Suppression of neuronal hyperexcitability and associated delayed neuronal death by adenoviral expression of GABA(C) receptors

Abstract

The excessive neuronal excitation underlying several clinically important diseases is often treated with GABA allosteric modulators in an attempt to enhance inhibition. An alternative strategy would be to enhance directly the sensitivity of postsynaptic neurons to GABA. The GABA(C) receptor, normally found only in the retina, is more sensitive to GABA and demonstrates little desensitization compared with the GABA(A) receptor. We constructed an adenovirus vector that expressed cDNA for both the GABA(C) receptor rho(1) subunit and a green fluorescent protein (GFP) reporter and used it to transduce cultured hippocampal neurons. Transduced neurons were identified by fluorescence, double immunocytochemistry proved colocalization of the rho(1) protein and the reporter, Western blot verified the expected molecular masses, and electrophysiological and pharmacological properties confirmed the presence of functional GABA(C) receptors. rho(1)-GFP transduction resulted in an increased density of GABA(A) receptors as well as expression of novel GABA(C) receptors. This effect was not reproduced by addition of TTX or Mg(2+) to the culture medium to reduce action potentials or synaptic activity. In a model of neuronal hyperexcitability induced by chronic blockade of glutamate receptors, expression of GABA(C) receptors abolished the hyperactivity and the consequent delayed neuronal death. Adenovirus-mediated neuronal GABA(C) receptor engineering, via its dual mechanism of inhibition, may offer a way of inhibiting only those hyperexcitable neurons responsible for clinical problems, avoiding the generalized nervous system depression associated with pharmacological therapy.

Fig. 1.

Adenovirus-mediated expression of functional GABA_C receptors in cultured hippocampal neurons.A, Phase-contrast (top) and GFP-fluorescent (bottom) image pairs of Ad-GFP-transduced or Ad-ρ₁-GFP-transduced (both at 1 × 10⁻⁵ pfu/ml) live cells. GFP-positive cells (examples shown by white arrows) are transduced. B, Anti-GFP-antibody-reactive (top) or anti-ρ₁-antibody-reactive (bottom) fluorescent views of the same field from cultures 48 hr after viral transduction. White arrowspoint to the virally transduced neurons immunoreactive for GFP. The ρ₁-GFP-neuron is also immunoreactive to anti-ρ₁ antibody. C, Western blot of detergent-extracted membrane (top) and soluble (bottom) fractions of HEK293 cells transduced with Ad-ρ₁-GFP (lanes 1, 3) or Ad-GFP (lanes 2, 4). The relative molecular masses are denoted on the right. D, Whole-cell patch-clamp recordings from transduced neurons. The left traces are from GFP-neurons, and the right traces are from ρ₁-GFP-neurons. V_h = −50 mV. Drug applications are denoted by theblack horizontal bar. E, Current–voltage relationship (bottom graph) obtained from ρ₁-expressing neurons recorded with high (left) or low (right) intracellular chloride concentrations. Arrows point to the reversal potentials. The expected shift in Nernst chloride potential because of the ionic conditions used was 72 mV. Scale bar: A, B, 150 μm.

Fig. 2.

Pharmacological separation of GABA_Areceptor- and GABA_C receptor-mediated current components. Currents evoked by applications of 100 μm GABA alone (left column) or in the presence of 100 μmbicuculline (Bicuc; middle column) or 50 μm I4AA (right column). The three drug trials for eachrow are from the same cell. The representative current records are from neurons not transduced with virus (top row) or transduced with Ad-GFP (middle row) or Ad-ρ1-GFP (bottom row). The duration of drug application is denoted by the black horizontal linesabove the first row of traces.

Fig. 3.

Kinetic properties of GABA-evoked currents in virally transduced hippocampal neurons. A–C, GABA (100 μm)-evoked currents shown at different time resolutions to emphasize the activation (A), desensitization (B), and deactivation (C) phases. The superimposed traces are from nonfluorescent neurons, GFP-neurons, and ρ₁-GFP-neurons with coapplication of 50 μm I4AA or 100 μmbicuculline. The peaks of the current traces have been normalized to demonstrate better the kinetic properties of the currents. The kinetically distinct current is from a ρ₁-GFP-neuron with coapplication of GABA and bicuculline, whereas the remaining three traces superimpose.D, E, A kinetic fit of a triexponential (τ_act, τ_des1, and τ_des2) function to the activation and desensitization phases and a separate monoexponential (τ_deact) function fit to the current deactivation. The two traces are both from ρ₁-GFP-neurons with GABA coapplied either with I4AA (D) (i.e., the GABA_A component) or with bicuculline (E) (i.e., the GABA_Ccomponent). See Table 1 for summary of kinetic parameters.

Fig. 4.

Expression of ρ₁ subunit suppresses spontaneous action potentials in the kynurenate–Mg²⁺ model of hyperexcitable neurons.A, A representative whole-cell current-clamp recording from a hippocampal neuron grown under standard cell culture conditions. The bottom trace is an expanded view of the region denoted by the black bar. B–D, Neurons rendered hyperexcitable by kynurenate–Mg²⁺treatment (B) and transduction with Ad-GFP (C) or Ad-ρ₁-GFP (D). E, Action potential firing rate for the same cells shown above demonstrating the stability of excitability over time (nontransduced hyperexcitable neurons,black circles; Ad-GFP-neurons, white circles; Ad-ρ₁-GFP-neurons, black inverted triangles; and control nontreated, nontransduced neurons,white inverted triangles). F, A bar graph of average action potential frequency for the four conditions (n = 8–20 neurons for each). Kyn, Kynurenate.

Fig. 5.

Pharmacological properties of hyperexcitable hippocampal neurons. A, Whole-cell current-clamp recordings from control nontransduced neurons (top), Ad-GFP-neurons (middle), or Ad-ρ₁-GFP-neurons (bottom) exposed to GABA (left) or CACA (right) are shown.Black horizontal lines denote the duration of drug application. B, GABA_C antagonist I4AA, GABA_A antagonist bicuculline, I4AA + bicuculline, or the nonspecific GABA antagonist picrotoxin were applied to Ad-ρ₁-GFP-transduced kynurenate–Mg²⁺-treated neurons. A summary bar diagram (bottom) of the relative action potential frequency during GABA antagonist applications (n = 6–12 neurons for each) is shown. No statistically significant difference (ns, p > 0.05) was found except for I4AA + Bicuc and picrotoxin (PTXN).

Fig. 6.

GABA_A receptor-mediated current density is enhanced after transduction with the ρ1-GFP virus.A, Current density (peak current evoked by 100 μm GABA/cell input capacitance) at 2, 4, and 6 d after transduction with no virus (gray bars), Ad-GFP (black bars), or Ad-ρ1-GFP (hatched bars). An asterisk denotes statistical significance (Ad-ρ1-GFP vs no virus) at p < 0.02, 0.01, or 0.0001 for 2, 4, or 6 d, respectively. Ad-GFP versus no virus was not significantly different at any time points (p > 0.21). B, Pharmacological separation of the total current (circle) into GABA_A (square) and GABA_C(triangle) components at different time points after viral transduction. At the day 6 time point (asterisks), both GABA_A (p < 0.01) and GABA_C (p < 0.0001) components were significantly greater compared with those on day 0.C, Current density of neurons grown for 6 d in the control medium (Cont) supplemented as noted. Only Ad-ρ1-GFP-transduced neurons demonstrated significantly greater current density (p < 0.001). ForA–C, the numbers of cells are denoted inparentheses.

Fig. 7.

Suppression of hyperexcitability-induced delayed neuronal death. A, Phase-contrast and fluorescent image pairs of a culture dish rendered hyperexcitable for 0 min (top) or 30 min (bottom).Arrows point to the ethidium-homodimer-positive dead neurons. B, Summary bar diagram of the percentage of neuronal death versus the duration of hyperexcitability.C, Phase-contrast (top left), GFP (bottom), and ethidium (top right) images from an Ad-ρ₁-GFP-transduced culture subjected to 30 min of hyperexcitability. White arrows point to nontransduced dead (i.e., GFP-negative and ethidium-positive) neurons, and black arrows point to transduced live (i.e., GFP-positive and ethidium-negative) neurons. D, Summary bar diagram of delayed neuronal death in control and Ad-ρ₁-GFP-transduced neurons from three separate experiments (asterisk, significant atp < 0.0001).