J. K. Thompson1, H. Crawford3,4, M. Pasca Di Magliano1,2, F. Bednar1 1University Of Michigan,Surgery,Ann Arbor, MI, USA 2University Of Michigan,Cell And Developmental Biology,Ann Arbor, MI, USA 3University Of Michigan,Molecular And Integrative Physiology,Ann Arbor, MI, USA 4University Of Michigan,Internal Medicine,Ann Arbor, MI, USA
Introduction: KRAS is the primary oncogenic driver in pancreatic ductal adenocarcinoma (PDA). Pancreatic acinar cells are most susceptible to transformation by Kras in mouse models of PDA. The earliest stage of transformation consists of conversion of the acinar cells to a duct-like progenitor phenotype in a process called acinar-ductal metaplasia (ADM). Networks of developmental transcription factors (TFs) are involved in fate specification and maintenance of acinar cells. We hypothesize that oncogenic Kras alters these networks to establish the neoplastic cell state within the pancreas. In this work, we aimed to validate a genetically engineered mouse model that will allow us to analyze changes in gene regulatory networks driven by oncogenic Kras in early pancreatic neoplasia.
Methods: Mice with a pancreatic acinar cell-specific, tamoxifen inducible Cre recombinase (Ela-CreER) were bred with strains containing the oncogenic Kras G12D allele and the fluorescent protein tdTomato in the Rosa26 locus. To activate the oncogenic Kras and the tdTomato lineage tracer, Ela-CreER; Kras G12D/+; R26 tdTomato mice and littermates lacking oncogenic Kras, were treated with daily gavages of tamoxifen (4mg/day) for five consecutive days. Mice were sacked one week after the initial tamoxifen gavage. We analyzed the expression of tdTomato by fluorescence microscopy and fluorescence-activated cell sorting (FACS). Total cell RNA was isolated from the sorted cells using the RNeasy Micro Kit (QIAGEN) and complementary DNA (cDNA) was synthesized with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). We utilized a panel of 26 pancreatic developmental TFs and three acinar and duct markers (amylase, elastase, keratin 19 – CK19) with specific quantitative PCR (qPCR) TaqMan probes to characterize the isolated cells. Differences in TF expression between the wildtype and oncogenic Kras-containing mice were analyzed with a Kruskal-Wallis rank test. Statistical significance was set at p<0.05.
Results: Tamoxifen gavage induced high level of tdTomato expression in Ela-CreER mice. Fluorescence microscopy confirmed that the tdTomato+ cells were also amylase positive. RT-qPCR analysis after FACS sorting confirmed high amylase, high elastase, and low CK19 expression in the tdTomato+ cells. We also consistently found measurable levels of 23/26 (88%) of pancreatic developmental TFs in the isolated acinar cells. Preliminary analysis did not reveal significant differences in TF levels in oncogenic Kras-expressing pancreata versus wildtype controls.
Conclusion: We established a genetically engineered mouse model of early pancreatic neoplasia, which allows for specific isolation of acinar cells and their progeny. Early analysis of the model suggests that one-week activation of oncogenic Kras does not yet lead to significant developmental TF expression changes in the adult pancreas.