The present study was designed to investigate the efficacy of combination gene therapy using adenoviral vectors expressing gene products shown to possess apoptotic activity: E2F-1 (Ad-E2F-1) and a C-terminal deletion mutant of p21WAF1/cIP1 (Ad-p21-PCNA), on growth inhibition and apoptosis of human colon cancer cells in vitro and in vivo. Marked E2F-1 and p21-PCNA overexpression in response to adenovirus infection was evident by Western blot analysis. IC25 concentrations of each virus were used for each treatment in vitro to detect cooperative effects on cell death. Coexpression of E2F-1 and p21-PCNA resulted in an additive effect on cell death compared to infection with either virus alone. Cell cycle analysis, poly(ADP-ribose) polymerase (PARP) cleavage and analysis of cell morphology also revealed that coinfection with both Ad-E2F-1 and Ad-p21-PCNA enhanced cellular apoptosis compared to either virus alone. Interestingly, E2F-1 protein expression was markedly enhanced in the E2F-1/p21-PCNA adenovirus combination compared to Ad-E2F-1 infection alone. However, these same effects were not evident in cells coinfected with Ad-E2F-1 and an adenovirus expressing wild-type human p21WAF1/CIP1 (Ad-p21WT). The increase in E2F-1 expression with coexpression of E2F-1 and p21-PCNA was not a result of increased E2F-1 protein stability, but was related to increased transcriptional activity from the CMV promoter. Cell cycle analysis revealed G1 arrest 72 hours following single-gene therapy with either the wild-type or mutant p21, whereas increased accumulation of cells in G2/M phase was demonstrated in the E2F-1–overexpressing cells. In the combined therapies, E2F-1/p21-PCNA treatment still resulted in G1 arrest, but E2F-1 was able to counteract the G1 arrest when coinfected with p21WT. These results provide further evidence of the importance of the p21:PCNA-binding domain in mediating the complex cell cycle interaction between E2F-1 and p21. Simultaneous intratumoral injection of Ad-E2F-1 and Ad-p21-PCNA dramatically reduced tumor burden of SW620 xenografts compared to either treatment alone in our in vivo model but not in HT-29 colon cancer xenografts. When combined with Ad-p21-PCNA, E2F-1 adenovirus therapy resulted in approximately 95% decrease in tumor volume of SW620 tumor xenografts compared with controls (P<.05). In conclusion, although simultaneous delivery of E2F-1 and p21-PCNA transgenes results in increased E2F-1 expression and enhanced apoptosis of both SW620 and HT-29 colon cancer cells in vitro, this combination was only effective in the treatment of SW620 metastatic colon cancer in vivo. This may represent a potentially useful combination gene therapy strategy for metastatic colon cancer.
Adenovirus-mediated expression of the human e2f-1 gene has been shown to result in apoptotic cell death of a variety of cancer types in vitro and in vivo.1,2,3,4,5,6 The ability of E2F-1 to stimulate apoptosis through p53-dependent and -independent mechanisms is an appealing feature of e2f-1 gene–based therapy.1,4 Combining e2f-1 with a second gene that works cooperatively to enhance apoptotic tumor cell death may increase the efficacy of this treatment strategy.
Until the recent discovery of its apoptotic function, E2F-1 was known primarily for its role as a transcription factor involved in the coordination of cell cycle progression. By transactivating genes required for DNA replication (e.g., PCNA, DNA polymerase α, ribonucleotide reductase) and cell cycle control (e.g., CDK2 and 4, cyclins A, D, and E, and E2F-1 itself), E2F-1 regulates transition of the cell cycle from G1 to S phase.7,8,9 E2F-1–mediated cell cycle progression is affected by a number of cell cycle control proteins, including cyclin-dependent kinases (CDKs). In late G1 phase, the tumor suppressor pRb becomes hyperphosphorylated by the action of CDK4 and 6, thereby releasing and activating E2F-1.10,11 As the cell progresses through S phase, CDK2–cyclin A binds and phosphorylates E2F-1, thereby repressing the DNA-binding capacity and transcriptional activity of E2F-1.12,13,14,15
Cyclin-dependent kinase inhibitors (CKIs) also have the ability to target and regulate the activity of E2F-1. CKIs can inactivate E2F-1 in part by preventing the CDK-mediated phosphorylation of pRb, thus allowing pRb to bind and sequester E2F-1.16,17,18 The CKI, p21WAF1/cip1 (p21), has the capacity to arrest cells in G1 and G2 phases and, fittingly, has the ability to bind and inactivate the proliferating cell nuclear antigen (PCNA) and CDK2, both of which are E2F-1 transcriptional targets.19,20,21,22,23,24 In a reciprocal manner, E2F-1 can reverse the growth inhibitory effects of p21.25,26 Interestingly, a recent study demonstrated that E2F-1 directly transactivates p21, and subsequently, E2F-1 transcriptional activity is inhibited.27 Collectively, the evidence suggests a feedback loop mechanism between E2F-1 and p21 with apparently opposing activities that involve both direct and indirect interactions.26,28,29
The mechanism(s) of E2F-1–mediated apoptosis are not completely elucidated. E2F-1 has been shown to directly transactivate proapoptotic signals such as p14ARF, a tumor suppressor gene, and p73, a homolog of p53 that can transactivate similar target genes as p53.30,31,32 Interestingly, the apoptotic function of E2F-1 has been shown to be independent of its transcriptional function but dependent on its DNA-binding capability.33 In light of this finding, a mechanism involving E2F-1–mediated repression of antiapoptotic signals is emerging and is supported by the recent finding that E2F-1–induced apoptosis is associated with down-modulation of nuclear factor kappaB (NFkappaB) activity.34 Although numerous studies show that p21 inhibits apoptosis following a variety of apoptotic stimuli, the only data that exist regarding its effect on E2F-1–mediated apoptosis come from a study using glioma cells, which showed that p21 did not inhibit E2F-1–mediated cell death.35 The ability of p21 to bind and inhibit PCNA and CDK2 are required for the promotion of G1 and G2 cell cycle arrest;36 however, it is unclear if these functions of p21 are necessary for the inhibition of apoptosis.
A recent study by Zhang et al37 showed that treatment of cancer cells with the chemotherapeutic drug, camptothecin, resulted in caspase-mediated cleavage of p21 and conversion of the cancer cells from growth arrest to apoptosis. The same study further showed the cleaved p21 fragment lacked the C-terminus and could no longer bind PCNA. These results suggest a potential role for p21:PCNA interaction domain in the inhibition of apoptosis. In support of this theory, Prabhu et al38 demonstrated that adenovirus-mediated expression of a C-terminal deleted p21 gene product that is deficient in binding to PCNA (Ad-WAF1341, herein called Ad-p21-PCNA) inhibited cell cycle progression and induced apoptosis in colon adenocarcinoma cells.
In the present study, we sought to exploit the potential interaction of E2F-1 and p21-PCNA as a combination gene therapy strategy. Our results indicate that adenovirus-mediated coexpression of E2F-1 and p21-PCNA results in enhanced E2F-1 expression, with additive effects on cell growth inhibition and apoptosis in both colon cancer cell lines in vitro. In an animal model, however, this combination treatment enhanced tumor suppression only in the tumor xenografts established with the metastatic colon cancer cell line, SW620.
Materials and methods
Cells and culture conditions
The colon cancer cell lines, HT-29 (primary, mutant p53) and SW620 (metastatic, mutant p53) (ATCC, Rockville, MD) were routinely grown as monolayers in McCoy's 5A medium (Gibco, Grand Island, NY), supplemented with 10% heat-inactivated fetal bovine serum (Gibco)and penicillin (100 U/mL)/streptomycin (100 μg/mL). Both cell lines were cultured in a 5% CO2 incubator at 37°C and subcultured every 4–5 days. The transformed embryonic kidney cell line 293 was grown in minimum essential α-medium [medium with high glucose (4.6 g/L), supplemented with 10% FBS and 1% antibiotic/antimiotic].
Preparation of recombinant adenovirus
The adenovirus AdCMVE2F-1 (referred to herein as Ad-E2F-1) was a generous gift from Dr TJ Liu. AdCMVLacZ (referred to herein as Ad-LacZ) was a generous gift from Dr Brent French. Dr Walik S. El-Deiry generously provided the AdCMVWAF1 and AdCMVWAF1-341 viral vectors herein referred to as Ad-p21WT and Ad-p21-PCNA, respectively. For preparation of large virus stocks, 293 cells were infected at multiplicity of infection (MOI) 1–5 and harvested after the cytopathic effect became visible (24–48 hours). Cells were harvested and lysed in 1× PBS+2 containing 1% sucrose and virus aliquots were stored at −70°C. Titers were determined by plaque assay on 293 cells, as described previously.4,5
The cell lines infected with Ad-LacZ were assayed for β-galactosidase expression by the X-gal staining method as described previously.39 Briefly, 48 hours after infection, the tumor cells were washed with phosphate-buffered saline (PBS) and fixed in 2% (vol/vol) formaldehyde and 0.2% (vol/vol) glutaraldehyde in PBS, pH 7.4 for 5 minutes at 4°C. The cells were then washed and stained with X-gal solution (1 mg/mL 5-bromo-4-chloro-3-indolyl-b-galactopyranoside, 5 mM K4Fe(CN)6, 5 mM K3Fe(CN)6, 2 mM MgCl2 in PBS, pH 6.5) for 12–18 hours at 37°C. Blue staining of cell nuclei identified transduced cells. Mock-infected cells and cells transduced with other adenoviral vectors served as controls.
HT-29 and SW 620 cells were seeded in six-well plates at 1×105 total cells and incubated for 24 hours at 37°C. One six-well plate was infected with 200 μl media (mock); the remainder were infected with Ad-LacZ, Ad-E2F-1, Ad-p21WT, or Ad-p21-PCNA at increasing MOI (10, 25, 50, 100, and 150) for a period of 2 hours. Medium supplemented with 5% FBS and 1% penicillin/streptomycin was added and allowed to incubate for 5 days. After trypsinization, a cell count was obtained using a Coulter Counter ZM and trypan blue exclusion assays were performed. Approximate IC25 (concentration that inhibits 25% of viable cell growth compared to controls) values were calculated based on total viable cells and IC25 values were used for the remainder of the experiments.
Cells were seeded at 2.5×104 cells/mL in 75-mm flasks and allowed to incubate for 24 hours, at which time the cells were 35% confluent. Cells were treated with replication defective recombinant adenovirus vectors at IC25 concentrations for 2 hours and then fresh medium was added. Nonspecific viral toxicity was controlled by using LacZ adenovirus, thereby ensuring equivalent viral concentrations. Cells were harvested at 24-, 48-, 72- and 120-hour time points and prepared for various testing as outlined below.
Cell viability, cell count, and cell cycle analysis
Following viral infections, cells were harvested at 1-, 3-, and 5-day time points. For cell cycle analysis, 1×106 cells were washed twice with ice-cold PBS, fixed in 70% cold ethanol, and stored at 4°C until all time points were collected. Samples were then prepared for flow cytometric analysis as described previously.4,5 Flow cytometric analysis was performed using a FACscan (Becton Dickinson, Mountain View, CA). Data was analyzed using a CellFit cell-cycle analysis program (version 2.01.2).
Western blot analysis
For the detection of E2F-1 and p21 proteins, cells were lysed in lysis buffer as described previously.4,5 The protein concentrations were determined by the Bio-Rad protein assay (Bio-Rad, Hercules, CA). Thirty micrograms of total protein was loaded per lane and resolved in an 8% or 15% sodium dodecyl sulfate (SDS)–polyacrylamide gel. Proteins were transferred overnight to a polyvinylidene difluoride membrane (Amersham, Arlington Heights, IL) and immunoblotted using p21, E2F-1 mAbs and actin polyclonal antisera (Sigma, St. Louis, MO). Anti-mouse and anti-rabbit secondary antibodies conjugated to horseradish peroxidase were used and ECL reagents were used to detect the signals according to the manufacturer's instructions (Amersham).
Confirmation of apoptosis
Morphology associated with apoptotic cells was inspected by centrifuging 15,000 cells onto a microscope slide and staining with Wright/Giemsa stain. DNA fragmentation was confirmed by flow cytometry and apoptosis estimated by quantitation of the sub-G1 population. Poly(ADP-ribose) polymerase (PARP) cleavage assay has been show to be a sensitive method for detection of apoptosis.40,41,42 PARP cleavage assays were performed using a monoclonal mouse anti-PARP antibody (Calbiochem, Oncogene Research Products, La Jolla, CA) at a dilution of 1:100.
Protein stability assay
HT-29 and SW620 cells were seeded at 2.5×104 cells/mL in 75-mm flasks (three flasks per combination therapy) and allowed to incubate for 24 hours, at which time the cells were 35% confluent. Cells were treated with adenoviral vectors at IC25 concentrations for 2 hours and then fresh media was added. Five days following viral infection, cycloheximide was then added at a concentration of 100 μg/mL to each flask and then harvested at 0, 8, and 18 hours following drug treatment. Detection of E2F-1 protein by Western blot was carried out as described above.
Total RNA was extracted from cells 5 days following infection with the SV total RNA isolation system (Promega, Madison, WI). One microgram of total RNA was used in a two-enzyme reaction that synthesizes and amplifies cDNA from a specific target mRNA in one tube (Access RT-PCR system, Promega, Madison, WI). AMV reverse transcriptase is used for first-strand cDNA synthesis and Tf1 DNA Polymerase for second-strand cDNA synthesis in the presence of 50 pmol of transcript-specific primers for E2F-1 (forward primer: 5′-CTCGCAGCTCATCTC-3′ and reverse primer: 5′-ATGAGCTGGATGCCCTCAAG-3′). The RT-PCR conditions were established such that amplification of the cDNAs was dependent on the concentration of the corresponding RNA. RT-PCR of the housekeeping gene, β-actin, was also amplified to verify equal loading of RNA.
Transactivation of CMV promoter assay
To determine if adenoviral vectors had an effect on CMV promoter activity, we performed a dose–response assay in which SW620 cells were coinfected with a constant amount of Ad-LacZ (MOI 50), which was determined to result in β-galactosidase activity in the linear range, along with increasing MOI of Ad-E2F-1, Ad-p21WT, or Ad-p21-PCNA. Cells were allowed to incubate for 72 hours following treatment with adenoviruses to allow for β-galactosidase expression. The presence of β-galactosidase activity was determined by monitoring the hydrolysis of o-nitrophenyl-1-β-D-galactopyranoside (ONPG) quantitatively using a β-Gal assay kit (Invitrogen, Carlsbad, CA), as instructed by manufacturers. Briefly, cells were harvested 72 hours post infection with adenoviral combinations and then lyzed by resuspending in lysis buffer and subjecting to three freeze/thaw cycles in liquid nitrogen. Ten microliters cell lysate was then added to ONPG and 1× cleavage buffer and incubated at 37°C for 30 minutes. Following addition of 500 μl stop buffer, β-galactosidase activity in the supernatant was measured as absorbance at 420 nm. Protein concentration was then determined and specific activity of the lysate calculated.
In vivo combined treatment of transplanted human colon cancer cells
Tumors were formed by injecting 1×107 HT-29 and SW620 cells into athymic Balb/c nu/nu male mice (6–8 weeks, Charles River Laboratories, Wilmington, MA). Anesthesia was induced by intraperitoneal (i.p.) injection of ketamine (37.5 mg/kg) and xylazine (5 mg/kg). The cells were injected subcutaneously into the bilateral flanks of the mice. Ten days later, palpable tumors were randomized to be directly injected with Ad-E2F-1 (1×109 pfu), Ad-LacZ (1×109 pfu) alone, or a combination of Ad-E2F-1 (5×108 pfu), Ad-LacZ (5×108 pfu), or Ad-p21-PCNA (5×108 pfu), for a total MOI of 100. PBS was injected as a mock control. Subcutaneous injections were performed twice per week, with a total of six treatments. Each injection of purified virus was diluted in total volume of 100 μl PBS and administered in a single pass of a 27.5-gauge hypodermic needle using gentle, constant infusion pressure. Tumors were measured immediately before each treatment, and tumor size was estimated using the following formula: length (mm)×width (mm)/2. Animal survival was closely monitored and no signs of systemic toxicity were observed in animals receiving the recombinant adenoviruses. Animal experiments were performed in accordance with institutional guidelines and approved by the University Committee on the Use and Care of Animals.
Statistic analysis was performed by Student's t test. P<.05 was considered statistically significant.
Transduction efficiency of adenoviral vectors
To estimate the transduction efficiency of adenoviral vectors, the two human colon cell lines were infected with Ad-LacZ at MOI ranging from 25 to 150. At MOI of 20–75 or greater, over 85% of cells demonstrated nuclear staining for β-galactosidase in both cell lines. A dose-dependent increase in the intensity of β-galactosidase staining was demonstrated with increased MOI up to 100 (data not shown). No significant cytotoxicity was observed with MOI up to 150.
Dose response assays
Dose–response curves were generated for each of the viral vectors in both of the colon cancer cell lines using trypan blue exclusion assay to estimate total viable cells (data not shown). IC25 values were calculated and used in the remainder of the in vitro experiments to detect cooperative effects following the combination therapies. Approximate IC25 values were found to be: MOI 20, 20, and 10 for Ad-E2F-1, Ad-p21WT, and Ad-p21-PCNA, respectively, in both cell lines.
Adenovirus-mediated overexpression of E2F-1, p21WT, and p21-PCNA
Western blot analysis confirmed marked overexpression of E2F-1, p21WT, and p21-PCNA protein 5 days after infection in both the HT-29 and SW 620 cell lines (Fig 1). The E2F-1/p21WT combination therapy showed a significant decrease in E2F-1 protein expression compared to E2F-1 alone in the SW620 cell line but demonstrated similar levels in the HT-29 cells. Conversely, the E2F-1/p21-PCNA combination resulted in markedly increased E2F-1 expression compared to Ad-E2F-1 treatment alone in both cell lines. These results demonstrate a regulatory mechanism between p21 and E2F-1 in which the PCNA-binding domain of p21 has a significant effect on the expression level of E2F-1. Additionally, both cell lines showed a decrease in p21-PCNA protein levels in the E2F-1/p21-PCNA combination therapy compared to p21-PCNA gene therapy alone. Actin was used as an internal loading control.
Effect of adenoviral vector–mediated expression on cell viability and proliferation
Figure 2A shows viable cells (live cells/total cells) present 5 days post treatment. Cells infected with E2F-1/p21-PCNA combination therapy exhibited the greatest cell death (approximately 50% cell death in both cell lines) as assessed by trypan blue exclusion assay. Cells infected with Ad-E2F-1 or Ad-p21-PCNA alone exhibited 25% and 16% cell death, respectively. The E2F-1/p21-PCNA combination effect on cell death is greater than the single-therapy additive values (P<.05 for both cell lines), thus suggesting a cooperative effect on cell death. The combination of E2F-1/p21-PCNA resulted in significantly greater cell death compared to E2F-1/p21WT treatment (P<.05 for both cell lines).
Inhibition of cell proliferation was greatest following E2F-1/p21-PCNA combination therapy in both cell lines, as shown in Figure 2B, suggests a cooperative effect as well. In agreement with other studies,38 the overexpression of the p21-PCNA inhibited cell growth to a greater extent than overexpression of p21WT in both cell lines (P<.05). Although growth inhibition was seen following infection with all of the adenoviruses (excluding Ad-Lac-Z), maximal inhibition of cell proliferation was only achieved with the E2F-1/p21-PCNA combination. The E2F-1/p21-PCNA combination treatment resulted in significantly decreased cell proliferation compared to either treatment alone in both cell lines (P<.05). The combination of E2F-1/p21-PCNA resulted in significant inhibition of cell growth compared to E2F-1/p21WT treatment in HT-29 cells (P<.05), but not in SW620 cells. Notably, the cell growth in the SW620 cells infected with Ad-E2F-1/Ad-p21WT combination therapy exceeded that found in either of the separate therapies with Ad-E2F-1 or Ad-p21WT by 5 days post treatment, supporting the previously observed negative regulation between p21WT and E2F-1.25,26,27,28,29 This phenomenon was not observed in the HT-29 cells treated with E2F-1/p21WT combination therapy, which may be due to alterations in the p21/pRb/E2F-1 pathway in this cell line.
Effects of adenoviral vector–mediated expression on cell cycle
To further investigate the mechanisms of E2F-1, p21, and mutant p21 interactions, cell cycle distribution following the Ad-E2F-1, Ad-p21WT, and Ad-p21-PCNA infections alone or in combination was analyzed. Figure 3A shows control (mock and LacZ) cell cycle distribution in both cell lines at 3 days following treatment. Cells overexpressing p21WT or p21-PCNA exhibited an increase in the G1 population in both cell lines by 3 days post infection. However, the E2F-1–overexpressing cells demonstrated an increase in G2/M phase population with a concomitant decrease in G1 phase. In cells overexpressing both E2F-1 and p21WT, the p21-mediated G1 arrest was counteracted in both cell lines. However, E2F-1 could not inhibit the G1 arrest induced by p21-PCNA even though the protein expression level of E2F-1 was highest in these cells. Figure 3B illustrates the percentage of cells found in G1 phase 3 days following treatments. These data suggest that the ability of p21 to interact with PCNA is essential for E2F-1 counteraction of p21-induced G1 arrest.
Cooperative effect of E2F-1 and p21-PCNA on DNA fragmentation
By 5 days post infection, DNA fragmentation (subG1 population, Fig 4A was evident in the cells infected with E2F-1, alone and in combination with p21WT or p21-PCNA. Figure 4B graphically demonstrates the percentage of cells found in the subG1 population 5 days following gene transfer. The greatest subdiploid population was exhibited in the cells infected with the Ad-E2F-1/Ad-p21-PCNA combination therapy and was greater than the combined total of individual treatments with each virus. These findings suggest a cooperative effect between E2F-1 and p21-PCNA on apoptotic cell death. In both cell lines, E2F-1/p21-PCNA combination treatment resulted in significantly greater DNA fragmentation compared to either treatment alone in both cell lines (P<.05). The combination of E2F-1/p21-PCNA resulted in significantly greater DNA fragmentation compared to E2F-1/p21WT treatment (P<.05).
Cell death is due to apoptosis
We have demonstrated previously in a number of different cell lines that E2F-1–induced DNA fragmentation is due to apoptosis by using in situ TUNEL staining and Wright/Giemsa stain.4,5 To determine if the enhanced death and DNA fragmentation seen in the Ad-E2F-1/p21-PCNA combination therapy was due to apoptosis, Wright/Giemsa-stained cells were examined for morphological changes consistent with apoptosis including cytoplasmic blebbing, chromatin condensation, nuclear fragmentation, and the appearance of apoptotic bodies. To estimate the percentage of apoptotic cells in each sample, 1000 cells were counted and scored as either normal or apoptotic (condensed, intense chromatin staining, and nuclear fragmentation). Five days following infection with Ad-E2F-1, alone or in combination with p21 adenoviruses, evidence of apoptosis was observed (data not shown). The degree of the formation of apoptotic cells in each sample correlated with the extent of DNA fragmentation estimated in the flow cytometric analysis.
In addition to internucleosomal DNA fragmentation and the cellular morphological markers seen during apoptosis, a cascade of molecular events can serve as early biochemical markers, including activation of the caspasefamily of proteins resulting in PARP cleavage.40,41,42 Cleavage of the nuclear enzyme PARP by caspase-3 (Cpp32/YAMA/apopain) has been proposed as one of the earliest events in the execution phase of apoptosis.43,44,45,46 Western blot analysis confirmed cleavage of PARP by 5 days post treatment in SW620 cells treated with Ad-E2F-1 alone or in combination with both p21 adenoviruses (Fig 5B). Therefore, these data confirm apoptosis as the mechanism of cell death and the activation of the caspase cascade in SW620 cells.
The 89-kDa PARP apoptotic fragment was not detected in the HT-29 cell line. However, down-regulation of thefull-length (113-kDa) PARP protein was evident in all of the cells infected with Ad-E2F-1 alone or in combination with p21 adenoviruses. Although we could notdetect the apoptotic PARP fragment in the HT-29 cells, we believe apoptosis is the mechanism of death due to the morphological appearance of apoptotic bodies and DNA fragmentation.
Enhanced expression of E2F-1 protein in the E2F-1/p21-PCNA combination gene therapy is not due to altered protein stability
Studies show that p21 can affect protein stability of some ubiquinated proteins due to increased proteasome degradation24 and E2F-1 degradation is mediated by the ubiquitin–proteasome pathway.47 Therefore, we suspected that protein stability might be a factor related to the altered expression of E2F-1 in the combination therapies. E2F-1 protein half-life was measured by determining the level of protein at various time points following treatment with the protein synthesis inhibitor, cycloheximide. As shown in Figure 6, E2F-1 protein was similar at each time point following cycloheximide treatment (0, 8, 18 hours) in each sample and therefore protein stability does not appear to be a factor related to the increase or decrease of E2F-1 protein expression detected in the combination treatments.
RT-PCR analysis of E2F-1 mRNA expression
To investigate whether the enhanced protein expression of E2F-1 in the E2F-1/p21-PCNA combination therapy was associated with an increase in E2F-1 mRNA, we compared the abundance of E2F-1 mRNA by RT-PCR analysis following single and combination therapies with E2F-1. We reproducibly found increased expression of E2F-1 mRNA in the E2F-1/p21-PCNA–treated cells compared to the E2F-1– or E2F-1/p21WT–expressing cells (Fig 7). Although not strictly quantitative, these results imply that increased E2F-1 expression may be due to increased E2F-1 mRNA levels.
p21-PCNA enhances transactivation of CMV promoter
To investigate the ability of our adenoviral vectors to effect activity of the CMV promoter, coinfection with a constant amount of Ad-LacZ (MOI 50) and increasing amounts of Ad-E2F-1, Ad-p21WT, or Ad-p21-PCNA was performed and β-gal activity measured. Marked stimulation of the CMV promoter was observed (Fig 8) with increasing MOI of Ad-p21-PCNA. This increase occurred with or without coinfection of Ad-E2F-1. Coinfection with Ad-p21WT had little effect on β-gal activity with or without E2F-1 expression. These results suggest that overexpression of p21-PCNA (but not p21WT) activates CMV promoter transcription in these viral vectors.
In vivo treatment of human colon cancer cell xenografts
To determine whether our in vitro studies translated into a similar response in vivo, we evaluated the antitumor effects of combination gene therapy with E2F-1 and p21-PCNA adenoviral vectors. Single virus injections were performed at MOI 100 and combination virus injections were at MOI 50 each for a total of MOI 100. As shown in Figure 9, E2F-1 adenovirus alone inhibited tumor growth approximately 50% by day 16 in both HT-29 and SW620 cells. P21-PCNA adenovirus alone had no effect on tumor growth. When combined, E2F-1 and p21-PCNA adenovirus–mediated gene therapy caused approximately 95% decrease in tumor volume compared with LacZ control in the SW620 metastatic cell line, which was statistically significant (P<.05). The combined gene therapy of E2F-1 and p21-PCNA had no effect on tumor volume reduction in HT-29 xenografts compared to LacZ-treated tumors and, quite surprisingly, the presence of p21-PCNA completely inhibited any of the tumor suppressor effects seen with E2F-1 alone in vivo. In SW620 cells, the combination of E2F-1/p21-PCNA treatment significantly reduced tumor volume compared to E2F-1 treatment at MOI 100, as well as all other treatments (P<.05). Because the E2F-1/p21-PCNA combination treatment was not at all effective in inhibiting tumor growth in HT-29 cells, further experiments using the E2F-1 MOI 100 control were not performed for this cell line.
This study was designed to investigate the effectiveness of a combination therapy in vitro and in vivo using genes encoding the transcription factor, E2F-1, and a C-terminal deletion mutant of p21 (p21-PCNA) lacking the PCNA-binding domain. We used an adenoviral vector system to transfer the genes to two colon cancer cell lines, HT-29 (primary) and SW620 (metastatic), and analyzed the effect of this combination gene therapy on cell cycle regulation and apoptosis. We have shown previously that overexpression of E2F-1 efficiently triggers apoptotic cell death in a number of cell lines and involves caspase-3 activation and PARP cleavage.4,5 Here, for the first time, we show that coexpression of E2F-1 and p21-PCNA results in a cooperative effect on cellular apoptosis in vitro. However, this gene combination was only effective in tumor growth inhibition of SW620 (metastatic) colon cancer cell line in vivo. The results of this study also suggest a significant role for p21:PCNA interaction domain in the modulation of E2F-1–mediated regulation of cell cycle transition.
A growing body of evidence points to an opposing relationship between E2F-1 and p21 in both cell cycle control and mediation of apoptosis.16,17,18,19,20,21,22,23,24,25,26,27,28,29 Transciptional targets of E2F-1 include cdk2 and pcna genes favoring cell cycle progression.7,8,9 CDK2 is the catalytic partner of two cyclins involved in S phase progression, cyclin A and E, whereas PCNA plays an essential role in DNA replication as an auxiliary protein of Polδ, the enzyme responsible for the replication of chromosomal DNA. In vitro studies have shown that p21 inhibits processive DNA synthesis through a PCNA-dependent mechanism48 and the formation of cyclin–CDK complexes.19,20,21,22,23,24 Recently, much attention has been focused on the CDK2- and PCNA-binding domains of p21 and their importance in facilitating G1 and G2 cell cycle arrest. Experiments utilizing N- or C-terminal deletion mutants of p21, or p21 mutants containing point mutations in the CDK2- or PCNA-binding domains provide consistent data that show if p21 has lost the ability to translocate to the nucleus, it can no longer inhibit cell cycle progression.36,37,51 This observation is true even with the retention of p21's ability to inhibit CDK2 activity,36,51 which appears to be a requirement for inhibition of cell cycle progression in addition to nuclear localization. In the present study, we used an adenoviral vector expressing a C-terminal deletion mutant p21 gene (Ad-p21-PCNA). Characterization of the p21-PCNA protein reveals that it does not bind PCNA but still maintains the ability to inhibit CDK2 activity and translocate to the nucleus.38 Figure 3, A and B demonstrate that overexpression of p21-PCNA induced a G1 arrest that was equivalent or greater than the p21WT-induced G1 arrest. Thus, in agreement with previous studies by Prabhu et al,38 the binding and inactivation of PCNA by p21 is not required for the induction of a G1 arrest. Furthermore, as shown in Figure 3, A and B, when E2F-1 was coexpressed with p21WT, the cells accumulated in G2/M, but when coexpressed with p21-PCNA, the cells still accumulated in G1 phase. Therefore, our results also imply that the PCNA interaction domain of p21 is vital for the ability of E2F-1 to counteract a p21-induced G1 arrest.
In addition to, but apparently independent of its transcriptional function, E2F-1 possesses an apoptotic function.33 According to several studies, p21 has been shown to oppose apoptosis following a variety of apoptotic stimuli.40,50,51,52 However, little data exist regarding the effect of p21 on E2F-1–mediated apoptosis. Even less is known about the function of the CDK2- and PCNA-binding domains of p21 in the inhibition of an apoptotic signal, although the latter appears to be more important. Evidence from several studies indicate that there are cytoplasmic factors that p21 can deactivate to inhibit apoptosis including ASK1 (apoptosis signal–regulating kinase 1) and caspase-3.47,49 Interestingly, recent data indicate that treatment with the chemotherapeutic drug, camptothecin, results in caspase-mediated cleavage of p21 resulting in the production of a 15-kDa N-terminal fragment of p21 that does not bind to PCNA.37 The cleavage of p21 results in the conversion of cancer cells from growth arrest to apoptosis and is thought to be a critical step in the acceleration of chemotherapy-induced tumor cell death.37 This evidence suggests that the ability of p21 to bind to PCNA may be a requirement for p21-mediated inhibition of apoptosis. Our results support these studies. As demonstrated in Figure 4, wild-type p21 inhibited apoptosis induced by E2F-1 in the SW620 cell line, although it did not affect the level of E2F-1–mediated apoptosis in the HT-29 cells. However, p21-PCNA did notinhibit E2F-1–induced apoptosis but in fact, enhanced the apoptotic effect in both cell lines. These data suggest that the ability of p21 to bind and inactivate PCNA may play an important role in the inhibition of E2F-1–mediated apoptosis, and that removal of the PCNA-binding domain of p21 augments E2F-1–induced apoptosis. In addition, because apoptosis occurred even though the E2F-1/p21-PCNA–overexpressing cells arrested in G1, the mere inhibition of cell growth is not sufficient to inhibit apoptosis.
It is not clear why the p21 C-terminal deletion mutant enhanced E2F-1–mediated apoptosis. One explanation may be the increased levels of E2F-1 protein detected in the cells coinfected with Ad-E2F-1 and Ad-p21-PCNA. Because the adenoviruses used in this study contain a constitutive CMV promoter, we expected the expression levels of E2F-1 to be relatively similar in all therapy combinations. However, we consistently found significantly higher levels of E2F-1 protein in the cells overexpressing both E2F-1 and p21-PCNA. Further, in SW620 cells infected with Ad-E2F-1 and Ad-p21WT, an actual decrease in E2F-1 expression was detected. Because p21 has been shown to effect proteasome degradation,24 we suspected that protein stability might be a factor contributing to the altered levels of E2F-1. However, our protein stability experiments suggest otherwise. As demonstrated in Figure 6, protein stability of E2F-1 was similar at each time point in all combination treatments up to 18 hours following cycloheximide treatment.
Next we performed RT-PCR experiments to determine if the increased E2F-1 protein expression was associated with increased levels of E2F-1 mRNA. RT-PCR analysis demonstrated that E2F-1 mRNA levels are greatest in cells overexpressing both E2F-1 and p21-PCNA (Fig 9), and lowest in the E2F-1/p21WT–treated cells. Although only semiquantitative at best, this finding was reproducible and the RT-PCR results correspond well with protein expression levels (Fig 1). Furthermore, β-galactosidase activity assays confirmed the ability of p21-PCNA to up-regulate CMV promoter activity and accounts for the increase in E2F-1 mRNA and protein levels (Fig 7).
To determine if our in vitro studies translated into a similar response in vivo, we evaluated the antitumor effects of combination gene therapy with E2F-1 and p21-PCNA adenoviral vectors. It has been well documented that adenovirus-mediated E2F-1 has an antitumor efficacy in a variety of tumors attributable to both the E2F-1 tumor suppressor gene and the adenovirus delivery vector. In the animal model of human colon cancer, E2F-1 adenovirus alone inhibited tumor growth approximately 50% in both HT-29 and SW620 cells. P21-PCNA adenovirus alone had no effect on tumor growth. There was a significantly enhanced response when these modalities were combined to treat the SW620 metastatic cell line in vivo that was greater than the antitumoral effect of doubling the amount of Ad-E2F-1. However, Ad-p21-PCNA did not enhance E2F-1–mediated tumor suppression in the HT-29 xenografts but in fact, totally inhibited any antitumor effect demonstrated with Ad-E2F-1 alone. These data offer the possibility of enhanced antitumor activity with combined gene therapy in colon cancer metastases.
In conclusion, the coinfection of Ad-E2F-1 and Ad-p21-PCNA enhances E2F-1–mediated apoptosis and cell growth inhibition in colorectal cancer cells in vitro. This effect of Ad-p21-PCNA is associated with increased expression of E2F-1 when coinfection is performed, perhaps as a result of stimulation of the CMV promoter transcriptional activity by Ad-p21-PCNA. In vivo studies demonstrate that this combination is only effective against the xenografts established from the metastatic colon cancer cell line, SW620. These findings provide a potentially promising approach for the treatment of advanced metastatic colorectal cancer. In addition to the therapeutic implication of these results, this study also provides further evidence of the importance of the p21:PCNA-binding domain in mediating the complex cell cycle interaction between E2F-1 and p21.
Hunt KK, Deng J, Liu TJ et al. Adenovirus-mediated over expression of the transcription factor E2F-1 induces apoptosis in human breast and ovarian carcinoma cell lines and does not require p53 Cancer Res 1997 57: 4722–4726
Shan B, Lee WH . Deregulated expression of E2F-1 induces S-phase entry and leads to apoptosis Mol Cell Biol 1994 14: 8166–8173
Johnson DG, Schwartz JK, Cress WD, Nevins JR . E2F-1 overexpression in quiescent fibroblasts leads to induction of cellular DNA synthesis and apoptosis J Virol 1995 69: 2491–2500
Dong YB, Yang HL, Elliott MJ et al. Adenovirus-mediated E2F-1 gene transfer efficiently induces apoptosis in melanoma cells Cancer 1999 86: 2021–2033
Yang HL, Dong YB, Elliott MJ et al. Adenovirus-mediated E2F-1 gene transfer inhibits MDM2 expression and efficiently induces apoptosis in MDM2-overexpressing tumor cells Clin Cancer Res 1999 5: 2242–2250
Fueyo J, Comez-Manzano C, Yung WKA et al. Overexpression of E2F-1 in glioma triggers apoptosis and suppresses tumor growth in vitro and in vivo Nat Med 1998 4: 685–690
Nevins JW . E2F: a link between the Rb tumor suppressor protein and viral oncoproteins Science 1992 258: 424–429
Sherr CJ . Cancer cell cycles Science 1996 274: 1672–1677
Black AR, Azizkhan-Clifford J . Regulation of E2F: a family of transcription factors involved in proliferation control Gene 1999 237: 281–302
Weinberg RA . The retinoblastoma protein and cell cycle control Cell 1995 81: 323–330
Sherr CJ, Roberts JM . Inhibitors of mammalian G1 cyclin-dependent kinases Genes Dev 1995 9: 1149–1246
Krek W, Ewen ME, Shirodkar S, Arany Z, Kaelin WG Jr, Livingston DM . Negative regulation of the growth-promoting transcription factor E2F-1 by a stably bound cyclin A–dependent protein kinase Cell 1994 78: 161–172
Dynlacht BD, Flores O, Lees JA, Harlow E . Differential regulation of E2F transactivation by cyclin/cdk2 complexes Genes Dev 1994 8: 1772–1786
Xu M, Sheppard KA, Peng CY, Yee AS, Piwnica-Worms H . Cyclin A/CDK2 binds directly to E2F-1 and inhibits the DNA-binding activity of E2F-1/DP-1 by phosphorylation Mol Cell Biol 1994 14: 8420–8431
Kitagawa M, Higashi H, Suzuki-Takahashi I et al. Phosphorylation of E2F-1 by cyclin A-cdk2 Oncogene 1995 10: 229–236
Lukas J . Retinoblastoma-protein-dependent cell-cycle inhibition by the tumor suppressor p16 Nature 1995 375: 503–506
Chaung LS, Ian HI, Koh TW, Ng HH, Xu G, Li BF . Human DNA–(cytosine-5) methyltransferase–PCNA complex as a target for p21WAF1 Science 1997 277: 1996–2000
Xiong Y, Zhang H, Beach D . D type cyclins associate with multiple protein kinases and the DNA replication and repair factor, PCNA Cell 1992 71: 504–514
Serrano M, Hannon G, Beach D . A new regulatory motif in cell-cycle control causing specific inhibition of cyclin D/CDK4 Nature 1993 366: 704–708
Adams PD, Sellers WR, Sharma SK, Wu AD, Nalin CM, Kaelin WG Jr . Identification of a cyclin-cdk2 recognition motif present in substrates and p21-like cyclin-dependent kinase inhibitors Mol Cell Biol 1996 16: 6623–6633
Chen J, Jackson PK, Kirschner MW, Dutta A . Separate domains of p21 involved in the inhibition of Cdk kinase and PCNA Nature 1995 374: 386–388
Fotedar A, Cannella D, Fitzgerald PJ et al. Role of cyclin A–dependent kinase in DNA replication in human S phase cell extracts Biol Chem 1994 271: 31627–31637
Harper JW, Elledge SJ, Keyomarsi K et al. Inhibition of cyclin-dependent kinases by p21 Mol Biol Cell 1995 6: 387–400
Rousseau D, Cannella D, Boulaire J, Fitzgeral P, Fotedar A, Fotedar R . Growth inhibition by CDK–cyclin and PCNA binding domains of p21 occurs by distinct mechanisms and is regulated by ubiquitin–proteasome pathway Oncogene 1999 18: 4313–4325
Sekiguchi T, Hunter T . Induction of growth arrest and cell death by overexpression of the cyclin–CDK inhibitor p21 in hamster BHK21 cells Oncogene 1998 16: 369–380
Dimri GP, Nakanishi M, Desprez PY, Smith JR, Campisi J . Inhibition of E2F activity by the cyclin-dependent protein kinase inhibitor p21 in cells expressing or lacking a functional retinoblastoma protein Mol Cell Biol 1996 16: 2987–2997
Hiyama H, Iavarone A, Reeves S . Regulation of the cdk inhibitor p21 gene during cell cycle progression is under the control of the transcription factor E2F Oncogene 1998 16: 1513–1523
Shiyanov P, Bagchi S, Adami G et al. p21 disrupts the interaction between cdk2 and the E2F–p130 complex Mol Cell Biol 1996 16: 737–744
Delavaine L, La Thangue NB . Control of E2F activity by p21Waf1/Cip1 Oncogene 1999 18: 5381–5392
Bates S, Phillips AC, Clark PA et al. p14ARF links the tumour suppressors RB and p53 Nature 1998 395: 124–125
Sherr C . Tumor surveillance via the ARF–p53 pathway Genes Dev 1998 12: 2984–2991
Irwin M, Marin M, Phillips A et al. Role for the p53 homologue p73 in E2F-1–induced apoptosis Nature 2000 407: 645–648
Phillips AC, Bates S, Ryan KM, Helin K, Vousden KH . Induction of DNA synthesis and apoptosis are separable functions of E2F-1 Genes Dev 1997 11: 1853–1863
Phillps A, Ernst M, Bates S, Rice N, Vousden K . E2F-1 potentiates cell death by blocking antiapoptotic signaling pathways Mol Cell 1999 4: 771–781
Gomez-Manzano C, Fueyo J, Alamaeda F, Kyritsis AP, Yung WK . Gene therapy for gliomas: p53 and E2F-1 proteins and the target of apoptosis Bioorg Med Chem Lett 1999 3: 81–85
Cayrol C, Knibiehler M, Ducommun B . p21 binding to PCNA causes G1 and G2 cell cycle arrest in p53-deficient cells Oncogene 1998 16: 311–320
Zhang Y, Fujita N, Tsuruo T . Caspase-mediated cleavage of p21Waf1/Cip1 converts cancer cells from growth arrest to undergoing apoptosis Oncogene 1999 18: 1131–1138
Prabhu NS, Blagosklonny MV, Zeng YZ, Wu GS, Waldman T, El-Deiry WS . Suppression of cancer cell growth by adenovirus expressing p21WAF1/CIP1 deficient in PCNA interaction Clin Cancer Res 1996 2: 1221–1229
Adams DO, Edelson PH, Korren MS eds. Methods for studying mononuclear phagocytes: histochemical stains for macrophages in cell smears and tissue section: β-galactosidase, acid phosphatase, nonspecific esterase, succinate dehydrogenase, and cytochrome oxidase New York: Academic Press 1981 375–396
Polyak K, Waldman T, He TC, Kinzler KW, Volgelstein B . Genetic determinants of p53-induced apoptosis and growth arrest Genes Dev 1996 10: 1945–1952
Cohen GM . Caspases: the executioners of apoptosis Biochem J 1997 326: 1
Martin SJ, Green DR . Protease activation during apoptosis: death by a thousand cuts? Cell 1995 82: 349
Nicholson DW, Ali A, Thornberry NA et al. Identification and inhibition of the ICE/CED-3 protease necessary for mammalian apoptosis Nature 1995 376: 37
Kaufmann SH, Desnoyers S, Ottaviano Y, Davidson NE, Poirier GG . Specific proteolytic cleavage of poly(ADP-ribose) polymerase: an early marker of chemotherapy-induced apoptosis Cancer Res 1993 53: 3976
Tewari M, Quan LT, O'Rourke K et al. Salvesen that cleaves the death substrate poly(ADP-ribose) polymerase Cell 1995 Jun 2 81: 801–809
Lazebnik YA, Kaufmann SH, Desnoyers S, Poirier GG, Earnshaw WC . Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE Nature 1994 371: 346
Hateboer G, Kerkhoven RM, Shvarts A, Bernards R, Beijersbergen RL . Degradation of E2F by the ubiquitin–proteasome pathway: regulation by retinoblastoma family proteins and adenovirus transforming proteins Genes Dev 1996 10: 2960–2970
Kelman Z . PCNA: structure, functions and interactions Oncogene 1997 14: 629–640
Balbin M, Hannon Gj, Pendas AM et al. J Biol Chem 1996 271: 15782–15786
Lu Y, Yamagishi N, Yagi T, Takebe H . Mutated p21WAF1/cIP1/SDI1 lacking CDK-inhibitory activity fails to prevent apoptosis in human colorectal carcinoma cells Oncogene 1998 16: 705–712
Asada M, Yamada T, Ichijo H et al. Apoptosis inhibitory activity of cytoplasmic p21Cip1/WAF1 in monocytic differentiation EMBO J 1999 18: 1223–1234
Suzuki A, Tsutomi Y, Miura M, Akahane K . Caspase 3 inactivation to suppress Fas-mediated apoptosis: identification of binding domain with p21 and ILP and inactivation machinery by p21 Oncogene 1999 18: 1239–1244
We are grateful to Dr Brent French for providing the Ad5CMV-LacZ, Dr TJ Liu for providing the Ad5CMV-E2F-1, and Dr Wafik S El-Diery for providing both p21 adenoviral vectors AdCMV-WAF1 and AdCMV-WAF1-341. We thank Sherri Matthews for expert assistance with manuscript preparation. This study was supported by Grant 96-55 from the American Cancer Society, Grant 96-46 from the Alliant Community Trust Fund, the Mary and Mason Rudd Foundation Award, and the Center for Advanced Surgical Technologies (CAST) of Norton Hospital.
About this article
The Journal of Pathology (2018)
Bioscience Hypotheses (2009)
Developmental Neurobiology (2008)
Differential effects of cell cycle regulatory protein p21WAF1/Cip1 on apoptosis and sensitivity to cancer chemotherapy
Drug Resistance Updates (2003)
Journal of Biological Chemistry (2002)