In our previous work

acupuncture induced elevation of pain threshold could not be blocked by intra cerebroventricular injection of bicuculline

an antagonist f or GABA A receptors

suggesting that GABA in brain might not be involved in acup uncture analgesia

at least might not be mediated by activation of GABA A recept ors. In this paper

the effects of intra cerebroventricular and intrathecal inj e ction (icv and ith) of CGP 55845

a potent and selective antagonist against GABA B receptor

on acupuncture analgesia were further investigated by tail flick la tency tests in rats. There were 6～8 rats in each group. The results were as f ollows: 1. After icv and ith baclofen(Bac)

an agonist of GABA B receptor at doses of 0. 0 25 μg /5 μL and 0.25 μg /10 μL(+5 μL saline)

pain threshold (PT) were rai sed to 125.11±1.94% and 128.63± 0.93% of the basal control value respective ly. While subcutaneous injection of Bac

50 μg /kg was needed to produce significant analgesic effect (PT raised to 130.94±1.62%). It indicates that activation of GABA B receptor co uld produce analgesic effects mainly in the central nervous system. 2. After icv CGP 55845 at doses of 0.5

5.0

and 50 ng /5 μL

icv Bac (2.5 μg /5 μL) induced analgesic effect (PT raised to 136.24±1.96%) was blocked by 3 7.4%

77.0% and 75.1% respectively (PT raised to 130.43±1.72%

111.8±0.98% and 111.25±0.65%). It means that icv Bac induced analgesic effect could be signif icantly blocked by pretreatment of icv CGP 55845. 3. After electroacupuncture at bilateral "Ciliao"(BL 32) points (50 Hz

1～2 mA) for 10 minutes

PT were raised to 142.50±2.10 % without treatment

143.74±2.0 4% with pretreatment of saline and 142.47±1.18% with pre icv CGP 55845 at dose s of 0.5 ng/5 μL. When pretreated with icv CGP 55845 at doses of 5 and 50 ng/ 5 μL

acupuncture analgesic effects were significantly blocked by 73.7% and 71. 6% (PT raised to 111.19±1.20% and 112.09±1.12%). It indicates that GABA in brai n might be involved in acupuncture analgesia mainly mediated by activation of GA BA B receptors. 4. After electroacupuncture at bilateral "Ciliao"(BL 32) points

PT were ra ised to 138.15± 1.64% with pretreatment(ith) of saline and to 138.56±1 .21% with pretreatment(ith) of CGP 55845 5 ng/5 μL. When pretreated with ith CGP 55 845 at doses of 50 and 500 ng /5 μL

PT were raised to 119.04±1.04% and 109.08 ±1.94% respectively

indicating that acupuncture analgesia could be significant ly blocked by 52.6% and 76.7%. It means that GABA might also be involved in acup uncture analgesia mediated by activation of B receptors at spinal level.