Properties of associative long-lasting potentiation induced by cellular conditioning in the motor cortex of conscious cats

Abstract

Mechanisms of long-lasting potentiation of synaptic responses induced in the thalamocortical and recurrent collateral pathways of the pyramidal tract were studied in intracellular recordings from the motor cortex of unanesthetized, chronically implanted cats. The observations provide the first description of long-lasting potentiation in the unanesthetized neocortex in vivo. Monosynaptic excitatory postsynaptic potentials of 2–5 mV in amplitude were evoked as test responses by stimulation of the pyramidal tract and thalamic ventrolateral nucleus at 0.1–0.5 Hz frequency. Pressure microinjections of drugs and ions were also performed during intracellular recordings. In the first series of experiments, test synaptic responses were paired with intracellular current injection-induced action potentials at an interstimulus interval set between 0–200 ms and 0.1–0.5 Hz frequency. Pairings (30–100 ×) induced long-lasting potentiation of the test responses in 58% of cells. The increased synaptic responses typically initiated action potentials and their potentiation usually lasted over the period of recordings. Increases in amplitude of synaptic responses were not correlated with statistically significant changes in electrical membrane properties (resting potential, input resistance, time constant, spike threshold) or parameters of action potentials and their afterpotentials. The failure to induce increases in synaptic efficacy by unpaired stimuli (pseudoconditioning) demonstrated the associative property of the long-lasting potentiation. In a second series of experiments, differential cell conditioning was employed. This paradigm induced long-lasting potentiation of the explicitly paired synaptic response without noticeable modification of unpaired or pseudorandomly paired synaptic responses tested conjointly in the same neuron. These observations demonstrated the input-specificity of long-lasting potentiation. In a third series of experiments, sub-threshold depolarizing current pulses were summated with synaptic responses to induce firing in the recorded neuron during pairing. Long-lasting potentiation occurred in 55% of the summated synaptic inputs. Pseudoconditioning did not induce synaptic potentiation in these cells. In a fourth series of experiments, conditioning was employed in neurons in which firing activity was suppressed by an intracellularly injected lidocaine derivative. Long-lasting potentiation was induced in 50% of the attempts when synaptic responses were paired with current-induced depolarizations ≫ 30 mV. These results suggest that postsynaptic induction of long-lasting synaptic potentiation can be successful in the absence of postsynaptic sodium spikes in neurons of the motor cortex in vivo. In a fifth series of experiments, homosynaptic high-frequency tetanization (80–200 Hz for 5–15 s) was applied to the thalamocortical and recurrent pyramidal afferents. We were able to induce long-lasting potentiation by tetanization only in the pyramidal pathways. This is in contrast with low-frequency pairings that produced input-specific synaptic enhancement in both the thalamic and pyramidal pathways. The observations suggest that some mechanisms of the induction process of long-lasting potentiation may be different in these two pathways of the motor cortex.