Galectin-3 Shapes Antitumor Immune Responsesby Suppressing CD8þ T Cells via LAG-3 andInhibiting Expansion of Plasmacytoid DendriticCellsTheodore Kouo, Lanqing Huang, Alexandra B. Pucsek, Minwei Cao, Sara Solt,Todd Armstrong, and Elizabeth Jaffee
Abstract
Galectin-3 is a 31-kDa lectin that modulates T-cell responsesthrough several mechanisms, including apoptosis, T-cell receptor(TCR) cross-linking, and TCR downregulation. We found thatpatients with pancreatic ductal adenocarcinoma (PDA) whoresponded to a granulocyte-macrophage colony-stimulating fac-tor–secreting allogeneic PDA vaccine developed neutralizing anti-bodies to galectin-3 after immunization. We show that galectin-3binds activated antigen-committed CD8þ T cells only in thetumor microenvironment. Galectin-3–deficient mice exhibit
improved CD8þ T-cell effector function and increased expressionof several inflammatory genes. Galectin-3 binds to LAG-3, andLAG-3 expression is necessary for galectin-3–mediated suppres-sion of CD8þ T cells in vitro. Lastly, galectin-3–deficientmice haveelevated levels of circulating plasmacytoid dendritic cells, whichare superior to conventional dendritic cells in activating CD8þ Tcells. Thus, inhibiting galectin-3 in conjunctionwithCD8þT-cell–directed immunotherapies should enhance the tumor-specificimmune response. Cancer Immunol Res; 3(4); 412–23. �2015 AACR.
IntroductionWepreviously reported the clinical outcomes of a phase II study
testing an allogeneic granulocyte-macrophage colony-stimulatingfactor (GM-CSF)–secreting pancreatic ductal adenocarcinoma(PDA) vaccine in 60 patients with resected PDA. Twelve of the60 (20%) vaccinated patients demonstrated a greater than 3-yeardisease-free survival (DFS) associated with induction of T-cellresponses against the PDA-associated antigen, mesothelin (1). Inthe present study, we used the enriched vaccine-induced antibodyresponses to probe the vaccine cell lines for PDA-associatedantigens that are targets of the humoral immune response withthe goal of identifying proteins important to PDA development.Using this approach, we identified 11 new PDA-associated pro-teins for which specific antibody titers were elevated in thepostvaccination sera (2).
One of the proteins that we identified was galectin-3, a31-kDa lectin that is unique among the galectin family mem-bers due to the presence of both a carbohydrate recognitiondomain and an oligomerization domain that enables galectin-3to cross-link its binding targets. Galectin-3 has been shown
in vitro to possess several immunomodulatory functions, suchas reducing the affinity of the T-cell receptor (TCR) for itscognate MHC I–peptide ligand by sequestering the TCR from itsCD8þ coreceptor (3), inducing apoptosis (4), and internaliza-tion of the TCR (5). Galectin-3 also influences the strength ofantigen activation in dendritic cells (DC; refs. 6, 7). Thus, wesought to develop a mouse tumor model that would allow us toevaluate in vivo the role of galectin-3 in shaping the antitumorresponse in a tolerogenic setting.
We previously used the HER-2/neu transgenic (neu-N) mousemodel of mammary tumors to identify mechanisms of tumortolerance (8). In this study, we crossed the neu-N mice, whichdevelop neu-expressing mammary tumors, and high-avidityCD8þ TCR transgenic mice specific for the immunodominantepitope for neu onto galectin-3 knock out (KO) mice, and ana-lyzed CD8þ T-cell responses following treatment with a neu-targeted vaccine. This approach enabled us to examine genome-wide changes in protein expression caused by galectin-3. Here, wedemonstrate that in vivo depletion of galectin-3 increases both thenumber of functional CD8þ T cells found in the tumor microen-vironment (TME) and the expression of inflammatory proteins bythese T cells, leading to enhanced tumor rejection in galectin-3 KOmice when compared with galectin-3 wild-type (WT) mice. Fur-ther, we demonstrate that the effects of galectin-3 extend beyondmodulation of T-cell function to include expansion of plasmacy-toid DCs (pDC), which we show to be more potent activators ofCD8þ T cells than conventional DCs (cDC).
Materials and MethodsELISA
Costar 3690 96-well half-area EIA/RIA plates (Corning) werecoated at 4�C overnight with purified recombinant proteins at 5
Department of Oncology, Johns Hopkins University School of Medi-cine, Baltimore, Maryland.
Note: Supplementary data for this article are available at Cancer ImmunologyResearch Online (http://cancerimmunolres.aacrjournals.org/).
Corresponding Authors: Theodore Kouo, Sidney Kimmel Cancer Center atJohns Hopkins University School of Medicine, 1650 Orleans Street, CRB 1,4M06, Baltimore, MD 21287. Phone: 404-309-4286; Fax: 410-614-8216; E-mail:[email protected]; or Elizabeth Jaffee, [email protected]
mg/mL in bicarbonate/carbonate coating buffer. The protein-coat-ed plates were incubated with ELISA Blocker Blocking Buffer(Pierce Biotech) for 1 hour at room temperature. The wells werethen incubated with serial dilutions (1:100, 1:200, 1:400, and1:800) of sera for 2 hours at room temperature and with1:200,000 dilution of goat anti-human IgG (g-chain specific)peroxidase conjugate (Sigma; A8419) for 1 hour at room tem-perature. The wells were washed extensively with TBS-T betweenincubations. 3,305,50-tetramethylbenzidine liquid substrate (Sig-ma; T0440) was added to the wells and incubated in the dark for20 minutes at room temperature. The color development wasstopped by 1 N sulfuric acid. Absorbance at 450 nm (with areference wavelength of 570 nm) was measured on a PowerWave340 microplate reader (BioTek).
(galectin-3 KO) mice were purchased from The Jackson Labora-tories, bred, and housed in the Johns Hopkins animal facility.High-avidity neu-specific TCR transgenic mice were generated aspreviously described, and avidity was confirmed by tetramerstaining (9). Galectin-3 KO mice were backcrossed for 6 genera-tions using a marker-assisted selection (i.e., "speed congenic")approach. Mouse genomes were assessed at the DartMouse SpeedCongenic Core Facility at Dartmouth Medical School. Geneticbackground at the final backcross generation was determined tobe 99.75% for the desired FVB/N background. Backcrossed galec-tin-3 KOmice were bred with neu-N and high-avidity neu-specificTCR transgenicmice to generate galectin-3 KO neu-N and galectin-3 KO high-avidity neu-specific TCR transgenic mouse lines. Allexperiments were conducted with female mice between 6 to 12weeks of age according to protocols approved by the JohnsHopkins Animal Care and Use Committee.
Cell lines and mediaTheNT2.5 cell line is a neu-expressing tumor cell line generated
from spontaneously arising tumors in female neu-N mice aspreviously reported (10, 11). The 3T3neuGM vaccine line isgenetically modified from 3T3 fibroblast cells to secrete GM-CSFand express rat neu (12). The T2-Dq cell line is a TAP-2–deficientcell line that expresses the Dq MHC I allele, and was generated aspreviously reported (12). All cell lines are cultured and main-tained as previously reported (8).
Tumor, vaccine, chemotherapy, and adoptive transferFor tumor clearance studies, mice were tumor challenged with
the minimal tumorigenic dose of 5 � 104 NT2.5 tumor cellsinjected s.c. in the right upper mammary fat pad on day 0, given100 mg/kg cyclophosphamide i.p. on day 2, vaccinated with 3simultaneous s.c. injections of 1 � 106 3T3neuGM cells in thebottom and right limbs on day 3, and given 2 � 106 adoptivelytransferred high-avidity neu-specific CD8þ T cells on day 4. Fortumor-infiltrating lymphocytes (TIL) and lymph node experi-ments, mice were given 2 simultaneous s.c. injections of 2 �106 NT2.5 tumor cells in the right and left upper mammary fatpads on day 0, cyclophosphamide on day 8, vaccine on day 9, and6 � 106 adoptively transferred CD8þ T cells on day 10. High-avidity neu-specific Thy1.2þCD8þ T cells were negatively isolatedfrom spleens of female TCR transgenic mice and adoptivelytransferred as previously described (8).
Peptides and antibodiesRNEU420–429 (PDSLRDLSVF) and the negative control pep-
tide LCMV NP118–126 (RPQASGVYM) were produced in theJohns Hopkins Biosynthesis and Sequence Facility at a purity>95%. Antibodies used for flow cytometry studies were asfollows: anti–CD8-FITC (BD Biosciences), anti–CD8-PE (BDBiosciences), anti–CD8-PerCP (BD Biosciences), anti–CD8-APC (BD Biosciences), anti–galectin-3-AF647 (Biolegend),anti–galectin-3-PE (R&D), anti–PD1-PE (eBioscience), anti–LAG-3-PE (eBioscience), anti–Thy1.2-PerCP (Biolegend),anti–CD44-Pacific Blue (eBioscience), anti–CD11b-PE (BDBiosciences), anti–CD11c-FITC (BD Biosciences), anti–B220-APC (BD Biosciences), anti–Ly6C-PerCP-Cy5.5 (eBioscience),anti–IFNg-PE (BD Biosciences), anti–IFNg-Pacific Blue(eBioscience), and anti–Granzyme B-APC (BD Biosciences).Cellular division was assessed by labeling of high-avidity neu-specific CD8þ T cells with 1.5 mmol/L carboxyfluoresceindiacetate succinimidyl ester (CFSE; CellTrace CFSE cell pro-liferation kit; Invitrogen) before adoptive transfer (8). Per-meability was assessed with LIVE/DEAD Fixable Aqua DeadCell Stain (Invitrogen). Antibody staining was conducted at4�C for 20 minutes in FACS buffer (PBS, 5% FBS, 0.02%NaAzide).
Flow cytometry and intracellular cytokine stainingLymph nodes were dissected 3 and 5 days after adoptive
transfer, and tumors were dissected 5 days after adoptivetransfer, and homogenized by mashing through 40-micrometernylon cell strainers. Tumors were further processed by enzy-matic digestion using collagenase (1 mg/mL, Gibco) and hyal-uronidase (25 mg/mL, Sigma). After digestion, cells werewashed with RPMI before being trypsinized for 2 minutes with0.25% Trypsin-EDTA (Gibco). Lymphocytes were incubated for5 hours at 37�C in CTL media with RNEU420–429 or NP118–126peptide–pulsed T2-Dq target cells in the presence of monensin(GolgiStop; BD Biosciences) at a lymphocyte to target ratio of4:1. Cells were surface stained for Thy1.2 expression beforefixation/permeabilization using a mouse intracellular cytokinestaining (ICS) kit (BD Biosciences) to stain for intracellularIFNg and Granzyme B.
In vitro activation and suppression of CD8þ T cells withexogenous galectin-3
CD8þ T cells were negatively isolated from total splenocytesusing Dynal CD8þ negative isolation kits (Invitrogen) andstimulated for 3 days with anti-CD3/CD28 (Invitrogen) beadsat a T cell to bead ratio of 1:1 according to the manufacturer'srecommendations (13). For suppression studies, cells were alsoincubated with the indicated amount of recombinant mouse orhuman galectin-3. For human studies, cells were also incubatedwith a 5-fold molar concentration of purified IgG from patient'sserum, as described in the figures. After 3 days of activation,CD8þ T cells were assessed for IFNg production by ICS. Beadswere removed by magnetic separation, and cells were platedwith fresh anti-CD3/CD28 beads in a 96-well assay plate at a Tcell to bead ratio of 1:1 in the presence of monensin (Golgi-Stop; Invitrogen) for 5 hours at 37�C. Culture media were alsosupplemented with fresh galectin-3 at the indicated concentra-tions. Following the 5-hour incubation period, ICS studies wereconducted as above.
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 413
Cloning and purification of recombinant mouse galectin-3Total RNA was isolated from in vitro–activated high-avidity
neu-specific CD8þ T cells using the RNEasy Mini Kit (Qiagen).Galectin-3 cDNA was amplified with Superscript III First StrandSynthesis System (Invitrogen) and galectin-3–specific primerscontaining BamHI and NdeI restriction sites: 50-GGAATTCCA-TATGGCAGACAGCTTTTCGCTTAACGATG-30 (Forward) and50-CGGGATCCTTAGATCATGGCGTGGTTAGCGCTGGTGAGG-G-30 (Reverse). The galectin-3 cDNA was cloned into the pET-22B bacterial expression vector (Novagen), and protein expres-sion was carried out according to the manufacturer's instructions.Galectin-3 was purified from bacterial cell lysate materialby binding to lactosyl-agarose beads (Sigma) and eluting with200 mmol/L lactose. Purified material was dialyzed into PBS,and endotoxin was removed using the ToxinEraser EndotoxinRemovalKit (GenScript). Endotoxinwasquantified tobe less than1.0 EU/mL by the Limulus Amebocyte Lysate assay (Pierce).
Direct ex vivo antigen detection assayMice were treated as in tumor challenge experiments, but did
not receive cyclophosphamide or adoptive transfer. Four daysafter vaccination, CD8þDCs and pDCs were isolated from spleentissue using CD8þDC and pDC isolation kits (Miltenyi). CD8þ Tcells were negatively isolated from high-avidity neu-specific TCRtransgenic mice and labeled with CFSE as described above. Allcells were cocultured at a 1:1 ratio in CTL media for 3 days beforeevaluating CFSE dilution and cytokine production by FACS.
Coimmunoprecipitation of galectin-3 and LAG-3Ten microgram LAG-3–specific (Clone 410C9; ref. 14) or
galectin-3–specific (M3/38) antibody and corresponding isotypecontrols were conjugated to Protein G Dynabeads (Invitrogen) inPBS followed by cross-linking with 10 mmol/L BS3. CD8þ T cellswere isolated and activated as previously described. Cell surfaceproteins were cross-linked with 10 mmol/L BS(PEG)9 before celllysis with CelLytic M (Sigma) supplemented with 100 mmol/Llactose and protease inhibitor. Conjugated beads were incu-bated at 4�C overnight with CD8þ T-cell lysates. After washingbeads with TBST (Tris-buffered saline þ 0.1% Tween-20) thefollowing day, bound proteins were eluted by boiling in samplebuffer under reducing conditions. Standard Western blottingprocedures were followed, and protein interactions were shownafter developing membranes for 1 hour on high chemilumi-nescence film.
Gene expression analysisRNA was extracted using the Stratagene Absolutely RNA Nano-
prep Kit. Microarray hybridization and analyses were performedby the Johns Hopkins Deep Sequencing and Microarray CoreFacility using the NuGen amplification system and an AffymetrixExon 1.0 ST array. The data discussed in this publication areaccessible through Gene Expression Omnibus Series accessionnumber GSE59454.
Statistical analysisStudent t tests were performed using GraphPad Prism software
assuming equal variances. Log-rank tests were used for Kaplan–Meier plots. P values of less than 0.05 were considered to bestatistically significant.
ResultsVaccine-induced antibody responses against galectin-3 are
associated with improved DFS in surgically resected PDA patientstreated with a GM-CSF–secreting PDA vaccine.
We developed a functional proteomic approach to identifyPDA-associated proteins thatmight serve as targets of the immuneresponse (2). This approach utilizes pre- and posttreatment serafrom patients who received two GM-CSF–secreting pancreatictumor cell lines as vaccine and compared serologic reactiveproteins between treatment responders (disease-free survivors orDFS >3 years) and nonresponders (early recurrence after a singlevaccine or DFS <3 years). Paired sera from 60 subjects treated on arecently reported phase II study were evaluated by Western blot(Supplementary Fig. S1A–S1D). Proteins that demonstrated anincrease in posttreatment serologic responses were purified bytwo-dimensional gel andmass spectrometry. Elevenproteinswereidentified (Supplementary Fig. S1E andS1F; Supplementary TableS1), two of which were serologic targets recognized by multiplepatient sera. The first protein, Annexin A2, induces an epithelial tomesenchymal transition in PDA cells and promotes PDA metas-tases (2). The second protein, galectin-3, is a galactoside-bindingprotein and a known cancer-associated protein secreted by anumber of tumor types (15). Galectin-3, like Annexin A2, inducesposttreatment serologic responses that correlate with improvedDFS and overall survival (Fig. 1A–C).
Galectin-3 is known to inhibit T-cell activation by T-cell bind-ing (16). Thus, we hypothesized that posttreatment serologicresponses against galectin-3 improve DFS by enhancing CD8þ
T-cell function through inhibition of galectin-3 binding. First, weassessed the ability of recombinant human galectin-3 to inhibitIFNg production by CD8þ T cells maximally stimulated in vitrowith anti-CD3/CD28 beads. The minimal concentration neces-sary to observe a significant reduction in IFNg production was 25mg/mL (data not shown). Next, we tested whether purified IgGfrom prevaccinated and postvaccinated patient serum could pre-vent this reduction. IgG from postvaccine serum, but not pre-vaccine serum, attenuated the suppression observed at 25 mg/mLof galectin-3 (Fig. 1D).
Galectin-3 binds maximally activated CD8þ T cells with aterminally differentiated phenotype
We employed the neu-N mouse model to evaluate the role ofgalectin-3 in neu-specific effector T-cell responses within the TME.Surface galectin-3 was detected only on adoptively transferredhigh-avidity neu-specific T cells that had trafficked into the TME(Fig. 2A and Supplementary Fig. S2D). These cells have previouslybeen shown to be tumor antigen committed and activated incontrast with adoptively transferred cells remaining in other T-cellresiding sites 5 days after adoptive transfer (8). Because themajority of effector CD8þ T cells in the TME have already beenactivated and undergone terminal differentiation, we used lym-phocytes isolated from tumor-draining lymph nodes (TDN),which contain populations of T cells at different stages of division,to illustrate differences in galectin-3 binding during various stagesof T-cell activation in vivo. CFSE labeling of adoptively transferredT cells demonstrated that surface galectin-3 was associated withonly CFSE dilute and CD44þ cell populations, indicating thatthese are differentiated effector CD8þ T cells (Fig. 2B).
Next, we evaluated TILs for phenotypic differences associatedwith increased galectin-3 binding. High-avidity neu-specific
Kouo et al.
Cancer Immunol Res; 3(4) April 2015 Cancer Immunology Research414
CD8þ T cells in TILs were separated into galectin-3 high (galectin-3hi) versus galectin-3 low (galectin-3lo) populations and analyzedfor coexpression of programmed death 1 (PD-1) and lymphocyteactivation gene 3 (LAG-3) because these two markers are ofteninvolved in T-cell tolerance or terminal activation (17–19). PD-1and LAG-3 were increased on galectin-3hi CD8þ T cells whencompared with galectin-3lo cells (Fig. 2C). Thus, galectin-3 bind-ing onCD8þ T cells occurs almost exclusively in the TMEonT cells
displaying a terminally differentiated or tolerized phenotype withelevated expression of PD-1 and LAG-3.
LAG-3 is necessary for galectin-3–mediated suppression ofT-cell–secreted IFNg in vitro
We next evaluated whether PD-1 and LAG-3 are involved inmediating the downstream effects of galectin-3 on T-cell function.
Figure 1.Vaccine-induced galectin-3 antibodyresponses correlate with improvedDFS (1:400 dilution of sera wascompared from pre- andpostvaccination samples). A,prevaccine titers shown for patientswho were either disease free >3 years(DFS > 3 years), disease free less than3 years (DFS < 3years), patients whoreceived only a single vaccine beforedisease recurring (Single Vac), orhealthy donors (Donors). B,postvaccine titers for patients dividedby DFS >3 years (left) or <3 years(right) shown before the first vaccine(Pre-Vac), 28 days after the firstvaccine (Vac 1), after chemoradiationwhich was given after the first vaccine(Post-Rx), and after two additionalvaccines given after completingchemoradiation (Vac 3). C, titersshown over time for 8 of the12 patients demonstrated DFS > 3years. These patients received a totalof 5 vaccinations (Vac 1, Vac 2, Vac 3,Vac 4, and Vac 5), the first one beforechemoradiation, the second, third, andfourth each 1 month apart beginning1 month after completingchemoradiation, and the fifth,6 months after completing the fourthvaccination. D, antibodies werepurified from patient serum by ProteinG isolation and then incubated withhealthy donor T cells and recombinanthuman galectin-3 as described inMaterials and Methods. Top, gating ofIFNgþCD8þT cells on histogramplots.Bottom, percentages of IFNgþ CD8þ
T cells after a 3-day anti-CD3/CD28activation in the presence of variousconditions as shown in the graph. Allexperiments were performed at leasttwo independent times with threereplicates per sample. Bar graph data,mean � SD.
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 415
These molecules are heavily glycosylated and possess severalbinding sites for galectin-3 (20, 21). Although WT CD8þ T cellsexhibited a 20% reduction in IFNg production at exogenousgalectin-3 concentrations of 150 mg/mL or higher, LAG-3 KOCD8þ T cells produced maximum IFNg at all concentrations.
However, eliminating PD-1 in CD8þ T cells had minimal effectin blocking galectin-3 suppression (Fig. 3A and B). Coimmu-noprecipitation studies using activated CD8þ T-cell lysatesfurther demonstrated a physical interaction between LAG-3and galectin-3 (Fig. 3C). Next, cross-linking of cell surfaceproteins before cell lysis allowed us to immunoprecipitatea LAG-3–galectin-3 complex migrating between 150 and250 kDa with a LAG-3–specific antibody (Fig. 3D). Important-ly, we were not able to coimmunoprecipitate PD-1 with galec-tin-3 (data not shown). These data establish a link betweenLAG-3 and galectin-3 on antigen-committed CD8þ T cells andsuggest a new mechanism by which galectin-3 may regulateCD8þ T-cell function.
Depletion of galectin-3 leads to improved tumor-specific CD8þ
T-cell functionTo determine a functional role for galectin-3 in vivo, we bred
the neu-N and high-avidity neu-specific (referred to from here-on as simply "high-avidity") TCR transgenic mice onto agalectin-3 KO background. We observed an increase in thetotal number of IFNg- and Granzyme B–producing T cells onlywhen galectin-3 is knocked out in both the T cell and therecipient mouse. Depletion of galectin-3 in CD8þ T cells aloneor in the recipient mouse alone was insufficient to cause animprovement in CD8þ T-cell effector function as measured bycytokine production (Fig. 4A and B). This observation is con-sistent with the finding that galectin-3 elimination is requiredin both the recipient mouse and the adoptively transferredT cell to completely remove galectin-3 from the cell surface(Supplementary Fig. S2A–S2C).
Next, we tested the ability of high-avidity CD8þ T cells topromote long-term tumor control in a galectin-3–null envi-ronment. We previously determined that a minimum of 4 �106 high-avidity CD8þ T cells is required to achieve tumorcontrol in approximately 75% of neu-N mice (8). To illustratethe efficacy of CD8þ T cells in galectin-3–depleted mice, wetested if 2 � 106 galectin-3 KO high-avidity CD8þ T cells wouldbe sufficient to mediate a similar level of tumor control. Wecompared long-term tumor control in WT neu-N mice receivingWT high-avidity CD8þ T cells, galectin-3 KO neu-N mice receiv-ing galectin-3 KO high-avidity CD8þ T cells, and galectin-3 KOneu-N mice without adoptive transfer (Fig. 4C). Consistent withour functional data, we observed improved tumor-free survivalin galectin-3 KO mice that received galectin-3 KO CD8þ T cells,with 90% of mice remaining tumor free at the end of 60 daysversus 50% of control mice. Furthermore, because 100% ofgalectin-3 KO mice that did not receive high-avidity CD8þ Tcells rapidly developed tumors by day 20, we concluded thatthe survival benefit is mediated by tumor-specific high-avidityCD8þ T cells. We also compared tumor-free survival in galectin-3 WT mice adoptively transferred with either galectin-3 KO orgalectin-3 WT high-avidity CD8þ T cells. Consistent with ourfunctional studies, eliminating galectin-3 in high-avidity CD8þ
T cells alone did not result in improved survival because thisdoes not eliminate surface galectin-3 that occurs due to secre-tion of galectin-3 in the TME (Fig. 4D).
To validate these findings in endogenous antigen-specific T-cell responses, we used the Panc02 tumor model to compareendogenous antitumor T-cell responses in galectin-3 KO miceversus WT mice in response to whole-cell GM-CSF vaccination(22). Sixty days following treatment, a significantly higher
Figure 2.Activation of CD8þ T cells increases surface-bound galectin-3. A, tissueswereprocessed 5 days after adoptive transfer into neu-N mice according to theprocedure outlined in Materials and Methods. Histogram overlays showcomparison of surface galectin-3 mean fluorescence intensity from Thy1.2þ
high-avidity CD8þ T cells from TILs, TDN, and spleen. A LIVE/DEADpermeability stainwas included to gate exclusively on nonpermeabilized cellsidentified by low LIVE/DEAD staining. B, high-avidity CD8þ T cells werelabeled with 1.5 mmol/L CFSE before adoptive transfer into neu-N mice.Galectin-3 surface staining is shown with CFSE dilution and CD44 expressionfor cells gated on Thy1.2, 3 days after adoptive transfer. C, TILs wereextracted 5 days after adoptive transfer. Cells were gated on Thy1.2 anddivided into galectin-3hi and galectin-3lo staining populations based onisotype staining controls. A comparison of PD-1 and LAG-3 coexpression isshown for galectin-3hi and galectin-3lo population. Hi, high; lo, low;. Allexperiments were performed at least two independent times with 3 to 5miceper group.
Kouo et al.
Cancer Immunol Res; 3(4) April 2015 Cancer Immunology Research416
percentage of galectin-3 KO mice were tumor-free versus that ofcontrol WT mice (Fig. 4E). Together, data from these twomouse models provide evidence that galectin-3 can affectvaccine efficacy and may potentially do so via modulation ofCD8þ T cells.
Galectin-3 depletion leads to increased expression ofproinflammatory genes in CD8þ T cells
Because galectin-3 depletion results in increased numbers ofIFNg- and Granzyme B–producing high-avidity CD8þ T cells, wewere interested in identifyingother inflammatory signals thatmay
be affected by galectin-3. We used whole genome microarrayanalysis to compare galectin-3 WT CD8þ TILs isolated fromgalectin-3 WT neu-N mice (WT/WT), galectin-3 WT CD8þ TILsisolated from galectin-3 KO neu-N mice (WT/KO), and galectin-3KOCD8þ TILs isolated from galectin-3 KO neu-Nmice (KO/KO).Adoptively transferred CD8þ T cells in TILs were purified fromendogenous T cells by FACS sorting on Thy1.2, and global geneexpression patterns were compared between groups. We foundthat in both KO/KO and WT/KO CD8þ T cells, molecules asso-ciated with inflammatory response processes were increased rel-ative to that inWT/WTcells (Fig. 5A).Upstream regulator pathwayanalyses predicted activation of TNFa, IFNg , and IL6 signaling
Figure 3.LAG-3, but not PD-1, expression isrequired for galectin-3 suppression ofIFNg production by T cells. The effectof increasing concentrations ofextracellular galectin-3 on thepercentage of IFNg-producing CD8þ
T cells is shown for (A) PD-1 KO versusWT T cells and (B) LAG-3 KO versusWT T cells. Experiments wereperformed at least two independenttimes with three replicates per samplewith errors bars representing SD. C,coimmunoprecipitation of LAG-3 andgalectin-3 from activated CD8þ T-celllysates with either a galectin-3–specific antibody (M3/38) or a LAG-3–specific antibody (410C9). D,coimmunoprecipitation of a LAG-3–galectin-3 complex with LAG-3–specific antibody after cross-linkingof cell surface proteins on activatedCD8þ T cells before cell lysis asdescribed in Materials and Methods.
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 417
pathways in the KO/KO CD8þ T cells but not in WT/KO CD8þ Tcells (Supplementary Table S2).
We also show that S100A8 expression is increased almost 7-fold in galectin-3 KO/KO CD8þ T cells when compared with thatin WT/WT cells (Fig. 5B). S100A8 and S100A9 are EF-hand Ca2þ-bindingproteins that are heavily involved in several inflammatoryprocesses and possess tumoricidal properties in vitro. Theseinflammatory response genes are expressed by neutrophils, buthave not been reported to be expressed by T cells (23).
Inhibition of galectin-3 expression leads to an increase in pDCsin galectin-3 KO mice
The increased expression of several inflammatory proteins inWT high-avidity CD8 T cells adoptively transferred into galectin-3KO recipient mice suggests that other cell types are affected bygalectin-3 signaling. Galectin-3 has been reported to play aregulatory role in neutrophils and DCs (6, 7, 24).
We observed a significant increase in the numbers ofCD11cþ B220þ Ly6Cþ cells in the galectin-3 KO versus in the
Figure 4.Removal of surface galectin-3 fromhigh-avidity CD8þ T cells leads toimproved effector function and tumorclearance. Effector function of TILswas assayed after genetic eliminationof galectin-3 in the CD8 T cells, thetumor stroma, or in both. TILs wereisolated, and cytokine expression wasassessed according to the protocoloutlined in Materials andMethods. Theabsolute number is shown for Thy1.2þ
CD8þ T cells/mg tumor expressingeither (A) IFNg or (B) Granzyme B. Allexperiments were performed at leasttwo independent times with 3 to 5mice per group. Dot plot data, mean�SEM. C, all neu-N recipient mice weretreated as outlined in Materials andMethods. Adoptive transfer groups (n¼ 10) received 2 � 106 galectin-3 WTor KO high-avidity CD8þ T cells asindicated above on day 4. The controlgroup of galectin-3 KO mice (n ¼ 7)did not receive adoptive transfer toassess endogenous antitumorresponse. Mice were followed every 5days for palpable tumor formation. D,galectin-3 WT neu-N mice receivedadoptive transfer of either 2 � 106
galectin-3 WT high-avidity CD8þ Tcells (n ¼ 10) or 2 � 106 galectin-3 KOhigh-avidity CD8þ T cells (n¼ 10) afterreceiving treatment as outlined inMaterials and Methods. Mice werefollowed every 5 days for palpabletumor formation. E, WT C57Bl/6 mice(n¼ 15) or galectin-3 KOC57Bl/6mice(n ¼ 15) received 2.5 � 105 Panc02tumor cells on day 0, 100 mg/kgcyclophosphamide and 50 mg PC61 onday 2, and whole-cell GM-CSF vaccineon day 3. Mice were followed every 5days for development of palpabletumor formation at the injection site.One-tailed P values are shown.
Kouo et al.
Cancer Immunol Res; 3(4) April 2015 Cancer Immunology Research418
galectin-3 WT neu-N mice corresponding with pDCs. Further-more, the overall number of pDCs found in lymph nodes did notseem to vary significantly between na€�ve mice given tumor alone,and mice with tumors that are also treated with cyclophospha-mide and a GM-CSF vaccine (Fig. 6A). This finding indicates thatgalectin-3 may instead play an intrinsic role in pDC homeostasisrather than solely during immune stimulation, as was the case forCD8þ T cells.
A reciprocal relationship has been shown to exist betweenpDCs and myeloid-derived suppressor cells (MDSC; ref. 25).Because pDCs appear to be significantly elevated in the absence
of galectin-3, we determined if there was a correspondingdecrease in the number of immature myeloid cells, which havethe potential to develop into MDSCs. In contrast with what wasobserved with pDCs in galectin-3 KO mice, the number ofimmature myeloid cells found in lymph nodes is minimal inna€�ve and tumor-only mice. GM-CSF will induce MDSCs, andthis is seen in mice with tumors receiving the GM-CSF vaccineregardless of whether galectin-3 is knocked out. However, evenunder these conditions, the galectin-3 KO mice have signifi-cantly fewer immature myeloid cells than the WT mice, indi-cating that galectin-3 plays a role in GM-CSF recruitment of
Figure 5.Removal of galectin-3 increasesexpression of inflammatory moleculesin high-avidity CD8þ T cells. A,summary of microarray findingscomparing high-avidity galectin-3 WTCD8þ TILs from galectin-3 WT neu-Nmice (WT/WT; n ¼ 4), high-aviditygalectin-3 WT CD8þ TILs fromgalectin-3 KO neu-N mice (WT/KO;n¼ 4), and high-avidity galectin-3 KOCD8þ TILs from galectin-3 KO neu-Nmice (KO/KO; n ¼ 4). Analysis wasperformed using Ingenuity PathwayAnalysis software using a filter forlinear fold changes �2 and P � 0.05.Molecules listed above the arrowsbetween each group are immune-associated genes found to beupregulated between groups.Molecules listed below each group areimmune-associated genes found to beupregulated in comparison with theWT/WT control. CCL8, CXCL16,Gp49a/Lilrb4, Lyz1/Lyz2, and RNASE3were found to be upregulated inboth WT/KO and KO/KO groupsin comparison with WT/WT. � ,transcription factors. B, confirmationof S100A8 expression in KO/KO TILsby Western blot with S100A8-specificmAb (R&D). Total proteinconcentration quantified bymeasuring absorbance at 280 nm.Total protein (6 mg) was loaded persample.
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 419
immature myeloid cells, and potential development intoMDSCs (Fig. 6B).
Although pDCs are known to be involved in the promotion ofinflammatory processes important for reversing microbial infec-tions (26), they are also known to be immunosuppressive in theTME when galectin-3 is present (natural environment; refs. 27–29). To clarify the function of pDCs following GM-CSF vaccina-
tion in the absence of galectin-3, we evaluated their capacity toinduce proliferation of na€�ve high-avidity CD8þ T cells comparedwith CD11cþ CD8þ cDCs in vitro using a direct ex vivo antigendetection (DEAD) assay. Our data indicate that na€�ve T cellsproliferate to a greater extent and produce more IFNg whencocultured with both pDCs and cDCs than when cocultured witheither DC subtype alone (Fig. 6C and D). Furthermore, coculture
Figure 6.An increase in pDCs in galectin-3 KOneu-N mice is associated with lowernumbers of immature myeloid cells.Galectin-3 KO and WT neu-N micewere left untreated (n¼ 3), challengedwith 5� 104 NT2.5 tumor cells (n¼ 3),or challenged with 5 � 104 NT2.5tumor cells and treated withcyclophosphamide (CY) on day 2 andwhole-cell GM-CSF vaccine on day 3(n¼4). All micewere sacrificed onday6, and axillary and inguinal lymphnodes were dissected and pooled foreach mouse. Tissues were processedas described in Materials andMethods,and cells were stained for expressionof CD11b, CD11c, B220, and Ly6C. Thetotal number of cells was averaged foreach lymph node, and is shown for (A)pDCs (CD11cþ B220þ Ly6Cþ) and (B)immature myeloid cells (CD11bþ
Ly6Cþ). C, cells were cocultured at a 1:1ratio as specified in the figure beforebeing assessed for either CFSEdilutionor IFNg production by flow cytometry.Cytokine expression was assessed byICS as outlined in Materials andMethods. D, statistical comparison ofcell division in T cells þ pDC versus.T cells þ pDC þ cDC. Data, mean �SEM. This experiment was repeatedonce with similar results. Allexperiments were performed at leasttwo independent times with 3 to 5mice per group in A and B, and with3 replicates per group in C and D.
Kouo et al.
Cancer Immunol Res; 3(4) April 2015 Cancer Immunology Research420
of na€�ve T cells with pDCs alone activated a significantly greaternumber of T cells than with cDCs alone. These findings support anew role for galectin-3 as a regulator of pDCs via inhibiting theirfunction as T-cell activators in the promotion of anticancerinflammatory responses.
DiscussionOur data describe four novelfindings supporting galectin-3 as a
regulator of antigen-specific T-cell activation within the TME.First, galectin-3 binds T cells only after they have been activatedand trafficked into the TME. In addition, T cells and host-derivedcells within the TME, but not tumor cells, are themajor sources ofgalectin-3 that mediate T-cell suppression. Second, galectin-3regulation of activated T cells leads to genome-wide changes ininflammatory gene expression. Third, galectin-3 complexes withLAG-3 on activated terminally differentiated T cells, and func-tional LAG-3 is required for galectin-3–mediated T-cell suppres-sion. Fourth, inflammatory pDCs that enhance antigen-specific T-cell activation are increased in the absence of galectin-3.
We show that galectin-3 binding on T cells occurs primarily inthe TME, and only to tumor-specific T cells. One leading hypoth-esis is that activation of CD8þ T cells leads to changes in theglycosylation machinery that increases the number and accessi-bility of LacNAc motifs available for galectin-3 binding (30).However, glycan remodeling fails to explain why only a smallpercentage of activatedCD8þT cells in peripheral lymphoid tissuebind galectin-3, whereas the majority of activated CD8þ T cells inthe tumor bind galectin-3. Thus, the TME must provide otherconditions that are not satisfied elsewhere to promote galectin-3binding.
The leading assumption is that tumor cells are themajor sourcefor galectin-3 (31). Although this may be the case for certaintumor types, our data demonstrate that galectin-3 expression bypredominantly tumor-specific CD8þ T cells and stromal cellswithin the TME, but not tumor cells, results in cancer-specificCD8þ T-cell suppression. Fibroblasts have been shown inmodelsof fibrosis to secrete large amounts of galectin-3 and are alsoknown to be important for establishing the TME (32). Ourfindings highlight the importance of targeting not just tumorcells but also stromal cells for effective cancer immunotherapy.Future studies should focus on identifying the specific tumorstromal cell types that secrete galectin-3 as well as what otheradditional factors provided by the TME promote galectin-3 bind-ing to T cells.
We also performed genome-wide analysis to evaluate changesthat occur in tumor-specific CD8þ T cells resulting from bindingof extracellular galectin-3 in an in vivo setting. The data demon-strate that removal of galectin-3 leads to increased activation ofproinflammatory pathwayswithinCD8þ T cells. Recent data fromseveral groups demonstrate the capacity for galectin-3 to suppressT-cell function by inducing T-cell anergy via TCR clustering, andthat these T cells can be rescued by removing surface galectin-3(3, 33).Our gene arrayfindings demonstrate that themechanismsof T-cell regulation by galectin-3 extend beyond TCR signalingand include altering T-cell fate at the gene expression level. Thesefindings indicate that the addition of galectin-3–targeted therapyto existing cancer vaccines may need to occur before such altera-tions become irreversible. Additional studies should also furtherelucidate the role of the S100A8/9 signaling pathway as a poten-tial mediator of T-cell lytic activity.
Our in vivo data also demonstrate improved tumor-specificCD8þ T-cell effector function and tumor-free survival only whengalectin-3 is deleted from both adoptively transferred T cells andthe recipient mouse, which corresponds to our microarray find-ings demonstrating upregulation of proinflammatory pathwaysunder the same conditions. These conditions are associated withthe complete absence of galectin-3 on the CD8þ T-cell surface atthe time they are infiltrating the tumor. Therefore, these datasuggest a role for galectin-3 in modulating CD8þ T-cell functionby engaging surface glycoproteins. However, the absence ofendogenous galectin-3 may also enhance antitumor immunityby affecting other cell types involved with T-cell activation.Galectin-3 KO DCs have been shown to increase both T-cellnumber and cytokine production in helminthic infections, andpromote greater Th17 responses to fungal antigens (6, 7, 34).Furthermore, intracellular galectin-3 can also promote TCRdownregulation in CD4þ T cells via interactions with the proteinAlix at the immunological synapse (5), which would be expectedto result in less CD8þ T-cell activation.
PD-1 and LAG-3 are two major coreceptors that have beenshown to modulate T-cell function, and their coexpression hasbeen shown to be associated with regulating terminal T-cellactivation/exhaustion (17). Our finding that LAG-3 expressionis associated with galectin-3 binding, and further, is required forgalectin-3 suppression in vitroprovides a newmechanism throughwhich LAG-3 can regulate CD8þ T cells. LAG-3 is capable ofnegatively regulating the function of CD8þ T cells despite thefact that CD8þ T cells do not interact withMHC II, a known ligandof LAG-3. Anti–LAG-3 antibody therapywas shown to reverse thiseffect in CD4þ-depleted mice, which indicates a direct role forLAG-3 on CD8þ T cells (35). The mechanism by which LAG-3mediates CD8þ T-cell activity is unknown. LAG-3 can be exten-sively glycosylated, and as a result, would be a likely target forgalectin-3 binding. Our in vitro data suggest that CD8þ T-cellsuppression can be induced by galectin-3 cross-linking of LAG-3;however, whether this mechanism exists in vivo remains to beelucidated.
Our data also support a link between galectin-3 and activa-tion of inflammatory pDCs. Interestingly, LAG-3 has beenreported to negatively regulate pDC homeostasis (36). Thus,steady-state levels of extracellular galectin-3 may be necessaryto regulate pDC expansion via LAG-3. Further, because pDCsare either directly or indirectly involved with the homeostasis ofboth MDSCs and Tregs, dysregulation of pDCs may havedownstream consequences on these cell types (25). Indeed,our findings that immature myeloid cells proliferate signifi-cantly less in response to GM-CSF vaccination in galectin-3 KOmice are consistent with the notion of reciprocal regulationbetween pDCs and MDSCs.
The role of pDCs in tumor immunity remains unclear.Although they are important for antiviral immunity, they havealso largely been associated with tolerance induction and poorprognosis in cancer. It may be possible that pDCs have a stim-ulatory role in the absence of extracellular galectin-3. Our micro-array finding that the expression of several inflammatory genes inT cells is increased as a result of galectin-3 deficiency in only host-derived cells helps support this hypothesis. Although these resultsmay be due simply to a reduction of surface galectin-3 on the Tcell, a more compelling argument is that the expansion of host-derived pDCs in the absence of galectin-3 helps to skew theimmune response toward inflammation. Further studies will be
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 421
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
necessary to characterize differences in cytokine production andfunctionality of pDCs in the presence and absence of galectin-3.
In summary, these findings develop a comprehensive profile ofthe extensive impact of galectin-3 at every stage of an antitumorimmune response in a tolerogenicmousemodel.Wepropose thatpatients who develop anti–galectin-3 antibody titers in responseto vaccination are able to neutralize the immunosuppressiveeffects of galectin-3 either directly or by disruption of galectin-3/receptor lattices, and as a result, mount a more effective cyto-toxic CD8þ T-cell response against tumor cells. TCR engagement,coreceptor activation, and the cytokine milieu have long beenconsidered the three signals that determine T-cell fate. However,consideration of protein glycosylation as a "Signal 4" may nowhelp to uncover a new class of molecules for immunomodulation(Fig. 7).
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: T. Kouo, L. Huang, E. Jaffee
Development of methodology: T. Kouo, L. Huang, A.B. Pucsek, T. Armstrong,E. JaffeeAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): T. Kouo, L. Huang, A.B. Pucsek, M. Cao, S. Solt,T. Armstrong, E. JaffeeAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): T. Kouo, L. Huang, S. Solt, E. JaffeeWriting, review, and/or revision of the manuscript: T. Kouo, L. Huang,A.B. Pucsek, M. Cao, T. Armstrong, E. JaffeeAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): T. KouoStudy supervision: T. Kouo, T. Armstrong, E. Jaffee
Grant SupportThis work was funded NIH grants SPORE (P50CA062924) and
R01CA184926.The costs of publication of this article were defrayed in part by the
payment 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 14, 2014; revised January 23, 2015; accepted February 8,2015; published OnlineFirst February 17, 2015.
References1. Lutz E, Yeo CJ, Lillemoe KD, Biedrzycki B, Kobrin B, Herman J, et al. A
lethally irradiated allogeneic granulocyte-macrophage colony stimulatingfactor-secreting tumor vaccine for pancreatic adenocarcinoma. A phase IItrial of safety, efficacy, and immune activation.AnnSurg 2011;253:328–35.
Figure 7.N-glycosylation of coregulatory molecules serves as the signal 4 required for optimal T-cell activation. In the VDN, where na€�ve T-cell priming occurs, activatingsignals are delivered via TCR recognition of peptide–MHC I complexes presented by DCs (signal 1) and interactions between costimulatory molecules andtheir ligands (signal 2). Furthermore, cytokines secreted by regulatory cell populations also affect cellular differentiation (signal 3). pDCs are increased in number inthe absence of extracellular galectin-3, and the cytokines they secrete may skew T-cell differentiation in the direction of a more proinflammatory phenotype.Activation of CD8 T cells leads to increased expression of the coregulatory molecules PD-1 and LAG-3 by terminally differentiated effector CD8 T cells. In the TME,expression of PD-L1 by tumor cells can suppress T-cell function via PD-1. Changes in glycosylation of LAG-3 (signal 4) make it a more favorable target forgalectin-3 binding, which in turn leads to cross-linking and activation of the LAG-3 signaling complex. Cross-linking of LAG-3 results in suppression of effector CD8T-cell function. Lack of galectin-3 increases S100A8/9 expression, which may also lead to optimization of lytic function.
Cancer Immunol Res; 3(4) April 2015 Cancer Immunology Research422
2. Zheng L, Foley K, Huang L, Leubner L, Mo G, Olino K, et al., Tyrosine 23phosphorylation-dependent cell-surface localization of Annexin A2 isrequired for invasion and metastases of pancreatic cancer. PLoS ONE2011;6:e19390.
3. DemotteN,Wieers G, VanDer Smissen P,MoserM, SchmidtC, ThielemansK, et al. A galectin-3 ligand corrects the impaired function of human CD4and CD8 tumor-infiltrating lymphocytes and favors tumor rejection inmice. Cancer Res 2010;70:7476–88.
4. Fukumori T, Takenaka Y, Yoshii T, Kim HR, Hogan V, Inohara H, et al.CD29 andCD7mediate galectin-3-induced type II T-cell apoptosis. CancerRes 2003;63:8302–11.
5. Chen HY, Fermin A, Vardhana S, Weng IC, Lo KF, Chang EY, et al.Galectin-3 negatively regulates TCR-mediated CD4þ T-cell activationat the immunological synapse. Proc Natl Acad Sci U S A 2009;106:14496–501.
6. Breuilh L, Vanhoutte F, Fontaine J, van Stijn CM, Tillie-Leblond I, CapronM, et al. Galectin-3 modulates immune and inflammatory responsesduring helminthic infection: impact of galectin-3 deficiency on the func-tions of dendritic cells. Infect Immun 2007;75:5148–57.
7. Wu SY, Yu JS, Liu FT, Miaw SC, Wu-Hsieh BA. Galectin-3 negativelyregulates dendritic cell production of IL-23/IL-17-axis cytokines in infec-tion by Histoplasma capsulatum. J Immunol 2013;190:3427–37.
8. Weiss VL, Lee TH, Song H, Kouo TS, Black CM, Sgouros G, et al. Traffickingof high avidity HER-2/neu-specific T cells into HER-2/neu-expressingtumors after depletion of effector/memory-like regulatory T cells. PLoSONE 2012;7:e31962.
9. Ercolini AM, Ladle BH, Manning EA, Pfannenstiel LW, Armstrong TD,Machiels JP, et al. Recruitment of latent pools of high-avidity CD8(þ)T cells to the antitumor immune response. J Exp Med 2005;201:1591–602.
10. Machiels JP, Reilly RT, Emens LA, Ercolini AM, Lei RY, Weintraub D, et al.Cyclophosphamide, doxorubicin, and paclitaxel enhance the antitumorimmune response of granulocyte/macrophage-colony stimulating factor-secreting whole-cell vaccines in HER-2/neu tolerized mice. Cancer Res2001;61:3689–97.
11. Reilly RT, Gottlieb MB, Ercolini AM, Machiels JP, Kane CE, Okoye FI, et al.HER-2/neu is a tumor rejection target in tolerized HER-2/neu transgenicmice. Cancer Res 2000;60:3569–76.
12. Ercolini AM, Machiels JP, Chen YC, Slansky JE, Giedlen M, Reilly RT, et al.Identification and characterization of the immunodominant rat HER-2/neuMHCclass I epitopepresentedby spontaneousmammary tumors fromHER-2/neu-transgenic mice. J Immunol 2003;170:4273–80.
13. Trickett A, Kwan YL. T cell stimulation and expansion using anti-CD3/CD28 beads. J Immunol Methods 2003;275:251–5.
14. Woo SR, Li N, Bruno TC, Forbes K, Brown S,Workman C, et al. Differentialsubcellular localization of the regulatory T-cell protein LAG-3 and thecoreceptor CD4. Eur J Immunol 2010;40:1768–77.
15. Cay T. Immunhistochemical expression of galectin-3 in cancer: a review ofthe literature. Turk Patoloji Derg 2012;28:1–10.
16. Peng W, Wang HY, Miyahara Y, Peng G, Wang RF. Tumor-associatedgalectin-3 modulates the function of tumor-reactive T cells. Cancer Res2008;68:7228–36.
17. Woo SR, Turnis ME, Goldberg MV, Bankoti J, Selby M, Nirschl CJ, et al.Immune inhibitory molecules LAG-3 and PD-1 synergistically regulate T-cell function to promote tumoral immune escape. Cancer Res 2012;72:917–27.
18. Topalian SL, Drake CG, Pardoll DM. Targeting the PD-1/B7-H1(PD-L1)pathway to activate anti-tumor immunity. Curr Opin Immunol 2012;24:207–212.
20. Agata Y, Kawasaki A, Nishimura H, Ishida Y, Tsubata T, Yagita H, et al.Expression of the PD-1 antigen on the surface of stimulatedmouse T and Blymphocytes. Int Immunol 1996;8:765–72.
21. Baixeras E, Huard B, Miossec C, Jitsukawa S, Martin M, Hercend T, et al.Characterization of the lymphocyte activation gene 3-encoded protein. Anew ligand for human leukocyte antigen class II antigens. J Exp Med 1992;176:327–37.
22. Leao IC, Ganesan P, Armstrong TD, Jaffee EM. Effective depletion ofregulatory T cells allows the recruitment of mesothelin-specific CD8 Tcells to the antitumor immune response against a mesothelin-expressingmouse pancreatic adenocarcinoma. Clin Transl Sci 2008;1:228–39.
23. Lood C, Stenstrom M, Tyden H, Gullstrand B, Kallberg E, Leanderson T,et al. Protein synthesis of the pro-inflammatory S100A8/A9 complex inplasmacytoid dendritic cells and cell surface S100A8/A9 on leukocytesubpopulations in systemic lupus erythematosus. Arthritis Res Ther 2011;13:R60.
24. Sato S, Ouellet N, Pelletier I, Simard M, Rancourt A, Bergeron MG. Role ofgalectin-3 as an adhesion molecule for neutrophil extravasation duringstreptococcal pneumonia. J Immunol 2002;168:1813–22.
25. Ioannou M, Alissafi T, Boon L, Boumpas D, Verginis P. In vivo ablation ofplasmacytoid dendritic cells inhibits autoimmunity through expansion ofmyeloid-derived suppressor cells. J Immunol 2013;190:2631–40.
26. Takagi H, Fukaya T, Eizumi K, Sato Y, Sato K, Shibazaki A, et al. Plasma-cytoiddendritic cells are crucial for the initiation of inflammation andT cellimmunity in vivo. Immunity 2011;35:958–71.
27. Labidi-Galy SI, Treilleux I, Goddard-Leon S, Combes JD, Blay JY, Ray-Coquard I, et al. Plasmacytoid dendritic cells infiltrating ovarian cancer areassociated with poor prognosis. Oncoimmunology 2012;1:380–2.
28. Sisirak V, Faget J, Vey N, Blay JY, Menetrier-Caux C, Caux C, et al.Plasmacytoid dendritic cells deficient in IFNalpha production promotethe amplification of FOXP3 regulatory T cells and are associated with poorprognosis in breast cancer patients. Oncoimmunology 2013;2:e22338.
29. Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H, et al.Plasmacytoid dendritic cells from mouse tumor-draining lymph nodesdirectly activate mature Tregs via indoleamine 2,3-dioxygenase. J ClinInvest 2007;117:2570–82.
30. Antonopoulos A, Demotte N, Stroobant V, Haslam SM, van der Bruggen P,Dell A. Loss of effector function of human cytolytic T lymphocytes isaccompanied by major alterations in N- and O-glycosylation. J Biol Chem2012;287:11240–51.
31. Dumic J, Dabelic S, Flogel M. Galectin-3: an open-ended story. BiochimBiophys Acta 2006;1760:616–35.
33. Gordon-Alonso M, Demotte N, van derBruggen P. Sugars boost exhaustedtumor-infiltrating lymphocytes by counteracting immunosuppressiveactivities of galectins. Oncoimmunology 2014;3:e28783.
34. Fermin Lee A, Chen HY, Wan L, Wu SY, Yu JS, Huang AC, et al. Galectin-3modulates th17 responses by regulating dendritic cell cytokines. Am JPathol 2013;183:1209–22.
35. Grosso JF, Kelleher CC, Harris TJ, Maris CH, Hipkiss EL, De Marzo A, et al.LAG-3 regulates CD8þ T cell accumulation and effector function inmurineself- and tumor-tolerance systems. J Clin Invest 2007;117:3383–92.
36. WorkmanCJ,WangY, El KasmiKC, PardollDM,MurrayPJ,DrakeCG, et al.LAG-3 regulates plasmacytoid dendritic cell homeostasis. J Immunol 2009;182:1885–91.
www.aacrjournals.org Cancer Immunol Res; 3(4) April 2015 423
Galectin-3 Regulates CD8þ T Cells via LAG-3 and pDCs
