Genetic background determines the response to adenovirus ... - Nature

Genetic background determines the response to adenovirusmediated wild-type p53 expression in pancreatic tumor cells M. Cascallo ´,1 E. Mercade´,2 G. Capella `,

3

F. Lluı´s,3 C. Fillat,4 A. M. Go ´mez-Foix,1 and A. Mazo1

Departments of 1Biochemistry and Molecular Biology and 2Microbiology, University of Barcelona, Barcelona, Spain; 3Laboratory of Gastrointestinal Research, Hospital de la Santa Creu i Sant Pau Barcelona, Spain; and 4Department of Molecular Genetics, Cancer Research Institute (Institut de Recerca (IRO)), Oncolo `gı`ca, Barcelona, Spain. The development of new therapies is particularly urgent with regard to pancreatic tumors. Gene therapy approaches involving p53 replacement are promising due to the central role of p53 in the cellular response to DNA damage and the high incidence of p53 mutations in pancreatic tumors. Adenoviruses containing wild-type (wt) p53 cDNA (Ad5CMV-p53) were introduced into four human pancreatic cell lines to examine the impact caused by exogenous wt p53 on these cells. Introduction of wt p53 in mutant p53 cells (NP-9, NP-18, and NP-31) caused marked falls in cell proliferation and rises in the level of apoptosis. In contrast, overexpression of p53 did not induce apoptosis in NP-29 (wt p53). The presence of p16 contributes to the induction of apoptosis, as demonstrated by introduction of the wt p16 gene (Ad5RSV-p16). Analysis of cell cycle and apoptosis in etoposide-treated cells corroborated the inability of NP-29 to die by apoptosis, suggesting that this wt p53 cell line lacks p53 downstream functions in the apoptosis pathway. Taken together, our results indicate that the effects elicited by exogenous p53 protein depend upon the molecular alterations related to p53 actions on cell cycle and apoptosis. Therefore, knowledge of the genetic background of tumor cells is crucial to the development of efficient therapies based on the introduction of tumor suppressor genes.

Key words: Adenovirus; p53; p16; pancreatic cancer; cell cycle; apoptosis.

P

ancreatic cancer is one of the most aggressive human tumors and the fourth leading cause of cancer death in Europe and North America.1 Surgery is the only curative therapy. Despite improvements in diagnosis and surgical treatment, most patients are not candidates for surgery, and ,5% of all patients survive for 5 years.2 Therefore, the development of new therapeutic approaches is particularly urgent in pancreatic tumors. One of the most promising approaches to emerge from the improved understanding of cancer at the molecular level is the possibility of using gene therapy to selectively target and destroy tumor cells.3 It has became clear that tumorigenesis is driven by alterations in genes that control cell growth or cell death.4,5 Despite the multiplicity of these lesions within a cancer cell, the correction of a single critical genetic lesion may be sufficient to abrogate tumorigenicity in human cancer cells if the function restored succeeds in the reestablishment of growth control to an extent that would eventually lead to selective tumor cell death by apoptosis.6 The p53 tumor suppressor gene is the most frequent

Received July 22, 1998; accepted December 6, 1998. Address correspondence and reprints to Dr. A. Mazo, Department of Biochemistry and Molecular Biology, C/Martı´ i Franque´s 1, 08028 Barcelona, Spain. E-mail address: [email protected] © 1999 Stockton Press 0929-1903/99/$12.00/10

428

target for genetic alterations in human cancer.7 p53 protein has a central role in the cellular response to DNA damage.8,9 Induction of p53 might lead to growth arrest and/or to apoptosis depending upon p53 levels, cell type, and the cellular environment.10,11 It is not fully understood what functions of p53 are required for cell cycle arrest or apoptosis. The role of p53 in the G1 checkpoint, where it induces growth arrest mediated by the cyclin-dependent kinase inhibitor (CKI) p21Cip1, is well established.12,13 Several authors have reported the role of the CKI p16Ink4a in the control of this checkpoint and postulate a cooperation between p16 and p53 that leads to a more efficient induction of premature senescence or apoptosis.14,15 p53 has also been implicated in a G2/M checkpoint.16 –19 Controversial results have been published regarding the function of p53 in this checkpoint. Expression of p53 has been associated with growth arrest in the G2/M phase of the cell cycle,16,17 whereas other authors report a positive role for p53 in G2 exit after g-irradiation or treatment with the DNA topoisomerase II inhibitor etoposide.18,19 It is less clear how p53 induces apoptosis. Recently, Polyak et al have suggested a model for p53 apoptosis on the basis of the marked increase in several transcripts in p53-expressing cells.20 Tumor suppressor gene therapy requires efficient systems of gene delivery. Adenoviral vectors have many advantages over other viral and nonviral vectors in that

Cancer Gene Therapy, Vol 6, No 5, 1999: pp 428 – 436

´ , MERCADE´, CAPELLA`, ET AL: GENETIC BACKGROUND AND p53 REINTRODUCTION CASCALLO

Table 1. Status of Several Genes Involved in G1 Checkpoint in Human Pancreatic Tumor Cell Lines Cell line NP-9 NP-18 NP-29 NP-31

p53* 175 CGC-CAC 246 ATG-GTG wt Exon 7†

p21

p16

wt 11-base pair deletion (codons 27–31) wt wt wt 1-base pair deletion (codon 36) wt Homo‡ (exon 1,2,3)

*CGC, arginine; CAC, histidine; ATG, methionine; GTG, valine. †An abnormal SSCP pattern was evidenced in multiple experiments; no mutation was detected by direct sequencing and/or sequencing of the polymerase chain reaction product. ‡Homozygous deletion

high titers of virus can be obtained and the virus is stable and easy to handle.21 Adenoviruses (Ads) have been successfully assayed in vitro and have been administered safely in vivo to several organs, leading to efficient gene delivery in both systems.22–27 In this study, a recombinant replication-defective Ad (Ad5CMV-p53) was used to transfer wild-type (wt) p53 cDNA into four human pancreatic tumor cell lines. The exogenous gene was delivered efficiently, and the restoration of p53 function significantly inhibited the growth in the four cell lines, more strongly in mutant p53 cells. An arrest in the G1 phase of the cell cycle was observed in all cell lines, but the induction of apoptosis was dependent upon the genetic background of tumor cells. The overall results may have important implications for treatments aiming to restore p53 function in tumor cells. MATERIALS AND METHODS

Construction and preparation of recombinant Ads (rAds) p53 and b-galactosidase cDNA were fused to a simian virus 40 polyadenylation signal and cloned downstream of the cytomegalovirus promoter in a pACCMVpLpA plasmid. rAds were generated by cotransfection of this plasmid with pJM17 into subconfluent cultures of 293 cells using a modified calcium phosphate precipitation technique. Individual viral plaques were isolated and amplified in 293 cells, and rAd plaques containing human p53 cDNA and b-galactosidase cDNA were identified by polymerase chain reaction amplification and restriction enzyme digestion. Virus without insert (Ad-control) was generated in the same way. Ad5RSV-p16 was a kind gift of Dr. J. Fueyo (M.D. Anderson Cancer Center, Houston, Tex). Viral clones were propagated in 293 cells and recovered after 48 hours by three cycles of freezing and thawing. Titers were determined by plaque assay on 293 cells.

Cell lines and culture conditions: Infection conditions The NP-9, NP-18, NP-29, and NP-31 cell lines were derived from human pancreatic adenocarcinomas that had been perpetuated as xenografts in nude mice.28 NP-9 and NP-29 were maintained in Dulbecco’s modified Eagle’s medium/F12 medium (1:1) and NP-18 and NP-31 were maintained in RPMI 1640 (both supplemented with 10% fetal bovine serum) in a humidified atmosphere containing 5% CO2 at 37°C. The status

Cancer Gene Therapy, Vol 6, No 5, 1999

429

of the p53, p21, and p16 genes in these cell lines is summarized in Table 1. Cells were plated and subsequently infected 24 hours later with Ad for 12 hours in serum-free medium at 37°C. Optimal concentrations of virus causing expression in .90% of cells, without visible toxic effects, were determined by infection with Ad5CMV-lacZ and subsequent staining with 5-bromo-4chloro-3-indolyl b-D-galactoside. In the experiments carried out with two viral vectors, sequential incubations separated by 48 hours were performed.

Cell growth rate determination Pancreatic cell lines were seeded (10,000 cells/well) in 24-well culture plates. The cells were infected with either Ad5CMVp53 or Ad-control. Serum-free culture medium was used for mock infection. Quadruplicate dishes of each treatment were counted at regular intervals until 8 days postinfection in an autoanalyzer Multisizer (Coulter, Hialeah, Fla).

Immunocytochemistry assay After 36 hours, the infected cell monolayers were fixed in methanol (220°C) and treated with 3% H2O2 for 5 minutes. Immunodetection was performed after blocking with phosphate-buffered saline/bovine serum albumin 3% and by using a monoclonal antibody (Ab) against p53 (Ab-2), Oncogene Research Products, Cambridge, Mass) as the primary Ab. Mouse immunoglobulin G (IgG) was used as a control. The secondary Ab was a horseradish peroxidase (HRP)-conjugated anti-mouse IgG (Dako, Carpinteria, Calif). Peroxidase activity was determined with diaminobenzidine/H2O2 after incubation at 37°C for 30 minutes. The cells were counterstained with Harris hematoxylin.

Western blot analysis Total cell lysates obtained by incubation in RIPA buffer (150 mM NaCl, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 20 mM ethylenediaminetetraacetic acid, and 1% Igepal CA-630 in 50 mM tris(hydroxymethyl)aminomethane) (pH 7.4)) for 1 hour were sonicated at 50 W for three cycles of 30 seconds. After centrifugation at 14,000 3 g, supernatant proteins (25 mg/lane determined by bicinchoninic acid protein assay) (Pierce, Rockford, Ill) were electrophoretically separated on 10% polyacrylamide-sodium dodecyl sulfate gel and transferred to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany). Membranes were developed according to an enhanced chemiluminescence protocol (Amersham, Arlington Heights, Ill). Monoclonal Ab against human p53 Ab-2 and polyclonal Ab (pAb) against p21 (Santa Cruz Biotechnology, Santa Cruz, Calif) and actin (Sigma, St. Louis, Mo) were used as primary Abs; HRPconjugated anti-mouse IgG, HRP-conjugated anti-goat IgG, and HRP-conjugated anti-rabbit IgG (Sigma) were used as secondary Abs. Relative quantities of p53 and p21 were determined by using a scanner (SNAP SCAN 600-Agfa, Geraert Morstel, Belgium) and Phoretix 1D Advanced software (Non Linear Dynamics, Newcastle, UK) for data densitometric analysis.

Cell cycle analysis All cell cycle distribution measurements were performed on an Epics-XL flow cytometer (Coulter) equipped with an aircooled argon ion laser to give 488 nm light. Triplicates of cell cultures were incubated in etoposide at a final concentration of

430


Figure 1. Immunocytochemical and morphological analysis of pancreatic cell lines after Ad-mediated p53 gene transfer. Mock-infected control NP-18 cells (A) and NP-18 cells infected with 50 m.o.i. of Ad5CMVp53 (B) were immunostained with anti-p53 Ab. Cells were counterstained with hematoxylin. The analysis was performed at 36 hours postinfection (photomicrographs: 3800 magnification). Staining was similar in all infected cell lines. Morphological changes were examined for mock-infected control NP-9 cells (C) and for NP-9 cells infected with 50 m.o.i. of Ad5CMV-p53 (D). The cells were photographed at 72 hours postinfection (photomicrographs: 3400 magnification). Similar results were obtained with the other three cell lines.

2 mM for 24 hours before or after infection with rAds. Cells were harvested and stained in tris(hydroxymethyl)aminomethane-buffered saline containing propidium iodide (PI) (50 mg/ mL), ribonuclease A (10 mg/mL), and Igepal CA-630 (0.1%) for 1 hour at 4°C. Data from $10,000 cells were collected andanalyzed using Multicycle software (Phoenix Flow Systems, San Diego, Calif).

Apoptosis assays Chromatin structure was visualized by staining cells with Hoechst 33342 (5 mg/mL) for 1 hour at 37°C at 36 hours after infection with rAds. For Hoechst 33342 pulse labeling, 1 3 106 cells were trypsinized and resuspended in 1 mL of medium supplemented with 10% fetal bovine serum. Hoechst 33342 was added to a final concentration of 1 mg/mL, and cells were maintained at 37°C for 30 seconds. Uptake of the dye combined with cell size determination allows calculation of the percentage of apoptotic cells.29 Triplicates of each sample

were analyzed with an Elite flow cytometer using the 350-nm line of the argon ion laser. Elite software was used for data analysis.

RESULTS

Efficiency of adenoviral infection in pancreatic cancer cells and analysis of exogenous p53 expression Pancreatic cell lines are efficiently infected by Ad, because all four cell lines used in this work (NP-9, NP-18, NP-29, and NP-31) presented efficiencies of transduction that were .90% when they were infected with Ad5CMV-lacZ reporter virus at concentrations as low as 10 multiplicities of infection (m.o.i.) (data not shown). p53 expression was tested by immunocytochemistry and Western blotting in the four cell lines after infection



431

four cell lines from day 1 to at least day 8 and was maximal at 36 – 60 hours. Later on, p53 levels decreased slowly, probably due to the outgrowth of noninfected cells together with the episomal localization of viral DNA. A time course of p53 expression in NP-9-infected cells is shown in Figure 2B.

Effects of exogenous p53 on cell cycle progression and apoptosis

Figure 2. Western blot analysis of the expression of p53 and p21 proteins after infection with Ad5CMV-p53. A: Immunoblots of total extracts of mock-infected cells (2) and infected cells with 10 m.o.i. of Ad5CMV-p53 (1) corresponding to the four cell lines were analyzed using anti-p53, anti-p21, and anti-actin Abs. B: Time course of p53 expression in the NP-9 cell line transduced at the same conditions. C: Relative amounts of p53 and p21 were quantified by densitometry.

with Ad5CMV-p53. For immunocytochemical assay, the expression and the localization of the exogenous protein were evaluated 36 hours after the viral infection. The immunocytochemical pattern in virus-infected cells revealed a characteristic staining of p53 protein in the nucleus that was not seen in control cells (Fig 1, A and B). Cells expressing exogenous p53 also had a larger mean diameter than noninfected cells (Fig 1, C and D). Cellular extracts of control and Ad5CMV-p53-transduced cells were prepared 36 hours after viral infection to evaluate p53 expression by Western blot. p53 levels were markedly increased in all infected cells (Fig 2A). p21Cip1 protein has been reported as one of the most important transcriptional targets of p53.12,13 When its levels were analyzed, the expression of p21Cip1 was higher in transduced cells than in controls (Fig 2A), demonstrating that the newly expressed p53 protein was transcriptionally functional. Expression of the transgene was maintained in the


Various viral concentrations were assayed to find a dose that would cause a p53-mediated decrease in cell number without significant toxic effects. Optimal results were obtained with doses ranging from 40 to 80 m.o.i. (data not shown). When cell lines were infected with Ad5CMV-p53 (50 m.o.i.) and counted on alternate days postinfection (Fig 3), all four showed a significant decrease in cell number. The growth curves obtained when cells were infected with Ad-control supported the possibility that the inhibitions observed were a consequence of the increased expression of wt p53. In the cultures infected with Ad5CMV-p53, the numbers of cells at 4 days postinfection were 3.8%, 17.5%, and 7.2% in mutant p53 cell lines NP-18, NP-31, and NP-9, respectively, and 30.1% in the wt p53 NP-29 cells, with respect to the controls (Fig 3). These differences suggest that the effect of the expression of the transduced protein on the growth depended upon the p53 status of the cells. To determine the mechanism by which p53 overexpression inhibits growth, cell cycle profiles and apoptosis were analyzed. Flow cytometric profiles obtained at 36 hours after viral infections showed that the expression of exogenous p53 protein led to a significant G1 arrest in all cell lines (Fig 4). These profiles indicate that the decreases in cell number are due, at least in part, to an arrest in the G1 phase of the cycle. Apoptosis was analyzed in control cultures and in Ad5CMV-p53-transduced cells by incubation with the vital DNA dye Hoechst 33342. Characteristic morphological patterns with fragmented DNA in the nucleus and the appearance of apoptotic bodies were not present in control NP-18 cells but were evident in transduced NP-18 cells at 36 hours postinfection (Fig 5). To quantify the percentage of apoptotic cells in cultures, Hoechst 33342 pulse-labeling experiments were performed (Table 2). The results revealed that the difference in the percentage of apoptotic cells between control and infected cultures in NP-29 cells was negligible (6.8 6 2.2% vs. 8.3 6 0.8%), indicating that the expression of exogenous p53 did not induce apoptosis in this wt p53 cell line. In contrast, a marked increase in the number of apoptotic cells was observed in the Ad5CMV-p53-transduced cultures of mutant p53 cell lines with respect to control cells. However, the percentages of apoptotic cells for the cell lines were different (1.0 6 0.3% vs. 11.0 6 2.1% of apoptotic cells in NP-9; 6.9 6 0.5% vs. 51.0 6 2.4% in NP-18). To determine whether the significantly higher apoptotic level achieved by the p53-transduced NP-18 cell line

432


Figure 3. Growth curves of pancreatic cell lines transduced with Ad5CMV-p53. The mean of cell counts (n 5 4) were plotted against the number of days after infection with 50 m.o.i. of each virus. (A) NP18; (B) NP-31; (C) NP-9; (D) NP-29. Mock infected cells (f), Ad-control infected cells (●), and Ad5CMV-p53 infected cells (). Bars, SD. Initial density: 10,000 cells/well.

Figure 4. Cell cycle profile of pancreatic cell lines transduced with Ad5CMV-p53. The percentage of cells in each cell cycle phase was analyzed by flow cytometry of PI-stained cells. All infections were performed at 50 m.o.i of virus. At 36 hours after treatment, triplicate dishes of each condition were analyzed.



433

Figure 5. Morphological analysis of apoptosis in Ad5CMV-p53-transduced cells. The chromatin structure of cells was visualized by nuclear staining of mock-infected NP-18 cells (A) and NP-18 cells infected with 50 m.o.i. of Ad5CMV-p53 (B) with Hoechst 33342. Arrows indicate apoptotic nuclei. Photomicrograph: 3800 magnification.

compared with transduced NP-9 cells could be due to the presence of wt p16, the transduction of p16 (25 m.o.i. of Ad5RSV-p16) before the introduction of wt p53 (25 m.o.i. Ad5CMV-p53) was assayed in the NP-9 cell line. The results (Table 2) clearly indicate a significant increase in apoptotic level due to the presence of wt p16 in the NP-9 cell line.

Effects of exogenous p53 on cell cycle and apoptosis after etoposide treatment We have also determined the effects on cell cycle and apoptosis induced by exogenous p53 expression in response to DNA damage produced by the DNA topoisomerase II inhibitor, etoposide. Cultures of all cell lines were infected with Ad5CMVp53 or maintained in culture medium and subsequently treated for 24 hours with 2 mM of etoposide. The cell cycle profiles of the noninfected cells showed important differences in response to etoposide treatment. The mutant p53 cells (NP-9, NP-18, and NP-31) showed a clear arrest in G2. In contrast, the wt p53 cells (NP-29) did not show arrest in G2 in response to the same treatment (Fig 6A). In this cell line, treatment with increasing concentrations of the drug induced increases in the levels of the endogenous p53 protein (Fig 7). In Ad5CMV-p53-transduced mutant p53 cells, etoposide did not induce the clear G2 arrest that was detectable in

untransduced cells, whereas no significant differences were observed between transduced and untransduced NP-29 cells (Fig 6A). Identical etoposide treatments were performed before introduction of the wt p53 gene. The changes observed in the cell cycle profiles of mutant p53 cells indicate that G2 blocking progressed toward endoreduplication processes in the course of 3 days posttreatment; these processes were abolished when wt p53 was introduced (Fig 6B). No significant increases in apoptosis were observed in either control or Ad5CMV-p53-infected NP-29 cells, in contrast to mutant p53 cells (Table 2). This result corroborates the inability of NP-29 to die by apoptosis in response to the overexpression of wt p53. DISCUSSION Only a few studies on wt p53 gene replacement in pancreatic cancer have been reported, and the results are controversial.30 –32 Thus, Kimura et al did not detect apoptosis after wt p53 transduction either in vitro or in vivo.31 In contrast, very recently Lang et al reported multiple phenotypic changes in one pancreatic cell line (Panc-1) as a consequence of wt p53 transduction.32 The availability of new pancreatic cell lines with marked differences in their p53 and p16 gene status allows us to

Table 2. Percentage of Apoptotic Cells as Determined by Hoechst 33342 Pulse Labeling NP-9

NP-9 1 p16*

NP-18

NP-29

Mock-infected Ad5-control Ad5CMV-p53

1.0 6 0.3† 1.6 6 0.8 11.0 6 2.1

2.0 6 0.4 4.6 6 1.9 46.5 6 3.8

6.9 6 0.5 11.8 6 2.1 51.0 6 2.4

6.8 6 2.2 5.6 6 0.9 8.3 6 0.8

Mock-infected 1 EP‡ Ad-control 1 EP Ad5CMV-p53 1 EP

1.7 6 0.6 1.8 6 0.7 17.1 6 1.6

ND§ ND ND

14.9 6 3.8 21.8 6 3.9 64.3 6 2.0

6.9 6 1.2 8.2 6 1.8 11.2 6 1.5

*NP-9 cells infected with Ad5RSV-p16 †Percentages calculated 36 hours after viral infection. ‡EP, treatment with 2 mM of etoposide for 24 hours. §Not determined


434


Figure 7. Western blot analysis of p53 expression in NP-29 cells treated with various concentrations of etoposide. Total extracts of NP-29 cells that had been treated for 24 hours with different concentrations of etoposide were analyzed using the monoclonal Abs anti-p53 and anti-actin. Relative amounts of p53 were quantified by densitometry.

Figure 6. Cell cycle profile of pancreatic cell lines transduced with Ad5CMV-p53 and treated with etoposide. Cell cycle profiles were obtained by flow cytometry analysis of PI-stained cells. All infections were performed at 50 m.o.i of virus. A: Cells treated with 2 mM of etoposide after the viral infections. At 36 hours after etoposide treatment, triplicate dishes of each condition were analyzed. B: Cells treated with 2 mM of etoposide before the viral infections. At 72 hours after etoposide treatment, triplicate dishes of each condition were analyzed.

study the influence of the genetic background of the target cell on the impact of p53 overexpression. With the aid of a recombinant reporter gene Ad, Ad5CMV-lacZ, we determined the ability of Ad to infect pancreatic tumor cells. Remarkably, the transduction efficiency achieved was very high, which makes possible successful Ad-mediated p53 transfer without the toxicity associated with high virus concentrations. When the p53 gene was delivered to pancreatic cell lines, the expressed p53 proved to be functional. First, immunocytochemical analysis revealed that the newly

expressed wt p53 is located mainly in the nucleus, like the endogenous protein. Marked morphological changes were observed, providing the first evidence that exogenous p53 is expressed in a functional way. Western blot analysis demonstrated that expression of the wt p53 protein was significantly increased in all infected cells and maintained long enough to achieve notorious effects on cell proliferation. The growth of the four cell lines when transduced with Ad5CMV-p53 was significantly lower than that of cells transduced with control Ad, indicating that this effect was due to the overexpression of p53. The high expression of the newly transduced protein had different effects on the growth of the infected cell lines, depending upon the p53 status of the cells. The wt p53 cells showed less inhibition of proliferation than the mutant p53 cells, as described in some types of tumor cells22,33 but not in others.34 The mechanism by which wt p53 protein inhibits growth may be related to arrest of the cell cycle or to apoptosis.9,10 The mechanism underlying the cytocidal effect induced by p53 appeared to be dependent upon the cell types and the availability of downstream effectors.20,35 When the cell cycle profile was analyzed, all cell lines experienced arrest in the G1 cell cycle phase, which is consistent with the increase observed in p21Cip1 expression as assessed by Western blot. These observations clearly indicate that exogenous p53 expression can activate the transcription of this CKI, which contributes significantly to drive the effects of p53 on the cell cycle,12,13 and consequently, to growth inhibition. However, there was great variation in the degree of apoptosis



induced by the overexpression of exogenous wt p53 in the four cell lines. NP-29, with functional endogenous p53, did not apoptotize when exogenous wt p53 was expressed. This may explain the low level of inhibition of this cell line. The lack of response to overexpression of p53 is consistent with the fact that p53 is still functional in NP-29 cells, thus indicating that the altered factors responsible for the tumoral behavior of these cells must be located downstream of the p53-dependent apoptotic pathway. In our cells, no significant differences in bax or bcl-2 were found when their levels of expression were analyzed by Western blotting in nontransduced cell extracts of the four cell lines (data not shown). Recently, it has been postulated that one of the main functions of the CKI p21Cip1 is to modulate the timing and intensity of the p53 apoptotic signals.36 The ratios between p21 and p53 concentrations were very similar, ruling out the possibility that this factor was responsible for the lack of apoptosis in NP-29 cells. The differences observed between different mutant p53 cell lines could be related to alterations in other elements involved in cell cycle checkpoints.37,38 Sandig et al have recently reported that the Ad-mediated overexpression of wt p16 collaborates very efficiently with p53 in the induction of apoptosis in human hepatocellular and breast carcinoma cells.14 Our results with p53 suggested that the endogenous levels of wt p16 could condition apoptosis in response to exogenous p53. Indeed, the wt p16 cell line (NP-18) showed a significantly higher apoptotic level (51% of apoptotic cells) than the NP-9 cell line (11% of apoptotic cells), which presents a deleted p16 gene. The increases in apoptotic level observed when both p16 and p53 genes are introduced in NP-9 cells (46.5% of apoptotic cells) support this contribution. Although several hypotheses have been advanced as to how the actions of p16 render tumor cells more susceptible to apoptotic stimuli,14,39 the current knowledge of apoptosis-controlling pathways is insufficient to fully explain the synergism observed between the actions of p16 and p53. It is well established that p53 is a key factor in the connection between DNA damage and the cellular apparatus responsible for death by apoptosis.40,41 Etoposide increases the apoptotic response and may cause G2 blocking depending upon p53 status, thus throwing light on this connection. In our cell lines, the expression of wt p53 prevented the accumulation of cells in G2 provoked by etoposide, irrespective of the endogenous or exogenous origin of wt p53. The abrogation of G2 blocking when the p53 function is introduced before the drug treatment could indicate that wt p53 can override the etoposide-induced blocking of G2, as demonstrated by other authors in different cell types.19 However, the possibility that p53 G1 arrest predominates over G2 arrest in wt p53-expressing cells cannot be ruled out. The results obtained when p53 is introduced after the etoposide treatments support the first hypothesis, indicating that p53 is allowing DNA-damaged cells to exit from G2. The Hoechst pulse experiments performed in control and infected NP-9 and NP-18 cells indicate that a


435

significant increase in apoptosis accompanies the abrogation of G2 block achieved by the combined treatments in mutant p53 cell lines. This behavior may be explained if cell cycle deregulation is considered as part of the common pathway of apoptosis, and checkpoints may be actively bypassed after the cell is committed to die. On the contrary, although the level of endogenous p53 in NP-29 rose as a consequence of etoposide treatment and G2 block was not observed, no variations in apoptotic rate were observed in either control or infected cultures, reinforcing the suggestion that this wt p53 tumor cell line either lacks other downstream functions in the p53-dependent apoptosis pathway or that these functions are altered. The overall results indicate that the status of tumor suppressor genes, such as p16 and p53, in pancreatic tumor cells and the possibility of the restoration of their function may drastically modify their response to different treatments. The expression of p53 alone may have important cytotoxic activity toward tumors that conserve unaltered the elements necessary to develop the p53 function in G1 checkpoint and apoptosis processes. Preliminary results obtained in our laboratory show that p53 transduction in cells previously implanted into the subcutaneous tissue of nude mice reduces tumor growth in vivo. Taken together, our data may have important implications for gene therapy with p53 or other treatments aimed at restoring p53 function in tumor cells to induce apoptosis. ACKNOWLEDGMENTS We thank Drs. J. Fueyo, J. L. Gelpı´, and O. Bachs for criticism and careful reading of the manuscript. We acknowledge the help received from Dr. J. Comas, who is responsible for the flow cytometry section of the Serveis Cientifico Te`cnics de la Universitat de Barcelona. We thank Robin Rycroft for editorial help. M.C. is the recipient of a predoctoral fellowship from Generalitat de Catalunya. This research was supported by Marato ´ del Ca`ncer-TV3 Grant 46/95 and by Comision Interministerial de Ciencia y Tecnologia Grants SAF97/0214 and SAF98/042.

REFERENCES 1. Wingo PA, Tong T, Bolden S. Cancer statistics, 1995. CA Cancer J. Clin. 1995;45:8 –30. 2. Lionetto R, Pugliese V, Bruzzi P, Rosso R. No standard treatment is available for advanced pancreatic cancer. Eur J Cancer. 1995;31A:882– 887. 3. Roth JA, Cristiano RJ. Gene therapy for cancer: what have we done and where are we going? J Natl Cancer Inst. 1997;89:21–39. 4. Bishop JM. Molecular themes in oncogenesis. Cell. 1991; 64:235–248. 5. King KL, Cidlowski JA. Cell cycle and apoptosis: common pathways to life and death. J Cell Biochem. 1995;8:175–185. 6. Culver KW, Blaese RM. Gene therapy for cancer. Trends Genet. 1994;10:174 –178. 7. Hollstein M, Sidranski D, Vogelstein B, Harris CC. p53 mutations in human cancers. Science. 1991;253:49 –53.

436


8. Harris CC. Structure and function of the p53 tumor suppressor gene: clues for rational cancer therapeutic strategies. J Natl Cancer Inst. 1996;88:1442–1455. 9. Ko LJ, Prives C. p53: puzzle and paradigm. Genes Dev. 1996;10:1054 –1072. 10. Haffner R, Oren M. Biochemical properties and biological effects of p53. Curr Opin Genet Dev. 1995;5:84 –90. 11. Nielsen LL, Maneval DC. p53 tumor suppressor gene therapy for cancer. Cancer Gene Ther. 1998;5:52– 63. 12. El-Deiry WS, Tokino K, Velculescu VE, et al. WAF1, a potential mediator of p53 tumor suppression. Cell. 1993; 75:817– 825. 13. Xiong Y, Hannon GJ, Zhang H, Casso D, Kobayashi R, Beach D. p21 is a universal inhibitor of cyclin kinases. Nature. 1993;366:701–704. 14. Sandig V, Brand K, Herwig S, Lukas K, Bartek J, Strauss M. Adenovirally transferred p16INK4/CDKN2 and p53 genes cooperate to induce apoptotic tumor cell death. Nat Med. 1997;3:313–318. 15. Serrano M, Lin AW, McCurrah ME, Beach D, Lowe SW. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16Ink4a. Cell. 1997;88: 593– 602. 16. Steward N, Hicks GG, Paraskevas E, Mowat M. Evidence for a second cell cycle block at G2/M by p53. Oncogene. 1995;10:109 –115. 17. Agarwal ML, Agarwal A, Taylor WR, Stark GR. p53 controls both the G2/M ant the G1 cell cycle checkpoints and mediates reversible growth arrest in human fibroblasts. Proc Natl Acad Sci USA. 1995;92:8493– 8497. 18. Guillouf C, Rosselli F, Krisnaraju K, Moustacci E, Hoffman B, Liebermann DA. p53 involvement in control of G2 exit of the cell cycle: role in the DNA-damage induced apoptosis. Oncogene. 1995;10:2263–2270. 19. Skladanowski A, Larsen AK. Expression of a wild-type p53 increases etoposide cytotoxicity in M1 myeloid leukemia cells by facilitated G2 to M transition: implications for gene therapy. Cancer Res. 1997;57:818 – 823. 20. Polyak K, Xia Y, Zweier JL, Kinzler KW, Vogelstein B. A model for p53-induced apoptosis. Nature. 1997;389: 300 –305. 21. Becker TC, Noel RJ, Coats WS, et al. Use of recombinant adenovirus for metabolic engineering of mammalian cells. Methods Cell Biol. 1994;5:161–189. 22. Zhang WW, Fang XF, Mazur W, French BA, Georges RN, Roth JA. High efficiency gene transfer and high-level expression of wild-type p53 in human lung cancer cells mediated by recombinant adenovirus. Cancer Gene Ther. 1994;1:4 –15. 23. Zabner J, Petersen DM, Puga AP, et al. Safety and efficacy of repetitive adenovirus-mediated transfer of CFTR cDNA to airway epithelia of primates and cotton rats. Nat Genet. 1994;6:75– 83. 24. Wills KN, Maneval DC, Menzel P, et al. Development and characterization of recombinant adenoviruses encoding human p53 for gene therapy of cancer. Hum Gene Ther. 1994;5:1079 –1088. 25. Wang J, Bucana CD, Roth JA, Zhang WW. Apoptosis induced in human osteosarcoma cells is one of the mech-

26. 27.

28.

29.

30.

31.

32.

33.

34.

35. 36.

37. 38. 39. 40. 41.

anisms for the cytocidal effect of Ad5CMV-p53. Cancer Gene Ther. 1995;2:9 –17. Sobol R, Scanlon KJ. Cancer gene therapy, clinical trials. Cancer Gene Ther. 1995;2:5– 6. Kagawa S, Fujiwara T, Hizuta A, et al. p53 overexpression overcomes p21waf1/cip1-mediated G1 arrest and induces apoptosis in human cancer cells. Oncogene. 1997;5:1903– 1909. Reyes G, Villanueva A, Garcı´a C, et al. Orthotopic xenografts of human pancreatic carcinomas acquire genetic aberrations during dissemination in nude mice. Cancer Res. 1996;56:5713–5719. Darzynkiewicz Z, Li X, Gong J, Hara S, Traganos. Analysis of cell death by flow cytometry. In: Rickwood D, Hames BD, eds. Cell Growth and Apoptosis: A Practical Approach, 1st ed. NY: Oxford University Press, 1995: 143–167. Favrot M, Coll JL, Louis N, Negoescu A. Cell death and cancer: replacement of apoptotic genes and inactivation of death suppressor genes in therapy. Gene Ther. 1998;5: 728 –739. Kimura M, Tagawa M, Takenaga K, et al. Inability to induce the alteration of p53-null human pancreatic carcinoma cells after the transduction of wild-type p53 gene. Anticancer Res. 1997;17:879 – 884. Lang D, Miknyoczki SJ, Huang L, Ruggeri BA. Stable reintroduction of wild-type p53 (MTmp53ts) causes the induction of apoptosis and neuroendocrine-like differentiation in human ductal pancreatic carcinoma cells Oncogene. 1998;16:1593–1602. Go ´mez-Manzano C, Fueyo J, Kyritsis P, et al. Adenovirusmediated transfer of the p53 gene produces rapid and generalized death of human glioma cells via apoptosis. Cancer Res. 1996;56:694 – 699. Liu TJ, El-Naggar K, McDonnell TJ, et al. Apoptosis induction mediated by wild-type p53 adenoviral gene transfer in squamous cell carcinoma of the head and neck. Cancer Res. 1995;55:3117–3122. Polyak K, Waldman T, He TC, Kinzler KW, Vogelstein B. Genetic determinants of p53-induced apoptosis and growth arrest. Genes Dev. 1996;10:1945–1952. Go ´mez-Manzano C, Fueyo J, Kyritsis AP, et al. Characterization of p53 and p21 functional interactions in glioma cells en route to apoptosis. J Natl Cancer Inst. 1997;89: 1036 –1044. Sherr CJ. Cancer cell cycles. Science. 1996;274:1672–1677. Levine AJ. p53, the cellular gatekeeper for growth and division. Cell. 1997;88:323–331. Lukas J, Parry D, Aagaard L, et al. Retinoblastomaprotein-dependent cell-cycle inhibition by the tumour suppressor p16. Nature. 1995;375:503–506. Lowe SW, Ruley HE, Jacks T, Housman DE. p53-dependent apoptosis modulates the cytotoxicity of anticancer agents. Cell. 1993;74:957–967. Fujiwara T, Grimm EA, Mukhoadhyay T, Zhang WW, Owe-Schaub B, Roth JA. Induction of chemosensitivity in human lung cancer cells in vivo by adenovirus-mediated transfer of the wild-type p53 gene. Cancer Res. 1994;54: 2287–2291.
