Highly Ordered Neuronal Networks Implemented on Nanowire-Based Electrical Devices


The use of artificial, pre-patterned neuronal networks in vitro is a promising approach in the study of the development and dynamics of small neural systems, for the purpose of understanding the basic functionality of neurons and eventually of the brain. The present work describes a high-fidelity and robust method for controlling neuronal growth on solid substrates such as silicon wafers and polished glasses, enabling us to obtain mature and durable neural networks of individual cells of designed geometries. It has a number of advantages over other related techniques which have been reported, mainly as a result of its high yield and remarkable reproducibility. This approach, based on a highly-controlled surface-chemistry sequence, allows forming functional, tailor-made neural architectures with a micrometer-resolution partition. The main achievement of this methodology is the capability of creating large-scale neuronal networks of low cell densities that develop intact, typical neurites and mature electrically-active synapses with a relatively long-term survival of up to four weeks.


Neuronal Networks1

Cortical neuron cells aligned on a nanowire-device array. (A) Optical image of a 3 days old neuronal network chemically patterned according  to the NW based electrodes. (B-D) SEM images of individual cells crossing a single nanowire device (denoted by the white arrows).

 Neuronal Networks2Schematic illustration of the chemical surface patterning. (1) Creation of hydrophobic surface that repeals neuronal cells. (2) Formation of squares pattern. (3) Peeling of the hydrophobic layer from the square regions. (4) Coverage with poly-lysine for cells attachment. (5) Chemical pattern of poly-lysine squares surrounded by hydrophobic flourosilane layer.


Conclusions: In this work cortical and hippocampal neurons, 99.9% glia-free, have been successfully cultured and developed with intact neuronal physiology up to 3 weeks in defined and highly ordered large-scale network, forming functionally mature network. We were able to overcome the low survival rate that naturally is accompanied to low cell densities and the absence of glia cells. We were also able to successfully position the cells directly on nano-pillars and grow them on NW-FET devices with registration to the nanowires underneath, in the form of individual cells and as a connected network. This would allow in the future to explore at high resolution the electrical activity of the neurons at the level of a single cell and as an operating network and will assist to provide new and important insights to the field of neurobiology