ag-presenting cpg-activated pdcs prime th17 cells that induce … · suggesting that cpg-activated,...

Microenvironment and Immunology

Ag-Presenting CpG-Activated pDCs Prime Th17 Cells ThatInduce Tumor Regression

Leslie Gu�ery1, Juan Dubrot1, Carla Lippens1, Dale Brighouse1, Pauline Malinge2, Magali Irla1,3,Caroline Pot1,4, Walter Reith1, Jean-Marc Waldburger1, and St�ephanie Hugues1

AbstractPlasmacytoid dendritic cells (pDC) rapidly and massively produce type I IFN and other inflammatory cyto-

kines in response to foreign nucleic acids, thereby indirectly influencing T-cell responses. Moreover, antigen(Ag)-presenting pDCs directly regulate T-cell differentiation. Depending on the immune environment, pDCsexhibit either tolerogenic or immunogenic properties. Here, we show that CpG-activated pDCs promote efficientTh17 differentiation. Indeed, Th17 responses are defective in mice selectively lacking MHCII on pDCs uponantigenic challenge. Importantly, in those mice, the frequency of Th17 cells infiltrating solid tumors is impaired.As a result, the recruitment of infiltrating leukocytes in tumors, including tumor-specific cytotoxic T lymphocytes(CTL), is altered and results in increased tumor growth. Importantly, following immunization with tumor Agand CpG-B, MHCII-restricted Ag presentation by pDCs promotes the differentiation of antitumor Th17 cellsthat induce intratumor CTL recruitment and subsequent regression of established tumors. Our resultshighlight a new role for Ag presenting activated pDCs in promoting the development of Th17 cells andimpacting on antitumor immunity. Cancer Res; 74(22); 6430–40. �2014 AACR.

IntroductionPlasmacytoid dendritic cells (pDC) exhibit important innate

functions during infections, notably by producing largeamount of type I IFN (IFNI; ref. 1). However, recent evidencesdirectly involved these cells in adaptive immunity. pDCsupregulate MHC class II molecules (MHCII) upon inflamma-tion (2) and function as antigen (Ag)-presenting cells (APC)in vivo in several mouse models of diseases to induce bothT-cell–mediated immunity and tolerance. For instance, pDCspromote the initiation of myelin-induced Th17 responses andExperimental Autoimmune Encephalomyelitis (EAE; ref. 3).Furthermore, Ag targeting to pDCs via BST-2 in combinationwith Toll-like receptor (TLR) agonists provides an effectivevaccination (4). On the other hand, Treg induction by pDCswasshown to delay cardiac allograft rejection, dampen asthmaticreactions to inhaled Ags, and protect against graft versus hostdisease (5–7). In addition, Ag targeting to pDCs via Siglec-Hinhibits Th-cell–dependent autoimmunity (8). We recently

formally demonstrated that MHCII-restricted presentationof myelin Ags by pDCs promotes Treg expansion and inhibitsEAE (9).

The role of pDCs in antitumor immunity has been debated.In tumor microenvironment, pDCs exhibit a tolerogenic phe-notype characterized by low costimulatory molecule expres-sion and low IFNI production, and promote tumor growthpossibly through Treg induction (10–13). In contrast, specificAg delivery to pDCs using BST2 in combination with TLRagonists induced protective immunity against tumor growth(4). In humans, intranodal injections of pDCs activated andloaded with tumor Ag-associated peptides induced antitumorspecific T-cell responses (14). However, because targeting Agpresentation in pDCs also increased their production of IFNI(15), those studies could not determine the contribution of Agpresentation by pDCs in impacting tumor T-cell immunity. Inaddition, TLR-triggered pDCs exhibit potent antitumor effects.Tumor regression has been correlated for some (i.e., R848 andCpG-A), but not all (CpG-B) TLR ligands, with IFNI productionby pDCs (16). Thus, pDC-mediated antitumor potential mayrely on other IFNI independent pDC functions, possibly viatheir ability to present tumor Ags.

Here, we investigatedwhetherAg-presenting pDCcapacitiescould bemodulated and exploited to enhance antitumor T-cellimmunity. We show that upon immunization with an Ag andCpG-B, in vivo Ag-specific Th17 responses are significantlyimpaired in genetically modified mice lacking MHCII expres-sion onpDCs. In contrastwithwhatwas shown in conventionalDCs (cDC; ref. 17), MHCII deficiency does not impair innatepDC functions. Importantly, in mice further challengedwith tumors, pDC-primed Th17 cells control tumor growthby promoting the recruitment of immune cells, including

1Department of Pathology and Immunology, University of Geneva MedicalSchool, Geneva, Switzerland. 2NovImmune SA, Geneva, Switzerland.3Centre d'immunology deMarseille Luminy, Universit�e de la M�editerran�ee,Marseille, France. 4Division of Neurology, Department of Clinical Neuros-ciences, Geneva University Hospitals, Geneva, Switzerland.

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

Corresponding Author:St�ephanie Hugues, Department of Pathology andImmunology, University of Geneva Medical School, 1 rue Michel Servet,1211, Geneva, Switzerland. Phone: 41-2-23-79-58-93; Fax: 41-2-23-79-57-46; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-14-1149

�2014 American Association for Cancer Research.

CancerResearch

Cancer Res; 74(22) November 15, 20146430

on February 20, 2021. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 24, 2014; DOI: 10.1158/0008-5472.CAN-14-1149

http://cancerres.aacrjournals.org/

tumor-specific cytotoxic T lymphocytes (CTL), into thetumors. In addition, vaccination with an MHCII-restrictedtumor epitope ofmice bearing established tumors significantlyinducesTh17 cells that promote intratumoral CTL recruitmentand tumor growth inhibition, in a mechanism dependent onMHCII expression by pDCs. Thus, CpG-activated pDCs con-tribute as APCs to the induction of antitumoral Th17 cells thatdramatically dampen established tumor growth, a propertythat may be of interest in the establishment of antitumorimmunotherapies.

Materials and MethodsMiceAll mice [WT, H2-Aa�/� (18), pIIIþIV�/� (19), OT-I (20),

OT-II Rag2�/� (21), mMT (22), mMT pIIIþIV�/� (9)] were ofpure C57BL/6 background and maintained under specificpathogen-free conditions. All animal husbandry and experi-ments were approved by the animal research committee ofthe University of Geneva (Geneva, Switzerland).

Immunizations and T-cell transferBone marrow (BM) chimeric mice were generated as

described (9). Mice were immunized subcutaneously withOVAII peptide (ISQAVHAAHAEINEAGR; 2 mg; Polypeptide) inthe presence of CpG-B 1668 (5mmoles; Invivogen), and injectedi.v. 24 hours later with OT-II Rag2�/� LN cells labeled with10 mmol/L carboxyfluorescein diacetate succinimidyl ester(CFSE; Invitrogen). After 4 days, dLN and ndLN cells wererestimulated for 18 hours with OVAII (10 mg/mL).

In vitro cell differentiation and treatmentspDCs were differentiated from BM with Flt3-L (100 ng/mL;

Peprotech; ref. 9), sorted as CD11cintB220þPDCA-1þ pDCs,using a Moflow Astrios (Beckman Coulter) and treated with0.3 mmol/L CpG-B 1668, 0.3 mmol/L CpG-A 1585, or 1 mg/mLimiquimod (Invivogen). In some experiments, cells were load-ed with 10 mg/mL of OVAII peptide.

Flow cytometrySiglec-H (eBio440c), CD8a (53.6.7), TNFa (MP6-XT22),

Foxp3 (FJK-16s), CD45.1 (A20), CD3 (145-2C11), and PDCA-1(eBio927) antibodies were from eBioscience, CD11c (n418),B220 (RA3-6B2), CD19 (6D5), F4/80 (BM8), and CD11b (M1/70) antibodies from Biolegend and I-Ab (AF6-120.1), CD4(RM4.5), IL17 (TC11-18H10), and IFNg (XMG1.2) antibodiesfrom BD. H-2kb SIINFEKL pentamer was purchased fromProimmune.

ELISAsCytokine production was assessed in culture supernatants

using ELISA Kit from eBioscience (TNFa, IL6, IL17, IL10, andTGFb), BD (IFNg), or PBL (IFNa).

qRT-PCRTotal RNA was isolated and RT-PCR was performed as

described (9, 21). Primer sequences are provided in Supple-mentary Information.

Dendritic cell isolation from LNLN cells were isolated (9, 21), depleted of lymphocytes using

CD3 and CD19 antibodies and magnetic sorting (MiltenyiBiotec). pDCs (CD11cintB220þPDCA-1þ) were sorted using aMoflow Astrios (Beckman Coulter).

Tumor experimentsMice were immunized subcutaneously with CpG-B (5

mmoles) and OVAII (10 mg) either before (7 days) or after(between 7 days and 10 days) tumor cell transplantation(5 � 105 EG7 thymoma cells, s.c.). Tumor size was measuredwith a caliper [L (length) � l (width)]. For tumor—infiltratinglymphocytes (TIL) analysis, tumors were digested with colla-genase D (1 mg/mL) þ DNAse 1 (10 mg/mL; Roche) and TILswere enriched using lympholyteM (Cedarlane Laboratory) andrestimulated for 18 hourswith phorbol 12-myristate 13-acetate(PMA; 50 ng/mL; Sigma) and ionomycine (1 mg/mL; Sigma).When indicated,micewere depleted of CD8þT cells using anti-CD8mAbs (53-6.72), 100 mg i.p. at days 8, 11, and 14 after tumorchallenge. In some experiments, EG7 tumor-bearingmice wereadoptively transferred with OVAI-specific CD8

þ transgenic Tcells (2� 106) purified fromLN and spleen of OT-I CD45.1miceusing magnetic sorting (Miltenyi Biotec).

In vitro generation of Th17 cellsOT-II na€�ve T cells were extracted from LN using CD4þ

CD62Lþ T cells isolation kit (Miltenyi). Cells were cultured for3 days on anti-CD3 and anti-CD28 (BioXCell) coated platesin presence with IL6 (25 ng/mL) and TGFb (2.5 ng/mL;eBioscience). Cells (1.106) were then injected i.v. into suble-thally irradiated mice (500 Gy).

In vivo killing assayCFSE labeled splenocytes were loaded (2.5 mmol/L CFSE)

or not (0. 5 mmol/L CFSE) with OVAI peptide (SIINFEKL;10 mg/mL; Polypeptide) and 5 � 106 each (ratio 1:1) wasinjected i.v. into tumor-bearingmice. Twenty-hours later, CFSEcells were analyzed in tumor dLN (TdLN) and ndLN. Specificin vivo killing was calculated as [1�(% CFSE 2.5 mmol/LTdLN/% CFSE 0.5 mmol/L TdLN)/(% CFSE 2.5 mmol/LndLN/% CFSE 0.5 mmol/L TdLN)] � 100.

Statistical analysisStatistical significance was assessed by the Mann–Whitney

test or by the two-way ANOVA test with Bonferroni correc-tion for tumor growths using GraphPad software. �, P < 0.05;��, P < 0.01; ��, P < 0.001.

ResultspDCs induce Th17 cells upon CpG-B activation

We investigated whether the lack of MHCII on pDCs has animpactonTh responses upon inflammatory antigenic challenge.MHCII expression is regulated by the master regulator CIITA,itself under the control of cell-specific promoters (Supplemen-tary Fig. S1A; ref. 23).MHCII expression in pDCs strictly relies onpIII, whereas pI is expressed in macrophages and cDCs (23). Weused mice deficient for pIII and pIV of CIITA (pIIIþIV�/� mice;ref. 19). Because of the deletion of pIV that is expressed by

Ag-Presenting pDCs Promote Antitumor Th17 Cells

www.aacrjournals.org Cancer Res; 74(22) November 15, 2014 6431




cortical thymic epithelial cells, pIIIþIV�/� mice are devoid ofCD4þ T cells (24). Consequently, and as described before (9), werestored the CD4þ T-cell population by generating BM chimericmice using irradiated WT recipients. Furthermore, because pIIIis expressed in B cells, we have used BMprecursors derived fromB-cell deficient (mMT) mice that were backcrossed withpIIIþIV�/� mice. In both cases (mMT:WT and mMTpIIIþIV�/�:WT), BM chimeric mice lack B cells, the only dif-ference being presence (mMT:WT), or absence (mMT pIIIþIV�/

�:WT) of MHCII on pDCs (Supplementary Fig. S1B). In contrastwith pDCs, cDCs express normal MHCII levels in mMT:WT andmMT pIIIþIV�/�:WT mice (data not shown; ref. 9). Mice wereimmunized with OVAII peptide þ CpG-B and transferred withCFSE-labeled OVA-specific OT-II CD4þ T cells. T cells prolifer-ated similarly in draining lymph nodes (dLN), but not in non-dLN (ndLN), of mMT:WT and mMT pIIIþIV�/�:WT mice (Sup-plementary Fig. S2). The proportion of IFNg producing OT-IIcells, as well as the frequency of Foxp3þ OT-II cells in the dLN,was not affected by the lack of MHCII on pDCs (Fig. 1A and B).Thus, in contrast with our previous study performed in thecontext of EAE (9), in which CFA was used as an adjuvant, CpG-activated pDCs do not promote Treg. The frequency of IL17secreting proliferating OT-II cells, as well as ex vivo IL17production, was significantly impaired in dLN of mMTpIIIþIV�/�:WT compared with mMT:WT mice (Fig. 1A–C),suggesting that CpG-activated, MHCII-sufficient pDCs pres-ent Ag to CD4þ T cells to promote Th17 responses. Con-sistently, CpG-B–treated, OVA-loaded pDCs were foundcompetent to induce IL17 production by OT-II cells in vitro(Fig. 1D). We further demonstrated that impaired Th17responses observed in CpG-B context resulted from theselective abrogation of MHCII expression by pDCs. Indeed,levels of MHCII expression by other APCs (including CD8þ

DCs and macrophages) were rigorously similar in mMT:WTand mMTpIIIþIV:WT chimeric mice, first in steady-statebefore immunization, and then equally upregulated 12 hoursafter CpG-B treatment (Supplementary Fig. S3). CD11bþ DCsalready expressed very high levels of MHCII in na€�ve con-dition, and did not further upregulate MHCII after CpG-Btreatment, CD11bþ DCs from mMT:WT and mMTpIIIþIV:WTmice; however, expressing similar MHCII levels (not shown).Altogether, our results demonstrated that impaired Ag-spe-cific Th17 observed in mMTpIIIþIV:WT compared withmMT:WT immunized in the context of CpG-B resulted fromthe selective abrogation of MHCII expression by pDCs.

Innate pDC functions are not affected by the absence ofMHCII

In cDCs, MHCII molecules potentialize TLR-triggeredinnate functions (17). We tested both in vitro and in vivowhether pDC innate functions could be affected by the lossof MHCII, and explain altered Th17 responses in micelacking MHCII on pDCs. We first analyzed TLR-triggeredresponses in pDCs derived in vitro from BM of WT,pIIIþIV�/�, and H2-Aa�/� mice. Imiquimod and CpG-Btreatments induced a significant MHCII upregulation byWT, but not H2-Aa�/� and pIIIþIV�/�, pDCs (Fig. 2A).However, WT, pIIIþIV�/�, and H2-Aa�/� pDCs produced

equal levels of IL6, TNFa, and IFNa in response to bothTLR ligands after 6 and 24 hours (Fig. 2B and C). Intracel-lular TNFa staining showed similar results (Fig. 2D).Accordingly, WT, H2-Aa�/�, and pIIIþIV�/� pDCs express-ed comparable levels of Tnfa, Il6, Il1b, and Ifna4 mRNAsafter imiquimod or CpG-B stimulation (SupplementaryFig. S4A). As described (25), pDCs produce low IFNI levelsafter imiquimod or CpG-B treatment (Fig. 2B and C andSupplementary Fig. S4A). In contrast, expressions of Ifna4and IfnbmRNA were similarly strongly induced after CpG-Atreatment in both MHCII-sufficient and MHCII-deficientpDCs (Supplementary Fig. S4B).

In vivo, LN WT pDCs (Fig. 2E) expressed steady state levelsof MHCII, whereas pIIIþIV�/� pDCs were MHCII negative(Fig. 2F). WT and pIIIþIV�/� pDCs similarly increased theexpression levels of Il1b, Il6, Tnfa, and IfnbmRNAs upon CpG-Binjection (Fig. 2G), demonstrating that MHCII deficiency doesnot impair TLR-mediated pDC innate responses in vivo.

A

B

IFN

g+ce

lls(%

)

NS

0

5

10

15 *

IL17

+ce

lls(%

)

0

1

2

3

0

2

4

6

8

10NS

Fo

xp3+

cells

(%)

CFSE

Fo

xp3

IL17

IFN

gIF

Ng

mMT

:WT

mM

T p

III+I

V-/

-:W

T

6.4

CFSE

Fo

xp3

IL17

D

0

1

2

3

4

T T + pDC

NoneOVAII +CpG-B

IL17

(n

g/m

L)

C

IFN

g (n

g/m

L)

IL17

(n

g/m

L)

mMT:WT mMT pIII+IV-/- :WT

0

10

20

30

40NS

8.42.311.6

12.3 0.9

0

0.5

1

1.5

2 *

Figure 1. Mice lacking MHCII on pDCs exhibit altered Th17 responses invivo. A–C, mMT:WT and mMT pIIIþIV�/�:WT chimeras were immunized inthe flank (s.c.) with CpG-B and OVAII, and CFSE-labeled OT-II Rag2�/�

cells were adoptively transferred and analyzed 4 days later in dLN andndLN. A, representative flow-cytometric profiles showing IL17þ, IFNgþ,and Foxp3þ cells after gating on CFSEþ CD4þ T cells. B, graphsrepresent percentages of proliferating T cells expressing IL17, IFNg , orFoxp3. C, IL17 and IFNg production by T cells in culture supernatants.B and C, results show the means and SEM derived from 7 mice and arerepresentative of three independent experiments. D, BM-pDCs wereincubated or not for 24 hours with CpG-B and OVAII, washed, andcoculturedwithOT-II Rag2�/�LNcells for 4days. IL17 concentrationwasmeasured in culture supernatants.

Gu�ery et al.

Cancer Res; 74(22) November 15, 2014 Cancer Research6432




Altogether, our data demonstrate that TLR9-activated pDCsact as APCs to induce Th17 cell differentiation.

Tumor Ag presentation by CpG-activated pDCs inducesTh17 cells that control subsequent tumor challenge byrecruiting immune cells in solid tumorsTo determine whether pDC-induced Th17 responses

could be exploited in antitumor T-cell immunity, we immu-nized mMT:WT and mMT pIIIþIV�/�:WT mice with OVAII þCpG-B and further challenged mice with OVA-expressingtumors (EG7) in the same flank. Mice lacking MHCIIon pDCs exhibited a significant increase in tumor growth(Fig. 3A), suggesting that OVAII peptide presentation byCpG-activated pDCs controls tumor growth. Similar resultswere observed when BM chimeric mice were immunizedwith OVA protein instead of OVAII peptide (SupplementaryFig. S5A), further confirming that pDCs function as bona fideAPCs in our model. In addition, we observed a significantreduction in absolute CD45hi TIL numbers in mMTpIIIþIV�/�:WT compared with mMT:WT mice (Fig. 3B).Although frequencies of CD4þ T cells, CD8þ T cells, cDCs,and pDCs infiltrating solid tumors were identical whetherpDCs express MHCII or not, the dramatic reduction of TILs

in the absence of MHCII on pDCs led to a significant de-crease in absolute numbers of those cells (Fig. 3C–H).

To investigate whether tumor growth was controlled byTh17 cells primed by CpG-activated pDCs, we further analyzedeffector function of tumor-infiltrating CD4þ T cells. Weobserved a significant decrease in frequencies of both IL17þ

and IL17þIFNgþ CD4þ T cells (Fig. 4A, C, and E) in mMTpIIIþIV�/�:WT mice compared with mMT:WT mice. On thecontrary, the IFNgþ CD4þ T cells (Fig. 4A and 4D) and Foxp3þ

CD4þ T-cell frequencies remained unchanged (Fig. 4B and F),whereas absolute numbers of all these populations weredecreased in mMT pIIIþIV�/�:WT mice, reflecting the drasticreduction in total TILs (Fig. 4C–F). Consistent with impairedIL17þ and IL17þIFNgþ CD4þ T-cell frequencies, the produc-tion of IFNg and IL17 by TILs was significantly decreased inmice lacking MHCII on pDCs (Fig. 4G). In contrast, IL10 andTGFb productions were not affected, further confirming thatAg presentation by CpG-activated pDCs did not lead to Tregdevelopment (Fig. 4G).

Consistent with a role of tumor-specific Th17 cells in theinhibition of tumor growth, and as described in the mouse B16melanoma tumor model (26), we observed a significant reduc-tion in tumor size (Fig. 4H), as well as increased intratumor

A

MHCII

WT H2-Aα−/−

pIII+IV−/−

None CpG-BImiquimod D

B

C

TN

Fa

(ng

/mL

)

CpG-BImiCT0

1

2

3

TN

Fa

(ng

/mL

)

IL6

(ng

/mL

)

0

1

2

3

CpG-BImiCT

WT H2-Aα−/−

pIII+IV −/−

WT H2-Aα−/−

pIII+IV−/−

CpG-BImiCT

IL6

(ng

/mL

)

CpG-BImiCT

PDCA-1

TN

Fa

None Imiquimod CpG-B

WT

H2-Aα-/-

pIII+IV-/-

0.44

0.39

0.27

21.7

18.9

18.4

22.5

22.3

22.6

IFN

a (p

g/m

L)

TN

Fa+

pD

Cs

(%)

0

10

20

30

40

NSNS

WT H2-Aα−/−

pIII+IV−/−

CpG-BImiCT

IFN

a (p

g/m

L)

CpG-BImiCT0

50

100

150

200

250

NS

NS

0

20

40

60

80

100

CpG-BImiCT

0

2

4

6

0

0.5

1

1.5

2

FE

MHCII

pDC

WTpIII+IV-/-

SS

C

CD19

CD

11c

B220

PD

CA

-1

B220

cDCpDC

GF

old

incr

ease

0

1

2

3

Fo

ldin

crea

se0

1

2

3

4

5

Fo

ldin

crea

se

0

2

4

6

8

Fo

ldin

crea

se

0

1

2

3

4Tnfa Il1bIl6 Ifnb

WT pIII+IV-/- WT pIII+IV-/- WT pIII+IV-/- WT pIII+IV-/-

Figure2. MHCII-deficient pDCs exhibit normal TLR responses in vitro and in vivo. A–D, BM-pDCs fromWT, H2-Aa�/�, and pIIIþIV�/�mice were activatedor not with imiquimod or CpG-B. A, flow-cytometric profiles of MHCII expression after 24 hours. B and C, TNFa, IL6, and IFNa production after 6 hours(B) or 24 hours (C) of treatment. Results are pooled from 4 to 8 mice. D, representative flow-cytometric profiles of intracellular TNFa staining after 6hours of treatment. Histograms show percentages of TNFaþ cells among PDCA-1þ cells. A–D, results are representative of at least threeindependent experiments. E–G, WT and pIIIþIV�/� mice were injected in the flank (s.c.) with PBS or CpG-B, and pDCs from dLN were analyzed 12 hourslater. E, pDCs were sorted by flow cytometry as CD19�CD11cintB220þPDCA-1þ cells. F, MHCII expression by LN pDCs from PBS-injected mice.G, Tnfa, Il6, Il1b, and Ifnb mRNA fold increase in sorted pDCs from CpG-B relative to PBS-injected mice. The mean and SEM were derived from4 to 5 mice and are representative of three independent experiments.






IL17 cytokine levels (Fig. 4I), when in vitro differentiated Th17,but not Th0 cells (Supplementary Fig. S6A and S6B), wereadoptively transferred in mice bearing established tumors.

CpG-activated pDCprimed Th17 responses promote CTLrecruitment in tumors

Because tumor rejection is mainly mediated by CTLs, weinvestigated whether tumor-specific CD8þ T cells were affect-ed by the loss of MHCII expression by pDCs. Although thepercentage of IFNgþCD8þT cells was slightly increased in EG7tumors from mMT pIIIþIV�/�:WT compared with mMT:WT,

frequencies of OVA-specific (pentamerþ) CD8þ T cells weresimilar (Fig. 5A–D). Absolute numbers of IFNg producing andpentamerþ CD8þ T cells were lower in mMT pIIIþIV�/�:WT

B

CD

45h

ice

ll (N

um

ber

×10

3)

0

100

200

300**

AmMT:WTmMT pIII+IV-/- :WT

Time (days)

Tu

mo

rsi

ze (

mm

2)

Siglec-H

CD

11c

mMT pIII+IV-/- :WTmMT:WT

4.53

0.98

4.44

0.76


1.05

24.5

0.91

26.6

CD4

CD

8

DC

0

10

20

30

Nu

mb

er (

×103

)

NS

0

25

50

75 *E

0

0.5

1

1.5

2 NS

F

0

1

2

3*CD4+ CD8T cells + T cells

HG

0

2

4

6

8NS

0

5

10

15 *

0

1

2

3

4 *

0

0.5

1.0

1.5NS

mMT:WT mMT pIII+IV-/- :WT

sCDpsCDc

Fre

qu

ency

(%

)

Fre

qu

ency

(%

)F

req

uen

cy (

%)

Fre

qu

ency

(%

)

Nu

mb

er (

×10

3 )

Nu

mb

er (

×10

3 )N

um

ber

(×1

03 )

4 6 8 100

20

40

60

80

***

Figure 3. Tumor Ag presentation by CpG-activated pDCs controlssubsequent tumor challenge. A–H, mMT:WT and mMT pIIIþIV�/�:WTchimeras were immunized s.c. with OVAII and CpG-B. Seven days later,EG7 cells were implanted subcutaneously in an ipsilateral manner. A,tumor growth was measured every 1 to 2 days. Results show the meanand SEMderived from 18mice. B–H, at day 8 after tumor transplantation,TILs were extracted and analyzed. Results show the mean and SEMderived from 3 to 6 mice and are representative of three independentexperiments. B, histograms showabsolute numbers of CD45hi TILs. C–D,representative flow-cytometric profiles, after gating on CD45hi cells, ofCD4þ and CD8þ T cells (C) and cDCs (CD11cþ cells) or pDCs (SiglecHþ

cells; D). E–H, percentages and absolute numbers of CD4þ T cells (E),CD8þ T cells (F), cDCs (G), and pDCs (H).

A mMT pIII+IV-/- :WTmMT:WT

C D

E FNS

0

1

2

3

4

5NS

0

1

2

3 **

0

5

10

15

20

0

5

10 **

Fre

qu

ency

(%

)

Fre

qu

ency

(%

)F

req

uen

cy (

%)

Fre

qu

ency

(%

)

0

5

10

15

20 **

Nu

mb

er (

×103

)

Nu

mb

er (

×103

)N

um

ber

(×1

03)

Nu

mb

er (

×102

)

0

2

4

6

8

10 **

mMT:WT mMT pIII+IV-/-:WT

IFNg

IL17

**

0

0.5

1

1.5

0

2

4

6

8 **

14.6

85.4 82.5

17.51.66

3.9779

15.4 0.58

3.7985.6

10

CD4

Fo

xp3

mMT pIII+IV-/- :WTmMT:WT B

IL17

(n

g/m

L)

IFN

g (n

g/m

L)

IL10

(n

g/m

L)

TG

Fb

(ng

/mL

)

0

2

4

6

8 ** **

0

2

4

6

0

0.5

1

1,5

2NS

0

0.2

0.4

0.6 NSG

0 5 10 15 200

100

200

300

400

Time (days)

Th0Th17

Th celltransfer

Tu

mo

rsi

ze (

mm

2 )

H*

I

0

1

2

3

4

5

6

IL17

(p

g/m

L×1

02)

**

Th0 Th17

IL17+ cells IFNg + cells

IL17+ IFNg + cells Foxp3+ cells

Figure 4. pDC-primed Th17 cells control tumor growth. A–G, mMT:WT andmMT pIIIþIV�/�:WT chimeras were immunized s.c. with OVAII and CpG-B.Seven days later, EG7 cells were implanted s.c. in an ipsilateral manner. Atday 8 after tumor transplantation, tumors were extracted and TILs wererestimulated 18 hours with PMA/Iono. A, representative flow-cytometricprofiles of IL17 and IFNg expression after gating onCD45hiCD4þ T cells. B,representative flow-cytometric profiles of CD4 and Foxp3 expression aftergating on CD45hi cells. C–F, percentages and absolute numbers ofCD4þ T cells expressing IL17 (C), IFNg (D), IL17, IFNg (E), or Foxp3 (F).G, IL17, IFNg , IL10, andTGFbweremeasured inculturesupernatants.C–G,results show the mean and SEM derived from 3 to 6 mice and arerepresentative of three independent experiments. H and I, EG7 cells wereimplanteds.c. inWTmice.Sevendays later,micewere injected i.v.withOT-II T cells in vitro polarized into Th0 or Th17 cells. H, tumor growth wasmeasuredevery 1 to2days. I, TILswere restimulated in vitrowithPMA/Ionoand IL17 was quantified in culture supernatants. H and I, results arerepresentative of three independent experiments.

Gu�ery et al.





mice compared with mMT:WT mice (Fig. 5C–D). In dLN,percentages (Fig. 5E–F) and absolute numbers (not shown)of pentamer CD8þ T cells were, however, identical in bothgroups, suggesting an impaired recruitment of those cells intumors. Moreover, CTL cytotoxic activity was similar whether

pDCs express MHCII or not in an in vivo killing assay usingOVAI-loaded CFSE-stained target cells (Fig. 5G and H). Theseresults suggest that increased tumor-specific Th17 cellsinduced by CpG-activated pDCs did not alter priming of CTLand Th1 cells in dLN but impaired the recruitment of thosecells in solid tumors. Accordingly, total cells isolated fromtumor dLN migrated less efficiently toward EG7 tumor super-natant extracted from mMT pIIIþIV�/�:WT compared withmMT:WT chimeras (Supplementary Fig. S7A), although relativefrequencies of LN cell subsets were not affected (not shown). Inaddition, the IL17-induced chemokines CCL-2 and CXCL-2were found significantly decreased in tumor supernatants ofmMT pIIIþIV�/�:WT compared with mMT:WT chimeras (Sup-plementary Fig. S7B). Altogether, our data demonstrated thatCpG-activated pDCs promoted antitumor Th17 cells, intratu-moral recruitment of immune cells, including CTLs, andresulted in tumor growth inhibition.

Ag-presenting functions of CpG-activated pDCs can beexploited for antitumor immunotherapies

To determine whether CpG-activated pDC ability to pro-mote antitumor Th17 cells could be used as a therapeuticalstrategy, we next transferred CpG-activated OVAII-loaded WTor pIIIþIV�/� pDCs in EG7 tumor-bearing mice. Unfortunate-ly, we could not observe any regression of tumors from miceinjectedwithMHCII sufficient pDCs (not shown). Similarly, thevaccination of mice bearing established tumors with OVAII þCpG-B had no beneficial effect on tumor growth when per-formed in the flank in which the tumor developed (not shown),suggesting that pDC ability to prime effector T cells might havebeen altered by the tumor microenvironment. However, con-tralateral OVAII þ CpG-B vaccination led to a significantreduction in tumor size (Fig. 6A). In addition, we observedsimilar results when tumor-bearing mice were immunizedwith OVA protein þ CpG-B (Supplementary Fig. S5B). Impor-tantly, tumors from OVAIIþ CpG-B vaccinated mice exhibiteda dramatic increase in infiltrating CD45hi TILs, with a selectiveenrichment of Th17 cells (Fig. 6B–E). In agreement, IL17produced at the vaccination site, and consequently in tumorswas significantly increased (Fig. 6F). These data show thatOVAII þ CpG-B vaccination of tumor-bearing mice promotesTh17 cells that induce intratumor cell recruitment and tumorgrowth inhibition. Consistent with our previous observations,vaccination did not affect tumor-specific CTL frequencies(Fig. 6G–I). However, as a consequence of enhanced cellinfiltration, absolute numbers of Th1, Th17, pentamerþCD8þ,and IFNg-producing CD8þ T cells were significantly increased(Fig. 6C, D, and G–I).

To determine whether tumor growth inhibition observedfollowing vaccination of tumor-bearing mice was dependenton MHCII-mediated Ag presentation by pDCs, we performedsimilar experiments in mMT:WT and mMT pIIIþIV�/�:WTchimeras. First, tumor growth was found similar in unvacci-nated mMT:WT and mMTpIIIþIV�/�:WTmice, demonstratingthat, Ag presentation by pDCs in tumor microenviron-ment does not lead to antitumor T-cell immunity. However,EG7 tumor-bearing mMT:WT chimeras immunized withOVAIIþCpG-B in the contralateral flank exhibited dramatic

*

0

2

4

6

8

10

0

10

20

30

40

50

Nu

mb

er (

×102

)

Nu

mb

er (

×10

2 )

*

Fre

qu

ency

(%

)

Fre

qu

ency

(%

)NS

0

4

8

12

0

0.5

1

1.5

2

2.5 *

29.5

70.5

42.1

57.9


CD8

IFN

g

8.41

91

11.4

88.6


CD8

Pen

t


mMT pIII+IV-/-:WTmMT:WT

CD8

Pen

t

dLN

ndLN

0.09 0.10

0.01 0.01

CD

8+p

ent+

cells

(%)

mMT:WT

mMT pIII+IV−/−:WT

67.432.6

69.230.8

60.2

39.8

58.9

41.1


dLN

ndLN

CFSE CFSE

0

20

40

60

Sp

ecif

ic in

viv

o

killi

ng

(%)

mMT:WT

mMT pIII+IV-/-:WT

BA

DC

FE

HG

IFNg + CD8+ T cells Pent+ CD8+ T cells

0

0.05

0.10

0.15

Figure 5. Impaired CTL recruitment in mice lacking MHCII on pDCs. A–H,mMT:WT and mMT pIIIþIV�/�:WT chimeras were immunized s.c. withOVAII and CpG-B. Seven days later, EG7 cells were implanted s.c. in anipsilateral manner. B–D, at day 8 after tumor transplantation, TILs wererestimulated with PMA/Iono. A and B, representative flow-cytometricprofiles of CD8 and IFNg (A) or CD8 and pent-OVA (B) after gating onCD45hi cells. CandD,percentages andabsolute numbersofCD8þTcellsexpressing IFNg (C) or the Pentamer OVA (D). E and F, 8 days after tumortransplantation, dLN and ndLN were analyzed. E, representative flow-cytometric profiles of pent-OVA expression after gating on CD8þ T cells.F, percentages and absolute numbers of CD8þ T cells expressing pent-OVA (pentþ). G andH, at day 7 after tumor transplantation, CFSE-labeledtarget cells were injected i.v. and dLN were analyzed 20 hours later. G,representative flow-cytometric profiles of CFSE. H, histograms showpercentages of Ag-specific in vivo killing. A–H, results show themeanandSEMderived from3 to 6mice andare representative of three independentexperiments.






tumor growth inhibition, whereas, in contrast, no protectiveeffect was observed in mMTpIIIþIV�/�:WT, in which pDCs areMHCII deficient (Fig. 6J). Altogether, these results demon-strate that OVAIIþCpG-Bmediated tumor regression, and thatthis effect was dependent on pDC Ag-presenting functions.

Antitumor adaptive immune responses against EG7 tumorsare mainly CTL-dependent (27–29). To determine whetherTh17-mediated control of tumor growth was CTL dependent,CD8þ T cells were depleted in WT tumor-bearing mice vac-cinated or not with OVAII þ CpG-B. As expected, CD8 deple-tion led to a significant increase in tumor size (Fig. 7A).Importantly, we observed a total abrogation of OVAII þCpG-B vaccine efficacy in the absence of CD8þ T cells(Fig. 7A), suggesting that OVAII-mediated Th17 cells controltumor growth in a CD8-dependent manner. Most of the cancertherapies and vaccines developed so far have been directedto target specific CD8þ T cells (30). Thus, we next wonderedwhether having a concomitant antitumor Th17 and CD8 T-cellresponses improve tumor rejection. For that we adoptivelytransferred OT-I cells into WT EG7 tumor-bearing mice thatwe vaccinated or not with OVAII þ CpG-B. We injected sub-optimal OT-I amounts (2� 106), which only conferred a partialand time-limited control of tumor growth when transferredalone (Fig. 7B). OVAII þ CpG-B vaccination increased OT-Irecruitment into tumors (Fig. 7C), and consequently signifi-cantly ameliorated and prolonged OT-1–mediated tumorgrowth inhibition (Fig. 7B). Accordingly, OVAII þ CpG-Bvaccination dramatically improved OT-1–mediated tumorrejection in mMT:WT chimeras (Fig. 7D). In striking contrast,OVAII þ CpG-B vaccination had no effect on tumor growthin OT-I transferred mMT pIIIþIV�/�:WT (Fig. 7D), high-lighting the importance of Ag presentation by pDCs in thismodel.

Altogether, our results demonstrate that antitumor vacci-nation using MHCII-restricted epitopes in the presence ofCpG-B induce tumor antigen presentation by pDCs and sub-sequent antitumor Th17 responses. pDC-mediated Th17 cellsconsequently potentialize CTL-mediated tumor immu-notherapies by increasing intratumor CTL recruitment.

DiscussionpDCs are potent sensors of nucleic acid and become acti-

vated to produce large amounts of IFNI (31). Those cells,despite low Ag uptake capacity, also function as APCs inseveral conditions and, depending on their activation statusand the cytokinic environment, promote either T-cell tolerance(3, 4) or effective T-cell immunity (5–9). In tumors, it has beendemonstrated that IFNI displays antitumoral effects throughantiproliferative and proapoptotic functions, thus inhibitingtumor cells survival and angiogenesis (32, 33). Moreover, IFNI

0

1

2

3

4

00

1

2

3

0

10

20

30

Fre

qu

ency

(%

)

Fre

qu

ency

(%

)

Fre

qu

ency

(%

)F

req

uen

cy (

%)

Fre

qu

ency

(%

)F

req

uen

cy (

%)

0

200

400

600

800

Nu

mb

er (

×103

)

Nu

mb

er (

×103

)N

um

ber

(×1

03)

Nu

mb

er (

×103

)N

um

ber

(×1

03)

Nu

mb

er (

×103

)

0

1

2

3

4

5

0

5

10

15

20

25

0.2

0.4

0.6

0.8

1.0

*

0

5

10

15

20

25 *

0

1

2

3 *

0

10

20

30 *

*

*

d0

CTVAX

0

50

100

150

200

250

d2 d4

*

*

Tu

mo

r si

ze (

mm

2)

Cel

ls (

Nu

mb

er ×

106 )

CT

VAX

0

0.5

1

1.5

2

2.5

IL17

(p

g/m

L)

TumordLN

A B

C G

F

J

0

40

80

120

160

*

0

1

2

3

4

*

0

10

20

30

40

50

*

**

HD

IE

CD4+ CD8T cells + T cells

IFNg + CD4+ T cells Pent+ CD8+ T cells

IL17+ CD4+ T cells IFNg + CD8+ T cells

Tu

mo

rsi

ze (

mm

2 )

1614121080

30

60

90

120 mMT:WT

mMT pIII+IV-/-:WTmMT:WT VAX

mMT pIII+IV-/-:WT VAX

VAX

*

NS

***

Figure 6. MHCII-mediated pDC-dependent vaccination with CpG-B þOVAII inhibits established tumors. A–I, EG7 cells were implanted s.c. inWTmice. Ten days later, mice were immunized (VAX) or not (CT) s.c. withOVAIIþCpG-B in a contralateral manner. A, tumor growthwasmeasuredevery 1 to 2 days. B–I, at day 5 after immunization, TILs were analyzed.B, histograms show absolute numbers of CD45hi TILs. C–E, and G–I,percentages and absolute numbers of CD4þ T cells (C), IFNgþ CD4þ Tcells (D), IL17þCD4þ T cells (E), CD8þ T cells (G), PentþCD8þ T cells (H),and IFNgþCD8þ T cells (I). F, IL17 wasmeasured in culture supernatants.

B–I, results show the mean and SEM derived from 5 mice and arerepresentative of three independent experiments. J, EG7 cells wereimplanted s.c. in mMT:WT and mMT pIIIþIV�/�:WT mice. Ten days later,mice were immunized s.c. with OVAIIþCpG-B in a contralateral manner.Tumor growth was measured every 1 to 2 days. Results show the meanandSEMderived from5mice andare representative of three independentexperiments.

Gu�ery et al.





limits tumor progression by enhancing CD8aþ DCs mediatedcross-presentation of tumor Ags and subsequent CTL priming(34, 35). Consequently, IFNIs are used in the treatment ofvarious cancers (32). However, pDC ability to produce IFNI isimpaired in tumor microenvironment, therefore convertingthese cells into tolerogenic pDCs and leading to immuno-suppression in several different cancers (for review, see ref. 33).In addition, in breast tumors in particular, pDCs induceTreg expansion in IDO and/or ICOS-dependent mechanisms(10–12). Accordingly, pDC infiltration in tumors has beencorrelated with poor clinical outcome (36, 37). Thus, pro-moting pDC activation in tumor context would restore IFNIproduction, inhibit Treg expansion, and represent an effec-tive approach to induce antitumoral immune response. Inagreement, CpG-A–activated, IFNI-producing pDCs injecteddirectly into tumors initiate an effective and syste-mic antitumor immunity through the orchestration of animmune cascade involving the sequential activation of nat-ural killer (NK) cells, cDCs, and CD8þ T cells (38). Moreover,TLR7 ligands or CpG-B intratumoral injection inhibitmurine mammary tumor growth in a pDC-dependent man-ner (16). Interestingly, however, protection was linked torestoration of IFNI occurring after TLR7 ligand, but not afterCpG-B, administration. Thus, although the role of pDC-mediated IFNI in antitumor immunity has been convincinglydemonstrated, it remains unknown whether those cellscould directly contribute to antitumoral T-cell immunity asAPCs, and in particular after being activated by CpG-B.

Here, using mice lacking MHCII expression by pDCs, weshow that upon CpG-B activation, pDCs drive Th17 cell dif-ferentiation through MHCII-dependent Ag presentation. Forexperimental reasons, our mouse model lacks peripheral Bcells, which might not be neutral. However, we comparedMHCIIþ and MHCII� pDCs in a B-cell deficient background,and investigated the selective impact of MHCII-mediated Agpresentation by pDCs on CD4þ T-cell responses and tumorgrowth.

It has been previously demonstrated that TLR triggeredpDCs indirectly impact T-cell responses by activating otherAPCs (1). Our results, together with previous studies, highlighta direct role for pDCs as APCs in the modulation of T-helperactivation and outcome. Our data further establish that, incontrast with what was described for cDCs, innate pDC func-tions are not altered by the loss of MHCII. Because MHCIImachinery is differentially regulated in pDCs and cDCs (39, 40),regulation of TLR-triggered inflammatory responses may alsobe different in those cells, and, in contrast with cDCs (17),MHCII may not promote TLR signaling in pDCs. Altogether,our data demonstrate that loss of MHCII expression by pDCsleads to impaired Th17 responseswithout affecting pDC innatefunctions and that CpG-activated pDCs function as efficientpro-Th17 bona fide APCs.

VAX

6 8 10 12 14 160

30

60

90

120

150

180 CTCT VAXa-CD8a-CD8 VAX

Tu

mo

rsi

ze (

mm

2 ) NS

*

***

A

B

C OT1 OT1 + VAX

CD8

CD

45.1

0

10

20

30

40

CD

45.1

+C

D8+

cells

(%

)

OT1 OT1 +

VAX

11.4% 28.7%

6 8 10 12 140

40

80

120mMT:WT +OT1

mMT pIII+IV-/- :WT +OT1mMT:WT +OT1 +VAX

mMT pIII+IV-/- :WT +OT1 +VAX

OT1 +/- VAX

Tu

mo

rsi

ze (

mm

2 )

D

Time (days)

Time (days)

Time (days)

**

20151050

50

100

150

200

250 CpG-B

OT1 + CpG-B

OT1 + CpG-B + OVAII NS

*****

Tu

mo

rsi

ze (

mm

2 )

VAX

*

Figure 7. MHCII-mediated pDC-dependent beneficial effect of OVAII

vaccine to OT-I transfer in inducing tumor regression. A–C, EG7 cellswere implanted s.c. in WTmice. A, 8 days later, mice were injected or notwith depleting anti-CD8 mAbs (days 8, 11, 14) and immunized or not in acontralateral manner with OVAII þ CpG-B (day 8). Tumor growth wasmeasured every 1 to 2 days. Results show the mean and SEM derivedfrom 6 mice and are representative of two independent experiments. Band C, 8 days later, mice were transferred or not with OT-I CD45.1 cells(2�106) and immunized or not in a contralateral manner with OVAII þCpG-B. B, tumor growth was measured every 1 to 2 days. Results showthe mean and SEM derived from 8 mice and are representative of twoindependent experiments. C, OT-I cell frequencies (CD45.1þ cells) wereanalyzed in tumors 18 days after T-cell transfer. Representative flow-cytometric profiles and histograms represent a pool of two experimentswith 4 mice/group. D, EG7 cells were implanted s.c. in mMT:WT and mMTpIIIþIV�/�:WT mice. Eight days later, mice were transferred with OT-I

CD45.1 cells (2�106) and immunized or not in a contralateral manner withOVAII þ CpG-B. Tumor growth was measured every 1 to 2 days. Resultsshow the mean and SEM derived from 8 mice and are representative oftwo independent experiments.






Tissue infiltrating pDC numbers were correlated with thepresence of Th17 cells in mucosa and skin of GvHD patients(41, 42) and in mouse tumors (43, 44). pDC depletion duringEAE leads to impaired encephalitogenic Th17 responses anddecreased clinical scores (3). In vitro, pDCs activatedwith TLR7or TLR9 ligands induce Th17 cell differentiation from eithernaive or memory T cells (45, 46). In addition, TGFb-treatedpDCs transferred into collagen-induced arthritismice promoteTh17 cells and exacerbate the disease (47). Depleting antibodyexperiments demonstrated that production of some cytokinesby pDCs, namely TGFb, IL6, IL1b, IL23, IFNa, and TNFadepending on the studies, was involved in Th17 cell differen-tiation (44, 46, 47). We clearly determined that, in addition topDC-derived cytokines that are preserved in our model, pDCAg-presenting capacities are implicated in Th17 induction.However, mechanisms accounting for CpG-B–activated pDCability to drive Th17 cells as APCs remain to be elucidated.Future experiments, including careful analysis of proteinexpression and dynamic interactions with T cells, will decipherintrinsic pDCproperties, which allow them to promote distinctT-cell responses (Th1, Th17, Treg) depending on inflammatorycontexts.

Subsequent tumor cell challenge showed that pDC-primedTh17 cells control tumor growth. Moreover, our data suggestthat pDC-induced Th17 and Th1/17 cells promote inflamma-tion and intratumor recruitment of immune cells, includingAg-specific CTLs, resulting in an increased antitumor immu-nity. Th17 cells implication in tumor immunity remains con-troversial. For instance, IL17 production favors angiogenesisand tumor cell survival (48, 49). Conversely, Th17 mightimprove tumor rejection by recruiting other immune cellssuch as DCs, Th, CTLs, and NK cells in the tumor (50–52).Accordingly, the IL17-induced CCL-2 and CXCL-2 chemokines,were decreased in tumor supernatants from mice lackingMHCII on pDCs. Decrease in CCL-2 and CXCL-2, which inducethe recruitment of lymphocytes, myeloid cells (53–55), andneutrophils (56), respectively, may explain impaired recruit-ment of several immune cell populations into tumors in micelacking MHCII on pDCs. Altogether, our data demonstratedthat MHCIIþ pDC-derived, tumor-specific Th17 cells increaseoverall immune tumor infiltrate, including CTLs that wouldthen mediate tumor rejection. Finally, and in agreement withour results showing a decrease in IFNgþIL17þ Th cells inabsence of MHCII on CpG-activated pDCs, production of IFNgby Th17 cells was shown to be essential for Th17-dependenttumor rejection (26).

Tumor microenvironment maintains pDCs in a partiallyactivated phenotype that would prevent their ability to driveeffective antitumor T-cell responses (10). Accordingly, intumor-bearing mice, immunization with a tumor Ag in pres-ence of CpG-B in an ipsilateral manner, or transfer of CpG-Bactivated and tumor Ag-loaded pDCs, failed to inhibit tumorgrowth. However, tumor Ag immunization at a distal site leadsto induction of tumor-specific Th17 cells, massive recruitmentof immune cells in tumors, and inhibition of tumor growth.Importantly, those protective effects are abrogated in micelacking MHCII on pDCs, demonstrating requirement of pDCAg-presenting functions for effective tumor regression. Inter-

estingly, efficacy of therapeutic vaccines was more potentthan in a prevention setting. One probable explanation isthat when the vaccine is administrated to tumor-bearingmice, tumor-specific CTLs have already been primed, and assoon as some vaccine-derived Th17 cells are induced, andinfiltrate solid tumors to promote intratumor recruitment ofalready primed endogenous tumor-specific CTLs. In con-trast, in the case of preventive vaccine, Th17 cells are primedin the absence of any MHCI tumor Ags and consequentlyattract CTLs only once they have been generated, days aftertumor challenge. OVAII vaccine-induced protection is abro-gated in CD8-depleted mice, providing direct evidence thatpDC-mediated Th17-induced tumor rejection relies ontumor-specific CTLs.

Notably, intratumoral specific CTL recruitment, as well asCTL-mediated tumor rejection, is significantly enhanced byconcomitant vaccination with MHCII-restricted tumor Ag andCpG-B. Importantly, this effect is exclusively dependent ontumor Ag presentation by pDCs to CD4þ T cells and subse-quent induction of antitumor Th17 responses. Thus, strategiesaiming at enhancing tumor-specific CTL priming, such asMHCI tumor epitope vaccines, could be synergized by pro-moting CTL recruitment into tumors via induction of pDC-dependent tumor-specific Th17.

Here, we demonstrate that Ag-presenting activated pDCsinduce potent Ag-specific Th17 cells, suggesting that pDCscould be used not only as inflammatory cytokines producers,but also as efficient APCs, to improve tumor vaccine efficacy.Those results pave the way for future manipulation torestore the antitumor T-cell responses, notably by combin-ing targeting on both innate and adaptive pDCs functions.

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

Authors' ContributionsConception and design: L. Gu�ery, W. Reith, J.-M. Waldburger, S. HuguesDevelopment of methodology: C. Lippens, P. Malinge, C. PotAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): L. Gu�ery, J. Dubrot, C. Lippens, D. Brighouse, C. Pot,J.-M. WaldburgerAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): L. Gu�ery, J. Dubrot, D. Brighouse, S. HuguesWriting, review, and/or revision of the manuscript: L. Gu�ery, J. Dubrot,C. Lippens, M. Irla, C. Pot, W. Reith, J.-M. Waldburger, S. HuguesStudy supervision: S. Hugues

AcknowledgmentsThe authors thank J.P Aubry-Lachainaye for excellent assistance in flow

cytometry, F.V. Duraes and V. Bochet for help in experiments, N. Page andD. Merkler for providing OT-I mice, and N. Bendriss-Vermare and F. Benvenutifor helpful discussions.

Grant SupportThis work was supported by the Swiss National Science Foundation

(310030-127042) and the European Research Council (pROsPeCT 281365)grants (S. Hugues).

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

Received April 17, 2014; revised August 4, 2014; accepted August 29, 2014;published OnlineFirst September 24, 2014.

Gu�ery et al.





References1. Colonna M, Trinchieri G, Liu YJ. Plasmacytoid dendritic cells in

immunity. Nat Immunol 2004;5:1219–26.2. Villadangos JA, Young L. Antigen-presentation properties of plasma-

cytoid dendritic cells. Immunity 2008;29:352–61.3. Isaksson M, Ardesjo B, Ronnblom L, Kampe O, Lassmann H, Eloranta

ML, et al. Plasmacytoid DCpromote priming of autoimmune Th17 cellsand EAE. Eur J Immunol 2009;39:2925–35.

4. Loschko J, Schlitzer A, DudziakD, Drexler I, SandholzerN, BourquinC,et al. Antigen delivery to plasmacytoid dendritic cells via BST2 inducesprotective T cell-mediated immunity. J Immunol 2011;186:6718–25.

5. de Heer HJ, Hammad H, Soullie T, Hijdra D, Vos N, Willart MA, et al.Essential role of lung plasmacytoid dendritic cells in preventing asth-matic reactions to harmless inhaled antigen. J Exp Med 2004;200:89–98.

6. Hadeiba H, Sato T, Habtezion A, Oderup C, Pan J, Butcher EC. CCR9expression defines tolerogenic plasmacytoid dendritic cells able tosuppress acute graft-versus-host disease. Nat Immunol 2008;9:1253–60.

7. Ochando JC, Homma C, Yang Y, Hidalgo A, Garin A, Tacke F, et al.Alloantigen-presenting plasmacytoid dendritic cellsmediate toleranceto vascularized grafts. Nat Immunol 2006;7:652–62.

8. Loschko J, Heink S, Hackl D, Dudziak D, Reindl W, Korn T, et al.Antigen targeting to plasmacytoid dendritic cells via Siglec-H inhibitsTh cell-dependent autoimmunity. J Immunol 2011;187:6346–56.

9. Irla M, Kupfer N, Suter T, Lissilaa R, Benkhoucha M, Skupsky J, et al.MHC class II-restricted antigen presentation by plasmacytoid den-dritic cells inhibits T cell-mediated autoimmunity. J Exp Med2010;207:1891–905.

10. Sisirak V, Faget J, Gobert M, Goutagny N, Vey N, Treilleux I, et al.Impaired IFN-alpha production by plasmacytoid dendritic cells favorsregulatory T-cell expansion that may contribute to breast cancerprogression. Cancer Res 2012;72:5188–97.

11. Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H, et al.Plasmacytoid dendritic cells frommouse tumor-draining lymph nodesdirectly activate mature Tregs via indoleamine 2,3-dioxygenase. J ClinInvest 2007;117:2570–82.

12. Faget J, Bendriss-Vermare N, Gobert M, Durand I, Olive D, BiotaC, et al. ICOS-ligand expression on plasmacytoid dendritic cellssupports breast cancer progression by promoting the accumula-tion of immunosuppressive CD4þ T cells. Cancer Res 2012;72:6130–41.

13. Conrad C, Gregorio J, Wang YH, Ito T, Meller S, Hanabuchi S, et al.Plasmacytoid dendritic cells promote immunosuppression in ovariancancer via ICOS costimulation of Foxp3(þ) T-regulatory cells. CancerRes 2012;72:5240–9.

14. Tel J, AarntzenEH,BabaT,SchreibeltG, SchulteBM,Benitez-RibasD,et al. Natural human plasmacytoid dendritic cells induce antigen-specific T-cell responses in melanoma patients. Cancer Res2013;73:1063–75.

15. Tel J, Sittig SP, Blom RA, Cruz LJ, Schreibelt G, Figdor CG, et al.Targeting uptake receptors on human plasmacytoid dendritic cellstriggers antigen cross-presentation and robust type I IFN secretion.J Immunol 2013;191:5005–12.

16. LeMercier I, Poujol D, Sanlaville A, Sisirak V, Gobert M, Durand I, et al.Tumor promotion by intratumoral plasmacytoid dendritic cells isreversed by TLR7 ligand treatment. Cancer Res 2013;73:4629–40.

17. Liu X, Zhan Z, Li D, Xu L, Ma F, Zhang P, et al. Intracellular MHC class IImolecules promote TLR-triggered innate immune responses by main-taining activation of the kinase Btk. Nat Immunol 2011;12:416–24.

18. Kontgen F, Suss G, Stewart C, Steinmetz M, Bluethmann H. Targeteddisruption of the MHC class II Aa gene in C57BL/6 mice. Int Immunol1993;5:957–64.

19. LeibundGut-Landmann S, Waldburger JM, Reis e Sousa C, Acha-Orbea H, Reith W. MHC class II expression is differentially regulated inplasmacytoid and conventional dendritic cells. Nat Immunol2004;5:899–908.

20. Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ,Carbone FR. T cell receptor antagonist peptides induce positiveselection. Cell 1994;76:17–27.

21. BarndenMJ, Allison J, HeathWR, Carbone FR. Defective TCR expres-sion in transgenic mice constructed using cDNA-based alpha- andbeta-chain genes under the control of heterologous regulatory ele-ments. Immunol Cell Biol 1998;76:34–40.

22. Kitamura T, Takaku F, Miyajima A. IL-1 up-regulates the expression ofcytokine receptors ona factor-dependent humanhemopoietic cell line,TF-1. Int Immunol 1991;3:571–7.

23. Reith W, LeibundGut-Landmann S, Waldburger JM. Regulation ofMHC class II gene expression by the class II transactivator. Nat RevImmunol 2005;5:793–806.

24. Waldburger JM, Suter T, Fontana A, Acha-Orbea H, Reith W.Selective abrogation of major histocompatibility complex classII expression on extrahematopoietic cells in mice lacking promoterIV of the class II transactivator gene. J Exp Med 2001;194:393–406.

25. Kadowaki N, Antonenko S, Liu YJ. Distinct CpG DNA and poly-inosinic-polycytidylic acid double-stranded RNA, respectively,stimulate CD11c- type 2 dendritic cell precursors and CD11cþdendritic cells to produce type I IFN. J Immunol 2001;166:2291–5.

26. Muranski P, Boni A, Antony PA, Cassard L, Irvine KR, Kaiser A, et al.Tumor-specific Th17-polarized cells eradicate large established mel-anoma. Blood 2008;112:362–73.

27. Ramakrishnan R, Assudani D, Nagaraj S, Hunter T, Cho HI, Antonia S,et al. Chemotherapy enhances tumor cell susceptibility to CTL-medi-ated killing during cancer immunotherapy in mice. J Clin Invest2010;120:1111–24.

28. Hariharan K, Braslawsky G, Black A, Raychaudhuri S, Hanna N. Theinduction of cytotoxic T cells and tumor regression by soluble antigenformulation. Cancer Res 1995;55:3486–9.

29. Boissonnas A, Fetler L, Zeelenberg IS, Hugues S, Amigorena S. In vivoimaging of cytotoxic T cell infiltration and elimination of a solid tumor.J Exp Med 2007;204:345–56.

30. Palucka K, Banchereau J. Dendritic-cell-based therapeutic cancervaccines. Immunity 2013;39:38–48.

31. Gilliet M, Cao W, Liu YJ. Plasmacytoid dendritic cells: sensing nucleicacids in viral infection and autoimmune diseases. Nat Rev Immunol2008;8:594–606.

32. Wang BX, Rahbar R, Fish EN. Interferon: current status and futureprospects in cancer therapy. J Interferon Cytokine Res 2011;31:545–52.

33. Demoulin S, Herfs M, Delvenne P, Hubert P. Tumor microenvironmentconverts plasmacytoid dendritic cells into immunosuppressive/tolero-genic cells: insight into the molecular mechanisms. J Leukoc Biol2013;93:343–52.

34. Diamond MS, Kinder M, Matsushita H, Mashayekhi M, Dunn GP,Archambault JM, et al. Type I interferon is selectively required bydendritic cells for immune rejection of tumors. J Exp Med 2011;208:1989–2003.

35. Fuertes MB, Kacha AK, Kline J, Woo SR, Kranz DM, Murphy KM, et al.Host type I IFN signals are required for antitumor CD8þ T cellresponses through CD8{alpha}þ dendritic cells. J Exp Med 2011;208:2005–16.

36. 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.

37. Labidi-Galy SI, Sisirak V,MeeusP, GobertM, Treilleux I, Bajard A, et al.Quantitative and functional alterations of plasmacytoid dendritic cellscontribute to immune tolerance in ovarian cancer. Cancer Res2011;71:5423–34.

38. Liu C, Lou Y, Lizee G, Qin H, Liu S, Rabinovich B, et al. Plasmacytoiddendritic cells induceNK cell-dependent, tumor antigen-specific T cellcross-priming and tumor regression in mice. J Clin Invest 2008;118:1165–75.

39. Young LJ, Wilson NS, Schnorrer P, Proietto A, ten Broeke T,Matsuki Y, et al. Differential MHC class II synthesis and ubiqui-tination confers distinct antigen-presenting properties on conven-tional and plasmacytoid dendritic cells. Nat Immunol 2008;9:1244–52.






40. O'Keeffe M, Grumont RJ, Hochrein H, Fuchsberger M, Gugasyan R,Vremec D, et al. Distinct roles for the NF-kappaB1 and c-Rel tran-scription factors in the differentiation and survival of plasmacytoid andconventional dendritic cells activated by TLR-9 signals. Blood2005;106:3457–64.

41. Bossard C, Malard F, Arbez J, Chevallier P, Guillaume T, Delaunay J,et al. Plasmacytoid dendritic cells and Th17 immune response con-tribution in gastrointestinal acute graft-versus-host disease. Leukemia2012;26:1471–4.

42. Malard F, Bossard C, Brissot E, Chevallier P, Guillaume T, Delaunay J,et al. Increased plasmacytoid dendritic cells andRORgammat-expres-sing immune effectors in cutaneous acute graft-versus-host disease.J Leukoc Biol 2013;94:1337–43.

43. YokogawaM, Takaishi M, Nakajima K, Kamijima R, Digiovanni J, SanoS. Imiquimod attenuates the growth of UVB-induced SCC in micethrough Th1/Th17 cells. Mol Carcinog 2013;52:760–9.

44. Xu L,WangC, Zhou Y, Ren T,Wen Z. CpG oligonucleotides induce thedifferentiation of CD4(þ)Th17 cells by triggering plasmacytoid den-dritic cells in adoptively cell transfer immunotherapy. Immunol Lett2012;142:55–63.

45. Gautreau L, Chabannes D, Heslan M, Josien R. Modulation of regu-latory T cell-Th17 balance by plasmacytoid dendritic cells. J LeukocBiol 2011;90:521–7.

46. Yu CF, Peng WM, Oldenburg J, Hoch J, Bieber T, Limmer A, et al.Humanplasmacytoiddendritic cells support Th17cell effector functionin response to TLR7 ligation. J Immunol 2010;184:1159–67.

47. Bonnefoy F, Couturier M, Clauzon A, Remy-Martin JP, Gaugler B,Tiberghien P, et al. TGF-beta-exposed plasmacytoid dendritic cellsparticipate in Th17 commitment. J Immunol 2011;186:6157–64.

48. Numasaki M, Fukushi J, Ono M, Narula SK, Zavodny PJ, Kudo T, et al.Interleukin-17 promotes angiogenesis and tumor growth. Blood2003;101:2620–7.

49. Wang L, Yi T, Kortylewski M, Pardoll DM, Zeng D, Yu H. IL-17 canpromote tumor growth through an IL-6-Stat3 signaling pathway. J ExpMed 2009;206:1457–64.

50. Martin-Orozco N, Muranski P, Chung Y, Yang XO, Yamazaki T, Lu S,et al. T helper 17 cells promote cytotoxic T cell activation in tumorimmunity. Immunity 2009;31:787–98.

51. Kryczek I, Wei S, Szeliga W, Vatan L, Zou W. Endogenous IL-17contributes to reduced tumor growth and metastasis. Blood2009;114:357–9.

52. Nunez S, Saez JJ, Fernandez D, Flores-Santibanez F, Alvarez K, TejonG, et al. T helper type 17 cells contribute to anti-tumour immunity andpromote the recruitment of T helper type 1 cells to the tumour.Immunology 2013;139:61–71.

53. LancaT,Silva-SantosB.Recruitment of gammadeltaT lymphocytes totumors: A new role for the pleiotropic chemokine CCL2. Oncoimmu-nology 2013;2:e25461.

54. Ben-Baruch A. The Tumor-Promoting Flow of Cells Into, Withinand Out of the Tumor Site: Regulation by the Inflammatory Axisof TNFalpha and Chemokines. Cancer Microenviron 2012;5:151–64.

55. Ma Y, Mattarollo SR, Adjemian S, Yang H, Aymeric L, Hannani D,et al. CCL2/CCR2-dependent recruitment of functional antigen-presenting cells into tumors upon chemotherapy. Cancer Res2014;74:436–45.

56. Brackett CM, Muhitch JB, Evans SS, Gollnick SO. IL-17 promotesneutrophil entry into tumor-draining lymph nodes following inductionof sterile inflammation. J Immunol 2013;191:4348–57.


Gu�ery et al.




2014;74:6430-6440. Published OnlineFirst September 24, 2014.Cancer Res Leslie Guéry, Juan Dubrot, Carla Lippens, et al. Tumor RegressionAg-Presenting CpG-Activated pDCs Prime Th17 Cells That Induce

Updated version

10.1158/0008-5472.CAN-14-1149doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/09/23/0008-5472.CAN-14-1149.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/22/6430.full#ref-list-1

This article cites 56 articles, 26 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/22/6430.full#related-urls

This article has been cited by 4 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/74/22/6430To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-1149

http://cancerres.aacrjournals.org/content/suppl/2014/09/23/0008-5472.CAN-14-1149.DC1

http://cancerres.aacrjournals.org/content/74/22/6430.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/22/6430.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/22/6430


ag-presenting cpg-activated pdcs prime th17 cells that induce … · suggesting that cpg-activated,...

Documents