Chick embryos with an undeveloped blood-brain barrier were used to examine the down-regulation of GABAA receptors in vivo. The GABAA receptor agonist isoguvacine (5 mumol) was applied to the vascularized chorioallantoic membrane of 8 day embryos. This treatment was repeated on embryonic days 11, 14, and 17, and the embryos were sacrificed on day 18 (stage 42). Isoguvacine administration reduced the clonazepam-displaceable binding of [3H]flunitrazepam to washed cerebellar membranes by 34.0 +/- 3.0% compared to vehicle-treated controls. Binding reductions of lower magnitude were found in membranes from the cerebrum and optic lobes. Administration of isoguvacine had no significant effect on the wet weights of whole embryos or cerebella, the yield of cerebellar membranes, or the binding of [3H]N-methylscopolamine. The reduction of [3H]flunitrazepam binding to cerebellar membranes was dose-dependent, allowing a half saturation value of 8 microM isoguvacine to be estimated. Scatchard analysis showed that the Bmax for [3H]flunitrazepam binding was reduced by 28.3 +/- 6.7% compared to controls, without a change in the Kd. Embryonic exposure to isoguvacine also caused a reduction of 43.6 +/- 6.0% in the binding of the GABAA receptor channel ligand [35S]t-butylbicyclophosphorothionate to washed cerebellar membranes. Taken together, these results indicate that isoguvacine induces a down-regulation of the receptor subunits in vivo. However, measurements of cerebellar GABAA receptor mRNAs for the alpha 1, beta 2L, beta 2S, beta 4, gamma 1, gamma 2L, and gamma 2S subunits by reverse transcriptase-polymerase chain reaction (RT-PCR) revealed no significant alterations by isoguvacine administration. The data suggest that translational or post-translational mechanisms, rather than those modulating the synthesis or stability of subunit mRNAs, take precedence in establishing GABAA receptor down-regulation.