Developing Novel Therapies for High Risk Pediatric Cancers

Division Of Basic Sciences - Nci

Investigators

Linked publications, trials & patents

Abstract

Among other assays we will use the Incutyte and ACEA systems to monitor cell growth. For the CRISPR screens we are performing genome wide and epigenetic focused libraries. For the drug screen we will use single agent and combination responses of a panel of 2000 drugs (Mechanism Interrogation Plate (Libraries) developed by NCATS. The content of this library include FDA approved compounds, several of which are approved for cancer therapy, those in clinical trials (phase 1, 2 or 3), several kinase inhibitors. For all these compounds, the target or mechanism of action is known. The most promising targets will be further evaluated in patient derived xenograft animal models. In a complementary effort with NCATs, we have teamed up the Molecular Targets Laboratory to identify inhibitors of the PAX3-FOXO1 fusion gene. To do this we will utilize the PAX3-FOXO1 activity reporter cell line developed by our laboratory using the super-enhancer region within the ALK gene which was cloned upstream of a minimal CMV promoter driving a Green Fluorescent Protein (GFP)-Luciferase reporters. We confirmed that RH41 cells (FP-RMS) appeared green with the stably transduced reporter construct, whereas RD cells (FN-RMS) containing the reporter construct were negative. We also confirmed that the ALK enhancer luciferase activity in the RH41 cell line was suppressed rapidly upon induction of the shRNA against PAX3-FOXO1. Of note decrease in luciferase activity preceded reduction of cell number. This property allows for the cell line to be utilized in drug screening experiments. This is currently being performed in large scale. For epigenetic studies our group has recently showed that PAX3-FOXO1 reprograms the cis-regulatory epigenetic landscape by inducing de novo super enhancers in collaboration with the bromodomain and extra-terminal domain protein family member BRD4, freezing FP-RMS cells in a myoblast-like state. These studies proved the feasibility of using unbiased high-throughput screening approaches to identify small molecules that disrupt the PAX3-FOXO1 core regulatory circuitry. In particular, we demonstrated that PAX3-FOXO1 transcriptional activity depends on its physical interaction with BRD4. The BRD4 inhibitor JQ1 ablated this interaction resulting in decreased PAX3-FOXO1 protein levels that correlated with suppression of FP-RMS xenograft growth in mice. In addition to protein-protein interactions, protein homeostasis (encompassing protein synthesis, folding, trafficking and degradation) is influenced by many other factors including post-translational modifications. Therefore, selectively targeting modifications that will result in decreased stability or activity of PAX3-FOXO1 is an attractive approach toward development of novel therapies for FP-RMS. Our previous screens used indirect approaches as they assessed PAX3-FOXO1-driven target gene expression or FP-RMS cell viability, with PAX3-FOXO1 protein levels being characterized post-hoc. Assay readouts of tagged endogenous proteins obviate the limitations associated with exogenous reporters where overexpression may disrupt the natural stoichiometry of interacting proteins or result in aggregation or mislocalization. Further, it is critical that the functional activity of the protein not be impeded by the addition of the tag and thus smaller tags are desirable because of their presumably reduced impact. For these reasons, we chose to fuse the pro-luminescent HiBiT peptide to endogenous PAX3-FOXO1 using CRISPR/Cas9-mediated knockin. HiBiT is a small 11-amino acid peptide capable of producing a luminescence signal that is about 100-fold brighter than firefly or Renilla luciferases through high affinity complementation with LgBiT, an 18 kDa subunit derived from the NanoLuc luciferase, thereby allowing detection at high sensitivity. Plasmids encoding Cas9 and single guide RNAs targeting the carboxy terminus of FOXO1 together with a DNA construct for homology-directed HiBiT addition were transfected into two FP-RMS cell lines, RH4 and SCMC. The DNA homology construct encoding HiBiT was followed by a P2A "self-cleaving" peptide for coexpression of an mCherry fluorescent protein as a FACS-selectable marker. Single mCherry-positive clones were sorted into 96-well plates and expanded. Six clones were obtained for each cell line in which the HiBiT tag was demonstrated to be successfully appended to the carboxy terminus of PAX3-FOXO1 using the Nano-Glo HiBiT Blotting System. Nucleotide sequencing of clones RH4.P3F-HmC 1A9 (RH4.PAX3-FOXO1-HiBiT-P2A-mCherry, clone 1A9) and SCMC.P3F-HmC 3C4 (SCMC.P3F-HiBiT-PAX3-FOXO1-P2A-mCherry, clone 3C4) confirmed accurate in-frame addition of the HiBiT tag to PAX3-FOXO1. Principal component analysis of RNA-seq data showed that the edited clones faithfully recapitulated the transcriptome landscape of the parental FP-RMS cell lines. HiBiT signal was readily detectable and HiBiT-tagged PAX3-FOXO1 was reduced at the protein level upon treatment with JQ1 and the clinical BRD4 inhibitor CPI-0610. RH4.P3F-HmC 1A9 and SCMC.P3F-HmC 3C4 are being subjected to quantitative high-throughput screening using single agent and combination responses of a panel of 4500 FDA-approved and advanced investigational drugs toward the identification, and preclinical and clinical development of new treatment options for FP-RMS. For Rhabdomyosarcoma (RMS), FGFR4 is a rational target given that it is a key regulator of myogenic differentiation and muscle regeneration after injury; it is expressed in myoblasts, but not in differentiated skeletal muscle. We and others have found that FGFR4 is highly expressed in all RMS, and high expression is a diagnostic and prognostic biomarker. It is a direct target and strongly induced by PAX3-FOXO1, PAX3, and PAX7 and we reported that PAX3-FOXO1 established a super-enhancer at the gene's locus. We have reported that approximately 10% of FN-RMS have activating mutations in FGFR4 and that cells harboring FGFR4 mutations are oncogene addicted and sensitive to pharmacological inhibition by small molecules. Therefore, FGFR4 is a key cell surface tyrosine kinase receptor for RMS biology, growth and survival. We are developing monoclonal antibodies and human scFv binders. The majority detect the human FGFR4 protein by both ELISA and by FACS analysis . To mitigate for potential organ toxicity, we are examining FGFR4 expression levels in normal human organs. We are currently performing extensive RNAseq and immunohistochemistry (IHC) analysis of normal organ and rhabdomyosarcoma tissue arrays. We are testing our scFv binders as potential FGFR4 chimeric antigen receptors (CARs) to generate a second-generation CAR receptor lentiviral construct that contains the CD8 transmembrane region, 41BB and CD3zeta intracellular domains and a human EGFR extracellular domain. This design was chosen because of its efficacy in clinical trials and CAR T cell persistence in patient's peripheral blood for several months after therapy. The truncated EGFR in the CAR construct allows for the measurement of transduced T cells as well as therapeutic targeting of CAR T cells with Cetuximab in clinical trials in case of uncontrolled toxicity. Anti-FGFR4 CART cells could lyse RH30 but not RAJI, a FGFR4 negative Burkitt's lymphoma cell line (data not shown). Work is currently underway to validate FGFR4 CAR T cells in-vivo. We are also developing novel TCRs as therapies for pediatric solid tumors. If successful we anticipate the development of potent immunotherapeutic biologics and cell-ba *TRUNCATED*