curating Drugs@FDA original New Drug Application (NDA) and Biologic License Application (BLA) database (data extracted from August 2017 to June 2020) and cross-referencing with Drugbank, 20 IUPHAR and ChemBL databases, we were able to find the approved drugs associated with this class.
Over 500 GPCR drugs target class A and many of them act at >1 receptor: 75% are made against aminergic receptors and 10% for peptidic ligand receptors with indications ranging from analgesics, allergies, cardiovascular diseases, hypertension, pulmonary diseases, depression, migraine, glaucoma, Parkinson's disease to schizophrenia, cancer-related fatigue, etc. Approximately 500 novel drug candidates are in clinical trials. Of them, 134 are for peptide-activated GPCRs, while small molecules still occupy the majority. It is noted that 6% of class A members are sensory and alicarboxylic acid receptors that have broad untapped therapeutic potentials (Table S1). Chemokine, prostanoid and melanocortin receptors constitute >8% clinical trial targets in this class.
In the past 3 years, about 20 NDAs were approved targeting mostly peptide and aminergic receptors (Table 1). Siponimod and ozanimod provide alternatives to fingolimod (approved in 2010) for treating relapsing forms of multiple sclerosis by modulating sphingosine-1-phosphate receptor. Two radiolabeled ligands, gallium 68 dotatoc and lutetium 177 dotatate, have been approved for neuroendocrine tumor and pancreatic gastrointestinal cancer diagnosis, respectively. Pitolisant, a selective inverse Fig. 1 Phylogenetic tree of GPCRs as drug targets. Node represents GPCR named according to its gene name. Receptors with approved drugs on the market are highlighted by color. GPCRs are organized according to GPCR database. 4 Approved drug list was derived from previous publications, 4,11 complemented by additional search of newly approved entities at Drugs@FDA (accessdata.fda.gov) until June 2020. See Table  S2 for details substance-P receptor agonist of histamine receptor, is used to treat narcolepsy-related daytime sleepiness, while lemborexant, an orexin receptor antagonist, is used for insomnia management. Gilteritinib (ASP2215) is a small molecule inhibitor of tyrosine kinase. However, it also antagonizes serotonin receptors without any reported pharmacological consequences. Revefenacin is a longacting antagonist of muscarinic acetylcholine receptors (mAChRs) indicated for chronic obstructive pulmonary disease. Amisulpride, trialed for antiemetic and schizophrenia, was finally approved for antiemetic in 2020. This molecule is acting as an antagonist against dopamine and serotonin receptors. Fosnetupitant, a prodrug of netupitant, was approved for chemotherapy-induced nausea and vomiting. Cysteamine treats radiation sickness via modifying action of neuropeptide Y receptor. Cannabidiol is one the active constituents of the Cannabis plant and was trialed for schizophrenia, graft versus host disease, and anticonvulsant. It was eventually approved in 2018 for the treatment of severe forms of epilepsy-Lennox-Gastaut syndrome and Dravet syndrome. Meanwhile, fostamatinib, indicated for chronic immune thrombocytopenia, targets >300 receptors and enzymes, including adenosine receptor A3.
Class B This class of GPCRs is divided into two subfamilies: secretin (B1) and adhesion (B2), containing 15 and 33 members, respectively. 4,21 Secretin subfamily members are characteristic of large extracellular domains (ECDs) and bind to vasoactive intestinal peptide (VIP), pituitary adenylate cyclase-activating peptide (PACAP), corticotropin-releasing factor (CRF), parathyroid peptide hormone (PTH), growth hormone-releasing hormone (GHRH), calcitonin gene-related peptide (CGRP), glucagon, and glucagonlike peptides (GLPs), respectively. Adhesion subfamily has nine subgroups, possessing unique N-terminal motifs, such as epidermal growth factor, cadherin, and immunoglobulin domains. They are distinguished from other GPCRs due to their roles in cell adhesion and migration. 22,23 Apart from the long N-terminal domain, other unique features of the B2 subfamily are the GPCR autoproteolysis-inducing domain and the proteolysis site that are responsible for signaling activation through a Stachel sequence (a tethered agonist) and producing N-terminal fragment (NTF) and Cterminal fragment. The hallmarks of the B2 GPCR subfamily are a two-step activation model, the ligand-NTF interaction and the Stachel signaling/basal activity. Adhesion receptors can also signal independently of fragment dissociation and this has complicated pharmacological consequences. 22,24,25 In this class, receptors of glucagon family peptides, followed by CGRP, PTH, GHRH, CRF, VIP, and PACAP, constitute major targets for therapeutic intervention (Table S1) of various diseases, including obesity, T2DM, osteoporosis, migraine, depression, and anxiety. 26,27 To date, multiple GLP-1 receptor (GLP-1R) agonists have been developed by a combination of selective amino acid substitutions, enzymatic cleavage blockade, and conjugation to entities that increase binding to plasma proteins. These methods not only slow down fast renal clearance of the peptides but also extend their half-lives. Dose-dependent side effects such as nausea and gastrointestinal adverse events are the main drawbacks that are becoming more of a compliant with dose scaling. 28,29 For instance, one newly approved GLP-1R agonist, semaglutide, has a noticeable half-life of 168 h thereby allowing weekly subcutaneous administration, while oral semaglutide (approved in 2019) formulated using absorption enhancer shows a similar half-life but is dosed daily with reported side effects (Table 2). 30,31 One of the latest approaches to develop more efficacious therapeutics against T2DM and obesity relates to dual-and triagonists targeting two or more of GLP-1R, glucagon receptor (GCGR), and glucose-dependent insulinotropic peptide receptor (GIPR). Many of them are currently in different phases of clinical trials (Table 3). [32][33][34][35][36][37] Of note, in this receptor family, GLP-2 stimulates intestinal growth and an approved GLP-2R agonist, teduglutide, is used to treat short bowel syndrome. 38 CGRP family has a considerable clinical relevance. For instance, pramlintide that targets amylin receptor is utilized to treat both type 1 and type 2 diabetes. Salmon calcitonin has been explored as a treatment for Paget's disease and metabolic disorders. [39][40][41] Furthermore, the association of migraine and CGRP elevation led to FDA-approved monoclonal antibodies (mAbs) against its receptor, e.g., erenumab and eptinezumab, as well as several small molecule antagonists such as rimegepant and ubrogepant (Table 2). 42,43 Two approved diagnostic agents are analogs of CRF (corticorelin ovine triflutate peptide) and GHRH (sermorelin) for diagnosis of Cushing's disease or ectopic adrenocorticotropic hormone syndrome and growth hormone deficiency, respectively. 44,45 Tesamorelin, another synthetic form of GHRH, was approved in 2010 to treat human immunodeficiency virus (HIV)associated lipodystrophy. 44 PTH analogs, teriparatide and abaloparatide, were approved in 2002 and 2017, respectively, for postmenopausal osteoporosis with similar side effects. However, abaloparatide binds to parathyroid hormone 1 receptor (PTH1R) with higher affinity and selectivity that resulted in greater bone density. 46 No therapeutic agent from the adhesion subfamily has entered clinical trial to date (Table S1). 2,4,47 Although, adhesion GPCRs have shown coupling to heterotrimeric G proteins, the major challenge associated with this family is connecting G protein signals with biological activities. 24 This subfamily was found to play functional roles in the immune, cardiovascular, respiratory, nervous, musculoskeletal, reproductive, renal, integumentary, sensory, endocrine, and gastrointestinal systems, with implications in neurological and neoplastic disorders. 24 For instance, ADGRG1 and ADGRF1 are considered as potential drug targets due to their extensive pathogenetic involvement. Two ADGRG1/ADGRG5 modulators, dihydromunduletone and 3-α-acetoxydihydrodeoxygedunin developed via drug screening efforts, showed disease-related efficacy changes thereby calling for exploration of their activities in a pathological environment. 24,25 However, associated drug resistance may not only hamper disease but also offer insights into potential mechanisms of such resistance and strategies to tackle it.
Classes C and F Class C (glutamate) contains 22 receptors, which are further divided into 5 subfamilies including 1 calcium-sensing receptor (CaSR), 2 gamma-aminobutyric acid (GABA) type B receptors (GABA B1 and GABA B2 ), 3 taste 1 receptors (TS1R1-3), 8 metabotropic glutamate receptors (mGluR1-8), and 8 orphan GPCRs. 48 The distinctive features of glutamate subfamily are their large ECD and obligated constitutive dimer for receptor activation. 49 The structural information of ECD indicates the roles of conserved venus fly trap (VFT) and cysteine-rich domain (CRD) on the ligandbinding site. Two conserved disulfide bonds between VFT domains stabilize the homodimers or heterodimers of class F GPCRs. 50 The cryo-EM structures of the first full-length mGluR5 51 and more recently the GABA B Rs further revealed their assembly mechanism and overall architecture. 52 57 Only one class F GPCR (smoothed receptor SMO) has been validated as a drug target whose small molecule antagonists were approved as anti-neoplastic agents. 58 Other 10 members of this class are all Frizzled receptors (FZD1-10), which mediate Wnt signaling and are essential for embryonic development and adult organisms. FZDs together with cognate Hedgehog and Wnt signal are associated with a variety of diseases such as cancer, fibrosis, and neurodegeneration. 59 They share a conserved CRD in the extracellular part and ECD structures of SMO and FZD2/4/5/7/8 were determined. 60 However, only SMO, FZD4, and FZD5 have TMD structures. [61][62][63] Lack of full-length structures and complexity in signaling pathways impeded drug discovery initiatives. 60 Linking of Wnt with extracellular CRD would activate downstream signaling, while the dimerization process and the interaction between CRD and TMD remain elusive. 64 It is known that the downstream effectors of Wnt signaling consist of β-catenin, planar cell polarity, and Ca 2+ pathways, whereas receptor activation involves in Wnt, Norrin, FZD, LDL receptor-related protein 5/6, and many other co-factors. 64 Key breakthrough is thus required to advance our knowledge of these receptors.

MEDICINAL CHEMISTRY OF GPCR
Agent type Agents targeting GPCRs continue to expand in the past decades. Among them, exogenous small molecules, including traditionally developed synthetic organics, natural products, and inorganics, still dominate with a total percentage of 64% (Fig. 2). Nevertheless, the proportion of small molecules declines since 2010. In addition to traditional ligand discovery, several new modalities appear, though currently at the stage of academic research. Covalent ligands, with the embedding of reactive moieties that can be covalently linked to receptors, significantly enhance the weak binding of unoptimized leads. 65,66 Photoactive ligands, developed by the introduction of photo-responsive groups to drug candidates, bring a new interdisciplinary field, photopharmacology. Albeit in its infancy, it has already found in vivo applications. 67,68 In comparison, biologicals, such as peptides, antibodies, and metabolites, become more and more visible in the list. Particularly, the number of approved peptide drugs occupies approximately one third of the whole repertoire, with many more in different clinical stages as the pipeline 41,69 -most of them target classes A and B GPCRs. Naturally occurring peptides have been continually discovered from plants, animals, fungi, and bacteria. Although they act as efficient chemical messengers to modulate cellular functions, these peptides suffer from unfavorable pharmacokinetic and pharmacodynamics properties, such as very short plasma halflives and low plasma protein binding. Therefore, chemical modifications are required to promote the membrane permeability, brain penetration, and oral bioavailability. 70 Available strategies include peptide cyclization, N-methylation, palmitoylation, unnatural amino acid insertion, peptide-small molecule conjugation, and peptide self-assembly. By the way, developing peptidic agents may offer a new approach to de-orphanize certain orphan GPCRs. 71 mAbs represent a promising alternative in GPCR drug discovery. 72,73 Over small molecules, mAbs possess obvious advantages of improved specificity, affinity, and other pharmacological properties. Thus they are being developed against cancers, inflammation, and metabolic disorders. To date, three GPCRtargeting mAbs were approved (mogamulizumab, erenumab, and eptinezumab) while bi-specific antibodies, nanobodies, antibody-drug conjugates, and antibody-peptide conjugation are also in the development stage.
The emergence of many conceptually new molecular entities, such as RNA aptamer, provides not only powerful tool for biophysical study but also potential therapeutic candidates. 74 Usually, aptamer has great molecular diversity and little immunogenicity. 75 In addition, GPCRs are known to function by forming dimers (homodimers or heterodimers) and oligomers on the cell membrane. 76 Therefore, strategies to induce receptor dimerization and/or oligomerization have received attention using scaffolds based on DNA (aptamer), small molecule, and physical stimuli. 77 Structure-activity relationship (SAR) Studies of SARs are critical to the identification of drug-like molecules, especially when the crystal or cryo-EM structure of a drug target is not available. Given that many 3D GPCR structures have been solved in the past decade, most approved drugs were discovered without relevant structural information. Two examples are reviewed below to show the importance of SAR analysis.
Orexin-1 and orexin-2 receptors (also known as hypocretin receptors, OX 1 R and OX 2 R) are class A GPCRs for which two endogenous peptide ligands were identified, orexin A and orexin B (also known as hypocretin 1 and hypocretin 2). The orexin signaling system plays a crucial role in regulating the sleep/wake cycle-both OX 1 R and OX 2 R are involved while the precise contribution of each has yet to be defined. Therefore, dual antagonists were developed as potential treatment for insomnia. 78 Suvorexant (belsomra), the first-in-class dual orexin receptor antagonist, was launched in 2014. 79 The second, lemborexant/ E2006 (dayvigo) developed by Eisai, was approved by the FDA in 2020. 80 It started from hit compound 1 (6, Fig. 3) with modest binding affinity to OX 2 R (K i = 8.7 µM) and no affinity for OX 1 R. 81 The first round of SAR studies revealed that changing the ketone group to an amide led to a remarkable enhancement (~1000-fold) of binding affinity at both OX 1 R and OX 2 R (compound 2). Substitution of the aniline group with a 2-amino-5-cyano pyridine (compound 3) maintained OX 2 R affinity and reduced OX 1 R activity, but physicochemical properties were improved compared to compound 2. 81 Further SAR studies focused on the modification of all three aromatic substitutions in compound 3. 82 Changing the di-OMe-phenyl substituent to a pyrimidine group resulted in a significant loss of binding affinity, as shown with compound 4, but an improved overall profile due to reduced lipophilicity and enhanced solubility. Then replacing the cyano group to a fluorine regained the binding affinity for both receptors (compound 5), and finally adding a second fluorine to the benzene group significantly improved OX 1 R affinity and led to lemborexant. 82 Clearly, slight structural modifications may cause significant change of compound activity, and SAR studies coupled with optimization of physicochemical properties are useful steps to obtain druggable candidates.
CGRP is a 37-amino acid neuropeptide and its receptor is implicated in migraine. 83 The benzodiazapinone compound 7 was identified as a hit compound with modest CGRP receptor binding affinity (K i = 4.8 µM, Fig. 4). 84 Replacing the right-hand spirohydantoin structure with piperidyldihydroquinazolinone, a privileged structure for CGRP receptor antagonists, 85 an affinity boost of 100-fold was gained. 84 Further optimization of the benzodiazepinone core resulted in the caprolactam compound 9, which showed a K i of 25 nM. 86 Changing the piperidyldihydroquinazolinone moiety to a piperidylazabenzimidazolone led to compound 10, with a binding affinity of 11 nM. 87 Then by changing the N-substituent on the caprolactam and adding difluro substitutions on the lower benzene ring delivered compound MK-0974 (11, K i = 0.77 nM, Fig. 4), which entered clinical trials. Compound 12 (BMS-846372) shares the same piperidylazabenzimidazolone and the lower diflurobenzene substructures with 11 but differs from the latter with a carbamate core structure and a pyridine-fused-cyclopentane in replacement of the caprolactam. 88 Compound 11 displayed high binding affinity while suffered from poor physicochemical properties, such as low solubility. To improve this, a hydroxyl group was attached to the cycloheptane ring and it was discovered that the (S)-isomer 13 was more potent  than the (R)-OH compound 14. The -OH was finally replaced with an -NH 2 group, which led to the clinical compound rimegepant. 89 The latter was further developed for better safety and efficacy profiles and obtained regulatory approval by the FDA in 2020.
The above examples demonstrate that, starting from a modest affinity hit compound, systematic SAR studies could successfully lead to very potent GPCR ligands that qualify as clinical candidates. Slight modifications of chemical structures sometimes cause remarkable changes of binding affinity or potency, which could not always be accurately predicted by conventional methods, such as docking. Therefore, SAR studies will continue to play a critical role in drug discovery.

GPCR STRUCTURE
The structure of GPCRs is a crucial determinant for understanding the molecular mechanisms underlying ligand recognition and receptor activation. It provides a foundation for drug discovery. The first crystal structure of inactive state rhodopsin purified from bovine eyes was solved in 2000. 90 Although tremendous efforts have been made, elucidation of GPCR structures remains challenging due to several bottlenecks, including low receptor expression level, difficulties in extraction, highly flexible conformation, lack of crystal contacts, etc. The first crystal structure of GPCR extracted from exogenously expressed host cells, the human β2-adrenergic receptor (β 2 AR, gene name: ADRB2) bound to an antagonist, was disclosed in 2007, representing a milestone in GPCR structural biology. 7 Several innovative methods, especially the incorporation of a soluble fusion partner and lipidic cubic phase (LCP) crystallization, facilitated subsequent studies. Further technological breakthroughs in protein expression and purification, 91,92 receptor engineering, 8,93 application of Fab fragment and nanobody, 94,95 and GPCR crystallization 96 led to an exponential growth of this field.
The crystal structure of β 2 AR in complex with stimulatory G protein (G s ) solved in 2011 97 and rhodopsin bound to visual arrestin reported in 2015 98 revealed the molecular mechanism of GPCR interaction with G protein and arrestin, respectively. Notably, the wave of resolution revolution in the single-particle cryo-EM has brought a significant impact on the determination of GPCR complexes. 10 Over 90% of GPCR-transducer complex structures were solved using cryo-EM (Table 4). To date, a total of 455 structures from 82 GPCRs belonging to all classes except B2 have been reported (Table 4). Although GPCRs show extensive sequence diversity, they share a conversed structural architecture of a TMD composed of seven helices embedded in the cell membrane. The transmembrane (TM) helices, essential for signal transduction across the cell membrane, are linked by three extracellular loops (ECLs) and three intracellular loops (ICLs). However, distinct structural features exist among members from different classes despite their overall structural similarity.
Various ligands of class A GPCRs bind to similar orthosteric sites directly in the helix bundle. Structural variations in ECLs, TM helices, and side chains show a remarkable variety of sizes, shapes, and physicochemical properties of the ligand-binding pockets, leading to diversified mechanisms of ligand recognition. 99 For example, ligand binding, access, and selectivity are affected by ECL2. 100,101 Many published GPCR structures are in an antagonistbound inactive state, but the number of agonist-bound active state structures have been increased steadily in recent years due to the deployment of cryo-EM. Additionally, the structure of human M2R bound to a PAM (LY2119620) unveiled the allosteric ligand recognition mechanism. 102 A summary of complicated recognition and modulation mechanisms of class A GPCRs bound to agonist, antagonist, and PAM is illustrated in Fig. 5a.
Class A GPCRs are activated through a common pathway, which strings the conserved "micro-switches" together, including CWxP, PIF, Na + pocket, NPxxY, and DRY, thereby linking the ligand-binding pocket to the G protein-coupling region ( Fig. 5b-f). 99,103 The binding of diverse agonists triggers the rotameric switch of W 6.48 , a highly conserved residue in the "CWxP" motif, and the concomitant side chain rotation of F 6.44 (Fig. 5b). Upon stimulation by an agonist, conformational rearrangement occurs in the PIF (P 5.50 , I 3.40 , and F 6.44 , Fig. 5c) and the Na + pocket residues (D 2.50 , S 3.39 , N 7.45 , and N 7.49 , Fig. 5d). These reorganizations rigger the notable outward displacement of TM6, the hallmark of class A GPCR activation (Fig. 5b). The repacking of Na + pocket residues initiates the TM7 movement toward TM3. Upon receptor activation, the "NPxxY" residue Y 7.53 changes its rotamer conformation and points toward TM3, rendering new contact formation between Y 7.53 and residues in TM3 (L 3.43 , I 3.46 , and R 3.50 , Fig. 5e) and subsequently the enhanced packing of TM3-TM7. Finally, "DRY", one of the most conserved motifs in class A receptors, locates at the bottom of the 7TM and forms an intrahelical salt bridge between D/E 3.49 and R 3.50 . R 3.50 forms an additional inter-helical salt bridge with D 6.30 , known as the ionic lock, which connects the intracellular ends of TM3 and TM6 to stabilize receptors in an inactive state (Fig. 5f). These contacts are eliminated after agonist binding, and R 3.50 is released to interact with other residues to facilitate the G protein coupling. It is notable that an acidic residue at position 6.30 is less conserved in 30% of class A receptors. Alternatively, R 3.50 may form polar interactions with other polar residues in TM6 (i.e., T 6.34 in κ-OR and μ-OR) to mediate the activation. Collectively, these rearrangements and reorganizations of conserved motifs are critical to the activation of class A GPCRs.
Class B GPCRs contain a large ECD and a TMD bundle with the peptide ligand recognition by both domains. According to the two-domain-binding model, the C-terminus of the peptide interacts with the ECD and orient the N-terminus of the peptide toward the TMD bundle. It then engages with the TMD core to facilitate receptor activation. 104 The most remarkable structural feature of this class is the swing of ECD, accompanied by the corresponding shift of the peptide C-terminus (Fig. 6a, b). Conversely, the N-terminus inserts into a V-shape cavity within the helix bundle with a similar binding pose. Compared to small molecule-binding pocket of class A, that of class B is more solventaccessible with higher flexibility and larger volume to accommodate sizeable peptidic ligands. 9 In addition, structural studies also reveal an antagonist-binding pocket deep in the TMD bundle of CRF1R 105 and a common binding site for allosteric modulators of GCGR 106 and GLP-1R 107 located outside the TMD bundle between TM6 and TM7 (Fig. 6c).
A comparison of the full-length active receptor structures with that in the inactive state reveals a general activation mechanism for class B GPCRs (Fig. 6d, e). 9,108 The binding of a peptidic ligand causes a conformation rearrangement of the central polar network with simultaneous destabilization of the TM6 helix, thus initiating a sharp kink formation at the conserved motif P 6.47b xxG 6.50b . This central polar network is preserved across the class B receptors solved so far. However, the exact interactions may vary among different members in a ligand-and receptor-specific manner. The rearrangement of TM6 breaks the polar interaction of the conserved HETX motif and the TM2-6-7-helix 8 polar network, thereby inducing a notable outward displacement of TM6 and creating a cytoplasmic cavity to accommodate α5 helix of Gα s protein.
Class C GPCRs are distinguished by a characteristically large ECD that forms an obligate dimer. The ECD is distal to the TMD and contains an orthosteric ligand-binding pocket. It is composed of a ligand-binding VFT linked by the CRD to the TMD except for the metabotropic GABA B receptor (GABA B R), which lacks CRD (Fig.  7a-d). This structural feature results in a potentially unique ligand recognition mechanism. The full-length structures of mGlu5 in apo and agonist-bound states, 51 as well as several recently reported full-length structures of GABA B Rs, 52,53,55 have significantly The structures were updated in September 2020. The PDB codes of GPCR structures determined by cryo-EM are in bold. The structural data were collected from the Protein Data Bank (rcsb.org) 294 ADCYAP1R1 pituitary adenylate cyclase-activating polypeptide type I receptor, ADORA1 (A 1 AR) adenosine A1 receptor, ADRA2B α2B adrenergic receptor, ADRB1 (β 1 AR) β1-adrenergic receptor, AGTR2 (AT2R) angiotensin II receptor type 2, APLNR Apelin receptor, C5AR1 C5a anaphylatoxin chemotactic receptor 1, CALCR (CTR) calcitonin receptor, CCR1-9 C-C chemokine receptor (CCR) type 1-9, CHRM2 (M2R) muscarinic acetylcholine receptor M2, CHRM3 (M3R) muscarinic acetylcholine receptor M3, CHRM4 (M4R) muscarinic acetylcholine receptor M4, CNR2 (CB2) cannabinoid receptor 2, CRHR1 (CRF1R) corticotropin-releasing factor receptor 1, CRHR2 (CRF2R) corticotropin-releasing factor receptor 2, CXCR1-2 C-X-C chemokine receptor type 1-2, CYSLTR1-2 cysteinyl leukotriene receptor 1-2, DRD4 D4 dopamine receptor, EDNRB endothelin receptor type B, F2R proteinase-activated receptor 1, F2RL1 proteinase-activated receptor 2, FFAR1-2 free fatty acid receptor 1-2, GABBR2 (GABA B2 ) GABA type B receptor subunit 2, GHRHR 295 growth hormone-releasing hormone receptor, GLP-2R 296 glucagon-like peptide-2 receptor, GPBAR1 protein-coupled bile acid receptor, GPR52 G protein-coupled receptor 52, HCRTR2 (OX 2 R) orexin receptor type 2, HRH1 histamine H1 receptor, extended our understanding of the activation mechanism of the class C receptors. It is known that an agonist binds and stabilizes the conformation of the VFT, leading to compaction of the intersubunit dimer interface and proximity of the CRD (Fig. 7a, b). This conformation transition, in turn, triggers TMD rearrangement through interaction between ECL2 and CRD. [51][52][53]55 In contrast to mGlu5, the GABA B R undergoes a featured asymmetric activation. After the binding of agonist baclofen to GABA B1 (GB1) subunit, the latter only exhibits a negligible conformational change. Additionally, due to lacking CRD in the GABA B Rs, the relatively shorter stalk and ECL2 region may rigidify their conformations and mediate the transduction of conformational changes from VFT to 7TM. 53 In contrast, substantial conformational alterations occur at the stalk and TM3/4/5-ICL3 regions at the cytoplasmic part of GB2 (Fig. 7c,  d), which predominantly couples to G i1 heterotrimer. Interestingly, cholesterols are observed at the TMD interface of inactive GABA B Rs 52 (Fig. 7c), while two chained phospholipids occupy a binding site overlapped with the orthosteric binding pocket in class A GPCRs 52,53 (Fig. 7c, d). These cholesterols and phospholipids may contribute to the activity regulation of the GABA B R. Noteworthy, in contrast to other allosteric modulators that bind to the TMD core of class C GPCRs (Fig. 7e), [109][110][111] (+)-BHFF occupies a novel allosteric site at the interface of TMDs in GB1 and GB2 subunits 52 (Fig. 7d). This novel allosteric binding site may provide a promising template for the design of PAMs for GABA B Rs. Class F GPCRs include SMO and 10 FZDs in humans. Besides a canonical TMD across all classes of GPCRs, class F is characterized by a large ECD composed of a CRD and an ECD linker domain to connect with TMD (Fig. 8a, b). 112 It was reported that SMO has a unique allosteric modulation mechanism. 113 In fact, two ligandbinding sites have been identified: one in CRD and the other in TMD (Fig. 8b). SMO is activated by cholesterol via binding to CRD. The binding of an antagonist to TMD was proposed to trigger its conformation change thereby propagating to CRD and allosterically impeding the binding of cholesterol. 113 Recent structural studies reveal that cholesterol and oxysterol that are critical for SMO activation are located deep within the 7TM domain of SMO (Fig. 8b, c). 114,115 CRD of FZD can interact with lipoglycoprotein Wnt and Norrin (specific ligand for FZD4) to mediate the Wnt signaling. 61,116 Structures of CRD in complex with Wnt or Norrin provided molecular details of how they formed a symmetrical homodimer (2:2 complex) during ligand recognition (Fig. 8d,  e). 117,118 In contrast to SMO, the ligand recognition and receptor activation mechanisms of FZD remains elusive due to the absence of the full-length FZD structures. So far, only two apo TMD structures of FZD4 and FZD5 have been reported (Fig. 8f). 61,63 Structures of the full-length FZD in a ligand-bound state are required awaiting to provide mechanistic explanations.

GPCR PHARMACOLOGY
The explosion of 3D GPCR structures and computational simulations has revealed the dynamic conformations between inactive, intermediate, and active states of GPCRs. The detailed structural information illustrated that cholesterol, ion, lipids, and water also participate in receptor activation. 99,119 The flexibility of receptorbinding pocket endows the complex pharmacological mechanisms of ligand recognition and signal transduction. Biased signaling, allosteric modulation, and polypharmacology are helping us better understand how GPCRs bind to numerous ligands and how they transmit diverse signals to elicit physiological functions.

Polypharmacology
Ligand binding to multiple targets leads to antagonism, additive, or synergism pharmacological responses that could be positive or negative based on the mechanism of action. The paradigm of one drug vs. multiple targets has outpaced the time and cost associated with the conventional therapy. 120 Polypharmacology thus emerges to study acceptable degree of specificity toward multiple targets, interconnected signaling pathways that result in clinical benefit or cross-reactivity that may cause adverse events. 121,122 T2DM, obesity, cancer, and Alzheimer's disease are major indications for GPCR modulators. 4 These polygenic diseases The common activation pathway of class A GPCRs as exampled by the structures of inactive (gray, PDB code 3NYA) and active β 2 AR (green, PDB code 3SN6). The conformational changes of conserved "microswitches", including CWxP (b), PIF (c), Na + pocket (d), NPxxY (e), and DRY (f), are highlighted. Side chains of residues in "micro-switches" are shown as sticks. Red arrows indicate the shift and swing directions of elements in the active β 2 AR structure relative to the inactive one are not completely treatable by a single agent, while desirable efficacies may be achieved for certain respiratory conditions, central nervous system (CNS) disorders, and cardiovascular diseases through modulators directed against β 2 AR, DRD2, and AGTR1, 123 respectively.
It was shown that 5-hydroxytryptamine receptor 2 (5-HT 2 ) binds to selective inverse (ritanserin) and highly promiscuous (ergotamine) agonists but the interaction with ergotamine is broad. 121 This feature allows the development of pan serotonin receptor modulators to treat different diseases. [124][125][126][127] For instance, zolmitriptan as an anti-migraine drug is also used for hyperesthesia via binding to off-target site, 120 and lorcaserin (Belviq) is used to treat obesity while its therapeutic potential for depression, schizophrenia, and drug addiction is being investigated. 128,129  The common activation mechanism of class B GPCRs as exampled by the structures of inactive GCGR (gray, PDB code 3NYA) and active VIP1R (green, PDB code 6VN7). Side chains of residues in three conserved polar network are shown in stick presentation. The conserved P 6.47b xxG 6.50b motifs in TM6 are shown as single red spheres However, off-target activity, hallucinations, 130 and cardiac valvulopathy related to 5-HT 2A and 5-HT 2B modulation 129 should be carefully monitored. Atypical antipsychotics are mainly targeting both dopamine and serotonin receptors, usually as antagonist for DRD2 and antagonist or inverse agonist for 5-HT 2A. 131 Exemplified by clozapine 120 and aripiprazole, 132 haloperidol, amoxapine, and asenapine 4 display a diverse spectrum of receptor interaction. Additionally, carazolol, a member of aminergic division exerts its effects by interacting with multiple adrenergic receptors as inverse agonist or allosteric antagonist. 19,129 Istradefylline combined with L-DOPA/dopamine simultaneously target A 2A R, DRD1 and DRD2 in animal model of Parkinson's disease. 131 Amitryptyline, a tricyclic compound targeting muscarinic and histamine H1 receptors, 133 is used to treat depression and non-selective muscarinic receptor antagonists are trialed for bladder dysfunction. 4 Lorazepam, indicated for anxiety due to interaction with GABA A R, is also an allosteric modulator of the proton-sensing GPCR (GPR68) 134 and has been repurposed to treat pancreatic cancer. 2 6'-Guanidinonaltrindole (6'-GNTI) is an agonist with higher selectivity for δ/κ-opioid receptor heterodimer but not homodimer. Importantly, 6'-GNTI is an analgesic that offers additional benefit. In cardiovascular diseases, β blockers decrease catecholamine-induced heart rate elevation via interaction with valsartan (AT1R-mediated signaling). 135 It is of note that mono-, dual-, and tri-agonists for the glucagon family of receptors (GLP-1R, GCGR, and GIPR) have been developed and trialed for weight loss and glucose control (Table 3). Successful outcome will determine whether unimolecular polypharmacology is a practical approach to translate safety and efficacy of multiple agents into a single molecule. 136 Biased agonism Activated GPCRs can recruit multiple transducers (such as heterotrimeric G proteins, GPCR kinases, and β-arrestin) and consequently produce distinct biological responses. Ligands that preferentially engage one signaling pathway over others are regarded as bias and may show improved therapeutic outcomes. 137,138 Biased signaling that has been applied to drug discovery involve AT2R, µ-OR, κ-OR, β-adrenergic receptors, DRD2, CTR, CCR, and adenosine receptors. µ-OR is the best studied receptor for biased agonism. 137 Compounds that stimulate Gα i coupling and cAMP production but not β-arrestin recruitment are preferable to retain analgesia and reduce opioid-related side effects. 139 This G protein bias was also demonstrated with widely used drug tramadol, 140 whose active metabolite, desmetramadol, elicited maximum cAMP production without affecting β-arrestin 2 recruitment compared to fentanyl and morphine. Safety profile is improved with less adverse effect such as respiratory depression. 140 Another µ-OR-biased ligand, oliceridine (TRV130, Olinvo TM ), passed phase III clinical trial but did not get the FDA approval for safety concerns. 141 The NDA for oliceridine was resubmitted and a new counterpart, TRV734, is not only suitable for oral administration but also safer due to reduced dependency. 142 A fourth µ-OR-biased ligand, PZM21, cross-reacts with κ-OR and failed to reduce respiratory depression in C57BL and CD-1 mice. 143 Whether this relates to its residual but marked effect on β-arrestin 2 recruitment, as opposed to oliceridine whose action is negligible, 144 remains to be further studied.
Similar situation occurred with κ-OR as well whose agonists possess analgesic property and have a low risk of dependence and abuse but with adverse effects such as sedation, motor dysfunction, hallucination, and dysphoria. 145 G protein-biased agonists of κ-OR, 146 including RB-64, 145 mesyl salvinorin B, triazole 1.1, diphenethylamines and LOR17, 141 were reported to minimize the adverse effects in preclinical settings. One of such, nalfurafine, was approved in Japan (2015) as an anti-pruritic agent for patients with chronic liver diseases. 147 Carvedilol, known as a β1 and β2 adrenoceptor blocker, was found to be biased toward β-arrestin recruitment, G proteincoupled receptor kinase activation, and ERK1/2 phosphorylation. Joining its rank included alprenolol, bucindolol, and nebivolol, all are used to treat hypertension and congestive heart failure. 148 In the case of β3 adrenoceptor, CL316243 is cAMP-biased, whereas L748337 and SR59230 are ERK/p38 phosphorylation-biased. 149,150 Interestingly, CL316243 was also tested for treatment of obese mice. 151,152 However, none of them have advanced to the clinic.
In contrast to µ-OR, arrestin bias is desirable for AT1R to improve cardiac performance. 153 Nonetheless, clinical development of AT1R modulators either resulted in a phase IIb trial failure (TRV027) in 2017 154 or never reached to clinical stage (SBpa, SVdF, SI, sarmesin, saralasin, and SII). 155 Of note is that biased molecules may show species preference. For instance, CL316243 is more active in mice than in humans, 156 whereas nalfurafine works better in humans vs. rodents. 157 A list of therapeutic agents with biased signaling approved or advanced to clinical trials is shown in Table 5.

Allosteric modulation
In recent years, studies on allosteric GPCR modulators have gained unprecedented momentum. [158][159][160][161] An allosteric modulator is a ligand binding to a position other than the orthosteric site but can modify responses of a receptor to stimulus. Allosteric modulators that enhance agonist-mediated response are called PAMs, while those attenuate the response are called NAMs. This phenomenon is very common such that the Allosteric Database 2019 (ASD, http://mdl.shsmu.edu.cn/ASD) 162 records 37520 allosteric modulations on 118 GPCR members, covering all four classes.
Allosteric modulation is advantageous in terms of (i) using highly druggable pockets. In some cases, it is easier to design ligands at an allosteric site than the orthosteric site, such as class B GPCRs with orthosteric pockets wide open. For example, both PAM 163 and NAMs 107 binding to the same position at the TMD of GLP-1R were reported; (ii) improving selectivity. The orthosteric site and cognate ligand are often highly conserved, making it hard to discover very selective orthosteric binders. Meanwhile, nonconserved allosteric sites would be a better choice evidenced by discovery of many subtype selective allosteric modulators of acetylcholine 102,164 and cannabinoid receptors 165,166 ; (iii) introducing signal bias. Allosteric modulators with biased signaling were developed for prostaglandin F2α receptor 167 and chemokine receptor CXCR4. 168 Albeit still as an emerging concept, allosteric modulators have exhibited a great potential with some compounds being marketed or in clinical trials. 160 However, developing allosteric modulators of GPCRs remains challenging-molecules recorded in the ASD largely concentrate on two subfamilies, the mGluRs (8 members, 17,115 modulations), and mAChRs (5 members, 7666 modulations), accounting for nearly 2/3 of the total number. Some individual receptors also contribute a significant proportion, such as CB1 (1948 modulations), GABA B (1286 modulations), and follicle-stimulating hormone receptor (1233 modulations). Excluding these "easy cases," allosteric modulators are few in number. Furthermore, the structural diversity of the allosteric modulators is quite low, for many derivatives would be included soon after a parent compound is identified. The difficulty in developing allosteric modulators is partly due to the limitation of detecting allosteric behavior: Not every newly discovered active compound could be tested for its effect on binding affinity or EC 50 of an orthosteric agonist, therefore some allosteric modulators were not correctly identified. For instance, BPTU in P2RY1, the first GPCR NAM solved in complex structure (PDB code: 4XNV), 169 was not considered allosteric until the structure was obtained. To make things worse, NAMs may weaken the binding of an endogenous ligand thus behaving like a competitor, such as NDT9513727 in C5AR1 170 (PDB code: 5O9H). 171 The most effective way to identify the binding site of an allosteric modulator on a GPCR is solving the complex structure. Crystallography is an effective technique, while rapidly deployment of cryo-EM has started to deliver its promise (PDB codes: 6OIK 172 and 6U1N 173 ). To date, 17 GPCRs have reported structures in complex with allosteric modulators. Detailed analysis of complex structures before October 2018 was reported previously, 161 and here we focus on insights provided by newly published results. The most unusual allosteric-binding sites on GPCRs are at the lipidic interface embedded in cell membrane. Five different positions were identified by crystal structures (Fig. 9): UP12, UP34, LOW34, LOW345, and LOW67. Four of them were recently reviewed. 161 The LOW34 site was reported in 2019 for ORG27569 in CB1 (PDB code: 6KQI 166 ; Fig. 10a).
ORG27569 attracted much attention for its distinctive function: increasing the binding of orthosteric agonist CP55940 but making it act as inverse agonist. 165 Many attempts were made to locate the binding site of ORG27569 by mutagenesis but the results are conflicting: one study showed that the effect of ORG27569 on CP55940-induced [ 35 S]GTPγS binding was disturbed by mutations to multiple residues at the orthosteric site, leading to a hypothesis that ORG27569 stays in the same pocket close to CP55940. 174 Another study found that ORG27569 reduced the binding of a fluorescence-labeled orthosteric antagonist, and the effect was only disturbed by mutations at the lipidic interface close to the cytoplasmic end of CB1. 175 Besides, it was reported that the functions of ORG27569 were also affected by breaking a disulfide bond at the N-terminus 176 or by constitutive active/inactive mutations at the cytoplasmic interface. 174,175,177,178 The crystal structure exhibited that the position of ORG27569 is considerably overlapped with a cholesterol captured in another intermediate state (PDB code: 5XRA, 179 Fig. 10a), consistent with the site located by the fluorescence-labeled orthosteric antagonist. 175 At this site, the higher selectivity to CB1 over CB2 could be explained. Interestingly, ORG27569 is the only allosteric modulator at lipidic interface forming no hydrogen bond to the receptor.
There have been three more complex structures of allosteric modulators at lipidic interface since October 2018, all obtained by crystallography. Two are β 2 AR, with a NAM AS408 (PDB code: 6OBA 180 ) or a PAM Cmpd-6FA (PDB code: 6N48 181 ). Both allosteric modulators bind to the LOW345 site (Fig. 10b). The NAM stays at a position very similar to NAMs in C5AR1 (PDB codes: 5O9H, 171 6C1R, and 6C1Q 182 ) but the PAM is close to ICL2 and only partially overlaps with PAMs of FFAR1 (PDB codes: 5TZY 183 and 5KW2 184 ), showing a complex regulation nature at this site. The other complex structure is full-length GLP-1R with PF-06372222 (PDB code: 6LN2 185 ), a NAM previously used to co-crystallized with GLP-1R TMD (PDB code: 5VEW 107 ).
Even around the position of orthosteric ligands (among the helices and facing extracellular side), another ligand may occupy the space not taken by the endogenous ligand and act as an allosteric modulator. The very abundant PAMs/NAMs of mAChRs function in this mechanism. PAM LY2119620 in M2R (with the orthosteric agonist iperoxo and stabilized by a nanobody) was the first allosteric modulator to obtain complex structure with a class A GPCR (PDB code: 6MQT 102 ). Recently, LY2119620 was also observed in protein complexes of M2R with G protein (PDB code: 6OIK 172 ) or arrestin (PDB code: 6U1N 173 ) by cryo-EM.
CCR5 is a chemokine receptor and an important anti-HIV drug target. A marketed inhibitor, maraviroc, has long been recognized as a NAM of CCR5. There were hypotheses that small molecule NAMs, chemokine, and the HIV-binding protein have separate binding sites. 186 However, structures of CCR5 in complex with maraviroc (PDB code: 4MBS 187 ), chemokine analog antagonist (PDB code: 5UIW 188 ), or HIV envelope glycoprotein (PDB code: 6MEO 189 ) show that these ligands highly overlap in CCR5 pocket (Fig. 10c). Therefore, the noncompetitive behavior of maraviroc may be due to a very extensive interface of peptidic CCR5 agonist, thus a small molecule cannot diminish the binding even with this much collision. The results illustrate that allosteric behavior is not equal to totally separated binding positions, because partially overlapped sites with different key interactions are also allowed. The last case of allosteric modulator in extracellular pocket is PAM TT-OAD2 of GLP-1R (PDB code: 6ORV 190 ). This small molecule agonist only slightly collides with the endogenous peptide (PDB code: 5VAI 191 , Fig. 10d), consistent with its behavior that only partially displaces an orthosteric probe. 190 The cytoplasmic interface, where a GPCR interacts with intracellular partners, including Gα and β-arrestin, contains pockets suitable for drug design. So far, four small molecules have been validated by crystallography to bind at this position. The targets are three chemokine receptors (CCR2, PDB code: 5T1A 192 ; CCR7, PDB code: 6QZH 193 ; and CCR9, PDB code: 5LWE 194 ) and β 2 AR (PDB code: 5X7D 195 ). These ligands are all NAMs and  proximately share the same binding site (TM1, TM2, TM6, TM7, ICL1, and H8). Their binding position does not overlap with Gα, therefore they may stabilize the inactive state by blocking conformational changes required for receptor activation. This site is generally non-conserved in the GPCR superfamily, thus targeting here may provide some selectivity. Additionally, many nanobodies at the cytoplasmic interface were also developed for several receptors, including AGTR1 (PDB codes: 6DO1 196 and 6OS0 197 ), β 1 AR (PDB code: 6IBL 198 ), β 2 AR (PDB code: 6N48 181 ), and SMO (PDB code: 6O3C 114 ; for information before October 2018, see review 161 ).
Multi-domain regulation is an interesting topic in allosteric modulator discovery. Class C GPCRs use ECDs to recognize their cognate ligands, leaving the classic pocket of TMD for allosteric modulating. 161,199 This is the major reason why this class has a large number of allosteric modulators. In the case of mGluRs, both PAMs and NAMs have been widely reported, but only NAMs obtained complex structures-there is no solved active state structure. The full-length structures of mGlu5 (PDB codes: 6N51 and 6N52 51 ) displayed how the binding of orthosteric agonist to ECD triggers the change of interaction between two monomers, but the conformational change of TMD remains elusive.
SMO in class F is also a multi-domain receptor. The first reported ligand of SMO cyclopamine (an antagonist causing birth defects) binds to the classic TMD pocket (PDB code: 4O9R 200 ) shared by several other antagonists with different chemical scaffolds and an agonist (SAG). 62,113,201,202 ALLO-1, an antagonist identified as allosteric modulator not competitive to cyclopamine, was recently found to bind at a deeper position in the pocket by photo-affinity labeling combined with mass spectrometry (MS). 203 SMO has another pocket in ECD that interacts with steroids, including cholesterol (PDB codes: 5L7D 113 and 6D35 204 ) and 20(S)-hydroxycholesterol (PDB code: 5KZV 119 ). Since cholesterol has been the most favored candidate of SMO endogenous ligand, the ECD pocket is treated as orthosteric making the TMD pocket allosteric. However, newly obtained structures demonstrated that cholesterol or its analog can also bind to TMD pocket (PDB codes: 6O3C 114 and 6OT0 115 ), leaving the question open for which is the true orthosteric site.

Disease indication
GPCRs are involved in many human diseases and specific drug intervention is one of the most celebrating achievements in the pharmaceutical industry (Table S3 and Fig. S1). Among all available drugs targeting GPCRs, HRH1, DRD2, M1R, and ADRA1A are the most frequently addressed for indications such as hypertension, allergy, pain, and schizophrenia, and 33% of them have >1 indication with an overall average of 1.5. Although CNS diseases are still popular accounting for 26% of all approved indications, development focuses have now been shifted to T2DM, obesity, multiple sclerosis, smoking cessation, short bowel syndrome, and hypocalcemia. Repurposing of existing drugs for new indications also emerged to supplement discovery efforts.

STRUCTURE-BASED DRUG DESIGN
As two general types of computer-aided drug design techniques (Table 6), 205 SBDD and ligand-based drug design, exploit the structural information of protein targets and the knowledge of known ligands, respectively (Table 6). SBDD, on the basis of crystal/cryo-EM/NMR structures or homology models, first identifies key sites and important interactions responsible for target functions, then screens large virtual library/designed agents that disrupt or enhance such interactions to modulate relevant biological processes and/or signaling pathways by molecular docking, and finally discovers active leads with desired pharmacological properties.
Clearly, the past decade is a golden age for SBDD on GPCR. With the year of 2011 (when LPC crystallization, 206 fusion proteins, 8 and other key techniques collaborated to launch the outbreak of GPCR structure determination including the landmark β 2 AR-G s ) for watershed, 207 SBDD of GPCR evolves two distinct stages: rhodopsin-based homology model and truly authentic structure of individual receptors. Boosted by the fast-increasing number of high-quality GPCR structures, improved accuracy of combinational computational approaches, and better understanding of activation mechanism and pharmacology, SBDD is developing rapidly with fruitful scientific reports and increasing GPCR-targeted drugs contributed by this approach. Considering the length of time required for a drug to be available on the market (10-15 years) and the chance of applying structural biology information to hit discovery and lead optimization in the first 2-3 years of a drug discovery program, it is probably too early to see the approval of GPCR-targeted drugs being developed with the aid of a structure, and such situation is likely to change as the tremendous efforts from both academia and industry start to bear the fruits of successes. The following is a brief account of recent advances in three main aspects of SBDD (chemical space, receptor dynamics, and pose evaluation) in the context of their application in GPCR pharmacology.
Optimized virtual library Despite the vast chemical space (>10 63 drug-like molecules), only a nominal fraction has been explored by SBDD, where both the compound availability and insufficient diversity limited the number of screened ligands. To overcome these problems, ultralarge 208-210 and focused libraries [211][212][213] were employed. Lyu and colleagues 208 presented an excellent model of "bigger is better" in virtual drug screening. Based on the 130 well-characterized reactions, they generated 170 million make-on-demand compounds (http://zinc15.docking.org/), the resulting library is remarkably diverse with >10.7 million scaffolds unavailable before. By docking 138 million molecules against DRD4, they discovered 81 new chemotypes (24% hit rate), 30 of them showed submicromolar activity, including a 180-pM subtype-selective and G i -biased DRD4 agonist. This ultra-large library docking study provides important information: (i) hit rate fell almost monotonically with docking score; (ii) hit rate vs. score curve of DRD4 predicted that 1 from every 873 compounds may have a minimum affinity of 1 μM; and (iii) human visual evaluation improved the selected compound with higher affinities, efficacies, and potencies but not the hit rate. A follow-up study on MT 1 by docking >150 million "leadlike" molecules 209 identified 15 active leads (39% hit rate) with potencies ranging from 470 pM to 6 μM. Alternatively, focused library [210][211][212][213][214][215] including scaffold library, natural products, dark chemical matter (i.e., chemicals that have never shown bioactivity tested in over 100 assays), and fragment-and lead-like libraries were introduced in virtual screening (VS) for dozens of receptors. Focused on compound library of traditional Chinese medicine (TCM), Liu et al. found that salvianolic acids A and C antagonized the activity of both P2RY1 and P2RY12 purinoceptors in the low µM range, while salvianolic acid B antagonized the P2RY12 purinoceptor. Remarkably, these three salvianolic acids are major active components of the broadly used hemorheologic TCM Danshen (Salvia militorrhiza). Taking NTSR1 as an example, Ranganathan et al. found that the fragment library tended to have higher hit rate than that of the lead-like library (19%) but the affinities were ∼100-fold weaker. Collectively, these results demonstrate the importance and advantages of ultra-large and tailored libraries in discovering potent GPCR modulators.

Receptor dynamics
Emerging evidence from crystallography, spectroscopy, and molecular dynamics (MD) simulations have demonstrated the crucial roles of GPCR dynamics involved in ligand recognition, receptor activation, and allosteric modulation. [216][217][218] To consider the protein flexibility during GPCR-related SBDD, many computational approaches 217 221 For 5-HT 1A whose crystal structure is not available, Warszycki et al. applied MC and NMA to generate an ensemble of binding pockets with the input of a homology template and known active compounds and finally discovered two new active ligands through VS. 222 Pose evaluation Correctly selecting and ranking poses of docked compounds in the ligand-binding pockets have been a challenge for SBDD, especially for GPCR that is embedded in the cell membrane with significant conformational adaptability. To address this problem, many physics-based scoring functions [223][224][225][226] integrated with some user-friendly computer programs (e.g., Dock, GOLD, AutoDock, Glide, and rDock) were routinely adopted in SBDD. Recently, precise computational approaches including free energy calculation methods like molecular mechanics/Poisson-Boltzmann surface area (MM/PB(GB)SA), 227,228 free energy perturbation (FEP), 229 quantum mechanical/MM calculations, 230 and fragment molecular orbital 231,232 have been employed with improved performance. Compared to the empirical scoring functions, MM/PB(GB)SA and FEP are physically more rigorous free-energy calculation methods with an increased computational cost and have been adopted in the studies of DNA-ligand, protein-ligand, and protein-protein interactions. 233,234 By introducing of the minimization-based MM/ GBSA refining and rescoring of docked poses, Zhou et al. identified seven 5-HT 2B antagonists with novel chemical scaffolds and the most potent one has an IC 50 of 27.3 nM in a cellular assay. 227 Lenselink et al. used FEP to design A 2A R antagonists and identified a highly potent molecule with K i of 1.2 nM. 232 However, computational investigation across 20 class A crystal structures and 934 known ligands demonstrated that the correlations between predicted binding free energy by MM/PBSA and experimental data varied significantly. The observed variations exist between individual receptors and are highly system specific, 235 indicating that successful application of MM/PBSA may require additional efforts in validation of experimental data and optimization of simulation/calculation parameters. Alternatively, protein-ligand interaction fingerprints exacted from available crystal structures 225,236 fueled docking score with protein-ligand-binding mode information and resulted in improved VS virtual hit rates for β 2 AR (53%) and HRH1 (73%) with up to nM affinities and potencies. 225,236 Collectively, innovation in VS and pose evaluation, along with evolution of computational hardware, has significantly advanced SBDD and is expected to lift the discovery efficiency to a new height, since GPCRs have multiple downstream signaling pathways responsible for distinct functions or consequences. The high degree of sequence and pocket similarities between different subtypes demands for novel ligands with superior specificity and selectivity. In this regard, allosteric and biased modulators may offer additional pharmacological benefits.
Subtype selectivity It is known that GPCR subtypes share high sequence identities in orthosteric sites with distinct distribution and downstream signaling profiles. Cross-reactivity among subtypes could cause undesired side effects. For example, five MR subtypes display different G protein coupling features (G q/11 : M1R, M3R and M5R; G i/o : M2R and M4R) and organ distribution (CNS, M1R; peripheral tissues such as heart and colon, M2R), while their sequence identify (64-82%) and similarity (82-92%) in TMD are quite conserved. 172 Similar observations were seen among dopamine receptors (DRD1 to DRD5), histamine receptors (HRH1 to HRH4), and adenosine receptors (A 1 AR, A 2A R, A 2B R, and A 3 AR). Guided by structural information, rational design of subtype selective compounds progresses steadily. Using an extended DRD2specific binding pocket from the haloperidol-bound DRD2 crystal structure, Fan et al. discovered two highly selective DRD2 agonists (O4SE6 and O8LE6) that specifically activate DRD2 (EC 50 = 1 µM) after screening of 320 non-olfactory GPCRs. 237 Through VS of 3.1 million molecules against M2 and M3, Kruse et al. identified a partial M3 agonist without measurable M2 agonism, capable of stimulating insulin release from a mouse β-cell line. 238   Virtual screening of ultralarge libraries [224][225][226] Novel scaffolds and chemotypes, higher chances of finding potent ligands Substantial computational resources, compound synthesis Virtual screening of focused libraries [227][228][229] Less demand for computational resources, compound easy to purchase, specific scaffolds or origins Reduced diversity, less novelty Ensemble docking [235][236][237] Considering protein flexibility, improved enrichment factors, rescue of false-negative ligands Increased computational burden, pose evaluation, false positive Energy-based pose evaluation [248][249][250][251][252] Improved scoring and ranking ability Substantial computational burden, method validation target dependency G protein-coupled receptors: structure-and function-based drug discovery Yang et al.

Biased signaling
Recently, Suomivuori et al. performed extensive MD simulations to identify two major signaling conformations that couple effectively to arrestin or G protein, respectively. They then designed ligands via minor chemical modification resulting in strong arrestin-biased or G q -biased signal transduction. 240 Meanwhile, McCorvy and colleagues discovered that specific ECL2-ligand contacts are associated with β-arrestin recruitment, whereas blockage of TM5 interaction reduces the G i/o signaling. An arrestin-biased DRD2 modulator was thus made exhibiting a calculated bias factor of 20 relative to quinpirole. 241 In addition, Mannel et al. conducted a VS of a tailored virtual library bearing 2,3-dichlorophenylpiperazine for DRD2 and found that 18 compounds occupy both orthosteric and allosteric sites, and 4 of them stimulated β-arrestin recruitment (EC 50 = 320 nM, E max = 16%) without detectable G protein signaling. 214 Allosterism In the past 5 years, an increasing number of receptor-allosteric modulator complex structures revealed diversified positions of allosteric sites and a variety of binding modes, thereby deepening our understanding of allosteric modulation in terms of underlying mechanisms and structural basis. Further to conventional orthosteric ligands, allosteric modulators affect receptor function in different ways. While PAM may enhance maximal efficacy, NAM could reduce agonist signaling strength. 159,160,242 It was reported that a PAM of M2R is located above the orthosteric site and interacts with ECLs. Korczynska et al. screened 4.6 million molecules against the allosteric sites of M2R and identified a PAM that potentiated the action of antagonist N-methyl scopolamine (NMS). Subsequent optimization led to a subtype-selective compound 628 that increased NMS binding with a co-operativity factor of 5.5 and a K B of 1.1 µM. 16 Alternatively, Lückmann et al. carried out MD simulations of agonist-removed FFAR1 and found that closure of a potential allosteric site is associated with agonist bindingcompounds that bind to this site to prevent the closure functions as allosteric agonists. 243 Obviously, aided by >400 structures from 82 receptors, SBDD is now entering into a new era with substantial knowledge of GPCR signaling 103,110,244 and drug candidate attributes. 245 NOVEL SCREENING TECHNOLOGY As GPCRs represent the most prominent family of therapeutic targets, 132 innumerable efforts have been made in both industry and academia to screen for novel ligands that can modulate the activity of a specified GPCR and serve as lead compounds for drug development.
A diverse array of experimental technologies suitable for assaying protein-ligand interactions have been directly applied or tailored to GPCR-targeted ligand screening, and they can be classified into three main categories: binding-based, stabilitybased, and cell signaling-based assays (Table 7). Binding-based assays monitor the physical interactions between a GPCR protein typically in a purified recombinant form with individual test compounds. Cell signaling-based assays measure downstream effectors (e.g., cAMP, Ca 2+ , IP1/IP3) of specific intracellular signaling pathways known to be mediated by GPCR, which reflect the functional outcome of ligand binding to the receptor. Stabilitybased assays assess the variation of thermal stability for a purified protein when treated by test compounds. These different techniques vary in the ligand screening throughput and binding characteristics (Table 7). In the lead discovery stage, both bindingand signaling/activity-based assays are implemented in a parallel or sequential manner, as the multipronged use of complementary techniques would reduce the overall false-positive and falsenegative rates. 246 Here we summarize about 20 experimental screening technologies adapted to GPCR ligand discovery (Table 7) and highlight the most recent development of binding-based approaches. Notably, an update of assays assessing GPCR activation and signaling has been provided in a previous review 247 and will not be elaborated here. The structure-based VS is covered in the above section. Structural elucidation technologies (e.g., X-ray crystallography, single-particle cryo-EM, NMR, and HDX-MS) not suitable for high-throughput screening (HTS) are also excluded.
DNA-encoded library (DEL) Impressive technological advances have been made for bindingbased ligand screening over the past decade. Specifically, DEL has emerged as a powerful approach to drug discovery. [248][249][250][251] Created by split and pool synthesis, DEL usually contains hundreds of thousands to billions of distinct small molecule-DNA conjugates. A majority of DEL-based HTS reported to date involve incubation of an immobilized target protein with the library before the protein-ligand complexes are isolated. Encoding DNA tags associated with the immobilized target are then amplified and sequenced to assign relevant chemical structures. 250,251 Although DEL was predominantly applied to ligand screening against soluble proteins such as enzymes, successful adaptation of this technique to GPCRs was reported in a few cases. [252][253][254][255] Lefkowitz group reported the discovery of a NAM for β 2 AR by screening a DEL of 190 million. 252 This NAM not only has a unique chemotype but also exhibits low μM affinity and inhibits cAMP production as well as β-arrestin recruitment. Later on, the same team discovered the first small molecule PAM for β 2 AR through HTS of >500 million DEL compounds. 253 Both NAM and PAM demonstrated high selectivity. Of note is that the NAM was found using unliganded β 2 AR, whereas the PAM was unmasked via intentional application of β 2 AR with its orthosteric site occupied by an agonist thereby shifting the receptor to the active state. 252,253 These two studies elegantly demonstrated a proof-of-concept strategy for binding-based screening of allosteric modulators targeting different conformational states. 253 The power of DEL in the discovery of allosteric GPCR modulators was further demonstrated for PAR2. 254 Screening a billion-size library with a thermostabilized PAR2 mutant resulted in the identification of several agonists and antagonists, and some of them bind to an allosteric pocket in the TMD of PAR2. A similar approach was used to discover tachykinin receptor neurokinin-3 (NK3) antagonists of low nM potency involving NK3 overexpressing cells and a library containing tens of millions DNA-encoded compounds. 255 Clearly, DEL-based ligand screening against GPCRs and other integral membrane proteins offers great promises as it circumvents difficulties in receptor purification.
Affinity selection MS Due to the high sensitivity and high selectivity of modern MS for both protein and small molecule analysis, versatile MS-based technologies have been developed in the past two decades for screening ligands of a given protein target or characterization of ligand-binding properties (Table 7). Almost all of them were originally developed for measuring ligand interactions with soluble proteins, [256][257][258][259][260][261] and recently they have been adapted to more challenging GPCR drug discovery. The majority of MS-based technologies (e.g., automated ligand identification system (ALIS), ultrafiltration-liquid chromatography/MS, frontal affinity chromatography-MS, membrane-based affinity MS, and competitive MS binding) employ a methodology very similar to DEL as they all capture and detect ligands physically associated with a given GPCR, [262][263][264][265][266][267] except that native MS analyzes the entire ligandbound receptor complexes. 268,269 In general, these methods have several advantages over ligand binding or cell signaling assays: (i) unbiased and direct detection of ligand-receptor binding facilitates the identification of both orthosteric and allosteric modulators; (ii) confirmation of ligand identity with accurate mass measurement; (iii) no chemical labeling or DNA encoding of test compounds; and (iv) quantitative MS analysis enables ranking of ligand affinity or evaluation of binding characteristics.
ALIS is currently the most prevailing MS-based technique employed in pharmaceutical companies for HTS of large-scale synthetic compound libraries. 256,261,[270][271][272][273] This system integrates size exclusion chromatography for isolating protein-ligand complexes and reverse-phase chromatography for dissociating bound ligands, which are then identified by high-resolution MS. Not surprisingly, the application of ALIS to ligand screening for GPCRs substantially lagged behind soluble proteins due to difficulties in obtaining membrane receptors of sufficient purity and stability. The earliest application was ligand screening for M2R, 274 in which purified M2R was incubated with a 1500-compound pool in each round of affinity selection. After screening a total of 350,000 compounds, one orthosteric antagonist and one allosteric modulator were identified for AChR. Later on, a similar strategy was implemented to screen ligands for CXCR4 using two libraries comprised of 48,000 and 2.75 million compounds, respectively. Each reaction consumed 250 ng purified receptor incubated with a pool of 100 or 2000 compounds. Out of the 362 primary hits, 34 were subsequently confirmed to be new antagonists. 262 Membrane-based affinity MS developed by Shui's group enables ligand screening toward wild-type active GPCRs embedded in the cell membrane. 266 It features isolation of membrane fractions from cells expressing a GPCR at high yield and incubation of the cell membrane with a compound cocktail, thus keeping the receptor in its native conformation and eliminating the need of protein purification. Compounds associated with the receptor were then released and subjected to high-resolution MS for structural DEL DNA-encoded library, SPR surface plasmon resonance, MST microscale thermophoresis, ALIS automated ligand identification system, UF ultrafiltration, FAC frontal affinity chromatography, DSF differential scanning fluorimetry, DLS dynamin light scattering assignment (Fig. 11a). Each incubation consumed about 2 µg membrane-embedded GPCR protein with a pool of 480 compounds. Primary hits were selected based on the binding index (BI) derived from quantitative MS signals used to distinguish putative ligands from non-specific binders (Fig. 11b). Screening a small compound library with this approach led to the discovery of an antagonist for the 5-HT 2C receptor and four PAMs for GLP-1R that are not reported previously 266 (Fig. 11c). More recently, the same team devised another affinity MS strategy that enabled screening of 20,000 compounds in one pool. 275 Specifically, they modified the workflow by performing iterative rounds of affinity selection for compounds associated Fig. 11 Developing affinity MS approaches for GPCR ligand screening. a Experimental workflow of membrane-based affinity MS. b Membranebased affinity MS screening of 4333 compounds split into 9 cocktails against GLP-1R. Initial hits are indicated by red dots, while gray dots represent negatives. c Binding of one new ligand to 5-HT 2C (upper) and four new ligands to GLP-1R (lower) were validated by a radioligandbinding assay. d Experimental workflow of affinity MS-based screening of natural herb extracts. e Initial hits from screening fractionated herbal extracts toward 5-HT 2C . Aporphines are annotated with larger pink dots. BI binding index. f Structural validation of 1857 by MSMS analysis. g 1857 displayed selective agonism at 5-HT 2C . Source: adapted from Qin et al. 266 and Zhang et al. 282 with A 2A R. Similar to the previously described single-round affinity MS screening assay, quantitative measurement of BI renders detection of high-affinity ligands in this experiment. By comparing the selection of 16 benchmark A 2A R ligands from screening compound pools of 480-mix, 2400-mix, 4800-mix, and 20K-mix, they demonstrated that this accelerated affinity MS screening approach, using either the purified receptor or receptorexpressing cell membranes, allowed detection of most highaffinity A 2A R ligands (K d or K i <5 µM), and significant reduction of protein consumption and MS instrument time. 275 Three new antagonists for A 2A R were identified as a result. It is likely that the throughput of this method could be further increased to assay close to or above 1 million compounds in one pool. 275 The affinity MS technique has been widely employed to fish out and identify putative ligands toward various enzyme targets from complex extracts of natural products, which could promote lead discovery from TCM. [276][277][278][279][280][281] Indeed, this technique was successfully extended to GPCR ligand screening from herbal extracts. It involved the optimization of receptor construct and integration of affinity MS with metabolomics data mining workflow for sensitive and accurate ligand identification 282 (Fig. 11d). After screening a panel of herbal extracts, a naturally occurring aporphine compound (1857) displaying strong subtype selectivity for 5-HT 2C without affecting 5-HT 2A or 5-HT 2B was discovered ( Fig. 11e-g). Moreover, this new lead exhibited exclusive bias toward G protein signaling and showed in vivo efficacy for food intake suppression and weight loss. 282 Although not directly applied to GPCRs, a previously reported cell-based assay vascular endothelial growth factor receptor 2 (VEGFR2) is interesting. 283 It used a special one-bead-onecompound library of peptoids and cells overexpressing VEGFR2. Beads bound to the color-coded VEGFR2-expressing cells were selected under fluorescence microscopy and the attached ligands decoded by tandem MS analysis. Hits with low μM affinity to the soluble VEGFR2 ectodomain were identified subsequently. We envision that these membrane-based or cell-based screening platforms will make a major impact on GPCR drug discovery, especially when they are fully integrated.
Competitive MS binding assay employs a non-radioactive ligand to compete the binding of a test compound to a protein target. It resembles radioligand-binding assays but avoids the use of radioisotope. [284][285][286] When assaying, the marker ligand liberated from the target is measured by an multiple reaction monitoring-based MS method of high sensitivity and selectivity for compound detection. Not only established for a number of transporters and ion channels, [287][288][289] this approach is equally effective in addressing GPCRs as recently exemplified on A 1 AR/A 2A R and DRD1/2/ 5. 267,285,290,291 It was shown that unlabeled marker compounds could substitute their radiolabeled counterparts in all types of ligandbinding characterization studies, including saturation, displacement, dissociation, and competitive association, yielding results in excellent accordance with classic radioligand-binding assays. 267,290 EMERGING OPPORTUNITIES AND PROSPECTS Recent scientific and technological advancements in GPCR biology have provided an enormous amount of information that will benefit our current and future efforts in rational drug design. Integration and refinement of massive data by artificial intelligence is a clear direction to guide both virtual and experimental screening of efficacious therapeutic agents with new scaffolds and of novel chemotypes for all classes of GPCRs.
However, as described in this review, factors that influence GPCR drug discovery include, but not limited to, therapeutic target, chemical diversity, mechanism of signaling, ligand-binding site, mode of action, clinical indication, polypharmacology, etc. Future opportunities may arise from: (i) de-orphanization of orphan GPCRs to provide novel targets; (ii) new indication for drug intervention via discovery and/or repurposing efforts; (iii) development of lead compounds targeting classes B2 and F GPCRs to address unmet medical needs; and (iv) validation of polypharmacology may lead to improved drug therapies.