N-Terminal Region Sulfation in a Conserved Extracellular CCR2B Is

Home , Sulfation

Monocyte Chemotactic Protein-1 Receptor CCR2B Is a Glycoprotein That Has Tyrosine Sulfation in a Conserved Extracellular N-Terminal Region This information is current as of September 27, 2021. Alexander A. Preobrazhensky, Sofya Dragan, Tomonori Kawano, Mikhail A. Gavrilin, Irina V. Gulina, Leena Chakravarty and P. E. Kolattukudy J Immunol 2000; 165:5295-5303; ; doi: 10.4049/jimmunol.165.9.5295 Downloaded from http://www.jimmunol.org/content/165/9/5295

References This article cites 68 articles, 39 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/165/9/5295.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 27, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Monocyte Chemotactic Protein-1 Receptor CCR2B Is a Glycoprotein That Has Tyrosine Sulfation in a Conserved Extracellular N-Terminal Region

Alexander A. Preobrazhensky, Sofya Dragan, Tomonori Kawano, Mikhail A. Gavrilin, Irina V. Gulina, Leena Chakravarty, and P. E. Kolattukudy1

Monocyte chemotactic protein-1 (MCP-1) binding to its receptor, CCR2B, plays an important role in a variety of diseases involving infection, inﬂammation, and/or injury. In our effort to understand the molecular basis of this interaction and its biological consequences, we recognized a conserved hexad of amino acids at the N-terminal extracellular domain of several chemokine receptors, including CCR2B. Human embryonic kidney 293 cells expressing Flag-tagged CCR2B containing site-directed muta-

tions in this region, 21–26, including a consensus tyrosine sulfation site were used to determine MCP-1 binding and its biological Downloaded from consequences. The results showed that several of these amino acids are important for MCP-1 binding and consequent lamelli- .podium formation, chemotaxis, and signal transduction involving adenylate cyclase inhibition and Ca2؉ influx into cytoplasm Mutations that prevented adenylate cyclase inhibition and Ca2؉ influx did not significantly inhibit lamellipodium formation and chemotaxis, suggesting that these signaling events are not involved in chemotaxis. CCR2B was found to be sulfated at Tyr26; this sulfation was abolished by the substitution of Tyr with Ala and severely reduced by substitution of Asp25, a part of the consensus sulfation site. The expressed CCR2B was found to be N-glycosylated, as N-glycosidase F treatment of the receptor or growth of http://www.jimmunol.org/ the cells in tunicamycin reduced the receptor size to the same level, from 50 to 45 kDa. Thus, CCR2B is the first member of the CC chemokine receptor family shown to be a glycoprotein that is sulfated at the N-terminal Tyr. These post-translational mod- ifications probably have significant biological functions. The Journal of Immunology, 2000, 165: 5295–5303.

hemotactic cytokines (chemokines) recruit leukocytes to fated in proteins (6, 11). Several examples are known where sulfa- sites of injury, infection, and inflammation (1–4). Che- tion of tyrosines was shown to be important either for protein-protein C mokines are classified mainly in terms of the patterns of interaction (12–16) or for protein precursor processing (11). Cys residues, which form structurally important disulfide bonds. Monocyte chemotactic protein-1 (MCP-1)2 receptor (CCR2B), by guest on September 27, 2021 Those with an amino acid residue between the N-terminal pair of which is known to play an important role in a variety of diseases Cys residues, called ␣ or CXC chemokines, chemotax neutrophils involving inflammation and/or injury such as wound healing, ath- and nonhemopoietic cells, whereas those without an intervening erosclerosis, and arthritis (1–4), also contains a consensus sulfa- ␤ amino acid residue, called or CC chemokines, chemotax mono- tion domain at the N-terminal region. We noticed that a hexad of cytes, T cells, eosinophils, and NK cells. These two groups con- amino acid residues in this region is conserved in several chemo- stitute the majority of known chemokines. Chemokines specifi- kines, and therefore we suspected it to be functionally important. cally bind to seven transmembrane G protein-coupled receptors To test whether these residues are involved in the biological func- present on the target cells and cause chemotaxis and signal trans- tion of the receptor, site-directed mutagenesis was used. This mu- duction events that are not well understood. tagenesis involved the consensus sulfation site, including tyrosine, A chemokine receptor, CCR5 was recently reported to require the suspected site of sulfation, and the neighboring amino acid tyrosine sulfation for its function as a coreceptor for HIV-1 (5). residue that should be a part of the consensus required for enzy- Tyrosine sulfation of proteins is a modification that widely occurs matic sulfation by protein tyrosine sulfotransferases (6, 11). The in multicellular eukaryotic organisms (6, 7). Recently, two protein results presented here demonstrate that the hexad is important for tyrosine sulfotransferases, PTST-1 and PTST-2, have been cloned ligand binding, lamellipodium formation, chemotaxis, and signal (8–10). The sulfation of tyrosines occurs post-translationally, in the trans part of the Golgi network (6). The most important feature transduction. Structural alterations in this region of the receptor of tyrosine sulfation consensus sequences is the presence of an caused differential effects on lamellipodium formation and chemo- 2ϩ acidic or neutral amino acid residue directly before a tyrosine to be taxis vs Ca influx and adenylate cyclase inhibition, indicating sulfated, although not all theoretical consensus sequences are sul- that the receptor plays a role beyond the known function of causing ␣ ␤␥ dissociation of the trimeric G protein into Gi and G . Our results also show that tyrosine within the hexad is sulfated and that Neurobiotechnology Center and Departments of Biochemistry and Molecular and the mutation in the consensus residue thought to be required for Cellular Biochemistry, Ohio State University, Columbus, OH 43210 sulfation severely inhibits sulfation and consequently inhibits the Received for publication March 14, 2000. Accepted for publication August 9, 2000. biological function of the receptor. Some chemokine receptors, The costs of publication of this article were defrayed in part by the payment of page such as CXCR4, are known to be glycoproteins, although no CC charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 Address correspondence and reprint requests to Dr. P. E. Kolattukudy, Neurobio- technology Center, Ohio State University, 206 Rightmire Hall, 1060 Carmack Road, Columbus, OH 43210. E-mail address: [email protected] 2 Abbreviation used in this paper: MCP-1, monocyte chemotactic protein-1.

chemokine receptor has been shown to be glycosylated. Our stud- once with HEPES containing binding buffer (1 mM CaCl2, 5 mM MgCl2, ies show that CCR2B is a glycoprotein and is sulfated at 0.5% BSA, and 50 mM HEPES, pH 7.2) (24). The final assay volume (250 ␮ ϫ 6 tyrosine 26. l) contained 5 10 transfected and washed cells, various amounts of MCP-1, and 0.02 pmol of iodinated MCP-1 protein. Various transfected cell lines were incubated with labeled and unlabeled MCP-1. Following Materials and Methods incubation, the cells were passed through a Whatman GF/C filter, and Reagents radioactivity on the filter was measured in a Packard COBRA gamma

radiation counter (Downers Grove, IL). Dissociation constant (Kd) values The expression vector, pcDNA3 was obtained from Invitrogen (San Diego, for MCP-1 binding to CCR2B and mutated receptor in various cell lines of CA). Human embryonic kidney (HEK293) cells were obtained from Amer- HEK293 were determined using the LIGAND program (25). ican Type Culture Collection (ATCC CRL 1573; Manassas, VA). Lipo- fectamine, G418, penicillin-streptomycin solution, DMEM/F-12, BAC- ϩ TO-BAC baculovirus expression system, and FBS were purchased from Measurement of Ca2 influx induced by MCP-1 Life Technologies (Gaithersburg, MD). MEM/Eagle’s medium was obtained from ICN. Anti-Flag M2 mAb was purchased from Eastman Kodak Calcium influx into the cytoplasm was measured in HEK293 cell lines ␮ (Rochester, NY). Mouse IgG1, ␬ (MOPC-21), FITC-conjugated sheep anti- expressing CCR2B or its mutants after incubation with 2 M fura-2/ace- mouse IgG (whole molecule), phalloidin-FITC, chloramine-T, isobutyl- toxymethyl ester for 30 min at 37°C in a CO2 incubator. Fluorescence was methylxanthine, forskolin, QAE-Sephadex A-25, and Igepal CA-630 were measured after the addition of different concentrations of MCP-1 in a Per- obtained from Sigma (St. Louis, MO). Fura-2 and calcein acetoxymethyl kin-Elmer LS 3B fluorescence spectrometer (Palo Alto, CA) with constant ester were purchased from Molecular Probes (Eugene, OR). Na125I and stirring as previously described (18). The maximal fluorescence was mea- ␮ 2ϩ [3H]adenine were obtained from NEN (Boston, MA); Fluoromount-G sured after treating the cells with 50 M digitonin, and Ca influx is mounting medium was purchased from Southern Biotechnology Associates expressed as a percentage of the maximal fluorescence. 35

(Birmingham, AL). [ S]Na2SO4 was obtained from American Radiola- Downloaded from beled Chemicals (St. Louis, MO). Protein G-agarose and N-glycosidase F Adenylate cyclase assay deglycosylation kit were obtained from Roche (Indianapolis, IN). Polyvi- nylidene difluoride membrane was purchased from Millipore (Bedford, Stably transfected, confluent HEK293 cells were labeled overnight with 2 MA), and GF/C filters were obtained from Whatman (Clifton, NJ). The ␮Ci/ml [3H]adenine (31.7 Ci/mmol) in HEPES buffer containing DMEM/ enhanced chemiluminescence kit for immunodetection was obtained from F-12 medium supplemented with 10% FBS, 1% penicillin-streptomycin, Pierce (Rockford, IL). Anti-CCR2B Ab CCR2-03 was provided by Prof. and 0.4 mg/ml of G418 in 12-well tissue culture plates. The cells were M. Mellado (Universidad Autońoma de Madrid, Madrid, Spain). washed in the presence of 1 mM isobutylmethylxanthine and treated with

different concentrations of MCP-1 in the presence or the absence of an http://www.jimmunol.org/ Expression of MCP-1 activator, forskolin, for 30 min at room temperature. Intracellular [3H]ATP and [3H]cAMP, separated through Dowex 50W and neutral alumina col- A human MCP-1 cDNA clone (17) was used to express MCP-1 in the 3 Bac-To-Bac baculovirus expression system, and the recombinant protein umns, respectively (26–28), were assayed for H in a scintillation counter. The amount of cAMP was normalized to its own ATP pool for each frac- was puriﬁed in a two-step fast protein liquid chromatography protocol ϫ using Mono S HR 5/5 and Superdex-75 HR 10/30 columns (American tion and estimated by their ratios (cpm of cAMP/ATP 100). In each Pharmacia Biotech, Arlington Heights, IL) as described previously (18). experiment a full MCP-1 dose-response curve was generated and expressed as a percentage of the forskolin control. All data points were determined in Constructs for CCR2B receptor mutants duplicate, and results were plotted as a percentage of adenylate cyclase inhibition. CCR2B receptor was cloned using primers that were designed based on the

published sequence (19), and the first strand of cDNA was generated from by guest on September 27, 2021 total RNA isolated from the THP-1 monocytic leukemia cell line. We Chemotaxis assay cloned the product into pcDNA3 eukaryotic expression vector at BamHI and XbaI sites that were engineered into the primers. Prolactin leader se- Chemotaxis of transfected HEK293 cell lines was measured as described previously (18). Transfected cells (107/ml) were suspended in Gey’s bal- quence (20) (from Dr. I. F. Charo, University of Calfornia, San Francisco, ␮ CA) and a Flag tag epitope sequence (DYKDDDDK) were introduced at anced cell solution containing 0.2% BSA, incubated with 2 M calcein acetoxymethyl ester at 37°C for 30 min, washed with PBS buffer, and the N terminus immediately following the initiator amino acid, methionine. ϫ 6 The presence of an octapeptide Flag tag provides an Ab epitope allowing resuspended in the Gey’s BSA solution at 1 10 cells/ml. Various amounts of MCP-1 were placed in a 96-well Polytronic view plate immunological detection of the recombinant protein (21). The Flag se- ␮ quence was joined to the receptor by amplification of an N-terminal se- (UniPlate 350; Neuroprobe, Cabin John, MD), and 200 l of cell suspen- quence with both primers. Mutations at the conserved hexapeptide sion was added to the top wells of the chamber. Following 2-h incubation (TTFFDY) sequences located at positions 21–26 were performed by a two- at 37°C, the number of migrating cells was determined by measuring flu- step PCR method (22). Each amino acid at the hexapeptide region was orescence with a Cytofluor 2300 plate reader (Millipore). The quantitation mutated to alanine individually. In addition, Tyr 26 was mutated to phenyl- of HEK293 cell migration is useful only to detect major changes in cell alanine individually. mobility; therefore, we used the measured comparative chemotactic activ- ities only in this context. Transfections, cloning, and selection of highly expressing stable transformants Confocal microscopic analysis HEK293 cells were grown in a 1/1 mixture of DMEM and nutrient mix HEK293 cells expressing wild-type and mutant forms of CCR2B were F-12 containing 10% FBS and 1% penicillin/streptomycin solution in a stained with phalloidin-FITC to visualize polymerization of actin as pre- CO2 incubator at 37°C. Lipofectamine was used at 20 mg/ml to transfect viously described (29). The cells in culture (70% confluent) were harvested ␮ 60–80% confluent cells with 1 g of DNA by following the manufactur- and suspended in PBS containing 10 mM EDTA, washed twice with fresh er’s instructions. Transfected cells selected in the presence of 0.4 mg/ml culture medium, spread on plastic chamber slides, and grown for 24–36 h. ␮ G418 were analyzed by flow cytometry with the anti-Flag mAb (10 g/ The cells were incubated with or without MCP-1 (50 nM) for 15 min on the ϫ 6 ml/0.2 10 cells). Untransfected HEK293 cells were mixed with mouse slides, fixed by soaking the slides in ice-chilled PBS containing 4% para- ␬ ␮ IgG1 MOPC-21 at 10 g/ml as a control. Cells were incubated for 30 formaldehyde for 15 min, and washed twice in PBS. The slides were in- min at 4°C, washed with PBS, and labeled with 1/256 diluted FITC-con- cubated with 0.05 mg/ml phalloidin-FITC for 40 min at room temperature jugated sheep anti-mouse IgG at 4°C for 30 min in the dark and washed in a humidified chamber in the dark, washed three times in PBS for a total twice with PBS containing 5% FBS; expression of the receptor was deter- of 30 min, and mounted with glass coverslips using Fluoromount-G mount- mined by FACS. Cells were cloned, and the highly expressing clones were ing medium. Actin polymerization in the cells was visualized using a Bio- selected for producing the cell cultures used in the various experiments. Rad MRC 1024 laser scanning confocal microscope (Hercules, CA). FITC Before using each culture, expression of the receptor was measured by fluorescent signals were detected using a 480-nm excitation filter. Cells FACS analysis. forming lamellipodia (Ͼ12 ␮m) were counted under the confocal micro- MCP-1 binding assay scope. The frequency of lamellipodium formation (percentage) before the addition of MCP-1 was subtracted from that measured after MCP-1 treat- Radiolabeling of MCP-1 was performed by the chloramine-T method (23) ment. Samples were triplicated, and approximately 50 cells were examined with a modification (18). Stable transformants were washed in PBS and in each replicate. The Journal of Immunology 5297

Preparation of detergent extract HEK293 cells stably transfected with wild-type or mutant receptors were cultured in 75-cm2 ﬂasks in DMEM/F-12 medium until 50–75% conﬂu- ence. In some experiments tunicamycin was added to the cells at 1, 3.3, or 10 ␮g/ml for 2 days, with a change in medium after 24 h. The cells were removed with 10 mM EDTA solution in PBS, washed twice with PBS, and lysed in a buffer (50 mM Tris-HCl buffer, pH 7.5) containing 150 mM NaCl, 1% Igepal CA-630, and 0.5% deoxycholate. The following protein- ase inhibitors were added to the buffer just before use: PMSF (up to 1 mM), leupeptin (6.8 ␮g/ml), pepstatin (6.9 ␮g/ml), and aprotinin (5 ␮g/ml). Usu- ally 1 ml of buffer was used to lyse 5 ϫ 106 cells. The homogenate was centrifuged in an Eppendorf microfuge at 14,000 rpm for 15 min. 35 Ϫ2 Cell labeling with SO4 The monolayers of cells were washed three times with MEM/Eagle’s me-

dium supplied with nonessential amino acids, glutamine, and CaCl2 at 0.2, ␮ 35 10, and 1 mM, respectively. Then, 500 Ci of Na2 SO4 was added to the cell culture in 10 ml of the same medium, and the culture was incubated for 24 h. After immunoprecipitation, labeled protein preparations were subjected to SDS-PAGE, polyvinylidene difluoride blotted, and visualized by a phosphorimager. Downloaded from Immunoprecipitation of CCR2B by anti-Flag Ab-agarose FIGURE 1. Surface expression of CCR2B and its mutants in HEK293 cells. Cloned stably transfected HEK293 cells were grown and examined Anti-Flag M2 affinity gel (100 ␮l) was added to 1–3 ml of cell extract in for expression of the CCR2B and its mutants by flow cytometry after a test tube, and the suspension was slowly rotated overnight at 4°C. The gel incubation with the Flag mAb conjugated with FITC. Untransfected and was washed once with the lysis buffer and twice with 50 mM Tris-HCl transfected HEK293 are denoted by the bold solid line and dotted line, buffer, pH 7.5, containing 150 mM NaCl and 0.1% Igepal CA-630. The respectively. proteins were eluted either by incubation of the gel with an equal volume

of 2ϫ Laemmli sample buffer for 5 min at 100°C or twice by 0.15 ml of http://www.jimmunol.org/ 0.1 M glycine (pH 2.5) and 0.1% Igepal CA-630 for 10 min each elution, followed by neutralization with 1 M Tris and precipitation with ethanol. Substitution of all conserved amino acid residues with Ala abol- Immunodetection was performed using the enhanced chemiluminescence 2ϩ method. ished MCP-1 binding and Ca inﬂux into the cytoplasm caused by MCP-1 addition (data not shown). When mutants with single Indirect immunoprecipitation of CCR2B substitutions were compared, F24A caused a 5-fold decrease in After centrifugation of the homogenate, 25 ␮g of mAb M2 or a control binding, whereas F23A and D25A showed only minimal changes mAb was added to the extract, and the mixture was incubated on ice for in binding (Table I). At position 26, substitution of Tyr with Ala ␮ 1–3 h. Then, 100 l of protein G-agarose preincubated with 5% FBS in caused only a 10-fold decrease in binding, whereas substitution

lysis buffer was added to the extract, and the test tube with the mixture was by guest on September 27, 2021 slowly rotated overnight at 4°C. The gel was washed several times accord- with Phe virtually abolished binding. ing to the manufacturer’s protocol, and the proteins were eluted from the Effects of mutations in the conserved segment of CCR2B on the gel by adding equal volume of 2ϫ Laemmli sample buffer. actin-based cytoskeletal changes induced by MCP-1 were ana- Treatment CCR2B with N-glycosidase F lyzed by confocal microscopy. MCP-1 binding to wild-type CCR2B caused formation of very thin, finger-like protrusions of The N-glycosidase F deglycosylation kit was used according to the man- plasma membrane, called filopodia, in about 5 min (data not ufacturer’s (Roche) instructions for treatment of CCR2B isolated by M2 affinity gel precipitation. shown). In MCP-1-treated CCR2B cells, number of filopodia formed per cell reached 3.2, which is 2.6-fold greater than that in Results the control. After filopodia were formed, another type of actin- The N-terminal extracellular domain of chemokine receptors is based membrane protrusion, known as membrane ruffles or lamel- thought to play an important role in ligand binding. Within this lipodia, was formed within about 10 min (Fig. 2). The percentage region a segment is conserved among CCR1, CCR2, and the of cells forming lamellipodia extending Ͼ12 ␮m was used as the CMV-encoded US28. Five residues within the hexapeptide measure of the frequency of lamellipodium formation. To deter- TTFFDY located at positions 21–26 in CCR2B are identical in the mine whether this frequency of lamellipodium formation corre- three receptors. To test whether this conservation has any func- lated with chemotactic migration, the frequency observed in the tional significance, we used a site-directed mutagenesis approach. cells expressing wild-type CCR2B and its mutants were plotted We replaced all residues in the hexad with alanine, and each res- against their score for migration (Fig. 3). Y26A cells showed the idue was individually substituted by Ala. We expressed the Flag- highest level of MCP-1-induced formation of lamellipodia among tagged mutants fused to prolactin leader peptide in pcDNA3 vector all mutant receptors used in this study, and this level was similar in HEK293 cells after confirming by DNA sequencing that each to that observed with the wild-type CCR2B-expressing cells. This construct did not contain any unintended mutations. Stable trans- mutant also showed chemotactic activity somewhat comparable to formants were cloned, and the expression of the receptor by each that of the wild type. On the other hand, Y26F showed only about clone was tested by flow cytometry with anti-Flag M2 Ab. The 3% lamellipodium formation (Fig. 3), and this mutant showed very highly expressing clones were selected for producing the cells used little chemotactic activity. Untransfected cells and all mutants with for further studies on CCR2B and its mutants; the surface expres- drastically decreased chemotactic activity induced by MCP-1 did sions of all the mutants used for our studies were comparable (Fig. not show much MCP-1-induced lamellipodium formation (Fig. 3). 1 and Table I). The wild-type CCR2B and its mutants were tested On the other hand, cells expressing wild-type CCR2B and its mu- for ligand binding, induction of filopodia and lamellipodium for- tants that were active in MCP-1-induced chemotactic migration mation by MCP-1, chemotaxis in an MCP-1 gradient, induction of showed high frequency of lamellipodium formation (Fig. 3 and Ca2ϩ influx into the cytoplasm induced by this chemokine, and Table I). Lamellipodium formation showed good correlation with adenylate cyclase inhibition caused by it. chemotaxis. Neither the frequency of filopodium formation nor the 5298 SULFATION AND GLYCOSYLATION OF CCR2B

Table I. Effects of mutation in the conserved hexad of amino acids in the N-terminal domain of CCR2B on its biological activity

Adenylate Chemotactic Lamellipodium MCP-1 Binding Ca2ϩ Inﬂux (% of Cyclase Migration Formation Index Level of Ϯ Ϯ Ϯ Mutation KD (nM SE) maximum) IC50 (nM) (% SE) ( SE) Expression (%) TTFFDY (Wt) 4.4 Ϯ 1.0 100 0.3 100 31.8 Ϯ 7.8 88 ATFFDY (T21A)a 2.0 Ϯ 0.5 90 0.1 60 7.2 Ϯ 2.8 63 TAFFDY (T22A) ND 80 0.2 ND ND 62 TTAFDY (F23A) 2.0 Ϯ 0.3 0 ϾϾ1000 60 16.6 Ϯ 1.0 69 TTFADY (F24A) 19.0 Ϯ 1.7 75 8.0 8 0 Ϯ 2.3 76 TTFFAY (D25A) 6.0 Ϯ 3.0 0 ϾϾ1000 6 0 Ϯ 7.5 71 TTFFDA (Y26A) 42.0 Ϯ 27.0 0 ϱ 70 31.6 Ϯ 2.4 78 TTFFDF (Y26F) N/A 0 ϱ 15 3.0 Ϯ 1.4 80

a Bold letters indicate amino acid substitutions. N/A, Not applicable because there was no binding. length of the filopodia showed correlation with chemotaxis (data in the conserved N-terminal region. Ca2ϩ influx was dependent on not shown). Thus, formation of lamellipodia is probably involved the MCP-1 concentration; with the cells expressing CCR2B, 6 nM in MCP-1-induced chemotactic migration MCP-1 produced the maximal fura-2 fluorescence, which was 38%

Chemotaxis of HEK293 cells expressing the wild-type CCR2B of the maximal fluorescence observed when the membrane was Downloaded from and its mutants showed that chemotactic capability drastically de- disrupted with digitonin. T21A, T22A, and F24A showed only creased by the substitution of F24 or D25 by Ala and moderately small decreases in Ca2ϩ influx compared with that elicited by decreased by substitution of T22 with Ala (Table I). T21A and CCR2B, whereas F23A, D25A, Y26A, and Y26F showed no cal- F23A showed moderate chemotactic activity. Unexpectedly, sub- cium influx (Fig. 4A). stitution of Y26 with Phe caused a drastic decrease in chemotaxis, Since CCR2B is a G protein-coupled receptor that functions whereas substitution with Ala showed only a small decrease in with G , MCP-1 binding causes adenylate cyclase inhibition (Fig.

i http://www.jimmunol.org/ Ͻ chemotaxis. In all these cases, lamellipodium formation correlated 4B). A 50% inhibitory dose (IC50)of 1.0 nM was found for the with chemotaxis. wild-type CCR2B, and a similar IC50 was manifested by the mu- Calcium inﬂux into the cytoplasm caused by MCP-1 binding to tants, T21A and T22A. With T24A, a somewhat higher concen- the receptor expressed on HEK293 cells was affected by mutation tration of MCP-1 (8 nM) was required to achieve 50% inhibition, whereas F23A and D25A showed some inhibition, but the 50% level could not be achieved or calculated from the data (Fig. 4B). With Y26A and Y26F, absolutely no adenylate cyclase inhibition could be observed with any concentration of MCP-1.

Since the conserved N-terminal region studied includes a con- by guest on September 27, 2021 sensus sequence as a sulfation site, we tested whether CCR2B is sulfated at Y26. Incubation of the HEK293 cells that express CCR2B in [35S]sulfate-containing medium followed by immunological isolation of the solubilized receptor and SDS-PAGE showed that a 50-kDa protein was labeled with 35S (Fig. 5A). To determine whether this sulfation is located at Y26, mutants in

FIGURE 3. Correlation between lamellipodium formation and chemo- FIGURE 2. Confocal microscopic image of lamellipodia induced by tactic migration by MCP-1 treatment of HEK293 cells expressing CCR2B MCP-1 treatment of HEK293 cells expressing CCR2B and its mutants. and its mutants. HEK293 cells untransfected or transfected with CCR2B or HEK293 cells transfected with wild-type (Wt) CCR2B and its mutants its mutants were left resting or were stimulated with 50 nM MCP-1 for 15 were left resting (control) or were stimulated with 50 nM MCP-1 for 15 min on plastic slides. The number of cells forming lamellipodia was min on plastic slides, and actin-based morphological changes were visu- counted before and after addition of 50 nM MCP-1. Cells forming lamel- alized with phalloidin-FITC. Experiments were repeated three times, and lipodia (Ͼ12 ␮m) were counted under the confocal microscope, and this most representative images are presented. number was plotted against migration (percentage). The Journal of Immunology 5299

FIGURE 5. Immunoprecipitation of [35S]sulfate-labeled CCR2B and its mutants from HEK293 cells expressing CCR2B. A, Indirect precipitation using mAb M2 and protein G-agarose (lane 3), control Ab (lane 2), and total cell extract before immunoprecipitation (lane 1). B, Direct precipitation of 35S-labeled proteins with M2-agarose gel.

by SDS-PAGE (Fig. 7A). The untreated samples showed a major band at 50 kDa and a minor band at 38 kDa. The N-glycosidase-

treated samples showed immunologically cross-reacting major Downloaded from

species at 45 kDa and a minor one at a slightly lower Mr. We also tested the size of the receptor produced in the presence of N-glycosylation inhibitor. CCR2B was immunoprecipitated from cells growing for 2 days in the presence of 10 ␮g/ml of tunicamycin (Fig. 7B). After SDS-PAGE, the major receptor species from con-

trol cells appeared at 50 kDa, with a minor one at 38 kDa, whereas http://www.jimmunol.org/ in tunicamycin-treated cells, the major receptor was found at 45

kDa with a minor one at a slightly lower Mr than 38 kDa, exactly as seen after N-glycosidase treatment. The lowering of the molec- FIGURE 4. Measurement of calcium influx and adenylate cyclase inhi- ular size by N-glycosidase and tunicamycin treatments shows that bition by MCP-1 in HEK293 cells expressing CCR2B or its mutants. A, CCR2B contained N-glycosidically attached carbohydrates. The MCP-1 concentration dependence of fluorescence enhancement in To test whether the N-glycosidically attached carbohydrates on fura-2-labeled HEK293 cells expressing CCR2B or its mutants was mea- the receptor are sulfated, 35S-labeled CCR2B, immunologically sured. The calcium flux is expressed as a percentage of the maximum 35 Ϫ2 isolated from HEK293 cells expressing this receptor in SO4 fluorescence elicited by treating the cells with digitonin. B, MCP-1-depen- containing medium, was treated with N-glycosidase, and the prod- by guest on September 27, 2021 dent inhibition of adenylate cyclase was measured as described in Mate- ucts were subjected to SDS-PAGE and phosphorimaging (Fig. rials and Methods.IC values are shown in Table I. 50 7C). The molecular size of the 35S-labeled CCR2B was shifted from 50 to 45 kDa, with no indication of any loss in 35S. This which this Y is substituted with F or A were tested for sulfation result confirmed the glycoprotein nature of CCR2B and showed (Fig. 5B). Immunologically recovered mutant receptors contained that the sulfation is not on the carbohydrate. little label, showing that Y26 is the site of sulfation. To determine Surface expression of CCR2B decreased 2-fold after treatment whether the mutation in the neighboring position affects tyrosine with tunicamycin (Fig. 8). However, this treatment did not alter sulfation, the D25A mutant of CCR2B was tested for sulfation. Ca2ϩ influx into the cytoplasm and adenylate cyclase inhibition at The immunologically isolated D25A mutant of CCR2B showed a saturating concentrations of MCP-1 (data not shown). Thus, some very low level of label, demonstrating that this consensus acidic of the signaling events caused by MCP-1 treatment are not affected amino acid is required for optimal sulfation at Y26. In all these by the absence of glycosylation. Since tunicamycin is a general cases, immunological examination of the cells expressing the Flag- inhibitor of protein N-glycosylation that can affect many parame- tagged CCR2B and its mutants showed that they were all equally ters, we did not perform extensive studies on functional changes expressed on the surface; thus, the difference in labeling repre- that may be caused by tunicamycin treatment. Specific effects of sented the level of sulfation rather than any differences in expres- glycosylation of CCR2B on its biological function must await site- sion of the receptor. directed mutagenesis of the receptor that selectively prevents its Sulfation experiments showed that the 35S-labeled receptor and glycosylation. its mutant D25A migrated in SDS-PAGE as a protein considerably larger in size (50 kDa) than the size (41 kDa) expected from its amino acid composition. CCR2B immunoprecipitated from the detergent extract with anti-Flag mAb (M2) and protein G-agarose was subjected to SDS-PAGE and immunoblotted with the anti- Flag Ab or anti-CCR2B Ab (CCR2-03) (30). In both cases, a major 50-kDa species and a minor 38-kDa species were found (Fig. 6, lanes 1 and 2). On a control blot treated with an mAb against another Ag, only heavy and light chains of M2 Ab could be detected with the anti-mouse HRP conjugate (Fig. 6, lane 3). To test whether N-glycosidically attached carbohydrates contributed to FIGURE 6. Immunoprecipitation of CCR2B expressed in HEK 293 this larger size, the immunologically isolated receptor preparation cells by M2-agarose gel. Immunoblots were performed with mAb M2 (lane was treated with N-glycosidase, and the products were examined 1), mAb CCR2-03 (lane 2), and control Ab (lane 3). 5300 SULFATION AND GLYCOSYLATION OF CCR2B

Among the mutants in the conserved N-terminal region, only one mutant, Y26F, did not show ligand binding, signaling, and chemotaxis. Other mutations retain ligand binding, but differen- tially affect signaling and chemotaxis. Migration of cells transfected with CCR2B receptor showed maximal chemotactic activity at 1 nM MCP-1. Cells transfected with F24A, D25A, and Y26F mutant receptors showed a loss in their chemotactic activity. Four FIGURE 7. Decrease in the molecular size of CCR2B expressed in mutants (F23A, D25A, Y26A, and Y26F) were found to be de- HEK293 cells by treatment with N-glycosidase F (A and C) and by treat- fective in adenylate cyclase inhibition and calcium signaling, and ment of the cells with tunicamycin (B). ϩ, Treatment; Ϫ, control. In all three (F24A, D25A, and Y26F) were defective in cell migration. cases, immunodetection was with mAbM2. C, The transfected cells were Only D25A and Y26F were inactive in signaling and chemotaxis. grown in 35SO Ϫ2. 4 F23A and Y26A mutations caused a loss in signaling, but had only moderate effects on chemotaxis. However, F24A, which retained calcium elicitation and adenylate cyclase inhibition, showed no Discussion chemotaxis in response to MCP-1. These results show that Ca2ϩ The extracellular N-terminal domain of chemokine receptors is influx and chemotaxis elicited via MCP-1 binding to CCR2B are thought to play a significant role in binding the ligand (31, 32). In not necessarily coupled. Even though we cannot rule out the pos- 2ϩ our effort to explore possible functions of this N-terminal region, sibility that Ca influx, too low to detect by the methods we used, we noted that a segment within this region is conserved in CCR1, is involved in chemotaxis, we consider it unlikely. Whether the Downloaded from 2ϩ CCR2, and the virally encoded US28 that also binds MCP-1 and Ca influx associated with chemokine binding to cell surface re- RANTES (33–35). To test for a possible functional role of this ceptors is necessary for chemotaxis has been uncertain (39). Cell segment, site-directed mutagenesis was used. This approach indi- migration has been previously observed in the absence of calcium cated that some of the residues in this segment are important for influx, and calcium influx has been observed without chemotaxis the productive interaction of MCP-1 with CCR2B. One of the (40). Neutrophil chemotaxis toward a formylated peptide (41) and functions we tested involves the signaling elicited by MCP-1 bind- macrophage-derived chemokine-induced chemotaxis of eosino- http://www.jimmunol.org/ ing on the receptor CCR2B. Chemokines, like a large number of phils (42) were reported to occur without significant changes in 2ϩ hormones and neurotransmitters, exert their effect on cells by bind- Ca signaling. Arachidonic acid release (43) and phosphoinositol ing to a heterotrimeric G protein-coupled receptor (G␣␤␥) (36, 3-kinase activity (44), which are thought to be involved in che- 2ϩ 37). Interaction of the ligand with the seven-transmembrane re- motaxis of monocytes, were not correlated with the Ca influx ceptor alters GTP binding on the G␣ subunit, with subsequent induced by MCP-1. Reports that G␣ is not required for chemotaxis dissociation of the receptor-G protein complex and release of G␣ by Gi-coupled receptors (45) and recent observations that inhibi- and G␤␥ subunits (36). Both these complexes can then activate or tion of neutrophil chemotaxis was not accompanied by intracellu- 2ϩ inhibit signaling pathways by interacting with effectors such as lar Ca levels (46) are consistent with our results. All these re- adenylate cyclase, phospholipases, cyclic nucleotide phosphodies- sults are consistent with the idea that lamellipodium formation and by guest on September 27, 2021 ␣ chemotaxis are linked to G␤␥ function, whereas adenylate cyclase terases, and ion channels (37). The i subunit of Gi proteins is 2ϩ ␣ involved in the rapid rise in cytoplasmic calcium and the potent inhibition and Ca influx are linked to Gi . However, our obser- inhibition of adenylate cyclase. G␤␥ is an essential mediator of vation that CCR2B structural alterations by a mutation in it (F24A) 2ϩ chemotaxis (38). In the present study the rapid influx of calcium barely affects the adenylate cyclase inhibition/Ca influx, but pre- into the cytoplasm in response to MCP-1 was compared among the vents lamellipodium formation and chemotaxis, suggests that the cells transfected with the wild-type and mutant CCR2B receptors. receptor influences chemotaxis in ways other than by merely caus- ␤␥ ␣ Our data suggest that CCR2B and its mutants, T21A, T22A, and ing dissociation of G from Gi . One possibility is that the struc- F24A, elicit robust calcium influx and adenylate cyclase inhibition tural alteration affects association with a kinase necessary for che- when treated with MCP-1 in a dose-dependent manner. The other motaxis. However, such possibilities have not been tested. mutants, F23A, D25A, Y26A, and Y26F, showed neither calcium The regulation of leukocyte migration to the sites of inflamma- influx elicitation nor adenylate cyclase inhibition in response to tion involves the coordinated action of surface receptors, second MCP-1. We found a good correlation between calcium influx and messengers, and the cytoskeleton (4, 40, 47). Actin-based mor- adenylate cyclase inhibition elicited by all the receptor mutants. phological changes in various motile cells have been studied in many biological systems (48–51). Cells need to make certain types of contacts to extracellular matrix on which to crawl; therefore, the degree to which cells anchor themselves to a substrate is directly correlated with their motile activity (50, 52). It has been proposed that in the leading lamellipodia, orthogonally cross-linked actin filaments are contracted by myosin II, pulling the cell body for- ward (53). With gelsolin-null fibroblasts, evidence was presented indicating that the lamellipodia are necessarily involved in cell motility (54). Lamellipodia protrusion produced by actin polymerization that involves Rac1 activation was suggested to be required for migration of fibroblasts (55). The signaling events in chemotaxis in other cell types involve actin-based morphological FIGURE 8. Inhibition of the surface expression of CCR2B by tunicamycin. HEK293 cells transfected with CCR2B were incubated with 3.3 changes. For example, stimuli such as IL-8 (47), fMLP (56), and ␮g/ml tunicamycin for 2 days and stained with anti-Flag M2 and FITC- phosphatidic acid (57) induce actin-based morphological changes conjugated anti-mouse IgG Abs. The bold curve indicates staining in un- and chemotaxis in neutrophils. Recent results suggested that a treated (control) cells, and the light curve represents CCR2B expression in symmetrical establishment of sites of actin polymerization produce the presence of tunicamycin. directional migration of neutrophils in response to chemotactic The Journal of Immunology 5301 gradients (58). Very little is known about the role of actin poly- ifications occur in trans-Golgi, just before export of the protein to merization in MCP-1-induced chemotaxis of its target cells. Dur- the cell surface, consistent with the idea that the 50-kDa form is ing trans-endothelial migration of monocytes, polymerized actin probably surface expressed. We did not observe any decrease in and cell adhesion molecules are colocalized at the leading edge of 35S in the receptor band after treatment with N-glycosidase. This membrane protrusion (59). Our results show correlation between observation and the lack of 35S labeling of the receptor with Y26 lamellipodium formation and chemotaxis mediated by MCP-1 substitutions show that virtually all the sulfate label incorporated binding of CCR2B. They suggest that aa 24, 25, and 26 (F-D-Y), into CCR2B was located on Tyr26. The minor immunologically including the putative site of sulfation in the MCP-1 receptor, are cross-reacting component found at 38 kDa may be a degradation important for function of the receptor with respect to lamellipo- product. Since N-deglycosylation also reduced the size of this spe- dium formation and chemotaxis. However, Y26A showed the cies, it also is a glycosylated moiety. In our 35S labeling studies, highest responsiveness to MCP-1 in both morphological changes we did not find a sulfated 38-kDa species, possibly because it is and chemotaxis among all mutants tested here, suggesting that only a minor component. Since it is likely to be a minor degrada- tyrosine sulfation at position 26 is not essential in MCP-1-induced tion product, we did not pursue its identity. lamellipodium formation and cell migration in vitro. In addition, Tyrosine sulfation is probably not directly involved in lamelli- F23A, which is completely defective in induction of cytosolic podium formation and chemotaxis, as shown by our finding that Ca2ϩ influx, showed moderate responsiveness to MCP-1 in the cells expressing Y26A mutant, which is not sulfated, respond to induction of filopodia and lamellipodium formation. This finding MCP-1 binding by lamellipodium formation and are chemotaxed. suggested that the actin-mediated membrane protrusion induced by It is noteworthy, however, that Y26F was inactive in MCP-1 bind- 2ϩ MCP-1 is not regulated by the increase in cytosolic Ca . ing, signaling, lamellipodium formation, and chemotaxis, although Downloaded from It was shown in a recent paper that the chemokine receptor it was surface expressed, and its sequence was reconfirmed to be CCR5, a principal HIV-1 coreceptor, is post-translationally mod- correct. Substitution of the highly polar sulfated Y26 in the recep- ified by sulfation of tyrosines in the N-terminal part of the mole- tor with a highly hydrophobic F probably causes significant ab- cule (5). Sulfated tyrosines were shown to contribute to the binding normalities in the conformation of the N terminus that is known to of CCR5 to macrophage inflammatory protein-1␣ and -1␤ and be important for the function. Substitution with a much less hy-

HIV-1 gp120/CD4 complexes and to the ability of HIV-1 to enter drophobic A would cause much less conformational perturbation, http://www.jimmunol.org/ cells expressing CCR5 and CD4. CXCR4, another important and thus Y26A retains some of the functions of the receptor. In any HIV-1 coreceptor, was also shown to be sulfated. In the present case, the signaling involved in lamellipodium formation does not study we show that one of the N-terminal tyrosines of chemokine require Tyr 26 sulfation in CCR2B. When free MCP-1 binds to the receptor CCR2B, Y26, is sulfated, and this modiﬁcation may be receptor on the cell surface in the in vitro chemotaxis assays, sul- critical for the interaction of the receptor with MCP-1. In this fation of Tyr 26 of the receptor is not necessary, as Y26A forms segment of the receptor, there is another potential sulfation site at lamellipodia and is active in chemotaxis. In such assays, addition Y28. Since substitution of Y26 with F or A virtually abolished of heparin does not inhibit chemotaxis because the heparin bound sulfation of the receptor under experimental conditions, it is likely at the C-terminal basic amino acids (64), distal to the receptor that Y28 is not sulfated. Substitution of one of three sulfated ty- binding domains of MCP-1, does not interfere with binding to the by guest on September 27, 2021 rosines of P-selectin glycoprotein ligand-1 did not lead to lack of receptor. On the other hand, Tyr26 probably plays an important sulfation of two other tyrosines taking part in interaction with P- role in the trans-endothelial migration of monocytes in vivo. This selectin, and only substitution of all these tyrosines prevented process would involve the binding not of free MCP-1, but of binding of the ligand (60). MCP-1 that is tethered to the endothelial cells via interaction with According to the amino acid sequence of CCR2B, it should be the sulfates of the proteoglycans on the endothelial surface. We a 41-kDa protein. Previously, an Mr of 38 kDa had been reported have previously identiﬁed two basic amino acid residues in the for this receptor on the basis of SDS-PAGE (61). In our experi- C-terminal ␣-helix of MCP-1 that are involved in this ionic inter- ments two molecular species of CCR2B were detected after SDS- action (64). Probably, the sulfated N-terminal domain of the re- PAGE, the major at 50 kDa and the minor one at 38 kDa; both ceptor initially grabs the tethered MCP-1 by ionic interaction for species were visualized with anti-Flag Ab and anti-CCR2B Ab. the other extracellular interacting domains of CCR2B to complete

Another group reported a Mr of 50 kDa for chemokine receptor the ligand receptor binding that involves the other amino acid res- CXCR4, which has a theoretical Mr of 40 kDa according to the idues of MCP-1. This interaction with the other domains of amino acid sequence (62). The same Mr has been reported for CCR2B probably thermodynamically promotes transfer of the chemokine receptor CCR5 (63), which has a very similar number tethered MCP-1 to CCR2B. of amino acids with a calculated Mr of 40.6 kDa. When CCR2B Our results show that mutants of CCR2B that allow lamellipo- was indirectly immunoprecipitated with anti-Flag mAb M2, the dium formation retain chemotactic capability even when such mu- main band of receptor was clearly resolved from the heavy chain tants do not elicit detectable Ca2ϩ influx and adenylate cyclase of the mAb, which was at about 55 kDa. inhibition. Our finding that the actin-mediated membrane protru- Examination of the amino acid sequences of chemokines of CC sions induced by MCP-1 and chemotaxis are not regulated by the and CXC groups shows that most of the chemokines contain po- increase in cytosolic Ca2ϩ suggests that the biological conse- tential sites for N-glycosylation in their N-terminal segments. quences of MCP-1 binding to its receptor can be viewed as two There is only one consensus N-glycosylation site in the amino acid distinct steps 1) chemotaxis that involves lamellipodium forma- sequence of CCR2B (residues 14–17). Our results from N-glyco- tion, and 2) the other signaling events involving Ca2ϩ influx and sidase treatment of CCR2B and tunicamycin treatment of cells adenylate cyclase inhibition and subsequent events that probably expressing CCR2B very strongly suggest that this receptor con- lead to changes in gene expression resulting in activation of the tains an N-bound glycan. The loss of this glycan converted the target cells. In the recent transgenic experiments in which MCP-1 major 50-kDa receptor to a species migrating as 45 kDa on SDS production was targeted at specific tissues, chemotaxis of mono- gels. A similar shift in mobility was observed after N-glycosidase cytes/macrophages to the location of MCP-1 production was ob- treatment in the case of 35S-labeled receptor, showing that the served. However, these infiltrating cells were apparently not acti- sulfated receptor was also glycosylated. It is known that both mod- vated to cause any pathology (65–67), except in one case where 5302 SULFATION AND GLYCOSYLATION OF CCR2B targeted expression of MCP-1 in the cardiomyocytes caused heart coupled receptors for human rantes and monocyte chemoattractant protein-1. failure in older animals as a result of the events initiated by the J. Exp. Med. 177:851. 25. Munson, P. J., and D. Robard. 1980. Ligand: a versatile computerized approach infiltrating monocytes (68). The general conclusion from the trans- for characterization of ligand-binding systems. Anal. Biochem. 107:220. genic experiments has been that MCP-1 causes infiltration of target 26. Hung, D. T., Y. H. Wong, T.-K. H. Vu, and S. R. Coughlin. 1992. The cloned cells, but their activation requires additional signals. The molecu- platelet thrombin receptor couples to at least 2 distinct effectors to stimulate phosphoinositide hydrolysis and inhibit adenylyl cyclase. J. Biol. Chem. 267: lar events initiated in the monocytes/macrophages as a result of 20831. MCP-1 binding, the nature of the additional signals that are re- 27. Solomon, Y., C. Loudos, and M. Rodbell. 1974. A highly sensitive adenylate cyclase assay. Anal. Biochem. 58:541. quired to trigger activation, and the molecular events involved in 28. Myers, S. J., L. M. Wong, and I. F. Charo. 1995. Signal-transduction and ligand this activation remain to be elucidated. specificity of the human monocyte chemoattractant protein-1 receptor in transfected embryonic kidney-cells. J. Biol. Chem. 270:5786. 29. Wulf, E., A. Deboben, F. A. Bautz, H. Faulstich, and T. H. Weiland. 1979. Acknowledgments Fluorescent phalloidin, a tool for the visualization of cellular actin. Proc. Natl. We thank Drs. Cliff Beall and Janice Liu for assistance with preliminary Acad. Sci. USA 76:4498. experiments. We thank Dr. I. Charo for providing us with the prolactin 30. Frade, J. M. R., M. Mellado, G. del Real, J. C. Gutierrez-Ramos, P. Lind, and leader construct and Dr. M. Mellado for providing us with CCR2 Abs. C. Martinez-A. 1997. Characterization of the CCR2 chemokine receptor: functional CCR2 receptor expression in B cells. J. Immunol. 159:5576. 31. Monteclaro, F. S., and I. F. Charo. 1996. The amino-terminal extracellular do- References main of the MCP-1 receptor, but not the RANTES/MIP-1␣ receptor, confers 1. Murphy, P. M. 1994. The molecular biology of leukocyte chemoattractant recep- chemokine selectivity: evidence for a two-step mechanism for MCP-1 receptor tors. Annu. Rev. Immunol. 12:593. activation. J. Biol. Chem. 271:19084. 2. Rollins, B. J. 1997. Chemokines. Blood 90:909. 32. Monteclaro, F. S., and I. F. Charo. 1997. The amino-terminal domain of CCR2

3. Luster, A. D. 1998. Chemokines: chemotactic cytokines that mediate inflamma- is both necessary and sufficient for high affinity binding of monocyte chemoat- Downloaded from tion. N. Engl. J. Med. 338:436. tractant protein 1: receptor activation by a pseudo-tethered ligand. J. Biol. Chem. 4. Baggiolini, M. 1998. Chemokines and leukocyte traffic. Nature 329:565. 272:23186. 5. Farzan, M., T. Mirzabekov, P. Kolchinsky, R. Wyatt, M. Cayabyab, N. P. Gerard, 33. Kuhn, D. E., C. J. Beall, and P. E. Kolattukudy. 1995. The cytomegalovirus US28 C. Gerard, J. Sodroski, and H. Choe. 1999. Tyrosine sulfation of the amino protein binds multiple CC-chemokines with high-affinity. Biochem. Biophys. Res. terminus of CCR5 facilitates HIV-1 entry. Cell 96:667. Commun. 211:325. 6. Huttner, W. B. 1988. Tyrosine sulfation and the secretory pathway. Annu. Rev. 34. Bodaghi, B., T. R. Jones, D. Zipeto, C. Vita, L. Sun, L. Laurent, Physiol. 50:363. F. Arenzana-Seisdedos, J. L. Virelizier, and S. Michelson. 1998. Chemokine 7. Huttner, W. B. 1984. Determination and occurrence of tyrosine O-sulfate in pro- sequestration by viral chemoreceptors as a novel viral escape strategy: with- teins. Methods Enzymol. 107:200. drawal of chemokines from the environment of cytomegalovirus-infected cells. http://www.jimmunol.org/ 8. Ouyang, Y.-B., W. S. Lane, and K. L. Moore. 1998. Tyrosylprotein sulfotrans- J. Exp. Med. 188:855. ferase: purification and molecular cloning of an enzyme that catalyzes tyrosine 35. Billstrom, M. A., G. L. Johnson, N. J. Avdi, and G. S. Worthen. 1998. Intracel- O-sulfation, a common posttranslational modification of eukaryotic proteins. lular signaling by the chemokine receptor US28 during human cytomegalovirus Proc. Natl. Acad. Sci. USA 95:2896. infection. J. Virol. 72:5535. 9. Beisswanger, R., D. Corbeil, C. Vannier, C. Thiele, U. Dohrman, R. Kellner, 36. Hamm, H. E. 1998. The many faces of G protein signaling. J. Biol. Chem. 273: K. Ashman, C. Niehrs, and W. B. Huttner. 1998. Existence of distinct tyrosyl- 669. protein sulfotransferase genes: molecular characterization of tyrosylprotein sul- 37. Berman, D. M., and A. G. Gilman. 1998. Mammalian RGS proteins: barbarians fotransferase-2. Proc. Natl. Acad. Sci. USA 95:11134. at the gate. J. Biol. Chem. 273:1269. 10. Ouyang, Y.-B., and K. L. Moore. 1998. Molecular cloning and expression of 38. Neptune, E. R., and H. R. Bourne. 1997. Receptors induce chemotaxis by re- ␤␥ human and mouse tyrosylprotein sulfotransferase-2 and a tyrosylprotein sulfo- leasing the subunit of Gi, not by activating Gq or Gs. Proc. Natl. Acad. Sci. transferase homologue in Caenorhabditis elegans. J. Biol. Chem. 273:24770.

USA 94:14489. by guest on September 27, 2021 11. Bundgaard, J. R., J. Vuust, and J. F. Rehfeld. 1997. New consensus features for 39. Sozzani, S., M. Locati, D. Zhou, M. Rieppi, W. Luini, G. Lamorte, G. Bianchi, tyrosine O-sulfation determined by mutational analysis. J. Biol. Chem. N. Polentarutti, P. Allavena, and A. Mantovani. 1995. Receptors, signal trans- 272:21700. duction, and spectrum of action of monocyte chemotactic protein-1 and related 12. Stone, S. R., and J. Hofsteenge. 1986. Kinetics of the inhibition of thrombin by chemokines. J. Leukocyte Biol. 57:788. hirudin. Biochemistry 25:4622. 40. Needham, M., N. Sturgess, G. Cerillo, I. Green, H. Warburton, R. Wilson, 13. Hofsteenge, J., S. R. Stone, A. Donella-Deana, and L. A. Pinna. 1990. The effect L. Martin, D. Barratt, M. Anderson, C. Reilly, et al. 1996. Monocyte chemoat- of substituting phosphotyrosine for sulphotyrosine on the activity of hirudin. Eur. tractant protein-1: receptor interactions and calcium signaling mechanisms. J. Biochem. 188:55. J. Leukocyte Biol. 60:793. 14. Leyte, A., H. B. van Schijndel, C. Niehrs, W. B. Huttner, M. P. Verbeet, 41. Laffafian, I., and M. B. Hallett. 1995. Does cytosolic free Ca2ϩ signal neutrophil K. Mertens, and J. A. van Mourik. 1991. Sulfation of Tyr1680 of human blood chemotaxis in response to formylated chemotactic peptide? J. Cell Sci. 108:3199. coagulation factor VIII is essential for the interaction of factor VIII with von 42. Bochner, B. S., C. A. Bickel, M. L. Taylor, D. W. MacGlashan, Jr., P. W. Gray, Willebrand factor. J. Biol. Chem. 266:740. C. J. Raport, and R. Godiska. 1999. Macrophage-derived chemokine induces 15. Pouyani, T., and B. Seed. 1995. PSGL-1 recognition of P-selectin is controlled by human eosinophil chemotaxis in a CC chemokine receptor 3- and CC chemokine a tyrosine sulfation consensus at the PSGL-1 amino terminus. Cell 83:333. receptor 4-independent manner. J. Allergy Clin. Immunol. 103:527. 16. Wilkins, P. P., K. L. Moore, R. P. McEver, and R. D. Cumming. 1995. Tyrosine 43. Locati, M., D. Zhou, W. Luini, V. Evangelista, A. Mantovani, and S. Sozzani. sulfation of P-selectin glycoprotein ligand-1 is required for high affinity binding 1994. Rapid induction of arachidonic acid release by monocyte chemotactic pro- to P-selectin. J. Biol. Chem. 270:22677. tein-1 and related chemokines: role of Ca2ϩ influx, synergism with platelet-ac- 17. Li, Y.-S., Y.-J. Shyy, J. G. Wright, A. J. Valente, J. F. Cornhill, and tivating factor and significance for chemotaxis. J. Biol. Chem. 269:4746. P. E. Kolattukudy. 1993. The expression of monocyte chemotactic protein 44. Yen, H.-H., Y. Zhang, S. Penford, and B. J. Rollins. 1997. MCP-1-mediated (MCP-1) in human vascular endothelium in vitro and in vivo. Mol. Cell. Biochem. chemotaxis requires activation of non-overlapping signal transduction pathways. 126:61. J. Leukocyte Biol. 61:529. 18. Beall, C. J., S. Mahajan, D. E. Kuhn, and P. E. Kolattukudy. 1996. Site-directed 45. Neptune, E. R., T. Iiri, and H. R. Bourne. 1999. G␣ is not required for chemo- mutagenesis of monocyte chemoattractant protein-1 identifies two regions of the i taxis mediated by G -coupled receptors. J. Biol. Chem. 274:2824. polypeptide essential for biological activity. Biochem. J. 313:633. i 19. Charo, I. F., S. J. Myers, A. Herman, C. Franci, A. J. Connolly, and 46. Konishi, K., M. Urada, I. Adachi, and T. Tanaka. 2000. Inhibitory effect of S. R. Coughlin. 1994. Molecular cloning and functional expression of two mono- sanguiin H-11 on chemotaxis of neutrophil. Biol. Pharm. Bull. 23:213. cyte chemoattractant protein 1 receptors reveals alternative splicing of the car- 47. Sham, R. L., P. D. Phatak, T. P. Ihne, C. N. Abboud, and C. H. Packman. 1993. boxyl-terminal tails. Proc. Natl. Acad. Sci. USA 91:2752. Signal pathway regulation of interleukin-8-induced actin polymerization in neu- 20. Franci, C., J. Gosling, C.-L, Tsou, S. R. Coughlin, and I. F. Charo. 1996. Phos- trophils. Blood 82:2546. phorylation by a G protein-coupled kinase inhibits signaling and promotes inter- 48. Hall, A. 1998. Rho GTPases and the actin cytoskeleton. Science 279:509. nalization of the monocyte chemoattractant protein-1 receptor: critical role of 49. Parent, C. A., and P. N. Devreotes. 1999. A cell’s sense of direction. Science carboxyl-tail serines/threonines in receptor function. J. Immunol. 157:5606. 284:765. 21. Prickett, K. S., D. C. Amberg, and T. P. Hopp. 1989. A calcium-dependent 50. Small, J. V., K. Rottner, and I. Kaverina. 1999. Functional design in the actin antibody for identification and purification of recombinant proteins. BioTech- cytoskeleton. Curr. Opin. Cell Biol. 11:54. niques 7:580. 51. Waterman-Storer, C. M., and E. D. Salmon. 1999. Positive feedback interactions 22. Higuchi, R., B. Krummel, and R. K. Saiki. 1988. A general method of in vitro between microtubule and actin dynamics during cell motility. Curr. Opin. Cell preparation and specific mutagenesis of DNA fragments: study of protein and Biol. 11:61. DNA interactions. Nucleic Acids Res. 16:7351. 52. Huttenlocher, A., R. R. Sandborg, and A. F. Horwitz. 1995. Adhesion in cell 23. Hunter, W. M., and F. C. Greenwood. 1962. Preparation of iodine-131 labeled migration. Curr. Opin. Cell Biol. 7:697. human growth hormone of high specific affinity. Nature 194:495. 53. Svitkina, T. M., A. V. Verkhovsky, K. M. McQuade, and G. G. Borisy. 1997. 24. Van Riper, G., S. Siciliano, P. A. Fischer, R. Meurer, M. S. Springer, and Analysis of the actin-myosin II system in fish epidermal keratocytes: mechanism H. Rosen. 1993. Characterization and species distribution of high-affinity GTP- of cell body translocation. J. Cell Biol. 139:397. The Journal of Immunology 5303

54. Azuma, T., W. Witke, T. P. Stossel, J. H. Hartwig, and D. J. Kwiatkowski. 1998. 63. Zhao, J., L.-H. Ben, Y.-L. Wu, W. Hu, K. Ling, S.-M. Xin, H.-L. Nie, L. Ma, and Gelsolin is a downstream effector of rac for fibroblast motility. EMBO J. 17:1362. G. Pei. 1999. Anti-HIV agent trichosanthin enhances the capabilities of chemo- 55. Waterman-Storer, C. M., R. A. Worthylake, B. P. Liu, K. Burridge, and kines to stimulate chemotaxis and G protein activation, and this is mediated E. D. Salmon. 1999. Microtubule growth activates Rac1 to promote lamellipodial through interaction of trichosanthin and chemokine receptors. J. Exp. Med. 190: protrusion in fibroblasts. Nat. Cell. Biol. 1:45. 101. 56. Arcaro, A., and M. P. Wymann. 1993. Wortmannin is a potent phosphatidylino- 64. Chakravarty, L., L. Rogers, T. Quach, S. Breckenridge, and P. E. Kolattukudy. sitol 3-kinase inhibitor: the role of phosphatidylinositol 3,4,5-trisphosphate in 1998. Lysine 58 and histidine 66 at the C-terminal ␣-helix of monocyte che- neutrophil responses. Biochem. J. 296:297. moattractant protein-1 are essential for glycosaminoglycan binding. J. Biol. 57. Siddiqui, R. A., and D. English. 1997. Phosphatidic acid elicits calcium mobili- Chem. 273:29641. zation and actin polymerization through a tyrosine kinase-dependent process in human neutrophils: a mechanism for induction of chemotaxis. Biochim. Biophys. 65. Fuentes, M. E., S. K. Durham, M. R. Swergel, A. C. Lewin, D. C. Barton, Acta 1349:81. J. R. Megill, R. Bravo, and S. A. Lira. 1995. Controlled recruitment of monocytes 58. Weiner, O. D., G. Servant, M. D. Welch, T. J. Mitchison, J. W. Sedat, and and macrophages to specific organs through transgenic expression of monocyte H. R. Bourne. 1999. Spatial control of actin polymerization during neutrophil chemoattractant protein-1. J. Immunol. 155:5769. chemotaxis. Nat. Cell. Biol. 1:75. 66. Grewal, I. S., B. J. Rutledge, J. A. Fiorillo, L. Gu, R. P. Gladue, R. A. Flavell, 59. Sandig, M., E. Negrou, and K. A. Rogers. 1997. Changes in the distribution of and B. J. Rollins. 1997. Transgenic monocyte chemoattractant protein-1 LFA-1, catenins and F-actin during transendothelial migration of monocytes in (MCP-1) in pancreatic islets produces monocyte-rich insulitis without diabe- culture. J. Cell Sci. 110:2807. tes: abrogation by a second transgene expressing systemic MCP-1. J. Immunol. 60. Liu, W.-J., V. Ramachandran, J. Kang, T. K. Kishimoto, R. D. Cummings, and 159:401. R. P. McEver. 1998. Identification of N-terminal residues on P-selectin glycop- 67. Gunn, M. D., N. A. Helken, X. Liao, and L. T. Williams. 1997. Monocyte che- rotein ligand-1 required for binding to P-selectin. J. Biol. Chem. 273:7078. 61. Rodriguez-Frade, J. M., A. J. Vila-Coro, A. M. de Ana, J. P. Albar, moattractant protein-1 is sufficient for the chemotaxis of monocytes and lympho- C. Matrinez-A., and M. Mellado. 1999. The chemokine monocyte chemoattrac- cytes in transgenic mice but requires an additional stimulus for inflammatory tant protein-1 induces functional responses through dimerization of its receptor activation. J. Immunol. 158:376. CCR2. Proc. Natl. Acad. Sci. USA 96:3628. 68. Kolattukudy, P. E., T. Quach, S. Bergese, S. Breckenridge, J. Hensley, 62. Zhou, H., and H.-H. Tai. 1999. Characterization of recombinant human CXCR4 R. Altschuld, G. Gordillo, S. Klenotic, C. Orosz, and J. Parker-Thornburg. 1998. Downloaded from in insect cells: role of extracellular domains and N-glycosylation in ligand bind- Myocarditis induced by targeted expression of the MCP-1 gene in murine cardiac ing. Arch. Biochem. Biophys. 369:267. muscle. Am. J. Pathol. 152:101. http://www.jimmunol.org/ by guest on September 27, 2021