Intraductal papillary mucinous neoplasm (IPMN), the most common pancreatic cystic neoplasm, is known to progress to invasive ductal adenocarcinoma. IPMNs commonly harbor activating somatic mutations in GNAS and KRAS, primarily GNAS(R201H) and KRAS(G12D). GNAS encodes the stimulatory G-protein α subunit (Gsα) that mediates a stimulatory signal to adenylyl cyclase to produce cyclic adenosine monophosphate (cAMP), subsequently activating cAMP-dependent protein kinase A. The GNAS(R201H) mutation results in constitutive activation of Gsα. To study the potential role of GNAS in pancreatic tumorigenesis in vivo, we generated lines of transgenic mice in which the transgene consisted of Lox-STOP-Lox (LSL)-GNAS(R201H) under the control of the CAG promoter (Tg(CAG-LSL-GNAS)). These mice were crossed with pancreatic transcription factor 1a (Ptf1a)-Cre mice (Ptf1a(Cre/+)), generating Tg(CAG-LSL-GNAS);Ptf1a(Cre/+) mice. This mouse line showed elevated cAMP levels, small dilated tubular complex formation, loss of acinar cells and fibrosis in the pancreas; however, no macroscopic tumorigenesis was apparent by 2 months of age. We then crossed Tg(CAG-LSL-GNAS);Ptf1a(Cre/+) mice with LSL-Kras(G12D) mice, generating Tg(CAG-LSL-GNAS);LSL-Kras(G12D);Ptf1a(Cre/+) mice. We used these mice to investigate a possible cooperative effect of GNAS(R201H) and Kras(G12D) in pancreatic tumorigenesis. Within 5 weeks, Tg(CAG-LSL-GNAS);LSL-Kras(G12D);Ptf1a(Cre/+) mice developed a cystic tumor consisting of marked dilated ducts lined with papillary dysplastic epithelia in the pancreas, which closely mimicked the human IPMN. Our data strongly suggest that activating mutations in GNAS and Kras cooperatively promote murine pancreatic tumorigenesis, which recapitulates IPMN. Our mouse model may serve as a unique in vivo platform to find biomarkers and effective drugs for diseases associated with GNAS mutations.