Intact synaptic GABAergic inhibition and altered neurosteroid modulation of thalamic relay neurons in mice lacking delta subunit.

Robust GABA-mediated inhibitory postsynaptic currents (IPSCs) in neurons of the thalamic relay (TC) nuclei are important in sustaining oscillatory activity within thalamic and thalamocortical circuits. The biophysical properties and pharmacological sensitivities of these IPSCs both depend on the subunit combination of postsynaptic gamma-aminobutyric acid-A (GABA(A)) receptors. Recombinant GABA(A) receptors containing the delta subunit (heavily expressed in TC nuclei) have been shown to exhibit slowed desensitization rates and high affinity for GABA in heterologous expression systems. We tested whether the GABA(A)-mediated synaptic inhibition in TC neurons would be affected by loss of the delta subunit. Spontaneous and evoked IPSCs were recorded from neurons in the ventral basal complex (VB) of the thalamus from brain slices of wild-type (delta(+/+)) and homozygous delta subunit deficient mice (delta(-/-)). Spontaneous IPSCs (sIPSCs) from delta(-/-) mice had no significant differences in amplitude, duration, or frequency compared with their delta(+/+) counterparts. However, baseline noise (63% of control) and the relative contribution of the slow component to overall decay (79% of control) were significantly lower in delta(-/-) VB recordings. Evoked IPSCs (eIPSCs) in delta(-/-) neurons showed no difference in peak amplitude, but had an accelerated slow decay component (40- vs. 55-ms time constant). We further tested whether neurosteroid modulation of GABA(A) receptors was dependent on the presence of the delta subunit, as previously reported in recombinant systems. Pregnenolone sulfate (PS) significantly reduced eIPSC peak amplitude (-30%) and increased duration in delta(-/-), but not in delta(+/+) mice. sIPSCs were not affected in any neurons, delta(-/-) or delta(+/+). In contrast, 3-alpha,5-alpha-tetrahydrodeoxycorticosterone (THDOC) increased the durations of eIPSCs and sIPSCs in both delta(-/-) and delta(+/+) VB neurons. Our findings show that although the delta subunit confers a striking PS insensitivity to eIPSCs in VB neurons, it plays only a minor role in the synaptic inhibition of VB neurons. This suggests delta subunit containing GABA(A) receptors may be functionally limited to an extrasynaptic locus in VB neurons.

Actions of U-92032, a T-type Ca2+ channel antagonist, support a functional linkage between I(T) and slow intrathalamic rhythms.

Thalamic relay neurons express high levels of T-type Ca(2+) channels, which support the generation of robust burst discharges. This intrinsically mediated form of phasic spike firing is thought to be critical in the generation of slow (3-4 Hz) synchronous oscillatory activity of absence epilepsy. Recordings made from brain slices or whole animals have shown that slow synchronous absence-like activity can be abolished when Ca(2+)-dependent burst firing in relay neurons is interrupted by the pharmacological or genetic inactivation of T-channels. Because succinimide drugs act as incomplete and nonspecific antagonists, we tested whether the novel T-channel antagonist U-92032 could provide stronger support for a role of T-channels in slow oscillatory activity. Ca(2+)-dependent rebound (LTS) bursts were recorded using whole cell current clamp in relay cells of the ventral basal complex (VB) from thalamic slices of adult rats. We used LTS kinetics to measure the availability of T-channels in VB cells after TTX. U-92032 (1 and 10 microM) reduced the maximum rate of depolarization of the isolated LTS by 51% and 90%, respectively, compared with the 35% reduction due to 2 mM methylphenylsuccinimide (MPS), the active metabolite of the antiabsence drug methsuximide. U-92032 (1 and 10 microM) also suppressed evoked, slow oscillations in thalamic slices with a time course similar for observed intracellular effects. Unlike MPS, we observed no substantial effects of short-term U-92032 applications (< or =2 h) on the generation of action potentials in VB cells. Our findings show U-92032 is a more potent, effective, and specific T-channel antagonist than previously studied succinimide antiabsence drugs and that it dramatically reduces epileptiform synchronous activity. This suggests that U-92032 or other specific T-channel antagonists may provide effective drug treatments for absence epilepsy.

Resilient RTN fast spiking in Kv3.1 null mice suggests redundancy in the action potential repolarization mechanism.

Fast spiking (FS), GABAergic neurons of the reticular thalamic nucleus (RTN) are capable of firing high-frequency trains of brief action potentials, with little adaptation. Studies in recombinant systems have shown that high-voltage-activated K(+) channels containing the Kv3.1 and/or Kv3.2 subunits display biophysical properties that may contribute to the FS phenotype. Given that RTN expresses high levels of Kv3.1, with little or no Kv3.2, we tested whether this subunit was required for the fast action potential repolarization mechanism essential to the FS phenotype. Single- and multiple-action potentials were recorded using whole-cell current clamp in RTN neurons from brain slices of wild-type and Kv3.1-deficient mice. At 23 degrees C, action potentials recorded from homozygous Kv3.1 deficient mice (Kv3.1(-/-)) compared with their wild-type (Kv3.1(+/+)) counterparts had reduced amplitudes (-6%) and fast after-hyperpolarizations (-16%). At 34 degrees C, action potentials in Kv3.1(-/-) mice had increased duration (21%) due to a reduced rate of repolarization (-30%) when compared with wild-type controls. Action potential trains in Kv3.1(-/-) were associated with a significantly greater spike decrement and broadening and a diminished firing frequency versus injected current relationship (F/I) at 34 degrees C. There was no change in either spike count or maximum instantaneous frequency during low-threshold Ca(2+) bursts in Kv3.1(-/-) RTN neurons at either temperature tested. Our findings show that Kv3.1 is not solely responsible for fast spikes or high-frequency firing in RTN neurons. This suggests genetic redundancy in the system, possibly in the form of other Kv3 members, which may suffice to maintain the FS phenotype in RTN neurons in the absence of Kv3.1.