GARP Dampens Cancer Immunity by Sustaining Function and ... · Tumor Biology and Immunology GARP...

Tumor Biology and Immunology

GARP Dampens Cancer Immunity by SustainingFunction and Accumulation of Regulatory T Cellsin the ColonMohammad Salem1, Caroline Wallace1, Maria Velegraki1, Anqi Li1, Ephraim Ansa-Addo1,Alessandra Metelli1, Hyunwoo Kwon1, Brian Riesenberg1, Bill Wu1, Yongliang Zhang1,Silvia Guglietta1, Shaoli Sun2, Bei Liu1, and Zihai Li1,3

Abstract

Activated regulatory T (Treg) cells express the surfacereceptor glycoprotein-A repetitions predominant (GARP),which binds and activates latent TGFb. How GARP modu-lates Treg function in inflammation and cancer remainsunclear. Here we demonstrate that loss of GARP in Treg cellsleads to spontaneous inflammation with highly activatedCD4þ and CD8þ T cells and development of enteritis. Tregcells lacking GARP were unable to suppress pathogenic T-cellresponses in multiple models of inflammation, includingT-cell transfer colitis. GARP�/� Treg cells were significantlyreduced in the gut and exhibited a reduction in CD103expression, a colon-specific migratory marker. In the coli-

tis-associated colon cancer model, GARP on Treg cells damp-ened immune surveillance, and mice with GARP�/� Treg cellsexhibited improved antitumor immunity. Thus, GARPempowers the functionality of Treg cells and their tissue-specific accumulation, highlighting the importance of cellsurface TGFb in Treg function and GARP as a potentialtherapeutic target for colorectal cancer therapy.

Significance: These findings uncover functions ofmembrane-bound TGFb and GARP that tune the activity ofTreg cells, highlighting a potential treatment strategy in auto-immune diseases and cancer.

IntroductionTo maintain immune homeostasis and uphold tolerance to

self-antigens, CD4þ regulatory T (Treg) cells spearhead the adjust-ment of the immune responses generated by other effector T-cellpopulations (1). Treg cells are a potent suppressive cell subset thatis classically characterized by expression of the master regulatorytranscription factor forkhead box protein 3 (Foxp3; ref. 2). Loss-of-function mutations in Foxp3 gene lead to a severe systemicautoimmune disorder characterized by fatal tissue and organdamage (3). In cancer, increased numbers of infiltrating Treg cellsinto tumor microenvironments suppress antitumor responsesand have been clinically associated with poor prognosis (4).Consequently, identifying possiblemechanisms thatwould allowthe modulation of Tregs function is urgently needed.

TGFb is a preeminent cytokine with critical immunomodula-tory roles (5). TGFb is highly implicated in the induction, devel-opment, and maintenance of Treg cells. Mice with global TGFb

deletion develop a severe, generalized autoimmune disorder anddie within 3–4weeks after birth (6). Mice lacking TGFb receptor IIon T cells showed similar disease characteristics with aberrantT-cell activation and increased Th1 and Th2 cell populations (7),which suggests that TGFb plays an important role in immunesurveillance and suppression of effector T cells. Lack of TGFbreceptor on CD4þ T cells leads to disappearance of thymic Tregcells during postnatal days 3–5 (8).Moreover, peripheral Tregs aresignificantly reduced in numbers compared with thymic Tregs in8- to 10-day-old TGFb-deficientmice (9). In fact, TGFb is requiredto generate peripheral Treg cells (10) by activating the canonicalTGFb/SMADs signaling;micewith Smad2 and Smad3deletions onCD4þ T cells also developed severe autoimmune disease withreduced Foxp3 expression in the peripheral CD4þ T cells (11).Furthermore, Treg cells with deficient TGFb signaling failed tohome to the inflammatory sites in a T-cell transfer colitis mod-el (12). Despite the increased knowledge of how TGFb influencesthe suppressive function of Treg cells, underlying mechanismsthat regulate the bioavailability and activation of TGFb on Tregcells remain unclear.

Glycoprotein-A repetitions predominant (GARP), a transmem-brane protein encoded by the gene Leucine rich repeat containingprotein 32 (Lrrc32), is a cell surface docking receptor for latentTGFb, a tetra-peptide complex formed by the TGFbdimer and twocopies of latency-associated peptide (LAP; ref. 13). The release ofthe biologically active mature TGFb from this complex can occurby several factors including heat, acidic conditions, and integ-rins (14). Earlier reports have shown that GARP-deficient CD4þ

T cells mice did not impair the suppressive function of Treg cellsin vitro (15) and that silencing of its expression by RNA interfer-ence does not significantly affect Foxp3 expression in expanded

1Department of Immunology and Microbiology, Hollings Cancer Center, MedicalUniversity of South Carolina, Charleston, South Carolina. 2Department of Pathol-ogy and Laboratory Medicine, Medical University of South Carolina, Charleston,South Carolina. 3First Affiliated Hospital, Zhengzhou University School ofMedicine, Zhengzhou, China.

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

Corresponding Author: Zihai Li, Medical University of South Carolina, 86Jonathan Lucas Street, Charleston, SC 29425. Phone: 843-792-5342; Fax: 843-792-9588; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-18-2623

�2019 American Association for Cancer Research.

CancerResearch

Cancer Res; 79(6) March 15, 20191178

on April 13, 2020. © 2019 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 23, 2019; DOI: 10.1158/0008-5472.CAN-18-2623

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-18-2623&domain=pdf&date_stamp=2019-2-28

http://cancerres.aacrjournals.org/

Tregs (16). In contrast, other studies have reported that solubleGARP canhave abeneficial effect in sustaining Treg differentiationin xenogeneic Graft-versus-Host disease (17). Moreover, a mAbdirected against the GARP/latent TGFb complex blocked Treg cell-mediated TGFb production in the same mouse model (18).However, the roles of GARP in Treg development, lineage stabil-ity, and function have not been completely elucidated.

To address the above question, we generated Treg-specificGARP knockout mice and found that GARP plays an importantrole in immune homeostasis and aged mice with GARP�/� Tregcells develop spontaneous intestinal inflammation. Tregs lackingGARP show reduced ability to suppress inflammatory responsesand less accumulation in the intestinal tract. Using in vitro and invivo settings, we found that GARP modulates the expression ofCD103, an important molecule that is involved in homing of theT cells to the gut. As a result, deletion of GARP on Treg cellssignificantly improved the antitumor immunity against colorectalcancer. Overall, our data established that GARP plays an impor-tant role in empowering Treg cell function and promoting theiraccumulation in the colon.

Materials and MethodsAnimals

Lrrc32f/fFoxp3YFP-Creþ mice were generated in house. Lrrc32f/f

mice were obtained from Riken (15). Foxp3YFP-Cre, and Rag2�/�

mice were from Jackson Laboratory. Inducible GARP OE andglobal GARP deletion (R26-creERT2-GARP) mice were describedpreviously (19, 20). GARP OE was induced with doxycycline(50 mg/mL; Sigma) in 1% sucrose drinking water. GARP deletionin R26-creERT2-GARP cells was induced in vitro using 250 nmol/L4-Hydroxytamoxifen (4-HT). All mice were on a pure C57BL6/Jbackground. The animal procedures were conducted on anapproved protocol by the Medical University of South CarolinaInstitutional Animal Care and Use Committee.

Isolation of immune cells from tissuesThymus, spleen, mesenteric lymph nodes (mLN), and periph-

eral lymph nodes (pLN), were dissociated into a single-cellsuspension and RBC lysis buffer (Sigma) was used to remove redblood cells. For colons, tissues were dissected and incubated for30minutes at 37�Cwith collagenaseD (1mg/mL; Roche), dispase(0.05 U/mL; Worthington), and DNase I (100 mg/mL; Sigma-Aldrich). Lymphocyteswere collected from the interface of a 40%/80% Percoll gradient (GE Healthcare).

Flow cytometryFor surface staining, after Fixable ViabilityDye (Affymetrix) and

Fc-receptor blocking, cells were stained for surface markers. Anti-bodies against mouse CCR9 (eBioCW-1.2), CD4 (GK1.5), CD8a(53-6.7), CD11b (M1/70), CD44 (IM7),CD62L (MEL-14),CTLA-4 (UC10-4B9), Foxp3 (FJK-16), GARP (YGIC86), GITR (DTA-1),GL7 (GL7), and KLRG1 (2F1) were from Thermo Fisher Scientific.Those recognizing, CCR2 (48607), CD25 (PC61), Gr1 (RB6-8C5), ICOS (7E.17G9), IFNg (XMG1.2) Ki67 (SolA15), and TNF(MP6-XT22) were from BD Biosciences. B220 (RA3-6B2) anti-body was purchased from BioLegend. Antibodies against humanCD4 (RPA-T4), CD25 (M-A251), and CD103 (Ber-ACT8) werepurchased fromBDBiosciences. Antibodies against human Foxp3(236A/E7) and GARP (7B11) were purchased from eBioscienceand BioLegend, respectively.

For intracellular staining for transcription factors, Foxp3/Transcription Factor Staining Buffer Set (Affymetrix) was used.To assess the expression of the intracellular cytokines, cells werestimulated for 2 hours with 50 ng/mL PMA (phorbol 12-myr-istate-13-acetate) and 1 mg/mL ionomycin (Sigma-Aldrich) in thepresence of 5 mg/mL brefeldin A (BD Biosciences). The stainingwas then performed using BD Cytofix/Cytoperm Kit according tothe manufacturer's protocol (BD Biosciences). Samples wereanalyzed immediately on BD FACSDiva, and data analysis wasperformed using FlowJo Software (Tree Star).

In vitro suppression assayUsing MACS Treg isolation kit (Miltenyi Biotec), the na€�ve T

(CD4þCD25�) and Treg cells (CD4þCD25þ) were purified. Atotal of 1 � 105 CFSE (Thermo Fisher Scientific)-labeled WTCD4þCD25� T cells were stimulated with 2 mg/mL anti-CD3antibody (145-2C11; BD Biosciences) in the presence of 5 �104 Rag2�/� splenocytes. WT and GARP�/� Treg cells were addedin 1:1 to 1:8 Treg/Teff cell ratio. Activated CD4þCD25� T cellsonly, without Treg cells, were used as a positive control for T-cellproliferation. Three days poststimulation, CFSE dilutions ofT cells were analyzed and quantified by flow cytometry.

Generation of in vitro–inducible Treg cells and anti-GARPtreatment

CD4þCD25� T cells were isolated from 8- to 10-week-old WTor f/f creþ mice or from peripheral blood mononuclear cells ofhealthy individuals using MACS columns (Miltenyi Biotec). Atotal of 2 � 106 mouse or 4 � 105 human cells were seeded perwell in a 24- or 96-well plate, respectively, and stimulated with5 mg/mL of plate-bound anti-CD3 (145-2C11; BD Biosciencesfor mouse, and HITEa; BioLegend for human) and 2 mg/mLanti-CD28 (37.51; BD Biosciences for mouse and CD28.2;BioLegend for human) in RPMI medium supplementedwith 10% heat-inactivated FBS, 1,000 U/mL of IL2, anti-IL4(5 mg/mL), anti-IFNg (5 mg/mL), and TGFb (10 ng/mL) for 72hours. Human in vitro–induced Treg cells were treated with a poolof polyclonal mouse anti-human GARP antibody made in ourown laboratory (at 10 mg/mL each) or their isotype control for upto 10 days. As an additional control, cells were also treated with10 mg/mL of neutralizing anti-TGFb antibody. The media werereplenished with fresh antibody-containing media every 2 days.

Lupus inductionFemaleWT and f/f creþmice were given a single intraperitoneal

injection of pristane (500 mL/20 g mouse; Sigma). Mice weresacrificed after 4 months and peripheral blood, spleen, and lungwere collected for further analysis.

T-cell transfer colitisNa€�ve T cells (CD4þCD45RBhiCD25�) from WT mice were

sorted using FACSAria II (BD Biosciences). A total of 3.5 � 105

CD4þCD45RBhiCD25� T cells per mouse, alone or in combina-tion with 1.5 � 105 WT or GARP�/� Treg cells, were intraperito-neally injected intoRag2�/�mice. After T-cell reconstitution,micewere weighed weekly and monitored for signs of disease. At theend of the experiment, mice were sacrificed and their organs wereexamined histologically and by flow cytometry for evidence ofcolitis and inflammation. Standard hematoxylin and eosin-stained sections were examined and scored by an experiencedpathologist (S. Sun) in a blinded fashion. Grading was based on

Function of Treg-GARP in Tumor and Autoimmunity

www.aacrjournals.org Cancer Res; 79(6) March 15, 2019 1179




acute inflammation as absent (0), minimal (1), mild (2), mod-erate (3), or severe (4).

Colitis-associated colon cancerA total of 7–12 mice per group were given 12.5 mg/kg body

weight azoxymethane (AOM) with intraperitoneal injection onday 1. Dextran sodium sulfate (DSS) was added to drinking waterwas given at 2.5% for 5 days inweeks 2 and 5, and then at 2% for 4days on week 8. Mice were weighed and monitored weekly foroverall health. Mice were euthanized at 12 weeks. Tumors werecounted and measured. Tumor-infiltrating lymphocytes (TIL)were isolated by collagenase D (Sigma) digestion followed byHistopaque-1083 (Sigma) mediated density separation.

Generation of human GARP-Jurkat cellsHuman GARP coding sequence was inserted into the BLR

retroviral vector via BglII and EcoRI sites (Addgene). The retroviruswas generated in Platinum-A cells (Cell Biolabs) and then trans-duced into Jurkat cell. Human GARP-Jurkat cells were furtherselected by blasticidin treatment (Invivogen).

ELISAFor active and total TGFb, ELISA plates (Corning) were coated

overnight at 4�C with anti-mouse/human TGFb1 capture anti-body (BioLegend), followed by blockingwith 1%BSA in PBS for 2hours at room temperature. Samples were treated with HCl for 10minutes and neutralized with Tris/NaOH to obtain total TGFb.Both active and total TGFbwere detected using anti-mouse Biotin-TGFb antibody (BioLegend), Streptavidin-HRP (BioLegend), andTMB Substrate reagents (BD Biosciences).

For bacteria-specific IgA, a fecal supernatant from mMT micewas used as shown in ref. 20 and the protein concentration wasmeasured using Bradford assay (Bio-Rad). Each well of the ELISAplate was coated with 0.5 mg/mL mMT fecal supernatant. Serumwas serially diluted and bacteria-specific IgA was detected withanti-mouse IgA-HRP antibody (Southern Biotech).

RNA isolation and quantitative RT-PCRTotal RNA was isolated using TRIzol reagent according to the

manufacturer's instructions (Invitrogen). cDNA was synthe-sized using SuperScript III reverse transcriptase (Invitrogen)and analyzed by quantitative PCR on StepOne Plus machine(Applied Biosystems) using SYBR Green reagent (Bio Rad)and the following primers: human CD103 forward, 50-TGGGACTGGTGACAATTGTC-30; human CD103 reverse, 50-GAGTACAGGTGTGCAGCTCT-30; human b-actin forward,50-CCCTGGACTTCGAGCAA GAG-30; human b-actin reverse,50-CCAGGAAGGAAGGCTGGAAG -30.

Detection of ANANOVA Lite HEp-2–coated slides (INOVA Diagnostics) were

incubatedwith diluted sera (in PBS), followed by FITC-conjugatedgoat anti-mouse IgG or anti-mouse IgM antibody (ThermoFisher Scientific), and examined by fluorescence microscopy. Thefluorescence staining intensity was graded as 0 (negative), 1 (slightstaining), 2 (moderate staining), and 3 (bright staining).

HistologyMouse organs were fixed in 4% formalin, embedded in paraf-

fin, sectioned, and stained following standard hematoxylin andeosin staining protocol.

Statistical analysisData analysis was performed using GraphPad Prism software

(GraphPad Software, Inc.). Results are expressed as mean� SEM.P values were determined with Student t test. Two-way ANOVAwith Sidak multiple comparison test was used for multiple groupcomparisons in tumor curves and body weight graphs. P < 0.05was considered statistically significant (�, P< 0.05; ��,P <0.01; ��,P < 0.005).

ResultsTreg-specific GARP deletion leads to chronic intestinalinflammation

To directly determine the role of GARP in Treg cell–mediatedhomeostasis, we generatedmicewith specific deletion ofGARPonTreg cells by crossing mice with a conditional allele of Lrrc32withmice expressing Foxp3YFP-Cre (Lrrc32 flox/flox [f/f] x Foxp3YFP-Cremice, herein called f/f creþmice). TCR activation of CD4þCD25þ

Treg cells increased GARP levels (Supplementary Fig. S1A andS1B), in accordancewithwhat has previously reported, that GARPselectively identifies activated Treg cells (16). Young f/f creþ mice(4–6weeks old) developed normally with no apparent changes inthe systemic serum levels of active and total TGFb or any keycytokines including IL2, IL6, TNF, and IFNg (SupplementaryFig. S2A). The percentage of splenic CD4þ and CD8þ T cells waslower in f/f creþ comparedwithWTmice.No changes in frequencywere observed in the thymic single CD4þ or CD8þ T-cell popula-tions (Fig. 1A). Notably, the frequencies of CD4þFoxp3þ T cellswere elevated in both thymus and spleen of f/f creþ mice (Fig. 1Band C). CD4þ and CD8þ T cells displayed a more activatedeffector-like phenotype with an increase in the CD44þCD62L�

population in f/f creþ mice (Fig. 1D). Histologically, slightincreases in lymphocyte infiltration were detected in the liver,and the lung (Supplementary Fig. S2B). In addition, thevilli in the small intestine were slightly shorter (SupplementaryFig. S2B). Moreover, IgG antinuclear antibodies (ANA) in theserum were also increased in the KO mice (SupplementaryFig. S2C).

To further investigate whether these immune alterations pre-dispose the mice to the development of a chronic autoimmunedisease, themice were aged to one year.We observed an increasedfrequency of CD4þ T cells in both colon (not significant) andmLNs, and as well as an increased CD8þ T-cell percentage in thespleen (Fig. 1E). We also found that aged f/f creþ mice hadsignificantly higher levels of bacteria-specific IgA in the serum,suggesting increased gut permeability due to loss of gut tolerance(Fig. 1F). Histologically, we noted aberrant villi structure (thick,shortened, and fused) in small intestine of f/f creþ mice, whichreflects partial villous atrophy that is known to manifest inpatients with celiac or other inflammatory bowel diseases(Fig. 1G; ref. 21). Thus, lack of GARP on Treg cells resulted indisruption of immune homeostasis and made them incapable ofmaintaining intestinal homeostasis with aging, suggesting thatGARP might play a key role in maintaining Treg-mediatedimmune regulation.

GARP is important for optimal Treg differentiation andcontrolling lupus-mediated inflammation

Our data suggest that GARP deletion in Treg cells might affecttheir suppressive function. Using an in vitro suppression assay,we confirmed Edwards and colleagues data (15) that the lack of

Salem et al.

Cancer Res; 79(6) March 15, 2019 Cancer Research1180




Figure 1.

Deletion of GARP in Foxp3þ T cells leads to systemic inflammation. A, Percentage of CD4þ and CD8þ T cells in thymus and spleen of 6- to 8-week-oldWT and f/fcreþmice. B, Representative flow cytometry plots of CD4þFoxp3þ T cells in thymus and spleen. C,Quantification of percentages of CD4þFoxp3þ T cells in B. D,Representative flow cytometric plots and quantification of splenic CD4þ and CD8þ T cells based on CD44 and CD62L expression. E, Percentage of CD4þ andCD8þ T cells in spleen, colon, and mLNs from 1-year-oldWT and f/f creþmice (n¼ 5) was determined by flow cytometry and quantified. F, Gut bacteria–specificIgA was measured in the sera of mice by ELISA (n¼ 5–6). G, The small intestine ofWT and f/f creþmice were stained with hematoxylin and eosin. Arrows,stunted and fused villi. Statistical analyses were performed by unpaired Student t test (� , P < 0.05; �� , P < 0.01; �� , P < 0.001). Error bars, mean� SEM.






GARP on Treg cells did not alter their suppressive capacity onCD4þ T cells (Supplementary Fig. S3A). However, in vitro Tregsuppression assays do not necessarily recapitulate in vivo pro-cesses. When we tested the expression of different Treg markersgated on CD4þFoxp3þ cells, we found that CD25 (Fig. 2A),ICOS, and KLRG1 (Supplementary Fig. S3B and S3C) expres-sion were significantly lower in GARP�/� Treg cells comparedwith WT cells after 72-hour TCR stimulation. In fact, TGFbsignaling has been shown to induce the expression CD25 (22).KLRG1 defines terminally differentiated Treg cells and is linkedto IL2ra (CD25) (23). On the other hand, the expression ofGITR (24) and CTLA-4 was unchanged (Supplementary Fig.S3D and S3E). In addition, mice lacking GARP in Treg cellsshowed increased frequency of Foxp3þHeliosþ T cells (Fig. 2B),typically defined as thymic Treg cells (25), indicating a possibledysregulation in the generation of peripheral Treg cells, whichare TGFb-dependent (10). Indeed, the frequency of differenti-ated iTreg cells from primary CD4þCD25� T cells was signif-icantly less in the absence of GARP (Supplementary Fig. S3F).These data suggest that GARP may affect the function of Tregcells.

To this end and based on data showing increased systemic ANAlevel in aged f/f creþ mice (Supplementary Fig. S2C), we hypoth-esized thatmicewithGARP�/�Treg cellsmight be less suppressiveunder chronic inflammatory conditions such as systemic lupuserythematosus. Therefore, we employed a pristane-induced lupusmodel, which is a result of induction and proliferation of auto-reactive lymphocytes that produce proinflammatory cytokines,pathogenic autoantibodies, and immunocomplexes (20, 26). Weobserved that deletion of GARP in Treg cells leads to decreasedability to control the disease. Indeed, several hallmarks of auto-immunity, such as lymphocytopenia, systemic IgG, and elevatedGL7þ germinal center B cells levels were more prominent in f/fcreþ thanWTmice (Fig. 2C). In addition, immune cell infiltrationinto the lungs was more severe in f/f creþ mice (Fig. 2D). Thesedata imply that GARP is crucial for Treg cells to exert theirregulatory function during inflammation.

GARP deletion on Treg cells abrogates their ability to suppressT-cell responses during inflammation

Given that the intestinal compartment was clearly affected inmice lackingGARP on Treg cells, we hypothesized that Treg-GARP

Figure 2.

Lack of GARP affects the Tregdifferentiation and decreases theability to control the inflammationin pristane-induced Lupus model. A,Splenic CD4þCD25þ T cells fromWT and f/f creþwere TCR-activated for 72 hours. Theexpression level of CD25 wasdetermined by flow cytometry.Representative mean fluorescenceintensity (MFI) plots and dataquantifications are shown (n¼ 3).B, Flow cytometry analysis ofFoxp3 versus Helios expressionin splenic CD4þ T cells.Representative dot plots and bargraphs are presented. C, FemaleWT (n¼ 5) and f/f creþmice (n¼ 5)were given a single intraperitonealinjection of pristane. Mice weresacrificed 4 months later.Lymphocyte count was determinedby complete blood count, IgG1serum level was detected by ELISA,and GL7þ% [B220] was measuredby flow cytometry. D, Lungs ofWTand f/f creþmice were fixed in 4%paraformaldehyde, sectioned,and stained with hematoxylin andeosin. Each point represents anindividual mouse. Data represent 3–4 independent experiments.Statistical analyses were performedby unpaired Student t test(� , P < 0.05; �� , P < 0.01;�� , P < 0.001). Error bars,mean� SEM.

Salem et al.





might play an important role in managing immunopathologyduring ongoing intestinal inflammation. To address this pos-sibility, we tested their ability to suppress inflammatoryresponses in the intestine using a T-cell transfer model ofcolitis. CD4þCD45RBhiCD25� T cells from WT mice wereinjected intraperitoneally together with either WT or GARP�/�

Treg cells into RAG2�/� mice. Functional Treg cells have beenshown to reverse the disease in this mouse model (27). Asanticipated, the control group of RAG2�/� mice injected onlywith CD4þCD45RBhiCD25� T cells without Tregs (Fig. 3A)showed severe body weight loss (Fig. 3B). Interestingly,GARP�/� Treg cell cotransfer completely failed to control thedisease in the recipient mice, which showed significant weightloss (Fig. 3B). Thehistologic scorewas also indistinguishable fromcontrol mice that did not receive Treg cells (Fig. 3C). This wasassociated with increased CD4þ T cells in the colon (Fig. 3D).Notably, the frequency of GARP�/� Treg cells in the colon wassignificantly lower compared with WT Treg cells (Fig. 3E). Asexpected, this was associated with increased TNFaþ, IFNgþ CD4þ

T cells (Fig. 3F), tissue inflammatory monocytes/macrophages(CD11bþGr1int), and neutrophils (CD11bþGr1hi; Fig. 3G;ref. 28). These data clearly demonstrate that GARP is importantfor the suppressive function of Treg cells and may additionallyaffect their colonic accumulation.

GARP�/� Treg cells fail to accumulate in the colonOur data so far led us to hypothesize that GARP expression not

only shapes the function of Treg cells, but also impacts theirability to accumulate in the colon. To address this possibility, weexamined female GARP-heterozygous mice (f/f creþ/�). Giventhat Foxp3 is encoded on the X chromosome, one of whichundergoes inactivation in females, only 50% of their Treg cellsexpress GARP (Fig. 4A). The percentages of WT (YFP�) andGARP�/� (YFPþ) of CD4þFoxp3þ Treg cells in thymus, spleen,pLNs, mLNs, and colon were determined. Consistent with ourearlier finding, the ratio of YFPþ to YFP� Treg cells was lower inthe colon, whereas it was higher in pLNs (Fig. 4B and C),indicating that GARP�/� Treg cells are lacking the ability tomigrate to the colon. GARP�/� and WT Treg cells had similarproliferation rate based on Ki67þ staining, as well as survival asindicated by the expression of ICOS (Fig. 4D; ref. 29). KLRG1expression was lower on colonic GARP�/� Treg cells (Fig. 4D). Toassess the expression of markers implicated in Treg cell traffickingto the colon, we examined the expression level of chemokinereceptors (CCR) such as CCR2 (30), and CCR9 (31), which havebeen associated with colonic homing. In addition, we deter-mined CD103 expression; CD103 is a TGFb-regulated moleculenecessary for cell migration to the colon (12). No changes inCCR2 or CCR9 expression in GARP�/� Treg cells were observed(Fig. 4E); however, CD103 was significantly reduced in the KOmice (Fig. 4F). These data show that the ability of GARP�/� Tregcells to accumulate in the colon is altered, possibly due toreduced CD103-mediated homing.

GARP modulates CD103 expressionTo further assess whether GARP expression could modulate

CD103 expression, mRNA level of CD103 was determined inJurkat human T cells with GARP overexpression (OE). We foundthat CD103mRNA levels were higher inGARPOE comparedwithWT Jurkat cells (Fig. 5A). Next, we tested whether in vivoGARPOEcan also modify CD103 expression. To this end, we employed a

doxycycline-inducible GARP OE mouse model where a Tet-onelement is knocked into the GARP promoter (19, 20). After 6weeks of doxycycline treatment, the mice were analyzed andGARP expression was confirmed to be upregulated in the thymicand splenic CD4þ and CD8þ T cells (Fig. 5B). No differences wereobserved in the proportion of thymic Treg cells or their CD103expression (Fig. 5C). Notably, the frequencies of CD4þFoxp3þ

and CD103þFoxp3þ T cells were more abundant in the peripheryof the GARP OE mice, especially in the spleen and the colon(Fig. 5D–F). To confirm the modulatory role of GARP on CD103,we used another mouse model for inducible global GARP dele-tion (ROSA26-creERT2-GARP). Splenocytes were isolated fromERT2-creþ/� and GARP deletion was induced by 4-HT treatmentin vitro (Supplementary Fig. S4A). The percentage of CD103þ cellswas reduced in ERT2-creþ/� Tregs after 6 days of 4-HT treatment,following GARP deletion. (Supplementary Fig. S4B). As expected,nomodulation of either GARP or CD103was observed in controlcells after 4-HT treatment.

To validate that in human setting, in vitro–induced human Tregcells were generated from peripheral blood CD4þCD25� cellsfrom four healthy subjects and treated with a pool of polyclonalmouse anti-human GARP antibody made in our own laboratory(at 10 mg/mL each) or their isotype control for up to 10 days. As anadditional control, cells were also treated with 10 mg/mL ofneutralizing anti-TGFb antibody. The media were replenishedwith fresh antibody-containing media every 2 days. Both thepercentage of CD103þ Treg cells and their expression level ofCD103 were significantly decreased after 8 and 10 days of eitheranti-GARP or anti-TGFb antibody treatment compared with treat-ment with the isotype antibody for the same period of time(Fig. 5G). These data further support the modulatory functionof GARP on CD103 expression.

Deletion of GARP in Treg cells leads to reduced colitis-associated colon cancer development

We next investigate the impact of GARP on the colonicaccumulation of Treg cells in the context of colorectal tumor. WTand f/f creþ littermates were subjected to colitis-associated coloncancer (CAC) based on a combination of the mutagenic agentAOMandDSS (32). CACwas induced by injecting a single dose ofAOM, followed by three cycles of DSS administration (Fig. 6A).Themicewere sacrificed atweek 12 after initial AOM injection.Noweight difference was observed between WT and f/f creþ mice(Fig. 6B). While no changes were observed in splenocyte cellcounts (Fig. 6C), the cell numbers were significantly increased inthemLNs of f/f creþmice (Fig. 6D). Interestingly, micewithGARPdeletion in Treg cells developed fewer tumors and the tumor loadin these mice was significantly reduced compared with the WTmice (Fig. 6E). Notably, this was associated with a significantincrease in the percentage of CD4þ T cells, but not CD8þ T cells,within the tumor-infiltrating lymphocytes (TIL) population(Fig. 6F). Indeed, TILs from f/f creþmice had a significant decreasein the percentage of Treg cells compared with control littermates(Fig. 6G). Functional analysis of TILs also showed that bothCD4þ

and CD8þ T cells had higher TNFa expression (Fig. 6H). Alto-gether, these data are in accordance with the aforementionedresults that GARP improves the accumulation of Treg cells inthe gut.

To further confirm the association of GARP/CD103 andcolonic Treg cells, MC38 colon cancer cells were implantedsubcutaneously into the flanks of WT and f/f creþ mice. The






Figure 3.

GARP Expression is essential for Treg cell–suppressive function during intestinal inflammation. A, Experimental outline. Rag2�/�mice received 3.5� 105

CD4þCD45RBhiCD25�WT T cells (i.p) alone or in combination with 1.5� 105WT or GARP�/� Treg cells (CD4þCD25þ T cells). B, Body weight trend of theRAG2�/� recipient mice upon T-cell transfer ofWT or GARP�/� Treg cells (5–6mice in each group are displayed; mean� SEM; �, P < 0.05, two-way ANOVAwithmultiple comparison). C, Representative hematoxylin and eosin staining of colon tissue section and their histology score based on acute inflammation as absent(0), minimal (1), mild (2), moderate (3), or severe (4). D, Total cell number of colonic CD4þ T cells. E, Representative dot plots and cumulative frequency data ofWT and GARP�/� colonic CD4þFoxp3þ T cells. F, Representative dot plots and bar graphs showing percentage and number of TNFaþ and IFNgþ CD4þ T cells inthe colon. G, Number of monocytes/macrophages (CD11bþGr1int) and neutrophils (CD11bþGr1hi). Each point represents an individual mouse. The results arerepresentative of two independent experiments with 5–6mice in each group. Statistical analysis performed by unpaired Student t test (� , P < 0.05; �� , P < 0.01;�� , P < 0.001). Error bars, mean� SEM.

Salem et al.





tumor growth was similar (Supplementary Fig. S5A) with nochanges in tumor-infiltrating CD4þ, CD8þ, or CD4þFoxp3þ Tcells (Supplementary Fig. S5B and S5C). The percentageof Foxp3þCD103þ Treg cells was also comparable (Supplemen-tary Fig. S5D), consistent with the previous data that GARP ismore important for the accumulation of Treg cells specifically inthe gut.

DiscussionThe role of TGFb in the development of Treg cells and sup-

pression of various immune cell subsets including T cells, den-dritic cells, B cells, NK cells, and myeloid cells is well estab-lished (33–35). The presence of GARP specifically on Treg cellsunderlines its importance for these cells. However, it is not clear

how the absence of GARP on Treg cells modulates their functionand ability to operate under inflammatory conditions. Our studyis thefirst comprehensive research doneon the role ofGARP in thegeneration, maintenance, suppressive function, and migration ofTreg cells in various inflammatory and cancer settings. We foundthatGARP is required tomaintain systemic immune homeostasis,particularly in the intestine. GARP is also important for theoptimal generation of Treg cells from na€�ve CD4þ T cells. Weobserved that even though GARP deletion affected the expressionlevel of Foxp3 after activation, GARP�/� Treg cells were able tocontrol the proliferation of CD4þ T cells in vitro. However, theywere unable to control chronic inflammatory diseases such asinducible lupus. GARP�/� Treg cells were also unable to suppressT-cell–mediated colitis or antitumor responses in the colon. Thiswas not only due to the defect in the suppressive function, but also

Figure 4.

GARP�/� Treg cells are not able topersist within the colon. A,Experimental outline. The Treg cellsin Lrrc32f/fFoxp3YFP-Creþ/� femalemice constitute 50% ofWT Foxp3and 50% of Foxp3-YFP fusionprotein due to randomX-inactivation. B, Bar graph showsfrequency of GARP�/� relative toWT Treg cells. Value 1 indicates the1:1 frequency ratio of GARP�/� andWT Treg cells. Student unpaired ttest was used (paired two-tailedStudent t test). C, Representativeflow cytometry plots show Foxp3versus YFP staining gated onCD4þFoxp3þ Treg cells. D, Bargraphs show frequency ofWT orGARP�/� Treg cells expressing,Ki67, ICOS, and KLRG1. E,Frequencies ofWT or GARP�/�

Treg cells expressing, Ki67, ICOS,and KLRG1. F, Representativeflow cytometry plots showthe percentage of CD103þ cellsinWT and GARP�/� Treg cellsin the colon, and quantificationof Foxp3þCD103þ fromLrrc32f/fFoxp3YFP-Creþ/�mice.Statistical analysis performedby unpaired Student t testindicated (� , P < 0.05; �� , P < 0.01).






Figure 5.

GARPmodulates CD103 expression. A, Expression of CD103 was determined inWT and GARP OE Jurkat cells by qPCR. B, The expression levels of GARP inthymic and splenic CD4þ and CD8þ T cells in WT and Tet-inducible GARP-overexpressing (OE) mice were determined by flow cytometry. Representative meanfluorescence intensity (MFI) plots are shown. C–F, Percentage of Treg cells and CD103þ Treg cells in the thymus, spleen, pLNs, and colon was determined by flowcytometry and quantified.G, Graph bars show the change in the percentage of CD103þ and CD103 MFI in human Tregs treated with GARP antibodies. HumanTreg cells were generated in vitro from CD4þCD25� T cells of four different healthy individuals in the presence of anti-CD28/CD3, IL2, TGFb, anti-IL4, and anti-IFNg . After 72 hours, the media were removed and replenished with fresh T-cell media containing IL2 and either anti-GARP antibodymix or isotype antibody oranti-TGFb neutralizing antibody. The cells were harvested after 8 and 10 days of antibody treatment, stained, and analyzed by flow cytometry. Statistical analysiswas performed by unpaired Student t test (A–F) or paired t test (G; � , P < 0.05; �� , P < 0.01; �� , P < 0.001). Error bars, mean� SEM.

Salem et al.





Figure 6.

Mice with GARP deletion in Treg cells have better protective tumor immunity in a mouse colon cancer model. A, Experimental outline. Eight- to 12-week-oldWTand f/f creþmice were intraperitoneally injected with 12.5 mg/kg AOM, followed by three cycles of DSS administration. Mice were sacrificed at week 12 after AOMinjection. The results are representative of two independent experiments. B, Body weight change during the course of the experiment. C, Total cells count inspleen. D, Total cells count in mLNs. E, Total tumor count in each individual mouse and tumor load defined by the total tumor size in each mouse. F, CD4þ andCD8þ T-cell percentage in TILs gated on CD45þ cells. G, CD4þFoxp3þ T cells percentage in TILs gated on CD45þ cells. H, Bar graphs and representative flowcytometry plots show IFNg and TNFa expression in CD4þ and CD8þ T cells in TILs gated on CD45þ cells. Statistical analysis was performed by unpaired Student ttest (� , P < 0.05; �� , P < 0.01; �� , P < 0.001). Error bars, mean� SEM.






because of their defective accumulation in the colon, whichmightbe CD103-mediated.

GARP was previously shown to be expressed on activated Tregcells (16). The main function of GARP is to bind latent TGFb,indicating that GARP might play a role in tuning the function ofTreg cells. In our work, we used a transgenic mouse model inwhich GARP was specifically deleted on Foxp3 Treg cells, whichallowed us to elucidate important and previously unknownfunctions of this molecule on Treg cells. In particular, Foxp3expression in GARP�/� Treg cells was not affected under homeo-static conditions. When the cells were activated by TCR stimula-tion, we noticed that the expression levels of key markers such asCD25, ICOS, and KLRG1 were reduced, which might be associ-ated with the compromised function and stability of Treg cells.Accordingly, in chronic inflammatory disease models such as thepristine-induced lupus model, GARP�/� Treg cells were not ableto control the disease. In fact, previous studies have shown thatTreg cells from active lupus patients exhibited functional andstability abnormalities (36, 37). In parallel, the TGFb-mediatedsignaling in the cells defined by SMAD activation was conse-quentlyweaker. This signaling pathway is critically involved in theinduction of Foxp3 because deletion of both SMAD2 and SMAD3in CD4þ T cells leads to their failure to upregulate Foxp3 inresponse to TGFb (11). Another evidence of the abnormal TGFbsignaling as a consequence of GARP deletion is that in vitrogeneration of Treg cells, which is largely TGFb-dependent differ-entiation mechanism, was significantly reduced. In other words,by regulating of TGFb bioavailability, GARP might modulate thefunction of Treg cells. Although, TGFb is important for thegeneration, maintenance, and function of Treg cells, it has beenreported that cell-specific deletion of TGFb did not result in anyimmune abnormalities and the mice developed normally (38).Thus, more work is needed to clarify these controversial findings.

Another key observation from our study is that the expressionof GARP on Treg cells is a crucial factor to maintain the intestinalhomeostasis. We observed that young mice with GARP�/� Tregcells had shorter villi and increased lymphocytes. Upon aging, themice developed an intestinal inflammation mimicking celiacdisease in humans with thick, shortened, and fused villi. We alsodetected significantly increased bacterial-specific IgA in the serumof aged mice, indicating increased intestinal barrier permeabilitycompared with WT mice. This chronic disease development isinduced by TGFb signaling as also was shown in a recent study inmice lacking Tgfbr1 on Treg cells (12). In addition, aging itself hasbeen reported to decrease the function of Treg cells (39, 40).Therefore, further studies are needed todeterminehow the age canaffect the expression profile for TGFb-associated mediatorsincluding GARP on Treg cells.

A prior study (41) demonstrated that deletion of GARP onCD4þ T cells leads to a reduction Treg cells in themLNs and Peyerpatches during anoral tolerance diseasemodel. It suggests that thegeneration of Treg cells in these tissues is affected due to GARPdeletion. Our study brings to light unknown effects of GARPexpression on Treg cells. In fact, not only it is necessary for thegeneration of new Treg cells, but it is also important for theoptimal expression of CD103, which is an important factor incell recruitment to the intestine. Indeed, GARP�/� Treg cells werenot able to suppress T-cell transfer colitis because of the failure ofGARP�/� Treg cells to accumulate in the colon. Moreover, micewith specific GARP deletion in Treg cells exhibit better antitumorresponses in the AOM/DSS colon tumor model. Our finding is

well in line with previous murine and human studies showingthat depletion of Treg cells or a low density of Foxp3þ cells in thecolon is associatedwithbetter colorectal antitumor immunity (42,43). Furthermore, the role of TGFb in the induction of CD103has been previously reported in a variety of immune cells (44, 45).CD103 is not just functioning as a homing marker. Recent find-ings show that the Foxp3 expression level in Treg cells fromCD103�/� mice was significantly reduced and these Tregcells were not able to suppress murine contact hypersensitivityreactions during the elicitation phase (46). Moreover, in vitro–generated alloantigen-primed CD4þCD103þ cells coexpressedCD25. These cells suppressed T-cell activation, and containedmore Foxp3 mRNA than the CD103� CD25þ cells isolated fromthe same cultures (47). The link between GARP and CD103 istherefore remarkable. Our findings have a translational relevanceto imply GARP as a potential target for combination therapyagainst cancer to improve the current immunotherapy outcomes,especially because Treg cells constitute a major challenge inantitumor immunity. To this end, our laboratory has previouslyreported that targeting GARP potentiates protective immunityagainst bothmelanoma and colon cancer (48). In addition, it alsolimited metastasis in an orthotopic model of human breastcancer (49). Therefore, we believe that our study uncovers severalnovel aspects of GARP as a functionality tuning and traffickingenhancer by empowering TGFb signaling. It is an interestingfeature that can be utilized as a treatment strategy in immuno-therapy combinations for autoimmune diseases and cancer espe-cially after the recent findings that give clear structural andmechanistic insights on how GARP binds to TGFb on humanTreg cells (50).

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

Authors' ContributionsConception and design: Z. LiDevelopment of methodology: E. Ansa-AddoAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Wallace, E. Ansa-Addo, H. Kwon, B. Riesenberg,B. Wu, S. GugliettaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Wallace, B. Riesenberg, S. Sun, Z. LiWriting, review, and/or revision of the manuscript: C. Wallace, B. Riesenberg,S. Guglietta, S. Sun, Z. LiAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases):Study supervision: Z. LiOther (initiated, performed, and designed the experiments, analyzed thedata, and wrote the manuscript): M. SalemOther (designed and performed experiments): M. Velegraki, A. Li, A. Metelli,Y. ZhangOther (provided critical materials and scientific input): B. Liu

AcknowledgmentsWe thank the FACS core of Hollings Cancer Center, MUSC for their technical

assistance. The research was funded by multiple grants from the NIH:R01AI077283, R01CA213290, R01CA188419, and P01CA186866 (all to Z. Li.)

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received August 22, 2018; revised September 21, 2018; accepted January 17,2019; published first January 23, 2019.

Salem et al.





References1. Grant CR, Liberal R, Mieli-Vergani G, Vergani D, Longhi MS. Regulatory

T-cells in autoimmune diseases: challenges, controversies and–yet–unanswered questions. Autoimmun Rev 2015;14:105–16.

2. Fontenot JD,GavinMA,RudenskyAY. Foxp3programs thedevelopment andfunction of CD4þCD25þ regulatory T cells. Nat Immunol 2003;4:330–6.

3. Bennett CL, Christie J, Ramsdell F, Brunkow ME, Ferguson PJ, Whitesell L,et al. The immune dysregulation, polyendocrinopathy, enteropathy,X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet2001;27:20–1.

4. Saito T, Nishikawa H, Wada H, Nagano Y, Sugiyama D, Atarashi K, et al.Two FOXP3(þ)CD4(þ) T cell subpopulations distinctly control the prog-nosis of colorectal cancers. Nat Med 2016;22:679–84.

5. Chen W, Ten Dijke P. Immunoregulation by members of the TGFbsuperfamily. Nat Rev Immunol 2016;16:723–740.

6. Kulkarni AB, Huh CG, Becker D, Geiser A, Lyght M, Flanders KC, et al.Transforming growth factor beta 1 null mutation in mice causes excessiveinflammatory response and early death. Proc Natl Acad Sci U S A 1993;90:770–4.

7. Gorelik L, Flavell RA. Abrogation of TGFbeta signaling in T cells leads tospontaneous T cell differentiation and autoimmune disease. Immunity2000;12:171–81.

8. Liu Y, Zhang P, Li J, Kulkarni AB, Perruche S, ChenW. A critical function forTGF-beta signaling in the development of natural CD4þCD25þFoxp3þregulatory T cells. Nat Immunol 2008;9:632–40.

9. Marie JC, Letterio JJ, Gavin M, Rudensky AY. TGF-beta1 maintains sup-pressor function and Foxp3 expression in CD4þCD25þ regulatory T cells.J Exp Med 2005;201:1061–7.

10. Chen W, Jin W, Hardegen N, Lei KJ, Li L, Marinos N, et al. Conversion ofperipheral CD4þCD25- naive T cells to CD4þCD25þ regulatory T cells byTGF-beta induction of transcription factor Foxp3. J Exp Med 2003;198:1875–86.

11. Takimoto T,Wakabayashi Y, Sekiya T, Inoue N,Morita R, Ichiyama K, et al.Smad2 and Smad3 are redundantly essential for the TGF-beta-mediatedregulation of regulatory T plasticity and Th1 development. J Immunol2010;185:842–55.

12. Konkel JE, Zhang D, Zanvit P, Chia C, Zangarle-Murray T, Jin W, et al.Transforming growth factor-beta signaling in regulatory T cells controls THelper-17 cells and tissue-specific immune responses. Immunity 2017;46:660–674.

13. TranDQ, Andersson J,WangR, RamseyH,UnutmazD, Shevach EM.GARP(LRRC32) is essential for the surface expression of latent TGF-beta onplatelets and activated FOXP3þ regulatory T cells. Proc Natl Acad Sci U S A2009;106:13445–50.

14. Metelli A, SalemM,Wallace CH,WuBX, Li A, Li X, et al. Immunoregulatoryfunctions and the therapeutic implications of GARP-TGF-beta in inflam-mation and cancer. J Hematol Oncol 2018;11:24.

15. Edwards JP, Fujii H, Zhou AX, Creemers J, Unutmaz D, Shevach EM.Regulation of the expression of GARP/latent TGF-beta1 complexes onmouse T cells and their role in regulatory T cell and Th17 differentiation.J Immunol 2013;190:5506–15.

16. Wang R, Kozhaya L,Mercer F, KhaitanA, Fujii H,UnutmazD. Expression ofGARP selectively identifies activated human FOXP3þ regulatory T cells.Proc Natl Acad Sci U S A 2009;106:13439–44.

17. Hahn SA, Stahl HF, Becker C, Correll A, Schneider FJ, Tuettenberg A, et al.Soluble GARP has potent antiinflammatory and immunomodulatoryimpact on human CD4(þ) T cells. Blood 2013;122:1182–91.

18. Cuende J, Lienart S,DedobbeleerO, van derWoning B,De BoeckG, StockisJ, et al.Monoclonal antibodies against GARP/TGF-beta1 complexes inhibitthe immunosuppressive activity of human regulatory T cells in vivo.Sci Transl Med 2015;7:284ra56.

19. Wu BX, Li A, Lei L, Kaneko S, Wallace C, Li X, et al. Glycoprotein Arepetitions predominant (GARP) positively regulates transforming growthfactor (TGF) beta3 and is essential for mouse palatogenesis. J Biol Chem2017;292:18091–7.

20. Wallace CH, Wu BX, Salem M, Ansa-Addo EA, Metelli A, Sun S, et al. Blymphocytes confer immune tolerance via cell surface GARP-TGF-betacomplex. JCI Insight 2018 Apr 5 [Epub ahead of print].

21. Catassi C, Kryszak D, Bhatti B, Sturgeon C, Helzlsouer K, Clipp SL, et al.Natural history of celiac disease autoimmunity in a USA cohort followedsince 1974. Ann Med 2010;42:530–8.

22. Dupuy d'Angeac A, Reme T, Monier S, Gao Q, Duperray C, Jullien P, et al.Contrasting effect of transforming growth factor type beta 1 (TGF-b1) onproliferation and interleukin-2 receptor expression in activated and rapidlycycling immature (CD3-CD4-CD8-) thymocytes. J Cell Physiol 1993;154:44–52.

23. Cheng G, Yuan X, Tsai MS, Podack ER, Yu A, Malek TR. IL-2 receptorsignaling is essential for the development of Klrg1þ terminally differen-tiated T regulatory cells. J Immunol 2012;189:1780–91.

24. Ronchetti S, Ricci E, Petrillo MG, Cari L, Migliorati G, Nocentini G, et al.Glucocorticoid-induced tumour necrosis factor receptor-related protein: akey marker of functional regulatory T cells. J Immunol Res 2015;2015:171520.

25. Elkord E, Al-Ramadi BK. Helios expression in FoxP3(þ) T regulatory cells.Expert Opin Biol Ther 2012;12:1423–5.

26. Tsokos GC. Systemic lupus erythematosus. N Engl J Med 2011;365:2110–21.

27. Mottet C, Uhlig HH, Powrie F. Cutting edge: cure of colitis byCD4þCD25þ regulatory T cells. J Immunol 2003;170:3939–43.

28. Griseri T, McKenzie BS, Schiering C, Powrie F. Dysregulated hematopoieticstem and progenitor cell activity promotes interleukin-23-driven chronicintestinal inflammation. Immunity 2012;37:1116–29.

29. Redpath SA, van der Werf N, Cervera AM, MacDonald AS, Gray D, MaizelsRM, et al. ICOS controls Foxp3(þ) regulatory T-cell expansion, mainte-nance and IL-10 production during helminth infection. Eur J Immunol2013;43:705–15.

30. Johdi NA, Ait-Tahar K, Sagap I, Jamal R. Molecular signatures of humanregulatory T Cells in colorectal cancer and polyps. Front Immunol 2017;8:620.

31. Wermers JD, McNamee EN, Wurbel MA, Jedlicka P, Rivera-Nieves J.The chemokine receptor CCR9 is required for the T-cell-mediatedregulation of chronic ileitis in mice. Gastroenterology 2011;140:1526–35.

32. Neufert C, Becker C, Neurath MF. An inducible mouse model of coloncarcinogenesis for the analysis of sporadic and inflammation-driven tumorprogression. Nat Protoc 2007;2:1998–2004.

33. Tran DQ. TGF-b: the sword, the wand, and the shield of FOXP3(þ)regulatory T cells. J Mol Cell Biol 2012;4:29–37.

34. Onishi Y, Fehervari Z, Yamaguchi T, Sakaguchi S. Foxp3þ natural regula-tory T cells preferentially form aggregates on dendritic cells in vitro andactively inhibit their maturation. Proc Natl Acad Sci U S A 2008;105:10113–8.

35. Xu A, Liu Y, Chen W, Wang J, Xue Y, Huang F, et al. TGF-beta-inducedregulatory T cells directly suppress B cell responses through a noncytotoxicmechanism. J Immunol 2016;196:3631–41.

36. Lyssuk EY, Torgashina AV, Soloviev SK, Nassonov EL, Bykovskaia SN.Reduced number and function of CD4þCD25highFoxP3þ regulatory Tcells in patients with systemic lupus erythematosus. Adv Exp Med Biol2007;601:113–9.

37. Valencia X, Yarboro C, Illei G, Lipsky PE. Deficient CD4þCD25high Tregulatory cell function in patients with active systemic lupus erythema-tosus. J Immunol 2007;178:2579–88.

38. Gutcher I, Donkor MK, Ma Q, Rudensky AY, Flavell RA, Li MO. Autocrinetransforming growth factor-beta1 promotes in vivo Th17 cell differentia-tion. Immunity 2011;34:396–408.

39. Zhao L, Sun L, Wang H, Ma H, Liu G, Zhao Y. Changes ofCD4þCD25þFoxp3þ regulatory T cells in aged Balb/c mice. J Leukoc Biol2007;81:1386–94.

40. Sun L, Hurez VJ, Thibodeaux SR, Kious MJ, Liu A, Lin P, et al. Agedregulatory T cells protect from autoimmune inflammation despite reducedSTAT3 activation and decreased constraint of IL-17 producing T cells.Aging Cell 2012;11:509–19.

41. Edwards JP, Hand TW,Morais da Fonseca D, Glass DD, Belkaid Y, ShevachEM. The GARP/Latent TGF-beta1 complex on Treg cells modulates theinduction of peripherally derived Treg cells during oral tolerance. Eur JImmunol 2016;46:1480–9.

42. Pastille E, Bardini K, Fleissner D, Adamczyk A, Frede A, Wadwa M, et al.Transient ablation of regulatory T cells improves antitumor immunity incolitis-associated colon cancer. Cancer Res 2014;74:4258–69.

43. Yoon HH, Orrock JM, Foster NR, Sargent DJ, Smyrk TC, Sinicrope FA.Prognostic impact of FoxP3þ regulatory T cells in relation to CD8þ






T lymphocyte density in human colon carcinomas. PLoS One 2012;7:e42274.

44. Bain CC, Montgomery J, Scott CL, Kel JM, Girard-Madoux MJH, Martens L,et al. TGFbetaR signalling controls CD103(þ)CD11b(þ) dendritic celldevelopment in the intestine. Nat Commun 2017;8:620.

45. AnzD,MuellerW, GolicM, KunzWG, RappM, Koelzer VH, et al. CD103 isa hallmark of tumor-infiltrating regulatory T cells. Int J Cancer 2011;129:2417–26.

46. Braun A, Dewert N, Brunnert F, Schnabel V, Hardenberg JH, Richter B,et al. Integrin alphaE(CD103) is involved in regulatory T-cell functionin allergic contact hypersensitivity. J Invest Dermatol 2015;135:2982–91.

47. Allakhverdi Z, FitzpatrickD, Boisvert A, BabaN, Bouguermouh S, SarfatiM,et al. Expression of CD103 identifies human regulatory T-cell subsets.J Allergy Clin Immunol 2006;118:1342–9.

48. Rachidi S, Metelli A, Riesenberg B, Wu BX, Nelson MH, Wallace C, et al.Platelets subvert T cell immunity against cancer via GARP-TGFbeta axis.Sci Immunol 2017;2:pii: eaai7911.

49. Metelli A, Wu BX, Fugle CW, Rachidi S, Sun S, Zhang Y, et al. Surfaceexpression of TGFbeta docking receptor GARP promotes oncogenesis andimmune tolerance in breast cancer. Cancer Res 2016;76:7106–17.

50. Lienart S, Merceron R, Vanderaa C, Lambert F, Colau D, Stockis J, et al.Structural basis of latent TGF-beta1 presentation and activation by GARPon human regulatory T cells. Science 2018;362:952–56.


Salem et al.




2019;79:1178-1190. Published OnlineFirst January 23, 2019.Cancer Res Mohammad Salem, Caroline Wallace, Maria Velegraki, et al. Accumulation of Regulatory T Cells in the ColonGARP Dampens Cancer Immunity by Sustaining Function and

Updated version

10.1158/0008-5472.CAN-18-2623doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2019/01/23/0008-5472.CAN-18-2623.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/79/6/1178.full#ref-list-1

This article cites 49 articles, 18 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/79/6/1178.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/79/6/1178To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-18-2623

http://cancerres.aacrjournals.org/content/suppl/2019/01/23/0008-5472.CAN-18-2623.DC1

http://cancerres.aacrjournals.org/content/79/6/1178.full#ref-list-1

http://cancerres.aacrjournals.org/content/79/6/1178.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/79/6/1178


GARP Dampens Cancer Immunity by Sustaining Function and ... · Tumor Biology and Immunology GARP...

Documents

Transcript of GARP Dampens Cancer Immunity by Sustaining Function and ... · Tumor Biology and Immunology GARP...