Chemotherapy-Derived Inﬂammatory Responses Accelerate the ... · Microenvironment and Immunology Chemotherapy-Derived Inﬂammatory Responses Accelerate the Formation of Immunosuppressive

Microenvironment and Immunology

Chemotherapy-Derived Inflammatory ResponsesAccelerate the Formation of ImmunosuppressiveMyeloid Cells in the Tissue Microenvironment ofHuman Pancreatic CancerShintaro Takeuchi1,2, Muhammad Baghdadi2, Takahiro Tsuchikawa1, Haruka Wada2,Toru Nakamura1, Hirotake Abe1,2, Sayaka Nakanishi2, Yuu Usui2, Kohtaro Higuchi2,Mizuna Takahashi1, Kazuho Inoko1, Syoki Sato1, Hironobu Takano1, Toshiaki Shichinohe1,Ken-ichiro Seino2, and Satoshi Hirano1

Abstract

Pancreatic ductal adenocarcinoma (PDAC) is the most com-mon type of pancreatic malignancies. PDAC builds a tumormicroenvironment that plays critical roles in tumor progressionandmetastasis.However, the relationship between chemotherapyand modulation of PDAC-induced tumor microenvironmentremains poorly understood. In this study, we report a role ofchemotherapy-derived inflammatory response in the enrichmentof PDAC microenvironment with immunosuppressive myeloidcells. Granulocyte macrophage colony-stimulating factor (GM-CSF) is amajor cytokine associatedwithoncogenic KRAS in PDACcells. GM-CSF production was significantly enhanced in variousPDAC cell lines or PDAC tumor tissues from patients after

treatment with chemotherapy, which induced the differentiationof monocytes into myeloid-derived suppressor cells (MDSC).Furthermore, blockade of GM-CSF with monoclonal antibodieshelped to restore T-cell proliferationwhen coculturedwithmono-cytes stimulated with tumor supernatants. GM-CSF expressionwas also observed in primary tumors and correlated with poorprognosis in PDACpatients. Together, these results describe a roleof GM-CSF in the modification of chemotherapy-treated PDACmicroenvironment and suggest that the targeting of GM-CSFmaybenefit PDAC patients' refractory to current anticancer regimensby defeating MDSC-mediated immune escape. Cancer Res; 75(13);2629–40. �2015 AACR.

IntroductionPancreatic ductal adenocarcinoma (PDAC) is an aggressive

cancer characterized by highmortality and poor prognosis, wherein advanced cases the average of life expectancy is less than 1 year(1, 2). A recent study of cancer incidence and mortality hasprojected PDAC to become the second leading cause of cancer-related death by 2030 in the United States (3). In spite of recentprogress in treatment strategies, the current protocols of chemo-therapy regimens remain insufficient to cure the patients (4, 5).Recently, we and other groups have reported a new concept of"adjuvant surgery" in which PDAC patients are treated with pre-operative chemotherapy, followed by surgical resection, whichcontributes to long-term survival for locally advanced cases (6, 7).

Unfortunately, this procedure can be applied in only a smallpopulation of selected patients that were characterized with highoutcome of preoperative chemotherapy (6, 7). Thus, new thera-peutic strategies for improving chemotherapeutic response arecritically needed to improve the clinical outcomes in advancedPDAC, which in turn depend on the deep understanding ofchanges induced in tumor microenvironment under chemother-apeutic conditions. In this context, it has recently become clearthat anticancer chemotherapeutic agents can modify the tumormicroenvironment, and the therapeutic effects mediated by theseagents are considerably dependent on the host immunologicreaction (8, 9). In addition, the complex interaction betweentumor cells and other cellular components of tumor microenvi-ronment such as cancer-associated fibroblasts and myeloidcells has great impact on invasion, metastasis, and acquiring ofchemo-resistant phenotypes (10, 11). PDAC microenvironmentconstitutes molecular and cellular components with inflamma-tory features, such as pancreatic stellate cells and immune cells,which affect PDAC progress (12, 13). Accumulating evidence hasunveiled the role of KRAS oncogene in the formation of desmo-plastic and inflammatory microenvironment via the secretion ofmultiple cytokines and chemokines (14). Thus, the understand-ing of the interaction between tumor microenvironment andimmune cell and cytotoxic therapies is essential for the improve-ment of PDAC treatment.

Myeloid-derived suppressor cells (MDSC) are heterogeneouspopulations of immune cells derived from progenitor cells in

1Department of Gastroenterological Surgery II, Hokkaido UniversityGraduate School of Medicine, Sapporo, Japan. 2Division of Immuno-biology, Institute for Genetic Medicine, Hokkaido University, Sapporo,Japan.

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

Corresponding Authors: Ken-ichiro Seino, Division of Immunobiology, Institutefor Genetic Medicine, Hokkaido University, Kita-15, Nishi-7, Sapporo 060-0815,Japan. Phone: 811-1706-5532; Fax: 811-1706-7545; E-mail:[email protected]; and Satoshi Hirano, [email protected]

doi: 10.1158/0008-5472.CAN-14-2921

�2015 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2629

on April 1, 2020. © 2015 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst May 7, 2015; DOI: 10.1158/0008-5472.CAN-14-2921

http://cancerres.aacrjournals.org/

bone marrow, which accumulate in tumor microenvironmentvia various pathologic mechanisms, and contribute to tumorprogression by damping T-cell immunity and promoting angio-genesis (15, 16). Cytokines such as colony-stimulating factors(e.g., GM-CSF and granulocyte colony-stimulating factor) arekey molecules involved in the generation of MDSCs (17, 18).Oncogenic KRAS is the most frequently mutated gene in PDACand has been shown to be involved in PDAC development andgrowth (19, 20). Importantly, oncogenic KRAS is associatedwith overexpression of GM-CSF, which induces MDSC forma-tion in PDAC microenvironments, which in turn prompt thedevelopment and progression of PDAC in genetically engi-neered mouse models (21, 22). Moreover, targeted depletionof MDSCs was effective to increase the intratumoral accumu-lation of activated T cells and thus improved the therapeuticefficacies of immunotherapy in murine models of PDAC andother cancers (23). However, little is known about the role ofMDSCs in human PDAC, especially in clinical therapeuticsettings, for example, chemotherapy-treated conditions.

In the current study, we show phenotypic and functionalchanges of monocytes under chemotherapy-treated humanPDAC conditions. Human monocytes differentiated intoHLA-DRlow/negative MDSC phenotype when cultured in condi-tioned medium (CM) of human PDAC cells. Moreover, HLA-DRlow/negative cell formation was enhanced when human mono-cytes were cultured in CM of chemotherapy-treated humanPDAC cells. Gene and protein expression of GM-CSF or otherinflammatory factors in human PDAC cell lines were upregu-lated after treatment with anticancer cytotoxic agents such asgemcitabine andfluorouracil (5-FU). Blockade ofGM-CSF in thesupernatants of PDAC cell culture with specific monoclonalantibodies resulted in recovery of T-cell proliferation whencocultured with monocytes stimulated with PDAC CM. Consis-tent with these results, we found that PDAC tumor tissues inchemotherapy-treated cancer patients recruited more cells thatexpress MDSC markers compared with nontreated group.

In conclusion, targeting of PDAC with chemotherapy mayactivate inflammatory signals that induce the production ofmultiple sets of cytokines and chemokines in tumor cells. Amongthese, GM-CSF has emerged as a critical factor that link inflam-matory signals with the creation of immunosuppressive micro-environment via the acceleration of monocytes differentiationinto MDSCs. Together, our results give a new insight into howchemotherapymay results in counterproductive effects, and high-light the candidate molecules to be targeted in future improve-ment of PDAC treatment.

Materials and MethodsEthics

Human PDAC samples were obtained from surgical specimensafter obtaining informed consent fromall patients. Blood sampleswere obtained from healthy volunteers and PBMCs were sepa-ratedusing cell separating tube (BDBiosciences). Bothprocedureswere ethically approved by the committees in the InstitutionalReview Board of Hokkaido University Hospital (Sapporo, Japan;No. 013-0389, 013-0390).

Human PDAC tissue samplesFor tissue microarray (TMA), PDAC tissue samples were

obtained from 99 resected PDACs in our institute between

1994 and 2005. TMAs were constructed as described in ourprevious report (24). Patients without information about survivalor broken andpoor sampleswere omitted fromanalysis. A total of68 patients were subjected to analysis. The characteristics ofpatients for TMA study are summarized in Supplementary TableS1. Evaluation procedure was performed as previously reportedwith a little modification. The intensity of GM-CSF staining wasclassified according to a three-level scale: 0, weak or equivalentstaining compared with normal pancreas; 1þ, strong and partialstaining to cytoplasm of cancer cell; 2þ, strong and diffusestaining to cytoplasm. Scoring was evaluated by two independentinvestigators.

The 15 patients that underwent surgery in our institute andwere evaluated (Fig. 5) are overlap cohorts described in ourprevious report resected in our institute between 2006 and2010 (25). The characteristics of these patients are summarizedin Supplementary Table S2–S3. Immunohistochemical testingand evaluation of myeloid cells were performed according toprevious reports (25). Briefly, five areas of most abundant mye-loid cells distribution were selected in high-power field (�400).Average counted numbers of areas were compared. All specimenswere evaluated by two independent investigators.

Cell linesHuman PDAC cell lines (Capan-1, Capan-2, PANC-1, MIA-

PaCa-2, and BxPC-3), human cervical cancer cell line (HeLa), andhuman leukemia cell line (Jurkat) were purchased from ATCC.PK-45-P and PK-1 were purchased from RIKEN. PCI-43 and PCI-43-P5 were previously established from surgically resected pri-mary carcinoma tissues in our institute (26). All cell lines werecultured in an appropriate medium as indicated bymanufacturesor references. For CM used in monocyte culture, Capan-1 andPANC-1 cells were cultured in RPMI1640 (WAKO) supplementedwith 10% FBS (Cell Culture Bioscience), 1% penicillin/strepto-mycin, 10 mmol/L HEPES, 1% L-glutamine, 1 mmol/L sodiumpyruvate, 1% nonessential amino acids (all from Life Technolo-gies), and 50 mmol/L 2-mercaptoethanol (WAKO) in accordancewith optimizing conditions for monocytes.

In vitro human monocyte cultureTo examine the effects of PDAC-derived factors on monocyte

differentiation, we established the following in vitro models. Fornormal condition, the supernatants of PDAC cell culture wereharvested when cells became 80% confluent and passed through0.2 mm filter (Sartourius Stedim Biotech). To mimic clinicalpharmacologic settings in PDAC patients, gemcitabine (1–30mmol/L) or 5-FU (10 mg/mL) were applied at concentrationssimilar to that used in clinic (1–30 mmol/L). PDAC cells werepulsed with GEM or 5-FU for 60 minutes followed by washing 5times with sterilized PBS and changing to fresh media. After 72hours, supernatants were collected and passed through 0.2 mmfilter as described above. Human peripheral monocytes werepurified from PBMC of healthy donors using CD14þ selectionby magnetic cell sorting systems according to manufacturer'sprotocols (Miltenyi Biotec) and cultured in the presence of super-natants prepared from normal PDAC or chemotherapy-treatedPDAC cells for 6 days. On day 6, gene expression and proteinanalysiswere evaluatedbyquantitative RT-PCRorflowcytometry,respectively. In some experiments, cytokines in the supernatantsof PDAC cell culture were neutralized using anti-human GM-CSF(clone BVD2-23B6; Biolegend, 10mg/mL), anti-human IL6 (clone

Takeuchi et al.

Cancer Res; 75(13) July 1, 2015 Cancer Research2630




6708; R&D Systems, 2 mg/mL), or anti-human IL8 (clone 6217;R&D Systems, 2 mg/mL).

Flow cytometrySingle-cell suspensions were used for flow cytometry analysis

after treatment with human FcR blocker (Miltenyi Biotec) oranti-mouse CD16/32 (BD Biosciences) and stained with appro-priate fluorescent antibodies according to manufacturer'sinstructions. Fluorescent antibodies used for the staining ofhuman cell surface markers were purchased from BD Bio-sciences (anti-HLA-DR and anti-CD15), Beckman Coulter(anti-CD11b and anti-CD33), Miltenyi Biotec (anti-CD14), orBiolegend (anti-CCR2 and anti-CX3CR1). Fluorescent antibo-dies used for the staining of mouse cell surface markers werepurchased from Biolegend (anti-CD11b and anti-Gr1). Sam-ples were run on FACScanto II (BD Biosciences) and analyzedusing FlowJo software V7.6.5.

Quantitative RT-PCRRNA was extracted from cells using RNeasy Plus Mini Kit

(Qiagen) according to the manufacturer's protocol, and used forcDNA synthesis (Prime Script RT Master Mix, Takara Bio). cDNAproducts were used to amplify target genes using Power SYBRGreen (Life Technologies) and specific primer (SupplementaryTable S3). PCR reactions and data analysis were performed in aStepOne Real-Time PCR System (Applied Biosystems), using thecomparative Ct method and the housekeeping gene GAPDH.Primers used in this study are as follows:

GAPDH (forward: 50-AACAGCGACACCCACTCCTC-30;reverse: 50-ATACCAGGAAATGAGCTTGACAA-30), M-CSF (for-ward: 50-GCCTGCGTCCGAACTTCTA-30; reverse: 50-ACTGC-TAGGGATGGCTTTGG-30), GM-CSF (forward: 50-ATGATGGC-CAGCCACTACAA-30; reverse: 50-CTGGCTCCCAGCAGTCAAAG-30), IL6 (forward: 50-GGCACTGGCAGAAAACAACC-30; reverse:50-GCAAGTCTCCTCATTGAATCC-3), IL8 (forward: 50-CTGCG-CCAACACAGAAAATTA-30; reverse: 50-ATTGCATCTGGCAACC-TAC-30), IL1B (forward: 50-ATCACTGAACTGCACGTCC-30;reverse: 50-GCCCAAGGCCACAGGTATTT-30), PTCS2 (forward:50-GTTCCACCCGCAGTACAGAA-30; reverse: 50-AGGGCTTCAG-CATAAAGCGT-30), TNF (forward: 50-CACAGTGAAGTGCTGG-CAAC-30; reverse: 50-AGGAAGGCCTAAGGTCCACT-30), VEGF-A(forward: 50-CTACCTCCACCATGCCAAGT-30; reverse: 50-GCAG-TAGCTGCGCTGATAGA-30), CXCL-12 (forward: 50-CTACA-GATGCCCATGCCGAT-30; reverse: 50-CAGCCGGGCTACAATCT-GAA-30), SCF (forward: 50-AGCCAGCTCCCTTAGGAATGA-30;reverse: 50-TGCCCTTGTAAGACTTGGCTG-30), TGF-B1 (forward:50-GGGACTATCCACCTGCAAGA-30; reverse: 50-GAACCCGTT-GATGTCCACTT-30), CCL-2 (forward: 50-CAGCAAGTGTCCCA-AAGAAGCTG-30; reverse: 50-TGGAATCCTGAACCCACTTCTGC-30),NOS2 (forward: 50-TCCAAGGTATCCTGGAGCGA-30; reverse:50-AATGTGGGGCTGTTGGTGAA-30), ARG1 (forward: 50-ATGTT-GACGGACTGGACCCATCT-30; reverse: 50-TGCAACTGCTGTG-TTCACTGTTC-30), IL-10 (forward: 50-GAGATGCCTTCAGCA-GAGTGA-30; reverse: 50-ACATGCGCCTTGATGTCTGG-30). Pri-mers specificity was confirmed by peak melt curve before using.All experiments were performed in duplicate for each sample.

Cytokine measurementCytokines were measured using commercial ELISA kits accord-

ing to the manufacturer's instructions. The kits for GM-CSF andIL8were purchased fromBiolegend. The kit for IL6was purchased

from R&D Systems. All measurements were performed usingsupernatants from three independent cell cultures.

Western blottingTotal cell lysates were prepared using RIPA buffer supplemen-

tedwith protease inhibitors aprotinin andphenylmethylsulfonyl-fluoride. Protein sampleswere resolvedusing 10%SDS-PAGEandwere then transferred to polyvinylidene difluoridemembrane (GEHealthcare). Membranes were probed with primary antibodiesagainst target molecules followed by reaction with secondaryantibodies conjugated to horseradish peroxidase for appropriateincubation time. Antibodies against ERK, p-ERK, AKT, and p-AKTwere purchased from Cell Signaling Technology; antibodiesagainst b-actin were purchased from Millipore; secondary anti-bodies were purchased from Jackson ImmunoResearch. Immu-noreactivity was detected by an Enhanced ChemiluminescenceDetection System (GEHealthcare). Equal loading of proteins wasconfirmed with b-actin.

NF-kB luciferase reporter assayPromoter activities of NF-kB in cultured cells were monitored

using Ready-To-Glow secreted luciferase reporter system (Clon-tech). Briefly, Capan-1 cells were transfected with secreting lucif-erase reporter plasmid encodingNF-kBusing Lipofectamine 2000(Invitrogen), and stable clones were selected by G418. Stableclones were stimulated with GEM or 5-FU and luciferase activitiesin the supernatants were detected at the indicated time points.Luciferase activities were compensated by cell number.

T-cell proliferation assayAutologous reactions of monocytes and CD4þ or CD8þ T cells

were estimated by 3H-thymidine incorporation assay. Briefly,humanCD4þorCD8þT cellswere isolated fromPBMCof healthydonors using CD4þ T Cell Isolation Kit and CD8þ T Cell IsolationKit (Miltenyi Biotec). CD4þ or CD8þ T cells were cultured in thepresence of 3 mg/mL of anti-CD3 antibody (OKT3; eBioscience)and 1 mg/mL of anti-CD28 antibody (CD28.2; Biolegend). Stim-ulated CD4þ or CD8þ T cells were then cocultured with mono-cytes differentiated in the presence of tumor supernatants at theindicated conditions at different T-cell/monocyte ratios. 3H-thy-midine incorporations were counted after 72 hours culture.

Immunohistochemical staining of formalin-fixed paraffin-embedded tissues

Paraffin-embedded specimens were cut into thin slices andmounted on glass slide. Sections were deparaffinized in xyleneand rehydrated in ethanol. Antigen retrieval was performed byboiling for 20 minutes in citrate buffer (pH 6.0) or Tris-EDTAbuffer (pH9.0). Endogenous peroxidase activitywas blockedwith3% hydrogen peroxide in methanol. Nonspecific reactions wereblocked with original blocking cocktails; the equal quantity of10% normal goat serum (Nichirei), protein-block serum-freeready-to-use (Dako), and antibody diluent with backgroundreducing components (Dako). Immunohistochemical reactionswere carried out using the enzyme polymer methods with His-tofine series (Nichirei). Primary antibodies were mounted intoslides for 60 minutes at room temperature or overnight at 4�Cfollowed by 20 minutes incubation with secondary antibodies atroomtemperature. Antibodies used for FFPEwere purchased fromLSBio (GM-CSF: LS-C104671 clone), Abcam (CD14: ab49755clone, HLA-DR: EPR3692 clone), and Biolegend (CD66b:

GM-CSF and Modulation of PDAC Microenvironment

www.aacrjournals.org Cancer Res; 75(13) July 1, 2015 2631




G10F5), and used according to the manufacturer's instructions.The list of primary antibodies with their reactive conditions islisted in Supplementary Table S4. Immunohistochemical reac-tions were visualized with DAB or Fast Red II (Nichirei) followedby counterstaining with hematoxylin andmounted on coverslips.

Statistical analysisParametric statistics were applied for in vitro data and Student t

test was used for comparison between groups. For mouse orhuman data, nonparametric statistics were applied in whichMann–Whitney U test, Fisher exact test, or c2 test was used asappropriate. Overall survival was calculated from the date ofoperation to the date of last follow-up or date of patient death.The Kaplan–Meier method was used to estimate overall survival,and survival differences were estimated by the log-rank test.

Except where indicated, the values were presented as mean �SEM. P was considered statistically significant when <0.05. Alldata were analyzed using StatFlex software v6.0.

ResultsHuman monocytes differentiate into MDSCs when cultured inthe supernatants of PDAC cell culture

PDAC cells secrete multiple inflammatory cytokines andgrowth factors. To assess how PDAC cell–derived soluble factorsinfluence human myeloid cells differentiation, we generated invitro culture models using CM from two PDAC cell lines: Capan-1and PANC-1 (Fig. 1A). We found that human monocytes formeddifferentmorphologies in response to PDAC tumor supernatants.Monocytes differentiated into spindle adherent cells when

Figure 1.Supernatants of human PDAC cellculture induce the differentiation ofmonocytes into MDSCs. A, a schemeof culture protocol used to study theeffects of PDAC-derived factors onmonocytes differentiation. Humanperipheral CD14þ monocytes werepurified from healthy donor andcultured in PDAC CM for 6 days. B,representative photomicrographs ofmonocytes cultured for 6 days innormal medium, Capan-1 CM, orPANC-1 CM. Monocytes differentiateinto spindle macrophage-like cellswhen cultured in normal medium,whereas the supernatants of PDACcells inducemonocytes differentiationinto circular immature cells. Scalebars, 100 mm. C, flow cytometryanalysis of CD14, CD33, and HLA-DRexpression in monocytes cultured innormal medium (control), Capan-1CM, or PANC-1 CM. PDAC CM–treatedmonocytes were CD14þCD33þHLA-DRlow cells resembling mo-MDSC. D,HLA-DR expression levels in culturedmonocytes at day 6. HLA-DRexpressions were significantlydecreased when monocytes werecultured in PDACCM (n¼ 3 donors). E,flow cytometry analysis of NOS2 andARG1 in monocytes cultured in normalmedium (control), Capan-1, or PANC-1CM. Gray histogram, isotype; blackline, control medium; gray line, Capan-1 or PANC-1 CM. Capan-1 or PANC-1CM–treated monocytes show highlevels of NOS2 and ARG1 comparedwith control. F, flow cytometryanalysis of CD11b, CD15, CCR2 andCXCR1 expression in monocytescultured in Capan-1 or PANC-1 CM.Gray histogram, isotype; black line,Capan-1 or PANC-1 CM. PDACCM–treated monocytes showedexpression of CD11b andCCR2but lackthe expression of CD15 or CXCR1. Flowcytometry results are shown asrepresentative multiple independentexperiments. �, P < 0.05; �� , P < 0.01.

Takeuchi et al.





cultured in normal medium, while monocytes that were differ-entiated in the presence of Capan-1 or PANC-1 supernatantsformed floating immature cells (Fig. 1B). Previous reports sug-gested that PDAC induces the accumulation of MDSCs in tumorregions in genetically engineered mouse models (21, 22). Mono-cyte-derived MDSCs (mo-MDSCs) from cancer patients expressthe monocyte macrophage marker CD14 and the common mye-loid marker CD33, but lack or show lower expression of maturemyeloidmarkersHLA-DR (27).We found that humanmonocytesexpressed CD14 and CD33, while HLA-DR expression was rela-tively lower in monocytes cultured in the presence of PDACsupernatants compared with normal medium (Fig. 1C and D).Mo-MDSCs suppress T-cell immunity via nitric oxide synthase 2(NOS2) or Arginase 1 (ARG1; refs. 28, 29). Thus, we next eval-uated the expression levels of these two enzymes in monocytesinduced by PDAC CM. PDAC CM–treated monocytes showedhigh expression of both NOS2 and ARG1 (Fig. 1E). In addition,we examined the expression of other myeloid lineage markers,and found that PDAC CM–treated monocytes express the com-monmyeloid marker CD11b, chemokine receptor 2 (CCR2), butlack the expression of granulocyte or tissue resident macrophagemarker CD15 or CX3C chemokine receptor 1 (CX3CR1;

Fig. 1F; ref. 30). Together, these data demonstrated that humanperipheral monocytes differentiated into mo-MDSCs when stim-ulated with PDAC CM.

The supernatants of chemotherapy-treated PDAC cells enhancethe differentiation of human monocytes into MDSCs

Next, we examined whether the differentiation patterns ofmonocytes are altered in chemotherapy-treated PDAC microen-vironment. To do so, we established in vitro culture model usingCapan-1 cell line treated with gemcitabine or 5-FU (Fig. 2A).Interestingly, after 6 days of culture, monocytes showedmorpho-logic changes when cultured in the supernatants of chemother-apy-treated PDAC cells, represented by increased diameters (Fig.2B and Supplementary Fig. S1) and formation of cytoplasmicvacuoles that were not observed inmonocytes cultured in normalmedium or normal PDAC supernatant (Fig. 2C). These mono-cytes showed high forward and side scatter voltage signals in flowcytometry analysis, which was consistent with gross examination(Fig. 2D). In addition, the HLA-DRlow/negative fraction wasincreased in monocytes differentiated in the supernatants ofchemotherapy-treated PDAC cells (Fig. 2D and E and Supple-mentary Fig. S1). These changes are consistent with the phenotype

Figure 2.Supernatants of chemotherapy-treated PDAC cells inducemorphologic changes in monocyteswith enhanced MDSC markers. A, ascheme of culture protocol used tostudy the effects of chemotherapy-treated PDAC microenvironment onmonocytes differentiation. Capan-1cells were pulsed with gemcitabine(GEM; 1 mmol/L or 30 mmol/L) or 5-FU(10 mg/mL) for 1 hour, followed bycareful wash with sterilized PBS andchanged into fresh medium. CM wascollected after 72 hours and applied tohuman peripheral CD14þ monocytesas described above. B, morphologicchanges in monocytes cultured ingemcitabine-treated PDAC CM at day6. These cells were larger in size thanmonocytes cultured in PBS-treatedPDAC CM. Scale bars, 100 mm. C, MayGiemsa staining showed uniquecytoplasmic vacuoles in monocytescultured in gemcitabine (GEM)-treated PDACCM (red arrows) but notPBS-treated PDAC CM or normalmedium. Scale bars, 20 mm. D and E,flow cytometry analysis shows highforward and side scatter voltagesignals (top) and increasedfrequencies of HLA-DRlow/negative

fraction (bottom) in monocytescultured ingemcitabine-treated PDACCM compared with PBS-treated PDACCM. (n ¼ 3 donors). F, enhancedexpression of NOS2 in monocytescultured in the supernatants ofgemcitabine (GEM)-treated Capan-1cells. Data are shown asrepresentative of two independentexperiments. � , P < 0.05; ��, P < 0.01.






of HLA-DRlow/negative immature monocytes that have been previ-ously reported (27). To evaluate the immunosuppressive featuresofmonocytes differentiated in gemcitabine-treated PDACCM,weanalyzed expression levels of ARG1, IL-10, TGF-b1, and NOS2.Although no significant changes were observed in the expressionof ARG1, IL-10, or TGF-b1 (data not shown), NOS2 expressionwas significantly increased in monocytes differentiated in gemci-tabine-treated PDAC CM (Fig. 2F). MDSCs are usually character-ized by lack or low expression of HLA-DR and high expression ofNOS2 (28, 31). Accordingly, these data suggest that the super-natants of chemotherapy-treated PDAC cells accelerate the dif-ferentiation of monocytes into MDSCs with enhanced molecularpatterns.

Treatment with chemotherapy amplifies the expression of GM-CSF and other inflammatory cytokines in PDAC cells via theactivation of MAPK signaling pathway and NF-kB transcription

MDSCs are immunosuppressive myeloid cells that contributeto tumor progression and immune evasion. Accumulating evi-dence has unveiled that GM-CSF and other tumor-derived mole-cules are necessary for the induction of preferential expansion ofMDSCs in tumor microenvironment (32–34). To identify factorsin the supernatants of chemotherapy-treated PDAC cells respon-sible for monocytes differentiation into MDSCs, we investigatedexpression profiles of various cytokines and chemokines inCapan-1 or PANC-1 cell lines. Following stimulation with gem-citabine or 5-FU, several cytokines and chemokines were upre-gulated in both cell lines (Fig. 3A and B and Supplementary Fig.S2). In particular, the expression of GM-CSF, IL6, and IL8 wasincreased in the supernatants of chemotherapy-treated Capan-1cells (Fig. 3C and Supplementary Fig. S3). In the next experiment,we focused on GM-CSF since both cell lines showed a significantenhancement in GM-CSF production after treatment with gemci-tabine or 5-FU. In addition, GM-CSF is well known for its role asan essential factor of MDSC proliferation and differentiation inPDAC (22). In oncogenic KRAS-mediated PDAC murine model,GM-CSF is regulated by MAPK or PI3K signaling pathway, twomajor downstream pathways of KRAS oncogene (21). Thus, wenext compared the activation status of these two pathwaysthrough the evaluation of ERK phosphorylation as an indicatorfor MAPK pathway, or AKT for PI3K pathway in normal orchemotherapy-treated conditions. We found that gemcitabinetreatment enhances the phosphorylation of ERK (Fig. 3D) butnot AKT (data not shown) in a time-dependent manner. NF-kBis a major transcription factor that induces the expression ofinflammatory cytokines, including GM-CSF (35, 36). Thus, wenext examined whether gemcitabine treatment may induce pro-moter activities of NF-kB in PDAC cells. In a luciferase assay, wefound that NF-kB–luciferase activities were enhanced after che-motherapy treatment (Fig. 3E). These data indicate that chemo-therapy enhances the production of multiple inflammatory cyto-kines including GM-CSF by amplifying the activation status ofMAPK signaling pathway and NF-kB promoter activities in PDACcells.

Neutralization of GM-CSF in the supernatants ofchemotherapy-treated PDAC cells blocks monocytedifferentiation into MDSCs and helps recovery of T-cellproliferation

The supernatants of chemotherapy-treated PDAC cells wereenriched with GM-CSF, and induced morphologic and pheno-

typic changes in monocytes. To further examine the contribu-tion of GM-CSF in these changes, we utilized a specific mono-clonal antibody to neutralize GM-CSF in chemotherapy-treatedCapan-1 CM. Interestingly, we found that the neutralization ofGM-CSF has resulted in decreased forward and side scattervoltage signals as well as HLA-DRlow/negative fractions (Fig.4A), and abolished the formation of cytoplasmic vacuoles thatwere observed in the case of gemcitabine-treated Capan-1 CM(Fig. 4B). These data indicate that GM-CSF is one of the majorfactors of monocyte differentiation in the supernatants ofchemotherapy-treated PDAC cells.

MDSCs are heterogeneous populations of cells that are definedby their ability to potently suppress T-cell response by NOS2-dependentmechanism(31). As described above, the supernatantsof chemotherapy-treated PDAC cells were enrichedwith GM-CSF,and induced high expression of NOS2 in MDSCs differentiatedfrommonocytes. To confirm the immunosuppressive potential ofMDSCs generated from monocytes in the presence of PDACsupernatants, we cocultured these MDSCs with CD4þ or CD8þ

T cells and examined T-cell aggregation and proliferation afterstimulation. Interestingly, MDSCs generated from monocytes bynormal Capan-1 CM suppressed aggregation and proliferation ofstimulated CD4þ or CD8þ T cells, which was further suppressedbyMDSCs generatedbygemcitabine-treatedCapan-1CM(Fig. 4Cand D and Supplementary Fig. S4). Importantly, the neutraliza-tion of GM-CSF in gemcitabine-treated Capan-1 CMwas effectiveto abolish these immunosuppressive functions and contribute tothe recovery of T-cell function as observed by enhanced aggrega-tion and proliferation (Fig. 4C andD and Supplementary Fig. S4).Together, these data highlight the role of GM-CSF in the enhance-ment ofMDSCs formation in chemotherapy-treated PDACmicro-environment, and suggest that the neutralization of GM-CSFmaycontribute to block the formation ofMDSCs and thus the recoveryT-cell response.

GM-CSF is expressed in various human PDAC cell lines andtumor tissues and serves as a poor prognostic indicator forPDAC patients

To investigate whether GM-CSF expression is a commonfeature of PDAC cells, we examined the expression of GM-CSFin human samples. Quantitative PCR analysis showed highexpression of GM-CSF in all PDAC cell lines with some varia-tions (Fig. 5A). Next, immunohistochemistry staining was usedto examine protein levels of GM-CSF in PDAC tissues of 68resected primary tumors by TMA. PDAC tissues also showedvariety in GM-CSF expression (Fig. 5B). The intensity of GM-CSF staining was classified as high or low as described inMaterials and Methods (Fig. 5B and C), and scores were usedto generate Kaplan–Meier survival curve. We found that sur-vival rates were significantly lower in patients with high expres-sion of GM-CSF (Fig. 5D). These data suggest that GM-CSF, aMDSC-inducing cytokine, is generally expressed in humanPDAC, and correlates with poor prognosis.

Finally, to examine the impact of tumor microenvironment onMDSCs differentiation in human PDAC tissues under chemo-therapeutic conditions, we assessed MDSC marker expression intumor-infiltrating myeloid cells in PDAC patients treated withpreoperative chemotherapy including gemcitabine in our insti-tute (Supplementary Table S3). We found that tumor-infiltratingCD14þ cells in PDAC patients treated with preoperative chemo-therapy show no or weak expression of HLA-DR compared with

Takeuchi et al.





patients without preoperative chemotherapy treatment (Fig. 5Eand F). These data indicate that CD14þHLA-DR� cells constitute adominant fraction in PDAC tissues following chemotherapy, anobservation that might be a contrast to a previous mouse study(37). Furthermore, we investigated the expression of CD66b, amarker of granulocytic MDSC (G-MDSC; ref. 38), and found thatthe frequencies of tumor-infiltrating CD66bþ cells were signifi-cantly higher in PDAC patients after chemotherapy treatment(Fig. 5G and H). On the other hand, no significant difference wasobserved in the frequencies of CD68þ macrophages between thetwo groups (Supplementary Fig. S5). Taken together, these results

suggest that chemotherapy treatment accelerates the formation ofboth mo-MDSCs and G-MDSCs in human PDAC tissues, consis-tent with previous experiments.

DiscussionMost of PDAC cancer cases are diagnosed at late stages, which

make surgical resection of the tumor or the organ difficult if notimpossible (39). Chemotherapy has been suggested as a possiblestrategy for the treatment of PDAC patients; however, clinicalresponse mediated by anticancer cytotoxic agents against PDAC is

Figure 3.Chemotherapy treatment amplifiesthe expression of multiple MDSC-inducing cytokines in PDAC cells viaMAPK pathway-mediated signal. Aand B, quantitative RT-PCR analysisfor various cytokines and chemokinesin PBS or gemcitabine (GEM)-treated(A) or 5-FU–treated (B) Capan-1 cellsafter 72 hours of stimulation. Datafrom PBS-treated cells were set as 1.Data are shown as representative ofthree independent experiments. C,ELISA measurement of GM-CSF inthe supernatants of PBS orchemotherapy-treaded Capan-1 cellsafter 72 hours of stimulation. GM-CSFproduction is enhanced afterchemotherapy treatment in a dose-dependent manner. Data is shown asrepresentative of two independentexperiments. D,Western blotting of p-ERK or total ERK, p-AKT or total AKT,and b-actin of PBS or gemcitabine-treated Capan-1 cells stimulated forthe indicated time. Gemcitabineenhances the phosphorylation of ERKin a time-dependent manner. Similarresults were obtained from multipleindependent experiments. E, a timecourse of luciferase activity of Nfkbpromoter-luciferase reporter plasmidin Capan-1 cells stimulated withgemcitabine (top) or 5-FU (bottom).Data are shown as representative oftwo independent experiments.� , P < 0.05; �� , P < 0.01; �� , P < 0.001.






so limited, and it is unlikely that chemotherapy alone will providedurable clinical benefit for the majority of PDAC patients. Thus,new combination protocols are suggested to gain cumulative orsynergistic benefit in large population of patients. One goodexample is the treatment with radical surgery, which was accom-panied by favorable clinical outcomes in some clinical cases (6, 7).Moreover, recent progress has been achieved in the protocols of"neoadjuvant chemotherapy" against PDAC (40, 41). These newprotocols enable the analysis of molecular and pathologic patternsof chemotherapy-treated PDAC. For example, recent preoperativechemotherapy protocols helped to identify the molecular pat-terns of T cells, showing increased accumulation in tumor tissuesinPDACoroesophageal cancer patients (25, 42,43). In addition, inthis study we have reported for the first time the distribution ofMDSCmarkers in PDAC patients after chemotherapy treatment, in

whichMDSCs were the dominant cells in cancer regions. However,the real therapeutic effects of chemotherapy in PDAC treatmentstill poorly understood, since a large proportion of PDAC pati-ents develop chemoresistance and thus cannot receive surgicaltherapy. Therefore, further studies are critically needed to identifythe molecular mechanism of chemoresistance in PDAC.

It is now well established that the antitumor activities ofchemotherapy considerably rely on the complex interactionbetween tumor and immune systemof the host (9, 44).Moreover,accumulating evidence has unveiled the importance of the inter-action between tumor cells and myeloid cells in inducing che-moresistance andmetastasis (11, 45). This is also applicable in thecase of PDAC, and the deep understanding of this complexinteraction in tumor microenvironment is a key concept for theimprovement of chemotherapeutic response against PDAC. To

Figure 4.Blockade of GM-CSF contributes to the reversal of morphologic and phenotypic changes induced in monocytes by chemotherapy-treated PDAC CM. A, flowcytometry analysis shows decreased forward and side scatter voltage signals (top) and decreased frequencies of HLA-DRlow/negative fraction (bottom) inmonocytes cultured in gemcitabine-treated PDAC CM after depletion of GM-CSF (anti-GM-CSF, 10 mg/mL). B, microscopic examination and May Giemsa stainingshowed decrease in cell size (top) and disappearance of cytoplasmic vacuoles (bottom) that were observed in gemcitabine-treated PDAC CM after treatmentwith anti-GM-CSF. Scale bars, 100 mm for photomicrographs and 20 mm for May Giemsa staining. C, photomicrographs of T-cell aggregate. MDSCs werecocultured with autologous CD4þ T cells stimulated with anti-CD3/28 for 72 hours at the indicated ratio. Data are shown as representative of two independentexperiments. Scale bar, 10 mm. D, T-cell proliferation assay. MDSCs were cocultured with autologous CD4þ T cells stimulated with anti-CD3/28 for 72 hoursat the indicated ratio and T-cell proliferation was measured by H3-thymidine uptake. Neutralization of GM-CSF in gemcitabine-treated Capan-1 CM was effective toabolish the immunosuppressive functions and contribute to the recovery of CD4þ T-cell function as observed by enhanced aggregation and proliferation.Data are shown as representative of two independent experiments. � , P < 0.05; ��, P < 0.01; ��, P < 0.001.

Takeuchi et al.





understand how PDAC cells influence tumor microenvironmentin chemotherapy-treated condition, we first analyzed monocytedifferentiation patterns using in vitro culture models. When stim-ulatedwith the supernatants of chemotherapy-treated PDACcells,human monocytes differentiated into immunosuppressive cellsthat resemble MDSCs, showing similar morphology and sharedthe same molecular markers. Interestingly, the supernatants ofchemotherapy-treated PDAC cells were found to be enrichedwith GM-CSF and other inflammatory factors that induce thedifferentiation of monocytes into MDSCs. Consistent with this,

immunostaining of tumor tissues of PDAC patients treated withchemotherapy has shown enhancement in MDSC markers com-paredwithnormal tissues. Thus, chemotherapy itselfmay result incounterproductive effects in which the formation of immuno-suppressive and tumorigenic myeloid cells is enhanced at themicroenvironment of PDAC.

MDSCs are a heterogeneous population of immature mye-loid cells that negatively regulate the antitumor immuneresponses (15). MDSCs also support tumor immune evasionby suppressing T-cell immunity and promote angiogenesis and

Figure 5.GM-CSF expression is observed invarious PDAC cell lines and tumortissues of PDAC patients, and relatedto the enhancement of MDSC markersafter treatment with preoperativechemotherapy. A, quantitative RT-PCRanalysis of GM-CSF in various PDACcell and non-PDAC cell lines. GM-CSFexpression was normalized to GAPDH.Data are shown as representative ofthree independent experiments. B,immunohistochemical staining of GM-CSF in PDAC region or normal region ofpancreatic tissues from PDAC patients.Scale bar, 100 mm. C, the intensity ofGM-CSF staining was classifiedaccording to a three-level scale: 0, 1þ,2þ, and 71% of patients were GM-CSFhigh criteria. D, Kaplan–Meier survivalanalysis of overall survival in 68resected PDAC samples. GM-CSF–highpopulation showed significantly lowersurvival rates. E, immunohistochemicalstaining of CD14 and HLA-DR inpancreatic tissues of PDAC patientsbefore or after treatment withpreoperative chemotherapy. Scale bar,100 mm. F, frequencies of CD14þHLA-DRþ (left) and percentage of HLA-DRþ

cells to total CD14þ cells (middle) andtotal CD14þ (right) in pancreatictissues of PDAC patients before orafter preoperative chemotherapy.G, immunohistochemical staining ofCD66b in pancreatic tissues of PDACpatients before or after treatment withpreoperative chemotherapy. Scale bar,100 mm. H, frequencies of CD66bþ inpancreatic tissues of PDAC patientsbefore or after preoperativechemotherapy. For F and H, barsindicate the median value and the boxencompasses the 25th and 75thpercentiles. � , P < 0.05; �� , P < 0.01;�� , P < 0.001.






tumor progression (21, 22, 46). Accumulation of MDSCs hasbeen correlated with tumor progression in patients (39). Inaddition, a recent report has suggested that MDSCs contributeto senescence evasion and chemoresistance in tumor (11). InPDAC, MDSCs were found to be induced by MAPK or PI3Kpathway–dependent GM-CSF, and significantly correlated withtumor development and prognosis (21, 22). Importantly, wehave found that GM-CSF production was dramaticallyenhanced in several PDAC cell lines as well as tumor tissuesin PDAC patients after treatment with chemotherapy, whichwas accompanied by increased frequencies of MDSCs. Onepossible mechanism is the activation of MAPK and NF-kBsignaling pathway as a consequent of chemotherapy-inducedDNA-damage response (47). However, detailed mechanismshould be elucidated in future studies.

GM-CSF may play two different roles at the tumor micro-environment of PDAC. First, GM-CSF may help to induce oractivate anticancer immune responses through the priming ofimmunostimulatory dendritic cells. On the basis of this con-cept, GVAX, a GM-CSF gene-transferred tumor cell vaccine, hasbeen developed for the treatment of advanced PDAC patients,but the clinical outcome was lower than what was expected(48). Alternatively, GM-CSF may induce the formation ofMDSCs. One possible mechanism of these conflicting roles ofGM-CSF is the enrichment of PDAC microenvironment withdanger-associated molecular patterns (DAMP) after chemother-apy treatment. DAMPs are released from tumor cells killed byanticancer cytotoxic agents, and signaling mediated by theseDAMPs may be involved in the alteration of cellular differen-tiation pattern (49, 50), which should be clarified in futurestudies.

Our data indicate that MDSCs were increased after treatmentof PDAC with chemotherapy, which was related to enhance-ment in GM-CSF production. The neutralization of GM-CSFwith antibodies was effective to reduce MDSC frequencies, andhelp the recovery of T-cell function (Fig. 6). Depletion ofMDSCs has been recently suggested for PDAC treatment(23). In this context, the targeting of GM-CSF may constitutean additional option to further improve current protocols ofPDAC treatment.

In conclusion, our data identify a role of chemotherapy-derivedinflammatory response, in particular GM-CSF, in the enrichmentof PDAC microenvironment with MDSCs. Here, we suggest thatthe targeting of MDSCs by direct depletion and/or the neutrali-zation of tumor-derived GM-CSF in combination with currenttherapeutic regimens constitute a promising strategy for thetreatment of PDAC patients.

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

Authors' ContributionsConception and design: S. Takeuchi, M. Baghdadi, T. Tsuchikawa,T. Nakamura, K. Inoko, S. Sato, T. Shichinohe, K.-I. SeinoDevelopmentofmethodology: S. Takeuchi,M. Baghdadi, T.Nakamura,H.Abe,K. Inoko, S. SatoAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S. Takeuchi, M. Baghdadi, T. Nakamura, H. Abe,S. Nakanishi, Y. Usui, K. Higuchi, K. Inoko, S. Sato, K.-I. SeinoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Takeuchi, M. Baghdadi, T. Tsuchikawa, H. Wada,T. Nakamura, H. Abe, S. Nakanishi, Y. Usui, K. Higuchi, K. Inoko, S. Sato,S. Hirano, K.-I. SeinoWriting, review, and/or revision of the manuscript: S. Takeuchi, M. Baghdadi,T. Tsuchikawa, T. Nakamura, K. Inoko, S. Sato, S. Hirano, K.-I. SeinoAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S. Takeuchi, T. Nakamura, M. Takahashi,K. Inoko, S. Sato, H. TakanoStudy supervision: S. Takeuchi, T. Tsuchikawa, T. Nakamura, K. Inoko, S. Sato,T. Shichinohe, K.-I. Seino

AcknowledgmentsThe authors thank Hiraku Shida for technical assistance in making slides for

immunochemical staining and Tomohiro Shimizu and Katsuji Marukawa forkindly supporting in cytospin techniques. They also thank Dr. Masaki Miya-moto for the guidance and mentorship of this research.

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 indicatethis fact.

Received October 7, 2014; revised April 19, 2015; accepted April 22, 2015;published OnlineFirst May 7, 2015.

Figure 6.Mechanism of chemotherapy-mediated induction of MDSCs. Ascheme of mechanism by whichchemotherapy induces MDSCformation in PDAC microenvironmentis shown. Chemotherapy inducesactivation of MAPK signal pathwayand NF-kB promoter activities,leading to enhancement in GM-CSFproduction, which in turn enhance thedifferentiation of monocytes intoMDSCs. Anti-GM-CSF antibody mayoffer a promising tool to blockmonocytes differentiation intoMDSCs and thus help the recovery ofeffective antitumor T-cell response.

Takeuchi et al.





References1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin

2014;64:9–29.2. Shaib YH,Davila JA, El-SeragHB. The epidemiology of pancreatic cancer in

the United States: changes below the surface. Aliment Pharmacol Ther2006;24:87–94.

3. Rahib L, Smith BD, Aizenberg R, Rosenzweig AB, Fleshman JM, MatrisianLM. Projecting cancer incidence anddeaths to 2030: theunexpected burdenof thyroid, liver, and pancreas cancers in the United States. Cancer Res2014;74:2913–21.

4. Conroy T,Desseigne F, YchouM,Bouch�eO,GuimbaudR, B�ecouarn Y, et al.FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl JMed 2011;364:1817–25.

5. Paulson AS, TranCaoHS, TemperoMA, Lowy AM. Therapeutic advances inpancreatic cancer. Gastroenterology 2013;144:1316–26.

6. Kato K, Kondo S, Hirano S, Tanaka E, Shichinohe T, Tsuchikawa T, et al.Adjuvant surgical therapy for patientswith initially-unresectable pancreaticcancer with long-term favorable responses to chemotherapy. J Hepatobili-ary Pancreat Sci 2011;18:712–6.

7. Satoi S, Yamaue H, Kato K, Takahashi S, Hirono S, Takeda S, et al. Role ofadjuvant surgery for patients with initially unresectable pancreatic cancerwith a long-term favorable response to non-surgical anti-cancer treatments:results of a project study for pancreatic surgery by the Japanese Society ofHepato-Biliary-Pancreatic Surgery. J Hepatobiliary Pancreat Sci 2013;20:590–600.

8. Galluzzi LSenovilla L, Zitvogel L, Kroemer G. The secretally: immunosti-mulation by anticancer drugs. Nat Rev Drug Discov 2012;11:215–33.

9. Andre F, Dieci MV, Dubsky P, Sotiriou C, Curigliano G, Denkert C, et al.Molecular pathways: involvement of immune pathways in the therapeuticresponse and outcome in breast cancer. Clin Cancer Res 2013;19:28–33.

10. Quail DF, Joyce JA. Microenvironmental regulation of tumor progressionand metastasis. Nat Med 2013;19:1423–37.

11. Di Mitri D, Toso A, Chen JJ, Sarti M, Pinton S, Jost TR, et al. Tumour-infiltrating Gr-1þ myeloid cells antagonize senescence in cancer. Nature2014;515:134–7.

12. Apte MV, Wilson JS, Lugea A, Pandol SJ. A starring role for stellate cells inthe pancreatic cancer microenvironment. Gastroenterology 2013;144:1210–9.

13. Zheng L, Xue J, Jaffee EM,Habtezion A. Role of immune cells and immune-based therapies in pancreatitis and pancreatic ductal adenocarcinoma.Gastroenterology 2013;144:1230–40.

14. di Magliano MP, Logsdon CD. Roles for KRAS in pancreatic tumordevelopment and progression. Gastroenterology 2013;144:1220–9.

15. GabrilovichDI,Nagaraj S.Myeloid-derived suppressor cells as regulators ofthe immune system. Nat Rev Immunol 2009;9:162–74.

16. Corzo CA, Condamine T, Lu L, Cotter MJ, Youn J-I, Cheng P, et al.HIF-1a regulates function and differentiation of myeloid-derivedsuppressor cells in the tumor microenvironment. J Exp Med 2010;207:2439–53.

17. Serafini P, Carbley R, Noonan KA, Tan G, Bronte V, Borrello I. High-dosegranulocyte-macrophage colony-stimulating factor-producing vaccinesimpair the immune response through the recruitment of myeloid sup-pressor cells. Cancer Res 2004;64:6337–43.

18. Lesokhin AM, Hohl TM, Kitano S, Cortez C, Hirschhorn-Cymerman D,Avogadri F, et al. Monocytic CCR2þ myeloid-derived suppressor cellspromote immune escape by limiting activated CD8 T-cell infiltration intothe tumor microenvironment. Cancer Res 2012;72:876–86.

19. Iacobuzio-Donahue CA, Velculescu VE,Wolfgang CL, Hruban RH. Geneticbasis of pancreas cancer development and progression: insights fromwhole-exome and whole-genome sequencing. Clin Cancer Res 2012;18:4257–65.

20. Iacobuzio-Donahue CA. Genetic evolution of pancreatic cancer: lessonslearnt from the pancreatic cancer genome sequencing project. Gut 2012;61:1085–94.

21. Pylayeva-Gupta Y, Lee Kyoung E, Hajdu Cristina H, Miller G, Bar-Sagi D.Oncogenic Kras-induced GM-CSF production promotes the developmentof pancreatic neoplasia. Cancer Cell 2012;21:836–47.

22. Bayne Lauren J, BeattyGregory L, JhalaN,ClarkCarolyn E, RhimAndrewD,Stanger Ben Z, et al. Tumor-derived granulocyte-macrophage colony-stim-ulating factor regulates myeloid inflammation and T cell immunity inpancreatic cancer. Cancer Cell 2012;21:822–35.

23. Stromnes IM, Brockenbrough JS, Izeradjene K, Carlson MA, Cuevas C,Simmons RM, et al. Targeted depletion of an MDSC subset unmaskspancreatic ductal adenocarcinoma to adaptive immunity. Gut 2014;63:1769–81.

24. Tanaka K, Tsuchikawa T, Miyamoto M, Maki T, IchinokawaM, Kubota KC,et al. Down-regulation of Human Leukocyte Antigen class I heavy chain intumors is associated with a poor prognosis in advanced esophageal cancerpatients. Int J Oncol 2012;40:965–74.

25. Tsuchikawa T, Hirano S, Tanaka E,Matsumoto J, Kato K, Nakamura T, et al.Novel aspects of preoperative chemo(radiation)therapy improving anti-tumor immunity in pancreatic cancer. Cancer Sci 2013;104:531–5.

26. Shichinohe T, SenmaruN, Furuuchi K, Ogiso Y, Ishikura H, Yoshiki T, et al.Suppression of pancreatic cancer by the dominant NegativerasMutant,N116Y. J Surg Res 1996;66:125–30.

27. Almand B, Clark JI, Nikitina E, van Beynen J, English NR, Knight SC, et al.Increased production of immature myeloid cells in cancer patients: amechanism of immunosuppression in cancer. J Immunol 2001;166:678–89.

28. Youn J-I, Nagaraj S, CollazoM,GabrilovichDI. Subsets ofmyeloid-derivedsuppressor cells in tumor-bearing mice. J Immunol 2008;181:5791–802.

29. Virtuoso LP, Harden JL, Sotomayor P, SigurdsonWJ, Yoshimura F, EgilmezNK, et al. Characterization of iNOS(þ) Neutrophil-like ring cell in tumor-bearing mice. J Transl Med 2012;10:152.

30. Hart KM, Usherwood EJ, Berwin BL. CX3CR1 delineates temporallyand functionally distinct subsets of myeloid-derived suppressor cells ina mouse model of ovarian cancer. Immunol Cell Biol 2014;92:499–508.

31. Mazzoni A, Bronte V, Visintin A, Spitzer JH, Apolloni E, Serafini P, et al.Myeloid suppressor lines inhibit T cell responses by an NO-dependentmechanism. J Immunol 2002;168:689–95.

32. Dolcetti L, Peranzoni E,Ugel S,Marigo I, FernandezGomezA,MesaC, et al.Hierarchy of immunosuppressive strength among myeloid-derived sup-pressor cell subsets is determined by GM-CSF. Eur J Immunol 2010;40:22–35.

33. Sawanobori Y, Ueha S, Kurachi M, Shimaoka T, Talmadge JE, Abe J, et al.Chemokine-mediated rapid turnover of myeloid-derived suppressor cellsin tumor-bearing mice. Blood 2008;111:5457–66.

34. Song X, Krelin Y, Dvorkin T, Bjorkdahl O, Segal S, Dinarello CA, et al.CD11bþ/Gr-1þ immature myeloid cells mediate suppression of T cellsin mice bearing tumors of IL-1b-secreting cells. J Immunol 2005;175:8200–8.

35. Schreck R, Baeuerle PA.NF-kappaB as inducible transcriptional activator ofthe granulocyte-macrophage colony-stimulating factor gene. Mol Cell Biol1990;10:1281–6.

36. Thomas RS, Tymms MJ, McKinlay LH, Shannon MF, Seth A, Kola I. ETS1,NFkappaB and AP1 synergistically transactivate the human GM-CSF pro-moter. Oncogene 1997;14:2845–55.

37. Suzuki E, Kapoor V, Jassar AS, Kaiser LR, Albelda SM. Gemcitabineselectively eliminates splenic Gr-1þ/CD11bþmyeloid suppressor cells intumor-bearing animals and enhances antitumor immune activity. ClinCancer Res 2005;11:6713–21.

38. Dumitru CA, Moses K, Trellakis S, Lang S, Brandau S. Neutrophils andgranulocytic myeloid-derived suppressor cells: immunophenotyping, cellbiology and clinical relevance in human oncology. Cancer ImmunolImmunother 2012;61:1155–67.

39. Ryan DP, Hong TS, Bardeesy N. Pancreatic adenocarcinoma. N Engl J Med2014;371:1039–49.

40. White RR, Evans DB. Neoadjuvant chemotherapy for localized pancreaticcancer: too little or too long? Ann Surg Oncol 2014;21:1508–9.

41. Boeck S, Haas M, Ormanns S, Kruger S, Siveke JT, Heinemann V. Neoad-juvant chemotherapy in pancreatic cancer: innovative, but still difficult. Br JCancer 2014;111:1675–6 .

42. TsuchikawaT,MdMM,YamamuraY, Shichinohe T,Hirano S, KondoS. Theimmunological impact of neoadjuvant chemotherapy on the tumormicro-environment of esophageal squamous cell carcinoma. Ann Surg Oncol2012;19:1713–9.

43. Homma Y, Taniguchi K, Murakami T, Nakagawa K, Nakazawa M, Mat-suyama R, et al. Immunological impact of neoadjuvant chemoradiother-apy in patients with borderline resectable pancreatic ductal adenocarci-noma. Ann Surg Oncol 2014;21:670–6.






44. Halama N, Michel S, Kloor M, Zoernig I, Benner A, Spille A, et al.Localization and density of immune cells in the invasive margin of humancolorectal cancer liver metastases are prognostic for response to chemo-therapy. Cancer Res 2011;71:5670–7.

45. Qian B-Z, Pollard JW. Macrophage diversity enhances tumor progressionand metastasis. Cell 2010;141:39–51.

46. Lyden D, Hattori K, Dias S, Costa C, Blaikie P, Butros L, et al. Impairedrecruitment of bone-marrow-derived endothelial and hematopoietic precur-sor cells blocks tumor angiogenesis and growth. NatMed 2001;7:1194–201.

47. Gilbert LA, Hemann MT. DNA damage-mediated induction of a chemore-sistant niche. Cell 2010;143:355–66.

48. Middleton G, Silcocks P, Cox T, Valle J, Wadsley J, Propper D, et al.Gemcitabine and capecitabine with or without telomerase peptide vaccineGV1001 in patients with locally advanced or metastatic pancreatic cancer(TeloVac): an open-label, randomised, phase 3 trial. Lancet Oncol 2014;15:829–40.

49. Vernon PJ, Loux TJ, Schapiro NE, Kang R, Muthuswamy R, Kalinski P, et al.The receptor for advanced glycation end products promotes pancreaticcarcinogenesis and accumulation of myeloid-derived suppressor cells.J Immunol 2013;190:1372–9.

50. Krysko O, Love Aaes T, Bachert C, Vandenabeele P, Krysko DV. Many facesof DAMPs in cancer therapy. Cell Death Dis 2013;4:e631.


Takeuchi et al.




2015;75:2629-2640. Published OnlineFirst May 7, 2015.Cancer Res Shintaro Takeuchi, Muhammad Baghdadi, Takahiro Tsuchikawa, et al. Microenvironment of Human Pancreatic CancerFormation of Immunosuppressive Myeloid Cells in the Tissue Chemotherapy-Derived Inflammatory Responses Accelerate the

Updated version

10.1158/0008-5472.CAN-14-2921doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2015/05/07/0008-5472.CAN-14-2921.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/75/13/2629.full#ref-list-1

This article cites 50 articles, 17 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/75/13/2629.full#related-urls

This article has been cited by 12 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://cancerres.aacrjournals.org/content/75/13/2629To 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-14-2921

http://cancerres.aacrjournals.org/content/suppl/2015/05/07/0008-5472.CAN-14-2921.DC1

http://cancerres.aacrjournals.org/content/75/13/2629.full#ref-list-1

http://cancerres.aacrjournals.org/content/75/13/2629.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/75/13/2629


Documents

Chemotherapy-Derived Inﬂammatory Responses Accelerate the ... · Microenvironment and Immunology Chemotherapy-Derived Inﬂammatory Responses Accelerate the Formation of Immunosuppressive