S. R. Kelso

A combination of current- and voltage-clamp techniques applied to hippocampal brain slices was used to evaluate the role of postsynaptic electrogenesis in the induction of associative synaptic enhancement. In accordance with Hebb's postulate for learning, repetitive postsynaptic spiking enabled enhancement in just those synapses that were eligible to change by virtue of concurrent presynaptic activity. However, the essential postsynaptic electrogenic event that controlled the enhancement was shown to involve biophysical processes that were unknown when Hebb formulated his neurophysiological postulate. The demonstrated spatiotemporal specificity of this pseudo-Hebbian conjunctive mechanism can account qualitatively for the known neurophysiological properties of associative long-term potentiation in these synapses, which in turn can explain the "cooperativity" requirement for long-term potentiation.

1. N-Methyl-D-aspartate (NMDA) receptors were expressed in Xenopus oocytes injected with rat brain RNA. The modulation of NMDA-induced currents was examined by activating protein kinase C (PKC) either directly (using phorbol esters) or indirectly (via metabotropic glutamate agonists). 2. Bath application of the PKC activator, 4-beta-phorbol-12,13-dibutyrate (PDBu) resulted in a two-fold increase in the NMDA-evoked current at all holding potentials examined (-80 to 0 mV). The inactive (alpha) stereoisomer of phorbol ester was ineffective. 3. The increase was observed under conditions that eliminate the oocyte's endogenous calcium-dependent chloride current, which often contributes to the NMDA response in oocytes. 4. The PDBu effect was specific to the NMDA subclass of glutamate receptors in that no increase was observed in the responses to two other glutamate agonists, kainate and AMPA (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid). 5. Stimulation of PKC by activation of metabotropic receptors via either quisqualate or trans-ACPD (trans-1-aminocyclopentane-1,3-dicarboxylic acid) also led to an increase in NMDA currents. 6. Both methods of enhancement induced transient effects. PDBu effects lasted 10-45 min, depending upon both dose and length of application. Quisqualate and trans-ACPD effects were shorter, lasting less than 10 min under these conditions of application. 7. Both methods of enhancement were blocked by the PKC inhibitor, staurosporine. In addition, the phorbol ester-induced enhancement of NMDA responses occluded further enhancement by quisqualate. 8. The results suggest a role for metabotropic glutamate receptors in modulation of NMDA-mediated processes.

To understand the basis of hypothalamic neuronal thermosensitivity, single-unit activity was recorded in vitro, from constantly perfused tissue slices of rat preoptic area and anterior hypothalamus, PO/AH. The firing rate and thermosensitivity of individual PO/AH neurons was determined before, during, and after tissue perfusion with a synaptic blocking medium, containing elevated magnesium and decreased calcium concentrations. During synaptic blockade, thermosensitivity was retained in nearly all of the warm-sensitive neurons, and some temperature-insensitive neurons showed increased warm sensitivity. The thermosensitivity of all cold-sensitive neurons was lost during synaptic blockade. These results support the hypothesis that PO/AH cold-sensitive neurons depend on synapses from nearby warm-sensitive neurons for their temperature sensitivity; whereas warm sensitivity is an independent property of certain PO/AH neurons.

The biophysical mechanisms underlying long-term potentiation (LTP) were investigated in identifiable and monosynaptic excitatory inputs to hippocampal neurons. The results provide the first insights into the conductance changes that are responsible for the expression of LTP. Both current- and voltage-clamp measurements of the mossy fiber synaptic response in pyramidal neurons of region CA3 were made with a single-electrode-clamp system. The excitatory postsynaptic response was pharmacologically isolated by bathing hippocampal slices in saline containing 10 microM picrotoxin, which blocks the synaptic inhibition that normally accompanies the experimentally evoked mossy fiber response. LTP was induced by tetanically stimulating the mossy fiber input for 1 s at 100 Hz. Before and 20 min to 1 h after inducing LTP, we attempted to measure the mean excitatory postsynaptic potential (EPSP) amplitude, intrasomatic current-voltage relationship to a step (RN) or alpha function (AN) current waveform, membrane time constant (tau m), spike threshold (T50), peak excitatory postsynaptic current amplitude (IP), synaptic conductance increase (delta G), and synaptic reversal potential (VR); but adequate assessments of all eight of these were not always obtained for every cell that was studied. The induction of LTP increased the mean (+/- SE) EPSP amplitude form 10.5 +/- 1.4 mV during the control period to 16.8 +/- 2.4 mV after the induction of LTP (n = 14; P less than 0.05). This change was not accompanied by increases in the mean value of RN (63 +/- 11 M omega before and 61 +/- 11 M omega after induction; n = 8; P greater than 0.05); AN, which approximates the effective synaptic input resistance at the soma (10.0 +/- 1.50 M omega before and 10.5 +/- 1.60 M omega after; n = 10; P greater than 0.05); or tau m (22 +/- 2 ms before and 20 +/- 2 ms after; n = 8; P greater than 0.05). There was no significant change in T50, which was also assessed with an alpha function current waveform (1.48 +/- 0.11 nA before and 1.49 +/- 0.10 nA after; n = 6; P greater than 0.05). The mean value of IP increased from 1.1 +/- 0.2 nA during the control period to 1.8 +/- 0.3 nA after inducing LTP (n = 15; P less than 0.05). Similarly, delta G increased from 30 +/- 4 nS before to 47 +/- 4 nS after induction (n = 10; P less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

1. Intracellular neuronal activity was recorded in rat preoptic-anterior hypothalamic tissue slices. Thirty neurones were classified as warm sensitive, cold sensitive or temperature insensitive, based on their firing rate response to temperature changes. Seventy-seven per cent of the neurones were temperature insensitive, which included both spontaneously firing and silent neurones. Of all neurones, 10% were warm sensitive and 13% were cold sensitive. 2. Silent temperature-insensitive neurones had lower input resistances (126 +/- 21 M omega) than thermosensitive neurones (179 +/- 24 M omega). Regardless of neuronal type, however, resistance was inversely related to temperature. 3. Warm-sensitive neurones were characterized by a slow, depolarizing pre-potential, whose rate of rise was temperature dependent. This depolarizing potential disappeared during current-induced hyperpolarization, suggesting that intrinsic mechanisms are responsible for neuronal warm sensitivity. 4. Spike activity in cold-sensitive neurones correlated with putative excitatory and inhibitory postsynaptic potentials, whose frequency was thermosensitive. This suggests that cold sensitivity in these neurones depends on synaptic input from nearby neurones. 5. Like cold-sensitive neurones, action potentials of temperature-insensitive neurones often were preceded by short duration (less than 20 ms), rapidly rising pre-potentials, whose rates of rise were not affected by temperature. In some temperature-insensitive neurones, depolarizing current injection increased both firing rate (by 5-8 impulses s-1) and warm sensitivity, with pre-potentials having temperature-dependent rates of rise. We suggest that temperature-insensitive neurones employ two opposing, thermally dependent mechanisms: a voltage-dependent depolarizing conductance and a hyperpolarizing sodium-potassium pump.