Abstract
Selective activation of the M1 muscarinic acetylcholine receptor (mAChR) via a positive allosteric modulator (PAM) is a new approach for the treatment of the cognitive impairments associated with schizophrenia and Alzheimer’s disease. Herein, we describe the characterization of an M1 PAM radioligand, 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-(methyl-t3)pyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one ([3H]PT-1284), as a tool for characterizing the M1 allosteric binding site, as well as profiling novel M1 PAMs. 8-((1S,2S)-2-Hydroxycyclohexyl)-5-((6-methylpyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one (PT-1284 (1)) was shown to potentiate acetylcholine (ACh) in an M1 fluorometric imaging plate reader (FLIPR) functional assay (EC50, 36 nM) and carbachol in a hippocampal slice electrophysiology assay (EC50, 165 nM). PT-1284 (1) also reduced the concentration of ACh required to inhibit [3H]N-methylscopolamine ([3H]NMS) binding to M1, left-shifting the ACh Ki approximately 19-fold at 10 μM. Saturation analysis of a human M1 mAChR stable cell line showed that [3H]PT-1284 bound to M1 mAChR in the presence of 1 mM ACh with Kd, 4.23 nM, and saturable binding capacity (Bmax), 6.38 pmol/mg protein. M1 selective PAMs were shown to inhibit [3H]PT-1284 binding in a concentration-responsive manner, whereas M1 allosteric and orthosteric agonists showed weak affinity (>30 μM). A strong positive correlation (R2 = 0.86) was found to exist between affinity values generated for nineteen M1 PAMs in the [3H]PT-1284 binding assay and the EC50 values of these ligands in a FLIPR functional potentiation assay. These data indicate that there is a strong positive correlation between M1 PAM binding affinity and functional activity, and that [3H]PT-1284 can serve as a tool for pharmacological investigation of M1 mAChR PAMs.
Introduction
Muscarinic acetylcholine receptors (mAChRs) are members of the superfamily of G protein-coupled receptors (GPCR). Five mammalian subtypes have been identified and are referred to as M1–M5, with distribution in the central nervous system as well as the periphery. mAChR subtypes M1, M3, and M5 couple to Gq/11 proteins, which results in IP3 production and subsequent calcium mobilization, whereas subtypes M2 and M4 couple to Gi/o proteins, thereby inhibiting adenylyl cyclase activity (Bonner et al., 1987; Caulfield 1993; Caulfield and Birdsall, 1998). The M1 subtype is highly expressed in the hippocampus, striatum, and cerebral cortex (Wall et al., 1991; Levey 1996; Porter et al., 2002) and activation of these receptors in regions known for memory and cognitive function are expected to bestow a procognitive effect (Caulfield et al., 1983; Hagan et al., 1987; Messer et al., 1990).
The orthosteric agonist xanomeline achieved proof of concept in human clinical trials for improvements in positive, negative, and cognitive symptoms associated with schizophrenia (Shekhar et al., 2008), as well as improvements in cognitive function associated with Alzheimer’s disease (Bodick et al., 1997). Although reported as an M1/M4 mAChR-preferring agonist, xanomeline caused peripherally mediated adverse effects such as nausea, vomiting, salivation, and gastrointestinal distress, which led to a high dropout rate in these clinical trials. Alleviating these adverse effects and preserving biologic efficacy by developing subtype-selective orthosteric agonists has proven to be challenging (Messer, 2002) owing to complete homology of the muscarinic receptors at the orthosteric acetylcholine (ACh) site (Kruse et al., 2013).
A new approach for achieving subtype selectivity is the development of allosteric modulators that bind at sites distinct from the orthosteric binding site (Kruse et al., 2013). Unlike the orthosteric binding site, allosteric binding sites display heterogeneity across mAChR subtypes, providing opportunities for achieving desired subtype selectivity (Ma et al., 2009; Kruse et al., 2013). Therefore, by targeting allosteric binding sites, selective positive allosteric modulators (PAMs) could be identified to enhance both the affinity and intrinsic activity of the endogenous agonist, ACh, at the M1 mAChR without activation of other mAChR subtypes. This profile could potentially lead to therapeutics with improved safety and toleration profiles, compared with nonselective mAChR activators. In 2009, BQCA, a highly selective M1 PAM, was disclosed by Merck (Kenilworth, NJ; Ma et al., 2009). As a follow-on to BQCA, the second generation M1 PAM PQCA (Kuduk et al., 2011) was shown to improve cognition in animal models designed to measure recognition memory, spatial working memory, and executive function (Uslaner et al., 2013; Lange et al., 2015; Puri et al., 2015; Vardigan et al., 2015).
Availability of a potent and selective radioligand that binds to the M1 allosteric site will facilitate the understanding of M1 PAM pharmacology and the characterization of M1 selective PAMs (Trankle et al., 1996). Herein, we describe our efforts in the identification and in vitro pharmacology profiling of an M1-selective PAM radioligand lead, 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-methylpyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one [PT-1284 (1)], derived from a chemotype described by Merck (Kuduk et al., 2012), in human M1 (hM1) stable cell line assays and a native hippocampal brain slice assay, as well as the evaluation of the binding characteristics of [3H]PT-1284.
Materials and Methods
All chemicals and reagents were purchased from Sigma-Aldrich (St. Louis, MO) unless otherwise specified. All test compounds were synthesized at Pfizer, Inc. (Groton, CT) unless otherwise noted. Chinese hamster ovary (CHO) cell lines stably expressing hM1 ([3H]NMS saturable binding capacity (Bmax) = 3069 fmol/mg protein) and human M3 ([3H]NMS Bmax = 976.88 fmol/mg protein) were obtained from Wyeth Research (Princeton, NJ). Human embryonic kidney 293 cell lines stably expressing human M2 ([3H]NMS Bmax = 139.80 fmol/mg protein) and human M4 ([3H]NMS Bmax = 167.59 fmol/mg protein) were purchased from Promega (Madison, WI). [3H]NMS (85.5 Ci/mmol; 1 mCi/ml), Whatman GF/B filter plates, MicroScint-20, and Ultima Gold MV were purchased from PerkinElmer (Waltham, MA). Whatman GF/B filter sheets were purchased from Brandel (Gaithersburg, MD). F12 nutrient media and penicillin/streptomycin was purchased from Gibco/Thermo Fisher Scientific (Grand Island, NY). Fluo-8-AM dye was purchased from Molecular Probes (Grand Island, NY). [3H]PT-1284 (79.5 Ci/mmol; 1 mCi/ml) was custom synthesized and purchased from PerkinElmer. The Pierce BCA protein assay kit was purchased from Thermo Fisher Scientific.
Animals used were Sprague Dawley male rats aged 7–8 weeks purchased from Charles River Laboratories (Wilmington, MA). All procedures performed on animals in this study were in accordance with established guidelines and regulations, and were reviewed and approved by the Pfizer Institutional Animal Care and Use Committee. Pfizer animal care facilities that supported this work are fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International.
PT-1284 (1) Synthesis
All solvents and reagents were obtained from commercial sources and were used as received. All reactions were monitored by thin layer chromatography (TLC plates F254; Merck) or ultra high-performance liquid chromatography–mass spectrometry analysis (Waters Acquity, ESCi Multi-Mode Ionization Source). Proton nuclear magnetic resonance (1H NMR) and carbon-13 nuclear magnetic resonance (13C NMR) spectra were obtained using deuterated solvent on a Varian 400 MHz instrument (LabX, Midland, ON, Canada). All 1H NMR shifts are reported in δ units (ppm) relative to the signals for chloroform (7.27 ppm) and methanol (3.31 ppm). All 13C shifts are reported in δ units (ppm) relative to the signals for chloroform (77.0 ppm) and methanol (49.1 ppm) with 1H-decoupled observation. All coupling constants (J values) are reported in hertz. NMR abbreviations are as follows: br, broadened; s, singlet; d, doublet; t, triplet; q, quartet; p, pentuplet; m, multiplet; dd, doublet of doublets; ddd, doublet of doublet of doublets.
Compound 12.
6-Chloropyridin-3-yl)methyl)zinc(II) chloride (12 ml, 6.0 mmol, 0.5 M in tetrahydrofuran) was added to a solution of compound 11 (1.8 g, 5.0 mmol), which was prepared according to previously reported procedures (Kuduk et al., 2012), with palladium tris-tert-butylphosphine (128 mg, 0.25 mmol). The resulting mixture was stirred at 60°C under a nitrogen atmosphere for 18 hours. The mixture was cooled to room temperature and concentrated under reduced pressure. The crude material was purified by flash chromatography through silica using 5% methanol in dichloromethane to afford a brown solid, which was washed with methanol (10 ml) to afford compound 12 as a white solid. 1HNMR(400 MHz, CDCl3), δ 8.97 (d, J=2.76 Hz, 1 H), 8.33 (d, J=2.26 Hz, 1 H), 8.28 (d, J=7.28 Hz, 1 H), 7.80 (s, 1 H), 7.50 (dd, J=8.66, 4.14 Hz, 1 H), 7.34 (dd, J=8.28, 2.51 Hz, 1 H), 7.20 (d, J=8.28 Hz, 1 H), 4.83–5.05 (m, 2 H), 4.44 (s, 2 H), 4.14–4.26 (m, 1 H), 3.70–3.83 (m, 1 H), 3.50 (d, J=4.77 Hz, 1 H), 2.39 (d, J=6.53 Hz, 1 H), 2.22 (d, J=11.54 Hz, 1 H), 2.01 (d, J=12.30 Hz, 1 H), 1.86 (d, J=12.30 Hz, 2 H), 1.66–1.76 (m, 1 H), 1.32–1.56 (m, 2 H).
Compound 13.
Bis(tributyltin; 1.1 g, 1.84 mmol) was added to a solution of compound 12 [150 mg, 0.37 mmol) with tetrakis (triphenylphosphine) palladium (0); 43 mg, 0.037 mmol] in toluene (1 ml). The resulting mixture was heated at 110°C for 18 hours. The reaction mixture was cooled to room temperature, filtered to remove solids, and concentrated under reduced pressure. The crude material was purified by flash chromatography through a finely ground mixture of silica and potassium carbonate (9:1 by weight) using a gradient of 0% to 10% methanol in dichloromethane to afford compound 13 as a clear oil. 1HNMR(400 MHz, CDCl3) δ 8.96 (d, J=1.56 Hz, 1 H), 8.70 (s, 1 H), 8.36 (d, J=7.81 Hz, 1 H), 7.82 (s, 1 H), 7.68 (dd, J=12.10, 7.02 Hz, 2 H), 7.53–7.59 (m, 1 H), 5.31 (s, 2 H), 4.84–5.03 (m, 2 H), 4.44 (s, 1 H), 4.13–4.25 (m, 1 H), 3.71–3.82 (m, 1 H), 2.37 (d, J=6.24 Hz, 1 H), 2.22 (d, J=12.88 Hz, 1 H), 2.01 (d, J=12.88 Hz, 1 H), 1.86 (d, J=11.32 Hz, 2 H), 1.25–1.71 (m, 18 H), 1.01–1.19 (m, 6 H), 0.86 (t, J = 7.41 Hz, 6 H).
PT-1284 (1).
Iodomethane (177 mg, 0.08 mmol) was added to a solution of compound 13 [80 mg, 0.12 mmol) with tris(dibenzylideneacetone)palladium (0); 6 mg, 0.006 mmol], tri(o-tolyl)phosphine (4 mg, 0.012 mmol), and cesium fluoride (55 mg, 0.36 mmol) in dimethylformamide (DMF) (1 ml). The resulting mixture was stirred at room temperature for 45 minutes. The mixture was filtered and concentrated under reduced pressure. The crude material was purified by HPLC [column: Atlantis dC18 5-μm, 4.6 × 50 mm; retention time: 1.7 minutes; gradient: 95% H2O/5% acetonitrile (CH3CN) linear to 5% H2O/95% CH3CN over 4 minutes, hold at 5% H2O/95% CH3CN for an additional 1 minute (flow: 2.0 ml/min)] to afford PT-1284 (1) as a white solid. 1H NMR (400 MHz, methanol-d4) δ 8.98 (d, J=3.01 Hz, 1 H), 8.65 (d, J=8.53 Hz, 1 H), 8.35 (s, 1 H), 7.76 (s, 1 H), 7.66 (dd, J=8.78, 4.27 Hz, 1 H), 7.58 (dd, J=8.03, 2.01 Hz, 1 H), 7.25 (d, J=8.03 Hz, 1 H), 4.91–5.05 (m, 2 H), 4.60 (s, 3 H), 4.03–4.14 (m, 1 H), 3.83–3.92 (m, 1 H), 2.50 (s, 3 H), 2.18–2.12 (m 1 H), 1.71–1.97 (m, 4 H), 1.54 – 1.41 (m, 3 H).
Membrane Preparation
Stably transfected CHO cells overexpressing the human M1 mAChR subtype (M1 WYE clone 200-101, Wyeth Research) were grown adherently in M1 Dulbecco’s Modified Eagle Medium growth medium containing 10% fetal bovine serum (FBS), 1% nonessential amino acid, 1% penicillin/streptomycin, and 500 μg/ml genetecin. The cells were harvested by gently scraping in a physiologic phosphate-buffered saline solution. The collected cells were homogenized on ice using a Dounce glass/Teflon homogenizer. The homogenate was centrifuged at 1000g for 10 minutes and the pelleted fraction was collected and stored at –80°C in 50 mM Tris (pH 7.4) at a protein concentration of approximately 2 mg protein/ml, determined using the Pierce BCA protein assay kit. Rat, nonhuman primate, and dog brain membranes were also prepared as described earlier.
Fluorometric Imaging Plate Reader Functional Potentiation Assay
CHO cells expressing the hM1, hM3, and hM5 mAChR (HD Bioscience, Shanghai, China) were seeded at a density of 10K cells per well in black-wall, clear-bottomed 384-well plates in F12 nutrient media supplemented with 10% FBS and 1% penicillin/streptomycin. The cells were grown overnight at 37°C in the presence of 5% CO2. The cell media was subsequently removed and replaced with loading solution, containing 2 μM Fluo-8-AM dye, 2 mM probenecid, 1× acid red 1, in Hanks’ balanced salt solution, and the plate was incubated for 1 hour at 37°C in the dark. For PAM potentiation measurements, the cells were preincubated with test compound for 10 minutes before being stimulated with an EC20 concentration of ACh (the actual concentration was adjusted for each experiment, but was found to be between 1 and 3 nM). Ca2+ modulation was measured using a FLIPR Tetra fluorometric imaging plate reader (Molecular Devices, Sunnyvale, CA).
cAMP Assay
Human embryonic kidney 293 cells expressing hM2 and hM4 mAChR (cells expressed with Promega GloSensor cAMP technology) were seeded and cultured at a density of 25K cells per well in white 384-well poly-d-lysine coated plates in Dulbecco’s Modified Eagle Medium media supplemented with 10% FBS, 1% penicillin streptomycin, 500 μg/ml genetecin, 200 μg/ml hygromycin B, and 1% Glutamax for 24 hours at 37°C with 5% CO2. The culture media was replaced with equilibration media (88% CO2-independent media, 10% FBS, and 2% GloSensor cAMP reagent stock) for 2 hours at room temperature away from light. Following the 2-hour incubation, a mixture was added containing ACh at its EC20 and isoproterenol at its EC80 (concentrations were adjusted for each experiment) to the cell plate, which was then incubated for 10 minutes at room temperature. The plate was then read by Envision (PerkinElmer).
Hippocampal Slice Electrophysiology
Acute hippocampal slices were prepared from 7–10 week old rats whose brains were removed and immersed in ice-cold cutting solution (206 mM sucrose, 3 mM KCl, 1.25 mM NaH2PO4, 7 mM MgCl2, 26 mM NaHCO3, 10 mM d-glucose, 0.5 mM CaCl2, 1 mM l-ascorbate, 1 mM sodium pyruvate, bubbled continuously with 95% O2/5% CO2). Coronal slices (300 μm), containing the dorsal hippocampus were cut with a vibratome (VT1000S or 1200S; Leica, Buffalo Grove, IL) and allowed to recover in artificial cerebrospinal fluid (ACSF; 124 mM NaCl2, 3 mM KCl, 1.25 mM NaH2PO4, 1.3 mM MgCl2, 26 mM NaHCO3, 10 mM d-glucose, 2 mM CaCl2, 1 mM l-ascorbate, 1 mM sodium pyruvate, bubbled continuously with 95% O2/5% CO2) at 32°C for a recovery period of at least 1 hour. After recovery, slices were transferred individually to MED-P515A 64-electrode arrays (AutoMate Scientific, Inc., Berkeley, CA) and positioned with the CA1 stratum pyramidale directly over one row of eight electrodes and perfused continuously with recirculating ACSF warmed to 32°C. Electrophysiological signals were high-pass filtered at 0.1 Hz and digitized at 20 kHz using MED64 Software Mobius QT (WitWerx, Inc., Tustin, CA). Extracellular action potentials were detected offline using custom-written Matlab scripts. CA1 neuron spiking activity was measured as the multiunit firing rate for each electrode contacting the stratum pyramidale. Compound activity was assessed alone (“agonist mode”) or in the presence of an EC20 concentration (100 nM) of carbachol (“PAM mode”).
[3H]N-Methylscopolamine Binding Assay
[3H]N-methylscopolamine ([3H]NMS) displacement studies were conducted using 96-well microtiter plates in a total volume of 250 μl containing 0.2 nM [3H]NMS with increasing concentrations of the orthosteric ligand ACh in the absence or presence of PT-1284 (1) (0.1, 0.32, 1, 3.2, or 10 μM). The binding assay was initiated by the addition of 30 μg of CHO hM1 membranes in binding buffer (50 mM Tris, 2 mM MgCl2, pH 7.4) and incubated at room temperature for 60 minutes. Nonspecific binding was defined with 10 μM atropine sulfate. The reaction was terminated by rapid vacuum filtration through presoaked (0.5% polyethylenimine) Whatman GF/B filter plates using the Packard FilterMate Harvester 196 (PerkinElmer). The filters were washed with iced-cold 50 mM Tris, pH 7.4, and allowed to dry prior to adding MicroScint-20 scintillation cocktail. Filter-bound radioactivity was measured using the TopCount NXT scintillation counter (PerkinElmer).
[3H]PT-1284 Radiolabeling
8-((1S,2S)-2-Hydroxycyclohexyl)-5-((6-(tributylstannyl)pyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one (compound 13; MW 662.4, 60 mg, 0.09 mmol, yellow glue-like residue), tri(dibenzylideneacetone)-dipalladium (0; 5 mg, 0.00546 mmol), tri-(o-tolyl)phosphine (2.76 mg, 0.009 mM), and cesium fluoride (50 mg, 0.329 mM) were weighed into a 1-ml microflask and chilled to ice-cold temperature. DMF (0.5 ml) was added and immediately followed by introduction of [3H]-methyl iodide in DMF (1000 mCi at 85 Ci/mmol, 0.0116 mM, 1 ml toluene). The reaction was stirred at room temperature for 2 hours. Unreacted [3H]-methyl iodide was removed under vacuum. The residue was filtered and concentrated under reduced pressure. The dark-green residue was analyzed by radio-high-performance liquid chromatography (HPLC) analysis. Approximately 90% of the radioactivity coeluted with the reference standard. Purification to a radiochemical purity of 97.8% was achieved using a Luna C-18 column, 5-μm, 250 × 10 mm (Phenomenex P/N 00G-4252-NO, S/N 529990-1). Fractions were collected, analyzed for radiochemical purity, pooled, rotary-evaporated to dryness, and dissolved in ethanol to 1 mCi/ml. The radiochemical purity and the identity were measured by analytical HPLC (Phenomenex Luna C-18 column, 5-μm, 250 × 4.6 mm) at a flow rate of 1 ml/min and ultraviolet at 254 nm, using 85% of 1% triethylammonium acetate (pH 4) and 15–40% CH3CN and a 20-minute linear gradient and then 10 minutes at 100% CH3CN. Identity was confirmed by coelution with the reference standard. Specific activity was calculated by mass spectral analysis to be 79.5 Ci/mmol.
Saturation Binding Assay
[3H]PT-1284 saturation studies were conducted using 96-well microtiter plates in a total volume of 250 μl containing various concentrations of [3H]PT-1284 (0.4-50 nM) in the presence of ACh (800 nM or 1 mM). The binding assay was initiated by the addition of either 100 μg of CHO hM1 mAChR membranes or brain membranes in binding buffer (50 mM Tris, 2 mM MgCl2, pH 7.4) and incubated at room temperature for 60 minutes. Nonspecific binding was defined with 10 μM compound 9 (Kuduk et al., 2012). The reaction was terminated by rapid vacuum filtration through presoaked (0.5% polyethylenimine) Whatman GF/B filter sheets using the Brandel ML 96-well Harvester (Brandel, Gaithersburg, MD). The filters were washed with ice-cold 50 mM Tris, pH 7.4, and allowed to dry prior to addition of Ultima Gold MV scintillation cocktail. Filter-bound radioactivity was measured using the PerkinElmer Tri-Carb scintillation counter.
Displacement Binding Assay
[3H]PT-1284 displacement studies were conducted in the presence of ACh (800 nM and 1 mM). A total volume of 250 μl containing either 30 or 5 nM [3H]PT-1284 (corresponding to studies run with 800 nM ACh or 1 mM ACh, respectively) with increasing concentrations of compound added to 96-well microtiter plates. The binding assay was initiated by the addition of approximately 100 μg of CHO hM1 membranes in binding buffer (50 mM Tris, 2 mM MgCl2, pH 7.4). Nonspecific binding was defined with 10 μM compound 9. The assay was incubated at room temperature for 60 minutes. The reaction was terminated by rapid vacuum filtration through presoaked (0.5% polyethylenimine) Whatman GF/B filter plates. Filters were washed with ice-cold 50 mM Tris, pH 7.5, and allowed to dry prior to addition of MicroScint-20 scintillation cocktail. Radioactivity was measured using the PerkinElmer TopCount NXT scintillation counter. Reported Ki values are all mean ± S.E.M.
Autoradiographic Study of [3H]PT-1284
Male Sprague Dawley rat brains were rapidly removed and immediately placed on dry ice to freeze and then stored at –80°C. The brains were mounted, and 20-μm cryostat-cut (Zeiss, Thornwood, NY) coronal sections were transferred onto gelatin-coated slides and stored at -80°C. Prior to the assay, the sections were thawed in a 37°C incubator. Sections were then incubated for 30 minutes at room temperature in assay buffer (50 mM Tris, 2 mM MgCl2, pH 7.4) with 5 nM [3H]PT-1284 in the presence of 1 mM ACh. Nonspecific binding was defined with 10 μM compound 9. Following the incubation period, the sections were washed (2 × 2 minutes) in ice-cold assay buffer followed by a 30-second wash in ice-cold deionized water. The sections were dried rapidly under a cool air stream and further dried in a desiccant-containing vacuum. Labeled sections were exposed to Kodak BioMax MR Film for 47 days before being scanned on a Bio-Rad 800 densitometer. Sections were later analyzed using Quantity One 1-D Software (Bio-Rad, Hercules, CA).
Data Analysis
All data were analyzed using GraphPad Prism 6.03 (GraphPad Software, La Jolla, CA). Displacement binding curves were best fit to a one-site four-parameter model, where Ki values were determined from the Cheng-Prusoff relationship such that Ki = IC50/1 + [ligand]/Kd, where [ligand] is the concentration of the free radioligand used in the assay and Kd is the dissociation constant of the radioligand for the receptor (Cheng and Prusoff, 1973). Saturation curves were fitted using the one-site binding hyperbola model, where the Ki and Bmax were derived from the formula Y = (Bmax * X)/(Kd + X), where Y is the specific binding and X is the concentration of the ligand. To account for ligand depletion in [3H]NMS saturation binding analyses for the human M1, M3, and M4 mAChR cell lines and in [3H]PT-1284 saturation binding analysis for the human M1 mAChR cell line, a global-fit model with parameter-sharing was used (Motulsky and Christopoulos, 2004). Also, owing to ligand depletion, [3H]NMS displacement binding was best fit to a one-site heterologous binding with depletion model (Swillens, 1995). Fluorometric imaging plate reader (FLIPR) EC50 data were fitted to the compound percent effect using four-parameter logistic fit models.
Results
For the development of M1-selective PAM radioligands, we used a set of central nervous system (CNS) positron emission tomography (PET) ligand design-and-selection criteria that our group had previously published, considering the similar property requirements between a PET ligand and a [3H] radioligand (Zhang et al., 2013). Specifically, we targeted leads that possessed potent M1 PAM activity and high selectivity against other mAChR subtypes, and, importantly, demonstrated low nonspecific binding levels. Toward this end, a database of approximately 280 patent compounds with reported M1 PAM potency values were compiled, and their CNS PET multiparameter optimization (MPO) values were calculated. Upon filtering by potency (M1 PAM EC50 < 20 nM) and physicochemical property criteria (CNS PET MPO > 3), 18 leads were identified (Fig. 1). Notably, eight out of these 18 leads shared the tricyclic lactam core exemplified by compound 10 (Kuduk et al., 2012), and only three had a synthetic handle for tritiation, among which compound 10 emerged as the most promising lead, with a reported M1 PAM EC50 of 18 nM and good physicochemical properties [CNS PET MPO desirability of 3.15].
Analysis of known literature M1 PAMs comparing CNS PET multiparameter optimization (MPO) with reported potency data. A database of 280 compounds were compiled from the literature and their potencies were correlated with CNS PET MPO. Initial leads were limited to compounds with CNS PET MPO > 3 and M1 FLIPR calcium mobilization EC50 < 20 nM.
One of the leading causes of radioligand failure is high nonspecific binding. Fraction unbound in brain (Fu,b; Di et al., 2011) has been shown to be a useful predictor for nonspecific binding and, specifically, we targeted Fu,b > 0.05 to minimize this risk (Zhang et al., 2013). Toward this end, compound 10 was synthesized and its Fu,b was measured to monitor nonspecific binding risk. The relatively low Fu,b value of compound 10 (0.025) prompted us to synthesize and profile additional close-in analogs of compound 10 aiming for higher Fu,b, thus lowering the risk of high nonspecific binding, at the same time maintaining favorable M1 PAM potency and selectivity. As shown in Fig. 2, incorporation of a heteroaryl-N at the naphthalene core led to a quinoline analog, PT-1284 (1), that had comparable M1 PAM potency [PAM EC50 of 36 nM (n = 14) versus 18 nM for compound 10] with reduced lipophilicity (ELogD 1.83 versus 2.90 for compound 10; Lombardo et al., 2001), which led to much improved Fu,b (0.10 versus 0.025 for compound 10). The synthesis of PT-1284 (1) can be seen in Fig. 3. It is worth noting that PT-1284 (1) was expected to have low brain permeability (Feng et al., 2008) as it is a P-glycoprotein substrate (MDR BA/AB = 8.8; Johnson et al., 2014). [3H]PT-1284 showed no specific binding in vivo when dosed up to 100 μCi. However, it is fit-for-purpose to enable in-vitro or ex-vivo tissue radioligand binding assays for which brain permeability is not required and, subsequently can be used to facilitate the development of novel M1 PAM in vivo radiotracers and PET ligands.
Selection of PT-1284 (1). Incorporating a heteroaryl-N at the naphthalene core led to a quinoline analog PT-1284 (1) with comparable M1 PAM potency, much reduced lipophilicity, and a 4-fold higher Fu,b than compound 10.
Synthesis of PT-1284 (1). Pd(t-Bu)3, palladium tris-tert-butylphosphine; (Bu3Sn)2, Bis(tributyltin); Pd(PPh3)4, tetrakis (triphenylphosphine) palladium; Mel, Iodomethane; Pd2dba3, tris (-dibenzylideneacetone) palladium; CsF, cesium floride.
FLIPR Functional Potentiation Activity.
The effect of PT-1284 (1) on functional activity was assessed for selectivity among all mAChRs. PT-1284 (1) increased M1 calcium levels in a concentration-dependent manner whether applied alone (agonist mode) or in the presence of a low concentration (EC20) of the nonselective agonist ACh (PAM mode). In the presence of ACh at its EC20, signaling endpoints were generated using the FLIPR assay for M1, M3, and M5, whereas the cAMP accumulation assay was used to generate data for M2 and M4 (Fig. 4). In the agonist mode of our FLIPR assay, PT-1284 (1) displayed transient partial agonist activity that ranged from 125 nM to >13 μM with variable efficacy. Though this activity was noted, its relevance was uncertain and we were unable to quantify it. In response to an EC20 challenge with M1 endogenous agonist ACh, free-base PT-1284 (1) displayed robust potentiation of ACh with an EC50 of 36 nM ± 7 (mean ± S.E.M., n = 14). There were no functional PAM responses observed for PT-1284 (1) at the M2–M5 mAChR subtypes when tested up to 10 μM.
PAM functional concentration-response curves for hM1-M5 mAChR. Measurement of functional responses for PT-1284 (1) in the presence of an EC20 concentration of ACh. Data are expressed as percentage of functional effect, and the endpoints were generated using FLIPR for M1, M3, and M5 and cAMP assays for M2 and M4. Open circles for M1 data were not fitted to the curve; this “inverted U” effect occurred at concentrations at which direct agonist activity was observed using agonist mode assay conditions. Data shown are representative of two (M2–M5) or fourteen (M1) separate experiments and expressed as the mean ± S.E.M.
Hippocampal Slice Electrophysiology.
M1 receptors are expressed in hippocampal CA1 pyramidal neurons and interneurons, where they increase cell excitability (Langmead et al., 2008, Dasari and Gulledge, 2011). In acute hippocampal slices, this can be measured as an increase in the spontaneous-action-potential-firing rate of CA1 neurons (Langmead et al., 2008). Since acetylcholinesterase is present in rat brain slices and has the ability to catalyze the breakdown of the mAChR endogenous ligand ACh (Lockhart et al., 2001), carbachol was used as the cholinergic agonist to define PAM mode for this assay. In acute slices of dorsal hippocampus from adult rats, PT-1284 (1) increased spontaneous CA1 neuron firing in a concentration-dependent manner whether applied alone (agonist mode) or in the presence of a low concentration (∼EC10–EC30) of the nonselective cholinergic agonist carbachol (PAM mode). However, the potency and efficacy of the compound were both enhanced by the presence of carbachol, consistent with an agonist-PAM pharmacological profile. When the concentration-response data from each slice were fitted individually, the agonist EC50 was 1.1 ± 0.32 μM (mean ± S.E.M., n = 8 slices) and the maximal agonist effect was 35 ± 6 Hz, whereas the PAM EC50 was 165 ± 44 nM (mean ± S.E.M., n = 9 slices) and the maximal PAM effect was 96 ± 20 Hz (Fig. 5).
Hippocampal slice electrophysiology. PT-1284 (1) increased spontaneous firing of CA1 neurons in rat hippocampal slices. When the compound was applied alone (agonist mode), CA1 firing rates increased in a concentration-dependent manner (EC50 = 1.1 ± 0.32 μM, max = 35 ± 6 Hz; n = 8). When the compound was applied in the continuous presence of an EC20 concentration of carbachol (PAM mode), PT-1284 (1) increased spontaneous firing at much lower concentrations (EC50 = 165 ± 44 nM, max = 96 ± 20 Hz, n = 9). Data are shown as mean ± S.E.M.
[3H]NMS Displacement.
The effects of PT-1284 (1) on the affinity of ACh at the [3H]NMS orthosteric binding site were examined using CHO cells stably transfected with the hM1 mAChR subtype ([3H]NMS: Kd = 0.506 nM; Bmax = 3069 fmol/mg protein). Although PT-1284 (1) was shown to compete with the nonselective muscarinic receptor antagonist [3H]NMS with weak affinity (Ki = 7.84 ± 1.42 μM; mean ± S.E.M., n = 3; Fig. 6A), there were no changes to the maximal response of ACh, indicating that PT-1284 (1) did not compete with ACh for binding to M1 mAChR (Fig. 6B). PT-1284 (1) reduced the concentration of ACh required to inhibit [3H]NMS binding to M1 mAChR, left-shifting the EC50 of ACh by approximately 19-fold at 10 μM (Fig. 6B).
[3H]NMS binding to hM1 cell membranes. (A) PT-1284 (1) competes at the [3H]NMS orthosteric binding site. Inhibition of [3H]NMS binding to hM1 mAChR membranes by PT-1284 (1). (B) Inhibition of [3H]NMS binding to hM1 mAChR membranes by ACh in the absence or presence of various concentrations of PT-1284 (1). Data are expressed as %inhibition and analyzed for Ki values using the one-site heterologous with depletion model (Swillens 1995). Data shown are representative curves from three separate experiments performed in duplicate and expressed as the mean ± S.E.M.
Optimization of [3H]PT-1284 Binding Protocol.
Preliminary studies to ascertain assay conditions for [3H]PT-1284 binding to CHO cells stably expressing the hM1 mAChR were performed at radiolabel concentrations below PT-1284 (1)’s EC50 of 36 nM. Binding experiments were initially performed in the absence of the orthosteric agonist ACh and no specific binding was detected (data not shown). A concentration-response curve for ACh in this assay showed significantly increased specific binding with a maximal effect at 1 mM (Fig. 7), indicating a positive cooperation between the orthosteric and allosteric binding sites.
ACh concentration-response curve in the [3H]PT-1284 binding assay. [3H]PT-1284 (5 nM) binding to hM1 mAChR membranes by ACh. Data shown are representative of four separate experiments performed in duplicate and expressed as counts per minute (mean ± S.E.M.).
[3H]PT-1284 (5 nM)-specific binding levels were determined over a protein concentration range from 6 to 200 μg per well. Specific binding increased with increasing concentrations of membranes and plateaued between 50 and 200 μg per well. Kinetic studies showed the rate of association of [3H]PT-1284 (5 nM), which reached its equilibrium within 10 minutes, and remained stable for up to 120 minutes at room temperature (data not shown). Subsequent experiments used a radioligand concentration of 5 nM, membrane concentration of 100 μg per well, and incubations were performed at room temperature for 60 minutes.
Using these established conditions, in the presence of 1 mM ACh, the percent specific binding of [3H]PT-1284 to hM1 membranes was greater than 90%, with binding values of 49,523 ± 1659 dpm for total binding and 2067 ± 389 dpm for nonspecific binding determinations (mean ± S.E.M. of three experiments).
[3H]PT-1284 Saturation Binding.
[3H]PT-1284 saturation experiments were performed in the presence of 800 nM or 1 mM ACh (Table 1; Fig. 8, A–D). Using CHO cells stably expressing hM1 mAChR, [3H]PT-1284 bound with a higher affinity (Kd) when in the presence of 1 mM ACh (4.23 ± 0.17 nM, mean ± S.E.M., n = 3) and the Bmax increased nearly 3-fold (6383 ± 346 fmol/mg protein, mean ± S.E.M., n = 3) compared with saturation binding in the presence of 800 nM ACh (29 ± 3.15 nM; 2479 ± 415 fmol/mg protein, mean ± S.E.M., n = 3). [3H]PT-1284 bound to rat frontal cortex and hippocampus with similar affinity and saturable binding capacity [2.65 ± 0.34 nM and 415 ± 31 fmol/mg protein; 2.55 ± 0.57 nM and 410 ± 51 fmol/mg protein, respectively (mean ± S.E.M., n = 3)]. [3H]PT-1284 bound to nonhuman primate (NHP) and dog cortex with Kd and Bmax values of 2.28 ± 0.19 nM; 244 ± 36 fmol/mg protein and 2.88 ± 0.69 nM; 118.5 ± 7.72 fmol/mg protein, respectively (mean ± S.E.M., n = 3). Scatchard plots for all saturation experiments could be fitted by a straight line, indicating that [3H]PT-1284 bound to a single site in each case (Fig. 8, A–D, inset graphs). M1 mAChR specific binding was not detected in rat or nonhuman primate cerebellum when tested up to a [3H]PT-1284 concentration of 50 nM (data not shown).
[3H]PT-1284 saturation binding to hM1 mAChR, rat, nonhuman primate, and dog brain membranes
[3H]PT-1284 saturation binding values were obtained in the presence of 1 mM ACh unless otherwise noted. Data shown are the result of three separate experiments performed in triplicate unless otherwise noted and expressed as the mean ± S.E.M.
[3H]PT-1284 saturation curves and Scatchard analyses (inset graphs) for (A) hM1 cell membranes (total binding = 46,204 dpm and nonspecific binding = 2714 dpm at approximately 3 nM [3H]PT-1284), (B) rat frontal cortex membranes (total binding = 4329 dpm and nonspecific binding = 2559 dpm at approximately 3 nM [3H]PT-1284), (C) nonhuman primate cortex membranes (total binding = 1594 dpm and nonspecific binding = 735 dpm at approximately 3 nM [3H]PT-1284), (D) dog cortex membranes (total binding = 706 dpm and nonspecific binding = 360 dpm at approximately 3 nM [3H]PT-1284). Saturation binding was measured in the presence of 1 mM ACh. Data shown are a representative of three separate experiments performed in triplicate and expressed as the mean ± S.E.M.
[3H]PT-1284 Displacement Binding.
[3H]PT-1284 displacement studies were performed with a series of compounds previously determined to be PAMs, allosteric agonists, or orthosteric agonists at the M1 mAChR (data not shown). The rank order of potency was consistent with structural recognition of the [3H]PT-1284 binding site, in that PAMs showed higher affinity than compounds known to be allosteric or orthosteric agonists (Table 2; Fig. 9). M1 mAChR PAM affinities were 10-fold weaker when the assay was run in the presence of 800 nM ACh. For example, the affinity of compound 3 for the [3H]PT-1284 binding site decreased from 9.7 ± 0.23 nM to 84 ± 15 nM (mean ± S.E.M., n = 3; data not shown), respectively. As shown in Fig. 10, for a cohort of 19 M1 PAMs previously reported by our group (Davoren et al., 2016), a positive correlation (R2 = 0.86) was found between the EC50 values obtained using the M1 mAChR FLIPR PAM assay and the binding pKi values determined with the [3H]PT-1284 binding displacement assay.
[3H]PT-1284 displacement binding to hM1 mAChR membranes by positive allosteric modulators and orthosteric or allosteric binding ligands
[3H]PT-1284 displacement binding values (pKi) were obtained in the presence of 1 mM ACh. Data shown are the result of three separate experiments performed in duplicate and expressed as the mean pKi ± S.D.
Structures of compounds used in [3H]PT-1284 displacement binding assay. Structures include mAChR allosteric agonists, orthosteric agonists, and M1 PAMs.
Correlation between hM1 functional and [3H]PT-1284 displacement binding data. Positive allosteric modulators show a correlation between FLIPR calcium mobilization EC50 values and [3H]PT-1284 displacement binding. The dashed line indicates unity (R2 = 1.00). Solid line indicates best linear fit of the data generated (R2 = 0.86). Data shown are the mean values of at least three separate experiments.
Autoradiographic Analysis for [3H]PT-1284 Binding to Rat Brain.
Autoradiographic distribution of [3H]PT-1284 binding to coronal sections of rat brain is shown in Fig. 11. In representative sections shown in (A) and (B), the highest distribution of M1 PAM mAChR [3H]PT-1284 binding can be seen in the cortex (CTX), caudate (CPu), nucleus accumbens (Abc), and the hippocampus (Hip). [3H]PT-1284 binding was displaced by 10 μM compound 9 (Fig. 11C).
Autoradiographic analysis of representative rat brain coronal sections using [3H]PT-1284. (A) Cortex, caudate, and nucleus accumbens and (B) hippocampus represent autoradiograms for total specific binding. (C) Inhibition of specific binding by 10 μM compound 9. All binding was done in the presence of 1 mM ACh. (D) Density scale in the range of 3.7–2000 μCi/g protein.
Discussion
Although recent advancements have been made in the development of mAChR allosteric modulators to improve subtype selectivity, little has been reported on the development of radiolabeled modulators to study the actual allosteric binding pocket. With the characterization of the M2/M4 mAChR allosteric modulator radioligand [3H]LY2119620, great strides were taken to further the understanding of mAChR allosteric binding sites (Schober et al., 2014). As with other GPCR PAM radioligands such as the mGluR4 PAM, and [3H]PAM2 (Le Poul et al., 2012), these authors showed positive cooperativity between the orthosteric and allosteric binding sites and further showed that selectivity between M2 and M4 mAChR subtypes was dependent on the allosteric ligand used. This is the first report on the development and characterization of a novel M1-selective mAChR PAM radioligand that possesses high M1 PAM potency and broad spectrum selectivity, and demonstrated high M1-specific binding in vitro.
Herein, we presented evidence from several in vitro assays to show the pharmacological profile of PT-1284 (1) as an agonist-PAM. PT-1284 (1) was shown to have PAM characteristics in a FLIPR assay by potentiating M1 mAChR calcium mobilization when in the presence of an EC20 concentration of ACh. PT-1284 (1) also showed higher potency and induced a greater response in hippocampal CA1 neurons when applied in the presence of carbachol. In the past, the only way to assess allosteric binding activity of a compound such as BQCA was indirectly through the orthosteric binding site. Radioligand binding studies utilizing [3H]NMS, the nonselective orthosteric mAChR ligand, showed a positive cooperativity between PAM compounds and orthosteric agonists ACh, carbachol, or oxotremorine-M for M1, M2, and M4 mAChR subtypes (Ma et al., 2009; Canals et al., 2012; Abdul-Ridha et al., 2014; Schober et al., 2014). In line with literature values, the affinity of ACh in our [3H]NMS binding assay was 400 μM. In the current study, we also determined that PT-1284 (1) was positively cooperative with ACh for the M1 mAChR showing a 19-fold leftward shift of ACh’s activity at the [3H]NMS binding site.
It has been suggested that GPCR binding levels are directly proportional to the number of receptors in an active state, reflected by the coupling of the GPCR to the G protein (Christopoulos and El-Fakahany, 1999; Schober et al., 2014). [3H]PT-1284 binds to the allosteric pocket of the M1 mAChR and enhances G protein coupling and the affinity of ACh for this receptor. The [3H]NMS binding density for a CHO cell line overexpressing mAChR hM1 (3 pmol/mg protein) is in line with literature values ranging from 2 to 4 pmol/mg protein. At concentrations up to 50 nM, [3H]PT-1284 bound to a single, saturable site on human M1 receptors, with saturation binding levels increasing with higher concentrations of ACh present. This, in turn, increased the affinity of [3H]PT-1284 for the receptor, revealing subnanomolar affinity at maximal concentrations of ACh. In this study, we showed that cooperativity between our PAM ligand and the orthosteric ligand ACh may be improved by increasing the number of G protein-bound receptors. In spite of the fact that differences in ACh levels have been reported across species (Fujii et al., 1995), [3H]PT-1284 bound with similar affinity to native rat, dog, and NHP tissues when enhanced with the addition of ACh, which is important in that cross-species translation can be used to predict occupancy of a clinical compound in humans from rat in vivo binding or NHP PET receptor occupancy studies (Shaffer et al., 2014). Autoradiographic localization studies showed distribution in the cortex, caudate, nucleus accumbens, and hippocampus of the rat brain, which correlates with the observed binding capacities likewise found in our saturation binding studies. Also, with [3H]PT-1284 displacement studies, we observed greater affinity of M1 PAM compounds in the presence of higher concentrations of ACh. The rank order of affinities was consistent with structural recognition of the PT-1284 (1) binding site, in that PAMs showed higher affinity compared with compounds known to be allosteric or orthosteric agonists. Taken together, these data showed cooperation between the ACh orthosteric binding site and the PT-1284 (1) binding site. To demonstrate [3H]PT-1284 as a tool in characterizing novel M1 mAChR PAMs, a direct correlation of binding pKi and functional EC50 values (R2 = 0.86) was observed for a cohort of compounds.
In summary, we have demonstrated PT-1284 (1) to be a M1-selective PAM agonist that acts as a PAM when [3H]PT-1284 binding is performed using 5 nM. With the characterization of the radioligand and a direct correlation with our calcium mobilization assay, we have shown [3H]PT-1284 to be a tool in profiling future M1 mAChR PAMs. These findings could enable improvements in the treatment of symptoms of schizophrenia and cognitive deficits associated with Alzheimer’s disease.
Authorship Contributions
Participated in research design: Smith, Davoren, Lee, Zhang, Grimwood.
Conducted experiments: Smith, Edgerton, Lee, Neal.
Performed data analysis: Smith, Edgerton, Lazzaro, Neal.
Wrote or contributed to the writing of the manuscript: Smith, Davoren, Edgerton, Lee, Zhang, Grimwood.
Footnotes
- Received April 14, 2016.
- Accepted June 30, 2016.
Abbreviations
- ACh
- acetylcholine
- Bmax
- saturable binding capacity
- BQCA
- 1-(4-methoxybenzyl)-4-oxo-1,4-dihydroquinoline-3-carboxylic acid, benzyl quinolone carboxylic acid
- CHO
- Chinese hamster ovary
- CNS
- central nervous system
- compound 9
- 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-methylpyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-h]quinolin-7-one
- compound 10
- (2-((1S,2S)-2-hydroxycyclohexyl)-5-((6-methylpyridin-3-yl)methyl)-1,2-dihydro-3H-benzo[e]isoindol-3-one
- compound 12
- 5-((6-chloropyridin-3-yl)methyl)-8-((1S,2S)-2-hydroxycyclohexyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one
- compound 13
- 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-(tributylstannyl)pyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-hour]quinolin-7-one
- DMF
- dimethylformamide
- FBS
- fetal bovine serum
- FLIPR
- fluorometric imaging plate reader
- Fu,b
- fraction unbound in brain
- GPCR
- G protein-coupled receptors
- hM1
- human M1
- HPLC
- high-performance liquid chromatography
- [3H]PT-1284
- 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-(methyl-t3)pyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-h]quinolin-7-one
- mAChR
- muscarinic acetylcholine receptor
- MPO
- multiparameter optimization
- NMS
- N-methylscopolamine
- 1H NMR
- proton nuclear magnetic resonance
- PAM
- positive allosteric modulator
- PET
- positron emission tomography
- PQCA
- 1-{[4-cyano-4-(2-pyridinyl)-1-piperidinyl]methyl}-4-oxo-4H-quinolizine-3-carboxylic acid
- PT-1284 (1)
- 8-((1S,2S)-2-hydroxycyclohexyl)-5-((6-methylpyridin-3-yl)methyl)-8,9-dihydro-7H-pyrrolo[3,4-h]quinolin-7-one
- 13C NMR
- carbon-13 nuclear magnetic resonance
- Copyright © 2016 by The American Society for Pharmacology and Experimental Therapeutics
References
In this issue
[3H]PT-1284, a Novel M1 mAChR PAM Radioligand
