DIRECT MODULATION OF GABA-A RECEPTOR BY INTRACELLULAR ATP IN DISSOCIATED NUCLEUS-TRACTUS-SOLITARII NEURONS OF RAT

1. Effect of intracellular ATP on Cl- current (I(Cl) mediated by the GABA (gamma-aminobutyric acid) receptor subtype, GABA(A), was studied in dissociated nucleus tractus solitarii (NTS) neurones using the whole-cell mode of patch clamp. A concentration-jump technique termed 'Y tube' was used to rapidly apply agents externally. Dissociated neurones were obtained from 1- to 3-week-old rats. 2. When the patch-pipette solution contained 2 mM-ATP, the amplitude of I(Cl) elicited by 10(-5) M-GABA did not show any time-dependent decrease (apparent run-down), for more than 60 min after the initial recording. In the presence of ATP, the half-maximum concentration (K(D)) and Hill coefficient calculated from the GABA concentration-response curve were 9.12-mu-M and 1.47, respectively. 3. In the absence of intracellular ATP, the amplitude of GABA-induced I(Cl) decreased with time. The relative peak amplitudes after 20 and 60 min from the initial recording were 0.40 +/- 0.09 (n = 11) and 0.16 +/- 0.05 (n = 8) with respect to the initial response. 4. Removal of Mg2+ from the internal solution induced run-down of the GABA response even in the presence of 2 mM-intracellular ATP, suggesting that both intracellular ATP (2 mM or more) and Mg2+ are necessary to prevent run-down of the GABA response. 5. Activation of dephosphorylation processes by alkaline phosphatase (100-200-mu-M) did not affect the GABA response in neurones perfused with internal solution containing 2 mM-ATP and 3 mM-Mg2+. Blocking the dephosphorylation process by okadaic acid, a phosphatase inhibitor, did not prevent the run-down of the GABA response. 6. Calcium influxes passing through both the voltage-dependent L-type Ca2+ channel and the glutamate receptor-operated cation channel did not affect I(Cl) induced by GABA. 7. GABA-induced I(Cl) was also maintained by adding 2 mM-ADP or ATP-gamma-S (adenosine-5'-O-3-thiotriphosphate) to the internal solution containing Mg2+. Addition of 2 mM-adenosine, AMP, cyclic AMP, AMP-PNP (adenylimido-diphosphate) or ADP-beta-S (adenosine-5'-O-2-thiodiphosphate) to the internal solution did not prevent the run-down of the GABA response even in the presence of 3 mM-intracellular Mg2+. Based on the chemical specificity of these ATP analogues, it is suggested that there is an ATP-sensitive binding site (ATP receptor) in the cytoplasmic side of the cell membrane. 8. The internally perfused P2 type ATP receptor antagonist, quinidine (0.1 mM), did not have any effect on the GABA response in neurones perfused with an internal solution containing 5 mM-ATP and 6 mM-Mg2+. 9. Reversal potential of the GABA response (E(GABA) was -17 mV, which was close to the Cl- equilibrium potential (E(Cl), - 18.5 mV). E(GABA) was not affected by the presence or absence of intracellular ATP. 10. Decreasing the concentration of intracellular ATP ([ATP]i) shifted the GABA concentration-response curve to the right without changing the maximum response, indicating that decreased [ATP]i produces a reduction in the affinity of the GABA(A) receptor to GABA. The run-down of the GABA response did not show any voltage dependence. 11. The chloride current elicited by pentobarbitone was also attenuated with time in response to reduced [ATP]i. However, glycine-gated I(Cl) and glutamate-induced current were not modulated by changing [ATP]i. 12. It was concluded that intracellular ATP may regulate the affinity of the GABA(A) receptor-Cl- channel complex via an intracellular ATP receptor.