Synthesis and Preliminary Evaluations of a Triazole-Cored Antagonist as a PET Imaging Probe ([18F]N2B-0518) for GluN2B Subunit in the Brain

Abstract

GluN2B is the most studied subunit of N-methyl-d-aspartate receptors (NMDARs) and implicated in the pathologies of various central nervous system disorders and neurodegenerative diseases. As pan NMDAR antagonists often produce debilitating side effects, new approaches in drug discovery have shifted to subtype-selective NMDAR modulators, especially GluN2B-selective antagonists. While positron emission tomography (PET) studies of GluN2B-selective NMDARs in the living brain would enable target engagement in drug development and improve our understanding in the NMDAR signaling pathways between normal and disease conditions, a suitable PET ligand is yet to be identified. Herein we developed an ¹⁸F-labeled potent antagonist, 2-((1-(4-[¹⁸F]fluoro-3-methylphenyl)-1 H-1,2,3-triazol-4-yl)methoxy)-5-methoxypyrimidine ([¹⁸F]13; also called [¹⁸F]N2B-0518) as a PET tracer for imaging the GluN2B subunit. The radiofluorination of [¹⁸F]13 was efficiently achieved by our spirocyclic iodonium ylide (SCIDY) method. In in vitro autoradiography studies, [¹⁸F]13 displayed highly region-specific binding in brain sections of rat and nonhuman primate, which was in accordance with the expression of GluN2B subunit. Ex vivo biodistribution in mice revealed that [¹⁸F]13 could penetrate the blood-brain barrier with moderate brain uptake (3.60% ID/g at 2 min) and rapid washout. Altogether, this work provides a GluN2B-selective PET tracer bearing a new chemical scaffold and shows high specificity to GluN2B subunit in vitro, which may pave the way for the development of a new generation of GluN2B PET ligands.

Keywords: 18F-labeling; GluN2B subunit; PET imaging; Subtype-selective; autoradiography; spirocyclic iodonium ylide.

Figure 1.

Chemical structures of representative GluN2B-selective NMDAR PET tracers.

Figure 2.

Pharmacology studies of our GluN2B-selective NMDAR antagonists. (A) Concentration−response curves for antagonists 11−14 (0.03−1.0 μM) on human GluN1/GluN2B were plotted as the percent of the maximal response to glutamate/glycine (100 μM/100 μM) and fit by the Hill equation. (B) %Current responses to glutamate/glycine (100 μM/100 μM for NMDAR) or glutamate (100 μM for AMPAR and KAR) co-applied with compound solution (1 μM) of 11 or 13 were recorded in Xenopus oocytes expressing human GluN1/GluN2A receptors, human GluN1/GluN2B receptors, human GluN1/GluN2D receptors, rat GluN1/GluN2C receptors, rat GluA1(flip) subunit or rat GluK2(Q) subunit. The data were expressed as the percent of the maximal response to agonists. (C and D) Inhibition of triheteromeric receptors by compounds 11 (C) and 13 (D), respectively. Concentration−response curves were generated from the triheteromeric receptors including GluN1/GluN2A/GluN2A (2A/2A), GluN1/GluN2B/GluN2B (2B/2B), and GluN1/GluN2A/GluN2B (2A/2B) upon activated by glutamate/glycine (100 μM/100 μM). Data are mean ± SEM from 10–14 oocytes.

Figure 3.

Molecular docking structures of compounds 11 (purple in A) and 13 (orange in B) onto NMDAR subunit GluN1b and GluN2B interface.

Figure 4.

In vitro autoradiography results of [¹⁸F]13 (37 kBq/mL, 1.58 nM) in rat brain sections. Representative autoradiograms in rat brain sagittal sections: [¹⁸F]13 alone (A, baseline), and pre-blocked by cold compound 13 (C, self-blocking, 10 μM) and BMT-108908 (D, a NAM drug blocking, 10 μM). (B) GluN2B mRNA expression in mouse brain and the data were retrieved from mouse.brain-map.org (experiment: 69257725). (E) Quantitative analysis of baseline and blocking experiments. The value is expressed as the ratio of the radioactivity of different brain regions to that of the cerebellum (reference region). HIP, hippocampus; FCx, frontal cortex; STR, striatum; THM, thalamus; Cb, cerebellum; OLF, olfactory bulb. Asterisks indicate statistical significance: ***p <0.001, and **p <0.01 vs control.

Figure 5.

In vitro autoradiography results of [¹⁸F]13 (37 kBq/mL, 1.58 nM) in NHP brain sections. Representative autoradiograms in NHP brain sections: [¹⁸F]13 alone (A, baseline), and pre-blocked by cold compound 13 (B, self-blocking, 10 μM) and BMT-108908 (C, a NAM drug blocking, 10 μM). (D) Quantitative analysis of baseline and blocking experiments. The value is expressed as the ratio of the radioactivity of different brain regions to that of the white matter of forebrain. HIP, hippocampus; ECx, entorhinal cortex; ICx, insular cortex; CdN, caudate nucleus; CLA, claustrum; PUT, putamen; THM, thalamus; FWM, white matter of forebrain. Asterisks indicate statistical significance: ***p <0.001, and **p <0.01 vs control.

Figure 6.

(A) Ex vivo biodistribution in mice at four different time points (2, 10, 30, and 60 min) post injection of [¹⁸F]13. The results are expressed as the percentage of the injected dose per gram of wet tissue (% ID/g) with the exception that stomach uptake was expressed as % ID/organ. (B) The values of the brain, blood, and bone uptakes of [¹⁸F]13. The values for remaining organs of interest are included in Table S1 of Supporting Information.

Scheme 1.

Synthesis of GluN2B-selective NMDAR antagonists (11–14). Reagents and conditions: (a) NaNO2₂, NaN₃, HCl (6 N), 0–5 °C, 2 h, 89% yield. (b) Propargyl alcohol, DIPEA, CuI, THF, 40 °C, 2 h, 71% yield. (c) 2-Bromo-5-methoxypyrimidine, NaH, THF, 40 °C, 2–3 h, 29% yield.

Scheme 2

Synthesis of the precursors (17, 20, and 21) and radiosynthesis of [¹⁸F]11 and [¹⁸F]13. Reagents and conditions: (a) DIPEA, CuI, THF, 40 °C, 4 h, 25% yield. (b) [¹⁸F]F^-, Kryptofix₂₂₂/K₂CO₃ (7.5/1.5 mg), anhydrous DMF, 140 °C, 10 min, <3% yield (decay corrected). (c) 1) mCPBA, CHCl₃, rt, overnight; 2) SPIAd, Na₂CO₃ (10%), ethanol, pH = 10, rt, 6 h, 36% yield over two steps. (d) [¹⁸F]F^-, TEAB, anhydrous DMF, 120 °C, 10 min. (e) DIPEA, CuI, anhydrous DMF, 100 °C, 10 min, 35% and 39% yield for [¹⁸F]11 and [¹⁸F]13, respectively, over two steps (decay corrected).