Immunostimulatory RNA Oligonucleotides Induce an EffectiveAntitumoral NK Cell Response through the TLR71

Carole Bourquin,2,3* Laura Schmidt,2* Anna-Lisa Lanz,* Bettina Storch,*Cornelia Wurzenberger,* David Anz,* Nadja Sandholzer,* Ralph Mocikat,‡ Michael Berger,*†

Hendrik Poeck,*§ Gunther Hartmann,§ Veit Hornung,§ and Stefan Endres*

RNA oligonucleotides containing immune-activating sequences promote the development of cytotoxic T cell and B cell responsesto Ag. In this study, we show for the first time that immunostimulatory RNA oligonucleotides induce a NK cell response thatprevents growth of NK-sensitive tumors. Treatment of mice with immunostimulatory RNA oligonucleotides activates NK cells ina sequence-dependent manner, leading to enhanced IFN-� production and increased cytotoxicity. Use of gene-deficient miceshowed that NK activation is entirely TLR7-dependent. We further demonstrate that NK activation is indirectly induced throughIL-12 and type I IFN production by dendritic cells. Reconstitution of TLR7-deficient mice with wild-type dendritic cells restoresNK activation upon treatment with immunostimulatory RNA oligonucleotides. Thus, by activating both NK cells and CTLs, RNAoligonucleotides stimulate two major cellular effectors of antitumor immunity. This dual activation may enhance the efficacy ofimmunotherapeutic strategies against cancer by preventing the development of tumor immune escape variants. The Journal ofImmunology, 2009, 183: 6078–6086.

V iruses are recognized by the innate immune systemthrough endosomal and cytosolic receptors that selec-tively detect conserved pathogen structures (1). Molec-

ular patterns specific to nucleic acids from viruses are recognizedby the cytoplasmic helicases retinoic acid-inducible gene I (RIG-I)4 and melanoma differentiation-associated gene 5 (MDA-5) andby endosomal receptors of the TLR family, the TLRs 3, 7, 8, and9 (1–8). Binding of nucleic acid ligands to their respective receptoractivates intracellular signaling cascades that rapidly induce innateand adaptive immune responses (1).

An important goal of therapeutic cancer vaccines is the induc-tion of Ag-specific responses that mediate protective immunityagainst tumors. Because of their immunostimulatory properties,nucleic acid ligands for TLRs are powerful tools that can be used

to boost tumor-specific immune responses. Indeed, inclusion incancer vaccines of CpG oligodeoxynucleotides, that bind TLR9,leads to the production of IFN-� and potentiates immune responsesto tumor Ags in patients (9). Within the immune system, humanTLR9 is expressed mainly on B cells and a subset of dendritic cells(DC), the plasmacytoid DC. The receptors TLR7 and TLR8 thatare activated by ssRNA of viral origin are expressed in addition byhuman myeloid DC and monocytes that are essential for Ag pre-sentation and for the initiation of immune responses against tumorAgs (5, 6, 10, 11). We have recently described RNA sequencesthat stimulate immune responses through the TLRs 7 and 8 (12).We have further shown that ssRNA oligonucleotides that activateTLR7 can trigger the generation of CTL by inducing Th1-typeimmunity (13).

In addition to the Ag-specific immune responses effected byCTL, a major component of antitumoral immunity is the innate NKcell response (14). NK cells are involved both in tumor immuno-surveillance and in the rejection of established tumors and canprevent the dissemination of metastases (14, 15). The ideal NK celltargets are tumor cells that have lost expression of MHC class I orthat overexpress ligands for the activating receptor NKG2D (16).Upon interaction with target cells, NK cells exert cytotoxic func-tions that are determined by a balance of multiple activating andinhibitory signals (16). Two important antitumoral effector mech-anisms triggered by target cell interaction are the direct cytolyticactivity of NK cells and the production of IFN-�. Strategies toexploit NK cell-mediated immunity may thus hold strong potentialfor cancer therapy.

The therapeutic potential of TLR7 agonists is supported by en-couraging results with a recently developed class of antitumoragents, the imidazoquinolines, that acts in part through the activa-tion of TLR7 (17). The lead compound, imiquimod, is howeveronly approved for the treatment of skin tumors by topical use (18).Although some imidazoquinoline compounds support antitumoralNK responses in i.v. disseminated tumor models (19), in solidtumors injection of imiquimod is only effective locally but not at

*Center for Integrated Protein Science Munich and Division of Clinical Pharmacol-ogy, Department of Internal Medicine and †Dr. von Haunersches Kinderspital, De-partment of Pediatric Surgery, Ludwig-Maximilian University of Munich, Munich,Germany; ‡Helmholtz Zentrum Munchen, Institute of Molecular Immunology, Mu-nich, Germany; and §Institute of Clinical Chemistry and Pharmacology, UniversityHospital, University of Bonn, Bonn, Germany

Received for publication May 20, 2009. Accepted for publication September 4, 2009.

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 study was supported by grants from the German Research Foundation (Deut-sche Forschungsgemeinschaft En 169/7-2 and Graduiertenkolleg 1202 to C.B. andS.E.), the excellence cluster CIPSM 114 (to S.E.) and the SFB-TR 36 (to S.E.), theLMUexcellent (research professorship to S.E.), the Else-Kroner Fresenius Founda-tion, and the BayImmuNet (to C.B. and S.E.). This work is part of the doctoral thesisof L.S. and A.L.L. supported by Graduiertenkolleg 1202.2 C.B. and L.S. contributed equally to this study.3 Address correspondence and reprint requests to Dr. Carole Bourquin, Division ofClinical Pharmacology, University of Munich, Ziemsenstrasse 1, 80336 Munich, Ger-many. E-mail address: carole.bourquin@med.lmu.de4 Abbreviations used in this paper: RIG-I, retinoic acid-inducible gene I; DC, den-dritic cell; DOTAP, N-[1-(2,3-dioleoxy)propyl]-N,N,N-trimethylammonium methyl-sulfate; MDA-5, melanoma differentiation-associated gene 5; PD, phosphodiester;PTO, phosphorothioate; �2m, �2-microglobulin; wt, wild type; BMDC, bone marrow-derived DC.

Copyright © 2009 by The American Association of Immunologists, Inc. 0022-1767/09/$2.00

The Journal of Immunology

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0901594

distant sites (20). Due to the short half-life of these small mole-cules in vivo, frequent applications are necessary.

In this study, we show for the first time that immunostimulatoryRNA oligonucleotides elicit an effective antitumoral NK responsein vivo through TLR7. We demonstrate that RNA oligonucleotidetreatment selectively inhibits growth of MHC class I-negative tu-mors but not growth of MHC class I-expressing tumors. Stimula-tion of mice with immunostimulatory RNA oligonucleotides leadsto an activated NK phenotype with enhanced IFN-� productionand increased cytotoxicity. These effector functions are triggeredindirectly through activation of DC to produce IL-12 and type IIFN. We further show that oligonucleotide-activated DC are suf-ficient to promote NK activation in vivo. Thus, the therapeuticapplication of RNA oligonucleotides represents a promising strat-egy to stimulate NK cell responses to induce effective antitumorimmunity.

Materials and MethodsMice and cell lines

Female C57BL/6 mice were purchased from Harlan-Winkelmann. TLR7-deficient mice (C57BL/6 background) were a gift from S. Akira (OsakaUniversity, Osaka, Japan) and bred in the animal facilities of the MedicalDepartment of the Ludwig-Maximilian University of Munich (Munich,Germany). Female �2-microglobulin (�2m)-deficient mice and IL-12p40-deficient mice (both C57BL/6 background) were from Charles River Lab-oratories. IFNAR-deficient mice (129Sv background) and wild-type (wt)129Sv mice as controls were provided by Dr. H. J. Anders (Ludwig-Maxi-milian University of Munich, Munich, Germany). Mice were 5–12 wk ofage at the onset of the experiments. All animal studies were approved bythe local regulatory agency (Regierung von Oberbayern, Munich, Ger-many). The C57BL/6-derived T cell lymphoma cell line RMA and theTAP-deficient variant RMA-S were provided by Dr. J. Charo (Berlin,Germany).

Oligonucleotides

The 20-mer oligoribonucleotides 9.2dr (5�-UGUCCUUCAAUGUCCUUCAA-3�) and poly(A) in both the unmodified phosphodiester (PD) and fullyphosphorothioate (PTO) forms were purchased from CureVac. To protectagainst degradation, all oligoribonucleotides used were PTO unless indi-cated otherwise. The PTO-modified CpG oligodeoxyribonucleotide 1826(5�-TCCATGACGTTCCTGACGTT-3�) was obtained from the ColeyPharmaceutical Group (Langenfeld, Germany). Oligonucleotides werecomplexed with N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammo-nium methylsulfate (DOTAP; Roche) before use by incubation for 20 minat a ratio of 1:2–1:5 (w/w).

For stimulation of human NK cells, the sequence 9.3as (5�-UGGUAAUUGAAGGACAGGU-3�; CureVac) was used. Two hundred nanogramsof RNA was complexed with the polycationic polypeptide poly-L-arginineP7762 (Sigma-Aldrich). Stimulation with RNA was compared to stimula-tion with the small molecule TLR7 and TLR8 agonists 3M-001, 3M-002,and 3M-003 (3M Pharmaceuticals).

Generation of bone marrow-derived DC and cell isolations

To prepare bone marrow-derived DC (BMDC), bone marrow cells fromC57BL/6 mice were cultured in complete RPMI 1640 (10% FCS, 2 mML-glutamine, 100 �g/ml streptomycin, and 1 IU/ml penicillin) supple-mented with 20 ng/ml GM-CSF and 20 ng/ml IL-4 (Tebu Bio) (DC me-dium). On day 7, loosely adherent cells were harvested and washed.Splenic DC were isolated by MACS of splenocytes after labeling withanti-CD11c microbeads (Miltenyi Biotec). DC-depleted splenocytes con-tained �2% of CD11c� cells. NK cells were isolated from total spleno-cytes by sorting with anti-DX5 (panNK) microbeads (purity �95%; Milte-nyi Biotec).

Isolation and culture of human monocytes and NK cells

Human PBMCs were prepared by Ficoll-Hypaque density gradient cen-trifugation of heparinized blood from healthy donors. NK cells were pre-pared with a NK Isolation Kit II (Miltenyi Biotec). Additional anti-CD4microbeads were added to the kit to exclude contamination with monocytesand DC. Evaluation of the NK cell population showed purity of �97%(SEM, 2.5%) and absence of CD4� cells (�0.1%). Monocytes were pre-pared with a Monocyte Isolation Kit II (Miltenyi Biotec). Anti-CD56 mi-

crobeads were added to the kit to exclude contamination with NK cells.The purity of isolated monocytes (CD11c�, HLA-DR�, and CD14�) was96% (SEM, 2%); NK cells were absent (�0.1%). Omitting the additionalCD4� depletion (for NK cells) or CD56� depletion (for monocytes) re-sulted in an up to 2% contamination with the respective cell population.Cells were cultured in complete RPMI 1640 supplemented with 10 mM4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid (Sigma-Aldrich).

NK activation in vitro

Splenocytes, splenic DC, or BMDC were stimulated with 10 �g/ml 9.2dr orpoly(A) or 1 �g/ml CpG complexed with DOTAP for 6 h at 1 � 106/ml in 200�l of complete medium. For NK-DC cocultures, stimulated DC were washedtwice and 2 � 105 purified NK cells or 2 � 105 total splenocytes in 200 �l offresh medium/well were added. For cultures of NK cells with DC supernatant,200 �l of BMDC supernatant was added to 2 � 105 purified NK cells or totalsplenocytes. Cell activation was measured 18 h later. Neutralizing Abs againstIL-2 (20 �g/ml, clone JES6-1A12; R&D Systems), IL-6 (5 �g/ml, cloneMP520F3; R&D Systems), IL-12/IL-23p40 (40 �g/ml, clone C17.8;BioLegend), and IL-15 (2 �g/ml, clone AIO.3; MBL) were added to the su-pernatant, as indicated, 2 h before coincubation with NK cells.

Human NK cells (1 � 105 in 200 �l of medium) were cocultured withaccessory cells (monocytes) at the indicated ratios. Cells were stimulatedfor 12 or 36 h with 1 �g/ml 9.3as or 10 �M 3M-001, 3M-002, or 3M-003before cell activation or IFN-� production were measured.

In vivo NK activation and tumor therapy

For in vivo immunostimulation, 20–40 �g of oligonucleotides complexedwith DOTAP were injected i.v. or 100 �g of uncomplexed CpG wereinjected s.c. into TLR7-deficient or wt C57BL/6 mice. Single-cell suspen-sions were prepared from spleen 22 h after injection unless indicated oth-erwise. For DC transfer, TLR7-deficient mice were injected twice i.v. with2.5 � 107 wt BMDC in 200 �l of Hepes-buffered saline at a 2-h interval.Wild-type BMDC were either unstimulated or stimulated in vitro with 1�g/ml 9.2dr for 6 h before transfer. Two hours after the cell transfer, micewere injected i.v. with RNA oligonucleotides. For tumor therapy, groups offive mice were injected on day 0 with 106 TAP-deficient RMA-S or wtRMA tumor cells s.c. Mice were treated i.v. with RNA oligonucleotides ondays 0, 3, and 6 (RMA-S) and additionally on days 9 and 12 (RMA). Fordepletion of NK cells, 0.5 �g of the IL-2R chain-specific mAb TM�1 wasgiven i.p. 2 days before and 2 days after tumor challenge as describedpreviously (21). For depletion of CD8 T cells, 0.1 �g of the RmCD8 mAbwas given i.p. 2 and 1 days before and 7 days after tumor challenge.Efficacy of depletion was confirmed by flow cytometry. Tumor size wasexpressed as the product of the perpendicular diameters of individual tu-mors. For assessing NK activation after repeated injections of RNA oli-gonucleotides in tumor-bearing mice, 4 � 105 B16 cells were injected i.v.and animals were treated with RNA oligonucleotides eight times at 3-dayintervals.

Flow cytometry and ELISA

Single-cell suspensions from splenocytes or blood were treated with ammo-nium chloride buffer to lyse erythrocytes. RMA-S tumors were mechanicallydisrupted and incubated with collagenase (Sigma Aldrich) at 37°C for 20 min,then passed through a 40-�m cell strainer (BD Biosciences) to obtain single-cell suspensions. For analysis of activation markers, cells were stained withfluorochrome-coupled mAbs and analyzed by flow cytometry. Anti-mouseCD3-allophycocyanin, NK1.1-PerCP, and IFN-�-FITC, anti-mouse and anti-human CD69-PE, and anti-human leukocyte Ag DR-PerCP were from BDBiosciences.

To assess IFN-� production by NK cells, splenocytes were isolated 4 hafter RNA injection and incubated for 4 h with brefeldin A at a concen-tration of 1 �g/ml. Cells were stained with PerCP-conjugated anti-NK1.1(BD Biosciences), fixed with 2% paraformaldehyde, and permeabilized(0.5% BSA, 0.5% saponin, and 0.02% sodium azide in PBS). Fixed cellswere stained with FITC-conjugated anti-IFN-� Ab (BD Biosciences) for 25min. The percentage of IFN-�-positive NK cells was determined by flowcytometry. Levels of human or murine IFN-� in the supernatant were quan-tified by ELISA (BD Biosciences) according to the manufacturer’sprotocol.

Cytotoxicity assays

For in vivo determination of cytotoxicity, target splenocytes from wt and�2m-deficient mice were labeled for 15 min at room temperature with 15and 0.15 pM CFSE, respectively. Cells were washed twice with PBS andresuspended at 1 � 108/ml. The two populations were mixed at a 1:1 ratioand 1 � 107 cells were injected i.v. into wt C57BL/6 mice injected 4 h

6079The Journal of Immunology

previously with 9.2dr. CFSE staining in splenocytes was analyzed by flowcytometry 4–10 h after target cell injection. Specific lysis was calculated asfollows: specific lysis (percent) � 100 � [100 � (CFSElow cells in stim-ulated mice/CFSEhigh cells in stimulated mice)/(CFSElow cells in unstimu-lated mice/CFSEhigh cells in unstimulated mice)] (22).

For determination of cytotoxicity in vitro, splenic NK cells were eitherisolated from treated mice 4 h after oligonucleotide injection or were ac-tivated with 100 �l of supernatant from RNA-stimulated DC for 4 h. NKcells were plated at 5 � 105 cells/100 �l in complete RPMI 1640 withoutphenol red (PAA). YAC-1 target cells (1 � 106 cells/ml) were labeled with10 �M AM-calcein (Invitrogen) for 30 min at 37°C, washed three timeswith PBS, and resuspended at 5 � 104 cells/ml. NK cells were incubatedwith labeled target cells for 4 h at 37°C, 150 �l of supernatant was trans-ferred to a new plate, and fluorescence was measured at 485/535 nm aspreviously described (23). The background cytotoxic activity of TLR7-deficient NK cells from untreated mice was similar to that of TLR7-defi-cient NK cells from RNA-treated mice.

Statistics

Comparisons in tumor size among groups were made using the Mann-Whitney U test for various time points. For cytokine ELISAs and cytotox-icity assays, significance was assessed by Student’s t test. Statistical anal-ysis was performed using SPSS software. Error bars indicate SEM.

ResultsImmunostimulatory RNA sequences suppress growth ofMHC-negative tumors

We have recently described immunostimulatory sequences withinRNA oligonucleotides that activate a Th1-type immune response,an important effector of antitumoral immunity (12, 13). To assesswhether immunostimulatory RNA oligonucleotides can suppresstumor growth through activation of NK cells, the second majorcellular component of antitumoral immunity, we examined the ef-ficacy of RNA oligonucleotides in the treatment of RMA-S tu-mors. The RMA-S cell line is a MHC class I-negative variant of

the RMA lymphoma that is selectively targeted by NK cells,unlike the MHC-sufficient parent cell line (24). Mice weretreated with the highly immunostimulatory oligonucleotide9.2dr (12, 13) after inoculation of either NK-sensitive RMA-Scells or wt RMA cells. Three injections of the 9.2dr oligonu-cleotide complexed with DOTAP at 3-day intervals suppressedprogression of RMA-S tumors compared with untreated mice(Fig. 1A). In contrast, after injection of a control poly(A) se-quence of the same length, no effect on tumor growth was ob-served. Treatment with RNA oligonucleotides did not preventtumor progression in the MHC-sufficient RMA tumors (Fig.1B), suggesting that NK cells are the critical effectors of theantitumor effect of immunostimulatory RNA. Within the tumor,NK cells represented 11.6% of small lymphocytes (�2.4%; n �5). Results for one tumor are shown in Fig. 1C. Indeed, duringtreatment with 9.2dr, RMA-S tumors progressed more rapidlyin NK-depleted mice than in CD8-depleted mice or undepletedmice (Fig. 1D).

FIGURE 1. RNA oligonucleotides suppress growth of MHC-negativetumors. Mice were inoculated s.c. on day 0 with 106 RMA-S (A, C, and D)or RMA (B) tumor cells. Mice were treated at 3-day intervals from day 0with 20 �g PTO 9.2dr or poly(A) i.v. complexed with DOTAP. A and B,Data show the mean tumor sizes � SEM (n � 5) and are representative oftwo independent experiments. C, Percentage of NK cells (NK1.1�CD3�,gated on small lymphocytes) in RMA-S tumors was measured by flowcytometry. Data are shown for one representative tumor (n � 5). D, Miceinjected with NK-depleting (depl.) or CD8-depleting Abs were treated with9.2dr as in A. �, p � 0.05; ��, p � 0.01; ns, not significant; comparison topoly(A)-treated group (A), untreated (B), or undepleted group (D). SSC,Side scatter.

FIGURE 2. RNA oligonucleotides rapidly activate NK cells in vivo.Mice were injected i.v. with 20 �g of poly(A) or 9.2dr with a PD or PTObackbone complexed with DOTAP. A, Surface expression of CD69 onsplenic NK cells (NK1.1�CD3�) was measured by flow cytometry 22 hafter injection. Data show the mean of individual mice for each group (n �3 � SEM). Results are representative of four independent experiments. B,Percentage of IFN-�-producing NK cells in the spleen 4 h after injectionwithout further stimulation (n � 6). C and D, In vivo NK cytotoxicity. Fourhours after RNA injection, mice were injected with target cells from wt(CFSEhigh) and �2m-deficient mice (CFSElow). Splenocytes were analyzedby flow cytometry 10 h after cell transfer. C, One representative mouse pergroup is shown. Numbers indicate percentage of cells in the indicated gate.D, Specific lysis for individual mice (data points, n � 3) and mean (bar).B–D, Results are representative of two independent experiments. E, CD69was measured as in A on circulating and intratumoral NK cells in micebearing RMA-S tumors (n � 6) treated with 9.2dr as in Fig. 1. F, CD69was measured on circulating NK cells 22 h after either one injection or thelast of eight injections of 9.2dr (at 3-day intervals) in mice inoculated i.v.with B16 cells (n � 5). �, p � 0.05; ��, p � 0.01; and ���, p � 0.001. Nostim, No stimulation.

6080 ANTITUMOR NK RESPONSE BY RNA OLIGONUCLEOTIDES

Immunostimulatory RNA sequences rapidly activate NK cellsin vivo

To characterize the effect of RNA oligonucleotides on NK cells,mice were injected i.v. with RNA oligonucleotides complexedwith DOTAP. In addition to oligonucleotides comprising a PTObackbone, oligonucleotides of the same sequence with an unmod-ified PD backbone were used. As shown in Fig. 2A, a single ap-plication of 9.2dr PTO resulted in up-regulation of the early acti-vation molecule CD69 on �80% of splenic NK cells. CD69expression was also up-regulated on a small proportion of spleno-cytes in mice injected with the unmodified 9.2dr PD oligonucleo-tide. Furthermore, 9.2dr PTO enhanced production of IFN-� byNK cells as early as 4 h after injection (Fig. 2B). Notably, thisincrease in IFN-� production was observed directly ex vivo with-out the need for restimulation in culture by target cells. Activationof NK cells was clearly sequence dependent, since NK cells werenot activated by treatment with poly(A) oligonucleotides of thesame length in either the PTO or PD form. Because of their stron-ger immune-activating capacity, PTO oligonucleotides were usedfor all additional experiments.

To assess the cytotoxicity of NK cells following RNA oligonu-cleotide application, an in vivo assay was performed (22). In thisassay CFSE-labeled splenocytes from �2m-deficient mice were in-jected i.v. as target cells. Splenocytes from wt mice labeled with ahigher CFSE concentration were injected simultaneously as refer-ence. Blood was collected from recipients 10 h after target cellinjection and PBMC were analyzed for CFSE staining by flow

cytometry. Lysis of CFSElow �2m-deficient cells was selectivelyincreased in 9.2dr PTO-treated mice with a specific lysis of �60%(Fig. 2, C and D). Indeed, an increased specific lysis was detectedas early as 4 h after target cell injection (data not shown).

We then examined CD69 expression on NK cells from tumor-bearing mice. As with healthy animals, CD69 was up-regulated inmice bearing RMA-S tumors upon treatment with 9.2dr PTO (Fig.2E). Within the tumor, NK cells also showed increased CD69 ex-pression following treatment with 9.2dr PTO (Fig. 2E). CD69 up-regulation was also seen after the last of three RNA injections,suggesting that activation of NK cells by RNA oligonucleotides isnot subject to a decreased response after repeated administration.This was confirmed in another tumor model after i.v. injection ofB16 melanoma cells: NK activation was not impaired even aftereight applications of 9.2dr PTO over 21 days (Fig. 2F). Thus, RNAoligonucleotides activate NK cells in vivo in a sequence-dependentmanner, without the need for extensive restimulation by targetcells.

In vivo NK activation by RNA oligonucleotides is mediatedthrough TLR7

RNA from viral origin is recognized by a variety of different re-ceptors of the innate immune system, including TLR7 (1). To de-termine whether NK cell activation by RNA oligonucleotides ismediated through TLR7, TLR7-deficient and wt mice were in-jected with 9.2dr complexed with DOTAP. As shown in Fig. 3A,TLR7-deficient mice injected with 9.2dr showed no up-regulation

FIGURE 3. NK activation by RNA oligonucleotidesis mediated through TLR7. wt and TLR7-deficient micewere injected with 9.2dr as in Fig. 2 or with CpG. A,Surface expression of CD69 on splenic NK cells for onerepresentative mouse per group and mean � SEM (n �3). B, Percentage of IFN-�-producing NK cells in thespleen for one representative mouse per group. C, NKcell cytotoxicity was measured ex vivo. Splenic NKcells were isolated 4 h after injection and cytotoxicityagainst YAC-1 cells was measured for the indicated E:Tratios. ��, p � 0.01; comparison to TLR7�/� 9.2dr.Results are representative of two independent experi-ments. No stim, No stimulation.

6081The Journal of Immunology

of the early activation molecule CD69 on NK cells. In contrast,injection of the CpG oligodeoxynucleotide 1826, a ligand forTLR9, resulted in expression of CD69 on 80% of NK cells inTLR7-deficient mice, demonstrating that TLR7-deficient NK cellscan respond to activation through another TLR. Also, the percent-age of IFN-�-producing NK cells was not significantly increasedby 9.2dr in TLR7-deficient mice (Fig. 3B). Furthermore, NK cellsisolated 4 h after RNA oligonucleotide injection showed TLR7-dependent cytotoxicity. NK cells from 9.2dr-treated wt mice ef-fected a specific lysis of �60% against the NK-sensitive cell lineYAC-1 (Fig. 3C) in a standard in vitro cytotoxicity assay (25). Incontrast, NK cells from TLR7-deficient mice showed no cytotoxicactivity upon 9.2dr stimulation. Thus, NK cell activation by RNAoligonucleotides is TLR7-dependent.

DC are essential for NK cell activation

We have previously shown that stimulation with RNA oligonucle-otides leads to DC activation through TLR7 (12, 13). Because NKcells can be efficiently activated by DC (26, 27), we investigatedwhether DC play a role in NK-cell activation by RNA oligonu-cleotides. When total splenocytes from naive mice were stimulatedwith 9.2dr, NK cells showed strong CD69 up-regulation (Fig. 4A).In contrast, when DC were depleted from splenocytes by MACSbefore stimulation, CD69 up-regulation on NK cells was abol-

ished. We then activated BMDC with 9.2dr, washed them ex-tensively to remove oligonucleotides, and cocultured them withnaive splenocytes. The NK cell fraction within these spleno-cytes showed an activated phenotype that was associated withdose-dependent production of IFN-� (Fig. 4B). Similar resultswere observed when 9.2dr-stimulated splenic DC were used toactivate splenocytes (data not shown). Isolated NK cells werealso activated by RNA-stimulated DC and induced to producehigh amounts of IFN-� (Fig. 4B). Thus, DC are necessary andsufficient to induce NK cell activation and IFN-� productionthrough RNA oligonucleotides.

To investigate whether activation of NK cells by DC was me-diated by soluble factors, BMDC were activated with 9.2dr,washed to remove oligonucleotides, and cultured for a further timeperiod of 18 h. Supernatant from the culture was then added tounstimulated splenocytes or to purified NK cells. In both cases,supernatant from activated DC but not from unstimulated DC ledto up-regulation of CD69 on NK cells in a dose-dependent manner(Fig. 4C). Supernatant from DC activated with the higher dose ofRNA oligonucleotides led to IFN-� production by both spleno-cytes and purified NK cells (Fig. 4C). Furthermore, supernatantfrom activated DC but not from DC stimulated with poly(A) in-duced NK cell cytotoxicity (Fig. 4D). Thus, NK cells are stimu-lated through soluble factors produced by activated DC.

FIGURE 4. RNA oligonucleotides stimulate NKcells in a DC-dependent manner. A, Expression of CD69on NK cells was measured 6 h after stimulation of totalsplenocytes or CD11c-depleted splenocytes with 10�g/ml 9.2dr or poly(A) complexed with DOTAP. B,BMDC were stimulated with 9.2dr for 6 h or left un-stimulated. Cells were washed to remove oligonucleo-tides and cocultured for 18 h with splenocytes or puri-fied NK cells. CD69 expression on NK cells and IFN-�secretion were measured. C and D, BMDC were stim-ulated as in B, washed, and cultured for an additional18 h. BMDC culture supernatant was harvested andadded to splenocytes or purified NK cells. C, Expressionof CD69 on NK cells and IFN-� production were as-sessed after 18 h. D, NK cytotoxicity was measured af-ter 4 h against YAC-1 target cells at an E:T ratio of12.5:1. Results are representative of three independentexperiments. ��, p � 0.01; ns, not significant; No stim,no stimulation.

6082 ANTITUMOR NK RESPONSE BY RNA OLIGONUCLEOTIDES

Using a panel of neutralizing Abs, we showed that IFN-� pro-duction by NK cells was entirely IL-12 dependent (Fig. 5A). Incontrast, neutralizing Abs for IL-2, IL-6, and IL-15 had no effecton IFN-� production. To confirm the role of IL-12 in NK activa-tion, splenocytes from IL-12p40-deficient mice were stimulatedwith RNA oligonucleotides. Unexpectedly, activation of NK cellsby 9.2dr stimulation based on CD69 expression was not sup-pressed in IL-12-deficient splenocytes (Fig. 5B, left panel). IFN-�production was clearly dependent on IL-12, as IFN-� secretionwas strongly reduced in NK cells stimulated with DC from IL-12-deficient mice (Fig. 5B, middle panel). In contrast, cytotoxicity ofNK cells activated with the culture supernatant of RNA-stimulatedDC from IL-12-deficient mice was not impaired (Fig. 5B, rightpanel).

Since type I IFN has also been implicated in the activation ofNK cells (28, 29), we measured the activation of NK cells fromtype I IFN receptor-deficient mice upon RNA stimulation. BothCD69 up-regulation and cytotoxicity of NK cells were signifi-cantly reduced compared with wt NK cells (Fig. 5C). In contrast,IFN-� production in type I IFN receptor-deficient NK cells waspreserved. We conclude that the effector functions of NK cellsfollowing RNA oligonucleotide application are differently regu-lated. Although IFN-� production is dependent on IL-12, cytotox-icity depends on type I IFN. Thus, DC-derived IL-12 and type IIFN are both critical cytokines for the activation of NK cells byRNA oligonucleotides.

Immunostimulatory RNA activates human NK cells more rapidlythan small molecule agonists

In contrast to mice in which TLR7 is the sole receptor for ssRNAin the endosome, in humans TLR7 and TLR8 share the detectionof ssRNA (5, 6). Small molecule TLR7 and TLR8 agonists, theimidazoquinolines, have been shown to activate NK cells (30). Wecompared the effect of immunostimulatory RNA with three smallmolecule agonists on human NK cells. Peripheral blood monocytes

were used as accessory cells, because these cells express TLR8 andcan differentiate to monocyte-derived DC. ImmunostimulatoryRNA up-regulated CD69 expression on up to 60% of NK cells ina monocyte-dependent manner (Fig. 6A). Stimulation with smallmolecule TLR7/8 agonists also increased CD69 expression, albeitat lower levels. An increase in CD69 expression was also seenwhen NK cells were cocultured with plasmacytoid DC for bothimmunostimulatory RNA and imidazoquinolines (data not shown).Furthermore, stimulation of the monocyte-NK coculture with im-munostimulatory RNA induced IFN-� production by NK cells asearly as 12 h after stimulation (Fig. 6B). IFN-� induction by imi-dazoquinolines occurred later and did not reach the levels obtainedby immunostimulatory RNA (Fig. 6C). For both imidazoquino-lines and immunostimulatory RNA, stimulation was clearly de-pendent on the presence of monocytes.

DC mediate RNA activation of NK cells in vivo

To determine whether DC also mediate NK cell activation by RNAoligonucleotides in vivo, TLR7-deficient mice were reconstitutedwith wt BMDC before treatment with 9.2dr. CD69 expression onNK cells was up-regulated in these mice upon 9.2dr treatment (Fig.7, A and B). In contrast, we have shown that TLR7-deficient micewithout transferred DC show no NK cell activation following 9.2drinjection (Fig. 3A). NK cell activation was also seen when DCwere stimulated with 9.2dr before transfer, in addition to the invivo 9.2dr treatment (Fig. 7). Transfer of DC alone without RNAstimulation did not induce NK activation. Thus, DC are sufficientto mediate activation of NK cells by RNA oligonucleotidesin vivo.

DiscussionWe have recently described immune-activating sequences withinssRNA oligonucleotides that can act through TLR7 (12). Theseoligonucleotides induce Th1-type cytokines in vivo and trigger thegeneration of CTL and Ab production when coinjected with Ag

FIGURE 5. IL-12 and type I IFN mediate activationof NK cells by RNA-stimulated DC. A, BMDC werestimulated with 10 �g/ml 9.2dr or poly(A) complexedwith DOTAP for 6 h and splenocytes were activatedwith the supernatant in the presence of neutralizing Absagainst IL-2, IL-6, IL-12, and IL-15. IFN-� productionwas quantified by ELISA. B, Left panel, Splenocytesfrom wt or IL-12�/� mice were stimulated with 9.2drfor 18 h and expression of CD69 on NK cells was as-sessed. Middle panel, BMDC from wt or IL-12�/� micewere stimulated with 9.2dr for 6 h. Cells were washedto remove oligonucleotides and cocultured for 18 h withpurified NK cells from TLR7�/� mice. IFN-� produc-tion was quantified by ELISA. Right panel, wt splenicNK cells were activated for 4 h with supernatant fromstimulated wt or IL-12�/� BMDC and cytotoxicityagainst YAC-1 cells was measured at an E:T ratio of20:1. C, Left panel, Splenocytes from wt or IFNAR�/�

mice were stimulated with 9.2dr for 18 h and expressionof CD69 on NK cells was assessed. Middle panel, wtBMDC were stimulated with 9.2dr for 6 h. Cells werewashed and cocultured for 18 h with splenocytesfrom wt or IL-12�/� mice. IFN-� production wasquantified by ELISA. Right panel, Cytotoxicity of wtor IFNAR�/� NK cells activated with supernatantfrom stimulated wt BMDC was measured as in B.�, p � 0.05; ��, p � 0.01; ���, p � 0.001, ns, notsignificant, nd, not determined; No stim, no stimulation.

6083The Journal of Immunology

(13). In this study, we show for the first time that RNA oligo-nucleotides can inhibit tumor growth in a sequence-dependentmanner. Treatment of mice with TLR7-activating oligonucleo-tides selectively suppressed growth of NK-sensitive tumors, dem-onstrating that RNA oligonucleotides activate NK immunity (31,32). Two main mechanisms are responsible for NK-mediated an-titumor immunity, IFN-� production and direct cytotoxicity (16).IFN-� is a crucial mediator of antitumor immunity in experimentalmodels and elevated levels of IFN-� are associated with diseaseoutcome in several clinical studies (14, 33). The increase in IFN-�production by NK cells upon RNA oligonucleotide treatment sug-gests that IFN-� may play an important role in antitumor efficacy.Indeed, in a murine lung metastasis model, IFN-� was essential forthe NK-dependent antitumor effect of 3M-011, a TLR7 agonist(19). We further show that RNA oligonucleotides potentiate NKcytotoxic activity in vivo. The close contact of activated NK cellswith target cells is a prerequisite for cytotoxic activity (34) andindeed we clearly show the presence of activated NK cells withinthe s.c. tumors. Thus, both IFN-� production and direct cytotox-icity may synergize to provide NK-mediated antitumor immunityupon RNA oligonucleotide treatment.

In the absence of infection, NK cells present a naive phenotypeand do not efficiently up-regulate effector functions upon contactwith target cells expressing NK-activating receptors (35, 36). Toachieve efficient protection against infections and tumors, NK cells

therefore require an initial priming that can be effected by bacte-rial, viral, or even parasitic infectious agents (37–39). Primed NKcells are not directly activated to produce IFN-�, but they respondto triggering by target cells (37). In contrast to this two-step pro-cess, we show here that a single injection of RNA oligonucleotidesrapidly stimulates IFN-� production by NK cells in vivo withoutthe need for restimulation with target cells. Furthermore, an in-crease in cytotoxicity was seen in vivo as early as 4 h after injec-tion of the target cells. Thus, RNA oligonucleotides rapidly acti-vate NK cells in vivo while bypassing the need for additionalrestimulation. Activation was clearly sequence-specific, as apoly(A) sequence of the same length induced neither an activatedphenotype nor IFN-� production or cytotoxicity. We thus show forthe first time that RNA oligonucleotides can activate NK cells invivo in a sequence-dependent manner.

Because the oligonucleotide backbone itself, independently ofthe nucleotide sequence, may play a role in TLR7 and TLR9 rec-ognition and activation (40, 41), we compared in vivo NK activa-tion by oligonucleotides with the same sequence containing eitheran unmodified PD backbone or a PTO-modified backbone. Weobserved low activation with the PD backbone, while activationwas strongly enhanced by the use of the PTO backbone, althoughstimulation remained sequence-dependent. Similarly, we previ-ously observed stronger T and B cell activation with PTO RNAoligonucleotides (13). This probably relates to the increased sta-bility provided by the PTO backbone, since RNA is highly sus-ceptible to nuclease degradation (42).

RNA can be sensed by a variety of receptors of the innate im-mune system, including the TLRs 3 and 7, the cytoplasmic heli-cases RIG-I and MDA-5, the serine-threonine kinase PKR, and theNALP-3 inflammasome (3–8, 43–46). Importantly, we showedthat NK activation by the 9.2dr RNA oligonucleotide is entirelyTLR7-dependent, since the increase in NK effector function fol-lowing RNA treatment is absent in TLR7-deficient mice. Thisdemonstrates that cytosolic receptors do not participate in the invivo recognition of these RNA oligonucleotides. Thus, we showedfor the first time that RNA oligonucleotides activate NK cells in

FIGURE 6. Immunostimulatory RNA activates human NK cells morerapidly than small molecule agonists. In brief, 105 NK cells were cocul-tured with monocytes at the indicated ratio. Cells were stimulated with 1�g/ml of the immunostimulatory RNA oligonucleotide 9.3as or with 3M-001 (TLR7 agonist), 3M-002 (TLR8 agonist), or 3M-003 (TLR7 and 8agonist) (10 �M). A, After 36 h, expression of CD69 was analyzed on NKcells (gated as MHC class II-negative cells). B and C, Supernatant wasremoved after 12 h (B) and 36 h (C) of coculture and IFN-� production wasassessed. Data show the mean of three donors � SEM. n.d., Notdetermined.

FIGURE 7. DC mediate activation of NK cells by RNA oligonucleo-tides in vivo. A and B, Two hours before injection of 9.2dr, TLR7�/� micewere transferred with 5 � 107 wt BMDC that were either unstimulated orstimulated in vitro with 9.2dr for 6 h. Twenty-two hours after RNA injec-tion, expression of CD69 on splenic NK cells was measured. Data show theresults of (A) one representative mouse per group and (B) the mean � SEM(n � 3). No stim, No stimulation.

6084 ANTITUMOR NK RESPONSE BY RNA OLIGONUCLEOTIDES

vivo through TLR7. In vitro studies show that also human NK cellswithin PBMC can be stimulated by a TLR7-activating ssRNA se-quence derived from HIV-1 (47). This suggests that a therapeuticstrategy using RNA oligonucleotides to target NK cells may alsobe effective in humans.

Small molecule agonists of the imidazoquinoline family havebeen shown to activate NK cells through TLR7 and TLR8 (30).Previous in vitro studies on the mechanisms of NK cell activationby imidazoquinolines have however been inconsistent. Direct ac-tivation of human NK cells by imidazoquinolines without the needfor accessory cells has been shown in vitro but is dependent on thepresence of the cytokines IL-12, IL-15, or IFN-� that are generallyproduced by accessory cells (48, 49). Contrasting reports claimthat the induction of IFN-� production in vitro by imidazoquino-lines is dependent on activation of accessory cells such as mono-cytes (30, 50, 51). Furthermore, imidazoquinolines also activatethe NALP3 inflammasome, which leads to the production of bio-active IL-1� and IL-18 (52). Indeed, IL-18 was shown to contrib-ute to NK cell activation by imidazoquinoline compounds (51). Inthis study, we clearly show that reconstitution of TLR7-deficientmice with wt DC restores activation of NK cells by immunostimu-latory RNA. Thus, we provide in vivo data demonstrating that themain mechanism for in vivo activation of NK cells by immunos-timulatory RNA is entirely TLR7-dependent and is mediated byaccessory cells. Indeed, the induction of NK responses to patho-gens in vivo requires DC (37–39).

Although DC can activate NK cells by direct contact (29, 53),cell-cell contact is not necessary for activation of NK cells byRNA oligonucleotides. Instead, activation is mediated by DC-se-creted factors, as the stimulatory effect on NK cells can beachieved by transferring supernatant of activated DC. Several cy-tokines, the most important being IL-2, IL-10, IL-12, IL-15, IL-18,and IFN-�, have been implicated in the regulation of NK activa-tion (26, 54–57). Type I IFN in particular plays an important rolefor the in vivo priming of NK cells in different infectious models(37–39). Depending on the pathogen, type I IFN affects NK cellpriming by acting either on myeloid DC that are induced to pro-duce IL-15 (37) or directly on NK cells (39). The cellular sourceof type I IFN necessary to activate NK cells upon TLR stimulationremains however unclear (37). Interestingly, depletion of plasma-cytoid DC, the main producers of the type I IFN-�, does not impairactivation of NK cells in vivo upon infection (38). We have pre-viously shown that RNA oligonucleotides induce both IL-12 andIFN-� in vivo (13). We show here that the effector functions fol-lowing RNA oligonucleotide treatment are differently regulated:whereas IFN-� production is dependent on IL-12, cytotoxicity de-pends entirely on type I IFN.

The therapeutic potential of cancer treatment strategies targetingNK cells has gained momentum in recent years. NK cells are oneof the cellular mediators of Ab-dependent cell-mediated cytotox-icity, an important effector mechanism for the therapeutic use ofmAbs targeting tumor Ags. Other treatments targeting NK cellsinclude NK cell adjuvants such as cytokines, soluble TRAIL, c-kittyrosine kinase inhibitors, or even the transfer of preactivated NKcells (14). These therapeutic strategies could be profitably com-bined with RNA oligonucleotides to enhance NK cell activationin vivo.

In summary, RNA oligonucleotides possess an important ther-apeutic potential for the treatment of NK-sensitive tumors by trig-gering the activation of NK cells via the induction of cytokineproduction by DC. In addition to their direct antitumoral efficacy,NK cells favor the generation of CTL through IFN-� production(21, 58). This mechanism may contribute to the ability of RNAoligonucleotides to induce an Ag-specific CTL response (13). Be-

cause both MHC-negative and MHC-positive tumor cells can thusbe eliminated, the combined activation of CTL and NK cells byRNA-based therapies may help prevent tumor immune escape(59). Furthermore, RNA oligonucleotides can be designed to in-clude other antitumoral properties in the same molecule: RNA oli-gonucleotides containing a 5�-triphosphate lead to antitumoral ef-ficacy through activation of the RIG-I receptor (60). In addition,introduction of a specific inhibitory sequence (small interferingRNA) permits the knockdown of specific genes such as tumor-promoting genes (60). Thus, it will be possible to combineTLR7 activation with other potent antitumoral mechanisms inone RNA molecule, providing a promising mutifunctional ther-apeutic approach.

DisclosuresThe authors have no financial conflict of interest.

References1. Takeuchi, O., and S. Akira. 2007. Recognition of viruses by innate immunity.

Immunol. Rev. 220: 214–224.2. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto,

K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptorrecognizes bacterial DNA. Nature 408: 740–745.

3. 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.

4. Hornung, V., J. Ellegast, S. Kim, K. Brzozka, A. Jung, H. Kato, H. Poeck,S. Akira, K. K. Conzelmann, M. Schlee, et al. 2006. 5�-Triphosphate RNA is theligand for RIG-I. Science 314: 994–997.

5. Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira,G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of sin-gle-stranded RNA via Toll-like receptor 7 and 8. Science 303: 1526–1529.

6. Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innateantiviral responses by means of TLR7-mediated recognition of single-strandedRNA. Science 303: 1529–1531.

7. Andrejeva, J., K. S. Childs, D. F. Young, T. S. Carlos, N. Stock, S. Goodbourn,and R. E. Randall. 2004. The V proteins of paramyxoviruses bind the IFN-in-ducible RNA helicase, mda-5, and inhibit its activation of the IFN-� promoter.Proc. Natl. Acad. Sci. USA 101: 17264–17269.

8. Pichlmair, A., O. Schulz, C. P. Tan, T. I. Naslund, P. Liljestrom, F. Weber, andC. Reis e Sousa. 2006. RIG-I-mediated antiviral responses to single-strandedRNA bearing 5�-phosphates. Science 314: 997–1001.

9. Speiser, D. E., D. Lienard, N. Rufer, V. Rubio-Godoy, D. Rimoldi, F. Lejeune,A. M. Krieg, J. C. Cerottini, and P. Romero. 2005. Rapid and strong humanCD8� T cell responses to vaccination with peptide, IFA, and CpG oligode-oxynucleotide 7909. J. Clin. Invest. 115: 739–746.

10. Hornung, V., S. Rothenfusser, S. Britsch, A. Krug, B. Jahrsdorfer, T. Giese,S. Endres, and G. Hartmann. 2002. Quantitative expression of Toll-like receptor1–10 mRNA in cellular subsets of human peripheral blood mononuclear cells andsensitivity to CpG oligodeoxynucleotides. J. Immunol. 168: 4531–4537.

11. Krug, A., A. Towarowski, S. Britsch, S. Rothenfusser, V. Hornung, R. Bals,T. Giese, H. Engelmann, S. Endres, A. M. Krieg, and G. Hartmann. 2001. Toll-like receptor expression reveals CpG DNA as a unique microbial stimulus forplasmacytoid dendritic cells which synergizes with CD40 ligand to induce highamounts of IL-12. Eur. J. Immunol. 31: 3026–3037.

12. Hornung, V., M. Guenthner-Biller, C. Bourquin, A. Ablasser, M. Schlee,S. Uematsu, A. Noronha, M. Manoharan, S. Akira, A. de Fougerolles, S. Endres,and G. Hartmann. 2005. Sequence-specific potent induction of IFN-� by shortinterfering RNA in plasmacytoid dendritic cells through TLR7. Nat. Med. 11:263–270.

13. Bourquin, C., L. Schmidt, V. Hornung, C. Wurzenberger, D. Anz, N. Sandholzer,S. Schreiber, A. Voelkl, G. Hartmann, and S. Endres. 2007. ImmunostimulatoryRNA oligonucleotides trigger an antigen-specific cytotoxic T-cell and IgG2a re-sponse. Blood 109: 2953–2960.

14. Terme, M., E. Ullrich, N. F. Delahaye, N. Chaput, and L. Zitvogel. 2008. Naturalkiller cell-directed therapies: moving from unexpected results to successful strat-egies. Nat. Immunol. 9: 486–494.

15. Sobolev, O., P. Stern, A. Lacy-Hulbert, and R. O. Hynes. 2009. Natural killercells require selectins for suppression of subcutaneous tumors. Cancer Res. 69:2531–2539.

16. Vivier, E., E. Tomasello, M. Baratin, T. Walzer, and S. Ugolini. 2008. Functionsof natural killer cells. Nat. Immunol. 9: 503–510.

17. Hemmi, H., T. Kaisho, O. Takeuchi, S. Sato, H. Sanjo, K. Hoshino, T. Horiuchi,H. Tomizawa, K. Takeda, and S. Akira. 2002. Small anti-viral compounds acti-vate immune cells via the TLR7 MyD88-dependent signaling pathway. Nat. Im-munol. 3: 196–200.

18. Schon, M. P., and M. Schon. 2008. TLR7 and TLR8 as targets in cancer therapy.Oncogene 27: 190–199.

19. Dumitru, C. D., M. A. Antonysamy, K. S. Gorski, D. D. Johnson, L. G. Reddy,J. L. Lutterman, M. M. Piri, J. Proksch, S. M. McGurran, E. A. Egging, et al.2009. NK1.1� cells mediate the antitumor effects of a dual Toll-like receptor 7/8

6085The Journal of Immunology

agonist in the disseminated B16-F10 melanoma model. Cancer Immunol. Immu-nother. 58: 575–587.

20. Broomfield, S. A., R. G. van der Most, A. C. Prosser, S. Mahendran,M. G. Tovey, M. J. Smyth, B. W. Robinson, and A. J. Currie. 2009. Locallyadministered TLR7 agonists drive systemic antitumor immune responses that areenhanced by anti-CD40 immunotherapy. J. Immunol. 182: 5217–5224.

21. Adam, C., S. King, T. Allgeier, H. Braumuller, C. Luking, J. Mysliwietz,A. Kriegeskorte, D. H. Busch, M. Rocken, and R. Mocikat. 2005. DC-NK cellcross talk as a novel CD4� T-cell-independent pathway for antitumor CTL in-duction. Blood 106: 338–344.

22. Crozat, K., K. Hoebe, S. Ugolini, N. A. Hong, E. Janssen, S. Rutschmann,S. Mudd, S. Sovath, E. Vivier, and B. Beutler. 2007. Jinx, an MCMV suscepti-bility phenotype caused by disruption of Unc13d: a mouse model of type 3 fa-milial hemophagocytic lymphohistiocytosis. J. Exp. Med. 204: 853–863.

23. Neri, S., E. Mariani, A. Meneghetti, L. Cattini, and A. Facchini. 2001. Calcein-acetyoxymethyl cytotoxicity assay: standardization of a method allowing addi-tional analyses on recovered effector cells and supernatants. Clin. Diagn Lab.Immunol. 8: 1131–1135.

24. Glas, R., L. Franksson, C. Une, M. L. Eloranta, C. Ohlen, A. Orn, and K. Karre.2000. Recruitment and activation of natural killer (NK) cells in vivo determinedby the target cell phenotype: an adaptive component of NK cell-mediated re-sponses. J. Exp. Med. 191: 129–138.

25. Durdik, J. M., B. N. Beck, E. A. Clark, and C. S. Henney. 1980. Characterizationof a lymphoma cell variant selectively resistant to natural killer cells. J. Immunol.125: 683–688.

26. Granucci, F., I. Zanoni, N. Pavelka, S. L. Van Dommelen, C. E. Andoniou,F. Belardelli, M. A. Degli Esposti, and P. Ricciardi-Castagnoli. 2004. A contri-bution of mouse dendritic cell-derived IL-2 for NK cell activation. J. Exp. Med.200: 287–295.

27. Fernandez, N. C., A. Lozier, C. Flament, P. Ricciardi-Castagnoli, D. Bellet,M. Suter, M. Perricaudet, T. Tursz, E. Maraskovsky, and L. Zitvogel. 1999.Dendritic cells directly trigger NK cell functions: cross-talk relevant in innateanti-tumor immune responses in vivo. Nat. Med. 5: 405–411.

28. Andoniou, C. E., S. L. van Dommelen, V. Voigt, D. M. Andrews, G. Brizard,C. Asselin-Paturel, T. Delale, K. J. Stacey, G. Trinchieri, and M. A. Degli-Esposti.2005. Interaction between conventional dendritic cells and natural killer cells is in-tegral to the activation of effective antiviral immunity. Nat. Immunol. 6: 1011–1019.

29. Akazawa, T., T. Ebihara, M. Okuno, Y. Okuda, M. Shingai, K. Tsujimura,T. Takahashi, M. Ikawa, M. Okabe, N. Inoue, et al. 2007. Antitumor NK acti-vation induced by the Toll-like receptor 3-TICAM-1 (TRIF) pathway in myeloiddendritic cells. Proc. Natl. Acad. Sci. USA 104: 252–257.

30. Hart, O. M., V. Athie-Morales, G. M. O’Connor, and C. M. Gardiner. 2005.TLR7/8-mediated activation of human NK cells results in accessory cell-depen-dent IFN-� production. J. Immunol. 175: 1636–1642.

31. Borg, C., M. Terme, J. Taieb, C. Menard, C. Flament, C. Robert, K. Maruyama,H. Wakasugi, E. Angevin, K. Thielemans, et al. 2004. Novel mode of action ofc-kit tyrosine kinase inhibitors leading to NK cell-dependent antitumor effects.J. Clin. Invest. 114: 379–388.

32. Kelly, J. M., P. K. Darcy, J. L. Markby, D. I. Godfrey, K. Takeda, H. Yagita, andM. J. Smyth. 2002. Induction of tumor-specific T cell memory by NK cell-me-diated tumor rejection. Nat. Immunol. 3: 83–90.

33. Dunn, G. P., A. T. Bruce, K. C. Sheehan, V. Shankaran, R. Uppaluri, J. D. Bui,M. S. Diamond, C. M. Koebel, C. Arthur, J. M. White, and R. D. Schreiber. 2005.A critical function for type I interferons in cancer immunoediting. Nat. Immunol.6: 722–729.

34. Lieberman, J. 2003. The ABCs of granule-mediated cytotoxicity: new weapons inthe arsenal. Nat. Rev. Immunol. 3: 361–370.

35. Bryceson, Y. T., M. E. March, H. G. Ljunggren, and E. O. Long. 2006. Synergyamong receptors on resting NK cells for the activation of natural cytotoxicity andcytokine secretion. Blood 107: 159–166.

36. Fehniger, T. A., S. F. Cai, X. Cao, A. J. Bredemeyer, R. M. Presti, A. R. French,and T. J. Ley. 2007. Acquisition of murine NK cell cytotoxicity requires thetranslation of a pre-existing pool of granzyme B and perforin mRNAs. Immunity26: 798–811.

37. Lucas, M., W. Schachterle, K. Oberle, P. Aichele, and A. Diefenbach. 2007.Dendritic cells prime natural killer cells by trans-presenting interleukin 15. Im-munity 26: 503–517.

38. Schleicher, U., J. Liese, I. Knippertz, C. Kurzmann, A. Hesse, A. Heit,J. A. Fischer, S. Weiss, U. Kalinke, S. Kunz, and C. Bogdan. 2007. NK cellactivation in visceral leishmaniasis requires TLR9, myeloid DCs, and IL-12, butis independent of plasmacytoid DCs. J. Exp. Med. 204: 893–906.

39. Martinez, J., X. Huang, and Y. Yang. 2008. Direct action of type I IFN on NKcells is required for their activation in response to vaccinia viral infection in vivo.J. Immunol. 180: 1592–1597.

40. Haas, T., J. Metzger, F. Schmitz, A. Heit, T. Muller, E. Latz, and H. Wagner.2008. The DNA sugar backbone 2�-deoxyribose determines toll-like receptor 9activation. Immunity 28: 315–323.

41. Yasuda, K., M. Rutz, B. Schlatter, J. Metzger, P. B. Luppa, F. Schmitz, T. Haas,A. Heit, S. Bauer, and H. Wagner. 2006. CpG motif-independent activation ofTLR9 upon endosomal translocation of “natural” phosphodiester DNA. Eur.J. Immunol. 36: 431–436.

42. Lan, T., E. R. Kandimalla, D. Yu, L. Bhagat, Y. Li, D. Wang, F. Zhu, J. X. Tang,M. R. Putta, Y. Cong, et al. 2007. Stabilized immune modulatory RNA com-pounds as agonists of Toll-like receptors 7 and 8. Proc. Natl. Acad. Sci. USA 104:13750–13755.

43. Kato, H., O. Takeuchi, E. Mikamo-Satoh, R. Hirai, T. Kawai, K. Matsushita,A. Hiiragi, T. S. Dermody, T. Fujita, and S. Akira. 2008. Length-dependentrecognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-Iand melanoma differentiation-associated gene 5. J. Exp. Med. 205: 1601–1610.

44. Kato, H., O. Takeuchi, S. Sato, M. Yoneyama, M. Yamamoto, K. Matsui,S. Uematsu, A. Jung, T. Kawai, K. J. Ishii, et al. 2006. Differential roles of MDA5and RIG-I helicases in the recognition of RNA viruses. Nature 441: 101–105.

45. Katze, M. G., M. Wambach, M. L. Wong, M. Garfinkel, E. Meurs, K. Chong,B. R. Williams, A. G. Hovanessian, and G. N. Barber. 1991. Functional expres-sion and RNA binding analysis of the interferon-induced, double-stranded RNA-activated, 68,000-Mr protein kinase in a cell-free system. Mol. Cell Biol. 11:5497–5505.

46. Kanneganti, T. D., M. Body-Malapel, A. Amer, J. H. Park, J. Whitfield,L. Franchi, Z. F. Taraporewala, D. Miller, J. T. Patton, N. Inohara, and G. Nunez.2006. Critical role for Cryopyrin/Nalp3 in activation of caspase-1 in response toviral infection and double-stranded RNA. J. Biol. Chem. 281: 36560–36568.

47. Alter, G., T. J. Suscovich, N. Teigen, A. Meier, H. Streeck, C. Brander, andM. Altfeld. 2007. Single-stranded RNA derived from HIV-1 serves as a potentactivator of NK cells. J. Immunol. 178: 7658–7666.

48. Sawaki, J., H. Tsutsui, N. Hayashi, K. Yasuda, S. Akira, T. Tanizawa, andK. Nakanishi. 2007. Type 1 cytokine/chemokine production by mouse NK cellsfollowing activation of their TLR/MyD88-mediated pathways. Int. Immunol. 19:311–320.

49. Girart, M. V., M. B. Fuertes, C. I. Domaica, L. E. Rossi, and N. W. Zwirner.2007. Engagement of TLR3, TLR7, and NKG2D regulate IFN-� secretion but notNKG2D-mediated cytotoxicity by human NK cells stimulated with suboptimaldoses of IL-12. J. Immunol. 179: 3472–3479.

50. Berger, M., A. Ablasser, S. Kim, I. Bekeredjian-Ding, T. Giese, S. Endres,V. Hornung, and G. Hartmann. 2009. TLR8-driven IL-12-dependent reciprocaland synergistic activation of NK cells and monocytes by immunostimulatoryRNA. J. Immunother. 32: 262–271.

51. Gorski, K. S., E. L. Waller, J. Bjornton-Severson, J. A. Hanten, C. L. Riter,W. C. Kieper, K. B. Gorden, J. S. Miller, J. P. Vasilakos, M. A. Tomai, andS. S. Alkan. 2006. Distinct indirect pathways govern human NK-cell activationby TLR-7 and TLR-8 agonists. Int. Immunol. 18: 1115–1126.

52. Kanneganti, T. D., N. Ozoren, M. Body-Malapel, A. Amer, J. H. Park, L. Franchi,J. Whitfield, W. Barchet, M. Colonna, P. Vandenabeele, et al. 2006. BacterialRNA and small antiviral compounds activate caspase-1 through cryopyrin/Nalp3.Nature 440: 233–236.

53. Kijima, M., T. Yamaguchi, C. Ishifune, Y. Maekawa, A. Koyanagi, H. Yagita,S. Chiba, K. Kishihara, M. Shimada, and K. Yasutomo. 2008. Dendritic cell-mediated NK cell activation is controlled by Jagged2-Notch interaction. Proc.Natl. Acad. Sci. USA 105: 7010–7015.

54. Qian, C., X. Jiang, H. An, Y. Yu, Z. Guo, S. Liu, H. Xu, and X. Cao. 2006. TLRagonists promote ERK-mediated preferential IL-10 production of regulatory den-dritic cells (diffDCs), leading to NK-cell activation. Blood 108: 2307–2315.

55. Gerosa, F., A. Gobbi, P. Zorzi, S. Burg, F. Briere, G. Carra, and G. Trinchieri.2005. The reciprocal interaction of NK cells with plasmacytoid or myeloid den-dritic cells profoundly affects innate resistance functions. J. Immunol. 174:727–734.

56. Dalod, M., T. Hamilton, R. Salomon, T. P. Salazar-Mather, S. C. Henry,J. D. Hamilton, and C. A. Biron. 2003. Dendritic cell responses to early murinecytomegalovirus infection: subset functional specialization and differential reg-ulation by interferon �/�. J. Exp. Med. 197: 885–898.

57. Koka, R., P. Burkett, M. Chien, S. Chai, D. L. Boone, and A. Ma. 2004. Cuttingedge: murine dendritic cells require IL-15R� to prime NK cells. J. Immunol. 173:3594–3598.

58. Zhou, H., Y. Luo, C. D. Kaplan, J. A. Kruger, S. H. Lee, R. Xiang, andR. A. Reisfeld. 2006. A DNA-based cancer vaccine enhances lymphocyte crosstalk by engaging the NKG2D receptor. Blood 107: 3251–3257.

59. Diefenbach, A., E. R. Jensen, A. M. Jamieson, and D. H. Raulet. 2001. Rae1 andH60 ligands of the NKG2D receptor stimulate tumour immunity. Nature 413:165–171.

60. Poeck, H., R. Besch, C. Maihoefer, M. Renn, D. Tormo, S. S. Morskaya,S. Kirschnek, E. Gaffal, J. Landsberg, J. Hellmuth, et al. 2008. 5�-Triphosphate-siRNA: turning gene silencing and Rig-I activation against melanoma. Nat. Med.14: 1256–1263.

6086 ANTITUMOR NK RESPONSE BY RNA OLIGONUCLEOTIDES