sign in sign up
<<
Home , Oliceridine

  

Biased agonism of clinically approved mu- receptor and TRV130 is not controlled by binding and signaling kinetics

Pedersen, Mie Fabricius; Wrobel, Tomasz Marcin; Märcher-Rørsted, Emil; Pedersen, Daniel Sejer; Møller, Thor Christian; Gabriele, Federica; Pedersen, Henrik; Matosiuk, Dariusz; Foster, Simon Richard; Bouvier, Michel; Bräuner-Osborne, Hans

Published in: Neuropharmacology

DOI: 10.1016/j.neuropharm.2019.107718

Publication date: 2020

Document version Publisher's PDF, also known as Version of record

Document license: CC BY

Citation for published version (APA): Pedersen, M. F., Wrobel, T. M., Märcher-Rørsted, E., Pedersen, D. S., Møller, T. C., Gabriele, F., Pedersen, H., Matosiuk, D., Foster, S. R., Bouvier, M., & Bräuner-Osborne, H. (2020). Biased agonism of clinically approved mu- agonists and TRV130 is not controlled by binding and signaling kinetics. Neuropharmacology, 166, [107718]. https://doi.org/10.1016/j.neuropharm.2019.107718

Download date: 01. okt.. 2021 Neuropharmacology 166 (2020) 107718

Contents lists available at ScienceDirect

Neuropharmacology

journal homepage: www.elsevier.com/locate/neuropharm

Biased agonism of clinically approved μ-opioid receptor agonists and T TRV130 is not controlled by binding and signaling kinetics Mie Fabricius Pedersena,b, Tomasz Marcin Wróbela,c, Emil Märcher-Rørsteda, Daniel Sejer Pedersena, Thor Christian Møllera, Federica Gabrielea, Henrik Pedersend, ∗ ∗∗ Dariusz Matosiukc, Simon Richard Fostera, Michel Bouvierb, , Hans Bräuner-Osbornea, a Department of Drug Design and Pharmacology, University of Copenhagen, Copenhagen, Denmark b Department of Biochemistry and Molecular Medicine, Institute for Research in Immunology and Cancer, Université de Montréal, Montreal, QC, Canada c Department of Synthesis and Chemical Technology of Pharmaceutical Substances, Medical University of Lublin, Lublin, Poland d Lundbeck A/S, Valby, Denmark

HIGHLIGHTS

• Oliceridine ((R)-TRV130) and display equal extreme biased agonism. • Buprenorphine display high residence time compared to oliceridine ((R)-TRV130). • Signaling and binding kinetics does not alter observational bias. • μ-opioid receptor agonists display different dependence on GRK2 and GRK5 overexpression.

ARTICLE INFO ABSTRACT

Keywords: Binding and signaling kinetics have previously proven important in validation of biased agonism at GPCRs. Here Biased agonism we provide a comprehensive kinetic pharmacological comparison of clinically relevant μ-opioid receptor ago- Binding kinetics nists, including the novel biased oliceridine (TRV130) which is in clinical trial for management. We μ-opioid receptor demonstrate that the bias profile observed for the selected agonists is not time-dependent and that agonists with Oliceridine ((R)-TRV130) dramatic differences in their binding kinetic properties can display the same degree of bias. Binding kinetics analyses demonstrate that buprenorphine has 18-fold higher receptor residence time than oliceridine. This is thus the largest pharmacodynamic difference between the clinically approved drug buprenorphine andthe clinical candidate oliceridine, since their bias profiles are similar. Further, we provide the first pharmacological characterization of (S)-TRV130 demonstrating that it has a similar pharmacological profile as the (R)-form, oliceridine, but displays 90-fold lower potency than the (R)-form. This difference is driven by a significantly slower association rate. Finally, we show that the selected agonists are differentially affected by G protein- coupled receptor kinase 2 and 5 (GRK2 and GRK5) expression. GRK2 and GRK5 overexpression greatly increased μ-opioid receptor internalization induced by , but only had modest effects on buprenorphine and oliceridine-induced internalization. Overall, our data reveal that the clinically available drug buprenorphine displays a similar pharmacological bias profile in vitro compared to the clinical candidate drug oliceridine and that this bias is independent of binding kinetics suggesting a mechanism driven by receptor-conformations. This article is part of the Special Issue entitled ‘New Vistas in Opioid Pharmacology’.

1. Introduction and with ~20% of all drugs in clinical trials targeting these receptors (Hauser et al., 2017). The conventional model of GPCR activation de- G protein-coupled receptors (GPCRs) are a major drug target family scribes that binding of agonists stabilize a receptor state that leads to through which ~30% of all medicines in the clinic mediate their action, activation of its cognate G protein, whereas antagonist binding

∗ Corresponding author. Université de Montréal, 2950, chemin de Polytechnique, Pavillon Marcelle-Coutu, Montreal, Québec, H3T 1J4, Canada. ∗∗ Corresponding author. University of Copenhagen, Universitetsparken 2, 2100, Copenhagen East, Denmark. E-mail addresses: [email protected] (M. Bouvier), [email protected] (H. Bräuner-Osborne). https://doi.org/10.1016/j.neuropharm.2019.107718 Received 26 March 2019; Received in revised form 8 July 2019; Accepted 22 July 2019 Available online 24 July 2019 0028-3908/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/). M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718 stabilizes an inactive receptor conformation. Due to this, GPCR sig- MEM, Dulbecco's phosphate buffered saline (DPBS), Hank's balanced naling was traditionally interpreted in a linear fashion as a switch that salt solution (HBSS), Lipofectamine 2000, pcDNA™5/FRT/TO vector, could either turn a given signaling pathway “on” or “off” (Costa-Neto pOG44 Flp-Recombinase expression vector and the Flp-In™ T-REx™ et al., 2016). Today, increasing evidence shows that GPCR functionality 293 cell line were purchased from Thermo Fisher Scientific (Waltham, is pluridimensional, meaning GPCRs can interact with multiple G pro- MA, USA). Zeocin, hygromycin B, blasticidin S HCl and pluronic® F-68 teins, β-arrestins, G protein-coupled receptor kinases (GRKs) and other non-ionic surfactant were purchased from Life Technologies (Carlsbad, effectors (Daaka et al., 1997; Schönegge et al., 2017; Violin et al., 2006; CA, USA). Polyethylenimine (PEI) was purchased from Polysciences Inc. Zhang et al., 2015). Ubiquitous expression of GPCRs combined with the (Warrington, PA, USA). Coelenterazine 400a was purchased from pluridimensional signaling, can lead to activation of non-desired sig- NanoLight Technology (Pinetop, AZ, USA). Tag-lite® SNAP Lumi4-Tb® naling pathways in either therapeutically relevant or non-relevant tis- was purchased from CisBio (Codolet, France). MicroScint-0, GF/C filter sues upon drug administration. As such, selective activation of signaling plates and [3H]- were purchased from Perkin Elmer pathways by a GPCR has in recent years been shown to hold a great (Waltham, MA, USA). DAMGO and GppNHp were purchased from potential in drug discovery (DeWire et al., 2013; Violin et al., 2014). Abcam (Cambridge, United Kingdom). Buprenorphine, and Ligands were found to differentially regulate these different pathways morphine were obtained from Lundbeck A/S (Valby, Denmark) or through their action on a unique receptor leading to the concept of purchased from Sigma-Aldrich (St. Louis, MO, USA). The biased agonism (Costa-Neto et al., 2016). pcDNA3.1(+)-GRK2 and pcDNA3.1(+)-GRK5 constructs were kind The concept of biased agonism has opened up the possibility to gifts from Novo Nordisk A/S (Maaloev, Denmark). design agonists that selectively activate therapeutically relevant sig- naling pathways, while not engaging pathways responsible for un- 2.2. Synthesis of TRV130 desired effects that are mediated by the same receptor. Deciphering the contribution from individual effectors to physiological outcomes is (R)- and (S)-TRV130 were synthesized in-house by a modified complex, yet necessary to fully exploit the potential of biased agonism. method to that previously reported (Chen et al., 2013) to provide ra- Taking the example of the μ-opioid receptor (μ-OR), which has been cemic TRV130 followed by chiral preparative SFC to give both en- extensively studied in relation to bias, it has been demonstrated that β- antiomers with high enantiopurity (e.r. > 99:1). See supporting in- arrestin2 knock out in mice led to decreased side-effects such as re- formation for full experimental details and characterization. spiratory depression and constipation, as well as reducing morphine tolerance but not dependence (Bohn et al., 2000; Raehal et al., 2005). 2.3. Plasmids and constructs Further, morphine reward and dependence have been shown to be driven by GRK5 but not GRK3 whereas morphine-induced locomotor The parental pcDNA5/FRT/TO vector was modified by removal of activity is dependent on GRK6 (Glück et al., 2014; Raehal et al., 2009). an MluI site using site-directed mutagenesis. For the generation of Recently, G protein biased agonists for the μ-OR have been discovered SNAP-tagged receptor, the vector was additionally modified with the that have properties comparable to morphine, yet with re- inclusion of an expression cassette containing the 21 amino acid human duced side-effects such as nausea and constipation in rodents andhu- interleukin 2 receptor alpha (hIL-2Ra) signal peptide (MDSYLLMWG- mans (Chen et al., 2013; DeWire et al., 2013; Manglik et al., 2016; LLTFIMVPGCQA), FLAG-tag (DYKDDDDK) and SNAP-tag (Maurel et al., Soergel et al., 2014), and reduced respiratory depression in rodents 2008) in frame with unique 3′ MluI and 5’ NotI sites as for receptor (DeWire et al., 2013; Manglik et al., 2016). In humans, some oliceridine insertion. Construction of pcDNA5/FRT/TO encoding the N-terminally dose regimens have been reported to reduce opioid-induced respiratory FLAG-tagged human μ-OR or FLAG-SNAP-tagged human μ-OR were adverse effects compared to morphine in some (Singla et al., 2017, thereafter conducted by restriction cloning. Vectors encoding BRET

2019; Soergel et al., 2014; Viscusi et al., 2019, 2016) albeit not all constructs with human Gαi2, GαoA and β-arrestins sequences were de- (Singla et al., 2017) clinical studies. Thus, the μ-OR is a promising drug scribed previously (Galés et al., 2005; Namkung et al., 2016; Quoyer target in relation to biased agonism. et al., 2013). Until now, most studies on biased agonism have primarily focused on comparing signaling outcomes at their individually optimal activa- 2.4. Cell lines and culturing tion time for measurement. However, a recent study on the dopamine

D2 receptor showed that agonist-dependent signaling bias can change Parental Flp-In T-REx293 cells were maintained in DMEM supplied dramatically with time, and that this characteristic was linked to the with 10% dFBS, 1% PenStrep and 100 μg/ml zeocin at 37 °C and 5% binding kinetics of the agonists (Klein Herenbrink et al., 2016). For the CO2 in a humidified incubator. Flp-In T-REx293 cells were transfected μ-OR it has also been demonstrated that expression levels of GRK2, one with the pOG44 vector encoding the Flp-Recombinase and pcDNA5/ of the key mediators in β-arrestin recruitment and thereby receptor FRT/TO encoding the FLAG-tagged human μ-OR at a ratio of 9:1 with internalization, can influence the degree of bias (Manglik et al., 2016; Lipofectamine 2000 as the transfection reagent, according to the Thompson et al., 2016). This has led to the question whether bias is in manufacturer's instructions. Flp-In T-REx293-FLAG-μ-OR were main- fact driven by kinetics and cellular background rather than conforma- tained in DMEM supplied with 10% dFBS, 1% PenStrep, 15 μg/ml tion. blasticidin S HCl and 200 μg/ml hygromycin B at 37 °C and 5% CO2. In this study, we set out to explore and compare the bias profiles of HEK293A cells were maintained in DMEM supplied with 10% FBS and clinically relevant agonists and determine whether kinetic context is 1% PenStrep at 37 °C and 5% CO2. critically important when assessing their bias profiles and whether these agonists display differences in GRK2 and GRK5 sensitivity. 2.5. BRET measurements

2. Methods and materials HEK293A cells were transfected with PEI as previously described

(Namkung et al., 2016) with 20 ng Gαi2-RlucII or 10 ng GαoA-RlucII 2.1. Materials (BRET donor) along with 300 ng Gγ2-GFP10 (BRET acceptor) to detect G protein activation. For β-arrestin recruitment recordings, cells were All reagents were purchased from Sigma-Aldrich (St. Louis, MO, transfected with 20 ng β-arrestin1-RlucII or 20 ng β-arrestin2-RlucII USA) unless otherwise stated. All restriction endonucleases, Dulbecco's (BRET donor) along with 500 ng rGFP-CAAX (BRET acceptor). In both modified eagle medium (DMEM), penicillin/streptomycin (PenStrep), BRET assays, 200 ng pcDNA5/FRT/TO-FLAG-μ-OR were co-transfected dialyzed and non-dialyzed fetal bovine serum (dFBS and FBS), Opti- and the total DNA amount was adjusted to 1 μg with pcDNA3.1(+). For

2 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

G protein activation assays, 100 ng Gβ1 was additionally co-transfected. reaction volume of 250 μl in binding buffer (50 mM Tris, 10 mM MgCl2, Forty-eight hours post transfection, cells were washed in DPBS and 0.1 mM EDTA, 0.1% ascorbic acid) containing 100 μM GppNHp to en- incubated in assay buffer (HBSS, 1 mM CaCl2, 1 mM MgCl2, 20 mM sure all receptors were G protein-uncoupled and thereby ensure a single HEPES, 0.01% pluronic® F-68, pH 7.4) for 30 min prior to stimulation receptor state (Remmers and Medzihradsky, 1991). Unless otherwise with varying concentrations of ligand for the indicated times of kinetic stated, all setups were performed using 1 μg of Flp-In T-REx293-FLAG- measurements. Two minutes prior to BRET recordings coelenterazine hMOR-1R membranes and 2 nM [3H]naloxone, since these conditions 400a was added to a final concentration of 2.5 μM. All BRET mea- avoided ligand depletion. Bound and free [3H]naloxone were, in all surements were performed using a Mithras LB940 Multimode Micro- experiments, separated by fast-flow filtration through GF/C filters fol- plate Reader (Berthold Technologies, Bad Wildbad, Germany) equipped lowed by 4 washes with ice-cold washing buffer (50 mM Tris, 10 mM 2 with the filters for BRET : 400/70 nm (donor) and 515/20 nm (ac- MgCl2, 0.1 mM EDTA). The filter was dried for 1 h at 60 °C and filter- ceptor). Raw BRET ratios were determined by calculating the ratio of bound radioactivity was quantified by scintillation counting, after ad- light intensity emitted by GFP10 or rGFP divided with the light in- dition of MicroScint-0, using a Packard TopCount microplate scintilla- tensity emitted by RlucII. The BRET ratios were normalized by sub- tion counter (Perkin Elmer, Waltham, MA, USA). tracting BRET ratios obtained with buffer-stimulation and thereafter as The dissociation rate of [3H]naloxone was determined by allowing per cent of stimulation with maximum DAMGO concentration ((BRE- [3H]naloxone to reach equilibrium with Flp-In T-REx293-FLAG-μ-OR 3 Tligand/BRETDAMGO) × 100) to all individual time points. membranes, and re-association of [ H]naloxone was prevented by ad- dition of 10 μM . Radioactivity resulting from bound [3H]na- 2.6. TR-FRET real-time internalization loxone was measured at multiple time points after addition of fentanyl. 3 To determine kobs of [ H]naloxone, association was initiated by the Internalization of the μ-OR was tracked with a method previously addition of final concentrations 10 nM, 5 nM or 13 nMof[ H]naloxone described (Roed et al., 2014). HEK293A cells were seeded at a density to the membranes and terminated by filtration at various time points. 5 3 of 4 × 10 cells/well in a 6-well plate and transfected 24 h later with The kon of [ H]naloxone was derived from kobs using the koff value de- 300 ng of pcDNA5/FRT/TO-FLAG-SNAP-μ-OR and varying amounts of termined from dissociation binding experiments. pcDNA3.1(+)-GRK2 or pcDNA3.1(+)-GRK5. In all cases, the DNA Kinetic parameters of unlabeled agonists were determined by si- amount was equalized with empty pcDNA3.1(+). Cells were trans- multaneous addition of [3H]naloxone with unlabeled agonists to Flp-In fected with Lipofectamine 2000 according to the manufacturer's re- T-REx293-FLAG-μ-OR membranes. To ensure that the rate constants commendations. Twenty-four hours post transfection, cells were de- determined for the unlabeled agonists were independent of the ligand 3 tached and re-seeded into poly-D-lysine coated white 384-well plates at concentration, [ H]naloxone was mixed with three different con- a density of 2 × 104 cells/well over night. Surface expressed FLAG- centrations of unlabeled agonists. The binding reaction was terminated SNAP-μ-OR were labeled with 10 μl of 100 nM tag-lite® SNAP Lumi4- at different time points. Tb® in Opti-MEM for 60 min at 37 °C. Cells were washed once in 40 μl assay buffer (HBSS, 1 mM CaCl2, 1 mM MgCl2, 20 mM HEPES, 0.01% 2.9. Data analysis pluronic® F-68, pH 7.4) and incubated for 5 min with 10 μl of 100 μM fluorescein-O-acetic acid diluted in assay buffer. Thereafter the cells The results were analyzed using Prism 7.0 (GraphPad Software, San were stimulated with 10 μl pre-heated ligands at 2 × final concentra- Diego, CA, USA). tion. Internalization was recorded immediately and over a 90-min Concentration-response curves were fitted using the following four period using an EnVision plate reader (Perkin-Elmer, Waltham, MA, parameter model: USA). The reading chamber temperature was maintained at 37 °C (Top Bottom) throughout the entire reading period. Emission was measured at 520/ Y= Bottom + 1+ 10( logEC50 [ A ]) n (1) 8 nm and 615/8.5 nm after excitation at 340 nm. Internalization was calculated as the ratio between donor emission and acceptor emission where Bottom and Top represent the minimal and maximal asymptote of (615/520 nm). the concentration-response curve, [A] is the logarithm molar con- centration of agonist and EC50 is the concentration of agonist required 2.7. Membrane preparation to induce 50% of the maximal response. For quantification of ligand bias, concentration response-curves Flp-In T-REx293-FLAG-μ-OR cells were treated with 50 ng/ml dox- were fitted to the following form of the operational model of agonism ycycline hyclate to induce receptor expression for 24 h. Induced cells (Kenakin et al., 2012; Klein Herenbrink et al., 2016), originally de- were harvested by washing once in ice-cold DPBS and detached with a scribed by Black and Leff (1983): n non-enzymatic cell dissociation solution and centrifuged at 500×g for n (Emax basal)[] A 5 min. The cell pellet was resuspended in ice-cold lysis buffer (20 mM ()KA E= Basal + n n HEPES, 1 mM EDTA, pH 7.5) with added protease inhibitor cocktail. [A ]n 1 + []A Cells were then homogenized with a 15 ml glass-dounce and cen- ()()KA ()KA (2) trifuged at 500×g for 5 min. The supernatant was transferred to an SV- where Emax is the maximal possible response of the system, basal is the 34 tube and centrifuged at 40,000×g for 20 min. After centrifugation, basal level of response, KA the equilibrium dissociation constant of the the supernatant was removed and the resulting pellet was washed once agonist (A), τ represents the signalling efficacy of the agonist, and n is in binding buffer (50 mM Tris, 10 mM MgCl2, 0.1 mM EDTA, 0.1% as- the slope of the function. Emax and n were shared between the agonists. corbic acid) with added protease inhibitor cocktail. After an additional Data for all agonists within each pathway were fit globally to determine round of centrifugation (40,000×g, 20 min), the supernatant was re- τ and KA. Biased agonism was then quantified by subtracting the moved and the resulting pellet was resuspened in binding buffer with log ( /KA ) DAMGO from log ( /KA ) of the agonists tested giving the protease inhibitors. The resuspended pellet was homogenized using a log ( /KA ). The log ( /KA ) was calculated by subtracting the 2 ml glass-dounce, snap frozen and stored at −80 °C. Protein con- of two given pathways. centrations were determined using the method of Bradford (1976). 3 The agonist concentration that inhibited half of the total [ H]na- loxone binding was determined using the equation: 2.8. Determination of binding kinetic constants (Top Bottom) Y= Bottom + ([A ] log IC ) All radioligand binding assays were performed at 37 °C in a total 1+ 10 50 (3)

3 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Fig. 1. Chemical structure of selected agonists. Chemical structure of DAMGO, buprenorphine, morphine, loperamide, oliceridine ((R)-TRV130) and (S)-TRV130.

2 18 where Y denotes the percentage of the total binding, Bottom and Top KF =0.5( K + K + ( K K )+ 4 k1 k 3 [ L ][ I ] 10 ) represent the minimal and maximal asymptote, IC50 represent the agonist concentration inhibiting 50% of the total binding. IC50 values where the abbreviations used are: R, receptor; [L], radioligand in na- obtained from the fit were used to calculate the KI values using the nomolar; RL, receptor-radioligand complex; I, non-labeled competitor equation described by Cheng and Prusoff (1973): in nanomolar; k1, association rate of radioligand; k2 dissociation rate of radioligand; k3 association rate of non-labeled competitor; k4 dissocia- IC50 KI = tion rate of non-labeled competitor; N, total number of receptors; t, time 1 + []A KD (4) in minutes. Fixing [L], [I], k1 and k2 allowed calculation of k3, k4 and N. The residence time (t ) of the unlabeled agonists could then be where [A] represents the [3H]naloxone concentration in molar and K R D calculated using the equation: represents the equilibrium dissociation constant of [3H]naloxone-re- ceptor complex in molar concentration. 1 3 tR = The data obtained for [ H]naloxone in dissociation binding were koff (9) fitted to a one-phase exponential decay function and t1/2 was trans- formed into koff rate constants using the following equation: ln (2) 3. Results koff = t1/2 (5) 3.1. Buprenorphine and the TRV130 enantiomers are apparent extreme G 3 The data obtained for [ H]naloxone in association binding were protein biased agonists fitted to a one-phase exponential association function to calculate the observed association rate kobs. The association rate kon was calculated as To compare the bias profiles of clinically relevant agonists, we chose described by Hill (1909): agonists with different pharmacological profiles both in vivo and in vitro.

()kobs k off The agonists selected for the study were (D-Ala(2)-mephe(4)-gly-ol(5)) kon = (DAMGO), loperamide, morphine, buprenorphine, and both []radioligand (6) enantiomers of TRV130 (Fig. 1). DAMGO, a degradation-resistant μ-OR where the koff value was determined from dissociation binding. These selective analogue of the endogenous agonist met-enkephalin (Handa rate parameters could then be used to calculate the equilibrium dis- et al., 1981) was used as the reference agonist in all studies. Loper- sociation constant of the radioligand, KD: amide, which is used in the clinic as antidiarrheal medication, was selected as it exhibits different bioavailability compared to other opi- koff KD = ates and thereby non-analgesic effects (Regnard et al., 2011; Schinkel kon (7) et al., 1996). Morphine, buprenorphine and TRV130 were selected due Association and dissociation rates for the unlabeled agonists were to their very different analgesic and pharmacological profiles in vivo calculated by fitting the data from competition kinetic experiments toa and in vitro. Morphine, is a full agonist with respect to its analgesic two-component exponential equation described originally by Motulsky effects but is also known to be a partial agonist in β-arrestin2 recruit- and Mahan (1983): ment to the μ-OR (DeWire et al., 2013; McPherson et al., 2010; Molinari et al., 2010). Buprenorphine is a potent partial agonist in G protein Nk[ L ] 109 k ( K K ) (k K ) (k K ) []RL = 1 4 FS + 4 F e KF t 4 S e KS t recruitment/activation at the μ-OR, with distinct pharmacological KK KK K K FS FS F S properties such as a bell-shaped analgesic dose-response curve and a (8) ceiling effect with respect to respiratory depression (Dahan et al., 2006, with the following variables used: 2005; Lutfy et al., 2003). Further, buprenorphine has been shown to be biased against β-arrestin2 recruitment (McPherson et al., 2010; 9 K= k1 [ L ] 10 + k2 Molinari et al., 2010). TRV130 was selected since it is a biased μ-OR

9 clinical candidate having improved analgesic properties and reduced K= k3 [ I ] 10 + k4 side-effects compared to morphine in rodents (DeWire et al., 2013) and

2 18 to some degree in humans (Singla et al., 2017; Soergel et al., 2014; KS =0.5( K + K( K K )+ 4 k1 k 3 [ L ][ I ] 10 ) Viscusi et al., 2019, 2016). However, the FDA has recently evaluated

4 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Fig. 2. G protein activation and β-arrestin recruitment at the μ-OR. The effect of increasing concentrations of theμ-

OR agonists to induce (A) Gαi2 activation, (B) GαoA acti- vation, (C) β-arrestin1 recruitment and (D) β-arrestin2 recruitment in HEK293 cells transiently expressing the human FLAG-μ-OR. Cells were stimulated with agonists for 5 min before recording of G protein activation and for 60 min before β-arrestin recruitment. Data were normal- ized to the maximal response induced by DAMGO and are expressed as mean ± SEM of three to four independent experiments performed in duplicate.

and rejected the New Drug Application and requested additional safety observed bias of the selected agonists by time-resolved measurement of data (Trevena Inc., 2018). G protein activation and β-arrestin recruitment after agonist stimula- To keep the assay conditions consistent and to avoid bottlenecks in tion. DAMGO behaved as a full agonist for both Gαi2 and GαoA acti- terms of inhibitors in second messenger assays, we focused on the two vation, as well as β-arrestin1 and β-arrestin2 recruitment and possessed major pathways from the μ-OR, Gαi2 and GαoA activation as well as β- a stable potency over time (Fig. 3A–D and Table 1). In contrast, bu- arrestin1 and β-arrestin2 recruitment, all of which are known to be prenorphine behaved as a partial agonist to all time points in Gαi2 ac- activated/recruited by the μ-OR (Chakrabarti et al., 1995; Groer et al., tivation (Fig. 3E and Table 1) and at time points up to 20 min in GαoA 2011; Saidak et al., 2006). DAMGO, buprenorphine, morphine, loper- activation (Fig. 3F and Table 1). In both cases, the potency of bupre- amide and oliceridine ((R)-TRV130) all potently activated both Gαi2 norphine increased dramatically over time. Buprenorphine did not re- (Fig. 2A) and GαoA (Fig. 2B) proteins, whereas (S)-TRV130 was 90-fold cruit the β-arrestins at any time point (Fig. 3G and H and Table 1) and less potent compared to oliceridine for both G proteins (Fig. 2A and B). therefore no formal bias could be quantified. Yet, given the difference In β-arrestin recruitment, DAMGO behaved as a full agonist both in β- in potency between the G protein and β-arrestin assays, for DAMGO arrestin1 and β-arrestin2 recruitment, whereas morphine and loper- (8.75 vs 6.39) and that the efficacy of buprenorphine toward Gαi2 and amide behaved as partial agonists (Fig. 2C and D). From the con- GαoA was a least 75% of that of DAMGO, one would expect to see a centration responses of buprenorphine, oliceridine and (S)-TRV130 in measurable β-arrestin response starting at least at concentrations of β-arrestin1 and β-arrestin2 recruitment, no meaningful fit could be 1 μM. Yet no response could be detected even at 100 μM, strongly ar- obtained as none of these agonists reached a maximum plateau within guing in favor of bias that we qualify as apparent bias. The potency of the possible concentration range applied, indicating that they are very morphine and loperamide remained constant in all pathways and both weak agonists, if at all, for these pathways. behaved as partial agonists in recruitment of the β-arrestins over the time course (Fig. 3I-L and Fig. 3M-P and Table 1, respectively). Like for buprenorphine, the TRV130 enantiomers behaved as partial agonists in

3.2. Signaling measurement kinetics does not influence biased agonism at Gαi2 activation to all time points (Fig. 3Q and U and Table 1), but only the μ-OR at early time points in GαoA activation (Fig. 3R and V and Table 1). Further, the TRV130 enantiomers induced similar extremely biased Conventionally quantification of biased agonism has been calcu- responses of β-arrestin1 and β-arrestin2 recruitment over the time lated at single time points to which the optimal signaling responses are course (Fig. 3S and T and Fig. 3X and Y and Table 1). In contrast to obtained within the given signaling pathway. Consequently, the time buprenorphine, both the TRV130 enantiomers remained stable in po- point of measurement can vary greatly from seconds to hours. tency for both G proteins within the time course (Fig. 3Q-R and Fig. 3U- Analytical methods to describe bias assume that binding equilibrium V and Table 1). No meaningful fits could be obtained for buprenorphine has been established. However, the assumption of equilibrium must be and the TRV130 enantiomers in β-arrestin recruitment at any time considered when applying these models on data from dynamic biolo- point, as these agonists failed to give a maximal plateau within the gical systems (Lane et al., 2017). If the data are obtained under non- concentrations applied. Therefore, the apparent extreme bias of these equilibrium conditions, this can have a great impact on the apparent agonists was preserved irrespective of the time point of validation. ligand potency. Accordingly, it has been reported that kinetics can be Application of the operational model on the concentration response important for the observed bias (Klein Herenbrink et al., 2016). data where meaningful fits could be obtained and the bias factors Therefore, we examined the importance of signaling kinetics for the

5 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Fig. 3. Kinetic concentration-responses of μ-OR agonists. Responses induced by increasing concentrations of (A-D) DAMGO, (E-H) buprenorphine, (I-L) morphine, (M-P) loperamide, (Q-T) oliceridine ((R)-TRV130) and (U-Y) (S)-TRV130 in HEK293A cells transiently expressing the human FLAG-μ-OR were determined at various time points between 2 and 60 min. The effects of the agonists were measured with BRET-based sensors and the data were normalized to the respective timepointof DAMGO within each pathway. The values are expressed as mean ± SEM of three to four independent experiments performed in duplicate.

6 ..Pdre,e al. et Pedersen, M.F. Table 1 Emax (% of DAMGO within each time point) and pEC50 at various time points at the μ-OR. Concentration response curves of Gαi2 and GαoA activation as wells as β-arrestin1 and β-arrestin2 recruitment were measured at time points between 2 and 60 min. The data were fitted to a four-parameter model of agonism. Emax and pEC50 values represent mean ± SEM of three to four independent experiments. No fit could be obtained for buprenorphine, oliceridine (R)-TRV130 and (S)-TRV130 in β-arrestin recruitment. Emax of the agonist-induced responses compared to DAMGO were compared using one-way ANOVA followed by Dunnett's multiple comparisons test.

Gαi2 activation

2 min 5 min 10 min 20 min 40 min 60 min

Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM

DAMGO 98 2 8.75 0.06 98 1 8.65 0.04 97 1 8.56 0.04 98 1 8.58 0.03 99 1 8.60 0.03 99 1 8.61 0.03 Buprenorhine 74c 3 8.35 0.07 82c 1 8.72 0.03 77c 1 9.01 0.04 82c 2 9.10 0.05 85c 1 9.22 0.04 84c 1 9.26 0.03 Morphine 96 3 8.14 0.08 100 1 8.09 0.04 97 2 8.07 0.05 95 4 8.04 0.10 99 2 8.14 0.05 98 2 8.17 0.05 Loperamide 102 3 9.04 0.10 108b 3 8.97 0.07 103a 2 9.12 0.04 99 2 9.05 0.06 99 1 9.15 0.03 98 1 9.20 0.03 Oliceridine (R)-TRV130 75c 1 8.45 0.05 85b 2 8.51 0.08 81c 1 8.48 0.05 87b 1 8.46 0.03 85c 1 8.50 0.05 82c 1 8.49 0.05 (S)-TRV130 78c 2 6.49 0.05 89a 3 6.56 0.07 84c 2 6.44 0.05 89a 2 6.46 0.04 86c 2 6.49 0.04 85c 2 6.53 0.05

GαoA activation

2 min 5 min 10 min 20 min 40 min 60 min

Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM

DAMGO 99 1 8.90 0.02 98 1 8.88 0.02 98 1 8.86 0.02 98 1 8.86 0.02 98 1 8.81 0.03 96 1 8.82 0.03 Buprenorhine 93b 1 8.64 0.03 91b 2 8.98 0.04 91a 1 9.23 0.03 92a 1 9.28 0.03 94 1 9.36 0.03 93 1 9.43 0.03 Morphine 103 1 8.42 0.02 103 2 8.38 0.05 107b 2 8.42 0.06 104a 2 8.45 0.04 105a 2 8.42 0.04 106b 2 8.43 0.05 Loperamide 109c 1 9.12 0.02 105b 1 9.16 0.03 107b 2 9.26 0.06 104a 2 9.29 0.05 107b 2 9.34 0.07 105b 2 9.36 0.06 Oliceridine (R)-TRV130 89c 1 8.51 0.02 91b 1 8.70 0.02 94 1 8.71 0.03 95 1 8.74 0.02 93 1 8.70 0.03 96 1 8.70 0.03 a 7 (S)-TRV130 95 1 6.66 0.03 92 1 6.74 0.04 94 1 6.71 0.03 95 1 6.72 0.03 100 1 6.69 0.04 101 2 6.71 0.06

β-Arrestin1 recruitment

2 min 5 min 10 min 20 min 40 min 60 min

Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM

DAMGO 99 3 6.39 0.08 100 1 6.17 0.03 101 1 6.19 0.03 101 1 6.17 0.03 99 1 6.13 0.03 98 2 6.06 0.04 Buprenorhine n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d Morphine 30c 6 6.90 0.69 19c 1 6.29 0.16 21c 2 6.18 0.24 18c 1 6.15 0.11 22c 2 6.19 0.19 23c 2 6.10 0.18 Loperamide 59c 2 7.11 0.09 61c 3 6.74 0.08 56c 3 6.86 0.10 59c 2 6.87 0.05 52c 2 6.99 0.06 46c 2 7.05 0.07 Oliceridine (R)-TRV130 n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d (S)-TRV130 n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d

β-Arrestin2 recruitment

2 min 5 min 10 min 20 min 40 min 60 min Neuropharmacology 166(2020)107718

Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM Emax SEM pEC50 SEM

DAMGO 99 2 6.43 0.06 101 1 6.21 0.03 100 1 6.17 0.02 98 1 6.14 0.02 98 1 6.17 0.02 99 1 6.10 0.03 Buprenorhine n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d Morphine 25c 2 6.46 0.17 22c 1 6.16 0.09 21c 1 6.24 0.16 19c 1 6.18 0.17 22c 1 6.15 0.13 25c 1 6.00 0.11 Loperamide 66c 3 6.99 0.10 68c 2 6.77 0.06 60c 2 6.76 0.07 58c 3 6.73 0.08 55c 2 6.87 0.07 51c 2 6.94 0.06 Oliceridine (R)-TRV130 n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d (S)-TRV130 n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d n.d

a: P ≤ 0.05 compared to DAMGO, b: P ≤ 0.01 compared to DAMGO, c: P ≤ 0.001 compared to DAMGO. M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

(10 log ( /KA ) ) could be calculated, revealed that no bias could be de- 3.4. μ-OR agonists display different dependency on GRK2 and GRK5 tected for any of the agonists when Gαi2 and GαoA activation were overexpression compared (Fig. 4A). Further, for the agonists where bias to β-arrestin recruitment could be calculated (namely morphine and loperamide), no Phosphorylation of the μ-OR promotes β-arrestin recruitment and change in bias was observed for any combination of pathways internalization. Multiple kinases such as PKC and GRKs have been (Fig. 4B–F). Thus, for the agonists used here, kinetics was not a critical shown to phosphorylate specific residues in the C-terminal of the re- factor when comparing Gαi2 and GαoA activation, as well as β-arrestin1 ceptor in a ligand dependent manner (Bailey et al., 2009; Doll et al., and β-arrestin2 recruitment. Bias for buprenorphine and the TRV130 2012; Just et al., 2013). Since different cell types could have diverse enantiomers between G protein activation and β-arrestin recruitment kinase abundance, overexpression of these kinases and the effect on could not be calculated due to their lack of quantifiable concentration signaling have been intensely investigated. Recently, studies have de- response curves for the β-arrestin recruitment. Indeed, reliable quan- monstrated that the ability of ligands to recruit β-arrestins can change tification of bias is strictly dependent on a robust concentration re- profoundly with GRK2 overexpression, resulting in attenuation or dis- sponse curve reaching a plateau, and from which an EC50 value can be appearing of bias (Manglik et al., 2016; Thompson et al., 2016). To obtained (Kenakin et al., 2012; Winpenny et al., 2016). Since these determine whether overexpression of GRK2 could increase inter- criteria were not fulfilled for buprenorphine and the TRV130 en- nalization mediated by the strongly biased buprenorphine and TRV130 antiomers, no reliable fit of the bias parameters could be obtained. enantiomers and to compare the effect to the other μ-OR agonists, ex- pression of GRK2 was titrated in HEK293A cells and the impact on μ-OR internalization was investigated. 3.3. Buprenorphine display significant higher receptor residence time The agonist profile of μ-OR internalization was consistent withβ- compared to the other agonists arrestin recruitment, as DAMGO was a full agonist, morphine and lo- peramide were partial agonists and buprenorphine, oliceridine and (S)- Characterization of the binding parameters of [3H]naloxone was TRV130 induced minimal internalization (Fig. 6A). In all cases, over- determined to obtain the kinetic as well as the equilibrium binding expression of GRK2 increased the agonist-induced internalization of μ- parameters of the unlabeled agonists. The kinetic constants kon and koff OR. However, the degree to which the internalization increased with of [3H]naloxone were determined from association and dissociation GRK2 overexpression was variable for the agonists (Fig. 6B and binding (Supplementary Figs. S1A and B) and were calculated to be Supplementary Fig. S4). GRK2 overexpression increased the DAMGO- 13.4 × 107 M−1min−1 ± 0.9 × 107 M−1min−1 and 1.196 induced internalization span by 53 ± 17%, whereas the morphine-in- min−1 ± 0.1 min−1, respectively. Using these constants, the dissocia- duced internalization span was drastically increased by 105 ± 18%. tion constant pKD was calculated to be 8.05 ± 0.03, which was con- Loperamide was the least affected by GRK2 overexpression with a span sistent with data from saturation binding (pKI: 8.07 ± 0.04) increase of 21 ± 7%. Further, the internalization span induced by the (Supplementary Fig. S1C). To further confirm the kinetic parameters of biased agonists also increased upon GRK2 overexpression, with bu- 3 3 [ H]naloxone, kobs was plotted against the [ H]naloxone concentration prenorphine-induced internalization increasing by 46 ± 10% and oli- used in the association kinetic experiments. If the system follows the ceridine ((R)-TRV130) and (S)-TRV130 by 67 ± 9% and 55 ± 12%, law of mass action, kobs should increase in a linear manner with in- respectively. Both Emax and the span increase for buprenorphine, oli- creasing radioligand concentration (Hill, 1909; Motulsky and ceridine and (S)-TRV130 were not significantly different, suggesting Christopoulos, 2003). The slope of this correlation should equal kon and GRK2 overexpression impact these agonists equally in the system used the intercept with the ordinate axis should equal koff. Indeed, a linear here. Another important observation from the μ-OR internalization correlation was obtained between these variables (kon and koff calcu- experiments is that GRK2 overexpression altered morphine from a low lated as 11.6 × 107 M−1min−1 ± 3.3 × 107 M−1min−1 and 1.285 efficacy partial agonist into a full agonist, whereas the biased agonists min−1 ± 0.2 min−1, respectively) (Supplementary Fig. S1D). These buprenorphine and the TRV130 enantiomers were altered from being parameters were not statistically different from the kinetic constants inactive into medium efficacy partial agonists (Fig. 6B). Since GRK5 has obtained from association and dissociation kinetic data. been reported to be of importance for μ-OR phosphorylation, anti-no- To directly compare the kinetic binding parameters of the selected ciception, tolerance, and dependence (Doll et al., 2012; Glück et al., agonists, competition association binding of [3H]naloxone was con- 2014), we also assessed the effect of titrating this kinase. In general, the ducted as described by (Motulsky and Mahan, 1983). DAMGO, bupre- agonist-induced μ-OR internalization response increased in a similar norphine, morphine, loperamide and oliceridine bound the receptor manner upon GRK5 overexpression as observed for GRK2 over- with an association rate within the same magnitude, whereas (S)- expression (Fig. 6D and Supplementary Fig. S5). However, the actual TRV130 displayed a slow association rate that was ~45 times slower increase in μ-OR internalization was slightly less with a span increase of than oliceridine (Fig. 5A, Table 2 and Supplementary Fig. S2). This DAMGO-induced internalization of 34 ± 11%. Here morphine was difference in kon between the TRV130 enantiomers also explains the altered from being a low-efficacy partial agonist to a high-efficacy potency differences observed for these two agonists asthe koff values partial agonist with a span increase of 78 ± 8%. Further, under these are equal (Table 2). Buprenorphine displayed a slow dissociation rate conditions, loperamide was least affected with an increase of 17 ±8%. which agrees with previous observations (Tzschentke, 2002)(Table 2), The biased agonists buprenorphine, oliceridine, and (S)-TRV130 en- whereas DAMGO, morphine, loperamide and the TRV130 enantiomers antiomers went from being inactive to low efficacy agonists upon GRK5 all dissociated rapidly from the receptor (Fig. 5B, Table 2 and overexpression with increased responses of 28 ± 16%, 49 ± 12%,

Supplementary Fig. S2). These data revealed that the two strongly and 46 ± 26%, respectively. Again, the Emax and the span increase for biased agonists buprenorphine and oliceridine possessed residence buprenorphine, oliceridine and (S)-TRV130 were not significantly dif- times at the opposite ends of the spectrum of values obtained for the ferent, suggesting GRK5 overexpression, like GRK2 overexpression, agonists. Here, buprenorphine had an 18-fold higher residence time at impacts μ-OR internalization mediated by these biased agonists equally the μ-OR compared to oliceridine (Table 2). Comparison of the affinity in this system. Ligand-independent μ-OR internalization increased constants obtained by kinetic and equilibrium competition binding slightly with increasing amounts of GRK2 and GRK5, yet this increase (Supplementary Fig. S3) demonstrated an excellent linear correlation was not significantly different from the ligand-independent μ-OR in- (Fig. 5C). ternalization without co-transfection of GRKs (Fig. S4H and Fig. S5H). From the SNAP-tag donor signals, it could be observed that over- expression of GRK2 or GRK5 resulted in similar cell-surface expression of the μ-OR (Fig. 6C). However, further increases in GRK

8 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Fig. 4. μ-OR agonists display no kinetic change in bias. Concentration response curves obtained at various time points in G protein activation and β-arrestin recruitment were fitted to the operational model of agonism. The transduction coefficients log(τ/KA) were calculated for each agonist in each pathway and Δlog(τ/

KA) values were determined with DAMGO as the reference agonist. The Δlog(τ/KA) values obtained in the different pathways were subtracted to calculate ΔΔlog(τ/ log ( /KA ) KA). The bias values 10 between Gαi2 activation, GαoA activation, β-arrestin1 recruitment and β-arrestin2 recruitment were plotted in a bias web. The ΔΔlog

(τ/KA) values were analyzed by two-way ANOVA with all combinations of time and agonist and corrected for multiple comparison with Tukey's test (significance P < 0.05).

Fig. 5. Binding kinetic parameters for μ-OR agonists. Kinetic binding parameters were determined by competition kinetic binding between the indicated agonists and 3 [ H]-Naloxone on membranes from Flp-In T-REx293-FLAG-μ-OR cells. Kinetic traces were fitted to a two-component exponential equation to obtain (A) kon and (B) koff. (C) The dissociation constant (pKD) for each agonist was calculated from kon and koff and plotted against the dissociation constant (pKI) obtained from equilibrium competition binding. The values are expressed as mean ± SEM from three independent experiments performed in duplicate.

9 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Table 2 3 3 Kinetic parameters for [ H]naloxone and unlabelled μ-OR agonists. Association (kon) and dissociation (koff) parameters and affinity constants of[ H]naloxone and unlabelled agonists from kinetic and equilibrium competition binding. Data represent mean ± SEM from three independent experiments. Association and dis- sociation constants kon and koff from kinetic binding and law of mass action were compared with a Student's t-test and no significant difference was found (P > 0.05). 3 7 −1 −1 For determination of kon and koff of the unlabelled agonists the binding parameters of [ H]naloxone from kinetics were used (kon = 13.4 × 10 M min and −1 koff = 1.196 min ).

Kinetics Equilibrium Law of mass action

kon ( ± SEM) koff ( ± SEM) pKD ( ± SEM) tR ( ± SEM) pKI ( ± SEM) kon ( ± SEM) koff ( ± SEM) /× 107 M−1min−1 /min−1 /min /× 107 M−1min−1 /min−1

[3H]naloxone 13.4 ( ± 0.9) 1.196 ( ± 0.1) 8.05 ( ± 0.03) 0.85 ( ± 0.08) 8.07 ( ± 0.04) 11.6 ( ± 3.3) 1.285 ( ± 0.2) DAMGO 3.24 ( ± 0.6) 1.029 ( ± 0.2) 7.50 ( ± 0.02) 1.04 ( ± 0.20) 7.34 ( ± 0.2) Buprenorphine 7.97 ( ± 0.8) 0.106 ( ± 0.02) 8.9 ( ± 0.13) 10.68 ( ± 2.83) 8.37 ( ± 0.09) Morphine 1.78 ( ± 0.08) 1.388 ( ± 0.1) 7.11 ( ± 0.09) 0.73 ( ± 0.06) 7.02 ( ± 0.2) Loperamide 4.89 ( ± 0.8) 0.912 ( ± 0.09) 7.72 ( ± 0.06) 1.12 ( ± 0.13) 7.64 ( ± 0.2) Oliceridine (R)-TRV130 3.25 ( ± 0.7) 1.763 ( ± 0.2) 7.25 ( ± 0.08) 0.59 ( ± 0.09) 7.29 ( ± 0.1) (S)-TRV130 0.0732 ( ± 0.02) 1.896 ( ± 0.7) 5.62 ( ± 0.14) 0.75 ( ± 0.32) 5.73 ( ± 0.1)

overexpression were not possible in this system, as the highest amount loperamide and oliceridine all potently activated Gαi2 and GαoA, used here already led to a slight reduction in receptor cell-surface ex- whereas (S)-TRV130 was significantly less potent (90-fold compared to pression. oliceridine). In this study, buprenorphine and oliceridine behaved as

partial agonists to all time points in Gαi2 activation, and to distinct time points in GαoA activation with similar Emax. Although buprenorphine 4. Discussion has been shown to be a partial agonist compared to oliceridine in in- hibiting cAMP accumulation in a previous study (DeWire et al., 2013), Biased agonism has been increasingly studied over the last decade, the difference in measured efficacy between our study and DeWire etal. providing compelling evidence for the potential therapeutic utility of is likely caused by a difference in measurement point (G protein acti- this approach (Singla et al., 2017; Zhang et al., 2015). To the best of our vation vs. cAMP accumulation) and/or differences in receptor orG knowledge, this is the first study that directly compares the bias profile protein expression levels, which is well-known to affect measured ef- of clinically relevant μ-OR agonists possessing distinct pharmacological ficacy (Bräuner-Osborne et al., 1996). Furthermore, it should be noted profiles, taking into account the kinetic context and comparing theki- that the efficacy towards the G protein/cAMP pathways was measured netic properties with the biased agonist oliceridine, which has been in by assessing the specific Gαi2 and GαoA activation, and therefore it clinical trial as a novel treatment for acute severe pain (Singla et al., cannot be excluded that buprenorphine and the TRV130 enantiomers 2017, 2019; Soergel et al., 2014; Viscusi et al., 2019, 2016). Here, we possess a different activation pattern (more or less partiality) for other have demonstrated that DAMGO, buprenorphine, morphine,

Fig. 6. The effect of GRK2 or GRK5 overexpression on agonist-mediated μ-OR internalization. (A) Internalization of FLAG-SNAP-μ-OR transiently transfected into HEK293A cells with no co-transfection of GRK. (B and D) The effect of titration of human GRK2 or GRK5 on agonist-induced internalization of FLAG-SNAP-μ-OR. In both (A), (B) and (D) cells were stimulated with 100 μM agonist and data represent area under the curve (AUC) from real-time internalization over a time-course of 90 min for which the buffer has been subtracted. (C and E) donor signals from the labeled SNAP-tag from the different transfection conditions, representingthe number of receptors at the cell-surface. Maximum μ-OR internalization induced by oliceridine, (S)-TRV130 and buprenorphine was compared with one-way ANOVA corrected with Tukey's test for multiple comparison and no significant difference was found (P > 0.05). Data represent mean ± SEM from three independent experiments carried out in triplicate.

10 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

G proteins of the Gαi/o family. In β-arrestin1 and β-arrestin2 recruit- (Atwood et al., 2011) and studies have shown that overexpression of ment, DAMGO behaved as a full agonist and morphine and loperamide GRK2 in HEK293 cells can affect the signaling of biased agonists behaved as partial agonists. Buprenorphine, oliceridine and (S)-TRV130 (Manglik et al., 2016; Thompson et al., 2016). Therefore, we made a showed a very strong bias towards G protein activation, as they failed to direct comparison of the functional consequence of the selected ago- induce significant recruitment of any of the β-arrestins. nists upon GRK2 titration. These data revealed that the agonists dis- Since previous observations have suggested that apparent bias can played differences in GRK2-dependent potentiation of μ-OR inter- be highly dependent on the time point of comparison (Klein Herenbrink nalization. While morphine-induced receptor internalization was et al., 2016), we investigated the impact of kinetics on the signaling of increased dramatically, loperamide-induced internalization only in- the selected agonists using real-time BRET and receptor internalization creased slightly. In agreement with the negligible β-arrestin recruit- as well as kinetic radioligand binding assays. In general, all the selected ment by buprenophine and the TRV130 enantiomers, these agonists agonists sustained a stable concentration-dependent response of the only induced marginal receptor internalization without GRK2 over- signaling pathways investigated, with buprenorphine being the only expression. Yet, these G protein biased agonists all increased μ-OR in- agonist that displayed an increase in potency over time in both Gαi2 and ternalization upon GRK2 overexpression to a similar extent, which was GαoA activation. This could be explained by the low dissociation rate well below the increase observed for morphine. Interestingly, these data leading to increased receptor occupancy over time. Furthermore, bu- demonstrate that despite their pronounced G protein bias, buprenor- prenorphine and both TRV130 enantiomers were apparently extremely phine and the TRV130 enantiomers promote μ-OR internalization in biased against β-arrestin recruitment. Application of the operational cells with high GRK2 expression, which suggests that these biased drugs model on pathways where meaningful fits could be obtained, confirmed could induce receptor-internalization in tissues where GRK2 is more that there was no bias between signaling pathways for the agonists abundant, albeit to a lesser extent than morphine or DAMGO. Due to between Gαi2 and GαoA activation. Buprenorphine's time dependent the observed role of GRK5 in μ-OR biology (Doll et al., 2012; Glück increase in potency of G protein activation were similar for both G et al., 2014) we also assessed the impact of overexpressing this kinase. proteins, resulting in no generation of bias between the two G protein In general, GRK5 increased agonist-induced μ-OR internalization, albeit subtypes. Buprenorphine, oliceridine and (S)-TRV130 displayed an to a lesser extent than observed for GRK2. This could be explained by apparent extreme bias towards G protein activation, which preclude the relatively high abundance of endogenous GRK5 in HEK293 cells calculation of bias factors. For the agonists where bias factors could be (Atwood et al., 2011). calculated in β-arrestin recruitment, no change in bias was observed Collectively, our study is the first to directly compare bias of clini- between any of the pathways (Fig. 4). Therefore, the overall bias profile cally relevant agonists at the μ-OR and combine with a kinetic analysis for all the selected agonists was retained over time, which allowed us to approach. Overall, we show that kinetics was not of critical importance conclude that kinetic context was not of critical importance. for the bias profile, which remained the same irrespective of the time Using kinetic competition binding we could directly compare the kon point measured. Further, we demonstrate that buprenorphine and oli- and koff values of the μ-OR agonists. These data showed that all the ceridine displayed the same apparent extreme bias in vitro. Nonetheless, selected agonists possessed a kon within the same order of magnitude at this point, it cannot be ruled out that the reported beneficial clinical except (S)-TRV130, which had a significantly lower association rate. effects of buprenorphine, in terms of lower respiratory depression The slower association rate likely accounts for the lower potency of (S)- (Dahan et al., 2006, 2005; Megarbane et al., 2006) are due to its low TRV130 in Gαi2 and GαoA activation compared to the rest of the ago- efficacy and complex pharmacology (Khanna and Pillarisetti, 2015) nists. It also demonstrates that small changes in stereochemistry can rather than its biased profile (Wootten et al., 2018). A direct compar- have dramatic effects on the binding kinetics. However, the difference ison of the physiological effects of these two agonists could give novel in binding kinetics between oliceridine and (S)-TRV130 did not alter and valuable insights into the benefits of biased agonists in the clinic. the overall bias profile, indicating that the change in the 3D orientation However, it should also be kept in mind, that we have shown that the of TRV130 around its chiral center, that influences its affinity and po- two agonists differ in the kinetic binding profile. Thus, the biggest tency, does not affect its towards Gαi vs. β-ar- difference observed between the agonists in this study, beside their bias restin. Moreover, it clearly shows that the bias of TRV130 is not driven profile, is the residence times, where we observed an 18-fold difference by its potency. Calculation of koff revealed that DAMGO, morphine, between buprenorphine and the clinical candidate oliceridine, which loperamide, oliceridine and (S)-TRV130 have dissociation rates within potentially also could lead to significantly different clinical effects the same magnitude, whereas buprenorphine dissociates significantly (Copeland, 2016). It will therefore be important to determine if dif- more slowly from the μ-OR than the other ligands and accordingly had ferences in therapeutic profile is caused by differences in ligand bias, an 18-fold higher residence time compared to oliceridine. ligand binding kinetics or even other parameters such as ADME prop- Considering the differences in dissociation rates between bupre- erties. Finally, this study underlines that agonists can display a similar norphine and the TRV130 enantiomers in relation to their G protein overall degree of signaling bias despite possessing different inherent biased nature, it can be concluded that bias is not necessarily driven by binding kinetic properties. low koff values. This may however be different for other receptors and/ or ligands, as has been suggested for the dopamine receptor (Klein Herenbrink et al., 2016). Since binding kinetics are unlikely to be the Financial disclosures driving force behind the bias for the agonists selected in this study, it appears likely that the bias is more related to different conformational All authors reported no financial interest related to these studies and states of the receptor. The majority of research that supports a con- declared no conflicts of interests. formation-driven mechanism of bias has relied, like this study, on in- direct pharmacological data. Yet, increasing biochemical, biophysical and structural studies are appearing, which provide direct evidence of Author contributions ligand-dependent GPCR and associated transducer (G protein and β- arrestin) conformational states (Furness et al., 2016; Kahsai et al., 2011; M.F.P., M.B. and H.B.-O. designed the study; T.M.W., E.M.-R., Liang et al., 2018; Rahmeh et al., 2012; Shukla et al., 2008). D.S.P., H.P. and D.M. synthesized and purified the TRV130 en- Different cellular backgrounds can also result in differences inap- antiomers; M.F.P., T.C.M. and F.G. carried out pharmacological ex- parent bias (Gesty-Palmer et al., 2009, 2006). This could relate to dif- periments; S.R.F cloned the SNAP-tag expression cassette; M.F.P., M.B ferences in abundance of effector proteins. GRK2 expression has been and H.B.-O. wrote the manuscript. All authors thoroughly read and reported to be low in HEK293 cells compared to other cell types commented on the manuscript.

11 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Acknowledgements 167, 1259–1270. Furness, S.G.B., Liang, Y.-L., Nowell, C.J., Halls, M.L., Wookey, P.J., Dal Maso, E., Inoue, A., Christopoulos, A., Wootten, D., Sexton, P.M., 2016. Ligand-dependent modulation We thank H. Lundbeck A/S and Dr. Asuka Inoue for generous gifts of μ- of G protein conformation alters drug efficacy. Cell 167, 739–749 e11. OR ligands and HEK293A cells, respectively. H.B.-O. acknowledges fi- Galés, C., Rebois, R.V., Hogue, M., Trieu, P., Breit, A., Hébert, T.E., Bouvier, M., 2005. nancial support from the Independent Research Fund Denmark | Medical Real-time monitoring of receptor and G-protein interactions in living cells. Nat. Methods 2, 177–184. Sciences (4183-00131A), the Carlsberg Foundation (CF17-0132) and the Gesty-Palmer, D., Chen, M., Reiter, E., Ahn, S., Nelson, C.D., Wang, S., Eckhardt, A.E., Toyota Foundation (9310-F). M.F.P. thank the Oticon Foundation (16- Cowan, C.L., Spurney, R.F., Luttrell, L.M., Lefkowitz, R.J., 2006. Distinct β-arrestin- 0668) and the Knud Højgaards Fond (16-02-0142) for financial support. and G protein-dependent pathways for parathyroid hormone receptor-stimulated D.S.P., E.M.-R. and T.M.W thank the Lundbeck Foundation (R171-2014- ERK1/2 activation. J. Biol. Chem. 281, 10856–10864. Gesty-Palmer, D., Flannery, P., Yuan, L., Corsino, L., Spurney, R., Lefkowitz, R.J., Luttrell, 361) and the Carlsberg Foundation (CF14-0052) for financial support. L.M., 2009. A beta-arrestin-biased agonist of the parathyroid hormone receptor S.R.F. acknowledges funding from the Independent Research Fund (PTH1R) promotes bone formation independent of G protein activation. Sci. Transl. Denmark | Medical Sciences (4183-00243B), the Lundbeck Foundation Med. 1, 1ra1. Glück, L., Loktev, A., Moulédous, L., Mollereau, C., Law, P.Y., Schulz, S., 2014. Loss of (R181-2014-2826), and the Augustinus Foundation (16-0313). T.C.M. morphine reward and dependence in mice lacking G protein-coupled receptor kinase acknowledges funding from the European Union’s Horizon 2020 research 5. Biol. Psychiatry 76, 767–774. and innovation programme under the Marie Sklodowska-Curie grant Groer, C.E., Schmid, C.L., Jaeger, A.M., Bohn, L.M., 2011. Agonist-directed interactions with specific β-arrestins determine μ-opioid receptor trafficking, ubiquitination, and agreement No 797497. We acknowledge the financial support from a dephosphorylation. J. Biol. Chem. 286, 31731–31741. Canadian Institute for Health Research (CIHR) Foundation grant (FDN- Handa, B.K., Land, A.C., Lord, J.A., Morgan, B.A., Rance, M.J., Smith, C.F., 1981. 148431) to MB. MB also holds a Canada Research Chair in Signal Analogues of beta-LPH61-64 possessing selective agonist activity at mu- re- ceptors. Eur. J. Pharmacol. 70, 531–540. Transduction and Molecular Pharmacology. Hauser, A.S., Attwood, M.M., Rask-Andersen, M., Schiöth, H.B., Gloriam, D.E., 2017. Trends in GPCR drug discovery: new agents, targets and indications. Nat. Rev. Drug Appendix A. Supplementary data Discov. 16, 829–842. Hill, A.V., 1909. The mode of action of nicotine and curari, determined by the form of the contraction curve and the method of temperature coefficients. J. Physiol. 39, Supplementary data to this article can be found online at https:// 361–373. doi.org/10.1016/j.neuropharm.2019.107718. Just, S., Illing, S., Trester-Zedlitz, M., Lau, E.K., Kotowski, S.J., Miess, E., Mann, A., Doll, C., Trinidad, J.C., Burlingame, A.L., von Zastrow, M., Schulz, S., 2013. Differentiation of opioid drug effects by hierarchical multi-site phosphorylation. Mol. Pharmacol. 83, References 633–639. Kahsai, A.W., Xiao, K., Rajagopal, S., Ahn, S., Shukla, A.K., Sun, J., Oas, T.G., Lefkowitz, R.J., 2011. Multiple ligand-specific conformations of the β2-adrenergic receptor. Nat. Atwood, B.K., Lopez, J., Wager-Miller, J., Mackie, K., Straiker, A., 2011. Expression of G Chem. Biol. 7, 692–700. protein-coupled receptors and related proteins in HEK293, AtT20, BV2, and N18 cell Kenakin, T., Watson, C., Muniz-Medina, V., Christopoulos, A., Novick, S., 2012. A simple lines as revealed by microarray analysis. BMC Genomics 12, 14. method for quantifying functional selectivity and agonist bias. ACS Chem. Neurosci. Bailey, C., Oldfield, S., Llorente, J., Caunt, C., Teschemacher, A., Roberts, L., McArdle, C., 3, 193–203. Smith, F., Dewey, W., Kelly, E., Henderson, G., 2009. Involvement of PKCα and G- Khanna, I.K., Pillarisetti, S., 2015. Buprenorphine - an attractive opioid with under- protein-coupled receptor kinase 2 in agonist-selective desensitization of μ-opioid utilized potential in treatment of chronic pain. J. Pain Res. 8, 859–870. receptors in mature brain neurons. Br. J. Pharmacol. 158, 157–164. Klein Herenbrink, C., Sykes, D.A., Donthamsetti, P., Canals, M., Coudrat, T., Shonberg, J., Black, J.W., Leff, P., 1983. Operational models of pharmacological agonism. Proc. R.Soc. Scammells, P.J., Capuano, B., Sexton, P.M., Charlton, S.J., Javitch, J.A., Lond. Ser. B Biol. Sci. 220, 141–162. Christopoulos, A., Lane, J.R., 2016. The role of kinetic context in apparent biased Bohn, L.M., Gainetdinov, R.R., Lin, F.-T., Lefkowitz, R.J., Caron, M.G., 2000. μ-Opioid agonism at GPCRs. Nat. Commun. 7, 10842. receptor desensitization by β-arrestin-2 determines morphine tolerance but not de- Lane, J.R., May, L.T., Parton, R.G., Sexton, P.M., Christopoulos, A., 2017. A kinetic view pendence. Nature 408, 720–723. of GPCR allostery and biased agonism. Nat. Chem. Biol. 13, 929–937. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram Liang, Y.-L., Khoshouei, M., Glukhova, A., Furness, S.G.B., Zhao, P., Clydesdale, L., Koole, quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, C., Truong, T.T., Thal, D.M., Lei, S., Radjainia, M., Danev, R., Baumeister, W., Wang, 248–254. M.-W., Miller, L.J., Christopoulos, A., Sexton, P.M., Wootten, D., 2018. Phase-plate Bräuner-Osborne, H., Ebert, B., Brann, M.R., Falch, E., Krogsgaard-Larsen, P., 1996. cryo-EM structure of a biased agonist-bound human GLP-1 receptor–Gs complex. Functional partial agonism at cloned human muscarinic acetylcholine receptors. Eur. Nature 555, 121–125. J. Pharmacol. 313, 145–150. Lutfy, K., Eitan, S., Bryant, C.D., Yang, Y.C., Saliminejad, N., Walwyn, W., Kieffer, B.L., Chakrabarti, S., Prather, P.L., Yu, L., Law, P.Y., Loh, H.H., 1995. Expression of the mu- Takeshima, H., Carroll, F.I., Maidment, N.T., Evans, C.J., 2003. Buprenorphine-in- opioid receptor in CHO cells: ability of mu-opioid ligands to promote alpha-azidoa- duced antinociception is mediated by mu-opioid receptors and compromised by nilido[32P]GTP labeling of multiple G protein alpha subunits. J. Neurochem. 64, concomitant activation of opioid receptor-like receptors. J. Neurosci. 23, 2534–2543. 10331–10337. Chen, X.-T., Pitis, P., Liu, G., Yuan, C., Gotchev, D., Cowan, C.L., Rominger, D.H., Koblish, Manglik, A., Lin, H., Aryal, D.K., McCorvy, J.D., Dengler, D., Corder, G., Levit, A., Kling, M., Dewire, S.M., Crombie, A.L., Violin, J.D., Yamashita, D.S., 2013. Structure-ac- R.C., Bernat, V., Hübner, H., Huang, X.-P., Sassano, M.F., Giguère, P.M., Löber, S., tivity relationships and discovery of a G protein biased μ opioid receptor ligand, [(3- Duan, Da, Scherrer, G., Kobilka, B.K., Gmeiner, P., Roth, B.L., Shoichet, B.K., 2016. methoxythiophen-2-yl)methyl]({2-[(9R)-9-(pyridin-2-yl)-6-oxaspiro-[4.5]decan-9- Structure-based discovery of opioid with reduced side effects. Nature 537, yl]ethyl})amine (TRV130), for the treatment of acute severe pain. J. Med. Chem. 56, 185–190. 8019–8031. Maurel, D., Comps-Agrar, L., Brock, C., Rives, M.-L., Bourrier, E., Ayoub, M.A., Bazin, H., Cheng, Y., Prusoff, W.H., 1973. Relationship between the inhibition constant (K1) andthe Tinel, N., Durroux, T., Prézeau, L., Trinquet, E., Pin, J.-P., 2008. Cell-surface protein- concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic protein interaction analysis with time-resolved FRET and snap-tag technologies: ap- reaction. Biochem. Pharmacol. 22, 3099–3108. plication to GPCR oligomerization. Nat. Methods 5, 561–567. Copeland, R.A., 2016. The drug-target residence time model: a 10-year retrospective. Nat. McPherson, J., Rivero, G., Baptist, M., Llorente, J., Al-Sabah, S., Krasel, C., Dewey, W.L., Rev. Drug Discov. 15, 87–95. Bailey, C.P., Rosethorne, E.M., Charlton, S.J., Henderson, G., Kelly, E., 2010. μ-opioid Costa-Neto, C.M., Parreiras-e-Silva, L.T., Bouvier, M., 2016. A pluridimensional view of receptors: correlation of agonist efficacy for signalling with ability to activate inter- biased agonism. Mol. Pharmacol. 90, 587–595. nalization. Mol. Pharmacol. 78, 756–766. Daaka, Y., Luttrell, L.M., Lefkowitz, R.J., 1997. Switching of the coupling of the beta2- Megarbane, B., Marie, N., Pirnay, S., Borron, S., Gueye, P., Risede, P., Monier, C., Noble, adrenergic receptor to different G proteins by protein kinase A. Nature 390, 88–91. F., Baud, F., 2006. Buprenorphine is protective against the depressive effects of Dahan, A., Yassen, A., Bijl, H., Romberg, R., Sarton, E., Teppema, L., Olofsen, E., Danhof, on ventilation. Toxicol. Appl. Pharmacol. 212, 256–267. M., 2005. Comparison of the respiratory effects of intravenous buprenorphine and Molinari, P., Vezzi, V., Sbraccia, M., Grò, C., Riitano, D., Ambrosio, C., Casella, I., Costa, fentanyl in humans and rats. Br. J. Anaesth. 94, 825–834. T., 2010. Morphine-like selectively antagonize receptor-arrestin interactions. Dahan, A., Yassen, A., Romberg, R., Sarton, E., Teppema, L., Olofsen, E., Danhof, M., J. Biol. Chem. 285, 12522–12535. 2006. Buprenorphine induces ceiling in respiratory depression but not in analgesia. Motulsky, H.J., Christopoulos, A., 2003. Analyzing kinetic binding data. In: Motulsky, Br. J. Anaesth. 96, 627–632. H.J., Christopoulos, A. (Eds.), Fitting Models to Biological Data Using Linear and DeWire, S.M., Yamashita, D.S., Rominger, D.H., Liu, G., Cowan, C.L., Graczyk, T.M., Nonlinear Regression. Oxford University Press, pp. 245–251. Chen, X.-T., Pitis, P.M., Gotchev, D., Yuan, C., Koblish, M., Lark, M.W., Violin, J.D., Motulsky, H.J., Mahan, L.C., 1983. The kinetics of competitive radioligand binding pre- 2013. A G protein-biased ligand at the μ-opioid receptor is potently analgesic with dicted by the law of mass action. Mol. Pharmacol. 25, 1–9. reduced gastrointestinal and respiratory dysfunction compared with morphine. J. Namkung, Y., Le Gouill, C., Lukashova, V., Kobayashi, H., Hogue, M., Khoury, E., Song, Pharmacol. Exp. Ther. 344, 708–717. M., Bouvier, M., Laporte, S.A., 2016. Monitoring G protein-coupled receptor and β- Doll, C., Pöll, F., Peuker, K., Loktev, A., Glück, L., Schulz, S., 2012. Deciphering μ-opioid arrestin trafficking in live cells using enhanced bystander BRET. Nat. Commun.7, receptor phosphorylation and dephosphorylation in HEK293 cells. Br. J. Pharmacol. 1–12.

12 M.F. Pedersen, et al. Neuropharmacology 166 (2020) 107718

Quoyer, J., Janz, J.M., Luo, J., Ren, Y., Armando, S., Lukashova, V., Benovic, J.L., Viscusi, E.R., 2019. APOLLO -2: a randomized, placebo and active-controlled phase Carlson, K.E., Hunt, S.W., Bouvier, M., 2013. Pepducin targeting the C-X-C chemo- III study investigating oliceridine (TRV 130), a G protein-biased ligand at the μ-opioid kine receptor type 4 acts as a biased agonist favoring activation of the inhibitory G receptor, for management of moderate to severe acute pain following abdomino- protein. Proc. Natl. Acad. Sci. 110, E5088–E5097. plasty. Pain Pract. papr 12801. Raehal, K.M., Schmid, C.L., Medvedev, I.O., Gainetdinov, R.R., Premont, R.T., Bohn, L.M., Soergel, D.G., Subach, R.A., Burnham, N., Lark, M.W., James, I.E., Sadler, B.M., 2009. Morphine-induced physiological and behavioral responses in mice lacking G Skobieranda, F., Violin, J.D., Webster, L.R., 2014. Biased agonism of the μ-opioid protein-coupled receptor kinase 6. Drug Depend. 104, 187–196. receptor by TRV130 increases analgesia and reduces on-target adverse effects versus Raehal, K.M., Walker, J.K.L., Bohn, L.M., 2005. Morphine side effects in beta-arrestin 2 morphine: a randomized, double-blind, placebo-controlled, crossover study in knockout mice. J. Pharmacol. Exp. Ther. 314, 1195–1201. healthy volunteers. Pain 155, 1829–1835. Rahmeh, R., Damian, M., Cottet, M., Orcel, H., Mendre, C., Durroux, T., Sharma, K.S., Thompson, G.L., Lane, J.R., Coudrat, T., Sexton, P.M., Christopoulos, A., Canals, M., Durand, G., Pucci, B., Trinquet, E., Zwier, J.M., Deupi, X., Bron, P., Banères, J.-L., 2016. Systematic analysis of factors influencing observations of biased agonism at the Mouillac, B., Granier, S., 2012. Structural insights into biased G protein-coupled re- mu-opioid receptor. Biochem. Pharmacol. 113, 70–87. ceptor signaling revealed by fluorescence spectroscopy. Proc. Natl. Acad. Sci. 109, Trevena Inc, 2018. Trevena Receives Complete Response Letter for Oliceridine from FDA. 6733–6738. [WWW Document]. Press Release. http://www.trevena.com/news-details.php?id= Regnard, C., Twycross, R., Mihalyo, M., Wilcock, A., 2011. Loperamide. J. Pain Symptom 214, Accessed date: 24 January 2019. Manage. 42, 319–323. Tzschentke, T., 2002. Behavioral pharmacology of buprenorphine, with a focus on pre- Remmers, A.E., Medzihradsky, F., 1991. Reconstitution of high-affinity opioid agonist clinical models of reward and addiction. Psychopharmacology 161, 1–16. binding in brain membranes. Proc. Natl. Acad. Sci. 88, 2171–2175. Violin, J.D., Crombie, A.L., Soergel, D.G., Lark, M.W., 2014. Biased ligands at G-protein- Roed, S.N., Wismann, P., Underwood, C.R., Kulahin, N., Iversen, H., Cappelen, K.A., coupled receptors: promise and progress. Trends Pharmacol. Sci. 35, 308–316. Schäffer, L., Lehtonen, J., Hecksher-Soerensen, J., Secher, A., Mathiesen, J.M., Violin, J.D., Ren, X.-R., Lefkowitz, R.J., 2006. G-protein-coupled receptor kinase speci- Bräuner-Osborne, H., Whistler, J.L., Knudsen, S.M., Waldhoer, M., 2014. Real-time ficity for β-arrestin recruitment to the beta2-adrenergic receptor revealed byfluor- trafficking and signaling of the glucagon-like peptide-1 receptor. Mol.Cell. escence resonance energy transfer. J. Biol. Chem. 281, 20577–20588. Endocrinol. 382, 938–949. Viscusi, E.R., Skobieranda, F., Soergel, D.G., Cook, E., Burt, D.A., Singla, N., 2019. Saidak, Z., Blake-Palmer, K., Hay, D.L., Northup, J.K., Glass, M., 2006. Differential acti- APOLLO-1: a randomized placebo and active-controlled phase III study investigating vation of G-proteins by μ-opioid receptor agonists. Br. J. Pharmacol. 147, 671–680. oliceridine (TRV130), a G protein-biased ligand at the μ-opioid receptor, for man- Schinkel, A.H., Wagenaar, E., Mol, C.A., van Deemter, L., 1996. P-glycoprotein in the agement of moderate-to-severe acute pain following bunionectomy. J. Pain Res. 12, blood-brain barrier of mice influences the brain penetration and pharmacological 927–943. activity of many drugs. J. Clin. Investig. 97, 2517–2524. Viscusi, E.R., Webster, L., Kuss, M., Daniels, S., Bolognese, J.A., Zuckerman, S., Soergel, Schönegge, A.-M., Gallion, J., Picard, L.-P., Wilkins, A.D., Le Gouill, C., Audet, M., D.G., Subach, R.A., Cook, E., Skobieranda, F., 2016. A randomized, phase 2 study Stallaert, W., Lohse, M.J., Kimmel, M., Lichtarge, O., Bouvier, M., 2017. Evolutionary investigating TRV130, a biased ligand of the μ-opioid receptor, for the intravenous action and structural basis of the allosteric switch controlling β2AR functional se- treatment of acute pain. Pain 157, 264–272. lectivity. Nat. Commun. 8, 2169. Winpenny, D., Clark, M., Cawkill, D., 2016. Biased ligand quantification in drug dis- Shukla, A.K., Violin, J.D., Whalen, E.J., Gesty-Palmer, D., Shenoy, S.K., Lefkowitz, R.J., covery: from theory to high throughput screening to identify new biased μ opioid 2008. Distinct conformational changes in β-arrestin report biased agonism at seven- receptor agonists. Br. J. Pharmacol. 173, 1393–1403. transmembrane receptors. Proc. Natl. Acad. Sci. 105, 9988–9993. Wootten, D., Christopoulos, A., Marti-Solano, M., Babu, M.M., Sexton, P.M., 2018. Singla, N., Minkowitz, H., Soergel, D., Burt, D., Subach, R.A., Salamea, M., Fossler, M., Mechanisms of signalling and biased agonism in G protein-coupled receptors. Nat. Skobieranda, F., 2017. A randomized, Phase IIb study investigating oliceridine Rev. Mol. Cell Biol. 19, 638–653. (TRV130), a novel μ-receptor G-protein pathway selective (μ-GPS) modulator, for the Zhang, H., Sturchler, E., Zhu, J., Nieto, A., Cistrone, P.A., Xie, J., He, L., Yea, K., Jones, T., management of moderate to severe acute pain following abdominoplasty. J. Pain Res. Turn, R., Di Stefano, P.S., Griffin, P.R., Dawson, P.E., McDonald, P.H., Lerner, R.A., 10, 2413–2424. 2015. Autocrine selection of a GLP-1R G-protein biased agonist with potent anti- Singla, N.K., Skobieranda, F., Soergel, D.G., Salamea, M., Burt, D.A., Demitrack, M.A., diabetic effects. Nat. Commun. 6, 8918.

13