The functional characteristics of heteropentameric GABAA receptor subtypes are defined by their precise subunit composition. Anxiolytic benzodiazepines and endogenous neuroactive steroids allosterically modulate many subtypes, where acute exposure generally facilitates GABA-mediated transmission. Chronic exposure results in time-dependent and cell-specific changes in the expression of several receptor subunits, particularly the less common subunits.
We have shown that in rats exposed to diazepam for 28 days, changes in α4 RNA were mirrored by a change in the level of β1 RNA (Holt et al. 1996). Since the α4 subunit is often associated with the δ subunit, the present study has focused on the properties of α4, β1 and δ containing receptors. In a companion study reported at this meeting Griffiths et al. show that during the rat oestrous cycle the α4, β1 and δ subunits undergo parallel changes in expression in the periaqueductal grey matter, an area associated with initiating panic-like behaviour. Women suffering from premenstrual dysphoric disorder (PMDD), when challenged with flumazenil in the late luteal phase of their cycle, exhibit panic-like behaviour (Le Melledo et al. 2000). These observations suggest that flumazenil, an antagonist at the most common GABAA receptor subtypes, may act as an inverse agonist at the aberrant receptors expressed under these conditions.
Recombinant GABAA receptors, expressed in Xenopus oocytes by injection with 50 ng of total cRNA containing equal concentrations of each subunit-encoding RNA, were characterised by standard two-electrode voltage clamp procedures. The concentration dependence for GABA-mediated currents were measured for each receptor combination and the effects of either 100 nM flumazenil or flunitrazepam were measured at the GABA EC50.
The agonist response of the ubiquitous α1β2λ2L (EC50 32.8 ± 2.5 µM, mean ± S.E.M.) was enhanced by flunitrazepam (42.1+ 3.2 %, n = 7) but unaffected by flumazenil. After sequentially changing a single subunit in the expressed receptors we found that the agonist response of the α4β1δ receptor (EC50 2.02 ± 0.33 µM) was insensitive to flunitrazepam but flumazenil exhibited significant inverse agonist activity, decreasing the response to the GABA EC50 by 32.8 ± 6.6 % (n = 9). This receptor also displayed unusually slow desensitisation kinetics.
These results lend credence to the suggestion that aberrant GABAA receptors expressed during the late luteal phase of the menstrual cycle may explain the panicogenic effects of flumazenil challenge in PMDD.
This work was supported by the CIHR