Identification of CCR6, the Specific Receptor for a Novel Lymphocyte-directed CC Chemokine LARC*

L iver and a ctivation- r egulated c hemokine (LARC) is a recently identified CC chemokine that is expressed mainly in the liver. LARC functions as a selective che- moattractant for lymphocytes that express a class of receptors specifically binding to LARC with high affin- ity. To identifiy the receptor for LARC, we examined LARC-induced calcium mobilization in cells stably ex- pressing five CC chemokine receptors (CCR1-CCR5) and five orphan seven-transmembrane receptors. LARC spe- cifically induced calcium flux in K562 cells as well as 293/EBNA-1 cells stably expressing an orphan receptor GPR-CY4. LARC induced migration in 293/EBNA-1 cells stably expressing GPR-CY4 with a bi-modal dose-re- sponse curve. LARC fused with secreted alkaline phosphatase (LARC-SEAP) bound specifically to Raji cells stably expressing GPR-CY4 with a K d of 0.9 n M . Only LARC but not five other CC chemokines (MCP-1, RAN- TES, MIP-1 (cid:97) , MIP-1 (cid:98) , and TARC) competed with LARC-SEAP for binding to GPR-CY4. By Northern blot analy- sis, GPR-CY4 mRNA was expressed mainly

The chemokines are a group of structurally related approximately 70 -90-amino acid polypeptides involved in leukocyte recruitment and activation (1,2). The chemokines are grouped into two main subfamilies, CXC and CC, on the basis of the arrangement of the N-terminal two conserved cysteine residues. One amino acid separates the two cysteines in the CXC chemokines while the two cysteines are adjacent in the CC chemokines. Most CXC chemokines are potent neutrophil attractants while most CC chemokines recruit monocytes and also lymphocytes, basophils, and/or eosinophils with variable selectivity. Recently, a novel lymphocyte-specific chemotactic cytokine, lymphotactin/SCM-1, 1 has been reported, which car-ries only the second and the fourth of the four cysteine residues conserved in all other chemokines (3,4). This may suggest the existence of the C type chemokine subfamily.
Recently, we have identified a novel CC chemokine, LARC (liver and activation-regulated chemokine), and mapped its gene to chromosome 2q33-q37 (28). Expression of LARC mRNA was detected mainly in the liver among various human tissues and also induced in several human cell lines by phorbol 12myristate 13-acetate. LARC was chemotactic for lymphocytes but not for monocytes. LARC fused with the secreted alkaline phosphatase (LARC-SEAP) bound specifically to lymphocytes with a K d of 0.4 nM. Notably, the binding of LARC-SEAP was competed only by LARC and not by other chemokines so far tested (28). These results indicated the presence of a class of receptors specific for LARC on lymphocytes. In the present study, we have demonstrated that an orpan receptor GPR-CY4 2 is the LARC receptor that is selectively expressed on lymphocytes.

EXPERIMENTAL PROCEDURES
Cells-Human hematopoietic cell lines were maintained in RPMI 1640 supplemented with 10% fetal calf serum (FCS). 293/EBNA-1 cells were purchased from Invitrogen (San Diego, CA) and maintained in Dulbecco's modified Eagle's medium supplemented with 10% FCS. Peripheral blood leukocytes were fractionated by surface markers as de-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Calcium Mobilization Assay-This was carried out as described previously (25). In brief, cells were suspended at 3 ϫ 10 6 cells/ml in Hank's balanced salt solution (HBSS) containing 1 mg/ml of bovine serum albumin (BSA) and 10 mM HEPES, pH 7.4, (HBSS-BSA) and incubated with 1 M fura-PE3-AM (Texas Fluorescence Labs) at room temperature for 1 h in the dark. After washing twice with HBSS-BSA, cells were suspended in HBSS-BSA at 2.5 ϫ 10 6 cells/ml. 2 ml of the cell suspension in a quartz cuvette was placed in a luminescence spectrometer (Perkin-Elmer LS 50B) and fluorescence was monitored at 340 nm ( ex1 ), 380 nm ( ex2 ) and 510 nm ( em ) every 200 ms. To determine EC 50 , a dose-response curve was generated in each experiment by plotting data as percent maximum response.
Migration Assay-Cell migration was determined by using a 48-well microchemotaxis chamber as described previously (29). In brief, each chemo-attractant was diluted in Hepes-buffered RPMI 1640 supplemented with 1% BSA and placed in lower wells (25 l/well). Cells suspended in RPMI 1640 with 1% BSA at 2 ϫ 10 6 cells/ml were placed in upper wells (50 l/well). Upper and lower wells were separated by a polyvinylpyrrolidone-free polycarbonate filter with 8-m pores precoated with type IV collagen. Incubation was carried out at 37°C for 4 h in 5% CO 2 , 95% air. Filters were removed, washed, and stained with Diff-Quik. Migrated cells were counted in five randomly selected highpower fields (400 ϫ) per well. All determinations were done in triplicate.
Binding Assay-This was carried out as described previously (25,28,29). In brief, for displacement experiments, 2 ϫ 10 5 cells were incubated for 1 h at 16°C with 1 nM of SEAP(His) 6 or LARC-SEAP(His) 6 in the presence of increasing concentrations of unlabeled chemokines in 200 l of RPMI 1640 containing 20 mM Hepes, pH 7.4, 1% BSA, and 0.02% sodium azide. For saturation experiments, cells were incubated for 1 h at 16°C with increasing concentrations of LARC-SEAP(His) 6 in the presence or absence of 1 M unlabeled LARC. After incubation, cells were washed five times and lysed in 50 l of 10 mM Tris-HCl, pH 8.0, and 1% Triton X-100. Samples were heated at 65°C for 10 min to inactivate cellular phosphatase. After brief centrifugation to remove cell debris, alkaline phosphatase activity in 10 l of lysate was measured by chemiluminescent assay as described previously (28). All determinations were done in duplicate. The binding data were analyzed by the LIGAND program (34).
Northern Blot analysis-This was carried out as described previously (28,29). In brief, total RNA was prepared by using Trizol® reagent (Life Technologies, Inc.). RNA samples were separated by electrophoresis on a 1% agarose gel containing 0.66 M formaldehyde, blotted onto a filter membrane (Hybond N ϩ ) (Amersham Japan, Tokyo). Multiple tissue Northern blots and immune blots were purchased from CLONTECH (Palo Alto, CA). Hybridization was carried out with probes labeled with 32 P using Prime It II kit (Stratagene, La Jolle, CA) at 65°C in Quick-Hyb solution (Stratagene). After washing at 55°C with 0.2 ϫ SSC and 0.1% SDS, filters were exposed to x-ray films at Ϫ80°C with an intensifying screen. U45984), and GPR-9 -6 3 (GenBank TM accession number U45982). As shown in Fig. 1A, LARC specifically induced calcium flux in K562 cells expressing GPR-CY4 with complete desensitization against a rapid successive treatment with LARC. LARC did not induce calcium flux in parental K562 cells or those expressing CCR1, CCR2B, CCR3, CCR4, CCR5, or four other orphan receptors. On the other hand, MIP-1␣, MIP-1␤, MCP-1, eotaxin, or TARC did not induce calcium flux in K562 cells expressing GPR-CY4 (not shown). These chemokines, however, properly induced calcium flux in K562 cells expressing their respective CCRs even after treatment with LARC (Fig. 1A). Similar results were obtained by using a panel of 293/EBNA-1 cells stably expressing these cloned receptors (data not shown). As shown in Fig. 1B, 293/EBNA-1 cells stably expressing GPR-CY4 responded to LARC in calcium mobilization with an EC 50 of ϳ50 nM. These results clearly demonstrated that LARC was a specific functional ligand for GPR-CY4.

Induction of Calcium Mobilization by LARC-To
Induction of Chemotaxis by LARC-Previously, we showed that LARC induced chemotaxis in freshly isolated peripheral blood lymphocytes with a maximal effect at 1 g/ml (28). We therefore examined whether LARC was capable of inducing migration of 293/EBNA-1 cells stably expressing GPR-CY4. As shown in Fig. 2A, LARC induced migration in cells stably expressing GPR-CY4 with a typical bi-modal dose-response curve with a maximum effect at 1 g/ml and an EC 50 of ϳ100 ng/ml (ϳ12 nM). LARC did not induce migration in cells transfected with the vector alone. A checkerboard-type analysis revealed that the migration of GPR-CY4-transfected 293/ EBNA-1 cells toward LARC was mostly chemotactic (Fig. 2B).
Binding of LARC-Previously, we showed that LARC-SEA-P(His) 6 specifically bound to a single class of receptors expressed on lymphocytes with a K d of 0.4 nM (28). Importantly, the binding of LARC-SEAP(His) 6 was competed only by LARC and not by any other chemokines so far tested, indicating that the LARC receptor is not shared by other chemokines (28). We therefore examined the binding of LARC-SEAP(His) 6 to a panel of Raji cells stably expressing GPR-CY4 and other cloned receptors. LARC-SEAP(His) 6 bound specifically to cells expressing GPR-CY4 but not to parental cells or those expressing five CCRs or four other orphan receptors (data not shown). As shown in Fig. 3A, the binding of LARC-SEAP(His) 6 to GPR-CY4 was saturable when increasing concentrations of LARC-SEAP(His) 6 were incubated with Raji cells expressing GPR-CY4. The Scatchard analysis revealed a K d of 0.9 nM and 28,800 sites/cell (Fig. 3B). Unlabeled LARC fully competed with LARC-SEAP(His) 6 for GPR-CY4 with an IC 50 of 3.4 nM (Fig.  3C). Furthermore, no other CC chemokines, MCP-1, RANTES, MIP-1␣, MIP-1␤, or TARC, were capable of competing with LARC-SEAP(His) 6 for GPR-CY4 (Fig. 3D). These binding characteristics were highly consistent with those obtained from the endogenous class of LARC receptors expressed on lymphocytes (28).
Selective Expression of GPR-CY4 in T and B Cells-We have shown that the endogenous class of LARC receptors is expressed selectively on lymphocytes (28). Therefore, we examined the expression pattern of GPR-CY4 in various human tissues and leukocyte subsets by Northern blot analysis (Fig.  4). When blots for various tissues were hybridized with the 32 P-labeled GPR-CY4 cDNA probe (Fig. 4A), GPR-CY4 mRNA was found to be expressed strongly in the spleen and weakly in the lymph nodes. Weak expression was also detected in the testis (larger transcripts), small intestine, and PBL. Notably, the mRNA expression was very low, if any, in the liver where the LARC transcripts were mainly detected (28). When blots specific for various lymphoid tissues were probed, GPR-CY4 mRNA was detected strongly in the spleen, lymph nodes, and appendix, and weakly in the fetal liver (Fig. 4B). When the same lymphoid tissue blots were rehybridized with the 32 Plabeled LARC cDNA probe, LARC mRNA was detected moderately in the appendix and weakly in the lymph nodes, PBL, and fetal liver (Fig. 4B). Thus, the constitutive expression of GPR-CY4 and that of LARC overlap partly in the secondary lymphoid tissues. We then examined the expression of GPR-CY4 mRNA in various leukocyte subsets. T cells (both CD4 ϩ and CD8 ϩ T cells) and B cells were clearly positive, whereas NK cells, monocytes, or granulocytes were virtually negative even though some RNA loading differences were noted (Fig. 5A). We also examined the expression of GPR-CY4 mRNA in various human hematopoietic cell lines. Only a T cell line, Hut102, , and a megakaryocytic cell line (MEG-1) were virtually negative (not shown). Collectively, the expression of GPR-CY4 is mostly limited in the secondary lymphoid tissues and also in T and B lymphocytes. The expression pattern of GPR-CY4 is thus highly consistent to the lymphocyte-selective expression of the endogenous LARC receptor described previously (28).
Loetscher et al. (35) have reported that CD45RO ϩ T cells express CCR1 and CCR2 only after prolonged treatment with IL-2. They further showed that activation of T cells with PHA, anti-CD3, or anti-CD3 and anti-CD28 did not induce expression of CCR1 or CCR2 but rather suppressed the effect of IL-2 (35). We therefore examined the effect of IL-2 without or with PHA on expression of GPR-CY4 in CD4 ϩ and CD8 ϩ T cells (Fig.  5B). Expression of GPR-CY4 in both CD4 ϩ and CD8 ϩ T cells was strongly up-regulated by IL-2. The effect of IL-2 was, however, strongly suppressed by co-treatment with PHA. Thus, the expression of GPR-CY4 in T cells is positively regulated by IL-2 but negatively regulated by T-cell activation like those of CCR1 and CCR2 (35). DISCUSSION LARC is a novel CC chemokine with 20ϳ28% identity to other cloned human CC chemokines (28). LARC is mainly expressed in the liver and also induced in human cell lines, such as a monocytic cell line U937, by phorbol myristate acetate. Thus, we designated this chemokine as LARC from Liver and Activation-Regulated Chemokine (28). The present study has further demonstrated that LARC is constitutively expressed at relatively low levels in tissues such as the lymph nodes, appendix, and fetal liver (Fig. 4B). It remains to be explored what types of cells produce LARC in the liver and some lymphoid tissues and what kinds of cytokines and stimulants regulate LARC expression.
LARC is chemotactic for lymphocytes in vitro with a maximal activity at 1 g/ml (28). At high concentrations, LARC may also be chemotactic for neutrophils. However, LARC is totally inactive on monocytes (28). A similar relatively low potency in induction of chemotaxis in lymphocytes has been noted for TARC (29) and SDF-1/PBSF (36). Lymphocytes, especially resting ones, may be relatively inefficient in chemotactic responses to these chemokines. In keeping with the lymphocyte-selective activity of LARC, lymphocytes possess a class of receptors binding LARC with a high affinity (K d ϭ 0.4 nM) (28). Furthermore, the receptor expressed on lymphocytes is highly specific for LARC and not shared by other CC chemokines so far tested (28). Interestingly, TARC (29) and SDF-1/PBSF (36) are also the ones that possess receptors, CCR4 (25) and CXCR4 (12,13), respectively, that are not shared by other chemokines so far tested.
In the present study, we have demonstrated that an orphan receptor GPR-CY4 (GenBank TM accession number U45984) is the LARC receptor expressed on lymphocytes. Recently, the same receptor was also deposited in the data base as DRY6 4 (GenBank TM accession number U60000). It was also reported as an orphan receptor CKR-L3 (37). LARC induced calcium mobilization and chemotactic responses specifically in K562 cells and 293/EBNA-1 cells stably expressing GPR-CY4 (Figs. 1 and 2). LARC fused with SEAP bound specifically to Raji cells stably expressing GPR-CY4 with a K d of 0.9 nM (Fig. 3). Binding of LARC-SEAP to GPR-CY4 was blocked only by LARC and not by any other chemokines so far tested (Fig. 3). GPR-CY4 was found to be expressed strongly in the secondary lymphoid tissues such as the spleen, lymph nodes, and appendix, and also in the fetal liver (Fig. 4). A very similar result was reported for the tissue expression of CKR-L3 (37). Furthermore, GPR-CY4 was expressed highly selectively in peripheral blood lymphocytes, namely both CD4 ϩ and CD8 ϩ T cells and B cells (Fig.  5A). Collectively, these results clearly indicate that GPR-CY4 is the receptor that specifically binds LARC with high affinity and is expressed selectively on lymphocytes. We propose GPR-CY4 to be designated as CCR6.
Compared with the high affinity binding of LARC to CCR6 (K d ϭ 0.9 nM), LARC needed much higher concentrations to induce intracellular calcium mobilization (EC 50 ϭ ϳ50 nM) or chemotactic responses (EC 50 ϭ ϳ12 nM) in cells stably transfected with CCR6. At present, we do not know the exact causes of such discrepancies, but these may be due in part to differences in assay conditions such as temperature, duration of incubation, etc. Furthermore, Monteclaro and Charo (46) recently demonstrated a two-step mechanism for activation of CCR1 by MCP-1 in which high affinity binding of MCP-1 with the amino terminus of CCR1 allows subsequent low affinity interactions of MCP-1 with the extracellular loops/transmembrane domains of CCR1 that lead to receptor activation and signaling. A similar two-step mechanism may explain high affinity binding versus low signaling potency of LARC to CCR6.
As in the cases of CCR1 and CCR2 (35), IL-2 strongly induces the expression of CCR6 in resting T cells while PHA activation blocked the inducing effect of IL-2 (Fig. 5B). Thus, not antigenic stimulation per se, but subsequent IL-2-mediated expansion may enhance T-cell responsiveness to LARC. In mice, repeated injection of IL-2 was shown to induce massive lymphocyte infiltration in the liver and lung (38). Since LARC is expressed rather selectively in the liver and lung (28), LARC may be involved in the IL-2-induced lymphocyte infiltration in these organs.
Most CC chemokines are known to interact with multiple shared receptors (1,2,5,6). For example, MIP-1␣ binds to CCR1 and CCR5 (14,15,26,27,45), while RANTES binds to CCR1, CCR3, and CCR5 (14,15,21,22,26,27,45). Eotaxin apparently interacts only with CCR3 (20 -22), but CCR3 also binds RANTES, MCP-3, and MCP-4 (21)(22)(23)44). Thus, each chemokine may recruit multiple types of cells even if they express different types of receptors, whereas each cell may respond to multiple types of chemokines even by expressing a single type of receptor. The exact physiological meanings of such redundant and complex relationships between chemokines and their receptors are still unclear, but such partially overlapping specificities may have advantages in acute inflammatory responses where similar leukocyte subsets have to be rapidly recruited in a wide variety of settings and microenvironments even if there are considerable differences in the local pattern and spectrum of chemokine production. In this context, LARC (28) and also the recently identified T-cell-directed CC chemokine TARC (29) are quite unique because they interact with highly specific receptors, CCR6 (this paper) and CCR4 (25), respectively. In fact, LARC and TARC have a number of features in common that are unique among the known CC chemokines. They are constitutively expressed in certain organs and lymphoid tissues, with LARC mainly in the liver (28) and also in some secondary lymphoid tissues (Fig. 4B), whereas TARC is mainly expressed in the thymus and also most probably in some secondary lymphoid tissues (29). Both LARC and TARC act selectively on lymphocytes, LARC on both T and B cells (28, Fig. 5), whereas TARC acts mainly on CD4 ϩ T cells (25,29). Even though the genes for other CC chemokines are known to be clustered on human chromosome 17q11.2 (1, 2), the genes for LARC and TARC are mapped distinctly to chromosome 2q33-q37 and chromosome 16q13, respectively (28,39). Thus, LARC and TARC may constitute a new group of CC chemokines that have more specialized functions in lymphocyte trafficking and immune responses than other CC chemokines clustered on chromosome 17.
In this regard, the CXC chemokine SDF-1/PBSF is also the one that acts via its specific receptor CXCR4 (12,13), is constitutively expressed in various tissues (40), and is mapped to chromosome 10q (40) instead of chromosome 4 where the genes for other CXC chemokines are known to be clustered (1,2). Furthermore, the recently described C type chemokine lymphotactin/SCM-1 (3, 4) that also acts selectively on lymphocytes is distinctly mapped to human chromosome 1q23 (41). By generating gene-targeted mice, SDF-1/PBSF has been shown to be essential for B cell lymphopoiesis in the fetal liver and for myelopoiesis and B cell lymphopoiesis in the bone marrow during embryonic development (42). Thus, during embryogenesis, SDF-1/PBSF may be involved in generation of B cell progenitors in the fetal liver and in colonization of hematopoietic precursor cells into the bone marrow (42). Recently, genetargeted mice lacking a putative CXC chemokine receptor BLR1 that is expressed on mature B cells and a subpopulation of CD4 ϩ T cells have been shown to have anatomical defects such as lack of inguinal lymph nodes, impaired development of Peyer's patches, and defective formation of primary follicles and germinal centers in the spleen (43). When injected into wild type mice, B cells lacking BLR1 failed to migrate from the T cell-rich zone into B cell follicles in the spleen and Peyer's patches (43). Thus, BLR1 may be involved in B cell migration within specific anatomic compartments in the spleen and Peyer's patches. Likewise, TARC and LARC with their respective receptors, CCR4 and CCR6, may play roles not only in inflammatory and immunological responses but also in the normal lymphocyte trafficking and microenvironmental homing that are essential for development and maintenance of various lymphoid tissues.
Identification of the LARC receptor CCR6 now enables us to define the exact types of cells that respond to LARC. Generation of gene-targeted mice lacking LARC and CCR6 will be useful to address their in vivo functions.