Our Research

Metastases are the most deadly form of cancer, affording little or no hope of successful therapeutic intervention.  Improved treatment options for metastatic disease requires a deeper understanding of the essential rate-limiting effectors of the metastatic cascade and relevant pre-clinical animal models to test potential new therapies. In the Stewart Laboratory, we identify novel genes and signaling pathways that regulate cell proliferation, survival, migration and differentiation during embryonic development and translate these findings to animal models of cancer to determine if these mechanisms are “reactivated” in metastasis.  In addition, we perform drug screens with the new pre-clinical animal models of metastatic cancer, such as pediatric brain tumors, to identify novel therapeutics that specifically kill metastatic cancer cells.

Current models of metastatic disease suggest that it occurs in discrete stages with each stage requiring new genetic and/or epigenetic changes. These changes allow primary cancer cells to escape the tumor mass, mobilize in and out of circulation and seed foreign environments to eventually become an uncontrolled secondary growth.  One way for metastatic cells to coordinate these events is through the re-activation of conserved embryonic programs of cell migration, such as the epithelial-to-mesenchymal transition (EMT).  The EMT dramatically changes the repertoire of cell adhesion molecules on the surface of tumor cells, converting them into a more motile mesenchymal shape, which allows the cells to move away from the primary tumor.  Moreover, induction of EMT in tumor cells has an additional, devastating consequence for cancer patients as it promotes cell survival and self-renewal, which in turn allows tumor cells to evade apoptosis and remain dormant.  Other embryonic cell migration behaviors are also observed during cancer invasion and metastasis, such as contact inhibition of locomotion, however the molecular mechanisms regulating these cell behaviors, and their role in cancer metastasis are not known.

An excellent model for investigating cell migration behavior during embryogenesis is the neural crest, which is a multipotent cell population that migrates extensively in the vertebrate embryo to generate a variety of cell types, including pigment cells, neurons, glia and elements of the craniofacial skeleton. Neural crest progenitors are initially generated in the neuroepithelium of the neural tube, so they must first undergo an epithelial-mesenchymal transition (EMT) to form premigratory neural crest cells. These cells then divide and navigate through a number of embryonic tissues that secrete potential pro-apoptotic signals, before arriving at their final destination to differentiate. Thus, neural crest cells have evolved mechanisms to coordinate a number of cellular processes that involve regulation of cell survival, proliferation and migration, as well as modifications of cell-to-cell adhesion that involve dynamic interactions with the extracellular matrix. Disrupting these processes during human development causes a number of congenital diseases (neurocristopathies) and cancers such as melanoma and neuroblastoma. Importantly, recent studies have shown that reactivation of neural crest transcription factors in primary tumors promotes tumor invasiveness and metastasis.

Our understanding of the genetic pathways that activate cell migration in both embryonic development and metastasis remain incomplete.  In addition, we do not known the identity of the rate-limiting effectors of these processes, such as cell adhesion molecules and extracellular proteases, which would represent excellent therapeutic targets. This is due in part to the lack of vertebrate models that can directly test potential and required functions of genes in embryonic cell migration processes and metastasis.  To address these deficiencies we used forward genetic and genomic approaches in zebrafish embryos to identify novel genes required for EMT and collective neural crest cell migration.  These include cell signaling, cell adhesion and actin-binding molecules. In parallel, we generate metastasis models in zebrafish, including novel pediatric brain tumors resembling human primitive neuroectodermal tumors (PNETs) that metastasizes through the cerebral fluid and invades visceral tissues.  Our overarching hypothesize is that a subset of molecules required for normal embryonic EMT and cell migration movements are also required for cancer cell invasion and/or metastasis, which we can now directly test using new zebrafish cancer models. Targeting such molecules in pre-clinical metastasis models will enable us to identify new therapeutic targets that kill metastatic cells.  Our approach couples powerful embryonic techniques of the zebrafish system with newly established metastasis models to identify rational targets for blocking cancer invasion and metastasis in vivo that cannot be duplicated with other model systems.

We use zebrafish in our studies because the exceptional imaging qualities of zebrafish embryos and adult pigment mutants allow us to follow cancer cell growth and migration of in vivo using real-time imaging techniques. In addition, the molecular pathways underlying mammalian embryonic development and cancer are highly conserved in zebrafish. A unique strength of the zebrafish system as a pre-clinical cancer model is the ability to generate thousands of tumor-bearing animal, which will enable us to 1) selectively knock-out or ectopically express single or multiple genes from our embryonic assays to identify new therapeutic targets of metastasis and 2) perform high-throughput drug screens to identify small chemicals that inhibit or kill metastatic cells. We take advantage of the resources available at Huntsman Cancer Institute and the University of Utah to perform these studies (drug screening, imaging and zebrafish cores).

In summary, the ability to identify new compounds that selectively target metastatic cells will depend both on rational drug design (based on new mechanistic insights) and the generation of animal models of metastasis that can be exploited to interrogate drug and gene interactions. Together these approaches will have a significant impact on identifying new treatments and prolonging survival for patients with metastatic disease. Current projects in the lab include:

  1. Identify genes functioning downstream of Foxd3 that regulate neural crest migration and survival.  Foxd3 is a transcription factor essential for stem cell survival and neural crest EMT and cell migration, but the targets of Foxd3 are unknown. This project aims to identify the Foxd3 targets required for neural crest EMT and cell migration using both genetic and biochemical analyses in zebrafish and human cell culture. Methods include new gene knockout technologies, including TALENs and CrispR, and whole-genome ChIP.  Essential effectors of Foxd3 will be tested in the zebrafish tumor models for their ability to promote or inhibit cancer metastasis, to determine if they are relevant drug targets.
  2. Generation of new zebrafish pediatric tumor models.  We have generated new zebrafish models of pediatric brain tumors and neuroblastoma. This project aims to generate additional zebrafish models of pediatric brain tumors, including medulloblastoma and diffuse pontine glioma, to identify essential oncogenic drivers in these tumors and for use in future drug screens.
  3. Identify new therapeutics that inhibit embryonic EMT and cell migration.  We have generated novel zebrafish embryonic assays to visualized EMT and neural crest cell migration in real time with live-imaging approaches. This project aims to use chemical genetic approaches to identify new drugs that selectively inhibit EMT and cell migration.  These drugs will then be used in zebrafish models of pediatric brain tumors to determine if they can inhibit metastasis and reverse tumor growth.
Rodney Stewart
Rodney Stewart, PhD
Principal Investigator
Cancer Center Bio