NH2- and COOH-Terminal Truncations of Murine Granulocyte Chemotactic Protein-2 Augment the In Vitro and In Vivo Neutrophil Chemotactic Potency

Transcription

1 This information is current as of June 11, References Subscriptions Permissions Alerts NH2- and COOH-Terminal Truncations of Murine Granulocyte Chemotactic Protein-2 Augment the In Vitro and In Vivo Neutrophil Chemotactic Potency Anja Wuyts, Anne D'Haese, Valerie Cremers, Patricia Menten, Jean-Pierre Lenaerts, Arnold De Loof, Hubertine Heremans, Paul Proost and Jo Van Damme J Immunol 1999; 163: ; ; This article cites 44 articles, 27 of which you can access for free at: Information about subscribing to The Journal of Immunology is online at: Submit copyright permission requests at: Receive free -alerts when new articles cite this article. Sign up at: The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 9650 Rockville Pike, Bethesda, MD Copyright 1999 by The American Association of Immunologists All rights reserved. Print ISSN: Online ISSN:

2 NH 2 - and COOH-Terminal Truncations of Murine Granulocyte Chemotactic Protein-2 Augment the In Vitro and In Vivo Neutrophil Chemotactic Potency 1 Anja Wuyts, 2 * Anne D Haese, Valerie Cremers,* Patricia Menten,* Jean-Pierre Lenaerts,* Arnold De Loof, Hubertine Heremans, Paul Proost,* and Jo Van Damme* Chemokines are important mediators of leukocyte migration during the inflammatory response. Post-translational modifications affect the biological potency of chemokines. In addition to previously identified NH 2 -terminally truncated forms, COOH-terminally truncated forms of the CXC chemokine murine granulocyte chemotactic protein-2 (GCP-2) were purified from conditioned medium of stimulated fibroblasts. The truncations generated 28 natural murine GCP-2 isoforms containing residues, including most intermediate forms. Both NH 2 - and COOH-terminal truncations of GCP-2 resulted in enhanced chemotactic potency for human and murine neutrophils in vitro. The truncated isoform GCP-2(9 78) was 30-fold more potent than intact GCP-2(1 92)/LPS-induced CXC chemokine (LIX) at inducing an intracellular calcium increase in human neutrophils. After intradermal injection in mice, GCP-2(9 78) was also more effective than GCP-2(1 92)/LIX at inducing neutrophil infiltration. Similar to human IL-8 and GCP-2, murine GCP-2(9 78) and macrophage inflammatory protein-2 (MIP-2) induced calcium increases in both CXCR1 and CXCR2 transfectants. Murine GCP-2(9 78) could desensitize the calcium response induced by MIP-2 in human neutrophils and vice versa. Furthermore, MIP-2 and truncated GCP-2(9 78), but not intact GCP-2(1 92)/LIX, partially desensitized the calcium response to human IL-8 in human neutrophils. Taken together, these findings point to an important role of post-translationally modified GCP-2 to replace IL-8 in the mouse. The Journal of Immunology, 1999, 163: The migration of leukocytes from blood vessels into tissues is an essential part of the inflammatory response. Adhesion molecules and chemokines are key mediators for leukocyte migration. Chemokines are low m.w. proteins that specifically attract subsets of leukocytes. They are divided into four subfamilies depending on the position of the conserved cysteines (CXC, CC, C, and CX 3 C chemokines) (1 3). The CXC chemokine subfamily can further be divided into ELR and ELR CXC chemokines depending on the presence or the absence of the Glu-Leu- Arg (ELR) 3 motif, respectively. This motif, positioned in the NH 2 - terminal region just in front of the first cysteine residue, is Laboratories of *Molecular Immunology and Immunobiology, Rega Institute for Medical Research, and Laboratory for Developmental Physiology and Molecular Biology, Zoological Institute, University of Leuven, Leuven, Belgium Received for publication April 2, Accepted for publication September 13, The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by the Fund for Scientific Research of Flanders, the Concerted Research Actions of the Regional Government of Flanders, the InterUniversity Attraction Pole initiative of the Belgian Federal Government, and the Biomed Program of the European Community. P.P. is a senior research assistant with the Fund for Scientific Research of Flanders. 2 Address correspondence and reprint requests to Dr. Anja Wuyts, Rega Institute, Minderbroedersstraat 10, B-3000 Leuven, Belgium. address: anja.wuyts@ rega.kuleuven.ac.be 3 Abbreviations used in this paper: ELR motif, Glu-Leu-Arg motif; [Ca 2 ] i, intracellular calcium concentration; CXCR, CXC chemokine receptor; EMEM, Eagle s MEM with Earle s salts; FPLC, fast protein liquid chromatography; GCP-2, granulocyte chemotactic protein-2; HEK cells, human embryonic kidney cells; IL-8Rh, IL-8R homologue; LIX, LPS-induced CXC chemokine; MALDI-MS, matrix-assisted laser desorption/ionization mass spectrometry; MIP-2, macrophage inflammatory protein-2; NAP-2, neutrophil-activating protein-2; poly ri:rc, poly(riboinosinic acid) poly(ribocytidylic acid); PVDF, polyvinylidene fluoride; RP-HPLC, reverse phase HPLC. important for receptor binding (4). ELR CXC chemokines are the most selective chemotactic factors for neutrophil migration. In the human system, seven ELR CXC chemokines have been identified: IL-8; GRO, -, and - ; neutrophil-activating protein-2 (NAP-2); epithelial cell-derived neutrophil attractant-78; and granulocyte chemotactic protein-2 (GCP-2) (4). However, in the mouse, only three ELR CXC chemokines with neutrophil-activating properties are known: KC, the murine homologue of GRO (5 8); macrophage inflammatory protein-2 (MIP-2), the murine homologue of GRO / (9); and GCP-2 (10). The murine counterpart of IL-8, the most potent human ELR CXC chemokine, has not been identified yet. It is likely that this chemokine does not exist in the mouse and that the other murine ELR CXC chemokines replace IL-8. The most potent murine neutrophil chemotactic protein is GCP-2. Different NH 2 -terminally truncated forms of this chemokine were isolated from conditioned medium of thymic epithelial cells as well as of fibroblasts (10). Human ELR CXC chemokines activate their target cells by binding to two receptors: CXC chemokine receptor 1 (CXCR1) and CXCR2 (11, 12). IL-8 and GCP-2 are highly efficacious ligands for both CXCR1 and CXCR2, whereas the other ELR CXC chemokines are efficient ligands for CXCR2 only (13, 14). In the mouse, only one homologue of CXCRs (IL-8R homologue (IL- 8Rh); 68 and 71% similarity with CXCR1 and CXCR2, respectively) has been identified (15). IL-8Rh binds the murine chemokines KC and MIP-2 with high affinity (16, 17). In our previous study on mouse GCP-2, mixtures of NH 2 -terminally processed forms were evaluated in vitro (10). Here, we show that further analysis of natural GCP-2 additionally revealed COOH-terminal truncation. The biological significance of these post-translational modifications was studied both in vitro (microchamber assay, calcium release assay) and in vivo (intradermal Copyright 1999 by The American Association of Immunologists /99/$02.00

3 6156 IN VITRO AND IN VIVO ACTIVITY OF TRUNCATED FORMS OF mugcp-2 injection in mice) using pure recombinant GCP-2(1 92)/LPSinduced CXC chemokine (LIX) and synthetic GCP-2(9 78). Furthermore, the receptor usage of murine GCP-2 forms was determined using human neutrophils as well as CXCR1- and CXCR2-transfected cells. Materials and Methods Production, purification, and identification of natural murine GCP-2 The MO fibroblast cell line was grown in Eagle s MEM with Earle s salts (EMEM; Life Technologies, Paisley, Scotland) supplemented with 10% FCS (Life Technologies). To produce murine GCP-2, confluent monolayers (175 cm 2 ; Nunc, Roskilde, Denmark) were induced for 72 h in EMEM containing 2% FCS and supplemented with the dsrna poly(riboinosinic acid) poly(ribocytidylic acid) (poly ri:rc; P-L Biochemicals, Milwaukee, WI) at 50 g/ml plus LPS (Escherichia coli B4; Difco, Detroit, MI) at 10 g/ml (10). Murine GCP-2 was purified from the conditioned medium by adsorption to controlled pore glass beads and by heparin-sepharose affinity chromatography as previously described (18). As a third purification step, Mono-S cation exchange fast protein liquid chromatography (FPLC; Pharmacia, Uppsala, Sweden) at ph 4.0 was used. Proteins were eluted with a linear NaCl gradient (0 1 M) in 50 mm formate, ph 4.0 (1 ml/min, 1-ml fractions). Absorbance at 220 nm was measured as a parameter for the protein concentration (18). Alternatively, the heparin-sepharose fractions containing murine GCP-2 were purified by Mono-S cation exchange chromatography at ph 6.4. Proteins were eluted with a linear NaCl gradient in 50 mm malonate, ph 6.4 (1 ml/min, 1-ml fractions). GCP-2 was further purified to homogeneity by reverse phase (RP-) HPLC onac 8 Aquapore RP-300 column (Perkin-Elmer, Norwalk, CT) and eluted with an acetonitrile gradient (0 80% (v/v) acetonitrile in 0.1% (v/v) trifluoroacetic acid; 0.4 ml/min, 0.4-ml fractions). Absorbance at 220 nm was measured. Purified proteins were analyzed for molecular mass and purity by SDS- PAGE under reducing conditions on Tris/tricine gels (19). The relative molecular mass markers used were phosphorylase b (M r, 92,500), BSA (M r, 66,200), OVA (M r, 45,000), carbonic anhydrase (M r, 31,000), soybean trypsin inhibitor (M r, 21,500), and lysozyme (M r, 14,400) (Bio-Rad, Richmond, CA) and the low molecular mass marker (Pierce, Rockford, IL) aprotinin (M r, 6,500). Alternatively, the relative molecular mass markers OVA (M r, 45,000), carbonic anhydrase (M r, 31,000), -lactoglobulin (M r, 18,400), lysozyme (M r, 14,400), bovine trypsin inhibitor (M r, 6,200), and insulin (M r, 3,400) (Life Technologies) were used. The identities of purified proteins were determined by NH 2 -terminal amino acid sequence analysis on a pulsed liquid phase protein sequencer (477A/120A, Perkin-Elmer) with on-line detection of phenylthiohydantoin amino acids. The presence of a cysteine was obvious from the absence of any detectable signal (18). The molecular mass of pure proteins was determined by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). -Cyano-4-hydroxycinnamic acid and cytochrome c were used as matrix and internal standard, respectively. To identify the NH 2 -terminal amino acid sequence of GCP-2 forms that eluted at the same position during chromatographic purification, the proteins were separated by SDS-PAGE on Tris/tricine gels, electroblotted on polyvinylidene fluoride (PVDF) membranes (Problott, Perkin-Elmer), and stained with Coomassie brilliant blue R250. After destaining, membranes were rinsed five times with MilliQ water (Millipore, Bedford, MA). The protein bands were excised from PVDF blots and subjected to NH 2 -terminal amino acid sequence analysis. Murine GCP-2(9 78) was chemically synthesized by automated F-moc solid phase peptide synthesis on a 433A peptide synthesizer (Perkin-Elmer) as previously shown for human GCP-2 (20). After synthesis of the primary structure, the peptide was cleaved from the resin, and side-chain protecting groups were removed by stirring the resin-bound peptide under nitrogen in a cleavage mixture containing 0.75 g of phenol, 250 l of ethanedithiol, 500 l of thioanisole, 500 l of MilliQ water, and 10 ml of trifluoroacetic acid for 90 min. Disulfide bridges were formed by incubation of RP-HPLCpurified unfolded peptide in 150 mm Tris-HCl, ph 8.7, containing 1 mm EDTA, 0.3 mm oxidized glutathione, 3 mm reduced glutathione, and 1 M guanidinium chloride for 90 min. The folded peptide was purified by C 8 RP-HPLC. Recombinant murine GCP-2(1 92)/LIX was purchased from PeproTech (Rocky Hill, NJ). Chemotaxis assay The neutrophil chemotactic activity was tested in the 48-well microchamber (Neuro Probe, Cabin John, MD) chemotaxis assay. Human or murine neutrophils were purified from fresh heparinized peripheral blood from one donor or from pooled mouse blood obtained by cardiac punctures, respectively, as described previously (18). The lower compartments of the microchamber, filled with test samples or controls, were separated from the upper compartments, containing neutrophils/ml, by a 5- m pore size polycarbonate filter (Nuclepore, Pleasanton, CA). After incubation at 37 C for 45 min, migrated cells were fixed, stained, and counted in 10 microscopic fields/well. The chemotactic activity is expressed as a chemotactic index, i.e., the number of cells migrated to the test sample divided by the number of cells migrated to the negative control (18). Statistical analysis was performed using the Mann-Whitney U test. Measurement of intracellular calcium concentration ([Ca 2 ] i ) Changes in [Ca 2 ] i were measured using the fluorescent indicator fura-2 as described by Grynkiewicz et al. (21). Purified neutrophils (10 7 cells/ml) were incubated with 2.5 M fura-2/am (Molecular Probes Europe, Leiden, The Netherlands) and 0.01% pluronic F-127 (Sigma, St. Louis, MO) for 30 min at 37 C in EMEM containing 2% FCS. After incubation, cells were washed twice and resuspended (10 6 cells/ml) in HBSS (1 mm Ca 2 ; Life Technologies) supplemented with 0.1% FCS and buffered at ph 7.4 with 10 mm HEPES/NaOH. Fura-2 fluorescence was measured in an LS50B luminescence spectrophotometer fitted with a temperature-regulated stirred cuvette holder (Perkin-Elmer) after equilibration of the cells at 37 C for 10 min (20). The excitation wavelengths used were 340 and 380 nm; emission was measured at 510 nm. The [Ca 2 ] i was calculated using the Grynkiewicz equation (21). The K d used for calibration was 224 nm. In desensitization experiments, buffer or chemokine was added to the cells as a first stimulus, followed by the addition of an active chemokine concentration after 2 min. The percent inhibition of the increase in [Ca 2 ] i in response to the second stimulus by prestimulation of the cells was calculated. In addition to calcium measurements in neutrophils, changes in [Ca 2 ] i were determined in human embryonic kidney (HEK) cells transfected with either CXCR1 or CXCR2, supplied by Dr. J. M. Wang (National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, MD) (22). HEK cells were cultured in DMEM (Life Technologies) with 10% FCS and 800 g/ml geneticin (Life Technologies) to maintain the transfected characteristics. Cells were treated with trypsin/edta (Life Technologies), washed, and loaded with fura-2/am in DMEM with 10% FCS as described for neutrophils. In vivo inflammatory properties To evaluate the effects of GCP-2(9 78) and GCP-2(1 92)/LIX in vivo, C57BL/6 mice were shaved on the abdomen, and chemokines or lysozyme (negative control peptide) (diluted in 0.9% NaCl) or 0.9% NaCl were injected intradermally (50 l/site). After 2 h, the mice were sacrificed, and injection sites were excised. Skin biopsies were fixed for 24 h in Bouin s fixative. Standard paraffin embedding, sectioning, and staining with hematoxylin-eosin were performed, followed by microscopic examination of the sections at a magnification of 400. The granulocytes were counted in 20 fields for each injection site. Results are expressed as the mean number of granulocytes per field minus the mean number of granulocytes per field at the saline injection site. Statistical analysis was performed using Student s t test. Results Purification and identification of different isoforms of murine GCP-2 The MO fibroblast cell line was stimulated with LPS and poly ri:rc to produce murine GCP-2 (10). The chemokine was purified by adsorption to controlled pore glass beads, heparin-sepharose affinity chromatography, cation exchange chromatography at ph 4.0, and C 8 RP-HPLC based upon neutrophil chemotactic activity. Since in a previous study truncated GCP-2 forms remained as pools after purification (10), special care was taken to separate these and additional forms as much as possible. SDS-PAGE analysis of HPLC-purified GCP-2 showed protein bands of 6 kda eluting at 28% acetonitrile (fraction 49) and of 7 kda eluting at 29% acetonitrile (fraction 53) (Fig. 1A). NH 2 -terminal amino acid sequence analysis and mass determination by MALDI-MS revealed that the 6- and 7-kDa proteins corresponded to different isoforms of murine GCP-2 (Table I). The 7-kDa protein contained NH 2 - terminally intact GCP-2 (78 aa) and GCP-2 missing two NH 2 -

4 The Journal of Immunology 6157 FIGURE 1. Separation of different isoforms of murine GCP-2 by FPLC and RP-HPLC. A, Murine GCP-2, isolated from MO cell-conditioned medium by adsorption to CPG, heparin-sepharose affinity chromatography, and cation exchange chromatography at ph 4.0, was further purified by RP-HPLC. Proteins eluted from the column were evaluated for purity and relative molecular mass by SDS-PAGE on Tris/tricine gels (4 l/lane; reducing conditions; silver staining). B, Murine GCP-2, isolated from MO cell-conditioned medium by adsorption to CPG and heparin-sepharose affinity chromatography, was further purified by cation exchange chromatography at ph 6.4. Proteins eluting from the cation exchange column at M NaCl were further purified by RP-HPLC. The eluted proteins were analyzed by SDS-PAGE (20 l/lane; reducing conditions; silver staining). C, The protein doublet containing GCP-2(LS) and GCP-2(LL), purified from MO cell-conditioned medium by adsorption to controlled pore glass beads, heparin-sepharose affinity chromatography, cation exchange chromatography at ph 4.0, and RP-HPLC (A), was submitted to cation exchange chromatography at ph 6.4. Eluted proteins were analyzed by SDS-PAGE (20 l/lane; reducing conditions; silver staining). Relative molecular mass markers (100 ng each) are as indicated in Materials and Methods. terminal amino acids. These long isoforms were previously called murine GCP-2(L). The smaller 6-kDa protein corresponded to additional truncated forms of murine GCP-2 that were missing four to eight NH 2 -terminal residues and were previously designated murine GCP-2(S) (10). Furthermore, HPLC fractions 47 and 48 (eluting at 27.5% acetonitrile) contained, in addition to the 6-kDa GCP-2(S), an 8.5-kDa protein, whereas fractions 51 and 52 (eluting at 28.5% acetonitrile) contained a 9.5-kDa protein band in addition to the 7-kDa GCP-2(L) (Fig. 1A). Blotting of the latter fractions on PVDF membranes and sequencing of the proteins revealed that the 6- and 8.5-kDa proteins, on the one hand, and the 7- and 9.5-kDa proteins, on the other hand, have the NH 2 -terminal amino acid sequence, corresponding to the NH 2 -terminus of murine GCP-2(S) and GCP-2(L), respectively. The difference in relative molecular mass of the GCP-2(S) and GCP-2(L) doublets can be explained by additional amino acids at the COOH-terminus or, alternatively, by glycosylation. Murine GCP-2 does not contain N-glycosylation sites, and additionally, the cdna-deduced amino acid sequence of a protein called LIX is identical with that of natural GCP-2(L), except for a COOH-terminal extension of 14 aa (23). Thus, COOH-terminal extension is the most probable explanation for the higher molecular mass proteins. In an effort to separate the 8.5-kDa from the 6-kDa GCP-2(S) and the 9.5-kDa from the 7-kDa GCP-2(L), fibroblast-conditioned medium was purified using cation exchange chromatography at ph 6.4 as an alternative for Mono-S FPLC at ph 4.0. In these conditions, GCP-2 forms were better separated and were further purified by C 8 RP-HPLC. Purification of the GCP-2 forms eluting during FPLC at M NaCl by HPLC yielded pure GCP-2 of 8.5 kda as well as GCP-2 of 9.5 kda, as determined by NH 2 -terminal sequencing (Fig. 1B and Table I). However, by gel filtration a 15-kDa protein could not be separated from the 9.5-kDa GCP- 2(L). Possibly, the 15-kDa molecule corresponds to a dimer of this protein. Indeed, blotting and sequencing of this 15-kDa protein in the 9.5-kDa GCP-2(L) revealed the NH 2 -terminus of GCP-2(L) (data not shown). To delineate the exact identity of the 8.5-kDa GCP-2(S) and 9.5-kDa GCP-2(L), mass analysis was performed by MALDI-MS. This revealed the presence of protein isoforms of murine GCP-2 with an extended COOH-terminus for both 8.5- and 9.5-kDa GCP-2. The 8.5- and 9.5-kDa GCP-2 contained several COOH-terminally extended forms, including the one corresponding to LIX. Therefore, the 8.5- and 9.5-kDa proteins were designated GCP-2(SL) and GCP-2(LL), respectively. In retrospective, the 6- and 7-kDa GCP-2 forms were designated GCP-2(SS) and GCP-2(LS), respectively. To obtain pure murine GCP-2(LL) (without the presence of the dimer), RP-HPLC fractions 51 and 52, containing murine GCP-2(LS) and GCP-2(LL) (Fig. 1A), were further submitted to cation exchange chromatography at ph 6.4, yielding pure 7- and 9.5-kDa protein (Fig. 1C). Amino acid sequence analysis confirmed that the separated proteins contained the same NH 2 -terminus corresponding to the longer form of GCP-2. Comparison of the neutrophil chemotactic activity of NH 2 - and/ or COOH-terminally truncated forms of natural murine GCP-2 After purification to homogeneity, the different natural isoforms of murine GCP-2 were compared for their neutrophil chemotactic activity in the microchamber chemotaxis assay using human neutrophils (Fig. 2A). The minimal effective concentrations for murine GCP-2(SS), GCP-2(SL), GCP-2(LS), and GCP-2(LL) to induce significant neutrophil migration ( p 0.01 compared with negative control) were 15, 30, 60, and 100 ng/ml, respectively. These results confirm the previous findings that GCP-2(SS) is 4-fold more active than GCP-2(LS) (10) and in addition show that GCP-2(SL) is 3 times more potent than GCP-2(LL). Thus, NH 2 -terminal truncation induces an increase in chemotactic potency of murine GCP-2. Furthermore, GCP-2(SS) and GCP-2(LS) are more potent than GCP-2(SL) and GCP-2(LL), respectively, indicating that COOHterminal cleavage also increases the chemotactic potency of the chemokine. Although the latter effect is less pronounced than that of NH 2 -terminal truncation, the combined cleavage of murine GCP-2 at both termini of the protein generates a chemokine with a 10-fold higher potency than that of the intact protein (Fig. 2A). The most active and most prominent (10) natural isoform of murine GCP-2, i.e., GCP-2(9 78), was synthesized by solid phase peptide synthesis, whereas the least active form, GCP-2(1 92), corresponded to recombinant LIX. The neutrophil chemotactic activities of pure synthetic GCP-2(9 78) and recombinant GCP-2(1 92)/LIX were compared in the microchamber assay. As shown for the natural GCP-2 isoforms, the truncated protein was much more active than the intact protein (Fig. 2B). GCP-2(9 78) was active

5 6158 IN VITRO AND IN VIVO ACTIVITY OF TRUNCATED FORMS OF mugcp-2 Table I. Identification of different natural isoforms of murine GCP-2 NH 2 -Terminal Amino Acid Sequence Protein (M r ) Sequence Relative Amount (%) Determined Molecular Mass Corresponding COOH Terminus Theoretical Molecular Mass 9.5 kda (LL) APSSVIAATELRCVC a KKKAKRNALAVERTASVQ KKKAKRNALAVERTA KKKAKRNALAVERT KKKAKRNALA KKKAKRNAL kda (SL) AATELRCVC KKKAKRNALAVERTASV KKKAKRNALAVERTA KKKAKRNALAVERT KKKAKRNAL ATELRCVC KKKAKRNALAVERTASVQ KKKAKRNALAVERTA KKKAKRNALAVERT KKKAKRNALA KKKAKRNAL TELRCVC KKKAKRNALAVERTASVQ KKKAKRNALAVERTA KKKAKRNALAVERT KKKAKRNALAVER KKKAKRNALA KKKAKRNAL kda (LS) APSSVIAATELRCVC KKKA SSVIAATELRCVC KKKAK kda (SS) VIAATELRCVC 9 ND b KKKA IAATELRCVC 17 ND KKKA AATELRCVC KKKA ATELRCVC KKKA TELRCVC KKKA a The accuracy of the experimentally determined masses is 0.15%. b ND, not detected. from 3 ng/ml onward ( p 0.01), whereas 300 ng/ml of GCP- 2(1 92)/LIX was necessary to obtain significant migration. Thus, a 100-fold difference in chemotactic potency between GCP-2(1 92)/LIX and GCP-2(9 78) was observed, whereas natural GCP- 2(SS) and GCP-2(LL) showed a 10-fold difference in potency. This can be explained by the presence of other isoforms in GCP- 2(LL) and in GCP-2(SS) in addition to GCP-2(1 92)/LIX and GCP-2(9 78), respectively (see Table I). The chemotactic activities of murine GCP-2(9 78) and GCP- 2(1 92)/LIX were also compared in the microchamber assay using murine neutrophils (Fig. 2C). Again, it was found that GCP-2(9 78) is 100 times more potent than GCP-2(1 92)/LIX (minimal effective concentrations of 10 and 1000 ng/ml, respectively; p 0.01 compared with negative control). Intracellular signaling by murine GCP-2 forms in neutrophils and desensitization by murine MIP-2 and human IL-8 The ability of murine GCP-2(9 78) and GCP-2(1 92)/LIX to induce an increase in [Ca 2 ] i in human neutrophils was evaluated in parallel with that of natural murine MIP-2, purified to homogeneity from conditioned medium of WEHI-3 myelomonocytic cells (10), and with that of pure natural human IL-8 (24) (Fig. 3). As observed in the chemotaxis assay, GCP-2(1 92)/LIX was less potent at increasing the [Ca 2 ] i in neutrophils than GCP-2(9 78); the minimal effective concentrations were 500 and 15 ng/ml, respectively. In contrast to its weaker chemotactic potency (10), murine MIP-2 was 7-fold more potent than GCP-2(9 78) at stimulating an increase in [Ca 2 ] i (minimal effective concentration of 2 ng/ml), but remained less potent than human IL-8, which was still active at 0.5 ng/ml (Fig. 3). The higher sp. act. of human IL-8 compared with those of the murine chemokines may be due to species specificity. To determine whether murine GCP-2 recognizes the same receptors as human IL-8 and murine MIP-2 on human neutrophils, calcium desensitization experiments were performed. Cross-desensitization was observed among murine GCP-2(9 78), MIP-2, and human IL-8 (Table II). IL-8 at 5 ng/ml completely inhibited the calcium increase in response to 50 ng/ml murine GCP-2(9 78). However, GCP-2(9 78) at 150 ng/ml and MIP-2 at 20 ng/ml could only partially desensitize for the IL-8 response (1.5 ng/ml). In contrast, murine GCP-2(9 78) at 50 ng/ml completely abolished the calcium increase induced by 7 ng/ml MIP-2, and, inversely, 20 ng/ml MIP-2 inhibited the GCP-2-induced calcium response (50 ng/ml). These data indicate that murine GCP-2 and MIP-2 use an identical receptor(s) to activate human neutrophils and that GCP-2 shares at least one receptor with human IL-8. In contrast to murine GCP-2(9 78), GCP-2(1 92)/LIX at concentrations up to 1000 ng/ml did not decrease the calcium response to IL-8, suggesting a lower affinity of the latter protein for the IL-8Rs. CXCR1 and CXCR2 receptor usage by murine GCP-2 and MIP-2 The receptor usage by murine GCP-2(9 78) was further evaluated by performing calcium measurements in human CXCR1- or CXCR2-transfected HEK cells. As previously shown, human IL-8 and human GCP-2 efficiently induced an increase in [Ca 2 ] i in both CXCR1- and CXCR2-transfected cells (Fig. 4). IL-8 was a more potent stimulus for CXCR1 than for CXCR2, whereas human GCP-2 showed similar minimal effective concentrations for calcium induction through both receptors. Both murine GCP-2 and MIP-2 induced an increase in [Ca 2 ] i in CXCR1- as well as CXCR2-transfected cells, whereas the human homologue of

6 The Journal of Immunology 6159 FIGURE 2. Comparison of the neutrophil chemotactic activities of different isoforms of murine GCP-2. A, Different natural isoforms of murine GCP-2 (SS, LS, SL, LL) were compared for their neutrophil chemotactic activities in the microchamber chemotaxis assay using human neutrophils. Chemotactic indexes (mean SEM) are derived from six independent experiments. In each experiment, samples were tested in triplicate. B, Synthetic GCP-2(9 78) and recombinant GCP-2(1 92)/LIX were compared for neutrophil chemotactic activity in the microchamber chemotaxis assay using human neutrophils. Chemotactic indexes (mean SEM) are derived from seven independent experiments. In each experiment, samples were tested in triplicate. C, Synthetic GCP-2(9 78) and recombinant GCP-2(1 92)/LIX were compared for neutrophil chemotactic activity in the microchamber chemotaxis assay using murine neutrophils. Chemotactic indexes (mean SEM) are derived from six independent experiments. In each experiment, samples were tested in triplicate. Asterisks indicate a significant effect of GCP-2 compared with the negative control., p 0.01;, p MIP-2, GRO /, failed to efficiently affect CXCR1 (13). In contrast to IL-8, both murine GCP-2 and MIP-2 were more potent at inducing an increase in [Ca 2 ] i through CXCR2 than through CXCR1. Murine GCP-2 at 50 ng/ml increased the [Ca 2 ] i in CXCR2-transfected cells, whereas 150 ng/ml was necessary to observe an effect in CXCR1-transfected cells. Murine MIP-2 was 10 times more potent at stimulating a calcium increase in CXCR2- transfected cells than in CXCR1-transfected cells; the minimal effective concentrations were 7 and 70 ng/ml, respectively. Because human IL-8 was more efficient at signaling through CXCR1 in this test, the results suggest a predominant CXCR2 usage by murine GCP-2 and MIP-2. This confirms that functionally the murine IL- 8Rh is the equivalent of CXCR2. Comparison of the biological activities of murine GCP-2 (9 78) and GCP-2(1 92)/LIX in vivo To confirm the differences in potency between GCP-2 isoforms observed in vitro, mice were intradermally injected with different FIGURE 3. Induction of an increase in [Ca 2 ] i in human neutrophils by murine GCP-2 and MIP-2. Human neutrophils loaded with fura-2/am were stimulated with different concentrations (high to low concentration from top to bottom) of murine GCP-2(9 78) (150, 50, 15, and 5 ng/ml), GCP-2(1 92)/LIX (1000, 500, and 150 ng/ml), MIP-2 (20, 7, and 2 ng/ml), or human IL-8 (5, 1.5, 0.5, and 0.15 ng/ml) at about 60 s. The [Ca 2 ] i was calculated according to the Grynkiewicz equation (21). One representative experiment of three is shown. concentrations of murine GCP-2(1 92)/LIX or GCP-2(9 78) and with lysozyme and 0.9% NaCl as negative controls. Murine GCP- 2(9 78) induced significant ( p 0.01) granulocyte accumulation 2 h after injection at a dose of 10 ng, whereas GCP-2(1 92)/LIX was only active at a dose of 150 ng (Fig. 5). No cell types other than granulocytes were chemoattracted to the injection site (Fig. 6). Lysozyme (1000 ng) did not induce infiltration of granulocytes (data not shown). These data indicate that GCP-2(9 78) is also more potent than GCP-2(1 92)/LIX in vivo. However, the difference in potency may be less pronounced in vivo than in vitro. Discussion In the human system, IL-8 is the most potent neutrophil-activating chemokine, whereas the other ELR CXC chemokines GRO, -, and - ; NAP-2; epithelial cell-derived neutrophil attractant-78; and GCP-2 are weaker chemoattractants (4). In the mouse the equivalent of human IL-8 has not been identified. However, murine GCP-2 is equally potent to attract murine neutrophils as human IL-8 (10). This indicates that in the mouse other chemokines replace IL-8 during inflammation. Murine GCP-2, isolated from conditioned medium of epithelial cells or fibroblasts, occurs in several NH 2 -terminally truncated forms (10). In this study we show that fibroblasts stimulated with LPS plus poly ri:rc produce not only different NH 2 -terminally, but also COOH-terminally truncated forms of murine GCP-2. These NH 2 - and/or COOH-terminal truncations give existence to murine GCP-2 isoforms containing 69 (GCP-2(10 78)) (10) to 92 residues (GCP-2(1 92)/LIX), including intermediate forms. Naturally truncated forms shortened (S) at the NH 2 -terminus or COOH-terminus compared with longer (L) forms were designated SS, SL, LS, and LL, respectively. The 92-residue protein corresponds to the sequence of LIX, a cdna that was cloned from murine fibroblasts by Smith and Herschman (23). GCP-2(1 92)/LIX is expressed in several tissues during acute endotoxemia (25). However, natural GCP-2(1 92)/LIX has not previously been isolated or evaluated for its biological activity. GCP-2(1 92)/LIX has an extended COOH-terminus compared with other human and murine ELR CXC chemokines (Fig. 7). This is also the case for the murine

7 6160 IN VITRO AND IN VIVO ACTIVITY OF TRUNCATED FORMS OF mugcp-2 Table II. Desensitization of the GCP-2-induced increase in [Ca 2 ] i in neutrophilic granulocytes by IL-8 and MIP-2 and vice versa First Stimulus Concentration (ng/ml) Second Stimulus Concentration (ng/ml) % Inhibition of Second Stimulus exp. 1 exp. 2 mugcp-2(9 78) 100 mugcp-2(9 78) (30 ng/ml) huil-8 5 mugcp-2(9 78) (50 ng/ml) mugcp-2(9 78) 150 mumip ND 50 (7 ng/ml) mugcp-2(9 78) 150 huil (1.5 ng/ml) ND mugcp-2(1 92)/LIX 1000 huil (1.5 ng/ml) mumip-2 20 mugcp-2(9 78) (50 ng/ml) mumip-2 20 huil (1.5 ng/ml) and human ELR CXC chemokine MIG (26, 27). The COOH-terminal truncation of murine GCP-2 gives rise to a protein that is more similar in length to the other ELR CXC chemokines (Fig. 7). Previously, only the murine isoforms GCP-2(SS) and GCP- 2(LS) were identified and compared for their bioactivity. It was found that the NH 2 -terminally truncated GCP-2(SS) forms are more potent than the longer GCP-2(LS) forms. In this study we also compared the neutrophil chemotactic and activating potencies of COOH-terminally processed forms of murine GCP-2. Both NH 2 -terminal and COOH-terminal truncation result in an increase in neutrophil chemotactic potency, but COOH-terminal truncation has a somewhat less pronounced effect (2-fold) than cleavage at the NH 2 -terminus. The higher potency of the truncated isoforms of GCP-2 was confirmed in intracellular signaling experiments, yielding a 30-fold difference in sp. act. NH 2 -terminal cleavage has previously been reported for other CXC chemokines, including human IL-8 (24, 28), NAP-2 (24, 29), and GCP-2 (30, 31) and murine GCP-2 and KC (10). No natural chemokine truncation beyond the ELR motif, necessary for biological activity (32), has been observed. NH 2 -terminal truncation of ELR CXC chemokines often induces an increase in biological potency. The various natural NH 2 -terminally processed forms of IL-8 have never been completely separated chromatographically (33). However, using defined synthetic and recombinant material, the NH 2 -terminally truncated IL-8 forms were shown to be more active than the longer forms (32, 34). Platelet basic protein and connective tissue-activating protein-iii, precursors of NAP-2, need to be processed at the NH 2 -terminus to become neutrophil attractants (29, 33). In contrast, the different NH 2 -terminal isoforms of human GCP-2 did not show differences in potency in the chemotaxis assay (31). The FIGURE 4. Induction of an increase in [Ca 2 ] i in CXCR1 and CXCR2 transfectants by murine GCP-2 and MIP-2. HEK cells, transfected with CXCR1 or CXCR2, were stimulated with different concentrations of human IL-8 and GCP-2 and murine GCP-2(9 78) and MIP-2. The [Ca 2 ] i was calculated according to the Grynkiewicz equation (21). The detection limit for the increase in [Ca 2 ] i (15 nm) is indicated by a dotted line. Data are derived from two experiments.

8 The Journal of Immunology 6161 FIGURE 5. In vivo inflammatory properties of GCP-2(9 78) and GCP- 2(1 92)/LIX. Different amounts of GCP-2(9 78) and GCP-2(1 92)/LIX in 0.9% NaCl were injected intradermally in mice. Saline (0.9% NaCl) was injected as a negative control. After 2 h, mice were sacrificed. Granulocytes were counted in 20 microscopic fields/injection site. Data are derived from eight mice. Results are expressed as the mean number of granulocytes per field minus the mean number of granulocytes per field at the saline injection site SEM. Asterisks indicate a significant effect of GCP-2 compared with 0.9% NaCl:, p 0.01;, p increased biological activity of ELR CXC chemokines by NH 2 - terminal processing is in contrast with data on other chemokines, such as the ELR CXC chemokine stromal cell-derived factor-1 and the CC chemokines RANTES, macrophage-derived chemokine, and eotaxin, all showing impaired chemotactic activity and receptor binding after removal of their NH 2 -terminal dipeptide by CD26 (35, 36). COOH-terminal truncation has only been described for a few CXC chemokines. Natural COOH-terminally truncated forms of NAP-2 missing four and seven residues were isolated from the conditioned medium of platelet-containing mononuclear cells (37 39). These isoforms were 3 and 5 times more potent than intact NAP-2 to degranulate neutrophils and to compete for NAP-2 receptor binding. Recombinant forms of NAP-2 missing five to seven residues showed a 5-fold increase in potency (38, 39). Using synthetic isoforms of IL-8, it has been shown that removal of the three COOH-terminal amino acids induces an increase in biological activity, whereas further removal of amino acids gradually decreases the potency (32). A 3.6-kDa form of IL-8, purified from conditioned medium of fibroblasts, with a 50-fold lower potency than intact IL-8 may represent a COOHterminally truncated form (40). Finally, for the ELR CXC chemokine MIG, COOH-terminally truncated forms ( aa) were isolated from monocytes and THP-1 cells. This truncation resulted in a decrease in lymphocyte-activating potency (27). In the human system, two ELR CXC chemokine receptors (CXCR1 and CXCR2) have been identified. Human IL-8 and GCP-2 can efficiently activate cells through binding to both receptors, whereas the other ELR CXC chemokines are better ligands for CXCR2 (13, 14). In the mouse, only the functional homologue of CXCR2, IL-8Rh, but not that of CXCR1, is known (15). The IL-8Rh binds the murine chemokines KC and MIP-2 with high affinity (16, 17) and is important for neutrophil migration to inflammatory sites (41). Furthermore, mice lacking the IL-8Rh show lymphadenopathy and splenomegaly (41). KC efficiently binds to human CXCR2, but not to CXCR1 (8, 42). Substitution of the amino acid sequence of KC between cysteines 2 and 3 with the corresponding domain of IL-8 confers binding to CXCR1 (42). FIGURE 6. In vivo recruitment of granulocytes by GCP-2(9 78) and GCP-2(1 92)/ LIX. Mice were injected intradermally with 900 ng of GCP-2(9 78) (A), 100 ng of GCP- 2(9 78) (B), 500 ng of GCP-2(1 92)/LIX (C), or 0.9% NaCl (D). After 2 h, mice were sacrificed, and injection sites were excised. Skin biopsies were stained with hematoxylin-eosin. Magnification, 350.

9 6162 IN VITRO AND IN VIVO ACTIVITY OF TRUNCATED FORMS OF mugcp-2 FIGURE 7. underlined. Sequence alignment of murine ELR CXC chemokines and their human counterparts. The conserved ELR motif and cysteine residues are Furthermore, it has been shown that the presence of a basic residue, Arg 20 in human GCP-2 and Lys 20 in IL-8 (Fig. 7), is essential for signaling through CXCR1 (43). In the murine chemokines GCP-2 and MIP-2, a basic amino acid is present at this position, i.e., Lys 21 in murine GCP-2 and Arg 17 in murine MIP-2 (Fig. 7). This basic residue is not present in KC (Gly 17 ). According to these predictions, we indeed found that murine GCP-2 and MIP-2 can bind both CXCR1 and CXCR2 to activate cells. However, both chemokines were more potent to signal through CXCR2 than through CXCR1, but the efficiencies for both receptors were comparable. In contrast, others have shown that recombinant MIP-2 has only low affinity for CXCR1, whereas it bound CXCR2 with high affinity (44). This can be explained by the observation that high agonist potency and high affinity binding are distinct functions (45). Human GCP-2 induces neutrophil accumulation and plasma extravasation in rabbit skin (20, 30). Murine KC and MIP-2 were also shown to have chemotactic properties in vivo (8, 9). After intradermal injection of murine GCP-2(9 78) and GCP-2(1 92)/ LIX in mice, both isoforms induced significant neutrophil accumulation after 2 h; GCP-2(9 78) was more potent than GCP-2(1 92)/LIX. It has been shown that IL-8(6 77) is a more potent chemoattractant and activator of neutrophils than IL-8(1 77) in vitro, but in vivo both isoforms are equipotent, possibly due to rapid proteolytic processing of the 77-aa form (34). In addition to different proteases being involved in the processing of IL-8 and GCP-2, the fact that the effects of IL-8 were evaluated after 4 h, whereas the chemotactic activity of murine GCP-2 was evaluated after 2 h, may also explain the difference between these two experiments. The period of 2 h may be too short to completely convert the intact GCP-2 into truncated forms. However, the difference in potency between the two GCP-2 isoforms may be lower in vivo than in vitro. Because multiple intermediately processed forms (26 in total) have been purified from natural cellular sources, this might be due to partial cleavage of intact GCP-2 into truncated GCP-2. In conclusion, murine GCP-2 occurs as 28 different NH 2 - and COOH-terminally truncated forms. Truncation yields more potent GCP-2, both in vitro and in vivo. This indicates that during inflammation, the presence of both chemokines and proteases that can cleave chemokines will determine the efficiency of leukocyte accumulation and activation. Acknowledgments We thank the Laboratory of Clinical Immunology of the University of Leuven for providing blood samples. The technical assistance of René Conings and Marleen Groenen is greatly acknowledged. References 1. Rollins, B. J Chemokines. Blood 90: Baggiolini, M., B. Dewald, and B. Moser Human chemokines: an update. Annu. Rev. Immunol. 15: Luster, A. D Chemokines: chemotactic cytokines that mediate inflammation. N. Engl. J. Med. 338: Wuyts, A., P. Proost, and J. Van Damme Interleukin-8 and other CXC chemokines. In The Cytokine Handbook, 3rd Ed. A.W. Thomson, ed. Academic Press, London, p Cochran, B. H., A. C. Reffel, and C. D. Stiles Molecular cloning of gene sequences regulated by platelet-derived growth factor. Cell 33: Oquendo, P., J. Alberta, D. Wen, J. L. Graycar, R. Derynck, and C. D. Stiles The platelet-derived growth factor-inducible KC gene encodes a secretory protein related to platelet -granule proteins. J. Biol. Chem. 264: Heinrich, J. N., E. C. O Rourke, L. Chen, H. Gray, K. S. Dorfman, and R. Bravo Biological activity of the growth factor-induced cytokine N51: structurefunction analysis using N51/interleukin-8 chimeric molecules. Mol. Cell. Biol. 14: Bozic, C. R., L. F. Kolakowski, Jr., N. P. Gerard, C. Garcia-Rodriguez, C. von Uexkull-Guldenband, M. J. Conklyn, R. Breslow, H. J. Showell, and C. Gerard Expression and biologic characterization of the murine chemokine KC. J. Immunol. 154: Wolpe, S. D., B. Sherry, D. Juers, G. Davatelis, R. W. Yurt, and A. Cerami Identification and characterization of macrophage inflammatory protein 2. Proc. Natl. Acad. Sci. USA 86: Wuyts, A., A. Haelens, P. Proost, J.-P. Lenaerts, R. Conings, G. Opdenakker, and J. Van Damme Identification of mouse granulocyte chemotactic protein-2 from fibroblasts and epithelial cells. Functional comparison with natural KC and macrophage inflammatory protein-2. J. Immunol. 157: Holmes, W. E., J. Lee, W.-J. Kuang, G. C. Rice, and W. I. Wood Structure and functional expression of a human interleukin-8 receptor. Science 253: Murphy, P. M., and H. L. Tiffany Cloning of complementary DNA encoding a functional human interleukin-8 receptor. Science 253: Ahuja, S. K., and P. M. Murphy The CXC chemokines growth-regulated oncogene (GRO), GRO, GRO, neutrophil-activating peptide-2, and epithelial cell-derived neutrophil-activating peptide-78 are potent agonists for the type B, but not the type A, human interleukin-8 receptor. J. Biol. Chem. 271: Wuyts, A., P. Proost, J.-P. Lenaerts, A. Ben-Baruch, J. Van Damme, and J. M. Wang Differential usage of the CXC chemokine receptors 1 and 2 by interleukin-8, granulocyte chemotactic protein-2 and epithelial-cell-derived neutrophil attractant-78. Eur. J. Biochem. 255: Cerretti, D. P., N. Nelson, C. J. Kozlosky, P. J. Morrissey, N. G. Copeland, D. J. Gilbert, N. A. Jenkins, J. K. Dosik, and B. A. Mock The murine homologue of the human interleukin-8 receptor type B maps near the Ity-Lsh-Bcg disease resistance locus. Genomics 18: Bozic, C. R., N. P. Gerard, C. von Uexkull-Guldenband, L. F. Kolakowski, Jr., M. J. Conklyn, R. Breslow, H. J. Showell, and C. Gerard The murine interleukin 8 type B receptor homologue and its ligands: expression and biological characterization. J. Biol. Chem. 269: Lee, J., G. Cacalano, T. Camerato, K. Toy, M. W. Moore, and W. I. Wood Chemokine binding and activities mediated by the mouse IL-8 receptor. J.Immunol. 155: Wuyts, A., P. Proost, G. Froyen, A. Haelens, A. Billiau, G. Opdenakker, and J. Van Damme Purification and identification of human and mouse granulocyte chemotactic protein-2 isoforms. Methods Enzymol. 287: Schägger, H., and G. von Jagow Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kda. Anal. Biochem. 166: Wuyts, A., N. Van Osselaer, A. Haelens, I. Samson, P. Herdewijn, A. Ben-Baruch, J. J. Oppenheim, P. Proost, and J. Van Damme Characterization of synthetic human granulocyte chemotactic protein 2: usage of chemokine receptors CXCR1 and CXCR2 and in vivo inflammatory properties. Biochemistry 36: Grynkiewicz, G., M. Poenie, and R. Y. Tsien A new generation of Ca 2 indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: Ben-Baruch, A., K. Bengali, K. Tani, L. Xu, J. J. Oppenheim, and J. M. Wang IL-8 and NAP-2 differ in their capacities to bind and chemoattract 293 cells transfected with either IL-8 receptor type A or type B. Cytokine 9: Smith, J. B., and H. R. Herschman Glucocorticoid-attenuated response genes encode intercellular mediators, including a new C-X-C chemokine. J. Biol. Chem. 270: Van Damme, J., J. Van Beeumen, R. Conings, B. Decock, and A. Billiau Purification of granulocyte chemotactic peptide/interleukin-8 reveals N-terminal sequence heterogeneity similar to that of -thromboglobulin. Eur. J. Biochem. 181:337.

10 The Journal of Immunology Rovai, L. E., H. R. Herschman, and J. B. Smith The murine neutrophilchemoattractant chemokines LIX, KC, and MIP-2 have distinct induction kinetics, tissue distributions, and tissue-specific sensitivities to glucocorticoid regulation in endotoxemia. J. Leukocyte Biol. 64: Farber, J. M A macrophage mrna selectively induced by -interferon encodes a member of the platelet factor 4 family of cytokines. Proc. Natl. Acad. Sci. USA 87: Liao, F., R. L. Rabin, J. R. Yannelli, L. G. Koniaris, P. Vanguri, and J. M. Farber Human Mig chemokine: biochemical and functional characterization. J. Exp. Med. 182: Van Damme, J., B. Decock, R. Conings, J.-P. Lenaerts, G. Opdenakker, and A. Billiau The chemotactic activity for granulocytes produced by virally infected fibroblasts is identical to monocyte-derived interleukin 8. Eur. J. Immunol. 19: Walz, A., B. Dewald, V. von Tscharner, and M. Baggiolini Effects of the neutrophil-activating peptide NAP-2, platelet basic protein, connective tissueactivating peptide III, and platelet factor 4 on human neutrophils. J. Exp. Med. 170: Proost, P., C. De Wolf-Peeters, R. Conings, G. Opdenakker, A. Billiau, and J. Van Damme Identification of a novel granulocyte chemotactic protein (GCP-2) from human tumor cells: in vitro and in vivo comparison with natural forms of GRO, IP-10, and IL-8. J. Immunol. 150: Proost, P., A. Wuyts, R. Conings, J.-P. Lenaerts, A. Billiau, G. Opdenakker, and J. Van Damme Human and bovine granulocyte chemotactic protein-2: complete amino acid sequence and functional characterization as chemokines. Biochemistry 32: Clark-Lewis, I., C. Schumacher, M. Baggiolini, and B. Moser Structureactivity relationships of interleukin-8 determined using chemically synthesized analogs: critical role of NH 2 -terminal residues and evidence for uncoupling of neutrophil chemotaxis, exocytosis, and receptor binding activities. J. Biol. Chem. 266: Van Damme, J., M. Rampart, R. Conings, B. Decock, N. Van Osselaer, J. Willems, and A. Billiau The neutrophil-activating proteins interleukin 8 and -thromboglobulin: in vitro and in vivo comparison of NH 2 -terminally processed forms. Eur. J. Immunol. 20: Nourshargh, S., J. A. Perkins, H. J. Showell, K. Matsushima, T. J. Williams, and P. D. Collins A comparative study of the neutrophil stimulatory activity in vitro and pro-inflammatory properties in vivo of 72 amino acid and 77 amino acid IL-8. J. Immunol. 148: Proost, P., I. De Meester, D. Schols, S. Struyf, A.-M. Lambeir, A. Wuyts, G. Opdenakker, E. De Clercq, S. Scharpé, and J. Van Damme Aminoterminal truncation of chemokines by CD26/dipeptidyl-peptidase IV: conversion of RANTES into a potent inhibitor of monocyte chemotaxis and HIV-1-infection. J. Biol. Chem. 273: Van Damme, J., S. Struyf, A. Wuyts, E. Van Coillie, P. Menten, D. Schols, S. Sozzani, I. De Meester, and P. Proost The role of CD26/DPP IV in chemokine processing. Chem. Immunol. 72: Brandt, E., F. Petersen, and H.-D. Flad A novel molecular variant of the neutrophil-activating peptide NAP-2 with enhanced biological activity is truncated at the C-terminus: identification by antibodies with defined epitope specificity. Mol. Immunol. 30: Ehlert, J. E., F. Petersen, M. H. G. Kubbutat, J. Gerdes, H.-D. Flad, and E. Brandt Limited and defined truncation at the C terminus enhances receptor binding and degranulation activity of the neutrophil-activating peptide 2 (NAP-2): comparison of native and recombinant NAP-2 variants. J. Biol. Chem. 270: Ehlert, J. E., J. Gerdes, H.-D. Flad, and E. Brandt Novel C-terminally truncated isoforms of the CXC chemokine -thromboglobulin and their impact on neutrophil functions. J. Immunol. 161: Schröder, J.-M., M. Sticherling, H. H. Henneicke, W. C. Preissner, and E. Christophers IL-1 or tumor necrosis factor- stimulate release of three NAP-1/IL-8-related neutrophil chemotactic proteins in human dermal fibroblasts. J. Immunol. 144: Cacalano, G., J. Lee, K. Kikly, A. M. Ryan, S. Pitts-Meek, B. Hultgren, W. I. Wood, and M. W. Moore Neutrophil and B cell expansion in mice that lack the murine IL-8 receptor homolog. Science 265: Heinrich, J. N., and R. Bravo N51 competes 125 I-interleukin (IL)-8 binding to IL-8R but not IL-8R : structure-function analysis using N51/IL-8 chimeric molecules. J. Biol. Chem. 270: Wolf, M., M. B. Delgado, S. A. Jones, B. Dewald, I. Clark-Lewis, and M. Baggiolini Granulocyte chemotactic protein 2 acts via both IL-8 receptors, CXCR1 and CXCR2. Eur. J. Immunol. 28: Jerva, L. F., G. Sullivan, and E. Lolis Functional and receptor binding characterization of recombinant murine macrophage inflammatory protein 2: sequence analysis and mutagenesis identify receptor binding epitopes. Protein Sci. 6: Ahuja, S. K., J. C. Lee, and P. M. Murphy CXC chemokines bind to unique sets of selectivity determinants that can function independently and are broadly distributed on multiple domains of human interleukin-8 receptor B: determinants of high affinity binding and receptor activation are distinct. J. Biol. Chem. 271:225.