Emerging histiotypic properties of cultured neuronal networks

Abstract

In the past decade, substantial progress has been made with substrate-integrated micro-electrode arrays and the growth of viable networks on such arrays. However, such progress has not been accompanied by the acceptance of dissociated (primary) neuronal cultures as reliable systems for pharmacological and toxicological studies. Given that such quasi-monolayer cultures provide strong long-term adhesion stability, optical access to major components of the network circuitry, and high signal-to-noise ratios for multi-site recording of spatiotemporal action potential (spike) patterns, it is essential that such systems be validated as representative of the parent tissue. In addition, action potential waveshapes can also be monitored quantitatively over long periods of time for assessment of channel pharmacology and statistical surveys of discharges from different neuronal compartments. This chapter summarizes characteristic tissue specificities of native activity and histiotypic pharmacological and toxicological responses in order to demonstrate that appropriate dissociation and culture maintenance techniques can generate spontaneously active networks with remarkable similarity to parent tissue responses in vivo. Work with neuronal tissue in culture can be classified as part of two major mechanistic domains: (1) receptor-dependent studies and (2) circuit-dependent studies. The second domain has received limited attention, however, the first domain is less complex, has been explored by several laboratories, and is supported by a rapidly growing database attesting to the histiotypic nature of network responses in culture. In this chapter, we show that different tissues from the murine CNS have different native activity states and may also differ quantitatively in their pharmacological responses. Nevertheless, the overall pharmacological responses agree well with in vivo data. These results imply that, compared to the parent tissue in situ, primary cultures retain the same general ratio of cells (neurons and glia), receptor properties, synaptic mechanisms, and inherent cellular spike generation characteristics. The lower synaptic density and shallow three-dimensional tissue layer do not seem to impair or alter the type and character of pharmacological responses. The complexities seen in behavioral phenomena such as tinnitus caused by quinine and salicylate, or metal excitotoxicity due to exposure to mercury, or intoxication induced by ethanol can be studied in vitro in a much more controlled and simpler environment. This provides an opportunity to not only analyze global activity changes brought about by various chemicals, but also offers a platform for careful and objective scrutiny of changes in complex spatiotemporal patterns, and enables quantitative investigations of cell culture correlates of various behavioral phenomena.