Diminished G1 Checkpoint after γ-Irradiation and Altered Cell Cycle Regulation by Insulin-like Growth Factor II Overexpression*

High levels of insulin-like growth factor II (IGFII) mRNA expression are detected in many human tumors of different origins including rhabdomyosarcoma, a tumor of skeletal muscle origin. To investigate the role of IGFII in tumorigenesis, we have compared the mouse myoblast cell line C2C12-2.7, which was stably transfected with human IGFII cDNA and expressed high and constant amounts of IGFII, to a control cell line C2C12-1.1. A rhabdomyosarcoma cell line, RH30, which expresses high levels of IGFII and contains mutated p53, was also used in these studies. IGFII overexpression in mouse myoblast C2C12 cells causes a reduced cycling time and higher growth rate. After γ-irradiation treatment, C2C12-1.1 cells were arrested mainly in G0/G1 phase. However, C2C12-2.7 and RH30 cells went through a very short G1 phase and then were arrested in an extended G2/M phase. To verify further the effect of IGFII on the cell cycle, we developed a Chinese hamster ovary (CHO) cell line with tetracycline-controlled IGFII expression. We found that CHO cells with high expression of IGFII have a shortened cycling time and a diminished G1 checkpoint after treatment with methylmethane sulfonate (MMS), a DNA base-damaging agent, when compared with CHO cells with very low IGFII expression. It was also found that IGFII overexpression in C2C12 cells was associated with increases in cyclin D1, p21, and p53 protein levels, as well as mitogen-activated protein kinase activity. These studies suggest that IGFII overexpression shortens cell cycling time and diminishes the G1 checkpoint after DNA damage despite an intact p53/p21 induction. In addition, IGFII overexpression is also associated with multiple changes in the levels and activities of cell cycle regulatory components following γ-irradiation. Taken together, these changes may contribute to the high growth rate and genetic alterations that occur during tumorigenesis.

Insulin-like growth factor II (IGFII) 1 has been shown to play an important role in the development, growth, and survival of normal cells. IGFII is encoded by the imprinted Igf2 gene expressed only from the paternal allele in most tissues (1). The signaling of IGFII is mediated by the type I IGF receptor. We have previously shown that forced overexpression of IGFII in C2C12 cells leads to transformed characteristics (2). IGFII expressing cells exhibit a proliferative advantage. Moreover, elevated levels of IGFII have been detected in human tumors of various origins and may act through autocrine or paracrine signaling loops, which are often associated with increased levels of type I IGF receptor (3)(4)(5)(6)(7). In a transgenic carcinogenesis model involving simian virus 40 T antigen targeted to pancreatic ␤-islet cells, the initial proliferation switch is correlated with focal activation of IGFII, and reduced IGFII expression impairs tumor cell growth in vitro and in vivo (8). In addition, IGFII is further up-regulated in all pancreatic islet-cell tumors in the T antigen transgenic mice, and Igf2 gene disrupted mice developed fewer tumors of reduced size, a lower grade malignancy, and higher number of apoptotic cell bodies (8). IGFII may also act as a survival factor and inhibits apoptosis induced by cytokine deprivation, DNA damage, and a variety of chemotherapeutic agents, although this anti-apoptotic activity may not be a primary consequence of IGFII signaling (9 -11). The anti-apoptotic activity of IGFII could promote the accumulation of additional genetic abnormalities that lead to cell proliferation.
Cell cycle progression of eukaryotic cells is finely regulated by an intrinsic molecular clock comprised of cyclins and cyclinassociated kinases (12). The final decision of mammalian cells to replicate their DNA or to withdraw from the cell cycle with an unduplicated genome takes place in mid-to late G 1 phase, referred to as the restriction point (13). The commitment process at the G 1 checkpoint reflects a complicated integration of positive and negative extracellular and intracellular signals transduced by multiple cascades into the cell nucleus (14 -17). The existence of internal checkpoints at different stages of the cell cycle is an important feature to prevent the cell from prematurely entering the next phase before all the necessary macromolecular events have been completed. A key regulatory component of cell cycle progression is the tumor suppressor p53 which is normally expressed at very low levels in many different tissues due to the short half-life of the protein (18). Following DNA damage, p53 protein levels rise dramatically and promote the transcription of WAF1/CIP1 gene, the product of which, p21 WAF1CIP1 , causes growth arrest and delayed entry into the S phase until the genomic lesions are fully repaired (19,20).
When growth factors interact with their receptors at the cell surface, a cascade of phosphorylation events is triggered which transduce mitogenic signals to the nucleus, leading to DNA synthesis and subsequently to mitosis. Since all these extracellular stimuli must ultimately pass their signals to the cell cycle machinery itself, it is important to investigate the effects of growth factors on the cell cycle and to understand the molecular basis of the link between the upstream signaling pathways and the cell cycle clock . In our studies, we examined the mouse  myoblast cell line C2C12, stably transfected with human IGFII  cDNA, CHO cells with tetracycline-regulated IGFII expression,  and a RMS cell line, RH30, which expresses high levels of  IGFII and contains mutated p53. After DNA damage, cells with  IGFII overexpression had a diminished G 1 checkpoint arrest and an amplified G 2 /M phase despite intact p53/p21 induction. We also found that IGFII overexpression was associated with increases in basal levels of cyclin D1, p21, and p53 proteins, as well as mitogen-activated protein kinase (MAPK) activity. These effects of IGFII on the cell cycle may contribute to the high proliferation rate and accumulation of genetic changes during tumorigenesis.

MATERIALS AND METHODS
Stable Transfection of CHO-AA8 Tet-off Cells with pTRE-IGFII-CHO cells were transfected with two plasmids, Tet-off and pTRE-IGFII separately. CHO cells stably transfected with the Tet-off plasmid were purchased from CLONTECH (CHO-AA8 Tet-off, catalog number C3004-1). The tetracycline-regulatable IGFII plasmid pTRE-IGFII was generated by inserting a full-length human IGFII cDNA into the tetracycline-responsive vector pTRE (CLONTECH) between the EcoRI and XbaI site. The pTRE-IGFII was cotransfected with pTK-Hyg into the CHO-AA8 Tet-off cells as follows. Cells were trypsinized and resuspended in complete ␣-MEM medium at 10 million cells per ml. 0.4 ml of the resuspended cells was mixed with 40 g of pTRE-IGFII and 2 g of pTK-Hyg in a 0.4-cm cuvette. The cells were electroporated in a Bio-Rad Gene Pulsar at 960 microfarads, 0.22 kV/cm (t ϭ 20 -30 ms) and then allowed to stand at room temperature for 10 min. The cells were plated into 20 10-cm tissue culture plates containing 10 ml of complete ␣-MEM medium. The cells were allowed to grow for 48 h, and then hygromycin was added at 0.5 mg/ml. Cells were selected for 2 weeks or until single colonies were seen. About 200 clones were picked and subcultured in 6-well plates. Confluent clones were trypsinized and resuspended in 3 ml of complete media. One ml of clone was plated in duplicate into another 6-well plate. Complete ␣-MEM with 10 g/ml tetracycline was added to one plate and complete media to the other. The remaining 1 ml of clone was maintained in culture as a stock. After 48 h of treatment, the duplicate clone or set of clones with and without tetracycline were trypsinized with 50 -100 l of trypsin for 3 min, and then 1 ml of complete media was added to neutralize the trypsin. For dot blot, 50 l of the suspension was pipetted on to a dot blotter and filtered through a Nytran membrane presoaked in 10ϫ SSC. IGFII expression was checked by Northern blot as described below.
Cell Cultures and Treatment-The mouse myoblast cell lines, C2C12-1.1 and C2C12-2.7, were generated by C. P. Minniti and have previously been described (2). C2C12 cells were grown in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM glutamine, 50 units/ml penicillin, 50 mg/ml streptomycin, and 1 mg/ml G418 (geneticin) at 37°C, 5% CO 2 in a humidified incubator. CHO cells were grown in ␣-MEM supplemented with 10% tetracycline-free fetal bovine serum, 0.5 mg/ml G418, 1 mg/ml hygromycin B, and alternative 10 g/ml tetracycline. Cells were irradiated with a 137 Cs source at 5.0 Gray/min. In some experiments, cells were exposed to 1 g/ml nocodazole (Sigma) alone or plus MMS (Aldrich) at 40 g/ml for varying duration.
Cell Cycle Analysis-Exponentially growing cells were washed in PBS, fixed in ice-cold 70% ethanol for at least 1 h at 4°C. After washing in PBS containing 0.1% glucose, cells were treated with PBS staining buffer containing RNase A at 1 mg/ml, propidium iodide at 50 g/ml, 0.1% glucose in the dark at 4°C for 30 min, and filtered through a 35 M strainer cap (catalog number 2235, Becton Dickinson). A total of 10,000 stained cells were analyzed in a fluorescence-activated cell sorter (FAC-Scan; Becton Dickinson). The CELLQUEST software (Becton Dickinson) was used to determine the distribution of cells in the various cell cycle compartments as G 0 /G 1 , S, and G 2 /M.
Northern Blot-Total RNA was isolated by using the RNeasy Mini Kit (Qiagen) and loaded on a Nytran membrane (Schleicher & Schuell). The membrane was washed with 100 ml of 10ϫ SSC and UV crosslinked with Stratagene UV cross-linker. Dot blot or RNA blot was prehybridized in QuikHyb Hybridization Solution (Stratagene) at 65°C with 50 g/ml sheared salmon sperm DNA for 30 min and then hybridized with a random primer labeled probe (Amersham Pharmacia Bio-tech, rediprime DNA labeling system) in the QuikHyb Hybridization Solution (Stratagene) at 65°C for 2 h. The membrane was washed by using the QuikHyb washing procedures (Stratagene) and exposed to x-ray film. Quantitative results were obtained by using the NIH Image program. In some experiments, RNA transcript levels were normalized to glyceraldehyde-3-phosphate dehydrogenase expression.
Western Blot and Immunoprecipitation Assays-Exponentially growing cells were washed twice in cold PBS and lysed in lysis buffer (Tris-HCl 50 mM, pH 7.5; NaCl 120 mM; Nonidet P-40 0.5%; NaF 10 mM; Na 3 VO 4 1 mM; Na 4 P 2 O 7 1 mM; dithiothreitol 1 mM; 4-(2-amino ethyl)benzenesulfonyl fluoride hydrochloride 1 mM; leupeptin 20 g/ml; aprotinin 20 g/ml). Lysates were incubated for 30 min in ice (vortexed vigorously every 10 min) and clarified by centrifugation at 14,000 rpm for 2 min. Protein concentration was determined by the Bio-Rad Protein Assay (catalog number 500-0006, Bio-Rad). For Western blot analysis, proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Novex) and blotted onto a Protran nitrocellulose membrane (catalog number 00830N, Schleicher & Schuell). Membranes were blocked for 1 h in TBST (Tris-HCl 20 mM, pH 7.5; NaCl 138 mM; Tween 20 0.1%) containing 5% non-fat milk at room temperature, probed with primary antibody for 1 h, washed three times with TBST, probed again with horseradish peroxidase-conjugated secondary antibody for 45 min, and washed again three times in TBST. Antigenantibody reaction was revealed by using enhanced chemiluminescence (ECL) procedures according to the manufacturers' recommendation (Pierce). Mouse monoclonal anti-p53 and anti-p21 antibodies were purchased from Calbiochem. All other primary and secondary antibodies were purchased from Santa Cruz Biotechnology. For immunoprecipitation, 1 mg of cellular extracts was precleared by incubating extracts with protein A or protein G-agarose (Santa Cruz Biotechnology) for 1 h at 4°C. Extracts were then incubated with primary antibody overnight at 4°C, and immune complexes were collected by incubation (1 h, 4°C with rocking) with protein A-or protein G-agarose. Immunoprecipitates were washed four times with ice-cold lysis buffer, eluted in equal volume of 2ϫ SDS sample buffer (Tris-HCl 125 mM, pH 6.8; SDS 4%, glycerol 20%, and 0.005% bromphenol blue; Novex), and resolved on SDS-polyacrylamide gel electrophoresis. Antigen-antibody complexes were visualized with the enhanced chemiluminescence (ECL) detection system (Pierce).
Kinase Assay-Immunoprecipitation was performed as described above. Immunoprecipitates were washed three times with lysis buffer (content as described above) followed by washing three times with kinase buffer (Tris-HCl 20 mM, pH 7.5; MgCl 2 4 mM; dithiothreitol 1 mM). Protein A-or protein G-agarose beads were resuspended in 26 l of kinase buffer containing 3 Ci of [␥-32 P]ATP (NEN Life Science Products) and 3 g of histone H1 (Life Technologies, Inc.) for Cdk2 and Cdc2 kinases, 3 g of GST-Rb (Santa Cruz) for Cdk4 kinase, or 3 g of myelin basic protein (Sigma) for MAPK kinase. Kinase reactions were carried out for 30 min at 37°C and stopped by addition of 26 l of 2ϫ SDS sample buffer. The samples were heated (100°C, 5 min) and subjected to SDS-polyacrylamide gel electrophoresis. Gels were fixed and dried, and kinase activities were visualized and quantitated as described above. To verify the specificity of the Cdk4 kinase activity, we included a Cdk4 antibody-specific blocking peptide (Calbiochem) in the reaction and used the same amount of normal rabbit IgG (Santa Cruz) to replace Cdk4 antibody.

Cells Stably Expressing IGFII Show a Diminished G 1 Checkpoint after DNA Damage-
To study the effects of IGFII on cell cycle control and the growth properties of mammalian cells, we chose the mouse myoblast cell line C2C12. Cells stably transfected with human IGFII cDNA and expressing high and constant levels of IGFII were termed C2C12-2.7, and a control cell line was called C2C12-1.1 which contained the vector alone. An RMS cell line, RH30, which expresses high levels of IGFII and mutated p53, was also used in these experiments. IGFII expression altered the proliferation rate and the distribution of the cell cycle profile under normal culture conditions (Fig. 1A, 0 Gy). As demonstrated in Fig. 1A, increasing doses of ␥-irradiation leads to a much more prominent G 2 /M phase arrest in IGFII overexpressing C2C12-2.7 cells than in control C2C12-1.1 cells. Furthermore, the response to ␥-irradiation in C2C12-2.7 cells appears quite similar to the response pattern observed in the RMS tumor cell line RH30. To analyze further the effect of IGFII expression on cell response to DNA damage, we synchronized C2C12 cells by serum starvation for 48 h. Cells were then stimulated with serum and received ␥-irradiation at the same time. As shown in Fig. 1B, there is a prominent G 1 arrest in control C2C12 cells by 18 h following ␥-irradiation (compare C2C12-1.1 ϪIR to ϩIR). In contrast, the IGFII overexpressing cells have no detectable G 1 arrest following ␥-irradiation (compare C2C12-2.7 ϪIR to ϩIR). Since the C2C12-1.1 and C2C12-2.7 cells have the same genetic background except for different levels of IGFII expression, these results suggest that the diminished G 1 checkpoint in C2C12-2.7 cells after DNA damage is likely due to the overexpression of IGFII. It is noteworthy that no apoptosis was observed in either C2C12-1.1 or C2C12-2.7 cells 48 h following both serum depletion and 3-10 Gray ␥-irradiation (data not shown). Furthermore, in the absence of ␥-irradiation, 14.5% of C2C12-1.1 cells are in S/G 2 M compared with 23% of C2C12-2.7 cells 9 h following serum addition, and by 18 h, 29% of C2C12-2.7 cells have entered G 2M in contrast to only 15% of C2C12-1.1 cells, strongly suggesting that IGFII overexpression shortens G 1 phase (Fig. 1B).
To exclude that clonal variation between C2C12-1.1 and C2C12-2.7 cells was responsible for the observed changes, we developed another model using tetracycline-regulated IGFII expression in CHO cells. As shown in Fig. 2A, the addition of 10 g/ml tetracycline for 48 h (ϩTet) markedly decreased IGFII expression compared with CHO cells growing without tetracycline (ϪTet). The proliferation rate of CHO Ϫ tet cells is higher than in CHO ϩ tet cells (data not shown). Following 12 h of treatment with a microtubule polymerization inhibitor, nocodazole, CHO control cells (ϩtet) demonstrated a G 2 /M arrest as expected (Fig. 2B). When these cells were subjected to a DNA base-damaging agent, MMS, for 12 h, G 1 arrest was also seen (Fig. 2B, CHOϩtet, Noco ϩ MMS). In contrast, the addition of MMS for 12 h to nocodazole-treated CHO cells express-  2. IGFII expression shortens G 1 checkpoint both before and after DNA damage. A, Northern blot analysis of IGFII mRNA from CHO cells growing without tetracycline (ϪTet) or with 10 g/ml of tetracycline (ϩTet) for 48 h. After transferring total RNA to membrane, the blot was hybridized with the random primer-labeled IGFII probe. rRNA of 28 S and 18 S stained by ethidium bromide was displayed as a loading control. B, equal number of CHO cells was plated at low density and incubated for 48 h in medium with or without tetracycline (10 g/ml). CHO cells with tetracyclinecontrolled IGFII expression were then treated with 1 g/ml nocodazole (Noco) alone or plus 40 g/ml of base-damaging agent, MMS, for 12 h. Flow cytometric analysis of cells stained with propidium iodide was then performed.
ing high levels of IGFII (Ϫtet) caused no detectable G 1 block (Fig. 2B, CHO-tet, Noco ϩ MMS). When ␥-irradiation was used instead of MMS treatment, the same findings were observed (data not shown). Moreover, with 12 h nocodazole treatment, 91% of CHOϪtet cells accumulate in G 2 M phase compared with 60% of CHOϩtet cells. Furthermore, 18% of CHOϩtet cells stay in G 1 phase in comparison to 2.8% of CHOϪtet cells (Fig.  2B). These results further indicate that IGFII overexpression shortens G 1 phase of the cell cycle and diminishes the G 1 checkpoint after DNA damage.
Expression Levels of Cell Cycle Regulatory Proteins in C2C12 Cells-In order to investigate the role of IGFII on the kinetics of cell cycle regulatory protein expression after DNA damage, we prepared cell extracts from C2C12-1.1 and C2C12-2.7 cells at various time points following 7 Gy ␥-irradiation and determined the expression of a variety of cell cycle proteins by Western blotting. As shown in Fig. 3A, the basal cyclin D1 levels were consistently higher in C2C12-2.7 cells compared with control cells. Following exposure to ␥-irradiation, cyclin D1 levels increased slightly in both control and C2C12-2.7 cells. No changes were observed in the basal levels of cyclins E, A, B1, and p27 proteins between the two C2C12 cell lines. However, in both cell lines, cyclin A and cyclin B1 proteins increased at 8 h after irradiation and then decreased gradually to the level below the non-radiation group at 36 h. There was an additional lower molecular weight band of E2F-1 in C2C12-2.7 cells compared with C2C12-1.1 cells. After 7 Gray of ␥-irradiation, E2F-1 protein levels decreased in both cell lines. The cyclin-dependent kinases Cdk2, Cdc2, and Cdk4 had equivalent protein expression in both C2C12 cell lines. Furthermore, these levels did not change following ␥-irradiation (Fig. 3B). There were no changes in proliferating cell nuclear antigen protein levels both with and without ␥-irradiation (data not shown).
C2C12-2.7 cells with high levels of IGFII expressed elevated levels of p21 and p53 proteins (Fig. 4, A and B) compared with C2C12-1.1 control cells. In both high IGFII and control cell lines, p53 protein levels were increased after ␥-irradiation. The increases in p53 levels were observed at 3 h after irradiation and were dose-dependent in the range of 3-7 Gray. The relative increase of p53 protein after radiation compared with basal expression in C2C12-1.1 cells is higher than in IGFII overexpressing C2C12-2.7 cells. After 7 Gy ␥-irradiation, in both

FIG. 4. Changes in the inhibitors of cell cycle progression by IGFII expression after DNA damage. Immunoblot analyses of p53 protein level 24 h after indicated doses of ␥-irradiation (IR) treatment (A)
, p21 expression at indicated time points after 7 Gray of ␥-irradiation (B) were performed using antibodies specific for p53 (mouse monoclonal antibody) or p21 (rabbit polyclonal antibody). C, Northern blot analysis of p21 mRNA from C2C12-1.1 and C2C12-2.7 cells 24 h following indicated doses of ␥-irradiation. After transferring total RNA to a membrane, the blot was hybridized with a random primer-labeled p21 probe. D, histogram results were obtained by scanning the autoradiographic film of C and using 0 Gray of C2C12-1.1 cells treated as 100% control. E, Western blot analysis of pRb protein levels from C2C12-1.1 and C2C12-2.7 cells with or without ␥-irradiation.
C2C12 cell lines, p21 protein levels were also induced compared with the levels observed at time 0 h. However, the relative increase of p21 levels compared with basal expression in C2C12-1.1 cells was higher than in IGFII overexpressing C2C12-2.7 cells (seen on darker exposure). The induced p21 protein reached a maximum at 8 h and was sustained at least to 36 h. Constitutive levels of p21 RNA are also higher in IGFII overexpressing C2C12-2.7 cells compared with control cells. Twenty four hours following ␥-irradiation, the relative increase in p21 RNA was similar in C2C12-1.1 and C2C12-2.7 cells (Fig.  4, C and D). pRb protein level was similar in high IGFII expressing C2C12-2.7 cells compared with C2C12-1.1 control cells. After various doses of ␥-irradiation, pRb shifted from the hyperphosphorylated inactive form (higher band in Fig. 4E) to the hypophosphorylated active form (lower band in Fig. 4E) in both cell lines.
The Effects of IGFII Overexpression on MAPK Activity-Previous studies have demonstrated that the activation of the MEK-MAPK (also referred to as extracellular signal-regulated kinase, ERK) pathway is necessary for mitogen-induced G 1 /S phase cell cycle progression (21,22). Growth factors and oncogenes products can activate the MAPK pathway through a phosphorylation cascade involving Ras, Raf, and MAPKK for mitogenesis (14 -16, 23). We found that the MAPK (ERK-1 and ERK-2) protein levels were similar in C2C12-1.1 and C2C12-2.7 cells with or without irradiation (Fig. 5A). The basal MAPK (ERK-1 and ERK-2)-associated kinase activity toward the myelin basic protein substrate in IGFII-overexpressing C2C12-2.7 cells was higher than in C2C12-1.1 control cells (Fig. 5B, upper  panel). After 7 Gray ␥-irradiation, MAPK activity was decreased in extracts of both C2C12 cells. A Western blot of the immunoprecipitated MAPK used for the above kinase assay was performed to determine MAPK protein levels. As seen in Fig. 5B (lower panel), the alteration of kinase activity is not due to changes in protein levels.
IGFII Overexpression Partially Releases the Inhibition of Cdk Kinase Activities by ␥-Irradiation-We then examined the activities of cdks after ␥-irradiation in IGFII overexpressing and wild-type C2C12 cells. Histone H1-associated Cdk2 kinase activity decreased in C2C12-1.1 control cells after ␥-irradiation (Fig. 6, A and B). Only a mild decrease, however, in Cdk2 activity in IGFII overexpressing C2C12-2.7 cells was observed, and no change of Cdk2 activity occurred in RH30 tumor cells expressing high levels of IGFII. Similarly, histone H1-associ-ated Cdc2 kinase activity was significantly reduced in C2C12-1.1 control cells, but this was not seen in high IGFII expressing C2C12-2.7 cells nor in RH30 tumor cells even following 10 Gy of ␥-irradiation (Fig. 6, A and C). In addition, the kinase activity of Cdk4 toward the GST-Rb substrate decreased in C2C12-1.1 control cells 24 h after 3-7 Gy of ␥-irradiation, whereas there were no appreciable changes observed in C2C12-2.7 cells (upper panel in Fig. 6D). To verify the specificity of the Cdk4 kinase activity shown above, we included a Cdk4 antibodyspecific blocking peptide in the reaction or used normal rabbit IgG to replace Cdk4 antibody. It was found that Cdk4 antibodyspecific blocking peptide could block GST-Rb phosphorylation, and very weak GST-Rb phosphorylation was observed when Cdk4 antibody was replaced by normal rabbit IgG (lower panel in Fig. 6D).
Induced p21 Binds to Cyclin-Cdk Complexes-p21 has previously been shown to bind and inhibit the activities of cyclincdk complexes. To determine if elevated p21 in high IGFII expressing cells is associated with the cyclin-cdk complexes, we immunoprecipitated Cdk2, Cdc2, and Cdk4 from cell extracts that did not receive ␥-irradiation or 24 h after ␥-irradiation. We then examined the p21 protein levels in the cdk-containing complexes. Our results demonstrate that more p21 is associated with cdk complexes in IGFII-overexpressing C2C12-2.7 cells both under basal conditions and following ␥-irradiation when compared with C2C12-1.1 control cells (Fig. 7A). However, the relative increase of p21 protein after irradiation in C2C12-1.1 cells is higher than in IGFII-overexpressing C2C12-2.7 cells. We have performed a Western blot of immunoprecipitated cdks used in the above experiments. All Cdk2, Cdc2, and Cdk4 protein levels did not change (data not shown) as had been previously seen in Fig. 3B. However, because this assay reflects both free and p21-associated cdks, we next sought to compare the levels of cdks in the p21 complex. We immunoprecipitated p21 from cell extracts without or 24 h after ␥-irradiation. We then examined the cdk protein levels by Western blot. Our results demonstrate that more cdks are associated with p21 in C2C12-2.7 cells than C2C12-1.1 cells under normal growth conditions (Fig. 7C, 0 Gray). After ␥-irradiation, the levels of cdks associated with p21 were also higher in C2C12-2.7 cells especially following 10 Gray treatment (Fig. 7B). DISCUSSION We have shown that the IGFII mitogenic signaling pathway shortens the G 1 interval and appears to inhibit the block of the G 1 -to-S transition following DNA damage. IGFII overexpression is associated with relatively high cdk kinase activity after DNA damage which enables high IGFII expressing cells to quickly override the G 1 checkpoint. This short G 1 checkpoint arrest may lead to the lack of appropriate repair of damaged DNA, and subsequent propagation of genetic alterations may contribute to genetic instability in tumor cells overexpressing IGFII. Furthermore, the increased percentage of IGFII overexpressing cells in S and G 2 /M phase indicates that the cells have a higher proliferative rate.
We also found that cyclin D1 levels were increased in IGFII overexpressing cells which indicates that cyclin D1 may be a downstream target of the IGFII signaling pathway. Since cyclin D1 is required and is a rate-limiting factor for entry into S phase, the overexpression of cyclin D1 may accelerate the G 1to-S transition in high IGFII expressing cells (24 -26). Many alterations in G 1 -associated processes have been observed in cancer cells, of which cyclin D1 overexpression is the most common (27)(28)(29)(30). Overexpression of cyclin D1 contributes to the oncogenic transformation of cells in vitro and in vivo (31)(32)(33)(34). Induction of cyclin D expression is part of the mitogenic response of various mesenchymal or epithelial cells to serum or growth factors such as platelet-derived growth factor, epithelial growth factor, and insulin-like growth factor I (35)(36)(37)(38). All of these mitogens operate via signaling cascades involving tyrosine kinase receptors and G proteins at least in part through the Ras-Raf-MAPK pathway (14 -16). The role of D-type cyclins as potential sensors and integrators of diverse mitogenic stimuli with the cell cycle machinery is also confirmed by induction of cyclin D1 expression and G 1 phase acceleration by activated ras or raf oncogenes themselves (39,40).
Another significant finding of this study is that the p21 and p53 proteins are constitutively expressed in the high IGFIIexpressing cell line C2C12-2.7. The increase of p21 expression corresponds to an increased level of p21 mRNA. Despite the elevated basal levels of p53 and p21 protein, the proliferation rate of IGFII-overexpressing cells was higher compared with the control C2C12 cells. It has previously been reported that ectopic expression of cyclin D1 in asynchronously growing cells was accompanied by increased levels of the p53 tumor suppressor protein as well as the cyclin/cdk inhibitor p21 through an E2F transactivation mechanism (41,42). The induction of p21 does not lead to growth arrest of cells but rather to stabilization of cyclin D1/cdk function (41,42). It is known that p21 does not always function as an inhibitor of cyclin-cdk complexes. At low concentrations, p21 promotes the assembly of stable and active Cdk4 and D-type cyclin kinase complexes which allows the cells to continue to enter into S phase, resulting in a transformed phenotype. At high concentrations, such as following the stress of DNA damage, p21 inhibits kinase activity and cell growth (43)(44)(45)(46). A proposed explanation for this paradox is that cyclin D1-cdk complexes containing a single p21 molecule are still catalytically active, whereas complexes containing multiple p21 subunits are inactive (46,47). We found that following DNA damage, the relative increases in both p53 and p21 were higher in C2C12-1.1 cells than the increases in IGFII overexpressing C2C12-2.7 cells, which may lead to the longer G 1 checkpoint arrest in C2C12-1.1 cells compared with C2C12-2.7 cells. Following ␥-irradiation, cdk activities are decreased in both cell lines. However, the inhibition of cdks in high IGFIIexpressing C2C12-2.7 cells is diminished. This may be explained by two factors: higher cyclin D1 levels in IGFII-overexpressing C2C12-2.7 cells and a relative lower increase in p21 protein associated in all the cdk complexes after radiation when compared with C2C12-1.1 cells. Thus the ratio of p21 to cyclin D1-cdk complexes in IGFII-overexpressing C2C12-2.7 cells following DNA damage is less than the ratio in C2C12-1.1 control cells, which potentially leads to maintenance of stable, FIG. 7. Induced p21 binds to cdk complexes with or without ␥-irradiation. A, lysates from cells treated with different doses (0, 7, 10 Gray) of ␥-irradiation for 24 h were immunoprecipitated (IP) with various cdk antibodies followed by Western blot for p21 with monoclonal p21 antibody. B, after treatment with different doses (0, 7, 10 Gray) of ␥-irradiation for 24 h, cell lysates were immunoprecipitated (IP) with p21 antibody followed by Western blot for cdk proteins with various cdk antibodies. active cyclin D1-cdk-p21 complexes in C2C12-2.7 cells following ␥-irradiation.
It has been widely reported that expression of p21 is induced in a p53-dependent manner in response to DNA damage, resulting in G 1 /S arrest through inhibition of the cyclin-cdk complexes (48,49). On the other hand, p21 gene expression is also induced through a p53-independent mechanism during differentiation of a number of cell types, as well as in various murine tissues during development and in the adult animal (50 -53). Moreover, it was reported that expression of p21 is elevated following stimulation of quiescent cells with serum or purified growth factors, and in proliferating cells the majority of p21 protein is found in active cyclin-cdk complexes (46,52,54). Induction of p21 in these situations does not appear to require p53. In addition, ectopic expression of cyclin D1 correlated with increased levels of p21 protein, which did not lead to growth arrest (42). It has been reported that higher levels of p21 protein and RNA are observed in several different human tumors compared with the corresponding normal tissues and that the p21 is not mutated in these tumors, although the p21 gene possesses a polymorphism at codon 31 (55). The elevated p21 expression is related to tumor differentiation and is p53independent, which is a common feature of in vivo neoplasms (55). Taken together, our findings support the evidence that the stoichiometry of cyclin D1-cdk-p21 is critical in determining the activity of this complex.
We also observed that MAPK activity was induced in high IGFII-expressing C2C12-2.7 cells, which may be a result of the activation of the IGFII signaling pathway. It has been reported that the MAPK pathway signals, including Ras, Raf, MAPKK, and MAPK, can all transcriptionally induce p21 expression in response to growth factors such as epidermal growth factor, nerve growth factor, serum, and 12-O-tetradecanoylphorbol-13-acetate treatment (54). This report is consistent with the findings that forced expression of either v-Src or activated c-Raf in murine hematopoietic Baf-3 cells prevents down-regulation of p21 expression that normally occurs in these cells in response to growth factor withdrawal (56). In combination with other results in our study, it could be suggested that IGFII may directly activate mitogenic signaling pathways and drive cell cycle progression. However, it still remains to be investigated whether increased MAPK activity contributes to the elevated expression of cyclin D1, p21, and p53 in IGFII-overexpressing C2C12-2.7 cells.
It has been shown that E2F-1 can strongly transactivate the human p21 gene through E2F-binding sites that are located in the p21 gene and that the transactivation of the p21 gene is correlated with increased levels of endogenous E2F-1 and p21 proteins at the G 1 /S boundary (46). Curiously, we observed a lower molecular weight E2F-1 portion associated with the induced p21 protein levels seen in high IGFII expression cells compared with control cells. However, the role of E2F-1 in the regulation of p21 expression is not yet known. We did not observe obvious differences of cyclin E protein between high IGFII-expressing cells and control cells with or without DNA damage. Although cyclin E is also rate-limiting for entry into S phase and may contribute to phosphorylation of pRb, its peak abundance and its associated Cdk2 appears to function at the G 1 /S transition by phosphorylating some currently unknown substrate(s) other than pRb (37,45,57,58).
In conclusion, we found that IGFII overexpression in mammalian cells causes a shortened cycling time and increases in cyclin D1, p21, and p53 protein levels as well as increases in MAPK activity. However, the mechanisms underlying the induction of these proteins by IGFII overexpression and whether the increased levels of cyclin D1, p53, as well as MAPK activity participate directly or indirectly in the p21 induction are still under investigation. Following DNA damage, IGFII overexpression is associated with a diminished G 1 checkpoint, which may contribute to the high growth rate and genetic alterations during tumorigenesis.