| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology and Cell Biology, Forschungszentrum Borstel, Borstel, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-granule compartment of blood
platelets and that become liberated immediately after wounding in the
course of platelet activation. Two major CXC chemokines have been found
in platelets, a group of homologous proteins comprised under the term
-thromboglobulin Ag (-TG Ag) and platelet factor 4 (CXCL4) (for a
review, see Ref. 10). Both are secreted simultaneously by
activated platelets, and their normal serum concentrations are fairly
high, amounting to 1.6–4.8 and 0.4–1.9 µM, respectively. However,
several peculiarities with these platelet-derived chemokines render it
difficult to predict how they will actually influence
neutrophil-endothelial cell interaction. Thus, the -TG Ag proteins
are not secreted as active chemokines, but appear as N-terminally
elongated precursors, the quantitatively prevailing of which is the
connective tissue-activating peptide III (CTAP-III). Proteolytic
truncation of CTAP-III at the N terminus converts the precursor into
the active chemokine neutrophil-activating peptide 2 (CXCL7), a process
that occurs most rapidly (within minutes) in the presence of
neutrophils and appears to be catalyzed by a surface-bound, cathepsin
G-like enzyme on these cells (11). As a consequence,
CTAP-III-processing neutrophils are exposed to continuously increasing
concentrations of CXCL7. As we have shown previously, this does not
activate the processing cells, but results in their functional
desensitization toward subsequently administered higher dosages of
CXCL7 and to other ELR+ CXC chemokines. This
phenomenon is most likely due to down-regulation of chemokine receptors
CXCR-2 and possibly also CXCR-1 by the CXCL7 generated during
processing (12, 13). In contrast, CXCL7 has been shown to
stimulate neutrophils for chemotaxis over an extremely wide range of
stimulus concentrations (14), for the degranulation of
lysosomal enzymes (15), and to up-regulate adhesion
molecules on these cells (16). To date, nothing is known about the impact of CXCL7 and its precursors on neutrophil-endothelial cell interaction, while at least some information exists on CXCL4, the second major platelet chemokine. Representing an ELR- CXC chemokine, CXCL4 neither interacts with receptors CXCR-1 or CXCR-2 on neutrophils nor was it found to induce chemotaxis or lysosomal degranulation in these cells (17, 18). As we have previously shown, CXCL4 rather binds to a cell surface-expressed chondroitin sulfate proteoglycan and induces firm adhesion of neutrophils to cultured HUVEC (19). Notably, the neutrophil-expressed adhesion molecules involved by CXCL4 (the integrin LFA-1 and L-selectin) are different from those that become activated by classic ELR+ CXC chemokines such as CXCL8 (the integrin macrophage-1 Ag (MAC-1)), which may explain the higher resistance to shear force of CXCL4-stimulated cell adhesion as compared with that induced by CXCL8. Furthermore, in adherent neutrophils CXCL4 induces exocytosis of secondary granule contents, while CXCL8 is inactive in this respect (19). To date, nothing is known as to whether CXCL4 has any impact on neutrophil transendothelial migration.
To address their potential roles as early mediators of neutrophil recruitment, we investigated the impact of CXCL4, CTAP-III, and its derivative CXCL7 on neutrophil adhesion to and migration through monolayers of cultured HUVEC. In the present study, all experiments were conducted with unstimulated HUVEC under static conditions to mimick conditions likely to exist at the onset of inflammation immediately after platelet activation, i.e., in the absence of proinflammatory cytokines and after interruption of blood flow due to thrombus formation. We present data demonstrating that all of the above platelet-derived chemokines promote neutrophil adhesion to HUVEC, while only CXCL7 induces transendothelial migration. Most interestingly, CXCL4 appears to function as a regulator of cell adhesion induced by CXCL7 and CTAP-III, while CTAP-III significantly affects CXCL7-dependent transmigration. Altogether our results suggest that platelet-derived chemokines may act in concert to organize early neutrophil recruitment.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human natural CTAP-III, CXCL7, and CXCL4 were prepared in our
laboratory from release supernatants of purified thrombin-stimulated
platelets, as described previously. Briefly, CTAP-III (together with
other variants of -TG Ag) was absorbed by immunoaffinity
chromatography and then purified to homogeneity using sequential cation
exchange (11) and reversed-phase HPLC (20).
CXCL7 was then obtained by limited digestion of homogeneous CTAP-III
with chymotrypsin and purified by reversed-phase HPLC
(14). CXCL4 was isolated from the flow-through of the
immunocolumn obtained after absorption of -TG Ag and then further
purified by sequential heparin-Sepharose and reversed-phase
chromatography (17). Human rCXCL8 bearing a His tag at the
N terminus was produced in Escherichia coli according to
previously described methods (21), the His tag being
removed by digestion with endoproteinase ArgC to yield the 72-aa form
of the chemokine. All chemokine preparations exceeded 99% purity
according to overloaded silver-stained SDS-PAGE and automated
N-terminal sequence analysis.
Antibodies
Murine mAb against CXCR-2 (clone RII115) was generated in our
laboratory, as described previously (14), while mAb
against CXCR-1 (clone SE-2) was kindly provided by O. Götze
(University of Göttingen, Göttingen, Germany). mAbs
directed against CD18 (clone MHM23), CD11a (clone MHM24), and CD11b
(clone 2LPM19C) as well as murine isotype control Ab IgG2b were all
purchased from DAKO (Hamburg, Germany). Murine mAb anti-CD62L
(clone Dreg56) was obtained from Coulter-Immunotech (Hamburg, Germany).
A murine mAb against human IL-2 (clone B0-7) (22) served
as an IgG1 isotype control. A rabbit antiserum (R--TG) reacting
to all known variants of -TG Ag was raised in our laboratory against
a purified preparation of native -TG Ag.
Preparation and culture of human neutrophils and HUVEC
Neutrophils were routinely isolated from citrated blood of single healthy donors by gradient centrifugation on Ficoll-Hypaque to purity greater than 95% in all events, as previously described (11). Human endothelial cells were isolated from umbilical cord veins by collagenase treatment and cultured in dishes precoated with 0.04% gelatin, according to Jaffe et al. (23), with minor modifications as described by Petersen et al. (19). Briefly, the cells were maintained in HUVEC culture medium consisting of M199 supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine (all from Biochrom, Berlin, Germany), 10% FCS, 40 µg/ml endothelial cell growth factor (Boehringer Mannheim, Mannheim, Germany), and 50 µg/ml heparin (Sigma-Aldrich, Deisenhofen, Germany). Cells were subcultured after trypsinization (0.5% trypsin solution, supplemented with 0.2% EDTA; Biochrom) and used throughout passages 2 to 4.
Adhesion assay
Neutrophil adhesion to HUVEC was quantified, as described
previously (19). In brief, HUVEC were cultured in
microtiterplates for 2–4 days to form a monolayer. Cells were then
washed twice with warm (37°C) Dulbecco’s PBS (PBS-D)/0.1% BSA (low
endotoxin BSA; Serva, Heidelberg, Germany) supplemented with 0.9 mM
CaCl2 and 0.5 mM MgCl2.
Immediately after receiving stimuli, 150-µl aliquots of neutrophils
(2 x 105 cells) in the same buffer were
added to washed HUVEC. Neutrophils were allowed to adhere to HUVEC for
25 min at 37°C, and nonadherent cells were then pelleted to one edge
of the wells by centrifuging the plates (45° angle, 200 x
g, 1 min, at room temperature (RT)) and removed by
aspiration. Adherent cells were lysed and quantified by measurement of
neutrophil-specific endogenous -glucuronidase enzymatic activity
using p-nitrophenyl--glucuronide (Sigma-Aldrich) as a
substrate (24). Cell numbers were calculated by means of a
standard of lysed cells run in parallel. In some experiments,
stimulation with the chemokines described above was performed in the
presence of various Abs directed to adhesion molecules, as indicated in
the text. In experiments in which neutrophils were treated with
aprotinin or Abs to chemokine receptors, the respective cells were
preincubated with these reagents for 10 or 30 min, and were then added
to HUVEC, as described above.
Transmigration assay
Transendothelial migration was quantified using the Transwell
system (Costar, Cambridge, MA). HUVEC at 5 x
104 cells/insert were seeded in 100 µl HUVEC
culture medium on 0.1% gelatin-coated membranes (6.5 mm diameter,
polycarbonate membrane with 5-µm pores), and were placed in 24-well
culture plates (Costar), containing 600 µl/well HUVEC medium, which
was replaced after the first day of culture. HUVEC were cultured for
3–5 days at 37°C and 5% CO2 to form a tight
monolayer. Confluence was determined by measuring permeability to
FITC-labeled albumin. Therefore, HUVEC monolayer was washed twice with
warm (37°C) PBS-D/Ca/Mg (PBS-D containing 0.9 M
CaCl2 and 0.5 M MgCl2).
Then 100 µl FITC-labeled albumin in PBS-D was added into the insert
that had been placed onto a new 24-well culture plate containing 600
µl PBS-D/Ca/Mg/well. After 60 min of incubation at 37°C and 5%
CO2, the FITC-labeled albumin in the lower
chamber was photometrically monitored at 495 nm. Diffusion of albumin
was expressed as a percentage of equilibrium, which was simulated by
mixing 100 µl FITC-labeled albumin with 600 µl PBS-D/Ca/Mg.
Transwells were used when albumin diffusion was generally <5% of
equilibrium. To perform the transmigration assay, HUVEC monolayers were
washed with assay buffer (M199 containing 0.1% low endotoxin BSA) and
incubated for 60 min in assay buffer before each assay. Stimuli were
diluted in 600 µl assay buffer to various concentrations and placed
in the 24-well culture plate. HUVEC-covered filters were transferred
onto the stimuli after assay buffer was removed. Immediately afterward,
neutrophils at 5 x 105 cells in 100 µl assay
buffer were added to each insert. In some experiments, neutrophils were
pretreated at 37°C for 10 min with different concentrations of
chemokines or for 30 min with mAb against CD62L before added to the
insert. After incubation at 37°C in 5% CO2 for
45 min, the Transwell inserts were removed by gently scraping the
bottom surfaces of the filters several times against the well edge to
dislodge the cells adhering to the under surface of the endothelial
cell-covered filters, and to collect them in the lower chamber.
Migrated cells were lysed in assay buffer containing 0.1% Triton
X-100, and -glucuronidase enzymatic activity was measured, as
described for the neutrophil adhesion assay above. The number of
migrated cells was calculated from a standard of lysed cells run in
parallel.
Degranulation assay
Neutrophils (1 x 107/ml) suspended in PBS-D/0.1% BSA (low endotoxin) were preincubated for 10 min with 5 µg/ml cytochalasin B (Sigma-Aldrich) and supplemented with CaCl2 (1.8 mM) and MgCl2 (1 mM). Then 100 µl cell suspension was added to 100 µl samples preheated at 37°C. After 30 min of incubation, cells were sedimented and supernatants were assayed for their elastase enzymatic activity, as previously described (13). Release rates were determined as percentage of total elastase activity in neutrophil lysate obtained with 0.1% hexadecyl-trimethylammoniumbromide. Backgrounds, as determined in the presence of buffer alone, were subtracted.
Gel electrophoreses and immunoblotting
SDS-PAGE was performed according to Schägger and von Jagow
(25). Samples were reduced with 1 mM DTT for 60 min at RT,
treated for 30 min with 2% iodoacetamide at RT, and loaded onto a 13%
polyacrylamide gel with a 10% spacer gel and a 4% stacking gel on
top. Rainbow Protein Markers (Amersham Buchler, Braunschweig, Germany)
served as molecular mass markers. Western blotting was conducted as
described previously (26). Briefly, protein bands from the
gels were electrophoretically transferred onto polyvinylidene
difluoride membranes (Immobilon P; Millipore, Eschborn, Germany), and
-TG Ag polypeptides were immunochemically detected with a primary
antiserum R--TG Ag (see above). Bound Abs were detected by using
peroxidase-conjugated goat anti-rabbit IgG (Dianova, Hamburg,
Germany) and the diaminobenzidine reaction.
Processing of CTAP-III by HUVEC and neutrophils
HUVEC were cultured in microtiterplates to form a monolayer.
Before the assay, HUVEC monolayers (
3–4 x
104 cells/well) were washed twice with assay
buffer (PBS-D/0.1% BSA (low endotoxin) supplemented with 0.9 mM
CaCl2 and 0.5 mM MgCl2),
and then 100 µl aliquots of 1 µM CTAP-III in the same buffer were
added alone or in combination with neutrophils (2 x
105 cells) and incubated for 1 h at 37°C.
In parallel, CTAP-III was also added to neutrophils alone (2 x
105 cells in 100 µl assay buffer). The
recovered supernatants were analyzed for neutrophil-stimulating
activity in the degranulation assay and for the presence of CXCL7
protein by SDS-PAGE and immunoblotting (see above).
Flow cytometric analysis of CXCR-1 and CXCR-2 expression on neutrophils
Flow cytometric (FACS) analyses were performed as described (14). To determine the impact of CTAP-III and other chemokines on CXCR-1 and CXCR-2 cell surface expression, neutrophils (106/ml) suspended in PBS-D/0.1% BSA (low endotoxin) were first incubated with CTAP-III, CXCL7, CXCL8, or CXCL4, or left untreated for 10 min at 37°C. Subsequently, cells were incubated with mAb SE-2 against CXCR-1 (5 µg/ml) or mAb RII115 against CXCR-2 (2 µg/ml) for 1 h on ice, labeled with fluorescein-conjugated goat anti-mouse IgG (Dianova, Hamburg, Germany) at a final concentration of 15 µg/ml, and analyzed in a flow cytometer (model FACSCalibur; BD Biosciences, Heidelberg, Germany). Murine Ab IgG2b (5 µg/ml) was used as isotype control (DAKO).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have recently found that CXCL4, a major CXC chemokine released
by activated platelets, induces firm adhesion of neutrophils to
unstimulated HUVEC (19). In the current study, we were
interested in whether other platelet-derived CXC chemokines, in
particular the quantitatively prevailing CTAP-III and its N-terminally
truncated derivative CXCL7, would also affect neutrophil adhesion to
endothelial cells. To examine this, monolayers of cultured HUVEC were
incubated with isolated neutrophils in the presence and absence of
increasing concentrations of the chemokines. For comparison, CXCL8 and
CXCL4 were tested for their adhesion-inducing capacity in parallel. As
shown in Fig. 1
, CXCL7 and surprisingly
also CTAP-III induced neutrophil adhesion to HUVEC in a dose-dependent
manner. Within the concentration range tested, the chemokines showed
similar efficacies (32 ± 5.0% and 31 ± 8.2% of
specifically adhering neutrophils in response to CXCL7 and CTAP-III,
respectively), while their potencies differed considerably. In fact,
while 3.5 nM CXCL7 was sufficient to induce half-maximal cell adhesion,
a 100-fold higher dosage of CTAP-III (EC50 = 350
nM) was required. In this respect, CXCL7 behaved more similar to CXCL8
(EC50 
1 nM), whereas the potency of CTAP-III
was more comparable with that of CXCL4 (EC50
1000 nM). However, as a most characteristic common feature
discriminating them from CXCL8 as well as from CXCL4, both CXCL7 and
CTAP-III were active over an extremely wide range of stimulus
concentrations and exhibited very similar slopes of their dose-response
curves (see Fig. 1). The finding that apart from CXCL7 also its
precursor CTAP-III induces neutrophil adhesion to HUVEC is a most
unexpected one, because up to now, the latter chemokine was not found
to activate neutrophil functions, but was rather described to
desensitize these cells.
|
The above observation that not only CXCL7, but also its
N-terminally prolonged precursor CTAP-III was able to promote
neutrophil adhesion to HUVEC raised questions as to the mechanism(s)
underlying this phenomenon. Because neutrophils reportedly express no
specific receptors for CTAP-III (12), we suspected that
CXCL7, generated from CTAP-III through proteolytic processing, could be
responsible for the induction of cell adhesion. To examine this, we
first looked for the presence of CXCL7 protein and CXCL7 biological
activity in the supernatants of neutrophils coincubated with HUVEC in
the presence of CTAP-III (100 nM, 1 µM, 10 µM) under exactly the
same conditions as described for the adhesion assay. Separation of
supernatants derived from incubations with 1 or 10 µM CTAP-III by
SDS-PAGE, subsequent Western blotting, and immunodetection using
R--TG, an antiserum reactive to all variants of -TG-Ag, in
fact demonstrated the presence of two immunoreactive proteins, a major
one comigrating with a reference of purified CTAP-III, and a minor one
comigrating with a reference of purified CXCL7 (data not shown). No
CXCL7-like protein was detected in supernatants derived from
incubations with 100 nM CTAP-III, which scored positive for CTAP-III
only. Quite corresponding results were obtained upon testing the same
supernatants for CXCL7 biological activity using the neutrophil
degranulation assay. Although no CXCL7-like activity higher than the
detection limit of the assay (2 nM CXCL7) was measurable in
supernatants derived from incubations without or with 100 nM CTAP-III,
supernatants derived from incubations with 1 and 10 µM CTAP-III
scored positive, containing activities equivalent to 53 ± 23 and
331 ± 36 nM CXCL7, respectively. These data taken together
strongly indicate that biologically active CXCL7 becomes generated in
cocultures of neutrophils with HUVEC.
Although a previous study has revealed that among leukocytes
neutrophils have the highest capacity to proteolytically convert
CTAP-III into CXCL7 (12), to date nothing is known as to
whether HUVEC may also contribute to or modulate this process. To
investigate this, 1 µM CTAP-III was incubated with HUVEC monolayers
alone, with neutrophils alone, or with a combination of both cell types
for 1 h, and the recovered supernatants were analyzed for CXCL7
biological activity in the neutrophil degranulation assay. As can be
taken from the data in Table I, no
detectable CXCL7 activity was present in supernatants derived from
HUVEC alone, while those derived from neutrophils were highly active
and did not differ from those recovered from cocultures of HUVEC and
neutrophils. Activation of HUVEC by thrombin did not significantly
change these results, i.e., activated HUVEC alone did not generate
measurable CXCL7 from CTAP-III, and the amount generated in cocultures
of HUVEC with neutrophils was practically identical with that generated
by neutrophils alone (Table I). These results clearly show that HUVEC
lack the capacity to process CTAP-III into CXCL7 as well as to modulate
CTAP-III processing by neutrophils. Western blot analyses of the same
supernatants confirmed these results, because upon immunostaining with
antiserum R--TG, a protein comigrating with CTAP-III was the only
-TG variant detectable in HUVEC-derived supernatants, while those
derived from neutrophils cocultured with or without HUVEC additionally
contained an immunoreactive protein comigrating with CXCL7 (Fig. 2). In conclusion, CXCL7 generation in
cocultures of HUVEC and neutrophils exclusively depends on the
processing capacity of neutrophils.
|
|
To examine our hypothesis that CTAP-III-induced neutrophil
adhesion to HUVEC depends on the proteolytic formation of CXCL7 by the
neutrophils themselves, we used two different approaches. The first one
was to inhibit cathepsin G, the CTAP-III-processing protease expressed
on neutrophils, using the serine protease inhibitor aprotinin
(12). Therefore, neutrophils preincubated with aprotinin
for 10 min as well as untreated control cells were added to monolayers
of HUVEC and allowed to adhere for 25 min in the presence of either
CTAP-III or CXCL7 (control) or in the absence of stimulus. To obtain
comparable assay conditions, equipotent concentrations for CTAP-III (3
µM) and CXCL7 (300 nM) were chosen that represented dosages almost
sufficient to induce maximal cell adhesion (compare Fig. 1
). As shown
in Fig. 3, preincubation of neutrophils
with aprotinin significantly reduced their adhesion to HUVEC in
response to CTAP-III (by 50%), while there was no reduction in
response to CXCL7. Whereas these results indicated that the
CTAP-III-processing protease is involved in CTAP-III-induced neutrophil
adhesion, the failure of aprotinin treatment to completely inhibit the
CTAP-III-dependent effect did not exclude the possibility that other
mechanisms were also involved. Dosages of aprotinin higher than 10
µg/ml were not applicable because under these conditions the
inhibitor had a modulating effect on neutrophil adhesion by itself.
|
, pre-exposure of neutrophils to
-CXCR-1 Ab resulted in drastic reduction of cell adhesion in
response to CXCL8, but had no effect on CXCL7- or CTAP-III-induced
adhesion. In contrast, pre-exposure of neutrophils to -CXCR-2 Ab was
without effect on CXCL8-induced adhesion, while both CXCL7- as well as
CTAP-III-induced adhesion became drastically inhibited. In the
simultaneous presence of both Abs, no further enhancement of the
inhibitory effects induced by either Ab alone was visible (Table II).
Taken together, these results indicate that CXCL8 induces neutrophil
adhesion to HUVEC exclusively through CXCR-1, while CXCL7 induces cell
adhesion exclusively via CXCR-2. The observation that CTAP-III-induced
adhesion was also dependent on CXCR-2 provided a further hint that
CXCL7 might represent the active component. However, as seen above with
aprotinin, the inhibitory effect of -CXCR-2 on CTAP-III-induced
adhesion was only partial, and concentrations of the Ab higher than
those applied (50 µg/ml) were impractical, because of their intrinsic
modulatory effect on cell adhesion.
|
-CXCR-2 Ab simultaneously. As shown in Fig. 3, these conditions led to complete abrogation of CTAP-III-induced
neutrophil adhesion, while that induced by CXCL7 was not further
reduced to below the level of that remaining in the presence of
-CXCR-2 Ab alone. Furthermore, appropriate isotype controls
performed in the presence and absence of aprotinin made sure that the
inhibitor and Abs did not interfere with each other, i.e., the presence
of Ab isotype did not change the capacity of aprotinin to inhibit
CTAP-III-induced adhesion, while the presence of aprotinin did not
render the isotype reactive to CXCR-2. In conclusion, these results
altogether suggest that CTAP-III-induced adhesion depends on the
formation of CXCL7, and that interaction of CXCL7 with CXCR-2 signals
CTAP-III-dependent neutrophil adhesion to HUVEC. CXCL7- and CTAP-III-induced adhesion is dependent on MAC-1, but not LFA-1 or L-selectin
We have recently found that CXCL4-induced neutrophil adhesion to
HUVEC is dependent on the integrin LFA-1 and L-selectin, while adhesion
induced by CXCL8 requires the activation of the integrin MAC-1
(19). We therefore investigated what kinds of adhesion
molecules might be involved in CXCL7- and CTAP-III-induced adhesion by
performing blocking experiments using Abs against the
neutrophil-expressed integrin chains CD11a, CD11b, and CD18, as well as
against L-selectin (CD62L). As shown in Table III, CXCL7- as well as CTAP-III-induced
cell adhesion became drastically reduced by about or more than 90% in
the presence of Abs to either CD11b or CD18, while Abs to CD11a or
CD26L had practically no effect, as compared with the corresponding
isotype controls. These data demonstrate that cell adhesion induced by
CXCL7 and its precursor, just like that induced by CXCL8 (see Table III
for a comparison), is dependent on MAC-1 (CD11b/CD18), while LFA-1 and
L-selectin do not appear to play a role.
|
Because, along with CTAP-III, activated platelets also release
high amounts of CXCL4, it must be expected that neutrophils will become
exposed to both CXC chemokines practically simultaneously. We therefore
also investigated whether CXCL4 and CTAP-III or CXCL7 would mutually
influence each others’ capacity to induce neutrophil adhesion. For
this we exposed neutrophils to increasing dosages of CTAP-III (1–10
µM) or CXCL7 (0.1 and 1 µM) in the simultaneous presence or absence
of CXCL4. Two CXCL4 dosages were chosen, a lower one (0.5 µM) that by
itself had no capacity to promote neutrophil adhesion, and a 10-fold
higher one (5 µM) sufficient to induce near maximum cell adhesion
(compare Fig. 1). As shown in Fig. 4A, incubation of cells with
increasing concentrations of CTAP-III alone led to a
concentration-dependent increase in cell adhesion. By contrast, in the
simultaneous presence of 0.5 µM CXCL4, cell adhesion in response to
CTAP-III became almost completely abrogated. This did not occur in the
presence of 5 µM CXCL4, in which the chemokine induced a strong
adhesion response by itself, and stimulation of cells with CXCL4 and
CTAP-III in combination resulted in additive or slightly subadditive
responses as compared with the effect of either chemokine alone.
Corresponding experiments performed to assess the effect of CXCL4 on
CXCL7-induced adhesion led to a strikingly different result. As shown
in Fig. 4B, CXCL4 at a substimulatory dosage (0.5 µM)
considerably enhanced the capacity of CXCL7 to promote neutrophil
adhesion (by 100%), whereas at a high concentration (5 µM CXCL4)
the effects of CXCL4 and CXCL7 were approximately additive or slightly
superadditive. These results demonstrate that CXCL4, depending on its
actual concentration, may in fact drastically modulate CTAP-III- as
well as CXCL7-dependent cell adhesion. The striking observation that
the effect of CXCL4 on CTAP-III is just opposite to its effect on CXCL7
may indicate an important regulatory role for CXCL4 in early neutrophil
adhesion to HUVEC, as will be discussed later.
|
Having seen that CTAP-III, CXCL7, and CXCL4 all promote neutrophil
adhesion to HUVEC, we wondered whether these chemokines would likewise
be capable to stimulate neutrophil transendothelial migration. To study
this, we used the Transwell system, in which neutrophils were added to
confluent HUVEC monolayers cultured on gelatin-precoated polycarbonate
membranes, and these inserts were then placed into lower chambers
containing appropriate dilutions of the chemoattractants to be tested.
In agreement with published data (3, 4), CXCL8, included
in these assays for control reasons, induced neutrophil
transendothelial migration in a concentration-dependent manner, being
characterized by a typical optimum curve (Fig. 5). Maximum efficacy of the chemokine was
reached within a concentration range of 15–62 nM, and the stimulus
concentration inducing half-maximal cell migration
(EC50) was at 2 nM. Of the platelet-derived
chemokines, only CXCL7 was able to stimulate transendothelial migration
of neutrophils, while CTAP-III and CXCL4 were inactive within the
entire concentration range tested (10–10,000 nM). Interestingly,
CXCL7-stimulated cell migration differed from that induced by CXCL8,
inasmuch as no optimum curve was observed, but a continuous increase in
the number of migrated cells within a concentration range of 6 up to
4,000 nM CXCL7, in which maximum efficacy was reached. Although both
chemokines exhibited comparable efficacies (67% of migrated cells),
the potency of CXCL7 was 20-fold lower than that of CXCL8
(EC50 40 nM and EC50
2 nM, respectively; compare Fig. 5). Apart from showing that CXCL7
is efficient in stimulating neutrophil migration over an extremely wide
range of concentrations, these results altogether clearly demonstrate
that a chemokine’s ability to induce cell adhesion may not be
correlated to its ability to stimulate cell migration. Nevertheless, it
may not be excluded that CTAP-III and CXCL4, which scored negative in
the migration experiments, could be involved in regulating cell
migration by modulating CXCL7- and/or CXCL8-induced effects. This was
investigated as described below.
|
Upon intravascular platelet activation, neutrophils will first
become exposed to the released CTAP-III and CXCL4 before they may
migrate into the tissue in response to chemoattractants. To study the
impact of CTAP-III and CXCL4 on neutrophil transendothelial migration,
we therefore preincubated neutrophils for 10 min in the absence and
presence of either CTAP-III (1 µM) or CXCL4 (up to 5 µM), before
subjecting them to the transmigration assay. In these initial
experiments, pre-exposure of neutrophils to CTAP-III resulted in almost
complete abrogation of cell migration to CXCL7 (100 nM) and in
significant reduction of migration to an approximately equipotent
dosage of CXCL8 (10 nM), while pre-exposure to CXCL4 was without effect
(data not shown). In a more detailed approach, we then investigated the
concentration kinetics of CTAP-III, and for comparison also that of
CXCL7. As shown in Fig. 6A,
preincubation of neutrophils with increasing dosages of CTAP-III led to
a concentration-dependent reduction of cell migration in response to
CXCL7 as well as CXCL8. However, while the CXCL7-induced migration
response could be inhibited down to background levels with increasing
dosages of CTAP-III, this did not occur with the CXCL8-induced
response, which was much less potently affected by CTAP-III and
moreover became only partially inhibited (by 48%) even by highest
dosages of CTAP-III. Very similar inhibition kinetics were obtained
when neutrophils were pre-exposed to increasing dosages of CXCL7 (Fig. 6B), with the exception that considerably lower dosages of
the chemokine were required to obtain inhibition rates comparable with
that induced by CTAP-III. Thus, half-maximal inhibition of
CXCL7-induced migration was obtained following preincubation with 180
nM CTAP-III and 30 nM CXCL7, respectively. These results suggest, as
already shown for the induction of cell adhesion above (compare Fig. 3), that the impact of CTAP-III on neutrophil migration is indirect,
depending on its proteolytic cleavage into CXCL7 and interaction of the
latter with CXC chemokine receptors on these cells (compare Table II).
Generally, homologous desensitization among CXC chemokines correlates
with the down-regulation of their common receptors from the cell
surface. Correspondingly, as documented in Table IV, we observed that pre-exposure of
neutrophils to approximately equipotent dosages of CTAP-III (1 µM) or
CXCL7 (300 nM) resulted in significant down-regulation of CXCR-2,
with no or only minimal impact on CXCR-1 expression. This may explain
why the former chemokines desensitized CXCL8-induced cell migration
only partially, because CXCL8 has been shown to mediate neutrophil
chemotaxis by both receptors, whereas CXCL7 preferentially acts through
its high affinity receptor CXCR-2 (14). By contrast, in
keeping with its inability to desensitize cell migration to CXCL7 or
CXCL8, CXCL4 modulated neither CXCR-1 nor CXCR-2 expression on
neutrophils (Table IV).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Obviously, there is no contribution of endothelial cells to CXCL7
generation from CTAP-III. Neither did we find HUVEC to cleave CTAP-III
by themselves, nor did these cells modulate neutrophil-mediated
processing of the chemokine precursor. This was seen in experiments in
which neither unstimulated nor thrombin-stimulated HUVEC were able to
generate CXCL7 protein or CXCL7 biological activity from CTAP-III or to
change the amount of CXCL7 generated by CTAP-III-processing
neutrophils. Our finding that newly formed CXCL7 represented the active
principle in CTAP-III-induced cell adhesion also provided an
explanation for the fact that both chemokines involved the same
adhesion molecules on neutrophils, i.e., MAC-1, but not LFA-1 or
L-selectin as previously found with CXCL4 (19). Likewise,
the time-dependent conversion of CTAP-III into active CXCL7 during
incubation with neutrophils can explain the large difference in the
potencies of these chemokines as opposed to their practically identical
efficacies for the induction of cell adhesion. Because CTAP-III
cleavage by neutrophils is a continuous process proceeding for hours
(12, 28), the potency determined for CTAP-III in the
adhesion assay can only be considered an apparent one and does not
represent a constant, as its value will depend on the amount of CXCL7
generated during the actual incubation time chosen for the adhesion
assay. As a consequence of continuous CXCL7 formation from CTAP-III, it
may be envisaged that induction of cell adhesion by CTAP-III in vivo
will represent a dynamic self-amplifying process, in which CXCL7
generated by processing neutrophils will recruit additional neutrophils
that in turn will enhance CXCL7 formation, etc. Such a mechanism
differs largely from that observed with CXCL4. In comparison with CXCL7
and CTAP-III, this chemokine requires a high threshold concentration
for the induction of measurable cell adhesion (0.5 µM, as opposed
to 1 and 10 nM for the former ones, respectively) and reaches
maximum effects within a relatively narrow range of concentrations.
Thus, induction of cell adhesion by CXCL4 resembles an off/on reaction,
which phenomenon has been interpreted to ensure firm neutrophil
adhesion under conditions of rapid blood flow (19).
Irrespective of the latter notion, it is intriguing to find that
platelets secrete two different major CXC chemokines that promote
neutrophil adhesion through completely different mechanisms and
moreover involve different sets of adhesion molecules, which altogether
suggests that there exists considerable evolutionary pressure to
facilitate this important initial step in early neutrophil recruitment.
The usage of different receptors by CXCL4, which binds to a chondroitin
sulfate proteoglycan (30, 31), and CXCL7, which we in this
study found to induce cell adhesion exclusively through CXCR-2, may
also be within these lines. Corresponding to the latter observation,
our Ab inhibition experiments moreover demonstrated that CXCL8 acted
exclusively through CXCR-1. This kind of restriction to only one of the
ELR+ CXC chemokine receptors is not a rule,
because it was demonstrated previously that CXCL7 and CXCL8 both can
use either CXCR-1 or CXCR-2 to elicit neutrophil chemotaxis
(14) and degranulation (32).
Even more interestingly, we also obtained evidence that there existed
cross talk between the different chemokines, inasmuch as CXCL4 at a
dosage below the threshold to induce cell adhesion by itself was able
to modulate CTAP-III- as well as CXCL7-induced adhesion. At first
sight, it may appear paradoxical that CXCL4 acts strongly inhibitory on
CTAP-III-induced adhesion, while it enhances the cellular response to
the cleavage product CXCL7. However, with regard to the above notion
that cell adhesion in response to platelet-derived chemokines appears
to be a tightly regulated process, the ambivalent action of CXCL4 may
be quite instrumental for the control of cell adhesion directly at its
onset. As seen in a different study (E. Brandt, unpublished results),
already nanomolar CXCL4 has the capacity to inhibit the processing of
CTAP-III by neutrophils, which results in considerable reduction of
CXCL7 formation. On the one hand, this mechanism could serve to prevent
undesired neutrophil activation in situations of normal platelet
apoptosis or superficial release of -granule contents. In contrast,
together with the observed heterologous enhancement of CXCL7 activity
by CXCL4, it could contribute to sharpen the neutrophil adhesion
response by allowing the cells to mount a more intense reaction as soon
as a threshold value of CXCL7 has been reached. These mechanisms are
apparently overridden in situations of full platelet activation, in
which CTAP-III as well as CXCL4 become released at high concentrations
and, as shown by our data, exhibit additive effects on cell
adhesion.
Cross talk between platelet-derived chemokines also exists at the level of neutrophil transendothelial migration. As seen in our experiments using unstimulated HUVEC in a Transwell system, only CXCL7 had the capacity to promote transendothelial migration, while its precursor CTAP-III and also CXCL4 were totally inactive. These results were not unexpected, because we and others previously found corresponding characteristics for these chemokines concerning their ability to promote in vitro neutrophil chemotactic migration (14, 17, 18, 33). Moreover, we found that preincubation of neutrophils with CTAP-III drastically down-modulated their migratory response to CXCL7 as well as to CXCL8. This finding was reminiscent of results obtained in former work, in which we found CXCL7 precursors CTAP-III and platelet basic protein as well as CXCL7 itself to act as desensitizing agents for ELR+ CXC chemokines in Boyden chamber chemotaxis assays (14, 28). To date, all evidence suggests that CTAP-III-dependent down-regulation of neutrophil transendothelial migration likewise is based on the precursor’s cleavage into CXCL7, which then selectively down-regulates CXCR-2 from the neutrophil cell surface. Due to the preference of CXCL7 to induce cell migration through CXCR-2, while CXCL8 preferentially activates this function through CXCR-1, it appears consistent that CTAP-III at very high dosages was able to completely block the CXCL7-induced response, while it affected CXCL8-dependent migration only partially. In a similar system, Kitayama and coworkers (5) have found homologous desensitization of transendothelial migration between CXCL8 and CXCL1. With respect to the fact that CXCL1 predominantly acts through CXCR-2, their results are much alike ours, because they described luminally applied CXCL8 to completely suppress CXCL1-induced cell migration, whereas in a reverse setting CXCL1 could only partially desensitize CXCL8-dependent cell migration. Both CXCL8 as well as CXCL1 may become produced by endothelial cells during inflammation. Thus, it appears likely that regulatory principles serving to limit or to counteract excessive cell recruitment may be realized at different states of inflammation to make sure that neutrophil extravasation only occurs when a potent or a sufficient amount of stimulus signals an inflammatory situation. Altogether our results demonstrate that neutrophil recruitment by platelet-derived chemokines underlies multiple regulatory control, which appears quite logical, because these mediators are directly at the onset of an inflammatory reaction and thus have a decisive impact on its further enhancement or limitation.
|
Acknowledgments |
|---|
, I. von
Cube, and C. Engellenner for perfect technical assistance. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ernst Brandt, Department of Immunology and Cell Biology, Forschungszentrum Borstel, Parkallee 22, D-23845 Borstel, Germany. E-mail address: ebrandt{at}fz-borstel.de
3 Abbreviations used in this paper: CXCL, CXC chemokine ligand; ELR, Glu-Leu-Arg; -TG, -thromboglobulin; CTAP-III, connective tissue-activating peptide III; MAC-1, macrophage-1 Ag; PBS-D, Dulbecco’s PBS; RT, room temperature.
Received for publication March 11, 2002. Accepted for publication June 26, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, MCP-1, IP-10, and Mig are sequentially and differentially expressed during phase-specific infiltration of leukocyte subsets in human wound healing. Am. J. Pathol. 153:1849. and IL-8 mRNA in psoriasis: a model for neutrophil migration and accumulation in vivo. J. Invest. Dermatol. 107:778.[Medline]
-thromboglobulins and platelet factor 4: blood platelet-derived CXC chemokines with divergent roles in early neutrophil regulation. J. Leukocyte Biol. 67:471.[Abstract]
-thromboglobulin formed in cultures of stimulated mononuclear cells activates human neutrophils. Biochem. Biophys. Res. Commun. 159:969.[Medline]
renders human neutrophils responsive to platelet factor 4: comparison of PF-4 and IL-8 reveals different activity profiles of the two chemokines. J. Immunol. 156:1954.[Abstract]
potentiates neutrophil degranulation in response to host defense cytokines neutrophil-activating peptide 2 and IL-8 by modulating intracellular cyclic AMP levels. J. Immunol. 149:1356.[Abstract]
-thromboglobulin and their impact on neutrophil functions. J. Immunol. 161:4975.This article has been cited by other articles:
|
|
 |
|
  F. Schiemann, E. Brandt, R. Gross, B. Lindner, J. Mittelstadt, C. P. Sommerhoff, J. Schulmistrat, and F. Petersen The Cathelicidin LL-37 Activates Human Mast Cells and Is Degraded by Mast Cell Tryptase: Counter-Regulation by CXCL4 J. Immunol., August 15, 2009; 183(4): 2223 - 2231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Gleissner, P. von Hundelshausen, and K. Ley Platelet Chemokines in Vascular Disease Arterioscler Thromb Vasc Biol, November 1, 2008; 28(11): 1920 - 1927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Zhang, G. Cherryholmes, A. Mao, C. Marek, J. Longmate, M. Kalos, R. P. ST. Amand, and J. E. Shively High Plasma Levels of MCP-1 and Eotaxin Provide Evidence for an Immunological Basis of Fibromyalgia Experimental Biology and Medicine, September 1, 2008; 233(9): 1171 - 1180. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Woller, E. Brandt, J. Mittelstadt, C. Rybakowski, and F. Petersen Platelet factor 4/CXCL4-stimulated human monocytes induce apoptosis in endothelial cells by the release of oxygen radicals J. Leukoc. Biol., April 1, 2008; 83(4): 936 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Barish, A. R. Atkins, M. Downes, P. Olson, L.-W. Chong, M. Nelson, Y. Zou, H. Hwang, H. Kang, L. Curtiss, et al. PPAR{delta} regulates multiple proinflammatory pathways to suppress atherosclerosis PNAS, March 18, 2008; 105(11): 4271 - 4276. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Pitchford, S. Momi, S. Baglioni, L. Casali, S. Giannini, R. Rossi, C. P. Page, and P. Gresele Allergen Induces the Migration of Platelets to Lung Tissue in Allergic Asthma Am. J. Respir. Crit. Care Med., March 15, 2008; 177(6): 604 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Kasper, E. Brandt, S. Brandau, and F. Petersen Platelet Factor 4 (CXC Chemokine Ligand 4) Differentially Regulates Respiratory Burst, Survival, and Cytokine Expression of Human Monocytes by Using Distinct Signaling Pathways J. Immunol., August 15, 2007; 179(4): 2584 - 2591. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Ganser, M. A. Morgan, and E. M. Weissinger Going from genes to proteins in myelodysplastic syndromes PNAS, January 23, 2007; 104(4): 1109 - 1110. [Full Text] [PDF]
|
|
|
|
 |
|
 F. Schiemann, T. A. Grimm, J. Hoch, R. Gross, B. Lindner, F. Petersen, S. Bulfone-Paus, and E. Brandt Mast cells and neutrophils proteolytically activate chemokine precursor CTAP-III and are subject to counterregulation by PF-4 through inhibition of chymase and cathepsin G Blood, March 15, 2006; 107(6): 2234 - 2242. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Honda, E. Segi-Nishida, Y. Miyachi, and S. Narumiya Prostacyclin-IP signaling and prostaglandin E2-EP2/EP4 signaling both mediate joint inflammation in mouse collagen-induced arthritis J. Exp. Med., February 21, 2006; 203(2): 325 - 335. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Baltus, P. von Hundelshausen, S. F. Mause, W. Buhre, R. Rossaint, and C. Weber Differential and additive effects of platelet-derived chemokines on monocyte arrest on inflamed endothelium under flow conditions J. Leukoc. Biol., August 1, 2005; 78(2): 435 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. von Hundelshausen, R. R. Koenen, M. Sack, S. F. Mause, W. Adriaens, A. E. I. Proudfoot, T. M. Hackeng, and C. Weber Heterophilic interactions of platelet factor 4 and RANTES promote monocyte arrest on endothelium Blood, February 1, 2005; 105(3): 924 - 930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Z. Dudek, I. Nesmelova, K. Mayo, C. M. Verfaillie, S. Pitchford, and A. Slungaard Platelet factor 4 promotes adhesion of hematopoietic progenitor cells and binds IL-8: novel mechanisms for modulation of hematopoiesis Blood, June 15, 2003; 101(12): 4687 - 4694. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), CXCL8 (
), or CXCL4 (
), and allowed to attach for 25 min at
37°C. Nonadherent cells were removed from HUVEC monolayers by
centrifuging the plates at an angle of 45° (200 x
g for 1 min). Adherent cells were lysed (together with
HUVEC) and quantified by measurement of neutrophil-specific endogenous
) or 300 nM CXCL7 (
