Cancer Research A Galectin-3 Ligand Corrects the Impaired … · Microenvironment and Immunology A Galectin-3 Ligand Corrects the Impaired Function of Human CD4 and CD8 Tumor-Infiltrating

Micr

A GCD4Fav

NathaKris TPierre

Abst

Intro

Mosponsemulatthe prspecifnosisprogre

AuthorUniverCatholUniver4GynaeCliniqude Brude PhyGrandInstitute

Note:Resear

N. Dem

C. Schprovide

Corresfor CanBelgiumvanderb

doi: 10

©2010

Cance7476

Published OnlineFirst August 18, 2010; DOI: 10.1158/0008-5472.CAN-10-0761

Canceresearch

oenvironment and Immunology

alectin-3 Ligand Corrects the Impaired Function of Humanand CD8 Tumor-Infiltrating Lymphocytes and

R

ors Tumor Rejection in Mice

lie Demotte1, Grégoire Wieërs1, Patrick Van Der Smissen2, Muriel Moser6, Christopher Schmidt8,
hielemans3, Jean-Luc Squifflet4, Birgit Weynand5, Javier Carrasco7, Christophe Lurquin1, J. Courtoy2, and Pierre van der Bruggen1
ractHum

IFN-γmobilIndeedsecretantibodeveloImporIFN-γGCS-1

(5–7). Inss despite

s' Affiliatisité Cathoique de Losite i t Brucology anes Universitxelles, Instisiologie AnHôpital deof Medica

Supplemench Online (h

otte and G

midt, K. Thid tumor sa

ponding Acer Resear. Phone:ruggen@br

.1158/0008-

American A

r Res; 70(

Downloa

an CD8+ tumor-infiltrating T lymphocytes (TIL), in contrast with CD8+ blood cells, show impairedsecretion on ex vivo restimulation. We have attributed the impaired IFN-γ secretion to a decreasedity of T-cell receptors on trapping in a lattice of glycoproteins clustered by extracellular galectin-3., we have previously shown that treatment with N-acetyllactosamine, a galectin ligand, restored thision. We strengthened this hypothesis here by showing that CD8+ TIL treated with an anti–galectin-3dy had an increased IFN-γ secretion. Moreover, we found that GCS-100, a polysaccharide in clinicalpment, detached galectin-3 from TIL and boosted cytotoxicity and secretion of different cytokines.tantly, we observed that not only CD8+ TIL but also CD4+ TIL treated with GCS-100 secreted moreon ex vivo restimulation. In tumor-bearing mice vaccinated with a tumor antigen, injections of00 led to tumor rejection in half of the mice, whereas all control mice died. In nonvaccinated mice,00 had no effect by itself. These results suggest that a combination of galectin-3 ligands and therapeutic
GCS-1
vaccination may induce more tumor regressions in cancer patients than vaccination alone. Cancer Res; 70(19);7476–88. ©2010 AACR.

cytescomeresistafunctiDes

humaT-cell

duction

st cancer patients mount a spontaneous T-cell re-against their tumor, and antitumor T cells can accu-

e at metastatic sites (1–4). In several tumor types,esence of T-cell infiltrates, which may contain tumor-ic lymphocytes, seems to correlate with better prog-

patients with advanced cancer, tumorsthe presence of tumor-infiltrating lympho-

(10, 11TIL arcarcinassay,(12). Clymphified Mshort-cytomTILs wexpreszyme(5, 14–We

receptCTL cN-acefunctisecretface Tthe TC

ons: 1Ludwig Institute for Cancer Research andlique de Louvain, de Duve Institute, 2Universitéuvain, de Duve Institute, Cell Biology Unit, 3Vrijessel , Laboratory of Physiology-Immunology,d 5Pathology, Université Catholique de Louvain,aires Saint-Luc, Brussels, Belgium; 6Université Libretut de Biologie et de Médecine Moléculaire, Serviceimale, 7Department of Oncology and Hematology,Charleroi, Charleroi, Belgium; and 8Queensland

l Research, Brisbane, Queensland, Australia

tary data for this article are available at Cancerttp://cancerres.aacrjournals.org/).

. Wieërs are joint first authors.

elemans, J-L. Squifflet, B. Weynand, and J. Carrascomples.

uthor: Pierre van der Bruggen, Ludwig Institutech, 74 av. Hippocrate, UCL7459, B-1200 Brussels,32-2764-7431; Fax: 32-2762-9405; E-mail: pierre.u.licr.org.

5472.CAN-10-0761

ssociation for Cancer Research.

19) October 1, 2010

Research. on August 1, 202cancerres.aacrjournals.org ded from

(TIL). This suggests that the antitumoral T cells be-ineffective either because tumor cells have becoment to the immune attack or because TILs have becomeonally impaired (8, 9).pite this potentially adverse tumoral environment,n TIL can easily be expanded in vitro and the resultinglines and clones can lyse autologous tumor cells). However, reports on the ex vivo function of humane scarce. CD8+ T lymphocytes isolated from renal cellomas displayed no cytotoxicity in a redirected killingin contrast with blood CD8+ T cells of healthy donorsD8+ T cells, which were isolated from a metastaticnode of a melanoma patient and recognized the mod-elan-AMART-1 peptide, failed to produce IFN-γ on

term restimulation with peptide, as opposed to anti-egalovirus CD8+ T cells from the same tumors (13).ere also shown to differ from blood T cells for genesion patterns and for several markers such as gran-B and perforin, suggesting an impaired function17).previously described a physical dissociation of T-cellors (TCR) from coreceptors CD8 on human TIL andlones with impaired function (18). A galectin ligand,tyllactosamine (LacNAc), restored the ability of non-onal T cells to secrete cytokines, and this increasedion was associated with an increased proximity of sur-
CRs with the coreceptor CD8. We therefore attributedR-CD8 dissociation observed on nonfunctional CTL
1. © 2010 American Association for Cancer

http://cancerres.aacrjournals.org/

clonesa lattica lecttypestumorgalectSev

have bides elist oftheticand gbrancrefs. 2ber ofmodifextracical trWe

LacNAand Ctherap

Mate

Cells,Pati

bloodreviewascitesat −80modifted wL-arginstreptmicinserumwas frciatedpreser3 hourLiberalatedor solPoCASthaweHS, 6I (Sigmcells bouslymideby a pBiotecsamplples inin Fig.tion kor CD

recomfromunmovidedtics). Alaboranot ahttp://

FunctPol

bottommediuand BCD3/Ced aftBiosoupotPL

readetoxicitrivedwere land inti-CD3+ T celmeasuassay,cove'smous75,000inson(1:100also adbeled(PerCPallophinson)malde(Bectomated

DetacCell

Fc bloanti–gline M2 houGCS-1DickinPBS w(Therroomand iantibogalectat 4°C

Galectin-3 Ligand Corrects the Impaired Function of TIL

www.a


to a reduced mobility of the TCRs, which are trapped ine of glycoproteins clustered by extracellular galectin-3,in abundantly secreted by macrophages and variousof tumor cells (19, 20). Our data suggested that intra-al galectin-3 could impair T-cell function and thatin-3 ligands could improve antitumor immunity in vivo.eral galectin ligands, in particular galectin-3 ligands,een identified in the past decade, such as polysacchar-xtracted from plants or fungi (e.g., ref. 21; a completereferences is available as Supplementary Data). Syn-galectin-3 ligands were obtained by decorating lactosealactose cores with various moieties or by creatinghed structures such as glycodendrimers (reviewed in2, 23). A few of them were reported to reduce the num-tumor metastases in mice (24–29). Only GCS-100, aied citrus pectin, and DAVANAT, a galactomannanted from guar beans, have previously been used in clin-ials (30–32).here compared the effect of GCS-100 with that ofc on the cytotoxicity and cytokine secretion by CD8+

D4+ human TIL on ex vivo stimulation, and tested theeutic potential of GCS-100 in a murine tumor model.

rials and Methods

TILs, and reagentsent-derived solid tumor samples, tumor ascites, andcells were collected after approval by institutionalboards of all collaborating institutions. Cells fromwere concentrated by centrifugation and either frozen°C or resuspended in culture medium made of Iscove'sied Dulbecco's medium (Life Technologies) supplemen-ith 0.24 mmol/L of L-asparagine, 0.55 mmol/L ofine, 1.5 mmol/L of L-glutamine (AAG), 100 μg/mL ofomycin, 100 units/mL of penicillin, 30 μg/mL of genta-(culture medium; Sigma-Aldrich), and 2% human(HS) or ascitic fluid (ascites without cells). Ascitic fluidozen at −20°C. Solid tumors were mechanically disso-and either passed through a 0.1-mm sieve before cryo-vation in 10% DMSO/3.6% HS or incubated for 2 tos in culture medium supplemented with 0.14 units/mLse Blendzymes (Roche). Mononuclear cells were iso-from the blood of hemochromatosis patients, ascites,id tumor samples using Lymphoprep (Axis-Shield) and usually frozen at −80°C or −196°C. Cells wered in culture medium supplemented with 2% to 10%IU/mL of interleukin (IL)-2, and 5 units/mL of DNasea-Aldrich). T cells were isolated from mononuclear

y rosetting with sheep erythrocytes (BioMérieux) previ-treated with 2(2-aminoethyl) isothiourea dihydrobro-(Sigma-Aldrich). CD8+ and CD4+ T cells were isolatedositive selection strategy (AutoMACS System, Miltenyi) using magnetic beads coated with anti-CD8. Fores LB2990, LB2974, and LB2912 in Fig. 1A, all the sam-Fig. 1B, sample LB2974 in Fig. 1C, and sample LB29912, a depletion strategy using the CD8 untouched isola-
it (Miltenyi Biotec) was used. The purity of the CD4+
8+ T-cell samples was between 90% and 95%. Humanat 4°Cbodie

acrjournals.org

Research. on August 1, 202cancerres.aacrjournals.org Downloaded from

binant IL-2 was from Chiron Healthcare SAS, IL-7 wasR&D Systems, LacNAc was from Carbosynth, anddified citrus pectin was from Sigma. GCS-100 was pro-by Prospect Therapeutics, Inc. (now Solana Therapeu-ll the cell lines were obtained several years ago in ourtory and described previously, but the cell lines wereuthenticated according to guidelines described incancerres.aacrjournals.org/site/misc/ifora.xhtml.

ional assaysyclonal stimulations were performed in round-ed microwells with 10,000 T cells in 200 μL of culturem, containing 0–6 IU/mL of IL-2, with 3,000 (ELISAioplex) or 10,000 (ELISpot) beads coated with anti-D28 antibodies (Dynabeads, Invitrogen). IFN-γ secret-er overnight coculture was measured by ELISA usingrce Cytoset reagents (Invitrogen) or by ELISpot (ELIS-

US kit ALP; Mabtech) using the AID-ELISpot classicr software 5.0 (Autoimmun Diagnostika). In the cyto-y assay, the target cells were murine P1.azar cells de-from mastocytoma cell line P815 (33). Target cellsabeled for 1 hour with 50 μCi of Na[51Cr]O4, washed,cubated for 15 minutes at room temperature with an-monoclonal antibody (OKT3, 1 μg/mL; Mabtech). CD8ls were added to the targets, and chromium release wasred after 4 hours of incubation. For the degranulationCD8+ T cells (75,000) were stimulated for 5 hours in Is-modified Dulbecco's medium 2% HS AAG with 150,000e cells (P815) incubated with anti-CD3 antibody orCD3/CD28 beads. Brefeldin-A (GolgiPlug, Becton Dick-) and FITC-coupled anti-CD107a and anti-CD107b), or the control isotype (IgG1, Becton Dickinson), wereded. After 5 hours of stimulation, cells were washed, la-for 15 minutes at 4°C with peridinin chlorophyll protein)-coupled anti-CD3 (1:40, SK1; Becton Dickinson) andycocyanin-coupled anti-CD8 (1:40, RPA-T8; Becton Dick-antibodies, washed, and fixed in PBS and 1% parafor-hyde (PFA). Cells were analyzed on a FACSCaliburn Dickinson). The percentage of CD107+ cells was esti-for the CD3+CD8+ T cells.

hment of galectinss from ascites were treated for 15 minutes at 4°C withcking reagent (Miltenyi Biotec) before treatment withalectin-1 or anti–galectin-3 antibodies. Melanoma cellZ2-MEL.43 and cells from ascites were incubated forrs at 37°C under agitation with LacNAc (5 mmol/L),00 (25 or 5 μg/mL), or B2C10 (mouse IgG1; Bectonson); fixed for 5 minutes at room temperature inith 0.5 mg/mL dithio-bis-succinimidyl propionatemo Scientific); washed; incubated for 20 minutes attemperature with 25 mmol/L of glycine; washed again;ncubated with 5 μg/mL of either anti–galectin-3dy M3/38 (rat IgG2a; BioLegend) or polyclonal anti–in-1 rabbit IgG (Abcam). Cells were further incubatedfor 15 minutes, washed, and incubated for 15 minutes
with relevant anti-immunoglobulin secondary anti-
s coupled to Alexa Fluor 488 (Invitrogen), together

Cancer Res; 70(19) October 1, 2010 7477



FigurewhethebreastTILs weGCS-10superna

Demotte et al.

Cance7478


1. GCS-100 boosted IFN-γ secretion by CD8+ and CD4+ TIL. Columns, mean of triplicate measurements; bars, SD. Nonstimulated TIL or PBL,r treated or not, secreted <15 pg/mL of IFN-γ. A, CD8+ TILs were isolated from ascites (ovary, colon, pancreas, and melanoma VUB88),carcinoma pleural effusion, and solid melanomas. B, CD8+ PBLs were isolated from blood samples from hemochromatosis patients. C, CD4+

re isolated from ascites or solid melanomas. CD8+ and CD4+ T cells were treated for 2 h (bold) or 20 h with 5 mmol/L of LacNAc or 5 μg/mL of
0, and then 10,000 cells were cultured overnight in 200 μL of culture medium with 3,000 CD3/CD28 beads. The accumulation of IFN-γ in thetant was measured by ELISA. D, the number of IFN-γ–secreting cells among 10,000 cells was estimated by ELISpot.
r Res; 70(19) October 1, 2010 Cancer Research

Research. on August 1, 2021. © 2010 American Association for Cancercancerres.aacrjournals.org Downloaded from


Figure2 to 5 mbeadsascitic(5 mmowith La


www.a


2. Efficacy of GCS-100 treatment in different culture conditions. A, CD8+ TILs were isolated from ascites and treated for 2 h (bold) or 20 h withmol/L of LacNAc or 5 μg/mL of GCS-100 in culture medium containing 20% to 25% ascitic fluid, and 10,000 cells were stimulated with

as described in Fig. 1. B, frozen cells collected from ascites were thawed and incubated for 2 h in culture medium and then resuspended influid or in culture medium containing 2% HS throughout the following procedures. Cells were incubated for 2 to 20 h in the presence of LacNAc

+
l/L), GCS-100 (5 μg/mL), or medium. CD8 cells were isolated by positive selection, resuspended at 10,000 cells per round-bottomed microwellcNAc or GCS-100, and stimulated as described in Fig. 1.
Cancer Res; 70(19) October 1, 2010acrjournals.org 7479



with aanti–Cfinal wand anweighmolecMiller

FluorA d

The fomouseIgG2a;eBioscInvitrobodies

MiceP81

ted witoethacontaithe firthe firantigeP815 (

Resu

GCS-1CD8

solidnomaent hiGCS-1overnand ation bwas uisolateGCS-1tures,culturof GCGCS-1T cellincreacludedT cellon stimin parIn o

CD4+

samplGCS-1CD3/Ction wWe co

correcbut alTo

perforsamplovernithe adnumbthe TIon theTo

effectisupprinhibiic thesFirst,tin ligted wIFN-γ(Fig. 2testedCD8+,furthewerefluid.LacNAbeadsThe reIFN-γcarcin(Fig. 2

GCS-1In a

were acity, aGCS-1and tuTablesecretretionT cellsWe

targetOnly tcytotoTIL wbloodbut sisuredcentagexpresic gralizedgranucells w

Demotte et al.

Cance7480


nti–CD3-PerCP (1:40, SK7; Becton Dickinson) andD8-Alexa Fluor 647 (1:40, 3B5; Invitrogen). After aashing step, cells were fixed with 2% PFA in PBSalyzed on a FACSCalibur. We attributed a moleculart of 120 kDa (average) to GCS-100, according to theular weight described for another modified pectin byand colleagues (34).

escence resonance energy transfer microscopyetailed protocol is available as Supplementary Data.llowing antibodies were used: anti-TCRβ (1:40, IP-26,IgG1; eBioscience), anti-CD8α (1:40, UCHT-4, mouseSigma-Aldrich), anti-CD4 (1:40, RPA-T4, mouse IgG1;ience), anti–IgG2a-Alexa Fluor 488 (1:400, donor, green;gen), and anti–IgG1-Alexa Fluor 568 secondary anti-(1:400, acceptor, red; Invitrogen).

vaccination and GCS-1005 clone 3 cells were expanded in DMEM supplemen-th AAG, 10% FCS, glucose, and 75 μmol/L of mercap-nol (33). The recombinant adenovirus constructns the coding sequence of a fusion protein combiningst 81 amino acids of the mouse invariant chain withst 83 amino acids of P1A, a protein containing annic peptide, which is a tumor rejection antigen of35).

lts

00 boosts IFN-γ secretion by CD8+ and CD4+ TILs+ TILs were isolated to >90% purity from 18 tumors: 5melanomas and 1 melanoma ascites, 1 breast carci-pleural effusion, and 11 carcinoma ascites from differ-stologic origins. The lymphocytes were incubated with00 or LacNAc for 2 to 20 hours and then stimulatedight by the addition of beads coated with anti-CD3nti-CD28 antibodies (CD3/CD28 beads). IFN-γ secre-y nontreated TIL isolated from four of five solid tumorsndetectable, whereas secretion by nontreated TILd from ascites was more variable. The presence of00 boosted IFN-γ secretion by >3-fold in 13 of 18 cul-and the average increase in IFN-γ secretion for alles was 4-fold (Fig. 1A). Treatment with LacNAc insteadS-100 had a similar effect (Fig. 1A). When the same00 or LacNAc treatment was applied to blood CD8+

s isolated from patients with hemochromatosis, these in IFN-γ secretion was marginal (Fig. 1B). We con-that CD8+ TILs, compared with normal blood CD8+

s, are often impaired in their ability to secrete IFN-γulation. However, this defect can be reverted, at least

t, with the addition of GCS-100 or LacNAc.ur previous report (18), CD4+ TILs were not analyzed.TILs were therefore isolated from 13 of the 18 TILes shown in Fig. 1. They were also incubated with00 or LacNAc and then stimulated by the addition ofD28 beads. In the presence of GCS-100, IFN-γ secre-
as increased by >3-fold in 6 of 13 samples (Fig. 1C).nclude that the impaired secretion of IFN-γ can be
treatetreate

r Res; 70(19) October 1, 2010


ted by GCS-100 and LacNAc not only for CD8+ TILso for CD4+ TIL.estimate the number of IFN-γ–secreting T cells, wemed ELISpot experiments with a few TIL and bloodes. Isolated CD8+ and CD4+ T cells were incubatedght with GCS-100 or LacNAc and then stimulated bydition of CD3/CD28 beads. After sugar treatment, theer of IFN-γ–secreting cells increased by a factor of 2 forL, whereas the sugar treatment had a marginal effectblood CD8 T cells (Fig. 1D).

increase TIL function in patients, GCS-100 has to beve in a tumor environment, despite the presence ofessive factors found in ascitic fluid and cells that couldt TIL, such as tumor cells or regulatory T cells. To mim-e situations, two types of experiments were conducted.nine CD8+ TIL populations were incubated with galec-ands as above and stimulated in medium supplemen-ith 20% to 25% of cell-free carcinomatous ascitic fluid.secretion increased >3-fold in five of nine samplesA). Second, the effect of LacNAc and GCS-100 wason either CD2+ cells (including natural killer cells,and CD4+ T cells) or on the total cell population,r containing macrophages and tumor cells. The cellsresuspended either in culture medium or in asciticCD8+ TILs were purified after a first GCS-100 orc treatment and then stimulated with CD3/CD28in the presence of LacNAc or GCS-100 (Fig. 2B).sults suggest that both LacNAc and GCS-100 can boostsecretion by CD8+ TIL even in the presence of eitheromatous ascitic fluid or other cells present in ascitesB).

00 boosts other functions of CD8+ and CD4+ TILsddition to IFN-γ secretion, several other TIL functionslso tested (i.e., secretion of other cytokines, cytotoxi-nd degranulation). In comparison with untreated cells,00– and LacNAc-treated cells also secreted more IL-2mor necrosis factor-α, but no IL-4 (SupplementaryS1). With some TIL samples, we observed slightly moreion of IL-10 and IL-17 after treatment. Cytokine sec-was similar for treated and untreated blood CD8+

.estimated the cytotoxicity of CD8+ TIL using mousecells coated with an anti-CD3 antibody (Fig. 3A).he CD8+ TIL treated with GCS-100 had a significantxicity, whereas the cytotoxicity of untreated CD8+

as minimal. The cytotoxicity of CD8+ peripherallymphocytes (PBL) was higher than that of TIL,milar for treated and untreated cells. We also mea-degranulation, a prerequisite for cytolysis. The per-es of degranulating cells were identified by surfacesion of CD107a and CD107b, which reside in cytolyt-nule membranes within the cytoplasm and are mobi-to the cell surface following activation-inducedle exocytosis (Fig. 3B). The percentage of CD107+

as much higher in treated TIL compared with un-
d TIL, whereas this percentage was equivalent ford and untreated blood CD8+ cells. Finally, we
Cancer Research



measuzymea potea simboostidue tomenta

Thenomafor atof GCat a c

Figurecytotoxwere isovarianLB3059carcinomurineanti-CDand incindicatereleaseB, CD8stimulamurineincubatwith an75,000FITC-coand anand bresimultaAfter 5labeledpercentestimated for CD3+CD8+ T cells.


www.a


red the intracellular content of perforin and gran-B as phenotypic markers of cells that might havential to kill. Both treated and untreated TILs haveilar content of the two enzymes, indicating that theng effect of GCS-100 and LacNAc treatment was not
an effect on perforin or granzyme content (Supple-
ry Fig. S1).5 μg/mby TIL

acrjournals.org


toxicity of GCS-100 was evaluated with human mela-cells and T-cell clones. Their proliferation was measuredleast 4 weeks in the presence of various concentrationsS-100. Growth inhibition was observed starting at day 7oncentration of 125 μg/mL (Supplementary Fig. S2). At

3. GCS-100 boosted theicity of CD8+ TIL. CD8+ TILsolated from ascites of(LB2991, LB3015, and) and pancreatic (LB3033)mas. A, target cells wereP815 cells, coated with an3 antibody, 51Cr labeled,ubated with T cells at thed E:T ratios. Chromiumwas measured after 4 h.+ T cells (75,000) wereted either with 150,000P815 cells previouslyed (or not) for 15 minti-CD3 antibody or withCD3/CD28 beads.upled anti-CD107a

ti-CD107b antibodiesfeldin-A were addedneously to the stimulators.h of stimulation, cells werefor CD3 and CD8, and theage of CD107+ cells was

L, which was efficient in augmenting cytokine secretion, GCS-100 had no effect on proliferation.




An anby stiOur

CD8+

can beglycoping inmobiliationtionalafter LLacNAeffectwhichpresumthis pactivitulatedtibody(Fig. 4extentOth

tecteda roleto detthe readditicompagalectboth gof GCface g(159 μmatedmodifshownas lowgalectthat hIncu

25 μg/decreativelyGCS-1LacNAB2C10failedanti–gboostanti–gwhilegalectnot ex

GCS-1corecWe

cence

FigureIFN-γ streatedan isottreatmedescrib(5 μg/mThey wisotypeCD3+C

Demotte et al.

Cance7482


ti–galectin-3 antibody promotes IFN-γ secretionmulated CD8+ TILsworking hypothesis for impaired IFN-γ secretion byTIL on ex vivo stimulation is that galectin-3, whichsecreted by tumor cells and macrophages, binds to

roteins and oligomerizes on the T-cell surface, result-galectin–glycoprotein lattices that decrease TCRty (36). This hypothesis was also based on the dissoci-of TCR from CD8 coreceptors observed on nonfunc-CTL clones and the increased TCR/CD8 colocalizationacNAc treatment (18). However, because GCS-100 andc could compete with several galectins, we tested theof an anti–galectin-3 monoclonal antibody, B2C10,binds to the NH2-terminal domain of galectin-3 andably affects the oligomerization of the lectin: blocking

rocess should therefore decrease the T-cell inhibitoryy of galectin-3 (37). CD8+ TILs were incubated and stim-in the presence of either antibody B2C10, a control an-as negative control, or GCS-100 as a positive controlA). B2C10 boosted IFN-γ secretion to a comparableas GCS-100.er galectins (i.e., galectin-1 and galectin-9) were de-in human and/or murine tumors and thought to playin antitumor immunity (reviewed in ref. 19). We failedect galectin-8, galectin-9, or MGL (a lectin implicated ingulation of T-cell function) on TIL (38). However, inon to galectin-3, we detected galectin-1 (Fig. 5). Wered GCS-100 and LacNAc for their ability to detachin-3 and galectin-1 from melanoma cells that carryalectins (Supplementary Fig. S3). The concentrationsS-100 that were able to detach 50% (EC50) of the sur-alectin-3 and galectin-1 were estimated at 1.3 μmol/Lg/mL) and 2 μmol/L, respectively. EC50s were esti-for LacNAc at 9 and 18 μmol/L, respectively. Differentied citrus pectins and synthetic ligands were previouslyto detach galectin-3 in a solid-phase assay with EC50

as 1 mmol/L (22, 39–42). The most effective syntheticin-3 ligand seems to be a thiodigalactoside derivativeas a Kd of 33 nmol/L (39).bation of TIL for 2 hours with 5 mmol/L of LacNAc,mL of GCS-100, or 5 μg/mL of GCS-100 resulted in ase of galectin-3 staining by 67%, 40%, and 24%, respec-(Fig. 5). Staining of galectin-1 did not decrease after00 treatment, whereas it decreased by 38% afterc treatmen. Treatment with anti–galectin-3 antibodydetached galectin-3 but not galectin-1 (Fig. 4B). We

to detach galectin-1 by treating TIL with three differentalectin-1 antibodies, and these treatments did notIFN-γ secretion either (data not shown). Because thealectin-3 antibody seemed unable to detach galectin-1boosting TIL function, we conclude that detachingin-3 from TIL is sufficient to restore function whilecluding the contribution of other galectins.

00 treatment promotes colocalization of TCR andeptors CD8 or CD4
have previously used a flow cytometry–based fluores-resonance energy transfer approach to show a loss of
B2C10due toB2C10

r Res; 70(19) October 1, 2010


4. An anti–galectin-3 antibody can detach galectin-3 and boostecretion by TIL. A, CD8+ TILs were isolated from ascites andfor 2 h with either 5 μg/mL of anti–galectin-3 antibody B2C10,ype-matched control antibody, or 5 μg/mL of GCS-100. Afternt, 10,000 cells were stimulated with CD3/CD28 beads ased in Fig. 1. B, carcinoma ascites were treated for 2 h with B2C10L) before being labeled with anti-CD3 and anti-CD8 antibodies.ere also labeled with anti–galectin-1, anti–galectin-3 (M3/38), or-matched control antibodies. Cells shown in the figures areD8+ cells. Data are representative of three experiments. Aftertreatment, the decreased galectin-3 staining did not seem to be
competition between B2C10 and M3/38 antibodies becausewas detached from cells together with galectin-3.
Cancer Research



colocacompathe efbothrescen“accepwith Gbeledfluoropled twaveleto thefull accreasigionsincreatwo Cin eacWh

100, nceptorzationemisswith Gcreasesignifithoughthe reprevio(18, 43For

was altor phof TIL

with LLB302100 ortion wex vivo

Vacciwith GWe

bearin2 × 10were vtigen Pvaccinwith eand 50a weeable inwhichhad nvival,vaccineffect(Fig. 6

Discu

Ourtheirdisaccmericreport

Figureor medreprese


www.a


lization of TCR and CD8 at the surface of CD8+ TILred with blood CD8+ T cells (18). We examined herefect of GCS-100 or LacNAc on this colocalization inCD8+ and CD4+ TIL using a microscopy-based fluo-ce resonance energy transfer approach known astor photobleaching.” TILs were treated for 2 hoursCS-100 or LacNAc, attached to coverslips, and la-with an anti-TCRβ antibody coupled to an acceptorchrome and anti-CD8α or anti-CD4 antibodies cou-o a donor fluorochrome. Upon excitation at donorngth, energy could be transferred from the donoracceptor if closer than ∼10 nm. In these conditions,ceptor bleaching abrogates the energy loss, thus in-ng donor emission. For each analyzed cell, three re-of the membrane were bleached and donor emissionses were measured. Data are shown in Table 1 forD8+ and two CD4+ TIL samples and for two cellsh condition.en CD8+ TILs were not treated with LacNAc or GCS-o increase in donor emission was detected upon ac-photobleaching, indicating poor TCR-CD8 colocali-(Table 1). For TIL VUB147, increases in donor

ion were 20% and 14% after incubation for 2 hoursCS-100 and LacNAc, respectively, thus indicating in-d TCR-CD8 colocalization after treatment. Lower butcant increases were observed with TIL LB3022. Al-modest at first sight, these increases are well within

ported range in the literature and similar to ourusly reported value (26%) with a functional CTL clone, 44).CD4+ TILs not treated with LacNAc or GCS-100, thereso no increase in donor emission detected upon accep-

ium and washed before being labeled with anti-CD3, anti-CD8, and anti–gntative of three experiments.

otobleaching (Table 1). After 2 hours of treatmentVUB129, increases were 11% with GCS-100 and 11%

in bookines,

acrjournals.org


acNAc. Comparable values were obtained with TIL2. For both CD4+ and CD8+ TILs treated with GCS-LacNAc, increased TCR/CD8 or TCR/CD4 colocaliza-as paralleled by increased secretion of IFN-γ uponrestimulation (Table 1).

nated tumor-bearing mice that were also treatedCS-100 reject their tumorstested the therapeutic effect of GCS-100 in tumor-g mice. Forty mice were injected s.c. in the flank with6 P815 mastocytoma cells. On day 4, half of the animalsaccinated with an adenovirus encoding P815 tumor an-1A (Fig. 6A; ref. 35). On day 10, at which time all mice,ated or not, bore a tumor of ∼70 mm2, treatmentsither PBS or GCS-100 (200 μg s.c. close to the tumor0 μg i.p.) were initiated. Mice were injected three timesk. Three weeks later, the tumor had become undetect-6 of the 10 vaccinated mice treated with GCS-100, of5 were still alive after another 3 months. Vaccine aloneo effect on tumor growth, but slightly increased sur-with a median life span 10 days longer than in non-ated mice. In nonvaccinated mice, GCS-100 had noby itself, either on tumor size or on animal survivalB and C).

ssion

previous observations that human CD8 TIL increasedsecretion of IFN-γ after ex vivo treatment withharide LacNAc prompted us to test GCS-100, a poly-galectin ligand approved for clinical trials. The resultsed here indicate that GCS-100 is as efficient as LacNAc

-3 antibodies. Cells shown are CD3+CD8+ cells. Data are
5. Galectin-3 can be detached from TIL with GCS-100. Carcinoma ascites from LB3081 were treated for 2 h at 37°C with GCS-100, LacNAc,
alectin

sting TIL to be cytotoxic and to secrete different cyto-not only CD8+ but also CD4+ TIL. This effect was obtained




at coninduceprompWh

Our wby anexpreschangsurfacthe reharbobettersolid t

with smobilTCR aand thSev

treatmdetacincreaand LThird,both I

Tab

123

123

NOTweror 5beleantiandwasincrGCS0.1Lac

Demotte et al.

Cance7484


centrations (5 μg/mL) much lower than those used toapoptosis of tumor cells (>100 μg/mL), the activity thatted its administration to cancer patients (30).y do galectin-3 ligands improve human TIL function?orking hypothesis is that TILs have been stimulatedtigen recently and that activation could modify thesion of enzymes of the N-glycosylation pathway ande the structure of N-glycans exposed at the celle, as shown for murine T cells (45). We surmise thatcently activated TILs, compared with resting T cells,r a set of glycans that are either more numerous or
ligands for galectin-3. Galectin-3 is abundant in manyumors and carcinomatous ascites and can form lattices
tors Clizatio

le 1. C D

GC La

cr do is

C l 1 l 2 l 1 l 2 l 1 l 2 l

5 03 6 30

GC La

cr do is

C l 1 l 2 l 1 l 2 l 1 l 2 l

30 0

NAc-treated TIL. Secretion of IFN-γ was analyzed as described in Fig.

r Res; 70(19) October 1, 2010


urface glycoproteins that would thereby reduce TCRity. This could explain the absence of colocalization ofnd CD4/CD8 coreceptors, the impaired function of TIL,e boosting effect of galectin ligands.eral of our new results support this hypothesis. First,ent with an anti–galectin-3 antibody that was able toh galectin-3 from TIL is as efficient as GCS-100 insing IFN-γ secretion by TIL. Second, both GCS-100acNAc can detach galectin-3 from the surface of cells.treatment of TIL with GCS-100 or LacNAc increasedFN-γ secretion and colocalization of TCR with corecep-
D8 or CD4. In this respect, the increased TCR-CD4 coloca-n after GCS-100 or LacNAc treatment suggests a common
4
olocalization of TCR and coreceptors CD8 or C
s

1

s

1

1

1

on TIL

GC

l 2 ll 1

71

GC

l 2 ll 1

3 74

.

. © 2010 American

La

ll 2 ll 1

47

La

ll 2 ll 1

5

Cance

Association for Ca

TIL CD8 VUB147 TIL CD8 LB3022

/
S-100 cNAc / S-100 cNAc
ll 2

0

ll 2

6

r Researc

ncer

el
Cel Cel Cel
In

Cel

ease in

Cel

nor em

Cel

ion (%)

Cel
Ce Ce Ce Ce
−
0 2 22 8 14 −1 3 9 2 2 −1 5 1
3 1
13 932 15
1520 −4

−6 10

126 1

1414
−1 9 7 3
−3
20
Secretio

14

of IFN-γ by TIL po

−1

lation (pg/mL)

8
15
n pu

0 219 219
5 344 366
TIL CD4 VUB129 TIL CD4 LB3022

/
S-100 cNAc / S-100 cNAc
el
Cel Cel Cel
In

Cel

ease in

Cel

nor em

Cel

ion (%)

Cel
Ce Ce Ce Ce
1
1 8 11 5 11 2 4 0 1 2
1 1
9 74 10
1520 −

−7 3−1 1

9 710 7

109
0 1 1 7
1
11
Secretio

11

of IFN-γ by TIL po

−1

lation (pg/mL)

17
7
n pu

0 102 153 12 215 219

E: CD8+ or CD4+ TILs were isolated from ascites of carcinoma patients, frozen, and thawed 2 h before the experiment. Cellse subsequently incubated for 2 h at 37°C with medium containing 10% ascitic supernatant in the presence of 2 mmol/L LacNAcμg/mL GCS-100. After washing, some cells were tested for IFN-γ secretion, and the others were attached to coverslips, la-d on ice with anti-TCR antibodies (acceptor) and anti-CD8 or anti-CD4 antibodies (donor), fixed, and stained with secondarybodies. For each of the two cells that were analyzed, in three regions of interest, five images were recorded before bleachingfive others after bleaching of the acceptor fluorochrome. Three control regions were not bleached. Increase in donor emissioncalculated as follows: [1 − (donor intensity before bleaching/donor intensity after bleaching)] × 100. For CD8+ TIL, mean

ease in donor emission (in percentage) was −2.1 ± 1.6 for nontreated TIL, 14.5 ± 2.2 (P < 0.001, Mann-Whitney test) for-100–treated TIL, and 14.4 ± 1.5 (P < 0.001) for LacNAc-treated TIL. For CD4+ TIL, mean increase in donor emission was± 0.7 for nontreated TIL, 13.8 ± 2.6 (P < 0.001, Mann-Whitney test) for GCS-100–treated TIL, and 9.3 ± 1.3 (P < 0.001) for

h


Figurevaccinaantigensurvivetumorsmice wwith 2 ×P815 chalf ofwith anby i.d.each eastudy, mto 62, 7a weeks.c. in tand 50survivawas pegroup;C, tumoproducshortest tumor lengths.


www.aacrjournals.org

cancerres.aaDownloaded from


6. Tumor-bearing miceted against a tumorand treated with GCS-100d longer and rejected. A, forty 7-wk-old DBA2ere injected s.c. in the flank106 cells of mastocytoma

lone 3. Four days later,the mice were vaccinatedadenovirus encoding P1Ainjections, half a dose inr. In a double-blindedice received, from days 1000 μg GCS-100 three times: 200 μg were injectedhe periphery of the tumor0 μg i.p. B, Kaplan-Meierl curves. The experimentrformed once (n = 10 perP < 0.001, log-rank test).r size indicated as thet of the longest and the


Research. on August 1, 2021. © 2010 American Association for Cancercrjournals.org


mechaTIL, atThi

groupmice dzymethreshthe firlooserteringcolleagto secra trantherefOth

in humin antposurekine rFOXPdendrT lymimmun(51). Oto impshow t1 on Tbut noTIL. Bgalectof galestimulImp

binaticells.mentcreaserespecgrowtremaiferentent madd uThe

cytes otrialswith dmelanrespontumorthe Tfollowrespontumorin thefactorthem.the va

rectedthe fewnationand suThe

ligandin vacindicaencodresponin 57%ineffemice.9

of a S4 of 1and threportferredtion wtumortion orejectily affeThe

that Gon muGalecmetabthe prapoptculartin inadheslial cepresenmentdid nomice,cells ation apendeIt m

cally,Howeical tricer pamyelolymphapies,days. Streatmlogic t xicity. Skin rash was observed on two patients thatresolv with cessation of treatment or steroids.

Demotte et al.

Cance7486


nism behind impaired function of both CD4+ and CD8+

least for those cells responding to galectin ligands.s hypothesis agrees well with data from two others. Demetriou and colleagues observed that T cells fromeficient in N-acetylglucosaminyl-transferase V, an en-of the N-glycosylation pathway, have a lower activationold than T cells from wild-type mice. The authors werest to propose that reduced N-glycosylation, resulting inglycoprotein/galectin lattices, could improve TCR clus-at the immune synapse (46). In addition, Kuball andues (47) observed an increased frequency of cells ableete cytokines upon stimulation, if they were expressingsfected TCR lacking a glycosylation site. They proposedore that N-glycosylation modifies TCR avidity.er galectins (i.e., galectin-1 and galectin-9) were detectedan and/or murine tumors and thought to play a role

itumor immunity (reviewed in ref. 19). In humans, ex-to galectin-1 induced a shift from Th1 to Th2 cyto-

esponse, promoted the expansion of CD4+CD25high

3+ T cells, and increased the tolerogenic potential ofitic cells (41, 48–50). Galectin-1 was reported to inducephocyte apoptosis (48). In mice, tumor growth ande dysfunction correlated with the presence of galectin-1n the contrary, injections of galectin-9 were reportedrove the survival of tumor-bearing mice (52). Here, wehat, in addition to galectin-3, we also detected galectin-IL. At the concentration used for treating TIL, LacNAct GCS-100 was able to detach galectin-1 from treatedecause the anti–galectin-3 antibody did not detachin-1 from TIL, we conclude that the mere detachmentctin-3 could result in a boosted secretion of IFN-γ byated TIL.aired TIL function most probably results from a com-on of mechanisms, all ultimately depending on tumorIt is therefore remarkable that a few hours of treat-with GCS-100 or LacNAc was sufficient to strongly in-IFN-γ secretion by stimulated TIL. However, thetive contribution of other galectins, transformingh factor-β, IL-10, PD-1/PD-L1, or regulatory T cells,ns important to clarify in order to define whether dif-lymphocyte populations are impaired through differ-echanisms and whether several inhibitory pathwaysp in the same cells.identification of antigens recognized by T lympho-n human tumor cells has resulted in numerous clinicalinvolving the vaccination of tumor-bearing patientsefined tumor antigens. Among vaccinated metastaticoma patients, ∼5% show a complete or partial clinicalse, whereas an additional 10% show some evidence ofregression without clear clinical benefit. Analyses of-cell responses of melanoma patients suggest theing scenario (1, 53). Spontaneous tumor-specific T-cellses are frequent. Although they could control earlys, they eventually fail to do so and become impairedir function, most probably due to immunosuppressives present in the tumor. Galectin-3 could be one of
Thus, the tumors of the patients about to receiveccine already contain functionally impaired T cells di- 9 C. Uyt
r Res; 70(19) October 1, 2010


against tumor antigens. It is therefore possible that, inpatients who show tumor regression following vacci-

, some T cells generated by the vaccine reach the tumorcceed in reversing the local immunosuppression.relevance of combining vaccination and galectin

s is supported by the results obtained with GCS-100cinated mice bearing P815 tumors. Previous resultsted that prophylactic vaccination with an adenovirusing P815 tumor antigen P1A induced an anti-P1A T-cellse that conferred protection against tumor challengeof mice (54). However, therapeutic vaccination was

ctive at inducing tumor rejection in tumor-bearingInjection in mice bearing a subcutaneous P815 tumoremliki Forest virus encoding P1A was therapeutic for0 animals, provided that tumors were small (≤25 mm2)at intratumoral injections were used (55). In the present, injection of another P1A adenoviral construct con-only a mild survival advantage. However, in combina-ith GCS-100 injections, half of the mice rejected theirand appeared free of disease on day 130. The vaccina-f tumor-bearing mice with P1A was essential for tumoron, and we are tempted to believe that GCS-100 positive-cted the anti-P1A T cells.se experiments, however, do not provide any evidenceCS-100 exerts its therapeutic effects in mice by actingrine T cells or by disrupting galectin-glycan lattices.tins were shown to play pleiotropic roles in tumorolism. Myeloma cell growth was inhibited in vitro inesence of 500 μg/mL of GCS-100, which also triggeredosis and decreased cell migration in response to vas-endothelial growth factor (30, 56). Modified citrus pec-hibited the homoagglutination of melanoma cells,ion of tumor cells to extracellular matrix and endothe-lls, as well as the formation of capillary tubes in thece of galectin-3 (reviewed in ref. 57). Although treat-of nonvaccinated, tumor-bearing mice with GCS-100t provide a survival benefit compared with controlGCS-100 could have increased the apoptosis of tumornd inhibited neoangiogenesis and metastasis forma-nd therefore participated in the therapeutic effect inde-ntly of an effect on the immune system.ight seem risky to inject a galectin-3 ligand systemi-considering the pleiotropic role of galectin-3 (58, 59).ver, no life-threatening side effect was observed in a clin-al with GCS-100. GCS-100 has been administered to can-tients based on its potential to induce apoptosis ofma cells in vitro (30). Twenty-four patients with chronicocytic leukemia, who had relapsed after one or two ther-received 160 mg/m2 of GCS-100 i.v. for 5 days every 21ix of them experienced a partial clinical response. Theent was generally well tolerated with minimal hemato-
oed
tenhove and G. Warnier, personal communication.

Cancer Research



Takof canimpairthat vtion otumor

Disclo

No p

Ackn

WeparticuA. Cornand B.suggest

E. Marbaix for help in statistical analysis; N. Krack for editorialassistance; and L. Vanbiervliet, S. Ottaviani, C. Wildmann, V. Ha Thi,and T.

Grant

FondtifiqueProgram

Refe1. Lu

anpa20

2. Capager35

3. Roinfipa167

4. Caa ppa

5. Gationou

6. Cleinfineo

7. Zhrec200

8. Gaby

9. Matumtion

10. Yalymrien

11. Zopapri

12. VacyttumEx

13. Zipmafun

14. GoChsio


www.a


en together, our observations suggest that treatmentcer patients with galectin ligands could correct aned function of TIL. As observed in mice, it is possibleaccination combined with local and/or systemic injec-f a galectin ligand would be more effective at producingregressions than vaccination alone.

sure of Potential Conflicts of Interestfamily.rechercNationawas funnale, In(Belgium

Theof pageaccorda

otential conflicts of interest were disclosed.

owledgments

thank Prospect Therapeutics (now Solana Therapeutics), and inlar F. Tao, for providing us with GCS-100; D. Latinne and
et for ELISpot experiments; D. Godelaine, P. Coulie, T. Boon,Van den Eynde for critical reading; C. Uyttenhove for helpful
the recombinant adenovirus;Rece

OnlineF

rgun G, Holderried TA, Zahrieh D, Neuberg D, Gribben JG.ronic lymphocytic leukemia cells induce changes in gene expres-n of CD4 and CD8 T cells. J Clin Invest 2005;115:1797–805.

15. Hapropaoth

16. Luexpa sCa

17. Yeinfsub

18. Deofinf

19. Liusio

20. Liu3,ph

21. Satioch

22. Raamtosen

23. Pie20

24. Huglubo20

25. Nacamo

26. Ogme2:3

27. Gliitoap

28. GlicaRe

29. An

acrjournals.org


Lac for their assistance.

Support

ation contre le Cancer (Belgium), Fonds de la Recherche Scien-Médicale (Belgium), European Community Sixth Frameworkme, LIFESCIHEALTH-5-IP (contract number 518234), and DeloriP. van der Bruggen is a “fellow du fonds Allard-Janssen pour lahe sur le cancer.” G. Wieërs is supported by a grant from the Fondsl de la Recherche Scientifique (Belgium). The cell imaging platformded by FNRS, Région Bruxelloise, Région Wallonne, Loterie Natio-teruniversity Attraction Poles, and Université Catholique de Louvain).

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

ived 03/05/2010; revised 07/21/2010; accepted 07/25/2010; publishedirst 08/18/2010.
ions; G. Warnier for providing
rencesrquin C, Lethé B, Corbière V, et al. Contrasting frequencies ofti-tumor and anti-vaccine T cells in metastases of a melanomatient vaccinated with a MAGE tumor antigen. J Exp Med 2005;1:249–57.rrasco J, Van Pel A, Neyns B, et al. Vaccination of a melanomatient with mature dendritic cells pulsed with MAGE-3 peptides trig-s the activity of nonvaccine anti-tumor cells. J Immunol 2008;180:85–93.senberg SA, Packard BS, Aebersold PM, et al. Use of tumor-ltrating lymphocytes and interleukin-2 in the immunotherapy oftients with metastatic melanoma. N Engl J Med 1988;319:6–80.rcelain G, Rouas-Freiss N, Zorn E, et al. In situ T-cell responses inrimary regressive melanoma and subsequent metastases: a com-rative analysis. Int J Cancer 1997;72:241–7.lon J, Costes A, Sanchez-Cabo F, et al. Type, density, and loca-of immune cells within human colorectal tumors predict clinical

tcome. Science 2006;313:1960–4.mente CG, Mihm MC, Bufalino R, et al. Prognostic value of tumorltrating lymphocytes in the vertical growth phase of primary cuta-us melanoma. Cancer 1996;77:1303–10.ang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells,urrence, and survival in epithelial ovarian cancer. N Engl J Med3;348:203–13.jewski TF, Meng Y, Blank C, et al. Immune resistance orchestratedthe tumor microenvironment. Immunol Rev 2006;213:131–45.rincola F, Jaffee EM, Hicklin DJ, Ferrone S. Escape of human solidors from T-cell recognition: molecular mechanisms and func-al significance. Adv Immunol 2000;74:181–273.nnelli JR, Hyatt C, McConnell S, et al. Growth of tumor-infiltratingphocytes from human solid cancers: summary of a 5-year expe-ce. Int J Cancer 1996;65:413–21.rn E, Hercend T. A MAGE-6-encoded peptide is recognized by ex-nded lymphocytes infiltrating a spontaneously regressing humanmary melanoma lesion. Eur J Immunol 1999;29:602–7.n den Hove LE, Van Gool SW, Van Poppel H, et al. Phenotype,okine production and cytolytic capacity of fresh (uncultured)our-infiltrating T lymphocytes in human renal cell carcinoma. Clin

p Immunol 1997;109:501–9.pelius A, Batard P, Rubio-Godoy V, et al. Effector function of hu-n tumor-specific CD8 T cells in melanoma lesions: a state of localctional tolerance. Cancer Res 2004;64:2865–73.

rlin H, Kuna TV, Peterson AC, Meng Y, Gajewski TF. Tumorgression despite massive influx of activated CD8(+) T cells in atient with malignant melanoma ascites. Cancer Immunol Immun-er 2006;55:1185–97.scher U, Filgueira L, Juretic A, et al. The pattern of cytokine generession in freshly excised human metastatic melanoma suggeststate of reversible anergy of tumor-infiltrating lymphocytes. Int Jncer 1994;57:612–9.SW, Wang Y, Valmori D, et al. Ex-vivo analysis of CD8+ T cellsiltrating colorectal tumors identifies a major effector-memoryset with low perforin content. J Clin Immunol 2006;26:447–56.motte N, Stroobant V, Courtoy PJ, et al. Restoring the associationthe T cell receptor with CD8 reverses anergy in human tumor-iltrating lymphocytes. Immunity 2008;28:414–24.FT, Rabinovich GA. Galectins as modulators of tumour progres-n. Nat Rev Cancer 2005;5:29–41.FT, Hsu DK, Zuberi RI, et al. Expression and function of galectin-a β-galactoside-binding lectin, in human monocytes and macro-ages. Am J Pathol 1995;147:1016–28.thisha UV, Jayaram S, Harish Nayaka MA, Dharmesh SM. Inhibi-n of galectin-3 mediated cellular interactions by pectic polysac-arides from dietary sources. Glycoconj J 2007;24:497–507.binovich GA, Cumashi A, Bianco GA, et al. Synthetic lactuloseines: novel class of anticancer agents that induce tumor-cell apop-is and inhibit galectin-mediated homotypic cell aggregation anddothelial cell morphogenesis. Glycobiology 2006;16:210–20.ters RJ. Inhibition and detection of galectins. Chembiochem06;7:721–8.likova K, Benson V, Svoboda J, Sima P, Fiserova A. N-acetyl-D-cosamine-coated polyamidoamine dendrimer modulates anti-dy formation via natural killer cell activation. Int Immunopharmacol09;9:792–9.ngia-Makker P, Hogan V, Honjo Y, et al. Inhibition of humanncer cell growth and metastasis in nude mice by oral intake ofdified citrus pectin. J Natl Cancer Inst 2002;94:1854–62.uchi H, Toyokuni T, Dean B, et al. Effect of lactose derivatives ontastatic potential of B16 melanoma cells. Cancer Commun 1990;11–6.nsky VV, Kiriakova G, Glinskii OV, et al. Synthetic galectin-3 inhib-r increases metastatic cancer cell sensitivity to taxol-inducedoptosis in vitro and in vivo. Neoplasia 2009;11:901–9.nsky GV, Price JE, Glinsky VV, et al. Inhibition of human breast
ncer metastasis in nude mice by synthetic glycoamines. Cancers 1996;56:5319–24.dré S, Unverzagt C, Kojima S, et al. Determination of modulation



ofN-gin v

30. ChpedeCa

31. Coa fich27

32. Grga[ab

33. UyP8rat

34. Milhycoga421

35. Vacodno

36. Ahpeero10

37. Liuofdo

38. vanofC-t

39. Cugagainte

40. Anaslacbio

41. Iurgaan

42. Gliin aby

43. Antratio

44. EldtocresSo

45. Coangly

46. Dereggly

47. KuTCthe

48. StidisJ I

49. IlarerenoNa

50. Mo1ap30

51. Rugacepri

52. Nadevia

53. GeT ctio

54. NäboALproJ I

55. CoImP820

56. Strance

57. Glion

58. Du

Demotte et al.

Cance7488


l igand properties of synthetic complex-type biantennarylycans by introduction of bisecting GlcNAc in silico, in vitro andivo. Eur J Biochem 2004;271:118–34.auhan D, Li G, Podar K, et al. A novel carbohydrate-based thera-utic GCS-100 overcomes bortezomib resistance and enhancesxamethasone-induced apoptosis in multiple myeloma cells.ncer Res 2005;65:8350–8.tter F, Smith DA, Boyd TE, et al. Single-agent activity of GCS-100,rst-in-class galectin-3 antagonist, in elderly patients with relapsedronic lymphocytic leukemia [abstract 7006]. J Clin Oncol 2009;:15s.ous JJ, Redfern CH, Mahadevan D, Schindler J. GCS-100, alectin-3 antagonist, in refractory solid tumors: a phase I studystract 13023]. J Clin Oncol 2006;24:18s.ttenhove C, Maryanski J, Boon T. Escape of mouse mastocytoma15 after nearly complete rejection is due to antigen-loss variantsher than immunosuppression. J Exp Med 1983;157:1040–52.ler MC, Nesmelova IV, Platt D, Klyosov A, Mayo KH. The carbo-drate-binding domain on galectin-1 is more extensive for amplex glycan than for simple saccharides: implications forlectin-glycan interactions at the cell surface. Biochem J 2009;:211–21.

n den Eynde B, Lethé B, Van Pel A, De Plaen E, Boon T. The geneing for amajor tumor rejection antigen of tumor P815 is identical to thermal gene of syngeneic DBA/2mice. J ExpMed 1991;173:1373–84.mad N, Gabius HJ, Andre S, et al. Galectin-3 precipitates as antamer with synthetic multivalent carbohydrates and forms het-geneous cross-linked complexes. J Biol Chem 2004;279:841–7.FT, Hsu DK, Zuberi RI, et al. Modulation of functional propertiesgalectin-3 by monoclonal antibodies binding to the non-lectinmains. Biochemistry 1996;35:6073–9.Vliet SJ, Gringhuis SI, Geijtenbeek TB, van Kooyk Y. Regulation

effector T cells by antigen-presenting cells via interaction of theype lectin MGL with CD45. Nat Immunol 2006;7:1200–8.mpstey I, Sundin A, Leffler H, Nilsson UJ. C2-symmetrical thiodi-lactoside bis-benzamido derivatives as high-affinity inhibitors oflectin-3: efficient lectin inhibition through double arginine-areneractions. Angew Chem Int Ed Engl 2005;44:5110–2.dré S, Pieters RJ, Vrasidas I, et al. Wedgelike glycodendrimersinhibitors of binding of mammalian galectins to glycoproteins,tose maxiclusters, and cell surface glycoconjugates. Chem-chem 2001;2:822–30.isci I, Cumashi A, Sherman AA, et al. Synthetic inhibitors oflectin-1 and -3 selectively modulate homotypic cell aggregationd tumor cell apoptosis. Anticancer Res 2009;29:403–10.nsky GV, Mossine VV, Price JE, et al. Inhibition of colony formationgarose of metastatic human breast carcinoma and melanoma cellssynthetic glycoamine analogs. Clin ExpMetastasis 1996;14:253–67.
ikovsky M, Dale L, Ferguson S, Petersen N. Resonance energynsfer in cells: a new look at fixation effect and receptor aggrega-n on cell membrane. Biophys J 2008;95:1349–59.
Bio59. Hs

Im

r Res; 70(19) October 1, 2010


er AD, Domin A, Kaminski Schierle GS, et al. A quantitative pro-ol for dynamic measurements of protein interactions by Försteronance energy transfer-sensitized fluorescence emission. J Rc Interface 2009;6:S59–81.melli EM, Sutton-Smith M, Yan Q, et al. Activation of murine CD4+

d CD8+ T lymphocytes leads to dramatic remodeling of N-linkedcans. J Immunol 2006;177:2431–40.metriou M, Granovsky M, Quaggin S, Dennis JW. Negativeulation of T-cell activation and autoimmunity by Mgat5 N-cosylation. Nature 2001;409:733–9.ball J, Hauptrock B, Malina V, et al. Increasing functional avidity ofR-redirected T cells by removing defined N-glycosylation sites inTCR constant domain. J Exp Med 2009;206:463–775.

llman BN, Hsu DK, Pang M, et al. Galectin-3 and galectin-1 bindtinct cell surface glycoprotein receptors to induce T cell death.mmunol 2006;176:778–89.regui JM, Croci DO, Bianco GA, et al. Tolerogenic signals deliv-d by dendritic cells to T cells through a galectin-1-driven immu-regulatory circuit involving interleukin 27 and interleukin 10.t Immunol 2009;10:981–91.tran CC, Molinder KM, Liu SD, Poirier F, Miceli MC. Galectin-functions as a Th2 cytokine that selectively induces Th1optosis and promotes Th2 function. Eur J Immunol 2008;38:15–27.binstein N, Alvarez M, Zwirner NW, et al. Targeted inhibition oflectin-1 gene expression in tumor cells results in heightened Tll-mediated rejection: a potential mechanism of tumor-immunevilege. Cancer Cell 2004;5:241–51.gahara K, Arikawa T, Oomizu S, et al. Galectin-9 increases Tim-3+

ndritic cells and CD8+ T cells and enhances antitumor immunitygalectin-9-Tim-3 interactions. J Immunol 2008;181:7660–9.rmeau C, Ma W, Schiavetti F, et al. High frequency of anti-tumorells in the blood of melanoma patients before and after vaccina-n with tumor antigens. J Exp Med 2005;201:241–8.slund TI, Uyttenhove C, Nordstrom EK, et al. Comparative prime-ost vaccinations using Semliki Forest virus, adenovirus, andVAC vectors demonstrate differences in the generation of atective central memory CTL response against the P815 tumor.mmunol 2007;178:6761–9.lmenero P, Chen M, Castanos-Velez E, Liljestrom P, Jondal M.munotherapy with recombinant SFV-replicons expressing the15A tumor antigen or IL-12 induces tumor regression. Int J Cancer02;98:554–60.eetly MJ, Maharaj L, Joel S, et al. GCS-100, a novel galectin-3tagonist, modulates MCL-1, NOXA, cell cycle to induce myelomall death. Blood 2010;115:3939–48.nsky VV, Raz A. Modified citrus pectin anti-metastatic properties:e bullet, multiple targets. Carbohydr Res 2009;344:1788–91.mic J, Dabelic S, Flogel M. Galectin-3: an open-ended story.chim Biophys Acta 2006;1760:616–35.u DK, Chen HY, Liu FT. Galectin-3 regulates T-cell functions.
munol Rev 2009;230:114–27.
Cancer Research



2010;70:7476-7488. Published OnlineFirst August 18, 2010.Cancer Res Nathalie Demotte, Grégoire Wieërs, Patrick Van Der Smissen, et al. Tumor Rejection in MiceCD4 and CD8 Tumor-Infiltrating Lymphocytes and Favors A Galectin-3 Ligand Corrects the Impaired Function of Human

Updated version

10.1158/0008-5472.CAN-10-0761doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2010/08/18/0008-5472.CAN-10-0761.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/70/19/7476.full#ref-list-1

This article cites 59 articles, 18 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/70/19/7476.full#related-urls

This article has been cited by 18 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] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/70/19/7476To 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-10-0761

http://cancerres.aacrjournals.org/content/suppl/2010/08/18/0008-5472.CAN-10-0761.DC1

http://cancerres.aacrjournals.org/content/70/19/7476.full#ref-list-1

http://cancerres.aacrjournals.org/content/70/19/7476.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/70/19/7476


Documents

Cancer Research A Galectin-3 Ligand Corrects the Impaired … · Microenvironment and Immunology A Galectin-3 Ligand Corrects the Impaired Function of Human CD4 and CD8 Tumor-Infiltrating