| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-CD47 Interactions Are Required for the Transmigration of Monocytes Across Cerebral Endothelium1
Departments of
* Molecular Cell Biology and
Pathology, Vrije Universiteit Medical Center, Amsterdam, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-CD47 interactions in monocyte transmigration across brain
ECs. CD47 expression was observed in vivo on cerebral endothelium of
both control animals and animals suffering from experimental allergic
encephalomyelitis. To investigate whether SIRP-CD47 interactions are
instrumental in the trafficking of monocytes across cerebral EC
monolayers, in vitro assays were conducted in which the migration of
monocytes, but not adhesion, was found to be effectively diminished by
blocking SIRP and CD47 on monocytes and ECs, respectively. In this
process, SIRP was found to interact solely with its counterligand
CD47 on ECs. Overexpression of the CD47 molecule on brain ECs
significantly enhanced monocytic transmigration, but did not affect
adhesion. SIRP-CD47-mediated transendothelial migration involved Gi
protein activity, a known signaling component of CD47. Finally,
cross-linking of CD47 on brain ECs induced cytoskeletal reorganization
of the endothelium, a process that was Gi protein independent. These
data provide the first evidence that the interaction of CD47 with its
monocytic counterligand SIRP is of importance in the final step of
monocyte trafficking into the brain, a critical event in the
development of neuroinflammatory diseases. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4
1 and
m2, interacting with
their endothelial ligands VCAM and ICAM, has been described
(3, 4, 5, 6). However, other IgSF members may also participate
in the transendothelial migration of monocytes.
One of the candidate proteins involved in monocyte diapedesis is the
widely expressed CD47, also known as integrin-associated protein.
Structurally, CD47 contains a single extracellular Ig-like domain, five
transmembrane-spanning segments, and a short cytoplasmic tail.
Originally, CD47 was identified in association with the
v3 integrin. CD47 can
interact with thrombospondin (7), and it has a part in
3 integrin-mediated actions such as activation
of oxidative burst, FcR-dependent phagocytosis (8, 9), and
cell spreading and motility (10, 11, 12). Furthermore, CD47
plays an immunoregulatory role, as demonstrated in CD47-deficient mice
that appear to have a defect in neutrophil recruitment and therefore
suffer from increased susceptibility to infections, e.g.,
Escherichia coli peritonitis (13). In vitro
studies indicate that CD47 is involved in neutrophil migration across
peripheral ECs and monocyte migration across epithelial cells
(14, 15, 16). CD47 functions may involve the activation of
multiple signal transduction pathways, and a functional coupling of
CD47 to heterotrimeric Gi proteins, protein kinase C, and the small
GTPase Cdc42 has been demonstrated (10, 17, 18, 19).
Recently, signal-regulatory protein (SIRP), another member of the
IgSF, was identified as a cellular ligand for CD47
(20, 21, 22). SIRP, also called SH2 domain-containing
phosphatase substrate 1 or macrophage fusion receptor, is
exclusively expressed on myeloid cells and neurones
(22, 23, 24). SIRP is composed of three extracellular
Ig-like domains and a cytoplasmic domain carrying four immunoreceptor
tyrosine-based inhibitory motifs, which constitute binding sites for
the Src homology 2 domain-bearing phosphatases 1 and 2.
Therefore, SIRP is suggested to have a negative regulatory function
in, e.g., signaling by tyrosine kinase-dependent receptors (23, 25, 26). Recently, the ligation of SIRP on
macrophages by CD47 on target cells has been shown to prevent their
phagocytosis (27, 28).
Because the high expression of SIRP is merely restricted to
monocytic cells, we suggest that SIRP interacting with CD47 may
contribute to the recruitment of monocytes into tissue during
inflammatory diseases. In this study, we report that SIRP is indeed
involved in the transmigration of monocytes across brain endothelium
through interaction with its counterligand CD47, a molecule that upon
engagement induces signaling events in the endothelium.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ham’s F12 medium, RPMI 1640 medium, endothelial serum-free
medium (with L-glutamine), penicillin-streptomycin,
L-glutamine, FCS, HBSS, and trypsin/EDTA were all obtained
from Life Technologies (Rockville, MD). Pertussis toxin (PTX) in
50% in glycerol, collagenase type I, collagen type IV from calf skin
(0.1% solution in 0.1 N-acetic acid), and goat anti-mouse Ab was
obtained from Sigma (Zwijndrecht, The Netherlands).
Rhodamine-phalloidin and 2',7'bis(2-carboxyethyl)-5(and
6)-carboxyfluorescein acetoxymethyl were obtained from Molecular Probes
(Eugene, OR). IFN-
and IL-1 were a kind gift of U-Cytech
(Utrecht, The Netherlands) and Glaxo Wellcome (Basel, Switzerland),
respectively. The mAb ED9 (anti-SIRP; IgG1 isotype) was
generated in our laboratory and is commercially available from Serotec
(Oxford, U.K.). OX-8, OX-19, and OX-52 mAbs were obtained from Serotec.
OX101 (anti-CD47) IgG and OX41 (anti-SIRP; IgG2a) were
purified from hybridoma supernatants by protein A affinity
chromatography (20, 24).
SIRP-Fc protein generation
SIRP-Fc proteins were constructed and purified from Chinese
hamster ovary cell supernatants, as previously described
(29).
Animals and induction of EAE
Acute experimental allergic encephalomyelitis (EAE) was induced in 8- to 11-wk-old male Lewis rats (200 g) obtained from Harlan (Zeist, The Netherlands), as described before (30). Rats were injected s.c. in hind footpads with 20 µg synthetic myelin basic protein 63–88 peptide, 500 µg Mycobacterium tuberculosis type 37HRa (Difco, Detroit, MI), and 50 µl CFA (Difco) supplemented with PBS to reach a volume of 100 µl. Rats were examined daily (weight and clinical disease), and neurological aberrations were graded from 1 to 5, as described before (30). Clinical disease in EAE animals was apparent between days 10 and 19 after immunization with a maximum clinical score between days 14 and 15. Incidence of EAE was 100%. Animals were housed under standard laboratory conditions with water and food ad libitum. Control animals and EAE animals were sacrificed at the peak of the disease, day 15 postinfection, by CO2 inhalation. Spinal cords were dissected, snap frozen in liquid nitrogen, and stored at -80°C
Immunohistochemistry
Cryostat sections (8 µm thick) were melted onto gelatin-coated
glass slides and dried in containers with silica gel. Slides were fixed
in acetone (10 min), dried, and incubated with mAbs to detect the
expression of SIRP and CD47. Sections were incubated with saturating
concentrations of the mAb OX101 (anti-CD47) and the mAb ED9
(anti-SIRP) (24). As a conjugate, a
peroxidase-conjugated rabbit anti-mouse IgG (DAKO, Tilburg, The
Netherlands) was used. Peroxidase activity was demonstrated by
incubation with 0.5 mg/ml 3.3'-diaminobenzidine-tetrahydrochloride in
Tris buffer containing 0.03%
H2O2. Omission of the
primary mAb served as negative control. Sections were counterstained
with hematoxylin and mounted in Entalan.
Cell culture
The well-characterized immortalized Lewis rat brain EC line GP8.3 served a model for the blood-brain barrier in vitro (31, 32) and was maintained in Ham’s F-12 medium supplemented with 20% FCS (heat inactivated), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. GP8.3 cells were passaged into other flasks, multiwell plates, or eight-chamber slides (Labtek; Nalge Nunc International, Naperville, IL) coated with type I collagen 1/20 diluted in HBSS. GP8.3 cells were detached at 37°C with 2 ml trypsin/EDTA in PBS. Cells were collected in prewarmed culture medium, centrifuged, and resuspended in fresh culture medium. Cultures were grown to confluence at 37°C in 5% CO2, and medium was replaced every other day until the formation of monolayers.
The monocytic cell line NR8383 was obtained from the American Type Culture Collection (Manassas, VA). These nonadherent cells were cultured in RPMI 1640 medium supplemented with 10% FCS (heat inactivated), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin.
Construction of the retroviral construct LZRS-rCD47-IRES-EGFP and retroviral transduction of rat brain ECs
The cDNA of rat CD47 containing a V-G substitution at position 264 in a pcDNA 3.0 vector was a kind gift of E. Vernon-Wilson (School of Pathology, Oxford, U.K.). The complete cDNA was subcloned as an EcoRI fragment into the polylinker of the retroviral vector LZRS-polylinker-IRES-EGFP (33) (obtained from H. Spits, The Netherlands Cancer Institute, Amsterdam, The Netherlands), which is a modified version of the original pBMN-LZRS (34). Correct cloning was confirmed by restriction enzyme analysis. The Phoenix-A packaging cell line (34) (provided by G. Nolan, Stanford University, Stanford, CA) was transfected with the retroviral construct containing the rat CD47-IRES-EGFP or the control construct containing enhanced green fluorescent protein (EGFP) only. Amphotropic retroviral supernatants were harvested, as previously described (33), and cell-free aliquots were stored at -80°C until further use. The rat brain EC line GP8.3 was transduced with either the rat CD47-IRES-EGFP or the control-IRES-EGFP retrovirus using calcium phosphate according to the manufacturer’s instructions (Life Technologies). Transductants were subsequently selected by FACS sorting (FACStarPlus; BD Biosciences, San Jose, CA) on the basis of EGFP expression. Cell surface expression of CD47 was determined by FACS analysis using the OX101 mAb, as described below. FACS-sorted, mock-transduced, GP8.3 cells expressing EGFP only were used as a control in subsequent experiments. Ectopic expression of CD47 and/or EGFP was stable during the whole experimental period (data not shown).
Flow cytometry analysis
For flow cytometric analysis, normal GP8.3 cells were detached by incubation with 1 mg/ml collagenase (type I) in PBS/0.1% BSA for 5 min in the incubator and harvested with a needle (0.8 x 340 mm). Cells were centrifuged and resuspended in PBS/0.1% BSA. The monocytic cell line and brain EC line GP8.3 were incubated with 5 µg/ml OX101, ED9, and OX41, or 25 µg/ml SIRP-Fc in 100 µl PBS in a 96-well plate. CD47 and mock-transduced GP8.3 ECs were also stained using OX101 and were screened for their SIRP-Fc-binding properties. After an incubation of 30 min at 4°C, cells were washed three times with PBS. Ab or protein binding to the cells was detected after incubating with fluorescent-labeled conjugates (Jackson ImmunoResearch Laboratories, West Grove, PA) or rat anti-mouse IgG (F(ab')2)/R-PE (DAKO) or anti-human IgG-biotin and streptavidin/R-PE (Sigma) in PBS/0.1% BSA for 30 min at 4°C. For blocking of SIRP-Fc binding, a preincubation was performed with 20 µg/ml anti-CD47 (OX101). Binding was detected by using FACScan flow cytometry (BD Biosciences) and analyzed using CellQuest software (BD Biosciences).
Monocyte isolation
Female Wistar rats were obtained from Harlan and were kept under standard laboratory conditions with food and water ad libitum. Animals were used at a body weight of 250–350 g. Monocytes were isolated after perfusion of the rat, which was performed essentially according to Scriba et al. (35). Briefly, after anesthesia, the thorax was opened and two canulae (16G and 20G) were inserted in the left and right ventricle, respectively. The vasculature of the rat was perfused with 1 L prewarmed RPMI 1640 medium supplemented with 1% BSA and 20 mM HEPES. PBMCs were isolated as the interphase on a Ficoll-Paque (Amersham Pharmacia Biotech, Uppsala, Sweden) gradient. Monocytes were purified from PBMCs by negative selection using OX-8, OX-19, OX-52, and OX-33 Abs and goat anti-mouse Ig-coated magnetic beads (PerkinElmer, Wellesley, MA), resulting in a yield of about 20 x 106 monocytes per rat. The population consisted of >90% viable monocytes (ED9-positive cells) and <10% lymphocytes. All monocytes expressed the adhesion molecule complement receptor type 3, LFA-1, very late Ag-4, and platelet EC adhesion molecule, as assessed by FACS analysis.
Monocyte migration
The migratory capacity of monocytes to cross a monolayer of
brain ECs was assayed using time-lapse videomicroscopy, as described
previously (32). Briefly, monocytes (5 x
105/ml) were added to 96-well plates containing
nonstimulated or stimulated (48 h with 100 ng/ml IL-1 and 200 U/ml
IFN-) brain EC monolayers. Before migration, cells were preincubated
for 30 min with 10 µg/ml mAb OX101 (anti-CD47) and the
anti-SIRP mAbs ED9 or OX41. Monocyte migration was conducted in
presence of specific Abs, as mentioned. Migration was also performed in
the presence of 25 µg/ml SIRP-Fc protein and 25 µg/ml ED9
Fab.
Data are expressed as the mean and SEM of at least 12 individual wells, and significant differences between groups were determined by two-way ANOVA.
To block intracellular signaling via Gi/Go proteins, brain ECs were pretreated for 2 h with PTX (200 ng/ml), after which cells were extensively washed and kept in culture medium for 2 h before the migration assay.
Monocyte adhesion assay
The involvement of SIRP and CD47 in monocyte adhesion to
monolayers of control or cytokine-treated brain ECs was also
determined, as described previously (32). Freshly isolated
monocytes or NR8383 cells were fluorescently labeled with 1 µM
2',7'bis(2-carboxyethyl)-5(and 6)-carboxyfluorescein acetoxymethyl for
15 min at 37°C in F-12/2% BSA and were subsequently washed with
medium. To block CD47 and SIRP, cells were incubated for 30 min at
4°C with the anti-CD47 mAb OX101 and the anti SIRP mAbs ED9 or
OX41, or with isotype-matched IgGs, all at 10 µg/ml. Activation of
Gi/Go proteins was blocked by pretreatment of the brain ECs with PTX,
as described above, after which cells were used in the adhesion
assay.
Before the adhesion experiment, GP8.3 monolayers were washed twice with
prewarmed F-12 medium supplemented with 0.1% BSA. Subsequently,
fluorescently labeled cells (1 x 106
cells/ml) were added to both nonstimulated and stimulated (for 48
h with 200 U/ml IFN- and 100 ng/ml IL-1) monolayers and were allowed
to adhere for 30 min at 37°C and 5% CO2 in the
absence or presence of blocking Abs or Fab. After the incubation,
nonadherent cells were removed by gently washing the monolayers with
prewarmed F-12/0.1% BSA. The amount of adhered cells was determined by
lysing the cells with 0.1 M NaOH and by measuring the fluorescence
intensity in a Tecan X-Fluorscan (excitation wavelength, 485 nm;
emission wavelength, 535 nm). The number of adhered monocytes was
calculated using a calibration curve with various cell concentrations
ranging from 5 x 103 cells/ml to 1 x
106 cells/ml on monolayers of brain ECs. Data are
expressed as the mean and SEM of at least 12 individual wells, and
significant differences between groups were determined by two-way
ANOVA.
Cross-linking studies and F-actin localization
Before cross-linking experiments, monolayers of cerebral ECs were washed three times with HBSS and then left for 48 h at 37°C, 5% CO2 in serum-free endothelial-specific medium. Endothelial monolayers were incubated for 30 min at 37°C, 5% CO2 in either the absence or presence of 10 µg/ml OX101 (anti-CD47 mAb) in endothelial-specific medium. Cells were washed again three times with HBSS and then incubated with serum-free medium in the presence or absence of 10 µg/ml goat anti-mouse for 30 and 60 min, respectively. Control conditions with only the OX101 or goat anti-mouse were incubated for 30 min. After incubation, cells were washed three times with HBSS and fixated in 4% paraformaldehyde in PBS for 15 min, followed by gentle washing with PBS containing 1% BSA. Cells were subsequently permeabilized in 0.25% Triton X-100/PBS for 5 min and blocked for 30 min with PBS containing 10% FCS. Subsequently, cells were incubated with 0.1 µg/ml rhodamine-phalloidin for 1 h and then exhaustively washed in PBS containing 1% BSA. Cells were mounted in Fluostab embedding medium (ICN Biomedicals, Costa Mesa, CA) and viewed on an Eclipse E800 (Nikon, Badhoevedorp, The Netherlands) fluorescence microscope.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and its counterligand CD47 in this process
was studied.
Expression of CD47 and SIRP on spinal cords of EAE animals
The neuroinflammatory disease EAE, the animal model for multiple
sclerosis, was induced in Lewis rats. At the peak of the disease (i.e.,
day 15 after induction of EAE), animals were sacrificed, lumbar spinal
cords were dissected, and cryostat sections were stained with mAbs
directed against SIRP and CD47, ED9 and OX101, respectively. Lumbar
spinal sections of control animals were also stained for the expression
of CD47 and SIRP. Sections only incubated with the
peroxidase-conjugated secondary rabbit anti-mouse IgG showed no
immunoreactivity in control and EAE animals (data not shown).
CD47 was predominantly expressed at the level of the CNS capillaries,
and especially on the CNS ECs, in both control and EAE animals (Fig. 1
). No clear difference in
cerebrovascular CD47 expression between EAE animals and control animals
could be detected. Additionally, in EAE animals CD47 was also expressed
on infiltrated cells in the perivascular cuffs. SIRP was found to be
clearly expressed in perivascular infiltrates, but was not detected on
capillaries of either control or EAE animals (Fig. 1). Additionally,
SIRP was also expressed on neurones, as described previously (data
not shown) (24).
|
and CD47-dependent binding
properties of SIRP
To quantify expression levels of SIRP, and CD47 on the brain
ECs and monocytes in vitro, FACS analysis was performed. The monocytic
cell line NR8383 expressed high levels of SIRP, and CD47 compared
with their conjugate control (Fig. 2A). Brain ECs express high
levels of CD47 and little or no SIRP. Upon stimulation with the
proinflammatory cytokines IL-1 and IFN-, no significant increase
in the expression of CD47 and SIRP could be observed (Fig. 2A).
|
binding was determined using
SIRP-Fc fusion proteins (Fig. 2
B). SIRP-Fc proteins
were capable of binding to brain ECs, and the binding increased
slightly upon stimulation of the cerebral ECs with IL-1 and IFN-,
consistent with the constitutive CD47 expression on cytokine-stimulated
brain ECs. Protein binding could be prevented completely by
preincubation with the anti-CD47 mAb OX101, indicating SIRP is
the major ligand for CD47 expressed on brain ECs (Fig. 2B).
CD47-SIRP mediate monocyte transendothelial migration
Monocyte trafficking across monolayers of brain EC was assessed by
time-lapse video microscopy after 4 h of migration (Fig. 3). To identify whether CD47 and SIRP
contribute to monocyte transendothelial migration, the anti-CD47
mAb OX101 or anti-SIRP mAb ED9 was used. In the presence of
OX101, the migration of freshly isolated monocytes was reduced by
46 ± 6.6%, indicating that CD47 is involved in the migration
process (Fig. 3A). Moreover, in the presence of
anti-SIRP mAb ED9, the number of migrated cells was reduced by
63 ± 9.2%, indicating that SIRP also has a part in monocyte
migration. The anti-SIRP mAb OX41, recognizing a different
epitope than ED9 (24), had no significant blocking effect
on the migration process. In the presence of SIRP-Fc protein (25
µg/ml) and ED9 Fab (25 µg/ml), transendothelial migration of
freshly isolated monocytes was reduced by 33 ± 2.6% and by
39 ± 6.5%, respectively. Monocyte migration was not affected in
the presence of isotype-matched control Abs (data not shown).
|
and
IFN- increased 2-fold (Fig. 3B). Anti-CD47 (OX101) and
anti-SIRP (ED9) inhibited the migration of monocytes across both
control and cytokine-activated brain ECs to a similar degree, which was
comparable with that of freshly isolated monocytes. Again, in the
presence of anti-SIRP mAb OX41, no significant reduction could
be observed. Migration of freshly isolated rat monocytes and NR8383
cells in the presence of both the mAbs OX101 and ED9 revealed no
further decrease in the number of migrating cells, and the percentage
of inhibition was comparable with migration in the presence of ED9
alone (data not shown).
To further demonstrate the importance of CD47 in the transendothelial
migration, we transfected the brain EC line GP8.3 with the
LZRS-IRES-EGFP retroviral vector containing the rat CD47 construct, as
analyzed by their EGFP fluorescence. After FACS sorting, staining of
these cells with OX101 resulted in a 100-fold overexpression of the
CD47 as compared with GP8.3 cells transduced with the mock
LZRS-IRES-EGFP retroviral vector (geomean values; data not shown). The
transendothelial migration of NR8383 cells across CD47-overexpressing
endothelial monolayers increased 1.8-fold, whereas monocyte migration
across mock-transduced GP8 cells was at similar levels compared with
nontransduced GP8.3 cells (Fig. 3C). Monocyte passage over
monolayers of GP8.3 cells ectotropically overexpressing CD47 could be
blocked in the presence of OX101 (10 µg/ml) and by ED9 (10 µg/ml)
(Fig. 3C).
Monocyte adhesion to cerebral EC is not influenced by SIRP and
CD47
Adhesion assays were conducted to determine whether impaired
monocyte migration observed after the intervention of the CD47
interaction with SIRP was accompanied by a different adhesion
profile. Brain ECs and freshly isolated monocytes were pretreated with
blocking mAbs directed against SIRP and CD47, respectively. As shown
in Fig. 4, there were no significant
effects on monocyte adhesion to brain ECs, with the possible exception
of the anti-SIRP mAb ED9 that gave a slight inhibition of
monocyte adhesion to cytokine-activated endothelium. In agreement,
adhesion of the monocytic NR8383 cells to control or
cytokine-stimulated monolayers of cerebral ECs was not significantly
affected by mAbs directed against SIRP and CD47 (data not shown).
Furthermore, no induction of monocyte adhesion to the
CD47-overexpressing nor the mock-transduced GP8 cells was found (data
not shown).
|
Because CD47 is known to exert biological effects via Gi proteins
(18), monocyte migration was performed after blocking Gi
protein-dependent signaling by pretreatment of the cerebral ECs with
PTX. Migration of freshly isolated rat monocytes (Fig. 5A) as well as NR8383 cells
(data not shown) across cerebral ECs was significantly reduced upon
treatment of the ECs with PTX. Similarly, monocyte transmigration was
also inhibited in the CD47-overexpressing brain ECs. In the presence of
anti-SIRP mAb ED9, no additional inhibition of the migration
could be observed, indicating that SIRP-CD47-mediated transmigration
depends on Gi activity. Monocyte adhesion to the endothelium was not
influenced by the pretreatment of the cerebral ECs with PTX (Fig. 5B), indicating that only the transmigration of monocytes
across CNS-ECs requires the active participation of Gi proteins in
brain ECs.
|
A number of studies indicate that adhesion molecules may transduce
signals that lead to cytoskeletal reorganization in the brain EC, hence
facilitating cellular trafficking (36, 37). To investigate
whether the activation of CD47 induces such effects, the cross-linking
effects of CD47 in brain ECs were determined. A clear rearrangement of
F-actin leading to the formation of stress fibers was observed after
the cross-linking of CD47 with the secondary Ab (Fig. 6). Within 30 min 50% of the cells, and
after 60 min >80% of the brain EC, revealed the formation of stress
fibers induced by CD47 activation. Incubation of the brain EC with the
anti-CD47 mAb or secondary Ab alone showed no significant change in
F-actin staining (Fig. 6). The brain ECs were also subjected to
activation of CD47 by 25 µg/ml SIRP-Fc protein after cross-linking
with the secondary anti-human Fc Ab. Ligation of CD47 in this way
also induces the formation of stress fibers in 50% of the cells after
60 min (data not shown).
|
). These
findings suggest that CD47 is capable of inducing cytoskeletal changes
in the brain EC, which appear to be independent of Gi protein
activation. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expressed on
myeloid cells and the endothelial CD47 contributes to the
transmigration of monocytes across monolayers of brain EC, but that
this interaction is not contributing to monocyte adhesion to
ECs. In our study, CD47 appeared to be highly expressed on brain ECs, and its expression is not significantly increased upon treatment of the cells with proinflammatory cytokines, as observed earlier (14). Moreover, our observations indicate that the CD47 molecule is highly and constitutively expressed in vivo on CNS capillary endothelium of control animals and animals suffering from EAE.
Blocking the interaction between the CD47 molecule and its monocytic
counterligand SIRP in vitro leads to a significantly diminished
monocyte migration across the blood-brain barrier ECs, whereas their
firm adhesion was not affected in our assays. This suggests that
CD47-SIRP interactions are only required in the final postadhesion
step of the migration process, which is consistent with our
observations that CD47 is localized close to the tight junctions in
these brain EC monolayers (data not shown). Also, overexpression of
CD47 in the brain EC line significantly enhanced monocyte
transmigration, without affecting monocytic adhesion to these cells.
CD47 has also been shown to contribute to the spontaneous migration of
monocytes across the alveolar epithelial barrier (15, 16).
However, in this study, no role for the counterreceptor of CD47, the
myeloid protein SIRP, has been investigated. A recent study
suggested that the interaction between CD47 expressed on T cells and
SIRP expressed on HUVECs can also mediate the constitutive arrest of
T lymphocytes on inflamed endothelium by up-regulation of
41 integrins on T
cells (38). Although the monocytes used in our study also
express CD47, preincubation of monocytes with the blocking
anti-CD47Ab could not interfere with their migration, suggesting
that such an inverted event may not be involved in monocyte
transmigration. The role of SIRP-CD47 interaction in this
postadhesion event is unique among that of other adhesion molecules
involved in monocyte transmigration across brain endothelium. Previous
studies in our laboratory have shown that adhesion molecules such as
very late Ag-4/VCAM-1 and complement receptor type 3/ICAM-1
mediate the firm adhesion to as well as the transmigration of monocytes
across brain endothelium (41).
CD47 is a widely expressed molecule that can mediate a variety of
actions by triggering different transduction pathways (for review, see
Ref. 12), among which are the Gi proteins. Indeed, we
showed that only the final step in the CD47-SIRP transendothelial
migration process was shown to be fully dependent on Gi/Go protein
activity. Further analysis of the CD47-mediated activation of Gi
proteins and the role of this in the transmigration process will be
necessary.
The transendothelial migration of monocytes likewise requires the
active participation of the brain ECs to facilitate monocyte passage
(32). Recently, evidence has emerged that adhesion
molecules can mediate various intracellular signal transduction
pathways and can interact with cytoskeletal elements (36, 37). Until now, little is known regarding the effect of CD47
activation on ECs derived from brain capillaries. In this study, CD47
was activated by cross-linking with the OX101 mAb to mimic monocyte
attachment via SIRP. A clear reorganization of the cytoskeleton of
the cerebral ECs was observed, as shown by the formation of stress
fibers, indicating the activation of signaling cascades. The induced
changes by CD47 activation were found to be independent of Gi proteins,
and suggest the involvement of the small GTPases, because these can
actively regulate cytoskeletal changes (39). In other cell
types, CD47 has already been reported to functionally couple to
pathways that may regulate cytoskeletal organization, like Cdc42 of the
Rho-GTPases family in B cells (17). Recently, the ligation
of CD47 on T cells has been reported to lead to actin polymerization
(40).
Taken together, we show that the CD47-SIRP interaction contributes
to the transendothelial migration of monocytes into the brain under
inflammatory conditions. In contrast, CD47 is also capable of
activating other Gi-independent signaling pathways that mediate
cytoskeletal changes. Interference with CD47-SIRP interaction may
offer a novel approach to selectively suppress monocyte infiltration
into the brain during neuroinflammatory diseases such as multiple
sclerosis, thereby decreasing the clinical symptoms.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Helga E. de Vries, Department of Molecular Cell Biology, Vrije Universiteit Medical Center, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. E-mail address: HE.de_Vries.Cell{at}med.vu.nl
3 Abbreviations used in this paper: IgSF, Ig superfamily; EAE, experimental allergic encephalomyelitis; EC, endothelial cell; EGFP, enhanced green fluorescent protein; PTX, pertussis toxin; SIRP, signal-regulatory protein.
Received for publication December 27, 2001. Accepted for publication March 25, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
v3 integrin ligands to promote melanoma cell spreading and focal adhesion kinase phosphorylation. J. Biol. Chem. 274:22755.V3. J. Biol. Chem. 275:4796.v3, integrin-associated protein (CD47), and heterotrimeric G proteins. J. Cell Biol. 146:673.. Eur. J. Immunol. 30:2130.[Medline]
(SIRP) regulates Fc and complement receptor-mediated phagocytosis. J. Exp. Med. 193:855. determines its cellular binding-specificity. J. Cell Sci. 114:1321.[Abstract]
prevents or aggravates experimental autoimmune encephalomyelitis in Lewis rats. J. Immunol. 157:5721.[Abstract]
translocation in T cells. J. Biol. Chem. 276:7672. directly influences monocyte infiltration into the
central nervous system. J Neuroimmunol. In press.This article has been cited by other articles:
|
|
 |
|
  M. Stefanidakis, G. Newton, W. Y. Lee, C. A. Parkos, and F. W. Luscinskas Endothelial CD47 interaction with SIRP{gamma} is required for human T-cell transendothelial migration under shear flow conditions in vitro Blood, August 15, 2008; 112(4): 1280 - 1289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Su, M. Johansen, M. R. Looney, E. J. Brown, and M. A. Matthay CD47 Deficiency Protects Mice from Lipopolysaccharide-Induced Acute Lung Injury and Escherichia coli Pneumonia J. Immunol., May 15, 2008; 180(10): 6947 - 6953. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Schreibelt, G. Kooij, A. Reijerkerk, R. van Doorn, S. I. Gringhuis, S. van der Pol, B. B. Weksler, I. A. Romero, P.-O. Couraud, J. Piontek, et al. Reactive oxygen species alter brain endothelial tight junction dynamics via RhoA, PI3 kinase, and PKB signaling FASEB J, November 1, 2007; 21(13): 3666 - 3676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Rao, L. Yang, G. Garcia-Cardena, and F. W. Luscinskas Endothelial-Dependent Mechanisms of Leukocyte Recruitment to the Vascular Wall Circ. Res., August 3, 2007; 101(3): 234 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Subramanian, E. T. Boder, and D. E. Discher Phylogenetic Divergence of CD47 Interactions with Human Signal Regulatory Protein {alpha} Reveals Locus of Species Specificity: IMPLICATIONS FOR THE BINDING SITE J. Biol. Chem., January 19, 2007; 282(3): 1805 - 1818. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Reijerkerk, G. Kooij, S. M. A. van der Pol, S. Khazen, C. D. Dijkstra, and H. E. de Vries Diapedesis of monocytes is associated with MMP-mediated occludin disappearance in brain endothelial cells FASEB J, December 1, 2006; 20(14): 2550 - 2552. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Schreibelt, R. J. P. Musters, A. Reijerkerk, L. R. de Groot, S. M. A. van der Pol, E. M. L. Hendrikx, E. D. Dopp, C. D. Dijkstra, B. Drukarch, and H. E. de Vries Lipoic Acid Affects Cellular Migration into the Central Nervous System and Stabilizes Blood-Brain Barrier Integrity J. Immunol., August 15, 2006; 177(4): 2630 - 2637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Lahoud, A. I. Proietto, K. H. Gartlan, S. Kitsoulis, J. Curtis, J. Wettenhall, M. Sofi, C. Daunt, M. O'Keeffe, I. Caminschi, et al. Signal Regulatory Protein Molecules Are Differentially Expressed by CD8- Dendritic Cells J. Immunol., July 1, 2006; 177(1): 372 - 382. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Hagnerud, P. P. Manna, M. Cella, A. Stenberg, W. A. Frazier, M. Colonna, and P.-A. Oldenborg Deficit of CD47 Results in a Defect of Marginal Zone Dendritic Cells, Blunted Immune Response to Particulate Antigen and Impairment of Skin Dendritic Cell Migration J. Immunol., May 15, 2006; 176(10): 5772 - 5778. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. van Beek, F. Cochrane, A. N. Barclay, and T. K. van den Berg Signal Regulatory Proteins in the Immune System J. Immunol., December 15, 2005; 175(12): 7781 - 7787. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Sen, S. Subramanian, and D. E. Discher Indentation and Adhesive Probing of a Cell Membrane with AFM: Theoretical Model and Experiments Biophys. J., November 1, 2005; 89(5): 3203 - 3213. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Liu, I. Soto, Q. Tong, A. Chin, H.-J. Buhring, T. Wu, K. Zen, and C. A. Parkos SIRP{beta}1 Is Expressed as a Disulfide-linked Homodimer in Leukocytes and Positively Regulates Neutrophil Transepithelial Migration J. Biol. Chem., October 28, 2005; 280(43): 36132 - 36140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Alblas, H. Honing, C. Renardel de Lavalette, M. H. Brown, C. D. Dijkstra, and T. K. van den Berg Signal Regulatory Protein {alpha} Ligation Induces Macrophage Nitric Oxide Production through JAK/STAT- and Phosphatidylinositol 3-Kinase/Rac1/NAPDH Oxidase/H2O2-Dependent Pathways Mol. Cell. Biol., August 15, 2005; 25(16): 7181 - 7192. |
