A human iPS cell-based model of medulloblastoma demonstrates co-operativity between SHH signalling and mutation in an epigenetic modifier

Objective: Medulloblastoma (MB), the most common malignant pediatric brain tumor, is comprised of four subgroups (WNT, SHH, Group 3, Group 4).  While recurrent single nucleotide variations is rare in MB, mutation of... [ view full abstract ]

Objective: Medulloblastoma (MB), the most common malignant pediatric brain tumor, is comprised of four subgroups (WNT, SHH, Group 3, Group 4).  While recurrent single nucleotide variations is rare in MB, mutation of epigenetic modifiers is frequent, suggesting a mechanism for tumorigenesis. Genetically engineered mouse models (GEMMs) recapitulate some aspects of MB, but fail to replicate the epigenetics characteristic of human MB. Therefore, we aimed to develop the first human stem cell based model of MB using iPSC derived from Gorlin syndrome (GS), a disease that predisposes patients to SHH MB due to heterozygosity for PTCH1.
 
Methods: We differentiated iPSC from healthy (WT) or GS patients toward hindbrain neuroepithelial stem (NES) cells, a presumed cell of origin of MB. We then electroporated the NES cells with RNA-guided FokI Nucleases (RFN, CRISPR/dCas9-FokI) targeting GSE1 (formerly KIAA0182), a gene encoding an epigenetic modifier and co-mutated with PTCH1 in MB. These NSC were implanted orthotopically into the hindbrains of mice.
 
Results: By 92 days post-implantation, 80% of mice with GS NSC mutant for GSE1 developed tumors with histology consistent with MB. In contrast, GS NSC with wild-type GSE1 developed the first tumor at longer latency (> 140 days) while WT NSC have not generated tumors (> 300 days). 
 
Conclusion: Our human iPSC-based model of MB demonstrates a cooperative effect between SHH signaling and a mutation in an epigenetic modifier.  We are currently investigating similarities and differences between our human MB with GEMMs and elucidating the role of epigenetic alterations in MB tumorigenesis.