Key words: cell cycle; checkpoints; senescence; apoptosis; tumor-stromal interactions; mitosis; cyclin-dependent kinase inhibitors; p21; oncogene; chemoprevention; DNA damage; CDK; cyclin; aging

ABSTRACT
p21^{Waf1/Cip1/Sdi1} is best known as a broad-specificity inhibitor of cyclin/cyclin-dependent kinase complexes, but p21 also interacts with many other regulators of transcription or signal transduction. p21 induction, which is mediated by p53 and by p53-independent mechanisms, is essential for the onset of cell cycle arrest in damage response and cell senescence. The effects of p21 knockout in mice and its expression patterns in human cancer are consistent with a role for p21 as both a tumor suppressor and an oncogene. Several functions of p21 are likely to promote carcinogenesis and tumor progression. These include endoreduplication and abnormal mitosis that develop in tumor cells after release from p21-induced growth arrest, the ability of p21 to inhibit apoptosis through several different mechanisms, and its ability to stimulate transcription of secreted factors with mitogenic and anti-apoptotic activities. The latter effects of p21 show close resemblance to paracrine activities of senescent cells and to tumor-promoting functions of stromal fibroblasts. Therapeutic strategies targeting the oncogenic consequences of p21 expression may provide a new approach to chemoprevention and treatment of cancer.

p21 and its molecular interactions
p21^{Waf1/Cip1/Sdi1} is the first identified inhibitor of cyclin/cyclin-dependent kinase (CDK) complexes, which regulate transitions between different phases of the cell cycle. p21 was independently isolated as a CDK-binding protein [1] and as a growth-inhibitory gene which is upregulated by wild-type p53 [2] or overexpressed in senescent fibroblasts [3]. Although many other CDK inhibitors have since been discovered [4], p21 appears to be the only inhibitor capable of interacting with essentially all of the CDK complexes. The effects of p21 on CDKs, however, are not limited to simple binding and inhibition. For example, p21 binding, depending on its stoichiometry, may not only inhibit but also stimulate CDK4/6 complexes [5]. On the other hand, p21 affinity towards CDC2/CDK1 (the primary regulator of cellular entry into mitosis) is low, suggesting that p21 may not bind CDC2 under physiological conditions [6]. Nevertheless, p21 expression efficiently inhibits CDC2 in vivo; this effect appears to be mediated at least in part through the interference with the activating phosphorylation of CDC2 at Thr¹⁶¹ [7].

p21 targets, however, are not limited to CDKs. In particular, p21 has long been known to bind the replication/repair factor PCNA. PCNA binding by p21 interferes with DNA replication but not with the effects of PCNA on DNA repair; p21-PCNA binding also affects PCNA interactions with several nuclear regulatory proteins [8]. In recent years, p21 was found to bind many other regulatory proteins, such as SAPK, ASK1 (MEKK5) and CK2 kinases, calmodulin and GADD45, procaspase 3, and oncogenic protein SET that inhibits protein phosphatase PP2A [8]. Some of p21 interactions have pronounced effects on transcription (reviewed in [9]). In particular, p21 inhibits E2F factors indirectly (via Rb dephosphorylation that results from CDK inhibition) and through direct binding [10]; p21 also binds and inhibits Myc [11] and STAT3 [12]. While the above interactions inhibit transcription regulated by the corresponding factors, p21 has also been shown to stimulate NFkappaB-mediated transcription. This effect is mediated by the activation of transcriptional cofactors/histone deacetylases p300 and CBP that augment not only NFkappaB but also many other inducible transcription factors [9;13]. This transcription-activating effect of p21 had been initially attributed to the inhibition of CDK2 that interacts with p300 and CBP. A more recent study [14] demonstrated, however, that the stimulation of p300 by p21 is mediated not by the binding of CDK2 but through a separate repressor domain of p300 termed CRD1. Furthermore, Coqueret and Gascan [12] showed that p21 can bind CBP in vitro, although it is unknown if such direct binding is responsible for the in vivo effect of p21 on p300 and CBP. The effects of p21 on transcription factors and cofactors, as well as its binding to CDK complexes and mediators of signal transduction are likely to have pleiotropic effects on cellular gene expression. Indeed, p21 expression from an inducible promoter results in the inhibition or induction of specific groups of genes with related functions [15], which will be described later in this review.

p21 induction in the programs of damage response and cell senescence
p21 expression is increased by different cell types in response to a variety of stressful stimuli, including DNA-damaging drugs or ionizing radiation, agents affecting DNA replication (mimosine) or mitosis (taxol), TGF-beta, differentiating agents or oncogenic Ras. The responses to some of these factors (e.g. DNA damage or mutant RAS) are mediated primarily through transcriptional activation of the p21 gene by p53. Other p21 inducers, such as TGF-beta or mimosine, induce p21 in p53-independent manner [16;17]. p53-independent p21 induction is regulated in part at the level of transcription, through various cis-regulatory sites in the p21 promoter, and in part at the level of mRNA and protein stability (reviewed in [18]). p21 induction appears transient rather than sustained in all physiological situations, including not only damage-induced temporary growth arrest (quiescence) but also permanent arrest associated with senescence (see below) or with terminal differentiation of postmitotic cells [19]. This transient mode of expression suggests that p21 is involved at the early stages but not in the long-term maintenance of cell cycle arrest.

A biological phenomenon that involves p21 and is of special importance to the present discussion is cell senescence, the physiological process of terminal growth arrest, accompanied by specific changes in cell shape, adhesion and gene expression [20;21]. The “classical” form of senescence is replicative senescence, a slow process which is mediated primarily by telomere shortening [21]. Another form, accelerated senescence, occurs in response to DNA damage [22;23] or introduction of oncogenic RAS [24]. Accelerated senescence, which does not involve telomere shortening [25;26] and is not prevented by telomerase expression [27], has been recognized as an important anti-carcinogenic response, which permanently stops the growth of cells that have experienced potentially carcinogenic damage [21;28]. Not only normal cells but also different types of tumor cell lines develop senescence-like terminal growth arrest when treated with a variety of anticancer agents [29]. Tumor cells can also be induced to undergo senescence through the overexpression of several tumor suppressors [30-32] including p21 [33;34] or through inhibition of oncogenes [35].

Terminal growth arrest of normal cells undergoing replicative or accelerated senescence involves consecutive activation of p53, p21 and p16Ink4A. DNA damage or telomere shortening beyond a critical length activates p53, and p53 in its turn induces the transcription of p21. Strong but transient p21 induction coincides with the onset of growth arrest in senescent cells. p21 expression eventually decreases to the levels that are just a little higher than in non-senescent cells, at which time another CDK inhibitor, p16 becomes constitutively upregulated, suggesting that p16 expression maintains growth arrest of senescent cells upon the decay of p21 [23;36].

Role of p21 in cell cycle arrest and maintenance of genomic stability
Ectopic expression of p21 arrests the cell cycle either in G1 or in both G1 and G2, depending on the cell type [37]. While p21-induced G1 arrest can be readily explained by direct inhibition of G1-specific cyclin/CDK complexes, the mechanism of p21-induced G2 arrest is less obvious. As stated above, inhibition of the CDC2/cyclin B1 complex by p21 binding may require higher than physiological doses of p21, but p21 was found to inhibit the activating phosphorylation of CDC2 at Thr¹⁶¹ [7]. p21 binding to a Cyclin-B interactive protein CARB was also proposed to play a role in G2 arrest [38]. Another potential mechanism of the effect on p21 on G2 has been indicated by our finding that p21 induction rapidly inhibits the transcription of Topoisomerase II, Cyclin B1, CDC2 and many other genes required for the onset and execution of mitosis [15]. The ability of p21 to inhibit the expression of G2/M-specific genes agrees with the observations that p21 deficiency interferes with the inhibition of a large group of G2/M genes in response to ionizing radiation [39] or doxorubicin treatment [40]. The requirement for p21 in damage-induced transient G1 and G2 arrest has been demonstrated by the analysis of p21-/- cells. p21-nullizygous human [41] and mouse [42;43] cells fail to arrest in G1 in response to p53-activating treatment. p21-deficient human cells also fail to maintain G2 arrest after DNA damage; such cells instead enter mitosis, which is aberrant and culminates in the formation of multiple micronuclei and cell death (mitotic catastrophe) [44;45]. In addition, p21 knockout leads to the appearance of polyploid cells after DNA damage [46] or mitotic spindle disruption [47].

The role of p21 in permanent growth arrest associated with replicative senescence was demonstrated by the work of Brown et al. [48], who found that p21 knockout in normal human fibroblasts extends their lifespan in culture. In contrast to these results in human cells, p21 knockout was reported to have no effect on replicative senescence in mouse fibroblasts [49]. The latter conclusion, however, was disputed in a more recent study by Dulic et al. [50]. These authors found that p21-/- mouse fibroblasts, as well as human fibroblasts carrying the papillomavirus E6 gene that inhibits p21 expression, have a limited lifespan in culture, but proliferation of these cells is limited by “crisis-like events” rather than by terminal cell cycle arrest [50]. p21 is also involved in accelerated senescence of tumor cells, since its inhibition or knockout decrease by several-fold the induction of senescence by different anticancer agents [29]. The evidence for the role of p21 as a positive regulator of cell cycle arrest in damage response and senescence is therefore very convincing.

Effects of p21 knockout on carcinogenesis in the mouse
The role of p21 in carcinogenesis has been investigated in mouse models through gene knockout studies. In light of the established role of p21 in cell cycle arrest, it was a big surprise that p21-/- mice were initially reported to remain tumor-free until at least 7 months of age [42]. Recently, however, Martin-Caballero et al. [51] showed that p21-/- mice develop spontaneous tumors at an average age of 16 months, whereas control animals remain tumor-free for over 2 years. The tumors of p21-deficient mice were mostly of hematopoietic and endothelial origin, with about 10% epithelial tumors [51]. p21 knockout was also found to increase the rate of tumorigenesis in RB-haploinsufficent [52], p18^Ink4c-deficient [53], APC-haploinsufficient [54] and v-Ha-ras transgenic backgrounds [55]. These findings have provided formal evidence for classifying p21 as a tumor suppressor. In contrast to the above results, p21 deficiency did not increase the frequency of tumorigenesis in mice that are also deficient for the Werner syndrome gene [56]. Most strikingly, p21 knockout actually reduced the development of radiation-induced tumors in ATM-deficient [57] and wild type mice [51]. In a related finding, p21 levels were significantly increased in 9 of 12 mouse fibroblast lines that were neoplastically transformed by ionizing radiation [58]. Hence, mouse p21 in specific situations appears to display the properties of an oncogene rather than a tumor suppressor.

p21 in human cancer
Several p21 mutations have been reported in human cancers [59-63], and some of these mutations were shown to abrogate p21 activity as a CDK inhibitor [64]. Nevertheless, the overall conclusion from various large-scale studies that have searched for p21 mutations in human cancer has been that p21 mutations are exceedingly rare [65-67]. This paucity of mutations sets p21 at a striking contrast to other tumor suppressors, such as p53 or p16, which act in the same pathways of checkpoint control or senescence but are very frequently mutated in cancers. Since p21 expression is one of the most prominent markers for the functional activity of p53, many studies have analyzed p21 expression in different types of human cancer, in the expectation that a loss of p21 would show a correlation with carcinogenesis and negative prognosis. The outcome of such studies, however, has been anything but straightforward. In some reports, the lack of p21 was indeed shown to correlate with tumor progression and negative prognosis, as reported, for example, for small-cell lung [68], colorectal [69], cervical [70], and head and neck cancers [71]. Such correlations may be the strongest when the absence of p21 is seen along with the expression of p53, as an indication for the loss of p53 function [72-74]. In many cases, however, p21 expression was concluded to have no prognostic value [75-79]. Strikingly, several reports have indicated that increased p21 expression was associated with tumor progression and negative prognosis in prostate [80;81], ovarian [82], cervical [83;84], breast [85] and esophageal squamous cell carcinomas [86], as well as in brain tumors [87]. To complicate things further, different histological subtypes of soft tissue sarcomas were found in the same study to have either increased or decreased p21 expression relative to normal cells [88], and p21 expression was concluded to be a negative prognostic marker in superficial bladder cancers but a positive marker in invasive bladder cancers [89]. Of special interest, a study on extrahepatic bile duct carcinomas has concluded that both low and high p21 levels are significant markers of negative prognosis, whereas moderate levels of p21 provide a more favorable indication [90]. The results of these studies are consistent with the dual role of p21 as both a tumor suppressor and an oncogene.

Destabilizing effects of p21 induction on mitosis and genome stability
How could p21, one of the principal checkpoint control proteins, contribute to carcinogenesis and tumor progression? As mentioned above, lower levels of p21 promote the assembly of active Cyclin D-CDK4/6 complexes, suggesting that p21 may potentially promote cell growth [5]. Nevertheless, all the studies on p21 knockout or overexpression support the role of p21 as an inducer of cell cycle arrest but not as a positive regulator of cell cycle progression. A possible exception is the work of Weiss et al. [91], who reported that p21 inhibition with an antisense oligonucleotide in vascular smooth muscle cells inhibits both cyclin D complex assembly and growth factor-stimulated proliferation. The latter study, however, has not examined whether the effect of anti-p21 oligonucleotides on cell proliferation could be due to changes in apoptosis (a well-documented function of p21, see below) rather than cell cycle progression. On the other hand, analysis of ectopic p21 expression has revealed several other p21 activities that are likely to have an oncogenic effect.

As mentioned above, p21 deficiency promotes an increase in DNA ploidy following DNA damage. Paradoxically, endoreduplication, an unscheduled round of DNA replication leading to the doubling of DNA content, was also found to occur as a consequence of p21 overexpression [37;92]. Comparison of the effects of p21 in different tumor cell lines suggested that endoreduplication in p21-induced cells may require RB deficiency [37]. Nevertheless, endoreduplication was also induced in RB-positive HT1080 fibrosarcoma cells when these cells re-entered the cycle after transient induction of p21 [45].

HT1080 cells recovering after prolonged p21 induction develop not only endoreduplication but also numerous mitotic abnormalities, indicative of abnormal centrosome amplification and chromosome segregation [45]. p21-induced abnormal mitosis is attributable at least in part to the ability of p21 to inhibit the expression of genes involved in different stages of mitosis [15]. This inhibition occurs at the level of transcription, and it is mediated in part through negative regulatory elements CDE and CHR, which are found in the promoters of many of these genes [93]. Because of this repression, the intracellular pool of proteins required for both the execution and the quality control of mitosis is depleted upon prolonged p21 expression. These proteins are synthesized de novo after release from p21, but their re-synthesis appears to be poorly synchronized in tumor cells. Such cells therefore enter mitosis with insufficient amounts of mitosis-control proteins (such as spindle checkpoint component Mad2), resulting in abnormal chromosome segregation [45]. p21-induced abnormal mitosis may also be related to transcriptional inhibition of centrosome-controlling proteins, such as aurora kinases AIM1 and AIK1/AIM2 [15] or to p21-mediated inhibition of centrosome duplication at the G1-S boundary [94]. Since transient p21 induction promotes endoreduplication and abnormal chromosome segregation in tumor cells, such induction may serve as a genome-destabilizing factor contributing to tumor progression. There is as yet no direct evidence whether p21 induction may also have a genome-destabilizing effect in normal cells, which may potentially promote neoplastic transformation. Interestingly, normal fibroblast cultures derived from old people or from patients with progeria (a premature aging syndrome) were found to differ from similar cultures of young individuals by nuclear abnormalities and decreased expression of the same mitotic genes that are inhibited by p21 [95]. These age-associated changes in mitotic gene expression may potentially reflect the lifetime history of p21 induction and the resulting transcriptional inhibition of mitotic genes. Such inhibition could have become stabilized, for example, through changes in DNA methylation. We hypothesize that preventing cells that have experienced prolonged p21 induction from potentially abnormal reentry into the cycle may represent one of the monitoring functions of the program of cell senescence. This mechanism, which is maintained to a large extent by p16, would be abolished by p16-inactivating mutations, with the corresponding carcinogenic consequences.

p21 as an inhibitor of apoptosis.
Probably the best-documented oncogenic function of p21 is its ability to protect cells from apoptosis [8]. Importantly, p21 is cleaved by caspase 3 at the onset of apoptosis, losing its apoptosis-suppressing activity [96]. The anti-apoptotic function of p21 can explain why p21 knockout decreases the rate of radiation-induced tumorigenesis in p21 knockout mice [51] and increases the sensitivity to radiation-induced apoptosis in ATM-/- mouse cells [57].

Many studies on the role of p21 in apoptosis have utilized apoptosis-inducing factors that require cell cycle progression for their effect, such as DNA-damaging agents or anti-microtubular drugs. In such cases, the anti-apoptotic effect of p21 can be interpreted as a consequence of its growth-inhibitory function. For example, p21 knockout in human HCT116 colon carcinoma cells results in a drastic increase in apoptosis upon treatment with doxorubicin [46] or radiation [44]. This effect, however, can be attributed primarily to the debilitation of damage-induced G2 checkpoint arrest in p21-/- cells [44]. As a result, doxorubicin-treated or irradiated cells enter mitosis prior to repairing DNA damage and undergo mitotic catastrophe, a process which is not only lethal per se but also activates cellular self-destruction through apoptosis [97]. p21, however, also protects cells from apoptosis induced by a variety of other factors, at least some of which may not require cell cycle progression. Such factors include p53 overexpression [98;99], low density culture [100] or growth factor deprivation [101], butyrate [102], CD95 ligand [103] and anti-IgM antibodies (in B-lymphoma cells) [104]. p21 has also been implicated in the prevention of apoptosis during monocyte differentiation [105] and skeletal myogenesis [106]. These observations indicate that p21 expression has an anti-apoptotic effect, which is distinct from its growth-inhibitory activity.

The anti-apoptotic functions of p21 may be attributed to several of its molecular interactions. These include the binding of procaspase 3 and inhibition of its conversion to mature caspase 3, interactions with caspases 8 and 10, inhibition of apoptosis-regulating kinases, such as SAPKs and ASK1, and inhibition of apoptosis-stimulating transcription factors, such as Myc and E2F (see [8] for a detailed discussion). In addition to these intracellular anti-apoptotic functions of p21, the effects of p21 on transcription result in the production of secreted anti-apoptotic factors, as described below.

Paracrine effects of p21 induction: parallels with cell senescence and tumor-promoting functions of stromal fibroblasts.
Additional biological functions of p21 have been revealed by cDNA microarray analysis of the effects of p21 on gene expression in HT1080 fibrosarcoma cells [15]. p21 expression, which induces a senescent phenotype in this cell line, was found to upregulate a large group of genes encoding extracellular matrix components and other secreted proteins, many of which have been associated with cell senescence and organism aging. p21-inducible genes include a group of anti-apoptotic and mitogenic factors, such as connective tissue growth factor (CTGF), epithelin/granulin, galectin-3, activin A, prosaposin and several others. In agreement with the overexpression of such factors, conditioned media from p21-induced but not from uninduced cells protected an apoptosis-prone transformed cell line from serum starvation-induced apoptosis. The same conditioned media also stimulated cell growth and thymidine incorporation in a slow-growing human fibrosarcoma line [15]. Induction of the same genes and anti-apoptotic activity of conditioned media were also observed in normal human fibroblasts infected with a p21-expressing adenoviral vector (M.E. Swift, B.D. Chang, E.V. Broude, and I.B.R., unpublished). In light of the paracrine mitogenic effects of p21 expression, it is interesting to note that transient p21 expression in postmitotic tissues has been localized to narrow zones adjacent to proliferative compartments [19;107]. Upregulation of p21-inducible genes can be reproduced through the use of promoter-reporter constructs, indicating that this effect of p21 is exerted at least in part at the level of transcription (J.C. Poole, B.D. Chang and I.B.R., unpublished). Activation of such genes is observed not only after ectopic expression of p21 but also as a result of drug-induced p21 induction. In particular, polyamine depletion in a human melanoma cell line was found to induce p21, followed by the development of the senescent phenotype and activation of 13 of 14 tested p21-inducible genes, including the genes for anti-apoptotic and mitogenic factors [108]. Upregulation of several tumor-promoting paracrine factors (TGFa, CYR61, prosaposin, and several proteases) was also associated with p21 induction in HCT116 colon carcinoma cells that underwent accelerated senescence after treatment with doxorubicin. Induction of most (albeit not all) genes in this group was delayed or diminished by p21 knockout [40]. Doxorubicin-treated HCT116 cells, however, induce only a minority of the genes that are activated by p21 in other cell types. HCT116 cells carry a mutation of the p21-stimulated transcriptional coactivator p300 [109], suggesting that this apparently dominant p300 mutation may have diminished the transcription-activating function of p21 [40]. The observed paracrine effects of p21 expression parallel some of the known features of cell senescence. As described above, this process involves consecutive activation of p21 and p16; our preliminary studies indicate that the latter CDK inhibitor shares with p21 most of its transcription-activating effects (B.D. Chang, M. Shen, and I.B.R., unpublished). Some of the genes that are highly expressed in senescent cells have long-range, pleiotropic effects, including degradative enzymes, inflammatory cytokines and growth factors, which may contribute to age-dependent increase in carcinogenesis [20;21]. In agreement with the latter hypothesis, senescent but not presenescent fibroblasts strongly promote the growth of preneoplastic and malignant (but not normal) epithelial cells, as based on in vitro coculture and in vivo tumorigenesis assays [110]. This tumor-promoting activity has been observed with several types of normal human fibroblasts that were made to senesce by different means [110], including both replicative senescence and accelerated senescence, which was induced by hydrogen peroxide or by ectopic expression of oncogenic Ras (which leads to p21 induction [24]) or tumor suppressor p14^ARF, which activates p53 and consequently p21 [111]. The role of senescent cells in promoting carcinogenesis has also been suggested by a clinical study, where expression of a senescence marker in normal human hepatocytes was found to be strongly correlated with the presence of hepatocellular carcinoma in the surrounding liver [112]. The paracrine activities found in p21-induced or senescent fibroblasts have been previously associated with stromal elements of human tumors (see [113] for a recent review). Stromal fibroblasts (sometimes described morphologically as myofibroblasts) were shown to secrete paracrine mitogenic and anti-apoptotic factors and proteases that promote growth and survival of carcinoma cells [114-117]. These factors have been identified in fibroblast-conditioned media [118] and in co-culture studies. In particular, Olumi et al. [117] showed that co-culture of prostate carcinoma cells with normal prostate fibroblasts strongly decreases carcinoma cell death and promotes xenograft tumor formation. The paracrine effects of fibroblasts not only stimulate tumor growth but also promote neoplastic transformation of initiated cells [119]. This anti-apoptotic and tumor-promoting activity may be related to the production of different secreted growth factors, proteases and angiogenic factors by stromal fibroblasts [113;120].

Conversion of resting fibroblasts into tumor-promoting myofibroblasts was shown to be stimulated by TGFbeta [121;122]. In agreement with these findings, TGFbeta expression in breast cancer is localized preferentially to the advancing edges of primary tumors and to lymph node metastases, suggesting that TGFbeta plays a role in tumor interactions with neighboring stromal cells [123]. The tumor-promoting effect of TGFbeta appeared paradoxical, since TGFbeta inhibits the proliferation of most cell types by inducing the expression of several CDK inhibitors [124]. These inhibitors include p21, which is induced by TGFbeta in p53-independent manner [16]. TGFbeta also induces senescence in fibroblasts [125]. The tumor-promoting activity of mouse mammary stroma was also found to be activated by ionizing radiation [126]. Like TGFbeta, ionizing radiation induces both accelerated senescence [22;29] and p21 expression in different cell types, including human mammary fibroblasts [127]. It appears therefore that paracrine activities associated with cellular senescence are similar or identical to the tumor-promoting functions of stromal fibroblasts, and that p21 induction (superceded by p16 induction in senescent cells) is a common element in both situations (Figure 1).

Effects of p21 on stress sensitivity and age-related diseases.
p21 induction activates not only tumor-promoting factors but also a number of genes associated with different aspects of aging [15]. One of these genes is p66^Shc, a mediator of oxidative stress. The knockout of p66 expands the lifespan of mice by about 30% and greatly increases in vivo survival of a toxin (paraquat) [128]. Remarkably, p21 knockout mice also show much improved survival of toxic treatment, in this case ionizing radiation [51]. Radiation survival of p21-/- mice has been attributed both to a decrease in radiation-induced cancer relative to the wild-type animals and to the ability of radiation to suppress glomerulonephritis, the primary cause of death in p21-/- mice [51]. The latter condition is associated with an autoimmune process resulting from abnormal T-cell proliferation in p21-/- mice [129]. In contrast to their susceptibility to autoimmune glomerulonephritis, p21 knockout mice were found to be resistant rather than sensitive to chronic renal failure produced by partial renal ablation [130] and to diabetic nephropathy induced by streptozotocin [131]. Remarkably, both p21 and p21-inducible proteins CTGF, fibronectin and plasminogen activator inhibitor 1 (PAI-1) have been implicated in an in vitro model of diabetic nephropathy [132]. These results suggest that p21 induction may mediate the sensitivity of the organism to several forms of stress, at least in part through the effects of p21 on gene expression.

Many p21-inducible genes have also been implicated in age-related diseases. One of the genes that are induced upon p21 expression [15] and in cells undergoing replicative [133] or accelerated [40] senescence encodes the precursor of beta amyloid peptide, an essential culprit of Alzheimer’s disease. p21 expression also upregulates serum amyloid A, a stress response protein implicated in amyloidosis, atherosclerosis and arthritis, and tissue transglutaminase, which cross-links amyloid peptides leading to plaque formation in both Alzheimer’s disease and amyloidosis. p21 also induces CTGF and galectin-3 involved in atherosclerosis and several genes that were found to be upregulated in arthritis. It appears therefore that p21 induction, an event that repeatedly occurs in different cell types over the lifetime, may contribute to the development not only of cancer but also of other age-related diseases [15].

p21 and p21-mediated pathways as potential targets for therapeutic intervention
As described above, the expression of p21, which is induced in response to many different forms of stress, has pleiotropic effects that are not directly related to the role of p21 in cell cycle arrest. These effects, which include genome destabilization, prevention of apoptosis, overexpression of secreted anti-apoptotic and mitogenic factors, and activation of genes involved in age-related diseases, may contribute to tumor progression, carcinogenesis, and some of the detrimental side effects of p21-inducing anticancer agents. These observations indicate the desirability of developing therapeutic approaches to counteract these pathogenic effects of p21. Can p21 itself be used as a target for such therapeutic strategies? Several considerations warrant caution in the use of inhibitors that target p21 directly. (a) The evidence for tumor-suppressing functions of p21 in the control of genome stability and in the anti-carcinogenic program of senescence is quite strong. It is conceivable therefore that general inhibition of p21 function or expression (using, for example, p21-targeting oligonucleotides) could promote tumor progression or de novo carcinogenesis. (b) There is as yet no evidence that the growth-arrest functions of p21 and its tumor-promoting effects on apoptosis or gene expression involve different domains in the p21 molecule, and it may be impossible therefore to develop p21-targeting agents that will selectively inhibit its oncogenic functions but not its tumor-suppressing activities. (c) Preliminary evidence from our laboratory suggests that at least some of the effects of p21 are likely to be shared by other CDK inhibitors, including those that are also induced in situations relevant to cancer (e.g. p16 is highly expressed in senescent cells, whereas p27^Kip1 and p15^Ink4B are stimulated by TGFbeta). p21-specific inhibitors therefore may have a limited effect in the situations that involve the induction of other CDK inhibitors. Despite the above misgivings, it is still possible that p21-targeting agents, if applied for a limited time in a well-defined therapeutic context, may have a beneficial impact that will outweigh their potential side effects. This optimism is based primarily on the promising results obtained with pifithrin alpha, a chemical agent that inhibits transcription-activating functions of p53 (including p21 induction). Although the potential undesirable effects of p53 inhibition are even more numerous than those of p21 inhibition, pifithrin alpha was found to protect mice from the toxicity of ionizing radiation (due to the interference with p53-dependent apoptosis), without promoting tumor formation [134].

Since most of the oncogenic functions of p21 are related to its effects on gene expression, it should also be possible to target the regulatory pathways that mediate these effects of p21 instead of targeting p21 itself. Agents aimed at these pathways should have the advantages of not interfering with the growth arrest functions of p21 and of being effective in the situations where such pathways are activated by non-p21 CDK inhibitors or by some other factors. Such agents may potentially be used over a long period of time, as not only therapeutic but also chemopreventive agents.

In principle, agents targeting the transcriptional effects of p21 can be identified by mechanism-independent screening using p21-responsive promoter-reporter constructs (such an approach has been used to develop pifithrin alpha; [134]). Alternatively, one could use rational drug design to target the specific molecular interactions, which mediate the undesirable effects of p21. In particular, p21 stimulates p300/CBP histone acetylases and factors that require their activity, including NFkappaB [9;13;14]. Remarkably, NFkappaB inhibition is a documented effect of several widely used non-steroidal anti-inflammatory drugs such as aspirin, sodium salicylate or sulindac [135;136], which have shown considerable promise in cancer chemoprevention. While the chemopreventing efficacy of these drugs has been attributed to their effect on cyclooxygenase 2 [137], we are not aware of any evidence that would rule out NFkappaB inhibition as another mechanism of action for these chemopreventive agents. It is presently unknown to what extent the effects of p21 on p300/CBP and NFkappaB can account for the induction of different disease-associated genes. Furthermore, p21 induction does not affect all of p300/CBP-responsive promoters [14], suggesting the existence of other factors that determine the specificity of this effect of p21. Further research is likely to elucidate the molecular network of p21-affected regulatory pathways and reveal the most suitable targets for therapeutic intervention with the oncogenic effects of p21.

References
[1] J.W.Harper, G.R.Adami, N.Wei, K.Keyomarsi, S.J.Elledge, The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin- dependent kinases. Cell 75 (1993) 805-816.
[2] W.S.El Deiry, T.Tokino, V.E.Velculescu, D.B.Levy, R.Parsons, J.M.Trent, D.Lin, W.E.Mercer, K.W.Kinzler, B.Vogelstein, WAF1, a potential mediator of p53 tumor suppression. Cell 75 (1993) 817-825.
[3] A.Noda, Y.Ning, S.F.Venable, O.M.Pereira-Smith, J.R.Smith, Cloning of senescent cell-derived inhibitors of DNA synthesis using an expression screen. Exp.Cell Res. 211 (1994) 90-98.
[4] C.J.Sherr and J.M.Roberts, CDK inhibitors: positive and negative regulators of G1-phase progression. Genes Dev. 13 (1999) 1501-1512.
[5] J.LaBaer, M.D.Garrett, L.F.Stevenson, J.M.Slingerland, C.Sandhu, H.S.Chou, A.Fattaey, E.Harlow, New functional activities for the p21 family of CDK inhibitors. Genes Dev 11 (1997) 847-862.
[6] J.W.Harper, S.J.Elledge, K.Keyomarsi, B.Dynlacht, L.H.Tsai, P.Zhang, S.Dobrowolski, C.Bai, L.Connell-Crowley, E.Swindell, Inhibition of cyclin-dependent kinases by p21. Mol.Biol.Cell 6 (1995) 387-400.
[7] V.A.Smits, R.Klompmaker, T.Vallenius, G.Rijksen, T.P.Makela, R.H.Medema, p21 inhibits thr¹⁶¹ phosphorylation of cdc2 to enforce the G2 DNA damage checkpoint. J.Biol.Chem. 275 (2000) 30638-30643.
[8] G.P.Dotto, p21(WAF1/Cip1): more than a break to the cell cycle? Biochim.Biophys.Acta 1471 (2000) M43-M56.
[9] N.D.Perkins, Not just a CDK inhibitor: regulation of transcription by p21^{Waf1/Cip1/Sdi1}. Cell Cycle in press (2002).
[10] L.Delavaine and N.B.La Thangue, Control of E2F activity by p21^Waf1/Cip1. Oncogene 18 (1999) 5381-5392.
[11] H.Kitaura, M.Shinshi, Y.Uchikoshi, T.Ono, S.M.Iguchi-Ariga, H.Ariga, Reciprocal regulation via protein-protein interaction between c-Myc and p21(cip1/waf1/sdi1) in DNA replication and transcription. J.Biol.Chem. 275 (2000) 10477-10483.
[12] O.Coqueret and H.Gascan, Functional interaction of STAT3 transcription factor with the cell cycle inhibitor p21^{WAF1/CIP1/SDI1}. J.Biol.Chem. 275 (2000) 18794-18800.
[13] N.D.Perkins, L.K.Felzien, J.C.Betts, K.Leung, D.H.Beach, G.J.Nabel, Regulation of NF-kappaB by cyclin-dependent kinases associated with the p300 coactivator. Science 275 (1997) 523-527.
[14] A.W.Snowden, L.A.Anderson, G.A.Webster, N.D.Perkins, A novel transcriptional repression domain mediates p21(WAF1/CIP1) induction of p300 transactivation. Mol.Cell Biol. 20 (2000) 2676-2686.
[15] B.D.Chang, K.Watanabe, E.V.Broude, J.Fang, J.C.Poole, T.V.Kalinichenko, I.B.Roninson, Effects of p21^{Waf1/Cip1/Sdi1} on cellular gene expression: implications for carcinogenesis, senescence, and age-related diseases. Proc.Natl.Acad.Sci.U.S.A 97 (2000) 4291-4296.
[16] M.B.Datto, Y.Li, J.F.Panus, D.J.Howe, Y.Xiong, X.F.Wang, Transforming growth factor beta induces the cyclin-dependent kinase inhibitor p21 through a p53-independent mechanism. Proc.Natl.Acad.Sci.U.S.A 92 (1995) 5545-5549.
[17] R.S.Alpan and A.B.Pardee, p21^{WAF1/CIP1/SDI1} is elevated through a p53-independent pathway by mimosine. Cell Growth Differ. 7 (1996) 893-901.
[18] A.L.Gartel and A.L.Tyner, Transcriptional regulation of the p21 (WAF1/CIP1)gene. Exp.Cell Res. 246 (1999) 280-289.
[19] A.L.Gartel, M.S.Serfas, M.Gartel, E.Goufman, G.S.Wu, W.S.El Deiry, A.L.Tyner, p21 (WAF1/CIP1) expression is induced in newly nondividing cells in diverse epithelia and during differentiation of the Caco-2 intestinal cell line. Exp.Cell Res. 227 (1996) 171-181.
[20] J.Campisi, The biology of replicative senescence. Eur.J.Cancer 33 (1997) 703-709.
[21] R.A.DePinho, The age of cancer. Nature 408 (2000) 248-254.
[22] A.Di Leonardo, S.P.Linke, K.Clarkin, G.M.Wahl, DNA damage triggers a prolonged p53-dependent G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev. 8 (1994) 2540-2551.
[23] S.J.Robles and G.R.Adami, Agents that cause DNA double strand breaks lead to p16INK4a enrichment and the premature senescence of normal fibroblasts. Oncogene 16 (1998) 1113-1123.
[24] M.Serrano, A.W.Lin, M.E.McCurrach, D.Beach, S.W.Lowe, Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 88 (1997) 593-602.
[25] E.Michishita, K.Nakabayashi, H.Ogino, T.Suzuki, M.Fujii, D.Ayusawa, DNA topoisomerase inhibitors induce reversible senescence in normal human fibroblasts. Biochem.Biophys.Res.Commun. 253 (1998) 667-671.
[26] K.Suzuki, I.Mori, Y.Nakayama, M.Miyakoda, S.Kodama, M.Watanabe, Radiation-induced senescence-like growth arrest requires TP53 function but not telomere shortening. Radiat.Res. 155 (2001) 248-253.
[27] C.P.Morales, S.E.Holt, M.Ouellette, K.J.Kaur, Y.Yan, K.S.Wilson, M.A.White, W.E.Wright, J.W.Shay, Absence of cancer-associated changes in human fibroblasts immortalized with telomerase. Nat.Genet. 21 (1999) 115-118.
[28] R.A.Weinberg, The cat and mouse games that genes, viruses, and cells play. Cell 88 (1997) 573-575.
[29] B.D.Chang, E.V.Broude, M.Dokmanovic, H.Zhu, A.Ruth, Y.Xuan, E.S.Kandel, E.Lausch, K.Christov, I.B.Roninson, A senescence-like phenotype distinguishes tumor cells that undergo terminal proliferation arrest after exposure to anticancer agents. Cancer Res. 59 (1999) 3761-3767.
[30] M.M.Sugrue, D.Y.Shin, S.W.Lee, S.A.Aaronson, Wild-type p53 triggers a rapid senescence program in human tumor cells lacking functional p53. Proc.Natl.Acad.Sci.U.S.A 94 (1997) 9648-9653.
[31] H.J.Xu, Y.Zhou, W.Ji, G.S.Perng, R.Kruzelock, C.T.Kong, R.C.Bast, G.B.Mills, J.Li, S.X.Hu, Reexpression of the retinoblastoma protein in tumor cells induces senescence and telomerase inhibition. Oncogene 15 (1997) 2589-2596.
[32] L.Uhrbom, M.Nister, B.Westermark, Induction of senescence in human malignant glioma cells by p16INK4A. Oncogene 15 (1997) 505-514.
[33] L.Fang, M.Igarashi, J.Leung, M.M.Sugrue, S.W.Lee, S.A.Aaronson, p21^{Waf1/Cip1/Sdi1} induces permanent growth arrest with markers of replicative senescence in human tumor cells lacking functional p53. Oncogene 18 (1999) 2789-2797.
[34] B.D.Chang, Y.Xuan, E.V.Broude, H.Zhu, B.Schott, J.Fang, I.B.Roninson, Role of p53 and p21^waf1/cip1 in senescence-like terminal proliferation arrest induced in human tumor cells by chemotherapeutic drugs. Oncogene 18 (1999) 4808-4818.
[35] E.C.Goodwin, E.Yang, C.J.Lee, H.W.Lee, D.DiMaio, E.S.Hwang, Rapid induction of senescence in human cervical carcinoma cells. Proc.Natl.Acad.Sci.U.S.A. 97 (2000) 10978-10983.
[36] D.A.Alcorta, Y.Xiong, D.Phelps, G.Hannon, D.Beach, J.C.Barrett, Involvement of the cyclin-dependent kinase inhibitor p16 (INK4a) in replicative senescence of normal human fibroblasts. Proc.Natl.Acad.Sci.U.S.A 93 (1996) 13742-13747.
[37] A.B.Niculescu, III, X.Chen, M.Smeets, L.Hengst, C.Prives, S.I.Reed, Effects of p21(Cip1/Waf1) at both the G1/S and the G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in preventing endoreduplication. Mol.Cell Biol. 18 (1998) 629-643.
[38] A.McShea, T.Samuel, J.T.Eppel, D.A.Galloway, J.O.Funk, Identification of CIP-1-associated regulator of cyclin B (CARB), a novel p21-binding protein acting in the G2 phase of the cell cycle. J.Biol.Chem. 275 (2000) 23181-23186.
[39] S.M.de Toledo, E.I.Azzam, P.Keng, S.Laffrenier, J.B.Little, Regulation by ionizing radiation of CDC2, cyclin A, cyclin B, thymidine kinase, topoisomerase IIalpha, and RAD51 expression in normal human diploid fibroblasts is dependent on p53/p21^Waf1. Cell Growth Differ. 9 (1998) 887-896.
[40] B.D.Chang, M.E.Swift, M.Shen, J.Fang, E.V.Broude, I.B.Roninson, Molecular determinants of terminal growth arrest induced in tumor cells by a chemotherapeutic drug. Proc. Natl. Acad. Sci. U.S.A. in press (2002).
[41] T.Waldman, K.W.Kinzler, B.Vogelstein, p21 is necessary for the p53-mediated G1 arrest in human cancer cells. Cancer Res. 55 (1995) 5187-5190.
[42] C.Deng, P.Zhang, J.W.Harper, S.J.Elledge, P.Leder, Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82 (1995) 675-684.
[43] J.Brugarolas, C.Chandrasekaran, J.I.Gordon, D.Beach, T.Jacks, G.J.Hannon, Radiation-induced cell cycle arrest compromised by p21 deficiency. Nature 377 (1995) 552-557.
[44] F.Bunz, A.Dutriaux, C.Lengauer, T.Waldman, S.Zhou, J.P.Brown, J.M.Sedivy, K.W.Kinzler, B.Vogelstein, Requirement for p53 and p21 to sustain G2 arrest after DNA damage. Science 282 (1998) 1497-1501.
[45] B.D.Chang, E.V.Broude, J.Fang, T.V.Kalinichenko, R.Abdryashitov, J.C.Poole, I.B.Roninson, p21^{Waf1/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 (2000) 2165-2170.
[46] T.Waldman, C.Lengauer, K.W.Kinzler, B.Vogelstein, Uncoupling of S phase and mitosis induced by anticancer agents in cells lacking p21. Nature 381 (1996) 713-716.
[47] Z.A.Stewart, S.D.Leach, J.A.Pietenpol, p21(Waf1/Cip1) inhibition of cyclin E/Cdk2 activity prevents endoreduplication after mitotic spindle disruption. Mol.Cell Biol. 19 (1999) 205-215.
[48] J.P.Brown, W.Wei, J.M.Sedivy, Bypass of senescence after disruption of p21CIP1/WAF1 gene in normal diploid human fibroblasts. Science 277 (1997) 831-834.
[49] C.Pantoja and M.Serrano, Murine fibroblasts lacking p21 undergo senescence and are resistant to transformation by oncogenic Ras. Oncogene 18 (1999) 4974-4982.
[50] V.Dulic, G.E.Beney, G.Frebourg, L.F.Drullinger, G.H.Stein, Uncoupling between phenotypic senescence and cell cycle arrest in aging p21-deficient fibroblasts. Mol.Cell Biol. 20 (2000) 6741-6754.
[51] J.Martin-Caballero, J.M.Flores, P.Garcia-Palencia, M.Serrano, Tumor susceptibility of p21(Waf1/Cip1)-deficient mice. Cancer Res. 61 (2001) 6234-6238.
[52] J.Brugarolas, R.T.Bronson, T.Jacks, p21 is a critical CDK2 regulator essential for proliferation control in Rb-deficient cells. J.Cell Biol. 141 (1998) 503-514.
[53] D.S.Franklin, V.L.Godfrey, D.A.O'Brien, C.Deng, Y.Xiong, Functional collaboration between different cyclin-dependent kinase inhibitors suppresses tumor growth with distinct tissue specificity. Mol.Cell Biol. 20 (2000) 6147-6158.
[54] W.C.Yang, J.Mathew, A.Velcich, W.Edelmann, R.Kucherlapati, M.Lipkin, K.Yang, L.H.Augenlicht, Targeted inactivation of the p21(WAF1/CIP1) gene enhances Apc-initiated tumor formation and the tumor-promoting activity of a Western-style high-risk diet by altering cell maturation in the intestinal mucosal. Cancer Res. 61 (2001) 565-569.
[55] J.Adnane, R.J.Jackson, S.V.Nicosia, A.B.Cantor, W.J.Pledger, S.M.Sebti, Loss of p21^WAF1/CIP1 accelerates Ras oncogenesis in a transgenic/knockout mammary cancer model. Oncogene 19 (2000) 5338-5347.
[56] M.Lebel, R.D.Cardiff, P.Leder, Tumorigenic effect of nonfunctional p53 or p21 in mice mutant in the Werner syndrome helicase. Cancer Res. 61 (2001) 1816-1819.
[57] Y.A.Wang, A.Elson, P.Leder, Loss of p21 increases sensitivity to ionizing radiation and delays the onset of lymphoma in atm-deficient mice. Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 14590-14595.
[58] B.Krolewski and J.B.Little, Overexpression of p21 protein in radiation-transformed mouse 10T(1/2) cell clones. Mol.Carcinog. 27 (2000) 141-148.
[59] K.Bhatia, S.Fan, G.Spangler, M.Weintraub, P.M.O'Connor, J.G.Judde, I.Magrath, A mutant p21 cyclin-dependent kinase inhibitor isolated from a Burkitt's lymphoma. Cancer Res. 55 (1995) 1431-1435.
[60] X. Gao, Y.Q. Chen, N. Wu, D.J. Grignon, W. Sakr, A.T. Porter, K.V. Honn, Somatic mutations of the WAF1/CIP1 gene in primary prostate cancer. Oncogene 11 (1995) 1395-1398.
[61] M.J.Vidal, F.Loganzo, Jr., A.R.de Oliveira, N.K.Hayward, A.P.Albino, Mutations and defective expression of the WAF1 p21 tumour-suppressor gene in malignant melanomas. Melanoma Res 5 (1995) 243-250.
[62] M.Balbin, G.J.Hannon, A.M.Pendas, A.A.Ferrando, F.Vizoso, A.Fueyo, C.Lopez-Otin, Functional analysis of a p21^{WAF1,CIP1,SDI1} mutant (Arg94 --> Trp) identified in a human breast carcinoma. Evidence that the mutation impairs the ability of p21 to inhibit cyclin-dependent kinases. J.Biol.Chem. 271 (1996) 15782-15786.
[63] S.B.Malkowicz, J.E.Tomaszewski, A.J.Linnenbach, T.A.Cangiano, Y.Maruta, T.W.McGarvey, Novel p21^WAF1/CIP1 mutations in superficial and invasive transitional cell carcinomas. Oncogene 13 (1996) 1831-1837.
[64] M.Welcker, J.Lukas, M.Strauss, J.Bartek, p21WAF1/CIP1 mutants deficient in inhibiting cyclin-dependent kinases (CDKs) can promote assembly of active cyclin D/CDK4(6) complexes in human tumor cells. Cancer Res. 58 (1998) 5053-5056.
[65] M.Shiohara, W.S.El Deiry, M.Wada, T.Nakamaki, S.Takeuchi, R.Yang, D.L.Chen, B.Vogelstein, H.P.Koeffler, Absence of WAF1 mutations in a variety of human malignancies. Blood 84 (1994) 3781-3784.
[66] K.E.McKenzie, A.Siva, S.Maier, I.B.Runnebaum, R.Seshadri, S.Sukumar, Altered WAF1 genes do not play a role in abnormal cell cycle regulation in breast cancers lacking p53 mutations. Clin Cancer Res 3 (1997) 1669-1673.
[67] A.Patino-Garcia, E.Sotillo-Pineiro, L.Sierrasesumaga-Ariznabarreta, p21^WAF1 mutation is not a predominant alteration in pediatric bone tumors. Pediatr.Res. 43 (1998) 393-395.
[68] T.Komiya, Y.Hosono, T.Hirashima, N.Masuda, T.Yasumitsu, K.Nakagawa, M.Kikui, A.Ohno, M.Fukuoka, I.Kawase, p21 expression as a predictor for favorable prognosis in squamous cell carcinoma of the lung. Clin.Cancer Res. 3 (1997) 1831-1835.
[69] T.K.Zirbes, S.E.Baldus, S.P.Moenig, S.Nolden, D.Kunze, S.T.Shafizadeh, P.M.Schneider, J.Thiele, A.H.Hoelscher, H.P.Dienes, Prognostic impact of p21/waf1/cip1 in colorectal cancer. Int.J.Cancer 89 (2000) 14-18.
[70] X.Lu, T.Toki, I.Konishi, T.Nikaido, S.Fujii, Expression of p21^WAF1/CIP1 in adenocarcinoma of the uterine cervix: a possible immunohistochemical marker of a favorable prognosis. Cancer 82 (1998) 2409-2417.
[71] N.Kapranos, G.P.Stathopoulos, L.Manolopoulos, E.Kokka, C.Papadimitriou, A.Bibas, J.Yiotakis, G.Adamopoulos, p53, p21 and p27 protein expression in head and neck cancer and their prognostic value. Anticancer Res. 21 (2001) 521-528.
[72] O.Caffo, C.Doglioni, S.Veronese, M.Bonzanini, A.Marchetti, F.Buttitta, P.Fina, R.Leek, L.Morelli, P.D.Palma, A.L.Harris, M.Barbareschi, Prognostic value of p21(WAF1) and p53 expression in breast carcinoma: an immunohistochemical study in 261 patients with long-term follow-up. Clin.Cancer Res. 2 (1996) 1591-1599.
[73] M.Ogawa, N.Onoda, K.Maeda, Y.Kato, B.Nakata, S.M.Kang, M.Sowa, K.Hirakawa, A combination analysis of p53 and p21 in gastric carcinoma as a strong indicator for prognosis. Int.J.Mol.Med. 7 (2001) 479-483.
[74] M.A.Anttila, V.M.Kosma, J.Hongxiu, J.Puolakka, M.Juhola, S.Saarikoski, K.Syrjanen, p21/WAF1 expression as related to p53, cell proliferation and prognosis in epithelial ovarian cancer. Br.J.Cancer 79 (1999) 1870-1878.
[75] R.M.Elledge and D.C.Allred, Prognostic and predictive value of p53 and p21 in breast cancer. Breast Cancer Res.Treat. 52 (1998) 79-98.
[76] M.Baekelandt, R.Holm, C.G.Trope, J.M.Nesland, G.B.Kristensen, Lack of independent prognostic significance of p21 and p27 expression in advanced ovarian cancer: an immunohistochemical study. Clin.Cancer Res. 5 (1999) 2848-2853.
[77] U.J.Gohring, A.Bersch, M.Becker, W.Neuhaus, T.Schondorf, p21(waf) correlates with DNA replication but not with prognosis in invasive breast cancer. J.Clin.Pathol. 54 (2001) 866-870.
[78] P.V.Kaye, K.Radebold, S.Isaacs, D.M.Dent, Expression of p53 and p21^waf1/cip1 in gastric carcinoma: lack of inter- relationship or correlation with prognosis. Eur.J.Surg.Oncol. 26 (2000) 39-43.
[79] A.Haitel, H.G.Wiener, B.Neudert, M.Marberger, M.Susani, Expression of the cell cycle proteins P21, P27, and PRB in clear cell renal cell carcinoma and their prognostic significance. Urology 58 (2001) 477-481.
[80] G.B.Baretton, U.Klenk, J.Diebold, N.Schmeller, U.Lohrs, Proliferation- and apoptosis-associated factors in advanced prostatic carcinomas before and after androgen deprivation therapy: prognostic significance of p21/WAF1/CIP1 expression. Br.J. Cancer 80 (1999) 546-555.
[81] S.Aaltomaa, P.Lipponen, M.Eskelinen, M.Ala-Opas, V.M.Kosma, Prognostic value and expression of p21(waf1/cip1) protein in prostate cancer. Prostate 39 (1999) 8-15.
[82] G.Ferrandina, A.Stoler, A.Fagotti, F.Fanfani, R.Sacco, A.De Pasqua, S.Mancuso, G.Scambia, p21^WAF1/CIP1 protein expression in primary ovarian cancer. Int.J.Oncol. 17 (2000) 1231-1235.
[83] T.H.Cheung, K.W.Lo, M.M.Yu, S.F.Yim, C.S.Poon, T.K.Chung, Y.F.Wong, Aberrant expression of p21(WAF1/CIP1) and p27(KIP1) in cervical carcinoma. Cancer Lett. 172 (2001) 93-98.
[84] D.S.Bae, S.B.Cho, Y.J.Kim, J.D.Whang, S.Y.Song, C.S.Park, D.S.Kim, J.H.Lee, Aberrant expression of cyclin D1 is associated with poor prognosis in early stage cervical cancer of the uterus. Gynecol.Oncol. 81 (2001) 341-347.
[85] C.Ceccarelli, D.Santini, P.Chieco, C.Lanciotti, M.Taffurelli, G.Paladini, D.Marrano, Quantitative p21(waf-1)/p53 immunohistochemical analysis defines groups of primary invasive breast carcinomas with different prognostic indicators. Int.J.Cancer 95 (2001) 128-134.
[86] M.Sarbia and H.E.Gabbert, Modern pathology: prognostic parameters in squamous cell carcinoma of the esophagus. Recent Results Cancer Res. 155 (2000) 15-27.
[87] J.M.Jung, J.M.Bruner, S.Ruan, L.A.Langford, A.P.Kyritsis, T.Kobayashi, V.A.Levin, W.Zhang, Increased levels of p21^WAF1/Cip1 in human brain tumors. Oncogene 11 (1995) 2021-2028.
[88] J.A.Pindzola, J.P.Palazzo, A.J.Kovatich, B.Tuma, M.Nobel, Expression of p21^WAF1/CIP1 in soft tissue sarcomas: a comparative immunohistochemical study with p53 and Ki-67. Pathol.Res.Pract. 194 (1998) 685-691.
[89] P.Korkolopoulou, A.E.Konstantinidou, E.Thomas-Tsagli, P.Christodoulou, P.Kapralos, P.Davaris, WAF1/p21 protein expression is an independent prognostic indicator in superficial and invasive bladder cancer. Appl.Immunohistochem.Molecul.Morphol. 8 (2000) 285-292.
[90] X.Li, A.Hui, T.Takayama, X.Cui, Y.Shi, M.Makuuchi, Altered p21(WAF1/CIP1) expression is associated with poor prognosis in extrahepatic bile duct carcinoma. Cancer Lett. 154 (2000) 85-91.
[91] R.H.Weiss, A.Joo, C.Randour, p21(Waf1/Cip1) is an assembly factor required for platelet-derived growth factor-induced vascular smooth muscle cell proliferation. J.Biol.Chem. 275 (2000) 10285-10290.
[92] S.Bates, K.M.Ryan, A.C.Phillips, K.H.Vousden, Cell cycle arrest and DNA endoreduplication following p21^Waf1/Cip1 expression. Oncogene 17 (1998) 1691-1703.
[93] H.Zhu, B.D.Chang, T.Uchiumi, I.B.Roninson, Identification of promoter elements responsible for transcriptional inhibition of Polo-like kinase 1 and Topoisomerase II genes by p21^{WAF1/CIP1/SDI1}. Cell Cycle in press (2002).
[94] Y.Matsumoto, K.Hayashi, E.Nishida, Cyclin-dependent kinase 2 (Cdk2) is required for centrosome duplication in mammalian cells. Curr.Biol. 9 (1999) 429-432.
[95] D.H.Ly, D.J.Lockhart, R.A.Lerner, P.G.Schultz, Mitotic misregulation and human aging. Science 287 (2000) 2486-2492.
[96] Y.Zhang, N.Fujita, T.Tsuruo, Caspase-mediated cleavage of p21^Waf1/Cip1 converts cancer cells from growth arrest to undergoing apoptosis. Oncogene 18 (1999) 1131-1138.
[97] I.B.Roninson, E.V.Broude, B.D.Chang, If not apoptosis, then what? Treatment-induced senescence and mitotic catastrophe in tumor cells. Drug Resistance Updates in press (2001).
[98] K.Polyak, T.Waldman, T.C.He, K.W.Kinzler, B.Vogelstein, Genetic determinants of p53-induced apoptosis and growth arrest. Genes Dev 10 (1996) 1945-1952.
[99] M.Gorospe, C.Cirielli, X.Wang, P.Seth, M.C.Capogrossi, N.J.Holbrook, p21(Waf1/Cip1) protects against p53-mediated apoptosis of human melanoma cells. Oncogene 14 (1997) 929-935.
[100] C.van den Bos, S.Silverstetter, M.Murphy, T.Connolly, p21(cip1) rescues human mesenchymal stem cells from apoptosis induced by low-density culture. Cell Tissue Res. 293 (1998) 463-470.
[101] K.Ozaki and S.Hanazawa, Porphyromonas gingivalis fimbriae inhibit caspase-3-mediated apoptosis of monocytic THP-1 cells under growth factor deprivation via extracellular signal-regulated kinase-dependent expression of p21CIP1/WAF1. Infect.Immun. 69 (2001) 4944-4950.
[102] M.Mahyar-Roemer and K.Roemer, p21 Waf1/Cip1 can protect human colon carcinoma cells against p53- dependent and p53-independent apoptosis induced by natural chemopreventive and therapeutic agents. Oncogene 20 (2001) 3387-3398.
[103] T.Glaser, B.Wagenknecht, M.Weller, Identification of p21 as a target of cycloheximide-mediated facilitation of CD95-mediated apoptosis in human malignant glioma cells. Oncogene 20 (2001) 4757-4767.
[104] R.Marches, R.Hsueh, J.W.Uhr, Cancer dormancy and cell signaling: induction of p21(waf1) initiated by membrane IgM engagement increases survival of B lymphoma cells. Proc.Natl.Acad.Sci.U.S.A 96 (1999) 8711-8715.
[105] M.Asada, T.Yamada, H.Ichijo, D.Delia, K.Miyazono, K.Fukumuro, S.Mizutani, Apoptosis inhibitory activity of cytoplasmic p21(Cip1/WAF1) in monocytic differentiation. EMBO J. 18 (1999) 1223-1234.
[106] Z.Jiang, P.Liang, R.Leng, Z.Guo, Y.Liu, X.Liu, S.Bubnic, A.Keating, D.Murray, P.Goss, E.Zacksenhaus, E2F1 and p53 are dispensable, whereas p21(Waf1/Cip1) cooperates with Rb to restrict endoreduplication and apoptosis during skeletal myogenesis. Dev.Biol. 227 (2000) 8-41.
[107] W.S.El Deiry, T.Tokino, T.Waldman, J.D.Oliner, V.E.Velculescu, M.Burrell, D.E.Hill, E.Healy, J.L.Rees, S.R.Hamilton, Topological control of p21^WAF1/CIP1 expression in normal and neoplastic tissues. Cancer Res. 55 (1995) 2910-2919.
[108] D.L.Kramer, B.D.Chang, Y.Chen, P.Diegelman, K.Alm, A.R.Black, I.B.Roninson, C.W.Porter, Polyamine depletion in human melanoma cells leads to G1 arrest associated with induction of p21^{WAF1/CIP1/SDI1}, changes in the expression of p21-regulated genes, and a senescence-like phenotype. Cancer Res. 61 (2001) 7754-7762.
[109] S.A.Gayther, S.J.Batley, L.Linger, A.Bannister, K.Thorpe, S.F.Chin, Y.Daigo, P.Russell, A.Wilson, H.M.Sowter, J.D.Delhanty, B.A.Ponder, T.Kouzarides, C.Caldas, Mutations truncating the EP300 acetylase in human cancers. Nat.Genet. 24 (2000) 300-303.
[110] A.Krtolica, S.Parrinello, S.Lockett, P.Y.Desprez, J.Campisi, Senescent fibroblasts promote epithelial cell growth and tumorigenesis: A link between cancer and aging. Proc.Natl.Acad.Sci.U.S.A 98 (2001) 12072-12077.
[111] J.D.Weber, L.J.Taylor, M.F.Roussel, C.J.Sherr, D.Bar-Sagi, Nucleolar Arf sequesters Mdm2 and activates p53. Nat.Cell Biol. 1 (1999) 20-26.
[112] V.Paradis, N.Youssef, D.Dargere, N.Ba, F.Bonvoust, J.Deschatrette, P.Bedossa, Replicative senescence in normal liver, chronic hepatitis C, and hepatocellular carcinomas. Hum.Pathol. 32 (2001) 327-332.
[113] B.Elenbaas and R.A.Weinberg, Heterotypic signaling between epithelial tumor cells and fibroblasts in carcinoma formation. Exp.Cell Res. 264 (2001) 169-184.
[114] M.Gregoire and B.Lieubeau, The role of fibroblasts in tumor behavior. Cancer Metastasis Rev. 14 (1995) 339-350.
[115] J.L.Camps, S.M.Chang, T.C.Hsu, M.R.Freeman, S.J.Hong, H.E.Zhau, A.C.von Eschenbach, L.W.Chung, Fibroblast-mediated acceleration of human epithelial tumor growth in vivo. Proc.Natl.Acad.Sci.U.S.A 87 (1990) 75-79.
[116] A.Noel, A.Hajitou, C.L'Hoir, E.Maquoi, E.Baramova, J.M.Lewalle, A.Remacle, F.Kebers, P.Brown, C.M.Calberg-Bacq, J.M.Foidart, Inhibition of stromal matrix metalloproteases: effects on breast-tumor promotion by fibroblasts. Int.J.Cancer 76 (1998) 267-273.
[117] A.F.Olumi, P.Dazin, T.D.Tlsty, A novel coculture technique demonstrates that normal human prostatic fibroblasts contribute to tumor formation of LNCaP cells by retarding cell death. Cancer Res. 58 (1998) 4525-4530.
[118] L.W.Chung, Fibroblasts are critical determinants in prostatic cancer growth and dissemination. Cancer Metastasis Rev. 10 (1991) 263-274.
[119] A.F.Olumi, G.D.Grossfeld, S.W.Hayward, P.R.Carroll, T.D.Tlsty, G.R.Cunha, Carcinoma-associated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer Res. 59 (1999) 5002-5011.
[120] R.Montesano, M.S.Pepper, L.Orci, Paracrine induction of angiogenesis in vitro by Swiss 3T3 fibroblasts. J.Cell Sci. 105 ( Pt 4) (1993) 1013-1024.
[121] A.Desmouliere, A.Geinoz, F.Gabbiani, G.Gabbiani, Transforming growth factor-beta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J.Cell Biol. 122 (1993) 103-111.
[122] M.B.Vaughan, E.W.Howard, J.J.Tomasek, Transforming growth factor-beta1 promotes the morphological and functional differentiation of the myofibroblast. Exp.Cell Res. 257 (2000) 180-189.
[123] B.I.Dalal, P.A.Keown, A.H.Greenberg, Immunocytochemical localization of secreted transforming growth factor- beta 1 to the advancing edges of primary tumors and to lymph node metastases of human mammary carcinoma. Am.J.Pathol. 143 (1993) 381-389.
[124] I.Reynisdottir, K.Polyak, A.Iavarone, J.Massague, Kip/Cip and Ink4 Cdk inhibitors cooperate to induce cell cycle arrest in response to TGF-beta. Genes Dev. 9 (1995) 1831-1845.
[125] C.Frippiat, Q.M.Chen, S.Zdanov, J.P.Magalhaes, J.Remacle, O.Toussaint, Subcytotoxic H2O2 stress triggers a release of transforming growth factor-beta 1, which induces biomarkers of cellular senescence of human diploid fibroblasts. J.Biol.Chem. 276 (2001) 2531-2537.
[126] M.H.Barcellos-Hoff and S.A.Ravani, Irradiated mammary gland stroma promotes the expression of tumorigenic potential by unirradiated epithelial cells. Cancer Res. 60 (2000) 1254-1260.
[127] K.M.Meyer, S.M.Hess, T.D.Tlsty, S.A.Leadon, Human mammary epithelial cells exhibit a differential p53-mediated response following exposure to ionizing radiation or UV light. Oncogene 18 (1999) 5795-5805.
[128] E.Migliaccio, M.Giorgio, S.Mele, G.Pelicci, P.Reboldi, P.P.Pandolfi, L.Lanfrancone, P.G.Pelicci, The p66^shc adaptor protein controls oxidative stress response and life span in mammals. Nature 402 (1999) 309-313.
[129] D.Balomenos, J.Martin-Caballero, M.I.Garcia, I.Prieto, J.M.Flores, M.Serrano, A.Martinez, The cell cycle inhibitor p21 controls T-cell proliferation and sex- linked lupus development. Nat.Med. 6 (2000) 171-176.
[130] J.Megyesi, P.M.Price, E.Tamayo, R.L.Safirstein, The lack of a functional p21(WAF1/CIP1) gene ameliorates progression to chronic renal failure. Proc.Natl.Acad.Sci.U.S.A 96 (1999) 10830-10835.
[131] M.Al Douahji, J.Brugarolas, P.A.Brown, C.O.Stehman-Breen, C.E.Alpers, S.J.Shankland, The cyclin kinase inhibitor p21WAF1/CIP1 is required for glomerular hypertrophy in experimental diabetic nephropathy. Kidney Int. 56 (1999) 1691-1699.
[132] M.Murphy, C.Godson, S.Cannon, S.Kato, H.S.Mackenzie, F.Martin, H.R.Brady, Suppression subtractive hybridization identifies high glucose levels as a stimulus for expression of connective tissue growth factor and other genes in human mesangial cells. J.Biol.Chem. 274 (1999) 5830-5834.
[133] M.J.Adler, C.Coronel, E.Shelton, J.E.Seegmiller, N.N.Dewji, Increased gene expression of Alzheimer disease beta-amyloid precursor protein in senescent cultured fibroblasts. Proc.Natl.Acad.Sci.U.S.A 88 (1991) 16-20.
[134] P.G.Komarov, E.A.Komarova, R.V.Kondratov, K.Christov-Tselkov, J.S.Coon, M.V.Chernov, A.V.Gudkov, A chemical inhibitor of p53 that protects mice from the side effects of cancer therapy. Science 285 (1999) 1733-1737.
[135] E.Kopp and S.Ghosh, Inhibition of NF-kappa B by sodium salicylate and aspirin. Science 265 (1994) 956-959.
[136] Y.Yamamoto, M.J.Yin, K.M.Lin, R.B.Gaynor, Sulindac inhibits activation of the NF-kappaB pathway. J.Biol.Chem. 274 (1999) 27307-27314.
[137] W.Dempke, C.Rie, A.Grothey, H.J.Schmoll, Cyclooxygenase-2: a novel target for cancer chemotherapy? J.Cancer Res.Clin.Oncol. 127 (2001) 411-417.

Acknowledgements.
I thank all the members of our laboratory whose work has been described in this review. The author’s studies have been supported by grants R01 CA89636, R01 CA62099 andRR37 CA40333 from the U.S. National Cancer Institute.

Figure legends

Figure 1. p21 expression and tumor-stromal interactions.

Back . . .