HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Weinberg, W. C. Articles by Denning, M. F. Search for Related Content PubMed PubMed Citation Articles by Weinberg, W. C. Articles by Denning, M. F.

P21^WAF1 CONTROL OF EPITHELIAL CELL CYCLE AND CELL FATE

¹ Laboratory of Immunobiology, Division of Monoclonal Antibodies, Center for Biologics Evaluation and Research, FDA, NIH Bldg. 29B, Room 3NN04, HFM-564, Bethesda, MD 20892; and ² Department of Pathology and the Cardinal Bernardin Cancer Center, Loyola University Medical Center, 2160 S. First Avenue, Maywood, IL 60153;

*corresponding author, DMA/CBER/FDA, 29 Lincoln Drive, Bldg. 29B/Room 3NN04, HFM-564, Bethesda, MD 20892-4555, weinberg{at}cber.fda.gov

Abstract

(I) Introduction

(II) Cloning and Functional Domains

(III) Regulation of G1 Progression by p21WAF1

(IV) Regulation of p21WAF1

(V) Biological Consequences of p21WAF1 Expression

(A) P21WAF1 AND TERMINAL DIFFERENTIATION

(B) P21WAF1 AND APOPTOSIS

(VI) Gene Regulation by p21WAF1

(VII) p21WAF1 Family Members

(VIII) Conclusions

REFERENCES

Top Next

Key words. p21^WAF1, cell cycle, keratinocytes, cyclin-dependent kinase, cell transformation

(I) Introduction

Top Previous Next

 
Cell and tissue homeostasis are dependent on a tightly coordinated network of signaling pathways, resulting in a spatial and temporal balance of proliferation, growth arrest, differentiation, senescence, and apoptosis. Aberrant regulation of any of these biological processes can contribute to the onset and progression of tumorigenesis. These cell fate decisions are critically linked to cell cycle regulation, which is achieved by sequential activation and inactivation of cyclin-dependent kinases (cdks). Cdk inhibitors provide one of the mechanisms for inactivating cdks, and include the WAF1/CIP/KIP family, which exercises broad-acting inhibition of cdks involved in the G1/S transition and includes p21^WAF1, p57^KIP2, and p27^KIP1; and the INK4 family, which specifically inhibits cdk4 and cdk6 and includes p15^INK4b, p16^INK4a, p18^INK4c, and p19^INK4d (Sherr and Roberts, 1999).

A direct link between cdk regulation and cancer development was made with the discovery of p21^WAF1 as a transcriptional target of the tumor suppressor p53, which suggested that p21^WAF1 could function as the effector of p53-mediated tumor suppression (El Deiry et al., 1993; Gu et al., 1993; Harper et al., 1993; Xiong et al., 1993a). p53 is lost or mutated in a majority of human cancers, including those of the head and neck (Hainaut et al., 1998), and functional consequences of p53 inactivation can include dysregulation of each of the pathways regulating tissue homeostasis noted above. A role for p21^WAF1 in p53-mediated tumor suppression was further supported by the ability of p21^WAF1 to block tumor cell growth in vitro and in vivo (El Deiry et al., 1993; Cardinali et al., 1998). Accumulating evidence suggests that it is the primary mediator of growth suppression by p53, though not of other p53-mediated responses such as apoptosis (Deng et al., 1995; Attardi et al., 1996; Cox, 1997).

The majority of cancers of the head and neck are epithelial in origin, and > 90% of these are squamous cell carcinomas. Significant understanding of signals influencing squamous cell biology has been derived from studies of murine epidermis (Fuchs and Byrne, 1994; Yuspa, 1994; Greenhalgh et al., 1996b). Features of this model include the structural compartmentalization of proliferation and differentiation characteristic of epithelial tissues, the availability of well-defined molecular probes, and the ability to reproduce and manipulate the tightly regulated control of keratinocyte differentiation in vitro (Dlugosz et al., 1995). Furthermore, several experimental approaches, including in vitro selection assays, transgenic mouse lines, and experimental carcinogenesis studies, have been developed and have allowed for dissection of the multi-step nature of cancer development (Yuspa, 1994). In this review, we will present an overview of the structure and regulation of p21^WAF1, its mechanistic role in the regulation of cell cycle progression and gene expression, its place among other members of this cdk inhibitor family, and how expression of this protein translates into biological differences in cell behavior. Because of the preponderance of epithelially derived tumors among cancers arising from tissues of the head and neck, and the usefulness of epidermal cells as a model system for epithelial growth regulation, we will concentrate on the biological consequences of alterations in p21^WAF1 expression in studies of epithelial systems, primarily those utilizing the well-characterized squamous cell model of epidermal keratinocytes.

(II) Cloning and Functional Domains

Top Previous Next

 
The multifunctionality of the p21^WAF1 protein is underscored by its discovery in six independent laboratories, using unique approaches to explore distinct cell pathways. The protein was identified in three laboratories by virtue of its interaction with cell cycle machinery. Harper et al. (1993) used the method of two-hybrid screening to identify genes encoding proteins that interact with cdk2. Gu et al. (1993) used biochemical methods to purify and identify the same protein from cdk2/cyclin complexes. Xiong et al. (1993a) purified the protein by its co-precipitation with cyclin D followed by microsequencing and subsequent cloning. All three laboratories identified the 21-kilodalton protein product as a cdk2 inhibitor, and designated it, respectively, CIP1 (cdk interacting protein 1), CAP20 (cdk2-associated protein-20), and p21. El-Diery et al. (1993) set out to identify genes induced by over-expression of p53, and, applying a subtractive hybridization approach, cloned the identical gene, which they designated WAF1 (wild-type p53 activated-fragment 1). This brought to full circle the earlier observation that p21 protein was lost from cdk complexes isolated from p53-deficient cells (Xiong et al., 1993b). The cDNA, referred to as sdi1 (senescent cell-derived inhibitor 1), was also cloned from senescent fibroblasts, and over-expression was shown to inhibit DNA synthesis of cycling cells (Noda et al., 1994). The gene is also referred to as mda6 (melanoma differentiation antigen 6), following its cloning by subtractive hybridization from differentiating melanocytes (Jiang et al., 1995). The isolation of the gene from senescing and differentiating cells suggested a role for this protein in the growth arrest associated with normal biological processes.

p21^WAF1 is recognized to be a component of a quaternary complex including cyclins, cdks, and PCNA (a subunit of DNA polymerase delta) (Xiong et al., 1993a; Sherr and Roberts, 1995). p21^WAF1 has also been shown to form a binary complex with PCNA by direct binding of the carboxyl terminus, and through this is capable of directly regulating DNA replication (Waga et al., 1994). The domains of p21^WAF1 involved in interacting with cyclin-cdk and PCNA have been identified and found to be distinct (Chen et al., 1995; Goubin and Ducommun, 1995; Luo et al., 1995; Nakanishi et al., 1995; Zakut and Givol, 1995; Lin et al., 1996). The interaction with cyclin-cdk complexes occurs via the amino-terminal half of the molecule, which alone is required for growth inhibition of tumor cells (Chen et al., 1995; Zakut and Givol, 1995).

The carboxyl terminus of p21^WAF1 includes 2 phosphorylation sites, and differential phosphorylation of these sites may mediate cellular distribution, binding to PCNA, and cyclin-cdk complex formation, with functional consequences (Li et al., 2001; Rossig et al., 2001; Zhou et al., 2001). In addition, the carboxyl terminus harbors a caspase cleavage site that is cleaved during apoptosis, yielding a truncated protein that is unable to mediate growth arrest (Poon and Hunter, 1998; Y Zhang et al., 1999).

(III) Regulation of G1 Progression by p21WAF1

Top Previous Next

 
Cell cycle progression through the G1 phase into S is a major checkpoint for proliferating cells, and is under multiple levels of control by p21^WAF1 (Sherr and Roberts, 1999) (Fig. 1). In the absence of growth factors or other mitogenic stimuli, cells finishing mitosis are arrested in G0, an early, reversible stage of G1. During sustained mitogenic stimulation, expression of D-type cyclins is induced and gives rise to the formation of cyclin D/cdk4 and cyclin D/cdk6 complexes, which, upon phosphorylation by cyclin-activating kinase, CAK, become components of an active tetrameric complex among the cyclin, cdk, PCNA, and p21^WAF1. The active cyclin D complexes rapidly phosphorylate pRb, allowing for subsequent phosphorylation of pRb by cyclin E/cdk2 complexes. These cyclin E complexes must also be phosphorylated by CAK to be active and further phosphorylate pRb, allowing for progression past the restriction point where mitogenic stimulation is no longer required. Once pRb becomes hyperphosphorylated by the sequential activities of cdk4 and cdk2, E2F is released from inhibition by pRb, and the expression of genes required for the S phase is induced.

View larger version (22K): [in this window] [in a new window]   Figure 1. Positive and negative regulation of G1 progression by p21WAF1. p21WAF1 exerts a negative effect on G1 progression by inhibiting the activity of cyclin E/cdk2 complexes which phosphorylate pRb in mid to late G1 (1), as well as the function of PCNA in S phase (4). However, p21WAF1 can facilitate G1 phase progression as an assembly factor for cyclin D/cdk4 complexes (2). The induction of D-type cyclins in early G1 recruits p21WAF1 into these active cyclin D/cdk4 or cyclin D/cdk6 complexes, depleting the free pool of p21WAF1, and relieving the inhibition on cyclin E/cdk2 (3).

The positive effects of p21^WAF1 on G1 phase progression are largely due to its function as an assembly factor for active cyclin D/cdk complexes. Thus, p21^WAF1 stimulates the assembly, and is a component, of active cyclin D/cdk complexes (LaBaer et al., 1997; Cheng et al., 1999). The incorporation of p21^WAF1 into these cyclin D/cdk complexes can titrate free p21^WAF1 from its inhibitory effects on cyclin E/cdk2 and promote progression late in G1. Consistent with its requirement for progression through G1, p21^WAF1 is very low in quiescent cells and is induced by mitogens, along with its binding partners the D-type cyclins.

p21^WAF1 also has effects at blocking cell cycle progression in the S phase, due to its ability to bind to and inhibit PCNA (Waga et al., 1994). p21^WAF1 also inhibits the interaction between PCNA and DNA methyltransferase, suggesting that p21^WAF1 may regulate the recruitment of DNA methyltransferase to sites of newly synthesized DNA (Chuang et al., 1997). Finally, p21^WAF1 can interact with E2F and inhibit transcription from the cyclin A promoter, suggesting that p21^WAF1 can also have direct effects on the transcription of S phase genes (Delavaine and La Thangue, 1999) (see also Section VI, Gene Regulation by p21^WAF1).

p21^WAF1 is strongly implicated in DNA damage-induced growth arrest of cells in the G1 and S phases. p21^WAF1 is induced in response to a variety of DNA damaging insults, and can inhibit the activity of G1 cdks, as well as the DNA replication activity of PCNA. p21^WAF1 co-localizes with PCNA in the nucleus of cells following DNA damage, but does not interfere with the function of PCNA in DNA repair (Li et al., 1996). p21^WAF1 has also been demonstrated to inhibit the activity of the stress-activated protein kinase SAPK/JNK, and thus may be involved in integrating the MAP kinase pathway with the DNA damage response (Shim et al., 1996). Mouse embryo fibroblasts deficient in p21^WAF1 have a defect in G1 arrest following DNA damage (Brugarolas et al., 1995, 1999; Deng et al., 1995).

The p53 dependence of DNA damage-induced G1 arrest is supported by a lack of G1 arrest in keratinocytes that harbor mutant p53 (Herzinger et al., 1995), and oral squamous cell carcinoma cell lines that lack p53 (Courtois et al., 1997). However, there are p53-dependent and -independent mechanisms to induce p21^WAF1, and it is less clear what the contribution is of p21^WAF1 to the DNA damage response in cells with mutant or deleted p53. For example, p21^WAF1 protein and mRNA are induced by UV radiation in mouse keratinocytes lacking p53 in vivo and in vitro, and this induction is postulated to provide a back-up protective mechanism in cells with inactive p53 (Liu et al., 1999). It remains to be determined if keratinocytes from p53 null mice that induce p21^WAF1 in response to UV radiation also fail to undergo a G1 arrest.

Elevated levels of p21^WAF1 have been observed in some squamous carcinomas, suggesting an inability of p21^WAF1 to inhibit cell cycle progression in cancer cells. The reason for this decreased responsiveness is still unclear and is an important topic for future research [see also Section (V) (C) p21^WAF1 and Carcinogenesis].

(IV) Regulation of p21WAF1

Top Previous Next

 
The dual role of p21^WAF1 in growth suppression and cell cycle progression suggests a delicately balanced regulation of p21^WAF1 protein levels. p21^WAF1 was initially isolated on the basis of its induction by p53, and, as discussed above, is now understood to be a well-defined cellular response to DNA damage (El Deiry et al., 1993). However, several p53-independent pathways have been identified (Russo et al., 1995; Alpan and Pardee, 1996; Zeng and El Deiry, 1996; Haapajarvi et al., 1999), and induction of p21^WAF1 in development has been shown to occur independently of p53 (El Deiry et al., 1995; Halevy et al., 1995; Macleod et al., 1995; Parker et al., 1995). Multiple factors involved in growth regulation, development, and differentiation up-regulate p21^WAF1 independently of p53, including growth factors (Michieli et al., 1994; Y Liu et al., 1996), cytokines such as TGF-ß (Datto et al., 1995a; Li et al., 1995; Hu et al., 1999), glucocorticoids (Corroyer et al., 1997), and retinoids (M Liu et al., 1996; Yang et al., 1999).

Induction of p21^WAF1 by p53 requires the transcription factor Sp1 (Koutsodontis et al., 2001) and an intact p53 binding site localized far (> 1.9 kb) upstream of the coding sequence (El Deiry et al., 1993; Macleod et al., 1995). A proximal promoter sequence located between -119 bp and the p21^WAF1 transcriptional start site that includes 6 Sp1 binding sites and the TATA box plays a major role in p53-independent transcriptional regulation of p21^WAF1 (Gartel and Tyner, 1999). Within the proximal promoter, a 78-base-pair GC-rich sequence has been defined that binds the Sp1 and Sp3 transcription factors. Binding of either Sp1 or Sp3 activates basal promoter activity, while binding of Sp3 is critical for promoter inducibility upon induction of differentiation in keratinocytes by elevated extracellular calcium (Prowse et al., 1997). This promoter sequence encompasses the consensus Sp1/Sp3 binding site identified for p21^WAF1 inducibility by TGF-ß in HaCat cells (Datto et al., 1995b), and by the retinoblastoma gene product in MDCK cells (Decesse et al., 2001), and is distinct from the p53 consensus sequence (Datto et al., 1995a). The transcriptional co-activator p300 is also required for the calcium-induced, p53-independent induction of p21^WAF1 in murine keratinocytes (Missero et al., 1995). p21^WAF1 transcription is also subject to down-regulation by c-myc over-expression (Mitchell and El Deiry, 1999), probably due to interaction with and sequestration of Sp1 (Gartel et al., 2001); this suppression may contribute to c-myc-mediated transformation.

p21^WAF1 levels are also regulated post-transcriptionally. p21^WAF1 is subject to proteosome-dependent degradation (Blagosklonny et al., 1996; Rousseau et al., 1999; Sheaff et al., 2000), which can be differentially modulated by interaction with cdks or PCNA (Cayrol and Ducommun, 1998). This mechanism may contribute to the down-regulation of p21^WAF1 that has been associated with terminal differentiation of keratinocytes (Di Cunto et al., 1998; Harvat et al., 1998). Protein stability can also be modulated by phosphorylation status (Li et al., 2001; Rossig et al., 2002), and expression levels can be enhanced by increased mRNA stability, as by treatment with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) (Park et al., 2001) and glucocorticoids (Corroyer et al., 1997).

(V) Biological Consequences of p21WAF1 Expression

Top Previous Next

 
    (A) P21^WAF1 AND TERMINAL DIFFERENTIATION
Differentiation of multiple tissue types, including squamous epithelia, is associated with an increase in p21^WAF1 expression (El Deiry et al., 1995; Parker et al., 1995). Terminal differentiation in stratified epithelia is a complex process involving irreversible growth arrest and sequential changes in gene expression, which accompany upward migration of cells from the basal to the superficial layers. p21^WAF1 null mice develop normally (Deng et al., 1995), and there are no apparent abnormalities in the morphology, in vivo labeling index, or expression patterns of differentiation-specific genes of the epidermis of these mice (Weinberg et al., 1999; Paramio et al., 2001). However, a decrease in senescence-associated beta-galactosidase activity of epidermal cells has been observed by histochemistry (Paramio et al., 2001).

The tightly coordinated program of gene expression in normal skin can be mimicked in vitro by manipulation of the calcium concentration in the medium, such that keratinocytes cultured under low calcium conditions maintain a cuboidal morphology and continue to proliferate. When the extracellular calcium concentration is increased from 0.05 mM to > 0.1 mM, murine keratinocytes cease proliferating and, within 4-8 hrs, begin to develop a squamous morphology. This is accompanied by many of the molecular changes in differentiation marker gene expression that are observed in vivo, including increases in p21^WAF1 levels (Hennings et al., 1980; Yuspa et al., 1989; Missero et al., 1995; Weinberg et al., 1995). Differentiation can also be triggered in vitro by the placement of cells into suspension culture; spontaneously detached cells can also be analyzed (Green, 1977; Rheinwald and Beckett, 1980). This disruption of cell-matrix adhesion is a very potent differentiation signal that may mimic the detachment from the basement membrane that occurs during terminal differentiation in vivo. Additional means whereby keratinocytes can be induced to differentiate in vitro include exposure to TPA and to interferon . These agents also trigger a transient p21^WAF1 induction and irreversible growth arrest (Todd and Reynolds, 1998; Harvat and Jetten, 2001; Wan et al., 2001).

Several studies have examined the expression kinetics of endogenous cyclin-dependent kinase inhibitors during keratinocyte terminal differentiation in vitro (Missero et al., 1995, 1996; Weinberg et al., 1995; Alani et al., 1998; Harvat et al., 1998; Martinez et al., 1999), and the contribution of p21^WAF1 to growth arrest and induction of epidermal differentiation gene products has been tested in both knockout mice and over-expression systems. Although the exact kinetics differs between calcium- and suspension-induced differentiation, in general, a sequential induction of p21^WAF1, p27^KIP1, and p16^INK4a is associated with differentiation of normal keratinocytes in vitro (Missero et al., 1996; Hauser et al., 1997; Harvat et al., 1998; Martinez et al., 1999). The induction of p21^WAF1 is transient, reaching a maximum at 4-12 hrs after differentiation is triggered, and returns to baseline by 24 hrs. These in vitro data are supported by the restricted staining pattern of p21^WAF1 immediately above the proliferating cells in the hair follicles and sebaceous glands of human skin (Ponten et al., 1995). The induction of p27^KIP1 and p16^INK4a is delayed but progressive, suggesting that these cdk inhibitors may be responsible for maintaining terminally differentiated cells in a post-mitotic state. Nevertheless, p21^WAF1-deficient keratinocytes have been shown to re-initiate DNA synthesis after exposure to elevated calcium for 3-6 days, suggesting that p21^WAF1 is required for initiating the irreversible growth arrest of differentiated keratinocytes (Topley et al., 1999). Whether the lack of p21^WAF1 prevents the sustained induction of p27^KIP1 and p16^INK4a is unclear, and determining the mechanism for the failure of p21^WAF1 -/- cells to execute cell cycle withdrawal warrants further investigation.

In vitro studies from independent laboratories have yielded differing results regarding the proliferative capacity and differentiation potential of p21^WAF1 null keratinocytes (Missero et al., 1996; Paramio et al., 2001). In studies by Paramio et al. (2001), p21^WAF1 gene status alone had no effect on the cell labeling indices or growth rates of monolayer cultures of primary keratinocytes under basal conditions or following exposure to elevated calcium or TGF-ß. In addition, no effect was observed on differentiation potential, as determined by induction of differentiation markers K10 and filaggrin (Paramio et al., 2001). These results are consistent with findings reported in vivo (Weinberg et al., 1999). In contrast, in a separate study with a similar culture model, Brdu incorporation levels and FACS analysis revealed a two-fold increase in the nuclear labeling index and S phase population of p21^WAF1 null keratinocytes. No change in the time or extent of calcium-induced growth arrest or TGF-beta responsiveness was observed, though p21^WAF1 null keratinocytes displayed a shortened G1 phase (Missero et al., 1996). However, a dramatic decrease in induction of differentiation markers keratin 1, loricrin, and involucrin, but not filaggrin, was observed in the p21^WAF1 null keratinocytes (Missero et al., 1996). It has been suggested that these findings are due to an altered balance between growth and differentiation rather than to impaired progression through differentiation (Di Cunto et al., 1998).

The differences observed across laboratories in differentiation potential of p21^WAF1 null keratinocytes might be at least partly explained by variations in culture conditions. For example, in the studies by Missero et al. (1995, 1996), the culture medium included epidermal growth factor (EGF) as a component. Selective concentrations of EGF can induce p21^WAF1 (Michieli et al., 1994; Fan et al., 1995; Jakus and Yeudall, 1996). Furthermore, EGF and EGF-receptor ligands re-direct the differentiation program of primary murine epidermal keratinocytes in response to elevated calcium concentrations (Cheng et al., 1993). Thus, the inclusion of EGF in culture medium might accentuate any subtle differences due to p21^WAF1 gene loss.

A dramatic block in expression of 4 differentiation-specific gene products in the differentiating cell population was also observed following over-expression of p21^WAF1 by an adenoviral vector (Di Cunto et al., 1998). The dual role of p21^WAF1 in G1 progression may help explain the conflicting results from p21^WAF1 over-expression and deletion studies, both of which showed a block in differentiation marker expression (Missero et al., 1996; Di Cunto et al., 1998), since over-expression and complete absence of p21^WAF1 can both impede cells from progressing through G1. However, in studies of suspension cultures of primary human epidermal keratinocytes, over-expression of p21^WAF1 by adenoviral vector did not block expression of the differentiation-specific proteins cornifin or involucrin (Harvat et al., 1998).

p21^WAF1 appears also to be involved in the induction of growth arrest by differentiation signaling intermediates such as NF-B and Notch1. NF-B activation occurs in the first differentiating layer of normal epidermis, and ectopic expression of NF-B subunits in the basal layer of mouse epidermis triggers growth arrest accompanied by the induction of p21^WAF1 (Seitz et al., 1998, 2000). Over-expression of active NF-B did not induce p27^KIP1, p57^KIP2, or p16^INK4A, suggesting that p21^WAF1 is the cdk inhibitor responsible for the growth arrest. The NF-B-induced growth arrest was attenuated in the epidermis of p21^WAF1 -/- mice, supporting the role of p21^WAF1 in at least part of the growth inhibition by NF-B. A similar situation exists for Notch1 (Rangarajan et al., 2001). Notch1 is required for induction of p21^WAF1 in mouse keratinocytes, and forced expression of Notch1 can trigger p21^WAF1 and growth arrest. Furthermore, Notch1 fails to trigger growth arrest in p21^WAF1 -/- keratinocytes (Rangarajan et al., 2001). Taken together, these results suggest that the induction of p21^WAF1 is a common intermediate for multiple signaling pathways that trigger growth arrest.

    (B) P21^WAF1 AND APOPTOSIS
p21^WAF1 clearly has a role in responding to apoptotic stimuli by being a p53 target gene induced in response to DNA damage, and aiding in cell cycle arrest (El Deiry et al., 1994; Brugarolas et al., 1995, 1999). However, p21^WAF1 is also a substrate for the effector caspase-3 (Y Zhang et al., 1999), and proteolysis of p21^WAF1 may be involved in the abandonment of cell cycle arrest in favor of apoptotic cell death. Human p21^WAF1 is cleaved by caspase-3 in apoptotic cells at DHVD¹¹² to remove 52 amino acids from its C terminus (Y Zhang et al., 1999). This truncated p21^WAF1 lacks the PCNA binding domain, the second cyclin binding domain, and the nuclear localization signal, and thus is cytoplasmic and inactive at mediating growth arrest (Poon and Hunter, 1998; Y Zhang et al., 1999). The inactivation of p21^WAF1 during the terminal phase of apoptosis may be an important switch between growth arrest and apoptosis.

p21^WAF1 can also form complexes with the inactive procaspase-3 to inhibit the activation of caspase-3 (Suzuki et al., 1998, 1999). Formation of the p21^WAF1/procaspase-3 complex is in part promoted by protein kinase A phosphorylation of p21^WAF1 (Suzuki et al., 2000). Thus, the induction of p21^WAF1 in response to apoptotic insults arrests cell cycle progression via the inhibition of cdk2 and cdk4, and may also help prevent the execution phase of apoptosis by binding to and inhibiting procaspase-3. The inhibition of procaspase-3 activation may be overcome by a decrease in ATP production due to mitochondrial dysfunction, which would inhibit the phosphorylation of p21^WAF1 and its binding to procaspase-3. This would make procaspase-3 available for activation, which in turn could then cleave p21^WAF1 and disrupt its ability to mediate cell cycle arrest, creating a very tight, all-or-nothing commitment to apoptosis. Consistent with this, cells derived from p21^WAF1 null mice display an enhanced apoptotic response to radiation of both intestinal cells and thymocytes, compared with wild-type mice (Wang et al., 1997; Martin-Caballero et al., 2001). This tight control over apoptosis is an important safeguard against malignant transformation, and cells resistant to apoptosis are prone to transformation. Indeed, p21^WAF1 null mice display a delayed incidence of radiation-induced thymic lymphomas compared with wild-type mice (Wang et al., 1997; Martin-Caballero et al., 2001).

(C) P21^WAF1 AND CARCINOGENESIS

In contrast to p53, p21^WAF1 mutations rarely occur in human cancers (Shiohara et al., 1994). However, the protein has been shown by immunohistochemistry to be over-expressed in human cutaneous squamous cell carcinomas (Tron et al., 1996), and dysregulation of p21^WAF1 protein expression has been observed by immunohistochemical analysis in a majority of squamous cell carcinomas of the head and neck and adjacent oral dysplasias (Erber et al., 1997; van Oijen et al., 1998; Yook and Kim, 1998). This protein expression is independent of p53 status and is regulated post-transcriptionally. Unlike normal mucosa (El Deiry et al., 1995; Parker et al., 1995), p21^WAF1 expression in these lesions is observed in proliferating cells, as determined by co-expression with PCNA or Brdu; thus, the growth-inhibitory function of p21^WAF1 in these tumors is abrogated. This may reflect a direct association of p21^WAF1 and PCNA in binary complexes, changes in the stoichiometry of quaternary p21^WAF1 and cdk complexes, preferential expression of an alternative form of p21^WAF1, altered processing of the protein, changes in the ability of tumor cells to recruit p21^WAF1 into cdk2 complexes, and/or altered expression of other cell cycle regulatory molecules such as cyclin D1 (Erber et al., 1997; Poon and Hunter, 1998; van Oijen et al., 1998; Yook and Kim, 1998; Harvat et al., 2001). These observations suggest that altered expression of p21^WAF1 may be an early event contributing to or reflecting the malignant transformation of head and neck squamous epithelium, and may serve as a therapeutic target or predictive indicator.

The availability of genetically engineered mice harboring null mutations in the p21^WAF1 gene (Deng et al., 1995) has allowed for the further analysis of p21^WAF1 gene contribution to tumor formation. Whereas p53 null mice develop tumors of multiple cell lineages, initial reports of p21^WAF1 null mice suggested that the loss of p21^WAF1 does not increase spontaneous tumor frequency (Deng et al., 1995). However, a recent study evaluating tumor formation found that p21^WAF1 null mice demonstrate an increased incidence of tumor formation when evaluated over a two-year time frame, with tumors being a contributing cause of death in 55% of males and 27% of females (Martin-Caballero et al., 2001). The higher incidence of tumors among males reflects an elevated incidence of glomerulonephritis, with increased severity in females due to abnormal T-cell proliferation occurring prior to tumor formation, and does not necessarily reflect a gender-specific susceptibility to tumor formation. The majority of tumors that arise in the p21^WAF1 null mice are histiocytic sarcomas, hemangiomas, hemangiosarcomas, and B-cell lymphomas. In contrast to p53 null mice, none of the tumors observed was T-cell in origin, and approximately 10% of tumors were derived from epithelium. Of 65 tumors arising in 151 mice, 2 were epidermal tumors, inclu-ding one benign keratoacanthoma and one sebaceous gland adenoma. Thus, while the tumor latency is substantially longer in p21^WAF1 null mice (13-18 mos) vs. p53 null mice (5 mos), p21^WAF1 null mice appear to have a modest increase in spontaneous tumor formation. p21^WAF1 loss does not appear, however, to increase spontaneous transformation of epidermal cells.

The contribution of p21^WAF1 to epidermal homeostasis might be more readily detected in the presence of an additional genetic change that might increase the stress response. To this end, several laboratories have evaluated the phenotype of murine keratinocytes following introduction of the v-Ha-ras oncogene, which by itself is capable of transforming wild-type keratinocytes and conferring a benign papilloma phenotype. When primary keratinocytes from p21^WAF1 (+/+), (+/-), and (-/-) were transduced in vitro with a retroviral vector encoding a mutant v-Ha-ras oncogene, the three genotypes’ growth arrested to the same extent in response to elevation of extracellular calcium to 1.3 mM (Weinberg et al., 1997). However, p21^WAF1 (-/-) keratinocytes expressing mutant v-Ha-ras displayed a decreased responsiveness to growth suppression by TGF-beta, most notable at doses > 1 nM. When these cultures were grafted with normal dermal cells onto the dorsal surfaces of nude mice, papillomas developed that were phenotypically indistinguishable from those in genetically matched p21^WAF1 (+/+) control cultures. This was in contrast to similar studies with p53 null keratinocytes, which rapidly formed undifferentiated carcinomas (Weinberg et al., 1994). In a similar approach, Paramio et al. (2001) observed a decrease in senescence-associated beta-galactosidase staining in v-Ha-ras-transduced p21^WAF1 null cells, reflecting a decrease in the v-Ha-ras-induced senescent cell population (31% vs. 55% in p21^WAF1+/+) (Paramio et al., 2001). When v-Ha-ras-transduced p21^WAF1 null primary keratinocytes were injected subcutaneously into nude mice, only 1/6 mice developed a tumor, which histologically was a well-differentiated squamous cell carcinoma (Paramio et al., 2001). This is consistent with the inability of benign papillomas to grow in a subcutaneous environment (Strickland et al., 1988; Paramio et al., 2001). These findings were also consistent with results from an in vitro transformation assay in which there was only a slight increase in focus formation from v-Ha-ras-transduced primary keratinocytes cultured for 2 wks in 0.5% serum (Paramio et al., 2001). In contrast to the studies by Paramio et al. (2001) and Weinberg et al. (1997), Missero et al. (1996) found that p21^WAF1 null keratinocytes displayed an aggressive tumorigenic phenotype following introduction of the v-Ha-ras oncogene and subcutaneous injection into nude mice. Again, the differences among these studies may be attributable in part to differences in culture conditions, in that the addition of EGF to the culture medium used by Missero et al. (1996) could select for different target cell populations with enhanced potential for transformation. In addition, elevated v-Ha-ras levels have been correlated with increased malignancy (Brissette et al., 1993), and the viral titers may have varied across studies. This could have magnified effects if the target cell populations display genotype-specific differences in proliferative potential, as has been described (Missero et al., 1996).

Several groups have applied a classic two-stage protocol of skin tumor induction in p21^WAF1 null mice to evaluate the stages of carcinogenesis during which p21^WAF1 plays a critical role. This regimen dissects the process of carcinogenesis into four defined stages: initiation, defined as mutation of target cells; promotion, or selective growth of the initiated population; conversion of the benign tumor cell population into a malignant carcinoma; and progression of the carcinoma into a more poorly differentiated phenotype. The model entails applying a single dose of carcinogen to the dorsum of mice as an initiating agent, followed by multiple applications of a tumor promoter, in this case the phorbol ester TPA. TPA alone induces p21^WAF1 expression in murine keratinocytes both in vitro (Wan et al., 2001) and in vivo (Greenhalgh et al., 1996a), and p21^WAF1 protein expression is increased in papillomas and carcinomas arising from this protocol (Philipp et al., 1999).

In a study by Weinberg et al. (1997), an increase in benign tumor formation was observed in p21^WAF1 null mice, with no influence on tumor growth rates or conversion frequency, suggesting that p21^WAF1 functions to block papilloma formation. A similar enhancement of tumor formation with p21^WAF1 gene inactivation has been observed in intestinal (Yang et al., 2001) and mammary tumor models (Adnane et al., 2000). Following a similar protocol, Philipp et al. (1999) found a small increase in papilloma formation in p21^WAF1 null that was statistically insignificant. This study used a four-fold-higher dose of initiating agent and 2.5-fold more TPA, which, along with the genetic background of the strains used, resulted in higher overall levels of papillomas that may have reduced any p21^WAF1-mediated differences in susceptibility. Consistent with Weinberg et al. (1999), there was no observed change in the growth of the resulting tumors. In both studies, the percentage of mice developing carcinomas was increased in the p21^WAF1 null mice; this is presumably a reflection of the increased papilloma yield in the p21^WAF1 null animals, since the percentage of papillomas converting to carcinomas remained low and was similar across genotypes in both studies (Weinberg et al., 1997; Philipp et al., 1999). Keratinocytes of each genotype behaved similarly as well in an in vitro model of malignant conversion (Weinberg et al., 1999). These studies suggest that p21^WAF1 does not act to block the conversion stage of carcinogenesis.

Conflicting results were reported in another study, in which p21^WAF1 loss resulted in an increased susceptibility to skin carcinoma formation with decreased papilloma development (Topley et al., 1999). The duration of the promotion phase in this study was two months, compared with a minimum of 15 weeks in the other two studies. It has been established that decreasing the period of tumor promotion by TPA results in a lower tumor yield; however, the resulting papillomas exhibit an increased predisposition to malignant conversion (Hennings et al., 1985). The mice in the study by Topley et al. (1999) were promoted with TPA for only 8 wks; thus, the resulting tumors may represent papillomas with a "high risk" phenotype. The consequence of p21^WAF1 loss in this study may be due to effects on progression and may become apparent only after conversion to carcinoma has occurred. This is consistent with the decreased differentiation levels observed in carcinomas arising in p21^WAF1 null mice, with no concomitant increase in carcinoma frequency (Philipp et al., 1999). The same model could also explain the growth of malignant cells arising from subcutaneous injection of v-Ha-ras-transduced keratinocytes, since this system requires in vivo selection of malignant tumor cells, as previously discussed.

The low level of spontaneous tumor formation in p21^WAF1 null mice suggests that, unlike p53, p21^WAF1 is not a general mediator of tumor suppression. Rather, the contribution of p21^WAF1 to p53-mediated tumor suppression may be dependent on the carcinogenic stimuli. In the case of a p53 stress response to apoptotic stimuli, such as radiation exposure, loss of p21^WAF1 does not enhance tumor formation, as demonstrated by the delayed onset of thymic lymphomas (Martin-Caballero et al., 2001). Alternatively, if p53 tumor suppression relies on growth arrest and blocking proliferative stimuli, such as appears to be the case in two-stage carcinogenesis of epidermis, p21^WAF1 loss enhances neoplasia (Philipp et al., 1999; Weinberg et al., 1999). p21^WAF1 may also contribute to maintaining a differentiated phenotype in tumor cells later in the process, subsequent to malignant conversion (Missero et al., 1996; Philipp et al., 1999).

We have integrated the somewhat conflicting data on p21^WAF1 and carcinogenesis into a working model that fits the majority of the data (Fig. 2). p21^WAF1 expression during normal cell differentiation, growth arrest, and senescence is an integral part of normal cellular physiology (El Deiry et al., 1995; Parker et al., 1995), but its loss can be compensated for by related proteins (P Zhang et al., 1999). Perhaps due to its response to DNA damage, p21^WAF1 has an inhibitory effect on papilloma formation. This is supported by the small increase in spontaneous tumors observed in aging p21^WAF1 null mice (Martin-Caballero et al., 2001) and the increased papilloma yield in chemical carcinogenesis studies of p21^WAF1 null mouse skin (Philipp et al., 1999; Weinberg et al., 1999). In contrast, conversion of benign tumors to carcinoma does not appear to be under p21^WAF1 regulation, and the conversion frequency of papillomas to carcinomas in p21 null mice is not elevated (Weinberg et al., 1997; Philipp et al., 1999). However, the progression to less-differentiated tumors (i.e., spindle cell carcinoma) may be under negative regulation by p21^WAF1, as supported by the increase in carcinoma grade observed in p21^WAF1 -/- mice in some studies (Missero et al., 1996; Philipp et al., 1999).

View larger version (9K): [in this window] [in a new window]   Figure 2. Stage-specific roles for p21WAF1 in epithelial carcinogenesis. p21WAF1 is involved in several normal cellular processes, including cell cycle progression, cell differentiation, and senescence. p21WAF1 also has an inhibitory effect on the initiation/promotion stage(s) of carcinogenesis, thereby reducing the formation of papillomas, but does not affect conversion of these benign tumors to carcinomas. Once a carcinoma is established, however, p21WAF1 helps maintain it in a well-differentiated state, thus impeding progression to spindle cell carcinomas.

Top Previous Next

A growing body of work has given strong support to the ability of p21^WAF1 to regulate, both positively and negatively, the expression of genes involved in growth arrest, senescence, and aging. Over-expression of p21^WAF1 is sufficient to induce a senescence-like growth arrest in many cell types (Chang et al., 2000a,b), and the expression of numerous genes associated with senescence and aging is up-regulated (McConnell et al., 1998; Chang et al., 2000b). The genes induced by p21^WAF1 include many extracellular matrix components and secreted proteases, suggesting a paracrine-growth-stimulating activity (Chang et al., 2000b). Equally impressive is the down-regulation of a large number of genes involved in DNA replication, repair, and mitosis by ectopic p21^WAF1 expression (Harvat et al., 1998; Chang et al., 2000a,b). Furthermore, anti-cancer drugs that trigger growth arrest also induce p21^WAF1, in addition to the same battery of senescence-related genes induced following p21^WAF1 over-expression (Chang et al., 1999; Kramer et al., 2001). The coordinated changes in gene expression induced by ectopic p21^WAF1 expression are consistent with the profound effects of p21^WAF1 on growth arrest, and may represent a major function of p21^WAF1 in growth arrest independent of its inhibitory activity of cyclin-dependent kinases.

(VII) p21WAF1 Family Members

Top Previous Next

 
p21^WAF1 belongs to the Kinase Inhibitor Protein (KIP) class of cdk inhibitors, which includes p27^KIP1 and p57^KIP2 (Sherr and Roberts, 1999). The KIP class of cdk inhibitors shares structural and functional similarities, most notably a domain in their amino-terminus that mediates binding to cyclin and cdk subunits, and their ability to inhibit cdk activity when complexed with cyclin A, B, D, or E. These features distinguish them from the INK4 family of cdk inhibitors (p16^INK4a, p15^INK4b, p18^INK4c, and p19^INK4d) that inhibit only cyclin-D-associated ckd2, ckd4, or cdk6 (Sherr and Roberts, 1999). Similar to p21^WAF1, p27^KIP1 was identified by multiple approaches, including its ability to bind and inhibit the activity of cyclin E/ckd2 complexes (Polyak et al., 1994a), and by a yeast two-hybrid screen with cdk4 as the "bait" (Toyoshima and Hunter, 1994). In addition to the extensive sequence homology in their amino-terminal domains, p27^KIP1 shares many similar activities with p21^WAF1, such as inhibition of CAK, and the ability to promote cdk/cyclin D activity at stoichiometric levels (Polyak et al., 1994b; Sherr and Roberts, 1995, 1999). p27^KIP1 is also induced during keratinocyte differentiation and can trigger growth arrest (Hauser et al., 1997; Harvat et al., 1998), although antisense p27^KIP1 inhibited suspension-induced differentiation, but not growth arrest (Hauser et al., 1997). Over-expression of p27^KIP1 in keratinocytes did not affect differentiation marker expression (Harvat et al., 1998), while p21^WAF1 over-expression has been shown to suppress marker expression in some studies (Di Cunto et al., 1998), but not others (Harvat et al., 1998). Both p21^WAF1 and p27^KIP1 have a role in carcinogenesis, with p27^KIP1 -/- mice developing earlier and higher-grade carcinomas in the two-stage chemical carcinogenesis protocol (Philipp et al., 1999). p21^WAF1 and p27^KIP1 also have differences in their regulation, with p27^KIP1 levels decreasing in response to mitogens while p21^WAF1 is induced. p57^KIP2 also contains the characteristic amino-terminal cdk inhibitory domain, and appears not to be as widely expressed; however, expression in keratinocytes is up-regulated at the protein level during calcium-induced differentiation (Martinez et al., 1999).

Individual INK4 family members are induced during growth arrest by a variety of agents, during cell cycle progression, and during development, but their ability to arrest cells in G1 depends on pRb function (Sherr and Roberts, 1999). p16^INK4a is closely associated with preventing escape of normal cells from replicative senescence, and the gene is mutated, deleted, or silenced in most immortalized keratinocytes (Symington and Carter, 1995; Loughran et al., 1996; Enders et al., 1996; Munro et al., 1999; Dickson et al., 2000). The p16^INK4a locus also encodes the tumor suppressor ARF in an alternative reading frame (Serrano et al., 1996), and ARF is able to induce a cell cycle arrest by indirectly causing the stabilization of p53 (Sharpless and DePinho, 1999). The p16^INK4a locus can cooperate with p21^WAF1 in the induction of senescence (Paramio et al., 2001). Keratinocytes that are deficient in p21^WAF1 and both p16^INK4a/ARF gene products resist senescence and form poorly differentiated spindle tumors when transduced with the v-Ha-ras oncogene (Paramio et al., 2001). Thus, p21^WAF1 and p16^INK4a/ARF gene products may be able to serve compensatory roles in triggering growth arrest.

(VIII) Conclusions

Top Previous Next

 
p21^WAF1 is aberrantly expressed in human cancers, and this expression appears to be unlinked to p53 status. It is unclear whether the dysregulation of p21^WAF1 expression in cancer contributes to tumorigenesis by positive regulation of cell cycle progression or inhibition of apoptosis, as noted above, or represents an attempt at a well-defined protective cellular response to escape tumorigenesis by blocking cell cycle progression. Nonetheless, numerous growth-regulatory, chemopreventive, and chemotherapeutic agents appear to invoke p21^WAF1 as at least part of their mechanism of action (Datto et al., 1995a; Li et al., 1995; M Liu et al., 1996; Lepley and Pelling, 1997; Chang et al., 1999; Yang et al., 1999). Furthermore, forced over-expression of p21^WAF1 in xenograft models of brain (Chen et al., 1996) and prostate (Eastham et al., 1995) cancer as well as squamous cell carcinomas of the head and neck (Cardinali et al., 1998) suppressed or delayed tumor growth. Taken together, these findings suggest that p21^WAF1 could be a strategic target in cancer therapy.

Recent advances in the development of low-molecular-weight mimetics based on p21^WAF1 structure have been shown to be effective at binding cyclin-cdk complexes and inhibiting kinase activity, as well as inhibiting tumor cell growth (Ball et al., 1997; Bonfanti et al., 1997; Mutoh et al., 1999). These peptides can be administered alone or in combination with DNA-damaging agents. The latter approach would exploit the dual effects of p21^WAF1 on cell cycle progression and could serve to decrease adverse side-effects of chemotherapy by increasing specificity to the target cancer cells which may be less responsive to a p21^WAF1-induced block in cell cycle regulation than their normal counterparts (Blagosklonny and Pardee, 2001).

In experimental models, p21^WAF1 has been shown to act at specific stages of carcinogenesis (see Fig. 2). p21^WAF1 loss can be compensated for by other family members in development and tumor suppression (P Zhang et al., 1999; Franklin et al., 2000), but can also cooperate with other genetic changes in modulating tumorigenesis (Wang et al., 1997; Adnane et al., 2000; Franklin et al., 2000; Yang et al., 2001; Paramio et al., 2001). Thus, knowledge of specific genetic alterations correlating with defined stages of the neoplastic process may prove useful in determining the application of p21^WAF1 to the optimal strategies for chemoprevention and cancer therapy.

Acknowledgments

 
We thank Drs. Chuxia Deng and Henry Hennings for critical reading of the manuscript.

REFERENCES

Top

 
Adnane J, Jackson RJ, Nicosia SV, Cantor AB, Pledger WJ, Sebti SM (2000). Loss of p21^WAF1/CIP1 accelerates Ras oncogenesis in a transgenic/knockout mammary cancer model. Oncogene 19:5338–5347.[Medline]

Alani RM, Hasskarl J, Munger K (1998). Alterations in cyclin-dependent kinase 2 function during differentiation of primary human keratinocytes. Mol Carcinog 23:226–233.[Medline]

Alpan RS, Pardee AB (1996). p21^WAF1/CIP1/SDI1 is elevated through a p53-independent pathway by mimosine. Cell Growth Differ 7:893–901.[Abstract]

Attardi LD, Lowe SW, Brugarolas J, Jacks T (1996). Transcriptional activation by p53, but not induction of the p21 gene, is essential for oncogene-mediated apoptosis. EMBO J 15:3693–3701.[Medline]

Ball KL, Lain S, Fahraeus R, Smythe C, Lane DP (1997). Cell-cycle arrest and inhibition of Cdk4 activity by small peptides based on the carboxy-terminal domain of p21^WAF1. Curr Biol 7:71–80.[Medline]

Blagosklonny MV, Pardee AB (2001). Exploiting cancer cell cycling for selective protection of normal cells. Cancer Res 61:4301–4305.[Abstract/Free Full Text]

Blagosklonny MV, Wu GS, Omura S, El Deiry WS (1996). Proteasome-dependent regulation of p21^WAF1/CIP1 expression. Biochem Biophys Res Commun 227:564–569.[Medline]

Bonfanti M, Taverna S, Salmona M, D’Incalci M, Broggini M (1997). p21^WAF1-derived peptides linked to an internalization peptide inhibit human cancer cell growth. Cancer Res 57:1442–1446.[Abstract/Free Full Text]

Brissette JL, Missero C, Yuspa SH, Dotto GP (1993). Different levels of v-Ha-ras p21 expression in primary keratinocytes transformed with Harvey sarcoma virus correlate with benign versus malignant behavior. Mol Carcinog 7:21–25.[Medline]

Brugarolas J, Chandrasekaran C, Gordon JI, Beach D, Jacks T, Hannon GJ (1995). Radiation-induced cell cycle arrest compromised by p21 deficiency. Nature 377:552–557.[Medline]

Brugarolas J, Moberg K, Boyd SD, Taya Y, Jacks T, Lees JA (1999). Inhibition of cyclin-dependent kinase 2 by p21 is necessary for retinoblastoma protein-mediated G1 arrest after gamma-irradiation. Proc Natl Acad Sci USA 96:1002–1007.[Abstract/Free Full Text]

Cardinali M, Jakus J, Shah S, Ensley JF, Robbins KC, Yeudall WA (1998). p21(WAF1/Cip1) retards the growth of human squamous cell carcinomas in vivo. Oral Oncol 34:211–218.[Medline]

Cayrol C, Ducommun B (1998). Interaction with cyclin-dependent kinases and PCNA modulates proteasome-dependent degradation of p21. Oncogene 17:2437–2444.[Medline]

Chang BD, Xuan Y, Broude EV, Zhu H, Schott B, Fang J, et al. (1999). Role of p53 and p21waf1/cip1 in senescence-like terminal proliferation arrest induced in human tumor cells by chemotherapeutic drugs. Oncogene 18:4808–4818.[Medline]

Chang BD, Broude EV, Fang J, Kalinichenko TV, Abdryashitov R, Poole JC, et al. (2000a). p21Waf1/Cip1/Sdi1-induced growth arrest is associated with depletion of mitosis-control proteins and leads to abnormal mitosis and endoreduplication in recovering cells. Oncogene 19:2165–2170.[Medline]

Chang BD, Watanabe K, Broude EV, Fang J, Poole JC, Kalinichenko TV, et al. (2000b). Effects of p21Waf1/Cip1/Sdi1 on cellular gene expression: implications for carcinogenesis, senescence, and age-related diseases. Proc Natl Acad Sci USA 97:4291–4296.[Abstract/Free Full Text]

Chen J, Jackson PK, Kirschner MW, Dutta A (1995). Separate domains of p21 involved in the inhibition of Cdk kinase and PCNA. Nature 374:386–388.[Medline]

Chen J, Willingham T, Shuford M, Bruce D, Rushing E, Smith Y, et al. (1996). Effects of ectopic overexpression of p21(WAF1/CIP1) on aneuploidy and the malignant phenotype of human brain tumor cells. Oncogene 13:1395–1403.[Medline]

Cheng C, Tennenbaum T, Dempsey PJ, Coffey RJ, Yuspa SH, Dlugosz AA (1993). Epidermal growth factor receptor ligands regulate keratin 8 expression in keratinocytes, and transforming growth factor alpha mediates the induction of keratin 8 by the v-rasHa oncogene. Cell Growth Differ 4:317–327.[Abstract]

Cheng M, Olivier P, Diehl JA, Fero M, Roussel MF, Roberts JM, et al. (1999). The p21(Cip1) and p27(Kip1) CDK ’inhibitors’ are essential activators of cyclin D-dependent kinases in murine fibroblasts. EMBO J 18:1571–1583.[Medline]

Chuang LS, Ian HI, Koh TW, Ng HH, Xu G, Li BF (1997). Human DNA-(cytosine-5) methyltransferase-PCNA complex as a target for p21^WAF1. Science 277:1996–2000.[Abstract/Free Full Text]

Corroyer S, Nabeyrat E, Clement A (1997). Involvement of the cell cycle inhibitor CIP1/WAF1 in lung alveolar epithelial cell growth arrest induced by glucocorticoids. Endocrinology 138:3677–3685.[Abstract/Free Full Text]

Courtois SJ, Woodworth CD, Degreef H, Garmyn M (1997). Early ultraviolet B-induced G1 arrest and suppression of the malignant phenotype by wild-type p53 in human squamous cell carcinoma cells. Exp Cell Res 233:135–144.[Medline]

Cox LS (1997). Multiple pathways control cell growth and transformation: overlapping and independent activities of p53 and p21Cip1/WAF1/Sdi1. J Pathol 183:134–140.[Medline]

Datto MB, Li Y, Panus JF, Howe DJ, Xiong Y, Wang XF (1995a). Transforming growth factor beta induces the cyclin-dependent kinase inhibitor p21 through a p53-independent mechanism. Proc Natl Acad Sci USA 92:5545–5549.[Abstract/Free Full Text]

Datto MB, Yu Y, Wang XF (1995b). Functional analysis of the transforming growth factor beta responsive elements in the WAF1/Cip1/p21 promoter. J Biol Chem 270:28623–28628.[Abstract/Free Full Text]

Decesse JT, Medjkane S, Datto MB, Cremisi CE (2001). RB regulates transcription of the p21/WAF1/CIP1 gene. Oncogene 20:962–971.[Medline]

Delavaine L, La Thangue NB (1999). Control of E2F activity by p21Waf1/Cip1. Oncogene 18:5381–5392.[Medline]

Deng C, Zhang P, Harper JW, Elledge SJ, Leder P (1995). Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82:675–684.[Medline]

Di Cunto F, Topley G, Calautti E, Hsiao J, Ong L, Seth PK, et al. (1998). Inhibitory function of p21Cip1/^WAF1 in differentiation of primary mouse keratinocytes independent of cell cycle control. Science 280:1069–1072.[Abstract/Free Full Text]

Dickson MA, Hahn WC, Ino Y, Ronfard V, Wu JY, Weinberg RA, et al. (2000). 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 20:1436–1447.[Abstract/Free Full Text]

Dlugosz AA, Glick AB, Tennenbaum T, Weinberg WC, Yuspa SH (1995). Isolation and utilization of epidermal keratinocytes for oncogene research. Meth Enzymol 254:3–20.[Medline]

Eastham JA, Hall SJ, Sehgal I, Wang J, Timme TL, Yang G, et al. (1995). In vivo gene therapy with p53 or p21 adenovirus for prostate cancer. Cancer Res 55:5151–5155.[Abstract/Free Full Text]

El Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, et al. (1993). WAF1, a potential mediator of p53 tumor suppression. Cell 75:817–825.[Medline]

El Deiry WS, Harper JW, O’Connor PM, Velculescu VE, Canman CE, Jackman J, et al. (1994). WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis. Cancer Res 54:1169–1174.[Abstract/Free Full Text]

El Deiry WS, Tokino T, Waldman T, Oliner JD, Velculescu VE, Burrell M, et al. (1995). Topological control of p21^WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res 55:2910–2919.[Abstract/Free Full Text]

Enders GH, Koh J, Missero C, Rustgi AK, Harlow E (1996). p16 inhibition of transformed and primary squamous epithelial cells. Oncogene 12:1239–1245.[Medline]

Erber R, Klein W, Andl T, Enders C, Born AI, Conradt C, et al. (1997). Aberrant p21(CIP1/WAF1) protein accumulation in head-and-neck cancer. Int J Cancer 74:383–389.[Medline]

Fan Z, Lu Y, Wu X, DeBlasio A, Koff A, Mendelsohn J (1995). Prolonged induction of p21Cip1/WAF1/CDK2/PCNA complex by epidermal growth factor receptor activation mediates ligand-induced A431 cell growth inhibition. J Cell Biol 131:235–242.[Abstract/Free Full Text]

Franklin DS, Godfrey VL, O’Brien DA, Deng C, Xiong Y (2000). Functional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol Cell Biol 20:6147–6158.[Abstract/Free Full Text]

Fuchs E, Byrne C (1994). The epidermis: rising to the surface. Curr Opin Genet Dev 4:725–736.[Medline]

Gartel AL, Tyner AL (1999). Transcriptional regulation of the p21((WAF1/CIP1)) gene. Exp Cell Res 246:280–289.[Medline]

Gartel AL, Ye X, Goufman E, Shianov P, Hay N, Najmabadi F, et al. (2001). Myc represses the p21(WAF1/CIP1) promoter and interacts with Sp1/Sp3. Proc Natl Acad Sci USA 98:4510–4515.[Abstract/Free Full Text]

Goubin F, Ducommun B (1995). Identification of binding domains on the p21Cip1 cyclin-dependent kinase inhibitor. Oncogene 10:2281–2287.[Medline]

Green H (1977). Terminal differentiation of cultured human epidermal cells. Cell 11:405–416.[Medline]

Greenhalgh DA, Wang XJ, Donehower LA, Roop DR (1996a). Paradoxical tumor inhibitory effect of p53 loss in transgenic mice expressing epidermal-targeted v-rasHa, v-fos, or human transforming growth factor alpha. Cancer Res 56:4413–4423.[Abstract/Free Full Text]

Greenhalgh DA, Wang XJ, Roop DR (1996b). Multistage epidermal carcinogenesis in transgenic mice: cooperativity and paradox. J Investig Dermatol Symp Proc 1:162–176.[Medline]

Gu Y, Turck CW, Morgan DO (1993). Inhibition of CDK2 activity in vivo by an associated 20K regulatory subunit. Nature 366:707–710.[Medline]

Haapajarvi T, Kivinen L, Heiskanen A, des BC, Datto MB, Wang XF, et al. (1999). UV radiation is a transcriptional inducer of p21(Cip1/Waf1) cyclin-kinase inhibitor in a p53-independent manner. Exp Cell Res 248:272–279.[Medline]

Hainaut P, Hernandez T, Robinson A, Rodriguez-Tome P, Flores T, Hollstein M, et al. (1998). IARC database of p53 gene mutations in human tumors and cell lines: updated compilation, revised formats and new visualisation tools. Nucleic Acids Res 26:205–213.[Abstract/Free Full Text]

Halevy O, Novitch BG, Spicer DB, Skapek SX, Rhee J, Hannon GJ, et al. (1995). Correlation of terminal cell cycle arrest of skeletal muscle with induction of p21 by MyoD. Science 267:1018–1021.[Abstract/Free Full Text]

Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ (1993). The p21