[CANCER RESEARCH 54, 552-559, January 15, 1994] Analysis of the Tumor Suppressor Activity of the K-rev-1 Gene in Human Tumor Cell Lines 1 Kevin Y. Sato, Paul G. Polakis, Heinz Haubruck, Clare L. Fasching, Frank McCormick, and Eric J. Stanbridge 2 Department of Microbiology and Molecular Genetics, College of Medicine, University of California, lrvine, Irvine, California 92717 [K. Y. S., C. L. E, E. J. S.], and Onyx Pharmaceuticals, Richmond, California 94806 [P. G. P., H. H., F. M.] ABSTRACT Overexpression of the human K-rev-1 gene in v-Ki-ras-transformed NIH 3T3 cells has been reported to result in the reversal of transformation and tumor suppression. To address whether human K-rev-1 is a tumor suppressor gene of human tumor cells, we have systematically transfected epitope-tagged wild-type or activated mutant K-rev-1 complementary DNA expression vectors into a series human tumor cell lines that express an activated ras oncogene, namely HT1080, E J, and SW480. Using the epitope-tag-specific monoclonal antibody, it is shown that the K-rev-1 protein localizes to the medial/trans-Golgi network. Ectopic expression of the wild-type or activated mutant K-rev-1 protein did not significantly affect the morphology or in vitro growth of any clones. Furthermore, all clones expressing the wild-type or activated mutant K-rev-1 protein were tumorigenic. Western blot analysis of tumor reconstitutes demonstrated that there was no decrease or loss of introduced K-rev-1 protein expres- sion. The results in the present study demonstrate that expression of K-rev-1 does not reverse the transformed phenotype or significantly affect the tumorigenic phenotype of human tumor cell lines that express endog- enous ras oncogenes. INTRODUCTION Mutational activation of cellular ras is one of the most commonly observed events in clinical tumors, ras mutations have been observed in greater than 50% of all oncogene-associated tumors (1). It has been hypothesized that activated ras plays an important role in the devel- opmental biology of cancer, due to its association with many different human tumors. A considerable amount of research has been focused on understanding the biology, function, and regulation of the ras protein, p21. Currently, very little is known about the negative regulation of ras p21 activity and signal transduction beyond the known GTPase-acti- vating proteins (2). To isolate factors involved in negatively regulating ras p21 function, Noda et al. (3) transfected a v-Ki-ras-transformed NIH 3T3 cell line (DT) with a normal human fibroblast cDNA 3 expression library. Since the only known transforming gene in the DT cells is the v-Ki-ras oncogene, factors expressed by the cDNA library that affected the transformed phenotype were assumed to be acting either directly or indirectly upon the oncogenic ras gene product. This procedure resulted in the isolation of the human K-rev-1 gene (4). Other groups had previously cloned this gene and termed it smg21a and rapla, respectively (5, 6). Based on Northern blot analyses, the ability of the K-rev-1 gene product to revert DT cells to the nontrans- formed phenotype was dependent upon high levels of mRNA expres- sion. Specific amino acid substitutions in the K-rev-1 protein [Gly 12 to Val 12 (V12) or Gin 63 to Glu 63 (E63)] enhanced its reversion activity, as evidenced by decreased mRNA levels required for rever- Received 9/21/93; accepted 12/2/93. 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 This work was supported by NIH Predoctoral Training Fellowship GM07134-17 (K. Y. S.) and USPHS Grant CA-19401 from the National Cancer Institute (E. J. S.). 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: cDNA, complementary DNA; GAP, GTPase-activating protein; CMV, cytomegalovirus; mAb, monoclonal antibody; DME, Dulbecco's modified Eagle's medium; RER, rough endoplasmic reticulum; FITC, fluorescein isothiocyanate. sion of transformation and tumor suppression (7). 4 The E63-activated K-rev-1 mutant was also found to have a greater suppressive activity than the V12-activated mutant. The human K-rev-1 gene encodes a protein that shares greater than 50% overall amino acid sequence identity with ras p21. Most of the sequence similarities lie within regions that are functional do- mains in ras p21, such as the domains responsible for GTP binding and GTPase activity, a carboxyl-terminal isoprenylation signal se- quence, and a domain that shares complete identity to the rasGAP/ effector binding domain (4). Biochemically, the K-rev-1 protein binds GTP and has a weak intrinsic GTPase activity (5, 8). A spe- cific GTPase-activating protein (rapGAP) and a GDP dissociation- stimulating factor have also been isolated for the K-rev-1 protein (9- 13). The K-rev-1 protein is posttranslationally modified by the addition of a geranylgeranyl group to the cysteine residue of the C181-Ali-mli-X (Ali, aliphatic amino acid; X, any amino acid) con- sensus isoprenylation signal sequence (14). Like ras p21, the K-rev-1 protein associates with membranes via the isoprenyl group (14). Isoprenylation is required for its transformation and tumor sup- pressor effects in DT cells (7). Although the K-rev-1 protein associ- ates with membranes, its cellular localization is different from that of ras p21. Using an antiserum that recognizes both K-rev-1 and raplb proteins, B6ranger et al. (15) provided evidence that K-rev-1 may lo- calize to the medial/trans-Golgi network. Mutagenesis and K-rev-1/ras p21 chimera studies revealed that the common effector/rasGAP binding domain was required for K-rev-1 transformation reversion activity (7, 16). The K-rev-1 protein binds to rasGAP with more than a 100-fold greater affinity than ras p21 in vitro (8, 17). This binding was found to be dependent upon K-rev-1 protein association with GTP, but rasGAP did not stimulate its GT- Pase activity, rapGAP associates with the K-rev-1 protein at a site outside of the effector/rasGAP binding domain and can stimulate its GTPase activity in the presence of a prebound rasGAP. 5 Collectively, these studies suggest that the K-rev-1 protein reverses the transformed phenotype and affects tumorigenicity by binding to a ras p21 effector. This would result in sequestering the factor and attenuation of the oncogenic ras p21 proliferative signal. Recently, Sakoda et aL (18) demonstrated that the K-rev-1 protein inhibited c-Ki-ras p21-associ- ated activation of the c-los promoter/enhancer in NIH 3T3 cells, which suggests that the K-rev-1 protein might antagonize the ras p21 signal transduction pathway in vivo. To investigate the role of the human K-rev-1 protein in human tumor growth suppression, we have systematically transfected either wild-type or activated mutant K-rev-1 cDNA expression vectors into a series of human tumor cell lines that express one of the activated ras oncogenes. Unlike earlier studies, this analysis, in part, uses peptide- tagged K-rev-1 proteins that allow for specific detection of the intro- duced K-rev-1 protein. Antiserum to these tagged proteins have al- lowed us to determine that the K-rev-1 protein indeed localizes to the medial/trans-Golgi network. 4 M. Noda, personal communication. 5 p. Polakis, unpublished data. 552 Research. on November 29, 2021. © 1994 American Association for Cancer cancerres.aacrjournals.org Downloaded from