Tumor Microenvironment Network: Projects

The Proneural Perivascular Niche and Therapeutic Response

Principal Investigator: Eric Holland, MD, PhD

We are investigating the in vivo expression patterns of the tumor vasculature and stromal cells of the perivascular niche in response to therapy, aiming to identify genes that predict response and survival in patients with glioblastoma multiforme (GBM).  

Specific Aims:

  • Using mouse models of proneural GBMs, we will determine which cells (endothelial cells, microglia, or reactive astrocytes) of the perivascular stroma produce the gene products that drive the resistance phenotype and predict survival in human GBM. 
  • Using human proneural GBM surgical samples, we will verify the perivascular cell types in which the survival genes are produced. Using immunohistochemistry and fluorescence-activated cell sorting with cell surface markers identified in mouse models, we will isolate these stromal cell types from human proneural GBMs and determine whether the cells express these genes.
  • Using mouse models of proneural GBMs, we will determine whether expression of the survival genes in their respective cell types is regulated by the aggressiveness of the tumor cells, or whether the tumor responds to treatment with radiation and temozolomide over time. 
  • Using mouse models of proneural GBMs, we will knock down the genes in the appropriate perivascular cell types to determine whether any of the genes are causally related to therapeutic resistance. We will use cell type mixing experiments in culture and implanted into mice in vivo.
  • The four investigations listed above, which focus on the biology of therapeutic response in the PDGF/proneural GBM subgroup, will be extended to NF1/mesenchymal and EGFR/classical GBMs.

Brain Metastasis Microenvironment and Mechanisms

Principal Investigator: Joan Massagué, PhD

We are investigating the role of tumor vasculature in facilitating a tumor’s metastasis to the brain, and in modulating the therapeutic response of metastatic tumors.

Specific Aims:

  • We recently identified a number of genes that are associated with brain metastasis. Sets of these genes will be screened for mediators of blood-brain barrier extravasation, vascular cooption, astrocyte and microglia engagement, and brain parenchyma infiltration. In addition, these mediators will be pre-clinically targeted.
  • We will define in situ cancer cell viability pathways and reactive endothelial pathways in the brain metastatic niche, and target these pathways pre-clinically.
  • We will identify genes that are specifically expressed in situ in brain metastasis cancer cells, endothelial cells, astrocytes, or microglia.
  • We will define the role of brain endothelium derived “angiocrine” factors and other stroma derived factors that support brain metastatic outgrowth, and target these factors pre-clinically.
  • We will define the impact of radiation treatment on the expression of mediators described above. In addition, we will elucidate the mechanisms that support the viability of residual disease and its eventual outgrowth after radiation therapy.

The Role of Endothelial-Derived Angiocrine Factors in the Progression of Brain Tumors

Principal Investigator: Shahin Rafii, MD

We are using genetic tools to modify endothelial cells in a manner that enables us to investigate the role of angiocrine signaling in the progression of both primary and metastatic brain tumors.

Specific Aims:

  • By employing both in vitro and in vivo angiogenic models, we will determine the potential of conditional selective activation of mryAkt1 in endothelial cells (ECs) in sustaining the growth of primary and metastatic brain tumors.
  • We will identify angiocrine factors produced by tumor-activated brain ECs that induce proliferation and invasiveness of primary and metastatic brain tumors, including Wnt5a, Notch-ligands (Jagged-1 and Jagged-2), EGF, PDGF-β, and PDGF-β.
  • We will assess whether conditional deletion of Notch-ligands, PDGFs, Wnt5a selectively in ECs (VE-cadherin-Cre, CD133-Cre), stromal cells (Nestrin-Cre, SMA-Cre), or monocytes (LysM-Cre) abrogates metastasis to the brain and impairs the growth of gliomas.
  • We will identify as yet unrecognized angiocrine factors specifically induced in tumor brain ECs in brain tumor models established in the Holland (PDGF/proneural GBMs) and Massagué labs (metastatic breast and lung tumors).
  • We will determine the effectiveness of targeting angiocrine factors in knockout models or via systemic delivery of highly efficient shRNA against angiocrine factors in blocking the growth and reversing chemotherapy resistance of brain tumors in in vitro co-culture and in vivo brain tumor models.