acquisition of resistance of pancreatic cancer …...(pi3k)/akt pathways (6, 13). 2-methoxyestradiol...

DNA Damage and Cellular Stress Responses

Acquisition of Resistance of Pancreatic Cancer Cells to2-Methoxyestradiol Is Associated with the Upregulation ofManganese Superoxide Dismutase

Jianhong Zhou and Yuchun Du

AbstractAcquired resistance of cancer cells to anticancer drugs or ionizing radiation (IR) is one of the major obstacles in

cancer treatment. Pancreatic cancer is an exceptional aggressive cancer, and acquired drug resistance in this cancer iscommon. Reactive oxygen species (ROS) play an essential role in cell apoptosis, which is a key mechanism by whichradio- or chemotherapy induce cell killing.Mitochondria are themajor source of ROS in cells. Thus, alterations in theexpression of mitochondrial proteins, involved in ROS production or scavenging, may be closely linked to theresistance of cancer cells to radio- or chemotherapy. In the present study, we generated a stable cell line by exposingpancreatic cancer cells to increasing concentrations of ROS-inducing, anticancer compound 2-methoxyestradiol(2-ME) over a 3-month period. The resulting cell line showed strong resistance to 2-ME and contained an elevatedlevel of ROS. We then used a comparative proteomics method to profile the differential expression of mitochondrialproteins between the parental and the resistant cells. One protein identified to be upregulated in the resistant cells wasmanganese superoxide dismutase (SOD2), a mitochondrial protein that converts superoxide radicals to hydrogenperoxides. Silencing of SOD2 resensitized the resistant cells to 2-ME, and overexpression of SOD2 led the parentalcells to 2-ME resistance. In addition, the 2-ME–resistant cells also showed resistance to IR. Our results suggest thatupregulation of SOD2 expression is an important mechanism by which pancreatic cancer cells acquire resistance toROS-inducing, anticancer drugs, and potentially also to IR. Mol Cancer Res; 10(6); 768–77. �2012 AACR.

IntroductionPancreatic cancer is an exceptionally aggressive cancer,

with the 5-year survival rate of approximately 2%, andmedian survival after diagnosis of less than 6 months (1).Intrinsic and acquired resistance of most patients withpancreatic cancer to radio- and chemotherapy is one of themajor challenges in pancreatic cancer treatment. Currently,less than 10%of patients with pancreatic cancer benefit fromradiation therapy, and 20% of patients from chemotherapy(2, 3). Because of lack or poor therapeutic responses ofpancreatic cancer to therapies, the long-term survival rate forpatients with organ-confined pancreatic cancer is only 20%,much lower than those formany othermajor types of cancers(4). Thus, understanding the mechanisms underlying theresistance of pancreatic cancer to radio- and chemotherapyand developing novel treatment strategies directed againstthis devastating disease are greatly needed.

Reactive oxygen species (ROS), such as superoxide(O2

*�), hydrogen peroxide (H2O2), and hydroxyl radical(HO

*

), are the reactive chemical molecules that are con-stantly produced and eliminated in all living cells in abalanced manner. A loss of the balance between produc-tion and elimination would have severe consequences forcells. A moderate increase of cellular ROS accumulationstimulates cell growth and proliferation and promotes thedevelopment of cancers (5). However, acute and excessiveROS accumulation triggers apoptotic pathways and leadsto cell death (6). Because of oncogenic signaling, increasedmetabolic activity, mitochondrial malfunction, and thehypoxia growth environment of cancer cells, cancer cellsusually contain elevated levels of ROS compared withnormal cells (7, 8). This unique difference between cancercells and healthy cells may provide opportunities todevelop novel therapeutic strategies to selectively killcancer cells by inducing cellular ROS burst (6, 7, 9).Ionizing radiation (IR) and many anticancer drugs areknown to kill cancer cells largely through ROS-mediatedapoptosis (10). Thus, identification of the proteins/enzymes that play a critical role in ROS metabolism inrelation to cancer drug resistance may provide valuableinformation on designing novel anticancer drugs to over-come chemo- and radioresistance in cancer.Mitochondria are the organelles where most cellular ROS

are generated and play a central role in ROS-mediatedapoptosis (11). Approximately 90%of the oxygen consumed

Authors' Affiliation: Department of Biological Sciences, University ofArkansas, Fayetteville, Arkansas

Note: Supplementary data for this article are available at Molecular CancerResearch Online (http://mcr.aacrjournals.org/).

Corresponding Author: Yuchun Du, Department of Biological Sciences,University of Arkansas, Fayetteville, AR 72701. Phone: 479-575-6944; Fax:479-575-4010; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-11-0378

�2012 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 10(6) June 2012768

on June 24, 2020. © 2012 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst April 30, 2012; DOI: 10.1158/1541-7786.MCR-11-0378

http://mcr.aacrjournals.org/

within a eukaryote is used in mitochondrial respiration, andabout 1% to 2% of the overall oxygen consumption is usedfor producing ROS (12). ROS formation in mitochondria ismainly due to electron leakage naturally occurring in com-plexes I and III of the respiratory chain located in themitochondrial membrane (12). Incomplete oxygen reduc-tion within the mitochondrial respiration chain leads to theformation of the first molecule of ROS, superoxide radicals(O2

*�). The superoxide radicals are then dismutased toproduce hydrogen peroxides (H2O2) by mitochondrialmanganese superoxide dismutase (SOD2), and the pro-duced H2O2 is further converted into water through gluta-thione peroxidase (GPx) and other molecules (12). Therelatively long-lived H2O2 can pass mitochondrial mem-brane by diffusion into the cytosol and nucleus to activate arange of signaling pathways, including mitogen-activatedprotein kinase (MAPK) and phosphoinositide 3-kinase(PI3K)/AKT pathways (6, 13).2-Methoxyestradiol (2-ME) is a natural physiologic

metabolite of estrogen, 17b-estradiol. A variety of cancercells, including pancreatic cancer cells, were shown to besensitive to 2-ME (14–16). 2-ME specifically affects cancercells with no effects on normal cells (17), and its anticancereffects are largely linked to its ability to trigger mitochon-drion-dependent apoptosis through inducing accumulationof cellular ROS (15, 18). To explore the molecular mech-anism underlying the acquired resistance of pancreaticcancer cells to ROS-inducing anticancer drugs, we generateda stable cell line by exposing pancreatic cancer cells toincreasing concentrations of 2-ME over a period of 3months. The resulting cell line showed strong resistance to2-ME and contained an elevated level of ROS.We then useda 2-dimensional gel electrophoresis (2-DE)–based proteo-mics method to identify the mitochondrial proteins thatwere differentially expressed between the parental and theresistant cells. One protein identified was SOD2. Functionalanalysis revealed that SOD2 played a critical role in protect-ing the 2-ME–resistant cells from the excessive oxidativestress, thus conferring resistance to ROS-induced apoptosisto the cells.

Materials and MethodsCell culture and generation of stable cells with acquiredresistance to 2-METhe pancreatic cancer cells that we used to generate stable

cells were MIA PaCa-2, which is an undifferentiated, pri-mary pancreatic cancer cell line sensitive to 2-ME (19, 20).The MIA PaCa-2 cells were routinely maintained in Dul-becco's Modified Eagle's Medium (DMEM) supplementedwith 10% FBS and 1% penicillin and streptomycin. 2-ME–resistant stable cells were generated by continuously expos-ing the MIA PaCa-2 cells to stepwise increasing 2-MEconcentrations in the range of 0.5 to 2.5 mmol/L over aperiod of 3 months. After the cells adapted to one 2-MEconcentration, typically requiring 2 to 3 weeks, exposure to2-ME was increased by 0.5 mmol/L. The process wasrepeated until 2-ME concentration reached 2.5 mmol/L.

Beyond 2.5 mmol/L, no stable cells were successfullyrecovered.

ROS measurementCellular ROS concentrations were measured with a flow

cytometer (BD Biosciences) after incubating cells with afluorescence probe 20,70-dichlorofluorescein-diacetate(DCFH-DA; Invitrogen) as described by Gottlieb andcolleagues (21). The parental and the resistant cells werecultured with 2.5 mmol/L 2-ME for 1 week, then harvested,washed with cold PBS, and incubated with 10 mmol/LDCFH-DA for 30 minutes. Propidium iodide was addedto each sample immediately before flow cytometric analysisto differentiate dead and live cells. Fluorescence was analyzedonly in live cells. The cells were also incubated with propi-dium iodide alone, and the mean fluorescence of live cells ofthe single-stained cells was subtracted from the live cells ofthe double-stained cells. For polyethylene glycol (PEG)-catalase or PEG-SOD treatments (22), the parental and theresistant cells were pretreated with control media or mediacontaining 100 U/mL PEG-catalase or 100 U/mL PEG-SOD (Sigma) for 24 hours, followed by an incubation with2.5 mmol/L 2-ME in the media for an additional 16 hours.The cells were then washed and incubated with 5 mmol/LCM-H2DCFDA (Invitrogen) for 20minutes at 37�C. Afterthis, the cells were harvested, washed twice with cold PBS,and analyzed by flow cytometry as described above.

Cell viability assay and clonogenic survival assayCells were cultured with the indicated concentrations of

2-ME for 48 hours and then used for a viability assay, whichwas conducted with a CellTiter-Glo Luminescent CellViability Assay Kit (Promega). Luminescence was measuredwith a SPECTRAmax GEMINI XS fluorescence microplatereader (Molecular Devices). The clonogenic survival assaywas conducted according to Franken and colleagues (23).Briefly, the parental and the resistant cells were plated in the60-mm dishes in DMEM and cultured for 6 hours. 2-MEwas then added to each plate resulting in a final 2-MEconcentration of 0, 0.25, 0.5, 1.0, or 1.5 mmol/L in theculture medium (or the cells were exposed to different dosesof g-ray). The cells were cultured for 21 days with a mediumchange every 4 days. Colonies with more than 50 cells werecounted, and surviving fractions were normalized withplating efficiencies. Triplicate plates were used for each2-ME concentration or irradiation dose.

Isolation of mitochondriaMitochondria were purified from cells using differential

centrifugation and nonlinear sucrose ultracentrifugation(Supplementary Fig. S1; refs. 24, 25).

2-DEThree hundred micrograms of mitochondrial protein

dissolved in a rehydration buffer was loaded onto a 17-cmReadyStrip IPG strip (pH 3–10), which was in turn kept atroom temperature overnight. Other procedures were con-ducted as described previously (26, 27).

SOD2 and Acquired Resistance of Pancreatic Cancer to ROS

www.aacrjournals.org Mol Cancer Res; 10(6) June 2012 769




Liquid chromatography/tandem mass spectrometryanalysis and database searchThe liquid chromatography/tandem mass spectrometry

(LC/MS-MS) analysis was carried out using a LTQ-XL massspectrometer (Thermo) in the Proteomic Facility at theUniversity of Arkansas for Medical Sciences (Little Rock,AR). Proteins were in-gel digested with trypsin (Promega)overnight at 37�C, and the resultingpeptidesweredissolved in20 mL 0.1% formic acid for LC/MS-MS analysis. Mascot(V2.2; Matrix Science) was used to search against the Inter-national Protein Index (IPI) human protein database (V3.68)using LC/MS-MS data as described (26, 27). The parametersfor database searching were as follows: (i) 2.0-Da mass errortolerance forMS and 0.65-Da forMS-MS, (ii) a maximumofone missed cleavage, and (iii) variable modifications: acetyla-tion at peptide N-terminus, phosphorylation on tyrosine/serine/threonine, and oxidation on methionine.

Overexpression and small hairpin RNA silencing ofSOD2For overexpression, the coding sequence of human SOD2

was cloned into the HindIII and XhoI sites of plasmidpcDNA3.1, and the resulting construct was transfected intothe parental cells with calcium phosphate method. Forsilencing of SOD2, shSOD2 sequence (20) was annealedand cloned into the SalI and XbaI sites of a safety-modifiedretroviral vector pSuppressorRetro (Imgenex). Cotranfec-tion of the retroviral construct with a packaging vector pCL-10A1 into 293T cells and the subsequent infection of theresistant cells with the resulting viral particles were con-ducted according to the manufacturer's instructions (Imge-nex). A negative control plasmid harboring a scrambledsequence that does not show significant homology to humangene sequences from the same company was transfected andinfected in the same way as described above. For bothoverexpression and small hairpin RNA (shRNA) silencing,stable cells were selected with 1 mg/mL G418, and SOD2-overexpressing clones or the clones with knockdown ofSOD2 were screened using Western blotting with anti-SOD2 antibodies.

Western blottingWestern blotting was conducted as described previously

(26, 27). Antibodies were purchased from the followingsources: anti-SOD2, peroxiredoxin III (Prx III) and actinfrom Santa Cruz Biotechnology, and anti-PARP from BDPharmingen.

Measurements of SOD2 and GPx activitiesThe parental or resistant cells (�1.8 � 107) were washed

with ice-cold PBS twice and harvested with a rubber police-man. The cells were lysed in 3 packed cell pellet volumes of0.05 mol/L phosphate buffer, pH 7.8, by 3 cycles ofsonication (Branson Digital Sonifier 450) on ice with amicrotip at 15% power (10 � 0.5 s pulses in each cycle).After centrifugation at 10,000 � g for 15 minutes at 4�C,protein concentration in the supernatant was determined bya BCAProtein Assay Kit (Thermo Scientific). SOD2 activity

was measured using a SOD Assay Kit-WST (DojindoMolecular Technologies). The measurements were con-ducted according to manufacturer's instructions with thefollowing specifications: (i) the cell extract was incubatedwith 5 mmol/L potassium cyanide for 40 minutes at roomtemperature to inactivate Cu/Zn-SOD activities; (ii) 25 to40 mg protein was used in each reaction; (iii) the standardcurve was generated using serial dilutions of SOD frombovine erythrocytes, and one unit of SOD is defined asthe amount of enzyme that inhibits the rate of cytochromec reduction by 50% at pH 7.8 and 25�C (Sigma). Enzymaticactivity was expressed as units per mg of protein.GPx activity wasmeasured bymonitoring the oxidation of

NADPHwith aUV spectrophotometer at 340 nm (28). Thereaction mixture contained 50 mmol/L Tris-HCl, pH 8.0,0.5 mmol/L EDTA, 0.25 mmol/NADPH, 2.1 mmol/Lreduced glutathione, 0.5 U glutathione reductase (Sigma),and 0.5 to 1mg cell extract prepared as described above. Thereaction was initiated by adding 300 mmol/L t-butyl hydro-peroxide, and the final volume of the reaction mixture was 1mL. The enzyme-catalyzed oxidation of NADPH was mon-itored for 5 minutes at 25�C.

Statistical analysisP values were calculated by one-way ANOVA (PSI-

PLOT).

ResultsGeneration of stable cell line with acquired resistance toROS-inducing compound 2-METo eliminate the complications caused by genetic back-

ground variations, we generated a pancreatic cancer isogenicstable cell line that differs in sensitivity to ROS-inducingcompound 2-ME. By exposing the pancreatic cancer MIAPaCa-2 cells to increasing concentrations of 2-ME (0.5–2.5mmol/L) over a 3-month period, we generated an isogeniccell linewith acquired resistance to 2-ME, designated asMIAPaCa-2R. As shown in Fig. 1A, the resistant MIA PaCa-2Rcells showed strong resistance to 2-ME–mediated cell killingcompared with the parental MIA PaCa-2 cells. We also usedWestern blotting to detect the cleavage of PARP to monitorcell apoptosis. As shown in Fig. 1B, the MIA PaCa-2R cellswere more resistant to the 2-ME–mediated apoptosis thanthe parental cells. Apoptosis is not the only form of cell deathin most tumors, and clonogenic survival assay is widelyregarded as the "gold standard" cellular sensitivity assay dueto its ability to measure long-term cell survival and all modesof cell death. We conducted clonogenic survival assay on theparental and the resistant cells. As shown in Fig. 1C, theMIAPaCa-2R cells showed markedly enhanced clonogenic sur-vival over the parental cells after the cells were exposed toincreasing concentrations of 2-ME.

The resistant cells contain a higher level of cellular ROSthan the parental cellsTo examine whether ROS were involved in the resistance

of theMIA PaCa-2R cells to 2-ME–mediated cell killing and

Zhou and Du

Mol Cancer Res; 10(6) June 2012 Molecular Cancer Research770




apoptosis, we measured cellular ROS levels with flow cyto-metry using a fluorescent probe DCFH-DA (21). Inthis analysis, once the cell-permeable fluorescent probe

DCFH-DA is in the cells, it is cleaved by cellular esteraseinto its nonfluorescent and nonmembrane-permeable form,DCFH, and thus is trapped in the cells. The DCFH is thenoxidized by cellular oxidants into its fluorescent form, DCF.After incubating cells withDCFH-DA, it was shown that thefluorescence intensity in the MIA PaCa-2R cells was almost3-fold of that in the MIA PaCa-2 cells (Fig. 2A). To makesure the DCF fluorescence intensity represented the ROSlevels in cells, we incubated cells with cell-permeable PEG-catalase or PEG-SOD to examine whether they could inhibitthe fluorescence signal intensity. The results showed thatpretreatments of both the parental and resistant cells withPEG-catalase or PEG-SOD significantly inhibited the DCFfluorescent signals (Fig. 2B and C; for uncharacterizedreasons, PEG-catalase pretreatments resulted in more pro-nounced decrease in fluorescence signal intensity in theparental cells than in the resistant cells), suggesting thatDCF fluorescence signal intensities we measured repre-sented the ROS levels in the cells. In short, results in thissection suggest that cellular ROS levels were significantlydifferent between the parental and resistant cells, and cellularROS may be closely related to the acquired resistance of theMIA PaCa-2R cells to 2-ME.

Isolation of mitochondria from cellsFor conducting comparative proteomic analysis using

purified mitochondria as starting materials, we used differ-ential centrifugation and nonlinear sucrose ultracentrifuga-tion to purify mitochondria from theMIA PaCa-2 andMIAPaCa-2R cells (24, 25). Western blot analysis with anti-bodies against marker proteins showed that the purifiedmitochondria were essentially free of cytosolic and nuclearcontaminations but contained endoplasmic reticulum pro-teins (Supplementary Fig. S1), which are usually difficult toremove using differential centrifugation (24).

SOD2 is upregulated in the resistant MIA PaCa-2R cellsWe then used 2-DE to profile protein expression

between the MIA PaCa-2 cells and MIA PaCa-2R cellsusing the purified mitochondria as starting materials. Ofmore than 650 protein spots resolved, multiple proteinspots showed at least a 50% difference in the normalizedvolume between the MIA PaCa-2 cells and MIA PaCa-2Rcells. LC/MS-MS analysis and database search revealedthat one protein that showed approximately a 2-foldincrease in the normalized volume in the MIA PaCa-2R gel over the counterpart in parental cell gel was SOD2(Fig. 3A). SOD2 is the mitochondrial enzyme that con-verts superoxide (O2

*�) into H2O2, which is furthereliminated by GPx and other molecules (12). Increasesof SOD2 expression have been detected in primary cancertissues of different types of cancers (14, 29, 30). Because ofthe critical roles of SOD2 in ROS metabolism in normaland cancer cells, we selected this enzyme for furtheranalysis. First, we examined the expression of SOD2 byWestern blotting with anti-SOD2 antibodies and con-firmed that the expression of SOD2 in the MIA PaCa-2Rcells was indeed upregulated (Fig. 3B). To examine

Figure 1. Different 2-ME sensitivities between the parental MIA PaCa-2and the resistant MIA PaCa-2R cells. A, MIA PaCa-2 and MIA PaCa-2Rcells were cultured with the indicated concentrations of 2-ME for 72hours, and cell viabilities were examined with the CellTiter-GloLuminescent Cell Viability Assay. As shown, the MIA PaCa-2R cells weremore resistant to 2-ME than theMIA PaCa-2 cells. B, the parental and theresistant cellswere culturedwith the indicatedconcentrations of 2-ME for48 hours, and proteins from each population were analyzed by Westernblotting. While 2.5 mmol/L 2-ME induced a significant amount of the MIAPaCa-2 to apoptosis, theMIAPaCa-2R cells showed no apoptosis underthe same condition. b-Actin was used as a loading control. Equalamounts of total protein were loaded in each lane (40 mg/lane).C, clonogenic survival assay. MIA PaCa-2 and MIA PaCa-2R cells werecultured with the presence of indicated concentrations of 2-ME for 21days. Colonies with more than 50 cells were counted. As shown, the MIAPaCa-2R cells were more 2-ME–resistant than the parental cells.






whether the increased expression would lead to anenhanced SOD2 enzymatic activity in the resistant cells,we measured the catalytic activity of SOD2. Consistentwith the expression data, SOD2 enzymatic activity in theMIA PaCa-2R cells was also approximately 2-fold of thatin the parental cells (Fig. 3C).

Ectopic overexpression of SOD2 decreases the sensitivityof MIA PaCa-2 cells to 2-METo examine whether the upregulated SOD2 protected the

MIA PaCa-2R cells from 2-ME–mediated cell killing andapoptosis (Fig. 1), we generated a cell line that stably over-expresses SOD2 in MIA PaCa-2 cells (Fig. 4A). Enzymaticactivity measurements confirmed that cells overexpressingSOD2 contained significantly (P < 0.001) higher SOD2activity than the control cells (Fig. 4B).Western blot analysisof PARP cleavage showed that overexpression of SOD2 ledMIA PaCa-2 cells to become resistant to 2-ME–mediatedapoptosis (Fig. 4C). Consistent with this, cell viability assayshowed that overexpression of SOD2 caused the parentalcells to be significantly less sensitive to the 2-ME–mediatedcell killing as compared with the control cells (Fig. 4D;compare "SOD2over" with control; P < 0.001). However,overexpression of SOD2 alone did not promote the 2-MEresistance of the MIA PaCa-2 cells to the level of theresistance of the MIA PaCa-2R cells (Fig. 4D; compare"SOD2over" with MIA PaCa-2R), suggesting that other

Figure 2. The resistant MIA PaCa-2R cells contain a high level of ROS. A,MIA PaCa-2 and MIA PaCa-2R cells were cultured with 2.5 mmol/L 2-MEfor 1week and then harvested for ROSmeasurement with flow cytometryusing a fluorescent probe DCFH-DA (see Materials and Methods). Tenthousand cells were evaluated and data were expressed as fold increaseof ROS generation over that in the parental cells. B, pretreatments of theMIAPaCa-2cellswith PEG-catalase or PEG-SOD inhibit DCF fluorescentsignal intensities (thusROS levels). Datawere expressedas fold decreaseof ROS generation over the untreated control. C, same as (B) except thatthe MIA PaCa-2R cells were tested. Values are the mean � SE of 3separate sample preparations. ��, ��, and � denote statisticalsignificance of P < 0.001, P < 0.01, and P < 0.05, respectively.

Figure 3. SOD2 is upregulated in the resistant MIA PaCa-2R cells. A,portions of the 2-DE gels showing the expression levels of SOD2 in theMIA PaCa-2 cells and the MIA PaCa-2R cells. B, Western blot analysisvalidationof SOD2expression in theMIAPaCa-2 cells and theMIAPaCa-2R cells. b-Actin was used as a loading control. Equal amounts of totalprotein were loaded in each lane (40 mg/lane). C, SOD2 enzymaticactivities in the MIA PaCa-2 cells and the MIA PaCa-2R cells. Values arethe mean � SE of 3 separate sample preparations. ��, statisticalsignificance of P < 0.01.

Zhou and Du





factors also contributed to the resistance of MIA PaCa-2Rcells to 2-ME. We also conducted clonogenic survival assayto compare long-term cell survival between the "SOD2over"cells and the parental cells. The results showed that over-expression of SOD2 enhanced the long-term clonogenicsurvival of the "SOD2over" cells under different concentra-tions of 2-ME compared with the parental cells (Fig. 4E).Taken together, results in this section suggest that theupregulated SOD2 in theMIA PaCa-2R cells plays a criticalrole in conferring the resistance of the MIA PaCa-2R cells tothe 2-ME–mediated cell killing and apoptosis.

Suppression of SOD2 expression moderately resensitizesMIA PaCa-2R cells to 2-METo ensure that SOD2 is indeed involved in the acquired

resistance to 2-ME, we selectively suppressed SOD2 expres-sion inMIA PaCa-2R cells with retrovirus-mediated shRNA(Fig. 5A). SOD2 enzymatic activity assay confirmed that thestable cells with SOD2being silenced contained significantlylower activity of SOD2 compared with the resistant cells(Fig. 5B). Western blot analysis showed that silencing of theSOD2 expression resensitized the MIA PaCA-2R cells to 2-ME–mediated apoptosis (Fig. 5C). In agreement with this,suppression of SOD2 in the MIA PaCa-2R cells led to amoderate increase in sensitivity to 2-ME–mediated cellkilling (Fig. 5D; compare "shSOD2" with MIA PaCa-2R;

P¼ 0.0712). However, the suppression of SOD2 expressiondid not completely reverse the 2-ME–resistant phenotype ofthe MIA PaCa-2R cells (Fig. 5D; compare "shSOD2" withMIA PaCa-2). These results were consistent with the over-expression data (Fig. 4), and both sets of results imply thatSOD2 plays an important role in the acquired resistance ofpancreatic cancer cells to 2-ME but is not the sole factorresponsible for the acquired resistance.

MIA PaCa-2R cells also show a tendency to have IRresistanceIn addition to mediating drug-induced apoptosis and cell

killing, ROS are also major components involved in IR-induced apoptosis and cell killing (10). To test whether thesensitivity of pancreatic cells to 2-ME is linked to theirsensitivity to IR, we first treated the MIA PaCa-2 and MIAPaCA-2R cells with 5 Gy of g-radiation and measured cellapoptosis and viability. The results showed the 2-ME–resistant MIA PaCa-2R cells were also more resistant toIR-induced apoptosis (Fig. 6A) and showed moderatelyhigher cell viability after the exposure (Fig. 6B; P ¼0.0975). We then conducted a clonogenic survival assay tocompare the long-term survival of the 2 types of cellsafter exposing the cells to different doses of IR. As shownin Fig. 6C, the resistant MIA PaCa-2R cells showed signif-icantly (P < 0.01) enhanced long-term clonogenic survival

Figure 4. Overexpression of SOD2in MIA PaCa-2 cells enhancesresistance to 2-ME–mediated cellkilling and apoptosis. A,Western blotanalysis of SOD2 expression invector-transformed cells (control)and the stable cells thatoverexpresses SOD2 (SOD2over). B,enzymatic activities of SOD2 in theparental and the SOD2over cells. C,the control and the SOD2over cellswere cultured with indicatedconcentrations of 2-ME for 48 hoursand then harvested for analysis ofPARP cleavage by Western blotting.In both A and C, equal amounts oftotal proteinwere loaded in each lane(40 mg/lane), and b-actin was used asa loading control. D, the control,SOD2over, and MIA PaCa-2R cellswere cultured with 2.5 mmol/L 2-MEfor 48 hours and then harvested forviability analysis. E, clonogenicsurvival of the parental cells and theSOD2over cells. The parental and theSOD2over cells were cultured inpresence of 2-ME at the indicatedconcentrations for 21 days, and thecolonieswithmore than 50 cells werecounted for calculation ofcell survival. Values in B, D, andE are the mean � SE of 3separate sample preparations.��, statistical significanceof P < 0.001.






over the parental cells after the cells were exposed to IR ofmore than 3 Gy. These results suggest that the acquired2-ME resistance of pancreatic cancer cells is not restricted toROS-inducing compound 2-ME, but also linked to radio-resistance of the cells. The observed radioresistance of theMIA PaCa-2R cells might be related to the increasedexpression of SOD2 in the resistant cells, as overexpressionof SOD2 has been shown to result in IR resistance indifferent types of cells (31).

The expression or activity of important proteins involvedin detoxifying H2O2 in mitochondria is not enhancedH2O2, the product of SOD2, is also toxic to cells.

Multiple factors are known to be involved in convertingH2O2 into water in mammalian cells, including catalase,GPx, and peroxiredoxin (32). Catalase, a major H2O2detoxifying enzyme in cells, is mainly located in peroxisomes(33) and has not been found in the mitochondria of mosttissues except for heart tissue (34, 35). Even in heartmitochondria, the contribution of catalase to H2O2 detox-ification is negligible (36). In mitochondria, GPx and amitochondrion-specific peroxiredoxin, Prx III, seem to playa major role in removing H2O2 (12, 36, 37). To examinewhether these enzymes are involved in the 2-ME resistanceof the MIA PaCa-2R cells, we compared the expression ofPrx III and the activity of GPx between the parental cells andthe resistant cells. The expression of Prx III was similarbetween the parental and the resistant cells (Fig. 7A). Incontrast to the changes in SOD2expression and activity (Fig.1), the catalytic activity of GPx in the resistant cells wassignificantly (P < 0.05) lower than that in the parental cells(Fig. 7B). A decrease in GPx catalytic activity could result inincreased sensitivity of cells to ROS (instead of enhanced

resistance to ROS). These results suggest that Prx III andGPx may not positively contribute to the increased 2-MEresistance of the MIA PaCa-2R cells.

DiscussionThrough exposure of pancreatic cancer MIA PaCa-2 cells

to increasing concentrations of ROS-inducing anticancercompound 2-ME over a 3-month period, we generated a cellline, MIA PaCa-2R, that was strongly resistant to 2-ME(Fig. 1). The ROS level in theMIA PaCa-2R cells was almost3-fold of that in the parental cells (Fig. 2). Comparativeproteomic analysis revealed that the expression of SOD2, akey antioxidant enzyme in mitochondria, was upregulated(Fig. 3). Functional analyses of SOD2 suggest that SOD2plays a critical role in the acquired resistance of the MIAPaCa-2R cells to 2-ME (Figs. 4 and 5) and potentially also toIR (Fig. 6).Compared with other major types of cancers, one unique

feature of pancreatic cancer is that approximately 90% ofpancreatic cancers contain Kras mutations (38), the highestin all cancers. Kras is involved in the G-protein signaltransduction pathway, modulating cellular proliferation anddifferentiation. Mutations of Kras result in constitutiveactivation of the MAPK and PI3K/AKT signal transductionpathways and, consequently, unregulated proliferation andimpaired differentiation (39), which lead to increased gen-eration of ROS (7). The elevated levels of ROS in turn canfurther activate MAPK and PI3K/AKT pathways (6, 13),potentially forming a positive feedback loop. The activatingmutations of Kras usually occur at a very early stage ofpancreatic cancer development, suggesting that Kras onco-genes play a key role in the pancreatic cancer pathogenesis(38). Tumor cells harboring activated Ras mutation are

Figure 5. Suppression of SOD2expression by shRNA moderatelyresensitizes MIA PaCa-2R cells to2-ME–mediated apoptosis and cellkilling. A, SOD2 expression in theMIA PaCa-2R cells and the cellswith SOD2 knockdown by shRNA(shSOD2). B, enzymatic activitiesof SOD2 in the resistant and theshSOD2cells.C,MIAPaCa-2Randthe shSOD2 cells were culturedwith 2.5 mmol/L 2-ME for 48 hoursand then harvested for Westernblot analysis with anti-PARPantibody. In both A and C, equalamounts of total protein wereloaded in each lane (40 mg/lane),and b-actin was used as a loadingcontrol. D, the indicated cells werecultured with 2.5 mmol/L 2-ME for48hours and thenharvested for cellviability assay. Values in B and Dare the mean� SE of 3 separatesample preparations. ��, statisticalsignificance of P < 0.01.

Zhou and Du





usually resistant to chemo- and radiotherapy (40). In linewith this notion, transformation of human ovarian epithelialcells with mutated Ras increases the threshold of ROS

tolerance (hence the resistance to the oxidative agent–medi-ated cell killing and apoptosis) by upregulating the overallantioxidant capacity of cells (41). Furthermore, a recentstudy showed that the IR resistance of cancer stem cells wasassociated with low levels of ROS resulting from enhancedROS defense in the cells (42). All these lines of evidencesuggest that ROS may play a critical role in the exceptionalintrinsic and acquired resistance of pancreatic cancer cells toIR and ROS-inducing chemodrugs. SOD2 plays a vital rolein ROS metabolism in that it converts highly toxic super-oxide (O2

*�) into less toxic hydrogen peroxides (H2O2) inthe mitochondria, a site that is vulnerable to ROS attack andimportant to cell apoptosis. SOD2-knockout mice diewithin 10 days after birth (43). Consistent with its essentialfunctions, enforced expression of SOD2 suppresses themalignant phenotypes of several types of human cancer cellsincluding pancreatic cancer, suggesting that SOD2 mayfunction as a tumor suppressor (19, 44, 45). However, inthe present study, we found that SOD2 expression wasupregulated instead of being downregulated in the resistantMIA PaCa-2R cells (Fig. 3). This contradiction suggests thatSOD2 must function as a tumor suppressor different frommost other conventional tumor suppressors, which areusually inhibited or lost in malignant cancer cells (39).Because the resistant MIA PaCa-2R cells contain high levelsof ROS and are thus under excessive oxidative stress, the cellsmust have evolved a molecular mechanism to defend againstthe excessive stress for survival (i.e., to evade the ROS-induced apoptosis). Upregulation of SOD2 expression must

Figure 6. The 2-ME–resistant MIA PaCa-2R cells are also moderatelyresistant to g-ray–mediated apoptosis and cell killing. A and B, the MIAPaCa-2 cells and the MIA PaCa-2R cells were treated with 5 Gy of g-ray,followed by a recovery culture at 37�C for 48 hours. The cells were thenharvested for Western blotting with anti-PARP antibodies (A) and cellviability assay (B). InWestern blotting, equal amounts of total proteinwereloaded in each lane (40 mg/lane), and b-actin was used as a loadingcontrol. C, clonogenic survival of the parental and the resistant cells. Theparental and the resistant cells were treated with the indicated doses ofg-ray and then cultured for 21 days. The colonies with more than 50 cellswere counted for calculation of cell survival. Values in B and C are themean � SE of 3 separate sample preparations. The P value in (B) was0.0975, and the P value at each dose >3 Gy in (C) was < 0.01.

Figure 7. The expression or activity of important proteins involved indetoxifying H2O2 in the mitochondria. A, the expression of Prx III in theparental and the resistant cells. Equal amounts of total protein wereloaded in each lane (40 mg/lane), and b-actin was used as a loadingcontrol. B, the activity ofGPx in theparental and the resistant cells. Valuesare the mean � SE of 3 separate sample preparations. �, statisticalsignificance of P < 0.05.






be one of those molecular mechanisms evolved in the MIAPaCa-2R cells. From this perspective, it is logical to speculatethat the increased expression of SOD2 in the MIA PaCa-2Rcells is an adaptive response of the cells to the excessiveoxidative cellular environment, which can be reflected by thehigher levels of ROS in the cells (Fig. 2). Because IR andmany anticancer drugs kill cancer cells largely throughinducing ROS (6, 9, 10), upregulation of SOD2 is likelyto be a common molecular mechanism underlying acquiredradio- or chemoresistance in cancer cells. This may beparticularly true for those types of cancers that harbor amutated Ras.It has been well established that 2-ME induces cellular

ROS accumulation (15, 18), and ROS triggers the expres-sion of SOD2 (29, 46). Concomitantly, ROS have alsobeen shown to activate transcription factor NF-kB, whichin turn controls the expression of SOD2 (47, 48). Thus,the increased expression of SOD2 in the MIA PaCa-2Rcould be a result of NF-kB activation induced by themarkedly elevated ROS levels in the resistant cells (Fig. 2).In addition to NF-kB, the promoter of SOD2 alsocontains binding site(s) for transcription factor AP-1(49). Furthermore, mutations or methylations of theSOD2 promoter sequence were also reported to affectthe expression of SOD2 in cancer cells (20, 50). Theseresults suggest that multiple factors, including NF-kB,AP-1, promoter mutations, and methylations, may func-tion independently or in combination to regulate theexpression of SOD2 in different tissues or under differentenvironments. Noticeably, SOD2 expression in primarypancreatic cancer cells (including MIA PaCa-2) isrepressed (19, 20). Further investigations on how theSOD2 expression is derepressed during the developmentof acquired drug resistance may provide valuable infor-mation on designing novel therapeutic strategies to resen-

sitize drug-resistant pancreatic cancer cells to radio- andchemotherapy.2-ME has been shown to trigger O2

*� generation in cellsvia the mitochondrial electron transport chain (18). Inter-estingly, in the present study, we found that the enzymaticactivity of GPx, a key enzyme in detoxifying H2O2 in themitochondria (12, 36), in the resistant cells was significantly(P < 0.05) decreased compared with the parental cells (Fig.7B). The decreased GPx could potentially contribute to theincreased ROS levels in the MIA PaCa-2R cells (Fig. 2).Future studies on examining the relationship between thereduced activity of GPx and the increased ROS level in theMIA PaCa-2R cells may provide new information on howcellular ROS is built up when cancer cells are exposed toROS-inducing compounds such as 2-ME.

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

Authors' ContributionsConception and design: J. Zhou, Y. DuDevelopment of methodology: J. Zhou, Y. DuAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): J. Zhou, Y. DuAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): J. Zhou, Y. DuWriting, review, and/or revision of the manuscript: Y. DuAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): Y. DuStudy supervision: Y. Du

Grant SupportThis work was supported by NIH grants 1R03CA135549-01A1 and

1R15GM087671-01, and a grant from the Arkansas Biosciences Institute (to Y. Du).The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received August 9, 2011; revised April 21, 2012; accepted April 23, 2012;published OnlineFirst April 30, 2012.

References1. Mangray S, King TC. Molecular pathobiology of pancreatic adenocar-

cinoma. Front Biosci 1998;3:D1148–60.2. Jessup JM, Steele G Jr, Mayer RJ, Posner M, Busse P, Cady B, et al.

Neoadjuvant therapy for unresectable pancreatic adenocarcinoma.Arch Surg 1993;128:559–64.

3. White R, Lee C, Anscher M, Gottfried M, Wolff R, Keogan M , et al.Preoperative chemoradiation for patients with locally advanced ade-nocarcinoma of the pancreas. Ann Surg Oncol 1999;6:38–45.

4. Jemal A, Clegg LX, Ward E, Ries LA, Wu X, Jamison PM, et al. Annualreport to the nation on the status of cancer, 1975–2001, with a specialfeature regarding survival. Cancer 2004;101:3–27.

5. Nicotera TM, Privalle C,Wang TC, OshimuraM, Barrett JC. Differentialproliferative responses of Syrian hamster embryo fibroblasts to para-quat-generated superoxide radicals depending on tumor suppressorgene function. Cancer Res 1994;54:3884–8.

6. Fruehauf JP, Meyskens FL Jr. Reactive oxygen species: a breath of lifeor death? Clin Cancer Res 2007;13:789–94.

7. Pelicano H, Carney D, Huang P. ROS stress in cancer cells andtherapeutic implications. Drug Resist Updat 2004;7:97–110.

8. Zhou Y, Hileman EO, Plunkett W, Keating MJ, Huang P. Free radicalstress in chronic lymphocytic leukemia cells and its role in cellularsensitivity to ROS-generating anticancer agents. Blood 2003;101:4098–104.

9. Benhar M, Engelberg D, Levitzki A. ROS, stress-activated kinases andstress signaling in cancer. EMBO Rep 2002;3:420–5.

10. Kong Q, Lillehei KO. Antioxidant inhibitors for cancer therapy. MedHypotheses 1998;51:405–9.

11. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281:1309–12.

12. Turrens JF. Mitochondrial formation of reactive oxygen species. JPhysiol 2003;552:335–44.

13. Gupta A, Rosenberger SF, Bowden GT. Increased ROS levels con-tribute to elevated transcription factor and MAP kinase activities inmalignantly progressedmouse keratinocyte cell lines. Carcinogenesis1999;20:2063–73.

14. Hileman EO, Liu J, Albitar M, Keating MJ, Huang P. Intrinsic oxidativestress in cancer cells: a biochemical basis for therapeutic selectivity.Cancer Chemother Pharmacol 2004;53:209–19.

15. Qanungo S, Basu A, Das M, Haldar S. 2-Methoxyestradiol inducesmitochondria dependent apoptotic signaling in pancreatic cancercells. Oncogene 2002;21:4149–57.

16. Schumacher G, Kataoka M, Roth JA, Mukhopadhyay T. Potent anti-tumor activity of 2-methoxyestradiol in human pancreatic cancer celllines. Clin Cancer Res 1999;5:493–9.

17. Seegers JC, LotteringML, Grobler CJ, van Papendorp DH, HabbersettRC, Shou Y, et al. The mammalian metabolite, 2-methoxyestradiol,

Zhou and Du





affects P53 levels and apoptosis induction in transformed cells but notin normal cells. J Steroid Biochem Mol Biol 1997;62:253–67.

18. Chauhan D, Li G, Sattler M, Podar K, Mitsiades C, Mitsiades N, et al.Superoxide-dependent and -independentmitochondrial signalingdur-ingapoptosis inmultiplemyelomacells.Oncogene2003;22:6296–300.

19. Cullen JJ, Weydert C, Hinkhouse MM, Ritchie J, Domann FE, Spitz D,et al. The role of manganese superoxide dismutase in the growth ofpancreatic adenocarcinoma. Cancer Res 2003;63:1297–303.

20. Hurt EM, Thomas SB, Peng B, Farrar WL. Molecular consequences ofSOD2 expression in epigenetically silenced pancreatic carcinoma celllines. Br J Cancer 2007;97:1116–23.

21. Gottlieb E, Vander Heiden MG, Thompson CB. Bcl-x(L) prevents theinitial decrease in mitochondrial membrane potential and subsequentreactive oxygen species production during tumor necrosis factoralpha-induced apoptosis. Mol Cell Biol 2000;20:5680–9.

22. Cai H, Li Z, Dikalov S, Holland SM, Hwang J, Jo H, et al. NAD(P)Hoxidase-derived hydrogen peroxide mediates endothelial nitric oxideproduction in response to angiotensin II. J Biol Chem 2002;277:48311–7.

23. Franken NA, Rodermond HM, Stap J, Haveman J, van Bree C.Clonogenic assay of cells in vitro. Nat Protoc 2006;1:2315–9.

24. AntonssonB,Montessuit S, SanchezB,Martinou JC.Bax is present asa high molecular weight oligomer/complex in the mitochondrial mem-brane of apoptotic cells. J Biol Chem 2001;276:11615–23.

25. Rezaul K, Wu L, Mayya V, Hwang SI, Han D. A systematic character-ization of mitochondrial proteome from human T leukemia cells. MolCell Proteomics 2005;4:169–81.

26. WangY, Zhou J, RuanC, DuY. Inhibition of type I interferon productionvia suppressing IKK-gamma expression: a new strategy for counter-acting host antiviral defense by influenza A viruses? J Proteome Res2012;11:217–23.

27. Zhou Z, Zhou J, Du Y. Estrogen receptor alpha interacts with mito-chondrial protein HADHB and affects beta-oxidation activity. Mol CellProteomics. 2012 Feb 27. [Epub ahead of print].

28. ParkSY,Chang I,KimJY,KangSW,ParkSH,SinghK, et al. Resistanceof mitochondrial DNA-depleted cells against cell death: role of mito-chondrial superoxide dismutase. J Biol Chem 2004;279:7512–20.

29. Hu Y, Rosen DG, Zhou Y, Feng L, Yang G, Liu J, et al. Mitochondrialmanganese-superoxide dismutase expression in ovarian cancer: rolein cell proliferation and response to oxidative stress. J Biol Chem2005;280:39485–92.

30. Malafa M, Margenthaler J, Webb B, Neitzel L, Christophersen M.MnSOD expression is increased in metastatic gastric cancer. J SurgRes 2000;88:130–4.

31. Sun J, Chen Y, Li M, Ge Z. Role of antioxidant enzymes on ionizingradiation resistance. Free Radic Biol Med 1998;24:586–93.

32. Andreyev AY, Kushnareva YE, Starkov AA. Mitochondrial metabolismof reactive oxygen species. Biochemistry (Mosc) 2005;70:200–14.

33. Chance B, Sies H, Boveris A. Hydroperoxide metabolism in mamma-lian organs. Physiol Rev 1979;59:527–605.

34. Phung CD, Ezieme JA, Turrens JF. Hydrogen peroxide metabolism inskeletal muscle mitochondria. Arch Biochem Biophys 1994;315:479–82.

35. Radi R, Turrens JF, Chang LY, Bush KM, Crapo JD, Freeman BA.Detection of catalase in rat heart mitochondria. J Biol Chem1991;266:22028–34.

36. Antunes F, Han D, Cadenas E. Relative contributions of heartmitochondria glutathione peroxidase and catalase to H(2)O(2)detoxification in in vivo conditions. Free Radic Biol Med 2002;33:1260–7.

37. Chang TS,ChoCS,Park S, YuS, KangSW,RheeSG. Peroxiredoxin III,a mitochondrion-specific peroxidase, regulates apoptotic signaling bymitochondria. J Biol Chem 2004;279:41975–84.

38. Motojima K, Tsunoda T, Kanematsu T, Nagata Y, Urano T, Shiku H.Distinguishing pancreatic carcinoma from other periampullary carci-nomas by analysis of mutations in the Kirsten-ras oncogene. Ann Surg1991;214:657–62.

39. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.

40. Brunner TB, Cengel KA, Hahn SM, Wu J, Fraker DL, McKenna WG,et al. Pancreatic cancer cell radiation survival and prenyltransferaseinhibition: the role of K-Ras. Cancer Res 2005;65:8433–41.

41. Young TW, Mei FC, Yang G, Thompson-Lanza JA, Liu J, Cheng X.Activation of antioxidant pathways in ras-mediated oncogenic trans-formation of human surface ovarian epithelial cells revealed by func-tional proteomics and mass spectrometry. Cancer Res 2004;64:4577–84.

42. Diehn M, Cho RW, Lobo NA, Kalisky T, Dorie MJ, Kulp AN, et al.Association of reactive oxygen species levels and radioresistance incancer stem cells. Nature 2009;458:780–3.

43. Li Y, Huang TT, Carlson EJ, Melov S, Ursell PC, Olson JL, et al. Dilatedcardiomyopathy and neonatal lethality in mutant mice lacking man-ganese superoxide dismutase. Nat Genet 1995;11:376–81.

44. Church SL, Grant JW, Ridnour LA, Oberley LW, Swanson PE, MeltzerPS, et al. Increased manganese superoxide dismutase expressionsuppresses the malignant phenotype of human melanoma cells. ProcNatl Acad Sci U S A 1993;90:3113–7.

45. Weydert C, Roling B, Liu J, Hinkhouse MM, Ritchie JM, Oberley LW,et al. Suppression of the malignant phenotype in human pancreaticcancer cells by the overexpression of manganese superoxide dismu-tase. Mol Cancer Ther 2003;2:361–9.

46. Chang LY, Kang BH, Slot JW, Vincent R, Crapo JD. Immunocyto-chemical localization of the sites of superoxide dismutase induction byhyperoxia in rat lungs. Lab Invest 1995;73:29–39.

47. Das KC, Lewis-Molock Y, White CW. Thiol modulation of TNF alphaand IL-1 inducedMnSODgene expression and activation of NF-kappaB. Mol Cell Biochem 1995;148:45–57.

48. Warner BB, Stuart L, Gebb S, Wispe JR. Redox regulation of manga-nese superoxide dismutase. Am J Physiol 1996;271:L150–8.

49. Wan XS, Devalaraja MN, St Clair DK. Molecular structure and orga-nization of the human manganese superoxide dismutase gene. DNACell Biol 1994;13:1127–36.

50. Xu Y, Krishnan A, Wan XS, Majima H, Yeh CC, Ludewig G, et al.Mutations in the promoter reveal a cause for the reduced expression ofthe human manganese superoxide dismutase gene in cancer cells.Oncogene 1999;18:93–102.






2012;10:768-777. Published OnlineFirst April 30, 2012.Mol Cancer Res Jianhong Zhou and Yuchun Du Manganese Superoxide Dismutase2-Methoxyestradiol Is Associated with the Upregulation of Acquisition of Resistance of Pancreatic Cancer Cells to

Updated version

10.1158/1541-7786.MCR-11-0378doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2012/04/30/1541-7786.MCR-11-0378.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/10/6/768.full#ref-list-1

This article cites 49 articles, 19 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/10/6/768.full#related-urls

This article has been cited by 3 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/10/6/768To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-11-0378

http://mcr.aacrjournals.org/content/suppl/2012/04/30/1541-7786.MCR-11-0378.DC1

http://mcr.aacrjournals.org/content/10/6/768.full#ref-list-1

http://mcr.aacrjournals.org/content/10/6/768.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/10/6/768


acquisition of resistance of pancreatic cancer …...(pi3k)/akt pathways (6, 13). 2-methoxyestradiol...

Documents