View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Cell, Vol. 103, 897–907, December 8, 2000, Copyright 2000 by Cell Press The Mammalian UV Response: c-Jun Induction Is Required for Exit from p53-Imposed Growth Arrest

Eitan Shaulian,* Martin Schreiber,† Fabrice Piu,* exposure rapidly stimulates c-Jun and ATF2 N-terminal Michelle Beeche,‡ Erwin F. Wagner,† phosphorylation, a modification that enhances their and Michael Karin*§ transactivation potential (Devary et al., 1992; Gupta et *Laboratory of Regulation and Signal al., 1995; Karin, 1995). These phosphorylation events Transduction are mediated by c-Jun N-terminal kinases (JNKs) (Hibi University of California, San Diego et al., 1993; De´ rijard et al., 1994; Gupta et al., 1995). 9500 Gilman Drive Although the mechanism of UV-mediated JNK activa- La Jolla, California 92093 tion is not fully understood, it involves activation of a † Research Institute of Molecular Biology MAP kinase (MAPK) signaling cascade by membrane Dr. Bohr-Gasse 7 proximal events rather than a response to DNA damage Vienna, A-1030 per se (Devary et al., 1992, 1993; Sachsenmaier et al., Austria 1994; Rosette and Karin, 1996). Several other genotoxic ‡ Laboratory of agents, including ionizing (IR) and mitomycin The Salk Institute C (MMC), do not activate the same signaling pathways 10010 N. Torrey Pines Road as UVC or UVB (Liu et al., 1996; Shaulian and Karin, La Jolla, California 92037 1999). Given the similarity in the type of and signaling pathways that are activated by UV radiation and mito- Summary gens, the UV response has been regarded as a “pseudo growth response,” but its exact physiological function The mammalian UV response results in rapid and dra- remained enigmatic. It was proposed that the UV re- matic induction of c-jun. Induction of a protooncogene, sponse may be similar in function to a wound healing normally involved in mitogenic responses, by a geno- response and therefore may facilitate tissue remodeling toxic agent that causes growth arrest seems paradoxi- and regeneration after extensive damage (Herrlich et al., cal. We now provide an explanation for the role of 1997). Induction of jun, fos, and AP-1 transcriptional c-Jun in the UV response of mouse fibroblasts. c-Jun is activity in response to UV irradiation seem to contribute necessary for cell-cycle reentry of UV-irradiated cells, to remodeling of sun-exposed human skin (Fisher et al., but does not participate in the response to ionizing 1996). radiation. Cells lacking c-Jun undergo prolonged cell- In addition to MAPK activation and induction of cycle arrest, but resist , whereas cells that growth-associated immediate-early genes, UV expo- express c-Jun constitutively do not arrest and undergo sure results in DNA damage through formation of pyrimi- apoptosis. This function of c-Jun is exerted through dine dimers and 6–4 photoproducts (Friedberg, 1995). negative regulation of p53 association with the UV-induced DNA damage results in either transient ar- promoter. Cells lacking c-Jun exhibit prolonged p21 rest of actively proliferating cells or elimination of cells induction, whereas constitutive c-Jun inhibits UV-medi- with irreparable DNA damage (Lane, 1992; Lu and Lane, ated p21 induction. 1993; Kaufmann and Wilson, 1994). Transient cell-cycle arrest provides cells with ample time for DNA repair before proceeding to replicate damaged DNA (Hartwell Introduction and Kastan, 1994; Elledge, 1996). The response to UV differs from the response to IR, which causes a different Analogous to the bacterial SOS response, the mamma- type of lesion: single- and double-strand breaks (Fried- lian UV response entails induction of gene expression berg, 1995). In fibroblasts, exposure to IR results in a in response to short wavelength (UV) radiation senescence-like state, including a prolonged growth ar- (Herrlich et al., 1997). Curiously, the program of UV- rest, known as replicative death (Nagasawa and Little, mediated gene induction is similar to the one activated 1983; Di Leonardo et al., 1994). In mammalian cells, the by phorbol ester tumor promoters and (Karin key regulatory involved in induction of growth and Herrlich, 1989; Holbrook and Fornace, 1991; Angel, arrest by UV or IR is p53 (Ko and Prives, 1996; Levine, 1995). Some of the genes most rapidly induced by UVC 1997; Oren, 1999). or UVB radiation are also immediate-early regu- Exposure to UV or IR results in rapid p53 accumula- lated genes, including fos and jun (reviewed by Holbrook tion, caused by stabilization of this otherwise short- and Fornace, 1991; Herrlich et al., 1992). In fact, the lived protein (Maltzman and Czyzyk, 1984; Kastan et al., c-jun protooncogene is one of the most UV-responsive 1991). p53 is a sequence-specific transcriptional regula- genes identified thus far (Devary et al., 1991). tor (Kern et al., 1991) and its accumulation causes induc- UV-mediated c-jun induction requires one or two di- tion of several target genes including p21waf1, , vergent AP-1 binding sites within the c-jun promoter, Gadd45, and Bax (see El-Deiry, 1998 and references recognizable by Jun:ATF2 heterodimers (Devary et al., within). p21 is an inhibitor of -dependent kinases 1991, 1992; van Dam et al., 1993; Herr et al., 1994). UV (Cdks), whose overexpression results in G1 and G2 ar- rests (Niculescu et al., 1998). Cells lacking functional § To whom correspondence should be addressed (e-mail: karinoffice@ p21 alleles fail to arrest in response to DNA damage ucsd.edu). (Brugarolas et al., 1995) and become very sensitive to Cell 898

induction of apoptosis (Waldman et al., 1996). Thus, p21 tive in nature (Devary et al., 1992). We wanted to deter- induction protects cells from the cytotoxic effects of mine the role of c-Jun in cell survival after UV irradiation. UVC, whereas p53 deficiency sensitizes mouse fibro- As c-junϪ/Ϫ MEFs undergo very early senescence after blasts to UV-induced cell death (Sheikh et al., 1997; 2–3 passages (Johnson et al., 1993; Schreiber et al., Bissonnette and Hunting, 1998). However, as continu- 1999), it was necessary to conduct these experiments ous p21 expression inhibits cell proliferation (El-Deiry with immortalized (3T3-like) c-junϩ/ϩ and c-junϪ/Ϫ fibro- et al., 1993; Niculescu et al., 1998) and causes premature blasts. All of the cell lines used in the present study senescence (Brown et al., 1997; Fang et al., 1999), an display normal p53 induction and have wt p53 alleles efficient protective response requires transient p21 in- (Schreiber et al., 1999). Three independently derived duction. c-junϪ/Ϫ fibroblast lines exhibited significantly reduced clo- The response to IR and other agents that induce DNA nogenic survival in comparison to two wt (c-junϩ/ϩ) cell strand breaks is dependent on ATM, the product of the lines (Figure 1A). After exposure to 18 J/m2 of UVC, Ataxia telangiectasia (AT) gene. ATM is a the average clonogenic survival of c-junϪ/Ϫ cells was that is activated in response to DNA strand breaks, re- approximately 3% in comparison to 20% for wt cells. sulting in either direct or indirect p53 phosphorylation By contrast, no significant differences were detected and stabilization (Banin et al., 1998; Canman et al., 1998). between the clonogenic survival of c-junϩ/ϩ and c-junϪ/Ϫ This pathway, however, is not activated by UV radiation cells exposed to IR (Figure 1B). Unlike UV radiation, IR (Kastan et al., 1992; Banin et al., 1998; Canman et al., does not increase c-Jun expression in mouse fibroblasts 1998). Induction of p53 by UV also differs in its kinetics (Figure 1E and see Liu et al., 1996; Shaulian and Karin, and magnitude from the response triggered by IR or 1999). The poor clonogenic survival of UV-irradiated radiomimetic agents, in being more prolonged and ro- c-junϪ/Ϫ cells was indeed due to lack of c-Jun, because bust (Lu and Lane, 1993; Zhan et al., 1993). An even less c-junϪ/Ϫ cells stably transfected with a human c-Jun ex- understood aspect of the UV response is the mecha- pression vector, the previously described C4 and C6 nisms responsible for cell-cycle reentry following p53 subclones (Schreiber et al., 1999), exhibited equal or and p21 induction. Such mechanisms are needed to better clonogenic survival than wt cells (Figure 1C). C4 avoid p21-induced senescence. and C6 cells constitutively express c-Jun at levels that Previous studies revealed that c-Jun is required for are similar to those achieved several hours after expo- progression from G1 to under normal culture sure to UVC (see Figure 3E). This effect was specific to conditions (Kovary and Bravo, 1991; Schreiber et al., c-Jun, as elevated expression of another oncoprotein, 1999). Surprisingly, genetic and biochemical analyses Ha-ras, had no effect on clonogenic survival (Figure 1D). attributed the major mitogenic activity of c-Jun to its Ϫ/Ϫ ability to downregulate p53 transcription; c-jun fibro- Ϫ Ϫ c-jun / Cells Are Less Sensitive to UV-Induced blasts were found to express higher basal levels of p53 Cell Death and p21 and, as a result, have lower G -cyclin-Cdk activ- 1 Clonogenic survival is a complex endpoint that is influ- ity (Schreiber et al., 1999). In addition, the loss of p53 enced by rates of cell death and/or proliferation, DNA relieves the proliferation defect of c-junϪ/Ϫ mouse em- and protein repair, and the ability to escape growth bryo fibroblasts (MEFs). Despite the changes in basal arrest and re-enter the . There is no straightfor- p53 transcription, p53 is normally induced in response ward correlation between apoptotic cell death and clo- to genotoxic damage in both c-jun null cells and cells nogenic survival (Finkel, 1999). We therefore examined that express c-Jun constitutively (Schreiber et al., 1999). Because the dramatic and rapid induction of c-jun the ability of the different cell lines to undergo UV- transcription is the hallmark of the mammalian UV re- induced cell death using a short term assay, based on sponse, we focused our efforts on understanding its annexin V staining (Vermes et al., 1995). Whereas expo- 2 physiological function. Using c-junϩ/ϩ and c-junϪ/Ϫ fibro- sure of wt cells to 20 J/m of UVC resulted in a 6-fold Ϫ/Ϫ blasts as well as c-junϪ/Ϫ cells programmed to reexpress increase in the rate of cell death, exposure of c-jun human c-Jun constitutively, we found that the major cells to the same dose of UVC did not significantly in- function of c-Jun in UV-irradiated cells is to promote crease the extent of annexin V staining measured at 24 cell-cycle reentry. This function is specific to UV-irradi- hr post exposure (Figures 2A and 2B). It should be noted, Ϫ/Ϫ ated cells, as c-Jun expression has little effect on the however, that c-jun cells have a somewhat higher response to IR, and is related to the unexpected ability basal rate of cell death than wt cells. By contrast, expo- of c-Jun to repress UV-induced p53-mediated p21 in- sure of c-Jun reexpressing cells (C4 and C6) to the same duction. c-Jun expression has little effect on p21 induc- dose of UVC resulted in a 10- to 18-fold increase in the tion by IR. Chromatin immunoprecipitation (ChIP) exper- extent of cell death (Figures 2A and 2B). Similar results iments (Braunstein et al., 1993; Alberts et al., 1998) were obtained using a dye exclusion assay (data not revealed that in UV-irradiated cells, c-Jun modulates shown), or a caspase 3 activation assay (Figure 2C), the association of p53 with the p21 promoter, thereby indicating that c-junϪ/Ϫ cells are relatively resistant to controlling the duration and intensity of p21 induction. UV-induced cell death and that constitutive c-Jun ex- pression in these cells restores and even increases UV Results sensitivity. The contribution of p53 to UV-induced cell death was Decreased Clonogenic Survival of UV-Irradiated examined by comparing c-junϪ/Ϫ p53Ϫ/Ϫ cells to c-junϩ/ϩ c-junϪ/Ϫ Cells p53Ϫ/Ϫ cells. The absence of p53 sensitized c-junϪ/Ϫ Previous experiments examining clonogenic survival as cells to UV-induced cell death and there was no differ- an endpoint suggested that the UV response is protec- ence in the extent of cell death between c-junϪ/Ϫ p53Ϫ/Ϫ c-Jun Mediates Cell-Cycle Exit by Antagonizing p53 899

Figure 1. c-Jun Increases the Clonogenic Survival of UV-Irradiated Mouse Fibroblasts Sparsely plated mouse fibroblasts that are either wt (c-junϩ/ϩ, two independent cell lines), c-jun null (c-junϪ/Ϫ, three independent cell lines), constitutive c-Jun reexpressors (C4 and C6, two independent cell lines), or Ha-ras transfectants were exposed to short wavelength UV light (A, C, or D) or IR (B) and colony formation was scored 2 weeks later. The number of colonies formed by untreated cultures of each cell type was given an arbi- trary value of 100% and all other values are depicted relative to that value. (E) c-junϩ/ϩ cells were exposed to the indicated doses of UVC or IR and the levels of c-Jun were determined by immunoblotting after 6 hr.

and c-junϩ/ϩp53Ϫ/Ϫ cells (Figure 2A). No further increase synthesis in c-junϪ/Ϫ cells to any greater extent than in in the extent of UV-induced cell death was observed wt cells (Figure 3B). upon reexpression of c-Jun in c-junϪ/Ϫp53Ϫ/Ϫ cells, while Once UV-irradiated cells have repaired their damaged reexpression of c-Jun in c-junϪ/Ϫp53ϩ/ϩ cells strongly DNA, they are expected to resume DNA synthesis and enhanced UV-induced death (Figure 2B). Thus, the stim- eventually re-enter the cell cycle. We exposed the cells ulation of UV-induced cell death by c-Jun is dependent to 16 J/m2 of UVC and measured BrdU incorporation on the presence of p53. into DNA during 12 hr periods at 24–36 hr and 48–60 hr past exposure. Consistent with the results described ϩ ϩ Ϫ Ϫ c-Jun Attenuates UV-Induced Growth Arrest above, DNA synthesis in both c-jun / and c-jun / cells and Enhances Cell-Cycle Reentry was inhibited 24 hr after irradiation, and the extent of Ϫ Ϫ To understand why c-junϪ/Ϫ cells are more susceptible inhibition was substantially greater in c-jun / cells (Fig- ϩ ϩ to UV-induced replicative death while being resistant to ure 3C). While DNA synthesis in c-jun / cells returned UV-induced cell death, we first examined rates of DNA to normal at 48 hr post irradiation, it remained sup- repair. In fact, c-junϪ/Ϫ cells were somewhat more effi- pressed for at least 60 hr in c-junϪ/Ϫ cells. In other experi- cient than wt cells in repair of UV-induced lesions, and ments, we found that DNA synthesis in c-junϪ/Ϫ cells both cells types have repaired most of the damage remained suppressed for at least 72 hr post UV exposure caused by 20 J/m2 UVC by 16 hr post irradiation (A. (data not shown). By contrast, c-Jun reexpressing cells Haghigi, R. Gjerset, and E. S., unpublished results). We (C4 or C6) did not exhibit significant inhibition of DNA therefore examined the effect of UV on the proliferative synthesis after UV irradiation (Figure 3C). Similar results capacity of the different cells. Cells were exposed to were obtained by cell-cycle analysis (Figure 3D). In re- increasing doses of UVC and their ability to incorporate sponse to UV irradiation, c-junϪ/Ϫ cells arrested mostly bromodeoxyuridine (BrdU) into DNA at 24 hr post irradia- at the and remained so for at least 3 days. By tion was measured. As expected, UV exposure resulted contrast, c-junϩ/ϩ cells started to re-enter the cycle by in dose-dependent inhibition of DNA synthesis, but the 48 hr and by 3 days exhibited similar cell-cycle distribu- inhibitory effect was much greater in c-junϪ/Ϫ cells (Fig- tion to nonirradiated cells. The constitutive c-Jun ex- ure 3A). By contrast, exposure to IR did not inhibit DNA pressors, C4 and C6, had returned to normal cell-cycle Cell 900

Figure 2. c-Jun Is Required for p53-Dependent UV-Induced Cell Death The indicated cell lines were exposed to 20 J/m2 of UVC and the extent of annexin V staining or caspase 3 activation was determined at 24 hr post irradiation. Annexin V staining was used to determine the rate of cell death in panels (A) and (B) and caspase 3 activity was measured in panel (C). The data are presented as (A) percentage of cell death before (solid bars) and after (open bars) UV exposure, or (B and C) raw flow cytometry results in which the gray and the black bordered areas represent the staining intensity of untreated and UV-irradiated cells, respectively. Fold-increase in annexin V staining or caspase 3 activity after UV irradiation is indicated. To measure caspase 3 activity, the cells were incubated with a fluorogenic caspase 3 substrate. distribution even earlier. Therefore, the differences in the different cells at 8 hr post irradiation (Figure 4B). As the clonogenic survival of the different cell lines are most predicted by the analysis of p21 levels (Figure 3E), basal likely due to their inherently different responses to UV and induced levels of p21 mRNA were higher in c-junϪ/Ϫ radiation. While in wt cells DNA synthesis is transiently cells than in c-junϩ/ϩ cells. Surprisingly, little or no induc- inhibited after UV exposure, c-junϪ/Ϫ cells undergo pro- tion of p21 mRNA was found in the C4 and C6 constitu- longed UV-induced growth arrest and this arrest is abro- tive c-Jun expressors (Figure 4B). Analysis of p21 mRNA gated in c-Jun reexpressing cells. levels at 24 hr post UV exposure yielded similiar results The p21 CDK inhibitor plays a major role in cellular (data not shown). Similar differences were observed for responses to DNA damage and its p53-induced expres- two other p53 target genes, Bax and Mdm2 (Figure 4B). sion is critical for inhibition of DNA synthesis (Brugarolas Transcriptional run-off experiments confirmed that c-Jun et al., 1995). We therefore compared the kinetics of p21 is a potent negative regulator of p21 transcription (data induction and expression in the different cell lines. c-Jun not shown). null cells expressed higher levels of p21 than wt cells p53 are quite common and were suggested after UV irradiation (Figure 3E). Furthermore, while p21 to contribute to immortalization of many rodent cell lines expression declined after 24 hr in wt cells, it remained (Harvey and Levine, 1991). However, the c-junϪ/Ϫ as well unchanged for at least 48 hr in c-junϪ/Ϫ cells. These as the C4 and C6 cells do not contain mutant p53 alleles differences in the kinetics and levels of p21 expression (Schreiber et al., 1999). Furthermore, and quite surpris- between c-junϩ/ϩ and c-junϪ/Ϫ cells were highly repro- ingly, all of the cell lines including C4 and C6 exhibited ducible and the decline in p21 expression in UV-irradi- normal induction of p21 mRNA in response to IR (Figure ated wt cells correlated with accumulation of c-Jun, 4C). This induction was observed after exposure to ei- which peaked at 24 hr post irradiation. Extremely weak ther high (18 Gray) or low (8 Gray) doses of IR. Normal and delayed p21 induction was found in c-Jun reex- induction of Mdm2 was also observed in response to pressing cells. treatment of C4 and C6 cells with MMC (data not shown), which like IR is a poor JNK and c-Jun activator in mouse c-Jun Inhibits p53 Transcriptional Activity fibroblasts (Liu et al., 1996). As p53 is the major regulator of p21 expression, we compared p53 accumulation in UV-irradiated cells of the c-Jun Regulates the Association of p53 various genotypes. As previously reported (Schreiber et with the p21 Promoter al., 1999), the basal level of p53 was higher in c-junϪ/Ϫ We attempted to examine the effect of c-Jun on p53 cells but its UV-induced level was quite similar in all transcriptional activity by transient transfection assays, cell types (Figure 4A). Nevertheless, striking differences using c-junϪ/Ϫp53Ϫ/Ϫ mouse fibroblasts and a luciferase were observed at the level of p21 mRNA expressed by reporter driven by the p21 promoter. However, no signifi- c-Jun Mediates Cell-Cycle Exit by Antagonizing p53 901

Figure 3. c-Jun Is Required for Resumption of DNA Synthesis after UV Exposure The indicated MEF cultures (2 c-junϩ/ϩ and2c-junϪ/Ϫ) were exposed to UVC (A) or IR (B). After 24 hr, the cells were labeled for a 12 hr period with BrdU to determine rates of DNA synthesis. The rate of DNA synthesis by untreated cultures was given an arbitrary value of 100% and all other values are depicted relative to this. (C) The indicated cell lines were exposed to 16 J/m2 of UVC and the rates of DNA synthesis were determined as above by labeling for 12 hr with BrdU at 24 and 48 hr post exposure. The 0 time point indicates the basal rates of DNA synthesis in nonirradiated cells. (D) c-junϪ/Ϫ, c-junϩ/ϩ, C4, and C6 cells were exposed to 16 J/m2 of UVC and their cell-cycle profiles were determined by propidium iodide staining and FACS analysis at the indicated time points. Only surviving cells that remained attached to the plate were analyzed. (E) Cells of the indicated genotypes were exposed to 12 J/m2 of UVC and the levels of p21, c-Jun, and expression at the indicated time points (hr post irradiation) were determined by immunoblotting. Cell 902

Figure 4. c-Jun Is a Negative Regulator of p53-Mediated Transacti- vation in UV-Irradiated Cells Figure 5. c-Jun Is a Negative Modulator of p53 Association with the p21 Promoter in UV-Irradiated Cells Cells of the indicated genotypes were exposed or not to UVC (12 J/m2) or IR (18 Grays). (A) Cells of the indicated genotypes were exposed to IR (18 Grays) 2 (A) p53 expression by nonirradiated and irradiated cells at 6 hr or UVC (12 J/m ) and p53 protein levels and association with the post exposure was examined by immunoblotting. Actin serves as p21 promoter were determined by immunoblotting (left panel) or a a loading control. ChIP assay (right panel). The ethidium bromide stained bands (right (B) Expression of p21, Mdm2, and Bax mRNAs was determined at panel) represent relative amounts of p21 promoter DNA recovered 8 hr post exposure by Northern blot hybridization. GAPDH mRNA by PCR amplification of DNA extracted from p53 immunoprecipi- serves as a loading control. tates. 2 (C) Expression of p21 mRNA before and 5 hr after exposure to IR (B) Cells of the indicated genotype were exposed to UVC (12 J/m ) was determined as above. and the amount of p53-immunoprecipitated p21 promoter DNA was determined at the indicated time points after UV exposure (upper panel). p53 levels at the same time points are shown at the bottom panel. cant effect of c-Jun on p53-mediated transactivation (C) Total p53 DNA binding activity in the indicated cell cultures was was detected (data not shown). As regulation of tran- determined before and 6 hr after UV exposure (12 J/m2) by a mobility scription from a chromosomal DNA template can differ shift assay using an oligonucleotide probe corresponding to a p53 binding site from the p21 promoter. Binding specificity is demon- from that of a transiently transfected template (Alberts strated by competition with nonlabeled wt and mutant oligonucleo- et al., 1998), we investigated the regulation of the endog- tides and antibody supershifting. Loading and extract quality were enous p21 gene by examining the interaction of p53 controlled by measuring NF-l DNA binding activity. with its promoter using the ChIP assay (Braunstein et al., 1993). Cross-linked p53-DNA complexes were im- were slightly reduced (Figure 5A). As previously de- munoprecipitated by an anti-p53 antibody and the - scribed (Lu and Lane, 1993), UV irradiation resulted in tive amount of precipitated p21 promoter DNA was de- prolonged and more substantial p53 induction than the termined by PCR amplification with specific primers. one caused by IR. In wt cells exposed to IR, the amount Control experiments indicated that the p53-specific an- of p53 bound p21 promoter DNA paralleled the total tibody pAb421 specifically detected the association of level of p53 (Figure 5A). While UV irradiation also induced p53 with the p21 promoter, giving rise to a very low level the association of p53 with the p21 promoter, in this case of nonspecific signal in nonirradiated cells and no signal promoter occupancy did not parallel total p53 levels and in p53Ϫ/Ϫ cells (data not shown). This allowed the reliable at 6 hr post irradiation only a small amount of p53 was comparison of p53 binding to the p21 promoter to the associated with the p21 promoter (Figure 5A). By con- kinetics of p53 accumulation. In wt cells, high p53 levels trast to the transient interaction of p53 with the p21 were detected 2.5 hr after IR treatment, and after 5 hr promoter in UV-irradiated wt cells, the level of promoter c-Jun Mediates Cell-Cycle Exit by Antagonizing p53 903

Figure 6. c-Jun Is No Longer Required for Cell-Cycle Reentry in UV-Irradiated p53-Deficient Cells (A) Clonogenic survival of c-junϩ/ϩp53Ϫ/Ϫ and c-junϪ/Ϫp53Ϫ/Ϫ cells exposed to the indicated doses of UVC was determined as described in Figure 1. (B) DNA synthesis of the indicated cells was determined by BrdU incorporation at different times before and after exposure to 16 J/m2 of UVC, as described in Figure 3. (C) Cell-cycle distribution of the indicated cell cultures before and after exposure to 16 J/m2 of UVC was determined by BrdU and propidium iodide staining and analyzed by flow cytometry. Lower left quadrant: cells in G1. Lower right quadrant: cells in G2. Upper quadrants: cells in S phase. (D and E) The indicated cell cultures were exposed to UVC (12 J/m2) and the levels of c-Jun, p21, and actin were determined at the indicated time points (hr) (D) or 24 hr (E) after UV exposure by immunoblotting. bound p53 in c-junϪ/Ϫ cells remained unchanged for at modifications that affect bulk DNA binding activity, were least 6 hr post UV exposure (Figure 5A). Even at 24 hr unaltered in C4 and C6 cells (data not shown). post UV irradiation, c-junϪ/Ϫ cells exhibited higher levels of p21 promoter bound p53 then c-junϩ/ϩ cells, despite p53 Deficiency Abrogates the Need for c-Jun similar levels of p53 expression (Figure 5B). These re- for Cell-Cycle Reentry sults are consistent with the prolonged induction of p21 To test whether the major function of c-Jun during the in c-junϪ/Ϫ cells (Figure 3D). Consistent with the defective UV response is to allow cells to exit p53-imposed growth induction of p21 mRNA (Figure 4B), only a very small arrest, we compared clonogenic survival and the ability amount of p21 promoter DNA was associated with p53 of c-junϩ/ϩp53Ϫ/Ϫ and c-junϪ/Ϫp53Ϫ/Ϫ fibroblasts to re- in the constitutive c-Jun expressors after UV exposure enter the cell cycle following UV-induced growth arrest. (Figure 5A). However, IR induced binding of p53 to the Unlike c-junϪ/Ϫp53ϩ/ϩ cells, which were very sensitive p21 promoter in both C4 and C6 cells. As an additional to UV-induced replicative death, all of the other cells, control for specificity of the ChIP assay, we examined including c-junϪ/Ϫp53Ϫ/Ϫ cells, formed colonies even after the interaction of c-Jun with the c-jun promoter. We exposure to 40 J/m2 of UVC (Figure 6A). In addition, both detected induction of c-Jun binding to its own promoter c-junϩ/ϩp53Ϫ/Ϫ and c-junϪ/Ϫp53Ϫ/Ϫ cell lines underwent following UV exposure in all cell types except c-junϪ/Ϫ partial inhibition of DNA synthesis after UV exposure cells (data not shown). (Figure 6B). This arrest may be due to a p53-independent Despite the striking differences in p21 promoter occu- growth arrest (Loignon et al., 1997). Both cell lines re- pancy, normal induction of p53 DNA binding activity sumed DNA synthesis after 48 hr, reaching the level was observed in vitro in both C4 and C6 cells after found in nonirradiated cells by 72 hr. Cell-cycle analysis exposure to UV irradiation (Figure 5C). Thus, the bulk demonstrated that the cells were arrested mainly at the of p53 in C4 or C6 cells is not covalently modified in a G2 phase following UV exposure (Figure 6C). In correla- way that compromises its DNA binding activity. Indeed, tion with the BrdU incorporation results, p53Ϫ/Ϫ c-junϪ/Ϫ both phosphorylation on serine 389 and p53 , cells resume normal cell-cycle distribution faster than Cell 904

This response, which resembles the response to mito- gens (Karin and Herrlich, 1989; Holbrook and Fornace, 1991; Herrlich et al., 1992; Angel, 1995), had seemed enigmatic and paradoxical for many years. After all, like other DNA damaging agents, UVC inhibits, rather than stimulates, cell proliferation. Using mouse fibroblast cul- tures that differ in their c-Jun levels, we elucidated the physiological function of c-jun induction. c-Jun expres- sion is essential for the ability of UV-irradiated cells to exit p53-imposed growth arrest and re-enter the cell cycle. While c-jun null cells undergo a prolonged growth arrest and assume a senescent-like phenotype in re- sponse to UV radiation resulting in poor clonogenic po- tential, cells that constitutively reexpress c-Jun fail to undergo UV-induced cell-cycle arrest and exhibit an in- creased clonogenic potential. The two cell types also differ in their ability to undergo UV-induced death: while c-junϪ/Ϫ cells are resistant to UV-induced apoptosis, c-Jun reexpressing cells show an increased apoptotic response. Remarkably, the c-Jun expression status only affects the outcome of UV exposure and has no bearing on clonogenic survival after exposure to IR. Although both UV and IR lead to p53 and p21 induction, they do so via different pathways. The present results show that even the termination of p21 induction in response to the two types of radiation occurs via different mechanisms, and that c-Jun is required only for terminating p21 induc- tion and promotion of cell-cycle reentry of UV-irradiated cells. As prolonged p21 expression can result in senes- cence (Brown et al., 1997; Fang et al., 1999), a common occurrence in fibroblasts exposed to IR (Nagasawa and Little, 1983), the proper termination of p21 induction is essential for an efficient protection of cells from the Figure 7. Regulation of p53 Function by c-Jun consequences of UV exposure. Indeed, c-junϪ/Ϫ cells (A) A scheme summarizing our major findings. UV irradiation results assume a senescent-like phenotype after exposure to in JNK activation and c-Jun induction, as well as stabilization and UVC (data not shown). induction of p53. Upon reaching a certain threshold, c-Jun inhibits the transcriptional activity of p53, thereby preventing further p21 As summarized in Figure 7, the function of c-Jun in induction and allowing cells to exit growth arrest. c-Jun, however, stimulating cell-cycle reentry of UV-irradiated cells is has no effect on the proapoptotic activity of p53. mediated via negative regulation of p21 induction by (B) In c-junϪ/Ϫ cells, the dominant p53 activated pathway is the one p53. The extended cell-cycle arrest of UV-irradiated leading to p21 induction and growth arrest. This pathway protects c-junϪ/Ϫ cells correlates with prolonged p21 induction, cells from apoptosis. In cells that express c-Jun constitutively (con- whereas in wt cells p21 induction is transient, in concert stitutive c-Jun), p21 induction is blocked and the predominant p53 with the ability of wt cells to re-enter the cell cycle after activated pathway is the one leading to apoptosis. UV exposure. Furthermore, UV-induced cell-cycle arrest and p21 induction are abrogated in cells that express c-Jun constitutively. Interestingly, this defect is observed p53Ϫ/Ϫ c-junϩ/ϩ cells. Furthermore, although p53-defi- only in UV-irradiated cells and c-Jun reexpressing cells cient cells express much lower levels of p21 than p53ϩ/ϩ exhibit normal p21 induction in response to IR or MMC. cells, the p53-independent induction of p21 was not Importantly, c-Jun expression levels have no bearing on affected at all by c-Jun (Figures 6D and 6E). the response to UV radiation (either DNA synthesis or These results further support the conclusion that the p21 expression) in p53 null cells. In other words, the major function of UV-induced c-Jun expression is to effects of c-Jun on UV-induced growth arrest or cell negatively regulate p53’s transcriptional activity in UV- death are largely dependent on expression of p53, and irradiated cells, leading to the termination of p21 induc- therefore, based on genetic considerations, the target tion and cell-cycle reentry (Figure 7). In the absence of for c-Jun in this pathway is p53. p53, c-Jun is no longer required to allow cell-cycle reen- Although c-Jun is suggested to affect basal levels of try of UV-irradiated cells and does not have an effect p53 mRNA (Schreiber et al., 1999), normal accumulation on p21 expression. of p53 protein in response to UV radiation or other geno- toxic challenges was observed in all the cell lines used Discussion in this study. Thus, the induced levels of p53 in UV- irradiated cells are similar in c-jun null, wt, and c-Jun A hallmark of the mammalian UV response is a rapid reexpressing cells. Nevertheless these cells vary greatly induction of immediate-early protooncogenes, such as in their ability to induce three different p53 target genes, c-jun and c-fos (Buscher et al., 1988; Devary et al., 1991). p21, Bax, and Mdm2, in response to UV radiation. c-Jun Mediates Cell-Cycle Exit by Antagonizing p53 905

Repression of p53-mediated transactivation by c-Jun mm in diameter) was scored 1–2 weeks after irradiation according is not due to a covalent modification that interferes with to the cells’ growth rates. Each experiment was repeated at least the DNA binding activity of bulk p53. Nevertheless, ChIP three times. experiments revealed a clear negative effect of c-Jun Cell Death Assays on association of p53 with the p21 promoter, paralleling 2·105 cells per 6 cm dish were plated 24 hr before irradiation. its effect on p53-dependent p21 transcription. In UV- Determination of death rate was performed 24 hr after irradiation irradiated wt cells, the interaction of p53 with the p21 by staining the cells with annexin V FITC conjugated antibodies promoter was transient. However, this interaction was (Pharmingen), according to manufacturer’s instructions or by incu- significantly prolonged in c-junϪ/Ϫ cells. Conversely, little bating the cells with caspase 3 flurogenic substrate PhiPhilux-G1D2 (Alexis Biochemicals) for 1 hr. The levels of FITC and PhiPhilux- or no p53 bound to the p21 promoter in UV-irradiated G1D2 fluorescence staining were assessed by flow cytometry (FACS- c-Jun reexpressing cells. Scan, Becton Dickinson). Both adherent and floating cells were The ability of c-Jun to inhibit p21 induction also ex- collected for analysis. plains why overexpression of c-Jun promotes the death of UV-irradiated cells and why c-jun null cells are resis- Proliferation Assay tant to UV-induced death. Abrogation of UV-induced Cells were plated on glass cover slides at a density of 1 · 105 cells growth arrest would allow cells with damaged DNA to per 3.5 cm dish and were UV-irradiated 24 hr later. BrdU was added at a final concentration of 10 ␮M for 12 hr periods at the indicated progress via the cell cycle and thus trigger p53-mediated times. Labeled cells were fixed with methanol, washed twice with apoptosis or a mitotic catastrophe. Indeed, p21-defi- PBS, incubated for 5 min with 4N HCl, washed twice with PBS, cient cells are more susceptible to DNA damage induced and blocked with 1% BSA in PBS. Cells were stained with FITC- apoptosis than p21-expressing cells (Waldman et al., conjugated anti BrdU-antibody and counterstained with DAPI for 1996). On the other hand, p53Ϫ/Ϫ cells are resistant to 30 min at room temp. and washed intensively. Cells presenting the proapoptotic effect of c-Jun. It should also be noted normal nuclear DAPI staining were counted first and were scored for BrdU staining. At least 100 cells were counted for each point. that some of the proapoptotic activity of p53 does not The fraction of BrdU labeled cells in the nontreated control popula- require gene induction (Caelles et al., 1994; Haupt et al., tion was given an arbitrary value of 100%. Each experiment was 1995) and therefore should not be opposed by c-Jun performed in duplicates. (Figure 7). Although c-Jun expression was also sug- gested to promote apoptosis via Fas induction Chromatin Immunoprecipitation Assay (ChIP) (Kolbus et al., 2000), the abrogation of the proapoptotic 106 cells were plated per 10 cm dish and were irradiated or not 24 effect of c-Jun in p53Ϫ/Ϫ cells suggests that, in UV-irradi- hr later. DNA and were crosslinked by addition of formalde- hyde (1% final concentration) 10 min before harvesting. Cells were ated mouse fibroblasts, this effect of c-Jun is mostly scraped off the plate, resuspended in hypotonic buffer, and passed Ϫ/Ϫ p53-dependent. We also find that, like c-jun fibro- 20 times through a 26 gauge needle. Nuclei were spun down, resus- blasts, JNK-deficient fibroblasts also overexpress p21 pended in 100 ␮l SDS lysis buffer (1% SDS, 10 mM EDTA, 50 mM and are resistant to UV-induced apoptosis (E. S., unpub- Tris-HCl, pH 8.1 and a protease inhibitor cocktail), and sonicated lished results). The proapoptotic effect of JNK in UV- to generate 500–2000 bp DNA fragments. After centrifugation, the irradiated mouse fibroblasts may also be mediated cleared supernatant was diluted 10-fold with immunoprecipitation buffer (50 mM Tris-HCl, pH 8, 150 mM NaCl, 5 mM EDTA, 0.5% through induction of c-Jun and repression of p53-medi- NP40). The cell lysate was precleared by incubation at 4ЊC for 45 ated p21 induction. min with protein G beads preabsorbed with sonicated single- Our findings also explain why exposure of mouse fi- stranded DNA. The cleared lysates were incubated for an additional broblasts or human normal diploid fibroblasts to high 2 hr with pAb421. Immune complexes were precipitated with protein doses of UV results in considerable p53 induction with- G beads preabsorbed with sonicated single-stranded DNA. After out concomitant p21 induction (Lu et al., 1996). Elevated centrifugation, the beads were washed and the antigen was eluted as described (Braunstein et al., 1993). DNA-protein cross-links were c-Jun was also shown to repress transcription of the reversed by heating at 65ЊC for 4–5 hr and DNA was phenol extracted tromobospondin 1 gene (Mettouchi et al., 1994), which and ethanol precipitated. Levels of p21 promoter DNA were deter- is activated by p53 (Dameron et al., 1994). Thus, negative mined by PCR using oligonucleotides spanning the p53 binding site: regulation of p53 target genes by c-Jun may be a rather 5Ј-GAGGATACCTTGCAAGGCTGCA-3Ј and 5Ј-GCACACCATTGCA prevalent but previously underappreciated phenome- CGTGAATGT-3. An internal control of nonstimulated cells was in- Ϫ/Ϫ non. This pathway may even be used by the E5 onco- cluded in every experiment and extracts from p53 cells were also tested as negative controls. protein of human papilloma , which is known both to induce c-jun (probably via its effects on EGF In Vitro DNA Binding Analysis signaling) and repress p21 expression (Tsao et Cell extracts for DNA binding analysis were prepared as described al., 1996). (Woo et al., 1998). 15 ␮g of protein extracts was incubated with p21-derived oligonucleutide probe in binding buffer composed of: Experimental Procedures 12.5 mM HEPES pH 7.6, 0.3 mM DTT, 3.3% glycerol, 6.6 ng/␮l salmon sperm DNA, 30 ng/␮l BSA (final concentrations), for 30 min Cell Culture and Antibodies at room temp. The reactions were than analyzed on 4% acrylamide Cells used in this study were grown in DMEM ϩ 10% fetal calf gel. Binding specificity was determined by appearance of shifted serum, and maintained at 37ЊC under 5% CO2. Rabbit polyclonal band only after incubation with pAb421 and also by competition antibodies (all from Santa Cruz Antibodies) were used to detect experiments with wt and mutated oligonucleotides. p53, p21, and c-Jun by immunoblotting. pAb421 mouse monoclonal antibody ( Research Products) was used to immunopre- Acknowledgments cipitate p53. We thank A. Haghigi and R. A. Gjerset for measurement of DNA Clonogenic Survival Assays repair rates; J. Aguilera for help with IR; M. Kapoor and G. Lozano 500 cells were plated per 6 cm plate and irradiated 24 hr later. for the anti-phospho 389 antibody; and T. Hunter and G. Wahl for Medium was changed every 3 days. The number of colonies (1–3 critical reading of the manuscript. Research was supported by Cell 906

grants from the NIH (CA76188), the DOE (DE-FG03-86ER60429), and (1993). WAF1, a potential mediator of p53 tumor suppression. Cell the State of California Research Program (99-00529V-10249) 75, 817–825. to M. K. M. K. is the Frank and Else Schilling-American Cancer Elledge, S.J. (1996). Cell cycle checkpoints–preventing an identity Society Research Professor. crisis. Science 274, 1664–1672. Fang, L., Igarashi, M., Leung, J., Sugrue, M.M., Lee, S.W., and Aaron- Received March 27, 2000; revised October 25, 2000. son, S.A. (1999). p21Waf1/Cip1/Sdi1 induces permanent growth ar- rest with markers of replicative senescence in human tumor cells References lacking functional p53. Oncogene 18, 2789–2797. Finkel, E. (1999). Biomedicine–does cancer therapy trigger cell sui- Alberts, A.S., Geneste, O., and Treisman, R. (1998). Activation of cide? Science 286, 2256–2258. SRF-regulated chromosomal templates by Rho-family GTPases re- Fisher, G.J., Datta, S.C., Talwar, H.S., Wang, Z.Q., Varani, J., Kang, quires a signal that also induces H4 hyperacetylation. Cell 92, S., and Voorhees, J.J. (1996). Molecular basis of sun-induced prema- 475–487. ture skin aging and retinoid antagonism. Nature 379, 335–339. Angel, P. (1995). The role and regulation of the Jun proteins in Friedberg, E.C. (1995). DNA Repair and Mutagenesis, E.C. Friedberg, response to phorbol ester and UV light. In Inducible Gene Expression G.C. Walker, and W. Siede, eds. (Washington, D.C.: ASM Press). I, P.A. Baeuerle, ed. (Basel, Switzerland: Birkhauser Verlag), pp. 62–92. Gupta, S., Campbell, D., De´ rijard, B., and Davis, R.J. (1995). Tran- scription factor ATF2: regulation by the JNK Banin, S., Moyal, L., Shieh, S.Y., Taya, Y., Anderson, C.W., Chessa, pathway. Science 267, 389–393. L., Smorodinsky, N.I., Prives, C., Reiss, Y., Shiloh, Y., and Ziv, Y. Hartwell, L.H., and Kastan, M.B. (1994). Cell cycle control and can- (1998). Enhanced phosphorylation of p53 by ATM in response to cer. Science 266, 1821–1828. DNA damage. Science 281, 1674–1677. Harvey, D.M., and Levine, A.J. (1991). p53 alteration is a common Bissonnette, N., and Hunting, D.J. (1998). p21-induced cycle arrest event in the spontaneous immortalization of primary BALB/c murine in G(1) protects cells from apoptosis induced by UV-irradiation or embryo fibroblasts. Genes Dev. 5, 2375–2385. RNA polymerase II blockage. Oncogene 16, 3461–3469. Haupt, Y., Rowan, S., Shaulian, E., Vousden, K.H., and Oren, M. Braunstein, M., Rose, A.B., Holmes, S.G., Allis, C.D., and Broach, (1995). Induction of apoptosis in Hela cells by trans-activation-defi- J.R. (1993). Transcriptional silencing in yeast is associated with cient p53. Genes Dev. 9, 2170–2183. reduced nucleosome acetylation. Genes Dev. 7, 592–604. Herr, I., Vandam, H., and Angel, P. (1994). Binding of promoter- Brown, J.P., Wei, W., and Sedivy, J.M. (1997). Bypass of senescence associated AP-1 is not altered during induction and subsequent after disruption of p21CIP1/WAF1 gene in normal diploid human repression of the c-Jun promoter by TPA and UVC-irradiation. Carci- fibroblasts. Science 277, 831–834. nogenesis 15, 1105–1113. Brugarolas, J., Chandrasekaran, C., Gordon, J.I., Beach, D., Jacks, Herrlich, P., Ponta, H., and Rahmsdorf, H.J. (1992). DNA damage- T., and Hannon, G.J. (1995). Radiation-induced cell cycle arrest induced gene expression: signal transduction and relation to growth compromised by p21 deficiency. Nature 377, 552–557. factor signaling. Rev. Physiol. Biochem. Pharmacol. 119, 187–223. Buscher, M., Rahmsdorf, H.J., Litfin, M., Karin, M., and Herrlich, P. Herrlich, P., Blattner, C., Knebel, A., Bender, K., and Rahmsdorf, (1988). Activation of the c-fos gene by UV and phorbol ester: different H.J. (1997). Nuclear and non-nuclear targets of genotoxic agents in signal transduction pathways converge to the same enhancer ele- the induction of gene expression. Shared principles in yeast, ro- ment. Oncogene 3, 301–311. dents, man and plants. Biol. Chem. 378, 1217–1229. Caelles, C., Helmberg, A., and Karin, M. (1994). p53-dependent apo- Hibi, M., Lin, A., Smeal, T., Minden, A., and Karin, M. (1993). Identifi- ptosis in the absence of transcriptional activation of p53-target cation of an oncoprotein-responsive and UV-responsive protein ki- genes. Nature 370, 220–223. nase that binds and potentiates the c-Jun activation domain. Genes Canman, C.E., Lim, D.S., Cimprich, K.A., Taya, Y., Tamai, K., Saka- Dev. 7, 2135–2148. guchi, K., Appella, E., Kastan, M.B., and Siliciano, J.D. (1998). Activa- Holbrook, N.J., and Fornace, A.Y., Jr. (1991). Responses to adver- tion of the ATM kinase by and phosphorylation of sity: molecular control of gene activation following genotoxic stress. p53. Science 281, 1677–1679. New Biology 3, 825–833. Dameron, K.M., Volpert, O.V., Tainsky, M.A., and Bouck, N. (1994). Johnson, R.S., van Lingen, B., Papaioannou, V.E., and Spiegelman, Control of in fibroblasts by p53 regulation of throm- B.W. (1993). A null at the c-jun causes embryonic bospondin-1. Science 265, 1582–1584. lethality and retarded cell growth in culture. Genes Dev. 7, 1309– De´ rijard, B., Hibi, M., Wu, I.-H., Barrett, T., Su, B., Deng, T., Karin, 1317. M., and Davis, R.J. (1994). JNK1: a protein kinase stimulated by UV Karin, M. (1995). The regulation of AP-1 activity by mitogen-activated light and Ha-Ras that binds and phosphorylates the c-Jun activation protein kinases. J. Biol. Chem. 270, 16483–16486. domain. Cell 76, 1025–1037. Karin, M., and Herrlich, P. (1989). Cis and trans-acting genetic ele- Devary, Y., Gottlieb, R.A., Lau, L., and Karin, M. (1991). Rapid and ments responsible for induction of specific genes by tumor promot- preferential activation of the c-jun gene during the mammalian UV ers, serum factors and stress. In Genes and Signal Transduction response. Mol. Cell. Biol. 11, 2804–2811. in Multistage . (New York: Marcel Dekker Inc.), pp. Devary, Y., Gottlieb, R., Smeal, T., and Karin, M. (1992). The mamma- 415–440. lian ultraviolet response is triggered by activation of Src tyrosine Kastan, M.B., Onyekwere, O., Sidransky, D., Vogelstein, B., and kinases. Cell 71, 1081–1091. Craig, R.W. (1991). Participation of p53 protein in the cellular re- Devary, Y., Rosette, C., DiDonato, J.A., and Karin, M. (1993). NF-kB sponse to DNA damage. Cancer Res. 51, 6304–6311. activation by ultraviolet light not dependent on a nuclear signal. Kastan, M.D., Zhan, Q., El-Deiry, W.S., Carrier, F., Jacks, T., Walsh, Science 261, 1442–1445. W.V., Plunkett, B.S., Vogelstein, B., and Fornace, A.J., Jr. (1992). A Di Leonardo, A., Linke, S.P., Clarkin, K., and Wahl, G.M. (1994). DNA mammalian pathway utilizing p53 and GADD45 damage triggers a prolonged p53-dependent G1 arrest and long- is defective in Ataxia-Telangiectasia. Cell 71, 587–597. term induction of Cip1 in normal human fibroblasts. Genes Dev. 8, Kaufmann, W.K., and Wilson, S.J. (1994). G(1) arrest and cell-cycle- 2540–2551. dependent clastogenesis in UV-irradiated human fibroblasts. Mutat. El-Deiry, W.S. (1998). Regulation of p53 downstream genes. Semin Res. 314, 67–76. Cancer Biol. 8, 345–357. Kern, S.E., Kinzler, K.W., Bruskin, A., Jarosz, D., Friedman, P., Prives, El-Deiry, W.S., Tokino, T., Velculescu, V.E., Levy, D.B., Parsons, R., C., and Vogelstein, B. (1991). Identification of p53 as a sequence- Trent, J.M., Lin, D., Mercer, W.E., Kinzler, K.W., and Vogelstein, B. specific DNA-binding protein. Science 252, 1708–1711. c-Jun Mediates Cell-Cycle Exit by Antagonizing p53 907

Ko, L.J., and Prives, C. (1996). p53: puzzle and paradigm. Genes of phosphatidylserine expression on early apoptotic cells using fluo- Dev. 10, 1054–1072. rescein labelled Annexin V. J. Immunol. Meth. 184, 39–51. Kolbus, A., Herr, I., Schreiber, M., Debatin, K.M., Wagner, E.F., and Waldman, T., Lengauer, C., Kinzler, K.W., and Vogelstein, B. (1996). Angel, P. (2000). c-Jun-dependent CD95-L expression is a rate- Uncoupling of S phase and induced by anticancer agents limiting step in the induction of apoptosis by alkylating agents. Mol. in cells lacking p21. Nature 381, 713–716. Cell. Biol. 20, 575–582. Woo, R.A., McLure, K.G., LeesMiller, S.P., Rancourt, D.E., and Lee, Kovary, K., and Bravo, R. (1991). The jun and fos protein families P.W.K. (1998). DNA-dependent protein kinase acts upstream of p53 are both required for cell cycle progression in fibroblasts. Mol. Cell. in response to DNA damage. Nature 394, 700–704. Biol. 11, 4466–4472. Zhan, Q.M., Carrier, F., and Fornace, A.J. (1993). Induction of cellular Lane, D.P. (1992). Cancer. p53, guardian of the . Nature 358, p53 activity by DNA-damaging agents and growth arrest. Mol. Cell. 15–16. Biol. 13, 4242–4250. Levine, A.J. (1997). p53, the cellular gatekeeper for growth and divi- sion. Cell 88, 323–331. Liu, Z.-G., Baskaran, R., Lea-Chou, E.T., Wood, L.D., Chen, Y., Karin, M., and Wang, J.Y.J. (1996). Three distinct signalling responses by murine fibroblasts to genotoxic stress. Nature 384, 273–276. Loignon, M., Fetni, R., Gordon, A.J.E., and Drobetsky, E.A. (1997). A p53-independent pathway for induction of p21(waf1cip1) and con- comitant G(1) arrest in UV-irradiated human skin fibroblasts. Cancer Res. 57, 3390–3394. Lu, X., and Lane, D.P. (1993). Differential induction of transcription- ally active p53 following UV or ionizing radiation: defects in chromo- some instability syndromes? Cell 75, 765–778. Lu, X., Burbidge, S.A., Griffin, S., and Smith, H.M. (1996). Dis- cordance between accumulated p53 protein level and its transcrip- tional activity in response to UV radiation. Oncogene 13, 413–418. Maltzman, W., and Czyzyk, L. (1984). UV irradiation stimulates levels of p53 cellular in nontransformed mouse cells. Mol. Cell. Biol. 4, 1689–1694. Mettouchi, A., Cabon, F., Montreau, N., Vernier, P., Mercier, G., Blangy, D., Tricoire, H., Vigier, P., and Binetruy, B. (1994). Sparc and thrombospondin genes are repressed by the c-Jun oncogene in rat embryo fibroblasts. EMBO J. 13, 5668–5678. Nagasawa, H., and Little, J.B. (1983). Comparison of kinetics of X-ray-induced cell killing in normal, ataxia telangiectasia and heredi- tary retinoblastoma fibroblasts. Mutat. Res. 109, 297–308. Niculescu, A., Chen, X., Smeets, M., Hengst, L., Prives, C., and Reed, S. (1998). 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, 629–643. Oren, M. (1999). Regulation of the p53 tumor suppressor protein. J. Biol. Chem. 274, 36031–36034. Rosette, C., and Karin, M. (1996). Ultraviolet light and osmotic stress: activation of the JNK cascade through multiple and cytokine receptors. Science 274, 1194–1197. Sachsenmaier, C., Radler-Pohl, A., Zinck, R., Nordheim, A., Herrlich, P., and Rahmsdorf, H.J. (1994). Involvement of growth factor recep- tors in the mammalian UVC response. Cell 78, 963–972. Schreiber, M., Kolbus, A., Piu, F., Szabowski, A., Mo¨ hle-Steinlein, U., Tian, J.M., Karin, M., Angel, P., and Wagner, E.F. (1999). Control of cell cycle progression by c-Jun is p53 dependent. Genes Dev. 13, 607–619. Shaulian, E., and Karin, M. (1999). Stress-induced JNK activation is independent of Gadd45 induction. J. Biol. Chem. 274, 29595–29598. Sheikh, M.S., Chen, Y.Q., Smith, M.L., and Fornace, A.J. (1997). Role of p21(Waf1/Cip1/Sd1) in cell death and DNA repair as studied using a tetracycline-inducible system in p53-deficient cells. Oncogene 14, 1875–1882. Tsao, Y.P., Li, L.Y., Tsai, T.C., and Chen, S.L. (1996). Human papillo- mavirus type 11 and 16 E5 represses P(21wafi/Sdii/Cipi) gene ex- pression in fibroblasts and keratinocytes. J. Virol. 70, 7535–7539. van Dam, H., Duyndam, M., Rottier, R., Bosch, A., de Vries-Smits, L., Herrlich, P., Zantema, A., Angel, P., and van der Eb, A.J. (1993). Heterodimer formation of cJun and ATF-2 is responsible for induc- tion of c-jun by the 243 adenovirus E1A protein. EMBO J. 12, 479–487. Vermes, I., Haanen, C., Steffens-Nakken, H., and Reutelingsperger, C. (1995). A novel assay for apoptosis. Flow cytometric detection