To investigate the antinociceptive effect of adenosine agonist R-phenylisopropyl-adenosine (R-PIA) given to conscious rats by intracerebroventricular (ICV) and intrathecal (IT), and identify the effect of R-PIA on minimum alveolar concentration (MAC) of halothane with pretreatment of A1 receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) or K+ channel blocker 4-aminopyridine (4-AP). Sprague-Dawley rats were implanted with 24-gauge stainless steel guide cannula using stereotaxic apparatus and ICV method, and an IT catheter (PE-10, 8.5 cm) was inserted into the lumbar subarachnoid space, while the rats were under pentobarbital anesthesia. After one week of recovery from surgery, rats were randomly assigned to one of the following protocols: MAC of halothane, or tail-flick latency. All measurements were performed after R-PIA (0.8-2.0 microg) microinjection into ICV and IT with or without pretreatment of DPCPX or 4-AP. Microinjection of adenosine agonist R-PIA in doses of 0.8-2.0 microg into ICV and IT produced a significant dose- and time-dependent antinociceptive action as reflected by increasing latency times and ICV administration of adenosine agonist R-PIA (0.8 microg) reducing halothane anesthetic requirements (by 29%). The antinociception and reducing halothane requirements effected by adenosine agonist R-PIA was abolished by DPCPX and 4-AP. ICV and IT administration of adenosine agonist R-PIA produced an antinociceptive effect in a dose-dependent manner and decreased halothane MAC with painful stimulation through activation of A1 receptor subtype, and the underlying mechanism involves K+ channel activation.