Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The chemokine families of proteins are emerging as essential factors in the recruitment and activation of immune and inflammatory cells (Oppenheim et al., 1991; Baggiolini et al., 1994). These proteins produce their biological effect by an interaction with selected members of the seven transmembrane G protein-coupled family of receptors (Horuk, 1995; Ahuja et al., 1994). To date, five human CCRs have been cloned, expressed and characterized (Neote et al., 1993; Myers et al., 1995; Power et al., 1995; Daugherty et al., 1996; Samson et al., 1996). Although this approach has yielded information about the function of these receptors, it has limitations. For example, even though a cell may express mRNA for a receptor, the resultant protein may not be correctly processed so that active protein reaches the membrane (Myers et al., 1995). Furthermore, the use of heterologous expression systems has generated conflicting data with regard to the action of certain chemokines on these receptors (Daugherty et al., 1996; Kitura et al., 1996). Thus it is important to identify cells or cell lines that have these receptor endogenously to verify and extend the observations made in the recombinant systems.

Eosinophils have been implicated in the pathophysiology of asthma and allergic disorders (Gleich, 1990). CC-chemokines have been shown to recruit and activate these cells (Kameyoshi et al., 1992; Pattison et al., 1995). RANTES was the first CC-chemokine shown to produce eosinophil chemotaxis (Rot et al., 1992; Pattison et al., 1995). In addition to RANTES, other members of the CC family of chemokines have been shown to be potent eosinophil chemoattractants, among them eotaxin (Ponath et al., 1996; Jose et al., 1994; Garcia-Zepeda et al., 1996), MCP-3 (Dahinden et al., 1994; Weber et al., 1995), MCP-2 (Noso et al., 1994; Weber et al., 1995) in vitro and MIP-1 in vivo (Lukacs et al., 1995). Using cloned receptors, it has been demonstrated that RANTES acts as an agonist on at least 4 of the 5 known CC chemokine receptors (Proudfoot et al., 1995; Power et al., 1995; Daugherty et al., 1996; Samson et al., 1996). In addition, MCP-3 appears to interact with at least three of the cloned receptors (Daugherty et al., 1996; Xu et al., 1995; Combadiere et al., 1995; Franci et al., 1995), and eotaxin is a selective agonist at CCR3 (Daugherty et al., 1996; Kitura et al., 1996). Peripheral blood eosinophils appear to express mRNA for two of these receptors, CCR1 and CCR3 (Daugherty et al., 1996; Kitura et al., 1996). Unfortunately, it is difficult to obtain sufficiently large quantities of eosinophils to characterize these endogenously expressed receptors fully and to compare this pharmacology with that obtained in recombinant expression systems. To study these receptors, we identified a cell line, EoL-3 cells, that when differentiated with sodium butyrate express high levels of these receptors. More important for drug discovery, these receptors are coupled to endogenous signal transduction machinery.

The EoL-3 cell line was produced from an individual with eosinophilic leukemia (Saito et al., 1985; Mayumi, 1992). This cell line is hyperdiploid with the karyotype of 49 XY (+4, +8, 9q, +13) (Saito et al., 1985; Mayumi, 1992). Unlike a sister cell line (EoL-1), EoL-3 cells cannot be induced to differentiate into eosinophilic granule-containing cells (Mayumi, 1992). EoL-3 cells do express CD23, and its expression can be increased by IL-4 treatment and decreased by TGF- (Mayumi, 1992) Like normal eosinophils, EoL-3 express FCRII with little or no expression of FCRI or FcRIII; however, IFN- will induce the expression of FCRI in these cells (Mayumi, 1992). Both eosinophils and EoL-3 cells express FcR receptors (Mayumi, 1992). Thus, although they do not possess all the characteristics of eosinophils, EoL-3 cells provide an useful model with which to study function (Saito et al., 1985; Mayumi, 1992).

The purpose of the present studies was to examine the expression of CC-chemokine receptors in the EoL-3 cell line and to characterize the pharmacology of these receptors. As determined by 3RACE with nested PCR, EoL-3 cells express mRNA for CCR1, CCR2 and CCR3 with a low signal detected for CCR5 and no signal for CCR4. Furthermore, using the results obtained in both binding studies and functional experiments, we were able to demonstrate the presence of CCR1 and CCR2 in these cells. Thus EoL-3 cells can serve as a useful model system to identify selective receptor antagonists for these receptors.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

EoL-3 Cell Differentiation

EoL-3 basal cells were grown in RPMI 1640 medium with 10% FBS and 25 mM HEPES (pH 7.6) and antibiotics for 2 to 3 weeks. The cells were differentiated by growing in the same medium with 0.5 mM sodium butyrate for an additional 10 days (Saito et al., 1985; Fischkoff et al., 1986).

CC Receptor Expression

Total RNA was prepared from dEoL-3 cells using the RNazole kit according to the manufacturer's protocol. One microgram of total RNA (1 mg/ml) and 1 µl of adapter primer (10 µM, Life Technologies, Grand Island, NY) in a volume of 11 µl was heated at 70°C for 10 min and then cooled on ice for 5 min, followed by the addition of 2 µl of 10X PCR buffer (200 mM Tris-HCl, pH 8.4, 500 mM KCl) 2 µl of 25 mM MgCl₂, 1 µl of 10 mM dNTPs and 2 µl of 0.1 M DTT. The reaction was incubated at 42°C for 5 minutes, followed by the addition of 1 µl of SuperScript II reverse transcriptase (Life Technologies), and then incubated at 42°C for 50 min. The reaction was terminated by a 15-min incubation at 70°C, followed by chilling on ice. The RNA was degraded by the addition of 1 µl of RNaseH for 20 min at 42°C. One microliter of each reverse transcription reaction was then subjected to 30 cycles of PCR (95°C, 30 sec; 45°C, 2 min; 72°C, 1.5 min), followed by an extension reaction of 10 min at 72°C in a 50-µl reaction mixture containing 5 µl of 10X PCR buffer, 3 µl of 25 mM MgCl₂, 1 µl of 10 mM dNTPs, 1 µl of the 10 µM abridged universal amplification primer (AUAP, Life Technologies) and 1 µl of transmembrane (TM) 2 primer (5-GGG AAT TCA ACC TGG CC(A/T) T(T/G)(T/G) C(T/G)G ACC T-3). Each 3RACE reaction was diluted 10-fold in 10 mM Tris-HCl, pH 8, 1 mM EDTA, pH 8. One microliter of each diluted 3RACE reaction was then subjected to 30 cycles of PCR (95°C, 30 sec.; 48°C, 2 min; 72°C, 1.5 min), followed by an extension reaction of 10 min at 72°C in a 100-µl reaction mixture containing 10 µl of 10X PCR buffer I (100 mM Tris-HCl, pH 8.3, 500 mM KCl, 25 mM MgCl₂), 8 µl of 2.5 mM dNTPs, 2 µl of 10 µM of TM2 primer (5-GGG AAT TCA ACC TGG CC(A/T) T(T/G)(T/G) C(T/G)G ACC T-3), 2 µl of 10 µM TM3 primer (5-GCC CTC GAG GAC (G/A)AT (G/A)GC CAG GTA (C/T/A)C(T/G) (G/A)TC-3) and 1 µl of AmpliTaq DNA polymerase. To control for the efficiency of the reverse transcriptase and 3RACE reactions, glyceraldehyde-3-phosphate dehydrogenase mRNA levels were examined using forward (5-ACC ACA GTC CAT GCC ATC AC-3) reverse (5-TCC ACC ACC CTG TTG CTG TA-3) primers. After nested PCR, the reaction products were separated on a 1.5% agarose gel in 1X TBE, and the DNA was transferred onto a nytran filter by capillary action in 20X SSC and then hybridized at 65°C in 0.5 M sodium phosphate, pH 7.2, 1% BSA, 7% SDS, 1 mM EDTA, pH 8, using ³²P-labeled probes generated by random priming (Promega). After hybridization, the blot was washed at room temperature in 1X SSC + 0.1% SDS, 3 times for 30 min, and then washed at 65°C in 0.1X SSC + 0.1% SDS, 3 times for 30 min. The blot was exposed for 1 day with X-ray film.

Radioligand Binding Studies

Whole-cell binding. Washed dEoL-3 cells (Dulbecco's phosphate-buffered saline) were resuspended in RPMI with 0.1% BSA, and 25 mM HEPES (pH 7.4) and 0.05% NaN₃ (reaction buffer) at 1 to 2 × 10⁷ cells/ml. Cells (0.5-2 × 10⁶) were incubated with ¹²⁵I-RANTES (0.3 nM) or ¹²⁵I-MIP-1 (0.15 nM) in the absence or presence of unlabeled chemokine (30-100 nM) for 90 min at room temperature in an Eppendorf microcentrifuge tubes (final reaction volume 200 µl). The binding reaction was terminated by placing the incubation mixture over a 10% sucrose cushion (750 µl) and centrifuging at 14,000 rpm for 2 to 3 min to separate bound from free ligand. The resultant supernant fraction was discarded, and the amount of the radioactivity associated with the pellet was determined by gamma scintillation spectrometery. Alternatively, the identical reaction was carried out in a 96-well plate on an orbital shaker (150 rpm). In this case, the reaction was terminated by rapid filtration using a 96-well plate harvester (Packard Unifilter-96 Harvester). Filters were washed 12 times with phosphate-buffered saline containing 0.1% BSA and 0.05% NaN₃. The filters were allowed to dry overnight, and 50 µl of scintillation fluid (Packard's Micro Scint 20) was added to each well. The amount of radioactivity bound to the filters was determined by liquid scintillation spectrometry. Nonspecific binding was determined in the presence of 30 to 100 nM unlabeled chemokine. Qualitatively similar results were obtained with both methods of separation.

EoL-3 cell Ca⁺⁺ mobilization. dEoL-3 cells were cultured as previously described and washed with phosphate-buffered saline. Cells were suspended at 1 × 10⁶ cells/ml in KRH buffer (118 mM NaCl, 4.6 mM KCl, 25 mM NaHCO₃, 1 mM KH₂PO₄ and 11 mM glucose) containing 50 mM HEPES, pH 7.4, 1 mM CaCl₂, 1 mM MgCl₂, 0.1% BSA and 2 µM Fura-2AM and incubated for 45 min at 37°C. Cells were centrifuged at 200 × g for 3 min, resuspended in the same buffer without Fura-2AM, incubated for 15 min at 37°C to complete the hydrolysis of intracellular Fura-2AM and centrifuged as before. Cells (5 × 10⁵ cells/ml) were resuspended in cold KRH with 50 mM HEPES (pH 7.4), 1 mM CaCl₂, 1 mM MgCl₂ and 0.1% gelatin and maintained on ice until assayed. Chemokine-induced fluorescence was measured in a fluorometer (Johnson Foundation Biomedical Group, Philadelphia, PA) with magnetic stirring, and the temperature was maintained at 37°C. Excitation was set at 340 nm and emission at 510 nm. Maximal Ca⁺⁺ attained after agonist stimulation was calculated as described by the method of Grynkiewicz et al., 1985.

Materials

¹²⁵I-RANTES (specific activity 2200 Ci/mmol), ¹²⁵I-MIP-1 (specific activity 2200 Ci/mmol) and ¹²⁵I-MCP-1 (specific activity 2200 Ci/mmol) were obtained from either New England Nuclear Research Products (RANTES and MIP-1; Boston, MA) or Amersham (MCP-1, Arlington, Heights, IL). The recombinant chemokines RANTES, MIP-1, MIP-1, MCP-1, MCP-2 and MCP-3 were purchased from either Preprotech (Rocky Hill, NJ) or R&D Systems Inc. (Minneapolis, MN). Fura-2AM was purchased from Calbiochem (San Diego, CA). Buffers, salts and protease inhibitors were obtained from Sigma Chemical Corp. (St. Louis, MO). 3RACE primers were obtained from Life Technologies. Filter plates and scintillation fluid were obtained from Packard Instrument Co. (Meriden, CT).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

To examine the expression of CCRs in dEoL-3 cells, we used a modified RT-PCR analysis that consisted of two separate and distinct PCR reactions. The first PCR reaction was based on 3RACE methodology that used degenerate primer to the conserved amino acid sequence NLAISDL within transmembrane region 2 and nondegenerate primer complementary to 5 end of the oligo dTprimer used to prime the first strand cDNA reaction. The 3RACE PCR step is followed by a nested PCR reaction using the same degenerate primer to TM2 and another degenerate primer to the conserved amino acid sequence DRYLAIV in transmembrane region 3. These primers were chosen because the amino acid sequence of the extracellular loop region between transmembrane regions 2 and 3 is highly divergent among the chemokine receptors and thus serves as a signature region. This protocol enabled us to analyze the distribution of several chemokine receptors simultaneously, and the nested PCR increased the sensitivity of this assay. Southern hybridization of the PCR products demonstrated that CCR1 and CCR2 receptors were expressed in EoL-3 cells (fig. 1). Furthermore, we detected mRNA for CCR3 and low levels of mRNA for CCR5 (fig. 1). Thus dEol-3 cells express mRNA for 4 of the 5 well-characterized CC chemokine receptors.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Representative Southern analysis of the expression of mRNA for CCRs in dEoL-3 cells. Expression of CCRs using a modified RT-PCR and Southern hybridization assay as described in "Materials and Methods."

To examine whether the receptors detected by Southern analysis were expressed on the surface of dEoL-3 cells, we examined the binding of ¹²⁵I-RANTES, ¹²⁵I-MIP-1 and ¹²⁵I-MCP-1 to EoL-3 cells. Labeled RANTES and MIP-1 demonstrated saturable binding to whole EoL-3 cells (fig. 2). Scatchard analyses of these data yielded the following K_d values: MIP-1, 1.4 nM; RANTES, 7 nM. These data were consistent with the presence of CCR1 and possibly CCR3 on the surface of dEoL-3 cells (Proudfoot et al., 1995; Neote et al., 1993; Daugherty et al., 1996). To differentiate between these two receptors, we examined the ability of CC chemokines to displace both ¹²⁵I-MIP-1 and ¹²⁵I-RANTES. If both ligands interact with CCR1, then the rank order for displacement of ¹²⁵I-MIP-1 and ¹²⁵I-RANTES would be similar. On the other hand, if MIP-1 does not bind to CCR3, then the rank order of chemokines for displacement of MIP-1 binding should be distinct from that observed for RANTES binding. As shown in figure 3, several CC chemokines had exactly the same ability to inhibit ¹²⁵I-RANTES or MIP-1. Unlabeled MIP-1, RANTES and MCP-3 were equipotent in displacing both ligands. MIP-1 was about 10-fold less potent, and MCP-1 and the CXC chemokine IL-8 were inactive.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Saturation binding of 125I-RANTES and 125I-MIP-1 to dEoL-3 cells. Increasing concentrations of 125I-RANTES (panel A) and 125I-MIP-1 (panel B) were added to dEoL-3 cells in the absence () or presence () or 100 nM unlabeled ligand. Specific binding () was defined as the difference between total binding and that observed in the presence of 100 nM unlabeled chemokine. The graph is representative of three experiments run in duplicate.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Competition binding of 125I-RANTES or 125I-MIP-1 to dEoL-3 cells. dEoL-3 cells were incubated with 0.3 nM 125I-RANTES (panel A) or 0.15 nM 125I-MIP-1 (panel B) in the absence or presence of increasing concentrations of RANTES, (); MIP-1, (); MIP-1, (); MCP-3, (); MCP-1, () and IL-8 (). The data represent the mean ± S.E.M. of two to six experiments.

We examined the binding of ¹²⁵I-MCP-1 to these cells because Southern analysis indicated the presence of CCR2b mRNA. ¹²⁵I-MCP-1 demonstrated saturable binding to dEoL-3 cells with a K_d value of 0.4 nM (fig. 4). Furthermore, when the ability of CC chemokines to displace ¹²⁵I-MCP-1 binding to dEoL-3 cells was examined, MCP-1, MCP-2 and MCP-3 produced a concentration-dependent displacement of labeled MCP-1 (fig. 5). At concentrations to 100 nM, MIP-1, MIP-1 and RANTES had no significant effect on this binding. The rank order of the affinity of these CC chemokines for this site was MCP-3  MCP-1 > MCP-2 >>> MIP-1 ~ MIP-1 ~ RANTES.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Saturation binding of 125I-MCP-1 to dEoL-3 cells. Increasing concentrations of 125I-MCP were added to dEoL-3 cells in the absence () or presence () of 100 nM unlabeled MCP-1. Specific binding () was defined as the difference between the binding that occurred in the absence of unlabeled 00 nM MCP-1 and that which occurred in its presence.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Competition binding of 125I-MCP-1 to dEoL-3 cells. dEoL-3 cells were incubated in the presence of 0.7 nM 125I-MCP-1 in the absence or presence of increasing concentrations of MCP-1, (); MCP-2, () and MCP-3, (). The data are expressed as a percent of control binding and are the mean ± S.E.M. of three experiments.

One of the drawbacks when using radioligand binding to study receptors is that although this technique provides information on the affinity of a given ligand for a receptor, it does not supply any data on the efficacy or effectiveness of the ligand at that receptor. Furthermore, it is difficult to prove experimentally, using ligand binding of agonists alone, whether two substances are interacting at the same site (Swillens et al., 1995). To explore the functional properties of the chemokine receptors present on dEoL-3 cells, we determined the ability of recombinant chemokines to increase intracellular Ca⁺⁺. Addition of MCP-1, MCP-3, RANTES or MIP-1, but not MIP-1, produced a concentration-dependent increase in intracellular Ca⁺⁺ concentration in dEoL-3 cells (fig. 6). Although the maximal response produced by MIP-1 and that produced by RANTES were similar, the magnitude of the Ca⁺⁺ response elicited by either MCP-1 or MCP-3 was approximately doubled. Furthermore, in a result consistent with the binding data, MIP-1 was more potent than RANTES, and MCP-3 was more potent than MCP-1 (fig. 6). To identify the receptor responsible for the Ca⁺⁺ signal produced by addition of these chemokines, we used the technique of receptor desensitization. Prior treatment of dEoL-3 cells with a maximal dose (100 nM) of either RANTES or MIP-1 abolished the subsequent response of those cells to a second challenge with the same agonist or to the alternative agonist (fig. 7A). Neither MCP-1 nor MIP-1 inhibited the response to MIP-1 or RANTES (fig. 7B). These results suggest that both MIP-1 and RANTES interact at the same receptor and support the hypothesis that this receptor is CCR1. In the same experiment, we also examined the ability of RANTES and MIP-1 to abolish or inhibit the response to MCP-3, because we had shown previously that this chemokine was capable of displacing both ¹²⁵I-MIP-1 and ¹²⁵I-RANTES. Using either RANTES or MIP-1, we observed that initial challenge of dEoL-3 cells with a maximal concentration of MCP-3 (10 nM) abolished the response to either MIP-1 or RANTES (fig. 7C). However, after initial challenge with either RANTES or MIP-1, the response to MCP-3 was inhibited but not abolished (fig. 7C). These results are in agreement with the data obtained from the ligand binding experiments demonstrating that MCP-3 inhibited the binding of MIP-1 or RANTES. These data also suggest that MCP-3 is interacting at a second distinct site (Combadiere et al., 1995; Ben-Baruch et al., 1995). Because the binding studies demonstrated that MCP-3 was able to displace ¹²⁵I-MCP-1 binding to dEoL-3 cells, desensitization studies with MCP-1 and MCP-3 were performed. In a manner similar to that seen with MIP-1 and with RANTES, prior challenge with MCP-1 inhibited but did not abolish the response to MCP-3 (fig. 8A). Pretreatment of dEoL-3 cells with both RANTES and MCP-1 or both MCP-1 and MIP-1 abolished the response to MCP-3 (fig. 8B). These data are consistent with MCP-3 interacting with both CCR1 and CCR2. The functional studies, combined with the binding experiments, support the hypothesis that dEoL-3 cells express at least two functional CC chemokine receptors, probably CCR1 and CCR2.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Calcium mobilization in dEoL-3 cells in response to CC chemokines. dEoL-3 cells were loaded with Fura-2 and the maximal Ca++ concentration achieved for increasing concentrations of RANTES, (); MIP-1, (); MIP-1, (); MCP-1, () and MCP-3, (). The data are representative of three to six experiments.

View larger version (38K): [in this window] [in a new window]   Fig. 7.   Desensitization of Ca++ transients in dEoL-3 cells to characterize the CCR1 receptor. Maximal Ca++ levels were achieved and determined after first and second challenge. A) Cross-desensitization of Ca++ transients by either RANTES or MIP-1. B) Inability of MIP-1 or MCP-1 to abolish the RANTES-induced Ca++ transient. C) MCP-3 abolishes the response to either MIP-1 or RANTES, whereas these chemokines inhibit but do not abolish the MCP-3-induced Ca++ transient. Unless indicated in the graph, the concentration of chemokine was 100 nM. All concentrations used were maximally effective concentrations. The data are representative of three separate tests.

View larger version (45K): [in this window] [in a new window]   Fig. 8.   Desensitization of Ca++ transients in dEoL-3 cells to characterize the CCR2 receptor. Maximal Ca++ levels were determined after first and second challenge. A) Demonstration of the ability of MCP-1 and MCP-3 to produce Ca++ transients through CCR2. B) Demonstration of the ability of MCP-3 to signal through both CCR1 and CCR2 on dEoL-3 cells. Concentrations used were maximally effective concentrations as presented on the figure. The data presented are from a representative experiment of three separate tests.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

There has been a virtual explosion in the number of proteins shown to recruit and activate leukocytes. Many of these proteins and their receptors have been identified from molecular cloning efforts and the use of heterologous expression systems. With regard to these novel 7TM receptors, the use of heterologous expression systems is not without its pitfalls. For example, although mRNA for CCR2a can be found in HEK 293 cells transfected with the appropriate clone, it appears that little or no protein is expressed on the cell surface (Charo et al., 1994; Myers et al., 1995). Furthermore, depending on the level of surface expression of CCR3, different investigators have found that this receptor responds or fails to respond to MCP-3 and RANTES (Daugherty et al., 1996; Kitura et al., 1996). Thus it has become important to identify cells or cell lines that endogenously express these receptors to confirm the observations obtained in the expression systems and to provide additional systems for examining the function of novel chemokines.

In this report, we characterized the CCRs present on the dEoL-3 cell line. This cell line was obtained from an individual with eosinophilic leukemia and expresses certain characteristics found in eosinophils (Mayumi, 1992). Preliminary experiments with basal EoL-3 cells demonstrated significant specific binding of radiolabeled RANTES; however, no functional response could be obtained (data not shown). Culture of EoL-3 cells in the presence of 0.5 mM sodium butyrate, an eosinophilic differentiation protocol originally described for HL-60 cells (Fischkoff et al., 1986; Saito et al., 1985), transforms the appearance of these cells from blast cells into cells with a condensed nucleus and nonspecific staining granules. dEoL-3 cells grown under these conditions increased the number of RANTES/MIP-1a receptors, and the CC chemokines were now capable of producing a functional response. Differentiation of HL-60 cells into "eosinophilic-like" cells will also increase the number of RANTES/MIP-1 receptors (CCR1) (Van Riper et al., 1994). Thus dEoL-3 cells became an interesting cell system for characterizing the expression and function of chemokine receptors for the CC chemokines, e.g., RANTES, MIP-1, MCP-1 and MIP-1.

To begin to characterize these receptors, we examined the mRNA expression pattern in EoL-3 cells. Using a combination of 3RACE with nested PCR and Southern analysis for detection, we detected a significant level of expression of CCR1, CCR2, and CCR3 mRNA with a much lower level of CCR5 and no expression of CCR4. This expression pattern is between that observed in human monocytes and that observed in human eosinophils (Daugherty et al., 1996; Wang et al., 1993). Peripheral blood monocytes express message for CCR1, CCR2 and CCR5 but not for CCR3 or CCR4 (Van Riper et al., 1993; Samson et al., 1996; Daugherty et al., 1996; Power et al., 1995). Human eosinophils, on the other hand, express CCR1 and CCR3 message but do not express CCR2, CCR4 or CCR5 mRNA (Daugherty et al., 1996; Kitura et al., 1996). Thus the expression pattern of chemokine receptors parallels the other differentiation markers obtained with EoL-3 cells in that these characteristics are associated with both eosinophils and monocytes (Mayumi, 1992).

Inasmuch as binding studies indicated the presence of a RANTES/MIP-1 receptor on dEoL-3 cells, its pharmacological properties were compared with those of the initial CCR cloned, CCR1. CCR1 was cloned from a HL-60 cell library (Neote et al., 1993) and, when expressed in HEK 293 cells, demonstrated an apparently equal affinity for both RANTES and MIP-1 (Proudfoot et al., 1995; Neote et al., 1993). MIP-1 was much less potent at this receptor, and MCP-1 showed little or no activity (Neote et al., 1993; Gao et al., 1993). Interestingly, in functional assays, MIP-1 appeared more potent than RANTES in stimulating a calcium flux; however, MIP- and RANTES abolished the calcium response to themselves and the other chemokine (Proudfoot et al., 1995; Neote et al., 1993). Furthermore, subsequent studies demonstrated that MCP-3 was also a potent agonist at CCR1 and competed with equal affinity for either RANTES or MIP-1 binding (Combadiere et al., 1995; Ben-Baruch et al., 1995). The present radioligand binding and cellular functional studies indicate that the RANTES/MIP-1 site identified on dEoL-3 cells matches the characteristics of the cloned CCR1 receptor. It is noteworthy that the similarity between the chemokines in their ability to compete for ¹²⁵I-RANTES binding and ¹²⁵I-MIP-1 binding demonstrates labeling to CCR1 and not CCR3, because even though RANTES can bind to CCR3, MCP-3 is less potent at that receptor, and MIP-1 does not appear to interact with CCR3 (Daugherty et al., 1996; Kitura et al., 1996).

¹²⁵I-MCP-1 binding to dEoL-3 cells was examined because mRNA to CCR2 was detected by Southern analysis. Labeled MCP-1 demonstrated saturable and displaceable binding to dEoL-3 cells. Pharmacological characterization of this site indicated that MCP-3 and MCP-2, but not RANTES, MIP-1 or MIP-1, would compete for MCP-1 binding. The binding characteristics that CCR2 expressed in HEK 293 cells appeared quite similar to those observed in the present study (Ben-Baruch et al., 1995; Combadiere et al., 1995; Myers et al., 1995; Yamagami et al., 1997), and this suggests that the MCP-1 site detected on dEoL-3 cells is CCR2. Functional studies confirmed this hypothesis, because MCP-1 and MCP-3 inhibited the calcium signal to a subsequent challenge with MCP-1. In agreement with their inability to inhibit MCP-1 binding, RANTES, MIP-1 and MIP-1 did not inhibit the calcium response to MCP-1.

Although dEoL-3 cells contain message for CCR3, we were unable to confirm, with either ligand binding experiments or functional assays, that dEoL-3 cells express CCR3 on the cell surface. The reasons for this are unclear; however, it has been observed in heterologous expression systems that CCR2a mRNA can be found in transfected cells even though little or no CCR2a protein is detected on the cell surface (Myers et al., 1995; Charo et al., 1994).

In summary, we have demonstrated the expression of both CCR1 and CCR2 in human dEoL-3 cells and have shown that this cell line can be used to investigate the actions of novel chemokines. Furthermore, this model system can be used for the identification of selective receptor antagonists for two CC chemokine receptors, and such selective antagonists will become useful tools for determining the role of the various CCRs in regulating inflammatory cell function.