Cancer Research Cancer Health Disparities Conference 2009
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ravanko, K.
Right arrow Articles by Hölttä, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ravanko, K.
Right arrow Articles by Hölttä, E.
[Cancer Research 60, 5244-5253, September 15, 2000]
© 2000 American Association for Cancer Research

Molecular Biology and Genetics

Loss of p27Kip1 from Cyclin E/Cyclin-dependent Kinase (CDK) 2 but not from Cyclin D1/CDK4 Complexes in Cells Transformed by Polyamine Biosynthetic Enzymes1

Kirsi Ravanko, Kristiina Järvinen, Aino Paasinen-Sohns and Erkki Hölttä2

Department of Pathology, Haartman Institute, University of Helsinki, FIN-00014 Helsinki, Finland


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cancer cells are known to display up-regulation of ornithine decarboxylase (ODC) and S-adenosylmethionine decarboxylase (AdoMetDC), the key enzymes in the biosynthesis of polyamines that are essential for cellular proliferation. We have shown previously that overexpression of ODC or AdoMetDC alone can induce tumorigenic transformation of rodent fibroblasts. Because the subversion of normal cell cycle control is thought to be a crucial event in cancer development, we examined ODC- and AdoMetDC-transformed fibroblasts for alterations in the cell cycle components. The level of cyclin D1 and cyclin D1-dependent kinase and total cyclin-dependent kinase (CDK) 4 activities were elevated in the ODC transformants and particularly in the AdoMetDC transformants. Cyclin E content was not elevated, but a moderate increase in cyclin E-dependent kinase activity was seen in both cells. Total CDK2 activity was increased only in the ODC-transformed cells. The amount of the p27Kip1 CDK inhibitor was greatly decreased in both transformants. Nevertheless, p27Kip1 was present in the active cyclin D1/CDK4 complexes in the cells but absent from the cyclin E/CDK2 complexes. Restoration of p27Kip1 expression in the ODC- and AdoMetDC-transformed cells by transfection resulted in growth inhibition, but not in morphological reversion. An elevation in the level of hyperphosphorylated retinoblastoma protein was observed mainly in the ODC-transformed cells. These results suggest that the expression of ODC or AdoMetDC may affect cell cycle regulation in many ways. However, the largest common effect, which is therefore potentially relevant to some aspects of transformation, appears to be the constitutive down-regulation of p27Kip1 and its loss from the cyclin E/CDK2 complexes.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cyclins, CDKs,3 CKIs, and pRb are central components of the cell cycle clock assumed to govern the proliferation of normal cells. Progression through the cell cycle is controlled by sequential synthesis of a series of cyclins and concurrent down-regulation of CKIs, resulting in the assembly and activation of specific cyclin/CDK complexes, which, in turn, are thought to direct the phosphorylation of certain key cell cycle effectors, such as pRb. There is ample evidence that the G1 phase of the cell cycle, which is regulated by stimulatory and inhibitory growth factors, is the most important step with respect to the control of cell proliferation (reviewed in Refs. 1 and 2 ).

In normal growth-stimulated cells, the levels of various D-type cyclins (cyclin D1, D2, and D3) are increased in mid-G1, the levels of cyclin E are increased in late G1, the levels of cyclin A are increased in S phase, and the levels of cyclin B are increased in G2-M phase (reviewed in Ref. 2 ). Subsequently, the D-type cyclins and cyclin E associate with CDK4/6 (3 , 4) and CDK2 (reviewed in Ref. 2 ), respectively, which are expressed at a constant rate during the cell cycle, to form the active kinase complexes. The activity of CDKs is also regulated by specific dephosphorylation and phosphorylation reactions of the CDKs and by interactions with distinct CKIs (5) as well as through cyclin binding. There are two classes of CKIs: (a) the INK4 family consisting of p15, p16, p18, and p19; and (b) the Cip/Kip family of proteins comprised of p21Cip1, p27Kip1, and p57Kip2. The INK4 family members exclusively inhibit cyclin D-associated kinase activity (reviewed in Ref. 6 ), whereas the Cip/Kip family proteins inhibit a broader range of CDKs including cyclin D/CDK4/6, cyclin E/CDK2, and cyclin A/CDK2 complexes (reviewed in Ref. 6 ; see Refs. 7, 8, 9, 10 ). The activity of the cyclin D/CDK4/6 reaches its maximum in mid-to-late G1, and the activity of the cyclin E/CDK2 complex reaches its maximum at the G1-S-phase transition and then decreases in S phase, G2, and M phase (11 , 12) . One major function of the cyclin D/CDK4/6 and cyclin E/CDK2 complexes is thought to be phosphorylation of pRb, although cyclin E/CDK2 may also have other important substrates (13 , 14) . In quiescent cells, pRb is in a hypophosphorylated active form, but after growth stimulation, it starts to become phosphorylated in mid-G1, and maximal phosphorylation occurs at the G1-S-phase junction. Recent evidence suggests that pRb must first be partially phosphorylated by cyclin D/CDK4/6 complexes before it can serve as the substrate for additional phosphorylations by the cyclin E/CDK2 complex (15) . However, cyclin E/CDK2 may also recognize unphosphorylated pRb and phosphorylate it (16) . Nevertheless, normally, the full phosphorylation of pRb requires the action of both these CDKs. The increased phosphorylation of pRb leads to its functional inactivation, resulting in the release of S-phase-specific transcription factors, such as E2F, that are bound to and sequestered by unphosphorylated pRb during G1. The phosphorylation of pRb is continued during the S phase and G2 phase by the action of the cyclinA/CDK2 complex. Thus, pRb eventually becomes phosphorylated at multiple different sites. This is likely to have distinct effects on the interactions of pRb (17) . Finally, pRb is dephosphorylated in the later stages of mitosis.

In transformed cells, the cell cycle clock is typically derailed. A multitude of genetic alterations and other changes in the cell cycle components have been observed in different cancer cells. For example, cyclins D1 and D2 and CDK4 may show activating mutations, rearrangements, amplifications, or deregulated expression in various human malignancies, resulting in increased cyclinD/CDK4 activity (reviewed in Refs. 1 and 18 ). Similarly, cyclin E is often overexpressed in various human cancers, such as breast cancer (reviewed in Ref. 19 ). Several transformed cell lines have also been found to show alterations in the composition of the cyclin/CDK complexes (20 , 21) . Furthermore, inactivating mutations or deletions are frequently found in genes encoding p53 (22) , which regulates the p21Cip1 levels, pRb (23) , and the INK4a gene products p16 and p19Arf (1 , 24, 25, 26) in different human cancer cell lines and malignant tumors. Moreover, pRb is functionally inactivated by phosphorylation in many cancer cell lines. However, it is not often clear whether the observed changes are a cause or a consequence of transformation, nor is it known by which mechanisms these cell cycle alterations cause cellular transformation.

In this work, we studied the possible disturbances of the cell cycle clock in NIH3T3 and Rat-1 cells transformed by overexpression of ODC or AdoMetDC, the two key regulatory enzymes of polyamine biosynthesis. The polyamines (putrescine, spermidine, and spermine) are known to be essential for normal cell proliferation, and growth stimulation of normal cells is invariably associated with a transient activation of ODC (and AdoMetDC, to a lesser degree; reviewed in Refs. 27, 28, 29 ). In contrast, cells transformed by various carcinogens and oncogenes such as v-src, neu, myc, and ras seem to exhibit a growth factor-independent constitutive increase in ODC activity (30, 31, 32) . These results, combined with the fact that overexpression of ODC (33 , 34) or AdoMetDC alone can induce tumorigenic transformation of rodent fibroblasts,4 raise the possibility that ODC and AdoMetDC may contribute to the cellular transformation induced by many different factors. This makes the ODC- and AdoMetDC-transformed cells a good model for studies of alterations in the cell cycle apparatus associated with cell transformation and for studying the question of whether or not there could be a common pathway leading to transformation.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Culture.
NIH3T3 cell lines were from Dr. Clifford Tabin (Center for Cancer Research, Massachusetts Institute of Technology, Cambridge, MA) or the ATCC (ATCC CRL 1658; Manassas, VA). The former NIH3T3 cells were stably transfected with the neomycin resistance gene (neo) alone (referred to as N1) or together with human ODC cDNA [referred to as Odc (33) ]. The Odc-n cell line was derived from a tumor induced by ODC-overexpressing NIH3T3 cells in nude mice (34) . For comparison, the NIH3T3 cell line from ATCC was also stably transfected with the ODC expression vector. The expression level/activity of ODC achieved in the latter transformants was lower (i.e., less than half) than that achieved in the former ones (33) , and the latter transformants also had a less transformed phenotype. Nevertheless, similar results, but with less marked changes, were obtained with these transformants as well. Results are shown for the N1 and Odc-n pair of cells (33) , which provided a better model for studies of transformation-associated changes. Transfection of the NIH3T3 cells from ATCC with human AdoMetDC cDNA resulted in full transformation, appropriate for this study. The cells transfected with neo were designated as 4N cells, and those transfected with AdoMetDC cDNA were designated as Amdc-s cells.4 It is also of note that the ODC- and AdoMetDC-transformed cells differ in their polyamine patterns: the former display a preferential increase in putrescine (34) ; and the latter display a preferential increase in spermine.4 In addition, Rat-1 cells stably transfected with AdoMetDC cDNA (called Rat-1 Amdc-s)4 were used in the studies.

The cells were cultured in DMEM containing penicillin, streptomycin, gentamicin, and 5% (v/v) FCS or newborn calf serum (Life Technologies, Inc.) at 37°C in a 5% CO2 atmosphere.

Extraction of Whole Cell Proteins.
The cells were grown for 2 or 3 days, harvested by centrifugation, washed twice with PBS, and then suspended directly into LSB lacking 2-mercaptoethanol. The samples were sonicated for 10 s and clarified by centrifugation at maximal speed in an Eppendorf microcentrifuge for 10 min at 4°C. Protein concentrations were determined by using the BCA Protein Assay Reagent (Pierce), and then 2-mercaptoethanol was added to a final concentration of 5%. All of the analyses below were repeated at least three times.

Immunoprecipitation of pRb.
The cells were collected as described above, washed twice with PBS, and lysed in an immunoprecipitation buffer [50 mM HEPES (pH 7.0), 150 mM NaCl, 10% glycerol, 1% Triton X-100, 0.25% sodium deoxycholate, 1.5 mM MgCl2, 1 mM EGTA, 100 mM NaF, 2 mM Na3VO4, 10 µg/ml aprotinin, 10 µg/ml leupeptin, and 1 mM p-aminoethylbenzenesulfonyl fluoride]. The samples were kept on ice for 30 min and clarified by centrifugation at maximal speed in an Eppendorf microcentrifuge for 10 min at 4°C. Protein concentrations were determined by using the Bio-Rad Protein assay kit.

Equal amounts of total soluble proteins (1.5 mg) were first preadsorbed with 5 µl of normal rabbit serum at 4°C for 1 h with gentle rotation. Thereafter, pRb was incubated with 2 µg of rabbit polyclonal anti-Rb antibody (C-15; Santa Cruz Biotechnology, Inc.) for 2 h at 4°C with rotation. Immunocomplexes were harvested with goat antirabbit IgG-agarose (Sigma, St. Louis, MO), washed twice with lysis buffer, and suspended in LSB. Samples were heated at 100°C for 5 min and subjected to SDS-PAGE.

Western Blotting.
The whole cell protein extracts (50 µg) or the immunoprecipitates were separated on 8–12% SDS-polyacrylamide gels and transferred to nitrocellulose filters by fast semidry blotting (Biometra) or using the Mini Trans-Blot cell (Bio-Rad). The filters were incubated in blocking buffer [25 mM Tris (pH 8.0), 125 mM NaCl, 0.1% Tween 20, 2% BSA, and 0.1% NaN3] overnight at room temperature and then incubated with the specific antibody diluted in blocking buffer for 2 h at room temperature. The filters were rinsed five times in washing buffer [10 mM Tris (pH 8.0), 150 mM NaCl, 0.05% NP40, and 0.05% Tween 20] and incubated with horseradish peroxidase-conjugated rabbit antimouse IgGs (DAKO) or swine antirabbit IgGs (DAKO) for 30 min at room temperature. The low concentrations of p21Cip1 were probed using a sandwich system of biotinylated secondary antibodies (DAKO; 1:4000) and horseradish peroxidase-conjugated streptavidin (Sigma; 1:2000). Finally, the filters were rinsed five times with the washing buffer, rinsed for 15 min with high-salt buffer [10 mM Tris (pH 8.0) and 300 mM NaCl], and rinsed three times with TBS [10 mM Tris (pH 8.0) and 150 mM NaCl]. The bands were visualized by enhanced chemiluminescence (Pierce) and by exposing Fuji RX film to the filters. Equal loading was assessed by staining the membranes with Ponceau S solution (Sigma) and blotting with actin (see below).

The antibodies used were: (a) rabbit polyclonal antibodies to CDK2 (M-2), CDK4 (C-22), CDK6 (C-21), cyclin D2 (M-20), cyclin D3 (C-16; all from Santa Cruz Biotechnology, Inc.), and cyclin E [M-20 (Santa Cruz Biotechnology, Inc.) or 06-459 (Upstate Biotechnology, Inc.)]; and (b) mouse monoclonal antibodies to cyclin D1 [clone DCS-6 (from Dr. J. Partek), 72-13G (Santa Cruz Biotechnology, Inc.), or Ab3 (Calbiochem)], p27Kip1 [clone 57 (Transduction Laboratories)], p21Cip1 (sx118), pRb [G3-245 (PharMingen)], and actin [Ab-1 (Oncogene Research Products)].

cDNA Microarray and Northern Blot Analysis.
Polyadenylated mRNA was isolated by oligodeoxythymidylic acid cellulose chromatography (31 , 32) . The mRNA expression levels were initially examined using the Atlas mouse cDNA expression array I (Clontech Laboratories, Palo Alto, CA). Before probe synthesis by reverse transcription, the polyadenylated RNA samples were treated with DNase I as described in the Clontech Laboratories expression array user manual. cDNA probes were synthesized using [{alpha}-32P]dATP (Amersham), and the membranes were prehybridized, hybridized, and washed in a hybridization oven according to the manufacturer’s instructions. Gene expression was determined by scanning with a Fuji BAS-2500 phosphorimager using MacBas 2.5 software. The signal intensities on the arrays were normalized and quantified relative to the control housekeeping genes ubiquitin and ß-actin. Genes with more than 2-fold induction or repression in repeated experiments were selected for further analysis.

Northern blot analysis was used to give a more accurate assessment of the changes in mRNA expression. mRNA (8 µg) was separated by 0.8% agarose/formaldehyde gel electrophoresis, transferred to a nylon membrane (Hybond-N; Amersham Pharmacia Biotech), and hybridized with [{alpha}-32P]dCTP (Amersham)-labeled cDNA inserts. The cDNA of cyclin D1 (pHsCYCD1-H123) was from Dr. D. Beach (Institute of Child Health, London, United Kingdom), p27Kip1 cDNA (pSG5/p27) was from Dr. M. Laiho (University of Helsinki, Helsinki, Finland), and actin was from Clontech Laboratories. Kodak Biomax MS film and Fuji BAS-2500 phosphorimager plates were exposed to the filters for quantitation.

In Vitro CDK Assay.
Immunocomplex CDK assays were performed essentially as described by Matsushime et al. (3) , with minor modifications. The cells were collected by scraping and centrifugation, washed twice with PBS, suspended in immunoprecipitation lysis buffer [50 mM HEPES (pH 7.5), 150 mM NaCl, 1 mM EDTA, 0.1% Tween 20, 10% glycerol, 1 mM DTT, 10 µg/ml aprotinin, 10 µg/ml leupeptin, 1 mM AEBSF, 10 mM ß-glycerophosphate, 50 mM NaF, and 1 mM Na3VO4] and sonicated on ice three times for 5 s each. The samples were then frozen and thawed once and clarified by centrifugation at maximal speed in an Eppendorf microcentrifuge for 10 min at 4°C. Protein concentrations were determined using the Bio-Rad Protein assay kit. This protocol was found to efficiently extract the nuclear CDKs.

Equal amounts of proteins (0.8–1.5 mg) were incubated with 2 µg of anti-cyclin D1 [DCS-11 (Neo Markers) or 72-13G], anti-CDK2 (M-2), anti-cyclin E (M-20), or anti-CDK4 (C-22). The immunoreactions were carried out at 4°C for 2 h with rotation. Immunocomplexes were harvested with goat antirabbit or antimouse IgG-agarose (Sigma) and washed three times with the lysis buffer and twice with the kinase reaction buffer [50 mM HEPES (pH 7.5), 10 mM MgCl2, 5 mM MnCl2, 1 mM DTT, and 10 mM ß-glycerophosphate]. The immunocomplexes were then suspended on ice in 25 µl of kinase reaction buffer containing 20 µM ATP, 5 µCi of [{gamma}-32P]ATP, and 2 µg of GST-Rb fusion protein (Santa Cruz Biotechnology, Inc.) or histone H1 (Boehringer Mannheim) and incubated at 30°C for 30 min. The samples were centrifuged at maximal speed in an Eppendorf microcentrifuge for 20 s, and the supernatant was suspended in 5x LSB and boiled. Alternatively, the reaction was stopped by boiling the samples directly in LSB, with similar results. The reaction products were resolved in 10% SDS-PAGE and transferred to a nitrocellulose filter (Bio-Rad Trans-Blot Transfer Medium) followed by exposure of Fuji RX film.

Analysis of the Cyclin/CDK Complexes for the Presence of p27Kip1 and p21Cip1.
After decay of the radioactivity, the above-mentioned CDK assay filters or fresh, nonradioactive filters from the respective immunoprecipitations were analyzed for the composition of the immunoprecipitated protein complexes. The immunoprecipitates of anti-cyclin D1 were immunoblotted with polyclonal antibodies to CDK4 (C-22) and CDK6 (C-21), and the immunoprecipitates of anti-cyclin E were immunoblotted with antibodies to CDK2 (M-2). In addition, all of the immunoprecipitates mentioned in the previous section were analyzed for the presence of p27Kip1 and p21Cip1 by immunoblotting with monoclonal antibodies to p27Kip1 and p21Cip1.

Cell Cycle Analysis.
For analysis of the distribution of the cells in various phases of the cell cycle, the cells were suspended in a solution of 25 mM Tris-HCl (pH 7.4), 10 mM NaCl, 0.5% NP40, 5 mM MgCl2, and 0.2 mg/ml ethidium bromide and treated with 100 µg/ml RNase for 30 min at 37°C. The relative DNA content was determined by flow cytometric analysis (FACScan; Becton Dickinson, Mountain View, CA) using either the SFIT or SOBR (sum of broadened rectangles) model.

Transfection of the Cells with p27Kip1 Plasmids.
ODC- and AdoMetDC-transformed cells were transfected with two different p27Kip1 expression plasmids [pSG5/p27 (see above) or pRcKipA (from René Bernards, The Netherlands Cancer Institute, Amsterdam, the Netherlands)]. As a control, cells were transfected with the empty vector. Cells were grown on 6-well plates to 80% confluence and transfected with different amounts (0.2–1 µg) of p27Kip1 plasmid DNA and 0.1 µg of the puromycin resistance plasmid pBABE-puro (35) using the LipofectAMINE Plus-kit (Life Technologies, Inc.). The day after transfection, 100,000 cells were transferred to 9-cm-diameter plates to study the effect of p27Kip1 on cell growth. The rest of the cells were used for analysis of p27Kip1 expression by immunoblotting. Cell morphology was monitored daily, and puromycin selection (1.5 µg/ml puromycin) was started 2 days after transfection. After selection, the cells were photographed or fixed with 3.5% paraformaldehyde for 30 min and stained with 0.5% crystal violet for 2 h to count the colonies (>50 cells).


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The Levels of Cyclins, CDKs, and CKIs in NIH3T3 Cells Transformed by Overexpression of ODC or AdoMetDC.
Our understanding of cell cycle regulation is derived mostly from studies of serum-starved or density-arrested cells stimulated to reenter the cell cycle by serum addition. The analysis of transformed cells for possible alterations in cell cycle parameters is somewhat problematic because transformed cells, in contrast to normal cells, can grow (at least to a certain degree) independently of serum growth factors and without contact inhibition. Here we decided to examine the levels of various components of the cell cycle machinery, including the cyclins, CDKs, and CKIs, in the parental and ODC- and AdoMetDC-transformed NIH3T3 cells growing normally in the presence of serum growth factors in an exponential growth phase. To also test for any possible variations in the expression of cell cycle components at different times, the cells were grown for 2 or 3 days and then harvested for analysis. Whole cell protein extracts were analyzed by SDS-PAGE and immunoblotting with specific antibodies to cyclins D1, D2, D3, E, and A (Fig. 1A)Citation ; CDK2, CDK4 and CDK6 (Fig. 1B)Citation ; p27Kip1 and p21Cip1 (Fig. 2)Citation . The level of cyclin D1 was markedly increased in AdoMetDC-overexpressing cells, and a reproducible small increase in the cyclin D1 level was also seen in the ODC-transformed cells, as compared with the respective normal cells (Fig. 1A)Citation . In contrast, the level of cyclin D2 was surprisingly decreased, particularly in the AdoMetDC-transformed cells with respect to the normal controls. This is, however, in accordance with a recent report showing that increased expression of cyclin D2 may be associated with the induction of growth arrest or maintenance of a nonproliferative state (36) . Similarly, cyclin D3 levels were decreased in both transformed cells but were relatively more decreased in the ODC transformants than in the AdoMetDC transformants. Notably, a high abundance of cyclin D3 has recently been observed in quiescent and differentiating cells (37) , making sense of the finding of cyclin D3 reduction in transformed cells. The levels of cyclin E, cyclin A (Fig. 1A)Citation , and cyclin B (data not shown) remained essentially unchanged in the ODC- and AdoMetDC-overexpressing cells, although a small decrease in cyclin A was seen in the AdoMetDC transformants in some experiments. The amounts of CDK2 and CDK4 also remained unchanged in both the transformed cell lines, whereas the level of CDK6 was consistently decreased in the AdoMetDC-overexpressing cells, but not in the ODC-overexpressing cells (Fig. 1B)Citation . The significance of the decrease in CDK6 in the AdoMetDC-transformed cells, if any, remains to be elucidated.



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 1. Levels of different cyclins and CDKs in the ODC- and AdoMetDC-transformed cells versus normal NIH3T3 cells. Whole cell protein extracts (50 µg) were resolved on SDS-PAGE, transferred to nitrocellulose membranes, and immunoblotted with specific antibodies to (A) cyclins D1, D2, D3, E, or A and (B) CDK2, CDK4, or CDK6. 4N and N1, normal NIH3T3 controls; Amdc-s, NIH3T3 cells overexpressing AdoMetDC cDNA (in a sense orientation); Odc-n, NIH3T3 cells overexpressing ODC cDNA (see "Materials and Methods"). Molecular weight standards are indicated on the right.

 


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. The amount of p27Kip1 is profoundly decreased in the ODC- and AdoMetDC-transformed cells. A, NIH3T3 cells were grown to near confluence (95–100%) or to semiconfluence. B, normal and AdoMetDC-transformed Rat-1 cells were grown to near confluence (95–100%). Total cellular proteins (50 µg) were resolved on SDS-PAGE and immunoblotted with specific antibodies to p27Kip1 and p21Cip1. Equal loading was confirmed by immunoblotting the membrane with an antibody to actin. Molecular weight standards are indicated on the right.

 
Interestingly, the level of p27Kip1 was greatly decreased in both ODC- and AdoMetDC-transformed NIH3T3 cells (Fig. 2A)Citation . A similar decrease in the p27Kip1 level has been found in cells transformed by the Ras (38) , Myc (39) , and v-Src oncoproteins (40) , all of which are known to potently activate ODC (30 , 32 , 41) . It would be interesting to see whether the effects of the ras, myc, and v-src oncogenes on p27Kip1 could be mediated in part by ODC. Notably, we have also found in normal serum-stimulated fibroblasts that the activity of ODC can modulate the levels of p27Kip1 (unpublished data; see also Ref. 42 ), indicating that regulation of p27Kip1 by ODC can occur at physiological levels of ODC expression as well. Because the p27Kip1 level is known to increase in contact-inhibited normal cells, we studied whether the decrease in p27Kip1 in the ODC and AdoMetDC transformants could simply reflect the loss of contact inhibition in transformed cells by analyzing the p27Kip1 levels in half-confluent cell cultures. As seen in Fig. 2ACitation the amount of p27Kip1 protein was also decreased in the ODC- and AdoMetDC-overexpressing NIH3T3 cells relative to the normal cells unrestrained by growth density. This result was not limited to the NIH3T3 cells; the level of p27Kip1 was also strongly decreased in the AdoMetDC-transformed Rat-1 cells (Fig. 2B)Citation . Interestingly, in the immunoblottings, these AdoMetDC transformants appeared to show a faint protein band with a molecular weight slightly less than 27,000. The origin of this protein band remains to be elucidated, but it is possible that it is the recently reported degradation product of p27Kip1 (43) .

In contrast to p27Kip1, the amount of the p21Cip1 was increased in the AdoMetDC-overexpressing NIH3T3 cells, whereas no marked change in p21Cip1 was observed in the ODC-overexpressing NIH3T3 cells (Fig. 2A)Citation . The increase in p21Cip1 in the AdoMetDC transformants appeared to depend on the intensity of AdoMetDC expression (data not shown), similar to that found previously in the case of high signaling of the Ras and Raf proteins (44 , 45) .

The above-mentioned pattern of the cell cycle component alterations in the ODC- and AdoMetDC-transformed cells was the same, regardless of whether the analysis took place after 2 or 3 days of culture. It was further confirmed by cell cycle analyses with FACS (Fig. 3)Citation that there was no specific accumulation of the ODC- or AdoMetDC-transformed cells at any cell cycle phase that could have explained the observed differences in cell cycle parameters relative to the normal cells. Indeed, in all of the cell lines, about 60% of the total cell population was in the G1 phase at 3 days of culture (Fig. 3)Citation . After 2 days of culture, the percentage of G1 cells was slightly lower, and the proportion of the S-phase cells was respectively higher (data not shown), but this did not significantly affect the expression patterns of the cell cycle components or the conclusions made regarding their changes in the transformed cells.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 3. The cell cycle distribution of ODC- and AdoMetDC-transformed cells. Normal and ODC- or AdoMetDC-overexpressing cells were cultured in the presence of 5% serum for 3 days, and then their cell cycling profiles were analyzed by FACS. Y axis, cell number; X axis, DNA content.

 
Microarray and Northern Blot Analysis of Cyclin, CDK, and CKI Gene Expressions.
To study the mechanisms underlying the observed changes in the cell cycle proteins, we first screened the levels of mRNA expression by cDNA microarray analysis using the Atlas mouse cDNA expression array I from Clontech Laboratories. Pertinent sections from the arrays are shown in Fig. 4ACitation . Expression of cyclin D1 mRNA was found to be increased approximately 7-fold in the AdoMetDC-transformed cells, whereas the ODC transformants showed only a small increase in repeated experiments. The mRNA levels of the other cyclins (cyclins A, D2, D3, and E) as well as those of p21Cip1, p27Kip1 (Fig. 4A)Citation , and CDK4 (data not shown) did not show any appreciable (>=2-fold) changes after transformation.



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 4. cDNA microarray analysis of selected gene expression levels. Representative sections of the Atlas mouse cDNA expression array I (Clontech Laboratories) scanned with a phosphorimager are shown. In the array, each cDNA is spotted in duplicate. The top panel shows the expression of the different cyclins and p21Cip1 and p27Kip1, and the control genes are shown in the bottom panel. The expression levels were normalized to those of the housekeeping genes (ß-actin) and quantified densitometrically. B, Northern blot analysis of cyclin D1 and p27Kip1 mRNA expression. Equal amounts (8 µg) of polyadenylated RNA from normal and ODC- or AdoMetDC-transformed cells were hybridized with cyclin D1 and p27Kip1 probes. As a loading control, the membrane was hybridized with an actin cDNA probe.

 
For a more precise analysis, we then performed Northern blot analyses of the expression levels of cyclin D1 and p27Kip1. The analyses confirmed that cyclin D1 mRNA was increased 4.7-fold in the AdoMetDC-overexpressing cells, whereas the ODC-overexpressing cells showed no increase (Fig. 4B)Citation . The levels of p27Kip1 mRNA were found to be slightly (<=30%) decreased in the transformed cells as compared with their normal counterparts (Fig. 4B)Citation .

The Activities of the Cyclin D/CDK4 and Cyclin E/CDK2 Complexes in ODC- and AdoMetDC-transformed NIH3T3 Cells.
Next we analyzed the in vitro activities of the different G1 phase CDKs in the ODC- and AdoMetDC-transformed cells relative to normal cells. The cell lysates were immunoprecipitated with anti-cyclin D1, anti-cyclin E, anti-CDK2, or anti-CDK4 antibodies, and the kinase activities of the immunocomplexes were determined with [{gamma}-32P]ATP and GST-Rb fusion protein as a substrate. In addition, the activities of the anti-cyclin E and anti-CDK2 immunoprecipitates were determined with histone H1 as a substrate.

Total CDK4 activity was found to be elevated in both ODC- and AdoMetDC-overexpressing NIH3T3 cells as compared with normal NIH3T3 cells (Fig. 5)Citation . Likewise, analysis of the anti-cyclin D1 immunoprecipitates revealed an increase in kinase activity in both transformants, particularly in the AdoMetDC-transformed cells that overexpressed cyclin D1 (Fig. 5)Citation . The magnitude of the increases in the kinase activities was found to show some interexperimental variation due to an apparent oscillation of the CDK activities in the normal cells. This is to be expected because there may be changes in the cell cycle components in normal cells within a relatively narrow time period (at least in cells synchronized by serum starvation), although, on the other hand, cyclin D-associated kinase activity in continuously cycling cells (analyzed here) has been found to persist throughout the cell cycle (46) . Because cyclin D1 can form a complex with CDK4 or CDK6, we determined which one of these two kinases is in complex with cyclin D1 in these cells. Immunoblottings of anti-cyclin D1 immunoprecipitates with anti-CDK4 and anti-CDK6 revealed the presence of CDK4 but not CDK6 in the complexes (data not shown). This is in agreement with earlier studies showing that CDK4 is the major partner of cyclin D1 in rodent fibroblasts (3) .



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 5. Cyclin D1- and cyclin E-dependent kinase and total CDK2 and CDK4 activities in the ODC- and AdoMetDC-transformed versus normal NIH3T3 cells. Equal amounts (0.8–1.5 mg) of total proteins from the cell lysates were immunoprecipitated with 2 µg of {alpha}-cyclin D1, {alpha}-CDK4, {alpha}-cyclin E, or {alpha}-CDK2 antibodies. The in vitro immunocomplex kinase assays were performed with GST-pRb fusion protein as the substrate, as described in "Materials and Methods." The reaction products were resolved on SDS-PAGE, transferred to nitrocellulose membranes, and exposed to X-ray film.

 
The cyclin E-associated CDK2 activity was also found to be modestly elevated in the ODC- and AdoMetDC-overexpressing NIH3T3 cells (Fig. 5)Citation . The total CDK2 activity (resulting from both the cyclin E- and cyclin A-dependent kinase activities) was increased in the ODC-transformed cells, but not in the AdoMetDC-transformed cells (Fig. 5)Citation . Similar results were obtained when histone H1 was used as a substrate (data not shown).

p27Kip1 Is Present in the Active Cyclin D1/CDK4 Complexes but not in the Cyclin E/CDK2 Complexes in the ODC and AdoMetDC Transformants.
Recent studies have indicated that besides acting as inhibitors of CDKs, the Kip family proteins p21Cip1 and p27Kip1 can also promote the assembly of cyclin D-dependent kinase complexes and facilitate the nuclear accumulation of cyclin D (47) . Having found a marked decrease in total p27Kip1 in the ODC- and AdoMetDC-transformed cells, we investigated how this might affect the amount of p27Kip1 in the various cyclin/CDK complexes. We reprobed the filters of the anti-cyclin D1, anti-cyclin E, anti-CDK2, and anti-CDK4 immunoprecipitates used for analysis of the kinase activities with an antibody to p27Kip1. As seen in Fig. 6Citation A, p27Kip1 was present in the complexes immunoprecipitated with anti-cyclin D1 and anti-CDK4 antibodies in both the normal and transformed cells. In the cells transformed by AdoMetDC, the amount of p27Kip1 was increased in the cyclin D1/CDK4 complexes in relation to that seen in the normal cells, evidently due to the increase in these complexes as a result of the increased cyclin D1 expression. Hence, despite the pronounced decrease in p27Kip1 in the transformed cells, the cyclin D1/CDK4 complexes retained p27Kip1. Respective blottings with an antibody to p21Cip1 revealed that in AdoMetDC-transformed cells (showing an increase in p21Cip1), the cyclin D1/CDK4 complexes also contained p21Cip1 in increased amounts (Fig. 6B)Citation . These data, together with the finding of increased cyclin D-dependent kinase activity in the immunocomplex kinase assays in these transformants, are in accord with the recently presented findings suggesting that p27Kip1 and p21Cip1 are essential activators of cyclin D-dependent kinases (47 , 48) . In contrast to the cyclin D/CDK4 immunocomplexes, the anti-cyclin E and anti-CDK2 immunoprecipitates from the ODC- and AdoMetDC-transformed cells showed a profound decrease in p27Kip1 as compared with normal cells (Fig. 6A)Citation . p21Cip1 was not detected in the cyclinE/CDK2 complexes from either normal or transformed cells (data not shown). Reprobing the filters with anti-cyclin E and anti-CDK2 antibodies confirmed that the immunoprecipitates from normal and transformed cells contained the same amounts of cyclin E and CDK2, as expected from the analysis of the total content of these proteins in the cells. Thus, the transformed cells appear to show a selective loss of p27Kip1 from the cyclin E/CDK 2 complexes.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 6. Loss of p27Kip1 from the cyclin E/CDK2 complexes but not from the cyclin D/CDK4 complexes in the ODC- and AdoMetDC-transformed NIH3T3 cells. Equal amounts of total proteins from the cell lysates were immunoprecipitated with {alpha}-cyclin D1, {alpha}-CDK4, {alpha}-cyclin E, or {alpha}-CDK2 antibodies as described in the Fig. 5Citation legend. The immunoprecipitated proteins were resolved on SDS-PAGE and immunoblotted with {alpha}-p27Kip1 (A) and {alpha}-p21Cip1 (B).

 
Ectopic Expression of p27Kip1 in ODC- and AdoMetDC-transformed Cells Inhibits Growth but Does Not Reverse the Morphology of the Cells.
To study the significance of the loss of p27Kip1 for different aspects of transformation, the ODC- and AdoMetDC-overexpressing cells were transfected with two different p27Kip1 expression vectors. In an attempt to restore the physiological levels of p27Kip1 and not only to induce its overexpression, different doses of the expression vectors and empty vectors (controls) were introduced into the cells. pBABE-puro plasmid was cotransfected for selection. The cell morphology was then monitored by microscopy, and the number of puromycin-resistant colonies in the p27Kip1-transfected and mock-transfected cells was calculated. The ectopic expression of the p27Kip1gene in the ODC- and AdoMetDC-overexpressing NIH3T3 cells (Fig. 7A)Citation did not revert the transformed morphology of the cells to normal (Fig. 8)Citation but markedly reduced the growth rate of the cells (Fig. 7B)Citation .



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 7. The effect of ectopic expression of p27Kip1 on the ODC- and AdoMetDC-transformed cells. A, ODC- and AdoMetDC-overexpressing NIH3T3 cells were transfected with a p27Kip1 expression vector (pSG5/p27) or an empty vector as described in "Materials and Methods." The amount of p27Kip1 was analyzed by immunoblotting with {alpha}- p27Kip1. Equal loading was confirmed by immunoblotting the membranes with an antibody to actin. B, the effect of p27Kip1 on the growth of the cells. Cells (105) transfected with the p27Kip1 plasmid or empty vector were selected in the presence of puromycin. The outgrowth of puromycin-resistant colonies (>50 cells) was calculated after staining with 0.5% crystal violet.

 


View larger version (144K):
[in this window]
[in a new window]
  		Fig. 8. Ectopic expression of p27Kip1 in the ODC- and AdoMetDC-transformed cells does not revert their morphology to normal. ODC- and AdoMetDC-overexpressing NIH3T3 cells were transfected with a p27Kip1 expression vector (pSG5/p27) or an empty vector (see Fig. 7Citation ), and stable transfectants were photographed.

 
Rb Phosphorylation in ODC- and AdoMetDC-transformed Cells.
Because pRb is the only known substrate for cyclin D1dependent kinase and is assumed to be the key substrate of the cyclin E-dependent kinase as well, we analyzed the in vivo phosphorylation status of pRb in the ODC and AdoMetDC transformants. Because the amount of pRb is low in NIH3T3 cells, pRb was first immunoprecipitated from the cell lysates containing 1.5 mg of total protein with a polyclonal anti-Rb antibody and then subjected to SDS-PAGE and immunoblotted with a monoclonal antibody to pRb recognizing both the unphosphorylated and phosphorylated forms of pRb. The ODC-transformed cells displayed a clear increase in the hyperphosphorylated form of pRb, which migrated in the gel more slowly than the hypophosphorylated form, when compared with the normal parental cells (N1), which had pRb predominantly in a hypophosphorylated form (Fig. 9)Citation . It should be noted that using a specific inhibitor of ODC ({alpha}-difluoromethylornithine), we have also found evidence that ODC promotes the phosphorylation of pRb in normal, serum-stimulated fibroblasts (unpublished data; see also Refs. 49 and 50 ). Unexpectedly, the AdoMetDC-transformed cells showed only a minor increase in the hyperphosphorylated form of pRb as compared with normal controls (4N; Fig. 9Citation ). However, this result is in accordance with the finding that in these cells there is only a small increase in cyclin E-dependent kinase and total CDK2 activities (Fig. 5)Citation , contributing for the most part to pRb phosphorylation (15) .



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 9. The phosphorylation status of pRb in the ODC- and AdoMetDC-transformed and normal NIH3T3 cells. Equal amounts (1.5 mg) of total protein extracts were immunoprecipitated with 2 µg of polyclonal {alpha}-Rb. The resulting immunoprecipitates were resolved on SDS-PAGE and immunoblotted with monoclonal {alpha}-Rb antibody. Molecular weight standards are indicated on the right.

 

   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
It is currently believed that the loss of normal cell cycle control plays an important role in the genesis of most, if not all, cancers. However, in many cases, it is difficult to say whether the observed failure in cell cycle regulation is a cause or a consequence of transformation. Nevertheless, it is firmly established that most human cancers show inactivation of two central regulatory pathways of the cell cycle, the pRb and p53 pathways. However, the perturbation of these two pathways is not sufficient to induce malignant transformation of normal, diploid human cells, and additional genetic alterations are required (51) . It remains to be seen whether these pathways could finally converge to a common molecular change(s) in the cell cycle critical for transformation. The transformation of immortalized rodent fibroblasts can instead be achieved by activation/inactivation of one or two collaborating oncogenes/tumor suppressor genes. Here we used the ODC- and AdoMetDC-transformed NIH3T3 and Rat-1 cells as a model to study possible cell cycle-associated alterations related to transformation. The fact that ODC is essential for normal cell proliferation, up-regulated in a variety of human cancer cells (reviewed in Ref. 52 ), activated by various oncogenes (including v-src, ras, raf, neu, and myc), (30, 31, 32) and capable of transforming established cell lines by itself (33 , 34) and of transforming primary cells in conjunction with other oncogenes (53, 54, 55) makes the ODC-transformed cell particularly well suited for studying possible universal cell cycle perturbation(s) associated with transformation.

Our analyses of the different cyclins revealed that the expression of cyclin D1 was markedly increased in the AdoMetDC-overexpressing NIH3T3 cells and also that the ODC-transformed cells showed a modest increase in cyclin D1 expression. In AdoMetDC transformants, the elevation in cyclin D1 protein levels was found to be due, at least in part, to enhanced mRNA expression. Interestingly, the increase in cyclin D1 was selective among the cyclin D family members, and the amounts of cyclins D2 and D3 were even decreased. The level of CDK4, the main partner of cyclin D1, remained unchanged. A similar up-regulation of cyclin D1 with a constant CDK4 level has been observed in ras oncogene-transformed NIH3T3 cells (56 , 57) . Moreover, the increase in cyclin D1 has recently been shown to be necessary, although not sufficient, for oncogenesis induced by ras in mouse skin (56 , 58) . Many cyclin D1 gene transfer experiments with cultured cells or animals have likewise indicated that cyclin D1 may be involved in cellular transformation and tumor formation. For example, overexpression of cyclin D1 in mammary epithelial cells of transgenic mice is reported to result in abnormal cell proliferation and development of breast adenocarcinomas (59) . The cyclin D1 gene is also a direct target of the ß-catenin/LEF-1 pathway implicated in the development of colon cancer (60) . Conversely, expression of the antisense cyclin D1 cDNA construct in colon carcinoma cells can cause loss of tumorigenicity of the cells in nude mice (61) . Cyclin D1 is also frequently overexpressed in various human cancers, such as parathyroid adenoma, lymphoma, and breast cancer, as a result of different genetic changes (see Ref. 1 and references therein). Hence, a substantial body of evidence indicates that cyclin D1 can have oncogenic activity. It is thus possible that the marked increase in cyclin D1 in the AdoMetDC-overexpressing cells is one important factor contributing to cellular transformation. Why then is the cyclin D1 level preferentially increased in AdoMetDC-transformed cells but not in cells overexpressing ODC, which could rather be expected from the temporal order of expression of these two enzymes in relation to cyclin D1 expression in normal cells during the G1 phase of the cell cycle? One explanation could be that cyclin D1 is not only acting to regulate the activity of CDKs but contributes to some aspects of transformation independently of the CDKs (62, 63, 64) . In this context, it is interesting to note that increased expression of cyclin D1 has recently been linked to the invasiveness of tumor cells (57) , which we have found to be one distinct difference between the AdoMetDC- and ODC-transformed cells.4

The activity of cyclin D1/CDK4 complexes was found to be elevated in both the ODC- and AdoMetDC-transformed cells. This has previously been shown to be true for cells transformed by ras (65) and myc (46) , suggesting that an increase in cyclin D1/CDK4 activity may be a relatively common event in cellular transformation. However, the observed changes in cyclin D1/CDK4 activity in the ODC-transformed cells were not very impressive, and the magnitude of the increase varied somewhat, calling into question its overall significance for transformation.

The activity of the cyclin E-dependent kinase was also found to be modestly elevated in both the ODC and AdoMetDC transformants in the in vitro immunocomplex kinase assays. The same finding has been reported previously for ras- and c-myc-overexpressing cells (66) . Notably, ODC is a direct transcriptional target of c-Myc (41) and is also potently up-regulated by activated ras (32) , making it tempting to speculate that the effects of Ras and Myc could be mediated in part through ODC.

Most strikingly, the ODC- and AdoMetDC-transformed cells displayed a profound decrease in p27Kip1, which can inhibit the activity of all CDKs, although it preferentially inhibits the activity of cyclin E/CDK2 (6) . The level of p27Kip1 has also been found to be decreased in rat fibroblasts after the activation of v-src (40) , ras, and myc (66) . Moreover, down-regulation of p27Kip1 is frequently seen in various human cancers, such as prostate (67 , 68) , breast (69) , non-small cell lung (70) , colorectal (71) , gastric (72) , and oral carcinomas (73) . Indeed, the amount of p27Kip1 present has been found to be a good prognostic indicator in various types of cancer (reviewed in Refs. 74 and 75 ). Hence, p27Kip1 has tumor suppressor-like properties. Indeed, p27Kip1 may be a novel type of tumor suppressor that is haploinsufficient for tumor suppression (76) . However, mutations in p27Kip1 seem to be rare in the human cancer cells. However, targeted disruption of the p27Kip1 gene is known to result in enhanced growth of mice, multiple organ hyperplasia, and predisposition to tumors (77, 78, 79) . In our experiments, transfection of p27Kip1 into the ODC- and AdoMetDC-transformed cells did not return the transformed morphology of the cells to normal but significantly reduced the growth rate of the cells. Therefore, p27Kip1 may not be directly involved in regulation of the actual transformation process but may be involved in regulation of the proliferative capacity of the transformed cells.

Most studies have shown that the level of p27Kip1 in the cells is regulated mainly at the posttranslational level by proteolytic degradation (9 , 74) . This is also probably true for the ODC- and AdoMetDC-transformed cells because we found only a small (<=30%) decrease in the p27Kip1 mRNA levels in these cells. The degradation of p27Kip1 is known to occur primarily through the ubiquitin-proteasome pathway (43 , 80) , although other mechanisms may also contribute to its degradation (43 , 81) . What signals p27Kip1 to undertake the degradation is still poorly understood. Phosphorylation of p27Kip1 is probably one important means of marking the protein for degradation. The phosphorylation of p27Kip1 may be brought about by cyclin E/CDK2 (13 , 39 , 82 , 83) , although it is likely that phosphorylation of p27Kip1 may also be brought about by other kinases (84) .

Intriguingly, recent studies have shown that p27Kip1 (together with p21Cip1) is necessary for the assembly of the cyclin D/CDK4/6 complexes (47) . Thus, a decrease in p27Kip1 could lead to a failure in the formation of these complexes. However, we observed that despite a significant decrease in p27Kip1 in the ODC- and AdoMetDC-transformed cells, the formation of the cyclin D1/CDK4/6 complexes was normal. In contrast, there was a strong reduction in p27Kip1 in the complexes of cyclin E/CDK2 in these two transformants as compared with that in normal cells. This suggests that the cellular transformation is preferentially associated with an altered function of the latter kinase complex. In normal serum-stimulated fibroblasts, p27Kip1 has been found to dissociate from cyclin E/CDK2 complexes in a Ras-regulated manner (38) . Because Ras is known to increase the amount of the cyclin D1/CDK4 complexes, which require p21Cip1 and/or p27Kip1, the latter of which is suggested to become titrated from the cyclin E/CDK2 complexes. Similarly, c-Myc has been shown to transiently induce the expression of cyclin D1 and/or cyclin D2, causing sequestration of p27Kip1 from cyclin E complexes (85 , 86) . The same could also hold true for the AdoMetDC-transformed cells showing a constitutive marked increase in p27Kip1 in the cyclinD1/CDK complexes. However, this kind of binding and sequestering of p27Kip1 by the cyclin D1/CDK4 complexes cannot solely explain the loss of p27Kip1 from the cyclin E/CDK2 complexes in the ODC-transformed cells, which displayed only a slight increase in the cyclin D1/CDK4 complexes. Therefore, other p27Kip1-dissociating regulatory mechanisms are likely to exist.

pRb is considered to be the major target of the cyclin D- and E-dependent kinases. In this study, we found a clear increase in the phosphorylation of pRb in the ODC-transformed cells, whereas the AdoMetDC-overexpressing cells showed only a marginal elevation in the hyperphosphorylated form of pRb. Hence, the phosphorylation status of pRb did not seem to strictly correlate with the transformation state of these cells. The loss of p27Kip1 from the cyclin E/CDK2 complexes in both the ODC- and AdoMetDC-transformed cells gives us yet another reason to speculate that there could be a substrate(s) other than pRb that may become specifically phosphorylated by the cyclin E-dependent kinase in the transformed cells. For example, one possibility is that p27Kip1 not only inhibits CDK2 but also affects the localization of the CDK2 complexes and thereby affects the substrate availability or specificity of the cyclin E-dependent kinase. On the other hand, the possibility that p27Kip1 could also have growth-regulatory functions unrelated to CDK activity cannot be excluded. Interestingly, p27Kip1 has recently been shown to induce an as yet unknown protease that can cleave cyclin A (87) . However, we did not detect any significant amounts of cyclin A cleavage product correlating with the p27Kip1 levels in our cells. Altogether, the overall constitutive down-regulation of p27Kip1 and its specific loss from the cyclin E/CDK2 complexes represent the largest alteration of the cell cycle machinery in common for the ODC and AdoMetDC transformants and could therefore be potentially relevant to some aspects of transformation. However, it is clear from the present results that overexpression of ODC and AdoMetDC affects the cell cycle in multiple ways, all of which may contribute to transformation. Notably, unlike the ODC-transformed cells (33) the AdoMetDC-transformed cells do not show an increase in their proliferation rate, which could indicate that the observed cell cycle component changes do not only reflect the proliferation differences between normal and transformed cells but could somehow be specifically related to transformation. The mechanisms by which ODC and AdoMetDC bring about these changes, and which of these changes are primary or secondary ones, remain to be elucidated.

Note Added in Proof
Recently, S. K. Gilmour et al. (88) have also reported that ODC overexpression stimulates cyclin E/CDK2 activity and proliferation in the skin of transgenic mice. However, in contrast to our data, they paradoxically found an increase in the levels of the CKIs p21Cip1 and p27Kip1 that, as speculated, could be due to the observed induction of differentiation or apoptosis of some cells within the skin (specifically, the follicular cells directed to overexpress ODC).


   ACKNOWLEDGMENTS
 
We thank Jiri Partek for the cyclin D1 antibody, Marikki Laiho and René Bernards for p27Kip1 plasmids, and Monica Schoulz for FACS analyses.


   FOOTNOTES
 
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.

1 Supported by the University of Helsinki, the Finnish Cancer Organizations, and the Finnish Academy of Sciences. Back

2 To whom requests for reprints should be addressed, at Haartman Institute, Department of Pathology, University of Helsinki, P. O. Box 21 (Haartmaninkatu 3), FIN-00014 University of Helsinki, Finland. Phone: 358-9-1912-6516; Fax: 358-9-1912-6675; E-mail: Erkki.Holtta{at}Helsinki.fi Back

3 The abbreviations used are: CDK, cyclin-dependent kinase; CKI, CDK inhibitor; ODC, ornithine decarboxylase; pRb, retinoblastoma protein; Rb, retinoblastoma; AdoMetDC, S-adenosylmethionine decarboxylase; ATCC, American Type Culture Collection; LSB, Laemmli sample buffer; GST, glutathione S-transferase; FACS, fluorescence-activated cell sorting. Back

4 A. Paasinen-Sohns, T. Eloranta, A. Laine, O. A. Jänne, M. Birrer, and E. Hölttä, submitted for publication. Back

Received 12/ 6/99. Accepted 7/18/00.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sherr C. J. Cancer cell cycles. Science (Washington DC), 274: 1672-1677, 1996.[Abstract/Free Full Text]
  2. Sherr C. J. G1 phase progression: cycling on cue. Cell, 79: 551-555, 1994.[Medline]
  3. Matsushime H., Quelle D. E., Shurtleff S. A., Shibuya M., Sherr C. J., Kato J. D-type cyclin-dependent kinase activity in mammalian cells. Mol. Cell. Biol., 14: 2066-2076, 1994.[Abstract/Free Full Text]
  4. Meyerson M., Harlow E. Identification of G1 kinase activity for cdk6, a novel cyclin D partner. Mol. Cell. Biol., 14: 2077-2086, 1994.[Abstract/Free Full Text]
  5. Morgan D. O. Principles of CDK regulation. Nature (Lond.), 374: 131-134, 1995.[Medline]
  6. Sherr C. J., Roberts J. M. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev., 9: 1149-1163, 1995.[Free Full Text]
  7. Xiong Y., Hannon G. J., Zhang H., Casso D., Kobayashi R., Beach D. p21 is a universal inhibitor of cyclin kinases. Nature (Lond.), 366: 701-704, 1993.[Medline]
  8. Harper J. W., Adami G. R., Wei N., Keyomarsi K., Elledge S. J. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell, 75: 805-816, 1993.[Medline]
  9. Polyak K., Lee M. H., Erdjument-Bromage H., Koff A., Roberts J. M., Tempst P., Massague J. Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell, 78: 59-66, 1994.[Medline]
  10. Toyoshima H., Hunter T. p27 a novel inhibitor of G1 cyclin-Cdk protein kinase activity, is related to p21. Cell, 78: 67-74, 1994.[Medline]
  11. Dulic V., Lees E., Reed S. I. Association of human cyclin E with a periodic G1-S phase protein kinase. Science (Washington DC), 257: 1958-1961, 1992.[Abstract/Free Full Text]
  12. Koff A., Giordano A., Desai D., Yamashita K., Harper J. W., Elledge S., Nishimoto T., Morgan D. O., Franza B. R., Roberts J. M. Formation and activation of a cyclin E-cdk2 complex during the G1 phase of the human cell cycle. Science (Washington DC), 257: 1689-1694, 1992.[Abstract/Free Full Text]
  13. Sheaff R. J., Groudine M., Gordon M., Roberts J. M., Clurman B. E. Cyclin E-CDK2 is a regulator of p27Kip1. Genes Dev., 11: 1464-1478, 1997.[Abstract/Free Full Text]
  14. Zhao J., Dynlacht B., Imai T., Hori T., Harlow E. Expression of NPAT, a novel substrate of cyclin E-CDK2, promotes S-phase entry. Genes Dev., 12: 456-461, 1998.[Abstract/Free Full Text]
  15. Lundberg A. S., Weinberg R. A. Functional inactivation of the retinoblastoma protein requires sequential modification by at least two distinct cyclin-cdk complexes. Mol. Cell. Biol., 18: 753-761, 1998.[Abstract/Free Full Text]
  16. Geng Y., Whoriskey W., Park M. Y., Bronson R. T., Medema R. H., Li T., Weinberg R. A., Sicinski P. Rescue of cyclin D1 deficiency by knockin cyclin E. Cell, 97: 767-777, 1999.[Medline]
  17. Harbour J. W., Luo R. X., Dei Santi A., Postigo A. A., Dean D. C. Cdk phosphorylation triggers sequential intramolecular interactions that progressively block Rb functions as cells move through G1. Cell, 98: 859-869, 1999.[Medline]
  18. Hunter T., Pines J. Cyclins and cancer. II. Cyclin D and CDK inhibitors come of age. Cell, 79: 573-582, 1994.[Medline]
  19. Donnellan R., Chetty R. Cyclin E in human cancers. FASEB J., 13: 773-780, 1999.[Abstract/Free Full Text]
  20. Xiong Y., Zhang H., Beach D. Subunit rearrangement of the cyclin-dependent kinase is associated with cellular transformation. Genes Dev., 7: 1572-1583, 1993.[Abstract/Free Full Text]
  21. Xiong Y., Kuppuswamy D., Li Y., Livanos E. M., Hixon M., White A., Beach D., Tlsty T. D. Alteration of cell cycle kinase complexes in human papillomavirus E6- and E7-expressing fibroblasts precedes neoplastic transformation. J. Virol., 70: 999-1008, 1996.[Abstract]
  22. Greenblatt M. S., Bennett W. P., Hollstein M., Harris C. C. Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res., 54: 4855-4878, 1994.[Free Full Text]
  23. Horowitz J. M., Park S. H., Bogenmann E., Cheng J. C., Yandell D. W., Kaye F. J., Minna J. D., Dryja T. P., Weinberg R. A. Frequent inactivation of the retinoblastoma anti-oncogene is restricted to a subset of human tumor cells. Proc. Natl. Acad. Sci. USA, 87: 2775-2779, 1990.[Abstract/Free Full Text]
  24. Chin L., Pomerantz J., DePinho R. A. The INK4a/ARF tumor suppressor: one gene—two products—two pathways. Trends Biochem. Sci., 23: 291-296, 1998.[Medline]
  25. Ruas M., Peters G. The p16INK4a/CDKN2A tumor suppressor and its relatives. Biochim. Biophys. Acta, 1378: F115-F177, 1998.[Medline]
  26. Sherr C. J. Tumor surveillance via the ARF-p53 pathway. Genes Dev., 12: 2984-2991, 1998.[Free Full Text]
  27. Tabor C. W., Tabor H. Polyamines. Annu. Rev. Biochem., 53: 749-790, 1984.[Medline]
  28. Pegg A. E. Recent advances in the biochemistry of polyamines in eucaryotes. Biochem. J., 234: 249-262, 1986.[Medline]
  29. Heby O., Persson L. Molecular genetics of polyamine synthesis in eukaryotic cells. Trends Biochem. Sci., 15: 153-158, 1990.[Medline]
  30. Hölttä E., Auvinen M., Andersson L. C. Polyamines are essential for cell transformation by pp60v-src: delineation of molecular events relevant for the transformed phenotype. J. Cell Biol., 122: 903-914, 1993.[Abstract/Free Full Text]
  31. Sistonen L., Hölttä E., Lehväslaiho H., Lehtola L., Alitalo K. Activation of chimeric EGF-R/neu tyrosine kinase induces the fos/jun transcription factor complex, glucose transporter and ornithine decarboxylase. J. Cell Biol., 109: 1911-1919, 1989.[Abstract/Free Full Text]
  32. Hölttä E., Sistonen L., Alitalo K. The mechanisms of ornithine decarboxylase deregulation in c-Ha-ras oncogene-transformed NIH 3T3 cells. J. Biol. Chem., 263: 4500-4507, 1988.[Abstract/Free Full Text]
  33. Auvinen M., Paasinen A., Andersson L. C., Hölttä E. Ornithine decarboxylase activity is critical for cell transformation. Nature (Lond.), 360: 355-358, 1992.[Medline]
  34. Auvinen M., Laine A., Paasinen-Sohns A., Kangas A., Kangas L., Saksela O., Andersson L. C., Hölttä E. Human ornithine decarboxylase-overproducing NIH3T3 cells induce rapidly growing, highly vascularized tumors in nude mice. Cancer Res., 57: 3016-3025, 1997.[Abstract/Free Full Text]
  35. Morgenstern J. P., Land H. Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res., 18: 3587-3596, 1990.[Abstract/Free Full Text]
  36. Meyyappan M., Wong H., Hull C., Riabowol K. T. Increased expression of cyclin D2 during multiple states of growth arrest in primary and established cells. Mol. Cell. Biol., 18: 3163-3172, 1998.[Abstract/Free Full Text]
  37. Bartkova J., Lukas J., Strauss M., Bartek J. Cyclin D3: requirement for G1/S transition and high abundance in quiescent tissues suggest a dual role in proliferation and differentiation. Oncogene, 17: 1027-1037, 1998.[Medline]
  38. Aktas H., Cai H., Cooper G. M. Ras links growth factor signaling to the cell cycle machinery via regulation of cyclin D1 and the Cdk inhibitor p27KIP1. Mol. Cell. Biol., 17: 3850-3857, 1997.[Abstract]
  39. Müller D., Bouchard C., Rudolph B., Steiner P., Stuckmann I., Saffrich R., Ansorge W., Huttner W., Eilers M. Cdk2-dependent phosphorylation of p27 facilitates its Myc-induced release from cyclin E/cdk2 complexes. Oncogene, 15: 2561-2576, 1997.[Medline]
  40. Johnson D., Frame M. C., Wyke J. A. Expression of the v-Src oncoprotein in fibroblasts disrupts normal regulation of the CDK inhibitor p27 and inhibits quiescence. Oncogene, 16: 2017-2028, 1998.[Medline]
  41. Bello-Fernandez C., Packham G., Cleveland J. L. The ornithine decarboxylase is a transcriptional target of c-Myc. Proc. Natl. Acad. USA, 90: 7804-7808, 1993.[Abstract/Free Full Text]
  42. Ray R. M., Zimmerman B. J., McCormack S. A., Patel T. B., Johnson L. R. Polyamine depletion arrests cell cycle and induces inhibitors p21Waf1/Cip1, p27Kip1 and p53 in IEC-6 cells, Am. J. Physiol., 276: C684-C691, 1999.
  43. Shirane M., Harumiya Y., Ishida N., Hirai A., Miyamoto C., Hatakeyama S., Nakayama K-i., Kitagawa M. Down-regulation of p27Kip1 by two mechanisms, ubiquitin-mediated degradation and proteolytic processing. J. Biol. Chem., 274: 13886-13893, 1999.[Abstract/Free Full Text]
  44. Sewing A., Wiseman B., Lloyd A. C., Land H. High-intensity Raf signal causes cell cycle arrest mediated by p21Cip1. Mol. Cell. Biol., 17: 5588-5597, 1997.[Abstract]
  45. Woods D., Parry D., Cherwinski H., Bosch E., Lees E., McMahon M. Raf-induced proliferation or cell cycle arrest is determined by the level of Raf activity with arrest mediated by p21Cip1. Mol. Cell. Biol., 17: 5598-5611, 1997.[Abstract]
  46. Mateyak M. K., Obaya A. J., Sedivy J. M. c-Myc regulates cyclin D-Cdk4 and -Cdk6 activity but affects cell cycle progression at multiple independent points. Mol. Cell. Biol., 19: 4672-4683, 1999.[Abstract/Free Full Text]
  47. Cheng M., Olivier P., Diehl J. A., Fero M., Roussel M. F., Roberts J. M., Sherr C. J. The p21Cip1 and p27Kip1 CDK "inhibitors" are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J., 18: 1571-1583, 1999.[Medline]
  48. LaBaer J., Garrett M. D., Stevenson L. F., Slingerland J. M., Sandhu C., Chou H. S., Fattaey A., Harlow E. New functional activities for the p21 family of CDK inhibitors. Genes Dev., 11: 847-862, 1997.[Abstract/Free Full Text]
  49. Omura T., Yano Y., Hasuma T., Kinoshita H., Matsui-Yuasa I., Otani S. Involvement of polyamines in retinoblastoma protein phosphorylation. Biochem. Biophys. Res. Commun., 250: 731-734, 1998.[Medline]
  50. Kramer D. L., Vujcic S., Diegelman P., Alderfer J., Miller J. T., Black J. D., Bergeron R. J., Porter C. W. Polyamine analogue induction of the p53–p21WAF1/CIP1-Rb pathway and G1 arrest in human melanoma cells. Cancer Res., 59: 1278-1286, 1999.[Abstract/Free Full Text]
  51. Hahn W. C., Counter C. M., Lundberg A. S., Beijersbergen R. L., Brooks M. W., Weinberg R. A. Creation of human tumour cells with defined genetic elements. Nature (Lond.), 400: 464-468, 1999.[Medline]
  52. Meyskens F. L., Jr., Gerner E. W. Development of difluoromethylornithine (DFMO) as a chemoprevention agent. Clin. Cancer Res., 5: 945-951, 1999.[Abstract/Free Full Text]
  53. Clifford A., Morgan D., Yuspa S. H., Soler A. P., Gilmour S. Role of ornithine decarboxylase in epidermal tumorigenesis. Cancer Res., 55: 1680-1686, 1995.[Abstract/Free Full Text]
  54. O’Brien T. G., Megosh L. C., Gilliard G., Soler A. P. Ornithine decarboxylase overexpression is a sufficient condition for tumor promotion in mouse skin. Cancer Res., 57: 2630-2637, 1997.[Abstract/Free Full Text]
  55. Hölttä E., Paasinen-Sohns A., Povelainen M., Järvinen K., Ravanko K., Kangas A. Cell transformation by ornithine decarboxylase is associated with phosphorylation of the transactivation domain of c-Jun. Biochem. Soc. Trans., 26: 621-627, 1998.[Medline]
  56. Liu J. J., Chao J. R., Jiang M. C., Ng S. Y., Yen J. J. Y., Yang-Yen H. F. Ras transformation results in an elevated level of cyclin D1 and acceleration of G1 progression in NIH 3T3 cells. Mol. Cell. Biol., 15: 3654-3663, 1995.