ABSTRACT
The β1-adrenoceptor (β1AR) is a G protein-coupled receptor (GPCR) activated by the hormone noradrenaline, resulting in the coupling of the heterotrimeric G protein Gs1. G protein-mediated signalling is terminated by phosphorylation of the receptor C-terminus and coupling of β-arrestin 1 (βarr1, also known as arrestin-2), which displaces Gs and induces signalling through the MAP kinase pathway2. The ability of synthetic agonists to induce signalling preferentially through either G proteins or arrestins (biased agonism)3 is important in drug development, as the therapeutic effect may arise from only one signalling cascade, whilst the other pathway may mediate undesirable side effects4. To understand the molecular basis for arrestin coupling, we determined the electron cryo-microscopy (cryo-EM) structure of the β1AR-βarr1 complex in lipid nanodiscs bound to the biased agonist formoterol5, and the crystal structure of formoterol-bound β1AR coupled to the G protein mimetic nanobody Nb806. βarr1 couples to β1AR in a distinct manner to how Gs couples to β2AR7, with the finger loop of βarr1 occupying a narrower cleft on the intracellular surface closer to transmembrane helix H7 than the C-terminal α5 helix of Gs. The conformation of the finger loop in βarr1 is different from that adopted by the finger loop in visual arrestin when it couples to rhodopsin8, and its β-turn configuration is reminiscent of the loop in Nb80 that inserts at the same position. β1AR coupled to βarr1 showed significant differences in structure compared to β1AR coupled to Nb80, including an inward movement of extracellular loop 3 (ECL3) and the cytoplasmic ends of H5 and H6. In the orthosteric binding site there was also weakening of interactions between formoterol and the residues Ser2115.42 and Ser2155.46, and a reduction in affinity of formoterol for the β1AR-βarr1 complex compared to β1AR coupled to mini-Gs. These differences provide a foundation for the development of small molecules that could bias signalling in the β-adrenoceptors.
Ligand bias arises through differential activation of the G protein pathway compared to the arrestin pathway and has been observed in ligands binding to many different GPCRs such as the μ-opioid receptor9, the angiotensin receptor (AT1R)10 and the β-adrenoceptors, β1AR and β2AR5,11. The ligands can show complex pharmacology. For example, carvedilol is an inverse agonist of β1AR when G protein activity is measured, but it is a weak agonist of the MAPK pathway activated by βarr111. In contrast formoterol is an agonist of both pathways, but stimulates the βarr1 pathway more than the G protein pathway5. Current theories favour the hypothesis that GPCRs exist in an ensemble of conformations and that ligands preferentially stabilise specific conformations12. This suggests that the conformation of a receptor bound to a heterotrimeric G protein could be different from the conformation that binds βarr. There is considerable spectroscopic evidence to support the existence of an ensemble of conformations of a GPCR even in the absence of ligands, and that specific ligands selectively stabilise specific conformations13-15. A notable study on AT1R using electron paramagnetic resonance also supports the notion that βarr biased ligands stabilise a different subset of conformations compared to G protein biased ligands16. Structures have been determined of many different GPCRs and a number of distinct states have been identified. For example, X-ray crystallography has identified two major structural states of β-adrenoceptors, comprising a number of inactive states17-19 and a number of active states6,7,20-22. In the absence of a G protein, the most thermodynamically stable state of the agonist-bound receptor is very similar to the inactive state, except for a small contraction of the ligand binding pocket. The active states have to be stabilised by binding of a G protein or a G protein-mimetic (e.g. nanobody Nb80 or Nb6B9)6,22 at the intracellular cleft that opens transiently on the cytoplasmic face of the receptor with increased frequency upon agonist binding13. Although the details of how an agonist activates the β-adrenoceptors is known in great detail6,7,17 and also the molecular basis for how G protein coupling increases ligand affinity21, there are few molecular details alluding to the mechanism of biased agonism. The structure of carvedilol-bound β1AR has been determined in the inactive state, and this suggested that interactions of the ligand with the extracellular end of H7 could promote the biased effect of the ligand as this is not observed in other unbiased ligands23. However, an active-state structure of β1AR coupled to βarr1 is required to further understanding of biased signalling, so we determined the cryo-EM structure of the formoterol-bound β1AR-βarr1 complex.
A complex of βarr1 coupled to purified β1AR could not be formed effectively in detergent, so β1AR was inserted into nanodiscs (see Methods). The β1AR construct contained six mutations to improve thermostability and a sortase sequence to allow ligation of a phosphorylated peptide (V2R6P) identical to the C-terminal sequence of the vasopressin receptor V2R. Pharmacological analysis of purified β1AR in nanodiscs indicated that coupling of βarr1 caused an increase in agonist affinity (Extended Data Fig. 1) as observed for coupling of mini-G proteins to detergent-solubilised β1AR21. This implied that the receptor coupled to arrestin was in an active state, as has been also observed for other GPCRs24. The structure of the formoterol-bound β1AR-βarr1 complex in nanodiscs (Fig. 1a-c) was determined by cryo-EM (see Methods and Extended Data Figs. 2-5), in the presence of the antibody fragment Fab30 that locks arrestin into an active conformation25 and was essential to provide sufficient mass for the alignment of the particles. The overall resolution was 3.3 Å, with the best-resolved regions of the cryo-EM map at the interface between β1AR and βarr1, reaching a local resolution of 3.2 Å (Extended Data Fig. 2 and Table 1). We also determined the structure of formoterol-bound β1AR-Nb80 complex in detergent by X-ray crystallography at 2.9 Å resolution (Fig. 1d and Table 2) to allow a direct comparison between β1AR coupled to different proteins but bound to the same ligand. The β1AR-Nb80 complex is too small (∼50 kDa of ordered protein) for structure determination by single-particle cryo-EM. The overall structures of formoterol-bound β1AR in the β1AR-Nb80 and β1AR-βarr1 complexes were virtually identical (rmsd 0.6 Å).
However, there were small significant differences in the extracellular surface, intracellular surface and in the orthosteric binding site (discussed below). β1AR in complex with βarr1 was also very similar to other structures of β1AR in an active state (rmsds ∼0.6 Å) and also to β2AR coupled to heterotrimeric Gs (PDB 3SN6; rmsd 0.8 Å)7,21. This allowed a detailed comparison between β1AR-βarr1, β2AR-Gs and β1AR-Nb80 (see below).
The overall structure of βarr1 coupled to β1AR (Fig. 2) is very similar to the X-ray structure of S-arrestin (also known as visual arrestin) coupled to rhodopsin8 (PDB 5W0P; rmsd 1.3 Å, 1853 atoms) and the structure of activated βarr1 coupled to Fab30 and the V2Rpp peptide25 (PDB 4JQI; rmsd 1.1 Å, 1861 atoms). The buried surface area of β1AR that makes contact to βarr1 (∼1200 Å2; excluding the phosphopeptide interface) is slightly smaller than the surface area of rhodopsin making contact to S-arrestin (∼1400 Å2). In addition, there is a 20° difference in tilt of arrestin relative to rhodopsin compared to β1AR (Extended Data Fig. 6). However, the regions of β1AR and rhodopsin that make contact to either βarr1 or S-arrestin respectively are conserved, as are the positions on the arrestin molecules that make contacts to the receptors (Extended Data Fig. 7 and 8). The position of the C-terminal V2R6P peptide in the cryo-EM structure is also virtually identical to the position of the peptide in the crystal structure of the βarr1-Fab30-V2Rpp complex (Extended Data Fig. 6), with the exception of the phosphate on Thr359. Phospho-Thr359 does not make contacts to βarr1 in the βarr1-Fab30-V2Rpp complex, but it appears to make contact with the tip of the lariat loop (Lys294, His295) of βarr1 in the β1AR-βarr1 complex. No density was observed in the β1AR-βarr1 cryo-EM structure equivalent to the N-terminal region of V2Rpp (RTpPPSpLGP) that is adjacent to the finger loop in the βarr1-Fab30-V2Rpp structure; this would clash with the new orientation of the finger loop and the receptor in the β1AR-βarr1 complex. The most significant difference between these three structures is the orientation and structure of the finger loop region (Fig. 2). In the activated non-receptor-bound βarr1-Fab30-V2Rpp structure, the finger loop forms an unstructured region that does not superpose with either the finger loop in S-arrestin coupled to rhodopsin or with βarr1 coupled to β1AR. In contrast, the receptor-bound finger loop of S-arrestin and βarr1 superpose, but they adopt different structures (Fig. 2). The finger loop of S-arrestin contains a short α-helical region whereas in βarr1 it forms a β-hairpin. When the arrestin molecules are aligned, it also appears that the tip of the β-hairpin of βarr1 protrudes about 5 Å deeper into the receptor than the α-helical region of S-arrestin. An interesting observation is that the CDR3 loop of nanobody Nb80 that inserts into the receptor bears a remarkable resemblance to the finger loop of βarr1 (Fig. 2), although the polypeptides run in antiparallel directions and bear little sequence similarity except for the Val-Leu residues at the tip of the loops.
Structure determination of the β1AR-βarr1 complex in nanodiscs showed it in relation to the lipid bilayer surface (Fig. 1a). This allowed the identification of 32 amino acid residues in βarr1 that potentially interact with lipid head groups, although ordered density for lipids was not observed. Two regions of arrestins had previously been suggested as interacting with lipids, the ‘344-loop’ (s18s19 loop) and the ‘197-loop’ (s11s12 loop)26,27. The nomenclature in parentheses is that implemented for arrestins in GPCRdb28. Both these loops in βarr1 contain backbone and side chains apparently buried in the head group region of the lipid bilayer and are flanked by residues where only the side chain interacts (Extended Data Figs. 8 and 9). Ten residues at the tip of the s18s19 loop were disordered and their position could therefore not be determined. Both the s11s12 and s18s19 loops in βarr1 are distant from the receptor in the complex. In contrast, two other regions of βarr1 that also appeared to interact with lipids (s8s9 and s15s16) contained residues that also interact with β1AR. The mutation L68C at the base of the finger loop is also accessible to the lipid bilayer, and is consistent with monobromobimane labelling studies, which shows a change in fluorescence when labelled arrestin couples to a GPCR29. Finally, the β-sheet s15 and the loop s14s15 contain many residues that could also potentially interact with lipids. It is notable that of the 18 residues in βarr1 that might interact with lipids through their side chains, 10 are either Lys or Arg. This suggests that negatively charged lipids such as phosphatidylinositols and/or phosphatidylserine, may play an important role in arrestin coupling30 in vivo, as has been suggested for coupling of the G protein Gs31.
The structure of the formoterol-bound β1AR-βarr1 complex was compared with the β2AR-Gs complex7. The amino acid sequences of β1AR and β2AR are 59% identical (excluding the N-terminus and C-terminus) and have very similar inactive state structures (rmsds 0.4-0.6 Å)17,19 and active state structures coupled to nanobodies (rmsds 0.4-0.6 Å)6,21,22 so the comparison is reasonable. Superposition of β1AR and β2AR from the respective complexes (rmsd 1.0 Å, 1634 atoms) shows that the long axis of βarr1 is at ∼90° angle to the long axis of Gs (Extended Data Fig. 6). This could have an influence on the coupling efficiency of G proteins compared to arrestins if the GPCR form dimers32 and whether coupling occurs or not could be dictated by which transmembrane helices form the interface. The only significant difference in secondary structure between the different receptors is that the cytoplasmic end of H6 is an additional 7 Å away from the receptor in β2AR coupled to Gs compared to β1AR coupled to βarr (Fig. 2). The cleft in the intracellular face is thus 8 Å narrower when βarr1 is coupled to β1AR compared to Gs coupled to β2AR (measured between the Cα of Ser346-Arg284 in β1AR and the Cα of Ser329-Lys267 in β2AR). The amino acid residues forming the interface between β1AR and βarr1 are very similar to those forming the interface between β2AR and Gs (Fig. 3a). In particular, both complexes rely on extensive contacts between ICL2 and the cytoplasmic end of H3 of the receptor and either βarr1 or Gs. However, there are contacts between the cytoplasmic ends of H2, H3, H7 and H8 in β1AR and βarr1 that are absent in the β2AR-Gs complex. There are also more extensive contacts between H5 and H6 of β2AR to Gs compared to the β1AR-βarr1 complex. The amino acid side chains in β1AR and β2AR at the respective interfaces are also similar in position, with the exception of Arg3.50 (Arg139 in β1AR and Arg131 in β2AR) in the DRY motif. In the β2AR-Gs complex and also in related complexes such as between the adenosine A2A receptor and Gs33, Arg3.50 extends away from the helix axis of H3 to form an interface between Tyr391 of the G protein and the hydrophobic interior of the receptor (Fig. 3c). In contrast, Arg3.50 in the β1AR-βarr1 complex adopts a different rotamer making extensive polar interactions with Asp1383.51 and Thr762.39 in the receptor and with Asp69 of βarr1 in the finger loop (Fig. 3b). The rotamer of Arg3.50 and its interactions to other β1AR side chains in the β1AR-βarr1 complex is virtually identical to those observed in inactive state structures of β1AR and in active state structures stabilised by nanobodies17,21. One final observation in the comparison between the β1AR-βarr1 complex and the β2AR-Gs structure is that the α5 helix of Gs does not overlap precisely with the position of the finger loop of βarr1 in the β1AR-βarr1 complex, with the finger loop lying closer to H7-H8 than the α5-helix (Figs. 3d & 3e).
Experimental data suggest that formoterol, carmoterol and carvedilol are arrestin biased ligands and that isoprenaline is a balanced agonist signalling equally between the G protein and arrestin pathways5,34 (see Fig. 4d for ligand structures). To identify elements that may be involved in biased signalling we therefore compared formoterol-bound β1AR coupled to βarr1 and to Nb80 (Nb80 being a known mimetic of the G protein Gs6). The largest differences were observed on the intracellular face of β1AR where the ends of H5 and H6 were closer to the receptor core by 6.7 Å (Cα of Ile241) and 1.9 Å (Cα of Lys284), respectively, when βarr1 was coupled compared to when Nb80 was coupled (Fig 4b). On the extracellular face of the receptor (Fig 4c), the largest change is in ECL3 that occludes the entrance to the orthosteric binding pocket through a 2.2 Å shift in its position (as measured at Cα of Arg317). There was no significant density for side chains in ECL3 of the β1AR-βarr1 structure, so we cannot compare changes in their interactions. In the orthosteric binding site there was a significant difference in the interactions between formoterol and the residues Ser2115.41 and Ser2155.46 (Fig. 4a). Three potential hydrogen bonds between formoterol and either Ser2155.46 (one hydrogen bond) or Ser2115.41 (two hydrogen bonds) are reduced in length by 0.4 Å, 0.8 Å and 0.2 Å, respectively, which is consistent with the decreased affinity of formoterol when βarr1 is coupled compared to mini-Gs (Extended Data Fig. 1). This also correlates with the inward movement of the cytoplasmic end of H5 by 6.7 Å. In comparison to isoprenaline, formoterol also makes extra contacts to both H6 (Val3267.36 and Phe3257.35) and ECL2 (D200), similar to the additional contacts observed in β1AR structures with bound carmoterol17 or carvedilol23 (Fig. 4a). Previous work identified a causal link between G protein coupling and changes in the extracellular positions of H6-ECL3-H7 that resulted in the decreased rates of ligand association/dissociation35 and an increase in ligand affinity21. The β1AR-βarr1 structure suggests that this region is also implicated in ligand bias, with the arrestin biased ligands of β-adrenoceptors possessing moieties that interact with the extracellular end of H6 and ECL2. This would be anticipated to affect the dynamics of the extracellular region which, in conjunction with the increased length of hydrogen bonds to H5, would affect the structure on the intracellular face where G proteins and arrestins couple.
The structure of β1AR-βarr1 suggests possibilities of designing biased agonists that result in repositioning of H5 compared to balanced agonists, and also new opportunities for developing drugs that bind on the intracellular surface of receptors. The difference in position of the α5 helix in β2AR-Gs away from H7-H8 offers the possibility of a small molecule drug binding in this region that would not affect Gs coupling, but would sterically block βarr1 binding, resulting in G protein-biased signalling. Targeting the interfaces between H6 and other helices to prevent the larger displacement of H6 in the β2AR-Gs structure compared to the β1AR-βarr1 structure would result in arrestin-biased signalling. These approaches are plausible given that allosteric ligands are known to interfere with receptor signalling when bound to the intracellular region of a receptor36 whilst others bind to the lipid-exposed surface of transmembrane domains37. The challenge remains in designing compounds that specifically target these sites.
Author contributions
Y.L. performed receptor, arrestin and zap1 expression, purification, nanodisc reconstitution and complex formation, cryo-EM grid preparation, data collection, data processing and model building. T.W. performed receptor and nanobody expression, purification and complex formation, crystallization, cryo-cooling of the crystals, X-ray data collection, data processing, and X-ray structure refinement. Y.L. and T.W. performed the pharmacological analyses. S.P. and H.D.-A. performed Fab expression, purification and validation. R.N. developed the sortase ligation of phosphorylated peptides onto β1AR. P.C.E purified mini-GS. J.G.-N. advised on cryo-EM data collection, data processing and model building. A.G.W.L. advised on X-ray data processing, structure solution and analysis. Y.L. and C.G.T. carried out structure analysis and manuscript preparation. A.K.S. managed the production of Fab. C.G.T. analysed data and managed the overall project. The manuscript was written by C.G.T and Y.L., and included contributions from all the authors.
Author information
Reprints and permissions information is available at www.nature.com/reprints. The authors declare the following competing interests: CGT is a shareholder, consultant and member of the Scientific Advisory Board of Heptares Therapeutics, who also partly funded this work. Correspondence and requests for materials should be addressed to cgt@mrc-lmb.cam.ac.uk.
Materials and Methods
Cloning, expression and purification of β1AR
The turkey (Meleagris gallopavo) β1AR constructs used for crystallization (see Extended Data Fig. 10) of the β1AR-Nb80 complex (trx-β1AR) was based on β44-m2317. The construction of trx-β1AR has been described previously21. Relative to wild-type β1AR, trx-β1AR contains truncations at the N- and C-termini (upstream of A33 and downstream of L367, respectively), and in intracellular loop 3 (C244 to R271, inclusive). Thioredoxin (E. coli trxA, with mutations C32S & C35S) was attached to the N-terminus via the linker EAAAK. trx-β1AR also contains the four thermostabilising mutations (R68S1.59, M90V2.53, F327A7.37 and F338M7.48), as well as two additional mutations C116L3.27 and C358A8.59. A hexahistidine tag is fused to the C-terminus of trx-β1AR.
The turkey β1AR construct used for electron cryo-microscopy of the β1AR-βarr1-Fab30 complex (β83S) was also based on β44-m2317. β83S (see Extended Data Fig. 10) shares the same truncations at the N-terminus and in intracellular loop 3 as trx-β1AR. β83S contains six thermostabilising mutations (M44C1.35, M90V2.53, V103C2.66, D322K7.32, F327A7.37 and F338M7.48), as well as three additional mutations C116L3.27, E130W3.41 and C358A8.59. The sequence downstream of C358A8.59 has been replaced with a linker sequence mimicking the C-terminal tail of vasopressin receptor 2 (V2R). The sequence contains a sortase recognition site (LPETG) followed by a heptahistidine tag [ARGRPLPETGGGRRHHHHHHH]. The sortase site is positioned in order to maintain the relative distance between H8 in V2R and the conserved phosphoserine triad motif following sortase assembly (Extended Data Fig. 4). β84S is identical to β83S except for an N-terminal MBP domain fusion constructed with a flexible linker.
The generation of trx-β1AR baculovirus and its expression and subsequent purification was performed as described previously21. It was solubilized and purified in decylmaltoside (DM, Generon) and eluted off the alprenolol sepharose ligand affinity column as described previously17,38,39 with 100 µM formoterol. The β83S construct was cloned into the baculovirus transfer vector pBacPAK8 (Clontech). Baculovirus expressing β83S was prepared using the flashBAC ULTRA system (Oxford Expression Technologies Ltd). β83S was expressed in Trichoplusia ni cells (Expressions Systems). Cells were grown in suspension in ESF 921 medium (Expressions Systems) to a density of 3 x 106 cells/ml. Cultures were supplemented with 5% BSA prior to infection with β83S baculovirus and incubated for 40 h. β83S was solubilized in 2% dodecylmaltoside (DDM, Generon) from the membrane fraction and further purified in 0.02% DDM by Ni2+-affinity chromatography and alprenolol sepharose ligand affinity chromatography. It was eluted from the alprenolol sepharose column with 100 µM alprenolol. β83S was further polished on a Superdex 200 Increase column to remove excess alprenolol.
β836P was generated by sortase A-mediated covalent assembly40 of purified β83S with a synthetic phosphopeptide, V2R6P (GGGDE[pS]A[pT][pT]A[pS][pS][pS]LAKDTSS, Tufts University Core Facility). The expression plasmid for sortase A(P94S, D160N, D165A, K196T) was a gift from S. Eustermann and D. Neuhaus. β83S (1 mg/ml) was incubated overnight on ice with 10-fold molar excess of V2R6P and His-tagged sortase A at 1:10 (mol/mol) enzyme:receptor ratio. The assembly mixture was pre-adjusted with NaOH to pH 7.5 prior to the addition of receptor. Unreacted receptor and enzyme were removed by negative Ni2+-affinity chromatography. β836P was further polished on a Superdex 200 Increase column.
Expression and purification of nanobody Nb80
A synthetic gene (Integrated DNA Technologies) for Nb806 was cloned into plasmid pET-26b(+) (Novagen) with a N-terminal His6 tag followed by a thrombin protease cleavage site. Expression in E. coli strain BL21(DE3)RIL (Agilent Technologies) and purification from the periplasmic fraction was as described elsewhere22, but with removal of the His6 tag was by a thrombin (Sigma) protease cleavage step before concentration to 40 mg/ml.
Formation of formoterol-bound trx-β1AR-nanobody complex and purification with detergent exchange by size exclusion chromatography
Preparation of receptor-nanobody complex was performed as described previously21. Formoterol-bound trx-β1AR (1.5 mg) was mixed with 1.5-fold molar excess nanobody (0.65 mg), cholesteryl hemisuccinate (Sigma) was added to 0.1 mg/ml in a final volume of 150 µL, and then incubated for 2 h at room temperature.
After incubation, size exclusion chromatography (SEC) was performed to separate receptor-nanobody complex from excess nanobody and to exchange the detergent from DM to HEGA-10 (Anatrace) for crystallization by vapour diffusion. A Superdex 200 10/300 GL Increase column (GE Healthcare) was used at 4 °C, the column was equilibrated with SEC buffer (10 mM Tris-Cl- pH 7.4, 100 mM NaCl, 0.1 mM EDTA, 0.35% HEGA-10) supplemented with 10 µM formoterol. Samples containing complex were mixed with 200 µl SEC buffer and centrifuged (14,000 x g, 5 minutes) immediately prior to SEC (flow rate 0.2 ml/minute), with a run time of one hour which was sufficient for a near-complete detergent exchange as indicated by quantitation of residual glycosidic detergent41. Peak fractions corresponding to complex were concentrated to 15 mg/ml for crystallization by vapour diffusion using Amicon Ultra-4 50 kDa centrifugal filter units (EMD-Millipore).
Crystallization of receptor-nanobody complex, data collection, processing and refinement
Crystals were grown in 150 + 150 nl sitting drops by vapour diffusion at 18 °C against reservoir solutions containing 0.1 M HEPES-Na+ pH 7.5 and 21-24% PEG1500. Crystals usually appeared within 2 h and grew to full size (up to 200 µm in length) within 48 h. Crystallization plates were equilibrated to 4 °C for at least 24 hours before cryo-cooling. Crystals were picked with LithoLoops (Molecular Dimensions Ltd) and dipped in 0.1 M HEPES-Na+ pH 7.5, 25% PEG1500, 5% glycerol before plunging into liquid nitrogen.
Diffraction data for the trx-β1AR-Nb80 complex with formoterol bound were collected at ESRF, Grenoble using the autonomous beamline MASSIF-142. X-ray diffraction data were collected from a single point on the crystal using automatic protocols for the location and optimal centring of crystals43. The beam diameter was selected automatically to match the crystal volume of highest homogeneous quality and was therefore collimated to 30 µm, and strategy calculations accounted for flux and crystal volume in the parameter prediction for complete data sets44. Diffraction data were processed using MOSFLM45 and AIMLESS46, the structure was solved using PHASER47 with use of the crystal structures of the active state β2AR stabilized with nanobody Nb806 and wild-type thioredoxin (PDB codes 3P0G, 2H6X) as search models. Diffraction was anisotropic, as indicated by the estimated resolution limits (CC1/2=0.3) in h,k,l directions (Extended Data Table 2). In order to retain statistically significant diffraction data, but eliminating reflections in less well diffracting directions, the data were truncated anisotropically using the UCLA Diffraction Anisotropy Server (http://services.mbi.ucla.edu/anisoscale/). Model refinement and rebuilding were carried out with REFMAC548 and COOT49.
Cloning, expression and purification of human β-arrestin-1. Wild-type human βarr1 was cloned into the pTrcHisB vector with a TEV protease-cleavable N-terminal His6 and FLAG tag. Two mutations were introduced by site-directed mutagenesis: L68C, a finger loop-mutation commonly used in the functional labelling of arrestins29, and R169E, which disrupts the polar core and predisposes arrestin to activation50. Arrestin was expressed in BL21 cells. Cells were grown in LB medium supplemented with 100 µg/ml ampicillin at 25 °C. Expression was induced with 30 µM IPTG at a cell density of OD600 0.5. The temperature was lowered to 15 °C and the cells allowed to grow for an additional 20 h. Cells were harvested and flash frozen in liquid nitrogen and stored at –80 °C. Arrestin was purified sequentially by Ni2+-affinity chromatography, TEV protease-cleavage of its N-terminal affinity tags, and Heparin chromatography, eluting off the Heparin column using 1 M NaCl. Purified arrestin was further polished on a Superdex 200 prep grade column (GE Healthcare) equilibrated in 20 mM Tris-Cl- pH 8.0, 0.1 M NaCl, 10% glycerol, 0.5 mM DTT. Peak fractions were pooled and concentrated to 20 mg/ml and flash frozen as aliquots in liquid nitrogen and stored at –80 °C.
Expression and purification of zebra fish apo-lipoprotein A-1
Zebra fish apo-lipoprotein A-1 (zap1) was expressed and purified as previously described51. Briefly, a pET-28a vector harbouring zap1 with a HRV-3C protease-cleavage N-terminal His6 tag was transformed into E. coli BL21(DE3)RIL cells. Cells were grown at 37°C in TB medium supplemented with kanamycin. Expression was induced at OD600 1-1.5 with 1 mM IPTG. The temperature was lowered to 25°C and the culture was allowed to grow for 3 h. Cells were lysed by sonication in the presence of 1% (v/v) Triton X-100. Cell lysate was clarified by centrifugation and passage through a 0.22µm filter prior to loading onto a HisTrap-FF column. The pellet from the previous centrifugation step was resuspended in buffer containing 6 M guanidine hydrochloride (GnHCl), clarified by centrifugation, and loaded onto the HisTrap column. The column was washed in successive buffers (base: 20 mM Tris-Cl- pH 7.5, 0.3 M NaCl, 20 mM imidazole) containing first, 6 M GnHCl, then 0.2% Triton X-100, followed by 50 mM Na-cholate, before eluting in 20 mM Tris-Cl- pH 7.5, 150 mM NaCl, 20 mM Na-cholate, 0.3 M imidazole. Purified zap1 was treated with HRV-3C protease in the presence of 0.5 mM TCEP to remove the His6 tag prior to polishing on a Superdex 200 Increase column equilibrated in 20 mM Tris-Cl- pH 7.5, 150 mM NaCl, 20 mM Na-cholate.
Expression and purification of Fab30
The coding region of Fab30 was synthesized by GenScript based on previously published crystal structure (PDB 4JQI). For large-scale purification, Fab30 was expressed in the periplasmic fraction of E. coli 55244 cells (ATCC) and purified using Protein L (GE Healthcare) gravity flow affinity chromatography as published previously29. Briefly, Cells harboring Fab30 plasmid were used to inoculate 2xYT and grown overnight at 30 °C. Fresh 2xYT was inoculated with 5% initial inoculum and grown for an additional 8 h at 30°C. Cells were harvested and resuspended in an equal volume of CRAP medium supplemented with ampicillin, and grown for 16 h at 30 °C.
For Fab30 purification, cells were lysed in Fab-lysis buffer (50 mM HEPES-Na+ pH 8.0, 0.5 M NaCl, 0.5% (v/v) Triton X-100, 0.5 mM MgCl2) by sonication.
Crude cell lysate was heated in a 65°C water bath for 30 min and chilled immediately on ice for 5 min. Subsequently, lysate was clarified by centrifugation at 20,000 x g and passaged through pre-equilibrated Protein L resin packed gravity flow affinity column. After binding at room temperature, beads were washed extensively with wash buffer (50 mM HEPES-Na+ pH 8.0, 0.5 M NaCl). Protein was eluted with 100 mM acetic acid into tubes containing 10% vol. neutralization buffer (1 M HEPES pH 8.0). Following elution, sample was desalted into Fab-storage buffer (20 mM HEPES pH 8.0, 0.1 M NaCl) using a pre-packed PD-10 column (GE Healthcare). Purified Fab30 was flash frozen stored in buffer supplemented with 10% glycerol.
Functional validation of purified Fab30
Functionality of purified Fab30 was assessed using co-immunoprecipitation with their reactivity towards V2Rpp-bound βarr1 as readout following a previously published protocol52. Briefly, Fab30 (1.5 µg) was incubated with purified βarr1 (2.5 µg) in the presence or absence of V2Rpp (pre-incubated with 5-10 fold molar excess compared to βarr1) in 100-200µl reaction volume. After 1h incubation at room temperature, pre-equilibrated Protein L beads were added to the reaction mixture and incubated for an additional 1 h. Subsequently, Protein L beads were washed 3-5 times using 20 mM HEPES-Na+ pH 7.4, 150 mM NaCl, 0.01% MNG to remove any non-specific binding. Bound proteins were eluted using 2×SDS loading buffer and separated by SDS-PAGE. Interaction of Fab30 with activated βarr1 was visualized using Coomassie-staining and Western blot.
Reconstitution of purified β1AR into nanodiscs and complex formation
Reconstitution was performed by adapting established protocols53. Stocks of 16:0-18:1 PC (POPC) and 16:0-18:1 PG (POPG, Avanti Polar Lipids) in chloroform were dried down under a nitrogen stream and fully solubilised in 20 mM HEPES-Na+, 150 mM NaCl, 100 mM Na-cholate to make 50 mM lipid stocks. β836P (500 µg) was reconstituted into zap1-supported nanodiscs containing 7:3 (mol/mol) POPC:POPG. Receptor, zap1 and lipids at a molar ratio of 1:10:1000 (net. 18 mM cholate) were mixed and incubated for an hour on ice. A three-fold excess of damp, pre-equilibrated Bio-Beads SM-2 (Bio-Rad) was added in batch and the sample was mixed end-over-end overnight at 4 °C. An absorption capacity of 80 mg cholate/g was used to calculate the requisite amount of polystyrene beads54. The reconstituted sample, composed of a mixture of β836P-incorporated nanodiscs and zap1/lipid-only nanodiscs, was further polished on a Superdex 200 Increase column equilibrated in 20 mM HEPES-Na+, 150 mM NaCl, 5 µM formoterol.
The nanodisc mixture was supplemented with a further 10 µM formoterol and incubated with a 2-fold excess of βarr1(L68C, R169E) for 1 h on ice. A 2-fold excess of His-tagged Fab30 was added and the mixture incubated for 1 h. Subsequently, the mixture was left to incubate in batch with 1 mL Ni-NTA resin (QIAGEN) overnight at 4°C. A pull-down of β836P-βarr1-Fab30 complex in nanodisc was performed by Ni2+-chromatography exploiting His-tagged Fab30 to remove tag-free zap1/lipid-only nanodiscs. The nanodisc-embedded ternary complex was separated from excess Fab30 on a Superdex 200 Increase column equilibrated in 10 mM HEPES-Na+ pH 7.5, 20 mM NaCl, 2 µM formoterol. SEC fractions were either used immediately for cryo-EM grid preparation or divied into aliquots and flash frozen and stored at –80°C. Grids prepared with freshly isolated complex or samples that had been subjected to a freeze/thaw cycle were identical in apparent quality.
β1AR-β-arrestin1-Fab30 cryo-grid preparation and data collection
Cryo-EM grids were prepared by applying 3 µL sample (at a protein concentration of 1 mg/ml) on glow-discharged holey gold grids (Quantifoil Au 1.2/1.3 300 mesh). Excess sample was removed by blotting with filter paper for 2-3 s before plunge-freezing in liquid ethane (cooled to −181 °C) using a FEI Vitrobot Mark IV maintained at 100% relative humidity and 4 °C. Data collection was carried out on grids made from a single preparation of β1AR-βarr1-Fab30 complex. Images were collected on a FEI Titan Krios microscope at 300 kV using a GIF Quantum K2 (Gatan) in counting mode. Data were collected in 3 independent sessions—two on LMB-Krios2; one on Diamond eBIC-Krios1—to give a total of 18,581 movies. When processing previous datasets, particles were assessed by cryoEF55 to have an orientation distribution efficiency, Eod ∼0.55, indicating moderately severe preferential orientation of the particles in freestanding ice. In order to improve orientation distribution, micrographs in this study were collected with a 30°-stage tilt. On LMB-Krios2, automated data acquisition was performed using serialEM56. Grid squares were subdivided into 3×3 grid hole-matrices. Stage shift was used to align the central grid hole within the acquisition template. Subsequently, image shift with active beam-tilt compensation was used to record from the nine holes. Large changes in sample height due to stage-tilt were compensated for by an equivalent degree of defocus adjustment, pre-determined and applied so as to normalise to the target defocus value. On eBIC-Krios1, data collection was performed using EPU (Thermo Fisher Scientific). Stage shift was used to centre individual grid holes. In all sessions, two non-overlapping exposures, aligned along the tilt axis, were collected per grid hole. Micrographs were collected with a total accumulated dose of ∼45-50 e-/Å2. Each micrograph was collected as dose-fractionated movie frames (∼1.0 e-/Å2/frame) at a dose rate of 4.5 e-/pix/sec (LMB) or 3.3 e-/pix/sec (eBIC) with an energy selection slit width of 20 eV. The datasets were collected at a magnification of 105,000× (1.1 Å/pix, LMB) and 130,000× (1.047 Å/pix, eBIC).
Data processing and model building
RELION-3.0.7 was used for all data processing unless otherwise specified57. Drift, beam-induced motion and dose-weighting were corrected in Warp-1.0.6 using a spatial resolution of 5×5 and a temporal resolution equal to the number of movie frames58. CTF estimation and determination of the focus gradient was performed in Warp using movie frame input, with 5×5 spatial resolution and a temporal resolution of 1. Micrographs were curated for quality based on ice contamination, CTF fitting quality, estimated resolution, and astigmatism, resulting in a trimmed dataset of 18,101 micrographs. Auto-picking was performed with a Gaussian blob as a template, which resulted in optimal particle picking. The CTF parameters for the picked coordinates were interpolated from the focus gradients modelled in Warp. Particles were extracted in a box-size equivalent to 264 Å and downscaled initially to 4.4 Å/pix. For each LMB-Krios2 session, micrographs were further separated into two halves, generating a total of five groups of particle stacks. Each group was processed independently. For each group, particles were subjected to two rounds of 3D classification in 6 classes using an ab initio model as reference. In the second round of 3D classification, particle distribution appeared to be dictated in part by the size of the nanodisc component (Extended Data Fig. 3). Aberrant classes of particles, such as C4 and C5 (Extended Data Fig. 3) were excluded from subsequent rounds of processing, as they probably arose from distorted nanodiscs (C4) or aggregation effects (C5) that arose during grid preparation. Particles of varying nanodisc sizes were combined, re-extracted with downscaling to 1.69 Å/pix, and refined to achieve an overall consensus alignment. Clear density could be observed for the transmembrane helices as well as two protrusions from the lipid boundary corresponding to ECL2 and ICL3, demarcating the volumes corresponding to receptor and zap1/lipid. Particles were subjected to Bayesian polishing before further refinement. Correcting for per-particle beam-induced motion consistently improved resolution by two resolution shells (according to a gold-standard FSC of 0.143). Signal subtraction was performed to remove most of the non-receptor component of the nanodisc, facilitating refinement of the β1AR-βarr1-Fab30 complex that included a thin annular layer of lipid. Subsequently, 3D classification without alignment into 6 classes (regularisation parameter, T=20) identified a subset of particles (∼8%) that refined to high resolution and showed fine map details in the receptor and arrestin regions. On trace-back, this subset of good particles constituted roughly an equal proportion of the class averages identified in the preceding round of 3D classification (i.e. class distributions based loosely on nanodisc morphology). At this stage, the good particles from the five groups were combined, re-extracted with downscaling of 1.1 Å/pix, and processed as a single dataset. The merged particle set was split according to microscope session for independent Bayesian polishing before re-merging for downstream processing. Following signal subtraction of the nanodisc and refinement, the model reached a resolution of 3.43 Å. Subsequently, refined particles were imported into and processed in RELION-3.1. On account of the image shift collection strategy used in LMB-Krios2, the particles from those two sessions were assigned to 1 of 18 optical groups—by sessions and based on position within their respective 3×3 matrices. Including the eBIC-Krios1 particles, this produced 19 optical group assignments, which were corrected independently for residual beam-tilt, anisotropic magnification, per-micrograph astigmatism, and per-particle CTF estimation. In the final refinement sequence, half maps were locally filtered between refinement iterations using SIDESPLITTER (K. Ramlaul, C.M. Palmer & C.H.S. Aylett, manuscript in preparation), an adaption of the LAFTER algorithm (Ramlaul et al. 2019) that maintains gold-standard separation between the two half maps. The final model contained 403,991 particles and reached an overall resolution of 3.3 Å with side chains visible for most of the complex (Extended Data Figs. 2 and 4). Local resolution estimates were calculated with RELION-3.1 showing the β1AR-βarr1 and βarr1-Fab30 interfaces at ∼3.2 Å and rising gradually to ∼3.7 Å at the level of the β1AR orthosteric binding site; H1 and the extracellular regions of the receptor, the C-distal end of arrestin, and CL-CH1 domains of Fab30 are at poorer resolution, with the worst regions reaching ∼4.5 Å at the most exposed edges. The final particle set was assessed to have an orientation distribution efficiency, Eod ∼0.72.
Model building and refinement was carried out using the CCP-EM59 and PHENIX60 software suites. The formoterol-bound trx-β1AR-Nb80 and βarr1-Fab30-V2Rpp crystal structures were used as starting models (PDB 6IBL and 4JQI). β1AR was modelled from A42 to A358 with a gap from R243 to R279 (inclusive) on account of weak density. The V2R6P portion of β1AR has been modelled from E372 to D384. The intervening linker region to A358 was too flexible to be resolved. Density for all phosphoresidues was well resolved. βarr1 was modelled from T6 to E359, with a gap between R331 and S340 (inclusive), which constitutes a region encompassing the C-distal “344-loop” that potentially interacts with the lipid head group region. Initial manual model building was performed in COOT49 following a jelly-body refinement in REFMAC548. Restraints for formoterol were generated using AM1 optimisation in eLBOW61. In order to better maintain geometry in the regions of weak density, secondary structure restraints, Ramachandran restraints and rotamer restraints were applied during real space refinement in PHENIX. The model followed iterative cycles of manual modification in COOT and restrained refinement in PHENIX. The final model achieved good geometry (Extended Data Table 1) with validation of the model performed in PHENIX, Molprobity62 and EMRinger63. The goodness of fit of the model to the map was carried out using PHENIX using a global model-vs-map FSC (Extended Data Fig. 4). Over fitting in refinement was monitored64 using FSCwork/FSCtest by refining a ‘shaken’ model against half map-1 and calculating a FSC of the resulting refined model against half map-2.
Expression and purification of mini-Gs
Mini-Gs (construct R393) was expressed in E. coli strain BL21(DE3)RIL and purified by Ni2+-affinity chromatography, removal of the His tag using TEV protease and negative purification on Ni2+-NTA for TEV and undigested mini-Gs removal; and final SEC to remove aggregated protein65. Purified mini-Gs was concentrated to a final concentration of 100 mg/ml in 10 mM HEPES-Na+ pH 7.5, 100 mM NaCl, 10% v/v glycerol, 1 mM MgCl2, 1 µM GDP and 0.1 mM TCEP.
Radioligand binding studies on β1AR in nanodiscs
Purified β84S was inserted into nanodiscs and ligated to the phosphorylated peptide as described for β83S. Zap1 with lipid only (no receptor) was used to determine background binding. Nanodiscs containing either empty nanodiscs, β846P, or β846P in the presence of βarr1 were prepared in the absence of ligands and then diluted into assay buffer for radioligand saturation binding studies as previously described for insect cell membranes21.
Competition binding assays
Nanodiscs containing β846P were resuspended in 20 mM HEPES-Na+ pH 7.5, 50 mM NaCl, 2.5 mM MgCl2, 0.1% BSA. Aliquots were supplemented with mini-GS construct R393 or βarr1 (final concentration 25 µM), either formoterol or isoprenaline (8 points, with the final concentration between 1 pM and 100 µM), and apyrase (final concentration 0.1 U/ml; only with mini-GS) to give a final volume of 120 µL or 220 µl. Samples were incubated at 20 °C for 1 h, before adding [3H]-DHA (Perkin Elmer) with concentrations of competing ligand in the range 1-2.5 x KD. Non-specific binding was determined by measuring binding in the presence of 100 µM unlabelled ligand. Samples were incubated at 20 °C for 1-5 h, before filtering through 96-well Multiscreen HTS GF/B filter plates (Merck Millipore) pre-soaked in 0.1% (w/v) polyethyleneimine, separating bound from unbound [3H]-DHA. Filters were washed three times with 200 µl chilled assay buffer, dried, and then punched into scintillation vials with 4 ml Ultima Gold scintillant (Perkin Elmer). Radioligand binding was quantified by scintillation counting with a Tri-Carb Liquid Scintillation Analyser (Perkin Elmer) and Ki values were determined using GraphPad Prism version 7.
Acknowledgements
The work in C.G.T.’s laboratory was funded by a grant from the European Research Council (EMPSI 339995), Heptares Therapeutics Ltd and core funding from the Medical Research Council [MRC U105197215]. The research program in A.K.S’s laboratory is supported by an Intermediate Fellowship of the Wellcome Trust/DBT India Alliance Fellowship (grant number IA/I/14/1/501285), the Science and Engineering Research Board (SERB) (EMR/2017/003804), Innovative Young Biotechnologist Award from the Department of Biotechnology (DBT) (BT/08/IYBA/2014-3) and the Indian Institute of Technology, Kanpur. H.D.-A. is supported by the National Post-Doctoral Fellowship of SERB (PDF/2016/002930) and DBT-BioCaRE grant (BT/PR31791/BIC/101/1228/2019). We thank Diamond Light Source (UK) for access and support of the cryo-EM facilities at eBIC (proposal EM17434) funded by the Wellcome Trust, MRC and BBSRC. We thank the beamline staff at the European Synchrotron Radiation Facility MASSIF-1 for help with X-ray diffraction data collection. We thank T. Nakane and P. Kolb for helpful discussions, D. Gloriam for access to unreleased data from GPCRdb, and G. Cannone from the LMB EM facility and J. Grimmett and T. Darling from LMB scientific computing for technical support during this work.