Escape from Cellular Quiescence
Quiescent: From Latin quies, referring to a state of being at rest, dormant, inactive, quiet, still (Merriam-Webster, 2009, Online Dictionary: http://www.merriam-webster.com/dictionary/quiescent). This term refers to a state of dormancy as opposed to a proliferative state. However, quiescent cells are in any other regard metabolically active. In many tissues with relative fast cell renewal rates the primary function of a small group of undifferentiated cells is limited to self-renewal (stem cells). These cells remain quiescent most of the time dividing only occasionally. In other tissues, key cell types perform fundamental tissue functions while remaining quiescent. Both stem cells and cells from tissues that renew via simple duplication can remain quiescent for long periods of time while retaining the capacity to re-enter the cell cycle. This chapter will discuss the mechanisms emerging as responsible for the maintenance of quiescence as well as those pathways that mediate quiescence entry and exit. We will also review signaling pathways deregulated during infection by Simian Virus 40 (SV40) and oncogenic transformation, which result in unscheduled exit from quiescence into the cell cycle, with focus on SV40 small t antigen.
KeywordsQuiescent Cell Mitogenic Stimulation Normal Human Fibroblast Cell Cycle Exit Pocket Protein
1.1 Quiescence: The Reversible State
It is well established that the quiescent state is associated with an increase in the expression of the CDK inhibitor p27 (Sherr and Roberts, 1999). Unexpectedly, the study of the gene expression fingerprints that characterize quiescence has also revealed that quiescence is not equivalent to growth arrest induced via inhibition of CDKs. Cells ectopically expressing the p21/p27 CDK inhibitors exhibit a distinctive program of gene expression that includes a portion of the genes found downregulated by all quiescent signals mentioned above, but it does not induce upregulation of genes that suppress differentiation or inhibit apoptosis (Coller et al., 2006). In agreement with the observation that CKI inhibitors are upregulated during differentiation along particular lineages, overexpression of p21 in dermal fibroblasts induced growth arrest but did not prevent MyoD-induced differentiation. In contrast, fibroblasts forced into quiescence by contact inhibition or mitogenic withdrawal are resistant to differentiation signals (Coller et al., 2006). These results show that cellular quiescence is not a mere consequence of cell cycle exit but rather a unique resting state that preserves cells in environments that are not suitable for proliferation.
More recently, the mechanisms that control the reversibility of cellular quiescence have started to be unveiled. Because the transcriptional repressor Hairy and Enhancer of Split1 (HES1) is induced by signals that force fibroblasts into quiescence but is not regulated when cell cycle exit is induced by overexpression of CKIs (Coller et al., 2006), Sang et al. tested whether HES1 modulates the reversibility of cellular quiescence (Sang et al., 2008). Remarkably, it was found that ectopic expression of HES1 in dermal fibroblasts prevents p21-induced irreversible senescence, although it cannot reverse this phenotype if senescence is attained prior to HES1 expression. More importantly, their work also demonstrated that MyoD-induced differentiation of proliferating fibroblasts is prevented by ectopic expression of HES1 and that inactivation of HES1 in quiescent fibroblasts is sufficient to induce spontaneous senescence or trigger myogenic differentiation in response to MyoD activation. Thus, HES1 emerges as a pivotal candidate to control the reversibility of the quiescent state.
1.2 Overcoming the Restriction Point
1.2.1 The Restriction Point
In unicellularorganisms such as yeast, the availability of nutrients in the environment primarily determines their proliferation rate. In contrast, nutrients in the environment of cells in multicellular organisms are not typically limiting, and thus proliferation rates are determined by mitogens produced by other cells or by genetic developmental programs. The cell cycle can be subdivided in two functionally distinct parts based on their dependency on mitogens for cell cycle progression (Fig. 1.1). The mitogen-dependent phase spans the period of the cell cycle beginning with initiation of post-mitotic G1 to the Restriction point (R), which was first defined by Arthur Pardee (Pardee, 1974). Once cells surpass R, they are committed to a round of DNA replication and cell division, and the progression and continuity from one phase to the next depend solely on the cell’s efficiency to faithfully complete DNA replication, chromosomal segregation, cytokinesis, and other required intermediate steps. On the other hand, normal post-mitotic early G1 cells that encounter an environment with limiting mitogens, extracellular substrate attachment, or space, enter a reversible quiescence state.
The main challenge faced by a cell exiting quiescence is to synthesize de novo all the gene products required for successful cell cycle entry and passage through R. E2F transcription factors control the expression of many genes whose products are essential, or at least important, for cell cycle progression. In quiescent cells, repressor E2Fs (E2Fs 4 and 5) form complexes with pocket proteins (typically p130 and the retinoblastoma protein, pRB) which silence E2F-dependent gene expression (reviewed in Graña et al., 1998; Mulligan and Jacks, 1998; Blais and Dynlacht, 2004; Rowland and Bernards, 2006). Mitogens activate intracellular signaling pathways that trigger activation of G1 cyclin/CDK complexes, which in turn disrupt E2F/pocket protein complexes via phosphorylation of the pocket protein. It is thought that once the E2F-gene expression program is set in motion to warrant the expression of sufficient levels of DNA replication enzymes and other cell cycle proteins and regulators, cell cycle progression becomes insensitive to both positive and negative external mitogenic stimuli (Fig. 1.1).
1.2.2 G1-Cyclins/CDK, pRB, and E2F Transcription Factors
Since this book contains a chapter devoted to the interplay between CDKs and E2F-dependent transcription, the focus of this section will be restricted to the events important for quiescence exit/entry.
G1-cyclins, together with their catalytic partners, the CDKs, are the key effectors of mitogenic signaling that drive cells out of quiescence in propitious environmental conditions. There are three mammalian isoforms of cyclin D (D1, D2, and D3) that exhibit tissue-specific expression. D-type cyclins bind to CDK4 or CDK6 (CDK4-6) and are activated in mid-G1. E-type cyclins, E1 and E2, bind to CDK2 leading to its activation later in G1.
Overexpression of G1 cyclins is common in primary tumors and derived tumor cell lines (Malumbres and Barbacid, 2001). Considering that mitogenic signaling converges in the activation of G1 cyclin/CDK complexes, deregulation of G1 cyclins in tumor cells may reduce the threshold of mitogenic stimulation required for passage through the R or for escaping quiescence. In this regard, early studies showed that overexpression of either D1 or E shortens G1 phase upon mitogenic stimulation. However, quiescent primary non-transformed fibroblasts that ectopically express cyclin D1 and/or E do not exit quiescence if the environment is deprived of mitogens or if the cells are arrested by growth to high density (Ohtsubo and Roberts, 1993; Quelle et al., 1993; Resnitzky et al., 1994; Sotillo et al., 2008, 2009). In contrast, similar expression of cyclin D1 and E in certain tumor cell lines is sufficient to trigger exit from quiescence in the absence of any mitogenic stimulation (Calbó et al., 2002). Experiments performed in our laboratory have shown that in quiescent tumor-derived T98G cells forced expression of cyclin E leads to formation of active CDK2 complexes, pocket protein phosphorylation, and activation of the E2F program concomitantly with cell cycle entry. Under the same conditions expression of cyclin E in quiescent normal human fibroblasts (NHF) leads to formation of inactive complexes failing to trigger cell cycle entry. Concentrations of serum as low as 0.1% make quiescent NHF responsive to deregulated cyclin E expression, suggesting that other mitogen-dependent events, besides cyclin E accumulation, are required to fully activate CDK2 and exit G0. This is consistent with previous work showing that microinjection of active G1 cyclin/CDK complexes into the nucleus of primary human WI38 fibroblasts is sufficient to induce DNA synthesis (Connell-Crowley et al., 1998).
Despite the clear important role of G1 CDKs in mediating passage through R and triggering E2F-dependent gene expression, ablation of G1 CDKs and cyclins in mice has evidenced a high level of functional redundancy and compensation among these G1 cyclin/CDK complexes in triggering inactivation of pocket proteins and other essential events during the cell cycle (Malumbres and Barbacid, 2009). Targeted disruption of D-type cyclins, E-type cyclins, CDK4-6, or CDK2 reduces inactivation of pocket proteins but not below a threshold that could prevent E2F-dependent gene expression in both proliferating and serum starved and re-stimulated MEFs (Lee and Sicinski, 2006; Berthet and Kaldis, 2007; Malumbres and Barbacid, 2009). Indeed, even fibroblasts obtained from mouse embryos that simultaneously lack expression of CDK2, CDK4, and CDK6 proliferate and exit quiescence in response to serum stimulation (Santamaria et al., 2007). Thus, CDK1 via its binding with cyclin E appears sufficient to inactivate pocket proteins and induce passage through R. Of note, serum-starved cyclin E1–/–; E2–/– double knock-out MEFs are unable to re-enter cell cycle when stimulated with mitogens. However, this is due to a defect in loading of MCM2 onto chromatin, as pocket proteins are inactivated and E2F-dependent genes expressed (Geng et al., 2003). These results indicate great plasticity and compensation among cyclins and CDKs in many cell types, with function of some of them only essential in particular cell types.
1.2.3 Is Inactivation of Pocket Proteins Beyond a Certain Threshold Sufficient for Passage Through R?
Ablation of the three pocket proteins in MEFs makes these cells bypass cell cycle exit signals induced by mitogen withdrawal, contact inhibition, and loss of attachment, but cells become apoptotic (Dannenberg et al., 2000; Sage et al., 2000). Pocket protein mutant MEFs also fail to arrest in response to a variety of signals that cause G1 arrest. Thus, pocket proteins are important for establishment of a G1 growth arrest and exit into quiescence, as triple mutant MEFs fail to become quiescent in response to three independent signals that mediate reversible growth arrest. Ectopic expression of E2F1 drives quiescent rodent fibroblasts into S phase (Johnson et al., 1993). Activator E2Fs (E2F1-3) are likely required for cell cycle re-entry, while E2F-repressor activities are critical for contact inhibition induced cell cycle exit and other signals that induce G1 arrest (Rowland and Bernards, 2006). Therefore, there is firm evidence that pocket protein/E2F pathways play critical roles in cell cycle re-entry/exit, as loss of these pathways makes cells insensitive to extracellular control in both cycling and quiescent cells. It is important to stress that despite the high degree of compensation among pocket proteins, it has become clear that distinct complexes play specialized functions. An evolutionarily conserved complex designated DREAM has been identified in mammalian cells that contains p130, E2F4, and mammalian homologs of Caenorhabditis elegans synthetic multivulva class B (synMuvB) gene products, including LIN-9, LIN-37, LIN-52, LIN-53, and LIN-54 (Litovchick et al., 2007). The DREAM complex binds the promoters of cell cycle-regulated genes in serum starved quiescent human T98G cells and is required for their repression. It is yet to be known if other G0 and G1 arrest-inducing signals result in formation of the same DREAM complex, and whether assembly of this complex is associated with formation of the common gene expression signature that defines quiescence in NHF (Coller et al., 2006). Conceivably, E2F-repressor complexes formed in response to G1 arrest signals that do not result in cell cycle exit into quiescence may be different from the DREAM complex in one or several subunits. It is also possible that expression of HES1 in cells exiting into G0 is independent of the formation of DREAM, but it might play a role in defining the type of E2F/pocket protein complexes that assemble at cell cycle promoters.
1.2.4 What Are Cells Doing as They Exit Quiescence Back into G1?
The classical video-microscopy experiments of Zetterberg and Larsson defined the period of time between post-mitotic G1 and R in NIH 3T3 cells (Zetterberg and Larsson, 1985). In these experiments, removal of mitogens for 1 h affected cell cycle length of only very young post-mitotic cells progressing through early G1. Mitogen removal for 1 h in early G1 resulted in elongation of the cell cycle by as much as 8 h. Others have shown that the time that cells require to exit quiescence is proportional to the time that they have spent in this state (Owen et al., 1989). One logical explanation that was drawn from these early studies is that cells need time to de novo transcribe and translate the gene products required for passage to R. More recently, changes in the assembly of DNA pre-replication factors onto replication origins have also been linked to quiescence. Blow and Hodgson proposed to define quiescence as a reversible withdrawal from the cell cycle characterized by unlicensed origins and lack of CDK activity (Blow and Hodgson, 2002). Origins of replication in metazoans are bound by ORC complexes in quiescent and cycling cells, but other components of the pre-replication complex are not loaded onto chromatin in quiescent cells (see Chapter 3 by McClendon et al., this volume). These include CDC6 and CDT1, both required for assembly of the multisubunit helicase composed of MCM2-7. Expression of MCM-2 in cells of the colonic crypt corroborates findings of cultured cells, as MCM2 is expressed at high levels in amplifying cells that are actively proliferating and is not expressed in terminally differentiated cells. MCM2 expression levels in the stem cells at the base of the crypts are lower than in the amplifying cells, which are consistent with their infrequent division (reviewed in Blow and Hodgson, 2002). A number of the subunits of the pre-replication complex are not expressed in quiescent cells because their genes are repressed by E2F-dependent mechanisms and because their protein products are targets of the APCCdh1 ubiquitin ligase, which is active when CDK activity is low (Diffley, 2004). In this regard, it is important to highlight that CDK-mediated phosphorylation of CDC6 has been linked to its stabilization and accumulation, as it prevents APC-mediated ubiquitination (Mailand and Diffley, 2005). CDC6 stabilization, accumulation, and loading onto replication origins may occur concomitantly with or downstream of its E2F-dependent transcription. Alternatively, CDC6 accumulation may be a mitogenically regulated CDK event partially independent of the E2F program. It is also notable that CDC6 expression has been linked to an “attachment checkpoint” that apparently operates at least partially independently of E2F-dependent transcription in NRK fibroblasts (Jinno et al., 2002). Thus, assembly of pre-RC emerges as another process linked to G1 checkpoints that mediate quiescence entry and exit.
1.3 Oncogenes That Cooperate to Bypass Quiescence
Cellular oncogenic transformation is associated with unresponsiveness to antiproliferative and differentiation signals, bypassing of mitogenic extracellular requirements and an increase in proliferative lifespan. As multicellular organisms consist mostly of quiescent cells, critical oncogenic alterations may primarily deal with ensuring that cells initiating the transformation process remain in a stable proliferative state most of the time. In this regard, as the term “Restriction Point” was coined it was already proposed to be bypassed during malignancy (Pardee, 1974).
A number of studies have focused on identifying activated oncogenes, viral transforming proteins, and/or tumor suppressor genes that, when deregulated alone or in combination, confer mitogen independency, the ability to bypass contact inhibition, and insensitivity to lack of substratum attachment. In other words, defining genetic alterations that bypass signals that ensure entry into quiescence in normal cells. The most informative studies have been performed testing the effect of combinations of oncogenes/inactivation of tumor suppressor genes in primary cells of different species in culture. That is, testing the ability of these cells to grow in suboptimal concentrations of serum, to form foci in cell monolayers and/or to grow in an anchorage-independent manner. Often, these studies have been followed up by testing if cells that appear transformed in culture form tumors when injected into nude mice. If so, these cells are typically designated “malignantly transformed.” A conclusion of these studies is the fact that the ability of certain oncogenes to bypass cellular quiescence varies among species, and human cells are more resistant to oncogenic transformation than murine cells (Hahn and Weinberg, 2002). It is also important to mention here that cells need to become immortal for malignant transformation. This is accomplished in many human cells via stabilization of chromosomal ends (telomeres), which shorten with each DNA replication cycle, as somatic cells do not express telomerase. In contrast, the majority of human tumor cells express telomerase or exhibit an alternative mechanism (ALT) to maintain chromosomal length (Stewart and Weinberg, 2006; Johnson and Broccoli, 2007) (see Chapter 8 by Denchi, this volume). In contrast, most studies performed using murine cells show that these cells become immortal by overcoming stress checkpoints, as most cell types used in these studies exhibit telomerase activity. For an in-depth analysis on immortalization, senescence, cancer, and aging readers are directed to specific reviews (Sherr and DePinho, 2000; Hahn, 2002; Blasco and Hahn, 2003; Serrano and Blasco, 2007).
Thus, what oncogenes or inactivated tumor suppressor genes help bypass quiescence? Ectopic expression of c-MYC and an activated RAS oncogene in normal immortal rat fibroblasts (REF-52 cells) induce cyclin E/CDK2 activity and exit from quiescence in low concentrations of serum (Leone et al., 1997). In quiescent REF-52 cells, cellular RAS is required for activation of CDK activity and E2F-dependent gene expression in response to mitogenic stimulation, but activated RAS is insufficient to induce CDK activation in quiescent REF-52 cells placed in low serum. Coexpression of c-MYC allows cyclin E/CDK2 activation likely via downregulation of CKIs (Leone et al., 1997). Others have shown that ectopic expression of an oncogenic version of RAS induces premature senescence in human and mouse cells through activation of the RAF/MAPK pathway leading to activation of p53/ARF. This arrest is bypassed by depletion of p53 function in mouse cells, but disruption of both the pRB and p53 pathways is necessary to bypass RAS-induced senescence in human cells (Serrano, 1997, 1998) (see Chapter 9 by Adams, this volume). Thus, the c-MYC/RAS pair fails to transform mouse fibroblasts unless the p53/ARF pathway is mutated, and human cells require alteration of both the p53 and pRB pathways, as well as activation of telomerase in order to become transformed (Hahn et al., 2002; Hahn and Weinberg, 2002). This likely explains why normal human fibroblasts as opposed to immortal REF-52 cells fail to exit quiescence upon coexpression of c-MYC and an activated RAS oncogene (Sotillo et al., 2008). Thus, because the same oncogenic signals that induce escape from quiescence also induce senescence, alteration of the p53/ARF pathway is critical for oncogenesis in mouse cells, while alteration of both the p53 and pRB pathways is required in human cells. Thorough transformation assays from the Hahn and Weinberg laboratories have demonstrated that expression of oncogenic RAS, SV40 large T (LT) and small t (st) antigens, and the catalytic subunit of telomerase (hTERT) is sufficient to transform NHF (Hahn et al., 1999), human mammary epithelial cells (Elenbaas et al., 2001), and kidney epithelial cells (Hahn et al., 2002). Subsequent studies have shown that in this setting LT and st can be substituted by inactivation of p53, pRB, and PTEN plus constitutive expression of c-MYC (Boehm et al., 2005). The role of the SV40 tumor antigens will be discussed in more detail in the next section.
Examination of NHF with various combinations of genetic alterations showed that NHF-expressing hTERT, a dominant negative version of p53 and deregulated c-MYC, exit quiescence in low concentrations of serum, but subsequent disruption of pRB via shRNA made these cells insensitive to serum starvation (Boehm et al., 2005). Interestingly, as mentioned above disruption of the three pocket proteins in MEFs appears to be required to make these cells largely insensitive to mitogen withdrawal, as disruption of either pRB or p130/p107 delays but does not prevent cell cycle exit (Sage et al., 2000). Interestingly, in addition to its role in irreversible senescence and apoptosis, p53 has been suggested to participate in cell cycle exit into quiescence, as p53 expression and activity increase in quiescent NHF, and p53 inactivation by different means delays reversible cell cycle exit in response to mitogenic withdrawal (Itahana et al., 2002).
c-MYC has been shown to induce expression of the so-called E2F-activators (E2F1-3) and to directly interact with their promoters upon mitogenic stimulation (Leone et al., 2001; Fernandez et al., 2003; Leung et al., 2008). Of note, c-MYC binding to E2F-promoters seems critical for the loading of E2F1 to these promoters. Besides those genes required for cell cycle entry, E2F1 also regulates a group of genes involved in apoptosis. It has been shown that activation of the PI3K/AKT pathway during mitogenic stimulation inhibits E2F1 pro-apoptotic targets, favoring the role of E2F1 as an inducer of proliferation rather than apoptosis (Hallstrom et al., 2008; Hallstrom and Nevins, 2009).
The importance that activation of PI3K/AKT pathway has on oncogenic transformation is underscored by the fact that mutations in PTEN, a key negative regulator of this pathway, and amplification and abnormal activation of PI3K and AKT are associated with many types of human cancers (Keniry and Parsons, 2008; Yuan and Cantley, 2008). It has been shown that ectopic expression of the active subunit of PI3K, p110a, can substitute for st when coexpressed with LT and hTERT in human epithelial cells, promoting growth in low concentrations of serum as well as proliferation in soft agar (Zhao et al., 2003). In this scenario, st can also be substituted by coexpression of activated alleles of AKT1 and RAC, a downstream effector of the PI3K/AKT pathway. Also, forced coactivation of the RAS/RAF/MEK pathway with AKT elicited a robust proliferative response leading to activation of G1 cyclin/CDKs resulting from cyclin D1 accumulation and p27 repression, as well as removal of p21 from cyclin E/CDK2 complexes (Mirza et al., 2004). In this regard, the PI3K pathway negatively regulates FOXO transcription factors via AKT-mediated phosphorylation and exclusion from the nucleus. Activation of FOXO transcription factors is associated with cell cycle exit into quiescence in non-hematopoietic cells and has been implicated in the transcriptional activation of p27 as well as downregulation of D-type cyclins (Medema et al., 2000; Schmidt et al., 2002). Moreover, the PI3K pathway has been shown to cooperate with c-MYC in the expression of c-MYC-dependent genes by inactivating FOXO, which is implicated in the negative regulation of multiple c-MYC genes (Bouchard et al., 2004). FOXO transcription factors have been implicated in long-term survival of quiescent cells (Burgering and Medema, 2003).
The adenoviral oncoprotein E1A has also been long known to have the ability to stimulate exit from cellular quiescence. Two recent studies suggest how E1A triggers cell cycling and inhibits the cellular antiviral response and differentiation (Ferrari et al., 2008; Horwitz et al., 2008). E1A expression in quiescent fibroblasts results in the global relocation of pocket proteins and the p300/CBP acetyltransferase on cellular promoters. This process occurs in a sequential manner, leading to acetylation of lysine 18 on histone H3 and promoter transactivation of a restricted number of genes involved in proliferation and growth. However, both E1A and SV40 LT cause a global decrease in the acetylation of histone H3 at this site, which is apparently due to the restriction of HATs to the subset of proliferation/growth genes concomitant to the exclusion of these proteins on other gene promoters. Thus, hypoacetylation at histone H3 lysine 18 may be a general consequence of DNA tumor oncogenesis, which is linked to quiescence exit (reviewed in Ferrari et al., 2009).
It is also important to highlight recent findings that suggest that HES1, whose increased expression in quiescent cells has been associated with conferring the reversible nature of this state, is found expressed at high levels in rhabdomyosarcoma tumors and derived cell lines (Sang et al., 2008). Rhabdomyosarcomas are aggressive tumors that express the muscle differentiation factor MyoD, but exhibit a block in myogenic differentiation. Forced inactivation of HES1 in a rhabdomyosarcoma cell line via expression of a dominant negative HES1 mutant or pharmacological inhibition of Notch, which positively regulates HES1 expression, promoted MHC expression and differentiation of these cells.
In summary, to endow normal cells with the capability of exiting quiescence in unfavorable environments, oncogenes/inactivated tumor suppressor genes must mimic mitogenic signals. Cells with these alterations may produce their own mitogens, force surrounding cells to do so, or exhibit constitutively activated downstream signaling pathways independently of mitogenic stimulation. Some of these alterations as well as others will also help bypass antiproliferative signals from the environment or will make the cell independent of substrate feeding. Phosphorylation of pocket proteins and activation of the E2F transcription program are critical, but exit from quiescence does not always lead to effective proliferation. Thus, other pathways, such as those that inhibit apoptosis and cell senescence and/or are involved in monitoring faithful DNA replication, must also be altered.
1.4 SV40 and Exit from Quiescence
1.4.1 SV40 Tumor Antigens and Their Cellular Targets
As described in Section 1.3, transformation assays designed to identify the precise combination of altered pathways that are required for transformation of a variety of normal human cells have revealed that expression of SV40 LT and st antigens, oncogenic RAS, and hTERT suffices to ensure the required alterations (Hahn et al., 1999; Elenbaas et al., 2001; Hahn et al., 2002). Transformation is also attained with expression of oncogenic RAS, c-MYC, hTERT, st and inactivation of both p53 and pRB (Boehm et al., 2005). However, expression of st, hTERT, and pRB inactivation is not required when comparable transformation assays are performed using rodent cells. The effects of st on transformation of human cells are thought to be, at least in part, due to its ability to facilitate proliferation in conditions that promote quiescence. This indicates that the ability of st to facilitate a bypass of the quiescent state may be uniquely critical for transformation of human cells. Thus, in this section we will discuss the effects of expression of SV40 antigens in human cells with a major focus on st.
In the early 1960s, Polyomavirus Simian Virus 40 (SV40) was discovered as a viral contaminant during the production of poliovirus vaccines from cultures from rhesus monkey kidney cells (Eddy et al., 1962). Soon after its discovery, SV40 was shown to induce tumors in newborn hamsters (Girardi et al., 1962; Gerber, 1963). However, it took several years and work from multiple laboratories to demonstrate that the expression of proteins encoded in the Early Region (ER) of SV40 was responsible for oncogenic transformation (reviewed in Chen and Hahn, 2003).
1.4.2 SV40 Small t Antigen Promotes Exit from Quiescence
The effects of st on transformation of human cells are thought to be, at least in part, due to its ability to facilitate proliferation in reduced concentrations of growth factors. This function is dependent on its ability to bind and inhibit PP2A, as st mutants unable to bind PP2A fail to cooperate with LT driving cell proliferation in a limiting mitogenic environment (Skoczylas et al., 2004). Expression of st stimulates growth of monkey kidney cells (CV-1) maintained in 0.1% serum to an extent comparable to serum. In this scenario quiescence exit is accompanied by activation of the RAF/MAPK pathway (Sontag et al., 1993). It was later shown that activation of PKCζ through st-mediated inhibition of PP2A would also contribute to further activate the MAPK pathway, as well as activate NFkB-dependent gene expression (Sontag et al., 1997) in both CV-1 and NIH3T3 cells in 0.1% serum. In the same study it was determined that pharmacological inhibition of the PI3K pathway or a dominant negative mutant of p85 blocks st-mediated activation of PKCζ and NFκB, as well as st-induced cell proliferation (Fig. 1.3). The importance of the PI3K pathway for st-mediated transformation has been discussed earlier in this chapter when referring to combinations of oncogenes able to induce exit from quiescence in normal cells (Fig. 1.3).
Progress has more recently been made identifying PP2A heterotrimers that could mediate st transforming activities. In human embryonic kidney (HEK) cells, PP2A/B56γ has been identified as a potential target of st transforming activity. In these cells, knockdown of B56γ substitutes for st in transformation assays with defined combinations of altered genes (Chen et al., 2004). However, it was subsequently shown that B56γ cannot substitute for st in cells maintained in suboptimal mitogen concentrations (Moreno et al., 2004), suggesting that st has other targets important to bypass serum requirements. A second member of the same family of B subunits, B56α, has been identified as a negative regulator of c-MYC protein stability, providing a mechanism to explain how st increases c-MYC expression. Interestingly, a stable c-MYC mutant that cannot be dephosphorylated substitutes for st in transformation assays (Yeh et al., 2004; Arnold and Sears, 2006).
Downstream targets of st have also been studied. A number of reports have described transcriptional activation of cyclin D1 and cyclin A promoters in reporter assays (Porras et al., 1996; Watanabe et al., 1996; Skoczylas et al., 2005). Cyclin A protein levels were also shown to be upregulated by st in density-arrested human fibroblasts stimulated with serum, concomitantly with downregulation of p27. In this scenario, cyclin A is inactive and cells do not enter the cell cycle unless LT is coexpressed (Porras et al., 1999).
In search for oncogenes that cooperate to bypass quiescence induced by complete depletion of mitogens in NHF, our lab found that coexpression of st and cyclin E, but not their individual expression, was sufficient to bypass quiescence and induce DNA synthesis (Sotillo et al., 2008). This same combination of oncogenes bypassed quiescence induced by growth to high density and lead to continued proliferation and foci formation in hTERT-NHF. These events are at least partially dependent on PP2A inhibition. Expression of st alone in quiescent NHF did not lead to accumulation of cyclins A or D1, but strikingly led to accumulation and loading of the essential replication factor CDC6 (Sotillo et al., 2009). As MCM2 loading onto chromatin was also observed in density-arrested cells expressing st, these results suggest that st induces steps toward licensing of replication origins, a characteristic of cells exiting G0 into G1 (see Section 1.2.4 above). When cyclin E and st were coexpressed in quiescent NHF, CDC6 further accumulated coinciding with CDK2 activation and DNA synthesis. In addition, we and others have observed that CDC6 expression, as well as the expression of other pre-RC components, is linked to phosphorylation of CDK2 on its activating T-loop (Nevis et al., 2009; Sotillo et al., 2009). Therefore, deregulation of cyclin E expression in the context of normal cells apparently driven out of quiescence by st leads to the cooperative and coordinated activation of an essential pre-replication complex factor (CDC6) and an activity required for origin firing (CDK2) (Sotillo et al., 2009). Importantly, it was also found that in the context of this oncogenic-driven exit from G0 and proliferation, CDK2 activity appeared to be essential (Sotillo et al., 2008). While the direct target of st in this case is unknown the current data suggest that the selective accumulation of the CDC6 transcript is dependent on E2F promoter elements but independent of CDK activation. This suggests that st controls factors that can selectively regulate the expression of a gene(s) whose expression is associated with exit from G0. Finally, it is important to point out that despite the observation that CDC6 expression is required for passage through an “attachment checkpoint” in NRK fibroblasts (Jinno et al., 2002), expression of cyclin E and st failed to induce anchorage independent growth of hTERT NHF, indicating that this oncogene pair does not reverse quiescence induced by all signals. Altogether these studies show that st changes a fundamental property of quiescent cells that differentiates them from post-mitotic G1 cells, which is the status of the cell replication origins. By facilitating passage through the G0/G1 transition, st may trick certain cells to create an environment proper for viral replication despite extracellular signaling that would otherwise keep cells in a quiescent state. Understanding how st disrupts the mechanisms that ensure maintenance of the quiescent state will provide insight into the molecular nature of this state, increase our understanding of how DNA tumor viruses promote their own replication, and may unveil novel mechanisms for cellular transformation that are exclusive to human cells.
1.5 Future Directions
The defining characteristic of cells in the quiescent state is maintenance of their ability to re-enter the cell cycle in propitious conditions. Recent work has shown that while cells exiting the cell cycle into quiescence initiate programs of gene expression that are coupled to the quiescent inducing signal, there is a common gene expression signature that emerges in a time-dependent manner. This signature appears associated with acquisition of resistance to differentiation, senescence, and cell death. A transcription factor designated HES1 has been shown to be upregulated during cell cycle exit into quiescence and appears to be required for maintenance of the quiescent state, as it blocks both differentiation and senescence in human fibroblasts. This factor is found deregulated in rhabdomyosarcomas and mediates a block in myogenic differentiation exhibited by these tumor cells. Work over the past several decades has led to the identification of oncogenes and tumor suppressor genes that, when deregulated, facilitate mitogen-independent cell cycle progression and insensitivity to other signals that induce exit into quiescence. SV40 st antigen has emerged as a key player in the specific transformation of human cells, as its unique transforming activity is not required in rodent cells. st appears to facilitate transformation in certain environments that favor cell cycle exit into quiescence, such as mitogen starvation and contact inhibition. Recent progress in identification of the PP2A heterotrimers that are targeted by st, as well as the downstream cell cycle players that mediate st activities that drive cells out of quiescence are likely to provide important insights in the near future.
We thank Manuel Serrano, David G. Johnson, Alison Kurimchak, and Judit Garriga for critically reading this manuscript and helpful suggestions. Work in this lab has been supported by a grant project under CA095569 and a Career Development Award (K02 AI01823) to XG of the National Institutes of Health.
- Hahn WC (2002) Immortalization and transformation of human cells. Mol Cell 13: 351–361.Google Scholar
- Leone G, Sears R, Huang E, Rempel R, Nuckolls F, Park CH, Giangrande P, Wu L, Saavedra HI, Field SJ, Thompson MA, Yang H, Fujiwara Y, Greenberg ME, Orkin S, Smith C, Nevins JR (2001) Myc requires distinct E2F activities to induce S phase and apoptosis. Mol Cell 8: 105–113.CrossRefPubMedGoogle Scholar
- Lin AW, Barradas M, Stone JC, van Aelst L, Sarrano M, Lowe SW (1998) Premature senescence involving p53 and p16 is activated in response to constitutive MEK/MAPK mitogenic signalling. Genes Dev 12: 3008–3019.Google Scholar
- Moreno CS, Ramachandran S, Ashby DG, Laycock N, Plattner CA, Chen W, Hahn WC, Pallas DC (2004) Signaling and transcriptional changes critical for transformation of human cells by simian virus 40 small tumor antigen or protein phosphatase 2A B56gamma knockdown. Cancer Res 64: 6978–6988.CrossRefPubMedGoogle Scholar
- Pallas DC, Weller W, Jaspers S, Miller TB, Lane WS, Roberts TM (1992) The third subunit of protein phosphatase 2A (PP2A), a 55-kilodalton protein which is apparently substituted for by T antigens in complexes with the 36- and 63-kilodalton PP2A subunits, bears little resemblance to T antigens. J Virol 66: 886–893.PubMedGoogle Scholar
- Porras A, Bennett J, Howe A, Tokos K, Bouck N, Henglein B, Sathyamangalam S, Thimmapaya B, Rundell K (1996) A novel simian virus 40 early-region domain mediates transactivation of the cyclin A promoter by small-t antigen and is required for transformation in small-t antigen-dependent assays. J Virol 70: 6902–6908.PubMedGoogle Scholar
- Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW (1997) Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 88: 593–602.Google Scholar
- Yeh E, Cunningham M, Arnold H, Chasse D, Monteith T, Ivaldi G, Hahn WC, Stukenberg PT, Shenolikar S, Uchida T, Counter CM, Nevins JR, Means AR, Sears R (2004) A signalling pathway controlling c-Myc degradation that impacts oncogenic transformation of human cells. Nat Cell Biol 6: 308–318.CrossRefPubMedGoogle Scholar