Regulation of EPSPs by the synaptic activation of GABA(B) autoreceptors in rat hippocampus

1. Intracellular recording was used to study the influence of GABA(B) autoreceptor-mediated regulation of monosynaptic GABA(A) and GABA(B) receptor-mediated hyperpolarizing inhibitory postsynaptic potentials (IPSP(A)s and IPSP(B)s, respectively) on alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) and N-methyl-D-aspartate (NMDA) receptor-mediated excitatory postsynaptic potentials (EPSP(A)s and EPSP(N)s, respectively) in the CA1 region of rat hippocampal slices. To achieve this, synaptic potentials were evoked monosynaptically by near stimulation following blockade of either EPSP(N)s, by the NMDA receptor antagonist (R)-2-amino-5-phosphonopentanoate (AP5; 0.05 mM), or EPSP(A)s, by the AMPA/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 0.01 mM). 2. Paired-pulse stimulation at 3-50 Hz caused an increase in the duration (paired-pulse widening) of EPSP(A)s, which paralleled the time course of paired-pulse depression of monosynaptic IPSCs, and a potentiation of the amplitude (paired-pulse potentiation) of EPSP(A)s, which did not. Paired-pulse stimulation also caused frequency-dependent changes in EPSP(N)s. At frequencies >40 Hz it produced paired-pulse depression of EPSP(N)s, along with marked summation of IPSPs, and at frequencies <40 Hz it caused paired-pulse enlargement of EPSP(N)s, concomitant with a reduction in IPSPs. 3. Paired-pulse potentiation of EPSP(A)s at 50 Hz was enhanced by picrotoxin (0.1 mM) but was not significantly affected by 3-amino-propyl(diethoxymethyl)phosphinic acid (CGP 35348; 1 mM). Paired-pulse depression of EPSP(N)s at 50 Hz was converted to paired-pulse enlargement by picrotoxin but was unaffected by CGP 35348. These effects can be explained by block of IPSP(A)s by picrotoxin. 4. Paired-pulse widening of EPSP(A)s at 5 Hz was occluded by picrotoxin and abolished by CGP 35348. Similarly, paired-pulse enlargement of EPSP(N)s at 5 Hz was occluded, and in some cases converted to paired-pulse depression, by picrotoxin. The effects of CGP 35348 were more complex in that this antagonist reduced paired-pulse enlargement of EPSP(N)s in control medium whereas it eliminated paired-pulse depression of EPSP(N)s in the presence of picrotoxin, effects consistent with its block of GABA(B) autoreceptors and IPSP(B)s, respectively. 5. 'Priming' using a 'priming stimulation protocol' (a single 'priming stimulus' followed at 1-50 Hz ('priming frequency') by a 'primed burst' of four shocks at 20-100 Hz ('burst frequency')) caused an increase in both 'primed' EPSP(A)s and EPSP(N)s compared with 'unprimed' EPSP(A)s and EPSP(N)s. This effect was optimal when the respective priming and burst frequencies were 5 and 100 Hz. 6. In the presence of either picrotoxin or CGP 35348 the primed EPSP(A)s and EPSP(A)s resembled unprimed EPSP(A)s and EPSP(N)s, respectively. This was because picrotoxin occluded whereas CGP 35348 blocked the effect of priming on EPSPs. 7. CGP 35348 had only modest effects on EPSP(A)s but enhanced EPSP(N)s evoked by a tetanus (20 stimuli at 100 Hz), in either the presence or absence of picrotoxin. In the absence of picrotoxin, CGP 35348 also promoted depolarization by enhancing a depolarizing GABA(A) receptor-mediated component (IPSPD). These effects can all be attributed to block of IPSP(B)s by CGP 35348. 8. CGP 35348 blocked the induction of long-term potentiation (LTP) of extracellularly recorded field EPSPs elicited by a priming stimulation protocol in control medium but was ineffective in the presence of picrotoxin. CGP 35348 was also ineffective at preventing tetanus-induced LTP (100 Hz, 1 s), in both the absence and presence of picrotoxin. 9. These data demonstrate the complex regulation of AMPA and NMDA receptor-mediated EPSPs during various patterns of synaptic activation caused by the dynamic changes in GABA-mediated synaptic inhibition, which are orchestrated by GABA(B) autoreceptors in a frequency-dependent manner.