GITR Pathway Activation Abrogates Tumor Immune Suppression through … · Research Article GITR Pathway Activation Abrogates Tumor Immune Suppression through Loss of Regulatory T-cell

Research Article

GITR Pathway Activation Abrogates Tumor ImmuneSuppression through Loss of Regulatory T-cell LineageStability

David A. Schaer1, Sadna Budhu1, Cailian Liu1, Campbell Bryson2, Nicole Malandro1,2, Adam Cohen4,Hong Zhong1, Xia Yang1, Alan N. Houghton1, Taha Merghoub1, and Jedd D. Wolchok1,2,3

AbstractLigation of GITR (glucocorticoid-induced TNF receptor-related gene, or TNFRSF18) by agonist antibody

has recently entered into early-phase clinical trials for the treatment of advanced malignancies. Although theability of GITR modulation to induce tumor regression is well documented in preclinical studies, theunderlying mechanisms of action, particularly its effects on CD4þFoxp3þ regulatory T cells (Treg), have notbeen fully elucidated. We have previously shown that GITR ligation in vivo by agonist antibody DTA-1 causesmore than 50% reduction of intratumor Tregs with downmodulation of Foxp3 expression. Here, we show thatthe loss of Foxp3 is tumor dependent. Adoptively transferred Foxp3þ Tregs from tumor-bearing animals loseFoxp3 expression in the host when treated with DTA-1, whereas Tregs from na€�ve mice maintain Foxp3expression. GITR ligation also alters the expression of various transcription factors and cytokines importantfor Treg function. Complete Foxp3 loss in intratumor Tregs correlates with a dramatic decrease in Heliosexpression and is associated with the upregulation of transcription factors, T-Bet and Eomes. Changes inHelios correspond with a reduction in interleukin (IL)-10 and an increase in IFN-g expression in DTA-1–treated Tregs. Together, these data show that GITR agonist antibody alters Treg lineage stability inducing aninflammatory effector T-cell phenotype. The resultant loss of lineage stability causes Tregs to lose theirintratumor immune-suppressive function, making the tumor susceptible to killing by tumor-specific effectorCD8þ T cells. Cancer Immunol Res; 1(5); 320–31. �2013 AACR.

IntroductionThe immune system is capable of recognizing malignant

cells, but inmost situations, tumors develop strategies to avoidelimination and escape immune surveillance (1). Recentadvances in immunotherapy have succeeded in shifting thebalance from tumor immune escape to tumor elimination.Instead of treating the tumor directly by inhibiting cellgrowth, immunotherapeutic approaches modulate a patient'simmune system to induce tumor regression. The success ofthis approach is highlighted by the U.S. Food and DrugAdministration approval of the CTLA-4–blocking antibody,

ipilimumab, the first therapy to show enhanced overall surv-ival for patients with melanoma (2). Serving as a proof-of-principle, CTLA-4 blockade has led to targeting of otherimmune checkpoints (PD-1/PD-L1) alone or in combinationwith CTLA-4, with very promising results in early-phase clin-ical trials (3–6). Although coinhibitory receptor blockadehas shown durable clinical efficacy, a significant number ofpatients (�50%–80%) remain refractory to these treatmentsand some tumor types do not respond as robustly as others(3, 7). To further potentiate antitumor immune responses andextend clinical benefit, activating costimulatory molecules,such as TNF receptor (TNFR) superfamily members GITR(glucocorticoid-induced TNF receptor-related gene), OX40,and 4-1BB, represent a logical next step (8, 9).

GITR became an attractive target for cancer immunother-apy after the agonistic anti-GITR antibodyDTA-1was shown toblock the suppressive effects of regulatory T cells (Treg; ref. 10).Subsequently, DTA-1 was shown to enhance tumor immunityin a concomitant immunitymodel ofmelanoma. In addition topreventing growth of secondary tumor challenges, DTA-1treatment also caused the regression of some of the primarytumor challenge (11). This observation has been extended intomultiple tumor models and various combinatorial strategieswith vaccines, adoptive T-cell transfer, and concurrent CTLA-4blockade (12). With preclinical success of GITR tumor immu-notherapy, it has been entered into early-phase clinical trials

Authors' Affiliations: 1Swim Across America – Ludwig CollaborativeLaboratory, Immunology Program, Sloan-Kettering Institute for CancerResearch; 2Weill Cornell Medical College; 3Ludwig Center for CancerImmunotherapy at Memorial Sloan-Kettering Cancer Center, New York,New York; and 4Perelman Center for Advanced Medicine, University ofPennsylvania, Philadelphia, Pennsylvania

Note: Supplementary data for this article are available at Cancer Immu-nology Research Online (http://cancerimmunolres.aacrjournals.org/).

T. Merghoub and J.D. Wolchok have co-senior authorship in this article.

Corresponding Author: Jedd D. Wolchok, Memorial Sloan-KetteringCancer Center, 1275 York Avenue, New York, NY 10065. Phone: 646-888-2315; Fax: 646-422-0453; E-mail: [email protected]

doi: 10.1158/2326-6066.CIR-13-0086

�2013 American Association for Cancer Research.

CancerImmunology

Research

Cancer Immunol Res; 1(5) November 2013320

on June 12, 2021. © 2013 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst September 16, 2013; DOI: 10.1158/2326-6066.CIR-13-0086

http://cancerimmunolres.aacrjournals.org/

for the treatment of advanced malignancies. Despite its ther-apeutic potential, the mechanism of action on Tregs asopposed to effector T cells (Teff) has not been fully elucidated.Understanding its activity on Tregs is a necessary step toinform the effective use of GITR therapy in humans.Whether or not GITR immunotherapy targets GITR solely

on Teffs, or on both Teffs and Tregs, has been an area ofinvestigation. Because GITR is constitutively expressed athigh levels on Tregs, it was assumed that DTA-1 directlyinhibited Treg-suppressive function in vitro (10). However,GITR is also upregulated on CD4 and CD8 Teffs followingactivation and acted as costimulatory receptor (13). Throughthe use of GITR�/� Tregs, it was determined that the costi-mulatory role of GITR enabled Teffs to resist Treg suppres-sion while having no direct effect on Tregs (14). Thus, initialreports of enhanced tumor immunity resulting from GITRligation by agonist antibody DTA-1 were attributed to themodulation of Teffs (15, 16). Nevertheless, we and others haverecently shown that direct modulation of Tregs is an import-ant consequence of DTA-1 therapy (17, 18). DTA-1 treatmentcauses more than 50% reduction of intratumor Tregs anddown modulation of Foxp3. In addition, the effects of DTA-1are attenuated if either Teffs or Tregs is GITR�/� (17). Ourdata suggest that the efficacy of DTA-1 comes not only fromits effect on Teffs, but also from its modulation of Tregs.Here, we show that GITR ligation by DTA-1 induces

intratumor Treg lineage instability. DTA-1 causes loss ofFoxp3 in a tumor-dependent manner and is preceded by theloss of the transcription factor Helios. This results in theacquisition of a Th1 effector-like profile and prevents Treg-mediated intratumor suppression of the antitumor immuneresponse. Our results show that modulation of Tregs, alongwith Teffs, is important and necessary for the efficacy ofGITR immunotherapy.

Materials and MethodsMiceC57BL/6: CD45.1, Thy1.2þ, Thy1.1þ, and OT-1 TCR trans-

genic mice were obtained from Jackson Laboratory. Pmel-1T-cell receptor transgenic mice were a gift from Dr. NicohlasRestifo [National Cancer Institute (NCI), Bethesda, MD].Foxp3-GFP knockin mice were a gift from Dr. A. Rudensky[Memorial Sloan-Kettering Cancer Center (MSKCC), NewYork, NY]. GITR�/� and GITRþ/þ littermates (Sv129 �C57BL/6 background) were a gift from Dr. P.P. Pandolfi(MSKCC) and were backcrossed more than 10 generationsand onto Pmel-1 Thy1.1þ C57BL/6 background using aspeed congenic system. Mice were maintained accordingto NIH Animal Care guidelines, under a protocol (# 96-04-017) approved by the MSKCC Institutional Animal Care andUse Committee.

Cell lines, tumor challenge, and DTA-1 therapyB16F10/LM3 (hereafter called B16) is derived from the

B16F10 line provided by I. Fidler (MD Anderson CancerCenter, Houston, TX), and transfected with OVA to generateB16-OVA (19). Tumor cells were cultured in RPMI-1640 medi-

um containing 7.5% FBS (for up to 2weeks after thawing). Eachmouse received 150,000 cells in 150 mL of growth factor–reduced Matrigel (BD Biosciences) injected subcutaneously.Four days after tumor challenge, mice were injected intraper-itoneally with either 1 mg of affinity-purified DTA-1 or PurifiedRat immunoglobulin G (IgG; Sigma-Aldrich) in 500 mL PBS.

Lymphocyte isolationSpleens, tumor-draining lymph nodes (TDLN), and tumors

were excised on days indicated in the text. Tumors wereweighed, and then tissue was homogenized through 40-mmstrainers to produce single-cell suspensions. Red blood cellswere lysed from spleens using an ACK lysis buffer (Lonza).Cells were washed with media, and tissue cell counts werecalculated using Guava cell counter (Millipore). Cells werethen either sorted for Tregs, stained immediately by fluores-cence-activated cell sorting (FACS) or for cytokine recall,stimulated with phorbol 12-myristate 13-acetate (PMA) andionomycin for 4 hours, and then treated with monensin beforeFACS staining.

Antibodies and FACS analysisAnti-GITR (DTA-1, S. Sakaguchi, Osaka University, Osaka,

Japan) and anti-OX40 (OX86, A. Weinberg, Earle ChilesResearch Institute, Portland, OR) were produced by theMSKCC Monoclonal Antibody Core Facility, and anti-4-1BB(LOB12.3) was procured from Bioxcell. Foxp3 Staining Kit(eBioscience) was used for intracellular staining. Antibodiesto antigens listed in figures were from BD Biosciences exceptFoxp3 (eBioscience), Helios, CD45.2 (Biolegend), and Nrp1(R&D systems). Dead cell exclusion was done using the AquaLIVE/DEAD Fixable Dead Cell Stain Kit (Invitrogen). Sampleswere acquired on 12-color LSRII cytometer, and analyzed usingFlowJo (Tree Star).

Treg adoptive transfersTumor-experienced or na€�ve Foxp3-GFP Tregs were isolated

from spleens and TDLNs of untreated Foxp3-GFPmice bearingB16 tumors 7 to 8 days after tumor challenge, or non–tumor-bearing. CD4þ GFPþ Tregs were isolated by enriching CD4þ

cells by CD4-positive or -negative MACS microbead separa-tion kits (Miltenyi) before sorting for GFP expression on aCytomation MoFlo or BD FACS Aria cell sorter in the MSKCCFlow Cytometry Core Facility. For cotransfer experiments(Fig. 1C), na€�ve Tregs were isolated from Thy 1.1þ C57BL/6mice using MACS microbead Treg isolation kit (Miltenyi).Tregs were then injected intravenously (5–7 � 105cell/mousefor each Treg type being transferred) in 200 mL of sterile PBS.

Collagen-fibrin gel-killing assayThe collagen-fibrin gel-killing assay is described in depth by

Budhu and colleagues (20) and was adapted for ex vivo tumors.Briefly, B16-Ova tumors isolated on day 10 or 11 after tumorchallenge were cut into small pieces, incubated for 5 minutesin 250 mg/mL collagenase in PBS containing Ca2þMg2þ, andhomogenized through 100-mm cell strainers to create single-cell suspensions. Viable tumor cells and tumor-infiltratinglymphocytes were counted by Trypan blue exclusion. A total of

GITR Induced Treg Lineage Instability

www.aacrjournals.org Cancer Immunol Res; 1(5) November 2013 321




A

C

Naïve

D

E

B d

donors

donor

donor

Naïve host

recipients

Figure 1. Foxp3 loss induced by DTA-1 is enhanced by tumor growth and is increased in the tumor microenvironment. A, fresh frozen sections of B16 tumorsfrom control IgG (IgG)- or DTA-1–treated Foxp3-GFP mice at day 10 of tumor growth, labeled for Foxp3 and DAPI described by Cohen and colleagues (17).Scale bar, 25 mm. Lack of Foxp3 staining and non-nuclear GFP label is seen in DTA-1–treated sections. B, representative FACS plots showCD4þ transferredTregs in spleen of recipients after gating on live, CD45þ CD3þ, MHC-IINEG, CD11bNEG cells. (Legend continued on the following page.)

Schaer et al.

Cancer Immunol Res; 1(5) November 2013 Cancer Immunology Research322




104 viable tumor cells, together with all infiltrating cells,were coembedded into collagen-fibrin gels with or without1 to 5 � 105 CD8þ T cells activated in vitro by cognatepeptide þ interleukin (IL)-2. Duplicate gels were lysed every24 hours for 3 days, and viable remaining tumor cells werediluted and plated in 6-well plates for colony formation. Sevendays later, plates were fixed with 3.7% formaldehyde andstained with 2% methylene blue before counting as describedby Budhu and colleagues (20).

Rate of tumor cell killingKilling constant k is calculated as described in ref. (20).

Briefly, k is calculated according to the following equation:bt ¼ b0e�kptþgt, where bt ¼ the concentration of B16 cells attime t; b0 ¼ the concentration of B16 cells at time 0; k ¼ thekilling rate constant (or killing efficiency) for CD8 T cells;p¼ the concentration of in vitro activated CD8 T cells; g¼ thegrowth rate constant for B16 cells.

Quantitative PCRIndividual tumors and pooled control spleenswere collected

and stained with anti-CD45, anti-CD4, and 40, 6-diamidino-2-phenylindole (DAPI) before CD45þ CD4þ GFPþ Treg or GFP�

Teff control, and were FACS sorted on BD FACS Aria directlyinto Trizol reagent (Invitrogen). Total RNA was prepared andreversed transcribed into cDNA using a High Capacity cDNAReverse Transcription Kit (Applied Biosystems). The primer-probe sets were from TaqMan Gene Expression Assays(Applied Biosystems). Quantitative real-time PCR reactionswere prepared with FasStart Universal Probemaster (Rox) mix(Roche) according to the manufacturer's instructions anddone using the ABI 7500 Real-Time PCR system (AppliedBiosystems). Each gene was amplified in duplicate and repeat-ed in two separate experiments. cDNA concentration differ-ences were normalized to glyceraldehyde—3—phosphatedehydrogenase (GAPDH). Relative gene expression of thetarget genes was calculated by the formula 2 – DCt [DCt ¼Ct (target gene) – Ct (GAPDH)].

ResultsTumor growth sensitizes Tregs to DTA-1–induced Foxp3lossWe previously showed that optimal GITR agonist antibody

DTA-1 treatment of early established (day 4) B16 tumorscaused intratumor Tregs to lose Foxp3 expression. By treatingtumors grown in mice where Tregs express GFP fused inframe to Foxp3 (Foxp3-GFP), we were able to detect remnantGFP in former Tregs, after Foxp3 had been degraded (Fig. 1A).GFP remains, whereas Foxp3 is degraded, because it is lesssusceptible to proteolytic degradation (Fig. 1A; ref. 17). DTA-1–induced Foxp3 loss is not seen in peripheral tissues, sug-

gesting that entry into the tumor microenvironment pro-motes Treg instability and increases susceptibility to mod-ulation (17). In addition, adoptively transferred Tregs sortedfrom spleens and TDLNs of tumor-bearing Foxp3-GFP micelose Foxp3 expression within 48 hours of infiltrating tumorsin DTA-1–treated hosts (17). To determine what rendersTregs susceptible to GITR modulation, we used the adoptivetransfer system to track the highly purified previouslyuntreated Tregs and probe the specific conditions permittingDTA-1–induced Foxp3 loss.

Tregs have been described to contain a minor populationthat is less stable and characterized by low CD25 expression.This population is susceptible to Foxp3 loss after long-termtransfer (4 weeks) into Rag�/� hosts (21). It has recentlybeen shown that stimulation with Fc-GITR-L can augmentFoxp3 loss after CD4þ T-cell transfer into a RAG�/� model ofinflammatory bowel disease (22). Therefore, we first askedwhether during our short-term (48 hours) adoptive transferconditions, DTA-1 exclusively modulates only the minorCD25 low population (which at most accounts for only�10% of Tregs). Consistent with published reports, there isa slight loss in the percentage of Foxp3þ Tregs in control IgG-treated mice 2 days after transfer (�90% pretransfer to�77%after transfer, Fig. 1B). In contrast, DTA-1 treatment induceda pronounced reduction in Foxp3þ Tregs (�90% Foxp3þpretransfer to �10% Foxp3þ after transfer, Fig. 1B). Thistranslates to an average of a 6 (� 0.58 SEM)-fold decrease inthe percentage of Foxp3þ transferred Tregs in DTA-1–trea-ted hosts compared with controls. These data confirm thatDTA-1 has the potential to modulate all Tregs and is notrestricted to the minor, unstable CD25low Treg fraction.

Having established the effect of DTA-1 on Treg Foxp3 lossin lymphopenic conditions, we next assessed how the pres-ence of the tumor and/or tumor infiltration affects Tregvulnerability to DTA-1. To accomplish this, we cotransferredcongenically marked (with CD45 and Thy1) Tregs isolatedfrom spleens and TDLNs of tumor-bearing (CD45.2þ) orna€�ve (Thy1.1þCD45.2þ) donors into lymphoreplete na€�veor tumor-bearing recipients (CD45.1þ) following thescheme in Fig. 1C. Na€�ve Thy 1.1 donor Tregs transferredinto na€�ve recipients displayed negligible loss of Foxp3 inperipheral tissues 48 hours after transfer [88% Foxp3þ,pretransfer (Fig. 1C) vs. �86%–91% Foxp3þ, in IgG (Fig.1D)]. DTA-1 treatment induced a maximum of 17% to 18%reduction in na€�ve Foxp3þ Tregs under these conditions(Fig. 1D, spleen IgG vs. DTA-1). In contrast, tumor-experi-enced CD45.2þ Tregs transferred into control IgG-treatedhosts displayed a greater loss of Foxp3 expression [94%Foxp3þ pretransfer (Fig. 1C) vs. �72% and �48% Foxp3þTregs in the spleen and lymph nodes, respectively (Fig. 1D)].DTA-1 treatment enhanced Foxp3 loss in tumor-experienced

(Continued.) C, Tregs were isolated from 25 naïve Thy1.1 CD45.2 donors and 25 tumor-bearing Foxp3-GFP CD45.2 donors, mixed 1:1, and transferred intonaïve (D), or tumor-bearing (E) CD45.1 recipients treated with IgG or DTA-1. Transferred Tregs were identified by gating on live, CD45þ, CD3þ, CD4þ

cells, and then onCD45.2þ (tumor-experienced Tregs), or CD45.2þ Thy1.1þ (naïve Tregs; D and E, left). Representative examples of Foxp3 andCD25 stainingare shown in the spleen (D and E, middle). Graphs show mean � SEM for percent of Foxp3þ donor Tregs recovered in each tissue from a representativeexperiment (D and E, left). Experiments were repeated three times with 4 to 5 per group. �, P < 0.01; ��, P < 0.001; ��, P < 0.0001. LN, lymph node.






Tregs, which was most evident in the lymph node, wherethere was approximately 50% reduction in Foxp3þ Tregs inDTA-1–treated animals, as compared with animals treatedwith control IgG (Fig. 1D). In tumor-bearing recipients, DTA-1 treatment further potentiated the decrease in Foxp3 expres-sion in tumor-experienced Tregs in the spleen {34% vs. 23%,IgG vs. DTA-1, P ¼ 0.025 comparing %foxp3NEG [(Post Foxp3purity-Post Transfer Foxp3%)/Post Foxp3 purity] in Fig. 1D andE}. Na€�ve Tregs only displayed a significant loss of Foxp3expression upon entering the tumor [(Fig. 1E) 27% drop IgGvs. DTA-1 compared with pretransfer]. Taken together, ourdata strongly suggest that preconditioning of Tregs in the pre-sence of tumor or in the tumormicroenvironment before DTA-1treatment is important to their susceptibility to Foxp3 loss.

Transferred Tregs do not display cleaved caspase-3, andequal numbers of transferred Tregs are recovered after DTA-1 treatment for both tumor-experienced and na€�ve Tregs(Supplementary Fig. S1A and S1B). Foxp3 loss in tumor-experienced Tregs sorted by high CD25 expression seemedto be comparable with those sorted by Foxp3-GFP (Supple-mentary Fig. S1C). Moreover, to address the possibility thatFoxp3NEG Tregs result from DTA-1–induced proliferation ofcontaminating Teffs, we monitored the proliferation oftransferred Teff (CD4þ GFPNEG) sorted from Foxp3-GFPtumor-bearing mice. Supplementary Figure 1D shows thatCD4þ GFPNEG Teffs (sorted from Foxp3-GFP tumor-bearingmice) do not proliferate or accumulate after transfer andDTA-1 treatment. In sum, these data indicate that Foxp3NEG

Tregs come directly from the Foxp3þ Tregs, the frequency ofwhich is not reduced as a consequence of cell death, deple-tion, or proliferation of contaminating Teffs.

In addition, we found that the DTA-1–induced Foxp3 lossoccurs in a dose-dependent manner (Supplementary Fig. S1E).Interestingly, agonist antibodies to GITR-related TNFR familymembers, 4-1BB and OX40, did not affect the frequency ofFoxp3þTregs (Supplementary Fig. S1F). Thus, this effect seemsto be uniquely associated with GITR stimulation.

Foxp3 loss correlates with a loss of Helios expressionThe data above suggest that lymphopenic conditions and

the presence of tumor sensitize Tregs to the effects of DTA-1.In addition, the data imply that DTA-1 has the ability tomodulate a large percentage of the Treg population, whichremains viable after loss of Foxp3. Therefore, we hypothesizedthat in the tumor therapy setting, even the intratumor Tregsthat maintain Foxp3 expression after DTA-1 treatment wouldbe affected by GITR stimulation. In fact, we have previouslyshown that Foxp3 expression in the remaining Tregs is signif-icantly lower in DTA-1– versus control IgG-treated tumors,supporting this concept (17). To better understand the out-come of DTA-1–induced Treg instability, we investigatedwhether there were changes in other markers associated withTreg stability, function, and/or ontogeny such as the transcrip-tion factor Helios, and expression of the cell surface VEGFcoreceptor neuropilin 1 (Nrp1). Expression of Nrp1 has beenreported to distinguish between thymus-derived (tTreg) andperipherally derived Tregs (pTreg) and is important for Tregtrafficking to B16 tumors (23–25). Although the exact role of

the Ikaros family transcription factor Helios remains unre-solved, it has been described as amarker of Treg activation andidentifies the most suppressive population of tumor-infiltrat-ing Tregs (26, 27). In control animals, intratumor Tregs areuniformly HeliosHIGH with a majority being Nrp1HIGH, at thepeak of B16 immune infiltration compared with peripheralTregs (spleen, 10–11 days after tumor challenge; ref. 28), sug-gesting a highly activated tTreg phenotype (Fig. 2A; refs. 25, 27).In contrast, DTA-1 treatment causes a clear loss of Heliosexpression in the remaining Foxp3þ intratumorTregs (Fig. 2A).Nrp1 expression did not seem to be as significantly affected asHelios, which may be related to its role in Treg trafficking (24).

Using changes in Helios expression as a surrogate markerto identify DTA-1–modulated Tregs in addition to Foxp3 loss,we expanded our analysis to early phases of Treg tumor in-filtration to determine the kinetics of Helios loss and its possiblecorrelation with Treg survival and function. At day 7 of tumorgrowth (3 days after DTA-1 treatment), there is already asignificant increase in the HeliosLOW Treg cell population(�18% compared with �55% in IgG vs. DTA-1 treatment,respectively, Fig. 2B, left). By day 10, there is an approximately55% to 60% loss of Foxp3þ Tregs (Supplementary Fig. 2A), andthe remaining Foxp3þ Tregs (�45%–50%) are HeliosLOW inDTA-1–treated tumors. Taken together, these data suggest thatapproximately 75% to 80% of the Tregs (compared with controlIgG) in the tumor have been modulated by DTA-1.

HeliosLOW Tregs in DTA-1–treated tumors express moreof the prosurvival genes BCL-2 and BCLXL than Helios

HIGH ortotal IgG Tregs (Fig. 2B). This phenotype extends to the peakof immune infiltration at day 10, but by day 14, even thoughthe tumors are regressing, the majority of the remainingTregs are HeliosHIGH (Fig. 2B). Tregs with the lowest levelsof Helios at day 7 also displayed a pronounced reduction ofFoxp3 and CD25 (Fig. 2C). Helios loss seems to parallel theextent to which free cell surface GITR is saturated/modulatedby DTA-1, preventing further staining on Tregs (Fig. 2D). Atday 14, the 1 mg/mouse dose of DTA-1 no longer saturatesavailable GITR, and intratumor Tregs in DTA-1–treated micedisplay similar Helios expression compared with IgG (day 14posttumor challenge in Fig. 2B and D). This supports theconclusion that Tregs lose Foxp3 expression after tumorinfiltration, with gross changes in Helios expression being areliable marker of GITR modulation. Increased expression ofBCL-2 and BCLXL, combined with the lack of activated cas-pase-3 (Supplementary Fig. S1), shows that modulated Tregsmaintain a prosurvival phenotype.

GITR stimulation alters Treg lineage stabilityTregs naturally co-opt and express inflammatory T-cell

lineage transcription factors (T-bet, RORgt) to facilitate thesuppression of the corresponding Teff program (29, 30). WhenFoxp3 expression is ablated in Tregs, they have been shown torevert to cells with a Teff phenotype. In addition, Helios hasbeen shown to stabilize Treg programming and suppress IL-2expression (31, 32). Therefore, DTA-1–treated Tregs that havelost or are losing Foxp3 expression could acquire a Teff-likephenotype. In contrast with the reduced expression of Foxp3and CD25 in DTA-1–treated mice, HeliosLOW Tregs showed

Schaer et al.





increased protein expression of T-cell lineage transcriptionfactors such as T-bet, RORgt, and Eomes, compared withHeliosHIGH Tregs (Fig. 3A). Expression in HeliosLOW Tregs wasalso higher than in Tregs in the IgG control groups at multipletime points (day 7 for T-bet, days 7–14 for Eomes, days 10 and14 for RORgt; Fig. 3A).

To determine whether increased T-bet, RORgt, and Eomesprotein levels in Tregs has biologic consequence, we con-ducted a cytokine recall assay on cells isolated from tumors10 days after DTA-1 treatment. Foxp3-GFP mice were usedfor this experiment because the staining for Foxp3 andHelios is diminished and unreliable after PMA/ionomycin

A

B

C D

Figure 2. DTA-1–modulated Tregs show reduced Helios expression and a prosurvival phenotype. A, representative FACS plots show the percentage ofHeliosLOW Tregs (live, CD45þ, CD3þ, CD4þ Foxp3þ) in pooled spleens and individual tumors of DTA-1- and IgG–treated mice 11 days after tumor challenge.B, example FACS histograms (top) show Helios expression of IgG- (gray filled) and DTA-1–treated (black line) tumor-infiltrating Tregs on indicated dayafter tumor challenge. Bottom histograms show comparison of BCL-2 expression in HeliosHIGH (dashed line) with HeliosLOW (solid black line) DTA-1–treatedTregs. Mean � SEM for the percentage of HeliosLOW Tregs in IgG- and DTA-1–treated tumors, mean fluorescence intensity (MFI) of BCL-2 and BCLXLfor IgG Tregs, compared with HeliosLOW DTA-1 Tregs at each time point is shown in the graphs. C, representative Foxp3 and CD25 expression of IgGTregs (gray filed) versus DTA-1–treated HeliosHIGH (dashed line) and HeliosLOW (black solid line) Tregs, 7 days after tumor challenge. D, example FACSplots show Foxp3 and GITR (DTA-1-PE-Cy7) staining of CD4 T cells in IgG tumors (day 10 post tumor challenge) compared with DTA-1–treatedtumors on days 7, 10, and 14 posttumor challenge. Experiments were repeated three times with 4 to 5 per group, with one representative experiment shown.�, P < 0.01; ��, P < 0.001; ��, P < 0.0001. TC, tumor challenge.






stimulation (Supplementary Fig. S2B). Using Foxp3-GFPmice also allowed us to circumvent this technical hurdleas low levels of Foxp3 expression correlate with loss of

Helios (Fig. 2C), allowing us to subset our analysis toFoxp3-GFPLOW and Foxp3-GFPHIGH Tregs. GFPLOW Tregs(HeliosLOW) in DTA-1–treated mice showed a more than

A

B C

D

,

,

,

,

,

,,

,

g

Day post TC Day post TC Day post TC

g-

g-

IFN-g

Figure 3. DTA-1–treated Tregs display a Teff-like profile. A, Helios expression compared with T-bet, RORgt, and Eomes in Tregs (CD45þ, CD3þ, CD4þ) fromDTA-1–treated tumors is shown in representative plots. Graphs show the mean� SEM for mean fluorescence intensity (MFI) of these markers for IgG Tregs,compared with DTA-1–modulated HeliosLOW Tregs at each time point. B, IFN-g recalls expression in GFP high (gray shaded) compared with GFP low Tregs(black line) from day 10 IgG- and DTA-1–treated tumors. Graph shows themean� SEM IFN-g expression in IgG Tregs, compared with GFP lowDTA-1 Tregs.C and D, Tregs and Teffs were sorted from individual mice as described in Materials and Methods from indicated tissue and time points for gene expressionanalysis. Graphs compare the level of IL-10, IFN-g (C), Foxp3, and Helios (D) expression in IgG- compared with DTA-1–treated tumors. Splenic Tregs andtumor Teffs are provided as controls. Experimentswere repeated three timeswith 4 to 5per group (A andB) and two timeswith 10per group (CandD),with onerepresentative experiment shown. �, P < 0.01; ��, P < 0.001; ��, P < 0.0001. TC, tumor challenge; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Schaer et al.





2-fold increase in IFN-g production compared with controlIgG-treated Tregs (Fig. 3B). Although there was no differ-ence in the IFN-g expression between GFPHIGH and GFPLOW

cells in IgG control tumors (Fig. 3B, top), IFN-g expressionwas restricted to GFPLOW in DTA-1–treated Tregs (Fig. 3B,bottom). Despite increased RORgt expression in HeliosLOW

Tregs, we did not detect any significant difference betweenIgG and DTA-1–treated Tregs in its related cytokine IL-17(data not shown). To confirm this result and more closelymeasure the changes in Treg lineage phenotype, we sortedFoxp3-GFP Tregs from individual tumors and measuredthe expression of relevant Tregs and Teff genes. Using thisapproach, we found a maximum of 4-fold upregulation inIFN-g expression (day 7) and approximately 2-fold decreasein IL-10 expression (day 10) in DTA-1–treated Tregs (Fig.3C). Other markers, such as GITR, IL-2, IL-17, TNF-a, TGF-b,and SATB1, were expressed to equivalent levels in DTA-1–treated and IgG-treated Tregs (data not shown). AlthoughHelios protein levels after DTA-1 treatment correlated withreduced Helios gene expression, there was no major differ-ence in Foxp3 gene expression (Fig. 3D). This would indicatethat GITR signaling may cause a posttranscription modifi-cation that leads to reduced Foxp3 protein expression.Regardless of the mechanism responsible for the loss ofFoxp3 and Helios expression, these results suggest thatDTA-1 induces Treg lineage instability and acquisition ofa Teff-like profile.

DTA-1–induced lineage instability removes Treg-suppressive function from the tumorTo determine whether the phenotypic changes described

above alter Treg-suppressive function in vivo, we used an exvivo collagen-fibrin gel matrix culture to measure CD8þ cyto-lytic T-cell (CTL) effector function against tumor cells fromcontrol IgG- or DTA-1–treated mice (20). Collagen-fibrin gelsmimic a three-dimensional tissue-like environment and aremore sensitive than packed cell-pellet assays at measuringCD8þ CTL effector function (20). Furthermore, we have foundthat collagen-fibrin gel cultures of explanted B16 or B16-expressing OVA (B16-OVA) tumors, which include all infiltrat-ing cells, are resistant to killing by a 10- to 50-fold excess ofin vitro cognate antigen-activated CD8þ CTL, recapitulatingthe suppression that exists in vivo (Fig. 4A; and Budhu andSchaer; unpublished data).Consistent with prior results, control IgG-treated tumors

become resistant to killing by in vitro activated CTLs andproliferate in the collagen gels after 24 hours, with the numberof tumor cells increasing overtime (Fig. 4A; Budhu and Schaer;unpublished data). In contrast, DTA-1 treatment causedtumors to remain susceptible to ex vivo killing by activatedCTLs, and the number of viable tumor cells continued todecrease at 48 and 72 hours (2-fold and 3-fold, respectively,vs. 0 hour; Fig. 4A). Calculation of the killing efficiency, k (asdescribed in Materials and Methods and in ref. 20) highlightsthe differences between DTA-1– and control IgG-treatedtumors. Killing efficiency of CTLs in DTA-1–treated tumorsincreases over 2-fold at 48 hours (5.3� 10�10 at 24 hours to 1.3� 10�9 at 48 hours; Fig. 4A) in contrast with that in IgG-treated

mice, which maintains suppression. Ex vivo addition of DTA-1had no effect on the killing of DTA-1–treated tumors, controlIgG-treated tumors, or cultured B16 cells, and GITR�/� CTLkilled tumor cells from DTA-1–treated tumors and culturedB16 cells at the same rate asGITRþ/þCTL (Fig. 4B, dashed linesand green lines vs. red lines). This suggests that killing isindependent of GITR stimulation by DTA-1 on CTL (Fig.4B). Combined, our data support the conclusion that GITRmodulation of Tregs by DTA-1 removes their suppressiveinfluence in the tumor microenvironment.

DiscussionThe overarching goal of cancer immunotherapy has been

the activation of tumor-specific immunity that is able toovercome the hurdles established by tumors to evade immunedestruction. GITR activation seems to reach an importantbalance by enhancing tumor immunity while inhibitingimmune suppression in a tumor-dependent manner. Theresearch presented here shows that in addition to its estab-lished role in modulating Teffs, DTA-1 treatment causes Tregsto lose lineage stability, reducing their suppressive influenceover the tumor microenvironment.

Our data suggest that conditions present in tumor-bearingmice and the tumor microenvironment are responsible formaking Tregs susceptible to GITR-induced Foxp3 loss.Reduced IL-2 levels have been shown to be important for Tregstability and homeostasis (33, 34). However, we do not believethat the lack of IL-2 accounts for Treg instability in our systembecause transferred Tregs lose Foxp3 in the periphery evenafter transfer into lymphoreplete hosts. In addition, equalnumbers of cotransferred tumor-experienced and na€�veTregs are recovered from DTA-1–treated animals, despite theloss of Foxp3 expression in tumor-experienced Tregs. Thissuggests that DTA-1 does not simply deplete Foxp3þ Tregs(Fig. 1D and Supplementary Fig. S1B). Only upon tumorinfiltration in DTA-1–treated animals do na€�ve donor Tregsmanifest significant Foxp3 loss, highlighting further the role oftumor conditioning on Tregs and even at steady state. There-fore, although the detailed mechanism of GITR signaling-induced Foxp3 loss requires further investigation, it is evidentthat tumor preconditioning and the tumor microenvironmentplay amajor role in permitting GITR-dependentmodulation ofFoxp3 expression.

The reduction of CD25 expression and the production ofIFN-g observed in intratumor Tregs during DTA-1 therapy(Figs. 2 and 3) are similar to what has been reported whenFoxp3 is deleted inmature Tregs (35). There has been evidencesuggesting that inflammatory environments cause Tregs tolose stability and convert to a Teff-like phenotype (29); how-ever, recent research has brought these findings into question.Results from Miyao and colleagues and Zhou and colleaguessuggest that the conversion of Tregs into Teffs is actually due toa transient expression of Foxp3 in non-Tregs (36, 37). It isunlikely that the DTA-1–induced Treg lineage conversion weobserve here is an artifact of lineagemarking. The Treg transferand gene expression analysis experiments (Figs. 1 and 3) rely onsorting an entire Foxp3-GFP–positive Treg population and donot use a lineage marking Cre recombinase system. In fact, we






were unable to use Foxp3-Cre mice due to the "leaky" lineagemarking seen during backcrossing to the C57BL/6 background(data not shown). Thus, we believe the results presented hereillustrate that DTA-1–mediated GITR stimulation causestumor-specific reprogramming of Tregs into a Teff-like phe-notype. As we were unable to isolate or phenotype repolarizedFoxp3� Tregs using the Foxp3-Cre lineage marking mice, itremains to be established whether the conversion of Tregs toa Teff-like profile is necessary or secondary to the loss ofFoxp3/suppressive function. Development of complex geneticmodels would be needed to answer this question and deter-

mine whether former DTA-1–modulated Tregs work to poten-tiate antitumor immunity after losing suppressive capacity.

How DTA-1–induced GITR signaling leads to Foxp3 degra-dation is an important question. Expression levels of Foxp3mRNA were comparable between control IgG- and DTA-1–treated mice, but there is a marked reduction in Foxp3 proteinlevels (Figs. 3B and 2C). This would suggest that downstreamsignaling fromGITR imposes posttranscriptional or posttrans-lational control of Foxp3 protein expression. Although down-stream signaling from GITR induced by GITR-L was recentlyshown to alter Treg-suppressive function through the

A

B

Figure 4. Treg lineage instability removes intratumor immune suppression. A and B, experiment schematic: tumors were isolated and dissociated, and10,000 live tumor cells were then embedded along with all tumor-infiltrating cells (�3–5 times tumor cell counts) in collagen-fibrin gel together with or withoutCTLs as described in Materials and Methods. After 24, 48, and 72 hours, gels were lysed and viable cells were cultured in a colony-forming assay. Nokilling would appear with 100þ colonies, and killing would show very few colonies. A, graphs show number of viable tumor cells recovered at indicated timepoints for IgG (left) and DTA-1 (middle) for total tumors alone (blue line) or with activatedOT-1 T cells (red line). Right, rate of B16 cell killing byOT-1CTL (killingconstant k, as calculated inMateriala andMethods), of IgG (dark gray) andDTA-1 (white) tumors is showncomparedwith primary tissue cultureB16cells alone(light gray). B, viable tumor cells recovered from cultures of IgG- and DTA-1–treated total tumors alone (blue), with GITRþ/þ Pmel-1 (red), and GITR�/� Pmel-1(green). Dashed lines indicate cultures that included the ex vivo addition of 10 mg/mL of DTA-1; solid ones indicate control cultures. Experiments wererepeated three times with tumors pooled from 3 to 5 mice for each experiment. Mean and � SEM of three experiments is shown in A; a representativeexperiment is shown in B. �, P < 0.01.

Schaer et al.





activation of c-jun-NH2-kinase (JNK), it is unclear whetherDTA-1 causes a similar effect (38). JNK activation after long-term GITR-L stimulus resulted in reduced Foxp3 mRNAexpression to a level that we did not observe with DTA-1treatment. GITR and TNFR family members use TNFR-asso-ciated factor (TRAF) proteins to transmit downstream signals(8, 39). Because many TRAF proteins function as E3 ubiquitinligases, one hypothesis could be that overstimulation of GITRby DTA-1 could cause an intersection of this cascade withFoxp3 protein and targeting it for degradation. Because intra-tumor Tregs express less Foxp3 mRNA than peripheral Tregs(Fig. 3B), this may make them uniquely sensitive to GITR-induced degradation of Foxp3.A propensity to modulate pTregs over tTregs would be a

logical assumption considering their unstable nature (29).However, in the case of B16 melanoma, it seems that themajority of intratumor Tregs have a tTreg-like phenotype, ashas been seen in 4T1 tumors, and without a minor pTregpopulation as seen in other tumors (25). In fact, transferexperiments into Rag�/� mice established that a majority ofTregs can be rendered susceptible to GITR-induced loss ofFoxp3. We found a similar result, with 75% to 80% of Tregsmodulated in the tumor microenvironment during DTA-1therapy in wild-type mice (% of intratumor Treg Foxp3 lossþ % Foxp3þHeliosLOW Tregs; Supplementary Fig. S2A andS2B). This suggests that the effects of DTA-1 are not limitedto a minor subset of Tregs, such as pTregs. Regardless, DTA-1treatment caused Tregs to lose Helios protein and geneexpression, corresponding with increased levels of inflamma-tory T-cell transcription factors, T-bet, RORgt, and Eomes.Treg expression of T-bet or RORgt is not unprecedented, andthe expression of these transcription factors is important forthe Treg-suppressive function (29). Surprisingly, Eomes, tra-ditionally thought of as a CD8þ CTL transcription factor, ishighly upregulated in the DTA-1–treated Tregs. We havereported recently that simulation of the closely related TNFRfamily member OX40 has the ability to induce Eomes in CD4Teffs (40). Even though there has been evidence that Tregscould control immunity through granzyme-dependent killingof B cells, to date no role for Eomes in Treg function has beendescribed (41). The significance of Eomes expression in DTA-1modulation of Tregs will require further investigation; how-ever, it exemplifies the level to which overstimulation ofGITR on susceptible Tregs can alter their lineage programThe end result of Treg lineage instability caused by GITR

immunotherapy is the removal of intratumor suppressionmediated by Tregs, as shown by the collagen-fibrin gel killingassay (Fig. 4). Using the same approach, we recently deter-mined that intratumor immune suppression in B16 tumorsis Treg dependent, as specific in vivo depletion of Tregsrestores killing of explanted tumors (Budhu and Schaer;unpublished data). Whether or not the DTA-1 effect is dueto reduced intratumor Treg numbers, Treg lineage instabil-ity, or a combination of both remains to be determined.Interestingly, even though GITR treatment removes Tregsuppression and DTA-1–treated tumors are regressingin vivo, tumor cells cocultured with total infiltrates continueto grow ex vivo (Fig. 4). We interpret the need for additional

input of Teffs to continue killing as evidence that for optimalin vivo therapy, GITR's ability to enhance CD8þ T-cell numbersand persistence also plays an important role (42). Consequent-ly, targeting Tregs seems to be a major mechanism for DTA-1treatment along with its intrinsic effects on CD8þ T cells. Thisconclusion is in agreement with our prior results showing thatboth Tregs and Teffs must express GITR for the optimal effectsof DTA-1 (17).

Development of new immunotherapies that accelerateantitumor immunity is important, as checkpoint blockadedoes not benefit all patients (2, 3). Our data show that ligationof GITR can accomplish both goals. By inducing Treg lineageinstability, DTA-1 releases an important source of suppres-sion of tumor immunity. At the same time, we and others haveshown that GITR ligation by DTA-1 accelerates antitumorimmunity to take advantage of the now permissive tumormicroenvironment (12, 17). The unique ability of GITR ligationto target both axes, modulating Tregs primarily in the tumormicroenvironment, supports the continued clinical develop-ment of GITR agonist agents. Accordingly, in collaborationswith GITR Inc., we are currently investigating the agonist anti-human GITR antibody, TRX-518, in a phase I first-in-humantrial (GITR Inc., Clinical trials.gov: NCT01239134). We believethat the knowledge gained from our study in understandingGITR mechanism of action will help facilitate the develop-ment of appropriate biomarkers and inform rational designof future clinical trials.

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

Authors' ContributionsConception and design: D.A. Schaer, C. Liu, A.D. Cohen, A.N. Houghton,T. Merghoub, J.D. WolchokDevelopment of methodology: D.A. Schaer, C. Liu, A.D. Cohen, T. MerghoubAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): D.A. Schaer, S. Budhu, C. Liu, C.F. Bryson, N.M.Malandro, A.D. Cohen, H. Zhong, X. Yang, T. MerghoubAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): D.A. Schaer, S. Budhu, C. Liu, C.F. Bryson, N.M.Malandro, A.D. Cohen, T. Merghoub, J.D. WolchokWriting, review, and/or revision of the manuscript: D.A. Schaer, S. Budhu,T. Merghoub, J.D. WolchokAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): C. Liu, H. ZhongStudy supervision: A.N. Houghton, T. Merghoub, J.D. Wolchok

AcknowledgmentsThe authors thank current and former Wolchok Lab members: Dr.

Stephanie Terzulli, Andre Burey, Judith Murphy, Kelly Crowley, RodgerPellegrini, Drs. Arvin Yang, and Francesca Avogadri for their support on theGITR project; Rudensky lab members: Drs. Steve Josefowicz, Rachel Niec,Ashutosh Chaudhry, and Robert Samstein for generous sharing of reagentsand always being available for advice and thoughtful discussion about Treglineage stability; Dr. Joe Ponte for valuable shared insight on the mechanismof GITR immunotherapy; Dr. Michael Curran for assistance with experimen-tal design; members of the MSKCC flow cytometry, and molecular cytologycore facilities; and Dr. Roberta Zappasodi for her very helpful comments,critical reading, and editing of this manuscript.

Grant SupportThis work was supported by NIH grants R01CA56821, P01CA33049, and

P01CA59350 (to A.N. Houghton and J.D. Wolchok), D.A. Schaer was supportedby the NIH Clinical Training for Scholar Grant K12 CA120121-01, and receivedsupport from the NIH/NCI Immunology Training Grant T32 CA09149-30and JohnD. Proctor Foundation:Margaret A. Cunningham ImmuneMechanismsin Cancer Research Fellowship Award; Swim Across America; the Mr. William






H.Goodwin andMrs. AliceGoodwin and theCommonwealth Cancer Foundationfor Research and the Experimental Therapeutics Center of MSKCC (to J.D.Wolchok).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received June 26, 2013; revised August 23, 2013; accepted September 8, 2013;published OnlineFirst September 16, 2013.

References1. Schreiber RD, Old LJ, Smyth MJ. Cancer immunoediting: integrating

immunity's roles in cancer suppression and promotion. Science 2011;331:1565–70.

2. Hodi FS, O'Day SJ, McDermott DF, Weber RW, Sosman JA, HaanenJB, et al. Improved survival with ipilimumab in patients with metastaticmelanoma. N Engl J Med 2010;363:711–23.

3. Topalian SL, Hodi FS, Brahmer JR, Gettinger SN, Smith DC, McDer-mott DF, et al. Safety, activity, and immune correlates of anti-PD-1antibody in cancer. N Engl J Med 2012;366:2443–54.

4. Brahmer JR, Tykodi SS, Chow LQ, HwuWJ, Topalian SL, Hwu P, et al.Safety and activity of anti-PD-L1 antibody in patients with advancedcancer. N Engl J Med 2012;366:2455–65.

5. HamidO, Robert C, DaudA, Hodi FS, HwuW-J, Kefford R, et al. Safetyand tumor responses with lambrolizumab (Anti–PD-1) in melanoma. NEngl J Med 2013;369:134–44.

6. Wolchok JD, Kluger H, Callahan MK, Postow MA, Rizvi NA, LesokhinAM, et al. Nivolumab plus ipilimumab in advancedmelanoma. N Engl JMed 2013;369:122–33.

7. Royal RE, Levy C, Turner K, Mathur A, Hughes M, Kammula US, et al.Phase 2 trial of single agent Ipilimumab (anti-CTLA-4) for locallyadvanced or metastatic pancreatic adenocarcinoma. J Immunother2010;33:828–33.

8. Snell LM, Lin GHY, McPherson AJ, Moraes TJ, Watts TH. T-cellintrinsic effects of GITR and 4-1BB during viral infection and cancerimmunotherapy. Immunol Rev 2011;244:197–217.

9. Weinberg AD, Morris NP, Kovacsovics-Bankowski M, Urba WJ, CurtiBD. Science gone translational: the OX40 agonist story. Immunol Rev2011;244:218–31.

10. Shimizu J, Yamazaki S, Takahashi T, Ishida Y, Sakaguchi S. Stimu-lation of CD25(þ)CD4(þ) regulatory T cells through GITR breaksimmunological self-tolerance. Nat Immunol 2002;3:135–42.

11. TurkMJ, Guevara-Patino JA, RizzutoGA, EngelhornME, Sakaguchi S,Houghton AN. Concomitant tumor immunity to a poorly immunogenicmelanoma is prevented by regulatory T cells. J Exp Med 2004;200:771–82.

12. Schaer DA, Murphy JT, Wolchok JD. Modulation of GITR for cancerimmunotherapy. Curr Opin Immunol 2012;24:217–24.

13. Kanamaru F, Youngnak P, Hashiguchi M, Nishioka T, Takahashi T,Sakaguchi S, et al. Costimulation via glucocorticoid-induced TNFreceptor in both conventional and CD25þ regulatory CD4þ T cells.J Immunol 2004;172:7306–14.

14. Stephens GL, McHugh RS, Whitters MJ, Young DA, Luxenberg D,Carreno BM, et al. Engagement of glucocorticoid-induced TNFRfamily-related receptor on effector T cells by its ligand mediatesresistance to suppression by CD4þCD25þ T cells. J Immunol 2004;173:5008–20.

15. Ramirez-Montagut T, Chow A, Hirschhorn-Cymerman D, Terwey TH,Kochman AA, Lu S, et al. Glucocorticoid-induced TNF receptor familyrelated gene activation overcomes tolerance/ignorance to melanomadifferentiation antigens and enhances antitumor immunity. J Immunol2006;176:6434–42.

16. Cohen AD, Diab A, Perales MA, Wolchok JD, Rizzuto G, Merghoub T,et al. Agonist anti-GITR antibody enhances vaccine-induced CD8(þ)T-cell responses and tumor immunity. Cancer Res 2006;66:4904–12.

17. CohenAD, Schaer DA, LiuC, Li Y, Hirschhorn-CymmermanD,KimSC,et al. Agonist anti-GITR monoclonal antibody induces melanomatumor immunity in mice by altering regulatory T cell stability andintra-tumor accumulation. PLoS ONE 2010;5:e10436.

18. Coe D, Begom S, Addey C, White M, Dyson J, Chai JG. Depletion ofregulatory T cells by anti-GITR mAb as a novel mechanism for cancerimmunotherapy. Cancer Immunol Immunother 2010;59:1367–77.

19. Wang S, Bartido S, Yang G, Qin J, Moroi Y, Panageas KS, et al. A rolefor a melanosome transport signal in accessing the MHC class IIpresentation pathway and in eliciting CD4þ T cell responses. J Immu-nol 1999;163:5820–6.

20. Budhu S, Loike JD, Pandolfi A, Han S, Catalano G, Constantinescu A,et al. CD8þ T cell concentration determines their efficiency in killingcognate antigen-expressing syngeneic mammalian cells in vitro and inmouse tissues. J Exp Med 2010;207:223–35.

21. Komatsu N,Mariotti-Ferrandiz ME,Wang Y,Malissen B,WaldmannH,Hori S. Heterogeneity of natural Foxp3þ T cells: a committed regu-latory T-cell lineage and an uncommitted minor population retainingplasticity. Proc Natl Acad Sci 2009;106:1903–8.

22. Ephrem A, Epstein AL, Stephens GL, Thornton AM, Glass D, ShevachEM. Modulation of Treg cells/Teffector function by GITR signaling iscontext-dependent. Eur J Immunol 2013May30. [Epub aheadof print].

23. Yadav M, Louvet C, Davini D, Gardner JM, Martinez-Llordella M,Bailey-Bucktrout S, et al. Neuropilin-1 distinguishes natural and induc-ible regulatory T cells among regulatory T cell subsets in vivo. J ExpMed 2012;209:1713–22.

24. Hansen W, Hutzler M, Abel S, Alter C, Stockmann C, Kliche S, et al.Neuropilin 1 deficiency on CD4þFoxp3þ regulatory T cells impairsmouse melanoma growth. J Exp Med 2012;209:2001–16.

25. Weiss JM, Bilate AM, Gobert M, Ding Y, Curotto de Lafaille MA,Parkhurst CN, et al. Neuropilin 1 is expressed on thymus-derivednatural regulatory T cells, but not mucosa-generated induced Foxp3þT reg cells. J Exp Med 2012;209:1723–42.

26. GottschalkRA,Corse E, Allison JP. Expression ofHelios in peripherallyinduced Foxp3þ regulatory T cells. J Immunology 2012;188:976–80.

27. Zabransky DJ, Nirschl CJ, Durham NM, Park BV, Ceccato CM, BrunoTC, et al. Phenotypic and functional properties of Heliosþ regulatory Tcells. PLoS ONE 2012;7:e34547.

28. SchaerDA, Li Y,MerghoubT, RizzutoGA,ShemeshA,CohenAD, et al.Detection of intra-tumor self antigen recognition during melanomatumor progression in mice using advanced multimode confocal/twophoton microscope. PLoS ONE 2011;6:e21214.

29. Hori S. Stability of regulatory T-cell lineage. Adv Immunol 2011;112:1–24.

30. Chaudhry A, Samstein RM, Treuting P, Liang Y, Pils MC, Heinrich JM,et al. Interleukin-10 signaling in regulatory T cells is required forsuppression of Th17 cell-mediated inflammation. Immunity 2011;34:566–78.

31. Getnet D, Grosso JF, GoldbergMV, Harris TJ, Yen HR, Bruno TC, et al.A role for the transcription factor Helios in human CD4þCD25þregulatory T cells. Mol Immunol 2010;47:1595–600.

32. Baine I, Basu S, Ames R, Sellers RS, Macian F. Helios inducesepigenetic silencing of il2 gene expression in regulatory T cells.J Immunol 2013;190:1008–16.

33. Setoguchi R, Hori S, Takahashi T, Sakaguchi S. Homeostatic main-tenance of natural Foxp3þ CD25þ CD4þ regulatory T cells by inter-leukin (IL)-2 and induction of autoimmune disease by IL-2 neutraliza-tion. J Exp Med 2005;201:723–35.

34. Fontenot JD, Rasmussen JP, Gavin MA, Rudensky AY. A function forinterleukin 2 in Foxp3-expressing regulatory T cells. Nat Immunol2005;6:1142–51.

35. Williams LM, Rudensky AY. Maintenance of the Foxp3-dependentdevelopmental program inmature regulatory Tcells requires continuedexpression of Foxp3. Nat Immunol 2007;8:277–84.

36. Miyao T, Floess S, Setoguchi R, Luche H, Fehling HJ, Waldmann H,et al. Plasticity of Foxp3(þ) T cells reflects promiscuous Foxp3 expres-sion in conventional T cells but not reprogrammingof regulatory Tcells.Immunity 2012;36:262–75.

Schaer et al.





37. Zhou X, Bailey-Bucktrout SL, Jeker LT, Penaranda C, Martinez-Llor-della M, Ashby M, et al. Instability of the transcription factor Foxp3leads to the generation of pathogenic memory T cells in vivo. NatImmunol 2009;10:1000–7.

38. Joetham A, Ohnishi H, Okamoto M, Takeda K, Schedel M, Dome-nico J, et al. Loss of T regulatory cell suppression following sig-naling through glucocorticoid-induced tumor necrosis receptor(GITR) is dependent on c-Jun N-terminal kinase activation. J BiolChem 2012;287:17100–8.

39. Hacker H, Tseng PH, Karin M. Expanding TRAF function: TRAF3 as atri-faced immune regulator. Nat Rev Immunol 2011;11:457–68.

40. Hirschhorn-Cymerman D, Budhu S, Kitano S, Liu C, Zhao F, Zhong H,et al. Induction of tumoricidal function in CD4þ T cells is associatedwith concomitant memory and terminally differentiated phenotype.J Exp Med 2012;209:2113–26.

41. Zhao D-M, Thornton AM, DiPaolo RJ, Shevach EM. ActivatedCD4þCD25þ T cells selectively kill B lymphocytes. Blood 2006;107:3925–32.

42. Snell LM, McPherson AJ, Lin GH, Sakaguchi S, Pandolfi PP, RiccardiC, et al. CD8 T cell-intrinsic GITR is required for T cell clonal expansionand mouse survival following severe influenza infection. J Immunol2010;185:7223–34.






2013;1:320-331. Published OnlineFirst September 16, 2013.Cancer Immunol Res David A. Schaer, Sadna Budhu, Cailian Liu, et al. through Loss of Regulatory T-cell Lineage StabilityGITR Pathway Activation Abrogates Tumor Immune Suppression

Updated version

10.1158/2326-6066.CIR-13-0086doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerimmunolres.aacrjournals.org/content/suppl/2013/09/17/2326-6066.CIR-13-0086.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerimmunolres.aacrjournals.org/content/1/5/320.full#ref-list-1

This article cites 41 articles, 19 of which you can access for free at:

Citing articles

http://cancerimmunolres.aacrjournals.org/content/1/5/320.full#related-urls

This article has been cited by 20 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerimmunolres.aacrjournals.org/content/1/5/320To request permission to re-use all or part of this article, use this link


http://cancerimmunolres.aacrjournals.org/lookup/doi/10.1158/2326-6066.CIR-13-0086http://cancerimmunolres.aacrjournals.org/content/suppl/2013/09/17/2326-6066.CIR-13-0086.DC1http://cancerimmunolres.aacrjournals.org/content/1/5/320.full#ref-list-1http://cancerimmunolres.aacrjournals.org/content/1/5/320.full#related-urlshttp://cancerimmunolres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerimmunolres.aacrjournals.org/content/1/5/320http://cancerimmunolres.aacrjournals.org/

Documents

GITR Pathway Activation Abrogates Tumor Immune Suppression through … · Research Article GITR Pathway Activation Abrogates Tumor Immune Suppression through Loss of Regulatory T-cell