The effect of anesthesia on neuronal communication

One challenging aspect in the analysis of neuronal circuits is the lack of quantitative and objective measurements of network activity to be translated into functional states. For example, the clinical assessment of the consciousness state in a brain-injured, unresponsive patient can be hardly analyzed at the cellular and network level. General anesthesia employs different classes of molecules to modulate at various levels neuronal functional states. General anesthetics (GA) are known to progressively and selectively reduce consciousness, perception and motor control. In this work we have investigated in a simplified neuronal circuit the effect of GA on information transfer. The Shannon mutual information (MI) was used to evaluate how much the neuron response reflected the input stimuli versus its intrinsic variability, providing a statistical tool to dissect the contribution of spike timing to neural information transmission. The cerebellum granule cell (GrC), due to its limited number of excitatory inputs, can be used to calculate the Mutual Information (MI) and its variation during a perturbed state (e.g. under anesthesia). The MI was experimentally assessed by detecting action potentials elicited in response to specific inputs through whole-cell patch-clamp recordings in rat acute cerebellar slices (P18-24). In order to test the action of the application of GA, GABAergic currents elicited by inhibitory afferent connections were recorded. The action of GA (in particular Sevoflurane and Desflurane) increased (+120%) post-synaptic inhibitory currents (IPSCs) in less than 10 sec and was fully recovered in 30 sec. Furthermore, the action of GA was to markedly reduce the MI measured in control condition (-57.4%). This control condition was fully recovered after removal the anesthetics, therefore leaving unaltered neuronal activity. This approach will be applied to larger circuits and investigated with other techniques (e.g. Multielectrode array recordings or cellular imaging), moreover different concentration of anesthetics could lead to the identification of multiple functional states.