Effects of ectopic overexpression of p21(WAF1/CIP1) on aneuploidy and the malignant phenotype of human brain tumor cells

p21WAF1/CIP1 is a downstream effector of the p53 tumor suppressor gene and a universal cyclin-dependent kinase (CDK) inhibitor. To determine the ability of p21WAF1/CIP1 to function as a tumor suppressor, we constructed a replication-defective adenovirus vector containing p21WAF1/CIP1 (Adp21WAF1/CIP1) to effect ectopic overexpression in a p53-defective human astrocytoma cell line, U-373MG. We observed a marked decrease in CDC2 and CDK2 kinase activity associated with a corresponding decrease in the amount of CDC2 but not CDK2 protein; a decreased growth potential of Adp21WAF1/CIP1-infected cells demonstrated by diminished [3H]thymidine incorporation, increased cell doubling time and G1-arrested cell cycle; an association between Adp21WAF1/CIP1-infected cells and inhibition of aneuploid cell accumulation; and an alteration of the malignant phenotype of cells was evidenced by the loss of anchorage-independent growth in soft agar and the failure to induce tumorigenesis in both peripheral and intracerebral xenograft models, including the prevention of tumor formation Adp21WAF1/CIP1 infection 2 days post tumor cell implantation. Adp21WAF1/CIP1. Adp21WAF1/CIP1 appears to be a strong candidate for gene therapy studies based on these studies indicating that Adp21WAF1/CIP1 inhibits proliferation, tumorigenicity and aneuploidy in human brain tumor cells.     

Analysis of p21WAF1/CIP1 in primary bladder tumors

The p21WAF1/CIP1 gene is regulated by p53 and encodes a cyclin-dependent kinase (Cdk)-inhibitor involved in senescence and cell quiescence. The role of p21 as a negative regulator of cell proliferation suggests that it may function as a tumor suppressor gene. However, only a few mutations of the p21WAF1/CIP1 gene have been reported to date. In order to assess potential p21WAF1/CIP1 gene alterations in human bladder cancer, we have examined this gene and its encoded product in a well-characterized cohort of 27 primary bladder tumors. Mobility shifts by single-strand conformation polymorphism in the p21WAF1/CIP1 gene were identified in 2 cases. Sequencing analyses revealed that one of these cases had point mutations in the 3' untranslated region, while the other case had a frame shift mutation at positions 322 (C to A) and a deletion of 8 nucleotides (323-->331; CCG-->ACG, codon 81 Arg-->Thr) that produced a stop signal at codon 83 (Gly--Stop). This tumor had a p21-negative phenotype by immunohistochemistry, but did not lose any allele. We further characterized these cases by the study of TP53 mutations using single-strand conformation polymorphism (PCR-SSCP) and sequencing, as well as immunohistochemical assays. Seven mobility shifts were identified and seven cases showed p53 nuclear accumulation. The two cases displaying mutated p21WAF1/CIP1 had wild-type TP53. It is concluded that p21WAF1/CIP1 gene aberrations are infrequent in bladder carcinoma but may be occasionally identified in primary bladder tumors.     

Dexamethasone-mediated protection from drug cytotoxicity: association with p21WAF1/CIP1 protein accumulation?

Dexamethasone (DEX)-mediated inhibition of drug-induced, but not CD95 ligand-induced, apoptosis in malignant glioma cells correlates with wild-type p53 status. Here, we examined mechanisms underlying DEX-mediated protection from apoptosis. DEX did not induce p53 expression in two p53 wild-type cell lines (U87MG, LN-229) and did not alter drug-induced p53 accumulation. Forced expression of temperature-sensitive p53val135 in mutant conformation failed to prevent accumulation of endogenous wild-type p53 but acted in a transdominant negative manner to inhibit p53-mediated, camptothecin-induced p21WAF1/CIP1 expression. p53val135-transfected cells retained responsiveness to DEX at restrictive temperature, suggesting that p53 activity is not required for cytoprotection. Forced expression of wild-type p53val135 abrogated the protective effect of DEX, suggesting redundant cytoprotective effects of DEX and p53. Indeed, DEX induced moderate accumulation of p21WAF1/CIP1 in U87MG, LN-229 and p53 mutant LN-18 cells, but not in p53 mutant LN-308 or T98G cells. LN-18 is also the p53 mutant cell line with the best cytoprotective response to DEX. p21WAF1/CIP1 accumulation occurred in the absence of changes in p21WAF1/CIP1 mRNA expression. Wild-type p53 was not required for this DEX effect since DEX induced p21WAF1/CIP1 accumulation in p53val135-transfected LN-229 cells, too. DEX failed to induce p21WAF1/CIP1 expression or cytoprotection in untransformed rat astrocytes. The same lack of modulation of p21WAF1/CIP1 expression and drug toxicity was observed in p21(+/+), p21(+/-) and p21(-/-) human colon carcinoma cells. Paradoxically, while only p21(+/+) and p21(+/-) mouse embryonic fibroblasts showed enhance p21WAF1/CIP1 levels after exposure to DEX, only p21(-/-) fibroblasts were protected from drug toxicity by DEX. The present study links DEX-mediated protection from cancer chemotherapy to a p53-independent pathway of regulating p21WAF1/CIP1 expression in glioma cells but this effect appears to cell type-specific.     

Platelet-derived growth factor (PDGF)-signaling mediates radiation-induced apoptosis in human prostate cancer cells with loss of p53 function

Platelet-derived growth factor (PDGF) signals a diversity of cellular responses in vitro, including cell proliferation, survival, transformation, and chemotaxis. PDGF functions as a "competence factor" to induce a set of early response genes expressed in G1 including p21WAF1/CIP1, a functional mediator of the tumor suppressor gene p53 in G1/S checkpoint. For PDGF-stimulated cells to progress beyond G1 and transit the cell cycle completely, progression factors in serum such as insulin and IGF-1 are required. We have recently shown a novel role of PDGF in inducing apoptosis in growth-arrested murine fibroblasts. The PDGF-induced apoptosis is rescued by insulin, suggesting that G1/S checkpoint is a critical determinant for PDGF-induced apoptosis. Because recent studies suggest that radiation-induced signal transduction pathways interact with growth factor-mediated signaling pathways, we have investigated whether activation of the PDGF-signaling facilitates the radiation-induced apoptosis in the absence of functional p53. For this study we have used the 125-IL cell line, a mutant p53-containing, highly metastatic, and hormone-unresponsive human prostate carcinoma cell line. PDGF signaling is constitutively activated by transfection with a p28v-sis expression vector, which was previously shown to activate PDGF alpha- and beta- receptors. Although the basal level of p21WAF1/CIP1 expression and radiation-induced apoptosis were not detectable in control 125-IL cells as would be predicted in mutant p53-containing cells, activation of PDGF-signaling induced expression of p21WAF1/CIP1 and radiation-induced apoptosis. Our study suggests that the level of "competence" growth factors including PDGF may be one of the critical determinants for radiation-induced apoptosis, especially in cells with loss of p53 function at the site of radiotherapy in vivo.     

The geranylgeranyltransferase-I inhibitor GGTI-298 arrests human tumor cells in G0/G1 and induces p21(WAF1/CIP1/SDI1) in a p53-independent manner

Recently we have shown that in fibroblasts (NIH 3T3 and Rat-1 cells) inhibition of protein geranylgeranylation leads to a G0/G1 arrest, whereas inhibition of protein farnesylation does not affect cell cycle distribution. Here we demonstrate that in human tumor cells the geranylgeranyltransferase-I (GGTase-I) inhibitor GGTI-298 blocked cells in G0/G1, whereas the farnesyltransferase (FTase) inhibitor FTI-277 showed a differential effect depending on the cell line. FTI-277 accumulated Calu-1 and A-549 lung carcinoma and Colo 357 pancreatic carcinoma cells in G2/M, T-24 bladder carcinoma, and HT-1080 fibrosarcoma cells in G0/G1, but had no effect on cell cycle distribution of pancreatic (Panc-1), breast (SKBr 3 and MDAMB-231), and head and neck (A-253) carcinoma cells. Furthermore, treatment of Calu-1, Panc-1, Colo 357, T-24, A-253, SKBr 3, and MDAMB-231 cells with GGTI-298, but not FTI-277, induced the protein expression levels of the cyclin-dependent kinase inhibitor p21WAF. HT-1080 and A-549 cells had a high basal level of p21WAF, and GGTI-298 did not further increase these levels. Furthermore, GGTI-298 also induces the accumulation of large amounts of p21WAF mRNA in Calu-1 cells, a cell line that lacks the tumor suppressor gene p53. There was little effect of GGTI-298 on the cellular levels of another cyclin- dependent kinase inhibitor p27KIP as well as cyclin E and cyclin D1. These results demonstrate that GGTase-I inhibitors arrest cells in G0/G1 and induce accumulation of p21WAF in a p53-independent manner and that FTase inhibitors can interfere with cell cycle events by a mechanism that involves neither p21WAF nor p27KIP. The results also point to the potential of GGTase-I inhibitors as agents capable of restoring growth arrest in cells lacking functional p53.     

Polyomavirus large T antigen overcomes p53 dependent growth arrest

Polyomavirus transforms cells in culture and induces tumors in mice without apparent interaction with or inactivation of the p53 tumor suppressor protein. In this report we investigate the ability of polyomavirus T antigens to overcome the growth suppression function of p53. A temperature sensitive p53 gene was introduced into mouse embryo fibroblasts derived from a p53 null mouse, resulting in expression of a protein with a mutant conformation at 37 degrees C and a functionally wild-type conformation at 32 degrees C. We found that expression of p53 at 32 degrees C induced the cyclin-dependent kinase inhibitor p21/WAF1 and arrested cell growth in the G1/G0 phase of the cell cycle. Only the under-phosphorylated form of the retinoblastoma tumor suppressor protein (pRB) was detected in these growth arrested cells. We introduced both polyomavirus large T (LT) and middle T (MT) antigens into this cell line and showed that LT overcame p53-dependent growth arrest, while MT did not. In cells grown at 32 degrees C, LT expession led to cell proliferation and phosphorylation of pRB in the presence of p21. A mutant LT containing a defective pRB binding domain failed to overcome the growth arrest, indicating that interaction of LT with RB proteins is required to override p53 function. Although the polyomavirus T antigens do not interact with p53 directly, our results indicate that the virus, through LT, is able to interfere with the growth suppressive activity of p53.     

Cytotoxic effects of recombinant adenovirus p53 and cell cycle regulator genes (p21 WAF1/CIP1 and p16CDKN4) in human prostate cancers

PURPOSE: Overexpressing or restoring the basal levels of tumor suppressor genes in cancer cells can suppress tumorigenicity of cancer cells. In this communication, we compared tumor suppressive activities of three well-defined tumor suppressive genes (p53, p21WAF1/CIP1, and p16CDKN2) delivered individually to prostate cancer cells with adenoviral vector (Ad). MATERIALS AND METHODS: Efficacy of growth inhibition by recombinant adenoviruses bearing p53, p21WAF1/CIP1, or p16CDKN2 (Ad5CMV-p53, Ad5CMV-p21, Ad5CMV-p16) genes were tested in vitro on androgen-dependent (LNCaP) and androgen-independent (C4-2, DU-145, and PC-3) human prostate cancer cells, ex vivo and in vivo on PC-3 tumor. RESULTS: Ad5CMV-p53 was observed to exert the greatest growth inhibitory action on all of the cell lines tested; inhibition appeared to be cytolytic. In comparison to control Ad5CMV-PA added samples, the growth inhibitory action of Ad5CMV-p21 and Ad5CMV-p16 appeared to be cytostatic. Ad5CMV-p53 is more effective than Ad5CMV-p16 and Ad5CMV-p21 in inhibiting the tumor "take" rate. A similar order of antitumor activity was observed when recombinant adenoviruses were injected intratumorily to previously established PC-3 tumors in vivo. CONCLUSION: p53 is the most effective tumor suppressor gene to target human prostate cancer.     

Release of cell cycle constraints in mouse melanocytes by overexpressed mutant E2F1E132, but not by deletion of p16INK4A or p21WAF1/CIP1

Compared to normal melanocytes, melanoma cell lines exhibit overexpression of hyperphosphorylated retinoblastoma tumor suppressor protein (Rb) or a marked decrease in, or lack of, expression of Rb. Hyperphosphorylation of Rb results in increased E2F-mediated transactivation of target genes and cell cycle progression. Using a combination of gene disruption and ectopic expression in growth factor-dependent mouse melanocytes, we studied the roles of E2F1 and the p16INK4A and p21WAF1/CIP1 CKIs (cyclin dependent kinase inhibitors) in the acquisition of TPA (12-O-tetradecanoyl phorbol-13-acetate)-independent growth in culture, a hallmark of melanomas. Surprisingly, melanocytes from p16INK4A- or p21WAF1/CIP1-null mice remained TPA-dependent, and disruption of p21WAF1/CIP1 accelerated cell death in the absence of this mitogen. Disruption of E2F1 had the most profound effect on melanocyte growth, resulting in a fourfold decrease in growth rate in the presence of TPA. Furthermore, enforced overexpression of the DNA-binding-defective E2F1E132 mutant conferred TPA-independence upon melanocytes and was associated with sequestration of Rb and constitutive expression of E2F1 target genes, including p21WAF1/CIP1. We conclude that neutralization of Rb by E2F1E132, but not the disruption of p16INK4A or p21WAF1/CIP1, resulted in the accumulation of free E2F and cell cycle progression. Thus, mechanisms other than the loss of p16INK4A or p21WAF1/CIP1 that activate E2F may play an important role in melanomas.     

Novel antipsychotics and new onset diabetes

Mutations of p53 tumor suppressor gene occur in a subset of aggressive prostatic carcinomas and are detectable by immunohistochemistry. However, it is uncertain whether p53 overexpression really reflects p53 gene mutation or loss of p53 function. p21WAF1, an inhibitor of cyclin-dependent kinases, is activated by wild-type p53 protein, not by mutant type. Therefore, it is possible that combined analysis of p21WAF1 and p53 proteins aids in determining the functional status of p53 immunostaining. Routinely processed prostatic tissues from 60 patients with prostatic adenocarcinomas were examined by immunohistochemistry for p21WAF1 and p53 expression. As for tissue distribution, p21WAF1 protein was expressed mostly in the luminal layers, in contrast, p53 protein was restricted to the basal layers of benign prostatic glands. In prostatic adenocarcinomas, p21WAF1 protein was more likely to be expressed in well-differentiated areas; in contrast, p53 protein was more likely in poorly differentiated areas in the tumors. The percentage of positive nuclear areas for p21WAF1 and p53 proteins in prostatic adenocarcinomas, assessed by CAS200 computerized image analyzer, were 8.6+/-10% and 16+/-14% (mean+/-SE), respectively. The survival study showed that the p53+/ p21- phenotype showed poorer prognosis than p53+/p21+. Multivariate analysis showed that p21WAF1 expression, clinical stage, and Gleason score were independent prognosticators. In conclusion, p21WAF1 immunohistochemistry is a useful method for interpretation of p53 immunohistochemical results. Combined analysis by p21WAF1 and p53 immunostaining would predict the patient survival more accurately than p53 immunostaining alone.     

Wild-type p53 triggers a rapid senescence program in human tumor cells lacking functional p53

The p53 tumor suppressor gene has been shown to play an important role in determining cell fate. Overexpression of wild-type p53 in tumor cells has been shown to lead to growth arrest or apoptosis. Previous studies in fibroblasts have provided indirect evidence for a link between p53 and senescence. Here we show, using an inducible p53 expression system, that wild-type p53 overexpression in EJ bladder carcinoma cells, which have lost functional p53, triggers the rapid onset of G1 and G2/M growth arrest associated with p21 up-regulation and repression of mitotic cyclins (cyclin A and B) and cdc2. Growth arrest in response to p53 induction became irreversible within 48-72 h, with cells exhibiting morphological features as well as specific biochemical and ultrastructural markers of the senescent phenotype. These findings provide direct evidence that p53 overexpression can activate the rapid onset of senescence in tumor cells.     

IFNgamma inhibition of cell growth in glioblastomas correlates with increased levels of the cyclin dependent kinase inhibitor p21WAF1/CIP1

Glioblastoma is a highly aggressive form of brain cancer characterized by uncontrolled cell growth resulting from a loss of cell cycle regulation. In this study we determined the antiproliferative effects of interferon gamma (IFNgamma) on the glioblastoma cell lines T98G, SNB-19 and U-373, focusing on the ability of IFNgamma to increase levels of p21WAF1/CIP1, an important negative regulator of cell cycle events. IFNgamma was found to inhibit the growth of all cell lines, with inhibition ranging from 82.2% to 45.4%. Flow cytometry analysis showed that IFNgamma treatment caused a cell cycle delay in the G1 or S phases. The strength of this delay varied, correlating with the degree by which IFNgamma inhibited proliferation of each cell line. IFNgamma treatment increased the production of the cyclin dependent kinase inhibitor (CKI) p21WAF1/ CIP1 in all cell lines, the level and kinetics of production of which correlated with the degree and stage of inhibition of cellular proliferation. Further, immunoprecipitation of p21WAF1/CIP1 in complexes of p21WAF1/CIP1/cyclin-dependent kinase 2 (cdk2)/cyclin showed that the amount of p21WAF1/CIP1 in the complexes and the inhibition of cdk2-cyclin kinase activity correlated with the level of p21WAF1/CIP1 produced in the cells by IFNgamma. These results show that IFNgamma has significant antiproliferative effects on the glioblastoma cell lines and suggest that p21WAF1/CIP1 plays a role in mediating these effects.     

WAF1/CIP1 is induced in p53-mediated G1 arrest and apoptosis

The tumor growth suppressor WAF1/CIP1 was recently shown to be induced by p53 and to be a potent inhibitor of cyclin-dependent kinases. In the present studies, we sought to determine the relationship between the expression of WAF1/CIP1 and endogenous regulation of p53 function. WAF1/CIP1 protein was first localized to the nucleus of cells containing wild-type p53 and undergoing G1 arrest. WAF1/CIP1 was induced in wild-type p53-containing cells by exposure to DNA damaging agents, but not in mutant p53-containing cells. The induction of WAF1/CIP1 protein occurred in cells undergoing either p53-associated G1 arrest or apoptosis but not in cells induced to arrest in G1 or to undergo apoptosis through p53-independent mechanisms. DNA damage led to increased levels of WAF1/CIP1 in cyclin E-containing complexes and to an associated decrease in cyclin-dependent kinase activity. These results support the idea that WAF1/CIP1 is a critical downstream effector in the p53-specific pathway of growth control in mammalian cells.     

The abnormal p53 proteins expressed in CML cell lines are non-functional

Inactivation of the tumor suppressor function of the p53 gene is found in association with 20-40% cases of chronic myeloid leukemia (CML) in blast crisis. A common mechanism of p53 inactivation in CML is by complete deletion of one p53 allele in association with a point mutation which produces a mutant p53 protein on the remaining allele. Whether the mutant p53 protein, which is generally expressed at an elevated level, plays any role in the pathogenesis of blastic transformation or in maintaining the neoplastic proliferation, as it does in some solid tumors, is unknown. By using an antisense oligonucleotide approach, we investigated the cellular function of known abnormal forms of p53 protein, both mutant and truncated, expressed in CML cell lines. We found that the introduction of p53 antisense oligonucleotides can specifically inhibit the translation of the p53 mRNA. However, inhibiting p53 expression had no effect on cell proliferation, cell viability, and colony formation. There was no change in cell doubling time when the cells were maintained in serum-free medium (SFM) in the presence of antisense oligonucleotides compared with cells maintained in SFM alone. We conclude that the mutant or truncated p53 proteins expressed in the blast cells of CML have no growth-promoting effect and are not required for cell survival and proliferation. We further speculate that the loss of the tumor suppressor function of p53 might be the only mechanism by which p53 is involved in the transition from chronic phase to blast crisis.     

Expression of bcl-2 and bcl-X in bladder cancer

PURPOSE: TP53 and RB1 gene mutations in bladder transitional cell carcinoma (TCC) are correlated with grade, stage, recurrence, and survival and may correlate with tumor cell apoptotic potential. Overexpression of the bcl-2 and bcl-X anti-apoptotic genes has been correlated with poor prognosis and chemotherapy resistance in other systems. Similar studies have not been performed in TCC. We thus sought to determine expression of bcl-2 and bcl-X in TCC and correlate these with stage, survival and abnormal pRb or p53 expression. MATERIALS AND METHODS: Forty-two TCC samples (19 Ta and 23 locally advanced tumors) and normal urothelial controls were examined. Immunohistochemistry for p53, pRb, bcl-2 and bcl-X was performed on an automated system using indirect streptavidin biotin/horseradish peroxidase staining. Western immunoblot analysis was performed on bladder cancer cell lines to further characterize bcl-X expression. Recurrence-free and disease-specific survival were retrospectively determined. Kaplan-Meier survival curves were compared using the log rank test, and correlation of abnormal staining with stage and p53 or pRb status was determined using Fisher's exact test. RESULTS: Bcl-2 was expressed in less than 1% of normal urothelial cells, but moderate expression of bcl-x was found in all normal urothelial samples. Only 7.0% of TCC samples (1/19 Ta and 2/23 locally advanced tumors) demonstrated bcl-2 overexpression. Bcl-X overexpression was observed in 45.2% of TCC (8/19 Ta and 11/23 locally advanced tumors). Western blot analysis also revealed that both the long (29 kDa) anti-apoptotic form and short (19 kDa) pro-apoptotic form were overexpressed in bladder cancer cell lines and normal human urothelial cells. Bcl-X overexpression was weakly correlated with normal p53 expression (p = 0.06). There were no correlations of bcl-2 and bcl-X overexpression with abnormal p53, pRb, or tumor stage. There were no differences in recurrence-free or overall survival in patients with abnormal bcl-X staining. CONCLUSIONS: Bcl-2 overexpression is rare in TCC. Bcl-X overexpression is common, likely reflecting its expression pattern in normal urothelium, but is not correlated with stage or abnormal p53 or pRb staining. Within the power limitations of this small study, bcl-X overexpression is not correlated with recurrence or survival.     

Growth arrest and suppression of tumorigenicity of bladder-carcinoma cell lines induced by the P16/CDKN2 (p16INK4A, MTS1) gene and other loci on human chromosome 9

Wild-type P16/CDKN2 (p16INK4A, MTS1) cDNA, directed by the cytomegalovirus (CMV) immediate early promoter, was transfected into RT4 and RT112 bladder-carcinoma cell lines bearing a mutated endogenous P16/CDKN2 gene and lacking endogenous P16/CDKN2 respectively. In both cases, only transfected clones with rearranged exogenous P16/CDKN2 cDNA could be grown and propagated in cell culture. This result is reminiscent of transfection of wild-type p53 into cells with a deleted or mutated endogenous gene and suggests that P16/CDKN2, over-expressed under control of the strong CMV promoter, induces growth arrest in RT4 and RT112 cells. Transfer of human chromosome 9 to RT4 cells produced RT4/H9 hybrid clones retaining the P16/CDKN2 gene, since in RT4/H9 cell clones P16/CDKN2-gene expression is modulated by the physiological control of chromosomal regulatory sequence. All the RT4/H9 clones lost the entire chromosome 9, except clone 4 and clone 5, which maintained a deleted and an intact chromosome 9 respectively. Loss of several loci in 9p21, including P16/CDKN2, in tumors induced in nude mice by clone 4 and clone 5 suggests that P16/CDKN2 or other genes in 9p21 suppress tumorigenicity in bladder-carcinoma cells. Tumors induced by clone 4 and clone 5 show loss of markers in 9q. The regions 9q22.3, 9q32-33 and 9q34.2, which were maintained in the 2 clones and lost in their derived tumors, may contain tumor-suppressor genes relevant in bladder carcinoma. The results of this study suggest that the P16/CDKN2 gene controls growth of bladder-carcinoma cells when it is over-expressed, and may be involved in the development of bladder carcinoma, but other genes in 9p21 and 9q may participate in bladder-cancer progression.     

Potential role of WAF1/Cip1/p21 as a mediator of TGF-beta cytoinhibitory effect

Transforming growth factor-beta (TGF-beta) inhibits cell cycle progression of many types of human cells by arresting them in the G1 phase of the cell cycle. The arrest is mediated through interactions of various cyclin-dependent protein kinases (CDKs) and their inhibitors. We demonstrate that treatment with TGF-beta induces increased levels of WAF1/Cip1/p21, a potent inhibitor of various cyclin-CDK kinase activities, in two colon cancer cell lines (LS1034 and LS513), which are sensitive to TGF-beta-induced growth arrest. The induction in at least one of these cell lines (LS1034,p53-) is p53-independent. No WAF1 induction was observed after TGF-beta treatment in a third cell line (HT-29), which is completely insensitive to the cytoinhibitory effect of TGF-beta. In both LS513 and LS1034, WAF1 induction correlated with reduced cyclin E-associated kinase activity in vitro and suppression of the retinoblastoma susceptibility gene (Rb) protein phosphorylation in vivo. In addition, WAF1 was physically associated with cyclin E in the two sensitive cell lines. These results suggest that WAF1/Cip1/p21 is a mediator of cellular sensitivity to TGF-beta.