Pharmacologic Evidence for a Putative Conserved Allosteric

1521-0111/93/2/157–167$25.00 https://doi.org/10.1124/mol.117.109561MOLECULAR PHARMACOLOGY Mol Pharmacol 93:157–167, February 2018Copyright ª 2018 by The American Society for Pharmacology and Experimental Therapeutics

Pharmacologic Evidence for a Putative Conserved Allosteric Siteon Opioid Receptors

Kathryn E. Livingston,1 M. Alexander Stanczyk, Neil T. Burford, Andrew Alt,2

Meritxell Canals, and John R. TraynorDepartment of Pharmacology and Edward F. Domino Research Center, University of Michigan, Ann Arbor, Michigan (K.E.L.,M.A.S., J.R.T.); Research and Development/Discovery, Bristol-Myers Squibb Company, Wallingford, Connecticut (N.T.B., A.A.);and Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, Victoria, Australia (M.C.)

Received June 2, 2017; accepted November 27, 2017

ABSTRACTAllosteric modulators of G protein–coupled receptors, includingopioid receptors, have been proposed as possible therapeuticagents with enhanced selectivity. BMS-986122 is a positiveallosteric modulator (PAM) of the m-opioid receptor (m-OR).BMS-986187 is a structurally distinct PAM for the d-opioidreceptor (d-OR) that has been reported to exhibit 100-foldselectivity in promoting d-OR over m-OR agonism. We usedligand binding and second-messenger assays to show thatBMS-986187 is an effective PAM at them-OR and at the k-opioidreceptor (k-OR), but it is ineffective at the nociceptin receptor.The affinity of BMS-986187 for d-ORs and k-ORs is

approximately 20- to 30-fold higher than for m-ORs, determinedusing an allosteric ternary complex model. Moreover, we provideevidence, using a silent allosteric modulator as an allostericantagonist, that BMS-986187 and BMS-986122 bind to a similarregion on all three traditional opioid receptor types (m-OR, d-OR,and k-OR). In contrast to the dogma surrounding allostericmodulators, the results indicate a possible conserved allostericbinding site across the opioid receptor family that can accom-modate structurally diverse molecules. These findings haveimplications for the development of selective allostericmodulators.

IntroductionAllosteric modulation of G protein–coupled receptors

(GPCRs) is a promising new avenue to develop safer drugs.With this aim in mind, we have discovered positive allostericmodulators (PAMs) of both m-opioid (m-OR) and d-opioid(d-OR) receptors. These allosteric modulators (Burford et al.,2013, 2015; Livingston and Traynor, 2014; Bisignano et al.,2015) bind to a site on the receptor distinct from theorthosteric site for endogenous opioid peptides and traditionalopiate analgesics drugs to enhance the affinity and/or efficacyof various orthosteric ligands in an agonist-dependent man-ner, described as probe dependence. BMS-986122 (Fig. 1) isselective for m-OR and has no detectable activity at the closelyrelated d-OR (Burford et al., 2013), whereas the d-OR PAM,BMS-986187 (Fig. 1), possesses a 100-higher potency at d-ORover m-OR (Burford et al., 2015).

The selectivity of the PAMs is not surprising as enhancedselectivity is one of the touted benefits of allosteric modulationof GPCRs over traditional orthosteric activation. Ligand-binding domains were evolved and maintained to bindendogenous ligands, whereas sites for small allosteric ligandsface less evolutionary pressure to be conserved. Therefore,even closely related receptors of the same family that havevery similar orthosteric binding pockets can have differentallosteric binding sites (for review, see Conn et al., 2009). Forexample, the m- and d-ORs share more than 64% sequenceidentity and are nearly identical in the transmembrane (TM)domains (Stevens, 2009), with highly conserved orthostericbinding pockets as determined from multiple inactive-statecrystal structures (Lin et al., 2009; Manglik et al., 2012;Fenalti et al., 2014), as well as an active-state structure ofm-OR (Huang et al., 2015). In addition, the opioid peptidesLeu- and Met-enkephalin are endogenous ligands for bothreceptors, and the signaling pathways downstream of thereceptors are quite similar (Chen et al., 1993). On the otherhand, the allosteric ligands discovered to date for the tworeceptors are highly structurally distinct (Fig. 1).In previous work, we have demonstrated that the action of

BMS-986122 atm-OR involves a disruption with the binding ofNa1 ions (Livingston and Traynor, 2014) that modulate theactivity of many class A GPCRs, including the opioid receptor

This work was supported by National Institutes of Health, NationalInstitute on Drug Abuse Institute [Grants R01 DA39997 and T32 DA007267]and an EDGE Fellowship from the Endowment for the Basic Sciences,University of Michigan.

1Current affiliation: Department of Psychiatry, Genentech Hall, SanFrancisco, California.

2Current affiliation: Arvinas Inc, New Haven, Connecticut.https://doi.org/10.1124/mol.117.109561.

ABBREVIATIONS: AC, adenylate cyclase; CHO, Chinese hamster ovary (cells); DPN, diprenorphine; FBS, fetal bovine serum; GPCR, G proteincoupled receptor; GTPgS, guanosine-59-O-(3-thio)triphosphate; HEK, human embryonic kidney (cells); mAchR, muscarinic acetylcholine receptor;NOPR, nociceptin receptor; d-OR, delta opioid receptor; k-OR, kappa opioid receptor; m-OR, mu opioid receptor; PK, Pro-Link; PAM, positiveallosteric modulator; SAM, silent allosteric modulator; TM, transmembrane.

157

at ASPE

T Journals on M

ay 3, 2022m

olpharm.aspetjournals.org

Dow

nloaded from

https://doi.org/10.1124/mol.117.109561

https://doi.org/10.1124/mol.117.109561

http://molpharm.aspetjournals.org/

family (for review, see Katritch et al., 2014). Na1 binds at awell described site within the 7-TM bundle and contributes tostabilization of the receptor in an inactive state with reducedaffinity for agonists (Pert et al., 1973; Pert and Snyder, 1974;Liu et al., 2012). We previously hypothesized (Livingston andTraynor, 2014) that allosteric disruption of Na1 binding maybe a common mechanism for PAMs of class A GPCRs.Although BMS-986187 possesses selectivity for d-OR over

m-OR, it does appear to have some PAMactivity at m-ORwhenendomorphin 1 is used as the orthosteric ligand (Burford et al.,2015). This and the structural and functional similaritiesbetween the m-OR and d-OR led us to examine the action ofBMS-986187 at m-OR and to test the hypothesis of a generalmechanism of action of allosteric modulation of GPCRs bystudying whether the binding of the d-OR-PAM BMS-986187and Na1 ions are mutually incompatible at both d-OR andm-OR. Finally, we askedwhether the allosteric sites on the tworeceptors are similar by using the silent allosteric modulatorBMS-986124, which acts as an antagonist at the allosteric sitein m-OR (Burford et al., 2013). We also investigated whetherthese modulators have activity at the closely related k-opioidreceptor (k-OR) and nociception receptor (NOPR).Overall, our findings confirm that the selective d-PAMBMS-

986187 has PAM activity at m-OR, and we further show thatthis compound is also a PAM at k-OR but not at NOPR. Inaddition, we have found that the m-PAM BMS-986122 is asilent allosteric modulator at d-OR and k-OR. Together, thesedata suggest that the allosteric binding sites on m-OR, d-OR,and k-ORmay be conserved. The results have ramifications forthe identification of the allosteric site(s) and for drug devel-opment of allosteric modulators for the opioid receptors.

Materials and MethodsMaterials. [3H]-Diprenorphine (DPN) and GTPg35S [(guanosine-

59-O(3-thio)triphosphate] were from PerkinElmer Life Sciences (Cam-bridge, MA). All tissue culture medium, penicillin-streptomycin,geneticin (G148), trypsin, and fetal bovine serum (FBS) were fromInvitrogen (Carlsbad, CA). DAMGO, naloxone, and morphine sulfatewere from Sigma-Aldrich (St. Louis, MO). PathHunter detectionreagents were from DiscoveRx (Freemont, CA). Lance-Ultra cAMPdetection reagents were from PerkinElmer Life Sciences. BMS-986122, BMS-986124, and BMS-986187 (structures in Fig. 1) weresynthesized or obtained as previously described (Burford et al., 2013,2015). Other drugs were from the Opioid Basic Research Center at theUniversity of Michigan. All other chemicals, unless otherwise speci-fied, were purchased from Sigma.

Cell Lines and Membrane Preparation. The generation andmaintenance of C6 rat glioma cells stably transfectedwith ratm-opioidreceptor (m-OR) or rat d-opioid receptor (d-OR) were performed asdescribed (Clark et al., 2008). Chinese hamster ovary (CHO) cells

expressing human k-OR, human embryonic kidney (HEK) 293T cellsexpressing human ORL1, and Flp-In CHO expressing human d-OR(CHO-d) were generated and maintained Dulbecco’s modified Eagle’smedium with 10% FBS in the presence of 0.4 mg/ml of Geneticin and0.1% penicillin/streptomycin. Cell membranes were prepared forbinding assays as described (Livingston and Traynor, 2014). Briefly,cells were grown to confluence andwashed twicewith 37°C phosphate-buffered saline (pH 7.4). Cells were detached in harvesting buffer(20 mMHEPES, 150 mMNaCl, 0.68 mM EDTA, pH 7.4) and pelletedby centrifugation at 200g for 3 minutes at room temperature. Thepellet was resuspended in ice-cold 50 mM Tris (pH 7.4) and homog-enized using a Tissue Tearor (Dremel, Mount Prospect, IL). Thishomogenate was centrifuged at 20,000g at 4°C for 20 minutes. Thepellet was then resuspended, homogenized, and centrifuged oncemore. The final pellet was resuspended in 50mMTris (pH 7.4) using aglass Dounce homogenizer, and aliquots were flash-frozen in liquidnitrogen. Aliquots were stored at 280°C until use. Protein concentra-tion was determined using the bicinchoninic acid quantificationmethod with bovine serum albumin as the standard.

CHO PathHunter cells expressing enzyme acceptor (EA)-taggedb-arrestin-2 and ProLink (PK)-tagged m-OR receptor (CHO-m) werefrom DiscoveRx. Cells were grown in F-12 media (11765; Invitrogen)containing Hyclone FBS 10%, Hygromycin 300 mg/ml, Geneticin(G418) 800 mg/ml and maintained at 37°C in a humidified incubatorcontaining 5%CO2. These cells were used for b-arrestin-2 recruitmentassays and inhibition of forskolin-stimulated cAMP accumulationassays described as follows.

Radioligand Binding Assays. Ligand binding assays were per-formed using the cell membrane homogenates described in thepreceding section. Competition binding assays were performed aspreviously described (Clark et al., 2003). Briefly, 3H-DPN (0.2–0.3 nM)was incubated in assay buffer (50 mM Tris, pH 7.4, 100 mM NaCl,5 mM MgCl2, 1 mM EDTA, 10 mM GTPgS) with 10 mg of membraneprotein, orthosteric ligand, and allosteric ligand (or vehicle) to pro-mote a low-affinity state of the receptor. In some experiments, thek-OR ligand 3H-U69,593 was examined in Tris-HCl buffer (pH 7.4)only. Nonspecific binding was determined in the presence of 10 mMnaloxone. Assays were incubated at room temperature for 75–90 minutes to reach equilibrium and then terminated and countedas described (Livingston and Traynor, 2014). Three independentexperiments, each done in duplicate, were performed, and the valueswere pooled to generate the mean curves as displayed in the figures.

GTPg35S Assays. GTPg35S binding experiments were performedas described (Traynor and Nahorski, 1995) using cell-membranehomogenates prepared as described. Briefly, 10-mg aliquots of mem-brane protein were incubated for 1 hour at 25°C in buffer (50mMTris,pH 7.4, 100 mM NaCl, 5 mM MgCl2, 1 mM EDTA) with 0.1 nMGTPg35S 30 mM GDP, orthosteric ligand, and allosteric ligand (orvehicle). An internal standard at 10 mM (DAMGO [([D-Ala2,N-MePhe4, Gly-ol]-enkephalin)] for m-OR, SNC80 for d-OR, U69,593for k-OR, and nociceptin for NOPR) was used to define maximalactivation, water, or vehicle-defined basal binding. The assays wereterminated and counted as described (Livingston and Traynor, 2014).Three independent experiments, each done in duplicate, were per-formed, and the values were pooled to generate mean curves asdisplayed in the figures.

PathHunter b-Arrestin-2 Assay. Confluent flasks of CHO-mcells were harvested with TrypLE Express and resuspended in F-12media supplemented with 10%FBS and 25mMHEPES at a density of6.67 � 105 cells/ml and plated (3 ml/well) into white solid TC-treated1536-well plates (Corning Inc., Corning, NY). Plates were incubatedovernight at 37°C in a 5% CO2 humidified incubator. The next day,increasing concentrations of BMS-986187 (40 nl of 100 � finalconcentration in 100% dimethylsulfoxide) were added to separaterows of the assay plates by acoustic dispense using an Echo-550(Labcyte, Sunnyvale, CA) from Echo-qualified 1536-well source plates(Labcyte). Next, 1 ml of increasing concentrations of DAMGO (4� finalconcentration in assay buffer) were added to separate columns of the

Fig. 1. Structure of (A) BMS-986187, (B) BMS-986122, and (C) BMS-986124.

158 Livingston et al.

at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


assay plates containing cells. Plates were covered with a lid andincubated at room temperature for 90 minutes. Incubations wereterminated by the addition of 2 ml of PathHunter Reagent (DiscoveRx,Fremont, CA). Luminescence was measured 1 hour later using aViewlux imaging plate reader (PerkinElmer).

Inhibition of Forskolin-Stimulated cAMP AccumulationAssays. CHO-m cells were grown to confluence, harvested, andresuspended at 106 cells/ml in assay buffer (Hanks’ balanced saltsolution 1 25 mM HEPES, 1 0.05% bovine serum albumin). In-creasing concentrations of BMS-986187 (30 nl of 100 � final concen-tration in 100% dimethylsulfoxide) were added to separate rows of1536-well white solid NT plates by acoustic dispense using an Echo-550 (Labcyte). Next, 1 ml of increasing concentrations of DAMGO (at3� final concentration in assay buffer) was added to separate columnsof the plates. Then, 1 ml of cells (1000 cells/well) was added to all wells,followed by 1 ml of forskolin (3� final concentration in assay buffer).Plates were lidded and incubated for 45minutes at room temperature.Incubations were terminated by the addition of Lance-Ultra cAMPdetection reagent (PerkinElmer) (1.5 ml of Eu-cryptate-labeled cAMPtracer in lysis buffer, followed by 1.5 ml of U-light conjugated anti-cAMP antibody in lysis buffer). After a 1-hour incubation at roomtemperature, time-resolved fluorescence was detected on a Viewlux orEnvision plate reader (PerkinElmer) with excitation at 337 nm andemission reads at 615 and 665 nm. The ratiometric data (665 nmread/615 nm read)� 10,000were then converted to cAMP (nanomolar)based on a standard curve for cAMP run at the same time underidentical conditions.

Data Analysis and Reporting. Data were analyzed usingGraphPad Prism version 6.01 (GraphPad, SanDiego, CA). Orthostericligand affinity (Ki) values and potency (EC50) values were determinedusing nonlinear regression, with basal and maximal values con-strained and are presented as mean6 S.E.M. or with 95% confidencelimits. Functional cooperativity values (ab) and the affinities ofallosteric modulators for the orthosteric agonist-unoccupied receptor(KB) were obtained using the allosteric ternary complex model(Christopoulos and Kenakin, 2002; Leach et al., 2007) or a derivationof the allosteric ternary complex model (Leach et al., 2010). Samplesize was predetermined to be a minimum of three independentexperiments. None of the functional assays was blinded to investiga-tors, and no data were excluded.

ResultsActivity of the d-PAM, BMS-986187, at m-OR. BMS-

986187 is reported to have a 100-fold selectivity for d-ORcompared with m-OR, although initial data on BMS-986187suggested that the compound is an efficacious, yet low, affinityPAM at m-OR (Burford et al., 2015). To confirm the PAM activityof BMS-986187 at m-OR, we studied the effects of BMS-986187using cell-membrane preparations fromC6 rat glioma cells stablyexpressing rat mOR (C6m, Lee et al., 1999). Since we havepreviously shown that modulators act by allosterically displacingNa1 ions from inactive m-OR, binding was measured by compe-tition assay using the orthosteric antagonist [3H]-diprenorphine([3H]-DPN) in the presence of Na1 ions and GTPgS to uncoupleheterotrimeric G protein and so stabilize inactive (R) conforma-tions of the receptor (Livingston and Traynor, 2014). We verifiedthat BMS-986187 does not compete with [3H]-DPN, (Fig. 2A) forthe orthosteric site on m-OR or stimulate GTPg35S binding (Fig.2B). Conversely, BMS-986187 significantly increased the bindingaffinity (Ki) of agonists for m-OR, consistent with a positiveallosteric ligand. As such, in the presence of 10 mM BMS-986187, the affinity of the m-OR agonist DAMGO was enhanced11-fold from 724 to 63 nM, whereas the affinity of methadonewasincreased 24-fold, from 603 to 25 nM. In contrast, the affinity of

morphine, a partial agonist, was enhanced by a nonsignificant3-fold, from 229 to 71 nM (Fig. 2; Table 1).The allosteric modulatory actions of BMS-986187 at m-OR

were then investigated in three signaling assays. BMS-986187(1 mM) significantly enhanced the potency of DAMGO by 6-fold,from 91 to 16 nM;methadone by 20-fold, from 203 to 10 nM, butit shifted morphine only a nonsignificant 3-fold, from 120 to38 nM in the GTPg35S binding assay in membranes from C6mcells. Therewas also a significant increase in themaximal effectof the partial agonist morphine, from 70% to 90% of theDAMGO response (Fig. 2; Table 1). The larger shift seen withmethadone allowed us to repeat the assay in the presence ofincreasing concentrations of BMS-986187 to generate a limitedseries of concentration-response curves (Fig. 3) that, whenanalyzed using a derivation of the allosteric ternary complexmodel (Leach et al., 2010; seeMaterials andMethods), resultedin a pKB value, representing the affinity of BMS-986187 for theunoccupiedm-ORof 5.2360.19 and a logab value of 1.1660.13for the cooperativity between methadone and BMS-986187.We then used high throughput methods to generate a more

comprehensive series of concentration-effect curves to studythe action of the modulator on adenylate cyclase (AC) in-hibition and b-arrestin-2 recruitment, which enabled us todetermine binding affinity and allosteric cooperativity usingthe full operational model of allostery (Leach et al., 2007; seeMaterials and Methods). Using the inhibition of AC in CHO-mcells as a downstream measure of Gi/o signaling, the BMS-986187 concentration dependently enhanced the potency ofDAMGO by 10-fold (EC50 from 66 to 7 pM) (Fig. 4A). Thesedata afforded a pKB value for BMS-986187 of 4.786 0.15 and alogab value for cooperativity of 1.576 0.17. Analysis using thesimplified operational model (Leach et al., 2010) gave thesame result (pKB 5 4.83 6 0.15; logab 5 1.52 6 0.17). Whenrecruitment of b-arrestin-2 was measured as a signalingoutput in the same cells, BMS-986187 was also observed toconcentration dependently enhance the potency of DAMGO(EC50 from 170 to 3 nM), representing a 58-fold shift inDAMGO potency (Fig. 4B). Analyses of these data afforded apKB value for BMS-986187 of 4.616 0.06 (Fig. 4) and a log abvalue of cooperativity of 2.1 6 0.05.The coupling efficiency of the AC assay is much higher than

in the b-arrestin-2 or GTPg35S assays as evidenced by the2500-fold higher potency of DAMGO in the former assay.Because of this high coupling efficiency, BMS-986187 aloneinhibited adenylate cyclase (Fig. 4C). In comparison, noagonist activity was observed in the GTPg35S assay (Fig. 2B)or the b-arrestin-2 assay (Fig. 4D). This direct agonist activityin the AC assay at m-OR is not due to an action of BMS-986187at the orthosteric site on m-OR, as shown by the 3H-DPNdisplacement assays (Fig. 2A), and so the compound should bedesignated an “ago-PAM” (Christopoulos et al., 2014), remi-niscent of its activity at the d-OR, where it is capable ofG-protein activation, AC inhibition, and mitogen-activatedprotein kinase activation in the absence of orthosteric agonist,in addition to allosterically enhancing orthosteric agonistaffinity and activity (Burford et al., 2015). The ago-PAMefficacy of BMS-986187 at m-OR is low, such that this canonly be observed whenmeasured using themore amplified ACsignaling output in cells overexpressing m-OR receptors.BMS-986187 Acts to Disrupt Na1 Binding at Both

m-OR and d-OR. Na1 ion binding is highly coordinated andcontributes to the stabilization of inactive receptor (R; Fenalti

Evidence for a Conserved Opioid Receptor Allosteric Site 159

at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


Fig. 2. BMS-986187 enhances the affinity, potency, and/or maximal stimulation of several opioid ligands at m-OR expressed in C6 cell membranes. (A)BMS-986187 alone did not displace 3H-DPN from the orthosteric site or (B) stimulate GTPg35S binding. The presence of 10 mM BMS-986187 (filledsymbols) increased the binding affinity and potency or maximal effect to stimulate GTPg35S for DAMGO (C and D), methadone (E and F), and morphine(G and H) compared with control conditions (open symbols). Data are presented as percent 3H-DPN bound for the binding data or percentage ofstimulation of a maximal concentration (10 mM) of the full-agonist DAMGO for the GTPg35S data. Nonlinear regression analysis using GraphPad Prism6.01 fit all curves to one site. Data shown aremeans6 S.E.M. of three (3H-DPN) or four (GTPg35S) independent experiments each performed in duplicate.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


et al., 2014; Katritch et al., 2014). Disruption of Na1 co-ordination leads to an increased level of active-state m-OR (R*)and therefore an increase in the binding affinity of orthostericagonists (Pert et al., 1973; Pert and Snyder, 1974). We havepreviously demonstrated that the binding of them-PAM,BMS-986122, and Na1 ions are incompatible leading BMS-986122to promote R* (Livingston and Traynor, 2014). Here we showthat the potency (EC50) of Na1 ions to inhibit the binding of them-OR agonist DAMGO to membranes from C6m cells isreduced by 5-fold in the presence of BMS-986187 [from 3.8(0.6–20) to 20.2 (12–34) mM] (Fig. 5A), although because of asmall window in the control group, the effect did not reachsignificance. Likewise, at the d-OR expressed in CHO cells,there was a negative relationship between Na1 ions andBMS-986187 such that BMS-986187 caused a concentration-dependent 4-fold rightward shift in the NaCl concentration-response curve [from 45 (35–59) to 172 (100–300) mM] toinhibit the constitutive receptor-mediated GTPg35S bindingthat occurs in Na1-free buffer (Szekeres and Traynor, 1997)(Fig. 5B).BMS-986122 and BMS-986187 are Competitive at

m-OR and d-OR. Although our initial hypothesis predictedthat BMS-986122 andBMS-986187 have distinct binding siteson m-OR based on their diverse structural features (Fig. 1), the

fact that they show a similar pattern of probe dependence intheir interaction with full versus partial agonists versusantagonists and a similar interaction with Na1 ions couldindicate that they engage mOR at the same allosteric site(s). To

TABLE 1Effects of BMS-986187 on ligand affinity and [35S]GTPgS binding at m, d, k, and NOP receptors

Ki (nM) Potency (GTPg35S) (nM) Emax (GTPg35S) (% Standard)

Control BMS-986187 Control BMS-986187 Control BMS-986187

m-ORaDAMGO 724

(562–912)63

(13–316) f91

(55–154)16

(6–38) f111

(103–119)110

(105–115)Methadone 603

(288–1259)25

(19–32) f203

(121–284)10

(3–24) f105

(98–111)107

(102–111)Morphine 229

(178–302)71

(43–120) f120

(85–166)38

(10–151)70

(67–74)90

(83–97) f

d-ORb

Leu-enkephalin

221(119–324)

7(3–12) f

28(18–42)

3.6(1.6–8.2) f

87(79–95)

94(89–100)

SNC80 71(20–122)

5(3–7) f

NT NT NT NT

TAN67 10(7–14)

3(0.2–5.8) f

NT NT NT NT

k-ORc

U69,593d 0.9(0.3–1.6)

0.3(0.2–0.5) g

881(480–1601)

64(35–121) f

105(82–128)

104(89–118)

Dynorphin A NT NT 15(11–19)

1.0(0.5–2) f

100(95–105)

95(86–104)

Salvinorin A NT NT 322(221–471)

25.9(11–60) f

110(97–122)

116(101–131)

Diprenorphine NT NT 46(25–37)

8.9(1.9–26)

31(25–37)

56(47–67) f

NOPRe

Nociceptin NT NT 2.6(2.0–4)

2.1(1.5–3)

102(94–110)

105(100–111)

Ro-64-6198 NT NT 13(8.0–19)

18(7.8–45)

109(102–115)

108(93–123)

NT, not tested. Values in parentheses represent 95% confidence intervals.aIn C6 rat glioma cells expressing rat m-OR.bData taken from Burford et al. (2015).cIn CHO cells expressing human k-OR.dKi values determined by saturation binding in Tris-buffer.eHEK293 cells expressing human NOPR.fSignificantly different from control conditions (nonoverlapping confidence intervals).gP = 0.03 compared with control (paired t test).

Fig. 3. BMS-986187 has a concentration-dependent effect on the potencyof methadone to activate G protein. Stimulation of GTPg35S binding inC6m cell membranes by methadone was performed in the presence ofincreasing concentrations (0.3–10 mM) of BMS-986187. Data were ana-lyzed using a modified allosteric ternary complex model as described inMaterials and Methods. Points shown are means 6 S.E.M. of threeindependent experiments, each in duplicate.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


evaluate this possibility, we used the m-silent allosteric modu-lator (m-SAM) BMS-986124, a close analog of BMS-986122 (Fig.1), which has been previously reported to block BMS-986122action atm-OR but having no effects alone (Burford et al., 2013).We asked whether BMS-986124 could inhibit the action ofBMS-986187 at m-OR and even at d-OR.BMS-986187, at 10 mM, produced a 6-fold increase in the

potency of DAMGO to stimulate GTPg35S binding in C6m cellmembranes (from 91 to 16 nM) (Table 1). The addition of them-SAM increased the EC50 of DAMGO [to 59.4 (47.5–74.2) nM]such that the PAM effects of BMS-986187 were no longersignificant (Fig. 6A). Using the same assay in membranesfrom C6d cells BMS-986187 (300 nM) displayed no agonismalone, but it enhanced the potency of the orthosteric agonistleu-enkephalin to stimulate GTPg35S by 6-fold [from378 (287–498) to 65.4 (44.9–95.4) nM]. This effect was re-versed in the presence of BMS-986124 [217 (154–295) nM](Fig. 6B). Furthermore, since BMS-986124 blocks the action ofBMS-986122 at m-OR and BMS-986187 at d-OR, we predictedBMS-986122, because of its structural similarity to BMS-986124 (Fig. 1), might bind to d-OR but lacks allostericefficacy. In a separate set of experiments in C6d cell mem-branes, the presence of BMS-986187 (300 nM) increased thepotency of Leu-enkephalin [from 715 (548–933) to 57.7(42.1–79.0) nM], but this increase was decreased by theaddition of 30 mM BMS-986122 [to 176 (136–228) nM](Fig. 6C). The fact that we observed only partial reversalmay be explained by the finding that BMS-986122 might notbe entirely silent at d-OR. For example, in a separateexperiment, the compound slightly enhanced the potency ofLeu-enkephalin [from 667 (525–848) to 293 (234–374) nM]

(Fig. 6D). BMS-986187 is an ago-PAM at d-OR and alone willstimulate GTPg35S binding via the allosteric site. Both them-SAM (BMS-986124) and them-PAM (BMS986122) were ableto inhibit the ago-PAM activity of BMS-986187 (Fig. 6E).BMS-986187 Is a Positive Allosteric Modulator of

k-OR. Since BMS-986187 has activity at both m-OR andd-OR, we sought to determine whether it also has activity atthe k-opioid receptor (k-OR) since these three receptors sharea high degree of homology as well as endogenous ligands. InCHO cells expressing the human k-OR, GTPg35S bindingstimulated by the k-OR-preferring peptide dynorphin A(1–17)was enhanced 15-fold (from 15 to 1.0 nM) in the presence of10 mM BMS-986187 with no change in the maximal response(Fig. 7; Table 1). Similarly, the activities of the k-OR agonistU69,593, the nonbasic nitrogen-containing agonist salvinorinA, and the partial agonist DPN (Traynor et al., 1987) wereenhanced in the presence of 10 mMBMS-986187. The potencyof U69,593 was shifted 14-fold, from 881 to 64 nM, and thepotency of salvinorin A was shifted by 12-fold, from 322 to26 nM, whereas the maximal responses to both compoundswas unaltered (Fig. 7; Table 1). In contrast, the potency of thepartial agonist DPNwas not significantly shifted (46–8.9 nM),but the maximal stimulation was enhanced from 31% to 56%of that seen with U69,593 (Fig. 7; Table 1). Analysis of theconcentration-response curves for the stimulation of GTPg35Sbinding by U69,593, in the presence of BMS-986187 (Fig. 7B)using the simplified operational model (Leach et al., 2010),afforded a pKB value for themodulator of 6.226 0.19 and a logab value of 1.11 6 0.13.We could not investigate the effect of BMS-986187 using

binding assays with 3H-DPN as the tracer ligand at k-OR

Fig. 4. Effects of BMS-986187 on opioid-mediated adenylate cyclase (AC) inhibition and b-arrestin-2 recruitment in CHO-m cells. (A) Concentration-response curves for the full m-OR agonist DAMGO to inhibit AC in the absence or presence of increasing concentrations of BMS-986187. (B)Concentration-response curves of DAMGO to recruit b-arrestin-2 were performed in the absence or presence of increasing concentrations of BMS-986187as listed in (A). (C andD) Effect of BMS-986187 on inhibition of AC or recruitment of b-arrestin-2 in the absence of orthosteric agonist. Datawere analyzedusingGraphPad Prism 6.01, and points shown aremeans6 S.E.M. of three independent experiments, each done in quadruplicate. Data in (A andB) wereanalyzed using an allosteric ternary complex model as described in Materials and Methods.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


because DPN has agonist activity at this receptor and thisaction is sensitive to BMS-986187 (Fig. 7D). Consequently, westudied the effects of the modulators on the binding of3H-U69,593. This was examined in Tris-buffers in the absenceof added Na1 ions and GTP. Even so, BMS-986187 enhancedthe affinity of 3H-U69,593 to k-OR fromKD5 0.96 0.2 to 0.360.1 nM (P 5 0.03; Fig. 7E; Table 1) without a change in theBmax. The effect is small because receptors will exist mostly inthe R* state under these conditions. Neither BMS-986122 norBMS-986124 affected 3H-U69,593 binding alone, but BMS-986124 prevented the BMS-986187–mediated enhancementof binding (Fig. 7F), suggesting that it is also a k-SAM.BMS-986122 and BMS-986187 Have no Detectable

PAM Activity at NOPR. The nociceptin receptor (NOPR) isnow recognized as a fourth member of the opioid receptorfamily (Cox et al., 2015) and, like the other family members, iscoupled to Gai/o proteins. Using HEK293T cells expressingNOPR, concentration-response curves of the abilities of theendogenous ligand nociceptin and the small-molecule syn-thetic agonist R064-6198 to stimulate GTPg35S binding weredetermined in the presence or absence of BMS-986187 (Fig. 8;

Table 1). Neither the potency nor maximal effect of eitherNOPR ligand was altered.

DiscussionOur results show that d-PAM BMS-986187 is also a m-PAM

and k-PAM, providing the first description of a k-PAM.Moreover, the m-PAM BMS-986122 is a SAM, or allostericantagonist, at d-OR, and the m-SAM BMS-986124 is anallosteric antagonist at m-OR, d-OR, and k-OR. In addition,the probe dependence of BMS-986187 at the m-OR and k-OR isthe same as seen with BMS-986122 at the m-OR (Livingstonand Traynor, 2014) and BMS-986187 at the d-OR (Burfordet al., 2015). In particular, an increase in ligand affinity andpotency was seen for full agonists but an increase occurred inthe maximal effect for the partial agonists, morphine at m-ORand diprenorphine at the k-OR, with little change in theiraffinity or potency. Finally, the mechanism of action ofthe PAMs at m-OR and d-OR involves destabilization of theinactive Na1-bound receptor. Thus, it appears that theallosteric ligand binding site on the m-, d-, and k-OR is similarenough to recognize the same ligands and interact with theorthosteric site but also sufficiently promiscuous to accommo-date different chemical scaffolds as exemplified by BMS-986122 and BMS-986187 (Fig. 1). The findings providepharmacologic evidence for a possible conserved allosteric siteacross the three naloxone-sensitive members of the opioidreceptor family.The average affinity of BMS-986187 for orthosteric-agonist–

free m-OR (pKB) determined from the GTPg35S, b-arrestin-2and AC assays is 4.756 0.07. In contrast, the pKB at the k-ORderived from GTPg35S data in 6.22, similar to the reportedaverage pKB value for BMS-986187 at d-OR (i.e., 6.02)(Burford et al., 2015). Since the pKB for BMS-986187 isindependent of the ligand occupying the orthosteric site, wecan compare these values to show that BMS-986187 isapproximately 20- to 30-fold selective for binding to d-ORand k-OR over m-OR. The cooperativity factor (log ab) betweenDAMGO and BMS-986187 at m-OR for the GTPg35S andadenylate cyclase assays is 1.57, whereas there is a greatercooperativity in the b-arrestin-2 assay (logab 5 2.1). Thecooperativity factor (log ab) between U69,593 and BMS-986187 at k-OR derived from GTPg35S assay data is 1.1. Thiscompares with full agonists (Leu-enkephalin and SNC80,respectively) at the d-OR that afforded log ab values of 1.18and 1.33 in the b-arrestin-2 assay, 1.67 and 1.0 in the [35S]GTPgS assay, and 2.8 and 2.1 in the AC assay (Burford et al.,2015). Although it is not possible to compare these valuesdirectly because of the different orthosteric ligands used, thesimilarity between log ab values at m-OR, d-OR and k-ORsuggests that the preference of BMS-986187 for the d-OR andk-OR ismainly due to its higher binding affinity. It is worthy ofnote that the cooperativity values at m-OR are greater forb-arrestin-2 thanAC,whereas the opposite is true at the d-OR.Previously, a 100-fold preference was reported for the

ability of BMS-986187 to act as a PAM at the d-OR comparedwith the m-OR. This selectivity was determined by comparingthe potency ratios for the compounds as PAMs at the m-ORusing endomorphin-1 and the d-OR using Leu-enkephalin(Burford et al., 2015), and this could vary depending on theprobes used, a facet that is complicated by the many endog-enous ligands for the opioid receptor family. On the other

Fig. 5. Antagonism between Na+ ions and BMS-986187. (A) The displace-ment byDAMGO of 3H-DPN binding tom-OR is inhibited byNa+ ions (opencircles). BMS-986187 (10 mM) enhances the binding of DAMGO andreduces the potency of Na+ ions to inhibit DAMGO binding (closed circles).(B) The ability of increasing concentrations of NaCl to decrease constitu-tive d-OR–mediated GTPg35S binding in the absence (open circles) orpresence of increasing concentrations of BMS-986187 was measured inmembranes from CHO-d cells. Data were analyzed using GraphPad Prism6.01, and points shown are means 6 S.E.M. of three independentexperiments, each in duplicate.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


hand, the affinity values reported here are for the agonistunoccupied receptor and do not depend on the orthostericligand used; thus, the discrimination obtained is the actualselectivity for BMS-986187 between the m-OR and the k- ord-OR.The apparent lack of selectivity of the allosteric modulator

between k- and d-OR and the low level of selectivity for thesereceptors over m-OR is counter to the many observationsacross other GPCRs that allosteric modulators can selectivelytarget closely related receptors that traditional orthostericligands have been unable to achieve (for review, seeMay et al.,2007; Keov et al., 2011). The idea of nonselectivity at allostericsites has been previously seen with the muscarinic acetylcho-line receptor (mAchR) family of five receptors that, like theopioid receptors, share a very high level of homology. Theallosteric modulator C7/3-phth acts at all subtypes of mAchRs,

although it has the highest affinity for the M2R mAChR(Christopoulos et al., 1999) and LY2033298 is a PAM at bothM2R andM4RmAChRs (Chan et al., 2008; Valant et al., 2012).Indeed, recent crystallographic studies with mAchRs havefound that the allosteric sites of these family members arequite homologous (Thal et al., 2016), giving a rational expla-nation for the difficulty in developing selective allostericligands for these receptors as well as orthosteric ligands;however, selectivity may arise from the cooperativity ofallosteric ligands with orthosteric sites and their ligands.For example, an allosteric ligand may bind different receptorsbut show cooperativity with only the orthosteric site of onereceptor, and this may, or may not, be dependent on theorthosteric ligand.One of the most pressing questions to understand more

completely the mechanism of the allosteric modulators and

Fig. 6. Interactions between BMS-986187 and related allosteric modulators. (A) The ability of 10 mMBMS-986187 to enhance the potency of DAMGO-mediated GTPg35S stimulation at the m-OR in C6m cell membranes was measured in the presence or absence of 30 mM of the m-SAMBMS-986124. (B) InC6d cell membranes, the effect of 300 nM BMS-986187 to enhance the potency of Leu-enkephalin–mediated GTPg35S stimulation was measured in thepresence or absence of the m-SAM 30 mM BMS-986124. (C) The ability of 300 nM BMS-986187 to enhance the potency of Leu-enkephalin–mediatedGTPg35S stimulation at d-ORwasmeasured in the presence or absence of 30 mMof the m-PAMBMS-986122. (D) 30 mMBMS-986122 has a small effect onthe potency of Leu-enkephalin to activate G protein. (E) BMS-986122 and BMS-986124 inhibit the ago-PAM action of BMS-986187 to stimulate GTPg35Sbinding toCHOd cellmembranes. *BMS-986187-mediated stimulation of GTPg35S binding is different from the vehicle control and significantly inhibitedby BMS-986122 or BMS-986122 ( analysis of variance with Tukey’s post hoc test). Data shown are mean 6 S.E.M. from three experiments in duplicate.Potency values were obtained from fitting the data by linear regression with Hill slopes of unity using GraphPad Prism 6.01.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


the rational design of modulators is identification of theirbinding site(s) on the opioid receptors. Comparisons of tworecent and independent molecular dynamics simulationssuggest that the binding of BMS-986122 at m-OR (Bartuziet al., 2016) and BMS-986187 at d-OR (Shang et al., 2016) relyon the same residues at the top of TM domains 2 and7 (Livingston and Traynor, 2017). Indeed, the striking simi-larity in theoretical binding pockets for these ligands at twodifferent opioid receptors supports our pharmacologic evi-dence of a conserved site. In the proposed region, Tyr 2.64, His7.36, and Ile 7.39 (numbers refer to Ballesteros-Weinsteingeneric numbering scheme) (Ballesteros andWeinstein, 1995)are conserved across m-OR, d-OR, and k-OR. In particular, Tyr2.64 may have an important role and is seen to undergo largeconformational changes upon m-OR receptor activation whencomparing the inactive and active crystal structures of thisreceptor (Manglik et al., 2012; Huang et al., 2015). It is alsoworth noting that Tyr 7.35 in k-OR, an amino acid implicated

in the putative allosteric sites on d-OR and m-OR (Bartuziet al., 2016; Shang et al., 2016), is involved in hydrogen bondformationwith the orthosteric antagonist JD-Tic (Wuet al., 2012),which could indicate that JD-Tic is bitopic and reaches into theallosteric site of k-OR. This residue 7.35, as a Trp or Tyr, is crucialfor the binding and function of allosteric modulators and agonist-mediated conformational changes at a number of GPCRs(reviewed in Livingston and Traynor, 2017), including M1(Abdul-Ridha et al., 2014), M2 (Jäger et al., 2007; Haga et al.,2012;Dror et al., 2013),M4 receptors (Thal et al., 2016), andb2ARand M2R (DeVree et al., 2016).The allosteric site for BMS-986122 and BMS-986187 does

not appear to be conserved on NOPR as we failed to seeallosteric activity of the ligands. Differences in amino-acidcomposition are found when comparing the reported putativeallosteric sites in m-OR and d-OR with the same region of theNOPR. This finding is not surprising as NOPR shares theleast amount of homology with the other opioid receptors, and

Fig. 7. BMS-986187 is a PAM at k-OR. (A) Using CHO cells expressing k-OR, the ability of dynorphin A (1–17) to active G protein as measured by theGTPg35S binding assay was determined in the absence or presence of 10 mM BMS-986187. Using the same cell membranes, concentration-responsecurves were generated for (B)U69,593, (C) salvinorin A, and (D) DPN in the presence of vehicle or 10mMBMS-986187. (E) Representative experiments (offour each in duplicate) showing the affinity but not maximal binding of 3H-U69,593 were enhanced in the presence of BMS-986187. (F) BMS-986122 orBMS-986124 did not alter the binding of 3H-U69,593 but blocked the PAMeffect of BMS-986187. *Analysis by ANOVA followed by Tukey post hoc showedsignificant differences between BMS-986187 alone and all other conditions.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


classically it has not been characterized as an opioid receptorowing to its inability to bind the antagonist naloxone (Coxet al., 2015). The orthosteric binding sites of the four opioidreceptors are generally conserved but with one strikingdifference (reviewed in Cox, 2013). In m-OR, d-OR, and k-OR,a His in TM6 is joined by a two-water-molecule bridge to thephenolic hydroxyl on Tyr1 of the opioid peptides or on thearomatic ring of small-molecule opioids, including naloxone.In NOPR, this His is replaced by Gln and so does not form thewater bridge linking to the phenolic hydroxyl. Instead, NOPRligands have a Phe in position 1 that is suggested to interactwith a Tyr in TM3 via p–stacking (Thompson et al., 2012).This difference could account for the lack of cooperativity ofNOPR orthosteric agonists with BMS-986122 and BMS-986187. Nevertheless, it cannot be ruled out that theseallosteric ligands might be modulators of NOPR but havemuch lower affinity so that higher concentrations are requiredor, owing to probe dependence, may exhibit cooperativity withligands other than Ro64-6198 and the endogenous ligandnociceptin that we examined. For example, the CCR5 receptornegative allosteric modulator aplaviroc fully prevents bindingof the chemokine CCR3 but has almost no effect on CCL5binding (Watson et al., 2005). In addition, BMS-986187 andBMS-986122 might be SAMs at NOPR, but without a knownPAM for NOPR, testing this is not currently possible.Few examples of endogenous allosteric modulators for

GPCRs have been reported (for review, see van der West-huizen et al., 2015) including dynorphin A, a putativeallosteric modulator of the M2 mAChR (Hu andEl-Fakahany, 1993); glutathione, an allosteric ligand for thecalcium-sensing receptor (Wang et al., 2006; Broadhead et al.,2011); and lipoxin A4, which has been proposed as anendogenous cannabinoid receptor modulator (Pamplona

et al., 2012). Since the evidence presented suggests that thethree classic members of the opioid family of receptors share asimilar allosteric binding site, it is tempting to speculate thatthis may be caused by evolutionary pressure from an endog-enous allosteric modulator that binds the opioid receptors,although this may be vestigial.

Authorship Contributions

Participated in research design: Livingston, Traynor.Conducted experiments: Livingston, Stanczyk, Burford, Alt.Contributed new reagents or analytic tools: Burford, Alt.Performed data analysis: Livingston, Canals, Traynor.Wrote or contributed to the writing of the manuscript: Livingston,

Traynor.

References

Abdul-Ridha A, Lane JR, Mistry SN, López L, Sexton PM, Scammells PJ, Christo-poulos A, and Canals M (2014) Mechanistic insights into allosteric structure-function relationships at the M1 muscarinic acetylcholine receptor. J Biol Chem289:33701–33711.

Ballesteros JA and Weinstein H (1995) Integrated methods for the construction ofthree-dimensional models and computational probing of structure-function rela-tions in G protein-coupled receptors. Methods Neurosci 25:366–428.

Bartuzi D, Kaczor AA, and Matosiuk D (2016) Interplay between two allosteric sitesand their influence on agonist binding in human m opioid receptor. J Chem InfModel 56:563–570.

Bisignano P, Burford NT, Shang Y, Marlow B, Livingston KE, Fenton AM, RockwellK, Budenholzer L, Traynor JR, Gerritz SW, et al. (2015) Ligand-based discovery ofa new scaffold for allosteric modulation of the m-opioid receptor. J Chem Inf Model55:1836–1843.

Broadhead GK, Mun HC, Avlani VA, Jourdon O, Church WB, Christopoulos A,Delbridge L, and Conigrave AD (2011) Allosteric modulation of the calcium-sensingreceptor by gamma-glutamyl peptides: inhibition of PTH secretion, suppression ofintracellular cAMP levels, and a common mechanism of action with L-amino acids.J Biol Chem 286:8786–8797.

Burford NT, Clark MJ, Wehrman TS, Gerritz SW, Banks M, O’Connell J, Traynor JR,and Alt A (2013) Discovery of positive allosteric modulators and silent allostericmodulators of the m-opioid receptor. Proc Natl Acad Sci USA 110:10830–10835.

Burford NT, Livingston KE, Canals M, Ryan MR, Budenholzer LM, Han Y, Shang Y,Herbst JJ, O’Connell J, Banks M, et al. (2015) Discovery, synthesis, and molecularpharmacology of selective positive allosteric modulators of the d-opioid receptor.J Med Chem 58:4220–4229.

Chan WY, McKinzie DL, Bose S, Mitchell SN, Witkin JM, Thompson RC, Christo-poulos A, Lazareno S, Birdsall NJM, Bymaster FP, et al. (2008) Allosteric modu-lation of the muscarinic M4 receptor as an approach to treating schizophrenia. ProcNatl Acad Sci USA 105:10978–10983.

Chen Y, Mestek A, Liu J, Hurley JA, and Yu L (1993) Molecular cloning and func-tional expression of a mu-opioid receptor from rat brain. Mol Pharmacol 44:8–12.

Christopoulos A, Changeux J-P, Catterall WA, Fabbro D, Burris TP, Cidlowski JA,Olsen RW, Peters JA, Neubig RR, Pin J-P, et al. (2014) International Union ofBasic and Clinical Pharmacology. XC. multisite pharmacology: recommendationsfor the nomenclature of receptor allosterism and allosteric ligands. Pharmacol Rev66:918–947.

Christopoulos A and Kenakin T (2002) G protein-coupled receptor allosterism andcomplexing. Pharmacol Rev 54:323–374.

Christopoulos A, Sorman JL, Mitchelson F, and El-Fakahany EE (1999) Character-ization of the subtype selectivity of the allosteric modulator heptane-1,7-bis-(di-methyl-39-phthalimidopropyl) ammonium bromide (C7/3-phth) at clonedmuscarinic acetylcholine receptors. Biochem Pharmacol 57:171–179.

Clark MJ, Harrison C, Zhong H, Neubig RR, and Traynor JR (2003) Endogenous RGSprotein action modulates m-opioid signaling through Galphao: effects on adenylylcyclase, extracellular signal-regulated kinases, and intracellular calcium path-ways. J Biol Chem 278:9418–9425.

Clark MJ, Linderman JJ, and Traynor JR (2008) Endogenous regulators of G proteinsignaling differentially modulate full and partial mu-opioid agonists at adenylylcyclase as predicted by a collision coupling model. Mol Pharmacol 73:1538–1548.

Conn PJ, Christopoulos A, and Lindsley CW (2009) Allosteric modulators of GPCRs:a novel approach for the treatment of CNS disorders.Nat Rev Drug Discov 8:41–54.

Cox BM (2013) Recent developments in the study of opioid receptors. Mol Pharmacol83:723–728.

Cox BM, Christie MJ, Devi L, Toll L, and Traynor JR (2015) Challenges for opioidreceptor nomenclature: IUPHAR Review 9. Br J Pharmacol 172:317–323.

DeVree BT, Mahoney JP, Vélez-Ruiz GA, Rasmussen SGF, Kuszak AJ, Edwald E,Fung J-J, Manglik A, Masureel M, Du Y, et al. (2016) Allosteric coupling from Gprotein to the agonist-binding pocket in GPCRs. Nature 535:182–186.

Dror RO, Green HF, Valant C, Borhani DW, Valcourt JR, Pan AC, Arlow DH, CanalsM, Lane JR, Rahmani R, et al. (2013) Structural basis for modulation of aG-protein-coupled receptor by allosteric drugs. Nature 503:295–299.

Fenalti G, Giguere PM, Katritch V, Huang X-P, Thompson AA, Cherezov V, Roth BL,and Stevens RC (2014) Molecular control of d-opioid receptor signalling. Nature506:191–196.

Haga K, Kruse AC, Asada H, Yurugi-Kobayashi T, Shiroishi M, Zhang C, Weis WI,Okada T, Kobilka BK, Haga T, et al. (2012) Structure of the human M2 muscarinicacetylcholine receptor bound to an antagonist. Nature 482:547–551.

Fig. 8. BMS-986187 (10 mM) failed to alter the potency of the NOPRagonists (A) nociceptin or (B) Ro-64-6198 to stimulate GTPg35S binding inmembranes from HEK293 cells stably expressing human NOPR. Datashown are means 6 S.E.M. from four experiments in duplicate fitted bylinear regression with Hill slopes of unity using GraphPad Prism 6.01.


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from


Hu J and el-Fakahany EE (1993) Allosteric interaction of dynorphin and myelin basicprotein with muscarinic receptors. Pharmacology 47:351–359.

Huang W, Manglik A, Venkatakrishnan AJ, Laeremans T, Feinberg EN, SanbornAL, Kato HE, Livingston KE, Thorsen TS, Kling RC, et al. (2015) Structural in-sights into m-opioid receptor activation. Nature 524:315–321.

Jäger D, Schmalenbach C, Prilla S, Schrobang J, Kebig A, Sennwitz M, Heller E,Tränkle C, Holzgrabe U, Höltje H-D, et al. (2007) Allosteric small molecules unveila role of an extracellular E2/transmembrane helix 7 junction for G protein-coupledreceptor activation. J Biol Chem 282:34968–34976.

Katritch V, Fenalti G, Abola EE, Roth BL, Cherezov V, and Stevens RC (2014)Allosteric sodium in class A GPCR signaling. Trends Biochem Sci 39:233–244.

Keov P, Sexton PM, and Christopoulos A (2011) Allosteric modulation of G protein-coupled receptors: a pharmacological perspective. Neuropharmacology 60:24–35.

Leach K, Loiacono RE, Felder CC, McKinzie DL, Mogg A, Shaw DB, Sexton PM,and Christopoulos A (2010) Molecular mechanisms of action and in vivo validationof an M4 muscarinic acetylcholine receptor allosteric modulator with potentialantipsychotic properties. Neuropsychopharmacology 35:855–869.

Leach K, Sexton PM, and Christopoulos A (2007) Allosteric GPCRmodulators: takingadvantage of permissive receptor pharmacology. Trends Pharmacol Sci 28:382–389.

Lee KO, Akil H, Woods JH, and Traynor JR (1999) Differential binding properties oforipavines at cloned mu- and delta-opioid receptors. Eur J Pharmacol 378:323–330.

Lin H, Higgins P, Loh HH, Law P-Y, and Liao D (2009) Bidirectional effects offentanyl on dendritic spines and AMPA receptors depend upon the internalizationof mu opioid receptors. Neuropsychopharmacology 34:2097–2111.

Liu W, Chun E, Thompson AA, Chubukov P, Xu F, Katritch V, Han GW, Roth CB,Heitman LH, IJzerman AP, et al. (2012) Structural basis for allosteric regulation ofGPCRs by sodium ions. Science 337:232–236.

Livingston KE and Traynor JR (2014) Disruption of the Na1 ion binding site as amechanism for positive allosteric modulation of the mu-opioid receptor. Proc NatlAcad Sci USA 111:18369–18374.

Livingston KE, and Traynor JR (2017) Allostery at opioid receptors: modulation withsmall molecule ligands. Br J Pharmacol DOI: 10.1111/bph.13823 [published aheadof print].

Manglik A, Kruse AC, Kobilka TS, Thian FS, Mathiesen JM, Sunahara RK, Pardo L,Weis WI, Kobilka BK, and Granier S (2012) Crystal structure of the m-opioid re-ceptor bound to a morphinan antagonist. Nature 485:321–326.

May LT, Leach K, Sexton PM, and Christopoulos A (2007) Allosteric modulation of Gprotein-coupled receptors. Annu Rev Pharmacol Toxicol 47:1–51.

Pamplona FA, Ferreira J, Menezes de Lima O, Jr, Duarte FS, Bento AF, Forner S,Villarinho JG, Bellocchio L, Wotjak CT, Lerner R, et al. (2012) Anti-inflammatorylipoxin A4 is an endogenous allosteric enhancer of CB1 cannabinoid receptor. ProcNatl Acad Sci USA 109:21134–21139.

Pert C and Snyder S (1974) Opiate receptor binding of agonists and antagonistsaffected differentially by sodium. Mol Pharmacol 1:868–879.

Pert CB, Pasternak G, and Snyder SH (1973) Opiate agonists and antagonists dis-criminated by receptor binding in brain. Science 182:1359–1361.

Shang Y, Yeatman HR, Provasi D, Alt A, Christopoulos A, Canals M, and Filizola M(2016) Proposed mode of binding and action of positive allosteric modulators atopioid receptors. ACS Chem Biol 11:1220–1229.

Stevens CW (2009) The evolution of vertebrate opioid receptors. Front Biosci 14:1247–1269.

Szekeres PG and Traynor JR (1997) Delta opioid modulation of the binding of gua-nosine-59-O-(3-[35S]thio)triphosphate to NG108-15 cell membranes: characteriza-tion of agonist and inverse agonist effects. J Pharmacol Exp Ther 283:1276–1284.

Thal DM, Sun B, Feng D, Nawaratne V, Leach K, Felder CC, Bures MG, Evans DA,Weis WI, Bachhawat P, et al. (2016) Crystal structures of the M1 and M4 mus-carinic acetylcholine receptors. Nature 531:335–340.

Thompson AA, Liu W, Chun E, Katritch V, Wu H, Vardy E, Huang XP, Trapella C,Guerrini R, Calo G, et al. (2012) Structure of the nociceptin/orphanin FQ receptorin complex with a peptide mimetic. Nature 485:395–399.

Traynor JR, Corbett AD, and Kosterlitz HW (1987) Diprenorphine has agonist ac-tivity at opioid kappa-receptors in the myenteric plexus of the guinea-pig ileum.Eur J Pharmacol 137:85–89.

Traynor JR and Nahorski SR, Sr (1995) Modulation by mu-opioid agonists of gua-nosine-59-O-(3-[35S]thio)triphosphate binding to membranes from human neuro-blastoma SH-SY5Y cells. Mol Pharmacol 47:848–854.

Valant C, Felder CC, Sexton PM, and Christopoulos A (2012) Probe dependence inthe allosteric modulation of a G protein-coupled receptor: implications for detectionand validation of allosteric ligand effects. Mol Pharmacol 81:41–52.

van der Westhuizen ET, Valant C, Sexton PM, and Christopoulos A (2015) Endoge-nous allosteric modulators of G protein-coupled receptors. J Pharmacol Exp Ther353:246–260.

Wang M, Yao Y, Kuang D, and Hampson DR (2006) Activation of family C G-protein-coupled receptors by the tripeptide glutathione. J Biol Chem 281:8864–8870.

Watson C, Jenkinson S, Kazmierski W, and Kenakin T (2005) The CCR5 receptor-based mechanism of action of 873140, a potent allosteric noncompetitive HIV entryinhibitor. Mol Pharmacol 67:1268–1282.

Wu H, Wacker D, Mileni M, Katritch V, Han GW, Vardy E, Liu W, Thompson AA,Huang XP, Carroll FI, et al. (2012) Structure of the human k-opioid receptor incomplex with JDTic. Nature 485:327–332.

Address correspondence to: John R. Traynor, Department of Pharmacol-ogy, University of Michigan, 1150 West Medical Center Drive, 2301 MSRB III,Ann Arbor, MI 48109. E-mail: [email protected]


at ASPE

T Journals on M

ay 3, 2022m


Dow

nloaded from

mailto:[email protected]


Documents

Pharmacologic Evidence for a Putative Conserved Allosteric