A Non-Glycosaminoglycan-Binding Variant of CC Chemokine Ligand 7 (Monocyte Chemoattractant Protein-3) Antagonizes Chemokine-Mediated Inflammation

Production of chemokines

Both wt and mutant CCL7 (K44AK46AR49A) were synthesized by the Merrifield solid phase method using a fully automated peptide synthesizer by Albachem. The final purity of both sequences was analyzed by high-pressure liquid chromatography and in both cases was >95%; the m.w. of each sequence was verified by electrospray-ionization mass spectrometry (ABI/Sciex QTrap; University of Newcastle Proteomics Unit). Each of the reagents used for preparation of the proteins was certified as endotoxin-free (<0.02 endotoxin units (EU)/ml).

Generation of non-heparin-binding CCL7 point mutants

Human CCL7 cDNA was cloned into the pBKS II plasmid (Stratagene) and verified by sequencing. The point mutations were introduced by inverse PCR as described previously (27). In brief, the DNA was alkali-denatured and diluted to a concentration of 100 pg per reaction to avoid the incorporation of unmutated DNA into the transformation reaction. The mutagenesis primers used are shown below with the mutated bases shown in bold and underlined: K44A (sense), 5′-GCTGTAATCTTCGCGACCAAACTGGAC-3′; (antisense), 5′-TTCCCGGGGACAGTGGCTACTGGTGGTC-3′; K46A (sense), 5′-ATCTTCAAGACCGCACTGGACAAGGAG; (antisense), 5′-TACAGCTTCCCGGGGACAGTGGCTACT-3′; K49A (sense), 5′-ACCAAACTGGACGCGGAGATCTGTGCTG-3′; (antisense), 5′-CTTGAAGATTACAGSTTCCCGGGGACAG-3′.

Amplification was performed in a DNA thermal cycler (Hybaid) for 35 cycles using Pfu Turbo DNA polymerase. The PCR products were purified and DNA was ligated and transformed into Epicurian coli competent cells. The sequence of the polyhistidine-tagged mutants was verified by DNA sequence analysis using an AB377 DNA sequencer XL (Applied Biosystems). The mutant proteins were purified as described previously (27); the purity and authenticity of the mutants was verified by reverse-phase HPLC and mass spectroscopy.

Heparin sepharose and MonoS (cation exchange) chromatography

Wild-type (wt) CCL7 or each mutant sequence (10 μg) was injected onto either a 1 ml HiTrap Heparin HP column or a MonoS cation exchange column (Amersham Pharmacia Biotech) connected to a fast protein liquid chromatography system and submitted to gradient elution with 0–2 M NaCl in 10 mM sodium phosphate (pH 7) at a flow rate of 0.5 ml/min. Eluted proteins were detected at 280 nM; fractions corresponding to absorbance peaks were verified by Western blotting.

Radioligand-binding assay

HEK-293 cells (ATCC CRL-1573) were stably transfected with human CCR1, CCR2b (28), or CCR3 (29) for use in ligand-binding experiments. Confluent cells were detached with PBS containing 3 mM EDTA, washed and resuspended in binding buffer (HBSS; 10 mM HEPES, 0.1% BSA) at a concentration of 5 × 10⁶ cells/ml. Cold-ligand competition assays were performed in a total volume of 150 μl of binding buffer using 1.5 × 10⁵ cells per sample and 100 pM ¹²⁵I-labeled CCL5 for CCR1- and CCR3-expressing transfectants, and ¹²⁵I-labeled CCL2 for CCR2b-expressing cells together with a variable concentration of either wt or mutant CCL7. After incubation of the cells at 37°C for 90 min with shaking the cells were harvested onto 0.8-μm pore cellulose acetate filter strips (Costar) using a vacuum manifold. The cells were further washed twice with HBSS containing 10 mM HEPES and 0.5M NaCl to remove any non-receptor-bound ¹²⁵I chemokines. The filters were then dried, transferred to test tubes, and the radioactivity was measured in a gamma counter (Clinnigamma; LKB-Wallac). Radioligand-binding parameters were calculated using Prism 3 software (GraphPad).

Ca²⁺ mobilization

Intracellular concentrations of Ca²⁺ were measured using Indo-1/AM. PBMC were isolated from fresh anticoagulant-treated blood using Ficoll-Hypaque (Amersham Pharmacia Biotech), washed twice in HBSS, and resuspended in HBSS containing 1 mM Ca²⁺, 1 mM MgCl₂, and 1% FCS (all from Sigma-Aldrich). The cells were loaded with 3 μM Indo-1/AM for 45 min at room temperature. After washing by centrifugation, the cells were resuspended in HBSS at a concentration of 10⁶ cells/ml and the temperature was equilibrated at 37°C for 5 min.

Chemokines were added at concentrations between 10 and 1000 ng/ml and changes in the concentration of intracellular Ca²⁺ were detected using a four-laser flow cytometer (LSRII; BD Biosciences) to monitor light emission at 440 and 530 nm following excitation at 350 nm. Each experiment was repeated three times on different days with ionomycin (10 μg/ml; Sigma-Aldrich) being used to generate a positive control signal.

In vitro chemotaxis assays

The ability of wt and mutant CCL7 to stimulate chemotaxis of PBMC or THP-1 cells (human myelomonocytic cell line; ATCC TIB-202) was assessed using a 24-well transwell system (Falcon). Chemokines were dissolved at 10 μg/ml in sterile PBS and diluted as required in HEPES-buffered RPMI 1640 (pH 7.2) supplemented with 1% BSA before adding 800 μl to the lower compartment of the assay system. The upper compartment was filled with 300 μl of a cell suspension at a concentration of 0.5 × 10⁶ cells/ml. These cells were allowed to migrate through a 3-μm pore membrane for 90 min. After this time the filter was stripped to remove cells from the upper surface and the cells on the lower surface were fixed with methanol and stained with hemotoxylin. Quantification of the chemotaxis was performed by counting the mean number of migrant cells per high power field. All assays were performed in triplicate.

For some experiments, THP-1 monocytes were incubated with no chemokine or with either wt or mutant CCL7 at 100 ng/ml in free solution for 30 min and then washed by centrifugation. These cells were then used in chemotaxis assays with either 100 ng/ml wt CCL7, mutant CCL7, or CXCL12 (PeproTech) in the lower compartment. Flow cytometric studies were also performed to assess the potential of both wt and mutant CCL7 to modulate the cell surface expression of CCR1 and CXCR4 using mAb reagents (clone 53504 and 44716; R&D Systems) together with appropriate fluorochrome-conjugated secondary Abs and isotype-matched control reagents.

Transendothelial chemotaxis

In a modified chemotaxis assay, human microvascular endothelial cells (HMEC-1) (30) were grown on the upper surface of 3-μm pore transwell filters to establish a confluent monolayer. These cells were then stimulated for 24 h with TNF-α (100 IU/ml) and IFN-γ (100 IU/ml) to up-regulate expression of the major adhesion molecules found on inflamed venules (31) and to optimize chemokine presentation. Chemokine-induced migration of the mononuclear cells was investigated as above except each assay was incubated for 2 h.

Cell recruitment to murine air pouches

Eight-week-old female BALB/c mice (Charles River) were used for generation of air pouches as described previously (32) in full compliance with U.K. Home Office regulations for animal experimentation. Briefly, air pouches were induced by injecting 3 ml of sterile air s.c. into the back of each animal followed by 1 ml of air on three further occasions (days 2, 4, and 5 respectively); this produced stable fluid-filled pouches. On day 6, each pouch was injected with 1 ml of PBS containing either 10 μg of wt CCL7, a similar quantity of the non-GAG-binding variant of CCL7 or a combination of 10 μg of both chemokine sequences. Identical age-matched control mice were injected with PBS alone. Six and 24 h later, recruited cells were recovered by gently lavaging the pouch with 1 ml of PBS containing 1 mM EDTA. The exudates were centrifuged at 1000 × g for 5 min and the supernatants removed. The cell pellets were resuspended in 1 ml of PBS for counting and assessment of viability. Two hundred microliters of each cell suspension was then smeared by cytocentrifugation (Shandon) and the cells were stained with Diff Quick (Dade; Baxter Healthcare) for differential cell counting. Further cytocentrifuge smears were labeled with Abs specific for CD3 (Serotec) and F4/80 (Serotec) to allow enumeration of T cells and mononuclear phagocytes, respectively.

An additional series of air pouch experiments was performed to examine the cell migration produced in response to 5 μg of mutant CCL7, wt CCL5 (PeproTech), or wt CXCL12 in the presence or absence of 5 μg of mutant CCL7. In all cases a group size of between three and five animals was used to allow experimental outcomes to be assessed reliably.

Immunohistochemistry

Ten-micrometer cryostat sections of dissected air pouch tissue were immunostained to detect T cells and monocytes/macrophages. Briefly, sections were fixed in acetone, treated to minimize endogenous peroxidase activity and nonspecific Ab binding before overnight incubation with primary anti-CD3 or F4/80 Abs. The secondary Ab was biotinylated sheep anti-rat IgG (Sigma-Aldrich) diluted 1/200 in normal lamb serum (NLS) containing 5% normal mouse serum. The immune complex was visualized with streptavidin-biotin-peroxidase complex (DAKO) and nickel-enhanced diaminobenzidine. Control sections were incubated with irrelevant, isotype-matched Abs in place of the primary Ab before normal development and showed negative labeling.

Synovial fluid samples

Following informed consent, synovial fluid was collected into sterile tubes from the knee joint of patients with active rheumatoid arthritis undergoing therapeutic aspiration. The samples were centrifuged and the cell-free supernatant was stored at −20°C. The fluid was diluted 1/10 into PBS for use in the lower chamber of transendothelial chemotaxis assays, as described above.

Statistical tests

All results are expressed as mean ± SEM of replicate samples and the significance of changes was assessed by the application of Student’s t test. All data were analyzed using Prism 3 software.

Identification of the heparin-binding site in CCL7

Alignment of the amino acid sequences of a series of CC-chemokines allowed location of a series of conserved basic amino acid residues conforming to the BXBXXB motif (residues 44, 46, and 49) within the CCL7 sequence. Modeling the three-dimensional structure (9) of CCL7 also implicated residues K46 and K49 as contributing to a putative GAG-binding domain. To determine the relative contribution to GAG binding of the consensus basic amino acids K44, 46, or 49, each was separately mutated to alanine by inverse PCR (27). The mass of each polyhistidine-tagged protein was verified by mass spectrometry (wt CCL7: 12,018 Da; K44A: 11,962 Da; K46A: 11,961 Da; K49A: 11,961 Da). The potential of each sequence to bind HS was measured using HiTrap Heparin HP columns. It was found that polyhistidine tagged wt CCL7 eluted at a salt concentration of 0.81 ± 0.01 M. Elution of the polyhistidine-tagged K44A mutant from the column was not significantly different to the wt protein (0.82 ± 0.01 M; p > 0.9). However, both the K46A and K49A CCL7 mutants were eluted from the column at a significantly lower salt concentration than the wt CCL7 (0.76 ± 0.01 M, p < 0.014 and 0.76 ± 0.00 M, p < 0.024, respectively).

Generation of a non-heparin-binding variant of CCL7

Molecular modeling studies (〈www.expasy.org/swissmod/SWISS-MODEL.html〉) suggested that combined mutation to alanine of all three conserved basic residues should produce a marked and localized reduction in the positive charge of the chemokine. To investigate the functional significance of this both wt CCL7 and the K44A/K46A/K49A (⁴⁴ATALDA⁴⁹) mutant protein were chemically synthesized. The final protein sequences were both >95% pure as defined by reverse-phase HPLC; Table I⇓ shows a close concordance between the predicted and experimentally defined mass for each of the synthesized proteins.

To assess the impact of this mutation on the potential for specific GAG binding both the wt and mutant sequences were assessed using heparin affinity columns. The results in Table II⇓ show that while the wt protein readily bound to heparin affinity columns and required 0.59 M NaCl for elution, the ⁴⁴ATALDA⁴⁹ mutant showed no significant affinity for heparin with the majority (95%) remaining unbound. To assess the specificity of the interaction with heparin, both wt CCL7 and the ⁴⁴ATALDA⁴⁹ sequence were also eluted from a cation exchange resin. Removal of the basic amino acid residues resulted in a decrease in the NaCl concentration required to elute the mutant sequence from the Mono S column (Table II⇓). The difference in NaCl concentration required for elution from the Mono S column and that required for elution from the heparin column for the ⁴⁴ATALDA⁴⁹ sequence was 0.45 M. This positive value indicates that the residues selected for mutation play a role in the specific interaction of CCL7 with heparin. The ⁴⁴ATALDA⁴⁹ sequence was defined as mutant CCL7′ in all subsequent assays.

Receptor ligation of a non-heparin-binding variant of CCL7

A series of radioligand binding experiments was performed to determine the impact of mutation of the three basic amino acid residues on the affinity of CCL7 for its specific receptors CCR1, CCR2b, and CCR3. A representative heterologous competition assay for CCR3 is shown in Fig. 1⇓, and indicates IC₅₀ values of 1.6 nM for wt CCL7 and 0.9 nM for the mutant sequence. The summary results in Table III⇓ are derived from three separate assays for each receptor and ligand permutation and indicate a marginally higher affinity of the mutant than the wt sequence for CCR1 (3.5-fold) and CCR3 (2-fold), while the wt protein had a 3.7-fold higher affinity than the mutant for CCR2b; in no case did these differences reach statistical significance (all p > 0.05).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Radioligand-binding assays. Results from a representative heterologous cold competition-binding assay conducted using HEK-293 cells transfected to stably express CCR3, ¹²⁵I-labeled CCL5 as the radiolabeled ligand, and a variable concentration of either unlabeled wt CCL7 (▴) or mutant CCL7 (▪) as competitors.

In vitro analysis of the cellular response to the non-heparin-binding CCL7 mutant

Ca²⁺ flux.

To investigate a possible linkage between GAG-binding capacity and cell activation, both wt and mutant CCL7 were tested in free solution for their ability to stimulate a dose-dependent increase in intracellular-free calcium in human mononuclear leukocytes. Indo-1-loaded cells were incubated with varying concentrations of either chemokine sequence and time-dependent changes in the concentration of intracellular-free calcium from baseline levels were monitored; the maximal value was recorded. A representative fluorescence profile generated in response to the addition of wt CCL7 at 1000 ng/ml to labeled PBMC is shown in Fig. 2⇓a. No difference was apparent between the maximal response to wt CCL7 and the non-GAG-binding mutant (Fig. 2⇓b), with the wt sequence generating a marginally smaller response than the mutant at 100 ng/ml and a slightly larger response at 1000 ng/ml. An additional experiment was performed to measure the intracellular Ca²⁺ flux generated by a 1:1 mixture of wt and mutant CCL7 (1000 ng/ml of each); importantly, it was found that the flux was similar to that produced by either chemokine in isolation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Calcium flux results. a, Representative changes in the intracellular Ca²⁺ concentration immediately after stimulation of PBMC with CCL7 (1000 ng/ml). b, Dose response analysis showing changes in the maximal change in intracellular Ca²⁺ concentration measured in PBMC following stimulation with either wt or mutant CCL7. The mean values are plotted as a ratio of the fluorescence emission at 440/530 nm; the data are from one of three separate experiments.

Trans-filter chemotaxis.

A series of experiments was performed to determine how mutation of the GAG binding site altered the ability of CCL7 to stimulate chemotaxis of THP-1 monocytes. The assay involved establishment of a chemokine concentration gradient by solute diffusion across a porous filter; 10 or 100 ng/ml of either chemokine sequence was added below the filter and leukocytes were introduced on the other side. It was found that the number of leukocytes stimulated to migrate by wt CCL7 was not different from that stimulated by the mutant sequence (p > 0.05; Fig. 3⇓a); similar results were observed for migration assays set up with PBMC.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

In vitro chemotaxis experiments. a, Measurement of mononuclear cell migration across cytokine-activated endothelial cell monolayers grown on filters (transendothelial) or across filters (trans-filter). Human THP-1 monocytes were stimulated with varying concentrations of chemokines at 37°C for 90 min in the case of trans-filter assays or for 2 h in the case of transendothelial assays. Similar results were observed using human PBMC. b, Examination of monocyte migration across resting endothelial cells in response to wt CCL7, mutant CCL7, and a mixture of both sequences. All assays were performed in triplicate and the number of migrant cells per high power field (×400) was counted for each membrane; data is presented as mean ± SEM.

Investigation of the role of GAG binding in chemokine-mediated inflammation

Air pouches were formed by s.c. injection of air into the back of mice; these develop a lining which resembles the synovial membrane (32). Injection of 10 μg of wt CCL7 induced significant recruitment of leukocytes within 6 h (not shown), with maximal recruitment observed after 24 h (p < 0.05; Fig. 4⇓a); at this time wt CCL7 attracted almost three times more leukocytes to the air pouches than was recruited following the injection of PBS (Table IV⇓). The recruitment of cells in response to the mutant CCL7 sequence was not significantly greater than that by injection of PBS (p > 0.05; Fig. 4⇓a); the failure of the mutant sequence to recruit leukocytes in vivo was apparent for all the major cell subpopulations enumerated by differential counting (Table IV⇓). Importantly, injection of a 1:1 mixture of wt and mutant CCL7 elicited no more leukocyte infiltration than was observed with PBS alone (Fig. 4⇓a; p > 0.05); again this effect extended to all leukocyte subpopulations (Table IV⇓).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Comparison of the potential of wt and mutant CCL7 to recruit cells using a murine air pouch model. a, Air pouches were injected with wt CCL7, mutant CCL7, or a 1:1 mixture of wt and mutant CCL7 (10 μg each), or PBS; in each case n = 5. The recruitment of cells was evaluated 24 h after administration of the chemokines. Representative data is shown from one of three separate experiments; the bars show mean values ± SEM. b, Air pouches were injected with mutant CCL7, wt CXCL12, wt CCL5 (5 μg each), or 1:1 mixtures of wt CXCL12 and mutant CCL7 or wt CCL5 and mutant CCL7 (5 μg each); in each case n = 3. The recruitment of cells was evaluated 24 h after administration of the chemokines. Representative data is shown from one of three separate experiments; the bars show mean values ± SEM.

To investigate further the potential of the non-GAG-binding mutant CCL7 to antagonize the inflammatory activity of wt chemokines, a further series of air pouch experiments was established in which the mutant sequence was mixed with wt CCL5 (which shares specific receptors with CCL7) or with wt CXCL12 (which shares no receptors with CCL7). Both wt chemokines were found to induce significant leukocyte infiltration of treated air pouches within 24 h (Fig. 4⇑b). However, mixture of either wt chemokine species with an equal quantity of mutant CCL7 reduced leukocyte recruitment in both cases to background levels (Fig. 4⇑b).

Fig. 5⇓ shows representative sections from tissue surrounding air pouches 24 h after the injection of wt CCL7 (Fig. 5⇓, a, c, and e) or mutant CCL7 (Fig. 5⇓, b, d, and f). H&E (Fig. 5⇓, a and b) staining showed a dramatic difference in the extent of infiltration of the tissues, with the mutant sequence (Fig. 5⇓b) producing minimal inflammation. The infiltrate shown in section (Fig. 5⇓a) is mainly restricted to the dermis and the disrupted layer of fatty tissue lying above the muscle. Immunoperoxidase labeling of CD3⁺ T cells (black) is shown in sections (Fig. 5⇓c) and (Fig. 5⇓d); a dense T cell infiltrate was observed in the dermis in tissue exposed to wt CCL7 but T cells were largely absent from the dermis of pouches injected with the mutant CCL7 sequence. Immunoperoxidase detection of the F4/80 Ag revealed macrophages within the dermis of sections (Fig. 5⇓e) and (Fig. 5⇓f); in contrast with the pouch contents (Table IV⇑), little difference was observed in the number of resident macrophages between the tissue from pouches exposed to wt CCL7 and mutant CCL7.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Histological evaluation of murine air pouches. Ten-micrometer cryostat sections of dissected air pouches were prepared 24 h after chemokine administration. a, c, and e, Sections taken from mice injected with wt CCL7; b, d, and f, sections from mice injected with non-GAG-binding mutant CCL7. Sections were stained with H&E (a and b), or labeled with Ab to identify CD3⁺ T cells (c and d) or Ab to identify F4/80⁺ mononuclear phagocytes (e and f).

Examination of the potential of mutant CCL7 to inhibit the chemotactic potential of synovial fluid from rheumatoid arthritis

Synovial fluid from patients with active rheumatoid arthritis is known to contain a range of chemoattractant factors, including several chemokines (33). Each of three synovial fluid samples was shown to induce significant transendothelial migration of PBMC (Fig. 6⇓a) or THP-1 monocytes (Figs. 6⇓, b and c) when used at a 1/10 dilution (all p < 0.01). Addition of 1000 ng/ml wt CCL7 did not enhance the migratory response (p > 0.05; Fig. 6⇓a) suggesting a maximal response was generated by the synovial fluid alone; addition of wt CCL7 also had no effect on monocyte migration toward synovial fluid sample b (p > 0.05; not shown) but marginally increased migration toward sample c (p < 0.05; not shown). Importantly, the addition of mutant CCL7 significantly inhibited almost all this chemotactic activity (Fig. 6⇓) when present at final concentrations of 1000 ng/ml (mean 94.3 ± 1.9% inhibition), 100 ng/ml (92.9 ± 2.4% inhibition), and 10 ng/ml (74.3 ± 9.3% inhibition).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Inhibition of the chemotactic response of mononuclear cells toward synovial fluid from three patients with active rheumatoid arthritis. The transendothelial migration of PBMC (a) or THP-1 monocytes (b and c) toward synovial fluid (1/10 dilution) was measured in the presence 10, 100, and 1000 ng/ml non-GAG-binding mutant CCL7. Each graph shows data using synovial fluid from a separate patient (a–c). Data in a demonstrate that addition of wt CCL7 did not alter migration toward the synovial fluid. All bars show the mean number of migrant cells per field ± SEM.

Examination of the antagonistic activity of mutant CCL7

Homologous and heterologous desensitization of the chemotactic response.

Fig. 7⇓ shows the results of trans-filter chemotaxis assays using monocytes which had first been incubated for 30 min in suspension with either 100 ng/ml wt or mutant CCL7 before removal of the chemokine by washing. As expected, it was found that untreated cells migrated efficiently when stimulated using 100 ng/ml of either wt or mutant CCL7 or CXCL12 in the lower compartment. However, cells which had been pretreated with either wt or mutant CCL7 showed no significant migration above background levels when stimulated with either the CCL7 sequence or CXCL12, suggesting induction of tolerance to stimulation of “homologous” CCL7 receptors and the “heterologous” receptor, CXCR4.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Demonstration of homologous and heterologous desensitization of the chemotactic response. Results from trans-filter chemotaxis assays performed using resting or chemokine pretreated monocytes which are subsequently stimulated with either medium (□), 100 ng/ml wt CCL7 (▪), 100 ng/ml mutant CCL7 (▦) or 100 ng/ml CXCL12 (▨). The bars show mean values which, for comparison, are normalized to baseline control levels ± SEM; the data are representative of three similar assays.