To test the hypothesis that suppression of ornithine decarboxylase (ODC) activity blocks the promotion of target cells in the outer root sheath of the hair follicle initiated by Raf/MEK/ERK activation, we crossed mice overexpressing an activated MEK mutant in the skin (K14-MEK mice) with two transgenic lines overexpressing antizyme (AZ), which binds to ODC and targets it for degradation. K14-MEK mice develop spontaneous skin tumors without initiation or promotion. These mice on the ICR background were crossed with K5-AZ and K6-AZ mice on both the carcinogenesis-resistant C57BL/6 background and the sensitive DBA/2 background. Expression of AZ driven by either the K5 or K6 promoter along with K14-MEK dramatically delayed tumor incidence and reduced tumor multiplicity on both backgrounds compared with littermates expressing the MEK transgene alone. The effect was most remarkable in the MEK/K6-AZ mice from the ICR/D2 F1 cross, where double transgenic mice averaged less than one tumor per mouse for more than 8 weeks, while K14-MEK mice averaged over 13 tumors per mouse at this age. Putrescine was decreased in MEK/AZ tumors, while spermidine and spermine levels were unaffected, suggesting that the primary role played by AZ in this system is to inhibit putrescine accumulation. MEK/AZ tumors did not show evidence of apoptosis, but there was a 15-20% decrease in S-phase cells and a 40-60% decrease in mitotic cells in MEK/AZ tumors. These results indicate that the principal effect of AZ may be to slow cell growth primarily by increasing G2/M transit time.

Elevated polyamine content and increased ornithine decarboxylase (ODC) activity have been associated with neoplastic growth in numerous animal models and human tissues. Antizyme (AZ) is a negative regulator of polyamine metabolism that inhibits ODC activity, stimulates ODC degradation, and suppresses polyamine uptake. Preliminary evidence, obtained from transgenic mice with tissue specific overexpression of AZ indicates that tumor development can be suppressed by AZ. To extend these studies, we have examined the effect of keratin 5 (K5)- or K6-driven AZ transgenes on 7,12-dimethylbenz[a]anthracene (DMBA)/12-O-tetradecanoylphorbol-13-acetate (TPA) chemical carcinogenesis of the skin, in promotion-resistant C57BL/6 and promotion-sensitive DBA/2 mice. On both genetic backgrounds, K6-AZ mice showed a reduction in tumor multiplicity, with 85% fewer tumors than wild-type controls on the C57BL/6 background and 50% fewer tumors on the DBA/2 background. K5-AZ mice developed 50% fewer tumors than controls on both backgrounds. The percent of mice with tumors and tumor size were also reduced in the K5-AZ and K6-AZ groups. Tumor and TPA-treated skin sections from K6-AZ mice exhibited the strongest AZ expression, with localization mainly in suprabasal keratinocytes. K6-AZ mice also had slightly reduced cell proliferation rates in tumors and TPA-treated skin. The lack of a more pronounced effect on cell proliferation is probably explained by the observation that AZ staining did not colocalize with proliferating cell nuclear antigen (PCNA), a marker for the proliferative compartment. These studies demonstrate a tumor-suppressive effect of AZ in C57BL/6 and DBA/2 mice, and confirm the importance of ODC and polyamines in tumor development.

A transgenic mouse line overexpressing a constitutively active mutant of MEK1, a downstream effector of Ras, driven by the keratin 14 (K14) promoter, has been used to test the hypothesis that ornithine decarboxylase (ODC) induction during tumor promotion following a single initiating event [i.e., the activation of the Raf/mitogen-activated protein kinase kinase/extracellular signal-regulated kinase (Raf/MEK/ERK) pathway], is a necessary step in skin carcinogenesis. K14-MEK mice exhibit moderate hyperplasia, with spontaneous skin tumor development within 5 weeks of birth. Analysis of epidermis and dermis showed induction of MEK protein and ERK1/ERK2 phosphorylation, but no change in Akt-1, suggesting that the PI 3-kinase pathway, another pathway downstream of ras, is not activated. Examination of tumors revealed high levels of ODC protein and activity, indicating that activation of signaling cascades dependent on MEK activity is a sufficient stimulus for ODC induction. When K14-MEK mice were given α-difluoromethylornithine (DFMO), a suicide inactivator of ODC, in the drinking water from birth, there was a dramatic delay in the onset of tumor growth (∼6 weeks), and only 25% of DFMO-treated mice developed tumors by 15 weeks of age. All untreated K14-MEK mice developed tumors by 6 weeks of age. Treatment of tumor-bearing mice with DFMO reduced both tumor size and tumor number within several weeks. Tumor regression was the result of both inhibition of proliferation and increased apoptosis in tumors. The results establish ODC activation as an important component of the Raf/MEK/ERK pathway, and identify K14-MEK mice as a valuable model with which to study the regulation of ODC in ras carcinogenesis.