Synthesis and Evaluation of Two Novel Radioligands for Neuroimaging of GABA Transporter-1

Introduction: Gamma-aminobutyric acid (GABA), as the major inhibitory neurotransmitter, plays an essential role in multiple physiological functions, and its concentrations in the brain are modulated by receptors and transporters. Alterations in the distribution and expression of the GABA transporters (GATs) have been implicated in a variety of human diseases, such as epilepsy and schizophrenia. Availability of PET radioligands will facilitate the elucidation of GATs in diseases and the development of effective therapeutic agents. We have previously reported the preliminary evaluation of brain-penetrant radioligands for the major GABA transporter in the brain, GAT-1 (J. Nucl. Med. 2021, 62, (Supplement 1), 6). Here we describe our continued effort in the development of novel radioligands for imaging GAT-1 in the brain.

Methods: Two radioligand candidates suitable for ¹⁸F-labeling were selected from a library of >100 tiagabine derivatives based on a comprehensive evaluation of their binding affinity to GAT-1, lipophilicity (LogD), passive membrane permeability (PAMPA assay), and efflux ratio by human P-glycoprotein (P-gp). The novel radioligands [¹⁸F]GATT-31 and [¹⁸F]GATT-39 were synthesized by Cu(OTf)₂-catalyzed ¹⁸F-fluorination of their respective aryltrimethyltin precursors, followed by removal of the ester protecting group. PET imaging in rhesus macaques were performed on the FOCUS-220 scanner to evaluate their pharmacokinetic and binding properties. Binding specificity was assessed in blocking studies with the GAT-1 inhibitor tiagabine (0.5 mg/kg). Blood samples were drawn for metabolite analysis and construction of plasma input function, which were used in quantitative kinetic analysis with the one-tissue compartment (1TC) model to calculate the regional volume of distribution (V_T, mL/cm³). The occupancy plots were used to estimate target occupation by tiagabine in the blocking studies, and the non-displaceable volume of distribution (V_ND). Regional binding potential (BP_ND) was then calculated using V_ND, where BP_ND=(V_T/V_ND)-1.

Results: [¹⁸F]GATT-31 and [¹⁸F]GATT-39 were synthesized in 2 steps with >98% radiochemical purity. Molar activity at the end of synthesis was 86 ± 34 GBq/µmol (n=2) for [¹⁸F]-GATT-31 and 172 ± 82 GBq/µmol (n=4) for [¹⁸F]GATT-39. Metabolism was slow for both radioligands, with the parent fraction accounting for >75% of plasma activity even at the end of the 180 min scan. Uptake in the brain was slow, and increased steadily with time, reaching SUV of ~3.0 for [¹⁸F]GATT-31 and ~2.0 for [¹⁸F]GATT-39 at 180 min (Figure 1). Distribution was heterogeneous across brain regions, with higher uptake in the cortex, moderate in cerebellum, and lower in thalamus and hippocampus. Regional time-activity curves were well fitted with the 1TC model to derive regional V_T values (Table 1). Specific binding of both radiotracers was demonstrated in blocking scans with tiagabine (0.5 mg/kg), which reduced regional uptake and V_T across brain regions, with V_ND of 0.47 and 0.51 mL/cm³, respectively, for [¹⁸F]GATT-31 and [¹⁸F]GATT-39. Regional BP_ND values calculated with V_ND were 25% higher on average for [¹⁸F]GATT-39 than [¹⁸F]GATT-31.

Conclusions: We have successfully synthesized two radioligands for GAT-1 and evaluated their pharmacokinetic and binding properties in rhesus monkeys. Both radiotracers appeared to be metabolically stable and demonstrated high levels of specific binding in non-human primate brain. Comprehensive evaluation of these two radiotracers, together with [¹⁸F]GATT-34 and [¹⁸F]GATT-44 reported previously, are underway to select the most appropriate one for translation to human studies.

• Download figure
• Open in new tab
• Download powerpoint