Vol. 11, 239 –246, May 2000 Cell Growth & Differentiation 239 Role of p53 in Cell Cycle Regulation and Apoptosis following Exposure to Proteasome Inhibitors1 Feng Chen, Daniel Chang, Meidee Goh, Under normal conditions, the p53 tumor suppressor pro- Sergey A. Klibanov, and Mats Ljungman2 tein is rapidly degraded by the 26S proteasome (4 –7) in a Department of Radiation Oncology, Division of Cancer Biology, process mediated by MDM2 (8 –10) and jun kinase (11). It has University of Michigan Comprehensive Cancer Center [F. C., D. C., S. A. K., M. L.] and Section of Urology [M. G.], Program in Cellular and been reported that nuclear export may be required for the Molecular Biology [M. L.], University of Michigan Medical School, Ann efficient degradation of p53 (12–14), suggesting that cyto- Arbor, Michigan 48109-0396 plasmic but not nuclear proteasomes are responsible for the degradation of p53. After cellular stresses, such as exposure to DNA-damaging agents, the half-life of the p53 protein is Abstract significantly increased, and p53 accumulates in the nucleus In this study, we explored what effect inhibitors of the 26S proteasome have on cell cycle distribution and of treated cells (15, 16). The mechanism by which the stability induction of apoptosis in human skin fibroblasts and of p53 is increased after cellular stress is not fully under- colon cancer cells differing in their p53 status. We stood, but it is likely that modifications of the p53 protein found that proteasome inhibition resulted in nuclear itself are involved (7). Phosphorylation of Ser15, Ser20, and accumulation of p53. This was surprising because it is Ser37 has been suggested to result in attenuated interaction thought that the degradation of p53 is mediated by between p53 and the negative regulator MDM2 (17–20). In- cytoplasmic 26S proteasomes. Nuclear accumulation terference with components of the degradation pathway, of p53 was accompanied by the induction of both such as phosphorylation of MDM2 (21), could also result in p21WAF1 mRNA and protein as well as a decrease in increased stability of p53. Furthermore, blockage of nuclear cells entering S phase. Interestingly, cells with export of MDM2 by the ARF protein (22) or by treatment with compromised p53 function showed a marked increase leptomycin B, an inhibitor of the CRM1-dependent nuclear in the proportion of cells in the G2-M phase of the cell export machinery (12–14), leads to the accumulation of p53 cycle and an attenuated induction of apoptosis after in the nucleus. proteasome inhibition. Taken together, our results Inhibition of the 26S proteasome results in the rapid suggest that proteasome inhibition results in nuclear accumulation of p53 (4, 5, 23) and of p53-inducible gene accumulation of p53 and a p53-stimulated induction of products such as p21WAF1, MDM2, and Bax (5, 23–25). In both G1 arrest and apoptosis. addition, certain cell types have been shown to undergo apoptosis after treatment with proteasome inhibitors (23, Introduction 26, 27) through a process that has been suggested to be The ubiquitin-dependent protein degradation pathway in- p53 dependent (24). Thus, accumulation of p53 by default volving the 26S proteasome plays an important role in the by simply inhibiting its degradation appears to activate regulation of various cellular processes such as cell cycle downstream events. However, the induction of p21WAF1, progression, cell differentiation, signal transduction, stress MDM2, and Bax by proteasome inhibition may not be responses, and apoptosis (1, 2). The orderly progression entirely dependent on p53-mediated transactivation (18, through the cell cycle is orchestrated by a tightly regulated 25, 28). Because these proteins are normally subjected to ubiquitin-mediated proteolysis of both cell cycle inhibitors regulation by proteasome-mediated degradation, it is pos- and cyclins (3). Thus, inhibition of the proteasome-mediated sible that the accumulation observed was due to their pathway may influence cell cycle progression as well as increased half-lives in the absence of proteasome-medi- many other cellular functions. ated degradation (25, 29). Furthermore, recent studies have questioned the requirement of p53 in the induction of apoptosis resulting from inhibition of proteasome function (30, 31). Received 9/29/99; revised 2/16/00; accepted 4/5/00. To address some of these controversies, we examined the The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked role of p53 in the induction of p21WAF1, cell cycle arrest, and advertisement in accordance with 18 U.S.C. Section 1734 solely to indi- induction of apoptosis after proteasome inhibition using a cate this fact. panel of cell lines differing in their p53 status. We show that 1 Supported by Seed Grant 836 from the Michigan Memorial-Phoenix Project, by a grant from the University of Michigan Comprehensive Cancer p53 actually accumulated in the cell nucleus rather than the Center’s Institutional Grant from the American Cancer Society, by NIH cytoplasm after proteasome inhibition. Furthermore, protea- Grant CA82376-01, and by start-up funds supplied by the Department of Radiation Oncology, University of Michigan. some inhibition resulted in the p53-dependent induction of 2 To whom requests for reprints should be addressed, at the Department p21WAF1, G1 arrest, and apoptosis, whereas a proteasome- of Radiation Oncology, Division of Cancer Biology, University of Michigan Comprehensive Cancer Center, 4306 CCGC, 1500 East Medical Center mediated arrest in G2-M phase was revealed in cells with Drive, Ann Arbor, MI 48109-0396. E-mail: firstname.lastname@example.org. compromised p53 function. 240 p53 and Proteasome Inhibitors though p53 is thought to be exported to cytoplasmic pro- teasomes, inhibition of proteasome activity results in nuclear accumulation of p53. Accumulation of p21WAF1 by LLnL3 or Lactacystin Is Stimulated by Wild-Type p53. To examine whether the drug-induced accumulation of p53 resulted in the induction of p21WAF1 expression, a panel of isogenic cell lines differing in their functional p53 status was used. First we character- ized the RKO parental and human papillomavirus 16 E6- transfected RKO cell lines in terms of their ability to induce cellular accumulation of p53 and/or p21WAF1 after 6 h of incubation with the proteasome inhibitors LLnL (30 M) or lactacystin (10 M). These doses were chosen because they have been shown to be sufficient to inhibit proteasome ac- tivity in cells (33, 34). The E6 protein interacts with the cellular protein E6AP to target the p53 protein for ubiquitylation and subsequent proteasome-mediated degradation (35, 36). It has been shown that the E6-mediated degradation in these cells overrides the p53 stabilization normally seen after exposure to ionizing radiation, UV light, or actinomycin D (37, 38). Here we show that the basal level of p53 was significantly lower in the E6-expressing cells (RKO-E6) than in the paren- tal RKO cell line (Fig. 2A). LLnL and lactacystin induced p53 to high levels in the parental RKO cell line, whereas the Fig. 1. Exposure of cells to lactacystin resulted in nuclear accumulation absolute p53 protein levels in the RKO-E6 cells were not of p53. Diploid human fibroblasts were mock-treated (A), irradiated with 30 J/m2 UVC and incubated for 16 h (B) or 6 h with 10 M lactacystin (C–F), increased as much as in the parental RKO cell line (Fig. 2A). followed by fixation and immunocytochemistry with anti-p53 antibodies Because the untreated RKO-E6 cells had such a low basal and FITC-conjugated antimouse antibodies. A–C were photographed at level of p53, the relative induction of p53 protein by protea- 40 magnification, and D–F were photographed at 100 magnification. Images A–D were obtained using fluorescence microscopy and anti-p53 some inhibition may rival that of the parental cells. However, antibody 1801, whereas images E and F were captured digitally using this quantification could not be done accurately due to the scanning confocal microscopy and anti-p53 antibodies AB-1 and FL-393, low p53 level in the untreated RKO-E6 cells. respectively. We next investigated the cellular levels of the p53-induc- ible cyclin-dependent kinase inhibitor p21WAF1 and the ability of LLnL to stimulate its expression. The basal level of Results p21WAF1 was markedly lower in the RKO-E6 cells than in the Nuclear Accumulation of p53 in Cells Treated with Lac- RKO cells (Fig. 2B). Treatment of the cells with LLnL strongly tacystin. Under nonstressed conditions, p53 is thought to induced p21WAF1 expression in the parental RKO cells, be actively translocated from the nucleus to the cytoplasm, whereas we observed only a slight induction of p21WAF1 in where it is degraded by the 26S proteasome (12–14). After the E6-expressing cells. As for the accumulation of p53, the drug-induced inhibition of the proteasome, p53 would be relative induction of p21WAF1 by LLnL may be similar be- expected to accumulate in the cytoplasm, as suggested tween the two cell lines, whereas the absolute protein levels previously by transient transfection experiments using p53- were much higher in the parental RKO cells. expressing vectors (32). To explore the localization of p53 We then assessed the effect of lactacystin on the cellular after proteasome inhibition, we treated diploid human fibro- levels of p53 and p21WAF1 in primary human fibroblasts blasts with the proteasome-specific inhibitor lactacystin, fol- derived from a normal individual (NFs) or from a Li-LFS lowed by immunohistochemistry with various anti-p53 anti- patient lacking functional p53 expression. Lactacystin bodies. Using fluorescence microscopy it was found that strongly induced p53 and p21WAF1 in NF cells, whereas no untreated fibroblasts showed only a faint staining (Fig. 1A), expression was seen in the LFS cells either with or without whereas cells irradiated with 30 J/m2 UVC light presented lactacystin (Fig. 2C). The mutant p53 gene in the LFS cells strong nuclear staining of p53 (Fig. 1B). Interestingly, lacta- has a frameshift mutation that leads to a truncated p53 cystin treatment for 6 h led to a strong nuclear accumulation protein that is very unstable in the cells (39). Finally, in a cell of p53 (Fig. 1, C–F). At 100 magnification, using either line overexpressing mutant p53 (HT29), which exhibits a fluorescence (Fig. 1D) or scanning confocal fluorescence significant baseline level of p21WAF1 expression, lactacystin microscopy (Fig. 1, E and F), it can be seen that p53 appears to accumulate in hundreds of foci throughout the nucleus, 3 with some areas devoid of staining. Similar nuclear accumu- The abbreviations used are: LLnL, N-acetyl-L-leucinyl-L-leucinyl-L- norleucinal; BrdUrd, bromodeoxyuridine; PI, propidium iodide; LFS, Li- lation of p53 was observed with the specific 26S proteasome Fraumeni syndrome; NF, normal fibroblast; PBSBT, 5 g of bovine albumin inhibitor MG132 (data not shown). We conclude that al- and 500 l Tween 20/liter PBS. Cell Growth & Differentiation 241 protein level of p21WAF1 observed in these cells. Taken to- gether, these results suggest that both protein stabilization and, to some extent, transcriptional up-regulation are re- sponsible for the accumulation of p21WAF1 by proteasome inhibitors. Modest p53-dependent G1 Arrest after Proteasome In- hibition. Because the RKO and RKO-E6 cells showed a differential expression of p53 and p21WAF1 after exposure to LLnL (Fig. 2, A and B), these cells were used to explore the role of p53 in cell cycle regulation after exposure to LLnL. The RKO and RKO-E6 cell lines were incubated with LLnL for 6 or 24 h, followed by a 15-min BrdUrd incubation to specifically label nascent DNA synthesis. Cells were then fixed and stained for DNA content, and anti-BrdUrd antibodies and secondary antibodies conjugated with FITC were used to identify cells synthesizing DNA at the time of labeling. Using two-parameter flow cytometry, the cell cycle effects of LLnL incubation were then analyzed. It was expected that LLnL would induce a G1 arrest in the parental RKO cells because this treatment resulted in a sig- nificant accumulation of p21WAF1 (Fig. 2B). Using one- parameter flow cytometry of PI-stained RKO, RKO-E6, and mutant p53 RKO-M cells (Fig. 3A), no obvious G1 arrest was observed in the cell lines tested. However, when using two- parameter flow cytometry of cells pulse-labeled with BrdUrd, we observed a marked difference between parental RKO and RKO-E6 cells in the G1-S-phase compartment of the cell cycle (Fig. 3B; Table 1). Whereas the RKO-E6 cells were stimulated to enter S phase after 6 h of treatment, the RKO cells were not. In fact, 24 h of LLnL treatment led to a visible decline in RKO cells occupying the S-phase compartment. At this time the number of parental RKO cells synthesizing DNA had dropped from 45% to 19% (Table 1). Thus, com- pared with the RKO-E6 cells, the parental RKO cells ap- peared to be blocked from entering the S phase, perhaps as a result of the strong induction of p21WAF1 in these cells (Fig. 2B). The mean FITC signal of the BrdUrd-positive cells, which is an indication of the rate of BrdUrd incorporation and thus the rate of DNA synthesis, was lower for the parental RKO cells than for the RKO-E6 cells after 24 h of LLnL incubation Fig. 2. Induction of p21WAF1 accumulation by proteasome inhibition is (Table 1). Thus, not only were fewer parental RKO cells attenuated in cells with compromised p53 function. Parental RKO cells and E6-expressing RKO-E6 cells were incubated with 30 M LLnL or 10 entering S phase, the cells that entered did synthesize DNA M lactacystin for 6 h before cells were collected, and the cellular levels of less efficiently. By multiplying the mean FITC signal with the p53 (A) or p21WAF1 (B) were assessed by Western blot. Human fibroblasts number of BrdUrd-positive cells, we obtained an estimation from a normal (NF) or LFS (LFS) individual (C) or human colon carcinoma HT29 cells with mutant p53 (D) were treated with 10 M lactacystin for 6 h of the relative amount of DNA synthesis in the different cell before the cellular levels of p53 or p21WAF1 were assessed by Western populations. LLnL did not affect DNA synthesis negatively blot. E, Northern blot of p21WAF1 mRNA (top) or -actin (bottom) expres- sion in cells mock-treated or treated with 10 M lactacystin for 6 h. The during the first 6 h of incubation in either cell line (Table 1). expression of ribosomal RNAs in these cells is shown in the left panel. However, after 24 h of LLnL incubation, the DNA synthesis in parental RKO cells was reduced to 26% of the level found in untreated control cells. In contrast, the LLnL-treated RKO-E6 did not induce a significant increase in p21WAF1 protein levels cells synthesized DNA at a rate comparable with that of (Fig. 2D). untreated control cells (Table 1). To examine whether the accumulation of p21WAF1 by pro- Proteasome Inhibition Results in a G2-M-phase Arrest teasome inhibition was due to up-regulation of the p21WAF1 in p53-compromised Cells. Whereas LLnL treatment led to gene, we performed Northern blots with p21WAF1-specific a modest G1 arrest in the wild-type p53-expressing RKO probes (Fig. 2E). Although lactacystin caused a slight up- cells, the RKO-M and RKO-E6 cells showed a significant regulation of the p21WAF1 gene as reported previously (5), it accumulation in the G2-M phase of the cell cycle (Fig. 3A). clearly did not fully account for the dramatic increase of the The percentages of cells in G2-M phase increased from 15% 242 p53 and Proteasome Inhibitors Table 1 The relative DNA synthesis is reduced in a p53-dependent manner after LLnL treatment % BrdUrd-incorporating Mean Relative DNA Cells Time (h) cellsa FITCa synthesisb RKO 0 45 36 100 6 52 41 132 24 19 22 26 RKO-E6 0 37 63 100 6 64 63 173 24 43 64 118 a Data taken from window H in Fig. 3B. b Relative DNA synthesis was calculated by multiplying the mean FITC value by the number (percentage) of cells in the H window and then expressing the results relative to untreated control cells. of the cell cycle after proteasome inhibition. Treatment with 10 M lactacystin increased the percentage of cells in G2-M phase from 9% in untreated cells to 27% in treated cells (Fig. 3C). We conclude that LLnL and lactacystin induce an ac- cumulation of p53-deficient cells in the G2-M phase of the cell cycle, whereas cells with wild-type p53 do not reveal a G2-M phase arrest, presumably because these cells arrest in G1 instead. Wild-Type p53 Sensitizes Cells to Proteasome Inhibi- tor-induced Apoptosis. Previous studies have shown that treatment of human and mouse cancer cells with protea- some inhibitors induces apoptosis (26, 27, 30, 31). Further- more, induction of apoptosis by proteasome inhibitors ap- pears to be p53 dependent in some cell types (24) but not in others (30, 31) and may be selective for transformed cells (31). Here we investigated the effect of LLnL on the induction of apoptosis in the parental RKO cells and compared it with the effect on the E6-expressing cells. Cells were incubated for 48 h in the presence of 30 M LLnL before both floating and attached cells were collected, fixed, and stained with PI. Flow cytometric analysis of the percentage of cells contain- ing sub-G1 DNA content as a measure of apoptosis revealed that LLnL induced apoptosis in both cell types (Fig. 4). How- ever, the induction of apoptosis was significantly higher (P 0.05) in the parental RKO cells (58%) as compared with the Fig. 3. p53-mediated G1 arrest and p53-independent G2-M-phase arrest RKO-E6 cells (38%). Thus, induction of apoptosis by LLnL in after proteasome inhibition. A, parental RKO, RKO-M, and RKO-E6 cells these human colon cancer cells did not require wild-type p53 were incubated with 30 M LLnL for 0, 6, or 24 h, followed by fixation, staining with PI, and flow cytometry. The position of G1, S phase, and function. However, wild-type p53 significantly enhanced G2-M phase in the flow diagrams is shown in the top left panel. B, RKO LLnL-induced apoptosis. and RKO-E6 cells were treated with LLnL for 6 or 24 h or untreated We next repeated the experiments using lactacystin in- (control) before BrdUrd was added to the media for 15 min to pulse-label cells actively synthesizing DNA. Cell cycle analysis was then performed stead of LLnL to more specifically inhibit 26S proteasome using two-parameter flow cytometry. The amount of BrdUrd incorporation activity. We found that 10 M lactacystin induced a more per cell is expressed on the Y axis (LOG FITC), whereas the DNA content per cell is expressed on the X axis (PI-DNA). The different encircled severe induction of apoptosis in the RKO cells than was regions represent the following: E, G1; F, S-phase DNA-containing cells observed after incubation with 30 M LLnL. At 24 h after not synthesizing DNA; G, G2-M; and H (the whole encircled region above adding lactacystin, 84% of the RKO cells were already the horizontal line), S-phase cells actively synthesizing DNA. C, HT29 cells treated with or without 10 M lactacystin for 24 h before fixation, PI scored as apoptotic, whereas 55% of the RKO-E6 cells had staining, and cell cycle analysis using flow cytometry. undergone apoptosis (Fig. 5A). These results are in agree- ment with the preferential induction of apoptosis in the wild- type p53-expressing RKO cells after LLnL treatment. in untreated RKO-E6 cells to 27% and 49% after 6 and 24 h Finally, we explored the induction of apoptosis in NF and of drug treatment, respectively. During the same time period, LFS cells treated with 10 M lactacystin. Again, the NF cells the population of cells in G1 was severely diminished. with wild-type p53 exhibited a slightly higher induction of Using HT29 colon cancer cells, which harbor mutant p53, apoptosis than the LFS cells, although this difference was we also observed an accumulation of cells in the G2-M phase not dramatic. Furthermore, whereas no significant cell cycle Cell Growth & Differentiation 243 Fig. 5. Lactacystin treatment resulted in accumulation of p53-compro- mised cells in the G2-M phase and a slightly higher percentage of wild- Fig. 4. p53 stimulates LLnL-induced apoptosis in RKO cells. A, parental type p53-expressing cells undergoing apoptosis compared with p53- RKO and E6-expressing RKO-E6 cells were incubated with 30 M LLnL for compromised cells. A, RKO and RKO-E6 cells were treated with 10 M 48 h before apoptosis was assessed by measuring the percentage of cells lactacystin for 24 h before both attached and floating cells were collected, with sub-G1 DNA content using PI staining and flow cytometry. B, quan- fixed, and stained with PI. Flow cytometry analysis revealed massive tification of multiple experiments show that a difference between the apoptosis in both cell lines and accumulation of nonapoptotic cells in the amount of apoptosis induced by the two cell lines was significant using G2-M phase of the cell cycle in the RKO-E6 cells. B, normal human skin the Student’s t test (P 0.05). The values are the mean of two (control) or fibroblasts (NF) or a cell line derived from a LFS patient (LFS) was treated three (treated) different biological samples, with error bars showing the with 10 M lactacystin. After 24 h of incubation, detached and attached sample SD. cells were collected, fixed, and stained with PI. The percentages of cells with a sub-G1 DNA are indicated. perturbation was observed in the nonapoptotic population of However, this scenario is less plausible because p53 has NF cells, a clear accumulation of cells in the G2-M phases been shown to accumulate in the cell nucleus after overex- was observed in the LFS cell population (Fig. 5B), supporting pression of wild-type p53 (40) or inhibition of the nuclear the findings obtained in the RKO-E6 cells after LLnL treat- export machinery with leptomycin B (12–14). Finally, it is ment (Fig. 3A). possible that the nuclear accumulation of p53 by protea- some inhibition is due to a direct or an indirect inhibition of a Discussion proteasome-mediated step involved in nuclear export. Ad- In this study, we investigated the role of p53 in mediating ditional studies are needed to elucidate the true mechanism cellular responses after inhibition of the 26S proteasomes. by which proteasome inhibition leads to nuclear accumula- We found that proteasome inhibition resulted in a significant tion of p53. accumulation of p53 in the nucleus of treated cells. This Both lactacystin and LLnL strongly induced the accumu- finding was surprising, considering that p53 is thought to be lation of wild-type p53 in the parental RKO cell line and in degraded primarily in the cytoplasm and not in the cell nu- NFs. In contrast, only a marginal increase or no increase in cleus (12–14). One hypothesis for the nuclear accumulation the absolute p53 protein level was observed in the E6- of p53 after proteasome inhibition is that p53 proteins tar- expressing cells, the LFS cells, or the HT29 cells containing geted for degradation by cytoplasmic proteasomes may be mutant p53. Furthermore, induction of p21WAF1 by protea- “recycled” and recruited back into the nucleus. Increased some inhibition was seen only in the cell cultures in which concentration of nuclear p53 would favor p53 tetrameriza- p53 accumulated, suggesting that the induction of p21WAF1 tion, which would block nuclear export (14) and further aug- may have been, at least in part, p53 dependent. This finding ment nuclear accumulation. Alternatively, p53 may, under is similar to those in studies in which forced expression of normal conditions, be degraded by nuclear proteasomes. exogenous p53 in cells (41) or treatment of cells with the 244 p53 and Proteasome Inhibitors nuclear export inhibitor leptomycin B (12, 28) resulted in normally subjected to degradation by the ubiquitin-protea- induction of p21WAF1 expression. However, the induction of some pathway (3, 25, 45, 46). p21WAF1 after proteasome inhibition could only partially be A profound arrest in the G2-M phase of the cell cycle was explained by an up-regulation of the p21WAF1 gene (Fig. 2E). found for cells with compromised p53 status but not in the Thus, the accumulation of p21WAF1 protein cannot be entirely wild-type p53-expressing cells after treatment with either explained by p53-mediated transactivation but may also re- LLnL or lactacystin (Figs. 3 and 5). One explanation for these sult from increased stability of the p21WAF1 protein in the results is that inhibition of proteasome activity causes a absence of proteasome degradation (5). Although p53 iso- G2-M-phase arrest that is only revealed in cells lacking wild- lated from proteasome inhibitor-treated cells has been re- type p53. Because the entry of wild-type p53-expressing ported to be fully capable of binding to oligonucleotides cells into S phase was reduced after drug treatment, a containing p21WAF1 promoter sequences (18), some studies smaller percentage of cells would be expected to reach the suggest that these p53 proteins are fairly poor in transacti- G2-M-phase compartment of the cell cycle. The mechanism vating either endogenous (18) or exogenous p21WAF1 pro- for the G2-M-phase arrest in the p53-compromised cells moters (28). The limited transactivation activity of p53 after after proteasome inhibition may be due to the inability of the proteasome inhibition may result from conformational mis- cells to exit mitosis. The exit of the M phase requires pro- folding of nascent p53 in the presence of proteasome inhib- teasome activity to degrade the targets of the anaphase itors (32). Alternatively, the relatively low transactivation ac- promoting complex such as cyclins A and B (3). An alterna- tivity of drug-accumulated p53 may be due to the lack of tive hypothesis for the lack of G2-M-phase arrest in the protein modifications, such as phosphorylation or acetylation wild-type p53-expressing cells could be that the exit from the (19, 42), which are thought to stimulate the transactivation G2-M phase is stimulated by wild-type p53 expression. This function of p53 (43). hypothesis is supported by findings that p53 may play a role Because the stability of p21WAF1 is increased after protea- in the exit from DNA damage-induced G2 arrest (47). some inhibition, it is expected that p21WAF1 levels should The role of p53 in the induction of apoptosis by protea- increase, even in cells with compromised p53 function. How- some inhibitors is controversial (24, 30, 31). Proteasome ever, the rather small increase of p21WAF1 protein levels in inhibitors have been shown to induce apoptosis in a number LLnL-treated RKO-E6 cells and the absence of such an of human cell lines (26, 27, 30, 31). However, the induction of increase in the LFS cells may have been related to the rather apoptosis in sympathetic neurons and thymocyte cultures low basal expression of this protein due to the absence of has been shown to be inhibited by proteasome inhibitors (48, functional p53. However, our result using the p53 mutant cell 49). Our results suggest that wild-type p53 function is not line HT29 shows that the detectable basal level of p21WAF1 required for the induction of apoptosis by LLnL or lactacystin did not increase after incubation with lactacystin (Fig. 2D). in the cells used in this study (Figs. 4 and 5). However, This would argue against protein stabilization as the major wild-type p53 function clearly stimulated the induction of mechanism of p21WAF1 induction after proteasome inhibi- apoptosis after proteasome inhibition. It has been shown that tion. The marginal induction of p53 expression after treat- certain apoptosis-promoting proteins such as Bax are sub- ment with proteasome inhibitors in the E6-expressing cells is jected to proteasome-mediated degradation (25). It is thus in disagreement with a previous study in which E6-trans- possible that inhibition of proteasome-mediated degradation fected normal fibroblasts were shown to readily accumulate of Bax together with increased expression of Bax in the p53 after incubation with the proteasome inhibitor MG132 presence of high levels of p53 may explain the stronger (5). The reason for this discrepancy is not clear, but it may be induction of apoptosis in the wild-type p53-expressing cells. that the RKO-E6 cell line used in our study expressed the E6 It is also possible that the potential p53-stimulated exit from protein to a higher level than the E6-transfected fibroblasts. the G2-M phase of the cell cycle may enhance apoptosis, as Whereas Maki et al. (5) readily detected p53 in the untreated has been suggested previously (47). E6-expressing fibroblasts used in their study, no p53 protein In conclusion, we have found that proteasome inhibition expression was detected in the RKO-E6 cells (Fig. 2A; Refs. leads to nuclear accumulation of p53, a p53-stimulated in- 37 and 38). duction of p21WAF1, and cell cycle arrest in the G1 phase of Whereas wild-type p53-expressing RKO cells tended to the cell cycle. Furthermore, cells with compromised p53 arrest in G1 after treatment with LLnL, RKO-E6 cells ap- function accumulated in the G2-M phase of the cell cycle and peared to be stimulated to enter S phase (Fig. 2B; Table 1). were somewhat protected against the induction of apoptosis This finding that LLnL blocked cells from entering S phase in by proteasome inhibition. Because proteasome inhibitors are wild-type p53 cells is in agreement with a previous study (44). potentially useful as antitumor agents (50, 51), the true mech- The most likely explanation for this G1 arrest is that the anisms of how these agents induce apoptosis and how p53 accumulation of the cyclin-dependent kinase inhibitor modifies these events warrants further exploration. p21WAF1 after LLnL treatment blocked the G1 to S-phase transition of these cells. The mechanism responsible for the Materials and Methods increased S-phase entry of the RKO-E6 cells after LLnL Cell Culture and Chemicals. Three isogenic human colon cancer cell treatment is not clear, but it may involve accumulation of lines, the parental cell line RKO, the mutant p53-expressing cell line RKO-M, and the human papillomavirus 16 E6-expressing cell line RKO- some cellular component(s) driving the progression of cells E6, were generously given to us by Dr. Albert Fornace, Jr., (National from G1 into S phase. Such components may include cyclin Cancer Institute, NIH, Bethesda, MD) via Drs. Ted Lawrence and Mary D1, cyclin E, and/or the transcription factor E2F1, which are Davis (University of Michigan). These cell lines have been described in Cell Growth & Differentiation 245 detail previously (37, 52, 53). The human colon carcinoma cell line HT29 References (mutant p53) and a normal diploid skin fibroblast cell strain were gener- 1. Varshavsky, A. The ubiquitin system. Trends Biochem. Sci., 22: 383– ously provided to us by Drs. Ted Lawrence and Mary Davis (University of 387, 1997. Michigan). Finally, the spontaneously immortalized LFS fibroblast cell 2. Hilt, W., and Wolf, D. H. Proteasomes: destruction as a programme. strain MDAH041, which harbors a single mutant copy of the p53 gene (p53 Trends Biochem. Sci., 21: 96 –102, 1996. /mt) was a generous gift from Michael Tainsky (Wayne State University, Detroit, MI). Cells were grown as monolayers in MEM (Life Technologies, 3. Koepp, D., Harper, J., and Elledge, S. How the cyclin became a cyclin: Inc.) supplemented with 10% fetal bovine serum, 1 vitamins, 1 amino regulated proteolysis in the cell cycle. Cell, 97: 431– 434, 1999. acids, and 1 antibiotics. Subconfluent cells seeded 2 days before the 4. Chowdary, D., Dermody, J., Jha, K., and Ozer, H. Accumulation of p53 experiments were treated with LLnL (Sigma), which is a reversible inhibitor in a mutant cell line defective in the ubiquitin pathway. Mol. Cell. Biol., 14: of the chymotryptic site on the proteasome (33), and the irreversible 1997–2003, 1994. proteasome inhibitor lactacystin (Ref. 33; Oncogene Research Products). 5. Maki, C. G., Huibregtse, J. M., and Howley, P. M. In vivo ubiquitination Both drugs were dissolved in DMSO. and proteasome-mediated degradation of p53. Cancer Res., 56: 2649 – Immunocytochemistry. After incubation with lactacystin or UV irra- 2654, 1996. diation, cells grown on coverslips were fixed in 100% methanol and stored 6. Brown, J., and Pagano, M. Mechanism of p53 degradation. Biochim. at 20°C for 1 h. The coverslips were then rinsed in PBS, followed by the Biophys. Acta, 1332: 1– 6, 1997. addition of 100 l of undiluted supernatant from hybridoma cells express- ing mouse anti-p53 mAb 1801 (a gift from Dr. Jiayuh Lin, University of 7. Ljungman, M. Dial 9-1-1 for p53: mechanisms of p53 activation by Michigan). After a 1-h incubation, the antibody solution was aspirated, and cellular stress. Neoplasia, 2: 208 –225, 2000. the coverslips were incubated on a rocker platform three times for 5 min 8. Haupt, Y., Maya, R., Kazaz, A., and Oren, M. Mdm2 promotes the rapid with PBSBT. The coverslips were then incubated for 1 h in the dark with degradation of p53. Nature (Lond.), 387: 296 –299, 1997. 100 l of a secondary FITC-conjugated antimouse IgG antibody (Sigma; 9. Kubbutat, M. H. G., Jones, S. N., and Vousden, K. H. Regulation of p53 1 1000 dilution in PBSBT), followed by aspiration and incubation three stability by Mdm2. Nature (Lond.), 387: 299 –303, 1997. times for 5 min in PBSBT. The coverslips were then mounted on micro- 10. Honda, R., Tanaka, H., and Yasuda, H. Oncoprotein MDM2 is a scope slides in one drop of Vectashield (Vector Laboratories, Inc.) and ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett., 420: 25–27, viewed and photographed using a fluorescence microscope (Nikon 1997. Eclipse E600) or a confocal microscope (MRC 600; BioRad/Nikon). Western Blot. Western blots were performed as described previously 11. Fuchs, S. Y., Adler, V., Buschmann, T., Yin, Z. M., Wu, X. W., Jones, (54) using 12% SDS-PAGE for analysis of p53 and 15% SDS-PAGE for S. N., and Ronai, Z. JNK targets p53 ubiquitination and degradation in analysis of p21WAF1. For the detection of p53 and p21WAF1 proteins, nonstressed cells. Genes Dev., 12: 2658 –2663, 1998. anti-p53 [Ab-2 (Oncogene Research Products) and FL-393 (Santa Cruz 12. Freedman, D. A., and Levine, A. J. Nuclear export is required for Biotechnology, Inc.)] and anti-p21WAF1 (Oncogene Research Products) as degradation of endogenous p53 by MDM2 and human papillomavirus E6. well as horseradish peroxidase-conjugated secondary antibodies (Sigma) Mol. Cell. Biol., 18: 7288 –7293, 1998. were used together with enhanced chemiluminescence (SuperSignal CL- 13. Tao, W. K., and Levine, A. J. Nucleocytoplasmic shuttling of onco- HRP Substrate System; Pierce) and X-ray film. The protein content of each protein Hdm2 is required for Hdm2-mediated degradation of p53. Proc. sample was determined by the use of a UV spectrophotometer and equal Natl. Acad. Sci. USA, 96: 3077–3080, 1999. protein loading, and transfer was confirmed by staining the nylon mem- 14. Stommel, J. M., Marchenko, N. D., Jimenez, G. S., Moll, U. M., Hope, branes with Coomassie Blue after the completion of the film exposure. T. J., and Wahl, G. M. A leucine-rich nuclear export signal in the p53 Northern Blot. Total RNA was isolated from untreated NFs and NFs tetramerization domain: regulation of subcellular localization and p53 treated with 10 M lactacystin using Trizol (Life Technologies, Inc.). Equal activity by NES masking. EMBO J., 18: 1660 –1672, 1999. amounts of RNA (10 g/lane) were loaded on a 1.2% agarose/formalde- hyde gel, and after electrophoresis, the RNA was transferred onto Nytran 15. Maltzman, W., and Czyzyk, L. UV irradiation stimulates levels of p53 nylon membranes (Schleicher & Schuell). Blots were hybridized with cellular antigen in nontransformed mouse cells. Mol. Cell. Biol., 4: 1689 – [ -32P]dCTP (Amersham Pharmacia Biotech) radiolabelled cDNA probes 1694, 1984. p21WAF1 and -actin. The cDNA probes for p21WAF1 and -actin were 16. Kastan, M., Onyekwere, O., Sidransky, D., Vogelstein, B., and Craig, generated using reverse transcription-PCR and isolated through agarose R. Participation of p53 protein in the cellular response to DNA damage. gel extraction. Cancer Res., 51: 6304 – 6311, 1991. BrdUrd Incubation, Cell Fixation, and Flow Cytometry. Cells were 17. Shieh, S. Y., Ikeda, M., Taya, Y., and Prives, C. DNA damage-induced pulse-labeled with 30 mM BrdUrd (Sigma) at 37°C for 15 min as described phosphorylation of p53 alleviates inhibition by MDM2. Cell, 91: 325–334, previously (38). The plates were then rinsed with PBS, and then the cells 1997. were collected by scraping them with a rubber policeman in 2 ml of PBS. 18. Siliciano, J. D., Canman, C. E., Taya, Y., Sakaguchi, K., Appella, E., After centrifugation, the cells were fixed by adding ice-cold 70% ethanol and Kastan, M. B. DNA damage induces phosphorylation of the amino dropwise under vortexing. Staining of BrdUrd-containing DNA was then terminus of p53. Genes Dev., 11: 3471–3481, 1997. performed as described previously (38, 55). The cell samples were ana- lyzed for BrdUrd content (FITC) and DNA content (PI) using two-parameter 19. Shieh, S. Y., Taya, Y., and Prives, C. DNA damage-inducible phos- phorylation of p53 at N-terminal sites including a novel site, Ser20, re- flow cytometry (Coulter Elite ESP Cell Sorter). quires tetramerization. EMBO J., 18: 1815–1823, 1999. Quantification of Apoptosis. Cells were treated with 30 M LLnL for 48 h or 10 M lactacystin for 24 h before cells were harvested and fixed, 20. Unger, T., Juven-Gershon, T., Moallem, E., Berger, M., Sionov, R. V., and the DNA was stained by PI as described previously (54, 56). Samples Lozano, G., Oren, M., and Haupt, Y. Critical role for Ser20 of human p53 were then analyzed using flow cytometry and cells with sub-G1 DNA in the negative regulation of p53 by Mdm2. EMBO J., 18: 1805–1814, content were scored as apoptotic. 1999. 21. Mayo, L. D., Turchi, J. J., and Berberich, S. J. Mdm-2 phosphorylation by DNA-dependent protein kinase prevents interaction with p53. Cancer Acknowledgments Res., 57: 5013–5016, 1997. We thank Dr. Al Fornace and Dr. Ted Lawrence for their kind gifts of the 22. Zhang, Y. P., and Xiong, Y. Mutations in human ARF exon 2 disrupt RKO and HT29 cell lines and Michael Tainsky for kindly supplying us with its nucleolar localization and impair its ability to block nuclear export of the Li-Fraumeni cell strains used in this study. We also thank the members MDM2 and p53. Mol. Cell, 3: 579 –591, 1999. of the Flow Cytometry Core at the University of Michigan for excellent 23. Ljungman, M., Zhang, F. F., Chen, F., Rainbow, A. J., and McKay, technical assistance and Dr. Bruce McKay for valuable input into this B. C. Inhibition of RNA polymerase II as a trigger for the p53 response. study. Oncogene, 18: 583–592, 1999. 246 p53 and Proteasome Inhibitors 24. Gazos Lopes, U., Erhardt, P., Yao, R., and Cooper, G. p53-dependent 41. El-Deiry, W., Tokino, T., Velculescu, V., Levy, D., Parsons, R., Trent, induction of apoptosis by proteosome inhibitors. J. Biol. Chem., 272: J., Lin, D., Mercer, W., Kinzler, K., and Vogelstein, B. WAF1, a potential 12893–12896, 1997. mediater of p53 tumor suppression. Cell, 75: 817– 825, 1993. 25. Chang, Y. C., Lee, Y. S., Tejima, T., Tanaka, K., Omura, S., Heintz, 42. Sakaguchi, K., Herrera, J. E., Saito, S., Miki, T., Bustin, M., Vassilev, N. H., Mitsui, Y., and Magae, J. mdm2 and bax, downstream mediators of A., Anderson, C. W., and Appella, E. DNA damage activates p53 through the p53 response, are degraded by the ubiquitin-proteasome pathway. a phosphorylation-acetylation cascade. Genes Dev., 12: 2831–2841, Cell Growth Differ., 9: 79 – 84, 1998. 1998. 26. Shinohara, K., Tomioka, M., Nakano, H., Tone, S., Ito, H., and Ka- 43. Lambert, P. F., Kashanchi, F., Radonovich, M. F., Shiekhattar, R., and washima, S. Apoptosis induction resulting from proteasome inhibition. Brady, J. N. Phosphorylation of p53 serine 15 increases interaction with Biochem. J., 317: 385–388, 1996. CBP. J. Biol. Chem., 273: 33048 –33053, 1998. 27. Drexler, H. C. A. Activation of the cell death program by inhibition of proteasome function. Proc. Natl. Acad. Sci. USA, 94: 855– 860, 1997. 44. Dietrich, C., Bartsch, T., Schanz, F., Oesch, F., and Wieser, R. J. p53-dependent cell cycle arrest induced by N-acetyl-L-leucinyl-L-leucinyl- 28. Lain, S., Midgley, C., Sparks, A., Lane, E. B., and Lane, D. P. An L-norleucinal in platelet-derived growth factor-stimulated human fibro- inhibitor of nuclear export activates the p53 response and induces the blasts. Proc. Natl. Acad. Sci. USA, 93: 10815–10819, 1996. localization of HDM2 and p53 to U1A-positive nuclear bodies associated with the PODs. Exp. Cell. Res., 248: 457– 472, 1999. 45. Diehl, J., Zindy, F., and Sherr, C. Inhibition of cyclin D1 phosphoryl- 29. Maki, C. G., and Howley, P. M. Ubiquitination of p53 and p21 is ation on threonine-286 prevents its rapid degradation via the ubiquitin- differentially affected by ionizing and UV radiation. Mol. Cell. Biol., 17: proteosome pathway. Genes Dev., 11: 957–972, 1997. 355–363, 1997. 46. Hofmann, F., Martelli, F., Livingston, D., and Wang, Z. The retinoblas- 30. Herrmann, J. L., Briones, F., Jr., Brisbay, S., Logothetis, C. J., and toma gene product protects E2F-1 from degradation by the ubiquitin- McDonnell, T. J. Prostate carcinoma cell death resulting from inhibition of proteosome pathway. Genes Dev., 10: 2949 –2959, 1996. proteasome activity is independent of functional Bcl-2 and p53. Onco- 47. Guillouf, C., Rosselli, F., Krishnaraju, K., Moustacchi, E., Hoffman, B., gene, 17: 2889 –2899, 1998. and Liebermann, D. A. p53 involvement in control of G2 exit of the cell 31. An, B., Goldfarb, R., Siman, R., and Dou, P. Novel dipeptidyl protea- cycle: role in DNA damage-induced apoptosis. Oncogene, 10: 2263– some inhibitors overcome Bcl-2 protective function and selectively accu- 2270, 1995. mulate the cyclin-dependent kinase inhibitor p27 and induce apoptosis in 48. Grimm, L. M., Goldberg, A. L., Poirier, G. G., Schwartz, L. M., and transformed, but not normal, human fibroblasts. Cell Death Differ., 5: Osborne, B. A. Proteasomes play an essential role in thymocyte apopto- 1062–1075, 1998. sis. EMBO J., 15: 3835–3844, 1996. 32. Magae, J., Illenye, S., Tejima, T., Chang, Y. C., Mitsui, Y., Tanaka, K., Omura, S., and Heintz, N. H. Transcriptional squelching by ectopic ex- 49. Sadoul, R., Fernandez, P. A., Quiquerez, A. L., Martinou, I., Maki, M., pression of E2F-1 and p53 is alleviated by proteasome inhibitors MG-132 Schroter, M., Becherer, J. D., Irmler, M., Tschopp, J., and Martinou, J. C. and lactacystin. Oncogene, 15: 759 –769, 1997. Involvement of the proteasome in the programmed cell death of NGF- deprived sympathetic neurons. EMBO J., 15: 3845–3852, 1996. 33. Rock, K., Gramm, C., Rothstein, L., Clark, K., Stein, R., Dick, L., Hwang, D., and Goldberg, A. Inhibitors of the proteasome block the 50. Adams, J., Palombella, V., Sausville, E., Johnson, J., Destree, A., degradation of most cell proteins and the generation of peptides pre- Lazarus, D., Maas, J., Pien, C., Prakash, S., and Elliott, P. Proteasome sented on MHC class I molecules. Cell, 78: 761–771, 1994. inhibitors: a novel class of potent and effective antitumor agents. Cancer 34. Fenteany, G., Standaert, R., Lane, W., Choi, S., Corey, E., and Schre- Res., 59: 2615–2622, 1999. iber, S. Inhibition of proteasome activities and subunit-specific amino- 51. Meng, L., Kwok, B., Sin, N., and Crews, C. Eponemycin exerts its terminal threonine modification by lactacystin. Science (Washington DC), antitumor effect through the inhibition of proteasome function. Cancer 268: 726 –731, 1995. Res., 59: 2798 –2801, 1999. 35. Scheffner, M., Werness, B., Huibregtse, J., Levine, A., and Howley, P. 52. Kuerbitz, S., Plunkett, B., Walsh, V., and Kastan, M. Wild-type p53 is The E6 oncoprotein encoded by human papillomavirus types 16 and 18 a cell cycle checkpoint determinant following irradiation. Proc. Natl. Acad. promotes the degradation of p53. Cell, 63: 1129 –1136, 1990. Sci. USA, 89: 7491–7495, 1992. 36. Scheffner, M., Huibregtse, J. M., and Howley, P. M. Identification of a 53. Kessis, T., Slebos, R., Nelson, W., Kastan, M., Plunkett, B., Han, S., human ubiquitin-conjugating enzyme that mediates the E6-AP-dependent Lorincz, A., Hedrick, L., and Cho, K. Human papillomavirus 16 E6 expres- ubiquitination of p53. Proc. Natl. Acad. Sci. USA, 91: 8797– 8801, 1994. sion disrupts p53-mediated cellular response to DNA damage. Proc. Natl. 37. Smith, M. L., Chen, I. T., Zhan, Q. M., Oconnor, P. M., and Fornace, Acad. Sci. USA, 90: 3988 –3992, 1993. A. J. Involvement of the p53 tumor suppressor in repair of u.v.-type DNA damage. Oncogene, 10: 1053–1059, 1995. 54. Ljungman, M., and Zhang, F. Blockage of RNA polymerase as a possible trigger for uv light-induced apoptosis. Oncogene, 13: 823– 831, 38. Chang, D., Chen, F., Zhang, F. F., McKay, B. C., and Ljungman, M. 1996. Dose-dependent effects of DNA-damaging agents on p53-mediated cell cycle arrest. Cell Growth Differ., 10: 155–162, 1999. 55. Hoy, C., Seamer, L., and Schimke, R. Thermal denaturation of DNA for 39. Loignon, M., Fetni, R., Gordon, A. J. E., and Drobetsky, E. A. A immunochemical staining of incorporated bromodeoxyuridine (BrdUrd): p53-independent pathway for induction of p21waf1cip1 and concomitant critical factors that affect the amount of fluorescence and the shape of G1 arrest in UV-irradiated human skin fibroblasts. Cancer Res., 57: 3390 – BrdUrd/DNA histogram. Cytometry, 10: 718 –725, 1989. 3394, 1997. 56. Chung, D. H., Zhang, F. F., Chen, F., McLaughlin, W. P., and Ljung- 40. Liang, S. H., Hong, D., and Clarke, M. F. Cooperation of a single lysine man, M. Butyrate attenuates BCLXL expression in human fibroblasts and mutation and a C-terminal domain in the cytoplasmic sequestration of the acts in synergy with ionizing radiation to induce apoptosis. Radiat. Res., p53 protein. J. Biol. Chem., 273: 19817–19821, 1998. 149: 187–194, 1998.
Pages to are hidden for
"Role of p53 in Cell Cycle Regulation and Apoptosis"Please download to view full document