Oral Presentation Advances in Neuroblastoma Research Congress 2016

A novel neuroblastoma metastatic mouse model identifies genes, pathways and drugs regulating metastasis (#14)

Alex BK Seong 1 2 , Kelly Fathers 3 , Robin Hallett 3 , Christina Yung 4 , Samar Mouaaz 3 , David Kaplan 3 5 , Meredith Irwin 1 2 6
  1. Medical Biophysics, University of Toronto, Toronto, ON, Canada
  2. Cell Biology, Hospital for Sick Children, Toronto, ON, Canada
  3. Neurosciences and Mental Health, Hospital for Sick Children, Toronto, ON, Canada
  4. Informatics and Bio-computing, Ontario Institute of Cancer Research, Toronto, Onatrio, Canada
  5. Molecular Genetics, University of Toronto, Toronto, ON, Canada
  6. Paediatrics, University of Toronto, Toronto, ON, Canada

Metastatic relapse is the major cause of death in Neuroblastoma (NB), yet there are no therapies to specifically target metastases. Most NB murine models do not recapitulate the bone, bone marrow and brain metastases seen in patients. We generated a NB metastatic mouse model using in vivo selection by intra-cardiac injection of NB cells expressing a GFP-Luciferase reporter. Cells from specific metastatic sites (bone, CNS) were isolated, GFP-sorted, and injected into a second cohort of mice. Bone and CNS metastases were confirmed by ex vivo bioluminescence, histology, microCT, and MRI. The metastatic subpopulations derived from bone and CNS exhibited enhanced metastatic burden and decreased latency in vivo as determined by bioluminescence monitoring. In comparison to the parental cells, these metastatic subpopulations showed increased migration, invasion, and chemoresistance in vitro, but not proliferation.

Gene expression profiling and unbiased clustering revealed two distinct subtypes, parental and metastatic, with differential regulation of 412 genes and multiple pathways including Hippo and integrin signaling. Using univariate Cox-regression, we identified a 28-gene “metastatic signature” that predicts NB patient outcome in multiple datasets.

We validated candidate genes that were upregulated including GJA1, SPHK1, and YAP/TAZ and downregulated including CADM1 in our metastatic subpopulations, and each independently predicted patient survival. Functional experiments with cells where GJA1, SPHK1, or YAP/TAZ expression was knocked down with shRNA or CRISPR, or CADM1 was overexpressed by lentiviral-mediated infection rescued metastatic phenotypes in culture and in some cases in mice. In addition, treatment with pharmacological inhibitors SKI II and Verteporfin that target SPHK1 and YAP/TAZ, respectively, inhibited NB metastasis in vivo. We have also identified drugs that selectively inhibit the growth of the metastatic subpopulations in culture.

Our metastatic NB model provides a platform to identify genes that regulate NB metastases and candidate drugs to target recurrent NB in the bone and brain.