Novel insights into the inhibitory effects of Galectin-1 on
neutrophil recruitment under flow
Dianne Cooper,1Lucy V. Norling, and Mauro Perretti
William Harvey Research Institute, Barts and The London, London, United Kingdom
binding protein endowed with anti-inflammatory
properties. The purpose of this study was to inves-
tigate the effects of endogenous and exogenous
Gal-1 on neutrophil recruitment onto TNF-treated
endothelium. The effect of human recombinant
(hr)Gal-1 on markers of neutrophil activation
(CD11b expression, P-selectin glycoprotein ligand
1, and L-selectin shedding) was also assessed. Gal-1
inhibited the platelet-activating factor-induced in-
crease in CD11b expression in a concentration-
dependent manner, as assessed by flow cytometry.
To determine the effects of Gal-1 on neutrophil
recruitment, an in vitro flow chamber was used:
Preincubation of neutrophils with hrGal-1 signifi-
cantly decreased the extent of capture, rolling, and
adhesion on activated endothelial monolayers. This
inhibition was shared with the endogenous protein,
as knockdown of endothelial Gal-1 using small in-
terfering RNA resulted in a significant increase in
the number of cells captured and rolling. To verify
the effects of Gal-1 in an in vivo system, intravital
microscopy of Gal-1 null mice and their wild-type
counterparts was performed. Leukocyte adhesion
and emigration were increased significantly in the
cremasteric circulation of Gal-1 null mice inflamed
with IL-1?. These findings indicate that Gal-1 func-
tions to limit neutrophil recruitment onto a TNF-
treated endothelium, a property that may under-
line its inhibitory effects in acute inflammation. J.
Leukoc. Biol. 83: 1459–1466; 2008.
Galectin-1 (Gal-1) is a ?-galactoside-
Key Words: endothelial cells ? inflammation ? adhesion molecules
The hallmark and function of an acute inflammatory response
are the recruitment of neutrophils from the peripheral circula-
tion to the site of injury. The adhesion of circulating inflam-
matory cells to the vessel wall and their subsequent migration
into the interstitium are complex, tightly regulated processes
that involves specific ligand/receptor interactions between
blood cells and endothelial cells. Just as a host of mediators
exists to induce and promote inflammation, a growing number
of endogenous, anti-inflammatory mediators also act to limit
and turn off the inflammatory response [1–3]. Evidence sug-
gests that Galectin-1 (Gal-1) may belong to this family of
endogenous, anti-inflammatory mediators.
Originally identified in organs of electric eel and termed
electrolectin , Gal-1 now belongs to a family of 15 lectins,
classified by their carbohydrate recognition domains and their
affinity for ?-galactoside. The anti-inflammatory/immunoregu-
latory nature of Gal-1 has been identified in various animal
models of chronic inflammatory and autoimmune diseases. The
administration of human recombinant (hr)Gal-1 in vivo pre-
vents the development of chronic inflammation and ongoing
disease in models of experimental autoimmune myasthenia
gravis and experimental autoimmune encephalomyelitis [5, 6].
In collagen-induced arthritis, i.p. injection of hrGal-1 or ex-
pression of mouse Gal-1 from engineered syngeneic fibroblasts
attenuated the development of experimental arthritis. This
effect was associated with induction of T cell apoptosis and a
switch from a Th1- to a Th2-driven condition .
Pharmacological intervention with Gal-1 also exerts protec-
tive actions in models of tissue injury, e.g., acute hepatotoxicity
following Con A administration , a mouse model of entero-
colitis , and experimental autoimmune uveitis . Again,
the underlying mechanism was proposed to be inhibition of T
cell activation and a shift from a Th1 to a Th2 lymphocyte
profile [9, 10]. Our own work has shown that hrGal-1 inhibits
neutrophil chemotaxis and transmigration through endothelial
monolayers and reduces leukocyte recruitment onto IL-1-
treated, mesenteric vessels . Here, we extend these find-
ings by determining the potential of hrGal-1 to alter neutrophil
adhesion molecule expression as well as the recruitment of this
cell type to the endothelium under flow conditions. In addition,
we complemented this analysis by addressing, for the first time,
the role of endogenous Gal-1 in an inflamed microcirculation.
MATERIALS AND METHODS
hrGal-1 was kindly provided by Prof. Richard D. Cummings (Department of
Biochemistry, Emory University School of Medicine, Atlanta, GA, USA). The
1Correspondence: William Harvey Research Institute, Barts and The Lon-
don, School of Medicine and Dentistry, Charterhouse Square, London EC1M
6BQ, UK. E-mail: firstname.lastname@example.org
Received December 14, 2007; revised February 8, 2008; accepted February
0741-5400/08/0083-1459 © Society for Leukocyte Biology
Journal of Leukocyte Biology
Volume 83, June 2008
anti-Gal-1 rabbit polyclonal antibody was raised against hrGal-1 in-house and
IgG-purified (1.18 mg/ml) as described previously  and validated .
Assessment of hrGal-1 binding to
polymorphonuclear neutrophil by flow cytometry
Determination of cell-surface binding sites on neutrophils was assessed using
biotinylated hrGal-1 (biotGal-1; using classical biochemical protocols, Pierce,
Rockford, IL, USA). The local research ethics committee approved experi-
ments with healthy volunteers. Informed consent was provided according to the
Declaration of Helsinki. Blood was collected into 3.2% sodium citrate and
diluted 1:1 in RPMI 1640 (Sigma-Aldrich, Poole, USA) before separation
through a double density gradient as described previously . After neutro-
phil isolation and washing, contaminating erythrocytes were removed by hy-
potonic lysis. Neutrophils were incubated with or without platelet-activating
factor (PAF; 10?9M; C16 form: C26H54NO7P, Sigma-Aldrich) for 30 min at
37°C. Cells were then plated at a density of 2 ? 105cells per well in 96-well
plates and incubated with biotGal-1 (4 ?g/ml) for 1 h on ice. To determine
whether Gal-1 binding was carbohydrate-dependent in nature, neutrophils
were coincubated with Gal-1 in the presence of 30 mM thiodigalactoside or
sucrose. Following three washes with FACS buffer (PBS containing 0.2% BSA
and 1.3 mM CaCl2), cells were incubated with PE-conjugated streptavidin
(Caltag, Burlingame, CA, USA) for 45 min on ice. Unstained controls were also
prepared for accurate calibration of the FACS machine. Flow cytometry was
performed using a FACScan II analyzer (Becton Dickinson, Cowley, UK) with
an air-cooled, 100 mW argon laser tuned to 488 nm, connected to an Apple
Macintosh G3 computer (Cupertino, CA, USA) running CellQuest II (Becton
Dickinson, Franklin Lakes, NJ, USA). Gal-1 binding was recorded as units of
fluorescence, where the median fluorescence intensity for 10,000 cells was
measured in the fluorescence 2 (FL2) red channel (590 nm).
Assessment of adhesion molecule expression by
Peripheral blood neutrophils, isolated as for the Gal-1-binding studies, were
incubated with or without hrGal-1 (0.04–4 ?g/ml) or PAF (10?9M) for 30 min
at 37°C. Cells were then plated at a density of 2 ? 105cells per well in 96-well
plates and incubated with purified mAb: mouse anti-human L-selectin (20
?g/ml, clone FMC46, AbD Serotec, Oxford, UK), mouse anti-human CD11b (5
?g/ml, clone ICRF44, AbD Serotec), or mouse anti-human PE-conjugated
P-selectin glycoprotein ligand 1 (PSGL-1; 10 ?g/ml, clone KPL-1, BD PharM-
ingen, Erembodegem, Belgium) for 1 h on ice, prior to staining with FITC-
conjugated F(ab?)2goat anti-mouse IgG (1:200, AbD Serotec). Isotype and
unstained controls were also prepared for accurate calibration of the FACS
machine. Flow cytometry was performed as described above with the following
exceptions: L-selectin and CD11b expression was recorded as units of fluo-
rescence where the median fluorescence intensity for 10,000 cells was mea-
sured in the FL1 green channel (548 nm). In the case of the anti-PSGL-1
antibody, the red FL2 channel was used (590 nm).
Flow chamber assay
Confluent HUVEC (Glycotech, Gaithersburg, MD, USA) monolayers (up to
passage 4) were stimulated with TNF-? (10 ng/ml, Sigma-Aldrich) for 4 h.
Neutrophils were isolated as for flow cytometry experiments, diluted to 1 ?
106/ml in Dulbecco’s PBS supplemented with Ca2?and Mg2?, and incubated
with or without hrGal-1 (0.04–4 ?g/ml) for 10 min prior to flow at 37°C. The
flow chamber assay was run as described previously . In brief, the chamber
was placed under an Eclipse TE3000 microscope (Nikon, Melville, NY, ISA)
with 40? magnification, and neutrophils (1?106/ml) were perfused over the
endothelial monolayers at a constant rate of 1 dyne/cm2using a syringe pump
(Harvard Apparatus Inc., South Natick, MA, USA). After 8 min of perfusion,
six random fields were recorded for 10 s each using a JVC TK-C1360B digital
color video camera, ready for off-line analysis.
Video sequences were transferred to a computer and loaded into ImagePro-
Plus software (Media Cybernetics, Wokingham, Berkshire). Neutrophils were
manually tagged and their movements on the endothelium monitored. The total
number of interacting cells was quantified as initial cell capture and further
classified as rolling or firmly adherent (cells that remained stationary for the
10-s observation period) as described in the literature .
Knockdown of Gal-1 using small interfering
HUVEC were seeded 24 h before transfection at a density of 2 ? 105cells in
antibiotic-free media. Transfections were performed with nontargeting or a pool
of three Gal-1 target-specific siRNAs (Santa Cruz Biotechnology, Santa Cruz,
CA, USA; sense strand 1: CAGCAACCUGAAUCUCAAA, sense strand 2: CCA-
GAUGGAUACGAAUUCA, sense strand 3: GUGUGGCCUUUGACUGAAA) with
cells at 60–80% confluence, according to the manufacturer’s instructions. Knock-
down of Gal-1 protein expression was monitored by Western blotting, as described
previously using a rabbit anti-Gal-1 polyclonal antibody . Flow chamber
experiments were performed as described above, 48 h post-transfection.
Breeding founders of the Gal-1 null mouse colony were received from the
Consortium for Functional Glycomics, and a colony was established at B&K
Universal (Hull, UK). These mice are on a homogenous C57Bl6 background,
and age- and sex-matched wild-type (WT) mice were purchased from B&K
Universal. All experiments were performed with male animals (body weight
25–30 g) strictly following the Home Office regulations (Guidance on the
Operation of Animals, Scientific Procedures Act 1986).
Intravital microscopy was used to observe IL-1?-induced leukocyte responses
within the cremasteric microcirculation of Gal-1 null and WT mice. IL-1? was
used for these studies rather than TNF-?, as at least part of the adhesive
response to TNF-? within the cremasteric microcirculation results from a
direct activation of neutrophils . IL-1? (30 ng in 400 ?l saline) was
injected intrascrotally as described previously  at Time 0, and the micro-
circulation was observed at 2, 4, and 6 h post-injection. The cremaster was
prepared for intravital microscopy as recently described . In brief, mice
were anesthetized with a mixture of xylazine (7.5 mg/kg) and ketamine (150
mg/kg); the cremaster was then dissected free of skin and fascia, opened
longitudinally, and pinned against the viewing platform of a plexiglass stage.
The preparation was mounted on a Zeiss Axioskop “FS” microscope (original
magnification, 40?, Carl Zeiss, Welwyn Garden City, UK) and transillumi-
nated with a 12-V, 100-W halogen light source. To avoid drying out, the
cremaster muscle was superfused with bicarbonate-buffered saline (pH 7.4,
37°C, gassed with 5% CO2/95% N2) at a rate of 2 ml/min.
Following a 30-min stabilization period, video recordings were made with a
Hitachi charged-coupled device color camera (KPC571, Tokyo, Japan) and a
S-VHS video-recorder (SVO-9500MDP) for subsequent off-line analysis. Leu-
kocyte rolling flux, firm adhesion, and transmigration in post-capillary venules
with a wall shear rate ?500 per s and diameter 20–40 ?m were quantified as
described previously . Briefly, rolling flux was quantified as the number of
rolling leukocytes to pass a defined point on the venular wall per minute.
Firmly adherent leukocytes were classified as those remaining stationary for
30 s within a given 100-?m vessel segment and transmigration as the number
of leukocytes that had emigrated up to 50 ?m on either side of a 100-?m vessel
segment. In each animal, responses from several vessel segments (three to five)
and multiple vessels (three to five) were quantified.
Statistical difference across the different treatments was analyzed by one-way
ANOVA, followed if significant, by Bonferroni’s post-hoc test. Where two
variables were analyzed, a Student’s t-test was used. All data are reported as
mean ? SEM of n experiments performed for the in vitro analyses in duplicate.
P ? 0.05 was considered significant.
Gal-1 binds to neutrophils in a carbohydrate-
No differences were observed between the levels of biotGal-1
bound to naı ¨ve compared with PAF-stimulated neutrophils
(Fig. 1): Quantified binding was 219 ? 16 and 223 ? 30 mean
1460Journal of Leukocyte Biology
Volume 83, June 2008
fluorescence intensity (MFI) units, respectively (n?4, not sig-
nificant). Gal-1 binding was carbohydrate-dependent in nature,
as it was markedly inhibited in the presence of 30 mM thiodi-
galactoside (47?8 MFI, n?4, P?0.001).
Exogenous Gal-1 inhibits PAF-induced CD11b
expression on neutrophils
Initially, we determined the effects of exogenous Gal-1 on
neutrophil adhesion molecule expression in the presence and
absence of PAF. PSGL-1 levels were not significantly altered
in response to PAF in the absence of hrGal-1; however, addi-
tion of 4 ?g/ml Gal-1 in the presence of PAF resulted in a
significant decrease in PSGL-1 levels compared with neutro-
phils treated with 4 ?g/ml Gal-1 alone (Fig. 2A). PAF incu-
bation resulted in a significant increase in CD11b expression
and L-selectin shedding by the neutrophils (Fig. 2, B and C,
respectively): hrGal-1 also inhibited the PAF-induced CD11b
expression in a concentration-dependent manner, such that
with 4 ?g/ml hrGal-1, there was no significant difference
observed between neutrophils incubated with or without PAF
(Fig. 2B). The hrGal-1 did not cause any significant changes to
L-selectin expression (Fig. 2C).
Exogenous Gal-1 inhibits neutrophil:HUVEC
interactions under flow
To determine whether the effects on CD11b expression were
translatable to a functional effect, an in vitro flow chamber
was used to assess the effects of hrGal-1 on neutrophil:
HUVEC interactions under flow. Preincubation of neutro-
phils with hrGal-1 resulted in a significantly lower extent of
cell capture by the endothelium—significant at the lower
concentrations of hrGal-1 applied: 0.04 and 0.4 ?g/ml (Fig.
3A). With regards to neutrophil rolling, an inhibitory effect
was observed with the lowest (0.04 ?g/ml) and highest (4
?g/ml) concentrations of hrGal-1, significantly reducing the
number of rolling neutrophils (Fig. 3B). Again, concentra-
tion-dependent effects were observed with regards to neu-
trophil adhesion, and only the lowest concentration of Gal-1
(0.04 ?g/ml) was significantly active (Fig. 3C). The highest
concentration of hrGal-1 used (4 ?g/ml) caused a slight
increase in firm adhesion that was significant compared with
the two lowest concentrations of Gal-1 used (0.04 and 0.4
Fig. 1. Gal-1 binding to neutrophils. Isolated neutrophils (1?106/ml) were
incubated in the presence/absence of PAF (10?9M) for 30 min at 37°C prior
to incubation with biotGal-1 (4 ?g/ml) for 1 h on ice. Thiodigalactoside (TDG;
30 mM) or sucrose (Suc; 30 mM) was added to some samples at the same time
as the Gal-1, and Gal-1 binding was assessed by flow cytometry. Histograms
represent fluorescence intensity in the FL2 channel and are representative of
separate experiments from four different donors.
Fig. 2. Effect of Gal-1 on neutrophil adhesion molecule expression. Iso-
lated neutrophils (1?106/ml) were incubated with PAF (10?9M) for 30 min
at 37°C in the presence or absence of hrGal-1 (0.04–4 ?g/ml). Neutrophils
analyzed by flow cytometry for PSGL-1 (A), CD11b (B), and L-selectin (C)
expression. Results are expressed as percentage of control. Experiments
were performed in duplicate from three different donors.#, P ? 0.05 vs. 4
?g/ml Gal-1; *, P ? 0.05, versus Control (Ctrl);†, P ? 0.05 vs. 0.04 ?g/ml;
‡, P ? 0.05 vs. 0.4 ?g/ml.
Cooper et al.
Endothelial Galectin-1 and cell trafficking1461
?g/ml) but not to control when analyzed by one-way
ANOVA followed by Bonferroni’s post-hoc test. These ef-
fects of Gal-1 were found to be carbohydrate-dependent, as
they could be inhibited by coincubation with 30 mM thiodi-
galactoside (58, 116, and 30% of reversal on the hrGal-1
effect on neutrophil capture, rolling, and adhesion; n?3
tested on 0.04 ?g and 15, 66, and 46% of reversal on
hrGal-1 effect on neutrophil capture, rolling, and adhesion;
n ? 3 tested on 4 ?g/ml hrGal-1).
Knockdown of endothelial Gal-1 leads to
enhanced neutrophil recruitment
To determine whether endogenous Gal-1 plays a role in limit-
ing neutrophil recruitment, siRNA was used to knockdown
Gal-1 within HUVEC. Application of a pool of three target-
specific siRNAs maximally suppressed endothelial Gal-1 by
?50% at 48 h and 72 h post-transfection (Fig. 4B), without
affecting the expression of Gal-3 or the housekeeping gene
?-actin, as detected by Western blotting (Fig. 4A). Forty-eight
hours post-transfection (time-point chosen for optimum conflu-
ency for flow chamber assays), the attenuated endothelial Gal-1
expression led to a substantial increase (89%) in the initial
capture of neutrophils to the endothelium; a prerequisite for
further interactions (Fig. 5A) with a subsequent increase in
cell rolling also observed (Fig. 5B). No changes were observed
with regards to firm adhesion of neutrophils (Fig. 5C).
Increased leukocyte emigration in Gal-1 null mice
To determine the effects of endogenous Gal-1 in an in vivo
system, intravital microscopy studies were carried out in Gal-1
null mice and their WT counterparts. Following administration
of murine rIL-1? intrascrotally, the number of rolling (Fig.
6A), adherent (Fig. 6B), and emigrated (Fig. 6C) leukocytes
was quantified. The number of rolling and emigrated leuko-
cytes showed a time-dependent increase in both strains of mice
with maximal responses observed at the 6-h time-point. Leu-
Fig. 3. Exogenous Gal-1 inhibits neutrophil:HUVEC interactions under flow.
Effects of hrGal-1 on neutrophil capture (A), rolling (B), and adhesion (C).
Isolated neutrophils (1?106/ml) were incubated with Gal-1 (0.04–4 ?g/ml) for
10 min at 37°C prior to perfusion over TNF-? (10 ng/ml; 4 h)-stimulated
HUVECs. Interactions were quantified from six random fields/treatment. Re-
sults are expressed as percentage of control of four independent experiments.
*, P ? 0.05, versus control;#, P ? 0.05, versus 0.04 ?g/ml Gal-1 and 0.4
?g/ml Gal-1, n ? 4.
Fig. 4. Selective decrease of endothelial Gal-1 protein expression. Knock-
down of Gal-1, but not Gal-3, in HUVEC as determined by Western blotting
(A). Cumulative data from greater than or equal to three individual experi-
ments. Data are mean ? SEM; *, P ? 0.05, versus 48 h nontransfected control
1462Journal of Leukocyte Biology
Volume 83, June 2008
kocyte adhesion peaked at the 4-h time-point in Gal-1 null and
WT mice. There was no significant difference between the
numbers of rolling leukocytes in either genotype, whereas
significantly more leukocytes became adherent in the postcap-
illary venules of Gal-1 null mice at the 4-h time-point, the peak
of the adhesive response. With regards to leukocyte emigration,
significantly higher numbers of leukocytes were observed to
emigrate in Gal-1 null mice at the 2-h and 6-h time-points.
The effectiveness of pharmacological intervention with Gal-1
has been proven in a number of models for autoimmune con-
ditions including rheumatoid arthritis , diabetes , and
graft-versus-host disease , all of which have a strong T cell
component. Although these and other studies have reported the
effectiveness for hrGal-1 in dampening Th1-driven patholo-
gies, largely as a result of induction of T cell apoptosis, few
studies have investigated Gal-1 effects during acute inflamma-
tion. In the present study, we have investigated for the first time
the effect of Gal-1 on neutrophil-endothelium interactions un-
der flow using an in vitro human system that resembles phys-
iological events occurring in the inflamed microcirculation. We
report here a role for the exogenous and endogenous protein in
limiting neutrophil recruitment as evident in in vitro and in
vivo systems. These findings implicate Gal-1 as a novel, en-
dogenous, anti-inflammatory mediator involved in controlling
the trafficking of neutrophils during an inflammatory response.
Fig. 5. Decreased endothelial Gal-1 enhances neutrophil:HUVEC interac-
tions under flow. Knockdown of Gal-1 in HUVEC affects neutrophil capture
(A), rolling (B), and adhesion (C), as determined 48 h post-transfection.
Neutrophils were isolated as described in Materials and Methods and were
perfused over TNF-? (10 ng/ml; 4 h)-stimulated HUVECs. Interactions were
quantified from six random fields/treatment. Results are expressed as percent-
age of control of greater than or equal to three independent experiments; *,
P ? 0.05, versus nontransfected control.
Fig. 6. Lack of endogenous Gal-1 leads to increased leukocyte emigration in
vivo. Gal-1 null mice and their WT counterparts were injected with IL-1?
intrascrotally for the time-points shown. Leukocyte flux (A), adhesion (B), and
emigration (C) were quantified in three to five segments of three to five vessels
per mouse by intravital microscopy of the cremaster muscle. Results are
expressed as mean ? SEM of four to seven mice/group; *, P ? 0.05, versus WT.
Cooper et al.
Endothelial Galectin-1 and cell trafficking 1463
The effects of Gal-1 on neutrophil function have been ad-
dressed in few studies. It has been demonstrated that although
Gal-1 induces phosphatidylserine exposure on the neutrophil
plasma membrane, it does not, in contrast to activated lym-
phocytes, induce neutrophil apoptosis [19, 20]. This effect of
hrGal-1 renders the cells sensitive to phagocytic recognition
and removal ; hence, if confirmed in an in vivo setting, it
would be a proresolving, anti-inflammatory signal. Gal-1 has
also been demonstrated to induce NAD(P)H oxidase and sub-
sequently, superoxide generation in exudated neutrophils in
what could be described as a proinflammatory role for Gal-1
. These effects, however, were only evident at high (40 ?M)
concentrations, which far exceed the maximal concentration
used in this study (maximal concentration used of 275 nM). As
NAD(P)H oxidase induction was only observed in primed
neutrophils and coincided with granule mobilization, it was
suggested that priming resulted in up-regulation of a receptor
for Gal-1, which was not present on naı ¨ve neutrophils. A
receptor(s) for Gal-1 on neutrophils has not been identified to
date. In agreement with the studies by Almkvist et al.  and
Dias-Baruffi et al. , who described an increase in Gal-1
binding to neutrophils collected from skin blister exudates or
following treatment with fMLP, we have also demonstrated
increased Gal-1 binding to neutrophils post-adhesion to the
endothelium . However, in the present study, we did not
observe increased Gal-1 binding after stimulation with PAF,
nor in our previous study following PMA stimulation ,
indicating that up-regulation of a receptor might not be a
necessary prerequisite to unveil Gal-1 effects on the neutro-
phil; in line with this hypothesis, hrGal-1 produces an increase
in intracellular calcium in naı ¨ve and activated neutrophils
. The Gal-1 binding to naı ¨ve neutrophils observed in the
present study was diminished significantly by coincubation
with thiodigalactoside, indicating that the binding to neutro-
phils is carbohydrate-dependent. Moreover, these data are in
line with the abolition of the effects of Gal-1 in the flow
chamber by thiodigalactoside.
In light of our previous studies in which hrGal-1 inhibited
neutrophil chemotaxis and transmigration , we addressed
the effect of hrGal-1 on neutrophil adhesion molecule expres-
sion as key determinants of neutrophil recruitment. The quan-
titative increase in CD11b expression on neutrophils in re-
sponse to agonists such as PAF or fMLP, although not neces-
sarily sufficient to support neutrophil adhesion, serves as an
indication of neutrophil activation [22, 23]. In conjunction with
an increase in CD11b expression, neutrophil activation also
results in L-selectin and PSGL-1 shedding [24, 25]. Although
L-selectin shedding is thought to limit neutrophil activation
and thus adhesion during inflammation [26, 27], the function of
PSGL-1 shedding has still to be fully elucidated. Gal-1 did not
have any effect on L-selectin levels on naı ¨ve or PAF-stimulated
neutrophils; however, it did appear to enhance PAF-stimulated
PSGL-1 shedding in a concentration-dependent manner. The
mechanism responsible for PSGL-1 shedding is distinct from
that of L-selectin shedding , which may account for the
differing effects on these two adhesion molecules. Although the
full functional impact of PSGL-1 shedding is not clear, it has
been linked to a decrease in leukocyte rolling and adhesion
[24, 28]. The lack of an effect on PSGL-1 shedding in naı ¨ve
neutrophils suggests that this is an unlikely mechanism for
Gal-1 inhibition of neutrophil rolling in the flow chamber. The
loss of the PAF-induced CD11b expression in response to
Gal-1 suggests that Gal-1 may function to limit some aspects of
neutrophil activation. These effects of Gal-1 are in contrast to
another galectin, Gal-3, which has been shown to induce
L-selectin shedding and IL-8 production in naı ¨ve and primed
neutrophils . These opposing effects are not limited to
adhesion molecule expression; Gal-3 has also been demon-
strated to promote neutrophil adhesion to endothelial monolay-
ers in vitro  and to enhance neutrophil apoptosis . The
effects of Gal-1 on adhesion molecule expression were only
apparent at the highest concentration of Gal-1 used (4 ?g/ml),
whereas its effects in the flow chamber were apparent at lower
concentrations. Specific dose-dependent effects of Gal-1 are
not unusual with numerous reports in the literature indicative
of this [32–34]. It is difficult to directly compare the effects of
Gal-1 on adhesion molecule expression in a static, single-cell
system with those on cell recruitment under flow conditions. It
may be that when exposed to shear stress, neutrophils are more
sensitive to lower concentrations of Gal-1 than under static
With this study, we have highlighted another role for Gal-1
with regards to neutrophil recruitment in acute inflammation.
These effects were observed at relatively low concentrations
(2.75–275 nM) of Gal-1 in comparison with those used in other
studies. However, we and others have previously demonstrated
activity of Gal-1 at concentrations as low as 2.75 nM in
inhibiting neutrophil transmigration , T cell adhesion ,
and IL-2 production by T cells . The present study is in line
with others that suggest biphasic, concentration-dependent
effects of Gal-1 [33, 34]. The lowest concentrations of Gal-1
used in the present study resulted in a significant reduction in
all three parameters measured: neutrophil capture, rolling, and
firm adhesion. However, these effects were lost at the highest
concentration of Gal-1 used, with only a significant reduction
in cell rolling apparent at the highest concentration of 4 ?g/ml.
Along with this reduction in neutrophil rolling, there was also
an increase in firm adhesion at this highest concentration,
suggesting that the decreased rolling is not a result of de-
creased neutrophil capture, as observed at lower concentra-
tions, but a result of conversion from a rolling to adherent state,
suggesting that high concentrations of Gal-1 may play a role in
bridging neutrophils to the endothelium, an effect that has
been observed for another galectin, Gal-3 . Gal-1 is known
to exist in a monomer-dimer equilibrium with a Kdof ?7 ?M
. At lower concentrations, the homodimeric protein spon-
taneously dissociates into a monomeric form but retains its
carbohydrate-binding specificity . It is therefore likely that
at the concentrations used in the present study, hrGal-1 was
acting in monomeric form. As alluded to above, the effects of
Gal-1 are often opposing at low versus high concentrations, as
is the case with neutrophil recruitment: low concentrations
inhibiting cell recruitment and high concentrations promoting
recruitment. Low concentrations of Gal-1 have previously been
shown to be mitogenic, whereas high concentrations inhibit
cell growth and/or induce apoptosis [32, 33]. In an inflamma-
tory context, high Gal-1 concentrations have been observed to
trigger macrophage apoptosis and inhibit cytokine generation,
1464Journal of Leukocyte Biology
Volume 83, June 2008
and low concentrations decrease inflammatory macrophage ac-
tivity without affecting viability in response to parasitic infec-
It is becoming increasingly apparent that the effects of Gal-1
are complex, and its molecular mechanism of action is likely
dependent on target cell, concentration applied, and carbohy-
drate dependency. From the present study, it can be concluded
that the inhibitory effects of Gal-1 on neutrophil recruitment
are apparent at low concentrations of Gal-1 and are carbohy-
drate-dependent in nature. The mechanism of action of Gal-1
at different concentrations has still to be elucidated; however,
depending on the concentration of Gal-1 used, it is likely that
it may act by inducing the cross-linking of proteins leading to
inhibition or promotion of a cellular response or by enhancing
or sterically hindering the binding of cell surface receptors
with their ligands, as may be the case when it interacts with
adhesion molecules such as CD43 . With regards to the
effects of Gal-1 on neutrophil recruitment, other than the
implication of an effect in the study of Rabinovich et al. 
and our previous data showing Gal-1 inhibits neutrophil che-
motaxis and transmigration , this has not been addressed.
In determining the effects of Gal-1 on neutrophil recruitment
under flow, we have been able to observe that the specific steps
of neutrophil rolling and firm adhesion are down-regulated by
As endothelial cells, in contrast to neutrophils, express high
amounts of Gal-1 [11, 40], it is attractive to hypothesize that
endogenous Gal-1 presented by the endothelium can act to
limit recruitment of this cell type (even more if primed) during
an ongoing, inflammatory response. The expression of endo-
thelial Gal-1 is increased upon activation by proinflammatory
cytokines  or exposure to tumor cell-conditioned medium
. Little is known, however, about the function of endothelial
cell-derived Gal-1 during inflammation. Overexpression of
Gal-1, induced by conditioned medium from prostate cancer
cells, reduces T cell migration across the endothelial mono-
layer, an effect that can be reversed by antiserum to Gal-1 .
Our data presented here also suggest that endogenous Gal-1
limits neutrophil recruitment with higher numbers of cells
recruited to HUVEC monolayers in which Gal-1 has been
depleted. How knockdown of endothelial Gal-1 would augment
neutrophil recruitment is not yet apparent and clearly warrants
further investigation. We were unable to duplicate the effects
of knocking down endothelial Gal-1 by pretreating endothelial
monolayers with thiodigalactoside, suggesting that merely
blocking exposed Gal-1 on the endothelial surface is not suf-
ficient to promote neutrophil recruitment. As levels of endo-
thelial Gal-1 are not increased upon short-term TNF-? treat-
ment (our unpublished observations), the levels of Gal-1
present in endothelial cells used for the flow assays are equiv-
alent to basal and as such, will not reflect a more chronic
situation where endothelial Gal-1 is increased ; during
such conditions, blocking surface Gal-1 may be more effective.
The implication of a role for endogenous Gal-1 was corrobo-
rated further by the increased leukocyte adhesion and emigra-
tion observed in Gal-1 null mice in response to IL-1?. Lack of
Gal-1 in endothelial cells in vitro or in Gal-1 null mice led to
an enhanced cell recruitment, although different aspects of the
leukocyte recruitment cascade were modulated in these two
models. In vitro, a decrease in endothelial Gal-1 levels led to
an increase in cell capture and rolling with no apparent effect
on firm adhesion. However, in vivo, an increase in firm adhe-
sion and emigration was observed with no apparent effect on
the number of rolling cells. There are various explanations for
these discrepancies. As the flow chamber system consists of
just two cell types, and only one of these has been depleted of
Gal-1 (in fact, human neutrophils express little if any detect-
able Gal-1), it can be deduced that a lack of endothelial Gal-1
results in an increase in the number of neutrophils captured
and rolling on the endothelium. In the in vivo scenario, it is
impossible to rule out the contribution of Gal-1 derived from
cell types other than endothelial cells. It is possible that a lack
of Gal-1, derived from macrophages, for example, may also
affect leukocyte recruitment. As shown for other inflammatory
mediators, such as NO, the source and amount of mediator
produced have important bearings on its actions. Furthermore,
it is difficult to directly compare the events observed in the in
vitro setting of the flow chamber with those in vivo. It is well
known that cells such as erythrocytes and platelets affect how
leukocytes interact with the vessel wall under inflamed and
noninflamed conditions; it cannot be ruled out therefore that in
vivo, another cell type, such as platelets, is also involved in the
increased recruitment of leukocytes to the vessel wall leading
to an apparent effect on different aspects of the recruitment
cascade. To the best of our knowledge, this is the first study
investigating the vascular inflammatory response in Gal-1 null
mice to date. Although a specific role for endothelial-derived
Gal-1 cannot be derived from these results, they are a further
indication that Gal-1 functions to negatively regulate leukocyte
recruitment during acute inflammation, and further studies are
In conclusion, we have revealed here a novel, endogenous,
anti-inflammatory pathway centered on endothelial Gal-1 that
targets neutrophil trafficking in an inflammatory context. Anal-
yses of single-cell or dual-cell systems have shown, by large,
an inhibitory function for hrGal-1 on neutrophil activation,
although the variety of effects observed at distinct concentra-
tions of the protein may indicate the involvement of multiple
receptors and signaling circuits. Irrespective of a specific mo-
lecular mechanism(s), we propose that during inflammation,
modulation of Gal-1 levels on endothelial cells would be part of
a “negative check-point” aimed at limiting the influx of leu-
kocytes into the surrounding tissues, thereby contributing to a
moderation of the inflammatory response.
This work was supported by the Research Advisory Board of
Barts and The London Charity (Nonclinical Fellowship
RAB03/F2 to D. C. and Ph.D. studentship RAB03/MRes to
L. V. N.), the Arthritis Research Campaign (Nonclinical Career
Development Fellowship 18103 to D. C.), and in part by the
William Harvey Research Foundation (M. P.). The Consortium
for Functional Glycomics (grant number GM62116) provided
the breeder founders of the mouse colony. We thank Dr. R.
Cummings (Emory University, Atlanta, GA, USA) for the gen-
erous supply of hrGal-1.
Cooper et al.
Endothelial Galectin-1 and cell trafficking1465
1. Gilroy, D. W., Lawrence, T., Perretti, M., Rossi, A. G. (2004) Inflammatory
resolution: new opportunities for drug discovery. Nat. Rev. Drug Discov. 3,
2. Perretti, M. (1997) Endogenous mediators that inhibit the leukocyte-
endothelium interaction. Trends Pharmacol. Sci. 18, 418–425.
3. Serhan, C. N. (2007) Resolution phase of inflammation: novel endogenous
anti-inflammatory and proresolving lipid mediators and pathways. Annu.
Rev. Immunol. 25, 101–137.
4. Teichberg, V. I., Silman, I., Beitsch, D. D., Resheff, G. (1975) A ?-D-
galactoside binding protein from electric organ tissue of Electrophorus
electricus. Proc. Natl. Acad. Sci. USA 72, 1383–1387.
5. Levi, G., Tarrab-Hazdai, R., Teichberg, V. I. (1983) Prevention and
therapy with electrolectin of experimental autoimmune myasthenia gravis
in rabbits. Eur. J. Immunol. 13, 500–507.
6. Offner, H., Celnik, B., Bringman, T. S., Casentini-Borocz, D., Nedwin,
G. E., Vandenbark, A. A. (1990) Recombinant human ?-galactoside
binding lectin suppresses clinical and histological signs of experimental
autoimmune encephalomyelitis. J. Neuroimmunol. 28, 177–184.
7. Rabinovich, G. A., Daly, G., Dreja, H., Tailor, H., Riera, C. M., Hiraba-
yashi, J., Chernajovsky, Y. (1999) Recombinant galectin-1 and its genetic
delivery suppress collagen-induced arthritis via T cell apoptosis. J. Exp.
Med. 190, 385–398.
8. Santucci, L., Fiorucci, S., Cammilleri, F., Servillo, G., Federici, B.,
Morelli, A. (2000) Galectin-1 exerts immunomodulatory and protective
effects on concanavalin A-induced hepatitis in mice. Hepatology 31,
9. Santucci, L., Fiorucci, S., Rubinstein, N., Mencarelli, A., Palazzetti, B.,
Federici, B., Rabinovich, G. A., Morelli, A. (2003) Galectin-1 suppresses
experimental colitis in mice. Gastroenterology 124, 1381–1394.
10. Toscano, M. A., Commodaro, A. G., Ilarregui, J. M., Bianco, G. A.,
Liberman, A., Serra, H. M., Hirabayashi, J., Rizzo, L. V., Rabinovich,
G. A. (2006) Galectin-1 suppresses autoimmune retinal disease by pro-
moting concomitant Th2- and T regulatory-mediated anti-inflammatory
responses. J. Immunol. 176, 6323–6332.
11. La, M., Cao, T. V., Cerchiaro, G., Chilton, K., Hirabayashi, J., Kasai, K.,
Oliani, S. M., Chernajovsky, Y., Perretti, M. (2003) A novel biological
activity for galectin-1: inhibition of leukocyte-endothelial cell interactions
in experimental inflammation. Am. J. Pathol. 163, 1505–1515.
12. Hirabayashi, J., Ayaki, H., Soma, G., Kasai, K. (1989) Production and
purification of a recombinant human 14 kDa ?-galactoside-binding lectin.
FEBS Lett. 250, 161–165.
13. Hayhoe, R. P., Kamal, A. M., Solito, E., Flower, R. J., Cooper, D., Perretti,
M. (2006) Annexin 1 and its bioactive peptide inhibit neutrophil-endo-
thelium interactions under flow: indication of distinct receptor involve-
ment. Blood 107, 2123–2130.
14. Patel, K. D. (1999) Mechanisms of selective leukocyte recruitment from
whole blood on cytokine-activated endothelial cells under flow conditions.
J. Immunol. 162, 6209–6216.
15. Young, R. E., Thompson, R. D., Nourshargh, S. (2002) Divergent mech-
anisms of action of the inflammatory cytokines interleukin 1-? and tumor
necrosis factor-? in mouse cremasteric venules. Br. J. Pharmacol. 137,
16. Chatterjee, B. E., Yona, S., Rosignoli, G., Young, R. E., Nourshargh, S.,
Flower, R. J., Perretti, M. (2005) Annexin 1-deficient neutrophils exhibit
enhanced transmigration in vivo and increased responsiveness in vitro.
J. Leukoc. Biol. 78, 639–6d46.
17. Perone, M. J., Bertera, S., Tawadrous, Z. S., Shufesky, W. J., Piganelli,
J. D., Baum, L. G., Trucco, M., Morelli, A. E. (2006) Dendritic cells
expressing transgenic galectin-1 delay onset of autoimmune diabetes in
mice. J. Immunol. 177, 5278–5289.
18. Baum, L. G., Blackall, D. P., Arias-Magallano, S., Nanigian, D., Uh, S. Y.,
Browne, J. M., Hoffmann, D., Emmanouilides, C. E., Territo, M. C.,
Baldwin, G. C. (2003) Amelioration of graft versus host disease by
galectin-1. Clin. Immunol. 109, 295–307.
19. Dias-Baruffi, M., Zhu, H., Cho, M., Karmakar, S., McEver, R. P., Cum-
mings, R. D. (2003) Dimeric galectin-1 induces surface exposure of
phosphatidylserine and phagocytic recognition of leukocytes without in-
ducing apoptosis. J. Biol. Chem. 278, 41282–41293.
20. Stowell, S. R., Karmakar, S., Stowell, C. J., Dias-Baruffi, M., McEver,
R. P., Cummings, R. D. (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. Almkvist, J., Dahlgren, C., Leffler, H., Karlsson, A. (2002) Activation of
the neutrophil nicotinamide adenine dinucleotide phosphate oxidase by
galectin-1. J. Immunol. 168, 4034–4041.
22. Repo, H., Rochon, Y. P., Schwartz, B. R., Sharar, S. R., Winn, R. K.,
Harlan, J. M. (1997) Binding of human peripheral blood polymorphonu-
clear leukocytes to E-selectin (CD62E) does not promote their activation.
J. Immunol. 159, 943–951.
23. Vedder, N. B., Harlan, J. M. (1988) Increased surface expression of
CD11b/CD18 (Mac-1) is not required for stimulated neutrophil adherence
to cultured endothelium. J. Clin. Invest. 81, 676–682.
24. Davenpeck, K. L., Brummet, M. E., Hudson, S. A., Mayer, R. J., Bochner,
B. S. (2000) Activation of human leukocytes reduces surface P-selectin
glycoprotein ligand-1 (PSGL-1, CD162) and adhesion to P-selectin in
vitro. J. Immunol. 165, 2764–2772.
25. Kishimoto, T. K., Jutila, M. A., Berg, E. L., Butcher, E. C. (1989)
Neutrophil Mac-1 and MEL-14 adhesion proteins inversely regulated by
chemotactic factors. Science 245, 1238–1241.
26. Hafezi-Moghadam, A., Thomas, K. L., Prorock, A. J., Huo, Y., Ley, K.
(2001) L-selectin shedding regulates leukocyte recruitment. J. Exp. Med.
27. Smalley, D. M., Ley, K. (2005) L-selectin: mechanisms and physiological
significance of ectodomain cleavage. J. Cell. Mol. Med. 9, 255–266.
28. Van Genderen, H., Wielders, S. J., Lindhout, T., Reutelingsperger, C. P.
(2006) Rolling and adhesion of apoptotic monocytes is impaired by loss of
functional cell surface-expressed P-selectin glycoprotein ligand-1. J.
Thromb. Haemost. 4, 1611–1617.
29. Nieminen, J., St-Pierre, C., Sato, S. (2005) Galectin-3 interacts with naive
and primed neutrophils, inducing innate immune responses. J. Leukoc.
Biol. 78, 1127–1135.
30. Sato, S., Ouellet, N., Pelletier, I., Simard, M., Rancourt, A., Bergeron,
M. G. (2002) Role of galectin-3 as an adhesion molecule for neutrophil
extravasation during streptococcal pneumonia. J. Immunol. 168, 1813–
31. Fernandez, G. C., Ilarregui, J. M., Rubel, C. J., Toscano, M. A., Gomez,
S. A., Beigier Bompadre, M., Isturiz, M. A., Rabinovich, G. A., Palermo,
M. S. (2005) Galectin-3 and soluble fibrinogen act in concert to modulate
neutrophil activation and survival: involvement of alternative MAPK path-
ways. Glycobiology 15, 519–527.
32. Adams, L., Scott, G. K., Weinberg, C. S. (1996) Biphasic modulation of
cell growth by recombinant human galectin-1. Biochim. Biophys. Acta
33. Biron, V. A., Iglesias, M. M., Troncoso, M. F., Besio-Moreno, M., Patrig-
nani, Z. J., Pignataro, O. P., Wolfenstein-Todel, C. (2006) Galectin-1:
biphasic growth regulation of Leydig tumor cells. Glycobiology 16, 810–
34. Zuniga, E., Gruppi, A., Hirabayashi, J., Kasai, K. I., Rabinovich, G. A.
(2001) Regulated expression and effect of galectin-1 on Trypanosoma
cruzi-infected macrophages: modulation of microbicidal activity and sur-
vival. Infect. Immun. 69, 6804–6812.
35. Rabinovich, G. A., Ariel, A., Hershkoviz, R., Hirabayashi, J., Kasai, K. I.,
Lider, O. (1999) Specific inhibition of T-cell adhesion to extracellular
matrix and proinflammatory cytokine secretion by human recombinant
galectin-1. Immunology 97, 100–106.
36. Cho, M., Cummings, R. D. (1995) Galectin-1, a ?-galactoside-binding
lectin in Chinese hamster ovary cells. I. Physical and chemical charac-
terization. J. Biol. Chem. 270, 5198–5206.
37. Leppanen, A., Stowell, S., Blixt, O., Cummings, R. D. (2005) Dimeric
galectin-1 binds with high affinity to ?2,3-sialylated and non-sialylated
terminal N-acetyllactosamine units on surface-bound extended glycans.
J. Biol. Chem. 280, 5549–5562.
38. He, J., Baum, L. G. (2006) Endothelial cell expression of galectin-1
induced by prostate cancer cells inhibits T-cell transendothelial migra-
tion. Lab. Invest. 86, 578–590.
39. Rabinovich, G. A., Sotomayor, C. E., Riera, C. M., Bianco, I., Correa, S. G.
(2000) Evidence of a role for galectin-1 in acute inflammation. Eur.
J. Immunol. 30, 1331–1339.
40. Baum, L. G., Seilhamer, J. J., Pang, M., Levine, W. B., Beynon, D.,
Berliner, J. A. (1995) Synthesis of an endogeneous lectin, galectin-1, by
human endothelial cells is up-regulated by endothelial cell activation.
Glycoconj. J. 12, 63–68.
41. Clausse, N., van den Brule, F., Waltregny, D., Garnier, F., Castronovo, V.
(1999) Galectin-1 expression in prostate tumor-associated capillary endo-
thelial cells is increased by prostate carcinoma cells and modulates
heterotypic cell-cell adhesion. Angiogenesis 3, 317–325.
1466 Journal of Leukocyte Biology
Volume 83, June 2008