Cyclins, CDKs and cancer
Jonathon Pines
The connections between cancer and the basic machinery of the cell cycle have taken a surprisingly long time to become apparent. However, the past 2 years has en a dramatic increa in the number of cell cycles regulators that have been implicated as either protooncogenes or as tumour suppressor genes. In this review I will attempt to show how perturbations in the known cell cycle regulators may play a part in the process of oncogenesis.
Key words: cancer / cell cycle / cyclins / CDKs / kinas /oncogenesis
C ANCER is a cell cycle dia. Cancer cells still progress through the four broadly defined phas of the cell cycle: the G1 pha before DNA replication,DNA replication itlf (S pha), the G2 pha before cell division, and finally cell division (M pha).However, some of the controls exerted on progression through the cell cycle are lost when cells become transformed; in particular the controls or ‘check-points’ at the G1-S and the G2-M transitions are less stringent or even abnt in cancer cells. We are now beginning to understand the molecular changes responsible for the changes. It has recently been shown that some of the most fundamental changes in transformation are in the protein ki
na complexes that are thought to regulate cell cycle checkpoints.The most intensively studied of the regulators are complexes between members of the cyclin and the cyclin-dependent kina (CDK) families.
Cyclins and CDKs
The cyclins are a growing family of proteins that all share homology in a ~100 amino acid region called the cyclin box (Table 1). The cyclin box binds members of a well-conrved family of protein kinas that have the defining property of requiring a cyclin partner for their activation. Hence their designation
as cyclin-dependent kinas (Table 1). Both cyclins and CDKs have been, and are, continuously and extensively reviewed (not least by the author), so I will give no more than a brief summary here. (Figure 1).For more extensive reviews e ref 1 and 2.
From the crystal structure of CDK2 cyclins are predicted to activate their partner CDK by altering the conformation of the N-terminal lobe of the kina.3Cyclin binding probably achieves two purpos. Firstly,it reorientates bound ATP such that the scissile γ- phosphate bond becomes susceptible to nucleophilic attack from the substrate hydroxyl group. Secondly, it probably helps to displace a regio
n of the CDK, the T loop, that partially obscures the substrate binding site.A conrved threonine residue in the T loop is also phosphorylated by CAK (CDK activating kina) as part of the activation process. In mammalian cells there are veral cyclins and veral CDKs; some types of cyclin are able bind a variety of CDKs, and vice versa (Table 1). There is some evidence that the type of cyclin which is bound may affect the substrate specificity of the CDK. Without doubt, different cyclins are able to target a CDK to different compart-ments and thus different substrates in the cell. This is likely to be important in defining the specific roles for each cyclin–CDK complex, becau in vitro most of the cyclin-CDKs phosphorylate the same basic connsus quence; (K/R) - S/T-P - X - (K/R), where the basic residues are preferred but not esntial. The sub-strates of the CDKs have been recently reviewed.4
Some, perhaps all, cyclin–CDK complexes are further negatively regulated by phosphorylation of the CDK on residues in the ATP binding region. This regulation has been most clearly delineated for the mitotic cyclin B-cdc2 complex (reviewed in ref 5). In mammalian cells, the cyclin B-cdc2 complex accumu-lates in an inactive form through S and G2 phas,becau the protein kina activity of cdc2 is inhibited by phosphorylation on threonine 14 and tyrosine 15.The protein kinas responsible for the inhibitory phosphorylation of Y15 in fission yeast are the products of the weel and mik1genes,
and a human homologue has been identified. Until recently this was thought to be a weel homolog, but a full length clone now shows more homology to fission yeast mikl From the Wellcome/CRC Institute, Tennis Court Road, Cam-bridge, CB21QR, UK
minars in CANCER
BIOLOGY,Vol 6, 1995: pp 63–72报价单怎么写
©1995 Academic Press Ltd
白接骨>手机骷髅1044-579X/95/020063+10$08.00/0
T a b l e 1.C y c l i n -C D K c o m p l e x e s a n d t h e i r c o n n e c t i o n w i t h c a n c e r
C y c l i n C e l l c y c l e r o l e
A s s o c i a t e d C D K
A s s o c i a t e d p r o t e i n s S t a b i l i t y D e g r a d a t i o n L i n k t o c a n c e r
A -S +G 2→M
c d c 2 & C D K 2p 107 + E 2F , p 21, P C N A U n s t a b l e i n m i t o s i s U b i q u i t i n S t a b i l i z a t i o n i n h e p a t o c e l l u l a r c a r c i n o m a . C o m p l e x w i t h E 2F d i s r u p t e d b y E 1A . p 21 & P C N A n o t a s s o c i a t e d i n c a n c e r B 1-G 2→M c d c 2
p 21, P C N A U n s t a b l e i n m i t o s i s U b i q u i t i n N o B 2G 2→M p 21, P C N A U n s t a b l e i n m i t o s i s U b i q u i t i n N o C -G 1(?)C D K X ?N /D N /D P E S T N /D D 1G 1C D K 4 (2,5,6)R b , p 21, P C N A , p 16R a p i d t u r n o v e r P E S T
P R A D 1, b c l 1 p r o t o -o n c o g e n e s D 2G 1C D K 4 (2,5,6)R b , p 21, P C N A , p 16R a p i d t u r n o v e r v i n -1 p r o t o -o n c o g e n e
D 3G 1C D K 4 (2,5,6)
R b , p 21, P C N A , p 16R a p i d t u r n o v e r
A l l D -t y p e s女生微信名
p 21 & P C N A n o t a s s o c i a t e d i n c a n c e r E -G 1+G 1→S
C D K 2p 107+E 2F , p 21, P C N A p 27 a f t e r T G F ß t r e a t m e n t
R a p i d t u r n o v e r P E S T
英镑用英语怎么说
A l t e r e d l e v e l s a n d p r o t e i n i n t u m o u r s . C o m p l e x w i t h E 2F d i s r u p t e d b y E 1A . p 21 & P C N A n o t a s s o c i a t e d i n c a n c e r
F N /D N /D G
N /D N /D
I n d u c e d b y p 53
J. Pines
(T. Hunter, personal communication). The mammal-ian T14 kina has only been partially purified to date.The phosphata responsible for dephosphorylating Y15 is the product of the cdc25gene in fiss
ion yeast (reviewed in ref 5). In mammalian cells, three cdc25homologues have been identified, cdc25A, B and C,which appear to act at different points in the cell cycle (Figure 1). The cdc25C protein and cyclin B-cdc2 are both substrates for each other and form a positive feedback loop in the initiation of mitosis. Similarly,recent data show that cyclin E-CDK2 is activated by cdc25A in S pha.
The protein phosphata PP2A is intimately con-cerned in preventing the premature activation of cdc25C, and perhaps the other types of cdc25 too.When PP2A is inhibited by adding okadaic acid to cells or frog egg extracts, cdc25C is activated and in turn this caus the premature activation of cyclin
B-cdc2, the mitotic kina.6Okadaic acid is a powerful tumour promoter in vivo , and this may be related to its interference with the cell cycle machinery.
The cyclin–CDK complexes are inactivated by the cell cycle specific degradation of the cyclin partner.(The CDKs are stable proteins that tend to be in excess of the cyclin partners.) The cyclins can be roughly divided into two types depending on the manner of their degradation; the G1 cyclins and the mitotic cyclins. The G1 cyclins, D and E, are rapidly turned over (t 1~30 min), and thus their level is determined by the rate of their transcription. Their instability is conferred by ‘PEST’ quence
s in a region C-terminal to the cyclin box. In contrast the mitotic cyclins, A and B, are stable throughout interpha, but are rapidly and specifically degraded during mitosis. This property is conferred by a region called the destruction box, looly conrved between
Figure 1. Points of action of the cyclin–CDK complexes in the mammalian cell cycle. Schematic of the stages of the mammalian cell cycle, and the points at which the various cyclin-CDK complexes are thought to act. Also illustrated are the p21 and p27 CDIs acting in G1 pha. Note that p21 may also act on DNA replication complexes directly (e Figure 2) and potentially on cyclins A and B in G2 pha. The regulation of cyclin B-cdc2 by the phosphorylation state of Y15, through the antagonism of wee1/mik1and cdc25C is shown, but note that this regulation also applies to cyclin E-CDK2, and potentially other cyclin–CDK complexes.
Cyclins, CDKs and cancer
mitotic cyclins, which targets cyclins A and B for destruction in an ubiquitin-dependent manner in mitosis.7
Recently another mechanism by which cyclin–CDK complexes are regulated has come to light. A ries of cyclin-CDK inhibitors (CDIs) have been identified that bind and inhibit the kina complexes
directly. (Reviewed in ref 8). One of the, p16INK4, has been found to be rearranged/deleted in many of human tumours9,10(e later). Another CDI, p21, has proved to be an important player in the p53-mediated respon to DNA damage11,12and the p27KIP1CDI is involved in the negative growth respon to TGF .13 Cyclin–CDK complexes have been implicated in the regulation of the initiation and completion of DNA replication (S pha), of cell division (mitosis), and a major control point in G1 pha called START. At START, the cell commits itlf to a further round of DNA replication rather than the alternative fates of quiescence or, in yeast, mating. In mammalian cells the restriction point (R) can be thought of as the rough equivalent to START in yeast; after R the cell is able to go on to S pha in the abnce of rum. START is the point in the cell cycle at which signals from the external environment are integrated by the cell, and the balance of positive and negative growth signals determine whether the cell proliferates or becomes quiescent or differentiates. Tumour sup-pressors such as p53 and retinoblastoma protein (Rb), and proto-oncogenes such as myc, are thought to impinge upon the cell cycle at this stage. Thus it comes as no surpri that the cell cycle components most strongly implicated in oncogenesis are involved in the regulation of G1 pha. The are the D-type cyclin–CDK complexes (reviewed in more detail in this volume by Stewart Bates and Gordon Peters; Chapter 2). There is rather less evidence that changes in the other types of cyclins or CDKs, with the exception of the p16 and p21 CDIs, are common in oncogenesis.
D-type cyclins: the Bcl1, PRAD1 and vin-1 proto-oncogenes
The D-type cyclins are most likely to be involved in the regulation of the R point. There are three types of D cyclin; D1, D2 and D3 which are cell type specific.14 Most cells express D3, and either D1 or D2 (but not all three). Cyclin D1 has been identified as the PRAD1 proto-oncogene15and as the most likely candidate for the Bcl-1 proto-oncogene.16Cyclin D2 has been shown to be the mou vin-1 proto-oncogene.17 Human cyclin D3 has not yet been identified as a proto-oncogene. Recent work has shown that over-expressing D type cyclins alone is not sufficient to transform a cell, perhaps not surprising given that oncogenes need to co-operate in tumorogenesis.18 D-type cyclins have a very short half life (~25 min) and their synthesis is highly growth factor dependent; when growth factors are withdrawn, cyclin D synthesis ceas immediately.19This has led to the idea that the D-type cyclins may act as growth factor nsors. The D-type cyclins are unusual among the other types of cyclin in that they bind to Rb, through an L-X-C-X-E motif in their N-terminus.20,21The D-type cyclins are able to bind to veral different CDK partners; CDK2, CDK4, CDK5 and CDK6.22-24Of the, their main and consistent partner appears to be CDK4; in many cell types CDK2, CDK5 and CDK6 are not associated with cyclin D. CDK4 is unusual amongst the CDKs in that it associates with its partner cyclin for only a short period in the cell cycle, in late G1 and early S pha,25 and its 金针菇凉拌菜
synthesis is subject to regulation by negative growth factors such as TGF .2Cyclin D-CDK com-plexes have a very limited substrate specificity; in vitro the best substrate found to date is Rb.27
1787年p16INK4, a protein with four ankyrin repeats, specifi-cally binds and inhibits CDK4.28p16 was initially described as binding more cyclin D-CDK4 complexes in T Ag transformed cells, behaviour that is somewhat at odds with its recent propod role as a tumour suppressor gene.9,10There are two cloly linked genes for p16 that map to chromosome 9p129and are rearranged, deleted or mutated in a majority of gliomas, leukaemias and melanomas.9,10Indeed, p16 is a leading candidate for the melanoma susceptibility gene. Little is known of the physiological role of p16.
D-type cyclins as regulators of Rb
It is likely that D-type cyclins are important in the regulation of Rb in the cell cycle. RB is under-phosphorylated throughout G1 pha, is phosphory-lated at the G1/S transition and remains phosphory-lated until late mitosis. The phosphorylated and unphosphorylated forms of Rb have different proper-ties. Unphosphorylated Rb is able to block cells in G1 pha, and binds a bewildering variety of proteins (reviewed in this volume by T. Kouzarides in Chapter 4), whereas the phosphorylated form of Rb cannot. Perhaps the most relevant to the cell cycle are the
J. Pines
family of transcription factors collectively known as E2F. E2F has been implicated as responsible for transcribing a t of genes at the end of G1 pha that are required for DNA synthesis (reviewed in this issue by William Kaelin in Chapter 5). Thus one role for the D-type cyclin kina might be to phosphorylate Rb in order to relea E2F to turn on genes required for S pha.
D-type cyclins are involved in differentiation and co-operate with oncogenes
D-type cyclins, and Rb itlf, play an important role in the switch between proliferation and differentiation. 32D myeloid cells normally express cyclins D2 and D3 in a growth factor dependent manner, and proliferate in culture until G-CSF is added, which induces them to differentiate. When the cells are transfected with either cyclin D2 or D3 under a constitutive promoter, the cells are unable to differentiate in the prence of G-CSF; they continue to proliferate until they die.29 Constitutive expression of cyclin D1 has no effect on their differentiation, nor does expression of cyclin D2 and D3 mutants that are unable to bind Rb.
When D-type cyclin synthesis becomes constitutive and independent of the prence of growth factors, the cell cycle machinery may perceive this as a signal that growth factors are constantly pres
五角星拼音ent, and so cells would proliferate rather than differentiate. This is one cellular change that facilitates oncogenesis.18Errors in cyclin D expression would therefore be expected to co-operate with other oncogenes in transformation, and recent data show that deregulated cyclin D1 will co-operate with myc in transgenic mice,30and with ras and a defective E1A protein in cultured cells.31(See Chapter 2 by Stewart Bates and Gordon Peters for a more detailed discussion).
D- and E-type cyclins co-ordinate progress through G1 pha
Overexpressing D-type cyclins alone only moderately accelerates the cell’s entry into S pha, although Rb is phosphorylated much earlier than usual.32The E-type cyclins are thought to act at the G1-S transition itlf and be important in the initiation of DNA replication. However, like the D-type cyclins, over-expressing cyclin E only moderately accelerates entry into S pha.33But if both D- and E-type cyclins are overexpresd, there is a marked acceleration through G1 pha and into S pha.34This suggests that the D- and E-type cyclin-CDK complexes regulate different aspects of G1 pha.
Cyclin E-CDK2
Cyclin E forms an active protein kina complex exclusively with CDK2. Cyclin E synthesis begins in
late G1 pha and continues until the cells have entered S pha. Once the cells are in S pha, cyclin E is rapidly degraded. The cyclin E-CDK2 complex is heavily implicated as the protein kina that initiates S pha. Perhaps the most convincing evidence for this is that Drosophila which are mutant in cyclin E arrest in development with their cells blocked in late G1 pha.35Corroborating data come from studies on TGF growth repression in mammalian cells. The cyclin E-CDK2 complex appears to be the main target of TGF repression.36TGF treatment activates the p27KIP1CDI, which is prent in latent form in untreated cells. p27KIP1is structurally related to p21 in its N-terminus,37,38the part of the protein which interacts with the cyclin–CDK complex, and so the two CDIs may bind and inhibit the cyclin-CDKs in a similar fashion. p27KIP1binds and inhibits cyclin E-CDK2, and thus blocks the cell in late G1 pha.13There are data to suggest that this is also the means by which cells are normally arrested when they reach con-fluence. At prent there are no data to show whether p27Kip1is altered in tumour cells, which would not be surprising becau cancer cells are often unresponsive to both contact-inhibition and to TGF . Potentially, p27 forms a link between cyclin D and cyclin E, in that there are data to suggest that cyclin D2-CDK4 is able to quester p27, which would relieve its inhibition of cyclin E-CDK2.13
There is some evidence that cyclin E differs in normal compared with transformed cells. Cyclin E lev
els are incread in a large fraction of tumours, and some tumours have mutations in cyclin E.39Addition-ally, in common with cyclin A (e later), cyclin E-CDK2 associates with the Rb-related protein p107 and the transcription factor E2F.40The complexes are disrupted by the adenovirus E1A proteins in transformed cells,41but it is unclear what effect cyclin E-CDK2 has on E2F in normal cells. However, there is no evidence that cyclin E itlf is a proto-oncogene.
Cyclins, CDKs and cancer
