DEPRESSION OF THE FAST IPSP UNDERLIES PAIRED-PULSE FACILITATION IN AREA CA1 OF THE RAT HIPPOCAMPUS

1. Intracellular recordings from CA1 pyramidal neurons in the rat hippocampal slice have been used to study synaptic transmission after maximal orthodromic stimulation of the Schaffer collateral-commissural fibers. Paired-pulse stimulation was used to investigate how the first (conditioning) stimulation influenced the response to the second (test) stimulation. 2. When the test stimulation was delivered up to approximately 4 s after the conditioning stimulation, the late phase of the excitatory postsynaptic synaptic potential (EPSP) was increased ("late-phase facilitation") whereas the fast (f-) and the slow (s-) inhibitory postsynaptic potentials (IPSPs) were depressed. 3. In terms of appearance and time course, facilitation of the intracellularly recorded EPSP was similar to that of the extracellularly recorded field EPSP in stratum radiatum. 4. The s-IPSP is not involved in facilitation of the EPSP. To show this, we counteracted the s-IPSP either by repolarizing the membrane potential to the resting level or by intracellularly injecting the quaternary lignocaine derivative QX 314. Facilitation of the late phase of the EPSP was unaffected by either procedure. 5. The conditioned response was modified in two ways when the stimulation was delivered at the equilibrium potential for the f-IPSP (E(f-IPSP)) and the s-IPSP had been blocked by intracellular injection of QX 314. The amplitude of the EPSP was increased, and the repolarizing phase was delayed with an apparent depolarizing shift of E(f-IPSP). This effect was present at pulse intervals > 20 ms and was maximal after 150 ms. Facilitation could be detected at interpulse intervals of up to 4 s. 6. The gamma-aminobutyric acid-B (GABA(B)) agonist baclofen (1-mu-M) reduced late-phase facilitation by preferentially increasing the unconditioned response, such that this came to resemble a conditioned response in control medium. 7. The f-IPSP was isolated pharmacologically to investigate its role in the facilitation of the EPSP. This was done by blocking the s-IPSP with QX 314 and the EPSP with a mixture of the N-methyl-D-aspartate (NMDA) receptor blocker, 2-amino-5-phosphonovaleric acid (APV, 50-mu-M), and the non-NMDA receptor blocker 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10-mu-M). An f-IPSP was then evoked by stimulating the interneurons directly. This potential could be blocked by the GABA(A) receptor antagonist bicuculline (20-mu-M), thereby confirming the successful isolation of GABA(A) ergic transmission. 8. With paired-pulse stimulation, the amplitude of the conditioned f-IPSP was depressed. The initial rising phases of the unconditioned and conditioned f-IPSPs were identical, but the time-to-peak and the recovery phase of the conditioned f-IPSP were both shorter. These kinetics were not mimicked by other maneuvers to reduce the f-IPSP, such as increasing the membrane potential or decreasing the stimulation intensity. 9. A depolarizing shift in E(f-IPSP) was also found to make a minor contribution to the depression of the f-IPSP. However, the depression of the conditioned f-IPSP was present at membrane potentials on either side of E(f-IPSP), indicating that the reduction was mainly due to a decrease of the GABA(A) conductance. 10. Paired-pulse depression of the isolated GABA(A) ergic inhibition was also studied with the use of single-electrode voltage-clamp recordings. With paired-pulse stimulation, the peak current as well as the total charge transfer of the conditioned response was depressed. The time course of the depression reflected that of the late-phase facilitation of the EPSP. It is concluded that paired-pulse facilitation of the late phase of the EPSP is largely accounted for by a depression of the f-IPSP. 11. Paired-pulse depression of the f-IPSP at an interpulse interval of 100 ms was greatly reduced by baclofen (1-mu-M). This effect was mainly due to a reduction of the unconditioned f-IPSP, which came to resemble a conditioned f-IPSP in control medium. Baclofen had only a minor effect on the conditioned f-IPSP. 12. It has not been possible to decide whether pre- or postsynaptic factors account for the difference in kinetics between the unconditioned and conditioned f-IPSPs. However, our results are in accordance with the proposed existence of presynaptic GABA(B) receptors on the inhibitory nerve terminals (autoreceptors). It is postulated that these receptors are activated by GABA, which is released in response to the conditioning stimulation. 13. Extracellular recordings of field EPSPs in stratum radiatum faithfully reflect depression of the GABA(A) receptor-mediated f-IPSP by a mechanism involving GABA(B) receptors. Measurement of the field EPSP provides a technically easy method for investigating putative GABA(B) receptor modulators.