Cdk2 knockout mice are viable

BACKGROUND: Cyclin-dependent kinases (Cdks) and their cyclin regulatory subunits control cell growth and division. Cdk2/cyclin E complexes are thought to be required because they phosphorylate the retinoblastoma protein and drive cells through the G1/S transition into the S phase of the cell cycle. In addition, Cdk2 associates with cyclin A, which itself is essential for cell proliferation during early embryonic development. RESULTS: In order to study the functions of Cdk2 in vivo, we generated Cdk2 knockout mice. Surprisingly, these mice are viable, and therefore Cdk2 is not an essential gene in the mouse. However, Cdk2 is required for germ cell development; both male and female Cdk2(-/-) mice are sterile. Immunoprecipitates of cyclin E1 complexes from Cdk2(-/-) spleen extracts displayed no activity toward histone H1. Cyclin A2 complexes were active in primary mouse embryonic fibroblasts (MEFs), embryo extracts and in spleen extracts from young animals. In contrast, there was little cyclin A2 kinase activity in immortalized MEFs and spleen extracts from adult animals. Cdk2(-/-) MEFs proliferate but enter delayed into S phase. Ectopic expression of Cdk2 in Cdk2(-/-) MEFs rescued the delayed entry into S phase. CONCLUSIONS: Although Cdk2 is not an essential gene in the mouse, it is required for germ cell development and meiosis. Loss of Cdk2 affects the timing of S phase, suggesting that Cdk2 is involved in regulating progression through the mitotic cell cycle.     

Combined loss of Cdk2 and Cdk4 results in embryonic lethality and Rb hypophosphorylation

Mouse knockouts of Cdk2 and Cdk4 have demonstrated that, individually, these genes are not essential for viability. To investigate whether there is functional redundancy, we have generated double knockout (DKO) mice. Cdk2-/- Cdk4-/- DKOs die during embryogenesis around E15 as a result of heart defects. We observed a gradual decrease of Retinoblastoma protein (Rb) phosphorylation and reduced expression of E2F-target genes, like Cdc2 and cyclin A2, during embryogenesis and in embryonic fibroblasts (MEFs). DKO MEFs are characterized by a decreased proliferation rate, impaired S phase entry, and premature senescence. HPV-E7-mediated inactivation of Rb restored normal expression of E2F-inducible genes, senescence, and proliferation in DKO MEFs. In contrast, loss of p27 did not rescue Cdk2-/- Cdk4-/- phenotypes. Our results demonstrate that Cdk2 and Cdk4 cooperate to phosphorylate Rb in vivo and to couple the G1/S phase transition to mitosis via E2F-dependent regulation of gene expression.     

Androgen Suppresses the Proliferation of Androgen Receptor-Positive Castration-Resistant Prostate Cancer Cells via Inhibition of Cdk2, CyclinA,and Skp2

The majority of prostate cancer (PCa) patient receiving androgen ablation therapy eventually develop castration-resistant prostate cancer (CRPC). We previously reported that androgen treatment suppresses Skp2 and c-Myc through androgen receptor (AR) and induced G1 cell cycle arrest in androgen-independent LNCaP 104-R2 cells, a late stage CRPC cell line model. However, the mechanism of androgenic regulation of Skp2 in CRPC cells was not fully understood. In this study, we investigated the androgenic regulation of Skp2 in two AR-positive CRPC cell line models, the LNCaP 104-R1 and PC-3^AR Cells. The former one is an early stage androgen-independent LNCaP cells, while the later one is PC-3 cells re-expressing either wild type AR or mutant LNCaP AR. Proliferation of LNCaP 104-R1 and PC-3^AR cells is not dependent on but is suppressed by androgen. We observed in this study that androgen treatment reduced protein expression of Cdk2, Cdk7, Cyclin A, cyclin H, Skp2, c-Myc, and E2F-1; lessened phosphorylation of Thr14, Tyr15, and Thr160 on Cdk2; decreased activity of Cdk2; induced protein level of p27^Kip1; and caused G1 cell cycle arrest in LNCaP 104-R1 cells and PC-3^AR cells. Overexpression of Skp2 protein in LNCaP 104-R1 or PC-3^AR cells partially blocked accumulation of p27^Kip1 and increased Cdk2 activity under androgen treatment, which partially blocked the androgenic suppressive effects on proliferation and cell cycle. Analyzing on-line gene array data of 214 normal and PCa samples indicated that gene expression of Skp2, Cdk2, and cyclin A positively correlates to each other, while Cdk7 negatively correlates to these genes. These observations suggested that androgen suppresses the proliferation of CRPC cells partially through inhibition of Cyclin A, Cdk2, and Skp2.     

Unmasking the Redundancy Between Cdk1 and Cdk2 at G2 Phase in Human Cancer Cell Lines

A series of studies published in 2003 has challenged the essentiality of Cdk2. A recently published work indicates that cyclin E-Cdk1 compensates for Cdk2’s function at G1/S transition in Cdk2-/- Mefs. In this study, we uncovered a redundant mechanism between Cdk1 and Cdk2 at G2 in multiple cancer cell lines. When either Cdk2 or Cdk1 is ablated using RNAi, there were complex shifts of cyclin A towards its reciprocal partner, i.e., when Cdk2 is ablated, cyclin A redistributes to Cdk1; when Cdk1 is ablated, cyclin A forms more abundant complexes with Cdk2. Further, cyclin B redistributes to Cdk2 upon Cdk1 knockdown. These redistributions bring about increased kinase activities of corresponding complexes. Elimination of the compensatory mechanism by knockdown of both Cdk1 and Cdk2 using RNAi reveals phenotypes at G2 phase. The results suggest that the redistributed complexes contribute to the cyclin B-Cdk1 activation when either Cdk1 or Cdk2 alone is ablated and this redundancy masks Cdk2’s role when Cdk2 is singly ablated. It is also worth noting that the predominant G2 arrest described here, unlike those Cdk1-Cdk2 double ablated Mefs, raises a question of whether different Cdk activities are required for G1/S or G2/M progression in normal vs. cancer cells.     

The oncogenic activity of cyclin E is not confined to Cdk2 activation alone but relies on several other,distinct functions of the protein

We have previously shown that cyclin E can malignantly transform primary rat embryo fibroblasts in cooperation with constitutively active Ha-Ras. In addition, we demonstrated that high level cyclin E expression potentiates the development of methyl-nitroso-urea-induced T-cell lymphomas in mice. To further investigate the mechanism underlying cyclin E-mediated malignant transformation, we have performed a mutational analysis of cyclin E function. Here we show that cyclin E mutants defective to form an active kinase complex with Cdk2 are unable to drive cells from G(1) into S phase but can still malignantly transform rat embryo fibroblasts in cooperation with Ha-Ras. In addition, Cdk2 activation is not a prerequisite for the ability of cyclin E to rescue yeast triple cln mutations. We also find that the oncogenic properties of cyclin E did not entirely correspond with its ability to interact with the negative cell cycle regulator p27(Kip1) or the pocket protein p130. These findings suggest that the oncogenic activity of cyclin E does not exclusively rely on its ability as a positive regulator of G(1) progression. Rather, we propose that cyclin E harbors other functions, independent of Cdk2 activation and p27(Kip1) binding, that contribute significantly to its oncogenic activity.     

Phosphorylation of threonine 61 by cyclin a/Cdk1 triggers degradation of stem-loop binding protein at the end of S phase

Histone mRNA levels are cell cycle regulated, and a major regulatory mechanism is restriction of stem-loop binding protein (SLBP) to S phase. Degradation of SLBP at the end of S phase results in cessation of histone mRNA biosynthesis, preventing accumulation of histone mRNA until SLBP is synthesized just before entry into the next S phase. Degradation of SLBP requires an SFTTP (58 to 62) and KRKL (95 to 98) sequence, which is a putative cyclin binding site. A fusion protein with the 58-amino-acid sequence of SLBP (amino acids 51 to 108) fused to glutathione S-transferase (GST) is sufficient to mimic SLBP degradation at late S phase. Using GST-SLBP fusion proteins as a substrate, we show that cyclin A/Cdk1 phosphorylates Thr61. Furthermore, knockdown of Cdk1 by RNA interference stabilizes SLBP at the end of S phase. Phosphorylation of Thr61 is necessary for subsequent phosphorylation of Thr60 by CK2 in vitro. Inhibitors of CK2 also prevent degradation of SLBP at the end of S phase. Thus, phosphorylation of Thr61 by cyclin A/Cdk1 primes phosphorylation of Thr60 by CK2 and is responsible for initiating SLBP degradation. We conclude that the increase in cyclin A/Cdk1 activity at the end of S phase triggers degradation of SLBP at S/G(2).     

Recognition of Cdk2 by Cdk7

Cdk7, a member of the cyclin dependent protein kinase family, regulates the activities of other Cdks through phosphorylation on their activation segment, and hence contributes to control of the eukaryotic cell cycle. Cdk7 is itself phosphorylated on the activation segment. Cdk7 phosphorylates Cdk1, Cdk2, Cdk4, and Cdk6, but only Cdk1 and Cdk2 can phosphorylate Cdk7 and none of them is able to auto-phosphorylate. The activation segments of the Cdks are very similar in sequence. Their specificity does not appear to be dictated by the sequences surrounding the phosphorylation sites but by structural determinants at remote sites. Through mutagenesis studies, we have identified regions in Cdk2 responsible for its interaction with Cdk7. A model has been built that explains the molecular basis for the specificity observed in Cdk recognition. The two kinases are arranged in a quasi-symmetric head-to-tail arrangement in which the N-terminal lobe from one kinase docks against the C-terminal lobe from the other kinase, and the activation segments are within reach of the opposite catalytic sites. Further experiments demonstrate that cyclin A hydrophobic pocket is not a recruitment site for Cdk7.     

Hepatocyte growth factor releases mink epithelial cells from transforming growth factor beta1-induced growth arrest by restoring Cdk6 expression and cyclin E-associated Cdk2 activity

Transforming growth factor beta (TGF-beta) potently suppresses Mv1Lu mink epithelial cell growth, whereas hepatocyte growth factor (HGF) counteracts TGF-beta-mediated growth inhibition and induces Mv1Lu cell proliferation (J. Taipale and J. Keski-Oja, J. Biol. Chem. 271:4342-4348, 1996). By addressing the cell cycle regulatory mechanisms involved in HGF-mediated release of Mv1Lu cells from TGF-beta inhibition, we show that increased DNA replication is accompanied by phosphorylation of the retinoblastoma protein and alternative regulation of cyclin-Cdk-inhibitor complexes. While TGF-beta treatment decreased the expression of Cdk6, this effect was counteracted by HGF, followed by partial restoration of cyclin D2-associated kinase activity. Notably, HGF failed to prevent TGF-beta induction of p15 and its association with Cdk6. However, HGF reversed the TGF-beta-mediated decrease in Cdk6-associated p27 and cyclin D2-associated Cdk6, suggesting that HGF modifies the TGF-beta response at the level of G1 cyclin complex formation. Counteraction of TGF-beta regulation of Cdk6 by HGF may in turn affect the association of p27 with Cdk2-cyclin E complexes. Though HGF did not differentially regulate the total levels of p27 in TGF-beta-treated cells, p27 immunodepletion experiments suggested that upon treatment with both growth factors, less p27 is associated with Cdk2-cyclin E complexes, in parallel with restoration of the active form of Cdk2 and the associated kinase activity. The results demonstrate that HGF intercepts TGF-beta cell cycle regulation at multiple points, affecting both G1 and G1-S cyclin kinase activities.     

Cell Cycle Sibling Rivalry: Cdc2 Versus Cdk2

It has been long believed that the cyclin-dependent kinase 2 [Cdk2] binds to cyclin E or cyclin Aand exclusively promotes the G1/S phase transition and that Cdc2/cyclin B complexes play a majorrole in mitosis. We now provide evidence that Cdc2 binds to cyclin E [in addition to cyclin A & B]and is able to promote the G1/S transition. This new concept indicates that both Cdk2 and/or Cdc2can drive cells through G1/S phase in parallel. In this review we discuss the classic cell cycle modeland how results from knockout mice provide new evidence that refute this model. We focus on newroles of Cdc2 and p27 in regulating the mammalian cell cycle and propose a new model for cellcycle regulation that accommodates these novel findings.     

Vitamin D inhibits G1 to S progression in LNCaP prostate cancer cells through p27Kip1 stabilization and Cdk2 mislocalization to the cytoplasm

1,25-(OH)2 vitamin D3 (1,25-(OH)2D3) exerts antiproliferative effects via cell cycle regulation in a variety of tumor cells, including prostate. We have previously shown that in the human prostate cancer cell line LN-CaP, 1,25-(OH)2D3 mediates an increase in cyclin-dependent kinase inhibitor p27Kip1 levels, inhibition of cyclin-dependent kinase 2 (Cdk2) activity, hypophosphorylation of retinoblastoma protein, and accumulation of cells in G1. In this study, we investigated the mechanism whereby 1,25-(OH)2D3 increases p27 levels. 1,25-(OH)2D3 had no effect on p27 mRNA levels or on the regulation of a 3.5-kb fragment of the p27 promoter. The rate of p27 protein synthesis was not affected by 1,25-(OH)2D3 as measured by luciferase activity driven by the 5'- and 3'-untranslated regions of p27 that regulate p27 protein synthesis. Pulse-chase analysis of 35S-labeled p27 revealed an increased p27 protein half-life with 1,25-(OH)2D3 treatment. Because Cdk2-mediated phosphorylation of p27 at Thr187 targets p27 for Skp2-mediated degradation, we examined the phosphorylation status of p27 in 1,25-(OH)2D3-treated cells. 1,25-(OH)2D3 decreased levels of Thr187 phosphorylated p27, consistent with inhibition of Thr187 phosphorylation-dependent p27 degradation. In addition, 1,25-(OH)2D3 reduced Skp2 protein levels in LNCaP cells. Cdk2 is activated in the nucleus by Cdk-activating kinase through Thr160 phosphorylation and by cdc25A phosphatase via Thr14 and Tyr15 dephosphorylation. Interestingly, 1,25-(OH)2D3 decreased nuclear Cdk2 levels as assessed by subcellular fractionation and confocal microscopy. Inhibition of Cdk2 by 1,25-(OH)2D3 may thus involve two mechanisms: 1) reduced nuclear Cdk2 available for cyclin binding and activation and 2) impairment of cyclin E-Cdk2-dependent p27 degradation through cytoplasmic mislocalization of Cdk2. These data suggest that Cdk2 mislocalization is central to the antiproliferative effects of 1,25-(OH)2D3.     

A kinetic model of the cyclin E/Cdk2 developmental timer in Xenopus laevis embryos

Early cell cycles of Xenopus laevis embryos are characterized by rapid oscillations in the activity of two cyclin-dependent kinases. Cdk1 activity peaks at mitosis, driven by periodic degradation of cyclins A and B. In contrast, Cdk2 activity oscillates twice per cell cycle, despite a constant level of its partner, cyclin E. Cyclin E degrades at a fixed time after fertilization, normally corresponding to the midblastula transition. Based on published data and new experiments, we constructed a mathematical model in which: (1) oscillations in Cdk2 activity depend upon changes in phosphorylation, (2) Cdk2 participates in a negative feedback loop with the inhibitory kinase Wee1; (3) cyclin E is cooperatively removed from the oscillatory system; and (4) removed cyclin E is degraded by a pathway activated by cyclin E/Cdk2 itself. The model's predictions about embryos injected with Xic1, a stoichiometric inhibitor of cyclin E/Cdk2, were experimentally validated.     

Multisite phosphorylation by Cdk2 and GSK3 controls cyclin E degradation

Autophosphorylation-triggered ubiquitination has been proposed to be the major pathway regulating cyclin E protein abundance: phosphorylation of cyclin E on T380 by its associated CDK allows binding to the receptor subunit, Fbw7, of the SCFFbw7 ubiquitin ligase. We have tested this model in vivo and found it to be an inadequate representation of the pathways that regulate cyclin E degradation. We show that assembly of cyclin E into cyclin E-Cdk2 complexes is required in vivo for turnover by the Fbw7 pathway; that Cdk2 activity is required for cyclin E turnover in vivo because it phosphorylates S384; that phosphorylation of T380 in vivo does not require Cdk2 and is mediated primarily by GSK3; and that two additional phosphorylation sites, T62 and S372, are also required for turnover. Thus, cyclin E turnover is controlled by multiple biological inputs and cannot be understood in terms of autophosphorylation alone.     

Binding of HTm4 to cyclin-dependent kinase (Cdk)-associated phosphatase (KAP).Cdk2.cyclin A complex enhances the phosphatase activity of KAP, dissociates cyclin A, and facilitates KAP dephosphorylation of Cdk2

Cyclin-dependent kinase 2 (cdk2) activation requires phosphorylation of Thr160 and dissociation from cyclin A. The T-loop of cdk2 contains a regulatory phosphorylation site at Thr160. An interaction between cdc-associated phosphatase (KAP) and cdk2 compromises the interaction between cdk2 and cyclin A, which permits access of KAP, a Thr160-directed phosphatase, to its substrate, cdk2. We have reported that KAP is bound and activated by a nuclear membrane protein, HTm4. Here, we present in vitro data showing the direct interaction between the HTm4 C terminus and KAP Tyr141. We show that this interaction not only facilitates access of KAP to Thr160 and accelerates KAP kinetics, but also forces exclusion of cyclin A from the KAP.cdk2 complex.     

Non-dependence of cyclin E/Cdk2 kinase activity on the initiation of oocyte maturation in goldfish

Cdk2 kinase activity increases during oocyte maturation but neither cyclin A nor B is associated with Cdk2 in mature oocytes in goldfish. As a potential Cdk2 partner in meiosis, a cyclin E homolog was isolated from a goldfish oocyte cDNA library. A monoclonal antibody was raised against bacterially produced full-length goldfish cyclin E. Both cyclin E and Cdk2 were already present in immature oocytes and their protein levels did not change remarkably during oocyte maturation. Cyclin E formed a complex mainly with Cdk2 just at the time of germinal vesicle breakdown (GVBD) in association with the increase in Cdk2 kinase activity, although a fraction of cyclin E bound to Cdk(s) other than Cdk2 and Cdc2. Ectopic activation of cyclin E/Cdk2 by the injection of cyclin E messenger RNA (mRNA) into immature oocytes did not induce maturation-promoting factor (MPF) activation and GVBD. Furthermore, inhibition of cyclin E/Cdk2 kinase activity by the injection of p21SDI1 into the oocytes treated with 17alpha,20beta-dihydroxy-4-pregnen-3-one had no effect on MPF activation and GVBD. These results indicate that cyclin E/Cdk2 kinase activity is insufficient and unnecessary for initiating goldfish oocyte maturation.     

Differential control of Eg5-dependent centrosome separation by Plk1 and Cdk1

Cyclin-dependent kinase 1 (Cdk1) is thought to trigger centrosome separation in late G2 phase by phosphorylating the motor protein Eg5 at Thr927. However, the precise control mechanism of centrosome separation remains to be understood. Here, we report that in G2 phase polo-like kinase 1 (Plk1) can trigger centrosome separation independently of Cdk1. We find that Plk1 is required for both C-Nap1 displacement and for Eg5 localization on the centrosome. Moreover, Cdk2 compensates for Cdk1, and phosphorylates Eg5 at Thr927. Nevertheless, Plk1-driven centrosome separation is slow and staggering, while Cdk1 triggers fast movement of the centrosomes. We find that actin-dependent Eg5-opposing forces slow down separation in G2 phase. Strikingly, actin depolymerization, as well as destabilization of interphase microtubules (MTs), is sufficient to remove this obstruction and to speed up Plk1-dependent separation. Conversely, MT stabilization in mitosis slows down Cdk1-dependent centrosome movement. Our findings implicate the modulation of MT stability in G2 and M phase as a regulatory element in the control of centrosome separation.     

Differential regulation of Cdc2 and Cdk2 by RINGO and cyclins.

Cyclin-dependent kinases (Cdks) are key regulators of the eukaryotic cell division cycle. Cdk1 (Cdc2) and Cdk2 should be bound to regulatory subunits named cyclins as well as phosphorylated on a conserved Thr located in the T-loop for full enzymatic activity. Cdc2- and Cdk2-cyclin complexes can be inactivated by phosphorylation on the catalytic cleft-located Thr-14 and Tyr-15 residues or by association with inhibitory subunits such as p21(Cip1). We have recently identified a novel Cdc2 regulator named RINGO that plays an important role in the meiotic cell cycle of Xenopus oocytes. RINGO can bind and activate Cdc2 but has no sequence homology to cyclins. Here we report that, in contrast with Cdc2- cyclin complexes, the phosphorylation of Thr-161 is not required for full activation of Cdc2 by RINGO. We also show that RINGO can directly stimulate the kinase activity of Cdk2 independently of Thr-160 phosphorylation. Moreover, RINGO-bound Cdc2 and Cdk2 are both less susceptible to inhibition by p21(Cip1), whereas the Thr-14/Tyr-15 kinase Myt1 can negatively regulate the activity of Cdc2-RINGO with reduced efficiency. Our results indicate that Cdk-RINGO complexes may be active under conditions in which cyclin-bound Cdks are inhibited and can therefore play different regulatory roles.     

FOXM1c is activated by cyclin E/Cdk2, cyclin A/Cdk2, and cyclin A/Cdk1, but repressed by GSK-3alpha

Two different inhibitory domains, N-terminus and central domain, keep FOXM1c almost inactive despite its strong transactivation domain. Here, we demonstrate that cyclin E/Cdk2, cyclin A/Cdk2, and cyclin A/Cdk1 activate FOXM1c. Cyclin E/Cdk2 does not target its transactivation domain or its DNA-binding domain. Instead, its activating effect strictly depends on the presence of either the central domain or the N-terminus of FOXM1c and thus is completely lost if both inhibitory domains are deleted. Cyclin E/Cdk2 activates FOXM1c by releasing its transactivation domain from the repression by these two inhibitory domains. However, it does not directly increase the transactivation potential of the TAD. The DNA-binding is not affected by cyclin E/Cdk2, neither directly nor indirectly. These two activating effects of cyclin E/Cdk2 via central domain and N-terminus are additive. Cyclin A/Cdk2 and cyclin A/Cdk1 show similar characteristics. GSK-3alpha, another proliferation-associated kinase, represses FOXM1c.