[CANCER RESEARCH 60, 4602– 4609, August 15, 2000] Mutational and Nonmutational Activation of p21 ras in Rat Colonic Azoxymethane- induced Tumors: Effects on Mitogen-activated Protein Kinase, Cyclooxygenase-2, and Cyclin D1 1 Marc Bissonnette, 2 Sharad Khare, 2 Friederike C. von Lintig, Ramesh K. Wali, Lan Nguyen, Yingchun Zhang, John Hart, Susan Skarosi, Nissi Varki, Gerry R. Boss, and Thomas A. Brasitus 3 Departments of Medicine and Pathology [Y. Z., J. H.], University of Chicago, Chicago, Illinois 60637 [Y. Z., J. H.], and Department of Medicine [F. C. v. L., N. V., G. R. B.], University of California, San Diego, California 92093 [M. B., S. K., R. K. W., L. N., S. S., T. A. B.] ABSTRACT Azoxymethane (AOM)-induced colonic carcinogenesis involves a num- ber of mutations, including those in the K-ras gene and CTNNB1, that codes for -catenin. Prior in vitro studies have also demonstrated that wild type p21 K-ras can be activated by epigenetic events. We identified 15 K-ras mutations in 14 of 84 AOM-induced colonic tumors by three independent methods. By single strand conformational polymorphism, we also ob- served mutations in 22 of 68 tumors in exon 3 of CTNNB1. A highly sensitive method was then used to measure p21 ras activation levels. All tumors assayed possessing K-ras mutations had significantly higher p21 ras activation levels (8.8 1.5%; n 13) compared with that of control colon (3.7 0.4; n 6; P < 0.05) or tumors without such mutations (4.2 0.4%; n 70; P < 0.05). Among tumors with wild-type K-ras, there was a subset of tumors (18 of 70) that had significantly higher p21 ras activation levels (8.0 0.9%; n 18) compared with control colons. In three of four tumors examined with activated wild-type p21 ras , we ob- served increased c-erbB-2 receptor expression and decreased Ras-GAP expression. In contrast, only one of eight tumors examined with wild-type ras and nonactivated p21 ras demonstrated these alterations. Mitogen- activated protein kinase (MAPK) activation and cyclooxygenase-2 (COX-2) expression were increased in tumors with mutated or activated wild-type p21 ras , compared with their nonactivated counterparts. Al- though -catenin mutations did not alter COX-2 expression or MAPK activity, mutations in either K-ras or -catenin significantly increased cyclin D1 expression. In contrast, in tumors with wild-type but activated p21 ras , cyclin D1 expression was not enhanced. Thus, the spectrum of changes in MAPK, COX-2, and cyclin D1 is distinct among tumors with ras or -catenin mutations or nonmutational activation of p21 ras . INTRODUCTION Colonic malignant transformation involves activating mutations in proto-oncogenes, such as K-ras, and genetic alterations in tumor suppressor genes, including the APC 4 gene and CTNNB1, the gene coding for -catenin, a downstream effector of APC signaling (1– 4). K-ras mutations have been detected in 50% of large human colonic adenomas and adenocarcinomas (2), and mutations in APC or -cate- nin occur in most colon cancers (1). Mutations in the K-ras gene (5, 6) and in CTNNB1 (7) can also be detected in colonic tumors of rats administered the colonic procarcinogen, 1,2-dimethylhydrazine, or its proximate metabolite, AOM. Several lines of evidence from both human (8 –10) and experimen- tal systems (11–14) have also indicated that epigenetic alterations in a number of important signal transduction elements, including cyclin D1 and p21 ras , may lead to the clonal expansion of a variety of malignant cell types. Recent in vitro studies, for example, have found that expression of cyclin D1 is increased by mutations in K-ras or APC/-catenin (15–17). Cyclin D1 activates cyclin-dependent ki- nase-4 and cyclin-dependent kinase-6 and, thereby, promotes the G 1 to S transition. This growth enhancing regulator is also increased in a subset of AOM-induced tumors (18). The K-ras proto-oncogene codes for p21 K-ras , a small monomeric GDP/GTP-binding (G) protein, which is involved in the regulation of a number of important normal cellular functions, including prolifer- ation, differentiation, and apoptosis (19, 20). In its GTP-bound form, it serves as an active signal transducer, whereas in its GDP-bound state, p21 ras is inactive (20, 21). The conversion of GTP-bound to GDP-bound p21 K-ras is stimulated by GAPs (20). Inactive p21 K-ras , in turn, can be reactivated by replacement of its bound GDP by GTP, via GNEFs (20, 21). Several receptor tyrosine kinases, including EGF and c-erbB-2 receptors, have been shown to activate p21 ras by increasing GNEF activity (22, 23). Although GAPs and GNEFs can, thereby, regulate the activation of wild-type p21 K-ras , these two regulatory proteins do not influence oncogenic p21 K- ras , which is constitutively activated by mutation (20). Because only 50% of colonic carcino- mas have mutant p21 K-ras , we hypothesized that alterations in the activity and/or expression of Ras-GAP and/or GNEFs, such as Sos, could lead to sustained activation of wild-type p21 K-ras in colonic malignant cells not possessing K-ras mutations. To date, however, these possibilities have not been experimentally demonstrated in models of colonic carcinogenesis in vivo. Activation of p21 K-ras , in turn, might be expected to stimulate a number of its downstream effectors including, for example, the ERK family of MAPK by activating the MEK kinase, Raf-1, thereby phosphorylating and activating MEK (reviewed in Ref. 24). This latter dual functioning kinase activates two isoenzymes of MAPK/ERK, pp44 ERK-1 and pp42 ERK-2 . Once activated, these MAPKs translocate to the nuclei of cells, and their phosphorylated substrates, in turn, lead to transactivation of genes involved in the regulation of cellular proliferation, differentiation, and malignant transformation (25). We, therefore, determined whether the activation status of p21 K-ras in AOM-induced tumors was related to alterations in the activities of these MAPKs. The gene coding for COX-2, an inducible isoform of the enzyme that catalyzes the conversion of arachidonic acid to prostaglandins and other eicosanoids (26), is another important downstream target of p21 ras . COX-2, for example, has been shown to be increased in a number of ras-transformed epithelial cells (27). Several lines of evidence, moreover, have implicated COX-2, which has been shown Received 1/26/00; accepted 6/20/00. 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 These studies were funded in part by the following grants: USPHS Grant CA36745 (to T. A. B. and M. B.), DK42086 (to T. A. B., Digestive Diseases Research Core Center), and CA69532 (to M. B.), as well as by the Samuel Freedman Research Laboratories for Gastrointestinal Cancer Research. T. A. B. is the recipient of a Merit Award from the NIH. 2 These authors contributed equally to this work. 3 To whom requests for reprints should be addressed, at Department of Medicine, MC 4076, University of Chicago Hospitals and Clinics, 5841 South Maryland Avenue, Chicago, IL 60637. Phone: (773) 702-9898; Fax: (773) 702-2182; E-mail: tbrasitu@ medicine.bsd.uchicago.edu. 4 The abbreviations used are: APC, adenomatous polyposis coli; AOM, azoxymethane; EGF, epidermal growth factor; EGFR, EGF receptor; ASOH, allele-specific oligohybrid- ization; PM-RFLP, primer-mediated RFLP; SSCP, single strand conformational polymor- phism; GAP, GTPase activating protein; GNEF, guanine nucleotide exchange factor; COX-2, cyclooxygenase-2; MAPK, mitogen-activated protein kinase; MEK, MAPK kinase; ERK, extracellular signal regulated kinase; IHC, immunohistochemistry. 4602 Research. on August 16, 2015. © 2000 American Association for Cancer cancerres.aacrjournals.org Downloaded from