Engineering network complexity to decode emergent neuronal dynamics and early dysfunction in ALS

Neurodegenerative diseases such as Amyotrophic Lateral Sclerosis (ALS) arise from progressive alterations in neural network organization that precede overt neurodegeneration. Identifying the mechanisms behind these early dysfunctions remains challenging, as neural behavior emerges from cross-scale organizational interactions. In vitro models simplify this complexity while preserving essential physiological features and are therefore critical for dissecting the principles underlying neural network dynamics and their pathological disruption. In this thesis, I developed a progressive experimental framework based on in vitro models to investigate how neural network topology shapes the emergence of functional activity and how deviations from these principles may characterize early disease states. First, I developed two-dimensional (2D) engineered modular cortical networks to examine how the balance between segregation and integration influences network maturation. By controlling the timing of structural coupling between modules, I showed that developmental timing critically determines the emergence of coordinated network organization. I then extended these principles to scaffold-free three-dimensional cortical–hippocampal assembloid-like systems that mimic the spatial and cellular organization found in vivo. In this modular three-dimensional (3D) system, neuronal activity evolved from predominantly local synchronization to coordinated inter-regional dynamics, accompanied by the emergence of structured oscillatory activity. The establishment of these reliable physiological models provided a necessary reference for identifying early pathological alterations. Within this context, I developed and characterized an in vitro ALS murine model, which revealed an early divergence from physiological maturation — marked by transient hyperexcitability, impaired astrocytic support, and fragmented network topology — ultimately leading to progressive network instability. Together, these findings highlight how engineered in vitro systems provide a controlled framework to identify fundamental principles of physiological network organization and to detect early circuit-level alterations associated with neurodegenerative diseases.