A long-standing goal of neuroscience is a theory that explains the formation of the minicolumns in the cerebral cortex. Minicolumns are the elementary computational units of the mature neocortex. Here, we use zinc oxide nanowires with controlled topography as substrates for neural-cell growth. We observe that neuronal cells form networks where the networks characteristics exhibit a high sensitivity to the topography of the nanowires. For certain values of nanowires density and fractal dimension, neuronal networks express small world attributes, with enhanced information flows. We observe that neurons in these networks congregate in superclusters of approximately 200 neurons. We demonstrate that this number is not coincidental: the maximum number of cells in a supercluster is limited by the competition between the binding energy between cells, adhesion to the substrate, and the kinetic energy of the system. Since cortical minicolumns have similar size, similar anatomical and topological characteristics of neuronal superclusters on nanowires surfaces, we conjecture that the formation of cortical minicolumns is likewise guided by the interplay between energy minimization, information optimization and topology. For the first time, we provide a clear account of the mechanisms of formation of the minicolumns in the brain.
opeNThere are amendments to this paper Scientific RepoRts | (2019) 9:4021 | https://doi.org/10.1038/s41598-019-40548-z www.nature.com/scientificreports www.nature.com/scientificreports/ architectures can direct, control and, in some cases, improve the organization of simple elements into clusters [38][39][40][41][42][43] . Nano-topography is especially important in guiding cell fate at the bio-interface, and is therefore of interest in neural tissue engineering, bio computing, biosensors operations and neural cell based sensors, the diagnosis and analysis of neurodegenerative disorders, neural development [44][45][46] . In previously reported studies 39,47,48 , we examined patterns of neuroblastoma N2A cells on meso-porous silicon. We observed that N2A cells on a surface with nano-scale motifs display an increased ability to create patterns in which the nodes of the patterns form highly clustered groups and the elements of the groups are connected by a finite, and generally low, number of steps. Networks with similar characteristics are named small world networks 49,50 . In ref. 46 we demonstrated that neural networks with a small world topology on rough silicon substrates feature enhanced signal propagation speed and computational capabilities compared to regular grids of the same size on flat surfaces.Here, we present zinc oxide nanowire surfaces with variable nanowire density that achieve a tight interface with hippocampal neurons to direct their assembly into clusters. Resulting neuronal networks show a high sensitivity to the geometrical characteristics of the nanowires. For certain combinations of fractal dimension and nanowires density, neurons accumulate into clusters with ~200 neurons per clu...
