High local polyamine concentrations may suppress cell growth of primary prostatic carcinomas. When cell growth assays were conducted in defined serum-free medium, spermine inhibited the growth of poorly metastatic rat prostate carcinoma cells, causing cell cycle arrest in the G1 phase. In contrast, highly metastatic prostate carcinoma cells were resistant to the growth inhibitory activity of spermine. Ornithine decarboxylase antizyme levels, measured by Western blotting, were elevated 1–2 h after spermine treatment of spermine-sensitive cells but not spermine-resistant cells. Spermine uptake was similar in both the sensitive and resistant cell lines. These results suggest that failure to induce antizyme correlates with spermine resistance in prostate carcinoma.

We previously purified an endogenous growth inhibitor from human prostate and identified it as spermine, a polyamine secreted in high concentrations by the prostatic ductal epithelium (1). Spermine was effective in inhibiting prostate cancer cell growth both in vitro and in vivo. This work suggested that spermine might act as an endogenous repressor of prostate carcinoma cell growth. In this study, we investigated whether certain prostatic carcinoma cells might lose their sensitivity to spermine as they progress to the metastatic state. Our results suggest that highly metastatic prostate carcinoma cells may become spermine resistant and consequently grow more easily in the spermine-rich microenvironment. Resistant cells failed to undergo growth arrest or apoptosis after spermine treatment.

Elevated intracellular spermine levels generally up-regulate levels of ODC2 antizyme (2). We consequently investigated whether antizyme levels were altered in the spermine-resistant prostate cancer cells. Our results imply that tumor progression in prostate carcinoma involves faulty antizyme regulation resulting in spermine-resistant cells that can grow preferentially in the spermine-rich prostatic environment.

Cell Culture.

Cell lines derived from the Dunning rat prostate carcinoma were obtained from Dr. John Isaacs, Johns Hopkins University, Baltimore, MD (AT2.1, AT3.1, MatLyLu) and Dr. Michael Freeman, Children’s Hospital, Boston, MA (NbE, FB2) and maintained as described previously (3, 4).

Cell Proliferation Assays.

Cells were plated at a density of 6 × 104 cells/well in Corning 6-well culture dishes in RPMI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 10% calf serum and 250 nm dexamethasone (Sigma Chemical Co., St. Louis, MO) overnight. The cells were then washed once and incubated in serum-free RPMI 1640 containing 5 μg/ml human recombinant epidermal growth factor (Intergen, Purchase, NY), 5 μg/ml bovine insulin (Sigma), 5 μg/ml transferrin (Sigma), and 5 ng/ml sodium selenite (Sigma). Some of the cultures received varying concentrations of spermine tetrahydrochloride (Sigma) or Taxol (Sigma). Seventy-two h later, the cells were trypsinized and counted with a Coulter particle counter (Coulter Electronics Inc., Hialeah, FL). Cell viability was determined by staining with 0.2% trypan blue (Sigma) and examining immediately under bright-field microscopy.

Cell Cycle Analysis.

Single cell suspensions were prepared by trypsinization and resuspended in Krishan’s reagent (0.05 mg/ml propidium iodide-0.1% sodium citrate-0.02 mg/ml RNase A-0.3% NP-40, pH 8.3). Cell nuclei were then analyzed for DNA content using a Becton Dickinson FACScan cell sorter.

Annexin V Staining.

Cells were cultured on glass chamber slides (Nunc, Naperville, IL) that had been coated overnight with 1% gelatin (Difco Laboratories, Detroit, MI) solution in PBS. After cells were washed once with PBS, they were incubated with 0.5 μg/ml fluorescein-conjugated Annexin V (Clontech Laboratories, Inc., Palo Alto, CA) for 15 min at room temperature in the dark. Cells were washed with PBS and fixed with 2% formaldehyde before visualization. Cells were observed under fluorescent illumination using a filter set with absorption at 485 nm and emission at 510 nm.

Spermine Transport.

AT2.1 and AT3.1 cells were plated at a density of 104 cells/well in a 96-well plate (Corning Costar, Cambridge, MA). After 24 h, the cells were incubated in serum-free medium containing 10 μm spermine along with 10 nm [14C]spermine (Amersham Pharmacia, Piscataway, NJ) for the indicated times. Cells were then washed twice in PBS, lysed with 1N NaOH, and counted in a liquid scintillation counter.

Immunoblotting.

Antiantizyme antibodies were obtained from Dr. John Mitchel, Northwestern University, Chicago, IL (5) and from Dr. Philip Coffino, University of California, San Francisco, CA (6). Similar results were obtained with both antibodies. AT2.1 and AT3.1 cells were incubated in the presence or absence of spermine for 0–8 h and lysed in buffer containing 50 mm Tris-HCl (pH 7.4), 150 mm NaCl, 1.0% NP-40, 0.25% deoxycholic acid, 1.0 mm EDTA, 1.0 mm phenylmethylsulfonyl fluoride, 1.0 μg/ml Apopain, 1.0 μg/ml Leupepsin, 1.0 μg/ml Pepstatin, 1.0 mm Na3VO4, and 1.0 mmdl-DTT. Twenty μg of protein were subjected to SDS-PAGE and transferred to nitrocellulose membranes. The membranes were blocked with 10% milk, followed by 1-h incubations with primary and secondary (donkey antirabbit antibody conjugated to horseradish peroxidase; Amersham) antibodies and finally developed using the Amersham ECL Western blotting detection kit.

Differential Responses of Prostate Carcinoma Cell Lines to Spermine.

We previously showed that spermine treatment could cause a reduction in cell growth and viability in human and rat prostate carcinoma cell lines (1). These studies were performed in serum-supplemented medium and were potentially complicated by the action of serum amine oxidases, which can catabolize spermine to toxic byproducts, including acrolein and hydrogen peroxide. To demonstrate that spermine could directly inhibit prostate cancer cell lines in the absence of serum, we used a defined serum-free medium containing insulin, transferrin, sodium selenite, and recombinant epidermal growth factor to support the growth of Dunning rat prostate carcinoma cell lines. As shown in Fig. 1 a, this medium supported the growth of the AT2.1 and AT3.1 Dunning carcinoma cell lines.

Previous work by Isaacs et al.(3) showed that these lines differ in their metastatic potential, with the AT2.1 lines being poorly metastatic whereas the AT3.1 cells show high metastatic potential. As shown in Fig. 1,b, 3-day spermine treatment of the Dunning carcinoma cells results in a dose-dependent decrease in the number of cells remaining on the dish. Intriguingly, the poorly metastatic AT2.1 cells were more sensitive to spermine treatment by nearly two orders of magnitude. Complete cell loss was seen at 10 μm spermine with AT2.1 cells (IC50 = 2.5 μm) compared with the 1.0 mm spermine (IC50 = 150 μm) necessary to cause complete loss of the AT3.1 cells. Other poorly metastatic prostate tumor cell lines, including Nbe (IC50 = 6.3 μm) and FB2 (IC50 = 5.3 μm; Ref. 4) showed a response similar to that of the AT2.1 cells. Another highly metastatic rat prostate cancer cell line, MatLyLu (IC50 =650 μm), showed reduced sensitivity to spermine. Together, these results suggest that spermine resistance is related to the degree of metastatic potential in Dunning rat carcinoma cells, with spermine resistance developing as metastatic potential increases. These results do not apply to all types of death stimuli because, as shown in Fig. 1 c, the AT2.1 and AT3.1 cells are equally sensitive to inhibition by Taxol, a reagent that disrupts microtubule formation (7).

Cell Growth Arrest in Spermine-treated Prostate Carcinoma Cells.

The decrease in cell number observed after spermine treatment of spermine-sensitive prostate carcinoma cells could result from an inhibition of cell growth, an induction of cell death, or a combination of these two mechanisms. To determine whether spermine-treated cells undergo cell cycle arrest, we used fluorescence-activated cell sorting to analyze the DNA content of AT2.1 and AT3.1 cells grown in the presence or absence of spermine. In untreated, unsynchronized AT2.1 cells, 50.6% of the cell population was in the G1 phase, with the remaining cells in the S, G2, or M phase (Fig. 2). In contrast, after overnight exposure to 10 μm spermine, the proportion of cells in the G1 phase increased to 70.1%, indicating that spermine-treated cells undergo G1 growth arrest (Fig. 2,a). As expected, the percentage of spermine-resistant AT3.1 cells in the G1 phase was not substantially altered after spermine treatment (Fig. 2 b).

Differential Regulation of ODC-Antizyme in Spermine-treated Prostate Carcinoma Cells.

ODC antizyme (antizyme) is rapidly up-regulated after cellular exposure to elevated polyamine levels (8). To determine whether the growth inhibitory effects of spermine on Dunning rat carcinoma cell lines correlates with the induction of antizyme, we examined the levels of antizyme protein at various time points after spermine treatment of spermine-sensitive and spermine-resistant prostate carcinoma cell lines. Within 1 h after spermine treatment, the 23-kDa isoform of antizyme could be detected by Western blot analysis in the spermine-sensitive AT2.1 cells (Fig. 3,A). In AT3.1 cells, however, no significant spermine-induced increase in antizyme levels could be detected (Fig. 3 A). Thus, spermine resistance in highly metastatic prostate carcinoma cells correlates with the failure to induce antizyme subsequent to spermine treatment.

One possible explanation for the failure of AT3.1 cells to induce antizyme in response to spermine treatment would be provided if these cells have a defect in spermine transport. To test this hypothesis, we measured spermine uptake in AT2.1 and AT3.1 cells. As shown in Fig. 3 B, spermine uptake is not significantly different between these two cell lines over a time period sufficient to induce substantial antizyme levels in the spermine-sensitive AT2.1 cells.

Evidence for Apoptosis in Spermine-treated Prostate Carcinoma Cells.

Antizyme recently has been demonstrated to cause cell death when overexpressed in certain cell lines such as ras-transfected NIH3T3 cells (9). We therefore investigated whether spermine treatment of prostate carcinoma cells led to apoptosis in spermine-sensitive and spermine-resistant prostate cancer cells. Our results showed signs of apoptosis, including nuclear condensation revealed by electron microscopy and membrane blebbing in the spermine-treated AT2.1 cells but not in the resistant AT3.1 cells (not shown). We did not, however, detect caspase activation in these cells (not shown). One indication of apoptosis is the translocation of phosphatidyl serine from the inner leaflet of the plasma membrane to the outer leaflet where it can be visualized after binding to FITC-conjugated Annexin V (10, 11). As shown in Fig. 4, Annexin V binds to AT2.1 cells within 1 h after treatment, and the staining intensifies over the next 8 h. In contrast, no significant Annexin V staining was observed with the spermine-resistant AT3.1 cells. Taken together, our findings indicate that spermine-resistant cells fail to undergo either cell cycle arrest or apoptosis after spermine treatment. Such resistant cells might be expected to have an in vivo growth advantage in the spermine-rich prostatic microenvironment.

We showed previously that spermine could inhibit prostate cancer cell growth in vitro and in vivo and hypothesized that spermine may act as an endogenous inhibitor of prostate cancer cell growth in the primary site in vivo. Because some late-stage prostate cancers do acquire the ability to grow more rapidly in the prostate, we decided to investigate whether there was a class of prostate carcinoma cells that was resistant to spermine inhibition. We now demonstrate a differential sensitivity to spermine in prostate cancer cell lines. Aggressive, highly metastatic prostate cancer cell lines are relatively insensitive to the inhibitory effects of spermine, whereas other cell lines are highly sensitive, showing both inhibition of cell proliferation and apoptosis after spermine treatment.

It previously has been suggested that the cytotoxic effects of spermine are mediated by reactive oxygen species produced by the action of amine oxidases present in serum (12, 13). To determine whether this was true for prostate carcinoma cells, we cultured the cells in defined, serum-free medium that supported proliferation of both spermine-sensitive and spermine-resistant cell lines. Under these conditions, spermine-sensitive cells exposed to 10 μm spermine showed cell cycle arrest with subsequent cell death, suggesting that serum-derived enzymes do not account for the spermine-induced cell growth inhibition.

The spermine-sensitive prostate carcinoma cells showed several responses to spermine at a relatively low (10 μm) concentration. These included accumulation of cells in the G1 phase of the cell cycle and loss of viability characterized by markers of the apoptotic pathway including membrane blebbing, nuclear condensation, and translocation of Annexin V to the outer plasma membrane. We were unable to associate this apoptotic pathway with caspase activity. Caspase-independent apoptosis has been reported previously for cells with ubiquitin-independent protein degradation (14) and in BAX-induced cell death (15). In contrast to the spermine-sensitive cells, the highly metastatic, spermine-resistant cells showed no cell cycle inhibition and no loss of viability when treated for the same time periods with the same concentrations of spermine. These results suggest that cells can develop resistance to the inhibitory effects of spermine as they progress toward the metastatic phenotype. If this was to occur in vivo, such spermine-resistant cells would be expected to have a growth advantage in the spermine-rich prostatic environment (16).

A potential mechanism for the development of spermine resistance may be found in the differential induction of the protein ODC-antizyme (antizyme) in spermine-sensitive and spermine-resistant cells. Antizyme is rapidly up-regulated in many cells in response to elevated spermine levels (2). After up-regulation, antizyme binds to the polyamine-synthetic enzyme ODC (17, 18) and facilitates its degradation by the 26S proteasome (19). When overexpressed, antizyme inhibits cell growth and reduces viability both in vitro and in vivo(9, 20). It is interesting, therefore, that our results show that antizyme levels are up-regulated in spermine-sensitive but not in spermine-resistant prostate carcinoma cells.

Our results suggest that failure to up-regulate antizyme after spermine exposure may render certain prostate tumor cells resistant to spermine growth inhibition. The failure to up-regulate antizyme does not result from a primary defect in polyamine transport because there is no significant difference in spermine uptake between the spermine-sensitive and spermine-resistant cell lines at the time of antizyme induction. We recognize, however, that defects in spermine transport may occur in some prostate cancer cells and could represent an additional mechanism for spermine resistance in prostate cancer.

Antizyme may also have a role in regulating tumor cell growth and survival in other tumors. In A375 melanoma cells, Interleukin-1 inhibition of cell growth was shown to correlate with antizyme up-regulation, indicating that antizyme regulation of cell growth can occur in response to stimuli other than polyamines (21). The recent finding that antizyme levels are reduced in carcinogen-induced oral cancers provides additional evidence that antizyme levels may regulate tumor development in vivo(22). Our findings extend these results to prostatic carcinoma cells that usually grow slowly in the polyamine-rich prostatic microenvironment. We speculate that failure to up-regulate antizyme in prostate carcinoma cells can lead to a subpopulation of cells that are no longer growth inhibited by spermine and that consequently grow preferentially in the later stages of prostate cancer progression.

Fig. 1.

Response of Dunning rat prostate carcinoma cell lines to spermine and Taxol. a, poorly metastatic AT2.1 and highly metastatic AT3.1 cells were grown in defined serum-free growth medium that supports the growth of these cells equally well. b, treatment with increasing doses of spermine reveals that significantly lower spermine concentrations cause a reduction in the number of AT2.1 (○) cells compared with the AT3.1 (•) cells, demonstrating differential sensitivity to spermine. In contrast, both cell lines demonstrate similar responsiveness to Taxol (c). Bars, SD.

Fig. 1.

Response of Dunning rat prostate carcinoma cell lines to spermine and Taxol. a, poorly metastatic AT2.1 and highly metastatic AT3.1 cells were grown in defined serum-free growth medium that supports the growth of these cells equally well. b, treatment with increasing doses of spermine reveals that significantly lower spermine concentrations cause a reduction in the number of AT2.1 (○) cells compared with the AT3.1 (•) cells, demonstrating differential sensitivity to spermine. In contrast, both cell lines demonstrate similar responsiveness to Taxol (c). Bars, SD.

Close modal
Fig. 2.

Spermine-sensitive cells arrest in the G1 phase after spermine treatment. Cells were exposed to 10 μm spermine in serum-free growth medium for the indicated times, and then trypsinized, labeled with Krishan’s reagent, and analyzed by flow cytometry. The percentage of cells in the G1 phase increases as the percentage of cells in S/G2-M decreases after spermine treatment in the spermine-sensitive AT 2.1 cells (a). In contrast, no significant change was seen with AT3.1 cells (b).

Fig. 2.

Spermine-sensitive cells arrest in the G1 phase after spermine treatment. Cells were exposed to 10 μm spermine in serum-free growth medium for the indicated times, and then trypsinized, labeled with Krishan’s reagent, and analyzed by flow cytometry. The percentage of cells in the G1 phase increases as the percentage of cells in S/G2-M decreases after spermine treatment in the spermine-sensitive AT 2.1 cells (a). In contrast, no significant change was seen with AT3.1 cells (b).

Close modal
Fig. 3.

Up-regulation of antizyme (AZ) after spermine treatment of spermine-sensitive prostate carcinoma cells. A, ODC antizyme (antizyme) levels are differentially up-regulated in spermine-sensitive AT2.1 cells, with higher levels of antizyme appearing at early time points in the AT2.1 cells relative to spermine-resistant AT3.1 cells. Anti-β-actin antibody was used to confirm equal protein loading on the same gel. B, analysis of [14C]spermine uptake reveals no substantial difference in spermine uptake in AT2.1 (•) versus AT3.1 (▵) cells at the time of antizyme induction. Bars, SD.

Fig. 3.

Up-regulation of antizyme (AZ) after spermine treatment of spermine-sensitive prostate carcinoma cells. A, ODC antizyme (antizyme) levels are differentially up-regulated in spermine-sensitive AT2.1 cells, with higher levels of antizyme appearing at early time points in the AT2.1 cells relative to spermine-resistant AT3.1 cells. Anti-β-actin antibody was used to confirm equal protein loading on the same gel. B, analysis of [14C]spermine uptake reveals no substantial difference in spermine uptake in AT2.1 (•) versus AT3.1 (▵) cells at the time of antizyme induction. Bars, SD.

Close modal
Fig. 4.

Annexin V staining reveals phosphatidylserine translocation after spermine treatment of spermine-sensitive prostate carcinoma cells. Phosphatidylserine translocation, a correlate of apoptosis, is revealed by annexin V staining of spermine-treated AT2.1 cells. Annexin V staining increases with time after spermine treatment of the AT2.1 cells (left) but not of the AT3.1 cells (right).

Fig. 4.

Annexin V staining reveals phosphatidylserine translocation after spermine treatment of spermine-sensitive prostate carcinoma cells. Phosphatidylserine translocation, a correlate of apoptosis, is revealed by annexin V staining of spermine-treated AT2.1 cells. Annexin V staining increases with time after spermine treatment of the AT2.1 cells (left) but not of the AT3.1 cells (right).

Close modal

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.

2

The abbreviation used is: ODC, ornithine decarboxylase.

We thank Drs. Ruchi Newman, Marsha Moses, and Michael Freeman for critical reading of this manuscript, and Dr. Roy Smith for helpful discussions. We also recognize the helpful suggestions and comments received from Dr. Senya Matsufuji, Tokyo Jikei University School of Medicine, Tokyo, Japan. Antibodies to antizyme were generously provided by Dr. John L. A. Mitchell, Northern Illinois University, Chicago, IL, and Dr. Phillip Coffino, University of California San Francisco, San Francisco, CA.

1
Smith R. C., Litwin M. S., Lu Y., Zetter B. R. Identification of an endogenous inhibitor of prostatic carcinoma cell growth.
Nat. Med.
,
1
:
1040
-1045,  
1995
.
2
Hayashi S., Murakami Y., Matsufuji S. Ornithine decarboxylase antizyme: a novel type of regulatory protein.
Trends Biochem. Sci.
,
21
:
27
-30,  
1996
.
3
Isaacs J. T., Isaacs W. B., Feitz W. F., Scheres J. Establishment and characterization of seven Dunning rat prostatic cancer cell lines and their use in developing methods for predicting metastatic abilities of prostatic cancers.
Prostate
,
9
:
261
-281,  
1986
.
4
Freeman M. R., Bagli D. J., Lamb C. C., Guthrie P. D., Uchida T., Slavin R. E., Chung L. W. Culture of a prostatic cell line in basement membrane gels results in an enhancement of malignant properties and constitutive alterations in gene expression.
J. Cell Physiol.
,
158
:
325
-336,  
1994
.
5
Mitchell J. L., Judd G. G., Bareyal-Leyser A., Ling S. Y. Feedback repression of polyamine transport is mediated by antizyme in mammalian tissue-culture cells.
Biochem. J.
,
299
:
19
-22,  
1994
.
6
Hua S. B., Li X., Coffino P., Wang C. C. Rat antizyme inhibits the activity but does not promote the degradation of mouse ornithine decarboxylase in Trypanosoma brucei.
J. Biol. Chem.
,
270
:
10264
-10271,  
1995
.
7
Haldar S., Chintapalli J., Croce C. M. Taxol induces bcl-2 phosphorylation and death of prostate cancer cells.
Cancer Res.
,
56
:
1253
-1255,  
1996
.
8
Hayashi S., Murakami Y. Rapid and regulated degradation of ornithine decarboxylase.
Biochem. J.
,
306
:
1
-10,  
1995
.
9
Iwata S., Sato Y., Asada M., Takagi M., Tsujimoto A., Inaba T., Yamada T., Sakamoto S., Yata J., Shimogori T., Igarashi K., Mizutani S. Anti-tumor activity of antizyme which targets the ornithine decarboxylase (ODC) required for cell growth and transformation.
Oncogene
,
18
:
165
-172,  
1999
.
10
Koopeman G., Reutelingsperger C. P. M., Kuijten G. A. M., Keehnen R. M. J., Pals S. T., van Oers M. H. J. Annexin V for flow cytometric detection of phosphatidylserine on B cells undergoing apoptosis.
Blood
,
84
:
1415
-1420,  
1994
.
11
Martin S. J., Reutelingsperger C. P., McGahon A. J., Rader J. A., van Schie R. C., LaFace D. M., Green D. R. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl..
J. Exp. Med.
,
182
:
1545
-1556,  
1995
.
12
Webber M. M., Chaproniere-Rickenberg D. Spermine oxidation products are selectively toxic to fibroblasts in culture of normal human prostatic epithelium.
Cell Biol. Int. Rep.
,
4
:
185
-193,  
1980
.
13
Brunton V. G., Grant M. H., Wallace H. M. Mechanisms of spermine toxicity in baby-hamster kidney (BHK) cells. The role of amine oxidases and oxidative stress.
Biochem. J.
,
280
:
193
-198,  
1991
.
14
Monney L., Otter I., Olivier R., Ozer H. L., Haas A. L., Omura S., Borner C. Defects in the ubiquitin pathway induce caspase-independent apoptosis blocked by Bcl-2.
J. Biol. Chem.
,
273
:
6121
-6131,  
1998
.
15
Xiang J., Chao D. T., Korsmeyer S. J. BAX-induced cell death may not require interleukin 1 β-converting enzyme-like proteases.
Proc. Natl. Acad. Sci. USA
,
93
:
14559
-14563,  
1996
.
16
Heston W. D. W. Prostatic polyamines and polyamine targeting as a new approach to therapy of prostatic cancer.
Cancer Surveys
,
11
:
217
-238,  
1991
.
17
Li X., Coffino P. Regulated degradation of ornithine decarboxylase requires interaction with the polyamine-inducible protein antizyme.
Mol. Cell. Biol.
,
12
:
3556
-3562,  
1992
.
18
Murakami Y., Tanaka K., Matsufuji S., Miyazaki Y., Hayashi S. Antizyme, a protein induced by polyamines, accelerates the degradation of ornithine decarboxylase in Chinese-hamster ovary-cell extracts.
Biochem. J.
,
283
:
661
-664,  
1992
.
19
Murakami Y., Matsufuji S., Kameji T., Hayashi S., Igarashi K., Tamura T., Tanaka K., Ichihara A. Ornithine decarboxylase is degraded by the 26S proteasome without ubiquitination.
Nature (Lond.)
,
360
:
597
-599,  
1992
.
20
Murakami Y., Matsufuji S., Miyazaki Y., Hayashi S. Forced expression of antizyme abolishes ornithine decarboxylase activity, suppresses cellular levels of polyamines and inhibits cell growth.
Biochem. J.
,
304
:
183
-187,  
1994
.
21
Yang D., Hayashi H., Takii T., Mizutani Y., Inukai Y., Onozaki K. Interleukin-1-induced growth inhibition of human melanoma cells. Interleukin-1-induced antizyme expression is responsible for ornithine decarboxylase activity down-regulation.
J. Biol. Chem.
,
272
:
3376
-3383,  
1997
.
22
Tsuji T., Todd R., Meyer C., McBride J., Liao P. H., Huang M. F., Chou M. Y., Donoff R. B., Wong D. T. Reduction of ornithine decarboxylase antizyme (ODC-Az) level in the 7,12-dimethylbenz(a)anthracene-induced hamster buccal pouch carcinogenesis model.
Oncogene
,
16
:
3379
-3385,  
1998
.