Development of 11C-labeled mGluR2negative allosteric modulators for PET imaging

Objectives: Metabotropic glutamate receptor subtype 2 (mGluR2) is involved in the pathology of several CNS disorders and neurodegenerative diseases, including Alzheimer’s disease.[1] We have identified two series of small molecule library of mGluR2 negative allosteric modulators (NAMs) with quinoline carboxamides and naphthyridine carboxamides as core scaffold, respectively.[2] Our objective was to synthesize a focused library of mGluR2 NAMs, evaluate potency and selectivity using functional assays, and to radiolabel and perform preliminary evaluation of one representative mGluR2 NAM by PET. Methods: We have designed two efficient synthetic routes to access quinoline carboxamides and naphthyridine carboxamides, respectively. For quinoline carboxamides, oxidation of quinoline and substitution of consequence at α-position were first employed to get 3 in 74% and 60% yields, respectively. Bromination with NBS generated bromide 4 in 60% yield, followed by succinimide substitution to give key intermediate 5 in 80% yield. Suzuki cross coupling with (2-fluoro-4-methoxyphenyl)boronic acid gave nitrile (6a) in 65% yield, which was hydrolyzed to afford the final standard (7a) in 55% yield. The analogous procedure was applied to obtain phenolic precursor (7b) in 46% yield. To get naphthyridine carboxamides NAMs, acid mediate cyclization was employed to get naphthyridine ester (9) in 64% yields, followed by reductive hydronation and N-alkylation to afford 10 in 40% yield over two steps. Suzuki cross-coupling with (2-fluoro-4-methoxyphenyl)boronic acid and (2-fluoro-4-hytroxyphenyl)boronic acid yielded 11a and 11b in 25% and 32% yields, respectively, followed by amination to obtain final standard (12a) and phenolic precursor (12b) in 72% and 63% yields, respectively. In vitro activity and selectivity towards mGluR2 was evaluated by a thallium flux assay in human embryonic kidney 293 (HEK) cells expressing heteromeric G-protein coupled inwardly rectifying potassium (GIRK) channels. In addition to [11C]QCA,[3] we preformed 11C-radiolabeling of naphthyridine carboxamide 12a from precursor 12b (1 mg) with [11C]CH3I in the presence of NaOH in DMF (0.2 mL) at 80°C for 5 min. Results: A library of quinoline carboxamides and naphthyridine carboxamides was obtained in overall yields from 7.6% to 5.9%, respectively. In vitro screening of candidate NAMs showed potency toward mGluR2 with EC50 values from 18 nM to 113 nM. No other subtype selectivity (>100 fold selectivity) towards mGluR1, mGluR3-mGluR8 was detected. Radiolabeling with [11C]CH3I yielded [11C]7a and [11C]12a in average 6% and 5% isolated radiochemical yields, respectively (ca. 740-800 MBq) with greater than 40 GBq/µmol specific activity within 45 min. No signs of radiolysis was observed up to 90 min after formulation (10% ethanol in saline). Preliminary PET imaging studies of [11C]12a in rat brain showed peak uptake of ca. 2 SUV with rapid washout to ca. 0.6 SUV at 30 min post injection. The regional distribution was heterogeneous with signal levels in the decreasing order of cortex, hippocampus, thalamus, cerebellum and pons. Conclusions: We have successfully prepared two series of potent and selective NAMs for mGluR2 and performed preliminary radiolabeling and PET imaging studies. Further characterization including blocking studies with structurally-diverse mGluR2NAMs, ex vivo whole body distribution and radiometabolite analysis will be performed in order to evaluate and develop potent mGluR2 receptor PET tracers. References:[1] Neuropharmacol. 2004, 46, 907-917; [2] WO2013066736; [3] ACS Chem Neurosci. 2017, 8, 1937-1948.SNMMI 2018 Annual Meetin