UCSF

My thesis sought to advance our understanding of the genetic, molecular, and cellular factors governing the acquisition of an invasive growth phenotype during tumor development and progression using a genetically engineered mouse model of pancreatic neuroendocrine tumorigenesis (PNET) known as RIP-Tag. In one line of investigation, I compared the genome wide transcriptional profiles of non-invasive and highly invasive PNETs in RIP-Tag mice, which revealed that the expression of multiple genes encoding components of desmosomes, a cell-cell adhesion complex, is downregulated in invasive PNETs. Genetic deletion in RIP-Tag mice of one of these genes, desmoplakin, results in increased local tumor invasion, demonstrating that desmosomes can act as a distinct suppressor of tumor invasion. In a second line of investigation, I show that tumor invasion in the RIP-Tag model can be modified by genetic background. Using genetic mapping analysis, I identify a polymorphic locus on mouse chromosome 17 that is significantly associated with the development of invasive PNETs in RIP-Tag mice. I show that one gene residing in this locus, the anaplastic lymphoma kinase (Alk), is expressed at significantly higher levels in mice inbred into the C57Bl/6 genetic background versus the C3HeB/Fe genetic background and that higher levels of Alk expression are correlated with the development of invasive PNETs in RIP-Tag mice. Additionally, I demonstrate that pharmacological inhibition of Alk results in reduced tumor invasiveness in RIP-Tag mice. Collectively, my thesis research adds two new functional components to the growing list of factors that affect the hallmark capability of tumor invasion, one component of a malignant tumor phenotype. My thesis work demonstrates that tumor invasion is regulated at both the molecular and genetic levels and establishes desmosomal adhesion and signaling from the anaplastic lymphoma kinase as modulators of invasive tumor growth.