The trafficking of lymphocyte populations is a complex process controlled by a vast array of molecules. In this process, cells must be able to sense small changes in chemoattractant gradients. Migration through a chemotactic gradient probably employs an on-off mechanism in which chemokine receptor desensitization, internalization, and recycling may be important steps. This multistep process requires the coordinated action of many factors, including G protein-coupled receptor kinases, arrestins, clathrin, and GTP-hydrolyzing proteins such as dynamin. In this report, we show that RANTES and its derivative, aminooxypentane (AOP)-RANTES, a potent RANTES antagonist as well as an inhibitor of HIV-1 infection, both promote CCR5 desensitization involving G protein-coupled receptor kinases-2 and β-arrestin equally well. An important difference between the two molecules is that (AOP)-RANTES is more efficient than RANTES in promoting Ser/Thr phosphorylation of the receptor and association of G protein-coupled receptor kinases-2, β-arrestin, and clathrin to the CCR5. After stimulation with either ligand, we observe rapid, transient association of dynamin to CCR5, implicating this protein in receptor sensitization, but this association is faster and longer-lasting following (AOP)-RANTES stimulation. In summary, we show that chemokine receptor internalization takes place through the formation of clathrin vesicles and involves dynamin activity. We provide compelling evidence that the differences between RANTES and (AOP)-RANTES in Gαi activation condition subsequent signaling events, including internalization and receptor recycling.

The chemokines are a family of low m.w. proteins that were first described as inducible mediators of leukocyte attraction in inflammation, but several constitutively expressed chemokines were later identified that are involved in the normal traffic and homing of lymphocytes. Classification of the members of this protein family is based on the presence of four conserved cysteines; the first two can be adjacent (CC chemokines), separated by one (CXC) or three amino acid residues (CX3C), or can even lack one of the cysteines (C) (1, 2, 3). RANTES is a CC chemokine that induces migration and activation of specific leukocyte subsets by binding to several chemokine receptors (CCR1, CCR3, CCR4, and CCR5) (4). It is a potent chemoattractant for CD4+ and CD8+ lymphocytes, as well as for monocytes, NK cells, and eosinophils (5, 6, 7). RANTES can prevent infection by macrophage-tropic HIV-1 strains (8, 9), and its overexpression has been demonstrated in various chronic pathologies.

An N-terminal-modified RANTES, termed aminooxypentane (AOP)3-RANTES, has recently been described (10). This analogue binds to the CCR5 as does RANTES, but does not induce chemotaxis, thus acting as a RANTES antagonist (10, 11). We have shown that both RANTES and (AOP)-RANTES activate early signaling events following binding to CCR5, including triggering of tyrosine phosphorylation and activation of the Janus activated kinase (JAK)/STAT pathway. They also promote dimerization and tyrosine phosphorylation of the CCR5 (12), a phenomenon that has been described in other chemokine receptors following chemokine binding (13). Whereas both ligands promote Ca2+ influx, only RANTES induces sustained cell polarization and chemotaxis. (AOP)-RANTES promotes the activation and association of Gαi to the CCR5 more efficiently, although both association and dissociation occur more rapidly (12). This conditions the availability of free, active βγ subunits, which are required to induce specific ligand-induced responses such as chemotaxis (14).

After RANTES or (AOP)-RANTES binding to CCR5, there is a striking difference in their ability to trigger receptor down-modulation (11). (AOP)-RANTES thus causes a rapid decrease of >90% of cell surface-expressed CCR5, an effect that RANTES promotes less efficiently. This is due to the fact that (AOP)-RANTES inhibits recycling of internalized CCR5 to the cell surface, whereas RANTES does not (11). This effect has practical consequences, as (AOP)-RANTES is a strong inhibitor of macrophage infection by HIV-1 under conditions in which RANTES is barely effective (11).

Little is known of the regulation mechanisms in the cellular response to chemokines, or of the role of desensitization in lymphocyte migration. The trafficking of lymphocyte populations is a complex process controlled by a vast array of molecules. In this process, cells must be able to sense continuously small changes in chemoattractant gradients. Migration through a chemotactic gradient probably employs an on-off mechanism in which chemokine receptor desensitization, internalization, and recycling may be important steps. For a large number of related G protein-coupled receptors (GPCR), including chemokine receptors, rapid desensitization appears to involve agonist-promoted receptor phosphorylation by GPCR kinases (GRK) (15, 16). We demonstrated that desensitization of CCR2 in monocytes is mediated by GRK2, which translocates to the membrane following monocyte chemoattractant protein-1 (MCP-1) stimulation and phosphorylates the Ser/Thr residues in the carboxyl tail of CCR2, increasing its affinity for β-arrestin (17). This macromolecular complex prevents any further coupling between the receptor and G proteins; uncoupled receptors are subsequently removed from the plasma membrane through internalization.

In this report, we extend this model by demonstrating that RANTES- and (AOP)-RANTES-promoted CCR5 desensitization involves GRK2 and β-arrestin. We also show that chemokine receptor internalization takes place through the formation of clathrin vesicles, involving dynamin activity, and provide evidence that the differences in Gi activation between RANTES and (AOP)-RANTES condition subsequent signaling events, including internalization and receptor recycling.

Human embryonic kidney (HEK) 293 cells (TIB202) were obtained from the American Type Culture Collection (Manassas, VA), and CCR5-transfected HEK 293 cells were donated by Dr. J. Gutierrez (Department of Immunology and Oncology, Centro Nacional de Biotecnología, Madrid, Spain). Abs used include anti-phosphoserine and anti-threonine mAb (Biomol, Plymouth Meeting, PA), anti-GRK2 Ab AB9 and anti-β-arrestin-1 Ab Ab186 were donated by Dr. F. Mayor, Jr. (Centro de Biologia Molecular, Madrid, Spain) (17), anti-JAK2 polyclonal Ab (Upstate Biotechnology, Lake Placid, NY), anti-clathrin and anti-dynamin II mAb (Transduction Laboratories, Lexington, KY), anti-MHC class I (W6/32; American Type Culture Collection), and anti-β2-microglobulin (PharMingen, San Diego, CA). The anti-hGH receptor (hGHR-05) used as isotype-matched control Ab in FACS analysis was developed in our laboratory (18). Anti-CCR5 mAb CCR5-02 and CCR5-03 were generated in our laboratory (12). Recombinant human RANTES and stromal cell-derived factor (SDF)-1α were from Peprotech (London, U.K.), and (AOP)-RANTES was kindly donated by Drs. R. Offord, A. E. I. Proudfoot, and T. Wells (Serono, Geneva, Switzerland).

Cells were centrifuged (250 × g, 10 min, room temperature) and plated in V-bottom 96-well plates (2.5 × 105 cells/well). Cells untreated or treated with RANTES (10 nM), SDF-1α (10 nM), or (AOP)-RANTES (10 nM) for 30 min at 4°C were incubated with 50 μl/well biotin-labeled mAb (CCR5-03 or hGHR-05 as an isotype-matched control, 5 μg/ml, 60 min, 4°C). Cells were washed twice in PBS with 2% BSA and 2% FCS and centrifuged (250 × g, 5 min, 4°C). FITC-labeled streptavidin (Southern Biotechnologies Associates, Birmingham, AL) was added, incubated (30 min, 4°C), and plates washed twice. Cell-bound fluorescence was determined in a Profile XL flow cytometer at 525 nm (Coulter Electronics, Miami, FL).

Changes in intracellular calcium concentration were monitored using the fluorescent probe Fluo-3AM (Calbiochem, La Jolla, CA). Cells (0.5 × 106 cells/ml) were treated with 10 μg/ml of CCR5-03 mAb or 10 μg/ml of an isotype control mAb (hGHR-05) during 30 min at 37°C. After washing, cells (2.5 × 106 cells/ml) were resuspended in RPMI 1640 containing 10% FCS and 10 mM HEPES and incubated with 10 μl of Fluo-3AM/106 cells (300 mM in DMSO, 15 min, 37°C) (19). After incubation, cells were washed, resuspended in complete medium containing 2 mM CaCl2, and maintained at 4°C until just before RANTES or (AOP)-RANTES addition to minimize membrane trafficking and to eliminate spontaneous Ca2+ entry. Calcium mobilization in response to 5 nM RANTES or (AOP)-RANTES was determined at 37°C in an EPICS XL flow cytometer (Coulter) at 525 nm and includes background level stabilization and determination of the probe loading level for each sample. Only samples with a similar load, as assessed by ionophore-induced Ca2+ mobilization, were used (5 μg/ml ionomycin; Sigma, St. Louis, MO). For calcium mobilization in cells pretreated with RANTES, SDF-1α, or (AOP)-RANTES, cells were loaded with Fluo-3AM, treated with 10 nM of the different chemokines (15 min, 37°C), and, after washing, calcium flux was determined as above.

For receptor internalization analysis, CCR5-transfected HEK 293 cells were serum-starved (0.5 × 106 cells/ml, 60 min, 37°C) in DMEM medium with 0.1% BSA. After washing, cells were incubated for the times indicated with 10 nM RANTES or (AOP)-RANTES at 37°C or 4°C. Cells were washed, and CCR5 expression was determined by flow cytometry analysis using an anti-CCR5 mAb (CCR5-03) compared with an isotype-matched mAb (hGHR-05). Results are expressed as percentage of staining observed in untreated CCR5-transfected HEK 293 cells.

RANTES- or (AOP)-RANTES-stimulated cells (20 × 106) were lysed in a detergent buffer (20 mM triethanolamine, pH 8.0, 300 mM NaCl, 2 mM EDTA, 20% glycerol, 1% digitonin, with 10 mM sodium orthovanadate, 10 μg/ml leupeptin, and 10 μg/ml aprotinin) for 30 min at 4°C with continuous rocking, then centrifuged (15,000 × g, 15 min). Immunoprecipitations were performed essentially as described earlier (19). Protein extracts precleared by incubation with 20 μg of anti-mouse IgM-agarose (Sigma) were centrifuged (15,000 × g, 1 min), immunoprecipitated with the CCR5-03 mAb (5 μg/sample, 120 min, 4°C), and followed by anti-mouse IgM-agarose. Immunoprecipitates or protein extracts were separated in SDS-PAGE and transferred to nitrocellulose membranes. Western blot analysis was performed as described (18), using 2% BSA in TBS as a blocking agent for the anti-phosphoserine/threonine analysis. When stripping was required, membranes were incubated (60 min, 60°C) with 62.5 mM Tris-HCl, pH 7.8, containing 2% SDS and 0.5% 2-ME. After washing with 0.1% Tween-20 in TBS for 2 h, membranes were reblocked, reprobed with the appropriate Ab, and developed as above. In all cases, protein loading was controlled using a protein detection kit (Pierce, Rockford, IL) and, when necessary, by reprobing the membrane with the immunoprecipitating Ab.

Chemokine receptors undergo internalization after ligand activation. RANTES and (AOP)-RANTES promote CCR5 internalization in both freshly isolated primary cells and in transfected CHO cells (11). We analyzed this process in CCR5-transfected HEK 293 cells. Cells were starved for 60 min before stimulation for different periods with RANTES, (AOP)-RANTES, or SDF-1α as control. After washing to eliminate free ligand, cell-surface CCR5 was measured by FACS analysis using the specific CCR5-03 Ab. This mAb does not interfere with RANTES binding to CCR5 at either 4°C or 37°C, as assessed by FACS (Fig. 1,A) and Ca2+ mobilization assays (Fig. 1 B).

FIGURE 1.

Expression of functional CCR5 on stably transfected HEK 293 cells: effect of CCR5-03 mAb. A, Serum-starved CCR5-transfected HEK 293 cells were treated as indicated with RANTES (10 nM), (AOP)-RANTES (10 nM), or SDF-1α (10 nM; control) at 4°C. After washing, cells were incubated with biotin-labeled CCR5-03 mAb, followed by streptavidin-PE; an isotype-matched mAb was used as negative control (shaded area). One representative experiment of four performed is shown. B, CCR5-transfected HEK 293 cells were pretreated as indicated with 20 μg/ml of CCR5-03 mAb or an isotype-matched control mAb for 30 min at 37°C before determination of Ca2+ mobilization in response to 10 nM RANTES or 10 nM (AOP)-RANTES. Ca2+ mobilization was determined in a flow cytometer. Results are expressed as a percentage of the maximum chemokine response. Equivalent Fluo-3AM loading was determined as indicated (Materials and Methods). The result of one of three experiments performed is shown.

FIGURE 1.

Expression of functional CCR5 on stably transfected HEK 293 cells: effect of CCR5-03 mAb. A, Serum-starved CCR5-transfected HEK 293 cells were treated as indicated with RANTES (10 nM), (AOP)-RANTES (10 nM), or SDF-1α (10 nM; control) at 4°C. After washing, cells were incubated with biotin-labeled CCR5-03 mAb, followed by streptavidin-PE; an isotype-matched mAb was used as negative control (shaded area). One representative experiment of four performed is shown. B, CCR5-transfected HEK 293 cells were pretreated as indicated with 20 μg/ml of CCR5-03 mAb or an isotype-matched control mAb for 30 min at 37°C before determination of Ca2+ mobilization in response to 10 nM RANTES or 10 nM (AOP)-RANTES. Ca2+ mobilization was determined in a flow cytometer. Results are expressed as a percentage of the maximum chemokine response. Equivalent Fluo-3AM loading was determined as indicated (Materials and Methods). The result of one of three experiments performed is shown.

Close modal

Both RANTES and (AOP)-RANTES caused removal of CCR5 from the cell surface, reaching 60% reduction after 30 min (Fig. 2, A and B). However, (AOP)-RANTES was more effective than RANTES, showing 20% down-regulation at 5 min and 55% at 15 min, whereas only 5% and 15% down-regulation was promoted by RANTES at the same times (Fig. 2, A and B). When experiments were performed at 4°C, a condition that inhibits receptor internalization (20), disappearance of surface CCR5 was minimal following either stimulus (Fig. 2,A). This indicates that the decrease induced in cell-surface CCR5 was due to receptor internalization and not to ligand-mediated inhibition of CCR5-03 binding. This receptor down-modulation has functional consequences; cells pretreated for 15 min with (AOP)-RANTES display minimal calcium mobilization in response to RANTES, whereas cells pretreated with RANTES still mobilize calcium in response to RANTES, although this response is reduced compared with the response when cells are pretreated with SDF-1α (Fig. 2 C).

FIGURE 2.

Down-regulation of CCR5 from the surface of stably transfected HEK 293 cells. A, Serum-starved CCR5-transfected HEK 293 cells were incubated for the times indicated with RANTES (10 nM) or (AOP)-RANTES (10 nM) at 37°C or 4°C. Surface CCR5 was detected by FACS analysis using biotin-labeled CCR5–03 mAb, followed by streptavidin-PE; an isotype-matched mAb was used as control. Results are expressed as the percentage of maximum binding obtained in the absence of chemokines, with SD indicated. B, A representative flow cytometry figure is shown for data from A, corresponding to treatment with RANTES and (AOP)-RANTES (15 min, 37°C). C, Fluo-3AM-loaded cells were treated at 37°C as in A and stimulated with 10 nM RANTES. Calcium mobilization was assessed as in Fig. 1. The result of one of five experiments performed is shown.

FIGURE 2.

Down-regulation of CCR5 from the surface of stably transfected HEK 293 cells. A, Serum-starved CCR5-transfected HEK 293 cells were incubated for the times indicated with RANTES (10 nM) or (AOP)-RANTES (10 nM) at 37°C or 4°C. Surface CCR5 was detected by FACS analysis using biotin-labeled CCR5–03 mAb, followed by streptavidin-PE; an isotype-matched mAb was used as control. Results are expressed as the percentage of maximum binding obtained in the absence of chemokines, with SD indicated. B, A representative flow cytometry figure is shown for data from A, corresponding to treatment with RANTES and (AOP)-RANTES (15 min, 37°C). C, Fluo-3AM-loaded cells were treated at 37°C as in A and stimulated with 10 nM RANTES. Calcium mobilization was assessed as in Fig. 1. The result of one of five experiments performed is shown.

Close modal

We previously demonstrated that the rapid uncoupling of the CCR2 chemokine receptor is mediated by MCP-1-promoted translocation of GRK2 to CCR2 and subsequent receptor phosphorylation (17). Analysis of the CCR5 carboxyl-terminal sequence reveals serine and threonine residues that may be potential GRK phosphorylation sites, as are also found in the CCR2. Therefore, we explored whether RANTES and (AOP)-RANTES treatment of CCR5-transfected HEK 293 cells promotes CCR5 phosphorylation in Ser/Thr residues. Cells treated for different times with either ligand were lysed and extracts immunoprecipitated with the CCR5-specific Ab (CCR5-03). Western blot analysis of immunoprecipitates with a mixture of anti-phosphoserine/threonine Abs showed a phosphorylated protein in the position corresponding to that of CCR5. The time course of phosphorylation differed depending on the ligand used to stimulate the cells. Whereas CCR5 phosphorylation after RANTES stimulation peaks/reaches maximum at 15 min, similar to the times described for MCP-1-activated CCR2 (17), (AOP)-RANTES appears to be more efficient; phosphorylated CCR5 Ser/Thr residues were apparent as soon as 5 min after stimulation (Fig. 3), confirming the ability of (AOP)-RANTES to activate GRK2 more rapidly. Protein loading equivalence was confirmed by reprobing the membrane with specific anti-CCR5 mAb (CCR5-02).

FIGURE 3.

Time course of Ser/Thr phosphorylation of CCR5. CCR5-transfected HEK 293 cells were stimulated at the times indicated, lysed, and cell extracts were immunoprecipitated with an anti-CCR5 mAb. After SDS-PAGE and transfer, the membrane was developed with a mixture of anti-phosphoserine/threonine Abs (upper panel). The Mr of CCR5 is 38 kDa. As a control, a RANTES-stimulated cell lysate immunoprecipitated with anti-β2-microglobulin was analyzed in Western blot using the same anti-phosphoserine/threonine Ab mixture. CCR5 protein loading was controlled by stripping and reprobing membranes with mAb CCR5-02 (lower panel). One representative experiment of four performed is shown.

FIGURE 3.

Time course of Ser/Thr phosphorylation of CCR5. CCR5-transfected HEK 293 cells were stimulated at the times indicated, lysed, and cell extracts were immunoprecipitated with an anti-CCR5 mAb. After SDS-PAGE and transfer, the membrane was developed with a mixture of anti-phosphoserine/threonine Abs (upper panel). The Mr of CCR5 is 38 kDa. As a control, a RANTES-stimulated cell lysate immunoprecipitated with anti-β2-microglobulin was analyzed in Western blot using the same anti-phosphoserine/threonine Ab mixture. CCR5 protein loading was controlled by stripping and reprobing membranes with mAb CCR5-02 (lower panel). One representative experiment of four performed is shown.

Close modal

As we previously demonstrated that GRK2 is involved in Ser/Thr phosphorylation of CCR2 in both Mono Mac 1 and CCR2-transfected HEK 293 cells (17), we tested whether this is also the case for CCR5. As expected, a fraction of cytoplasmic GRK2 associates with the ligand-stimulated receptor in the plasma membrane. When extracts from RANTES- and (AOP)-RANTES-stimulated cells are immunoprecipitated with the CCR5-03 mAb and tested for the presence of GRK2 in Western blot, GRK2 association to CCR5 is observed in a ligand-dependent manner (Fig. 4). The association is highest at 15 min after RANTES induction, but (AOP)-RANTES is more efficient, because GRK2 association is evident at 5 min after activation. Assay specificity was demonstrated by immunoprecipitating the same cell extracts with an unrelated control mAb (anti-β2-microglobulin) (Fig. 4). The GRK2 AB9 Ab shows slight cross-reactivity with GRK3, but this band migrates at a slightly different molecular mass, as previously described (17). Changes in GRK2 association are not a consequence of variation in the amount of CCR5 immunoprecipitated, because Western blot analysis of the same membrane after reprobing with the CCR5-02 mAb confirms equivalent CCR5 protein loading (Fig. 4). These data demonstrate that 1) GRK2 forms a complex with CCR5 as consequence of ligand stimulation and 2) that the (AOP)-RANTES derivative is more efficient than RANTES in triggering this association.

FIGURE 4.

GRK2 association to the CCR5 after RANTES or (AOP)-RANTES activation. Whole-cell lysates from CCR5-transfected HEK 293 cells stimulated with 10 nM RANTES or 10 nM (AOP)-RANTES for the times indicated were precipitated with the anti-CCR5 mAb CCR5-03. Immunoprecipitates were analyzed in Western blot with anti-GRK2 AB9 Ab (upper panel). A CCR5-transfected HEK 293 cell lysate was included as a control of GRK2 migration in the same gel. As negative control, RANTES-stimulated cell lysates were immunoprecipitated with anti-β2-microglobulin mAb. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of five performed is shown.

FIGURE 4.

GRK2 association to the CCR5 after RANTES or (AOP)-RANTES activation. Whole-cell lysates from CCR5-transfected HEK 293 cells stimulated with 10 nM RANTES or 10 nM (AOP)-RANTES for the times indicated were precipitated with the anti-CCR5 mAb CCR5-03. Immunoprecipitates were analyzed in Western blot with anti-GRK2 AB9 Ab (upper panel). A CCR5-transfected HEK 293 cell lysate was included as a control of GRK2 migration in the same gel. As negative control, RANTES-stimulated cell lysates were immunoprecipitated with anti-β2-microglobulin mAb. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of five performed is shown.

Close modal

Desensitization is a process that also involves ligand-induced association of β-arrestin1 to the receptor-GRK2 complex (17). Therefore, we studied RANTES and (AOP)-RANTES-induced association of β-arrestin1 to CCR5. Extracts of CCR5-transfected HEK 293 cells were immunoprecipitated with CCR5-03 mAb and analyzed in Western blot with the anti-arrestin mAb. β-arrestin1 coimmunoprecipitates with both RANTES- and (AOP)-RANTES-stimulated CCR5; this association gives a maximum signal at 30 min after RANTES stimulation, whereas the maximum peak occurs at 15 min after (AOP)-RANTES stimulation (Fig. 5). In both cases, equal protein loading in the gel was assessed by reprobing the membrane with the CCR5-02 mAb. Control experiments with the unrelated anti-β2-microglobulin mAb confirm the specificity of the β-arrestin-CCR5 interaction (Fig. 5).

FIGURE 5.

Association of β-arrestin to CCR5 after RANTES or (AOP)-RANTES activation. Whole-cell lysates from CCR5-transfected HEK 293 cells stimulated with 10 nM RANTES or 10 nM (AOP)-RANTES as in Fig. 3 were immunoprecipitated with CCR5-03 mAb. Immunoprecipitates were analyzed in Western blot with an anti-arrestin mAb. The Mr of β-arrestin is ∼50 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of β-arrestin migration. As a negative control, RANTES-stimulated cell lysates were immunoprecipitated with anti-β2-microglobulin mAb. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of three performed is shown.

FIGURE 5.

Association of β-arrestin to CCR5 after RANTES or (AOP)-RANTES activation. Whole-cell lysates from CCR5-transfected HEK 293 cells stimulated with 10 nM RANTES or 10 nM (AOP)-RANTES as in Fig. 3 were immunoprecipitated with CCR5-03 mAb. Immunoprecipitates were analyzed in Western blot with an anti-arrestin mAb. The Mr of β-arrestin is ∼50 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of β-arrestin migration. As a negative control, RANTES-stimulated cell lysates were immunoprecipitated with anti-β2-microglobulin mAb. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of three performed is shown.

Close modal

RANTES and (AOP)-RANTES induce rapid internalization of CCR5 (11), although as shown above, RANTES is less efficient than (AOP)-RANTES in promoting CCR5 down-regulation (Fig. 2A, B). To examine whether the internalization process occurs via clathrin vesicles, RANTES- or (AOP)-RANTES-stimulated cell extracts were immunoprecipitated with CCR5-03 mAb and tested for clathrin in Western blot. Clathrin associates to CCR5 following stimulation by either ligand; this association is most evident at 30 min after RANTES induction, whereas in (AOP)-RANTES-stimulated cells, clathrin associates to the receptor by 5 min, with a maximum at 15 min (Fig. 6). Assay specificity was demonstrated by immunoprecipitating the same extracts with the unrelated anti-β2-microglobulin mAb used as control; equivalent protein loading was confirmed by stripping and reprobing the membranes with anti-CCR5 mAb (Fig. 6).

FIGURE 6.

Clathrin association to CCR5 after RANTES or (AOP)-RANTES activation. RANTES (10 nM)- or (AOP)-RANTES (10 nM)-induced CCR5-transfected HEK 293 cell lysates were immunoprecipitated with anti-CCR5 mAb, CCR5-03, and developed in Western blot with an anti-clathrin H chain mAb (upper panel). The Mr of clathrin H chain is 180 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of clathrin H chain migration. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of four performed is shown.

FIGURE 6.

Clathrin association to CCR5 after RANTES or (AOP)-RANTES activation. RANTES (10 nM)- or (AOP)-RANTES (10 nM)-induced CCR5-transfected HEK 293 cell lysates were immunoprecipitated with anti-CCR5 mAb, CCR5-03, and developed in Western blot with an anti-clathrin H chain mAb (upper panel). The Mr of clathrin H chain is 180 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of clathrin H chain migration. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of four performed is shown.

Close modal

Endocytosis of many GPCR requires the GTPase activity of dynamin (21). Receptor stimulation leads to recruitment of cytosolic dynamin to coated pits, where it induces constriction of the pits and fission of vesicles (22, 23). We also analyzed whether dynamin associates to CCR5 after RANTES and (AOP)-RANTES activation. CCR5-transfected HEK 293 cells, untreated or treated with RANTES or (AOP)-RANTES, were lysed, and cell extracts immunoprecipitated with CCR5-03 mAb were tested for the presence of dynamin in Western blot. Dynamin associates to CCR5 following stimulation by either ligand. This association is evident earlier for RANTES than for (AOP)-RANTES, but the latter ligand triggers a more prolonged association. Whereas RANTES promotes maximum association at 5 min, decreasing thereafter, maximum association with (AOP)-RANTES occurs at 30 min and is maintained throughout the activation studies (Fig. 7). Thus, we conclude that (AOP)-RANTES triggers more rapid and sustained dynamin association to CCR5. Assay specificity was demonstrated by immunoprecipitating cell extracts from treated cells with the unrelated anti-β2-microglobulin mAb used as control; equivalent protein loading was confirmed by reprobing the membrane with anti-CCR5 mAb (Fig. 7).

FIGURE 7.

Dynamin associates with CCR5 after RANTES or (AOP)-RANTES activation. RANTES (10 nM)- and 10 nM (AOP)-RANTES-induced CCR5-transfected HEK 293 cell lysates were immunoprecipitated with anti-CCR5 mAb, CCR5-03, and developed in Western blot with an anti-dynamin II mAb. The Mr of dynamin II is 100 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of dynamin migration. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of five performed is shown.

FIGURE 7.

Dynamin associates with CCR5 after RANTES or (AOP)-RANTES activation. RANTES (10 nM)- and 10 nM (AOP)-RANTES-induced CCR5-transfected HEK 293 cell lysates were immunoprecipitated with anti-CCR5 mAb, CCR5-03, and developed in Western blot with an anti-dynamin II mAb. The Mr of dynamin II is 100 kDa. A CCR5-transfected HEK 293 cell lysate was included in the gel as a control of dynamin migration. CCR5 protein loading was controlled by reprobing the membrane with CCR5-02 mAb. One representative experiment of five performed is shown.

Close modal

Desensitization and recycling of chemotactic receptors appears to be a critical mechanism by which leukocytes maintain their ability to sense the chemoattractant gradient during an inflammatory response. Internalization of GPCR, the receptor family to which the chemokine receptors belong, is generally believed to require binding of an agonist to the receptor, thus activating the events that initiate receptor endocytosis. Briefly, this pathway is initiated by phosphorylation by GRK on Ser/Thr residues in the intracellular carboxyl-terminal domain of GPCR; this triggers recruitment of arrestin proteins, causing sequestration of the phosphorylated receptors into clathrin-coated pits. Previous studies demonstrated that chemotactic receptors of the CXC (24) and CC (25) family were phosphorylated in serine residues by GRK and that GRK2 and β-arrestin1 bind the MCP-1-activated CCR2, initiating a process that leads to CCR2 desensitization (17). (AOP)-RANTES, a RANTES analogue obtained by chemical coupling of a pentacarbo alkyl chain to oxidized N-terminal serines (10), is known to promote CCR5-mediated calcium mobilization, although it lacks the ability to trigger chemotaxis of human monocytes (10, 11). A more detailed study demonstrated that this analogue triggers CCR5 dimerization, JAK/STAT pathway activation, CCR5 tyrosine phosphorylation, and activation of the Gi pathway, as does RANTES, although with some differences (12). (AOP)-RANTES is more efficient than RANTES in triggering Gαi association to and dissociation from the CCR5. This gives rise to differences in the pool of free, active βγ subunits, subsequently affecting long-term ligand-induced responses such as chemotaxis, internalization, and recycling.

To understand the mechanism by which these two ligands trigger different responses, we used anti-CCR5-specific mAb to analyze CCR5 desensitization and internalization following RANTES or (AOP)-RANTES stimulation in CCR5-transfected HEK 293 cells. As has been shown for freshly isolated primary cells and CCR5-transfected Chinese hamster ovary cells (11), both ligands induce receptor down-regulation in transfected HEK 293 cells, although (AOP)-RANTES is more efficient in promoting faster disappearance of CCR5 from the cell surface. These differences have functional implications, because RANTES- and AOP-RANTES-induced loss of CCR5 surface expression correlates with a reduced RANTES-induced calcium signal at this time point. This observation concurs with previously described differences in the ability of other CCR5 ligands (macrophage inflammatory protein (MIP)-1α, MIP-1β, Met-RANTES) to promote calcium flux, receptor phosphorylation, and receptor down-modulation (11, 26).

We next analyzed the mechanism implicated in CCR5 down-regulation and show that following stimulation both RANTES and (AOP)-RANTES trigger the formation of a macromolecular complex between CCR5, GRK2, the regulatory protein β-arrestin, and clathrin. These results extend and confirm earlier data for the MCP-1-stimulated CCR2 (17) and assign a clear role to the complex associated to the chemokine receptor during the internalization process.

Supporting these results, recent reports also show that GRK2 and arrestin colocalize with another GPCR, the β-adrenergic receptor, in internalization vesicles (27, 28), and GRK3 has been associated to ligand-promoted internalization of CCR5 in transfected rat basophilic leukemia-2H3 cells (26). Our data also lead to the conclusion that the improved capacity of (AOP)-RANTES to promote CCR5 internalization is a consequence of improved ability to trigger complex formation, that is, GRK2 association, phosphorylation in CCR5 Ser/Thr residues, and association of β-arrestin1 and clathrin. It has recently been shown that the CCR5 C-terminal serine residues are phosphorylated following agonist stimulation (26). In addition, Gβγ acts as a docking complex, providing an interface for the GPCR, which would facilitate GPCR interaction with diverse signaling pathways (29). This is the case for GRK coupling to other receptors (29), in which βγ interacts with the third intracellular loop of the M2-M3 muscarinic receptors. The association of βγ with the activated GPCR allows formation of a ternary complex with GRK2, which is required for effective receptor Ser/Thr phosphorylation, necessary in turn for β-arrestin recruitment. Our previous observations showed that (AOP)-RANTES is more efficient than RANTES in promoting both Gαi association to and dissociation from CCR5 (12). We now link this to the differential efficiency of these molecules to activate the Gi pathway, thus making active βγ subunits available, which recruit activated GRK2. Given that Gαi association to CCR5 is a consequence of ligand-induced receptor activation, it appears obvious that receptor activation, probably as a result of conformational changes, may differ following RANTES or (AOP)-RANTES binding. This concurs with the report of affinity differences between RANTES and (AOP)-RANTES in their interaction with CCR5, reflected by differences in their binding properties (10), because RANTES shows biphasic binding to the receptor whereas (AOP)-RANTES has a single phase competition curve.

All together, our data explain the increased efficiency of (AOP)-RANTES in promoting CCR5 down-regulation. However, the enhanced ability of (AOP)-RANTES to trigger GRK2 translocation and β-arrestin-1/clathrin association to the CCR5 does not explain the capacity of (AOP)-RANTES to inhibit normal CCR5 recycling, which does not occur when receptors are activated by RANTES (11). Although endocytosis of chemokine receptors has not been studied in detail, there is evidence that these receptors can either recycle or enter a degradation pathway via the lysosomal compartment; in fact, CXCR2 is reported to undergo degradation after internalization (30). It is believed that receptor recycling from endocytic organelles requires ligand dissociation from the receptor and subsequent dephosphorylation. Here again, we show clear evidence that explains the enhanced efficiency of (AOP)-RANTES-induced internalization, because clathrin association to CCR5 persists longer after (AOP)-RANTES treatment. Nonetheless, this is insufficient to account for the absence of recycling after (AOP)-RANTES stimulation.

Synaptic vesicle recycling and endocytosis of many receptors, including GPCR, require the GTPase activity of dynamin (31, 32); in fact, coexpression of CXCR2 and a dominant negative mutant of dynamin inhibits receptor internalization (33). After ligand activation, and as a consequence of Src-mediated tyrosine phosphorylation (21), dynamin is recruited to coated pits, where it binds to the appendage domain of α-adaptin, a component of the clathrin-coated pits (34). In endocytosis, dynamin catalyzes a GTP-dependent pinching-off of endocytic vesicles from the plasma membrane (31). Dynamin is depleted from coated vesicles relative to coated pits and there is a pool of available cytosolic dynamin; it is thus assumed that GDP-bound dynamin is disassembled and recycled via the cytosol for repeated rounds of vesicle budding (32). This process also takes place in RANTES- and (AOP)-RANTES-mediated CCR5 internalization. Nonetheless, whereas RANTES promotes transient association between the dynamin and CCR5 in the clathrin vesicles, (AOP)-RANTES promotes sustained association between dynamin, clathrin, and the receptor. The outcome of the differential dynamin association to CCR5 is that normal CCR5 recycling is altered when cells are stimulated with (AOP)-RANTES. This suggests that dynamin association to clathrin is a critical step in normal CCR5 recycling. Furthermore, the loss of receptor recycling alters receptor expression and therefore the ability of cells to respond to a continuous signal generated over a concentration gradient (33). Thus, we conclude that distinct ligands are able to trigger the differential association of signaling complexes that further modulate biological responses, and we identify the molecular complexes that control these processes.

We thank Drs. Federico Mayor, Jr., Ana Aragay, Robin Offord, Amanda E. I. Proudfoot, and Tim Wells for gifts of reagents and helpful discussion of this work. We also thank M. C. Moreno and I. López for help with flow cytometry and C. Bastos and C. Mark for secretarial and editorial assistance, respectively.

1

This work was partially supported by grants from the Spanish Comision Interministerial de Ciencia y Tecnologia, the European Union, and the Comunidad de Madrid. The Department of Immunology and Oncology was founded and is supported by the Spanish Research Council (Consejo Superior de Investigaciones Cientificas) and Pharmacia and Upjohn.

3

Abbreviations used in this paper: AOP, aminooxypentane; JAK, Janus activated kinase; GPCR, G protein-coupled receptor; GRK, GPCR kinase; MCP, monocyte chemoattractant protein; SDF, stromal cell-derived factor; MIP, macrophage inflammatory protein; HEK, human embryonic kidney.

1
Schall, T. J., K. B. Bacon.
1994
. Chemokines, leukocyte trafficking and inflammation.
Curr. Opin. Immunol.
6
:
865
2
Schall, T. J..
1994
. The chemokines.
The Cytokine Handbook
2nd Ed.
181
Academic Press, London.
3
Baggiolini, M..
1998
. Chemokines and leukocyte traffic.
Nature
392
:
565
4
Rollins, B..
1997
. Chemokines.
Blood
90
:
909
5
Ugoccioni, M., M. D’Apuzzo, M. Loetscher, B. Dewald, M. Baggiolini.
1995
. Actions of the chemotactic cytokines MCP-1, MCP-2, MCP-3, RANTES, MIP-1α and MIP-1β on human monocytes.
Eur. J. Immunol.
25
:
64
6
Schall, T. J., K. Bacon, K. J. Toy, D. V. Goeddel.
1990
. Selective attraction of monocytes and T lymphocytes of the memory phenotype by cytokine RANTES.
Nature
347
:
669
7
Taub, D. D., T. J. Sayers, C. R. Carter, J. R. Ortaldo.
1995
. α and β chemokines induce NK cell migration and enhance NK-mediated cytolysis.
J. Immunol.
155
:
3877
8
Alkhatib, G., C. Combadiere, C. C. Broder, Y. Feng, P. E. Kennedy, P. M. Murphy, E. A. Berger.
1996
. CC CKR5: A RANTES, MIP-1α, MIP-1β receptor as a fusion cofactor for macrophage-tropic HIV-1.
Science
272
:
1955
9
Cocchi, F., A. L. DeVico, A. Garzino-Demo, S. K. Arya, R. C. Gallo, P. Lusso.
1995
. Identification of RANTES, MIP-1α, and MIP-1β as the major HIV-suppressive factors produced by CD8+ T cells.
Science
270
:
1811
10
Simmons, G., P. R. Clapham, L. Picard, R. E. Offord, M. M. Rosenkilde, T. W. Schwartz, R. Buse, T. N. C. Wells, A. E. I. Proudfoot.
1997
. Potent inhibition of HIV-1 infectivity in macrophages and lymphocytes by a novel CCR5 antagonist.
Science
276
:
276
11
Mack, M., B. Lucknow, P. J. Nelson, J. Cihak, G. Simmons, P. R. Clapham, N. Signoret, M. Marsh, M. Stangassinger, F. Borlat, T. N. C. Wells, D. Schlöndorff, A. E. I. Proudfoot.
1998
. AOP-RANTES induces CCR5 internalization but inhibits recycling: a novel inhibitory mechanism of HIV infectivity.
J. Exp. Med.
187
:
1215
12
Rodriguez-Frade, J. M., A. Vila-Coro, A. Martin de Ana, M. Nieto, F. Sanchez-Madrid, A. E. I. Proudfoot, T. N. C. Wells, C. Martínez-A., M. Mellado.
1999
. Similarities and differences in RANTES- and (AOP)-RANTES-triggered signals: implications for chemotaxis.
J. Cell Biol.
144
:
755
13
Rodriguez-Frade, J. M., A. J. Vila-Coro, A. Martín de Ana, J. P. Albar, C. Martínez-A., M. Mellado.
1999
. The chemokine monocyte chemoattractant protein-1 induces functional responses through dimerization of its receptor CCR2.
Proc. Natl. Acad. Sci. USA
96
:
3628
14
Neptune, E. R., T. Iiri, H. R. Bourne.
1999
. Gαi is not required for chemotaxis mediated Gi-coupled receptors.
J. Biol. Chem.
274
:
2824
15
Lefkowitz, R. J..
1993
. G protein-coupled receptor kinases.
Cell
74
:
409
16
Premont, R. T., J. Inglese, R. J. Lefkowitz.
1995
. Protein kinases that phosphorylate activated G protein-coupled receptors.
FASEB J.
9
:
175
17
Aragay, A. M., M. Mellado, J. M. Rodriguez-Frade, A. M. Martín, C. Martínez-A., F. Mayor, Jr.
1998
. Monocyte chemoattractant protein-1-induced CCR2B receptor desensitization mediated by the G protein-coupled receptor kinase-2.
Proc. Natl. Acad. Sci USA
95
:
2985
18
Mellado, M., J. M. Rodriguez-Frade, L. Kremer, C. von Kobbe, A. Martín de Ana, I. Mérida, C. Martínez-A.
1997
. Conformational changes required in the human growth hormone receptor for growth hormone signaling.
J. Biol. Chem.
272
:
9189
19
Mellado, M., J. M. Rodríguez-Frade, A. M. Aragay, G. del Real, A. Vila-Coro, A. Martin de Ana, A. Serrano, F. Mayor, Jr, C. Martínez-A.
1998
. The chemokine MCP-1 triggers tyrosine phosphorylation of the CCR2B receptor and the JAK2/STAT3 pathway.
J. Immunol.
161
:
805
20
Roettger, B. F., R. U. Rentsch, D. Pinon, E. Holicky, E. Hadac, J. M. Larkin, L. J. Miller.
1995
. Dual pathways of internalization of the cholecystokinin receptor.
J. Cell Biol.
128
:
1029
21
Seungkirl, A., S. Maudsley, L. M. Luttrell, R. J. Lefkowitz, Y. Daaka.
1999
. Src-mediated tyrosine phosphorylation of dynamin is required for β2-adrenergic receptor internalization and mitogen-activated protein kinase signaling.
J. Biol. Chem.
274
:
1185
22
McClure, S. J., P. J. Robinson.
1996
. Dynamin, endocytosis and intracellular signalling.
Mol. Membr. Biol.
13
:
189
23
Urrutia, R., J. R. Henley, T. Cook, M. A. McNiven.
1997
. The dynamins redundant or distinct functions for an expanding family of related GTPases?.
Proc. Natl. Acad. Sci. USA
94
:
377
24
Haribabu, B., R. M. Richardson, Y. Fisher, S. Sozzani, S. C. Peiper, R. Horuk, H. Ali, R. Snyderman.
1997
. Regulation of human chemokine receptorsCXCR4: role of phosphorylation in desensitization and internalization.
J. Biol. Chem.
272
:
28726
25
Franci, C., J. Gosling, C-L. Tsou, S. R. Coughlin, I. Charo.
1996
. Phosphorylation by a G protein-coupled kinase inhibits signaling and promotes internalization of the monocyte chemoattractant protein-1 receptor.
J. Immunol.
157
:
5606
26
Oppermann, M., M. Mack, A. E. I. Proudfoot, H. Olbrich.
1999
. Differential effects of CC chemokines on CC chemokine receptor 5 (CCR5) phosphorylation and identification of phosphorylation sites on the CCR5 carboxyl terminus.
J. Biol. Chem.
274
:
8875
27
Ruiz-Gómez, A., F. Mayor, Jr.
1997
. β-adrenergic receptor kinase (GRK2) colocalizes with β-adrenergic receptors during agonist-induced receptor internalization.
J. Biol. Chem.
272
:
9601
28
García-Higuera, I., F. Mayor, Jr.
1994
. Rapid desensitization of neonatal rat liver β-adrenergic receptors: a role for β-adrenergic receptor kinase.
J. Clin. Invest.
93
:
937
29
Wu, G., J. L. Benovic, J. D. Hildebrandt, S. M. Lanier.
1998
. Receptor docking sites for G-protein βγ subunits: implications for signal regulation.
J. Biol. Chem.
273
:
7197
30
Mueller, S.G., W. P. Schaw, A. Richmond.
1995
. Activation of protein kinase C enhances the phosphorylation of the type B interleukin-8 receptor and stimulates its degradation in non-hematopoietic cells.
J. Biol. Chem.
270
:
10439
31
Zhang, J., S. S. G. Ferguson, L. S. Barak, L. Ménard, M. G. Caron.
1996
. Dynamin and β-arrestin reveal distinct mechanisms for G protein-coupled receptor internalization.
J. Biol. Chem.
271
:
18302
32
Schmid, S. L., M. A. McNiven, P. De Camilli.
1998
. Dynamin and its partners: a progress report.
Curr. Opin. Cell Biol.
10
:
504
33
Yang, W., D. Wang, A. Richmond.
1999
. Role of clathrin-mediated endocytosis in CXCR2 sequestration, resensitization and signal transduction.
J. Biol. Chem.
274
:
11328
34
Wang, L.H., T. C. Südhof, R. G. Anderson.
1995
. The appendage domain of α-adaptin is a high affinity binding site for dynamin.
J. Biol. Chem.
270
:
10079