The monocyte chemotactic protein-1 (MCP-1) receptor (MCP-1R) is expressed on monocytes, a subpopulation of memory T lymphocytes, and basophils. Two alternatively spliced forms of MCP-1R, CCR2A and CCR2B, exist and differ only in their carboxyl-terminal tails. To determine whether CCR2A and CCR2B receptors function similarly, Jurkat T cells were stably transfected with plasmids encoding the human CCR2A or CCR2B gene. Nanomolar concentrations of MCP-1 induced chemotaxis in the CCR2B transfectants that express high, intermediate, and low levels of MCP-1R. Peak chemotactic activity was shifted to the right as receptor number decreased. Five-fold more MCP-1 was required to initiate chemotaxis of the CCR2A low transfectant, but the peak of chemotaxis was similar for the CCR2A and CCR2B transfectants expressing similar numbers of receptors. MCP-1-induced chemotaxis was sensitive to pertussis toxin, implying that both CCR2A and CCR2B are Giα protein coupled. MCP-1 induced a transient Ca2+ flux in the CCR2B transfectant that was partially sensitive to pertussis toxin. In contrast, MCP-1 did not induce Ca2+ flux in the CCR2A transfectant. Since MCP-1 can stimulate chemotaxis of the CCR2A transfectant without inducing Ca2+ mobilization, Ca2+ flux may not be required for MCP-1-induced chemotaxis in the Jurkat transfectants. These results indicate that functional differences exist between the CCR2A and CCR2B transfectants that can be attributed solely to differences in the carboxyl-terminal tail.
Chemokines (chemoattractant cytokines) are involved in the attraction and activation of leukocytes during the inflammatory process (1). They are known to be involved in hemopoiesis, T cell activation, and neutrophil degranulation. Chemokines are small 8- to 10-kDa secreted proteins that are structurally related by a four-cysteine motif (2, 3, 4, 5). The chemokine superfamily is divided into two major subfamilies, α (C-X-C) and β (C-C), based on the presence or absence of an intervening amino acid between the first two of the four conserved cysteines. Two minor chemokine subfamilies have been described recently that lack the typical cysteine distribution. They are the C, γ chemokines represented by lymphotactin (6) and the CX3C, δ chemokines represented by fractalkine (7). For the most part, C-X-C chemokines (e.g., IL-8, melanocyte growth stimulatory activity/growth-regulated oncogene α, epithelial cell-derived neutrophil-activating peptide-78, and neutrophil-activating peptide-2) are potent neutrophil chemoattractants and C-C chemokines (e.g., monocyte chemotactic protein-1 (MCP-1),2 MCP-2, MCP-3, macrophage inflammatory protein-1α, macrophage inflammatory protein-1β, RANTES, and eotaxin) are monocyte, lymphocyte, basophil, eosinophil, and/or NK cell chemoattractants (8). The C chemokine lymphotactin lacks two of the four conserved cysteines and has chemotactic activity for lymphocytes, but not for monocytes or neutrophils (6). The CX3C chemokine fractalkine is composed of a membrane-bound form that is induced on activated primary endothelial cells and promotes strong adhesion of leukocytes and a soluble or shed form that has potent chemoattractant activity for T cells and monocytes (7).
MCP-1 is made by almost all cells or tissues examined upon stimulation by a variety of agents, but its targets are limited to monocytes/macrophages, basophils, mast cells, T lymphocytes, NK cells, and dendritic cells that express CCR2 (9, 10, 11, 12). MCP-1 was first designated monocyte chemotactic and activating factor because it stimulated chemotactic migration of human monocytes, but not neutrophils, and could activate monocytes to kill tumor targets in vitro (13). Although MCP-1 was first identified as a monocyte chemoattractant, Carr et al. (11) later identified MCP-1 as a major chemoattractant for T cells of the memory phenotype. MCP-1 also activates adhesion of T lymphocytes to fibronectin through activation of β1 integrins (14, 15).
The chemokine receptors belong to the serpentine family of G protein-coupled receptors (4). There are two alternatively spliced forms of the MCP-1R, designated MCP-1RA (CCR2A) and MCP-1RB (CCR2B), which differ only in their carboxyl-terminal tails (16). Myers et al. (17) studied the ligand specificity and signal transduction of human embryonic kidney-293 (HEK-293) cells transfected with CCR2B. In this study MCP-3 was defined as a ligand for MCP-1RB. Subsequent studies have shown that in addition to MCP-1 and MCP-3, MCP-2 and MCP-4 are ligands for the MCP-1R (18, 19, 20). Myers et al. (17) did not study the chemotaxis of HEK-293 transfectants expressing CCR2A or CCR2B; therefore, the question remained of whether MCP-1 can induce chemotaxis of CCR2A transfectants.
Since MCP-1R is normally expressed on T cells, we stably transfected the Jurkat T cell line with either human CCR2A or CCR2B and tested the transfectants for their ability to bind MCP-1, flux Ca2+, or chemotax following stimulation with MCP-1. In this report we show that Jurkat CCR2A and CCR2B transfectants both bind MCP-1 with high affinity and chemotax in response to MCP-1 stimulation. When CCR2A and CCR2B transfectants expressing similar numbers of receptors are compared, MCP-1 induces a transient Ca2+ flux in the CCR2B transfectant, which is not seen in the CCR2A transfectant. The chemotactic response is completely blocked by pertussis toxin, whereas the MCP-1-induced Ca2+ flux is only partially inhibited by pertussis toxin. Therefore, Ca2+ flux may not be required for MCP-1-induced chemotaxis in the Jurkat transfectants.
Materials and Methods
Recombinant human MCP-1 was purchased from PeproTech (Rocky Hill, NJ) or obtained from LeukoSite (Cambridge, MA). Propidium iodide and pertussis toxin were purchased from Calbiochem (San Diego, CA). Fluo-3, acetoxymethyl ester, was purchased from Molecular Probes (Eugene, OR). BSA (fraction V) was purchased from Sigma (St. Louis, MO).
mAb 5A11 (γ2a isotype) was produced against the human MCP-1R by immunizing mice with a synthetic peptide corresponding to the first 32 N-terminal amino acids of the MCP-1R (21). Purified 5A11 IgG was prepared from ascites by protein A affinity chromatography. mAb 1D9 (γ2a isotype) was produced against the human MCP-1R by immunizing mice with L1.2 transfectants expressing MCP-1RB (G. LaRosa, manuscript in preparation). A γ2a-negative control mAb was purchased from Becton Dickinson (San Jose, CA). FITC-labeled goat anti-mouse IgG (γ-chain-specific) Ab was purchased from Southern Biotechnology Associates (Birmingham, AL).
Cell culture and stable transfections
The T cell line Jurkat JE64-6A, a subclone of E6-1, was obtained from Dr. Yoji Shimizu (University of Minnesota, Minneapolis, MN). Jurkat JE64-6A cells were maintained in RPMI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 10% FBS (HyClone, Logan, UT), 2 mM l-glutamine, and 100 U/ml penicillin/streptomycin (complete medium). The medium was further supplemented with G418 (1 mg/ml) for transfected cultures. Cells were maintained in log phase (0.5 × 105 to 5 × 105//ml) and were seeded at 2 × 105 cells/ml the day before an experiment.
Human MCP-1R cDNA constructs (containing the coding sequence for MCP-1RA or MCP-1RB with CD5 signal peptide at the N-terminus) were cut out of the pcDNA3 vector (Invitrogen, San Diego, CA) with BamHI and XhoI and placed in the pMH-Neo vector (22). The plasmid was transformed into DH5α, isolated by CsCl density centrifugation, and linearized with ScaI. Restriction enzymes were removed using the PCR Magic prep kit (Promega, Madison, WI). Jurkat cells (5 × 106) were electroporated with 20 μg of linearized plasmid (MCP-1RA plasmid in pMH-Neo or pMH-Neo plasmid), 280 μg of salmon sperm DNA (Sigma), in 0.8 ml of HeBS (20 mM HEPES, 137 mM NaCl, 5 mM KCl, 0.7 mM Na2HPO4, and 6 mM dextrose, pH 7.05) at 280 V, and 960 μF (Bio-Rad Gene Pulser; Bio-Rad Laboratories, Hercules, CA). For MCP-1RB transfections Jurkat cells were electroporated with 40 μg of plasmid in 200 μl of Opti-MEM I (Life Technologies) at 200 V and 960 μF as previously described (23). Cells were seeded at 5 × 105 cells/ml in T-25 flasks and G418 selection was begun 2 days later. On day 20 the MCP-1RA and MCP-1RB transfectants were stained with anti-MCP-1R mAb 5A11, and the brightest 0.1 and 2%, respectively, were collected (see below).
Flow cytometric analysis and cell sorting
For FACS analysis cells were washed in staining buffer (Dulbecco’s PBS (D-PBS), 10% heat-inactivated human serum, and 0.1% NaN3), stained with saturating concentrations of mAbs, and analyzed after staining with a 1/25 dilution of FITC-conjugated goat anti-mouse IgG (Southern Biotechnology Associates) on a FACScan (Becton Dickinson) equipped with CellQuest data analysis software. Propidium iodide was included to gate out dead cells. Ten thousand viable cells were acquired and analyzed in D-PBS and 0.1% NaN3 containing 1% FBS. An isotype-matched negative control mAb stain was performed for each transfectant.
For sterile cell sorting, cells were stained as described above with staining buffer that did not contain NaN3. Propidium iodide (2.5 μg/ml final) was added to cells resuspended in D-PBS containing 5% FBS just before the cells were sorted on an Coulter EPICS Elite ESP flow cytometer (Coulter, Miami, FL) equipped with an autoclone. For each batch sort, the brightest 2% of the MCP-1R-positive population was collected and grown, except for the first two batch sorts for MCP-1RA, when <1% of the cells were positive. The MCP-1RA transfectants were batch-sorted eight times, and single-cell clones were isolated using the autoclone. The Jurkat MCP-1RB transfectants were single-cell cloned by limiting dilution after batch sorting one time.
MCP-1 binding assay
Membranes used in the MCP-1 binding assay were prepared from three MCP-1RB clones (D3-3s, F6-5s, and E7-3s) that expressed high, intermediate, and low levels of MCP-1RB, respectively, and one MCP-1RA clone (G5-5) that expressed low levels of MCP-1RA. Cells were harvested by centrifugation and washed twice in ice-cold PBS. Cells were resuspended in ice-cold lysis buffer (5 mM HEPES (pH 7.5); 2 mM EDTA; 5 μg/ml each of leupeptin, aprotinin, and chymostatin; and 100 μg/ml PMSF) at a concentration of 5 × 107 cells/ml. The cell suspension was homogenized 10–15 times using a Dounce homogenizer (Kontes; Vineland, NJ) and a B pestle on ice. Nuclei and debris were removed by centrifugation at 500-1000 × g for 10 min at 4°C. The supernatant was transferred to a fresh tube and centrifuged at 25,000 × g for 30 min at 4°C. The supernatant was aspirated, and the pellet was resuspended in freezing buffer (10 mM HEPES (pH 7.5); 300 mM sucrose; 1 μg/ml each of leupeptin, aprotinin, and chymostatin; and 10 μg/ml PMSF) using a homogenizer until all clumps were resolved. Membranes were aliquoted and frozen at −70–85°C until needed. Typical binding assays used 2.5–40 μg protein/well.
For MCP-1 competition assay, different amounts of membranes were included in the binding reaction along with 0.36 nM 125I-labeled MCP-1 and increasing concentrations of cold MCP-1 (0.1–500 nM). Binding reactions were performed in a final volume of 100 μl in a binding buffer containing 10 mM HEPES (pH 7.2), 1 mM CaCl2, 5 mM MgCl2, and 0.5% BSA. To determine inhibition of MCP-1 binding by guanosine 5′-o-(3-thiotriphosphate) (GTPγS), Jurkat membranes were incubated with 125I-labeled MCP (0.36 nM) with and without 5 μM GTPγS. After 30 min at room temperature, the binding reactions were filtered through GF/C filters (glass fiber filters, type C; Whatman, Clifton, NJ) that had been presoaked with 0.3% polyethyleneimine and washed twice with binding buffer containing 0.5 M NaCl. Filters were dried and counted in a β-Max scintillation counter using MicroScint 20 (Packard, Meriden, CT) scintillation fluid. Data were analyzed using GraphPad Prism (version 3.0; GraphPad Software, San Diego, CA).
Fluorometric imaging plate reader Ca2+ flux assay
Ca2+ flux was measured in Jurkat T cell transfectants using the fluorescent Ca2+ indicator dye fluo-3. Briefly, cells were washed and resuspended in complete supplemented medium prewarmed to 37°C at a final concentration of 3.75 × 106 cells/ml. The cells were loaded with 4 μM fluo-3 from a 1 mM stock solution in DMSO containing 10% pluronic acid. After 1 h in a CO2 incubator the cells were diluted 1/1 with prewarmed Ca2+ flux buffer (140 mM NaCl, 5 mM KCl, 1 mM MgSO4, 1 mM CaCl2, 20 mM HEPES, 1 mM Na2HPO4, and 5.5 mM glucose, pH 7.4) and incubated at 37°C for another 15 min. Cells were washed four times with Ca2+ flux buffer after loading with dye, resuspended at 2 × 106 cell/ml in Ca2+ flux buffer, and plated into 96-well, black-wall microplates (Corning Costar, Cambridge, MA) at 300,000 cells/well. The plate was spun for 5 min in a centrifuge (200 × g). Just before the addition of stimulus, the cells were washed twice using a Denley CELLWASH 018 (soak = height of dispense = 8; pause = speed of dispense = F; number of washes = 2; volume of wash = F). The final aspiration from the Denley cell washer was adjusted to leave 150 μl of residual buffer (same settings as above except soak = 0 and volume = 0). The microtiter plate was placed in a fluorescence-imaging plate reader (FLIPR; Molecular Devices, Sunnyvale, CA) that determined cell-associated fluorescence simultaneously in all wells of a 96-well microtiter plate maintained at 37°C (24). An additional plate was prepared that contained triplicate wells of stimulus dissolved in Ca2+ flux buffer (MCP-1 dilutions, ionomycin and buffer alone). Stimulus (50 μl) was pipetted from this plate to the cell plate. The cell monolayer was excited with an argon ion laser (488 nm), and cell-associated fluorescence was measured for 3 min (every 1 s for the first 60 s and every 5 s for the next 120 s). The instrument software normalized the fluorescent reading to give equivalent initial readings at time zero. Fluorescent counts (arbitrary units) obtained with buffer alone have been subtracted from all samples. Peak fluorescence counts were used to determine agonist activity.
Chemotactic migration assay
Cell migration was measured using Costar 24-well Transwell inserts (no. 3421, 6.5-mm diameter, 5-μm pore size; Corning Costar). Briefly, 60 μl of stimulus diluted in chemotaxis buffer (RPMI 1640 without bicarbonate, 50 mM HEPES, and 0.5% BSA, pH 7.4) from a 1 mg/ml stock (in d-PBS) was added to 0.54 ml of chemotaxis buffer in the bottom well. Cells (5 × 106/ml) in the same medium were added to the top of the insert in 200-μl aliquots. After 6-h incubation at 37°C in humidified 5% CO2, migrated cells were harvested from the bottom well and counted on a FACScan (Becton Dickinson). Cells were collected on the FACScan at a flow rate setting of 60 μl/min. Data represent the number of cells collected over 103 s. Forward and side scatters were used to exclude debris and dead cells. Data were analyzed using CellQuest software (Becton Dickinson).
Pertussis toxin treatment
For pertussis toxin studies, cells were preincubated overnight in complete supplemented medium at a density of 2 × 105 cells/ml in a tissue culture incubator. The cells were washed and loaded with fluo-3 as described for measurement of Ca2+ flux or resuspended in chemotaxis buffer when used in the chemotactic migration assay. Ca2+ flux induced by ionomycin was used to normalize cell numbers when varying concentrations of pertussis toxin-treated cells were analyzed for inhibition of MCP-1-induced Ca2+ flux.
A dose-response model was fit to the data for each competition assay. Each model produced estimates of the Kd, binding capacity (Bmax), and IC50 (50% inhibitory concentration) parameters, along with associated SEs. Kd, and Bmax estimates were each compared between the CCR2Alow and CCR2Blow transfectant binding assays via a Z statistic. The level of statistical significance was considered p < 0.0125.
MCP-1R is expressed on the surface of Jurkat CCR2A and CCR2B transfectants
By flow cytometric analysis MCP-1R can be detected on the surface of Jurkat T cell clones stably transfected with the human CCR2B or CCR2A gene (Fig. 1). The CCR2B clones D3-3s, F6-5s, and E7-3s express high, intermediate, and low levels of MCP-1RB, respectively. The CCR2A clone G5-5s expresses low levels of MCP-1RA. Clones expressing high or intermediate levels of MCP-1RA were not found. This is probably due to the fact that the cytoplasmic tail of MCP-1RA causes it to be retained within the cytoplasm of the cell (25). MCP-1R is not detected on the surface of Jurkat T cells stably transfected with the pMH-Neo vector (data not shown). Full-length sequencing, RT-PCR, and RNase protection analysis confirmed that MCP-1RB was expressed in the CCR2B transfectants, and MCP-1RA was expressed in the CCR2A transfectant (data not shown).
Jurkat CCR2A and CCR2B transfectants bind 125I-labeled MCP-1 with high affinity
125I-labeled MCP-1 bound to Jurkat CCR2B transfectants with IC50 values of 0.7–1.9 nM; this was competed with cold MCP-1 (Fig. 2). Scatchard analysis of equilibrium binding data indicated a single high affinity binding site with Kd values of 0.7–1.6 nM. The Bmax values for the CCR2B transfectants were 4.6, 3.4, and 0.8 μmol/mg protein for the high, intermediate, and low transfectants, respectively. Similarly, 125I-labeled MCP-1 bound to the Jur-kat CCR2A transfectant with an IC50 of 5 nM, which was competed with cold MCP-1 (Fig. 2). Scatchard analysis of equilibrium binding data indicated a high affinity binding site with a Kd of 3.2 nM (Fig. 2). The Bmax for the CCR2A transfectant was 0.9 μmol/mg protein. MCP-1 binding data confirms that the CCR2B and CCR2A Jurkat transfectants bind MCP-1 with high affinity (nanomolar range). Although a significant difference was detected between the Kd values for CCR2Alow and CCR2Blow transfectants in two experiments (2-fold difference), there were two experiments that did not show a statistical difference between the Kd values. Therefore, the 2-fold difference in Kd values in only 50% of the experiments is not considered biologically relevant. A statistically significant difference did not exist between the Bmax values for the CCR2Alow and CCR2Blow transfectants; therefore, the low CCR2A and CCR2B transfectants express equivalent numbers of receptors.
MCP-1 stimulates Ca2+ mobilization in the Jurkat CCR2B transfectants
MCP-1 stimulated a rapid mobilization of Ca2+ in the CCR2B transfectants that was dose dependent (Fig. 3). Ca2+ flux was observed with 10 nM MCP-1; the response was maximal at 300 nM MCP-1. Ca2+ flux was not observed with 1 nM MCP-1. The dose-response curves indicate an EC50 for MCP-1-mediated Ca2+ flux of 12–14 nM for the CCR2Bhigh and CCR2Bint clones (Fig. 4), whereas the CCR2Blow clone had an EC50 of 37 nM. Although high levels of MCP-1 were required to maximally stimulate Ca2+ flux in the CCR2B transfectants, it was specific because 300 nM MCP-1 did not stimulate Ca2+ flux in the Jurkat pMH-Neo transfectant (data not shown). In contrast to the CCR2B transfectant, Ca2+ flux was not observed in the CCR2A transfectant, when concentrations as high as 300 nM were tested (data not shown).
MCP-1 stimulates chemotaxis of Jurkat CCR2A and CCR2B transfectants
Jurkat CCR2A and CCR2B transfectants migrated in response to dilutions of MCP-1, with a typical bell-shaped dose-response curve (Fig. 5). Optimal migration was observed between 50 and 100 ng/ml MCP-1 (6.0–12 nM) for the MCP-1RBhigh transfectant and between 400 and 800 ng/ml (48–96 nM) for the CCR2Bint and CCR2Blow transfectants. Checkerboard analysis (26) indicated that the MCP-1-induced migration of CCR2B transfectants was due primarily to chemotaxis at concentrations of ≥10 ng/ml (1.2 nM) and was not due to random locomotion (Fig. 5). The CCR2A dose-response curve was shifted to the right, with optimal migration observed at 800 ng/ml MCP-1 (96 nM; Fig. 5). Checkerboard analysis indicated that MCP-1-induced migration of the CCR2A transfectant was due primarily to chemotaxis at concentrations of ≥50 ng/ml (Fig. 5). Approximately 50% of the cells migrated at the peak for the CCR2Bhigh transfectant, whereas approximately 30% of the cells migrated at the peak for the CCR2Bint and CCR2Blow transfectants. Approximately 15% of the cells migrated at the peak for the CCR2A transfectant in Fig. 5. In other experiments as many as 30% of the cells migrated at the peak for the CCR2A transfectant. Although more total cells migrated at the peak for the CCR2Bhigh transfectant, the migration index was only 7.5 (ratio of cells migrated in the presence of MCP-1/cells migrated in the absence of MCP-1). A similar migration index was obtained for the CCR2Blow transfectant. Migration indexes of approximately 18- and 15-fold were obtained for the CCR2Bint transfectant and the CCR2Alow transfectant, respectively.
Pertussis toxin inhibits MCP-1-induced Ca2+ flux and MCP-1-induced chemotaxis
MCP-1induced Ca2+ flux in the CCR2B transfectants was inhibited by pertussis toxin with an IC50 of 1–2 pg/ml (Fig. 6). Approximately 40% of the MCP-1-induced Ca2+ flux was pertussis toxin sensitive in the CCR2Bhigh transfectant, whereas approximately 70% of the MCP-1-induced Ca2+ flux was pertussis toxin sensitive in the CCR2Bint transfectant. Similarly, chemotaxis was inhibited by pertussis toxin treatment with IC50 values between 1–5 pg/ml for the CCR2A and CCR2B transfectants. In contrast to the MCP-1-induced Ca2+ flux, almost all the chemotaxis (≥90%) was sensitive to pertussis toxin inhibition (Fig. 7). This suggests that CCR2 couples to Giα and other G proteins, but the chemotactic response is through the Giα-coupled response.
GTPγS differentially inhibits MCP-1 binding to Jurkat CCR2A and CCR2B transfectants
GTPγS (5 μM) inhibited MCP-1 binding to Jurkat CCR2Bhigh and CCR2Bint transfectants similarly (inhibition, 93.9 ± 2.3% and 88.3 ± 1.9% (±SEM), respectively; Fig. 8). Significant differences were observed between CCR2Bhigh and CCR2Blow transfectants (inhibition, 74.3 ± 6.2%) and between CCR2Blow and CCR2Alow transfectants (inhibition, 36.7 ± 10.0%). This suggests that Jurkat CCR2B transfectants with low receptor numbers may couple less efficiently to G proteins. Since similar receptor numbers are expressed on the CCR2Alow and CCR2Blow transfectants (see Fig. 2), the data suggest that CCR2A may couple less efficiently than CCR2B to G proteins in the Jurkat T cell.
A previous report by Wong et al. (25) suggested that CCR2A and CCR2B expressed in HEK-293 transfectants have similar functions. Both receptors bound MCP-1 with high affinity, mobilized Ca2+ in response to nanomolar concentrations of MCP-1, and mediated agonist-dependent inhibition of adenylate cyclase in a dose-dependent manner. The EC50 values for Ca2+ mobilization and adenylate cyclase inhibition were approximately 4.3- and 7.3-fold higher, respectively, for the CCR2A HEK-293 transfectants compared with the CCR2B HEK-293 transfectants. This could have been due to the lower number of CCR2A receptors expressed on the surface of the HEK transfectants compared with the CCR2B HEK-293 transfectants.
In this report we found that CCR2A and CCR2B expressed in a Jurkat T cell bound MCP-1 with high affinity, but only CCR2B transfectants mobilized Ca2+ in response to MCP-1. To exclude the possibility that low receptor number was responsible for the lack of MCP-1-induced Ca2+ flux in the Jurkat CCR2A transfectant (CCR2Alow), we examined a CCR2B transfectant (CCR2Blow) that expressed equivalent receptor numbers. MCP-1 induced Ca2+ flux in the CCR2Blow transfectant, although the EC50 was approximately 4-fold higher than the EC50 obtained for MCP-1-induced Ca2+ flux in the Jurkat CCR2Bhigh and CCR2Bint transfectants. The EC50 for Ca2+ flux in the Jurkat CCR2B transfectants ranged from approximately 10–40 nM depending on the level of surface receptor expression, whereas an EC50 of about 3.4 nM was obtained for Ca2+ flux in CCR2B HEK-293 transfectants that express high levels of CCR2B (17). Peripheral blood monocytes mobilize Ca2+ in response to low nanomolar concentrations of MCP-1 (27, 28, 29), whereas concentrations as high as 10 nM MCP-1 did not mobilize Ca2+ in normal T cells from peripheral blood (27). Therefore, receptor number as well as differences in the receptor coupling between cell types may contribute to differences in functional activity between peripheral T cells and monocytes as well as between CCR2 expressed in Jurkat and HEK-293 cells.
We examined the pertussis toxin sensitivity of the MCP-1-induced Ca2+ flux in the Jurkat CCR2Bhigh and CCR2Bint transfectants and found that it was only partially sensitive to pertussis toxin (∼40% in the CCR2B high transfectant and ∼70% in the CCR2B intermediate transfectant). This suggests that CCR2B can couple to G proteins that are not pertussis toxin sensitive in the Jurkat T cell. Myers et al. (17) reported that 20% of the MCP-1 stimulated Ca2+ flux in CCR2B HEK-293 transfectants was insensitive to pertussis toxin. This may be due to overexpression of MCP-1R in transfected cell lines, which alternatively couple to pertussis toxin-insensitive G proteins. Arai et al. (30) identified pertussis toxin-resistant G proteins that could couple to the MCP-1 receptor by performing cotransfection experiments in COS-7 and HEK-293 cells. Both CCR2A and CCR2B couple to pertussis toxin-insensitive Gqα and G16α in COS-7 cells, whereas Gqα coupling is only observed in HEK-293 cells. These authors conclude that MCP-1 receptors couple to multiple G proteins and the coupling is cell type specific. Therefore, we postulate that differences in coupling of CCR2A and CCR2B may exist between Jurkat T cells and HEK-293 cells.
Since MCP-1 acts as a major chemoattractant of T lymphocytes in vitro (11), we examined the ability of Jurkat CCR2A and CCR2B transfectants to chemotax in response to MCP-1. Chemotaxis was observed in both CCR2A and CCR2B transfectants, which was inhibitable by pertussis toxin (≥90%), demonstrating that CCR2A, if expressed on the surface of a Jurkat T cell, couples to Giα proteins. The peak of chemotaxis was shifted to the right as CCR2B receptor number decreased. We observed a difference in chemotaxis between the CCR2Alow and the CCR2Blow transfectant even though receptor numbers were equivalent. Five-fold more MCP-1 was required to initiate chemotaxis of the CCR2Alow transfectant. Equivalent amounts of pertussis toxin (EC50 = ∼1–2 pg/ml) inhibited the chemotaxis of the CCR2Alow and CCR2Blow transfectants, whereas 2.5- to 5-fold more pertussis toxin was required to inhibit chemotaxis of the CCR2Bhigh and CCR2Bint transfectants. Arai et al. (31) showed that MCP-1-stimulated chemotaxis as well as Ca2+ mobilization in the mouse pre-B cell line, 300.19, are mediated through Giα proteins almost exclusively, whereas the MCP-1-stimulated Ca2+ flux is mediated through pertussis toxin-sensitive and -insensitive G proteins in the Jurkat, HEK, and COS-7 CCR2B transfectants (30, 31). Our data would suggest that MCP-1-stimulated chemotaxis of both CCR2A and CCR2B Jurkat transfectants is almost exclusively mediated through Giα, whereas MCP-1-stimulated Ca2+ mobilization in the Jurkat CCR2B transfectants is mediated by pertussis toxin-sensitive and -insensitive G proteins.
Since receptor number and Kd for binding MCP-1 are similar in the CCR2Alow and CCR2Blow transfectants, functional differences between CCR2Alow and CCR2Blow transfectants could be explained by differences in coupling efficiencies to G proteins within the cell. MCP-1 binding to membranes from CCR2Blow transfectants was more sensitive to inhibition with GTPγS than MCP-1 binding to membranes from CCR2Alow transfectants (Fig. 8). This may explain why the dose-response curve for CCR2A chemotaxis is shifted to the right compared with CCR2B. In addition, GTPγS inhibits MCP-1 binding to CCR2Blow transfectant membranes less efficiently than MCP-1 binding to CCR2Bhigh transfectant membranes. Therefore, CCR2B receptor number may play a role in the coupling efficiency of CCR2B to G proteins within the Jurkat T cell.
The cell background in which the MCP-1R is expressed may determine its function. This is supported by the work of Arai et al. (31), which showed that when receptors that do not normally mediate chemotaxis in the host are transfected into hemopoietic cell lines they mediate chemotaxis. Calcium mobilization alone was not sufficient to induce directed migration. Activation of Giα-coupled receptors and the subsequent release of Gβγ dimers is required to initiate signal transduction, leading to directed cell migration. Therefore, receptors that are not Giα-coupled cannot mediate chemotaxis. In addition, a recent report by Gerszten et al. (32) showed that MCP-1 and IL-8 can each rapidly cause rolling monocytes to adhere firmly onto monolayers expressing E-selectin, whereas other chemokines do not. The effects do not correlate with either the induction of a calcium transient or chemotaxis. They are mediated by activation of leukocyte integrins. Since MCP-1 can induce chemotaxis in the Jurkat CCR2A transfectant without inducing calcium mobilization, we suggest that MCP-1-induced calcium mobilization is not required for chemotaxis of the Jurkat CCR2A transfectants.
Wong et al. (25) used RT-PCR to examine the expression of CCR2A and CCR2B transcripts in monocyte cell lines and freshly isolated monocytes. CCR2B was the predominant transcript. Monocytes that differentiated into macrophages in culture showed decreased mRNA expression of CCR2A and CCR2B. One cell type that was not examined in the previous study was the T lymphocyte. CCR2A protein might be expressed on the surface of the T lymphocyte after activation. When CCR2A is transfected into multiple cell types, it is mostly retained in the cytoplasm of the cell (25). It is only with overexpression that surface protein can be detected. There may be an activation stimulus that is required for the CCR2A protein to be expressed on the cell surface. Thus, it remains to be determined whether activated T cells can be induced to express CCR2A protein on the cell surface, which can modulate the chemotactic response to MCP-1.
We thank Linda Coughenour for help with analysis of the MCP-1 binding data, Joann Schmidt for FLIPR discussions, Aleida Perez for RNase protection analysis, Paul Juneau for statistical analysis, and Karen Brown for assistance with assembling the manuscript.
Abbreviations used in this paper: MCP, monocyte chemoattractant protein; HEK, human embryonic kidney; D-PBS, Dulbecco’s PBS; FLIPR, fluorescent imaging plate reader; Bmax, binding capacity; GTPγS, guanosine 5′-o-(3-thiotriphosphate).