A molecular switch of chemokine receptor selectivity. Chemical modification of the interleukin-8 Leu25 --> Cys mutant.

Interleukin-8 (IL-8), a member of the CXC chemokine family, is a key activator of neutrophils. We have previously shown that two novel CC chemokine-like properties, namely monocyte chemoattraction and binding to CC CKR-1, are introduced into IL-8 by mutating Leu to the conserved tyrosine present in CC chemokines. To further investigate the role of this position in receptor selectivity, we have mutated Leu to cysteine. The protein folds correctly with two disulfide bonds and a free thiol group at Cys. This mutant behaves overall like wild-type IL-8, with little change in neutrophil chemotaxis and IL-8 receptor binding, and has no effect on CC CKR-1. These data are consistent with cysteine being approximately isosteric with the natural amino acid leucine. However, modification of the cysteine by addition of a fluorescent N-methyl-N-(2-N-methyl, N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)aminoethyl)acetamido (NBD) group lowers potency in neutrophil chemotaxis and affinity in IL-8 receptor binding assays by 2 orders of magnitude. This Leu Cys-NBD mutant introduces monocyte chemoattractant activity and the ability to displace I-labeled macrophage inflammatory protein-1α from the recombinant CC CKR-1 receptor. Additionally, we show a specific interaction between the fluorescent mutant and the N-terminal 34-amino acid peptide from CC CKR-1. This confirms the importance of this region in IL-8 in receptor binding and in conferring specificity between CXC and CC chemokines. Circular dichroism spectra of the IL-8 mutants having CC chemokine-like activity show a consistent drop in α-helical content compared with the spectra for wild-type IL-8. This suggests that distortion of the C-terminal helix may play a role in chemokine receptor-ligand selectivity.

IL-8R-B (2), have been identified that bind with nanomolar affinity to IL-8. The regions necessary for receptor activation and subsequent signaling through G-proteins have been localized at the flexible amino terminus of IL-8 (Glu 4 -Leu 5 -Arg 6 ) by mutagenesis and peptide synthesis studies (3,4).
Members of the CC chemokine family, such as macrophage inflammatory protein-1␣ (MIP-1␣) and RANTES, activate a variety of cell types including monocytes during chronic inflammation. The CC chemokines mediate this effect through a different family of receptors including CC CKR-1, which binds MIP-1␣ and RANTES (5,6); CC CKR-2, which binds MCP-1 and MCP-3 (7-9); CC CKR-3, which binds RANTES, MIP-1␣, and MIP-1␤ (10); and CC CKR-4, which responds to RANTES, MCP-1, and MIP-1␣ (11). As in CXC chemokines, it is the amino-terminal region of the ligand that is responsible for receptor activation. Truncated N-terminal mutants of MCP-1 have been shown to be antagonists of MCP-1-induced monocyte chemotaxis (12). Mutants of RANTES that have an additional methionine residue at the amino terminus have been shown to be antagonists of THP-1 and T-cell activation (13).
To date, no natural human CXC ligand has been found to bind a CC chemokine receptor. To investigate the molecular basis of this selectivity, we have compared the primary sequences of CXC and CC chemokines. In the region of IL-8 corresponding to the inner ␤-sheet, there is a leucine residue, corresponding to Leu 25 , that is conserved as a small hydrophobic amino acid in CXC chemokines, but is always a tyrosine in CC chemokines. We have made the Leu 25 3 Tyr mutant and shown its ability to attract peripheral blood monocytes and displace MIP-1␣ from its CC CKR-1 receptor, two activities that IL-8 does not possess (14). We have further investigated the molecular basis of this receptor selectivity by making the Leu 25 3 Cys mutation. Cysteine is approximately isosteric with leucine and would be expected to have similar activity compared with wild-type IL-8. However, the free thiol group is chemically reactive and can be modified with a variety of reagents such as the fluorescent group N,NЈ-dimethyl-N-(iodoacetyl)-NЈ-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)ethylenediamine (IANBD-amide) that we added onto Cys 25 .
We show here that the Leu 25 3 Cys mutant binds to IL-8R-A and IL-8R-B and activates neutrophils while having no effect on monocytes. Chemical modification of the cysteine residue with a fluorescent NBD group, however, introduces monocyte chemoattraction and the ability to displace MIP-1␣ from its CC CKR-1 receptor. This fluorescent probe has been used to study heterodimer formation when unlabeled IL-8 is added as well as binding to a 34-amino acid peptide from the C terminus of the CC CKR-1 receptor. Finally, by comparison of the CD spectra, we can show that IL-8 mutants showing CC chemokine activity have a lower ␣-helical content compared with IL-8. This indicates that a distortion of the C-terminal helix is important in * 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.
altering the selectivity between CXC and CC chemokines and suggests a possible mechanism for receptor selectivity.

EXPERIMENTAL PROCEDURES
Reagents-Unless otherwise stated, all chemicals were purchased from Sigma.
Construction and Expression of IL-8 Leu 25 3 Cys-Human IL-8, cloned by polymerase chain reaction from a peripheral blood monocyte gt11 cDNA library, was expressed in Escherichia coli B-cells using the trp promoter (19). Mutant proteins were made by resynthesizing a cassette of DNA between the NdeI and SacI sites coding for amino acids 1-50. The Leu 25 3 Cys mutant was constructed using the following oligonucleotides, and the sequence was verified by the dideoxy chain termination method: 5Ј-CCTTTCCACCCCAAATTTATCAAAGAATGT-AGAGTGATTGAGAGT-3Ј (upper strand) and 5Ј-TGGTCCACTCTC-AATCACTCTACATTCTTTGATAAATTTGGGGTG-3Ј (lower strand). Both wild-type IL-8 and the Leu 25 3 Cys mutant are highly expressed using this system.
Purification of IL-8 Leu 25 3 Cys-The purification using 100 g of E. coli cells, wet weight, was similar to the method described previously (14). The glutathione used in the renaturation step was shown to be attached to the thiol and was consequently removed by dissolving the protein lyophilysate in PBS (140 mM NaCl, 3 mM KCl, 8 mM Na 2 HPO 4 , 1.5 mM KH 2 PO 4 , pH 7.4), adding dithiothreitol to a final concentration of 1 mM, and incubating for 1 h at 37°C. The sample was desalted on a PD-10 gel filtration column in the same buffer, dialyzed against 0.1% trifluoroacetic acid, and lyophilized. The molecular mass of the final protein was verified by electrospray ionization mass spectroscopy (15).
Analytical Methods-Amino acid analysis was performed after gasphase hydrolysis of 1 nmol of protein in 6 M HCl containing 1 mg/ml phenol at 112°C for 24 h. The resultant amino acids were quantified using a Beckman 6300 system. Titration of thiols was done using 5,5Ј-dithiobis(2-nitrobenzoate) (Ellman's reagent) (16). For the determination of accessible free thiols, 1 mg of protein was dissolved in 0.1 M Tris-HCl, pH 7.5, containing 1% sodium dodecyl sulfate. A fresh solution of 5,5Ј-dithiobis(2-nitrobenzoate) (0.01 M in 0.05 M sodium phosphate buffer, pH 7.0) was added to the protein solution, and the increase in absorbance at 412 nm was monitored until a stable value was reached. The calculation of the number of free thiol groups was carried out using an absorbance coefficient of ⑀ ϭ 13,000 M Ϫ1 cm Ϫ1 . To determine the total number of sulfhydryl groups, 1 mg of protein was reduced with 1 mM dithiothreitol and denatured in 6 M guanidine hydrochloride at 60°C for 1 h. Excess dithiothreitol was removed on a Brownlee C-8 HPLC column (220 ϫ 2.1 mm), and the protein was lyophilized. The lyophilysate was taken up in the above-mentioned Tris buffer, and the determination of thiols was carried out as described above.
Coupling of IANBD-amide to IL-8 Leu 25 3 Cys-IANBD-amide (Molecular Probes, Inc., Eugene, OR) is a hydrolytically stable iodoacetamide that has fluorescein-like spectra with excitation and emission maxima at 472 and 536 nm, respectively. The iodoacetamide group reacts irreversibly with free thiols. 0.84 mg of 100 M IL-8 Leu 25 3 Cys was dissolved in 1 ml of 50 mM borate buffer, pH 8.5. IANBD-amide dissolved in N,N-dimethylformamide was added to the protein solution in 10-fold excess. The solution was incubated at room temperature for 1 h in the dark, and the reaction was quenched by addition of 10 l of 1 M cysteine. The modified protein was dialyzed against 0.1% trifluoroacetic acid for 6 h and then lyophilized. Mass spectroscopy was carried out to verify the correct molecular mass of the mutant protein coupled to IANBD-amide, and peptide mapping was carried out to verify that the modified residue was Cys 25 and not one of the four other cysteines present in IL-8. For the peptide mapping, 4 g of Staphylococcus aureus V8 protease (Boehringer Mannheim; sequencing-grade) was added to 10 nM protein solubilized in 25 mM NH 4 HCO 3 , pH 7.8 (protein/enzyme ratio of 20:1), and the solution was incubated at 37°C for 18 h in the dark. The peptides generated by proteolytic digestion were separated by reverse-phase HPLC on a Brownlee C-18 column (220 ϫ 2.1 mm). The peptides were eluted with 0.1% trifluoroacetic acid using a gradient from 0 to 55% over 60 min with 90% acetonitrile, 10% H 2 O containing 0.1% trifluoroacetic acid. The peptides were monitored using a diode array detector at 214 and 480 nm and collected manually for sequencing. The peptide showing absorbance at 480 nm was sequenced using an Applied Biosystems Model 477A pulsed liquid-phase protein sequencer with a Model 120A on-line phenylthiohydantoin (PTH)-derivative analyzer. One-third of the PTH-derivatives were collected in the internal fraction collector. Cycle 25 showed an amino acid with a novel PTHderivative retention time. The absorption spectrum of this PTH-derivative was measured on a Hewlett-Packard 8452A diode array spectro-photometer against a blank containing 0.1% trifluoroacetic acid.
Monocyte and Neutrophil Chemotaxis Assays-The purification of human polymorphonuclear leukocytes and monocytes was carried out using fresh blood by sedimentation over dextran sulfate followed by Ficoll gradient centrifugation (14). Neutrophil chemotaxis was measured using a colorimetric assay described by Shi et al. (17). A 96-well chamber in which a microtiter plate can be placed (NeuroProbe, Cabin John, MD) was used. Serial dilutions of the chemoattractants were made in RPMI 1640 medium containing 2 mM L-glutamine, 25 mM Hepes, and 10% heat-inactivated fetal calf serum. 370 l of chemoattractant was added to the lower chamber of the assay wells and covered with a polyvinylpyrrolidone-free polycarbonate membrane (pore size of 3 m). 200 l of 10 6 cells/ml was added to the top wells, and the assay plate was incubated at 37°C for 1 h. The liquid in the upper wells was removed by aspiration, and 100 l of ice-cold PBS containing 20 M EDTA was added to each well in order to detach the cells from the underside of the membrane. After 30 min of incubation at 4°C, the liquid was removed by aspiration. Centrifugation at 500 ϫ g for 10 min allowed the migrating cells to collect in the bottom wells. The supernatant was removed by aspiration, and 100 l of 5 mg/ml dye solution (Promega) in RPMI 1640 medium was added to the cells and incubated overnight at 37°C in the dark. The reaction that monitors the conversion of tetrazolium blue into its formazan product was stopped with 100 l of propan-2-ol acid (2 mM HCl), and the plate was left at room temperature for 3 h. Absorbance was recorded at 540 nm using a microtiter plate reader (Thermomax, Molecular Devices, Menlo Park, CA). The protocol for the monocyte chemotaxis assay using the 48-well micro-Boyden chamber was described previously (14).

IL-8 Receptor Binding
Assay-This assay was performed using HL-60 cells transfected by electroporation with the cDNA sequence for IL-8R-A and IL-8R-B. A final volume of 150 l was added to 96-well multiscreen filter plates containing 3 ϫ 10 5 HL-60 cells, 0.23 nM 125 Ilabeled IL-8, and varying concentrations of chemokine. After incubation at 4°C, the cells were washed, and the bound radioactivity was counted (14). The data obtained were fitted using Grafit Version 3.01 (18)  CC CKR-1 Receptor Binding Assay-Binding of 0.3 nM 125 I-labeled MIP-1␣ to CC CKR-1 receptors on 3 ϫ 10 5 COS-7 cells was competed by addition of varying concentrations of unlabeled chemokine. The activity retained on the filters after washing was counted as described previously (14).
Fluorescence Studies of the Interaction of IL-8 Leu 25 3 Cys-NBD-The fluorescence of the NBD group was measured at 20°C on a Jasco FP-777 spectrofluorometer (Omnilab AG, Geneva) using a quartz cuvette of 1-cm path length with an excitation wavelength of 472 and detecting emitted light at 536 nm. The mutant protein was dissolved at a concentration of 7.5 nM in PBS, pH 7.4. The effect of added IL-8 was studied by incubating the Leu 25 3 Cys-NBD mutant with increasing concentrations of wild-type IL-8.
The interaction of the fluorescent NBD mutant with the amino terminus of CC CKR-1 was studied using a synthetic peptide corresponding to the N-terminal extracellular domain of this receptor. This amino-terminal peptide, M34A (Neosystem S. A., Strasbourg, France), has the following sequence: 1 METPNTTEDYDTTTEFDYGDATPCQK-VNERAFGA 34 . The change in fluorescence intensity was monitored when varying concentrations of the receptor peptide (10 Ϫ5 to 10 Ϫ11 M) were added to 10 nM Leu 25 3 Cys-NBD mutant in PBS, pH 7.4, using the same conditions described above. Two control studies were carried out; one used IL-8 that was labeled with NBD at the N terminus using a procedure similar to that described by Alouani et al. (19). The second control study used RANTES that had been coupled with a NBD group via a lysine spacer to the oxidized N-terminal serine of RANTES using the same method.
Far-ultraviolet Circular Dichroism-CD spectroscopy measurements were made in the far-ultaviolet range (198 -250 nm), where the spectrum of proteins in solution is sensitive to their secondary structure. The measurements were done on an Aviv Model 62DS circular dichroism spectrometer using quartz cuvettes of 1-cm path length at 20.5°C. Lyophilized samples were resuspended in 10 mM Tris-HCl, pH 7.5, at 0.03 mg/ml. A modified method of Hennessey and Johnson (20) was used to analyze the CD spectra in which 26 data points were used to describe the measured CD spectrum between 198 and 250 nm. These points were then fitted with a reference data set of 15 proteins of known three-dimensional structure (21).

Mutagenesis and
Expression-A trp expression system was used to express human recombinant IL-8 in E. coli B-cells (19). The IL-8 Leu 25 3 Cys mutant was assembled using an oligonucleotide cassette method. The purification and renaturation methods were identical for both proteins. The molecular mass of the IL-8 Leu 25 3 Cys mutant was found to be 8676.3 Ϯ 0.9 Da instead of 8372 Da. This additional mass of 304 could be attributed to the presence of glutathione (predicted additional molecular mass ϭ 305.3) used in the renaturation step attached to the free thiol. Upon removal of the glutathione, electrospray ionization mass spectroscopy for the mutant gave the expected mass of 8371.0 Ϯ 1.3 Da. 100 g of cell paste yielded ϳ5 mg of pure protein, which was judged to be 98% pure by SDS-polyacrylamide gel electrophoresis (Fig. 1).
Analysis of IL-8 Leu 25 3 Cys-The determination of free sulfhydryl groups was performed using 5,5Ј-dithiobis(2-nitrobenzoate) with wild-type IL-8 as a control (16). The total number of thiols and the free number of thiols were found to be 3.95 and 0.02 for wild-type IL-8, respectively. This increased to 4.63 and 0.65 for the Leu 25 3 Cys mutant protein. Amino acid analysis of the IL-8 Leu 25 3 Cys protein confirmed the expected composition.
Fluorescent Labeling of IL-8 Leu 25 3 Cys with IANBDamide-Analysis of the HPLC profile of the mutant protein after labeling with IANBD-amide indicated that all the initial material had reacted (Fig. 2). The IL-8 Leu 25 3 Cys mutant migrated as a single species on reverse-phase HPLC, with a retention time of 32.78 min. After incubation with IANBD, a new peak appeared at 34.97 min on the HPLC trace. Base-line separation of the starting material and product indicated that the neutrophil activity seen for the modified protein was not due to contamination with the IL-8 Leu 25 3 Cys mutant. To confirm the position of the incorporation of the NBD group, a peptide digest was performed using Glu-C protease. The fragments were analyzed by reverse-phase HPLC, detecting absorbance at 495 and 214 nm. Only one peptide had significant absorbance at 495 nm, indicating the presence of the NBD group. Sequencing of this peptide gave the sequence Glu 24 -Xaa-Arg-Val-Ile-Glu 29 . The unidentified PTH-derivative at position 25 had an absorption spectrum showing a maximum at 495 nm, confirming the incorporation of NBD at this position (Fig. 3).
Assay for Chemotactic and Receptor Binding Activity-In neutrophil chemotaxis, IL-8 shows a saturating dose-response curve with a midpoint at ϳ1.2 nM and a maximum efficacy of 12 ( Fig. 4). The IL-8 Leu 25 3 Cys mutant has a midpoint at 3 nM, and the IL-8 Leu 25 3 Cys-NBD mutant has a midpoint at 27 nM. In these assays, RANTES showed no ability to cause neutrophil chemotaxis.
Receptor binding was assayed by displacement of 125 I-labeled IL-8 from IL-8R-A or IL-8R-B on HL-60 cells. IL-8 shows equal affinity for both receptors, with IC 50 values of 1.4 Ϯ 0.1 nM for IL-8R-A and 1.9 Ϯ 0.3 nM for IL-8R-B (Fig. 5). Under the conditions of the assay, where the concentration of 125 I-labeled IL-8 is much lower than its K d value, the IC 50 values for the mutants equal the K d values, to a first approximation (28). The IL-8 Leu 25 25 3 Cys mutant before and after labeling with IANBD-amide. The initial material migrated with a retention time of 32.78 min. As it was labeled with the fluorescent probe, a new peak appeared at 34.97 min. The peptides were separated on a reverse-phase Brownlee C-4 column (30 ϫ 2.1 mm). 25 3 Cys. A, a single major peak was seen when absorbance was monitored at 495 nm; B, Edman sequencing of this peptide showed the release of a PTH-derivative in cycle 25 with a characteristic absorbance at 495 nm, confirming the site of modification.
In a chemotaxis assay using peripheral blood monocytes, RANTES gives a bell-shaped curve with a maximum at 1 nM and a maximal efficacy of 6 (Fig. 6). The IL-8 Leu 25 3 Cys-NBD mutant protein is also able to induce monocyte chemotaxis, with a maximum at 12 nM and an efficacy similar to that of RANTES. This mutant also displaces MIP-1␣ from the MIP-1␣/RANTES (CC CKR-1) receptor (Fig. 7). MIP-1␣ can displace the radioactive ligand from the receptor, with an IC 50 of 0.97 Ϯ 0.03 nM, and the IL-8 Leu 25 3 Cys-NBD mutant displaces 125 I-labeled MIP-1␣, with an IC 50 of 118 Ϯ 0.8 nM. Both IL-8 and the IL-8 Leu 25 3 Cys mutant show no activity in these two assays.
Fluorescence Studies-Addition of varying amounts of wildtype IL-8 to a solution of 7.5 nM IL-8 Leu 25 3 Cys-NBD caused a concentration-dependent increase in fluorescence (Fig. 8). These data fitted well with the binding isotherm, ϭ final ϩ ( initial Ϫ final ) ϫ L/([L]/IC 50 ϩ [L]), where initial ϭ 1217 units was the initial fluorescence of the IL-8 Leu 25 3 Cys-NBD in solution and final ϭ 2052 units was the limiting fluorescence at infinite concentration of added IL-8. The IC 50 or midpoint concentration was 110 nM. When IL-8 Leu 25 3 Cys-NBD was mixed with a 34-amino acid peptide from the amino terminus of CC CKR-1, there was a dose-dependent fluorescence decrease from 802 to 461 units (Fig. 9). The data were fitted to the binding isotherm, and a value of 156 nM was calculated for the IC 50 . Similar data were obtained when RANTES specifically labeled at the N terminus was used, giving an IC 50 of 40 nM. However, no change in fluorescence was observed with Nterminally labeled IL-8, confirming that the change in fluorescence was due to a specific interaction with the N-terminal receptor peptide.
CD Measurements-We have carried out CD measurements in the far-ultraviolet range (198 -250 nm) of 0.03 mg/ml samples of IL-8, RANTES, MIP-1␤, and the three IL-8 variants at 20.5°C in 10 mM Tris-HCl, pH 7.5. As can be seen in Fig. 10, there are clear differences between the spectra obtained for CXC and CC chemokines. In addition, the spectra for the IL-8 mutants showing monocyte chemotactic activity and CC CKR-1 binding show a third class of spectra compared with the wild- type chemokine data. The results were compared with a standard data set (21) in order to determine the relative amounts of the secondary structural elements such as ␣-helix and ␤-sheet.
To validate our results, we compared the results obtained experimentally by CD with the theoretical values from known NMR data of IL-8 (22), RANTES (23), and MIP-1␤ (24) ( Table  I). We see that in all three cases, the experimental and theoretical values are only different by 1%. The IL-8 Leu 25 3 Cys mutant shows similar structure to IL-8, but with a slight loss of ␣-helix. The IL-8 Leu 25 3 Tyr and IL-8 Leu 25 3 Cys-NBD mutants, however, show a marked decrease in the ␣-helical content with percent averages of 6 and 8, respectively, which is even lower than that for the CC chemokines. The other two secondary structural classes remain unchanged when compared with IL-8.

DISCUSSION
Chemokines play a key role in inflammatory diseases by selectively recruiting and activating a wide variety of cells, including leukocytes. The initial stage of this cellular activation involves the binding of chemokines to a family of seven transmembrane domain G-protein-coupled receptors. To date, two receptors have been cloned for CXC chemokines: IL-8R-A (1) and IL-8R-B (2). cDNAs for four human CC chemokine receptors have also been cloned (5)(6)(7)(8)(9)(10)(11). The available data using these recombinant receptor clones clearly show that CXC chemokines do not bind to CC chemokine receptors or vice versa. The only receptor that has been found that binds both classes of ligand is a ubiquitous chemokine receptor on erythrocytes known as the Duffy antigen (25). So far, however, this has not been shown to induce a signaling response to any chemokine.
In the attempt to understand the molecular basis of this selectivity, the three-dimensional structures of the ligands were initially studied. These studies show that the monomeric structures of both CXC and CC chemokines are similar, even though their sequence identity is lower than 25% in many cases. However, the dimeric interfaces for CXC and CC chemokines are different, suggesting that factors such as the hydrophobicity of the dimeric interface play a role in receptor selectivity. Multiple sequence alignments have enabled us to identify a conserved small hydrophobic amino acid in CXC chemokines, corresponding to Leu 25 in IL- 8 (14), that is always replaced by a much larger tyrosine in CC chemokines. The Leu 25 3 Tyr mutation introduces two novel CC chemokine-like activities into IL-8, namely monocyte chemotaxis and the ability to bind CC CKR-1.
To further investigate the role of this important residue, we have made the Leu 25 3 Cys mutant. Since leucine and cysteine are approximately isosteric, we would predict that this mutation would produce only a minimal change in the activity of IL-8. However, this mutation introduces a chemically reactive group, which could be modified with a variety of reagents such as a hydrophobic fluorescent group. This in turn would enable us to monitor the binding of mutant IL-8 to form receptor complexes.
The Leu 25 3 Cys mutant refolds correctly with two disulfide   (20). The standard deviation is Ϯ5%. Values in italics were calculated from the NMR structures for these three chemokines (22)(23)(24) bonds and one free thiol that can be attributed to Cys 25 . There should be no disulfide bond formation across the dimeric interface since the distance between the two thiols (calculated from the structure of wild-type IL-8) is 5.8 Å. This distance is much longer than the 3 Å normally seen in disulfide bonds. The Leu 25 3 Cys mutant is overall similar to wild-type IL-8. It binds IL-8R-A 50-fold and IL-8R-B 10-fold weaker than wild-type IL-8 and activates neutrophils with only 2-fold less potency. No effect was seen in monocyte chemotaxis assays, and the mutant was unable to displace MIP-1␣ from its CC CKR-1 receptor. The modification of the free cysteine with the fluorescent NBD group was shown to be stoichiometric. HPLC purification confirmed that there was no unlabeled starting material in the final product. Addition of the bulky aromatic NBD group caused a dramatic 100-fold decrease in binding to IL-8R-A and IL-8R-B and a concomitant 30-fold decrease in potency in neutrophil chemotaxis assays. These results are consistent with those reported for the IL-8 Leu 25 3 Tyr mutation (14). In addition, the NBD-modified mutant can compete with MIP-1␣ for binding to CC CKR-1, with a 118-fold lower potency compared with MIP-1␣. The mutant protein can also signal through the receptor, as can be seen by its ability to attract monocytes.
The fluorescent NBD probe is sensitive to its local environment (26), and this property can be used to study the interaction of IL-8 Leu 25 3 Cys-NBD with other proteins. When the labeled mutant protein is incubated with wild-type IL-8, there is an increase in the fluorescent signal, which corresponds to a local increase in hydrophobicity around residue 25. We predict that the proteins are forming heterodimers and that the NBD group is buried in the hydrophobic pocket between the two IL-8 C-terminal helices. We are currently trying to crystallize the mutant protein to verify this hypothesis.
The four CC chemokine receptors show over 50% amino acid identity along their entire length, but have N-terminal extracellular regions that are very different. We have synthesized the N-terminal 34-amino acid extracellular region of CC CKR-1 and studied the effect of adding this peptide to Leu 25 3 Cys-NBD. The N-terminal extracellular region has been shown to be important in determining the ligand specificity of IL-8R-A and IL-8R-B. NMR studies have shown that an interaction occurs between a peptide corresponding to the amino-terminal fragment of the type 1 human IL-8R and IL-8 complexes (27). Addition of the CC CKR-1 peptide to Leu 25 3 Cys-NBD causes a decrease in fluorescence intensity, suggesting that a complex is formed when the NBD group enters a more polar environment. The specificity of the receptor is maintained in this peptide since a similar change in fluorescence can be seen when N-terminally labeled RANTES is used as the ligand. However, no change is fluorescence is observed with N-terminally labeled IL-8. We are currently attempting to obtain high levels of receptor expression for CC CKR-1, which would enable us to study the interaction with the receptor in the membrane.
To analyze the possible three-dimensional effects of the mutations, we have studied the CD spectra of wild-type and mutant chemokines. When the secondary structure compositions of the wild-type chemokines such as IL-8, RANTES, and MIP-1␤ were calculated from the CD data, there was a very good agreement of the ␣-helical and ␤-sheet content as compared with the NMR determinations. The CD spectra of the CXC and CC chemokines are clearly different: the CC chemo-kines show a much lower ␣-helical content, consistent with a much shorter C-terminal helix. However, the IL-8 mutants that bind to CC CKR-1 and that can induce monocyte chemotaxis (Leu 25 3 Tyr and Leu 25 3 Cys-NBD) show a characteristic third class of CD spectra with a lower ␣-helical content compared with wild-type IL-8. Since in the IL-8 structure, Leu 25 is close to the C-terminal helix, it is tempting to suggest that the introduction of a large aromatic and hydrophobic group close to the helix is causing some distortion of the helix/ sheet interface. This helical distortion in turn may lead to a distortion of the amide bonds and a characteristic new CD spectrum. We are currently solving the structures of these mutants by x-ray crystallography in order to investigate the three-dimensional basis of the change in receptor selectivity.