Thesis defense by Maxime Poinsot 17/12/2024

Tuesday 17 December at 2:30 PM, Amphi CERIMED

Maxime Poinsot (SONIC team)

Standardized Modeling of Neuropathologies on a Chip Using a Multimodal Multiplexed Analysis Platform

Jury
Sophie HALLIEZ, LilNCog, Lille, Rapporteuse
Vincent STUDER, IINS, Bordeaux, Rapporteur
Stéphanie DESCROIX, Institut Curie, Paris, Examinatrice
Laurent MALAQUIN, LAAS, Toulouse, Examinateur
Sophie CHAUVET, IBDM, Marseille, Présidente du Jury
Eduardo GASCON GONZALO INT, Marseille, Directeur de thèse
Maxime CAZORLA, INT, Marseille, Co-Directeur de thèse
 William CESAR Fluigent, Paris, Invité

Abstract
Neural networks play a fundamental role in cognitive functions such as learning, memory, and decision-making. Their dysfunction is implicated in various neurodegenerative (e.g., Parkinson’s disease), neurodevelopmental (notably autism spectrum disorders), and psychiatric conditions. Among these networks, the cortico-striatal network holds a central position. It is characterized by specific unidirectional projections connecting cortical regions to the striatum, forming a complex topographical organization essential for motor control, motivation, and cognition.

However, the study of human neural networks is limited by current techniques. In vivo models allow the observation of neuronal alterations in a physiological context but offer limited access to precise cellular mechanisms and raise ethical concerns. In contrast, traditional in vitro models, while valuable for mechanistic studies, struggle to replicate the structural and functional complexity of neural networks, particularly their topographical organization.

Organs-on-a-chip, and more specifically brains-on-a-chip, offer an innovative solution. These microfluidic devices recreate the brain’s microarchitecture in a controlled environment, enabling the simulation of specific biological functions with high precision. Nevertheless, the specific topography of the cortico-striatal network has not yet been fully modeled. This thesis aims to develop a brain-on-a-chip model capable of accurately reproducing this complex organization to better understand the pathological mechanisms associated with neurodevelopmental disorders.

In parallel, a second objective, in collaboration with the company Fluigent, is to design an autonomous microfluidic system to manage this model in a multiplexed manner. This versatile platform will standardize the use of organs-on-a-chip, making the technology more accessible and better suited to the needs of both research laboratories and the industry.

The subthalamic nucleus (STN) has long been considered a motor brain structure and is the target for Deep Brain Stimulation (DBS) treatment of Parkinson’s Disease. However, work in rats has identified the STN as a target for reducing motivation for cocaine while increasing it for sucrose. STN DBS also curbs behaviors mimicking addiction, like compulsive drug seeking and loss of control over consumption. The contribution of STN and its network in reward processes remains to be elucidated. In this thesis, we showed that high-frequency DBS and STN inhibition can induce self-stimulation in rats, highlighting the role of STN in reward processing. Using calcium imaging and optogenetic inhibition of the various hyperdirect pathways linking the prefrontal cortex to the STN, we showed their involvement in reward-seeking behavior.