Regulation of epileptiform activity by two distinct subtypes of extrasynaptic GABAA receptors

Abstract

Background: GABAergic deficit is one of the major mechanisms underlying epileptic seizures. Previous studies have mainly focused on alterations of synaptic GABAergic inhibition during epileptogenesis. Recent work suggested that tonic inhibition may also play a role in regulating epileptogenesis, but the underlying mechanism is not well understood.

Results: We employed molecular and pharmacological tools to investigate the role of tonic inhibition during epileptogenesis both in vitro and in vivo. We overexpressed two distinct subtypes of extrasynaptic GABAA receptors, α5β3γ2 and α6β3δ receptors, in cultured hippocampal neurons. We demonstrated that overexpression of both α5β3γ2 and α6β3δ receptors enhanced tonic inhibition and reduced epileptiform activity in vitro. We then showed that injection of THIP (5 μM), a selective agonist for extrasynaptic GABAA receptors at low concentration, into rat brain also suppressed epileptiform burst activity and behavioral seizures in vivo. Mechanistically, we discovered that low concentration of THIP had no effect on GABAergic synaptic transmission and did not affect the basal level of action potentials, but significantly inhibited high frequency neuronal activity induced by epileptogenic agents.

Conclusions: Our studies suggest that extrasynaptic GABAA receptors play an important role in controlling hyperexcitatory activity, such as that during epileptogenesis, but a less prominent role in modulating a low level of basal activity. We propose that tonic inhibition may play a greater role under pathological conditions than in physiological conditions in terms of modulating neural network activity.

Figure 1

Tonic GABA current increased after the overexpression of α5β3γ2 GABA_A receptors. A, Typical GABA (100 μM) induced currents in HEK293T cells transfected with α5β3γ2 subunits (left panel), which could be largely blocked by α5 subunit-specific inverse agonist L655,708 (100 nM, right panel). B, Summarized data showing GABA-induced α5β3γ2 receptor currents in HEK293T cells significantly inhibited by L655,708 (Control, 437.1 ± 63.1 pA, n = 10; L655,708, 142.8 ± 25.7 pA, n = 10; ***, p < 0.001). C, Typical GABA current traces in cultured hippocampal neurons transfected with mCherry or plus the α5β3γ2 subunits. D, Bar graphs showing no significant difference between the total whole-cell GABA currents in neurons transfected with mCherry or plus the α5β3γ2 subunits. E, Representative tonic GABA currents revealed by rapid application of GABA_A-R blocker bicuculline (100 μM) in hippocampal neurons transfected with mCherry or plus the α5β3γ2 subunits. F, Summarized data showing that tonic GABA current in α5β3γ2-transfected neurons (24.5 ± 4.2 pA, n = 8) was significantly increased in comparison with the control neurons (13.1 ± 1.7 pA, n = 8; *, p < 0.05).

Figure 2

Inhibition of epileptiform activity in cultured hippocampal neurons overexpressing α5β3γ2 receptors. A, Typical traces from two hippocampal neurons showing the epileptiform burst activity after chronic CTZ treatment (5 μΜ, 24 h). Panel b shows the expanded view of a single epileptiform burst from panel a. Epileptiform burst is characterized by a train of action potentials on a large depolarization shift. B, Representative traces showing the lack of epileptiform bursts in two hippocampal neurons transfected with the α5β3γ2 receptors. C, Bar graphs illustrating that overexpression of α5β3γ2 receptors significantly reduced the percentage of neurons showing epileptiform activity after chronic CTZ treatment (mCherry, ~90%, n = 29; α5β3γ2, ~33%, n = 33; ***, p < 0.001). D, Neurons transfected with α5β3γ2 receptors showing lower burst frequency (0.49 ± 0.16 per min, n = 33), compared to mCherry controls after CTZ treatment (1.83 ± 0.35 per min, n = 29; ***, p < 0.001).

Figure 3

Overexpression of α6β3δ subunits results in large tonic GABA current in cultured hippocampal neurons. A, Typical recordings showing whole-cell currents induced by rapid application of GABA (20 μM) in GFP control and α6β3δ-transfected neurons in the presence of TTX (1 μM) and DNQX (10 μM). B, Summarized data showing no significant difference in whole-cell GABA currents between the two groups. C, Typical traces of tonic GABA currents, revealed by application of Bic (40 μM) in the presence of TTX (1 μM) and DNQX (10 μM), recorded from a GFP control neuron and a α6β3δ-transfected neuron. Holding potential = −70 mV. D, Summarized data showing that the average amplitude of tonic GABA currents was significantly increased in α6β3δ-transfected neurons (28.1 ± 3.6 pA, n = 10), compared to the GFP controls (7.4 ± 1.1 pA, n = 11; ***, p < 0.0001). E, Tonic currents activated by THIP (5 μM) in GFP control and α6β3δ-transfected neurons. F, Summarized data showing a significant increase of THIP-induced tonic currents after transfection of α6β3δ subunits (962 ± 130 pA, n = 12), compared to the GFP controls (35.3 ± 7.4 pA, n = 11; ***, p < 0.0001).

Figure 4

Overexpression of α6β3δ receptors inhibits epileptiform bursting activity in cultured hippocampal neurons. Aa, Representative traces showing the typical recurrent epileptiform bursts after chronic pretreatment with CTZ (5 μM, 24 h) in two different GFP-transfected pyramidal neurons. Ab, A single epileptiform burst in (a) was expanded to show a train of action potentials overlaying on a large depolarization shift. Ba, Representative traces showing the lack of typical epileptiform bursts in two α6β3δ-transfected hippocampal neurons. Bb, Expanded view of the boxed activity in (a). C, Bar graph showing the percentage of neurons with epileptiform bursting activity after chronic treatment with CTZ (5 μM for 24 h). *** p < 0.001, Pearson Chi-Square test. D, Bar graph showing a significant reduction of the average epileptiform burst frequency in neurons transfected with α6β3δ receptors, comparing to GFP controls after CTZ treatment. ** p < 0.01.

Figure 5

THIP inhibits CTZ-induced epileptiform activity and seizure behavior. Aa-d, Typical traces showing ‘silent’ baseline activity of hippocampal CA1 neurons in control condition (a), synchronized epileptiform bursting activities induced by CTZ (5 μmol) (b), inhibition of THIP (4 mg /kg) on burst activity (c), and the recovery after the THIP injection (d). Ae, Bar histogram showing group data of THIP inhibition on CTZ-induced epileptiform burst activities. B, Bar histogram showing group data of CTZ-induced (5 μmol) seizure behavioral score and its significant attenuation by pre-treatment with THIP (10 mg/kg). THIP(5): 5 mg/kg THIP; THIP(10): 10 mg/kg THIP. * p < 0.05 and ** p < 0.01 in comparison with CTZ injection alone; # p < 0.05 for recovery, in comparison with CTZ + THIP.

Figure 6

THIP at low concentration has no effect on the amplitude and frequency of mIPSCs. A, B, Representative traces showing mIPSCs recorded in control and THIP (5 μM) treated neurons. Note that in the presence of THIP, baseline was always noisier than control traces, indicating the activation of extrasynaptic GABA_A-Rs. C, D, No significant difference between the mIPSC amplitude (p > 0.6) and frequency (p > 0.4) in control and THIP-treated neurons.

Figure 7

THIP effect on basal and elevated neuronal activity. A, B, THIP had no significant effect on the basal level of action potential firing (p > 0.4, n = 12). C, D, THIP significantly inhibited CTZ-induced action potentials (p < 0.01, n = 17). E, F, THIP also inhibited KA-induced action potentials (p < 0.03, n = 10).