Generating hiPSC-derived ATRT models to investigate cell of origin and identify therapeutic vulnerabilities - Project Summary Atypical teratoid rhabdoid tumors (ATRTs) are aggressive pediatric brain cancers that lack standardized treatment regimens. After radiotherapy treatments, survivors suffer from long-term neurocognitive defects. Thus, there is a critical need for dissecting the underlying biology of ATRTs to identify novel therapeutic strategies. ATRTs are driven by a differentiation block caused by biallelic inactivation of SMARCB1 with no other recurrent mutations, resulting in dysregulated cells that fail to terminally differentiate and consequently acquire oncogenic states. Naturally, neural progenitor cells (NPCs) are suspected to be the cells of origin for ATRTs, as their stalled differentiation has been implicated in pediatric brain tumorigenesis. However, ATRTs present unique features that imply a cell of origin that is not restricted to the central nervous system (CNS). Despite being driven by only SMARCB1 loss, ATRTs are comprised of three molecular subgroups, with different clinical outcomes, of which only one displays neural features: ATRT-Sonic hedgehog (SHH). Furthermore, ATRTs are molecularly identical to extracranial malignant rhabdoid tumors (MRTs). As a potential explanation for these nonneural features, the neural crest cell (NCC) is a putative cell of origin for both ATRTs and extracranial MRTs, as it emerges from the neuroectoderm but then migrates throughout the embryo. Therefore, the hypotheses of this proposal are that SMARCB1 loss interacts with cell identity and anatomical location during tumorigenesis, and that there are targetable vulnerabilities that can enable SMARCB1-depleted cells to overcome their differentiation block. Previously, the Furnari lab engineered human induced pluripotent stem cells (hiPSCs) with DOX-inducible SMARCB1 knockdown (KD hiPSCs). NPCs derived from these hiPSCs, that were differentiated without SMARCB1 expression (KD NPCs), exhibited an ATRT-SHH transcriptome, presented a block in neuronal differentiation, and formed orthotopic tumors. This hiPSC-derived SMARCB1 knockdown platform will be leveraged to investigate the hypotheses of this proposal. To unveil interactions between cell identity and SMARCB1 loss, RNAseq analyses will be performed on hiPSCs that were differentiated, with or without SMARCB1 expression, into NPCs and NCCs. To characterize interactions between anatomical location and SMARCB1 loss during tumorigenesis, KD NPCs and NCCs will be engrafted intracranially and subcutaneously. Resulting tumors will be analyzed via RNAseq. Since KD NPCs are unable to differentiate further into NCAM+ neurons, a high-throughput, pooled CRISPR screen will be performed on KD NPCs to identify targetable vulnerabilities that can overcome their differentiation block. Candidate genes will be validated in vitro via IPTG inducible knockdown and drug treatment studies, followed by in vivo validation by treating orthotopically engrafted KD NPC brain tumors with candidate drugs. If successful, this proposal will provide novel insights into the underlying biology that drives intertumoral heterogeneity of ATRTs and identify novel drug targets that overcome stalled differentiation, a fundamental feature of ATRTs.
