Vitamin E D-Tocotrienol Augments the Antitumor Activity of Gemcitabine and Suppresses Constitutive NF-Kb Activation in Pancreatic Cancer

Published OnlineFirst October 4, 2011; DOI: 10.1158/1535-7163.MCT-11-0424

Molecular Cancer Preclinical Development Therapeutics

Vitamin E d-Tocotrienol Augments the Antitumor Activity of Gemcitabine and Suppresses Constitutive NF-kB Activation in Pancreatic Cancer

Kazim Husain1, Rony A. Francois1, Teruo Yamauchi1, Marta Perez1, Said M. Sebti2, and Mokenge P. Malafa1,2

Abstract The NF-kB transcription factor functions as a crucial regulator of cell survival and chemoresistance in pancreatic cancer. Recent studies suggest that tocotrienols, which are the unsaturated forms of vitamin E, are a promising class of anticancer compounds that inhibit the growth and survival of many cancer cells, including pancreatic cancer. Here, we show that tocotrienols inhibited NF-kB activity and the survival of human pancreatic cancer cells in vitro and in vivo. Importantly, we found the bioactivity of the four natural tocotrienol compounds (a-, b-, d-, and g-tocotrienol) to be directly related to their ability to suppress NF-kB activity in vitro and in vivo. The most bioactive tocotrienol for pancreatic cancer, d-tocotrienol, significantly enhanced the efficacy of gemcitabine to inhibit pancreatic cancer growth and survival in vitro and in vivo. Moreover, we found that d-tocotrienol augmentation of gemcitabine activity in pancreatic cancer cells and tumors is associated with significant suppression of NF-kB activity and the expression of NF-kB transcriptional targets

(Bcl-XL, X-linked inhibitor of apoptosis, and survivin). Our study represents the ﬁrst comprehensive preclinical evaluation of the activity of natural vitamin E compounds in pancreatic cancer. Given these results, we are conducting a phase I trial of d-tocotrienol in patients with pancreatic cancer using pancreatic tumor cell survival and NF-kB signaling components as intermediate biomarkers. Our data also support future clinical investigation of d-tocotrienol to augment gemcitabine activity in pancreatic cancer. Mol Cancer Ther; 10(12); 2363–72. 2011 AACR.

Introduction pancreatic cancer (11–14). However, translation of these studies to the clinic has been challenging due to the low Pancreatic cancer is a leading cause of cancer mortality bioavailability of some of these natural compounds in with less than 5% of patients surviving 5 years after humans. diagnosis (1). Gemcitabine is a mainstay treatment for Tocotrienols, a group of 4 (a-, b-, d-, and g-tocotrienol) pancreatic cancer (2). However, tumor resistance to gem- unsaturated naturally occurring vitamin E compounds citabine therapy is common, making this a critical chal- (Fig. 1A), have received increasing attention for their lenge for pancreatic cancer management. potential as nontoxic dietary anticancer agents (15, 16). Prosurvival signaling mediated by the transcription Our group recently showed that oral administration of factor NF-kB is a key player in gemcitabine resistance in 100 mg/kg/d of d-tocotrienol to mice resulted in levels pancreatic cancer (3–11). Therefore, targeting dysregu- that were 10 times higher in pancreas than in subcutane- lated NF-kB signaling is an important strategy to improve ously implanted tumor tissue, suggesting that these com- the efficacy of gemcitabine therapy and thereby improve pounds will have reasonable bioavailability for pancreatic outcomes for patients with pancreatic cancer. Several tumor intervention (17). In this study, we investigated the studies have combined natural compounds that inhibit potential of the 4 natural tocotrienols to inhibit pancreatic NF-kB, such as genistein, curcumin, fisetin, and green tea, cancer and NF-kB activation in vitro and in vivo. In addi- with gemcitabine to investigate synergy in treating tion, we investigated the potential of the most bioactive tocotrienol to augment gemcitabine activity in vitro and in vivo. Authors' Affiliations: Departments of 1Gastrointestinal Oncology and 2Drug Discovery, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida Materials and Methods Corresponding Author: Mokenge P. Malafa. Department of Gastrointes- tinal Oncology, Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL Chemicals and cell lines 33612. Phone: 813-745-1432; Fax: 813-745-7229; E-mail: All chemicals and reagents were purchased from Sigma- fi mokenge.malafa@mof tt.org Aldrich unless otherwise specified. Tocotrienols (a, b, g, doi: 10.1158/1535-7163.MCT-11-0424 and d)anda-andd-tocopherol were obtained from Davos 2011 American Association for Cancer Research. Life Ltd. Gemcitabine-HCl was purchased from Eli Lilly

www.aacrjournals.org 2363

Husain et al.

CH3 140 ABHO AsPc-1 MiaPaCa-2 CH3 CH3 CH3 CH3 120 CH 3 O CH3 100 CH α # 3 -Tocotrienol # CH 80 # # 3 # * * * # HO # * * CH CH CH 60 * * 3 3 3 * * * * CH3 * * * * * * * O CH3 40 * * * % cell viability CH β * * 3 -Tocotrienol 20 * * HO CH CH CH 0 CH 3 3 3 3 icle mol/L mol/L mol/L mol/L mol/L mol/L mol/L mol/L eh mol/L mol/L mol/L mol/L mol/L mol/L mol/L CH O CH μ μ μ V μ 3 3 0 10 20 μ 40 μ 50 μ 60 80 μ 10 μ 20 μ 40 μ 50 60 μ 80 μ γ 100 μ 100 μ CH3 -Tocotrienol 140 HPDE6 C7 HPDE6 C7-Kras HO 120 CH3 CH3 CH3 CH3 100 CH # O 3 # * CH δ-Tocotrienol 80 * 3 * 60 * * CH3 * * * HO 40 * CH3 CH3 CH3 % cell viability CH3 20 * CH3 O CH3 * 0 CH3 α-Tocopherol

mol/L mol/L μmol/Lμmol/L mol/L mol/Lμmol/L mol/L μmol/L mol/L mol/L mol/L μmol/L mol/L mol/L mol/L 0 μ 0 10 μ 20 40 50 μ 60 μ 80 10 μ 20 μ 40 μ 50 60 μ 80 μ C 100 μ 100 μ Caspase-8 activity Caspase-3 activity α-Tocotrienol β-Tocotrienol γ-Tocotrienol δ-Tocotrienol α-Tocopherol 160 180 ** 140 * * 160 ** * 120 140 D 100 120 80 100 80 60 60 700 * P < 0.001 vs. vehicle 40 P 40 ** < 0.02 vs. vehicle 20 600 *** P < 0.05 vs. vehicle 20 # P < 0.02 vs. γ-T3 0 0 500 *** Caspase-8 activity (%) V α-T3 β-T3 γ -T3 δ-T3 α-T Caspase-3 activity (%) V α-T3 β-T3 γ -T3 δ-T3 α-T 400 ** α β Cell death ELISA Vehicle -T3 -T3 300 14 V α-T3 β-T3 γ-T3 δ-T3 200 *,# 12 No. of colonies PARP1 100 10 cleaved 8 0 Vehicle α-T3 β-T3 γ-T3 δ-T3 α-T 6 Bax 4 β-Actin γ-T3 δ-T3 α-T 2 0 MiaPaCa-2 cells

10-fold increase in apoptosis α-T3 β-T3 γ-T3 δ-T3 α-T

Figure 1. A, chemical structures of tocotrienols and tocopherol. B, effect of tocotrienols on pancreatic cancer cell proliferation (MTT assay). Points, means; bars, SE (n ¼ 3–5; , P < 0.001; #, P < 0.01). C, effect of tocotrienols on caspase-3 and 8 activities in MiaPaCa-2 cells (top left and right). Bars, SE (n ¼ 3; , P < 0.05; , P < 0.01). Bottom left, effect of tocotrienols on pancreatic cancer apoptotic cell death (ELISA) in MiaPaCa-2 cells (n ¼ 3). Bottom right, Western blot analysis of PARP-1 cleavage and proapoptotic Bax protein expression in MiaPaCa-2 cells (n ¼ 3). D, effect of tocotrienols on pancreatic cancer malignant transformation assay in MiaPaCa-2 cells. d-Tocotrienol signiﬁcantly inhibited malignant transformation compared with vehicle (, P < 0.001) and b-or g-tocotrienol treatment (#, P < 0.02). In contrast, no effect was observed with a-tocotrienol or a-tocopherol treatment. Bars, SE (n ¼ 3). V, vehicle.

and Company. L-Glutamine, penicillin, streptomycin, and tary extract. MiaPaCa-2 and AsPc-1 cells were grown in HEPES buffer were purchased from Mediatech. FBS was monolayers with DMEM and RPMI media, respectively, purchased from HyClone. Dulbecco’s Modified Eagle’s supplemented with 10% FBS, 2 mmol/L L-glutamine, Medium (DMEM), RPMI-1640, keratinocyte serum-free penicillin (50 IU/mL), and streptomycin (50 mg/mL). medium, sodium pyruvate, trypsin-EDTA, and PBS were RPMI medium was also supplemented with 10 2 mol/L purchased from Invitrogen. Ethanol (100%) was purchased HEPES buffer and 10 3 mol/L sodium pyruvate. Cells from Aaper Alcohol and Chemical. Pancreatic cancer cell were cultured at 37C in a humidified atmosphere of 5% lines MiaPaCa-2 and AsPc-1 were obtained from American CO2 and 95% O2. Type Culture Collection (ATCC). Human pancreatic ductal epithelial cells (HPDE6 C7) and HPDE6 C7-KRas cells were Cell proliferation MTT assay gifts from Dr. Tsao, Ontario Cancer Institute, University of Cells were seeded in 96-well plates at a density of 3,000 Toronto, Toronto, Ontario, Canada. These cells were cells per well and allowed to attach overnight. Cells were authenticated by morphologic, cell proliferation, and Myco- then incubated for 72 hours with various concentrations of plasma tests, as recommended in ATCC Technical Bulletin a-, b-, g-, and d-tocotrienol and a-tocopherol (10 5 to 10 4 No. 8 (2007). mol/L). In other experiments, cells were incubated for 72 hours with gemcitabine and d-tocotrienol (10 5 to 10 4 Cell culture and growth mol/L) and their combination or ethanol (<5%) in PBS as HPDE6 C7 and HPDE6 C7-KRas cells were grown in vehicle control. Media were aspirated and replaced with keratinocyte growth media (Invitrogen) supplemented 20 mL of 1 mg/mL MTT and incubated for 2 to 4 hours at with human epidermal growth factor and bovine pitui- 37 C in a humidified atmosphere of 5% CO2. Media were

2364 Mol Cancer Ther; 10(12) December 2011 Molecular Cancer Therapeutics

Tocotrienol Activity in Pancreatic Cancer

aspirated, 200 mL of dimethyl sulfoxide was added to each Caspase enzyme activity assay well, plates were incubated for 5 minutes with shaking, Treated cells were lysed with lysis buffer (0.5 mmol/L and absorbance was read at 540 nm. Tris/pH 8.0, 5 mmol/L EDTA/pH 8, 0.15 mmol/L NaCl, 0.5% NP-40). The enzymatic activities of caspase-3, 8, and Isobologram analysis 9 were determined using specific fluorogenic substrates. Gemcitabine and d-tocotrienol were combined at The liberated fluorescent 7-amido-4-methyl-coumarin ratios of 1:1 and 1:2 of their IC50 values to plot the groups were quantified using a multiwell plate Versa- isobologram using fraction effects and combination Fluor fluorometer at 355-nm excitation and 460-nm emis- index (CI) through CalcuSyn software (Biosoft). Con- sion wavelengths (Bio-Rad). stant ratio (IC50 ratio) of the 2-drug combination is used. The dose range (different fraction of the IC50 values) and Cell lysis and protein determination effects (proliferation) are entered in CalcuSyn, which Cells were washed 3 times in cold PBS (pH 7) and then D m r displays the parameters ( m, ,and )aswellas lysed in mammalian protein extraction reagent lysis buffer D isobologram using CI, where m is the median effect (Pierce) containing EDTA and protease inhibitor cocktail. dose signifying the potency, m is measurement of the Protein concentration was determined using BCA reagents sigmoidicity (shape) of the dose–effect curve, and r is (Pierce) according to manufacturer’s instructions. linear correlation coefficient of the median effect plot. The software analyzes the quantitative measure of the Western blot analysis degree of drug interactions in terms of additive effect Extracted proteins from cells and tumor tissues (40 mg) (CI ¼ 1), synergism (CI < 1), or antagonism (CI > 1), were resolved on 12.5% SDS-PAGE running gel and 5% basedonapublishedmethod(18). stacking gel. Proteins were then electrotransferred onto nitrocellulose membranes. After blocking in 5% nonfat Trypan blue dye exclusion assay powdered milk for 1 hour, membranes were washed and Treated cells were trypsinized and washed with PBS, treated with antibodies to CK18, PARP-1, NF-kB subunits and cell suspension (50 mL) was mixed with 50 mLof (p65/p50), IkBa, p-IkBa, Bcl-XL, X-linked inhibitor of trypan blue dye and incubated for 2 minutes at room apoptosis (XIAP), survivin, and b-actin (1:1,000 and temperature. A 10-mL volume was loaded onto a hemo- 1:5,000) overnight at 4C (Axxora; Santa Cruz Biotechnol- cytometer, and cells were scored as live or dead on the ogy; Cell Signaling). After they were washed, blots were basis of trypan blue exclusion. incubated with horseradish peroxidase–conjugated sec- ondary antibody IgG (1:5,000 and 1:10,000) for 1 hour Cell death assay at room temperature. Washed blots were then treated Cytoplasmic DNA fragments and histone were with SuperSignal West Pico chemiluminescent substrate detected in cells using Cell Death Detection ELISA kit (Pierce) for positive antibody reaction. Membranes were from Roche Diagnostic. exposed to X-ray film (Kodak) for visualization and densitometric quantization of protein bands using Colonogenic survival (anchorage-independent) AlphaEaseFC software (Alpha Innotech). growth assay The standard soft agar colony formation assay was Animals and treatments conducted in MiaPaCa-2 cells. Cells were seeded at a Female NIH severe-combined immunodeficient (SCID) density of 5,000 per well in 12-well plates in 0.3% agar nude mice (4–5 weeks old, 20–25 g) were obtained from over a 0.6% bottom agar layer. Colonies were fed with Charles River and kept in our Center’s animal facility for 1 growth media and a-, b-, g-, and d-tocotrienol or week for quarantine. Mice (n ¼ 25) were injected with a/d-tocopherol (5 10 5 mol/L), and colony formation AsPc-1 cells (one million) with Matrigel (100 mL) to both growth was observed for 14 days. In other experiments, flanks. When tumor volume reached 100 mm3, mice were colonies were fed with growth media and gemcitabine randomly divided into the following 5 treatment groups: (2 10 5 mol/L) and d-tocotrienol (5 10 5 mol/L) and (i) normal controls: 100 mL oral gavage of ethanol extracted their combination, and colony formation growth was olive oil (vehicle) twice daily for 4 weeks, (ii) a-tocotrienol observed for 14 days. Colonies were photographed after treated, (iii) b-tocotrienol treated, (iv) g-tocotrienol trea- overnight incubation with 1 mg/mL MTT in the wells. ted, and (v) d-tocotrienol treated. Tocotrienols were Colonies were counted under stereomicroscope and com- administered at 200 mg/kg oral gavage twice daily for pared with controls. Experiments were done at least twice, 4 weeks. In other experiments, mice (n ¼ 20) were injected each in triplicate. with AsPc-1 cells (one million) with Matrigel (100 mL) to both flanks. When tumor volume reached 100 mm3, NF-kB (p65) binding activity assay mice were randomly divided into the following 4 treat- NF-kB binding activities in nuclear and cytosolic frac- ment groups: (i) normal controls: 100 mL oral gavage of tions of MiaPaCa-2 cells were determined using the Trans ethanol extracted olive oil (vehicle) twice daily for 4 AM ELISA Kit from Active Motif, according to manufac- weeks, (ii) d-tocotrienol treated (200 mg/kg oral gavage turer’s instructions. twice daily for 4 weeks), (iii) gemcitabine treated

www.aacrjournals.org Mol Cancer Ther; 10(12) December 2011 2365

Husain et al.

(100 mg/kg intraperitoneally twice a week for 4 weeks), potency of the tocotrienols in inhibiting colony formation and (iv) d-tocotrienol þ gemcitabine treated (200 mg/kg was 65% for d-tocotrienol, 30% for g-tocotrienol, 13% for oral gavage daily plus gemcitabine at 100 mg/kg intra- b-tocotrienol, and 2% for a-tocotrienol. peritoneally twice a week for 4 weeks). Tumor volumes were measured every other day. Animal studies were Effect of tocotrienols on growth and survival of approved by our Institutional Laboratory Animal Care pancreatic cancer cells in vivo and Use Committee and were conducted per the guide- On the basis of our in vitro studies, which strongly lines of the NIH. support differential effects of natural tocotrienols on pancreatic cancer cells, we examined the effects of tocotrienol Statistical analysis on the growth of AsPc-1 human pancreatic cancer xeno- Data are expressed as means SEM and were analyzed grafts in mice. Such studies have never been documented statistically using unpaired t tests or one-way ANOVA, in vivo to the best of our knowledge. A dose of 200 mg/kg where appropriate. ANOVA was followed by Duncan tocotrienol administered by intragastric gavage twice multiple range tests using SAS statistical software for a day was selected based on our previously publish- comparison between different treatment groups. Signifi- ed report (17). Treatment started after randomization cance was set at P < 0.05. and continued per experimental protocol for 34 days. Figure 2B shows a gradual increase in tumor volume in Results the control and a-tocotrienol groups. Tumor volumes in the b-, d-, and g-tocotrienol groups were also significantly Effect of tocotrienols on growth and survival of lower from day 15 (P < 0.05) to end of treatment (P < 0.01 pancreatic cancer cells in vitro for b- and g-tocotrienol and P < 0.001 for d-tocotrienol). Using the MTT assay, we found that natural tocotrie- On day 34, d-tocotrienol significantly reduced tumor nols have variable inhibitory effects on human pancre- volume by 50% (P < 0.001), g-tocotrienol reduced tumor atic cancer cells, with d-tocotrienol and g-tocotrienol volume by 42% (P < 0.01), and b-tocotrienol reduced exhibiting the most significant inhibitory effects (cell tumor volume by 32% (P < 0.01). In contrast, a-tocotrienol viability reduced up to 60% with 50 mmol/L) and did not significantly decrease tumor growth. b-tocotrienol having a moderate inhibitory effect (cell We next examined expression of cell apoptosis markers viability reduced up to 40% with 50 mmol/L; Fig. 1B). CK18 and Bax in AsPc-1 cells in vitro and in tumor tissues. Both a-tocotrienol and saturated vitamin E (a-tocoph- As shown in Fig. 2C and D, g- and d-tocotrienol signifi- erol) did not show any significant inhibitory effects. In cantly induced expression of CK18 and Bax. contrast, treatment of HPDE6 C7 cells resulted in essen- tially no effect on cell viability. HPDE6 C7-KRas cells Effect of tocotrienols on NF-kB activation in were sensitive to the effects of some of the tocotrienols, pancreatic cancer cells with 50 mmol/L tocotrienol resulting in 55%, 30%, 15%, Because previous studies have shown that g-tocotrie- and 0% growth inhibition for d-, g-, b-, and a-tocotrienol can suppress constitutive NF-kB activation in pan- nol, respectively. creatic cancer cells (19), the possibility that tocotrienol To further assess whether the loss of viability could, in bioactivity in pancreatic cancer is related to NF-kB part, be due to apoptosis, we evaluated the effects of suppression was considered. As illustrated in Fig. 3A, tocotrienols on apoptosis using histone-DNA ELISA, cas- tocotrienol treatment suppressed constitutive NF-kB pase-3, and caspase-8 activity as well as Western immu- activation in a tocotrienol compound–dependent man- noblotting for PARP-1 and the proapoptotic protein Bax in ner. The extent of NF-kB activity inhibition was found to MiaPaCa-2 cells. Figure 1C shows a significant increase in be significant for g-andd-tocotrienol in the nuclear apoptotic cells, closely paralleling the loss of viable cells extract and for b-, g-, and d-tocotrienol in the cytosolic following treatment with 50 mmol/L tocotrienol. The extract. Interestingly, a-tocotrienol and a-tocopherol activities of caspase-3 and 8 were also significantly had no effect on NF-kB activity. These results were increased after treatment of MiaPaCa-2 cells with d-toco- further confirmed by immunoblotting experiments in trienol and g-tocotrienol. Western immunoblotting AsPc-1 and MiaPaCa-2 cells (Fig. 3B) and in tumor revealed that tocotrienol treatment resulted in the appear- tissues of mice xenografted with AsPc-1 cells (Fig. ance of a cleaved active component of PARP-1 and induc- 3C). A crucial step in the activation of NF-kBproteins tion of Bax in MiaPaCa-2 cells treated with 50 mmol/L d-, is the phosphorylation of IkBa freeing NF-kBcomplexes g-, and b-tocotrienol for 72 hours. to translocate to the nucleus where they induce target Tocotrienols also influenced colony formation of gene expression, including prosurvival factors such as human pancreatic cancer cells in soft agar, which is one Bcl-XL, survivin, and XIAP. As shown in Fig. 3B and C, of the best in vitro indicators of malignant behavior. As d-tocotrienol consistently suppressed NF-kB/p65 and shown in Fig. 1D, treatment of MiaPaCa-2 cells with 50 phosphorylated IkBa expression in pancreatic cancer mmol/L tocotrienol resulted in significantly fewer and cells and tumor tissues, consistent with downregulation smaller colonies than shown in cells treated with vehicle of Bcl-XL, survivin, and XIAP in pancreatic cancer cells control or the saturated form of vitamin E. The relative and tumor tissues. Consistent with a previous report,

2366 Mol Cancer Ther; 10(12) December 2011 Molecular Cancer Therapeutics

Tocotrienol Activity in Pancreatic Cancer

A B Orthotopic injection 1,000 *P < 0.05 vs. vehicle control AsPc-1 cells Treatment Control vehicle **P < 0.01 vs. vehicle control Randomization Stop Sacrifice α-Tocotrienol ***P < 0.001 vs. vehicle control

) β-Tocotrienol

3 800 γ-Tocotrienol δ 1 Vehicle -Tocotrienol α-T3 600 ** 2 ** * β * 3 -T3 * ** ** * * 400 * 4 γ-T3 * * *** * ** 5 δ ** -T3 ** ** ** * **

Tumor volume (mm 200 ** ** ** d -7 d 0 d 32 d 33 * 0 35302520151050 Days posttreatment C V α-T3 β-T3 γ-T3 δ-T3 D V α-T3 β-T3 γ-T3 δ-T3 CK18 CK18

Bax Bax

β-Actin β-Actin

AsPc-1 cells Tumor tissue

Figure 2. A, effect of in vivo tocotrienols on AsPc-1 pancreatic tumor growth in SCID nude mice. B, tumor volume in d-, g-, and b-tocotrienol groups was significantly decreased compared with vehicle. Points, means; bars, SE (n ¼ 10; , P < 0.05; , P < 0.01; , P < 0.001). C, expression of apoptosis marker CK18 and proapoptotic Bax protein in AsPc-1 cells (n ¼ 3) by Western blot analysis. D, CK18 and Bax expression in mouse tumor tissues (n ¼ 5) by Western blot analysis. V, vehicle. g-tocotrienol also suppressed NF-kB proteins and their 4B). We further investigated the effects of gemcitabine target genes (19). However, a-andb-tocotrienol did not and d-tocotrienol alone and in combination on anchor- suppress the NF-kBproteins,p-IkBa or Bcl-XL, survi- age-independent growth of MiaPaCa-2 cells using soft vin, and XIAP, in a consistent fashion. agar colony formation assay (Fig. 4D). Gemcitabine (20 mmol/L) alone inhibited colony formation by 84%; d-Tocotrienol augments inhibition of pancreatic d-tocotrienol (50 mmol/L) alone inhibited colony formation cancer cell proliferation by gemcitabine by 67%. Gemcitabine (20 mmol/L) plus d-tocotrienol (50 Recently, g-tocotrienol was reported to augment gemci- mmol/L) resulted in 99% inhibition of anchorage-indepen- tabine activity and to inhibit NF-kB activation in pancreatic dent growth. cancer cells (19). Because we observed that both g-and d-tocotrienol significantly inhibited NF-kB activation, we d-Tocotrienol augments gemcitabine-induced investigated whether d-tocotrienol augmented gemcita- apoptosis in pancreatic cancer cells bine inhibition of pancreatic cancer growth. The effects of We next compared whether enhanced cytotoxicity d-tocotrienol and gemcitabine alone and in combination on by gemcitabine þ d-tocotrienol was mediated by AsPc-1 and MiaPaCa-2 cell proliferation were determined apoptosis. Figure 4C shows that, relative to single-agent by MTT. Cells were treated with vehicle, d-tocotrienol, gemcitabine, combination with d-tocotrienol elicited a gemcitabine, or d-tocotrienol þ gemcitabine at (0–100 significantly (P < 0.001, P < 0.01) higher percentage of cell mmol/L) for 72 hours (Fig. 4A). At 10 to 100 mmol/L, death in pancreatic cancer cell lines, suggesting that the d-tocotrienol þ gemcitabine significantly augmented gem- loss of viable cells is due to the induction of cell death citabine inhibition of cell proliferation. Furthermore, to pathway. We confirmed this by measuring the fold confirm synergism, we determined CI values for 2 combi- increase in apoptosis in MiaPaCa-2 cells using the cell nation treatment groups. Our results show that cells treated death ELISA and the caspase-3 activity assay. While with gemcitabine þ d-tocotrienol at a ratio of 1:1 had treatment of MiaPaCa-2 cells with gemcitabine and additive/antagonistic loss of cell viability (CI ¼ 1.32), d-tocotrienol as single agents increased apoptosis, gemci- whereas cells treated with gemcitabine þ d-tocotrienol at tabine þ d-tocotrienol significantly augmented the fold 1:2 showed synergistic loss of cell viability (CI ¼ 0.89; Fig. increase in apoptosis. Caspase-3 activity was also

www.aacrjournals.org Mol Cancer Ther; 10(12) December 2011 2367

Husain et al.

A 120 120 κ NF- B/p65 activity (nuclear) NF-κB/p65 activity (cytosolic) 100 100 * * ** 80 * 80 **

60 60

40 40 B/p65 activity (%) B/p65 activity (%) κ κ NF- NF- 20 20

0 0 α β γ δ α Vehicle -T3 -T3 -T3 -T3 -T Vehicle α-T3 β-T3 γ-T3 δ-T3 α-T

B α β γ δ V -T3 -T3 -T3 -T3 V α-T3 β-T3 γ-T3 δ-T3 NF-κB/p65 NF-κB/p65 NF-κB/p50 NF-κB/p50 p-IκBα (Ser32/36) p-IκBα (Ser32/36) IκBα IκBα β -Actin β-Actin AsPc-1 cells MiaPaCa-2 cells α β γ δ V -T3 -T3 -T3 -T3 V α-T3 β-T3 γ-T3 δ-T3 Bcl-XL Bcl-X Survivin L Survivin XIAP β -Actin XIAP AsPc-1 cells β-Actin C MiaPaCa-2 cells V α-T3 β-T3 γ-T3 δ-T3 NF-κB/p65 V α-T3 β-T3 γ-T3 δ-T3 NF-κB/p50 Bcl-XL p-IκBα (Ser32/36) Survivin IκBα XIAP β-Actin β-Actin Tumor tissue Tumor tissue

Figure 3. A, effect of tocotrienols on nuclear and cytosolic NF-kB/p65 binding activity in MiaPaCa-2 cells. b-, g-, and d-tocotrienol signiﬁcantly decreased p65 binding to DNA in the cytosol, whereas g- and d-tocotrienol signiﬁcantly decreased p65 binding to DNA in the nucleus compared with vehicle. Bars, SE (n ¼ 3; , P < 0.05; , P < 0.02). B, effect of tocotrienols on NF-kB subunits, IkBa phosphorylation, and NF-kB–associated gene products in AsPc-1 and MiaPaCa-2 cells (n ¼ 3) by Western blot analysis. C, effect of tocotrienol treatment on p65 and p50 subunit, p-IkBa, and antiapoptotic protein expression in mouse tumor tissue (n ¼ 5) by Western blot analysis. V, vehicle.

observed to be significantly elevated (P < 0.001) in the 72 hours, and their nuclear extracts were subjected to combination treatment group. PARP-1, a 116-kDa nuclear NF-kB DNA binding activity assay (ELISA). We found PARP-1, is one of the main cleavage targets of caspase-3 in that gemcitabine alone did not significantly affect NF- vivo, serving as a marker of cells undergoing apoptosis kB DNA binding activity (Fig. 5A). Interestingly, treat- (20). Our data confirm induction of apoptosis by gemci- ment with d-tocotrienol significantly inhibited binding tabine, d-tocotrienol, and the combination of the 2 agents activity, with combined gemcitabine and d-tocotrienol in AsPc-1 and MiaPaCa-2 cells. significantly inhibiting NF-kB DNA binding activity more than either drug alone. These results were further d-Tocotrienol downregulates constitutively confirmed by Western immunoblotting in AsPc-1 and activated NF-kB in gemcitabine-treated pancreatic MiaPaCa-2 cells. As shown in Fig. 5B, expression levels cancer cells of NF-kB/p65 and NF-kB/p50 proteins were signifi- MiaPaCa-2 cells were exposed to 50 mmol/L d-toco- cantly depleted in cells treated with d-tocotrienol and in trienol, 20 mmol/L gemcitabine, or their combination for those treated with gemcitabine þ d-tocotrienol. In

2368 Mol Cancer Ther; 10(12) December 2011 Molecular Cancer Therapeutics

Tocotrienol Activity in Pancreatic Cancer

A 120 120 B Combo - algebraic estimate AsPc-1 MiaPaCa-2 100 100 5.0 * * 4.0 1:1 80 80 * * 3.0 * * 60 * 60 * * * * * 2.0 a * * * a * * * * * * * * CI 1.96SD 40 a * * 40 a * # * # * 1.0 a cell viability% % cell viability cell viability % Vehicle * Vehicle a # * # * Gemcitabine Gemcitabine a 20 a # * 20 * # 0 δ-T3 * δ-T3 * a # * 0 0.2 0.4 0.6 0.8 1.0 Gem + δ-T3 * # Gem + δ-T3 a a * # 0 * 0 * * Fractional effect 0 10 20 40 50 60 80 100 μmol/L 0 10 20 40 50 60 80 100 μmol/L Combo - algebraic estimate 4.0

3.0 1:2 C 60 MiaPaCa-2 cells AsPc-1 cells a, c, d 60 a P 50 a-P < 0.001 vs. vehicle a, c, d 2.0 - < 0.001 vs. vehicle b b-P < 0.02 vs. gem -P < 0.02 vs. gem c c P δ a, b -P < 0.02 vs. δ-T3 a, b 40 - < 0.02 vs. -T3 CI 1.96 SD d P d-P < 0.01 vs. gem 1.0 40 - < 0.001 vs. gem 30 a 0 20 0 0.2 0.4 0.6 0.8 1.0 20 a % dead cells dead % % dead cells dead % Fractional effect 10

0 0 Vehicle Gemcitabine δ-T3 Gem + δ-T3 Vehicle Gemcitabine δ-T3 Gem + δ-T3 (20 μmol/L) (50 μmol/L) (20 μmol/L) (50 μmol/L) D 200 0.14 Caspase-3 activity **, *** MiaPaCa-2 cell death ELISA *P < 0.05 vs. vehicle ** 0.12 **P < 0.01 vs. vehicle 150 ***P < 0.001 vs. gem 0.10 *

0.08 100 0.06 Vehicle Gem (20 µmol/L) δ-T3 (50 µmol/L) Gem + δ-T3 0.04 50

0.02 Caspase-3 activity (%) 1,000 10-fold increase in apoptosis 0.00 0 Soft Agar Colony Formation Vehicle Gemcitabine δ-T3 Gem + δ-T3 Vehicle Gemcitabine δ-T3 Gem + δ-T3 (20 μmol/L) (50 μmol/L) (20 μmol/L) (50 μmol/L) 800 *P < 0.01 vs. vehicle **P < 0.001 vs. vehicle δ δ δ δ #P < 0.05 vs. gem V Gem -T3 Gem + -T3 V Gem -T3 Gem + -T3 600 ##P < 0.001 vs. gem or δ-T3 PARP1 PARP1 400 cleavage cleavage *, # Bax Number of colonies 200 Bax ** **,## β-Actin β 0 -Actin Vehicle Gemcitabine δ-T3 Gem + δ-T3 AsPc-1 MiaPaCa-2

Figure 4. A, gemcitabine (Gem) and d-tocotrienol signiﬁcantly (, P < 0.001) decreased proliferation of AsPc-1 and MiaPaCa-2 cells compared with vehicle, with the combination being more effective in inhibiting cell proliferation than either drug alone (#, P < 0.01; a, P < 0.01). B, effect of gemcitabine þ d-tocotrienol (1:1) and gemcitabine þ d-tocotrienol (1:2) as shown in isobologram with CI in MiaPaCa-2 cells. Points, means; bars, SE (n ¼ 3–5). C, effect of gemcitabine (gem) and d-tocotrienol alone and in combination on pancreatic cancer survival (trypan blue) in AsPc-1 and MiaPaCa-2 cells. d-Tocotrienol induced greater cell death than vehicle control (a, P < 0.001) or gemcitabine alone (b, P < 0.02). However, the combination resulted in greater effect on cell death than vehicle alone (a, P < 0.001) or gemcitabine or d-tocotrienol alone (c, P < 0.02; d, P < 0.01). Bars (means) SE (n ¼ 3–5). Apoptosis-inducing activity was 3-, 1-, and 6-fold for d-tocotrienol, gemcitabine, and the combination, respectively, compared with vehicle. Bars, means (n ¼ 3). Caspase-3 activity in MiaPaCa-2 cells treated with d-tocotrienol was signiﬁcantly increased (, P < 0.01) versus gemcitabine (, P < 0.05) and vehicle. However, the combination resulted in greater caspase-3 activity than vehicle (, P < 0.01) or with gemcitabine alone (, P < 0.001). Bars (means) SE (n ¼ 3). Western blot analyses show PARP-1 cleavage and proapoptotic Bax protein expression in MiaPaCa-2 and AsPc-1 cells (n ¼ 3).D,effectsofgemcitabineandd-tocotrienol alone and in combination on pancreatic cancer malignant transformation in MiaPaCa-2 cells. Gemcitabine is more effective in inhibiting colony formation (, P < 0.001) than d-tocotrienol (, P < 0.01; #, P < 0.02). However, the combination resulted in almost complete inhibition of the malignant transformation compared with vehicle (, P < 0.001) or with gemcitabine or d-tocotrienol alone (##, P < 0.001). Bars (means) SE (n ¼ 3). addition, we assessed the antiapoptotic molecules Bcl- d-Tocotrienol enhances the in vivo therapeutic XL, survivin, and XIAP. Relative to control and gemci- effects of gemcitabine in a pancreatic tumor model in tabine alone, Bcl-XL, survivin, and XIAP expression SCID nude mice levels were downregulated in cells exposed to d-toco- On the basis of our in vitro results, we designed trienol alone and in cells exposed to gemcitabine þ studies to determine the effects of d-tocotrienol and d-tocotrienol. These results strongly suggest that the gemcitabine in a human model of pancreatic tumor augmentation of gemcitabine activity in pancreatic can- in SCID nude mice (Fig. 6A). The dose of d-tocotrienol cer cells by d-tocotrienol is, in part, due to the inacti- was determined from our previous studies of d-toco- vation of NF-kB and its downstream genes by trienol pharmacokinetics in mice (17); our in vitro d-tocotrienol. data indicated that a 2:1 ratio of d-tocotrienol þ

www.aacrjournals.org Mol Cancer Ther; 10(12) December 2011 2369

Husain et al.

those shown in mice treated with gemcitabine or d-toco- in vitro A 120 trienol alone. The results are in accordance with the NF-κB/p65 activity (nuclear) results. P 100 * < 0.01 vs. Vehicle **P < 0.001 vs. vehicle We next examined the cleavage of PARP-1 protein, a ***P < 0.01 vs. Gem 80 * marker of apoptosis, in the tumor tissues of mice as well as the expression of the proapoptotic protein Bax. As shown 60 **, *** in Fig. 6C, d-tocotrienol, gemcitabine, and d-tocotrienol þ 40 gemcitabine induced PARP-1 cleavage in mouse tumors. κ B/P65 activity (%) Bax expression was signiﬁcantly induced in the d-toco- NF- 20 trienol alone and the gemcitabine þ d-tocotrienol tumors 0 compared with vehicle and gemcitabine alone. Vehicle Gemcitabine δ-T3 Gem + δ-T3 (20 μmol/L) (50 μmol/L) d þ B Effect of -tocotrienol gemcitabine treatment on Vehicle Gem δ-T3 Gem + δ-T3 NF-kB activation and NF-kB–regulated gene NF-κB/p65 products We further investigated whether the effects of combin-

κ AsPc-1 cells NF- B/p50 ing d-tocotrienol and gemcitabine on tumor growth in mice were associated with inhibition of NF-kB activation. Our Western blot analysis results clearly show that NF-kB/p65 and NF-kB/p50 proteins and the NF-kB tran- L scriptional targets Bcl-XL, survivin, and XIAP were mod- erately downregulated by d-tocotrienol alone. Similar to our in vitro studies, gemcitabine alone did not affect the Vehicle Gem δ-T3 Gem + δ-T3 expression of NF-kB or NF-kB–regulated gene products. in vitro NF-κB/p65 However, similar to our studies, constitutively

MiaPaCa-2 cells active NF-kB was almost completely suppressed in tumor NF-κB/p50 samples of mice treated with d-tocotrienol þ gemcitabine (Fig. 6D). Tumors also revealed downregulation of important NF-kB–regulated proapoptotic proteins (Bcl-XL, survivin, and XIAP). L Discussion

Our results clearly show that the 4 natural tocotrienols Figure 5. A, effect of gemcitabine (Gem) and d-tocotrienol alone and in have different abilities to inhibit cancer growth and sur- combination on NF-kB/p65 DNA binding activity in nuclear compartment vival. We show that d-andg-tocotrienol consistently inhib- of MiaPaCa-2 cells. Gemcitabine was refractory to p65 binding to DNA, ited cancer growth and survival in vitro and in vivo.In whereas d-tocotrienol significantly decreased p65 binding to DNA (, P < 0.01) compared with vehicle. However, the combination resulted in contrast, a-andb-tocotrienol and a-tocopherol do not greater inhibition of p65 binding to DNA than vehicle (, P < 0.001) or inhibit growth and survival of pancreatic cancer cells. Our gemcitabine alone (, P < 0.01). Bars (means) SE (n ¼ 3). B, expression data agreewith theresults reportedbyHussein andMo (21) – n ¼ of NF-kB subunit and NF-kB regulated antiapoptotic proteins ( 3). whoshowedthatd-tocotrienolsuppressedtheproliferation of Panc-1, MiaPaCa-2, and BxPc-3 human pancreatic cancer gemcitabine was synergistic in inhibiting pancreatic cells. Our results are also in agreement with Kunnumak- cancer growth. kara and colleagues (19) who reported that g-tocotrienol AsPc-1 cells were implanted in the subcutaneous tissue inhibited the growth and survival of Panc-1, MiaPaCa-2, of flanks of SCID nude mice. Mice were randomly and BxPc-3 human pancreatic cancer cells. Our data pro- assigned to the 4 treatment groups (see Materials and vide the first report of a direct comparison of the effects of Methods), with treatment continued per experimental all 4 of the vitamin E tocotrienol compounds on human protocol for 23 days. Animals were euthanized on the pancreatic cancer cells in vitro and in vivo. The results last day of the therapy. To determine the effects of treat- indicate that d-tocotrienol is the most bioactive tocotrienol ment on tumor development, we assessed tumor volume against human pancreatic cancer cells and provide the on days 0 to 23 of treatment. As shown in Fig. 6B, on day rationale for selecting d-tocotrienol as the lead tocotrienol 12, tumor volumes in the d-tocotrienol alone and d-toco- compound for further studies of the use of tocotrienols for trienol þ gemcitabine groups were significantly lower pancreatic cancer prevention and treatment. than tumor volumes in vehicle and gemcitabine alone Our in vitro and in vivo results show that the bioactivity treatment groups (P < 0.01 and P < 0.001). From treatment of tocotrienols in pancreatic cancer cells was directly day 17, tumor volumes in mice treated with d-tocotrienol related to inhibition of NF-kB activity. Our results agree þ gemcitabine were significantly reduced compared with with Kunnumakkara and colleagues (19) who reported

2370 Mol Cancer Ther; 10(12) December 2011 Molecular Cancer Therapeutics

Tocotrienol Activity in Pancreatic Cancer

AB 800 Vehicle 700 δ Gemcitabine ) -T3 AsPc-1 3 Gemcitabine (100 mg/kg, i.p. x twice/wk) 600 Sacrifice δ-T3 + gemcitabine cells ** 500 ** ** 400 * ** * ***c **b 300 * ** e d –7 d 0 d7 d14 d 21 d 23 a d ** 200 b Tumor volume (mm Tumor δ-T3 (200 mg/kg, p.o. twice/d) 100 0 50 10 15 20 25 Days postfeeding C D Vehicle δ-T3 Gem Gem + δ-T3

NF-κB/p65 δ δ Vehicle -T3 Gem Gem + -T3 NF-κB/p50 PARP1 cleavage Survivin

Bax XIAP

β-Actin Bcl-XL β-Actin Tumor tissue Tumor tissue

Figure 6. A, effect of gemcitabine (Gem) and d-tocotrienol alone and in combination on in vivo AsPc-1 tumor growth in mice. B, gemcitabine and d-tocotrienol alone significantly inhibited growth of AsPc-1 tumor (tumor volume) in vivo compared with vehicle (, P < 0.05; , P < 0.01; , P < 0.001). However, the combination resulted in greater inhibition of tumor growth than shown with vehicle (, P < 0.05; , P < 0.01; , P < 0.001), gemcitabine alone (a, P < 0.05; b, P < 0.02; c, P < 0.01), and d-tocotrienol alone (c, P < 0.01; d, P < 0.05; e, P < 0.01). Points (means); bars SE (n ¼ 8–10). C, apoptotic protein expression in AsPc-1 mouse tumor tissue by Western blot analysis. Both gemcitabine and d-tocotrienol increased PARP-1 cleavage and Bax protein expression. However, the combination resulted in greater increase in PARP-1 cleavage and Bax protein expression than shown with vehicle (n ¼ 4–5). D, expression of apoptotic protein, NF-kB subunits, and antiapoptotic protein in AsPc-1 mouse tumor tissues by Western blot analysis (n ¼ 4–5). p.o., per os. that g-tocotrienol inhibited NF-kB activity in vitro and in Several mechanisms by which tocotrienols exert their vivo. We show for the first time that d-tocotrienol also anticancer effects have been reported. Suppression of inhibited NF-kB activity and the expression of NF-kB– inflammatory transcription factor NF-kB, which is regulated gene products. In contrast, a- and b-tocotrienol involved in tumorigenesis, and inhibition of 3-hydroxy- had no significant effect on NF-kB activity. These results 3-methyl-glutaryl CoA (HMG-CoA) reductase, mamma- strongly suggest that the bioactivity of tocotrienols against lian DNA polymerase, and certain protein kinases are cancer cells is due, in part, to inhibition of the activity of potential targets to tocotrienols (16). In this study, we the transcription factor NF-kB. Because numerous studies found that the in vitro and in vivo anti–pancreatic cancer have shown that inhibition of NF-kB can enhance gemci- activity of tocotrienols can be attributed to inhibition of tabine activity in pancreatic cancer cells (3, 8, 9, 19, 22), we NF-kB signaling. Additional studies are being designed to investigated whether d-tocotrienol could enhance the clarify the mechanisms of inhibition of NF-kB signaling by activity of gemcitabine against human pancreatic cancer. d-tocotrienol in pancreatic cancer cells. One possibility is We found that d-tocotrienol augmented gemcitabine inhi- that d-tocotrienol inhibits the degradation of the protein bition of cell viability, as well as induction of apoptosis that inhibits NF-kB, the inhibitor of NF-kB(IkB). IkB in pancreatic cancer cells. We also found that d-tocotrienol sequesters NF-kB in the cytoplasm, and phosphorylation augmented gemcitabine inhibition of growth of human of IkB makes it a target for ubiquitination. The added pancreatic AsPc-1 tumors in SCID nude mice. Augmen- ubiquitin molecules render IkB for degradation by pro- tation of gemcitabine activity by d-tocotrienol is associat- teasomes. We observed that d-tocotrienol treatment ed with inhibition of constitutively active NF-kB and resulted in a decrease of phosphorylated IkB. Whether NF-kB–regulated gene products. Our results are in agree- d-tocotrienol affects the kinase that phosphorylates IkBis ment with Kunnumakkara and colleagues (19) who unknown. showed that g-tocotrienol inhibition of constitutively In summary, this is the first report evaluating the effects active NF-kB and NF-kB–regulated gene products was of all of the natural vitamin E tocotrienol compounds on associated with g-tocotrienol potentiation of gemcitabine- pancreatic cancer cells. Our results show that d- and induced apoptosis, as well as g-tocotrienol potentiation of g-tocotrienol inhibited NF-kB activity, cell growth, cell the effects of gemcitabine against human pancreatic survival, and tumor growth in nude mice. We further tumors in nude mice. show that d-tocotrienol augmented gemcitabine activity

www.aacrjournals.org Mol Cancer Ther; 10(12) December 2011 2371

Husain et al.

in vitro and in vivo. These results suggest that inhibition of Acknowledgments NF-kB signaling by d-tocotrienol may be an effective approach for the prevention and treatment of pancreatic The authors thank Rasa Hamilton (Moffitt Cancer Center) for editorial assistance. cancer. Our findings suggest evaluation of NF-kB signaling compounds as an endpoint biomarker in our ongoing Grant Support phase I trial of d-tocotrienol in patients with pancreatic tumors (ClinicalTrials.gov. Identifier: NCT00985777). The study was supported, in part, by National Cancer Institute/USPHS grant 1RO1 CA-129227-01A1 and DAVOS-69-15099-99-01. Disclosure of Potential Conflicts of Interest The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate M.P. Malafa and S.M. Sebti are named as inventors on US Patent "Delta- this fact. Tocotrienol Treatment and Prevention of Pancreatic Cancer" (June 26, 2007; OTML docket number 06A069) but do not have financial interest in the companies that have licensed this patent. No potential conflicts of Received June 8, 2011; revised August 24, 2011; accepted September 18, interest were disclosed by other authors. 2011; published OnlineFirst October 4, 2011.

References 1. Bilimoria KY, Bentrem DJ, Ko CY, Ritchey J, Stewart AK, Winchester tabine in an orthotopic model of pancreatic cancer through sup- DP, et al. Validation of the 6th edition AJCC Pancreatic Cancer Staging pression of proliferation, angiogenesis, and inhibition of nuclear System: report from the National Cancer Database. Cancer 2007;110: factor-kappaB-regulated gene products. Cancer Res 2007;67: 738–44. 3853–61. 2. Burris HA III, Moore MJ, Andersen J, Green MR, Rothenberg ML, 12. Mahon PC, Baril P, Bhakta V, Chelala C, Caulee K, Harada T, et al. Modiano MR, et al. Improvements in survival and clinical benefit with S100A4 contributes to the suppression of BNIP3 expression, che- gemcitabine as first-line therapy for patients with advanced pancreas moresistance, and inhibition of apoptosis in pancreatic cancer. Cancer cancer: a randomized trial. J Clin Oncol 1997;15:2403–13. Res 2007;67:6786–95. 3. Arlt A, Gehrz A, Muerkoster S, Vorndamm J, Kruse ML, Folsch UR, et al. 13. Kurbitz C, Heise D, Redmer T, Goumas F, Arlt A, Lemke J, et al. Role of NF-kappaB and Akt/PI3K in the resistance of pancreatic Epicatechin gallate and catechin gallate are superior to epigallocate- carcinoma cell lines against gemcitabine-induced cell death. Onco- chin gallate in growth suppression and anti-inflammatory activities in gene 2003;22:3243–51. pancreatic tumor cells. Cancer Sci 2011;102:728–34. 4. Bai J, Sui J, Demirjian A, Vollmer CM Jr, Marasco W, Callery MP. 14. Murtaza I, Adhami VM, Hafeez BB, Saleem M, Mukhtar H. Fisetin, a Predominant Bcl-XL knockdown disables antiapoptotic mechanisms: natural flavonoid, targets chemoresistant human pancreatic cancer tumor necrosis factor-related apoptosis-inducing ligand-based triple AsPC-1 cells through DR3-mediated inhibition of NF-kappaB. Int J chemotherapy overcomes chemoresistance in pancreatic cancer cells Cancer 2009;125:2465–73. in vitro. Cancer Res 2005;65:2344–52. 15. Miyazawa T, Shibata A, Sookwong P, Kawakami Y, Eitsuka T, Asai A, 5. Liu WS, Yan HJ, Qin RY, Tian R, Wang M, Jiang JX, et al. siRNA directed et al. Antiangiogenic and anticancer potential of unsaturated vitamin E against survivin enhances pancreatic cancer cell gemcitabine chemo- (tocotrienol). J Nutr Biochem 2009;20:79–86. sensitivity. Dig Dis Sci 2009;54:89–96. 16. Aggarwal BB, Sundaram C, Prasad S, Kannappan R. Tocotrienols, the 6. Shrikhande SV, Kleeff J, Kayed H, Keleg S, Reiser C, Giese T, et al. vitamin E of the 21st century: its potential against cancer and other Silencing of X-linked inhibitor of apoptosis (XIAP) decreases gemci- chronic diseases. Biochem Pharmacol 2010;80:1613–31. tabine resistance of pancreatic cancer cells. Anticancer Res 2006;26: 17. Husain K, Francois RA, Hutchinson SZ, Neuger AM, Lush R, Coppola 3265–73. D, et al. Vitamin E delta-tocotrienol levels in tumor and pancreatic 7. Wang CY, Guttridge DC, Mayo MW, Baldwin AS Jr. NF-kappaB induces tissue of mice after oral administration. Pharmacology 2009;83: expression of the Bcl-2 homologue A1/Bfl-1 to preferentially suppress 157–63. chemotherapy-induced apoptosis. Mol Cell Biol 1999;19:5923–9. 18. Tallarida RJ. Drug synergism: its detection and applications. J Phar- 8. Kong R, Sun B, Wang SJ, Pan SH, Wang G, Chen H, et al. [An macol Exp Ther 2001;298:865–72. experimental study of gemcitabine inducing pancreatic cancer cell 19. Kunnumakkara AB, Sung B, Ravindran J, Diagaradjane P, Deorukhkar apoptosis potentiated by nuclear factor-kappa B P65 siRNA]. Zhon- A, Dey S, et al. Gamma-tocotrienol inhibits pancreatic tumors and ghua Wai Ke Za Zhi 2010;48:128–33. sensitizes them to gemcitabine treatment by modulating the inflam- 9. Pan X, Arumugam T, Yamamoto T, Levin PA, Ramachandran V, Ji B, matory microenvironment. Cancer Res 2010;70:8695–705. et al. Nuclear factor-kappaB p65/relA silencing induces apoptosis and 20. Fulda S, Debatin KM. Extrinsic versus intrinsic apoptosis pathways in increases gemcitabine effectiveness in a subset of pancreatic cancer anticancer chemotherapy. Oncogene 2006;25:4798–811. cells. Clin Cancer Res 2008;14:8143–51. 21. Hussein D, Mo H. d-delta-Tocotrienol-mediated suppression of the 10. Harikumar KB, Kunnumakkara AB, Sethi G, Diagaradjane P, Anand P, proliferation of human PANC-1, MIA PaCa-2, and BxPC-3 pancreatic Pandey MK, et al. Resveratrol, a multitargeted agent, can enhance carcinoma cells. Pancreas 2009;38:e124–36. antitumor activity of gemcitabine in vitro and in orthotopic mouse 22. Kong R, Sun B, Jiang H, Pan S, Chen H, Wang S, et al. Downregulation model of human pancreatic cancer. Int J Cancer 2010;127:257–68. of nuclear factor-kappaB p65 subunit by small interfering RNA syner- 11. Kunnumakkara AB, Guha S, Krishnan S, Diagaradjane P, Gelovani gizes with gemcitabine to inhibit the growth of pancreatic cancer. J, Aggarwal BB. Curcumin potentiates antitumor activity of gemci- Cancer Lett 2010;291:90–8.

2372 Mol Cancer Ther; 10(12) December 2011 Molecular Cancer Therapeutics

Vitamin E δ-Tocotrienol Augments the Antitumor Activity of Gemcitabine and Suppresses Constitutive NF- κB Activation in Pancreatic Cancer

Kazim Husain, Rony A. Francois, Teruo Yamauchi, et al.

Mol Cancer Ther 2011;10:2363-2372. Published OnlineFirst October 4, 2011.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-11-0424

Cited articles This article cites 22 articles, 9 of which you can access for free at: http://mct.aacrjournals.org/content/10/12/2363.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/10/12/2363.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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/10/12/2363. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.