Our previous studies have shown that electroacupuncture (EA) at the Neiguan-Jianshi (P5-P6) acupoints inhibits sympathetic outflow and attenuates excitatory visceral cardiovascular reflexes through enkephalin- or beta-endorphin-related opioid receptors in the rostral ventrolateral medulla (rVLM). It is not known whether EA at these acupoints activates neurons containing enkephalin or beta-endorphin in the rVLM as well as in the periaqueductal gray (PAG) that are involved in EA-mediated central neural regulation of sympathetic activity. The present study evaluated activated neurons in the rVLM and PAG by detecting c-Fos immunoreactivity, and identified the relationship between c-Fos nuclei and neuronal structures containing enkephalin or beta-endorphin in these regions. To enhance the detection of cell bodies containing enkephalin or beta-endorphin, colchicine (90-100 microg/kg) was injected into the subarachnoid space in anesthetized cats 28-30 h prior to EA or the sham-operated control for EA. Following bilateral barodenervation and cervical vagotomy, EA (1-4 mA, 2 Hz, 0.5 ms) was performed at the P5-P6 acupoints (overlying median nerve; n=7) for 30 min. Identical procedures, with the exception of electrical stimulation, were carried out in five control animals. EA decreased blood pressure (BP) in four of seven cats (5-15 mm Hg) while the sham procedure for EA produced no responses. Perikarya containing enkephalin were found in the rVLM and rarely in the PAG, while no cell bodies labeled with beta-endorphin were identified in either region. Compared to animals in the control group, more c-Fos immunoreactivity, located principally in close proximity to fibers containing enkephalin or beta-endorphin, was observed in the rVLM and ventrolateral PAG (vlPAG) in EA-treated cats. Moreover, neurons double-labeled with c-Fos and enkephalin in the rVLM were significantly increased in cats following EA stimulation (P<0.05). These data indicate that EA at the P5-P6 acupoints activates neurons in the rVLM and vlPAG. These activated neurons contain enkephalin in the rVLM, and most likely interact with nerve fibers containing enkephalin or beta-endorphin in both the rVLM and vlPAG. The results from this study provide the first anatomical evidence showing that EA at the P5-P6 acupoints has the potential to influence neuronal structures (perikarya, axons and/or dendrites) containing enkephalin or beta-endorphin in specific regions of the brain stem. These neurons likely form the substrate for EA's influence on sympathoexcitatory cardiovascular reflexes.