Preclinical Evaluation of Benzazepine-Based PET Radioligands (R)- and (S)-11C-Me-NB1 Reveals Distinct Enantiomeric Binding Patterns and a Tightrope Walk Between GluN2B- and σ1-Receptor-Targeted PET Imaging

Abstract

The study aims to investigate the performance characteristics of the enantiomers of ¹¹C-Me-NB1, a recently reported PET imaging probe that targets the GluN2B subunit of N-methyl-d-aspartate (NMDA) receptors. Methods: Reference compound Me-NB1 (inhibition constant for hGluN1/GluN2B, 5.4 nM) and the phenolic precursor were prepared via multistep synthesis. Following chiral resolution by high-performance liquid chromatography, enantiopure precursor compounds, (R)-NB1 and (S)-NB1, were labeled with ¹¹C and validated in rodents using in vitro/ex vivo autoradiography, PET experiments, and dose-response studies. To illustrate the translational relevance, (R)-¹¹C-Me-NB1 was validated in autoradiographic studies using postmortem human GluN2B-rich cortical and GluN2B-deficient cerebellar brain slices. To determine target engagement, receptor occupancy was assessed at different plasma concentrations of CP101,606, a GluN2B receptor antagonist. Results: The radiosynthesis of (R)- and (S)-¹¹C-Me-NB1 was accomplished in 42% ± 9% (decay-corrected) radiochemical yields. Molar activity ranged from 40 to 336 GBq/μmol, and an excellent radiochemical purity of greater than 99% was achieved. Although (R)-¹¹C-Me-NB1 displayed heterogeneous accumulation with high selectivity for the GluN2B-rich forebrain, (S)-¹¹C-Me-NB1 revealed a homogeneous distribution across all brain regions in rodent brain autoradiograms and predominantly exhibited σ₁-receptor binding. Similar to rodent brain, (R)-¹¹C-Me-NB1 showed in postmortem human brain tissues higher binding in the cortex than in the cerebellum. Coincubation of the GluN2B-antagonist CERC-301 (1 μM) reduced cortical but not cerebellar binding, demonstrating the specificity of (R)-¹¹C-Me-NB1 binding to the human GluN2B-containing NMDA receptor. In vivo specificity of (R)-¹¹C-Me-NB1 in the GluN2B-expressing cortex, striatum, thalamus, and hippocampus was demonstrated by PET imaging in rodents. Applying GluN2B-antagonist eliprodil, an evident dose-response behavior was observed with (R)-¹¹C-Me-NB1 but not with (S)-¹¹C-Me-NB1. Our findings further underline the tightrope walk between GluN2B- and σ₁-receptor-targeted imaging, illustrated by the entirely different receptor binding behavior of the 2 radioligand enantiomers. Conclusion: (R)-¹¹C-Me-NB1 is a highly selective and specific PET radioligand for imaging the GluN2B subunit of the NMDA receptor. The entirely different receptor binding behavior of (R)-¹¹C-Me-NB1 and (S)-¹¹C-Me-NB1 raises awareness of a delicate balance that is underlying the selective targeting of either GluN2B-carrying NMDA or σ₁-receptors.

Keywords: GluN2B; PET imaging; neurodegeneration; receptor occupancy; σ1-receptor.