[CANCER RESEARCH 63, 2005–2009, May 1, 2003] Advances in Brief The Mutant K-ras Oncogene Causes Pancreatic Periductal Lymphocytic Infiltration and Gastric Mucous Neck Cell Hyperplasia in Transgenic Mice 1 Felix H. Brembeck, Franz S. Schreiber, Therese B. Deramaudt, Linden Craig, Ben Rhoades, Gary Swain, Paul Grippo, Doris A. Stoffers, Debra G. Silberg, and Anil K. Rustgi 2 Gastrointestinal Unit [F. H. B., F. S. S., T. B. D., B. R., G. S., D. G. S., A. K. R.], Department of Genetics [A. K. R.], Abramson Cancer Center and Family Cancer Research Institute [F. H. B., F. S. S., T. B. D., B. R., A. K. R.], and Division of Endocrinology, Diabetes and Metabolism [D. A. S.], University of Pennsylvania, Philadelphia, Pennsylvania 19014; University of Tennessee College of Veterinary Medicine, Tennessee [L. C.]; and Department of Surgery, Northwestern University, Chicago, IL 60611 [P. G.] Abstract A frequent genetic alteration found in premalignant stages of pancre- atic adenocarcinoma is K-ras oncogene point mutation. The mechanistic basis for the inability of K-ras mutation to transform pancreatic ductal cells is unclear, although cooperating events with p16 inactivation, p53 mutation, and SMAD 4 mutation are recognized to be necessary. We have generated a novel mouse model in which the cytokeratin 19 promoter, specifically active in pancreatic ductal cells but not other cell types of the pancreas, is fused to mutant K-ras. This is of direct relevance to human pancreatic cancer because premalignant lesions are found specifically in ductal cells. There is dramatic periductal lymphocytic infiltration in the pancreata of transgenic mice, predominantly CD4 T lymphocytes, which may act as an adaptive immune response to activated ras-mediated sig- naling. In addition, gene array analysis reveals an induction of N-cadherin in transgenic mice pancreatic ductal cells, the significance of which relates to promotion of cell adhesion and deterrence of cell migration. Apart from these important biological considerations, there is parallel activity of the cytokeratin 19 promoter in the stem cell region of the gastric epithelium, namely in mucous neck cells. Activated K-ras in this context causes mucous neck cell hyperplasia, a precursor to gastric adenocarcinoma. There is concomitant parietal cell decrease, which is a key step toward gastric adenocarcinoma. Taken together, we have defined how mutant K-ras signaling modulates important molecular events in the initiating events of pancreatic and gastric carcinogenesis. Introduction Pancreatic adenocarcinoma is a common gastrointestinal malig- nancy with 25,000 new cases annually in the United States. Unfor- tunately, there are 28,000 deaths related to the disease and its metastatic complications attributable to presentation of patients at advanced stages. Pancreatic adenocarcinomas arise in ductal epithelial cells and undergo a carefully orchestrated program of genetic alter- ations that arise in premalignant lesions. These eventually culminate in the development of cancer. K-ras oncogene point mutations are found in 90 –95% of all pancreatic ductal adenocarcinomas (1, 2) and, as such, also represent the earliest known genetic alteration in premalignant or precursor lesions. Although K-ras oncogene point mutations are important initiating events, it is clear that inactivation of the p16 and p53 tumor suppressor genes, as well as the SMAD genes, are also important for tumor development and progression (3–7). The ability to model pancreatic ductal adenocarcinoma and its precursor stages has been hampered by the inability to target mutant Ki-ras in a direct fashion to ductal cells. The elastase promoter is active in acinar cells and has been used to target Ha-ras to the pancreas (8). These mice develop acinar cell tumors. To that end, we have targeted mutant Ki-ras with the cytokeratin 19 promoter to ductal cells (9, 10) to recapitulate the premalignant stages of pancreatic adenocarcinoma. Because this promoter is also active in the gastric isthmus region, we also observe evidence of mucous neck hyper- plasia. Materials and Methods K19-K-ras V12 Transgene. For construction of the K19-K-ras-V12 Neo- Bam transgene vector, the pCMV-NeoBam vector (gift of B. Vogelstein) was used. The cytomegalovirus promoter in this vector was replaced with the 2.1-kb K19 5' flanking region plus promoter fragment after KpnI and SalI digestion. The 2.1-kb K19 fragment was generated by PCR using a 5'- CTAGTCTAGACGTCTCAAGTTCCTTTCTAAGACCCAC primer to am- plify at -1970 bp and introduce a 5' KpnI and BsmBI restriction site and a 5'-T ATGGCCGACGTCG ACGGAAAAAGGGACGCAGGTCTGA primer to amplify up to +46 bp of the K19 promoter linked to a SalI restriction enzyme site. PCR was performed using 20 ng of template DNA, 200 M each deoxynucleotide triphosphate, 0.5 M each primer, and 2.5 units of plaque- forming unit Turbo DNA polymerase (Stratagene) in a 1  reaction buffer provided by the manufacturer. After an initial 60 s denaturation step at 94°C, 30 cycles of amplification were done as follows: denaturation 60 s/94°C, annealing 60 s/60°C, extension 2 min/72°C, and a final extension at 72°C for 10 min. The K19 promoter PCR product was KpnI/SalI digested and subcloned into the NeoBam vector. The K-ras-V12 cDNA, with a substitution of Gly3Val at codon 12, was obtained by a BamHI digest of the pZIP-K-rasV12 vector (gift of C. Der) and subcloned in the BamHI restriction site of the K19-NeoBam vector between the rabbit -globin intron and polyadenylation sequence. The construct was verified by DNA sequencing by the dye termi- nator cycle sequencing method (ABI) in the automated DNA sequencing facility of the University of Pennsylvania. Transgenic Mice. For microinjection, the purified K19-K-rasV12 NeoBam vector was linearized by a BSMBI/EagI restriction enzyme digest to release the K19-K-rasV12 transgene, purified using Elutips, and used for microinjection. The DNA was injected into B6  SJL recipient oocytes and transferred to pseudopregnant females. Genotyping of founders and offspring was performed using PCR with primers specific for the transgene. The sense primer (5'- GGCTGGCGTGGAAATATTCT), spanning the -globin intronic sequence, and the antisense primer (5'-GCTGTATCGTCAAGGCACTC) recognize a specific sequence of the K-rasV12 cDNA. Transgenic mice demonstrated a single PCR band corresponding to the expected product of 360 bp. Transgene incorporation was confirmed by Southern blot analysis after a BglII restriction enzyme digest of 15 g of genomic DNA to release the integrated 2.7-kb transgene within the K19 promoter and at the polyadenylation signal. Blots were hybridized with the 360-bp PCR probe. Only mice with proven integra- tion of the transgene by Southern blot analysis and corroborated by PCR genotyping were analyzed further and compared with age-matched wild-type littermates. Three founder lines were established with generation of F1 and F2 mice. Received 10/25/02; accepted 3/18/03. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by NIH Grants R01 DK 60694, NIH R01 DK59539 (DGS), and P30 DK50306 (Morphology, Molecular Biology, Transgenic/Chimeric Mouse, and Cell Cul- ture Core Facilities) and the Abramson Family Cancer Research Institute. 2 To whom requests for reprints should be addressed, at 6 CRB, University of Pennsylvania, 415 Curie Boulevard, Philadelphia, PA 19014. Phone: (215) 898-0154; Fax: (215) 573-5412; E-mail: anil2@mail.med.upenn.edu. 2005 Research. on February 12, 2016. © 2003 American Association for Cancer cancerres.aacrjournals.org Downloaded from