Immortal, non-tumourigenic mouse mammary outgrowths express high levels of cyclin B1 and activation of cyclin B1/cdc2 kinase

Abstract. Neoplastic transformation of mouse mammary epithelial cells is the result of several identifiable phenotypic changes which presumably require sequential genetic alterations. In our model system, mammary cells progress from a mortal state (virgin duct) to several morphologically distinct intermediate states. The intermediate states are distinct cell populations that are phenotypically identified as immortal, non-tumourigenic (i.e. EL11), weakly tumourigenic ductal/alveolar hyperplasia (i.e. EL12) and moderately tumourigenic alveolar hyperplasiaa (i.e. TM12) to invasive tumours (i.e. EL12T/TM12T). We have studied the changes in total cyclin A and B1 levels, cyclin A and B1 complexed to cdc2, cyclin B1cdc2 kinase activity and cyclin D proteins in EL11 and EL12 immortalized outgrowth lines. Results revealed increased levels in total cyclin B1(> 5-fold), cyclin B1/cdc2 (3–4-fold) and cyclin B1/cdc2 kinase activity (2–3.5-fold) in EL11 and EL12 phenotypes when compared to control mammary gland (virgin). No changes in the levels of total cyclin A or cycln A associated to cdc2 were observed. Cyclin D1, D2 and D3 protein levels were low in the EL11 immortal ductal outgrowth. Exposure to hormones via a pituitary isograft stimulated the synthesis of cyclin D1 and D2 but not D3 associated to cdk4 as well as total cdk4 proteins. Bromodeoxyuridine (BrdUrd) labelling indices showed marked increases in immortal ductal outgrowths (EL11 and EL12) when compared to virgin, suggesting that epithelial cells are cycling in these cell populations. Even in the presence of hormone stimulation, EL11 outgrowths were not tumourigenic, suggesting that other events are necessary to drive the cells to a tumourigenic phenotype. The results suggest that increased levels of cyclin B1 and cyclin B1-cdc2 kinase activities are early events and may be an important marker for the immortalized phenotype.     

Increased D-type cyclin expression together with decreased cdc2 activity confers megakaryocytic differentiation of a human thrombopoietin-dependent hematopoietic cell line

At the late phase of megakaryocytopoiesis, megakaryocytes undergo endomitosis, which is characterized by DNA replication without cell division. Although a number of cell cycle regulatory molecules have been identified, the precise roles of these molecules in megakaryocytic endomitosis are largely unknown. In a human interleukin-3-dependent cell line transfected with the thrombopoietin (TPO) receptor c-mpl (F-36P-mpl), either treatment with TPO or the overexpression of activated ras (Ha-Ras(G12V)) induced megakaryocytic maturation with polyploid formation. We found that TPO stimulation or Ha-Ras(G12V) expression led to up-regulation of cyclin D1, cyclin D2, and cyclin D3 expression. In addition, expression levels of cyclin A and cyclin B were reduced during the total course of both TPO- and Ha-Ras(G12V)-induced megakaryocytic differentiation, thereby leading to decreased cdc2 kinase activity. Neither the induced expression of cyclin D1, cyclin D2, or cyclin D3 nor the expression of a dominant negative form of cdc2 alone could induce megakaryocytic differentiation of F-36P-mpl cells. In contrast, overexpression of dominant negative cdc2 together with cyclin D1, cyclin D2, or cyclin D3 facilitated megakaryocytic differentiation in the absence of TPO. These results suggest that both D-type cyclin expression and decreased cdc2 kinase activity may participate in megakaryocytic differentiation.     

Cyclin D1 is dispensable for G1 control in retinoblastoma gene-deficient cells independently of cdk4 activity.

To elucidate the regulator-versus-target relationship in the cyclin D1/cdk4/retinoblastoma protein (pRB) pathway, we examined fibroblasts from RB-1 gene-deficient and RB-1 wild-type littermate mouse embryos (ME) and in human tumor cell lines that differed in the status of the RB-1 gene. The RB+/+ and RB-/- ME fibroblasts expressed similar protein levels of D-type cyclins, cdk4, and cdk6, showed analogous spectra and abundance of cellular proteins complexed with cdk4 and/or cyclins D1 and D2, and exhibited comparable associated kinase activities. Of the two human cell lines established from the same sarcoma biopsy, the RB-positive SKUT1B cells contained cdk4 that was mainly associated with D-type cyclins, contrary to a predominant cdk4-p16INK4 complex in the RB-deficient SKUT1A cells. Antibody-mediated neutralization of cyclin D1 arrested the RB-positive ME and SKUT1B cells in G1, whereas this cyclin appeared dispensable in the RB-deficient ME and SKUT1A cells. Lack of requirement for cyclin D1 therefore correlated with absence of functional pRB, regardless of whether active cyclin D1/cdk4 holoenzyme was present in the cells under study. Consistent with a potential role of cyclin D/cdk4 in phosphorylation of pRB, monoclonal anti-cyclin D1 antibodies supporting the associated kinase activity failed to significantly affect proliferation of RB-positive cells, whereas the antibody DCS-6, unable to coprecipitate cdk4, efficiently inhibited G1 progression and prevented pRB phosphorylation in vivo. These data provide evidence for an upstream control function of cyclin D1/cdk4, and a downstream role for pRB, in the order of events regulating transition through late G1 phase of the mammalian cell division cycle.     

Activation of cyclin-dependent kinase 4 (cdk4) by mouse MO15-associated kinase.

The assembly of functional holoenzymes composed of regulatory D-type cyclins and cyclin-dependent kinases (cdks) is rate limiting for progression through the G1 phase of the mammalian somatic cell cycle. Complexes between D-type cyclins and their major catalytic subunit, cdk4, are catalytically inactive until cyclin-bound cdk4 undergoes phosphorylation on a single threonyl residue (Thr-172). This step is catalyzed by a cdk-activating kinase (CAK) functionally analogous to the enzyme which phosphorylates cdc2 and cdk2 at Thr-161/160. Here, we demonstrate that the catalytic subunit of mouse cdc2/cdk2 CAK (a 39-kDa protein designated p39MO15) can assemble with a regulatory protein present in either insect or mammalian cells to generate a CAK activity capable of phosphorylating and enzymatically activating both cdk2 and cdk4 in complexes with their respective cyclin partners. A newly identified 37-kDa cyclin-like protein (cyclin H [R. P. Fisher and D. O. Morgan, Cell 78:713-724, 1994]) can assemble with p39MO15 to activate both cyclin A-cdk2 and cyclin D-cdk4 in vitro, implying that CAK is structurally reminiscent of cyclin-cdk complexes themselves. Antisera produced to the p39MO15 subunit can completely deplete mammalian cell lysates of CAK activity for both cyclin A-cdk2 and cyclin D-cdk4, with recovery of activity in the resulting immune complexes. By using an immune complex CAK assay, CAK activity for cyclin A-cdk2 and cyclin D-cdk4 was detected both in quiescent cells and invariantly throughout the cell cycle. Therefore, although it is essential for the enzymatic activation of cyclin-cdk complexes, CAK appears to be neither rate limiting for the emergence of cells from quiescence nor subject to upstream regulatory control by stimulatory mitogens.     

A model for regulation of the cell cycle incorporating cyclin A, cyclin B and their complexes

Abstract. t. A mathematical model for the cell cycle is proposed that incorporates the known biochemical reactions involving both cyclin A and cyclin B, the interactions of these cyclins with cdc2 and cdk2, and the controlling effects of cdc25 and weel. The model also postulates the existence of an as yet unknown phosphatase involved in the formation of maturation promoting factor. The model produces solutions that agree qualitatively with a wide variety of experimentally observed cell-cycle behavior. Conditions under which the model could explain the initial rapid divisions of embryonic cells and the transition to the slower somatic cell cycle are also discussed.     

Cyclin D2 and Ha-Ras transformed rat embryo fibroblasts exhibit a novel deregulation of cell size control and early S phase arrest in low serum.

The D-type cyclins are growth factor-regulated delayed early functions which peak at the G1/S transition, are thought to regulate entry into S phase and have been implicated in tumorigenesis. Here, we show that cyclin D2 can co-operate with Ha-Ras to impose a novel transformed state on rat embryo fibroblasts (REF). While clonal cyclin D2/Ha-Ras REF transformants exhibit a characteristic transformed phenotype in high serum, in low serum they arrest cell proliferation and display profound morphological and cytological changes indicating loss of control of cell mass and deregulation of the G1/S transition. Notably, in low serum, despite re-establishment of actin cables and arrest of proliferation, cell mass continues to increase, creating giant cells up to 10 x normal size. Also, during low-serum culture the cells make a very gradual but progressive entry into S phase, reaching a 2.4N DNA content after 6 days. PCNA is expressed and 2N and 4N cells are largely absent, and thus the cells undergo a novel S phase arrest. While transfer to low serum induced the retinoblastoma protein to enter its dephosphorylated state, and cyclin A, cyclin B and cdc2 levels to decrease, all as normal, cyclin E, cdk4, cdk2 and the exogenous cyclin D2 persisted at high levels. These results indicate that cyclin D2 and Ha-Ras can transform cells when mitogenic signals from growth factors are provided. However, in low serum, co-operation of cyclin D2 and Ha-Ras provides only a subset of the progression signals and these are sufficient for G1-related cell mass increase and S phase entry, but are insufficient for full cell cycling.     

TGFbeta regulates the expression and activities of G2 checkpoint kinases in human myeloid leukemia cells

Transforming Growth Factor-beta (TGFbeta) is known to be a negative regulator of G1 cyclin/cdk activity. It is not clear whether TGFbeta has any effect on G2 checkpoint kinases. We have found that TGFbeta downregulated the expression of several G2 checkpoint kinases including cdc2, cyclin B1, and cdc25c without causing cell accumulation in G2/M phases in two human leukemia cell lines. The inhibition was time-dependent with a maximal inhibition being observed by 24h for cyclin B1 and cdc2 and by 48h for cdc25c. The inhibition was not a result of G1 arrest but a direct effect of TGFbeta which downregulates their expression at mRNA level. In proliferating cells, there was a significant formation of cdc2-pRb complexes, which was decreased to 30% of control levels by 48h after initiating TGFbeta treatment. Cdc2 showed a marked kinase activity on GST-Rb protein in proliferating cells detected by in vitro kinase assay, which was downregulated in response to TGFbeta. In addition, TGFbeta caused a rapid and transient dephosphorylation of cdc2 (Tyr15) and cdc25c (Ser216) for about 2-3h before a dramatic decrease of both molecules by 48h. Taken together, our data suggest that TGFbeta has a direct inhibitory effect on G2 checkpoint kinases, which is regulated at mRNA level. The transient activation of cdc2 and cdc25c and subsequent inhibition of cdc2, cyclin B1, and cdc25c could amplify TGFbeta-induced G1 arrest and growth inhibition.     

Coupling of the cell cycle and myogenesis through the cyclin D1-dependent interaction of MyoD with cdk4

Proliferating myoblasts express the muscle determination factor, MyoD, throughout the cell cycle in the absence of differentiation. Here we show that a mitogen-sensitive mechanism, involving the direct interaction between MyoD and cdk4, restricts myoblast differentiation to cells that have entered into the G0 phase of the cell cycle under mitogen withdrawal. Interaction between MyoD and cdk4 disrupts MyoD DNA-binding, muscle-specific gene activation and myogenic conversion of 10T1/2 cells independently of cyclin D1 and the CAK activation of cdk4. Forced induction of cyclin D1 in myotubes results in the cytoplasmic to nuclear translocation of cdk4. The specific MyoD-cdk4 interaction in dividing myoblasts, coupled with the cyclin D1-dependent nuclear targeting of cdk4, suggests a mitogen-sensitive mechanism whereby cyclin D1 can regulate MyoD function and the onset of myogenesis by controlling the cellular location of cdk4 rather than the phosphorylation status of MyoD.     

Cyclin B in fish oocytes: its cDNA and amino acid sequences, appearance during maturation, and induction of p34cdc2 activation.

Under the influence of maturation-inducing hormone (MIH) secreted from follicle cells, oocyte maturation is finally triggered by maturation-promoting factor (MPF), which consists of a homolog of the cdc2+ gene product of fission yeast (p34cdc2) and cyclin B. Two species of cyclin B clones were isolated from a cDNA library constructed from mature goldfish oocytes. Sequence comparisons revealed that these two clones are highly homologous (95%) and were found to be similar to Xenopus cyclin B1. Using monoclonal antibodies against Escherichia coli-produced goldfish cyclin B and the PSTAIR sequence of p34cdc2, we examined the levels of cyclin B and p34cdc2 proteins during goldfish oocyte maturation induced in vitro by 17 alpha, 20 beta-dihydroxy-4-pregnen-3-one (17 alpha, 20 beta-DP), a natural MIH in fish. Protein p34cdc2 was found in immature oocyte extracts and did not remarkably change during oocyte maturation. Cyclin B was not detected in immature oocyte extracts and appeared when oocytes underwent germinal vesicle breakdown. Cyclin B that appeared during oocyte maturation was labelled with [35S]methionine, indicating its de novo synthesis. Introduction of E. coli-produced cyclin B into immature oocyte extracts induced p34cdc2 (MPF) activation. Although the possibility that immature goldfish oocytes contain an insoluble cyclin B is not completely excluded, these results strongly suggest that 17 alpha, 20 beta-DP induces oocytes to synthesize cyclin B, which in turn activates preexisting p34cdc2, forming active MPF.     

Bovine papillomavirus E2 protein activates a complex growth-inhibitory program in p53-negative HT-3 cervical carcinoma cells that includes repression of cyclin A and cdc25A phosphatase genes and accumulation of hypophosphorylated retinoblastoma protein.

The bovine papillomavirus E2 protein can inhibit the proliferation of HT-3 cells, a p53-negative cervical carcinoma cell line containing integrated human papillomavirus type 30 DNA. Here, we analyzed HT-3 cells to explore the mechanism of p53-independent E2-mediated growth inhibition. Expression of the E2 protein repressed expression of the endogenous human papillomavirus type 30 E6/E7 genes. This was accompanied by hypophosphorylation and increased accumulation of p105Rb and repression of E2F1 expression. The E2 protein also caused reduced cyclin-dependent kinase (cdk) 2 activity, but this did not appear to be due to increased expression of cdk inhibitors. Rather, expression of cyclin A, which regulates cdk2 activity, and the cdc25A and cdc25B phosphatases, which are thought to activate cdk2, was significantly reduced at both the RNA and protein levels in response to E2 expression. The E2 protein reduced expression of cdc25A and cdc25B in both HT-3 and HeLa cells, but not in cells that were not growth-inhibited by the E2 protein. E2 point mutants unable to inhibit cell growth did not repress cdc25A and cdc25B expression, nor did the cell cycle inhibitors hydroxyurea and mimosine. Based on these results and the known properties of cell cycle components, we propose a model to account for E2-induced growth inhibition of cervical carcinoma cell lines.     

Restraint of apoptosis during mitosis through interdomain phosphorylation of caspase‐2

The apoptotic initiator caspase‐2 has been implicated in oocyte death, in DNA damage‐ and heat shock‐induced death, and in mitotic catastrophe. We show here that the mitosis‐promoting kinase, cdk1–cyclin B1, suppresses apoptosis upstream of mitochondrial cytochrome c release by phosphorylating caspase‐2 within an evolutionarily conserved sequence at Ser 340. Phosphorylation of this residue, situated in the caspase‐2 interdomain, prevents caspase‐2 activation. S340 was susceptible to phosphatase 1 dephosphorylation, and an interaction between phosphatase 1 and caspase‐2 detected during interphase was lost in mitosis. Expression of S340A non‐phosphorylatable caspase‐2 abrogated mitotic suppression of caspase‐2 and apoptosis in various settings, including oocytes induced to undergo cdk1‐dependent maturation. Moreover, U2OS cells treated with nocodazole were found to undergo mitotic catastrophe more readily when endogenous caspase‐2 was replaced with the S340A mutant to lift mitotic inhibition. These data demonstrate that for apoptotic stimuli transduced by caspase‐2, cell death is prevented during mitosis through the inhibitory phosphorylation of caspase‐2 and suggest that under conditions of mitotic arrest, cdk1–cyclin B1 activity must be overcome for apoptosis to occur.     

Role of cyclins in neuronal differentiation of immortalized hippocampal cells.

The proto-oncogene cyclin D1 and the neuron-specific cyclins p35 and p39 are expressed during brain maturation. To investigate the role of these cyclins in neuronal differentiation, we used a conditionally immortalized rat hippocampal cell line, H19-7, that expresses cyclin-dependent kinases 4 and 5 (cdk4 and -5). Cyclin D1, which activates cdk4 and binds but does not activate cdk5, was increased upon differentiation of the H19-7 cells. However, microinjection of either sense or antisense cyclin D1 cDNA or anti-cyclin D1 antibodies had no effect on morphological differentiation of the cells. On the other hand, neurite outgrowth was stimulated by expression of p35 or p39, both of which activate cdk5. A dominant-negative mutant of cdk5 blocked both p35- and p39-induced neurite extension as well as basic fibroblast growth factor (bFGF)-induced neuronal differentiation. However, of these cyclins, only antisense p39 prevented bFGF-induced neurite outgrowth. These studies indicate that cyclin D1 is neither necessary nor sufficient for morphological differentiation, that p35 is sufficient but not required, and that p39 is both necessary and sufficient for neurite outgrowth in the hippocampal cells. Taken together, these results represent the first demonstration of a specific role for p39 in neuronal differentiation, implicate the cyclin-activated kinase cdk5 in this process, and indicate that p39 is able to mediate neurite outgrowth in the presence or absence of cyclin D1.     

Mechanisms of cell cycle arrest in response to TGF-beta in progestin-dependent and -independent growth of mammary tumors

TGF-beta1 modulation of cell cycle components was assessed in an experimental model in which the synthetic progestin medroxyprogesterone acetate (MPA) induced mammary tumors in Balb/c mice. TGF-beta1 inhibited both MPA-induced proliferation of progestin-dependent C4HD epithelial cells and proliferation of the progestin-independent variant cell type C4HI, arresting cells in G(1) phase of the cell cycle. Progestin-independent 60 epithelial cells evidenced reduced response to TGF-beta1 antiproliferative effects. TGF-beta1 inhibition of cyclins D1 and A expression and up-regulation of p21(CIP1) levels were the common findings in all three cell types. In addition, a significant content reduction of cyclin D1/cdk4 and cyclin A/cdk2 complexes was found after TGF-beta1 inhibition of MPA-dependent and -independent proliferation. TGF-beta1 inhibited cyclin D2 expression and up-regulated p27(KIP1) levels only when acting as inhibitor of MPA-induced proliferation of C4HD cells. Regulation of these two cell cycle components resulted in decreased cyclin D2/cdk2 complex and in increased p27(KIP1) association with cdk2 in C4HD cells treated with TGF-beta1. These two molecular mechanisms, unobserved in progestin-independent growth of C4HI or 60 cells, were associated with a significantly higher degree of inhibition of cdk2 kinase activity in C4HD cells compared to that found in TGF-beta-treated C4HI or 60 cells. Reduced sensitivity of 60 cells to the growth-inhibitory effects of TGF-beta1 correlated with significantly lower levels of p15(INK4B), p21(CIP1), and p27(KIP1) expressed in these cells, compared to the levels present in C4HD or C4HI cells, and correlated as well with lack of expression of p16(INK4). Thus, common targets were found to exist in TGF-beta1 inhibitory action on breast cancer cells, but regulation of specific targets was found when TGF-beta1-inhibited proliferation driven by the progesterone receptor.     

p63 is necessary for the activation of human papillomavirus late viral functions upon epithelial differentiation

The late phase of the human papillomavirus (HPV) life cycle is linked to epithelial differentiation, and we investigated the factors that regulate this process. One potential regulator is p63, a member of the p53 family of proteins, which modulates epithelial development, as well as proliferation capability, in stem cells. In this study, we examined the role of p63 in the HPV life cycle using a lentiviral knockdown system for p63. In epithelial cells, the ΔN truncated isoforms of p63 predominate, while the full-length TA isoforms are present at very low levels. Upon the differentiation of normal keratinocytes, p63 levels rapidly decreased while higher levels were retained in HPV-positive cells. Our studies indicate that reducing p63 levels in differentiated HPV-positive cells resulted in the loss of viral genome amplification and late gene expression. p63 regulates the expression of cell cycle regulators, and we determined that cyclin A, cyclin B1, cdk1, and cdc25c were reduced in p63-deficient, HPV-positive keratinocytes, which suggests a possible mechanism of action. In addition, activation of the DNA repair pathway is necessary for genome amplification, and the expression of two members, BRCA2 and RAD51, was altered in the absence of p63 in HPV-positive cells. Our studies indicate that p63 is necessary for the activation of differentiation-dependent HPV late viral functions and provide insights into relevant cellular targets.     

Induction of cyclin mRNA and cyclin-associated histone H1 kinase during liver regeneration.

Cyclins and cyclin-associated cdc kinases are key regulators of oocyte maturation (Maller, J. L. (1990) in The Biology and Medicine of Signal Transduction (Nishizuka, Y., Endo, M., and Tanaka, C., eds) pp. 323-328, Raven Press, New York), yeast cell cycles (Nurse, P. (1990) Nature 344, 503-508), DNA replication in cell-free systems (D'Urso, F., Marraccino, R. L., Marshak, R. R., and Roberts, J. M. (1990) Science 250, 786-791), and amphibian cell proliferative transitions (Hunt, T. (1991) Nature 350, 462-463). The extent to which these regulatory molecules participate in the growth control of differentiated epithelial cells like hepatocytes is unknown. Therefore, we investigated the expression of "G1" (E, C, and D) and "G2/M" (A, B1, and B2) cyclin mRNAs, the relative levels of cyclin A- and B1-associated histone H1-kinase activity, and the appearance of cyclin-associated kinases (p32/p33cdk2 and p33/p34cdc2) in regenerating rat liver and in control tissues from sham hepatectomized rats. To do this, we exploited a battery of human cyclin cDNAs and cyclin antisera that recognize rat molecules. The results suggest an apparent sequence of regeneration-specific changes: 1) elevated and induced expression of cyclins E (2.1 kilobases (kb)) and C (4 kb), and D mRNAs (4 kb), within 12 h, respectively; 2) induction of cyclins A (3.4 and 1.8 kb), B1 (2.5 and 1.8 kb), and B2 (1.9 kb) mRNAs at 24 h; 3) induction of cyclin A- and B1-associated nuclear histone H1 kinase at 24 h; and 4) enhanced levels of PSTAIRE-containing proteins of Mr approximately 32-33 and 33-34 kDa in nuclear extracts from 24-h regenerating liver that co-immunoprecipitate with cyclin A and B1 antisera, respectively. These observations provide an intellectual framework that unifies the biology of hepatocyte mitogenesis, proto-oncogene expression, and the machinery of the cell cycle.     

Purification and biochemical analysis of catalytically active human cdc25C dual specificity phosphatase

We describe a reliable and efficient method for the purification of catalytically active and mutant inactive full-length forms of the human dual specificity phosphatase cdc25C from bacteria. The protocol involves isolating insoluble cdc25C protein in inclusion bodies, solubilization in guanidine HCL, and renaturation through rapid dilution into low salt buffer. After binding renatured proteins to an ion exchange resin, cdc25C elutes in two peaks at 350 and 450 mM NaCl. Analysis by gel exclusion chromatography and enzymatic assays reveals the highest phosphatase activity is associated with the 350 mM NaCl with little or no activity present in the 450 mM peak. Furthermore, active cdc25C has a native molecular mass of 220 kDa consistent with a potential tetrameric complex of the 55-kDa cdc25C protein. Assaying phosphatase activity against artificial substrates pNPP and 3-OMFP reveals a 220 kDa form of the phosphatase is active in a non-phosphorylated state. The protein effectively activates cdk1/cyclin B prokinase complexes in vitro in the absence of cdk1 kinase activity in an orthovanadate sensitive manner but is inactivated by A-kinase phosphorylation. In vitro phosphorylation of purified cdc25C by cdk1/cyclin B1, cdk2/cyclin A2 and cdk2/cyclin E shows that distinct TP/SP mitotic phosphorylation sites on cdc25C are differentially phosphorylated by these 3 cdk/cyclin complexes associated with different levels of cdc25C activation. Finally, we show that endogenous native cdc25C from human cells is present in high molecular weight complexes with other proteins and resolves mostly above 200-kDa. These data show that untagged cdc25C can be purified with a simple protocol as an active dual specificity phosphatase with a native molecular mass consistent with a homo-tetrameric configuration.     

The G2/M checkpoint phosphatase cdc25C is located within centrosomes

cdc25C is a phosphatase which regulates the activity of the mitosis promoting factor cyclin B/cdk1 by dephosphorylation, thus triggering G(2)/M transition. The activity and the sub-cellular localisation of cdc25C are regulated by phosphorylation. It is well accepted that cdc25C has to enter the nucleus to activate the cyclin B/cdk1 complex at G(2)/M transition. Here, we will show that cdc25C is located in the cytoplasm at defined dense structures, which according to immunofluorescence analysis, electron microscopy as well as biochemical subfractionation, are proven to be the centrosomes. Since cyclin B and cdk1 are also located at the centrosomes, this subfraction of cdc25C might participate in the control of the onset of mitosis suggesting a further role for cdc25C at the centrosomes.