Introduction: The mammalian brain contains abundant G protein-coupled cannabinoid CB(1) receptors that respond to Delta(9)-tetrahydrocannabinol, the active ingredient of cannabis. The availability of a positron emission tomography (PET) radioligand would facilitate studies of the addictive and medicinal properties of compounds that bind to this receptor. Among the known classes of ligands for CB(1) receptors, the pyrazoles are attractive targets for radiopharmaceutical development because they are antagonists and are generally less lipophilic than the other classes.
Methods: A convenient high-yield synthesis of N-(4-[(18)F]fluorophenyl)-5-(4-bromophenyl)-1-(2,4-dichlorophenyl)-1H-pyrazole-3-carboxamide (AM5144) was devised by coupling the appropriate pyrazole-3-carboxyl chloride compound with 4-[(18)F]fluoroaniline. The labeled precursor was synthesized from 1-[(18)F]fluoro-4-nitrobenzene in 60% radiochemical yield for 10 min using an improved procedure involving sodium borohydride reduction with cobalt chloride catalysis. The product was purified by HPLC to give a specific activity >400 mCi/micromol and a radiochemical purity >95%, and a PET study was conducted in a baboon.
Results: Although the regional uptake of AM5144 in baboon brain was consistent with binding to cannabinoid CB(1) receptors, absolute uptake at <0.003% injected radioactivity per cubic centimeter was lower than the previously reported uptake of the radioiodinated pyrazole AM281.
Conclusions: The relatively poor brain uptake of AM5144 and other pyrazole CB(1) receptor ligands is not surprising because of their high lipophilicity as compared with most brain PET radiotracers. However, for nine pyrazole compounds for which rodent data are available, brain uptake and calculated logP values are not correlated. Thus, high logP values should not preclude evaluation of radiotracers for targets such as the CB(1) receptor that may require very lipophilic ligands.