Presentation Synthesis and preclinical evaluation of 11C-labeled 1,3-dihydro-2H-benzo[d]imidazol-2-ones for development of TARP γ-8 dependent AMPAR PET tracers

Chen, Zhen  ,  Zhang, Xiaofei  ,  SHAO, TUO  ,  Shao, Yihan  ,  Tomita, Susumu  ,  Ming-Rong, Zhang  ,  Liang, Huan

Objectives: Transmembrane a-Amino-3-hydroxyl-5-methyl-4-isoxazolepropionic acid (AMPA) receptor regulatory proteins are a new family of scaffolding proteins, which regulate the trafficking, gating, protein levels and pharmacology of AMPA receptors (AMPARs). The goal of the project was to establish a focused library of TARP γ-8 dependent AMPAR antagonists amenable for radiolabeling with carbon-11 or fluorine-18, and carry out preliminary in vivo evaluation for imaging TARP γ-8 by PET. Methods: A focused library of 1,3-dihydro-2H-benzo[d]imidazol-2-ones and benzo[d]thiazol-2(3H)-ones were synthesized based on lead AMPAR antagonist JNJ-55511118.[1] We designed PdCl2(dtbpf)-catalyzed cross-coupling of aryl boric acid 1 with 4-bromobenzene-1,2-diamine 2 to afford desired [1,1'-biaryl]-3,4-diamine precursors 3-10 in moderate to excellent yields (42-86%). The cyclization of diamine 3-10 was achieved by treatment with 1,1’-carbonyldiimidazole (CDI), thus delivering urea-type candidate AMPAR inhibitors 11-18 in 52-82% yields. The carbamothioate-type candidate AMPAR inhibitors 21-23 were also obtained in 62-83% yields over two steps via protection of 6-bromobenzo[d]thiazol-2(3H)-one 19 with Boc2O followed by coupling with aryl boric acid 1. The focused library of 1,3-dihydro-2H-benzo[d]imidazol-2-ones 11-18 and benzo[d]thiazol-2(3H)-ones 21-23 was then subjected to in vitropharmacological evaluation using calcium flux assay and molecular docking studies by AutoDock. The most promising compounds were radiolabeled at the 11C-carbonyl position using [11C]COCl2. Dynamic PET imaging studies were performed in SD rats in a Siemens Inveon scanner for 60 min. Results: The urea-type standard compounds 11-18 and carbamothioate-type candidates 21-23 were obtained in 31-83% yields over two steps. The labeling precursor diamine 3 and 10 was assembled in 42% and 86% yield, respectively. Preliminary pharmacological evaluation demonstrated compounds 11, 12 and 14 as promising candidates, which possessed high potency to TARP γ-8 with IC50 values between 10-20 nM and excellent subtype selectivity between γ-8 and γ-2 dependent AMPARs. The homology binding pocket between TARP γ-8 and candidate compounds was established using SWISS-MODEL, with the best hit, pdb 5kk2, as the building template, to study the molecular interaction.[2] The overall G-factor average of the model was -0.26, indicating reasonable normality. As proof of concept, [11C]11 was radiolabeled in 13.4% decay-corrected RCY relative to starting [11C]CO2 with high radiochemical purity (>99%) and high specific activity (196.6 GBq/μmol). No radiolysis of [11C]11 was detected up to 90 min after formulation (10% ethanol in saline). Dynamic PET studies of [11C]11 exhibited reasonable brain penetration (~0.8 SUV). Conclusion: We have established a small library of TARP γ-8 dependent AMPAR antagonists, and identified several lead molecules for in vivoevaluation. The radiosynthesis of [11C]11 was achieved in excellent yield, and in vivoPET study confirmed brain permeability of [11C]1. Further radiolabeling of 12 and 14, followed by blocking studies and radiometabolite analysis will be performed to develop new TARP γ-8 dependent AMPA receptor PET tracers. References: [1] J Pharmacol Exp Ther. 2016, 357, 394-414; [2] ACS Chem Neurosci., 2017, 8, 2631-2647.
SNMMI 2018 Annual Meeting

Number of accesses :  

Other information