| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Institute for Biochemistry, and
Third Department of Internal Medicine, Justus-Liebig-Universität, Giessen, Germany;
The Sol Sherry Thrombosis Research Center, Temple University School of Medicine, Philadelphia, Pennsylvania, USA; and
Deutsches Herzzentrum, Technische Universität München, Munich, Germany
1Correspondence: Institut für Biochemie, Fachbereich Humanmedizin, Justus-Liebig-Universität, Friedrichstrasse 24, D-35392 Giessen, Germany. E-mail: triantafyllos.chavakis{at}innere.med.uni-giessen.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mß2) on neutrophils and exert antiadhesive properties by binding to the urokinase receptor (uPAR) and vitronectin. We define the molecular mechanisms for the antiadhesive effects of HK related to disruption of ß2-integrin-mediated cellular interactions in vitro and in vivo. In a purified system, HK and HKa inhibited the binding of soluble fibrinogen and ICAM-1 to immobilized Mac-1, but not the binding of ICAM-1 to immobilized LFA-1 (CD11a/CD18, Lß2). This inhibitory effect could be attributed to HK domain 5 and to a lesser degree to HK domain 3, consistent with the requirement of both domains for binding to Mac-1. Accordingly, HK, HKa, and domain 5 inhibited the adhesion of Mac-1 but not LFA-1-transfected K562 human erythroleukemic cells to ICAM-1. Moreover, adhesion of human monocytic cells to fibrinogen and to human endothelial cells was blocked by HK, HKa, and domain 5. By using peptides derived from HK domain 5, the sequences including amino acids H475-G497 (and to a lesser extent, G440-H455) were identified as responsible for the antiadhesive effect, which was independent of uPAR. Finally, administration of domain 5 into mice, followed by induction of thioglycollate-provoked peritonitis, decreased the recruitment of neutrophils by 
70% in this model of acute inflammation. Taken together, HKa (and particularly domain 5) specifically interacts with Mac-1 but not with LFA-1, thereby blocking Mac-1-dependent leukocyte adhesion to fibrinogen and endothelial cells in vitro and in vivo and serving as a novel endogenous regulator of leukocyte recruitment into the inflamed tissue.—Chavakis, T., Kanse, S. M., Pixley, R. A., May, A. E., Isordia-Salas, I., Colman, R. W., Preissner, K. T. Regulation of leukocyte recruitment by polypeptides derived from high molecular weight kininogen.
Key Words: HK • monoclonal antibody • leukocyte • Mac-1
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mß2, CR3) and LFA-1 (CD11a/CD18, Lß2), which interact with their counter-receptor ICAM-1 on the endothelial cells (1
, 2)
. Mac-1 also regulates leukocyte adhesion to provisional matrix substrates, including fibrinogen (FBG), which becomes deposited at the sites of inflammation and injury after increases in vascular permeability and damage. Divalent cations (Mn2+, Mg2+) or phorbol ester serve as stimuli of integrin-dependent leukocyte adhesion phenomena (3
4
5
6)
that are also profoundly affected by the presence of the urokinase receptor (uPAR), which has the ability to interact with and regulate the function of integrins (7
8
9
10
11)
. Moreover, uPAR-expressing leukocytic cells can directly adhere via this glycolipid-anchored protein to matrix-associated vitronectin, another component of the temporary wound-healing tissue (3
, 12)
.
After vascular injury, inflammation, or activation of complement in humoral immune defense, high molecular weight kininogen (HK) serves a nonenzymatic cofactor role in the initiation of the contact phase. HK is composed of six domains and is present in plasma at a concentration of 0.67 µM (13
14
15)
. Kallikrein can liberate the short-lived vasodilator peptide bradykinin from HK (16)
, thereby generating stoichiometric concentrations of two-chain, kinin-free HKa, which lacks most of its domain 4. Domain 5 in HKa is rich in His, Gly, and Lys, which enables HKa to bind to anionic surfaces, zinc ions, or heparin as well as to cell surface proteoglycans (17
18
19
20)
. Domains 3 and 5 mediate HK/HKa binding to cells (21
, 22)
, contributing to the regulation of pericellular plasmin generation by modulating plasma kallikrein-dependent formation of uPA (23)
, a reaction dependent on the binding of plasma prekallikrein to HK domain 6 (14)
. In addition to heparan sulfate proteoglycans on endothelial cells, the binding proteins for globular C1q (denoted gC1qR), uPAR, and cytokeratin 1 were identified as binding proteins for HK (24
25
26
27)
. HK (especially HKa) was previously reported to exert antiadhesive properties (28
29
30)
, and we recently identified HKa and particularly domain 5 to directly bind to vitronectin, thereby competing for uPAR-dependent leukocyte adhesion as well as integrin-dependent endothelial cell adhesion to vitronectin (31)
. Moreover, domains 3 and 5 of HK bind to the integrin Mac-1 on granulocytes and compete for fibrinogen binding (32
33
34)
.
These observations prompted us to investigate in more detail the antiadhesive properties of kininogen-derived fragments and peptides related to disruption of ß2-integrin-mediated cellular interactions. Our results indicate that kininogen, and particularly domain 5, specifically interact with Mac-1 but not LFA-1, thereby blocking Mac-1-dependent leukocyte adhesion to FBG and endothelial cells in vitro and in vivo. Thus, kininogen (especially its domain 5) plays a regulatory role in the recruitment of leukocytes to the inflamed tissue and could be important for therapeutic interventions in hyperinflammatory situations.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. GST was amino-terminally attached to the following sequences of HK: G235-M357 (domain 3), K420-S513 (domain 5), T503-S626 (domain 6), as well as K420-D474 and H475-K502 (domain 5 sequences). The mutants were purified on a glutathione column reaching more than 90% purity. Peptide synthesis and HPLC purification to more than 95% purity were performed by Dr. J. Lambris (University of Pennsylvania, Philadelphia, PA). Refolding of cysteine-containing peptides was carried out by air oxidation for 3 days at 4°C with continuous agitation in buffer containing 50 mM ammonium bicarbonate, pH 8.5, at a final concentration of 100 µg/ml, followed by freeze-drying. In addition to peptides HK406, HK440, HK475, HK483, and HK486 (Table 1
), the following control scrambled peptides were used: HK475M (GHHKKHGHGHH), HK483M (GHHGHKKNGKKKGNK), and HK486M (KGHKKNGKKNKGNHWGK).
|
ZnCl2 was from Sigma (Munich, Germany), vitamin D3 was from Biomol (Hamburg, Germany), transforming growth factor-ß was from R&D Systems (Boston, MA), and interleukin 3 (IL-3) was from PBH (Hannover, Germany). Isolated ICAM-1 was kindly provided by Dr. Carl Figdor (Nijmegen, The Netherlands). Isolated Mac-1, LFA-1, and ICAM-1-Fc (ICAM-1 was fused to the constant region of human IgG) were obtained from Dr. S. Bodary (Genentech, San Francisco, CA). FBG was purchased from Kabivitrum (Munich, Germany). The blocking monoclonal antibody (mAb) against human CD18, 60.3, was kindly provided by Dr. J. Harlan (Seattle, WA); the blocking mAb against mouse CD18, GAME46, as well as the blocking mAb against mouse Mac-1 (M1/70) and mouse LFA-1 (M17/4) used for in vivo inhibition studies were from PharMingen (Hamburg, Germany). Polyclonal antibodies and mAb against ICAM-1 and against FBG, peroxidase-conjugated streptavidin as well as secondary anti-mouse and anti-rabbit immunoglobulins were from DAKO (Hamburg, Germany). Phosphatidyl inositol specific phospholipase C (piPLC) was from Oxford Glyco Systems (Oxford, UK).
Cell culture
Myelomonocytic cells (U937) were from American Type Culture Collection (ATCC) (Rockville, MD) and were cultured as described by the supplier in RPMI 1640 medium containing 10% (v/v) fetal calf serum. BAF-3 cells were from ATCC and cultured in RPMI 1640 medium containing 10% fetal calf serum and 2 ng/ml IL-3. Mononuclear cells were isolated from peripheral blood by Ficoll gradient centrifugation (20 min, 4°C, 800 g). Human umbilical vein endothelial cells (HUVEC) were isolated and cultivated as described (36
, 37)
. All culture media were from Life Technologies (Eggenstein, Germany) or PromoCell (Heidelberg, Germany). K562 cells transfected with Mac-1 or LFA-1 were kindly provided by Dr. M. Smith (Celltech, Slough, UK) and Dr. Y. van Kooyk (University of Nijmegen, The Netherlands) and cultivated in a mixture of 75% RPMI containing 10% fetal calf serum and 25% ISCOVE’s medium containing 5% fetal calf serum.
Construction of uPAR-transfected BAF-3 cells
BAF-3 cells (IL-3-dependent mouse B cell line) were transfected by electroporation with uPAR cDNA in the sense orientation using the expression vector pCDNA3. Cells were selected in the presence of G418 (1.2 mg/ml) and found to express uPAR by FACS analysis, Northern blotting, and uPAR-ELISA, respectively (3
, 31)
.
ELISA for ligand–receptor interactions
Maxisorb plates (high binding capacity; Nunc, Naperville, IL) were coated with Mac-1 or LFA-1 (5 µg/ml) dissolved in 20 mM HEPES, 150 mM NaCl, 1 mM Mn2+, pH 7.2, then blocked with 3% (w/v) bovine serum albumin (BSA) in the same buffer. Binding of biotin-FBG (0.5 µg/ml) or ICAM-1-Fc (10 µg/ml) to the immobilized integrin was performed in a final volume of 50 µl (same buffer as above together with 0.05% (w/v) Tween-20) in the absence or presence of different kininogen forms without or together with 10–50 µM Zn2+. After incubation for 2 h at 22°C and washing, bound biotin-FBG was detected by the addition of 1:8000 diluted peroxidase-conjugated streptavidin. Bound ICAM-1-Fc was detected by the addition of 1:1000 diluted peroxidase-conjugated antibody against human immunoglobulins. Peroxidase was further quantified using 2,2'-azino-di-[3-ethylbenzthiazoline sulfonate (6)
] as substrate (Boehringer Mannheim, Germany), followed by development in a microplate reader (Molecular Devices, Menlo Park, CA) measuring absorbance at 405 nm. In other experiments, FBG (2 µg/ml) or ICAM-1 (10 µg/ml) binding to immobilized Mac-1 or LFA-1 was performed (same buffer as above). Binding was detected by a polyclonal antibody against FBG or ICAM-1, respectively. Polyclonal antibodies were further detected by 1:1000 diluted peroxidase-conjugated anti-rabbit immunoglobulins. When tested, binding of biotin-labeled HK or HKa (10 nM each) to immobilized Mac-1 (same buffer as above in the absence or presence of 10 µM Zn2+) was detected as described for biotin-FBG binding. Nonspecific binding to BSA-coated wells was used as blank and subtracted to calculate specific binding. For biotinylated ligands, nonspecific binding was also determined in the presence of a 100 molar excess of unlabeled ligand.
Cell adhesion assays
Cell adhesion to ICAM-1- and FBG-coated plates (and BSA-coated wells as control) was tested according to previously described protocols (3)
. Multiwell plates were coated with 5 µg/ml ICAM-1 or FBG (dissolved in bicarbonate buffer, pH 9.6), respectively, and blocked with 3% (w/v) BSA. U937 cells differentiated for 24 h with vitamin D3 (100 nM) and transforming growth factor ß (2 ng/ml), BAF-3, or K562 cells were washed in serum-free RPMI and plated onto the precoated wells for 60–90 min at 37°C in serum-free RPMI in the absence or presence of competitors. After the incubation period for the adhesion assay, the wells were washed, adherent cells were fixed with methanol/acetone (1:1) for 30 min at 4°C, stained for 45 min with crystal violet, and destained with a solution of acetic acid/methanol/H20 (10:30:60). The number of adherent cells was quantified in a microplate reader by measuring absorbance at 590 nm. Pretreatment of cells with piPLC was performed in serum-free medium (RPMI) containing 0.5 U/ml of piPLC for 2 h at 37°C and cells were washed with serum-free medium before proceeding with the standard adhesion assay.
Adhesion of human peripheral blood mononuclear cells and U937 cells to cultured HUVEC was performed according to a previously published protocol (11)
. HUVEC were seeded onto gelatin-coated 96-well plates for 48 h. Confluency was confirmed by microscopic inspection before each experiment. Freshly isolated mononuclear cells or U937 cells that had been differentiated for 24 h with vitamin D3 (100 nM) and transforming growth factor-ß (2 ng/ml) were washed twice in adhesion medium (serum-free RPMI 1640/HEPES 25 mM), followed by no pretreatment or stimulation with PMA (50 ng/ml). After washing, cells were added (7x105/ml) to HUVEC monolayers in the absence or presence of competitors. After 30 min coincubation (37°C, 5% CO2, 90% humidity), the plates were gently washed twice to remove nonadherent cells. Remaining adherent cells were quantified by counting 16 high-power fields using light microscopy.
In vivo peritonitis model
Experiments were performed according to a previously described protocol (38
, 39)
, in which 1 ml thioglycollate bouillon (Merck, Darmstadt, Germany) was administered intraperitoneally to female 8- to 10-wk-old NMRI mice (Charles River Wiga, Sulzfeld, Germany). For inhibition studies, 100 µg mAb against mouse Mac-1 or mouse LFA-1 in PBS or 40–100 µg of either GST domain 5 or GST in PBS was administered intravenously (i.v.) 30 min before the injection of thioglycollate. Control mice were treated with the same volume of PBS; some mice were treated with isotype-matched control antibodies. All reagents were endotoxin free. One and 4 h after injection of thioglycollate, cells were harvested from the peritoneum and the total number of emigrated neutrophils was determined. Mice were killed at the times indicated and peritoneal lavage was obtained by injecting 10 ml PBS, massaging the peritoneal wall, and removing the fluid. Total cell numbers were determined in a Casy Counter (Schärfe System, Germany). Fifty thousand cells were then transferred onto adhesion slides (Bio-Rad, Munich, Germany), fixed, and stained with Diff-Quick (DADE-Behring, Munich, Germany). Cells were differentially counted by microscopy, evaluating 300 cells per slide. The absolute number of neutrophils in the peritoneal lavage was calculated from the total cell count in the peritoneal lavage and the fraction of neutrophils. Peripheral neutrophil counts were not affected by any of the antibodies or peptides injected, as determined independently from whole blood cell count.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
33
34)
. The binding of biotin-FBG to immobilized Mac-1 was inhibited by HK (IC50: 230 nM), HKa (IC50: 130 nM), the isolated GST domain 5 (IC50: 140 nM) and to a lesser degree by GST domain 3 (IC50: 440 nM) in a Zn2+-dependent manner, whereas GST domain 6 or the isolated GST had no effect at all (Fig. 1
A). The possibility that HK competes with the binding of other Mac-1 ligands (such as ICAM-1) was tested: Similar to FBG binding, interaction of soluble ICAM-1-Fc with immobilized Mac-1 was blocked in a Zn2+-dependent manner by HK (IC50: 215 nM), HKa (IC50: 100 nM), domain 5 (IC50: 150 nM), and, to a lesser extent, domain 3 (IC50: 500 nM) (Fig. 1B
). Since ICAM-1 is also a ligand of LFA-1, and both Mac-1 and LFA-1 share homologies in their binding domains, we tested whether kininogen also interferes with the binding of ICAM-1 to immobilized LFA-1. No effect of different HK forms on the ICAM-1/LFA-1 interaction in the absence or presence of Zn2+ was found (not shown). Similar results with the different HK forms were obtained when binding to Mac-1 was performed with the isolated ligands, FBG and ICAM-1, and detected by respective polyclonal antibodies (not shown).
|
To further assess the significance of the partial inhibition observed with domain 3, inhibition of biotin-FBG binding to Mac-1 by increasing concentrations of domain 5 was tested in the absence or presence of a subthreshold concentration of domain 3 (100 nM). In this case, an additive inhibitory effect of both domains was observed when both of them were in subthreshold concentrations whereas no such effect was found when domain 5 was in the optimal concentration that provided maximal inhibition of FBG binding to Mac-1 (Fig. 2
A, similar data for ICAM-1 binding to Mac-1; not shown). Similar results were also observed in the reverse experiment, i.e., when increasing concentrations of domain 3 were tested in the absence or presence of a subthreshold concentration of domain 5 (100 nM).
|
Furthermore, in order to determine the exact concentration of Zn2+ needed for the inhibitory effect of HK, binding of biotin-FBG to Mac-1 was performed without or together with 500 nM HKa or domain 5 in the absence or presence of increasing concentrations of free zinc ions. From Fig. 2B, we
conclude that 10 µM of free Zn2+ are sufficient for maximal inhibition of FBG binding to Mac-1 by HKa or domain 5. In accordance, 10 µM Zn2+ were also required for the direct binding of HK/HKa to immobilized Mac-1 (data not shown).
To test the hypothesis that HK only interacts with Mac-1 and not with LFA-1 on the cell surface, the inhibitory capacity of different HK forms on adhesion of transfected erythroleukemic K562 cells was studied. Whereas nontransfected K562 cells did not adhere to FBG or ICAM-1, cells became adherent to both substrates on transfection with Mac-1, whereas LFA-1-transfected cells only adhered to ICAM-1. In both cases, cell adhesion was stimulated by Mn2+ (100 µM), a known activator of ß2-integrins (4
5
6
, 10
, 11)
. HK, HKa, domain 5 (and domain 3 to a lesser degree) decreased adhesion only of Mac-1-transfected K562 cells to both immobilized ligands, whereas adhesion of LFA-1-transfected cells to ICAM-1 was not affected at all (Fig. 3
). Taken together, HK, HKa and particularly domain 5 specifically interact with Mac-1 but not with LFA-1, thereby blocking the interaction of Mac-1 with its ligands ICAM-1 and FBG.
|
Inhibition of leukocyte adhesion to FBG and endothelial cells by kininogens
As previously established, the adhesion of myelomonocytic U937 cells [differentiated with TGFß (2 ng/ml) and vitamin D3 (100 nM) for 24 h] to immobilized FBG or endothelial cells is predominantly mediated by ß2-integrins, and adhesion is enhanced by Mn2+ or phorbol ester (PMA) (4
5
6
, 10
, 11)
. HK, HKa, and domain 5 dose-dependently inhibited adhesion of differentiated U937 cells to endothelial cells and FBG under control conditions or stimulation with PMA or Mn2+, whereas domain 3 had a weaker effect; domain 6 and GST were not inhibitory at all (Fig. 4
A, B). Similar results were obtained when the adhesion of human peripheral blood mononuclear cells to endothelial cells was tested (not shown). Moreover, inhibition by domain 5, HK or HKa and inhibition by a blocking monoclonal antibody against CD18 were additive if the components were at subthreshold concentrations but not additive when either component was at optimal concentration (Fig. 4C
). Since domain 5 could reproduce most of the antiadhesive capacity of HK/HKa, peptides derived from this domain were used to identify the responsible inhibitory sequence(s). Both parts of domain 5, K420-D474 and H475-K502, individually fused to GST at the amino-terminal end, exhibited antiadhesive function, with the second part of domain 5 being almost as potent as the entire domain 5 (K420-S513) (Fig. 5
A). Five synthetic peptides derived from both parts of domain 5 and displaying surface loops that could potentially participate in protein-protein interactions were tested for adhesion inhibition within a concentration range of 1–1000 nM (40)
(Table 1)
. Peptides HK475 and HK483 entailing the sequences H475-H485 and H483-G497 respectively, were the most potent ones and could completely inhibit monocyte adhesion to FBG. Whereas peptide HK406 (G406-N422) had no effect at all, peptides HK486 (G486-K502) and HK440 (G440-H455) provided a 50–60% inhibition at a concentration of 1000 nM (Fig. 5B
).
|
|
To test whether the behavior of the various HK peptides does not simply reflect their gly/his/lys content, the effect of peptides HK475, HK483 and HK486 was compared to the respective scrambled peptides HK475M, HK483M and HK486M. The ability of the scrambled peptides to inhibit adhesion to FBG was markedly decreased (Fig. 5B
); scrambled peptides HK475M and HK483M provided an inhibition of not more than 20%. Moreover, we investigated whether these peptides derived from domain 5 could directly interfere with HK/HKa binding to immobilized Mac-1. In a Zn2+-dependent manner HK and HKa bound to immobilized Mac-1, HKa having a higher affinity (EC50: 2.5 nM) than HK (EC50: 5 nM) (Fig. 6
A). The minimal concentration of Zn2+ required for optimal binding of HK and HKa to Mac-1 was 10 µM (data not shown). The peptides HK475 and HK483 blocked the interaction between HK/HKa and Mac-1, whereas the respective scrambled peptides had a markedly reduced inhibitory potency; again, peptides HK440 and HK486 provided a much weaker inhibition of HK/HKa binding to Mac-1 (Fig. 6B
) and peptide HK406 had no effect at all. Taken together, these results indicate that the sequence H475-G497 predominantly mediates the interaction of HK/HKa with Mac-1 and thereby contains the antiadhesive activity of HK domain 5 for Mac-1-dependent interactions and that the sequence G440-H455 may represent a secondary interaction site.
|
The influence of uPAR on the antiadhesive activity of kininogen
Recent reports provide evidence that uPAR regulates the activity particularly of ß2-integrins, thereby influencing leukocyte recruitment (7
, 8)
. Moreover, we have shown that uPAR is a binding protein for HKa on the cell surface, an interaction mediated by domain 5 (31)
. We tested the possibility that the antiadhesive effect of HK on Mac-1-dependent cell adhesion could be modulated in the presence of uPAR. Nontransfected BAF-3 cells or BAF-3 cells transfected with uPAR were compared for their ability to adhere to FBG via ß2-integrins, which is significantly augmented in the uPAR-expressing cells (Fig. 7
). With both uPAR-transfected and nontransfected BAF-3 cells a similar inhibition of cell adhesion by HK, HKa or domain 5 was observed, and the pattern of inhibition was reminiscent of the effect of HK forms on the adhesion of monocytes to FBG (see Fig. 4
). Although the different HK forms inhibited adhesion of both wild type and uPAR-expressing cells, there was a change in affinity in the presence of uPAR: The IC50 of HKa and domain 5 in the presence of uPAR (175 nM and 150 nM, respectively) was higher than in the absence of uPAR (IC50 of both HKa and domain 5, 50 nM). On the other hand, no change in the IC50 was observed for HK and domain 3 (100 nM and 480 nM, respectively). This is in accordance with the fact that HK and domain 3 do not bind to uPAR, whereas HKa and domain 5 bind to this protein (31)
. Thus, a portion of HKa and domain 5 is not available for binding to Mac-1 and thereby for inhibition of adhesion to FBG. In addition, the anti-CD18 blocking antibody showed a similar IC50 on both cell types.
|
PiPLC pretreatment of U937 monocytes resulted in removal of 80–90% of uPAR from the cell surface and in a reduction of Mac-1-dependent adhesion to FBG (11)
. Except for a slight decrease in IC50 of HKa and domain 5 (not shown), piPLC treatment did not further interfere with the ability of the different HK forms to block adherence to FBG (Fig. 8
A). Finally, the presence of soluble uPAR did not interfere with binding of biotin-labeled HK/HKa to immobilized Mac-1 (Fig. 8B
). Taken together, the antiadhesive effect of HK on Mac-1-dependent adhesion is mediated by the direct binding of kininogen to Mac-1 and not by the interaction of HK with uPAR, although a portion of HKa and domain 5 (the two HK forms that interact with uPAR) bind to uPAR on the cell surface and may therefore not be available for Mac-1 inhibition.
|
Inhibition of neutrophil recruitment by kininogen in the mouse model of acute thioglycollate-induced peritonitis
Thioglycollate induced peritonitis is a well-established mouse model for ß2-integrin-dependent leukocyte emigration into sites of acute inflammation. After thioglycollate stimulation, there is a significant increase in the total leukocyte count after 4 h in the peritoneum attributed predominantly to emigrated neutrophils; the portion of neutrophils among leukocytes is 50–60% at this time point, whereas 1 h after stimulation only 3–10% neutrophils have emigrated (38
, 39)
. The use of blocking antibodies against Mac-1 or LFA-1 30 min before the induction of peritonitis by thioglycollate resulted in a 60–75% inhibition of neutrophil extravasation into the inflamed peritoneum at the 4 h time point, whereas isotype-matched control antibody had no effect (Fig. 9
). Neutrophil recruitment into the peritoneum was significantly reduced in mice that were preinjected i.v. with GST domain 5, and maximal inhibition (70%) was obtained at 4 h with 100 µg per mouse of the HK fragment. Injection of GST alone did not affect neutrophil emigration to the peritoneum at all (Fig. 9)
. To prove the hypothesis that domain 5 blocks Mac-1, in vivo, the following experimental setting was used: Coinjection of blocking antibodies to Mac-1 and LFA-1 resulted in almost complete inhibition of neutrophil emigration. Mice treated with a mixture of antibody against Mac-1 and with domain 5 at subsaturating concentrations (domain 5 at 40 µg, anti-Mac-1 at 50 µg) showed an additive inhibition of neutrophil emigration (30–40% inhibition for each substance, 75% inhibition in combination), whereas no significant additive inhibition was observed when both substances were used at a concentration of 100 µg each. In contrast, the inhibition by domain 5 at 100 µg and the blocking antibody against LFA-1 at 100 µg were additive, providing an almost complete blockade of neutrophil emigration, reminiscent of the effect of coinjection of antibody against Mac-1 together with antibody against LFA-1 (Fig. 10
). Thus, there is in vivo evidence that domain 5 can block Mac-1-dependent neutrophil recruitment into inflamed tissues.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
including the induction of genes coding for adhesion receptors, their change in avidity as well as the modification of ligand binding affinities due to e.g., adhesion receptor clustering or sensitization by uPAR (7
, 8)
. Uncontrolled activation of leukocytes and/or endothelial cells results in pathological chronic inflammation causing atherosclerosis, rheumatoid arthritis and other disorders. In acute inflammation, Mac-1 and LFA-1 are the predominant integrins mediating leukocyte adhesion to endothelium by binding to their counter-receptor ICAM-1 (CD54). Mac-1 is constitutively expressed on neutrophils and monocytes, whereas LFA-1 is predominantly expressed on lymphocytes; however, recent data underline the important contribution of the latter integrin for neutrophil recruitment as well. Mac-1 is also a receptor for FBG, thereby mediating the adhesion and migration of neutrophils and monocytes onto fibrinogen/fibrin, a major provisional matrix component in inflamed or injured sites (41
42
43
44)
.
The present study demonstrates that single- or two-chain kininogen (HK or HKa) and particularly the domain 5 thereof specifically interact with Mac-1 but not with LFA-1, thereby blocking Mac-1-dependent leukocyte adhesion to FBG and to endothelial cells in vitro and interfering with neutrophil emigration during acute inflammation in vivo. The described antiinflammatory mechanisms of action of HK presented here can be of enormous importance considering the high plasma concentration of this factor (0.67 µM), as HK can serve to prevent or regulate excessive leukocyte recruitment and hyperinflammatory responses. Moreover, these mechanisms are in full agreement with the first description of a patient with high molecular weight kininogen (‘Fitzgerald factor’) deficiency, who showed excessive leukocyte migration using a skin window device (45
46)
. Moreover, the characterized peptide sequences in HK domain 5 involved in inhibition of Mac-1 function can provide new therapeutic substances to inhibit leukocyte infiltration in patients.
In a Zn2+-dependent manner HK, HKa, domain 5 (and domain 3 to a lesser extent) inhibited both, the binding of FBG and ICAM-1 to immobilized Mac-1, but did not affect the interaction of ICAM-1 with immobilized LFA-1. That kininogen specifically interacts with Mac-1 but not with LFA-1 was also shown by analyzing the adhesion of K562 erythroleukemic cells: Only cells that were transfected with Mac-1 were blocked in their ability to adhere to immobilized ICAM-1 by kininogen, whereas those transfected with LFA-1 remained unaffected. Our data are in accordance with previous reports showing that HK binds to neutrophils and competes with the binding of FBG (32
33
34)
and are also consistent with the requirement of both domains 3 and 5 for this association (47)
. Domain 5 plays a predominant role in mediating the effect of HK, since it is much more effective to displace FBG or ICAM-1 than domain 3. Although domain 3 is not able to completely block FBG or ICAM-1 binding to Mac-1, the interaction of domain 3 with Mac-1 plays a significant role, as evidenced by an additive inhibitory effect of both domains 3 and 5 when given in subthreshold concentrations. Both, the binding of HK to Mac-1 and the inhibitory effect of HK on the interaction between FBG or ICAM-1 with Mac-1 required the presence of Zn2+; maximal effects occurred around physiological concentrations of Zn2+ (<15 µM).
Due to their interaction with Mac-1, HK (and especially domain 5) could inhibit the adhesion of monocytes to FBG and endothelial cells. Furthermore, we extended these observations by providing evidence for the first time that HK/HKa or domain 5 can exert these activities in vivo: The emigration of neutrophils to the inflamed peritoneum in the mouse model of thioglycollate-induced peritonitis was effectively blocked by 70% due to the injection of the isolated domain 5 (as a GST fusion product). The extent of inhibition was comparable to the inhibitory effect observed with an antibody directed to Mac-1. Experiments using mixtures of domain 5 together with blocking antibodies against Mac-1 or LFA-1, respectively, were performed to prove the specificity of blockade of Mac-1 by domain 5 in vivo: The coinjection of domain 5 with anti-LFA-1 almost completely abolished neutrophil emigration, reminiscent of the effect of both ß2-integrin blocking antibodies together. Although ß2-integrin complexes are essential for neutrophil emigration, recent results from ß2-deficient mice demonstrated that Mac-1/LFA-1 independent pathways for cell recruitment could be used during acute inflammation in the peritoneum and the lung as well (48)
that need further characterization.
UPAR can regulate the activity particularly of ß2-integrins and thereby modulates leukocyte recruitment (7
8
9
10
11)
. Due to direct binding of HKa and domain 5 to uPAR (20
, 31)
, this glycolipid-anchored protein could be involved in the observed inhibition mechanisms, since ligation of uPAR by other ligands (such as antibodies or urokinase) induces conformational changes important for integrin function (11)
. No matter whether BAF-3 cells were transfected with uPAR or not, we could not detect a significant influence of the glycolipid-anchored protein on inhibition of cell adhesion by HK. Our data also indicate that the inhibition of Mac-1 function by HK or HKa is not mediated by the binding of kininogen to uPAR, but is due to the direct interaction of HK or HKa and Mac-1. This conclusion is also strengthened by the observation that 1) piPLC treatment of monocytes did hardly interfere with the ability of HK/HKa to block adhesion to FBG/endothelial cells, 2) isolated suPAR did not affect HK/HKa binding to Mac-1, and 3) inhibition of FBG and ICAM-1 binding to Mac-1 by the different kininogen forms was not affected in the presence of suPAR (not shown). Taken together, uPAR appears not to compete with HKa or domain 5 binding to Mac-1 but a portion of HKa or domain 5 interacts with cell surface-associated uPAR so that higher concentrations of the inhibitors are needed for Mac-1 blockade (Fig. 7)
. Finally, the possibility that the uPAR-HK interaction can be important in vivo should not be excluded, since uPAR serves a direct role in leukocyte adhesion to vitronectin, which is blocked by HK or domain 5 (31)
.
Using synthetic peptides the amino acid sequences responsible for the inhibitory effect of domain 5 on Mac-1 function were identified: The histidine-glycine-lysine rich region H475-G497 and to a lesser extent the histidine-glycine rich region G440-H455 reproduced the antiadhesive effect of domain 5. The former region has previously been described as endothelial cell binding region (22)
, whereas the latter region overlaps with one of two noncontiguous binding regions of domain 5 for neutrophils, which were recently identified by fine mapping (47)
. Moreover, the region H475-G497 also directly inhibited binding of HK/HKa to Mac-1, indicating that it is the major Mac-1 binding site on HK (with the region G440-H455 probably representing a secondary site). Scrambled peptides from the region H475-K502 had markedly reduced potency to inhibit HK binding to Mac-1 or Mac-1-dependent cell adhesion, indicating the sequence specificity for the indicated peptides. This is also strengthened by the fact that two similar peptides HK483 and HK486 had very different inhibitory potencies.
HKa also has a higher affinity for Mac-1 than HK, reflecting the conformational differences between both forms of kininogen that are likely to be due to the exposure of different binding sequences (e.g., H475-G497) in HKa. In accordance with Weisel et al. (49)
, HK appears to be in a curvilinear array of three linked globular regions (D1-D3, D5, and D6), whereas on kallikrein cleavage a conformational change takes place whereby HKa is composed of two disulfide-linked chains arranged in a triangular conformation accounting for the increased tendency of HKa to bind to cell surfaces and other proteins. Partial differences in the interaction of kininogens with Mac-1 in a purified system as opposed to neutrophils (47)
could be explained by additional HK/HKa binding proteins on these cells.
The region detected as the major Mac-1 binding site (H475-G497) in the present study is overlapping with the HK sequence (K480-T503) recently described to induce endothelial cell apoptosis and to inhibit angiogenesis (50)
. However, three major differences should be mentioned. 1) The peptide concentrations used to induce apoptosis are > 10-fold lower than those inhibiting adhesion. This could be because smaller amounts of HK are sufficient for interfering with signal transduction events leading to apoptosis, whereas the direct blockade of adhesion receptors on the cell surface requires more inhibitory molecules. 2) The proapoptotic effect is seen only with HKa, whereas in our studies HK can block Mac-1-dependent adhesion as well. 3) Induction of apoptosis is not mediated by any of the known receptors for kininogen and cannot be affected by integrins or their ligands, such as FBG, whereas in the present study HK competes with FBG for binding to the integrin Mac-1. Nevertheless, the HK region 475–503 reveals important features and further studies will demonstrate whether this portion can be used as leading substance for therapeutic interventions in hyperinflammatory or malignant situations in patients.
The present study provides new in vitro and in vivo evidence that HK/HKa, and particularly domain 5, can inhibit Mac-1-dependent leukocyte emigration and HK thereby serves as a potent antiinflammatory molecule. Since Mac-1 also 1) mediates phagocytosis and contributes to elevated natural killer cell activity against iC3b-coated target cells (51
, 52)
, 2) coordinates the activation of coagulation factor X independent of tissue factor and factor VII, leading to rapid fibrin formation (53
54
55)
, and 3) serves as binding partner for platelet ICAM-2 (56)
, potential inhibition of these multiple functions of Mac-1 by HK/HKa or domain 5 could interfere with neutrophil or monocyte activities. Moreover, locally generated HKa can balance the recruitment of leukocytes induced by bradykinin itself, thereby providing a physiological feedback mechanism. In addition, this study opens new approaches to interrupt one or more Mac-1 functions and will ultimately help in designing novel antiadhesive compounds for therapeutic interventions.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 30, 2001.
Revision received July 24, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Oehmcke, O. Shannon, M. von Kockritz-Blickwede, M. Morgelin, A. Linder, A. I. Olin, L. Bjorck, and H. Herwald Treatment of invasive streptococcal infection with a peptide derived from human high-molecular weight kininogen Blood, July 9, 2009; 114(2): 444 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. G. Coffman, J. C. Brown, D. A. Johnson, N. Parthasarathy, R. B. D'Agostino Jr., M. O. Lively, X. Hua, S. L. Tilley, W. Muller-Esterl, M. C. Willingham, et al. Cleavage of high-molecular-weight kininogen by elastase and tryptase is inhibited by ferritin Am J Physiol Lung Cell Mol Physiol, March 1, 2008; 294(3): L505 - L515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Khan, H. N. Bradford, I. Isordia-Salas, Y. Liu, Y. Wu, R. G. Espinola, B. Ghebrehiwet, and R. W. Colman High-Molecular-Weight Kininogen Fragments Stimulate the Secretion of Cytokines and Chemokines Through uPAR, Mac-1, and gC1qR in Monocytes Arterioscler Thromb Vasc Biol, October 1, 2006; 26(10): 2260 - 2266. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chavakis, A. Athanasopoulos, J.-S. Rhee, V. Orlova, T. Schmidt-Woll, A. Bierhaus, A. E. May, I. Celik, P. P. Nawroth, and K. T. Preissner Angiostatin is a novel anti-inflammatory factor by inhibiting leukocyte recruitment Blood, February 1, 2005; 105(3): 1036 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Kanse, R. L. Matz, K. T. Preissner, and K. Peter Promotion of Leukocyte Adhesion by a Novel Interaction Between Vitronectin and the {beta}2 Integrin Mac-1 ({alpha}M{beta}2, CD11b/CD18) Arterioscler Thromb Vasc Biol, December 1, 2004; 24(12): 2251 - 2256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. K. Lishko, N. P. Podolnikova, V. P. Yakubenko, S. Yakovlev, L. Medved, S. P. Yadav, and T. P. Ugarova Multiple Binding Sites in Fibrinogen for Integrin {alpha}M{beta}2 (Mac-1) J. Biol. Chem., October 22, 2004; 279(43): 44897 - 44906. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. J. H. Sachs, T. Chavakis, L. Fung, A. Lohrenz, J. Bux, A. Reil, A. Ruf, and S. Santoso Human alloantibody anti-Mart interferes with Mac-1-dependent leukocyte adhesion Blood, August 1, 2004; 104(3): 727 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Wetzel, T. Chavakis, K. T. Preissner, M. Sticherling, U.-F. Haustein, U. Anderegg, and A. Saalbach Human Thy-1 (CD90) on Activated Endothelial Cells Is a Counterreceptor for the Leukocyte Integrin Mac-1 (CD11b/CD18) J. Immunol., March 15, 2004; 172(6): 3850 - 3859. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Chavakis, A. Bierhaus, N. Al-Fakhri, D. Schneider, S. Witte, T. Linn, M. Nagashima, J. Morser, B. Arnold, K. T. Preissner, et al. The Pattern Recognition Receptor (RAGE) Is a Counterreceptor for Leukocyte Integrins: A Novel Pathway for Inflammatory Cell Recruitment J. Exp. Med., November 17, 2003; 198(10): 1507 - 1515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chavakis, S. Santoso, K. J. Clemetson, U. J. H. Sachs, I. Isordia-Salas, R. A. Pixley, P. P. Nawroth, R. W. Colman, and K. T. Preissner High Molecular Weight Kininogen Regulates Platelet-Leukocyte Interactions by Bridging Mac-1 and Glycoprotein Ib J. Biol. Chem., November 14, 2003; 278(46): 45375 - 45381. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Zhao, J.-L. Lew, L. Huang, J. Yu, T. Zhang, Y. Hrywna, J. R. Thompson, N. de Pedro, R. A. Blevins, F. Pelaez, et al. Human Kininogen Gene Is Transactivated by the Farnesoid X Receptor J. Biol. Chem., August 1, 2003; 278(31): 28765 - 28770. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Santoso, U. J.H. Sachs, H. Kroll, M. Linder, A. Ruf, K. T. Preissner, and T. Chavakis The Junctional Adhesion Molecule 3 (JAM-3) on Human Platelets is a Counterreceptor for the Leukocyte Integrin Mac-1 J. Exp. Med., September 2, 2002; 196(5): 679 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chavakis, R. A. Pixley, I. Isordia-Salas, R. W. Colman, and K. T. Preissner A Novel Antithrombotic Role for High Molecular Weight Kininogen as Inhibitor of Plasminogen Activator Inhibitor-1 Function J. Biol. Chem., August 30, 2002; 277(36): 32677 - 32682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Chavakis, N. Boeckel, S. Santoso, R. Voss, I. Isordia-Salas, R. A. Pixley, E. Morgenstern, R. W. Colman, and K. T. Preissner Inhibition of Platelet Adhesion and Aggregation by a Defined Region (Gly-486-Lys-502) of High Molecular Weight Kininogen J. Biol. Chem., June 21, 2002; 277(26): 23157 - 23164. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
