Synthesis and biological investigations of [18F]MR18445, a 5-HT3 receptor partial agonist

18F Labelled MR18445 (4-[4-(4-[18F]fluorobenzyl)piperazino]-7-methoxypyrrolo++ +[1,2-alpha] quinoxaline), a selective 5-HT3 receptor partial agonist with nanomolar affinity, was synthesized and examined as a potential radioligand for PET imaging of brain 5HT3 receptors. Radiotracer was prepared by N-alkylation of the MR18491 precursor with 4-[18F]fluorobenzyl iodide. This latter was synthesized in a three-step procedure from 4-[18F]fluorobenzaldehyde obtained by 18F-nucleophilic displacement of 4-nitrobenzaldehyde, subsequently reduced to 4-[18F]fluorobenzyl alcohol and converted into reactive 4-[18F]fluorobenzyl iodide. The reduction step was performed on a column filled with NaBH4/Al2O3 and 4-[18F]fluorobenzyl alcohol was obtained with high reproducible yield (82-93% from 4-[18F]fluorobenzaldehyde) if there were traces of water in the system. The radiosynthesis of [18F]MR18445 required approximately 120 min. Semi-preparative HPLC purification followed by formulation gave injectable solutions of [18F]MR18445 with a radiochemical purity > 99%. The overall yield of the synthesis was mainly dependent upon the first step efficiency of aromatic incorporation of 18F- and varied from 12% to 29%. All the synthetic procedure was realized on a ZYMARK robotic system. Biological in vivo studies in rats showed that uptake of [18F]MR18445 in brain was rapid and high. No evidence of radiolabeled metabolites could be observed in the brain as late as 40 min after injection, despite the rapid appearance of metabolites in the plasma. However, neither phosphorimaging autoradiographic studies in rats nor PET experiments in baboons revealed specific binding of the radiotracer in brain, suggesting [18F]MR18445 is not suitable for 5-HT3 receptors PET studies.