A microelectrode array (MEA) integrated with clustering structures for investigating in vitro neurodynamics in confined interconnected sub-populations of neurons

Abstract: Understanding how the information is coded in large neuronal networks is one of the major challenges for neuroscience. A possible approach to investigate the information processing capabilities of the neuronal ensembles is given by the use of dissociated neuronal cultures coupled to microelectrode arrays (MEAs). Here, we describe a new strategy, based on MEAs, for studying in vitro neuronal network dynamics in interconnected sub-populations of cortical neurons. The rationale is to sub-divide the neuronal network into communicating clusters while preserving a high degree of functional connectivity within each confined sub-population, i.e. to achieve a compromise between a completely random large neuronal population and a patterned network, such as currently used with conventional MEAs. To this end, we have realized and functionally characterized a Pt microelectrode array with an integrated EPON SU-8 clustering structure, allowing to confine five relatively large yet interconnected spontaneously developing neuronal networks (i.e. thousands of cells). The clustering structure consists of five chambers of 3mm in diameter interconnected via 800μm long and 300μm wide microchannels and is integrated on the MEA of 60 thin-film Pt electrodes of 30μm diameter. Tests of the Pt microelectrodes’ stability under stimulation showed a stable behavior up to 35,000 voltage stimuli and the biocompatibility was assessed with the cultures of dissociated rat''s cortical neurons achieving cultures’ viability up to 60 days in vitro. Compared to conventional MEAs, the monitoring of spontaneous and evoked activity and the computation of the Post-Stimulus Time Histogram (PSTH) within the clusters clearly demonstrates: (i) the capability to selectively activate (through poly-synaptic pathways) specific network regions and (ii) the confinement of the network dynamics mainly in the highly connected sub-networks. [Copyright &y& Elsevier]