Our previous work has demonstrated that 100-Hz electroacupuncture (EA) or 100-Hz transcutaneous electrical nerve stimulation (TENS) was very effective in ameliorating the morphine withdrawal syndrome in rats and humans. The mechanism was obscure. (1) Rats were made dependent on morphine by repeated morphine injections (5-140 mg/kg, s.c., twice a day) for eight days. They were then given 100-Hz EA for 30 min 24 h after the last injection of morphine. A marked increase in tail flick latency (TFL) was observed. This effect of 100-Hz EA could be blocked by naloxone (NX) at 20 mg/kg, but not at 1 mg/kg, suggesting that 100-Hz EA-induced analgesia observed in morphine-dependent rats is mediated by kappa-opioid receptors. (2) A significant decrease of the concentration of dynorphin A (1-17) immunoreactivity (-ir) was observed in the spinal perfusate in morphine-dependent rats, that could be brought back to normal level by 100-Hz EA. (3) 100-Hz EA was very effective in suppressing NX-precipitated morphine withdrawal syndrome. This effect of EA could be prevented by intrathecal administration of nor-BNI (2.5 micrograms/20 microliters), a kappa-opioid receptor antagonist, or dynorphin A (1-13) antibodies (25 micrograms/20 microliters) administered 10 min prior to EA. In conclusion, while the steady-state spinal dynorphin release is low in morphine-dependent rats, it can be activated by 100-Hz EA stimulation, which may be responsible for eliciting an analgesic effect and ameliorating morphine withdrawal syndrome, most probably via interacting with kappa-opioid receptor at spinal level.