angptl4 induction by prostaglandin e ... - cancer...

Microenvironment and Immunology

ANGPTL4 Induction by Prostaglandin E2 under HypoxicConditions Promotes Colorectal Cancer Progression

Sun-Hee Kim1, Yun-Yong Park2, Sang-Wook Kim3, Ju-Seog Lee2, Dingzhi Wang1, and Raymond N. DuBois1,4

AbstractProstaglandin E2 (PGE2), the most abundant COX-2–derived prostaglandin found in colorectal cancer,

promotes tumor cell proliferation and survival via multiple signaling pathways. However, the role of PGE2 intumor hypoxia is not well understood. Here, we show a synergistic effect of PGE2 and hypoxia on enhancingangiopoietin-like protein 4 (ANGPTL4) expression and that elevation of ANGPTL4 promotes colorectal cancergrowth. PGE2 induces ANGPTL4 expression at both the mRNA and protein levels under hypoxic conditions.Moreover, hypoxia induces one of the PGE2 receptors, namely EP1. Activation of EP1 enhances ANGPTL4expression, whereas blockage of EP1 by an antagonist inhibits PGE2 induction of ANGPTL4 under hypoxicconditions. Importantly, overexpression of ANGPTL4 promotes cell proliferation and tumor growth in vitro and invivo. In addition, treatment with ANGPTL4 recombinant protein increases colorectal carcinoma cell proliferationthrough effects on STAT1 signaling. The MAP kinase and Src pathways mediate ANGPTL4-induced STAT1expression and activation. These results are relevant to human disease because we found that the expression ofANGPTL4 and STAT1 are elevated in 50% of human colorectal cancers tested and there is a positive correlationbetween COX-2 and ANGPTL4 as well STAT1 expression in colorectal carcinomas. Collectively, these findingssuggest that PGE2 plays an important role in promoting cancer cell proliferation via ANGPTL4 under hypoxicconditions. Cancer Res; 71(22); 1–11. �2011 AACR.

Introduction

Recent evidence confirms a reduction in risk for colorectaland other cancers in individuals who regularly take aspirin orother nonsteroidal anti-inflammatory drugs. These drugs arepotent inhibitors of COX-1 and COX-2. Elevation of COX-2 wasfound in approximately 50% of colorectal adenomas and 85% ofadenocarcinomas (1, 2) and is associated with worse survivalamong patients with colorectal cancer (3). COX-2 converts freearachidonic acid into prostanoids, including prostaglandins(PG) and thromboxanes (TX). Prostaglandin E2 (PGE2) is themost abundant prostaglandin found in colorectal cancer (4). Anincreasingly large body of evidence has shown that PGE2mediates the tumor-promoting effects of COX-2 in colorectalcancer (5). PGE2 treatment dramatically increased both small-and large-intestinal adenoma burden in ApcMin/þ mice andsignificantly enhanced azoxymethane–induced colon tumorincidence and multiplicity (6, 7). PGE2 promotes tumor growth

by stimulating EP receptor signaling with subsequent enhance-ment of cellular proliferation, promotion of angiogenesis, inhi-bition of apoptosis, and stimulation of invasion/motility (8).

Hypoxia is one of the most common and critical factorsidentified thus far in the regulation of cancer progression andmetastasis (9, 10). Adaptation of tumor cells to hypoxia isachieved largely by undergoing genetic changes that enablethem to survive and become more malignant or invasive (11).Hypoxia induces COX-2 expression and PGE2 levels via hyp-oxic-inducible factor (HIF)-1, a major transcription factor thatregulates the hypoxic response, in colorectal carcinoma cells(12, 13). Increased levels of PGE2 during hypoxia, in turn,potentiates HIF-1 transcriptional activity and promotes colo-rectal carcinoma cell growth and survival (12), suggesting thatcross-talk exists between PGE2 and hypoxia. However, themechanism through which PGE2 promotes colorectal cancerprogression in the hypoxic tumor microenvironment isunclear. We postulated that PGE2 might mediate a pivotaltumor cellular adaptive response to hypoxia with the impli-cation for connecting tumor hypoxia to cancer progression.

Angiopoietin-like protein 4 (ANGPTL4) is a secretaryprotein that is cleaved into an N-terminal coiled-coil fragment(N-ANGPTL4) and a C-terminal fibrinogen-like domain(C-ANGPTL4) that modulates the disposition of circulatingtriglycerides (14). Although the role of ANGPTL4 in lipidmetabolism has been well characterized, its role in cancerprogression is poorly understood. Although recent studiessuggest that ANGPTL4 is involved in cancer progression, theprecise role of ANGPTL4 in angiogenesis and cancer progres-sion is still debated. For example, ANGPTL4 has been reported

Authors' Affiliations:Departments of 1Cancer Biology, 2Systems Biology,and 4Gastrointestinal Medical Oncology, The University of Texas MDAnderson Cancer Center, Houston, Texas; and 3Department of InternalMedicine, Medical School, Chonbuk National University, Jeonju, Korea

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

Corresponding Author: Raymond N. DuBois, The University of Texas MDAnderson Cancer Center, Unit 1492, 1515 Holcombe Boulevard, Houston,TX 77030. Phone: 713-745-4495; Fax: 713-745-1812; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-11-1262

�2011 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org OF1

Research. on May 7, 2018. © 2011 American Association for Cancercancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 21, 2011; DOI: 10.1158/0008-5472.CAN-11-1262

http://cancerres.aacrjournals.org/

to have proangiogenic effects and antiangiogenic effects indifferent models (15–17). Moreover, ANGPTL4 mediates TGF-b–induced lung metastasis of breast cancer (18) but inhibitsmetastasis of melanoma cells (19). These conflicting datasupport the need of additional research to address the roleof ANGPTL4 in cancer progression.

In this study, ANGPTL4 is identified as a novel focal pointof the cross-talk between PGE2 and hypoxia signaling path-ways in colorectal cancer cells. Our results show that PGE2and hypoxia synergistically induce ANGPTL4 expression viaEP1. Importantly, we show that the elevation of ANGPTL4promotes tumor growth in vitro and in vivo. Moreover, weidentified STAT1 as a new target of ANGPTL4 and found thatinhibition of STAT1 attenuated ANGPTL4 induction ofcolorectal cancer cell proliferation. We further elucidatedthe ANGPTL4 signaling pathways in response to STAT1expression and activation. The maximal induction ofANGPTL4 required both hypoxia and PGE2, indicating thecomplex nature of interactions within the tumor microen-vironment that affect cancer signaling pathways.

Materials and Methods

ReagentsPGE2, 17-phenyl-trinor-PGE2, and SC-19220 were purchased

from Cayman Chemical Company. U73122, BAPTA-AM,PD98059, SU6656, and diphenylene iodonium (DP1) werepurchased from EMD Chemicals. ONO-AE3-208 was kindlyprovided by Ono Pharmaceutical Co. Full-length and C-termi-nal ANGPTL4 recombinant proteins were obtained from R&DSystems.

Cell culture and treatmentLS-174T, HT-29, and HCT-116 were purchased from the

American Type Culture Collection. All cells were routinelymaintained in McCoy's 5A medium containing 10% FBS, 100units/mL penicillin, and 100 mg/mL streptomycin in a 5% CO2

humidified incubator at 37�C. Cells were exposed to hypoxia byplacing them in amixed-gas incubator thatwas infusedwith anatmosphere consisting of 94% N2, 5% CO2, and 1% O2. Cellswere treated with PGE2 at the same time.

Quantitative real-time PCRTotal RNA was isolated by using TRIzol (Invitrogen).

cDNA was synthesized from 2 mg of total RNA with High-Capacity cDNA Reverse Transcription Kits (Applied Biosys-tems) and mixed with TaqMan Gene Expression Assay Mix,sterile water, and TaqMan Fast Universal PCR Master Mix(Applied Biosystems). Quantitative real-time PCR (qRT-PCR)was carried out using a 7900 HT Fast (Applied Biosystems)system, and expression of target genes mRNA relative to 18srRNA was calculated.

Measurement of ANGPTL4 secretionCells were exposed to PGE2 or hypoxic conditions, condi-

tioned medium was collected, and then the amount ofANGPTL4 secreted into medium was measured using HumanANGPTL4 DuoSet ELISA kit (R&D Systems, Inc.).

DNA constructionANGPTL4 cDNA (Origene) was amplified with pfuUltra

Hotstart DNA polymerase (Stratagene). The PCR productswere digested with XhoI and NotI and were cloned intopBMN-I-GFP (addgen).

Establishment of stable cell linepBMN-I-GFP and pBMN-I-GFP-ANGPTL4 retroviral vectors

were transfected into Phoenix cells in 60-mm dishes usingLipofectamine reagent (Invitrogen) according to the manu-facturer's protocol. Culture medium containing virus particleswas collected 48 hours later and was added to LS-174T cells.Cells infected were sorted by green fluorescent protein (GFP)positivity to eliminate uninfected cells.

Western blottingWhole cell lysates were prepared for Western blot analyses

using lysis buffer containing 20 mmol/L Tris-HCl, pH 7.4, 150mmol/L NaCl, 1% TX-100, 1 mmol/L EDTA, pH 8.0, and1 mmol/L phenylmethylsulfonylfluoride. Samples were dena-tured in a SDS sample buffer. Total proteins were separated byloading 20mg of total cell lysate on a denaturing 10% SDS-PAGEand transferred to a nitrocellulose membrane. Membraneswere blocked with 5% nonfat dry milk in PBS containing0.1% Triton X-100 and incubated with primary antibodies thatrecognize STAT1, Src, phospho-Src (Cell Signaling Technolo-gy), ANGPTL4 (R&D Systems), ERk, phospho-ERk (Santa CruzBiotechnology), and Actin (Sigma-Aldrich). Secondary anti-body conjugated to horseradish peroxidase (Vector Laborato-ries Inc.) was used at 1:2,000 to detect primary antibodies, andenzymatic signals were visualized by chemiluminescence.

Cell viability assayNinety-six–well plates were seeded with 5,000 cells per well

and cells were treated without or with ANGPTL4 in serum-freemedium for 3 days. Cell viability was determined using CellProliferation Reagent WST-1 (Roche Applied Science).

ImmunocytochemistryCells were treated with ANGPTL4, washed with PBS, and

fixed with 4% paraformaldehyde for 30 minutes at room tem-perature, and subsequent blocking with PBS containing 1%bovine serum albumin and 0.1% TX-100 for 30 minutes at roomtemperature. The cells were incubated with STAT1 antibody(Cell Signaling Technology) for 2 hours at room temperature,followed by biotinylated secondary antibody (Vector Labora-tories Inc.) for 1 hour, then Fluorescein streptavidin (VectorLaboratories Inc.) for 30 minutes, and then 40,6-diamidino-2-phenylindole (DAPI; Invitrogen) for 5 minutes to visualizenuclei. Cells were examined under a fluorescence microscope(Nikon ECLIPSE TE300) to determine localization of STAT1.

ImmunohistochemistryParaffin-embedded specimens were treated with xylene and

ethanol to remove the paraffin. The slides were immersed inBorg decloaker solution (Biocare Medical, Inc.) and boiled in apressure cooker at 125�C for 5 minutes for antigen retrieval.Endogenous peroxidase activity was blocked by incubation in

Kim et al.

Cancer Res; 71(22) November 15, 2011 Cancer ResearchOF2




3% H2O2 containing PBS solution for 10 minutes. The slideswere blocked with 5% normal goat serum and incubated withanti-STAT1 (Cell Signaling), HIF-1a (BD Biosciences), andANGPTL4 (Adipobioscience) at 4�C overnight. After washingwith PBS, the slides were incubated with Goat anti-Rabbit HRP(Vector Laboratories). Afterwashing, the slideswere developedwith 3,30-diaminobenzidine reagent (Vector Laboratories) fol-lowed by counterstaining with hematoxylin.

RNA interferencesiGENOME SMARTpool siRNAs targeting STAT1 (M-

003543-01-000) were purchased from Dharmacon, Inc. LS174Tcells were transfected with 20 nmol/L of STAT1 siRNA ornontargeting siRNA using Lipofectamine RNAiMAX (Invitro-gen) according to the manufacturer's specifications. The effi-cacy of knockdown was confirmed by Western blot analysis.

Xenograft studyAll mice were housed and treated in accordance with proto-

cols approved by the Institutional Animal Care and Use Com-mittee at The University of TexasMDAndersonCancer Center.LS-174T cells (5 � 105) selected for the stable expression ofANGPTL4 (ANGPTL4/LS-174T) or control GFP (GFP/LS-174T)were injected s.c. into the flanks of nude mice. The tumor sizewas measured starting from 13 to 28 days after injection. Afterthe mice were euthanized using CO2 asphyxiation, necropsieswere done to remove tumors andmeasurements were taken oftumor weight and size.

Gene expression data of colon cancer patientsHuman colorectal carcinoma specimens were obtained

from the Tissue Procurement and Banking Facility (TPBF) atThe University of Texas MD Anderson Cancer Center. Equalamounts of mRNA were analyzed by qRT-PCR for COX-2,ANGPTL4, and STAT1. Microarray data from Moffit CancerCenter (Moffit cohort, GSE17536, N ¼ 177) were downloadedfrom GEO (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE17536). Kaplan–Meier plots and log-rank test wereused to estimate patient prognosis.

Statistical analysisEach experiment was done at least 3 times, and data are

presented as the mean � SE. Statistical significance wasdetermined using Student t test, 1-factor ANOVA, or 2-factorANOVA, where applicable. In Figs. 1–4, � indicates P < 0.05;�� indicates P < 0.01; and �� indicates P < 0.001.

Results

PGE2 enhances ANGPTL4 expression in colorectalcarcinoma cells under hypoxic conditionsTo help mimic the conditions present in the tumor micro-

environment, we examined the role of PGE2 under normoxicand hypoxic conditions. We conducted transcriptome analysiswith RNA isolated from LS-174T cells exposed to PGE2 and/orhypoxia. LS-174T cells normally have low COX-2 levels and donot produce PGE2 under normoxic and hypoxic conditions(data not shown). Analysis of ourmicroarray data revealed that

ANGPTL4 was one of the PGE2 downstream targets in cellsexposed to hypoxia. Multiple lines of evidence have shown thathypoxia induces ANGPTL4 expression in several cell linesincluding monocytes and adipocytes (20–22). As expected,hypoxia increased ANGPTL4 expression at both mRNA andprotein levels in LS-174T cells (Fig. 1A and B). Strikingly, PGE2treatment dramatically enhanced ANGPTL4 expression underhypoxic conditions at both mRNA and protein levels in a dose-dependent manner in LS-174T cells, whereas PGE2 had noeffect on ANGPTL4 expression under normoxic conditions(Fig. 1A and B). The synergistic effect of PGE2 and hypoxiaon ANGPTL4 protein expressionwas also observed inHCT-116and HT-29 colorectal carcinoma cells (Fig. 1C). These resultsshow that PGE2 and hypoxia synergistically induce ANGPTL4expression in colorectal cancer cells.

EP1 mediates PGE2-stimulated ANGPTL4 expressionduring hypoxia

To explore why PGE2 induces ANGPTL4 expression onlyunder hypoxic conditions but not under normoxic conditions,we first examined the effects of hypoxia on PGE2 receptorlevels. There are 4 known PGE2 receptors (EP1, EP2, EP3, andEP4) that are typical G protein-coupled receptors. Becauserecent reports showed that hypoxia induced EP1 expression inosteoblastic cells (23),we examined the effect of hypoxia onEP1expression in colorectal cancer cells. Indeed, hypoxia specifi-cally increased only EP1 expression and not other PGE2 recep-tors in LS-174T cells (Fig. 2A). Because LS-174T cells expresshigh levels of EP4, we examined which receptor mediates theeffects of PGE2 on upregulation of ANGPTL4 expression. Treat-ment of LS-174T cellswith anEP1 antagonist, SC19220, blockedPGE2-induced ANGPTL4 induction under hypoxic conditions,whereas an EP4 antagonist, ONO-AE3-208, had no effect onANGPTL4 expression (Fig 2B). Consistent with these results, aselective EP1 agonist (17-PTPGE2) has the same effects as PGE2on ANGPTL4 expression (Fig. 2C). Because EP1 is coupled tothe Gq/PLC/Ca2þ pathway (24), we tested whether inhibitionof this pathway attenuates PGE2 induction of ANGPTL4.Treatment of LS-174T cells with a PLC inhibitor (U73122) anda Ca2þ chelating agent (BAPTA) partially suppressed PGE2-induced ANGPTL4 expression under hypoxic conditions,respectively (Fig. 2D). These results reveal that EP1 is inducedby hypoxia and mediates PGE2 induction of ANGPTL4.

ANGPTL4 promotes tumor growth in vitro and in vivoTo determine the biologic role of ANGPTL4 in regulating

cancer cell proliferation, LS-174T cells were treated with full-length recombinant ANGPTL4 (F-ANGPTL4) and C-ANGPTL4,respectively. Our results showed that C-ANGPTL4 moredramatically promoted LS-174T cell proliferation thanF-ANGPTL4 (Fig. 3A, top). In addition, C-ANGPTL4 exhibitedsimilar effects on regulating HCT-116 and HT-29 cell prolifer-ation (Fig. 3A, bottom). In contrast, both F-ANGPTL4 andC-ANGPTL4 had no effects on regulating LS-174T cell apopto-sis (data not shown). To confirm the proliferative effect ofANGPTL4, we established LS-174T cells that overexpressedANGPTL4 (ANGPTL4/LS-174T) and control cells (GFP/LS-174T). C-ANGPTL4 (37 kDa) was easily detected in the

PGE2 and Hypoxia Synergistically Induce ANGPTL4 Expression

www.aacrjournals.org Cancer Res; 71(22) November 15, 2011 OF3




conditioned medium of ANGPTL4/LS-174T cells (Fig. 3B, top).Because ANGPTL4 is proteolytically cleaved upon secretionfrom cells, F-ANGPTL4 was strongly detected in the cell lysatebut not the conditioned medium (data not shown). Consistentwith above results, overexpression of ANGPTL4 enhanced cellproliferation (Fig. 3B, bottom). These results are consistentwith our observation that PGE2 increases cell proliferationunder hypoxic conditions (Fig. 3C).

To validate our in vitro results in vivo, we injected ANGPTL4/LS-174T or GFP/LS-174T cells into the flanks of nude mice. Thetumor size was measured starting from 13 to 28 days afterinjection. Tumor weight and C-ANGPTL4 levels were also mea-sured after mice were euthanized. As expected, C-ANGPTL4levels were higher in the tumor tissue of mice injected with cellsexpressing ANGPTL4 than in the controls (Fig. 3D, left). More-over, cells expressing ANGPTL4 formed larger tumors than thecontrol cells (Fig. 3D,middleand right), suggesting that elevationof ANGPTL4 promotes colorectal cancer growth. Taken togeth-er, these results suggest that ANGPTL4 accelerates colorectalcancer growth by enhancing cell proliferation.

STAT1 is important for ANGPTL4-mediated cellproliferation

Although several recent reports have shown the role ofANGPTL4 in cancer progression (15, 25, 26), little is knownabout how ANGPTL4 regulates cell proliferation and whatdownstream targets of ANGPTL4 are involved in regulatingcell proliferation. The microarray analysis revealed thatC-ANGPTL4 treatment increased IFN-related gene expression,

such as STAT1, IFI35, IFI9, and IRF1. These geneswere validatedby qRT-PCR (data not shown). We further showed thatC-ANGPTL4 treatment induced STAT1 expression at both themRNA and protein levels as well as its nuclear localization(Fig. 4A, top and middle). Immunohistochemical staining fur-ther revealed that STAT1 expression was elevated in tumortissue derived frommice implantedwithANGPTL4/LS-174 cellsas compared with the GFP control (Fig. 4A, bottom). BecauseSTAT1 is overexpressed in head and neck and lung cancer andhas been reported to have a proliferative effect (27, 28), weexamined the contribution of STAT1 to C-ANGPTL4-inducedcell proliferation.Knockdownof STAT1 significantly suppressedC-ANGPTL4-induced cell proliferation (Fig. 4B). Because extra-cellular signal–regulated kinase (ERK) and Src are known to bethe critical signaling molecules for cell proliferation or survival(29), we determined whether these pathways are involved inC-ANGPTL4 induction of STAT1 expression. As shown inFig. 4C, PGE2 was able to enhance the phosphorylation of Srcand ERK aswell as STAT1 expression under hypoxic conditions.Moreover, C-ANGPTL4 treatment induced phosphorylation ofSrc and ERK in time-dependentmanners (Fig. 4D, top), whereaspharmacologic inhibitors of Src (SU6656) and ERK (PD98059)blocked STAT1 expression stimulated by C-ANGPTL4, respec-tively (Fig. 4D, middle). Recently, ANGPTL4 has been reportedto stimulate NADPH-oxidase–dependent production of O2

�,which in turn triggers the Src and ERK pathway (30). Therefore,we examined whether ANGPTL4 regulated Src and ERK acti-vation and STAT1 expression via upregulating O2

� production.Indeed, inhibition of O2

� production by an NADPH oxidase

Figure 1. PGE2 enhances ANGPTL4expression under hypoxicconditions. A and B, LS-174T cellswere treated with the indicatedconcentrations of PGE2 andexposed to normoxia/hypoxia for12 hours (mRNA) or 48 hours(protein) to determine theexpression of ANGPTL4. ANGPTL4mRNA (A) and protein (B) levelswere determined by qRT-PCR andELISA analysis, respectively. C,ELISA analysis was done in HT-29(left) and HCT-116 (right) cellsexposed to normoxia/hypoxia ordimethyl sulfoxide (DMSO)/1 mmol/L PGE2 for 48 hours. �, P < 0.05;��, P < 0.01; ��, P < 0.001.

Kim et al.





inhibitor (DP1) suppressed ANGPTL4 stimulated the phos-phorylation of Src and ERK and the expression of STAT1(Fig. 4D, bottom). Collectively, these results indicate thatSTAT1 is a downstream target of C-ANGPTL4 for regulatingcancer cell proliferation and that STAT1 induction is depen-dent on NADPH-oxidase-mediated production of O2

�, Src,and MAP kinase pathways.

ANGPTL4 and STAT1 are elevated in human colorectalcancer specimensBecause increased levels of COX-2 and a downstream

product, PGE2, have been reported in most human colorectalcancers (1) and hypoxia is one of the key elements in thetumor microenvironment that promotes cancer develop-ment and progression (9, 10), we examined ANGPTL4 levelsin human colorectal cancer and the correlation of ANGPTL4with COX-2 and STAT1 in human colorectal cancer.ANGPTL4 mRNA levels were elevated in 16 of 30 cancer

specimens as compared with those in adjacent normalmucosa (Fig. 5A). Moreover, ANGPTL4 expression in humancolorectal cancer specimens correlates with COX-2 expres-sion (r¼ 0.2433, P¼ 0.0056). In addition, high STAT1 expres-sion was found in 17 of 30 tumor tissues and is also correlatedwith ANGPTL4 expression (r ¼ 0.6861, P � 0.0001). Tosupport these findings, we used a public microarray databaseto retrieve gene expression data of colorectal cancer patients.Patients in the Moffitt cohort (N ¼ 177) were dichotomizedaccording to expression level of ANGPTL4. The overall sur-vival rate of patients with higher expression of ANGPTL4 wassignificantly worse than that of those with lower expressionof ANGPTL4 (Fig. 5B). In addition, the expression of COX-2positively correlated with the expression of ANGPTL4 (r ¼0.434, P � 0.0001) and STAT1 (r ¼ 0.232, P ¼ 0.001) in theMoffitt cohort (Fig. 5C). Therefore, these results suggest thatelevation of ANGPTL4 could be relevant to human colorectalcancer progression.

Figure 2. EP1 mediates PGE2-induced ANGPTL4 expressionduring hypoxia. A, LS-174T cellswere exposed to hypoxia for 12hours and qRT-PCR was done todetermine the expression of EP1,EP2, EP3, and EP4 receptors (left).LS-174T cells were exposed tohypoxia, and EP1 protein levelswere analyzed by Western blottingover times as indicated (right). Actinwas used as a protein loadingcontrol. B, LS-174T cells werepretreated with EP1 (10 mmol/L SC-19220) or EP4 (10 mmol/L ONO-AE3-208) antagonists for 1 hourand then were treated with PGE2

under normoxic or hypoxicconditions, following qRT-PCR andELISA to detect ANGPTL4 mRNAand protein. C, ANGPTL4 levelswere determined in the growthmedium with a commercial ELISAafter treatment with the indicatedconcentration of EP1 agonist(17-PT-PGE2) under normoxic orhypoxic conditions. D, cells werepretreated with PLC inhibitor(10 mmol/L U73122) or Ca2þ

chelator (1 mmol/L BAPTA-AM) for 1hour and then treated with PGE2

under normoxia/hypoxia for 48hours. ANGPTL4 levels wereanalyzed by ELISA. �, P < 0.05;��, P < 0.01; ��, P < 0.001.






Discussion

Hypoxia in the tumor microenvironment contributes tocancer progression and metastasis by activating adaptive pro-grams that promote cancer cell proliferation, survival, andmoti-lity as well as tumor-associated angiogenesis. It has beenreported that there is a cross-talk between the COX-2 pathwayand signaling pathways affected by hypoxia. For example,hypoxia induces COX-2 expression in colorectal cancer cellsand vascular endothelial cells (12, 31). Reciprocally, COX-2–derived PGE2 regulates HIF-1 expression and transcriptionalactivity in those cells (12, 32). Although a connection betweenCOX-2 and hypoxia has been established, the precise mechan-isms bywhich PGE2 and hypoxia synergistically promote cancerprogression are largely unclear. In this study, we identifyANGPTL4 as a novel PGE2 downstream target under hypoxicconditions in colorectal cancer cells. We also found that hyp-oxia–induced EP1 mediates the effects of PGE2 on regulation ofANGPTL4. Elevation of ANGPTL4 promotes tumor growthin vitro and in vivo, suggesting that ANGPTL4 can act as adownstream effector of PGE2 for adaptation of cancer cells tohypoxia. Our results suggest that the COX-2-PGE2 pathway plays

an important role in adapting cancer cells to the tumor micro-environment such as hypoxia that promotes cancer progression.

PGE2 exerts its biologic effects in an autocrine or paracrinemanner by binding to its cognate cell surface receptors, EP1,EP2, EP3, and EP4. All 4 EP receptors have been implicated inregulating cancer progression in certain contexts. These recep-tors are categorized into 3 groups. EP1 couples to Gq toincrease intracellular calcium, EP2 and EP4 couple to Gasto induce intracellular cyclic AMP, and EP3 couples to Gai todecrease cyclic AMP. EP receptors have distinct functionsbecause each receptor has a specific tissue distribution andinduces a specific signaling pathways (some redundancy mayexist; ref. 33). Here, we found that hypoxia specifically enhanc-ed EP1 expression in colorectal carcinoma cells without sig-nificant effects on EP2, EP3, or EP4 (Fig. 2A). EP1 has beenreported to promote cell survival and proliferation in differenttypes of cancer cells (34, 35). A role for EP1 in colorectaltumorigenesis has been further confirmed in animal models.For example, inhibition of EP1 signaling by genetic deletion ortreatment of an EP1-selective antagonist reduced intestinaltumor burden inApcMin/þ and carcinogen-treated EP1�/�mice(36), showing that EP1 contributes to cancer development. Our

Figure 3. ANGPTL4 increases tumorgrowth in vitro and in vivo. A, LS-174T cells were treated with full-length (F–A) or C-terminal fragment(C–A) of recombinant ANGPTL4protein for 3 days to determine cellproliferation (top). HCT-116 andHT-29 cells were exposed to 100ng/mL C-ANGPTL4 for 3 days andcell proliferation was analyzed(bottom). B, LS-174T cells werestably transfected with GFP orANGPTL4 constructs. C-ANGPTL4expression was determined byWestern blotting after the collectionof conditioned medium (top). Thesecells were incubated in serum-freemedium for 3 days and then cellproliferationwas tested (bottom). C,1 mmol/L of PGE2 was treated inLS-174T cells exposed tonormoxia/hypoxia for 3 days andcell proliferation assaywas done. D,LS-174T cell stably transfectedwithGFP or ANGPTL4 constructs wereinjected into the flanks of nudemice. The tumor size (middle) wasmeasured from 13 to 26 days afterinjection, and tumor weight (right)was also measured and C-ANGPTL4 level (left) was detectedafter mice were euthanized.�, P < 0.05; ��, P < 0.01;��, P < 0.001.

Kim et al.





results are supported by the observation that hypoxia specif-ically increases EP1 expression in bone cells. Hypoxia-inducedEP1may play a role in bone remodeling and repair in regions ofcompromised or damaged bone, where O2 tension is low (23).Our data showed that only hypoxia-induced EP1 mediated

PGE2 induction of ANGPTL4, although EP4 expression is highin both normoxic and hypoxic conditions (Fig. 2). This resultexplains why PGE2 only induced ANGPTL4 under hypoxicconditions but not under normoxic conditions. Moreover, ourresults from immunochemistry studies showed that ANGPTL4

Figure 4. ANGPTL4 enhances cancer cell growth through STAT1 A, LS-174T cells were exposed to C-ANGPTL4 for 12 hours (mRNA) and 24 hours (protein)and qRT-PCR and Western blot were done for STAT1 mRNA and protein levels, respectively (top). Cells treated with vehicle (CTL) or C-ANGPTL4 werestained with antibody against STAT1 (green) and counterstained with DAPI (blue). Immunofluorescence was detected by a fluorescence microscope (NikonECLIPSE TE300; �400; middle). Immunohistochemistry was done to detect STAT1 expression in tumor tissue generating by injection of ANGPTL4/LS-174cells into the flanks of nude mice compared with the GFP control (bottom). B, cells were transfected with nontargeting and STAT1 siRNA, respectively. After24 hours, transfecting cells were treated with C-ANGPTL4 for 3 days and subjected to Western blotting for STAT1 (top) and cell proliferation analysis(bottom). C, Western blot analysis was done in LS-174T cells exposed to normoxia/hypoxia or DMSO/1 mmol/L PGE2 for 24 hours to determine theexpression of STAT1 and the phosphorylation of ERK and Src. D, Western blot analysis was carried out to examine the phosphorylation of ERK and Src inLS-174T cells in the presence or absence of C-ANGPTL4 (top). The cells were pretreated with 1 mmol/L SU6656 (SU) or 10 mmol/L PD98059 (PD) for 1 hourand then were treated with C-ANGPTL4. STAT1 levels were determined (middle). Cells were treated with 2 mmol/L of diphenylene iodonium (DP1) andC-ANGPTL4 and then the phosphorylation of ERK and Src (0.5 h) as well as the expression of STAT1 (24 h) were determined (bottom). �, P < 0.05;��, P < 0.01.






is expressed in hypoxic regions of the tumors (SupplementaryFig. S1) although HIF-1a, an indicator of hypoxia, andANGPTL4 did not exactly colocalize in same cells but over-lapped in the same area because HIF-1a is a transcriptionfactor and ANGPTL4 is a ligand to be secreted and diffused inthe tumors. Collectively, these findings imply that tumorhypoxia enhances PGE2 signaling through upregulating EP1receptor expression, which may allow for adaptation of cancercells to the tumor microenvironment.

The synergistic effects of hypoxia and other factors such asinsulin, platelet-derived growth factor, and TNF-a targetinggenes has been reported in multiple types of cells, includingcancer cells (37, 38). In contrast to PGE2, these factors alsoinduce the expression of target genes under normoxic condi-tions. Our results show that PGE2 induced ANGPTL4 expres-sion only under hypoxic conditions. Several studies have

revealed that ANGPTL4 is transcriptionally regulated inresponse to intracellular or extracellular signals (18, 39).Because HIF mediates hypoxia upregulation of ANGPTL4 indifferent types of cells and PGE2 enhances HIF-1 transcrip-tional activity (12, 13, 20–22), it is possible that HIF-1 mediatesthe effect of PGE2 on induction of ANGPTL4. However, ourresults show that inhibition of HIF-1 did not affect PGE2induction of ANGPTL4 (data not shown). On the basis of theobservations that ANGPTL4 is one of the peroxisome pro-liferator-activated receptor (PPAR) target genes (40) withseveral putative peroxisome proliferator response consensussequences on its promoter (41) and PGE2 transactivates PPARdactivity (7), we postulate that PPARd is a potential transcrip-tion factor that could regulate PGE2-mediated ANGPTL4expression under hypoxic conditions. Further investigation isneeded to test this hypothesis.

Figure 5. COX-2, ANGPTL4, andSTAT1 expressions in colorectalcancers. A, total RNA was isolatedfrom 30 individual human colorectalcancer tissues and correspondingnormal mucosa. Equal amounts ofmRNA were analyzed by qRT-PCRfor COX-2, ANGPTL4, and STAT1.The cross line indicates a ratio equalto 1. B, colorectal cancer patients inthe Moffitt cohort (n ¼ 177) weredichotomized by expression ofANGPTL4 and patient with relativehigh expression or relative lowexpression of ANGPTL4 wereconsidered for plotting. DFS,disease-free survival. C, correlationof COX-2 and STAT1 or ANGPTL4expression in Moffitt colorectalcancer patient cohort. Scatter plotsbetween COX-2 and STAT1 orANGPTL4 in Moffitt cohort(N ¼ 177).

Kim et al.





The functions of ANGPTL4 have been well established inregulating lipid metabolism through the inhibition of lipopro-tein lipase (LPL). N-ANGPTL4, which assembles into multi-meric structures, is essential to inhibit the activity of LPL (39).However, C-ANGPTL4 exists as a monomer and its functionalrole inmonomeric form is still unclear. In addition to the role ofANGPTL4 in lipid metabolism, several recent studies havelinked F-ANGPTL4 to angiogenesis and cancer development(15, 25, 26). However, few studies have delineated the biologicfunction of C- or N-ANGPTL4 in cancer progression. Here, weshow that C-ANGPTL4 induces colorectal cancer cell prolif-erationmore effectively than F-ANGPTL4 (Fig. 3A). In addition,our results show for the first time that C-ANGPTL4 inducedcolorectal cancer proliferation through upregulation of STAT1.STAT1 has been reported to stimulate cell growth and isoverexpressed in head and neck and lung cancer (27, 28). Inaddition, activation of STAT1 promotes cell survival in Wilmstumor (27). Moreover, radiation induces STAT1 expressionthrough activation of epidermal growth factor receptor inbreast cancer cells (42). STAT1 has been reported to regulatethe oncogenic genes Fos and Egr1 in the absence of STAT3 (43).However, little information regarding themechanism bywhichSTAT1 regulates cell proliferation and oncogenesis is available.Because our microarray date showed that C-ANGPTL4increased several oncogenic genes, including Fos and Egr1(data not shown), the effect of C-ANGPTL4 on these oncogenicgenes via STAT1 may contribute to cancer cell proliferation.Moreover, we identified novel C-ANGPTL4 downstream tar-gets, Src and ERK, which are critical for C-ANGPTL4 inductionof STAT1 expression (Fig. 4). It has been well establishedpreviously that the Src and ERK are responsible for cellproliferation. Therefore, the role of ANGPTL4 in cancer devel-opment is likely conferred by C-ANGPTL4 through activatingthe signaling pathways which regulate cell proliferation.Although C- and F-ANGPTL4 have a fibrinogen-like domainof angiopoietin that binds to angiopoietin receptor Tie1 andTie2, C- and F-ANGPTL4 do not have any significant affinity forthose receptors (44). A recent report shows that C-ANGPTL4interacts with integrin b1 and b5 and produces O2

� to regulateanoikis resistance of cancer cells (30). Therefore, confirmationof the C-ANGPTL4 receptor such as integrins will add to ourunderstanding of how C-ANGPTL4 regulates cell proliferationas well as angiogenesis and metastasis.It is generally accepted that hypoxia in the tumor micro-

environment promotes chemoresistance (45). It has beenreported that COX-2 overexpressing cancer cells are moreresistant to doxorubicin and gefitinib (46, 47). Moreover,STAT1 has been shown to play a role in chemo- and radio-resistance. A recent study has shown that the IFN-related

gene signature including STAT1 and its target genes isassociated with resistance to DNA damage across multiplecancer cell lines and can affect resistance to chemotherapy(48). In addition, radioresistant tumor cells expressed higherSTAT1 levels than control cells (28). In accordance withthese published results, we found for the first time thatSTAT1 was upregulated by ANGPTL4, and our preliminarydata showed that ANGPTL4 treatment suppressed oxalipla-tin-induced cell death in LS-147T cells (data not shown).Furthermore, the expression of COX-2, ANGPTL4, andSTAT1 in human colorectal cancer specimens seems to becorrelated (Fig. 5). On the basis of these observations, it isconceivable that ANGPTL4 may contribute to the hypoxia orCOX-2–induced drug resistance and may offer a therapeuticoption for colorectal cancer patients.

In conclusion, our studies reveal the molecular mechanismby which PGE2 signaling and hypoxia coordinately promotetumor growth. We found that PGE2 enhanced ANGPTL4expression via EP1 receptor signaling that was enhanced byhypoxia. Moreover, we provide in vitro and in vivo evidenceshowing that elevation of ANGPTL4 promotes tumor growth.These results indicate that PGE2 plays an important role inpromoting a cellular adaptive response to hypoxia for theirgrowth via ANGPTL4. Knowledge about the role of PGE2 intumor hypoxia may enhance our understanding of how cancercells adapt to hypoxia for their growth. It has not escaped ourattention that other genes are also more dramatically inducedunder hypoxic conditions following PGE2 treatment. Thisgroup of genes may reveal other important connectionsbetween inflammatory mediators and cancer progression inthe tumor microenvironment.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Marivonne Rodriguez, Sung-Nam Cho, and Hong Wu for technicalassistance.

Grant Support

Thiswork was supported byNIHGrants RO1DK 62112 (R.N. DuBois), P01-CA-77839 (R.N. DuBois), and R37-DK47297 (R.N. DuBois).

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 April 14, 2011; revised August 9, 2011; accepted August 26, 2011;published OnlineFirst September 21, 2011.

References1. Gupta RA, Dubois RN. Colorectal cancer prevention and treat-

ment by inhibition of cyclooxygenase-2. Nat Rev Cancer 2001;1:11–21.

2. Marnett LJ, DuBois RN. COX-2: a target for colon cancer prevention.Annu Rev Pharmacol Toxicol 2002;42:55–80.

3. Ogino S, Kirkner GJ, Nosho K, Irahara N, Kure S, Shima K, et al.Cyclooxygenase-2 expression is an independent predictor of poorprognosis in colon cancer. Clin Cancer Res 2008;14:8221–7.

4. Wang D, Dubois RN. Cyclooxygenase-2: a potential target in breastcancer. Semin Oncol 2004;31:64–73.






5. Wang D, Buchanan FG, Wang H, Dey SK, DuBois RN. ProstaglandinE2 enhances intestinal adenoma growth via activation of the Ras-mitogen-activated protein kinase cascade. Cancer Res 2005;65:1822–9.

6. Kawamori T, Uchiya N, Sugimura T, Wakabayashi K. Enhancement ofcolon carcinogenesis by prostaglandin E2 administration. Carcino-genesis 2003;24:985–90.

7. Wang D, Wang H, Shi Q, Katkuri S, Walhi W, Desvergne B, et al.Prostaglandin E(2) promotes colorectal adenoma growth via transac-tivation of the nuclear peroxisome proliferator-activated receptordelta. Cancer Cell 2004;6:285–95.

8. Wang D, Dubois RN. Eicosanoids and cancer. Nat Rev Cancer2010;10:181–93.

9. Giatromanolaki A, Harris AL. Tumour hypoxia, hypoxia signaling path-ways and hypoxia inducible factor expression in human cancer.Anticancer Res 2001;21:4317–24.

10. Harris AL. Hypoxia–a key regulatory factor in tumour growth. Nat RevCancer 2002;2:38–47.

11. Lendahl U, Lee KL, Yang H, Poellinger L. Generating specificity anddiversity in the transcriptional response to hypoxia. Nat Rev Genet2009;10:821–32.

12. Kaidi A, Qualtrough D, Williams AC, Paraskeva C. Direct transcrip-tional up-regulation of cyclooxygenase-2 by hypoxia-inducible fac-tor (HIF)-1 promotes colorectal tumor cell survival and enhancesHIF-1 transcriptional activity during hypoxia. Cancer Res 2006;66:6683–91.

13. Greenhough A, Smartt HJ, Moore AE, Roberts HR, Williams AC,Paraskeva C, et al. The COX-2/PGE2 pathway: key roles in the hall-marks of cancer and adaptation to the tumour microenvironment.Carcinogenesis 2009;30:377–86.

14. Lee EC, Desai U, Gololobov G, Hong S, Feng X, Yu XC, et al. Iden-tification of a new functional domain in angiopoietin-like 3 (ANGPTL3)and angiopoietin-like 4 (ANGPTL4) involved in binding and inhibition oflipoprotein lipase (LPL). J Biol Chem 2009;284:13735–45.

15. MaT, JhamBC, Hu J, Friedman ER, Basile JR, Molinolo A, et al. Viral Gprotein-coupled receptor up-regulates Angiopoietin-like 4 promotingangiogenesis and vascular permeability in Kaposi's sarcoma. ProcNatl Acad Sci U S A 2010;107:14363–8.

16. Le Jan S, Amy C, Cazes A, Monnot C, Lamande N, Favier J, et al.Angiopoietin-like 4 is aproangiogenic factor producedduring ischemiaand in conventional renal cell carcinoma. Am J Pathol 2003;162:1521–8.

17. Ito Y, Oike Y, Yasunaga K, Hamada K, Miyata K, Matsumoto S, et al.Inhibition of angiogenesis and vascular leakiness by angiopoietin-related protein 4. Cancer Res 2003;63:6651–7.

18. Padua D, Zhang XH, Wang Q, Nadal C, Gerald WL, Gomis RR, et al.TGFbeta primes breast tumors for lung metastasis seeding throughangiopoietin-like 4. Cell 2008;133:66–77.

19. Galaup A, Cazes A, Le Jan S, Philippe J, Connault E, Le Coz E, et al.Angiopoietin-like 4 prevents metastasis through inhibition of vascularpermeability and tumor cell motility and invasiveness. Proc Natl AcadSci U S A 2006;103:18721–6.

20. Knowles HJ, Cleton-Jansen AM, Korsching E, Athanasou NA. Hyp-oxia-inducible factor regulates osteoclast-mediated bone resorption:role of angiopoietin-like 4. FASEB J 2010;24:4648–59.

21. Murata M, Yudo K, Nakamura H, Chiba J, Okamoto K, Suematsu N,et al. Hypoxia upregulates the expression of angiopoietin-like-4 inhuman articular chondrocytes: role of angiopoietin-like-4 in theexpression of matrix metalloproteinases and cartilage degradation.J Orthop Res 2009;27:50–7.

22. Wang B, Wood IS, Trayhurn P. Dysregulation of the expression andsecretion of inflammation-related adipokines by hypoxia in humanadipocytes. Pflugers Arch 2007;455:479–92.

23. Genetos DC, Lee CM, Wong A, Yellowley CE. HIF-1alpha regulateshypoxia-induced EP1 expression in osteoblastic cells. J Cell Biochem2009;107:233–9.

24. Kimura M, Osumi S, Ogihara M. Prostaglandin E(2) (EP(1)) receptoragonist-induced DNA synthesis and proliferation in primary cultures ofadult rat hepatocytes: the involvement of TGF-alpha. Endocrinology2001;142:4428–40.

25. Nakayama T, Hirakawa H, Shibata K, Abe K, Nagayasu T, Taguchi T.Expression of angiopoietin-like 4 in human gastric cancer: ANGPTL4promotes venous invasion. Oncol Rep 2010;24:599–606.

26. Wang Z, Han B, Zhang Z, Pan J, Xia H. Expression of angiopoietin-like 4 and tenascin C but not cathepsin C mRNA predicts prognosisof oral tongue squamous cell carcinoma. Biomarkers 2010;15:39–46.

27. Timofeeva OA, Plisov S, Evseev AA, Peng S, Jose-Kampfner M,LovvornHN, et al. Serine-phosphorylatedSTAT1 is aprosurvival factorin Wilms' tumor pathogenesis. Oncogene 2006;25:7555–64.

28. Khodarev NN, Beckett M, Labay E, Darga T, Roizman B, Weichsel-baum RR. STAT1 is overexpressed in tumors selected for radioresis-tance and confers protection from radiation in transduced sensitivecells. Proc Natl Acad Sci U S A 2004;101:1714–9.

29. Roskoski R Jr. RAF protein-serine/threonine kinases: structure andregulation. Biochem Biophys Res Commun 2010;399:313–7.

30. Zhu P, Tan MJ, Huang RL, Tan CK, Chong HC, Pal M, et al. Angio-poietin-like 4 protein elevates the prosurvival intracellular O2(-):H2O2ratio and confers anoikis resistance to tumors. Cancer Cell 2011;19:401–15.

31. Ji YS, Xu Q, Schmedtje JF Jr. Hypoxia induces high-mobility-groupprotein I(Y) and transcription of the cyclooxygenase-2 gene in humanvascular endothelium. Circ Res 1998;83:295–304.

32. Ji R, Chou CL, Xu W, Chen XB, Woodward DF, Regan JW. EP1prostanoid receptor coupling to G i/o up-regulates the expression ofhypoxia-inducible factor-1 alpha through activation of a phosphoino-sitide-3 kinase signaling pathway. Mol Pharmacol 2010;77:1025–36.

33. Cha YI, DuBois RN. NSAIDs and cancer prevention: targets down-stream of COX-2. Annu Rev Med 2007;58:239–52.

34. Zhang L, Jiang L, Sun Q, Peng T, Lou K, Liu N, et al. Prostaglandin E2enhances mitogen-activated protein kinase/Erk pathway in humancholangiocarcinoma cells: involvement of EP1 receptor, calcium andEGF receptors signaling. Mol Cell Biochem 2007;305:19–26.

35. Makita H,MutohM,MaruyamaT, Yonemoto K, Kobayashi A, FujitsukaH, et al. A prostaglandin E2 receptor subtype EP1-selective antago-nist, ONO-8711, suppresses 4-nitroquinoline 1-oxide-induced rattongue carcinogenesis. Carcinogenesis 2007;28:677–84.

36. Kawamori T, Kitamura T, Watanabe K, Uchiya N, Maruyama T, Nar-umiya S, et al. Prostaglandin E receptor subtype EP(1) deficiencyinhibits colon cancer development. Carcinogenesis 2005;26:353–7.

37. Li X, Kimura H, Hirota K, Kasuno K, Torii K, Okada T, et al. Synergisticeffect of hypoxia and TNF-alpha on production of PAI-1 in humanproximal renal tubular cells. Kidney Int 2005;68:569–83.

38. Meissner U, Ostreicher I, Allabauer I, RascherW, Dotsch J. Synergisticeffects of hypoxia and insulin are regulated by different transcriptionalelements of the human leptin promoter. Biochem Biophys Res Com-mun 2003;303:707–12.

39. Koliwad SK, Kuo T, Shipp LE, Gray NE, Backhed F, So AY, et al.Angiopoietin-like 4 (ANGPTL4, fasting-induced adipose factor) is adirect glucocorticoid receptor target and participates in glucocorti-coid-regulated triglyceride metabolism. J Biol Chem 2009;284:25593–601.

40. Rieck M, Meissner W, Ries S, Muller-Brusselbach S, Muller R. Ligand-mediated regulation of peroxisome proliferator-activated receptor(PPAR) beta/delta: a comparative analysis of PPAR-selective agonistsand all-trans retinoic acid. Mol Pharmacol 2008;74:1269–77.

41. KaddatzK,AdhikaryT, Finkernagel F,MeissnerW,Muller-BrusselbachS,Muller R. Transcriptional profiling identifies functional interactions ofTGF beta and PPAR beta/delta signaling: synergistic induction ofANGPTL4 transcription. J Biol Chem 2010;285:29469–79.

42. Amorino GP, Hamilton VM, Valerie K, Dent P, Lammering G, Schmidt-Ullrich RK. Epidermal growth factor receptor dependence of radiation-induced transcription factor activation in human breast carcinomacells. Mol Biol Cell 2002;13:2233–44.

43. Schiavone D, Avalle L, Dewilde S, Poli V. The immediate early genesFos and Egr1 become STAT1 transcriptional targets in the absence ofSTAT3. FEBS Lett 2011;585:2455–60.

44. OikeY, AkaoM,KubotaY, SudaT. Angiopoietin-like proteins: potentialnew targets for metabolic syndrome therapy. Trends Mol Med2005;11:473–9.

Kim et al.





45. Cosse JP, Michiels C. Tumour hypoxia affects the responsiveness ofcancer cells to chemotherapy and promotes cancer progression.Anticancer Agents Med Chem 2008;8:790–7.

46. Singh B, Cook KR, Vincent L, Hall CS, Berry JA, Multani AS, et al.Cyclooxygenase-2 induces genomic instability, BCL2 expres-sion, doxorubicin resistance, and altered cancer-initiating cellphenotype in MCF7 breast cancer cells. J Surg Res 2008;147:240–6.

47. Kim YM, Park SY, Pyo H. Cyclooxygenase-2 (COX-2) negativelyregulates expression of epidermal growth factor receptor and causesresistance to gefitinib in COX-2-overexpressing cancer cells. MolCancer Res 2009;7:1367–77.

48. Weichselbaum RR, Ishwaran H, Yoon T, Nuyten DS, Baker SW,Khodarev N, et al. An interferon-related gene signature for DNA dam-age resistance is a predictive marker for chemotherapy and radiationfor breast cancer. Proc Natl Acad Sci U S A 2008;105:18490–5.






Published OnlineFirst September 21, 2011.Cancer Res Sun-Hee Kim, Yun-Yong Park, Sang-Wook Kim, et al. Conditions Promotes Colorectal Cancer Progression

under Hypoxic2ANGPTL4 Induction by Prostaglandin E

Updated version

10.1158/0008-5472.CAN-11-1262doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2011/11/07/0008-5472.CAN-11-1262.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/2011/11/04/0008-5472.CAN-11-1262To 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-11-1262

http://cancerres.aacrjournals.org/content/suppl/2011/11/07/0008-5472.CAN-11-1262.DC1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/early/2011/11/04/0008-5472.CAN-11-1262


angptl4 induction by prostaglandin e ... - cancer...

Documents