The effects of an intracerebroventricular (i.c.v.) administration of a non-selective full benzodiazepine receptor agonist, midazolam, and a neuroactive steroid, allopregnanolone, on picrotoxin-induced seizures and striatal dopamine metabolism, were studied in mice. It was found that acute i.c.v. injections of midazolam (ED50=38.25 nmol) and allopregnanolone (ED50=26.34 nmol) blocked picrotoxin-induced seizures to a similar extent. After repeated administration at the ED(85) doses (midazolam-56.6 nmol, allopregnanolone-94.2 nmol; once or twice daily for 5 days) tolerance developed to the anticonvulsant activity of midazolam (ED50=94.14 nmol) and allopregnanolone (ED50=186.70 nmol). Acute i.c.v. injections of midazolam and allopregnanolone (at the ED50 doses established in the model of picrotoxin seizures: 38.25 and 26.34 nmol, respectively), significantly decreased the concentration of dopamine metabolites: 3-methoxytyramine and 3,4-dihydroxyphenylacetic acid, as well as the dopamine turnover rate (homovanilic acid/dopamine ratio; by about 20%), in the mouse striatum. These findings together with the recently published data on the potentiation by midazolam and allopregnanolone of ethanol-induced sleep [Pharmacol. Biochem. Behav. 67 (2000) 345] indicate a very similar central effect profile of benzodiazepines and neurosteroids. Moreover, similar efficacy of allopregnanolone and midazolam at the GABA(A) receptors has been found. Overall, the results of the present study, along with the possibility of neurosteroid conversion in the brain into other steroid hormones (testosterone, estradiol, aldosterone), add to the accumulating evidence suggesting a less favorable pharmacological profile for this class of drugs than was previously thought.