of June 18, 2013.This information is current as

Production of Type I IFNEnhances Antitumor Activity through the Cell-Associated Double-Stranded RNA

JanssenSara McBride, Kasper Hoebe, Philippe Georgel and Edith

http://www.jimmunol.org/content/177/9/61222006; 177:6122-6128; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/177/9/6122.full#ref-list-1

, 17 of which you can access for free at: cites 45 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2006 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

Cell-Associated Double-Stranded RNA Enhances AntitumorActivity through the Production of Type I IFN1

Sara McBride,* Kasper Hoebe,† Philippe Georgel,† and Edith Janssen2*

The efficacy of tumor cell vaccination largely depends on the maturation and activation status of the dendritic cell. Here weinvestigated the ability of soluble and tumor cell-associated dsRNA to serve as an adjuvant in the induction of protective adaptiveantitumor responses. Our data showed that cell-associated dsRNA, but not soluble dsRNA, enhanced both tumor-specific CD8�

and CD4� T cell responses. The cell-associated dsRNA increased the clonal burst of tumor-specific CD8� T cells and endowedthem with an enhanced capacity for expansion upon a secondary encounter with tumor Ags, even when the CD8� T cells wereprimed in the absence of CD4� T cell help. The adjuvant effect of cell-associated dsRNA was fully dependent on the expressionof TLR3 by the APCs and their subsequent production of type I IFNs, as the adjuvant effect of cell-associated dsRNA wascompletely abrogated in mice deficient in TLR3 or type I IFN signaling. Importantly, treatment with dsRNA-associated tumor cellsincreased the number of tumor-infiltrating lymphocytes and enhanced the survival of tumor-bearing mice. The data from ourstudies suggest that using cell-associated dsRNA as a tumor vaccine adjuvant may be a suitable strategy for enhancing vaccineefficacy for tumor cell therapy in cancer patients. The Journal of Immunology, 2006, 177: 6122–6128.

T he use of autologous tumor vaccines is a promisingmethod to target the greatest number of potential tumorAgs without requiring their individual identification. Den-

dritic cells (DCs)3 are the most potent APCs to take up, process,and present tumor vaccine Ags to naive tumor-specific CD4� andCD8� T cells (1–4). However, the efficacy of these treatment strat-egies largely depends on the maturation and activation status of theDC. Immature DCs exhibit potent phagocytic but poor stimulatorycapabilities and generally mediate T cell tolerance, whereas matureDCs conversely exhibit poor phagocytic but potent T cell activa-tion capacities (5). Current efforts toward enhancing the efficacy oftumor vaccine research have thus sought to either increase thenumber of Ag-specific DCs or selectively induce their activationand maturation (1–4, 6). An important pathway for DC maturationinvolves the detection of specific molecules of microbial originthrough their cognate receptors, which include the TLR family (7,8). Upon TLR engagement, DCs mature and up-regulate MHC-peptide complexes and costimulatory molecules on the cell surfaceas well as the secretion of cytokines and chemokines required forT cell activation (9, 10).

DCs are comprised of a heterogeneous family of leukocytes, andthe DC subsets known to cross-present cellular Ags are derivedfrom both the myeloid and lymphoid lineage; in mice, the cross-priming DC subsets are restricted to CD8�� populations (11, 12).

These cross-priming DC are endowed with specific TLR reper-toires and expression patterns that change upon maturation or ac-tivation (9). It has recently been shown that the DC subsets thatrespond to TLR3, TLR4, and TLR9 ligands can enhance T cellpriming in vitro (13). In vivo studies have demonstrated that co-administration of dsRNA with Ag enhances expansion of CD8� Tcells in a murine transfer model (14). Additional in vivo modelshave shown that dsRNA enhances endogenous CD8� T cell re-sponses upon treatment with peptide-pulsed DCs or Ag-pulsed celllines (14–16). Despite similarities in TLR signaling pathways, en-gagement of some specific TLRs induces distinct activation statesand cytokine secretion patterns by DCs and thereby different en-vironments in which T cells become activated, as well as possiblyinfluencing T cell function and survival.

We have investigated the ability of dsRNA to serve as an ad-juvant in the induction of protective antitumor CD4� and CD8� Tcell responses in normal and immunocompromised mice. We fo-cused on those effector molecules that convey the dsRNA-medi-ated adjuvant effect and the subsequent changes in T cell prolif-eration, survival, and cytolytic capacity.

Materials and MethodsMice and cell lines

C57BL/6, B6/129 F1, B6/129S2-IL6tm1Kopf/J (IL-6�/�), and B6;129S-Tnfrsf1atm1ImxTnfrsf1btm1Imx/J (TNFRp55/p75) were obtained from TheJackson Laboratory. IL-12p40�/�, IL-10�/�, OT-1 Rag�/�, HY Rag�/�,TRIFlps2/lps2 (17), TRIFlps2/lps2MyD88�/� (all backcrossed to the C57BL/6background), and B6/129TRIFlps2/lps2 mice were bred in-house. Type I IFNreceptor-deficient (IFN�BR�/�) mice were kindly provided by Dr. J.Sprent (The Scripps Research Institute, La Jolla, CA). Mice were main-tained under specified pathogen-free conditions in accordance with theguidelines established by the Association for Assessment and Accredita-tion of Laboratory Animal Care International.

The previously described cell lines EL-4, EL-4mOVA, andMEC.B7.Sig-OVA were cultured in IMDM (Invitrogen Life Technologies)supplemented with 10% FCS, 50 �M 2-ME, 2 mM L-glutamine, 20 U/mlpenicillin, and 20 �g/ml streptomycin (18, 19).

Association of TLR ligands to cells

Irradiated (3000 rad) EL-4 and EL-4-mOVA cells were loaded with syn-thetic dsRNA (polyinosinic-polycytidylic acid (poly(I:C)); Amersham Bio-sciences) via electroporation as previously described (16). Briefly, cells

*La Jolla Institute for Allergy and Immunology, Developmental Immunology, LaJolla, CA 92037; and †The Scripps Research Institute, Department of Immunology,La Jolla, CA 92037

Received for publication June 20, 2006. Accepted for publication August 9, 2006.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported in part by Leukemia and Lymphoma Society Career De-velopment Award 3248-05 (to E.J.).2 Address correspondence and reprint requests to Dr. Edith Janssen, La Jolla Institutefor Allergy and Immunology, Developmental Immunology 1B, 9420 Athena Circle,La Jolla, CA 92037. E-mail address: EJanssen@liai.org3 Abbreviations used in this paper: DC, dendritic cell; BM, bone marrow; PKR, pro-tein kinase R; poly(I:C), polyinosinic-polycytidylic acid. RIG-I, retinoic acid-induc-ible gene I; TIL, tumor-infiltrating lymphocyte.

The Journal of Immunology

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

were electroporated (950 farads, 0.250 V) in the presence or absence ofpoly(I:C) (10 �g in a 200-�l volume). This protocol has been shown toresult in �1 ng of poly(I:C) per 105 cells (16). At the time of injection theirradiated and electroporated EL-4 and EL-4-mOVA cells are early apo-ptotic as determined by annexin V expression and do not express type IIFN or IL-2.

Immunization and tumor challenge

Mice of indicated strains were immunized s.c. with 20 � 106 irradiated andelectroporated EL-4-mOVA cells resuspended in PBS. After 7 days, micewere sacrificed and isolated splenocytes were analyzed for OVA-specific Tcells responses. In cases where B6/129 mice were used, age- and sex-matched littermates were used to control for minor histocompatibility dif-ferences between EL-4-mOVA cells and the B6/129 mice. Depletion ofCD4� T cells in vivo was performed by i.p. administration of 150 �g of themAb (mAb) GK1.5 on the first 3 days before immunization (18, 20). Intumor challenge experiments, CD4-depleted mice were immunized with5 � 106 live EL-4-mOVA cells. As soon as palpable tumors had formed,20 � 106 irradiated electroporated EL-4-mOVA cells were administereds.c. to the mice, and tumor growth was monitored daily with a verniercaliper. Mice were euthanized when tumors reached 1 cm3.

Function and enumeration of tumor-specific T cells

For tumor-specific CD8� T cell responses in the periphery, splenocyteswere stimulated for 6 days in vitro with irradiated syngeneic MEC.B7.Sig-OVA. Following restimulation, viable cells were collected over Ficollgradient (Lympholyte-M; Cedarlane Laboratories). Cells were incubatedeither directly ex vivo or following in vitro restimulation with theOVA257–264 peptide (SIINFEKL; 5 �g/ml) or control peptide GP33–41

(KAVYNFATC; both from A & A Laboratories) for 5 h in the presence ofbrefeldin A. Surface staining for CD8 and intracellular cytokine stainingfor IFN-� was performed using a Cytofix/Cytoperm kit (BD Pharmingen)according to the manufacturer’s instructions. The fold expansion of specificCD8� cells was calculated by dividing the absolute number of IFN-��

CD8� cells after in vitro culture by the absolute number of IFN-�� CD8�

cells at the start of the culture (12, 18, 20). The cytolytic activity ofrestimulated splenocytes was evaluated by a JAM test as previouslydescribed using [3H]thymidine-labeled EL-4 cells loaded with OVA257–264

or GP33–41 peptide (21). Specific killing was calculated as ((spontaneouscpm � experimental cpm) � 100)/spontaneous cpm.

OVA-specific CD4� Th1 and Th2 cells were enumerated by ELISPOTafter a 48-h in vitro stimulation with OVA323–339 (ISQAVHAAHAEINEAGR; 10 �g/ml), control peptide LLO190–201 (NEKYAQAYPNVS;10 �g/ml), or Con A (2 �g/ml; positive control) as previously described(22). For proliferative responses, 2 � 106 cells were cultured in 96-wellplates in the presence of titrated doses of OVA323–339. After 72 h of culture,cells were pulsed with 0.2 �Ci of [3H]thymidine, and [3H]thymidineincorporation was determined 16 h later.

Tumor-infiltrating lymphocytes (TILs)

Tumor cells and TILs were isolated as described previously (23). Thedegree of lymphocyte infiltration was determined by FACS analysis afterstaining for CD8�, TCR-��, CD19, and CD11b. For further quantitativeenumeration of CD8� TILs, quantitative PCR was performed on resectedtumors using SYBR Green and the following primers: mL32, 5�-GAAACTGGCGGAAACCCA-3� (forward) and 5�- GGATCTGGCCCTTGAACCTT-3 (reverse); �-actin, 5�-CTGAATGGCCCAGGTCTGA-3�(forward) and 5�- CCCTGGCTGCCTCAACAC-3� (reverse); CD8�, 5�-CCGTTGACCCGCTTTCTGT-3� (forward) and 5�-CGGCGTCCATTTTCTTTGGAA-3� (reverse); and OVA, 5�-GACTGAGCAAGAAAGCAAACCTGTG- 3� (forward) and 5�-TTGTCCCACTGGCAAATGGAAG-3� (reverse). The ratio of CD8� mRNA to OVA mRNA was de-termined after normalization to �-actin and L32 for each sample.

In vitro DC assays

Bone marrow (BM)-derived DCs from male mice of the indicated strainswere generated with GM-CSF (BD Biosciences) as previously described(24). To evaluate cytokine production, DC activation, and phagocytoticcapacity, 105 DCs were cultured in 96-well plates with 104 irradiated andelectroporated EL-4-mOVA cells in the absence or presence of dsRNA.Supernatants were collected at different time points for IL-6, TNF-�, IL-10, and IL-12 analysis by cytometric bead array assay (BD Pharmingen).The concentration of type I IFN was determined by using a cell line con-taining the IFN-stimulated response element luciferase reporter constructdiscussed elsewhere (25). DC activation was determined by expression ofthe surface molecules MHC I/II, CD40, CD80, and CD86 in combination

with CD11c and CD11b (all from eBioscience) by flow cytometric anal-ysis. For Ag presentation assays, 5 � 105 DCs were incubated with 2 � 105

irradiated electroporated EL-4-mOVA cells. Purified OT-1 or HY CD8 Tcells (Miltenyi Biotec) were labeled with CFSE and added to the culturesafter 36 h. Cell proliferation was determined by flow cytometric analysis ofCFSE intensity in combination with staining for CD8�, V�2, and HYTCR, respectively, at the indicated time points (19).

Statistics

Unless stated otherwise, the data are expressed as means � SEM andevaluated using ANOVA followed by a Dunnett test. A probability valueof p � 0.05 was considered statistically significant.

ResultsCell-associated dsRNA, but not soluble dsRNA, enhances CD4�

and CD8� T cell responses

To test the adjuvant effect of synthetic dsRNA provided in theform of poly(I:C) on the adaptive antitumor response, wild-typemice were immunized with syngeneic thymoma cells expressingfull-length OVA on their membrane (EL-4-mOVA). The EL-4-mOVA cells were either associated with dsRNA via electropora-tion or mixed with soluble dsRNA before injection. After 7 days,the frequencies and functions of OVA-specific CD4� and CD8� Tcells were determined.

Immunization with cell-associated Ags generally results in poorCD4� T cell responses but strong CD8� T cell responses (26).Splenocytes from EL-4-mOVA-immunized mice did not show aproliferative response upon in vitro stimulation with their cognateligand, whereas splenocytes from mice immunized with dsRNA-associated EL-4-mOVA demonstrated a small but significant re-sponse (Fig. 1A). To dissect whether this augmentation resultedfrom an increased number of Ag-specific CD4� T cells or an in-creased proliferative capacity per cell, ELISPOT assays were per-formed to enumerate the OVA-specific CD4� T cells. EL-4-mOVA immunization induced OVA-specific CD4� T cells thatpredominantly produced IFN-�. Association of dsRNA with theEL-4-mOVA cells significantly enhanced the number of OVA-specific CD4� T cells producing IFN-�, but not IL-4, indicatingthat dsRNA enhanced the induction OVA-specific CD4� T cellsand skewed the response to a Th1 cell phenotype (Fig. 1B). Noeffect of soluble dsRNA was seen on the induction of the OVA-specific CD4� T cells, and the observed responses were compa-rable to immunization with EL-4-mOVA cells only (data notshown).

Cellular association of dsRNA with EL-4-mOVA cells inducedcomparable adjuvant effects in the OVA-specific CD8� T cell pop-ulation. Cell-associated dsRNA, but not soluble dsRNA, doubledthe CD8� T cell response (control, 1.39 104 � 0.47 104; solublepoly(I:C), 1.08 104 � 0.36 104; and cell-associated poly(I:C), 3.28104 � 0.62 104 SIINFEKL-specific CD8 T cells/spleen; Fig. 1, Cand D). Analysis at different time points revealed that this was theresult of greater clonal expansion rather than increased expansionkinetics (data not shown). The phenotypic cytokine profile andcytolytic capacity of the OVA-specific CD8� T cells was compa-rable on a per cell basis within all groups (data not shown). How-ever, cell-associated dsRNA endowed the CD8� T cells with agreater capacity for secondary expansion when cultured in vitrowith OVA-expressing fibroblasts (Fig. 1, D and E).

Cell-associated dsRNA potentiates immune responses inCD4-depleted mice

CD4� T cell help during CD8� T cell priming plays an importantrole in clonal expansion, cytokine production, capacity for second-ary expansion, and generation of memory CD8� T cells (20, 27,28). To dissect whether the enhanced CD8� T cells response was

6123The Journal of Immunology

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

an intrinsic trait of the CD8� T cells or resulted from the enhancedCD4� T cell response, we used mice that were depleted of CD4�

T cells before immunization. Although CD8� T cells responseswere induced in the absence of CD4� T cells, their clonal expan-sion was diminished along with their capacity for secondary ex-pansion upon re-encounter with cognate Ag (Fig. 2, A and B) (20,27, 28). Cell-associated dsRNA, but not soluble dsRNA, enhancedthe clonal expansion of CD8� T cells (control, 5.5 103 � 1.1 103;soluble poly(I:C), 4.2 103 � 1.7 103; and cell-associated poly(I:C),11.3 103 � 1.8 103 SIINFEKL-specific CD8 T cells/spleen; Fig. 1,C and D). Importantly, these CD8� T cells regained their capacityfor secondary expansion and were able to kill OVA-pulsed targetcells in vitro (Fig. 2, B and C). These data indicate that cell-asso-ciated dsRNA intrinsically alters both the Ag-specific CD4� andCD8� T cell responses, resulting in a more robust anti-tumorresponse.

dsRNA sensing is essential for the adjuvanticity

Different pathways of sensing dsRNA have been suggested to exertadjuvant effects on the adaptive immune response. 129/TRIFlps/lps

mice, which are deficient in TLR3 signaling due to a mutation inthe adaptor protein TRIF (17) yet normally express protein kinaseR (PKR), retinoic acid-inducible gene I (RIG-I), and RNaseL,were used to analyze the relative contributions of distinct dsRNA-sensing pathways in mediating the adjuvant effect of cell-associ-ated dsRNA. Immunization with dsRNA-associated EL-4-mOVAcells enhanced primary CD8� T cell expansion in CD4-depleted129/WT mice but not in 129/TRIFlps/lps mice (Fig. 3A). In line withthis finding, the dsRNA-associated EL-4-mOVA-induced CD8� Tcells from CD4-depleted 129/WT mice but not from 129/TRIFlps/

lps mice regained their cytolytic activity and capacity for secondary ex-pansion (Fig. 3B and data not shown), showing a key role for TLR3 but notfor PKR, RIG-1, and RNaseL in the adjuvant effect of cell-associateddsRNA.

The adjuvant effect of dsRNA is mediated through type I IFN

When BM-derived DCs were cultured with dsRNA they showedsigns of maturation through up-regulation of MHC class I andclass II as well as the costimulatory molecules CD40, CD80, andCD86 (data not shown) (10). In addition, these DCs produced var-ious cytokines, including TNF-�, IL-6, IL-10 IL-12, and type IIFNs (10). In line with our previously published findings, the sens-ing of apoptotic cells in the form of irradiated, electroporated EL-4-mOVA induced small amounts of type I IFN in BM-derived DCs(12). Importantly, EL-4-mOVA cell-associated dsRNA but notsoluble dsRNA, in a concentration similar to the cell-associatedamount, significantly increased production of type I IFN by theDC. Comparable results were found for TNF-�, IL-6, and IL-12(Fig. 4A and data not shown). To dissect which cytokines wereelemental in conferring the dsRNA adjuvant effect, BM-DCs fromcytokine and cytokine receptor-deficient mice were cultured withEL-4-mOVA cells (either normal or associated with dsRNA), andthe proliferation of naive OVA-specific OT-1 CD8� T cells wasevaluated. DCs from all animals induced comparable CD8� T cellproliferation in the absence of dsRNA; 50–60% of the OT-1 cellsshowed signs of division based on their CFSE dilution profile. Inall DC subsets except those generated from IFN-��R�/� mice,cross-presentation of dsRNA-associated EL-4-mOVA resulted inenhanced proliferation, evident in both the percentages of dividingcells and the rounds of successful division (Fig. 4B).

FIGURE 1. Enhanced CD4� and CD8� T cell responses after immunization with cell-associated dsRNA. Wild-type mice were immunized s.c. with 2 �107 EL-4-mOVA cells that were electroporated (e.p.) in the presence or absence of synthetic dsRNA (poly(I:C)). As a control, EL-4-mOVA cells weremixed with dsRNA before immunization. After 7 days, splenocytes were isolated and Ag-specific CD4� and CD8� T cell functions were determined. A,Proliferation of Ag-specific CD4� T cells after a 3-day stimulation with the OVA323–339 epitope in vitro in splenocytes derived from mice immunized withEL-4-mOVA cells in the absence (�) or presence (■) of dsRNA. B, The frequency of Ag-specific IFN-� and IL-4 production by splenic CD4� T cellswas determined by ELISPOT upon stimulation with the MHC class II-restricted OVA323–339 epitope (■) and control peptide LLO 190–201 (�). C, Thefrequency of IFN-�� Ag-specific CD8� T cells in spleens 7 days after immunization upon a 5-h stimulation with OVA257–264-peptide (filled bars) asdetermined by intracellular cytokine staining. GP33–41 peptide stimulation (open bars) was used as control stimulation D, Flow cytometric analysis of thefrequency of IFN-�� OVA257–264-specific CD8� T cells among total splenocytes before and after a 6-day in vitro restimulation with MEC.B7.Sig-OVAcells. The value in the left panel represents the percentage of total splenocytes that are CD8�IFN-��. E, Fold expansion of OVA257–264-specific CD8� uponin vitro stimulation with MEC.B7.Sig-OVA cells. The fold expansion is calculated by dividing the absolute number of OVA257–264-specific CD8� cellsafter in vitro stimulation by the absolute numbers of OVA257–264-specific CD8� at the start of the culture. Values represent mean � SEM (n � 4) and arerepresentative of three experiments.

6124 dsRNA-INDUCED IFN ENHANCES ANTI-TUMOR RESPONSE

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

As observed in vivo, the effect of cell-associated dsRNA on Tcell proliferation was greater that that of soluble dsRNA. OT-1cells demonstrated greater proliferation when stimulated bydsRNA-associated EL-4-mOVA-fed DCs than normal EL-4-mOVA-fed DCs (Fig. 4C). The addition of soluble dsRNA re-sulted in only a small increase in OT-1 proliferation, whereas re-combinant IFN-� strongly enhanced the proliferative capacity,thus confirming a role for type I IFN in the adjuvant effect ofdsRNA. Interestingly, HY CD8� T cells, which recognize maleself-Ag consistently expressed by the DC, showed no difference inproliferation when stimulated with BM-derived DCs that were feddsRNA-associated EL-4-mOVA or EL-4-mOVA in the presenceof soluble dsRNA. These observations suggest that the adjuvanteffect of dsRNA sensing during presentation of self-Ags by cellsdiffers from the mechanism observed in cross-presentation.

Absence of dsRNA-mediated adjuvant effect in IFN-��R�/�

mice

To confirm the role of type I IFN in the cell-associated dsRNAadjuvant effect in vivo, CD4-depleted wild-type and IFN-��R�/�

mice were immunized with EL-4-mOVA cells (normal or dsRNAassociated). IL-12p40�/� mice, deficient in IL-12 and IL-23, bothof which are cytokines with known adjuvant effects on T cell prim-ing (29), were used as an additional control. Comparable to ob-servations in CD4-depleted wild-type mice, cell-associated dsRNAenhanced primary expansion and restored both secondary expan-sion and the cytolytic capacity of OVA-specific CD8� T cells inIL-12p40�/� mice, exclusive of IL-12 and IL-23, in the adjuvanteffect of cell-associated dsRNA (Fig. 5 and data not shown). Im-

portantly, cell-associated dsRNA did not enhance the OVA-spe-cific CD8� T cell responses in CD4-depleted IFN-��R�/� mice,nor did it restore either cytotoxicity or secondary expansion (Fig.5 and data not shown).

dsRNA-associated tumor vaccines inhibit tumor growth andincrease TILs

To test whether dsRNA association enhanced the therapeutic po-tential of a tumor vaccine, we developed a model in which CD4-depleted wild-type mice were inoculated with live EL-4-mOVAcells and treated with irradiated EL-4-mOVA cells (normal ordsRNA-associated) once palpable tumors had formed. Treatmentof mice with irradiated EL-4-mOVA cells (with or without solubledsRNA) failed to inhibit tumor growth or enhance survival com-pared with untreated mice, as did dsRNA-associated EL-4 cellsthat lack the major rejection Ag OVA (Fig. 6, A and B, and data notshown). In contrast, increased survival (40 days after tumor in-oculation) and tumor growth inhibition were observed among micetreated with dsRNA-associated EL-4-mOVA cells. Analysis ofdissected tumors 14 days postinoculation revealed interesting TILdifferences. Tumors from dsRNA-associated EL-4-mOVA-treatedmice contained nearly 2-fold greater TILs (absolute numbers) thantumors derived from control, EL-4, and EL-4-mOVA-treated mice.However, in a sample size of n � 8, this increase was not signif-icant ( p � 0.56, data not shown). Subsequent FACS analysisshowed that the percentage of CD8� T cells among the TILs wassignificantly increased in dsRNA-associated EL-4-mOVA-treatedmice (41.2 � 11.8%) in comparison with control, EL-4-mOVA-treated, and dsRNA-associated EL-4 treated mice (20.3 � 4.6%19.8 � 7.2%, and 21.83 � 5.2%, respectively). To obtain a morequantitative assessment of the CD8� T cell population infiltratingthe tumor, quantitative RT-PCR was performed and the ratio ofCD8� mRNA to tumor Ag (OVA) mRNA was determined. Nosignificant difference in CD8�:OVA mRNA ratio was observedbetween control, EL-4-mOVA, and dsRNA-associated EL-4treated mice. Corresponding with the flow cytometric data,dsRNA-associated EL-4-mOVA treatment significantly increasedthe CD8�:OVA mRNA ratio, demonstrating increased CD8�-ex-pressing cells infiltrating into the tumor.

These data show that cell-associated dsRNA can be used toinduce a protective CD8� T cell response to tumor Ags, even inimmunocompromised mice.

FIGURE 2. Cell-associated dsRNA enhances primary and secondarycytotoxic responses in CD4� T cell-deficient mice. Wild-type mice weredepleted of CD4� cells before immunization with 2 � 107 EL-4-mOVAcells that were electroporated (e.p.) in the presence of absence of syntheticdsRNA (poly(I:C)) as described in the Fig. 1 legend. A, The frequency ofIFN-� � Ag-specific CD8� T cells in spleens 7 days after immunizationupon a 5-h stimulation with OVA257–264-peptide (■) as determined byintracellular cytokine staining. GP33–41 peptide stimulation (�) was usedas a control. B, Fold expansion of OVA257–264-specific CD8� upon in vitrostimulation with MEC.B7.Sig-OVA cells. The fold expansion is calculatedby dividing the absolute number of OVA257–264-specific CD8� cells after invitro stimulation by the absolute numbers of OVA257–264-specific CD8� at thestart of the culture. C, In vitro cytolytic activity of restimulated splenocytesfrom mice 7 days after immunization. Various E:T ratios were tested for killingof syngeneic target cells pulsed with OVA257–264 (■ ) or GP33–41 (�). Valuesrepresent mean � SEM (n � 4) and are representative of four experiments.

FIGURE 3. Cell-associated dsRNA mediates its effect through TLR3 in129/TRIFlps2/lps2 mice. 129/TRIFlps2/lps2 and control 129/B6 mice were de-pleted of CD4� cells and immunized as described in the Fig. 1 legend withEL-4-mOVA cells. A, The frequency of IFN-� � Ag-specific CD8� T cells7 days after immunization as determined by intracellular cytokine staining.(■, OVA 257–264; �, GP33–41). B, Fold expansion and of OVA257–264-spe-cific CD8� upon in vitro stimulation with MEC.B7.Sig-OVA cells. Valuesrepresent mean � SEM (n � 4). WT, wild type; e.p, electroporation.

6125The Journal of Immunology

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

DiscussionIn this study we have shown that tumor cell vaccination with cell-associated dsRNA, but not soluble dsRNA, enhances both tumor-specific CD8� and CD4� T cell responses and inhibits tumorgrowth in a therapeutic fashion when administered after tumorinoculation. The cell-associated dsRNA not only increased theclonal burst of tumor-specific CD8� T cells but also endowedthem with an enhanced capacity for expansion upon secondaryencounter with tumor Ags, even when the CD8� T cells wereprimed in the absence of CD4� T cell help. The observed adjuvanteffect of cell-associated dsRNA was fully dependent on the ex-pression of TLR3 by the APCs and their subsequent production oftype I IFNs.

Because dsRNA is produced by all viruses at some point duringtheir replication phases, the immune system has developed manypathways to sense dsRNA. Interaction of dsRNA with TLR3,PKR, RNaseL, or RIG-I has been shown to induce type I IFNs thataffect viral replication and function as adjuvants in the priming ofvirus-specific adaptive immune responses (30–33). In APCs suchas DCs, PKR, RIG-I, and RNaseL are expressed in the cytoplasmand likely recognize dsRNA produced in the context of viral rep-lication (31, 32). In contrast, TLR3 localizes to an intracellularvesicular compartment that has been suggested to be endosomal,because inhibiting the acidification of endosomes abrogatespoly(I:C) signaling (34). TLR3 is believed to encounter dsRNA inthese vesicles through the phagocytosis of apoptotic infected cells.Only nonplasmacytoid DCs of the myeloid lineage express TLR3;in the mouse, CD8�� DCs that facilitate cross-priming constitu-tively express TLR3, and the expression level is up-regulated uponDC activation and maturation. These observations coincide withour finding that the adjuvant effect of cell-associated dsRNA iscompletely abolished in 29/TRIFlps2/lps2 mice deficient in TLR3signaling but normal for PKR, RIG-I, and RNaseL signaling.

Although type I IFNs were initially characterized as potent an-tiviral factors, they have also been shown to directly suppress tu-mor cell replication, induce tumor cell apoptosis, and reduce tumorgrowth via their anti-angiogenic properties (15, 35–37). In addi-tion, type I IFNs possess antitumor properties through activation ofthe innate immune system by stimulating NK cell-mediated tumorlysis and enhancing the tumoricidal properties of macrophages (38,39). Indeed, many studies showing a therapeutic antitumor effectof soluble dsRNA-induced type I IFNs failed to enhance tumor-specific adaptive responses, indicating that the effect resulted in-stead from either a direct impact on the tumor or activation of theinnate immune response (15, 37). Our data show that only cell-associated, not soluble, dsRNA enhances tumor-specific T cell in-duction that results in an antitumor effect in vivo. This observationindicates that the production of type I IFN was insufficient alone togenerate protective innate and adaptive antitumor responses, evenwhen the tumor vaccine was administered simultaneously with sol-uble dsRNA. However, using IFN-��R�/� DC and IFN-��R�/�

mice, we demonstrate that the adjuvant effect of cell-associateddsRNA is mediated through the production of type I IFN by thecross-priming DCs upon sensing dsRNA. Although the direct ef-fect of type I IFNs on T cells is controversial (40, 41), in ourstudies the type I IFNs affected DC function in an autocrine fash-ion, because adding either dsRNA or recombinant type I IFN toIFN-��R�/� DCs did not specifically affect CD8� T cell prolif-eration. Various studies have shown that type I IFNs induce up-regulation of MHC-peptide complexes, costimulatory molecules(including CD40 and CD80/86), and the production of cytokinesand chemokines by DCs. Notably, type I IFNs can induce fullactivation of DCs, transforming a DC with potent phagocytic ca-pacity but poor stimulatory capacity into a mature DC with poorphagocytic but potent T cell activation capacities (42, 43). Thisfinding demonstrates why the correct timing of dsRNA sensing

FIGURE 4. The adjuvant effect of cell-associated dsRNA is mediated by type I IFN. A, Wild-type (WT) BM-derived DCs were cultured with irradiatedEL-4-mOVA cells-associated to synthetic dsRNA (p(IC)), EL-4-mOVA cells with soluble (sol.) dsRNA, EL-4-mOVA cells, or dsRNA alone in a con-centration representative of the amount of cell-associated dsRNA. After 24 h, type I IFN concentrations were determined using a cell line containing anIFN-stimulated response element-luciferase reporter construct (nd, not detectable). B, BM-derived DCs from indicated strains were cultured for 24 h withEL-4-mOVA cells electroporated (e.p.) in the presence or absence of synthetic dsRNA (poly(I:C)). Purified OVA-specific OT-1 CD8� T cells labeled withCFSE were added for 64 h, and proliferation was determined by the CFSE dilution profile of the CD8�� V�2� cells. C, Timing of dsRNA availabilityaffects DC presentation of exogenous proteins but not self-proteins. BM-derived DCs from wild-type male mice were cultured with EL-4-mOVA cellselectroporated in the presence or absence of synthetic dsRNA (poly(I:C)) under the indicated conditions. After 30 h, purified OT-1 or HY-specific CD8�

T cells labeled with CFSE were added. Proliferation was determined after a 64-h culture by the CFSE dilution profile of the CD8�� TCR-��� cells.Representative histograms are shown of triplicate cultures and are representative of two experiments.

6126 dsRNA-INDUCED IFN ENHANCES ANTI-TUMOR RESPONSE

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

and type I IFN production is crucial in the adjuvant effect mediatedby dsRNA. Our studies show that activation of OVA-specificCD8� T cells in vitro is enhanced when dsRNA or recombinanttype I IFN is introduced after the DC has taken up the OVA-expressing tumor cells but not when soluble dsRNA is adminis-tered before the tumor cells. Additional studies showed that DCspreviously exposed to dsRNA have reduced phagocytosis of thetumor vaccine in vitro (data not shown). Importantly, HY-specificCD8� T cells that recognized self-protein on DCs exhibited en-hanced proliferation regardless of when the dsRNA was added tothe culture. Because epitopes of self-protein are constitutively pre-sented by the MHC in the absence of cross-priming, the dsRNAenhanced the T cell response by increasing the expression of co-stimulatory molecules by the DC. Our observations also explainwhy soluble dsRNA treatment together with peptide-pulsed DCshas been shown to be effective in the induction of antitumor CD8�

T cell responses, because peptides can bind directly to MHC classI and class II without the need for uptake or processing (2, 14).

DC activation is known to be crucial for the induction of pro-tective antitumor CD8� T cell responses (1–4). Under normalcross-priming conditions, DC activation is provided by CD4� Tcell help via the interaction of CD40 on the DC with CD40L on theCD4� T cell (44–46). We and others have previously shown thatCD8� T cell priming in the absence of T cell help yields CD8� Tcells that fail to secondarily expand (20, 27, 28) and instead un-dergo TRAIL-mediated apoptosis upon encounter with their cog-nate tumor Ag (18). Our current data reveals that not only doesdsRNA enhance the tumor-specific CD4� T cell response andthereby T cell help for the priming of CD8� T cells, but it alsoactivates DCs in such a fashion that the induction of sufficientanti-tumor memory CD8� T cells becomes Th cell independent.

Although it is not clear what signals are required to induce mem-ory CD8� T cells in the absence of Th cells, the induction ofCD8� T memory in our studies is most likely mediated by theenhanced Ag-presenting capacity of the DCs, which resulted fromup-regulation of MHC class I-peptide complexes and the costimu-latory molecules CD80 and CD86 by the type I IFNs. In addition,cytokines and chemokines induced by the type I IFNs may havecontributed to a more favorable environment for T cell recruitmentand retention of the cells in the lymph nodes and their subsequentactivation.

In light of the fact that many tumor-bearing patients have com-promised immune systems as a direct result of treatment regimensincluding chemotherapeutic agents and/or radiotherapy, incorpo-ration of dsRNA in tumor vaccines represents a feasible approachto amplifying antitumor CD8� T cell responses.

The data from our studies collectively suggest that using dsRNAas a tumor vaccine adjuvant may be a suitable strategy for enhanc-ing vaccine efficacy by combining the advantages of autologous

FIGURE 5. Type I IFN mediates dsRNA adjuvanticity in vivo. Wild-type (WT), IL-12p40�/�, and IFN-��R�/� mice were depleted of CD4�

cells and immunized as described in the Fig. 1 legend with EL-4-mOVAcells. A, The frequency of IFN-�� Ag-specific CD8� T cells 7 days afterimmunization as determined by intracellular cytokine staining (■,OVA257–264; �, GP33–41). B, Fold expansion of OVA257–264-specificCD8� upon in vitro stimulation with MEC.B7.Sig-OVA cells. Values rep-resent mean � SEM (n � 4). e.p., electroporation.

FIGURE 6. Cell-associated dsRNA inhibits tumor growth and enhanceslymphocyte infiltration. A and B, CD4-depleted mice were inoculated with5 � 106 live EL-4-mOVA cells. Upon palpable tumor formation, micewere treated s.c. with 2 � 107 control electroporated (e.p.) EL-4-mOVAcells (�), control electroporated EL-4-mOVA cells in combination withsoluble (sol) dsRNA (‚), electroporated (e.p.) EL-4-mOVA cells-associ-ated dsRNA (�), or electroporated EL-4 cells-associated dsRNA (E), andtumor growth (A) and survival (B) was monitored. C, Flow cytometricanalysis of TILs 14 days after tumor inoculation. Tumors were chemicallydissociated and stained for T cells by CD8�� and TCR-���. Gating basedon forward and side scatter was used to discriminate EL-4-mOVA tumorcells from TILs. The numbers in the right corner represent the percentageof CD8�� TCR-��� in the TIL population. FACS plots represent datafrom a single tumor-bearing mouse, with analogous findings in 24 tumor-bearing mice. D, mRNA was extracted from tumors and quantitative RT-PCR was performed to determine the relative amount of CD8� mRNA(representing CD8� TIL) and tumor Ag (OVA). Data were normalized forcDNA using both L32 and �-actin and are expressed as the ratio of CD8�to tumor Ag of at least five individual tumor-bearing mice.

6127The Journal of Immunology

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from

tumor cells with a stimulus for an innate receptor expressed by allpatients. Such an approach would deliver the dsRNA to DCs in anoptimal manner because, after phagocytosis of the tumor vaccine,the dsRNA becomes available in endosomes that express the cog-nate receptor. Importantly, we have shown in a preclinical thera-peutic protocol that this approach is able to enhance the survival oftumor-bearing immunocompromised mice. We believe that the re-sults described herein provide a strong foundation for the futuredevelopment of more clinically effective strategies for tumor celltherapy in cancer patients.

AcknowledgmentsWe thank K. R. Prilliman and S. P. Schoenberger for advice and criticalreading of the manuscript.

DisclosuresThe authors have no financial conflict of interest.

References1. Schuler, G., B. Schuler-Thurner, and R. M. Steinman. 2003. The use of dendritic

cells in cancer immunotherapy. Curr. Opin. Immunol. 15: 138–147.2. Sheng, K. C., G. A. Pietersz, M. D. Wright, and V. Apostolopoulos. 2005. Den-

dritic cells: activation and maturation–applications for cancer immunotherapy.Curr. Med. Chem. 12: 1783–1800.

3. Paczesny, S., H. Ueno, J. Fay, J. Banchereau, and A. K. Palucka. 2003. Dendriticcells as vectors for immunotherapy of cancer. Semin. Cancer Biol. 13: 439–447.

4. Banchereau, J., and A. K. Palucka. 2005. Dendritic cells as therapeutic vaccinesagainst cancer. Nat. Rev. Immunol. 5: 296–306.

5. Delamarre, L., H. Holcombe, and I. Mellman. 2003. Presentation of exogenous antigenson major histocompatibility complex (MHC) class I and MHC class II molecules isdifferentially regulated during dendritic cell maturation. J. Exp. Med. 198: 111–122.

6. Michiels, A., K. Breckpot, J. Corthals, S. Tuyaerts, A. Bonehill, C. Heirman,K. Thielemans, and J. L. Aerts. 2006. Induction of antigen-specific CD8� cyto-toxic T cells by dendritic cells co-electroporated with a dsRNA analogue andtumor antigen mRNA. Gene Ther. 13: 1227–1236.

7. Verdijk, R. M., T. Mutis, B. Esendam, J. Kamp, C. J. Melief, A. Brand, andE. Goulmy. 1999. Polyriboinosinic polyribocytidylic acid (poly(I:C)) induces stablematuration of functionally active human dendritic cells. J. Immunol. 163: 57–61.

8. Kadowaki, N., S. Ho, S. Antonenko, R. W. Malefyt, R. A. Kastelein, F. Bazan,and Y. J. Liu. 2001. Subsets of human dendritic cell precursors express differenttoll-like receptors and respond to different microbial antigens. J. Exp. Med. 194:863–869.

9. Reis e Sousa, C. 2004. Toll-like receptors and dendritic cells: for whom the bugtolls. Semin. Immunol. 16: 27–34.

10. Hoebe, K., E. M. Janssen, S. O. Kim, L. Alexopoulou, R. A. Flavell, J. Han, andB. Beutler. 2003. Up-regulation of costimulatory molecules induced by lipopoly-saccharide and double-stranded RNA occurs by Trif-dependent and Trif-inde-pendent pathways. Nat. Immunol. 4: 1223–1229.

11. den Haan, J. M., S. M. Lehar, and M. J. Bevan. 2000. CD8� but not CD8�

dendritic cells cross-prime cytotoxic T cells in vivo. J. Exp. Med. 192:1685–1696.

12. Janssen, E., K. Tabeta, M. Barnes, S. Rutschmann, S. McBride, K. Bahjat,S. P. Schoenberger, A. N. Theophilopoulos, B. Beutler, and K. Hoebe. 2006. Acell-death induced TLR-independent pathway drives efficient T cell activation.Immunity 24: 787–799.

13. Datta, S. K., V. Redecke, K. R. Prilliman, K. Takabayashi, M. Corr, T. Tallant,J. DiDonato, R. Dziarski, S. Akira, S. P. Schoenberger, and E. Raz. 2003. Asubset of Toll-like receptor ligands induces cross-presentation by bone marrow-derived dendritic cells. J. Immunol. 170: 4102–4110.

14. Salem, M. L., A. N. Kadima, D. J. Cole, and W. E. Gillanders. 2005. Defining theantigen-specific T-cell response to vaccination and poly(I:C)/TLR3 signaling:evidence of enhanced primary and memory CD8 T-cell responses and antitumorimmunity. J. Immunother. 28: 220–228.

15. Fujimoto, C., Y. Nakagawa, K. Ohara, and H. Takahashi. 2004. Polyriboinosinicpolyribocytidylic acid [poly(I:C)]/TLR3 signaling allows class I processing ofexogenous protein and induction of HIV-specific CD8� cytotoxic T lympho-cytes. Int. Immunol. 16: 55–63.

16. Schulz, O., S. S. Diebold, M. Chen, T. I. Naslund, M. A. Nolte, L. Alexopoulou,Y. T. Azuma, R. A. Flavell, P. Liljestrom, and C. Reis e Sousa. 2005. Toll-likereceptor 3 promotes cross-priming to virus-infected cells. Nature 433: 887–892.

17. Hoebe, K., X. Du, P. Georgel, E. Janssen, K. Tabeta, S. O. Kim, J. Goode, P. Lin,N. Mann, S. Mudd, et al. 2003. Identification of Lps2 as a key transducer ofMyD88-independent TIR signalling. Nature 424: 743–748.

18. Janssen, E. M., N. M. Droin, E. E. Lemmens, M. J. Pinkoski, S. J. Bensinger,B. D. Ehst, T. S. Griffith, D. R. Green, and S. P. Schoenberger. 2005. CD4�

T-cell help controls CD8� T-cell memory via TRAIL-mediated activation-in-duced cell death. Nature 434: 88–93.

19. van Stipdonk, M. J., E. E. Lemmens, and S. P. Schoenberger. 2001. Naive CTLsrequire a single brief period of antigenic stimulation for clonal expansion anddifferentiation. Nat. Immunol. 2: 423–429.

20. Janssen, E. M., E. E. Lemmens, T. Wolfe, U. Christen, M. G. von Herrath, andS. P. Schoenberger. 2003. CD4� T cells are required for secondary expansion andmemory in CD8� T lymphocytes. Nature 421: 852–856.

21. Matzinger, P. 1991. The JAM test. A simple assay for DNA fragmentation andcell death. J. Immunol. Methods 145: 185–192.

22. Taguchi, T., J. R. McGhee, R. L. Coffman, K. W. Beagley, J. H. Eldridge,K. Takatsu, and H. Kiyono. 1990. Detection of individual mouse splenic T cellsproducing IFN-� and IL-5 using the enzyme-linked immunospot (ELISPOT) as-say. J. Immunol. Methods 128: 65–73.

23. Belldegrun, A., A. Kasid, M. Uppenkamp, S. L. Topalian, and S. A. Rosenberg.1989. Human tumor infiltrating lymphocytes. Analysis of lymphokine mRNAexpression and relevance to cancer immunotherapy. J. Immunol. 142:4520–4526.

24. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, andG. Schuler. 1999. An advanced culture method for generating large quantities ofhighly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223:77–92.

25. Jiang, Z., P. Georgel, X. Du, L. Shamel, S. Sovath, S. Mudd, M. Huber, C. Kalis,S. Keck, C. Galanos, et al. 2005. CD14 is required for MyD88-independent LPSsignaling. Nat. Immunol. 6: 565–570.

26. Li, M., G. M. Davey, R. M. Sutherland, C. Kurts, A. M. Lew, C. Hirst,F. R. Carbone, and W. R. Heath. 2001. Cell-associated ovalbumin is cross-pre-sented much more efficiently than soluble ovalbumin in vivo. J. Immunol. 166:6099–6103.

27. Sun, J. C., and M. J. Bevan. 2003. Defective CD8 T cell memory following acuteinfection without CD4 T cell help. Science 300: 339–342.

28. Shedlock, D. J., and H. Shen. 2003. Requirement for CD4 T cell help in gener-ating functional CD8 T cell memory. Science 300: 337–339.

29. Langrish, C. L., B. S. McKenzie, N. J. Wilson, R. de Waal Malefyt,R. A. Kastelein, and D. J. Cua. 2004. IL-12 and IL-23: master regulators of innateand adaptive immunity. Immunol. Rev. 202: 96–105.

30. Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognitionof double-stranded RNA and activation of NF-�B by Toll-like receptor 3. Nature413: 732–738.

31. Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi,M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA helicase RIG-Ihas an essential function in double-stranded RNA-induced innate antiviral re-sponses. Nat. Immunol. 5: 730–737.

32. Clemens, M. J., and A. Elia. 1997. The double-stranded RNA-dependent proteinkinase PKR: structure and function. J. Interferon Cytokine Res. 17: 503–524.

33. Saunders, L. R., and G. N. Barber. 2003. The dsRNA binding protein family:critical roles, diverse cellular functions. FASEB J. 17: 961–983.

34. Matsumoto, M., K. Funami, M. Tanabe, H. Oshiumi, M. Shingai, Y. Seto,A. Yamamoto, and T. Seya. 2003. Subcellular localization of Toll-like receptor3 in human dendritic cells. J. Immunol. 171: 3154–3162.

35. De Clercq, E., Z. X. Zhang, K. Huygen, and R. Leyten. 1982. Inhibitory effect ofinterferon on the growth of spontaneous mammary tumors in mice. J. Natl. Can-cer Inst. 69: 653–657.

36. Davies, M. E., and A. K. Field. 1983. Effect of poly I:C/poly-L-lysine (poly ICL)on the development of murine osteogenic sarcoma. J. Interferon Res. 3: 89–95.

37. Bloksma, N., C. F. Kuper, F. M. Hofhuis, B. Benaissa-Trouw, and J. M. Willers.1983. Antitumour activity of endotoxin, concanavalin A and poly I:C and theirability to elicit tumour necrosis factor, cytostatic factors, and interferon in vivo.Cancer Immunol. Immunother. 16: 35–39.

38. Matsumoto, S., M. Miyagishi, H. Akashi, R. Nagai, and K. Taira. 2005. Analysisof double-stranded RNA-induced apoptosis pathways using interferon-responsenoninducible small interfering RNA expression vector library. J. Biol. Chem.280: 25687–25696.

39. Sivori, S., M. Falco, M. Della Chiesa, S. Carlomagno, M. Vitale, L. Moretta, andA. Moretta. 2004. CpG and double-stranded RNA trigger human NK cells byToll-like receptors: induction of cytokine release and cytotoxicity against tumorsand dendritic cells. Proc. Natl. Acad. Sci. USA 101: 10116–10121.

40. Bahl, K., S. K. Kim, C. Calcagno, D. Ghersi, R. Puzone, F. Celada, L. K. Selin,and R. M. Welsh. 2006. IFN-induced attrition of CD8 T cells in the presence orabsence of cognate antigen during the early stages of viral infections. J. Immunol.176: 4284–4295.

41. Le Bon, A., V. Durand, E. Kamphuis, C. Thompson, S. Bulfone-Paus,C. Rossmann, U. Kalinke, and D. F. Tough. 2006. Direct stimulation of T cellsby type I IFN enhances the CD8� T cell response during cross-priming. J. Im-munol. 176: 4682–4689.

42. Le Bon, A., and D. F. Tough. 2002. Links between innate and adaptive immunityvia type I interferon. Curr. Opin. Immunol. 14: 432–436.

43. Montoya, M., G. Schiavoni, F. Mattei, I. Gresser, F. Belardelli, P. Borrow, andD. F. Tough. 2002. Type I interferons produced by dendritic cells promote theirphenotypic and functional activation. Blood 99: 3263–3271.

44. Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, andW. R. Heath. 1998. Help for cytotoxic-T-cell responses is mediated by CD40signalling. Nature 393: 478–480.

45. Schoenberger, S. P., R. E. Toes, E. I. van der Voort, R. Offringa, and C. J. Melief.1998. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40Linteractions. Nature 393: 480–483.

46. Ridge, J. P., F. Di Rosa, and P. Matzinger. 1998. A conditioned dendritic cell canbe a temporal bridge between a CD4� T-helper and a T-killer cell. Nature 393:474–478.

6128 dsRNA-INDUCED IFN ENHANCES ANTI-TUMOR RESPONSE

by guest on June 18, 2013http://w

ww

.jimm

unol.org/D

ownloaded from