Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The chemokine family is composed of a number of structurally and functionally related chemotactic cytokines that regulate cell migration during healthy and pathological processes, including angiogenesis, hematopoiesis, infection, and cancer (Wuyts et al., 1999; Murphy et al., 2000; Rossi and Zlotnik, 2000). In particular, chemokines are key mediators of leukocyte migration during both the inflammatory response and normal homeostasis. The classification of chemokines in C, CC, CXC, and CX₃C subfamilies, according to the position of conserved cysteine residues, provides only a limited help to distinguish the functional activities of these mediators. Indeed, although many CXC chemokines [e.g., interleukin-8 (IL-8)], preferentially activate and chemoattract neutrophilic granulocytes, other members of this subfamily selectively stimulate migration of lymphocytes. Inversely, whereas most CC chemokines do not act on neutrophils, some [e.g., monocyte chemotactic protein-3 (MCP-3)] weakly chemoattract neutrophils in addition to monocytes and lymphocytes (Xu et al., 1995). This divergence in spectrum of target cells is dictated for each individual chemokine by the expression of specific G protein-coupled seven-transmembrane receptors that are recognized by the chemokine to exert its biological activities. In this respect, the CC chemokine receptor CCR1 is expressed on neutrophils, in addition to the CXC chemokine receptors CXCR1 and CXCR2 (Crisman et al., 1999; Zhang et al., 1999). Chemokine receptor recognition and hence biological potency is strongly affected by NH₂-terminal processing of chemokines by proteases (Van Damme et al., 1999; Van den Steen et al., 2000; Struyf et al., 2001a).

Regakine-1 is a recently discovered CC chemokine that is constitutively present at high concentrations in bovine plasma and to which neutrophil chemotactic activity has been ascribed (Struyf et al., 2001b). Although the natural 7.5-kDa protein of 70 residues contains the four conserved cysteines, it has less than 50% sequence identity with any known human or animal chemokine. No specific receptors for regakine-1 have been identified and no high affinity for the classical receptors on neutrophils (i.e., CXCR1 and CXCR2) was detected for regakine-1. However, a particular characteristic of the plasma-derived regakine-1 resides in its potential to synergize with IL-8, a ligand for both CXCR1 and CXCR2, to chemoattract neutrophils (Struyf et al., 2001b). CXCR2 is also functionally expressed on nonhematopoietic cells and accounts for the angiogenic activity of IL-8 and related CXCR2 agonists (Addison et al., 2000; Devalaraja et al., 2000). To better delineate the unexpected activity of this novel CC chemokine, parallel experiments were performed with either selective CXCR2 agonists [i.e., neutrophil activating protein-2 (NAP-2) or inflammatory mediators other than chemokines (e.g., complement factor C5a), both also present in human plasma]. It was found that NAP-2 and C5a synergize with regakine-1 in neutrophil chemotaxis, whereas other plasma- or tissue-derived CC chemokines failed to cooperate with C5a or regakine-1. This study demonstrates that regakine-1 possesses the unique capability to potentiate the inflammatory response in the blood circulation. It can be speculated that neutralization of this chemokine can reduce neutrophil-mediated injury.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cells. Neutrophilic granulocytes were isolated from buffy coats, freshly derived from blood of healthy donors (Blood Transfusion Center of Antwerp and Laboratory of Experimental Immunology, University of Leuven). Mononuclear cells and granulocytes were separated by gradient centrifugation for 30 min at 400g on Ficoll-sodium metrizoate (Lymphoprep; Invitrogen, Groningen, The Netherlands). Erythrocytes in the granulocyte pellet were removed by sedimentation for 30 min at 37°C in hydroxyethyl-starch solution (Plasmasteril; Fresenius AG, Bad Homburg, Germany). Remaining erythrocytes were lysed by hypotonic shock (30 s) in bidistilled water. Neutrophilic granulocytes were used in the chemotaxis assay at a concentration of 10⁶ cells/ml in Hanks' balanced salt solution (Invitrogen) supplemented with 1 mg/ml of human serum albumin (Belgian Red Cross).

Chemoattractants. Regakine-1 was isolated from fetal calf serum (Invitrogen) derived from Bos taurus by subsequent adsorption to silicic acid, heparin-Sepharose affinity chromatography, mono-S cation-exchange chromatography (Amersham Biosciences, Uppsala, Sweden) and reversed phase-high performance liquid chromatography (RP-HPLC) (Perkin-Elmer, Norwalk, CT) as described previously (Struyf et al., 2001b). Natural human IL-8 (CXCL8) and human NAP-2 (CXCL7) were purified to homogeneity from monocyte-derived conditioned medium (Van Damme et al., 1989). A fraction containing natural connective tissue-activating peptide III (CTAP-III) (CXCL7) was purified from blood platelets (Van Damme et al., 1989). Natural chemokines contained less than 2.5 pg of lipopolysaccharide/µg of protein, as determined in the Limulus amoebocyte lysate assay (QCL chromogenic method; BioWhittaker). Recombinant human hemofiltrate CC chemokine-1 (HCC-1, CCL14) was purchased from Peprotech (Rocky Hill, NJ) and the bacterial-derived chemotactic peptide fMLP was obtained from Sigma (St. Louis, MO). The CC chemokines MCP-3 (CCL7) and MIP-1/LD78 (CCL3-L1) were synthesized by solid-phase peptide synthesis using fluorenylmethoxycarbonyl chemistry (see next paragraph) as described previously (Struyf et al., 2001a). The anaphylatoxin C5a was either obtained from Sigma (recombinant anaphylatoxin C5a) or purified from human plasma by heparin-Sepharose affinity chromatography (Amersham Biosciences), Resource S cation-exchange chromatography (Amersham Biosciences) and RP-HPLC on a Resource RPC column (Amersham Biosciences). For identification, the NH₂-terminal residues of C5a (TLQKKIEEIA) were determined by Edman degradation on a capillary protein sequencer (Procise 491cLC; Applied Biosystems, Foster City, CA) and corresponded to that of intact C5a. The average relative molecular mass of natural C5a was determined by ion trap mass spectrometry (Esquire LC, Bruker Daltonik, Bremen, Germany) to be 10437 Da. Because the average molecular mass of unglycosylated C5a is 8268 Da, considerable posttranslational modification by glycosylation is evident.

Chemical Synthesis of Different Forms of Regakine-1. Intact regakine-1 and NH₂-terminally truncated forms missing the first two, four, or eight amino acids, designated regakine-1(1-70), regakine-1(3-70), regakine-1(5-70), and regakine-1(9-70), respectively, were chemically synthesized in a single run using amino acids with 9-fluorenylmethoxy-carbonyl-protected -amino groups on a model 433A solid-phase peptide synthesizer using the standard FastMoc programs with conditional double coupling and acetic anhydride capping (Applied Biosystems). The final deprotection and cleavage of the peptide from the resin was performed by incubating the synthesis product for 2 h at room temperature in the following cleavage mixture: 10 ml of trifluoroacetic acid, 0.5 ml of water, 0.5 ml of thioanisole, 0.25 ml of ethanedithiol, and 0.75 g of crystalline phenol. The synthetic chemokine was separated from the resin on a medium-porosity glass filter, precipitated into cold methyl t-butyl ether, washed, dissolved in water, and subsequently lyophilized. Crude full-length and shorter isoforms of synthetic regakine-1 were separated from peptide fragments by RP-HPLC on a Resource RPC column (Amersham Biosciences). After purification, the disulfide bridges were formed by incubation at room temperature for 1.5 h in 150 mM Tris, pH 8.6; 2 M ureum, 3 mM EDTA, 0.3 mM oxidized glutathione, and 3 mM reduced glutathione. The folded peptides were repurified by RP-HPLC on an Aquapore RP-300 column (Applied Biosystems). The purity and the molecular mass of folded intact or truncated regakine-1 proteins were confirmed by ion trap electrospray mass spectrometry (Fig. 1). For all regakine-1 forms, the experimentally determined average molecular mass differed by less than 1 Da from the theoretical average molecular mass, indicative of a correct synthesis and folding (Table 1). Purified synthetic regakine-1 isoforms were free of detectable lipopolysaccharide (< 2.5 pg of lipopolysaccharide/µg of regakine-1).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Mass spectrometry of folded synthetic Regakine-1 forms. RP-HPLC purified and folded synthetic regakine-1(9-70) (A), regakine-1(5-70) (B), regakine-1(3-70) (C), and regakine-1(1-70) (D) were subjected to electrospray ion trap mass spectrometry. Left, average of 800 to 1000 spectra of the multiple charged ions resulting in an expected error for the molecular mass of the uncharged molecules of less than 1 Da. The m/z values are indicated above the peaks that are derived from the regakine-1 proteins, whereas the number of charges for the multiple charged ions are indicated between brackets. The right part of the figure shows the uncharged deconvoluted spectra for the different regakine-1 forms with the average molecular mass indicated on top of the peaks.

Chemotaxis Assay. Cell migration was measured with the Boyden microchamber technique (Neuro Probe, Gaithersburg, MD). Cell fractions and samples were diluted in Hanks' balanced salt solution supplemented with 1 mg/ml human serum albumin (dilution buffer) and tested in triplicate. The lower compartment, containing the test sample or control dilution buffer, was separated from the upper compartment, containing neutrophils (1 × 10⁶ cells/ml), by a polyvinylpyrrolidone-free polycarbonate membrane with a 5-µm pore size (Nuclepore; Corning Costar, Acton, MA). After migration (45 min at 37°C) the filters were incubated, fixed and stained using Hemacolor solutions (Merck, Darmstadt, Germany). The migrated cells were counted microscopically in 10 oil immersion fields at a 500× magnification. The chemotactic activity was expressed as a chemotactic index (CI), the number of cells migrated to the test sample divided by the number of cells migrated to the dilution buffer (negative control). In experiments in which the synergistic effect of two chemoattractants was investigated, both molecules were added to the lower wells of the microchamber. In desensitization experiments, neutrophils were preincubated with regakine-1 for 10 min at 37°C and subsequently washed twice with dilution buffer before transfer of the cells to the chemotaxis chamber. To study the effect of protein kinase inhibitors on the synergistic effect, neutrophils were treated with the kinase inhibitors PD98059 (Calbiochem Merck Eurolab, Lutterworth, UK), wortmannin (Sigma-Aldrich, St. Louis, MO) or staurosporin (Sigma) and loaded into the upper wells of the Boyden chamber. Statistical differences between chemotactic indexes were determined by the Mann-Whitney U test.

Binding Assays. Competition for fMLP binding was measured using purified neutrophils or FPR transfected rat basophilic leukemia cells (RBL cells) as described previously (Le et al., 1999). Briefly, a cell suspension [FPR/RBL cells, 2 × 10⁶ cells/200 µl in RPMI 1640 medium containing 2 mg/ml bovine serum albumin (BSA; Sigma) and 5 mg/ml NaN₃] was incubated for 30 min at 37°C under constant rotation with 30 nM [³H]fMLP (PerkinElmer Life Sciences, Boston, MA) and varying concentrations of unlabeled fMLP or regakine-1. Each experiment was done in duplicate. After incubation, the samples were filtered onto Whatman GF/C discs (Whatman International, Kent, UK) on a 12-well manifold, followed by washing three times with 5 ml of ice-cold phosphate-buffered saline (PBS). The discs were air-dried, submerged in liquid scintillation fluid, and counted for -emission.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Synergy between Plasma-Derived Regakine-1 and NAP-2 in Neutrophil Chemotaxis. It has previously been observed that the chemotactic effect of IL-8 on neutrophils was significantly enhanced in the presence of the CC chemokine regakine-1 (Struyf et al., 2001b). To verify whether this synergistic effect is mediated specifically via one of the IL-8 receptors (i.e., CXCR2), the platelet-derived CXCR2 ligand NAP-2 was tested in combination with plasma-derived regakine-1. Regakine-1 alone had a weak but statistically significant neutrophil chemotactic activity (e.g., CI ± S.E.M. of 6.4 ± 2.8 and 5.3 ± 2.4 at 100 and 30 ng/ml, respectively; p = 0.01 versus control buffer) (Fig. 2). Regakine-1 dose-dependently increased the neutrophil chemotactic activity of the chemokine NAP-2 (Fig. 2). Indeed an agonistic concentration (30 or 100 ng/ml) of regakine-1 enhanced the chemotactic index of NAP-2 (30 and 100 ng/ml) 3- to 6-fold above the additive effect of the two agonists. For comparison, it was confirmed that the neutrophil chemotactic activity of IL-8 (5 ng/ml) was also enhanced by regakine-1 (100 ng/ml). It can be deduced that different CXC chemokines binding to CXCR2 (i.e., IL-8 and NAP-2) cooperate with regakine-1 in neutrophil chemotaxis experiments.

View larger version (66K): [in this window] [in a new window]   Fig. 2.   Regakine-1 synergizes with the CXC chemokine NAP-2 in neutrophil chemotaxis. NAP-2 () and IL-8 () were combined with different concentrations of regakine-1 in the lower compartment of the microchamber to measure human neutrophil chemotaxis. The chemotactic response is expressed as the mean CI ± S.E.M., derived from two to five (NAP-2) or two to seven (IL-8) independent experiments. On average, the S.E.M. did not exceed 30% of the mean CI and is not shown for clarity. Statistically significant differences in chemotactic indexes between NAP-2 alone or in combination with regakine-1, determined by the Mann-Whitney test, are indicated by  (p < 0.05).

Enhanced Inflammatory Response to C5a and fMLP in the Presence of Regakine-1. To verify whether a synergistic effect also exists between regakine-1 and neutrophil-activating inflammatory mediators other than chemokines, similar chemotaxis experiments were executed with the anaphylatoxin C5a. This potent chemoattractant, found in serum during septic shock, activates neutrophils expressing the unique C5a receptor (Gerard and Gerard, 1991). Natural C5a was purified to homogeneity from human plasma (see Materials and Methods). In Fig. 3, it is shown that regakine-1 (30 or 100 ng/ml) significantly enhanced the neutrophil chemotactic response toward C5a (30 or 100 ng/ml). With this combination, a maximal neutrophil chemotactic index of more than 100 was reached, whereas regakine-1 and C5a alone elicited maximal indexes of about 3 and 30, respectively. Similarly, regakine-1 was also capable to synergize with the bacterial peptide fMLP (Fig. 3).

View larger version (66K): [in this window] [in a new window]   Fig. 3.   Regakine-1 cooporates with C5a in neutrophil chemotaxis. C5a (), fMLP () or IL-8 () were combined with different concentrations of regakine-1 in the lower compartment of the microchamber to measure human neutrophil chemotaxis. The chemotactic response is expressed as the mean CI ± S.E.M., derived from four independent experiments. On average, the S.E.M. did not exceed 30% of the mean CI and is not shown for clarity. Statistically significant differences in chemotactic indexes between C5a, fMLP, or IL-8 alone or combined with regakine-1, determined by the Mann-Whitney test, are indicated by  (p < 0.05).

Lack of Synergy between C5a and Plasma Chemokines Other Than Regakine-1. In a further attempt to precisely delineate the spectrum of synergy between plasma chemokines and C5a, other chemokines constitutively present in plasma were evaluated for their synergistic capacity in C5a-induced neutrophil chemotaxis (Table 3). CTAP-III (15 ng/ml), NAP-2 (30 ng/ml), and HCC-1 (300 ng/ml) failed to enhance the chemotactic effect of C5a at a concentration (30 ng/ml) that synergized with regakine-1 (Fig. 3). This demonstrates that the neutrophil chemotactic effect of C5a is selectively enhanced by the CC chemokine regakine-1 and not by other plasma CC (HCC-1) or CXC (CTAP-III, NAP-2) chemokines. Indeed, only cumulative chemotactic indexes were reached when each of the three chemokines was combined with C5a (Table 3). Thus, unlike regakine-1, the CXCR2 agonist NAP-2 and the CC chemokine HCC-1, to which no neutrophil chemotactic activity has been ascribed, did not synergize with C5a when combined at concentrations present in the circulation.

The CC Chemokines MCP-3 and MIP-1/LD78 Lack Synergistic Activity with Neutrophil Chemoattractants, Including Regakine-1. To obtain further evidence for the specificity of the synergy observed with regakine-1, experiments were performed with MCP-3, another CC chemokine with weak neutrophil chemotactic potency (Xu et al., 1995). Table 4 shows that, in contrast to regakine-1 (Fig. 3), biologically active MCP-3 (30 and 100 ng/ml) did not enhance the neutrophil chemotactic index of C5a (10 and 30 ng/ml). In addition, MCP-3 also failed to synergize (Table 4) with fMLP (10 nM) or IL-8 (5 ng/ml), whereas the latter cooperates with regakine-1 in neutrophil chemotaxis (Fig. 2 and 3). These data further point to the unique capacity of regakine-1 to synergize with other neutrophil chemoattractants.

View this table: [in this window] [in a new window]   TABLE 5 Regakine-1 fails to synergize with other neutrophil chemoattracting CC chemokines in neutrophil chemotaxis MCP-3 or MIP-1/LD78 were combined with different concentrations of regakine-1 in the lower compartment of the microchamber to measure human neutrophil chemotaxis. The chemotactic response is expressed as the mean CI ± S.E.M., derived from four independent experiments.

Regakine-1 Does Not Compete for fMLP or C5a Binding Sites on Neutrophils and Receptor Transfectants. Because regakine-1 synergized with both fMLP and C5a, it was verified whether regakine-1 acts on the receptors of these chemoattractants on neutrophils or receptor-transfected cell lines. Regakine-1, in contrast to unlabeled fMLP, did not displace [³H]fMLP from RBL cells transfected with the high affinity fMLP receptor at concentrations as high as 750 ng/ml (Fig. 4, A and B). In addition, 1 nM of fMLP inhibited binding of [³H]fMLP to freshly isolated neutrophils for more than 80%, whereas regakine-1 at 750 ng/ml did not ( 10% reduction) affect [³H]fMLP binding (data not shown). Similarly, it was found that different concentrations of regakine-1 were unable to displace ¹²⁵I-C5a from neutrophils, whereas unlabeled recombinant C5a was dose-dependently preventing the binding of ¹²⁵I-C5a (Fig. 4, C and D).

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Regakine-1 fails to compete for [3H]fMLP binding or 125I-C5a binding. RBL cells transfected with the high affinity fMLP receptor were incubated with 30 nM [3H]fMLP and varying concentrations of regakine-1 (A: , natural regakine-1; , synthetic regakine-1) or unlabeled fMLP (B: ). Results are expressed as the percentage of [3H]fMLP specific binding (mean ± S.E.M. of two to four independent experiments). Neutrophils were incubated with 0.04 nM 125I-C5a and different concentrations of natural regakine-1 (C: ) or unlabeled recombinant C5a (D: ). Results are expressed as the percentage of 125I-C5a specific binding (mean ± S.E.M. of two to five independent experiments).

The Synergistic Activity of Regakine-1 Does Not Reside in Its NH₂-Terminal Region. For many CXC and CC chemokines, the NH₂-terminal region determines their binding capacity to G protein-coupled receptors. Post-translational processing of chemokines by proteases resulted in either reduced (e.g., MCP-1, MCP-2, eotaxin) or enhanced (IL-8, NAP-2, HCC-1, MIP-1/LD78) chemotactic potencies (Detheux et al., 2000; Van Damme et al., 1999; Van den Steen et al., 2000; Struyf et al., 2001a). Differently truncated isoforms of regakine-1 lacking two, four, or eight residues at the NH₂ terminus were therefore chemically synthesized and compared for their synergistic effect with C5a on neutrophil chemotaxis. It was found that regardless of the length of their NH₂-terminal region, all regakine-1 isoforms were equipotent at enhancing the chemotactic response of neutrophils to C5a (Fig. 5). For comparison, it was shown (Table 3) that both NAP-2 and its precursor CTAP-III did not synergize with C5a, whereas NAP-2 did cooperate with regakine-1 to chemoattract neutrophils (Fig. 2)

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Synergistic effect between NH2-terminally truncated forms of regakine-1 and C5a. Synthetic intact regakine-1(1-70) and the truncated isoforms regakine-1(3-70), regakine-1(5-70) and regakine-1(9-70), missing two, four, or eight amino acids, respectively, at the NH2 terminus, were combined with C5a (30 ng/ml) in the microchamber assay to measure neutrophil chemotaxis. Results represent the mean CI (± S.E.M.) of six or seven independent experiments. Statistically significant differences in chemotactic indexes between C5a alone or combined with regakine-1, determined by the Mann-Whitney test, are indicated by  (p < 0.05) and (p < 0.01).

The Effect of Protein Kinase Inhibitors on the Synergy between C5a and Regakine-1. To determine whether protein kinases mediate the chemotactic activity of regakine-1 and its enhancing effect on neutrophil responses to C5a, neutrophils were treated with different concentrations of either PD98059 (a p44/42 mitogen-activated protein kinase inhibitor), wortmannin (a phosphatidylinositol-3-kinase inhibitor) or staurosporin (a protein kinase A and C inhibitor) before transfer to the upper wells of the Boyden chamber. Figure 6 illustrates that the chemotactic response to 30 ng/ml C5a or 30 ng/ml C5a in combination with 100 ng/ml regakine-1 was not inhibited by any of these inhibitors. Nevertheless, a significant inhibition of 5 ng/ml IL-8-induced chemotaxis was observed with 50 µM PD98059. On the other hand, wortmannin and staurosporin (100 nM) did not inhibit IL-8-induced neutrophil chemotaxis.

View larger version (35K): [in this window] [in a new window]   Fig. 6.   The synergy between regakine-1 and C5a in neutrophil chemotaxis is not affected by protein kinase inhibitors. Regakine-1, C5a, a combination of both, or IL-8 was added to the lower compartment of the chemotaxis microchamber. The combination of C5a and regakine-1 yielded a statistically significant synergy in neutrophil chemotaxis (not indicated). Neutrophils were treated with PD98059 at 10 and 50 µM (A; n = 5), wortmannin at 10 and 100 nM (B; n = 9), staurosporin at 10 and 100 nM (C; n = 8), or were left untreated before transfer to the upper wells of the microchamber. The chemotactic indexes were calculated using the appropriate controls (individual protein kinase inhibitor treated versus untreated control cells) and are expressed as the mean CI ± S.E.M. Statistically significant differences between chemotactic responses in the presence or absence of protein kinase inhibitors were calculated with the Mann-Whitney test and are indicated by (p < 0.01).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Within the family of chemotactic cytokines, the classification as CXC or CC chemokine is based on biochemical structure rather than on functional properties. Consequently, inducible inflammatory and constitutive homeostatic chemokines belong to both the CXC and CC subfamilies, defined by the positioning of conserved cysteine residues in their sequence. The corresponding receptors were designated CXCR and CCR. Several of these receptors have multiple ligands, and one chemokine is often capable of binding to different receptors (Wuyts et al., 1999; Murphy et al., 2000; Rossi and Zlotnik, 2000). Only a few chemokines are constitutively present in normal plasma. These include the platelet-derived CXC chemokines platelet factor-4 and -thromboglobulin, known for more than 2 decades (Deuel et al., 1977; Begg et al., 1978), and the recently discovered CC chemokines HCC-1 and regakine-1 (Schulz-Knappe et al., 1996; Struyf et al., 2001b). For platelet factor-4 and regakine-1, no receptors have been identified, whereas for -thromboglobulin and HCC-1, post-translational processing by proteases is essential to become functionally active as the CXCR2 ligand NAP-2 (Walz et al., 1989) and the CCR1, -3, -5 ligand HCC-1(9-74) (Detheux et al., 2000), respectively. Such increase in biological potency by NH₂-terminal processing is a common phenomenon for other CXCR2 ligands [e.g., IL-8 after cleavage by gelatinase B (Van den Steen et al., 2000)], as well as for other CCR1 ligands such as MIP-1/LD78 (Struyf et al., 2001a) after processing by CD26/dipeptidyl peptidase IV (CD26/DPP IV). However, for most CC chemokines, removal of the NH₂-terminal dipeptide by CD26/DPP IV results in a partial or complete loss of receptor recognition and hence biological activity (Van Damme et al., 1999).

In this study, we have investigated the impact of NH₂-terminal truncation of regakine-1 on its unique property to attract neutrophils and to synergize with other proinflammatory mediators. Indeed, it was found that at a physiological concentration (30 ng/ml), regakine-1 synergized with the selective CXCR2 agonist NAP-2. Furthermore, significantly enhanced neutrophil chemotaxis was observed when regakine-1 was combined with the classical chemoattractant C5a at a concentration (30 ng/ml) detected in the blood circulation during an inflammatory response (Hogasen et al., 1995). Deletion of the two, four, or eight NH₂-terminal residues of regakine-1 (in front of the CC motif) neither enhanced the neutrophil chemotactic potency of regakine-1 nor altered its synergy with C5a. The relevance of this synergistic effect between regakine-1 and other chemoattractants binding to G protein-coupled receptors, including also the bacterial peptide fMLP, was endorsed by the finding that other CC chemokines with neutrophil chemotactic activity, such as MCP-3, did not possess the characteristic of regakine-1 to cooperate with C5a, fMLP, or IL-8 on neutrophils. The weak neutrophil agonists (MCP-3, MIP-1/LD78) did also not enhance the neutrophil chemotactic potency of regakine-1. Furthermore, other CXC (CTAP-III) and CC (intact HCC-1) plasma chemokines with poor chemokine receptor binding capacity and hence chemotactic activity failed to synergize with regakine-1.

It is expected that the synergy between regakine-1 and IL-8, NAP-2, C5a, or fMLP is dependent on the binding of these latter agonists to their known G protein-coupled receptors. To unravel whether the synergy depended on regakine-1 binding to the same receptor(s), binding competition experiments were executed. It was found that regakine-1 was unable to displace labeled C5a or fMLP from binding to neutrophils or cells transfected with the corresponding receptor. Furthermore, in an attempt to desensitize neutrophils by pretreatment with regakine-1, it was found that no inhibition of the calcium flux and chemotaxis induced by other chemokines, fMLP or C5a, could be obtained. Moreover, regakine-1 on its own was unable to induce a calcium signal. In this respect, regakine-1 behaved differently than other CC chemokines such as MIP-1, which induced a weak but consistent calcium flux in neutrophils (McColl et al., 1993). This effect was even more pronounced after NH₂-terminal truncation of MIP-1/LD78 by CD26/DPP IV (Struyf et al., 2001a). Different biological responses were observed with leukotactin-1, another CC chemokine that chemoattracts neutrophils. In contrast to regakine-1 and MIP-1, leukotactin-1 induced a potent calcium signal in neutrophils despite the fact that it shares CCR1, the functional receptor on these cells, with MIP-1/LD78 (Zhang et al., 1999). The lack of calcium signaling observed with regakine-1 is not unique, in that other CC chemokines and NH₂-terminally truncated chemokine isoforms that still bind to their receptor have been reported to induce weak or marginal calcium signals (Pettit and Fay, 1998; Blanpain et al., 1999; Van Damme et al., 1999). These NH₂-terminally processed chemokines are devoid of any chemotactic activity and function as natural inhibitors of the intact chemokines. Intact regakine-1 induced the release of gelatinase B from neutrophils and enhanced the shape change induced by C5a, confirming the role of regakine-1 in inflammation (data not shown). Finally, protein kinase inhibitors, such as PD98059, wortmannin, and staurosporin, had no effect on the synergy between regakine-1 and C5a. In all probability, protein kinase signaling is not the major pathway mediating the synergy between regakine-1 and C5a. However, the literature on the effect of protein kinase inhibitors in neutrophil chemotaxis is rather contradictory (Thelen et al., 1995; Knall et al., 1997; Zu et al., 1998; Nagata et al., 2001).

Taken together, our findings demonstrate that regakine-1, a CC chemokine constitutively present in plasma, but not other plasma or tissue chemokines, synergizes with various unrelated chemoattractants that bind G protein-coupled receptors. This is indicative of its unique role as amplifier of the inflammatory response. Synergistic effects in vitro and in vivo have been described between functionally related cytokines such as IL-1 and tumor necrosis factor- (Movat et al., 1987; Okusawa et al., 1988) and for IL-1 and interferon- (Struyf et al., 1998). Although for these cytokines, enhanced production of second messengers, including chemokines, is implicated, the precise molecular mechanisms remain partially unknown. For regakine-1, one can speculate that the synergy occurs at the extracellular level. The fact that regakine-1 does not interfere with binding to the G protein-coupled receptor of the cooperative chemoattractants implies a separate regakine-1 receptor. In addition, because chemokine receptor heterodimerization (e.g., between CCR2 and CCR5) has been demonstrated to increase the sensitivity and dynamics of the chemokine response (Mellado et al., 2001), this phenomenon might also be applicable to regakine-1. The optimal concentrations for synergy between regakine-1 and C5a are physiological, which suggests that this mechanism effectively occurs in the in vivo situation. Blockade of regakine-1 in the plasma can therefore be considered to down-modulate systemic inflammatory diseases, such as septic shock.