Structural basis of μ-opioid receptor targeting by a nanobody antagonist

Yu, Jun; Kumar, Amit; Zhang, Xuefeng; Martin, Charlotte; Van holsbeeck, Kevin; Raia, Pierre; Koehl, Antoine; Laeremans, Toon; Steyaert, Jan; Manglik, Aashish; Ballet, Steven; Boland, Andreas; Stoeber, Miriam

doi:10.1038/s41467-024-52947-6

Download PDF

Article
Open access
Published: 09 October 2024

Structural basis of μ-opioid receptor targeting by a nanobody antagonist

Nature Communications volume 15, Article number: 8687 (2024) Cite this article

12k Accesses
15 Citations
110 Altmetric
Metrics details

Subjects

Abstract

The μ-opioid receptor (μOR), a prototypical G protein-coupled receptor (GPCR), is the target of opioid analgesics such as morphine and fentanyl. Due to the severe side effects of current opioid drugs, there is considerable interest in developing novel modulators of μOR function. Most GPCR ligands today are small molecules, however biologics, including antibodies and nanobodies, represent alternative therapeutics with clear advantages such as affinity and target selectivity. Here, we describe the nanobody NbE, which selectively binds to the μOR and acts as an antagonist. We functionally characterize NbE as an extracellular and genetically encoded μOR ligand and uncover the molecular basis for μOR antagonism by determining the cryo-EM structure of the NbE-μOR complex. NbE displays a unique ligand binding mode and achieves μOR selectivity by interactions with the orthosteric pocket and extracellular receptor loops. Based on a β-hairpin loop formed by NbE that deeply protrudes into the μOR, we design linear and cyclic peptide analogs that recapitulate NbE’s antagonism. The work illustrates the potential of nanobodies to uniquely engage with GPCRs and describes lower molecular weight μOR ligands that can serve as a basis for therapeutic developments.

A µ-opioid receptor modulator that works cooperatively with naloxone

Article 03 July 2024

Profiling the proximal proteome of the activated μ-opioid receptor

Article 25 March 2024

Design and structural validation of peptide–drug conjugate ligands of the kappa-opioid receptor

Article Open access 06 December 2023

Introduction

G protein-coupled receptors (GPCRs) represent key therapeutic targets due to their central roles in cellular signaling and control over a plethora of physiological processes. Developing new ligands that bind a given GPCR with high selectivity remains a significant challenge in drug discovery^1,2,3. Small molecule ligands have historically dominated the landscape of GPCR-targeted drugs, but recently biologics, including antibodies and nanobodies (Nbs), have emerged as an alternative class of ligands that offer distinct advantages and hold promise for therapeutic developments^4,5. Nbs are single-___domain antibody fragments derived from heavy chain-only antibodies, which naturally occur in camelids and cartilaginous fish, and are characterized by small size, high antigen binding affinity, and binding loops that can access deep cavities on target proteins⁶. Nbs can show enhanced selectivity over small molecules due to their ability to interact with unique and extended epitope surfaces. Over the last decade, Nbs that bind GPCRs on their intracellular side have served as innovative research tools to uncover GPCR signal transduction mechanisms^7,8. For example, Nbs were used as crystallization chaperones or as fiducial markers in high-resolution structural studies^9,10,11,12. Conformation-selective Nbs were also repurposed into biosensors to report GPCR activity in living cells^13,14. Only recently, several Nbs that bind GPCRs as extracellular ligands and thereby modulate receptor function have been described^{15,16,17,18,19,20}. Generating knowledge on GPCR-targeting Nbs is key to unlocking their potential as both versatile research tools and therapeutic compounds.

Opioid receptors (ORs) are prototypical members of the rhodopsin-like GPCR family and function in pain modulation and analgesia^21,22. The OR family comprises four major receptor subtypes, including the μOR, δOR, κOR, and nociceptin-OR (NOPR), with the μOR representing the prime therapeutic target for pain relief. Approved drugs that target μORs are diverse small molecule compounds, including the widely used analgesics morphine and fentanyl. The ligand repertoire has recently been expanded through structure-based molecular docking, rational design, and high-throughput screening, delivering new OR ligands with distinct pharmacological profiles, including biased agonism, receptor subtype selectivity, and pharmacokinetic properties^23,24,25. Discovering additional and innovative modulators of μOR function, including agonists and antagonists, remains a pressing necessity for developing improved analgesics and compounds that can reduce or reverse the deleterious opioid side effects²⁶. Until now, no antibody or Nb ligand for ORs has been characterized in-depth, representing a hurdle in exploiting the unique features of biologics to effectively target ORs.

In this study, we functionally and structurally analyzed NbE, an extracellular μOR-targeting Nb ligand⁹. Using cellular binding and signaling assays, we identified that NbE selectively binds to μOR with nanomolar affinity and acts as an antagonist when added as ligand or expressed as a genetically-encoded cell surface displayed protein. Using cryogenic electron microscopy (cryo-EM), we then solved the structure of the NbE-μOR complex, which reveals that NbE deeply inserts into the orthosteric pocket and additionally interacts with two extracellular loops (ECL) of the μOR. We find that the ECL interactions significantly contribute to NbE binding and OR subtype selectivity. Based on NbE’s complementary-determining region 3 (CDR3), which inserts as a β-hairpin loop into the orthosteric pocket, we synthesized a set of linear and cyclic peptide mimetics of different lengths. The CDR3-based cyclic peptides were grafted on a semi-rigid β-turn-inducing scaffold. Thereby, we identify peptides that bind the μOR with an affinity similar to NbE and retain μOR-selective antagonism. The work uncovers a distinct ligand engagement profile at the μOR and provides a strategy for developing biologics-based µOR ligands.

Results

NbE binds the extracellular side of μOR and is an antagonist

We first tested whether NbE, which was part of a Nb library previously generated against the μOR⁹, binds to the extracellular side of μOR on the surface of living cells. We covalently conjugated purified NbE with Alexa Fluor 488 and incubated OR-expressing HEK293 cells with the fluorescently labeled NbE. Confocal microscopy analyses showed a pronounced NbE signal at the plasma membrane of cells expressing mouse or human μOR, while cells expressing the closely related δOR, κOR, or NOPR were not labeled by NbE (Fig. 1a and Supplementary Fig. 1a). Selective binding of NbE to μOR-expressing cells was also detected by flow cytometry analyzes after adding NbE at increasing concentrations (Fig. 1b). δOR-, κOR-, or NOPR-expressing cells gated for similar receptor surface levels did not exhibit NbE binding above control cells (Fig. 1b and Supplementary Fig. 1b). Since the NbE staining intensity of μOR expressing cells did not plateau at the highest NbE concentration tested (10 μM), we turned to grating-coupled interferometry (GCI) to determine the NbE binding strength. We immobilized biotinylated AVI-tagged NbE on a streptavidin-coated biosensor surface and perfused it with buffer containing purified murine μOR at different concentrations. From the binding curves, we extracted the affinity of the NbE-μOR interaction (K_D = 56 nM) and the kinetic parameters (k_a= 2.8 × 10³ M⁻¹s⁻¹ and k_d = 1.6 × 10⁻⁴ s⁻¹), which revealed a slow NbE off-rate (Supplementary Fig. 1c, d).

**Fig. 1: NbE binds the extracellular side of the μOR and acts as an antagonist.**

Next, we investigated if NbE binds competitively with orthosteric μOR ligands. We incubated μOR-expressing HEK293 cells with increasing concentrations of naloxone, subsequently added fluorescent NbE, and quantified the NbE signal by flow cytometry. NbE binding decreased in a naloxone concentration-dependent manner and was entirely abolished at high naloxone concentrations (Fig. 1c). The data suggested that the NbE binding site on the μOR potentially overlaps with the orthosteric ligand binding pocket, and thus we tested if NbE modulates μOR activity. We first assessed whether NbE behaves as an agonist by measuring μOR-mediated inhibition of cyclic AMP (cAMP) accumulation in living cells, a readout of Gi-driven OR signaling. Application of NbE did not activate μOR even at high concentrations, in contrast to the peptide agonist DAMGO (Fig. 1d). We then tested whether NbE acts as a μOR antagonist and reduces the signaling effects of opioid peptides and small molecule opioid drugs. Indeed, pre-incubation of μOR-expressing cells with NbE caused a concentration-dependent decrease in DAMGO- and morphine-induced Gi signaling (Fig. 1e). At high concentrations, NbE fully blocked DAMGO and morphine-driven μOR inhibition of cAMP production. Furthermore, NbE addition strongly reduced the recruitment of a G protein probe (miniGi) and β-arrestin2 to DAMGO-activated μOR, as measured by split nanoluciferase (NanoLuc)-based complementation assays (Supplementary Fig. 2a).

We hypothesized that NbE may also antagonize μOR function when ectopically expressed in cells and targeted to the extracellular leaflet of the plasma membrane via a glycolipid anchor. To test this, we fused NbE to a C-terminal glycosylphosphatidylinositol (GPI) signal peptide, separated by a flexible linker (Fig. 1f). We also added a secretory signal peptide at the N-terminus, followed by an HA-tag or a fluorescent protein to enable determining the subcellular localization of the NbE fusion proteins (Fig. 1f). Confocal microscopy analyses of NbE-GPI-expressing cells showed that NbE predominantly localized at the cell surface (Fig. 1g). Immunostaining of cells expressing HA-NbE-GPI with HA antibodies in non-permeabilizing conditions revealed that NbE was efficiently displayed on the extracellular side of the plasma membrane (Supplementary Fig. 2b). We also constructed a bivalent tandem NbE-NbE-GPI construct as well as a non-targeting Nb-GPI control (Nb-ctr), and both were predominantly cell surface localized (Supplementary Fig. 2c). μOR-expressing HEK293 cells transfected with monovalent or bivalent NbE-GPI constructs showed significantly higher cAMP levels after DAMGO treatment relative to cells expressing the Nb-GPI control, indicating a NbE-mediated block of μOR signaling (Fig. 1h).

Taken together, NbE specifically binds to the extracellular side of the μOR and competes with orthosteric opioid ligands. Furthermore, NbE acts as an antagonist when added to cells as an extracellular ligand or when ectopically expressed and targeted to the outer plasma membrane leaflet as GPI-anchored fusion protein.

Cryo-EM structure determination of the NbE-μOR complex

Advancement of structural biology techniques have led to the determination of several μOR structures in complex with either antagonists^11,27, partial agonists^25,28, or full agonists, including the endogenous peptides β-endorphin and endomorphin^9,28,29,30. All ligands bind to the orthosteric ligand binding pocket that is on the extracellular side and largely solvent exposed. To identify the precise molecular binding mode of NbE, we determined the structure of the NbE-μOR complex using cryo-EM. Initial structure determination was precluded due to the absence of large extracellular features of the complex. The surrounding detergent micelle additionally impaired accurate particle projection alignments, resulting in unsuccessful 3D reconstruction. To provide extra features for accurate particle alignment, we incubated the purified NbE-μOR complex with a Fab module consisting of a Nb-binding Fab fragment (NabFab) and an anti-Fab Nb, recently developed as a fiducial marker³¹. The resulting stable and homogenous NbE-µOR-Fab module complex was purified by size exclusion chromatography (Supplementary Fig. 3a). Cryo-EM analyzes enabled us to determine the structure of the NbE-µOR-Fab module complex at a global resolution of 3.1 Å (Supplementary Fig. 3b–f, Supplementary Fig. 4a–c, Supplementary Table 1). The EM-derived Coulomb potential map provides excellent side chain densities for most parts of the complex with the best resolved region at the NbE-µOR interface (Supplementary Fig. 3d, f and Supplementary Fig. 5a).

The overall structure of the NbE-µOR complex bound to the Fab module is shown in Fig. 2a. The NabFab attaches to the NbE scaffold at a site distal to the CDR loops. NabFab binding to NbE is ensured by introducing three point mutations in NbE’s scaffold, thereby assimilating NbE to the scaffold of Tc-Nb4, the original target antigen of NabFab³¹. Introducing the scaffold mutations did not significantly alter NbE binding (Ki) to µOR in cell membranes, as measured in homogeneous time-resolved fluorescence (HTRF) Tag-lite ligand competition binding assays (Supplementary Fig. 4d). The pre-assembled NbE-NabFab complex showed a modest reduction in the Ki (Supplementary Fig. 4d), possibly due to the larger ligand size and slight steric clashes between the membrane surrounding µOR and the NabFab fragment, as suggested by the EM density map (Fig. 2a and Supplementary Movie 1). The Fab module binds in a single and rigid conformation to NbE, enabling high-resolution structure determination of the NbE-µOR complex. For reasons of clarity, the Fab module is omitted in all subsequent figures (Fig. 2b and Supplementary Movie 1).

**Fig. 2: Architecture of the NbE-µOR complex.**

When bound to NbE, the µOR adopts an inactive conformation, which closely resembles the structures of the µOR bound to the morphinan antagonist β-funaltrexamine (β-FNA) or alvimopan (Supplementary Fig. 6a)^11,27. The average root mean square deviation (RMSD) for all C_α atoms of µOR between the NbE-µOR complex and the µOR complexes bound to antagonists is 0.78 Å, whereas the RMSD between NbE-µOR and agonist-bound structures is on average 1.72 Å (Supplementary Table 2). In particular, transmembrane helix 6 (TM6), which is displaced by roughly 10 Å in the activated state^9,29, superimposes well between the NbE-µOR and the β-FNA- or alvimopan-bound inactive µOR structures (Supplementary Fig. 6a, c). Moreover, a conserved core triad consisting of the amino acids I155^3.40, P244^5.50, and F289^6.44 (Ballesteros-Weinstein numbering scheme³²), which lies below the ligand binding pocket and propagates structural rearrangements involved in receptor activation⁹, superimposes well with the µOR in the inactive form (Supplementary Fig. 6b, d). Structural differences in the ligand binding pocket between agonists, partial agonists, and antagonists are relatively subtle^33,34, however, binding of NbE to the µOR induces several unique conformational changes in the orthosteric binding pocket and the extracellular loops.

Unique interaction profile of NbE with the inactive μOR

Binding of NbE to the µOR is mediated through its complementarity-determining regions (CDRs) CDR1 and CDR3, with the main interaction interface being formed between the β-hairpin loop of CDR3 and µOR’s TM helices 3, 5, 6, and 7 (Fig. 3a–d and Supplementary Table 3). CDR3 deeply inserts into the orthosteric ligand binding pocket with the three aromatic residues Y106^NbE, F107^NbE and Y108^NbE forming the tip of the loop (Fig. 3b, e and Supplementary Movie 1).

**Fig. 3: The NbE-µOR interaction interface.**

Similar to previously determined µOR structures bound to the ligands DAMGO²⁹, BU72⁹, β-FNA²⁷, alvimopan¹¹, PZM21, and FH210²⁵, as well as β-endorphin and endomorphin³⁰, H297^6.52 is positioned closely to a phenol hydroxyl group of the NbE ligand (Fig. 3b and Supplementary Fig. 5b). While binding of BU72 and β-FNA is mediated through a hydrogen bond network that involves two water molecules^9,27, in the NbE-µOR complex, H297^6.52and Y106^NbE are directly forming a hydrogen bond (Fig. 3b and Supplementary Fig. 5b). In addition, Y106^NbE is surrounded by mainly hydrophobic residues, including Y148^3.33, M151^3.36, V236^5.42, F237^5.43, I296^6.51, and V300^6.55 (Fig. 3b). F107^NbE, the second tip-forming aromatic residue, inserts itself into a neighboring hydrophobic cavity (Fig. 3b, e) and interacts with the key residues I143^3.30, Y148^3.33, M151^3.36, W293^6.48, I322^7.39 and Y326^7.43 of µOR (Fig. 3b). Interestingly, the aspartate D147^3.32, present in many class A GPCRs, and crucial for the recognition of DAMGO, β-FNA, BU72, PZM21, and endogenous peptides by forming a salt bridge with an amine group of the ligand, is rotated compared to all other µOR structures (Supplementary Fig. 6e). Instead of forming a salt bridge with the ligand, D147^3.32 stacks in this conformation onto F107^NbE, representing a distinct ligand-receptor interaction mode (Fig. 3b and Supplementary Fig. 6e). The third aromatic residue Y108^NbE is situated at the periphery of the orthosteric binding pocket (Fig. 3c). The aromatic ring of the phenol group stacks against L219^ECL2, part of ECL2, whereas the hydroxyl group forms a hydrogen bond with K233^5.39 (Fig. 3b, c and Supplementary Fig. 5b). K233^5.39 had previously been identified as the side chain to which the morphinan ligand β-FNA is covalently attached^27,35. Another notable difference in the ligand binding pocket involves W318^7.35 which is uniquely positioned, likely due to steric constraints when inserting the large NbE ligand into the narrow ligand binding pocket (Supplementary Fig. 6e). This specific rotamer conformation has only been observed when the pentameric peptide ligand DAMGO is bound to the µOR, however, in the NbE-µOR structure W318^7.35 is shifted by several Å. Because of the unusual positioning of W318^7.35 and its rotamer conformation, K303^6.58 that often stacks onto W318^7.35 is displaced by several Å, hydrogen bonding with S103^NbE of NbE instead (Fig. 3d and Supplementary Fig. 5b).

The three tip-forming aromatic residues of NbE are recognized by neighboring but distinct binding cavities in the orthosteric binding pocket (Fig. 3e). Here, Y106^NbE and F107^NbE are binding to sites usually occupied by small molecule ligands (Fig. 3f–h), whereas Y108^NbE stacks onto L219 of ECL2, thereby stabilizing the CDR3 β-hairpin loop in its conformation. The binding of ligands to this peripheral binding site has not yet been observed in other µOR structures. Conversely, the ligands alvimopan (antagonist), endomorphin, and fentanyl (agonists) are inserting aromatic functional groups into another binding cavity, which is not occupied by NbE (Fig. 3f–h).

To probe if the Nb ligand undergoes conformational rearrangements upon µOR binding, we crystallized NbE in its unbound form. The crystal contained three NbE molecules per asymmetric unit with all molecules virtually identical to each other, including the CDRs. A superposition between the unbound and bound NbE structures revealed that the CDR1 region and the CDR3 β-hairpin loop undergo a conformational shift when comparing the free and the µOR-bound state, while the Nb scaffold is nearly identical (Supplementary Fig. 7). The conformational differences suggest that CDR1 and CDR3 exhibit intrinsic flexibility, which could potentially be important for conformational selection of the inactive µOR and high-affinity binding of NbE.

In summary, NbE shows a unique interaction profile with the µOR. In particular, D147^3.32, L219^ECL2, and W318^7.35 of the µOR engage with NbE in a so far undetected ligand binding mode that can offer new possibilities for structure-guided drug design.

Extracellular loops confer NbE binding selectivity

To provide a rationale for NbE’s selectivity for µOR over other OR family members (Fig. 1a, b), we focused our structural analyses on the extracellular loops, which represent the most variable regions in the OR family. In particular, ECL2 shows variations in sequence and length between the receptors (Fig. 4a and Supplementary Fig. 8), providing a possible explanation for the observed µOR selectivity given the specific interactions of NbE with ECL2 and ECL3 (Fig. 3c, d). K209^ECL2, R211^ECL2 and Q212^ECL2 interact with E111^NbE, Y102^NbE, and T35^NbE respectively, thereby contributing to high-affinity binding (Fig. 4b and Supplementary Fig. 5b). Importantly, Q212^ECL2 is unique to the µOR. The centered L219^ECL2 of ECL2 stacks onto Y108^NbE of CDR3 (Fig. 3b, c). K303^6.58, also unique to µOR, forms a specific hydrogen bond with S103^NbE that likely stabilizes ECL3 in a conformation that allows several nonspecific side chain-main chain interactions between µOR and NbE, with E310^ECL3 being at ECL3’s center (Fig. 4c and Supplementary Fig. 5b). Binding of NbE to µOR induces clear shifts for ECL2, parts of TM6, ECL3 and TM7 compared to the other two antagonist-bound µOR structures (Supplementary Fig. 6a).

**Fig. 4: ECL regions confer NbE binding selectivity.**

To probe the relevance of µOR’s ECLs in NbE binding, we first substituted ECL2 residues K209^ECL2, R211^ECL2, Q212^ECL2 and L219^ECL2 (‘KRQL’ motif) by alanine and tested NbE binding to the µOR mutant by flow cytometry, gating for cells with similar µOR surface expression (Supplementary Fig. 9a, b). Compared to wild-type µOR, the KRQL-A mutant showed 50% reduced NbE binding, indicating a lower affinity for the ECL2 mutant (Fig. 4d). Exchanging E310^ECL3 in ECL3 with alanine also significantly reduced NbE binding (Fig. 4d). Combining both the ECL2 and ECL3 mutations did not lead to further reduction in NbE binding, highlighting the central role of the orthosteric pocket interactions (Fig. 4d). All mutants retained comparable responses to DAMGO, as assessed by flow cytometry assays evaluating DAMGO-induced µOR internalization (Supplementary Fig. 9c).

Next, we reasoned that assimilating the δOR to the µOR, based on the uncovered interaction interfaces, might transform the δOR from a non-binder to a NbE binder. Given the established role of the amino acid position 7.35 in opioid receptor subtype selectivity^36,37, we first mutated δOR’s L300^7.35 in the orthosteric pocket to the corresponding µOR residue W318^7.35 (δOR^L300W mutant). We then also converted the δOR residues D193^ECL2 (corresponds to µOR Q212^ECL2) and M199^ECL2 (corresponds to µOR T218^ECL2) into glutamine and threonine residues respectively (Fig. 4b, c). T218^ECL2 of µOR forms an intramolecular hydrogen bond with the guanidinium group of R211^ECL2. As a consequence of this loop stabilization, K209^ECL2 and R211^ECL2 of the µOR are ideally positioned to create the aforementioned hydrogen bond and salt bridge network with Y102^NbE and E111^NbE of NbE (Fig. 4b). Because T218^ECL2 is unique to the µOR, we speculated that mutating δOR M199^ECL2 into threonine might stabilize the ECL2 of the δOR in a conformation that favors NbE binding. We quantified fluorescent NbE binding to cells expressing µOR (control), wild-type δOR, or δOR mutants by flow cytometry in cells gated for similar receptor levels (Supplementary Fig. 9d). As expected, only non-specific background NbE binding was detected for cells expressing wild-type δOR (Fig. 4e). The NbE signal was not increased for cells expressing δOR^L300W. However, strong NbE binding was observed when cells expressed the triple mutant δOR^{D193Q, M199T, L300W} (Fig. 4e). The results show that mutating two residues in the ECL2^δOR, combined with the orthosteric L300W mutation, can convert the non-binding δOR into a strong NbE binder. To test the role of ECL3, we next substituted the entire ECL3 as well as residues of the connecting α-helix 7 of the δOR (residues 287–300) with residues of the µOR (306-318). The substitution led to a moderate but significant increase in NbE binding (Fig. 4e), indicating that µOR’s ECL3 region partially contributes to NbE binding.

In sum, the mutational and gain-of-function studies identify ECL2 and ECL3 as important contributors to NbE binding and receptor subtype selectivity.

Constrained peptide mimetics of NbE’s CDR3 efficiently antagonize µOR

The centering of key contacts on a single CDR makes NbE’s CDR3 a promising starting point for the design of ligands that downsize the Nb towards smaller peptides based on the antigen-binding paratope³⁸. First, we designed and synthesized a library of linear peptides of increasing length, with the shortest peptide based on the four residues ¹⁰⁵SYFY¹⁰⁸ that compose the β-turn segment at the tip of NbE’s CDR3 (Fig. 5a). We systematically extended each peptide by one N- and one C-terminal residue of the CDR3^NbE with the longest linear peptide spanning the 14 residues ¹⁰⁰KYYSGSYFYKSEYD¹¹³ (Fig. 5a). Peptide binding to µOR was assessed using the HTRF Tag-lite binding assay, which relies on fluorescence resonance energy transfer (FRET) between SNAP-tagged µOR labeled with terbium cryptate as FRET donor, and the red fluorescent opioid ligand naltrexone, serving as acceptor, with a decrease in FRET indicating competitive binding of a test compound. Binding of the control ligand naloxone and of NbE was readily detected with Ki values of 1.9 nM and 18 nM respectively (Fig. 5b and Supplementary Fig. 10a). In the Tag-lite binding assay, we detected significant binding of peptides 5 (12 residues) and 6 (14 residues) at Ki values of 1.3 µM and 3.1 µM, respectively, while the shorter peptides 1–4 showed little to no binding (Fig. 5b and Supplementary Fig. 10a, b).

**Fig. 5: CDR3-based peptide-mimetics exhibit µOR binding and antagonism.**

Linear peptides are intrinsically flexible molecules able to adopt multiple conformations, which commonly deviate from ordered secondary structures, such as β-hairpins. This was confirmed by circular dichroism (CD), showing predominantly random coil CD spectra for peptides 1-6 in water and in the α-helix stabilizing solvent TFE (2,2,2-trifluoroethanol) (Supplementary Fig. 10c, e). To mimic the more rigid CDR3 β-hairpin structure observed in NbE, cyclic peptides 7 and 8 were designed. Here, the longest loop sequences (i.e., peptides 5 and 6, respectively) were grafted on the β-turn-inducing D-Pro-L-Pro dipeptide³⁹. The position of the scaffold in these sequences was chosen to create a conformational bias for the observed hydrogen bonds in the NbE CDR3 β-hairpin structure, comparable to the conformational rigidity imposed by the Nb framework (Fig. 5a, c). CD measurements confirmed the beneficial effect of cyclization, as typical CD spectra for a β-hairpin structure were observed for cyclic peptide 8 (Supplementary Fig. 10d, f). The increased tendency of cyclic peptides 7 and 8 to adopt a β-hairpin structure compared to the linear peptides, as imposed by their design, also resulted in higher affinities for the µOR with Ki values of 245 nM and 39 nM, respectively, with the latter closely approaching the parental NbE affinity (Ki 18 nM) (Fig. 5d and Supplementary Fig. 10a). Consequently, this CDR3 peptidomimetic recapitulates closely the binding properties of the parental Nb. Previous work on such mimetics already delivered proof-of-concept, but never with equivalent binding properties^40,41.

Next, we tested whether the peptide mimetics retain the µOR-selective antagonism observed for NbE. Peptides 5 to 8 caused a concentration-dependent decrease in DAMGO-driven μOR signaling, with peptides 7 and 8 fully reversing μOR-mediated inhibition of cAMP production at high concentrations (Fig. 5e). Furthermore, DPDPE-driven δOR signaling was not antagonized by the peptides, showing that the CDR3 analogs retain receptor subtype selectivity (Fig. 5f). Taken together, the extensive interaction interface between the CDR3 of NbE and the µOR allowed the design of downsized linear and cyclic peptide mimetics that retain high-affinity µOR binding and antagonism. The NbE-based paratope mimetics provide a distinct source of small peptide opioid ligands and demonstrate the value of Nbs for biologics-based drug development.

Discussion

The emergence of biologics as therapeutic agents has opened new avenues in GPCR drug discovery⁴. In this study, we present an in-depth molecular characterization of the single-___domain antibody fragment NbE that functionally interacts with the µOR. NbE selectively binds to the extracellular domains of µOR with nanomolar affinity and antagonizes its function. The cryo-EM structure of the NbE-μOR complex shows that NbE stabilizes the µOR in an inactive conformation that resembles µOR structures bound to the small molecule antagonists β-FNA or alvimopan^11,27. Yet NbE exerts distinct effects on key residues in the orthosteric binding pocket as well as the extracellular regions of µOR. The interactions of NbE with µOR’s ECL2 and ECL3 critically contribute to selective binding, paralleling recent observations for the role of extracellular vestibules in serving as selectivity filters of ORs for different opioid peptide agonists³⁰. NbE exhibits an unprecedented ligand interaction profile with the µOR, centering almost all contacts on a hairpin loop formed by CDR3. The work opens the possibility to therapeutically target µOR with a biologic and provides an important structural template for new ligand design.

NbE’s binding kinetics are characterized by a relatively slow on-rate and by nearly irreversible binding on the time scale studied, pointing to unique biomolecular recognition. Based on the conformational differences that we detect in NbE’s unbound and µOR-bound state, in particular with regards to CDR3, it is possible that µOR binding drives NbE towards a conformation that is more complementary to the orthosteric pocket. Moreover, since GPCRs are known to be inherently dynamic and sample an ensemble of conformations in the unliganded state, NbE’s slow on-rate may also reflect conformational selection of a weakly populated µOR conformation. The slow off-rate makes NbE a unique µOR ligand, holding the potential for long-lasting µOR-specific antagonism and distinct physiological effects. Underlining the therapeutic potential of pseudo-irreversible ligands, the small molecule antagonist methocinnamix, which very slowly dissociates from the µOR but shows rapid reversal at δOR and κOR⁴², was recently shown to persistently attenuate the effect of opioid drugs for days to weeks following a single administration in animal models^43,44. Therefore, antagonists with a slow-off rate have therapeutic potential in the long-term treatment of opioid use disorders.

Recent structures of class A GPCRs bound to extracellular Nbs begin to reveal the diversity in Nb-driven GPCR modulation (Supplementary Fig. 11). For example, Nbs targeting the α1A-adrenergic receptor⁴⁵, rhodopsin¹⁸, or orexin receptor 2⁴⁶ interact with extracellular loops and allosterically modulate receptor function without directly contacting the orthosteric pocket. In contrast, extracellular Nbs of the apelin receptor (APJ)¹⁷ or the angiotensin II type 1 receptor (AT1R)¹⁹ occupy the orthosteric pocket and mimic the binding of a peptide ligand. Similar to the NbE-µOR interaction, the APJ and AT1R Nbs access the orthosteric pocket via their long CDR3 loops. Yet APJ and AT1R Nbs both exhibit extensive additional interactions via CDR1 and CDR2, which contrasts with the limited interactions outside the CDR3 region detected for NbE. We find that linear and cyclic CDR3 peptide analogs retain biological activity and are selective for the µOR, confirming the importance of the CDR3 sequence in NbE binding and, more broadly, demonstrating that CDR paratopes of Nbs can spur the development of low molecular weight antibody mimetics. Until now, OR peptide antagonists have been developed through structural modification and conversion of natural agonist peptides or by functional screening of synthetic short peptide libraries, often yielding peptides with limited OR subtype selectivity or reduced affinity^47,48,49. The functional NbE-derived peptides described here comprise the NbE CDR3 residues that interact with the orthosteric pocket and include flanking Y102 and E111 that interact with µOR’s ECL2, providing a rationale for their µOR selectivity. As expected, taking NbE’s CDR3 sequence out of the rigid Nb framework resulted in a drop in affinity and potency. This effect could be compensated by conformational rigidification through cyclization with a β-hairpin stabilizing template, thereby functionally mimicking the parental NbE. Future steps towards fine-tuning the three-dimensional peptide structure will very likely give access to further improved binding and antagonist potency.

Beyond NbE’s action as an extracellular ligand, we present a strategy to employ NbE when ectopically expressed in cells to modulate OR activity, which is broadly applicable to other GPCR-modulatory Nbs. Linkage of NbE to a GPI-anchor effectively displays NbE on the extracellular plasma membrane leaflet of transfected cells and blocks OR function, implementing a promising strategy to inhibit µOR in individual neurons. In the future, NbE can be fused with subcellular targeting motifs to achieve µOR inhibition at sub neuronal compartments, which would deliver high-precision genetically encoded modulators to untangle µOR signaling promoted by endogenous and exogenous ligands in functional studies in vitro and in vivo.

In sum, the present results reinforce the emerging notion that Nbs can uniquely engage with clinically important target proteins and constitute a new class of GPCR ligands. Mechanistic insights into Nb-GPCR engagement can instruct novel targeting strategies poised to impact future drug development.

Methods

Mammalian cell culture conditions and plasmids

HEK293 cells (CRL-1573, ATCC, female) were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco), supplemented with 10% fetal bovine serum (FBS, Thermo Fisher). HEK293 cells stably expressing N-terminally signal sequence FLAG (ssf)-tagged murine µOR (HEK293-µOR) were cultured in the presence of 250 μg/ml Geneticin (Gibco)¹⁴. A clonal HEK293 cell line stably expressing N-terminal signal sequence (ss) SNAP-tagged murine µOR was generated by cloning a ‘CAG promoter, ssSNAP-µOR, RGK promoter, PuroR’ cassette into a piggyBac transposon plasmid (Addgene #84239), co-transfection of the plasmid with piggyBac transposase, followed by puromycin selection and isolating single cell clones by FACS. For transient DNA expression, Lipofectamine 2000 (Invitrogen) was used according to the manufacturer’s instructions. Published plasmids used in this study: Murine ssf-δOR, ssf-µOR, ssf-κOR, and ssf-NOPR in pcDNA3.1, NbE-6xHis in pXAP100⁹, LgBiT-β-arrestin2 (human) in pNBe3⁵⁰, µOR-SmBiT (mouse), and miniGi1-LgBiT in pcDNA3.1⁵¹. Plasmids generated in this study: mNbE (N45K, Q117K, Q120P)-6xHis in pXAP100, GPI-anchored constructs: NbE and non-targeting Nb-ctr (NbALFA⁵²) fused with a C-terminal GPI signal peptide from CD59, separated by a 15 aa (GGGGS)3 flexible linker, with N-terminal IL-2 secretory signal peptide and HA- or mRuby2-tag in pcDNA3.1. Mutant versions of murine ssf-µOR and ssf-δOR, and human ssf-µOR and ssf-δOR in pcDNA3.1. Plasmids were generated by In-Fusion cloning (Takara) or QuikChange Site-Directed Mutagenesis (Agilent). Plasmid sequences are available in the Yareta data repository (see Data Availability paragraph).

NbE purification and labeling

Escherichia coli WK6 cells were transformed with NbE (originally named ‘Nb35’ ⁹) cloned into the pXAP100 plasmid, containing an N-terminal pelB signal sequence for periplasmic expression and C-terminal 6xHis-tag for purification. A single colony was grown in Terrific Broth media (1.2% w/v tryptone, 2.4% w/v yeast extract, 0.6% glycerol, 0.1% w/v glucose, 2 mM MgCl₂) at 37 °C with 50 μg/mL ampicillin to an optical density (OD A₆₀₀) of 0.7, followed by induction with 1 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) and incubated overnight at 37 °C at 180 rpm shaking. Cell pellets were resuspended in TES buffer (0.2 M Tris, 5 mM EDTA, 0.5 M sucrose, pH 8.0) equivalent to 5% of bacterial culture volume, and kept on a shaker at 4 °C for at least 60 min. Cells underwent osmotic shock by adding double the volume of TES diluted 1:4 with H₂O and kept on the 4 °C shaker for 45 min. Cells were pelleted and discarded, and NbE contained in periplasmic lysate purified by Ni²⁺-affinity (HisPur Ni NTA Resin; Thermo Fisher Scientific, catalog no. 88221) and eluted by low pH by adding acetate buffer (50 mM NaAc,1 M NaCl, pH 4.5-4.7) and immediately adding 1 M Tris (pH 7.5) to the eluates. NbE was further purified by size-exclusion chromatography (Superdex 75, Cytiva) and purity assessed by SDS-PAGE gel electrophoresis and gel staining with Coomassie blue (Bio-Rad, catalog no. 1610400). NbE was covalently conjugated at primary amines of lysines with Alexa Fluor 488 (Invitrogen, catalog no. A10235). For conjugation, 1 mg NbE was incubated with AF488 for 1 h at room temperature (RT) and NbE-AF488 separated from free dye using gel filtration. The concentration and degree of labeling was determined with a NanoDrop spectrophotometer (Thermo Scientific) and NbE-AF488 subsequently used in fluorescence microscopy- and flow cytometry-based binding assays.

Flow cytometry-based binding assay

HEK293 cells were transiently transfected with ssf-tagged murine µOR, δOR, κOR, and NOPR or with µOR and δOR mutants. 24 h post transfection, cells were detached with PBS-EDTA and resuspended in ice-cold assay buffer (PBS, 1 mM Ca²⁺, 0.5 mM Mg²⁺) at 8 million live cells/ml. Cells were incubated for 30 min at RT with NbE-AF488 at various concentrations and 10 min anti-FLAG M1 antibody (Sigma- Aldrich, catalog no. F-3040) conjugated to Alexa Fluor 647 (Invitrogen, catalog no. A20173) (anti-FLAG M1-AF647). Cells were washed twice with PBS and resuspended in FACS buffer (PBS, 1 mM Ca²⁺, 0.5 mM Mg²⁺, BSA 0.5%). Flow cytometry was performed using a Beckman Colter CytoFLEX Flow Cytometer. Data were analyzed with FlowJoTM v10 software and cells were gated for i) singlets, ii) living cells, and iii) comparable receptor expression using the anti-FLAG M1-AF647 signal. Conditions were normalized to the AF488 signal from µOR wt-expressing cells treated with 10 μM (Fig. 1b) or 1 μM (Fig. 4d, e) NbE-AF488. To probe competitive binding with naloxone, HEK293 cells stably expressing µOR were seeded at 5 × 10⁴ live cells/cm² density. Cells were detached and resuspended in ice-cold assay buffer. Cells were then pre-treated with different concentrations of naloxone (Sigma-Aldrich, N7758) for 5 min, followed by addition of 1 μM NbE-AF488 for 30 min. Cells were washed twice with PBS, resuspended in FACS buffer, and analyzed using the Beckman Colter CytoFLEX Flow Cytometer. Data were analyzed with FlowJoTM v10 software and cells were gated for i) singlets and ii) living cells. Conditions were normalized to the maximum AF488 signal from the cells treated with 1 μM NbE-AF488 without addition of naloxone. The analyzed data was plotted using GraphPad Prism Version 9.2 software.

Luminescence-based cAMP assay

HEK293-µOR cells were transfected with a plasmid encoding pGloSensor-20F cAMP reporter (Promega). Cells were harvested 24 h post transfection and resuspended at a 1.5 × 10⁶ live cells/ml in assay media (DMEM without phenol red or FluoroBrite DMEM, 250 μg/ml luciferin). 100 μl cell suspension was plated into each well of a clear bottom 96-well plate, and equilibrated for 60 min at 37 °C. To probe the reversal of 10 nM DAMGO (Sigma-Aldrich, E7384) or 30 nM morphine (morphine HCl, Sintetica) (concentrations correspond to their respective EC50 values in the assay, agonist stock in H₂O, diluted in assay media), cells were pre-incubated for 15 min with NbE at different concentrations. Before addition of ligands, sequential luminescence images were collected to obtain baseline luminescence values using the FDSS/μCELL kinetic plate imager (Hamamatsu) with an integrated simultaneous dispensing head and simultaneous detection across the plate or using the Hidex Sense microplate reader. Cells were then treated with ligands and 2.5 μM Forskolin (FSK, Sigma-Aldrich, catalog no. F6886), followed by continuous luminescence imaging for 10 min, using two technical replicates per condition. Luminescence signals from cells not stimulated with FSK were considered as background and subtracted from all experimental conditions. The difference in luminescence signal between the baseline (average luminescence before addition of drugs (FDSS) or at beginning of acquisition (Hidex)) and the maximum signal in each condition was determined, and cells treated with 2.5 μM FSK (no agonist or NbE) normalized to 100%. To illustrate data on peptide-driven reversal of the DAMGO or DPDPE effect, the inhibition mediated by 10 nM DAMGO or 1 nM DPDPE in the absence of peptides was normalized as 100% μOR or 100% δOR signaling. GraphPad Prism Version 9.2 software was used for data analysis, statistical tests and graph plotting.

Split NanoLuc-based recruitment assay

HEK293 cells were seeded into 6-well plates and wells were transfected with 0.25 μg of µOR-SmBiT and (i) 0.25 μg of miniGi1-LgBiT or (ii) 0.25 μg of LgBiT-β-arrestin2 using 3 μl of Lipofectamine 2000. 24 h after transfection, cells were seeded into black, clear-bottom 384-well plates (20,000 cells per well) in assay media (FluoroBrite DMEM and the NanoLuc substrate NanoGlo (Promega, catalog no. N2012)), and incubated for 45 min at 37 °C. Cells were pre-incubated with different concentrations of NbE. Luminescence was recorded using the FDSS/μCELL kinetic plate imager (Hamamatsu) with an integrated simultaneous dispensing head and simultaneous detection across the plate. After acquiring baseline luminescence for 3 min, DAMGO (100 nM for miniGi1 and 1 μM for β-arrestin2 recruitment) was added to cells followed by continuous luminescence imaging for 10 min, using two technical replicates per condition. Luminescence signals from cells without any treatment were considered as background and subtracted from all experimental conditions. The difference in luminescence signal between the baseline (average luminescence before addition of DAMGO) and the maximum signal in each condition was determined, and cells treated with only DAMGO (no NbE) were normalized to 100%. GraphPad Prism Version 9.2 software was used for data analysis and graph plotting.

Confocal microscopy-based live cell binding assay

HEK293 cells were seeded on poly-L-lysine-coated 35-mm Cellvis glass-bottomed dishes (IBL, catalog no. 220.110.022) and, after 24 h, transfected with mouse ssf-µOR, -δOR, -κOR, or -NOPR (0.8 μg DNA) using 3 μl of Lipofectamine 2000. 16 to 24 h after transfection, cells were incubated at 37 °C with 1 μM NbE-AF488 for 30 min and anti-FLAG M1-AF647 antibody (1:1000 dilution) for 10 min in HBS imaging solution (Hepes-buffered saline with 135 mM NaCl, 5 mM KCl, 0.4 mM MgCl₂,1.8 mM CaCl₂, 20 mM Hepes, 1 mM d-glucose, 1% FBS, adjusted to pH 7.4). Cells were subsequently washed with PBS and imaged with a spinning disk confocal microscope (Nipkow, Zeiss) using Plan Apo 63x/1.4 Oil DICIII objective in a temperature and CO₂-controlled environment (37 °C, 5% CO₂).

Immunofluorescence and fixed cell binding assay

HEK293 cells were seeded onto 15-mm glass coverslips in 12-well plates and, after 24 h, transfected with human ssf-µOR or human ssf-δOR (0.4 μg of DNA using 1.5 μl of Lipofectamine 2000). 16 to 24 h after transfection, cells were treated with HA-tagged NbE for 30 min at 37 °C followed by washes with PBS and fixed using 4% formaldehyde (FA) in PBS. Cells were blocked with 1.5% BSA in PBS and incubated overnight at 4 °C with primary antibodies: HA-tag (6E2) mouse mAb (1:100 dilution; Cell Signaling Technology (CST), catalog no. 2367), Anti-FLAG rabbit (1:200 dilution; Merck Millipore Sigma, catalog no. F7425) in blocking solution. After three washes, cells were incubated with goat anti-rabbit immuno-globulin G (IgG) Alexa Fluor 647 (Thermo Fisher Scientific, catalog no. A32733) and goat anti-mouse IgG Alexa Fluor 568 (Thermo Fisher Scientific, catalog no. A11031) secondary antibodies (1:2000 dilution) for 45 min at RT. Samples were mounted using ProLong glass antifade mountant (Thermo Fisher Scientific, catalog no. P36982) and imaged with a spinning disk confocal microscope (Nipkow, Zeiss) using an Alpha Plan Apochromat 100x/1.46 Oil DIC. To check the localization of NbE-GPI constructs, HEK293 cells were seeded onto 15-mm glass coverslips in 12-well plates and, after 24 h, were transfected with NbE-GPI, 2xNbE-GPI, or Nb-ctr either fused to HA tag or the red fluorescent protein mRuby2. Cells were fixed using 4% FA in PBS. Cells were blocked with 1.5% BSA in PBS and incubated overnight at 4 °C with primary antibodies: mouse anti HA-tag conjugated with Alexa Fluor 488 (1:200, Life Technologies, catalog no. A-21287), and mouse anti-FLAG M1-647 (1:1000 dilution) to label the surface µOR. After three washes, cells were mounted using ProLong mountant and imaged with a spinning disk confocal microscope (Nipkow, Zeiss) using an Alpha Plan Apochromat 100x /1.46 Oil DIC.

Flow cytometry-based internalization assay

HEK293 cells were seeded onto 12 well-plate and transiently transfected with ssf-tagged murine wt μOR or μOR mutants. 24 h post transfection, cells were treated with 10 μM DAMGO or vehicle control for 25 min. Cells were immediately washed with ice-cold PBS, and surface receptors labeled for 10 min with anti-FLAG M1-AF647 (1:1000 dilution) at 4 °C. Cells were then resuspended in 500 µl FACS buffer (PBS, 1 mM Ca²⁺, 0.5 mM Mg²⁺, BSA 0.5%) and analyzed by flow cytometry (Beckman Colter Cytoflex). Data was analyzed with FlowJoTM v10 software and cells were gated for singlets and further gated for receptor expression. The area under curve (AUC) was calculated as a measure of the receptor level from the histogram of the gated population using the formula: AUC = sum of counts x width of the histogram bin; where the width of the bin was calculated using Sturges’ formula. The receptor levels of the vehicle control (no DAMGO) was normalized to 100% for each condition and each experimental replicate.

Tag-lite HTRF binding assay

HEK293 cells stably expressing SNAP-µOR were labeled with 100 nM Tag-lite SNAP-Lumi4-Tb (Revvity, catalog no. SSNPTBC). Labeled cells were resuspended at a concentration of 1 × 10⁶ cells/ml. 10 µl of the resuspension was plated in each well of low volume 96-well plates or 384-well plates to obtain 10'000 cells per well. 5 µl of labeled antagonist (Revvity, catalog no. L0005RED) was added to each well, resulting in a final ligand concentration of 3 nM, followed by addition of different concentrations of the test compound, and incubation for 3 h at RT, protected from light. Each experimental condition was performed in technical triplicates. Upon reaching equilibrium after 3 h, the FRET signal at 620 nm and 660 nm was read using a SpectraMax Paradigm Multi-Mode Microplate Reader (Molecular Devices). The ratio of the acceptor and donor emission signals (HTRF ratio signal) for each individual well was calculated. HTRF signals were plotted against concentrations. All conditions were normalized to the maximum HTRF ratio from the wells with 3 nM labeled reference antagonist (L0005RED). The signal from 1 µM naloxone condition was considered as background. GraphPad Prism Version 9.2 software was used for data analysis, statistical tests, and graph plotting.

Grating coupled interferometry assay

Grating coupled interferometry (GCI) experiments were conducted on a Creoptix WAVE delta system using 4PCP chips (Creoptix AG). The chips were conditioned with 100 mM sodium borate (pH 9.5) and 1 M NaCl (Xantec). Neutravidin (100 μg/μl in 10 mM sodium acetate, pH 5.0) was immobilized on the chip surface using standard amine coupling. This included 420 s of surface activation with a 1:1 mix of 400 mM EDC and 100 mM NHS (Xantec), 420 s neutravidin injection, 420 s BSA (0.5%) injection, and a final 420 s surface passivation with 1 M ethanolamine at pH 8.0. Subsequently, biotin-NbE was captured at three concentrations: 100 μg/μl, 20 μg/μl, and 1 μg/μl, yielding surface masses of approximately 3700, 3500, and 2300 pg/mm², respectively. Any remaining neutravidin sites were filled with 100 μg/μl biotin-BSA. All preparation steps were performed at 10 μl/min flow rate. The analyte, containing purified µOR or the TCR-CD3 complex (negative control), was injected in a 1:2 dilution series from 7.8 nM to 1000 nM at 45 μl/min. The running buffer contained 25 mM HEPES (pH 7.5), 100 mM NaCl, 0.01% LMNG, and 0.001% CHS. Blank injections and a reference channel were used for double referencing, and a 0-2% dimethylsulfoxide (DMSO) calibration curve was employed for bulk refractive index correction. Data were analyzed with Creoptix WAVEcontrol software, with corrections made for X and Y offset, DMSO calibration, and double referencing. A 1:1 Langmuir binding model with bulk correction was used for all experiments. Consistent results were observed across the three NbE concentrations.

Purification of the anti-Fab Nb

A fragment encoding the anti-Fab nanobody³¹ was synthesized by GeneArt (Thermo Fisher Scientific) and subsequently cloned into the pET28a vector downstream of a 6xHis tag. The resulting fusion proteins were expressed in Escherichia coli BL21 (DE3) at 18 °C overnight. Protein expression was induced with 0.5 mM IPTG at an OD A₆₀₀ of 0.6. The fusion proteins were first purified in batch by a Ni²⁺-affinity step (HisPur Ni-NTA Resin; Thermo Fisher Scientific), before being further purified by size-exclusion chromatography (Superdex 75 Increase 10/300 GL). The final protein buffer solution contained 25 mM Tris-HCl (pH 7.5) and 150 mM NaCl (no reducing agents). Monomeric anti-Fab nanobody fractions were pooled, concentrated to ~3.8 mg/ml, flash-frozen in liquid nitrogen, and stored at −70 °C.

Purification of the NabFab

The NabFab plasmid was kindly provided by the group of Prof. Kaspar Locher, ETH Zürich³¹. Chemically competent C43 Escherichia coli cells were used for protein expression. Four liters of TB autoinduction media (Terrific Broth containing 0.4% glycerol, 0.01% glucose, 0.02% lactose,1.25 mM MgSO₄ and 100 μg/ml ampicillin) were inoculated with overnight cultures from single colonies and incubated for 6 h at 37 °C, before shifting to 30 °C for expression overnight at 180 rpm shaking. Cells were resuspended in lysis buffer containing 50 mM Tris-HCl pH 7.5, 200 mM NaCl, protease inhibitor cocktail tablets (PIC) (complete EDTA-free, Roche Diagnostics) and 5 units/ml supernuclease (Novagen) and lysed by sonication. The lysate was incubated in a water bath at 65 °C for 40 min and spun down at 20,000x g for 30 min. The filtered supernatant was loaded on a HiTrap Protein L column (Cytiva), pre-equilibrated with 50 mM Tris-HCl (pH 7.5) and 500 mM NaCl. The protein was eluted using 0.1 M acetic acid and immediately loaded onto a Hitrap SP HP (Cytiva) column pre-equilibrated with buffer A (50 mM sodium acetate, pH 5.0). After a washing step, the NabFab was eluted using a salt gradient with buffer B (50 mM sodium acetate, 2 M NaCl, pH 5.0). The final protein buffer contained 25 mM Tris-HCl pH 7.5, 150 mM NaCl, the protein was concentrated to 7.0 mg/ml, flash-frozen and stored at −70 °C.

Purification of µOR

DNA coding for the full-length murine µOR was subcloned into the pF1 vector. The resulting construct comprises an N-terminal signal sequence, a Flag epitope tag and a C-terminal HRV-3C protease cleavage site followed by a Strep II tag and 8x His tag. µOR was expressed in Sf9 cells using baculoviruses in the presence of 10 µM naloxone. Typically, 20 ml of P3 viruses were used to infect 500 ml of Sf9 cells. Cells were infected at a cell density of roughly 3.0 × 10⁶ cells/ml, incubated for 48 h at 27 °C at 110 rev/min and harvested by centrifugation at a cell viability of 80–85%. The purification of µOR has been previously described²⁷. All purification steps were performed at 4 °C. Briefly, Sf9 cell pellets were lysed using hypotonic buffer containing 20 mM HEPES pH 7.5, 5 mM MgCl₂, PIC, 5 units/ml supernuclease, 10 μM naloxone (Sigma-Aldrich) and 2 mg/ml iodoacetamide (Sigma-Aldrich), and gently stirred for 1 h. Cell membranes were separated by ultracentrifugation at 100,000 × g for 40 min and resuspended in solubilization buffer made of 25 mM HEPES pH 7.5, 500 mM NaCl, 1% LMNG, 0.1% CHS, PIC, 10 μM naloxone and 2 mg/ml iodoacetamide using a Dounce homogenizer. The solution was gently stirred for 4 h. The insoluble debris was removed by ultracentrifugation at 100,000 × g for 1 h. The supernatant (supplemented with 20 mM imidazole) was incubated for 2-3 h with HisPur Ni-NTA Resin and extensively washed with wash buffer A (25 mM HEPES pH 7.5, 500 mM NaCl, 0.1% LMNG, 0.01% CHS, 1 μM naloxone, 10 mM imidazole), followed by buffer B (25 mM HEPES pH 7.5, 500 mM NaCl, 0.1% LMNG, 0.01% CHS, 1 μM naloxone, 10 mM MgCl₂, 10 mM ATP). Proteins were eluted using an imidazole gradient and subsequently loaded onto a 5 ml StrepTactin Superflow Cartridge (Qiagen) at a flow rate of 0.8 ml/min. The column was washed with wash buffer C (25 mM HEPES pH 7.5, 500 mM NaCl, 0.05% LMNG, 0.005% CHS) to remove contaminations and residual naloxone. µOR was eluted using 2.5 mM desthiobiotin. The µOR was further purified by size exclusion chromatography. A Superose 6 Increase 10/300 GL column (GE Healthcare Life Sciences) was pre-equilibrated with 20 mM HEPES pH 7.5, 100 mM NaCl, and 0.001% LMNG/0.0001% CHS. Fractions containing µOR were concentrated, flash-frozen, and stored at −80 °C.

Purification of the TCR-CD3 complex (control in GCI assay)

The two plasmids encoding the human TCRγδ and CD3γδε₂ζ₂ complex were transfected into the HEK293F cells. The cell pellets were subjected to repeated cycles of Dounce homogenization in hypoosmotic buffer (25 mM HEPES pH 7.5) supplemented with protease inhibitor cocktail (Roche). The cell membrane was homogenized in a solubilization buffer (25 mM HEPES pH 7.5, 500 mM NaCl, 1% LMNG, 0.1% CHS), and solubilized for 4 h in 4 °C. Cell debris and insoluble material was removed by centrifugation at 35,000 g for 30 min. The supernatant was gently mixed with strep-tactin resin at 4 °C overnight for binding. The resin was washed with buffer A (25 mM HEPES pH 7.5, 100 mM NaCl, 0.01% LMNG, 0.001% CHS), and eluted with buffer B (25 mM HEPES pH 7.5, 100 mM NaCl, 0.01% LMNG, 0.001% CHS, 50 mM biotin). The eluted sample underwent gel filtration using a Superose 6 Increase 10/300 GL column in buffer (25 mM HEPES, pH 7.5, 100 mM NaCl, 0.01% LMNG, 0.001% CHS) and was concentrated to around 1 mg/ml.

NbE-µOR-Fab module complex assembly

To assemble the NbE-µOR-Fab module complex, freshly prepared µOR was incubated with mNbE (containing framework mutations N45K, Q117K, Q120P to enable NabFab binding³¹), NabFab and anti-Fab Nb in a molar ratio of 1:2:2.5:3 overnight, gently mixed at 4 °C. A final size-exclusion step using a Superose 6 Increase 10/300 GL column was conducted. Fractions containing the complex were pooled, concentrated to around 2.8 mg/ml and immediately used for EM grid preparation. The assembled complex was analyzed by application to a Superose 6 Increase 5/150 column.

Cryo-EM sample preparation and data collection

3 µl of freshly purified NbE-µOR-Fab module complex were applied onto holey gold grids (Quantifoil UltrAuFoil R1.2/1.3, 300 mesh), front blotted for 3–4 s with 1 mm additional movement (95% humidity at 15 °C) before being plunged into liquid ethane using an EM GP2 automatic plunge freezer (Leica). The cryo-EM data sets of NbE-µOR-Fab module complex were acquired on a Thermo Scientific Talos Arctica Cryo-TEM at an accelerating voltage of 200 kV. A total of 5’938 movies were recorded using a Falcon III direct electron detector at a nominal magnification of 150,000x, resulting in a pixel size of 0.9759 Å. Data were collected using EPU (Thermo Fisher Scientific) with one image per hole, a defocus range of −0.6 to −2.0 μm and a total electron dose of 40e^-/Å² distributed over 44 frames per acquisition. Data acquisition was monitored on-the-fly pre-processing using CryoSPARC v.3.3.1⁵³.

Cryo-EM image processing

All data were processed using CryoSPARC v.3.3.1 and RELION 3.1^53,54. The data processing workflow is summarized in Supplementary Fig. 4. First, raw movies were aligned and dose-weighted using patch-based motion correction⁵⁵. Contrast transfer function (CTF) parameters were estimated by patch-based CTF estimation⁵³. Only micrographs with a CTF fit better than 4.0 Å resolution were selected for further processing, resulting in a set of 5’686 micrographs. Particles were initially picked using a blob picker with a minimum and maximum diameter of 70 Å and 200 Å, respectively. The resulting 2D classes were fed into the Topaz particle-picking pipeline to increase the number and accuracy of picked particles. After several rounds of 2D classification (and removal of duplicated particles) 704’786 particles were selected. Ab initio reconstruction with multiple classes was followed by heterogeneous refinement. The best class contained 509,809 particles. 3D classification helped to further clean up the particle set. 3D refinement using non-uniform refinement and CTF refinement (global and local CTF refinement) resulted in a reconstruction of 3.1 Å resolution. To further improve the density of the µOR, refined particles were imported into RELION 3.1 for 3D classification without alignment (K = 4 and T = 12), using a soft mask on the seven transmembrane helices of the µOR. The best class showing high resolution features was selected (445,766 particles) and particles were re-imported to CryoSPARC for non-uniform refinement followed by local refinement with a soft mask around the NbE-µOR. As a result, the density for the NbE-µOR interface was significantly improved. All maps were sharpened with deepEMhancer⁵⁶. All resolution estimations were derived from Fourier shell correlation (FSC) calculations between reconstructions from two independently refined half-sets, and reported resolutions are based on the FSC = 0.143 criterion. Local resolution estimations are obtained by ResMap⁵⁷.

Cryo-EM model building and refinement

The crystal structure of inactive µOR bound to a covalent antagonist (PDB: 4DKL) was used as an initial reference²⁷. Similarly, the crystal structure of the NbE published in this manuscript was used as an initial model. For NabFab and anti-Fab Nb, the cryo-EM structure of VcNorM complex was used as initial models³¹. All initial models were fitted into cryo-EM maps using Chimera X⁵⁸, then manually built in Coot⁵⁹, iteratively and real-space refined using PHENIX⁶⁰. Model validation was performed with MolProbity⁶¹. Structural figures were generated in Chimera X⁵⁸.

Crystallization and data collection

Purified NbE was concentrated to ~30 mg/mL for crystallization. Crystals were grown using the hanging drop vapor diffusion method at 16 °C in a temperature-controlled incubator. The best NbE crystals grew in 1 M succinic acid pH 7.0, 1-2% polyethylene glycol 2000 monomethyl ether (PEG 2000 MME) and 0.1 M Hepes, pH 7.0, reaching their final size within 3 days. Crystals of NbE were cryoprotected using the well solution supplemented with 25% glycerol and flash frozen in liquid nitrogen. Diffraction data was collected at the Advanced Photon Source on beamline 23-IDD on a Pilatus 6 M detector. The final dataset was collected from a single crystal of NbE, with reflections extending to 2.85 Å resolution.

Crystallographic data processing and model refinement

Reflection data was integrated using XDS⁶², and scaled and merged using Aimless as part of the CCP4 suite⁶³. Initial phases were obtained using molecular replacement in Phaser⁶⁴, using a homology model from SwissModel⁶⁵ yielding three copies of NbE in the asymmetric unit. Iterative rounds of manual model building and automated refinement were carried out in Coot⁶⁶ and Phenix⁶⁰, respectively. The final model was refined to a R_free of 0.331 with favorable geometry (96% Ramachandran favored, 3% allowed, and 0.3% outliers). Of the three chains in the asymmetric unit, one (chain A) is characterized by exemplary density for the resolution and was used as an initial model for building of the NbE-µOR-Fab module complex.

Synthesis of linear peptides 1-6

Peptides were synthesized using Fmoc-based solid phase peptide synthesis (SPPS) on a microwave assisted peptide synthesizer (CEM Liberty Lite). The synthesis was performed on 0.1 mmol scale using preloaded Wang resin or Rink Amide resin depending on the desired C-terminal end of the peptide, being a carboxylic acid or carboxamide. Fmoc deprotection was performed at 90 °C for 3 min using a solution of 20% 4-methylpiperidine in N,N-dimethylformamide (DMF) during the entire synthesis. Each coupling was done using 5 equivalents of Fmoc-protected amino acid, with 0.5 M N,N’-diisopropylcarbodiimide (DIC) and 1 M Oxyma Pure as coupling reagents. N-terminal acetylation has been done manually using 10 equivalents of acetic anhydride and 5 equivalents of N,N’-diisopropylethylamine (DIPEA) during 1 h in DMF. After completion of the sequence, the resin was washed several times with dichloromethane (DCM), followed by the cleavage using a cocktail solution consisting of 90% trifluoroacetic acid (TFA), 5% triisopropylsilane and 5% ultrapure water during 4 h. After freeze-drying, crude peptides were obtained and purified using preparative HPLC. More specifically, a Gilson HPLC system, equipped with Gilson 322 pumps and a Supelco Discovery BIO Wide Pore C18 column (25 cm×21.2 mm, 10 µm), was used. Crude peptides were dissolved in DMSO and purified using linear gradients of water (containing 0.1% TFA) in acetonitrile (containing 0.1% TFA) at a flow rate of 20 mL min⁻¹. Finally, fractions were collected by UV detection at 214 nm, and the accompanying purities were assessed by analytical RP-HPLC using a Chromolith High Resolution C18 column (50 mm × 4.6 mm, 1.1 μm, 150 Å), after which the pure fractions with a purify exceeding 95% were combined and lyophilized to obtain the final purified peptide as a powder (TFA salt). The identity of the peptides was confirmed by High Resolution Mass Spectrometry (HRMS) on a Waters Synapt XS QTOF system with an Electron Spray inlet. Lock Mass correction was performed using a Leucine Enkephalin (100 μg μL⁻¹) solution in H₂O:MeCN (1:1) with 0.1% formic acid.

Synthesis of cyclic peptides 7 and 8

Peptides were manually synthesized using Fmoc/tBu-based SPPS as linear variants using polypropylene syringes equipped with a polyethene frit. The synthesis was performed on a 0.15 mmol scale using a 2-chlorotrityl chloride resin. Between every step, the resin was washed with DMF (3 times) and DCM (3 times). Loading of the resin (2 equiv) was performed overnight with the appropriate Fmoc/tBu-protected amino acid (1 equiv) and DIPEA (2 equiv) in dry DCM. Fmoc deprotections were performed using a solution of 20% 4-methylpiperidine in DMF for 5 and 15 min. Amino acid couplings were carried out using 4 equiv of protected amino acid, 4 equiv of N,N,N’,N’-tetramethyl)-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate (HBTU) and 8 equiv of DIPEA in DMF for 1 to 3 h. After final Fmoc removal, the peptides were cleaved from the resin under their side chain-protected forms using a solution of 20% hexafluoro-2-propanol in DCM for 90 min. After evaporation under reduced pressure, a powder was obtained which was directly used for head-to-tail cyclization in solution. Cyclization of the linear, side-chain protected peptides was directly performed on the crude peptides (0.05-0.1 mmol scales) using PyOxim (5 equiv), Oxyma Pure (1 equiv) and DIPEA (12 equiv) in DCM (0.5 mM). After 21 h the solvent was removed under reduced pressure, followed by removal of the side chain protecting groups using a mixture consisting of 90% trifluoroacetic acid, 5% triisopropylsilane and 5% ultrapure water for 90 min. After evaporation under reduced pressure, the peptides were precipitated and washed with cold diethyl ether. After freeze-drying, crude peptides were obtained and purified by preparative HPLC as described above for the linear peptides 1-6, using a ReproSil 100 C18 column (25 cm × 20 mm, 5 μm) at a flow rate of 15 mL min⁻¹. Peptides with a purity exceeding 95% were analyzed, characterized and obtained as described for the linear peptides 1-6.

Circular dichroism of linear and cyclic peptides 1-8

Circular dichroism spectra were recorded on an Applied Photophysics ChiraScanTM-plus spectrometer (ChiraScan v.4.5.1848.0) at 20 °C with nitrogen gas purging at 4 L/min. Measurements were performed in a quartz cuvette with a path length of 1 mm containing 100 µL of the peptide solutions. The spectral range of 300–180 nm was sampled with a 1 nm step size, 1 nm bandwidth and 1 s collection time per step. The background was subtracted for all spectra. Raw data was obtained in ellipticity θ (mdeg) and converted towards the mean residue molar ellipticity [θ] considering the number of amide bonds.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The EM maps of the NbE-µOR-Fab module complex have been deposited in the Electron Microscopy Data Bank (EMDB) under the accession code EMD-18541. Protein coordinates for NbE-µOR-Fab module complex structure and the crystal structure of the isolated nanobody NbE have been deposited in the Protein Data Bank (PDB) under accession codes 8QOT and 8V8K, respectively. The raw and analyzed experimental data are available in the Yareta repository [https://doi.org/10.26037/yareta:swsqivwptvga7g6yx5eb2zovqu]. Source data are provided with this paper.

References

Santos, R. et al. A comprehensive map of molecular drug targets. Nat. Rev. Drug Discov. 16, 19–34 (2017).
Article CAS PubMed Google Scholar
Sriram, K. & Insel, P. A. G protein-coupled receptors as targets for approved drugs: how many targets and how many drugs? Mol. Pharmacol. 93, 251–258 (2018).
Article CAS PubMed PubMed Central Google Scholar
Hauser, A. S., Attwood, M. M., Rask-Andersen, M., Schiöth, H. B. & Gloriam, D. E. Trends in GPCR drug discovery: new agents, targets and indications. Nat. Rev. Drug Discov. 16, 829–842 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hutchings, C. J., Koglin, M., Olson, W. C. & Marshall, F. H. Opportunities for therapeutic antibodies directed at G-protein-coupled receptors. Nat. Rev. Drug Discov. 16, 661 (2017).
Article CAS PubMed Google Scholar
Laeremans, T. et al. Accelerating GPCR drug discovery with conformation-stabilizing VHHs. Front Mol. Biosci. 9, 863099 (2022).
Article CAS PubMed PubMed Central Google Scholar
Jovčevska, I. & Muyldermans, S. The therapeutic potential of nanobodies. BioDrugs 34, 11–26 (2020).
Article PubMed Google Scholar
Manglik, A., Kobilka, B. K. & Steyaert, J. Nanobodies to study G protein-coupled receptor structure and function. Annu. Rev. Pharmacol. Toxicol. 57, 19–37 (2017).
Article CAS PubMed Google Scholar
Heukers, R., De Groof, T. W. M. & Smit, M. J. Nanobodies detecting and modulating GPCRs outside in and inside out. Curr. Opin. Cell Biol. 57, 115–122 (2019).
Article CAS PubMed Google Scholar
Huang, W. et al. Structural insights into µ-opioid receptor activation. Nature 524, 315–321 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Rasmussen, S. G. F. et al. Structure of a nanobody-stabilized active state of the β(2) adrenoceptor. Nature 469, 175–180 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Robertson, M. J. et al. Structure determination of inactive-state GPCRs with a universal nanobody. Nat. Struct. Mol. Biol. 29, 1188–1195 (2022).
Article CAS PubMed Google Scholar
Kruse, A. C. et al. Activation and allosteric modulation of a muscarinic acetylcholine receptor. Nature 504, 101–106 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Irannejad, R. et al. Conformational biosensors reveal GPCR signalling from endosomes. Nature 495, 534–538 (2013).
Article ADS CAS PubMed Google Scholar
Stoeber, M. et al. A genetically encoded biosensor reveals ___location bias of opioid drug action. Neuron 98, 963–976.e5 (2018).
Article CAS PubMed PubMed Central Google Scholar
McMahon, C. et al. Synthetic nanobodies as angiotensin receptor blockers. Proc. Natl Acad. Sci. 117, 20284–20291 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Scholler, P. et al. Allosteric nanobodies uncover a role of hippocampal mGlu2 receptor homodimers in contextual fear consolidation. Nat. Commun. 8, 1967 (2017).
Article ADS PubMed PubMed Central Google Scholar
Ma, Y. et al. Structure-guided discovery of a single-___domain antibody agonist against human apelin receptor. Sci. Adv. 6, eaax7379 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Wu, A. et al. Structural basis for the allosteric modulation of rhodopsin by nanobody binding to its extracellular ___domain. Nat. Commun. 14, 5209 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
M. A. Skiba et al. Antibodies expand the scope of angiotensin receptor pharmacology. Nat. Chem. Biol. https://doi.org/10.1038/s41589-024-01620-6 (2024).
Schlimgen, R. R. et al. Structural basis for selectivity and antagonism in extracellular GPCR-nanobodies. Nat. Commun. 15, 4611 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Corder, G., Castro, D. C., Bruchas, M. R. & Scherrer, G. Endogenous and exogenous opioids in pain. Annu. Rev. Neurosci. 41, 453–473 (2018).
Article CAS PubMed PubMed Central Google Scholar
Kieffer, B. L. & Evans, C. J. Opioid receptors: from binding sites to visible molecules in vivo. Neuropharmacology 56, 205–212 (2009).
Article CAS PubMed Google Scholar
Manglik, A. et al. Structure-based discovery of opioid analgesics with reduced side effects. Nature 537, 185–190 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Faouzi, A. et al. Structure-based design of bitopic ligands for the µ-opioid receptor. Nature 613, 767–774 (2023).
Article ADS CAS PubMed Google Scholar
Wang, H. et al. Structure-based evolution of G protein-biased μ-opioid receptor agonists. Angew. Chem. Int. Ed. Engl. 61, e202200269 (2022).
Article CAS PubMed PubMed Central Google Scholar
Volkow, N. D. & Collins, F. S. The role of science in addressing the opioid crisis. N. Engl. J. Med. 377, 391–394 (2017).
Article PubMed Google Scholar
Manglik, A. et al. Crystal structure of the µ-opioid receptor bound to a morphinan antagonist. Nature 485, 321–326 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhuang, Y. et al. Molecular recognition of morphine and fentanyl by the human μ-opioid receptor. Cell 185, 4361–4375.e19 (2022).
Article CAS PubMed Google Scholar
Koehl, A. et al. Structure of the µ-opioid receptor-Gi protein complex. Nature 558, 547–552 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, Y. et al. Structures of the entire human opioid receptor family. Cell 186, 413–427.e17 (2023).
Article CAS PubMed Google Scholar
J. S. Bloch et al. Development of a universal nanobody-binding Fab module for fiducial-assisted cryo-EM studies of membrane proteins. Proc. Natl. Acad. Sci. USA. 118 (2021).
J. A. Ballesteros, H. Weinstein, “Integrated methods for the construction of three-dimensional models and computational probing of structure-function relations in G protein-coupled receptors” in Methods in Neurosciences, S. C. Sealfon, (ed). 25, 366–428 (Academic Press, 1995).
Manglik, A. Molecular Basis of Opioid Action: From Structures to New Leads. Biol. Psychiatry 87, 6–14 (2020).
Article CAS PubMed Google Scholar
Z. Li et al. Three-dimensional structural insights have revealed the distinct binding interactions of agonists, partial agonists, and antagonists with the µ opioid receptor. Int. J. Mol. Sci. 24, 7042 (2023).
Chen, C. et al. Determination of the amino acid residue involved in [3H]beta-funaltrexamine covalent binding in the cloned rat mu-opioid receptor. J. Biol. Chem. 271, 21422–21429 (1996).
Article CAS PubMed Google Scholar
Marie-Pepin, C., Yue, S. Y., Roberts, E., Wahlestedt, C. & Walker, P. Novel “restoration of function” mutagenesis strategy to identify amino acids of the δ-opioid receptor involved in ligand binding. J. Biol. Chem. 272, 9260–9267 (1997).
Article Google Scholar
Granier, S. et al. Structure of the δ-opioid receptor bound to naltrindole. Nature 485, 400–404 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Van Holsbeeck, K., Martins, J. C. & Ballet, S. Downsizing antibodies: towards complementarity-determining region (CDR)-based peptide mimetics. Bioorg. Chem. 119, 105563 (2022).
Article PubMed Google Scholar
Obrecht, D., Chevalier, E., Moehle, K. & Robinson, J. A. β-Hairpin protein epitope mimetic technology in drug discovery. Drug Discov. Today Technol. 9, e63–e49 (2012).
Van Holsbeeck, K. et al. Nanobody loop mimetics enhance son of sevenless 1-catalyzed nucleotide exchange on RAS. Angew. Chem. Int. Ed. Engl. 62, e202219095 (2023).
Article PubMed Google Scholar
Martin, C. et al. Rational design of Nanobody80 loop peptidomimetics: Towards biased β2 adrenergic receptor ligands. Chemistry 23, 9632–9640 (2017).
Zamora, J. C. et al. Long-term antagonism and allosteric regulation of mu opioid receptors by the novel ligand, methocinnamox. Pharmacol. Res. Perspect. 9, e00887 (2021).
Article CAS PubMed PubMed Central Google Scholar
Maguire, D. R. et al. Effects of acute and repeated treatment with methocinnamox, a mu opioid receptor antagonist, on fentanyl self-administration in rhesus monkeys. Neuropsychopharmacology 45, 1986–1993 (2020).
Article CAS PubMed PubMed Central Google Scholar
Gerak, L. R. et al. Methocinnamox produces long-lasting antagonism of the behavioral effects of µ-opioid receptor agonists but not prolonged precipitated withdrawal in rats. J. Pharmacol. Exp. Ther. 371, 507–516 (2019).
Article CAS PubMed PubMed Central Google Scholar
Toyoda, Y. et al. Structural basis of α1A-adrenergic receptor activation and recognition by an extracellular nanobody. Nat. Commun. 14, 1–13 (2023).
Article Google Scholar
Hong, C. et al. Structures of active-state orexin receptor 2 rationalize peptide and small-molecule agonist recognition and receptor activation. Nat. Commun. 12, 815 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Dooley, C. T., Chung, N. N., Schiller, P. W. & Houghten, R. A. Acetalins: opioid receptor antagonists determined through the use of synthetic peptide combinatorial libraries. Proc. Natl Acad. Sci. Usa. 90, 10811–10815 (1993).
Article ADS CAS PubMed PubMed Central Google Scholar
Schiller, P. W. et al. Conversion of δ-, κ- and μ-receptor selective opioid peptide agonists into δ-, κ- and μ-selective antagonists. Life Sci. 73, 691–698 (2003).
Article CAS PubMed Google Scholar
Purington, L. C., Pogozheva, I. D., Traynor, J. R. & Mosberg, H. I. Pentapeptides displaying μ opioid receptor agonist and δ opioid receptor partial agonist/antagonist properties. J. Med. Chem. 52, 7724–7731 (2009).
Article CAS PubMed PubMed Central Google Scholar
Laschet, C., Dupuis, N. & Hanson, J. A dynamic and screening-compatible nanoluciferase-based complementation assay enables profiling of individual GPCR–G protein interactions. J. Biol. Chem. 294, 4079–4090 (2019).
Article CAS PubMed Google Scholar
Radoux-Mergault, A., Oberhauser, L., Aureli, S., Gervasio, F. L. & Stoeber, M. Subcellular ___location defines GPCR signal transduction. Sci. Adv. 9, eadf6059 (2023).
Article CAS PubMed PubMed Central Google Scholar
Götzke, H. et al. The ALFA-tag is a highly versatile tool for nanobody-based bioscience applications. Nat. Commun. 10, 4403 (2019).
Article ADS PubMed PubMed Central Google Scholar
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 (2017).
Article CAS PubMed Google Scholar
J. Zivanov et al. New tools for automated high-resolution cryo-EM structure determination in RELION-3. Elife 7, e42166 (2018).
Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron microscopy. Nat. Methods 14, 331–332 (2017).
Article CAS PubMed PubMed Central Google Scholar
Sanchez-Garcia, R. et al. DeepEMhancer: a deep learning solution for cryo-EM volume post-processing. Commun. Biol. 4, 874 (2021).
Article PubMed PubMed Central Google Scholar
Kucukelbir, A., Sigworth, F. J. & Tagare, H. D. Quantifying the local resolution of cryo-EM density maps. Nat. Methods 11, 63–65 (2014).
Article CAS PubMed Google Scholar
Meng, E. C. et al. UCSF ChimeraX: Tools for structure building and analysis. Protein Sci. 32, e4792 (2023).
Article CAS PubMed PubMed Central Google Scholar
Casañal, A., Lohkamp, B. & Emsley, P. Current developments in Coot for macromolecular model building of electron cryo-microscopy and crystallographic data. Protein Sci. 29, 1069–1078 (2020).
Article PubMed PubMed Central Google Scholar
Liebschner, D. et al. Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D. Struct. Biol. 75, 861–877 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Williams, C. J. et al. MolProbity: more and better reference data for improved all-atom structure validation. Protein Sci. 27, 293–315 (2018).
Article CAS PubMed Google Scholar
Kabsch, W. XDS. Acta Crystallogr. D. Biol. Crystallogr. 66, 125–132 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Agirre, J. et al. The CCP4 suite: integrative software for macromolecular crystallography. Acta Crystallogr D. Struct. Biol. 79, 449–461 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
McCoy, A. J. Solving structures of protein complexes by molecular replacement with phaser. Acta Crystallogr. D. Biol. Crystallogr. 63, 32–41 (2007).
Article ADS CAS PubMed Google Scholar
Schwede, T., Kopp, J., Guex, N. & Peitsch, M. C. SWISS-MODEL: an automated protein homology-modeling server. Nucleic Acids Res 31, 3381–3385 (2003).
Article CAS PubMed PubMed Central Google Scholar
Emsley, P. & Cowtan, K. Coot: model-building tools for molecular graphics. Acta Crystallogr. D. Biol. Crystallogr. 60, 2126–2132 (2004).
Article ADS PubMed Google Scholar

Download references

Acknowledgements

We thank A. Radoux-Mergault for valuable discussions and M. de Lapeyrière for generating preliminary data on NbE’s antagonism. We thank Y. Pfister for technical assistance, A. Howe for assistance with EM data collection at the DCI-Geneva (cryoGEnic) EM facility, the computing department for providing the infrastructure for cryo-EM analysis, N. Roggli for maintaining computing in the Molecular and Cellular Biology Department, and C. Bauer for his contributions to the cryoGEnic facility. W. Herrebout is acknowledged for kindly providing access to the CD spectrometer, and J. Bertouille is thanked for technical assistance. This work was supported by the Swiss National Science Foundation (PCEFP3_181282 to M.S. and 310030_185235 and TMSGI3_211581 to A.B.), the Helmut Horten Foundation, the Boninchi Foundation, and the EMBO Young Investigator Program (A.B. and M.S.). A.K. was supported by a Swiss Government Excellence PhD Scholarship (ESKAS-2021.0329) of the Swiss Federal Commission for Scholarships for Foreign Students. C.M., K.V.h. and S.B. acknowledge the Research Council of the Vrije Universiteit Brussel for the financial support through the Strategic Research Program (SRP50 and SRP95) and the Research Foundation Flanders (FWO) Hercules MZW OZR3939 grant. We thank members of the Stoeber and Boland groups for critical comments on the manuscript. Molecular graphics and analyses performed with UCSF ChimeraX, developed by the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco, with support from National Institutes of Health R01-GM129325 and the Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases.

Author information

These authors contributed equally: Jun Yu, Amit Kumar.
These authors jointly supervised this work: Andreas Boland, Miriam Stoeber.

Authors and Affiliations

Department of Molecular and Cellular Biology, University of Geneva, Geneva, Switzerland
Jun Yu, Xuefeng Zhang & Andreas Boland
Department of Cell Physiology and Metabolism, University of Geneva, Geneva, Switzerland
Amit Kumar & Miriam Stoeber
Departments of Chemistry and Bioengineering Sciences, Vrije Universiteit Brussel, Brussels, Belgium
Charlotte Martin, Kevin Van holsbeeck & Steven Ballet
Department of Plant Sciences, University of Geneva, Geneva, Switzerland
Pierre Raia
Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA, USA
Antoine Koehl
Confo Therapeutics N.V., Gent, Belgium
Toon Laeremans
Structural Biology Brussels, Vrije Universiteit Brussel, Brussels, Belgium
Jan Steyaert
VIB-VUB Center for Structural Biology, VIB, Brussels, Belgium
Jan Steyaert
Department of Pharmaceutical Chemistry, University of California, San Francisco, CA, USA
Aashish Manglik

Authors

Jun Yu
View author publications
Search author on:PubMed Google Scholar
Amit Kumar
View author publications
Search author on:PubMed Google Scholar
Xuefeng Zhang
View author publications
Search author on:PubMed Google Scholar
Charlotte Martin
View author publications
Search author on:PubMed Google Scholar
Kevin Van holsbeeck
View author publications
Search author on:PubMed Google Scholar
Pierre Raia
View author publications
Search author on:PubMed Google Scholar
Antoine Koehl
View author publications
Search author on:PubMed Google Scholar
Toon Laeremans
View author publications
Search author on:PubMed Google Scholar
Jan Steyaert
View author publications
Search author on:PubMed Google Scholar
Aashish Manglik
View author publications
Search author on:PubMed Google Scholar
Steven Ballet
View author publications
Search author on:PubMed Google Scholar
Andreas Boland
View author publications
Search author on:PubMed Google Scholar
Miriam Stoeber
View author publications
Search author on:PubMed Google Scholar

Contributions

J.Y. purified and reconstituted the NbE-µOR-NabFab complex and determined the cryo-EM structure. A.K. characterized NbE, receptor mutants, and peptidomimetics. X.Z. and P.R. performed GCI affinity measurements. C.M., K.V.h. and S.B. designed and synthesized peptide mimetics. A.M. and A.K. determined the NbE crystal structure. T.L. and J.S. generated NbE and performed binding characterization. All authors planned experiments and analyzed data. A.B. and M.S. supervised the project and wrote the manuscript with input from all other authors.

Corresponding authors

Correspondence to Andreas Boland or Miriam Stoeber.

Ethics declarations

Competing interests

M.S., A.B., S.B., A.K., J.Y., K.V.h. and C.M. have filed a priority patent application around the new cyclic peptides targeting the µOR on Sep 12, 2024, with filing number EP24200109.7. The remaining authors declare no other competing interests.

Peer review

Peer review information

Nature Communications thanks Christophe Stove, Fei Xu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Movie 1

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Yu, J., Kumar, A., Zhang, X. et al. Structural basis of μ-opioid receptor targeting by a nanobody antagonist. Nat Commun 15, 8687 (2024). https://doi.org/10.1038/s41467-024-52947-6

Download citation

Received: 16 December 2023
Accepted: 24 September 2024
Published: 09 October 2024
DOI: https://doi.org/10.1038/s41467-024-52947-6

This article is cited by

Rigid enlargement of sybodies with antibody fragments for cryo-EM analyses of small membrane proteins
- Fabian Ackle
- Sujani Thavarasah
- Markus A. Seeger
Scientific Reports (2025)