INTRINSIC INPUT–OUTPUT PROPERTIES OF LAYER 5 INHIBITORY NEURONS IN <EM>EX VIVO</EM> HUMAN NEOCORTICAL SLICES

Alterations in excitation–inhibition balance are hypothesized to contribute to epileptic hyperexcitability, yet underlying cellular mechanisms in human neocortex remain poorly understood. Recent human studies indicate that seizure initiation and propagation engage cortical layers in a layer-specific manner, motivating investigation of defined microcircuits. In this context, layer 5 (L5) inhibitory neurons regulate pyramidal neuron excitatory output, shaping local processing and long-range cortical communication.
We hypothesize that L5 inhibitory neurons in epileptic tissue exhibit altered intrinsic input–output transformations, including changes in gain, adaptation dynamics, or sensitivity to temporal input structure. To test this, L5 interneurons are recorded with whole-cell patch-clamp in acute human neocortical slices from patients undergoing surgical resection for drug-resistant epilepsy and brain tumors, enabling comparison across different diagnostic groups.
Intrinsic excitability is assessed with step current injections to characterize firing patterns, gain, and spike-frequency adaptation. Frequency-dependent responsiveness is tested using chirp stimuli to evaluate resonance and filtering properties. To probe input–output transformations under physiologically-relevant conditions, neurons are additionally stimulated with sinusoidal inputs combined with stochastic fluctuations, enabling assessment of dynamic gain and temporal sensitivity. Recorded neurons are biocytin-filled and reconstructed to relate electrophysiological properties to morphology.
Preliminary results show distinctive input-output properties of L5 inhibitory interneurons, paving the way towards a better understanding of potential cell-autonomous contributions to disrupted excitation–inhibition balance in epileptic networks. By dissecting the intrinsic input–output properties of human L5 inhibitory neurons, this work contributes a single-cell framework complementary to network-level observations and supports a multiscale understanding of human neocortical dysfunction.