Cell Cycle Re-Entry Following Chemically-Induced Cell Cycle Synchronization Leads to Elevated P53 and P21 Protein Levels

Home , Cell synchronization

SHORT REPORT Cell cycle re-entry following chemically-induced cell cycle synchronization leads to elevated p53 and p21 protein levels

Chuan Ji1, Lawrence J Marnett1,2 and Jennifer A Pietenpol1

Departments of 1Biochemistry and 2Chemistry, Center in Molecular Toxicology, and The Vanderbilt Cancer Center, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0146, USA

Mimosine (MIM) and aphidicolin (APH) are two agents G1 cell cycle checkpoint (Kastan and Kuerbitz, 1995; frequently used in tissue culture-based experiments to Agarwal et al., 1995) and also in G2/M checkpoint achieve cell synchronization at late G1 and S phases. (Stewart et al., 1995; Agarwal et al., 1995). In cells Following MIM or APH treatment of human cancer cell maintaining a normal p53 gene, p53 protein levels are lines, a reversible growth arrest in late G1 and S phases increased as a response to DNA damage (Kastan et al., of the cell cycle was correlated with moderate increases 1991; Kuerbitz et al., 1992) and this is correlated with in p53 and p21 protein levels. Both p53-dependent and growth arrest. The tumor suppressive function of p53 -independent increases in p21 were observed following is thought to be linked to its ability to function as a treatment with either agent. However, a striking increase sequence-speci®c transcriptional activator (Pietenpol et in p21 protein levels and a continuous elevation in both al., 1994) of a number of genes, including p21 (El- p53 and p21 protein levels were observed over 48 h after Deiry et al., 1993). p21 is a cyclin-dependent kinase cells re-entered the cell cycle following the chemically- inhibitor (Kato and Sherr, 1993; Harper et al., 1993) induced synchronization. In addition, the increase in p21 and when overexpressed can also induce growth arrest protein levels typically seen following treatment of cells (El-Deiry et al., 1993). Several studies have demon- with DNA damaging agents, was enhanced when cells strated that p21 mRNA and protein levels are elevated were treated with genotoxic agents following MIM or in a p53-dependent manner following treatment of cells APH synchronization. These ®ndings suggest that with genotoxic agents (El-Deiry et al., 1994; Dulic et caution should be exercised when interpreting results al., 1994), although additional studies have shown from experiments using cell synchronization agents, in regulation of p21 that is p53-independent (Steinman et particular, studies designed to investigate p53- and p21- al., 1994; Macleod et al., 1995; Zhang et al., 1995; regulatory pathways. Parker et al., 1995; Russo et al., 1995). In this study, we demonstrate that treatment of cells with either APH Keywords: p53; p21; cell cycle synchronization; or MIM resulted in elevation of p53 and p21 protein mimosine; aphidicolin levels. More signi®cantly, the elevated levels of p53 and p21 protein were maintained over a 48 h time period following release of cells from chemically-induced synchronization. Mimosine (MIM) and aphidicolin (APH) are com- Initially, we determined the kinetics of cell cycle monly used as potent and reversible late G1 and S release following treatment of the human colorectal phase blockers of the cell cycle. Several modes of cancer cell line RKO with MIM (100 mM)orAPH action for these two compounds have been proposed, (12 mM). Cell cycle progression was evaluated by ¯ow however, the molecular mechanisms are not completely cytometric analysis. Following a 24 h MIM or APH understood. Aphidicolin is a known inhibitor of DNA treatment, a decrease in the percentage of RKO cells in polymerases (Wang, 1991). Mimosine is a plant amino the G2/M phase and an increase in the percentage of acid thought to block DNA synthesis by decreasing cells in the G1 and S phases of the cell cycle were cellular dNTP concentrations through regulation of observed (Figure 1a). ribonucleotide reductase (Dai et al., 1994; Gilbert et The cell cycle blockade induced by MIM or APH al., 1995). Previously, many studies have used these was released by washing the cell monolayers and agents to synchronize cells to examine cell cycle incubating the cultures in fresh medium. In order to regulation of gene expression and protein activity. determine the kinetics of release, experiments were However, little attention has been given to the direct conducted in which cell cycle distribution was analysed eect of these agents or the eect of cell cycle re-entry at distinct times following removal of MIM (Figure 1b) following the induction of arrest by these agents on or APH (Figure 1c). Over time following release from checkpoint regulatory proteins. the MIM block, cell populations in G1 phase decreased The mammalian cell cycle maintains internal check in parallel with a S phase increase. By 12 h post- points at which cells arrest when previous events have release, over 80% of the cells were in S followed by a not been completed in response to external signals 50% increase in the G2/M phase population at 15 h (Hartwell and Weinert, 1989). Numerous studies have (Figure 1b). A similar pattern of cell cycle progression demonstrated a role for the tumor suppressor p53 in was observed following APH treatment. By 5 ± 10 h post APH release, 475% of the cell population was in S phase and by 12 ± 15 h, 460% of the cell population Correspondence: JA Pietenpol was in G2/M (Figure 1c). We performed dose-response Received 24 March 1997; revised 18 July 1997; accepted 22 July 1997 experiments with both agents and obtained similar p53 and p21 increase after cell cycle synchronization CJiet al 2750

Figure 1 MIM and APH induction of G1 and S phase blocks in RKO cells and kinetics of release. RKO cells were incubated in the presence or absence of MIM (100 mM) and APH (12 mM) for 24 h and cells harvested for ¯ow cytometric determination (a). Data presented are representative of at least three independent experiments. Kinetics of cell cycle release following MIM (b) and APH (c) blocks are shown. G1, &;S,~; G2/M, *. RKO cells were incubated with MIM or APH at 378C for 24 h, washed two times with growth medium and incubated in drug-free medium for the times indicated. RKO cells were grown in McCoy's 5A medium supplemented with 10% bovine serum, 50 unit/ml penicillin and 50 mg/ml streptomycin. Cells were seeded 24 h or 48 h before chemical treatment, and were 60 ± 80% con¯uent at the time of treatment. MIM (10 mM) and APH (15 mM) stock solution were made with PBS and methanol, respectively, and diluted with medium prior to application. ADR or MDA was directly dissolved in a fresh medium and added to cultures immediately. Flow cytometric determination of cell-cycle was performed by staining the DNA with propidium iodide. Brie¯y, following indicated treatment, cells were trypsinized and washed with ice-cold PBS and ®xed in 70% ethanol (7208C for at least 1 h). Cells were resuspended in PBS, incubated with RNase (1 mg/ml) at room temperature for 15 min, and stained with propidium iodide (50 mg/ml) at room temperature for 30 min. Cell-cycle determination was performed using a Becton-Dickinson ¯uorescence-activated cell Sorting (FACScan) ¯ow cytometer p53 and p21 increase after cell cycle synchronization CJiet al 2751 a

b Control MDA ADR MIM1 MIM2 APH1 APH2

P53 Treated 24 h

P21

P53 Post-Treatment 10 h P21

P53 Post-Treatment 24 h P21

P53 Post-Treatment 34 h P21

P53

Post-Treatment 48 h P21

Figure 2 Cell cycle kinetics following release from a MIM or APH block. RKO cells were treated with MIM (MIM1, 100 mM; MIM2, 250 mM), APH (APH1, 12 mM; APH2, 24 mM), MDA (1 mM), or ADR (0.4 mM) for 24 h. Cells were either harvested or washed with medium twice and re-incubated in fresh medium for 10 h, 24 h, 34 h and 48 h. Control indicates exponentially growing cells. From each sample, equal number of cells were processed for: (a) Flow cytometric analysis of cell cycle distribution as described in Figure 1 legend, and (b) Western blot analysis of p21 and p53 protein levels as previously described (Pietenpol et al., 1996) using anti-p21 EA10 and anti-p53 DO-1 (Oncogene Research Products, Cambridge, MA) p53 and p21 increase after cell cycle synchronization CJiet al 2752 kinetics of synchronization and release with dosages of ¯ow cytometric analysis and Western analysis. Pro®les 100 mM to 400 mM MIM and 8 mM to 24 mM APH of cell cycle distribution and corresponding Western (data not shown). blots from the samples are shown in Figure 3a and b, To examine the eect of synchronization agents on respectively. Cells synchronized by MIM and APH and p53 and p21 protein levels, RKO cells were treated released were predominantly found in the G2/M phase, with MIM and APH for 24 h followed by release for 22 h post-release. Treatment of cells with either MDA increasing periods of time over a 48 h time course. For or ADR during release altered the cell cycle pro®le a positive control, in parallel experiments, RKO were treated with the following genotoxic agents: adriamycin (ADR), or malondialdehyde (MDA) a product of lipid biosynthesis and eicosanoid biosynthesis (Janero, 1990). Treatment of cells with either ADR or MDA a has been demonstrated to induce DNA damage and increase both p53 and p21 protein levels (El-Deiry et al., 1994; Chuan et al., 1997). From each treatment condition, cell samples were divided for ¯ow cytometric analysis (Figure 2a) and Western analysis (Figure 2b). Administration of ADR (0.4 mM) and MDA (1 mM) resulted in predominantly a G2/M phase arrest and a S and G2/M phase arrest, respectively. The cell cycle arrest induced by these agents was not reversible even after 48 h incubation in drug-free medium (Figure 2A). Administration of MIM and APH resulted in cell cycle arrest at G1 and S phases and the arrest was reversed after agent withdrawal (Figure 2a). Ten hours following MIM and APH withdrawal, cell populations moved from G1/S to S and G2/M phases. Whereas the cells appear to recover from MIM treatment, cells appeared to accumulate and arrest at G2/M following APH treatment. Corresponding to these cell cycle progression changes, the level of p53 protein was signi®cantly increased. Twenty-four hours after MIM and APH treatment, the level of p53 protein was increased by all drugs in a dose-dependent manner (data not shown). After removal of the synchronization agents, only slight increases in p53 protein levels were observed. The increase in p53 protein levels during synchroniza- b tion was likely due to post-translational stabilization of the protein as previously shown for other agents (Maltzman and Czyzyk, 1984).

p21 protein levels were not altered signi®cantly Control MIM APH MDA MIM-MDA APH-MDA ADR MIM-ADR APH-ADR following a 24 h MIM treatment (100 mM or 250 mM) or lower dose APH treatment (12 mM) but increased p53 moderately following 24 mM APH exposure. In contrast, the p21 protein levels were signi®cantly increased as early as 10 h after release from MIM- and APH-induced cell cycle arrest (Figure 2b). Further incubation for 24 h and 36 h post-treatment, resulted in even greater accumulation of p21 protein that began p21 to decrease after 48 h of growth in drug-free medium. The increase in p21 protein levels correlated with an Figure 3 H1299 cell cycle distribution and p53-independent observed G2/M accumulation of cells, in particular expression of p21 protein following release from MIM or APH following release from APH blockade. In addition, the block. H-1299 cells were synchronized with MIM (250 mM)or increase in p21 protein is paralleled by an increase in APH (12 mM) for 24 h. Cells were washed twice to remove synchronization agents from the medium. Two hours after release p21 RNA levels (data not shown). from MIM or APH blockade, cells were incubated in the presence To determine if the observed increase in p21 protein or absence of MDA (1 mM) or ADR (1.2 mM) for 22 h. Cells from was due to a p53-dependent mechanism, experiments the same sample were divided into two portions: (a) Flow were conducted with a human large cell lung cytometric analysis of cell cycle distribution as described in Figure 1 legend; and (b) Western blot analysis of p21 protein levels as carcinoma cell line, H1299, that has a homozygous described in Figure 2 legend. Control represents exponentially deletion of the p53 gene. Following induction of a cell growing cells; MIM and APH represents cells that were treated cycle block by a 24 h treatment of MIM (250 mM)or with these agents for 24 h and released for 24 h; ADR and MDA APH (12 mM), cells were washed and incubated in drug represent cells that were treated with these agents for 22 h; MIM- MDA, APH-MDA, MIM-ADR, and APH-ADR represent cells free medium for 2 h, and then cultured in the presence treated with the synchronization agents for 24 h, washed and or absence of MDA (1 mM) or ADR (1.2 mM) for an incubated for 2 h, followed by treatment with either ADR or additional 22 h. Cells were harvested and analyzed by MDA for 22 h p53 and p21 increase after cell cycle synchronization CJiet al 2753 signi®cantly. Following synchronization induced by alteration of p53 and p21 protein levels for 2 days MIM, MDA blocked cells predominantly in S phase, following release from MIM- or APH-induced cell whereas ADR blocked in G2/M phase. However, cells synchronization indicates that there may be potential were blocked predominantly in S phase by ADR misinterpretation in conclusions reached by studies that following synchronization by either agent. Interest- employ MIM and APH to block cell cycle progression. ingly, the cell samples that underwent a predominant Thus, while these agents remain useful as tools for G2 accumulation had parallel increases in p21 protein probing select cell cycle regulatory events or aspects of levels (Figure 3b, APH-MDA and MIM-ADR). the DNA replication machinery, caution should be In summary, during MIM- and APH-induced exercised in the interpretation of the results. synchronization, p53 protein levels increased. The direct induction of p53 and p21 by MIM is consistent with previous studies (Lavin et al., 1994; Linke et al., 1996; Alpan and Pardee, 1996). However, we Acknowledgements determined that following release of cells from the This work was supported by National Institutes of Health Grants CA47479 (to LJM), CA68450, CA68485, and MIM or APH block, a striking increase in p21 protein ES00267 (core services), CA70856 (to JAP), and The levels was evident. The increase in p21 protein levels Council for Tobacco Research, USA Grant 4349 (to was observed in parallel with an accumulation of cells JAP). We wish to thank Jim Price for the ¯ow cytometric in G2/M, and occurred by both p53-dependent and analysis and Laura Neidernhofer for the MDA used in -independent pathways. In addition, the signi®cant these studies.

References

Agarwal ML, Agarwal A, Taylor WR, and Stark GR. (1995). Kuerbitz SJ, Plunkett BS, Walsh WV and Kastan MB. Proc. Natl. Acad. Sci. USA, 92, 8493 ± 8497. (1992). Proc. Natl. Acad. Sci. USA, 89, 7491 ± 7495. Alpan RS and Pardee AB. (1996). Cell Growth Dier., 7, Lavin MF, Khanna KK, Beamish H, Teale B, Hobson K and 893 ± 901. Watters D. (1994). Int. J. Radiat. Biol., 66, S151 ± S156. Chuan J, Rouzer CA, Marnett LJ and Pietenpol JA. (1997). LinkeSP,ClarkinKC,DiLeonardoA,TsouAandWahl Proc. Am. Assoc. Cancer Res., 38, 1324. GM. (1996). Genes Dev., 10, 934 ± 947. Dai Y, Gold B, Vishwanatha JK and Rhode S. (1994). Virol., Macleod KF, Sherry N, Hannon G, Beach D, Tokino T, 205, 210 ± 216. Kinzler K, Vogelstein B and Jacks T. (1995). Genes Dev., Dulic V, Kaufmann WK, Wilson SJ, Tlsty TD, Lees E, 9, 935 ± 944. Harper JW, Elledge SJ and Reed SI. (1994). Cell, 76, Maltzman W and Czyzyk L. (1984). Mol. Cell. Biol., 4, 1013 ± 1023. 1689 ± 1694. El-Deiry WS, Harper JW, O'Connor PM, Velculescu VE, Parker SB, Eichele G, Zhang P, Rawls A, Sands AT, Bradley Canman CE, Jackman J, Pietenpol JA, Burrell M, Hill A, Olson EN, Harper JW and Elledge SJ. (1995). Science, DE, Wang Y, Wiman KG, Mercer WE, Kastan MB, Kohn 267, 1024 ± 1027. KW, Elledge SJ, Kinzler KW and Vogelstein B. (1994). Pietenpol JA, Tokino T, El-Deiry WS, Kinzler KW and Cancer Res., 54, 1169 ± 1174. Vogelstein B. (1994). Proc. Natl. Acad. Sci. USA, 91, El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons 1998 ± 2002. R, Trent JM, Lin D, Mercer WE, Kinzler KW and Pietenpol JA, Lengauer L, Jordan J, Kinzler KW and Vogelstein B. (1993). Cell, 75, 817 ± 825. Vogelstein B. (1996). Proc. Natl. Acad. Sci. USA, 93, Gilbert DM, Neilson A, Miyazawa H, DePamphelis ML and 8390 ± 8394. Burhans WC. (1995). J. Biol. Chem., 270, 9597 ± 9606. Russo T, Zambrano N, Esposito F, Ammendola R, Cimino Harper JW, Adami GR, Wei N, Keyomarsi K and Elledge F, Fiscella M, Jackman J, O'Connor PM, Anderson CW SJ. (1993). Cell, 75, 805 ± 816. and Appella E. (1995). J. Biol. Chem., 270, 29386 ± 29391. Hartwell LH and Weinert TA. (1989). Science, 246, 629 ± Steinman RA, Homan B, Iro A, Guillouf C, Liebermann 634. DA and El-Houseini ME. (1994). Oncogene, 9, 3389 ± Janero DR. (1990). Free Radical Biol. Med., 9, 515 ± 540. 3396. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B and Stewart N, Hicks GG, Paraskevas F and Mowat M. (1995). Craig RW. (1991). Cancer Res., 51, 6304 ± 6311. Oncogene, 10, 109 ± 115. Kastan MB and Kuerbitz SJ. (1995). Environ. Health Wang TF. (1991). Annu. Rev. Biochem., 60, 513 ± 552. Perspect., 5, 55 ± 58. Zhang W, Grasso L, McClain CD, Gambel AM, Cha Y, Kato J and Sherr CJ. (1993). Proc. Natl. Acad. Sci. USA, 90, Travali S, Deisseroth AB and Mercer WE. (1995). Cancer 11513 ± 11517. Res., 55, 668 ± 674.