The Involvement of Endoplasmic Reticulum Stress in the Suppression of Colorectal Tumorigenesis by Tolfenamic Acid

Published OnlineFirst October 8, 2013; DOI: 10.1158/1940-6207.CAPR-13-0220

Cancer Prevention Research Article Research

Xiaobo Zhang1,2, Seong-Ho Lee3, Kyung-Won Min2, Michael F. McEntee2, Jin Boo Jeong3, Qingwang Li1, and Seung Joon Baek2

Abstract The nonsteroidal anti-inﬂammatory drug tolfenamic acid has been shown to suppress cancer cell growth and tumorigenesis in different cancer models. However, the underlying mechanism by which tolfenamic acid exerts its antitumorigenic effect remains unclear. Previous data from our group and others indicate that tolfenamic acid alters expression of apoptosis- and cell-cycle arrest–related genes in colorectal cancer cells. þ Here, we show that tolfenamic acid markedly reduced the number of polyps and tumor load in APCmin/ mice, accompanied with cyclin D1 downregulation in vitro and in vivo. Mechanistically, tolfenamic acid promotes endoplasmic reticulum (ER) stress, resulting in activation of the unfolded protein response (UPR) signaling pathway, of which PERK-mediated phosphorylation of eukaryotic translation initiation factor 2a (eIF2a) induces the repression of cyclin D1 translation. Moreover, the PERK-eIF2a-ATF4 branch of the UPR pathway plays a role in tolfenamic acid-induced apoptosis in colorectal cancer cells, as silencing ATF4 attenuates tolfenamic acid-induced apoptosis. Taken together, these results suggest ER stress is involved in tolfenamic acid-induced inhibition of colorectal cancer cell growth, which could contribute to antitumorigenesis in a mouse model. Cancer Prev Res; 6(12); 1–11. 2013 AACR.

Introduction and has received much attention since the inhibition of Colorectal cancer is the third-leading cause of cancer- COX-2 activity produces adverse effects (5). related death in the United States (1). Taking nonsteroidal Tolfenamic acid is a conventional NSAID that has been anti-inflammatory drugs (NSAID) has been associated with long used for treatment of migraines. Compared with the a reduced risk of colorectal tumorigenesis in epidemiologic other NSAIDs, tolfenamic acid exhibits fewer upper gastro- studies (2), and a large amount of evidence from cell culture intestinal side effects (6). Increasing evidence has shown and animal studies has also shown that NSAID treatments that tolfenamic acid also suppresses tumorigenesis in sev- can inhibit growth of cancer cells and tumors (3). Many eral cancer models (7–12), and that this suppression seems mechanisms have been proposed to elucidate the effect of to be independent of COX inhibition. For example, sup- NSAIDs in antitumorigenesis (4). One such proposed pression of specificity proteins Sp1, Sp3, and Sp4, and their mechanism is through the inhibition of COX-2, which is target genes, has been considered to contribute to tolfe- overexpressed in human tumors and plays a role in carci- namic acid-induced anticancer activity (7). Previous results nogenesis. However, a COX-independent function of from our lab also indicate that tolfenamic acid inhibits cell NSAIDs also plays an important role in antitumorigenesis proliferation and induces apoptosis in human colorectal cancer cells, which is in part mediated by the induction of the tumor suppressor proteins NAG-1 and ATF3 (13, 14). These data suggest that tolfenamic acid could exert antican- Authors' Affiliations: 1College of Animal Science and Technology, Northwest A&F University, Yangling, Shaanxi, China; 2Department of Bio- cer activity through various molecular mechanisms, and medical and Diagnostic Sciences, College of Veterinary Medicine, Univer- prompted us to further investigate the additional molecular 3 sity of Tennessee, Knoxville, Tennessee; and Department of Nutrition and mechanisms by which tolfenamic acid induces anticancer Food Science, University of Maryland, College Park, Maryland activity. Note: Supplementary data for this article are available at Cancer Prevention Cyclin D1, an oncogenic protein, is often overexpressed Research Online (http://cancerprevres.aacrjournals.org/). in various cancer cells and tumor tissues. The activated Corresponding Authors: Seung Joon Baek, Department of Biomedical and Diagnostic Sciences, College of Veterinary Medicine, University of cyclin D1-CDK4/6 complex phosphorylates Rb protein and Tennessee, 2407 River Drive, Knoxville, TN 37996-4542. Phone: 865-974- subsequently induces the expression of E2F-target genes 8216; Fax: 865-974-5616; E-mail: [email protected]; Qingwang Li, College that are necessary for DNA synthesis (15). In a recent study, of Animal Science and Technology, Northwest A&F University, 22 Xinong Road, Yangling, Shaanxi 712100, People's Republic of China. Phone: 86- cyclin D1 was also shown to play a role in DNA repair 137-0912-9117; Fax: 86-029-8709-2164; E-mail: through binding to DNA repair proteins, such as BRCA2 [email protected] and RAD51 (16), indicating an extra role of cyclin D1. A doi: 10.1158/1940-6207.CAPR-13-0220 variety of compounds, including NSAIDs, have been docu- 2013 American Association for Cancer Research. mented to mediate cyclin D1 degradation and cell growth

www.aacrjournals.org OF1

Downloaded from cancerpreventionresearch.aacrjournals.org on September 23, 2021. © 2013 American Association for Cancer Research. Published OnlineFirst October 8, 2013; DOI: 10.1158/1940-6207.CAPR-13-0220

Zhang et al.

inhibition (17). It has been reported that tolfenamic acid PERK, and PARP were from Cell Signaling Technology; and also downregulated cyclin D1 expression in esophageal and antibody for HA tag was from GenScript. Control siRNA breast cancer cells (8, 9); however, the detailed molecular (#6201) and siRNA for PERK (#9024) were obtained from mechanism(s) is not clear. Elucidation of the molecular Cell Signaling Technology. siATF4 (sc-35112) was purchased mechanism(s) underlying cyclin D1 downregulation by from Santa Cruz Biotechnology. tolfenamic acid would be beneficial to further understand the antitumorigenic activity of tolfenamic acid, and to Cell culture develop novel derivatives, which could exert more pro- Human cancer cell lines (HCT-116, HT-29, SW480, nounced anticancer effects. LoVo, Caco-2, A549, PC-3, and AsPC-1) were purchased Endoplasmic reticulum (ER) stress inhibits cyclin D1 from American Type Culture Collection (ATCC). ATCC translation and cell-cycle progression (18). Moreover, ER tests the authenticity of these cell lines using short tandem stress initiates apoptosis by activating transcription factor 4 repeat analyses. HCT-116 and HT-29 cells were maintained (ATF4)-dependent C/EBP homologous transcription factor in McCoy’s 5A medium (Bio Whittaker). SW480, A549, (CHOP), apoptosis signal-regulating kinase 1 (ASK1), and PC-3, and AsPC-1 cells were maintained in RPMI-1640 caspase-12 (19). Interestingly, two ER stress-inducible genes medium (Mediatech). Caco-2 and LoVo cells were kept in (EGR1 and ATF3) have been previously identified as the Eagle’s minimum essential medium and Ham’s F12 medi- anticancer targets of tolfenamic acid in colon cancer (13, um (HyClone), respectively. All cells were cultured in 14). Thus, ER stress could be involved in tolfenamic acid- media supplemented with 10% FBS, 100 U/mL penicillin, mediated cancer cell growth suppression. Indeed, some and 100 mg/mL streptomycin in a 5% CO2 atmosphere at NSAIDs like indomethacin and celecoxib have been shown 37C. to trigger ER stress response (20, 21). In contrast, diclofenac blocked ER stress-induced apoptosis in SH-SY5Y cells (22), Plasmid, mutagenesis, and transient transfection and pranoprofen also suppressed ER stress-induced glu- The pRcCMV-cyclin D1-HA plasmid was generously cose-regulated protein 78 (GRP78) and CHOP expression provided by Dr. E. Dmitrovsky (Department of Pharma- (23). Therefore, how ER stress is involved in tolfenamic cology and Toxicology, Dartmouth Medical School, Han- acid’s effect on cell growth remains unclear and needs to be over, NH; ref. 25). The pCMV-cyclin D1 and ubiquitin further elucidated. (Ub)-HA were gifts from Dr. Richard G. Pestell (Kimmel In the current studies, we first evaluated the chemo- Cancer Center, Thomas Jefferson University, Philadel- preventive effect of tolfenamic acid in a mouse model of phia, PA). Cyclin D1 transversion from threonine to colorectal cancer. As a result, we found that tolfenamic acid alanine at the 286 amino acid position was generated þ inhibited polyp formation in APCmin/ mice, along with a using the QuickChange site-directed mutagenesis kit dramatic decrease in cyclin D1 expression in tumors. Con- (Strategene) according to the manufacturer’s protocol. sistently, we have also shown that tolfenamic acid down- ATF4 expression construct was purchased from OriGene, regulates cyclin D1 expression in different cancer cells in and p5xATF6-GL3 luciferase reporter construct (plasmid vitro, which is associated with the increase in Rb activity. We #11976) was obtained from Addgene (26). Transient further investigated the underlying mechanism, showing transfection was carried out using TransIT-2020 transfec- that tolfenamic acid repressed cyclin D1 translation through tion reagent (Mirus Bio LLC) according to the manufac- activation of the ER stress-mediated unfolded protein turer’s protocol. For luciferase assay, cells were seeded in responses (UPR) pathway. In addition, ER stress played a 12-well plates at a density of 1.0 105 cells per well. After role in tolfenamic acid-induced apoptosis in colorectal transfection with 1 mg p5xATF6-GL3 and pRL-null, cells cancer cells. Therefore, our data strongly suggest that ER were treated with dimethyl sulfoxide (DMSO) and tolfe- stress response could contribute to the antitumorigenic namic acid for 24 hours. Cell lysates were harvested using activity of tolfenamic acid. 1XPassive lysis buffer (Promega), and then were subjected to luciferase activity analysis using DualGlo Luciferase Assay Kit (Promega). Materials and Methods Reagents and antibodies Caspase-3/7 activity assay Tolfenamic acid, SC-560, and lactacystin were purchased Cells were seeded in a 96-well plate at a density of 1.0 from Cayman Chemical, and 5,5-dimethyl-3-(3-fluorophe- 104 cells per well. Empty vector and ATF4 expression vector nyl)-4-(4-methylsulphonyl)phenyl-2(5H)-furanone (DFU) were transiently transfected using TransIT-2020 transfec- was described previously (24). All other NSAIDs, cyclohex- tion reagent (Mirus Bio LLC) for 24 hours. Caspase-3/7 imide, and propidium iodide (PI) were purchased from activity was determined by caspase-Glo 3/7 reagent (Pro- Sigma-Aldrich. Epoxomicin was obtained from Calbiochem. mega). Briefly, 100 mL reagent was added to each well. After All other chemicals were purchased from Fisher Scientific. incubating the plate at room temperature for 1 hour in the RNase A was purchased from 5PRIME. Antibodies for dark, luminescence was measured using a FLX-800 micro- cyclin D1, ATF4, Bcl-2, and Actin were from Santa plate reader (Bio-Tek). Fold change compared with empty Cruz Biotechnology; antibodies for pRb (Ser780), total Rb, vector transfection is represented as the mean SD of three p-Cyclin D1 (Thr286), BiP, CHOP, p-eIF2a,totaleIF2a, wells.

OF2 Cancer Prev Res; 6(12) December 2013 Cancer Prevention Research

Tolfenamic Acid and ER Stress

Determination of sub-G1 cells Tris–HCl pH 7.4, 150 mmol/L NaCl, 1 mmol/L EDTA, 1% Cells were seeded in 6-well plates at a density of 3.0 105 Triton X-100, 1% sodium deoxycholate, 0.1% SDS) sup- cells per well in three replicates, and cultured to 60% to 80% plemented with 1 protease inhibitor cocktail solution confluence. Then, cells were treated with vehicle and tolfe- (Calbiochem) and phosphatase inhibitor (1 mmol/L namic acid (50 mmol/L) for 24 hours, then harvested and Na3VO4, 1 mmol/L NaF), and centrifuged at 13,000 g fixed in 70% ethanol. After being stored at 20C overnight, for 10 minutes at 4C. Protein concentration was deter- the fixed cells were washed with PBS and stained with mined by the BCA protein assay (Pierce) using Bovine propidium iodide (PI, 70 mmol/L) solution containing Serum Albumin as the standard. Protein (30 mg) was mixed RNase A (1 mg/mL) for 15 minutes at room temperature. with an equal amount of 2 SDS-PAGE sample loading The sub-G1 phase was determined by a Beckman Coulter buffer and boiled for 5 minutes. After separation by SDS- Epixs XL flow cytometer equipped with ModFit LT software. PAGE, the proteins were transferred to nitrocellulose membranes (Osmonics). The membranes were incubated with a RNA interference specific primary antibody in TBS containing 0.05% Tween HCT-116 cells were seeded on 6-well plates at a density 20 (TSB-T) and 5% nonfat dry milk at 4C overnight. After of 3.0 105 cells per well overnight. Control siRNA, three washes with TBS-T, the blots were incubated with siATF4, or siPERK was transfected at a final concentration peroxidase-conjugated IgG for 1 hour at room temperature, of 100 nmol/L using PepMute siRNA & DNA Transfection visualized using ECL (Amersham Biosciences), and quan- Reagent (SignaGen) according to the manufacturer’s tified by Scion Image Software (Scion Corp.). instruction. After 24- (siATF4) or 48-hour (siPERK) transfection, cells were treated with vehicle and tolfenamic Immunoprecipitation acid (50 mmol/L). The cells were harvested using lysis buffer (0.025 mol/L Tris, 0.15 mol/L NaCl, 0.001 mol/L EDTA, 1% NP40, 5% RNA isolation, semi-quantitative reverse transcription Glycerol, pH7.4) containing with 1 protease inhibitor PCR, and real-time PCR cocktail solution (Calbiochem) and phosphatase inhibitor Total RNA of HCT-116 and SW480 cells treated by (1 mmol/L Na3VO4, 1 mmol/L NaF), and then kept on ice DMSO and tolfenamic acid were isolated by an E.Z.N.A for 30 minutes. After being spun down for 10 minutes, the Total RNA Kit (Omega Bio-Tek) according to the manu- suspension was precleared using protein A/G PLUS-agarose facturer’s protocol. Then, RNA (1 mg) was reverse tran- (Santa Cruz Biotechnology) for 30 minutes at 4C. Protein scribed using a Verso cDNA synthesis Kit (ThermoScienti- concentration was determined as described above. Protein fic). PCR was performed using GoTaq Green Master Mix lysates (1,000 mg) were incubated with 5 mg primary anti- PCR Reaction Mixture (Promega) with primers for human body and IgG control for 1 hour at 4C, followed by adding cyclin D1, EGR-1, ATF3, XBP1, and GAPDH as follows: 50 mL resuspended protein A/G PLUS-agarose overnight. cyclin D1, forward 50-ATGGAACACCAGCTCCTGTGCTGC- Immunoprecipitates were collected by centrifuging at 1000 30 and reverse 50-TCAGATGTCCACGTCCCGCACGT-30; g for 5 minutes at 4C. After washing five times with lysis EGR-1 forward 50-CTGCGACATCTGTGGAAGAA-30 and buffer, the pellets were resuspended with 50 mL2 SDS- reverse 50-TGTCCTGGGAGAAAAGGTTG-30; ATF3: forward PAGE sample loading buffer. The samples were boiled for 5 50-GTTTGAGGATTTTGCTAACCTGAC-30, and reverse 50- minutes, and 20 mL of samples were subjected to Western AGCTGCAATCTTATTTCTTTCTCGT-30; XBP1: forward 50- blot analysis. CCTTGTAGTTGAGAACCAGG-30, and reverse 50-GGGGC- TTGGTATATATGTGG-30; GAPDH, forward 50-GGGCT- Animal study þ GCTT TTAACTCTGGT-30 and reverse 50-TGGCAGGTTTTT- The APCMin/ mice (C57BL/6background) were pur- CTAGACGG-30. Real-Time PCR was performed using iTaq chased from Jackson Laboratory. To investigate the long- Univesal SYBR Green Supermix (Bio-Rad) with different term effect of tolfenamic acid on tumor formation, mice primer sets for human cyclin D1, EGR-1, or GAPDH as aged 6 to 8 weeks were randomly assigned to three groups, follows: cyclin D1, forward 50-GGCGGAGGAGAACAAA- and 0, 25, or 50 mg/kg of tolfenamic acid was given by oral CAGA-30 and reverse 50- TGTGAGGCGGTAGTAGGACA- gavage with 0.5% methylcellulose (vehicle) every 2 days for 30; EGR-1 forward 50-CACCTGACCGCAGAGTCTTT-30 and 4 weeks. To investigate the short-term effect of tolfenamic reverse 50-CTGACCAAGCTGAAGAGGGG-30; GAPDH, for- acid on tumor formation and protein expression, mice aged ward 50-GGGAGCCAAAAGGGTCATCA-30 and reverse 50- 16 to 18 weeks were randomly divided into two groups, and TGATGGCATGGACTGTGGTC-30. Gene expression levels 0 or 50 mg/kg body weight of tolfenamic acid was given by were calculated and GAPDH was used as a house-keeping oral gavage with 0.5% methylcellulose once per day for 3 gene, using MyiQ thermal cycler (Bio-Rad). The fold days. At the end of the experiment, the mice were eutha- changes in mRNA levels were calculated using the DDCt nized by overdose of CO2, and the entire intestinal tract was method. removed, flushed with cold saline, and opened longitudi- nally. The number and size (diameter) of polyps and tumors Western blot analysis were scored blindly under a dissecting microscope as Cells were washed with PBS and cell lysates harvested described previously (27, 28). Tumor load was calculated using radioimmunoprecipitation assay buffer (50 mmol/L as number of tumors average diameter. Both normal and

www.aacrjournals.org Cancer Prev Res; 6(12) December 2013 OF3

Zhang et al.

A B T TA A

4 6–8 10–12 (wks) 4 16–18 (wks) Wean Euthanasia Wean Euthanasia Group 1: 0.5% methylcellulose (n = 6) Group 1: 0.5% methylcellulose (n = 8) Group 2: 25 mg/kg BW of TA (n = 7) Group 2: 50 mg/kg BW of TA (n = 7) Group 3: 50 mg/kg BW of TA (n = 6) 80 140 40 40 70 120 35 35 60 100 30 30 50 25 25 80 40 ** 20 20 60 30

15 15 load Tumor ** Tumor load Tumor 20 40 10 10 *** of polyp/mouse No. 20 No. of polyp/mouse No. 5 5 *** 10 *** *** 0 0 0 0 02550 0 25 50 0 50 0 50 TA (mg/kg) TA (mg/kg) TA (mg/kg) TA (mg/kg)

C Normal Tumor

Veh TA Veh TA

123456 123456 Cyclin D1

COX-2

Actin

Figure 1. Tolfenamic acid suppresses colorectal tumorigenesis in a mouse model. A, mice aged 6–8 weeks were randomly divided into three groups, and respectively administered 0.5% methylcellulose, 25 mg/kg body weight, or 50 mg/kg body weight of tolfenamic acid (TA) every other day for 4 weeks (total 14 times). The number of polyps and tumor load were calculated as described in Materials and Methods. Data indicate mean SD of all experiments. Results were considered statistically signiﬁcant at , P < 0.001, compared to group 1. B, mice aged 16–18 weeks were randomly divided into two groups, and respectively treated with 0.5% methylcellulose or 50 mg/kg body weight tolfenamic acid once per day for 3 days (total three times). The number of polyps and tumor load were calculated, and the data are expressed as mean SD of all experiments. Results were considered statistically signiﬁcant at , P < 0.01, compared to group 1. C, proteins prepared from tissues of mice at age 16–18 weeks fed vehicle or 50 mg/kg body weight tolfenamic acid were subjected to Western blot analysis for COX-2, cyclin D1, and actin expression.

þ neoplastic tumors were kept under RNAlater solution for polyps and tumor load of APCmin/ mice treated with molecular analysis. All animal research procedures were vehicle and tolfenamic acid. In the first experiment, we approved by the University of Tennessee Animal Care treated mice aged 6 to 8 weeks with tolfenamic acid for 4 and Use Committee and were in accordance with NIH weeks. As shown in Fig. 1A, tolfenamic acid caused a guidelines. striking reduction in the total number of polyps and tumor load in a dose-dependent manner. Next, we treated Statistical analysis mice at 16 to 18 weeks of age with tolfenamic acid for a Statistical analysis was performed using the Student short period (3 days) to get a tumor sample and observe unpaired t test with statistical significance set at , P < the short-term effect of tolfenamic acid treatment on 0.5; , P < 0.01; and , P < 0.001. tumorigenicity. As shown in Fig. 1B, tolfenamic acid still significantly reduced the number of polyps and tumor load, compared with the control group. Many genes, Results including b-catenin and Smad, had altered expression in Tolfenamic acid suppressed polyp formation in an tissue samples (data not shown). However, the most þ APCmin/ mouse model interesting and consistent gene alteration was cyclin D1 To evaluate the chemopreventive activity of tolfenamic (Fig. 1C). Both cyclin D1 and COX-2 were overexpressed acid in a mouse model, we examined the number of in tumor tissues, which is in agreement with a previous

OF4 Cancer Prev Res; 6(12) December 2013 Cancer Prevention Research

Tolfenamic Acid and ER Stress

A B Conventional selective Cox-2 selective Cox-1 HCT-116 SW480 LoVo Caco-2 HT-29

Cyclin D1

Actin DMSO Diclofenac Ibuprofen Aspirin Naproxen Piroxicam acid Tolfenamic SC-236 DFU SC-560

Cyclin D1

Actin

C TA (μmol/L) TA (50 μmol/L)

0 5102550 0 1 3 6 12 24 (h)

Cyclin D1 Cyclin D1 Actin Actin

D HCT-116 Caco-2 PC-3 AsPC-1 A549

TA (μmol/L) 0 5 25 50 0 5 25 50 0 5 25 50 0 5 25 50 0 5 25 50

Cyclin D1 Actin

Figure 2. Tolfenamic acid downregulates cyclin D1 in cancer cells. A, endogenous cyclin D1 expression in different colorectal cancer cells. Western blot analysis was carried out for analyzing expression of cyclin D1 and actin in the indicated cell lines. B, SW480 cells were treated with various NSAIDs at 50 mmol/L for 24 hours. The cell lysates were harvested and subjected to Western blot analysis for cyclin D1 and actin. C, SW480 cells were treated with 0, 5, 10, 25, or 50 mmol/L tolfenamic acid for 24 hours (left) and 50 mmol/L tolfenamic acid for 0, 1, 3, 6, 12, or 24 hours (right). Cell lysates were analyzed by Western blot using cyclin D1 and actin antibodies. D, colorectal (HCT-116, Caco-2), prostate (PC-3), pancreatic (AsPC-1), and lung (A549) cancer cells were treated with 0, 5, 25, or 50 mmol/L of tolfenamic acid for 24 hours. Western blot analysis was carried out for analyzing cyclin D1 and actin. report (29); however, only cyclin D1 was dramatically Tolfenamic acid caused rapid cyclin D1 downregulation in suppressed in tolfenamic acid-treated tumor samples. a dose- and time-dependent manner, and the decrease occurred within 1 hour (Fig. 2C). To further evaluate the Tolfenamic acid downregulated cyclin D1 expression in effect of tolfenamic acid on other cancer cells, we treated cancer cells colorectal cancer cells (HCT-116 and Caco-2), prostate As cyclin D1 acts as a pro-oncogenic factor, it is not cancer cells (PC-3), pancreatic cancer cells (AsPC-1), and surprising that colorectal cancer cells harbor overexpressed lung cancer cells (A549) with tolfenamic acid. As shown cyclin D1 (Fig. 2A), and its downregulation may contribute in Fig. 2D, tolfenamic acid dose-dependently downregu- to the antiproliferative effect of NSAIDs. As it has been lated cyclin D1 expression in different cancer cells. documented that NSAIDs differ in their ability to suppress cyclin D1 expression or cell proliferation (30), we treated Tolfenamic acid affected phosphorylation of Rb SW480 cells for 24 hours with the same dose (50 mmol/L) protein of different NSAIDs: conventional (diclofenac, ibuprofen, The cyclin D1-CDK4/6 complex phosphorylates and aspirin, naproxen, piroxicam, tolfenamic acid), COX-2 inactivates Rb, facilitating the G1–S phase transition. Spe- selective (SC-236, DFU), or COX-1 selective (SC-560). ciﬁcally, cyclin D1 is required for phosphorylating the Tolfenamic acid and SC-560 completely decreased cyclin Ser780 amino acid position of Rb (31). We observed that D1 expression, suggesting both of them were the most tolfenamic acid decreased phosphorylated Rb at Ser780 in potent agents of those we tested with respect to cyclin D1 both SW480 and HCT-116 cells (Fig. 3A). To further inves- downregulation (Fig. 2B). As we have shown that tolfena- tigate whether cyclin D1 plays a role in the loss of Rb mic acid can suppress cyclin D1 in an in vivo model (Fig. 1), phosphorylation by tolfenamic acid, we transiently trans- we selected this compound for subsequent studies. fected cyclin D1 expression vector containing an alanine for

www.aacrjournals.org Cancer Prev Res; 6(12) December 2013 OF5

Zhang et al.

A SW480 HCT-116 TA (50 μmol/L) 0 1 3 6 12 24 0 1 3 6 12 24 (h) Figure 3. Tolfenamic acid (TA) activates Rb via p-Rb (Ser780) dephosphorylation. A, HCT-116 and SW480 cells were treated with Total Rb 50 mmol/L tolfenamic acid for 0, 1, 3, 6, 12, or 24 hours. Cell lysates p-RB/Total Rb: 1.0 0.5 0.2 0.2 0.2 0.2 1.0 0.5 0.4 0.2 0.1 0.0 were harvested and analyzed by Western blot for phospho-Rb B Mock Cyclin D1-HA (T286A) (Ser780) and total Rb. B, HCT-116 cells were transiently transfected TA (50 μmol/L) 0 6 12 24 0 6 12 24 (h) with Mock and cyclin D1-HA (T286A) for 24 hours, followed by p-Rb (Ser780) treatment with 50 mmol/L Total Rb tolfenamic acid for 0, 6, 12, or 24 hours. Western blot analysis was HA performed for phosphor-Rb (Ser780), total Rb, HA, and actin. Actin

p-RB/Total Rb: 1.00.7 0.3 1.00.1 0.9 0.8 0.3

threonine substitution at codon 286 (cyclin D1-HA pathway, which is mainly responsible for protein transla- [T286A]), which is more stable than the wild-type form tion. However, we did not observe that tolfenamic acid (32). Overexpression of the stable form of cyclin D1 suppressed this pathway (data not shown). Another impor- restored phosphorylated Rb at Ser780, suggesting that tant protein translation control is mediated by eIF2a, cyclin D1 downregulation by tolfenamic acid, at least in phosphorylation of which can lead to nearly global trans- part, facilitates dephosphorylation of Rb at Ser780 (Fig. 3B). lation inhibition. As ER stress usually induces phosphorylated eIF2a, we therefore asked whether tolfenamic acid Tolfenamic acid inhibited cyclin D1 at the translational promoted ER stress in colorectal cancer cells. The ER stress- regulation in colorectal cancer cells activated UPR pathway consists of a large variety of down- To elucidate the molecular mechanism by which tolfe- stream responses, including ATF6-mediated transcription- namic acid downregulated cyclin D1 expression in colorec- al activation, splicing of x-box-binding protein-1(XBP-1) tal cancer cells, we treated HCT-116 and SW480 cells with mRNA, and induction of target gene expression, such as tolfenamic acid for 1 hour and examined the expression of BiP, ATF4, and CHOP (33). As expected, treatment of HCT- cyclin D1 transcripts. As shown in Fig. 4A, cyclin D1 116 and SW480 cells with tolfenamic acid resulted in the transcript levels were not altered by tolfenamic acid within induction of ATF6 transcriptional activity, XBP-1 mRNA 1 hour, although there was a dramatic increase in EGR-1 and splicing, and expression of speciﬁc genes (BiP, ATF4, and ATF3 mRNA expression as a control (13, 14). Given cyclin CHOP) in a time-dependent manner (Fig. 5A–C), indicat- D1 is a short-lived protein, we examined whether tolfe- ing tolfenamic acid could be a potential activator of ER namic acid triggered the cyclin D1 degradation. Results stress. in Fig. 4B and C indicate that neither endogenous nor exogenous cyclin D1 stability was affected by tolfenamic Cyclin D1 suppression and ATF4 induction by acid. Furthermore, treatment of inhibitors responsible for tolfenamic acid are mediated by PERK-eIF2a pathway cyclin D1 degradation did not restore cyclin D1 expression Given that ER stress would activate the kinase PERK that in the presence of tolfenamic acid (Fig. 4D and Supplemen- resides in the ER lumen, resulting in the phosphorylation of tary Fig. S1). In addition, tolfenamic acid did not affect eIF2a and subsequently the repression of cyclin D1 trans- Thr286 phosphorylation nor ubiquitination of cyclin D1, lation (18), we examined the relationship of PERK/eIF2a which has been well documented to be associated with signaling and cyclin D1 expression in the presence of cyclin D1 degradation (Fig. 4E and Supplementary Fig. S2). tolfenamic acid. As shown in Fig. 6A, tolfenamic acid caused Taking these results together, we found that tolfenamic acid rapid phosphorylation of eIF2a coupled with a dramatic did not affect transcription nor protein degradation path- decrease of cyclin D1 expression in a time-dependent man- way of cyclin D1; rather tolfenamic acid may inhibit cyclin ner. In addition, thapsigargin, a well-known activator of ER D1 translation initiation. stress had an effect on cyclin D1 expression similar to that of tolfenamic acid (Fig. S3). Moreover, silence of PERK using Tolfenamic acid promoted ER stress response speciﬁc siRNA slightly restored cyclin D1 expression in the To further understand the cellular signaling pathways presence of tolfenamic acid (Fig. 6B), suggesting the PERK- involved in tolfenamic acid-mediated cyclin D1 translation, eIF2a pathway was involved in tolfenamic acid-mediated we investigated the effect of tolfenamic acid on the mTOR cyclin D1 downregulation. The UPR is a prosurvival

OF6 Cancer Prev Res; 6(12) December 2013 Cancer Prevention Research

Tolfenamic Acid and ER Stress

A HCT-116 SW480 30 40 TA (μmol/L) 0 525500 52550 HCT116 SW480 25 30 Cyclin D1 20 Cyclin D1 Cyclin D1 EGR1 EGR1 EGR-1 15 20 ATF3 ATF3 ATF3 10

Fold induction Fold induction Fold 10 5 GAPDH 0 0 0 5 25 50 (μmol/L) 0 5 25 50 (μmol/L) BCTA (μmol/L) 01025 DMSO TA (50 μmol/L) CHX (10 μg/mL) 0306090 0 30 60 90 0 30 60 90 (min) CHX (10 μg/mL) 0 30 60 90 120 0 30 60 90 120 (min) Cyclin D1 Cyclin D1-HA Actin GFP

DEIP: Input IgG Cyclin D1 TA (50 μmol/L) –+ –+–+ Cl

4 WB with:

DMSO DMSO Lactacystin Epoxomicin NH z-VAD-fmk Calpeptin Leupeptin KU55933 Staurosporine HA TA (50 μmol/L) –++++ +++++ Cyclin D1 Cyclin D1 Actin p-Cyclin D1 (Thr 286)

Figure 4. Tolfenamic acid suppresses cyclin D1 translation. A, HCT-116 and SW480 cells were treated with 0, 5, 25, or 50 mmol/L of tolfenamic acid for 1 hour. Semi-quantitative RT-PCR was carried out for analyzing cyclin D1, EGR-1, ATF3, and GAPDH mRNA expression using specific primers as described in Materials and Methods. Real-time PCR was also performed as described in Materials and Methods (right). B, SW480 cells were treated with 0, 10, or 25 mmol/L of tolfenamic acid for 1 hour, followed by incubation with 10 mg/mL cycloheximide (CHX) for the indicated times. Cell lysates were subjected to Western blot using cyclin D1 and actin antibodies. C, cyclin D1-HA construct was transiently transfected into HCT-116 cells along with GFP expression vector serving as a transfection control for 24 hours. Cells were treated with DMSO and 50 mmol/L tolfenamic acid for 1 hours, followed by incubation with 10 mg/mL cycloheximide (CHX) for the indicated times. Cell lysates were analyzed using HA (cyclin D1) and GFP antibodies. D, SW480 cells were treated with DMSO, lactacystin (10 mmol/L), epoxomicin (1 mmol/L), NH4Cl (20 mmol/L), z-VAD-fmk (50 mmol/L), calpeptin (50 mmol/L), leupeptin (200 mmol/L), KU55933 (10 mmol/L), or staurosporine (2 mmol/L) for 1 hour, followed by incubation with 50 mmol/L of tolfenamic acid for 1 hour, as indicated. Cell lysates were analyzed for cyclin D1 and actin using Western blot analysis. E, the pCMV-cyclin D1 and Ub-HA plasmids were cotransfected into HCT-116 cells for 24 hours. Cell lysates were immunoprecipitated using either cyclin D1 antibody (IP: cyclin D1) or nonspecific IgG antibody (IP: IgG). Cyclin D1 ubiquitination, phosphorylation of cyclin D1, and total amount of cyclin D1 were determined with the indicated antibody. response to allow cells to remove unfolded proteins and re- Bcl-2 downregulation, and attenuated apoptosis as mea- establish homeostasis. However, if the primary stimuli sured by cleaved PARP (Fig. 6E). These data indicated causing unfolded proteins are persistent and the stress tolfenamic acid caused cell apoptosis via, at least in part, cannot be resolved, signaling switches from prosurvival to the PERK-eIF2a-ATF4 pathway in colorectal cancer cells. proapoptosis (34). Of the UPR signaling pathway branches, PERK-eIF2a-ATF4 upregulates CHOP expression, which subsequently induces apoptosis in part through downregu- Discussion lating BCL-2 gene expression (19). As shown in Fig. 6B, We and others have previously reported that tolfenamic tolfenamic acid-induced CHOP expression is mediated by acid exhibits anticancer activity via different molecular the PERK-eIF2a-ATF4 axis. Consistent with CHOP upregu- targets, including Sps (7, 37), ATF2/ATF3 (14), NAG-1 lation, tolfenamic acid also downregulated Bcl-2 expression (13), NF-kB (38), and Smads (39). In the present study, accompanied with the induction of apoptosis as measured we further confirmed that tolfenamic acid is a powerful by cleaved PARP and the percentage of sub-G1 fraction cells chemopreventive or therapeutic agent against colorectal (Fig. 6C). In addition, overexpression of ATF4 increased cancer in vitro and in vivo, and that tolfenamic acid-mediated caspase-3/7 activity, supporting the previously documented growth inhibition might be associated with ER stress proapoptotic role of ATF4 in ER stress (35, 36; Fig. 6D). response, especially the PERK-eIF2a axis, which resulted Finally, knockout of ATF4 using specific siRNA blocked in downregulation of prosurvival proteins like cyclin D1, tolfenamic acid-induced CHOP expression and subsequent and upregulation of proapoptotic proteins like CHOP. In

www.aacrjournals.org Cancer Prev Res; 6(12) December 2013 OF7

Zhang et al.

Figure 5. Tolfenamic acid promotes ER stress response. A, HCT-116 80 10 A HCT-116 *** SW480 and SW480 cells were transiently 8 ** 60 transfected with p5ATF6-GL3 and 6 40 pRL-null for 24 hours, followed by

RLU 4 *** RLU ** treatment with 0, 5, 25, or 50 mmol/L 20 2 * of tolfenamic acid for 24 hours. *** 0 0 Luciferase activity was measured as DMSO 5 μmol/L 25 μmol/L 50 μmol/L DMSO 5 μmol/L 25 μmol/L 50 μmol/L described in Materials and Methods. RLU, relative luciferase unit. Results B HCT-116 SW480 were considered statistically TA (μmol/L) 0 52550 TA (μmol/L) 0 5 25 50 signiﬁcant at , P < 0.5; , P < 0.01; US and , P < 0.001, compared with US XBP1 XBP1 S DMSO. B, HCT-116 and SW480 S cells were treated with 0, 5, 25, or GAPDH GAPDH 50 mmol/L of tolfenamic acid for 24 hours. Reverse transcription- C μ SW480, TA (50 μmol/L) PCR was carried out for analyzing HCT-116, TA (50 mol/L) XBP-1 and GAPDH mRNA ﬁ 01361224(h) 01361224(h) expression using speci cprimersas described in Materials and Methods. BiP BiP C, HCT-116 and SW480 cells were treated with 50 mmol/L of tolfenamic ATF4 ATF4 acid for 0, 1, 3, 6, 12, or 24 hours. CHOP CHOP The cell lysates were subjected to Actin Western blot analysis for analyzing Actin BiP, ATF4, CHOP, and actin expression.

addition, we found that tolfenamic acid markedly down- machinery (42). After exposure of cells to ER stress, PERK is regulated cyclin D1 in tumors and a large variety of cell lines, mainly responsible for phosphorylating eIF2a, which med- and that this downregulation partially contributed to the iates cyclin D1 translation repression (43). Indeed, silenc- loss of phosphorylated Rb. Given the key role of cyclin D1 in ing PERK in HCT-116 cells partially restored cyclin D1 G1–S phase transition, downregulation of cyclin D1 by expression in the presence of tolfenamic acid. Because we tolfenamic acid could induce cell-cycle arrest. Indeed, there could not completely block the expression of PERK (Fig. are several reports showing that treatment with tolfenamic 6B), it is likely that remaining PERK activity contributed to acid results in G0–G1 cell-cycle arrest (11, 40, 41). On the the loss of cyclin D1 in HCT-116 cells treated with tolfe- other hand, we also provided direct evidence that tolfenamic acid. In addition, it is possible that other kinases namic acid induced apoptosis via activation of the PERK- would be able to phosphorylate eIF2a in the absence of eIF2a-ATF4 pathway. PERK, resulting in cyclin D1 repression (43). Therefore, ER It has been suggested that tolfenamic acid suppresses stress-sensing pathways are very complicated, and the detail erbB2 and Sp transcription factors, which could result in pathways need to be elucidated. the suppression of cyclin D1 expression (8, 9). However, Accumulating evidence suggests that the activation of our data clearly indicate that tolfenamic acid-induced cyclin UPR exerts contradictory consequences to tumor devel- D1 downregulation is an erbB2- and Sp-independent path- opment. There is some evidence that UPR signaling con- way; although tolfenamic acid did not affect the cyclin D1 tributes to tumor growth through increasing tumor cell transcription, it significantly decreased protein expression tolerance to hypoxia (44, 45). On the other hand, various (Figs. 3 and 4A), and tolfenamic acid-induced cyclin D1 compounds harboring anticancer activity have been downregulation occurred much earlier than tolfenamic shown to induce ER stress-mediated cell-cycle arrest and acid-mediated Sp downregulation (1 vs. 24 hours) (unpub- apoptosis (46–49). Our data indicate that tolfenamic acid lished data). These results further indicate that tolfenamic exerts anticancer activity, in part dependent on the acti- acid-mediated cyclin D1 downregulation is independent of vation of the UPR signaling pathway. Similar findings transcription factors such as Sp1 and erbB2, and even other were previously reported in cells treated with the COX-2 transcription factors b-catenin or NF-kB. selective inhibitor celecoxib and its analog (21, 50), Interestingly, we found that cyclin D1 degradation was suggesting tolfenamic acid and celecoxib might exhibit not enhanced in the presence of tolfenamic acid (Fig. 4B–E). comparable antineoplastic activity in a COX-independent On the basis of our data, we suggest that tolfenamic acid manner. could be able to suppress cyclin D1 translation, at least in We have identified EGR1 and ATF3 as mediators of colorectal cancer cells. Tolfenamic acid triggered an ER tolfenamic acid-induced apoptosis in colorectal cancer cells stress response, resulting in the phosphorylation of eIF2a, (13, 14). Interestingly, both of them are inducible under ER which leads to the repression of protein translation through stress (51, 52). Tolfenamic acid might transcriptionally Met limiting the delivery of initiator met-tRNAi to translation upregulate expression of those two proapoptotic genes

OF8 Cancer Prev Res; 6(12) December 2013 Cancer Prevention Research

Tolfenamic Acid and ER Stress

A B Control PERK siRNA μ TA (50 μmol/L) –+–+ TA (50 mol/L) PERK 0 10 30 60 (min) 0 1 3 6 12 24 (h) p-elF2α (Ser51) Cyclin D1 Total elF2α Cyclin D1 ATF4 Actin CHOP

Actin

CDE EV ATF4 TA (50 μmol/L) Control ATF4 siRNA ATF4 01361224(h) –+–+TA (50 μmol/L) Bcl-2 Actin ATF4 c-PARP Actin CHOP ** ** 150,000 Bcl-2 15 100,000 c-PARP 10 content 50,000 1 5 0 Actin Luminescence 0 EV ATF4

% Sub-G DMSO TA (50 μmol/L)

Figure 6. Tolfenamic acid affects cyclin D1 and ATF4 expression via PERK-eIF2a axis. A, HCT-116 cells were treated with 50 mmol/L tolfenamic acid for the indicated times. Western blot analysis was performed for analysis of phosphorylated eIF2a, total eIF2a, cyclin D1, and actin expression. B, knockout of PERK was performed as described in Materials and Methods in HCT-116 cells, followed by incubating with 50 mmol/L tolfenamic acid for 24 hours, as indicated. Cell lysates were analyzed for PERK, Cyclin D1, ATF4, CHOP, and actin expression. C, HCT-116 cells were treated with 50 mmol/L tolfenamic acid for the indicated times. Western blot analysis was carried out using Bcl-2, PARP, and actin antibodies. The proportion of the sub-G1 phase of cells treated with 50 mmol/L tolfenamic acid for 24 hours was determined by flow cytometric analysis as described in Materials and Methods. The data are expressed as mean SD of three replicates. Results were considered statistically significant at , P < 0.01, compared with DMSO. D, The ATF4 construct was transfected into HCT-116 cells. Transfection efficiency was examined by Western blot using ATF4 antibody. Caspase-3/7 activity was measured as described in Materials and Methods, and the data are expressed as mean SD of three wells. Results were considered statistically significant at , P < 0.01, compared with EV. E, HCT-116 cells were transfected with siRNA against ATF4 for 24 h, followed by treatment with 50 mmol/L tolfenamic acid for 24 hours, as indicated. Cell lysates were subjected to Western blot analysis using ATF4, CHOP, Bcl-2, PARP, and actin antibodies. through regulation of ATF4 and CHOP, thereby causing model of intestinal neoplasia. The potential relevance of apoptosis. In addition, tolfenamic acid also activated ER our results to human familial and spontaneous colon stress-associated kinases (JNK) in colorectal cancer cells, cancer is provocative. Cyclin D1 overexpression has been which would trigger apoptosis as well (14). Although Sp demonstrated in adenomas from familial adenomatous transcriptional factors played an important role in tolfe- polyposis patients (53), and more than 60% of sponta- namic acid-mediated cell growth inhibition, that specific neous human colonic adenocarcinomas exhibit APC inhibition is likely to occur at a later stage. Here, we mutations (54). Our data are consistent with the recent provided evidence that ER stress response was an early event report that tolfenamic acid exhibits anticancer activity, of treatment with tolfenamic acid, whereas Sp down- as assessed by nude mice experiments (37). Along with regulation is later event affected by tolfenamic acid in the fact that each NSAID exhibits distinctive action in antitumorigenesis. anticancer activity, tolfenamic acid has great potential to We also reported here that tolfenamic acid dramatically be translated to the clinic for chemopreventive or thera- reduced the intestinal tumor load and polyp number in an peutic intervention of cancer. Moreover, our findings þ APCMin/ mouse model. Although we treated older mice suggestthattolfenamicacidwouldbeaverypowerful with tolfenamic acid for only 3 days, the number of polyps and effective chemopreventive agent for colorectal cancer and tumor load were still significantly decreased. The development, and inactivation of cyclin D1 would pro- results described here extend the involvement of cyclin vide one mechanism to support tolfenamic acid effects on þ D1 to the development of benign lesions in the APCMin/ chemoprevention.

www.aacrjournals.org Cancer Prev Res; 6(12) December 2013 OF9

Zhang et al.

In conclusion, the current study provides further evidence Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): K.-W. Min, M.F. McEntee, J. Jeong, for the antitumorigenic effect of tolfenamic acid on colorec- S.J. Baek tal cancer and elucidates the underlying molecular mechan- Study supervision: Q. Li, S.J. Baek isms of its action. Together with previous reports, our results suggest tolfenamic acid would be a promising chemopreven- Acknowledgments The authors thank Misty Bailey for her critical reading of this manuscript. tive agent for cancer. Novel tolfenamic acid derivatives that show similar safety profiles and enhanced anticancer effects would be attractive in ongoing investigations. Grant Support This work was supported by The University of Tennessee Center of fl Excellence in Livestock Diseases and Human Health grant (to S.J. Baek), Disclosure of Potential Con icts of Interest NIH grant R01CA108975 (to S.J. Baek), American Cancer Society grant (RSG- No potential conflicts of interest were disclosed 11-133-01-CCE; to S.-H. Lee), and a grant from the Ministry of Agriculture of the People’s Republic of China (2011ZX08008-003; to Q. Li). Financial Authors' Contributions support for X. Zhang was provided by the Program in Organizational or Conception and design: X. Zhang, S.-H. Lee, K.-W. Min, S.J. Baek Personal Cooperation with Foreign Counterparts (2010630161), China Development of methodology: M.F. McEntee, S.J. Baek Scholarship Council, China. Acquisition of data (provided animals, acquired and managed patients, The costs of publication of this article were defrayed in part by the provided facilities, etc.): M.F. McEntee payment of page charges. This article must therefore be hereby marked Analysis and interpretation of data (e.g., statistical analysis, bio- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate statistics, computational analysis): X. Zhang, S.-H. Lee, M.F. McEntee, this fact. S.J. Baek Writing, review, and/or revision of the manuscript: X. Zhang, K.-W. Min, Received June 20, 2013; revised August 30, 2013; accepted September 17, M.F. McEntee, S.J. Baek 2013; published OnlineFirst October 8, 2013.

References 1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 14. Lee SH, Bahn JH, Whitlock NC, Baek SJ. Activating transcription factor 2009. CA Cancer J Clin 2009;59:225–49. 2 (ATF2) controls tolfenamic acid-induced ATF3 expression via MAP 2. Kraus S, Arber N. Cancer: do aspirin and other NSAIDs protect against kinase pathways. Oncogene 2010;29:5182–92. colorectal cancer? Nat Rev Gastroenterol Hepatol 2011;8:125–6. 15. Musgrove EA, Caldon CE, Barraclough J, Stone A, Sutherland RL. 3. Mann JR, Backlund MG, DuBois RN. Mechanisms of disease: inflam- Cyclin D as a therapeutic target in cancer. Nat Rev Cancer 2011;11: matory mediators and cancer prevention. Nat Clin Pract Oncol 2005; 558–72. 2:202–10. 16. Jirawatnotai S, Hu Y, Michowski W, Elias JE, Becks L, Bienvenu F, et al. 4. Wang X, Baek SJ, Eling T. COX inhibitors directly alter gene expres- A function for cyclin D1 in DNA repair uncovered by protein inter- sion: role in cancer prevention? Cancer Metastasis Rev 2011;30: actome analyses in human cancers. Nature 2011;474:230–4. 641–57. 17. Alao JP. The regulation of cyclin D1 degradation: roles in cancer 5. Jones MK, Wang H, Peskar BM, Levin E, Itani RM, Sarfeh IJ, et al. development and the potential for therapeutic invention. Mol Cancer Inhibition of angiogenesis by nonsteroidal anti-inflammatory drugs: 2007;6:24. insight into mechanisms and implications for cancer growth and ulcer 18. Brewer JW, Hendershot LM, Sherr CJ, Diehl JA. Mammalian unfolded healing. Nat Med 1999;5:1418–23. protein response inhibits cyclin D1 translation and cell-cycle progres- 6. Hansen PE. Tolfenamic acid in acute and prophylactic treatment of sion. Proc Natl Acad Sci USA 1999;96:8505–10. migraine: a review. Pharmacol Toxicol 1994;75:81–2. 19. Szegezdi E, Logue SE, Gorman AM, Samali A. Mediators of endo- 7. Abdelrahim M, Baker CH, Abbruzzese JL, Safe S. Tolfenamic acid and plasmic reticulum stress-induced apoptosis. EMBO Rep 2006;7: pancreatic cancer growth, angiogenesis, and Sp protein degradation. 880–5. J Natl Cancer Inst 2006;98:855–68. 20. Tsutsumi S, Gotoh T, Tomisato W, Mima S, Hoshino T, Hwang HJ, et al. 8. Papineni S, Chintharlapalli S, Abdelrahim M, Lee SO, Burghardt R, Endoplasmic reticulum stress response is involved in nonsteroidal Abudayyeh A, et al. Tolfenamic acid inhibits esophageal cancer anti-inflammatory drug-induced apoptosis. Cell Death Differ 2004;11: through repression of specificity proteins and c-Met. Carcinogenesis 1009–16. 2009;30:1193–201. 21. Tsutsumi S, Namba T, Tanaka KI, Arai Y, Ishihara T, Aburaya M, et al. 9. Liu X, Abdelrahim M, Abudayyeh A, Lei P, Safe S. The nonsteroidal Celecoxib upregulates endoplasmic reticulum chaperones that inhibit anti-inflammatory drug tolfenamic acid inhibits BT474 and SKBR3 celecoxib-induced apoptosis in human gastric cells. Oncogene 2006; breast cancer cell and tumor growth by repressing erbB2 expression. 25:1018–29. Mol Cancer Ther 2009;8:1207–17. 22. Yamazaki T, Muramoto M, Oe T, Morikawa N, Okitsu O, Nagashima T, 10. Colon J, Basha MR, Madero-Visbal R, Konduri S, Baker CH, Herrera et al. Diclofenac, a non-steroidal anti-inflammatory drug, suppresses LJ, et al. Tolfenamic acid decreases c-Met expression through Sp apoptosis induced by endoplasmic reticulum stresses by inhibiting proteins degradation and inhibits lung cancer cells growth and caspase signaling. Neuropharmacology 2006;50:558–67. tumor formation in orthotopic mice. Invest New Drug 2011;29: 23. Hosoi T, Sasaki M, Baba S, Ozawa K. Effect of pranoprofen on 41–51. endoplasmic reticulum stress in the primary cultured glial cells. Neu- 11. Sankpal UT, Abdelrahim M, Connelly SF, Lee CM, Madero-Visbal R, rochem Int 2009;54:1–6. Colon J, et al. Small molecule tolfenamic acid inhibits PC-3 cell 24. Baek SJ, Kim JS, Moore SM, Lee SH, Martinez J, Eling TE. Cycloox- proliferation and invasion in vitro, and tumor growth in orthotopic ygenase inhibitors induce the expression of the tumor suppressor mouse model for prostate cancer. Prostate 2012;72:1648–58. gene EGR-1, which results in the up-regulation of NAG-1, an anti- 12. KangSU,ShinYS,HwangHS,BaekSJ,LeeSH,KimCH.Tolfe- tumorigenic protein. Mol Pharmacol 2005;67:356–64. namic acid induces apoptosis and growth inhibition in head and 25. Feng Q, Sekula D, Muller R, Freemantle SJ, Dmitrovsky E. Uncovering neck cancer: involvement of NAG-1 expression. PLoS ONE 2012;7: residues that regulate cyclin D1 proteasomal degradation. Oncogene e34988. 2007;26:5098–106. 13. Lee SH, Bahn JH, Choi CK, Whitlock NC, English AE, Safe S, et al. ESE- 26. Wang Y, Shen J, Arenzana N, Tirasophon W, Kaufman RJ, Prywes R. 1/EGR-1 pathway plays a role in tolfenamic acid-induced apoptosis in Activation of ATF6 and an ATF6 DNA binding site by the endoplasmic colorectal cancer cells. Mol Cancer Ther 2008;7:3739–50. reticulum stress response. J Biol Chem 2000;275:27013–20.

OF10 Cancer Prev Res; 6(12) December 2013 Cancer Prevention Research

Tolfenamic Acid and ER Stress

27. Baek SJ, Okazaki R, Lee SH, Martinez J, Kim JS, Yamaguchi K, et al. the expression of c-Met and survivin through suppressing specificity Nonsteroidal anti-inflammatory drug-activated gene-1 over expres- protein transcription factors. Gynecol Oncol 2011;122:163–70. sion in transgenic mice suppresses intestinal neoplasia. Gastroenter- 42. Harding HP, Zhang Y, Ron D. Protein translation and folding are ology 2006;131:1553–60. coupled by an endoplasmic-reticulum-resident kinase. Nature 1999; 28. Sukhthankar M, Yamaguchi K, Lee SH, McEntee MF, Eling TE, Hara Y, 397:271–4. et al. A green tea component suppresses posttranslational expression 43. Hamanaka RB, Bennett BS, Cullinan SB, Diehl JA. PERK and GCN2 of basic fibroblast growth factor in colorectal cancer. Gastroenterology contribute to eIF2alpha phosphorylation and cell cycle arrest after 2008;134:1972–80. activation of the unfolded protein response pathway. Mol Biol Cell 29. Wong NACS, Pignatelli M. b-catenin—A linchpin in colorectal 2005;16:5493–501. carcinogenesis? Am J Pathol 2002;160:389–401. 44. Romero-Ramirez L, Cao H, Nelson D, Hammond E, Lee AH, Yoshida H, 30. Takada Y, Bhardwaj A, Potdar P, Aggarwal BB. Nonsteroidal anti- et al. XBP1 is essential for survival under hypoxic conditions and is inflammatory agents differ in their ability to suppress NF-kappaB required for tumor growth. Cancer Res 2004;64:5943–7. activation, inhibition of expression of cyclooxygenase-2 and cyclin 45. Bi M, Naczki C, Koritzinsky M, Fels D, Blais J, Hu N, et al. ER stress- D1, and abrogation of tumor cell proliferation. Oncogene 2004;23: regulated translation increases tolerance to extreme hypoxia and 9247–58. promotes tumor growth. EMBO J 2005;24:3470–81. 31. Geng Y, Yu Q, Sicinska E, Das M, Bronson RT, Sicinski P. Deletion of 46. Lee AH, Iwakoshi NN, Anderson KC, Glimcher LH. Proteasome inhi- the p27Kip1 gene restores normal development in cyclin D1-deficient bitors disrupt the unfolded protein response in myeloma cells. Proc mice. Proc Natl Acad Sci USA 2001;98:194–9. Natl Acad Sci USA 2003;100:9946–51. 32. Diehl JA, Zindy F, Sherr CJ. Inhibition of cyclin D1 phosphorylation on 47. Nawrocki ST, Carew JS, Dunner K Jr, Boise LH, Chiao PJ, Huang P, threonine-286 prevents its rapid degradation via the ubiquitin-protea- et al. Bortezomib inhibits PKR-like endoplasmic reticulum (ER) kinase some pathway. Gene Dev 1997;11:957–72. and induces apoptosis via ER stress in human pancreatic cancer cells. 33. Zhang K, Kaufman RJ. Protein folding in the endoplasmic reticulum Cancer Res 2005;65:11510–9. and the unfolded protein response. Handb Exp Pharmacol 2006;172: 48. Nawrocki ST, Carew JS, Pino MS, Highshaw RA, Dunner K Jr, 69–91. Huang P, et al. Bortezomib sensitizes pancreatic cancer cells to 34. Kim I, Xu W, Reed JC. Cell death and endoplasmic reticulum stress: endoplasmic reticulum stress-mediated apoptosis. Cancer Res disease relevance and therapeutic opportunities. Nat Rev Drug Discov 2005;65:11658–66. 2008;7:1013–30. 49. Stolfi C, Sarra M, Caruso R, Fantini MC, Fina D, Pellegrini R, et al. 35. Armstrong JL, Flockhart R, Veal GJ, Lovat PE, Redfern CP. Regulation Inhibition of colon carcinogenesis by 2-methoxy-5-amino-N-hydro- of endoplasmic reticulum stress-induced cell death by ATF4 in neu- xybenzamide, a novel derivative of mesalamine. Gastroenterology roectodermal tumor cells. J Biol Chem 2010;285:6091–100. 2010;138:221–30. 36. Carracedo A, Gironella M, Lorente M, Garcia S, Guzman M, Velasco G, 50. Pyrko P, Kardosh A, Liu YT, Soriano N, Xiong W, Chow RH, et al. et al. Cannabinoids induce apoptosis of pancreatic tumor cells via Calcium-activated endoplasmic reticulum stress as a major compo- endoplasmic reticulum stress-related genes. Cancer Res 2006;66: nent of tumor cell death induced by 2,5-dimethyl-celecoxib, a non- 6748–55. coxib analogue of celecoxib. Mol Cancer Ther 2007;6:1262–75. 37. Pathi S, Li X, Safe S. Tolfenamic acid inhibits colon cancer cell and 51. Jiang HY, Wek SA, McGrath BC, Lu D, Hai T, Harding HP, et al. tumor growth and induces degradation of specificity protein (Sp) Activating transcription factor 3 is integral to the eukaryotic initi- transcription factors. Mol Carcinog 2013 May 14. [Epub ahead of print]. ation factor 2 kinase stress response. Mol Cell Biol 2004;24: 38. Jeong JB, Yang X, Clark R, Choi J, Baek SJ, Lee SH. A mechanistic 1365–77. study of the proapoptotic effect of tolfenamic acid: involvement of NF- 52. Reimertz C, Kogel D, Rami A, Chittenden T, Prehn JH. Gene expression kappaB activation. Carcinogenesis 2013;34:2350–60. during ER stress-induced apoptosis in neurons: induction of the BH3- 39. Zhang X, Min KW, Liggett J, Baek SJ. Disruption of the transforming only protein Bbc3/PUMA and activation of the mitochondrial apoptosis growth factor-beta (TGF-beta) pathway by tolfenamic acid via the pathway. J Cell Biol 2003;162:587–97. ERK MAP kinase pathway. Carcinogenesis 2013 July 16. [Epub ahead 53. Zhang T, Nanney LB, Luongo C, Lamps L, Heppner KJ, DuBois RN, of print]. et al. Concurrent overexpression of cyclin D1 and cyclin-dependent 40. Eslin D, Sankpal UT, Lee C, Sutphin RM, Maliakal P, Currier E, et al. kinase 4 (Cdk4) in intestinal adenomas from multiple intestinal neo- Tolfenamic acid inhibits neuroblastoma cell proliferation and induces plasia (Min) mice and human familial adenomatous polyposis patients. apoptosis: a novel therapeutic agent for neuroblastoma. Mol Carcinog Cancer Res 1997;57:169–75. 2013;52:377–86. 54. Powell SM, Petersen GM, Krush AJ, Booker S, Jen J, Giardiello FM, 41. Basha R, Ingersoll SB, Sankpal UT, Ahmad S, Baker CH, Edwards JR, et al. Molecular diagnosis of familial adenomatous polyposis. N Engl J et al. Tolfenamic acid inhibits ovarian cancer cell growth and decreases Med 1993;329:1982–7.

www.aacrjournals.org Cancer Prev Res; 6(12) December 2013 OF11

The Involvement of Endoplasmic Reticulum Stress in the Suppression of Colorectal Tumorigenesis by Tolfenamic Acid

Xiaobo Zhang, Seong-Ho Lee, Kyung-Won Min, et al.

Cancer Prev Res Published OnlineFirst October 8, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1940-6207.CAPR-13-0220

Supplementary Access the most recent supplemental material at: Material http://cancerpreventionresearch.aacrjournals.org/content/suppl/2013/10/08/1940-6207.CAPR -13-0220.DC2

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://cancerpreventionresearch.aacrjournals.org/content/early/2013/11/19/1940-6207.CAPR- 13-0220. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.