the vitamin e analog a-tea stimulates tumor - cancer research

Microenvironment and Immunology

The Vitamin E Analog a-TEA Stimulates Tumor Autophagyand Enhances Antigen Cross-Presentation

Yuhuan Li1, Tobias Hahn1, Kendra Garrison1, Zhi-Hua Cui1, Andrew Thorburn2, Jacqueline Thorburn2,Hong-Ming Hu1, and Emmanuel T. Akporiaye1

AbstractThe semisynthetic vitamin E derivative alpha-tocopheryloxyacetic acid (a-TEA) induces tumor cell apoptosis

andmay offer a simple adjuvant supplement for cancer therapy if its mechanisms can be better understood. Herewe report that a-TEA also triggers tumor cell autophagy and that it improves cross-presentation of tumorantigens to the immune system. a-TEA stimulated both apoptosis and autophagy in murine mammary and lungcancer cells and inhibition of caspase-dependent apoptosis enhanced a-TEA–induced autophagy. Cell exposureto a-TEA generated double-membrane–bound vesicles indicative of autophagosomes, which efficiently cross-primed antigen-specific CD8þ T cells. Notably, vaccination with dendritic cells pulsed with a-TEA–generatedautophagosomes reduced lungmetastases and increased the survival of tumor-bearingmice. Taken together, ourfindings suggest that both autophagy and apoptosis signaling programs are activated during a-TEA–inducedtumor cell killing. We suggest that the ability of a-TEA to stimulate autophagy and enhance cross-priming ofCD8þ T cells might be exploited as an adjuvant strategy to improve stimulation of antitumor immune responses.Cancer Res; 72(14); 1–11. �2012 AACR.

IntroductionAlpha-tocopheryloxyacetic acid (a-TEA) is a stable semi-

synthetic ether derivative of naturally occurring vitamin E(a-tocopherol). Although vitamin E has been pursued as ananticancer agent because of its antioxidant properties, theavailable epidemiologic data as well as in vivo studies inexperimental animal tumor models have showed a limitedrole for vitamin E in cancer prevention or control (1–3) andmay even be harmful as shown in the aborted SELECT clinicaltrial that examined the effect of high-dose vitamin E onprostate cancer (4). a-TEA is derived from vitamin E by achemical modification, which involves the replacement of thehydroxyl group at the number 6 carbon of the phenolic ring ofthe chroman head by an acetic acid residue linked by an etherbond (Supplementary Fig. S1; ref. 5). This modification rendersa-TEA, in contrast to vitamin E, redox silent but active against

tumors of various origins (5–7). The presence of a noncleavableether bond ensures the stability of a-TEA, allowing it to bedelivered via the oral route in a biologically active form. Wereported for the first time that when incorporated into mousechow, and supplied to mice in the diet, a-TEA significantlyinhibited the growth of transplanted and spontaneously-aris-ing metastatic breast cancers and dramatically reduced theincidence of spontaneous lung metastases before and afterprimary tumor establishment without overt toxicity (7, 8).

Reports fromnumerous laboratories including our ownhaveshowed that apoptosis is a primary mode of a-TEA–inducedtumor cell death (7–10), a process that is initiated by mito-chondrial depolarization followed by release of cytochrome cto the cytosol and activation of the caspase execution pathway(reviewed in ref. 11). However, the observation that the anti-tumor activity of a-TEA cannot be completely blocked usingpan or caspase-specific inhibitors (12, 13) suggests the involve-ment of additional pathway(s) in a-TEA–mediated tumor cellkilling.

Autophagy is normally a protective survival mechanismused by cells undergoing various forms of stress, includingchemotherapy, to sequester, process, and recycle damagedcellular organelles and misfolded and long-lived proteins toprovide nutrients to the cell (reviewed in ref. 14). It has recentlybecome clear that apoptosis and autophagy are not mutuallyexclusive events (reviewed in ref. 15) and that both could leadto cell death (14, 16). The formation of autophagosomesinvolves 3 major steps: The first (initiation stage) is thede novo formation of an isolation membrane, which is regu-lated by the mTOR and the Beclin-1 (Atg6)/class III phospo-hoinsitol-3 kinase (PI3K) complex. The second stage involveselongation and expansion of the phagophore to enclose

Authors' Affiliations: 1Robert W. Franz Cancer Research Center, Earle A.Chiles Research Institute, Providence Portland Medical Center, Portland,Oregon; and 2Department of Pharmacology, University of Colorado Den-ver, Aurora, Colorado

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Y. Li and T. Hahn contributed equally to this work.

Corresponding Authors: Emmanuel T. Akporiaye, Robert W. Franz Can-cer Research Center, Research Institute, Providence Portland MedicalCenter, 4805 NE Glisan St., Portland, OR 97213. Phone: 503-215-4595;Fax: 503-215-6841; E-mail: [email protected]; andHong-Ming Hu. Phone: 503-215-6531; Fax: 503-215-6841; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-11-3103

�2012 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org OF1

Research. on December 24, 2018. © 2012 American Association for Cancercancerres.aacrjournals.org Downloaded from

Published OnlineFirst June 28, 2012; DOI: 10.1158/0008-5472.CAN-11-3103

http://cancerres.aacrjournals.org/

cytosolic components including damaged organelles and mis-folded proteins and requires the conjugation of Atg5 to Atg12.The final stage is the formation of a mature autophagosomeready for fusion with lysosomal vesicles, which requires con-version of soluble LC3-I (Atg8) to the membrane-bound formLC3-II (17).

During autophagy, cellular components including viral orendogenous tumor-associated antigens (TAA) become avail-able for cross-presentation by professional antigen-presentingcells (APC) to prime antigen- or tumor-specific T-cellresponses (14, 18). Although autophagy is known to play anessential role inmajor histocompatibility complex (MHC) classII–restricted antigen presentation (19), only recently has itsrole in MHC class I–restricted stimulation of CD8þ T cells(cross-presentation) become appreciated (18, 20).

In this study, we investigated whether a-TEA stimulatestumor cell autophagy and enhances antigen cross-presenta-tion by dendritic cells (DC). We show that a-TEA inducestumor cell autophagy and that the a-TEA–derived autophago-some-enriched supernatant fraction (a-TAGS) stimulates effi-cient antigen cross-presentation. We describe here a novelmechanism of immune activation by a-TEA that involves thestimulation of tumor cell autophagy and enhanced cross-priming of CD8þ T cells.

Materials and MethodsPreparation of a-TEA

a-TEA [(2,5,7,8-tetramethyl-(2R-(4R,8R,12-trimethyltridecyl)chroman-6-yloxy) acetic acid)] was synthesized using a combi-nation of previously described methods (2, 5) and vesiculateda-TEA (Va-TEA) was generated as previously described (8).

MiceSix- to 8-week-old female BALB/c and C57BL/6 mice were

purchased from Harlan Laboratories. OT-I TCR transgenicbreeders were purchased from the Jackson Laboratory. CT-TCR transgenic breeders were kindly provided by Dr. Jill E.Slansky (University of Colorado Denver School of Medicine,Denver, CO; ref. 21). All mice were maintained and used inaccordance with the Principles of Animal Care (NIH publica-tion No. 85–23) and all studies were approved by the EACRIInstitutional Animal Care and Use Committee.

Tumor cell lines and cell cultureThe poorly immunogenic and highly metastatic 4T1 tumor

cell line was kindly provided by Dr. FredMiller of the MichiganCancer Foundation (Detroit, MI). The Lewis Lung carcinoma(3LL) cell line was kindly provided by Dr. Lea Eisenbach ofthe Weizmann Institute of Science (Rehovot, Israel).

DNA construction and transduction of cell lines3LL or 4T1 tumor cells were transduced with lentiviral

vectors that either expressed the ubiquitin-methionine greenfluorescence protein-ovalbumin (Ub-M-GFP-OVA) antigenor LC3-GFP as previously described (18). Ub-M-GFP-OVAantigen- or LC3-GFP-expressing cells were enriched for GFPpositivity by fluorescence-activated cell sorting. LC3-GFP wastransiently expressed in 3LL tumor cells after transfection

using Lipofectamine-LTX (Invitrogen) following the manufac-turer's instructions.

4T1 tumor cells stably transduced with Atg12-specific shorthairpin RNA (shRNA; 4T1-Atg12shRNA) and scrambled (con-trol) shRNA (4T1-CTRLshRNA) were generated previously.Atg12-specific shRNA expression is doxycycline-inducible andcoexpressed with red fluorescent protein, which was moni-tored by epifluorescence. Atg12 gene knockdown was deter-mined byWestern immunoblotting. The transduced cells (4T1-LC3-GFP, 4T1-OVA-GFP, 4T1-Atg12shRNA, 4T1-CTRLshRNA,and 3LL-OVA-GFP) were maintained in culture medium con-taining 2 mg/mL puromycin (Sigma-Aldrich).

Quantification of LC3-GFP punctae4T1-LC3-GFP or 3LL-LC3-GFP cells were cultured overnight

on Lab-Tek II chamber slides (Thermo Fisher Scientific). Thecells were then treated with Va-TEA or 100 mmol/L trehalose(Sigma-Aldrich) with the addition of 10 nmol/L bafilomycin-A1(Sigma-Aldrich) for the last 2 hours of incubation. Cells werefixed and the fraction of punctate-positive cells perfield of viewwas determined by counting punctate-positive and total num-ber of cells in 12 to 30 random high-powered fields (�100) ofview per treatment by epifluorescence microscopy.

Apoptosis assayTumor cells were pretreated with 50 mmol/L zVAD-fmk

(Sigma-Aldrich) for 2 hours before the addition of 40 mmol/LVa-TEA. Nonadherent and adherent cells were collectedusing 10 mmol/L EDTA, pooled and stained using theAPC-Annexin-V/7AAD Apoptosis Detection Kit (BD Phar-mingen) according to the manufacturer's instructions. Cellspositively staining with APC-Annexin-V were considered tobe apoptotic.

Clonogenic cell survival assay4T1-Atg12shRNA or 4T1-CTRLshRNA tumor cells (1.25� 105)

were cultured with 2 mg/mL doxycycline (Sigma-Aldrich) for72 hours to induce Atg12 knockdown and were then treatedwith 60 mmol/L Va-TEA. After 24 hours, nonadherent andadherent cells were collected (300� g, 5 minutes). Cell numberand viability were determined by trypan blue dye exclusionand 1.25 � 102, 2.5 � 102, 5 � 102, 103, and 104 viable cells(trypan blue negative) from each treatment group were platedin triplicate (100 mm dishes) and incubated (7% CO2, 37�C)for 7 days. The resulting colonies were methanol-fixed, stainedwith Giemsa (Sigma-Aldrich), and counted. The surviving cellfraction was determined as previously described (7).

Generation and collection of a-TAGSTumor cells were treatedwith 20mmol/L (4T1) or 80mmol/L

(3LL) Va-TEA for 24 hours. The culture supernatant wascollected and cleared of dead cells and cell debris (300 � g,10 minutes). To obtain the autophagosome enriched fraction(a-TAGS), the supernatant containing the crude large vesicleswas centrifuged at 10,000 � g for 15 minutes. The a-TAGSpellet was resuspended in PBS and protein content wasdetermined by bicinchoninic acid (BCA) protein assay(Thermo Scientific). For the control cells [vehicle (PBS)-

Li et al.

Cancer Res; 72(14) July 15, 2012 Cancer ResearchOF2




treated] in which cells were not dying, culture medium wascollected from twice as many cells. To block autophagy, cellswere pretreated with 10 mmol/L 3-methyladenine (3-MA;Sigma-Aldrich) for 16 hours before a-TEA treatment.

Transmission electron microscopya-TAGS was prepared from 4T1 or 3LL tumor cells treated

with 20 or 80 mmol/L Va-TEA respectively, for 24 hours andprepared for transmission electron microscopy (TEM) asdescribed in the Supplementary Methods. Microscopy wasconducted at 60 kV on a Philips Morgagne TEM (FEI Inc.),equipped with a charge-coupled device camera and imageswere collected at original magnifications of�1,000 to�37,000.

Western blot analysisTo detect LC3 conversion, 3LL or 4T1 tumor cells were

treatedwith 40 or 60mmol/LVa-TEAor 400 nmol/L rapamycin(Sigma-Aldrich) for 16 hours with the addition of bafilomycin-A1 (Sigma-Aldrich) for the last 2 hours of culture. Cell lysateswere prepared using Complete Lysis-M Buffer (Roche AppliedSciences) containing protease inhibitors. For in vivo LC3conversion, 4T1 tumor–bearing mice received a-TEA in thediet (�6mg/mouse/day) for 5 days. Subsequently, tumorswereresected and lysed using CelLytic Mammalian Tissue LysisReagent (Sigma-Aldrich) containing protease inhibitors and arotor-stator homogenizer. To detect Beclin-1 and cleavage ofnuclear PARP, tumor cells were preincubated with zVAD-fmk(80 mmol/L) for 2 hours before treatment with a-TEA (80mmol/L) for 16 hours. To detect Atg12, 4T1-Atg12shRNA and4T1-CTRLshRNA were cultured with 2 mg/mL doxycycline for72 hours before a-TEA treatment. All lysates were clarified(14,000 � g, 15 minutes, 4�C) and protein content was deter-mined by BCA (Thermo Scientific). After electrophoretic sep-aration, proteins were detected by Western immunoblotting.To detect endogenous LC3 in a-TAGS, a-TAGS was lysed in

RIPA buffer and clarified. Protein content was determined byBCA (Thermo Scientific) and equal amounts (5 mg) of proteinwere electrophoretically separated and LC3 was detected byWestern immunoblotting and quantified using Image-J soft-ware (22).

Cross-presentation assayFor the in vitro cross-presentation assay, splenic DCs were

generated as previously described (23). OT-I splenocyteswere carboxyfluorescein succinimidyl ester (CFSE)-labeled(5 mmol/L) and coincubated with antigen-loaded DCs(20 mg a-TAGS/1 � 106 DC; 6 hours) in 10 mL tissue cultureflat tubes (Techno Plastic Products AG) for 4 or 6 days. CFSEdilution was measured by flow cytometry as previouslydescribed (18). To determine in vivo cross-presentation,a-TAGS (20 mg protein) was injected into both inguinal lymphnodes of naive C57BL/6 mice. On the same day, 3 � 106 CFSE-labeled Thy1.1þ OT-I T cells were adoptively transferred byintravenous injection. Lymphnodeswere collected 5 days later,and single-cell suspensions were prepared and CFSE dilutionof Thy1.1þ CD8þ cells was determined by flow cytometry. Datawere acquired and analyzed using BD CellQuest Pro software(BD Biosciences).

IFN-g ELISA4T1-Atg12shRNA and 4T1-CTRLshRNA cells were incubated

with 2 mg/mL doxycycline for 72 hours before preparation ofa-TAGS as described earlier. DCs (3� 106) were pulsed with 20mg a-TAGS for 6 hours, and washed 3 times. T cells (3 � 106)isolated fromCT-TCRmicewere then coincubatedwith theDCfor 48 hours and IFN-g secretion was detected by ELISA(eBioscience).

Generation of DCs and animal vaccination studiesFor the 4T1 tumor model experiments, bone marrow–

derived DC (24) were either left untreated (nonpulsed-DC,npDC), pulsed overnight with 40 mg a-TAGS per 1 � 106 DC(a-TAGS-DC) or tumor cell freeze–thaw (3 cycles) lysate(Lysate-DC) at a ratio of 3 tumor cell equivalents per DC.Subsequently, the DC were matured by adding 200 U/mL TNF-a (PeproTech) for 24 hours. 4T1 tumor cells (5 � 104) wereinjected subcutaneously (s.c.) into the right mammary fat padof female BALB/c mice. The mice then received s.c. vaccina-tions of 1 � 106 a-TAGS-DC, Lysate-DC, or npDC on days 7, 9,and 11 posttumor injection.Mice injected subcutaneously witha-TAGS (40 mg) were included as controls. For the 3LL experi-ments, DC were left untreated (npDC), or pulsed with either 20mg a-TAGS per 3 � 106 DC (a-TAGS-DC) or tumor cell freeze-thaw lysate (Lysate-DC) in the presence of 200 ng/mL IFN-g(PeproTech). 3LL tumor cells (2 � 105) were injected intrave-nously into female C57BL/6 mice. On day 3 posttumor injec-tion, the mice were vaccinated subcutaneously with 3 � 106

a-TAGS-DC, Lysate-DC, npDC, or 20 mg a-TAGS. Visiblemetastatic nodules in the lungs were enumerated on day 28.

Statistical analysisStatistical significance of differences among data sets was

assessed by Student t test for pair wise comparisons or forcomparisons of multiple groups by 1-way ANOVA with TukeyHSD test. Survival was assessed according to Kaplan andMeierincluding log-rank test. All analyses were done using Prismsoftware (GraphPad). Probability values (P) of �0.05 wereconsidered indicative of significant differences.

Resultsa-TEA stimulates autophagy in tumor cells

To assess whether a-TEA stimulates autophagy in tumorcells, 3LL andmurinemammary tumor (4T1) cells were treatedwitha-TEA and the levels of LC3-I and LC3-II were determinedby Western immunoblotting. Induction of autophagy is asso-ciated with conversion of the cytosolic LC3-I to themembrane-bound LC3-II (17). The data show that a-TEA induced a 7-foldincrease in LC3-II conversion in a-TEA–treated 3LL tumorcells (Fig. 1A). LC3-II conversion in a-TEA–treated 4T1 tumorcells was less robust andwas�3-fold higher than vehicle (PBS)-treated cells (Fig. 1C). Aswitha-TEA, 4T1 tumor cells were alsoless responsive than 3LL tumor cells to rapamycin, a knownautophagy inducer. To confirm autophagy induction ina-TEA–treated tumor cells, we assessed the formation ofLC3-GFP punctae in LC3-GFP transgene-expressing 3LL and4T1 cells. a-TEA treatment caused a dose-dependent increase

a-TEA–Induced Tumor Cell Autophagy and Cross-Presentation

www.aacrjournals.org Cancer Res; 72(14) July 15, 2012 OF3




Untreated Trehalose

P < 0.05

P < 0.05

P < 0.05P < 0.05

P < 0.05

P < 0.049

--

α-TEA

α-

α-

Untreated

α-TEA (4

0 μm

ol/L)

α-TEA (6

0 μm

ol/L)

60 μm

ol/L

30 μm

ol/L12

0 μm

ol/L

60 μm

ol/L

Trehalose

α-TEA

α

l

fc

A B

C D

E

Figure 1. a-TEA induces tumor cell autophagy. A, 3LL tumor cellswere treatedwitha-TEA for 16 hours. LC3-I to LC3-II conversionwas determined byWesternimmunoblotting. Actin was used as loading control. Bar graphs indicate fold change of LC3-II (normalized to actin) over untreated cells. B, accumulation ofLC3-II-GFP punctae. 3LL cells transiently expressing LC3-GFP fusion protein (3LL-LC3-GFP) were treated with a-TEA or trehalose for 16 hourswith the addition of bafilomycin-A1 for the last 2 hours of treatment. Formation of punctaewas determinedby epifluorescencemicroscopy, and the proportionof punctate-positive cells per field of view was determined. Representative pictures (magnification, �100) of untreated, trehalose-, and a-TEA (30 mmol/L)-treated cells. C, increased conversion of LC3-I to LC3-II in 4T1 tumor cells after a-TEA treatment was determined as in A. D, formation of LC3-II-GFPpunctae was determined as in B in 4T1 mammary tumor cells stably expressing a LC3-GFP fusion protein (4T1-LC3-GFP). Representative pictures(magnification, �100) of untreated, trehalose-, and a-TEA (60 mmol/L)-treated cells. Data are from 3 independent experiments. E, 4T1 mammary tumor–bearingmice (n¼ 3 per group) received a-TEA in the diet for 5 days. Subsequently, tumors were resected, and LC3-I to LC3-II conversion was determined byWestern immunoblotting. 4T1 tumor cells treated in vitro with rapamycin were used as positive control. Glyceraldehyde-3-phosphate dehydrogenase(GAPDH) was used as loading control. The bar graph represents mean fold change � SD of LC3-II bands (normalized to GAPDH) from 3 mice per group.

Li et al.





in LC3 punctae formation indicative of autophagy (25) in both3LL (Fig. 1B) and 4T1 (Fig. 1D) tumor cells. Punctae formationin a-TEA–treated 3LL cells was similar or higher than punctaeformation induced by the known autophagy inducer trehalose(Fig. 1B). a-TEA treatment of 4T1 cells also significantlyincreased punctae formation to levels comparable to trehalose(Fig. 1D). To determine if a-TEA induces autophagy in vivo,4T1 tumor–bearing animals received an oral dose of a-TEA(6 mg/day) that we have previously shown to reduce tumorgrowth and prolong survival (2, 7). LC3 conversion in the tumorwas determined after 5 days of a-TEA treatment and we foundapproximately 3-fold increased LC3-II levels in 4T1 mammarytumors from a-TEA–treated mice compared with untreatedmice (Fig. 1E). Taken together, these 3 lines of evidenceindicate that a-TEA induced autophagy in tumor cells.

Autophagy precedes apoptosis during a-TEA treatmentOn the basis of the documented property of a-TEA as a

potent apoptosis-inducer (7, 8, 12, 13, 26), we compared thekinetics of apoptosis and autophagy induction during a-TEAtreatment. The data (Fig. 2A and B) show that a-TEA–inducedautophagy precedes apoptosis. At 4 hours of a-TEA treatmentwhenno change in the percentage of apoptotic cells is observedcompared with untreated cells (Fig. 2A), the percentageof punctate-positive cells, indicative of autophagy is �25%

(Fig. 2B); peak levels of apoptosis and autophagy are observedat 16 hours of a-TEA treatment. Our observation that a-TEAstimulates autophagy together with a recent report that treat-ment of stressed cells with the pan-caspase inhibitor zVAD-fmk enhanced autophagy (27, 28), prompted us to evaluate theeffect of apoptosis inhibition on a-TEA–induced autophagy.For this purpose, 4T1-LC3-GFP cells were treated with a-TEAin the presence of the pan-caspase inhibitor zVAD-fmk. a-TEAinduced caspase-dependent apoptosis, which reached a max-imum at 16 hours, was inhibited �55% by zVAD-fmk. Incontrast, increase in the percentage of LC3-GFP punctate-positive cells, which was detected as early as 4 hours aftera-TEA treatment (Fig. 2B) was enhanced by zVAD-fmk treat-ment. It has recently been reported that apoptosis inductionmay inhibit autophagy via regulation of Beclin-1 levels (27, 28).Therefore, we determined Beclin-1 protein levels in a-TEA–treated tumor cells. The data (Fig. 2C) show a modest reduc-tion in Beclin-1 protein in a-TEA–treated tumor cells com-pared with untreated cells (Fig. 2C). Pretreatment with zVAD-fmk seems to restore Beclin-1 expression (Fig. 2C).

a-TEA–induced autophagy decreases tumor cell survivalBecause autophagy can either be protective or lead to tumor

cell death (14, 29), we assessed the impact of a-TEA treatmenton tumor cell survival during autophagy inhibition. This was

α α

% a

po

pto

tic

cells

% p

un

ctat

e-p

osi

tive

cel

lsP = 0.0323

P = 0.0195

P = 0.0482

P = 0.0048 P = 0.0481

P = 0.0001P = 0.0001

N

A

C

B

N

2

Figure2. Effect of pan-caspase inhibitionona-TEA–induced tumor autophagy. A, 4T1-LC3-GFPcellswerepreincubatedwith thepan-caspase inhibitor zVAD-fmk for 2 hours and then treated with a-TEA for 4, 10, 16, or 24 hours in the presence of zVAD-fmk. Apoptosis was determined by flow cytometry afterstaining with 7-aminoactinomycin D (7-AAD) and Annexin-V-PE. B, 4T1-LC3-GFP cells were treated as in A, and formation of LC3-GFP punctae wasdetermined by epifluorescence microscopy. Proportion of punctate-positive cells per field of view (magnification, �100) was determined. Data arefrom 3 independent experiments. C, 4T1 tumor cells were treated with a-TEA for 10 or 16 hours in the presence of zVAD-fmk. PARP cleavage and Beclin-1levels were determined by Western immunoblotting. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as loading control. Bar graphsindicate fold change of Beclin-1 (normalized to GAPDH) over untreated cells.






achieved by doxycycline-induced shRNA knockdown of theautophagy pathway gene, Atg12 (Fig. 3A), which is essential forthe membrane extension step that leads to autophagosomalvesicle formation (14). Tumor cell survival was assessed using along-term clonogenicity assay, which provides the most accu-rate assessment of cell death. a-TEA treatment concurrentwith Atg12 knockdown significantly increased colony forma-tion of a-TEA–treated cells (P > 0.0131), suggesting thatautophagy contributed to a-TEA–mediated tumor cell death(Fig. 3B).

a-TAGS is an efficient antigen carrierfor cross-presentation

Given that we (18, 30) and others (19, 31) have previouslyshown that autophagy induction in tumor cells can play apivotal role in cross-presentation of tumor-associated antigen,we hypothesized thata-TEA–generated autophagosomes aug-ment antigen-specific CD8þ T-cell activation. We first deter-mined if a-TEA treatment of 4T1 and 3LL tumor cells resultedin release of autophagosomes. Examination of the precleared,high-speed (10,000� g) pellet fraction revealed the presence ofdouble-membrane structures (Fig. 4A) typical of autophago-somes (25). In addition, immunoblot analysis showed that theautophagosome-enriched fraction (a-TAGS) contained elevat-ed amounts of LC3-II (Fig. 4B).

To determine whether a-TAGS is an antigen carrier and canstimulate cross-presentation to CD8þ T cells, we pulsed DCswith a-TAGS collected from 3LL or 4T1 tumor cells that stablyexpress the OVA protein and tested their ability to stimulateOVA-specific transgenic OT-I CD8þ T cells (Fig. 5A). DCspulsed with a-TAGS from 3LL-OVA and 4T1-OVA cells stim-ulated vigorousOT-I CD8þT-cell proliferationwith over 60%ofT cells undergoing cell division (Fig. 5A and B). In contrast, theequivalent high-speed fraction from a-TEA–treated parental4T1 or 3LL tumor cells (data not shown) or untreated OVA-expressing (4T1-OVA, 3LL-OVA) cells did not significantlystimulate OT-I T-cell proliferation compared with unpulsed

DC (Fig. 5B). Furthermore, a-TAGS from OVA-expressingtumor cells stimulated proliferation of OT-I CD8þ T cells inan a-TEA dose-dependent manner. Peak T-cell stimulationwas achieved using 3LL- and 4T1–derived a-TAGS after treat-ment with 80 and 20 mmol/L a-TEA, respectively (data notshown). This finding suggests that a-TAGS is an efficientcarrier of tumor antigens for cross-presentation.

Tumor cell autophagy plays an important role ina-TAGSactivation of CD8þ T cells

We next confirmed the role of autophagy in cross-presen-tation of a-TAGS by blocking tumor cell autophagy using 3-MA, which inhibits class III phosphoinositide 3-kinase (PI3K)required for initiation of autophagy. Inhibition of autophagy by3-MA significantly inhibited OT-I CD8þ T-cell activation bya-TAGS obtained from both 3LL and 4T1 tumor cells (Fig. 6A).To further consolidate the role of autophagy in CD8þ T-cellactivation, a-TAGS was collected from doxycycline-treated4T1-Atg12shRNA tumor cells in which the autophagy pathwaywas inhibited by RNA interference–mediated knockdown ofAtg12. Activation of CT-TCR CD8þ T that are specific for theAH1 peptide derived from the envelope (GP70) of an endog-enous ecotropic murine leukemia virus in 4T1 tumor cells (21),by a-TAGS–pulsed DC was monitored by measuring IFN-glevels in the culture supernatant. Inhibition of autophagy byAtg12 gene knockdown dramatically diminished IFN-g pro-duction by CT-TCR CD8þ T cells (Fig. 6B). In contrast, CT-TCRCD8þT cells stimulated with DCs pulsedwitha-TAGS isolatedfrom cells expressing nonsilencing shRNA (4T1-CTRLshRNA)stimulated equivalent amounts of IFN-g in the absence orpresence of doxycycline (Fig. 6B).

Antitumor efficacy of a-TAGS–pulsed DC vaccinationFollowing our findings that a-TEA induces tumor cell

autophagy and that autophagosomes derived from a-TEA–treated tumor cells are antigen carriers that stimulate cross-presentation in vitro, we wanted to first evaluate whether

α

μ

P = 0.0131

mol/L)

A B

Figure 3. a-TEA–induced autophagy decreases tumor cell survival. A, the autophagy pathway was inhibited by doxycycline-inducible shRNA knockdownof Atg12 protein expression in 4T1 tumor cells (4T1-Atg12shRNA). 4T1-CTRLshRNA cells that express a nonspecific shRNA were included as control.After 72-hour doxycycline incubation, Atg12 protein levels were determined by Western immunoblot detection of the Atg12-Atg5 complex using an Atg12-specific antibody. B, the cells were then treated with a-TEA (60 mmol/L) for 24 hours. Nonadherent and adherent cells were collected, and thesurviving fraction (SF) was determined after 7 days of culture without a-TEA. Mean SF � SD from 4 independent experiments is shown. GAPDH,Glyceraldehyde-3-phosphate dehydrogenase.

Li et al.





a-TAGS can induce antigen-specific CD8þ T-cell proliferationin vivo.We found thata-TAGS isolated from3LL-OVAand 4T1-OVA cells that was injected into the inguinal lymph nodes ofnaive mice induced proliferation of over 90% and 50% ofadoptively transferred OT-I CD8þ T cells respectively, suggest-ing a-TAGS was efficiently cross-presented to CD8þ T cells invivo (Fig. 7A). We next determined the impact of a-TAGS–pulsed DC (a-TAGS-DC) vaccination on the growth of exper-imental 3LL lungmetastases. Mice were injected intravenouslywith 3LL tumor cells, received a-TAGS-DC vaccinations 3 dayslater, and visible lung metastases were enumerated 28 dayposttumor injection (Fig. 7B). The results show that a-TAGS-DC vaccination significantly reduced the number of lungmetastases �5-fold (P < 0.05) compared with no treatment(Control). Furthermore, the number of lung metastases inthe a-TAGS-DC vaccination group was significantly lower

(P < 0.05) compared with the DC alone (npDC), a-TAGS alone,or DC pulsed with freeze-thaw tumor cell lysate (Lysate-DC)groups. After these encouraging results, we wanted to deter-mine if a-TAGS-DC vaccination is also efficacious againstestablished tumors. For this purpose, 4T1 mammary tumor–bearing mice received 3 subcutaneous a-TAGS-DC injections.The data (Fig. 7C) show that a-TAGS-DC treatment reducedthe mean tumor size by 37% to 110 mm2 from 175 mm2 meantumor size in untreated mice. Lysate-DC and a-TAGS vacci-nation resulted in 34% and 32% tumor size reduction, respec-tively; npDC vaccination had aminimal effect on tumor growth(148 mm2). In addition, a-TAGS-DC treatment significantly (P< 0.05) prolonged the median survival to 37 days posttumorinjection compared with 29 days median survival of controlmice (Fig. 7D). Injections of npDC or a-TAGS alone had noeffect on survival (median survival in both groups was 31 days).Lysate-DC vaccination prolonged the median survival to35 days.

DiscussionIn this study, we investigated a-TEA–induced tumor cell

autophagy and its role in the enhancement of the antitumor_ + _ +

0.1 µm

A

B

0.1 µm

α

Figure 4. The high-speed (10,000 � g) supernatant fraction from a-TEA–treated tumor cells contains autophagosomes. A, 3LL or 4T1 tumor cellswere treated with 80 or 20 mmol/L a-TEA for 24 or 48 hours, respectively.The cell culture supernatant was collected, and dead cells and debriswere removed by centrifugation at 300� g for 10minutes. Subsequently,the low-speed supernatant was centrifuged at 10,000� g for 15minutes,and the high-speed pellet was examined by TEM. Double-membranevesicles (arrows) with a variety of morphology were present. Their sizesranged between 100 nm and 1 mm. Images were collected at originalmagnifications of �1,000 to �37,000. B, the high-speed fractions fromuntreated and a-TEA–treated 3LL or 4T1 cells were solubilized in lysisbuffer, and equal amounts of protein (5 mg) were electrophoreticallyseparated, and the autophagosome marker, LC3-II, was detected byWestern immunoblotting.

A

B

α3.6% 16.0%

H

6.3% 7.2% 78.4%

74.8%3LL-OVA

4T1-OVA

H H

NoNo

*P < 0.05 compared with all other groups

Figure 5. a-TAGS stimulates cross-presentation. 3LL and 4T1 tumor cellsthat stably express the OVA peptide (3LL-OVA, 4T1-OVA) were treatedwith a-TEA, and the cell culture supernatant was subjected to centrifugalfractionation. The a-TEA-generated autophagosome-enrichedsupernatant fraction (a-TAGS) was resuspended in culture medium andpulsed ontoDCs for 6 hours. TheDCswere thenwashedandcoincubatedwith CFSE-labeled OVA-specific TCR transgenic OT-I CD8þ T cells.UnpulsedDCs (no antigen) were included as controls. Proliferation of OT-ICD8þ T cells was assessed by determination of CFSE dilution using flowcytometry. A, representative histograms of percentage of proliferatingOT-I CD8þ T cells are shown. B, mean percentage of divided cells� SD isshown for 3 to 5 independent experiments.






immune response.We show for thefirst time that in addition toits well-described induction of tumor cell apoptosis, a-TEAinduces autophagy in lung and mammary tumor cells and thatautophagy-mediated cell death is partially responsible for thecytotoxic properties of a-TEA.

Autophagy is considered a prosurvival mechanism for recy-cling damaged organelles and proteins by cells undergoingvarious forms of stress including nutrient deprivation. How-ever, recent evidence suggests that under certain circum-stances and depending on the stressor, extensive, and irre-versible cellular damage can occur culminating in autophagiccell death (32, 33). The contribution of autophagy to cell deathhas been documented in studies in which inhibition of theautophagy pathway genes, Atg5, Atg6, or Atg7 by RNAi resultedin increased cell survival (34–36). Our finding that shRNAknockdown of Atg12 increased the clonogenicity of a-TEA–treated tumor cells is supportive of the role of autophagy ina-TEA–mediated tumor cell death.

Recently, it has become increasingly clear that autophagyand apoptosis are not independently regulated and that cross-talk may exist between both pathways (14, 15, 37). The simul-taneous induction of autophagy and apoptosis by a-TEA inour study has also been reported for other anticancer drugsincluding paclitaxel (38), mitoxantrone (39), oxaliplatin (39),

melphalan (38), arsenic trioxide (40), imatinib (41), and theproteasome inhibitor MG132 (42). In our study, we found thatautophagy preceded apoptosis during a-TEA treatment as hasbeen previously reported for other stress triggers includinggrowth factor starvation and apoptosis-inducing agents(27, 43). However, unlike previous reports, which showedautophagy inhibition in association with progressive apoptosis(27, 43), we showed that a-TEA–induced autophagy remainedhigh over time. The finding in our study that blockade ofapoptosis using the pan-caspase inhibitor, zVAD-fmk furtherincreased the formation of LC3-II punctates in a-TEA–treatedtumor cells, suggested cross-talk between both pathwaysduring a-TEA–mediated tumor cell cytotoxicity. Interestingly,in our study, a-TEA treatment resulted in only a very modestdecrease in Beclin-1 protein without detectable Beclin-1 cleav-age products (data not shown). In the studies by Wirawan andcolleagues (27) and Zhu and colleagues (43) where progressiveapoptosis correlated with autophagy inhibition, caspase-dependent cleavage of Beclin-1 was observed. The discrepancybetween their findings and ours could be due, in part, todifferences in the cell lines and stress triggers used in thestudies. Our results suggest that autophagy and apoptosisinduction by a-TEA occur in parallel and that Beclin-1 islikely not involved in regulation of cross-talk between bothpathways.

We also show here for the first time that the autophago-some-enriched fraction (a-TAGS), generated by treating tumorcells in vitrowitha-TEA, was an efficient tumor antigen carrier.DC pulsed with a-TAGS derived from OVA-expressing tumorcells stimulated antigen cross-presentation evidenced byenhanced proliferation of OVA-specificOT-I CD8þT cells. ThisCD8þ T-cell activation could be partially blocked with 3-MA,suggesting that autophagy is essential for efficient antigencross-presentation. Cross-presentation of an endogenoustumor-associated antigen (GP70) was also induced by a-TAGSand was inhibited by shRNA knockdown of Atg12, a geneproduct essential for autophagy. Thus, we show in our studyby both pharmacologic and genetic inhibition of autophagythat a-TEA–induced autophagy is necessary for a-TAGS–mediated stimulation of cross-presentation. The improvedantigen cross-presentation by a-TAGS–pulsed DC (a-TAGS-DC) observed in vitro was also evident in vivo as intranodallyinjected a-TAGS derived from 3LL-OVA and 4T1-OVA tumorcells induced proliferation of adoptively transferredOT-I CD8þ

T cells, suggesting thata-TAGS–associated tumor antigens arecross-presented in vivo. More importantly, a-TAGS-DC vacci-nation was significantly more efficacious than tumor lysate–pulsed DC (Lysate-DC) at reducing the incidence of experi-mental lung metastasis in the 3LL tumor model. a-TAGS-DCvaccination was less effective at suppressing established 4T1tumors and its ability to prolong animal survival was compa-rable to that of Lysate-DC that have long been recognizedto induce tumor-specific antitumor immune responses (44).Taken together, these results suggest that tumor antigensin a-TAGS were cross-presented in vivo to stimulate an anti-tumor immune response. These results are also in agreementwith our earlier report (2) showing that a-TEA treatmentof tumor-bearing mice enhanced the antitumor immune

B

NoNo

H

A

H

P <

P < 0.0085

Figure 6. Inhibition of autophagy before a-TEA treatment reduces cross-presentation by a-TAGS. A, the autophagy pathway was inhibited in3LL-OVA and 4T1-OVA cells by overnight incubation with the autophagyinhibitor 3-methyladenine (3-MA). Subsequently, a-TAGS was preparedand a cross-presentation assay was conducted. Mean percentage ofdivided OT-I CD8þ T cells � SD is shown from 3 independentexperiments. B, a-TAGS was collected from 4T1-Atg12shRNA cells inwhich the autophagy pathway was inhibited by doxycycline (DOX)-induced shRNA knockdown of Atg12 protein. a-TAGS was pulsed ontoDCs. TheDCswerewashed and coincubated for 48 hours withCT-TCRTcells specific for the 4T1 endogenous AH1 antigen. a-TAGS from4T1-CTRLshRNA cells that express a nonspecific shRNA were includedas control. IFN-g in the supernatant was detected by ELISA and meanIFN-g levels � SD from 2 independent experiments are shown.

Li et al.





response and resulted in increased cytokine secretion andtumor specific cytotoxicity and that the antitumor efficacyof a-TEA treatment was partially dependent on a functionalT-cell response (2).It has recently been reported that cytoreductive che-

motherapeutics such as gemcitabine (45) and doxorubicin(46) can have immune-stimulatory effects and the emergingconsensus is that these effects depend on the specificcytotoxic mechanisms of a given drug (47). Here we provideevidence for the first time that a-TEA induces tumor cellautophagy and that the autophagosome-enriched fractionstimulated tumor-specific cross-presentation that wasdependent on a functional autophagy pathway. Because ofits ability to stimulate autophagy and enhance cross-primingof CD8þ T cells, a-TEA chemotherapy may have clinicalrelevance as a new strategy for generating tumor-associatedantigens for cross-presentation by endogenous DC. Com-bining a-TEA therapy with immune modulators such asanti-CTLA-4, anti-OX40, or anti-4-1BB, which promote T-cell expansion, effector function, and survival could be aneffective strategy for enhancing the antitumor response.

Some of these agents either have already gained Food andDrug Administration approval (anti-CTLA-4; ref. 48) or are invarious stages of clinical testing [anti-OX40 (ref. 49;NCT01416844), anti-4-1BB (ref. 50; NCT01471210), anti-pro-grammed death ligand-1 (NCT00729664), anti-programmeddeath-1 (NCT00730639)] and could potentially be moreeffective when combined with a-TEA. Plans are underwayat our institute to test the safety and tolerability of a-TEA ina first-in-human phase I trial in patients with advancedcancer.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: Y. Li, T. Hahn, H.-M. Hu, E.T. AkporiayeDevelopment of methodology: Y. Li, T. Hahn, K. Garrison, Z.-H. Cui, E.T.AkporiayeAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Li, T. Hahn, K. Garrison, Z.-H. Cui, E.T. AkporiayeAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Li, T. Hahn, E.T. AkporiayeWriting, review, and/or revision of themanuscript: Y. Li, T. Hahn, Z.-H. Cui,A. Thorburn, H.-M. Hu, E.T. Akporiaye

P < 0.05

A B

C D

P < 0.05

P < 0.05

Days posttumor injection

% d

ivid

ed O

T-I

Per

cen

t su

rviv

al

Nu

mb

er o

f lu

ng

met

asta

ses

* P < 0.05 compared with all other groups

* P < 0.05 compared with control, npDC, α-TAGS

Figure 7. a-TAGS induces CD8þ T-cell immune responses in vivo and DCs pulsed with a-TAGS show therapeutic antitumor effect. A, in vivo cross-presentation of a-TAGS antigens. a-TAGS from 3LL-OVA, 3LL, 4T1-OVA, and 4T1 tumor cells were injected into both inguinal lymph nodes of naiveC57BL/6 or BALB/c mice. CFSE-labeled Thy1.1þ OT-I splenocytes were adoptively transferred by i.v. injection, and proliferation of OT-I CD8þ T cells in thelymph nodes was analyzed by flow cytometry 5 days post–T-cell transfer. Data represent mean � SEM of 3 mice per group. B, to determine the effectof DCs pulsed with a-TAGS on 3LL lung metastases, C57BL/6 mice were i.v. injected with 3LL tumor cells. Three days later, the mice received s.c.vaccination with unpulsed DC (npDC), DC pulsed with a-TAGS derived from 3LL tumor cells (a-TAGS-DC), DC pulsed with freeze-thaw tumor cell lysate(Lysate-DC) or a-TAGS alone. Numbers of lung metastases of individual mice (n ¼ 6 per group) on day 28 posttumor injection. Boxed numbersindicate means of lung metastases � SEM. C, to assess the effect of DCs pulsed with a-TAGS on 4T1 tumor growth, BALB/c mice with established4T1 mammary tumors (�13 mm2), received s.c. injections of npDC, a-TAGS-DC, Lysate-DC, or a-TAGS alone on days 7, 9, and 11 posttumor injection.Individual tumor areas on day 28 posttumor injection. Boxed numbers indicate mean tumor areas � SEM. D, Kaplan–Meier analysis of survival.






Administrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): Y. Li, T. Hahn, K. Garrison, Z.-H. Cui,E.T. AkporiayeStudy supervision: H.-M. Hu, E.T. AkporiayeProduction of reagents: A. Thorburn, J. Thorburn

AcknowledgmentsThe authors thank Dr. EdwinWalker and the staff of the ImmuneMonitoring

Laboratory at the EACRI for their technical assistance. Furthermore, the authorsthank the staff of the EACRI vivarium. The authors also thank Dr. Eric Barklis,Mike Webb, and the Oregon Health and Science University Electron MicroscopyCore Facility for electron microscopy imaging.

Grant SupportThis study was supported by grants 5R01CA120552 to E.T. Akporiaye;

R01CA10724375 and R21CA141278 to H.-M. Hu; and 2R01CA111421 and1R01CA150925 to A. Thorburn from the NIH.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact.

Received September 14, 2011; revised March 28, 2012; accepted April 22, 2012;published OnlineFirst June 28, 2012.

References1. Kline K, Lawson KA, Yu W, Sanders BG. Vitamin E and breast cancer

prevention: current status and future potential. J Mammary Gland BiolNeoplasia 2003;8:91–102.

2. Hahn T, Jagadish B, Mash EA, Garrison K, Akporiaye ET. Alpha-tocopheryloxyacetic acid: a novel chemotherapeutic that stimu-lates the anti-tumor immune response. Breast Cancer Res 2011;13:R4.

3. Kimmick GG, Bell RA, Bostick RM. Vitamin E and breast cancer: areview. Nutr Cancer 1997;27:109–17.

4. Klein EA, Thompson IM Jr, Tangen CM, Crowley JJ, Lucia MS,Goodman PJ, et al. Vitamin E and the risk of prostate cancer: theSelenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA2011;306:1549–56.

5. Lawson KA, Anderson K, Menchaca M, Atkinson J, Sun L, Knight V,et al. Novel vitamin E analogue decreases syngeneicmousemammarytumor burden and reduces lung metastasis. Mol Cancer Ther2003;2:437–44.

6. Jia L, YuW,Wang P, Sanders BG, Kline K. In vivo and in vitro studies ofanticancer actions of alpha-TEA for human prostate cancer cells.Prostate 2008;68:849–60.

7. Hahn T, Szabo L, Gold M, Ramanathapuram L, Hurley LH, AkporiayeET. Dietary administration of the proapoptotic vitamin e analogue{alpha}-tocopheryloxyacetic acid inhibits metastatic murine breastcancer. Cancer Res 2006;66:9374–8.

8. Hahn T, Fried K, Hurley LH, Akporiaye ET. Orally active {alpha}-tocopheryloxyacetic acid suppresses tumor growth and multiplicityof spontaneous murine breast cancer. Mol Cancer Ther 2009;8:1570–8.

9. Kline K, YuW, Sanders BG. Vitamin E: mechanisms of action as tumorcell growth inhibitors. J Nutr 2001;131:161S–3S.

10. Neuzil J, Tomasetti M, Mellick AS, Alleva R, Salvatore BA, Birringer M,et al. Vitamin E analogues: a new class of inducers of apoptosis withselective anti-cancer effects. Curr Cancer Drug Targets 2004;4:355–72.

11. Armstrong JS.Mitochondrialmembranepermeabilization: the sinequanon for cell death. Bioessays 2006;28:253–60.

12. Jia L, Yu W, Wang P, Li J, Sanders BG, Kline K. Critical roles forJNK, c-Jun, and Fas/FasL-signaling in vitamin E analog-inducedapoptosis in human prostate cancer cells. Prostate 2008;68:427–41.

13. Wang P, Yu W, Hu Z, Jia L, Iyer VR, Sanders BG, et al. Involvement ofJNK/p73/NOXA in vitamin E analog-induced apoptosis of humanbreast cancer cells. Mol Carcinog 2008;47:436–45.

14. Maiuri MC, Zalckvar E, Kimchi A, Kroemer G. Self-eating and self-killing: crosstalk between autophagy and apoptosis. Nat Rev Mol CellBiol 2007;8:741–52.

15. Eisenberg-Lerner A, Bialik S, Simon HU, Kimchi A. Life and deathpartners: apoptosis, autophagy and the cross-talk between them. CellDeath Differ 2009;16:966–75.

16. Eisenberg-Lerner A, Kimchi A. The paradox of autophagy and itsimplication in cancer etiology and therapy. Apoptosis 2009;14:376–91.

17. Eskelinen EL. New insights into the mechanisms of macro-autophagy in mammalian cells. Int Rev Cell Mol Biol 2008;266:207–47.

18. Li Y, Wang LX, Yang G, Hao F, Urba WJ, Hu HM. Efficient cross-presentation depends on autophagy in tumor cells. Cancer Res2008;68:6889–95.

19. Munz C. Enhancing immunity through autophagy. Annu Rev Immunol2009;27:423–49.

20. Kepp O, Tesniere A, Schlemmer F, MichaudM, Senovilla L, Zitvogel L,et al. Immunogenic cell death modalities and their impact on cancertreatment. Apoptosis 2009;14:364–75.

21. Jordan KR, McMahan RH, Oh JZ, Pipeling MR, Pardoll DM, Kedl RM,et al. Baculovirus-infected insect cells expressing peptide-MHC com-plexes elicit protective antitumor immunity. J Immunol 2008;180:188–97.

22. Rasband WS. Image J. U. S. National Institutes of Health, Bethesda,Maryland, USA, http://imagej.nih.gov/ij/, 1997–2012.

23. He Y, Pimenov AA, Nayak JV, Plowey J, Falo LD Jr, Huang L.Intravenous injection of naked DNA encoding secreted flt3 liganddramatically increases the number of dendritic cells and natural killercells in vivo. Hum Gene Ther 2000;11:547–54.

24. Kobie JJ, Wu RS, Kurt RA, Lou S, Adelman MK, Whitesell LJ, et al.Transforming growth factor beta inhibits the antigen-presenting func-tions and antitumor activity of dendritic cell vaccines. Cancer Res2003;63:1860–4.

25. Klionsky DJ, Abeliovich H, Agostinis P, Agrawal DK, Aliev G, AskewDS, et al. Guidelines for the use and interpretation of assays formonitoring autophagy in higher eukaryotes. Autophagy 2008;4:151–75.

26. Tiwary R, YuW, Sanders BG, Kline K. Alpha-TEA cooperates withMEKor mTOR inhibitors to induce apoptosis via targeting IRS/PI3K path-ways. Br J Cancer 2011;104:101–9.

27. Wirawan E, Vande Walle L, Kersse K, Cornelis S, Claerhout S, Vano-verberghe I, et al. Caspase-mediated cleavage of Beclin-1 inactivatesBeclin-1-induced autophagy and enhances apoptosis by promotingthe release of proapoptotic factors from mitochondria. Cell Death Dis2010;1:e18.

28. Cho DH, Jo YK, Hwang JJ, Lee YM, Roh SA, Kim JC. Caspase-mediated cleavage of ATG6/Beclin-1 links apoptosis to autophagy inHeLa cells. Cancer Lett 2009;274:95–100.

29. Thorburn A. Studying autophagy's relationship to cell death. Autop-hagy 2008;4:391–4.

30. Li Y, Wang LX, Pang P, Twitty C, Fox BA, Aung S, et al. Cross-presentation of tumor associated antigens through tumor-derivedautophagosomes. Autophagy 2009;5:576–7.

31. Munz C. Autophagy and antigen presentation. Cell Microbiol2006;8:891–8.

32. Klionsky DJ, Emr SD. Autophagy as a regulated pathway of cellulardegradation. Science 2000;290:1717–21.

33. Gozuacik D, Kimchi A. Autophagy and cell death. Curr Top Dev Biol2007;78:217–45.

34. Yu L, Alva A, Su H, Dutt P, Freundt E, Welsh S, et al. Regulation of anATG7-beclin 1 program of autophagic cell death by caspase-8. Sci-ence 2004;304:1500–2.

35. Shimizu S, Kanaseki T, Mizushima N,Mizuta T, Arakawa-Kobayashi S,Thompson CB, et al. Role of Bcl-2 family proteins in a non-apoptoticprogrammed cell death dependent on autophagy genes. Nat Cell Biol2004;6:1221–8.

Li et al.





36. Pyo JO, Jang MH, Kwon YK, Lee HJ, Jun JI, Woo HN, et al. Essentialroles of Atg5 and FADD in autophagic cell death: dissection of autop-hagic cell death into vacuole formation and cell death. J Biol Chem2005;280:20722–9.

37. Djavaheri-Mergny M, Maiuri MC, Kroemer G. Cross talk betweenapoptosis and autophagy by caspase-mediated cleavage of Beclin1. Oncogene 2010;29:1717–9.

38. Tang D, Kang R, Cheh CW, Livesey KM, Liang X, Schapiro NE, et al.HMGB1 release and redox regulates autophagy and apoptosis incancer cells. Oncogene 2010;29:5299–310.

39. Michaud M, Martins I, Sukkurwala AQ, Adjemian S, Ma Y, Pellegatti P,et al. Autophagy-dependent anticancer immune responses inducedbychemotherapeutic agents in mice. Science 2011;334:1573–7.

40. Qian W, Liu J, Jin J, Ni W, Xu W. Arsenic trioxide induces not onlyapoptosis but also autophagic cell death in leukemia cell lines via up-regulation of Beclin-1. Leuk Res 2007;31:329–39.

41. Basciani S, Vona R, Matarrese P, Ascione B, Mariani S, Cauda R, et al.Imatinib interferes with survival of multidrug resistant Kaposi's sarco-ma cells. FEBS Lett 2007;581:5897–903.

42. Yang W, Monroe J, Zhang Y, George D, Bremer E, Li H. Proteasomeinhibition induces both pro- and anti-cell death pathways in prostatecancer cells. Cancer Lett 2006;243:217–27.

43. ZhuY, Zhao L, Liu L, Gao P, TianW,Wang X, et al. Beclin 1 cleavage bycaspase-3 inactivates autophagy and promotes apoptosis. ProteinCell 2010;1:468–77.

44. Fields RC, Shimizu K, Mule JJ. Murine dendritic cells pulsed withwhole tumor lysates mediate potent antitumor immune responsesin vitro and in vivo. Proc Natl Acad Sci U S A 1998;95:9482–7.

45. Nowak AK, Robinson BW, Lake RA. Synergy between chemotherapyand immunotherapy in the treatment of established murine solidtumors. Cancer Res 2003;63:4490–6.

46. Casares N, Pequignot MO, Tesniere A, Ghiringhelli F, Roux S,Chaput N, et al. Caspase-dependent immunogenicity of doxorubi-cin-induced tumor cell death. J Exp Med 2005;202:1691–701.

47. Zitvogel L, Kepp O, Kroemer G. Decoding cell death signals in inflam-mation and immunity. Cell 2010;140:798–804.

48. Hodi FS, O'Day SJ, McDermott DF, Weber RW, Sosman JA, HaanenJB, et al. Improved survival with ipilimumab in patients with metastaticmelanoma. N Engl J Med 2010;363:711–23.

49. Sugamura K, Ishii N, Weinberg AD. Therapeutic targeting of theeffector T-cell co-stimulatory molecule OX40. Nat Rev Immunol2004;4:420–31.

50. Cheuk AT, Mufti GJ, Guinn BA. Role of 4-1BB:4-1BB ligand in cancerimmunotherapy. Cancer Gene Ther 2004;11:215–26.






Published OnlineFirst June 28, 2012.Cancer Res Yuhuan Li, Tobias Hahn, Kendra Garrison, et al. and Enhances Antigen Cross-Presentation

-TEA Stimulates Tumor AutophagyαThe Vitamin E Analog

Updated version

10.1158/0008-5472.CAN-11-3103doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2012/07/10/0008-5472.CAN-11-3103.DC1

Access the most recent supplemental material at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/early/2012/06/27/0008-5472.CAN-11-3103To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-11-3103

http://cancerres.aacrjournals.org/content/suppl/2012/07/10/0008-5472.CAN-11-3103.DC1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/early/2012/06/27/0008-5472.CAN-11-3103


the vitamin e analog a-tea stimulates tumor - cancer research

Documents