Using conventional autoradiographic and tissue counting techniques, the experimental quantitation of in vivo kinetics of prospective or established radioligands for PET is animal and labour intensive. The present study tested the feasibility of using PET itself to quantitate the specific binding of [11C]raclopride to rat striatum and to study the effects of experimental manipulation of endogenous dopamine on binding parameters. Carbon-11-labeled raclopride was given i.v. to anaesthetised rats, positioned in a PET camera and dynamic emission scans acquired over 60 min. Time-activity curves were generated for selected regions of interest, representing striatum and cerebellum and the striatal data fitted to a compartmental model, using cerebellum as the input function, thus circumventing the need for individual metabolite-corrected plasma curves. In control rats, the binding potential (BP), defined as the ratio of the rate constants for transfer from "free to bound" and "bound to free" compartments, was of the order of 0.6. This was reduced threefold by predosing with nonradioactive raclopride. Increasing extracellular dopamine levels by predosing with d-amphetamine resulted in a significant decrease in BP whereas reducing extracellular dopamine by predosing with gamma-butyrolactone caused a significant increase. Thus, despite the limitation in spatial resolution of PET, specific binding of raclopride could be assessed from regional time-activity curves from individual rats. The system was sufficiently sensitive that changes in BP could be detected following modulation of endogenous dopamine levels, a finding of potential relevance to the interpretation of clinical PET data.