Positive Versus Negative Modulation of Different Endogenous Chemokines for CC-chemokine Receptor 1 by Small Molecule Agonists through Allosteric Versus Orthosteric Binding*

7 transmembrane-spanning (7TM) chemokine receptors having multiple endogenous ligands offer special opportunities to understand the molecular basis for allosteric mechanisms. Thus, CC-chemokine receptor 1 (CCR1) binds CC-chemokine 3 and 5 (CCL3 and CCL5) with Kd values of 7.3 and 0.16 nm, respectively, as determined in homologous competition binding assays. However, CCL5 appears to have a >10,000-fold lower affinity in competition against 125I-CCL3. Mutational mapping revealed that CCL3 and CCL5 both are strongly affected by systematic truncations of the N-terminal extension, whereas only CCL5 and not CCL3 activation is affected by substitutions in the main ligand binding pocket including the conserved GluVII:06 anchor point. A series of metal ion chelator complexes were found to act as full agonists on CCR1 and to be critically affected by the same substitutions in the main ligand binding pocket as CCL5 but not by mutations in the extracellular domain. In agreement with the overlapping binding sites, the small non-peptide agonists displaced radiolabeled CCL5 with high affinity. Interestingly, the same compounds acted as allosteric enhancers of the binding of CCL3, with which they did not overlap in binding site, leading to an increased Bmax and affinity of this chemokine mainly due to an increased association rate. It is concluded that a small molecule agonist through binding deep in the main ligand binding pocket can act as an allosteric enhancer for one endogenous chemokine and at the same time as a competitive blocker of the binding of another endogenous chemokine.

Chemokine receptors belong to class A rhodopsin-like 7 transmembrane (7TM) 3 receptors and are important for the control of leukocyte development, differentiation, and migration (1). The chemokine system is very complex with ϳ20 receptors and 50 endogenous ligands identified to date and with redundancy and promiscuity as key features. The chemokines are 8 -12-kDa proteins with usually two highly conserved stabilizing disulfides. They are divided into subfamilies based upon the absence (CCLs), or presence of one (CXCLs) or three (CX3CL) amino acids between the first two cysteines. Chemokines adopt a surprisingly conserved tertiary structure as shown by NMR and x-ray crystallography despite a low overall sequence homology. Thus, they all have a disordered N terminus (Ͻ10 amino acids prior to the first Cys residue), which is essential for signaling, and which is followed by the so-called N-loop, a 3 10 helix, a three-stranded ␤-sheet, and a C-terminal ␣-helix (2). The receptors are classified according to their preferred ligands in CCR1-11, CXCR1-7, CX3CR1, and XCR1 (2). CCR1 is very promiscuous and binds at least 8 endogenous CCLs with highest affinity for the two first identified: CCL3 4 (MIP-1␣) and CCL5 (RANTES) (3). Like many other chemokine receptors, CCR1 is subject of massive academic and industrial research as target for a large number of inflammatory and autoimmune diseases. Several CCR1 antagonists have been discovered, exemplified by the virus-encoded vCCL2 (4), and many different small non-peptide antagonists, of which BX471 is the most potent (5).
Small molecule non-peptide agonists within the chemokine system have been highly useful tools in studies of molecular mechanisms of 7TM receptor activation, for example in the CCR5 and CCR8 and a metal ion site mutated variant of the CXCR3 receptor (6 -8). In the latter case, small aromatic metal ion chelator complexes could be made highly efficacious agonists provided that they were anchored in the main ligand binding pocket at a position corresponding to the presumed binding site for agonistic catecholamines in the ␤2-adrenergic receptor (6 -9). In all cases the small molecule ligands were assumed to act as agonists by constraining TM-VI and -VII toward TM-III in an inwardly bent, proposed active conformation, in accordance with the toggle switch model (10).
In principle, allosteric modulators bind to sites that are located topographically distinct from the orthosteric binding sites (11). Small molecule ligands in the chemokine field are usually allosteric, due to the large size and broad "velcro-like" interaction of the chemokines with mainly the extracellular domains of their respective receptors. In classical terms, allosteric modulators do not possess efficacy, and either increase affinity and/or efficacy of the endogenous (orthosteric) ligands as so-called allosteric enhancers (or positive allosteric modulators) in a positive cooperative manner, or decrease the binding and/or efficacy of the orthosteric ligands as negative allosteric modulators (11,12). However, it has become increasingly evident that a number of allosteric enhancers in addition possess efficacy on their own as agonists and are able to activate their target receptor independently of the orthosteric agonists (13)(14)(15)(16).
In the present study we describe a series of metal ion chelator complexes that activate CCR1 with the same efficacy as the endogenous CCL3 and CCL5 chemokines. Mutational analysis of the CCR1 N terminus and the major and minor binding pockets indicate that the small molecule agonists bind in the major binding pocket, with the metal ion anchoring to the highly conserved Glu 287 located at the extracellular end of TM-VII (in position VII:06 5 ). In contrast, the N terminus of CCR1 is essential for chemokine binding and action in general, whereas the major binding pocket in addition is important for CCL5 but not for CCL3 binding. Consistent with the overlapping binding sites, the small molecule agonists displaced radiolabeled CCL5 with high affinity. Importantly, the same compounds acted as allosteric enhancers of CCL3 binding, with which they did not overlap in binding. To our knowledge, this is the first example of allosteric modulators that act in a dual manner with regard to endogenous agonists acting as allosteric enhancers for one endogenous chemokine and at the same time as a competitive blocker of the binding of another endogenous chemokine.
Site-directed Mutagenesis-Point mutations were introduced in the receptors by the polymerase chain reaction overlap extension technique using the wild-type CCR1 as template. All reactions were carried out using the Pfu polymerase (Stratagene) under conditions recommended by the manufacturer. The generated mutations were cloned into the eukaryotic expression vector pcDNA3.1ϩ. The mutations were verified by restriction endonuclease digestion and DNA sequencing (ABI 310, Perkin-Elmer Life Sciences).
Binding Experiments-COS-7 cells were transferred to culture plates 1 day after transfection. The number of cells seeded per well was determined by the apparent expression efficiency of the receptors and was aimed at obtaining 5-10% specific binding of the added radioactive ligand (1 ϫ 10 4 to 3 ϫ 10 5 cells/well for the different CCR1 constructs). Two days after transfection, cells were assayed by competition binding for 3 h at 4°C using 10 -15 pM 125 I-CCL5 or 125 I-CCL3 plus unlabeled ligands in 0.4 ml of a 50 mM Hepes buffer, pH 7.4, supplemented with 1 mM CaCl 2 , 5 mM MgCl 2 , and 0.5% (w/v) bovine serum albumin. After incubation, cells were washed quickly two times in 4°C binding buffer supplemented with 0.5 M NaCl. Nonspecific binding was determined as the binding in the presence of 0.1 M unlabeled CCL5 or CCL3, respectively. Determinations were made in duplicates.
Inositol Phosphate Turnover (IP Turnover)-COS-7 cells were transfected according to the procedure mentioned above. Briefly, 6 ϫ 10 6 cells were transfected with 20 g of receptor cDNA in addition to 30 g of the promiscuous chimeric G-protein, G␣ ⌬6qi4myr (abbreviated as G qi4myr ), which turns the G␣ icoupled signal, most common pathway for endogenous chemokine receptors, into the G␣ q pathway (phospholipase C activation measured as IP turnover) (17,18). One day after transfection, COS-7 cells (1.5 ϫ 10 5 cells/well) were incubated for 24 h with 2 Ci of myo[ 3 H]inositol in 0.3 ml of growth medium per well. Cells were washed twice in 20 mM Hepes, pH 7.4, supplemented with 140 mM NaCl, 5 mM KCl, 1 mM MgSO 4 , 1 mM CaCl 2 , 10 mM glucose, and 0.05% (w/v) bovine serum albumin; and were incubated in 0.3 ml of buffer supplemented with 10 mM LiCl at 37°C for 90 min in the presence of various concentrations of ligands. Cells were extracted by addition of 1 ml of 10 mM formic acid to each well followed by incubation on ice for 30 -60 min. The generated [ 3 H]inositol phosphates were purified on AG 1-X8 anion exchange resin. Determinations were made in duplicates.
Receptor Surface Expression by Enzyme-linked Immunosorbent Assay (ELISA)-COS-7 cells were transiently transfected with the N-terminal FLAG-tagged variants of CCR1. The cells were washed once in TBS (50 mM Tris-base, 150 mM NaCl, pH 7.6), fixed in 4% glutaraldehyde for 15 min following three washes in TBS and incubation in blocking solution (2% bovine serum albumin in TBS) for 30 min at room temperature. The cells were subsequently incubated 2 h with anti-FLAG (M1) antibody (2 g/ml) in TBS supplemented with 1% bovine serum albumin and 1 mM CaCl 2 at room temperature. After three washes in TBS with 1 mM CaCl 2 , the cells were incubated with goat anti-mouse horseradish peroxidase-conjugated antibody in the same buffer as the anti-FLAG antibody for 1 h. After three washes in TBS supplemented with 1 mM CaCl 2 the immune reactivity was revealed by the addition of horseradish peroxidase substrate according to the manufacturer's instruction.
Calculations-IC 50 and EC 50 values were determined by nonlinear regression and B max values calculated using the GraphPad-Prism 3.0 software (GraphPad Software, San Diego).

RESULTS
CCL3 and CCL5 Bind to CCR1 in a Ligand-dependent Manner-The binding of CCL3 and CCL5 was probed in homologous and heterologous competition binding experiments in transiently transfected COS-7 cells. As previously reported (3) CCL5 binds to CCR1 with high affinity, K d of 0.16 nM (Log K d Ϫ9.8 Ϯ 0.14, n ϭ 7), measured against itself ( 125 I-CCL5) as radioligand. In contrast, CCL5 was unable to compete against 125 I-CCL3 and a Ͼ10000-fold decrease in apparent affinity (n ϭ 7) was observed (Fig. 1A). On the other hand, the observed affinity of CCL3 was less dependent upon applied radioligand, because the affinities varied from a K d of 7.3 nM (Log K d Ϫ8.1 Ϯ 0.03, n ϭ 12) against itself ( 125 I-CCL3) to a K i of 9.9 nM (Log K i Ϫ8.0 Ϯ 0.15, n ϭ 7) against 125 I-CCL5 (Fig. 1B). Despite these similar affinities, the changes in the Hill coefficients (from Ϫ1.0 Ϯ 0.07 to Ϫ0.42 Ϯ 0.07, respectively) indicated that the binding of CCL3 also depended on the applied radioligand. The broad spectrum virus-encoded chemokine antagonist vCCL2 was probed against both radioligands, and displayed similar binding pattern independent upon the radioligand with K i of 8.1 nM (Log K i Ϫ8.1 Ϯ 0.15, n ϭ 3) and Hill coefficients of Ϫ0.71 Ϯ 0.06 using 125 I-CCL3 as radioligand, and K i of 7.6 nM (Log K i Ϫ8.1 Ϯ 0.12, n ϭ 3), and Hill coefficients of Ϫ0.71 Ϯ 0.14 using 125 I-CCL5 as radioligand (Fig. 1C), showing that this chemokine antagonist competes equally well with both agonists. With regard to the maximal binding capacities (B max ) of the two agonists, a 390-fold higher B max was identified for CCL3 compared with CCL5 as adapted and calculated from the homologous competition binding experiments (B max of 489 Ϯ 131 fmol/10 5 cells for CCL3 and 1.25 Ϯ 0.43 fmol/10 5 cells for CCL5).
Agonistic Properties of Chemokines and Small Molecule Agonists for CCR1-The efficacy of the two endogenous chemokines (CCL3 and CCL5) were determined in transiently transfected COS-7 cells with measurement of inositol phosphate turnover (IP turnover) in cells co-transfected with the chimeric G qi4myr (18). This read-out has previously been used with success in other chemokine receptors (6,19). As shown in Fig. 2A, CCL3 and CCL5 stimulated CCR1 with potencies similar to those previously published (3) and with equal efficacies. A series of small molecule metal ion chelator complexes consisting of copper or zinc in complex with bipyridine or phenanthroline: CuBip, CuPhe, ZnBip, and ZnPhe were all identified as efficacious agonists for CCR1. In fact, they stimulated CCR1 with micromolar potencies (EC 50 from 11 to 28 M) and with efficacies similar to the endogenous chemokines (Fig. 2B). Importantly, the activation depended upon complex formation, because no activation was observed for the metal ions or for the chelators alone (Fig. 2C).

Mutational Mapping of the Molecular Basis for the Agonism of Chemokine Agonists and Small
Molecule Agonists-A series of mutations were created in distinct regions of CCR1 to determine regions of importance for the action of chemokines and of small molecule agonists (Fig. 3). The mutated residues were chosen based on current models for chemokine:receptor interaction (2,20), and for interaction of small molecule agonists with chemokine receptors (6 -8). All mutations were probed for the agonistic properties of the orthosteric    AUGUST 22, 2008 • VOLUME 283 • NUMBER 34

JOURNAL OF BIOLOGICAL CHEMISTRY 23123
chemokines and the small molecule ligands in IP turnover experiments (Table 1). In the CCR1 N terminus, five sequential truncations (⌬4, ⌬8, ⌬14, ⌬23, and ⌬24-CCR1) were created, and as expected (2,20), both chemokines were affected in a gradually increasing manner by the large truncations (⌬14, ⌬23, and ⌬24-CCR1), with no activation at all for the two latter truncations. In contrast, the minor truncations (⌬4 and (⌬8) resulted in a minor decrease in the potencies of CCL3 (Fig. 4A and Table 1), and a minor increase for CCL5 ( Fig. 4B and Table  1). A completely different pattern was observed for the small molecule agonists, as they were not affected by any of the N-terminal truncations (Fig. 5C and Table 1). Interestingly, the ⌬4-CCR1 truncation resulted in a ϳ2-fold increase in efficacy for both chemokines as well as all four small molecule agonists without any changes in receptor surface expression (data not shown).
Glu 287 in position VII:06 is conserved among chemokine receptors and located in the bridge area between the major and minor binding pocket (21). It is essential for the majority of small molecule ligands that usually contain centrally located positively charge groups (21). In CCR1, a substitution to Ala [E287A]CCR1 impaired the activation by CCL5 and the four metal ion chelator complexes with a Ͼ25-fold decrease in potencies (Fig. 4, E and F), whereas the potency of CCL3 was unaffected by this substitution (Fig. 4D).
A steric hindrance approach of residues facing toward the major binding pocket was chosen to elucidate the impact of this region. In TM-III, Ser 110 in position III:05 was substituted to Trp (and to Ala), [S110W]CCR1 (and [S110A]CCR1). In TM-IV, Ser 172 in position IV:20 was substituted to Tyr, [S172Y]CCR1, and in TM-V, Gly 207 in position V:12 was substituted to Phe (and to Ala), [G207F]CCR1 (and [G207A]CCR1). Mutations of the residues located in the "left" part delimited by TM-IV and TM-V (Ser 172 and Gly 207 ) of the major binding pocket had no impact on chemokines or metal ion chelator complexes (Table 1). In contrast, size introduction in position III:05 in the "right" part, [S110W]CCR1, severely impaired the activation by CCL5 and by the metal ion chelator complexes with Ͼ100-fold decrease in the potencies of CCL5 and the phenanthroline-containing complexes, whereas the potencies of the bipyridine complexes were impaired Ͼ25-fold (Table 1). In contrast, the potency of CCL3 was only affected to a minor degree (Ͻ10-fold and Table 1). Introduction of Ala at this position, [S110A]CCR1, had no effect on any of the ligands, indicating that size (Ser to Trp steric hindrance), rather than polarity (Ser to Ala) matters at this position. In the minor bind-  Table 1 together with the EC 50 values for the three other metal ion chelator complexes (CuPhe, ZnBip, and ZnPhe).

TABLE 1 Activation of CCR1 wt and mutations by endogenous orthosteric chemokines and allosteric small molecule agonists
IP turnover was measured in COS-7 cells cotransfected with CCR1 constructs and the promiscuous G-protein G qi4myr . The position of each mutation/truncation is given according to the nomenclatures of Schwartz (Sch) and Ballesteros and Weinstein (B&W), see text for references. The number of experiments are shown in parentheses, and F mut indicates the difference between the potency on wt CCR1 compared to mutant CCR1. Red background indicates a F mut Ͼ50. Orange indicates a F mut from 25 to 50, and yellow indicates a F mut from 10 to 25.
ing pocket (surrounded by TM-I, -II, and -VII) a steric hindrance approach was chosen for Leu 87 in position II:17 that protrudes right into the minor binding pocket, [L87F]CCR1, and as indicated in Table 1, this substitution had no effect on chemokines, or small molecule agonists.
In summary, the mutational analysis showed that both chemokines, but not the small molecule agonists depended upon the receptor N terminus (Fig. 4, A and B). Substitutions of the residues in the "right" part of the major binding pocket (size introduction at position III:05 [S110W]CCR1, and Ala substitution of GluVII:06: [E287A]CCR1) primarily affected CCL5 and the small molecules (but not CCL3), whereas substitutions of residues in the "left" part of the major binding pocket (size introduction at positions IV:20 [S172F]CCR1 and V:12 [G207F]CCR1) and in the minor binding pocket (size introduction at position II:17 [L87F]CCR1) had no effect on any of the ligands (Table 1).
Metal Ion Chelator Complexes Act as Allosteric Enhancers of CCL3 Binding, but at the Same Time Block CCL5 Binding-The affinities of the metal ion chelator complexes were determined in competition binding experiments using both endogenous chemokines as radioligands ( 125 I-CCL3 and 125 I-CCL5). Surprisingly, all four complexes were found to increase the specific binding of 125 I-CCL3 up to 3-fold with high and rather similar affinities (K i from 18 to 52 M, Table 2, Fig. 5, A-D). The enhancement was dependent upon complex formation, because neither metal ions nor the chelators alone resulted in similar enhancement (Fig. 5).
A completely different binding profile was observed when the metal ion chelators were probed against 125 I-CCL5. Thus, in contrast to the allosteric enhancement observed for 125 I-CCL3, all four metal ion chelator complexes instead displaced 125 I-CCL5 from CCR1 in a dose-dependent manner (Fig. 5,  E-H), consistent with the identified overlapping binding sites (Table 1), with affinities almost identical to the affinities observed for the enhancement of 125 I-CCL3 binding ( Table 1). The high affinity displacement of 125 I-CCL5 was also dependent upon metal ion chelator complex formation (Fig. 5).
Affinity-determined Increase in CCL3 Binding-Homologous competition binding experiments with CCL3 were performed in the presence and absence of a fixed concentration (100 M) of two selected metal ion chelator complexes (CuBip and CuPhe) to determine the effect of the allosteric modulators on affinity and B max of CCL3. Because of the influence of the percentage of bound radioligand on K i (K i increase when binding percentage increase) (22), the number of cells were adjusted to obtain similar binding percentages in the absence and presence of the metal ion chelator complexes (Table 3). Under these conditions, a small increase in the affinity of CCL3 was observed in the presence of CuBip and CuPhe, and consistent with Fig. 5, A-D, a corresponding 2-4-fold increase in B max was identified ( Table 3).

Determination of Association and Dissociation Rates for CCL3 in the Absence and Presence of Allosteric Enhancers-
The kinetic parameters of CCL3 binding to CCR1 were determined in transiently transfected COS-7 cells under the same conditions as for the competition binding experiments (Figs. 1 and 5). The association and dissociation rates were determined over a 4-h period, and the binding experiments were stopped at multiple time points (ϳ20). As summarized in Table 4, the association rate (k on ) was increased in the presence of 100 M CuBip or CuPhe, whereas the dissociation rate (k off ) was only affected with a small decrease in the presence of CuPhe, and no changes in the presence of CuBip. Thus, the affinity changes for CCL3 observed in the presence of the allosteric enhancers (Fig. 3) were mainly determined by an increase in the association rate, which is in contrast to the often observed decrease in dissociation rates for orthosteric ligands in the presence of allosteric enhancers (11) ( Table 4).

DISCUSSION
In the present report, we describe the interaction of three different types of agonists with CCR1, two endogenous chemokines, CCL3 and CCL5, and a group of structurally related small molecule metal ion chelator complexes which act as full agonists (Fig. 6). The chemokines were, as expected, found to over-  But whereas CCL5 was also highly dependent upon residues in the main ligand binding pocket for binding, this was not the case for CCL3. The small molecule agonists clearly bound only down in the main ligand binding crevice. In agreement with their orthosteric binding mode in relation to CCL5, the small molecule agonists competed normally for binding against this endogenous ligand. Surprisingly, but in agreement with their allosteric binding mode in relation to CCL3, the small molecule agonists acted as allosteric enhancers by increasing both the B max and the affinity of this endogenous ligand. Thus, in this case the binding of the same small molecule agonist in the normal main ligand binding pocket of 7TM receptors can act either as an allosteric enhancer or as a competitive blocker of different endogenous ligands depending on their different binding modes (Fig. 6).

Complex Pattern of Orthosteric Binding Sites for Endogenous
Agonists Acting on the Same Receptor-The chemokine system is rather unique among all 7TM receptor families because of the large promiscuity, both within the CC as well as the CXC branch. CCR5 for example, binds up to 10 different CCLs and the chemokine CCL5 interacts with three different receptors (CCR1, -3, and -5) (2). Although the majority of 7TM receptors only bind one endogenous ligand (and thus only encompass one orthosteric binding site), certain other receptor families contain two or even more endogenous ligands, as observed for the Y2 and the NK1 receptors (23,24).
The chemokine:receptor interaction is largely driven by electrostatics between the electropositive chemokines and electronegative extracellular parts of the receptors and in some cases the chemokine N terminus docks into the major binding pocket (2,20,25). In CCR1 we find, that the binding sites of CCL3 and CCL5 are partly overlapping, as both depend on the receptor N terminus in consistency with current models for chemokine binding (2,20). However, despite these partly overlapping binding sites, CCL5 is unable to compete against 125 I-CCL3 (Fig. 1), a phenomenon that could be explained by the 390-fold higher B max for CCL3 compared with CCL5. In fact, this phenomenon is highly overlooked and has been identified in other chemokine receptors, for example CXCR2 where the affinity for CXCL7 (NAP-2) varied up to 2000-fold depending upon whether it was measured against 125 I-CXCL5 (ENA-78) or against 125 I-CXCL8 (IL-8) (26). It has also been observed in the virus-encoded ORF74-HHV8 and US28-hCMV chemokine receptors (27,28). Outside the chemokine system it has been observed in for example the NK1 receptor, where neurokinin A (NKA) and neurokinin B (NKB) bound with high affinity, but were unable to compete against 125 I-substance P (23). Besides being attributed to huge differences in B max values, the most tempting explanation for this phenomenon is that ligands bind to distinct receptor conformations that do not interchange readily, as observed in the NK1 receptor, where mutations along the inner CCL3 Alone 0,011 Ϯ 0,0008 0,00036 Ϯ 0,00004 0,0053 Ϯ 0,0012 (4) CCL3 ϩ CuBip 0,014 Ϯ 0,0012 0,00060 Ϯ 0,00009 0,0050 Ϯ 0,0010 (4) CCL3 ϩ CuPhe 0,020 Ϯ 0,0035 0,00103 Ϯ 0,00017 0,0037 Ϯ 0,0005 (4) face of TM-II all resulted in an equilibrium shift between agonist and antagonist binding conformations (17).
Complex Pattern of Orthosteric Binding Sites for the Same Endogenous Ligand Acting on Different Receptors-Our data suggest that parts of CCL5, most likely the N-terminal (2,25), dock into the major binding pocket, whereas CCL3 solely depend on extracellular regions of CCR1. These findings are quite different from observations in CCR5, where mutational analysis suggests that the N terminus of CCL3, but not CCL5 mediates receptor activation by interacting with the transmembrane regions (29). However, consistent with our data in CCR1, studies with a CCL3/CCL5 chimera (CCL3 for the 8 amino acids prior to the CC motif, and CCL5 for the rest of the molecule) have shown that CCL5 interacts differently with CCR5 and CCR1 with the N terminus being essential for CCR1, but not CCR5 binding (25). Furthermore, N-terminal modifications of CCL5 (Met-and AOP-CCL5) and a naturally occurring N-terminal truncation (3-68 CCL5) do not affect the affinity for CCR5, but clearly decrease the affinity for CCR1 (30,31). Finally, studies of CCL3 interaction with CCR1 by photoaffinity and N-terminally fluorescent-labeled CCL3 analogues have shown that the N terminus of CCL3 interacts with extracellularly located regions of CCR1 (32). Thus, our findings are consistent with these previously published data of CCL5 and CCL3 interaction with CCR1, and confirm, that CCL5 interacts differently with CCR1 and CCR5, a phenomenon also observed for example for NPY and PYY in their interaction with Y1, Y2, and Y5 receptors (33).
The Molecular Mechanism of Action for the Small Molecule Agonists-The mutational analysis in CCR1 suggests that the small molecule agonists bind in the major binding pocket with the conserved GluVII:06 (21) as interaction partner for the metal ions (Cu 2ϩ and Zn 2ϩ ). Importantly, this binding is overlapping with the identified binding sites of small molecule agonists for CCR5, CCR8, and CXCR3 (6 -8), that activate these receptors by constraining the extracellular parts of TM-VI and -VII toward TM-III in consistency with current toggle switch models for receptor activation (9,10). The enhanced binding of CCL3 by the small molecule agonists is most likely due to a stabilization of CCR1 in a conformation that favors CCL3 binding, whereas the decreased binding of CCL5 is due to partly overlapping binding pockets (competition in the major binding pocket area around Glu VII:06) (Fig. 6).
Allosteric Modulators as Future Drugs-Although most approved 7TM drugs act at orthosteric binding sites (34), the allosteric modulators have several advantages compared with ligands acting at orthosteric site(s). First, the allosteric modulators tune the binding affinity/efficacy of the orthosteric endogenous ligands (provided absence of intrinsic efficacy!) and therefore require the presence of this ligand, which lower the risk of receptor desensitization and toxic effects (11). Second, the allosteric sites have not been under the same evolutionary pressure as the orthosteric sites to accommodate endogenous ligands, and therefore show higher divergence across receptor subtypes, facilitating development of selective ligands (13). Thus, allosteric positive and negative modulators are clearly advantageous and will be important for future pharmacological treatments. Presently, allosteric modulators have been described within clinically important 7TM receptors, for example CCR5, ghrelin, and cannabinoid receptors (16,36,37).
Impact of the Dual Modulation of Endogenous Agonists by Allosteric Modulators-Intrinsic activity for allosteric enhancers, as observed in current study, has been observed previously, for example within the GABAb, the ghrelin, the muscarinic, and the 5-hydroxytryptamine receptors (14 -16). It appears that this phenomenon, that many small molecule agonists in fact have some allosteric properties as well, is more common than previously anticipated (38). Dually acting allosteric compounds have also been observed previously, for instance in the cannabinoid CB1 receptor where the allosteric modulators Org27569, Org27759, and Org29647 all increased the binding of the agonist CP55940, and decreased the binding of the inverse agonist rimonabant (37). However, the current study is, to the best of our knowledge, the first description of dual action for allosteric modulators of structurally related endogenous agonists. This phenomenon obviously has important implications for drug discovery, as compounds acting on receptors having more than one endogenous ligand may have different effects on these and, accordingly, should be tested against all available endogenous ligands.