The synthetic peptide Trp-Lys-Tyr-Met-Val-Met-NH2 specifically activates neutrophils through FPRL1/lipoxin A4 receptors and is an agonist for the orphan monocyte-expressed chemoattractant receptor FPRL2.

Neutrophils express the G protein-coupled N-formyl peptide receptor (FPR) and its homologue FPRL1, whereas monocytes express FPR, FPRL1, and FPRL2, an orphan receptor sharing 83% amino acid identity with FPRL1. FPRL1 is a promiscuous receptor activated by serum amyloid A and by different synthetic peptides, including the hexapeptide Trp-Lys-Tyr-Met-Val-d-Met-NH(2) (WKYMVm). By measuring calcium flux in HL-60 cells transfected with FPR, FPRL1, or FPRL2, we show that WKYMVm activated all three receptors, whereas the l-conformer WKYMVM activated exclusively FPRL1 and FPRL2. The functionality of FPRL2 was further assessed by the ability of HL-60-FPRL2 cells to migrate toward nanomolar concentrations of hexapeptides. The half-maximal effective concentrations of WKYMVM for calcium mobilization in HL-60-FPRL1 and HL-60-FPRL2 cells were 2 and 80 nm, respectively. Those of WKYMVm were 75 pm and 3 nm. The tritiated peptide WK[3,5-(3)H(2)]YMVM bound to FPRL1 (K(D) approximately 160 nm), but not to FPR. The two conformers similarly inhibited binding of (125)I-labeled WKYMVm to FPRL2-expressing cells (IC(50) approximately 2.5-3 micrometer). Metabolic labeling with orthophosphoric acid revealed that FPRL1 was differentially phosphorylated upon addition of the l- or d-conformer, indicating that it induced different conformational changes. In contrast to FPRL1, FPRL2 was already phosphorylated in the absence of agonist and not evenly distributed in the plasma membrane of unstimulated cells. However, both receptors were internalized upon addition of either of the two conformers. Taken together, the results indicate that neutrophils are activated by WKYMVM through FPRL1 and that FPRL2 is a chemotactic receptor transducing signals in myeloid cells.


I-labeled WKYMVm to FPRL2-expressing cells (IC
ϳ 2.5-3 M). Metabolic labeling with orthophosphoric acid revealed that FPRL1 was differentially phosphorylated upon addition of the L-or D-conformer, indicating that it induced different conformational changes. In contrast to FPRL1, FPRL2 was already phosphorylated in the absence of agonist and not evenly distributed in the plasma membrane of unstimulated cells. However, both receptors were internalized upon addition of either of the two conformers. Taken together, the results indicate that neutrophils are activated by WKYMVM through FPRL1 and that FPRL2 is a chemotactic receptor transducing signals in myeloid cells.
The extravasation of leukocytes from the peripheral blood stream to inflammatory sites is a key feature in the innate immune response to infection (1). Different chemoattractants (e.g. N-formylated peptides, C5a, interleukin-8, leukotriene B 4 , and platelet-activating factor) and chemokines induce leukocyte infiltration and activation through binding to G proteincoupled seven-transmembrane cell-surface receptors (2,3). The chemoattractant-mediated dissociation of G␣ i2 from the G␤␥ subunit complex results in the activation of several downstream signaling effector enzymes that promote intracellular calcium mobilization, modifications in the metabolism of phosphoinositides, and activation of mitogen-activated protein kinases (4). The integration by the cell of the different chemoattractant-activated signaling pathways results in directed cell migration, recruitment of new receptors from the granules to the cell surface, release of proteolytic enzymes, production of large amounts of superoxide by the neutrophil NADPH oxidase, and increased gene transcription (5)(6)(7)(8). The extent of the cellular response is dependent on the identity of the agonist and on the level of expression and desensitization of the receptors involved in the activation process (9).
Two synthetic hexapeptides, Trp-Lys-Tyr-Met-Val-Met-NH 2 (WKYMVM)andTrp-Lys-Tyr-Met-Val-D-Met-NH 2 (WKYMVm), that stimulate phosphoinositide hydrolysis in myeloid cells were identified by screening a peptide library (10,11). The D-methionine-containing hexapeptide (WKYMVm) was found to be a very potent activator of several leukocyte effector functions such as chemotaxis, mobilization of complement receptor-3, and activation of the NADPH oxidase (11). The peptide WKYMVm activates neutrophils through both the N-formyl peptide receptor (FPR) 1 and FPRL1 (N-formyl peptide receptor-like-1) (12,13). The latter was originally cloned from human phagocytes by low-stringency hybridization of a cDNA library with the FPR cDNA sequence, and it was initially defined as an orphan receptor (14 -16). FPRL1 was later referred to as the LXA 4 receptor since it was shown to bind lipoxin A 4 with high affinity (17). In addition, several different peptides/proteins have been reported to stimulate this receptor. These include a leucine zipper-like domain of the HIV-1 envelope gp41 (18); a peptide derived from the HIV-1 envelope gp120 (19); serum amyloid A, a protein secreted during the acute phase of inflammation (20); the mitochondrial necrotactic peptide MYFINILTL (21); LL-37, a cleaved antimicrobial peptide of a human antimicrobial protein of the cathelicidin family (22); and the neurotoxic prion peptide fragment 106 -126 (23). The binding of the peptide/protein agonists to FPRL1 results in a G protein-mediated signaling cascade leading to chemotaxis, mobilization of granules, and release of reactive oxygen species. Thus, in many ways, the FPR and FPRL1/LXA 4 receptors show great similarities in their signaling abilities.
Based on the knowledge that the L-methionine-containing peptide WKYMVM (but not N-formyl-Met-Leu-Phe) is able to activate phosphoinositide hydrolysis in the human myeloma B-cell line U266, it has been suggested that WKYMVM activates a receptor distinct from FPR (10). Here, we show that radioactive WKYMVM binds to FPRL1, but not to FPR. The hexapeptides WKYMVM and WKYMVm induced a different pattern of phosphorylation and internalization of FPRL1 in transfected cells, suggesting that the two peptides bring about different structural changes in the receptor. In addition, we show that the L-and D-conformers elicit both calcium flux in and chemotaxis of undifferentiated HL-60 cells transfected with the FPRL2 (N-formyl peptide receptor-like-2). Thus, this orphan receptor, which shares 83% amino acid identity with FPRL1 and is expressed on monocytes, but not on neutrophils or differentiated neutrophil-like HL-60 cells (24), is a functional chemotactic receptor.
Stable Expression of FPR, FPRL1, and FPRL2 in Undifferentiated HL-60 Cells-The stable expression of FPR and FPRL1 in undifferentiated cells has been previously described (13). The cDNA encoding FPRL2 was cloned by polymerase chain reaction from genomic DNA isolated from undifferentiated HL-60 cells using the sense primer 5Јcgggatccgcagacaagatggaaaccaacttctccattcc-3Ј and the reverse primer 5Јgctctagagctcacattgcttgtaactccgtctcctc-3Ј. After cleavage with BamHI and XbaI, the polymerase chain reaction product was cloned in pcDNA3.1 and completely sequenced. The FPRL2 cDNA was further ligated into the pEFneo plasmid cleaved by XbaI (26). Transfection of HL-60 cells was performed by electroporation with a Bio-Rad Gene Pulser apparatus according to a slightly modified version (27) of the technique described by Tonetti et al. (28). Following electroporation, cells were allowed to recover in 20 ml of culture medium for 48 h prior to selection in a medium containing G418 (1 mg/ml). G418-resistant clones were obtained by limited dilution into 24-well microtiter plates, and FPRL2-expressing clones were identified for their ability to mobilize intracellular calcium upon addition of WKYMVm (100 nM final concentration). Cells were cultured in RPMI 1640 medium/glutaMax I, and the maximal density was maintained below 2 ϫ 10 6 cells/ml. Cells were centrifuged at each passage.
Stable Overexpression of FPR, FPRL1, and FPRL2 in RINm5F Cells-The stable expression of FPR in the insulin-secreting cell line RINm5F has been previously described (29). The FPR-expressing RINm5F cells expressed ϳ1.4 ϫ 10 6 receptors/cell. FPRL1-and FPRL2expressing RINm5F cells were obtained by transfecting cells with 1 g of pSV2neo DNA and 20 g of CDM8-FPRL1 or with 2 g of pEFneo-FPRL2, respectively. Cells were cultured in RPMI 1640 medium/glu-taMax I with 400 g/ml G418 and 10% heat-inactivated fetal calf serum. Resistant clones were screened for FPRL1 or FPRL2 expression by immunofluorescence using a rabbit polyclonal antibody directed against the last 10 amino acid residues of FPRL1.

Preparation of Tritiated Hexapeptide Trp-Lys-[3,5-3 H 2 ]Tyr-Met-Val-Met-NH 2 -Diiodinated
Trp-Lys-Tyr-Met-Val-Met-NH 2 (1.1 mg, 1 mmol) was treated overnight with 99.5% tritium gas (4.2 hectopascal) over palladium oxide (5 mg) in freshly dried methanol (2 ml) and triethylamine (30 ml). After elimination of the residual tritium gas under reduced pressure, the catalyst was removed by filtration through a Millipore FG filter. The filtrate was evaporated under vacuum and rinsed three times with methanol. The resulting crude product (7.1 mCi) was dissolved in methanol (7.2 ml). Purification by analytical reverse-phased HPLC (Vydac C 8 column, flow rate of 1 ml/min, linear gradient of 0.1% trifluoroacetic acid in water to 0.01% trifluoroacetic acid in acetonitrile from 18:82 to 50:50 in 30 min) gave pure tritiated Trp-Lys-Tyr-Met-Val-Met-NH 2 (0.308 mCi) with a specific activity of 8.5 Ci/mmol, which was compared with an authentic sample of Trp-Lys-Tyr-Met-Val-Met-NH 2 . Chemical and radiochemical purities were assessed by analytical HPLC (same conditions as above). The retention time for Trp-Lys-Tyr-Met-Val-Met-NH 2 was 12.52 min; the purity was 99%; and the retention time for diiodinated Trp-Lys-Tyr-Met-Val-Met-NH 2 was 19.68 min.
Peptide Iodination-The highly potent FPRL1 agonist Trp-Lys-Tyr-Met-Val-D-Met-NH 2 was labeled with 125 I using the solid-phase oxidizing agent IODO-GEN (30). Briefly, 10 nmol of IODO-GEN were solubilized in 100 l of dichloromethane and coated as a film on the wall of a polypropylene iodination vial by evaporation to dryness in a desiccator. Sodium phosphate buffer (0.01 M, 75 l, pH 8.0) was added to the vial, followed by Na 125 I (1 mCi) and 4 nmol of Trp-Lys-Tyr-Met-Val-D-Met-NH 2 in 25 l of sodium phosphate buffer. The iodination was allowed to proceed for 10 min and terminated by the addition of 500 l of water. After 5 min, the diluted mixture was passed through a Waters Sep-Pak C 18 cartridge, followed by a wash with 10 ml of potassium iodide (5 mM) and 10 ml of water. The cartridge was then turned upside down, and the 125 I-labeled peptide was eluted with 2 ml of methanol. The iodinated peptide migrated with R F ϭ 0.35 on a Whatman Silica Gel K6 plate with a solvent system of methanol/water/acetic acid (20:70:10, v/v/v). When higher amounts of IODO-GEN were used, the peptide was oxidized during the iodination procedure, and species with smaller R F values were observed. The iodinated peptide could be stored for several weeks at Ϫ20°C in methanol containing 0.5% ␤-mercaptoethanol. A specific activity of 100 -130 Ci/mmol was routinely obtained.
Binding Assays of Cells Stably Expressing FPR, FPRL1, or FPRL2-Equilibrium binding in RINm5F cells expressing either FPR or FPRL1 was essentially performed on adherent cells as previously described (31). Briefly, cells were incubated with increasing concentrations of Trp-Lys-[3,5-3 H 2 ]Tyr-Met-Val-Met-NH 2 in Dulbecco's modified Eagle's medium and 0.5% bovine serum albumin supplemented or not with an excess of nonradioactive peptide for 1 h at 4°C. Plates were washed three times with chilled phosphate-buffered saline; cells were solubilized in scintillation mixture; and the radioactivity bound to the cells was quantified with a ␤-counter. Competitive binding assays were performed on adherent cells at 4°C for 1 h using a single concentration of 125 I-labeled Trp-Lys-Tyr-Met-Val-D-Met-NH 2 in Dulbecco's modified Eagle's medium and 0.5% BSA containing increasing concentrations of unlabeled ligands.
Metabolic Labeling and Immunoprecipitation-RINm5F cells expressing either FPRL1 or FPRL2 were metabolically labeled with [ 32 P]orthophosphoric acid (0.3-0.5 mCi/ml) in 6-well plates as previously described (32). Undifferentiated HL-60-FPRL2 cells were metabolically labeled in suspension as previously described (33). About 40 ϫ 10 6 HL-60-FPRL2 cells were required for each condition of immunoprecipitation. Receptor phosphorylation was initiated by the addition of either WKYMVm or WKYMVM, and phosphorylated receptors were immunoprecipitated with affinity-purified rabbit IgGs directed against the last 10 residues of FPRL1 (34).
Neutrophil NADPH Oxidase Activity-The NADPH oxidase activity was determined using an isoluminol-enhanced chemiluminescence system (35). The chemiluminescence activity was measured in a six-channel Biolumat LB 9505 apparatus (Berthold Co., Wildbad, Germany) using disposable 4-ml polypropylene tubes with a 900-l reaction mix-ture containing 1-2 ϫ 10 5 neutrophils, horseradish peroxidase (4 units), and isoluminol (2 ϫ 10 Ϫ5 M) (36). The tubes were equilibrated in the Biolumat for 5 min at 37°C, after which the stimulus (0.1 ml) was added. By a direct comparison of the superoxide dismutase-inhibitable reduction of cytochrome c and superoxide dismutase-inhibitable chemiluminescence, 7.2 ϫ 10 7 cpm were found to correspond to production of 1 nmol of superoxide (a millimolar extinction coefficient for cytochrome c of 21.1 was used). Details about the chemiluminescence technique is given in Ref. 35.
Determination of Changes in Cytosolic Calcium in HL-60 Cells Expressing FPR, FPRL1, or FPRL2-HL-60 cells at a density of 1-3 ϫ 10 6 cells/ml were washed with phosphate-buffered saline. The cell pellets were resuspended at a density of 2 ϫ 10 7 cells/ml in RPMI 1640 medium without phenol red and containing 0.1% BSA and loaded with 2 M Fura-2/AM (Molecular Probes, Inc., Eugene, OR) for 30 min at 37°C. Cells were then diluted with 2 volumes of the same medium without BSA, washed once with KRG buffer, and resuspended in RPMI 1640 medium without phenol red at a density of 2 ϫ 10 7 cells/ml. Calcium measurements were carried out with a Spex FluoroMAX fluorescence spectrophotometer with an excitation wavelength of 340 nm; an emission wavelength of 505 nm; and slit widths of 5 and 10 nm, respectively. Intracellular free calcium concentrations were calculated using the following formula: where F max is the fluorescence in the presence of 0.04% Triton X-100 and F min is the fluorescence obtained after the addition of 5 mM EGTA plus 30 mM Tris-HCl, pH 7.4.
Chemotaxis Assay-Undifferentiated HL-60 cells stably transfected with pEFneo or pEFneo-FPRL2 were centrifuged, resuspended in fresh complete RPMI 1640 medium (10 6 cells/ml), and incubated overnight with 10 g/ml 1,1Ј-dioctadecyl-3,3,3Ј,3Ј-tetramethylindocarbocyanine perchlorate (Molecular Probes, Inc.). Cells were centrifuged and washed once with complete medium and twice with chemotaxis buffer (RPMI 1640 medium supplemented with 1% BSA and 20 mM HEPES, pH 7.5). 2 ϫ 10 5 cells in 100 l of chemotaxis buffer were loaded into the upper chamber of FluoroBlok inserts of 3-m pore size (Becton Dickinson, Le Pont-de-Claix, France). 500 l of chemotaxis buffer with or without the WKYMVm hexapeptide (at concentrations indicated in the figures) were loaded in the lower chamber. The FluoroBlok inserts were incubated at 37°C for 5 h. Migrating cells dropped from the filters to the bottom of the lower chamber. They were counted in five fields (magnification ϫ 100) using an inverted fluorescence microscope equipped with an image analyzer (SP-Eye, Photonic Science). The experiments were performed six times, and the results are presented as chemotaxis indexes (number of cells migrating in response to the chemoattractant divided by the number of HL-60-FPRL2 cells spontaneously migrating toward control medium) for different chemoattractant concentrations.
Microscopic Analysis of Internalization of FPRL1 and FPRL2-Ligand-mediated internalization of FPRL1 and FPRL2 was followed by an indirect immunofluorescence microscopy staining technique. Cells were seeded on polyornithine-coated coverslips 48 h prior to the experiments. To stop protein synthesis, cells were incubated in RPMI 1640 medium containing 1% BSA and 100 g/ml cycloheximide for 3 h and maintained in the same medium throughout the experiment. To study receptor internalization, RINm5F-FPRL1 cells and RINm5F-FPRL2 cells were treated with either WKYMVM or WKYMVm at 37°C for various time periods ranging from 0 to 30 min. Cells were finally fixed and permeabilized with acetone at Ϫ20°C for 30 s. Permeabilized cells were treated for immunofluorescent staining with a 1:400 dilution of affinitypurified rabbit anti-FPRL1 polyclonal antibody and a 1:500 dilution of BODIPY-labeled goat anti-rabbit antibody as previously described (32). For detailed microscopic analysis, we used a Leica TCS-SP2 confocal scanning microscope. BODIPY-labeled antibody was excited at 488 nm with an argon laser, and its fluorescence was collected between 500 and 600 nm.

Peptide-induced NADPH Oxidase Activity in Neutrophils-
The formylated peptide fMLF is a potent activator of the neutrophil NADPH oxidase, and the generated reactive oxygen species are secreted from the cells. When exposing neutrophils to WKYMVM, a similar superoxide production was achieved (Fig. 1). The time courses of the responses induced by the two peptides were very similar, but WKYMVM was a somewhat less potent activator, with an EC 50 of 75 nM compared with 50 nM for fMLF (Fig. 1, inset).
The fMLF antagonist cyclosporin H is known to block FPRevoked responses (Refs. 37 and 38; for a review, see Ref. 3). Accordingly, cyclosporin H (1 M) blocked the NADPH oxidase activity induced by fMLF, whereas the inhibitor had no effect on the response induced by WKYMVM (data not shown). To further determine the specificity in the activation route for the WKYMVM and fMLF peptides, we performed cross-desensitization experiments. Upon binding of fMLF to its receptor, the occupied receptor is phosphorylated (33) and interacts with the cytoskeleton (39 -41). Cells are subsequently desensitized and unable to generate oxidase-activating signals through the same receptor. As illustrated in Fig. 2A, neutrophils stimulated with fMLF were unable to generate a second burst of superoxide when challenged 10 min later with the same agonist (homologous desensitization). However, a second superoxide burst was observed when cells were first stimulated with fMLF and then challenged with WKYMVM (Fig. 2B). When the first stimulus was WKYMVM, no second respiratory burst was observed in response to the same agonist (homologous desensitization) (Fig. 2C), but a response was obtained with fMLF ( Fig. 2D). Thus, activation by one of the peptides did not induce desensitization to the other peptide (no heterologous or cross-desensitization), suggesting, as already proposed by Baek et al. (10), that the two peptides act through different receptors.
Peptide-induced Activation of FPR or Its Homologues (FPRL1 and FPRL2) in Transfected HL-60 Cells-In addition to the high-affinity FPR, human neutrophils express a structurally related receptor originally known as FPRL1 (15). The second homologue (FPRL2) is 83% identical to FPRL1, but its capacity to transduce signals has not yet been assessed (24). FPRL2 is not expressed in neutrophils or neutrophil-like HL-60 cells, but is present on monocytes (24). FPRL1 is promiscuous and binds several different peptides, including the D-methionine-containing chemoattractant WKYMVm (12,13). The hexapeptide WKYMVm is at least 6000-fold more potent than fMLF for the activation of FPRL1. However, it can also activate phagocytes through FPR (12,13). To investigate whether WKYMVM is able to activate receptors of the FPR family (FPR, FPRL1, and FPRL2), we stably expressed these receptors in myeloid HL-60 cells. Although WKYMVM has been shown to activate phosphoinositide hydrolysis in undifferentiated HL-60 cells (10), we did not observed a WKYMVM-mediated calcium mobilization in control cells, i.e. naive non-transfected cells or cells transfected with CXCR2 (data not shown). 2 Likewise, no calcium rise was observed when FPR-expressing cells were stimulated with 100 nM WKYMVM, whereas an increase in intracellular calcium concentration was observed when cells were stimulated with fMLFK (Fig. 3A). In contrast, WKYMVM induced a rise in intracellular calcium in FPRL1-expressing cells (Fig. 3B). The FPR antagonist cyclosporin H (1 M) had no inhibitory effect on the WKYMVM-induced calcium mobilization in HL-60-FPRL1 cells (data not shown). In FPRL1-expressing cells, WKYMVM mobilized calcium with an EC 50 of ϳ2 nM, whereas WKYMVm had an EC 50 of ϳ75 pM (Fig. 4). We thus conclude that WKYMVM is ϳ25-fold less potent than WKYMVm; but, in contrast to the latter, it appears to stimulate cells without activating FPR.
The three peptides fMLFK, WKYMVM, and WKYMVm were further tested on HL-60 cells that stably expressed FPRL2. In a previous study, Durstin et al. (24) have shown the lack of fMLF-induced calcium mobilization in frog oocytes injected with FPRL2 cRNA. Consistently, no fMLFK-induced response could be obtained in HL-60-FPRL2 cells. However, both WKYMVM and WKYMVm were able to induce a sustained increase in intracellular calcium (Fig. 3, C and D) with EC 50 values of ϳ80 and 3 nM, respectively (Fig. 4). Interestingly, although 1 M recombinant serum amyloid A elicited a marked increase in intracellular calcium (ϳ200 nM) in HL-60-FPRL1 cells, no calcium flux could be observed in HL-60-FPRL2 cells (data not shown), indicating that HL-60-FPRL2 cells do not constitutively expressed FPRL1. Thus, the ability of the hexapeptides to induce calcium mobilization demonstrates for the first time that the orphan monocyte-expressed receptor FPRL2 transduces signals.
To further assess the functionality of FPRL2 in myeloid cells, we examined the ability of WKYMVm to elicit the migration of HL-60-FPRL2 cells through 3-m pores into a chamber containing various concentrations of hexapeptide. Although naive HL-60 cells did not migrate to the lower chambers, cells expressing FPRL2 migrated in a bell-shaped dose-dependent manner in response to hexapeptide concentrations ranging from 1 nM to 1 M. Maximal cell migration was observed at 10 -50 nM WKYMVm (Fig. 5) and 500 -1000 nM WKYMVM (data not shown). To make sure that migration to the lower chamber was not due to chemokinesis, we added increasing concentrations of hexapeptide to the upper chamber. This did not increase spontaneous migration toward the control medium, but prevented migration to lower chambers containing the hexapeptide (data not shown), suggesting that migration of HL-60-FPRL2 cells in a concentration gradient is chemotaxis rather than chemokinesis.
Binding of Trp-Lys-Tyr-Met-Val-Met-NH 2 and Trp-Lys-Tyr-Met-Val-D-Met-NH 2 to FPRL1 and FPRL2-The binding parameters of the hexapeptides were determined with RINm5F cells that stably expressed either of the FPRs. The inability of WKYMVM to bind to FPR was assessed with the tritiated hexapeptide Trp-Lys-[3,5-3 H 2 ]Tyr-Met-Val-Met-NH 2 . The FPR-expressing cell line (RIN9815) has previously been shown to expose 1.4 ϫ 10 6 receptors/cell and to bind an N-formyl peptide derivative (42) with a K D of 21 nM (29). Despite a high level of expression of FPR in this cell line, no specific binding of [ 3 H]WKYMVM could be measured for concentrations ranging from 1 to 300 nM (Fig. 6). Specific binding was easily measured with RINm5F-FPRL1 cells, where Ͼ95% of the bound radioactivity could be displaced by an excess of nonradioactive peptide (Fig. 6). Plotting the data from the equilibrium binding study according to Scatchard revealed a K D of 160 nM and a level of expression of ϳ1.8 ϫ 10 6 receptors/cell (Fig. 6, inset). Due to the low specific radioactivity of the tritiated ligand and to a low level of surface receptor expression, we were unable to determine the binding parameters of [ 3 H]WKYMVM with HL-60-FPRL1 cells. However, a competitive displacement assay showed that the half-maximal concentration of WKYMVM to 2 No calcium response was observed when cells were centrifuged (and not simply diluted) at each passage. When this step was omitted, a low level of constitutive expression of FPRL1 occurred since WKYMVM induced a small calcium rise in control or FPR-expressing HL-60 cells. This is probably due to the fact that a small population of cells becomes differentiated under these conditions. However, the amplitude of the response remained small (between 25 and 75 nM) compared with that obtained with FPRL2-expressing cells ([Ca 2ϩ ] i Ͼ 550 nM). displace 125 I-labeled Trp-Lys-Tyr-Met-Val-D-Met-NH 2 was ϳ120 -180 nM (data not shown). This is consistent with the K D value determined with RINm5F-FPRL1 cells.
The calcium mobilization assay suggests that the L-and D-methionine-containing hexapeptides have a lower affinity for FPRL2 than for FPRL1. We therefore determined the concentration of each peptide that half-maximally inhibited the binding of 125 I-labeled Trp-Lys-Tyr-Met-Val-D-Met-NH 2 in RINm5F-FPRL1 and RINm5F-FPRL2 cells. As shown in Fig. 7, WKYMVm was the best competitor for binding of 125 I-labeled WKYMVm to FPRL1 (IC 50 ϳ 30 -40 nM versus 130 -150 nM for WKYMVM). This result is consistent with the observation that the half-maximal concentration of WKYMVm to mobilize calcium in HL-60-FPRL1 cells is lower than that of the L-methionine-containing conformer. Although WKYMVm was more potent than WKYMVM for the induction of a calcium increase in FPRL2-expressing HL-60 cells, the two hexapeptides had a similar ability to compete for the binding of 125 I-labeled WKYMVm to FPRL2 (IC 50 ϳ 2.5-3 M) (Fig. 7). Thus, the data indicate altogether that the hexapeptide WKYMVM is able to specifically bind to and activate neutrophils through FPRL1 and that both L-and D-methionine-containing hexapeptides are low-affinity agonists for the monocyte-restricted receptor, FPRL2.
Hexapeptide-induced Phosphorylation of FPRL1 and FPRL2-We further examined whether FPRL1 and FPRL2 were phosphorylated upon addition of either of the two hexapeptides. After metabolic labeling with [ 32 P]orthophosphoric acid, cells were incubated for various periods of time with no agonist, WKYMVM, or WKYMVm. After cell lysis, receptors were immunoprecipitated with affinity-purified rabbit immunoglobulins directed against the last 10 residues (PPAETELQAM) of FPRL1. These IgGs were found to cross-react with the corresponding sequence in FPRL2 (PPEETELQAM), but not with the C-terminal sequence of FPR (LPSAEVELQAK).
In the absence of ligand binding, there was no basal phosphorylation of FPRL1. After the addition of the L-or D-hexapeptide, FPRL1 was rapidly phosphorylated. As shown in Fig. 8A, the time courses of the responses induced by the two peptides were very similar (t1 ⁄2 ϳ 5-10 min). The immunoprecipitated materials migrated on SDS-polyacrylamide gel as two radioactive species with relative molecular masses of 50 and 100 kDa (Fig. 8B). The band at 100 kDa is specific of FPRL1 and may represent a dimeric form of the receptor since it was absent from immunoprecipitates that derived from stimulated RINm5F-FPRL2 cells or HL-60-FPRL2 cells (see Fig. 9). Such a pattern of migration is very common with G protein-coupled receptors (GPCRs) whether receptors are glycosylated or not. Dimeric species have been documented for a number of GPCRs, including the ␤ 2 -adrenergic receptor (43) and the C5a anaphy-latoxin receptor (32, 34) (for a review, see Ref. 44). Whether these higher molecular mass species have a functional role in signal transduction is still a matter of debate.
Unexpectedly, increasing concentrations of either WKYMVM or WKYMVm had a differential effect on the phosphorylation of FPRL1. Although FPRL1 phosphorylation steadily increased with the concentration of WKYMVM, the level of receptor phosphorylation reached a maximum at 0.1 M WKYMVm and then decreased at higher concentrations (Fig.  8B). The same discrepancy in the pattern of phosphorylation was observed when FPRL1 was expressed in Chinese hamster ovary cells (data not shown).
Likewise, RINm5F-FPRL2 cells were stimulated with increasing concentrations of either WKYMVM or WKYMVm. Although FPRL2 and FPRL1 have the same number of amino acids, FPRL2 had a slower electrophoretic mobility, which is consistent with an additional glycosylation site in the N-terminal region. In the absence of stimulation, FPRL2 was immunoprecipitated as an already phosphorylated protein that migrated on SDS-polyacrylamide gel with a relative molecular mass of 70 -80 kDa (Fig. 9A). This band was not recovered from lysates of RINm5F-FPRL1 cells (see Fig. 8B). As in the case of FPRL1, phosphorylated species with higher molecular masses were detected that may represent a dimeric form or receptor aggregates. After the addition of either of the two hexapeptides, the level of receptor phosphorylation increased only slightly. A similar pattern of phosphorylation was observed when FPRL2 was immunoprecipitated from lysates of undifferentiated HL-60-FPRL2 cells (Fig. 9B). This suggests that the high level of FPRL2 phosphorylation in unstimulated cells is not linked to receptor overexpression since HL-60-FPRL2 cells expressed physiologic amounts of receptors (10 4 to 2 ϫ 10 4 receptors/cells). There was no trace of phosphorylated FPRL1 present in the immunoprecipitates from lysates of undifferen- tiated HL-60 and HL-60-FPRL2 cells, confirming that FPRL1 is not constitutively expressed in undifferentiated HL-60 cells and undifferentiated HL-60-FPRL2 cells.
Hexapeptide-mediated Internalization of FPRL1 and FPRL2 in RINm5F Cells-Previous studies with chemoattractant and chemokine receptors have indicated that the phosphorylation of the receptor C-terminal domain is a key step for agonistmediated receptor internalization (32,(45)(46)(47). Therefore, we examined the subcellular distribution of FPRL1 and FPRL2 in RINm5F cells by confocal immunofluorescence microscopy. A complete absence of staining was observed with non-transfected RINm5F cells or cells transfected with FPR, indicating that there was no cross-reactivity of the antibody with the C-terminal extremity of FPR (data not shown). In the absence of agonist, FPRL1 remained evenly distributed at the cell surface, as illustrated by the bright fluorescent staining of the plasma membrane (Fig. 10, A and B, upper left panels). The addition of WKYMVM rapidly led to the formation of small receptor-containing vesicles that moved to the subcortical area after 10 min. Larger vesicles that accumulated in the perinuclear region were observable by 20 and 30 min. When cells were stimulated with WKYMVm, a majority of the cells still presented a bright fluorescent staining of the plasma membrane after 10 min (Fig. 10B, upper right panel). By 20 min, subcortical and cytoplasmic fluorescent vesicles were nevertheless observable; and by 30 min, the perinuclear region was brightly stained in a majority of the cells (Fig. 10B, lower panels). Thus, despite an apparently poor ability to induce FPRL1 phosphorylation, receptors are internalized upon WKYMVm application.
Likewise, the hexapeptide-mediated internalization of FPRL2 was examined after stable expression in RINm5F cells (Fig. 11). Unlike FPRL1, FPRL2 was not evenly distributed at the plasma membrane of unstimulated cells. A punctate fluorescence at or underneath the plasma membrane was observed, which may represent receptors already in endocytic vesicles. In ϳ10 -20% of the cells, fluorescent vesicles were seen farther down in the cytoplasm. The addition of either L-or D-methionine-containing peptide led to the accumulation of fluorescent cytoplasmic vesicles that were best observed in a majority of the cells after 20 min. Thus, like FPRL1, FPRL2 is internalized upon binding of either of the two hexapeptides. This provides additional support to the notion that FPRL2 is a functional receptor that can be activated by the synthetic hexapeptides WKYMVM and WKYMVm and transduce signals similar to those mediated through FPR and FPRL1.

DISCUSSION
The G protein-coupled seven-transmembrane receptors FPR, FPRL1, and FPRL2 belong to the N-formyl peptide chemoattractant receptor family (48). The first two of them have been identified in both neutrophils and monocytes, whereas FPRL2 is expressed in monocytes, but not in neutrophils and differen- tiated HL-60 cells (24). The prototypical chemoattractant receptor FPR binds to fMLF with high affinity, whereas the homologue FPRL1 binds the same agonist with very low affinity (16). Regarding FPRL2, no known natural or synthetic ligand has thus far been documented. Here, we show that FPRL2 can trigger calcium mobilization in myeloid cells following application of nanomolar concentrations of the newly described neutrophil and monocyte synthetic activators, Trp-Lys-Tyr-Met-Val-D/L-Met-NH 2 (WKYMVm/M) (11).
In the past few years, several studies have shown that FPRL1 is able to bind a broad spectrum of ligands that do not bear any sequence homologies to each other or to the classical FPR agonist fMLF. LXA 4 , a lipid mediator originally reported to inhibit neutrophil activities mediated by other chemoattractants (for example, fMLF-induced chemotaxis and granule secretion), was first described as a high-affinity ligand for FPRL1. However, LXA 4 has not been documented to initiate chemotaxis of and intracellular calcium increases in either neutrophils (17) or FPRL1-transfected HEK293 cells (20). In contrast, all peptide and protein agonists of FPRL1 (see the Introduction) have the capacity to induce chemotaxis of FPRL1-expressing cells and to trigger intracellular calcium elevation despite relatively modest affinities for FPRL1 (K D values in the high nanomolar to micromolar range).
The results of this study establish that the hexapeptide WKYMVM activates neutrophils via FPRL1, but not via FPR. First, WKYMVM does not induce intracellular calcium mobilization in FPR-expressing HL-60 cells. Second, WKYMVM does not desensitize the fMLF-mediated superoxide release in neutrophils, indicating that it does not cross-react with FPR. Third, [ 3 H]WKYMVM does not bind to FPR-expressing cells. In monocytes, WKYMVM is an agonist for FPRL1 and FPRL2. We show here that FPRL2 is a functional receptor since it triggers calcium flux in and migration of myeloid cells in response to WKYMVm/M binding.
In view of the high degree of amino acid identity between FPRL1 and FPRL2 and judging from the ability of WKYMVm/M to mediate both calcium mobilization in and migration of FPRL1-expressing cells (12,13) and undifferentiated FPRL2-expressing cells, it is likely that these two receptors share many signal transduction steps and that they have similar roles in host defense. It is worth noting here that, despite previous studies showing that the WKYMVm/M hexapeptides mediate phosphoinositide hydrolysis and calcium mobilization in HL-60 cells, even undifferentiated HL-60 cells (10,11), we exclude the possibility that the hexapeptide-mediated activation of HL-60-FPRL2 cells could result from constitutive expression of FPRL1. First, in contrast to Ryu and co-workers (10, 11), we did not observe calcium mobilization with undifferentiated control cells, 2 i.e. naive non-transfected cells or cells transfected with either CXCR4 (data not shown) or FPR (Fig.  3A). Second, if FPRL1 had been constitutively expressed in HL-60-FPRL2 cells, an fMLFK-mediated calcium response would have been observed (Fig. 3C) since fMLFK also activates FPRL1 (13). Third, HL-60-FPRL2 cells did not respond to serum amyloid A, an agonist that elicits calcium mobilization in FPRL1-expressing HEK cells (20) and HL-60-FPRL1 cells (data not shown). Finally, phosphorylated FPRL1 was not detected in lysates of WKYMVM-stimulated HL-60-FPRL2 cells.
Attenuation of the chemoattractant-evoked responses rapidly occurs despite a persistent stimulation of the receptor. This attenuation of responsiveness is thought to result from the desensitization of receptors through their phosphorylation and rapid removal from the cell surface by endocytosis (49,50). Agonist binding to GPCRs is thought to bring about a conformational change that allows interactions with either the cognate G protein(s) or a specific kinase(s) that seems critical for GPCR regulation (49,51,52).
Surprisingly, the stimulation of RINm5F-FPRL1 cells with increasing concentrations of either WKYMVM or WKYMVm leads to distinct patterns of FPRL1 phosphorylation. The higher efficacy of WKYMVm for ligand displacement and intracellular calcium mobilization is not correlated with a higher ability to induce FPRL1 phosphorylation and internalization. At first glance, this departs from the general model for GPCR regulation since there is, in general, a good correlation between the efficacy of ligands in GPCR activation and phosphorylation (53,54). This paradox could be explained if one considers the possibility that the binding of WKYMVm (but not WKYMVM) induces a conformational change that impairs the recognition of FPRL1 by the antibody we used. However, it has not been possible to directly test this hypothesis by immunoblot analysis of FPRL1 in immunoprecipitates since FPRL1 migrates at the same level as IgG heavy chains. An impaired interaction between FPRL1 and the antibody could result from the WKYMVm-mediated phosphorylation of the threonine residue present in the antigenic epitope (PPAETELQAM). Such a scenario could also explain why FPRL1 appears to be distributed in the plasma membrane and not in fluorescent vesicles by 10 min after the addition of WKYMVm. Judging from the time course of phosphorylation, FPRL1 is not fully phosphorylated after 10 min. Consequently, the antibody would stain the plasma membrane, i.e. non-phosphorylated receptors, but not the endocytic vesicles that should exclusively contain phosphorylated receptors. Endocytic vesicles would remain undetectable until receptors become dephosphorylated at later time points.
Thus, it appears that closely related agonists can induce FIG. 11. Internalization of FPRL2 in RINm5F cells. Shown are confocal immunofluorescence microscopic images of the medial plane of cells that were not stimulated or stimulated with 10 M WKYMVM or WKYMVm for 20 min at 37°C. FPRL2 was unevenly distributed at/ near the plasma membrane of unstimulated cells. After stimulation with either of the two peptides, FPRL2 accumulated in intracellular vesicles. distinct activated conformations of FPRL1 that result in a differential phosphorylation. Conformationally distinct forms of FPRL1 may be involved in coupling with the cognate G protein(s) and in the interactions with a specific kinase(s). This could explain the disproportion between the affinities for FPRL1 and the relative efficacy of the two peptides for calcium mobilization. Although the affinity of WKYMVm is only 3-4fold higher than that of WKYMVM, WKYMVm is a 25-fold more potent agonist for calcium mobilization. This suggests that WKYMVm induces a conformation allowing a better coupling with the G protein.
Surprisingly, despite a high level of amino acid identity to FPRL1, FPRL2 has a distinct behavior regarding the phosphorylation process and its distribution in unstimulated cells. In contrast to FPRL1, FPRL2 displays a strong basal phosphorylation in the absence of agonist, suggesting that it may undergo constitutive phosphorylation on a number of residues without any visible interference in the coupling with the G protein. This hypothesis is further supported by the observation that the level of receptor phosphorylation is only slightly increased upon agonist binding, as if only a few additional serine/threonine residues were unmasked and able to incorporate radioactive phosphate. Constitutive phosphorylation would be consistent with the punctate pattern of receptor distribution at/near the plasma membrane of unstimulated RINm5F-FPRL2 cells. In addition, fluorescent vesicles with an intracellular localization are observed in ϳ10 -20% of cells. Such a pattern suggests that FPRL2 may be internalized at a low rate in the absence of stimulation. An agonist-mediated internalization nevertheless occurs since FPRL2 is rapidly concentrated inside the cells following stimulation with either of the two hexapeptides.
In summary, the hexapeptide WKYMVM appears to be a more specific agonist that the D-methionine-containing peptide. WKYMVM provides a means to investigate more thoroughly how the binding of LXA 4 to FPRL1 can inhibit neutrophil functions otherwise activated through the same receptor by other agonists. The orphan receptor FPRL2 is temporarily adopted by the synthetic hexapeptides WKYMVm/M. It is clear that FPRL2 is a functional receptor with regard to chemotaxis of myeloid cells and agonist-mediated calcium flux and internalization. The next challenge will be the isolation of a natural ligand able to activate this receptor. This may take time, but the availability of radioactive synthetic agonists will undoubtedly be of great use in the screening of antagonists.