Advertisement
Research ArticleArticle

Differential Pharmacology and Binding of mGlu2 Receptor Allosteric Modulators

Daniel E. O’Brien, Douglas M. Shaw, Hyekyung P. Cho, Alan J. Cross, Steven S. Wesolowski, Andrew S. Felts, Jonas Bergare, Charles S. Elmore, Craig W. Lindsley, Colleen M. Niswender and P. Jeffrey Conn
Molecular Pharmacology May 2018, 93 (5) 526-540; DOI: https://doi.org/10.1124/mol.117.110114

Abstract

Allosteric modulation of metabotropic glutamate receptor 2 (mGlu2) has demonstrated efficacy in preclinical rodent models of several brain disorders, leading to industry and academic drug discovery efforts. Although the pharmacology and binding sites of some mGlu2 allosteric modulators have been characterized previously, questions remain about the nature of the allosteric mechanism of cooperativity with glutamate and whether structurally diverse allosteric modulators bind in an identical manner to specific allosteric sites. To further investigate the in vitro pharmacology of mGlu2 allosteric modulators, we developed and characterized a novel mGlu2 positive allosteric modulator (PAM) radioligand in parallel with functional studies of a structurally diverse set of mGlu2 PAMs and negative allosteric modulators (NAMs). Using an operational model of allosterism to analyze the functional data, we found that PAMs affect both the affinity and efficacy of glutamate at mGlu2, whereas NAMs predominantly affect the efficacy of glutamate in our assay system. More importantly, we found that binding of a novel mGlu2 PAM radioligand was inhibited by multiple structurally diverse PAMs and NAMs, indicating that they may bind to the mGlu2 allosteric site labeled with the novel mGlu2 PAM radioligand; however, further studies suggested that these allosteric modulators do not all interact with the radioligand in an identical manner. Together, these findings provide new insights into the binding sites and modes of efficacy of different structurally and functionally distinct mGlu2 allosteric modulators and suggest that different ligands either interact with distinct sites or adapt different binding poses to shared allosteric site(s).

Introduction

The group II metabotropic glutamate receptors, mGlu2 and mGlu3, have emerged as attractive pharmacological targets for the treatment of multiple brain disorders. These G protein–coupled receptors (GPCRs) function by binding their endogenous agonist, glutamate, in a large Venus Fly Trap orthosteric domain. This extracellular signal is then transduced through the receptors’ seven transmembrane domain (7TM) to Gi/o G proteins to inhibit adenylyl cyclase and regulate ion channels, among other functions (Niswender and Conn, 2010). Selective orthosteric mGlu2/3 agonists and antagonists have efficacy in preclinical models of neurologic and neuropsychiatric disorders (Nickols and Conn, 2014). However, these agonists and antagonists target the highly conserved glutamate orthosteric binding pocket and do not provide mGlu2 or mGlu3 receptor subtype specificity. In addition, most of these agents are amino acid analogs and do not display optimal central nervous system drug properties.

Allosteric modulation of GPCRs has emerged as a new pharmacological strategy to selectively target individual receptor subtypes (Leach et al., 2007; Conn et al., 2009, 2014; Christopoulos, 2014). These modulators bind to sites within the 7TM domain that are less highly conserved than the orthosteric binding site, providing mGlu2 positive allosteric modulators (PAMs) and negative allosteric modulators (NAMs) that can increase and decrease mGlu2 activity, respectively, with good receptor subtype selectivity and improved central nervous system exposure (Hemstapat et al., 2007; Lundström et al., 2011; Andrés et al., 2012; Justinova et al., 2015). mGlu2 PAMs and mGlu2/3 NAMs have efficacy in animal models of neurologic and neuropsychiatric disorders, providing further impetus for optimizing highly selective allosteric modulators (Galici et al., 2005; Benneyworth et al., 2007; Campo et al., 2011; Goeldner et al., 2013; Caprioli et al., 2015; Justinova et al., 2015). To further advance efforts to optimize and develop mGlu2-specific allosteric modulators, it is critical to develop a clear understanding of the molecular mechanisms by which these compounds modulate mGlu2 signaling.

Allosteric modulators of GPCRs regulate receptor-mediated signaling by altering the agonist’s affinity (α) and/or efficacy (β) (Leach et al., 2007; Conn et al., 2009; Kenakin, 2013, 2016, 2017; Zhang and Kavana, 2015). Various modes of allosteric modulation have been described for mGlu receptor allosteric modulators. Neither mGlu1 nor mGlu5 NAMs appear to exhibit affinity cooperativity with glutamate (Litschig et al., 1999; Lavreysen et al., 2003; Bradley et al., 2011; Gregory et al., 2012). Similarly, the affinities of mGlu5 PAMs are unaffected by glutamate, suggesting that they function by modulating the efficacy of endogenous ligands (Gregory et al., 2012). Conversely, mGlu4 allosteric modulators alter agonist affinity for mGlu4 (Poutiainen et al., 2015; Rovira et al., 2015). Previous studies of mGlu2 ligands demonstrated that both mGlu2 PAMs and mGlu2/3 NAMs affect agonist affinity (Lundström et al., 2011; Lavreysen et al., 2013). However, studies have yet to directly test how mGlu2 PAMs and NAMs affect the affinity and efficacy of the endogenous ligand, glutamate, for mGlu2. Due to potential probe-dependent effects and the possible influence of pharmacological mode on in vivo effects (Gregory and Conn, 2015; Rook et al., 2017), it is critical to fully characterize the allosteric modulators’ mode of cooperativity with glutamate.

Additionally, previous studies have proposed the existence of multiple, topographically unique allosteric binding sites in the 7TM region of mGlu1 and mGlu5 (O’Brien et al., 2004; Hemstapat et al., 2006) and interactions with nonidentical binding sites on mGlu4 have been suggested to be associated with distinct modes of allosteric modulator pharmacology (Rovira et al., 2015). Previous studies of mGlu2 allosteric modulators have suggested that mGlu2 PAMs and mGlu2/3 NAMs bind to an overlapping site (Lundström et al., 2011; Farinha et al., 2015; Doornbos et al., 2016). However, it is unclear whether this overlapping allosteric site on mGlu2 is identical for all mGlu2 allosteric modulators. Moreover, homology models for mGlu2 predict that PAMs form a hydrogen bond with Asn735 (Farinha et al., 2015; Doornbos et al., 2016); however, mGlu2/3 NAMs are not predicted to use this residue, suggesting that some NAMs could bind to a topographically unique site (Lundström et al., 2011). Thus, further studies are needed to determine whether all known mGlu2 allosteric modulators share a single allosteric site or can bind to distinct sites.

In this study, we characterized a chemically diverse set of mGlu2 PAMs and NAMs, demonstrating that mGlu2 PAMs act through affinity and efficacy modulation, whereas the mGlu2 NAMs that were tested act predominantly through efficacy modulation. Further mutagenesis studies demonstrated that some mGlu2 PAMs, as well as NAMs, require the Asn735 residue for their functional activity whereas others do not, suggesting that these allosteric modulators could either exhibit different binding modes or bind to distinct sites. To test for distinct modes of binding among these allosteric modulators, we developed a novel mGlu2 PAM radioligand and used this radioligand along with functional studies to show that mGlu2 allosteric modulators can bind to mGlu2 with different binding poses or at distinct allosteric sites.

Materials and Methods

Cell Culture.

Unless otherwise noted, cell culture reagents from Invitrogen (Carlsbad, CA) were used for the below studies. To establish stably expressing wild-type (WT) rat mGlu2 cell lines, the WT construct (Yin et al., 2014) was cloned into the pIRESpuro3 vector, transfected into human embryonic kidney (HEK) cells expressing the G protein inwardly rectifying potassium (GIRK) channel, and stable clones were selected using puromycin. For transient transfections, WT or N735D rat mGlu2 DNA was cloned into the pIRESpuro3 vector and transfected into HEK cells expressing the GIRK channel as outlined below. Cells were cultured as previously described (Niswender et al., 2008) with the addition of nonessential amino acids.

Drugs and Radioligands.

All drugs were synthesized in house through previously described methods unless otherwise noted. The mGlu2/3 NAMs (MNI-137 [4-(7-bromo-4-oxo-4,5-dihydro-3H-benzo[b][1,4]diazepin-2-yl)-pyridine-2-carbonitrile] and decoglurant (5-[2-[7-(trifluoromethyl)-5-[4-(trifluoromethyl)phenyl]pyrazolo[1,5-a]pyrimidin-3yl]ethynyl]pyridin-2-amine)) and mGlu2 NAMs (MRK-8-29 (4-(2-fluoro-4-methoxyphenyl)-7-(2-(2-methylpyrimidin-5-yl)ethyl)quinoline-2-carboxamide) and VU6001192 (6-((cis-2,6-dimethylmorpholino)methyl)-1-(4-fluorophenyl)-4-oxo-1,4-dihydroquinoline-3-carboxamide)) were synthesized as described previously (Hemstapat et al., 2007; Felts et al., 2015; Walker et al., 2015). Likewise, the mGlu2 PAMs, AZD8529 (7-methyl-5-[3-(piperazin-1-ylmethyl)-1,2,4-oxadiazol-5-yl]-2-[[4-(trifluoromethoxy)phenyl]methyl]-3H-isoindol-1-one), AZD8418 (5-{7-chloro-2-[(1S)-1-cyclopropylethyl]-1-oxo-2,3-dihydro-1H-isoindol-5-yl}-N,N-dimethyl-1,2-oxazole-3-carboxamide), BINA (3′-[[(2-cyclopentyl-6,7-dimethyl-1-oxo-2,3-dihydro-1H-inden-5-yl)oxy]methyl]biphenyl-4-carboxylic acid), JNJ-42491293 (8-chloro-3-(cyclopropylmethyl)-7-[4-(3,6-difluoro-2-methoxy-phenyl)-1-piperidinyl]-1,2,4-triazolo[4,3-a]pyridine), and AZ12559322 [N-(3-(2-isopropyl-7-methyl-1-oxoisoindolin-5-yl)phenyl)methanesulfonamide], were synthesized per previously described routes (Galici et al., 2006; Empfield et al., 2007; Andrés et al., 2012; Justinova et al., 2015). Glutamate and LY341495 [(2S)-2-amino-2-[(1S,2S)-2-carboxycycloprop-1-yl]-3-(xanth-9-yl) propanoic acid] were purchased from Tocris Biosciences (Bristol, UK), and [3H]-LY341495 was purchased from American Radiolabled Chemicals Inc. (St. Louis, MO).

Synthesis of the novel AZ12559322 radioligand was performed as described below. A solution of AZ12559322 (23 mg, 0.065 mmol) and N-iodosuccinimide (284 mg, 1.26 mmol) in dimethylsulfoxide (DMSO) (0.5 ml) was stirred at room temperature for 4 days, at which time an additional portion of N-iodosuccinimide (147 mg, 0.652 mmol) was added. The solution was stirred for an additional week and was then diluted with CH2Cl2 (5 ml). The solution was washed with saturated aqueous Na2S2O3 (5 ml) and twice with brine (5 ml). The organic layer was concentrated to dryness and the residue was purified by semi-preparative high-performance liquid chromatography (19 × 250 mm, XBridge C18 5μ OBD, with gradient elution from 10% to 75% MeCN-0.1% aqueous trifluoroacetic acid over 30 minutes; Waters, Milford, MA). The fractions containing di-iodinated products were combined and concentrated to dryness via lyophilization. The solid was then purified by a second semi-preparative high-performance liquid chromatography analysis using the same conditions to give two di-iodinated compounds. Each fraction was concentrated to dryness, taken up in CH2Cl2, and passed through a phase separator. The resulting CH2Cl2 solutions were concentrated to dryness to give 9.2 mg of the early eluting peak N-(2,4-diiodo-5-(2-isopropyl-7-methyl-1-oxoisoindolin-5-yl)phenyl)methanesulfonamide 1H nuclear magnetic resonance (NMR) (400 MHz, CD3OD) δ 1.33 (s, 3H), 1.35 (s, 3H), 2.69 (s, 3H), 3.09 (s, 3H), 4.46 (s, 2H), 4.52–4.64 (m, 1H), 5.49 (s, 1H), 6.94 (s, 1H), 7.10 (s, 1H), 7.26 (d, J = 8.5 Hz, 1H), 7.99 (d, J = 8.5 Hz, 1H). 13C NMR (101 MHz, CD3OD) δ 170.2, 153.3, 152.6, 144.1, 141.1, 140.6, 138.5, 131.9, 130.8, 127.5, 122.7, 101.2, 95.0, 46.1, 44.3, 41.6, 20.8, 17.6. Liquid chromatography-mass spectrometry (LCMS) (M+H) = 611.2, LCMS(M−H) = 609.2 and 7.6 mg of the late eluting peak (N-(2,4-diiodo-3-(2-isopropyl-7-methyl-1-oxoisoindolin-5-yl)phenyl)methanesulfonamide. 1H NMR (400 MHz, DMSO) δ 1.23 (d, J = 6.8 Hz, 6H), 2.65 (s, 3H), 3.05 (s, 3H), 4.38–4.47 (m, 3H), 7.16 (s, 1H), 7.27 (s, 1H), 7.32 (s, 1H), 8.44 (s, 1H), 9.42 (s, 1H). 13C NMR (101 MHz, DMSO) δ 166.9, 148.1, 146.2, 144.3, 142.3, 139.8, 135.8, 130.3, 129.3, 127.1, 121.3, 98.5, 44.3, 42.1, 41.3, 20.4, 16.8. LCMS(M+1) = 611.2, LCMS(M−1) = 609.3 (Scheme 1).

Scheme 1.
Scheme 1.

Synthesis of novel AZ12559322 radioligand.

5′-O-(3-[35S]thio)triphosphate Binding to mGlu2.

To measure activity of AZ12559322 at mGlu2, we used 5′-3-O-(thio)triphosphate (GTPγS) binding assays in membranes from a Chinese hamster ovary cell line expressing human mGlu2 as described previously (Justinova et al., 2015). Briefly, membranes from Chinese hamster ovary cells expressing human mGlu2 were incubated for 15 minutes at 30°C with AZ12559322 in 500 μl assay buffer containing 20 mM HEPES, 100 mM NaCl, 10 mM MgCl2 with 10 μM guanosine diphosphate, and 0.1 nM 5′-O-(3-[35S]thio)triphosphate. After this preincubation, an EC10 concentration of glutamate was added and allowed to incubate for 30 minutes at 30°C. Reactions were terminated through vacuum filtration from a Packard 96-well harvester onto Unifilter-96, GF/B filter microplates (Perkin Elmer, Waltham, MA). After washes with ice-cold water, plates were dried and then 40 μl scintillation fluid (MicroScint 20) was added to the plate. Radioactive counts per minute were then counted on a TopCount plate reader.

mGlu Selectivity.

To assess the selectivity of AZ12559322 in the mGlu receptor subfamily, we used a fluorescence-based cell assay in HEK293 cells expressing human mGlu constructs as described previously (Justinova et al., 2015). Briefly, for all mGlu receptors other than mGlu2, HEK293 cells expressed a chimeric mGlu receptor fused to the human calcium receptor. In these chimeric proteins, the extracellular and transmembrane domains were composed of the human mGlu receptor, whereas the intracellular domain was made up of the human calcium receptor fused to the promiscuous Gqi5 protein. Agonist-induced changes in intracellular calcium levels were measured using a fluorometric imaging plate reader (Molecular Devices, Sunnyvale, CA) as described previously (Levinthal et al., 2009).

Transient Transfections.

On the day before transfections, cells expressing GIRK1 and GIRK2 channels were plated in dishes for either functional assays (10-cm dish with 2 million cells/dish) or membrane preparations (10-mm dish with 5 million cells/dish). The following day, transfections were performed using FuGENE 6 Transfection Reagent (Promega, Madison, WI) at a ratio of 6 μl reagent to 1 μg DNA per Promega’s optimized protocol. Briefly, FuGENE 6 Transfection Reagent and Opti-MEM media were combined, mixed, and incubated at room temperature for 5 minutes. The appropriate amount of DNA was then added, mixed, and incubated for 15 minutes at room temperature. The FuGENE/DNA mixture was then added to dishes (600 μl/10 cm dish and 1500 μl/10 mm dish) and immediately returned to the 37°C incubator. For the functional assay, cells were plated on 384-well plates with poly(d-lysine) (Corning, Corning, NY) as described below. For membrane preparations, cells were incubated for 48 hours and then membranes were prepared as described below.

Functional GIRK-Mediated Thallium Flux Assay.

mGlu2-mediated thallium flux was measured as previously described using HEK293 cells transiently transfected with WT or N735D mGlu2 (Niswender et al., 2008). Briefly, HEK293 cells transiently transfected with a mGlu2 construct were plated onto 384-well plates with poly(d-lysine) (Corning) 24 hours after transfection with 30,000 cells in 20 μl per well in plating media containing Dulbecco’s modified Eagle’s medium, 10% dialyzed fetal bovine serum, 20 mM HEPES, 1 mM sodium pyruvate, and penicillin/streptomycin (all reagents from Life Technologies, Carlsbad, CA). The following day, using an ELX405 microplate washer (BioTek, Winooski, VT), plating media were exchanged for 20 μl/well assay buffer containing 1× Hanks’ balanced salt solution (Gibco, Grand Island, NY) and 20 mM HEPES (pH 7.4) with FluoZin-2 AM dye (330 nM final, prepared as a DMSO stock and mixed as a 1:1 ratio with pluronic acid F-127). After a 1 hour incubation at room temperature, this assay buffer containing dye was exchanged for 20 μl/well assay buffer containing only 1× Hanks’ balanced salt solution (Gibco) and 20 mM HEPES. A Functional Drug Screening System 7000 (Hamamatsu, Hamamatsu City, Japan) was then used to measure thallium flux at room temperature. After two baseline images were taken at 1 Hz (excitation, 470 ± 20 nm; emission, 540 ± 30 nm), 20 μl test compound (2×) was added to each well and incubated for 140 seconds. Then, 10 μl glutamate concentrations (5×) diluted in thallium buffer were added to each well. Data were collected for an additional 2.5 minutes and analyzed post hoc in Microsoft Excel (Microsoft Corp., Redmond, WA) as described previously (Niswender et al., 2008). Functional data were then graphed into GraphPad Prism software (GraphPad Software Inc., La Jolla, CA) and fitted to the operational model of allosterism as described below.

Operational Modeling of Allosterism.

Functional data for NAMs were fitted to the operational model of allosterism as defined by the equation from Leach et al. (2007): Embedded Imagewhere A is the agonist concentration, B is the allosteric modulator concentration, KA is the agonist’s equilibrium dissociation constant, and KB is the allosteric modulator’s equilibrium dissociation constant. Allosteric modulation was defined by both affinity cooperativity (α) and efficacy cooperativity (β), wherein the NAMs’ affinity cooperativity (α) variable was set to equal 1 due to their lack of affinity cooperativity with glutamate in radioligand binding assays. The terms τA and τB quantitate the ability of the agonist and allosteric modulator to induce a response on their own, respectively. Basal, Em, and n represent the basal response of the system, maximal system response, and transducer coefficient relating occupancy to a response, respectively.

Functional data for PAMs were fitted to a simplified version of the operational model of allosterism calculating a composite cooperativity term (αβ):Embedded ImageUsing this equation, a composite cooperativity term (αβ) was calculated that combined both affinity (α) and efficacy (β) cooperativity. All other parameters were the same as defined above.

Functional Potencies of PAMs and NAMs.

Functional potencies for PAMs (EC50) and NAMs (IC50) were determined using the curve fits generated from the operational model of allosterism. For each allosteric modulator, the EC20 (PAMs) or EC80 (NAMs) glutamate concentration was interpolated from the curve fit of the DMSO control. At each subsequent concentration of allosteric modulator (1 nM to 1 μM), the response at this EC20 (PAMs) or EC80 (NAMs) glutamate concentration was interpolated from the curve fits in GraphPad Prism software. Data were plotted in GraphPad Prism and a concentration-response curve was fitted to a four-point parameter logistic equation to determine potencies for each PAM (EC50) and NAM (IC50).

Membrane Preparation.

Once cells reached approximately 80%–90% confluence, dishes were washed four times with ice-cold phosphate-buffered saline and then cells were scraped off of the dishes. Cells were collected using a 5-minute centrifugation at 1000g, the supernatant was removed, and the cell pellet was stored in an −80°C freezer. To prepare membrane protein, cell pellets were thawed, suspended in 10 ml membrane binding buffer (50 mM Tris-HCl with 5 mM MgCl2 and 0.9% NaCl), and then homogenized using three 10-second bursts from a TR-10 polytron (Tekmar, Vernon, BC, Canada). After this, a 10-minute centrifugation at 1000g in an Avanti JE Beckman centrifuge (Beckman Coulter, Brea, CA) was performed at 4°C and the pellet was discarded. The resultant supernatant was centrifuged for 30 minutes at 20,000g at 4°C and the supernatant was discarded. The remaining pellet was resuspended in 1 ml membrane binding buffer and the protein concentration was determined for each sample in triplicate using a BCA assay (Bio-Rad, Hercules, CA) measured by a SmartSpec Plus Spectrophotometer (Bio-Rad). Samples were then stored in an −80°C freezer until used for one of the radioligand binding studies described below.

[3H]-AZ12559322 Binding.

Saturation binding experiments were performed to determine whether [3H]-AZ12559322 binds in a saturable manner to rat mGlu2 (RmGlu2) and/or any other mGlu receptor. Briefly, for initial saturation binding assays, 10 μg RmGlu2-expressing membrane was incubated with 25 μM glutamate at room temperature for 15 minutes prior to adding a range of radioligand concentrations (0.05–20 nM). For saturation binding experiments testing the effect of a glutamate concentration-response curve on binding, 10 μg RmGlu2-expressing membrane was incubated with a fixed concentration of glutamate (0, 1, 3, 10, 30, or 100 μM) prior to adding a range of radioligand concentrations (0.05–20 nM). For selectivity studies, 10 μg mGlu receptor-expressing membrane was incubated with a supersaturating concentration of glutamate (100 μM) prior to adding a range of radioligand concentrations (0.05–20 nM). After radioligand addition, the reaction was allowed to incubate at room temperature for 60 minutes based on association binding studies that demonstrate that this allows for the reaction to reach equilibrium (data not shown). For all of these saturation binding experiments, nonspecific binding was assessed using the mGlu2 PAM, AZD8529. For these assays and all subsequent radioligand binding studies, the reactions were terminated by vacuum filtration through a 96-well harvester (Packard) with ice-cold membrane buffer (50 mM Tris-HCl with 5 mM MgCl2 and 0.9% NaCl). Plates were then dried overnight and 50 μl scintillation fluid (MicroScint 20) was added to the plate. Radioactive counts per minute were then counted on a TopCount plate reader. Saturation binding data were fitted to a one-site binding (hyperbola) in GraphPad Prism.

For displacement binding assays, 10 μg RmGlu2-expressing membrane was incubated with 25 μM glutamate and a concentration-response curve of allosteric modulator at room temperature for 15 minutes. After this incubation, approximately 1 nM [3H]-AZ12559322 was added and incubated with this reaction at room temperature for 60 minutes. Nonspecific binding was assessed using the mGlu2 PAM, AZD8529. Data were fitted to a one-site competition binding curve in GraphPad Prism to determine each allosteric modulator’s affinity (Ki).

Dissociation binding studies were performed at 4°C since the radioligand dissociated too rapidly at room temperature (data not shown). Ten micrograms of RmGlu2-expressing membrane was incubated with a fixed concentration of glutamate (25 μM) prior to adding approximately 6 nM [3H]-AZ12559322. The reaction was allowed to equilibrate for at least 4.5 hours prior to initiating radioligand dissociation. For time points between 5 and 180 minutes before terminating the reaction, 10 μM AZD8529 with or without allosteric modulator (1–10 μM as outlined in the text) experiments were used to compare dissociation rates between control (10 μM AZD8529) and plus allosteric modulator conditions (10 μM AZD8529 plus 1–10 μM allosteric modulator). Data were fitted in GraphPad Prism to either a one-phase or two-phase exponential decay depending on goodness of fit (F-test) as determined in GraphPad Prism. An F-test was used to compare the dissociation constants (koff) between the control and plus allosteric modulator conditions.

[3H]-LY341495 Binding.

For glutamate affinity fold-shift experiments, 20 μg RmGlu2-expressing membrane was incubated for 60 minutes in the absence or presence of allosteric modulator (1 or 10 μM), a concentration-response curve of glutamate (1.5 nM to 1 mM), and approximately 2 nM [3H]-LY341495. Nonspecific binding was determined using a saturating concentration of LY341495 (10 μM). Data were fitted to a one-site competition binding curve to determine glutamate’s affinity (Ki) in the absence and presence of allosteric modulator using GraphPad Prism.

Data Analysis.

All data represent the mean ± S.E.M. from three or more independent experiments unless otherwise noted in the figure legends. Data were analyzed and graphed using Microsoft Excel and GraphPad Prism. Statistical significance was determined using either an unpaired t test or one-way analysis of variance (ANOVA) as outlined in the text unless otherwise noted, and a P value less than 0.05 was considered statistically significant.

Results

mGlu2 PAMs, But Not mGlu2/3 or mGlu2 NAMs, Modulate Glutamate Affinity for mGlu2.

Previous studies have demonstrated that mGlu2 allosteric modulators can exhibit affinity cooperativity (α) with mGlu2 agonists (Lundström et al., 2011; Lavreysen et al., 2013). However, allosteric modulators of other mGlu receptor subtypes often mediate their primary actions by efficacy cooperativity (β) and it is not known whether mGlu2 allosteric modulators affect both the affinity and efficacy of the endogenous orthosteric agonist, glutamate. Moreover, it is not clear whether a range of structurally diverse PAMs and NAMs act through similar or distinct mechanisms. To assess how a range of mGlu2 allosteric modulators bind to mGlu2 and affect its downstream signaling, we identified a set of chemically diverse, previously described mGlu2/3 NAMs, mGlu2 NAMs, and mGlu2 PAMs for further study (Bonnefous et al., 2005; Hemstapat et al., 2007; Lundström et al., 2011; Andrés et al., 2012; Felts et al., 2015; Justinova et al., 2015; Walker et al., 2015) (Fig. 1). To assess affinity cooperativity between allosteric modulators and glutamate, we measured mGlu2 affinity for glutamate (log Ki) by assessing its ability to displace the mGlu2 antagonist radioligand, [3H]-LY341495, in the absence or presence of allosteric modulator. Neither the mGlu2/3 NAMs (MNI-137 and decoglurant) nor the mGlu2-specific NAMs (MRK-8-29 and VU6001192) altered glutamate’s affinity for mGlu2, suggesting that these NAMs do not exhibit any appreciable affinity cooperativity with glutamate at mGlu2 (Fig. 2, A–D). In direct contrast to the NAMs, each of the three structurally distinct mGlu2 PAMs caused robust increases in glutamate’s affinity for mGlu2, demonstrating that these PAMs exhibit strong affinity cooperativity with glutamate at mGlu2 (Fig. 2, E–G; unpaired t test comparing log Ki in absence and presence of PAM; P < 0.001 for each PAM; n = 3 experiments). Fold shifts were calculated to determine the change of glutamate’s affinity in the presence of each PAM (BINA: 5.68 ± 0.66; AZD8418: 5.79 ± 0.91; and JNJ-42491293: 10.47 ± 1.15). Importantly, in the absence of glutamate (Fig. 2; vehicle condition), none of the allosteric modulators completely, or even significantly, displaced binding of [3H]-LY341495 to mGlu2, thereby confirming that these compounds do not directly bind to the orthosteric binding site.

Fig. 1.
Fig. 1.

Chemical structures of mGlu2/3 NAMs (A), mGlu2 NAMs (B), and mGlu2 PAMs (C).

Fig. 2.
Fig. 2.

mGlu2 PAMs, but not mGlu2 NAMs, significantly affect glutamate’s affinity for mGlu2. (A–D) The effect of mGlu2/3 NAMs (A and B) and mGlu2 NAMs (C and D) on glutamate affinity was tested in mGlu2-expressing cells by measuring the Ki of glutamate to compete off the antagonist radioligand, [3H]-LY341495, in the absence or presence of a maximally effective concentration of mGlu2 NAM (1 or 10 μM). (E–G) The effect of mGlu2 PAMs on glutamate affinity was tested in mGlu2-expressing cells by measuring the Ki of glutamate to compete off the antagonist radioligand, [3H]-LY341495, in the absence or presence of a maximally effective concentration of mGlu2 PAM (1 μM). All data are normalized to total binding in the absence of allosteric modulator (Veh) condition and represent the mean ± S.E.M.

Functional Activity of Structurally Distinct mGlu2 NAMs.

Previous studies suggest that Asn735 is a critical residue in the binding site for mGlu2 PAMs; however, some mGlu2/3 NAMs may not use this residue, instead using a different binding mode or a distinct binding site (Schaffhauser et al., 2003; Galici et al., 2006; Hemstapat et al., 2007; Lundström et al., 2011; Farinha et al., 2015). Interestingly, distinct binding sites have been posited to account for different modes of pharmacology for mGlu1 and mGlu4 (Hemstapat et al., 2006; Rovira et al., 2015). Based on this, it is possible that mGlu2 NAMs and PAMs bind to distinct sites, and that this could be important for the difference between NAMs and PAMs in terms of their affinity cooperativity with glutamate. To further verify the necessity of Asn735 for functional activity on mGlu2/3 NAMs and extend this to novel mGlu2 NAMs, we performed a progressive fold-shift experiment assessing NAM activity using the validated GIRK thallium flux assay in cells transiently transfected with either WT or the N735D mGlu2 (Fig. 3). Fitting the data to the operational model of allosterism (see Table 1 for operational model parameters and functional potencies in WT mGlu2) suggested that the mGlu2/3 NAMs, MNI-137 (Fig. 3A) and decoglurant (Fig. 3C), acted as partial NAMs at WT mGlu2 through efficacy modulation. Moreover, consistent with previous studies of mGlu2/3 NAMs (Hemstapat et al., 2007; Lundström et al., 2011), both MNI-137 (Fig. 3B) and decoglurant (Fig. 3D) retained functional activity at N735D mGlu2. Fitting data to the operational model (Table 1) also suggested that the mGlu2-selective NAMs, MRK-8-29 (Fig. 3E) and VU6001192 (Fig. 3G), functioned through efficacy modulation, but as full NAMs. However, whereas MRK-8-29 did retain full, albeit less potent, NAM activity at the N735D mutant (Fig. 3F), VU6001192 completely lost activity at this mutant receptor (Fig. 3H). Thus, not all mGlu2 NAMs retain their functional activity at this mutant receptor, demonstrating that an interaction with Asn735 is not necessary for their negative allosteric modulation.

Fig. 3.
Fig. 3.

mGlu2/3 and mGlu2 NAMs act primarily by modulating glutamate efficacy and some, but not all, retain activity in a BINA binding site mutant. (A–D) The effect of mGlu2/3 NAMs (MNI-137 and decoglurant) and selective mGlu2 NAMs (VU6001192 and MRK-8-29) was tested using the GIRK thallium flux in HEK293 cells transiently transfected with either WT (A and C) or N735D mGlu2 mutant (B and D) as outlined in the Materials and Methods. (E–H) The effect of selective mGlu2 NAMs (VU6001192 and MRK-8-29) was tested using the GIRK thallium flux in HEK293 cells transiently transfected with either WT (E and G) or N735D mGlu2 mutant (F and H) as outlined in the Materials and Methods. Data represent the mean ± S.E.M. and were fitted to the operational model of allosterism using the equation described in the Materials and Methods to determine allosteric modulator affinity (KB) and efficacy cooperativity (β) (see Table 1). Veh, vehicle.

View this table:
TABLE 1

mGlu2 and mGlu2/3 NAMs function predominantly through glutamate efficacy modulation

Functional Activity of Structurally Distinct mGlu2 PAMs.

Based on prior studies, all previously characterized mGlu2 PAMs require Asn735 for their functional activity (Schaffhauser et al., 2003; Galici et al., 2006; Hemstapat et al., 2007; Rowe et al., 2008; Farinha et al., 2015). Further homology modeling studies predict that mGlu2 PAMs form a hydrogen bond with this Asn735, indicating that this residue may be critical for binding affinity of PAMs to the allosteric pocket (Farinha et al., 2015; Doornbos et al., 2016). Thus, to validate our previous findings with BINA and extend them to structurally distinct mGlu2 PAMs, we performed further progressive fold-shift experiments assessing PAM activity in cells transiently transfected with either WT or the N735D mGlu2. At WT mGlu2, BINA (Fig. 4A) and AZD8418 (Fig. 4C) exhibited PAM activity that appeared to be through both efficacy and affinity cooperativity based on a comparison of the cooperativity (αβ) from the operational model to the glutamate affinity fold shifts (see Table 2 for operational model parameters and functional potencies in WT mGlu2; BINA fold shift: 5.68 ± 0.66 and AZD8418 fold shift: 5.79 ± 0.91). In contrast, JNJ-42491293 (Fig. 4E) exhibited PAM activity that, based on a comparison between the cooperativity (αβ) from the operational model to the glutamate affinity fold shifts, may result predominantly from affinity cooperativity (see Table 2 for operational model parameters; JNJ-42491293 fold shift: 10.47 ± 1.15). Moreover, all PAMs demonstrated agonist-like activity in the absence of glutamate, indicating that, in this assay, they may function as agonist-PAMs (Fig. 4, A, C, and E). Consistent with previous findings, we found that BINA completely loses its functional activity in cells expressing the N735D mGlu2 mutant (Hemstapat et al., 2007) (Fig. 4B). Likewise, AZD8418 completely lost functional activity in the cells expressing the N735D mutant, suggesting that this residue is necessary for AZD8418 PAM activity (Fig. 4D). In contrast, JNJ-42491293 retained its functional activity in cells expressing the N735D mutant, providing evidence of an mGlu2 PAM that either acts through a different binding mode or a distinct, nonidentical binding site (Fig. 4F). Taken together, the NAM and PAM functional activity in the N735D mutant suggests either distinct binding modes or multiple binding sites for allosteric modulators on mGlu2.

Fig. 4.
Fig. 4.

mGlu2 PAMs exhibit intrinsic agonist activity primarily through modulating glutamate affinity, but not all are affected by a BINA binding site mutant. (A–F) The effect of mGlu2 PAMs on a glutamate concentration-response curve was tested using the GIRK thallium flux assay in HEK cells transiently expressing either WT (A, C, and E) or N735D mGlu2 mutant (B, D, and F) as outlined in the Materials and Methods. Data represent the mean ± S.E.M. and were fitted to the operational model of allosterism using the equation described in the Materials and Methods to determine allosteric modulator affinity (KB) and cooperativity (αβ) (see Table 2).

View this table:
TABLE 2

mGlu2 PAMs function predominantly through glutamate affinity modulation

To directly test the hypothesis that these mGlu2 modulators exhibit distinct binding interactions with mGlu2, we sought to develop an mGlu2 allosteric modulator radioligand. Based on functional affinities (KB) and potencies (EC50 and IC50) determined using the operational model of allosterism (see Tables 1 and 2 for KB, EC50, and IC50 values), none of the allosteric modulators possessed the low nanomolar affinity necessary for a radioligand. Although functional affinities and potencies do not necessarily correspond to binding affinity (Kenakin, 2014), functional affinity has been shown to directly relate to affinity for other mGlu receptors (Gregory et al., 2012). Yet previous findings demonstrated that JNJ-42491293 exhibited low nanomolar affinity for mGlu2, suggesting that functional affinity may not correspond to binding affinity for mGlu2 (Andrés et al., 2012). Although JNJ-42491293 was developed as a radiotracer, it was recently shown to exhibit nonspecific binding in vivo, suggesting that it may not be a suitable for radioligand binding studies (Leurquin-Sterk et al., 2017). Thus, further chemical optimization efforts were undertaken to develop a more suitable mGlu2 allosteric modulator for radiolabeling.

AZ12559322 Is a Potent and Selective mGlu2 PAM.

Chemical optimization efforts led to the discovery of AZ12559322 (Fig. 5A) as a novel mGlu2 PAM that has properties suggesting that it may be suitable for use as a radioligand. Complementary functional and binding experiments were conducted in cells expressing mGlu2 to further study this compound. We performed progressive fold-shift experiments assessing the PAM activity of AZ12559322 in cells transiently transfected with either WT or the N735D mGlu2 using the validated GIRK thallium flux assay. Similar to the structurally distinct PAMs described above, AZ12559322 exhibited apparent agonist-PAM activity in cells transiently expressing WT mGlu2 (Fig. 5B; see Table 3 for operational model parameters and its functional potency derived as described in the Materials and Methods). Moreover, its functional activity was completely abrogated in cells expressing the N735D mGlu2 mutant, demonstrating that, like BINA and AZD8418, AZ12559322 requires this residue for PAM activity (Fig. 5C). In membranes expressing WT mGlu2, AZ12559322 increased glutamate’s affinity for mGlu2 as evidenced by the shift in a glutamate’s ability to compete off the mGlu2 antagonist radioligand, [3H] LY341495 (Fig. 5D, unpaired t test comparing log Ki in absence and presence of PAM; P < 0.001 for AZ12559322; n = 3 experiments). The affinity fold shift was calculated to determine the change of glutamate’s affinity in the presence of AZ12559322 (fold shift = 3.42 ± 0.73). Importantly, AZ12559322 did not significantly impact the binding of the antagonist radioligand, [3H]-LY341495, on its own, thereby verifying its binding to an allosteric site (Fig. 5D, vehicle treatment).

Fig. 5.
Fig. 5.

AZ12559322 is a novel mGlu2 PAM acting through glutamate affinity modulation and is dependent on N735 for its activity. (A) Chemical structure of AZ12559322. (B and C) The effect of AZ12559322 on glutamate responses was tested in HEK cells transiently expressing WT mGlu2 (B) or N735D mGlu2 (C) by measuring mGlu2-induced GIRK thallium flux. DMSO or 1 nM to 1 μM AZ12559322 was added to cells 140 seconds prior to a serial dilution of glutamate. Data from measured GIRK thallium flux responses are normalized to the maximal response to 1 mM glutamate. Symbols represent vehicle, 1 μM, and 300, 100, 30, 10, and 3 nM. (D) The effect of AZ12559322 on glutamate affinity was tested in mGlu2-expressing cells by measuring the Ki of glutamate to compete off the antagonist radioligand, [3H]-LY341495, in the absence or presence of a maximally effective concentration of AZ12559322 (1 μM). Binding data are normalized to total binding in the absence of allosteric modulator (Veh) condition. For all panels, data are graphed as the mean ± S.E.M.

View this table:
TABLE 3

AZ12559322 is a potent and selective mGlu2 PAM

Further experiments were performed to assess whether AZ12559322 was selective for mGlu2 over other mGlu receptor subtypes. In vitro GTPγS binding studies on cells expressing human mGlu2 showed that AZ12559322 potentiated an EC20 concentration of glutamate with potency of 10 nM and produced 77.5% of the maximal glutamate effect (Table 3, mean of n = 2 experiments). To determine whether AZ12559322 was selective for mGlu2, we performed further in vitro binding studies. As outlined in Table 3, AZ12559322 demonstrated no PAM or agonist activity at any other mGlu receptor subtype (mGlu1, mGlu3, mGlu4, mGlu6, mGlu7, mGlu8; Table 3, EC50 > 25 μM, n = 2 experiments). However, AZ12559322 did exhibit weak PAM activity at mGlu5 with an EC50 of 769 nM and produced 66.8% of the maximal glutamate effect (Table 3, mean of n = 2 experiments). Thus, comparing its PAM potencies at mGlu2 and mGlu5, AZ12559322 exhibited 76.9-fold selectivity for mGlu2 over mGlu5. Interestingly, although AZ12559322 did exhibit robust PAM activity in the mGlu5 assay, no agonist-PAM activity was noted up to 25 µM.

[3H]-AZ12559322 Is a Suitable In Vitro mGlu2 PAM Radioligand Tool Compound.

Due to its functional potency, selectivity, and amenable physiochemical properties, further studies were conducted to radiolabel and validate AZ12559322 as a suitable radioligand for mGlu2. After successful radiolabeling, we confirmed that, under experimental conditions, the binding of [3H]-AZ12559322 to mGlu2 reached equilibrium in the presence of a supersaturating (25 μM) concentration of the orthosteric agonist, glutamate (data not shown). Under these conditions in membranes stably expressing mGlu2, saturation binding experiments demonstrated that [3H]-AZ12559322 bound with a high affinity (Kd = 0.80 ± 0.07 nM, n = 3 experiments) and to a significant number of binding sites on mGlu2 (Bmax = 5306 ± 174 fmol/mg, n = 3 independent experiments) (Fig. 6A). Previous studies of mGlu2 and other GPCRs illustrated that the binding of such PAM radioligands is affected by concentration of an orthosteric agonist; however, whether the agonist affects the affinity and/or number of binding sites has not been consistent across radioligands even at mGlu2 (Lavreysen et al., 2013; Doornbos et al., 2016; Smith et al., 2016). Thus, in further saturation binding studies, [3H]-AZ12559322 affinity and number of binding sites were measured in the presence of different concentrations of the endogenous orthosteric agonist, glutamate (1–100 μM). As evidenced in the representative saturation binding experiment depicted in Fig. 6B, increasing concentrations of glutamate enhanced the affinity of [3H]-AZ12559322 for mGlu2 without affecting the number of binding sites. Data from three to four independent experiments are quantified in Fig. 6, C and D, illustrating that glutamate did not alter the number of binding sites (Bmax, Fig. 6C; one-way ANOVA) but did increase the affinity of [3H]-AZ12559322 for mGlu2 (Fig. 6D; one-way ANOVA with a Dunnett post-test compared with 0 μM glutamate, P < 0.05). To ensure that [3H]-AZ12559322 bound specifically to mGlu2 over other mGlu receptors, additional saturation binding studies were performed using a supersaturating concentration of glutamate (100 μM) in membranes expressing different mGlu receptors. As in Fig. 6, A and B, [3H]-AZ12559322 bound with a high affinity to mGlu2; however, it showed no appreciable binding to any of the other mGlu receptors (Fig. 6E).

Fig. 6.
Fig. 6.

[3H]-AZ12559322 binds with high affinity to mGlu2 in a glutamate-dependent manner and selectively binds to mGlu2 over other mGlu receptors. (A) Representative saturation binding experiment with [3H]-AZ12559322 demonstrates saturable binding in the presence of 25 μM glutamate (Bmax = 5306 ± 174 fmol/mg and Kd = 0.80 ± 0.07; n = 3 experiments). (B) Representative saturation binding experiment with [3H]-AZ12559322 with varying concentrations of glutamate (0–100 μM) demonstrates that glutamate concentration affects Kd (D), but not Bmax (C) (data from n = 3 to 4 experiments, one-way ANOVA with a Dunnett’s post hoc test, *P < 0.05). (E) Representative saturation binding experiment demonstrates that, in the presence of 25 μM glutamate, [3H]-AZ12559322 binds selectively to mGlu2 over other mGlu receptors. (F and G) Saturation binding data from three experiments show Bmax (mean ± S.E.M.) for [3H]-AZ12559322 (F) and [3H]-LY341495 (G) in HEK cells transiently transfected with WT or N735D mGlu2 (unpaired t tests, *P < 0.05).

Since the functional activity of AZ12559322 was completely abolished at the N735D mGlu2 mutant, further experiments were performed to determine whether Asn735 is necessary for AZ12559322 binding as well as its functional activity. Saturation binding experiments using [3H]-AZ12559322 in membranes transiently transfected with either WT or N735D mGlu2 demonstrated appreciable binding of [3H]-AZ12559322 only at WT mGlu2 (Fig. 6F; unpaired t test, P < 0.05). As a critical control, further experiments using the mGlu2 orthosteric antagonist radioligand, [3H] LY341495, illustrated that the Bmax, or number of binding sites, was comparable in the same membranes transiently transfected with WT and N735D mGlu2 (Fig. 6G; unpaired t test, P = 0.80). Taken together, these radioligand binding studies show that the selective and high-affinity binding of [3H]-AZ12559322 to mGlu2 requires Asn735.

We next tested whether all of the structurally diverse allosteric modulators could completely compete off [3H]-AZ12559322. These studies aimed to verify that binding of [3H]-AZ12559322 is reversible and to assess whether mGlu2 allosteric modulators act at a distinct site. Interestingly, displacement binding experiments revealed that all of the structurally distinct PAMs completely inhibited [3H]-AZ12559322 binding to mGlu2 (Fig. 7A; Table 4 for Ki values). Similarly, all structurally distinct mGlu2/3 and mGlu2 NAMs completely inhibited [3H]-AZ12559322 binding to mGlu2 (Fig. 7B; Table 4 for Ki values).

Fig. 7.
Fig. 7.

Affinities (Ki) of structurally distinct mGlu2 PAMs and NAMs for mGlu2. (A) Determination of selective mGlu2 PAM affinities using competition binding studies with the mGlu2 PAM radioligand, [3H]-AZ12559322. (B) Determination of mGlu2/3 NAMs (decoglurant and MNI-137) and selective mGlu2 NAM (VU6001192 and MRK-8-29) affinities using competition binding studies with the mGlu2 PAM radioligand, [3H]-AZ12559322. Data are representative of three experiments expressed as the mean ± S.E.M.

View this table:
TABLE 4

mGlu2 affinities (Ki) of mGlu2 PAMs and NAMs determined by [3H]-AZ12559322 displacement experiments

Data represent the mean ± S.E.M. of three individual experiments performed in triplicate.

mGlu2 Allosteric Modulator Binding to mGlu2 Suggests Nonidentical Modes of Interaction.

Although each allosteric modulator inhibited binding of the novel PAM radioligand, this does not necessarily indicate that these allosteric modulators all bind to the same site that is labeled by [3H]-AZ12559322 (Limbird, 2005). Based on the mutagenesis studies, as well as the finding that [3H]-AZ12559322 binding is lost at the N735D mGlu2 mutant, it is possible that some allosteric modulators inhibit [3H]-AZ12559322 binding through an allosteric mechanism rather than through competitive interactions. Thus, these allosteric modulators may bind to distinct sites and ligands at these sites may exhibit mutual negative cooperativity. To test this hypothesis, we performed dissociation binding experiments to evaluate the off-rate dissociation kinetics (Koff) of [3H]-AZ12559322 in the absence and presence of selected allosteric modulators. For these experiments, the selected PAMs and NAMs were added to the assay buffer at concentrations far in excess of predicted saturating concentrations (100×–1000× determined Ki values; see Table 4) to determine whether their addition affects the off-rate kinetics of [3H]-AZ12559322. A decrease or increase in the [3H]-AZ12559322 off-rate would indicate positive or negative cooperativity with a nonidentical allosteric site, respectively. Consistent with data from the mutagenesis studies, 1 μM JNJ-42491293 increased the off-rate of [3H]-AZ12559322 compared with control conditions, providing evidence that JNJ-42491293 binds to a nonidentical allosteric site with negative cooperativity with the site defined by AZ12559322 (Fig. 8A; F-test comparing koff values in Table 5, P < 0.001). In agreement with functional data from N735D mGlu2, 10 μM AZD8418 did not alter dissociation kinetics compared with control dissociation conditions, consistent with AZD8418 binding to the site in a manner that is similar to that of AZ12559322 (Fig. 8B; F-test comparing koff values in Table 5, P = 0.68). Likewise, 1 μM BINA did not affect dissociation kinetics compared with control conditions, consistent with AZ12559322 binding to the canonical BINA binding site (Fig. 8C; F-test comparing koff values in Table 5, P = 0.68). In contrast, both decoglurant and MRK-8-29 increased the off-rate of [3H]-AZ12559322 compared with control conditions, providing evidence that both NAMs bind to mGlu2 in a manner that is not identical to that of the radioligand and that they exhibit negative cooperativity with AZ12559322 (Fig. 8, D; F-test comparing koff values in Table 5, P < 0.001 for both NAMs). Interestingly, both decoglurant and MRK-8-29 changed the kinetics of dissociation from a one-phase decay to two-phase decay (see Table 5). However, in contrast with predictions from the mutagenesis data (see Fig. 3H), VU6001192 increased the off-rate of [3H]-AZ12559322 compared with control conditions, providing evidence that VU6001192 binds to a nonidentical allosteric site and confers negative cooperativity with regard to [3H]AZ12559322 binding (Fig. 8F; F-test comparing koff values in Table 5, P < 0.001). This could represent negative cooperativity between two distinct binding sites for VU6001192 and AZ12559322, or it could reflect changes in dissociation due to other actions, including effects on interactions to two protomers of the mGlu2 homodimer. However, regardless of the specific mechanism, when taken together, these data provide further evidence for nonidentical modes of interaction with allosteric sites on mGlu2 that are not specifically defined by whether Asn735 is necessary for the allosteric modulators’ function.

Fig. 8.
Fig. 8.

The dissociation rate of [3H]-AZ12559322 is affected by certain mGlu2 PAMs or NAMs, suggesting the presence of multiple allosteric binding sites. (A) Saturating concentrations of JNJ-42491293 (1 μM) increase the dissociation rate of [3H]-AZ12559322 from mGlu2-expressing membranes compared with the control dissociation induced by 10 μM AZD8529. (B and C) Neither a saturating concentration of AZD8418 (10 μM) nor BINA (1 μM) altered the dissociation rate of [3H]-AZ12559322 from mGlu2-expressing membranes compared with control dissociation induced by 10 μM AZD8529. (D–F) The mGlu2 NAMs, decoglurant (D; 3 μM), MRK-8-29 (E; 3 μM), and VU6001192 (F; 10 μM) all increase the dissociation rate of [3H]-AZ12559322 from mGlu2-expressing membranes compared with the control dissociation induced by 10 μM AZD8529. Data are representative of three experiments expressed as the mean ± S.E.M. (F-test, ***P < 0.001).

View this table:
TABLE 5

Dissociation binding kinetics (Koff) of [3H]-AZ12559322 in absence and presence of allosteric modulator

Data represent the mean ± S.E.M. of three individual experiments performed in triplicate. Values in parentheses are t1/2 in minutes.

Discussion

In this study, we characterized the in vitro molecular pharmacology properties of a chemically diverse set of mGlu2 PAMs and NAMs to probe allosteric mechanisms and inform further chemical optimization. We found that each of the mGlu2 PAMs evaluated appears to exhibit intrinsic agonist activity and modulate both affinity and efficacy of glutamate at mGlu2. Interestingly, the mGlu2/3 and mGlu2 NAMs studied did not alter glutamate’s affinity, and functional studies suggested that the NAMs predominantly act through efficacy modulation, in contrast to mGlu2 PAMs. Moreover, through the characterization of a novel mGlu2 PAM and its radioligand, we demonstrated that an mGlu2 PAM and multiple mGlu2 NAMs exhibit distinct modes of binding to mGlu2, which could reflect binding to a single site with different binding poses or interactions with distinct allosteric sites from the prototypical mGlu2 PAM, BINA. Taken together, these data further refine our understanding of mGlu2 allosteric modulation, serving to inform further drug discovery efforts.

Although our data suggest the possibility that mGlu2 PAMs possess intrinsic agonist activity, this apparent intrinsic activity of our mGlu2 PAMs could result from potentiation of glutamate released by cells or a large receptor reserve in our overexpression system (Noetzel et al., 2012; Rovira et al., 2015). Indeed, in overexpression systems with large receptor reserves, some mGlu5 PAMs exhibit robust agonist-PAM activity that is not seen in native tissues presumably due to lower receptor expression levels (Noetzel et al., 2012; Rook et al., 2013). However, other optimized mGlu5 ago-PAMs have demonstrated intrinsic agonist activity in native tissues that can have a major impact in the overall in vivo activity (Rook et al., 2013). Additionally, M1 PAMs also exhibit intrinsic agonist activity in native tissues and induce robust seizure activity (Davoren et al., 2016). In future studies, it will be important to systematically evaluate the propensity of different mGlu2 PAMs to display allosteric agonist activity and determine the impact of agonist activity in overall in vivo effects.

Consistent with previous findings, all mGlu2 PAMs, including our novel PAM radioligand, demonstrated robust affinity cooperativity with the orthosteric site. Previous studies reported such affinity cooperativity for mGlu2 PAMs with orthosteric agonists; however, the mechanism through which various PAMs affected agonist affinity differs from one PAM to another (Lavreysen et al., 2013; Doornbos et al., 2016). Similar to our reported findings, a PAM radioligand’s affinity for mGlu2 increased in the presence of agonist without a corresponding change in the number of binding sites (Lavreysen et al., 2013). In contrast, another structurally distinct mGlu2 PAM radioligand displayed an agonist-dependent increase in the number of binding sites without any change in its affinity for the receptor (Doornbos et al., 2016). These findings may reflect differential PAM effects on receptor active conformations, wherein PAMs can either increase the agonist’s affinity for the receptor or shift the population of receptors in the active state. Indeed, direct comparison of Bmax values determined from mGlu2 PAM, agonist, and antagonist radioligands support this hypothesis. Based on studies of mGlu receptor dimers, PAM binding to one protomer of the homodimer is sufficient to potentiate receptor-mediated downstream signaling, suggesting that PAMs may bind to only one side of the homodimer. In support of this, Lundström et al. (2011) found that an mGlu2 PAM radioligand recognized approximately 40% of the binding sites observed with an mGlu2 agonist radioligand, consistent with the PAM binding to one protomer of the active-state homodimer receptors or to a site spanning the dimer interface. Additionally, this group demonstrated that an mGlu2 agonist radioligand recognizes nearly half of the binding sites on mGlu2 compared with an mGlu2 antagonist radioligand, suggesting that, as seen with other GPCRs, the antagonist binds both active and inactive receptor states (Lundström et al., 2009). Taken together, these data may explain how some PAMs increase the number of agonist-bound sites and our finding that [3H]-AZ12559322 recognizes less than 20% of the sites seen with the orthosteric antagonist radioligand.

In contrast to PAMs, mGlu2/3 and mGlu2 NAMs act predominantly through efficacy cooperativity, thereby delineating the allosteric modes of action for PAMs and NAMs. Based on functional and binding data, mGlu2 PAMs appear to function through a combination of affinity and efficacy modulation since the PAM-induced fold shift of glutamate’s affinity for mGlu2 cannot account for the cooperativity (αβ) determined using the operational model of allosterism. Conversely, neither mGlu2/3 nor mGlu2 NAMs affect glutamate’s affinity, indicating that they function predominantly through efficacy modulation. However, it is important to avoid generalizing the results of studies with the mGlu2 NAMs included here to all mGlu2 NAMs. Indeed, a previous study using two mGlu2/3 NAMs that were not included here found that these NAMs could reduce the affinity of the agonist radioligand, [3H]-LY354740, for the orthosteric site (Lundström et al., 2011). This could also suggest that some NAMs can act by affinity modulation. However, it is also possible that this reflects a probe dependence of NAM effects on glutamate versus LY354740 binding or the use of human mGlu2 in this GTPγS. Probe dependence is a common phenomenon for allosteric modulators and has been posited to account for differential affinity modulation seen with mGlu5 PAMs (Gregory and Conn, 2015). Specifically, when assessing PAM affinity using inhibition of the [3H]-MPEP binding, quisqualate, but not glutamate, increased the affinity of mGlu5 PAMs for the receptor (Bradley et al., 2011; Gregory et al., 2012). Indeed, for the class A GPCR M5, both a PAM and NAM displayed probe-dependent effects on agonist affinity (Berizzi et al., 2016). Overall, these findings illustrate differences in mGlu2 PAM and NAM pharmacology and highlight the importance of using the endogenous ligand to assess pharmacological parameters of allosteric modulators.

Understanding the modes of action for allosteric modulators is important since it varies between allosteric modulators of other GPCRs and it could influence in vivo pharmacological properties. Among the class C mGlu receptors, neither mGlu1 nor mGlu5 allosteric modulators that have been evaluated exhibit affinity cooperativity with the endogenous agonist (Litschig et al., 1999; Gregory et al., 2012); however, both l-(+)-2-amino-4-phosphonobutyric acid and glutamate demonstrated affinity cooperativity with mGlu4 PAMs (Rovira et al., 2015). The affinity cooperativity seen with mGlu4 and our mGlu2 PAMs may be critically important in the further development of in vivo radiotracers to measure receptor occupancy since its binding could depend on local agonist concentration and thus muddle the interpretation of in vivo occupancy data. Thus, high-affinity mGlu2 NAMs may be better tools for further development of radiotracers due to their lack of affinity cooperativity with the endogenous ligand, glutamate. This may also apply more broadly to development of allosteric modulator radiotracers for other GPCRs since allosteric modulators of the class A receptors, M1, M4, and M5, all exhibit affinity cooperativity with the endogenous ligand and the binding of a recent M1 PAM radioligand is dependent on agonist concentration (Bubser et al., 2014; Berizzi et al., 2016; Smith et al., 2016; Rook et al., 2017). The allosteric modulator mode of action may also be important for in vivo physiologic effects including adverse side effect profile. A recent study demonstrated that the seizure liability of structurally distinct M1 PAMs varied potentially due to subtle differences in pharmacological properties including differing degrees of affinity cooperativity with acetylcholine (Rook et al., 2017). Thus, drug discovery efforts should fully characterize the cooperativity of allosteric modulators and optimize these properties based on how the cooperativity corresponds to behavioral outcomes on target-by-target basis.

The presence of multiple modes of interaction of allosteric modulators with binding sites on mGlu2 raises further questions and complications that must be addressed to inform additional drug discovery efforts. Previous homology modeling studies of mGlu2 allosteric modulators posited that both PAMs and NAMs occupied a common and overlapping binding pocket (Lundström et al., 2011; Farinha et al., 2015). Interestingly, our dissociation binding data indicate that some mGlu2 allosteric modulators bind to a nonidentical site from AZ12559322; however, this does not rule out the possibility that these distinct sites are overlapping. Regardless, the presence of multiple sites complicates future in vivo studies since Rook et al. (2015) demonstrated that an mGlu5 NAM positron emission tomography ligand was not displaced by mGlu5 PAMs that bind to overlapping, but nonidentical, sites. Therefore, with the further development of mGlu2 radiotracers, in vitro experiments must ensure that allosteric modulators occupy the identical allosteric site to accurately establish in vivo occupancy-efficacy relationships. Another potential consequence of multiple binding sites is that allosteric modulators that bind to one site may exhibit different pharmacological properties from those binding at another. As a recent example of this, the degree of mGlu4 PAM cooperativity corresponded with their predicted binding to distinct allosteric sites based on homology modeling (Rovira et al., 2015). While our data do not suggest a relationship between cooperativity of these nonidentical sites, it remains possible that such a relationship between pharmacology and binding site could exist. Regardless, further development of mGlu2 allosteric modulators must take into account the multiple binding sites and should better define these sites using mutagenesis and homology modeling.

In conclusion, this study provides a further characterization of mGlu2 allosteric modulators and demonstrates the presence of nonidentical allosteric sites on mGlu2. These findings should be considered in further drug discovery efforts especially for the development of novel radiotracers. Yet questions remain about the physiologic relevance of potential ago-PAM activity, cooperativity, and the distinct allosteric sites on mGlu2. Thus, further studies with novel in vivo tool compounds are necessary to define if and how the discrete in vitro pharmacological properties translate to behavioral responses.

Acknowledgments

The authors thank members of the Conn laboratory for their invaluable input, especially Samantha E. Yohn and Daniel J. Foster.

Authorship Contributions

Participated in research design: O’Brien, Cross, Wesolowski, Elmore, Niswender, Conn.

Conducted experiments: O’Brien, Shaw, Cho.

Contributed new reagents or analytic tools: Felts, Bergare, Elmore, Lindsley.

Performed data analysis: O’Brien, Shaw, Cho.

Wrote or contributed to the writing of the manuscript: O’Brien, Conn.

Footnotes

    • Received August 8, 2017.
    • Accepted March 12, 2018.

  • This work was supported by the National Institutes of Health National Institute of Neurological Disorders and Stroke [Grant R37-NS031373 (to P.J.C.)], the National Institutes of Health National Institute of Mental Health [Grant T32-MH093366 (to P.J.C. and D.E.O.)], the PhRMA Foundation [Postdoctoral Fellowship (to D.E.O.)], and AstraZeneca. AstraZeneca also supplied reagents and compounds for this study. Over the past year, C.W.L. consulted for Abbott. P.J.C., C.W.L., and C.M.N. received research/salary support from AstraZeneca and are inventors on multiple composition of matter patents protecting allosteric modulators of G protein–coupled receptors. A.J.C., S.S.W., J.B., and C.S.E. are/were employees of AstraZeneca during this study and may hold stock options.

  • https://doi.org/10.1124/mol.117.110114.

Abbreviations

7TM
seven transmembrane domain
α
affinity cooperativity
ANOVA
analysis of variance
AZ12559322
N-(3-(2-isopropyl-7-methyl-1-oxoisoindolin-5-yl)phenyl)methanesulfonamide
AZD8418
5-{7-chloro-2-[(1S)-1-cyclopropylethyl]-1-oxo-2,3-dihydro-1H-isoindol-5-yl}-N,N-dimethyl-1,2-oxazole-3-carboxamide
AZD8529
7-methyl-5-[3-(piperazin-1-ylmethyl)-1,2,4-oxadiazol-5-yl]-2-[[4-(trifluoromethoxy)phenyl]methyl]-3H-isoindol-1-one
β
efficacy cooperativity
Bmax
concentration of specific binding sites for the radioligand
BINA
3′-[[(2-cyclopentyl-6,7-dimethyl-1-oxo-2,3-dihydro-1H-inden-5-yl)oxy]methyl]biphenyl-4-carboxylic acid
DMSO
dimethylsulfoxide
GIRK
G protein inwardly rectifying potassium channel
GPCR
G protein–coupled receptor
GTPγS
5′-3-O-(thio)triphosphate
HEK
human embryonic kidney
JNJ-42491293
8-chloro-3-(cyclopropylmethyl)-7-[4-(3,6-difluoro-2-methoxy-phenyl)-1-piperidinyl]-1,2,4-triazolo[4,3-a]pyridine
LY341495
(2S)-2-amino-2-[(1S,2S)-2-carboxycycloprop-1-yl]-3-(xanth-9-yl) propanoic acid
mGlu
metabotropic glutamate receptor
MNI-137
4-(7-bromo-4-oxo-4,5-dihydro-3H-benzo[b][1,4]diazepin-2-yl)-pyridine-2-carbonitrile
MRK-8-29
4-(2-fluoro-4-methoxyphenyl)-7-(2-(2-methylpyrimidin-5-yl)ethyl)quinoline-2-carboxamide; decoglurant, 5-[2-[7-trifluoromethyl)-5-[4-(trifluoromethyl)phenyl]pyrazolo[1,5-a]pyrimidin-3-yl]ethynyl]pyridin-2-amine
NAM
negative allosteric modulator
NMR
nuclear magnetic resonance
PAM
positive allosteric modulator
P-gp
P-glycoprotein
RmGlu
rat metabotropic glutamate receptor
VU6001192
6-((cis-2,6-dimethylmorpholino)methyl)-1-(4-fluorophenyl)-4-oxo-1,4-dihydroquinoline-3-carboxamide
WT
wild type

References

  1. ↵
    1. Andrés J-I,
    2. Alcázar J,
    3. Cid JM,
    4. De Angelis M,
    5. Iturrino L,
    6. Langlois X,
    7. Lavreysen H,
    8. Trabanco AA,
    9. Celen S, and
    10. Bormans G
    (2012) Synthesis, evaluation, and radiolabeling of new potent positive allosteric modulators of the metabotropic glutamate receptor 2 as potential tracers for positron emission tomography imaging. J Med Chem 55:8685–8699.
    OpenUrlPubMed
  2. ↵
    1. Benneyworth MA,
    2. Xiang Z,
    3. Smith RL,
    4. Garcia EE,
    5. Conn PJ, and
    6. Sanders-Bush E
    (2007) A selective positive allosteric modulator of metabotropic glutamate receptor subtype 2 blocks a hallucinogenic drug model of psychosis. Mol Pharmacol 72:477–484.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Berizzi AE,
    2. Gentry PR,
    3. Rueda P,
    4. Den Hoedt S,
    5. Sexton PM,
    6. Langmead CJ, and
    7. Christopoulos A
    (2016) Molecular mechanisms of action of M5 muscarinic acetylcholine receptor allosteric modulators. Mol Pharmacol 90:427–436.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Bonnefous C,
    2. Vernier JM,
    3. Hutchinson JH,
    4. Gardner MF,
    5. Cramer M,
    6. James JK,
    7. Rowe BA,
    8. Daggett LP,
    9. Schaffhauser H, and
    10. Kamenecka TM
    (2005) Biphenyl-indanones: allosteric potentiators of the metabotropic glutamate subtype 2 receptor. Bioorg Med Chem Lett 15:4354–4358.
    OpenUrlCrossRefPubMed
  5. ↵
    1. Bradley SJ,
    2. Langmead CJ,
    3. Watson JM, and
    4. Challiss RA
    (2011) Quantitative analysis reveals multiple mechanisms of allosteric modulation of the mGlu5 receptor in rat astroglia. Mol Pharmacol 79:874–885.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Bubser M,
    2. Bridges TM,
    3. Dencker D,
    4. Gould RW,
    5. Grannan M,
    6. Noetzel MJ,
    7. Lamsal A,
    8. Niswender CM,
    9. Daniels JS,
    10. Poslusney MS, et al.
    (2014) Selective activation of M4 muscarinic acetylcholine receptors reverses MK-801-induced behavioral impairments and enhances associative learning in rodents. ACS Chem Neurosci 5:920–942.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Campo B,
    2. Kalinichev M,
    3. Lambeng N,
    4. El Yacoubi M,
    5. Royer-Urios I,
    6. Schneider M,
    7. Legrand C,
    8. Parron D,
    9. Girard F,
    10. Bessif A, et al.
    (2011) Characterization of an mGluR2/3 negative allosteric modulator in rodent models of depression. J Neurogenet 25:152–166.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Caprioli D,
    2. Venniro M,
    3. Zeric T,
    4. Li X,
    5. Adhikary S,
    6. Madangopal R,
    7. Marchant NJ,
    8. Lucantonio F,
    9. Schoenbaum G,
    10. Bossert JM, et al.
    (2015) Effect of the novel positive allosteric modulator of metabotropic glutamate receptor 2 AZD8529 on incubation of methamphetamine craving after prolonged voluntary abstinence in a rat model. Biol Psychiatry 78:463–473.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Christopoulos A
    (2014) Advances in GPCR allostery: from function to structure. Mol Pharmacol 86:463–478.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Conn PJ,
    2. Christopoulos A, and
    3. Lindsley CW
    (2009) Allosteric modulators of GPCRs: a novel approach for the treatment of CNS disorders. Nat Rev Drug Discov 8:41–54.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Conn PJ,
    2. Lindsley CW,
    3. Meiler J, and
    4. Niswender CM
    (2014) Opportunities and challenges in the discovery of allosteric modulators of GPCRs for treating CNS disorders. Nat Rev Drug Discov 13:692–708.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Davoren JE,
    2. Lee CW,
    3. Garnsey M,
    4. Brodney MA,
    5. Cordes J,
    6. Dlugolenski K,
    7. Edgerton JR,
    8. Harris AR,
    9. Helal CJ,
    10. Jenkinson S, et al.
    (2016) Discovery of the potent and selective M1 PAM-agonist N-[(3R,4S)-3-hydroxytetrahydro-2H-pyran-4-yl]-5-methyl-4-[4-(1,3-thiazol-4-yl)benzyl]pyridine-2-carboxamide (PF-06767832): evaluation of efficacy and cholinergic side effects. J Med Chem 59:6313–6328.
    OpenUrl
  13. ↵
    1. Doornbos MLJ,
    2. Pérez-Benito L,
    3. Tresadern G,
    4. Mulder-Krieger T,
    5. Biesmans I,
    6. Trabanco AA,
    7. Cid JM,
    8. Lavreysen H,
    9. IJzerman AP, and
    10. Heitman LH
    (2016) Molecular mechanism of positive allosteric modulation of the metabotropic glutamate receptor 2 by JNJ-46281222. Br J Pharmacol 173:588–600.
    OpenUrl
  14. ↵
    1. Empfield JR,
    2. Folmer JJ,
    3. Arnold JR,
    4. Clayton J,
    5. Slassi A,
    6. Isaac M,
    7. Egle I, and
    8. Ma F
    (2007) inventors, AstraZeneca, assignee. Metabotropic glutamate receptor-potentiating γsoindolones. Patent WO2007095024 A1. 2007 Aug 23.
  15. ↵
    1. Farinha A,
    2. Lavreysen H,
    3. Peeters L,
    4. Russo B,
    5. Masure S,
    6. Trabanco AA,
    7. Cid J, and
    8. Tresadern G
    (2015) Molecular determinants of positive allosteric modulation of the human metabotropic glutamate receptor 2. Br J Pharmacol 172:2383–2396.
    OpenUrl
  16. ↵
    1. Felts AS,
    2. Rodriguez AL,
    3. Smith KA,
    4. Engers JL,
    5. Morrison RD,
    6. Byers FW,
    7. Blobaum AL,
    8. Locuson CW,
    9. Chang S,
    10. Venable DF, et al.
    (2015) Design of 4-oxo-1-aryl-1,4-dihydroquinoline-3-carboxamides as selective negative allosteric modulators of metabotropic glutamate receptor subtype 2. J Med Chem 58:9027–9040.
    OpenUrl
  17. ↵
    1. Galici R,
    2. Echemendia NG,
    3. Rodriguez AL, and
    4. Conn PJ
    (2005) A selective allosteric potentiator of metabotropic glutamate (mGlu) 2 receptors has effects similar to an orthosteric mGlu2/3 receptor agonist in mouse models predictive of antipsychotic activity. J Pharmacol Exp Ther 315:1181–1187.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Galici R,
    2. Jones CK,
    3. Hemstapat K,
    4. Nong Y,
    5. Echemendia NG,
    6. Williams LC,
    7. de Paulis T, and
    8. Conn PJ
    (2006) Biphenyl-indanone A, a positive allosteric modulator of the metabotropic glutamate receptor subtype 2, has antipsychotic- and anxiolytic-like effects in mice. J Pharmacol Exp Ther 318:173–185.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Goeldner C,
    2. Ballard TM,
    3. Knoflach F,
    4. Wichmann J,
    5. Gatti S, and
    6. Umbricht D
    (2013) Cognitive impairment in major depression and the mGlu2 receptor as a therapeutic target. Neuropharmacology 64:337–346.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Gregory KJ and
    2. Conn PJ
    (2015) Molecular insights into metabotropic glutamate receptor allosteric modulation. Mol Pharmacol 88:188–202.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Gregory KJ,
    2. Noetzel MJ,
    3. Rook JM,
    4. Vinson PN,
    5. Stauffer SR,
    6. Rodriguez AL,
    7. Emmitte KA,
    8. Zhou Y,
    9. Chun AC,
    10. Felts AS, et al.
    (2012) Investigating metabotropic glutamate receptor 5 allosteric modulator cooperativity, affinity, and agonism: enriching structure-function studies and structure-activity relationships. Mol Pharmacol 82:860–875.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Hemstapat K,
    2. Da Costa H,
    3. Nong Y,
    4. Brady AE,
    5. Luo Q,
    6. Niswender CM,
    7. Tamagnan GD, and
    8. Conn PJ
    (2007) A novel family of potent negative allosteric modulators of group II metabotropic glutamate receptors. J Pharmacol Exp Ther 322:254–264.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Hemstapat K,
    2. de Paulis T,
    3. Chen Y,
    4. Brady AE,
    5. Grover VK,
    6. Alagille D,
    7. Tamagnan GD, and
    8. Conn PJ
    (2006) A novel class of positive allosteric modulators of metabotropic glutamate receptor subtype 1 interact with a site distinct from that of negative allosteric modulators. Mol Pharmacol 70:616–626.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Justinova Z,
    2. Panlilio LV,
    3. Secci ME,
    4. Redhi GH,
    5. Schindler CW,
    6. Cross AJ,
    7. Mrzljak L,
    8. Medd A,
    9. Shaham Y, and
    10. Goldberg SR
    (2015) The novel metabotropic glutamate receptor 2 positive allosteric modulator, AZD8529, decreases nicotine self-administration and relapse in squirrel monkeys. Biol Psychiatry 78:452–462.
    OpenUrlPubMed
  25. ↵
    1. Kenakin T
    (2013) Analytical pharmacology and allosterism: the importance of quantifying drug parameters in drug discovery. Drug Discov Today Technol 10:e229–e235.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Kenakin T
    (2014) What is pharmacological ‘affinity’? Relevance to biased agonism and antagonism. Trends Pharmacol Sci 35:434–441.
    OpenUrlCrossRef
  27. ↵
    1. Kenakin T
    (2016) The mass action equation in pharmacology. Br J Clin Pharmacol 81:41–51.
    OpenUrl
  28. ↵
    1. Kenakin T
    (2017) The quantitative characterization of functional allosteric effects. Curr Protoc Pharmacol 76:9.22.1–9.22.10.
    OpenUrl
  29. ↵
    1. Lavreysen H,
    2. Janssen C,
    3. Bischoff F,
    4. Langlois X,
    5. Leysen JE, and
    6. Lesage ASJ
    (2003) [3H]R214127: a novel high-affinity radioligand for the mGlu1 receptor reveals a common binding site shared by multiple allosteric antagonists. Mol Pharmacol 63:1082–1093.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Lavreysen H,
    2. Langlois X,
    3. Ahnaou A,
    4. Drinkenburg W,
    5. te Riele P,
    6. Biesmans I,
    7. Van der Linden I,
    8. Peeters L,
    9. Megens A,
    10. Wintmolders C, et al.
    (2013) Pharmacological characterization of JNJ-40068782, a new potent, selective, and systemically active positive allosteric modulator of the mGlu2 receptor and its radioligand [3H]JNJ-40068782. J Pharmacol Exp Ther 346:514–527.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Leach K,
    2. Sexton PM, and
    3. Christopoulos A
    (2007) Allosteric GPCR modulators: taking advantage of permissive receptor pharmacology. Trends Pharmacol Sci 28:382–389.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Leurquin-Sterk G,
    2. Celen S,
    3. Van Laere K,
    4. Koole M,
    5. Bormans G,
    6. Langlois X,
    7. Van Hecken A,
    8. Te Riele P,
    9. Alcázar J,
    10. Verbruggen A, et al.
    (2017) What we observe in vivo is not always what we see in vitro: development and validation of 11C-JNJ-42491293, a novel radioligand for mGluR2. J Nucl Med 58:110–116.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Levinthal C,
    2. Barkdull L,
    3. Jacobson P,
    4. Storjohann L,
    5. Van Wagenen BC,
    6. Stormann TM, and
    7. Hammerland LG
    (2009) Modulation of group III metabotropic glutamate receptors by hydrogen ions. Pharmacology 83:88–94.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Limbird LE
    (2005) Cell Surface Receptors: A Short Course on Theory and Methods, 3rd ed, Springer Science+Business Media, Inc., New York.
  35. ↵
    1. Litschig S,
    2. Gasparini F,
    3. Rueegg D,
    4. Stoehr N,
    5. Flor PJ,
    6. Vranesic I,
    7. Prézeau L,
    8. Pin JP,
    9. Thomsen C, and
    10. Kuhn R
    (1999) CPCCOEt, a noncompetitive metabotropic glutamate receptor 1 antagonist, inhibits receptor signaling without affecting glutamate binding. Mol Pharmacol 55:453–461.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Lundström L,
    2. Bissantz C,
    3. Beck J,
    4. Wettstein JG,
    5. Woltering TJ,
    6. Wichmann J, and
    7. Gatti S
    (2011) Structural determinants of allosteric antagonism at metabotropic glutamate receptor 2: mechanistic studies with new potent negative allosteric modulators. Br J Pharmacol 164:521–537.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Lundström L,
    2. Kuhn B,
    3. Beck J,
    4. Borroni E,
    5. Wettstein JG,
    6. Woltering TJ, and
    7. Gatti S
    (2009) Mutagenesis and molecular modeling of the orthosteric binding site of the mGlu2 receptor determining interactions of the group II receptor antagonist (3)H-HYDIA. ChemMedChem 4:1086–1094.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Nickols HH and
    2. Conn PJ
    (2014) Development of allosteric modulators of GPCRs for treatment of CNS disorders. Neurobiol Dis 61:55–71.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Niswender CM and
    2. Conn PJ
    (2010) Metabotropic glutamate receptors: physiology, pharmacology, and disease. Annu Rev Pharmacol Toxicol 50:295–322.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Niswender CM,
    2. Johnson KA,
    3. Luo Q,
    4. Ayala JE,
    5. Kim C,
    6. Conn PJ, and
    7. Weaver CD
    (2008) A novel assay of Gi/o-linked G protein-coupled receptor coupling to potassium channels provides new insights into the pharmacology of the group III metabotropic glutamate receptors. Mol Pharmacol 73:1213–1224.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Noetzel MJ,
    2. Rook JM,
    3. Vinson PN,
    4. Cho HP,
    5. Days E,
    6. Zhou Y,
    7. Rodriguez AL,
    8. Lavreysen H,
    9. Stauffer SR,
    10. Niswender CM, et al.
    (2012) Functional impact of allosteric agonist activity of selective positive allosteric modulators of metabotropic glutamate receptor subtype 5 in regulating central nervous system function. Mol Pharmacol 81:120–133.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. O’Brien JA,
    2. Lemaire W,
    3. Wittmann M,
    4. Jacobson MA,
    5. Ha SN,
    6. Wisnoski DD,
    7. Lindsley CW,
    8. Schaffhauser HJ,
    9. Rowe B,
    10. Sur C, et al.
    (2004) A novel selective allosteric modulator potentiates the activity of native metabotropic glutamate receptor subtype 5 in rat forebrain. J Pharmacol Exp Ther 309:568–577.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Poutiainen P,
    2. Kil K-E,
    3. Zhang Z,
    4. Kuruppu D,
    5. Tannous B, and
    6. Brownell A-L
    (2015) Co-operative binding assay for the characterization of mGlu4 allosteric modulators. Neuropharmacology 97:142–148.
    OpenUrl
  44. ↵
    1. Rook JM,
    2. Abe M,
    3. Cho HP,
    4. Nance KD,
    5. Luscombe VB,
    6. Adams JJ,
    7. Dickerson JW,
    8. Remke DH,
    9. Garcia-Barrantes PM,
    10. Engers DW, et al.
    (2017) Diverse effects on M1 signaling and adverse effect liability within a series of M1 Ago-PAMs. ACS Chem Neurosci 8:866–883.
    OpenUrl
  45. ↵
    1. Rook JM,
    2. Noetzel MJ,
    3. Pouliot WA,
    4. Bridges TM,
    5. Vinson PN,
    6. Cho HP,
    7. Zhou Y,
    8. Gogliotti RD,
    9. Manka JT,
    10. Gregory KJ, et al.
    (2013) Unique signaling profiles of positive allosteric modulators of metabotropic glutamate receptor subtype 5 determine differences in in vivo activity. Biol Psychiatry 73:501–509.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Rook JM,
    2. Tantawy MN,
    3. Ansari MS,
    4. Felts AS,
    5. Stauffer SR,
    6. Emmitte KA,
    7. Kessler RM,
    8. Niswender CM,
    9. Daniels JS,
    10. Jones CK, et al.
    (2015) Relationship between in vivo receptor occupancy and efficacy of metabotropic glutamate receptor subtype 5 allosteric modulators with different in vitro binding profiles. Neuropsychopharmacology 40:755–765.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Rovira X,
    2. Malhaire F,
    3. Scholler P,
    4. Rodrigo J,
    5. Gonzalez-Bulnes P,
    6. Llebaria A,
    7. Pin JP,
    8. Giraldo J, and
    9. Goudet C
    (2015) Overlapping binding sites drive allosteric agonism and positive cooperativity in type 4 metabotropic glutamate receptors. FASEB J 29:116–130.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Rowe BA,
    2. Schaffhauser H,
    3. Morales S,
    4. Lubbers LS,
    5. Bonnefous C,
    6. Kamenecka TM,
    7. McQuiston J, and
    8. Daggett LP
    (2008) Transposition of three amino acids transforms the human metabotropic glutamate receptor (mGluR)-3-positive allosteric modulation site to mGluR2, and additional characterization of the mGluR2-positive allosteric modulation site. J Pharmacol Exp Ther 326:240–251.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Schaffhauser H,
    2. Rowe BA,
    3. Morales S,
    4. Chavez-Noriega LE,
    5. Yin R,
    6. Jachec C,
    7. Rao SP,
    8. Bain G,
    9. Pinkerton AB,
    10. Vernier JM, et al.
    (2003) Pharmacological characterization and identification of amino acids involved in the positive modulation of metabotropic glutamate receptor subtype 2. Mol Pharmacol 64:798–810.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Smith DL,
    2. Davoren JE,
    3. Edgerton JR,
    4. Lazzaro JT,
    5. Lee CW,
    6. Neal S,
    7. Zhang L, and
    8. Grimwood S
    (2016) Characterization of a novel M1 muscarinic acetylcholine receptor positive allosteric modulator radioligand, [3H]PT-1284. Mol Pharmacol 90:177–187.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Walker AG,
    2. Wenthur CJ,
    3. Xiang Z,
    4. Rook JM,
    5. Emmitte KA,
    6. Niswender CM,
    7. Lindsley CW, and
    8. Conn PJ
    (2015) Metabotropic glutamate receptor 3 activation is required for long-term depression in medial prefrontal cortex and fear extinction. Proc Natl Acad Sci USA 112:1196–1201.
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Yin S,
    2. Noetzel MJ,
    3. Johnson KA,
    4. Zamorano R,
    5. Jalan-Sakrikar N,
    6. Gregory KJ,
    7. Conn PJ, and
    8. Niswender CM
    (2014) Selective actions of novel allosteric modulators reveal functional heteromers of metabotropic glutamate receptors in the CNS. J Neurosci 34:79–94.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Zhang R and
    2. Kavana M
    (2015) Quantitative analysis of receptor allosterism and its implication for drug discovery. Expert Opin Drug Discov 10:763–780.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleArticle

mGlu2 Allosteric Modulator Pharmacology and Binding Sites

Daniel E. O’Brien, Douglas M. Shaw, Hyekyung P. Cho, Alan J. Cross, Steven S. Wesolowski, Andrew S. Felts, Jonas Bergare, Charles S. Elmore, Craig W. Lindsley, Colleen M. Niswender and P. Jeffrey Conn
Molecular Pharmacology May 1, 2018, 93 (5) 526-540; DOI: https://doi.org/10.1124/mol.117.110114
Reddit logo Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement