Synthesis and characterization of [¹⁸F]mG2P026 as a high contrast PET imaging ligand for metabotropic glutamate receptor 2

Objectives: Metabotropic glutamate receptor 2 (mGluR2) is presynaptically expressed in neurons and astrocytes where it modulates the synaptic transmission and participates in synaptic plasticity. Activation of this receptor via positive allosteric modulators (PAMs) represents an attractive strategy in enhancing mGluR2 function. Although several PAMs showed promising results in preclinical animal models, the clinical proof of concept is still absent. The current work aims to develop a high contrast positron emission tomography (PET) imaging ligand to elucidate the mGluR2 neurobiology in vivo and promote the drug discovery on PAMs.

Methods: An array of triazolopyridines were synthesized as potential mGluR2 PAMs and their structure-activity relationship (SAR) was analyzed. The best candidate mG2P026 was radiolabeled via the Cu(I)-mediated radiofluorination of organoboranes in the TRACERLab^TM FXFN module. Characterization of [¹⁸F]mG2P026 was first conducted in Sprague-Dawley rats. The binding specificity of [¹⁸F]mG2P026 was tested by the pretreatment of three structurally distinct PAMs, i.e., mG2P001, BINA and mG2P026, before tracer injection. The following PET studies in cynomolgus monkeys investigated the binding profile of [¹⁸F]mG2P026 across different brain areas. The arterial blood was sampled to characterize the radiometabolite and provide input function for quantitative modeling. One-, two-tissue compartment models and Logan plot were investigated to generate reliable regional total volume of distribution (V_T).

Results: mG2P026 was a potent mGluR2 PAM (EC₅₀ = 12 nM) with an apparent mGluR2 agonist activity (IC₅₀ = 102 nM). Synthesis of [¹⁸F]mG2P026 was achieved with good radiochemical yield (RCY = 12 ± 4%, non-decay-corrected) and excellent chemical and radiochemical purities (> 95%) at the end of synthesis (EOS, 55 min, n > 5). In rats, [¹⁸F]mG2P026 distributed in the mGluR2-expressing regions of striatum, thalamus, cortex, hypothalamus, hippocampus and cerebellum. All the blocking experiments, however, showed enhanced PET signals, the extent of which varied by the class of injected PAMs. Specifically, the self-blocking elicited the most enhancement, which was similar under two different doses (2.0 mg/kg versus 4.0 mg/kg). The mG2P001 showed a lower enhancement than mG2P026. Interestingly, although BINA had the least enhancement, it showed a dose-dependent characteristic, where a higher dose (2 mg/kg) led to a higher increase across the brain regions of interest than those at a lower dose (1 mg/kg) except in the cortex. In the primate studies, the metabolite chromatogram revealed a moderate in vivo stability of [¹⁸F]mG2P026 with the presence of two more polar radiometabolites. The time-activity curves (TACs) were well described via the two-tissue compartment model (2T4k1v) and the Logan plot linearized well. A comparison of the PET SUV images between [¹⁸F]mG2P026 and our recently published [¹⁸F]JNJ-46356479 at the time window from 30 to 60 min further demonstrated the superior binding profile of [¹⁸F]mG2P026 for imaging mGluR2.

Conclusions: We have identified a potent mGluR2 ago-PAM mG2P026 from the SAR study of the triazolopyridines. mG2P026 can be radiolabeled via fluorine-18 in the automated synthetic module. Preclinical studies in rats demonstrated the unique feature of structurally distinct mGluR2 PAMs in affecting the radioactivity uptake of [¹⁸F]mG2P026. Further imaging studies in primates revealed the excellent contrast of [¹⁸F]mG2P026 in mapping the mGluR2 distribution in the brain. Acknowledgement: The research was supported by NIH grants [1R01EB021708, 1R01NS100164, 1S10RR023452-01 and 1S10OD025234-01]. The NIH grants [S10OD018035 and P41EB022544] supported the blood counting and metabolite analysis equipment used in the primate studies.