Prostaglandin J2 Induces Death Receptor 5 Expression Through Mrna Stabilization Independently Ofppar ; and Potentiates TRAIL-Induced Apoptosis

1827

12,14 15-Deoxy-# -prostaglandin J2 induces death receptor 5 expression through mRNA stabilization independently ofPPAR ; and potentiates TRAIL-induced apoptosis

Susumu Nakata,1 Tatsushi Yoshida,1 and the induction of apoptotic cell death enhanced by 1,3 1 Takumi Shiraishi, Mano Horinaka, cotreatment with 15d-PGJ2 and TRAIL. Moreover, a double- Junji Kouhara,1,2 Miki Wakada,1 stranded small interfering RNA targeting DR5 gene, which and Toshiyuki Sakai1 suppressed DR5 up-regulation by 15d-PGJ2, significantly attenuated apoptosis induced by cotreatment with 15d-PGJ2 1 Department of Molecular-Targeting Cancer Prevention and and TRAIL. These results suggest that 15d-PGJ2 is a potent 2Division of Digestive Surgery, Department of Surgery, sensitizer of TRAIL-mediated cancer therapeutics through 3 Graduate School of Medical Science and Department of Urology, DR5 up-regulation. [Mol Cancer Ther 2006;5(7):1827–35] Kyoto Prefectural University of Medicine, Kyoto, Japan Introduction Abstract Prostaglandins are a family of naturally occurring cyclic #12,14 15-Deoxy- -prostaglandin J2 (15d-PGJ2), the termi- 20-carbon fatty acids that are synthesized mainly from nal derivative of the PGJ series, is emerging as a potent arachidonate released from membrane phospholipids by antineoplastic agent among cyclopentenone prostaglan- the action of phospholipases (1). Cyclopentenone prosta- dins derivatives and also known as the endogenous ligand glandins are potent bioactive molecules that possess anti- of peroxisome proliferator-activated receptor ; (PPAR;). inflammatory (2) and antiviral (3) activity. In addition, On the other hand, death receptor 5 (DR5) is a specific cyclopentenone prostaglandins have been shown to induce receptor for tumor necrosis factor–related apoptosis- cell growth arrest (4–6) and apoptosis (1) in several neo- inducing ligand (TRAIL), which is one of the most pro- plastic cell types. In particular, the terminal metabolite of 12,14 mising candidates for new cancer therapeutics. Here, we prostaglandin J2,15-deoxy-D -prostaglandin J2 (15d-PGJ2), report that 15d-PGJ2 induces DR5 expression at both is emerging as the most potent antineoplastic agent of mRNA and protein levels, resulting in the synergistic this class of prostaglandins (4). Antineoplastic activity of sensitization of TRAIL-induced apoptosis in human neo- 15d-PGJ2 has been reported both in vitro and in vivo in a plastic cells, such as Jurkat human leukemia cells or PC3 multiplicity of tissues, including the prostate (7), hemato- human prostate cancer cells. 15d-PGJ2 significantly in- poietic (8), breast (9), colon (10), gastric (11), brain (12). In creased DR5 mRNA stability, whereas it did not activate most types of cancer, 15d-PGJ2 inhibits tumor cell prolif- DR5 promoter activity. Synthetic PPAR; agonists, such as eration and induces apoptosis; however, the mechanism pioglitazone or rosiglitazone, did not mimic the DR5- for antineoplastic activity has not been fully elucidated. ; inducing effects of 15d-PGJ2, and a potent PPAR Peroxisome proliferator-activated receptor g (PPARg), a inhibitor GW9662 failed to block DR5 induction by 15d- member of the nuclear receptor superfamily, functions as a ; PGJ2, suggesting PPAR -independent mechanisms. transcription factor mediating ligand-dependent transcrip- Cotreatment with 15d-PGJ2 and TRAIL enhanced the tional regulation. 15d-PGJ2 was shown to be a natural endo- sequential activation of caspase-8, caspase-10, caspase- genous ligand for PPARg and to exert some of its effects by 9, caspase-3, and Bid. DR5/Fc chimera protein, zVAD-fmk binding to PPARg (13, 14). More recent evidence, however, pancaspase inhibitor, and caspase-8 inhibitor efficiently indicates that 15d-PGJ2 can also act independently of blocked the activation of these apoptotic signal mediators PPARg activation (1). Tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) induces apoptosis selectively in cancer cells in vitro and in vivo while sparing most normal cells Received 1/13/06; revised 5/12/06; accepted 5/16/06. (15, 16). Therefore, TRAIL is one of the most promising Grant support: Japanese Ministry of Education, Culture, Sports, Science new candidates for cancer therapeutics and is currently and Technology. undergoing clinical trials. The costs of publication of this article were defrayed in part by the Death receptor (DR) 5 (also called TRAIL-R2, Apo2, payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to TRICK2, or KILLER) is a member of the tumor necrosis indicate this fact. factor receptor family and is a receptor for TRAIL (17, 18). Requests for reprints: Toshiyuki Sakai, Department of Molecular-Targeting Apo2L/TRAIL triggers apoptosis through binding to the Cancer Prevention, Graduate School of Medical Science, Kyoto Prefectural DRs DR4 (19) and/or DR5 (20). These receptors contain a University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku, Kyoto 602-8566, Japan. Phone: 81-75-251-5339; Fax: 81-75-241-0792. cytoplasmic death domain that recruits adaptor molecules E-mail: [email protected] involved in caspase activation (21). The resulting active Copyright C 2006 American Association for Cancer Research. caspase-8 (21) or caspase-10 (22) cleaves and activates doi:10.1158/1535-7163.MCT-06-0023 effector caspases, such as caspase-3, caspase-6, and

Mol Cancer Ther 2006;5(7). July 2006

caspase-7. On the other hand, Bid, a proapoptotic Bcl-2 Western Blot Analysis family member, is also cleaved by caspase-8 and then acti- Western blot analysis was done as described previously vates the mitochondrial apoptotic signaling pathway (23). (27). Rabbit polyclonal anti-DR5 antibody (Cayman Chem- Interestingly, some studies have reported that DR5 is ical), mouse monoclonal anti-Bid, anti-caspase-8, anti- expressed more abundantly in cancer cells than in normal caspase-9, and anti-caspase-10 antibodies (MBL, Nagoya, cells (24, 25). Thus, the expression of DR5 may partially Japan), and rabbit monoclonal anti-caspase-3 antibody contribute to the tumor-selective induction of apoptosis (Cell Signaling, Beverly, MA) were used as the primary mediated by TRAIL. Recently, a study based on receptor- antibodies. For subcellular fractionation, we used the selective TRAIL variants revealed that DR5 might contrib- Subcellular Proteome Extraction kit (Calbiochem, San ute more than DR4 to TRAIL-induced apoptosis in cancer Diego, CA) according to the manufacturer’s instructions. cells that express both DRs (26). Accordingly, DR5 is Quantification of PPAR; Activity considered to be a major DR on most tumor cells and an Nuclear extracts from Jurkat cells that were treated with attractive target for cancer therapeutics. pioglitazone and/or GW9662 were prepared by using In this study, we show that 15d-PGJ2 up-regulates DR5 Nuclear Extract kit (Active Motif Japan, Tokyo, Japan) expression at both mRNA and protein levels in human according to the manufacturer’s instructions. We measured malignant tumor cells. This up-regulation is not mediated PPARg DNA-binding activity using Trans AM kit (Active through PPARg activation by 15d-PGJ2 but through Motif Japan) according to the manufacturer’s instructions. increasing DR5 mRNA stability. DR5 up-regulation results Briefly, nuclear extracts (10 Ag) were applied to a multiwell in the synergistic sensitization of soluble recombinant plate coated with a oligonucleotide of the consensus- human TRAIL-induced apoptosis. binding sequence for PPARg, peroxisome proliferator response element. DNA-bound PPARg was detected by a Materials and Methods colorimetric assay using an anti-PPARg antibody and a Reagents secondary antibody conjugated to horseradish peroxidase.

15d-PGJ2 (Cayman Chemical, Ann Arbor, MI) and Transfection and Luciferase Assay trichostatin A (Wako, Osaka, Japan) were dissolved in As described previously (28), a digested SacI-NcoI ethanol. GW9662 (2-chloro-5-nitrobenzanilide), rosiglita- fragment from the DR5 promoter region of genomic DNA zone (Alexis Biochemicals, San Diego, CA), pioglitazone was subcloned into the SacI-NcoIsite of the pGVB2 (kindly provided by Takeda Chemical Industries, Osaka, luciferase assay vector (Toyo Ink, Tokyo, Japan) to produce Japan), and actinomycin D (Sigma, St. Louis, MO) were pDR5PF (pDR5/SacI). pDR5PF and vacant vector plasmids dissolved in DMSO. Soluble recombinant human Apo2L/ (1.0 Ag) were transfected into PC3 cells using Lipofect- TRAIL was purchased from PeproTech (London, United AMINE Plus (Invitrogen, Carlsbad, CA). After 24 hours, the Kingdom). Human recombinant DR5 (TRAIL-R2)/Fc chi- cells were treated with or without 15d-PGJ2 or trichostatin mera protein and the caspase inhibitors zVAD-fmk and A for 24 hours and then harvested. Luciferase assays were zIETD-fmk were purchased from R&D Systems (Minneap- then done as described previously (27). Data were analyzed olis, MN). using Student’s t test. Cell Lines and Cell Culture Quantification of Apoptosis Jurkat human T-cell acute lymphoblastic leukemia cells DNA fragmentation was quantified by the percentage of and PC3 human prostate cancer cells were maintained in cells with hypodiploid DNA (sub-G1). In brief, cells were RPMI1640 with 10% fetal bovine serum, 2 mmol/L fixed with 70% ethanol and treated with RNase A (Sigma), glutamine, 100 units/mL penicillin, and 100 Ag/mL and the nuclei were stained with propidium iodide (Sigma). streptomycin and incubated at 37jC in a humidified The DNA content was measured using a FACSCalibur flow atmosphere of 5% CO2. Normal human peripheral blood cytometer and CellQuest software (Becton Dickinson, Moun- mononuclear cells were isolated using a Ficoll-Paque PLUS tain View, CA). For all assays, 10,000 cells were counted. (Amersham Biosciences, Piscataway, NJ) density gradient Small Interfering RNAs according to the manufacturer’s instructions and main- The DR5 and LacZ small interfering RNA (siRNA) tained in X-VIVO 20 (Cambrex, Walkersville, MD) serum- sequences were described previously (synthesized by Pro- free medium with 0.2% bovine serum albumin (Sigma) at ligo, Kyoto, Japan; ref. 29). DR5 or LacZ siRNAs (25 nmol/L) j 37 C in a humidified atmosphere of 5% CO2. were transfected into cells using a modified Oligofectamine Northern Blot Analysis and RNase Protection Assay protocol (Invitrogen) as described previously (30). Twenty- Northern blot analysis was done as described previously four hours after transfection, cells were treated with 15d-PGJ2 (27). For the DR5 mRNA stability assay, the relative band (18 Amol/L) and/or TRAIL (5 ng/mL) for 24 hours and then intensity was assessed by densitometric analysis of harvested. Data were analyzed using Student’s t test. digitalized autographic images using Scion Image software (Scion Corp., Frederick, MD). The RNase protection assay was done using an RNase protection assay III kit (Ambion, Results Austin, TX) and labeled RNA probes generated with 15d-PGJ 2 Increases the Expression of DR5 hAPO3d template sets (BD PharMingen, San Diego, CA) We first investigated the effects of 15d-PGJ2 on the according to the manufacturer’s instructions. expressions of DR-related genes using an RNase protection

Mol Cancer Ther 2006;5(7). July 2006

assay. As shown in Fig. 1A, 15d-PGJ2 significantly up- DR5, a membrane receptor protein, was increased by regulated DR5 mRNA in Jurkat human T-cell acute 15d-PGJ2 treatment, especially in the membrane protein lymphoblastic leukemia cells carrying inactivated p53 fraction using the subcellular fractionation technique with a point mutation. The expressions of DR4, TRAIL, (Fig. 1D). DcR1, and DcR2 did not increase, although the expressions 15d-PGJ 2 Does Not Induce the DR5 Gene through of Fas and receptor-interacting protein slightly decreased. PPAR; Activation We next carried out Northern blot analysis to confirm the We hypothesized that 15d-PGJ2 might induce DR5 up-regulation of DR5 mRNA not only in Jurkat cells but expression through nuclear receptor PPARg activation also in PC3, androgen-independent human prostate cancer because 15d-PGJ2 is a natural endogenous ligand for cells that are p53-null mutant (Fig. 1B). As shown in PPARg. Previous reports have shown that PPARg is Fig. 1C, we showed that 15d-PGJ2 increased DR5 expres- abundantly expressed in both Jurkat and PC3 cells (31, 32). sion at the protein level in both Jurkat and PC3 cells by However, a PPARg agonist, such as pioglitazone nor Western blot analysis. Furthermore, we confirmed that rosiglitazone, failed to up-regulate DR5 expression in both

Figure 1. 15d-PGJ2 increases the expression of DR5. A, RNase protection assay. Lane 1, probes not treated with RNases; lane 2, RNase- protected probes following hybrid- ization with yeast tRNA; lanes 3 to 4, total RNA from Jurkat cells treated with or without 15d-PGJ2 (18 Amol/L) for 24 h. The housekeeping genes glyceraldehyde-3-phosphate dehydrogenase and L32 are controls. B, Northern blot analysis. Jurkat or PC3 cells were treated with 15d-PGJ2 at indicated concentrations for 24 h. 18Sand 28Sare loading controls. C, Western blot analysis. Jurkat or PC3 cells were treated with 15d-PGJ2 at the indicated concentrations for 24 h. West- ern blot analysis was carried out with anti-DR5 antibody. h-Actin is a loading control. D, Western blot analysis with subcellular fractionation. Jurkat cells were treated with 15d-PGJ2 (18 Amol/L) for 24 h. Cytosol and membrane protein fractions were separated using a subcellular proteome extraction kit. Coomassie bril- liant blue staining of the gel (CBB)is shown to ensure equal loading within each fraction. UT, untreated; ET, treatment with solvent ethanol.

Mol Cancer Ther 2006;5(7). July 2006

Jurkat and PC3 cells even at very high concentrations 15d-PGJ2-treated PC3 cells was strikingly prolonged, and (Fig. 2A-B). Furthermore, pretreatment by GW9662, an significant decay of DR5 mRNA could not be observed irreversible potent inhibitor of PPARg,didnotblock within 10 hours. These data suggest that the 15d-PGJ2- DR5 induction by 15d-PGJ2 at all (Fig. 2C). We confirmed mediated up-regulation of DR5 mRNA occurs via post- the potency of these reagents that we used by quantifying transcriptional control involving the increase of DR5 mRNA PPARg DNA-binding activity to peroxisome proliferator stability. response element. Indeed, pioglitazone could increase 15d-PGJ 2 Sensitizes TRAIL-Induced Apoptosis in a PPARg DNA-binding activity, and GW9662 could block Synergistic Fashion

the activation (Fig. 2D). These results strongly suggest that We hypothesized that 15d-PGJ2 might enhance exoge- 15d-PGJ2 does not induce the DR5 gene through PPARg nous TRAIL-induced apoptosis due to its ability to increase activation. DR5 expression. As shown in Fig. 4A–B, we investigated 15d-PGJ 2 Increases DR5 mRNA Stability the induction of apoptosis by concurrent treatment with To investigate the further mechanism underlying DR5 15d-PGJ2 and exogenous recombinant human TRAIL or up-regulation by 15d-PGJ2, we next examined the effect of each alone in both Jurkat and PC3 cells. Statistical analysis 15d-PGJ2 on DR5 promoter activity. We used the DR5 using factorial ANOVA showed the significant interactions promoter-luciferase reporter plasmid described previously between the administrations of 15d-PGJ2 and TRAIL in (28) for the transient luciferase assay. As shown in Fig. both Jurkat and PC3 cells (P < 0.001). The results indicated 3A, 15d-PGJ2 did not increase DR5 promoter activity, that 15d-PGJ2 strongly sensitized exogenous recombinant whereas trichostatin A increased as we previously TRAIL-induced apoptosis in a synergistic fashion. Further- reported (27). This result is also consistent with 15d-PGJ2 more, we examined the effect of the combined treatment not inducing DR5 mRNA through nuclear receptor with 15d-PGJ2 and TRAIL on normal human peripheral PPARg, a transcription factor. We then investigated the blood mononuclear cells (Fig. 4C, left). Whereas f20% of A effect of 15d-PGJ2 on DR5 mRNA stability. The relative apoptosis was observed by treatment with 18 mol/L 15d- decay rates of DR5 mRNA were determined by Northern PGJ2, any enhancement of apoptosis was not observed blot analysis in PC3 cells under basal conditions and after when cotreated with 50 ng/mL TRAIL. On the other hand, they had been exposed to 15d-PGJ2 for 18 hours. As shown in significant induction of DR5 expression was not observed Fig. 3B, in untreated PC3 cells, DR5 mRNA decayed with a in normal peripheral blood mononuclear cells that were f half-life of 5 hours after transcription was blocked with treated with 15d-PGJ2 in the same conditions with Jurkat actinomycin D. In contrast, the half-life of DR5 mRNA in cells (Fig. 4C, right).

Figure 2. PPARg is not involved in DR5 induction by 15d-PGJ2. A, Western blot analysis. Jurkat cells were treated with pioglitazone at the indicated concentrations for 24 and 48 h. B, Western blot analysis. PC3 cells were treated with pioglitazone or rosiglitazone for 24 h. C, Jurkat or PC3 cells were pretreated with GW9662 (20 Amol/L) for 1 h and then treated with 15d-PGJ2 (18 Amol/L) for 24 h. D, Jurkat cells were treated with 20 Amol/L pioglitazone and/or 20 Amol/L GW9662 for 24 h. PPARg DNA-binding activities were measured as described in Materials and Methods. Columns, mean of triplicate experiments; bars, SD. DM, treatment with solvent DMSO.

Mol Cancer Ther 2006;5(7). July 2006

Figure 3. 15d-PGJ2 does not increase DR5 promoter activity but increases DR5 mRNA stability. A, PC3 cells were transiently transfected with the DR5 promoter-luciferase reporter plasmid (pDR5PF) or vacant vector pGVB2. Cells were treated with or without 15d-PGJ2 (18 Amol/L) or trichostatin A (1 Amol/L) for 24 h, harvested, and assayed as described in Materials and Methods. Columns, mean of triplicate experiments; bars, SD. *, P < 0.001. B, DR5 mRNA half-life in untreated or 15d-PGJ2- treated PC3 cells. Untreated or 15d- PGJ2-treated (18 Amol/L 15d-PGJ2 for 18 h) cells were exposed to actinomycin D (4 Ag/mL) for 2.5, 5, 7.5, and 10 h before cell harvest for total RNA extraction. The DR5 mRNA decay rates were then determined by Northern blotting. DR5 mRNA levels were normalized to 28Sand plotted relative to that at time 0, which was taken as 100%. Points, mean of triplicate experiments; bars, SD. CT, treatment with solvent ethanol.

Synergistic Apoptosis Induced by Combined through specific interactions between TRAIL and its Treatment with 15d-PGJ2 and TRAIL Is Mediated by receptors. We also showed that zVAD-fmk pancaspase TRAIL-DR Interaction, Activation of Caspases, and Bid inhibitor and caspase-8 inhibitor efficiently interrupted the Cleavage apoptosis induced by 15d-PGJ2 and TRAIL. These obser- To confirm that the enhancement of apoptosis by 15d- vations were consistent with the characteristic features of PGJ2 and TRAIL is mediated through specific interactions TRAIL-induced apoptosis (33). As shown in Fig. 5 (bottom), between TRAIL and its receptors, we used human we did Western blot analysis of apoptotic signal mediators recombinant DR5/Fc chimera protein. As shown in Fig. 5 in the same conditions with quantitative assays of (top), human recombinant DR5/Fc chimera protein effi- apoptosis (Fig. 5, top). DR5 up-regulation was also ciently blocked apoptosis caused by 15d-PGJ2 and TRAIL. observed when 15d-PGJ2 was given with TRAIL. It was This result indicated that the enhancement of apoptosis clearly shown that combined treatment with 15d-PGJ2 and was mediated not by nonspecific toxicity of TRAIL but TRAIL significantly enhanced the activation of caspase-8,

Mol Cancer Ther 2006;5(7). July 2006

DR5 gene, which induces degradation of DR5 mRNA, could efficiently block DR5 up-regulation through mRNA stabilization by 15d-PGJ2. This reduction of DR5 expression significantly attenuated the apoptotic response for combined treatment with 15d-PGJ2 and TRAIL (Fig. 6B). This result indicates that DR5 up-regulation, at least in part, plays a role in the sensitization of TRAIL-induced apoptosis by 15d-PGJ2.

Discussion TRAIL is an attractive candidate for cancer treatment due to its ability to induce apoptosis selectively in cancer cells. The agonistic anti-DR5 or anti-DR4 monoclonal antibodies are also promising for cancer treatment (24, 34). The initial targets of these strategies are DRs expressed on cancer cells; however, some cancer cell types are resistant to TRAIL- induced apoptosis (35). Previous reports showed that DR5 overexpression using an expression vector in TRAIL- resistant cancer cells restored TRAIL sensitivity (36, 37). DR5 expression was also highly correlated with TRAIL sensitivity in several Jurkat subclones (38). These results raise possibilities of a strategy to induce DR5 expression in cancer cells to enhance the susceptibility toward TRAIL or the anti-DR5 agonistic monoclonal antibody. In this study, we show that 15d-PGJ2, the most potent antineoplastic agent among cyclopentenone prostaglandins derivatives, markedly up-regulates DR5 at both mRNA and protein levels (Fig. 1). This DR5 induction indeed results in the sensitization of TRAIL-induced apoptosis that is mediated by the activation of caspase-8, caspase-10, Figure 4. 15d-PGJ2 sensitizes TRAIL-induced apoptosis in a synergistic caspase-9, caspase-3, and Bid. Moreover, we show that fashion. Sub-G1 populations (apoptotic cells) were detected using flow cytometry analysis as described in Materials and Methods. Columns, siRNA targeting DR5 attenuates the apoptotic response for mean of triplicate experiments; bars, SD. A and B, Jurkat or PC3 cells cotreatment with 15d-PGJ2 and TRAIL (Fig. 6). Our data were treated with 15d-PGJ2 and/or TRAIL at the indicated concentrations show the significance of DR5 up-regulation for the for 24 h. C, left, normal peripheral blood mononuclear cells (PBMC) were enhancement of susceptibility by 15d-PGJ2 for TRAIL- treated with 18 Amol/L 15d-PGJ2 and/or 50 ng/mL TRAIL for 24 h; right, Western blot analysis. Normal peripheral blood mononuclear induced apoptosis in human cancer cells. Interestingly, in cells and Jurkat cells were treated with 15d-PGJ2 at the indicated normal peripheral blood mononuclear cells, DR5 up- concentrations for 24 h. Western blot analysis was carried out with anti- regulation and significant sensitization of TRAIL-induced DR5 antibody. For all of samples, protein (60 Ag) was loaded. h-Actin is a loading control. apoptosis by 15d-PGJ2 were not observed. On the other hand, thiazolidinediones, potent synthetic PPARg stimulants, do not mimic the DR5-inducing effects caspase-10, caspase-9, and caspase-3 and Bid cleavage. of 15d-PGJ2 (Fig. 2A–B). In addition, potent irreversible Moreover, this Western blot analysis also showed that the PPARg inhibitor, GW9662, does not attenuate DR5 induc- activation of these apoptotic signal members was blocked tion by 15d-PGJ2 (Fig. 2C). These observations suggest by human recombinant DR5/Fc chimera protein, zVAD- that DR5 induction by 15d-PGJ2 does not occur through fmk pancaspase inhibitor, and caspase-8 inhibitor. These nuclear receptor PPARg. Recent studies showed that results indicate that the enhanced apoptosis by 15d-PGJ2 troglitazone, a compound of thiazolidinediones, enhanced and TRAIL depends on TRAIL-DR interaction and caspase TRAIL-induced apoptosis by reducing the survivin level activation. via cyclin D3 repression and cell cycle arrest in a PPARg- DR5 Up-Regulation Contributes toward the Enhance- independent manner (39), whereas the precise mechanisms ment of Apoptosis by Combined Treatment with 15d- responsible for the enhancement of apoptosis by piogli- PGJ2 and TRAIL tazone were not elucidated (31, 40). Thus, it is suggested To address whether DR5 up-regulation by 15d-PGJ2 that DR5 up-regulation is a characteristic mechanism contributes toward the sensitization of TRAIL-induced responsible for sensitization of TRAIL-induced apoptosis apoptosis, we used transient transfection of a double- by 15d-PGJ2. stranded siRNA targeting DR5 gene. The Western blot Furthermore, we show that 15d-PGJ2 markedly increases analysis shown in Fig. 6A showed that the siRNA targeting DR5 mRNA stability (Fig. 3B). To our knowledge, no

Mol Cancer Ther 2006;5(7). July 2006

particular gene regulated by 15d-PGJ2 through mRNA stabilization has been reported to date. Therefore, our data provide a novel mechanism by which 15d-PGJ2 regulates gene expression. In addition, few agents capable of inducing DR5 expression through mRNA stabilization have been reported thus far. Only one report has shown that thapsigargin, which causes endoplasmic reticulum stress, induces DR5 expression partially through prolong- ing the DR5 mRNA half-life (41). However, no precise mechanism underlying the stabilization of DR5 mRNA by these agents has been elucidated yet. To date, AU-rich elements in the 3¶-untranslated region of a variety of human short-lived mRNA are important for the regulation of mRNA stability (42, 43). Indeed, the 3¶-untranslated region

Figure 6. DR5 up-regulation contributes toward the enhancement of apoptosis by combined treatment with 15d-PGJ2 and TRAIL. PC3 cells were transiently transfected with double-stranded siRNA targeting DR5 or LacZ or treated with transfection reagent (Oligofectamine) alone (mock). Twenty-four hours after transfection, cells were treated with 15d-PGJ2 (18 Amol/L) for 24 h. A, Western blotting was done as described in Materials and Methods. h-Actin is a loading control. B, sub-G1 populations were detected using flow cytometry analysis as described in Materials and Methods. Columns, mean of triplicate experiments; bars, SD. *, P < 0.001.

of the human DR5 gene contains AU-rich elements, at least two overlapping copies of the UUAUUUAUU nonamer. Therefore, it is possible to postulate that 15d-PGJ2 might effect AU-rich elements in 3¶-untranslated region of the human DR5 gene post-transcriptionally. The DR5 gene was reported as a p53-regulated gene (44–46). Previous reports showed that some conventional chemotherapeutic drugs, including etoposide or doxorubicin, can induce DR5 expression and enhance TRAIL- induced apoptosis in a p53-dependent manner in certain cancer cell types, such as lung cancer and lymphocytic leukemia cells (46, 47). In this study, we show that 15d-PGJ2 can induce DR5 expression in both PC3, p53-null mutant, and Jurkat cells harboring p53 inactivated with a point mutation. Thus, our results suggest that 15d-PGJ2 induces Figure 5. Combined treatment with 15d-PGJ2 and TRAIL induces synergistic apoptosis coupled with activation of caspase-3, caspase-8, DR5 expression not through p53-mediated transactivation caspase-10, and caspase-9 and Bid cleavage that are blocked by DR5/Fc but through DR5 mRNA stabilization. These results chimera and caspase inhibitors. Top, PC3 cells were treated with 15d- implicate possibilities that 15d-PGJ2 can be expected to PGJ2 (18 Amol/L) and/or TRAIL (5 ng/mL) for 24 h with or without various inhibitors. DR5/Fc, treated with DR5/Fc chimera protein (1 Ag/mL); zVAD- induce the DR5 gene in a relatively broader spectrum of fmk, treated with zVAD-fmk pancaspase inhibitor (20 Amol/L); caspase- cancer cell types than conventional chemotherapeutic A 8 inhibitor, treated with zIETD-fmk (20 mol/L). Sub-G1 populations were drugs, such as etoposide or doxorubicin, because p53 is detected using flow cytometry analysis as described in Materials and Methods. Columns, mean of triplicate experiments; bars, SD. Bottom, often inactivated in more than half of malignant cancer PC3 cells were treated with 15d-PGJ2 and/or TRAIL with or without cell types. various inhibitors in the same conditions as in top, and then total cell In summary, we have shown that 15d-PGJ2 is a potent extracts were harvested. Western blot analysis was carried out with anti- p53-independent sensitizer of TRAIL-induced apoptosis DR5, anti-caspase-3, anti-caspase-8, anti-caspase-10, anti-caspase-9, and anti-Bid antibodies. h-Actin is a loading control. Active forms of via DR5 up-regulation caused by the enhancement of caspases are shown. DR5 mRNA stability in a PPARg-independent manner.

Mol Cancer Ther 2006;5(7). July 2006

Our results imply that combined treatment of 15d-PGJ2 18. Walczak H, Degli-Esposti MA, Johnson RS, et al. TRAIL-R2: a novel with TRAIL is a promising strategy to enhance the apoptosis-mediating receptor for TRAIL. EMBO J 1997;16:5386 – 97. 19. Pan G, O’Rourke K, Chinnaiyan AM, et al. The receptor for the susceptibility of cancer cells to TRAIL-mediated cancer cytotoxic ligand TRAIL. Science 1997;276:111 – 3. therapeutics. 20. Sheridan JP, Marsters SA, Pitti RM, et al. Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors. Science 1997; 277:818 – 21. Acknowledgments 21. Kischkel FC, Lawrence DA, Chuntharapai A, Schow P, Kim KJ, We thank Dr. Kyohei Hayashi (Department of Epidemiology for Community Ashkenazi A. Apo2L/TRAIL-dependent recruitment of endogenous Health and Medicine, Kyoto Prefectural University of Medicine, Kyoto, FADD and caspase-8 to death receptors 4 and 5. Immunity 2000; Japan) for very helpful advice for statistical analysis. 12:611 – 20. 22. Kischkel FC, Lawrence DA, Tinel A, et al. Death receptor recruitment of endogenous caspase-10 and apoptosis initiation in the absence of References caspase-8. J Biol Chem 2001;276:46639 – 46. 23. LeBlanc HN, Ashkenazi A. Apo2L/TRAIL and its death and decoy 1. Straus DS, Glass CK. Cyclopentenone prostaglandins: new insights receptors. Cell Death Differ 2003;10:66 – 75. on biological activities and cellular targets. Med Res Rev 2001;21: 185 – 210. 24. Ichikawa K, Liu W, Zhao L, et al. Tumoricidal activity of a novel anti- human DR5 monoclonal antibody without hepatocyte cytotoxicity. Nat 2. Gilroy DW, Colville-Nash PR, Willis D, Chivers J, Paul-Clark MJ, Med 2001;7:954 – 60. Willoughby DA. Inducible cyclooxygenase may have anti-inflammatory properties. Nat Med 1999;5:698 – 701. 25. Koornstra JJ, Kleibeuker JH, van Geelen CM, et al. Expression of TRAIL (TNF-related apoptosis-inducing ligand) and its receptors in 3. Santoro MG. Antiviral activity of cyclopentenone prostanoids. Trends normal colonic mucosa, adenomas, and carcinomas. J Pathol 2003; Microbiol 1997;5:276 – 81. 200:327 – 35. 12 4. Narumiya S, Fukushima M. D -Prostaglandin J2, an ultimate metab- 26. Kelley RF, Totpal K, Lindstrom SH, et al. Receptor-selective mutants olite of prostaglandin D2 exerting cell growth inhibition. Biochem Biophys of apoptosis-inducing ligand 2/tumor necrosis factor-related apoptosis- Res Commun 1985;127:739 – 45. inducing ligand reveal a greater contribution of death receptor (DR) 5 than 5. Fukushima M, Sasaki H, Fukushima S. Prostaglandin J2 and related DR4 to apoptosis signaling. J Biol Chem 2005;280:2205 – 12. compounds. Mode of action in G1 arrest and preclinical results. Ann N Y 27. Nakata S, Yoshida T, Horinaka M, Shiraishi T, Wakada M, Sakai T. Acad Sci 1994;744:161 – 5. Histone deacetylase inhibitors upregulate death receptor 5/TRAIL-R2 and 6. Matsuzaki Y, Koyama M, Hitomi T, Takaoka Y, Kawanaka M, Sakai sensitize apoptosis induced by TRAIL/APO2-L in human malignant tumor 12,14 T. 15-deoxy-D -prostaglandin J2 activates the expression of cells. Oncogene 2004;23:6261 – 71. INK4b p15 gene, a cyclin-dependent kinase inhibitor. Int J Oncol 2005; 28. Yoshida T, Maeda A, Tani N, Sakai T. Promoter structure and 27:497 – 503. transcription initiation sites of the human death receptor 5/TRAIL-R2 gene. 7. Butler R, Mitchell SH, Tindall DJ, Young CY. Nonapoptotic cell death FEBSLett 2001;507:381 – 5. associated with S-phase arrest of prostate cancer cells via the peroxisome 12,14 29. Wang S, El-Deiry WS. Requirement of p53 targets in chemosensitiza- proliferator-activated receptor-g ligand, 15-deoxy-D -prostaglandin J2. tion of colonic carcinoma to death ligand therapy. Proc Natl Acad Sci U S A Cell Growth Differ 2000;11:49 – 61. 2003;100:15095 – 100. 8. Piva R, Gianferretti P, Ciucci A, Taulli R, Belardo G, Santoro MG. 15- 12,14 30. Shiraishi T, Yoshida T, Nakata S, et al. Tunicamycin enhances tumor deoxy-D -prostaglandin J2 induces apoptosis in human malignant B necrosis factor-related apoptosis-inducing ligand-induced apoptosis in n cells: an effect associated with inhibition of NF- B activity and down- human prostate cancer cells. Cancer Res 2005;65:6364 – 70. regulation of antiapoptotic proteins. Blood 2005;105:1750 – 8. 31. Okano H, Shiraki K, Inoue H, et al. Peroxisome proliferator-activated 9. Clay CE, Namen AM, Atsumi G, et al. Influence of J series receptor-g augments tumor necrosis factor family-induced apoptosis in prostaglandins on apoptosis and tumorigenesis of breast cancer cells. hepatocellular carcinoma. Anticancer Drugs 2002;13:59 – 65. Carcinogenesis 1999;20:1905 – 11. 32. Shiau CW, Yang CC, Kulp SK, Chen KF, Chen CS, Huang JW. 10. Shimada T, Kojima K, Yoshiura K, Hiraishi H, Terano A. Character- Thiazolidenediones mediate apoptosis in prostate cancer cells in part istics of the peroxisome proliferator activated receptor-g (PPARg) ligand through inhibition of Bcl-xL/Bcl-2 functions independently of PPARg. induced apoptosis in colon cancer cells. Gut 2002;50:658 – 64. Cancer Res 2005;65:1561 – 9. 11. Liu JD, Lin SY, Ho YS, et al. Involvement of c-jun N-terminal kinase 12,14 33. Ashkenazi A, Dixit VM. Death receptors: signaling and modulation. activation in 15-deoxy-D -prostaglandin J2-and prostaglandin A1- Science 1998;281:1305 – 8. induced apoptosis in AGSgastric epithelial cells. Mol Carcinog 2003;37: 16 – 24. 34. Ohtsuka T, Buchsbaum D, Oliver P, Makhija S, Kimberly R, Zhou T. Synergistic induction of tumor cell apoptosis by death receptor antibody 12. Kim EJ, Park KS, Chung SY, et al. Peroxisome proliferator-activated and chemotherapy agent through JNK/p38 and mitochondrial death receptor-g activator 15-deoxy-D12,14-prostaglandin J inhibits neuroblas- 2 pathway. Oncogene 2003;22:2034 – 44. toma cell growth through induction of apoptosis: association with extracellular signal-regulated kinase signal pathway. J Pharmacol Exp 35. Zhang L, Fang B. Mechanisms of resistance to TRAIL-induced Ther 2003;307:505 – 17. apoptosis in cancer. Cancer Gene Ther 2005;12:228 – 37. 13. Forman BM, Tontonoz P, Chen J, Brun RP, Spiegelman BM, Evans 36. Mitsiades N, Poulaki V, Mitsiades C, Tsokos M. Ewing’s sarcoma 12,14 RM. 15-deoxy-D -prostaglandin J2 is a ligand for the adipocyte family tumors are sensitive to tumor necrosis factor-related apoptosis- determination factor PPARg. Cell 1995;83:803 – 12. inducing ligand and express death receptor 4 and death receptor 5. Cancer 14. Kliewer SA, Lenhard JM, Willson TM, Patel I, Morris DC, Lehmann Res 2001;61:2704 – 12. JM. A prostaglandin J2 metabolite binds peroxisome proliferator- 37. Kuang AA, Diehl GE, Zhang J, Winoto A. FADD is required for activated receptor-g and promotes adipocyte differentiation. Cell 1995; DR4- and DR5-mediated apoptosis: lack of TRAIL-induced apoptosis in 83:813 – 9. FADD-deficient mouse embryonic fibroblasts. J Biol Chem 2000;275: 15. Ashkenazi A, Pai RC, Fong S, et al. Safety and antitumor activity of 25065 – 8. recombinant soluble Apo2 ligand. J Clin Invest 1999;104:155 – 62. 38. Jang YJ, Park KS, Chung HY, Kim HI. Analysis of the phenotypes of 16. Walczak H, Miller RE, Ariail K, et al. Tumoricidal activity of tumor Jurkat clones with different TRAIL-sensitivities. Cancer Lett 2003;194: necrosis factor-related apoptosis-inducing ligand in vivo. Nat Med 1999; 107 – 17. 5:157 – 63. 39. Lu M, Kwan T, Yu C, et al. Peroxisome proliferator-activated receptor- 17. MacFarlane M, Ahmad M, Srinivasula SM, Fernandes-Alnemri T, g agonists promote TRAIL-induced apoptosis by reducing survivin levels Cohen GM, Alnemri ES. Identification and molecular cloning of two via cyclin D3 repression and cell cycle arrest. J Biol Chem 2005;280: novel receptors for the cytotoxic ligand TRAIL. J Biol Chem 1997;272: 6742 – 51. 25417 – 20. 40. Goke R, Goke A, Goke B, El-Deiry WS, Chen Y. Pioglitazone inhibits

Mol Cancer Ther 2006;5(7). July 2006

growth of carcinoid cells and promotes TRAIL-induced apoptosis by damage-inducible p53-regulated death receptor gene. Nat Genet 1997;17: induction of p21waf1/cip1. Digestion 2001;64:75 – 80. 141 – 3. 41. He Q, Lee DI, Rong R, et al. Endoplasmic reticulum calcium pool 45. Wu GS, Burns TF, McDonald ER III, et al. Induction of the TRAIL depletion-induced apoptosis is coupled with activation of the death receptor KILLER/DR5 in p53-dependent apoptosis but not growth arrest. receptor 5 pathway. Oncogene 2002;21:2623 – 33. Oncogene 1999;18:6411 – 8. 42. Shaw G, Kamen R. A conserved AU sequence from the 3¶ 46. Takimoto R, El-Deiry WS. Wild-type p53 transactivates the KILLER/ untranslated region of GM-CSF mRNA mediates selective mRNA degra- DR5 gene through an intronic sequence-specific DNA-binding site. dation. Cell 1986;46:659 – 67. Oncogene 2000;19:1735 – 43. 43. Zhang T, Kruys V, Huez G, Gueydan C. AU-rich element-mediated 47. Johnston JB, Kabore AF, Strutinsky J, et al. Role of the TRAIL/ translational control: complexity and multiple activities of trans-activating APO2-L death receptors in chlorambucil- and fludarabine-induced factors. Biochem Soc Trans 2002;30:952 – 8. apoptosis in chronic lymphocytic leukemia. Oncogene 2003;22: 44. Wu GS, Burns TF, McDonald ER III, et al. KILLER/DR5 is a DNA 8356 – 69.

Mol Cancer Ther 2006;5(7). July 2006

Downloaded from mct.aacrjournals.org on September 24, 2021. © 2006 American Association for Cancer Research. 15-Deoxy-∆12,14-prostaglandin J2 induces death receptor 5 expression through mRNA stabilization independently of PPARγ and potentiates TRAIL-induced apoptosis

Susumu Nakata, Tatsushi Yoshida, Takumi Shiraishi, et al.

Mol Cancer Ther 2006;5:1827-1835.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/5/7/1827

Cited articles This article cites 47 articles, 17 of which you can access for free at: http://mct.aacrjournals.org/content/5/7/1827.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/5/7/1827.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/5/7/1827. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.