Understanding the mechanism of cisplatin (CDDP) action may improve therapeutic strategy for ovarian cancer. Although p53 and FLICE-like inhibitory protein (FLIP) are determinants of CDDP sensitivity in ovarian cancer, the interaction between p53 and FLIP remains poorly understood. Here, using two chemosensitive ovarian cancer cell lines and various molecular and cellular approaches, we show that CDDP induces p53-dependent FLIP ubiquitination and degradation, and apoptosis in vitro. Moreover, we showed that Itch (an E3 ligase) forms a complex with FLIP and p53 upon CDDP treatment. These results suggest that p53 facilitates FLIP down-regulation by CDDP-induced FLIP ubiquitination and proteasomal degradation. [Cancer Res 2008;68(12):4511–7]
- ubiquitin-proteasome pathway
- ovarian cancer
Cisplatin (CDDP) is the first-line anticancer agent for human ovarian cancer and is known to act in part by induction of apoptosis ( 1). In CDDP-sensitive cells, it also decreases cellular levels of FLICE-like inhibitory protein (FLIP), a FADD-binding suppressor of apoptosis, which is present as two splice variants: FLIPL (55 kDa) and FLIPS (28 kDa). FLIP contains two NH2-terminal death effector domains (DED), which prevent caspase-8 activation through DED-DED interaction ( 2). Although FLIP down-regulation is an important factor in CDDP-mediated apoptosis, its mechanism remained unclear ( 3).
The ubiquitin-proteasome pathway (UPP) could be involved in regulating FLIP. UPP is a major regulatory mechanism for intracellular protein level. This process is mediated by an E1 (Ub activating enzyme), E2 (Ub conjugase), and E3 (Ub ligase) complex. Typically, proteins modified by polyubiquitin chains are recognized and degraded by the proteasome ( 4). Importantly, Itch, a member of the HECT family of E3 ligases, interacts with FLIP and is believed to mediate its degradation as well as tumor necrosis factor α–induced apoptosis ( 5).
p53 is a transcription factor that regulates cell cycle progression, DNA repair, and apoptosis. p53 is maintained at low levels by its negative regulator, MDM2, which ubiquitinates p53, targeting it for proteasomal degradation ( 6). TP53 mutations are frequently observed in human ovarian cancer cells ( 7) and associated with decreased chemoresponsiveness ( 8).
In the present study, we investigated the involvement of Itch and p53 in CDDP-induced FLIP down-regulation. We showed that CDDP enhances FLIP-p53-Itch interaction, inducing FLIP ubiquitination and degradation in a p53- and Itch-dependent manner. These results suggest that the modulation of FLIP content may be an effective strategy to overcome chemoresistance in ovarian cancer.
Materials and Methods
Reagents. MG132, Lactacystin, and Epoxomicin were from Calbiochem (Ab-1). Cell Signaling, Inc., Ambion, and Dharmacon, Inc. provided siRNA for p53, Itch, and control, respectively. Ribojuice and Lipofectamine Plus were from Novagen and Invitrogen, respectively. HA-tagged ubiquitin was provided by Dr. Qiao Li (University of Ottawa, Canada). Adenoviral cDNAs were synthesized at the University of Ottawa Neuroscience Research Institute. Primary antibodies for immunoblots were mouse monoclonal anti-p53 (DO-1; Santa Cruz Biotechnologies), anti-FLIP (NF6; Alexis; Ab-1) and anti-Itch (BD Bioscience), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH; ab8245; Abcam), and anti-V5 (Invitrogen). Antibodies for precipitation were goat polyclonal anti-HA and anti-V5 (Bethyl Laboratories) and anti-p53 (C-19; Santa Cruz Biotechnologies), and for immunofluorescence were rabbit polyclonal anti-FLIP (Cell Signaling Technology) and anti-p53 (C-19; Santa Cruz Biotechnologies), and mouse monoclonal anti-Itch (BD Bioscience). Donkey secondary antibodies were from Jackson Immunoresearch.
Cell culture and adenovirus infection. Chemosensitive ovarian cancer cells (OV2008 and A2780s) were cultured as previously reported ( 8). Cells were infected with appropriate adenoviral constructs as indicated in the text. LacZ adenovirus was used to normalize each treatment group to same total adenoviral concentration. Adenovirus infection efficiency [multiplicity of infection (MOI), 5; 24 h] was >90% ( 8).
FLIP ubiquitination analysis. Cells were transfected with HA-ubiquitin ( 3). After 24 h, spent medium was replaced with fresh RPMI 1640 or DMEM F12 containing CDDP (0–10 μM; 0–9 h) and Epoxomicin. Cells were harvested for FLIP ubiquitination analyses ( 9). The cell pellet was resuspended in boiling 1% SDS in PBS, heated (100°C for 5 min), and suspended in 1% Triton X-100 in PBS (1:10). DNA was sheared by sonication, and after centrifugation (14,000 × g for 15 min), the supernatant [diluted with 1% Triton X-100 and 0.5% bovine serum albumin in PBS (1:1)] was incubated (overnight for 4°C) with primary antibody. The beads were washed (6 times with 1% Triton X-100 in PBS), and the precipitated proteins were immunoblotted.
RNA interference. Cells transfected for 24 h with p53 siRNA (100 nmol/L), Itch siRNA (100 nmol/L), or control siRNA (100 nmol/L; ref. 3) were treated with CDDP and harvested for subsequent analysis.
Western blotting. Western blotting (WB) was done as previously described ( 3). Membranes were incubated with anti-FLIP (NF60; 1:500), anti-GAPDH (1:20,000), anti-p53 (1:1,000), anti-Itch (1:500), or anti-V5 (1:5,000; overnight for 4°C), and with horseradish peroxidase–conjugated anti-rabbit or anti-mouse secondary antibody (1:1,000–10,000; 1 h at room temperature). Peroxidase activity was visualized with the enhanced chemiluminescent kit (Amersham Biosciences) and analyzed (Scion Image software; Scion, Inc.).
FLIP mRNA analysis. Relative differences in FLIP mRNA levels in experimental groups were determined semiquantitatively by reverse transcription-PCR (RT-PCR). Total RNA was reverse transcribed followed by PCR ( 10).
Assessment of apoptosis. Apoptosis was determined morphologically, using Hoechst 33258 nuclear stain (3). The counter was “blinded” to avoid experimental bias.
Immunoprecipitation. Immunoprecipitation IP was performed on whole cell lysates using anti-V5 and -p53 antibodies ( 11) and immunoblotted for p53, Itch, and V5.
Immunocytochemistry and confocal microscopy. Cultured OV2008 cells were fixed with methanol (10 min at −20°C). Nonspecific binding was blocked with 0.8% (w/v) serum albumin and 1% gelatin in PBS (30 min at room temperature; ref. 12). Cells were then incubated with anti-p53 (1:50), anti-FLIP (1:25), and anti-Itch (1:25; overnight at 4°C) and subsequently with secondary donkey-conjugated antibodies: anti-goat (Cy5), anti-mouse (Cy3), and anti-rabbit (FITC; 1:25; 2 h at room temperature). Cells were mounted with Vectashield (Vector). Cells incubated without primary antibodies served as negative controls. Fluorescence images (1,024 × 1,024 pixels) were acquired using a LSM 510 confocal laser-scanning microscope (Zeiss) with a ×63 oil immersion objective (numerical aperture, 1.4; ref. 13). Channel images were merged using Adobe Photoshop 7.01. Immunofluorescence profiles (arbitrary unit) were obtained using the Zeiss LSM 510 software (Zeiss; ref. 14).
Statistical analyses. All results are expressed as mean ± SE of at least three independent experiments. Data were analyzed by two-way ANOVA and Bonferroni posttest (PRISM software version 3.0; Graph Pad). Statistical significance was inferred at a P value of <0.05.
Results and Discussion
CDDP down-regulates FLIP protein content via FLIP proteasomal degradation. Our recent study shows that CDDP down-regulates FLIP content and induces apoptosis in ovarian cancer cells ( 3). To better understand how CDDP decreases FLIP level in ovarian cancer cells, two ovarian cancer cell lines (OV2008 and A2780s) were cultured with CDDP (0–10 μmol/L; 24 hours) and FLIP content was determined. CDDP decreases FLIPL and FLIPS levels in a concentration-dependent manner ( Fig. 1A ). FLIP down-regulation was associated with increased apoptosis ( Fig. 1A). The down-regulation of FLIP by CDDP (0–10 μmol/L; 0–24 hours) does not seem to be associated with any change in FLIPL and FLIPS mRNA abundance ( Fig. 1B, top).
To determine whether proteasomal degradation could be involved in CDDP-induced FLIP down-regulation, OV2008 cells were pretreated for 30 minutes with proteasome inhibitors MG132 (5 μmol/L), Lactacystin (10 μmol/L), or Epoxomicin (25 nmol/L) and cultured with CDDP for an additional 12 hours. Although the proteasome inhibitors displayed no effect on basal FLIP content, they significantly attenuated the decrease in FLIPL and FLIPS level induced by CDDP ( Fig. 1B, bottom), suggesting that proteasomal degradation may be responsible for the decreased FLIP content after CDDP challenge. Similarly, CDDP induces the proteasomal degradation of the antiapoptotic protein Xiap without affecting its mRNA abundance ( 15).
CDDP enhances FLIP-Itch interaction and FLIP ubiquitination. Because proteasomal degradation of most proteins is preceded by ubiquitination, we examined whether CDDP induces FLIPL and FLIPS ubiquitination. OV2008 and A2780s cells were transfected with HA-ubiquitin (2 μg; 24 hours); infected (MOI=25; 24 hours) with adenoviral V5-FLIPL, V5-FLIPS, or LacZ (as control); and treated with CDDP (0–10 μmol/L) in the presence of Epoxomicin (25 nmol/L) for 1.5 and 3 hours, respectively. Ubiquitinated FLIP was immunoprecipitated with anti–HA-ubiquitin and immunoblotted with anti–V5-FL or anti–V5-FS. Although FLIP ubiquitination was not detectable with nonspecific IgG or in cells infected with LacZ or HA-ub alone, CDDP enhanced FLIPL and FLIPS ubiquitination ( Fig. 1C–D, bottom), demonstrating that CDDP-induced FLIP degradation in ovarian cancer cells is associated with FLIP ubiquitination.
Because Itch possesses E3 ubiquitin ligase activity and is involved in protein ubiquitination ( 5), we determined whether Itch plays a role in CDDP-induced FLIP ubiquitination by targeting Itch with siRNA ( Fig. 1C–D, top). Although CDDP increased FLIPL and FLIPS ubiquitination in cells transfected with control siRNA, these responses were markedly suppressed by Itch depletion ( Fig. 1C–D, bottom), indicating a requirement for Itch in CDDP-induced FLIP ubiquitination. Although Itch has been shown to facilitate FLIPL degradation in hepatocytes ( 5), FLIPS ubiquitination was not examined nor was the effect of CDDP. The present study is the first to show that Itch plays a vital role in the ubiquitination of both FLIPL and FLIPS after CDDP treatment.
To determine whether Itch directly interacts with FLIP and if such an interaction leads to CDDP-dependent FLIP decrease, coprecipitation studies were carried out with OV2008 and A2780s cells transfected with HA-ubiquitin (2 μg; 24 hours) and subsequently infected (MOI=25; 24 hours) with adenoviral V5-FLIPL, V5-FLIPS, or LacZ and treated with CDDP (0–10 μmol/L) in the presence of Epoxomicin (25 nmol/L). As shown in Fig. 2A and C , FLIP-Itch interaction was not detected with nonspecific IgG or in cells infected with LacZ and was slightly detectable in absence of CDDP. This response was enhanced by CDDP after 1.5 hours for FLIPS ( Fig. 2C, top) and 3 hours for FLIPL ( Fig. 2A, top), and associated with an enhancement in CDDP-induced FLIPL and FLIPS ubiquitination ( Fig. 2B and D). These results suggest that FLIP-Itch binding may be crucial for FLIP regulation, and that CDDP decreases FLIP content by enhancing FLIP-Itch interaction, FLIP ubiquitination, and proteasomal degradation. Our findings are consistent with previous findings that FLIPS possesses a shorter half-life than FLIPL due to the presence of a unique c-terminus tail (Lys 193 and Lys195; ref. 9). Although FLIP ubiquitination has been detected in several cell subtypes ( 16– 18), the precise mechanism underlying FLIP ubiquitination was not determined. In the present study, we show for the first time that Itch binds to FLIPS and FLIPL to facilitate FLIP ubiquitination in a time-dependent manner.
p53 is required for CDDP-induced FLIP ubiquitination. Although some studies suggest p53 could control FLIP protein levels via the regulation of FLIP proteasomal degradation ( 19), direct evidence for such a conclusion was lacking. We thus investigated whether p53 is involved in the regulation of CDDP-induced FLIP degradation and apoptosis by transfecting ovarian cancer cells with p53 siRNA (0–100 nmol/L; 24 hours) before CDDP treatment (10 μmol/L; 24 hours). As expected, CDDP up-regulated p53, decreased FLIPL and FLIPS protein content, and induced apoptosis in cells transfected with control siRNA. In contrast, all these responses were attenuated in cells depleted of p53 ( Fig. 3A ). These results strongly suggest that p53 is an important regulator of FLIP and is required for CDDP-induced FLIP down-regulation and apoptosis. Our studies extend the observations from several studies demonstrating a direct correlation between p53 content and FLIP down-regulation ( 19, 20).
To examine if p53 interacts with FLIP in response to CDDP, we investigated FLIP-p53 interaction by IP (p53 and V5) and WB (V5 and p53, respectively). OV2008 and A2780s cells exhibited lower basal p53 levels, which were up-regulated by CDDP in a time-dependent manner ( Fig. 2A and C). Furthermore, although FLIP-p53 interaction was barely detectable in the absence of CDDP, this response was increased by CDDP at 1.5 hours for FLIPS ( Fig. 2C) and 3 hours for FLIPL ( Fig. 2A). CDDP also enhanced p53-Itch binding in a time-dependent manner ( Fig. 2A and C). These responses were associated with increased FLIPL and FLIPS ubiquitination in response to CDDP ( Fig. 2B and D). Our results show that CDDP increases p53-FLIP-Itch interaction, which may be important for FLIP ubiquitination and degradation.
We further elucidated whether CDDP-induced FLIP ubiquitination is p53 dependent by depleting p53. Our coprecipitation studies show that CDDP enhances p53-FLIP interaction ( Fig. 3B–C). These responses were associated with increased CDDP-induced FLIPL and FLIPS ubiquitination but attenuated after p53 down-regulation by p53 siRNA (100 nmol/L; 24 hours; Fig. 3B–C). Interestingly, p53 has been shown to facilitate FLIPL degradation and ubiquitination in colon cancer cells ( 19) and temperature-sensitive Germ cell-2 ( 20). These studies are consistent with our contention that, in response to CDDP, p53 binds to FLIP and Itch to facilitate FLIP ubiquitination and proteasomal degradation.
Our results clearly show that p53 plays an important role in CDDP-induced FLIP ubiquitination and proteasomal degradation, a phenomenon that involves Itch. To further test the above hypothesis, we examined FLIP-p53-Itch colocalization by triple immunolabeling on OV2008 cells treated with or without CDDP (0–2.5 μmol/L; 2 hours; Fig. 4 ). In control cells (DMSO; Fig. 4), FLIP-, Itch-, and p53-immunoreactivities (-IR) mostly displayed diffuse staining throughout the cytoplasm and at the cell membrane. Individual clusters of FLIP, Itch, and p53 along with a few triple colocalizations were observed in the cytoplasm. In contrast, cells treated with CDDP exhibited numerous FLIP, Itch, and p53 clusters at the cell membrane, with frequent triple colocalizations. Interestingly, CDDP induces p53 nucleus localization at 6 hours but not earlier (data not shown). Our findings are consistent with the contention that FLIP is recruited by Itch at the cell membrane in a p53-dependent manner. A similar recruitment phenomenon for FLIP by FADD at the Fas receptor has been reported and is believed to suppress caspase-8 activation (2). These results strongly argue in favor of a mechanism involving FLIP-p53-Itch colocalization after CDDP treatment. Moreover, p53, as a docking protein, could facilitate FLIP-Itch interaction, FLIP ubiquitination and degradation, and ultimately apoptosis.
In summary, we report that (a) CDDP down-regulates FLIP protein content via FLIP proteasomal degradation, (b) CDDP enhances FLIP-p53-Itch interaction and FLIP ubiquitination, and (c) p53 and Itch are required for CDDP-induced FLIP ubiquitination. These results are of high relevance for the modulation of FLIP as a possible modality in overcoming chemoresistance in ovarian cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Grant support: Canadian Research Society (#11181 to B.K. Tsang) and Canadian Institutes of Health Research (#MOP-79360 to R. Bergeron) and a New Investigator Award (R. Bergeron). M.R. Abedini was the recipient of a Ministry of Health and Medical Education Scholarship, Iran.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Qiao Li (University of Ottawa, Canada) and Dr. Kenneth Walsh (St. Elizabeth's Medical Centre, Boston, MA) for providing HA-tagged ubiquitin and adenoviral HA-tagged DN-Akt, respectively.
- Received February 23, 2008.
- Revision received April 22, 2008.
- Accepted April 23, 2008.
- ©2008 American Association for Cancer Research.