Cyclin A1, the alternative A-type cyclin, contributes to G1/S cell cycle ...

Oncogene (2005) 24, 2739–2744

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

Cyclin A1, the alternative A-type cyclin, contributes to G1/S cell cycle progression in somatic cells Ping Ji1, Shuchi Agrawal1, Sven Diederichs1, Nicole Baümer1, Annette Becker1, Thomas Cauvet1, Sascha Kowski2, Carmela Beger2, Karl Welte2, Wolfgang E Berdel1, Hubert Serve1 and Carsten Mu¨ller-Tidow*,1 1

Department of Medicine, Hematology/Oncology, University of Mu¨nster, Domagkstr. 3, 48149 Mu¨nster, Germany; 2Department of Pediatric Hematology and Oncology, MHH Hannover, 30625 Hannover, Germany

Cyclin A1 is an alternative A-type cyclin that is essential for spermatogenesis, but it is also expressed in hematopoietic progenitor cells and in acute myeloid leukemia. Its functions during cell cycle progression of somatic cells are incompletely understood. Here, we have analysed the cell cycle functions of cyclin A1 in transformed and nontransformed cells. Murine embryonic fibroblasts derived from cyclin A1-deficient mice were significantly impaired in their proliferative capacity. In accordance, cyclin A1/ cells accumulated in G1 and G2/M phase while the percentage of S phase cells decreased. Also, lectin stimulated splenic lymphocytes from cyclin A1/ mice proliferated slower than their wild-type counterparts. Forced cyclin A1 overexpression in NIH3T3 cells and in U937 leukemic cells either by transient transfection or by retroviral infection enhanced S phase entry. Consequently, siRNA mediated silencing of cyclin A1 in highly cyclin A1 expressing ML1 leukemic cells significantly slowed S phase entry, decreased proliferation and inhibited colony formation. Taken together, these analyses demonstrate that cyclin A1 contributes to G1 to S cell cycle progression in somatic cells. Cyclin A1 overexpression enhances S phase entry consistent with an oncogenic function. Finally, cyclin A1 might be a therapeutic target since its silencing inhibited leukemia cell growth. Oncogene (2005) 24, 2739–2744. doi:10.1038/sj.onc.1208356 Published online 27 December 2004 Keywords: cyclin A1; leukemia; cell cycle; siRNA

Main text The cell cycle is a major target for alterations in the pathogenesis of cancer. Almost all cancers contain mutations in tumor suppressors or oncogenes that regulate G1/S transition (Knudson, 2002). Cyclins play an important role in the regulation of the cell cycle as *Correspondence: C Mu¨ller-Tidow; E-mail: [email protected] Received 20 July 2004; revised 28 October 2004; accepted 29 October 2004; published online 27 December 2004

regulatory subunits of CDKs. For example, the sequential phosphorylation of Rb by CDK4/6 and CDK2 releases E2F from E2F/Rb complexes and enables progression from G1 to S phase (Weinberg, 1995; Dyson, 1998; Helin, 1998; Lania et al., 1999). Cyclin A2, also known as cyclin A, is the major A-type cyclin in mammals. Cyclin A1 is an alternative CDK2 associated A-type cyclin. It is highly expressed in testis and plays an important role in male meiosis (Sweeney et al., 1996). In addition, cyclin A1 is also expressed at high levels in acute myeloid leukemia (Yang et al., 1997). Here, it is a direct target of the PML-RARa oncoprotein in acute promyelocytic leukemia (Mu¨ller et al., 2000). Also, the Six1 oncogene induces cyclin A1 expression (Coletta et al., 2004). At low levels, cyclin A1 can be detected in most tissues. In vitro, cyclin A1 and cyclin A2 share many target genes. We have previously shown that cyclin A1/CDK2 phosphorylate and interact with Rb, E2F-1 and B-Myb similar to cyclin A2 (Yang et al., 1999a; Mu¨ller-Tidow et al., 2001). In addition, we recently identified several novel interaction partners of the cyclin A1/CDK2 complex (Diederichs et al., 2004). In a mouse model, a cyclin A1 transgene directed to the bone marrow contributed to a myeloproliferative disorder and subsequent leukemia development (Liao et al., 2001). Male cyclin A1 knockout mice are infertile but no other phenotype has been reported so far (Liu et al., 1998). More recent data show that cyclin A1 acts as a dose-dependent regulator for the production of haploid cells (Van der Meer et al., 2004). These data suggested that cyclin A1 functions are mainly relevant in leukemogenesis and spermatogenesis. However, several recent mouse models with deletion of either one of the important cell cycle regulators CDK2 or cyclin E developed normally (Berthet et al., 2003; Geng et al., 2003; Ortega et al., 2003; Parisi et al., 2003). Particularly interesting, ablation of CDK2 in the mouse impaired spermatogenesis while the mice were viable. These findings prompted us to evaluate the cell cycle effects of cyclin A1 in further detail. We used murine embryonic fibroblasts from cyclin A1/ mice to analyse the effects of cyclin A1 deficiency in isolated cells, which proliferate exponentially. Cyclin A1/ MEFs showed significantly decreased proliferative activity in [3H]thymidine incorporation experiments

Cyclin A1 functions in cell cycle progression P Ji et al

2740

(P ¼ 0.016, t-test) (Figure 1a). These analyses were reproduced with several batches of MEFs obtained from crosses of heterozygous females with heterozygous males. In addition, growth curve analyses revealed that MEFs from cyclin A1/ mice were significantly impaired proliferative responses (data not shown). Cell cycle analyses were performed to elucidate the affected phases of the cell cycle. Notably, cyclin A1/ cells cycled with a decreased percentage of cells in S phase (Figure 1b). The observed differences were consistent but relatively small, most likely due to the lack of synchronization (P ¼ 0.065). Therefore, MEF cells were synchronized by serum starvation before refeeding and analysis by BrdU incorporation assay. These analyses clearly showed impaired S phase entry in cyclin A1 deficient cells (P ¼ 0.027, paired t-test) (Figure 1c). To verify that cyclin A1/ cells were generally impaired in proliferative responses, spleen lymphocytes were isolated from cyclin A1/ and wild-type mice. Lymphocytes were stimulated with concanavalin A (Con A) or phytohemagglutinin (PHA) for 4 days before [3H]thymidine incorporation assays were performed. Again,

b

30 25

80 70 60 50 40 30 20 10 0 Cyclin A1 +/+ -/G1

+/+

-/-

20 15 10 5

0 Cyclin A1 +/+ -/- +/+ -/- +/+ -/12h 16h 20h Time after stimulation (h)

+/+ -/S

+/+ -/G2/M

d 6000 5000 4000 3000 2000 1000 0 Cyclin A1

[3H] incorporation (cpm)

c Relative S phase entry (%)

4000 3500 3000 2500 2000 1500 1000 500 0

Percentage of cells

a [3H] incorporation (cpm)

lymphocytes from cyclin A1/ mice exhibited a lower proliferative response compared to matched wild-type mice (Figure 1d). Without reaching statistical significance (P ¼ 0.113 and 0.058). Take together, these experiments suggested that low levels of cyclin A1 play a role in G1/S transition and that cyclin A1-deficient cells are impaired in proliferation under optimal growth conditions. Next, we asked whether cyclin A1 overexpression would influence S phase entry in somatic cells. For this purpose, we used transient transfections to determine the effects of cyclin A1 overexpression in nontransformed cell lines. NIH3T3 fibroblasts and 32D myeloid progenitor cells were transfected with pcDNA3-cyclin A1 or pcDNA3 control vector. Transfection efficiency was above 40% as indicated by co-transfected CD4 expression plasmid pMACS-4.1-CD4 (Figure 2a). Cells were harvested after 24 h and analysed for cell cycle distribution using propidium iodide staining. We observed that ectopic expression of cyclin A1 in NIH3T3 cells led to an increased proportion of cells in S phase compared to control vector transfected cells.

+/+ -/Con A

+/+ -/PHA

Figure 1 Function of cyclin A1 in somatic cells. (a) Proliferation analysis in murine embryonic fibroblasts (MEF). MEFs were isolated from day 13 pregnant cyclin A1 þ / females that were bred with cyclin A1 þ / males (van der Meer et al., 2004). MEFs from two cyclin A1 wild type (MEF A1 þ / þ ) and two knockout cyclin A1 (MEF A1/) embryos were arrested in G0/G1 phase by serum starvation. In total, 10 000 cells were seeded in each well, and cells were refed with fresh medium containing 10% FCS and 1 mCi [3H]thymidine. Cells were cultured for another 6 h and incorporated radioactivity was determined in a b-scintillation counter. The data represent the average of six wells and plus standard error for two MEF A1 þ / þ and two MEF A1/ cell lines in passage 6 (P ¼ 0.016, ttest). (b) Inhibition of cell cycle progression in cyclin A1/ MEF cells. In passage 6, MEF cells were arrested by contact inhibition, 1 106 cells were freshly seeded in 60-mm dishes and cultured for 24 h. Cells were stained with propidium iodide and cell cycle analysis was performed by flow cytometry. Mean percentages of cell cycle distribution plus standard error of three independently performed experiments are indicated. (c) Delayed entry into S phase in cyclin A1/ MEF cells. In the sixth passage, two MEF A1 þ / þ and two MEF A1/ cell lines were serum starved for 60 h and subsequently stimulated with medium containing 10% FCS (time point 0 h). The cells were pulse labelled with 20 mM BrdU for 1 h at the times indicated after addition of serum. Cells were stained with anti-BrdU antibodies and propidium iodide and analysed by flow cytometry. The percentage of cells in S phase at each time point was calculated. The data represent the mean of two MEF cell lines plus standard error in two independent experiments. (d) Impaired proliferation of cyclin A1/ spleen lymphocytes. Splenic lymphocytes were isolated from spleens of 8-week-old wild type or cyclin A1-deficient mice. Cells were stimulated with ConA or PHA for 4 days to induce proliferation. After addition of 1 mCi [3H]thymidine for another 6 h, incorporated radioactivity was determined in a b-scintillation counter. The data are shown as mean plus standard error of two mice in each group analysed in triplicates Oncogene


2741

Figure 2 Acceleration of G1 to S phase transition by ectopic expression of cyclin A1. (a) Transfection of NIH3T3 cells. NIH3T3 cells were co-transfected with pcDNA3 or pcDNA-A1 together with pMACS-4.1-CD4 using Superfact. At 24 h post-transfection, cells were stained with fluorescein isothiocyanate (FITC)-labelled anti-CD4 antibody and propidium iodide. More than 40% of the cells were routinely transfected. (b) Cell cycle distribution patterns in transfected NIH3T3 cells. Cells were synchronized by starving for 3 days in 0.5% FCS plus nocodazole treatment for 24 h, and transfected with pcDNA3 or pcDNA3-A1, analysed for cell cycle distribution at 24 h using propidium iodide staining in three independent experiments. (c) Kinetic analyses of S phase entry in transfected NIH3T3 cells. Cells were synchronized in G2/M phase, transfected with pcDNA3 or pcDNA3-A1 by Amaxa nucleoporation (AMAXA), and pulse-labelled with 10 mM BrdU for 3 h at 9 h as a starting points, and ended at 20 h before G2/M phase. S phase profiles of a representative experiment are shown

The S phase increase was associated with a decrease in the percentage of G1 cells indicating accelerated G1/S transition (Figure 2b). Similar effects were observed in 32D cells in which cyclin A1 also accelerated G1/S phase transition (data not shown). To confirm this finding, analyses of S phase entry were performed by BrdU incorporation assay at two time points (Figure 2c).

These data showed that cyclin A1 overexpression accelerated cell cycle progression in transient transfections. To further determine the functions of cyclin A1 in myeloid cells, we retrovirally established stable cell lines in U937 leukemia cells. Cells transduced with pLXINcyclin A1 were named U937-cyclin A1, whereas empty vector transduced cells were named U937-control. Cells were selected in G418 containing media and were analysed as bulk culture to exclude clone-specific effects. Cyclin A1 overexpression was confirmed on the mRNA (Figure 3a) as well as on the protein level (Figure 3b). Other cell cycle factors were not altered on the mRNA level in the cyclin A1 overexpressing U937 cell line (Figure 3a). Also, no changes on the protein level of either CDK2 or cyclin A2 were observed (Figure 3b). In line with the experimental results shown above, U937cyclin A1 cells consistently showed a higher percentage of S phase cells compared to U937-control cells as shown by conventional cell cyle assays (data not shown) and flow cytometric BrdU incorporation assays (Figure 3c and d, P ¼ 0.001, paired t-test). Primary acute myeloid leukemia (AML) blasts and several leukemia cell lines express high levels of cyclin A1 (Yang et al., 1997, 1999b). This prompted us to investigate whether a decrease of cyclin A1 expression levels in leukemic cell lines would inhibit leukemia cell growth and proliferation. The ML1 cell line that expresses high levels of cyclin A1 (Yang et al., 1997) was transfected with pRNAT-H1.1-RNAi cyclin A1 (Si-A1) or control vector pRNAT-H1.1. Stable cell lines were established by G418 selection and sorted for EGFP positivity. ML1 Si-A1 cells expressed significantly lower cyclin A1 mRNA (Figure 4a) and protein, whereas no change was found in the expression level of cyclin A2 (Figure 4b). To analyse cell cycle distribution in cyclin A1 silenced cell lines, ML1 Si-A1 cells were synchronized by starvation with 1% FCS media for 72 h. At this time, cells were released in 10% FCS supplemented fresh media. In a time course analysis, cells were pulse-labeled with BrdU and stained with a FITC-labeled anti-BrdU antibody and propidium iodide. Leukemic cells with decreased cyclin A1 levels consistently showed an impaired S phase entry as shown by BrdU incorporation assays (Figure 4c and d, P ¼ 0.032, paired t-test) and conventional cell cycle analyses (data not shown). To analyse the proliferative capacity of ML1 cells that decreased cyclin A1 expression, colony assays were performed. These assays showed that reduced cyclin A1 expression significantly decreased colony formation in methylcellulose (Figure 4e, Po0.001). Our findings provide evidence that under conditions, which induce rapid cell cycling, cyclin A1-deficient cells are significantly impaired in their proliferative response. These findings in a genetic mouse model were backed up by siRNA mediated cyclin A1 silencing in a human leukemia cell line. Further experiments showed that cyclin A1 overexpression significantly accelerated G1 to S progression. These data indicate a direct role of oncogenic cyclin A1 in accelerating G1- to S-phase entry. Since these effects were demonstrated in myeloid Oncogene


2742

Figure 3 Cyclin A1 overexpression in U937 leukemia cells. (a) Overexpression of cyclin A1 mRNA in U937-cyclin A1 cells. Human cyclin A1 cDNA was cloned into the pLXIN vector. Retrovirus was produced by transfection of 293T cells using the calciumphosphate method (Calphos: BD Clontech) with gag-pol, VSV-G and pLXIN (containing cyclin A1 or without any insert as a control). For transduction, a total of 4 105 U937 cells were plated into Retronectin-coated six-well plates (Takara; Gennevilliers, France). Cells were transduced by replacing cell culture media by 2.5 ml of retroviral supernatant 24 h after cell plating. After two rounds of transduction, cells were selected in medium containing 500 mg/ml G418. Expression levels of cyclin A1 mRNA and other cell cycle regulators were analysed in U937-control and U937-cyclin A1 cells by quantitative real-time RT–PCR as described (Mu¨ller-Tidow et al., 2004). U937-cyclin A1 significantly overexpressed cyclin A1 but none of the other cycle regulators. (b) Overexpression of cyclin A1 protein. Total protein extracts (28 mg) from U937-control or U937-cyclin A1 were electrophoresed and transferred to nitrocellulose as described (Mu¨ller-Tidow et al., 2004). Blots were probed with the indicated antibodies. (c and d) Increased S phase entry in U937cyclin A1 cells. U937-control and U937-cyclin A1 were serum starved for 72 h and released in 10% FCS media at the indicated time points, and pulse-labeled for 1 h with 10 mM BrdU before staining. Cell cycle distributions of a representative experiment are shown. (d) Relative G1- to S phase transition was calculated by the percentages of S phase cells at each time point minus the background (S phase after starvation)

and nonmyeloid cells, cyclin A1 induced G1/S transitions appear to be cell-type independent. These data are consistent with those published for other cyclins involved in G1/S transition. For example, cyclin E and cyclin D overexpression enhanced S phase entry in transfected cells (Resnitzky et al., 1994). We have recently demonstrated that cyclin A1 similar to other CDK2 associating cyclins can phosphorylate Rb and the Rb family members p107 and p130 (Yang et al., 1999a). In addition, cyclin A1-CDK2 interacts with and activates the transcription factor B-Myb that is also involved in G1/S transition (Mu¨ller-Tidow et al., 2001). These biochemical findings complement our current Oncogene

findings that cyclin A1 overexpression accelerates S phase entry in a wide variety of cells. Enhanced cell cycling mediated via cyclin A1 induction might be one of the oncogenic mechanisms in acute myeloid leukemia. Finally, cyclin A1 might be a suitable target for therapeutic intervention in acute myeloid leukemia. Acknowledgements This work is supported by grants from the Deutsche Forschungsgemeinschaft (Mu 1328/2-3, Se 600/3), the Deutsche Krebshilfe (10-1539-Mu¨3), the IMF and the IZKF at the University of Mu¨nster. C. Mu¨ller-Tidow is supported by the DFG-Heisenberg program (Mu1328/3-1).


2743

Figure 4 Inhibition of S phase progression by RNAi-mediated silencing of cyclin A1. (a) Decrease of cyclin A1 mRNA expression by siRNA silencing. Oligonucleotides containing 21 nt specific sense (italic) and antisense (bold) sequence against human cyclin A1 were synthesized, (Primer 1: GATCCCGTGTGCCGGTGTCTACTTCATTTGATATCCGATGAAGTAGACACCGGCACACTTTTTTCCAAA and Primer 2: AGCTTTTGGAAAAAAGT GTGCCGGTGTCTACTTCATCGGATATCAAATGAAGTAGACACCGGCACACGG. Double-stranded oligonucleotides were inserted into the pRNAT-H1.1 plasmid (BamHI and HindIII sides) (GenScript) under H1 promoter control and the resulting plasmid was named pRNAT-Si-A1. ML1 leukemia cells were transfected with pRNATSi-A1 or control vector and selected with G418. After FACS-sorting, EGFP þ cells were analysed as bulk culture. RNA was isolated and real-time RT–PCR was performed to analyse the cyclinA1 expression levels in the different cell lines. The cyclin A1-RNAi transfected cell line (ML1 Si-A1) expressed consistently less cyclin A1 compared to control vector transfected cell line (ML1 Si-C). The data are shown as mean plus standard error in triplicates. (b) Decrease of cyclin A1 protein by siRNA mediated silencing. Total protein extracts (60 mg) from ML1 Si-C and ML1 Si-A1 cells were electrophoresed and transferred on to nitrocellulose membrane. The blot was probed with polyclonal anti-cyclin A1 antibody, monoclonal anti-cyclin A2 antibody (Sigma clone CY-A1) and anti-actin antibody staining (Sigma) served as loading control. (c) Inhibition of S phase entry in ML1 Si-A1 cells. ML1 Si-C (control) and ML1 Si-A1 (siRNA against cyclin A1) were serum starved and released in media with 10% FCS. At the indicated times, cells were pulselabeled for 1 h with 10 mM BrdU before labeling with propidium iodide and flow cytometric analysis. The X-axis represents DNA content, and the Y-axis represents DNA synthesis (BrdU). Representative DNA profiles are shown. (d) G1- to S phase transition in ML-1 cells. Kinetics of S phase entry was performed at the indicated time points by BrdU incorporation. The percentage of G1/S transition was calculated by the percentages of S phase cells at each time point minus the background (S phase after starvation). (e) Decrease of colony formation in ML1 Si-A1 cell line. Colony assays in methylcellulose were performed with 1000 ML1 Si-C or ML1 Si-A1 cells in three dishes. The bars show mean plus standard error in triplicates. The number of colonies formed by the ML1 Si-C cell line was set as 100%

Oncogene


2744 References Berthet C, Aleem E, Coppola V, Tessarollo L and Kaldis P. (2003). Curr. Biol., 13, 1775–1785. Coletta RD, Christensen K, Reichenberger KJ, Lamb J, Micomonaco D, Huang L, Wolf DM, Mu¨ller-Tidow C, Golub TR, Kawakami K and Ford HL. (2004). Proc. Natl. Acad. Sci. USA, 101, 6478–6483. Diederichs S, Baümer N, Ji P, Metzelder SK, Idos GE, Cauvet T, Wang W, Gromoll J, Schrader MG, Koeffler HP, Berdel WE, Serve H and Mu¨ller-Tidow C. (2004). J. Biol. Chem., 279, 33727–33741. Dyson N. (1998). Genes Dev., 12, 2245–2262. Geng Y, Yu Q, Sicinska E, Das M, Schneider JE, Bhattacharya S, Rideout WM, Bronson RT, Gardner H and Sicinski P. (2003). Cell, 114, 431–443. Helin K. (1998). Curr. Opin. Genet. Dev., 8, 28–35. Knudson AG. (2002). Am. J. Med. Genet., 111, 96–102. Lania L, Majello B and Napolitano G. (1999). J. Cell. Physiol., 179, 134–141. Liao C, Wang XY, Wei HQ, Li SQ, Merghoub T, Pandolfi PP and Wolgemuth DJ. (2001). Proc. Natl. Acad. Sci. USA, 98, 6853–6858. Liu D, Matzuk MM, Sung WK, Guo Q, Wang P and Wolgemuth DJ. (1998). Nat. Genet., 20, 377–380. Mu¨ller C, Yang R, Park DJ, Serve H, Berdel WE and Koeffler HP. (2000). Blood, 96, 3894–3899. Mu¨ller-Tidow C, Steffen B, Cauvet T, Tickenbrock L, Ji P, Diederichs S, Sargin B, Ko¨hler G, Stelljes M, Puccetti E,

Oncogene

Ruthardt M, deVos S, Hiebert SW, Koeffler HP, Berdel WE and Serve H. (2004). Mol. Cell. Biol., 24, 2890–2904. Mu¨ller-Tidow C, Wang W, Idos GE, Diederichs S, Yang R, Readhead C, Berdel WE, Serve H, Saville M, Watson R and Koeffler HP. (2001). Blood, 97, 2091–2097. Ortega S, Prieto I, Odajima J, Martin A, Dubus P, Sotillo R, Barbero JL, Malumbres M and Barbacid M. (2003). Nat. Genet., 35, 25–31. Parisi T, Beck AR, Rougier N, McNeil T, Lucian L, Werb Z and Amati B. (2003). EMBO J., 22, 4794–4803. Resnitzky D, Gossen M, Bujard H and Reed SI. (1994). Mol. Cell. Biol., 14, 1669–1679. Sweeney C, Murphy M, Kubelka M, Ravnik SE, Hawkins CF, Wolgemuth DJ and Carrington M. (1996). Development, 122, 53–64. Van der Meer T, Chan WY, Palazon LS, Nieduszynski C, Murphy M, Sobczak-Thepot J, Carrington M and Colledge WH. (2004). Reproduction, 127, 503–511. Weinberg RA. (1995). Cell, 81, 323–330. Yang R, Morosetti R and Koeffler HP. (1997). Cancer Res., 57, 913–920. Yang R, Mu¨ller C, Huynh V, Fung YK, Yee AS and Koeffler HP. (1999a). Mol. Cell. Biol., 19, 2400–2407. Yang R, Nakamaki T, Lu¨bbert M, Said J, Sakashita A, Freyaldenhoven BS, Spira S, Huynh V, Mu¨ller C and Koeffler HP. (1999b). Blood, 93, 2067–2074.