Human neuronal networks on micro-electrode arrays are a highly robust tool to study disease-specific genotype-phenotype correlations in vitro

Abstract

Summary Micro-electrode arrays (MEAs) are increasingly used to characterize neuronal network activity of human induced pluripotent stem cell (hiPSC)-derived neurons. Despite their gain in popularity, MEA recordings from hiPSC-derived neuronal networks are not always used to their full potential in respect to experimental design, execution, and data analysis. Therefore, we benchmarked the robustness of MEA-derived neuronal activity patterns from ten healthy individual control lines, and uncover comparable network phenotypes. To achieve standardization, we provide recommendations on experimental design and analysis. With such standardization, MEAs can be used as a reliable platform to distinguish (disease-specific) network phenotypes. In conclusion, we show that MEAs are a powerful and robust tool to uncover functional neuronal network phenotypes from hiPSC-derived neuronal networks, and provide an important resource to advance the hiPSC field toward the use of MEAs for disease phenotyping and drug discovery.

Real-Time Monitoring of Levetiracetam Effect on the Electrophysiology of an Heterogenous Human iPSC-Derived Neuronal Cell Culture Using Microelectrode Array Technology

“…Microelectrode arrays (MEAs) are non-invasive biosensors that can reliably detect the electrophysiological spontaneous activity of thousands of neurons [ 8 ]. MEAs consist of a high number of electrodes that can be of different materials (e.g., gold, poly(3,4-ethylenedioxythiophene) polystyrene sulfonate—PEDOT) [ 9 ] usually implanted at the base of tissue culture wells that allows to record (and eventually stimulate) the activity of cells capable to generate electrical signals (e.g., neurons, cardiomyocytes) [ 10 , 11 , 12 ] in a label-free manner and in real-time. In neuronal cell culture, MEAs allow one to record the Extracellular Action Potentials (EAP, also called spikes) [ 10 , 13 ].…”

Functional Characterization of Human Pluripotent Stem Cell-Derived Models of the Brain with Microelectrode Arrays

“…The drawbacks of the Ngn2 and Dlx2/Ascl1-based differentiation methods are that they produce very simplistic models with no other cell types and the produced human neurons are typically cultivated on rat astrocytes. Still, the speed, efficacy and cell-line-to-cell-line repeatability of the methods [ 7 ] highlight their feasibility in mechanistic and pharmacological studies of human neuronal networks.…”

“…Several approaches to the differentiation of hiPSCs into cells with specific properties have been proposed over the years, allowing the formation of different cells of the nervous system, including astrocytes, microglia and several neuronal subpopulations [ 5 , 136 , 137 , 138 , 139 , 140 ]. Considering all of the above, it is evident that hPSC-derived neuronal 2D-cultures can produce both mixed and homogenous populations of neurons (as well as glia) that recapitulate basic firing and bursting activity, as well as the functional connectivity of human neuronal networks on MEA [ 7 , 128 , 135 ]. However, it is difficult to compare the exact spike and burst counts to in vivo studies because of the lack of SUA analysis in studies with hPSC-derived cultures.…”

“…While organoids and other 3D models can provide a more complex model of the human brain, hPSC-derived 2D neural cultures on MEAs will continue to provide relevant information on the function of human neuronal networks in the future. This is shown by the large number of recent publications utilizing the approach to gain insights into pathological network function [ 7 , 134 , 135 , 143 , 176 , 179 , 180 ]. However, while it is true that 2D models have considerably less detectable neurons per electrode than corresponding 3D models and the in vivo brain, it would be beneficial to use signal-to-noise ratio-enhancing methods combined with spike sorting to discriminate SUA from 2D models.…”

“…This not only widened the availability of hPSCs and reduced the associated ethical burden, but significantly facilitated the development of personalized medicine as the human-induced pluripotent stem cells (hiPSCs) could be differentiated into patient-specific in vitro disease models of practically any tissue or organ, including the brain [ 5 ]. Lately, research has been focused on the self-organizing three-dimensional (3D) hPSC-derived brain models known as organoids [ 6 ], even though two-dimensional (2D) cultures of human neural cells continue to serve as functional models of the brain [ 7 ].…”