While recent tumor sequencing efforts have provided a catalog of mutations that occur across many different tumor types, the challenge now is to understand which mutations directly contribute to tumorigenesis. It has also become clear that tissue and cellular context contribute a great deal to the ability of a mutation to cause cancer, making it essential to study these effects in robust and relevant model systems. Despite arising from the same organ, the two most frequent cancers of the pancreas are distinct, with different incidences and natural histories, as well as unique mutation patterns. Pancreatic ductal adenocarcinomas (PDACs) have nearly penetrant gain-of-function mutation in the KRAS oncogene, and loss-of-function mutations in the TP53 tumor suppressor gene. As these are amongst the most frequent mutations in human cancers, robust mouse models have been built for this disease. Pancreatic neuroendocrine tumors (PanNETs), on the other hand, have frequent loss of function mutations in the epigenetic regulators MEN1, DAXX and ATRX. These were the first tumors identified to have mutations in DAXX and ATRX, and as such our understanding of how these mutations contribute to cancer is limited. DAXX and ATRX interact and function as a chaperone for the histone variant H3.3, specifically loading H3.3 into heterochromatic regions of DNA, including telomeres and centromeres. This proposal aims to better understand the physiological consequences and molecular mechanism downstream of DAXX loss that contribute to tumorigenesis and to address the provocative question of why tumorigenesis in the endocrine pancreas is dramatically different than in the exocrine pancreas. Using our recently developed conditional Daxx allele in the mouse, we will use genome-wide ChIP- sequencing experiments for both Daxx and H3.3 to define the specific role of Daxx as an epigenetic regulator in the pancreas. We will also use this new allele to build a relevant mouse tumor model of Daxx mutant PanNETs. We will determine if Daxx loss alone, or in cooperation with Men1 loss (lesions that co-occur in a subset of human tumors), can promote tumorigenesis. Additionally, we will conduct a CRISPR/Cas9 screen to identify novel cooperating lesions and mechanisms in vivo. As PanNETs are rare, having a robust and faithful model become even more valuable for understanding the pathogenesis of the disease and to identify novel vulnerabilities and therapeutic targets. Combined, this work aims to uncover the potential cell-type specific differences that impact tumorigenesis in the pancreas to advance our understanding of the molecular mechanisms through which Daxx loss contributes to PanNETs and to develop novel models and reagents for future studies.

Public Health Relevance

To advance and improve the treatment of human cancers, we require a more comprehensive understanding of the kinds of mutations that occur and how these mutations contribute to cancer. This study will use a mouse model we have recently developed to study tumor formation in the pancreas and better understand why mutations in DAXX are frequently found in a rare form of pancreatic cancer (pancreatic neuroendocrine tumors) and not the most common type (pancreatic ductal adenocarcinoma).

National Institute of Health (NIH)
National Cancer Institute (NCI)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Special Emphasis Panel (ZCA1)
Program Officer
Okano, Paul
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Texas MD Anderson Cancer Center
United States
Zip Code