Purpose: Use of mGluR5 receptor radiotracers to determine whether an in vitro binding assay is able to predict how good a radiotracer is likely to be in imaging receptor in the central nervous system (CNS) via positron emission tomography (PET).
Procedures: Saturation and equilibrium competition studies in rat and rhesus membranes were used to determine receptor concentrations and tracer affinities. In addition, specific binding of metabotropic receptor subtype 5 (mGluR5) radioligands in rhesus and rat brain sections was determined using a "no-wash protocol," and the in vivo binding signal in rats was determined using micro-PET.
Results: Affinity values were determined for a series of mGluR5 antagonists (1-5) and ranged from 0.1 to 11 nM in rat. A previously reported "no-wash protocol" was then employed to determine specific binding in tissue sections following a 20-min incubation, and the regional distribution of these mGluR5 radiotracers determined in rat brain via autoradiography. The analogs 1b, 2b, 3b, and 4b, but not 5b, displayed good signal-to-noise ratios under these conditions with high density of binding in caudate, cortex, and hippocampus and lower density in cerebellum. With this information it was predicted that 1c, 2c, 3b, and 4b would display measurable signal-to-noise ratios in vivo, and that the larger in vitro signals for 3b and 4b would translate to 3b and 4b yielding the best in vivo signals. These predictions were investigated using micro-PET imaging in rat. Compound 1c showed a rapid wash-in and rapid wash-out profile in rat brain. Compound 2c showed similar signal-to-noise ratio as 1b, but slower washout. Compounds 3b and 4b showed the best signal-to-noise ratio in vivo, while 5b did not provide a significant signal, as predicted. In vivo occupancy estimates for 2-methyl-6-(phenylethynyl)-pyridine (MPEP) following intravenous administration were determined using radiolabeled compounds 1c, 2c, and 3b; they were essentially the same and were on the order of 1 mg kg(-1) (ID(50)).
Conclusions: An in vitro screen of several mGluR5 tracers was used to rapidly predict whether radiolabeled mGluR5 analogs would be useful as PET radiotracers. Results provided an extension to previously reported data. Two of the four radiotracers with the best in vitro "no-wash" results also showed the best potential as measured noninvasively using micro-PET.