A biased allosteric modulator functions as a molecular glue to induce β2AR dimerization

A biased allosteric modulator functions as a molecular glue to induce β2AR dimerization | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Biological Sciences - Article A biased allosteric modulator functions as a molecular glue to induce β 2 AR dimerization Brian Kobilka, Jiemin Shen, Teja Peddada, Konstantin Komolov, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8001844/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Family A G-protein coupled receptors (GPCRs) are typically described as monomers, yet growing evidence suggests they can form dimers with distinct signaling properties1-3. The mechanisms and therapeutic potential of such dimerization, however, remain poorly understood. Here, we show that AP-7-168, an optimized derivative of a β-arrestin-biased negative allosteric modulator of the β2-adrenergic receptor (β2AR) that sustains bronchorelaxation in cell and tissue models4, functions as a molecular glue to promote β2AR homodimerization. Cryo-EM structures reveal a unique binding mode in which two AP-7-168 molecules pack within a pocket formed by transmembrane helices 3, 4, and 5 of two protomers, stabilizing a dimeric conformation that selectively prevents β-arrestin coupling. In cells, AP-7-168 robustly induces β2AR dimerization and drives enlarged nanocluster formation. Combined with extensive functional studies, our findings unveil a novel allosteric mechanism by which a small molecule biases β2AR signaling through dimerization, highlighting ligand-induced dimerization as a strategy for GPCR modulation. Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy Biological sciences/Biochemistry/Proteins/G protein-coupled receptors Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction G-protein coupled receptors (GPCRs) represent the largest superfamily of transmembrane (TM) proteins, exhibiting complex signaling behaviors. Depending on the ligand, GPCRs can interact with G-proteins or bind to effector proteins such as arrestins and kinases, mediating G-protein-independent pathways and giving rise to diverse pharmacological and physiological responses 5,6 . Over 800 human GPCRs have been identified and systematically classified into five families based on their sequence features and structural characteristics 7 : i) family A Rhodopsin-like, ii) family B1 Secretin-like, iii) family B2 Adhesion-like, iv) family C Glutamate-like, and v) family F Frizzled-like receptors. Although family C receptors are known to be obligate dimers 8 , mediated primarily by their large Venus flytrap extracellular domains (ECDs), dimerization of other families is not well described. Over the past few decades, many studies suggest that family A receptors also possess the ability to form dimers 9-11 . Recent cryo-EM studies on rhodopsin 1 , apelin receptor (APJR) 2 , and GPR3 3 provide direct evidence for the dimerization of these family A GPCRs. However, unlike family C receptors, family A receptors do not have a significant ECD and therefore rely on interactions between their TM domains to undergo dimerization 12 . Despite the various studies on family A dimers, their physiological relevance is still debated. The development of ligands that stabilize family A receptor dimers remains elusive 13,14 , and it is difficult to reliably reproduce these dimers and to comprehensively study their functional, biochemical, and biophysical properties. We previously identified a class of negative allosteric modulators (NAMs) selective for the β 2 -adrenergic receptor (β 2 AR), a family A GPCR, that exhibited biased inhibition of β-arrestin recruitment with minimal impact on stimulatory G protein (Gs)-mediated cAMP production in cell-based assays 4 . Through comprehensive structure–activity relationship (SAR) studies, we developed AP-7-168 (hereafter referred to as AP), a compound that has high potency and biased effects. Activation of β 2 AR upon coupling with Gs triggers the downstream cAMP signaling pathway, leading to many physiological effects including the relaxation of airway smooth muscle, which is the core mechanism underlying the use of β 2 AR agonists in treating airway diseases such as asthma and chronic obstructive pulmonary disease. However, upon prolonged agonist stimulation, β 2 AR undergoes desensitization via arrestin-mediated inhibition of Gs coupling and receptor endocytosis, resulting in a diminished bronchodilatory effect. Our ex vivo tissue assays demonstrated that AP effectively maintained agonist-induced β 2 AR activation and the subsequent bronchodilation while significantly delaying receptor desensitization 4 , highlighting its potential clinical value. Despite these promising findings, the molecular mechanism by which AP modulates β 2 AR function remains unclear, posing a significant obstacle to the further development and optimization of AP as a therapeutic agent. In this study, using cryogenic electron microscopy (cryo-EM) combined with various biochemical and biophysical approaches, we demonstrate that AP functions as a molecular glue by stabilizing a homodimer complex of β 2 AR, which in turn alters the downstream signaling behaviors. These findings reveal a novel allosteric modulation mechanism in β 2 AR that may have broad implications in drug development for family A GPCRs. Results Structure determination of AP-bound β 2 AR As depicted in Fig. 1A, AP and its parent compound difluorophenyl quinazoline (DFPQ) 4 have the pharmacophore characterized by a quinazoline ring with 2- and 4-amino substitutions. AP features a 3,4-difluorophenyl group at the 2-amino position, a cyclohexane substituent at the 4-amino position, and a bromine atom at the 6-position of the quinazoline ring. AP exhibited a remarkable biased NAM effect in β 2 AR-expressing cells, being approximately 1,000-fold more effective at inhibiting β-arrestin2 recruitment versus cyclic adenosine monophosphate (cAMP) production when stimulated with the β-agonist isoproterenol (Iso) (Fig. 1B). To elucidate the molecular mechanism underlying this biased inhibition, we sought to determine the structure of β 2 AR bound to AP. We initially prepared the cryo-EM sample using detergent-solubilized and size-exclusion chromatography (SEC)-purified β 2 AR incubated with AP in the presence of carazolol, a β-antagonist. Given the small size (< 50 kDa) of a monomeric receptor, we included nanobody 60 (Nb60), which is specific to an inactive conformation of β 2 AR 15 , for steering particle alignment in cryo-EM data processing. Unexpectedly, the extracted particles from the dataset revealed two predominant 2D class averages, with approximately 70% of monomers and 30% of dimers (Supplementary Fig. 1A). Subsequent 3D classification and reconstruction showed the dimer consists of two parallel protomers with a symmetric TM3, TM4, and TM5 interface (Supplementary Fig. 1). Owing to the larger size and the C2 symmetry of the dimer, we obtained a cryo-EM map at a resolution of 2.9 Å (Supplementary Fig. 1A). The map revealed two distinct AP densities located at the dimer interface near the center of the membrane (Fig. 1D and Supplementary Fig. 1A). This finding suggests that AP may promote β 2 AR dimerization. We also managed to reconstruct a 4.3 Å resolution map for monomeric β 2 AR bound to Nb60 in the same dataset (Supplementary Fig. 1A). Despite the modest resolution, a distinct non-protein density, comparably strong to that of TM helices, was observed near the AP binding site identified in the dimer structure (Supplementary Fig. 2A,B). This indicates that AP can bind to monomeric β 2 AR at a similar position as in the dimer. Surface plasmon resonance (SPR) measurements further confirmed the direct binding of AP to the monomeric receptor with a K D of approximately 30 μM (Supplementary Fig. 2C). Encouraged by the initial dimeric structure in detergent, we sought to produce a homogeneous dimer sample and solve its structure in a native-like environment. Although incubation of AP with purified monomeric β 2 AR in detergent did not result in efficient dimerization as assessed by SEC (Fig. 1C), we achieved almost complete dimerization when β 2 AR was pretreated with AP in a membrane environment before detergent extraction (Fig. 1C and Methods). This highlights the critical role of a continuous fluid membrane in facilitating TM contacts mediated by AP and maintained in detergent after solubilization. We reconstituted the dimer into lipid nanodiscs and solved its structure in complex with an agonist, BI-167107 (BI), without nanobodies (Supplementary Fig. 1B). Cryo-EM analysis confirmed the presence of predominantly dimeric species, with no monomer detected in 2D classification (Supplementary Fig. 1B). The final cryo-EM map achieved a resolution of 2.5 Å, with a local resolution of ~2.2 Å near the AP binding site (Fig. 1D,E and Supplementary Fig. 1B). Structural basis of AP binding The high-resolution map enabled precise modeling of two AP molecules in the dimer, revealing a novel ligand binding mode (Fig. 1E, Supplementary Fig. 3, and Supplementary Table 1). The two AP molecules are packed against each other through π–π and van der Waals interactions and fit in an allosteric pocket formed by TM3, TM4, and TM5 of two β 2 AR protomers (Fig. 1E,2A–C). The two AP molecules also form extensive hydrophobic and van der Waals interactions with a series of non-polar residues in TM3 (C125 3.44 , V126 3.45 , V129 3.48 ), TM4 (M156 4.48 , M157 4.49 ), and TM5 (V206 5.45 , V210 5.49 , I214 5.53 ) (Fig. 2A–C). In addition, the 2-amino group in AP forms hydrogen bond interactions with E122 3.41 in TM3 (Fig. 2A). These specific interactions in our structural model rationalize previously reported SAR and mutagenesis data 4 . For example, removing the bromine atom reduced AP’s activity, likely because of the loss of interactions with V206 5.45 (Fig. 2A). The difluorophenyl substitution forms optimal contacts with V126 3.45 and V129 3.48 (Fig. 2C), whereas larger halogen atoms reduce ligand potency 4 . Similarly, V129 3.48 L or swapping TM3 with β 1 -adrenergic receptor (β 1 AR), where isoleucine replaces the valine residue, introduces steric clashes with the difluorophenyl group, therefore significantly reducing the potency of AP. Moreover, the E122 3.41 W mutation also leads to almost complete loss of response to AP 4 . Interestingly, the binding site of AP on β 2 AR closely overlaps with a previously identified binding site for an unbiased NAM, AS408 16 (Fig. 2D,E). Structural alignment of one of the β 2 AR protomers in the AP-bound dimer with the β 2 AR-AS408 complex (PDB ID: 6OBA) reveals a highly similar overall conformation, with an overall Cα root mean square deviation (RMSD) of only 0.53 Å (Fig. 2E). Both AP and AS408 possess a bromide substitution at the C6 position of the quinazoline ring (Fig. 2D,E), and this bromide is positioned in close proximity to V206 5.46 in both structures (Fig. 2E). However, despite the similarities in the quinazoline ring and the bromide position, there are notable differences in their binding orientations. The quinazoline rings of AP and AS408 are nearly parallel, but they are flipped 180 degrees relative to each other (Fig. 2E). This difference in orientation is likely due to the absence of a bulky substitution at the 4-amino position in AS408, in contrast to AP. Consequently, in the AS408-bound structure, the 4-amino group of AS408 forms a hydrogen bond with E122 3.41 , while in the AP-bound structure, it is the 2-amino group of AP that interacts with E122 3.41 (Fig. 2E). E122 3.41 has been proposed to be a critical conformational hub in β 2 AR 16 . It serves as a key link that connects the extracellular orthosteric ligand-binding pocket to the intracellular transducer-coupling interface. The shift in hydrogen bonding partners, from the 4-amino group of AS408 to the 2-amino group of AP, stabilizes E122 3.41 in a distinct conformation (Fig. 2E). This unique conformation of E122 3.41 may have significant implications for modulating the interactions of β 2 AR with different intracellular transducers, such as Gs and β-arrestins. Moreover, the cyclohexane group in AP, which is not present in AS408, plays a crucial role in ligand packing and hydrophobic contacts with the adjacent β 2 AR protomer. These differences may explain why AS408 lacks the ability to promote receptor dimerization and the biased NAM effect of AP. AP-mediated β 2 AR dimer interface The AP-mediated β 2 AR dimer interface is primarily built by TM3, TM4, and TM5, along with the loop regions ECL2 and ICL2 (Fig. 2F). In addition to the AP-mediated interactions around the central binding pocket (Fig. 2F–H), there are extensive intermolecular interactions between the two β 2 AR protomers from both extracellular and intracellular sides. As shown in Fig. 2F–H, in the extracellular side, L167 4.59 and M171 4.63 in TM4 and W173 in ECL2 of one protomer form extensive hydrophobic and van der Waals interactions with Q197 5.36 , A198 5.37 , I201 5.40 , and I205 5.44 in TM5 of the other protomer. On the intracellular side, K149 4.41 in TM4 forms a hydrogen-bonding interaction with the conserved Y132 3.51 (of the DRY motif) in TM3 of the adjacent protomer. In addition, F133 3.52 in TM3 and F139 34.51 , K140 34.51 , L144 34.56 , and L145 34.57 in the ICL2 loop create an extensive hydrophobic network between the two protomers. Moreover, we observed clear lipid-like densities, running almost parallel to the membrane plane, bridging hydrophobic contacts between the two protomers (Fig. 2F–H), which further stabilize the dimer conformation of β 2 AR. The β 2 AR dimer interface captured in our cryo-EM structures differs from the potential interface observed in the lipid cubic phase (LCP) crystal structure of β 2 AR 17 (Supplementary Fig. 4A). The AP-stabilized β 2 AR dimer interface also contrasts with those in the APJR dimers 2,12,18,19 (Supplementary Fig. 4B), rhodopsin dimer 1 (Supplementary Fig. 4C), and GPR3 dimer 3,20,21 (Supplementary Fig. 4D), while it is similar to a potential β 1 AR dimer interface in the LCP crystal structure 22 (Supplementary Fig. 4E). While previous biophysical studies have shown that the β 2 AR could form dimers and higher-order oligomers by itself 10,23 , it is possible that the new interface in our structures represents a native yet transient one for β 2 AR. Validation of AP-mediated β 2 AR dimerization in liposomes and in cells The above structures of AP-bound β 2 AR dimer were resolved in detergent micelles or nanodiscs. To validate whether the AP-mediated β 2 AR dimer adopts the same structural arrangement in a continuous membrane, we applied single-molecule fluorescence resonance energy transfer (smFRET) microscopy to probe its conformation in liposomes. As shown in Fig. 3A, the TM helices in the dimer structure resolved by cryo-EM form a parallelogram-like arrangement when viewed perpendicular to the membrane plane, with two diagonals of different lengths. Fluorescent donor and acceptor dyes labeled on residues near different vertices should yield distinct inter-dye distances and corresponding FRET values. To label each protomer in a site-specific manner, single-cysteine mutations were introduced into a minimal cysteine construct of β 2 AR 24 and “clicked” with maleimide-conjugated fluorescent dyes. Stochastic labeling is expected to result in half of the doubly labeled dimers having one donor and one acceptor dye. The labeled proteins were surface-immobilized and imaged using objective-based total internal reflection fluorescence (TIRF) microscopy. TM5-labeled samples exhibited a homogeneous Gaussian distribution of FRET values centered at ~0.9, while helix 8 (H8)-labeled samples showed a distribution centered at ~0.5 (Fig. 3A), with the higher FRET consistent with a shorter distance between the fluorophore pairs. These results are consistent with the cryo-EM structure and indicate the conformational stability of the dimer. To assess long-term stability, we measured ensemble FRET changes using dye-labeled samples in detergent (Supplementary Fig. 5A). When dimers were diluted into AP-free buffer, the relative FRET values remained nearly unchanged over 24 hours (Supplementary Fig. 5B,C). Moreover, incubation with Gs or activated β-arrestin could not dissociate the dimer (Supplementary Fig. 5D). The cryo-EM structure and biophysical data presented above provide strong evidence that AP promotes β 2 AR dimerization; however, it remains unclear whether this effect occurs in cells. To further address this, we employed two techniques to assess β 2 AR dimerization in cells. First, we analyzed the size of β 2 AR nanoclusters with or without AP treatment. GPCRs are known to form mesoscale nanoclusters in cells, which act as signaling hubs to locally concentrate effectors and enhance signaling efficiency 25,26 . Molecular mechanisms underlying nanoclustering are poorly understood but likely involve weak but specific interactions among receptors (Supplementary Fig. 6A). We hypothesized that AP stabilizes the dimer as a single unit that has increased interaction sites, facilitating prolonged contacts and the assembly of larger receptor clusters (Supplementary Fig. 6A). To visualize nanocluster distribution on cell membranes, we labeled β 2 AR with a gold nanoparticle (AuNP)-conjugated antibody and prepared cryo-EM grids with cytoplasmic content removed via unroofing. As shown in Supplementary Fig. 6B,C, cryo-EM imaging revealed distinct AuNP clusters rather than random scatters. Statistics of AuNPs in each cluster demonstrated that AP significantly increased nanocluster size (Fig. 3B) as a result of AP-induced dimerization of β 2 AR. The resulting enlarged nanoclusters may amplify signaling outcomes compared to isolated monomers or dimers combined. We also performed cell-based Bioluminescence Resonance Energy Transfer (BRET) assays to quantify β 2 AR dimerization by measuring distance-dependent BRET signals between Rluc-tagged and GFP-tagged receptors (Fig. 3C). Cells co-expressing differently tagged β 2 AR exhibited AP-dependent dimerization with a pEC 50 of 5.93 ± 0.18, n = 4 (Fig. 3D). Control experiments confirmed the specificity of this effect, showing no fluorescence interference from AP, no dimerization induced by Iso alone, and no expression-level-dependent self-homodimerization (Supplementary Fig. 7A–C). To further validate dimerization specificity, we tested a β 2 AR mutant (V129 3.48 L) that disrupts the AP binding pocket due to steric clashes. As expected, this mutation completely abolished AP-mediated dimerization (Fig. 3E). Similar effects were observed when swapping TM3 from β 1 AR into β 2 AR (Supplementary Fig. 7D) or assessing AP-mediated β 1 AR dimerization (Supplementary Fig. 7E). These results collectively demonstrate that AP robustly promotes β 2 AR dimerization in native cellular environments. Higher-order oligomerization of β 2 AR Apart from the dimer species in the cryo-EM dataset, a subset of 2D classes displayed top views of higher-order oligomers, including tetramers and hexamers (Supplementary Fig. 8A). We reconstructed a low-resolution map of the tetramer, revealing a dimer-of-dimers organization (Supplementary Fig. 8B). Docking of two dimer structures into the map shows symmetric interactions between the dimers, mediated by the intracellular regions of TM5–TM6 and TM1–ICL1 (Supplementary Fig. 8C). Mass photometry of diluted dimer samples in detergent confirmed the presence of a significant fraction of tetramer in solution (Supplementary Fig. 8D–F). Consistent with these findings, immunogold imaging of cell membranes showed enlarged nanoclusters upon AP treatment (Fig. 3B and Supplementary Fig. 6), supporting the notion that the AP-induced dimer serves as a structural unit for higher-order oligomerization. Together, these results indicate that AP-bound β 2 AR can assemble into higher-order oligomers, providing a structural basis for receptor nanocluster formation in cells. The β 2 AR dime r functions as a biased signaling species To understand the structural basis of the biased NAM activity of AP, we first compared our dimer structure with previously resolved monomeric β 2 AR structures. One protomer in the dimer structure aligns well with the previous crystal structure of β 2 AR in an inactive conformation (PDB ID: 2RH1), with an overall Cα RMSD of 0.53 Å (Fig. 4A). Notably, the structure of the dimer in nanodiscs, resolved without nanobodies or engineered fusions, exhibits a distinct ICL2 conformation from the crystal structure with an Cα RMSD of 1.5 Å (Fig. 4A). F139 34.51 on ICL2 participates in dimer contact with the same residue on the opposing protomer (Fig. 2F,H), resulting in an outward shift away from the central cavity involved in G-protein engagement (Fig. 4B). This shift is accompanied by other conformational changes in ICL2 (Fig. 5B and Supplementary Fig. 3B): Y141 34.53 forms hydrogen bond and π–cation interactions with R131 3.50 in TM3, which normally forms a hydrogen bond with H269 6.31 in TM6; K263 6.25 at the intracellular end of TM6 appears to replace K140 34.52 on ICL2 in neutralizing the helical dipole of TM3 (Supplementary Fig. 3B). These local conformational rearrangements may alter the interplay between β 2 AR and its transducers. Structural analysis of β 2 AR bound to its transducers provided additional insights into how the dimer achieves biased inhibition. Alignment of β 2 AR–Gs 27 , β 1 AR–β-arrestin1 28 , and neurotensin receptor type 1 (NTSR1)–G protein-coupled receptor kinase 2 (GRK2) 29 structures with one protomer in the dimer (Fig. 4C) revealed significant steric clashes between β-arrestin1 or GRK2 and the intracellular side of the other protomer, preventing their engagement with the receptor. In contrast, docked Gs appeared to be accommodated. However, dimerized β 2 AR adopts an inactive-like conformation. Structural comparison with the Gs-bound receptor (Supplementary Fig. 9A) revealed that local conformational changes breaking dimer symmetry are needed to allow for the loop-to-helix transition in ICL2 (Supplementary Fig. 9B,C), necessary for engagement with Gs in the canonical state, while the outward movements of TM5 and TM6 are not restricted by the dimer contacts. Together, these data suggest that dimerization of the β 2 AR contributes to the biased NAM effect of AP. To further confirm the functional consequences of the AP-induced dimerization, we conducted a series of biochemical assays to investigate the in vitro activities of the β 2 AR monomer and dimer in isolation. Unlike cell-based assays, where β 2 AR may exist as a mixture of monomers and dimers after AP treatment, the homogeneous and stable dimer sample that we obtained from purification enabled us to characterize the functions of dimer and monomer individually. First, we demonstrated that the dimer could form a stable complex with nucleotide-free Gs. This complex eluted as a single peak in SEC (Supplementary Fig. 10A), and the peak fractions contained the bands for Gs heterotrimer on SDS-PAGE (Supplementary Fig. 10B). Estimation of the relative ratios of β 2 AR to Gα on gels indicated a 2:1 stoichiometry in the dimer–Gs complex, consistent with a docking model in which two Gs molecules cannot simultaneously engage the dimer (Fig. 4C). Preliminary cryo-EM analysis of the dimer–Gs complex shows 2D class averages with one Gs density associated with an oval-shaped micelle housing one dimer (Supplementary Fig. 10C). We then used bio-layer interferometry (BLI) to measure the binding affinities and kinetics of guanosine diphosphate (GDP)-bound Gs to monomeric or dimeric β 2 AR, with and without AP (Fig. 4D and Supplementary Fig. 10D–F). All samples exhibited μM-level affinity for Gs(GDP) (Supplementary Table 2). Despite adopting an inactive conformation in the structural models, the β 2 AR dimer retains the ability to engage Gs, suggesting that the dimeric β 2 AR can adopt a Gs-favorable conformation. The dimer showed a modest (~3-fold) decrease in affinity for Gs (dissociation constant, K D = 3.11 μM) compared to the monomer ( K D = 0.96 μM). AP at 10 μM caused a slight reduction in the monomer–Gs binding affinity ( K D = 1.7 μM). Interestingly, dimer–Gs binding exhibited both a slower association rate ( k on ) and dissociation rate ( k off ) compared to monomer–Gs binding, while AP's effect on the monomer only reduced k on (Supplementary Table 2). These results suggest that AP partially inhibits recruitment of Gs by β 2 AR, with a more pronounced effect when tightly bound within the dimer. The smaller k off observed in dimer–Gs dissociation may reflect a cooperative mechanism, where Gs engagement with one protomer promotes a Gs-favorable conformation in the adjacent protomer, thereby increasing Gs residence time during dissociation. We further performed radioligand binding assays to assess the allosteric effect of AP on orthosteric ligand binding using isolated monomer or dimer reconstituted in nanodiscs. The AP-bound dimer showed an ~5-fold higher affinity for [ 3 H]-dihydroalprenolol (DHA), a β 2 AR antagonist, in the saturation binding, while AP at 10 μM did not significantly affect [ 3 H]-DHA binding to the monomer (Supplementary Fig. 11A). In agonist competition assays with Iso, the dimer exhibited a ~5-fold increase in the inhibitory constant ( K i ), but no significant effect of AP on the monomer (Supplementary Fig. 11B). These findings suggest that AP stabilizes β 2 AR in a more inactive-like conformation in the dimer. We next evaluated the potency of Gs in stabilizing the active state of β 2 AR by allosterically modulating the orthosteric pocket to reduce antagonist binding. We titrated Gs(GDP) into β 2 AR pre-bound with saturated [ 3 H]-DHA and quantified the remaining bound [ 3 H]-DHA. The results showed that Gs modulates the orthosteric pocket in both dimeric and monomeric β 2 AR, irrespective of AP presence (Fig. 4E). This suggests that while AP biases β 2 AR toward an inactive-like conformation, it does not significantly hinder Gs engagement at equilibrium. We then assessed the effects of AP on the function of β 2 AR as a guanine nucleotide exchange factor (GEF), which facilitates GDP-GTP exchange in Gs and accelerates GTP hydrolysis. Using the GTPase-Glo™ assay to monitor GTP turnover, we found that while AP significantly slows down the GTP turnover rate of monomeric β 2 AR, it did not affect the final efficacy. However, we observed much stronger inhibition in GTP turnover for the dimeric β 2 AR (Fig. 4F). This likely stems from two factors: (1) the slower association rate k on of Gs with dimeric β 2 AR as observed by BLI (Supplementary Fig. 10F) and (2) AP’s biasing of β 2 AR toward an inactive-like state, potentially introducing additional rate-limiting steps in a full productive cycle. To test if AP impedes β-arrestin core engagement with β 2 AR, we performed competition radioligand binding assays measuring β-arrestin–induced positive allosteric modulation of agonist binding. Using the sortase ligation strategy, we attached the synthetic phosphorylated vasopressin receptor 2 (V 2 R) peptide (V 2 Rpp) to monomeric or dimeric β 2 AR (Fig. 4G and Supplementary Fig. 12A), ensuring full tail engagement of β-arrestin, a prerequisite for efficient core engagement. Monomeric β 2 AR in nanodiscs displayed a ~7-fold increase in Iso affinity in the presence of β-arrestin1 (Supplementary Fig. 12B), consistent with previous reports 30,31 . AP reduced this increase in affinity to ~2-fold in the monomer and abolished it in the dimer (Fig. 4G), suggesting AP partially hinders β-arrestin core interactions on monomers and completely prevents it in dimers. These data align with structural predictions where steric clashes between β-arrestin and the adjacent protomer in the dimer prevent core engagement. Previously, we showed that DFPQ inhibits GRK-mediated phosphorylation of β 2 AR, a prerequisite for β-arrestin recruitment, in cells 4 . To identify which species is responsible for this inhibition, we measured GRK5 activity towards purified dimeric and monomeric β 2 AR. AP-bound β₂AR dimers exhibited near-complete inhibition of GRK5 phosphorylation in both detergent and nanodiscs (Fig. 4H and Supplementary Fig. 13A), whereas GRK5 activity toward monomer in nanodiscs remained unaffected by AP (Fig. 4H and Supplementary Fig. 13A). The slight reduction in GRK5 activity observed for monomer in detergent at high AP concentrations (Fig. 4I) likely reflects the inefficient dimerization of β₂AR in detergent (Supplementary Fig. 1A), as we previously observed (Fig. 1C). However, the dose-dependent inhibition observed for monomer reconstituted in bicelles (Fig. 5I) is a result of AP-induced dimerization in membrane, underscoring the requirement of a continuous lipid bilayer for effective receptor dimerization. The relatively large diameter of typical bicelles (~40 nm) compared to nanodiscs (~12 nm) increases the likelihood of incorporation of multiple receptors and provides a fluid membrane environment conducive to dimer formation. AP-dependent inhibition of GRK5 activity was further corroborated by a BRET-based assay monitoring AP effect on direct GRK5 recruitment to the β₂AR (Supplementary Fig. 13B,C). These findings are consistent with structural analyses indicating that steric clashes in the dimer hinder GRK engagement, further supporting the role of the dimer in preventing β-arrestin recruitment. Negative allosteric effect of AP on β 2 AR conformation dynamics Modulation of protein dynamics often underpins how ligands achieve biased effects in highly dynamic proteins like GPCRs. To provide further insights into the mechanism of AP’s biased NAM effect, we explored how AP tunes the dynamic behaviors of β 2 AR. To this end, we performed smFRET microscopy to monitor conformational changes and double electron–electron resonance (DEER) spectroscopy to measure residue-level distance distributions. Using the established protocols in our lab 24,32 , we placed fluorescent dyes for smFRET or spin labels for DEER near the intracellular ends of TM4 and TM6 on the minimal cysteine construct of β 2 AR. This allowed us to monitor TM6 outward movement, a hallmark of GPCR activation. Increased distances between TM4 and TM6 sites correspond to high-to-low FRET state transitions in smFRET experiments. As shown in Fig. 5A and detailed in the Methods, we used a two-affinity-tag purification strategy to produce an AP-mediated β 2 AR “heterodimer”, where only one protomer contained double-cysteine mutants for dye labeling. The recorded traces from the TM4–TM6 sensors of β 2 AR predominantly fall into two FRET states (Fig. 5B), representing an active and inactive state. In the absence of an agonist, β 2 AR was primarily in the inactive state; however, monomers without AP exhibited a significantly higher fraction of active species compared to dimers, which showed almost no active population (Fig. 5C,D). Adding AP to monomers substantially reduced the active state population from 26.5 ± 2.9% to 11.2 ± 3.2% (Fig. 5D), whereas adding Iso to dimers failed to increase active state species (Fig. 5C). These results indicate that AP stabilizes an inactive conformation of TM6, and Iso alone fails to drive TM6 transition into an active state in the dimer. This is consistent with the observed enhancement of antagonist binding in the dimer (Supplementary Fig. 11A,B). Upon adding Gs to the dimer or the monomer with AP, β 2 AR transitioned predominantly to an active state, although dimers retained a higher fraction of inactive species (Fig. 5C,D), in line with Gs engaging only one protomer of the dimer, while the other may remain in inactive or partially active conformation. This indicates the potential of Gs to overcome the AP-induced stabilization of the inactive conformation, driving receptor activation even in the dimeric form. Washing Gs away from dimers reversed the population to a primarily inactive state, resembling the condition without Gs coupling (Fig. 5C). DEER measurements on the β 2 AR (Supplementary Fig. 14), with spin labels reporting on the TM4–TM6 conformation (Fig. 5E), revealed that AP causes changes that are consistent to those observed by smFRET. Specifically, AP biases Iso-bound receptors toward a more homogeneous inactive-like conformation (Fig. 5F), while having a minimal effect on receptors bound to the super-agonist BI (Fig. 5G). The addition of Gs caused a fraction of the fully active conformer to become populated, regardless of AP presence (Fig. 5H). In addition to conformational effects, the modulation depth parameter in pulsed dipolar EPR spectroscopy provides a quantitative measure of spin cluster size, with higher values reflecting increased local concentrations of spin labels. Agonist-bound receptor exhibited a reduced modulation depth relative to the Apo receptor, suggesting that receptor activation is associated with decreased dimerization. Addition of AP to agonist-bound receptor resulted in an increase in modulation depth, consistent with enhanced dimerization (Supplementary Fig. 14). Notably, Gs binding largely inhibited the AP-induced increase in modulation depth, indicating that Gs coupling counteracts the dimer-promoting effect of AP. The magnitude of the shifts in modulation depth indicates slight changes in receptor monomer-dimer equilibrium consistent with a predominantly monomeric receptor, as expected for receptor in detergent. However, the clear trend in modulation depth indicates a correlation between the monomer-dimer and inactive-active equilibria. Together, these findings demonstrate that Gs can couple to both dimers and monomers with AP, driving TM6 outward movement. Moreover, the inhibitory effect of AP on the TM6 outward movement of β 2 AR monomer may also contribute to the level of biased signaling observed. This is supported by the observation that β-arrestin inhibition emerges at lower AP concentrations compared to dimerization in cell-based assays. Notably, the effect of AP on β 2 AR dynamics resembles the previous observation in μ-opioid receptor (μOR), where G protein-biased agonists are less efficient than balanced full agonists in stabilizing TM6 outward movement in both smFRET and DEER studies 33 . Discussion Our cryo-EM structures reveal that AP, a β-arrestin–biased NAM, promotes β 2 AR dimerization, thereby modulating downstream signaling. AP shows strong bias toward β-arrestin inhibition in cell-based assays, and its parent compound, DFPQ, produces sustained relaxation of contracted airway smooth muscle (ASM) in both human ASM cells and mouse airway models by preventing β 2 AR desensitization 4 , suggesting its physiological relevance. We elucidate the mechanistic basis underlying this biased effect: AP robustly induces β 2 AR dimerization. Biochemical and biophysical analyses demonstrate that the dimer acts as a biased signaling species, strongly inhibiting GRK-mediated phosphorylation and β-arrestin recruitment while permitting G protein coupling. Structural analysis indicates that steric hindrance imposed by the adjacent protomer prevents GRK and β-arrestin from adopting their canonical binding modes. These findings offer a novel perspective for achieving signaling bias: besides fine-tuning local conformational states, AP exploits the quaternary structure of receptor to selectively restrict transducer interactions. Based on our results, we propose an activation model for β 2 AR in the presence of AP (Fig. 6). We simplify the energy landscape into two states—an inactive ground state and a fully transducer-bound active state—and outline how AP reshapes this profile. In the absence of AP, the balanced agonist drives the receptor into an active conformation that readily accommodates transducer binding due to a low energy barrier. However, the energy barrier to achieve the fully transducer-bound active state is increased by AP. In the monomer, AP binds a similar pocket as in the dimer, biasing the receptor toward an inactive-like conformation. GRK phosphorylation remains largely intact; however, the reduced allosteric enhancement of agonist binding by β-arrestin suggests a partially destabilized engagement state. AP also slows β 2 AR-catalyzed nucleotide exchange in Gs, indicating a higher activation barrier for Gs, although Gs binding and activation remain largely unaffected, suggesting an unaltered active-state free energy. At higher concentrations, AP promotes stable β 2 AR dimerization, resulting in near-complete loss of GRK phosphorylation and blockade of β-arrestin core engagement. This reflects not only increased energy barriers but also a disrupted β-arrestin-bound state, due to steric clashes with the adjacent protomer. GTP turnover by Gs is further slowed in the dimer, again reflecting a higher activation barrier, while the free energy of the Gs-bound state remains similar to that of the monomer. In cells, where monomers and dimers coexist, AP shifts the receptor ensemble toward the more strongly biased dimeric form in a concentration-dependent manner. Despite reduced GTP turnover by Gs, signal amplification likely compensates and enables near-maximal cAMP production even under conditions where dimer is dominant. Thus, the loss of GRK and β-arrestin engagement in the dimer biases the receptor exclusively toward sustained Gs signaling, highlighting dimerization as a mechanism to selectively rewire family A receptor signaling. The dimer interface observed in our AP-bound β 2 AR complex displays unique features compared to other family A GPCR dimers. While previous studies have reported dimer structures in receptors such as APJR 12,18 and rhodopsin 1 by cryo-EM, our AP-stabilized dimer predominantly involves interactions on TM3, TM4, TM5, and ICL2, a configuration distinct from the earlier β 2 AR crystal structure 17 and other family A GPCR dimers. Importantly, the outward mobility of TM5 and TM6 is not restricted in our dimer configuration, which is essential for efficient transducer engagement. Notably, recent work 34 on the platelet-activating factor receptor using a cysteine crosslinking strategy reported that dimerization significantly biases signaling toward G protein pathways while limiting β-arrestin recruitment; one of their proposed dimer models resembles our structure with a TM3–TM4–TM5 interface. Similarly, recent studies used a computational and mutagenesis approach to design stable TM4–TM5 dimers of CXCR4, which biased signaling towards Gi 35 . Beyond discrete dimer formation, cryo-EM imaging reveals that AP-mediated β 2 AR dimerization drives the assembly of larger nanoclusters, and our structural model of the dimer-of-dimer provides a potential molecular basis for specific interactions that underlie this mesoscale organization. Such nanoclusters likely serve as dynamic signaling platforms that modulate both the amplitude and duration of receptor-mediated responses 25 . Recent papers 36-38 on higher-order transient structures (HOTS) highlight that dynamic oligomerization is a conserved strategy in membrane proteins, which may be beneficial for signaling efficiency and specificity in cells. Consistent with HOTS, AP-induced β 2 AR nanoclusters may fine-tune the balance between receptor activation, desensitization, and internalization. Although our GTP turnover assays on isolated AP-mediated dimers showed reduced turnover activities, the enlarged nanoclusters observed in cells may locally concentrate G proteins, thereby supporting more productive signaling events. Further investigations into the physiological relevance and functional consequences of these nanoclusters will help uncover the potential for harnessing such mechanisms across other receptor systems. Our findings extend the emerging paradigm of molecular glues in pharmacology 39,40 . AP functions as a GPCR molecular glue by bridging two β 2 AR protomers to form a dimeric assembly with biased functions. Although molecular glues that stabilize GPCR interactions with transducers 41 , including G proteins 42,43 , β‑arrestin 44,45 , and GRK 29 , have been reported, AP represents the first example of a ligand mediating TM interactions that drive dimerization of a family A GPCR. This mechanism is reminiscent of that observed in the membrane-bound molecular glue, NVS‑STG2 46 , for stimulator of interferon genes (STING). NVS‑STG2 induces high-order oligomerization of STING, robustly triggering downstream immune signaling 46 . Both examples illustrate how small molecules can harness and reshape the intrinsic propensity of membrane proteins to form oligomeric assemblies, thereby selectively modulating signal transduction. In summary, our study deciphers the structural basis by which AP acts as a biased NAM for β 2 AR and provides broader insights into how receptor dimerization impacts GPCR signaling. Such insights not only deepen our understanding of structure–function relationships in GPCR modulation but also highlight potential strategies for designing next-generation ligands that exploit oligomerization as a route to bias receptor signaling. Methods Expression and Purification of β 2 AR in Sf9 cells The β 2 AR construct PN1 was expressed and purified as previously described 32,47 . Briefly, Sf9 cells were infected with a PN1-containing baculovirus produced via BestBac method. Cells were then harvested and resuspended in chilled lysis buffer containing 10 mM HEPES, pH 7.4, 1 mM EDTA, 1 μM alprenolol, and protease inhibitors (leupeptin and benzamidine). Lysed cells were then pelleted at 18,600 rpm for 20 minutes and dounced to homogeneity in chilled solubilization buffer containing 20 mM HEPES pH 7.4, 350 mM NaCl, 1% n-dodecyl β-D-maltoside (DDM), 0.1% cholesteryl hemisuccinate (CHS), 2 mM MgCl 2 , 1 μM alprenolol, protease inhibitors, and benzonase. After stirring for 90 minutes at 4 o C and centrifugation at 18,600 rpm for 30 minutes, 2 mM CaCl 2 was added to the soluble fraction, which was then applied to anti-FLAG (DYKDDDDK) M1 immunoaffinity resin. The receptor was then washed (20 mM HEPES, pH 7.4, 350 mM NaCl, 0.1% DDM, 0.01% CHS, 2 mM CaCl 2 , leupeptin, and benzamidine), eluted (20 mM HEPES, pH 7.4, 350 mM NaCl, 0.1% DDM, 0.01% CHS, 5 mM EDTA, and 200 ug/mL FLAG peptide), and further purified on Superdex 200 10/300 Increase gel filtration column equilibrated in NH buffer (20 mM, HEPES pH 7.4, 100 mM NaCl) plus 0.1% DDM and 0.01% CHS. To produce homogeneous β 2 AR dimer, Sf9 cells expressing PN1 were resuspended at room temperature (RT) in 20 mM HEPES, pH 7.4, 150 mM NaCl, 10% glycerol, 1 μM alprenolol, protease inhibitors, and 10 μM AP. After a 30-minute incubation at RT, membrane solubilization was initiated by adding 1% lauryl maltose neopentyl glycol (LMNG) and 0.1% CHS. The purification steps followed the same protocol as for the monomeric receptor, except that 0.01% LMNG replaced DDM in all buffers, and 10 μM AP was maintained throughout the purification. Expression and Purification of heteromeric Gαsβ1γ2 As previously described 48,49 , heterotrimeric Gs was expressed and purified from Trichoplusia ni Hi5 cells. Briefly, two baculoviruses were generated from the BestBac method, one encoding the wild-type human Gαs subunit and the other encoding the wild-type human β1γ2 subunits containing a histidine tag on the N-terminus of the β subunit. Cells were infected with both viruses for 48 hours and harvested via centrifugation. The pellet was then resuspended and stirred for 30 minutes at 4 o C in hypotonic buffer containing 10 mM HEPES pH 7.4, 100 μM MgCl 2 , 5 mM β-mercaptoethanol (BME), 20 μM GDP, and protease inhibitors. Lysed cells were then pelleted at 18,600 rpm for 15 minutes and dounced to homogeneity in chilled NH buffer plus 1% sodium cholate, 0.05% DDM, 1 mM MgCl 2 , 5 mM BME, 20 μM GDP, and protease inhibitors. After solubilization for 1.5 hours while stirring at 4 o C and centrifugation at 18,600 rpm for 35 minutes, 20 mM of imidazole was added to the soluble fraction, which was then allowed to batch-bind to washed nickel-chelated Sepharose for 2 hours. Pelleted resin was then loaded into a narrow column, washed with buffers containing gradually declining amounts of cholate, and eluted with NH buffer plus 0.05% DDM, 1 mM MgCl 2 , 20 μM GDP, 100 μM tris(2-carboxyethyl)phosphine (TCEP), and 250 mM imidazole. Human rhinovirus 3C protease was added to cleave the histidine tag and the eluate was dialyzed overnight at 4C in 2 L of dialysis buffer (NH buffer plus 1 mM MgCl 2 , 0.05% DDM, 20 μM GDP, and 100 μM TCEP). The protein solution was run through a second nickel-chelated Sepharose column, washed with dialysis buffer supplemented with 20 mM imidazole, and dephosphorylated for 30 minutes on ice with lambda protein phosphatase, calf intestinal phosphatase, and Antarctic phosphatase with 1 mM manganese chloride. The heterotrimer was further purified from excess βγ subunits by using ion exchange chromatography on a MonoQ 10/100 GL column. Sample was loaded and washed with 20 mM HEPES, pH 7.4, 1 mM MgCl 2 , 0.05% DDM, 100 μM TCEP, and 20 μM GDP. Heterotrimer Gs was then eluted with a linear salt gradient from 50 mM NaCl to 500 mM NaCl. Expression and Purification of Nb60 Nb60 was expressed and purified as previously described 15 . Briefly, Nb60 was expressed in Escherichia coli BL21(DE3) cells. Lysate was then purified on nickel-chelated Sepharose column and subsequently on a Superdex 200 Increase 10/300 column in 20 mM HEPES, pH 7.4, and 150 mM NaCl. Sample preparation for cryo-EM For the sample in detergent, purified PN1 in 0.1% DDM/0.01% CHS was loaded onto anti-FLAG M1 immunoaffinity resin equilibrated in the same detergent mixture supplemented with 2 mM CaCl 2 . A detergent exchange was then performed, during which the receptor bound to the resin was washed with increasing ratios of TTG-T10 to DDM, ultimately transitioning the receptor into 0.01% TTG-T10/0.001% CHS. Each detergent exchange buffer contained NH buffer with 2 mM CaCl 2 , 1 μM carazolol, and 10 μM AP. The receptor was subsequently eluted in NH buffer plus 0.01% TTG-T10, 0.001% CHS, 1 μM carazolol, 10 μM AP, 5 mM EDTA, and 200 ug/mL FLAG peptide. After the detergent exchange, PN1 was incubated for 1 hour with 2-fold molar excess of Nb60. Excess Nb60 was cleared on anti-FLAG M1 immunoaffinity resin after washing with NH buffer plus 0.0006% TTG-T10, 0.00006% CHS, 1 μM carazolol, 10 μM AP, and 2 mM CaCl 2 . The complex was then eluted off the resin with NH buffer plus 0.0006% TTG-T10, 0.00006% CHS, 1 μM carazolol, 10 μM AP, 5 mM EDTA, and 200 μg/mL FLAG peptide. The final sample was concentrated to >10 mg/mL and used immediately for cryo-EM grid preparation. To prepare the AP-bound β 2 AR dimer in lipid nanodiscs, the purified dimer in LMNG was reconstituted into nanodiscs following a previous protocol 47 with modifications. Lipids were prepared by mixing 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC, Avanti), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS, Avanti), and cholesterol (Sigma-Aldrich) at a molar ratio of 7:2:1, followed by drying under argon and vacuum desiccation for 2 hours. Lipids were resuspended in NH buffer containing 14 mM DDM at 20 mg/mL. The receptor was diluted to 10 μM, incubated with 50 μM BI (MedChemExpress) for 10 minutes on ice, and then combined with membrane scaffold protein (MSP) 1E3D1 and lipids at a molar ratio of 1:2.5:100. The mixture was incubated on ice for 1 hour to allow nanodisc assembly. Detergent was removed by sequential addition of semi-wet Bio-Beads SM2 (60 mg/mL) three times over 3-hour period, followed by overnight incubation at 4°C. The next day, Bio-Beads were removed, and empty nanodiscs were separated by M1 affinity purification. The dimeric β₂AR in nanodiscs were eluted in buffer containing 1 μM BI, 10 μM AP, 5 mM EDTA, and 200 μg/mL FLAG peptide. The eluate was further polished by SEC in NH buffer plus 1 μM BI and 10 μM AP. Peak fractions were pooled and concentrated to 5 mg/mL for grid freezing. The cryo-EM grids were prepared using Vitribot Mark IV (Thermo Fisher Scientific). Quantifoil R1.2/1.3 Au grids were glow-discharged with air for 90 s at 10 mA using Plasma Cleaner (PELCO EasiGlowTM). Aliquots of 3 μL protein sample were applied to the glow-discharged grids. After blotting with filter paper (Ted Pella, Inc) for 3.0 s, the grids were plunged into liquid ethane cooled with liquid nitrogen. Cryo-EM data collection and processing For detergent- and nanodisc-reconstituted β 2 AR samples, a total of 7,139 and 17,997 micrograph stacks, respectively, were collected using a Titan Krios G3i operating at 300 kV on a Falcon 4i direct electron detector or on a K3 camera (Gatan) with a Quantum energy filter. Micrographs were recorded at a nominal magnification of 130,000×, corresponding to a calibrated pixel size of 0.95 Å (detergent) or 0.83 Å (nanodisc) with defocus values ranging from –2.0 μm to –0.8 μm. Each movie stack received a total electron dose of ~50 e⁻/Ų over 40 frames. Motion correction and contrast transfer function (CTF) estimation were performed using the patch motion and patch CTF in cryoSPARC 50 . Particle picking was carried out using the blob picker, and extracted particles were binned 4× for initial 2D classification. 2D classes with recognizable structural features were manually selected, and monomeric and dimeric particles were grouped separately. Ab initio reconstruction was performed using ~10% of the selected particles to generate four initial 3D references. Iterative rounds of heterogeneous refinement were conducted until poor-quality classes accounted for less than 5% of input particles. For the detergent dataset, 48,805 dimeric and 25,201 monomeric particles were selected for non-uniform (NU) refinement 51 . For the nanodisc dataset, 237,408 particles of dimeric species were selected. Particles were imported into RELION 4 52,53 for Bayesian polishing 54 and subsequently returned to cryoSPARC for further NU refinement. C2 symmetry was imposed for the final NU refinement of dimer species After handedness correction, local and CTF refinement were performed using a soft mask centered on the transmembrane domains. Final resolutions were estimated using the gold-standard Fourier shell correlation (FSC) 0.143 criterion. Local resolution estimation was carried out in cryoSPARC. Model building and refinement Initial models of β 2 AR (PDB ID: 2RH1) were docked into cryo-EM maps using UCSF Chimera 55 and manually adjusted in COOT 56 to fit the density, including placement of ligands. Real-space refinement was performed in PHENIX 57,58 with secondary structure and geometry restraints. Model validation was conducted using EMRinger 59 to assess side chain density fitting. Structural figures were prepared using PyMOL and ChimeraX 60 . Mass photometry Mass photometry measurements were performed using a Refeyn TwoMP instrument (Refeyn Ltd.) and the AcquireMP software (v2.3) following an established protocol 61 . Microscope coverslips (24 × 50 mm, #1.5 thickness; Corning) were cleaned with deionized water and isopropanol, then dried prior to use. Silicone gaskets were applied to the coverslips to form individual wells immediately before sample loading. The instrument was calibrated using NativeMark unstained protein standards (Thermo Fisher Scientific) following the manufacturer's instructions. Each measurement was conducted by first pipetting 10 μL of NH buffer into a well, followed by focal alignment and locking. Then, 1 μL of β 2 AR sample at a final concentration of 20 nM after dilution was added, mixed gently, and data were acquired for 60 seconds. At least 2,000 binding events were recorded per sample. Data processing and molecular mass determination were performed using the DiscoverMP software suite (Refeyn). Cell unroofing and immunogold cryo-EM Cell unroofing and immunogold labeling were performed as previously described 36,37,62 , with modifications. Quantifoil R1.2/1.3 300-mesh gold EM grids were glow-discharged for 15 seconds, rinsed three times with 70% ethanol, and washed four times with DPBS. Grids were incubated with 0.1 mg/mL poly-D-lysine (Gibco) for 1 hour at RT, followed by four washes in Dulbecco's phosphate-buffered saline (DPBS, Gibco). Laminin (Sigma-Aldrich, 15 μg/mL) was applied to the grids and incubated at 37°C for 2 hours, then washed again with DPBS. HEK293F cells in suspension were seeded onto prepared grids and cultured until reaching 60–70% confluency. β 2 AR was overexpressed via baculovirus transduction following the BacMam protocol (Thermo Fisher Scientific). Around 16 hours after transfection, cells were rinsed with DPBS containing calcium and magnesium. Grids bearing adherent cells were held with tweezers and dipped into a hypotonic swell buffer (6 mM HEPES-KOH pH 7.4, 43.3 mM K-gluconate, 1.6 mM NaCl, 0.6 mM MgCl 2 ) for 30 seconds. An additional 6 μL of swell buffer was added to each grid, followed by blotting with Whatman Grade 5 filter paper (Sigma-Aldrich) to remove the apical membrane. Unroofed samples were blocked with 3% goat serum (Thermo Fisher Scientific) in DPBS containing protease inhibitors (Thermo Fisher Scientific) for 20 minutes at RT. Primary antibody against β 2 AR (Thermo Fisher Scientific) diluted in DPBS with 1% goat serum was applied for 1 hour. After three washes in DPBS, grids were incubated with a goat anti-mouse secondary antibody conjugated to colloidal gold (Ted Pella, Inc) diluted in DPBS with 1% goat serum for 1 hour, then washed again three times. Grids were plunge-frozen in liquid ethane using a Leica EM GP2 system and stored in liquid nitrogen. Cryo-EM imaging was performed on a Glacios G2 operated at 200 kV, equipped with a Falcon 4i detector, at a nominal magnification of 11,000×. Cell-based BRET To evaluate the effects of AP on cAMP production and β-arrestin recruitment, HEK293 cells endogenously expressing β 2 AR were transiently transfected with either the BRET-based intramolecular cAMP sensor CAMYEL or with plasmids encoding β-arrestin2–GFP10 and β 2 AR–RlucII. CAMYEL comprises both donor and acceptor fused to the cAMP-binding domain of EPAC and undergoes a conformational change upon cAMP binding that alters the BRET signal 63 . Forty-eight hours post-transfection, cells were pre-incubated with increasing concentrations of AP (0.03–100 μM) for 30 minutes, followed by stimulation with 1 μM isoproterenol for 30 minutes in the presence of 5 μM coelenterazine H (for CAMYEL) or deep blue coelenterazine (DBC) (Cayman Chemical, Ann Arbor, USA) for the β-arrestin assay. For BRET-based β 2 AR dimerization studies, HEK293 cells were co-transfected with 10 ng/well of either β 2 AR–RlucII, β 2 AR_V129L–RlucII, β 2 AR_TM3–RlucII, or β 1 AR–RlucII (BRET donors), along with 20 ng/well of their respective GFP-tagged counterparts (BRET acceptors). Forty-eight hours post-transfection, cells were incubated with increasing concentrations of AP (0.03–100 μM) or Iso (0.1–100 μM) for 1 hour, followed by addition of DBC substrate and incubation for 20 minutes. To control for potential AP autofluorescence effects on the BRET signal, HEK293 cells transfected with β 2 AR–GFP were treated with AP (0.03–100 μM) for 30 minutes, and the induced GFP signal was directly measured and represented as fold change over basal. For saturation studies of β 2 AR homodimerization, cells were co-transfected with a fixed amount of β 2 AR–RlucII (10 ng/well) and increasing amounts of GFP-tagged β 2 AR plasmid (0–100 ng/well). BRET measurements were taken 48 hours post-transfection after incubation with DBC for 30 minutes. To evaluate GRK5 recruitment to the β 2 AR, cells were co-transfected with β 2 AR–Rluc and GRK5–GFP. Cells were either stimulated with increasing concentrations of Iso for 30 minutes followed by DBC incubation for 20 minutes or pre-treated with increasing concentrations of AP (0.03–100 μM) for 1 hour followed by 1 μM Iso stimulation for 30 minutes in the presence of the Rluc substrate DBC. All BRET signals were recorded at 395 nm (donor emission) and 510 or 530 nm (acceptor emission) using an Infinite F500 plate reader (Tecan, Männedorf, Switzerland). Induced BRET changes were calculated by subtracting the basal BRET signal (in the absence of ligand) from the BRET signal measured after stimulation. For the AP-mediated cAMP and β-arrestin assays as well as GRK5 recruitment, results are expressed as percent of the Iso-alone response. For dimerization studies and Iso-induced GRK5 recruitment, data are presented as percentage of the maximal BRET signal (% of MAX). All dose–response curves were fitted using the log(agonist/inhibitor) vs. response (three parameters) function in GraphPad Prism. Data are shown as mean ± S.E.M., n = 3/4. SPR analysis on AP binding SPR measurements were performed using a Biacore T200 system. Monomeric β 2 AR was captured onto a high-affinity streptavidin (SA) sensor chip (Cytiva) via a biotinylated anti-FLAG M2 antibody (Sigma-Aldrich), yielding a final response of ~1500 resonance units (RU). The running buffer consisted of 20 mM HEPES (pH 7.5), 100 mM NaCl, 0.01% LMNG. AP was injected at increasing concentrations ranging from 0.6 to 10 μM at a flow rate of 30 μL/min. Association and dissociation phases were recorded for each injection. All sensorgrams were processed using double-referencing to correct for nonspecific binding and instrument artifacts. This was accomplished by subtracting both (i) the response obtained from a blank injection of running buffer over the active surface (to account for bulk refractive index changes and injection artifacts), and (ii) the response from the compound injection over a reference surface without immobilized protein (to correct for nonspecific binding to the surface and matrix effects). Sensorgrams were processed using Biacore Evaluation Software, and binding curves were fit globally using a steady-state affinity model. BLI on Gs binding BLI measurements were conducted at 30°C with continuous shaking at 1000 rpm using an Octet RED384 system (FortéBio). SA biosensor tips (Sartorius) were coated with 10 nM biotinylated anti-FLAG M1 fragment antigen-binding region (Fab) in the NH buffer with 0.01% LMNG and 0.001% CHS for 300 seconds. FLAG-tagged β 2 AR was then captured by incubating the tips in 100 nM β 2 AR with 10 μM Iso for 600 seconds. Following receptor immobilization, the biosensors were transferred into wells containing a concentration series of Gs (30 nM to 10 μM) in the binding buffer containing 10 μM GDP and 0.1% bovine serum albumin (BSA, Sigma-Aldrich) for 180 seconds (association phase), followed by transfer into buffer-only wells for 300 seconds (dissociation phase). Control channels lacking either Gs or immobilized β 2 AR were used for double-reference subtraction. Association and dissociation kinetics were fitted with a single-exponential model to derive apparent k on and k off . Equilibrium binding responses were used to determine the dissociation constant K D . Radioligand binding assay For saturation binding studies, 50-100 femtomoles of monomeric or dimeric β 2 AR reconstituted in nanodiscs, following the protocol described in the sample preparation for cryo-EM, were incubated with increasing concentration of [ 3 H]-DHA at RT for 1 hour in a buffer containing 20 mM HEPES, 100 mM NaCl, 0.5% BSA. Non-specific binding of the radioligand was determined by adding 10 μM alprenolol in the same reaction system. For monomeric β 2 AR, the assay was performed with or without AP. For competition binding studies, monomeric or dimeric β 2 AR reconstituted in nanodiscs were incubated with 1 nM [ 3 H]-DHA and increasing concentrations of Iso or GDP-bound Gs in the same buffer as saturation binding. Nanodiscs were separated from excess [ 3 H]-DHA on Whatman GF/B filters using a Brandel 96-well harvester. The bound radioligand were read on a liquid scintillation counter (MicroBeta Jet, PerkinElmer). Data were analyzed by GraphPad Prism 10. GTP Turnover The GTPase GLO assay was performed using a modified GTPase-GLO TM assay from Promega as previously described 24,32 . Briefly, 100 nM of monomeric PN1, reconstituted in nanodiscs with MSP1E3D1 following the protocol described in the sample preparation for cryo-EM, was incubated for 1 hour at RT with 20 μM Iso (Sigma-Aldrich) and a range of concentrations of AP in NH buffer plus 0.2% DMSO, and 20 μM GTP. Simultaneously, a 1 μM stock of heterotrimeric Gs protein was prepared in a buffer consisting of NH buffer plus 0.04% DDM, 200 μM TCEP, 20 mM MgCl 2 , and 20 μM GDP. Equal volumes of PN1 and Gs were then mixed and incubated for 60 minutes. The final reaction consisted of 50 nM of ligand-bound PN1 and 500 nM Gs in NH buffer plus 0.1% DMSO, 0.02% DDM, 100 μM TCEP, 10 μM MgCl 2 , 10 μM GTP, and 10 μM GDP. An equal volume of GTPase-Glo reagent in NH buffer plus 0.02% DDM and 5 μM ATP was then added and incubated for 30 minutes. Detection reagent was subsequently added and incubated for 10 minutes. Luminescence was detected using the MicroBeta counter. For the time course assay, the experimental setup remained identical except 200 nM of AP-mediated dimeric PN1 was included as a condition and the PN1-Gs reactions occurred for 30, 60, 90, and 120 minutes. Ligation of V 2 Rpp to receptor and β-arrestin competition radioligand binding β 2 AR constructs modified with a C-terminal sortase recognition sequence (LPETGHH inserted after residue 365) were expressed in Sf9 cells and purified as described above for monomeric and AP-induced dimeric receptors. Sortase-mediated ligation of synthetic V 2 Rpp to receptor was performed as previously described 30,31 . For ligation reaction, 10 μM purified receptor was incubated in NH buffer supplemented with 0.01% LMNG, 0.001% CHS, and 5 mM CaCl 2 with 50 μM synthetic GGG–V 2 Rpp peptide and 2 μM evolved sortase A pentamutant (eSrtA) 64 . The mixture was incubated overnight at 4 °C. Unreacted receptor and eSrtA (bearing the C-terminal His tag) was removed by binding to nickel-chelated Sepharose resins. Labeled monomeric or dimeric β 2 AR–V 2 Rpp samples were reconstituted into nanodiscs following the protocol described in the sample preparation for cryo-EM. The equilibrium competition radioligand binding assays were performed with β 2 AR–V 2 Rpp in nanodiscs in the presence of 2 nM [ 3 H]-DHA, increasing concentrations of Iso, and 1 μM C-tail-truncated β-arrestin1(382), prepared as previously described 65 . 10 μM AP was added where applicable. After incubation at RT for 1 hour, samples were harvested, and radioactivity was measured as described in the previous section to calculate the K i values of Iso. GRK5 radiometric phosphorylation assays To evaluate the effect of β 2 AR dimerization on receptor phosphorylation, β 2 AR monomers or dimers (1 μM), purified in LMNG micelles or reconstituted in nanodiscs following the protocol described in the sample preparation for cryo-EM, were incubated for 5 min at 30°C with purified C-terminally Strep-tagged GRK5 (50 nM) in a reaction buffer containing 20 mM Tris-HCl, pH 7.4, 5 mM MgCl 2 , 30 mM NaCl, 0.5 mM EDTA, 100 μM [γ 32 P]ATP (1,000 to 2,000 cpm/pmol), and 25 μM BI. The β 2 AR samples in LMNG micelles were additionally supplemented with 20 μM C8-PIP 2 to increase efficiency of β 2 AR phosphorylation in detergent. To evaluate the effect of AP on β 2 AR phosphorylation, purified β 2 AR monomers (1 μM) in LMNG micelles were reconstituted into bicelles with PIP 2 66 and AP concentration was varied from 0 to 24 μM. Reactions were quenched with SDS sample buffer, and samples were separated by SDS-PAGE. Gels were stained with Coomassie blue (Sigma-Aldrich), dried, exposed to autoradiography film, and 32 P-labeled proteins were excised and counted to determine the amount of phosphate transferred. Reaction rates were normalized to phosphorylation of the β 2 AR monomers (β 2 AR monomers and dimers) or to phosphorylation in the absence of AP (AP effect). Sample preparation for fluorescence measurements Site-specific fluorophore labeling of β 2 AR was performed using engineered cysteine mutants on a minimal cysteine background (Δ6), as previously described 24,32 . For smFRET experiments, β 2 ARΔ6 constructs were cloned into the pcDNA-Zeo-tetO vector and transfected into Expi293 cells stably expressing the tetracycline repressor (Thermo Fisher Scientific, A14635). Transfections were carried out using the Expifectamine kit according to the manufacturer’s protocol. Two days post-transfection, receptor expression was induced with 4 μg/mL doxycycline and 5 mM sodium butyrate in the presence of 1 μM alprenolol. Cells were harvested 40 hours after induction and immediately processed for purification. For studies on the β 2 AR dimer in liposomes, single-cysteine mutants were introduced at TM5 (R228C) or H8 (I334C). Homogeneous dimers were expressed in Sf9 cells and purified as described above. Labeling was performed by incubating 10 μM purified receptor with a 5-fold molar excess of a pre-mixed maleimide-conjugated dye pair: DY549P1 (Dyomics) and Alexa Fluor 647 (Thermo Fisher Scientific) at a 1:1.5 ratio. The reaction was incubated for 30 minutes at RT and quenched with 5 mM L-cysteine. Excess dye was removed by SEC (Superdex 200 Increase 10/300) in 20 mM HEPES (pH 7.4), 150 mM NaCl, and 0.01% LMNG/0.001% CHS. Labeled dimers were reconstituted into liposomes consisting of POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE, Avanti) , and cholesterol at a molar ratio of 6:3:1 using an established protocol 67 . For studies on TM6 dynamics, double-cysteine mutants (N148C on TM4 and L266C on TM6) were introduced. Monomeric receptor was expressed in Expi293 cells and purified following the Sf9 purification procedure. For the dimer, Expi293 cells were co-transfected at a 1:1 plasmid ratio with constructs encoding an 8×His-tagged β 2 ARΔ6 (no FLAG tag) and a FLAG-tagged β 2 ARΔ6 carrying the N148C/L266C mutations. Heterodimers were isolated via tandem affinity purification. Clarified lysates were first incubated with nickel-chelated Sepharose resins and washed with buffer containing 20 mM imidazole. Proteins were eluted with 250 mM imidazole, then supplemented with 2 mM CaCl 2 and subjected to anti-FLAG M1 immunoaffinity purification. This two-step procedure enriched for heterodimers containing only one protomer with the double-cysteine mutations. Fluorophore labeling was performed as described above for the single-cysteine mutant dimer sample. Ensemble fluorescence measurements Ensemble FRET experiments were conducted in Fluoromax 4C spectrofluorometer (Horiba Scientific) with excitation and emission slit widths set to 5 nm and 3 nm, respectively. Emission spectra were recorded upon excitation at 532 nm. AP-bound β 2 AR dimers labeled with donor and acceptor fluorophores (I334C H8 sensor) were diluted 1,000-fold in NH buffer plus 0.01% LMNG without AP to a final concentration of 1 nM. Fluorescence spectra were collected at 1, 5, 30, 120 minutes, and 16- and 24-hours post-dilution. All spectra were normalized to donor intensity. To assess the effects of transducer binding, samples were incubated with 100 μM Iso and i) 10 μM Gs (in the presence of apyrase) or ii) 20 μM β-arrestin1(382), together with V 2 Rpp and Fab30, which stabilizes the active V 2 Rpp-bound β-arrestin1 conformation. β-arrestin1(382) and Fab30 were prepared as previously described 65 . Samples were incubated for 1 hour in the dark to allow full equilibration prior to measurement. All experiments were performed in triplicate. smFRET microscopy Flow chambers for smFRET experiments were assembled using mPEG-passivated glass coverslips (VWR), doped with biotin-PEG16 (Laysan Bio), as described previously 24,32 . Prior to use, coverslips were incubated with 1 mg/mL NeutrAvidin (Thermo Fisher Scientific), followed by 10 nM biotinylated anti-FLAG M1 Fab. Labeled β 2 AR samples were diluted to 100–500 pM in NH buffer plus 2 mM CaCl 2 and added to the chambers. After achieving optimal surface density, unbound receptor was washed out using imaging buffer supplemented with 100 μM cyclooctatetraene (Sigma-Aldrich) and an oxygen scavenging system (1% D-glucose, 1 mg/mL glucose oxidase, 0.04 mg/mL catalase). Fluorescence imaging was performed on a custom-built, objective-based TIRF microscope as reported previously 68 . The setup is built on a Zeiss Axiovert S100 TV platform with a 100×, 1.45 NA oil-immersion objective (Zeiss). Donor and acceptor fluorophores were excited with 532 nm and 637 nm lasers (OBIS LS 150 mW and LX 140 mW, Coherent). Emissions were separated by a 652 nm dichroic beamsplitter (Semrock), filtered through 580/60 nm and 731/137 nm bandpass filters, and split using an OptoSplit II beamsplitter (Cairn Research) onto an EMCCD camera (iXon DU897E, Andor). Data acquisition was controlled by μManager via custom BeanShell scripts, and movies were recorded as stacked TIFFs in frame-transfer mode at 100 ms exposure per frame. Laser power was tuned to balance high signal-to-noise with photobleaching timescales of tens of seconds. Each slide typically yielded 10–20 movies per channel. All imaging was performed at RT. Fluorescence traces were analyzed using custom Python scripts. Donor and acceptor channels were aligned using registration images, and individual molecules were identified as local intensity maxima within a 5-pixel neighborhood. Donor-only spots were excluded. For each fluorophore pair, intensities were background-corrected using a local circular region (35-pixel diameter). Donor leakage into the acceptor channel (~7%) was subtracted. Donor excitation was used to monitor emission for 80 seconds, followed by direct acceptor excitation for 1 second to confirm fluorophore identity. Traces were selected for analysis based on the following criteria: (1) signal-to-noise ratio ≥5; (2) single-step acceptor photobleaching prior to donor bleaching; (3) γ factor between 0.5 and 2.5; (4) anticorrelated donor and acceptor intensity fluctuations; and (5) single-step donor bleaching, if present. FRET efficiency ( E ) was calculated as E = I a /(I a +γI d ) , where I a and I d are the background-corrected acceptor and donor intensities, respectively. γ-correction was applied as described previously. For each trace, FRET values were binned into 30 intervals across the range [–0.25, 1.25] and normalized by the total number of data points. Ensemble FRET histograms were generated by averaging the normalized histograms from individual molecules and fit to a two-Gaussian distribution model. Sample preparation for DEER For DEER measurements, β 2 ARΔ6-N148C/L266C was expressed and purified as described above in Sf9 cells. To exchange detergent from 0.1% DDM/0.01% CHS to 0.01% (w/v) LMNG/0.001% CHS, the receptor was extensively washed with a progressive gradient of DDM: LMNG buffer. In parallel, while the receptor was bound to the resin, alprenolol was removed by washing with saturating concentrations of the low-affinity antagonist atenolol. Because of the fast dissociation kinetics of atenolol from the β 2 AR, subsequent washes with ligand-free buffer yielded unliganded β 2 AR for spin labeling. The flag eluted receptor was labeled with the spin label reagent IAP in the presence of 100 μM TCEP in buffer containing 20 mM HEPES, pH 7.4, 150 mM NaCl, and 0.01% LMNG/0.001% CHS. Twenty-fold molar excess of 3-(2-iodoacetamido)-proxyl (IAP) was added to 10 μM β 2 ARΔ6 receptor for 3 hours at RT. After quenching of the reaction with 5 mM final L-cysteine, the receptor was separated from the excess spin label by SEC (Superdex 200 10/300) in SEC buffer (20 mM HEPES, pH 7.4, 150 mM NaCl, and 0.01% LMNG/0.001% CHS) prepared with D 2 O. The sample was concentrated using a 50 kDa concentrator to a concentration > 25 μM. D8-glycerol was added as a cryoprotectant to 25 % (v/v). 13 μL of sample was added to a borosilicate capillary 1.4 mm ID × 1.7 mm OD (VitroCom, Inc) and flash-frozen in liquid nitrogen. DEER spectroscopy DEER experiments were conducted as previously described 32 at Q-band (~33.68 GHz) using a Bruker Elexsys 580 spectrometer equipped with a SpinJet AWG, EN5107D2 resonator, variable-temperature cryogen-free cooling system (ColdEdge Technologies Inc.), and a 300 W TWT amplifier (Applied Systems Engineering Inc.). All measurements were performed at 50 K. Dipolar evolution data were acquired using a dead-time-free 4-pulse DEER sequence with gaussian pulses 69 and with 16-step phase cycling. The experimental parameters used for DEER data collection were: π/2, π obs , and π pump pulse lengths of 40 ns; a frequency offset (Δv) of 90 MHz; d1 = 250 ns; d2 = 5150 ns; shot repetition time = 2000 μs; shots per point = 4; and integration window = 40 ns. The optimal microwave power (i.e., pulse amplitude) for the π/2, π obs , and π pump pulses were determined using transient nutation experiments, where pulse amplitudes were adjusted to maximize the inversion of the Hahn echo 70 . Pump pulses were applied to the maximum intensity of the field swept echo detected absorption spectrum. Observe pulses were applied at a frequency 90 MHz lower than the pump pulses. DEER data were processed using DeerAnalysis 2022 71 , which employs two fitting routines: neural network analysis (DEERNet 72 , Spinach revision 5662) and Tikhonov regularization (DeerLab 0.9.1) 73 . The consensus fit represents the mean of both methods, with reported 95% confidence intervals also incorporating errors from both methods. Time traces were normalized to signal intensity at t = 0, and distance distributions were area normalized. Custom Python scripts were used for plotting the dipolar evolution time traces and the distance distributions. Declarations Data availability The cryo-EM map for β 2 AR_dimer(AP) in nanodisc has been deposited to the Electron Microscopy Data Bank under the accession code EMD-72202. The corresponding atomic model has been deposited in the Protein Data Bank under the accession code 9Q3L. Any additional data reported in this paper are available from the lead contact, Brian K. Kobilka ( [email protected] ), upon request. Acknowledgements We thank R. Qiu of R. S. Lewis’s lab for help with smFRET experiments, E. S. Bruguera for instructions on BLI measurements, M. R. Eckart and J. Tran of Stanford Protein and Nucleic Acid (PAN) Facility for SPR data collection, P. A. N. Reddy and J. M. Salvino for providing AP-7-168, and B. Singal and C. Zhang for support on cryo-EM data collection at the Stanford cryo-EM center (cEMc) and Stanford-SLAC Cryo-EM Center (S2C2), which is supported by the US National Institute of General Medical Sciences (1R24GM154186). We also thank M. Bouvier for providing pcDNA-β-arrestin2-GFP10 and pcDNA3-β 2 AR-RlucII, L. I. Jiang for providing the cAMP intramolecular BRET sensor CAMYEL, G. Milligan for providing pcDNA3-β 1 AR-GFP, and P. S. Chae for providing TTG-T10 detergent. J. Xu is an investigator of SUSTech Institute for Biological Electron Microscopy. This research was supported by National Institutes of Health (NIH) awards R35NS137408 (B.K.K.), R01GM083118 (B.K.K.), P01HL114471 (J.L.B.), R01AI161296 (J.L.B.), R01GM135581 (M.T.L.), S10OD025260 (M.T.L.), and American Heart Association (AHA) Postdoctoral Fellowship 25POST1411512 (J.S.). Author Contributions Conceptualization, J.S., T.N.P., J.X., J.L.B., and B.K.K.; methodology, J.S., T.N.P., J.X., K.E.K., F.D.P., and A.M.G.; investigation: J.S., T.N.P., J.X., K.E.K, F.D.P, A.M.G., and H.W.; data curation and formal analysis, J.S., T.N.P., J.X., K.E.K., F.D.P., and A.M.G.; writing – original draft: J.S., J.X., and, T.N.P.; writing – review & editing, all authors; supervision and funding acquisition, B.K.K., J.L.B., and M.T.L. Competing interests A patent on the reported compounds was submitted by J.L.B. and others in 2022. B.K.K. is a cofounder of and consultant for ConfometRx Inc. All other authors declare no competing interests. References Zhao, D. Y. et al. Cryo-EM structure of the native rhodopsin dimer in nanodiscs. J Biol Chem 294 , 14215-14230, doi:10.1074/jbc.RA119.010089 (2019). Yue, Y. et al. Structural insight into apelin receptor-G protein stoichiometry. Nat Struct Mol Biol 29 , 688-697, doi:10.1038/s41594-022-00797-5 (2022). Chang, H. et al. Structural basis of oligomerization-modulated activation and autoinhibition of orphan receptor GPR3. Cell Rep 44 , 115478, doi:10.1016/j.celrep.2025.115478 (2025). Ippolito, M. et al. Identification of a β-arrestin-biased negative allosteric modulator for the β2-adrenergic receptor. Proc Natl Acad Sci U S A 120 , e2302668120, doi:10.1073/pnas.2302668120 (2023). Hilger, D. The role of structural dynamics in GPCR‐mediated signaling. The FEBS Journal 288 , 2461-2489, doi:10.1111/febs.15841 (2021). Wootten, D., Christopoulos, A., Marti-Solano, M., Babu, M. M. & Sexton, P. M. Mechanisms of signalling and biased agonism in G protein-coupled receptors. Nat Rev Mol Cell Biol , doi:10.1038/s41580-018-0049-3 (2018). Fredriksson, R., Lagerstrom, M. C., Lundin, L. G. & Schioth, H. B. The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol 63 , 1256-1272, doi:10.1124/mol.63.6.1256 (2003). Ellaithy, A., Gonzalez-Maeso, J., Logothetis, D. A. & Levitz, J. Structural and Biophysical Mechanisms of Class C G Protein-Coupled Receptor Function. Trends in biochemical sciences 45 , 1049-1064, doi:10.1016/j.tibs.2020.07.008 (2020). Zeng, F. Y. & Wess, J. Molecular aspects of muscarinic receptor dimerization. Neuropsychopharmacology 23 , S19-S31, doi:Doi 10.1016/S0893-133x(00)00146-9 (2000). Angers, S. et al. Detection of β2 adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). P Natl Acad Sci USA 97 , 3684-3689, doi:DOI 10.1073/pnas.060590697 (2000). Hiller, C., Kuhhorn, J. & Gmeiner, P. Class A G-protein-coupled receptor (GPCR) dimers and bivalent ligands. Journal of medicinal chemistry 56 , 6542-6559, doi:10.1021/jm4004335 (2013). Yue, Y. et al. Structural insights into the regulation of monomeric and dimeric apelin receptor. Nat Commun 16 , 310, doi:10.1038/s41467-024-55555-6 (2025). Huang, B. S., St Onge, C. M., Ma, H. G. & Zhang, Y. Design of bivalent ligands targeting putative GPCR dimers. Drug discovery today 26 , 189-199, doi:10.1016/j.drudis.2020.10.006 (2020). Ullmann, T. et al. Homobivalent Dopamine D(2) Receptor Ligands Modulate the Dynamic Equilibrium of D(2) Monomers and Homo- and Heterodimers. ACS Chem Biol 16 , 371-379, doi:10.1021/acschembio.0c00895 (2021). Rasmussen, S. G. et al. Structure of a nanobody-stabilized active state of the β2 adrenoceptor. Nature 469 , 175-180, doi:10.1038/nature09648 (2011). Liu, X. et al. An allosteric modulator binds to a conformational hub in the β2 adrenergic receptor. Nature chemical biology , doi:10.1038/s41589-020-0549-2 (2020). Cherezov, V. et al. High-Resolution Crystal Structure of an Engineered Human β2-Adrenergic G Protein–Coupled Receptor. Science 318 , 1258-1265, doi:10.1126/science.1150577 (2007). Ji, S.-Y. et al. Dynamic Monomer-Dimer Transition in Ligand-induced Apelin Receptor Activation. bioRxiv , 2025.2002.2019.639188, doi:10.1101/2025.02.19.639188 (2025). Yue, Y. et al. Mechanistic insights into the versatile stoichiometry and biased signaling of the apelin receptor-arrestin complex. Nat Commun 16 , 7403, doi:10.1038/s41467-025-62870-z (2025). Qiu, Z. et al. Vol. 14 RP107185 (eLife, 2025). Chen, G. et al. Mechanism and function of GPR3 regulated by a negative allosteric modulator. Nat Commun 16 , 7988, doi:10.1038/s41467-025-63422-1 (2025). Huang, J., Chen, S., Zhang, J. J. & Huang, X. Y. Crystal structure of oligomeric β1-adrenergic G protein-coupled receptors in ligand-free basal state. Nat Struct Mol Biol 20 , 419-425, doi:10.1038/nsmb.2504 (2013). Fung, J. J. et al. Ligand-regulated oligomerization of beta(2)-adrenoceptors in a model lipid bilayer. EMBO J 28 , 3315-3328, doi:10.1038/emboj.2009.267 (2009). Gregorio, G. G. et al. Single-molecule analysis of ligand efficacy in β2AR-G-protein activation. Nature 547 , 68-73, doi:10.1038/nature22354 (2017). Sungkaworn, T. et al. Single-molecule imaging reveals receptor-G protein interactions at cell surface hot spots. Nature 550 , 543-547, doi:10.1038/nature24264 (2017). Calebiro, D. & Koszegi, Z. The subcellular dynamics of GPCR signaling. Mol Cell Endocrinol 483 , 24-30, doi:10.1016/j.mce.2018.12.020 (2019). Rasmussen, S. G. et al. Crystal structure of the β2 adrenergic receptor–Gs protein complex. Nature 477 , 549-555, doi:10.1038/nature10361 (2011). Lee, Y. et al. Molecular basis of β-arrestin coupling to formoterol-bound β1-adrenoceptor Nature 583 , 862-866, doi:10.1038/s41586-020-2419-1 (2020). Duan, J. et al. GPCR activation and GRK2 assembly by a biased intracellular agonist. Nature 620 , 676-681, doi:10.1038/s41586-023-06395-9 (2023). Staus, D. P. et al. Sortase ligation enables homogeneous GPCR phosphorylation to reveal diversity in β-arrestin coupling. Proc Natl Acad Sci U S A 115 , 3834-3839, doi:10.1073/pnas.1722336115 (2018). Li, Y. et al. Chemical Synthesis of a Full-Length G-Protein-Coupled Receptor β2-Adrenergic Receptor with Defined Modification Patterns at the C-Terminus. J Am Chem Soc 143 , 17566-17576, doi:10.1021/jacs.1c07369 (2021). Casiraghi, M. et al. Structure and dynamics determine G protein coupling specificity at a class A GPCR. Sci Adv 11 , eadq3971, doi:10.1126/sciadv.adq3971 (2025). Zhao, J. et al. Ligand efficacy modulates conformational dynamics of the µ-opioid receptor. Nature 629 , 474-480, doi:10.1038/s41586-024-07295-2 (2024). Liu, J. et al. Biased signaling due to oligomerization of the G protein-coupled platelet-activating factor receptor. Nat Commun 13 , 6365, doi:10.1038/s41467-022-34056-4 (2022). Paradis, J. S. et al. Computationally designed GPCR quaternary structures bias signaling pathway activation. Nat Commun 13 , 6826, doi:10.1038/s41467-022-34382-7 (2022). Zhang, Y. et al. Higher-order transient membrane protein structures. Proc Natl Acad Sci U S A 122 , e2421275121, doi:10.1073/pnas.2421275121 (2025). Zhang, Y. & MacKinnon, R. Higher-order transient structures and the principle of dynamic connectivity in membrane signaling. Proc Natl Acad Sci U S A 122 , e2421280121, doi:10.1073/pnas.2421280121 (2025). Mandala, V. S., Fu, Z. & MacKinnon, R. Molecular contacts in self-assembling clusters of membrane proteins. Proc Natl Acad Sci U S A 122 , e2507112122, doi:10.1073/pnas.2507112122 (2025). Schreiber, S. L. The Rise of Molecular Glues. Cell 184 , 3-9, doi:10.1016/j.cell.2020.12.020 (2021). Schreiber, S. L. Molecular glues and bifunctional compounds: Therapeutic modalities based on induced proximity. Cell Chem Biol 31 , 1050-1063, doi:10.1016/j.chembiol.2024.05.004 (2024). Roth, B. L. & Krumm, B. E. Molecular glues as potential GPCR therapeutics. Biochem Pharmacol 228 , 116402, doi:10.1016/j.bcp.2024.116402 (2024). Krumm, B. E. et al. Neurotensin Receptor Allosterism Revealed in Complex with a Biased Allosteric Modulator. Biochemistry 62 , 1233-1248, doi:10.1021/acs.biochem.3c00029 (2023). Zhao, L. H. et al. Conserved class B GPCR activation by a biased intracellular agonist. Nature 621 , 635-641, doi:10.1038/s41586-023-06467-w (2023). Sun, D. et al. Molecular mechanism of the arrestin-biased agonism of neurotensin receptor 1 by an intracellular allosteric modulator. Cell Res 35 , 284-295, doi:10.1038/s41422-025-01095-7 (2025). He, G. et al. A GPCR-G protein-β-arrestin megacomplex induced by an allosteric modulator. bioRxiv , 2025.2003.2024.645131, doi:10.1101/2025.03.24.645131 (2025). Li, J. et al. Activation of human STING by a molecular glue-like compound. Nat Chem Biol 20 , 365-372, doi:10.1038/s41589-023-01434-y (2024). Strohman, M. J. et al. Local membrane charge regulates β2 adrenergic receptor coupling to Gi3. Nat Commun 10 , 2234, doi:10.1038/s41467-019-10108-0 (2019). Hilger, D. et al. Structural insights into differences in G protein activation by family A and family B GPCRs. Science 369 , doi:10.1126/science.aba3373 (2020). Krishna Kumar, K. et al. Negative allosteric modulation of the glucagon receptor by RAMP2. Cell 186 , 1465-1477 e1418, doi:10.1016/j.cell.2023.02.028 (2023). Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods 14 , 290-296, doi:10.1038/nmeth.4169 (2017). Punjani, A., Zhang, H. & Fleet, D. J. Non-uniform refinement: adaptive regularization improves single-particle cryo-EM reconstruction. Nat Methods 17 , 1214-1221, doi:10.1038/s41592-020-00990-8 (2020). Scheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. J Struct Biol 180 , 519-530, doi:10.1016/j.jsb.2012.09.006 (2012). Kimanius, D., Dong, L., Sharov, G., Nakane, T. & Scheres, S. H. W. New tools for automated cryo-EM single-particle analysis in RELION-4.0. Biochem J 478 , 4169-4185, doi:10.1042/BCJ20210708 (2021). Zivanov, J., Nakane, T. & Scheres, S. H. W. A Bayesian approach to beam-induced motion correction in cryo-EM single-particle analysis. IUCrJ 6 , 5-17, doi:10.1107/S205225251801463X (2019). Pettersen, E. F. et al. UCSF Chimera--a visualization system for exploratory research and analysis. Journal of computational chemistry 25 , 1605-1612, doi:10.1002/jcc.20084 (2004). Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta crystallographica. Section D, Biological crystallography 66 , 486-501, doi:10.1107/S0907444910007493 (2010). Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta crystallographica. Section D, Biological crystallography 66 , 213-221, doi:10.1107/S0907444909052925 (2010). Afonine, P. V. et al. Real-space refinement in PHENIX for cryo-EM and crystallography. Acta Crystallogr D Struct Biol 74 , 531-544, doi:10.1107/S2059798318006551 (2018). Barad, B. A. et al. EMRinger: side chain-directed model and map validation for 3D cryo-electron microscopy. Nature methods 12 , 943-946, doi:10.1038/nmeth.3541 (2015). Pettersen, E. F. et al. UCSF ChimeraX: Structure visualization for researchers, educators, and developers. Protein Sci 30 , 70-82, doi:10.1002/pro.3943 (2021). Sonn-Segev, A. et al. Quantifying the heterogeneity of macromolecular machines by mass photometry. Nat Commun 11 , 1772, doi:10.1038/s41467-020-15642-w (2020). Peitsch, C. F., Beckmann, S. & Zuber, B. iMEM: Isolation of Plasma Membrane for Cryoelectron Microscopy. Structure 24 , 2198-2206, doi:10.1016/j.str.2016.09.016 (2016). Jiang, L. I. et al. Use of a cAMP BRET sensor to characterize a novel regulation of cAMP by the sphingosine 1-phosphate/G13 pathway. J Biol Chem 282 , 10576-10584, doi:10.1074/jbc.M609695200 (2007). Chen, I., Dorr, B. M. & Liu, D. R. A general strategy for the evolution of bond-forming enzymes using yeast display. Proc Natl Acad Sci U S A 108 , 11399-11404, doi:10.1073/pnas.1101046108 (2011). Janetzko, J. et al. Membrane phosphoinositides regulate GPCR-β-arrestin complex assembly and dynamics Cell 185 , 4560-4573 e4519, doi:10.1016/j.cell.2022.10.018 (2022). Komolov, K. E. et al. Structural and Functional Analysis of a β2-Adrenergic Receptor Complex with GRK5. Cell 169 , 407-421 e416, doi:10.1016/j.cell.2017.03.047 (2017). Goddard, A. D., Dijkman, P. M., Adamson, R. J., dos Reis, R. I. & Watts, A. Reconstitution of membrane proteins: a GPCR as an example. Methods Enzymol 556 , 405-424, doi:10.1016/bs.mie.2015.01.004 (2015). van Dorp, S. et al. Conformational dynamics of auto-inhibition in the ER calcium sensor STIM1. Elife 10 , doi:10.7554/eLife.66194 (2021). Teucher, M. & Bordignon, E. Improved signal fidelity in 4-pulse DEER with Gaussian pulses. J Magn Reson 296 , 103-111, doi:10.1016/j.jmr.2018.09.003 (2018). Teucher, M. et al. Strategies to identify and suppress crosstalk signals in double electron-electron resonance (DEER) experiments with gadolinium(III) and nitroxide spin-labeled compounds. Magn Reson (Gott) 1 , 285-299, doi:10.5194/mr-1-285-2020 (2020). Jeschke, G. et al. DeerAnalysis2006 - a comprehensive software package for analyzing pulsed ELDOR data. Appl Magn Reson 30 , 473-498, doi:10.1007/Bf03166213 (2006). Worswick, S. G., Spencer, J. A., Jeschke, G. & Kuprov, I. Deep neural network processing of DEER data. Sci Adv 4 , eaat5218, doi:10.1126/sciadv.aat5218 (2018). Fabregas Ibanez, L., Jeschke, G. & Stoll, S. DeerLab: a comprehensive software package for analyzing dipolar electron paramagnetic resonance spectroscopy data. Magn Reson (Gott) 1 , 209-224, doi:10.5194/mr-1-209-2020 (2020). Additional Declarations Yes there is potential Competing Interest. A patent on the reported compounds was submitted by J.L.B. and others in 2022. B.K.K. is a cofounder of and consultant for ConfometRx Inc. All other authors declare no competing interests. Supplementary Files b2APsup103125.docx Supplementary information Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8001844","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Biological Sciences - Article","associatedPublications":[],"authors":[{"id":550555865,"identity":"4d607343-40b4-4417-9d50-4d7533e5ae87","order_by":0,"name":"Brian Kobilka","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5UlEQVRIiWNgGAWjYBACCQYehgMPDjDIgXkVDAyMDQwMBoS1JBxgMGZgA/LOEKuFAaglsYFoLZLtvQcPJJyxS58/v/nZgwMV92Qb2Ju3SeDTIs1zLuFAwo3k3A3H2MwNDpwpNm7gOVaGV4ucRI7BgYQPzLkb2BjMpD+2JSQ2SOSYEaOlPl2+jf2bxEGQFvk3+LVIg7XcOJzAcIzHDKJFgge/FskekF/OHDfccCynTOLAmQTjNp60Ygt8WiSO9x7+8OFYtbx88/FtEgcqEmT72Q9vvIFPCyZgI035KBgFo2AUjAJsAADfoFHHvDdAFgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0001-5958-3990","institution":"Stanford University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Brian","middleName":"","lastName":"Kobilka","suffix":""},{"id":550555866,"identity":"b2f5827e-86e2-4981-a132-4a34177059de","order_by":1,"name":"Jiemin Shen","email":"","orcid":"https://orcid.org/0000-0002-3977-0681","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Jiemin","middleName":"","lastName":"Shen","suffix":""},{"id":550555867,"identity":"24ef2482-3791-47c1-a9f6-c0b38725e2bb","order_by":2,"name":"Teja Peddada","email":"","orcid":"","institution":"Stanford University","correspondingAuthor":false,"prefix":"","firstName":"Teja","middleName":"","lastName":"Peddada","suffix":""},{"id":550555868,"identity":"ecd625c9-82f4-44e9-8610-3b0175a5b1e4","order_by":3,"name":"Konstantin Komolov","email":"","orcid":"","institution":"Thomas Jefferson University Sidney Kimmel Medical College","correspondingAuthor":false,"prefix":"","firstName":"Konstantin","middleName":"","lastName":"Komolov","suffix":""},{"id":550555869,"identity":"faa5812e-a894-4f33-92e2-bb348173395e","order_by":4,"name":"Francesco De Pascali","email":"","orcid":"","institution":"Thomas Jefferson University Sidney Kimmel Medical College","correspondingAuthor":false,"prefix":"","firstName":"Francesco","middleName":"","lastName":"De Pascali","suffix":""},{"id":550555870,"identity":"87f5df70-6827-4f7c-9c1e-74108c3ff3e0","order_by":5,"name":"Alexander Garces","email":"","orcid":"","institution":"Medical College of Wisconsin","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Garces","suffix":""},{"id":550555871,"identity":"105a7d01-70b3-4111-ba3d-da81979f92bf","order_by":6,"name":"Haoqing Wang","email":"","orcid":"","institution":"Stanford University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Haoqing","middleName":"","lastName":"Wang","suffix":""},{"id":550555872,"identity":"e139f83b-9085-4d4f-b0b4-76d13c7b2770","order_by":7,"name":"Michael Lerch","email":"","orcid":"","institution":"Medical College of Wisconsin","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Lerch","suffix":""},{"id":550555873,"identity":"467b9ae7-98cf-4a60-adf8-1fe3f7db8471","order_by":8,"name":"Jeffrey Benovic","email":"","orcid":"","institution":"Thomas Jefferson University","correspondingAuthor":false,"prefix":"","firstName":"Jeffrey","middleName":"","lastName":"Benovic","suffix":""},{"id":550555874,"identity":"04b593c6-486a-4f26-bea9-09096a7679ca","order_by":9,"name":"Jun Xu","email":"","orcid":"https://orcid.org/0000-0002-1600-583X","institution":"South University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2025-10-31 22:20:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8001844/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8001844/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96916235,"identity":"97b0ac2d-2e69-47eb-875a-fffcf9a14935","added_by":"auto","created_at":"2025-11-27 14:08:17","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2490843,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCryo-EM structures reveal that AP, a β-arrestin–biased NAM, mediates β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR dimerization.\u003c/strong\u003e \u003cstrong\u003eA\u003c/strong\u003e, Chemical structure of AP-7-168 (AP). Atom numbering on the quinazoline ring is indicated. \u003cstrong\u003eB\u003c/strong\u003e, Dose-dependent effects of AP on β-arrestin recruitment (red) and cAMP production (black) in HEK cells stimulated with 1 µM isoproterenol (Iso). AP exhibits ~930-fold lower IC\u003csub\u003e50\u003c/sub\u003e for β-arrestin recruitment. Data are normalized to Iso-alone controls and presented as mean values with error bars representing the standard error of the mean (SEM), n = 3. \u003cstrong\u003eC\u003c/strong\u003e, Size-exclusion chromatography (SEC) profiles of β\u003csub\u003e2\u003c/sub\u003eAR treated with AP on the cell membrane (red) versus detergent-solubilized β\u003csub\u003e2\u003c/sub\u003eAR with subsequent AP addition (black). \u003cstrong\u003eD\u003c/strong\u003e, Cryo-EM map of AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer reconstituted in lipid nanodiscs. Densities for two protomers are colored in green and orange, densities for AP are colored in magenta and densities for the lipids are colored in gray. \u003cstrong\u003eE\u003c/strong\u003e, Overall structure of the AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/02e81564f9b9cab025c9f104.png"},{"id":96791769,"identity":"4f6d5c97-90e2-4c33-9f3e-94d601e157fc","added_by":"auto","created_at":"2025-11-26 06:58:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3022600,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStructural analysis of AP binding site and β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR dimer interface. A–C\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eEnlarged AP binding pocket from side view (\u003cstrong\u003eA\u003c/strong\u003e), extracellular view (\u003cstrong\u003eB\u003c/strong\u003e) and intracellular view (\u003cstrong\u003eC\u003c/strong\u003e). The two AP molecules are shown in magenta and cyan respectively. Residues within 4 Å of AP were shown in sticks. Red dashed lines represent hydrogen bond interactions. \u003cstrong\u003eD\u003c/strong\u003e, Chemical structure of the unbiased NAM, AS408. \u003cstrong\u003eE\u003c/strong\u003e, Structural comparison of one protomer from the AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer (green) with monomeric β\u003csub\u003e2\u003c/sub\u003eAR bound to AS408 (purple, PDB ID: 6OBA). Enlarged view of the ligand binding pocket is shown on the left. C4 atoms of the quinazoline ring are labeled to highlight the 180° flip between AP and AS408. \u003cstrong\u003eF–H\u003c/strong\u003e,\u003cstrong\u003e \u003c/strong\u003eDetailed interactions between two protomers in the dimer interface from different views: side view (\u003cstrong\u003eF\u003c/strong\u003e), extracellular view (\u003cstrong\u003eG\u003c/strong\u003e), and intracellular view (\u003cstrong\u003eH\u003c/strong\u003e). Residues that are within 4 Å between the two protomers are shown as sticks. Red dashed lines represent hydrogen bond interactions. The lipid molecules in the dimer interface are shown in gray sticks.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/932c26dfceb33c76ab2bdc6d.png"},{"id":96791768,"identity":"3ad746b0-3d96-4fe1-a561-c0f79186ded1","added_by":"auto","created_at":"2025-11-26 06:58:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1455621,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAP promotes β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR dimerization in continuous membranes and in cells.\u003c/strong\u003e \u003cstrong\u003eA\u003c/strong\u003e, smFRET assay showing the conformation of AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer in proteoliposomes. Shaded parallelograms indicate the geometry of the dimer viewed from the intracellular side. Dye labeling sites (red) on TM5 (top) and H8 (bottom) and their respective inter-protomer distances (dotted lines) are indicated on the structure, with corresponding FRET population histograms shown on the right. Inset: schematic of β\u003csub\u003e2\u003c/sub\u003eAR dimer in a liposome with inter-protomer FRET; purple rectangles represent AP molecules. \u003cstrong\u003eB\u003c/strong\u003e, Cumulative probability plots of β\u003csub\u003e2\u003c/sub\u003eAR cluster size on the cell membrane. Inset: frequency distributions of cluster size. Blue, vehicle; red, treatment with 10 µM AP. ***, p \u0026lt; 0.001. \u003cstrong\u003eC\u003c/strong\u003e, Schematic of the cell-based BRET assay for receptor dimerization. \u003cstrong\u003eD\u003c/strong\u003e, Dose-response curve of AP-promoted β\u003csub\u003e2\u003c/sub\u003eAR dimerization in HEK cells. \u003cstrong\u003eE\u003c/strong\u003e, The V129\u003csup\u003e3.48\u003c/sup\u003eL mutation, which introduces steric clashes with AP, abolishes β\u003csub\u003e2\u003c/sub\u003eAR dimerization.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/6b905ab3f1db028c79e25fd6.png"},{"id":96791767,"identity":"c4c72f3e-88f4-47dd-b40b-735c19e6d925","added_by":"auto","created_at":"2025-11-26 06:58:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2074211,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAP-bound β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR dimer is a species with biased activity.\u003c/strong\u003e \u003cstrong\u003eA\u003c/strong\u003e, Overall structure comparison of the AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer (one protomer, green) with the carazolol-bound β\u003csub\u003e2\u003c/sub\u003eAR crystal structure (2RH1, gray) reveals an inactive state conformation.\u003cstrong\u003e B\u003c/strong\u003e, Comparison of the ICL2 conformation between the AP-bound β\u003csub\u003e2\u003c/sub\u003eAR (green) and the carazolol-bound inactive β\u003csub\u003e2\u003c/sub\u003eAR (gray). Yellow dashed lines represent hydrogen bond interactions. \u003cstrong\u003eC\u003c/strong\u003e, Structural models of β\u003csub\u003e2\u003c/sub\u003eAR dimer (green and orange) aligned with known transducer-bound GPCR structures: Gs (wheat), β-arrestin (lavender), and GRK (dark blue). Models were generated by aligning monomeric effector-bound structures to one protomer of the AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer. Original structures: β\u003csub\u003e2\u003c/sub\u003eAR–Gs (PDB: 3SN6), β\u003csub\u003e1\u003c/sub\u003eAR–β-arrestin1 (PDB: 6TKO), NTR1–GRK2 (PDB: 8JPB). \u003cstrong\u003eD\u003c/strong\u003e, Steady-state binding curves of GDP-bound Gs to β\u003csub\u003e2\u003c/sub\u003eAR measured by BLI. \u003cstrong\u003eE\u003c/strong\u003e, Dose-dependent negative allosteric effect of Gs binding on antagonist binding to the orthosteric site of β\u003csub\u003e2\u003c/sub\u003eAR. Measurements were performed on β\u003csub\u003e2\u003c/sub\u003eAR monomers and dimers reconstituted in lipid nanodiscs. \u003cstrong\u003eF\u003c/strong\u003e, GTP turnover assays show the different NAM effects of AP on the GEF activity of monomeric and dimeric β\u003csub\u003e2\u003c/sub\u003eAR toward Gs. Reactions were quenched at indicated time points and residual amounts of GTP were quantified. \u003cstrong\u003eG\u003c/strong\u003e, Schematic of monomeric or dimeric β\u003csub\u003e2\u003c/sub\u003eAR–V\u003csub\u003e2\u003c/sub\u003eRpp in nanodiscs with AP, showing β-arrestin tail engagement but no core interactions. Right: Competition binding of [³H]-DHA with agonist Iso in monomer (blue) or dimer (red) in the presence (dash line) or absence (solid line) of β-arrestin1. 10 μM AP was included. \u003cstrong\u003eH\u003c/strong\u003e, Radiometric kinase assay showing phosphorylation of b\u003csub\u003e2\u003c/sub\u003eAR monomers and dimers solubilized in LMNG detergent or reconstituted into nanodiscs. \u003cstrong\u003eI\u003c/strong\u003e, Dose-dependent effect of AP on b\u003csub\u003e2\u003c/sub\u003eAR phosphorylation. b\u003csub\u003e2\u003c/sub\u003eAR monomers were solubilized in MNG detergent or reconstituted into bicelles or nanodiscs and treated with AP for 30 min before the reaction.\u003cem\u003e \u003c/em\u003eTop: \u003csup\u003e32\u003c/sup\u003eP autoradiography of b\u003csub\u003e2\u003c/sub\u003eAR. Data are shown as mean with SEM; n = 3 in \u003cstrong\u003eE\u003c/strong\u003e, n = 6 in \u003cstrong\u003eF\u003c/strong\u003e, n = 3 in \u003cstrong\u003eG\u003c/strong\u003e, and n = 5 in \u003cstrong\u003eH\u003c/strong\u003e and \u003cstrong\u003eI\u003c/strong\u003e.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/d79af0c98a68d02cfdf7a6c1.png"},{"id":96916701,"identity":"1d8739e8-57d6-485e-942b-833a5bcfce93","added_by":"auto","created_at":"2025-11-27 14:08:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1522689,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAP biases TM6 conformational dynamics of β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR in both dimer and monomer.\u003c/strong\u003e \u003cstrong\u003eA\u003c/strong\u003e, Schematic of AP-bound β\u003csub\u003e2\u003c/sub\u003eAR dimer labeled on TM4 and TM6 with fluorophores (red and blue spheres) for smFRET analysis. \u003cstrong\u003eB\u003c/strong\u003e, Representative fluorescence traces from donor (green) and acceptor (red) channels (top), with corresponding FRET efficiency values (black line) and fitted state model (red line) shown below. Dotted lines indicate distinct FRET states. \u003cstrong\u003eC–D\u003c/strong\u003e, FRET population histograms of dimeric (\u003cstrong\u003eC\u003c/strong\u003e) and monomeric (\u003cstrong\u003eD\u003c/strong\u003e) β\u003csub\u003e2\u003c/sub\u003eAR showing effects of ligands and Gs on TM6 conformational dynamics. Solid colored lines are two-Gaussian model fits. \u003cstrong\u003eE\u003c/strong\u003e, Schematic of β\u003csub\u003e2\u003c/sub\u003eAR with spin labels on TM4 and TM6 for DEER studies. \u003cstrong\u003eF–H\u003c/strong\u003e, Distance distributions from DEER measurements of monomeric β\u003csub\u003e2\u003c/sub\u003eAR with spin labels on TM4 and TM6 under different ligand and Gs conditions.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/fd7275b2c7d9b968c83185dd.png"},{"id":96791772,"identity":"cb4d058d-b130-404a-8159-dee334bdbdaa","added_by":"auto","created_at":"2025-11-26 06:58:41","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":844418,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProposed model for AP biasing β\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003eAR signaling. \u003c/strong\u003eAP (purple) promotes β\u003csub\u003e2\u003c/sub\u003eAR dimerization, which reduces receptor desensitization in human ASM cells and mouse airway models. Simplified free-energy landscapes illustrate how AP alters activation energy barriers and state equilibria of the apo (red) and agonist/transducer-bound (blue) conditions under three scenarios: β\u003csub\u003e2\u003c/sub\u003eAR without AP (left), AP-bound monomer (middle), and AP-bound dimer (right).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/d8f4adaccb358aa1604028bb.png"},{"id":96922836,"identity":"1d2e250d-a84d-45d2-875c-28a9529f4212","added_by":"auto","created_at":"2025-11-27 14:20:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":15599023,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/67d51647-afef-4658-a1d5-bbea847e0927.pdf"},{"id":96791770,"identity":"ed6337ca-d67c-4782-8c0c-e22d05b7f897","added_by":"auto","created_at":"2025-11-26 06:58:41","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3789902,"visible":true,"origin":"","legend":"Supplementary information","description":"","filename":"b2APsup103125.docx","url":"https://assets-eu.researchsquare.com/files/rs-8001844/v1/f529e2aea9e251cdbd90a4cf.docx"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nA patent on the reported compounds was submitted by J.L.B. and others in 2022. B.K.K. is a cofounder of and consultant for ConfometRx Inc. All other authors declare no competing interests.","formattedTitle":"\u003cp\u003eA biased allosteric modulator functions as a molecular glue to induce β\u003csub\u003e2\u003c/sub\u003eAR dimerization\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eG-protein coupled receptors (GPCRs) represent the largest superfamily of transmembrane (TM) proteins, exhibiting complex signaling behaviors. Depending on the ligand, GPCRs can interact with G-proteins or bind to effector proteins such as arrestins and kinases, mediating G-protein-independent pathways and giving rise to diverse pharmacological and physiological responses\u003csup\u003e5,6\u003c/sup\u003e. Over 800 human GPCRs have been identified and systematically classified into five families based on their sequence features and structural characteristics\u003csup\u003e7\u003c/sup\u003e: i) family A Rhodopsin-like, ii) family B1 Secretin-like, iii) family B2 Adhesion-like, iv) family C Glutamate-like, and v) family F Frizzled-like receptors. Although family C receptors are known to be obligate dimers\u003csup\u003e8\u003c/sup\u003e, mediated primarily by their large Venus flytrap extracellular domains (ECDs), dimerization of other families is not well described.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOver the past few decades, many studies suggest that family A receptors also possess the ability to form dimers\u003csup\u003e9-11\u003c/sup\u003e. Recent cryo-EM studies on rhodopsin\u003csup\u003e1\u003c/sup\u003e, apelin receptor (APJR)\u003csup\u003e2\u003c/sup\u003e, and GPR3\u003csup\u003e3\u003c/sup\u003e provide direct evidence for the dimerization of these family A GPCRs. However, unlike family C receptors, family A receptors do not have a significant ECD and therefore rely on interactions between their TM domains to undergo dimerization\u003csup\u003e12\u003c/sup\u003e. Despite the various studies on family A dimers, their physiological relevance is still debated. The development of ligands that stabilize family A receptor dimers remains elusive\u003csup\u003e13,14\u003c/sup\u003e, and it is difficult to reliably reproduce these dimers and to comprehensively study their functional, biochemical, and biophysical properties.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe previously identified a class of negative allosteric modulators (NAMs) selective for the \u0026beta;\u003csub\u003e2\u003c/sub\u003e-adrenergic receptor (\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR), a family A GPCR, that exhibited biased inhibition of \u0026beta;-arrestin recruitment with minimal impact on stimulatory G protein (Gs)-mediated cAMP production in cell-based assays\u003csup\u003e4\u003c/sup\u003e. Through comprehensive structure\u0026ndash;activity relationship (SAR) studies, we developed AP-7-168 (hereafter referred to as AP), a compound that has high potency and biased effects. Activation of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR upon coupling with Gs triggers the downstream cAMP signaling pathway, leading to many physiological effects including the relaxation of airway smooth muscle, which is the core mechanism underlying the use of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR agonists in treating airway diseases such as asthma and chronic obstructive pulmonary disease. However, upon prolonged agonist stimulation, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR undergoes desensitization via arrestin-mediated inhibition of Gs coupling and receptor endocytosis, resulting in a diminished bronchodilatory effect. Our \u003cem\u003eex vivo\u003c/em\u003e tissue assays demonstrated that AP effectively maintained agonist-induced \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR activation and the subsequent bronchodilation while significantly delaying receptor desensitization\u003csup\u003e4\u003c/sup\u003e, highlighting its potential clinical value. Despite these promising findings, the molecular mechanism by which AP modulates \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR function remains unclear, posing a significant obstacle to the further development and optimization of AP as a therapeutic agent.\u003c/p\u003e\n\u003cp\u003eIn this study, using cryogenic electron microscopy (cryo-EM) combined with various biochemical and biophysical approaches, we demonstrate that AP functions as a molecular glue by stabilizing a homodimer complex of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, which in turn alters the downstream signaling behaviors. These findings reveal a novel allosteric modulation mechanism in \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR that may have broad implications in drug development for family A GPCRs.\u0026nbsp;\u003c/p\u003e\n"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eStructure determination of AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs depicted in Fig. 1A, AP and its parent compound difluorophenyl quinazoline (DFPQ)\u003csup\u003e4\u003c/sup\u003e have the pharmacophore characterized by a quinazoline ring with 2- and 4-amino substitutions. AP features a 3,4-difluorophenyl group at the 2-amino position, a cyclohexane substituent at the 4-amino position, and a bromine atom at the 6-position of the quinazoline ring. AP exhibited a remarkable biased NAM effect in \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR-expressing cells, being approximately 1,000-fold more effective at inhibiting \u0026beta;-arrestin2 recruitment versus cyclic adenosine monophosphate (cAMP) production when stimulated with the \u0026beta;-agonist isoproterenol (Iso) (Fig. 1B). To elucidate the molecular mechanism underlying this biased inhibition, we sought to determine the structure of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR bound to AP.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe initially prepared the cryo-EM sample using detergent-solubilized and size-exclusion chromatography (SEC)-purified \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR incubated with AP in the presence of carazolol, a \u0026beta;-antagonist. Given the small size (\u0026lt; 50 kDa) of a monomeric receptor, we included nanobody 60 (Nb60), which is specific to an inactive conformation of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003csup\u003e15\u003c/sup\u003e, for steering particle alignment in cryo-EM data processing. Unexpectedly, the extracted particles from the dataset revealed two predominant 2D class averages, with approximately 70% of monomers and 30% of dimers (Supplementary Fig. 1A). Subsequent 3D classification and reconstruction showed the dimer consists of two parallel protomers with a symmetric TM3, TM4, and TM5 interface (Supplementary Fig. 1). Owing to the larger size and the C2 symmetry of the dimer, we obtained a cryo-EM map at a resolution of 2.9 \u0026Aring; (Supplementary Fig. 1A). The map revealed two distinct AP densities located at the dimer interface near the center of the membrane (Fig. 1D and Supplementary Fig. 1A). This finding suggests that AP may promote \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe also managed to reconstruct a 4.3 \u0026Aring; resolution map for monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR bound to Nb60 in the same dataset (Supplementary Fig. 1A). Despite the modest resolution, a distinct non-protein density, comparably strong to that of TM helices, was observed near the AP binding site identified in the dimer structure (Supplementary Fig. 2A,B). This indicates that AP can bind to monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR at a similar position as in the dimer. Surface plasmon resonance (SPR) measurements further confirmed the direct binding of AP to the monomeric receptor with a \u003cem\u003eK\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e of approximately 30 \u0026mu;M (Supplementary Fig. 2C).\u003c/p\u003e\n\u003cp\u003eEncouraged by the initial dimeric structure in detergent, we sought to produce a homogeneous dimer sample and solve its structure in a native-like environment. Although incubation of AP with purified monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in detergent did not result in efficient dimerization as assessed by SEC (Fig. 1C), we achieved almost complete dimerization when \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was pretreated with AP in a membrane environment before detergent extraction (Fig. 1C and Methods). This highlights the critical role of a continuous fluid membrane in facilitating TM contacts mediated by AP and maintained in detergent after solubilization. We reconstituted the dimer into lipid nanodiscs and solved its structure in complex with an agonist, BI-167107 (BI), without nanobodies (Supplementary Fig. 1B). Cryo-EM analysis confirmed the presence of predominantly dimeric species, with no monomer detected in 2D classification (Supplementary Fig. 1B). The final cryo-EM map achieved a resolution of 2.5 \u0026Aring;, with a local resolution of ~2.2 \u0026Aring; near the AP binding site (Fig. 1D,E and Supplementary Fig. 1B).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStructural basis of AP binding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe high-resolution map enabled precise modeling of two AP molecules in the dimer, revealing a novel ligand binding mode (Fig. 1E, Supplementary Fig. 3, and Supplementary Table 1). The two AP molecules are packed against each other through \u0026pi;\u0026ndash;\u0026pi; and van der Waals interactions and fit in an allosteric pocket formed by TM3, TM4, and TM5 of two \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR protomers (Fig. 1E,2A\u0026ndash;C). The two AP molecules also form extensive hydrophobic and van der Waals interactions with a series of non-polar residues in TM3 (C125\u003csup\u003e3.44\u003c/sup\u003e, V126\u003csup\u003e3.45\u003c/sup\u003e, V129\u003csup\u003e3.48\u003c/sup\u003e), TM4 (M156\u003csup\u003e4.48\u003c/sup\u003e, M157\u003csup\u003e4.49\u003c/sup\u003e), and TM5 (V206\u003csup\u003e5.45\u003c/sup\u003e, V210\u003csup\u003e5.49\u003c/sup\u003e, I214\u003csup\u003e5.53\u003c/sup\u003e) (Fig. 2A\u0026ndash;C). In addition, the 2-amino group in AP forms hydrogen bond interactions with E122\u003csup\u003e3.41\u003c/sup\u003e in TM3 (Fig. 2A). These specific interactions in our structural model rationalize previously reported SAR and mutagenesis data\u003csup\u003e4\u003c/sup\u003e. For example, removing the bromine atom reduced AP\u0026rsquo;s activity, likely because of the loss of interactions with V206\u003csup\u003e5.45\u0026nbsp;\u003c/sup\u003e(Fig. 2A). The difluorophenyl substitution forms optimal contacts with V126\u003csup\u003e3.45\u003c/sup\u003e and V129\u003csup\u003e3.48\u003c/sup\u003e (Fig. 2C), whereas larger halogen atoms reduce ligand potency\u003csup\u003e4\u003c/sup\u003e. Similarly, V129\u003csup\u003e3.48\u003c/sup\u003eL or swapping TM3 with \u0026beta;\u003csub\u003e1\u003c/sub\u003e-adrenergic receptor (\u0026beta;\u003csub\u003e1\u003c/sub\u003eAR), where isoleucine replaces the valine residue, introduces steric clashes with the difluorophenyl group, therefore significantly reducing the potency of AP. Moreover, the E122\u003csup\u003e3.41\u003c/sup\u003eW mutation also leads to almost complete loss of response to AP\u003csup\u003e4\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInterestingly, the binding site of AP on \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR closely overlaps with a previously identified binding site for an unbiased NAM, AS408\u003csup\u003e16\u003c/sup\u003e (Fig. 2D,E). Structural alignment of one of the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR protomers in the AP-bound dimer with the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR-AS408 complex (PDB ID: 6OBA) reveals a highly similar overall conformation, with an overall C\u0026alpha; root mean square deviation (RMSD) of only 0.53 \u0026Aring; (Fig. 2E). Both AP and AS408 possess a bromide substitution at the C6 position of the quinazoline ring (Fig. 2D,E), and this bromide is positioned in close proximity to V206\u003csup\u003e5.46\u0026nbsp;\u003c/sup\u003ein both structures (Fig. 2E). However, despite the similarities in the quinazoline ring and the bromide position, there are notable differences in their binding orientations. The quinazoline rings of AP and AS408 are nearly parallel, but they are flipped 180 degrees relative to each other (Fig. 2E). This difference in orientation is likely due to the absence of a bulky substitution at the 4-amino position in AS408, in contrast to AP. Consequently, in the AS408-bound structure, the 4-amino group of AS408 forms a hydrogen bond with E122\u003csup\u003e3.41\u003c/sup\u003e, while in the AP-bound structure, it is the 2-amino group of AP that interacts with E122\u003csup\u003e3.41\u003c/sup\u003e (Fig. 2E).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eE122\u003csup\u003e3.41\u003c/sup\u003e has been proposed to be a critical conformational hub in \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003csup\u003e16\u003c/sup\u003e. It serves as a key link that connects the extracellular orthosteric ligand-binding pocket to the intracellular transducer-coupling interface. The shift in hydrogen bonding partners, from the 4-amino group of AS408 to the 2-amino group of AP, stabilizes E122\u003csup\u003e3.41\u003c/sup\u003e in a distinct conformation (Fig. 2E). This unique conformation of E122\u003csup\u003e3.41\u003c/sup\u003e may have significant implications for modulating the interactions of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR with different intracellular transducers, such as Gs and \u0026beta;-arrestins. Moreover, the cyclohexane group in AP, which is not present in AS408, plays a crucial role in ligand packing and hydrophobic contacts with the adjacent \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR protomer. These differences may explain why AS408 lacks the ability to promote receptor dimerization and the biased NAM effect of AP.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAP-mediated\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;dimer interface\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe AP-mediated \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer interface is primarily built by TM3, TM4, and TM5, along with the loop regions ECL2 and ICL2 (Fig. 2F). In addition to the AP-mediated interactions around the central binding pocket (Fig. 2F\u0026ndash;H), there are extensive intermolecular interactions between the two \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR protomers from both extracellular and intracellular sides. As shown in Fig. 2F\u0026ndash;H, in the extracellular side, L167\u003csup\u003e4.59\u003c/sup\u003e and M171\u003csup\u003e4.63\u0026nbsp;\u003c/sup\u003ein TM4 and W173 in ECL2 of one protomer form extensive hydrophobic and van der Waals interactions with Q197\u003csup\u003e5.36\u003c/sup\u003e, A198\u003csup\u003e5.37\u003c/sup\u003e, I201\u003csup\u003e5.40\u003c/sup\u003e, and I205\u003csup\u003e5.44\u0026nbsp;\u003c/sup\u003ein TM5 of the other protomer. On the intracellular side, K149\u003csup\u003e4.41\u0026nbsp;\u003c/sup\u003ein TM4\u003csup\u003e\u0026nbsp;\u003c/sup\u003eforms a hydrogen-bonding interaction with the conserved Y132\u003csup\u003e3.51\u003c/sup\u003e (of the DRY motif) in TM3 of the adjacent protomer. In addition, F133\u003csup\u003e3.52\u003c/sup\u003e in TM3 and F139\u003csup\u003e34.51\u003c/sup\u003e, K140\u003csup\u003e34.51\u003c/sup\u003e, L144\u003csup\u003e34.56\u003c/sup\u003e, and L145\u003csup\u003e34.57\u0026nbsp;\u003c/sup\u003ein the ICL2 loop create an extensive hydrophobic network between the two protomers. Moreover, we observed clear lipid-like densities, running almost parallel to the membrane plane, bridging hydrophobic contacts between the two protomers (Fig. 2F\u0026ndash;H), which further stabilize the dimer conformation of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer interface captured in our cryo-EM structures differs from the potential interface observed in the lipid cubic phase (LCP) crystal structure of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003csup\u003e17\u003c/sup\u003e (Supplementary Fig. 4A). The AP-stabilized \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer interface also contrasts with those in the APJR dimers\u003csup\u003e2,12,18,19\u003c/sup\u003e (Supplementary Fig. 4B), rhodopsin dimer\u003csup\u003e1\u003c/sup\u003e (Supplementary Fig. 4C), and GPR3 dimer\u003csup\u003e3,20,21\u003c/sup\u003e (Supplementary Fig. 4D), while it is similar to a potential \u0026beta;\u003csub\u003e1\u003c/sub\u003eAR dimer interface in the LCP crystal structure\u003csup\u003e22\u003c/sup\u003e (Supplementary Fig. 4E). While previous biophysical studies have shown that the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR could form dimers and higher-order oligomers by itself\u003csup\u003e10,23\u003c/sup\u003e, it is possible that the new interface in our structures represents a native yet transient one for \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eValidation of AP-mediated \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization in liposomes and in cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe above structures of AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003ewere resolved in detergent micelles or nanodiscs.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eTo validate whether the AP-mediated \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer adopts the same structural arrangement in a continuous membrane, we applied single-molecule fluorescence resonance energy transfer (smFRET) microscopy to probe its conformation in liposomes. As shown in Fig. 3A, the TM helices in the dimer structure resolved by cryo-EM form a parallelogram-like arrangement when viewed perpendicular to the membrane plane, with two diagonals of different lengths. Fluorescent donor and acceptor dyes labeled on residues near different vertices should yield distinct inter-dye distances and corresponding FRET values. To label each protomer in a site-specific manner, single-cysteine mutations were introduced into a minimal cysteine construct of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003csup\u003e24\u003c/sup\u003e and \u0026ldquo;clicked\u0026rdquo; with maleimide-conjugated fluorescent dyes. Stochastic labeling is expected to result in half of the doubly labeled dimers having one donor and one acceptor dye. The labeled proteins were surface-immobilized and imaged using objective-based total internal reflection fluorescence (TIRF) microscopy. TM5-labeled samples exhibited a homogeneous Gaussian distribution of FRET values centered at ~0.9, while helix 8 (H8)-labeled samples showed a distribution centered at ~0.5 (Fig. 3A), with the higher FRET consistent with a shorter distance between the fluorophore pairs. These results are consistent with the cryo-EM structure and indicate the conformational stability of the dimer. To assess long-term stability, we measured ensemble FRET changes using dye-labeled samples in detergent (Supplementary Fig. 5A). When dimers were diluted into AP-free buffer, the relative FRET values remained nearly unchanged over 24 hours (Supplementary Fig. 5B,C). Moreover, incubation with Gs or activated \u0026beta;-arrestin could not dissociate the dimer (Supplementary Fig. 5D).\u003c/p\u003e\n\u003cp\u003eThe cryo-EM structure and biophysical data presented above provide strong evidence that AP promotes \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization; however, it remains unclear whether this effect occurs in cells. To further address this, we employed two techniques to assess \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization in cells. First, we analyzed the size of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR nanoclusters with or without AP treatment. GPCRs are known to form mesoscale nanoclusters in cells, which act as signaling hubs to locally concentrate effectors and enhance signaling efficiency\u003csup\u003e25,26\u003c/sup\u003e. Molecular mechanisms underlying nanoclustering are poorly understood but likely involve weak but specific interactions among receptors (Supplementary Fig. 6A). We hypothesized that AP stabilizes the dimer as a single unit that has increased interaction sites, facilitating prolonged contacts and the assembly of larger receptor clusters (Supplementary Fig. 6A). To visualize nanocluster distribution on cell membranes, we labeled \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR with a gold nanoparticle (AuNP)-conjugated antibody and prepared cryo-EM grids with cytoplasmic content removed via unroofing. As shown in Supplementary Fig. 6B,C, cryo-EM imaging revealed distinct AuNP clusters rather than random scatters. Statistics of AuNPs in each cluster demonstrated that AP significantly increased nanocluster size (Fig. 3B) as a result of AP-induced dimerization of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR. The resulting enlarged nanoclusters may amplify signaling outcomes compared to isolated monomers or dimers combined.\u003c/p\u003e\n\u003cp\u003eWe also performed cell-based Bioluminescence Resonance Energy Transfer (BRET) assays to quantify \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization by measuring distance-dependent BRET signals between Rluc-tagged and GFP-tagged receptors (Fig. 3C). Cells co-expressing differently tagged \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR exhibited AP-dependent dimerization with a pEC\u003csub\u003e50\u003c/sub\u003e of 5.93 \u0026plusmn; 0.18, n = 4 (Fig. 3D). Control experiments confirmed the specificity of this effect, showing no fluorescence interference from AP, no dimerization induced by Iso alone, and no expression-level-dependent self-homodimerization (Supplementary Fig. 7A\u0026ndash;C). To further validate dimerization specificity, we tested a \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR mutant (V129\u003csup\u003e3.48\u003c/sup\u003eL) that disrupts the AP binding pocket due to steric clashes. As expected, this mutation completely abolished AP-mediated dimerization (Fig. 3E). Similar effects were observed when swapping TM3 from \u0026beta;\u003csub\u003e1\u003c/sub\u003eAR into \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (Supplementary Fig. 7D) or assessing AP-mediated \u0026beta;\u003csub\u003e1\u003c/sub\u003eAR dimerization (Supplementary Fig. 7E). These results collectively demonstrate that AP robustly promotes \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization in native cellular environments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHigher-order oligomerization of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eApart from the dimer species in the cryo-EM dataset, a subset of 2D classes displayed top views of higher-order oligomers, including tetramers and hexamers (Supplementary Fig. 8A). We reconstructed a low-resolution map of the tetramer, revealing a dimer-of-dimers organization (Supplementary Fig. 8B). Docking of two dimer structures into the map shows symmetric interactions between the dimers, mediated by the intracellular regions of TM5\u0026ndash;TM6 and TM1\u0026ndash;ICL1 (Supplementary Fig. 8C). Mass photometry of diluted dimer samples in detergent confirmed the presence of a significant fraction of tetramer in solution (Supplementary Fig. 8D\u0026ndash;F). Consistent with these findings, immunogold imaging of cell membranes showed enlarged nanoclusters upon AP treatment (Fig. 3B and Supplementary Fig. 6), supporting the notion that the AP-induced dimer serves as a structural unit for higher-order oligomerization. Together, these results indicate that AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR can assemble into higher-order oligomers, providing a structural basis for receptor nanocluster formation in cells.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dime\u003c/strong\u003e\u003cstrong\u003er functions as a biased signaling species\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo understand the structural basis of the biased NAM activity of AP, we first compared our dimer structure with previously resolved monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR structures. One protomer in the dimer structure aligns well with the previous crystal structure of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in an inactive conformation (PDB ID: 2RH1), with an overall C\u0026alpha; RMSD of 0.53 \u0026Aring; (Fig. 4A). Notably, the structure of the dimer in nanodiscs, resolved without nanobodies or engineered fusions, exhibits a distinct ICL2 conformation from the crystal structure with an C\u0026alpha; RMSD of 1.5 \u0026Aring; (Fig. 4A). F139\u003csup\u003e34.51\u003c/sup\u003e on ICL2 participates in dimer contact with the same residue on the opposing protomer (Fig. 2F,H), resulting in an outward shift away from the central cavity involved in G-protein engagement (Fig. 4B). This shift is accompanied by other conformational changes in ICL2 (Fig. 5B and Supplementary Fig. 3B): Y141\u003csup\u003e34.53\u003c/sup\u003e forms hydrogen bond and \u0026pi;\u0026ndash;cation interactions with R131\u003csup\u003e3.50\u003c/sup\u003e in TM3, which normally forms a hydrogen bond with H269\u003csup\u003e6.31\u003c/sup\u003e in TM6; K263\u003csup\u003e6.25\u003c/sup\u003e at the intracellular end of TM6 appears to replace K140\u003csup\u003e34.52\u003c/sup\u003e on ICL2 in neutralizing the helical dipole of TM3 (Supplementary Fig. 3B). These local conformational rearrangements may alter the interplay between \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR and its transducers.\u003c/p\u003e\n\u003cp\u003eStructural analysis of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR bound to its transducers provided additional insights into how the dimer achieves biased inhibition. Alignment of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;Gs\u003csup\u003e27\u003c/sup\u003e, \u0026beta;\u003csub\u003e1\u003c/sub\u003eAR\u0026ndash;\u0026beta;-arrestin1\u003csup\u003e28\u003c/sup\u003e, and neurotensin receptor type 1 (NTSR1)\u0026ndash;G protein-coupled receptor kinase 2 (GRK2)\u003csup\u003e29\u003c/sup\u003e structures with one protomer in the dimer (Fig. 4C) revealed significant steric clashes between \u0026beta;-arrestin1 or GRK2 and the intracellular side of the other protomer, preventing their engagement with the receptor. In contrast, docked Gs appeared to be accommodated. However, dimerized \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR adopts an inactive-like conformation. Structural comparison with the Gs-bound receptor (Supplementary Fig. 9A) revealed that local conformational changes breaking dimer symmetry are needed to allow for the loop-to-helix transition in ICL2 (Supplementary Fig. 9B,C), necessary for engagement with Gs in the canonical state, while the outward movements of TM5 and TM6 are not restricted by the dimer contacts. Together, these data suggest that dimerization of the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR contributes to the biased NAM effect of AP.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo further confirm the functional consequences of the AP-induced dimerization, we conducted a series of biochemical assays to investigate the \u003cem\u003ein vitro\u003c/em\u003e activities of the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomer and dimer in isolation. Unlike cell-based assays, where \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR may exist as a mixture of monomers and dimers after AP treatment, the homogeneous and stable dimer sample that we obtained from purification enabled us to characterize the functions of dimer and monomer individually. First, we demonstrated that the dimer could form a stable complex with nucleotide-free Gs. This complex eluted as a single peak in SEC (Supplementary Fig. 10A), and the peak fractions contained the bands for Gs heterotrimer on SDS-PAGE (Supplementary Fig. 10B). Estimation of the relative ratios of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR to G\u0026alpha; on gels indicated a 2:1 stoichiometry in the dimer\u0026ndash;Gs complex, consistent with a docking model in which two Gs molecules cannot simultaneously engage the dimer (Fig. 4C). Preliminary cryo-EM analysis of the dimer\u0026ndash;Gs complex shows 2D class averages with one Gs density associated with an oval-shaped micelle housing one dimer (Supplementary Fig. 10C).\u003c/p\u003e\n\u003cp\u003eWe then used bio-layer interferometry (BLI) to measure the binding affinities and kinetics of guanosine diphosphate (GDP)-bound Gs to monomeric or dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, with and without AP (Fig. 4D and Supplementary Fig. 10D\u0026ndash;F). All samples exhibited \u0026mu;M-level affinity for Gs(GDP) (Supplementary Table 2). Despite adopting an inactive conformation in the structural models, the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer retains the ability to engage Gs, suggesting that the dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR can adopt a Gs-favorable conformation. The dimer showed a modest (~3-fold) decrease in affinity for Gs (dissociation constant, \u003cem\u003eK\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e = 3.11 \u0026mu;M) compared to the monomer (\u003cem\u003eK\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e = 0.96 \u0026mu;M). AP at 10 \u0026mu;M caused a slight reduction in the monomer\u0026ndash;Gs binding affinity (\u003cem\u003eK\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e = 1.7 \u0026mu;M). Interestingly, dimer\u0026ndash;Gs binding exhibited both a slower association rate (\u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e) and dissociation rate (\u003cem\u003ek\u003c/em\u003e\u003csub\u003eoff\u003c/sub\u003e) compared to monomer\u0026ndash;Gs binding, while AP\u0026apos;s effect on the monomer only reduced \u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e (Supplementary Table 2). These results suggest that AP partially inhibits recruitment of Gs by \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, with a more pronounced effect when tightly bound within the dimer. The smaller \u003cem\u003ek\u003c/em\u003e\u003csub\u003eoff\u003c/sub\u003e observed in dimer\u0026ndash;Gs dissociation may reflect a cooperative mechanism, where Gs engagement with one protomer promotes a Gs-favorable conformation in the adjacent protomer, thereby increasing Gs residence time during dissociation.\u003c/p\u003e\n\u003cp\u003eWe further performed radioligand binding assays to assess the allosteric effect of AP on orthosteric ligand binding using isolated monomer or dimer reconstituted in nanodiscs. The AP-bound dimer showed an ~5-fold higher affinity for [\u003csup\u003e3\u003c/sup\u003eH]-dihydroalprenolol (DHA), a \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR antagonist, in the saturation binding, while AP at 10 \u0026mu;M did not significantly affect [\u003csup\u003e3\u003c/sup\u003eH]-DHA binding to the monomer (Supplementary Fig. 11A). In agonist competition assays with Iso, the dimer exhibited a ~5-fold increase in the inhibitory constant (\u003cem\u003eK\u003c/em\u003e\u003csub\u003ei\u003c/sub\u003e), but no significant effect of AP on the monomer (Supplementary Fig. 11B). These findings suggest that AP stabilizes \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in a more inactive-like conformation in the dimer. We next evaluated the potency of Gs in stabilizing the active state of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR by allosterically modulating the orthosteric pocket to reduce antagonist binding. We titrated Gs(GDP) into \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR pre-bound with saturated [\u003csup\u003e3\u003c/sup\u003eH]-DHA and quantified the remaining bound [\u003csup\u003e3\u003c/sup\u003eH]-DHA. The results showed that Gs modulates the orthosteric pocket in both dimeric and monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, irrespective of AP presence (Fig. 4E). This suggests that while AP biases \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR toward an inactive-like conformation, it does not significantly hinder Gs engagement at equilibrium. We then assessed the effects of AP on the function of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR as a guanine nucleotide exchange factor (GEF), which facilitates GDP-GTP exchange in Gs and accelerates GTP hydrolysis. Using the GTPase-Glo\u0026trade; assay to monitor GTP turnover, we found that while AP significantly slows down the GTP turnover rate of monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, it did not affect the final efficacy. However, we observed much stronger inhibition in GTP turnover for the dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (Fig. 4F). This likely stems from two factors: (1) the slower association rate \u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e of Gs with dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR as observed by BLI (Supplementary Fig. 10F) and (2) AP\u0026rsquo;s biasing of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR toward an inactive-like state, potentially introducing additional rate-limiting steps in a full productive cycle.\u003c/p\u003e\n\u003cp\u003eTo test if AP impedes \u0026beta;-arrestin core engagement with \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, we performed competition radioligand binding assays measuring \u0026beta;-arrestin\u0026ndash;induced positive allosteric modulation of agonist binding. Using the sortase ligation strategy, we attached the synthetic phosphorylated vasopressin receptor 2 (V\u003csub\u003e2\u003c/sub\u003eR) peptide (V\u003csub\u003e2\u003c/sub\u003eRpp) to monomeric or dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (Fig. 4G and Supplementary Fig. 12A), ensuring full tail engagement of \u0026beta;-arrestin, a prerequisite for efficient core engagement. Monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in nanodiscs displayed a ~7-fold increase in Iso affinity in the presence of \u0026beta;-arrestin1 (Supplementary Fig. 12B), consistent with previous reports\u003csup\u003e30,31\u003c/sup\u003e. AP reduced this increase in affinity to ~2-fold in the monomer and abolished it in the dimer (Fig. 4G), suggesting AP partially hinders \u0026beta;-arrestin core interactions on monomers and completely prevents it in dimers. These data align with structural predictions where steric clashes between \u0026beta;-arrestin and the adjacent protomer in the dimer prevent core engagement.\u003c/p\u003e\n\u003cp\u003ePreviously, we showed that DFPQ inhibits GRK-mediated phosphorylation of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, a prerequisite for \u0026beta;-arrestin recruitment, in cells\u003csup\u003e4\u003c/sup\u003e. To identify which species is responsible for this inhibition, we measured GRK5 activity towards purified dimeric and monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR. AP-bound \u0026beta;₂AR dimers exhibited near-complete inhibition of GRK5 phosphorylation in both detergent and nanodiscs (Fig. 4H and Supplementary Fig. 13A), whereas GRK5 activity toward monomer in nanodiscs remained unaffected by AP (Fig. 4H and Supplementary Fig. 13A). The slight reduction in GRK5 activity observed for monomer in detergent at high AP concentrations (Fig. 4I) likely reflects the inefficient dimerization of \u0026beta;₂AR in detergent (Supplementary Fig. 1A), as we previously observed (Fig. 1C). However, the dose-dependent inhibition observed for monomer reconstituted in bicelles (Fig. 5I) is a result of AP-induced dimerization in membrane, underscoring the requirement of a continuous lipid bilayer for effective receptor dimerization. The relatively large diameter of typical bicelles (~40 nm) compared to nanodiscs (~12 nm) increases the likelihood of incorporation of multiple receptors and provides a fluid membrane environment conducive to dimer formation. AP-dependent inhibition of GRK5 activity was further corroborated by a BRET-based assay monitoring AP effect on direct GRK5 recruitment to the \u0026beta;₂AR (Supplementary Fig. 13B,C). These findings are consistent with structural analyses indicating that steric clashes in the dimer hinder GRK engagement, further supporting the role of the dimer in preventing \u0026beta;-arrestin recruitment.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNegative allosteric effect of AP on \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR conformation dynamics\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eModulation of protein dynamics often underpins how ligands achieve biased effects in highly dynamic proteins like GPCRs. To provide further insights into the mechanism of AP\u0026rsquo;s biased NAM effect, we explored how AP tunes the dynamic behaviors of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR. To this end, we performed smFRET microscopy to monitor conformational changes and double electron\u0026ndash;electron resonance (DEER) spectroscopy to measure residue-level distance distributions. Using the established protocols in our lab\u003csup\u003e24,32\u003c/sup\u003e, we placed fluorescent dyes for smFRET or spin labels for DEER near the intracellular ends of TM4 and TM6 on the minimal cysteine construct of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR. This allowed us to monitor TM6 outward movement, a hallmark of GPCR activation. Increased distances between TM4 and TM6 sites correspond to high-to-low FRET state transitions in smFRET experiments. As shown in Fig. 5A and detailed in the Methods, we used a two-affinity-tag purification strategy to produce an AP-mediated \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR \u0026ldquo;heterodimer\u0026rdquo;, where only one protomer contained double-cysteine mutants for dye labeling. The recorded traces from the TM4\u0026ndash;TM6 sensors of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR predominantly fall into two FRET states (Fig. 5B), representing an active and inactive state. In the absence of an agonist, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was primarily in the inactive state; however, monomers without AP exhibited a significantly higher fraction of active species compared to dimers, which showed almost no active population (Fig. 5C,D). Adding AP to monomers substantially reduced the active state population from 26.5 \u0026plusmn; 2.9% to 11.2 \u0026plusmn; 3.2% (Fig. 5D), whereas adding Iso to dimers failed to increase active state species (Fig. 5C). These results indicate that AP stabilizes an inactive conformation of TM6, and Iso alone fails to drive TM6 transition into an active state in the dimer. This is consistent with the observed enhancement of antagonist binding in the dimer (Supplementary Fig. 11A,B). Upon adding Gs to the dimer or the monomer with AP, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR transitioned predominantly to an active state, although dimers retained a higher fraction of inactive species (Fig. 5C,D), in line with Gs engaging only one protomer of the dimer, while the other may remain in inactive or partially active conformation. This indicates the potential of Gs to overcome the AP-induced stabilization of the inactive conformation, driving receptor activation even in the dimeric form. Washing Gs away from dimers reversed the population to a primarily inactive state, resembling the condition without Gs coupling (Fig. 5C). DEER measurements on the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (Supplementary Fig. 14), with spin labels reporting on the TM4\u0026ndash;TM6 conformation (Fig. 5E), revealed that AP causes changes that are consistent to those observed by smFRET. \u0026nbsp;Specifically, AP biases Iso-bound receptors toward a more homogeneous inactive-like conformation (Fig. 5F), while having a minimal effect on receptors bound to the super-agonist BI (Fig. 5G). The addition of Gs caused a fraction of the fully active conformer to become populated, regardless of AP presence (Fig. 5H). In addition to conformational effects, the modulation depth parameter in pulsed dipolar EPR spectroscopy provides a quantitative measure of spin cluster size, with higher values reflecting increased local concentrations of spin labels. Agonist-bound receptor exhibited a reduced modulation depth relative to the Apo receptor, suggesting that receptor activation is associated with decreased dimerization. Addition of AP to agonist-bound receptor resulted in an increase in modulation depth, consistent with enhanced dimerization (Supplementary Fig. 14). Notably, Gs binding largely inhibited the AP-induced increase in modulation depth, indicating that Gs coupling counteracts the dimer-promoting effect of AP. The magnitude of the shifts in modulation depth indicates slight changes in receptor monomer-dimer equilibrium consistent with a predominantly monomeric receptor, as expected for receptor in detergent. However, the clear trend in modulation depth indicates a correlation between the monomer-dimer and inactive-active equilibria. Together, these findings demonstrate that Gs can couple to both dimers and monomers with AP, driving TM6 outward movement. Moreover, the inhibitory effect of AP on the TM6 outward movement of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomer may also contribute to the level of biased signaling observed. This is supported by the observation that \u0026beta;-arrestin inhibition emerges at lower AP concentrations compared to dimerization in cell-based assays. Notably, the effect of AP on \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dynamics resembles the previous observation in \u0026mu;-opioid receptor (\u0026mu;OR), where G protein-biased agonists are less efficient than balanced full agonists in stabilizing TM6 outward movement in both smFRET and DEER studies\u003csup\u003e33\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur cryo-EM structures reveal that AP, a \u0026beta;-arrestin\u0026ndash;biased NAM, promotes \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization, thereby modulating downstream signaling. AP shows strong bias toward \u0026beta;-arrestin inhibition in cell-based assays, and its parent compound, DFPQ, produces sustained relaxation of contracted airway smooth muscle (ASM) in both human ASM cells and mouse airway models by preventing \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR desensitization\u003csup\u003e4\u003c/sup\u003e, suggesting its physiological relevance. We elucidate the mechanistic basis underlying this biased effect: AP robustly induces \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization. Biochemical and biophysical analyses demonstrate that the dimer acts as a biased signaling species, strongly inhibiting GRK-mediated phosphorylation and \u0026beta;-arrestin recruitment while permitting G protein coupling. Structural analysis indicates that steric hindrance imposed by the adjacent protomer prevents GRK and \u0026beta;-arrestin from adopting their canonical binding modes. These findings offer a novel perspective for achieving signaling bias: besides fine-tuning local conformational states, AP exploits the quaternary structure of receptor to selectively restrict transducer interactions.\u003c/p\u003e\n\u003cp\u003eBased on our results, we propose an activation model for \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in the presence of AP (Fig. 6). We simplify the energy landscape into two states\u0026mdash;an inactive ground state and a fully transducer-bound active state\u0026mdash;and outline how AP reshapes this profile. In the absence of AP, the balanced agonist drives the receptor into an active conformation that readily accommodates transducer binding due to a low energy barrier. However, the energy barrier to achieve the fully transducer-bound active state is increased by AP. In the monomer, AP binds a similar pocket as in the dimer, biasing the receptor toward an inactive-like conformation. GRK phosphorylation remains largely intact; however, the reduced allosteric enhancement of agonist binding by \u0026beta;-arrestin suggests a partially destabilized engagement state. AP also slows \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR-catalyzed nucleotide exchange in Gs, indicating a higher activation barrier for Gs, although Gs binding and activation remain largely unaffected, suggesting an unaltered active-state free energy. At higher concentrations, AP promotes stable \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization, resulting in near-complete loss of GRK phosphorylation and blockade of \u0026beta;-arrestin core engagement. This reflects not only increased energy barriers but also a disrupted \u0026beta;-arrestin-bound state, due to steric clashes with the adjacent protomer. GTP turnover by Gs is further slowed in the dimer, again reflecting a higher activation barrier, while the free energy of the Gs-bound state remains similar to that of the monomer. In cells, where monomers and dimers coexist, AP shifts the receptor ensemble toward the more strongly biased dimeric form in a concentration-dependent manner. Despite reduced GTP turnover by Gs, signal amplification likely compensates and enables near-maximal cAMP production even under conditions where dimer is dominant. Thus, the loss of GRK and \u0026beta;-arrestin engagement in the dimer biases the receptor exclusively toward sustained Gs signaling, highlighting dimerization as a mechanism to selectively rewire family A receptor signaling.\u003c/p\u003e\n\u003cp\u003eThe dimer interface observed in our AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR complex displays unique features compared to other family A GPCR dimers. While previous studies have reported dimer structures in receptors such as APJR\u003csup\u003e12,18\u003c/sup\u003e and rhodopsin\u003csup\u003e1\u003c/sup\u003e by cryo-EM, our AP-stabilized dimer predominantly involves interactions on TM3, TM4, TM5, and ICL2, a configuration distinct from the earlier \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR crystal structure\u003csup\u003e17\u003c/sup\u003e and other family A GPCR dimers. Importantly, the outward mobility of TM5 and TM6 is not restricted in our dimer configuration, which is essential for efficient transducer engagement. Notably, recent work\u003csup\u003e34\u003c/sup\u003e on the platelet-activating factor receptor using a cysteine crosslinking strategy reported that dimerization significantly biases signaling toward G protein pathways while limiting \u0026beta;-arrestin recruitment; one of their proposed dimer models resembles our structure with a TM3\u0026ndash;TM4\u0026ndash;TM5 interface. Similarly, recent studies used a computational and mutagenesis approach to design stable TM4\u0026ndash;TM5 dimers of CXCR4, which biased signaling towards Gi\u003csup\u003e35\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eBeyond discrete dimer formation, cryo-EM imaging reveals that AP-mediated \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization drives the assembly of larger nanoclusters, and our structural model of the dimer-of-dimer provides a potential molecular basis for specific interactions that underlie this mesoscale organization. Such nanoclusters likely serve as dynamic signaling platforms that modulate both the amplitude and duration of receptor-mediated responses\u003csup\u003e25\u003c/sup\u003e. Recent papers\u003csup\u003e36-38\u003c/sup\u003e on higher-order transient structures (HOTS) highlight that dynamic oligomerization is a conserved strategy in membrane proteins, which may be beneficial for signaling efficiency and specificity in cells. Consistent with HOTS, AP-induced \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR nanoclusters may fine-tune the balance between receptor activation, desensitization, and internalization. Although our GTP turnover assays on isolated AP-mediated dimers showed reduced turnover activities, the enlarged nanoclusters observed in cells may locally concentrate G proteins, thereby supporting more productive signaling events. Further investigations into the physiological relevance and functional consequences of these nanoclusters will help uncover the potential for harnessing such mechanisms across other receptor systems.\u003c/p\u003e\n\u003cp\u003eOur findings extend the emerging paradigm of molecular glues in pharmacology\u003csup\u003e39,40\u003c/sup\u003e. AP functions as a GPCR molecular glue by bridging two \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR protomers to form a dimeric assembly with biased functions. Although molecular glues that stabilize GPCR interactions with transducers\u003csup\u003e41\u003c/sup\u003e, including G proteins\u003csup\u003e42,43\u003c/sup\u003e, \u0026beta;‑arrestin\u003csup\u003e44,45\u003c/sup\u003e, and GRK\u003csup\u003e29\u003c/sup\u003e, have been reported, AP represents the first example of a ligand mediating TM interactions that drive dimerization of a family A GPCR. This mechanism is reminiscent of that observed in the membrane-bound molecular glue, NVS‑STG2\u003csup\u003e46\u003c/sup\u003e, for stimulator of interferon genes (STING). NVS‑STG2 induces high-order oligomerization of STING, robustly triggering downstream immune signaling\u003csup\u003e46\u003c/sup\u003e. Both examples illustrate how small molecules can harness and reshape the intrinsic propensity of membrane proteins to form oligomeric assemblies, thereby selectively modulating signal transduction.\u003c/p\u003e\n\u003cp\u003eIn summary, our study deciphers the structural basis by which AP acts as a biased NAM for \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR and provides broader insights into how receptor dimerization impacts GPCR signaling. Such insights not only deepen our understanding of structure\u0026ndash;function relationships in GPCR modulation but also highlight potential strategies for designing next-generation ligands that exploit oligomerization as a route to bias receptor signaling.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eExpression and Purification of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR in Sf9 cells\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR construct PN1 was expressed and purified as previously described\u003csup\u003e32,47\u003c/sup\u003e. Briefly, Sf9 cells were infected with a PN1-containing baculovirus produced via BestBac method. Cells were then harvested and resuspended in chilled lysis buffer containing 10 mM HEPES, pH 7.4, 1 mM EDTA, 1 \u0026mu;M alprenolol, and protease inhibitors (leupeptin and benzamidine). Lysed cells were then pelleted at 18,600 rpm for 20 minutes and dounced to homogeneity in chilled solubilization buffer containing 20 mM HEPES pH 7.4, 350 mM NaCl, 1% n-dodecyl \u0026beta;-D-maltoside (DDM), 0.1% cholesteryl hemisuccinate (CHS), 2 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 1 \u0026mu;M alprenolol, protease inhibitors, and benzonase. After stirring for 90 minutes at 4 \u003csup\u003eo\u003c/sup\u003eC and centrifugation at 18,600 rpm for 30 minutes, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e was added to the soluble fraction, which was then applied to anti-FLAG (DYKDDDDK) M1 immunoaffinity resin. The receptor was then washed (20 mM HEPES, pH 7.4, 350 mM NaCl, 0.1% DDM, 0.01% CHS, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, leupeptin, and benzamidine), eluted (20 mM HEPES, pH 7.4, 350 mM NaCl, 0.1% DDM, 0.01% CHS, 5 mM EDTA, and 200 ug/mL FLAG peptide), and further purified on Superdex 200 10/300 Increase gel filtration column equilibrated in NH buffer (20 mM, HEPES pH 7.4, 100 mM NaCl) plus 0.1% DDM and 0.01% CHS.\u003c/p\u003e\n\u003cp\u003eTo produce homogeneous \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer, Sf9 cells expressing PN1 were resuspended at room temperature (RT) in 20 mM HEPES, pH 7.4, 150 mM NaCl, 10% glycerol, 1 \u0026mu;M alprenolol, protease inhibitors, and 10 \u0026mu;M AP. After a 30-minute incubation at RT, membrane solubilization was initiated by adding 1% lauryl maltose neopentyl glycol (LMNG) and 0.1% CHS. The purification steps followed the same protocol as for the monomeric receptor, except that 0.01% LMNG replaced DDM in all buffers, and 10 \u0026mu;M AP was maintained throughout the purification.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExpression and Purification of heteromeric G\u0026alpha;s\u0026beta;1\u0026gamma;2\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs previously described\u003csup\u003e48,49\u003c/sup\u003e, heterotrimeric Gs was expressed and purified from \u003cem\u003eTrichoplusia ni\u003c/em\u003e Hi5 cells. Briefly, two baculoviruses were generated from the BestBac method, one encoding the wild-type human G\u0026alpha;s subunit and the other encoding the wild-type human \u0026beta;1\u0026gamma;2 subunits containing a histidine tag on the N-terminus of the \u0026beta; subunit. Cells were infected with both viruses for 48 hours and harvested via centrifugation. The pellet was then resuspended and stirred for 30 minutes at 4 \u003csup\u003eo\u003c/sup\u003eC in hypotonic buffer containing 10 mM HEPES pH 7.4, 100 \u0026mu;M MgCl\u003csub\u003e2\u003c/sub\u003e, 5 mM \u0026beta;-mercaptoethanol (BME), 20 \u0026mu;M GDP, and protease inhibitors. Lysed cells were then pelleted at 18,600 rpm for 15 minutes and dounced to homogeneity in chilled NH buffer plus 1% sodium cholate, 0.05% DDM, 1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 5 mM BME, 20 \u0026mu;M GDP, and protease inhibitors. After solubilization for 1.5 hours while stirring at 4 \u003csup\u003eo\u003c/sup\u003eC and centrifugation at 18,600 rpm for 35 minutes, 20 mM of imidazole was added to the soluble fraction, which was then allowed to batch-bind to washed nickel-chelated Sepharose for 2 hours. Pelleted resin was then loaded into a narrow column, washed with buffers containing gradually declining amounts of cholate, and eluted with NH buffer plus 0.05% DDM, 1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 20 \u0026mu;M GDP, 100 \u0026mu;M tris(2-carboxyethyl)phosphine (TCEP), and 250 mM imidazole. Human rhinovirus 3C protease was added to cleave the histidine tag and the eluate was dialyzed overnight at 4C in 2 L of dialysis buffer (NH buffer plus 1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 0.05% DDM, 20 \u0026mu;M GDP, and 100 \u0026mu;M TCEP). The protein solution was run through a second nickel-chelated Sepharose column, washed with dialysis buffer supplemented with 20 mM imidazole, and dephosphorylated for 30 minutes on ice with lambda protein phosphatase, calf intestinal phosphatase, and Antarctic phosphatase with 1 mM manganese chloride. The heterotrimer was further purified from excess \u0026beta;\u0026gamma; subunits by using ion exchange chromatography on a MonoQ 10/100 GL column. Sample was loaded and washed with 20 mM HEPES, pH 7.4, 1 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 0.05% DDM, 100 \u0026mu;M TCEP, and 20 \u0026mu;M GDP. Heterotrimer Gs was then eluted with a linear salt gradient from 50 mM NaCl to 500 mM NaCl.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExpression and Purification of Nb60\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNb60 was expressed and purified as previously described\u003csup\u003e15\u003c/sup\u003e. Briefly, Nb60 was expressed in \u003cem\u003eEscherichia coli\u003c/em\u003e BL21(DE3) cells. Lysate was then purified on nickel-chelated Sepharose column and subsequently on a Superdex 200 Increase 10/300 column in 20 mM HEPES, pH 7.4, and 150 mM NaCl.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSample preparation for cryo-EM\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the sample in detergent, purified PN1 in 0.1% DDM/0.01% CHS was loaded onto anti-FLAG M1 immunoaffinity resin equilibrated in the same detergent mixture supplemented with 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e. A detergent exchange was then performed, during which the receptor bound to the resin was washed with increasing ratios of TTG-T10 to DDM, ultimately transitioning the receptor into 0.01% TTG-T10/0.001% CHS. Each detergent exchange buffer contained NH buffer with 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, 1 \u0026mu;M carazolol, and 10 \u0026mu;M AP. The receptor was subsequently eluted in NH buffer plus 0.01% TTG-T10, 0.001% CHS, 1 \u0026mu;M carazolol, 10 \u0026mu;M AP, 5 mM EDTA, and 200 ug/mL FLAG peptide. After the detergent exchange, PN1 was incubated for 1 hour with 2-fold molar excess of Nb60. Excess Nb60 was cleared on anti-FLAG M1 immunoaffinity resin after washing with NH buffer plus 0.0006% TTG-T10, 0.00006% CHS, 1 \u0026mu;M carazolol, 10 \u0026mu;M AP, and 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e. The complex was then eluted off the resin with NH buffer plus 0.0006% TTG-T10, 0.00006% CHS, 1 \u0026mu;M carazolol, 10 \u0026mu;M AP, 5 mM EDTA, and 200 \u0026mu;g/mL FLAG peptide. The final sample was concentrated to \u0026gt;10 mg/mL and used immediately for cryo-EM grid preparation.\u003c/p\u003e\n\u003cp\u003eTo prepare the AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer in lipid nanodiscs, the purified dimer in LMNG was reconstituted into nanodiscs following a previous protocol\u003csup\u003e47\u003c/sup\u003e with modifications. Lipids were prepared by mixing 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC, Avanti), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS, Avanti), and cholesterol (Sigma-Aldrich) at a molar ratio of 7:2:1, followed by drying under argon and vacuum desiccation for 2 hours. Lipids were resuspended in NH buffer containing 14 mM DDM at 20 mg/mL. The receptor was diluted to 10 \u0026mu;M, incubated with 50 \u0026mu;M BI (MedChemExpress) for 10 minutes on ice, and then combined with membrane scaffold protein (MSP) 1E3D1 and lipids at a molar ratio of 1:2.5:100. The mixture was incubated on ice for 1 hour to allow nanodisc assembly. Detergent was removed by sequential addition of semi-wet Bio-Beads SM2 (60 mg/mL) three times over 3-hour period, followed by overnight incubation at 4\u0026deg;C. The next day, Bio-Beads were removed, and empty nanodiscs were separated by M1 affinity purification. The dimeric \u0026beta;₂AR in nanodiscs were eluted in buffer containing 1 \u0026mu;M BI, 10 \u0026mu;M AP, 5 mM EDTA, and 200 \u0026mu;g/mL FLAG peptide. The eluate was further polished by SEC in NH buffer plus 1 \u0026mu;M BI and 10 \u0026mu;M AP. Peak fractions were pooled and concentrated to 5 mg/mL for grid freezing.\u003c/p\u003e\n\u003cp\u003eThe cryo-EM grids were prepared using Vitribot Mark IV (Thermo Fisher\u0026nbsp;Scientific). Quantifoil R1.2/1.3 Au grids were glow-discharged with air for 90 s at 10 mA using Plasma Cleaner (PELCO EasiGlowTM). Aliquots of 3 \u0026mu;L protein sample were applied to the glow-discharged grids. After blotting with filter paper (Ted Pella, Inc) for 3.0 s, the grids were plunged into liquid ethane cooled with liquid nitrogen.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCryo-EM data collection and processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor detergent- and nanodisc-reconstituted \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR samples, a total of 7,139 and 17,997 micrograph stacks, respectively, were collected using a Titan Krios G3i operating at 300 kV on a Falcon 4i direct electron detector or on a K3 camera (Gatan) with a Quantum energy filter. Micrographs were recorded at a nominal magnification of 130,000\u0026times;, corresponding to a calibrated pixel size of 0.95 \u0026Aring; (detergent) or 0.83 \u0026Aring; (nanodisc) with defocus values ranging from \u0026ndash;2.0 \u0026mu;m to \u0026ndash;0.8 \u0026mu;m. Each movie stack received a total electron dose of ~50 e⁻/\u0026Aring;\u0026sup2; over 40 frames.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMotion correction and contrast transfer function (CTF) estimation were performed using the patch motion and patch CTF in cryoSPARC\u003csup\u003e50\u003c/sup\u003e. Particle picking was carried out using the blob picker, and extracted particles were binned 4\u0026times; for initial 2D classification. 2D classes with recognizable structural features were manually selected, and monomeric and dimeric particles were grouped separately. Ab initio reconstruction was performed using ~10% of the selected particles to generate four initial 3D references. Iterative rounds of heterogeneous refinement were conducted until poor-quality classes accounted for less than 5% of input particles. For the detergent dataset, 48,805 dimeric and 25,201 monomeric particles were selected for non-uniform (NU) refinement\u003csup\u003e51\u003c/sup\u003e. For the nanodisc dataset, 237,408 particles of dimeric species were selected. Particles were imported into RELION 4\u003csup\u003e52,53\u003c/sup\u003e for Bayesian polishing\u003csup\u003e54\u003c/sup\u003e and subsequently returned to cryoSPARC for further NU refinement. C2 symmetry was imposed for the final NU refinement of dimer species After handedness correction, local and CTF refinement were performed using a soft mask centered on the transmembrane domains. Final resolutions were estimated using the gold-standard Fourier shell correlation (FSC) 0.143 criterion. Local resolution estimation was carried out in cryoSPARC.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModel building and refinement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInitial models of\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (PDB ID: 2RH1)\u0026nbsp;were docked into cryo-EM maps using UCSF Chimera\u003csup\u003e55\u003c/sup\u003e and manually adjusted in COOT\u003csup\u003e56\u003c/sup\u003e to fit the density, including placement of ligands. Real-space refinement was performed in PHENIX\u003csup\u003e57,58\u003c/sup\u003e with secondary structure and geometry restraints. Model validation was conducted using EMRinger\u003csup\u003e59\u003c/sup\u003e to assess side chain density fitting. Structural figures were prepared using PyMOL and ChimeraX\u003csup\u003e60\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMass photometry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMass photometry measurements were performed using a Refeyn TwoMP instrument (Refeyn Ltd.) and the AcquireMP software (v2.3) following an established protocol\u003csup\u003e61\u003c/sup\u003e. Microscope coverslips (24 \u0026times; 50 mm, #1.5 thickness; Corning) were cleaned with deionized water and isopropanol, then dried prior to use. Silicone gaskets were applied to the coverslips to form individual wells immediately before sample loading. The instrument was calibrated using NativeMark unstained protein standards (Thermo Fisher Scientific) following the manufacturer\u0026apos;s instructions. Each measurement was conducted by first pipetting 10 \u0026mu;L of NH buffer into a well, followed by focal alignment and locking. Then, 1 \u0026mu;L of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR sample at a final concentration of 20 nM after dilution was added, mixed gently, and data were acquired for 60 seconds. At least 2,000 binding events were recorded per sample. Data processing and molecular mass determination were performed using the DiscoverMP software suite (Refeyn).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell unroofing and immunogold cryo-EM\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCell unroofing and immunogold labeling were performed as previously described\u003csup\u003e36,37,62\u003c/sup\u003e, with modifications. Quantifoil R1.2/1.3 300-mesh gold EM grids were glow-discharged for 15 seconds, rinsed three times with 70% ethanol, and washed four times with DPBS. Grids were incubated with 0.1 mg/mL poly-D-lysine (Gibco) for 1 hour at RT, followed by four washes in Dulbecco\u0026apos;s phosphate-buffered saline (DPBS, Gibco). Laminin (Sigma-Aldrich, 15 \u0026mu;g/mL) was applied to the grids and incubated at 37\u0026deg;C for 2 hours, then washed again with DPBS. HEK293F cells in suspension were seeded onto prepared grids and cultured until reaching 60\u0026ndash;70% confluency. \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was overexpressed via baculovirus transduction following the BacMam protocol\u0026nbsp;(Thermo Fisher Scientific). Around 16 hours after transfection, cells were rinsed with DPBS containing calcium and magnesium. Grids bearing adherent cells were held with tweezers and dipped into a hypotonic swell buffer (6 mM HEPES-KOH pH 7.4, 43.3 mM K-gluconate, 1.6 mM NaCl, 0.6 mM MgCl\u003csub\u003e2\u003c/sub\u003e) for 30 seconds. An additional 6 \u0026mu;L of swell buffer was added to each grid, followed by blotting with Whatman Grade 5 filter paper (Sigma-Aldrich) to remove the apical membrane. Unroofed samples were blocked with 3% goat serum (Thermo Fisher Scientific) in DPBS containing protease inhibitors (Thermo Fisher Scientific) for 20 minutes at RT. Primary antibody against \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR (Thermo Fisher Scientific) diluted in DPBS with 1% goat serum was applied for 1 hour. After three washes in DPBS, grids were incubated with a goat anti-mouse secondary antibody conjugated to colloidal gold (Ted Pella, Inc) diluted in DPBS with 1% goat serum for 1 hour, then washed again three times.\u003c/p\u003e\n\u003cp\u003eGrids were plunge-frozen in liquid ethane using a Leica EM GP2 system and stored in liquid nitrogen. Cryo-EM imaging was performed on a Glacios G2 operated at 200 kV, equipped with a Falcon 4i detector, at a nominal magnification of 11,000\u0026times;.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell-based BRET\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the effects of AP on cAMP production and \u0026beta;-arrestin recruitment, HEK293 cells endogenously expressing \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR were transiently transfected with either the BRET-based intramolecular cAMP sensor CAMYEL or with plasmids encoding \u0026beta;-arrestin2\u0026ndash;GFP10 and \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;RlucII. CAMYEL comprises both donor and acceptor fused to the cAMP-binding domain of EPAC and undergoes a conformational change upon cAMP binding that alters the BRET signal\u003csup\u003e63\u003c/sup\u003e. Forty-eight hours post-transfection, cells were pre-incubated with increasing concentrations of AP (0.03\u0026ndash;100 \u0026mu;M) for 30 minutes, followed by stimulation with 1 \u0026mu;M isoproterenol for 30 minutes in the presence of 5 \u0026mu;M coelenterazine H (for CAMYEL) or deep blue coelenterazine (DBC) (Cayman Chemical, Ann Arbor, USA) for the \u0026beta;-arrestin assay.\u003c/p\u003e\n\u003cp\u003eFor BRET-based \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization studies, HEK293 cells were co-transfected with 10 ng/well of either \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;RlucII,\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR_V129L\u0026ndash;RlucII,\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR_TM3\u0026ndash;RlucII, or\u0026nbsp;\u0026beta;\u003csub\u003e1\u003c/sub\u003eAR\u0026ndash;RlucII (BRET donors), along with 20 ng/well of their respective GFP-tagged counterparts (BRET acceptors). Forty-eight hours post-transfection, cells were incubated with increasing concentrations of AP (0.03\u0026ndash;100\u0026nbsp;\u0026mu;M) or Iso (0.1\u0026ndash;100\u0026nbsp;\u0026mu;M) for 1 hour, followed by addition of DBC substrate and incubation for 20 minutes. To control for potential AP autofluorescence effects on the BRET signal, HEK293 cells transfected with \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;GFP were treated with AP (0.03\u0026ndash;100\u0026nbsp;\u0026mu;M) for 30 minutes, and the induced GFP signal was directly measured and represented as fold change over basal.\u003c/p\u003e\n\u003cp\u003eFor saturation studies of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR homodimerization, cells were co-transfected with a fixed amount of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;RlucII (10 ng/well) and increasing amounts of GFP-tagged\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR plasmid (0\u0026ndash;100 ng/well). BRET measurements were taken 48 hours post-transfection after incubation with DBC for 30 minutes.\u003c/p\u003e\n\u003cp\u003eTo evaluate GRK5 recruitment to the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, cells were co-transfected with \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;Rluc and GRK5\u0026ndash;GFP. Cells were either stimulated with increasing concentrations of Iso for 30 minutes followed by DBC incubation for 20 minutes or pre-treated with increasing concentrations of AP (0.03\u0026ndash;100\u0026nbsp;\u0026mu;M) for 1 hour followed by 1 \u0026mu;M Iso stimulation for 30 minutes in the presence of the Rluc substrate DBC.\u003c/p\u003e\n\u003cp\u003eAll BRET signals were recorded at 395 nm (donor emission) and 510 or 530 nm (acceptor emission) using an Infinite F500 plate reader (Tecan, M\u0026auml;nnedorf, Switzerland). Induced BRET changes were calculated by subtracting the basal BRET signal (in the absence of ligand) from the BRET signal measured after stimulation.\u003c/p\u003e\n\u003cp\u003eFor the AP-mediated cAMP and \u0026beta;-arrestin assays as well as GRK5 recruitment, results are expressed as percent of the Iso-alone response. For dimerization studies and Iso-induced GRK5 recruitment, data are presented as percentage of the maximal BRET signal (% of MAX). All dose\u0026ndash;response curves were fitted using the log(agonist/inhibitor) vs. response (three parameters) function in GraphPad Prism. Data are shown as mean \u0026plusmn; S.E.M., n = 3/4.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSPR analysis on AP binding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSPR measurements were performed using a Biacore T200 system. Monomeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was captured onto a high-affinity streptavidin (SA) sensor chip (Cytiva) via a biotinylated anti-FLAG M2 antibody (Sigma-Aldrich), yielding a final response of ~1500 resonance units (RU). The running buffer consisted of 20 mM HEPES (pH 7.5), 100 mM NaCl, 0.01% LMNG. AP was injected at increasing concentrations ranging from 0.6 to 10 \u0026mu;M at a flow rate of 30 \u0026mu;L/min. Association and dissociation phases were recorded for each injection. All sensorgrams were processed using double-referencing to correct for nonspecific binding and instrument artifacts. This was accomplished by subtracting both (i) the response obtained from a blank injection of running buffer over the active surface (to account for bulk refractive index changes and injection artifacts), and (ii) the response from the compound injection over a reference surface without immobilized protein (to correct for nonspecific binding to the surface and matrix effects). Sensorgrams were processed using Biacore Evaluation Software, and binding curves were fit globally using a steady-state affinity model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBLI on Gs binding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBLI measurements were conducted at 30\u0026deg;C with continuous shaking at 1000 rpm using an Octet RED384 system (Fort\u0026eacute;Bio). SA biosensor tips (Sartorius) were coated with 10 nM biotinylated anti-FLAG M1 fragment antigen-binding region (Fab) in the NH buffer with 0.01% LMNG and 0.001% CHS for 300 seconds. FLAG-tagged \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was then captured by incubating the tips in 100 nM \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR with 10 \u0026mu;M Iso for 600 seconds. Following receptor immobilization, the biosensors were transferred into wells containing a concentration series of Gs (30 nM to 10 \u0026mu;M) in the binding buffer containing 10 \u0026mu;M GDP and 0.1% bovine serum albumin (BSA, Sigma-Aldrich) for 180 seconds (association phase), followed by transfer into buffer-only wells for 300 seconds (dissociation phase). Control channels lacking either Gs or immobilized \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR were used for double-reference subtraction. Association and dissociation kinetics were fitted with a single-exponential model to derive apparent \u003cem\u003ek\u003c/em\u003e\u003csub\u003eon\u003c/sub\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003eand \u003cem\u003ek\u003c/em\u003e\u003csub\u003eoff\u003c/sub\u003e. Equilibrium binding responses were used to determine the dissociation constant \u003cem\u003eK\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRadioligand binding assay\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor saturation binding studies, 50-100 femtomoles of monomeric or dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR reconstituted in nanodiscs, following the protocol described in the sample preparation for cryo-EM, were incubated with increasing concentration of [\u003csup\u003e3\u003c/sup\u003eH]-DHA at RT for 1 hour in a buffer containing 20 mM HEPES, 100 mM NaCl, 0.5% BSA. Non-specific binding of the radioligand was determined by adding 10 \u0026mu;M alprenolol in the same reaction system. For monomeric\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, the assay was performed with or without AP. For competition binding studies, monomeric or dimeric\u0026nbsp;\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR reconstituted in nanodiscs were incubated with 1 nM [\u003csup\u003e3\u003c/sup\u003eH]-DHA and increasing concentrations of Iso or GDP-bound Gs in the same buffer as saturation binding. Nanodiscs were separated from excess [\u003csup\u003e3\u003c/sup\u003eH]-DHA on Whatman GF/B filters using a Brandel 96-well harvester. The bound radioligand were read on a liquid scintillation counter (MicroBeta Jet, PerkinElmer). Data were analyzed by GraphPad Prism 10.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGTP Turnover\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe GTPase GLO assay was performed using a modified GTPase-GLO\u003csup\u003eTM\u003c/sup\u003e assay from Promega as previously described\u003csup\u003e24,32\u003c/sup\u003e. Briefly, 100 nM of monomeric PN1, reconstituted in nanodiscs with MSP1E3D1 following the protocol described in the sample preparation for cryo-EM, was incubated for 1 hour at RT with 20 \u0026mu;M Iso (Sigma-Aldrich) and a range of concentrations of AP in NH buffer plus 0.2% DMSO, and 20 \u0026mu;M GTP. Simultaneously, a 1 \u0026mu;M stock of heterotrimeric Gs protein was prepared in a buffer consisting of NH buffer plus 0.04% DDM, 200 \u0026mu;M TCEP, 20 mM MgCl\u003csub\u003e2\u003c/sub\u003e, and 20 \u0026mu;M GDP. Equal volumes of PN1 and Gs were then mixed and incubated for 60 minutes. The final reaction consisted of 50 nM of ligand-bound PN1 and 500 nM Gs in NH buffer plus 0.1% DMSO, 0.02% DDM, 100 \u0026mu;M TCEP, 10 \u0026mu;M MgCl\u003csub\u003e2\u003c/sub\u003e, 10 \u0026mu;M GTP, and 10 \u0026mu;M GDP. An equal volume of GTPase-Glo reagent in NH buffer plus 0.02% DDM and 5 \u0026mu;M ATP was then added and incubated for 30 minutes. Detection reagent was subsequently added and incubated for 10 minutes. Luminescence was detected using the MicroBeta counter. For the time course assay, the experimental setup remained identical except 200 nM of AP-mediated dimeric PN1 was included as a condition and the PN1-Gs reactions occurred for 30, 60, 90, and 120 minutes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLigation of V\u003csub\u003e2\u003c/sub\u003eRpp to receptor and \u0026beta;-arrestin competition radioligand binding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR constructs modified with a C-terminal sortase recognition sequence (LPETGHH inserted after residue 365) were expressed in Sf9 cells and purified as described above for monomeric and AP-induced dimeric receptors. Sortase-mediated ligation of synthetic V\u003csub\u003e2\u003c/sub\u003eRpp to receptor was performed as previously described\u003csup\u003e30,31\u003c/sup\u003e. For ligation reaction, 10 \u0026mu;M purified receptor was incubated in NH buffer supplemented with 0.01% LMNG, 0.001% CHS, and 5 mM CaCl\u003csub\u003e2\u003c/sub\u003e with 50 \u0026mu;M synthetic GGG\u0026ndash;V\u003csub\u003e2\u003c/sub\u003eRpp peptide and 2 \u0026mu;M evolved sortase A pentamutant (eSrtA)\u003csup\u003e64\u003c/sup\u003e. The mixture was incubated overnight at 4 \u0026deg;C. Unreacted receptor and eSrtA (bearing the C-terminal His tag) was removed by binding to nickel-chelated Sepharose resins. Labeled monomeric or dimeric \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;V\u003csub\u003e2\u003c/sub\u003eRpp samples were reconstituted into nanodiscs following the protocol described in the sample preparation for cryo-EM.\u003c/p\u003e\n\u003cp\u003eThe equilibrium competition radioligand binding assays were performed with \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026ndash;V\u003csub\u003e2\u003c/sub\u003eRpp in nanodiscs in the presence of 2 nM [\u003csup\u003e3\u003c/sup\u003eH]-DHA, increasing concentrations of Iso, and 1 \u0026mu;M C-tail-truncated \u0026beta;-arrestin1(382), prepared as previously described\u003csup\u003e65\u003c/sup\u003e. 10 \u0026mu;M AP was added where applicable. After incubation at RT for 1 hour, samples were harvested, and radioactivity was measured as described in the previous section to calculate the \u003cem\u003eK\u003c/em\u003ei values of Iso.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGRK5 radiometric phosphorylation assays\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the effect of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimerization on receptor phosphorylation, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomers or dimers (1 \u0026mu;M), purified in LMNG micelles or reconstituted in nanodiscs following the protocol described in the sample preparation for cryo-EM, were incubated for 5 min at 30\u0026deg;C with purified C-terminally Strep-tagged GRK5 (50 nM) in a reaction buffer containing 20 mM Tris-HCl, pH 7.4, 5 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 30 mM NaCl, 0.5 mM EDTA, 100 \u0026mu;M [\u0026gamma;\u003csup\u003e32\u003c/sup\u003eP]ATP (1,000 to 2,000 cpm/pmol), and 25 \u0026mu;M BI. The \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR samples in LMNG micelles were additionally supplemented with 20 \u0026mu;M C8-PIP\u003csub\u003e2\u003c/sub\u003e to increase efficiency of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR phosphorylation in detergent. To evaluate the effect of AP on \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR phosphorylation, purified \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomers (1 \u0026mu;M) in LMNG micelles were reconstituted into bicelles with PIP\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e66\u003c/sup\u003e and AP concentration was varied from 0 to 24 \u0026mu;M. Reactions were quenched with SDS sample buffer, and samples were separated by SDS-PAGE. Gels were stained with Coomassie blue (Sigma-Aldrich), dried, exposed to autoradiography film, and \u003csup\u003e32\u003c/sup\u003eP-labeled proteins were excised and counted to determine the amount of phosphate transferred. Reaction rates were normalized to phosphorylation of the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomers (\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR monomers and dimers) or to phosphorylation in the absence of AP (AP effect).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSample preparation for fluorescence measurements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSite-specific fluorophore labeling of \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR was performed using engineered cysteine mutants on a minimal cysteine background (\u0026Delta;6), as previously described\u003csup\u003e24,32\u003c/sup\u003e. For smFRET experiments, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026Delta;6 constructs were cloned into the pcDNA-Zeo-tetO vector and transfected into Expi293 cells stably expressing the tetracycline repressor (Thermo Fisher Scientific, A14635). Transfections were carried out using the Expifectamine kit according to the manufacturer\u0026rsquo;s protocol. Two days post-transfection, receptor expression was induced with 4 \u0026mu;g/mL doxycycline and 5 mM sodium butyrate in the presence of 1 \u0026mu;M alprenolol. Cells were harvested 40 hours after induction and immediately processed for purification.\u003c/p\u003e\n\u003cp\u003eFor studies on the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimer in liposomes, single-cysteine mutants were introduced at TM5 (R228C) or H8 (I334C). Homogeneous dimers were expressed in Sf9 cells and purified as described above. Labeling was performed by incubating 10 \u0026mu;M purified receptor with a 5-fold molar excess of a pre-mixed maleimide-conjugated dye pair: DY549P1 (Dyomics) and Alexa Fluor 647 (Thermo Fisher Scientific) at a 1:1.5 ratio. The reaction was incubated for 30 minutes at RT and quenched with 5 mM L-cysteine. Excess dye was removed by SEC (Superdex 200 Increase 10/300) in 20 mM HEPES (pH 7.4), 150 mM NaCl, and 0.01% LMNG/0.001% CHS. Labeled dimers were reconstituted into liposomes consisting of POPC, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE, Avanti) , and cholesterol at a molar ratio of 6:3:1 using an established protocol\u003csup\u003e67\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFor studies on TM6 dynamics, double-cysteine mutants (N148C on TM4 and L266C on TM6) were introduced. Monomeric receptor was expressed in Expi293 cells and purified following the Sf9 purification procedure. For the dimer, Expi293 cells were co-transfected at a 1:1 plasmid ratio with constructs encoding an 8\u0026times;His-tagged \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026Delta;6 (no FLAG tag) and a FLAG-tagged \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026Delta;6 carrying the N148C/L266C mutations. Heterodimers were isolated via tandem affinity purification. Clarified lysates were first incubated with nickel-chelated Sepharose resins and washed with buffer containing 20 mM imidazole. Proteins were eluted with 250 mM imidazole, then supplemented with 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e and subjected to anti-FLAG M1 immunoaffinity purification. This two-step procedure enriched for heterodimers containing only one protomer with the double-cysteine mutations. Fluorophore labeling was performed as described above for the single-cysteine mutant dimer sample.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnsemble fluorescence measurements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEnsemble FRET experiments were conducted in Fluoromax 4C spectrofluorometer (Horiba Scientific) with excitation and emission slit widths set to 5 nm and 3 nm, respectively. Emission spectra were recorded upon excitation at 532 nm. AP-bound \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR dimers labeled with donor and acceptor fluorophores (I334C H8 sensor) were diluted 1,000-fold in NH buffer plus 0.01% LMNG without AP to a final concentration of 1 nM. Fluorescence spectra were collected at 1, 5, 30, 120 minutes, and 16- and 24-hours post-dilution. All spectra were normalized to donor intensity. To assess the effects of transducer binding, samples were incubated with 100 \u0026mu;M Iso and i) 10 \u0026mu;M Gs (in the presence of apyrase) or ii) 20 \u0026mu;M \u0026beta;-arrestin1(382), together with V\u003csub\u003e2\u003c/sub\u003eRpp and Fab30, which stabilizes the active V\u003csub\u003e2\u003c/sub\u003eRpp-bound \u0026beta;-arrestin1 conformation. \u0026beta;-arrestin1(382) and Fab30 were prepared as previously described\u003csup\u003e65\u003c/sup\u003e. Samples were incubated for 1 hour in the dark to allow full equilibration prior to measurement. All experiments were performed in triplicate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003esmFRET microscopy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFlow chambers for smFRET experiments were assembled using mPEG-passivated glass coverslips (VWR), doped with biotin-PEG16 (Laysan Bio), as described previously\u003csup\u003e24,32\u003c/sup\u003e. Prior to use, coverslips were incubated with 1 mg/mL NeutrAvidin (Thermo Fisher Scientific), followed by 10 nM biotinylated anti-FLAG M1 Fab. Labeled \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR samples were diluted to 100\u0026ndash;500 pM in NH buffer plus 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e and added to the chambers. After achieving optimal surface density, unbound receptor was washed out using imaging buffer supplemented with 100 \u0026mu;M cyclooctatetraene (Sigma-Aldrich) and an oxygen scavenging system (1% D-glucose, 1 mg/mL glucose oxidase, 0.04 mg/mL catalase).\u003c/p\u003e\n\u003cp\u003eFluorescence imaging was performed on a custom-built, objective-based TIRF microscope as reported previously\u003csup\u003e68\u003c/sup\u003e. The setup is built on a Zeiss Axiovert S100 TV platform with a 100\u0026times;, 1.45 NA oil-immersion objective (Zeiss). Donor and acceptor fluorophores were excited with 532 nm and 637 nm lasers (OBIS LS 150 mW and LX 140 mW, Coherent). Emissions were separated by a 652 nm dichroic beamsplitter (Semrock), filtered through 580/60 nm and 731/137 nm bandpass filters, and split using an OptoSplit II beamsplitter (Cairn Research) onto an EMCCD camera (iXon DU897E, Andor). Data acquisition was controlled by \u0026mu;Manager via custom BeanShell scripts, and movies were recorded as stacked TIFFs in frame-transfer mode at 100 ms exposure per frame. Laser power was tuned to balance high signal-to-noise with photobleaching timescales of tens of seconds. Each slide typically yielded 10\u0026ndash;20 movies per channel. All imaging was performed at RT.\u003c/p\u003e\n\u003cp\u003eFluorescence traces were analyzed using custom Python scripts. Donor and acceptor channels were aligned using registration images, and individual molecules were identified as local intensity maxima within a 5-pixel neighborhood. Donor-only spots were excluded. For each fluorophore pair, intensities were background-corrected using a local circular region (35-pixel diameter). Donor leakage into the acceptor channel (~7%) was subtracted.\u003c/p\u003e\n\u003cp\u003eDonor excitation was used to monitor emission for 80 seconds, followed by direct acceptor excitation for 1 second to confirm fluorophore identity. Traces were selected for analysis based on the following criteria: (1) signal-to-noise ratio \u0026ge;5; (2) single-step acceptor photobleaching prior to donor bleaching; (3) \u0026gamma; factor between 0.5 and 2.5; (4) anticorrelated donor and acceptor intensity fluctuations; and (5) single-step donor bleaching, if present.\u003c/p\u003e\n\u003cp\u003eFRET efficiency (\u003cem\u003eE\u003c/em\u003e) was calculated as \u003cem\u003eE = I\u003csub\u003ea\u003c/sub\u003e/(I\u003csub\u003ea\u003c/sub\u003e+\u0026gamma;I\u003csub\u003ed\u003c/sub\u003e)\u003c/em\u003e, where \u003cem\u003eI\u003csub\u003ea\u003c/sub\u003e\u003c/em\u003e and \u003cem\u003eI\u003csub\u003ed\u003c/sub\u003e\u003c/em\u003e are the background-corrected acceptor and donor intensities, respectively. \u0026gamma;-correction was applied as described previously. For each trace, FRET values were binned into 30 intervals across the range [\u0026ndash;0.25, 1.25] and normalized by the total number of data points. Ensemble FRET histograms were generated by averaging the normalized histograms from individual molecules and fit to a two-Gaussian distribution model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSample preparation for DEER\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor DEER measurements, \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026Delta;6-N148C/L266C was expressed and purified as described above in Sf9 cells. To exchange detergent from 0.1% DDM/0.01% CHS to 0.01% (w/v) LMNG/0.001% CHS, the receptor was extensively washed with a progressive gradient of DDM: LMNG buffer. In parallel, while the receptor was bound to the resin, alprenolol was removed by washing with saturating concentrations of the low-affinity antagonist atenolol. Because of the fast dissociation kinetics of atenolol from the \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR, subsequent washes with ligand-free buffer yielded unliganded \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR for spin labeling. The flag eluted receptor was labeled with the spin label reagent IAP in the presence of 100 \u0026mu;M TCEP in buffer containing 20 mM HEPES, pH 7.4, 150 mM NaCl, and 0.01% LMNG/0.001% CHS. Twenty-fold molar excess of 3-(2-iodoacetamido)-proxyl (IAP) was added to 10 \u0026mu;M \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR\u0026Delta;6 receptor for 3 hours at RT. After quenching of the reaction with 5 mM final L-cysteine, the receptor was separated from the excess spin label by SEC (Superdex 200 10/300) in SEC buffer (20 mM HEPES, pH 7.4, 150 mM NaCl, and 0.01% LMNG/0.001% CHS) prepared with D\u003csub\u003e2\u003c/sub\u003eO. The sample was concentrated using a 50 kDa concentrator to a concentration \u0026gt; 25 \u0026mu;M. D8-glycerol was added as a cryoprotectant to 25 % (v/v). 13 \u0026mu;L of sample was added to a borosilicate capillary 1.4 mm ID \u0026times; 1.7 mm OD (VitroCom, Inc) and flash-frozen in liquid nitrogen.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDEER spectroscopy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDEER experiments were conducted as previously described\u003csup\u003e32\u003c/sup\u003e at Q-band (~33.68 GHz) using a Bruker Elexsys 580 spectrometer equipped with a SpinJet AWG, EN5107D2 resonator, variable-temperature cryogen-free cooling system (ColdEdge Technologies Inc.), and a 300 W TWT amplifier (Applied Systems Engineering Inc.). All measurements were performed at 50 K. Dipolar evolution data were acquired using a dead-time-free 4-pulse DEER sequence with gaussian pulses\u003csup\u003e69\u003c/sup\u003e and with 16-step phase cycling.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe experimental parameters used for DEER data collection were: \u0026pi;/2, \u0026pi;\u003csub\u003eobs\u003c/sub\u003e, and \u0026pi;\u003csub\u003epump\u003c/sub\u003e pulse lengths of 40 ns; a frequency offset (\u0026Delta;v) of 90 MHz; d1 = 250 ns; d2 = 5150 ns; shot repetition time = 2000 \u0026mu;s; shots per point = 4; and integration window = 40 ns. The optimal microwave power (i.e., pulse amplitude) for the \u0026pi;/2, \u0026pi;\u003csub\u003eobs\u003c/sub\u003e, and \u0026pi;\u003csub\u003epump\u003c/sub\u003e pulses were determined using transient nutation experiments, where pulse amplitudes were adjusted to maximize the inversion of the Hahn echo\u003csup\u003e70\u003c/sup\u003e. Pump pulses were applied to the maximum intensity of the field swept echo detected absorption spectrum. Observe pulses were applied at a frequency 90 MHz lower than the pump pulses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDEER data were processed using DeerAnalysis 2022\u003csup\u003e71\u003c/sup\u003e, which employs two fitting routines: neural network analysis (DEERNet\u003csup\u003e72\u003c/sup\u003e, Spinach revision 5662) and Tikhonov regularization (DeerLab 0.9.1)\u003csup\u003e73\u003c/sup\u003e. The consensus fit represents the mean of both methods, with reported 95% confidence intervals also incorporating errors from both methods. Time traces were normalized to signal intensity at t = 0, and distance distributions were area normalized. Custom Python scripts were used for plotting the dipolar evolution time traces and the distance distributions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe cryo-EM map for \u0026beta;\u003csub\u003e2\u003c/sub\u003eAR_dimer(AP) in nanodisc has been deposited to the Electron Microscopy Data Bank under the accession code EMD-72202. The corresponding atomic model has been deposited in the Protein Data Bank under the accession code 9Q3L. Any additional data reported in this paper are available from the lead contact, Brian K. Kobilka ([email protected]), upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank R. Qiu of R. S. Lewis\u0026rsquo;s lab for help with smFRET experiments, E. S. Bruguera for instructions on BLI measurements, M. R. Eckart and J. Tran of Stanford Protein and Nucleic Acid (PAN) Facility for SPR data collection, P. A. N. Reddy and J. M. Salvino for providing AP-7-168, and B. Singal and C. Zhang for support on cryo-EM data collection at the Stanford cryo-EM center (cEMc) and Stanford-SLAC Cryo-EM Center (S2C2), which is supported by the US National Institute of General Medical Sciences (1R24GM154186). We also thank M. Bouvier for providing pcDNA-\u0026beta;-arrestin2-GFP10 and pcDNA3-\u0026beta;\u003csub\u003e2\u003c/sub\u003eAR-RlucII, L. I. Jiang for providing the cAMP intramolecular BRET sensor CAMYEL, G. Milligan for providing pcDNA3-\u0026beta;\u003csub\u003e1\u003c/sub\u003eAR-GFP, and P. S. Chae for providing TTG-T10 detergent. J. Xu is an investigator of SUSTech Institute for Biological Electron Microscopy. This research was supported by National Institutes of Health (NIH) awards R35NS137408 (B.K.K.), R01GM083118 (B.K.K.), P01HL114471 (J.L.B.), R01AI161296 (J.L.B.), R01GM135581 (M.T.L.), S10OD025260 (M.T.L.), and American Heart Association (AHA) Postdoctoral Fellowship 25POST1411512 (J.S.).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, J.S., T.N.P., J.X., J.L.B., and B.K.K.; methodology, J.S., T.N.P., J.X., K.E.K., F.D.P., and A.M.G.; investigation: J.S., T.N.P., J.X., K.E.K, F.D.P, A.M.G., and H.W.; data curation and formal analysis, J.S., T.N.P., J.X., K.E.K., F.D.P., and A.M.G.; writing \u0026ndash; original draft: J.S., J.X., and, T.N.P.; writing \u0026ndash; review \u0026amp; editing, all authors; supervision and funding acquisition, B.K.K., J.L.B., and M.T.L.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA patent on the reported compounds was submitted by J.L.B. and others in 2022. B.K.K. is a cofounder of and consultant for ConfometRx Inc. All other authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eZhao, D. Y.\u003cem\u003e et al.\u003c/em\u003e Cryo-EM structure of the native rhodopsin dimer in nanodiscs. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e294\u003c/strong\u003e, 14215-14230, doi:10.1074/jbc.RA119.010089 (2019).\u003c/li\u003e\n\u003cli\u003eYue, Y.\u003cem\u003e et al.\u003c/em\u003e Structural insight into apelin receptor-G protein stoichiometry. \u003cem\u003eNat Struct Mol Biol\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 688-697, doi:10.1038/s41594-022-00797-5 (2022).\u003c/li\u003e\n\u003cli\u003eChang, H.\u003cem\u003e et al.\u003c/em\u003e Structural basis of oligomerization-modulated activation and autoinhibition of orphan receptor GPR3. \u003cem\u003eCell Rep\u003c/em\u003e \u003cstrong\u003e44\u003c/strong\u003e, 115478, doi:10.1016/j.celrep.2025.115478 (2025).\u003c/li\u003e\n\u003cli\u003eIppolito, M.\u003cem\u003e et al.\u003c/em\u003e Identification of a \u0026beta;-arrestin-biased negative allosteric modulator for the \u0026beta;2-adrenergic receptor. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, e2302668120, doi:10.1073/pnas.2302668120 (2023).\u003c/li\u003e\n\u003cli\u003eHilger, D. The role of structural dynamics in GPCR‐mediated signaling. \u003cem\u003eThe FEBS Journal\u003c/em\u003e \u003cstrong\u003e288\u003c/strong\u003e, 2461-2489, doi:10.1111/febs.15841 (2021).\u003c/li\u003e\n\u003cli\u003eWootten, D., Christopoulos, A., Marti-Solano, M., Babu, M. M. \u0026amp; Sexton, P. M. Mechanisms of signalling and biased agonism in G protein-coupled receptors. \u003cem\u003eNat Rev Mol Cell Biol\u003c/em\u003e, doi:10.1038/s41580-018-0049-3 (2018).\u003c/li\u003e\n\u003cli\u003eFredriksson, R., Lagerstrom, M. C., Lundin, L. G. \u0026amp; Schioth, H. B. The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. \u003cem\u003eMol Pharmacol\u003c/em\u003e \u003cstrong\u003e63\u003c/strong\u003e, 1256-1272, doi:10.1124/mol.63.6.1256 (2003).\u003c/li\u003e\n\u003cli\u003eEllaithy, A., Gonzalez-Maeso, J., Logothetis, D. A. \u0026amp; Levitz, J. Structural and Biophysical Mechanisms of Class C G Protein-Coupled Receptor Function. \u003cem\u003eTrends in biochemical sciences\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 1049-1064, doi:10.1016/j.tibs.2020.07.008 (2020).\u003c/li\u003e\n\u003cli\u003eZeng, F. Y. \u0026amp; Wess, J. Molecular aspects of muscarinic receptor dimerization. \u003cem\u003eNeuropsychopharmacology\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, S19-S31, doi:Doi 10.1016/S0893-133x(00)00146-9 (2000).\u003c/li\u003e\n\u003cli\u003eAngers, S.\u003cem\u003e et al.\u003c/em\u003e Detection of \u0026beta;2 adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). \u003cem\u003eP Natl Acad Sci USA\u003c/em\u003e \u003cstrong\u003e97\u003c/strong\u003e, 3684-3689, doi:DOI 10.1073/pnas.060590697 (2000).\u003c/li\u003e\n\u003cli\u003eHiller, C., Kuhhorn, J. \u0026amp; Gmeiner, P. Class A G-protein-coupled receptor (GPCR) dimers and bivalent ligands. \u003cem\u003eJournal of medicinal chemistry\u003c/em\u003e \u003cstrong\u003e56\u003c/strong\u003e, 6542-6559, doi:10.1021/jm4004335 (2013).\u003c/li\u003e\n\u003cli\u003eYue, Y.\u003cem\u003e et al.\u003c/em\u003e Structural insights into the regulation of monomeric and dimeric apelin receptor. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 310, doi:10.1038/s41467-024-55555-6 (2025).\u003c/li\u003e\n\u003cli\u003eHuang, B. S., St Onge, C. M., Ma, H. G. \u0026amp; Zhang, Y. Design of bivalent ligands targeting putative GPCR dimers. \u003cem\u003eDrug discovery today\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 189-199, doi:10.1016/j.drudis.2020.10.006 (2020).\u003c/li\u003e\n\u003cli\u003eUllmann, T.\u003cem\u003e et al.\u003c/em\u003e Homobivalent Dopamine D(2) Receptor Ligands Modulate the Dynamic Equilibrium of D(2) Monomers and Homo- and Heterodimers. \u003cem\u003eACS Chem Biol\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 371-379, doi:10.1021/acschembio.0c00895 (2021).\u003c/li\u003e\n\u003cli\u003eRasmussen, S. G.\u003cem\u003e et al.\u003c/em\u003e Structure of a nanobody-stabilized active state of the \u0026beta;2 adrenoceptor. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e469\u003c/strong\u003e, 175-180, doi:10.1038/nature09648 (2011).\u003c/li\u003e\n\u003cli\u003eLiu, X.\u003cem\u003e et al.\u003c/em\u003e An allosteric modulator binds to a conformational hub in the \u0026beta;2 adrenergic receptor. \u003cem\u003eNature chemical biology\u003c/em\u003e, doi:10.1038/s41589-020-0549-2 (2020).\u003c/li\u003e\n\u003cli\u003eCherezov, V.\u003cem\u003e et al.\u003c/em\u003e High-Resolution Crystal Structure of an Engineered Human \u0026beta;2-Adrenergic G Protein\u0026ndash;Coupled Receptor. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e318\u003c/strong\u003e, 1258-1265, doi:10.1126/science.1150577 (2007).\u003c/li\u003e\n\u003cli\u003eJi, S.-Y.\u003cem\u003e et al.\u003c/em\u003e Dynamic Monomer-Dimer Transition in Ligand-induced Apelin Receptor Activation. \u003cem\u003ebioRxiv\u003c/em\u003e, 2025.2002.2019.639188, doi:10.1101/2025.02.19.639188 (2025).\u003c/li\u003e\n\u003cli\u003eYue, Y.\u003cem\u003e et al.\u003c/em\u003e Mechanistic insights into the versatile stoichiometry and biased signaling of the apelin receptor-arrestin complex. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 7403, doi:10.1038/s41467-025-62870-z (2025).\u003c/li\u003e\n\u003cli\u003eQiu, Z.\u003cem\u003e et al.\u003c/em\u003e Vol. 14 RP107185 (eLife, 2025).\u003c/li\u003e\n\u003cli\u003eChen, G.\u003cem\u003e et al.\u003c/em\u003e Mechanism and function of GPR3 regulated by a negative allosteric modulator. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 7988, doi:10.1038/s41467-025-63422-1 (2025).\u003c/li\u003e\n\u003cli\u003eHuang, J., Chen, S., Zhang, J. J. \u0026amp; Huang, X. Y. Crystal structure of oligomeric \u0026beta;1-adrenergic G protein-coupled receptors in ligand-free basal state. \u003cem\u003eNat Struct Mol Biol\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 419-425, doi:10.1038/nsmb.2504 (2013).\u003c/li\u003e\n\u003cli\u003eFung, J. J.\u003cem\u003e et al.\u003c/em\u003e Ligand-regulated oligomerization of beta(2)-adrenoceptors in a model lipid bilayer. \u003cem\u003eEMBO J\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 3315-3328, doi:10.1038/emboj.2009.267 (2009).\u003c/li\u003e\n\u003cli\u003eGregorio, G. G.\u003cem\u003e et al.\u003c/em\u003e Single-molecule analysis of ligand efficacy in \u0026beta;2AR-G-protein activation. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e547\u003c/strong\u003e, 68-73, doi:10.1038/nature22354 (2017).\u003c/li\u003e\n\u003cli\u003eSungkaworn, T.\u003cem\u003e et al.\u003c/em\u003e Single-molecule imaging reveals receptor-G protein interactions at cell surface hot spots. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e550\u003c/strong\u003e, 543-547, doi:10.1038/nature24264 (2017).\u003c/li\u003e\n\u003cli\u003eCalebiro, D. \u0026amp; Koszegi, Z. The subcellular dynamics of GPCR signaling. \u003cem\u003eMol Cell Endocrinol\u003c/em\u003e \u003cstrong\u003e483\u003c/strong\u003e, 24-30, doi:10.1016/j.mce.2018.12.020 (2019).\u003c/li\u003e\n\u003cli\u003eRasmussen, S. G.\u003cem\u003e et al.\u003c/em\u003e Crystal structure of the \u0026beta;2 adrenergic receptor\u0026ndash;Gs protein complex. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e477\u003c/strong\u003e, 549-555, doi:10.1038/nature10361 (2011).\u003c/li\u003e\n\u003cli\u003eLee, Y.\u003cem\u003e et al.\u003c/em\u003e Molecular basis of \u0026beta;-arrestin coupling to formoterol-bound \u0026beta;1-adrenoceptor \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e583\u003c/strong\u003e, 862-866, doi:10.1038/s41586-020-2419-1 (2020).\u003c/li\u003e\n\u003cli\u003eDuan, J.\u003cem\u003e et al.\u003c/em\u003e GPCR activation and GRK2 assembly by a biased intracellular agonist. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e620\u003c/strong\u003e, 676-681, doi:10.1038/s41586-023-06395-9 (2023).\u003c/li\u003e\n\u003cli\u003eStaus, D. P.\u003cem\u003e et al.\u003c/em\u003e Sortase ligation enables homogeneous GPCR phosphorylation to reveal diversity in \u0026beta;-arrestin coupling. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e115\u003c/strong\u003e, 3834-3839, doi:10.1073/pnas.1722336115 (2018).\u003c/li\u003e\n\u003cli\u003eLi, Y.\u003cem\u003e et al.\u003c/em\u003e Chemical Synthesis of a Full-Length G-Protein-Coupled Receptor \u0026beta;2-Adrenergic Receptor with Defined Modification Patterns at the C-Terminus. \u003cem\u003eJ Am Chem Soc\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 17566-17576, doi:10.1021/jacs.1c07369 (2021).\u003c/li\u003e\n\u003cli\u003eCasiraghi, M.\u003cem\u003e et al.\u003c/em\u003e Structure and dynamics determine G protein coupling specificity at a class A GPCR. \u003cem\u003eSci Adv\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, eadq3971, doi:10.1126/sciadv.adq3971 (2025).\u003c/li\u003e\n\u003cli\u003eZhao, J.\u003cem\u003e et al.\u003c/em\u003e Ligand efficacy modulates conformational dynamics of the \u0026micro;-opioid receptor. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e629\u003c/strong\u003e, 474-480, doi:10.1038/s41586-024-07295-2 (2024).\u003c/li\u003e\n\u003cli\u003eLiu, J.\u003cem\u003e et al.\u003c/em\u003e Biased signaling due to oligomerization of the G protein-coupled platelet-activating factor receptor. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6365, doi:10.1038/s41467-022-34056-4 (2022).\u003c/li\u003e\n\u003cli\u003eParadis, J. S.\u003cem\u003e et al.\u003c/em\u003e Computationally designed GPCR quaternary structures bias signaling pathway activation. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6826, doi:10.1038/s41467-022-34382-7 (2022).\u003c/li\u003e\n\u003cli\u003eZhang, Y.\u003cem\u003e et al.\u003c/em\u003e Higher-order transient membrane protein structures. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2421275121, doi:10.1073/pnas.2421275121 (2025).\u003c/li\u003e\n\u003cli\u003eZhang, Y. \u0026amp; MacKinnon, R. Higher-order transient structures and the principle of dynamic connectivity in membrane signaling. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2421280121, doi:10.1073/pnas.2421280121 (2025).\u003c/li\u003e\n\u003cli\u003eMandala, V. S., Fu, Z. \u0026amp; MacKinnon, R. Molecular contacts in self-assembling clusters of membrane proteins. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2507112122, doi:10.1073/pnas.2507112122 (2025).\u003c/li\u003e\n\u003cli\u003eSchreiber, S. L. The Rise of Molecular Glues. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e184\u003c/strong\u003e, 3-9, doi:10.1016/j.cell.2020.12.020 (2021).\u003c/li\u003e\n\u003cli\u003eSchreiber, S. L. Molecular glues and bifunctional compounds: Therapeutic modalities based on induced proximity. \u003cem\u003eCell Chem Biol\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 1050-1063, doi:10.1016/j.chembiol.2024.05.004 (2024).\u003c/li\u003e\n\u003cli\u003eRoth, B. L. \u0026amp; Krumm, B. E. Molecular glues as potential GPCR therapeutics. \u003cem\u003eBiochem Pharmacol\u003c/em\u003e \u003cstrong\u003e228\u003c/strong\u003e, 116402, doi:10.1016/j.bcp.2024.116402 (2024).\u003c/li\u003e\n\u003cli\u003eKrumm, B. E.\u003cem\u003e et al.\u003c/em\u003e Neurotensin Receptor Allosterism Revealed in Complex with a Biased Allosteric Modulator. \u003cem\u003eBiochemistry\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e, 1233-1248, doi:10.1021/acs.biochem.3c00029 (2023).\u003c/li\u003e\n\u003cli\u003eZhao, L. H.\u003cem\u003e et al.\u003c/em\u003e Conserved class B GPCR activation by a biased intracellular agonist. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e621\u003c/strong\u003e, 635-641, doi:10.1038/s41586-023-06467-w (2023).\u003c/li\u003e\n\u003cli\u003eSun, D.\u003cem\u003e et al.\u003c/em\u003e Molecular mechanism of the arrestin-biased agonism of neurotensin receptor 1 by an intracellular allosteric modulator. \u003cem\u003eCell Res\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 284-295, doi:10.1038/s41422-025-01095-7 (2025).\u003c/li\u003e\n\u003cli\u003eHe, G.\u003cem\u003e et al.\u003c/em\u003e A GPCR-G protein-\u0026beta;-arrestin megacomplex induced by an allosteric modulator. \u003cem\u003ebioRxiv\u003c/em\u003e, 2025.2003.2024.645131, doi:10.1101/2025.03.24.645131 (2025).\u003c/li\u003e\n\u003cli\u003eLi, J.\u003cem\u003e et al.\u003c/em\u003e Activation of human STING by a molecular glue-like compound. \u003cem\u003eNat Chem Biol\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 365-372, doi:10.1038/s41589-023-01434-y (2024).\u003c/li\u003e\n\u003cli\u003eStrohman, M. J.\u003cem\u003e et al.\u003c/em\u003e Local membrane charge regulates \u0026beta;2 adrenergic receptor coupling to Gi3. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 2234, doi:10.1038/s41467-019-10108-0 (2019).\u003c/li\u003e\n\u003cli\u003eHilger, D.\u003cem\u003e et al.\u003c/em\u003e Structural insights into differences in G protein activation by family A and family B GPCRs. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e369\u003c/strong\u003e, doi:10.1126/science.aba3373 (2020).\u003c/li\u003e\n\u003cli\u003eKrishna Kumar, K.\u003cem\u003e et al.\u003c/em\u003e Negative allosteric modulation of the glucagon receptor by RAMP2. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e186\u003c/strong\u003e, 1465-1477 e1418, doi:10.1016/j.cell.2023.02.028 (2023).\u003c/li\u003e\n\u003cli\u003ePunjani, A., Rubinstein, J. L., Fleet, D. J. \u0026amp; Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. \u003cem\u003eNat Methods\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 290-296, doi:10.1038/nmeth.4169 (2017).\u003c/li\u003e\n\u003cli\u003ePunjani, A., Zhang, H. \u0026amp; Fleet, D. J. Non-uniform refinement: adaptive regularization improves single-particle cryo-EM reconstruction. \u003cem\u003eNat Methods\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1214-1221, doi:10.1038/s41592-020-00990-8 (2020).\u003c/li\u003e\n\u003cli\u003eScheres, S. H. RELION: implementation of a Bayesian approach to cryo-EM structure determination. \u003cem\u003eJ Struct Biol\u003c/em\u003e \u003cstrong\u003e180\u003c/strong\u003e, 519-530, doi:10.1016/j.jsb.2012.09.006 (2012).\u003c/li\u003e\n\u003cli\u003eKimanius, D., Dong, L., Sharov, G., Nakane, T. \u0026amp; Scheres, S. H. W. New tools for automated cryo-EM single-particle analysis in RELION-4.0. \u003cem\u003eBiochem J\u003c/em\u003e \u003cstrong\u003e478\u003c/strong\u003e, 4169-4185, doi:10.1042/BCJ20210708 (2021).\u003c/li\u003e\n\u003cli\u003eZivanov, J., Nakane, T. \u0026amp; Scheres, S. H. W. A Bayesian approach to beam-induced motion correction in cryo-EM single-particle analysis. \u003cem\u003eIUCrJ\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 5-17, doi:10.1107/S205225251801463X (2019).\u003c/li\u003e\n\u003cli\u003ePettersen, E. F.\u003cem\u003e et al.\u003c/em\u003e UCSF Chimera--a visualization system for exploratory research and analysis. \u003cem\u003eJournal of computational chemistry\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 1605-1612, doi:10.1002/jcc.20084 (2004).\u003c/li\u003e\n\u003cli\u003eEmsley, P., Lohkamp, B., Scott, W. G. \u0026amp; Cowtan, K. Features and development of Coot. \u003cem\u003eActa crystallographica. Section D, Biological crystallography\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 486-501, doi:10.1107/S0907444910007493 (2010).\u003c/li\u003e\n\u003cli\u003eAdams, P. D.\u003cem\u003e et al.\u003c/em\u003e PHENIX: a comprehensive Python-based system for macromolecular structure solution. \u003cem\u003eActa crystallographica. Section D, Biological crystallography\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 213-221, doi:10.1107/S0907444909052925 (2010).\u003c/li\u003e\n\u003cli\u003eAfonine, P. V.\u003cem\u003e et al.\u003c/em\u003e Real-space refinement in PHENIX for cryo-EM and crystallography. \u003cem\u003eActa Crystallogr D Struct Biol\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 531-544, doi:10.1107/S2059798318006551 (2018).\u003c/li\u003e\n\u003cli\u003eBarad, B. A.\u003cem\u003e et al.\u003c/em\u003e EMRinger: side chain-directed model and map validation for 3D cryo-electron microscopy. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 943-946, doi:10.1038/nmeth.3541 (2015).\u003c/li\u003e\n\u003cli\u003ePettersen, E. F.\u003cem\u003e et al.\u003c/em\u003e UCSF ChimeraX: Structure visualization for researchers, educators, and developers. \u003cem\u003eProtein Sci\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 70-82, doi:10.1002/pro.3943 (2021).\u003c/li\u003e\n\u003cli\u003eSonn-Segev, A.\u003cem\u003e et al.\u003c/em\u003e Quantifying the heterogeneity of macromolecular machines by mass photometry. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1772, doi:10.1038/s41467-020-15642-w (2020).\u003c/li\u003e\n\u003cli\u003ePeitsch, C. F., Beckmann, S. \u0026amp; Zuber, B. iMEM: Isolation of Plasma Membrane for Cryoelectron Microscopy. \u003cem\u003eStructure\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 2198-2206, doi:10.1016/j.str.2016.09.016 (2016).\u003c/li\u003e\n\u003cli\u003eJiang, L. I.\u003cem\u003e et al.\u003c/em\u003e Use of a cAMP BRET sensor to characterize a novel regulation of cAMP by the sphingosine 1-phosphate/G13 pathway. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e282\u003c/strong\u003e, 10576-10584, doi:10.1074/jbc.M609695200 (2007).\u003c/li\u003e\n\u003cli\u003eChen, I., Dorr, B. M. \u0026amp; Liu, D. R. A general strategy for the evolution of bond-forming enzymes using yeast display. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e108\u003c/strong\u003e, 11399-11404, doi:10.1073/pnas.1101046108 (2011).\u003c/li\u003e\n\u003cli\u003eJanetzko, J.\u003cem\u003e et al.\u003c/em\u003e Membrane phosphoinositides regulate GPCR-\u0026beta;-arrestin complex assembly and dynamics \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e185\u003c/strong\u003e, 4560-4573 e4519, doi:10.1016/j.cell.2022.10.018 (2022).\u003c/li\u003e\n\u003cli\u003eKomolov, K. E.\u003cem\u003e et al.\u003c/em\u003e Structural and Functional Analysis of a \u0026beta;2-Adrenergic Receptor Complex with GRK5. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e169\u003c/strong\u003e, 407-421 e416, doi:10.1016/j.cell.2017.03.047 (2017).\u003c/li\u003e\n\u003cli\u003eGoddard, A. D., Dijkman, P. M., Adamson, R. J., dos Reis, R. I. \u0026amp; Watts, A. Reconstitution of membrane proteins: a GPCR as an example. \u003cem\u003eMethods Enzymol\u003c/em\u003e \u003cstrong\u003e556\u003c/strong\u003e, 405-424, doi:10.1016/bs.mie.2015.01.004 (2015).\u003c/li\u003e\n\u003cli\u003evan Dorp, S.\u003cem\u003e et al.\u003c/em\u003e Conformational dynamics of auto-inhibition in the ER calcium sensor STIM1. \u003cem\u003eElife\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, doi:10.7554/eLife.66194 (2021).\u003c/li\u003e\n\u003cli\u003eTeucher, M. \u0026amp; Bordignon, E. Improved signal fidelity in 4-pulse DEER with Gaussian pulses. \u003cem\u003eJ Magn Reson\u003c/em\u003e \u003cstrong\u003e296\u003c/strong\u003e, 103-111, doi:10.1016/j.jmr.2018.09.003 (2018).\u003c/li\u003e\n\u003cli\u003eTeucher, M.\u003cem\u003e et al.\u003c/em\u003e Strategies to identify and suppress crosstalk signals in double electron-electron resonance (DEER) experiments with gadolinium(III) and nitroxide spin-labeled compounds. \u003cem\u003eMagn Reson (Gott)\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 285-299, doi:10.5194/mr-1-285-2020 (2020).\u003c/li\u003e\n\u003cli\u003eJeschke, G.\u003cem\u003e et al.\u003c/em\u003e DeerAnalysis2006 - a comprehensive software package for analyzing pulsed ELDOR data. \u003cem\u003eAppl Magn Reson\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 473-498, doi:10.1007/Bf03166213 (2006).\u003c/li\u003e\n\u003cli\u003eWorswick, S. G., Spencer, J. A., Jeschke, G. \u0026amp; Kuprov, I. Deep neural network processing of DEER data. \u003cem\u003eSci Adv\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, eaat5218, doi:10.1126/sciadv.aat5218 (2018).\u003c/li\u003e\n\u003cli\u003eFabregas Ibanez, L., Jeschke, G. \u0026amp; Stoll, S. DeerLab: a comprehensive software package for analyzing dipolar electron paramagnetic resonance spectroscopy data. \u003cem\u003eMagn Reson (Gott)\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 209-224, doi:10.5194/mr-1-209-2020 (2020).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8001844/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8001844/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Family A G-protein coupled receptors (GPCRs) are typically described as monomers, yet growing evidence suggests they can form dimers with distinct signaling properties1-3. The mechanisms and therapeutic potential of such dimerization, however, remain poorly understood. Here, we show that AP-7-168, an optimized derivative of a β-arrestin-biased negative allosteric modulator of the β2-adrenergic receptor (β2AR) that sustains bronchorelaxation in cell and tissue models4, functions as a molecular glue to promote β2AR homodimerization. Cryo-EM structures reveal a unique binding mode in which two AP-7-168 molecules pack within a pocket formed by transmembrane helices 3, 4, and 5 of two protomers, stabilizing a dimeric conformation that selectively prevents β-arrestin coupling. In cells, AP-7-168 robustly induces β2AR dimerization and drives enlarged nanocluster formation. Combined with extensive functional studies, our findings unveil a novel allosteric mechanism by which a small molecule biases β2AR signaling through dimerization, highlighting ligand-induced dimerization as a strategy for GPCR modulation.","manuscriptTitle":"A biased allosteric modulator functions as a molecular glue to induce β2AR dimerization","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-26 06:58:36","doi":"10.21203/rs.3.rs-8001844/v1","editorialEvents":[],"status":"published","journal":{"display":false,"email":"[email protected]","identity":"nature","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nature","sideBox":"Learn more about [Nature](http://www.nature.com/nature/)","snPcode":"","submissionUrl":"","title":"Nature","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"facbe7b2-c222-4d4d-8bfa-20c59d46cdc3","owner":[],"postedDate":"November 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":58626493,"name":"Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy"},{"id":58626494,"name":"Biological sciences/Biochemistry/Proteins/G protein-coupled receptors"}],"tags":[],"updatedAt":"2026-05-05T02:45:28+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-26 06:58:36","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8001844","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8001844","identity":"rs-8001844","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}