The effect of intrathecal (i.t.) and systemic (i.p. and i.v.) administration of morphine, aminophylline, dibutyryl cyclic adenosine monophosphate (DBcAMP) and dibutyryl cyclic guanosine monophosphate (DBcGMP) on motor and sensory responses of the spinal nociceptive system was studied in rats. Motor responses were assessed in the tail-flick test performed on rats with an intact spinal cord, or as flexor reflex activity elicited in the electromyogram of the tibialis anterior muscle by supramaximal electrical stimulation of the sural nerve in rats in which the spinal cord was transected at the lower thoracic level. The sensory response consisted of activity in single ascending axons of the spinal cord evoked by electrical stimulation of afferent C fibres in spinal rats. Morphine (20 micrograms i.t. or 2 mg/kg i.p.) prolonged the tail-flick latency and aminophylline (25 mg/kg i.p. or 50 micrograms i.t.) prevented the antinociceptive effect of morphine. Aminophylline alone, administered by i.t. injection, reduced the tail-flick latency in a dose-dependent way. Morphine (2 mg/kg i.v. or 10 micrograms i.t.) reduced flexor reflex activity, and this reduction was abolished by aminophylline (25 mg/kg i.v. or 50 micrograms i.t.). Morphine (2 mg/kg i.v.) depressed spontaneous and evoked activity in single ascending axons responding to stimulation of afferent C fibres. This depressant effect of morphine was not abolished by aminophylline (50 micrograms i.t.); the depression was antagonized by naloxone (10 micrograms i.t.). DBcAMP (5 to 100 ng i.t.) dose-dependently prolonged the tail-flick latency. The antinociceptive effect of DBcAMP (50 ng i.t.) was prevented by aminophylline (50 micrograms i.t.) or naloxone (5 micrograms i.t.).(ABSTRACT TRUNCATED AT 250 WORDS)