ORNITHINE DECARBOXYLASE OVER-EXPRESSION STIMULATES MITOGEN-ACTIVATED PROTEIN KINASE AND ANCHORAGE-INDEPENDENT GROWTH OF HUMAN BREAST EPITHELIAL CELLS Andrea MANNI 1 *, Rita WECHTER 1 , Susan GILMOUR 4 , Michael F. V ERDERAME 1 , David MAUGER 3 and Laurence M. DEMERS 2 1 Department of Medicine, Milton S. Hershey Medical School, Pennsylvania State University, Hershey, PA 2 Department of Pathology, Milton S. Hershey Medical School, Pennsylvania State University, Hershey, PA 3 Center for Biostatistics and Epidemiology, Milton S. Hershey Medical School, Pennsylvania State University, Hershey, PA 4 Lankenau Medical Research Center, Wynnewood, PA, USA In these experiments we tested the hypothesis that consti- tutive activation of polyamine(PA) biosynthesis may contrib- ute to mammary carcinogenesis. Spontaneously immortal- ized normal human MCF-10A breast epithelial cells were infected with the retroviral vector pLO SN containing a cDN A which codes for a truncated and more stable ornithine decarboxylase (O DC), the rate-limiting enzyme in PA synthe- sis. Upon chronic selective pressure with a-difluoromethyl- ornithine (DFMO) (an irreversible inhibitor of ODC), infected MCF-10A cells exhibited an approximately 250-fold increase in ODC activity, which persisted despite discontinuation of DFMO. ODC-over-expressing MCF-10A cells showed a mod- est decrease in S-adenosylmethionine decarboxylase and an increase in spermidine/spermineN 1 -acetyltransferase. Analy- sis of cellular PA profile revealed a selective accumulation of putrescine without alterations in spermidine and spermine contents. Lesser degrees of increased ODC activity were obtained reproducibly by re-exposing the cellsto incremental small doses of DFMO. W e observed a bell-shaped dose- related positive effect of O DC activity on clonogenicity in soft agar of MCF-10A cells. Since anchorage-dependent growth was actually reduced, such positive influence on this feature of transformation was not a non-specific consequence of a growth advantage provided by ODC over-expression. In addition, we observed a close parallelism between the dose- dependent effects of ODC expression on clonogenicity and activity of the ERK-2 kinase, a central element of the MAPK cascade. Our data demonstrate an interaction between PA and the MAPK signalling pathway and suggest that the latter may be involved in ODC-induced transformation of mam- mary epithelial cells. Int. J. Cancer, 70:175–182, 1997. r 1997 Wiley-Liss, Inc. There is evidence to indicate that polyamines (putrescine, spermidine and spermine) are critically involved in breast-cancer- cell proliferation (Manni and Wright, 1984). Furthermore, we have shown that activation of polyamine biosynthesis may lead to tumor progression characterized by the acquisition of a less hormone- responsive and more aggressive breast cancer phenotype (Manni et al., 1995a, b, c). On the basis of these findings and the increasingly recognized oncogenic role of ODC (Clifford et al., 1995), the first rate-limiting enzyme in polyamine biosynthesis, we hypothesized that activation of the polyamine pathway may contribute to breast cancer development. To test this hypothesis, we induced ODC over-expression in the spontaneously immortalized MCF-10A human mammary epithelial cell line. MCF-10A cells, which exhibit a completely normal phenotype (Soule et al., 1990), have been extensively used to test the influence of activation of specific genes on the process of mammary carcinogenesis (Basolo et al., 1991). The potential clinical relevance of this experimental system is highlighted by the resemblance of its progressive stages of transformation to those of the human disease, i.e., hyperplasia without and with atypia, carcinoma in situ, and infiltrative cancer (Miller et al., 1993). Our results indicate that induction of ODC over-expression in MCF-10A cells confers anchorage-independent growth and that this effect may be mediated through the MAPK signalling cascade. MATERIAL AND METHODS Cell lines and culture conditions The parent MCF-10A1 cell line (Karmanos Cancer Institute, Detroit, MI) and its genetically engineered derivatives, were grown in Dulbecco’s minimal essential medium (DMEM)/F-12 (1:1) supplemented with 5% equine serum, 0.1 μg/ml cholera toxin, 10 μg/ml insulin, 100 U/ml penicillin, 100 μg/ml streptomycin, 2.5 μg/ml amphotericin B (GIBCO, Grand Island, NY), 0.5 μg/ml hydrocortisone (Sigma, St. Louis, MO) and 10 ng/ml epidermal growth factor (Collaborative Research, Palo Alto, CA). Infection technique and DFMO selection MCF-10A1 cells were infected either with the control retroviral expression vector pLXSN or with the same vector in which a 1.8-kb mouse ODC cDNA had been sub-cloned into the EcoRI restriction site (Clifford et al., 1995). This cDNA codes for a more stable enzymatically active truncated ODC molecule (Ghoda et al., 1989). We empirically observed that, in order to maintain ODC over-expression, it was necessary to chronically expose our in- fected cells to DFMO treatment. Therefore, ODC-infected MCF- 10A1 cells were grown in the presence of 0.1 mM DFMO. The drug was washed off for 3 days or longer before the cells were plated in the experimental dishes. We have reported that a similar experimental approach was necessary to preserve ODC over-expression in transfected MCF-7 breast-cancer cells (Manni et al., 1995c). Anchorage-dependent growth In these experiments, the cells were plated in triplicate at a density of 4 3 10 4 cells per 35-mm dish in the various experimental conditions described in ‘‘Results’’ and in the figure legends. Triplicate dishes were harvested 24, 48, 72 and 96 hr after plating. The number of cells was counted using a Coulter (Hialeah, FL) counter. Anchorage-independent growth Clonogenicity in soft agar was tested as described (Manni et al., 1995a). Cells were plated in triplicate at a density of 1.5 3 10 4 cells per 35-mm dish. The number of colonies (aggregates greater than 50 cells) was scored after 21 to 25 days. Measurement of cellular enzymatic activity and polyamine levels In these experiments, the cells were plated in duplicate at a density of 3 3 10 5 cells per 100-mm dish. For determination of enzymatic activity, the cells were washed 3 times with ice-cold PBS and then re-suspended in buffer containing 50 mM Tris, 2.5 mM DTT, 0.1 mM EDTA, pH 7.5, and stored at 270°C until use. At the time of the assays, the cells were frozen and thawed twice. The cell lysates were centrifuged at 13,600 g for 20 minutes. Contract grant sponsor: the National Institutes of Health; contract grant number: CA 40011-11; contract grant number: R-29CA55066. *Correspondence to: Division of Endocrinology, Diabetes, and Metabo- lism, Hershey Medical Center, 500 University Drive, Hershey, PA 17033, USA. Fax: 717-531-5726. Received 8 July 1996; revised 10 September 1996. Int. J. Cancer: 70, 175–182 (1997) r 1997 Wiley-Liss, Inc. Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer