[Cancer Biology & Therapy 2:2, 124-130, March/April 2003]; ©2003 Landes Bioscience
Review
Role of the Retinoblastoma Protein in Differentiation and Senescence ABSTRACT
2Department
of Pathology; Harvard Medical School; Boston, Massachusetts USA
rib ut io n.
fo
*Correspondence to: P.W. Hinds; Associate Professor; Harvard Medical School; Department of Pathology; Armenise 433; 200 Longwood Ave.; Boston, Massachusetts 02115 USA; Email:
[email protected]
is t
1University of Melbourne Department of Medicine; St Vincent’s Hospital; Melbourne, Australia
The retinoblastoma protein pRb is functionally inactivated in most human cancers. Numerous studies in cell culture and animal models suggest that pRb has a unique ability to encourage and enforce permanent cell cycle withdrawal, consistent with its role as a tumor suppressor protein. This cell cycle withdrawal has a generic component involving repression of transcription of genes required for proliferation. In addition, numerous studies hint at additional specific roles for pRb in differentiation of certain tissue types. Further, pRb appears to play a central role in the process of cellular senescence, a tumorsuppressive process characterized by proliferative arrest and phenotypic changes. Both differentiation and senescence pathways influenced by pRb involve direct and indirect interactions with the core machinery involved in cell-type-specific differentiation and cell shape control. This review focuses on pRb’s role as an participant in osteoblast differentiation illustrative of a broader role in terminal differentiation. In addition, novel pathways activated by pRb in its role as an inducer of cellular senescence will be discussed.
rd
David M. Thomas1 Hai-Su Yang2 Kamilah Alexander2 Philip W. Hinds2,*
Received 05/02/03; Accepted 05/03/03
ot
Previously published online as a CB&T “Paper in Press” at: http://landesbioscience.com/journals/cbt/
N
CELL CYCLE EXIT AND DIFFERENTIATION
©
20 03
La
nd
es
B
io
sc
ie
This work was supported by USPHS grant AG20208
The relationships between differentiation, proliferation and senescence appear fundamental to cell biology, recapitulated in multiple tissue systems. Our understanding of these relationships, and their implications for human disease, has expanded enormously in the past 15 years as a consequence of seminal advances in the molecular understanding of both cell cycle control and differentiation. Molecules identified either as regulators of differentiation or of cellular proliferation, have turned out to possess pleiotropic and often unexpected roles in other systems. Targeted disruption of any one component may have consequences for both normal tissue function and the emergence of tumors. For example, the retinoblastoma protein (pRb) was initially identified as a tumour suppressor protein critical to the genesis of the childhood tumour, retinoblastoma,1 but this has since been clearly shown to play key roles both in cell cycle control and development. The elucidation of mechanisms regulating cell cycle progression constitutes a major scientific achievement, but predictably has raised as many new questions as it solved. Our current understanding of the role of pRb in regulating transit through the G1/S phase of the cell cycle is generic: it is conceived of as operating in essentially all postembryonic cells capable of proliferation. Briefly, pRb belongs to the pocket protein family, which includes p107 and p130. pRb is thought to act as a corepressor controlling a family of transcription factors, collectively referred to as E2F. The E2F proteins are thought to bind to pRb through the pocket domain. These proteins, with their partners, DP1 and DP2, transactivate genes important for G1- to S-phase transition.2-4 Phosphorylation of pRb, and concomitant release of E2F repression, is catalyzed by cyclin D-dependent kinases, cdk4 and 6, in response to extracellular signals, and by the cyclin E- and Adependent kinase, cdk2. The activity of these kinases is in turn inhibited by proteins of the INK4 and CIP/KIP families. The INK4 family, whose members inhibit the D-kinase complexes, comprises p15INK4B, p16INK4A, p18INK4C and p19INK4D. These proteins appear to play roles in mediating growth arrest in response to extracellular signals and cell states. In ways that are still not well understood, p16INK4A levels are associated with the in vitro phenomenon of senescence, and appears to be a bona fide tumor suppressor protein whose function is lost in many cancers and clonal cell lines unlike other members of the INK4 family. The CIP/KIP family consists of p21CIP1, p27KIP1 and p57KIP2. These proteins bind and inhibit active cdk2 complexes, thereby maintaining pRb in an inactive state. Hypophosphorylated pRb is recruited to promoters by E2F, and in turn recruits histone deacetylase (HDAC) which deacetylates chromatin leading to active transcriptional repression. One significant
e.
Senescence, Differentiation, Retinoblastoma, Cell Cycle Exit, Tumorigenesis
nc
KEY WORDS
124
Cancer Biology & Therapy
2003; Vol. 2 Issue 2
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
target of this repression is the cdk2 activator cyclin E, suggesting the presence of a feed-forward loop in the pRb pathway: initial inactivation of pRb can lead to elevated levels of cyclin E, which in turn further inactivates pRb and performs other functions required for S phase progression.5 Indeed there is in vivo evidence that improper expression of cyclin E and subsequent activation of cdk2 can circumvent the need for pRb regulation, although this critical role of cyclin E/cdk2 may be subverted in at least some cancer cells.6-8 The cell cycle regulatory pathway centered around pRb is inactivated in almost all human cancers.2,3,9,10 Despite this, it is clear that the phenotypic consequences of loss-of-function of pRb are highly tissue-specific, both in murine models and in human disease. Furthermore, pRb seems to be critical for many aspects of differentiation in vitro and in vivo in ways that are not obviously accounted for by a generic role in regulating cell cycle progression.9 It is tempting to speculate that the tissue specificity of the phenotype observed following loss of function of pRb could be accounted for by relationships to differentiation processes within those tissues, which are by definition tissue-restricted. Indeed, the recent demonstration that many of the development defects originally observed in pRb-null animals may be secondary to placental defects11 does not explain the observation that muscle formation is profoundly disrupted in a variety of animal models of pRb loss including those that circumvent the aforementioned placental defect.11,12 Similarly, clear roles for pocket proteins and their E2F targets are emerging in models of adipogenesis.5,13-15 Thus it is clear that the pRb pathway can influence differentiation and development with likely direct impact on tumor susceptibility. Interestingly, the individual components of the pRb pathway targeted in the course of tumorigenesis seem specific to the tumor type. For instance, in melanoma loss of p16INK4A predominates, and in breast cancer overexpression of cyclin D1 is the most common alteration in the pRb pathway. pRb itself is most frequently inactivated in a highly-restricted subset of human tumors, including retinoblastomas, osteosarcomas, small lung cell carcinomas and bladder carcinomas. Among individuals with inherited heterozygous loss of the RB gene, osteosarcoma is the second most common tumor after retinoblastoma itself, with a 500-fold greater incidence of osteosarcoma than the general population.16-17 In addition, loss of pRb occurs in up to 60% of sporadic osteosarcomas.18-20 These data suggest a tissue-specific function of pRb in bone that is important to its role as a tumor suppressor. In light of comprehensive recent reviews of pRb and E2F function in cell cycle progression and development,5,12 this review will focus on recent gains in understanding of the molecular role of pRb in osteoblast differentiation and its implications for osteosarcoma.
ROLES FOR PRB IN OSTEOBLAST DIFFERENTIATION AND OSTEOSARCOMA Osteosarcoma, or primary bone cancer, is an uncommon tumor. It predominantly affects young adults, with a peak incidence in the 2nd decade of life, and occurs more frequently in males than females. It has long been noted in clinical cancer care that differentiation state identified pathologically predicts outcomes for a broad range of tumors. This is also true for osteosarcoma. Regardless of nodal status or the presence of metastases, either poorly differentiated or undifferentiated osteosarcoma heralds a 10–15% decrease in 5-year survival. It is not known why the state of differentiation of tumors in general should confer such powerful prognostic significance. www.landesbioscience.com
There are two possible interpretations of this phenomenon. One holds that the differentiated state of tumors is an epiphenomenon of genomic instability. In this model, loss of phenotype-specific genes occurs as a byproduct of increased rates of mutational events in cancers. Those cells that sustain loss-of-function mutations that lead to a growth disadvantage will be deleted from the population; cells that sustain mutations leading to a growth advantage will be enriched within the population; but those that sustain mutations within irrelevant genes in a stochastic fashion will neither be selected for nor discriminated against. Due to the lack of negative selection pressure in the face of continued mutagenic events, over time the population as a whole will tend to loss of differentiation. In the extreme case, only those genes will be preserved that are necessary to cell survival and proliferation. The factors governing the de-differentiation of a tumor will include both the rate of stochastic mutational events and the “age” since the transforming event. The alternate proposition suggests that differentiation imposes a constraint upon cell growth. Bypassing this constraint confers a growth advantage upon the tumor cell. Hanahan and Weinberg21 proposed that all tumors are characterised by six features: self-sufficiency in growth signals; insensitivity to anti-growth signals; evasion of apoptosis; limitless replicative capacity; angiogenesis; and the capacity to support tissue invasion and metastasis. Differentiation programs may affect two of these six features. First, differentiation may lead to decreased rates of cell division compared to undifferentiated cells. For example, in embryonic development, initial cell cycles are characterised by rapid cell division with minimal evidence of G1 or G2 phases. Second, it is a fundamental characteristic of stem cells that they possess unlimited replicative capacity (cellular immortality); as cells differentiate, mortality is acquired in some way at the point of commitment to a particular lineage. In vitro once cells reach this replicative limit, known as the Hayflick limit, they undergo a terminal egress from the cell cycle (known as cellular senescence). It is probable that a relationship exists between differentiation and the imposition of an “age limit” upon daughter cells. Precisely what this relationship is remains unknown but is likely to involve common regulation of factors that support the irreversibility of cell cycle exit observed in both differentiation and senescence. There is considerable circumstantial evidence to suggest that pRb may be involved in the relationship between cell cycle exit and differentiation in bone-forming cells. As noted above, much of this is predicated on epidemiologic observations in human cancer. Further, experimental evidence suggests the pocket proteins play roles in the differentiation of mesenchymal lineages, specifically chondrogenesis,22 myogenesis23 and adipogenesis.13-15,24,25 Little is currently understood about this relationship in bone, but nearly a decade of research in myogenesis has established a conceptual framework for investigating the interactions between cell cycle and differentiation machinery. Myogenesis involves the terminal withdrawal of myoblasts from the cell cycle, and formation of multinucleated myotubes. Coordination of cell cycle arrest with critical steps in differentiation is essential for proper muscle development. MyoD, a basic helix-loop-helix transcription factor, may both interact with the retinoblastoma tumor suppressor protein,26 and alter the action of cyclin-dependent kinase complexes.27 Disruption of cell cycle control through loss of pRb also disrupts the myogenic differentiation program.28,29 MyoD itself directly transcriptionally upregulates expression of the cyclin-dependent kinase (CDK) inhibitor p21CIP1, which in turn inhibits CDK2-complexes and thereby activates a pRb-dependent growth arrest.30 Thus the cellular machinery required for the myogenic phenotype also directly contributes to the
Cancer Biology & Therapy
125
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
inhibition of cell proliferation prerequisite for terminal myogenic differentiation. Using muscle as a template, it is interesting to speculate on the mechanisms by which progressive differentiation of osteoblasts is linked to withdrawal from the cell cycle. There are several observations that strongly suggest that osteoblast differentiation or function may be affected by the integrity of the pathways in which pRb is crucial. Viral oncoproteins that target the pocket proteins inhibit osteoblast differentiation. In osteoblast cell lines conditionally immortalized by temperature-sensitive SV40 large T antigens, differentiation is contingent upon deactivation of the oncoprotein.31-32 Evidence directly implicating pRb is suggested by the dependence on the pocket protein-binding domain of the adenoviral E1A 12S protein for the anti-differentiating effects of this protein.33 Consistent with the high frequency with which pRb is mutated in osteosarcoma, osteosarcoma-derived cell lines usually make poor models for study of late aspects of osteoblast differentiation. Primary human osteosarcoma biopsies have been reported to express low or absent levels of osteocalcin, a late marker of differentiation.34 Further, re-expression of pRb in SAOS2 osteosarcoma cells induces a senescent phenotype35-36 that displays markers suggestive of bone differentiation.37 Many cell culture models of differentiation exist in which it is possible to recapitulate various aspects of the osteogenic differentiation program. Additionally, primary cell cultures of calvarial osteoblasts can be induced to differentiate in vitro by various protocols. These models are used to generate matrix synthesis and mineralization, and are associated with expression of late markers of osteogenic differentiation, such as osteocalcin. The protocols used in these studies have in common the necessity of inducing growth arrest, invariably by cell contact inhibition. Indeed, continuous passage of these lines from subconfluent cultures completely prevents mineralization or osteocalcin production. Taken together, these data suggest that growth arrest is essential for full expression of the osteoblast phenotype in vitro. In this connection it is interesting to note that in vitro models of myogenesis also depend on growth arrest, usually induced by culture in low serum conditions.
THE RETINOBLASTOMA PROTEIN IS REQUIRED FOR OSTEOGENIC DIFFERENTIATION IN VITRO We recently reported direct evidence of a role for pRb in osteoblast differentiation.38 This study provides a model for understanding why pRb is selectively targeted in bone cancer. We investigated the effect of disrupting pRb function in MC3T3-E1 preosteoblasts, which lack the cyclin-dependent kinase inhibitor p16INK4A and are thus immortal in the presence of wild-type pRb. Upon differentiation of MC3T3E1 cells with ascorbic acid and β-glycerophosphate, one observes an increase in the activity of RUNX2, a key transcriptional regulator of bone formation,39 as well as increased expression of osteocalcin and mineralization of matrix in vitro.38,40 Expression of the pocket-protein inactivating HPV16 E7 oncoprotein in these cells resulted in loss of expression of markers of the differentiated osteoblast phenotype, including attenuated alkaline phosphatase (ALP) activity, osteopontin protein levels, type I collagen mRNA levels, osteocalcin mRNA and protein expression, and mineralization. Although these data are consistent with a role for pRb in osteoblast differentiation, HPV16 E7 affects other proteins besides pRb, including p107, p130 and p21CIP1. To better define the specific role of pRb in osteogenic differentiation, we used recombinant bone morphogenetic protein 2 (BMP-2) to induce differentiation in murine embryo fibroblasts (MEFs) from wild-type and RB-/- litter 126
mates, and MEFs lacking p107 or p130. BMP-2, a member of the transforming growth factor-β superfamily, has potent osteogenic properties,41 including induction of RUNX2, ALP, osteocalcin and mineralization in vitro and in vivo.42-43 MEFs lacking pRb did not undergo BMP-2 induced differentiation, particularly with respect to expression of osteocalcin and mineralization, while the loss of p107 or p130 individually appeared to enhance differentiation. Interestingly, p107/p130 double null MEFs resembled pRb-null MEFs. The defects in the BMP-2 responsiveness of MEFs lacking the RB gene specifically implicate pRb in regulation of terminal differentiation, rather than commitment to the osteoblast lineage per se. In our studies, BMP-2 increased the expression of the bonespecific transcription factor RUNX2 and ALP activity in both RB-/and wild-type MEFs but strikingly failed to induce late markers of differentiation, osteocalcin and mineralization, in RB-/- MEFs. These observations are consistent with data from human tumors. Osteosarcomas typically express high levels of ALP in the absence of pRb,44 while in contrast osteocalcin expression is usually reduced or absent34 (Thomas D, Hinds P, unpublished data). Thus, markers of earlier stages of osteoblast differentiation may be expressed in osteosarcoma-derived cells, and even vary with proliferative status, but late markers are less commonly expressed. These data are consistent with the mutual exclusivity of the terminally differentiated state, at least as measured by osteocalcin expression in bone-derived cells, and tumorigenesis.
THE RETINOBLASTOMA PROTEIN ACTS AS A TRANSCRIPTIONAL COACTIVATOR OF RUNX2 The runt family member RUNX2 plays an essential role in osteoblastic differentiation in vitro and in vivo.45-54 In mouse skeletal development, RUNX2 is first observed in early mesenchymal condensations, and then appears to be restricted to osteoblastic and chondroblastic cells.46 Mice nullizygous for RUNX2 exhibit a complete lack of ossification, and die immediately after birth from respiratory failure.50-51 Heterozygous mutant mice exhibit skeletal abnormalities comparable to the human disease cleidocranial dyplasia.54 Furthermore, RUNX2 maintains osteoblastic function postnatally by regulating expression of several bone-specific genes, such as osteopontin and osteocalcin, and by controlling bone extracellular matrix deposition.45,47,55,56 We found that neither basal nor BMP-2-induced expression of RUNX2 is dependent on pRb, suggesting that the defect in differentiation of RB-/- MEFs is downstream of RUNX2 induction.38 A physical association of RUNX2 and pRb exists both in vitro and in vivo, mediated by their C-terminal domains (aa383–516 of RUNX2 and aa833–882 of pRb). Interestingly, the spontaneous mutation of pRb in SAOS2 cells targets its RUNX2 interaction domain. It is important to note that RUNX2 did not associate with the related pocket proteins, p107 and p130, supporting the biological data described above. Furthermore, deconvolution microscopy indicates that RUNX2 and pRb co-localize in vivo. Tying this interaction to functional consequences, the capacity of pRb to associate with the osteoblast-specific promoters in vivo as assessed by chromatin immunoprecipitation is dependent on expression of RUNX2, suggesting that pRb is recruited to promoter sequences by RUNX2. The physical interaction of pRb with RUNX2, the cellular dependence of osteocalcin expression on pRb, and the recruitment of pRb to osteoblast-specific promoters in vivo suggested that pRb may positively regulate the function of RUNX2 as a transcription factor. Using a synthetic osteoblast-specific promoter45 (6OSE2-luc)
Cancer Biology & Therapy
2003; Vol. 2 Issue 2
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
or the native osteocalcin promoter we studied the effect of pRb with and without RUNX2 in cell lines in which both pRb and RUNX2 were either absent or dysfunctional (C33A cervical carcinoma cells, RB-/- 3T3 fibroblasts, and SAOS2). pRb alone failed to activate this promoter in any cell line, consistent with the absence of a DNA binding domain or previous evidence of direct transcriptional activity. However, RUNX2 and pRb together increased transcription approximately 2-fold over RUNX2 alone. This effect was not due to changes in RUNX2 levels, but rather to increased transcriptional activity of the protein. Interestingly, when chromosomally integrated osteocalcin promoter-reporter was employed, the magnitude of pRb activation of RUNX2 was much greater, but was not observed until 5–10 days after transfection of RB and RUNX2 expression constructs. Together, these data implicate chromatin remodelling events in pRb's ability to activate RUNX2, but the nature of these changes and of the proteins that cause them remains unclear. It is noteworthy that pRb, p107 and p130 all induced a similar growth arrest in SAOS2 cells but only pRb could augment RUNX2 activity, suggesting that growth arrest per se is insufficient to increase transcriptional activity of RUNX2. Deletion mutant studies indicated that activation of RUNX2 by pRb likely depends on the physical interaction of these two proteins through their respective C-termini. Interestingly, certain mutations of the pocket domain also abolished the synergism with RUNX2 without altering pRb/RUNX2 association. This may indicate that a pocket-domain function of pRb contributes to the transcriptional effect, in addition to the required physical interaction. These data suggest that pRb contributes directly to expression of osteoblast-specific genes by acting as a transcriptional co-activator for RUNX2. This effect is not shared by p107 and p130, identifying a unique function of pRb that may contribute to its role as a tumor suppressor gene.
OSTEOBLAST DIFFERENTIATION AND GROWTH ARREST The data presented above indicate that a key component of cell cycle machinery, pRb, interacts with RUNX2 and potentiates RUNX2-dependent transcription from the osteocalcin promoter. It is less clear whether the initiation of the osteoblast differentiation program contributes to growth arrest. As with myogenic differentiation noted earlier, there is evidence that osteogenic differentiation is associated with changes in levels of cyclin-dependent kinase inhibitors. BMP-2 has been reported to increase p21CIP1 levels, leading to the hypophosphorylation of pRb.57 The mechanism of this effect is not known. In an in vitro model, p27KIP1 has been observed to increase with osteoblast differentiation58 (Thomas and Hinds, unpublished data). p27KIP1 levels were increased during the matrix formation and mineralisation phases, while both p27KIP1 and p21CIP1 levels were increased by vitamin D treatment of rat calvarial cultures. Close analysis of the bones of p27KIP1-deficient mice revealed increased bone size and marrow osteoprogenitor cell populations. In vitro, p27KIP1-null osteoblasts appeared to compensate for the absence of p27KIP1 by increased levels of p21CIP1. However, the lack of p27KIP1 did not affect mineralisation of osteoblast cultures in vitro; indeed, mineralisation was enhanced in p27KIP1-/- osteoblast cultures, concomitant with an increase in cell density per well. It is not clear whether the mineralisation was reduced on a per cell basis. These data were interpreted to indicate that p27KIP1 played a role in limiting the proliferative capacity of osteoprogenitors and dividing osteoblasts, leading to accelerated expansion of this cell population. It is noteworthy that the late marker of the osteoblast lineage, osteocalcin, appears to inhibit bone formation in vivo.59 The expression www.landesbioscience.com
Figure 1. A model for pRb function in differentiation and terminal cell cycle exit of osteoblasts. Both E2F-dependent and independent functions of pRb are proposed to contribute to loss of proliferative capacity during the differentiation of osteoblasts. This general schema is likely to apply to other cell types as well. At left, recruitment of pRb to promoters of genes required for S phase progression is achieved through interaction with E2F/DP complexes; red crescent represents subsequent recruitment of chromatin-modifying enzymes that repress transcription. At right, physical interaction of pRb with RUNX2 leads to recruitment of additional unknown factors (?) that promote transcription of osteocalcin and perhaps other late markers of osteoblast differentiation. A third repressive function is provided by increased expression of p27KIP1 in response to RUNX2 and pRb activation; the mechanism of this activation is unknown. A further postulate of this model is that cyclin-dependent kinase-mediated regulation of pRb differs for the various functions described leaving pRb as the optimum choice for mutation in the genesis of osteosarcomas.
of osteocalcin was not examined in the study by Drissi et al.,58 so we do not know whether osteocalcin expression was altered in p27KIP1null osteoblasts. We have found that forced expression of RUNX2 leads to inhibition of cell growth in primary human fibroblasts. This effect is mediated by increased expression of p27KIP1, with associated inhibition of cyclin A-associated kinase activity and dephosphorylation of pRb. Indeed, suppression of colony formation due to ectopic expression of RUNX2 is dependent on the presence of pRb. Taken together, these data suggest that the increase in p27KIP1 associated with osteogenic differentiation is not strictly required for differentiation, or that some functional redundancy exists perhaps due to compensatory upregulation of p21CIP1 or p57KIP2. Nevertheless p27KIP1 may be a major mediator of cell cycle exit in response to pRb and RUNX2 action in osteoblasts but the mechanism through which increased p27KIP1 occurs remains unknown. Clearly, significant insight into pRb-mediated regulation of osteoblast differentiation and cell cycle exit as summarized in Figure 1 awaits the elucidation of the role of pRb/RUNX2 in synthesis of anti-proliferative proteins like p27KIP1 as well as of osteoblast-specific markers like osteocalcin.
SENESCENCE AND PRB As mentioned above, the terminal cell cycle exit associated with differentiation is likely accompanied by that conferred by the program of senescence as a significant mode of tumor suppression regulated by pRb. Proliferative exhaustion of primary cells in culture, termed cellular replicative senescence, has been observed upon the culture of many cell types from a variety of species.60 Senescent cells can typically be identified by an enlarged, flattened phenotype and the expression of an enzymatic β-galactosidase activity at pH 6.0 (SA-β-gal) of unknown regulation. In addition, senescent cells are characterized by an irreversible G1 growth arrest involving the repression of genes that drive cell cycle progression and the upregulation of cell cycle
Cancer Biology & Therapy
127
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
inhibitors like p16INK4a, p53, and its transcriptional target p21CIP1.61 The irreversibility of the senescent cellular state suggests that senescence may be anti-oncogenic. Indeed, this antiproliferative response must be overcome if cells are to become transformed.62 Further, senescence may play a significant role in response to cancer therapy.63 In human cells, the bypass of senescence requires inactivation of the p16INK4a/pRb pathway.63 Indeed, these negative regulators of proliferation accumulate with continued cell division in cultured primary cells, suggesting that they are part of a mechanism that counts the number of cell divisions and thus limits proliferative capacity. The initiating signal of this molecular “clock” has been attributed to the shortening of telomeres, the progressive erosion of which is thought to eventually trigger growth arrest. In support of this hypothesis, hTERT overexpression and the subsequent restoration of telomerase activity blocks telomere shortening and ultimately immortalizes cells.64 In addition, though hTERT activity is undetectable in normal cells, it is upregulated in tumor cells further suggesting that dysregulation of hTERT activity is involved in the malignant transformation of cells.65-66 Significantly, it has been found that immortal cells arising out of hTERT transduced cells require spontaneous or oncogene-induced disruption of the pRb pathway.67-70 Thus, pRb likely controls a senescence-instigating pathway that synergizes with, but is distinct from, that engendered by telomere loss. The strongest evidence for a telomere-independent senescence pathway is provided by the observation that primary cells will senesce prematurely upon oncogenic stimuli. For instance, the oncogenic, persistently-activated form of the small GTPase Ras, which initially causes proliferation, eventually triggers cell cycle arrest and premature senescence in both mouse embryo fibroblasts (MEFs) and human diploid fibroblasts (HDFs).71 While either p53 or pRb is dispensable in Ras-induced senescence in MEFs, only ablation of pRb function in HDFs by E1A results in senescence abrogation.71 Further, fibroblasts from patients with constitutive inactivating mutations in p16INK4a are resistant to Ras-induced senescence, although hTERT expression is still required for immortalization.72 This work underscores the critical role of p16INK4a/pRb pathway in premature senescence and its obligate inactivation in the immortalization of human cells. Significantly, the role of p16INK4a/pRb in senescence of primary cells can be recapitulated in tumor cells. The reintroduction of pRb or p16INK4a into tumor cells that have lost expression of either protein induces a premature senescence requiring p21CIP1 or, in the absence of an intact p53 pathway, p27KIP1.73-75 Thus, the growth arrest accompanying senescence has much in common with that observed in differentiating cells exiting the cell cycle as described above. Intriguingly, cyclin dependent kinase inhibitors like p14ARF, p21CIP1, and p27KIP1, which are required for senescence, can induce markers of senescence on their own. However, they cannot mediate the senescent shape change demonstrating that these two processes in senescence are separable.73,74,76 Despite clear evidence for a role for the p16INK4a/pRb pathway in senescence and a fundamental role for CIP/KIP inhibitors in mediating cell cycle exit, little is known about the mechanism of induction of the senescent phenotype characterized by expression of SAβ-gal and morphological alteration, nor about the impact of these cellular phenotypes on proliferation and tumorigenesis. Our recent work has involved an analysis of downstream effectors of cell shape changes in response to senescence induction and has resulted in the 128
definition of biochemical events likely to play significant roles in the tumor suppressive properties of senescence. Using several model systems of senescence, including long-term passage and acute expression of Ras or pRb, we have found that cdk5, a kinase that hitherto has been considered to be almost solely involved in regulating neuronal activities,77 plays a central role in both SA-β-gal induction and the shape change of senescent cells (Alexander and Hinds, unpublished). Cdk5 activity increases in cells induced to senesce by a variety of stimuli and inhibition of cdk5 activity blocks the expression of the senescence marker SA-β-gal. Further, we find in a variety of different types of senescence that cdk5-mediated repression of Rac1 activity is necessary for proper acquisition of the cytoskeletal changes accompanying senescence. Together, these observations indicate that cdk5, acting through Rac1, can function as a central regulator of multiple aspects of senescence. Although the exact mechanisms by which pRb leads to cdk5 activation and Rac1 inactivation remain mysterious, we have identified a cdk5 substrate in senescent cells. Modification of ezrin, and likely the related proteins radixin and moesin that all together comprise the ERM proteins, occurs on a specific residue (T235 of ezrin) and results in the membrane localization of the ERM protein.78 We postulate that this activation and localization of ERMs participates in cellular shape changes, and is likely tied to Rac1 inactivation and ultimately influences cellular proliferation or mobility. Indeed, ERM proteins have recently emerged as important cytoskeletal- and membrane-associated proteins involved in proliferation control, and, through the action of the related protein Nf2/merlin, in tumor suppression.79 These proteins play a role in the formation of microvilli, cell-cell junctions, and membrane ruffles, and also regulate substrate adhesion and motility.80-82 It has most recently become clear that the ERM proteins regulate and respond to proliferative signals, both in a positive and negative manner.82,83 Phosphorylation has been proposed to regulate ERM activation, since phosphorylation of ERM proteins correlates with their cytoskeletal association.84,85 Several observations have suggested that phosphorylation of serine/threonine residues is important for the activity of ERM proteins.82 Phosphorylation of T567 in ezrin has been found to be critical for conversion of ezrin to the active, open form competent for membrane localization and actin binding.86 Indeed, structural studies suggest that phosphorylation of T567 would sterically interfere with intramolecular N-terminal and C-terminal (referred to as N-ERMAD and C-ERMAD) interactions.87 Thus, regulation of ERM proteins through phosphorylation is likely critical to membrane-cytoskeleton signalling, and this in turn will have a pleiotropic impact on cell shape, motility, and proliferation. The pRb-responsive, cdk5-dependent phosphorylation on ezrin T235 that we have discovered is predicted to similarly provoke the open conformation of ezrin, based on the published structure,87 and indeed our experimental evidence strongly supports this.78 Unanswered by these observations is the mechanism by which ERM phosphorylation and membrane localization leads to cell shape changes in senescence. A clue to this comes from reports that the ERM proteins regulate (and are regulated by) the small GTP binding proteins of the rho family. Because the N-ERMAD has been reported to be involved in such regulation,88-89 the phosphorylation of T235 reported here may impact this function of ezrin. Further, such regulation of rho family proteins might impact upon p27KIP1 synthesis, given p27KIP1’s long-standing association with cell attachment, a process thought to involve ERM-mediated signalling.82 Indeed, translation control of p27kip1 has been reported to be subject to regulation by rho,90 providing an intriguing link between
Cancer Biology & Therapy
2003; Vol. 2 Issue 2
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
References
Figure 2. The emerging role of pRb in senescence. Inhibition of pRb phosphorylation by cdk4/6 is achieved through expression of p16INK4a; this also liberates p27Kip1 (and p21Cip1 if expressed) to inhibit cdk2 and other unidentified cdks. Recent data show that activation of pRb through this pathway results in activation of cdk5 with consequent inhibition of Rac1 and phosphorylation of ERM proteins. The impact of this on cell physiology is under investigation, but it is clear that cytoskeletal reorganization and cell shape is dependent on this process. Concomitantly, pRb increases the production of p27Kip1 to augment cell cycle arrest; crosstalk between the cdk5 arm and the p27Kip1 arm is suspected but still under investigation, as is crosstalk between Rac1 and ERM regulation. Finally, p27Kip1 regulation of cell cycle exit and senescent phenotype is likely to extend beyond inhibition of cdk2, but other targets in this process remain to be discovered.
ERM regulation by pRb and cdk5 and the processes of cell shape change and cell cycle exit in senescent cells. Together, this work, summarized in Figure 2, begins to provide a framework for the molecular events by which the process of senescence occurs downstream of senescence inducers. It seems clear that this will be found to be a complex process involving diverse signalling pathways normally observed to be used in a different context. Each of these events provides significant opportunity for further mechanistic study as well as potential entry points for therapies aimed at exploiting the processes of senescence and differentiation in cancer control. Acknowledgements Sincere apologies are offered to the myriad investigators whose work could not be cited here for the sake of brevity. Profuse thanks go to JongSeo Lee, Pedro Santiago, Gabriel Gutierrez, Bill Forrester, Karl Münger, Andi McClatchey and Jackie Lees for numerous stimulating conversations enjoyed during the course of the studies described here.
www.landesbioscience.com
1. Friend SH, Bernards R, Rogelj S, Weinberg RA, Rapaport JM, Albert DM, Dryja TP. A human DNA segment with properties of the gene that predisposes to retinoblastoma and osteosarcoma. Nature 1986; 323:643-6. 2. Weinberg RA. The retinoblastoma protein and cell cycle control. Cell 1995; 81:323-30. 3. Sherr CJ. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 2000; 60:3689-95. 4. Dyson N. The regulation of E2F by pRB-family proteins. Genes Dev. 1998; 12:2245-62. 5. Trimarchi JM, Lees JA. Sibling rivalry in the E2F family. Nat Rev Mol Cell Biol 2002; 3:11-20. 6. Geng Y, Whoriskey W, Park MY, Bronson RT, Medema RH, Li T, Weinberg RA, Sicinski P. Rescue of cyclin D1 deficiency by knockin cyclin E. Cell 1999; 97:767-77. 7. Tetsu O, McCormick F. Proliferation of cancer cells despite CDK2 inhibition. Cancer Cell 2003; 3:233-45. 8. Hinds PW. Cdk2 dethroned as master of S phase entry. Cancer Cell 2003; 3:305-307. 9. Lipinski MM, Jacks T. The retinoblastoma gene family in differentiation and development. Oncogene 1999; 18:7873-82. 10. Hickman ES, Moroni MC, Helin K. The role of p53 and pRB in apoptosis and cancer. Curr. Opin Genet Dev 2002;12(1):60-6. 11. Wu L, de Bruin A, Saavedra HI, Starovic M, Trimboli A, Yang Y, et al. Extra-embryonic function of Rb is essential for embryonic development and viability. Nature 2003; 421:942-7. 12. Classon M, Harlow E. The retinoblastoma tumour suppressor in development and cancer. Nat. Rev. Cancer 2002; 2:910-7. 13. Fajas L, Egler V, Reiter R, Hansen J, Kristiansen K, Debril MB,et al. The retinoblastomahistone deacetylase 3 complex inhibits PPARgamma and adipocyte differentiation. Dev Cell 2002; 3:903-10. 14. Fajas L, Landsberg RL, Huss-Garcia Y, Sardet C, Lees JA, Auwerx J. E2Fs regulate adipocyte differentiation. Dev Cell 2002; 3:39-49. 15. Rosen ED, Spiegelman BM. Molecular regulation of adipogenesis. Annu Rev Cell Dev Biol 2000; 16:145-71. 16. Abramson DH, Ellsworth RM, Kitchin FD, Tung G. Second nonocular tumors in retinoblastoma survivors. Are they radiation-induced? Ophthalmology 1984; 91:1351-5. 17. Gurney JG, Severson RK, Davis S, Robison LL. Incidence of cancer in children in the United States. Sex-, race-, and 1-year age-specific rates by histologic type. Cancer 1995; 75:2186-95. 18. Hansen MF, Koufos A, Gallie BL, Phillips RA, Fodstad O, Brogger A, et al. Osteosarcoma and retinoblastoma: a shared chromosomal mechanism revealing recessive predisposition. Proc Natl Acad Sci USA 1985; 82:6216-20. 19. Toguchida J, Ishizaki K, Sasaki MS, Nakamura Y, Ikenaga M, Kato M, et al. Preferential mutation of paternally derived RB gene as the initial event in sporadic osteosarcoma. Nature 1989; 338:156-8. 20. Wadayama B, Toguchida J, Shimizu T, Ishizaki K, Sasaki MS, Kotoura Y, et al. Mutation spectrum of the retinoblastoma gene in osteosarcomas. Cancer Res 1994; 54:3042-8. 21. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000; 100:57-70. 22. Cobrinik D, Lee MH, Hannon G, Mulligan G, Bronson RT, Dyson N, et al. Shared role of the pRB-related p130 and p107 proteins in limb development. Genes Dev 1996; 10:1633-44. 23. Walsh K. Coordinate regulation of cell cycle and apoptosis during myogenesis. Prog Cell Cycle Res 1997;3:53-8. 24. Chen PL, Riley DJ, Chen Y, Lee WH. Retinoblastoma protein positively regulates terminal adipocyte differentiation through direct interaction with C/EBPs. Genes Dev 1996; 10:2794-804. 25. Richon VM, Lyle RE, McGehee RE Jr. Regulation and expression of retinoblastoma proteins p107 and p130 during 3T3-L1 adipocyte differentiation. J Biol Chem 1997; 272:10117-24. 26. Gu W, Schneider JW, Condorelli G, Kaushal S, Mahdavi V, Nadal-Ginard B. Interaction of myogenic factors and the retinoblastoma protein mediates muscle cell commitment and differentiation. Cell 1993; 72:309-24. 27. Zhang JM, Wei Q, Zhao X, Paterson BM. Coupling of the cell cycle and myogenesis through the cyclin D1-dependent interaction of MyoD with cdk4. EMBO J 1999;18:92633. 28. Novitch BG, Mulligan GJ, Jacks T, Lassar AB. Skeletal muscle cells lacking the retinoblastoma protein display defects in muscle gene expression and accumulate in S and G2 phases of the cell cycle. J Cell Biol 1996; 135:441-56. 29. Novitch BG, Spicer DB, Kim PS, Cheung WL, Lassar AB. pRb is required for MEF2dependent gene expression as well as cell-cycle arrest during skeletal muscle differentiation. Curr ?ol 1999; 9:449-59. 30. Halevy O, Novitch BG, Spicer DB, Skapek SX, Rhee J, Hannon GJ, et al. Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science 1995; 267:1018-21. 31. Bodine PV, Trailsmith M, Komm BS. Development and characterization of a conditionally transformed adult human osteoblastic cell line. J Bone Miner Res 1996; 11:806-19. 32. Feuerbach D, Loetscher E, Buerki K, Sampath TK, Feyen JH. Establishment and characterization of conditionally immortalized stromal cell lines from a temperature-sensitive TAg transgenic mouse. J Bone Miner Res 1997; 12:179-90. 33. Beck GR Jr, Sullivan EC, Moran E, Zerler B. Relationship between alkaline phosphatase levels, osteopontin expression, and mineralization in differentiating MC3T3-E1 osteoblasts. Cell Biochem 1998; 68:269-80.
Cancer Biology & Therapy
129
ROLE OF THE RETINOBLASTOMA PROTEIN IN DIFFERENTIATION AND SENESCENCE
34. Hopyan S, Gokgoz N, Bell RS, Andrulis IL, Alman BA, Wunder JS. Expression of osteocalcin and its transcriptional regulators core-binding factor alpha 1 and MSX2 in osteoidforming tumours. J Orthop Res 1999; 17:633-8. 35. Hinds PW, Mittnacht S, Dulic V, Arnold A, Reed SI, Weinberg RA. Regulation of retinoblastoma protein functions by ectopic expression of human cyclins. Cell 1992; 70:993-1006. 36. Xu HJ, Zhou Y, Ji W, Perng GS, Kruzelock R, Kong CT, Bast RC, et al. Reexpression of the retinoblastoma protein in tumor cells induces senescence and telomerase inhibition. Oncogene 1997; 15:2589-96. 37. Sellers WR, Novitch BG, Miyake S, Heith A, Otterson GA, Kaye FJ, et al. Stable binding to E2F is not required for the retinoblastoma protein to activate transcription, promote differentiation, and suppress tumor cell growth. Genes Dev 1998; 12:95-106. 38. Thomas DM, Carty SA, Piscopo DM, Lee JS, Wang WF, Forrester WC, et al. The retinoblastoma protein acts as a transcriptional coactivator required for osteogenic differentiation. Mol Cell 2001; 8:303-16. 39. Karsenty G, Wagner EF. Reaching a genetic and molecular understanding of skeletal development. Dev Cell 2002; 2:389-406. 40. Xiao G, Cui Y, Ducy P, Karsenty G, Franceschi RT. Ascorbic acid-dependent activation of the osteocalcin promoter in MC3T3-E1 preosteoblasts: requirement for collagen matrix synthesis and the presence of an intact OSE2 sequence. Mol Endocrinol 1997; 11:110313. 41. Wozney JM. The bone morphogenetic protein family and osteogenesis. Mol Reprod Dev 1992; 32:160-7. 42. Yamaguchi A, Katagiri T, Ikeda T, Wozney JM, Rosen V, Wang EA, et al. Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro. J Cell Biol 1991; 113:681-7. 43. Gori F, Thomas T, Hicok KC, Spelsberg TC, Riggs BL. Differentiation of human marrow stromal precursor cells: bone morphogenetic protein-2 increases OSF2/CBFA1, enhances osteoblast commitment, and inhibits late adipocyte maturation. J Bone Miner Res 1999; 14:1522-35. 44. Spjut HJ, Dorfman HD, Fechne, RE, Ackerman LV. Tumors of bone and cartilage. In Atlas of Tumor Pathology, H.I. Firminger, Ed. (Publisher: Armed Forces Institute of Pathology, Washington), 1971; pp.141-173. 45. Ducy P, Karsenty G. Two distinct osteoblast-specific cis-acting elements control expression of a mouse osteocalcin gene. Mol Cell Biol 1995; 15:1858-69. 46. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 1997; 89:747-54. 47. Ducy P, Starbuck M, Priemel M, Shen J, Pinero G, Geoffroy V, Amling M, Karsenty G. A Cbfa1-dependent genetic pathway controls bone formation beyond embryonic development. Genes Dev. 1999;13:1025-36. 48. Ducy P, Schinke T, Karsenty G. The osteoblast: a sophisticated fibroblast under central surveillance. Science 2000; 289:1501-4. 49. Levanon D, Negreanu V, Bernstein Y, Bar-Am I, Avivi L, Groner Y. AML1, AML2, and AML3, the human members of the runt domain gene-family: cDNA structure, expression, and chromosomal localization. Genomics 1994; 23:425-32. 50. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T. Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 1997; 89:755-64. 51. Otto F, Thornell AP, Crompton T, Denzel A, Gilmour KC, Rosewell IR, et al. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development. Cell 1997; 89:765-71. 52. Kobayashi H, Gao Y, Ueta C, Yamaguchi A, Komori T. Multilineage differentiation of Cbfa1-deficient calvarial cells in vitro. Biochem. Biophys. Res Commun 2000; 273:630-6. 53. Yamaguchi A, Komori T, Suda T. Regulation of osteoblast differentiation mediated by bone morphogenetic proteins, hedgehogs, and Cbfa1. Endocr Rev 2000; 21:393-411. 54. Mundlos S, Otto F, Mundlos C, Mulliken JB, Aylsworth AS, Albright S, et al. R. Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia. Cell 1997; 89:773-9. 55. Sato M, Morii E, Komori T, Kawahata H, Sugimoto M, Terai K, et al. Transcriptional regulation of osteopontin gene in vivo by PEBP2alphaA/CBFA1 and ETS1 in the skeletal tissues. Oncogene 1998; 17:1517-25. 56. Frendo JL, Xiao G, Fuchs S, Franceschi RT, Karsenty G, Ducy P. Functional hierarchy between two OSE2 elements in the control of osteocalcin gene expression in vivo. J Biol Chem 1998; 273:30509-16. 57. Yamato K, Hashimoto S, Okahashi N, Ishisaki A, Nonaka K, Koseki T, et al. Dissociation of bone morphogenetic protein-mediated growth arrest and apoptosis of mouse B cells by HPV-16 E6/E7. Exp Cell Res 2000; 257:198-205. 58. Drissi H, Hushka D, Aslam F, Nguyen Q, Buffone E, Koff A, et al. The cell cycle regulator p27kip1 contributes to growth and differentiation of osteoblasts. Cancer Res 1999; 59:3705-11. 59. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, et al. Increased bone formation in osteocalcin-deficient mice. Nature 1996; 382:448-52. 60. Campisi, J. Replicative senescence: an old lives' tale? Cell 1996; 84:497-500. 61. Bringold F, Serrano M. Tumor suppressors and oncogenes in cellular senescence. Exp Gerontol 2000; 35:317-29. 62. Lundberg A, Hahn W, Gupta P, Weinberg R. Genes involved in senescence and immortalization. Curr Opin Cell Biol 2000; 12:705-9.
130
63. Schmitt CA, Fridman JS, Yang M, Lee S, Baranov E, Hoffman RM, et al. A senescence program controlled by p53 and p16INK4a contributes to the outcome of cancer therapy. Cell 2002; 109:335-46. 64. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB, et al. Extension of lifespan by introduction of telomerase into normal human cells. Science 1998; 279:349-52. 65. Meyerson M, Counter CM, Eaton EN, Ellisen LW, Steiner P, Caddle SD, et al. hEST2, the putative human telomerase catalytic subunit gene, is up-regulated in tumor cells and during immortalization. Cell 1997; 90:785-95. 66. Kolquist KA, Ellisen LW, Counter CM, Meyerson M, Tan LK, Weinberg RA, et al. Expression of TERT in early premalignant lesions and a subset of cells in normal tissues. Nat Genet 1998; 19:182-6. 67. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA, Klingelhutz AJ. Both Rb/p16INK4a inactivation and telomerase activity are required to immortalize human epithelial cells. Nature 1998; 396:84-8. 68. Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA, et al. Human keratinocytes that express hTERT and also bypass a p16(INK4a)-enforced mechanism that limits life span become immortal yet retain normal growth and differentiation characteristics. Mol Cell Biol 2000; 20:1436-47. 69. Opitz OG, Suliman Y, Hahn WC, Harada H, Blum HE, Rustgi AK. Cyclin D1 overexpression and p53 inactivation immortalize primary oral keratinocytes by a telomerase-independent mechanism. J Clin Invest 2001; 108:725-32. 70. Rheinwald JG, Hahn WC, Ramsey MR, Wu JY, Guo Z, Tsao H, et al. A two-stage, p16(INK4A)- and p53-dependent keratinocyte senescence mechanism that limits replicative potential independent of telomere status. Mol Cell Biol 2002; 22:5157-72. 71. Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 1997; 88:593-602. 72. Brookes S, Rowe J, Ruas M, Llanos S, Clark PA, Lomax M, et al. INK4a-deficient human diploid fibroblasts are resistant to RAS-induced senescence. EMBO J 2002; 21:2936-45. 73. Alexander K, Hinds PW. Requirement for p27(KIP1) in retinoblastoma protein-mediated senescence. Mol Cell Biol 2001; 21:3616-31. 74. Collado M, Medema R, Garcia-Cao I, Dubuisson M, Barradas M, Glassford J, et al. Inhibition of the phosphoinositide 3-kinase pathway induces a senescence-like arrest mediated by p27Kip1. J Biol Chem 2000; 275:21960-8. 75. Dai C, Enders G. p16 INK4a can initiate an autonomous senescence program. Oncogene 2000; 19:1613-22. 76. Dulic V, Beney G, Frebourg G, Drullinger L, Stein G. Uncoupling between phenotypic senescence and cell cycle arrest in aging p21-deficient fibroblasts. Mol Cell Biol 2000; 20:6741-54. 77. Dhavan R, Tsai LH. A decade of CDK5. Nat. Rev. Mol. Cell. Biol. 2001; 2:749-59. 78. Yang H-S, Hinds PW. Increased ezrin expression and activation by cdk5 coincident with acquisition of the senescent phenotype. Mol Cell 2003; (in press). 79. Louvet-Vallee S. ERM proteins: from cellular architecture to cell signaling. Biol. Cell 2000; 92:305-16. 80. Lamb RF, Ozanne BW, Roy C, McGarry L, Stipp C, Mangeat P, et al. Essential functions of ezrin in maintenance of cell shape and lamellipodial extension in normal and transformed fibroblasts. Curr Biol 1997; 7:682-8. 81. Bretscher A. Regulation of cortical structure by the ezrin-radixin-moesin protein family. Curr Opin Cell Biol 1999; 11:109-16. 82. Gautreau A, Louvard D, Arpin M. ERM proteins and NF2 tumor suppressor: the Yin and Yang of cortical actin organization and cell growth signaling. Curr Opin Cell Biol 2002; 14:104-109. 83. Gusella JF, Ramesh V, MacCollin M, Jacoby LB. Merlin: the neurofibromatosis 2 tumor suppressor. Biochim Biophys Acta 1999; 1423:M29-36. 84. Simons PC, Pietromonaco SF, Reczek D, Bretscher A, Elias L. C-terminal threonine phosphorylation activates ERM proteins to link the cell's cortical lipid bilayer to the cytoskeleton. Biochem Biophys Res Commun 1998; 253:561-5. 85. Yonemura S, Hirao M, Doi Y, Takahashi N, Kondo T, Tsukita S. Ezrin/radixin/moesin (ERM) proteins bind to a positively charged amino acid cluster in the juxta-membrane cytoplasmic domain of CD44, CD43, and ICAM-2. J Cell Biol 1998; 140:885-95. 86. Matsui T, Maeda M, Doi Y, Yonemura S, Amano M, Kaibuchi K, Tsukita S. Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J Cell Biol 1998; 140:647-57. 87. Pearson MA, Reczek D, Bretscher A, Karplus PA. Structure of the ERM protein moesin reveals the FERM domain fold masked by an extended actin binding tail domain. Cell 2000; 101:259-70. 88. Takahashi K, Sasaki T, Mammoto A, Hotta I, Takaishi K, Imamura H, et al. Interaction of radixin with Rho small G protein GDP/GTP exchange protein Dbl. Oncogene 1998; 16:3279-84. 89. Takahashi K, Sasaki T, Mammoto A, Takaishi K, Kameyama T, Tsukita S, et al. Direct interaction of the Rho GDP dissociation inhibitor with ezrin/radixin/moesin initiates the activation of the Rho small G protein. J Biol Chem 1997; 272:23371-5. 90. Vidal A, Millard SS, Miller JP, Koff A. Rho activity can alter the translation of p27 mRNA and is important for RasV12-induced transformation in a manner dependent on p27 status. J Biol Chem 2002; 277:16433-40.
Cancer Biology & Therapy
2003; Vol. 2 Issue 2
