Redundant and Antagonistic Functions of
Galectin-1, -3, and -8 in the Elicitation of T
Cell Responses
This information is current as
of June 18, 2017.
María Virginia Tribulatti, María Gabriela Figini, Julieta
Carabelli, Valentina Cattaneo and Oscar Campetella
J Immunol 2012; 188:2991-2999; Prepublished online 22
February 2012;
doi: 10.4049/jimmunol.1102182
http://www.jimmunol.org/content/188/7/2991
References
Subscription
Permissions
Email Alerts
http://www.jimmunol.org/content/suppl/2012/02/23/jimmunol.110218
2.DC1
This article cites 38 articles, 25 of which you can access for free at:
http://www.jimmunol.org/content/188/7/2991.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2012 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Supplementary
Material
The Journal of Immunology
Redundant and Antagonistic Functions of Galectin-1, -3,
and -8 in the Elicitation of T Cell Responses
Marı́a Virginia Tribulatti,1 Marı́a Gabriela Figini,1 Julieta Carabelli, Valentina Cattaneo,
and Oscar Campetella
G
alectins (Gals) constitute a family of secreted mammalian lectins with affinity for b-galactosides that bind
glycoreceptors on target cells and induce multiple
biological activities that include a broad spectrum of cellular responses, such as proliferation, apoptosis, differentiation, adhesion,
migration, and cytokine secretion (1, 2). Gals are characterized
by the presence of conserved carbohydrate-recognition domains
(CRDs) and are structurally classified into three subgroups: prototype (one CRD, such as Gal-1, -2, and -7), tandem repeat (two linked
CRDs, such as Gal-4, -8, and -9), and chimera (one CRD fused to
a nonlectin domain; Gal-3). Prototypical and chimeric Gals can
homodimerize or multimerize, respectively, thus participating in
many cellular processes, such as cell–cell and cell–matrix interactions, by acting as bivalent or multivalent factors. Another remarkable feature of this family of proteins is the formation of
Instituto de Investigaciones Biotecnológicas, Universidad Nacional de San Martı́n,
and Consejo Nacional de Investigaciones Cientı́ficas y Técnicas, B1650WGA San
Martı́n, Buenos Aires, Argentina
1
M.V.T. and M.G.F. contributed equally to this work.
Received for publication July 27, 2011. Accepted for publication January 19, 2012.
This work was supported by grants from the Agencia Nacional para la Promoción
Cientı́fica y Tecnológica, Argentina. O.C. is a researcher and V.C. and M.V.T. are
fellows at Consejo Nacional de Investigaciones Cientı́ficas y Técnicas, Argentina. J.C.
is a fellow at Agencia Nacional para la Promoción Cientı́fica y Tecnológica.
Address correspondence address and reprint requests to Dr. Marı́a Virginia Tribulatti,
Instituto de Investigaciones Biotecnológicas, Universidad Nacional de San Martı́n,
25 de Mayo y Francia, B1650WGA San Martı́n, Buenos Aires, Argentina. E-mail
address: [email protected]
The online version of this article contains supplemental material.
Abbreviations used in this article: C-CRD, C-terminal carbohydrate-recognition domain; CRD, carbohydrate-recognition domain; DC, dendritic cell; Gal, galectin; NCRD, N-terminal carbohydrate-recognition domain; PKC, protein kinase C; PTPase,
protein tyrosine phosphatase; TDG, thiodigalactoside.
Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00
www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102182
ordered cell surface Gal-glycan structures termed lattices, which
engage specific cell surface glycoconjugates by traditional ligand–
receptor interactions. These structures have been involved in the
control of receptor endocytosis, host–pathogen interactions, and
immune system homeostasis (3).
Gals have arisen as important regulators of the immune response
(2). Different members of this family have been found in primary
and secondary lymphoid organs, as well as in circulating cells, and
were shown to modulate several pathological processes, such as
autoimmunity, chronic inflammation, infection, and tumor progression (4). Gal-1 and -3 are the most abundantly expressed and
extensively studied members of the Gal family. With regard to
T cell homeostasis, Gal-1 was shown to promote apoptosis of
immature CD42CD82 and CD4+CD8+ thymocytes, thus contributing to maintenance of self-tolerance (5). In the periphery, Gal-1
displayed anti-inflammatory effects by inducing apoptosis of activated T cells (6). Additionally, Gal-1 was shown to exert its immunoregulatory role by negatively modulating proinflammatory
cytokine expression, skewing the balance from a Th1 toward a
Th2 or T regulatory response (7).
Gal-3 is another key regulator of T cell homeostasis, with the
peculiarity that it can induce T cell proliferation or apoptosis,
depending on whether it acts intracellularly or extracellularly,
respectively. When added exogenously, Gal-3 is able to induce
apoptosis of several human T leukemia cell lines, as well as activated mouse cells (8). Moreover, this Gal also contributes to the
maintenance of self-tolerance by triggering apoptosis of the
CD42CD82 thymocyte subset (9).
Gal-8, which belongs to the tandem-repeat group, is intrinsically
a heterodimer, because it has an N-terminal CRD (N-CRD) and
a C-terminal CRD (C-CRD) joined by a hinge linker peptide of
variable length. It is expressed in different organs and tissues
under physiological or pathological conditions, such as several
human cancers (10, 11). We previously reported the expression of
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
Galectins, a family of mammalian lectins, have emerged as key regulators of the immune response. We previously demonstrated
that galectin (Gal)-8, from the tandem-repeat subgroup, exerts two well-defined effects on mouse naive peripheral CD4 T cells:
Ag-specific costimulation and Ag-independent proliferation. These stimulatory signals on naive T cells have not been described for
any other Gal. Therefore, we investigated whether Gal-1 and Gal-3, two prominent members of the Gal family, share the
stimulatory effects exerted by Gal-8 on naive T cells. We found that Gal-1 costimulated Ag-specific T cell responses similarly
to Gal-8, as evaluated in the DO11.10 TCROVA-transgenic mouse model, by acting simultaneously on APCs and target CD4 T cells.
In contrast, Gal-3 failed to costimulate Ag-specific T cell responses; moreover, it antagonized both Gal-1 and Gal-8 signals. We
observed that both Gal-1 and Gal-3 were unable to induce Ag-independent proliferation; however, when two Gal-1 molecules were
covalently fused, the resulting chimeric protein efficiently promoted proliferation. This finding indicates that Gal-1 might eventually induce proliferation and, moreover, stresses the requirement of a tandem-repeat structure. Remarkably, a single dose of
recombinant Gal-1 or Gal-8 administered together with a suboptimal Ag dose to DO11.10 mice strengthened weak responses
in vivo. Taken together, these findings argue for the participation of Gals in the initiation of the immune response and allow the
postulation of these lectins as enhancers of borderline Ag responses, thus representing potential adjuvants for vaccine formulations. The Journal of Immunology, 2012, 188: 2991–2999.
2992
Materials and Methods
Mice, cell lines, and cell purification
C57BL/6J, C.Cg-Tg(DO11.10)10Dlo/J (DO11.10), and BALB/cJ breeding
pairs were obtained from The Jackson Laboratory (Bar Harbor, ME) and
bred in our facilities. DO11.10 T CD4 cell hybridoma was kindly provided
by Dr. P. Marrack (Howard Hughes Medical Center, Denver, CO). For
mouse splenocyte purification, spleens from 4–8-wk-old animals were
removed and disrupted against a stainless steel mesh in RPMI 1640 medium (Invitrogen, Carlsbad, CA). The cell suspension was washed and
incubated with RBC lysis buffer (Sigma, St. Louis, MO) and washed again
with medium. MiniMACS columns and anti-CD4–coupled paramagnetic
particles (Miltenyi Biotec, Bergisch Gladbach, Germany) were used for
CD4 T cell purification and depletion, following the manufacturer’s
instructions. Cell purity was checked by FACS. All assays involving animals were approved by the Ethical Committee Board of our Institute.
Reagents
ERK inhibitor [1,4-diamino-2,3-dicyano-1,4-bis(2-aminophenylthio)butadiene] (U0126) was from Cell Signaling Technology (Beverly, MA). Src
family tyrosine kinases selective inhibitor 1-(1,1-dimethylethyl)-1-(4methylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine, PI3K inhibitor [2-
(4-morpholinyl)-8-phenyl-4H-1-benzopyran-4-one] (LY 294002), protein
kinase C (PKC) inhibitor 3-[1-[3-(dimethylamino)propyl]-5-methoxy-1Hindol-3-yl]-4-(1H-indol-3-yl)-1H-pyrrole-2,5-dione (Go-6983), and p38
MAPK inhibitor [4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyridyl)1H-imidazole] (SB-203580) were from Enzo Life Sciences (Farmingdale, NY). The CD45 protein tyrosine phosphatase (PTPase) inhibitor
N-(9,10-dioxo-9,10-dihydrophenanthren-2-yl)-2,2-dimethyl-propionamide
was from Calbiochem (Darmstadt, Germany). Thiodigalactoside (TDG)
was from Carbosynth (Berkshire, U.K.). Con A and Mitomycin C were
from Sigma.
Recombinant Gals
Recombinant Gal-1-8-1 chimera was designed on a synthetic construction
by connecting two sequences of full-length mouse Gal-1 with the 9-aa linker
peptide from mouse Gal-8L isoform (12). Mouse Gal-1 was cloned using
mRNA from thymic epithelial cells as a template (for details see Ref. 12),
and mouse Gal-3 was obtained by a synthetic gene construction (Genscript, Piscataway, NJ). Conditions for protein expression and purification
by lactosyl-Sepharose (Sigma), followed by immobilized metal-affinity
chromatography (GE Healthcare, Uppsala, Sweden), were as described
for mouse Gal-8 (12). After the first step of purification, Gal-1 was alkylated in the presence of 0.1 M iodoacetamide (Sigma) overnight at 4˚C.
The use of alkylated Gal-1 prevents oxidation that leads to a loss of activity
and, at the same time, overcomes the need to use reducing agents, such as
DTT, which is known to introduce deleterious effects in cell cultures (20,
21). Gal-8 was biotinylated using EZ-Link Sulfo-NHS-Biotin (Pierce,
Rockford, IL), following the manufacturer’s instructions. Lectin activity of
these recombinant proteins was tested by hemagglutination assays.
Binding assays
DO11.10 hybridoma cells (1 3 106) were resuspended in 100 ml cold PBSsodium azide and treated with 0.25 mM biotin-labeled Gal-8, alone or in
the presence of Gal-1 or Gal-3. Cells were incubated for 30 min on ice and
washed with ice-cold PBS. After a 1-h incubation with FITC-streptavidin
(eBiosciences, San Diego, CA), cells were washed and fixed in 2% paraformaldehyde in PBS prior to flow cytometry analysis.
Cell proliferation and costimulation assays
Proliferation assays on mouse cells were performed as described previously
(13). Briefly, splenocytes (5 3 105 cells) from C57BL/6J mice were cultured at 37˚C in 5% CO2 for 48 h in flat-shaped, 96-well plates in 0.2 ml
RPMI 1640 medium in the presence of 10% FBS (Invitrogen), 2 mM
glutamine, and 5 mg/ml gentamicin (complete medium). For costimulatory
assays, splenocytes (3 3 105 cells) from DO11.10 mice were cultured for
48 h in 0.2 ml complete medium in the presence of the cognate OVA323–339
peptide at 1 mg/ml (Genscript). When BALB/cJ splenocytes were used as
APCs, they were pretreated with 100 mg/ml Mitomycin C in RPMI 1640
media for 1 h on ice, using siliconized propylene tubes to avoid binding of
adherent cell subpopulations, and cells were washed three times with icecold PBS. Complete blocking of cell division was checked by inhibition of
Con A-activated proliferation. Gal treatment was performed by incubating
APCs or purified CD4 T cells with the indicated amounts of Gal-1 or Gal-8
in RPMI 1640 complete media for 30 min on ice, and cells were washed to
eliminate unbound lectin. In all proliferation assays, 1 mCi [3H]methylthymidine (New England Nuclear, Newton, MA) was added to each well
18 h before harvesting. TDG- and signaling pathway-specific inhibitors
were added 30 min before the addition of recombinant proteins, with the
exception of the U0126 inhibitor, which was added 1 h before. Unstimulated cells’ basal response ranged from 200 to 1000 cpm and was subtracted in all experiments. Assays were performed in quadruplicate.
Inhibition of cell proliferation
For T cell activation, splenocytes (5 3 105) from C57BL/6J mice were
cultured, as described in proliferation assays (see above), in the presence of
2.5 mg/ml Con A for 24 h. Then, preactivated cells were treated with the
indicated amounts of Gals for an additional 24 h. Inhibition of cell proliferation was assessed by adding 1 mCi [3H]methyl-thymidine to each
well 7 h before harvesting.
In vivo costimulation
Initially, to determine suboptimal Ag dose, 6–8-wk-old female DO11.10
mice were immunized i.p. with decreasing amounts (from 5 to 0.05 mg) of
OVA (Sigma) in 0.2 ml PBS; spleens were collected 5 d after immunization. In vitro restimulation of splenocytes was performed basically as for
costimulation assays (see above), using two doses of OVA323–339 cognate
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
two Gal-8 isoforms in mouse thymus and spleen, which differed
only in the length of the linker region. Both isoforms were able to
induce apoptosis of the immature CD4highCD8high thymocyte
subpopulation involving caspase pathway activation, suggesting
that this Gal could also participate in central negative selection
(12). At the periphery, we found that Gal-8 exerts two actions on
naive CD4 T cells: at high concentrations, it induces strong Agindependent proliferation, whereas at low concentrations, it costimulates T cells in the presence of APCs and the corresponding
Ag. These activities are mediated by interaction with the T cell
surface glycoprotein CD45 and involve the activation of ZAP70and ERK signaling pathways (13). The molecular requirements
for both of Gal-8’s stimulatory effects were recently assessed; its
tandem-repeat structure is only essential for proliferation and not
for costimulation (14).
Interestingly, Gal-8’s proliferative and costimulatory signals on
primary T cells have not been described for any other Gal and
suggest a possible involvement of this lectin in inflammatory and
autoimmune processes. Moreover, these stimulatory activities
differentiate Gal-8 from other Gals, particularly Gal-1 and Gal-3,
which are known to display immunoregulatory functions (2). As
discussed above, the ability to kill activated peripheral T cells
endowed Gal-1 and Gal-3 with anti-inflammatory properties,
which were reflected by their therapeutic effects on experimental
models of T cell-mediated autoimmune disorders and inflammatory diseases (15). In this regard, we recently showed that, although Gal-8 is able to induce strong proliferation of freshly
isolated human PBMCs, it promoted cell death when these cells
were preactivated with PHA or CD3/CD28 stimulators (14). Thus,
the effector phase of ongoing responses could be limited, not only
by Gal-1 and Gal-3, but also by Gal-8.
The analysis of the existence of redundant or antagonistic
functions between Gals is a major concern, because these proteins
can converge under normal or pathological conditions, such as
inflammatory foci or certain tumors (16–19). The proapoptotic
activity of Gal-1 and Gal-3 has been extensively characterized,
although most experiments were performed using activated T cells
or T cell lines. Thus, there is scarce or contradictory information
regarding the effect of Gal-1 and Gal-3 on primary naive T cells.
Therefore, the aim of the present work was to contrast the effects
exerted by Gal-8 with Gal-1 and Gal-3 on primary T cell proliferation and Ag-specific stimulation to dissect whether these Gals
are cooperating or antagonizing at the very beginning of the T cell
response induction/elicitation.
GALECTIN INVOLVEMENT IN T CELL RESPONSES
The Journal of Immunology
2993
peptide (0.1 and 0.25 mg/ml). For in vivo costimulation, mice received 0.5
mg OVA i.p. in combination with 10–25 mg Gal-8 (OVA + Gal-8) or 25 mg
Gal-1 (OVA + Gal-1). Control groups consisted of mice that received Ag
only (OVA), vehicle (PBS), or Gals (Gal-1 and Gal-8). In vitro restimulation was performed in the presence of 0.25 mg/ml OVA cognate peptide.
Flow cytometry
A FlowMax cytometer PASIII (Partec, Münster, Germany) and WinMdi 2.9
software were used.
Statistical analysis
The Student t test was used; p values , 0.05 were considered significant.
Results
Gal-1 costimulates Ag-specific T cell response
Gals activate TCR downstream signaling pathways to induce
costimulation
To better characterize Gal-1 and Gal-8 costimulation activities,
we performed assays in the presence of lck (Src)-, PI3K-, PKC-,
p38 MAPK-, and ERK-specific inhibitors, as representative TCR
downstream signal transducers. We also tested CD45 PTPase inhibitor, because we previously determined that Gal-8 interacts with
FIGURE 1. Gal-1 costimulates TCR response at suboptimal Ag dose.
(A) Splenocytes from DO11.10 mice were cultured for 48 h in the presence
of the cognate OVA peptide (1 mg/ml) together with Gal-8 or increasing
amounts of Gal-1. (B) Inhibition of Gal-1–induced costimulation in the
presence of 30 mM TDG. Gal-1 was used at 20 mM. (C) Gal-1 (20 mM) or
Gal-8 (0.2 mM) was added either simultaneously (from a mixture containing both Gals, +) or separately (1 h incubation between Gals, -.), in
costimulation assays. (D) Gal-8 and Gal-1 exhibit an antiproliferative effect on Con A-activated T cells. Splenocytes from C57BL/6J mice were
stimulated with Con A for 24 h, and the indicated concentrations of Gal-1
and Gal-8 were added. The reduction in cell proliferation was determined
after 18 h by 3[H]thymidine incorporation. Depicted assays are representative of at least five (A) or three (B–D) independent experiments and were
carried out each time with different recombinant protein preparations.
*p , 0.05, **p , 0.001, ***p , 0.0001.
CD45 and that its PTPase activity is involved in Gal-8–induced
costimulation and proliferation (13). As expected, inhibition of
TCR signaling pathways resulted in a decreased OVA-induced
response (Fig. 2). The Gal-8 and Gal-1 costimulatory effect was
strongly affected in the presence of these inhibitors, thus supporting
that both Gals may reinforce the same intracellular pathways, activated by the corresponding Ag, to promote T cell costimulation.
Gals induce costimulation by simultaneously acting on APCs
and CD4 T cells
To gain insight into the mechanism by which Gals mediate their
costimulatory activity, we performed a set of experiments to determine whether these lectins are actually acting on APCs, T cells,
or both. First, we designed costimulation assays in which Mitomycin C-treated splenocytes from BALB/c mice were used as
APCs, and highly purified CD4 T cells from TCROVA DO11.10
mice were used as target. As observed in Fig. 3A, Gal-1 and Gal-8
readily stimulated CD4 T cells in the presence of OVA Ag, confirming that these cells are targets for the costimulatory effect.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
In our previous work, Gal-8 costimulatory activity on naive T cells
became evident when splenocytes from DO11.10 mice were incubated in the presence of 0.1–0.2 mM of Gal-8, together with
suboptimal doses of OVA cognate peptide (13). We first investigated whether Gal-1 was able to display a similar activity in this
model. Remarkably, when Gal-1 was tested in costimulation
assays, a dose-dependent effect similar to that of Gal-8 was observed (Fig. 1A). However, 100 times more Gal-1 was needed to
equal the activity of Gal-8. Preincubation with TDG almost
completely prevented Gal-1–induced costimulation, indicating
that this effect relies upon lectin–glycan interaction (Fig. 1B). In
the search for a synergistic effect in costimulatory activity, we
tested the simultaneous addition of Gal-1 and Gal-8; a similar
response was observed to that obtained with Gal-8 alone. Moreover, when lectins were added separately with a 1-hour delay, only
a minor, if any, additive effect was observed (Fig. 1C). These findings suggest that these Gals probably recognize similar receptors at
the T cell surface to induce the same activation pathways. In fact,
we observed that the presence of Gal-1 reduced biotin-labeled Gal8 binding to CD4 T cells, independently from the addition order,
indicating that these lectins actually compete for their binding at the
cell surface (Supplemental Fig. 1A).
Because this was the first incidence, to our knowledge, of T cell
costimulatory activity described for Gal-1, and it seemed to contradict its previously reported immunoregulatory function, we
assessed whether both functions are mutually exclusive. For this
purpose, we incubated splenocytes in the presence of the T cell
mitogen Con A for 24 h and exposed cells to different concentrations of Gal-1. Gal-8 was also tested, because we previously
demonstrated that it is able to inhibit cell proliferation of PHA or
anti-CD3/anti-CD28–stimulated human PBMCs (14). As shown in
Fig. 1D, both Gals inhibited the proliferation of activated T cells,
an effect prevented by the addition of TDG, thus supporting the
involvement of Gals and cell–glycan interactions. Unexpectedly,
this antiproliferative effect could not be ascribed to cell death
induction, because no increment of apoptotic cells was observed
after Gal-1 or Gal-8 treatment of activated splenocytes, as assessed by propidium iodide incorporation or caspase-3 activation
(data not shown).
Taken together, these results support a dual role for Gal-1 and
Gal-8 in the immune response by enhancing initial T cell responses
but limiting those that become exacerbated.
2994
Next, APCs and CD4 T cells were separately preincubated with
Gal-8 or Gal-1 for 30 min on ice to allow Gal binding to the cell
surface and then unbound lectins were washed out. Interestingly,
when Gals were bound to only one of the populations involved,
costimulation was no longer achieved. Furthermore, costimulation
was also absent when APCs and T cells treated separately with
Gals were mixed together. These findings indicate that to completely costimulate the Ag response, Gals need to bind on both
APCs and T cells, at the very same time, during the Ag-presenting
process. To confirm these observations, we repeated this assay
using CD4 T cell-depleted splenocytes from DO11.10 mice as
APCs (Fig. 3B). Again, Gal treatment on either APCs or T cells
was ineffective in inducing costimulation.
A tandem-repeat Gal-1 chimera displays Ag-independent
proliferative activity
In addition to the costimulatory activity, Gal-8 induces strong
proliferation of mouse resting CD4 T cells in the absence of an-
FIGURE 3. Gal-1 and Gal-8 bind to both APCs and T cells to costimulate Ag response. A total of 2 3 105 Mitomycin C-treated splenocytes
from BALB/cJ mice (A) or CD4 T cell-depleted splenocytes from
DO11.10 mice (B) were cocultured with 1 3 105 highly purified CD4
T cells for 48 h in the presence of 1 mg/ml cognate OVA peptide together
with Gal-1 (20 mM) or Gal-8 (0.2 mM). Where indicated, APCs or CD4
T cells were treated before cocultures with 20 mM Gal-1 [APC(Gal-1) or
T(Gal-1)] or 0.2 mM Gal-8 [APC(Gal-8) or T(Gal-8)]. Depicted assays
are representative of two independent experiments.
tigenic stimulus, when used at relatively high concentrations (13).
To test whether Gal-1 also shares this effect, we incubated C57BL/
6J mouse splenocytes in the presence of increasing amounts of
Gal-1. Even at the highest concentrations used (20 mM), Gal-1
was unable to induce the strong cell proliferation observed with 2
mM of Gal-8 (Fig. 4A). It is important to point out that Gal-1 was
assayed at those concentrations widely reported to be effective in
most biological activities tested. It should be kept in mind that,
although it homodimerizes under certain conditions, Gal-1 contains only one CRD, whereas Gal-8 is a tandem of two CRDs
fused by a linker peptide, thus being intrinsically a dimer. This
difference could explain why Gal-1 can only display costimulatory, but not proliferative, activity on T cells (14). In fact, we recently demonstrated that both Gal-8 single N-CRDs and C-CRDs
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 2. Gal-1 and Gal-8 costimulate TCR-specific response by
reinforcing downstream signaling pathways. Splenocytes from DO11.10
mice were preincubated with specific inhibitors: 5 mM 1-(1,1-dimethylethyl)-1-(4-methylphenyl)-1H-pyrazolo[3,4-d]pyrimidin-4-amine (Lck),
5 mM Ly-294002 (PI3K), 1 mM GO-6983 (PKC), 5 mM SB-203580 (p38
MAPK), 1 mM (CD45 PTPase) and 10 mM U0126 (ERK). Then, 1 mg/ml
of the cognate OVA peptide, alone or together with 10 mM Gal-1 (A) or 0.2
mM Gal-8 (B), was added to splenocytes and cultured for 48 h. Depicted
assays are representative of at least three independent experiments, which
were carried out each time with different recombinant protein preparations.
**p , 0.005, ***p , 0.0001.
GALECTIN INVOLVEMENT IN T CELL RESPONSES
The Journal of Immunology
2995
Gal-3 antagonizes Ag-specific T cell response
FIGURE 4. Gal-1 stable dimerization confers proliferative activity. (A)
Splenocytes from C57BL/6J mice were incubated for 48 h in the presence
of 20 mM Gal-1 or 2 mM Gal-8. (B) Schematic representation of the
recombinant chimera Gal-1-8-1. (C) Proliferation assays were performed
with the indicated amounts of Gal-1-8-1 or Gal-8. (D) Splenocytes from
DO11.10 mice were cultured for 48 h in the presence of the cognate OVA
peptide (1 mg/ml) together with increasing amounts of Gal-1 or Gal-1-8-1.
Con A was used at 2.5 mg/ml as positive control. TDG was used at 30 mM.
(E) Phase-contrast microscopy images of cultured mouse splenocytes
(original magnification 3200). After purification, 0.5 3 106 cells were
plated onto 96-well plates and stimulated for 18 h in complete media with
the indicated amounts of Gal-1, Gal-8, or Gal-1-8-1. Assays are representative of more than three independent experiments and were carried out
each time with different recombinant protein preparations. *p , 0.05,
**p , 0.005, ***p , 0.0001.
Next, we investigated whether Gal-3, another prominent member
of the Gal family, was also able to exert costimulation of naive
T cells. In contrast to Gal-1 and Gal-8, Gal-3 was unable to costimulate Ag-specific responses at any of the concentrations tested
(5–10 mM, data not shown), and it inhibited the OVA response
when the cognate peptide dose was #1 mg/ml (Fig. 5A). It was
reported that the lattice formed by Gal-3 negatively regulates the
TCR response by restricting receptor lateral motility (23–25).
These reports support the lack of costimulation in the presence of
Gal-3, as well as its inhibitory effect on the Ag-specific T cell
response. Subsequently, we investigated the addition of Gal-3 on
Gal-1 and Gal-8 costimulatory activity using low (1 mg/ml) and
high (2 mg/ml) OVA peptide concentrations. Under low-Ag conditions, at which Gal-3 inhibits the OVA response per se, it also
prevented Gal-1 and Gal-8 costimulatory effects (Fig. 5B, 5C).
Gal-3 inhibited Gal-1’s effect, despite whether the lectins were
added simultaneously or were separated by 1 hour and independently of the addition order. The costimulatory effect of Gal-8 was
inhibited when Gal-3 was added simultaneously or 1 h before, but
not when added 1 h after (Fig. 5B). Under high-Ag conditions,
even when Gal-3 was unable to affect the OVA response per se, it
was able to prevent either Gal-1 or Gal-8 costimulation when
added simultaneously. However, Gal-3 was unable to affect either
Gal-1– or Gal-8–induced costimulation when added separately
(Fig. 5D, 5E). These results suggest that Gal-3 negatively reg-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
are sufficient to costimulate T cell response to cognate peptide but,
in strong contrast, they are unable to trigger Ag-independent
proliferation. In contrast, chimeric homodimers N-CRD–N-CRD
and C-CRD–C-CRD triggered proliferation, indicating that Gal-8
tandem-repeat structure is essential for Ag-independent proliferation but not for costimulation induction (14). This prompted us to
investigate whether the stable dimerization of Gal-1 could overcome its inability to trigger T cell proliferation. For this purpose,
we designed a chimeric protein in which two Gal-1 CRDs are
covalently fused by the Gal-8L “long” linker peptide (12), Gal-18-1 (Fig. 4B). Gal-1-8-1 retained lectin activity and exhibited 2fold greater hemagglutinating activity than did Gal-1 (6.25 and
12.5 mg/ml, respectively). Remarkably, Gal-1-8-1 displayed a significant proliferative effect from 10 mM, which was inhibited by
preincubation with the Gal inhibitor, TDG (Fig. 4C). Of note, a 10fold greater amount of chimeric Gal-1-8-1 was needed to equal
Gal-8’s proliferative activity (20 and 2 mM, respectively). Moreover, Gal-1-8-1 induced costimulation at a similar rate as did Gal-1
(Fig. 4D), in agreement with the fact that Gal-8 monomeric CRDs
are sufficient to trigger this effect (14). Taken together, these
results indicate that stable linkage of two CRDs provides Gal-1
with a proliferative ability, although it does not necessarily imply
an increase in its costimulatory activity. Alternatively, differences
in Gal-1 and Gal-8 potency might be explained by a fine distinction among CRDs’ specificity rather than Gal molecular requirements. In this regard, the induced cellular phenotype of cultured
splenocytes was completely different in the presence of Gal-1 or
Gal-8: although Gal-8 induced an adhesive and spread phenotype,
both Gal-1 and Gal-1-8-1 promoted strong cell agglutination,
without adhesive or spread phenotype (Fig. 4E). This is in agreement with the fact that Gal-8 contains two CRDs with different
glycan specificities that enables it to promote cell adhesion by
linking the extracellular matrix and the cell surface (22). The
strikingly different cellular phenotypes induced by Gal-1-8-1 and
Gal-8 suggest that they might involve distinct mechanisms of
cellular activation to induce the proliferation of T cells.
2996
GALECTIN INVOLVEMENT IN T CELL RESPONSES
stronger signaling response that was not affected by the subsequent addition of Gal-3. Interestingly, even when Gal-3 was no
longer able to inhibit the OVA response (Fig. 5D, 5E), it was still
able to counteract the other Gals’ activity, although only when
present simultaneously. A possible explanation is that Gal-3 might
compete with the other Gals more efficiently when they are
present at the same time; however, Gal-3 was shown to reduce
biotin-labeled Gal-8 binding to CD4 T cells, regardless of the
addition order (Supplemental Fig. 1B). Like Gal-1, Gal-3 was
unable to induce splenocyte proliferation when tested in Agindependent proliferation assays. Notably, the presence of Gal-3
did not affect Gal-8–induced Ag-independent proliferation (Fig.
5F). Taken together, these findings demonstrate that Gal-3
antagonizes Gal-1 and Gal-8 costimulatory activity but not the
proliferative effect of Gal-8 on T cells.
Gal-8 and Gal-1 increase T cell-specific responses in vivo
Discussion
FIGURE 5. Gal-3 inhibits both Gal-1– and Gal-8–induced costimulation.
(A) Splenocytes from DO11.10 mice were cultured for 48 h in the presence
of increasing amounts of the cognate OVA peptide (OVA), together with Gal3. The inhibitory effect of Gal-3 on Gal-1– and Gal-8–induced costimulation
was tested using two concentrations of peptide: 1 mg/ml (B, C) or 2 mg/ml
(D, E). Combinations of Gal-3 and Gal-8 (B and D) or Gal-3 and Gal-1 (C
and E) were added either simultaneously (from a mixture containing both
Gals, +) or separately (1 h incubation between gals, -.). (A–E) Gal-1 and
Gal-3 were used at 10 mM, and Gal-8 was used at 0.2 mM. (F) Splenocytes
from C57BL/6J mice were incubated for 48 h in the presence of 10 mM Gal3, 2 mM Gal-8, or a combination of both Gals. Assays are representative of at
least three independent experiments. *p , 0.01, **p , 0.001.
ulates the TCR that directly affects the costimulatory activity of
Gal-1 and Gal-8. However, Gal-8 seemed to trigger a rapid and
Gals have arisen as key mediators of the innate and adaptive
immune response that participate in several processes, such as
host–pathogen interaction, inflammatory and autoimmune disorders, host-versus-graft disease, fetal–maternal tolerance, and
self-tolerance maintenance (2). Therefore, how Gals control immune cell homeostasis and the underlying mechanisms are crucial.
Given that some Gals can be simultaneously present in the same
microenvironment, the cross-talk among these lectins is another
key aspect that is now emerging. In the present work, we demonstrated that Gal-1, as well as Gal-8, can costimulate borderline
Ag-specific T cell responses. Blockade of established TCR signaling pathways (Lck, PI3K, PKC, p38 MAPK, and ERK) prevented Gal-1– and Gal-8–induced costimulation, indicating that
these Gals actually enhance weak signals from borderline AgTCR engagement. CD45, an abundant and heavily glycosylated
cell surface mucin, is a common ligand for Gal-1 and Gal-8 (6,
13), and its PTPase activity was previously shown to be involved
in Gal-8–induced proliferation and costimulation, probably by
lowering the TCR activation threshold by dephosphorylation of
the Lck Tyr505 inhibitory site (13). As shown in Fig. 2A, CD45
PTPase activity was also involved in Gal-1–induced costimulation. Although CD45 is a well-established Gal-1 counterreceptor,
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
We found that both Gal-8 and Gal-1 display costimulatory activity
in T cell Ag-specific responses. Next, we investigated whether
these Gals could costimulate an in vivo-induced immune response.
For this purpose, DO11.10 mice were immunized with a single
dose of OVA whole protein alone or mixed with recombinant Gals.
After 5 d, splenocytes were restimulated in vitro with the cognate
OVA peptide. The Ag dose used to immunize animals (0.5 mg)
was pinpointed in preliminary assays to generate a slight response
upon in vitro restimulation with 0.25 mg/ml of cognate peptide
(Fig. 6A). Remarkably, splenocytes derived from mice that were
immunized with a combination of Ag and 25 mg of Gal-1 or Gal-8
displayed an increased response after in vitro restimulation compared with those derived from mice that received Ag alone (Fig.
6B). Splenocytes from mice that received only Gal-1 or Gal-8
recombinant proteins (without Ag) responded similarly to those
from mice that received PBS (data not shown). Remarkably, when
larger doses of Ag were used to immunize animals, Gal-induced
costimulation was no longer evident (data not shown), stressing
its ability to costimulate otherwise borderline responses. These
results correlate with Gal-1 and Gal-8 costimulatory properties
observed in vitro (Fig. 1A) and strongly suggest that both Gal-1
and Gal-8 have the potential to costimulate T cell specificresponses in vivo.
The Journal of Immunology
its participation in Gal-1–induced apoptosis remains controversial.
Recently, Earl et al. (26) reported that Gal-1 binding to CD45 and
subsequent T cell death signaling were controlled by developmentally regulated glycosylation, as well as expression of specific
CD45 glycoforms. Nevertheless, how Gal-1 and Gal-8 modulate
CD45 PTPase activity to induce T cell costimulation needs further
investigation.
Despite the fact that CD4 T cells were identified as target cells
of the costimulatory effect, Gal-1 and Gal-8 have to bind APCs
and CD4 T cells simultaneously to completely achieve TCR activation in the presence of Ag (Fig. 3). Possibly, binding of Gal-8 and
Gal-1 stabilizes the immune synapses during Ag presentation,
thus strengthening TCR-downstream signaling, which results in an
increased cell-activation response, particularly when Ag signals
are weak.
The fact that Gal-3 affected the Ag-specific T cell response
agrees with previous reports demonstrating that the lattice built by
extracellular Gal-3 with surface glycoproteins compartmentalizes
TCR complexes in such a way that inhibits downstream-signaling
activation (24, 25). Apoptosis was discounted as a mechanism
responsible for the antagonistic effects of Gal-3, because this Gal
was shown to induce apoptosis of activated, but not resting, T cells
(data not shown). In agreement, Gal-3 failed to induce phospha-
tidylserine exposure in resting T cells, indicating that T cell activation is required to sensitize cells to Gal-3–induced apoptosis
(21). Although the negative effect of Gal-3 on TCR activation was
absent in the presence of a higher dose of Ag, Gal-3 was still able
to inhibit Gal-1– and Gal-8–induced costimulation only when they
were added simultaneously. Given that it was previously shown
that Gal-3 partially inhibits Gal-1 binding to the T cell surface (9),
we reasoned that Gal-3 might displace Gal-1 or Gal-8 more efficiently when present simultaneously. However, we observed that
Gal-3 partially reduced Gal-8 binding on T cells independently of
the addition order, not fully explaining our findings. Alternatively,
differences in the quaternary structures and cross-linking activities
might explain the inhibitory effects of Gal-3 on the activities of
Gal-1 and Gal-8. In this regard, the observation that exogenously
added Gal-3 antagonizes the apoptotic effects of Gal-1 in susceptible T cells was previously related to a greater avidity of the
Gal-3 pentamer compared with the Gal-1 homodimer for glycoprotein receptors on the cell surface (27).
The fact that 100-fold more Gal-1 is required to equal Gal-8
costimulatory activity could be explained by the observation that
tandem-repeat Gals, which have two CRDs stably fused by a linker
peptide, are more potent effectors than monomeric Gals. In fact, it
has been demonstrated that two covalently linked Gal-1 CRDs
induce apoptosis more efficiently than the monomeric form (28–
30). However, chimeric Gal-1-8-1 did not induce greater costimulation than native Gal-1, indicating that structure is not the only
important factor; rather, a fine distinction among CRDs’ specificity may account for the differences observed.
We previously established several molecular requirements for
Gal-8 to induce its costimulation and T cell-proliferation properties: although the tandem-repeat structure is essential for Agindependent proliferation, both single N-CRDs and C-CRDs are
able to induce Ag-specific costimulation (14). We also determined
that Gal-8’s single N-CRD is sufficient to trigger platelet activation (31). In agreement, in this study we showed that Gal-1 costimulated T cells in the presence of the cognate peptide, but it was
unable to trigger proliferation of naive splenocytes, even when
tested at a high concentration (20 mM). Notably, Gal-1 stable
dimerization stabilized with Gal-8 peptide linker (Gal-1-8-1)
overcame its inability to trigger proliferation in the absence of
Ag, although 10-fold more Gal-1-8-1 was still needed to equal
Gal-8 activity. Again, differences in potency could be related to
distinct CRD specificity among these Gals, because we observed
that proliferating splenocytes displayed an adhesive and spread
phenotype in the presence of Gal-8, whereas cells were only agglutinated in the presence of Gal-1-8-1.
In the present work, Gal-1–stimulating activity on resting T cells
is supported by demonstration of Gal-1–induced costimulation
and chimeric Gal-1-8-1–induced proliferation. This constitutes
a novel role for this Gal, because the majority of previous reports
focused on its proapoptotic role on immature or activated, but not
naive, T cells. In agreement with our results, Perillo et al. (6)
demonstrated in an early report that Gal-1 induces apoptosis on
PHA-activated PBMCs, but it is unable to exert the same effect on
naive peripheral human T cells. In addition, it was shown that,
although Gal-1 induces apoptosis of 2-d Con A-activated T cells,
it promotes the survival of resting T cells without inducing proliferation (32). In contrast to our observations and those from
other investigators, Matarrese et al. (33) observed that Gal-1
sensitized human resting T lymphocytes to Fas-mediated cell
death and that, at high doses, it was capable of inducing apoptosis
in these primary cells. Discrepancies might reflect the different
species used (human versus mouse), different protein preparations,
or the method used to assess cell death, because phosphati-
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
FIGURE 6. Gal-1 and Gal-8 increased Ag-specific response in vivo. (A)
Determination of Ag dose for immunization. DO11.10 mice were immunized i.p. with the indicated amounts of OVA whole protein dissolved in
PBS. After 5 d, spleens were removed, and splenocytes were restimulated
in vitro with two concentrations of cognate OVA peptide (0.1 and 0.25 mg/
ml) for 48 h. Stimulation was assessed by 3[H]thymidine incorporation.
Doses of 0.5 mg/ml of OVA whole protein and 0.25 mg/ml of OVA cognate
peptide were chosen for further immunization and restimulation, respectively. (B) Mice were immunized as in (A): PBS (n = 4), OVA (n = 6), OVA +
Gal-8 (n = 5), and OVA + Gal-1 (n = 6). *p , 0.02, versus OVA, **p ,
0.001, versus OVA.
2997
2998
Acknowledgments
We thank Dr. Philippa Marrack (Howard Hughes Medical Center, Denver,
CO) for providing DO11.10 T cell hybridoma and Dr. Alejandro Cassola
for carefully reading the manuscript. Technical assistance with animal
care provided by Fabio Fraga is highly appreciated.
Disclosures
The authors have no financial conflicts of interest.
References
1. Elola, M. T., C. Wolfenstein-Todel, M. F. Troncoso, G. R. Vasta, and
G. A. Rabinovich. 2007. Galectins: matricellular glycan-binding proteins linking
cell adhesion, migration, and survival. Cell. Mol. Life Sci. 64: 1679–1700.
2. Liu, F. T., and G. A. Rabinovich. 2010. Galectins: regulators of acute and chronic
inflammation. Ann. N. Y. Acad. Sci. 1183: 158–182.
3. Rabinovich, G. A., M. A. Toscano, S. S. Jackson, and G. R. Vasta. 2007.
Functions of cell surface galectin-glycoprotein lattices. Curr. Opin. Struct. Biol.
17: 513–520.
4. Toscano, M. A., J. M. Ilarregui, G. A. Bianco, L. Campagna, D. O. Croci,
M. Salatino, and G. A. Rabinovich. 2007. Dissecting the pathophysiologic role
of endogenous lectins: glycan-binding proteins with cytokine-like activity?
Cytokine Growth Factor Rev. 18: 57–71.
5. Perillo, N. L., C. H. Uittenbogaart, J. T. Nguyen, and L. G. Baum. 1997.
Galectin-1, an endogenous lectin produced by thymic epithelial cells, induces
apoptosis of human thymocytes. J. Exp. Med. 185: 1851–1858.
6. Perillo, N. L., K. E. Pace, J. J. Seilhamer, and L. G. Baum. 1995. Apoptosis of
T cells mediated by galectin-1. Nature 378: 736–739.
7. Toscano, M. A., A. G. Commodaro, J. M. Ilarregui, G. A. Bianco, A. Liberman,
H. M. Serra, J. Hirabayashi, L. V. Rizzo, and G. A. Rabinovich. 2006. Galectin-1
suppresses autoimmune retinal disease by promoting concomitant Th2- and T
regulatory-mediated anti-inflammatory responses. J. Immunol. 176: 6323–6332.
8. Fukumori, T., Y. Takenaka, T. Yoshii, H. R. Kim, V. Hogan, H. Inohara,
S. Kagawa, and A. Raz. 2003. CD29 and CD7 mediate galectin-3-induced type II
T-cell apoptosis. Cancer Res. 63: 8302–8311.
9. Stillman, B. N., D. K. Hsu, M. Pang, C. F. Brewer, P. Johnson, F. T. Liu, and
L. G. Baum. 2006. Galectin-3 and galectin-1 bind distinct cell surface glycoprotein receptors to induce T cell death. J. Immunol. 176: 778–789.
10. Bidon-Wagner, N., and J. P. Le Pennec. 2004. Human galectin-8 isoforms and
cancer. Glycoconj. J. 19: 557–563.
11. Hadari, Y. R., R. Arbel-Goren, Y. Levy, A. Amsterdam, R. Alon, R. Zakut, and
Y. Zick. 2000. Galectin-8 binding to integrins inhibits cell adhesion and induces
apoptosis. J. Cell Sci. 113: 2385–2397.
12. Tribulatti, M. V., J. Mucci, V. Cattaneo, F. Agüero, T. Gilmartin, S. R. Head, and
O. Campetella. 2007. Galectin-8 induces apoptosis in the CD4(high)CD8(high)
thymocyte subpopulation. Glycobiology 17: 1404–1412.
13. Tribulatti, M. V., V. Cattaneo, U. Hellman, J. Mucci, and O. Campetella. 2009.
Galectin-8 provides costimulatory and proliferative signals to T lymphocytes.
J. Leukoc. Biol. 86: 371–380.
14. Cattaneo, V., M. V. Tribulatti, and O. Campetella. 2011. Galectin-8 tandemrepeat structure is essential for T-cell proliferation but not for co-stimulation.
Biochem. J. 434: 153–160.
15. Bianco, G. A., M. A. Toscano, J. M. Ilarregui, and G. A. Rabinovich. 2006.
Impact of protein-glycan interactions in the regulation of autoimmunity and
chronic inflammation. Autoimmun. Rev. 5: 349–356.
16. Eshkar Sebban, L., D. Ronen, D. Levartovsky, O. Elkayam, D. Caspi, S. Aamar,
H. Amital, A. Rubinow, I. Golan, D. Naor, et al. 2007. The involvement of CD44
and its novel ligand galectin-8 in apoptotic regulation of autoimmune inflammation. J. Immunol. 179: 1225–1235.
17. Norling, L. V., M. Perretti, and D. Cooper. 2009. Endogenous galectins and the
control of the host inflammatory response. J. Endocrinol. 201: 169–184.
18. Remmelink, M., L. de Leval, C. Decaestecker, A. Duray, E. Crompot,
N. Sirtaine, S. André, H. Kaltner, X. Leroy, H. J. Gabius, and S. Saussez. 2011.
Quantitative immunohistochemical fingerprinting of adhesion/growth-regulatory
galectins in salivary gland tumours: divergent profiles with diagnostic potential.
Histopathology 58: 543–556.
19. Satelli, A., P. S. Rao, P. K. Gupta, P. R. Lockman, K. S. Srivenugopal, and
U. S. Rao. 2008. Varied expression and localization of multiple galectins in
different cancer cell lines. Oncol. Rep. 19: 587–594.
20. Stowell, S. R., S. Karmakar, C. J. Stowell, M. Dias-Baruffi, R. P. McEver, and
R. D. Cummings. 2007. Human galectin-1, -2, and -4 induce surface exposure of
phosphatidylserine in activated human neutrophils but not in activated T cells.
Blood 109: 219–227.
21. Stowell, S. R., Y. Qian, S. Karmakar, N. S. Koyama, M. Dias-Baruffi, H. Leffler,
R. P. McEver, and R. D. Cummings. 2008. Differential roles of galectin-1 and
galectin-3 in regulating leukocyte viability and cytokine secretion. J. Immunol.
180: 3091–3102.
22. Zick, Y., M. Eisenstein, R. A. Goren, Y. R. Hadari, Y. Levy, and D. Ronen. 2004.
Role of galectin-8 as a modulator of cell adhesion and cell growth. Glycoconj. J.
19: 517–526.
23. Chen, H. Y., A. Fermin, S. Vardhana, I. C. Weng, K. F. Lo, E. Y. Chang,
E. Maverakis, R. Y. Yang, D. K. Hsu, M. L. Dustin, and F. T. Liu. 2009. Galectin3 negatively regulates TCR-mediated CD4+ T-cell activation at the immunological synapse. Proc. Natl. Acad. Sci. USA 106: 14496–14501.
24. Chen, I. J., H. L. Chen, and M. Demetriou. 2007. Lateral compartmentalization
of T cell receptor versus CD45 by galectin-N-glycan binding and microfilaments
coordinate basal and activation signaling. J. Biol. Chem. 282: 35361–35372.
25. Demetriou, M., M. Granovsky, S. Quaggin, and J. W. Dennis. 2001. Negative
regulation of T-cell activation and autoimmunity by Mgat5 N-glycosylation.
Nature 409: 733–739.
26. Earl, L. A., S. Bi, and L. G. Baum. 2010. N- and O-glycans modulate galectin-1
binding, CD45 signaling, and T cell death. J. Biol. Chem. 285: 2232–2244.
27. Ahmad, N., H. J. Gabius, S. André, H. Kaltner, S. Sabesan, R. Roy, B. Liu,
F. Macaluso, and C. F. Brewer. 2004. Galectin-3 precipitates as a pentamer with
synthetic multivalent carbohydrates and forms heterogeneous cross-linked
complexes. J. Biol. Chem. 279: 10841–10847.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017
dylserine exposure is not necessarily associated with Gal-induced
apoptosis (34).
The observation that Gal-1 and Gal-8 stimulated naive T cells,
but displayed an antiproliferative effect when these cells became
activated (Fig. 1), strongly suggests a dual role for these lectins:
enhancing normal or physiological immune responses, especially
when stimulus is limited, and restraining the effector phase of
ongoing or exacerbated responses. However, the antiproliferative
effect could not be ascribed to cell death induction, because apoptosis was only observed when activated splenocytes were incubated in the presence of Gal-3 and not with Gal-1 or Gal-8. Of
note, our results are in line with a previous report that Gal-3, but
not Gal-1, induces apoptosis of primary activated T cells (21).
Therefore, further studies are necessary to properly address the
mechanism by which these Gals are exerting their antiproliferative
effect on activated T cells.
A single dose of 25 mg of Gal-1 or Gal-8, administered with
a suboptimal dose of Ag in mice, was sufficient to increase the
T cell response, demonstrating that Gals can prime resting T cells
in vivo to enhance borderline Ag responses. Although this outcome could be mediated by Gal-induced costimulation processes
observed in vitro, it should be considered that Gals can exert
different functions on many cell types in vivo. For example, Gals
might also be stimulating different APCs to sustain T cell activation after priming. In this regard, it was reported that dendritic
cells (DCs) engineered to express transgenic Gal-1 displayed an
enhanced ability to stimulate naive T cells and, conversely, induced apoptosis of activated T cells (35). These Gal-1–expressing
DCs exhibited a mature phenotype, as reflected by the increase in
MHC and costimulatory molecule expression, as well as enhanced
levels of proinflammatory cytokines. It was also demonstrated that
Gal-1 stimulates T cell proliferation directly through its ability
to promote DC maturation, because pretreatment of DCs with
recombinant Gal-1 was sufficient to trigger T cell stimulation.
Furthermore, Gal-1–treated human DCs displayed an enhanced
migration through extracellular matrix, suggesting that it could
participate in initiating the immune response (36, 37). In line with
this evidence, it was reported very recently that mature and immature human DCs constitutively express Gal-8 protein (38);
moreover, we observed that Gal-8 treatment induced the expression of activation markers on murine bone marrow-derived DCs
(J. Carabelli, M.V. Tribulatti, and O. Campetella, unpublished
observations). With regard to B cells, it was reported that both
Gal-1 and Gal-8, but not Gal-3, promoted plasma cell differentiation (39), probably contributing to the humoral response as well.
Nevertheless, we cannot discount a direct effect of Gals on naive
T cells during the Ag response in vivo, because Gal-8 was demonstrated to induce purified CD4 T cell proliferation (13) and,
as described in this article, chimeric Gal-1-8-1 also induced
splenocyte proliferation.
Finally, the findings reported in this article allow us to postulate
these lectins as enhancers of borderline Ag responses, thus representing suitable candidates as additives for vaccines preparations,
among other applications.
GALECTIN INVOLVEMENT IN T CELL RESPONSES
The Journal of Immunology
28. Bi, S., L. A. Earl, L. Jacobs, and L. G. Baum. 2008. Structural features of
galectin-9 and galectin-1 that determine distinct T cell death pathways. J. Biol.
Chem. 283: 12248–12258.
29. Bättig, P., P. Saudan, T. Gunde, and M. F. Bachmann. 2004. Enhanced apoptotic activity of a structurally optimized form of galectin-1. Mol. Immunol. 41:
9–18.
30. Earl, L. A., S. Bi, and L. G. Baum. 2011. Galectin multimerization and
lattice formation are regulated by linker region structure. Glycobiology 21:
6–12.
31. Romaniuk, M. A., M. V. Tribulatti, V. Cattaneo, M. J. Lapponi, F. C. Molinas,
O. Campetella, and M. Schattner. 2010. Human platelets express and are activated by galectin-8. Biochem. J. 432: 535–547.
32. Endharti, A. T., Y. W. Zhou, I. Nakashima, and H. Suzuki. 2005. Galectin-1
supports survival of naive T cells without promoting cell proliferation. Eur. J.
Immunol. 35: 86–97.
33. Matarrese, P., A. Tinari, E. Mormone, G. A. Bianco, M. A. Toscano, B. Ascione,
G. A. Rabinovich, and W. Malorni. 2005. Galectin-1 sensitizes resting human
T lymphocytes to Fas (CD95)-mediated cell death via mitochondrial hyperpolarization, budding, and fission. J. Biol. Chem. 280: 6969–6985.
34. Stowell, S. R., S. Karmakar, C. M. Arthur, T. Ju, L. C. Rodrigues, T. B. Riul,
M. Dias-Baruffi, J. Miner, R. P. McEver, and R. D. Cummings. 2009. Galectin-1
2999
35.
36.
37.
38.
39.
induces reversible phosphatidylserine exposure at the plasma membrane. Mol.
Biol. Cell 20: 1408–1418.
Perone, M. J., A. T. Larregina, W. J. Shufesky, G. D. Papworth, M. L. Sullivan,
A. F. Zahorchak, D. B. Stolz, L. G. Baum, S. C. Watkins, A. W. Thomson, and
A. E. Morelli. 2006. Transgenic galectin-1 induces maturation of dendritic cells
that elicit contrasting responses in naive and activated T cells. J. Immunol. 176:
7207–7220.
Fulcher, J. A., M. H. Chang, S. Wang, T. Almazan, S. T. Hashimi, A. U. Eriksson,
X. Wen, M. Pang, L. G. Baum, R. R. Singh, and B. Lee. 2009. Galectin-1 coclusters CD43/CD45 on dendritic cells and induces cell activation and migration
through Syk and protein kinase C signaling. J. Biol. Chem. 284: 26860–26870.
Fulcher, J. A., S. T. Hashimi, E. L. Levroney, M. Pang, K. B. Gurney,
L. G. Baum, and B. Lee. 2006. Galectin-1-matured human monocyte-derived
dendritic cells have enhanced migration through extracellular matrix. J. Immunol. 177: 216–226.
Mobergslien, A., and M. Sioud. 2011. Galectin-1 and -3 gene silencing in immature and mature dendritic cells enhances T cell activation and interferon-g
production. J. Leukoc. Biol.
Tsai, C. M., C. H. Guan, H. W. Hsieh, T. L. Hsu, Z. Tu, K. J. Wu, C. H. Lin, and
K. I. Lin. 2011. Galectin-1 and galectin-8 have redundant roles in promoting
plasma cell formation. J. Immunol. 187: 1643–1652.
Downloaded from http://www.jimmunol.org/ by guest on June 18, 2017