CX3CR1 Are Potent HIV Coreceptors Two Novel Fully Functional

Two Novel Fully Functional Isoforms of CX3CR1 Are Potent HIV Coreceptors Alexandre Garin, Nadine Tarantino, Sophie Faure, Mehdi Daoudi, Cédric Lécureuil, Anne Bourdais, Patrice Debré, This information is current as Philippe Deterre and Christophe Combadiere of October 2, 2021. J Immunol 2003; 171:5305-5312; ; doi: 10.4049/jimmunol.171.10.5305 http://www.jimmunol.org/content/171/10/5305 Downloaded from

References This article cites 49 articles, 26 of which you can access for free at: http://www.jimmunol.org/content/171/10/5305.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 2, 2021

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 © 2003 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Two Novel Fully Functional Isoforms of CX3CR1 Are Potent HIV Coreceptors1

Alexandre Garin, Nadine Tarantino, Sophie Faure, Mehdi Daoudi, Ce´dric Le´cureuil, Anne Bourdais, Patrice Debre´, Philippe Deterre, and Christophe Combadiere2

We identified two novel isoforms of the human chemokine receptor CX3CR1, produced by alternative splicing and with N- terminal regions extended by 7 and 32 aa. Expression of the messengers coding these isoforms, compared with that of previously described V28 messengers, is lower in monocytes and NK cells, but higher in CD4؉ T lymphocytes. CX3CR1 and its extended isoforms were expressed in HEK-293 cells and compared for expression, ligand binding, and cellular responses. In steady state experiments, all three CX3CR1 isoforms bound CX3CL1 with similar affinity. In kinetic binding studies, however, kon and koff were significantly greater for the extended CX3CR1 isoforms, thereby suggesting that the N-terminal extensions may alter the functions induced by CX3CL1. In signaling studies, all three CX3CR1 isoforms mediated agonist-dependent calcium mobilization, Downloaded from but the EC50 was lower for the extended than for the standard isoforms. In addition, chemotactic responses for these extended isoforms shifted left, also indicating a more sensitive response. Finally, the longer variants appeared to be more potent HIV coreceptors when tested in fusion and infection assays. In conclusion, we identified and characterized functionally two novel isoforms of CX3CR1 that respond more sensitively to CX3CL1 and HIV viral envelopes. These data reveal new complexity in CX3CR1 cell activation and confirm the critical role of the N-terminal domain of the chemokine receptors in ligand recognition and cellular response. The Journal of Immunology, 2003, 171: 5305–5312. http://www.jimmunol.org/

hemokines are involved in diverse physiological and CCR9 isoforms, termed CCR9A and CCR9B, which differ by 12 pathological conditions, mainly through their deploy- aa residues, differ in their ligand-binding affinity, Ca2ϩ flux, and C ment of leukocytes (1–3). They mediate their effects via chemotaxis (17). Similar modifications were observed for CXCR4- specific interactions with extracellular regions of chemokine re- lo, an N-terminal protein isoform of CXCR4 (18). Together these ceptors expressed at the leukocyte cell surface. The N-terminal data suggest that N-terminal isoforms of chemokine receptors can domain of these receptors has been described as a crucial deter- alter biochemical functions. Post-transcriptional modifications in minant in ligand binding and signaling (4–7), and thus alterations the carboxyl-terminal regions have also been reported to affect the by guest on October 2, 2021 in this domain may be of critical importance. Furthermore, the regulation of chemokine receptor functions (19). N-terminal domain plays a key role in interacting with the HIV We previously reported allelic isoforms of CX3CR1 associated envelope gp120 (8–10), at least for CCR5 and CXCR4, the two with rapid HIV disease progression (20) and reduced susceptibility principal HIV coreceptors (11–13). Abs raised against the N ter- to acute coronary events (21). CX3CR1 is the receptor for minals of CXCR4 and CCR5 inhibit cell fusion and infection with CX3CL1, the only member of the CX3C subfamily (22). CX3CR1 HIV-1 (8, 14), and the first 20 N-terminal residues of CCR5 func- is preferentially expressed in monocytes and NK cells (23) as well tion as a coreceptor when grafted onto a CCR2 backbone (15, 16). ϩ as in CD8 T cells (24), glial cells and astrocytes (25) and may A current model of chemokine binding suggests a two-step mech- directly modulate neuronal activity and survival (26, 27). CX3CL1 anism involving initial recognition of the chemokine receptor N- and the chemokine CXCL16 (28) exist in both soluble and mem- terminal domain, followed by more specific interactions with the other extracellular domains. Recent findings of post-transcriptional brane-anchored forms. The soluble form of CX3CL1 recruits lym- modifications in the N-terminal domain of chemokine receptors phocytes and monocytes (22, 29), whereas membrane-bound indicate that regulation of receptor function is still more complex. CX3CL1 directly mediates the capture and firm adhesion of CX3CR1-expressing leukocytes (23, 30); this discovery revealed a novel pathway for leukocyte activation. Recent reports point out physiopathological correlates to these functions: CX3CL1 knock- Laboratoire d’Immunologie Cellulaire et Tissulaire, Institut National de la Santeétde out mice are less susceptible than their wild-type littermates to la Recherche Me´dicale, Unite´543, Hoˆpital Pitie´-Salpeˆtriere, Paris, France cerebral ischemia-reperfusion injury (31), while CX3CR1 KO Received for publication October 25, 2002. Accepted for publication September 16, 2003. mice reject cardiac transplants more slowly in the presence of cy- The costs of publication of this article were defrayed in part by the payment of page closporine (32) and form fewer atherosclerotic lesions in an apo- charges. This article must therefore be hereby marked advertisement in accordance lipoprotein E background (33, 34). CX3CR1 plays a role in yet with 18 U.S.C. Section 1734 solely to indicate this fact. another disease; like CCR5 and CXCR4, it serves as an HIV co- 1 This work was supported by a grant from the Agence National de Recherche sur le receptor (35, 36). CX3CL1 specifically blocks CX3CR1 interaction SIDA and an Action Concertée Incitative grant for Jeunes Chercheurs from the French Ministry of Research. A.G. and S.F. are fellows of the Agence National de with the HIV envelope gp120 (37). The physiological relevance of Recherche sur le SIDA, and M.D. is a fellow of the Ministry of Research. CX3CR1 in HIV nonetheless remains unclear (20, 38, 39). 2 Address correspondence and reprint requests to Dr. Christophe Combadiere, Labo- Here we identify and functionally characterize two novel ratoire d’Immunologie Cellulaire et Tissulaire, Institut National de la Santeétdela Recherche Me´dicale, Unite´543, Hoˆpital Pitie´-Salpeˆtriere, 91 boulevard de l’Hoˆpital, CX3CR1 isoforms that differ from the classic form of CX3CR1 (re- AP-HP, 75634 Paris Cedex 13, France. E-mail address: [email protected] ferred to here as Met ϩ1) by N-terminal additions of 7 and 32 aa.

Materials and Methods Generation of specific Met Ϫ32 isoform polyclonal Abs Cells and culture conditions Rabbit polyclonal Abs were raised against the human Met Ϫ7 and Met Ϫ32 CX3CR1 isoforms (AGRO-BIO, La Ferte´ Saint Aubin, France) and Resting PBMCs of heparinized venous blood taken from healthy volun- further purified. teers were isolated by a one-step centrifugation on a Ficoll separating so- ϩ ϩ lution (Biochrom, Berlin, Germany). CD4 T lymphocytes, CD8 T lym- Western blot analysis of CX3CR1 protein phocytes, and NK cells were then isolated by positive selection through the use, respectively, of anti-CD4 mAbs, anti-CD8 mAbs, and a mixture of HEK-293 cells were suspended at 108 cells/ml in 20 mM Tris (pH 7.5), 1 anti-CD16 and anti-CD56 mAbs coupled with magnetic beads (Miltenyi mM EDTA, and 1 mM DTT supplemented with complete protease inhib- Biotec, Paris, France). Negative selection, using a mixture of anti-CD2, itor cocktail (Roche, Meylan, France). The cells were then homogenized at -CD7, -CD16, -CD19, and -CD56 mAbs coupled with magnetic beads 4°C in a Dounce homogenizer (Kontes, Vineland, NJ). Nuclear and cellular (Dynal, Compiegne, France), was used to enrich monocytes. Flow cyto- debris was removed by centrifugation for 10 min at 10,000 ϫ g. The metric analysis confirmed that the enriched cell populations were 95% supernatant was then centrifuged at 100,000 ϫ g at 4°C for 30 min. The pure. U373MG-CD4, HeLa, and HEK-293 cell lines were maintained in pellet was suspended in lysis buffer (20 mM Tris, 140 mM NaCl, 1 mM DMEM (Life Technologies SARL, Cergy Pontoise, France), supplemented EDTA, and 0.1% Nonidet P-40, pH 7.4) supplemented with complete pro- with 10% FCS, 1 mM L-glutamine, and penicillin-streptomycin (100 tease inhibitor cocktail. The samples were then assayed for protein content U/ml). Cells used for HIV fusion and infection were provided by Dr. M. and resuspended in sample buffer (50 mM Tris (pH 7), 3% SDS, 10% Alizon (Institut National de la Santé et de la Recherche Médicale, Unité glycerol, 5% 2-ME, and bromophenol blue). Proteins were separated by 567, Universite Paris V-Rene Descartes, Paris, France). The U373MG- SDS-PAGE with Porzio and Pearson’s method (40). Acrylamide gels CD4 cells express the CD4 receptor and the lacZ gene driven by the long (10%) were electrotransferred to Hybond-P nitrocellulose membrane (Am- terminal repeat element, HeLa-EnvLAI cells express HIV-1 LAI, HeLa- ersham Pharmacia Biotech, Orsay, France), and the blots were probed with EnvADA cells express HIV-1 ADA envelope glycoproteins, and both ex- either polyclonal Abs raised against a peptide corresponding to aa 2–21 of press HIV-1 trans-activator Tat. the human CX3CR1 (Chemicon International, Temecula, CA), in accor- Downloaded from dance with the manufacturer’s instructions or with our polyclonal Abs Ϫ Ϫ RNA isolation and cDNA synthesis raised against the human Met 7orMet 32 isoform. For detection, we used HRP-conjugated goat anti-mouse IgG (Bio-Rad, Marnes-la-Coquette, Total RNA was extracted with the RNeasy Kit (Qiagen, Hilden, Germany) France) and an ECL detection system (Amersham Pharmacia Biotech, Pis- from 107 CD4ϩ T lymphocytes, monocytes, and NK cells from two healthy cataway, NJ), in accordance with the manufacturer’s instructions, on Curix blood donors. Total RNA of 107 CD8ϩ T lymphocytes was extracted from Blue x-ray film (Agfa, Mortsel, Belgium). two other healthy blood donors. Treatment with DNase I (30 U) prevented any genomic DNA contamination. Samples were then heated to denature Radioligand competition CX3CR1 binding assay http://www.jimmunol.org/ ␮ DNase I. For each sample 1 g of total RNA was reverse transcribed for Binding assays used [125I]CX3CL1 (Amersham Pharmacia Biotech) in du- 1hat37°C with an RT-PCR kit (Stratagene, La Jolla, CA). plicate with 5 ϫ 104 CX3CR1-expressing HEK cells as previously described (37). Briefly, cells were incubated in a total volume of 200 ␮lof SYBR-Green quantitative PCR PBS containing 1 mg/ml BSA and 0.01% azide (pH 7.4) with 50 pM [125I]CX3CL1 and with or without increasing concentrations of unlabeled The SYBR-Green PCR kit (Applied Biosystems, Foster City, CA) was human CX3CL1 (PeproTech, Rocky Hill, NJ) or 50 nM unlabeled chemo- used according to the manufacturer’s instructions in a model 7700 se- kines (CCL2, CCL5, CCL20, CCL21, CXCL1, CXCL8, CXCL12, and quence detector (Applied Biosystems). For each PCR, 150 nM forward v-macrophage inflammatory protein-II; PeproTech). Azide was added to (clone10-SP1, AGAGCTCTCTGGCTTCTGGGTGGAGAA; V28-SP1, prevent ligand-mediated internalization of CX3CR1. After2hat37°C,

TGGCTGACTGGCAGATCCAGAGGTT) and reverse (RACE-SP1, GGC- unbound chemokines were separated from cells by centrifugation in 1 ml by guest on October 2, 2021 CAATGGCAAAGATGACGGAGTA) primers were used to amplify 1 nM of PBS complemented with 10% sucrose. Gamma emissions were then cDNA. Fluorescence was detected at the end of each elongation phase. counted in the cell pellet. For association studies, cells were incubated with [125I]CX3CL1 for increasing periods of time and washed. For dissociation CX3CR1 isoform constructs studies, cells were incubated for 2 h and washed to remove excess [125I]CX3CL1. Cells were then incubated in PBS/BSA/azide and harvested The open reading frame (ORF)3 corresponding to the CX3CR1 protein iso- after various periods of incubation. After washing, cells were processed for forms Met ϩ1, Met Ϫ7, and Met Ϫ32 were produced from cDNA specific radioactivity measurements. from clone 10 mRNA. Forward primers, designed to be specifictothe5Ј end of each isoform, were HindIII-tailed in the 5Ј end (ϩ32 clone, GCGCATATA Intracellular calcium measurements ϩ AGCTTGCCACCATGAGAGAACCCCTGGAGGCGT; 7 clone, GCG Intracytoplasmic free calcium was measured with Fura-2/AM (Molecular CATATAAGCTTGCCACCATGGCCAGTGGGGCCTTCAC). The reverse Probes, Leiden, The Netherlands). HEK-293 cells (3 ϫ 106) were washed Ј primer, common to all isoforms, was chosen to be specifictothe3 end of the once and loaded for 45 min at 37°C in the dark with 2 ␮M Fura-2/AM and Ј CX3CR1 ORF and was XhoI-tailed in its 5 end (LT3-CX3CR1, GCG 2 ␮M pluronic acid in 1 ml of HBSS buffer supplemented with 10 mM CATATAAGCTTGCCACCATGGATCAGTTCCCTGAATCAGTG). The HEPES, 0.5 mM MgCl2, and 1 mM CaCl2. Cells were centrifuged, and the PCR conditions were as follows: denaturation at 95°C for 30 s, annealing at pellet was resuspended in 2 ml of the same buffer and transferred to a 60°C for 30 s, and extension at 72°C for 1 min. The PCR products were then quartz cuvette for reading. CX3CL1 was added to the 2-ml cell volume at digested with HindIII/XhoI, subcloned into the mammalian expression vector ϩ various concentrations. Fluorescence was monitored with a spectroflu- pCDNA3.1 (Invitrogen, Leek, The Netherlands), and then sequenced onto orometer (SAFAS, Monaco) in cuvettes thermostatically controlled at both strands with the BigDye Terminator kit (Applied Biosystems, War- 37°C and stirred continuously. The cell suspension was excited alternately rington, U.K.). at 340 and 380 nm, and fluorescence was measured at 510 nm. Ten-nanom- eter slit widths were used for both excitation and emission. Graphic rep- Stable expression of CX3CR1 isoforms by HEK-293 cells resentations of intracellular calcium concentrations were computed with 2ϩ ϭ ϫ Ϫ Ϫ ϫ the equation: Ca concentration 225 (R )/(Rmax R) Sf380/Sb380, Two micrograms of each expression vector were transfected by 6 ␮gof previously determined by Grynkiewicz et al. (41). R was the ratio of the Transfast (Promega, Charbonnieres, France) into 106 HEK-293 cells grown fluorescence measured at the 340 and 380 nm excitations. R was as- ␮ max in DMEM. Two days later, transfected cells were selected with 1.5 g/ml sessed by lysing the cells with 0.5% Triton X-100 for Rmax, and Rmin was G-418 (Life Technologies, Gaithersburg, MD), and cell clones were iso- determined by adding the excess EGTA. Sb380 and Sf380 were the fluores- lated and tested for CX3CR1 expression by FACS analysis with FITC- cence levels with 380 nm excitation, both determined under the same anti-CX3CR1 Ab (MBL, Nagoya, Japan). Clones expressing each isoform conditions. protein were maintained in DMEM containing 0.5 ␮g/ml G-418. Three stable clones of each CX3CR1 isoform were independently assayed in Chemotactic activity of CX3CL1 on CX3CR1-transfected further functional assays. HEK-293 cells Chemotaxis was assayed in a 96-well chemotaxis chamber with a filter porosity of 10 ␮m (NeuroProbe, Cabin John, MD). HEK-293 cells were 3 Abbreviation used in this paper: ORF, open reading frame. washed twice with PBS, resuspended in serum-free RPMI 1640 containing The Journal of Immunology 5307

0.1% BSA, then labeled for 30 min at 37°C with CFSE (Molecular Probes) in PBS. Cells were then washed in PBS and resuspended in RPMI 1640 medium (106 cells/ml); 60 ␮l of cell suspension was then loaded onto the filter. A final volume of 28 ␮l of medium with various concentrations of CX3CL1 was placed in the lower chamber. The 96-well plate was incubated for 2–3hat37°C in 100% humidity and 5% CO2. The filter-top surface was rinsed with PBS, and the plate was centrifuged for 2 min at 1500 rpm. Fluorescence was measured with a Packard Fusion microplate analyzer (PerkinElmer, Boston, MA). Fusion and infection assays Fusion and infection assays were performed as previously described by Sol et al. (42). Briefly, for both assays, 2 ϫ 105 U373MG-CD4 cells were transiently transfected in DMEM with 1 ␮g of each CX3CR1 isoform and 3 ␮g of Transfast (Promega, Charbonieres, France). After 48 h, 5 ϫ 104 of these cells were cocultured with HeLa-P4.2-EnvLAI and HeLa-P4.2-En- ϫ 4 vADA at a 1:1 ratio for the fusion assay. At the same time, 5 10 of these cells were infected in DMEM (DEAE-dextran, 60 ␮g/ml) for 3 h with 10 ng of LAI, VIA, or V7 virus (respectively, R5-tropic, X4-tropic, and triple- tropic) and then extensively washed in DMEM to remove adsorbed virus. Twenty-four hours later, cells were lysed with 100 ␮l of lysis buffer (5 mM ␤ MgCl2 and 0.1% Nonidet P-40), and chlorophenolred- -D-galactopyrano-

side (Calbiochem-Novabiochem, Bad Soden, Germany) was added at a Downloaded from ﬁnal concentration of 600 mM to assess ␤-galactosidase activity. This reading was made at an excitation length of 570 nm and an emission length of 690 nm, with an Emax precision microplate reader (Molecular Devices, Wokingham, U.K.). For all experiments, the low background signal obtained with target cells transfected by pCDNA3 was subtracted from the signal reported. Results were also standardized according to CX3CR1 expression, measured as previously described, to estimate the total number of

CX3CL1 binding sites. http://www.jimmunol.org/ Statistical analysis Data handling, analysis, and graphic representation were performed using PRISM 2.01 (GraphPad Software, San Diego, CA). Curve regression and analysis used nonlinear regression analysis. Statistical analyses assessed the means by paired two-sample t tests. FIGURE 1. Identification and cell distribution of a new CX3CR1 mes- Ј Results senger. A,5 sequence and deduced translation of CX3CR1 clone 10 messenger. Uppercase letters represent the potential translated mRNA se- Expression of novel human CX3CR1 transcripts quence. Methionine codons are in black boxes, and Kozak consensus by guest on October 2, 2021 We previously identified three CX3CR1 transcripts that differ only sequences are underlined. The positions of codons are indicated above, in their 5Ј-untranslated regions and demonstrated that they are con- starting at the methionine of the classic form of CX3CR1 (Met ϩ1). The trolled by three independent promoters (43). Interestingly, one of amino acid charge is indicated under its respective codon. B, Comparison the three messengers, clone 10, contains two other in-frame AUG of V28 and clone 10 messenger expression by SYBR-Green quantitative PCR analysis. Total RNA extracted from various cell populations was reverse codons that can extend the protein by 7 or 32 aa (Fig. 1A). The first Ϫ transcribed, and cDNA products were subjected to specific PCR amplification. AUG codon is flanked by an A in 3, essential for the 40S ribo- Data are expressed as the level of V28 and clone 10 messenger expression after somal subunit trapping and for fixation of the 60S ribosomal sub- normalization based upon V28 messenger expression within NK cells, accord- unit. Downstream, the AUG is flanked by an A in ϩ4, which is less (C NK-V28-C X) ϫ ing to the following calculations: 2e T T 100, where CT repre- optimal for initiating protein translation than a G (44). This context sents the cycle at which a significant increase in SYBR-Green signal intensity is reported to be favorable for leaky scanning, so that initiation of is first detected, and X is the messenger within a subpopulation tested. Data are translation may also occur at the downstream AUG codon. The the mean Ϯ SEM of three independent experiments with cDNA from four second AUG conforms to a fully functional Kozak consensus se- healthy blood donors (N1 to N4). quence (A in Ϫ3andGinϩ4), which should preclude leaky scanning. Finally, the third AUG, which corresponds to the initiating AUG previously described for the classic CX3CR1 isoform, To determine the relative expression of these isoforms in PB- should not be used by clone 10. Thus, clone 10 is probably trans- MCs, we performed SYBR-Green quantitative PCR with primers lated in two isoforms of CX3CR1, which we call Met Ϫ32 and designed specifically to amplify clone 10, which contains the three Met Ϫ7, both longer than Met ϩ1, the classic isoform. AUG and the previously described V28 transcripts that translate A key factor affecting chemokine receptor functions is the pres- only CX3CR1 Met ϩ1. Both transcripts were detected in each of ence of charged amino acids in the extracellular N terminus. The the leukocyte subpopulations tested: CD4ϩ T lymphocytes, CD8ϩ first 35 aa are thus known to be critical for ligand binding and T lymphocytes, NK cells, and monocytes. As expected, V28 mes- signaling. Interestingly, the global charge of the additional amino sengers were expressed mainly by NK cells and monocytes, while acids in Met Ϫ32 is neutral: five are basic, and four are acidic (Fig. clone 10 messenger expression was more homogeneous through- 1A). Likewise, all the additional amino acids in the CX3CR1 Met out the subpopulations (Fig. 1B). Nonetheless, the relative expres- Ϫ7 isoform are neutral. Hence, the Met Ϫ7 and Met Ϫ32 isoforms sion of V28 and clone 10 messengers varied in each population. do not change the global charge of CX3CR1 and thus prevent V28 messengers were expressed principally in NK cells (14–35 repulsion. Nevertheless, the presence of four serines in Met Ϫ32 times more than clone 10 messengers) and in monocytes (3–5 and one in Met Ϫ7 might provide additional O-glycosylated po- times more), CD4ϩ T lymphocytes contained 2–4 times more tential sites that could enhance ligand binding, as reported for clone 10 messengers. Surprisingly, the CD8ϩ T lymphocytes from CCR5 (45). blood donor 3 contained 3.5 times more V28 messengers than 5308 CHARACTERIZATION OF CX3CR1 ISOFORMS clone 10 messengers, but this ratio was reversed for blood donor 4 (28 fold less), a finding that suggests a subtle regulation of CX3CR1 transcription within CD8ϩ T cells. These results indicate that clone 10 messengers are present in each leukocyte subpopulation we tested and may thus cause the production of CX3CR1 Met Ϫ7 and CX3CR1 Met Ϫ32 isoforms.

CX3CR1 isoforms identification To test whether the CX3CR1 isoforms were functional, we generated stable HEK-293 cell lines that expressed each isoform and tested them for the presence of cell surface CX3CR1 by flow cy- tometric analysis. Despite the N-terminal extension, both Met Ϫ7 and Met Ϫ32 isoforms could be detected with polyclonal CX3CR1-specific Abs. Three clones of each isoform, chosen for their high expression on FACS analysis, underwent further functional testing to ensure that any hypothetical function differences were not clone specific. The presence of specific CX3CR1 isoforms in the transfected HEK-293 cells was tested by Western blotting (Fig. 2, A and B). Downloaded from The commercial Ab revealed two specific bands, at ϳ27 and 30 kDa, in the CX3CR1 Met ϩ1 cell extract compared with the control CCR5 transfected HEK-293 cells. The weaker intensity of the 30-kDa band may reflect post-translational modifications occur- ring with Met ϩ1 or a CX3CR1 protein denaturation defect. Sim- ilarly, two specific bands with slightly higher molecular masses appeared in the Met Ϫ7 cell extract, and two more with apparent http://www.jimmunol.org/ molecular weights of ϳ31 and 34 kDa in the Met-32 cell extract. With our polyclonal Abs specific to the Met Ϫ32 isoform, no bands were detected in either the CX3CR1 Met ϩ1 or CCR5 cell extracts, and a single specific band was observed for the CX3CR1 Met Ϫ32 cell extract (Fig. 2B). Polyclonal Abs raised against the Met Ϫ7 isoform had no activity. Moreover, the polyclonal Abs raised against the Met Ϫ32 isoform did not reveal any identifiable FIGURE 3. Specific binding of 125I-labeled CX3CL1 to different CX3CR1 isoforms. A, HEK-293 cells stably transfected with CX3CR1 specific bands within the primary cell extract, probably because by guest on October 2, 2021 CX3CR1 expression in these cells was so low compared with that isoforms were incubated with increasing concentrations of unlabeled 125 in the HEK-transfected cells (data not shown). CX3CL1 in the presence of 0.05 nM [ I]CX3CL1. Cells were then washed, and the bound labeled CX3CL1 was quantitated in a gamma Together, these data indicate that the extended isoforms of counter. These data were analyzed with PRISM software (GraphPad) and CX3CR1 are effectively produced in HEK-transfected cells and are the means Ϯ SEM of four or six independent experiments performed in suggest that the Met Ϫ32 isoform may be produced within primary duplicate. The IC50 calculated for each isoform is indicated at the right of cells. the curves. B, CX3CR1-transfected HEK-293 cells were incubated with 0.05 nM [125I]CX3CL1 for the times indicated, and cell-associated radio- CX3CR1 isoforms bind CX3CL1 activity was measured. Data are the mean Ϯ SEM of three independent To compare the capacity of each protein isoform to bind CX3CL1, experiments performed in duplicate. C, CX3CR1-transfected HEK-293 125 we performed binding assays on CX3CR1-transfected HEK-293 cells were incubated with 0.05 nM [ I]CX3CL1 for 2 h, washed, and cells. In steady state experiments, the three isoforms bound the resuspended in chemokine-free binding medium. Cells were then harvested at the times indicated, and cell-associated radioactivity was measured. Data are the mean Ϯ SEM of three independent experiments performed in duplicate.

[125I]CX3CL1, which in competition assays was displaced by increasing concentrations of nonradioactive CX3CL1. As the competition curves show (Fig. 3A), the affinity was similar for all three isoforms, with a slight tendency toward better affinity for the elon- ϭ Ϯ Ϫ ϭ Ϯ gated forms (IC50 0.4 0.1 nM for Met 7 and IC50 0.4 Ϫ ϭ 0.1 nM for Met 32) compared with the classic form (IC50 1.3 Ϯ 0.6 nM for Met ϩ1). These modest increases in affinity did FIGURE 2. Identification of CX3CR1 isoforms by Western blot anal- not, however, reach statistical significance. In contrast, the simi- ysis. Membrane extracts from different CX3CR1 isoform-transfected larity of the total number of CX3CL1 binding sites within each HEK-293 cells were separated by SDS-PAGE and electrotransferred to a nitrocellulose membrane. The CCR5-HEK-293 clone was used as a neg- isoform clone suggests that N-terminal additions do not affect ative control. The protein markers are indicated in kilodaltons. The blots membrane expression. In addition, because the N-terminal regions were probed with polyclonal Abs raised against a peptide corresponding to of chemokine receptors are critically involved in ligand recogni- aa 2–21 of the human CX3CR1 (A) and also against a peptide correspond- tion, we tested whether N-terminal extension altered CX3CR1 ing to aa Ϫ25/Ϫ12 of the Met Ϫ32 isoform (B). specificity. None of the other human chemokines (CCL2, CCL5, The Journal of Immunology 5309

CCL20, CCL21, CXCL1, CXCL8, and CXCL12) or the HHV8- CX3CL1 is a better agonist for extended isoforms of CX3CR1 produced viral ligand v-macrophage inflammatory protein-II (test- 125 To determine whether changes in binding affected functional moi- ed up to 50 nM) competed significantly against [ I]CX3CL1 eties, we monitored calcium fluxes in CX3CR1 isoform-trans- (data not shown). fected HEK-293 clones. Stimulation of each CX3CR1 isoform Because the IC50 for the extended isoforms showed slight dif- with 10 nM CX3CL1 elicited a transient increase in intracellular ferences in the steady state experiments, we performed kinetic ex- calcium concentration, of similar intensity in each case (Fig. 4A). periments to characterize in more detail CX3CL1 affinity to each Interestingly, as Fig. 4B shows, the dose-response curves revealed CX3CR1 isoform. Surprisingly, the association rate constants (kon) that the threshold, half-maximal, and saturating concentrations of Ϫ Ϫ ϭ Ϯ for Met 7 and Met 32 were higher (kon 5.13 0.57 and CX3CL1 were lower for the extended CX3CR1 isoforms (EC for Ϯ ϩ ϭ 50 6.20 0.53 nM/min, respectively) than that for Met 1(kon Met Ϫ7 ϭ 0.13 Ϯ 0.05 nM; EC for Met Ϫ32 ϭ 0.08 Ϯ 0.04 nM) Ϯ Ϫ 50 3.17 0.63 nM/min; Fig. 3B), thereby indicating that Met 7 and than for Met ϩ1 (EC ϭ 0.57 Ϯ 0.19 nM; p Ͻ 0.05). We also Ϫ ϩ 50 Met 32 associate more slowly with CX3CL1 than does Met 1. performed desensitization analyses with all three isoforms. A con- Reciprocally, when the excess ligand was washed, and dissociation centration of 10 nM CX3CL1 fully abrogated the response to a Ϫ Ϫ was monitored, Met 7 and Met 32 had higher dissociation rate subsequent stimulation with 100 nM CX3CL1, whereas 1 nM in- ϭ Ϯ Ϯ Ϫ1 constants (koff 199 51 and 398 84 min , respectively) hibited 50% of the calcium mobilization in response to a second ϩ ϭ Ϯ Ϫ1 than Met 1(koff 66 18 min ; Fig. 3C); this indicates that optimal challenge (data not shown). Comparison of the desensiti- Ϫ Ϫ the Met 7 and Met 32 isoforms dissociated more slowly. To- zation curves of each isoform showed no differences gether, these data show that the CX3CR1 N-terminal addition between them. modestly affected the binding and release of CX3CL1. It is nonetheless possible that such modest changes in binding may affect Chemotactic activity of CX3CL1 on CX3CR1 isoforms Downloaded from isoform signaling and downstream functions. All CX3CR1 isoforms responded to CX3CL1 chemotactically in a dose-dependent fashion, producing the expected bell-shaped curve with similar maximal chemotaxis indexes (ϳ1.70; Fig. 5). The Met Ϫ32 HEK clones responded to lower CX3CL1 concentrations than Met ϩ1orMetϪ7 HEK clones. The half-maximal chemotactic http://www.jimmunol.org/ response for Met Ϫ32-transfected HEK occurred at ϳ0.1 nM CX3CL1 while at that concentration, Met ϩ1 and Met Ϫ7 cell responses represented only 15% of the maximal migration ( p Ͻ 0.05). Moreover, desensitization of the migration required lower concentrations of CX3CL1 for Met Ϫ7 and Met Ϫ32 HEK clones than for Met ϩ1 HEK clones (mid-desensitization at ϳ10 nM compared with ϳ100 nM; p Ͻ 0.05). Maximum migration of the isoforms was observed at different CX3CL1 concentrations (ϳ0.5

nM for extended CX3CR1 and ϳ1 nM for CX3CR1 Met ϩ1). by guest on October 2, 2021 Together, these data suggest that cells that express extended isoforms of CX3CR1 are more prone to migration in response to CX3CL1. Extended forms of CX3CR1 are potent HIV fusion and infection cofactors CX3CR1 has been described as an HIV-1 coreceptor for a limited number of viral envelopes. Because N-terminal regions of CCR5

FIGURE 4. Agonist-dependent calcium mobilization in HEK-293 cells expressing the different CX3CR1 isoforms. A, Stably transfected HEK-293 FIGURE 5. Chemotaxis activity of CX3CR1 isoform-transfected HEK- were loaded with Fura-2, and intracellular calcium concentrations were 293 in response to CX3CL1. CFSE-labeled HEK-293 cells stably trans- monitored as described in Materials and Methods. Cells were stimulated fected with CX3CR1 isoforms were tested against increasing concentra- with 10 nM CX3CL1 at the time indicated by an arrow. The data show one tions of CX3CL1 in a 96-well chemotaxis chamber. After loading 60,000 assay that is representative of five independent experiments. B, CX3CL1 cells, the 96-well plate was incubated for3hat37°C. Fluorescence in the potency in CX3CR1 isoform-transfected HEK-293. The amplitude of the bottom well was measured. Results are expressed as a chemotactic index, transient calcium peak elicited by the concentration of CX3CL1 indicated representing the number of cells migrating in response to CX3CL1 relative in CX3CR1 isoform-transfected HEK-293 is shown. Data are the mean Ϯ to the number of cells migrating in the absence of the chemokine. Data are SEM of five independent experiments. the mean Ϯ SEM of six independent experiments performed in triplicate. 5310 CHARACTERIZATION OF CX3CR1 ISOFORMS and CXCR4 have been involved in interactions with HIV enve- characterized two novel CX3CR1 isoforms that differ from the lopes, we compared the CX3CR1 isoforms in both fusion and in- form previously described by N-terminal additions of 7 and 32 aa. fection assays to investigate their HIV-1 coreceptor potencies. The These extended isoforms of CX3CR1 are fully functional and may results were normalized to the relative binding capacities to iodin- be more sensitive to CX3CL1, as binding, calcium, and chemo- ated CX3CR1 of U373MG-CD4 cells transfected with each iso- taxis assays show. These novel isoforms are potent in vitro HIV form. Surprisingly, U373MG-CD4 cells transfected with Met Ϫ7 coreceptors. Ϫ and Met 32 isoforms fused with both HeLa-EnvLAI and HeLa- Both isoforms may be translated from a single mRNA (clone ϩ Ͻ EnvADA cells 5 times more than did Met 1(p 0.05), with no 10) that also contains, in-frame, the AUG used for the translation difference in terms of viral envelope specificity (Fig. 6A). Fusions of the classic CX3CR1 form. Accordingly, we measured the rel- with extended isoforms of CX3CR1 occurred at one-third and one- ative expression of clone 10 and V28 transcripts in several leuko- fifth the rates of those obtained with CCR5- and CXCR4-trans- cyte subpopulations. Clone 10 messengers were present within ev- fected cells, respectively (data not shown). Similarly, both ex- ery leukocyte subpopulation tested and were expressed tended CX3CR1 isoforms responded more vigorously to infection predominantly by CD4ϩ T lymphocytes. If leaky scanning occurs (4.5- to 15-fold increased) from three types of viruses, LAI, VIA, from the far upstream AUG, the Met Ϫ7 isoform may be translated and V7 (Fig. 6B). These data showed that both forms of extended as well as the Met Ϫ32 isoform. The environment of the second CX3CR1 may serve as potent HIV-1 coreceptors for R5 and X4- AUG, however, precludes any leaky scanning and thus prevents tropic strains. translation of the Met ϩ1 form. Therefore, the latter may be syn- thesized only by the other two alternative messengers described Discussion (V28 and clone 6). The presence of multiple consecutive start Downloaded from We previously showed the existence of three distinct CX3CR1 codons within messengers is a well-known mechanism in leuko- mRNAs among leukocyte subpopulations and demonstrated that cyte translation regulation, as Kozak noted for lymphocytes (46). they were controlled by three different promoters (43). Here, we Surprisingly, Abs raised against the N terminus of the Met ϩ1 have identified a novel complexity in CX3CR1 regulation; we form can also detect the extended CX3CR1 isoforms on HEK clones. Consistent with the presence of both V28 and clone 10 messengers within the leukocyte subpopulation, CX3CR1 Abs may bind only to the Met ϩ1 form without detecting the longer http://www.jimmunol.org/ isoforms. Nevertheless, the increased molecular mass in Fig. 2 and the functional differences in Figs. 3–6 reveal the presence of the extended isoforms. The functional differences between the isoforms may be underestimated, because all three isoforms may be present simultaneously. The CX3CR1 isoforms described here differ from previously reported chemokine receptor isoforms that are generated by alternative splicing. Two isoforms of CXCR4 have been described by guest on October 2, 2021 (18): one results from a splicing event between a 5Ј exon coding for the first five N-terminal amino acids and a 3Ј exon coding for the remaining ORF, and the second transcript, called CXCR4-lo, arises from an intron-coding sequence that exchanges the first nine N-terminal amino acids. Assays of intracellular calcium release and of chemotaxis in transfected cells have shown that both isoforms are responsive to CXCL12. Another example of splicing variants that code for chemokine receptor isoforms is CCR9A and CCR9B, both of which are functional (17). Nevertheless, CCR9B, which is 12 aa shorter than CCR9A, is more sensitive to its ligand CCL25. CCR2 splice variants, known as CCR2A and CCR2B, also result from alternative splicing and differ in the carboxyl-terminal region (19). While both forms are functional, CCR2A is far less abundant than CCR2B in monocytes and macrophages. CX3CR1 Met ϩ1 and the extended isoforms, on the other hand, are driven by two independent promoters. This fact may allow different levels FIGURE 6. Extended CX3CR1 isoforms are potent HIV coreceptors. of expression during cell differentiation or maturation and may Fifty thousand U373MG-CD4-lacZ cells transfected with CX3CR1 iso- explain the differential expression we found in leukocyte subpopu- forms were assessed for fusion or infection. A, Coculture was performed lations. The two promoters may be differentially regulated upon with 50,000 HeLa-Tat cells expressing the prototypical R5 (EnvADA)or stimulation as well; this hypothesis remains to be elucidated. XR4 envelope (EnvLAI). Fusion was measured after 24 h of cell coculture Despite their N-terminal extensions, the two longer isoforms by measuring ␤-galactosidase activities with a microplate reader. B, The still bind CX3CL1 in steady state experiments with an affinity infection assay was performed at 2 h with 0.5 ng of virus to infect 50,000 similar to that of the Met ϩ1 form. The higher association and CX3CR1-transfected U373MG-CD4-lacZ cells. Infection was measured dissociation rate constants found in the extended isoforms, how- after 24 h of incubation by measuring ␤-galactosidase activities with a ever, indicate that they not only bind CX3CL1 more slowly, but microplate reader. Data are expressed as fusion or infection activity and represent the following calculations: OD observed in the presence of also retain the ligand much longer. The higher kon and koff that we CX3CR1 isoforms Ϫ OD observed in the presence of pcDNA3/binding observed may result from sterical environment modifications, observed in presence of CX3CR1 isoforms Ϫ binding observed in presence since the level of the kinetic constant correlates directly with the of pcDNA3. Data are the mean Ϯ SEM of eight independent experiments size of the isoforms rather than with additional amino acid charges. p Ͻ 0.05. Our results indicate that amino extension may not drastically affect ,ء .performed in triplicate The Journal of Immunology 5311 specific recognition of CX3CL1, but may instead decrease its ac- In conclusion, we describe two novel CX3CR1 isoforms with cessibility. Nevertheless, the extended tail may anchor the ligand increased sensitivity to their CX3CL1 and gp120 ligands, as shown in the binding pocket and limit its release. Moreover, it remains in binding and functional assays. The physiological advantages of possible that different degrees of post-translational modifications these more sensitive receptors remain to be determined. Further of CX3CR1 isoforms may affect isoform binding differently. Fong work will address specific distributions of these isoforms, the bi- et al. (47) showed that tyrosine sulfation of residues 14 and 22 is ological relevance of this polymorphism, and its physiological im- a key element in CX3CL1 binding. No post-translational modifi- plications for HIV. cations of the extended forms of CX3CR1 are currently known, although Western blotting experiments suggest the existence of References such phenomena (Fig. 2). The multiple serine residues present in 1. Murphy, P. M. 1994. The molecular biology of leukocyte chemoattractant recep- CX3CR1 Met Ϫ32 may be susceptible to O-glycosylation and may tors. Annu. Rev. Immunol. 12:593. affect ligand recognition, as proposed for CCR5 (45). However, 2. Luster, A. D. 1998. Chemokines: chemotactic cytokines that mediate inflammation. N. Engl. J. Med. 338:436. neither treatment with glycosidase to prevent O-glycosylation nor 3. Baggiolini, M. 2001. Chemokines in pathology and medicine. J. Intern. Med. growing the cells in the absence of sulfate to limit tyrosine sulfa- 250:91. tion affected the number and size of the bands detected. This sug- 4. LaRosa, G. J., K. M. Thomas, M. E. Kaufmann, R. Mark, M. White, L. Taylor, G. Gray, D. Witt, and J. Navarro. 1992. Amino terminus of the interleukin-8 gests either that other post-translational modifications occurred or receptor is a major determinant of receptor subtype specificity. J. Biol. Chem. that CX3CR1 proteins were denatured in various ways. 267:25402. 5. Hebert, C. A., A. Chuntharapai, M. Smith, T. Colby, J. Kim, and R. Horuk. 1993. In addition to their modestly increased affinity, the CX3CR1 Partial functional mapping of the human interleukin-8 type A receptor: identifi-

extended isoforms appeared slightly more sensitive than the classic cation of a major ligand binding domain. J. Biol. Chem. 268:18549. Downloaded from form in such functional assays as calcium response and chemo- 6. Monteclaro, F. S., and I. F. Charo. 1996. The amino-terminal extracellular domain of the MCP-1 receptor, but not the RANTES/MIP-1␣ receptor, confers taxis. The threshold, half-maximal, and saturating concentrations chemokine selectivity: evidence for a two-step mechanism for MCP-1 receptor required lower levels of CX3CL1. It also remains possible that the activation. J. Biol. Chem. 271:19084. signaling pathways mediated by the CX3CR1 isoforms generate 7. Pease, J. E., J. Wang, P. D. Ponath, and P. M. Murphy. 1998. The N-terminal extracellular segments of the chemokine receptors CCR1 and CCR3 are deter- different signaling and biological events in response to CX3CL1. minants for MIP-1␣ and eotaxin binding, respectively, but a second domain is essential for efﬁcient receptor activation. J. Biol. Chem. 273:19972.

We did not, however, observe any significant difference in the http://www.jimmunol.org/ 8. Feng, Y., C. C. Broder, P. E. Kennedy, and E. A. Berger. 1996. HIV-1 entry phosphorylation of ERK1/2 or in the adherence of CX3CR1 cells cofactor: functional cDNA cloning of a seven-transmembrane, G protein-coupled to immobilized CX3CL1 (data not shown). Hence, the existence of receptor. Science 272:872. these CX3CR1 isoforms may expand the range of CX3CL1 con- 9. Wu, L., G. LaRosa, N. Kassam, C. J. Gordon, H. Heath, N. Ruffing, H. Chen, J. Humblias, M. Samson, M. Parmentier, et al. 1997. Interaction of chemokine centrations over which a cell can migrate and thus allow more receptor CCR5 with its ligands: multiple domains for HIV-1 gp120 binding and sensitive responses. Although the CX3CR1 isoforms were more a single domain for chemokine binding. J. Exp. Med. 186:1373. sensitive in terms of calcium efflux and chemotaxis, the physio- 10. Farzan, M., H. Choe, L. Vaca, K. Martin, Y. Sun, E. Desjardins, N. Ruffing, L. Wu, R. Wyatt, N. Gerard, et al. 1998. A tyrosine-rich region in the N terminus logical importance of these slight differences is an unresolved of CCR5 is important for human immunodeficiency virus type 1 entry and me- point. We did observe, however, that every difference occurred at diates an association between gp120 and CCR5. J. Virol. 72:1160. physiological CX3CL1 concentrations. 11. Garzino-Demo, A., A. L. DeVico, and R. C. Gallo. 1998. Chemokine receptors by guest on October 2, 2021 and chemokines in HIV infection. J Clin. Immunol. 18:243. Besides its physiological roles, CX3CR1 is subverted by HIV to 12. Hoffman, T. L., and R. W. Doms. 1998. Chemokines and coreceptors in HIV/ allow virus entry. We and others have shown that CX3CR1 is a SIV-host interactions. AIDS 12:S17. minor in vitro coreceptor for HIV strains (35–37). Here, we limited 13. Berger, E. A., P. M. Murphy, and J. M. Farber. 1999. Chemokine receptors as HIV-1 coreceptors: roles in viral entry, tropism, and disease. Annu. Rev. Immu- our investigations to biochemical studies relevant to HIV infection, nol. 17:657. namely, the role of CX3CR1 in both fusion and infection assays 14. Lehner, T. 2002. The role of CCR5 chemokine ligands and antibodies to CCR5 coreceptors in preventing HIV infection. Trends Immunol. 23:347. and therefore its physical involvement as an HIV coreceptor, and 15. Atchison, R. E., J. Gosling, F. S. Monteclaro, C. Franci, L. Digilio, I. F. Charo, we found that both the Met Ϫ7 and Met Ϫ32 isoforms were more and M. A. Goldsmith. 1996. Multiple extracellular elements of CCR5 and HIV-1 potent coreceptors than the classic form for the R5, X4, and X4R5 entry: dissociation from response to chemokines. Science 274:1924. 16. Rucker, J., M. Samson, B. J. Doranz, F. Libert, J. F. Berson, Y. Yi, R. J. Smyth, strains of HIV-1. To date, CCR5 and CXCR4 are the principal R. G. Collman, C. C. Broder, G. Vassart, et al. 1996. Regions in ␤-chemokine HIV coreceptors used by the virus to enter cells (48). According to receptors CCR5 and CCR2b that determine HIV-1 cofactor specificity. Cell our data, CCR5 and CXCR4 were only 3–5 times more potent in 87:437. Ϫ Ϫ 17. Yu, C. R., K. W. Peden, M. B. Zaitseva, H. Golding, and J. M. Farber. 2000. the fusion assay than the Met 7 and Met 32 isoforms. These CCR9A and CCR9B: two receptors for the chemokine CCL25/TECK/Ck␤-15 extended isoforms of CX3CR1 may therefore be physiologically that differ in their sensitivities to ligand. J. Immunol. 164:1293. relevant for HIV infection in vivo. Although the presence of mes- 18. Gupta, S. K., and K. Pillarisetti. 1999. Cutting edge: CXCR4-Lo: molecular cloning and functional expression of a novel human CXCR4 splice variant. J. Im- sengers could not be correlated to the protein expression, our data munol. 163:2368. showed that clone 10 messengers were expressed more than V28 19. Charo, I. F., S. J. Myers, A. Herman, C. Franci, A. J. Connolly, and ϩ S. R. Coughlin. 1994. Molecular cloning and functional expression of two mono- messengers in CD4 T lymphocytes and only 3- to 5-fold less in cyte chemoattractant protein 1 receptors reveals alternative splicing of the car- monocytes, both primary targets of HIV. The presence of Met Ϫ7 boxyl-terminal tails. Proc. Natl. Acad. Sci. USA 91:2752. and Met Ϫ32 isoforms in these CD4-expressing cells may thus 20. Faure, S., L. Meyer, D. Costagliola, C. Vaneensberghe, E. Genin, B. Autran, J. F. Delfraissy, D. H. McDermott, P. M. Murphy, P. Debre, et al. 2000. Rapid promote virus entry. Further characterization of the mechanism by progression to AIDS in HIVϩ individuals with a structural variant of the che- which fusion increases with these isoforms and further study of mokine receptor CX3CR1. Science 287:2274. their involvement in HIV infection would be useful. For example, 21. Moatti, D., S. Faure, F. Fumeron, M. Amara, P. Seknadji, D. H. McDermott, P. Debre, M. C. Aumont, P. M. Murphy, D. de Prost, et al. 2001. Polymorphism post-translational modifications of the N-terminal domain are crit- in the fractalkine receptor CX3CR1 as a genetic risk factor for coronary artery ical for CCR5 fusion with gp120 (49). It remains possible that, as disease. Blood 97:1925. for CCR5, tyrosine sulfation of the extended CX3CR1 isoforms 22. Bazan, J. F., K. B. Bacon, G. Hardiman, W. Wang, K. Soo, D. Rossi, D. R. Greaves, A. Zlotnik, and T. J. Schall. 1997. A new class of membrane- enhances coreceptor function. These N-terminal extensions and bound chemokine with a CX3C motif. Nature 385:640. modifications may lead to a better understanding of the coreceptor 23. Imai, T., K. Hieshima, C. Haskell, M. Baba, M. Nagira, M. Nishimura, M. Kakizaki, S. Takagi, H. Nomiyama, T. J. Schall, et al. 1997. Identification and function and may be targets for new therapeutic approaches. Stud- molecular characterization of fractalkine receptor CX3CR1, which mediates both ies now in progress will help to clarify this issue. leukocyte migration and adhesion. Cell 91:521. 5312 CHARACTERIZATION OF CX3CR1 ISOFORMS

24. Nishimura, M., H. Umehara, T. Nakayama, O. Yoneda, K. Hieshima, HIV-2 (ROD/B): use of the 7-transmembrane receptors CXCR-4, CCR-3, and M. Kakizaki, N. Dohmae, O. Yoshie, and T. Imai. 2002. Dual functions of frac- V28 for entry. Virology 231:130. ϩ ϩ talkine/CX3C ligand 1 in trafficking of perforin /granzyme B cytotoxic effector 37. Combadiere, C., K. Salzwedel, E. D. Smith, H. L. Tiffany, E. A. Berger, and lymphocytes that are defined by CX3CR1 expression. J. Immunol. 168:6173. P. M. Murphy. 1998. Identification of CX3CR1: a chemotactic receptor for the 25. Nishiyori, A., M. Minami, Y. Ohtani, S. Takami, J. Yamamoto, N. Kawaguchi, human CX3C chemokine fractalkine and a fusion coreceptor for HIV-1. J. Biol. T. Kume, A. Akaike, and M. Satoh. 1998. Localization of fractalkine and Chem. 273:23799. CX3CR1 mRNAs in rat brain: does fractalkine play a role in signaling from 38. McDermott, D. H., J. S. Colla, C. A. Kleeberger, M. Plankey, P. S. Rosenberg, neuron to microglia? FEBS Lett. 429:167. E. D. Smith, P. A. Zimmerman, C. Combadiere, S. F. Leitman, R. A. Kaslow, et 26. Meucci, O., A. Fatatis, A. A. Simen, and R. J. Miller. 2000. Expression of al. 2000. Genetic polymorphism in CX3CR1 and risk of HIV disease. Science CX3CR1 chemokine receptors on neurons and their role in neuronal survival. 290:2031. Proc. Natl. Acad. Sci. USA 97:8075. 39. Hendel, H., C. Winkler, P. An, E. Roemer-Binns, G. Nelson, P. Haumont, S. 27. Boehme, S. A., F. M. Lio, D. Maciejewski-Lenoir, K. B. Bacon, and P. J. Conlon. O’Brien, K. Khalilli, D. Zagury, J. Rappaport, et al. 2001. Validation of genetic 2000. The chemokine fractalkine inhibits Fas-mediated cell death of brain mi- case-control studies in AIDS and application to the CX3CR1 polymorphism. croglia. J. Immunol. 165:397. J. Acquired Immune Defic. Syndr. 26:507. 28. Matloubian, M., A. David, S. Engel, J. E. Ryan, and J. G. Cyster. 2000. A transmembrane CXC chemokine is a ligand for HIV-coreceptor Bonzo. Nat. Im- 40. Porzio, M. A., and A. M. Pearson. 1977. Improved resolution of myofibrillar munol. 1:298. proteins with sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Bio- 29. Pan, Y., C. Lloyd, H. Zhou, S. Dolich, J. Deeds, J. A. Gonzalo, J. Vath, chim. Biophys. Acta 490:27. 2ϩ M. Gosselin, J. Ma, B. Dussault, et al. 1997. Neurotactin, a membrane-anchored 41. Grynkiewicz, G., M. Poenie, and R. Y. Tsien. 1985. A new generation of Ca chemokine upregulated in brain inflammation. Nature 387:611. indicators with greatly improved fluorescence properties. J. Biol. Chem. 30. Fong, A. M., L. A. Robinson, D. A. Steeber, T. F. Tedder, O. Yoshie, T. Imai, and 260:3440. D. D. Patel. 1998. Fractalkine and CX3CR1 mediate a novel mechanism of leu- 42. Sol, N., F. Ferchal, J. Braun, O. Pleskoff, C. Treboute, I. Ansart, and M. Alizon. kocyte capture, firm adhesion, and activation under physiologic flow. J. Exp. 1997. Usage of the coreceptors CCR-5, CCR-3, and CXCR-4 by primary and cell Med. 188:1413. line-adapted human immunodeficiency virus type 2. J. Virol. 71:8237. 31. Soriano, S. G., L. S. Amaravadi, Y. F. Wang, H. Zhou, G. X. Yu, J. R. Tonra, 43. Garin, A., P. Pellet, P. Deterre, P. Debre, and C. Combadiere. 2002. Cloning and V. Fairchild-Huntress, Q. Fang, J. H. Dunmore, D. Huszar, et al. 2002. Mice functional characterization of the human fractalkine receptor promoter regions. deficient in fractalkine are less susceptible to cerebral ischemia-reperfusion in- Biochem. J. 368:753. Downloaded from jury. J. Neuroimmunol. 125:59. 44. Kozak, M. 1997. Recognition of AUG and alternative initiator codons is aug- 32. Haskell, C. A., W. W. Hancock, D. J. Salant, W. Gao, V. Csizmadia, W. Peters, mented by G in position ϩ4 but is not generally affected by the nucleotides in K. Faia, O. Fituri, J. B. Rottman, and I. F. Charo. 2001. Targeted deletion of positions ϩ5 and ϩ6. EMBO J. 16:2482. CX(3)CR1 reveals a role for fractalkine in cardiac allograft rejection. J. Clin. 45. Bannert, N., S. Craig, M. Farzan, D. Sogah, N. V. Santo, H. Choe, and Invest. 108:679. J. Sodroski. 2001. Sialylated O-glycans and sulfated tyrosines in the NH2-termi- 33. Combadiere, C., S. Potteaux, J. L. Gao, B. Esposito, S. Casanova, E. J. Lee, nal domain of CC chemokine receptor 5 contribute to high affinity binding of P. Debre, A. Tedgui, P. M. Murphy, and Z. Mallat. 2003. Decreased atheroscle- chemokines. J. Exp. Med. 194:1661.

rotic lesion formation in CX3CR1/apolipoprotein E double knockout mice. Cir- http://www.jimmunol.org/ 46. Kozak, M. 1996. Interpreting cDNA sequences: some insights from studies on culation 107:1009. translation. Mamm. Genome. 7:563. 34. Lesnik, P., C. A. Haskell, and I. F. Charo. 2003. Decreased atherosclerosis in CX3CR1Ϫ/Ϫ mice reveals a role for fractalkine in atherogenesis. J. Clin. Invest. 47. Fong, A. M., S. M. Alam, T. Imai, B. Haribabu, and D. D. Patel. 2002. CX3CR1 111:333. tyrosine sulfation enhances fractalkine-induced cell adhesion. J. Biol. Chem. 35. Rucker, J., A. L. Edinger, M. Sharron, M. Samson, B. Lee, J. F. Berson, Y. Yi, 277:19418. B. Margulies, R. G. Collman, B. J. Doranz, et al. 1997. Utilization of chemokine 48. Moore, J. P. 1997. Coreceptors: implications for HIV pathogenesis and therapy. receptors, orphan receptors, and herpesvirus-encoded receptors by diverse human Science 276:51. and simian immunodeﬁciency viruses. J. Virol. 71:8999. 49. Farzan, M., T. Mirzabekov, P. Kolchinsky, R. Wyatt, M. Cayabyab, N. P. Gerard, 36. Reeves, J. D., A. McKnight, S. Potempa, G. Simmons, P. W. Gray, C. A. Power, C. Gerard, J. Sodroski, and H. Choe. 1999. Tyrosine sulfation of the amino T. Wells, R. A. Weiss, and S. J. Talbot. 1997. CD4-independent infection by terminus of CCR5 facilitates HIV-1 entry. Cell 96:667. by guest on October 2, 2021