HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Zaslaver, A. Articles by Ben-Baruch, A. Search for Related Content PubMed PubMed Citation Articles by Zaslaver, A. Articles by Ben-Baruch, A.

Actin Filaments Are Involved in the Regulation of Trafficking of Two Closely Related Chemokine Receptors, CXCR1 and CXCR2

Department of Cell Research and Immunology, George S. Wise Faculty of Life Sciences, Tel-Aviv University, Tel-Aviv, Israel

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The ligand-induced internalization and recycling of chemokine receptors play a significant role in their regulation. In this study, we analyzed the involvement of actin filaments and of microtubules in the control of ligand-induced internalization and recycling of CXC chemokine receptor (CXCR)1 and CXCR2, two closely related G protein-coupled receptors that mediate ELR-expressing CXC chemokine-induced cellular responses. Nocodazole, a microtubule-disrupting agent, did not affect the IL-8-induced reduction in cell surface expression of CXCR1 and CXCR2, nor did it affect the recycling of these receptors following ligand removal and cell recovery at 37°C. In contrast, cytochalasin D, an actin filament depolymerizing agent, promoted the IL-8-induced reduction in cell surface expression of both CXCR1 and CXCR2. Cytochalasin D significantly inhibited the recycling of both CXCR1 and CXCR2 following IL-8-induced internalization, the inhibition being more pronounced for CXCR2 than for CXCR1. Potent inhibition of recycling was observed also when internalization of CXCR2 was induced by another ELR-expressing CXC chemokine, granulocyte chemotactic protein-2. By the use of carboxyl terminus-truncated CXCR1 and CXCR2 it was observed that the carboxyl terminus domains of CXCR1 and CXCR2 were partially involved in the regulation of the actin-mediated process of receptor recycling. The cytochalasin D-mediated inhibition of CXCR2 recycling had a functional relevance because it impaired the ability of CXCR2-expressing cells to mediate cellular responses. These results suggest that actin filaments, but not microtubules, are involved in the regulation of the intracellular trafficking of CXCR1 and CXCR2, and that actin filaments may be required to enable cellular resensitization following a desensitized refractory period.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Chemokines are small chemotactic cytokines that mediate inflammatory processes, hematopoiesis and angiogenesis, and play a crucial role in AIDS pathogenesis (1, 2, 3, 4). Chemokine-mediated responses are induced by the activation of specific G protein-coupled receptors (GPCR),² which are expressed on target cells (5). In similarity to other members of the GPCR superfamily, the ability of chemokine receptors to transmit intracellular signals may be rapidly attenuated, and is tightly regulated by different mechanisms, including receptor phosphorylation-dependent G protein-uncoupling and receptor internalization (6, 7, 8, 9). Agonist-induced internalization of GPCR depletes the plasma membrane of receptors for the agonist and, therefore, may contribute to the desensitization of the functions mediated through these receptors (6, 7, 8, 9). Moreover, if the ligand that induced receptor internalization is removed, the receptors may be dephosphorylated, recycled back to the plasma membrane and resensitized upon a subsequent exposure to the same ligand (9, 10, 11).

Chemokine-induced receptor internalization and recycling were shown to play a significant role in the regulation of inflammatory processes, as indicated for example by studies on neutrophils, which consist of the major cellular infiltrate in the course of acute inflammation. Several investigations demonstrated that neutrophil responses could be regulated and desensitized in processes involving chemokine receptor internalization (12, 13, 14). Chemokine receptor internalization and recycling were also shown to be involved in the regulation of HIV-1 infectivity, as illustrated by the ability of the RANTES antagonist aminooxypentane-RANTES to induce a highly potent inhibition of HIV-1 entry to target cells. The aminooxypentane-RANTES-mediated inhibition was demonstrated to be a direct result of its ability to induce elevated levels of CCR5 internalization, and to prevent its recycling back to the plasma membrane, thereby removing a key element of the fusion complex (15).

Recent interest in the mechanisms regulating ligand-induced GPCR internalization has led to observations suggesting that some GPCR undergo endocytosis via clathrin-coated pits, in a process that involves the binding of arrestins to the receptors (8, 9, 16, 17). Other factors that are potential regulators of the internalization and recycling processes of GPCR are yet to be defined. In that regard, elucidation of the involvement of actin filaments and/or microtubules in ligand-induced GPCR intracellular trafficking is of major importance. The plasma membrane is functionally integrated with the cell "cortex" that consists of actin-based cytoskeleton (18). Therefore, the trafficking of vesicles at the plasma membrane may necessitate the active rearrangement of actin filaments, which may then be followed by actin-assisted vesicle budding and fusion at the plasma membrane. Moreover, microtubules may also play a direct role in trafficking of receptors, and were shown to facilitate transport along the trans-golgi network plasma membrane pathway (19, 20, 21, 22, 23).

Although of potentially major importance for the regulation of ligand-induced receptor intracellular trafficking, the involvement of cytoskeleton elements in GPCR trafficking was very minimally studied so far. Of high significance is the fact that the contribution of such elements to the regulation of trafficking of chemokine receptors was not addressed in any manner. In that respect, our study is focused on the elucidation of the role of cytoskeleton elements in the ligand-induced internalization and recycling processes of CXC chemokine receptor (CXCR)1 and CXCR2, two closely related receptors that mediate the migration of neutrophils to inflammatory sites in response to ELR-expressing CXC (ELR⁺-CXC) chemokines (1, 5, 24). The use of CXCR1 and CXCR2 allowed us to compare between two chemokine receptors that have similar general characteristics, but nevertheless are differentially regulated at multiple levels, including their internalization properties (25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37).

The similarity between CXCR1 and CXCR2 is illustrated by the fact that both receptors bind with high affinity the ELR⁺-CXC chemokine IL-8, and mediate potent cellular migration in response to this chemokine. However, the two receptors diverge in their ability to bind other members of the ELR⁺-CXC subfamily of chemokines, in the intracellular signals they transduce and in the mechanisms that regulate their desensitization and internalization following the binding of high concentrations of ligands (25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37). Based on their ability to undergo internalization and recycling, it was suggested that CXCR1 is involved in mediating IL-8-induced chemotaxis at the site of inflammation, where the concentration of IL-8 is high, whereas CXCR2 has an active role in the initiation phase of neutrophil migration, distant from the site of inflammation, where the concentration of IL-8 is at the picomolar level (33). In search for further elucidation of the mechanisms regulating CXCR1 and CXCR2 internalization, our previous studies demonstrated that potent CXCR1 internalization is induced by IL-8, but not by other ELR⁺-CXC chemokines, such as granulocyte chemotactic protein-2 (GCP-2) and neutrophil-activating protein-2. In contrast, CXCR2 undergoes potent internalization in response to both IL-8 and GCP-2, but not to neutrophil-activating protein-2 (31, 32).

In the present study we investigated by pharmacological means the involvement of microtubules and of actin filaments in the regulation of the cell surface expression of CXCR1 and CXCR2, following ELR⁺-CXC chemokine-induced internalization of these receptors. Our study provides the first evidence for the involvement of actin filaments, but not microtubules, in the trafficking of chemokine receptors. Functionally intact actin filaments were demonstrated to be required for potent recycling processes of both CXCR1 and CXCR2. The critical role of actin filaments in the regulation of receptor recycling was further substantiated by functional assays, suggesting that following a desensitized refractory period, the resensitization of CXCR1 and CXCR2-expressing cells requires functionally intact actin filaments that will enable potent receptor re-expression, as well as migratory responses to occur.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
DNAs for human CXCR1 and human CXCR2

Wild-type (WT) CXCR1, 19 CXCR1, WT CXCR2 and 20 CXCR2 DNAs were generated using PCR, shuttled into the expression vector pRc/CMV (Invitrogen, San Diego, CA), and subjected to full-length sequencing as previously described (31, 38). The primary amino acid sequences of the predicated carboxyl terminus domains of the receptors are shown in Table II.

Human embryonal kidney 293 (HEK 293) cells (kindly donated by Dr. P. Gray, ICOS, Bothell, WA) were grown and stably transfected as previously described (31, 38). FACS analyses showed that a high percentage (over 85%) of the transfected cells (WT CXCR1, 19 CXCR1, WT CXCR2, 20 CXCR2) expressed the receptor on the cell surface. All the transfected cells (WT CXCR1, 19 CXCR1, WT CXCR2, 20 CXCR2) bound IL-8 with high affinity (31, 38). Control transfections were performed with the vector (pRc/CMV) alone, and the resulting cells did not specifically bind IL-8, GCP-2, or Abs specific for human CXCR1 or CXCR2 (29, 31, 38, 39).

Parental and stable CXCR1-expressing 300-19 pre B cells (kindly donated by Dr. M. Wolf, Theodor Kocher Institute, University of Bern, Bern, Switzerland) were grown in RPMI 1640 medium, supplemented with 10% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, 250 ng/ml amphotericin, 2 mM L-glutamine, 1x non essential amino acids (all purchased from Biological Industries, Beit Haemek, Israel), and 0.05% mM 2-ME (Sigma, Rehovot, Israel). The DNA for CXCR1 transfection was expressed in a SR--puro vector. Puromycin (1 µg/ml; Sigma) was used for selection. No expression of human CXCR1 or CXCR2 was detected on the parental 300-19 cells, as determined by monoclonal mouse anti-CXCR1 or anti-CXCR2 Abs (R&D Systems, Minneapolis, MN).

Preparation of human neutrophils

The preparation of human neutrophils was performed as previously described (40). Briefly, neutrophils (94% purity) were separated by Ficoll-Hypaque density gradient centrifugation followed by dextran sedimentation and hypotonic lysis of remaining erythrocytes. Because NH₄Cl, a lysosomotropic agent, may inhibit receptor recycling on human neutrophils (33), the neutrophils used in these experiments were prepared without NH₄Cl.

Determination of receptor down-modulation by FACS analysis

Stable receptor-expressing HEK 293 cells were split, and a day later were taken for the experiments. All the experimental steps were performed while the cells were in suspension. The analysis of receptor down-modulation was performed as previously described (31). Briefly, aliquots of stable HEK 293 transfectants were removed and supplemented with 1000 ng/ml IL-8 (Dainippon, Japan) diluted in BSA medium (RPMI 1640 medium containing 1% BSA and 25 mM HEPES), while no IL-8 was added to control tubes. The cells were incubated at 37°C for 60 or 90 min (see Results), washed in cell sorter buffer (CSB: PBS containing 1% FCS, 0.02% NaN₃, and 25 mM HEPES) and incubated at 4°C with monoclonal mouse anti-CXCR1 or anti-CXCR2 Abs (R&D Systems). Baseline staining was obtained by adding CSB to the cells instead of anti-CXCR1 or anti-CXCR2 Abs. Following incubation and washings, the cells were incubated with FITC-conjugated goat anti-mouse IgG Abs (Jackson ImmunoResearch Laboratories, West Grove, PA), washed, and resuspended. The effects of cytoskeleton-disrupting drugs were determined by preincubating the cells with 10 µM nocodazole or 1 µM cytochalasin D (Sigma) for 30 min at 37°C, as well as in the course of exposure to IL-8. Because both reagents were resuspended in DMSO and used in a 1:1000 final dilution, a control of 1:1000 dilution of DMSO was also included. Incubations with 100 ng/ml and 500 ng/ml pertussis toxin (PTx) were performed by incubating the cells with the drug for 120 min at 37°C, followed by removal of the drug. In preliminary experiments, it was observed that nocodazole, cytochalasin D, and PTx did not have a notable effect on the basal expression of CXCR1 and CXCR2. Analysis of the cells following nocodazole, cytochalasin D, and PTx treatments indicated that cell viability was not affected by these compounds. FACSort (Becton Dickinson, San Jose, CA), was used to analyze 5000 live cell events. Percent reduction in cell surface expression was calculated from the mean channel fluorescence values of cells treated with ligand at 37°C vs the mean channel fluorescence of cells not treated with ligand, under similar conditions. Values of p were calculated by the Student t test.

Determination of receptor re-expression on the plasma membrane

The procedure used to determine receptor re-expression in HEK 293 cells was similar to that used to evaluate receptor down-modulation, only that matched samples were allowed to undergo a receptor recovery process that was performed as previously described (31). Briefly, internalization was induced by a 60- or 90-min exposure (see Results) to 1000 ng/ml IL-8, or by a 2-h exposure to 1000 ng/ml GCP-2. To allow receptor recovery, the stable receptor transfectants were washed once, resuspended in BSA medium, and incubated for 90 min at 37°C in BSA medium. Following the incubation, the cells were washed and labeled at 4°C with anti-CXCR1 or anti-CXCR2 Abs as described above. The effects of cytoskeleton-disrupting drugs were determined by incubating the cells at the time of cell recovery (but not of induction of internalization) with 10 µM nocodazole or 1 µM cytochalasin D. Because both reagents were resuspended in DMSO and used in a 1:1000 final dilution, a control of 1:1000 dilution of DMSO was also included. Analysis of the cells following nocodazole or cytochalasin D treatments indicated that cell viability was not affected by these compounds. Similar analysis of receptor re-expression in neutrophils was performed by a 10-min exposure to 1000 ng/ml IL-8 at 37°C, followed by removal of the ligand and recovery of the cells for 40 min at 37°C. FACSort was used to analyze 5000 live cell events. Mean fluorescence values of cells that were not exposed to IL-8, and of cells exposed to IL-8 and allowed to undergo recovery, were used to determine the level of receptor re-expression. The percent of inhibition of receptor recovery was calculated for each experiment as follows: the percent inhibition = 1 - (the percent of total expression in cells that underwent recovery in the presence of cytochalasin D - the percent of total expression in cells that underwent internalization)/(the percent of total expression in cells that underwent recovery in the presence of DMSO - the percent of total expression in cells that underwent internalization) x 100. Values of p were calculated by the Student t test.

Confocal analyses of receptor down-modulation

Stable CXCR2-expressing cells were split, and a day later, were taken for the experiments. All the experimental steps were performed while the cells were in suspension. Aliquots of cells were either incubated in the presence, or in the absence of 1000 ng/ml IL-8 for 90 min at 37°C. The cells that were exposed to IL-8 were divided into two subgroups before exposure to IL-8: the cells of one group were exposed to 1 µM cytochalasin D for 30 min at 37°C before, and in the course of incubation with IL-8, while the cells of the other group were not exposed to cytochalasin D. All the procedures from this stage forward were performed at room temperature. The cells were rinsed in PBS, fixed with 4% paraformaldehyde for 15 min and centrifuged onto 0.5% gelatin-coated slides. Permeabilized cells were obtained by incubating the cells with 0.2% Triton X-100 (Sigma) for 30 min, followed by blocking with 3% goat serum (Biological Industries), 0.25% gelatin (Sigma), and 0.15% saponin (Sigma) in PBS, for 60 min. After rinsing the cells in washing buffer (0.25% gelatin and 0.15% saponin in PBS), the cells were stained with polyclonal rabbit anti human CXCR2 Abs (0.5 µg/60 µl; Santa Cruz Biotechnology, Santa Cruz, CA) for 90 min. Following additional rinsing in washing buffer, the cells were stained with Rhodamine-conjugated goat anti-rabbit IgG (3.5 µg/60 µl, Jackson ImmunoResearch Laboratories) and Alexa Fluor 488 phalloidin (Molecular Probes, Eugene, OR) for 30 min. Following additional washings, stained cells were analyzed using a Zeiss confocal laser scanning microscope (Oberkochen, Germany). Zeiss LSM 410 invert was equipped with a 25 mW krypton-argon laser (488 and 568 maximum lines) and 10 mW HeNe laser (633 maximum line). A 40x NA/1.2 C-apochromat water-immersion lens (Axiovert 135 M, Zeiss) was used for all imaging.

Chemotaxis assays

The migration of HEK 293 cells was assessed by a 48-well microchemotaxis chamber technique as previously described (38). Briefly, the lower compartment of the chamber was loaded with aliquots of BSA medium, 100 ng/ml IL-8 (for CXCR1), or 50 ng/ml IL-8 (for CXCR2) diluted in BSA medium, while the upper compartment of the chamber was loaded with cells (resuspended in BSA medium). The two compartments were separated by a 10-µm pore-sized polycarbonate polyvinylpyrrolidone coated with 50 µg/ml rat collagen type I (Collaborative Biomedical Products, Bedford, MA). Following 5–6 h of incubation at 37°C the filter was removed, fixed and stained. The cells that were studied included: 1) cells that were not treated at all; 2) cells in which receptor down-modulation was induced by exposure to 1000 ng/ml IL-8; 3) cells in which receptor down-modulation was induced by exposure to 1000 ng/ml IL-8, and recovery was allowed (after removal of the ligand) for 90 min at 37°C in the absence of cytochalasin D; 4) cells in which receptor down-modulation was induced by exposure to 1000 ng/ml IL-8, and recovery was allowed (after removal of the ligand) for 90 min at 37°C in the presence of 1 µM cytochalasin D; and 5) cells that were incubated only with 1 µM cytochalasin D for 90 min at 37°C. All the cells were washed before loading into the chemotaxis chamber. In separate experiments, the effects of 100 ng/ml and 500 ng/ml PTx on migration were analyzed by the preincubation of the cells with the drug for 120 min at 37°C, followed by the removal of the drug, and loading of the cells in the chemotaxis chamber. Analysis of the cells following cytochalasin D and PTx treatments indicated that cell viability was not affected by the compounds. The baseline migration to BSA medium of the different types of cells was similar, and was subtracted from the response to IL-8. Values of p were calculated by the Student t test.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
The effects of nocodazole and cytochalasin D on IL-8-induced down-modulation of CXCR1 and CXCR2

To determine the effect of cytoskeleton-disrupting agents on the cell surface expression of CXCR1 and CXCR2 in the course of ligand-induced receptor internalization, we used CXCR1- or CXCR2-expressing HEK 293 cells. The analysis of the regulation of CXCR1 and CXCR2 in neutrophils may be impeded by the fact that both receptors are expressed together on these cells (27, 28, 33, 34, 39), providing a possibility for the existence of intracellular interactions and cross-talk between the two receptors. Unlike neutrophils, the cells of our system allow us to dissociate the mechanisms regulating each of the receptors, and to elucidate by using receptor mutants the mechanisms that are involved in the regulation of receptor trafficking by cytoskeleton elements. This well-characterized system was shown to resemble neutrophils in terms of ligand binding affinity, activation, homologous desensitization, certain characteristics of internalization of CXCR1 and CXCR2 and induction of migratory responses (29, 30, 31, 32, 38, 41, 42, 43, 44). In similarity to neutrophils, CXCR1- and CXCR2-expressing HEK 293 cells were shown to be activated in response to low concentrations of chemokines (10–100 ng/ml), and to undergo desensitization by exposure to high concentrations of chemokines (500–1000 ng/ml) (29, 30, 31, 32, 38, 42, 43). Furthermore, our previous studies have shown by FACS analysis that the exposure of CXCR1- and CXCR2-expressing HEK 293 cells to high concentrations of chemokines (CXCR1, 1000 ng/ml IL-8; CXCR2, 1000 ng/ml IL-8, 1000 ng/ml GCP-2), resulted in a prominent reduction in receptor cell surface expression (31, 32). Similar experiments that were performed at 4°C indicated that the preincubation of the cells with chemokines did not prevent the Abs from binding to cell surface-expressed CXCR1 and CXCR2. Confocal analyses performed under similar conditions indicated that potent chemokine-induced receptor internalization gave rise to the observed reduction in cell surface expression of both receptors (Ref. 32 and data not shown).

The analysis of the effects of nocodazole or cytochalasin D was performed by incubating WT CXCR1- or WT CXCR2-expressing cells with the drugs for 30 min at 37°C, followed by exposure to IL-8 for 90 min at 37°C. A highly notable IL-8-induced reduction in the cell surface expression of CXCR1 and CXCR2 was observed in the absence of nocodazole or of cytochalasin D (Fig. 1, A and B). The extent of IL-8-induced CXCR1 or CXCR2 down-modulation following the treatment with the drugs was compared with the extent of reduction in cell surface expression in their absence, or in the presence of DMSO, which was the solubilizer of the drugs and did not affect receptor down-modulation on its own (Fig. 1, A and B). In preliminary experiments it was observed that nocodazole and cytochalasin D did not have a notable effect on the basal expression of CXCR1 and CXCR2.

View larger version (26K): [in this window] [in a new window]   FIGURE 1. The effects of nocodazole and cytochalasin D on WT CXCR1 and WT CXCR2 down-modulation. HEK 293 cells were exposed to nocodazole or cytochalasin D as described in Materials and Methods. Matched samples were incubated in the same conditions without exposure to the drugs, or in the presence of DMSO. Following these treatments, the cells were incubated with IL-8. Another group of cells was not exposed to any treatment. The cells were subjected to FACS analysis as described in Materials and Methods. The data are presented as the percent of reduction in cell surface expression as compared with cells that were not exposed to any treatment, calculated as described in Materials and Methods. A, CXCR1-expressing cells. **, p < 0.002 for treatment with cytochalasin D vs DMSO treatment. B, CXCR2-expressing cells. *, p = 0.026 for treatment with cytochalasin D vs DMSO treatment. Cyto-D, Cytochalasin D; Nocod., nocodazole. Each value represents the mean ± SD of three independent experiments.

Analysis of the effects of cytochalasin D on reduction in the cell surface expression of CXCR1 or CXCR2, induced by incubation with 1000 ng/ml IL-8 for 90 min, demonstrated that the drug induced a moderate, but significant, effect on the cell surface expression of both receptors. This was manifested by the fact that following the treatment with cytochalasin D, an increased reduction in cell surface expression of both CXCR1 and CXCR2 was observed (Fig. 1, A and B). Cytochalasin D enhanced the ability of IL-8 to induce reduction in cell surface expression of CXCR1 and of CXCR2 by 28.1% and by 29.3%, respectively, as compared with DMSO treatment (CXCR1, p < 0.002; CXCR2, p = 0.026). The ability of cytochalasin D to promote IL-8-induced reduction in cell surface expression of both receptors was time-dependent, because the drug had no effect when internalization was induced by 60 min-exposure to IL-8 (data not shown). Altogether, these results suggest that actin filaments, but not microtubules, are involved in processes regulating the cell surface expression of both CXCR1 and CXCR2 following IL-8-induced receptor internalization.

To visualize the process of receptor internalization in the presence of cytochalasin D, a confocal analysis was performed (the experiments were performed while the cells were in suspension). As shown in Fig. 2A, when CXCR2-expressing cells were not exposed to IL-8 and were not treated by the drug, a membranous expression of CXCR2 was noted, and an intact organization of actin filaments was observed. Following induction of internalization by 1000 ng/ml IL-8 (90 min, 37°C) in the absence of cytochalasin D, a clearly visible translocation of the receptors to intracellular cytoplasmatic regions of the cells was observed, while the actin network remained intact (Fig. 2B). A highly prominent intracellular localization of the receptors was observed also when IL-8-induced receptor internalization was performed in the presence of cytochalasin D, accompanied by a pronounced disruption of the actin cytoskeleton (Fig. 2C). These results provide evidence for notable receptor internalization occurring while actin depolymerization was induced by cytochalasin D.

View larger version (27K): [in this window] [in a new window]   FIGURE 2. Confocal analysis of WT CXCR2 internalization in the presence or in the absence of cytochalasin D. CXCR2-expressing HEK 293 cells were incubated in the absence or in the presence of IL-8. The incubation with IL-8 was performed either following exposure to cytochalasin D, or in its absence, as described in Materials and Methods. The cells were washed and stained with rabbit anti-CXCR2 Abs, followed by staining with rhodamine-conjugated goat anti-rabbit IgG, combined with Alexa Fluor 488 phalloidin. The cells were then subjected to confocal analysis as described in Materials and Methods. In all the pictures shown, the red color represents the expression of CXCR2 and the green color represents the distribution of actin filaments. A, Cells not exposed to cytochalasin D or to IL-8. B, Cells not exposed to cytochalasin D but incubated with IL-8. C, Cells exposed to cytochalasin D and incubated with IL-8. A representative experiment of three independent experiments performed is shown. The reference bar in the lower left corner represents 10 µm.

Studies of neutrophils, as well as our previous observations on our WT CXCR1- and WT CXCR2-transfected cells, indicated that if the internalization-inducing ligand was removed and the cells were allowed to recover at 37°C, the receptors were re-expressed on the cell membrane (31, 32, 33, 41, 42). Complete receptor recovery, especially for CXCR2, required prolonged incubation at 37°C (for 6 h), whereas partial re-expression was noticed following 90 min of cell recovery (31, 32). Further analysis demonstrated that receptor re-expression following this shorter incubation period (90 min) was insensitive to cycloheximide, indicating that receptor recovery under these conditions did not result of de novo receptor synthesis (31, 32). Moreover, confocal analysis demonstrated that the receptors are not stored in pre-existing granules in these cells (Ref. 32 and data not shown). On the whole, these observations suggest that CXCR1 and CXCR2 re-expression on the cell surface following 90 min of recovery was the direct result of receptor recycling back to the plasma membrane.

Determination of the effects of cytoskeleton-disrupting agents on WT CXCR1 and WT CXCR2 recycling was performed by inducing receptor internalization by 1000 ng/ml IL-8 for 60 min at 37°C (in the absence of nocodazole or cytochalasin D), and by allowing cell recovery for 90 min at 37°C, in the presence or in the absence of nocodazole or cytochalasin D. This time point of cell recovery was chosen due to the fact that, at this time, receptor re-expression is the result of receptor recycling (as mentioned for experiments with cycloheximide) and because in earlier time points the level of receptor recovery is not high and stable enough to allow for detection of relatively subtle effects of the cytoskeleton-disrupting agents. The levels of receptor recycling in the presence of the drugs were compared with the extent of receptor recycling in their absence or in the presence of DMSO, which was the solubilizer of the drugs and did not affect receptor recycling on its own (Figs. 3 and 4).

View larger version (25K): [in this window] [in a new window]   FIGURE 3. The effects of nocodazole and cytochalasin D on WT CXCR1 re-expression on the cell membrane following IL-8-induced internalization. CXCR1-expressing HEK 293 cells were incubated in the absence or in the presence of IL-8. One group of IL-8-exposed cells was washed and stained immediately after exposure to IL-8 with anti-CXCR1-specific Abs. Another group of IL-8-exposed cells was washed and allowed to recover in the presence of cytochalasin D, nocodazole, or DMSO, or in their absence. The cells were washed and subjected to FACS analysis as described in Materials and Methods. A, Recovery of CXCR1 expression following the different treatments. The data are presented as the percent of total receptor expression in cells that were not exposed to any treatment, calculated as described in Materials and Methods. *, p = 0.005 for treatment with cytochalasin D vs DMSO treatment. Cyto-D, Cytochalasin D; Nocod., nocodazole. Each value represents the mean ± SD of three independent experiments. B, FACS analysis demonstrating the expression of CXCR1 after the different treatments as follows: 1) cells that were not exposed to any treatment; 2) cells that underwent IL-8-induced receptor down-modulation and no recovery at 37°C; 3) cells that underwent IL-8-induced receptor down-modulation and recovery at 37°C in the absence of cytochalasin D, but in the presence of DMSO; and 4) cells that underwent IL-8-induced receptor down-modulation and recovery at 37°C in the presence of 1 µM cytochalasin D. Baseline represents cells stained with CSB instead of Abs to CXCR1, counts represents relative cell number, and FL1-H represents fluorescence. A representative experiment of three independent experiments performed is shown.

View larger version (21K): [in this window] [in a new window]   FIGURE 4. The effects of nocodazole and cytochalasin D on WT CXCR2 re-expression on the cell membrane following IL-8-induced internalization. CXCR2-expressing HEK 293 cells were incubated in the absence or in the presence of IL-8. One group of IL-8-exposed cells was washed and stained immediately after exposure to IL-8 with anti-CXCR2-specific Abs. Another group of IL-8-exposed cells was washed and allowed to recover in the presence of cytochalasin D, nocodazole or DMSO, or in their absence. The cells were washed and subjected to FACS analysis as described in Materials and Methods. A, Recovery of CXCR2 expression following the different treatments. The data are presented as the percent of total receptor expression in cells that were not exposed to any treatment, calculated as described in Materials and Methods. *, p = 0.02 for treatment with cytochalasin D vs DMSO treatment. Cyto-D, Cytochalasin D; Nocod., nocodazole. Each value represents the mean ± SD of three independent experiments. B, FACS analysis demonstrating the expression of CXCR2 after the different treatments as follows: 1) cells that were not exposed to any treatment; 2) cells that underwent IL-8-induced receptor down-modulation and no recovery at 37°C; 3) cells that underwent IL-8-induced receptor down-modulation and recovery at 37°C in the absence of cytochalasin D, but in the presence of DMSO; and 4) cells that underwent IL-8-induced receptor down-modulation and recovery at 37°C in the presence of 1 µM cytochalasin D. Baseline represents cells stained with CSB instead of Abs to CXCR2, counts represents relative cell number, and FL1-H represents fluorescence. A representative experiment of three independent experiments performed is shown.

Analysis of the involvement of cytoskeleton elements in the recovery of WT CXCR2 expression indicated that in similarity to WT CXCR1, microtubules were not involved in the recycling of CXCR2, as indicated by the inability of nocodazole to affect the recycling process (Fig. 4A). In contrast, a pronounced inhibition of CXCR2 recycling was induced by cytochalasin D (Fig. 4, A and B), as indicated by the fact that the treatment with this drug abolished CXCR2 recycling (98.8% inhibition) as compared with cells treated with DMSO (p = 0.02; Fig. 4A). On the whole, the results shown in Figs. 3 and 4 provide evidence for the significant involvement of actin filaments in processes that mediate CXCR1 and CXCR2 recycling to the cell surface. The extent of inhibition was more prominent for CXCR2 recycling as compared with CXCR1. The more rapid kinetics of CXCR1 recycling as compared with CXCR2 recycling (Fig. 3A vs Fig. 4A) suggest that the lower sensitivity of CXCR1 recycling to cytochalasin D results from the fact that recycling events of CXCR1 have taken place before the activity of drug has come into a complete effect.

Because the fine control of neutrophil responses results from the coordinated activity of several ELR⁺-CXC chemokines (25, 26, 28, 29, 30, 31, 32, 35, 36), we elucidated the role of actin filaments in the regulation of CXCR2 trafficking in response to GCP-2, the only other ELR⁺-CXC chemokine that was shown to induce potent internalization of this receptor. As shown in Fig. 5, CXCR2 recycling was inhibited not only when the internalization-inducing ligand was IL-8, but also when internalization was induced by 1000 ng/ml GCP-2 for 2 h at 37°C, and cell recovery was allowed for 90 min at 37°C (82.6% inhibition; p = 0.028), suggesting that actin filaments are involved in the trafficking of CXCR2 in a chemokine-nonspecific manner, provided that the chemokine is able to induce CXCR2 internalization and recycling.

View larger version (33K): [in this window] [in a new window]   FIGURE 5. The effect of cytochalasin D on WT CXCR2 re-expression on the cell membrane following GCP-2-induced internalization. CXCR2-expressing HEK 293 cells were incubated in the absence or in the presence of GCP-2. One group of GCP-2-exposed cells was washed and stained immediately after exposure to GCP-2 with anti-CXCR2-specific Abs. Another group of GCP-2-exposed cells was washed and allowed to recover at the presence of cytochalasin D, DMSO, or in their absence. The cells were washed and subjected to FACS analysis as described in Materials and Methods. The data are presented as the percent of total receptor expression in cells that were not exposed to any treatment, calculated as described in Materials and Methods. *, p = 0.028 for treatment with cytochalasin D vs DMSO treatment. Cyto-D, Cytochalasin D. Each value represents the mean ± SD of three independent experiments.

First, the ability of PTx to uncouple the binding of G_i was determined by evaluating its ability to inhibit IL-8-induced migration of CXCR1- and CXCR2-expressing cells. As shown in Fig. 6A, 100 ng/ml PTx completely abolished IL-8-induced chemotactic responses of both CXCR1 (100 ng/ml IL-8) and CXCR2 (50 ng/ml IL-8) (p < 0.001). Similar effects of the drug were observed in 500 ng/ml (data not shown). In contrast to its effects on IL-8-induced migratory responses, PTx did not interfere with the internalization of CXCR1 and CXCR2, induced by 60-min exposure to 1000 ng/ml of the chemokine (Fig. 6B). The inability of the drug to modify the level of internalization was observed not only when it was used in the concentration of 100 ng/ml (data not shown), but also in the dose of 500 ng/ml (Fig. 6B).

View larger version (31K): [in this window] [in a new window]   FIGURE 6. The role of G protein coupling in the regulation of IL-8-induced chemotaxis and internalization. A, The effect of PTx treatment on the migratory response of CXCR1- and CXCR2-expressing HEK 293 cells. The cells were exposed to 100 ng/ml PTx for 2 h at 37°C, washed, and subjected to chemotaxis assays in response to IL-8, as described in Materials and Methods. ***, p < 0.001 for PTx treatment vs no treatment. A representative experiment of three independent experiments performed is shown. B, The effect of PTx treatment on the IL-8-induced internalization of CXCR1 and CXCR2. The cells were exposed to 500 ng/ml PTx for 2 h at 37°C, washed and subjected to FACS analysis, as described in Materials and Methods. Each value represents the mean ± SD of two to three independent experiments.

The effects of cytochalasin D on CXCR1 and CXCR2 re-expression in cells of a hematopoietic origin

CXCR1 and CXCR2 are known to be expressed on cells of a hematopoietic origin, primarily on neutrophils (27, 28, 33, 34, 39). Although the expression of both CXCR1 and CXCR2 on neutrophils may complicate the analysis of the regulation of these receptors, similar experiments to those performed on HEK 293 cells were performed on human neutrophils, to determine the effects of cytochalasin D on CXCR1 and CXCR2 re-expression. The analysis was performed on healthy independent donors. The basal expression levels of CXCR1 and CXCR2 in all donors were highly similar to the reported expression levels of these receptors on neutrophils (33), demonstrating higher expression levels of CXCR1 than CXCR2 (data not shown). Of note is the fact that the kinetics of CXCR1 and CXCR2 internalization in neutrophils is more rapid than in HEK 293 cells, and their recycling is quickly reaching a plateau level in neutrophils (33). Moreover, CXCR1 was shown to undergo a more efficient recycling in human neutrophils than CXCR2, the later reaching only a maximal level of 40% recovery of receptor expression (33).

To determine the effects of cytochalasin D on CXCR1 and CXCR2 re-expression, the experimental conditions in neutrophils were set so that internalization and recycling levels would approximately accommodate with those used in HEK 293 cells. To this end, neutrophils were exposed to 1000 ng/ml IL-8 for 10 min at 37°C, followed by removal of the ligand and cell recovery for 40 min at 37°C. Cell recovery was performed in the presence of DMSO, 1 µM cytochalasin D, or in their absence. As in HEK 293 cells, DMSO did not affect the re-expression of CXCR1 and CXCR2 following removal of IL-8 (data not shown).

The exposure of neutrophils to IL-8 resulted in an average of 62.5% reduction in CXCR1 cell surface expression (Table I). The expression of CXCR1 was quickly restored following a 40-min recovery in the presence of DMSO, and reached an average of 51.8% of the basal expression level before induction of internalization (Table I). In each of the neutrophil samples analyzed, cytochalasin D did not affect the recycling process of CXCR1 (Table I). The inability of cytochalasin D to affect CXCR1 recycling may very well be affected by the rapid kinetics of the recycling process, and does not necessarily imply that the recycling of this receptor is not regulated by actin filaments (see Discussion).

To nevertheless determine whether the effects of cytochalasin D on IL-8 receptors recycling that were observed in HEK 293 cells are relevant to hematopoietic cells, similar analysis was performed on 300-19 pre B cells that were transfected to stably express CXCR1. Furthermore, the use of this system allowed us to elucidate the regulation of CXCR1 recycling in a hematopoietic cell system in which the regulation of CXCR1 could be dissociated from that of CXCR2. The results of this analysis indicated that in similarity to HEK 293 cells, the recycling process of CXCR1 in 300-19 pre B cells was partially dependent on actin filaments, as observed by the fact that the recycling was inhibited by 45.6% by exposure to cytochalasin D, but was not dependent on microtubules. Following the exposure of 300-19 pre B cells to 1000 ng/ml IL-8 for 90 min at 37°C, a marked decrease in CXCR1 cell surface expression was noted, the expression levels reduced to 21.0 ± 2.9% of the total expression before induction of internalization. The removal of the ligand and the recovery of the cells for 90 min at 37°C resulted in recovery of CXCR1 membrane expression to 53.9 ± 0.35% and to 55.2 ± 5% of the total receptor expression, observed when the recovery was performed in the absence or in the presence of DMSO, respectively. The cell surface expression level of CXCR1 determined following the exposure to cytochalasin D in the course of the recovery phase were restored to only 39.6 ± 3.4% of the total expression level (inhibition of 45.6%), whereas no significant effect of nocodazole was observed on CXCR1 expression (50.0 ± 8.5% of total expression level before induction of internalization). The fact that CXCR1 recycling in 300-19 pre B cells is regulated by similar mechanisms to those of HEK 293 cells supports the similarities previously observed between hematopoietic cells and HEK 293 cells, and suggests that HEK 293 cells provide a legitimate and a reliable system for the analysis of the regulation of chemokine receptor trafficking by cytoskeleton elements.

The carboxyl terminus domains of CXCR1 and CXCR2 are partially involved in the regulation of the actin-mediated process of receptor recovery

To determine the mechanisms that may be involved in the actin-mediated process of receptor recycling, analysis of the role of carboxyl terminus domains of CXCR1 and CXCR2 was performed. In preliminary experiments the optimal conditions for analyzing the response of WT and mutated CXCR1 and CXCR2 were determined. Based on these experiments, internalization was induced by exposure to IL-8 for 90 min at 37°C (unlike Figs. 3 and 4 in which internalization was induced by a 60-min exposure to IL-8), followed by cell recovery for 90 min at 37°C, in the presence or in the absence of cytochalasin D.

A mutated CXCR1, termed 19 CXCR1, was produced by carboxyl terminus truncation that resulted in the removal of eight of nine phosphorylation sites (Table II). High and similar levels of receptor internalization and recycling were observed for WT CXCR1 and 19 CXCR1. However, whereas the recovery of WT CXCR1 was inhibited by 30.0% by cytochalasin D, 19 CXCR1 recycling was not affected by the drug, demonstrating 0% inhibition (Table III). The levels of 19 CXCR1 recovery in the presence of DMSO were 93.9 ± 12.2%, and 95.1 ± 15.4% when performed in the presence of cytochalasin D (Table III). Of note is the fact that in contrast to our previous study, 19 CXCR1 demonstrated similar levels of internalization to WT CXCR1, probably resulting from the different experimental conditions used in the previous and present study, regarding primarily the densities of cells cultured for experiments that may affect actin polymerization and levels of internalization induced by exposure to IL-8 (data not shown).

Of note is the fact that, in similarity to the experiments shown in Table III (internalization, 90 min), reduced sensitivity of mutated receptor recycling to cytochalasin D, as compared with WT receptors was observed also when the internalization was performed by a 60-min exposure to the chemokine (data not shown). As demonstrated above, the overall sensitivity to cytochalasin D of all the cells included in this study, was reduced following a 90-min exposure to IL-8 (see Discussion), and enabled a clearer comprehension of the differences between the WT and the mutated receptors. Following a 60-min exposure to IL-8, the differences in sensitivity to cytochalasin D of WT and mutated receptors was 20% for CXCR1, as compared with 30% difference in a 90-min exposure to IL-8 (Table III; at the 60-min exposure to IL-8 there was a reduction from 50% sensitivity for the WT receptors to 30% sensitivity for the mutated receptor). For CXCR2, following a 60-min exposure to IL-8, about 60% difference in sensitivity to cytochalasin D was detected between the WT and mutated receptors, in similarity to 60% difference between the WT and mutated CXCR2 following a 90-min exposure to the ligand (Table III; at the 60-min exposure, reduction from 100% sensitivity for WT CXCR2 to 40% sensitivity for 20 CXCR2). In all, the results of these experiments suggest that the dependence of the recycling process of both CXCR1 and CXCR2 on actin filaments is partially determined by the carboxyl terminus domain of these receptors.

Cytochalasin D-induced inhibition of CXCR2 recycling results in impairment of CXCR2-mediated migratory responses

As with many other receptors, it is expected that the level of receptor expression on the plasma membrane of the cells will affect the ability of the cells to respond to stimuli that are mediated by these receptors. Because both CXCR1 and CXCR2 mediate cellular migration in response to IL-8 (25, 26, 29, 47), the level of their expression on the cell surface may affect the migratory response of the cells that express these receptors. As was already demonstrated (Fig. 1, A and B), exposure of CXCR1- and CXCR2-expressing cells to high concentrations of IL-8 resulted in receptor internalization, as manifested by a decrease in receptor cell surface expression. Following ligand removal and cell recovery, the receptors were recycled in a cytochalasin D-sensitive process, and their expression was partially restored (Figs. 3 and 4). To test whether the effects of cytochalasin D on receptor recycling have a functional relevance, we determined the ability of cells that underwent these different treatments to migrate in response to IL-8 (the cells were washed before loading in the chemotaxis chamber). Due to the fact that the cytochalasin D-mediated effects were more prominent on CXCR2, we focused in this part of the study only on CXCR2-expressing cells.

To determine whether the effect of cytochalasin D on CXCR2 recycling alters the migratory responses of these cells, CXCR2-expressing cells were exposed to internalization-inducing concentrations of IL-8, followed by recovery in the presence or in the absence of cytochalasin D. All the cells were washed before loading into the chemotaxis chamber.

The results shown in Fig. 7 demonstrate that untreated CXCR2-expressing cells potently migrate to IL-8 (lane 1). The level of migration was not affected by incubation with DMSO alone (data not shown). Upon exposure to 1000 ng/ml IL-8, in a treatment that resulted in a pronounced receptor down-modulation (Fig. 1B), a marked inhibition in the migratory response of the cells was observed (Fig. 7, lane 2). When IL-8-induced CXCR2 down-modulation was followed by ligand removal and cell recovery for 90 min at 37°C, the expression of the receptors was partially restored (Fig. 4), and this process was accompanied by an almost-complete recovery of migratory responses to IL-8 (Fig. 7, lane 3). However, if cytochalasin D was present in the course of receptor recycling, its ability to inhibit receptor recycling (Fig. 4) was manifested by a pronounced impairment of migratory responses to the chemokine (Fig. 7, lane 4; p < 0.0001 for migration after recovery in the absence of cytochalasin D (lane 3) vs migration after recovery in the presence of cytochalasin D (lane 4)).

View larger version (34K): [in this window] [in a new window]   FIGURE 7. Cytochalasin D-induced inhibition of CXCR2 recycling results in impairment of CXCR2-mediated migratory responses. CXCR2-expressing HEK 293 cells were either not incubated with IL-8 (lanes 1 and 5) or incubated with 1000 ng/ml IL-8 (lanes 2-4). The cells that were not exposed to IL-8 were either not exposed to any treatment (lane 1) or exposed to cytochalasin D for 90 min at 37°C (lane 5). The IL-8-exposed cells were subdivided into two groups. The cells of the first group were not allowed to undergo recovery at 37°C (lane 2). The cells of the other group were washed (to remove the ligand) and allowed to recover for 90 min at 37°C in the absence (lane 3), or in the presence of cytochalasin D (lane 4). The cells were washed and subjected to a chemotaxis assay in response to 50 ng/ml IL-8, as described in Materials and Methods. ***, p < 0.0001 for migration after recovery in the absence of cytochalasin D (lane 3) vs migration after recovery in the presence of the drug (lane 4), and for migration after recovery in the presence of cytochalasin D (lane 4) vs migration after the treatment with cytochalasin D alone (lane 5). HPF, High power field. A representative experiment of four independent experiments performed is shown.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
The two closely related chemokine receptors, CXCR1 and CXCR2 are desensitized by exposure to high ligand concentrations, in a process which is accompanied by potent receptor internalization. Following removal of the ligand, the recovery of the cells at 37°C allows the recycling of the receptors to the plasma membrane, and may enable the re-sensitization of cellular functions (29, 30, 31, 32, 33, 42, 43). Such regulatory processes may play a significant role in the fine tuning of chemokine-induced inflammatory responses, and necessitate a better understanding of the mechanisms that control the intracellular trafficking of these receptors. Our present study is the first to provide evidence for novel findings regarding the regulation of chemokine receptor trafficking, as indicated by the following three observations: 1) an increased reduction in cell surface expression of CXCR1 and CXCR2 was observed when receptor internalization was induced by IL-8 (without allowing receptor recycling) in the presence of an actin-disrupting agent but not of a microtubule-disrupting agent; 2) actin filaments are important contributors to the regulation of CXCR1 and CXCR2 recycling to the plasma membrane in HEK 293 cells and in 300-19 pre B cells. This observation has important physiological implications because the cytochalasin D-induced inhibition of CXCR2 recycling resulted in the inhibition of migratory responses. In contrast to actin filaments, microtubules apparently are not involved in the regulation of the recycling process of these receptors. 3) The carboxyl terminus domains of CXCR1 and CXCR2 were demonstrated to be partially involved in the regulation of the actin-dependent process of receptor recycling to the plasma membrane.

The cytochalasin D-induced promotion of IL-8-induced reduction in cell surface expression of CXCR1 and CXCR2 may be interpreted in two different ways. The first is that actin filaments should undergo a process of partial depolymerization to allow for a highly potent receptor internalization to occur. The second possibility, which is more likely in view of the direct involvement of actin filaments in the regulation of receptor recycling, is that the cytochalasin D-mediated promotion of IL-8-induced reduction in receptor cell surface expression could actually be the direct result of the ability of the drug to inhibit receptor recycling. With respect to this hypothesis, it is important to note that the level of receptor expression may be determined by rapid dynamics of receptor internalization and recycling back to the plasma membrane. Obviously, high levels of receptor recycling are observed following induction of internalization, removal of the ligand and recovery of the cells at 37°C. However, it is possible that upon exposure to internalization-inducing ligand, low levels of receptor recycling do occur concomitantly with high levels of receptor internalization. This may be manifested by the fact that in many systems of GPCR, including the one used in this study, exposure of the cells to ligands in conditions inducing maximal internalization, does not result in a complete abolishment of receptor expression, and reduction in cell surface expression often does not reach 100%. Therefore, the possibility exists that cytochalasin D treatment during induction of internalization actually inhibited the low levels of receptor recycling that occurred concomitantly with the highly prominent events of receptor internalization. The net effect of such activity will be the observed promotion of reduction in cell surface expression of CXCR1 and CXCR2. In support of this hypothesis is the observation that the cytochalasin D-mediated effect was time-dependent, and was observed only following longer incubation periods with IL-8, in which more pronounced recycling processes may have started to take place. Moreover, when the effects of cytochalasin D on receptor recycling were determined, it was observed that the sensitivity to the drug was reduced when the exposure to IL-8, in the course of internalization induction, was prolonged (Figs. 3 and 4 vs Tables II and III). This could be explained by the fact that when the internalization-inducing exposure to the chemokine was prolonged, recycling events were initiated during the internalization process, before the addition of cytochalasin D (added only in the course of recovery). Such processes may give rise to a receptor subpopulation that is not exposed to cytochalasin D, resulting in reduced sensitivity to the drug during receptor recycling.

The understanding of the role of actin filaments in the regulation of CXCR1 and CXCR2 recycling necessitates a thorough investigation of the components that may be involved. In that respect, it is important to note that a recent report by Nakagawa and Miyamoto (49) has demonstrated that actin filaments localize on sorting endosomes. Similarly, it is conceivable that a physical association between polymerized actin filaments and recycling endosomes is required to allow for the adequate trafficking of CXCR1 and CXCR2, and for their recycling back to the plasma membrane following their ligand-induced internalization.

The significant involvement of actin filaments in processes of receptor recycling, observed upon ligand removal following receptor internalization, was noticed for both CXCR1 and CXCR2. However, it is interesting to note that quantitative differences were observed between CXCR1 and CXCR2, regarding the effect of cytochalasin D on their recycling. The effect was considerably much more evident for CXCR2 than for CXCR1, supporting our previous observations, as well as findings by others, suggesting that the two receptors may be differently regulated at multiple levels (25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37). The quantitatively different effect of the drug on CXCR1 recycling, as compared with CXCR2 suggests that CXCR1 recycling is differentially regulated than that of CXCR2. Our results suggest that the difference between the two receptors is not the result of the divergent involvement of prime G_i-mediated signaling events in induction of processes that were initiated by ligand-induced internalization. However, one cannot exclude the possibility that non-G_i-mediated signaling events are involved in the regulation of this process. An alternative possible explanation for the different extent of actin filament involvement in the recycling process of each of the receptors may be that their intracellular trafficking following internalization is somewhat different. Indeed, it is possible that to some extent the two receptors undergo recycling through different endosomes, as indicated by the quicker recycling kinetics of CXCR1 as compared with CXCR2 (see Fig. 3A vs 4A; following 90 min of recovery, the expression of CXCR1 was restored to 71.4 ± 6.25% of total expression before induction of internalization, whereas the expression of CXCR2 was restored only to 46.5 ± 6.9% of total expression). Therefore, it is our intention to study the nature of the intracellular organelles that direct CXCR1 and CXCR2 in the course of IL-8-induced internalization and recycling, in analysis based on the expression of the Rab family of small GTPases. The quicker recycling process of CXCR1, as compared with CXCR2 further suggests that the lower ability of cytochalasin D to affect CXCR1 recycling results from the fact that actin-dependent processes of CXCR1 recycling occurred before the activity of the drug was complete (since the drug was added only at the time of recycling). In contrast, the slower kinetics of CXCR2 recycling may allow for cytochalasin D to fully depolymerize actin filaments, resulting in a more pronounced effect on the recycling process.

The use of HEK 293 cells for the analysis of the role of actin filaments in the internalization and recycling processes of CXCR1 and CXCR2 allowed us to dissociate the mechanisms regulating the intracellular trafficking of each of the receptors, independently of the other, and to gain insight into these processes by the use of mutated receptors. Similar analysis of the regulation of CXCR1 and CXCR2 recycling in neutrophils demonstrated that cytochalasin D did not affect the process of CXCR1 recycling. The above-mentioned suggestion that the sensitivity of the recycling process to cytochalasin D is affected by the kinetics of the process raises the possibility that the inability of the drug to affect CXCR1 recycling in neutrophils results from the rapid kinetics of the process in these cells. It is possible that the crucial events of CXCR1 recycling take place more quickly in neutrophils than in HEK 293 cells, before cytochalasin D activity has come into a complete effect, resulting in the apparent inability of the drug to affect the recycling process in neutrophils. If this is indeed the case, the inability of cytochalasin D to affect CXCR1 recycling does not necessarily mean that the process is actin-independent, but that the experimental approach taken cannot elucidate the role of actin filaments in this process. However, the possibility cannot be ruled out that the process is indeed independent of actin filaments in neutrophils, or that the cross-talk between CXCR1 and CXCR2, shown to be differently regulated in many other aspects of the trafficking process, hampers the ability to analyze this issue. Unfortunately, in contrast to other publications demonstrating low, but significant levels of CXCR2 recycling, no recycling of CXCR2 was observed in neutrophils in our study, therefore, making it impossible to determine the regulation of CXCR2 recycling by actin filaments in neutrophils.

The inability to conclusively determine the role of actin filaments in CXCR1 and CXCR2 recycling in neutrophils emphasizes the need for the use of other experimental systems. In that regard, it is important to note that the experiments performed on 300-19 pre B cells indicated that CXCR1 is subjected to similar regulation in other hematopoietic cells; the recycling of CXCR1 in these cells was sensitive to cytochalasin D treatment (but not to nocodazole), the sensitivity being in the range observed in HEK 293 cells. This observation supports the possibility that indeed the recycling of CXCR1 and CXCR2 is regulated by actin filaments and provides us with experimental tools to further investigate this issue. Moreover, because CXCR1 and CXCR2 are expressed not only on hematopoietic cells but also on adherent cells, such as endothelial cells and several types of tumor cells (50, 51), the analysis of the regulation of their trafficking in HEK 293 cells is of major relevance.

Further insight into the potential involvement of actin filaments in the recycling process was provided by the use of the 19 CXCR1 and the 20 CXCR2 mutants, indicating that the carboxyl terminus domains of CXCR1 and CXCR2 are partially involved in the regulation of the actin-mediated process of receptor recycling. These findings are of major interest in view of observations on other GPCR, suggesting that in the course of their recycling, these receptors are dephosphorylated to enable subsequent re-sensitization (9, 10, 11). Such a process of dephosphorylation may not be required for 19 CXCR1 and 20 CXCR2 recycling because they lack most of the carboxyl terminus phosphorylation sites that are involved in their functional desensitization (30, 52, 53). In such a case, it is possible that 19 CXCR1 and 20 CXCR2 undergo a divergent process of intracellular trafficking through different endosomes than the WT receptors, and that the trafficking of these endosomes is less dependent on functionally intact actin filaments. Therefore, our results on the recycling of 19 CXCR1 and 20 CXCR2 suggest that the carboxyl terminus domains of CXCR1 and CXCR2 are significantly involved in the trafficking of these receptors and on their dependence on cytoskeleton elements. This hypothesis is supported by recent findings demonstrating that the carboxyl terminus of several GPCR participate in the process of their intracellular sorting (54, 55, 56), and suggests that the removal of carboxyl terminus motifs that normally mediate the trafficking of the receptors into the recycling pathway, may result in their trafficking into the degradation pathway.

Because CXCR1 and CXCR2 both mediate migratory responses that necessitate polymerization of actin filaments (18, 48), our observations on the need for polymerized actin to participate in CXCR1 and CXCR2 recycling are of major importance. The treatment that was used to induce potent receptor internalization (1000 ng/ml IL-8) was shown in our previous studies, as well as in studies by others, to induce receptor phosphorylation that resulted in functional desensitization of the receptors (30, 38, 52, 53, 57). It was observed that the desensitization of chemoattractant receptors is accompanied by actin depolymerization (58). Our present observations indicate that when the desensitizing/internalizing chemokine was removed from the cell proximity, and the receptors were allowed to recover, functionally intact actin filaments were required to allow receptor recycling, providing a readily available pool of potentially active receptors that may undergo resensitization by subsequent stimuli. In the case of CXCR1 and CXCR2, resensitization will be manifested by the ability of the receptors to transmit signals for migratory responses that also require actin polymerization. Therefore, it is possible that following a desensitized refractory period, functionally assembled actin filaments are required for efficient receptor recycling, as well as for potent migratory responses to occur.

The involvement of functionally intact actin filaments in CXCR1 and CXCR2 recycling, as well as in migratory responses, suggests that chemokine receptors are subjected to a specialized regulation, under which both recycling and migration depend on an interdependent mechanism. Further studies are required to elucidate whether such regulation is a more generalized mechanism that participates in the trafficking of other chemoattractant receptors. As for the regulation of other members of the GPCR superfamily, the conflicting observations regarding the regulation of muscarinic cholinergic receptor and adrenergic receptor trafficking by cytoskeleton components, combined with our present study, suggest that different receptors that belong to the superfamily of GPCR are divergently regulated in this respect (54, 59, 60). However, it is interesting to note that in similarity to the role of actin filaments in the regulation of CXCR1 and CXCR2 recycling, the recycling process of ₂-adrenergic receptors was also dependent on actin filaments (54). Moreover, two additional reports, addressing the recycling of IgA receptor and of Tac (the IL-2 receptor subunit) indicated that depolymerization of actin filaments by cytochalasin D resulted in inhibition of receptor recycling (22, 61). When compared with CXCR1 and CXCR2, both IgA receptor and Tac are of a totally different nature in terms of composition and signaling. Tac is of special interest because it may be internalized in a clathrin-independent pathway (23, 61). In contrast to Tac, CXCR1, and CXCR2, are seven transmembrane receptors that were demonstrated to undergo clathrin-dependent internalization (17, 62). Therefore, it may be suggested that with regard to the regulation of their recycling, a line of similarity exists between receptors of totally different nature (CXCR1, CXCR2, ₂-adrenergic receptors, IgA receptor and Tac), implying that receptor recycling is an event that is tightly regulated by mechanisms that depend on the receptors and ligands that are involved.

	Acknowledgments

 
We thank Dr. Levi and the staff of her laboratory (Laboratory of Infectious Diseases, Department of Clinical Biochemistry, Faculty of Health Sciences, Ben-Gurion University of the Negev and Soroka Medical Center, Beer Sheva, Israel) for their highly appreciated assistance in providing human neutrophils, and would like to acknowledge that the 300-19 pre B cells used in this study were produced by Dr. Pellegrino (Theodor Kocher Institute, University of Bern). We also thank Dr. Ran for critically reviewing the manuscript, Dr. Mittelman for his assistance in the performance of confocal analysis, and Dr. Sagi-Assif for her assistance with FACS services.

	Footnotes

 
¹ Address correspondence and reprint requests to Dr. Adit Ben-Baruch, Department of Cell Research and Immunology, George S. Wise Faculty of Life Sciences, Tel-Aviv University, Tel-Aviv 69978, Israel.

² Abbreviations used in this paper: GPCR, G protein-coupled receptors; CSB, cell sorter buffer; CXCR, CXC chemokine receptor; ELR⁺-CXC, ELR-expressing CXC; GCP-2, granulocyte chemotactic protein-2; HEK 293, human embryonal kidney 293; PTx, pertussis toxin; WT, wild-type.

Received for publication April 10, 2000. Accepted for publication October 20, 2000.