| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Departments of Cellular and Molecular Physiology [D. J. F., L. M. S., A. E. P.] and Pharmacology [A. E. P.], Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given the absolute requirement of polyamines for cell growth and the potentially oncogenic consequences of their overproduction, it is not surprising that ODC activity and polyamine levels are subjected to extensive regulation. The amount of ODC and other polyamine-metabolizing enzymes is controlled at the level of transcription and translation (reviewed in Refs. 1 , 2 , and 16 ), but a unique regulatory protein termed AZ is responsible for posttranslational regulation of ODC as well (reviewed in Refs. 17, 18, 19, 20, 21 ). In response to increased cellular polyamine content, a +1 frameshifting event is stimulated in the translation of AZ mRNA, resulting in the synthesis of full-length AZ protein (reviewed in Ref. 22 ). AZ then binds to the ODC monomer, which prevents formation of the enzymatically active homodimer and therefore reduces ODC activity. In addition, AZ binding stimulates the degradation of ODC by the 26S proteasome in an ATP-dependent but ubiquitin-independent manner. Many cell types possess a highly inducible polyamine transport system, and the uptake of exogenous polyamines is inhibited by AZ as well. As a result of this ability, AZ may possess a greater capacity to restrict cellular polyamine levels than either DFMO or a dominant negative ODC. In addition, recent evidence suggests AZ may stimulate polyamine excretion (23) . Therefore, in an elegant feedback response to elevated polyamine levels, AZ suppresses polyamine synthesis and accumulation through multiple mechanisms including reduction of ODC activity, depletion of ODC protein, and inhibition of polyamine uptake.
The AZ mRNA contains two potential start codons near the 5' end and produces a protein of either Mr 29,500 or Mr 24,000 (AZ-1A or AZ-1B, respectively), depending on the start codon used (19 , 24) . Two additional AZ genes have recently been identified and designated AZ-2 and AZ-3 (24 , 25) to distinguish them from the original, which is now termed AZ-1. Functional studies of the AZ-2 protein reveal inhibition of ODC activity and polyamine transport, but its ability to stimulate ODC degradation in unclear (26) . AZ-3 also inhibits ODC, and its expression is limited to germ cells within the testis. Studies presented in this article use AZ-1.
The mouse skin chemical carcinogenesis model is an extensively characterized system for the study of genetic and epigenetic events that influence tumor initiation, promotion, and progression (reviewed in Refs. 27, 28, 29 ). In this model, tumors characterized as benign papillomas form on the skin after a single application of a mutagenic tumor initiator and repeated application of a tumor promoter. Numerous tumor promoters of varying potency have been identified, and their application to mouse skin invariably results in an induction of ODC activity (30, 31, 32) . Established tumors contain constitutively increased quantities of the enzyme as well as elevated polyamine levels (33 , 34) . Furthermore, transgenic mice expressing a stable ODC protein in the skin from either the K5 or K6 promoter develop papillomas after DMBA initiation without subsequent tumor promotion (35) . K6/ODC mice crossed with TG.AC mice, which express v-Ha-Ras and are phenotypically initiated, form tumors in the absence of chemical initiation or promotion (36) . Conversely, DFMO treatment inhibits skin tumor promotion by protein kinase C-binding phorbol esters such as TPA (32 , 37) and non-protein kinase C-binding anthrone derivatives such as chrysarobin (38) . Systemic DFMO exposure regresses DMBA-induced tumors on K5/ODC and K6/ODC transgenic mice and DMBA/TPA-induced tumors on normal mice (39) as well as spontaneous tumors on K6/ODC-TG.AC double transgenic mice (40) . Systemic DFMO exposure has the potential for growth-inhibitory effects on the vasculature in addition to those on the tumor cells (40 , 41) . Nevertheless, these numerous lines of evidence suggest that increased ODC activity and polyamine levels are both necessary and sufficient for the tumor promotion stage of skin carcinogenesis.
We have used cytokeratin promoter elements to direct the expression of the AZ regulatory protein to specific skin cell populations in transgenic mice. These mice allow us to evaluate (a) the ability of AZ to regulate ODC activity and polyamine levels in vivo and (b) the role in chemical carcinogenesis of increased ODC activity and polyamine levels within specific skin cell populations. We report here that K5-AZ and K6-AZ transgenic mice exhibit a reduced induction of skin ODC activity after application of the tumor promoter TPA. In addition, AZ transgenic mice are resistant to DMBA/TPA chemical carcinogenesis relative to normal littermates. Therefore, AZ-mediated suppression of cellular polyamine content inhibits the promotion of initiated cells to macroscopic tumors.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Transgenic Vector Construction.
Two plasmid vectors, pK5-AZ and pK6-AZ, were constructed to target AZ expression to the skin of transgenic mice using bovine cytokeratin promoter sequences. The AZ cDNA was amplified by PCR from plasmid pGEM-
T205 AZ (42)
for insertion into the SalI and ClaI sites of the pK5 (35)
and pK6 (43)
constructs, respectively (Fig. 1A)
. The AZ insert was a full-length rat cDNA with 15 bp of native 5'- and 3'-untranslated region. The cDNA was engineered to contain a single-nucleotide deletion (T205) that removes the TGA stop codon terminating the first open reading frame of the mRNA, thereby allowing constitutive AZ expression without the requirement for polyamine-stimulated frameshifting (42)
.
|
Propagation and Screening of Transgenic Lines.
Transgenic lines were established by breeding transgenic founders and their progeny to B6D2F1/J mice (The Jackson Laboratory, Bar Harbor, ME) to produce equal numbers of transgenic mice and normal littermates. The genomic DNA of the resulting offspring was extracted from tail clips using the DNeasy Tissue Kit (Qiagen, Inc., Valencia, CA) and screened for the presence of the transgene DNA by PCR or slot blotting. For PCR screening, a sense primer that binds in the promoter region (K5P1 or K6P1) and an antisense primer that binds in the AZ coding region (AZP1) were used to amplify a region of the transgene in potential K5-AZ or K6-AZ transgenic mice, respectively (Fig. 1A)
. All PCR screening was performed at least in duplicate. For slot blot identification of transgenic mice, tail DNA was hybridized with a PCR product that corresponded to the promoter/transgene junction (K5P1/AZP2 or K6P1/AZP2; Fig. 1A
). Binding of the fluorescein-labeled probe (Random Prime Labeling and Signal Amplification System; Amersham Pharmacia Biotech) was quantitated using a FluorImager 595 and ImageQuant application software (Molecular Dynamics, Sunnyvale, CA). Copy number estimates were derived by comparing the signal intensity of transgenic mouse DNA to a standard curve of the transgene construct spiked into normal mouse DNA (44)
. Primer sequences were as follows: (a) K5P1, 5'-AGGCTGAATTAGGAGGCGTTG-3'; (b) K6P1, 5'-GCAGAAGGAGGGGACAATTATCAC-3'; (c) AZP1, 5'-CGAAGCAGTGGCTGTTGAGG-3'; and (d) AZP2, 5'-GCTGGGAGCTCGATGTAGAG-3'.
TPA Treatment and Skin Sample Processing.
An area of dorsal skin was shaved with surgical clippers, and 17 nmol of TPA (Calbiochem-Novabiochem Corp., La Jolla, CA) in 200 µl of acetone were applied 12–16 h later. Mice were sacrificed at the indicated time after TPA application, and treated skin was excised from the mouse. The epidermis was then scraped from the dermis after exposure to 55°C water for 20 s (45)
. Both epidermis and dermis were divided into equal-sized samples for measurement of ODC activity and polyamine levels. The age and numbers of normal and transgenic mice for each experiment are indicated in the figure legends and represent a mix of male and female animals.
Epidermal tissue was thawed on ice and then resuspended in 400 µl of ice-cold ODC Buffer A [25 mM Tris-HCl (pH 7.5), 2.5 mM DTT, 0.1 mM EDTA, 1x Protease Inhibitor Cocktail Set I (Calbiochem-Novabiochem), and 0.01% Tween 80]. The tissue was lysed by sonication and centrifuged for 30 min at 30,000 x g and 4°C, and the supernatant was transferred to a new tube and spun again immediately before the ODC and protein assays. Protein quantitation was performed using a protein assay dye reagent (Bio-Rad Laboratories, Hercules, CA) with BSA as a standard. Dermal samples were minced with scissors, homogenized in 1 ml of ice-cold ODC Buffer A using a polytron, and then centrifuged in the same manner.
Measurement of ODC Activity and Protein, Polyamine Levels, and AZ Protein.
ODC activity of epidermal and dermal extracts was assayed in duplicate for each sample using a standard assay (46)
. Reactions contained 20 µM L-[1-14C]ornithine (47.70 mCi/mmol; NEN Life Science Products, Boston, MA), and activity is expressed as pmol CO2 liberated/30 min/mg protein extract. ODC protein was detected using an affinity-purified ODC polyclonal antibody (5)
.
Polyamine content was determined using an ion pair reverse-phase high-performance liquid chromatography method as described previously (47) and normalized to tissue wet weight (in grams). Normalization to mg protein/sample yielded similar results.
Rabbit AZ antiserum was produced at Cocalico Biologicals (Reamstown, PA) using a purified polyhistidine-tagged AZ fusion protein (6xHis-AZ; Ref. 48 ) and was further purified with a 6xHis-AZ affinity column (AminoLink Plus Immobilization Kit; Pierce, Rockford, IL). For Western blotting, 100 µg of cytosolic protein extract were fractionated on 15% SDS-PAGE gels, transferred to polyvinylidene difluoride-Plus membrane (Osmonics Laboratory Products, Westborough, MA), blotted with purified AZ antibody (0.5 µg/ml), and visualized using an enhanced chemifluorescence Western Blotting Reagent Pack (Amersham Pharmacia Biotech) and a FluorImager 595. Antibody specificity was confirmed by preabsorbing with the 6xHis-AZ antigen and by blotting with a second affinity-purified polyclonal AZ antibody preparation (49) .
Two-stage Chemical Carcinogenesis.
An area of skin of approximately 2.5 x 2.5 cm and 1 cm above the base of the tail was shaved with surgical clippers when mice were 7–9 weeks of age. After a 48-h rest period, mice were initiated with 200 nmol of DMBA (Kodak Laboratory Chemicals, Rochester, NY) applied in 200 µl of acetone. Promotion began 1 week later with 17 nmol of TPA (Calbiochem-Novabiochem) in 200 µl of acetone applied twice weekly. Tumors greater than 1 mm in diameter were counted weekly, and data are expressed as the percentage of mice in each group bearing at least one tumor (tumor incidence) and the mean number of tumors/animal in each group (tumors/mouse). Mice were sacrificed 1 week after the final TPA application, and multiple tumors from a single mouse were combined and processed for measurement of ODC activity as described above for dermal tissue.
Statistical Analysis.
ODC activity and polyamine levels of normal and transgenic mice taken at a single time point were compared using the two-tailed, unpaired Student’s t test. ODC activity and tumor counts taken over a time course were compared using two-way ANOVA. For the 0–37.5 h TPA time course with K5-AZ and K6-AZ transgenic mice (Figs. 2
and 3
), normal littermate epidermal ODC values from the two groups were not statistically different and were combined for the analysis. Tumor incidence data were compared using a log-rank test of the survival curves.
|
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The cDNA insert for the transgene constructs had a single-nucleotide deletion (T205) to remove the requirement for polyamine-stimulated frameshifting in the translation of AZ mRNA (42)
. Three potential K5-AZ founder animals were designated founders 14, 2, and 3. K5-AZ(2) did not transmit the transgene, and AZ expression was not detected in mice from the K5-AZ(3) line. Therefore, all studies in this article used the progeny of founder 14 (referred to as K5-AZ). Two K6-AZ founder animals were designated K6-AZ(18) and K6-AZ(52). The number of integrated transgene copies in the genomic DNA is approximately 2–5 for K5-AZ mice and 20–30 for mice from both K6-AZ lines (data not shown). No obvious phenotypic abnormalities were observed in any of the founder animals or their progeny.
AZ and ODC Expression in the Skin of K5-AZ and K6-AZ Transgenic Mice.
Epidermal and dermal ODC activities increase rapidly and transiently after application of the tumor promoter TPA, peaking at approximately 4.5 h after treatment (30)
. Because the ODC activity of normal skin is near the limit of detection with even the most sensitive assay conditions, K5-AZ and K6-AZ were screened for AZ expression by treating mice with TPA and looking for a blocked or attenuated induction of ODC activity.
The K5 promoter should direct constitutive AZ expression in the basal cell layer of the interfollicular epidermis and the ORS of the hair follicle, and the promoter is not responsive to tumor promoter application (50
, 51)
. Preliminary studies showed that transgenic mice from the K5-AZ line had a >90% reduction in epidermal ODC activity 18 h after TPA application compared with normal littermates (Fig. 1B)
. A reduction in dermal ODC activity was not detected for K5-AZ transgenic mice analyzed at this single time point. The K6 promoter should direct constitutive AZ expression in the ORS of the hair follicle. In addition, tumor promoter application will elevate K6-driven expression in the ORS and induce novel expression within the interfollicular epidermis (52
, 53)
. K6-AZ(18) transgenic mice and normal littermates were treated at 0 h to induce AZ expression from the K6 promoter and treated again at 43.5 h to induce ODC activity at a time of elevated AZ expression. ODC activity was markedly induced in the epidermis of normal mice at 48 h, but the K6-AZ(18) transgenic mice exhibited a 90% reduction in ODC activity (Fig. 1B)
. The absolute ODC activity was considerably lower in dermal extracts from normal mice, but the activity in the K6-AZ(18) transgenic mice was again reduced by 90%. Preliminary studies with mice from the K6-AZ(52) line indicate a similar reduction in TPA-induced ODC activity (data not shown). Therefore, AZ protein is expressed in the skin of both K5-AZ and K6-AZ transgenic mice.
In addition to inhibiting ODC activity in a stoichiometric manner, AZ also stimulates degradation of the ODC protein by the 26S proteasome. K6-AZ(18) mice and normal littermates were treated with TPA at 0 h to induce expression of the AZ transgene, and skin samples were collected after a second TPA application at 72 h. Western blot analysis for ODC protein detected a strong increase in epidermal ODC expression 4.5 h after TPA treatment of normal but not K6-AZ(18) mice (Fig. 1C)
. This result is entirely consistent with an accelerated rate of ODC degradation catalyzed by transgene-derived AZ. ODC was not strongly induced in the normal dermis, but reduced levels of ODC protein were again detected in transgenic extracts.
Polyamine Levels in the Skin of K5-AZ and K6-AZ Transgenic Mice.
It was postulated that even a minimal amount of AZ expression from the K5 promoter could bring about a reduction in epidermal and dermal polyamine content. Therefore, polyamine levels were measured in the skin of K5-AZ transgenic mice and normal littermates shaved 16 h before sacrifice of the animal. Shaving produces a modest increase in the polyamine content of normal mouse skin (data not shown). The content of putrescine, spermidine, and spermine was reduced (by 60%, 37%, and 10%, respectively) in both the epidermis and dermis of K5-AZ transgenic mice relative to normal littermates (Table 1)
; however, only the reduction in epidermal spermidine content was statistically significant. The epidermal polyamine content was also determined for K5-AZ mice and normal littermates treated with 17 nmol of TPA 18 h before sacrifice (as in Fig. 1B
). The putrescine and spermidine levels of normal epidermis were substantially increased, whereas spermine levels were slightly reduced compared with untreated mice (Table 1)
. Although putrescine, spermidine, and spermine levels were all lower (by 6%, 12%, and 23%, respectively) in K5-AZ mice compared with normal littermates, none of the changes were statistically significant. Dermal polyamine content was not determined in these mice because a difference in dermal ODC activity was not detected at this time point (Fig. 1B)
.
|
. Therefore, polyamine content was measured in these mice 72 h after TPA application to evaluate the ability of AZ to suppress the polyamine synthesis stimulated by TPA. For epidermal samples from normal mice, a single TPA application resulted in substantial increases in putrescine and spermidine content and a slight decrease in spermine compared with untreated animals (Table 1)
. K6-AZ(18) mice exhibited similar levels of epidermal putrescine and spermine, but the epidermal spermidine content was significantly reduced (by 37%) compared with that in normal mice treated with TPA. Similar results were observed in measurements of dermal samples. Polyamine levels were not measured in untreated K6-AZ mice because the constitutive expression from the K6 promoter is limited to the extremely small cell population within the ORS of the hair follicle (52
, 53)
.
TPA Induction of ODC Activity in K5-AZ and K6-AZ Mouse Skin.
To further characterize the relative abundance and location of AZ expression from the K5 and K6 promoters, the induction of epidermal and dermal ODC activity was measured after successive TPA applications at 0 and 24 h (Figs. 2
and 3
). TPA caused a dramatic increase in the epidermal ODC activity that peaked at 4.5–9 h in normal mice. A similar induction of lesser magnitude was observed after a second application at 24 h. This lesser induction is consistent with previous studies that demonstrate that the skin is refractory to a second TPA application for up to 24 h (54)
, possibly due to cell cycle-dependent changes in TPA responsiveness (55)
. The induction of ODC activity was slightly reduced in the epidermis of K5-AZ transgenic mice (Fig. 2A)
, but the change was not statistically significant. The peaks of dermal ODC activity were significantly reduced in K5-AZ mice (Fig. 2B)
, indicating AZ expression that was not detected at a single time point (Fig. 1B)
. Epidermal and dermal extracts from K5-AZ mice were also subjected to Western blot analysis, but bands consistent with the size of the AZ protein were not detected (data not shown). Therefore, AZ protein is below the limit of detection by Western blot analysis of K5-AZ mice, but AZ activity is readily detected by polyamine measurements and ODC activity assays.
K6-AZ(18) transgenic mice and normal littermates were treated in the same manner to examine the level of AZ expression (Fig. 3)
. The first peak of epidermal ODC activity is similar in magnitude for normal and K6-AZ(18) transgenic mice (Fig. 3A)
. However, the induction of activity is less prolonged for the transgenic animals, and a second TPA application does not result in any increase in ODC activity. This result suggests that AZ expression from the K6 promoter increases in the epidermis after TPA application, but not as rapidly as the induction of ODC activity. In the dermis of K6-AZ(18) transgenic mice, the first peak of ODC activity is reduced in magnitude, whereas the second induction is completely blocked (Fig. 3B)
. Therefore, AZ is constitutively expressed in the dermis of K6-AZ(18) mice, and TPA induces additional AZ expression in this tissue. Overall, the expression of AZ from the K6 promoter is consistent with the previously reported pattern of constitutive and inducible transgene expression (43
, 52
, 53)
. Comparison of the experiments with mice from the K5-AZ and K6-AZ(18) lines (Figs. 2
and 3
) indicates that the TPA-treated skin of K6-AZ(18) transgenic mice clearly contains more total AZ protein than the skin of K5-AZ transgenic mice. Polyamine content was also measured in epidermal and dermal samples from the K6-AZ(18) transgenic mice and their normal littermates (data not shown). Consistent with the suppression of ODC activity by AZ, putrescine content was significantly reduced in both the epidermis and dermis of transgenic animals during the time course from 0–37.5 h; however, there were no significant differences in the spermidine or spermine levels.
AZ expression in the epidermis and dermis of TPA-treated K6-AZ(18) transgenic mice was directly assessed by Western blot analysis as well (Fig. 4)
. Several nonspecific bands are detected in epidermal and dermal extracts from both normal and K6-AZ(18) mice. At time 0, epidermal extract from a K6-AZ(18) mouse is similar to extract from a normal littermate, indicating that constitutive AZ expression is below the limit of Western blot detection (Fig. 4A)
. However, AZ protein is detected beginning 4.5 h after TPA application and increases in intensity up to 24 h. The second TPA application at this time gives a very strong induction of epidermal AZ expression. The size of the main induced band (Mr 24,000) is consistent with translation initiation at the second AUG of the AZ mRNA (AZ-1B; Refs. 24
and 56
). An additional induced band of lesser intensity (Mr 29,500, 28.5 h sample) may represent rare initiation at the first AUG of the AZ mRNA (AZ-1A). Several AZ fragments (ranging in size between Mr 16,000 and Mr 23,000) are also seen in the transgenic extracts, although their origin and function remain unclear (19)
. The AZ fragments are detected when AZ expression is highest, and they are the only induced bands in dermal extracts (Fig. 4B)
. Comparable levels of AZ expression were detected in the skin of mice from the K6-AZ(52) line (data not shown).
|
and reducing the polyamine content (Table 1)
after TPA application. To evaluate the role of increased ODC activity and polyamines in skin carcinogenesis, the susceptibility to DMBA/TPA chemical carcinogenesis was determined for both lines of K6-AZ mice. Transgenic mice from the K6-AZ(18) line developed significantly fewer tumors per animal than their normal littermates over the course of the experiment (Fig. 5B
; P < 0.001). In the K6-AZ(52) line, which has a similar transgene copy number and level of AZ expression, the average number of tumors per animal was >80% lower than that for normal littermates (Fig. 5D
; P < 0.001). The striking reduction in tumor multiplicity for K6-AZ(52) mice is best demonstrated by pictures of normal and transgenic animals taken at the conclusion of the experiment (Fig. 6)
. Although these pictures suggest that tumors on K6-AZ(52) mice are smaller than tumors on normal littermates, the heterogeneity of tumor size in all groups makes it difficult to draw any conclusions. There was no significant difference between K6-AZ(18) transgenic mice and normal littermates when the percentage of animals with tumors was compared (Fig. 5A)
. However, because this parameter scores the appearance of even a single tumor, it is much less sensitive for detecting differences between groups. A significant delay in tumor onset was detected when the tumor incidence of K6-AZ(52) transgenic mice was compared with that of normal littermates (Fig. 5C
; P < 0.001).
|
|
; P < 0.001). The K5-AZ mice developed half as many tumors as normal littermates (P < 0.001). Combination of the K5-AZ and K6-AZ transgenes to create double transgenic K5-AZ/K6-AZ(18) mice resulted in significantly fewer tumors than normal mice (P < 0.001) as well as the individual K5-AZ and K6-AZ(18) lines (both P < 0.01). When tumor incidence was examined, a significant delay in tumor onset relative to normal littermates was detected for the K5-AZ, K6-AZ(18), and double transgenic mice (Fig. 5E
; all P < 0.05). Therefore, the expression of AZ from either the K5 or the K6 promoter clearly reduces the susceptibility of transgenic mice to DMBA/TPA chemical carcinogenesis.
ODC Activity in Tumors from Transgenic Mice and Littermates.
Whereas ODC activity increases transiently after tumor promoter application, constitutive induction of the enzyme is observed within tumors resulting from DMBA/TPA carcinogenesis (33
, 34)
. Tumors were collected from mice in Fig. 5
one week after the final TPA application to determine the level of constitutive ODC and AZ expression (Table 2)
. The ODC activity in papillomas from normal mice was clearly elevated compared with the undetectable levels of normal skin. ODC activity was reduced in tumors from K5-AZ mice, both lines of K6-AZ mice, and double transgenic animals; however, the effect was statistically significant only for K6-AZ mice. This reduction in ODC activity is consistent with AZ expression within the hyperplastic tumor environment and was expected due to the responsiveness of the K6 promoter to proliferative stimuli (52)
. Western blot analysis confirmed expression of the AZ protein within tumors from K6-AZ(18), K6-AZ(52), and double transgenic mice, but not in tumors from normal or K5-AZ mice (data not shown).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and the induction of ODC activity after TPA application is attenuated in both the epidermis and dermis of K5-AZ mice (Figs. 1B
and 2
). Strong AZ expression is detected in K6-AZ mice by assays of epidermal and dermal ODC activity (Fig. 3)
and by Western blots for both ODC (Fig. 1C)
and AZ (Fig. 4)
protein. Polyamine measurements in the skin of K5-AZ and K6-AZ mice indicate that the epidermal and dermal spermidine content is reduced compared with normal littermates after a single TPA application (Table 1)
. Transgenic mice from the K5-AZ, K6-AZ(18), and K6-AZ(52) lines are produced at the expected 50% frequency, develop normally to adulthood, and have a life span comparable to that of normal littermates. The normal growth and development of mice overexpressing the inhibitory AZ protein may be due to the restricted expression of AZ in specific cell populations or compensatory increases in ODC activity and/or polyamine transport.
There are several interesting points related to Western blots of AZ in the K6-AZ epidermis and dermis (Fig. 4)
. First, it is clear that translation initiation at the second start codon of the AZ mRNA resulting in synthesis of AZ-1B is preferred in mouse skin, consistent with previous in vivo studies with AZ overexpression in mouse heart (48)
and previous studies in cells that overproduce ODC (57
, 58)
. Second, AZ fragments are detected in the skin (Fig. 4)
and hearts (48)
of transgenic mice or in cultured cells expressing a stable ODC (58)
or transfected AZ.4
The sizes of the AZ fragments are consistent between studies, therefore they may represent protease-sensitive sites within the protein or common intermediates in its degradation.
Effect of AZ Expression on Chemical Carcinogenesis.
K5-AZ and K6-AZ transgenic mice were subjected to a chemical carcinogenesis protocol to evaluate the role in skin carcinogenesis of increased ODC activity and elevated polyamine content. Statistically significant reductions in tumor incidence and multiplicity were achieved for both K5-AZ and K6-AZ(18) transgenic mice (Fig. 5)
. K5-AZ mice developed 50% fewer tumors than normal littermates, and K6-AZ(18) mice averaged a 35% reduction in two independent experiments. Combination of the K5-AZ and K6-AZ(18) transgenes yielded an additive reduction (70%) in tumor multiplicity, suggesting that the level of AZ expression correlates with the amount of resistance to chemical carcinogenesis. The K6-AZ(52) transgenic mice exhibit the most impressive effect of AZ expression on tumor formation (Figs. 5
and 6
). In addition to a striking reduction in tumor multiplicity (80%), tumor incidence data showed that the formation of even a single tumor could be delayed significantly or even prevented in K6-AZ(52) mice. Overall, all three lines of mice with targeted AZ expression show a reduction in tumor multiplicity relative to normal littermates, clearly indicating that fewer initiated cells develop into macroscopic tumors in the presence of AZ. It is noteworthy that expression from both the K5 and K6 promoters is directed to the ORS of the hair follicle. This is a proposed location of stem cells targeted by chemical initiators (59, 60, 61)
, and immunohistochemical staining of ODC protein after repeated TPA applications is most prominent in the region corresponding to the upper ORS (62
, 63) .
Limitations of Polyamine and ODC Activity Measurements in Extracts.
K5-AZ mice exhibit modest reductions in epidermal and dermal ODC activity (Fig. 2)
and polyamine content (Table 1)
, and these mice are resistant to DMBA/TPA carcinogenesis (Fig. 5F)
. The K5 promoter targets expression to the ORS of the hair follicle and the basal cell layer of the interfollicular epidermis. AZ could completely block the induction of ODC activity by TPA within these particular cells in vivo yet at the same time not be present in sufficient excess to bind up all of the ODC in homogenized epidermis or dermis assayed in vitro. Similarly, because K5-derived AZ can only inhibit ODC activity and polyamine accumulation within a very small percentage of the total cell population, it is very difficult to detect changes in the total epidermal or dermal polyamine content. Carcinogenesis studies of K5-AZ transgenic mice suggest that the apparent shortage of AZ expression is overcome by exquisite targeting of the protein to the initiated cell population. However, recent studies have proposed a role for AZ in the degradation of cyclin D1 and cyclin-dependent kinase 4 (21)
. AZ may therefore affect a reduction in tumor growth by both depleting polyamine pools and impeding cell cycle progression.
In contrast to K5-AZ mice, AZ expression in K6-AZ(18) mice completely prevents the increases in epidermal and dermal ODC activity after a second TPA application (Fig. 3)
. The strong induction of the K6 promoter by TPA (Fig. 4)
leads to excess AZ production in certain cell populations, and this excess protein is able to bind and inhibit all of the ODC within the homogenized extracts assayed in vitro. K6-AZ(18) mice exhibit only a modest reduction in polyamine content 72 h after a single TPA application (Table 1)
because in vivo the initial peak of ODC activity is only blocked effectively within the limited cell population with constitutive K6 expression, the ORS of the hair follicle. Constitutive expression from the K6 promoter is absent from the basal cell layer of the interfollicular epidermis, and TPA-induced expression from the K6 promoter occurs primarily in suprabasal layers (52
, 53) . Therefore, K6-derived AZ presumably blocks the expansion of initiated cells within the ORS of the hair follicle rather than expansion of initiated cells that may exist within the basal cell layer of the interfollicular epidermis.
Influence of Genetic Background on ODC Induction and Chemical Carcinogenesis.
In the experiments presented in this article, the transgenes were maintained on a mixed C57BL/6J (B6) and DBA/2J (D2) genetic background, and these inbred mouse strains differ in terms of ODC induction after acute tumor promoter application (31)
, histological changes associated with chronic tumor promoter application (64)
, and susceptibility to two-stage chemical carcinogenesis (28)
. The documented difference in susceptibility to chemical carcinogenesis is likely due to variation in the response to tumor promotion rather than tumor initiation. It has been determined that as many as four loci influence susceptibility to TPA promotion (27)
. The nonuniform B6D2 background very likely accounts for the observed variability of ODC activity measurements (Figs. 2
and 3
) as well as the variable tumor counts of normal littermate controls in the chemical carcinogenesis studies (Fig. 5)
, and may also account for the different magnitude of resistance to chemical carcinogenesis observed in the two K6-AZ lines. Future studies in which these transgenes are expressed on inbred backgrounds will address these possibilities.
Conclusions and Future Directions.
Transgenic mice using the K5 or K6 promoters to drive ODC expression demonstrated that increased ODC activity is sufficient to promote tumor development from initiated skin cell populations (35
, 36) . Studies with K5-AZ and K6-AZ transgenic mice now indicate that increased ODC activity and polyamine levels are necessary events in the tumor promotion phase of two-stage chemical carcinogenesis. These experiments provide a significant mechanistic extension to similar conclusions reached in mice treated with DFMO (32
, 37) because in the transgenic model AZ suppresses polyamine production specifically within the skin cell populations targeted by tumor initiators. Other studies have shown that the inhibition of melanoma cell growth after interleukin-1 treatment is mediated by increased AZ expression (65)
, malignant oral keratinocytes have reduced levels of AZ mRNA compared with normal cells (66)
, and inducible expression of transfected AZ can stimulate apoptosis of Ras-transformed NIH-3T3 cells as well as human tumor cell lines (67)
. However, the results presented here directly demonstrate for the first time that AZ overexpression suppresses tumor development in an animal cancer model and strongly support the concept that ODC is a valid target for chemoprevention strategies.
The K5-AZ and K6-AZ mice therefore provide a valuable model system to evaluate the role of increased ODC activity and polyamines in the numerous transgenic mouse skin cancer models that have used the K5, K6, or other keratin promoters to express growth factors, signal transduction proteins, and cell cycle or apoptosis regulators. In addition, the role of ODC and polyamines in tumor progression to invasive squamous cell carcinomas and UV-induced skin carcinogenesis can be evaluated using these mice. Finally, these mice are a valuable model to study the role of ODC activity and polyamines in other skin functions such as wound healing.
|
ACKNOWLEDGMENTS |
|---|
T205-AZ plasmid. We thank Dr. John Mitchell (Department of Biological Sciences, Northern Illinois University, DeKalb, IL) for the affinity-purified AZ antibody. |
FOOTNOTES |
|---|
1 Supported by NIH Grants CA-18138 and GM-26290 (to A. E. P.) and CA-82768 (to L. M. S.). 
2 To whom requests for reprints should be addressed, at Department of Cellular and Molecular Physiology (H166), Room C4737, 500 University Drive, Hershey, PA 17033. Phone: (717) 531-8152; Fax: (717) 531-5157; E-mail: aep1{at}psu.edu
3 The abbreviations used are: ODC, ornithine decarboxylase; DFMO, 
-difluoromethylornithine, AZ, antizyme; TPA, 12-O-tetradecanoylphorbol-13-acetate; DMBA, 7, 12-dimethylbenz[a]anthracene; K5, keratin 5; K6, keratin 6; ORS, outer root sheath.
4 D. J. Feith and A. E. Pegg, unpublished results.
Received 2/14/01. Accepted 6/19/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Difluoromethylornithine, an irreversible inhibitor of ornithine decarboxylase, inhibits tumor promoter-induced polyamine accumulation and carcinogenesis in mouse skin. Proc. Natl. Acad. Sci. USA, 79: 6028-6032, 1982.-difluoromethylornithine. Biochem. Biophys. Res. Commun., 105: 969-976, 1982.[Medline]
-difluoromethylornithine, an irreversible inhibitor of ornithine decarboxylase, inhibits B16 melanoma-induced angiogenesis in ovo and the proliferation of vascular endothelial cells in vitro. Cancer Res., 50: 4131-4138, 1990.This article has been cited by other articles:
|
|
 |
|
  H. Tang, K. Ariki, M. Ohkido, Y. Murakami, S. Matsufuji, Z. Li, and K.-i. Yamamura Role of ornithine decarboxylase antizyme inhibitor in vivo Genes Cells, January 1, 2009; 14(1): 79 - 87. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, D. J. Feith, P. Welsh, C. S. Coleman, C. Lopez, P. M. Woster, T. G. O'Brien, and A. E. Pegg Studies of the mechanism by which increased spermidine/spermine N1-acetyltransferase activity increases susceptibility to skin carcinogenesis Carcinogenesis, November 1, 2007; 28(11): 2404 - 2411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Isome, M. J. Lortie, Y. Murakami, E. Parisi, S. Matsufuji, and J. Satriano The antiproliferative effects of agmatine correlate with the rate of cellular proliferation Am J Physiol Cell Physiol, August 1, 2007; 293(2): C705 - C711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. P. Ivanov and J. F. Atkins Ribosomal frameshifting in decoding antizyme mRNAs from yeast and protists to humans: close to 300 cases reveal remarkable diversity despite underlying conservation Nucleic Acids Res., March 19, 2007; 35(6): 1842 - 1858. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Petros, G. F. Graminski, S. Robinson, M. R. Burns, N. Kisiel, R. F. Gesteland, J. F. Atkins, D. L. Kramer, M. T. Howard, and R. S. Weeks Polyamine Analogs with Xylene Rings Induce Antizyme Frameshifting, Reduce ODC Activity, and Deplete Cellular Polyamines J. Biochem., November 1, 2006; 140(5): 657 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Lopez-Contreras, C. Lopez-Garcia, C. Jimenez-Cervantes, A. Cremades, and R. Penafiel Mouse Ornithine Decarboxylase-like Gene Encodes an Antizyme Inhibitor Devoid of Ornithine and Arginine Decarboxylating Activity J. Biol. Chem., October 13, 2006; 281(41): 30896 - 30906. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Kim, U. Mangold, C. Waghorne, A. Mobascher, L. Shantz, J. Banyard, and B. R. Zetter Regulation of cell proliferation by the antizyme inhibitor: evidence for an antizyme-independent mechanism J. Cell Sci., June 15, 2006; 119(12): 2583 - 2591. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. E. Pegg Regulation of Ornithine Decarboxylase J. Biol. Chem., May 26, 2006; 281(21): 14529 - 14532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Feith, S. Origanti, P. L. Shoop, S. Sass-Kuhn, and L. M. Shantz Tumor suppressor activity of ODC antizyme in MEK-driven skin tumorigenesis Carcinogenesis, May 1, 2006; 27(5): 1090 - 1098. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Curtin, M. Hanausek, Z. Walaszek, R. Zoltaszek, J. E. Swauger, A. T. Mosberg, and T. J. Slaga Short-Term Biomarkers of Cigarette Smoke Condensate Tumor Promoting Potential in Mouse Skin Toxicol. Sci., January 1, 2006; 89(1): 66 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Sgarlato, C. L. Eastman, and H. H. Sussman Panel of Genes Transcriptionally Up-regulated in Squamous Cell Carcinoma of the Cervix Identified by Representational Difference Analysis, Confirmed by Macroarray, and Validated by Real-Time Quantitative Reverse Transcription-PCR Clin. Chem., January 1, 2005; 51(1): 27 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Newman, A. Mobascher, U. Mangold, C. Koike, S. Diah, M. Schmidt, D. Finley, and B. R. Zetter Antizyme Targets Cyclin D1 for Degradation: A NOVEL MECHANISM FOR CELL GROWTH REPRESSION J. Biol. Chem., October 1, 2004; 279(40): 41504 - 41511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Y. Y. Fong, D. J. Feith, and A. E. Pegg Antizyme Overexpression in Transgenic Mice Reduces Cell Proliferation, Increases Apoptosis, and Reduces N-Nitrosomethylbenzylamine-induced Forestomach Carcinogenesis Cancer Res., July 15, 2003; 63(14): 3945 - 3954. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Y. Y. Fong, R. Mancini, H. Nakagawa, A. K. Rustgi, and K. Huebner Combined Cyclin D1 Overexpression and Zinc Deficiency Disrupts Cell Cycle and Accelerates Mouse Forestomach Carcinogenesis Cancer Res., July 15, 2003; 63(14): 4244 - 4252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Schenkel, S. Hanke, C. De Lorenzo, R. Schmitt, and B. M. Mechler P Elements Inserted in the Vicinity of or Within the Drosophila snRNP SmD3 Gene Nested in the First Intron of the Ornithine Decarboxylase Antizyme Gene Affect Only the Expression of SmD3 Genetics, June 1, 2002; 161(2): 763 - 772. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Shantz, Y. Guo, J. A. Sawicki, A. E. Pegg, and T. G. O'Brien Overexpression of a dominant-negative ornithine decarboxylase in mouse skin: effect on enzyme activity and papilloma formation Carcinogenesis, April 1, 2002; 23(4): 657 - 664. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Coleman, A. E. Pegg, L. C. Megosh, Y. Guo, J. A. Sawicki, and T. G. O'Brien Targeted expression of spermidine/spermine N1-acetyltransferase increases susceptibility to chemically induced skin carcinogenesis Carcinogenesis, February 1, 2002; 23(2): 359 - 364. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
