sign in sign up
<<
Home , Corepressor, Cyclin A, E2F, E2F2, E2F3, E2F4

Oncogene (2006) 25, 5220–5227 & 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00 www.nature.com/onc REVIEW RB and progression

C Giacinti1,2 and A Giordano1,2

1Sbarro Institute for Research and Molecular Medicine, Temple University, Philadelphia, PA, USA and 2Department of Human Pathology and , University of Siena, Siena, Italy

The Rb is a tumor suppressor, which plays a of a tumor suppressor. Several human tumors show pivotal role in the negative control of the cell cycle and in and deletions of the Rb , and inherited tumor progression. It has been shown that Rb protein allelic loss of Rb confers increased susceptibility to (pRb)is responsible for a major G1 checkpoint, blocking cancer formation (Dunn et al., 1988). The Rb gene is S-phase entry and . The family functionally inactivated in most human includes three members, Rb/p105, p107 and Rb2/p130, either by direct /deletion, such as in retino- collectively referred to as ‘pocket ’. The pRb , and small-cell lung , protein represses gene , required for transi- or indirectly through altered expression/activity of tion from G1 to , by directly binding to the upstream regulators (Liu et al., 2004). Nevertheless, domain of and by binding to the the Rb protein plays a pivotal role in the negative of these as a complex with E2F. pRb control of the cell cycle and in tumor progression. represses transcription also by remodeling It has been shown that Rb protein (pRb) is structure through interaction with proteins such as responsible for a major G1 checkpoint (restriction hBRM, BRG1, HDAC1 and SUV39H1, which are point) blocking S-phase entry and cell growth, promot- involved in remodeling, / ing terminal differentiation by inducing both cell cycle deacetylation and methylation, respectively. Loss of pRb exit and tissue-specific (Weinberg, functions may induce cell cycle deregulation and so lead to 1995). a malignant phenotype. Gene inactivation of pRB through The Rb gene family includes three members, Rb/p105, chromosomal mutations is one of the principal reasons p107 and Rb2/p130, collectively referred to as ‘pocket for retinoblastoma tumor development. Functional inacti- proteins’. The term ‘pocket proteins’ derives from the vation of pRb by viral oncoprotein binding is also shown in conserved binding pocket region through which pRb, many neoplasias such as cervical cancer, p107 and Rb2/p130 bind viral oncoproteins and cellular and AIDS-related Burkitt’s lymphoma. factors such as the E2F family of transcription factors. Oncogene (2006) 25, 5220–5227. doi:10.1038/sj.onc.1209615 Further studies have shown that overexpression of all three pocket proteins in cells can induce growth arrest in Keywords: Rb;cell cycle;E2F;cancer;Rb2/p130;tumor the of the cell cycle (Dyson, 1998). These suppressor studies helped define the role of the Rb gene family in regulating transition between cell proliferation and terminal differentiation. How pRb family members control cell cycle proliferation is not completely under- stood. Moreover, the interaction between the pRb Introduction family proteins and the E2F family transcription factors plays a central role in governing cell cycle progression The Rb gene is an archetypal and DNA replication by controlling the expression of that was first identified in a malignant tumor of the cell cycle E2F-dependent genes (Macaluso et al., 2005). retina known as retinoblastoma. In addition, pRb recruits factors Retinoblastoma is a sporadic or hereditary pediatric such as 1 (HDAC1) (Luo et al., arising from retinal cells, and Knudson (1971) 1998), SWI/SNF factors (Harbour and Dean, 2000; hypothesized that the tumor phenotype is not apparent De Luca et al., 1997), Polycomb group proteins unless both copies of the gene are damaged. The cloning (Dahiya et al., 2001) or methyltransferase (Nielsen of the retinoblastoma Rb gene and the identification of et al., 2001) that act on the nearby surrounding biallelic Rb mutations in retinoblastoma tumors confirm nucleosome structure. the hypothesis that such exerts the action The pRb protein has also a central role in various differentiation processes, including eye, lens, brain, peri- pheral nervous system, epidermis, melanocytes, hair cells, Correspondence: Dr A Giordano, Sbarro Institute for and Molecular Medicine, Temple University, Philadelphia, muscle and liver. For example, pRb is necessary for the PA 19122, USA. completion of the muscle differentiation program and E-mail: [email protected] for myogenic helix–loop–helix-dependent transcription RB and cell cycle progression C Giacinti and A Giordano 5221 (reviewed by De Falco, in this issue). The most from Rb family proteins. Rb proteins can be phos- convincing evidence of the importance of pRb in cellular phorylated both on A/B domains of the pocket region differentiation comes from studies of Rb knockout mice, and on the C-terminal domain. The main hypothesis is where the disruptions of the Rb gene cause death by day that the at the carboxy terminus of 14 of gestation, associated with defects in the develop- pRb by D/cdk4 (or ckd6) causes a conformational ment of the hematopoietic system and central nervous modification, which displaces the HDAC bound to pRb system (Lee et al., 1992). through its LXCXE motif. This facilitates the phos- phorylation of the A/B pocket region by (A)/ cdk2, leading to the release of E2F from pRb/p105. Pocket protein phosphorylation is often reversed by Rb family proteins dephosphorylation, which causes transient inactivation. Pocket protein dephosphorylation is known to take The Rb/p105 and Rb2/p130 map to human chromo- place in the from to G1 and to depend somal areas 13q14 and 16p12.2, respectively (Baldi upon protein phosphatase 1 (Ludlow et al., 1993); et al., 1996), in which mutations have been found in dephosphorylation occurs in response to growth in- several human neoplasias. The p107 maps to chromo- hibitory signals. In some situations, phosphorylation somal area 20q11.2, which is not frequently found to be can cause permanent inactivation, for example, by altered in human neoplasia, although there is a recent binding (Tedesco et al., 2002) and leading to report (Ichimura et al., 2000) of deletion or functional degradation. High levels of Cdk activity may result in inactivation for p107 in human tumors. The Rb family phosphorylation of pRb-Ser567, which exposes the Rb proteins share large regions of homology, especially in a protein to a proteolytic cleavage site and degradation by bipartite region, the so-called ‘pocket domain’ (Claudio an unidentified protease (Ma et al., 2003). et al., 2002). Rb2/p130 and p107 are more related to The C-terminal region of pRb proteins also plays an each other, with about 50% amino-acid identity, than important role in orchestrating the activity of these they are to Rb/p105 (30–35% identity). The Rb family proteins. The carboxy-terminal region differs in length proteins show a peculiar steric conformation owing to between Rb/p105 and the other two proteins Rb2/p130 the pocket region, which is responsible for many of the and p170, which are instead very similar in the amino- protein–protein interactions in the homeostasis of the acid sequence. As described above, analysis of structural cell cycle (Paggi et al., 1996). The pocket region is similarities among the Rb family members discloses that characterized structurally by two conserved functional p107 and Rb2/p130 are more strictly related to each domains (A and B) separated by a spacer (S), which other than to Rb. The carboxy-terminal region is quite substantially differs among the three Rb proteins, Rb/ important as it contains the nuclear localization signal p105, Rb2/p130 and p107. Members of the E2F family (NLS) and domains responsible for HADC1 and cyclin/ of transcription factors, which are cellular Rb-binding cdk complex binding. The NLS controls the transport of proteins, interact with the A and B pocket domains Rb proteins from the cytoplasm to the nucleus; (Puri et al., 1999), as well as proteins that possess the it consists of two cluster basic residues separated LXCXE motif, such as oncoproteins from by a stretch of amino acids including proline residues several DNA tumor viruses (Lee et al., 1998), the (Dingwall et al., 1991). Nuclear localization signal D-type (Weinberg, 1995) and HDAC. Rb proteins works as a carrier also for and proteins, can bind E2F and several other LXCXE-containing both of which do not present the NLS. proteins at the same time because the two binding sites It has been demonstrated that HDAC1 needs both the for E2F peptide and LXCXE peptide are close but LXCXE-binding motif of the pocket region and the distinct from each other (Lee et al., 1998;Gallo and C-terminus region in order to form a stable complex Giordano, 2005). (Stiegler et al., 1998) with Rb proteins. The binding sites Rb family members are post-translation regulated for E2F and HDAC1 are distinct from each other (Lee proteins and their gradual phosphorylation leads to et al., 1998), so we can hypothesize that Rb works as a their functional inactivation;Rb protein phosphoryla- link between HDAC1 and E2F proteins when associated tion is a cell cycle-dependent phenomenon and cyclin/ to E2F-responsive promoters. cyclin-dependent kinase (cdk) complexes are responsible for this regulation. In the early G1 phase, D-type cyclins interact with the A/B pocket of the Rb gene products via their LXCXE motif;in particular, they couple with the Rb and cell cycle kinase cdk4 or cdk6 and phosphorylate pRb/105 and Rb2/p130 (Dowdy et al., 1993). In the middle to late Two of the most important proteins involved in the cell G1 phase, cyclins E and A form complexes with cdk2, cycle machinery are cyclin-dependent kinases and which specifically target Rb/p105, Rb2/p130 and p107 cyclins. A variety of cyclin/cdk complexes are in fact (De Luca et al., 1997). This step is crucial for the able to guide the cdks to appropriate substrates and inactivation of members of the pRb family and the activate their catalytic activity (Figure 1). Cyclin/cdk release of E2F factors;however, it is still not evident complexes are formed during distinct phases of the cell how phosphorylation could induce a conformational cycle, and are specifically involved in the phosphoryla- change to allow physical dissociation of E2F factors tion of a distinct set of target proteins. The cyclin/cdk

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5222 G0 and complete its cell cycle. This point is termed ‘’ (R point);it is a central event in CyclinB CyclinC normal cellular proliferation control (Lundberg et al., cdk1 M cdk3 1999). It has been demonstrated that pRb is the Rb E2F molecular device that serves as the R point switch. pRb is hypophosphorylated in resting G0 cells, is G1 CyclinD increasingly phosphorylated during progression through G2 cdk4/6 G1 and is maintained in a hyperphosphorylated state CyclinA until late (Weinberg, 1995;Claudio et al., 1996, 2002). pRb phosphorylation seems to be related to cdk1 p p CyclinE mitogenic signals, which converge on the cell cycle S Rb cdk2 machinery, represented by the /cdk4 (cdk6) E2F complex in the early and mid-G1, and composed of cyclin E/cdk2 in late G1. cdk2 Like pRb, p107 is phosphorylated by the cyclin D1– Figure 1 Rb and cell cycle machinery. Rb and Rb-p represent the cdk4/6 complex activity, whereas Rb2/p130 phosphory- unphosphorylated and the phosphorylated forms of the retino- lation depends on the –cdk4 complex activity. blastoma protein. In G0 and early G1, Rb physically associates All of the three pocket proteins are phosphorylated by with E2F factors and blocks their transactivation domain. In late the cyclin E/cdk2 complex, and Rb2/p130 and p107, but G1, Rb-p releases E2F, allowing the expression of genes that products necessary for S-phase progression. not Rb/p105, are phosphorylated by cyclin A–cdk2. Pocket proteins may also regulate the earlier transition from G0 to G1 (Cobrinik, 2005). In G0, the RNA content is at low levels and hypophosphorylated pocket complexes orchestrate the advance of the cell through proteins contribute to the G0 state by repressing different phases of its growth cycle. Mammalian G1 ribosomal RNA and tRNA gene expression. Ren and cyclins D and E mediate progression through G1/ Rollins (2004) recently found that the cyclin C/cdk3 S phases. Three D-type cyclins exist (cyclin D1, D2 and complex is able to increase the RNA content while D3), which are expressed differently in various cell mediating G0 exit by phosphorylating pRb in the G0 lineages, with most cells expressing cyclin D3 and either state. As expression of cyclin C precedes that of the D1 or D2. Two types of cyclin E (E1 and E2) exist, , in keeping with these considerations, the pRb which show overlapping expression patterns in mouse phosphorylation by cyclin C–Cdk3 mediates the G0–G1 tissues and can be co-overexpressed in human tumors transition. (Geng et al., 2001). Mitotic cyclins A and B mediate When in its actively growth-suppressing hypophos- progression through the S/G2/M phases. is phorylated state, pRb physically associates with E2F expressed in and early embryogenesis, whereas factors and blocks their ability to activate expression of is found in proliferating somatic cells. Cyclin genes that encode products necessary for S-phase B2 probably plays a role in Golgi remodeling during progression. mitosis (Jackman et al., 1995), whereas The Rb proteins repress gene transcription, required controls other functions of this cyclin type. Until now, for transition from G1 to S phase, by directly binding to at least 16 cyclins have been discovered, for many of the transactivation domain of E2F and by binding to the which binding partners and functions have yet to be promoter of these genes as a complex with E2F (Adams identified. et al., 1995). Pocket proteins regulate G1–S transition D-type cyclins are short-lived proteins whose synthe- also through E2F-independent mechanisms: (1) p107 sis and assembly with Cdk4 or Cdk6 in G1 is dependent and p130 bind and inhibit the cyclin E/cdk2 and cyclin on mitogenic signaling (Hitomi and Stacey, 1999). A/cdk2 kinases and (2) pRb inhibits cdk activity and Cyclin D/Cdk activity goes on through the first and G1–S progression by increasing the expression of p27 subsequent cycles as long as mitogenic stimulation and stabilizing the p27 protein by binding the Skp2 continues. Cyclin E protein levels peak at the G1/S protein and interfering with the Skp2–p27 complex, thus progression, followed by an increase in cyclin A levels in avoiding p27 ubiquitination. Progression of the cells the S phase. Both cyclin E and A interact with and through G1 and S phases requires inactivation of Rb activate Cdk2, whereas cyclin A can also bind Cdk1 protein phosphorylation. The phosphorylation status of (Lees et al. 1993). At the G2/M boundary, levels these proteins is regulated by Cdk inhibitor (CKI) increase, resulting in activation of its partner, Cdk1. binding to the cyclin/cdk complexes (Sherr et al., 1995). This fluctuation in cyclin expression and the resultant The CKI inhibitors p15, and p17 specifically act on oscillation in Cdk activity form the basis of a cyclin D kinase activity, whereas , p27 and p57 act coordinated cell cycle progression. on all other cyclin/cdk complexes. It has been reported The cell responds to mitogenic stimuli and decides to that both cyclin D1 overexpression and related CKI advance through the various phases of the cell cycle only inhibitor alterations produce persistent hyperphosphory- during a limited phase of its cycle. In fact, the cell needs lation of pRb, resulting in cell cycle arrest. stimulation only during the first two-thirds of its Extracellular physiological signals induce the phos- G1 phase where it may decide to continue its advance phorylation of Rb protein, affecting the decision to

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5223 transit the R point. directly induce rapid group proteins (Dahiya et al., 2001) or methyltransfer- expression of cyclin D, which begins to inactivate pRb ase (Vandel et al., 2001). through its ability to associate with its previously cited cdk4 (cdk6) partners (Matsushime et al., 1991). Each type of cyclin D could receive different upstream signals Binding to E2F transcription factors that converge on it, so that we can say that the control The E2F family of transcription factors consists of at of cell proliferation in various cell types, known to be least seven E2F family members, , , , exercised by different mitogens, is modulated via E2F4, E2F5, E2F6 and E2F7. expression of a distinct D-type cyclin gene, which is A functional E2F consists of a under the control of a distinct transcriptional promoter. heterodimer containing an E2F polypeptide and a DP The common point is that all these different signals polypeptide (Helin et al., 1993). Each of the E2F converge to a common target, which is pRb. polypeptides can heterodimerize with either DP1 or There is no evidence that mitogenic stimuli directly DP2 (DP3) (Ormondroyd et al., 1995), although the modulate the levels of cyclin E or cyclin/cdk2 complex. E2F7 transcription factor binds DNA in a DP- Extracellular signals can negatively affect the cyclin/cdk independent manner. machinery as well. Serum starvation leads to a collapse There are functional and structural differences within of cyclin D levels and activity and to an increase of the E2F family. E2F1, E2F2 and E2F3 are transcrip- specific CKI (Resnitzky et al., 1994). tional activators, which interact with pRB. E2F4 and Rb activity in cell cycle machinery can also be E2F5 are transcriptional and mostly bind regulated at the transcriptional and protein levels by Rb2/p130 and p107 (Gaubatz et al., 2000). E2F6 seems the transcription factor ICBP90 (Hopnefer, 2000). not to have a pocket protein interaction domain;it ICBP90 (inverted CCAAT box binding protein of interacts with Polycomb proteins and represses trans- 90 kDa) is a transcriptional regulator of the topoiso- cription (Morkel et al., 1997). The recently identified merase II alpha gene and it has also two consensus binding E2F7 and E2F8 are also believed to repress specific sites for pRb in its primary sequence. The promoter of the promoters (Di Stefano et al., 2003). The DP hetero- Rb gene contains several putative binding sites for dimerization partner does not seem to determine the ICBP90. Overexpression of ICBP90 induces down- binding specificity of E2F factors for the pocket regulation of pRb expression in lung fibroblasts, as a proteins. However, DP factors do function in stabilizing result of mRNA decrease, and increases both the S- and the interaction between E2F factors and the Rb proteins G2/M-phase cell fractions of fibroblasts and the (Helin et al., 1993). topoisomerase IIalpha expressions. DNA chromatin During the cell proliferation response, there is a immunoprecipitation experiments show that ICBP90 different expression of the various E2Fs, presenting an binds to the Rb gene promoter under its methylated increase in E2F3 in early-G1 to mid-G1 and with E2F1 status. These results show that ICBP90 regulates Rb at and E2F2 increasing at the G1/S boundary (Johnson the protein and gene transcription levels, thus favoring et al., 1998). E2F4 and E2F5 are expressed throughout the entry into the S phase of the cells, and we can the cell cycle, but in G0 and early G1, they are bound to hypothesize that ICBP90 overexpression, found in the nucleus by p107 and Rb2/p130, forming transcrip- cancer cells, could be involved in the altered checkpoint tional complexes. p107 and Rb2/p130 recruit controls occurring in (Jeanblanc, 2005). E2F4 into the nuclei of the cell, inducing the repression of S-phase genes transcription and cell proliferation. Meanwhile, pRb is believed to bind E2F1–3 either at or sequestered away from E2F-responding promoters (De How does Rb control cell proliferation? La Luna et al., 1996). After stimulation, quiescent cells enter the cell cycle, How Rb proteins control cell proliferation is not and in the early-G1 phase, pRb is phosphorylated and completely understood. It has been established that consequently E2F1–3 transcriptional factors are re- RB protein associates with a wide variety of transcrip- leased, whereas only in mid-G1 to late G1, we can tion factors and chromatin-associated complexes. observe the loss of Rb2/p130 complexes. Therefore, in Rb protein binding inhibits the transcriptional activa- the late G1 phase, Rb proteins are phosphorylated and tion ability of E2F factors, masking the transcriptional dissociate from E2Fs;E2F4 and E2F5 move to the activation domain and, in some cases, converting E2F cytoplasm and E2F1–3 bind to the E2F-responding factors to transcription repressors. The importance of promoters in sites that often are different from the the Rb–E2F interaction in cell growth control is binding site of E2F repressor proteins (Cinti et al., demonstrated by the finding that all naturally occurring 2000). At the G1/S-phase transition, complexes contain- Rb mutants isolated from human tumors lack the ability ing p107 in association with E2F4 can still be detected to bind and negatively regulate E2F (Qian et al., 1992). (Mudryj, 1991). These E2F4–p107 complexes also Rb proteins are thought to inhibit expression of E2F- contain cyclin E or A and cdk2 (Shirodkar, 1992). regulated genes in two ways (Dyson et al., 2002): by E2F–p130 complexes have also been shown to accumu- directly binding and blocking the activation domain of late, as some cell types, including myoblasts and E2F proteins or by active repression through the melanocytes, undergo terminal differentiation (Shin recruitment of HDAC, SWI/SNF factors, Polycomb et al., 1995).

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5224 pRb and the related proteins p107 and Rb2/p130 et al., 1995), which are respectively involved in nucleo- do not have a fully redundant function in regulating some remodeling, histone acetylation/deacetylation and E2F genes transcription;instead they show a great methylation. redundancy at the cell cycle level. For example, in hBRM and BRG1 are mammalian homologs of pRbÀ/À fibroblasts, there is a compensatory increase in components of the yeast chromatin remodeling complex p107 protein levels during the arrest in G0/G1 phase SNF2/SWI2 (Strober et al., 1996). Both BRG1 and (Sage et al., 2003). The mechanism of this antiprolifera- hBRM have been shown to associate with pRB and they tive effect of p107 is unknown – p107 could substitute are implicated in pRb-mediated repression. A physical pRb in the regulation of E2F1–3 or it may repress some interaction between pRb and BRG1 is not required for E2F-responsive genes that are controlled by pRb (Lee pRB-induced growth arrest and transcriptional repres- et al., 2002). Furthermore, pRb can also bind E2F4 sion of E2F target genes (Kang et al., 2004). in vitro but fails to regulate the E2F-responsive Histone deacetylase 1 is an HDAC that has recently promoter in p107À/À and p130À/À cells. It has been been shown to be recruited to E2F complexes by pRb shown that the pRb–E2F4 complex, in p107À/À cells, and to function in repressing cyclin E gene expression fails to bind the same promoter sites that are usually (Magnaghi-Jaulin et al., 1998). are generally linked by the p107–E2F4 complex in wild-type cells hyperacetylated at the promoters of actively transcribed (Rayman et al., 2002). genes but are hypoacetylated at silenced genes. Histone However, pRb is involved in the repression of a deacetylase is generally recruited on gene promoters by limited set of E2F genes in G0 and G1 cells, and recently many transcriptional regulators. pRB associates with its role in cyclin E gene repression has been defined. The HDAC activity in vivo and binds to class I HDACs cyclin E promoter is in fact deregulated in cells that have (HDAC1–HDAC3) in vitro. pRB-mediated repression is lost pRb, even though the p107/p130–E2F4 complex is enhanced by HDAC1 in transient transfection experi- still available (Herrera et al., 1996). pRb mediates the ments, and to confirm this association between pRb and repression of cyclin E promoter by the recruitment of HDAC, it has been shown that inhibition of HDAC HDAC and histone methyltransferase on a particular activity with trichostatin A interferes with pRB- E2F-SP1 site that is exclusively linked by the pRb– mediated repression at a subset of E2F-regulated E2F1–3 complexes. Cyclin E is derepressed not only by promoters (Zhang et al., 2000). Histone deacetylase is the loss of pRb but also by the loss of E2F1–3 proteins, thus involved in modulating the repressive activity of so it is clear that E2F1–3 proteins act in this case as pRb on E2F gene promoters. The recruitment of HDAC repressors instead of activators (Wu et al., 2001). p107 may be indirect and accompanied by recruitment of or Rb2/p130 can substitute pRb only if they are present other binding proteins such as RBP1, corepressors in a sufficient amount to bind E2F1–3 proteins (Lee and chromatin remodeling proteins (Lai et al., 1999). et al., 2002). Rayman et al. (2002) have shown that in RbÀ/À cells, pRb can inhibit proliferation through an E2F- HDAC is on E2F gene promoters;this allowed us to independent mechanism. For example, pRb can induce hypothesize that pRb is not strictly necessary in E2F the formation of promyelocytic (PML) nuclear transcriptional repression, but that other mechanisms bodies (Fang et al., 2002), increase p27 expression (see could be involved. Histone acetyltransferase proteins above), suppress Ras signaling (Lee et al., 2002) and (CCBP, p300, Tip6, etc.) participate in the activation of cooperate with hLin-9, a protein that is involved, similar E2F proteins (Condorelli and Giordano, 1997);they to pRb, in suppressing transformation but in an E2F- interact with the activation domain of E2F proteins and independent manner (Gagrica et al., 2004). stimulate their activation. RB/E2F repressor complexes can also be regulated in SUV39H1 is a methyltransferase that specifically a Cdk-independent manner. For example, transforming methylates K9 of histone H3 and cooperates with pRb -b recruits E2F4/p107 and E2F5/p107 by in the repression of the promoter of E2F-responding the c- promoter irrespective of cell phase, and it is genes. In cells lacking pRB, there is no association of still stable in the presence of high activity of Cdk (Chen me-K9-H3 in these promoters (Rea et al., 2000);in et al., 2004). In human cells, it has been noticed that particular, these results have been developed by studying p107 and p130 are removed from cell cycle-regulated cyclin E, one of the best E2F target genes modulated by gene promoters during S phase, but are still located at pRb. Both pRb and SUV39H1 are directly involved in the promoters of apoptotic genes (Young et al., 2004). repressing cyclin E transcription, as in both pRbÀ/À The regulation of RB/E2F complex in a Cdk-indepen- and SUV39H1À/À double knockout cells, cyclin E is dent manner is not yet well understood. overexpressed. Rb proteins are also able to prevent the assembly of pre-initiation complexes, inhibiting the activity of transcription factors recruited on E2F gene Binding to proteins that modify chromatin structure promoters (Ross et al., 1999). Repression by Rb proteins requires the conserved A and The type of repression that Rb proteins exert on cell B domains of the pocket protein and appears to involve growth depends on the type of corepressor proteins that the association of other proteins in addition to E2F bind E2F target genes. For example, different types of factors. Rb proteins repress transcription by remodeling transcriptional repression can operate at the same chromatin structure through interaction with proteins promoter in a different cellular state or in a different such as hBRM, BRG1, HDAC1 and SUV39H1 (Shao cell phenotype (Dimova and Dyson, 2005).

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5225 The pRb/E2F complex can also promote stable could be functionally inactivated during HIV infection repression independently of cell cycle position or by (Cinti et al., 2000;De Falco et al., 2003;Lazzi et al., being resistant to cell cycle progression. Unfortunately, 2002;Cinti and Giordano, 2000). the mechanisms through which pRb is involved in In cervical cancer, the inactivation of pRb occurs processes that permanently repress E2F-dependent through the binding of the E7 oncoprotein, which is transcription are not well known until now. For associated with over 90% of these cancer cases (Kim example, E2F4 can be found at E2F-regulated promo- et al., 2005;Cobrinik et al., 1993). ters in S phase, but it is unclear whether it acts as an Inappropriate pRb phosphorylation due to rearran- activator. Senescent cells, for example, show a higher gement and overexpression of cyclin D1 also contributes level of H3K9 methylation in E2F-responsive promoters to the development of several types of human tumors, in comparison to quiescent cells (Narita et al., 2003). Rb including parathyroid , B-cell lymphomas and proteins mediate E2F-responsive genes both in a squamous cell (Hunter et al., 1994). The quiescent and senescent/differentiated state, the former overproduction of cyclin D1 and its partner cdk4 leads being characterized by low levels of H3K9 methylation, to a constitutive phosphorylation of Rb proteins and to and the latter by high levels of H3K9 methylation. The a deregulation of E2F transcriptional activity. It has identity of histone methyltransferase that is recruited on also been reported that deregulation of other G1 cyclins the promoter can also modulate the transition between a (D2, D3 and E) is associated with tumorigenesis quiescent cell and a senescent/differentiated cell. (Keyomarsi et al., 1995). The finding that the cyclin D/cdk inhibitor, p16/ INK4A, is lost in several different tumor types (for Rb and cancer example, breast carcinoma, hepatocellular carcinoma, etc.) suggests that the altered activity of cyclin/cdk Loss of pocket protein functions may induce cell cycle complexes in G1 phase can be oncogenic (Hunter et al., deregulation and lead to a malignant phenotype. Rb 1994). p16/INK4 controls Rb activity by binding cdk4/6 protein can be functionally inactivated by phosphory- and inhibiting the action of cyclin D;in this way, p16/ lation, mutations or viral oncoprotein binding. As a INK4 promotes the inhibition of pRb phosphorylation direct consequence, E2F transcription factors are and the formation of the pRb/E2F repressor complex. liberated by control of Rb protein and induce deregula- Loss of function of the kinase inhibitor enhances cyclin tion of the cell cycle. D/cdk4 kinase activity, which promotes hyperphosphor- Loss of heterozygosity of tumor suppressor genes is ylation of pRb that is no more able to regulate cell cycle observed in a variety of human tumors, including checkpoints. , and small-cell lung For example, most of pRb-positive cell carcinomas. In retinoblastoma, a malignant childhood lines are p16/INK4-negative. This would indicate that tumor, pRb is functionally inactivated by mutations or the pathogenesis of some lung occurs by the partial deletions (Bignon et al., 1993). Bilateral familial absence of p16 inhibitor, which functions to keep pRb retinoblastomas show in pRb, and hypophosphorylated (Caputi et al., 2005). The main role tumors develop when both the alleles are mutated of pRb in lung cancer is also evaluated by the notion (pRbÀ/À). These data are supported by studies on that in small lung cancer cells, the frequency of deletion animals in which heterozygous mutation of pRb or mutation of pRb is 90% (Slagia et al., 1998). (pRb þ /À) is not sufficient for the development of the In , no pRb rearrange- tumor, whereas pRb þ /À and p107À/À animals develop ments have been found, whereas Rb2/p130 results in a retinal , showing that p107 is able to prevent drastic reduction in the expression level of both this type of tumor in pRbÀ/À mice and so it acts as a nasopharyngeal cell line and in primary tumors owing tumor suppressor (Lee et al., 1996). Gene inactivation to a mutation in the C-terminal domain of the through chromosomal mutations is one of the principal oncosuppressor protein (Claudio et al., 2000). reasons for retinoblastoma loss of function in cancer. We can summarize that a breakdown of cell cycle, Other insults can act on pRb altering cell cycle owing to products of oncogenes (cyclins, cdks and growth. For example, in tumor cells, pRB/E2F interac- ) and tumor suppressor genes (Rb proteins tion is broken by pRb binding to DNA tumor virus and CKI) whose functions converge on alteration of oncoprotein such as virus E7 and displacing E2F genes, leads to a cancer phenotype. E2F transcription factors. Functional inactivation by viral oncoprotein binding is observed in many neopla- Acknowledgements sias such as cervical cancer, mesothelioma and AIDS- related Burkitt’s lymphoma. For example, in AIDS- This work is supported in part by grants from NIH and Sbarro related Burkitt’s lymphoma, Rb2/p130 and p107 are Health Research Organization. We thank Simone C, De Falco expressed at high levels, suggesting that these proteins G and Bagella L for their critical comments.

References

Adams PD, KAelin WG. (1995). Semin Cancer Biol 6: Baldi A, Boccia V, Claudio PP, De Luca A, Giordano A. 99–108. (1996). Proc Natl Acad Sci 93: 4629–4632.

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5226 Bignon YJ, Rio P. (1993). Bull Cancer 80: 704–712. Kim YT, Zhao M. (2005). Yonsei Med J 31: 597–613. Caputi M, Russo G, Esposito V, Mancini A, Giordano A. Knudson AG. (1971). Proc Natl Acad Sci USA 68: 820–823. (2005). J Cell Physiol 205: 319–327. Lai A, Lee JM, Yang WM, De Caprio JA, Keelin WG et al. Chen D, Livne-bar I, Vanderluit JL, Slack RS, Agochiya M, (1999). Mol Cell Biol 19: 6632–6641. Bremner R. (2004). 5: 539–551. Lazzi S, Bellan C, De Falco G, Cinti C, Ferrari F et al. (2002). Cinti C, Giordano A. (2000). Emerg Ther Targets 4: 765–783. Hum Pathol 33: 723–731. Cinti C, Leoncini L, Nwongo A. (2000). Am J Pathol 156: Lee EY, Chang CY, Hu N, Wang YC, Lai CC, Herrup K et al. 751–760. (1992). Nature 359: 288–294. Claudio PP, De Luca A, Howard CM, Baldi A, Firpo Ej, Koff Lee EY, Cam H, Ziebold U, Rayman JB, Lees JA, Dynlacht A et al. (1996). Cancer Res 56: 2003–2008. BD. (2002). Cancer Cell 2: 463–472. Claudio PP, Tonin T, Giordano A. (2002). Genome 3: 3012. Lee JO, Russo AA, Pavletich NP. (1998). Nature 391: 859–865. Claudio PP, Howard CM, Fu Y. (2000). Cancer Res 60: Lee MH, Williams Bo, Mulligan G. (1996). Genes Dev 16: 8–12. 1621–1632. Cobrinik D. (2005). Oncogene 24: 2796–2809. Lees W-H, Shew J-Y, Hong FD, Sery TW, Donoso Lees EM, Cobrinik D, White P, Peeper Ds, Jacks T, Weinberg RA. Harlow E. (1993). Mol Cell Biol 2: 1194–1201. (1993). Genes dev 7: 2393–2404. Liu H, Dibling B, Spike B, Dirlam A, Macleod K. (2004). Curr Condorelli G, Giordano A. (1997). J Cell Biochem 67: Opin Genet Dev 14: 55–64. 423–431. Ludlow JW, Glendening CL, Livingston DM, DeCarprio JA. Dahiya A, Wong S, Gonzalo S, Gavin M, Dean DC. (2001). (1993). Mol Cell Biol 13: 367–372. Mol Cell 8: 557–569. Lundberg AS, Weinberg RA. (1999). Eur J Cancer 35: De Falco G, Bellan C, Lazzi S, Caludio P, La Sala D, Cinti C 531–539. et al. (2003). Oncogene 22: 6214–6219. Luo RX, Postigo AA, Dean DC. (1998). Cell 92: 463–473. De Falco G, Comes F, Simone C. (2006). Oncogene, review in Ma D, Zhou P, Harbour JW. (2003). J Biol Chem 278: this issue. 19358–19366. De Luca A, MacLachlan JK, Bagella L, Dean C, Howard CM, Macaluso M, Montanari M, Cinti C, Giordano A. (2005). Claudio PP et al. (1997). J Biol Chem 272: 20971–20974. Semin Oncol 32: 452–457. De La Luna S, Burden Mj, Lee C-W, La Thangue NB. (1996). Magnaghi-Jaulin L, Groisman R, Naguibneva I, Robin P, J Cell Sci 109: 2443–2452. Lorain S, LeVillain JP et al. (1998). Nature 391: 601–604. Dimova DK, Dyson N. (2005). Oncogene 44: 30–33. Matsushime H, Roussel MF, Ashmun RA, Sherr CJ. (1991). Dingwall C, LAskey RA. (1991). Trends Biol Sci 16: 478–481. Cell 65: 701–703. DiStefano L, Jensen MR, Helin K. (2003). EMBO J 22: Morkel M, Wenkel J, Bannister AJ, Kouzarides T, Hagemeier 6289–6298. C. (1997). Nature 390: 567–568. Dowdy SF, Hinds PW, Louis K, Reed SI, Arnold A, Weinberg Mudryj M, Devoto SH, Hiebert SW, Hunter T, Pines J, RA. (1993). Cell 73: 499–511. Nevins JR. (1991). Cell 65: 1243–1253. Dunn JM, Phillips RA, Becker AJ, Gallie BL. (1988). Science Narita M, Nunez S, Heard E, Lin AW, Hearn SA, Spector DL 241: 1797–1800. et al. (2003). Cell 113: 703–716. Dyson N. (2002). Curr Opin Cell Biol 14: 684–691. Nielsen SJ, Schneider R, Bauer UM, Bannister AJ, Morrison Dyson N. (1998). Genes Dev 12: 2245–2262. A, O’Carroll D et al. (2001). Nature 412: 561–565. Fang W, Mori T, Cobrinik D. (2002). Oncogene 21: Ormondroyd E, de la Luna S, La Thangue NB. (1995). 5557–5565. Oncogene 11: 1437. Gallo G, Giordano A. (2005). J Cell Phys 2: 176–181. Paggi MG, Baldi A, Bonetto F, Giordano A. (1996). J Cell Gagrica S, Hauser S, Kolfschoten I, Osterloh L, Agami R, Biochem 62: 418–430. Gaubatz S. (2004). EMBO J 23: 4627–4638. Puri PL, Maclachlan TK, Levrero M, Giordano A. (1999) In: Gaubatz S, Lindeman GJ, Ishida S, Jakoi L, Nevins JR, Stein GS, Baserga R, Giordano A, Denhardt DT (eds). The Livingston DM et al. (2000). Mol Cell 6: 729–735. Molecular Basis of Cell Cycle and Growth Control. Wiley: Geng Y, Yu Q, Whoriskey W, Dick F, Tsai KY, Ford HI et al. New York, pp 15–79. (2001). Proc Natl Acad Sci USA 98: 13138–13143. Qian Y, Luckey C, Horton L, Esser M, Templeton DJ. (1992). Harbour JW, Dean DC. (2000). Cell 101: 79–89. Mol Cell Biol 12: 5363–5372. Helin K, Herlow E, Fattey A. (1993). Mol Cell Biol 13: Rayman JB, Takahashi Y, Indjeian VB, Dannenberg JH, 3149–3156. Catchpole S, Watson RJ et al. (2002). Genes Dev 16: Herrera RE, Tan VP, Williams BO, Makela T, Weinberg RA, 933–947. Jacks T. (1996). Mol Cell Biol 16: 2402–2407. Rea S, Eisenhaber F, O’Carroll D, Strahl BD, Sun ZW, Hitomi M, Stacey DW. (1999). Curr Biol 19: 1075–1084. Schmid Ren S et al. (2000). Nature 406: 593–599. Hopnefer R, Mousli M, Garnier JM, Redon R, du Manoir S, Ren S, Rollins BJ. (2004). Dev Cell 6: 607–608. Chatton B et al. 2000. Cancer Res 1: 121–128. Resnitzky D, Gossen M, Buiard H, Reed SI. (1994). Mol Cell Hunter T, Pines J. (1994). Cell 79: 573–582. Biol 14: 1669–1679. Ichimura K, Hanafusa H, Takimoto H, Ohgama Y, Akagi T, Ross JF, Liu X, Dynlacht BD. (1999). Mol Cell 3: 195–205. Shimizu K. (2000). Gene 251: 37–43. Sage J, Miller AL, Perez-Mancera PA, Wysocki JM, Jacks T. Jeanblanc M, Mousli M, Hopfner R, Bathami K, Martinet N, (2003). Nature 424: 223–228. Abbady et al. (2005). Oncogene 49: 7337–7345. Shao Z, Robbins PD. (1995). Oncogene 10: 221–228. Jackman M, Firth M, Pines J. (1995). EMBO J 14: 1646–1654. Sherr CJ, Roberts JM. (1995). Genes Dev 9: 1149–1463. Johnson DG, Schendeider-Broussard R. (1998). Front Biosci 3: Shin EK, Shin A, Paulding C, Schaffhausen B, Yee AS. (1995). D447–D458. Mol Cell Biol 15: 2252–2262. Kang H, Cui K, Zhao K. (2004). Mol Cell Biol 24: 1188–1199. Shirodkar S, Ewen M, DeCaprio JA, Morgan JA, Livingston Keyomarsi K, Conte Jr D, Toyofuku W, Fox MP. (1995). DM et al. (1992). Cell 68: 157–166. Oncogene 11: 941–950. Slagia R, Skarin AT. (1998). J Clin Oncol 16: 1207–1217.

Oncogene RB and cell cycle progression C Giacinti and A Giordano 5227 Stiegler P, De Luca A, Bagella L, Giordano A. (1998). Cancer Weinberg RA. (1995). Cell 81: 323–330. Res 58: 5049–5052. Wu L, Timmers C, Maiti B, Saavedra HI, Sang L, Chong GT Strober BE, Dunaief JL, Guha S, Goff SP. (1996). Mol Cell et al. (2001). Nature 414: 457–462. Biol 16: 1576–1583. Young AP, Longmore GD. (2004). Oncogene 23: 2587–2599. Tedesco D, Lukas J, Reed SI. (2002). Genes Dev 16: Zhang HS, Gavin M, Dahiya A, Postigo AA et al. (2000). Cell 2946–2957. 101: 79–89. Vandel L, Nicolas E, Vaute O, Ferreira R, Ait-Si-Ali S, Trouche D. (2001). Mol Cell Biol 21: 6484–6494.

Oncogene