Fate Determinants of Basal-Squamous Pancreas Cancer

About This Grant

Project Summary Pancreas cancer is now the third leading cause of cancer related deaths, due to late diagnosis and therapy resistance. While pancreatic ductal adenocarcinoma (PDAC) has been the main focus of the field, cells of different subtypes of disease can be mixed in PDAC tumors. Single-cell data demonstrate there are subclones within PDAC with a basal gene signature, which align with adenosquamous. The basal-squamous signature is associated with worse prognosis and resistance to gemcitabine. This resistance demonstrates the need to understand the establishment and maintenance of basal-squamous pancreas cancer so effective therapies may be developed. We completed RNA-sequencing analysis to find genes upregulated in basal-squamous patient samples compared to classical patient samples to identify candidates that may be involved in basal- squamous establishment and maintenance in a physiologically relevant model. We identified novel candidate genes, whose expression correlate with canonical basal-squamous genes, that we hypothesize may play a role in basal-squamous pancreas cancer growth and identity. The Reya lab previously defined Musashi2 (Msi2) as a functional marker for cancer stem cells in pancreas cancer and recently published a novel mouse model to study pancreas cancer development from a common precancerous pool of cells. This model induces the expression of stabilized MycT58A in Msi2+ cells (Msi2-Myc mice). Msi2-Myc mice can form multiple subtypes of pancreas cancer and reliably form basal-squamous tumors (68% of mice). We will use Msi2-Myc derived precancers, human PDAC cell lines, and PDAC patient samples to determine if candidate genes are sufficient for basal-squamous establishment. We will overexpress candidate genes in these models to determine if there is an emergence and/or acceleration of the basal-squamous state through in vitro and in vivo models. We will also determine if they are necessary to maintain the basal-squamous state and growth. We will knockdown candidate genes by shRNA in human squamous cell lines, Msi2-Myc derived tumor cells, and basal-squamous patient samples, and determine the impact on the basal-squamous state and growth in in vitro and in vivo models. We will conduct CLIP-sequencing to find direct targets and integrate with RNA-seq to elucidate the mechanism by which the basal-squamous state is maintained. Preliminary data show that knockdown of candidate genes in Msi2-Myc derived mouse cells and human cell lines significantly reduces growth in vitro and in vivo, suggesting their role in maintaining growth. By RNA-seq and qPCR analysis of human cell lines and histology and immunofluorescence of resultant Msi2-Myc tumors, we have preliminary evidence that suggest genes of interest may be necessary to maintain basal-squamous identity. Based on these data, the aims of this proposal are to test the hypotheses that candidate genes are (1) necessary for the maintenance of basal-squamous pancreas cancer growth and identity and (2) sufficient to drive the establishment of basal- squamous pancreas cancer.