Section: New Results

From the microscopic to the mesoscopic scale

Participants: Laure Buhry, Axel Hutt Amélie Aussel, Nathalie Azevedo Carvalho.

In collaboration with Radu Ranta (univ. Lorraine), Dominique Martinez (CNRS), Abderrahman Iggidr (Inria), Patrick Hénaff (univ. Lorraine), Beate Knauer and Motoharu Yoshida (Ruhr university) and LieJune Shiau (university of Houston)

Memory & sleep

We proposed a detailed anatomical and mathematical model of the hippocampal formation for the generation of sharp-wave ripples and theta-nested gamma oscillations [1], [7]. Indeed, the mechanisms underlying the broad variety of oscillatory rhythms measured in the hippocampus during the sleep-wake cycle are not yet fully understood. We proposed a computational model of the hippocampal formation based on a realistic topology and synaptic connectivity, and we analyzed the effect of different changes on the network, namely the variation of synaptic conductances, the variations of the CAN channel conductance and the variation of inputs. By using a detailed simulation of intracerebral recordings, we showed that this model is able to reproduce both the theta-nested gamma oscillations that are seen in awake brains and the sharp-wave ripple complexes measured during slow-wave sleep. The results of our simulations support the idea that the functional connectivity of the hippocampus, modulated by the sleep-wake variations in Acetylcholine concentration, is a key factor in controlling its rhythms. A presentation of this work received a best poster award at the 27th annual Computational Neuroscience Meeting, CNS'2018.

Parkinson's network

Using a Hodgkin and Huxley's model, we modeled pathological oscillations of Parkinson's disease in basal ganglia. Our hypothesis was that the pathological oscillations are generated by a MSN-GPeA-FSN circuit and then transferred to the STN by the GPeP. The normal state is represented by neurons of the MSN emitting at a frequency of $\sim 1Hz$ and the parkinsonian state is represented by MSN neurons that emit at a frequency of $\sim 15Hz$. Our results correspond to the experimental results of the rat. In the normal state, there is no visible synchronization, whereas in the parkinsonian state pathological synchronizations are formed at the level of the circuit. There is even a rhythm that is created, that is to say, the neurons of MSN emit first then those of the FSN and then those of the GPeA and so on. We performed large-scale simulations of 1.5 million neurons in the basal ganglia in rats using the Grid5000 (parallel computation platform).