Rodent and Human Neuronal Networks: Analysis of Functional Maturation, Synaptic Transmission, and Spontaneous Activity in Vitro

The development of neuronal networks is a complex process that occurs during the embryonic and postnatal stages in mammals. Abnormalities in this process are known to lead to neurodevelopmental disorders. Spontaneous network activity, observed as repetitive network-wide bursts (NBs), is a hallmark of the functional maturation of developing neocortical networks in both sensory cortices in vivo and dissociated cortical cultures in vitro. NBs emerge in parallel with the formation of the network structure and synaptic transmission. Synaptic transmission between neurons is driven by excitatory AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) and NMDA (N-methyl-D-aspartate acid) receptors as well as inhibitory GABAA (γ-amino butyric acid type A) receptors and their neurotransmitters. How these three ionotropic receptors affect the initiation, propagation, and termination of postnatal cortical NBs is not understood in detail.

The first aim of this compilation thesis is to characterize the role of synaptic transmission in spontaneous network activity in postnatal rat cortical networks in vitro. The second aim is to reinforce the utilization of multiunit time-series data obtained in the first aim in data-driven in silico modeling. The third aim of this thesis is to study the differentiation of human neuroblastoma cells towards neuronal phenotypes and networks and to characterize the key indicators of network structure and synapses.

In this thesis, I first characterized the contributions of synaptic AMPA, NMDA, and GABAA receptors to the initiation, propagation, and termination of NBs in postnatal rat cortical cultures. I recorded network activity with the microelectrode array (MEA) technique and pharmacologically antagonized synaptic receptors at the end of the third week in vitro. I then analyzed the recorded activity using several methods. Furthermore, I developed a data analysis procedure that supports the fitting and validation of computational spiking network models to quantitatively reproduce experimental results and study the effect of synaptic transmission on network dynamics in silico. Finally, I assessed the effects of estradiol, cholesterol, brain-derived neurotrophic factor, and retinoic acid on the level of neuronal differentiation in human neuroblastoma cells by analyzing the number of proliferating nuclei, synaptic vesicles, and neurites in human cell networks at 10 days in vitro.

The results of this thesis show that each synaptic receptor type has a specific contribution to the dynamics of spontaneous network activity. Excitatory AMPA receptors initiate bursts by rapidly recruiting cells for network-wide activity, while NMDA receptors maintain the activity that has already been initiated. GABAA receptors inhibit the early onset phase of AMPA receptor-mediated activity and attenuate the termination phase of NMDA receptor-mediated bursts. The results clearly show an interaction between fast AMPA receptor-mediated excitation and GABAA receptor-mediated inhibition. In the presence of this interaction, the burst propagation patterns are rich and diverse. In the absence of this interaction, in contrast, bursts originate and propagate along the same pathways. The results of the thesis further describe how multiunit time-series data recorded by the MEA technique on rat cortical networks under pharmacology are analyzed to support the development of spiking network models. The results of this thesis also suggest that human neuroblastoma cells can be morphologically differentiated into neuronal phenotypes. Differentiation of these human cells increases neurite outgrowth and branching, neuron-neuron interactions, and the ability of cells to induce synaptic vesicle recycling. These findings are a prerequisite for the formation of proper neuronal network.

In conclusion, the studies presented in this thesis reveal for the first time how the three main mediators of synaptic transmission AMPA, NMDA, and GABAA receptors shape the dynamics of network activity, including the initiation, propagation, and termination of bursts in postnatal rat cortical networks in vitro. The data analysis presented in this thesis promotes the development of data-driven spiking network models. Such data-driven models can be used to study the complex dynamics of cellular mechanisms in network activity as well as to create and design new hypotheses that can be studied in vitro and in vivo. Finally, the results on human cell differentiation contribute to future studies comparing human neuroblastoma cell-based neuronal networks to rodent networks.