Postsynaptic membrane shifts during frequency potentiation of the hippocampal EPSP

1. In some classes of central neurons, repetitive synaptic stimulation induces substantial changes in the postsynaptic membrane, in conjunction with robust frequency potentiation of the excitatory postsynaptic potential (EPSP). However, the nature and time course of these postsynaptic membrane shifts, or their possible contributions to EPSP frequency potentiation (e.g., by altering driving force or current pathways), have not been examined extensively. We therefore studied the simultaneous patterns of change in composite EPSP amplitude, postsynaptic input resistance (R(in)), and postsynaptic membrane potential during a 4-min train of 10-Hz monosynaptic stimulation in CA1 neurons of hippocampal slices. Slices were maintained in media containing either control (4 mM) or high (6.5 mM) concentrations of K⁺. 2. Potentiation of the EPSP, hyperpolarization of the membrane, and a decline of R(in), all developed rapidly during 10-Hz synaptic stimulation; these responses reached maximal levels by 5-15 s of the stimulation train. In most cells, a membrane depolarization phase occurred between 15 and 45 s of stimulation, followed by rehyperpolarization by 1 min of stimulation. During the depolarization phase, both EPSP potentiation and the decline in R(in) remained near maximal. No significant differences were seen as a function of K⁺ concentrations. 3. These results show that hyperpolarization is not invariably associated temporally with EPSP freguency potentiation. Moreover, if driving force and membrane conductance changes are assumed to be approximately similar in large dendrites and soma, then the increase in driving force due to membrane hyperpolarization was not sufficient to account for the three- and fourfold increases in EPSP amplitude seen during frequency potentiation. Further, based on similar assumptions and on dendritic models of EPSP attenuation, the decline in R(in) should reduce EPSP amplitude at the dendritic synaptic site and, to a proporionately greater extent, at the soma. 4. Studies in which the membrane was hyperpolarized with injected current to approximately the IPSP reversal potential, or in which bicuculline methiodide was applied to the slices, indicated that depression of the IPSP by repetitive stimulation did not account for frequency potentiation of EPSP amplitude. 5. These data are therefore consistent with the conclusion that the frequency potentiation of composite EPSPs in central neurons depends on presynaptic mechanisms, rather than on generalized postsynaptic changes. However, our findings do not rule out localized postsynaptic changes in receptors or spines as possible contributing factors. 6. Simultaneous intracellular and extracellular recording from the pyramidal cell somal layer (stratum pyramidale) showed that extracellular DC potential shifts did not correlate temporally with intracellular hyperpolarization. In addition, bicuculline, which blocks γ-aminobutyric acid-dependent Cl^- conductance, did not affect the initial hyperpolarization phase or conductance increase. Therefore, the hyperpolatization phase appears to result from the increase of bicuculine-resistant conductance. Because neuronal membrane potential closely correlated with glial cell membrane potential (which follows the K⁺ equilibrium potential) during 10-Hz stimulation but not during resting conditions, this conductance change may be due to increased K⁺ conductance. 7. The depolarization phase in neurons appears to result from the accumulation of extracellular K⁺, since it developed only following the onset of population spike activity (synchronous neuronal firing) and since, as noted, glial membranes exhibited a similar depolarization phase. 8. The rehyperpolatization phase correlated in time with the beginning of population spike decline, and may simply reflect the restoration of normal extracellular K⁺ concentration ([K⁺](0)). However, it is also possible that increased activation of an electrogenic Na⁺-K⁺-ATPase facilitated this rehyperpolarization, since a substantial (5-12 mV) and prolonged (0.5-3 min) hyperpolarization beyond the original resting potential was seen in glial cells at stimulation offset. 9. Thus, rather than contributing to hippocampal EPSP frequency potentiation, generalized postsynaptic changes (e.g., hyperpolarization, decreased membrane resistance) appear instead to function as negative-feedback mechanisms, which limit the effects of synaptic potentiation on postsynaptic excitation. Consequently, bicuculline-resistant postsynaptic mechanisms may play an important modulatory role in the processing and storage of information carried by high-frequency activity.