Abstracts

Mutations in the gene for the [gamma]2 subunit of the GABAA receptor [GABRG2(R43Q)] have been shown to underlie some forms of familial human absence epilepsy. We have previously demonstrated marked reductions in the binding potential (Bmax/KD) for the PET benzodiazepine receptor ligand [¹¹C]flumazenil in heterozygous family members. To disambiguate the effects of receptor density (Bmax) and affinity (KD), we examined the interaction between [³H]flumazenil and benzodiazepine receptors in wild type and GABRG2(R43Q) [lsquo]knockin[rsquo] mice. Wild type (RR; n=3), heterozygous (RQ; n=3) and homozygous (QQ; n=3) GABRG2(R43Q) [lsquo]knockin[rsquo] C57 Black 6 mice were studied at P15. Following light anesthesia and decapitation, the brain minus cerebellum was removed and homogenized. Following centrifugation and resuspension of the pellet, the protein concentration was determined using a bicinchoninic acid assay. Saturation experiments were performed using Millipore Multiscreen 96-well filter plates with 0.1 mg protein/well and varying concentrations of [³H]flumazenil (0.2- 30nM, 78.6Ci/mmol). Unlabeled flumazenil was used to determine non-specific binding. After a 30min incubation (20[deg]C) the plate was aspirated on a Vacuum Manifold and washed (3 x 200[mu]l TRIS-HCl). Radioactivity bound to filters was counted using a Beckman LS6500 scintillation counter after equilibration with Ultima Gold scintillation cocktail. Bmax and KD were estimated using a non linear least squares fit of the observed bound and free radioligand concentrations. Mean ([plusmn]standard deviation) Bmax values for RR, RQ and QQ mice were 400[plusmn]37, 235[plusmn]19 and 6[plusmn]2 pmol/mg protein respectively (p[lt]0.001 for Turkey[rsquo]s HSD for all two way comparisons following one way analysis of variance). Mean ([plusmn]standard deviation) KD values for RR and RQ mice were 2.6[plusmn]0.4 nM and 3.3[plusmn]0.6 nM respectively (p=0.21). Poor counting statistics resulting from almost non-existent binding precluded reliable estimates of KD for QQ animals. The GABRG2(R43Q) mutation results in dramatic changes in benzodiazepine receptor density. Receptor affinity for [³H]FMZ in the heterozygotes was equivalent to that in wild type mice. These findings support the hypothesis that reduced binding potential in PET studies of humans with the mutation results from a reduction in receptor density. Perturbations in [gamma]-aminobutyric acid type A (GABAA) receptor mediated inhibitory neurotransmission may underlie epileptogenic hyperexcitability in thalamocortical circuits. The observed changes in [gamma]2 GABAA subunit function may be a marker of these abnormalities or may have a more direct causative role if reduced binding by an endogenous receptor ligand modulates thalamocortical excitability. (Supported by National Health and Medical Research Council of Australia and Bionomics Limited)