PHARMACOLOGICAL SUBTYPES OF THE GAMMA-AMINOBUTYRIC ACID(A) RECEPTORS DEFINED BY A GAMMA-AMINOBUTYRIC-ACID ANALOG 4,5,6,7-TETRAHYDROISOXAZOLO[5,4-C]PYRIDIN-3-OL AND ALLOSTERIC COUPLING - CHARACTERIZATION USING SUBUNIT-SPECIFIC ANTIBODIES

Various alpha and beta 3 subunit-specific antibodies were used to char acterize some of the heterogeneous ligand-binding properties of gamma- aminobutyric acid, receptors. Polyclonal antibodies that were raised a gainst the cytoplasmic amino acid sequence (380-392) of the rat beta 3 subunit recognized a single polypeptide of molecular mass of 58 kDa i n Western blots with Ro7-1986 affinity-purified GABA(A) receptors from the rat brain, and a doublet of molecular mass of 54 kDa and 56 kDa i n receptors from the bovine cortex, hippocampus, and cerebellum. Degly cosylation of purified receptors from the bovine cortex with N-glycana se resulted in a single band immunostained at molecular mass of 52 kDa . These anti-beta 3 subunit antibodies immunoprecipitated similar to 5 0% of [H-3]flunitrazepam binding sites from soluble extracts of bovine cortex, whereas beta cyto antibodies, which probably recognize all be ta subunit isoforms, precipitated almost 100% of benzodiazepine bindin g sites. These results indicate heterogeneity of GABA(A) receptor subu nit composition with respect to the nature of beta subunits. The GABA analogue 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP), like G ABA(A) shows heterogeneous binding affinities in brain homogenates. Th e higher affinity sites were previously suggested as corresponding to a 58-kDa polypeptide in rat that is photoaffinity-labeled with [H-3]mu scimol, a band that comigrates with the one stained by anti-beta 3 ant ibodies. However, THIP affinity was not significantly different betwee n receptors containing beta 3 subunits and those lacking beta 3, as de monstrated by similar affinities in receptors that ere immunoprecipita ted by anti-beta 3 antibodies and those that were not. Also, THIP disp laced [H-3]muscimol binding with similar multiple affinities across br ain regions where different beta subunit variants are expressed with v arying abundances. These observations suggest that the property of hig h affinity THIP binding cannot be explained solely by beta 3 subunits. The coupling efficiency between GABA and benzodiazepine binding sites appears to be determined by the nature of alpha subunits rather than of beta subunits. GABA enhanced [H-3]flunitrazepam binding with differ ent efficacies and potencies in receptors immunoprecipitated by anti-a lpha 1, -alpha 2, and -alpha 3 subunit antibodies. In contrast, beta 3 subunit-enriched and disenriched receptors did not differ in this pro perty. [H-3]flunitrazepam binding in GABA(A) receptors containing alph a 2 and alpha 3 subunits was enhanced to a significantly greater exten t than were those with alpha 1. In addition, receptors containing alph a 1 and alpha 3 subunits had higher potencies of enhancement than did those with alpha 2 subunits. These binding results suggest that benzod iazepines are likely to show differential efficacies at different subt ypes of GABA(A) receptors in the brain, as suggested by studies on rec ombinant receptor expression.