Dottorato in SCIENZE DELLA VITA

Titolo della tesi: Development of human iPSC-based in vitro systems for modelling GNAO1 disease and drug testing

GNAO1 encephalopathy is a rare genetic disease caused by pathogenic variants in the GNAO1 gene, which encodes for the α subunit of a heterotrimeric guanine nucleotide-binding protein (Go). This protein has a central role in the nervous system, regulating both synaptic activity and neurodevelopment. Patients affected by this pathology develop various combinations of clinical symptoms, including developmental delay, hypotonia, epilepsy and hyperkinetic movement disorder. Currently, effective treatments are still lacking, and this is also due to a poor understanding of the pathological mechanism leading to GNAO1 disorders. The aim of this study is to create an iPSC-based platform for GNAO1 disease modelling and drug testing, taking advantage of both cortical differentiation protocol and micropatterning system. We obtained a collection of three different GNAO1 pathogenic variants with their respective isogenic controls, thanks to the use of CRISPR-Cas9 gene editing. Initially, we characterized how p.G203R affects cortical development since it is one of the most represented among GNAO1 pathogenic substitutions and the most severe. We observed altered expression of neural and WNT-related genes and concomitant defects in the formation of neural rosettes at an early stage of differentiation. At later time points, p.G203R neurons showed an imbalance between neural progenitors, neurons and astrocytes. Functional analysis exhibited lower basal intracellular calcium levels ([Ca2+]i), a reduction of spontaneous activity, and a smaller response to several neurotransmitters. Collectively these finding suggest defective differentiation processes leading to significant impairments of functional maturation and neuronal activity. Interestingly, we confirmed the early defects also in other two pathogenic variants, p.S47G and p.E246K. To the best of our knowledge, how these two mutations affect neurodevelopment has been never explored before. These data collectively suggest that different pathogenic variants could share common neurodevelopmental defects, may related to altered WNT signalling. We have then established micropatterning models of gastrulation and neurulations. Gastruloids carrying the p.G203R and the p.E246K showed a strong reduction of ectodermal area, the germ layer precursor of the nervous system, while p.E246K neuruloids exhibited an aberrant neuroectodermal arrangement. Interestingly, we observed that caffeine treatment was able to selectively increase the SOX2+ area in gastruloids carrying GNAO1 pathogenic substitutions but not in the wild-type control. This provide evidence that micropatterning systems, which are highly fast, scalable and reproducible, could be used for future drug screening. Once new drug candidates are identified, their efficacy could be then tested on cortical neurons.