| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review article |
1 Stanford University, Pathology Department, 300 Pasteur Drive, MC 5324, Stanford, CA 94305, USA
2 Paterson Institute for Cancer Research, Christie Hospital NHS Trust, Manchester M20 4BX, UK
3 Cancer Research Institute and Department of Cellular and Molecular Pharmacology, University of California San Francisco, CA 94143-0875, USA
Correspondence
Gerard I. Evan
GEvan{at}cc.ucsf.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT AN INTRODUCTION TO KSHVSUMMARY AND PERSPECTIVESREFERENCES |
|---|
Published online ahead of print on 19 March 2004 as DOI 10.1099/vir.0.79812-0.
|
AN INTRODUCTION TO KSHV |
|---|
|
TOPABSTRACTAN INTRODUCTION TO KSHVSUMMARY AND PERSPECTIVESREFERENCES |
|---|
KSHV was first detected in Kaposi's sarcoma (KS) (Chang et al., 1994
), a skin tumour of endothelial origin often found in immunosuppressed patients (Boshoff & Weiss, 2001). KSHV is now accepted to be the transmissible agent of KS (Sarid et al., 1999). Since then, KSHV has also been found to be associated with a rare form of B cell lymphoma called primary effusion lymphoma (PEL) or body cavity-based lymphoma (Cesarman et al., 1995) and an unusual B cell plasmablast disease called multicentric Castleman's disease (Soulier et al., 1995).
KSHV shows sequence similarity to oncogenic gammaherpesviruses, including Epstein–Barr virus (Moore et al., 1996). Its genome encodes at least 81 open reading frames (ORFs), of which 66 show significant sequence similarity to ORFs of the rhadinovirus herpesvirus saimiri (HVS), a squirrel monkey virus (Russo et al., 1996c). This classifies KSHV in the family Herpesviridae, subfamily Gammaherpesvirinae, genus Rhadinovirus. Besides HVS and KSHV, this family includes mouse herpesvirus 68 (MHV68), equine herpesvirus 2 and viruses in primates (Lacoste et al., 2000).
All rhadinoviruses contain ORFs with striking homology to cellular genes. These genes are often unspliced, suggesting that they represent captured host cell RNAs. They encode proteins that control DNA synthesis, immune regulation, apoptosis or cell cycle progression (Moore & Chang, 1998). HVS, MHV68 and KSHV all encode a cyclin D homologue with properties that overlap with but extend the function of D-type cyclins. We focus here on the KSHV cyclin, as it serves as a prototype of virus-mediated deregulation of the cell cycle.
Expression of cyclins and cyclin-dependent kinases
Cell cycle transitions and exit from G0 are driven by activities of cyclin/cyclin-dependent kinase (Cdk) complexes. Mammalian G1 cyclins D and E mediate progression through G1/S phases. Three D-type cyclins exist (cyclin D1, D2 and D3), which are expressed differently in various cell lineages, with most cells expressing cyclin D3 and either D1 or D2 (Sherr, 1993). No differential roles for these cyclins have yet been recognized and evidence from mice expressing only a single D-type cyclin suggests that they are functionally redundant in mouse development (Ciemerych et al., 2002). Two types of cyclin E (E1 and E2) exist, which show overlapping expression patterns in mouse tissues and can be co-overexpressed in human tumours (Geng et al., 2001). Future research will elucidate whether any of their functions are distinct.
Mitotic cyclins A and B mediate progression through the S/G2/M phases. Cyclin A1 is expressed in meiosis and early embryogenesis, whereas cyclin A2 is found in proliferating somatic cells (Yang et al., 1997). Cyclin B2 probably plays a role in Golgi remodelling during mitosis (Jackman et al., 1995), while cyclin B1 controls other functions of this cyclin type. Additional cyclins exist (totalling around 16), for many of which binding partners and functions have yet to be identified.
At present, nine mammalian Cdks with known functions have been identified (Morgan, 1997), only four of which (Cdk1, -2, -4 and -6) are directly involved in cell cycle control. Cdk7 contributes indirectly by acting as a Cdk-activating kinase (CAK) that phosphorylates other Cdks (see below). Since Cdk expression levels are relatively constant throughout the cell cycle, the primary regulation of Cdk activity is governed by sequential synthesis, post-translational modification and degradation of cyclins.
D-type cyclins are short-lived proteins whose synthesis and assembly with Cdk4 or Cdk6 in G1 is dependent on mitogenic signalling (Hitomi & Stacey, 1999). Cyclin D/Cdk activity persists through the first and subsequent cycles as long as mitogenic stimulation continues (Matsushime et al., 1991). Cyclin E protein levels peak at the G1/S progression, followed by an increase in cyclin A levels in the S phase. Both cyclin E and A interact with and activate Cdk2, while cyclin A can also bind Cdk1 (Lees & Harlow, 1993). At the G2/M boundary, cyclin B levels increase, resulting in activation of its partner, Cdk1. This fluctuation in cyclin expression and resultant oscillation in Cdk activity (Fig. 1) form the basis of a coordinated cell cycle progression and a platform for more subtle regulatory controls.
|
; Lees & Harlow, 1993). Thus far, cyclin structures of bovine and human cyclin A, cyclin H and three viral cyclin D homologues have been described (Brown et al., 1995; Card et al., 2000; Jeffrey et al., 1995, 2000; Kim et al., 1996; Schulze-Gahmen et al., 1999). A comparison of these structures suggests that the core of all cyclins contains two compact domains of five helices, called the ‘cyclin fold’ (Fig. 2). The first helix bundle corresponds to the conserved cyclin box. This central fold is flanked by additional helices at the N and C termini, whose position and secondary structure vary in different cyclins.
|
-sheets (N lobe), a C-terminal lobe rich in 
-helices (C lobe) and a deep catalytic cleft at the junction of the lobes (De Bondt et al., 1993). Crystallized cyclin/Cdk6 shows a similar Cdk structure, suggesting that the structural features of Cdk2 are conserved in all Cdks (Jeffrey et al., 2000; Schulze-Gahmen & Kim, 2002). The catalytic activity of the monomeric Cdk subunit is restrained by two mechanisms. First, a large flexible loop (T-loop) on the C-terminal lobe obstructs substrate binding at the catalytic cleft. Secondly, a stretch of helical amino acids in the N-terminal lobe (PSTAIRE helix), which contains residues required for ATP phosphate binding, is directed away from the cleft (De Bondt et al., 1993) (Fig. 2
).
v-Cyclin is a cyclin D homologue.
The viral homologue of cellular cyclin D encoded by KSHV is called K cyclin or viral cyclin (v-cyclin). The v-cyclin sequence shows around 53 % sequence similarity to cyclin D2, with the highest level of homology around the cyclin box (Li et al., 1997). This, together with the ability of v-cyclin to bind and activate Cdk4 and Cdk6 and direct their kinase activity towards Rb (Chang et al., 1996; Godden-Kent et al., 1997; Li et al., 1997), shows that v-cyclin resembles D-type cyclins both at the structural and at the functional level.
Regulatory mechanisms governing cyclin/Cdk activity
Cdk activation by cyclin binding
Cdk activity is regulated through multiple mechanisms, the primary mechanism being its binding to the cyclin subunit. The crystal structure of the human Cdk2/cyclin A1 complex revealed that cyclin binding changes the Cdk2 T-loop structure and position such that it no longer obstructs the catalytic cleft. Cyclin binding also rotates the PSTAIRE helix and moves it into the catalytic cleft, resulting in the correct positioning of the ATP-binding catalytic residues (Jeffrey et al., 1995) (Fig. 2). Although the major cyclin/Cdk binding and structural characteristics are conserved, variations in the length of the Cdk PSTAIRE helix and position of the N-terminal cyclin helix exist (Jeffrey et al., 2000; Kim et al., 1996). This may underlie the specificity of Cdk4/6 for D-type cyclins, Cdk2 for cyclins A and E and Cdk1 for cyclin B.
v-Cyclin preferentially interacts with Cdk6.
In vitro binding and kinase assays indicate that v-cyclin mainly forms active complexes with Cdk6 (Godden-Kent et al., 1997; Li et al., 1997). The preference for Cdk6 is remarkable in light of the KSHV host cell repertoire, since lymphoid cells express relatively high levels of Cdk6 and cyclin D/Cdk6 activity predominates in such cells (Meyerson & Harlow, 1994). Cdk6 expression is also detectable in KS lesions (Ojala et al., 1999).
v-Cyclin/Cdk6 complexes show enhanced kinase activity.
v-Cyclin/Cdk6 complexes show enhanced kinase activity towards their substrates as measured by in vitro kinase assays when compared with cyclin D/Cdk complexes (Jeffrey et al., 2000; Li et al., 1997; Swanton et al., 1999). This could be the result of increased interactions between v-cyclin and Cdk, with the more stable active conformation accounting for an increased kinase activity (Jeffrey et al., 2000).
v-Cyclin/Cdk6 complexes phosphorylate Cdk2-type substrates.
v-Cyclin/Cdk6 complexes phosphorylate an extended array of substrates that are not normally targets of cyclin D/Cdk6. These include the cyclin E/Cdk2 targets p27, histone H1 and Cdc25A (Ellis et al., 1999; Mann et al., 1999) and the cyclin A/Cdk2 targets of the replication machinery, Cdc6 and Orc1 (Laman et al., 2001). The activity of v-cyclin/Cdk6 therefore mimics the combined activities of G1 and S phase cyclin/Cdk complexes. What explains this expanded substrate specificity? The substrate specificity of Cdk2 differs depending on its binding to cyclin E or A, implying that the cyclin subunit is likely to contribute (Sarcevic et al., 1997; Zarkowska et al., 1997). Indeed, substrates bind to a conserved binding groove on the cyclin, called the hydrophobic patch (Schulman et al., 1998). Another potential determinant is the conformation of the catalytic site (Brown et al., 1999). By inference, v-cyclin binding to substrates as well as the effect of v-cyclin on conformation of the Cdk6 catalytic cleft are likely to determine its substrate repertoire.
Regulation of Cdk activity by phosphorylation
Complete activation of Cdk activity is mediated by phosphorylation of a conserved threonine residue (Thr-160 in Cdk2, Thr-161 in Cdk1) in the T-loop by the CAK enzyme (Fig. 2). The major mammalian candidate for CAK is the cyclin H/Cdk7/Mat1 complex (Nigg, 1996). CAK phosphorylation moves the T-loop outwards, resulting in additional Cdk–cyclin contacts and probably increases potential for substrate binding (Russo et al., 1996b). Dephosphorylation of this residue can be achieved by the Cdk-associated phosphatase, KAP (Poon & Hunter, 1995).
In addition, Cdks undergo inhibitory phosphorylation on the Thr-14 and Tyr-15 residues located in the roof of the ATP phosphate-binding site (Fig. 2). It is likely that this reduces the affinity for ATP, thereby preventing Cdk catalytic activity (Endicott et al., 1999). Phosphorylation of these sites is mediated by the Wee1 and Myt1 protein kinases (Lew & Kornbluth, 1996) and dephosphorylation is carried out by phosphatases of the Cdc25 family. Cdc25A and B are active during G1/S and Cdc25C is the mitotic isoform. Not surprisingly, this extensive phosphorylation network is amenable to various regulatory mechanisms, which are central to the execution of cell cycle checkpoints.
Non-CAK-phosphorylated v-cyclin/Cdk6 complexes are active in vitro.
Unlike D-type cyclin/Cdk complexes, v-cyclin/Cdk6 complexes do not require CAK phosphorylation to generate active holoenzymes in vitro (Child & Mann, 2001; Kaldis et al., 2001), which renders v-cyclin function less dependent on host kinase activity compared with cyclin D. Activation is, however, incomplete, because non-CAK-phosphorylated v-cyclin/Cdk6 does not target the full range of retinoblastoma protein (Rb) phosphorylation sites and does not enable S-phase entry (Child & Mann, 2001).
Inhibition by Ink4-family proteins
The Ink4 family of Cdk inhibitors (CKIs) contains four members, named according to their molecular masses, p16Ink4a, p15Ink4b, p18Ink4c and p19Ink4d (p16, p15, p18 and p19) and their ability to inhibit Cdk4. These CKIs specifically inhibit Cdk4 and Cdk6 and do not bind Cdk2 or Cdk1. Ink4 proteins are composed of multiple ankyrin repeats and bind to the non-catalytic side of Cdk4 and Cdk6 opposite the cyclin D binding site (Russo et al., 1998). Binding of Ink4 inhibitors to Cdk4 or Cdk6 prevents Cdk interaction with cognate D-type cyclins: Ink4 binding induces allosteric changes by rotating the two Cdk4/6 lobes 15° in the vertical axis, thus mislocating the cyclin-interacting sequences and distorting the ATP binding site (Jeffrey et al., 2000) (Fig. 3).
|
). Insight into the lack of inhibition by Ink4 proteins was provided by the crystallization of the v-cyclin/Cdk6/p18 ternary complex (Jeffrey et al., 2000). The authors showed that CAK-phosphorylated Cdk6/v-cyclin complexes are resistant to inhibition by p18 and suggested that CAK phosphorylation of Cdk6 increases the interactions between the N and C lobes of Cdk6. This, together with increased interactions between v-cyclin and the Cdk6 C lobe could stabilize the active arrangement of the catalytic cleft and prevent its disruption upon Ink protein binding. In agreement with this, the interface of the HVS viral cyclin/Cdk6 complex is 20 % larger than in the cyclin A/Cdk2 complex (Schulze-Gahmen et al., 1999).
Inhibition by Cip/Kip family proteins
Cdk inhibitors of the Cip/Kip family are more broadly acting Cdk inhibitors. This family consists of three members, p21Cip1, p27Kip1 and p57Kip1 (p21, p27 and p57). p21 and p27 are potent inhibitors of cyclin D-, E- and A- and weak inhibitors of cyclin B-associated kinase activity (el-Deiry et al., 1993; Gu et al., 1993; Noda et al., 1994; Xiong et al., 1993). p57 is thought to be functionally analogous to other Cip/Kip proteins, but fewer studies have focused on p57.
The Cip/Kip family inhibitors can bind isolated cyclin and Cdk subunits independently, although their binding affinity for cyclin/Cdk complexes is higher (Chen et al., 1996; Harper et al., 1995; Lin et al., 1996; Toyoshima & Hunter, 1994). The Cip/Kip N-terminal moiety is necessary and sufficient to bind to and inhibit Cdk complexes. This N-terminal region contains two subsections: a short motif required for cyclin binding (Cy1) and a more complex segment that binds the Cdk subunit (Chen et al., 1996; Fotedar et al., 1996). The crystal structure of an N-terminal p27 peptide in complex with cyclin A/Cdk2 provided a breakthrough in our understanding of the inhibitory mechanism (Russo et al., 1996a). The p27 peptide stretches across the top of the cyclin/Cdk complex in an extended conformation (Fig. 3). The N-terminal part of the p27 peptide binds a conserved hydrophobic groove on the cyclin box. This p27/cyclin interaction is likely to serve as an anchor to facilitate subsequent p27/Cdk binding, since it does not require conformational changes. The C-terminal part of the peptide interacts with the upper Cdk lobe, disrupting the conformation of the ATP binding site. In addition, a p27 helix inserts into the catalytic cleft and prevents ATP binding. Together, these interactions disrupt any catalytic potential of the cyclin/Cdk complex.
Multiple experiments have indicated that p21 and p27 can associate with active cyclin D/Cdk complexes and may therefore act as assembly factors (Sherr & Roberts, 1999). However, p21/p27–/– cells proliferate and are susceptible to p16-mediated growth inhibition (Cheng et al., 1999), implying that p21 and p27 are not strictly required for cyclin D/Cdk activity. Furthermore, the stability of cyclin D and Cdks is significantly reduced in the absence of p21 and p27 (Bagui et al., 2000; Cheng et al., 1999), complicating assessment of the effect of Cip/Kip binding on cyclin D/Cdk complex activity. Purification and crystallization of the cyclin D/Cdk/CKI complex will clarify the properties of the holoenzyme.
v-Cyclin/Cdk6 complexes are less susceptible to inhibition by Cip/Kip proteins.
Similar to Ink4 inhibitors, Cip/Kip proteins are also unable to inhibit v-cyclin/Cdk6-driven kinase activity towards Rb in protein extracts from baculovirus-infected Sf9 cells (Swanton et al., 1997). The resistance to Cip/Kip proteins correlates with the lack of p21/p27 binding to v-cyclin/Cdk complexes. This conclusion is supported by the superimposition of the cyclin A/p27 structure over the HVS viral cyclin structure (Schulze-Gahmen et al., 1999). While the p27 binding pocket of cellular cyclins is highly conserved, viral cyclins contain substitutions in residues that are crucial for p27 binding. Notably, Glu-66 of cyclin D (Glu-220 of cyclin A), which forms a hydrogen bond and a salt bridge with p27, is substituted for Ser-60 in v-cyclin (Swanton et al., 1999). This results in loss of the electronegative character of the p27 binding groove and reduces the number of possible contacts between v-cyclin and p27, together decreasing the likelihood of p27 binding to v-cyclin (Schulze-Gahmen et al., 1999; Swanton et al., 1999).
Cyclin/Cdk localization
Since many critical cyclin/Cdk targets are nuclear proteins, nucleocytoplasmic shuttling is a means of regulating cyclin/Cdk activity (Yang & Kornbluth, 1999). The G1/S cyclins A and E are expressed predominantly in the nucleus (Ohtsubo et al., 1995; Pines & Hunter, 1991), consistent with their roles in DNA replication and S-phase progression. In contrast, cyclin B is exclusively cytoplasmic during G1/S, followed by a dramatic nuclear relocation upon mitotic entry (Takizawa & Morgan, 2000). Differences in the shuttling behaviour of these cyclins are a consequence of distinct nuclear import and export mechanisms and speeds (Jackman et al., 2002).
Cyclin D/Cdk complexes accumulate in the nucleus throughout the G1 phase but then relocate to the cytoplasm during the remainder of interphase (Baldin et al., 1993). Since neither cyclins nor Cdks contain canonical nuclear localization sequences (NLSs), nuclear import may be mediated via their binding to NLS-containing proteins. Suggested cyclin D nuclear import factors are Cip/Kip proteins, which contain an NLS and promote cyclin D/Cdk nuclear accumulation (Cheng et al., 1999; LaBaer et al., 1997). However, absence of both p21 and p27 does not abolish cyclin D1 nuclear import (Cheng et al., 1999) and, instead, p21 and p27 have been suggested to promote cyclin D1 nuclear accumulation via inhibition of a glycogen synthase kinase (GSK)-3-triggered nuclear export mechanism (Alt et al., 2002). Another candidate import factor is the p34SEI-1 protein that binds to cyclin D1/Cdk4 complexes and contains an NLS (Sugimoto et al., 1999). Yet another possibility is that cyclin D contains an unidentified NLS, as suggested for cyclin E (Moore et al., 1999).
How is nuclear accumulation of v-cyclin regulated?
Cyclin D1 nuclear export is triggered upon GSK-3-mediated phosphorylation of Thr-286 in cyclin D1 (Alt et al., 2002). v-Cyclin does not contain a residue homologous to the cyclin D1 Thr-286 residue, perhaps implying that v-cyclin evades nuclear export. This could explain why nuclear accumulation of v-cyclin is pronounced (Child & Mann, 2001), even though it does not bind p21/p27 CKIs (Swanton et al., 1997).
Cyclin/Cdk/CKI stability
While Cdks are expressed constitutively and are relatively stable, cyclins are subject to regulated degradation. The degradation of cyclins is regulated by the ubiquitin–proteasome system. Ubiquitination involves the covalent attachment of ubiquitin chains to lysine residues in the target protein, which promotes its recognition and degradation by the proteasome (Harper et al., 2002) (Fig. 4A). Two ubiquitin ligases regulate cell cycle progression: the Skp/Cullin/F-box (SCF) complex targets proteins at the G1/S/G2 phases and the anaphase-promoting complex (APC) acts at G2/M (Fig. 4B). Target specificity of these ligases is governed by their binding to adaptor proteins. These comprise F-box sequence motif-containing proteins such as Skp2 in the case of SCF (SCFskp2), or WD-40 repeat-containing proteins such as Cdh1 or Cdc20 in case of APC (APCCdh1, APCCdc20).
|
(Diehl et al., 1997, 1998). Cyclin E/Cdk2 complexes can initiate cyclin E degradation via phosphorylation of Thr-380, thus forming a negative feedback loop to limit cyclin E expression to late G1 (Clurman et al., 1996; Won & Reed, 1996). Degradation of p27 is initiated by cyclin E/Cdk2-catalysed phosphorylation of Thr-187 (Vlach et al., 1997). The APC mediates degradation of cyclin A after nuclear envelope breakdown in prometaphase, while cyclin B1 degradation occurs during the metaphase to anaphase transition (Clute & Pines, 1999; den Elzen & Pines, 2001; Geley et al., 2001) Interestingly, the APC is activated upon phosphorylation of APC core subunits by Cdk2 and Cdk1 itself (Morgan, 1999). Thus, protein degradation is initiated by cell cycle proteins, ensuring timed regulation of protein activity.
Is v-cyclin more stable than cyclin D?
v-Cyclin does not contain residues homologous to the C-terminal cyclin D1 Thr-286 residue or cyclin E Thr-380 residue that have been implicated in the destruction of these cyclins. However, a direct comparison of the stability of v-cyclin compared with cyclin D1 has not yet been performed.
G1/S progression
The retinoblastoma protein
The most recognized function of cyclin D/Cdk activity is inactivation of the Rb tumour suppressor. Rb and its relatives p107 and p130 are called ‘pocket’ proteins (Harbour & Dean, 2000), reflecting the folding of conserved domains around a pocket that constitutes the protein-binding site. The most established function of pocket proteins is interaction with and inhibition of the E2F family of transcription factors.
Rb is thought to repress E2F-dependent transcription by at least two mechanisms. First, Rb directly binds the transactivation domain of E2F, thus blocking recruitment of the transcription machinery (Helin et al., 1993). Secondly, Rb binding to E2F actively represses transcription via its interaction with factors that influence chromatin structure. These include histone deacetylases (HDACs) (Kouzarides, 1999) and components of the SWI/SNF chromatin-remodelling complex (Dunaief et al., 1994; Singh et al., 1995; Zhang et al., 2000). Together, these factors mediate tight assembly of chromatin in nucleosomes, inhibiting access of the transcriptional apparatus.
Rb contains 16 potential Cdk phosphorylation sites, which regulate the binding of distinct proteins (Knudsen & Wang, 1997). Complete phosphorylation of Rb relieves its binding to E2F, triggering E2F-dependent transcription of G1/S genes. Successive phosphorylation by cyclin D/Cdk complexes in early G1 and cyclin E/Cdk2 complexes in late G1 is necessary for hyperphosphorylation of Rb (Lundberg & Weinberg, 1998), and phosphorylation by cyclin A/Cdk2 may maintain Rb hyperphosphorylation during S phase (Sherr, 1996). Although based on overexpression studies, a potential model is that phosphorylation by cyclin D/Cdk4 first disrupts the association of Rb with HDAC, thereby allowing cyclin E expression. The interaction between Rb and SWI/SNF is, however, maintained and is sufficient to repress the cyclin A promoter. Upon phosphorylation by cyclin E/Cdk2, Rb repression of the cyclin A promoter is relieved (Harbour et al., 1999; Zhang et al., 2000), ensuring orderly activation of cyclin E and A in S phase.
The E2F family of transcription factors
The E2F family of transcription factors regulates the transcription of genes involved in cell cycle progression, nucleotide biosynthesis and DNA replication. Six E2F family proteins exist (E2F1–6) which heterodimerize with DP proteins (DP1 and DP2) in all possible combinations (Trimarchi & Lees, 2002). E2F4 and E2F5 in complex with p130 mediate the transcriptional repression of E2F-responsive genes in quiescent or differentiated cells (Trimarchi & Lees, 2002). In contrast, E2F1, E2F2 and E2F3 are expressed specifically in dividing cells where they interact with Rb and p107. They are potent transcriptional activators and cells lacking expression of E2F1, E2F2 and E2F3 are unable to proliferate (Wu et al., 2001), confirming their role in S-phase progression.
The complexity of E2F-triggered responses is illustrated by results from microarray analyses (Ishida et al., 2001; Muller et al., 2001), which identified E2F-induced genes that function during mitosis (e.g. Cdc2 and cyclin B), as well as genes involved in apoptosis and development. In addition, analysis of E2F-repressed genes in quiescent cells identified mitotic genes, including regulators of cytokinesis (e.g. polo-kinase), chromosome condensation and segregation (e.g. securin) (Ren et al., 2002). These results suggest a broader role for E2Fs in mitosis as well as G1/S.
Orderly progression through G1/S phases
A (simplified) scheme of G1/S progression can be drawn based on the assimilation of the previous information (Fig. 5). Cell cycle gene expression is suppressed in quiescent cells via the activity of p130/E2F4/E2F5 complexes. Upon mitogen stimulation, cyclin D levels rise and cyclin D/Cdk activity increases. In addition, Cip/Kip protein increases promote cyclin D nuclear accumulation and stability. Here, cyclin D/Cdk complexes initiate hypophosphorylation of Rb, which stimulates cyclin E expression. Cyclin E/Cdk2 activity also increases because the Cip/Kip inhibitors are sequestered away from cyclin E/Cdk towards cyclin D/Cdk complexes. This drives a feedback loop towards further phosphorylation of Rb, further amplified by degradation of p27 upon cyclin E/Cdk2 phosphorylation. Together, these signals culminate in transcriptional activation of genes including cyclin A and E2F itself.
|
) and cyclin E overexpression is sufficient to overcome growth arrest induced by a phosphorylation-resistant Rb mutant (Lukas et al., 1997). This implies that the main downstream target of cyclin D/Cdk activity is Rb, while cyclin E/Cdk2 has additional S-phase targets. Indeed, cyclin E, like cyclin A, triggers phosphorylation of DNA replication proteins such as Cdc6, directly activating DNA replication (Bell & Dutta, 2002). This simple model is, however, challenged by findings that mice lacking cyclin E1 and E2 or Cdk2 expression are viable and that cell cycle progression is uncompromised in cells derived from these mice (Berthet et al., 2003; Geng et al., 2003; Ortega et al., 2003). Specific cyclin/Cdk activities may therefore exhibit previously unappreciated functional redundancies.
Active cyclin D/E- and A-dependent kinases are regulated by negative feedback loops: cyclin E/Cdk2 complexes phosphorylate cyclin E, directing it to SCF-mediated ubiquitination. Cyclin A/Cdk2 phosphorylation of components of the APC may initiate proteolysis of cyclin A itself. Importantly, E2Fs are inactivated around S/G2 by at least two mechanisms. First, cyclin A/Cdk2 phosphorylation of E2F1 inhibits its DNA binding activity (Krek et al., 1994). Secondly, E2F1 undergoes SCFSkp2-dependent ubiquitination and degradation (Marti et al., 1999). Negative regulation of G1/S progression can also be enforced upon CKI induction, for example upon the execution of checkpoints.
v-Cyclin expression deregulates G0/G1/S control
v-Cyclin/Cdk complexes exhibit properties that extend those of cellular cyclins and potentially enable productive KSHV propagation. It is unknown whether KSHV infects proliferating progenitor or resting quiescent cells that contain high levels of CKIs. If the latter is true, then v-cyclin could still form active complexes with endogenous Cdks and stimulate exit from G0. In addition, its broadened substrate specificity to include G1- and S-phase targets would allow progression through G1 and initiation of S phase. Functional studies support this hypothesis, as v-cyclin expression induces S-phase entry in quiescent cells and in p16- or p27-overexpressing, G1-arrested cells (Swanton et al., 1997). Furthermore, the evasion of Cdk inhibition by v-cyclin/Cdk complexes would be beneficial in case of the activation of anti-viral tumour suppressor pathways by the host cell. Importantly, transgenic expression of v-cyclin promotes tumorigenesis (Verschuren et al., 2002), showing that v-cyclin has intrinsic oncogenic properties that can enhance host cell proliferation.
v-Cyclin/Cdk6 phosphorylation of replication proteins may also directly control DNA replication. Indeed, addition of v-cyclin, but not cyclin D, to isolated G1 nuclei triggered replication initiation in a manner analogous to cyclin A (Laman et al., 2001). Initiation starts with the assembly of pre-replicative complexes at replication origins in a reaction known as ‘licensing’ (Bell & Dutta, 2002). Further elucidation of DNA replication mechanisms in human cells will clarify how v-cyclin controls origin licensing and/or firing.
Cell cycle progression upon v-cyclin expression still relies on the host cell cycle machinery. Cells expressing a p27 T187A mutant that cannot be degraded cannot complete S phase even when v-cyclin is expressed (Ellis et al., 1999; Mann et al., 1999). This implies that activation of endogenous Cdks upon p27 degradation is required for full S-phase progression. Complementing these data, v-cyclin expression is unable to bypass a block on endogenous Cdk2 imposed by the chemical Cdk inhibitor roscovitine or by dominant-negative Cdk2 (Ellis et al., 1999). Another factor is the dependence of v-cyclin/Cdk activity on CAK phosphorylation by endogenous enzyme (Jeffrey et al., 2000; Kaldis et al., 2001). Thus, v-cyclin expression is likely to create an environment that is more conducive to (viral) DNA replication, but requires endogenous cyclin/Cdk activity for full cell cycle progression.
G2/M progression
Mitotic entry
Mitotic events such as nuclear envelope breakdown coincide with cyclin B/Cdk1 nuclear import (Hagting et al., 1999). Cyclin B nuclear localization is thought to be initiated upon its phosphorylation by cyclin B/Cdc2 itself or polo-kinase (Plk). Plk phosphorylation of Cdc25C also stimulates the dephosphorylation of inhibitory phosphates on Cdk2, further enhancing cyclin B/Cdk1 activity (Kumagai & Dunphy, 1996; Qian et al., 2001) (Fig. 6). Targets of cyclin B/Cdc2 include structural proteins and nucleolar proteins that regulate nuclear envelope breakdown and spindle assembly (Nigg, 1993).
|
). In addition, degradation of cyclins A and B is essential for mitotic progression beyond anaphase (Morgan, 1999), as illustrated by findings that non-degradable versions of cyclin A and B arrest cells in metaphase or anaphase (Geley et al., 2001; Parry & O'Farrell, 2001; Sigrist et al., 1995). Mitotic cyclin degradation is coordinated by the sequential activation of the APCCdc20 and APCCdh1 and, again, oscillations in kinase activity are key (Harper et al., 2002). First, the APCCdc20 complex is activated upon phosphorylation by Plk and cyclin B/Cdk1. This may increase the binding of the Cdc20 adaptor to APC and stimulate APCCdc20 activity towards degradation of cyclins A and B. Plk thus acts as a master regulator of mitosis through triggering both the activation and degradation of cyclin B. Secondly, cyclin A/Cdk2 phosphorylation of Cdh1 inhibits Cdh1 binding to APC and the degradation of cyclin A therefore licenses activation of APCCdh1 and complete degradation of mitotic cyclins (Lukas et al., 1999) (Fig. 6). Thirdly, cytokinesis is executed upon removal of inhibitory phosphates on Cdh1 and cyclin B/Cdk1 mitotic substrates by the Cdc14 phosphatase (Visintin et al., 1998).
Although securin and mitotic cyclins are the best understood APC substrates, a large number of other proteins are degraded by the APC. Examples are Plks themselves and Aurora protein kinases (Bischoff & Plowman, 1999). Both kinases localize to mitotic structures such as centrosomes, which constitute the microtubule organizing centres of the cell (Doxsey, 2001; Hinchcliffe & Sluder, 2001) and mitotic spindles. Deregulation of Aurora and Plk kinases disrupts mitotic exit (Meraldi et al., 2002), and the ability of APCCdh1 to keep their levels in check is therefore important for a proper execution of cytokinesis.
v-Cyclin expression deregulates mitotic progression
v-Cyclin-expressing cells undergo continued DNA synthesis and nuclear division, yet do not undergo cytokinesis (Verschuren et al., 2002). Thus v-cyclin/Cdk complexes, as well as deregulating G1/S progression, also deregulate mitosis. The cytokinesis defect may involve v-cyclin-associated misexpression of mitotic regulators, most likely through an increase of E2F transcriptional targets. Indeed, defects in cell division and concomitant multinucleation have been reported upon overexpression of Plk and Aurora kinases (Meraldi et al., 2002; Mundt et al., 1997; Wang et al., 2002). However, the mechanistic explanation of these results remains unknown.
An outcome of mitotic deregulation in v-cyclin-expressing cells is activation of the p53 tumour suppressor, triggering apoptosis and growth arrest (Verschuren et al., 2002). Activation of p53 occurs in an E2F1- and p19ARF-independent manner and may instead occur as a response to the formation of tetraploid cells. Active p53 would normally trigger a G1 arrest via induction of p21 protein and consequent G1/S Cdk inhibition (Stewart & Pietenpol, 2001). However, v-cyclin/Cdk complexes are refractory to Cdk inhibitors, which most likely explains why v-cyclin-expressing cells become polyploid. The increase in ploidy correlates with an increase in the number of centrosomes, explained by the co-regulation of the centrosome duplication and DNA replication cycles (Hinchcliffe & Sluder, 2001). A second outcome of mitotic deregulation is therefore the induction of genomic instability through centrosome amplifications and consequent spindle defects.
|
SUMMARY AND PERSPECTIVES |
|---|
|
TOPABSTRACTAN INTRODUCTION TO KSHVSUMMARY AND PERSPECTIVESREFERENCES |
|---|
). These include functions that deregulate G1/S progression, such as enhanced kinase activity towards Rb, a promiscuous substrate specificity, escape from inhibition by CKIs and initiation of DNA replication. In addition, v-cyclin/Cdk activity may promote genomic instability by disrupting mitotic progression. Counteracting these pro-tumorigenic stimuli, v-cyclin expression sensitizes cells to undergo apoptosis (Ojala et al., 1999; Verschuren et al., 2002), in part by inactivating the anti-apoptotic cellular Bcl-2 pool (Ojala et al., 2000), and triggers a p53-dependent growth arrest (Verschuren et al., 2002). v-Cyclin-expressing cultured cells only survive and expand as an aneuploid population when p53 is inactive. These findings are confirmed in vivo, as Eµ v-cyclin expression drives lymphomagenesis only when the p53 pathway is disrupted (Verschuren et al., 2002), probably via allowing the expansion of a genetically altered population of lymphocytes (Verschuren et al., 2004).
|
; van Dyk et al., 2000). However, thus far MHV68 cyclin is the only viral cyclin studied in an in vivo context and no deductions for persistent KSHV infection in a physiological context can be made.
Interesting but largely speculative topics of discussion concern the interplay between virally expressed genes. The majority of tumour cells in KSHV-associated diseases are latently infected, with only 1–3 % of the cells showing lytic replication (Zhong et al., 1996). Latent gene expression is highly restricted and encompasses an ‘oncogenic cluster’ of three genes with overlapping transcripts: latency-associated nuclear antigen (LANA), v-cyclin and viral FLICE-inhibitory protein (v-FLIP) (Jenner et al., 2001; Paulose-Murphy et al., 2001). In addition, several RNA transcripts encoding interferon regulatory factors (IRFs) are detected. Since latent gene expression is widespread in KSHV diseases, the combination of latent protein expression may be critical for oncogenesis.
An in vivo functional interaction may exist between the latent proteins LANA and ORF K10 IRF, which bind and inhibit p53 (Friborg et al., 1999; Rivas et al., 2001), and v-cyclin. Indeed, LANA colocalizes with p53 in KS samples (Katano et al., 2001). This, together with the absence of recurrent p53 mutations in KSHV-associated disorders (Carbone et al., 1998; Katano et al., 2001; Kennedy et al., 1998), raises the intriguing possibility that v-cyclin-induced apoptosis and/or growth arrest are inhibited by these proteins. Secondly, LANA binds the Rb protein and activates E2F-dependent transcription (Radkov et al., 2000), and LANA negatively regulates GSK-3 function to increase -catenin protein stability and transcriptional activity (Fujimuro et al., 2003), constituting further proliferative stimuli that could enhance v-cyclin function. Related to this, inhibition of GSK-3 by LANA could promote the nuclear localization of cellular cyclin D. Thirdly, latent expression of v-FLIP may cooperate with v-cyclin to promote KSHV tumorigenesis via its ability to inhibit death receptor-triggered lymphocyte apoptosis (Djerbi et al., 1999).
The establishment of KSHV gene functions in vivo is further complicated by the fact that infected cells may support the growth of uninfected cells in trans (Ensoli & Sturzl, 1998). Endothelial cells expressing the KSHV-encoded G protein-coupled receptor (GPCR), for example, cooperate with cells expressing v-cyclin in sarcomagenesis through paracrine mechanisms (Montaner et al., 2001). This emphasizes the need to assess KSHV gene function in a physiological context.
While 10 % of spindle cells in early KS lesions are KSHV-positive (Dupin et al., 1999), around 90 % of the spindle cells in late stage nodular lesions are KSHV-infected, suggesting that the virus provides a growth advantage to infected cells (Boshoff & Weiss, 2001). Furthermore, KS and PEL begin as polyclonal hyperplasias that develop into monoclonal tumours (Gill et al., 1998; Judde et al., 2000). The all-encompassing question is therefore whether v-cyclin expression promotes the tumorigenic evolution of KSHV-infected cells by triggering genomic instability and clonal outgrowth of cells with an enhanced proliferative capacity.
|
REFERENCES |
|---|
|
TOPABSTRACTAN INTRODUCTION TO KSHVSUMMARY AND PERSPECTIVESREFERENCES |
|---|
Bagui, T. K., Jackson, R. J., Agrawal, D. & Pledger, W. J. (2000). Analysis of cyclin D3–cdk4 complexes in fibroblasts expressing and lacking p27kip1 and p21cip1. Mol Cell Biol 20, 8748–8757.
Baldin, V., Lukas, J., Marcote, M. J., Pagano, M. & Draetta, G. (1993). Cyclin D1 is a nuclear protein required for cell cycle progression in G1. Genes Dev 7, 812–821.
Bell, S. P. & Dutta, A. (2002). DNA replication in eukaryotic cells. Annu Rev Biochem 71, 333–374.[CrossRef][Medline]
Berthet, C., Aleem, E., Coppola, V., Tessarollo, L. & Kaldis, P. (2003). Cdk2 knockout mice are viable. Curr Biol 13, 1775–1785.[CrossRef][Medline]
Bischoff, J. R. & Plowman, G. D. (1999). The Aurora/Ipl1p kinase family: regulators of chromosome segregation and cytokinesis. Trends Cell Biol 9, 454–459.[CrossRef][Medline]
Boshoff, C. & Weiss, R. A. (2001). Epidemiology and pathogenesis of Kaposi's sarcoma-associated herpesvirus. Philos Trans R Soc Lond Ser B Biol Sci 356, 517–534.[Medline]
Brown, N. R., Noble, M. E., Endicott, J. A., Garman, E. F., Wakatsuki, S., Mitchell, E., Rasmussen, B., Hunt, T. & Johnson, L. N. (1995). The crystal structure of cyclin A. Structure 3, 1235–1247.[Medline]
Brown, N. R., Noble, M. E., Endicott, J. A. & Johnson, L. N. (1999). The structural basis for specificity of substrate and recruitment peptides for cyclin-dependent kinases. Nat Cell Biol 1, 438–443.[CrossRef][Medline]
Carbone, A., Cilia, A. M., Gloghini, A., Capello, D., Todesco, M., Quattrone, S., Volpe, R. & Gaidano, G. (1998). Establishment and characterization of EBV-positive and EBV-negative primary effusion lymphoma cell lines harbouring human herpesvirus type-8. Br J Haematol 102, 1081–1089.[CrossRef][Medline]
Card, G. L., Knowles, P., Laman, H., Jones, N. & McDonald, N. Q. (2000). Crystal structure of a gamma-herpesvirus cyclin–cdk complex. EMBO J 19, 2877–2888.[CrossRef][Medline]
Cesarman, E., Chang, Y., Moore, P. S., Said, J. W. & Knowles, D. M. (1995). Kaposi's sarcoma-associated herpesvirus-like DNA sequences in AIDS-related body-cavity-based lymphomas. N Engl J Med 332, 1186–1191.
Chang, Y., Cesarman, E., Pessin, M. S., Lee, F., Culpepper, J., Knowles, D. M. & Moore, P. S. (1994). Identification of herpesvirus-like DNA sequences in AIDS-associated Kaposi's sarcoma. Science 266, 1865–1869.
Chang, Y., Moore, P. S., Talbot, S. J., Boshoff, C. H., Zarkowska, T., Godden, K., Paterson, H., Weiss, R. A. & Mittnacht, S. (1996). Cyclin encoded by KS herpesvirus. Nature 382, 410.[CrossRef][Medline]
Chen, J., Saha, P., Kornbluth, S., Dynlacht, B. D. & Dutta, A. (1996). Cyclin-binding motifs are essential for the function of p21CIP1. Mol Cell Biol 16, 4673–4682.[Abstract]
Cheng, M., Olivier, P., Diehl, J. A., Fero, M., Roussel, M. F., Roberts, J. M. & Sherr, C. J. (1999). The p21Cip1 and p27Kip1 CDK ‘inhibitors' are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 18, 1571–1583.[CrossRef][Medline]
Child, E. S. & Mann, D. J. (2001). Novel properties of the cyclin encoded by Human Herpesvirus 8 that facilitate exit from quiescence. Oncogene 20, 3311–3322.[CrossRef][Medline]
Ciemerych, M. A., Kenney, A. M., Sicinska, E., Kalaszczynska, I., Bronson, R. T., Rowitch, D. H., Gardner, H. & Sicinski, P. (2002). Development of mice expressing a single D-type cyclin. Genes Dev 16, 3277–3289.
Clurman, B. E., Sheaff, R. J., Thress, K., Groudine, M. & Roberts, J. M. (1996). Turnover of cyclin E by the ubiquitin–proteasome pathway is regulated by cdk2 binding and cyclin phosphorylation. Genes Dev 10, 1979–1990.
Clute, P. & Pines, J. (1999). Temporal and spatial control of cyclin B1 destruction in metaphase. Nat Cell Biol 1, 82–87.[CrossRef][Medline]
De Bondt, H. L., Rosenblatt, J., Jancarik, J., Jones, H. D., Morgan, D. O. & Kim, S. H. (1993). Crystal structure of cyclin-dependent kinase 2. Nature 363, 595–602.[CrossRef][Medline]
den Elzen, N. & Pines, J. (2001). Cyclin A is destroyed in prometaphase and can delay chromosome alignment and anaphase. J Cell Biol 153, 121–136.
Diehl, J. A., Zindy, F. & Sherr, C. J. (1997). Inhibition of cyclin D1 phosphorylation on threonine-286 prevents its rapid degradation via the ubiquitin–proteasome pathway. Genes Dev 11, 957–972.
Diehl, J. A., Cheng, M., Roussel, M. F. & Sherr, C. J. (1998). Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization. Genes Dev 12, 3499–3511.
Djerbi, M., Screpanti, V., Catrina, A. I., Bogen, B., Biberfeld, P. & Grandien, A. (1999). The inhibitor of death receptor signaling, FLICE-inhibitory protein, defines a new class of tumor progression factors. J Exp Med 190, 1025–1032.
Doxsey, S. J. (2001). Centrosomes as command centres for cellular control. Nat Cell Biol 3, E105–108.[CrossRef][Medline]
Dunaief, J. L., Strober, B. E., Guha, S., Khavari, P. A., Alin, K., Luban, J., Begemann, M., Crabtree, G. R. & Goff, S. P. (1994). The retinoblastoma protein and BRG1 form a complex and cooperate to induce cell cycle arrest. Cell 79, 119–130.[CrossRef][Medline]
Dupin, N., Fisher, C., Kellam, P. & 10 other authors (1999). Distribution of human herpesvirus-8 latently infected cells in Kaposi's sarcoma, multicentric Castleman's disease, and primary effusion lymphoma. Proc Natl Acad Sci U S A 96, 4546–4551.
el-Deiry, W. S., Tokino, T., Velculescu, V. E. & 7 other authors (1993). WAF1, a potential mediator of p53 tumor suppression. Cell 75, 817–825.[CrossRef][Medline]
Ellis, M., Chew, Y. P., Fallis, L., Freddersdorf, S., Boshoff, C., Weiss, R. A., Lu, X. & Mittnacht, S. (1999). Degradation of p27Kip cdk inhibitor triggered by Kaposi's sarcoma virus cyclin–cdk6 complex. EMBO J 18, 644–653.[CrossRef][Medline]
Endicott, J. A., Noble, M. E. & Tucker, J. A. (1999). Cyclin-dependent kinases: inhibition and substrate recognition. Curr Opin Struct Biol 9, 738–744.[CrossRef][Medline]
Ensoli, B. & Sturzl, M. (1998). Kaposi's sarcoma: a result of the interplay among inflammatory cytokines, angiogenic factors and viral agents. Cytokine Growth Factor Rev 9, 63–83.[CrossRef][Medline]
Fotedar, R., Fitzgerald, P., Rousselle, T., Cannella, D., Doree, M., Msieser, H. & Fotedar, A. (1996). p21 contains independent binding sites for cyclin and cdk2: both sites are required to inhibit cdk2 kinase activity. Oncogene 12, 2155–2164.[Medline]
Friborg, J., Jr, Kong, W., Hottiger, M. O. & Nabel, G. J. (1999). p53 inhibition by the LANA protein of KSHV protects against cell death. Nature 402, 889–894.[Medline]
Fujimuro, M., Wu, F. Y., ApRhys, C., Kajumbula, H., Young, D. B., Hayward, G. S. & Hayward, S. D. (2003). A novel viral mechanism for dysregulation of beta-catenin in Kaposi's sarcoma-associated herpesvirus latency. Nat Med 9, 300–306.[CrossRef][Medline]
Geley, S., Kramer, E., Gieffers, C., Gannon, J., Peters, J. M. & Hunt, T. (2001). Anaphase-promoting complex/cyclosome-dependent proteolysis of human cyclin A starts at the beginning of mitosis and is not subject to the spindle assembly checkpoint. J Cell Biol 153, 137–148.
Geng, Y., Yu, Q., Whoriskey, W. & 10 other authors (2001). Expression of cyclins E1 and E2 during mouse development and in neoplasia. Proc Natl Acad Sci U S A 98, 13138–13143.
Geng, Y., Yu, Q., Sicinska, E. & 7 other authors (2003). Cyclin E ablation in the mouse. Cell 114, 431–443.[CrossRef][Medline]
Gill, P. S., Tsai, Y. C., Rao, A. P., Spruck, C. H., III, Zheng, T., Harrington, W. A., Jr, Cheung, T., Nathwani, B. & Jones, P. A. (1998). Evidence for multiclonality in multicentric Kaposi's sarcoma. Proc Natl Acad Sci U S A 95, 8257–8261.
Godden-Kent, D., Talbot, S. J., Boshoff, C., Chang, Y., Moore, P., Weiss, R. A. & Mittnacht, S. (1997). The cyclin encoded by Kaposi's sarcoma-associated herpesvirus stimulates cdk6 to phosphorylate the retinoblastoma protein and histone H1. J Virol 71, 4193–4198.[Abstract]
Gu, Y., Turck, C. W. & Morgan, D. O. (1993). Inhibition of CDK2 activity in vivo by an associated 20K regulatory subunit. Nature 366, 707–710.[CrossRef][Medline]
Hagting, A., Jackman, M., Simpson, K. & Pines, J. (1999). Translocation of cyclin B1 to the nucleus at prophase requires a phosphorylation-dependent nuclear import signal. Curr Biol 9, 680–689.[CrossRef][Medline]
Harbour, J. W. & Dean, D. C. (2000). The Rb/E2F pathway: expanding roles and emerging paradigms. Genes Dev 14, 2393–2409.
