P04.09.b uncovering group 4 medulloblastoma through human cerebellarorganoids

Abstract BACKGROUND Central nervous system tumors are the second most prevalent cancers in the pediatric population and among them, medulloblastoma (MB) is the most common and malignant one. MB is divided in four major subgroups: WNT, SHH, group 3 and group 4. Despite this classification, recent studies have highlighted many genetic features shared between the last two subgroups making the patients’ stratification less distinct. Particularly, group 4 MB is the most common subtype, but the least understood. Single-cell transcriptomics studies identified Unipolar Brush Cells (UBCs) as putative cell of origin of group 4 MB subtype. Despite G4 MB being the most common subtype, there is still a paucity of models. In this scenario, my aim is to implement and adapt the cerebellar organoids experimental design already established by Ballabio et al. for G3 MB towards G4 MB modeling. MATERIAL AND METHODS We generated cerebellar organoids starting from human pluripotent stem cells. Thanks to the PiggyBac/transposon system and the CRISPR/CAS9 technology we electroporated organoids achieving putative genes’ overexpression and downregulation. Organoids were live-imaged weekly checking for proliferative cells clusters formation and, after having reached the tumorigenic phenotype, they were engrafted in nude mice. The classification of the tumor obtained is based on DNA methylation analysis. Immunostainings were performed on mice cerebella and organoids to identify their cell types. RESULTS Because of the aggressiveness and the lack of knowledge of G4 MB, we focused on the development of a cerebellar organoids-based model genetically modified to achieve the tumorigenic phenotype. We studied the best time window in which Tbr2+ (UBCs marker) cells are most present in organoids, deciding to electroporate them at day 70 of differentiation. Based on RNA-seq data of G4 MB patients, organoids were modified with different combinations of genetic alterations. Interestingly, some of them showed the appearance of hyperproliferating engineered cells forming clusters. However, only the co-overexpression of MYCN and Otx2 developed a tumor when engrafted in nude mice. The following DNA methylation analysis attributed the tumor to G3 MB. Hence the need to clone the hTbr2 promoter to guide the expression of oncogenes through a Cre-LoxP system in the UBCs neural progenitors. We confirmed the new promoter specificity through an immunofluorescent quantitative analysis and then performed the genetic screening mentioned above to compare the outcome with our previous results. CONCLUSION Despite the timing, the mutational screening performed did not show the expected outcome: since G3 and G4 have common genetic features, we can’t rely on the strong ubiquitous non-specific promoter (CAG). Hence, we suppose that the hTbr2 promoter-guided gene expression could be the promising way toward G4 MB modeling.