earlydetectionoftumorcellsbyinnateimmunecellsleads to t ......in t reg trafficking and accumulation...

Microenvironment and Immunology

Early Detection of Tumor Cells by Innate Immune Cells Leadsto Treg Recruitment through CCL22 Production by TumorCells

Julien Faget1,2,3, Cathy Biota1,2,3, Thomas Bachelot1,2,3, Michael Gobert1,2,3, Isabelle Treilleux1,2,3,Nad�ege Goutagny1,2,3, Isabelle Durand1,3, Sophie L�eon-Goddard1,2,3, Jean Yves Blay1,2,3,Christophe Caux1,2,3, and Christine M�en�etrier-Caux1,2,3

AbstractIn breast carcinomas, patient survival seems to be negatively affected by the recruitment of regulatory T cells

(Treg) within lymphoid aggregates by CCL22. However, themechanisms underpinning this process, whichmay beof broader significance in solid tumors, have yet to be described. In this study, we determined how CCL22production is controlled in tumor cells. In human breast carcinoma cell lines, CCL22 was secreted at low basallevels that were strongly increased in response to inflammatory signals [TNF-a, IFN-g , and interleukin (IL)-1b],contrasting with CCL17. Primary breast tumors and CD45þ infiltrating immune cells appeared to cooperate indriving CCL22 secretion, as shown clearly in cocultures of breast tumor cell lines and peripheral bloodmononuclear cells (PBMC) or their supernatants. We determined that monocyte-derived IL-1b and TNF-a arekey players as monocyte depletion or neutralization of these cytokines attenuated secretion of CCL22. However,when purified monocytes were used, exogenous human IFN-g was also required to generate this responsesuggesting a role for IFN-g–producing cells within PBMCs. In this setting, we found that human IFN-g could bereplaced by the addition of (i) IL-2 or K562-activated natural killer (NK) cells or (ii) resting NK cells in thepresence of anti-MHC class I antibody. Taken together, our results show a dialogue between NK and tumor cellsleading to IFN-g secretion, which in turn associates with monocyte-derived IL-1b and TNF-a to drive productionof CCL22 by tumor cells and subsequent recruitment of Treg. As one validation of this conclusion in primarybreast tumors, we showed that NK cells and macrophages tend to colocalize within tumors. In summary, ourfindings suggest that at early times during tumorigenesis, the detection of tumor cells by innate effectors(monocytes and NK cells) imposes a selection for CCL22 secretion that recruits Treg to evade this early antitumorimmune response. Cancer Res; 71(19); 6143–52. �2011 AACR.

Introduction

Cancer immunosubversion is a process by which tumorcells escape destruction by the immune system through avariety of mechanisms including the production of immuno-suppressive cytokines and the alteration of dendritic cell (DC)functions (1, 2).Several studies have shown that immune cells are present

and functional in solid tumors and may promote both hu-

moral and cellular antitumor immune responses. As an ex-ample, high levels of CD8þ T cells within the tumors have beenassociated with a better clinical prognosis in colorectal cancer(3). However, in most of the cases these T cells are unable tocounteract tumor progression. In cancer patients, increasedlevels of CD4þCD25highFOXP3þ regulatory T cells (Treg), alymphocyte subset with immunosuppressive properties, aredescribed in the peripheral blood, the primary tumor micro-environment, and in the draining lymph nodes, supporting arole for Treg in cancer-induced immunosuppression. However,their effect on tumor progression varies according to thetumor type in humans. Treg have a negative impact on survivalin lung, pancreatic, gastric, liver, or ovarian carcinomapatients (4–7), whereas they may exert a beneficial role inB-cell lymphoma, head and neck, or colon carcinoma (8, 9) orhave no impact in colon, prostate, renal, or anal squamous cellcarcinoma (10, 11; for review, see ref. 12).

We recently obtained evidence, in breast carcinoma,that selectively activated Treg accumulation within lymphoidaggregates, but not in the tumor bed, has a negative impact onpatients’ survival (13). Elucidating the mechanisms involved

Authors' Affiliations: 1Centre L�eon B�erard; 2Universit�e Lyon 1; and3Institut National de la Sant6 et de la Recherche Mddicale U1052, Centrede Recherche en Canc�erologie de Lyon, F-69000 Lyon, France

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

Corresponding Author: Christine M�en�etrier-Caux, Centre L�eon B�erard,CRCL INSERM U1052/CNRS 5286, 28 rue Laennec, 69373 Lyon cedex08, France. Phone: 33-4-78-78-27-50; Fax: 33-4-78-78-27-20; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-11-0573

�2011 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6143

on July 10, 2021. © 2011 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 18, 2011; DOI: 10.1158/0008-5472.CAN-11-0573

http://cancerres.aacrjournals.org/

in Treg trafficking and accumulation in the breast tumorenvironment is thereby critical for innovative therapeuticdevelopment to fight tumor-induced immunosuppression.

Experiments in mice using Treg from CCR4�/� or condi-

tional CCR4 knockout in FOXP3þ Treg compartment haverecently identified the critical role of CCR4 in Treg traffickingin secondary lymphoid organs or tissues (14, 15). Curiel andcolleagues strongly suggested a role for CCR4/CCL22 axis inTreg recruitment in ovarian ascitis (4).

We recently showed the selective loss of membrane CCR4on tumor-associated Treg (TA-Treg) consecutive to an activerecruitment through a CC chemokine (CCL22), and that breasttumors lacking CCL22 are not colonized by Treg independentlyof their CCL17 expression status (13), strongly suggesting theimportance of CCL22 in TA-Treg recruitment within breasttumors.

CCL22, produced by myeloid DCs (mDC), B cells, macro-phages, keratinocytes, or epithelial cells (6, 16, 17) and CCL17closely related to CCL22, produced by monocyte-derived DC(18) and keratinocytes (19), are 2 ligands for CCR4 (6, 20)preferentially expressed on Th2 lymphocytes (21) and Treg (forreview, see ref. 12). In peripheral blood mononuclear cells(PBMC), CCL22 is upregulated by interleukin (IL)-4, whereas itis downregulated by IFN-g treatment (22). In contrast, IFN-gfavored CCL22 secretion by keratinocytes (16, 23) and intes-tinal epithelial cells (24).

In this study, we showed that breast tumor cell recogni-tion by NK cells leads to their activation and IFN-g secretion,which in turn triggers CCL22 production by tumor cellsthrough cooperation with monocyte-derived IL-1b andTNF-a.

Materials and Methods

Breast tumor cell lines cultureAll tumor cell lines used in this study originated from

American Type Culture Collection except CLB-SAV generatedin the laboratory. Cell lines were cultured in RPMI 1640(Invitrogen) completed with 10% FBS (Lonza), 100 U/mLpenicillin, and 100 mg/mL streptomycin (Invitrogen; completemedium) at 37�C in a 5% CO2 incubator.

Primary breast tumorsBreast tumor tissues collected at the Centre L�eon B�erard

after patient informed consent were mechanically dilaceratedto obtain "mechanic tumor disaggregation supernatants" andthen subjected to enzymatic digestion as previously described(13).

Flow cytometry analyses (ADP Cyan; Beckman Coulter)were conducted to assess the percentage of NK cells(CD3�NKp46þ) and macrophages (CD4þCD68þCD163þ; allfrom Becton Dickinson except for CD163 from eBiosciencesand CD68 from Dako Cytomation) within primary tumor cellsuspension after gating on CD45þ cells, and data were ana-lyzed with FlowJo Analysis Software (Tree Star).

Immune cells (CD45þ) or NK cells (CD3�CD56þ) andmacrophages (CD4lowCD163þ) were purified on single-cell

suspension from breast primary tumor or ascitis, respectively,by cell sorting (FACS Aria; Becton Dickinson).

Breast tumor supernatantsSingle-cell suspensions from primary or metastatic (ascitis,

pleural effusion) breast tumors were incubated at a finalconcentration of 15 � 106 cells/mL in complete medium inpetri dishes. Cultured tumor cell supernatants were collectedafter 48 hours, filtrated on 0.22 mm, and frozen.

ImmunohistochemistryExpression of CCL22 on paraffin-embedded sections of

breast tumor or peritumoral tissue was analyzed with a goatanti-CCL22 antibody (Ab; Santa Cruz) as previously described(13). Routinely used CD163 (mIgG1; Menarini Diagnostics)staining was carried out according to the manufacturer.NKp46þ was detected as previously described (25) cells witha goat immunoglobulin G Ab (R&D Systems). Hematoxylin-counterstained sections were dehydrated and mounted. Fornegative control slides, primary antibodies were replaced by anonimmune serum.

Cytokines and antibodiesRecombinant human granulocyte macrophage colony-

stimulating factor (rhGM-CSF; specific activity: 2 � 106U/mg, used at 100 ng/mL) and rhIL-4 (specific activity:106 U/mg, used at 50 ng/mL) were from Schering PloughResearch Institute. rhTNF-a (specific activity: 5 � 106 U/mg)at 10 ng/mL was provided by Cetus Corporation. rhIL-1b(109 IU/mg), rhIFN-g (2 � 107 IU/mg), and IL-1RA were fromPeprotech. rhCCL22, rhCCL17, rhCXCL12, and monoclonalAb (mAb) against CXCL12 and CCL22 as well as isotypecontrols used for neutralization experiments were from R&DSystems.

Purification of cell subsets from peripheral bloodTotal PBMCs were isolated from heparinized blood

obtained from healthy volunteers by Ficoll Hypaque densitygradient centrifugation (Dominique Dutscher). Purified mDCsandmonocyte fractions were obtained using positive selectionkits, whereas untouched NK cells were purified using negativeselection kit (Miltenyi Biotech) and purity was confirmed byflow cytometry. For depletion experiments, different cellsubsets [myeloid cells, (mDC), plasmacytoid DCs (pDC),monocytes, NK cells, and T cells] were specifically depletedfrom PBMCs using positive selection kits withmagnetic beads.The absence of remaining positive cells in the depletedfraction was confirmed by flow cytometry.

Culture conditionsTumor cell lines were cultured at 2 � 105 cells/mL in

complete medium in 48-well plates (Becton Dickinson) andincubated for 24 or 48 hours in medium condition or in thepresence of rhIFN-g (0.1–100 ng/mL depending on the stud-ies). Coculture experiments were carried out by incubating 105

tumor cells with 106 PBMC for 24 or 48 hours in the presenceof 100 ng/mL rhGM-CSF with or without rhIFN-g . To char-acterize the cell subset responsible for CCL22 secretion, PBMC

Faget et al.

Cancer Res; 71(19) October 1, 2011 Cancer Research6144




cell supernatants (PBMC-SN) and tumor cell supernatants(TUM-SN) were generated by 24-hour incubation of either 106

PBMC or 105 tumor cells in 48-well plates (500 mL).

Cytokine detectionCCL22, CCL17 and IL-1b, and TNF-a levels were quantified

in cell supernatants using ELISA from R&D Systems andBender MedSystems, respectively.

Migration in response to CCL22CCR4 expression on the CCRF-CEM cell line was confirmed

by flow cytometry (Supplementary Fig. S1). Migration assayswere conducted using Transwell (6.5-mm diameter; CoStar)with 5� 105 cells/well. After 2 hours of preincubation at 37�C,CCRF-CEM cells were placed in 3-mm pore size inserts (100mL) and tested for their ability to migrate in response torhCCL22 (1–50 ng/mL) or culture supernatants (50%) addedin the lower well. After 1 hour and 30 minutes of incubation at37�C, cells were collected in cold PBS–EDTA and resuspendedafter centrifugation in 100 mL. The number of migrated cellswas analyzed by flow cytometry. In blockade experiments,anti-CXCL12 or -CCL22 mAb or their isotype controls were

incubated for 30 minutes with culture supernatants beforeCCRF-CEM cells were added in the insert.

Results

High levels of CCL22 but not CCL17 are detectable inprimary breast tumors

Analyses of breast TUM-SN showed the production of highlevels of CCL22 in TUM-SN from primary tumors (Fig. 1A) thatdecreased in metastatic ones. CCL22 was also detectable insupernatants of mechanical tumor disaggregation from 27primary tumors (mean ¼ 1.02 ng/mL; range, 0.13–6.9 ng/mL).Low levels of CCL17 were detected in these TUM-SN. More-over, supernatant from nontumor tissues (healthy breasttissue and fibro-adenoma) did not produce significant levelsof CCL22 or CCL17.

We have previously shown that expression of CCL22, butnot that of CCL17 by tumor cells in breast tumors, correlateswith TA-Treg infiltration (13).

In contrast to primary breast tumors where a strong CCL22expression was observed by immunohistochemistry (IHC;

Figure 1. CCL22 and CCL17 aredifferently produced withinprimary breast tumors and breastTUM-SN. A, quantification ofCCL22 and CCL17 levels withinhealthy or tumor breast tissuesupernatants by ELISA; primarybreast tumors (C) as well asperitumoral area (B) frozensections were stained with CCL22Ab in green (�10). CB

A

Primary Metastasis Healthy

0

2

4

6

8

10

12

14

CC

L22

(n

g/m

L)

11

1

33

39

0

1

2

3

4

5C

CL

22 (

ng

/mL

)

33

10

39

1

Primary Metastasis Healthy

Breasttumor

Mammarygland

Breasttumor

Mammarygland

Innate Immune Recognition Triggers CCL22 by Breast Tumor Cells

www.aacrjournals.org Cancer Res; 71(19) October 1, 2011 6145




Fig. 1C; ref. 13), CCL22 displayed a weak apical expression inperitumoral area by luminal breast epithelial cells withinlobular acini (Fig. 1B).

Breast tumor cell lines produce functional CCL22 inresponse to Th1/innate immunity but not Th2 signals

Contrasting with primary tumors, spontaneous CCL22 se-cretion by breast tumor cell lines in vitro was low to unde-tectable (Fig. 2A), suggesting mechanisms of regulation and arole of the microenvironment in CCL22 expression by tumorcells. Indeed, addition of rhIFN-g , a CCL22 inducer on kera-tinocytes (16, 23) and intestinal epithelial cells (24), inducedstrong CCL22 secretion on 5 of 7 tested cell lines (0. 28–1.1ng/mL for IFN-g ; Fig. 2A). In contrast, in PBMCs, CCL22production was downregulated by rhIFN-g but upregulatedby rhIL-4 (Fig. 2B).

When bulk primary breast tumor disaggregation was used,the secretion of CCL22 was lost upon depletion of CD45þ

immune cells. This CCL22 production by CD45-negative pri-mary tumor was restored (9-fold increase) either by additionof associated CD45þ infiltrate (3 � 104 CD45þ for 8 � 104tumor cells) or rhIFN-g (Fig. 2C), strongly suggesting thecooperation between tumor cells and immune cells for spe-cific CCL22 secretion.

This observation was confirmed using breast tumor celllines. The addition of allogeneic PBMCs to breast tumor celllines strongly enhanced the production of CCL22 but not ofCCL17. We observed a 12.8-, 18-, and 121-fold increase forCLB-SAV, MDA-MB453, and MCF-7 cell lines, respectively, incoculture condition when compared with tumor cells alone.

This secretion was further enhanced by rhIFN-g addition (17-and 5.5-fold increase in CCL22 production for CLB-SAV andMDA-MB453 or MCF-7, respectively; Fig. 2D).

To test the functionality of the CCL22 secreted in PBMC/tumor cell coculture supernatants, we used the CCRF-CEM T-cell line that expresses CCR4 and migrates in response torhCCL22 in a dose-dependent manner (SupplementaryFig. S1A and B). The coculture supernatant favored theCCRF-CEM cell migration in a Transwell Assay (5-fold in-crease over background level) that was specifically blocked bypreincubation of these supernatants with an anti-CCL22neutralizing mAb but not with an anti-CXCL12 able to attractCXCR4þ CCRF-CEM (Supplementary Fig. S1C).

Soluble factors produced by activated PBMCs induceCCL22 secretion by tumor cells

Whereas PBMCs alone are devoid of IFN-g secretion (Sup-plementary Fig. S3, bottom), coculture with tumor cells in-creased this production favoring CCL22. Furthermore, aspreviously mentioned, addition of rhIFN-g strongly enhancedCCL22 production in PBMC/tumor cell coculture while de-creasing that of PBMCs (Fig. 2D). We therefore wished todecipher the relative role of PBMCs and tumor cells in CCL22secretion in the PBMC/tumor cell coculture in the presence ofrhIFN-g . We compared the impact of rhIFN-g-activatedPBMC-SN on CCL22 production by tumor cell lines with thatof rhIFN-g-activated TUM-SN on PBMCs. As shown in Fig. 3,CCL22 levels secreted by rhIFN-g-treated tumor cells werestrongly enhanced in the presence of rhIFN-g-PBMC-SN (2.75-,19.36-, and 16.46-fold, respectively, for MDA-MB453, MCF-7,

B

A

0Medium IL-4 IFN-γ

0.5

1.0

CC

L22

(n

g/m

L)

MDA-MB453CLB-SAVMCF-7BT474CAMA-1SKBR3BT20

Tumor cells

PBMC

C

00.20.40.60.81.0

Bulk Med IFN-γ

CC

L22

(n

g/m

L)

CD45+

Tumor cells (CD45–)

CD45+

PBMC

MDA-MB453

CLB-SAV

MCF-7

Tumorcells

Tumor cells + PBMC

Medium IFN-γ

Tumorcells

Tumor cells + PBMC

0

200

CC

L17

(p

g/m

L)

100

50

150

CC

L22

(n

g/m

L)

0

0.5

1.0

1.5

2.0

2.5D

CocultureC

CL

22 (

ng

/mL

)

0

1

2

3

4

5

6

Medium IL-4 IFN-γ

Figure 2. CCL22 secretion bybreast epithelial cell lines afteractivation. Breast epithelial celllines (105) as shown in A or healthyPBMCs (106) as shown in Bsecreted CCL22 after24-hour culture in response torhIL-4 (50 ng/mL) or rhIFN-g(100 ng/mL). C, quantification ofCCL22 secreted by primary tumordisaggregation (8 � 104/200 mL)and cell-sorted tumor cells(CD45�; 8 � 104/200 mL) witheither recombinant cytokines(IFN-g ) or CD45þ immune cells(3 � 104/200 mL). D, analysis ofCCL22 and CCL17 production, bybreast epithelial cell lines alone orin coculture with PBMCs with orwithout rhIFN-g (100 ng/mL).

Faget et al.





and CLB-SAV, respectively). In contrast, rhIFN-g-TUM-SNaddition did not affect the low CCL22 levels detected inrhIFN-g-treated PBMC cultures. Together, those data stronglysuggest that rhIFN-g-treated PBMCs produce soluble factorscapable of inducing CCL22 production by tumor cells.

Interestingly, the effects observed were specific for CCL22as CXCL8 that was produced by rhIFN-g-activated PBMCs wasdownregulated in the presence of tumor cells and not inducedwhen tumor cells were cultured with PBMC-SN (data notshown).

Monocytes and IFN-g are both required for thesecretion of CCL22 by tumor cells

To determine the major cell fraction within PBMC respon-sible for the effects observed on tumor cells, specific depletionsof myeloid cells (CD33þ), monocytes (CD14þ), mDC(BDCA1þBDCA3þ), pDC (BDCA2þ), NK cells (CD56þ), or Tcells (CD3þ) were carried out using magnetic beads. Eachdepleted fractionwas added on tumor cell lines in the presenceof rhIFN-g to assess the CCL22 production. As shown in Fig. 4A,whereas addition of PBMCs induced a strong CCL22 produc-tion (2.95 � 0.1 ng/mL), we observed a drop in this secretionwhen monocyte (CD14�) or myeloid cell (CD33�)-depletedfractions were used (86% and 75% inhibition, respectively). Incontrast, the depletion of NK cells or T cells did not decreasethe CCL22 secretion. The increase observed with T-cell deple-tion likely results from increased monocyte percentage in theculture. The depletion of mDCs reduced the basal level ofCCL22 produced by PBMCs alone as shown in SupplementaryFig. S2 (26) but did not affect the CCL22 production by tumorcells. These results suggest that monocytes are the main actorsin CCL22 production by tumor cells within the coculture.

0

1

2

3

4

5

6

CC

L22

(n

g/m

L)

Med

ium

MC

F-7

MD

A-M

B45

3

CLB

-SA

V

The

lma

Med

ium

MC

F-7

MD

A-M

B45

3

CLB

-SA

VT

helm

a

Med

ium

MC

F-7

MD

A-M

B45

3

CLB

-SA

VT

helm

a

Tumor cells+ PBMC

PBMC-SN +tumor cells

Tumor cell SN+ PBMC

Tumorcells

MC

F-7

MD

A-M

B45

3

CLB

-SA

V

The

lma

Figure 3. Soluble factors secreted by PBMCs favor the production ofCCL22 by breast tumor cells. A 48-hour culture period of 105 breast tumorcells [with rhIFN-g (100 ng/mL)] alone or in coculture with PBMCs (106) or24-hour PBMC-SN obtained in rhIFN-g (100 ng/mL) medium induced thesecretion of CCL22, whereas culture of PBMCs with 24-hour breastTUM-SN did not induce CCL22 secretion.

B

0

1

2

3

4

5

6

CC

L22

(n

g/m

L)

_ IFN-γ _ IFN-γ

VAS-BLC7-FCM

Medium

PBMC

CD14 +

mDC+

A

PBMC

CD14–CD33–mDC–

NK–

Tcell–Tumor cells

0

1

2

3

4

5

6

7

_ MCF-7 CLB-SAV

CC

L22

(n

g/m

L)

MCF-7 +PBMC-SN

PBMC-SN

CCL22 (ng/mL)

Medium

Ctrl mAb

anti-TNF-α

IL-1RA + Ctrl mAb

IL-1RA + anti-TNF-α

IL-1RA

0 0.5 1 1.5 2 2.5

C

Figure 4. Within PBMCs, monocytes are essential to favor CCL22 production by breast tumor cells but required the presence of rhIFN-g , and CCL22production in culture of tumor cells with PBMC-SN is dependent on IL-1b and TNF-a. A, 105 tumor cells (MCF-7 and CLB-SAV) were cultured in thepresence of rhIFN-g for 48 hours with medium or 106 PBMCs or fractions depleted in myeloid cells (CD33�; 106), in CD14 (CD14�; 9 � 105), in mDC(mDC�; 106), in NK cells (NK�; 9 � 105), or in T cells (Tcells�; 5 � 105) for analysis of CCL22 production. B, CCL22 production after culture of 105 breasttumor cells (CLB-SAV andMCF-7) with purifiedmonocytes (CD14þ; 105) or mDC (mDCþ; 105) for 48 hours inmediumor in the presence of rhIFN-g (100 ng/mL).C, 24-hour rhIFN-g preactivated MCF-7 tumor cells were incubated for 2 hours with IL-1RA or not. In parallel, SN of a 24-hour PBMC culture in the presenceof 100 ng/mL rhIFN-g (PBMC-SN) was preincubated with 10 mg/mL control (Ctrl) mAb or anti-TNF-a mAb and then added on tumor cells for 24 hours.CCL22 production was quantified at the end of the culture.






To confirm their role, purified monocytes were added totumor cells in the presence or not in the presence of rhIFN-g(Fig. 4B). Whereas purified monocytes were not able to mimicPBMC action on tumor cells, further addition of rhIFN-g-in-duced CCL22 levels comparable with those obtained withPBMC. This effect was specific to monocytes as mDC, evenin presence of IFN-g , did not reconstitute PBMC effect. Ofmost importance, these results suggest that monocytes act incooperation with other cell subsets (i.e., NK cells, NKT cells, orT cells) capable of IFN-g secretion to increase CCL22 secretionby tumor cells.

Involvement of rhIL-1b and TNF-a in inducing CCL22secretion by tumor cells

As shown above (Fig. 4A), depletion of monocytes stronglyreduced the ability of tumor cells to produce CCL22 incoculture. Monocytes are strong producers of IL-1b andTNF-a, previously described to cooperate with IFN-g inCCL22 production on epithelial cells or keratinocytes (24,27, 28). As shown in Supplementary Fig. S3, whereas PBMCproduced low IL-1b and TNF-a levels (6.5 � 0.2 and 99 � 1pg/mL, respectively), the addition of rhIFN-g increased theirsecretion (IL-1b, 133 � 12 pg/mL; and TNF-a, 879 � 53pg/mL). These 2 cytokines are also detected in tumor cellline/PBMC coculture in the presence of rhIFN-g . Moreover,the loss of CCL22 production in monocyte-depleted fractionwas associated with the absence of IL-1b and TNF-a secretionin the coculture (data not shown).

Whereas CCL22 production by MDA-MB453 cells was most-ly dependent on IFN-g (Supplementary Fig. S4A and C), theculture of MCF-7 (Supplementary Fig. S4B and D) or CLB-SAV(data not shown) with a cross range of recombinant cytokinesshowed an important impact of low doses of IL-1b (100 pg/mL)or TNF-a (10 ng/mL) on CCL22 production with an additiveeffect of IFN-g (Supplementary Fig. S4B and D).

To confirm a role for these 2 cytokines within rhIFN-g-PBMC-SN, in CCL22 secretion by tumor cells, we testedthe impact of IL-1 receptor antagonist (IL-1RA) or an anti-TNF-a blocking Ab, previously validated (SupplementaryFig. S5A), on tumor cell line cultures. As shown in Fig. 4C,treatment with either IL-1RA or anti-TNF-a mAb was able toblock 40% of CCL22 secretion induced by rhIFN-g-PBMC-SN.The simultaneous blockade of IL-1b and TNF-a decreasedCCL22 secretion up to 80%, showing the role of IL-1b andTNF-a contained in rhIFN-g-PBMC-SN on CCL22 secretion bytumor cells.

NK cells and monocytes cooperate to induce CCL22production by tumor cells

To better understand the mechanisms involved in IFN-gsecretion within PBMC/tumor cell coculture (SupplementaryFig. S3) that act in synergy with IL-1b and TNF-a, we hypothe-sized that NK cells could be activated and secrete IFN-g afterinteraction with tumor cells.

To mimic NK activation, NK cells were pretreated with IL-2for 16 hours. We tested their impact on CCL22 production bytumor cells in the presence of purified monocytes or mDC-depleted PBMC fraction. Whereas activated NK cells, mDC-

depleted fraction, or purified monocytes each alone (Fig. 5Aand B) did not trigger CCL22 production by tumor cells, acombination of activated NK cells with either mDC-depletedfraction (Fig. 5A) or purified monocytes (Fig. 5B) inducedCCL22 levels comparable with those obtained in the presenceof PBMCs or exogenous rhIFN-g (Fig. 5A). This suggests thatIFN-g released by activated NK cells cooperates with mono-cytes to promote CCL22 release by tumor cells.

Interaction of K562 tumor cell line with NK cells also favorstheir activation (29). The addition of K562 (1:1 K562:NK ratio)to resting NK cells in the presence of purified monocytes andtumor cells increased CCL22 secretion by tumor cells that wasdependent on IFN-g as anti-IFN-gR1 blocking mAb reversedthis effect (Supplementary Fig. S6).

Because tumor cell lines upregulate MHC class I in responseto IFN-g or TNF-a (Supplementary Fig. S7), we neutralizedMHC class I expression on tumor cells as an alternativeapproach to revert blockade of NK activation through MHCclass I/killer inhibitory receptor (KIR) interactions. The pre-incubation of tumor cells with blocking anti-MHC class I mAb(W6/32) before the addition of resting NK cells and monocytessignificantly increased the CCL22 production. This increasewas strictly dependent on the presence of NK cells and

MDA-MB453MCF-7Medium0

0.5

1.0

1.5

2.0

2.5

3.0

CC

L22

(n

g/m

L)

Medium

rhIFN-γ

mDC–

IL-2-NKmDC– +IL-2-NK

mDC–+ rhIFN-γ

0

0.2

0.4

0.6

0.8

1.0

1.2

MDA-MB453MCF-7MediumC

CL

22 (

ng

/mL

)

MediumCD14+IL-2-NKCD14+ + IL-2-NK

A

B

Figure 5. Cooperation of monocytes and activated NK cells topromote CCL22 secretion by tumor cells. A, mDC-depleted fraction(mDC�; 106), IL-2–activated NK cells (IL-2–NK; 105), or their combinationwere added on 24-hour medium–pretreated breast tumor cells (MCF-7,MDA-MB453; 105), and CCL22 secretion was analyzed after a 48-hourculture period. The control was carried out by addition of mDC-depletedfraction (106) on rhIFN-g-pretreated tumor cells. B, purified monocytes(CD14þ; 105), IL-2–activated NK cells (IL-2-NK; 105), or their combinationwere added on 24-hour medium pretreated breast tumor cells (MCF-7,MDA-MB453; 105), and CCL22 secretion was analyzed after 48 hours ofculture.

Faget et al.





monocytes (Fig. 6). In these experimental conditions, additionof IL-1RA, anti-IFN-gR1, and anti-TNF-amAb used alone haveall shown a moderate to strong effect depending on the cellline, but when combined they completely blocked anti-MHCclass I impact, showing the involvement of IFN-g , TNF-a, andIL-1b (Fig. 6).

NK cells and macrophages colocalize with tumor cellsin situAs showed in Fig. 7A and B, NK cells (CD3�NKp46þ) as well

as macrophages (CD163þCD68þ) were detected within theprimary tumor cell suspensions by flow cytometry [mean ¼3.72% (0.15%–8.2%) for NKp46þ and mean ¼ 11.7% (0.58%–37.1%) for CD163þ]. As shown by IHC on paraffin-embedded

tumor sections, NK cells (NKp46þ; Fig. 7C and D) as well asmacrophages (CD163þ; Fig. 7E and F) are localized in thevicinity of tumor cells. Moreover, purified in situ activatedascite-derived macrophages are able to cooperate with NKcells to favor a strong CCL22 production by breast tumor cellline (Fig. 7G). Together, these data suggest the potentialrecognition of tumor cells by NK cells favoring in combinationwith macrophages, the initiation of CCL22 secretion by thesetumor cells.

Discussion

In this study, we showed that recognition of transformedmammary epithelial cells favors NK cell activation and

Figure 6. Blockade of IL-1b,TNF-a, and IFN-g reverse tumorcell CCL22 secretion induced afterculture with purified monocytesand NK cells in the presence ofanti-MHC class I blocking Ab.Twenty-four hour medium–cultured breast tumor cells (MCF-7 and MDA-MB453; 105) weretreated with anti-class I mAb,Ctrl mAb, anti-IFN-gR mAb(10 mg/mL), IL-1RA (100 ng/mL),and their combination for 2 hoursbefore the addition of purifiedmonocytes (105) and NK cells(5 � 105) with or without anti-TNF-a mAb (10 mg/mL). CCL22secretion was analyzed after a48-hour culture period.

0

100

200

300

400

500

600

700

800

CC

L22

(p

g/m

L)

MediumMCF-7CLB-SAV

Ctrl mAb

Class I mAb

IFN-γR1 mAb

TNF-α mAb

IL-1RA

CD14+ + NK

-

+

-

+

+

-

+

+

-

+

+

+

-+--

----

-

+

+

+

+

++++

-

+

-

-

+

+

+

+

-

+

+

+

+

+

+

-

+

+

-

+

+

+

-

++

-

-

-

-

-

+

-

-

-

-

-

-

Figure 7. NK cells andmacrophages are detectablewithin breast tumors and arefunctional. A, NK cell(CD3�NKp46þ) and macrophage(CD68þCD163þ) detection aftergating on CD45þ cells withinprimary breast tumor enzymaticdisaggregation or in associatedperipheral blood. B, summary dataof NK and macrophagepercentages in 7 tumordilacerations by flow cytometry(C–F) localization of NK (NKp46; Cand D) and macrophages (CD163;E and F) by IHC on paraffin-embedded primary breast tumortissue sections [magnification:�20 (C and E) or�40 (D and F)]. G,CCL22 production after culture of105 breast tumor cells (MCF-7)with breast ascite purifiedmacrophages (105) and NK cells(5 � 105) or their combination for48 hours.

Blood TumorMacrophages

CD68

CD

163

30%14.4%

Blood Tumor

NKp46

FS

C

NK cells

1.88%5.4%

CD 163NKp46

x20

x40Me

dium

Microp

hages

Microp

hages

+ NKNK

CC

L22

(n

g/m

L)

0

0.5

1

1.5

2

2.5

3MediumMCF-7

Microp

hages NK

% C

D16

3+w

ith

in C

D45

+

0

12

3

45

67

8

9

0

5

10

15

20

25

30

35

40 % N

Kp

46+

with

in C

D45

+

A B

FD

ECG






subsequent IFN-g secretion associated with the release ofmonocyte-derived IL-1b and TNF-a that triggers CCL22 pro-duction by tumor cells. This tumor cell–associated CCL22secretion favors blood CCR4þ Treg recruitment leading to thedevelopment of a tolerogenic environment conducive to thetumor immunosubversion and development.

We previously reported a strong correlation betweenCCL22 expression by tumor cells and the presence of Tregwithin breast tumor environment. Breast tumors lackingCCL22 are not colonized by Treg independently of theirCCL17 expression status (13). Moreover, Treg recruitmentin tumor environment induces a loss of CCR4 expression, aphenomenon observed when Treg are cultured in vitro withCCL22 but not with CCL17 (30). Similar observations weremade for CCR7 expression that was downregulated onT cells after interaction with CCL19 but not with CCL21(31, 32). In agreement with our observation, CCR4 andCCL22 requirement for Treg recruitment was also reportedin the mouse model of inflammatory bowel disease, in whichthe inability of CCR4�/� Treg to migrate within the colontissue leads to disease exacerbation (15).

Our IHC analyses show, in peritumoral breast tissue sam-ples, polarized apical CCL22 secretion by healthy luminalepithelial cells within lobular acini as described for otherchemokines (CXCL8, GROb, GROg , GROa, ENA78, MIG,IP10, and RANTES) detected in the milk or the colostrum(33, 34). Moreover, nonhematopoietic cells such as keratino-cytes and epithelial cells can secrete CCL22 (16, 24, 35).Polarized CCL22 secretion toward the lumen has also beendescribed in colon epithelium (24). Moreover, the cyclichormonal modulation may also affect Treg recruitment withinthe mammary gland via CCL22 secretion by epithelial cells.Indeed, treatment of women with progesterone favors in theendometrium a high CCL22 production by stromal cells andglandular epithelial cells at the end of the hormonal cycle (35).Taken together, these results suggest that CCL22 secretionwithin the breast tissue may be part of the mammary glandphysiology controlling the local inflammation associated withtissue remodeling either at the end of the menstrual cycle orduring breastfeeding.

In accordance with structural disorganization characteris-tic of primary breast tumor tissue, we observed that CCL22secretion is no more polarized favoring its diffusion within thetumor environment that may favor recruitment of macro-phages, NK cells, Th2 cells (6), and Treg (for review, see ref. 12)expressing CCR4. Moreover CCL22 production is stronglyenhanced when compared with healthy tissue. This is con-sistent with the levels of CCL22 found either in primary breasttumor mechanical disaggregation SNs (median ¼ 1.16 ng/mL;range, 0.23–8.8 ng/mL) or in 48-hour culture primary breastTUM-SN with more than 40-fold increase in CCL22 levels(median ¼ 2.91 ng/mL; range, 0.53–12.4 ng/mL) in compar-ison with nontumor SN (median ¼ 0.07 ng/mL; range, 0.03–0.23 ng/mL; P ¼ 0.004). It is also important to notice thatCCL22 content is 10-fold higher than that of CCL17 (median¼0.3 ng/mL; range, 0–4.4 ng/mL). Importantly, we show usingeither primary breast tumor or tumor cell lines the cooper-ation between tumor cells and immune infiltrate to induce

high quantities of CCL22, whereas CCL17 secretion remainsbarely detectable. Interestingly, although healthy bronchialepithelial cells secrete CCL17 (36), their tumor counterpart inlung carcinoma pleural effusion produces CCL22 (37). Takentogether, these results suggest the capacity of the tumorenvironment to modulate the chemokine arsenal of epithelialcells to favor the migration of specific cell subsets. CCL17, viathe recruitment of Th2 CCR4þ cells, will favor a Th2 responseas described in atopic dermatitis (38), whereas CCL22 is morespecialized in the recruitment of Treg as observed in tumors(for review, see ref. 12).

IL-4 and IL-13, critically involved in the development ofcutaneous pathologies like atopic dermatitis, have been large-ly shown to induce CCL22 secretion by cells of myeloid origin(monocytes and mDCs; refs. 22, 39) and to favor CCL17production by fibroblasts (40). In contrast, we showed in thisstudy that IL-4 reduces the CCL22 production in breast tumorepithelial cell lines as previously described for immortalizedkeratinocytes (16, 23), colon epithelial cells (24), and gliomacell lines (41).

In this study, we deciphered the mechanisms involved inthe increased secretion of CCL22 within the tumor environ-ment. We showed the existence of a dialogue between tumorcells and circulating immune cells leading to CCL22 produc-tion by tumor cells and to Treg recruitment. We reported thatbreast tumor cell lines producedCCL22 in response to rhIFN-g ,as previously described for keratinocytes (16, 23). This secre-tion is strongly enhanced in coculture with PBMCs but is lostafter myeloid cell (CD33þ) or monocyte (CD14þ) depletionshowing the major role of monocytes in this CCL22 secretionalthough they do not secrete CCL22 by themselves.

NK cells constitute a unique component of the innateimmune system able, without specific sensitization, to recog-nize autologous cells undergoing various form of stress, suchas malignant transformation (42). Target recognition occursvia the integration of negative and positive signals mediatedby inhibitory (KIR) or activating (KAR) receptors expressed atthe surface of NK cells. Breast tumor cells expressing ULBP orMICA/MICB markers that bind NKG2D on NK cells willstimulate their IFN-g secretion (Supplementary Fig. S5; refs.43, 44). However, expression of MHC class I (SupplementaryFig. S7), a KIR ligand, by breast tumor cells reduced this IFN-gsecretion. In coculture of breast tumor cell lines with purifiedmonocytes, rhIFN-g could be omitted upon addition of NKcells in conditions leading to their activation, that is, in thepresence of IL-2, K562 NK target cell line, or anti-MHC class IAb. All these culture conditions lead to IFN-g secretionrequired for CCL22 production as shown by the use of block-ing anti-IFN-gR Ab.

In this cell line, MICA (NKG2D-L) expression in breasttumors has been associated with a poor prognosis (43). Thiscould result either from the production of soluble MICA thatblocks the killing function of NK cells or the impact of IFN-gsecretion by NK cells on Treg recruitment through CCL22secretion by tumor cells. In the Lewis Lung carcinoma mousemodel, depletion of NK cells blocked CCL22 production in thetumor environment; however, NK cells were proposed as themajor source of CCL22 (45). In contrast, we never detected, in

Faget et al.





our experimental set-up, CCL22 secretion by resting as well asactivated NK cells.The replacement of purified monocytes/macrophages by

the combination of rhIL-1b and rhTNF-a in the culture ofbreast tumor cells with rhIFN-g or NK cells strongly enhancedthe CCL22 production. This observation is in agreement withprevious publications on keratinocytes and colon epithelialcell lines (16, 24), whereas blockade of IL-1b and TNF-aabrogate this secretion in [PBMC/tumor cell] coculture. In-terestingly, IFN-g increased IL-1R1 and TNF receptor ontumor cell lines (data not shown), as previously described(46), suggesting a potential amplification loop of CCL22production. Taken together, these results suggest the impor-tance of inflammation in the high CCL22 levels in the breasttumor environment that will favor Treg recruitment leading toreduced specific antitumor immune response. This is inagreement with studies in colon tissue reporting the involve-ment of intestinal flora-mediated chronic inflammation in theincreased recruitment of Treg (for review, see ref. 47). Thissuggests that inflammation in the mammary gland mayparticipate in the tumor development. In favor of this,TNF-a secretion by leukocytes infiltrating tumors stronglycontributes to mammary carcinogenesis in murine mammarymodels (48). Importantly, the in situ analyses on primarybreast tumors allow us to show the presence of NK cellsand macrophages in the vicinity of tumor cells.Treg have been described to reduce NK cell cytotoxicity (for

review, see ref. 49) suggesting that CCL22 production by tumor

cells inducing Treg recruitment represents one of the mechan-isms elaborated by tumors to avoid its destruction through NKcell cytotoxicity.

Taken together, our results allow us to propose a model inwhich mammary epithelial cell transformation processes fa-vored activation of NK cells present in the breast tissue byreducing KIR and inducing KAR ligand expression and theirsubsequent IFN-g secretion, leading to the production ofTNF-a and IL-1b by resident monocytes/macrophages. Actingtogether, these 3 cytokines will favor CCL22 overproduction bytumor cells, allowing the recruitment of CCR4þ blood Treg thatfavor the development of a tolerogenic environment.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Grant Support

J. Faget and M. Gobert are grant holders of the Ligue Nationale contre leCancer. This work was financially supported in part by grants from "le comit�ed�epartemental du Rhône de Ligue Contre le Cancer," the ARC Association (ARC-5074), the Breast Cancer Research Foundation and Institut National du Cancergrant INCA ACI-63-04, ACI 2007-2009.

The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received February 21, 2011; revised July 1, 2011; accepted July 25, 2011;published OnlineFirst August 18, 2011.

References1. Thomachot MC, Bendriss-Vermare N, Massacrier C, Biota C, Treilleux

I, Goddard S, et al. Breast carcinoma cells promote the differentiationof CD34þ progenitors towards 2 different subpopulations of dendriticcells with CD1a(high)CD86(-)Langerin- and CD1a(þ)CD86(þ)Langerinþ phenotypes. Int J Cancer 2004;110:710–20.

2. Treilleux I, Blay JY, Bendriss-Vermare N, Ray-Coquard I, Bachelot T,Guastalla JP, et al. Dendritic cell infiltration and prognosis of earlystage breast cancer. Clin Cancer Res 2004;10:7466–74.

3. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A,Mlecnik B, Lagorce-Pages C, et al. Type, density, and location of immune cells withinhuman colorectal tumors predict clinical outcome. Science 2006;313:1960–4.

4. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, et al.Specific recruitment of regulatory T cells in ovarian carcinoma fostersimmune privilege and predicts reduced survival. Nat Med 2004;10:942–9.

5. Gao Q, Qiu SJ, Fan J, Zhou J, Wang XY, Xiao YS, et al. Intratumoralbalance of regulatory and cytotoxic T cells is associated with prog-nosis of hepatocellular carcinoma after resection. J Clin Oncol2007;25:2586–93.

6. Godiska R, Chantry D, Raport CJ, Sozzani S, Allavena P, Leviten D,et al. Human macrophage-derived chemokine (MDC), a novel che-moattractant for monocytes, monocyte-derived dendritic cells, andnatural killer cells. J Exp Med 1997;185:1595–604.

7. Mizukami Y, Kono K, Kawaguchi Y, Akaike H, Kamimura K, Sugai H,et al. Localisation pattern of Foxp3þ regulatory T cells is associatedwith clinical behaviour in gastric cancer. Br J Cancer 2008;98:148–53.

8. Badoual C, Hans S, Rodriguez J, Peyrard S, Klein C, Agueznay NH,et al. Prognostic value of tumor-infiltrating CD4þ T-cell subpopula-tions in head and neck cancers. Clin Cancer Res 2006;12:465–72.

9. Tzankov A, Meier C, Hirschmann P, Went P, Pileri SA, Dirnhofer S.Correlation of high numbers of intratumoral FOXP3þ regulatory T cells

with improved survival in germinal center-like diffuse large B-celllymphoma, follicular lymphoma and classical Hodgkin's lymphoma.Haematologica 2008;93:193–200.

10. Fox SB, Launchbury R, Bates GJ, Han C, Shaida N, Malone PR, et al.The number of regulatory T cells in prostate cancer is associated withthe androgen receptor and hypoxia-inducible factor (HIF)-2alpha butnot HIF-1alpha. Prostate 2006;67:623–9.

11. Grabenbauer GG, Lahmer G, Distel L, Niedobitek G. Tumor-infiltratingcytotoxic T cells but not regulatory T cells predict outcome in analsquamous cell carcinoma. Clin Cancer Res 2006;12:3355–60.

12. Menetrier-Caux C, Gobert M, Caux C. Differences in tumor regulatoryT-cell localization and activation status impact patient outcome.Cancer Res 2009;69:7895–8.

13. Gobert M, Treilleux I, Bendriss-Vermare N, Bachelot T, Goddard-LeonS, Arfi V, et al. Regulatory T cells recruited through CCL22/CCR4 areselectively activated in lymphoid infiltrates surrounding primary breasttumors and lead to an adverse clinical outcome. Cancer Res 2009;69:2000–9.

14. Sather BD, Treuting P, Perdue N, Miazgowicz M, Fontenot JD,Rudensky AY, et al. Altering the distribution of Foxp3(þ) regulatoryT cells results in tissue-specific inflammatory disease. J Exp Med2007;204:1335–47.

15. Yuan Q, Bromley SK, Means TK, Jones KJ, Hayashi F, Bhan AK, et al. .CCR4-dependent regulatory T cell function in inflammatory boweldisease. J Exp Med 2007;204:1327–34.

16. Fujii-Maeda S, Kajiwara K, Ikizawa K, Shinazawa M, Yu B, Koga T,et al. Reciprocal regulation of thymus and activation-regulated che-mokine/macrophage-derived chemokine production by interleukin(IL)-4/IL-13 and interferon-gamma in HaCaT keratinocytes is mediat-ed by alternations in E-cadherin distribution. J Invest Dermatol2004;122:20–8.

17. Hino R, Kobayashi M, Mori T, Orimo H, Shimauchi T, Kabashima K,et al. Inhibition of T helper 2 chemokine production by narrowband






ultraviolet B in cultured keratinocytes. Br J Dermatol 2007;156;830–7.

18. Sallusto F, Mackay CR, Lanzavecchia A. The role of chemokinereceptors in primary, effector, and memory immune responses. AnnuRev Immunol 2000;18:593–620.

19. Vestergaard C, Kirstejn N, Gesser B, Mortensen JT, Matsushima K,Larsen CG. IL-10 augments the IFN-gamma and TNF-alpha inducedTARC production in HaCaT cells: a possible mechanism in theinflammatory reaction of atopic dermatitis. J Dermatol Sci 2001;26:46–54.

20. Imai T, Chantry D, Raport CJ,WoodCL, NishimuraM, Godiska R, et al.Macrophage-derived chemokine is a functional ligand for the CCchemokine receptor 4. J Biol Chem 1998;273:1764–8.

21. Lloyd CM, Delaney T, Nguyen T, Tian J, Martinez A, Coyle AJ, et al. CCchemokine receptor (CCR)3/eotaxin is followed by CCR4/monocyte-derived chemokine in mediating pulmonary T helper lymphocyte type2 recruitment after serial antigen challenge in vivo. J Exp Med2000;191:265–74.

22. Andrew DP, Chang MS, McNinch J, Wathen ST, Rihanek M, Tseng J,et al. STCP-1 (MDC) CC chemokine acts specifically on chronicallyactivated Th2 lymphocytes and is produced by monocytes on stimu-lationwith Th2 cytokines IL-4 and IL-13. J Immunol 1998;161:5027–38.

23. Xiao T, Kagami S, Saeki H, SugayaM, Kakinuma T, Fujita H, et al. BothIL-4 and IL-13 inhibit the TNF-alpha and IFN-gamma enhanced MDCproduction in a human keratinocyte cell line, HaCaT cells. J DermatolSci 2003;31:111–7.

24. Berin MC, Dwinell MB, Eckmann L, Kagnoff MF. Production of MDC/CCL22 by human intestinal epithelial cells. Am J Physiol GastrointestLiver Physiol 2001;280:G1217–26.

25. Delahaye NF, Rusakiewicz S, Martins I, Menard C, Roux S, Lyonnet L,et al. Alternatively spliced NKp30 isoforms affect the prognosis ofgastrointestinal stromal tumors. Nat Med 2011;17:700–7.

26. Vulcano M, Albanesi C, Stoppacciaro A, Bagnati R, D’Amico G, StruyfS, et al. Dendritic cells as a major source of macrophage-derivedchemokine/CCL22 in vitro and in vivo. Eur J Immunol 2001;31:812–22.

27. Rodenburg RJ, Brinkhuis RF, Peek R, Westphal JR, Van Den HoogenFH, van Venrooij WJ, et al. Expression of macrophage-derived che-mokine (MDC) mRNA in macrophages is enhanced by interleukin-1beta, tumor necrosis factor alpha, and lipopolysaccharide. J LeukocBiol 1998;63:606–11.

28. Qi XF, Kim DH, Yoon YS, Li JH, Song SB, Jin D, et al. The adenylylcyclase-cAMP system suppresses TARC/CCL17 and MDC/CCL22production through p38 MAPK and NF-kappaB in HaCaT keratino-cytes. Mol Immunol 2009;46:1925–34.

29. Fauriat C, Long EO, Ljunggren HG, Bryceson YT. Regulation of humanNK-cell cytokine and chemokine production by target cell recognition.Blood 2010;115:2167–76.

30. Mariani M, Lang R, Binda E, Panina-Bordignon P, D’Ambrosio D.Dominance of CCL22 over CCL17 in induction of chemokine receptorCCR4 desensitization and internalization on human Th2 cells. Eur JImmunol 2004;34:231–40.

31. Britschgi MR, Link A, Lissandrin TK, Luther SA. Dynamic modulationof CCR7 expression and function on naive T lymphocytes in vivo.J Immunol 2008;181:7681–8.

32. Otero C, Groettrup M, Legler DF. Opposite fate of endocytosed CCR7and its ligands: recycling versus degradation. J Immunol 2006;177:2314–23.

33. Maheshwari A, Christensen RD, Calhoun DA. ELRþ CXC chemokinesin human milk. Cytokine 2003;24:91–102.

34. Takahata Y, Takada H, Nomura A, Nakayama H, Ohshima K, Hara T.Detection of interferon-gamma-inducible chemokines in human milk.Acta Paediatr 2003;92:659–65.

35. Jones RL, Morison NB, Hannan NJ, Critchley HO, Salamonsen LA.Chemokine expression is dysregulated in the endometrium of womenusing progestin-only contraceptives and correlates to elevated re-cruitment of distinct leukocyte populations. Hum Reprod 2005;20:2724–35.

36. Sekiya T, Miyamasu M, Imanishi M, Yamada H, Nakajima T, Yama-guchi M, et al. Inducible expression of a Th2-type CC chemokinethymus- and activation-regulated chemokine by human bronchialepithelial cells. J Immunol 2000;165:2205–13.

37. Qin XJ, Shi HZ, Deng JM, Liang QL, Jiang J, Ye ZJ. CCL22 recruitsCD4-positive CD25-positive regulatory T cells into malignant pleuraleffusion. Clin Cancer Res 2009;15:2231–7.

38. Gros E, Bussmann C, Bieber T, Forster I, Novak N. Expression ofchemokines and chemokine receptors in lesional and nonlesionalupper skin of patients with atopic dermatitis. J Allergy Clin Immunol2009;124:753–60.

39. Bonecchi R, Bianchi G, Bordignon PP, D’Ambrosio D, Lang R, BorsattiA, et al. Differential expression of chemokine receptors and chemo-tactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J ExpMed 1998;187:129–34.

40. Fukuda K, Fujitsu Y, Seki K, Kumagai N, Nishida T. Differentialexpression of thymus- and activation-regulated chemokine(CCL17) and macrophage-derived chemokine (CCL22) by humanfibroblasts from cornea, skin, and lung. J Allergy Clin Immunol2003;111:520–6.

41. Jordan JT, Sun W, Hussain SF, DeAngulo G, Prabhu SS, HeimbergerAB. Preferential migration of regulatory T cells mediated by glioma-secreted chemokines can be blocked with chemotherapy. CancerImmunol Immunother 2008;57:123–31.

42. Luci C, Tomasello E. Natural killer cells: detectors of stress. Int JBiochem Cell Biol 2008;40:2335–40.

43. Madjd Z, Spendlove I, Moss R, Bevin S, Pinder SE, Watson NF, et al.Upregulation of MICA on high-grade invasive operable breast carci-noma. Cancer Immun 2007;7:17.

44. Park SW, Bae JH, Kim SD, Son YO, Kim JY, Park HJ, et al. Com-parison of level of NKG2D ligands between normal and tumor tissueusing multiplex RT-PCR. Cancer Invest 2007;25:299–307.

45. Mailloux AW, Young MR. NK-dependent increases in CCL22 secre-tion selectively recruits regulatory T cells to the tumor microenviron-ment. J Immunol 2009;182:2753–65.

46. Chomarat P, Rissoan MC, Banchereau J, Miossec P. Interferongamma inhibits interleukin 10 production by monocytes. J ExpMed 1993;177:523–7.

47. Izcue A, Coombes JL, Powrie F. Regulatory lymphocytes and intes-tinal inflammation. Annu Rev Immunol 2009;27:313–38.

48. Sangaletti S, Tripodo C, Ratti C, Piconese S, Porcasi R, Salcedo R,et al. Oncogene-driven intrinsic inflammation induces leukocyte pro-duction of tumor necrosis factor that critically contributes to mam-mary carcinogenesis. Cancer Res 2010;70:7764–75.

49. Ralainirina N, Poli A, Michel T, Poos L, Andres E, Hentges F, et al.Control of NK cell functions by CD4þCD25þ regulatory T cells. JLeukoc Biol 2007;81:144–53.

Faget et al.





2011;71:6143-6152. Published OnlineFirst August 18, 2011.Cancer Res Julien Faget, Cathy Biota, Thomas Bachelot, et al.

Recruitment through CCL22 Production by Tumor CellsregEarly Detection of Tumor Cells by Innate Immune Cells Leads to T

Updated version

10.1158/0008-5472.CAN-11-0573doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2011/08/18/0008-5472.CAN-11-0573.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/71/19/6143.full#ref-list-1

This article cites 49 articles, 23 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/71/19/6143.full#related-urls

This article has been cited by 10 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/71/19/6143To 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-11-0573http://cancerres.aacrjournals.org/content/suppl/2011/08/18/0008-5472.CAN-11-0573.DC1http://cancerres.aacrjournals.org/content/71/19/6143.full#ref-list-1http://cancerres.aacrjournals.org/content/71/19/6143.full#related-urlshttp://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/71/19/6143http://cancerres.aacrjournals.org/

earlydetectionoftumorcellsbyinnateimmunecellsleads to t ......in t reg trafficking and accumulation...

Documents