Brain Slice Recordings on High-Density Microelectrode Arrays, from Single-Unit Activity Tracking to Network Investigations

Abstract

Neurons are the smallest building blocks of the information processing system in the brain. They intercommunicate via electrical signals called action potentials (APs). From single-neuron APs to a functional brain module, information needs to be integrated across multiple network scales. Local field potentials (LFPs) summate electrical activity over a local neural network and can be detected simultaneously with APs in extracellular recordings. LFPs can propagate across neural networks via the respective neuronal connections to form oscillations. Brain oscillations play important roles in integrating information from individual neurons to the network level by increasing neuronal synchronization at specific moments. Many deficits in brain oscillations are correlated with brain diseases, epilepsy being one of those diseases. Brain slices are effective models to study neuronal activity at different levels because they largely preserve the neural circuits and local functions of the brain regions from which they were obtained. On the other hand, brain slices can also be cultured over weeks, thereby providing a comparably realistic environment to observe neuronal activity over extended durations. Recently developed complementary-metal-oxide-semiconductor-based microelectrode-array (CMOS-MEA) technology, featuring high spatial electrode density and a large array area, provides advantages in identifying and isolating AP signals from specific neurons, even within brain slices exhibiting high neuron density. Additionally, CMOS-MEAs enable to record extracellular activity across different network scales, from individual neuronal APs to LFPs propagating across larger network areas. This thesis firstly reviews relevant scientific applications of existing CMOS-MEA technologies with a focus on applications in neuroscience research. The second part describes a label-free extracellular microelectrode-array-based method to track single-unit neuronal activity in organotypic hippocampal-slice cultures over weeks. The third part includes a method to investigate the spatiotemporal dynamics of epileptic seizures with initial AP and LFP activity emerging prior to the epileptic seizure onset and propagating as epilepsy oscillations across the slice regions. High-spatiotemporal-resolution electrical activity images were generated through the corresponding MEA recordings, and were used to track specific single-unit neuronal “footprints” over weeks, and to observe AP-LFP activity during epileptic seizures. The results show that (1) single-unit neuronal activity remains relatively stable within organotypic slice environments, which confirms the possibility to study chronic impacts of pharmacological or genetic modifications on individual neurons within slice preparations; (2) epileptic seizures most likely originate from the hippocampal cornu ammonis 3 (CA3) regions and that increased electrical activity starts a few hundred milliseconds before epileptic seizure onset. Additionally, the regions affected by seizure activity remained consistent during seizure propagations. These observations demonstrate the potential of the new method to investigate the dynamics of epileptic seizures with detailed spatiotemporal information on electrical activity and promote future research on initiation of epileptic seizures and epilepsy predictions.