Development of neuronal circuits in human brain organoids

Stem cells have the remarkable potential to differentiate into mature cell types, like neurons. In vitro culture methods enable stem cells to self-organize into three-dimensional cell assemblies referred to as organoids. Because brain organoids generated from stem cells of human origin recapitulate the embryonic development of the central nervous system, they offer an unprecedented opportunity to examine the unique features of the human brain and psychiatric disorders. This research project aims to understand the generation and properties of the neuronal circuits in human brain organoids at three functional levels. First, we will identify which types of neurons coexist at various stages of brain organoid development. Second, we will systematically investigate synaptic connections between different cell types to understand excitatory and inhibitory microcircuits. Third, we will study whether neuronal activity evoked by an external stimulation affects the structure and dynamics of developing networks. Functional methods of neuroscience have contributed to our understanding of the brain in animal models. When applied to in vitro brain organoids, intracellular techniques and silicon microelectrodes operate without complex or invasive surgical procedures. Thus, electrophysiological techniques will bring insight into the molecules, the cell types, and the functional circuits involved in neuronal communication. Also, neuronal stimulation via modern optogenetic methods will allow us to evaluate how early embryonic activity affects the development of neuronal networks. Because brain organoids will be bioengineered based on of human stem cells, the knowledge acquired will offer a direct translation of how the human brain functions and its associated diseases. The novelty of this project resides in studying developing human neuronal networks by bridging the most advanced efforts in the fields of circuit neuroscience and stem cell biology. Detailed and functional characterization of developing neuronal circuits in human brain organoids will increase our understanding of the unique features of our brains. Importantly, the pathogenesis of neurodevelopmental disorders can be understood if we gain information about the cell types, synaptic connections and circuit properties of human networks. Thus, deriving brain organoids from the stem cells of patients will open a door for the molecular identification of the mechanisms of neuropsychiatric diseases, and assist to design strategies for early therapeutic interventions.

We have developed a computer model to predict the development of epilepsy using electrical signals obtained from human in vitro preparations. To train the model in epileptic and non-epileptic forms, we used stem cells from epileptic patients and generated 3D differentiated neural tissue - brain organoids - that can be cultured in vitro. Stem cells were genetically modified to correct genes associated with epilepsy, thus allowing us to compare epileptic brain organoids with those genetically identical without the gene mutation. We engineered a miniaturized customized system to obtain electroencephalogram (EEG) signals in brain organoids. The system comprises 128 micro-electrodes allocated in less than one square millimeter permitting high resolutive invasive measures, which is impossible in standard clinical practice. We tested brain organoids from two patients with a genetic form of epilepsy and compared them to organoids whose cells were corrected for the epileptic gene. Organoids showed periodic electrical signals associated with spontaneous excitatory network events. Excitatory events were overrepresented in organoids where the epileptic gene was expressed. Because network hyperexcitability is a landmark of epilepsy in humans, we used the enhancement of excitatory events as a biomarker of epilepsy, just as in clinical practice. Thus, we could rapidly test epileptic drugs of clinical relevance when using brain organoids to model the disease. Accordingly, drugs blocking neural function or affecting excitatory communication abolished electrical signals, while drugs affecting inhibitory transmission had no effect. These findings suggest that this epileptic form can be alleviated using clinical treatments that target excitatory mechanisms and predicts that traditional epileptic drugs that affect inhibitory transmission may have a minor effect. To generalize to different epilepsy forms or study other epileptic genes, we trained a neural network with electrical events from epileptic and genetically corrected organoids. When training the network model with organoids from one of the patients, the network could accurately predict epileptic cases from the second patient. Furthermore, the model had a low false-positive rate because it discarded epileptic features in electrical signals from organoids derived from human stem cells in non-epileptic patients. In summary, we built a computer-classification algorithm that identifies electrical epileptic patterns in human brain organoids. Because it uses signals from human in vitro preparations, it may assist in the early detection of epilepsy and a personalized selection of appropriate pharmacological treatments.