Mammalian cell-cycle regulation: several Cdks, numerous ... - Nature

Oncogene (2009) 28, 2925–2939

& 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00 www.nature.com/onc

REVIEW

Mammalian cell-cycle regulation: several Cdks, numerous cyclins and diverse compensatory mechanisms A Satyanarayana1 and P Kaldis2 1

Mouse Cancer Genetics Program, Center for Cancer Research, National Cancer Institute-Frederick, Frederick, MD, USA and Laboratory of Cell Division and Cancer, Institute of Molecular and Cell Biology (IMCB), Singapore, Republic of Singapore

2

After a decade of extensive work on gene knockout mouse models of cell-cycle regulators, the classical model of cell-cycle regulation was seriously challenged. Several unexpected compensatory mechanisms were uncovered among cyclins and Cdks in these studies. The most astonishing observation is that Cdk2 is dispensable for the regulation of the mitotic cell cycle with both Cdk4 and Cdk1 covering for Cdk2’s functions. Similar to yeast, it was recently discovered that Cdk1 alone can drive the mammalian cell cycle, indicating that the regulation of the mammalian cell cycle is highly conserved. Nevertheless, cell–cycle-independent functions of Cdks and cyclins such as in DNA damage repair are still under investigation. Here we review the compensatory mechanisms among major cyclins and Cdks in mammalian cell-cycle regulation. Oncogene (2009) 28, 2925–2939; doi:10.1038/onc.2009.170; published online 29 June 2009 Keywords: cell cycle; cyclin; Cdk; DNA damage; meiosis; knockout mouse

Introduction Cell division is a precisely regulated process and cells of living organisms are required to divide and to produce daughter cells. The ability of the cells to divide in turn is mainly attributed to the presence of two classes of molecules, cyclin-dependent kinases (Cdks), a family of serine/threonine kinases and their binding partners, cyclins (Morgan, 1997). In yeast, only one major Cdk is expressed (designated Cdk1, equivalent to p34Cdc28 in Saccharomyces cerevisiae and p34Cdc2 in Schizosaccharomyces pombe). Cdk1 activity oscillates during cell-cycle progression and is capable of regulating diverse cellcycle transitions (G1/S, S and G2/M phases) by associating with different cell-cycle stage-specific Correspondence: Professor P Kaldis, Laboratory of Cell Division and Cancer, Institute of Molecular and Cell Biology (IMCB), 61 Biopolis Drive, Proteos, 3-10B, Singapore 138673, Republic of Singapore. E-mail: [email protected] or Dr A Satyanarayana, Mouse Cancer Genetics Program, Center for Cancer Research, National Cancer Institute-Frederick, Bldg. 560/22-69, 1050 Boyles Street, Frederick, MD 21702-1201, USA. E-mail: [email protected] Received 11 March 2009; revised 17 May 2009; accepted 23 May 2009; published online 29 June 2009

cyclins (for example in S. pombe Cdc13, Cig1 and Cig2). Besides Cdk1, yeast also expresses Cdks such as PHO85 and Kin28 that may affect cell-cycle regulation in an indirect way (Morgan, 1997; Solomon et al., 1988; Huang et al., 2007). Many functional homologues of yeast Cdk1 have been identified in higher organisms. These specialized G1, G1/S, S and G2/M phase-specific Cdks have replaced the major multifunctional Cdk1 of yeast. The discovery of approximately 20 Cdk-related proteins led to the concept that in higher eukaryotic cells, cell-cycle events occur by complex combinations of Cdks (Cdk1, Cdk2, Cdk3, Cdk5, Cdk4, Cdk6, Cdk7, Cdk8 and so on) and cyclins (A1, A2, B1, B2, B3, C, D1, D2, D3, E1, E2, F and so on) in different phases of cell cycle, which in turn provide additional control to the cell-cycle machinery (Figure 1). Cyclins were thought to confer substrate specificity and regulation (activation, inactivation, localization, binding of subunits and others) to Cdk/cyclin complexes. In accordance with this hypothesis, extensive research was conducted in eukaryotic cells and the classical model of cell-cycle regulation was established. According to this model, in early G1 phase Cdk4 and/ or Cdk6 are activated by D-type cyclins and initiate phosphorylation of the retinoblastoma protein (Rb) family (Rb, p107 and p130; Sherr and Roberts, 1999, 2004). This leads to the release of E2F transcription factors and results in the activation and transcription of E2F responsive genes required for cell-cycle progression (Weinberg, 1995; Dyson, 1998). Early E2F responsive genes include E- and A-type cyclins (Lundberg and Weinberg, 1998; Sherr and Roberts, 1999). In the late G1 phase, Cdk2 is activated by binding to cyclin E (for review see Sherr and Roberts, 1999, 2004) and completes the phosphorylation of Rb (pocket proteins) leading to further activation of E2F mediated transcription. This leads to passage through the restriction point at the boundary of the G1/S phase, and to S phase initiation. Later Cdk2 plays an important role in S phase progression by complexing with cyclin A. A-type cyclins are synthesized at the onset of the S phase and phosphorylate proteins involved in DNA replication (Petersen et al., 1999; Coverley et al., 2000). During the G2/M transition, Cdk1/cyclin A activity is required for the initiation of prophase (Furuno et al., 1999). Finally, Cdk1/cyclin B complexes actively participate in and complete mitosis (Riabowol et al., 1989). The activities and functions of Cdk/cyclin complexes under

Cyclins and Cdks in mammalian cell-cycle regulation A Satyanarayana and P Kaldis

2926

Compensatory mechanisms among cyclins A decade of extensive work on gene knockout mouse models of cyclins disclosed that there is little significant impact on cell-cycle regulation or survival of the organisms in the absence of several individual cyclins, because each family of cyclins consists of numerous members. Four major classes of cyclins were discovered in mammalian cells: D-, E-, A-, B-type cyclins, which display dynamic localization patterns. In mammalian cells, cyclin B resides mainly in the cytoplasm, whereas cyclins A, D and E are nuclear (Sherr, 1993; Pines and Hunter, 1994; Ohtsubo et al., 1995). D-type cyclins (D1, D2 and D3) are expressed in a variety of cell types and tissues with cyclin D1 being the most ubiquitously expressed (Waclaw and Chatot, 2004).

Figure 1 Mammalian Cdk family members and their interacting cyclin partners. Although more than 20 Cdk proteins have been identified in mammalian cells, here we are presenting only those Cdks whose cell-cycle and cell-cycle-independent functions are extensively studied.

normal as well as extreme conditions, such as stress, DNA damage, telomere dysfunction and others, are regulated by two families of Cdk inhibitors; the INK4 family (p16INK4a, p15INK4b, p18INK4c, p19INK4d) specifically bind to Cdk4 and Cdk6 and prevent D-type cyclin activity and the Cip/Kip family (p21Cip1/Waf1/Sdi1, p27Kip1, p57Kip2) inhibits Cdk2/cyclin E, Cdk2/cyclin A, Cdk1/ cyclin A, as well as Cdk1/cyclin B activity (Toyoshima and Hunter, 1994; Aprelikova et al., 1995; O’Connor, 1997). The presence of multiple Cdks and cyclins in higher eukaryotic cells led to the speculation that each Cdk/ cyclin complex harbors unique functions restricted to a particular phase of the cell cycle. In yeast, profound redundancies between the many cyclin molecules were uncovered but there are only few Cdks present. In contrast, mammals express approximately 20 Cdks and the specific function of each Cdk was not clear. Therefore, mouse models represented a useful approach to study the functions of Cdks in vivo. It was found that among the mammalian cyclins, widespread compensation was detected confirming the findings in yeast. Oncogene

D-type cyclins Cyclin D knockout mouse models uncovered an intricate situation where different members of the Cdk/D-type cyclin families play cell type-specific roles (Fantl et al., 1995; Sicinski et al., 1995). Mice lacking individual D-type cyclins were viable, developed normally and did not show any severe impairment of cell proliferation (Kozar et al., 2004) but displayed several cell type-specific abnormalities. Cyclin D1/ mice had reduced body size, displayed hypoplastic retinas and neurological abnormalities. The mammary glands of females failed to undergo normal lobuloalveolar development during pregnancy (Fantl et al., 1995; Sicinski et al., 1995). The cyclin D2/ mice displayed defects in B-lymphocyte proliferation (Sicinski et al., 1996; Solvason et al., 2000), pancreatic b-cell proliferation, cerebellar development (Huard et al., 1999) and adult neurogenesis (Kowalczyk et al., 2004). Contrary to cyclin D2/ mice, cyclin D3 deletion led to a defect in the T-lymphocyte development (Sicinska et al., 2003). It appears that loss of one of the D-type cyclins could simply be substituted by one of its siblings in the majority of the cell types to prevent severe consequences on cell proliferation and survival (Figure 2). Nevertheless, it is clear that individual D-type cyclins still play cell type-specific roles such as D2 has a specific role in B lymphocytes whereas D3 has a unique role in T lymphocytes. These minor cell type-specific defects in single knockouts might be due to a lack of expression or difference in the timing of expression of each of the D-type cyclins in that particular cell type. In support to this view, it was demonstrated that cyclin D2 rescues the loss of cyclin D1 when expressed from the D1 locus (Carthon et al., 2005). This indicates that under normal circumstances D2 can compensate for the loss of D1 but a slight difference in the timing of expression between D1 and D2 renders D2 unable to substitute for the loss of D1 in certain cell types. Cyclin D compound mice The speculation that compensatory mechanisms operate only among one type of cyclins proved wrong as cells lacking two of the three or all three D-type cyclins can


2927

Figure 2 Array of compensations among cyclins. The wide range of intra- and interfamily compensatory mechanisms among different members of cyclins are shown. Under normal circumstances the functions of cyclins are limited to a particular phase of the cell cycle, but in the absence of one set of cyclins, another set of cyclins will no longer restrict their function to one phase but rather operate at multiple phases of the cell cycle.

still divide and progress through different phases of cell cycle and the mice can survive at least until midgestation (Ciemerych et al., 2002; Kozar et al., 2004; Pagano and Jackson, 2004; Figure 3). The cyclin D1/D2 double knockout mice survived up to 3 weeks after birth. Similar to D1 and D2 single knockout mice, they displayed defects such as reduced body size, hypoplastic cerebella and others (Ciemerych et al., 2002; Pogoriler et al., 2006). In contrast, most of the cyclin D1/D3 double mutant mice died immediately after birth due to neurological defects (Ciemerych et al., 2002). D2/D3 double knockout mice died at E17.5 due to severe megaloblastic anemia (Ciemerych et al., 2002; Table 1). It seems that cyclin D3 in combination with D1 or D2 is more important for the survival of the organism, although D3 single knockout mice have relatively fewer abnormalities compared to either D1 or D2. Otherwise, it could be possible that several of D3’s functions are not yet fully explored in the D3 single knockouts. It appears that D3 can partially take over most of the cell-cycle and cell-specific functions of D1 and D2 as the D1/D2 double knockout mice are viable and can survive up to 3 weeks. These single and double cyclin D knockout mouse studies disclosed that although deletion of two D-type cyclins does not effect the cell cycle, at least two of the three D-type cyclins are necessary for complete survival of the organism (Figure 2). Mice lacking all three D-type (D1, D2, D3) cyclins were generated and the triple knockout mice died around E16.5 (Kozar et al., 2004). The majority of the cell types including

mouse embryonic fibroblasts (MEFs) proliferated relatively normally, though MEFs displayed higher dependence on serum than wild-type cells (Kozar et al., 2004). The proliferative defects were mainly limited to hematopoietic cells and myocardial cells, indicating that these cells critically require D-type cyclins (Kozar et al., 2004). These studies suggest that either their function is not essential or other cyclins such as E- or A-type cyclins can drive cell-cycle progression in the absence of D-type cyclins in the majority of cell types. This was directly proven in a study where genetic replacement of cyclin D1 by human cyclin E1 rescued the phenotypes of cyclin D1 knockout mice, indicating that E-type cyclins could be the primary compensators of D-type cyclins (Geng et al., 1999; Table 2). The lack of all three D-type cyclins presents an intricate situation with regard to translocation of Cdk4/ Cdk6 from the cytoplasm to the nucleus. Under normal circumstances cyclin D1 and Cdk4/Cdk6 assemble in the cytoplasm and translocate to the nucleus. The lack of nuclear localization signals (NLS) in both proteins is overcome by a direct interaction with p21 or p27, which contain NLS, and provide a possible mechanism for nuclear import although other translocation mechanisms remain possible (Cheng et al., 1999; Coqueret, 2003). After import to the nucleus, Cdk4-6/cyclin D complexes sequester p27 from resident Cdk2/cyclin E/p27 complexes. The impact of Cdk2/cyclin E activation is the initiation of targeted degradation of p27, due to phosphorylation at threonine 187 whereas p27 bound Oncogene


2928

Figure 3 Consequence of loss of cyclins and Cdks on the survival of mice. The impact of loss of one or more cyclins and Cdks on the survival of the animals is indicated. Solid arrows indicate that the mice are viable and survived similar to wild-type mice. Dotted lines indicate that the deletion of cyclin and Cdks led to embryonic lethality, death of neonates or premature death of adult mice.

to Cdk4-6/cyclin D is stable (Coqueret, 2003). p27 bound to Cdk4-6/cyclin D is released when cyclin D is degraded, and can subsequently bind new complexes. The export of p27 from the nucleus to the cytoplasm for degradation is also tightly regulated and recently it was discovered that cyclin D2 facilitates the translocation of p27 from the nucleus into the cytoplasm for its KPCdependent degradation (Susaki et al., 2007). However, defects were not observed in cell-cycle initiation and progression of D1/D2/D3 triple knockout cells indicating the presence of alternative mechanisms. Recent studies indicate that cyclin D1 might also have a specific cytoplasmic function in transformed cells but its exact functions are still not known (Dhar et al., 1999; Alao et al., 2006). Similarly, cyclin D3 was identified to be present exclusively in the cytoplasm of germ cells although these functions need to be explored further (Kang et al., 1997). E-type cyclins The next class of cyclins is the E-type cyclins and mice deficient of cyclins E1 or E2 are viable and develop normally (Geng et al., 2003), suggesting that there is a functional redundancy between E1 and E2. However, deletion of both E1 and E2 led to embryonic lethality at E11.5 (Figure 3), indicating that at least one E-type cyclin is essential to complete embryogenesis and other cyclins cannot substitute for E-type cyclins during mid and late embryogenesis (Parisi et al., 2003). Analysis of extra embryonic tissues in cyclin E double mutants revealed placental defects and when the placental defects Oncogene

are rescued, the embryos reached birth. Impairment in the endoreplication of megakaryocytes was also observed in the absence of cyclin E (Geng et al., 2003; Parisi et al., 2003). The double mutant cyclin E embryonic fibroblasts displayed defects in cell-cycle reentry from quiescence although the cells proliferate normally in continuous culture (Geng et al., 2003). This might be due to the timing and the expression levels of cyclin E function as a rate limiting step in G1 progression and G1/S transition and overexpression of cyclin E results in shortening of G1 phase and accelerated S phase entry (Ohtsubo et al., 1995). The studies of cyclin E mouse models indicate that E-type cyclins are not essential for cell proliferation and D- and A-type cyclins substitute for the absence of E-type cyclins (Figure 2). A-type cyclins Two A-type cyclins, A1 and A2, have been discovered in mammalian cells (Girard et al., 1991; Zindy et al., 1992; Sweeney et al., 1996; Ravnik and Wolgemuth, 1999) and mouse knockouts of A1 and A2 were generated. Cyclin A1 knockout mice are viable and develop normally. In contrast, deletion of cyclin A2 results in early embryonic lethality and the mice die at E5.5 (Figure 3). Cyclin A2/ embryos did not show any impairment until the preimplantation blastocyst stage but died immediately after implantation (Murphy et al., 1997; Liu et al., 1998). This suggests that cyclin A2 is essential for early embryogenesis and none of the other cyclins are able to compensate for its loss. During early preimplantation it


2929 Table 1

Consequence of deletion of Cdks and cyclins

Gene deleted

Uncompensated phenotypes

Survival/ lethality

Meiotic phenotypes

Primary cellular localization

References

Cdk1

None

BE2.5

Not known

Cytoplasm

Cdk2

Viable

Both males and females are sterile Male and female infertility

Nucleus

Viable

Fertile

Nucleus

Cdk2/4

Reduced body size, impaired neural progenitor cell proliferation Reduced body size, insulin deficient diabetes due to decreased pancreatic b-cells Hypoplasia of thymus and spleen and defects in hematopoiesis Heart defects

(Santamaria et al., 2007) (Berthet et al., 2003; Ortega et al., 2003) (Rane et al., 1999; Tsutsui et al., 1999)

E15.5

Cdk2/6

Reduced body size, hematopoietic defects Viable

Cdk4/6

Severe anemia

E14.5-E18.5

Cdk2/4/6

Heart defects, hematopoietic defects

E13.5

Cdk3 Cdk5

No defects Severe neurological defects

(Malumbres et al., 2004) (Malumbres et al., 2004) (Santamaria et al., 2007) (Ye et al., 2001) (Ohshima et al., 1996)

Cdk11

Mitotic defects

Viable Died immediately after birth E3.5

If viable, might be sterile similar to Cdk2/ ? Both male and female are sterile If viable, might be sterile similar to Cdk4/ ? If viable, might be sterile similar to Cdk2 & Cdk4 Fertile Nucleus Not known Nucleus Not known

(Li et al., 2004)

Cdk4 Cdk6

Cyclin Cyclin Cyclin Cyclin Cyclin

A1 A2 B1 B2 D1

No abnormalities Not-known Not-known No abnormalities Reduced body size, hypoplastic retina, neurological abnormalities, mammary gland defects Cyclin D2 Defects in B-lymphocyte proliferation, pancreatic b-cell proliferation, cerebellar development, adult neurogenesis and hypoplastic thymus Cyclin D3 Defects in T-lymphocyte development Cyclin D1/D2 Reduced body size, hypoplastic cerebella

Viable

Nucleus

Nucleus/ cytoplasm Nucleus Nucleus Cytoplasm Cytoplasm Nucleus

Viable E5.5 BE10.5 Viable Viable

Only males are sterile Not-known Not-known Fertile Fertile

Viable

Females sterile; males fertile with hypoplastic testes

Fertile Nucleus If viable, might be sterile similar to D2/ ?

Cyclin D1/D3 Neurological abnormalities

Viable Viable but die within 3 weeks after birth Die soon after birth

Cyclin D2/D3 Megaloblastic anemia

E17.5

Cyclin D1/ D2/D3 Cyclin E1 Cyclin E2 Cyclin E1/E2 Cyclin F

E16.5

Proliferative defects in hematopoietic cells and cardiac myocytes No abnormalities No abnormalities Severe defects in extraembryonic tissues Defects in extraembryonic tissues

Viable Viable E11.5 E10.5

is possible that cyclin A2 might not be needed or cyclin A1 could partially compensate for cyclin A2. It was also predicted that cyclin B3 might replace cyclin A2, which shares some homology with the A-type cyclins (Winston et al., 2000). On the other hand in cultured somatic cells, cyclin A2 appears to play a nonoverlapping role in S and G2/M phase of cell cycle (Pagano et al., 1992; Resnitzky et al., 1995; Furuno et al., 1999). B-type cyclins B-type cyclins represent essential components of mitosis. Three mammalian B-type cyclins have been identified, B1, B2 and B3 (Pines and Hunter, 1989; Chapman and

Nucleus

Might be fertile? If viable, might be sterile similar to D2/ ? If viable, might be sterile similar to D2/ ? Fertile Nucleus Reduced male fertility Nucleus Not known Not known Nucleus

(Malumbres et al., 2004) (Berthet et al., 2006)

(Liu et al., 1998) (Murphy et al., 1997) (Brandeis et al., 1998) (Brandeis et al., 1998) (Fantl et al., 1995; Sicinski et al., 1995) (Huard et al., 1999; Kowalczyk et al., 2004; Sicinski et al., 1996; Solvason et al., 2000) (Sicinska et al., 2003) (Ciemerych et al., 2002) (Ciemerych et al., 2002) (Ciemerych et al., 2002) (Kozar et al., 2004) (Parisi et al., 2003) (Geng et al., 2003) (Geng et al., 2003) (Tetzlaff et al., 2004)

Wolgemuth, 1992; Gallant and Nigg, 1992; Lozano et al., 2002; Nguyen et al., 2002). Cyclins B1 and B2 are expressed in the majority of proliferating cells and localize to different compartments of the cell. Cyclin B1 is associated with microtubules, whereas cyclin B2 associates with intracellular membranes (Ookata et al., 1993; Jackman et al., 1995). Cyclin B1 localizes to the nucleus in mitosis whereas cyclin B2 does not change its localization, suggesting that the two proteins play different roles during cell-cycle progression (Pines and Hunter, 1991; Toyoshima et al., 1998; Yang et al., 1998). Cyclins B1 and B2 interact with Cdk1 (Draetta et al., 1989; Pines and Hunter, 1989) but can also bind to Cdk2 (Aleem et al., 2005). It was described that Cdk1/cyclin B1 Oncogene


2930 Table 2 Cdk and cyclin knockin mice Knockin gene

Phenotypes rescued (or aggravated)

Survival

Meiotic phenotype

Reference

cyclin D1cyclin E cyclin D1cyclin D2

Rescued all the phenotypes associated with loss of cyclin D1 Mostly rescued the defective retina and mammary gland phenotype, and partially rescued neurological defects observed in cyclin D1 knockout Disruption of cyclin E regulation by Fbw7, increased cyclin E activity. Proliferative abnormalities including abnormal erythropoiesis, dysplasia and anemia. Delayed mammary gland involution Insensitive to INK4 inhibitors. Multiple tumors including endocrine tumors and hemangiosarcomas Early embryonic lethality. Failed to rescue loss of Cdk1

Viable Viable

Fertile Fertile

(Geng et al., 1999) (Carthon et al., 2005)

Viable

Fertile

(Minella et al., 2008)

Viable

Fertile

(Sotillo et al., 2001)

BE2.5

Both male and female are sterile

(Satyanarayana et al., 2008a)

Cyclin E Cdk4R24C Cdk1Cdk2

T74A T393A

complexes promote nuclear envelope breakdown, chromosome condensation, and mitotic spindle assembly. Cytoplasmic Cdk1/cyclin B2 complexes are essential for the mitotic reorganization of the Golgi apparatus (Draviam et al., 2001). Meanwhile it was described that cyclin B3 shares the properties of both A- and B-type cyclins (Gallant and Nigg, 1994; Nieduszynski et al., 2002). Cyclin B3 is mainly localized in the nucleus (Gallant and Nigg, 1994) and associates with Cdk2 (Nguyen et al., 2002). Cyclin B3 expression is restricted to testes and fetal ovaries (Nguyen et al., 2002). To explore the functions of B-type cyclins, B1 and B2 knockout mice were generated. Deletion of cyclin B1 led to embryonic lethality before E10.5 (Brandeis et al., 1998). However, the exact reason for embryonic lethality was not reported. Several gene knockout mice have been reported to die around E10.5 due to severe defects in extra embryonic tissues (Parekh et al., 2004; Argraves and Drake, 2005). In this regard it would be interesting to investigate whether a similar defect exists in the cyclin B1-deficient mice. In contrast to cyclin B1, cyclin B2 knockout mice are viable and did not display any growth abnormalities, suggesting that cyclin B1 fully compensates for the loss of B2 (Figure 2). The precise functions of cyclin B3 in mouse embryogenesis and development is still unclear because cyclin B3 knockout mice have not yet been generated. Of all the different cyclin members (A, B, D, E), cyclin A2 and B1 appear to be the most nonredundant cyclins, as deletion of either A2 (E5.5) or B1 (E10.5) led to embryonic lethality, suggesting an important role in development and/or cellcycle regulation (Figure 3). To dissect the specific functions of cyclin B1 and A2 it will be essential to generate conditional knockout mice to investigate their role during embryogenesis and in adult organs.

Compensatory mechanisms among Cdks In human cell lines, inactivation of Cdk1, 2 or 3 resulted in cell-cycle arrest in one specific phase. For example, expression of Cdk2D145N caused a G1 arrest, whereas Cdk1D146N resulted in a G2/M arrest (van den Heuvel and Harlow, 1993). Similarly, cell lines carrying a Oncogene

temperature-sensitive Cdk1 mutation arrested in G2 at the restrictive temperature (Th’ng et al., 1990; Yasuda et al., 1991). Comparable results were obtained by microinjecting antibodies against Cdks (or their corresponding cyclin partners) or by using chemical Cdk inhibitors. These results suggested that Cdks perform functions that are specific for one phase of the cell cycle. To study the validity of these conclusions in vivo, mice lacking the catalytic partners of cyclins, Cdks, were generated with the expectation of severe phenotypes because Cdks can interact with several cyclins. Therefore, loss of one Cdk would have a similar effect as knocking out a whole family of cyclins. Cdk4/Cdk6 Mice lacking catalytic partners of the D-type cyclins, Cdk4 and Cdk6, are viable and displayed defects in the same tissues and organs as those affected by the lack of D-type cyclins. The body size and size of several organs is decreased in Cdk4/ mice due to a reduced number of cells rather than a decrease in cell size. Cdk4/ MEFs proliferate normally with a 4 h delay in S phase entry from quiescence. This delay was indirectly linked to inhibition of Cdk2 activity due to redistribution of p27. Accordingly, combined deletion of Cdk4 and p27 restored Cdk2 activity and normal S phase entry (Tsutsui et al., 1999). Although Cdk4 is dispensable for proliferation of MEFs, it is essential for the proliferation of certain endocrine cell types. Cdk4/ mice displayed similar defects as observed in cyclin D2/ mice such as insulin-deficient diabetes due to a severe decrease in pancreatic b cells (Rane et al., 1999; Tsutsui et al., 1999). The requirement of Cdk4 activity for b-cell proliferation could be that Cdk6 might not be expressed in this cell type. Accordingly, mutations in negative regulators of Cdk4 activity lead to abnormal b-cell proliferation indicating a specific role of Cdk4 in this cell type (Franklin et al., 2000; Latres et al., 2000). These defects could be due to a cell autonomous requirement of Cdk4 in these cell types as reexpression of endogenous Cdk4 in b cells and in the pituitary gland of Cdk4/ mice restores cell proliferation and normoglycemic condition. However, reexpression of Cdk4 does not rescue the small body size of Cdk4/ mice,


2931

Figure 4 Array of compensations among Cdks. Under normal circumstances the functions of Cdks are limited to a particular phase of the cell cycle but in the absence of one or more Cdks, the remaining Cdks will take over the functions of lost Cdks and faithfully control the cell cycle. For example, in the absence of Cdk2, Cdk4 and Cdk6, Cdk1 alone can drive the cell cycle by complexing with all the phase-specific cyclins.

indicating that this phenotype is not because of endocrine defects. Therefore mammalian Cdk4 is not only involved in controlling proliferation of specific cell types but might also play a crucial role in establishing homeostasis. On the other hand mice lacking Cdk6 do not display any defects in cell-cycle progression but female mice were reduced in size. Consistent with a prescribed role for Cdk6 in blood cell differentiation, these mice displayed defects in hematopoiesis including decreased cellularity of the thymus, red blood cells and lymphocytes (Malumbres et al., 2004). The mild phenotype of this knockout mouse suggests that the functions of Cdk6 are mainly compensated by Cdk4, or possibly by Cdk2 (Figure 4), which was identified to be interacting with D-type cyclins in Cdk4/Cdk6 double knockout mice (Malumbres et al., 2004). Cdk4/Cdk6/ double mutant mice On the basis of these studies it was hypothesized that Cdk4 and Cdk6 could simply compensate for each other (Figure 4). Cdk4 and Cdk6 share 71% amino-acid identity, and both partner with all three D-type cyclins to phosphorylate Rb. Nevertheless, a few differences have been described between Cdk4 and Cdk6 including differences in timing or pattern of expression. Similarly, Cdk4 and Cdk6 also display distinct patterns of subcellular localization. For example in mouse astrocytes, Cdk6 was localized predominately to the cytoplasm whereas Cdk4 localized almost exclusively to the nucleus (Ericson et al., 2003). In T cells, Cdk6 was detected in both the nucleus and cytoplasm with only the nuclear fraction active as Rb kinase (Mahony et al., 1998). Cdk6 was also localized to the ruffling edge of spreading fibroblasts, suggesting a role for Cdk6 in

controlling cell movement (Fahraeus and Lane, 1999). These observed differences in subcellular localization could explain a mechanism for regulating kinase activity, or may be indicative of the presence of a cytoplasmic role of Cdk6. However, Cdk4 and Cdk6 are widely expressed in embryos and their expression is similar in most of the cell types, suggesting that these two kinases might be distinct regarding their substrate specificity. Accordingly it was reported that Cdk4 is more efficient in phosphorylating Rb than Cdk6 and displays different residue selectivity (Takaki et al., 2005). This could explain the divergence in phenotype of corresponding Cdk4- and Cdk6-deficient mice. Moreover, recent findings suggest a new role for Cdk6 in the differentiation of a variety of cell types. This function, which affects the transcription of genes involved in terminal differentiation, is not shared with Cdk4 and could be independent of Rb (Grossel and Hinds, 2006). All these evidences suggest that Cdk4 and Cdk6 may have independent functions in the maintenance of the delicate balance between cellular division and differentiation. In contrast to single knockouts, double deletion of Cdk4 and Cdk6 led to embryonic lethality and mice die during the late stages of embryonic development due to severe anemia (Table 1). Cdk4/Cdk6/ double knockout mice survived a few days longer than triple cyclin D/ mice perhaps due to the ability of D-type cyclins to interact with Cdk2 and/or Cdk1 in the absence of Cdk4 and Cdk6 (Malumbres et al., 2004). DKO embryos displayed normal organogenesis and most cell types proliferated normally. MEFs lacking Cdk4 and Cdk6 proliferate with a lower efficiency but became immortal upon serial passage (Malumbres et al., 2004). Overall, none of the components of the cyclin D-associated complexes are required in early embryogenesis and in cell-cycle progression, suggesting the presence of alternative mechanisms to initiate the cell cycle. Cdk2 The compensation of loss of Cdk4 by Cdk6 or vice versa is perhaps not very surprising because both proteins were identified to play a role in the early G1 phase by binding to the same D-type cyclin partners. In contrast, it was widely anticipated that deletion of Cdk2 could lead to severe phenotypes because it was believed that Cdk2 would be absolutely essential for the G1/S transition of the cell cycle. This hypothesis was mainly based on in vitro studies where inhibition of Cdk2 led to cell-cycle arrest (van den Heuvel and Harlow, 1993). Surprisingly, Cdk2 knockout mice are viable and develop normally with a reduced body size (Berthet et al., 2003; Ortega et al., 2003). Cell proliferation is only slightly affected with a 4 h delay in S phase entry in Cdk2-deficient MEFs released from quiescence. Later it was demonstrated that Cdk1 binds to cyclin E in the absence of Cdk2 and promotes the G1/S transition (Aleem et al., 2005). It appears that Cdk2 is not necessary for cell proliferation but it is not known whether Cdk2 is rate-limiting in certain cell types or at Oncogene


2932

certain stages of development. Recently it has been demonstrated that neural progenitor cells of the subventricular zone displayed impaired proliferation in the absence of Cdk2 in the adult animals although they do not require Cdk2 perinatally. Interestingly, in this cell type, Cdk4, but not Cdk1, compensates for Cdk2. This compensatory mechanism in the absence of Cdk2 persisted only until postnatal day 15 involving increased Cdk4 expression with concomitant Rb inactivation (Jablonska et al., 2007). It appears that the concept derived from MEFs, Cdk1 compensates for Cdk2, might not be universal and there are certain cell types, which still require Cdk2 and have not developed a full spectrum of compensatory mechanisms in response to loss of Cdk2. Cdk2 was also proposed to function as a master regulator of centrosome duplication (Hinchcliffe et al., 1999; Meraldi et al., 1999; Matsumoto and Maller, 2004). Nevertheless, it was described recently that Cdk2 is not required for normal centrosome duplication, maturation and bipolar mitotic spindle formation. In contrast, Cdk2 deficiency completely abrogated aberrant centrosome duplication induced by a viral oncogene. This indicates that normal and aberrant centrosome duplication have significantly different requirements for Cdk2, and it appears that in the absence of Cdk2 other kinases including cyclin A-associated Cdk activity may compensate for the loss of Cdk2 to allow normal centrosome duplication (Duensing et al., 2006). Compensation of Cdk2 functions by Cdk1 Although the compensation of Cdk2 by Cdk1 in the mitotic cell-cycle regulation appears to be a simple mechanism, it left us with additional questions mainly due to the discrepancies in the properties of the two kinases. Cdk2 translocates to the nucleus as soon as it is translated in complex with cyclin E and is present mainly in the nucleus irrespective of the cell-cycle stage. In contrast, Cdk1 localizes mainly to the cytoplasm but is found in the nucleus during mitosis after the breakdown of nuclear lamina (Moore et al., 1999; Satyanarayana et al., 2008b). If Cdk1 has to perform the mitotic cell-cycle functions of Cdk2, Cdk1 needs to be in the nucleus at the G1/S transition. Another important question arising in this context is whether Cdk1 will localize exclusively in the nucleus similar to Cdk2 to perform both G1/S and G2/M transitions, or will still shuttle between cytoplasm and nucleus? A recent study described that Cdk1 localizes to the nucleus earlier in the absence of Cdk2 and mainly resides in the nucleus (Satyanarayana et al., 2008b). The molecular mechanism behind Cdk1’s premature localization is puzzling especially due to the fact that Cdk1 not only binds to cyclin B and cyclin A but also to cyclin E. One possibility could be that a subpopulation of Cdk1 localizes to the nucleus by complexing with cyclin E and this complex functions in the G1/S transition. Later, during mitosis another subset of Cdk1 binds to cyclin B and localizes to the nucleus to execute mitosis. At this point, it can also be speculated that the kinase activity of Oncogene

Cdk2 might be essential to keep Cdk1 in the cytoplasm until mitosis, and in the absence of Cdk2, Cdk1 is free to localize to the nucleus prematurely. Although this is unlikely due to the fact that to date no NLS was identified in Cdk1 and it appears that Cdk1 localizes to the nucleus only after complexing with its cyclin partners. Moreover due to earlier localization of Cdk1 to the nucleus, cells could face catastrophic consequences because this could lead to premature initiation of mitosis. Interestingly, premature initiation of mitosis was not observed although Cdk1 is mainly present in the nucleus in the absence of Cdk2 (Satyanarayana et al., 2008b). The molecular mechanism behind prevention of premature initiation of mitosis is still unknown. One possibility is that the Cdk1/cyclin E complexes are perhaps unable to phosphorylate the substrates that are essential for initiation of mitosis. Alternatively even though Cdk1 is physically present in the nucleus, initiation components for mitosis may not yet be available to be phosphoylated by Cdk1. Cdk2/Cdk4/ double knockout mice In contrast to Cdk2 deletion, simultaneous deletion of Cdk2/Cdk4 leads to embryonic lethality and embryos die between E14.5 and E16.5 due to heart defects (Figure 3). Compound mutants of Cdk4 and Cdk2 leads to an impairment of cell-cycle progression due to a progressive decline of Rb phosphorylation (Berthet et al., 2006). This suggests that Cdk6 and Cdk1 can regulate cell proliferation, but these two kinases might not phosphorylate Rb to full extent, leading to repression of E2Fs and resulting eventually in decreased Cdk1 expression. The declining Cdk1 expression might act as a negative loop leading to proliferation defects. The fact that more Cdk6 is bound to cyclin D1, in the absence of Cdk4, is apparently not sufficient to compensate for the lack of Cdk2 and Cdk4. In contrast, when Cdk6 is deleted together with Cdk2, Cdk2/Cdk6/ mice are viable and exhibit similar phenotypes that were observed in respective single knockouts (Malumbres et al., 2004), and have no impact on cell proliferation. On the basis of these studies, the idea of a minimal-essential Cdk in mammalian cells, like in yeast, started to emerge. Recently, it was described that MEFs lacking all three interphase Cdks (Cdk2, Cdk4 and Cdk6) can progress through different phases of cell-cycle with a delay but complete cell division (Santamaria et al., 2007) (Figure 5). No apparent changes in the expression of Cdk1, Cdk7, Cdk9 as well as cyclins D1, D2, E1, A2 and B1 were detected in the triple knockout cells. In contrast, a reduction in the p27 levels was detected (Santamaria et al., 2007). In these cells, Cdk1 binds to all cyclins resulting in phosphorylation of the retinoblastoma protein Rb. However, they become immortal on continuous passage when grown at low oxygen. The triple knockout embryos survived up to E13.5 and died due to heart and severe hematopoietic defects (Santamaria et al., 2007). Even though it appears that Cdk1 can drive cell-cycle progression alone by binding to all phase-specific cyclins, it cannot be excluded the


2933

Figure 5 Mammalian cell-cycle regulation Now and Then. According to classical model of cell-cycle regulation Cdk4, Cdk6, Cdk2 and Cdk1 complex with phase-specific cyclins at different stages of cell cycle to coordinate initiation and progression of cell cycle. Recent evidences suggest that most of the Cdks and cyclins are dispensable for mammalian cell-cycle regulation and Cdk1 alone by complexing with minimum number of cyclins can drive cells through different phases of cell cycle. Although it appears that Cdk1 alone can drive the cell-cycle regulation it cannot be excluded that other Cdks such as Cdk3, Cdk7 and Cdk11 might participate more actively in cell-cycle regulation in the absence of Cdk2, Cdk4 and Cdk6.

participation of other Cdk/cyclin complexes such as Cdk3/cyclin C, Cdk7/cyclin H or Cdk11/cyclin L. In the absence of Cdk2, Cdk4 and Cdk6 there is a possibility that these Cdk/cyclin complexes might play more active roles (Figure 5). On the basis of the above result, one could expect that disruption of the Cdk1 gene could lead to early embryonic lethality. In accordance with this hypothesis, it was observed that deletion of Cdk1 (our unpublished results) or a gene trap mutation in the Cdk1 gene (Santamaria et al., 2007) led to early embryonic lethality and the embryos fail to reach even the blastocyst stage in the absence of Cdk1. These results indicate that Cdk1 can drive the mammalian cell cycle alone similar to yeast Cdk1 (Figure 4). Replacement of Cdk2 by Cdk1 So far all the studies were focused on identifying compensatory mechanisms for the loss of Cdk2, Cdk4 and Cdk6. It was not explored if any of the Cdks can compensate for the loss of Cdk1. However, it was described that deletion of Cdk1 leads to early embryonic lethality (Santamaria et al., 2007; our unpublished results), and no other Cdk, including Cdk2, can compensate for the loss of Cdk1 under normal conditions. However, the inability of Cdk2 to take over the functions of Cdk1 could be due to (1) the differences in the timing of expression of Cdk1 and Cdk2 during different phases of cell cycle, (2) substrate specificity or (3) subcellular localization. It is interesting to explore whether Cdk2 acquires some of the properties of Cdk1, if Cdk2 is expressed directly from the Cdk1 locus. Therefore, we generated a knockin mouse in which Cdk2 was knocked into the Cdk1 locus (Cdk1Cdk2KI) and studied whether Cdk2 can rescue the loss of Cdk1 in vivo. Substitution of both the

copies of Cdk1 by Cdk2 leads to early embryonic lethality similar to deletion of Cdk1 (Satyanarayana et al., 2008a; Table 2). It indicates that Cdk2 cannot substitute for Cdk1 functions even when it is expressed directly from the Cdk1 locus. This result suggests that there are intrinsic differences between Cdk1 and Cdk2 despite the fact that they bind to the same cyclins and inhibitors. Although differences in substrate specificity cannot be excluded, most likely Cdk2 cannot compensate because of its nuclear localization compared to the cytoplasmic localization of Cdk1. One could speculate that Cdk1 fulfills essential functions in the cytoplasm, which would be quite contrary to the current view. Future experiments will have to explore this possibility. At the same time, we also analysed how efficiently Cdk2 expressed from the Cdk1 locus could perform its meiotic function in germ cells. Analysis of Cdk2/ Cdk1 þ /Cdk2KI mice where one copy of Cdk2 and one copy of Cdk1 are expressed from the Cdk1 locus and the Cdk2 gene was deleted in the endogenous Cdk2 locus, revealed that both Cdk2/Cdk1 þ /Cdk2KI male and female mice are sterile, similar to Cdk2/ mice, although they express functional Cdk2 protein from Cdk1 locus. This indicates that the genetic relocation of Cdk2 from the Cdk2 to the Cdk1 locus abolished Cdk2’s own meiotic function (Satyanarayana et al., 2008a). It is still puzzling why Cdk2 is unable to perform its meiotic function when expressed from Cdk1 locus. One possibility could be that the timing of Cdk2 expression might be crucial to participate fully in meiosis. The studies in the knockout mouse models of cellcycle regulators clearly produced contrasting results compared to in vitro studies, where inhibition of Cdks by drugs, antibodies and RNAi led to dramatic effects on cell-cycle progression. Several possible reasons could explain why the wide range of compensatory mechanisms observed in vivo was not evident in vitro. One reason could be that when Cdks are inhibited in vitro, they do not make their cyclin subunits available for other Cdks to compensate. In contrast, in in vivo models, the complete absence of specific Cdks, cyclin subunits become available to form novel, compensatory Cdk complexes. In line with this hypothesis a recent study illustrates that acute elimination of a protein’s function in somatic cells may have consequences different than in the mouse knockout setting, where there may be compensation by related proteins during development (Sage et al., 2003). On the basis of these results it appears that both in vitro and in vivo systems have their own advantages. The in vivo systems help us to understand the effect of complete loss of a protein and the concomitant compensatory mechanisms during development and disease. On the other hand the in vitro systems, where the acute inhibition of proteins and the resulting effects on the function of the proteins, help us to develop drugs for different diseases because some of the drugs are designed to inhibit a particular protein or a group of proteins for a limited amount of time. Oncogene


2934

Cdks and cyclins in meiosis The specific functions of cyclins and Cdks in meiosis have become more apparent with the generation of knockout mouse models. Although Cdk and cyclin knockout mouse models were extensively used to study embryogenesis, development and mitotic cell-cycle regulation, much less work was done in these mouse models to understand the role of individual cyclins and Cdks in meiosis. These studies revealed that in contrast to the effect of loss of individual cyclins and Cdks on mitotic cell cycle, meiosis was much more affected. Cyclin D1 and D3 single knockout mice are fertile and no detectable meiotic defects were identified in both males and females. In contrast, cyclin D2-deficient females are sterile (Kozar et al., 2004). When subjected to superovulation, D2/ females failed to release oocytes, suggesting ovarian failure as the cause of infertility. Although the number of ovarian follicles and oocytes was normal in D2/ females, the number of somatic granulosa cells surrounding each oocyte was reduced, indicating defective proliferation of ovarian granulosa cells. Why do D1 and D3 fail to compensate for D2 function in these cell types is not clearly understood. In contrast, the D2/ males are fertile though they showed hypoplastic testes (Sicinski et al., 1996). Cyclin E1-deficient male and female mice are fertile but E2 knockout males displayed testicular abnormalities, testicular atrophy and reduced fertility whereas cyclin E2/ females are normal and fertile (Geng et al., 2003). Testes of E2/ males displayed thinned walls of the seminiferous tubules with a reduction in spermatogenetic cells. Abnormal meiotic figures in the spermatocyte layers and multinuclear giant cells in seminiferous epithelium were also observed. This clearly shows that E1 and E2 have nonredundant functions in meiosis unlike mitosis. Similar to the E2-deficient females, cyclin A1 females are fertile but the males are completely sterile with spermatogenesis arresting before the first meiotic division in males. Spermatocytes progress normally to the transition from prophase to metaphase but stop just before metaphase I. This arrest leads to extensive germ cell apoptosis and sterility. Cyclin A1-deficient spermatocytes displayed impaired Cdk1 activation at the end of the meiotic prophase. As a result, the first meiotic division failed to be induced (Liu et al., 1998). However, cyclin B2 knockout males and females are fertile, suggesting that there is no obvious meiotic defect (Brandeis et al., 1998). These studies disclosed that cyclin D2, E2 and A1 have specific functions in meiosis. Whether cyclin A2 and B1 have any function in meiosis is still unclear because deletion of either of the cyclins leads to early embryonic lethality (Table 1). The specific role of cyclin partners, Cdks, was also studied in meiosis in the Cdk-deficient mice. Cdk4/ females are sterile due to a decrease in prolactin producing pituitary lactotrophs (Rane et al., 1999; Tsutsui et al., 1999). These mice also display decreased male fertility due to defective spermatogenesis and lower numbers of Leydig cells. These defects are mainly Oncogene

because of a cell autonomous requirement of Cdk4 in these cell types as reexpression of Cdk4 restores cell proliferation and results in fertile animals. In contrast to Cdk4/ mice, Cdk6 male and female mice are fertile and no meiotic defects were reported (Malumbres et al., 2004). Although Cdk2 was identified to be dispensable for mitotic cell cycle, it was discovered that Cdk2 is essential for meiosis and both male and female Cdk2/ mice are sterile with 100% penetrance (Berthet et al., 2003; Ortega et al., 2003). The germ cells of males do not progress beyond the pachytene stage of prophase I. At this stage, primary spermatocytes undergo apoptosis due to a consequence of defective synaptonemal complex formation (Ortega et al., 2003). However, oocytes progress normally through pachytene, engage in normal synapsis, progress further to the late diplotene or dictyate stage and die by apoptosis. The meiotic phenotype of Cdk2/ mice suggests that Cdk2 is essential for germ cell development beyond meiotic prophase I in both males and females and Cdk1 cannot compensate for Cdk2 in germ cells. The assumption that Cdk1 cannot compensate for Cdk2’s meiotic function is further strengthened by a recent study, where genetic substitution of Cdk1 by Cdk2 leads to the loss of Cdk2’s own meiotic function (Satyanarayana et al., 2008a). Both male and female Cdk2/Cdk1 þ /Cdk2KI mice are sterile, similar to Cdk2/ mice, even though there is abundant expression of knockin Cdk2 from the Cdk1 locus in the testes. In adult Cdk2/Cdk1 þ /Cdk2KI mice, testes and ovaries are atrophic. The seminiferous tubules are smaller than normal with a substantial depletion of germ cells, similar to Cdk2/ mice. The atrophic testes lacked round spermatids and displayed extensive apoptosis of germ cells, indicating that although knockin Cdk2 was expressed, it was unable to complete the pachytene stage and instead the cells underwent apoptosis (Satyanarayana et al., 2008a). It appears that the timing of expression of Cdk2 is crucial for its meiotic function(s). These results suggest the existence of a certain time window for the requirement of Cdk2. When Cdk2 is not expressed at that particular time point, the cells fail to complete meiosis even though Cdk2 is expressed thereafter. This indicates that the genetic locus and timing of Cdk2 expression determine the meiotic functions of Cdk2. Whether Cdk1 has any specific role in meiosis is still not known due to early embryonic lethality of Cdk1 knockout mouse. Specific deletion of Cdk1 in germ cells of mice will disclose the functions of Cdk1 in meiosis in future studies.

Other Cdks and cyclins Although the Cdks and cyclins discussed above attracted major attention due to their crucial roles in cell-cycle regulation, several other Cdks and cyclins have also been discovered. These include Cdk3, Cdk5, Cdk7, Cdk8, Cdk9, Cdk10, Cdk11 and cyclins C, F, G, H, K, L, T1, T2a, T2b. These Cdk/cyclin complexes have diverse cellular functions, including cell-cycle regulation,


2935

neuronal development and regulation of transcription. The functions of Cdk8/cyclin E and Cdk9/cyclin T, Cdk9/cyclin K and Cdk10 in transcriptional regulation were extensively reviewed (Akoulitchev et al., 2000; Sano and Schneider, 2003; Fisher, 2005; Wang and Fischer, 2008). Cdk5 was identified to have minor roles in cell-cycle regulation and functions mostly in neuronal development (Dhavan and Tsai, 2001). Cdk5 is ubiquitously expressed in neurons and several lines of evidence indicate that Cdk5 contributes to normal brain development where it plays a role in axon guidance and neuronal migration (Dhavan et al., 2002). Cdk5 was identified to play an important role in neurotransmission and synaptic plasticity and thus functions in cognitive functions such as learning and memory (Fischer et al., 2005; Cheung et al., 2006). Aberrant Cdk5 signaling has been proposed to be a common contributor to a variety of central nervous system disorders (Cruz and Tsai, 2004; Qu et al., 2007). As a result of Cdk5’s crucial role in neuronal development, Cdk5/ mice exhibited severe defects in the central nervous system and 60% of the mice died in utero and the remaining 40% died immediately after birth (Ohshima et al., 1996). The brain defects in Cdk5/ mice include lack of cortical laminar structure and cerebellar foliation. Moreover, the large neurons in the brain stem and in the spinal cord displayed chromatolytic changes, indicating that Cdk5 is important for brain development and neuronal differentiation and might also play critical roles in neuronal cytoskeleton structure and organization (Ohshima et al., 1996).

(CAK) complex, responsible for the activating phosphorylation of Cdk1, Cdk2, Cdk4 and Cdk6, which is essential for cell-cycle progression (Fisher and Morgan, 1994; Fisher, 2005). CAK is also part of the general transcription factor TFIIH (Feaver et al., 1994; Fisher and Morgan, 1996), and CAK phosphorylates the RNA polymerase II large subunit C-terminal domain (CTD) (Lu et al., 1992). CTD phosphorylation by TFIIH is involved in promoter clearance and in progression of transcription from the preinitiation to the initiation stage (Goodrich and Tjian, 1994). Genetic studies in Drosophila, where inactivation of a temperature sensitive Cdk7 mutant (cdk7P140S) prevented activation of Cdk1 by T-loop phosphorylation and caused embryonic and larval lethality by blocking mitosis. In contrast, adult flies bearing the temperature-sensitive allele are viable and are phenotypically nearly normal at the restrictive temperature, indicating that no impairment of transcription has occurred (Larochelle et al., 1998). This suggests that Cdk7 may not be crucial for the TFIIH function. Whether deletion of mammalian Cdk7 can have similar effects or any of the other Cdks substitute for Cdk7 in its absence is not known because Cdk7 knockout mice have not been generated. Nevertheless, disruption of the murine Mat1 gene leads to periimplantation lethality (Rossi et al., 2001). Mat1/ trophoblast giant cells exhibited a rapid arrest in endoreduplication, which was characterized by an inability to enter S phase (Rossi et al., 2001). Whether deletion of Cdk7 leads to a similar phenotype needs to be resolved.

Cdk3 Cdk3/cyclin C, Cdk7/cyclin H and Cdk11/cyclin L appear to function in cell-cycle regulation. In G0 cells, in response to growth factors, Cdk3/cyclin C complexes become activated and phosphorylate Rb and also the other Rb family members p107 and p130. Cdk3/cyclin C-dependent phosphorylation of Rb family proteins is completed in early and late G1 by Cdk4/cyclin D and Cdk2/cyclin E complexes (Ren and Rollins, 2004). In support to the above studies it was observed that dominant-negative forms of human Cdk3 block cells in G1 phase, and this arrest is not overcome by overexpression of Cdk2 (van den Heuvel and Harlow, 1993). Furthermore, Cdk3/cyclin E complexes can promote S phase entry in quiescent cells as efficiently as Cdk2/ cyclin E complexes (Sage, 2004). However, most mouse laboratory strains harbor a point mutation in the Cdk3 gene that converts a conserved tryptophan (Trp187) to a premature termination codon but did not display any defects during development or cell-cycle kinetics indicating a redundancy with other Cdk family members such as Cdk2 and Cdk4 (Ye et al., 2001).

Cdk11 Another Cdk which has a dual role in cell proliferation and transcriptional control is Cdk11. The mouse Cdk11 is present in two different isoforms, Cdk11p110 and Cdk11p58, expressed from a single gene locus (cdc2l). Cdk11p110 associates with RNAP II, cyclin L, premRNA splicing proteins RNPS1 and 9G8, several transcriptional elongation factors and is mainly involved in regulation of RNA transcription and processing (Loyer et al., 1998; Hu et al., 2003). On the other hand, Cdk11p58 is mitosis specific and expressed only during mitosis (Sachs, 2000). Cdk11p110/p58/ mice died early during embryogenesis (E3.5). The blastocysts die between E3.5 and E4.0 due to apoptosis as a result of proliferative defects and mitotic arrest. It appears that Cdk11 is essential for proper cellular proliferation and cell-cycle progression during early embryonic development (Li et al., 2004). It can though not be excluded that in addition to cell-cycle defects, Cdk11p110/p58/ mice also have extensive defects in transcriptional regulation.

Cdk7 Cdk7 was identified to have a dual role and functions in cell-cycle regulation as well as transcriptional control. Cdk7 together with cyclin H and the assembly factor MAT1 (meńage a trios) forms the Cdk activating kinase

Cyclin F In addition to the above discussed Cdks and cyclins, another cyclin which also function in cell-cycle regulation is cyclin F. Cyclin F very closely resembles cyclin A in amino-acid sequence. Similar to other cyclins, the expression of cyclin F strictly occurs in a cell-cycleregulated fashion. Its expression begins in S phase, Oncogene


2936

peaks in G2 and disappears as cells enter mitosis (Bai et al., 1994). The fact that cyclin F also contains an F box and can form an SCF (Skp1-Cul1/Cdc53-F-box) complex in vivo further suggests that it may also function in proteolysis (Bai et al., 1996). Cyclin F/ mice displayed several developmental abnormalities due to defects in yolk sac and chorioallantoic placentation and die around E10.5. Conditional deletion of cyclin F in several specific tissues revealed that it was not required for development and function of a number of different embryonic and adult tissues. In contrast, cyclin F/ MEFs exhibited cell-cycle defects, including reduced population doubling time and a delay in cellcycle reentry from quiescence, indicating that cyclin F plays a role in cell-cycle regulation (Tetzlaff et al., 2004).

Concluding remarks Through the characterization of knockout mouse models of cell-cycle regulators, our knowledge of cellcycle regulation has greatly improved. Extensive analyses of these mouse models revealed that most of the Cdks and cyclins, originally thought essential for cellcycle regulation, are in fact largely dispensable. Cyclin A2 and B1 emerged as the most nonredundant cyclins whereas Cdk1 emerged as the master regulator of mammalian cell-cycle regulation and Cdk1 alone by complexing with all the cyclins can drive the cell cycle similar than in yeast. Other cyclins and Cdks have specific functions in certain cell types and these cells have not developed a full spectrum of compensatory mechanisms. Although the finding that Cdk2 is dispensable for cell-cycle progression is astonishing, the finding that Cdk2 loses its own meiotic function when expressed from Cdk1 locus is even more puzzling. It appears that Cdk2 and its regulation is vital for meiosis. However, the functions of Cdk1 in the mammalian cell-

cycle regulation and development were mainly revealed from other Cdk knockout mouse models. Cdk1 deletion leads to early embryonic lethality (BE2); as a result, the specific functions of Cdk1 in mouse development have not been fully studied. Generation of Cdk1 conditional knockout mice will allow us to explore several important functions of Cdk1. Could the deletion of Cdk1 in conditional Cdk1 MEFs lead to S phase, G2/M, general cell-cycle arrest (a senescence-like phenotype) and/or apoptosis? So far it appears that Cdk2 is essential in meiosis. However, by specifically deleting Cdk1 in germ cells, the functions of Cdk1 in meiosis could be identified. By specifically deleting Cdk1 in certain cell types by specific Cre crossings, it is possible to identify cell types, if any, that operate Cdk1-independent cellcycle regulation. Similarly, by crossing Cdk1 conditional knockout mice with transgenic mouse tumor models and inducible Cre lines, it is possible to study the effect of Cdk1 deletion on tumor initiation and progression. This would provide valuable information for the development and testing of Cdk1 inhibitors in human cancers. The most important question is whether the information derived from these studies is sufficient to specifically target cancer therapy. Would inhibition of Cdk1 alone be sufficient to prevent cancerous growth?

Conflict of interest The authors declare no conflict of interest. Acknowledgements We thank Kaldis lab members for discussions and support. This research was supported by A*STAR of Singapore (PK) and the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research (AS).

References Akoulitchev S, Chuikov S, Reinberg D. (2000). TFIIH is negatively regulated by cdk8-containing mediator complexes. Nature 407: 102–106. Alao JP, Gamble SC, Stavropoulou AV, Pomeranz KM, Lam EW, Coombes RC et al. (2006). The cyclin D1 proto-oncogene is sequestered in the cytoplasm of mammalian cancer cell lines. Mol Cancer 5: 7. Aleem E, Kiyokawa H, Kaldis P. (2005). Cdc2-cyclin E complexes regulate the G1/S phase transition. Nat Cell Biol 7: 831–836. Aprelikova O, Xiong Y, Liu ET. (1995). Both p16 and p21 families of cyclin-dependent kinase (CDK) inhibitors block the phosphorylation of cyclin-dependent kinases by the CDK-activating kinase. J Biol Chem 270: 18195–18197. Argraves WS, Drake CJ. (2005). Genes critical to vasculogenesis as defined by systematic analysis of vascular defects in knockout mice. Anat Rec A Discov Mol Cell Evol Biol 286: 875–884. Bai C, Richman R, Elledge SJ. (1994). Human cyclin F. EMBO J 13: 6087–6098. Bai C, Sen P, Hofmann K, Ma L, Goebl M, Harper JW et al. (1996). SKP1 connects cell cycle regulators to the ubiquitin proteolysis machinery through a novel motif, the F-box. Cell 86: 263–274. Oncogene

Berthet C, Aleem E, Coppola V, Tessarollo L, Kaldis P. (2003). Cdk2 knockout mice are viable. Curr Biol 13: 1775–1785. Berthet C, Klarmann KD, Hilton MB, Suh HC, Keller JR, Kiyokawa H et al. (2006). Combined loss of Cdk2 and Cdk4 results in embryonic lethality and Rb hypophosphorylation. Dev Cell 10: 563–573. Brandeis M, Rosewell I, Carrington M, Crompton T, Jacobs AM, Kirk J et al. (1998). Cyclin B2-null mice develop normally and are fertile whereas cyclin B1-null mice die in utero. Proc Natl Acad Sci USA 95: 4344–4349. Carthon BC, Neumann CA, Das M, Pawlyk B, Li T, Geng Y et al. (2005). Genetic replacement of cyclin D1 function in mouse development by cyclin D2. Mol Cell Biol 25: 1081–1088. Chapman DL, Wolgemuth DJ. (1992). Identification of a mouse B-type cyclin which exhibits developmentally regulated expression in the germ line. Mol Reprod Dev 33: 259–269. Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM et al. (1999). The p21Cip1 and p27Kip1 CDK ‘inhibitors’ are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 18: 1571–1583.


2937 Cheung ZH, Fu AK, Ip NY. (2006). Synaptic roles of Cdk5: implications in higher cognitive functions and neurodegenerative diseases. Neuron 50: 13–18. Ciemerych MA, Kenney AM, Sicinska E, Kalaszczynska I, Bronson RT, Rowitch DH et al. (2002). Development of mice expressing a single D-type cyclin. Genes Dev 16: 3277–3289. Coqueret O. (2003). New roles for p21 and p27 cell-cycle inhibitors: a function for each cell compartment? Trends Cell Biol 13: 65–70. Coverley D, Pelizon C, Trewick S, Laseky RA. (2000). Chromatinbound Cdc6 persists in S and G2 phases in human cells, while soluble Cdc6 is destroyed in a cyclin A-cdk2 dependent process. J Cell Sci 113: 1929–1938. Cruz JC, Tsai LH. (2004). Cdk5 deregulation in the pathogenesis of Alzheimer’s disease. Trends Mol Med 10: 452–458. Dhar KK, Branigan K, Parkes J, Howells RE, Hand P, Musgrove C et al. (1999). Expression and subcellular localization of cyclin D1 protein in epithelial ovarian tumour cells. Br J Cancer 81: 1174–1181. Dhavan R, Greer PL, Morabito MA, Orlando LR, Tsai LH. (2002). The cyclin-dependent kinase 5 activators p35 and p39 interact with the alpha-subunit of Ca2+/calmodulin-dependent protein kinase II and alpha-actinin-1 in a calcium-dependent manner. J Neurosci 22: 7879–7891. Dhavan R, Tsai LH. (2001). A decade of CDK5. Nat Rev Mol Cell Biol 2: 749–759. Draetta G, Luca F, Westendorf J, Brizuela L, Ruderman J, Beach D. (1989). Cdc2 protein kinase is complexed with both cyclin A and B: evidence for proteolytic inactivation of MPF. Cell 56: 829–838. Draviam VM, Orrechia S, Lowe M, Pardi R, Pines J. (2001). The localization of human cyclins B1 and B2 determines CDK1 substrate specificity and neither enzyme requires MEK to disassemble the Golgi apparatus. J Cell Biol 152: 945–958. Duensing A, Liu Y, Tseng M, Malumbres M, Barbacid M, Duensing S. (2006). Cyclin-dependent kinase 2 is dispensable for normal centrosome duplication but required for oncogene-induced centrosome overduplication. Oncogene 25: 2943–2949. Dyson N. (1998). The regulation of E2F by pRB-family proteins. Genes Dev 12: 2245–2262. Ericson KK, Krull D, Slomiany P, Grossel MJ. (2003). Expression of cyclin-dependent kinase 6, but not cyclin-dependent kinase 4, alters morphology of cultured mouse astrocytes. Mol Cancer Res 1: 654–664. Fahraeus R, Lane DP. (1999). The p16INK4a tumour suppressor protein inhibits alphavbeta3 integrin-mediated cell spreading on vitronectin by blocking PKC-dependent localization of alphavbeta3 to focal contacts. EMBO J 18: 2106–2118. Fantl V, Stamp G, Andrews A, Rosewell I, Dickson C. (1995). Mice lacking cyclin D1 are small and show defects in eye and mammary gland development. Genes Dev 9: 2364–2372. Feaver WJ, Svejstrup JQ, Henry NL, Kornberg RD. (1994). Relationship of CDK-activating kinase and RNA polymerase II CTD kinase TFIIH/TFIIK. Cell 79: 1103–1109. Fischer A, Sananbenesi F, Pang PT, Lu B, Tsai LH. (2005). Opposing roles of transient and prolonged expression of p25 in synaptic plasticity and hippocampus-dependent memory. Neuron 48: 825–838. Fisher RP. (2005). Secrets of a double agent: CDK7 in cell-cycle control and transcription. J Cell Sci 118: 5171–5180. Fisher RP, Morgan DO. (1994). A novel cyclin associates with MO15/ CDK7 to form the CDK-activating kinase. Cell 78: 713–724. Fisher RP, Morgan DO. (1996). CAK in TFIIH: crucial connection or confounding coincidence? Biochim Biophys Acta 1288: O7–O10. Franklin DS, Godfrey VL, O0 Brien DA, Deng C, Xiong Y. (2000). Functional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol Cell Biol 20: 6147–6158. Furuno N, den Elzen N, Pines J. (1999). Human cyclin A is required for mitosis until mid prophase. J Cell Biol 147: 295–306. Gallant P, Nigg EA. (1992). Cyclin B2 undergoes cell cycle-dependent nuclear translocation and, when expressed as a non-destructible mutant, causes mitotic arrest in HeLa cells. J Cell Biol 117: 213–224.

Gallant P, Nigg EA. (1994). Identification of a novel vertebrate cyclin: cyclin B3 shares properties with both A- and B-type cyclins. EMBO J 13: 595–605. Geng Y, Whoriskey W, Park MY, Bronson RT, Medema RH, Li T et al. (1999). Rescue of cyclin D1 deficiency by knockin cyclin E. Cell 97: 767–777. Geng Y, Yu Q, Sicinska E, Das M, Schneider JE, Bhattacharya S et al. 2003)). Cyclin E ablation in the mouse. Cell 114: 431–443. Girard F, Strausfeld U, Fernandez A, Lamb NJ. (1991). Cyclin A is required for the onset of DNA replication in mammalian fibroblasts. Cell 67: 1169–1179. Goodrich JA, Tjian R. (1994). Transcription factors IIE and IIH and ATP hydrolysis direct promoter clearance by RNA polymerase II. Cell 77: 145–156. Grossel MJ, Hinds PW. (2006). Beyond the cell cycle: a new role for Cdk6 in differentiation. J Cell Biochem 97: 485–493. Hinchcliffe EH, Li C, Thompson EA, Maller JL, Sluder G. (1999). Requirement of Cdk2-cyclin E activity for repeated centrosome reproduction in Xenopus egg extracts. Science 283: 851–854. Hu D, Mayeda A, Trembley JH, Lahti JM, Kidd VJ. (2003). CDK11 complexes promote pre-mRNA splicing. J Biol Chem 278: 8623–8629. Huang D, Friesen H, Andrews B. (2007). Pho85, a multifunctional cyclin-dependent protein kinase in budding yeast. Mol Microbiol 66: 303–314. Huard JM, Forster CC, Carter ML, Sicinski P, Ross ME. (1999). Cerebellar histogenesis is disturbed in mice lacking cyclin D2. Development 126: 1927–1935. Jablonska B, Aguirre A, Vandenbosch R, Belachew S, Berthet C, Kaldis P et al. (2007). Cdk2 is critical for proliferation and selfrenewal of neural progenitor cells in the adult subventricular zone. J Cell Biol 179: 1231–1245. Jackman M, Firth M, Pines J. (1995). Human cyclins B1 and B2 are localized to strikingly different structures: B1 to microtubules, B2 primarily to the Golgi apparatus. EMBO J 14: 1646–1654. Kang MJ, Kim MK, Terhune A, Park JK, Kim YH, Koh GY. (1997). Cytoplasmic localization of cyclin D3 in seminiferous tubules during testicular development. Exp Cell Res 234: 27–36. Kowalczyk A, Filipkowski RK, Rylski M, Wilczynski GM, Konopacki FA, Jaworski J et al. (2004). The critical role of cyclin D2 in adult neurogenesis. J Cell Biol 167: 209–213. Kozar K, Ciemerych MA, Rebel VI, Shigematsu H, Zagozdzon A, Sicinska E et al. (2004). Mouse development and cell proliferation in the absence of D-cyclins. Cell 118: 477–491. Larochelle S, Pandur J, Fisher RP, Salz HK, Suter B. (1998). Cdk7 is essential for mitosis and for in vivo Cdk-activating kinase activity. Genes Dev 12: 370–381. Latres E, Malumbres M, Sotillo R, Martin J, Ortega S, MartinCaballero J et al. (2000). Limited overlapping roles of P15INK4b and P18INK4c cell cycle inhibitors in proliferation and tumorigenesis. EMBO J 19: 3496–3506. Li T, Inoue A, Lahti JM, Kidd VJ. (2004). Failure to proliferate and mitotic arrest of CDK11(p110/p58)-null mutant mice at the blastocyst stage of embryonic cell development. Mol Cell Biol 24: 3188–3197. Liu D, Matzuk MM, Sung WK, Guo Q, Wang P, Wolgemuth DJ. (1998). Cyclin A1 is required for meiosis in the male mouse. Nat Genet 20: 377–380. Loyer P, Trembley JH, Lathi JM, Kidd VJ. (1998). The RNP protein, RNPS1, associates with specific isoforms of the p34cdc2-related PITSLRE protein kinase in vivo. J Cell Sci 111: 1495–1506. Lozano JC, Perret E, Schatt P, Arnould C, Peaucellier G, Picard A. (2002). Molecular cloning, gene localization, and structure of human cyclin B3. Biochem Biophys Res Commun 291: 406–413. Lu H, Zawel L, Fisher L, Egly JM, Reinberg D. (1992). Human general transcription factor IIH phosphorylates the C-terminal domain of RNA polymerase II. Nature 358: 641–645. Lundberg AS, Weinberg RA. (1998). Functional inactivation of the retinoblastoma protein requires sequential modification by at least two distinct cyclin-cdk complexes. Mol Cell Biol 18: 753–761. Oncogene


2938 Mahony D, Parry D, Lees E. (1998). Active cdk6 complexes are predominantly nuclear and represent only a minority of the cdk6 in T cells. Oncogene 16: 603–611. Malumbres M, Sotillo R, Santamaria D, Galan J, Cerezo A, Ortega S et al. (2004). Mammalian cells cycle without the D-type cyclindependent kinases Cdk4 and Cdk6. Cell 118: 493–504. Matsumoto Y, Maller JL. (2004). A centrosomal localization signal in cyclin E required for Cdk2-independent S phase entry. Science 306: 885–888. Meraldi P, Lukas J, Fry AM, Bartek J, Nigg EA. (1999). Centrosome duplication in mammalian somatic cells requires E2F and Cdk2cyclin A. Nat Cell Biol 1: 88–93. Minella AC, Loeb KR, Knecht A, Welcker M, Varnum-Finney BJ, Bernstein ID et al. (2008). Cyclin E phosphorylation regulates cell proliferation in hematopoietic and epithelial lineages in vivo. Genes Dev 22: 1677–1689. Moore JD, Yang J, Truant R, Kornbluth S. (1999). Nuclear import of Cdk/cyclin complexes: identification of distinct mechanisms for import of Cdk2/cyclin E and Cdc2/cyclin B1. J Cell Biol 144: 213–224. Morgan DO. (1997). Cyclin-dependent kinases: engines, clocks, and microprocessors. Annu Rev Cell Dev Biol 13: 261–291. Murphy M, Stinnakre M-G, Senamaud-Beaufort C, Winston NJ, Sweeney C, Kubelka M et al. (1997). Delayed early embryonic lethality following disruption of the murine cyclin A2 gene. Nat Genet 15: 83–86. Nguyen TB, Manova K, Capodieci P, Lindon C, Bottega S, Wang XY et al. (2002). Characterization and expression of mammalian cyclin b3, a prepachytene meiotic cyclin. J Biol Chem 277: 41960–41969. Nieduszynski CA, Murray J, Carrington M. (2002). Whole-genome analysis of animal A- and B-type cyclins. Genome Biol 3: RESEARCH0070. O’Connor PM. (1997). Mammalian G1 and G2 phase checkpoints. Cancer Surv 29: 151–182. Ohshima T, Ward JM, Huh CG, Longenecker G, Veeranna, Pant HC et al. (1996). Targeted disruption of the cyclin-dependent kinase 5 gene results in abnormal corticogenesis, neuronal pathology and perinatal death. Proc Natl Acad Sci USA 93: 11173–11178. Ohtsubo M, Theodoras AM, Schumacher J, Roberts JM, Pagano M. (1995). Human cyclin E, a nuclear protein essential for the G1-to-S phase transition. Mol Cell Biol 15: 2612–2624. Ookata K, Hisanaga S, Okumura E, Kishimoto T. (1993). Association of p34cdc2/cyclin B complex with microtubules in starfish oocytes. J Cell Sci 105( Pt 4): 873–881. Ortega S, Prieto I, Odajima J, Martin A, Dubus P, Sotillo R et al. (2003). Cyclin-dependent kinase 2 is essential for meiosis but not for mitotic cell division in mice. Nat Genet 35: 25–31. Pagano M, Jackson PK. (2004). Wagging the dogma; tissue-specific cell cycle control in the mouse embryo. Cell 118: 535–538. Pagano M, Pepperkok R, Verde F, Ansorge W, Draetta G. (1992). Cyclin A is required at two points in the human cell cycle. EMBO J 11: 961–971. Parekh V, McEwen A, Barbour V, Takahashi Y, Rehg JE, Jane SM et al. (2004). Defective extraembryonic angiogenesis in mice lacking LBP-1a, a member of the grainyhead family of transcription factors. Mol Cell Biol 24: 7113–7129. Parisi T, Beck AR, Rougier N, McNeil T, Lucian L, Werb Z et al. (2003). Cyclins E1 and E2 are required for endoreplication in placental trophoblast giant cells. EMBO J 22: 4794–4803. Petersen BO, Lukas J, Sorensen CS, Bartek J, Helin K. (1999). Phosphorylation of mammalian CDC6 by cyclin A/CDK2 regulates its subcellular localization. EMBO J 18: 396–410. Pines J, Hunter T. (1989). Isolation of a human cyclin cDNA: evidence for cyclin mRNA and protein regulation in the cell cycle and for interaction with p34cdc2. Cell 58: 833–846. Pines J, Hunter T. (1991). Human cyclins A and B1 are differentially located in the cell and undergo cell cycle-dependent nuclear transport. J Cell Biol 115: 1–17. Pines J, Hunter T. (1994). The differential localization of human cyclins A and B is due to a cytoplasmic retention signal in cyclin B. EMBO J 13: 3772–3781. Oncogene

Pogoriler J, Millen K, Utset M, Du W. (2006). Loss of cyclin D1 impairs cerebellar development and suppresses medulloblastoma formation. Development 133: 3929–3937. Qu D, Rashidian J, Mount MP, Aleyasin H, Parsanejad M, Lira A et al. (2007). Role of Cdk5-mediated phosphorylation of Prx2 in MPTP toxicity and Parkinson’s disease. Neuron 55: 37–52. Rane SG, Dubus P, Mettus RV, Galbreath EJ, Boden G, Reddy EP et al. (1999). Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in beta-islet cell hyperplasia. Nat Genet 22: 44–52. Ravnik SE, Wolgemuth DJ. (1999). Regulation of meiosis during mammalian spermatogenesis: the A-type cyclins and their associated cyclin-dependent kinases are differentially expressed in the germ-cell lineage. Dev Biol 207: 408–418. Ren S, Rollins BJ. (2004). Cyclin C/cdk3 promotes Rb-dependent G0 exit. Cell 117: 239–251. Resnitzky D, Hengst L, Reed SI. (1995). Cyclin A-associated kinase activity is rate limiting for entrance into S phase and is negatively regulated in G1 by p27Kip1. Mol Cell Biol 15: 4347–4352. Riabowol K, Draetta G, Brizuela L, Vandre D, Beach D. (1989). The cdc2 kinase is a nuclear protein that is essential for mitosis in mammalian cells. Cell 57: 393–401. Rossi DJ, Londesborough A, Korsisaari N, Pihlak A, Lehtonen E, Henkemeyer M et al. (2001). Inability to enter S phase and defective RNA polymerase II CTD phosphorylation in mice lacking Mat1. EMBO J 20: 2844–2856. Sachs AB. (2000). Cell cycle-dependent translation initiation: IRES elements prevail. Cell 101: 243–245. Sage J. (2004). Cyclin C makes an entry into the cell cycle. Dev Cell 6: 607–608. Sage J, Miller AL, Perez-Mancera PA, Wysocki JM, Jacks T. (2003). Acute mutation of retinoblastoma gene function is sufficient for cell cycle re-entry. Nature 424: 223–228. Sano M, Schneider MD. (2003). Cyclins that don’t cycle–cyclin T/cyclin-dependent kinase-9 determines cardiac muscle cell size. Cell Cycle 2: 99–104. Santamaria D, Barriere C, Cerqueira A, Hunt S, Tardy C, Newton K et al. (2007). Cdk1 is sufficient to drive the mammalian cell cycle. Nature 448: 811–815. Satyanarayana A, Berthet C, Lopez-Molina J, Coppola V, Tessarollo L, Kaldis P. (2008a). Genetic substitution of Cdk1 by Cdk2 leads to embryonic lethality and loss of meiotic function of Cdk2. Development 135: 3389–3400. Satyanarayana A, Hilton MB, Kaldis P. (2008b). p21 Inhibits Cdk1 in the absence of Cdk2 to maintain the G1/S phase DNA damage checkpoint. Mol Biol Cell 19: 65–77. Sherr CJ. (1993). Mammalian G1 cyclins. Cell 73: 1059–1065. Sherr CJ, Roberts JM. (1999). CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev 13: 1501–1512. Sherr CJ, Roberts JM. (2004). Living with or without cyclins and cyclin-dependent kinases. Genes Dev 18: 2699–2711. Sicinska E, Aifantis I, Le Cam L, Swat W, Borowski C, Yu Q et al. (2003). Requirement for cyclin D3 in lymphocyte development and T cell leukemias. Cancer Cell 4: 451–461. Sicinski P, Donaher JL, B PS, Li T, Fazeli A, Gardner H et al. (1995). Cyclin D1 provides a link between development and oncogenesis in the retina and breast. Cell 82: 621–630. Sicinski P, Donaher JL, Geng Y, Parker SB, Gardner H, Park MY et al. (1996). Cyclin D2 is an FSH-responsive gene involved in gonadal cell proliferation and oncogenesis. Nature 384: 470–474. Solomon MJ, Booher R, Kirschner MW, Beach D. (1988). Cyclin in fission yeast. Cell 54: 738–740. Solvason N, Wu WW, Parry D, Mahony D, Lam EW, Glassford J et al. (2000). Cyclin D2 is essential for BCR-mediated proliferation and CD5 B cell development. Int Immunol 12: 631–638. Sotillo R, Dubus P, Martin J, de la Cueva E, Ortega S, Malumbres M et al. (2001). Wide spectrum of tumors in knock-in mice carrying a Cdk4 protein insensitive to INK4 inhibitors. EMBO J 20: 6637–6647.


2939 Susaki E, Nakayama K, Nakayama KI. (2007). Cyclin D2 translocates p27 out of the nucleus and promotes its degradation at the G0-G1 transition. Mol Cell Biol 27: 4626–4640. Sweeney C, Murphy M, Kubelka M, Ravnik SE, Hawkins CF, Wolgemuth DJ et al. (1996). A distinct cyclin A is expressed in germ cells in the mouse. Development 122: 53–64. Takaki T, Fukasawa K, Suzuki-Takahashi I, Semba K, Kitagawa M, Taya Y et al. (2005). Preferences for phosphorylation sites in the retinoblastoma protein of D-type cyclin-dependent kinases, Cdk4 and Cdk6, in vitro. J Biochem 137: 381–386. Tetzlaff MT, Bai C, Finegold M, Wilson J, Harper JW, Mahon KA et al. (2004). Cyclin F disruption compromises placental development and affects normal cell cycle execution. Mol Cell Biol 24: 2487–2498. Th’ng JP, Wright PS, Hamaguchi J, Lee MG, Norbury CJ, Nurse P et al. (1990). The FT210 cell line is a mouse G2 phase mutant with a temperature-sensitive CDC2 gene product. Cell 63: 313–324. Toyoshima F, Moriguchi T, Wada A, Fukuda M, Nishida E. (1998). Nuclear export of cyclin B1 and its possible role in the DNA damage-induced G2 checkpoint. EMBO J 17: 2728–2735. Toyoshima H, Hunter T. (1994). p27, a novel inhibitor of G1 cyclinCdk protein kinase activity, is related to p21. Cell 78: 67–74. Tsutsui T, Hesabi B, Moons DS, Pandolfi PP, Hansel KS, Koff A et al. (1999). Targeted disruption of CDK4 delays cell cycle entry with enhanced p27Kip1 activity. Mol Cell Biol 19: 7011–7019. van den Heuvel S, Harlow E. (1993). Distinct roles for cyclindependent kinases in cell cycle control. Science 262: 2050–2054.

Waclaw RR, Chatot CL. (2004). Patterns of expression of cyclins A, B1, D, E and cdk 2 in preimplantation mouse embryos. Zygote 12: 19–30. Wang S, Fischer PM. (2008). Cyclin-dependent kinase 9: a key transcriptional regulator and potential drug target in oncology, virology and cardiology. Trends Pharmacol Sci 29: 302–313. Weinberg RA. (1995). The retinoblastoma protein and cell cycle control. Cell 81: 323–330. Winston N, Bourgain-Guglielmetti F, Ciemerych MA, Kubiak JZ, Senamaud-Beaufort C, Carrington M et al. (2000). Early development of mouse embryos null mutant for the cyclin A2 gene occurs in the absence of maternally derived cyclin A2 gene products. Dev Biol 223: 139–153. Yang J, Bardes ES, Moore JD, Brennan J, Powers MA, Kornbluth S. (1998). Control of cyclin B1 localization through regulated binding of the nuclear export factor CRM1. Genes Dev 12: 2131–2143. Yasuda H, Kamijo M, Honda R, Nakamura M, Hanaoka F, Ohba Y. (1991). A point mutation in C-terminal region of cdc2 kinase causes a G2-phase arrest in a mouse temperature-sensitive FM3A cell mutant. Cell Struct Funct 16: 105–112. Ye X, Zhu C, Harper JW. (2001). A premature-termination mutation in the Mus musculus cyclin-dependent kinase 3 gene. Proc Natl Acad Sci USA 98: 1682–1686. Zindy F, Lamas E, Chenivesse X, Sobczak J, Wang J, Fesquet D et al. (1992). Cyclin A is required in S phase in normal epithelial cells. Biochem Biophys Res Commun 182: 1144–1154.

Oncogene