Suppression of Intratumoral CCL22 by Type I Interferon Inhibits Migration

Published OnlineFirst October 2, 2015; DOI: 10.1158/0008-5472.CAN-14-3499 Cancer Microenvironment and Immunology Research

Suppression of Intratumoral CCL22 by Type I Interferon Inhibits Migration of Regulatory T Cells and Blocks Cancer Progression David Anz1,2, Moritz Rapp1, Stephan Eiber1,2, Viktor H. Koelzer1, Raffael Thaler1, Sascha Haubner1, Max Knott1, Sarah Nagel1, Michaela Golic1, Gabriela M. Wiedemann1, Franz Bauernfeind3, Cornelia Wurzenberger1, Veit Hornung3,4, Christoph Scholz5, Doris Mayr6, Simon Rothenfusser1, Stefan Endres1, and Carole Bourquin1,7

Abstract

The chemokine CCL22 is abundantly expressed in many tion. Mechanistic investigations indicated that Treg blockade types of cancer and is instrumental for intratumoral recruit- was a consequence of reduced intratumoral CCL22 levels ment of regulatory T cells (Treg), an important subset of caused by type I IFN. Notably, stable expression of CCL22 immunosuppressive and tumor-promoting lymphocytes. In abrogated the antitumor effects of treatment with RLR or this study, we offer evidence for a generalized strategy to TLR ligands. Taken together, our ﬁndings argue that type I blunt Treg activity that can limit immune escape and promote IFN blocks the Treg-attracting chemokine CCL22 and thus tumor rejection. Activation of innate immunity with Toll-like helps limit the recruitment of Treg to tumors, a ﬁnding with receptor (TLR) or RIG-I–like receptor (RLR) ligands prevented implications for cancer immunotherapy. Cancer Res; 75(21); 1– accumulation of Treg in tumors by blocking their immigra- 11. 2015 AACR.

Introduction effector T-cell function ex vivo (1). Indeed, suppression of anticancer immunity is mediated predominantly by intratumoral Treg The chemokine CCL22 is abundantly expressed in the tissue of that suppress CD8 T-cell responses locally at the tumor site (6). many types of human cancer (1, 2). Its cognate receptor is CCR4, a þ High numbers of tumor-infiltrating FoxP3 Treg correlate with transmembrane protein expressed predominantly and constitu- poor prognosis and have been identified as a significant predictor tively by regulatory T cells (Treg), a subpopulation of immuno- of patient death in several types of human cancer (1, 2). suppressive T lymphocytes (3). Several studies show that CCL22 The aim of cancer immunotherapy is to promote antitumor leads to the recruitment of Treg to the cancer tissue (1, 2). immunity and to overcome tumor-induced immunosuppression. Although the role of Treg in cancer remains to be definitely Activation of the innate immune system with ligands for pattern established, many reports on human and murine cancers dem- recognition receptors (PRR) or with cytokines such as IFNa can onstrate a tumor-promoting effect: Treg progressively accumulate suppress tumor growth (7, 8). In mice, CpG oligonucleotides in the blood and lymphoid organs (4) and abundantly infiltrate (CpG) that stimulate the Toll-like receptor 9 (TLR9) improve the the tumor tissue itself (5). These tumor-infiltrating Treg that efficacy of anticancer vaccines and can also be used as single agent express CD25 and the transcription factor FoxP3 strongly suppress to reduce tumor size (9, 10). In humans, imiquimod, a synthetic agonist for TLR7, is used to treat basal cell carcinoma and vulvar intraepithelial neoplasia, and the cytokines IL2 and IFNa are used 1Center of Integrated Protein Science Munich (CIPS-M), Division of for selected patients with melanoma or renal cell cancer (11, 12). € € Clinical Pharmacology, Klinikum der Universitat Munchen, Munich, Furthermore, many TLR activators are currently under investiga- Germany. 2Medizinische Klinik und Poliklinik IV, Klinikum der Uni- versitat€ Munchen,€ Munich, Germany. 3Institute of Molecular Medicine, tion in clinical trials (13). Universitätsklinikum Bonn, Bonn, Germany. 4Gene Center and Depart- The impact of PRR-activating ligands on Treg in the tumor- ment of Biochemistry, Ludwig-Maximlians-Universität, Munich, Ger- bearing host is still unclear. It has been shown that PRR activation many. 5Department of Obstetrics and Gynecology, Ulm University Medical Centre, Ulm, Germany. 6Department of Pathology, Ludwig- can prevent Treg-mediated suppression of T effector cell prolif- Maximilians-Universitat,€ Munich, Germany. 7Department of Medicine, eration in the lymph node (14, 15), but an effect of PRR activation University of Fribourg, Fribourg, Switzerland. on Treg in the tumor tissue has so far not been described. In Note: Supplementary data for this article are available at Cancer Research particular, it is unknown whether trafficking and tumor infiltra- Online (http://cancerres.aacrjournals.org/). tion of Treg can be affected by agonists for TLRs and RIG-I–like D. Anz and M. Rapp contributed equally to this article. helicases. Corresponding Author: Stefan Endres, Division of Clinical Pharmacology, Lindwurmstrasse 2a, 80337 Munich, Germany. Phone: 49-89-440057300; Fax: Materials and Methods 49-89-440057330; E-mail: [email protected] Mice and cell lines doi: 10.1158/0008-5472.CAN-14-3499 Female BALB/c and C57BL/6 mice were from Harlan-Winkel- 2015 American Association for Cancer Research. mann. IFNI receptor (IFNAR)-deficient mice on C57BL/6

www.aacrjournals.org OF1

Anz et al.

background were kindly provided by Dr. Z. Waibler (Paul-Ehrlich All patients had been treated surgically between 2004 and 2008 at Institute, Langen, Germany). Mice were 5 to 10 weeks of age at the the same institution (Department of Obstetrics and Gynecology onset of experiments. Animal studies were approved by the local Maistrasse, University of Munich, Munich, Germany). Tissue regulatory agency (Regierung von Oberbayern, Munich, Ger- was stained with the following unconjugated primary antibodies many). The human cell lines Jurkat, A-375, A-431, A-549, and the respective isotype control antibodies: rabbit anti-hCCL22 CAMA-1, MCF-7, MDA-MB-231, MDA-MB-435s, Panc1, SK-BR- (Peprotech), mouse anti-hCD14 (Novocastra), mouse anti- 3, and SW480 and the murine cell lines 4T1, CT26, EL-4 (EG-7), hCD68 (Dako), rat anti-hDC-LAMP (Dendritics), mouse anti- and B16-F1 were obtained from ATCC, where short tandem hDC-SIGN (Abcam), and purified rabbit IgG (Biozol). For detec- repeat (STR) analysis is used for authentication and were used tion, the following secondary antibodies were used: biotinylated within 6 months after resuscitation (ATCC). IMIM-PC1 was goat anti-rabbit (Vector), Cy2 goat anti-rat, and Alexa Fluor 488 kindly provided by Prof. P. Michl (University of Marburg, Mar- donkey anti-mouse (both Jackson ImmunoResearch). For imag- burg, Germany), Panc02 by Prof. C. Bruns, MethA by Prof. ing, a confocal laser scanning microscope (LSM 510, Carl Zeiss) W. Zimmermann, and mGC8 by Dr. J. Nockel€ (all Klinikum der was used and images were processed using Adobe Photoshop for Universit€at Munchen,€ Munich, Germany). Cell lines were authen- adjustment of contrast and size. ticated using STR (LGC Standards) and were cultured in complete DMEM or RPMI medium (PAA Laboratories) and routinely tested Flow cytometry and adoptive T-cell transfer for mycoplasma contamination by MycoAlert Mycoplasma Tumors were mechanically disrupted, incubated with 1 mg/mL Detection Kit (LONZA). The CT26-CCL22dox cell line was gen- collagenase and 0.05 mg/mL DNAse (both Sigma Aldrich), and erated by lentiviral transduction with a construct containing a passed through a cell strainer. Single-cell suspensions were resus- doxycycline (Dox)-inducible CCL22 expression cassette as pended in 44% Percoll (Biochrome) and layered over 67% Percoll described (16). For tumor induction, 0.25 106 (CT26 or prior to centrifugation at 800 g for 30 minutes. Lymphocytes CT26-CCL22dox), 1 106 (B16 and Panc02), and 7.5 106 from the interphase were stained with anti-CD3-PerCP, anti-CD4- (EG-7) tumor cells were injected subcutaneously into the flank. APC, anti-CD103-Fitc (all BD Biosciences) and anti-CCR4-Pe Mice with subcutaneous CT26-CCL22dox tumors were fed with a (Biolegend) followed by intracellular detection of FoxP3 using normal or 25 mg/kg doxycycline-containing diet (ssniff Spezial- either anti-FoxP3-PE or anti-FoxP3-Pacific Blue antibody and di€aten GmbH). Tumor size was expressed as the product of the premixed regulatory T-cell staining reagents (Ebioscience). Events perpendicular diameters of individual tumors (mm2). were measured on a FACS Canto II flow cytometer (BD Bio- sciences) and analyzed with FlowJo software (TreeStar). For Application of TLR ligands adoptive transfer, single-cell suspensions from the spleen and 0 The fully PTO-modified CpG oligonucleotide 1826 (5 -TCCAT- lymph nodes of healthy mice were labeled with carboxyfluores- 0 GACGTTCCTGACGTT-3 ) (Coley Pharmaceutical Group) was cein diacetate succinimidyl ester (CFSE; Invitrogen) according injected subcutaneously either peritumorally or in the contralat- to the manufacturer's instructions and injected intravenously eral flank (100 mg CpG in PBS). EG-7 tumor–bearing mice were (3 107 cells per recipient mouse) into tumor-bearing recipients injected with 10 mg CpG complexed with 50 mg of liposomal (mean tumor size, 80 mm2). transfection reagent DOTAP (Roche). Poly(I:C) (Amersham Bioscience) was applied intraperitoneally (250 mg). Lipopoly- Cytokine assays of tissue lysates saccharide (LPS; Sigma-Aldrich) and resiquimod (R848; Alexis Tissue homogenates were resuspended in lysis buffer (BioRad Biochemicals) were injected subcutaneously into the flank con- Laboratories) and centrifuged. Total protein concentration was tralaterally to the tumor (20 mg). measured by Bradford assay (BioRad Laboratories). All samples were diluted to a protein concentration of 10 mg/mL, and CXCL12, Immunohistology and cell counting CCL21, CCL17, and CCL22 concentrations were measured by ELISA To permit the evaluation of highly standardized tissue areas, (R&D Systems). The final cytokine concentration was calculated as mouse tumor samples were bisected perpendicular to the skin at nanograms cytokine per gram protein in the respective lysate. their widest diameter to obtain cryosections (5 mm) derived from fi the tumor center. The following primary antibodies were used: rat Isolation of tumor-in ltrating leukocytes from human ovarian anti-mFoxP3 (eBioscience), rat anti-mCD4 (Biolegend), Syrian cancer hamster anti-mCD3, and rat anti-mCD8. Cy3 goat anti-Syrian Pieces of freshly isolated ovarian cancer specimens resected by hamster IgG, biotinylated donkey anti-rat IgG, rhodamin red X open surgery were mechanically disrupted, incubated with 1 mg/mL streptavidin, alkaline phosphatase streptavidin (all Jackson collagenase and 0.05 mg/mL DNAse (both Sigma Aldrich), ImmunoResearch) and Alexa-488 or -633 streptavidin (Invitro- washed and incubated in PBS containing 2 mmol/L EDTA for gen) were used as detection reagents. Infiltrating cells were 20 minutes. After washing, a second incubation with collagenase assessed by systematically counting nonoverlapping visual fields and DNAse was performed before passing the material through a (high-power fields, hpf) of the entire tumor section using a cell strainer. The obtained single-cell suspension was washed and fluorescence microscope equipped with an 40 oil immersion used for cell culture or magnetic cell separation. The study was objective (Carl Zeiss). Necrotic areas and blood vessels were approved by the local ethics committee. excluded from evaluation. Counting was performed blinded by two independent investigators. Immunohistology on human In vitro TLR stimulation of tumor-infiltrating cells and in vitro paraffin-embedded sections was performed on biopsy tissue of AH1 antigen stimulation breast cancer tissue that was retrieved from the archives of the Tumor-infiltrating murine leukocytes were isolated as described Institute of Pathology of the Ludwig-Maximilians-Universit€at above for flow cytometry and cells were treated with 5 mg/mL (Munich, Germany). Age of patients ranged from 26 to 80 years. CpG 1826 (Coley Pharmaceutical Group), 1000 U/mL murine

OF2 Cancer Res; 75(21) November 1, 2015 Cancer Research

CCL22 Suppression Blocks Cancer Progression

IFNa (R&D Systems), 50 ng/mL murine IFNg (PeproTech), or 10 tumors treated with CpG peritumorally and upon CpG injection ng/mL murine IL1b, IL2, IL6, IL10, or IL12 (all PeproTech). Human distant from the tumor. The decrease was selective for Treg, as total þ tumor cell lines and single-cell suspensions derived from ovarian infiltrating CD3 T cells were in contrast increased after CpG cancer specimens were treated with 100 ng/mL human IFNg treatment (Fig. 1A and B). Selectivity for Treg was also shown þ (PeproTech), 5 mg/mL CpG2006 (Invivogen), 0.5 mg/mL R848 within the CD4 T-cell population as indicated by a decreased (Alexis Biochemicals) or transfected with 1,000 ng/mL poly FoxP3 to CD4 ratio (Supplementary Fig. S1). Using flow cyto- (I:C) (Invitrogen), 1,000 ng/mL 50-triphosphate RNA (50P- metry, we confirmed the decrease of intratumoral Treg, demon- þ GCAUGCGACCUCUGUUUGA-30) (synthesis with the Megashort- strated by a reduction in FoxP3 cells per gram tumor tissue and a script Kit, Ambion Life Technologies), or with 1,000 ng/mL of a proportional decrease within the CD3 and CD4 T cells (Fig. 1C; nonstimulating control RNA (50-AAAAAAAAAAAAAAAAAAA-30) Supplementary Fig. S1). Four days after a single CpG injection, the (CureVac). Prior to cell transfection, poly (I:C) and the RNA was absolute number of intratumoral Treg was still strongly decreased. complexed in 50-mL Opti-MEM medium (Invitrogen) containing Treg numbers returned to the level of untreated mice 8 days after 1% Lipofectamine 2000 (Invitrogen). To quantify antigen speci- one CpG injection, although the Treg to CD8 ratio remained ficity, tumor-infiltrating T cells were stimulated with AH1 peptide strongly decreased (Supplementary Fig. S1C). Decreased Treg cell (2 mmol/L, SPSYVYHQF, Anaspec) for 4 hours in presence of infiltration was not limited to the CT26 model, as a reduction in Brefeldin A (5 mg/mL), fixed and permeabilized (eBioscience tumor-infiltrating Treg was observed upon CpG treatment in two Staining Buffer Set), and incubated with anti-IFNg-PE-Cy7 anti- additional tumor models in mice of different genetic background body (Biolegend). (Supplementary Fig. S2). To test whether suppression of intratumoral Treg infiltration can be mediated by activation of other PRR, Cell sorting we injected mice with the MDA-5 and TLR3 ligand poly(I:C) (17), Microbead-based CD11c and CD8 T-cell isolation kits (Miltenyi the TLR7 ligand R848, and the TLR4 ligand LPS. Infiltrating þ þ Biotec) were used to isolate highly pure dendritic cells (DC; >95%) FoxP3 and CD3 cells were quantified in 750 visual fields and CD8 T cells (>90%) from single-cell suspensions of subcuta- of 23 tumors. As seen after CpG application, treatment with neous murine tumors. APC-labeled anti-CD11c antibody (Biole- poly(I:C) and R848 led to a striking reduction in tumor- þ gend) was used to confirm DC purity via FACS analysis. To isolate infiltrating FoxP3 cells with a strong decrease of the FoxP3 þ human CD14 cells, CD14 microbeads were used (Miltenyi Biotec). to CD3 ratio (Fig. 1D). No significant reduction was seen upon treatment with LPS, which may be due to the fact that LPS also RNA isolation and quantitative PCR analysis of tumor- induces Treg proliferation (18). In conclusion, PRR activation fi in ltrating cells with CpG, poly(I:C), or R848 clearly and selectively decreases fi Total RNA was extracted from tumor-in ltrating cells using the number of Treg within tumor tissues. High Pure RNA Isolation Kit (Qiagen). One microgram of RNA was converted to cDNA using the Revert Aid First strand cDNA Innate immune activation inhibits the migration of regulatory Synthesis Kit (Fermentas). Quantitative real-time PCR amplifica- T cells into the tumor tion was performed with the Light Cycler TaqMan Master (Roche To examine whether the decreased numbers of intratumoral þ Diagnostics) on a LightCycler 2.0 instrument (Roche Diagnostics) Treg result from a reduced migration of FoxP3 cells into the together with the Universal Probe Library System (Roche Diag- tumor, we analyzed the trafficking of adoptively transferred Treg nostics). Relative gene expression is shown as a ratio of the in tumor-bearing mice that had been treated with CpG. Mice expression level of the gene of interest to that of hypoxanthine bearing established CT26 tumors received one injection of CpG. phosphoribosyltransferase (HPRT) RNA. Primers were obtained Twenty-four hours later, CFSE-labeled cells from the spleen and from Metabion. lymph nodes of healthy donor mice were transferred intraven- þ ously. Tumors were resected 18 hours after transfer and CFSE T Statistical analysis cells were analyzed by flow cytometry (Fig. 2A). Strikingly, in the All data are presented as mean SEM and were analyzed as tumors of CpG-treated animals, the fraction of transferred Treg þ þ appropriate by the unpaired Student t test or by ANOVA test using within intratumoral CFSE CD3 T cells was reduced on average – – the Student Newman Keuls correction. Correlations were ana- almost 4-fold compared with untreated mice (Fig. 2A and B). In lyzed 2-tailed by Pearson correlation. Statistical analysis was contrast, no difference in the proportion of transferred Treg was performed using GraphPad Prism 5 (GraphPad Software). observed in the spleen. Within the intratumoral CD4 T cells, the þ proportion of transferred FoxP3 Treg was also reduced, whereas Results no decrease was observed for FoxP3neg T effector cells (Fig. 2C), Innate immunostimulation reduces tumor infiltration by Treg indicating a specific inhibition of Treg migration. In conclu- To investigate the effect of PRR stimulation on tumor infiltra- sion, these findings demonstrate that CpG suppresses the migra- þ tion by Treg, we quantified these cells in CT26 tumors from mice tion of FoxP3 Treg into the tumor. treated with the TLR9 ligand CpG. Intratumoral Treg and total þ CD3 T cells were visualized by immunofluorescence and Innate immune activation suppresses tumor-associated CCL22 counted according to a highly standardized protocol (see Materi- in mice and humans als and Methods). A majority of tumors from untreated mice The migration and retention of Treg in peripheral tissues is showed prominent Treg cell infiltrations with formation of focal mediated by the interaction of specific integrins and chemokine clusters (analysis of 384 visual fields from tissue sections of 11 receptors with their respective ligands (19). To characterize the individual tumors; Fig. 1A and B). In mice treated with CpG, Treg molecular mechanisms responsible for the decrease in Treg migra- were almost completely absent from the tumors (analysis of 160 tion into the tumor upon treatment with TLR ligands, we analyzed visual fields from 12 tumors). This effect was observed both in the expression of several Treg-related chemokines and chemokine

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 OF3

Anz et al.

A Untreated B

FoxP3 CD3 Untreated CpG p.t. CpG c.l. 40 FoxP3 ** * Untreated 30 CpG p.t. 20 CpG c.l. 30 mm field

2 10

CpG peritumoral (p.t.) 0.2 150 * CD3 100

Cells/0.25 mm 0.1 50

FoxP3:CD3 ratio ** **

30 mm 21 28 32 33 50 60 16124 11 23 8 41 27 13 38 41 17 18 10 12 15 14 (Counted fields/tumor)

C Untreated CpG p.t. Untreated 400 30 105 105 CpG p.t. + 34.0 10.8 25 104 104

cells 300 of +

+ 20 103 103 * * 200 15 2 2 10 10 FoxP3 0 0 + per g tumor 10 %FoxP3 2 3 4 5 2 3 4 5 100

010 10 10 10 010 10 10 10 intratumoral CD3 CD4 5 CD4 FoxP3 0 0

D Ø = Untreated Untreated Poly (I:C) R848 LPS P = Poly (I:C)

FoxP3 ** * R = R848 80 L = LPS

field 60

2 0.3

40 * 0.2 * 20 Cells/0.25 mm 0.1 FoxP3:CD3 ratio

24 20 19 32 13 28 39 64 18 14 15 19 Ø PLR 45 50 2924 20 3525 45 50 57 65 (Counted fields/tumor)

Figure 1. Ligands for TLR and RLR prevent tumor infiltration by Treg. Mice bearing CT26 tumors with an average size of 60 mm2 were treated three times at 2-day intervals with CpG, either peritumorally (p.t., n ¼ 8) or contralaterally (c.l., n ¼ 4), poly(I:C) (n ¼ 6), R848 (n ¼ 6), or LPS (n ¼ 5) or remained untreated (n ¼ 18). Two days after the last injection, tumors were prepared for histology and stained for FoxP3 and CD3 or analyzed by flow cytometry. A, representative tissue areas of þ þ þ þ untreated and CpG-treated CT26 tumors, FoxP3 (green), and CD3 (red). B, infiltrating FoxP3 and CD3 cells were counted in each tumor sample as described in Materials and Methods. Each data point represents the number of positive cells in one visual field (0.25 mm2). Each dataset (vertically aligned dots) represents a tumor from an individual mouse. Numbers at the bottom of each dataset indicate the number of visual fields counted per tumor. Bars indicate the mean cell count per visual field in one tumor. The ratio of FoxP3þ to CD3þ cells is indicated (histogram). C, the absolute number of tumor-infiltrating CD4þ FoxP3þ Treg per gram tumor and the proportion of FoxP3þ cells within tumor-infiltrating CD3þ T cells were determined by flow cytometry. Data for one representative mouse per group and average of all mice per group (n ¼ 7) are shown. D, CT26 tumors from poly(I:C)-, R848-, and LPS-treated mice were analyzed for FoxP3þ and CD3þ cell infiltration by immunohistology. Data are represented as in B. Error bars, SEM. P values were calculated relative to untreated mice (, P < 0.05; , P < 0.01).

OF4 Cancer Res; 75(21) November 1, 2015 Cancer Research

CCL22 Suppression Blocks Cancer Progression

FACS of tumor CCL22, and CXCL12 have been described to mediate homing 105 of Treg in different circumstances (19, 21–23). In addition, the 104 A d 0 d 21 d 22 +18 h chemokine CCL22 has been associated with homing of Treg to 103 malignant tumors (1, 19). We analyzed intratumoral levels of 102 Tumor CpG Transfer 0 all these chemokines in CT26 tumors of CpG-treated mice by 0102 103 104 105

CD3 ELISA. Strikingly, we found a strong suppression of CCL22 CFSE levels in the tumor tissue upon treatment with the TLR ligand Untreated CpG (Fig. 3B). A decrease was also observed for CCL17, although 105 105 15.3 3.8 this chemokine was expressed only at very low levels. No 104 4 10 suppression was seen for CCL5, CCL21, or CXCL12 (Fig. 103 103 3B). Substantial intratumoral levels of CCL22 were observed

102 102 in three other subcutaneous murine tumor models, and treat- 0 0 ment with CpG decreased intratumoral CCL22 in all three 0102 103 104 105 0102 103 104 105

CD4 models (Fig. 3C). Furthermore, suppression of intratumoral FoxP3 CCL22 was also seen upon injection of poly(I:C), R848, or B CFSE+ T cells LPS, showing that this also results from activation of other Tumor Spleen Untreated PRR (Fig. 3D). Interestingly, we observed a linear correlation between intra- 12 12 CpG þ tumoral CCL22 and the number of inﬁltrating FoxP3 cells 10 10 (Supplementary Fig. S3). To examine whether the reduction of 8 8 intratumoral Treg by TLR ligands is due to suppression of CCL22, 6 6 we established a CT26 tumor cell line with inducible expression of dox % Treg CCL22 (CT26-CCL22 ). In this model, exposure of the tumor 4 ** 4 cells to doxycycline induces secretion of CCL22 (Supplementary 2 2 Fig. S4A). When CT26-CCL22dox cells were implanted subcuta- 0 0 neously in mice, addition of doxycycline to the diet of the animals led to increased intratumoral CCL22 production and enhanced ﬁ C CFSE+ CD4 T cells in ltration of tumors with Treg (Supplementary Fig. S4B). Doxy- Tumor cycline-dependent induction of CCL22 expression in these ** Untreated tumors was used to prevent the suppression of intratumoral CCL22 by TLR ligands. Indeed, no suppression of CCL22 was CpG 15 80 observed upon treatment with CpG in mice bearing subcutaneous CT26-CCL22dox tumors that received doxycycline (Fig. 3E, left). 60 Strikingly, in this group of mice, no reduction of intratumoral Treg 10 was observed, demonstrating that suppression of CCL22 is indeed % Treg ** 40

% T effector responsible for the inhibition of Treg migration upon TLR stim- 5 20 ulation (Fig. 3E, right). We next examined whether suppression of tumor-associated 0 0 CCL22 is also a phenomenon that can be observed in humans. We Figure 2. prepared single-cell suspensions from fresh human ovarian cancer Treatment with CpG inhibits migration of FoxP3þ cells into the tumor. Mice samples and incubated these with ligands for different PRR. High bearing CT26 tumors (average size, 80 mm2) were treated at day 21 after levels of CCL22 were detected in the untreated culture super- tumor induction by one peritumoral injection of CpG (n ¼ 7) or remained natants, whereas application of R848, poly(I:C) and triphosphate n ¼ untreated ( 7). Twenty-four hours later, the animals received an RNA as ligand for RIG-I suppressed CCL22 (Fig. 3F). Interestingly, intravenous transfer of CFSE-labeled cells from the spleen and lymph nodes of healthy donor mice. Tumors were resected 18 hours after transfer, followed CCL22 suppression in the human tumor supernatants was absent by ﬂow cytometric analysis of intratumoral and splenic CFSEþ T cells. A and B, upon stimulation with CpG (Fig. 3F). This is probably due to the þ þ þ þ þ proportion of CFSE CD4 FoxP3 Treg within the CFSE CD3 T cells was more restricted expression pattern of the receptor for CpG, TLR9, analyzed in the tumor and spleen of untreated and CpG-stimulated mice. in humans. Thus, PRR activation suppresses cancer-associated Results are shown for one representative mouse per group (in tumor; A) CCL22 in both mice and humans. and average of all mice per group (n ¼ 7; B). C, analysis of the proportion of transferred FoxP3þ Treg and FoxP3-negative T effector cells within tumor-inﬁltrating CFSEþCD3þCD4þ T cells. Error bars, SEM. P values were Dendritic cells and macrophages produce CCL22 in murine and calculated relative to untreated mice (, P < 0.01). human tumors To elucidate the mechanisms of CCL22 suppression, we next aimed to identify the source of intratumoral CCL22. Culture receptors after TLR stimulation of tumor-bearing mice. Two supernatants of seven different murine tumor cell lines, including receptors expressed on Treg are critically involved in the recruit- the CT26, B16, EG-7, and Panc02 cells, were uniformly negative ment of these cells: the chemokine receptor CCR4 (3) and for CCL22, indicating that the tumor cells themselves do not the integrin CD103 that mediates retention of Treg within periph- produce the chemokine (Fig. 4A). As in secondary lymphatic þ eral tissues (20). Treatment of tumor-bearing mice with CpG did organs of mice CD11c dendritic cells are the producers of CCL22 þ not modify the expression of CCR4 or CD103 on FoxP3 cells (24), we hypothesized that these cells could also be responsible (Fig. 3A). Several chemokines such as CCL5, CCL17, CCL21, for the production of CCL22 in tumors. To test this hypothesis, we

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 OF5

Anz et al.

A Untreated B Untreated CpG CpG CD103 CCR4 CCL5 CCL17 CCL21 CCL22 CXCL12 ns ns 14 80 10 10 150 0.8 12 20 8 8 60 0.6 10 100 6 6 8 * ** 40 0.4 6 10 4 4 (ng/g protein) 50 % Positive Treg % Positive 4 20 0.2 2 2 2 0 0 0 0 0 0 0

C Untreated D CpG Ø = Untreated 40 300 40 20 P= Poly (I:C) R = R848 30 30 ** 15 200 L= LPS

20 20 10

100 ** * * *

(ng/g protein) 10 ** 10 5 (% of untreated) Intratumoral CCL22 Intratumoral CCL22 0 0 0 0 B16 Panc02 EG-7 ØPR L

E ns CT26-CCL22dox Untreated F CpG

1,000 * ns 150 100 100 ** 800 * 600 100 * 50 ** 50 400 ** 50 (ng/g protein) (% of untreated) Secreted CCL22 Secreted CCL22 (% of untreated) (% of untreated) 200 Intratumoral Treg ** Intratumoral CCL22

0 0 0 0 Ø Dox ØDox CpG R848 Tumor Medium 3p-RNA Poly (I:C) Untreated Control RNA

Figure 3. TLR stimulation suppresses intratumoral CCL22 in mice and humans. A and B, CT26 tumor–bearing mice were treated with CpG as described in Fig. 1 (n ¼ 6 mice per þ þ group). A, expression of CD103 and CCR4 on CD4 FoxP3 splenic Treg was assessed by flow cytometry 2 days after the last injection. B, homogenates of the tumor tissue were analyzed for the indicated chemokines by ELISA. C, intratumoral CCL22 levels of CpG-treated or untreated subcutaneous B16 (n ¼ 7), EG-7 (n ¼ 5), and Panc02 (n ¼ 7) tumors were measured by ELISA in tissue homogenates. D, CT26 tumor–bearing mice were treated with poly(I:C), R848, or LPS as in Fig. 1 and tumors were analyzed for CCL22 levels by ELISA (n ¼ 7 mice per group). E, mice (n ¼ 7 per group) bearing subcutaneous CT26-CCL22dox tumors were fed with a normal or doxycycline-containing diet. At day 21 after tumor induction, both groups were either treated with CpG (three times at þ a 3-day interval) or remained untreated. One day after the last injection, tumors were analyzed for intratumoral CCL22 and FoxP3 T cells by ELISA and flow cytometry, respectively. The levels of CCL22 and Treg in mice with untreated CT26-CCL22dox tumors were set to 100%. F, single-cell suspensions from freshly isolated human ovarian cancer specimens were incubated with CpG, R848, poly(I:C), 50-triphosphate RNA (3p-RNA), a nonactivating control RNA, or remained untreated. Error bars, SEM. P values were calculated relative to untreated samples (, P < 0.05; , P < 0.01; ns, not significant).

OF6 Cancer Res; 75(21) November 1, 2015 Cancer Research

CCL22 Suppression Blocks Cancer Progression

AB C CT26 B16 1.0 0.8 2.0 0.4

0.8 * 0.6 *** 1.5 0.3 0.6 0.4 1.0 0.2 0.4 CCL22/HPRT CCL22 (ng/mL)

0.2 0.5 CCL22 (ng/mL) 0.1 0.2

0 0.0 0.0 0 − − + + + 4T1 B16 EG7 CD8 CT26 IMIM mGC8 MethA A-375 A-549 A-431 BM-DC Jurkat Panc02 CD11c CD11c CD11c CD11c MCF-7 Karpas SW480 Unsorted Unsorted Panc01 SKBR-3 CAMA-1 SK-MEL23 MDA-MB-231 MDA-MB-435s

D CCL22 Isotype E 1,000 ** 800

600 50 mm 50 mm 400

CCL22 (pg/mL) 200

0 − + All 10 mm CD14 CD14

F CD14 CCL22 Merge Isotype DC-LAMP CCL22 Merge Isotype

15 mm 15 mm

CD68 CCL22 Merge Isotype DC-SIGN CCL22 Merge Isotype

15 mm 15 mm

Figure 4. Dendritic cells and macrophages produce CCL22 in murine and human tumors. A, culture supernatants of murine tumor cell lines and bone marrow–derived þ þ dendritic cells (BM-DC, used as positive control) were analyzed for secretion of CCL22 by ELISA. B, CD11c cells and CD8 T cells were isolated from subcutaneous CT26 and B16 tumors, and CCL22 mRNA levels were quantiﬁed by qRT-PCR. C, CCL22 levels in the culture supernatants of different human tumor cell lines þ were measured by ELISA. D, representative examples of human breast cancer tissue sections were stained for CCL22 (brown cells). E, CD14 cells isolated from single-cell suspensions of freshly dissected ovarian cancer specimens were cultured for 3 days (n ¼ 4 patients). CCL22 was then measured in the culture supernatants. F, breast cancer tissues were prepared for histology and stained for CCL22 (red) and CD14, CD68, DC-LAMP, DC-SIGN (green), or with respective isotype controls. Error bars, SEM. P values of isolated CD11cþ and CD14þ cells were calculated relative to the CD11c- and CD14-depleted fraction, respectively (, P < 0.05; , P < 0.01).

þ isolated CD11c cells from subcutaneous tumors. Expression of source of intratumoral CCL22 in mice (Fig. 4B). As a previous þ CCL22 was observed almost exclusively in the CD11c-positive study (25) suggested a potential role for CD8 intratumoral T cells fraction, indicating that tumor-inﬁltrating dendritic cells are the as source of CCL22, we also isolated CD8 T cells from the tumors

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 OF7

Anz et al.

ACB Untreated Untreated Untreated IFNα IFNα Poly (I:C) 1,000 2,000 5,000 1,500 200 *

800 4,000 1,500 150 1,000 600 3,000 1,000 100 ** 400 ** 2,000 ** 500 ** *

(% of untreated) CCL22 (pg/mL) 500 CCL22 (pg/mL) 50

200 1,000 Intratumoral CCL22

0 0 0 0 0 + − −/− CT26 B16 Panc02 CD11c CD11c WT IFNAR

Figure 5. Suppression of intratumoral CCL22 is mediated by type I IFN. A, tumor-inﬁltrating cells from subcutaneous CT26, B16, and Panc02 tumors. B, CD11c positively sorted single-cell suspensions from Panc02 tumors were treated with IFNa or remained untreated. After incubation for 3 days, CCL22 levels in the supernatants were determined by ELISA. C, subcutaneous B16 tumors were induced in C57BL/6 (n ¼ 10) or IFNAR mice (n ¼ 10). At day 21 after tumor induction, both groups were either treated with poly(I:C) (twice at a 3-day interval) or remained untreated. Two days after the last injection, tumors were prepared and CCL22 levels in the tissue homogenates were measured by ELISA. Error bars, SEM (, P < 0.05; , P < 0.01).

but could not detect CCL22 expression in this fraction (Fig. 4B). tion) suppressed CCL22 secretion by freshly isolated tumor- We further measured CCL22 levels in the supernatant of several infiltrating immune cells, suggesting that a secreted factor human cancer cell lines. No detectable CCL22 levels were seen in was responsible for CCL22 suppression (data not shown). To the majority of samples, except for two breast cancer and one identify this factor, we incubated the tumor-infiltrating immune melanoma cell line that secreted CCL22 (Fig. 4C). Because CCL22 cells with a set of recombinant cytokines that are typically secretion has been reported upon stimulation with IFNg in several released upon innate immune activation (27). A minor breast cancer cell lines (26), we stimulated all human cell lines decrease of CCL22 was seen upon addition of IL10 and IFNg with IFNg. This indeed enhanced the chemokine production of and no change or even an increase occurred with IL1, IL2, IL6, the CCL22-secreting breast cancer and melanoma cells (CAMA-1, and IL12 (data not shown). In contrast, potent suppression of MCF-7, and melanoma A431), but CCL22 secretion remained CCL22 was seen upon treatment of the tumor-infiltrating cells undetectable in all other cell lines tested (Supplementary Fig. S5). from CT26, B16, and Panc02 tumors with IFNa (Fig. 5A). CCL22 As CCL22 was not secreted by the majority of human cancer cells secretion was also suppressed by IFNa in isolated tumor-derived þ tested, it appeared likely that, as in mice, tumor-infiltrating CD11c cells, indicating that dendritic cells are directly targeted immune cells were responsible for the production of intratu- by IFNa (Fig. 5B). To evaluate the contribution of IFNa to moral CCL22. We performed immunohistology of ovarian and the suppression of CCL22 by innate immune activation, we breast cancer tissues and found CCL22 to be expressed by cells induced tumors in interferon receptor–deficient mice (IFNAR) with macrophage-like morphology (Fig. 4D). We therefore and treated these animals by PRR stimulation. Strikingly, no isolated CD14-positive cells from single-cell suspensions of suppression of intratumoral CCL22 and tumor growth was seen fresh ovarian cancer specimens, cultured the cells for 3 days, in IFNAR mice (Fig. 5C; Supplementary Fig. S6A), demonstrat- and determined CCL22 levels in the supernatants. CCL22 ing that type I IFN is a key mediator in the process of CCL22 production was restricted to the CD14-positive fraction, dem- suppression. Treatment of tumors with recombinant IFNa onstrating that these cells represent the main source of CCL22 tended to reduce intratumoral CCL22 and Treg numbers but in these human tumors (Fig. 4E). Double immunofluorescent did not significantly suppress them, indicating a higher efficacy imaging revealed colocalization of CCL22 with CD14 and of PRR-induced IFNa (Supplementary Fig. S6B). CD68, confirming the expression by macrophages (Fig. 4F). Interestingly, some mature dendritic cells, characterized by CCL22 suppression is required for effective antitumor staining of DC-LAMP, were also positive for CCL22, whereas immunotherapy no colocalization was shown for DC-SIGN, a marker for more The outcome of antitumor treatment with CpG is dependent on þ immature dendritic cells (Fig. 4F). In conclusion, intratumoral an effective CD8 T-cell response (ref. 9 and Supplementary Fig. CCL22 is secreted by infiltrating dendritic cells and macro- S7). We examined the contribution of the suppression of intra- phages rather than by the tumor cells themselves. tumoral CCL22 to this therapeutic effect of CpG. We treated mice bearing CT26-CCL22dox tumors with CpG and simultaneously Suppression of intratumoral CCL22 is mediated by type I IFN prevented suppression of CCL22 by feeding with doxycycline. As To investigate the mechanism of CCL22 suppression, we mea- expected, in the absence of doxycycline, a clear suppression of sured the secretion of this chemokine by tumor-infiltrating tumor growth was observed in the animals receiving CpG treat- immune cells under different culture conditions. First, we found ment (Fig. 6). In contrast, CpG treatment did not significantly that culture supernatants from TLR-stimulated splenocytes (the block tumor growth when intratumoral CCL22 secretion was TLR ligand itself was removed by washing 2 hours after stimula- maintained by doxycycline administration. Thus, we demonstrate

OF8 Cancer Res; 75(21) November 1, 2015 Cancer Research

CCL22 Suppression Blocks Cancer Progression

150 CT26-CCL22dox Treg. For instance, activation of TLR7 and TLR9 on dendritic cells renders T effector cells refractory to Treg suppression, whereas Dox stimulation with IFNa or ligands for TLR2 and MDA-5 expressed )

2 by Treg can directly inhibit their suppressive function or suppress 100 ns No Dox their proliferation (15, 32–34). We describe here the blockade of Treg migration as a novel mechanism of action in the process of Dox + CpG innate immune activation by ligands for PRR. We further show * that inhibition of Treg migration indeed contributes to tumor * 50 regression. The efficacy of modulating Treg and T effector cell Tumor size (mm No Dox + CpG activity for immune activation against cancer has recently been highlighted by the success of the CTLA-4 blocking antibody ipilimumab (35). Furthermore, blockade of the Treg-associated 0 receptor CCR4 by a small molecule enhanced tumor immunity 0315 0 (36). Thus, inhibition of Treg migration represents an interesting Time (days) future concept of anticancer immunotherapy. An analysis of several Treg-associated chemokines and homing Figure 6. CCL22 suppression is required for the therapeutic efficacy of CpG. BALB/c receptors during immune activation revealed a potent suppression mice were inoculated with subcutaneous CT26-CCL22dox tumors and fed with of intratumoral levels of CCL22, a chemokine that is crucial for the a normal (n ¼ 28) or doxycycline-containing (n ¼ 28) diet. At day 21 after recruitment of Treg to tumors: In mice, CCL22-mediated Treg tumor induction, both groups were either treated with CpG (three times at a migration has been shown in a model of lung cancer (37), and 3-day interval) or remained untreated. Tumor size was measured every in humans, high levels of intratumoral CCL22 have been observed 2 P second day and expressed as mm . Error bars, SEM. values were calculated in many types of cancer (1, 2, 26, 38). The cellular source of CCL22 for day 30 (, P < 0.05; ns, not significant). in the tumor tissue is however still unclear. We show here that with the exception of some types of breast cancer and melanoma, CCL22 is secreted not by the tumor cells or by tumor-infiltrating T cells as that suppression of CCL22 is necessary for the therapeutic efficacy previously reported (25) but by tumor-infiltrating dendritic cells of TLR ligands. and macrophages. Interestingly, it has been generally assumed that CCL22 secretion is enhanced rather than suppressed upon TLR stimulation (39, 40). This assumption is however based on obser- Discussion vations made using dendritic cells differentiated in vitro (39, 40), in Tumor-infiltrating Treg inhibit antitumor immunity (6) and which we also observed enhanced CCL22 secretion upon TLR are associated with poor prognosis in several types of human stimulation (data not shown). In primary cells isolated from cancer (1, 2). In the present study, we show that stimulation of murine or human tumors, however, we clearly show that CCL22 innate immunity through TLRs and RIG-I–like helicases strong- secretion by dendritic cells and macrophages is suppressed upon þ ly reduces the absolute number of intratumoral FoxP3 Treg in stimulation with CpG and other activators of innate immunity. several mouse tumor models in a highly selective manner. Suppression of CCL22 upon PRR stimulation was dependent þ Indeed, the number of FoxP3-negative effector CD8 T cells on type I IFN. The concept of CCL22 suppression by type I IFN is withinthetumorwasevenincreased.Thisisinaccordancewith supported by previous reports where IFNa-containing combina- previous reports showing higher proportions of T effector cells tions were used: CCL22 was reduced in CD3-stimulated T cells upon treatment with CpG (28). The TLR7 agonist imiquimod upon treatment with IL12 and IFNa (41), in peripheral human increased the T effector cell to Treg ratio in human squamous blood-derived dendritic cells upon incubation with an IFNa- cell cancer but did not alter the proportion of antigen-specificT containing maturation cocktail (42) and in tumor cell cultures effector and Treg cells in the lymph node in a murine antitumor through a COX1 inhibitor and IFNa (43). We clearly demonstrate vaccine model (29, 30). Interestingly, a suppression of Treg here that IFNa as a single agent directly suppresses CCL22 infiltration upon treatment with CpG has been reported in a secretion of tumor-infiltrating immune cells. Although TLR- model of leishmania infection of the skin: TLR stimulation of induced CCL22 suppression critically depends on IFNa, treat- þ the animals reduced the number of FoxP3 cells in the infected ment with recombinant IFNa alone was not as efficient as TLR skin and thus led to clearance of the pathogen (31). We show ligands for CCL22 suppression, a phenomenon that could be here that suppression of intratumoral Treg by TLR agonists explained by the different pharmacokinetics of these agents. A results from a blockade of Treg immigration into the tumor. major contribution of IFNa to the therapeutic efficacy of TLR This is to our knowledge the first report demonstrating a direct ligands such as CpG or poly(I:C) was shown previously and could inhibition of Treg migration upon innate immune activation. be confirmed here (refs. 44, 45 and Supplementary Fig. S6). We Takentogetherwiththeobservation that TLR stimulation now demonstrate that the type I IFN–dependent CCL22 suppres- suppresses Treg infiltration in leishmania infection, these find- sion contributes to the anticancer effect of innate immune acti- ings suggest that activation of the innate immune system vation. Indeed, cancer therapy with IFNa is established for mel- through TLRs may generally prevent Treg migration to sites of anoma and can also be used for other types of human cancer (8). inflammation. The local reduction of Treg in response to Our findings uncover CCL22 suppression as a so far unknown microbial patterns could thus permit an efficient response mechanism of action of anticancer therapy with IFNa. This has against pathogens within the affected tissues. important therapeutic implications as patients with high intra- It was previously demonstrated that ligands for PRR and tumoral CCL22 may derive a higher benefit from IFNa treatment. cytokines such as IFNa affect immunosuppression exerted by It will be interesting to investigate intratumoral CCL22 as a

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 OF9

Anz et al.

predictive biomarker for IFN treatment. In summary, our findings Administrative, technical, or material support (i.e., reporting or organizing identify inhibition of Treg migration by CCL22 suppression as a data, constructing databases): M. Rapp, F. Bauernfeind, D. Mayr so far unknown mechanism of innate immune activation with Study supervision: D. Anz, C. Bourquin important implications for cancer therapy. Acknowledgments Disclosure of Potential Conflicts of Interest The authors thank Nadja Sandholzer and Patrick Layritz for expert technical assistance. V. Hornung has ownership interest (including patents) in RIGONTEC. No potential conflicts of interest were disclosed by the other authors. Grant Support Authors' Contributions This study was supported by grants from LMUexcellent research profes- Conception and design: D. Anz, M. Rapp, V.H. Koelzer, C. Scholz, S. Rothen- sorship (S. Endres), from the Friedrich Baur Foundation (D. Anz), from the fusser, S. Endres, C. Bourquin Deutsche Forschungsgemeinschaft Graduiertenkolleg 1202 (D. Anz, M. Development of methodology: D. Anz, M. Rapp, S. Eiber, V.H. Koelzer, Rapp, V.H. Koelzer, R. Thaler, S. Haubner, C. Wurzenberger, S. Endres, C. R. Thaler, S. Haubner, S. Nagel, M. Golic, F. Bauernfeind, C. Wurzenberger, Bourquin) and DFG AN 801/2-1 (D. Anz, S. Endres), the Deutsche Krebshilfe V. Hornung, C. Scholz 111326 (D. Anz), the excellence cluster CIPS-M 114 (S. Endres), BayImmu- Acquisition of data (provided animals, acquired and managed patients, Net (S. Endres, C. Bourquin), the Swiss National Science Foundation No. provided facilities, etc.): D. Anz, M. Rapp, S. Eiber, V.H. Koelzer, R. Thaler, 138284 and 156372 (C. Bourquin), and Krebsforschung Schweiz 2910-02- S. Haubner, M. Knott, S. Nagel, G.M. Wiedemann, C. Wurzenberger, C. Scholz 2012 (C. Bourquin). Analysis and interpretation of data (e.g., statistical analysis, biostatistics, The costs of publication of this article were defrayed in part by the payment of computational analysis): D. Anz, M. Rapp, S. Eiber, V.H. Koelzer, S. Haubner, page charges. This article must therefore be hereby marked advertisement in M. Knott, S. Nagel, G.M. Wiedemann, C. Wurzenberger, C. Scholz, D. Mayr, accordance with 18 U.S.C. Section 1734 solely to indicate this fact. S. Rothenfusser Writing, review, and/or revision of the manuscript: D. Anz, M. Rapp, V.H. Received November 27, 2014; revised July 3, 2015; accepted August 6, 2015; Koelzer, S. Haubner, C. Scholz, D. Mayr, S. Rothenfusser, S. Endres, C. Bourquin published OnlineFirst October 2, 2015.

References 1. Curiel TJ, Coukos G, Zou L, Alvarez X, Cheng P, Mottram P, et al. 15. Anz D, Koelzer V, Moder S, Thaler R, Schwerd T, Lahl K, et al. Immunos- Specific recruitment of regulatory T cells in ovarian carcinoma fosters timulatory RNA blocks suppression by regulatory T cells. J Immunol immune privilege and predicts reduced survival. Nat Med 2004; 2009;184:939–46. 10:942–9. 16. Bauernfeind F, Rieger A, Schildberg FA, Knolle PA, Schmid-Burgk JL, 2. Gobert M, Treilleux I, Bendriss-VermareN,BachelotT,Goddard-LeonS, Hornung V. NLRP3 inflammasome activity is negatively controlled by Arfi V, et al. Regulatory T cells recruited through CCL22/CCR4 are miR-223. J Immunol 2012;189:4175–81. selectively activated in lymphoid infiltrates surrounding primary breast 17. Anz D, Thaler R, Stephan N, Waibler Z, Trauscheid JM, Scholz C, et al. tumors and lead to an adverse clinical outcome. Cancer Res Activation of melanoma differentiation-associated gene 5 causes rapid 2009;69:2000–9. involution of the thymus. J Immunol 2009;182:6044–50. 3. Iellem A, Mariani M, Lang R, Recalde H, Panina-Bordignon P, Sinigaglia F, 18. Caramalho I, Lopes-Carvalho T, Ostler D, Zelenay S, Haury M, Demengeot et al. Unique chemotactic response profile and specific expression of J. Regulatory T cells selectively express Toll-like receptors and are activated chemokine receptors CCR4 and CCR8 by CD4(þ)CD25(þ) regulatory T by lipopolysaccharide. J Exp Med 2003;197:403–11. cells. J Exp Med 2001;194:847–53. 19. Wei S, Kryczek I, Zou W. Regulatory T-cell compartmentalization and 4. Colombo MP, Piconese S. Regulatory-T-cell inhibition versus depletion: trafficking. Blood 2006;108:426–31. the right choice in cancer immunotherapy. Nat Rev Cancer 2007;7:880–7. 20. Suffia I, Reckling SK, Salay G, Belkaid Y. A role for CD103 in the retention of þ þ 5. Betts GJ, Clarke SL, Richards HE, Godkin AJ, Gallimore AM. Regulating the CD4 CD25 Treg and control of Leishmania major infection. J Immunol immune response to tumours. Adv Drug Deliv Rev 2006;58:948–61. 2005;174:5444–55. 6. Yu P, Lee Y, Liu W, Krausz T, Chong A, Schreiber H, et al. Intratumor 21. Schneider MA, Meingassner JG, Lipp M, Moore HD, Rot A. CCR7 is þ þ þ depletion of CD4 cells unmasks tumor immunogenicity leading to the required for the in vivo function of CD4 CD25 regulatory T cells. J Exp rejection of late-stage tumors. J Exp Med 2005;201:779–91. Med 2007;204:735–45. 7. Kanzler H, Barrat FJ, Hessel EM, Coffman RL. Therapeutic targeting of 22. Wysocki CA, Jiang Q, Panoskaltsis-Mortari A, Taylor PA, McKinnon KP, Su þ þ innate immunity with Toll-like receptor agonists and antagonists. Nat Med L, et al. Critical role for CCR5 in the function of donor CD4 CD25 2007;13:552–9. regulatory T cells during acute graft-versus-host disease. Blood 2005;106: 8. Tarhini AA, Gogas H, Kirkwood JM. IFN-alpha in the treatment of mela- 3300–7. noma. J Immunol 2012;189:3789–93. 23. Zou L, Barnett B, Safah H, Larussa VF, Evdemon-Hogan M, Mottram P, et al. þ þ 9. Heckelsmiller K, Rall K, Beck S, Schlamp A, Seiderer J, Jahrsdorfer B, et al. Bone marrow is a reservoir for CD4 CD25 regulatory T cells that traffic Peritumoral CpG DNA elicits a coordinated response of CD8 T cells and through CXCL12/CXCR4 signals. Cancer Res 2004;64:8451–5. innate effectors to cure established tumors in a murine colon carcinoma 24. Tang HL, Cyster JG. Chemokine Up-regulation and activated T cell attrac- model. J Immunol 2002;169:3892–9. tion by maturing dendritic cells. Science 1999;284:819–22. 10. Krieg AM. Toll-like receptor 9 (TLR9) agonists in the treatment of cancer. 25. Spranger S, Spaapen RM, Zha Y, Williams J, Meng Y, Ha TT, et al. Up- Oncogene 2008;27:161–7. regulation of PD-L1, IDO, and T(regs) in the melanoma tumor micro- 11. Schon MP, Schon M. TLR7 and TLR8 as targets in cancer therapy. Oncogene environment is driven by CD8(þ) T cells. Sci Transl Med 2013; 2008;27:190–9. 5:200ra116. 12. van Seters M, van Beurden M, ten Kate FJ, Beckmann I, Ewing PC, Eijkemans 26. Faget J, Biota C, Bachelot T, Gobert M, Treilleux I, Goutagny N, et al. Early MJ, et al. Treatment of vulvar intraepithelial neoplasia with topical imi- detection of tumor cells by innate immune cells leads to T(reg) recruitment quimod. N Engl J Med 2008;358:1465–73. through CCL22 production by tumor cells. Cancer Res 2011;71:6143–52. 13. Galluzzi L, Vacchelli E, Eggermont A, Fridman WH, Galon J, Sautes-Frid- 27. Klinman DM, Currie D, Gursel I, Verthelyi D. Use of CpG oligodeoxynu- man C, et al. Trial Watch: Experimental Toll-like receptor agonists for cleotides as immune adjuvants. Immunol Rev 2004;199:201–16. cancer therapy. Oncoimmunology 2012;1:699–716. 28. Garbi N, Arnold B, Gordon S, Hammerling GJ, Ganss R. CpG motifs as þ þ 14. Pasare C, Medzhitov R. Toll pathway-dependent blockade of CD4 CD25 proinflammatory factors render autochthonous tumors permissive for T cell-mediated suppression by dendritic cells. Science 2003;299:1033–6. infiltration and destruction. J Immunol 2004;172:5861–9.

OF10 Cancer Res; 75(21) November 1, 2015 Cancer Research

CCL22 Suppression Blocks Cancer Progression

29. Clark RA, Huang SJ, Murphy GF, Mollet IG, Hijnen D, Muthukuru M, et al. 38. Anz D, Eiber S, Scholz C, Endres S, Kirchner T, Bourquin C, et al. In breast þ þ Human squamous cell carcinomas evade the immune response by down- cancer, a high ratio of tumour-infiltrating intraepithelial CD8 to FoxP3 regulation of vascular E-selectin and recruitment of regulatory T cells. J Exp cells is characteristic for the medullary subtype. Histopathology 2011; Med 2008;205:2221–34. 59:965–74. 30. Perret R, Sierro SR, Botelho NK, Corgnac S, Donda A, Romero P. Adjuvants 39. Rodenburg RJ, Brinkhuis RF, Peek R, Westphal JR, Van Den Hoogen FH, van that improve the ratio of antigen-specific effector to regulatory T cells Venrooij WJ, et al. Expression of macrophage-derived chemokine (MDC) enhance tumor immunity. Cancer Res 2013;73:6597–608. mRNA in macrophages is enhanced by interleukin-1beta, tumor necrosis 31. Wu W, Weigand L, Belkaid Y, Mendez S. Immunomodulatory effects factor alpha, and lipopolysaccharide. J Leukoc Biol 1998;63:606–11. associated with a live vaccine against Leishmania major containing CpG 40. Vulcano M, Albanesi C, Stoppacciaro A, Bagnati R, D'Amico G, Struyf S, oligodeoxynucleotides. Eur J Immunol 2006;36:3238–47. et al. Dendritic cells as a major source of macrophage-derived chemokine/ 32. Bacher N, Raker V, Hofmann C, Graulich E, Schwenk M, Baumgrass R, et al. CCL22 in vitro and in vivo. Eur J Immunol 2001;31:812–22. Interferon-alpha suppresses cAMP to disarm human regulatory T cells. 41. Iellem A, Colantonio L, Bhakta S, Sozzani S, Mantovani A, Sinigaglia F, et al. Cancer Res 2013;73:5647–56. Inhibition by IL-12 and IFN-alpha of I-309 and macrophage-derived 33. Pasare C, Medzhitov R. Toll-like receptors: linking innate and adaptive chemokine production upon TCR triggering of human Th1 cells. Eur J immunity. Microbes Infect 2004;6:1382–7. Immunol 2000;30:1030–9. 34. Sisirak V, Faget J, Gobert M, Goutagny N, Vey N, Treilleux I, et al. Impaired 42. Muthuswamy R, Urban J, Lee JJ, Reinhart TA, Bartlett D, Kalinski P. Ability IFN-alpha production by plasmacytoid dendritic cells favors regulatory T- of mature dendritic cells to interact with regulatory T cells is imprinted cell expansion that may contribute to breast cancer progression. Cancer Res during maturation. Cancer Res 2008;68:5972–8. 2012;72:5188–97. 43. Muthuswamy R, Berk E, Junecko BF, Zeh HJ, Zureikat AH, Normolle D, 35. Graziani G, Tentori L, Navarra P. Ipilimumab: a novel immunostimulatory et al. NF-kappaB hyperactivation in tumor tissues allows tumor-selective monoclonal antibody for the treatment of cancer. Pharmacol Res 2012; reprogramming of the chemokine microenvironment to enhance the 65:9–22. recruitment of cytolytic T effector cells. Cancer Res 2012;72:3735–43. 36. Pere H, Montier Y, Bayry J, Quintin-Colonna F, Merillon N, Dransart E, 44. Hafner M, Zawatzky R, Hirtreiter C, Buurman WA, Echtenacher B, Hehlgans et al. A CCR4 antagonist combined with vaccines induces antigen-specific T, et al. Antimetastatic effect of CpG DNA mediated by type I IFN. Cancer þ CD8 T cells and tumor immunity against self antigens. Blood 2011;118: Res 2001;61:5523–8. 4853–62. 45. Xiao H, Peng Y, Hong Y, Huang L, Guo ZS, Bartlett DL, et al. Local 37. Mailloux AW, Young MR. NK-dependent increases in CCL22 secretion administration of TLR ligands rescues the function of tumor-infiltrating selectively recruits regulatory T cells to the tumor microenvironment. CD8 T cells and enhances the antitumor effect of lentivector immuniza- J Immunol 2009;182:2753–65. tion. J Immunol 2013;190:5866–73.

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 OF11

Suppression of Intratumoral CCL22 by Type I Interferon Inhibits Migration of Regulatory T Cells and Blocks Cancer Progression

David Anz, Moritz Rapp, Stephan Eiber, et al.

Cancer Res Published OnlineFirst October 2, 2015.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-14-3499

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2015/10/02/0008-5472.CAN-14-3499.DC1

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2015/10/20/0008-5472.CAN-14-3499. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.