Generation of tumorigenic porcine pancreatic ductal epithelial cells: toward a large animal model of pancreatic cancer

Background. A large animal model of pancreatic cancer would permit development of diagnostic and interventional technologies not possible in murine models, and also would provide a more biologically-relevant platform for penultimate testing of novel therapies, prior to human testing. Here, we describe our initial studies in the development of an autochthonous, genetically-defined, large animal model of pancreatic cancer, using immunocompetent pigs.Methods. Primary pancreatic epithelial cells were isolated from pancreatic duct of domestic pigs; epithelial origin was confirmed with immunohistochemistry. Three transformed cell lines subsequently were generated from these primary cells using expression of oncogenic KRAS and dominant negative p53, with/without knockdown of p16 and SMAD4. We tested these cell lines using in vitro and in vivo assays of transformation and tumorigenesis.Results. The transformed cell lines outperformed the primary cells in terms proliferation, population doubling time, soft agar growth, 2D migration, and Matrigel invasion, with the greatest differences observed when all four genes (KRAS, p53, p16, and SMAD4) were targeted. All three transformed cell lines grew tumors when injected subcutaneously in nude mice, demonstrating undifferentiated morphology, mild desmoplasia, and staining for both epithelial and mesenchymal markers. Injection into the pancreas of nude mice resulted in distant metastases, particularly when all four genes were targeted.Conclusions. Tumorigenic porcine pancreatic cell lines were generated. Inclusion of four genetic “hits” (KRAS, p53, p16, and SMAD4) appeared to produce the best results in our in vitro and in vivo assays. The next step will be to perform autologous or syngeneic implantation of these cell lines into the pancreas of immunocompetent pigs. We believe that the resultant large animal model of pancreatic cancer could supplement existing murine models, thus improving preclinical research on diagnostic, interventional, and therapeutic technologies.