Regulation of Differentiation of Pediatric Embryonal Tumors- Neuroblastoma

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Project Summary Specific Aim 1. To identify key differentiation genes. We hypothesized that epigenedysrSpecific Aim 1. To identify key differentiation genes. We hypothesized that epigenetic dysregulation of neural crest stem cells and/or sympathoadrenal progenitors contributes to neuroblastoma initiation, tumorigenesis and progression. Moreover, by targeting critical chromatin regulators that keep neuroblastoma in a self-renewal state we should be able to suppress proliferation associated with self-renewal and induce differentiation. To do this we performed an epigenetic focused siRNA screen to identify genes that control NB cell proliferation and differentiation using a high-throughput, high content imaging screen. We identified 53 candidate genes whose loss of expression results in a decrease in the number of NB cells and of these, 16 also induce morphologic and biochemical evidence of differentiation. A secondary screen using additional siRNAs excluded genes which may have resulted from off-target effects of siRNAs. Four of the candidates had already been shown to affect NB cell growth and differentiation. To prioritize those hits that would be amenable to drug development, we performed an additional screen of a tool compound library of 20 small molecule inhibitors of chromatin regulators. We evaluated the growth and differentiation in 8 NB cell lines and 2 immortal, but not transformed cell lines after exposure for 7days to 8 different drug concentrations. The secondary chemical screen identified EZH2 and SETD8 as druggable NB targets. The studies on the screen and the novel SETD8 target were published in Veschi et al Cancer Cell 2016. To characterize molecular mechanisms of action of growth and differentiation genes Our group was the first to identify that EZH2 expression was elevated in NB tumors and functioned to suppress tumor suppressor genes like CASZ1 as well as differentiation genes (Wang et al Cancer Res. 2012). We have continued our study of EZH2 as it was also found in our high-content imaging epigenetic screen that loss of EZH2 expression led to a decrease in cell growth and an increase in differentiation of NB cells. We collaborated with Dr. Kim Stegmaier's group at Dana-Farber Cancer Center and Boston Children's Hospital to examine more broadly the role of EZH2 in NB. Detailed mechanistic studies revealed that a driver of EZH2 expression was MYCN which is amplified in some 30% of high-risk NB patients and EZH2 functioned to suppress differentiation associated gene programs. A pediatric phase I evaluation of Tazemetostat is ongoing for INI deleted rhabdoid tumors and synovial sarcomas, and our study provides convincing pre-clinical rationale for the assessment in NB patients. This study was recently published in Chen, L. et al J. Clin. Invest. 128:458-462, 2018. Another hit from our screen was INCENP a member of the chromosom al complex CPC. INCENP encodes the Inner centromere protein (INCENP), which is the structural component of CPC where it is needed for the assembly of Survivin, Borealin and Aurora B kinase, and it is needed for the activation of Aurora B kinase. CPC is responsible for proper chromosomal alignment, segregation and cytokinesis during the mitosis. We followed up on this hit as 2 other members of this complex, Survivin (Birc5) and Aurora B kinase have been targeted in NB. We found high levels of INCENP indicated a poor prognosis in high-risk NB patients. Through our gain- and loss-of-function studies, we confirmed the siRNA screen result that INCENP depletion suppresses NB cell growth. Mechanistically we determined that it did so by causing polyploidization, apoptosis and senescence. Genetic INCENP inhibition also decreased NB xenograft growth. No small molecule inhibitors of INCENP binding have been identified to date. However, the identification of peptides specifically interacting with INCENP that disrupt its activation of Aurora B kinase may enable a PROTAC or degrader strategy to be developed that selectively degrades INCENP and destabilizes the CPC complex. This study was published Ming et al Cancer Research, 2019 Specific Aim 2- To determine the molecular mechanisms by which CASZ1 regulates differentiation Studies from our Casz1 knockout mouse indicated a role for Casz1 in muscle development. To explore this we evaluated CASZ1 expression in rhabdomyosarcoma (RMS) tumors. Embryonal rhabdomyosarcomas (ERMS) carrying mutant RAS have significantly lower levels of CASZ1 than fusion positive RMS(ARMS) and normal muscle. We find a contributing factor to the low levels of CASZ1 in ERMS is activated RAS/MEK pathway suppresses CASZ1 expression. Consistent with this, when MEK inhibitor (MEKi) treated ERMS differentiate into myotubes, there is a 5-10-fold increase in CASZ1 expression. Loss of CASZ1 (shRNA) attenuates the MEKi induced myogenic program in ERMS Yohe, 2018 #487. Restoration of CASZ1 expression inhibits ERMS cell growth in vitro and the growth of tumor xenografts in mice. Deep sequencing of RMS patient samples identified 4 SNVs in the CASZ1 gene in tumors (4/85). The R25C SNV occurs in the NLS region. We generated a CASZ1R25C expression vector and immuno-staining indicated high levels of CASZ1 expression that was almost completely retained in the cytoplasm. We find the CASZ1R25C SNV completely loses transcriptional activity. Furthermore, expression of this mutant CASZ1 allele in RMS cells fails to suppress xenograft tumor growth in vivo. Using mouse models of normal myoblast differentiation and models of RMS induced to differentiate with trametinib which blocks signaling pathways activated by mutant RAS, we performed the first genome-wide assessment of CASZ1 binding sites. We found CASZ1 forms a feed-forward differentiation regulatory circuit with MYOD and MYOG to enhance expression of genes key to myogenic differentiation. Additionally CASZ1 binds to neural genes and suppresses their expression in muscle cells. These data indicate that genetic alterations associated with RMS like mutant RAS or polymorphisms in CASZ1 alter the differentiation ability of rhabdomyosarcoma tumor cells to implement a full myogenic differentiation program contributing to tumorigenesis. This study was published Liu et al Nature Comm. 2020. As we previously have published that the functional suppression of CASZ1 either through 1p36LOH or via EZH2 mediated epigenetic silencing inhibits the differentiation potential of Neuroblastoma tumor cells, our current studies have been aimed at determining genome-wide assessments of CASZ1 binding in NB cells in NB tumor cells and upon restoration of CASZ1 expression which functionally inhibits growth and induces differentiation.