Microdialysis was used to sample morphine 6-glucuronide (M6G) and morphine in striatal extracellular fluid after systemic administration in awake, freely moving rats. Morphine or M6G (25-67 mumol/kg) was given subcutaneously, and blood and striatal dialysate were sampled repeatedly during 120 min. Blood samples were obtained by indwelling catheters in the inferior vena cava. Opiates in serum or brain dialysate were analyzed with high-performance liquid chromatography. The functional intactness of the blood-brain barrier was verified by the use of sodium technetate (Na99mTcO4). The fractional penetration into the brain of morphine and M6G was approximately 350- and 90-fold higher than that of Na99mTcO4, respectively, with a relative difference in the transfer of morphine and M6G of about 4. No hydrolysis of M6G to morphine was detected. Striatal dialysate-to-serum ratios of M6G did not differ after 25 or 67 mumol/kg. Serum AUC0-120 min was 10 times higher for M6G than for morphine. This reflects both a smaller volume of distribution (Vd) for M6G and a decreased rate of elimination compared with morphine. The median t1/2 from serum was 36 and 32 min for morphine and M6G, respectively. The striatal dialysate AUC0-120 min of M6G was 2.9 times greater than that of morphine after an equimolar subcutaneous dose. Dialysate tmax was delayed approximately 40 min relatie to serum tmax for both drugs, and the median t1/2 from the dialysate was 82 and 48 min for M6G and morphine, respectively. These results represent direct evidence for the penetration of M6G into the brain after systemic administration to living rats.