Molecular basis of self‐sustaining seizures and pharmacoresistance during status epilepticus: The receptor trafficking hypothesis revisited

During status epilepticus (SE), seizures become self-sustaining (Wasterlain, 1974), and pharmacoresistance to benzodiazepines develops progressively, and can result in a 20-fold reduction of response to diazepam after 30 min of experimental seizures (Kapur & Macdonald, 1997; Mazarati et al., 1998a,b). This time-dependent pharmacoresistance is not mediated by classical transporter mechanisms (Löscher, 2007), but may reflect seizure-induced internalization of synaptic γ-aminobutyric acid (GABA)A receptors (GABAAR) containing the β2–3 and/or γ2 subunits (Goodkin et al., 2005, 2008; Naylor et al., 2005). When we compared slices obtained from rats in lithium–pilocarpine SE for 60 min to slices from control rats, the miniature inhibitory postsynaptic currents (mIPSCs) recorded from dentate granule cells were of lower amplitude after SE (Fig. 1A). Mathematical modeling based on a 7-state model of the GABAAR suggested that this reflected a reduction of the number of GABAAR from 36 ± 11 per synapse in controls to 18 ± 4 per synapse in animals subjected to SE. Minor changes in kinetics were also observed. This reduction of the number of active receptors per synapse presents therapeutic challenges, since GABAAR internalization occurs quickly (Naylor, 2002; Naylor & Wasterlain, 2005). As a result, the therapeutic response to the drugs of choice against SE (benzodiazepines and other GABAergic drugs) declines, and this makes it difficult to control seizures, since very early treatment of SE is rarely possible. In hippocampal slices prepared from rats in lithium/pilocarpine–induced status epilepticus (SE) for 1 h, γ-aminobutyric acid (GABA)A miniature inhibitory postsynaptic currents (IPSCs) recorded from granule cells (A, reduced number of GABAA receptors from 36 ± 11 to 18 ± 4 per granule cell synapse). Exposure of hippocampal slices to GABA produces similar changes (B, upper middle graph). Tonic currents generated by extrasynaptic GABAA receptors are increased in slices from rats in SE, reflecting (at least in part) increased extracellular GABA concentration during SE (C, upper right). In control granule cells (D, bottom left) the β2–3 subunits of GABAA receptors (red) localize to the vicinity of the presynaptic marker synaptophysin (green), whereas after an hour of SE induced by lithium and pilocarpine (D, bottom right), many have moved to the cell interior. The mechanism of those changes is poorly understood. Exposure of control slices for 20 min to 3 μm GABA plus a GABA-uptake inhibitor (10 μm NO711), resulted in a 48% reduction of mIPSC amplitude, suggesting that increased extracellular GABA can generate changes similar to SE. This was surprising, since in tissue culture, agonist-induced internalization of GABAAR has not been reported. An indirect mechanism has to be considered, but its molecular basis is not clear. By contrast, in the same dentate granule cell population, the tonic currents, generated by extrasynaptic receptors, actually increased (Fig. 1C) in slices from rats in SE (Naylor et al., 2005; Goodkin et al., 2008). Some experiments have suggested that this reflects an increase in extracellular GABA concentration without any change in the number of extrasynaptic GABAARs (Naylor et al., 2005), whereas other studies indicate an actual increase in receptor number (Goodkin et al., 2008). Because the bulk of tonic currents are probably generated by extrasynaptic, δ subunit–containing GABAAR, this suggests a new treatment of SE, using selective ligands of those receptors (neurosteroids such as ganaxolone, Mihalek et al., 1999). However, if these receptors are fully saturated by extracellular GABA, one would expect this approach to have limited success, since the receptors may already be maximally activated. We visualized the β2–3 subunits of GABAAR by immunocytochemistry/confocal microscopy in hippocampal granule cells of controls (Fig. 1D) and of rats in lithium–pilocarpine SE for 60 min (Fig. 1E). The subunits move from the cell membrane to the cell's interior with SE. Similar internalization of GABAAR subunits was seen in CA1 and CA3 pyramidal cells. The γ2 subunits of GABAAR, which are also synaptically located in control brains, showed similar changes. Changes in glutamate receptors and in other transmitter systems also play a role in hippocampal hyperexcitability during SE. Hippocampal slices prepared from rats after 60 min of lithium–pilocarpine SE display an increase in the amplitude of N-methyl-d-aspartate (NMDA)–mediated miniature excitatory postsynaptic currents (mEPSCs) on granule cell somas, suggesting an increase in number of functional NMDA receptors (NMDARs) per synapse. After an hour of SE in vivo, we observed a movement of NMDARs (visualized by immunocytochemistry of the NR1 subunit and confocal microscopy) from the cytoplasm to the vicinity of synapses in granule cells, and in CA1 and CA3 pyramids, suggesting activity-dependent trafficking of NMDARs to synapses. Because both the loss of synaptic GABAARs and the increase in synaptic NMDARs are proconvulsant, this suggests that the traditional treatment of SE by enhancing GABA inhibition leaves glutamatergic excitation unchecked, and that polytherapy is needed to simultaneously stimulate the remaining GABAARs and reduce the activity of NMDARs. Further increases in treatment potency could be achieved by adding, to the GABAA agonist/NMDA antagonist combination, a third drug, which enhances inhibition by non-GABA, nonglutamatergic mechanisms. We are testing this hypothesis in a very severe model of SE (5 mEq/kg lithium, 320 mg/kg pilocarpine), which is refractory to anesthetic doses of diazepam (20 mg/kg). Our goal is to stop seizures without depressing consciousness, using low doses (devoid of any measurable effect on SE when given alone) of a GABAAR agonist combined with an NMDA antagonist and a third drug with a different mechanism of action. We are also examining whether receptor trafficking–targeted polytherapy can reduce the long-term sequelae of SE. Supported by VHA Research Service and by grants NS13515 and NS 059704 from NIH/NINDS. The authors confirm that they have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Disclosure: None of the authors has any conflicts of interest in relation to this paper to declare.