p53 Independent G1 Arrest Induced by dl -α-Difluoromethylornithine

Biochemical and Biophysical Research Communications 280, 848 – 854 (2001) doi:10.1006/bbrc.2000.4227, available online at http://www.idealibrary.com on

p53 Independent G 1 Arrest Induced by DL-␣-Difluoromethylornithine Takahiro Nemoto,* ,† Sachiko Kamei,† Yousuke Seyama,* and Shunichiro Kubota* ,1 *Department of Physiological Chemistry and Metabolism, Graduate School of Medicine, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan; and †School of Allied Health Sciences, Faculty of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo 113-8510, Japan

Received December 21, 2000

Ornithine decarboxylase (ODC), which catalyzes polyamine biosynthesis, plays an essential role in cell growth. DL-␣-Difluoromethylornithine (DFMO), a synthetic inhibitor of ODC, inhibits cell growth. However, the exact mechanism by which polyamine depletion by DFMO results in growth inhibition remains to be elucidated. We clarified the mechanisms by which DFMO inhibits human gastric cancer cell (MKN45) growth. DFMO induced MKN45 cell G 1 phase arrest after 48 h, and the percentage of G 1 arrest cells continued to increase until 72 h. Expression of p21 and phosphorylation of Stat1 were significantly induced by DFMO at 24 h. Luciferase assay and gel shift assay showed specific binding of Stat1 to the p21 promoter, and promoter activity was activated at 24 h. In dominant negative p53 expressing cells, DFMO significantly induced p21 expression, arrested cells at G 1 phase, and suppressed cell growth effectively. These results suggest that DFMO induced MKN45 cell arrest at G 1 phase in a p53 independent manner, and Stat1 is, at least in part, involved in G 1 arrest. © 2001 Academic Press Key Words: polyamime; DFMO; G 1 arrest; p53; Stat1.

Polyamines (putrescine, spermidine and spermine) and ornithine decarboxylase (ODC) are essential for cell proliferation. Many reports described increased polyamine and ODC levels in various cancers (1– 4). Striking anti-proliferative effects of DFMO on various kinds of cells were reported (3–7). We previously demonstrated remarkable effects of DFMO on gastric cancer cell growth in vitro (8), but its exact mechanism remains to be elucidated. Abbreviations used: ODC, ornithine decarboxylase; DFMO, DL-␣difluoromethyl-ornithine; Stat, signal transducer and activator of transcription; 7AAD, 7-amino-actinomycin D; PY, pyronin Y. 1 To whom correspondence should be addressed. E-mail: kubota@ bio.m.u-tokyo.ac.jp.

0006-291X/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

Wild type p53 but not mutant p53 binds to a specific DNA sequence, and stimulates promoters of genes containing p53-binding site (9). The DNA binding activity and the transcriptional activity of p53 are induced on exposure to several DNA damaging agents and polyamine depletion (9 –11). p21 is a cyclin-dependent kinase (CDK) inhibitor that directly interacts with cyclin-CDK complex and thus arrests cell proliferation (12). The p21 gene promoter contains at least two binding sites for the p53 and specific DNA motifs responsible for the extracellular growth factors and hormones (13, 14). Stat proteins can recognize the SIE (sis-inducible element) motif in the promoter of p21 and increase the expression of p21 transcription (13). The characteristics of Stats are the primary regulation of their activity through rapid tyrosine phosphorylation, dimerization, nuclear translocation, and DNA binding (15). Several papers described that DFMO induces cell cycle arrest and apoptosis. Li et al. and Rey et al. showed DFMO-induced p53 dependent G 1 arrest (10, 11) and Takahashi et al. showed that DFMO induced apoptosis of the gastric cancer cells in vivo (16). The mechanism how DFMO inhibits cell growth or induces apoptosis still remains to be clarified. We undertook this study to elucidate the mechanism how DFMO inhibits human gastric cancer cell growth in vitro. MATERIALS AND METHODS Cells and reagents. MKN45 and KATO III human gastric cancer cell lines were obtained from Health Science Research Resources Bank (Osaka, Japan). The antibodies against p53, p27, Stat1, Stat3 and phospho-Tyr were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The antibody against p21 was obtained from Signal Transduction Lab. (Lexington, KY). ECL system and DL-[l14 C]-ornithine were obtained from Amersham Pharmacia Biotech. (Buckinghamshire, UK). DFMO was donated by Dr. P. McCann, Merrell Dow Research Center (Cincinnati, OH). p53 expression vector set and pCMV␤ vector were purchased from Clontech Laboratories Inc. (Palo Alto, CA). p21 promoter constructs (pUC-CAT-p21 and pGL2-p21-⌬Sma I) were kindly provided by Dr. D. Kardassis (Uni-

848

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

versity of Crete, Greece) (14). The p21 promoters were subcloned at the Hind III sites of the luciferase vector (pGL3). Dual luciferase reporter assay system and TransFast transfection reagent were purchased from Promega (Madison, WI). Cell culture and general experimental protocols. MKN45 and KATO III cells were maintained in RPMI 1640 media supplemented with 10% FBS. Dishes were incubated at 37°C in a humidified atmosphere of 95% air-5% CO 2. MKN45 cells were plated at 5 ⫻ 10 5 cells/dish and grown in RPMI 1640 media containing 10% FBS. One or 5 mM DFMO or 1 mM DFMO plus 20 ␮M putrescine were added and incubated for 24, 48, and 72 h. ODC enzyme assay and polyamine analysis. ODC activity was determined using DL-[l- 14C]-ornithine as a substrate, as described previously (17). Polyamine analysis was determined using highpressure liquid chromatography as described (18). Cell cycle analysis. For cell cycle distribution, cells were fixed with chilled 70% ethanol on ice for 30 min, treated with RNase A (0.5 mg/ml) for 20 min at 37°C, and stained with 50 ␮g/ml propidium iodide for 30 min at 4°C. G 1/G 0 cell distribution was analyzed using 7-ammino-actinomycin D and pyronin Y (7AAD/PY) stain method (19). Stained cells were analyzed using EPICS XL flow cytometer (Beckman Coulter, Tokyo, Japan). Immunoprecipitation and Western blot analysis. For immunoprecipitation, cell extracts (150 ␮g protein) were incubated at 4°C for 8 h with specific antibodies prebound to protein A-sepharose beads prewashed in IP buffer (50 mM Hepes buffer, pH 7.4, containing 150 mM NaCl, 10 mM EDTA, 100 ␮M NaF, and 2 ␮M sodium orthovanadate). For Western blotting, extracted proteins (100 ␮g) were separated by SDS polyacrylamide gel electrophoresis, transferred to nitrocellulose membrane, probed with diluted antibody (1:1,000), and visualized by ECL or diaminobenzidine as a substrate. Quantification of the bands was performed using NIH Image. The protein concentration was measured using a Bradford assay kit (Bio-Rad, Hercules, CA). Gel shift analysis. Two ␮g of nuclear extracts were used for gel shift assay using DIG gel shift kit according to the manufacturer’s protocol (Roche Molecular Biochemicals, Tokyo, Japan). The p53 consensus oligonucleotides were purchased from Santa Cruz Biotechnology. The oligonucleotides for SIE motif derived from the p21 promoter were synthesized by Espec (Tsukuba, Japan) (13). Transient transfection and promoter assay. One ␮g of luciferase construct was transfected into MKN45 cells (5 ⫻ 10 5) using Transfast lipofection reagent according to the manufacturer’s instruction (Promega). After 24 h culture media were changed to RPMI 1640 media containing 1 mM DFMO or 1 mM DFMO plus 20 ␮M putrescine. Promoter activities were analyzed by dual luciferase reporter assay system according to the manufacturer’s protocol. pRL control vector was co-transfected for normalization of transfection efficiency. Transfection efficiency using one ␮g of pCMV␤ plasmid was 54.7 ⫾ 3.8%. Stable transfection. One ␮g of plasmid construct was transfected into MKN45 cells using Transfast lipofection reagent. After 48 h media were changed to RPMI 1640 media containing 400 ␮g/ml of G418. Ten clones were selected after 2 weeks. Only one clone (number 4) showed activities to suppress DNA binding to p53 consensus oligonucleotides. Clone number 4 expressed 2.6fold p53 protein compared with the expression of mock transfectant. Cell growth and cell cycle distribution were analyzed using the clone number 4. Statistical analysis. The results were expressed as means ⫾ SD from three independent experiments. The significance was determined by Student’s t-test.

FIG. 1. Effects of DFMO on MKN45 cell growth. (a) MKN45 cells were treated with 1 and 5 mM DFMO over a 72 h period, and cell number was counted. (b) Cell cycle was analyzed by a flow cytometer as described under Materials and Methods, and percentages of G 1 phase cells are shown. (c) G 1/G 0 analysis was done using 7AAD/PY stain method. Cells treated with 1mM DFMO for 72 h were analyzed by flow cytometer. Dot plots of DNA and RNA (FL-3 and FL-2 signals) are shown.

RESULTS Inhibition of MKN45 Cell Growth and Induction of G 1 Arrest by DFMO The effects of DFMO on MKN45 cell growth were studied. DFMO concentrations at 1 and 5 mM inhibited MKN45 cell growth 66.3% and 70.4% after 48 h, respectively (p ⬍ 0.005) (Fig. 1a). After 72 h, DFMO concentrations at 1 mM and 5 mM suppressed cell growth 83.6% and 84.5%, respectively (p ⬍ 0.002). The

849

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

growth suppression by 1 mM DFMO was completely restored by addition of 20 ␮M putrescine. Viability of the cells treated with 1 mM DFMO was 98.5 ⫾ 0.7%, 97.5 ⫾ 1.7%, and 98.3 ⫾ 2.6% after 24, 48, and 72 h, respectively. We also studied whether DFMO induced DNA ladder as a marker of apoptosis. After 72 h 1 mM DFMO did not induce DNA fragmentation (data not shown). We next analyzed the effect of DFMO on ODC activities and polyamine levels in cells. One mM DFMO treatment decreased ODC activities 95.4 ⫾ 0.9% at 24 h, and the activities were undetectable after 48 h. The putrescine levels were 3.3 ⫾ 1.9 nmol/10 5 cells before treatment and decreased 1.0 ⫾ 0.5, 0.4 ⫾ 0.1, and 0.1 ⫾ 0.1 nmol/10 5 cells at 24, 48 and 72 h, respectively. The spermidine levels were 3.4 nmol/10 5 cells before treatment and decreased 1.2 ⫾ 0.4, 0.3 ⫾ 0.1, and 0.2 ⫾ 0.1 nmol/10 5 cells at 24, 48, and 72 h, respectively. The spermine levels were 6.6 ⫾ 2.4 nmol/ 10 5 cells before treatment and increased 9.4 ⫾ 2.0, 10.9 ⫾ 2.6, and 11.8 ⫾ 2.4 nmol/10 5 cells at 24, 48, and 72 h, respectively. Cell cycle analysis was done using a flow cytometer. As shown in Fig. 1b, the percentages of G 1 phase cells at 48 and 72 h were significantly increased, 63.9% and 72.6% (p ⬍ 0.02), respectively, compared to 0 h (55.7%) and 24 h (58.0%). We also checked G 1/G 0 cell distribution using 7AAD/PY stain method. Figure 1c shows the results of DNA/RNA twodimensional analysis. The G 0 population was unchanged by 1 mM DFMO treatment for 72 h. These data suggest that DFMO treatment significantly caused G 1 arrest at 48 h and the percentage of G 1 arrest cells continued to increase till 72 h. Effects of Polyamine Depletion on Expression of Cell Cycle Regulator Proteins We next studied how DFMO affected the expression of cell cycle regulator proteins. One mM DFMO induced p53 expression 1.2- and 1.3-fold, after 24 and 48 h, respectively (Fig. 2a). One mM DFMO also induced p21 expression 2.0- and 2.1-fold, after 24 and 48 h, respectively (Fig. 2b). Addition of 20 ␮M putrescine suppressed increased expression of p53 and p21 (data not shown). Expression of p27 was not induced by DFMO treatment (Fig. 2c). p53 Independent Induction of p21 in DFMO Treated Cells We next checked whether p21 induction by DFMO was in a p53 dependent manner, although p53 induction by DFMO was slight. We prepared cells expressing dominant negative p53. pCMV-mock and pCMVp53mt135 constructs were transfected into MKN45 cells and the cells were selected in the presence of 400 ␮g/ml of G418 for two weeks. One clone (number 4)

FIG. 2. Effects of DFMO on expression of p53, p21, and p27 proteins. MKN45 cells were treated with 1 mM DFMO for 24 and 48 h. Cell lysates (100 ␮g) were subjected to Western blot analysis using the antibodies against p53 (a), p21 (b), and p27 (c). Signals were visualized by ECL.

showed activities to suppress DNA binding to p53 consensus oligonucleotides. Therefore, clone number 4 was used for the following experiments. The growth of p53mt135 expressing cells was significantly suppressed by 1 mM DFMO after 48 h (Fig. 3a) (p ⬍ 0.01). The percentages of the cells at G 1 phase were significantly elevated at 48 (68.8%) and 72 h (68.9%) (p ⬍ 0.01), respectively, compared to 0 h (52.5%) and 24 h (51.7%) (Fig. 3b). The growth arrest and cell cycle arrest were completely blocked by addition of 20 ␮M putrescine. Next, we checked whether p21 was induced by DFMO in cells expressing p53mt135 (Fig. 3c). The p21 was significantly induced by 1 mM DFMO treatment. The magnitude of induction at 24 and 48 h were 9.1- and 10.1-fold, respectively, compared to that observed at 0 h. We also studied whether p21 was in-

850

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

(p ⬍ 0.01), respectively, compared with that observed at 0 h (Fig. 4b). The promoter activation with pGLp21-fl after 48 h was blocked by 20 ␮M putrescine (Fig. 4b), suggesting that the ability of DFMO to activate the p21 promoter is mediated by polyamine depletion. Activation was not seen with a control vector (pGL mock vector) alone. We next analyzed which region of the p21 promoter is responsible for p21 activation by DFMO using deletion mutant constructs as shown in Fig. 4a. One mM DFMO transactivated cells expressing full length of p21 promoter 2.3-fold after 48 h compared with control (Fig. 4c) (p ⬍ 0.02). This transactivation was suppressed by 20 ␮M putrescine. Deletion of the region between Afl II sites (pGL-p21-⌬Afl II) resulted in 45.4% reduction of promoter activities compared to that observed with full length constructs (p ⬍ 0.01). Deletion of the region between Bst EII sites (pGL-p21-⌬Bst EII) resulted in 61.8% reduction of promoter activities compared to that observed with full length constructs (p ⬍ 0.01). One mM DFMO did not transactivate cells expressing the deletion mutant constructs (pGL-p21-⌬Afl II and pGL-p21-⌬Bst EII), which did not contain Stat-binding site. On the contrary, 1 mM DFMO transactivated cells expressing deletion mutant pGL-p21-⌬Sma I which contained Stat binding site 2.0-fold, compared with control. The results suggest that DFMO activated p21 promoter through the region between the two Bst EII sites which contained Stat-binding site. DFMO Activates Stat Proteins

FIG. 3. Effects of DFMO on cell growth, cell cycle arrest and p21 expression in MKN45 cells expressing dominant negative p53. MKN45 transfectants expressing mutant p53 mt135 or vector were exposed to DFMO for 24, 48, and 72 h. (a) The numbers of cells were counted at 24, 48, and 72 h after addition of 1 mM DFMO. (b) Cells were subjected to cell cycle analysis by flow cytometer. The percentages of cells in the G 1 phase are shown. (c) p21 expression induced by 1 mM DFMO was analyzed by Western blotting.

duced by DFMO in other gastric cancer cells (KATO III) which lacked p53 expression. One mM DFMO suppressed KATO III cell growth and cells were arrested at G 1 phase (data not shown), indicating that p21 induction by DFMO was p53 independent. Polyamine Depletion Activates p21 Promoter We analyzed how p21 was induced by DFMO by transfection with p21 promoter constructs into MKN45 cells (Fig. 4a). We first transfected the full-length promoter construct (pGL-p21-fl) containing p53, Stat, and Sp1 binding sites. Time course study revealed that p21 promoter was activated 1.3- and 2.3-fold, at 24 and 48 h

Because the region which contains Stat-binding site is responsible for p21 promoter activation by DFMO, we next studied which Stat proteins are induced by DFMO using gel mobility shift assay and antibody interference assay. Cell lysates stimulated with 1 mM DFMO were prepared and immunoprecipitated with anti phospho-Tyr monoclonal antibody. Precipitated proteins were analyzed by Western blotting using anti Stat1 antibody (Fig. 5a). Although phosphorylation of Stat1 was weakly detected at 0 h, the phosphorylation of Stat1 was clearly stimulated 24 h after DFMO treatment. The magnitude of expression of phosphorylated Stat1 at 24 and 48 h after DFMO treatment was 5.5and 4.5-fold, respectively, compared with that observed at 0 h. Without immunoprecipitation with anti phospho-Tyr antibody, the expression of Stat1 was not changed by DFMO treatment (Fig. 5b). To clarify whether Stat protein is involved in p21 induction by DFMO we performed gel shift assay with the oligonucleotides whose sequence is present in Stat-binding site (Fig. 5c). Shifted bands were observed at 24 and 48 h after DFMO treatment. Two ␮g of nuclear extracts prepared from DFMO-treated cells were preincubated with 0.2 ␮g of anti Stat1 antibody, anti Stat3

851

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

852

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

antibody, and control IgG at 4°C for 30 min before gel shift assay (Fig. 5d). One hundred molar excess oligonucleotides blocked the shift of the band. Anti-Stat1 antibody, but not anti-Stat3 antibody and control IgG, blocked the shift of the band. These results suggest that Stat1 is, at least, in part, involved in p21 promoter activation induced by DFMO. DISCUSSION ODC catalyzes the rate-limiting step in the biosynthesis of the naturally occurring polyamines, which regulate proliferation and are closely linked to neoplastic growth (1– 4, 17, 20). We previously showed that DFMO significantly decreased MKN45 cell growth in vitro (8), but the exact mechanism by which polyamine depletion resulted in growth inhibition remains to be demonstrated. In this study we clarified the mechanism of G 1 arrest induced by DFMO, showing that DFMO induced Stat1 phosphorylation, and Stat1 activated p21 promoter in a p53 independent manner. To our knowledge this is the first report that Stat1 is involved in DFMO-induced G 1 arrest. Pfeffer et al. demonstrated that DFMO activates Stat3 by tyrosine phosphorylation in rat intestine mucosa cells (21). Although they did not study the involvement of other Stat family members in activation of p21 promoter, the difference between our data and their data could be due to the use of different cells. The Erk and JNK pathways link cell surface and nuclear events. Ray et al. showed that JNK activity was induced within 5 h after DFMO treatment (11). Schulze-Lohoff et al. demonstrated that DFMO induced the expression of c-Fos, c-Jun, and Egr-1 in cultured rat mesangial cells (22). They observed that increased expression of JNK/MAPK induced p53. JNKs are known to be capable to associate with p53 (23). But we could not detect significant activation of JNK by DFMO (data not shown). Therefore, it is not clear whether JNK pathway is involved in DFMOinduced G 1 arrest of MKN45 cells. There are several reports describing that ODC inhibitor induces cell cycle arrest (24, 25). In these papers, DFMO induced G 1 arrest in HuTu-80, HT-29, MCF-7, A-427, and CHO cells. Li et al. and Ray et al. showed that DFMO induced G 1 arrest of normal intestine mucosa cells in a p53 dependent manner (10, 11). However, these two groups did not study whether p21 promoter activation and Stat1 expression are involved in

FIG. 5. Effects of DFMO on Stat1 induction in MKN45 cells. (a) Cell lysates (150 ␮g) were incubated with anti p-Tyr antibodycoupled sepharose beads and immunoblotted with anti Stat1 antibody as described under Materials and Methods. Signals were visualized using diaminobenzidine as a substrate. N and M indicate negative control (without cell lysate) and molecular weight markers, respectively. (b) Cell lysates (100 ␮g) were subjected to Western blot analysis using the antibodies against Stat1. Signals were visualized by ECL. (c) Two ␮g of nuclear extracts, prepared as described under Materials and Methods, were used for gel shift assay. Arrows indicate the position of the protein-DNA complexes. NE (⫺) indicates without nuclear extract. (d) Before gel shift reaction, antibodies (0.2 ␮g) against Stat1, Stat3, or control IgG were preincubated with nuclear extracts on ice for 30 min. Arrows indicate position of the protein-DNA complexes. One hundred excess molar of competitor oligonucleotides were used to assess specific binding.

FIG. 4. Activation of p21 promoter activities induced by DFMO. (a) Schematic representation of p21 promoter deletion constructs used in the luciferase assay. (b) Time course study of p21 promoter activities. One ␮g of constructs (pGL-mock and pGL-p21-fl) were transfected into MKN45 cells. Cells were assayed by dual luciferase assay system. (c) Transactivation of the p21 promoter by DFMO. One ␮g of the constructs were transfected into MKN45 cells and incubated with 1 mM DFMO for 48 h. Luciferase assay was done as described under Materials and Methods.

853

Vol. 280, No. 3, 2001

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

DFMO-induced G 1 arrest. In the present study DFMO did not induce apoptosis of MKN45 cells. On the contrary, Takahashi et al. showed that DFMO inhibited the KKLS gastric cancer cell growth by inducing apoptosis (16). Spermine was shown to protect against apoptosis, and reduction in cellular polyamine is associated with endonuclease activities and apoptosis (26). These results suggest that intercellular spermine level is important for trigger of apoptosis. In our study, cellular putrescine and spermidine levels were decreased in DFMO-treated MKN45 cells, but spermine level was elevated. This may be the reason why DFMO did not induce apoptosis of MKN45 cells. In summary, we report here that DFMO induced G 1 arrest of MKN45 cells in a p53 independent manner and G 1 arrest was mediated through phosphorylation of Stat1 and subsequent p21 induction.

12.

13.

14.

15.

16.

ACKNOWLEDGMENTS We thank Dr. P. McCann, Merrell Dow Research Center, and Dr. D. Kardassis, University of Crete, Greece, for the generous gift of DFMO, and pUC-CAT-p21 and pGL2-p21-⌬Sma I constructs, respectively. This work was supported by a grant under the Ministry of Education, Science, Sports, and Culture, Japan.

17.

18.

REFERENCES 1. Luk, G. D., and Casero, R. A., Jr. (1987) Polyamines in normal and cancer cells. Adv. Enzyme. Regul. 26, 91–105. 2. Luk, G. D., Moshier, J. A., and Ehrinpreis, M. N. (1988) Ornithine decarboxylase as a marker for colorectal polyps and cancer. Prog. Clin. Biol. Res. 279, 227–239. 3. Auvinen, M., Paasinen, A., Andersson, L. C., and Holtta, E. (1992) Ornithine decarboxylase activity is critical for cell transformation. Nature 360, 355–358. 4. Clifford, A., Morgan, D., Yuspa, S. H., Soler, A. P., and Gilmour, S. (1995) Role of ornithine decarboxylase in epidermal tumorgenesis. Cancer Res. 55, 1680 –1686. 5. Kubota, S., Ohsawa, N., and Takaku, F. (1987) Effects of DLalpha-difluoromethylornithine on the growth and metastasis of B16 melanoma in vivo. Int. J. Cancer 39, 244 –247. 6. McCann, P. P., and Pegg, A. E. (1992) Ornithine decarboxylase as an enzyme target for therapy. Pharmacol. Ther. 54, 195–215. 7. Pegg, A. E., Shantz, L. M., and Coleman, C. S. (1995) Ornithine decarboxylase as a target for chemoprevention. J. Cell. Biochem. 22, 132–138. 8. Kubota, S. (1992) Synergistic antiproliferative activity of human fibroblast interferon in combination with alpha-difluoromethylornithine against human gastric cancer cells in vitro. Cancer 69, 2395–2399. 9. Kaelin, W. G., Jr. (1999) The p53 gene family. Oncogene 18, 7701–7705. 10. Li, L., Li, J., Rao, J. N., Li, M., Bass, B. L., and Wang, J. Y. (1999) Inhibition of polyamine synthesis induces p53 gene expression but not apoptosis. Am. J. Physiol. 276, C946 –C954. 11. Ray, R. M., Zimmerman, B. J., McCormack, S. A., Patel, T. B.,

19.

20.

21.

22.

23. 24.

25.

26.

854

and Johnson, L. R. (1999) Polyamine depletion arrests cell cycle and induces inhibitors p21 (Waf1/Cip1), p27 (Kip1), and p53 in IEC-6 cells. Am. J. Physiol. 276, C684 –C691. El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. (1993) WAF1, a potential mediator of p53 tumor suppression. Cell 75, 817– 825. Chin, Y. E., Kitagawa, M., Su, W. C., You, Z. H., Iwamoto, Y., and Fu, X. Y. (1996) Cell growth arrest and induction of cyclindependent kinase inhibitor p21 WAF1/CIP1 mediated by STAT1. Science 272, 719 –722. Moustakas, A., and Kardassis, D. (1998) Regulation of the human p21/WAF1/Cip1 promoter in hepatic cells by functional interactions between Sp1 and Smad family members. Proc. Natl. Acad. Sci. USA 95, 6733– 6738. Heim, M. H., Kerr, I. M., Stark, G. R., and Darnell, J. E., Jr. (1995) Contribution of STAT SH2 groups to specific interferon signaling by the Jak-STAT pathway. Science 267, 1347–1349. Takahashi, Y., Mai, M., and Nishioka, K. (2000) Alphadifluoromethylornithine induces apoptosis as well as antiangiogenesis in the inhibition of tumor growth and metastasis in a human gastric cancer model. Int. J. Cancer 85, 243–247. Kubota, S., Kiyosawa, H., Nomura, Y., Yamada, T., and Seyama, Y. (1997) Ornithine decarboxylase overexpression in mouse 10T1/2 fibroblasts: Cellular transformation and invasion. J. Natl. Cancer Inst. 89, 567–571. Morgan, D. M. L. (1998) Determination of polyamines as their benzoylated derivatives by HPLC. In Methods in Molecular Biology, Vol. 79 (Morgan, D. M. L., Ed.), pp. 111–123, Humana Press, Totowa, NJ. Toba, K., Winton, E. F., Koike, T., and Shibata, A. (1995) Simultaneous three-color analysis of the surface phenotype and DNARNA quantitation using 7-amino-actinomycin D and pyronin Y. J. Immunol. Methods 182, 193–207. Kubota, S., Yamada, T., Kamei, S., and Seyama, Y. (1995) Ornithine decarboxylase is directly involved in mouse mammary carcinoma cell invasion in vitro. Biochem. Biophys. Res. Commun. 208, 1106 –1115. Pfeffer, L. M., Yang, C. H., Pfeffer, S. R., Murti, A., McCormack, S. A., and Johnson, L. R. (2000) Inhibition of ornithine decarboxylase induces STAT3 tyrosine phosphorylation and DNA binding in IEC-6 cells. Am. J. Physiol. 278, C331–C335. Schulze-Lohoff, E., Fees, H., Zanner, S., Brand, K., and Sterzel, R. B. (1994) Inhibition of immediate-early gene induction in renal mesangial cells by depletion of intracellular polyamines. Biochem. J. 298, 647– 653. Fuchs, S. Y., Fried, V. A., and Ronai, Z. (1998). Stress-activated kinases regulate protein stability. Oncogene 17, 1483–1490. Seidenfeld, J., Block, A. L., Komer, K. A., and Naujokas, M. F. (1986) Altered cell cycle phase distribution in cultured human carcinoma cells partially depleted of polyamines by treatment with difluoromethylornithine. Cancer Res. 46, 47–53. Fredlund, J. O., and Oredsson, S. M. (1996) Impairment of DNA replication within one cell cycle after seeding of cells in the presence of a polyamine-biosynthesis inhibitor. Eur. J. Biochem. 237, 539 –544. Brune, B., Hartzell, P., Nicotera, P., and Orrenius, S. (1991) Spermine prevents endonuclease activation and apoptosis in thymocytes. Exp. Cell Res. 195, 323–329.