Voltage-clamp analysis of mossy fiber synaptic input to hippocampal neurons

Abstract

Evoked synapatic responses were studied in hippocampal pyramidal neurons, using the in vitro slice preparation. Using both current- and voltage-clamp techniques, the amplitude and time course of the synaptic responses were measured as a function of the membrane potential of the postsynaptic cell. The results provide the 1st description of the postsynaptic conductance mechanisms responsible for the generation of synaptic signals in a mammalian cortical neuron. Intracellular recordings were made with low-resistance (10-30 M.OMEGA.) glass micropipettes, which were usually filled with 2 M Cs2SO4 or 4 M potassium acetate. Under direct visual control, the recording micropipette tips were positioned into the stratum pyramidale (the pyramidal cell body layer) of the CA3 region, and bipolar stimulating electrodes were placed into the stratum granulosum (granule cell body layer) of the dentate gyrus. Low-intensity current was used to stimulate the granule cells, whose mossy fiber axons are known to make monosynaptic en passant synapses onto the proximal apical dendrites of the pyramidal neurons. A 3-kHz time-share single-electrode clamp (SEC) was used for both current- and voltage-clamp experiments. In many of the experiments, Cs+ were injected into the postsynaptic pyramidal neurons to increase their input resistance and reduce outward rectification. Strict criteria were established for accepting voltage-clamp results, criteria that were satisfied in only 12 of the 298 cells studied. Under current-clamp conditions, when the membrane potential of the postsynaptic pyramidal cell was -60 to -70 mV or more negative, the postsynaptic potential (PSP) typically appeared to consist of a simple monophasic depolarization. When the pyramidal cell was depolarized by passing outward current, the PSP became biphasic, consisting of a depolarizing early phase followed by a hyperpolarizing late phase. The early phase is thought to represent the monosynaptic response to the granule cell mossy fiber input, while the late phase is believed to arise from recurrent or feedforward synaptic inhibition. When the pyramidal cell was further depolarized, the amplitude of the hyperpolarizing phase of the PSP greatly increased and the amplitude of the depolarizing phase decreased. When such cells were depolarized more positive that .apprx. -30 mV, the entire PSP waveform typically appeared to become hyperpolarizing. When the late phase was blocked pharmacologically by adding picrotoxin (5-10 .mu.M) or penicillin (3.3 mM) to the bath, the reversal potential of the remaining PSP shifted in a positive direction to a mean (.+-. SE) value of -0.3 .+-. 2.9 mV. Synaptic signal generation produced by granule cell stimulation results from a conductance increase mechanism and the currents display a clear reversal potential, consistent with a chemical mode of synaptic transmission. This characterization of the conductance waveforms and current reversal potentials is important for understanding the genesis of synaptic signals and their propagation throughout the dendritic arborization. The ability to separate and quantify the conductances responsible for the excitatory and inhibitory currents will permit tests of several of the possible mechanisms that may underlie the interesting forms of use-dependent synaptic plasticity in the hippocampus.