Merlin Regulation of Mouse Spinal Cord Neural Precursor Cell Function

Cancers of the spinal cord are uncommon neoplasms, the majority of which are glial cell tumors (ependymoma and astrocytoma) thought to arise from multipotent neuroglial progenitor (stem) cells (NPCs) within the spinal cord. Whereas many spinal ependymomas exhibit indolent behavior, the only treatment option for clinically-symptomatic tumors is surgery. Neurofibromatosis type 2 (NF2) is an autosomal dominant inherited cancer predisposition syndrome, caused by a germline mutation in the NF2 tumor suppressor gene, in which affected individuals develop spinal cord (SC) ependymomas. In this dissertation, we use NF2 as a tractable genetic model system to define the key intracellular signaling pathways that control SC NPC function relevant to SC ependymoma formation. In support of an essential role for the Nf2 protein (merlin/schwannomin) in SC tumorigenesis, we demonstrate that merlin loss in SC NPCs results in increased NPC growth and glial differentiation. We also show that Nf2-deficient NPCs exhibit high levels of ErbB2 activation. We further show that pharmacologic or shRNAi inhibition of ErbB2 restores Nf2-deficient NPC growth and glial differentiation to WT levels, and that NF2-associated spinal cord ependymomas exhibit increased ErbB2 activation. These findings establish ErbB2 activation as a key driver of NPC growth and gliogenesis in Nf2-deficient NPCs, and identify a unique mechanism for merlin growth regulation relevant to the design of future drug therapies for SC ependymomas. Secondarily, we demonstrate that merlin loss in brainstem NPCs does not result in increased NPC proliferation or glial differentiation. We also show that ErbB2 is not hyperactivated in brainstem NPCs. We further show that expression of a constitutively-activated ErbB2 (V659E) molecule by murine stem cell virus (MSCV) retroviral infection increases brainstem NPC proliferation and clonogenic expansion. These findings establish that merlin negatively regulates NPC function in a CNS region specific manner. Lastly, we developed genetically-engineered conditional knockout mice using Cre/LoxP technology in which Nf2 gene inactivation was targeted to various potential progenitor cell populations in the spinal cord. While robust Cre-mediated recombination and merlin loss was observed, Nf2 gene inactivation in GFAP or Olig2 transgene-positive cells did not lead to increased gliogenesis or tumor formation. These observations suggest that other temporal and or cooperative genetic events are necessary for SC ependymoma development