Complementary in vitro and computational modelling for the investigation of interacting neuronal networks

Objective. Many higher brain functions are attributed to the cerebral cortex, characterized not only by many neurons but also by an extensive connectivity with other brain areas. The finely regulated interactions between them are suggested to be at the basis of the rise of complex patterns of activity. Due to the complexity of the system itself, unravelling the mechanisms underlying brain functions requires us to devise realistic but simplified models that allow understanding how cells of different brain circuits interact. The goal of my work fits in this framework. The objective is to create workable models of interacting neuronal networks which allow the investigation of the effects of the intrinsic features of the brain on its emerging dynamics. Approach. To this end, I developed both experimental and computational models. First, I focused on dissecting the role of different brain key features, such as modularity, heterogeneity, and three-dimensionality, by creating different scaffolds. Then, I devised a new polymeric device that allows combining these elements for the study of 3D heterogeneous interacting populations with a realistic connectivity. Complementary, I developed a computational model to in part reproduce the in vitro experimental findings to try to infer the effect of the non-directly observable features and interaction between them thanks to the controllability of in silico systems. Main results. Firstly, I studied the effect of a modular network organization by plating cortical and hippocampal neurons over Micro-Electrode Arrays: the results highlighted important differences in how electrical signals are transmitted in different brain regions. Therefore, it was not surprising that interconnecting heterogeneous cultures with these different intrinsic characteristics would produce a very different modulation of their activity. Cortical activity undergoes a complementary modification in the cortico-hippocampal and the cortico-thalamic microcircuit. And yet a different effect can be observed in cortical dynamics if three-dimensional structures are considered: the temporal distribution of cortical events is quite influenced by the type of 3D scaffold. Secondly, my in silico results suggest that the introduction of three-dimensionality induces a global reduction in both firing and bursting rates compared to a 2D model. It also allowed to find out that the effects induced by a 3D organization of the cells is somewhat mitigated by different connectivity motifs of the network. Finally, the foundations for a more realistic model, comprehensive of all the listed key elements and a more in vivo-like connectivity, were laid out both experimentally and computationally. Significance. The proposed in vitro and in silico models could help the study of the biological mechanisms responsible for cognitive capabilities and the breakdown of these mechanisms in brain diseases. In the framework of neuroscience, a great effort is focused on creating “data ladders” to link the information available at different scales of organization to have an increasingly complete picture of how the brain computes. Reliable in vitro models will be a powerful tool to add information at the microcircuit level. Complementarily, computational models will allow manipulations and recordings that are impossible and/or ethically problematic otherwise, enabling a thorough investigation of the causal mechanisms to the rise of brain’s activity and pathologies. The available data coupled to in silico models could in fact allow the reliable prediction of the key parameters underlying brain phenomena. Moreover, the knowledge acquired with these techniques could be exploited for brain-inspired technologies, with a positive repercussion on industry and society.