Constitutive activation of CCR5 and CCR2 induced by conformational changes in the conserved TXP motif in transmembrane helix 2.

CCR5 is a G protein-coupled receptor for RANTES, MIP-1alpha, MIP-1beta, and MCP-2 that functions as the front line coreceptor for human immunodeficiency virus type 1 infection. To elucidate the mechanism for CCR5 activation, this coreceptor was expressed in yeast coupled to the pheromone response pathway and a constitutively active mutant (CAM) was derived by random mutagenesis. Conversion of Thr-82 in the highly conserved TXP motif in transmembrane helix 2 to Pro, His, Tyr, Arg, or Lys conferred autonomous signaling activity in yeast and mammalian cells. This substitution also imparted constitutive signaling to CCR2 in yeast and mammalian cells, but not CCR1, CCR3, CCR4, CXCR2, or CXCR4. The CCR5-CAM, but not the CCR2-CAM had a reduction in ligand binding affinity. Whereas the amplitude of calcium mobilization induced by RANTES stimulation was lower in the CCR5-CAM than the wild-type (WT) receptor, MCP-1 induced a higher signal in the CCR2-CAM than in CCR2-WT. The chemotactic response of CCR5-CAM(T82P) to RANTES was similar to that of CCR5-WT, but CCR5-CAM(T82K) was dramatically decreased. The chemotactic response of CCR2-WT and CCR2-CAM(T94K) were similar. These findings extend insight into the role of the TXP motif in the mechanism for CCR5 signaling. CCR2, the receptor most closely genetically related to CCR5, shared a similar signaling mechanism, but other receptors containing the TXP motif did not. The expression of CCR5 and CCR2 in yeast and the availability of variants with autonomous signaling represent critical tools for characterizing receptor antagonists and developing approaches to block their role in human diseases.

GPCRs have seven hydrophobic helices that function as transmembrane spanning domains. In addition to this hydrophobic core, the receptor contains three intracellular loops, three extracellular loops, an N-terminal extracellular domain, and a C-terminal cytoplasmic tail. Amino acid residues in the N-terminal extracellular domain and the second extracellular loop have been shown to be critical for CCR5 utilization as a coreceptor for M-tropic gp120 subunits (14 -16, 20, 21). The utilization of coreceptor domains also has been shown to be influenced by the conformation of transmembrane-spanning domains and residues predicted to occur at the helical-cytosolic interface (19,20). In addition to membrane fusion and viral entry, the interaction of CD4-activated gp120 with CCR5 results in receptor activation, which may increase viral replication (22)(23)(24). Some envelope glycoproteins demonstrate weak CCR5 signaling and have low level replication, which can be augmented by activating CCR5 with physiologic ligands (24).
The mechanism for GPCR signaling has been shown to involve re-orientation of hydrophobic helices and consequent alteration in the conformation of the cytoplasmic domains (25,26). The structural basis for the dynamic equilibrium between inactive and active conformations is best understood for rhodopsin. Although some insight into the mechanism for ligand binding has been developed, understanding of the mechanism for CCR5 activation is limited. Mutation of the Pro residue in a highly conserved sequence TXP in transmembrane domain (TM2) 2 of CCR5 has been shown to render the receptor refractory to conformational activation by ligand binding (27). In order to gain insight we have functionally expressed human CCR5 in Saccharomyces cerevisiae coupled to the pheromone response pathway by a hybrid G␣ subunit. CCR5 variants with autonomous signaling activity were selected using this system. A randomly generated constitutively active mutant (CAM) contained a conversion of Thr-82 to Pro in the TXP sequence in TM2. This CCR5 mutant also demonstrated constitutive signaling in mammalian cells. Analogous mutations conferred autonomous signaling to CCR2, but not CCR1, CCR3, or CCR4. The conformational shift in CCR5 induced by activating mutations in Thr-82 resulted in decreased binding affinity to its ligands, but the analogous substitution in CCR2 did not alter the binding of MCP-1. Exposure of the CCR5 and CCR2 CAMs to their respective ligands activated cytosolic mobilization of calcium ions. These findings provide insight into the mechanism for CCR5 and CCR2 activation and furnish valuable reagents for developing a new generation of antagonists.

EXPERIMENTAL PROCEDURES
Strains, Cell Lines, and Plasmids-For expression in yeast open reading frame (ORF) of human CCR5 or other chemokine receptors, was cloned into the yeast expression vector Cp4258 and the Saccharomyces cerevisiae strain CY12946 was transformed as described previously (28). For mammalian expression, CHO cells were transfected with pcDNA3.1-CCR5 or CCR2 constructs and single cell clones were derived by immunofluorescent sorting (MoFlo, Cytomation, Fort Collins, CO) with monoclonal antibodies (mAbs) 227R (ICOS, Bothell, WA) or antihuman CCR2 (R&D Systems, Minneapolis, MN), respectively.
Recombinant RANTES Production-The ORF encoding the mature protein human RANTES was amplified by PCR and subcloned into the pPICZ␣A shuttle vector (Invitrogen, Carlsbad, CA) using unique restriction sites, EcoRI and XbaI. Pichia Pastoris were transformed with the linearized vector by electroporation. Colonies with high level expression were selected on 1000 g/ml Zeocin YPDS plates. The protein was targeted for secretion into the culture medium and purified to homogeneity by cation exchange HPLC using an UNO™ S1 column (Bio-Rad), followed by reversed phase HPLC using a MICROSORB-MV™ C4 column (VARIAN, Walnut Creek, CA).
Construction and Screening of CCR5 Mutant Library-The CCR5 ORF was subjected to random mutagenesis by PCR in the presence of manganese resulting in a mutation rate of 0.1-0.3%. This mutant library was transformed into the histidine auxotrophic CY12946 yeast strain and functionally expressed upstream of the pheromone response pathway as previously described (28). Autonomous CCR5 activation of FUS1-HIS3 reporter gene allows selection of these colonies on histidine deficient medium. Plasmid DNA from positive colonies was extracted, sequenced, and retransformed into yeast for further characterization. Site-directed mutagenesis was performed using a QuikChange kit (Invitrogen).
Yeast Receptor Activation Assay-Ligand-induced or autonomous receptor activation in yeast strains carrying a FUS1-HIS3 reporter gene was evaluated by growth on histidine-deficient medium. Receptor activation in yeast strains carrying a FUS1-lacZ gene was performed using a fluorescein di-␤-D-galactopyranoside (Molecular Probes, Eugene, OR) as described previously (28). The experimental data were normalized using ␤-galactosidase activity of nonstimulated wild-type (WT) CCR5.
Ligand Binding and Displacement-CHO cells stably expressing CCR5 or CCR2 variants were incubated with 0.1 nM 125 I-MIP-1␣ or 125 I-MCP-1 (Amersham Biosciences, Piscataway, NJ) in the presence of incremental concentrations of RANTES (Leinco Tech., St. Louis, MO), MIP-1␣, MIP-1␤, and MCP-2 (PeproTech, Rocky Hill, NJ) for CCR5, or MCP-1 (PeproTech) for CCR2 as described previously (21). All binding experiments were conducted on ice to prevent ligand-induced receptor internalization. The affinity was calculated using Prism (GraphPad Software, San Diego, CA) and is expressed as the EC 50 Ϯ S.D. averaged from the results obtained in three independent experiments and based on duplicate samples of each concentration.
[ 35 S]GTP␥S Binding Assay-The [ 35 S]GTP␥S binding assay was carried out essentially as described (28,29). Briefly, 7 g of cell membranes were preincubated with 40 M GDP (Sigma) in the absence (basal binding) or presence of an agonist for 30 min at 30°C before addition of [ 35 S]GTP␥S (Amersham Biosciences) to a final concentration of 0.25 nM. Nonspecific binding was determined in the presence of 10 M GTP␥S (Sigma). After incubating for another 30 min, reactions were terminated by rapid filtration through GF/C filters (Whatman Inc., Clifton, NJ). Bound radioactivity was measured by scintillation counting. Cells were treated with 100 ng/ml pertussis toxin (Calbiochem, La Jolla, CA) for 14 -16 h to uncouple the G i/o ␣ from the receptors. The percentage of specific binding was calculated as 100 ϫ (sample cpm Ϫ nonspecific cpm) Ϭ (basal cpm Ϫ nonspecific cpm). The values are the means Ϯ S.E. of triplicate samples, and the results are representative of four independent experiments.
Calcium Mobilization-CHO cells were loaded with 2 g/ml fura-2 acetoxymethyl ester (Molecular Probes). Agonist-dependent increases in cytoplasmic calcium were determined as described (21). The results are representative of two independent experiments.
Chemotaxis Assay-Cells were resuspended in MEM␣ medium containing 0.5% bovine serum albumin. The cell density was adjusted to 2 ϫ 10 6 cells/ml and 100 l was added to the top chamber of 24-well transwell apparatus (6.5-mm diameter, 8.0-m pore size; Corning Inc., Corning, NY). Agonist was added to the lower chamber. The plates were incubated for 4 h at 37°C. Cells that passed through membranes into the lower chamber were collected and counted by flow cytometry. The chemotactic index was determined as a ratio of cells in lower chamber in the presence versus in the absence of agonist. The results are representative of three independent experiments.

RESULTS
Derivation of CCR5 CAMs in Yeast-CCR5 was expressed in yeast coupled to the pheromone response pathway by a hybrid G␣ subunit, as previously described for CXCR4 (28). Histidine auxotrophic yeast strains expressing human CCR5 and a pheromone-responsive FUS1-HIS3 reporter gene grew in the absence of histidine when exposed to RANTES (Fig. 1A), demonstrating that the receptor was functional and linked to the pheromone response pathway. High concentrations of RANTES were required to activate CCR5 signaling in yeast, but chemokines that do not bind CCR5 had no effect at any concentration (data not shown). Control cells (Fig. 1A) and yeast strains expressing CXCR4 (data not shown) were not stimulated to grow in the absence of histidine when exposed to RANTES, establishing the specificity of the system. This was confirmed by the ability of RANTES to activate human CCR5 in parallel experiments in yeast strains containing a FUS1-lacZ reporter gene, resulting in expression of ␤-galactosidase activity ( Fig. 2A, see inset).
In order to derive CCR5-CAMs, the CCR5 ORF was randomly mutated by PCR in the presence of manganese at a mutational rate of 0.1-0.3% and these cDNA pools were screened for the ability to confer histidine-independent growth in the histidine auxotrophic yeast strain containing the FUS1-HIS3 reporter gene. Screening of over 10 5 recombinant events for growth in the absence of histidine and CCR5 ligands yielded three clones with an A to C substitution resulting in the T82P mutation. No other activating mutation was identified among 18 autonomously growing clones that were sequenced. Yeast cells expressing CCR5(T82P) grew in histidine free medium in the absence of RANTES and this was not significantly altered by the addition of this ligand, as shown in Fig. 1B. The proliferation of cells expressing the CCR5-CAM without ligand stimulation was greater than that of those expressing CCR5-WT grown in the presence of 1 M RANTES. Cells expressing CCR5-WT did not grow in the absence of RANTES. The specificity of WT receptor and CCR5-CAM coupling to pheromone response pathway was confirmed by inhibition of yeast proliferation with 3-amino-1, 2,4-triazole, a competitive inhibitor of histidine production (data not shown).
Amino acid substitutions that confer CCR5 autonomous signaling were determined by saturation mutagenesis of Thr-82. As shown in Fig. 1C, substitution of Thr-82 with Lys or Arg resulted in high levels of basal ␤-galactosidase expression in yeast cells containing the FUS1-lacZ reporter gene. Conversion of Thr-82 to His, Tyr, or Pro, but not Gly, Ala, Val, Leu, Met, Trp, Asp, or Glu, resulted in CAM activity, albeit at lower levels. Exposure of CAMs to RANTES did not result in additional activation of receptor signaling. Activation of ␤-galactosidase expression in yeast cells expressing CCR5(T82K) or CCR5(T82P) was not significantly altered by exposure to RAN-TES at concentrations up to 10 M ( Fig. 2A).
CCR5 CAMs Selected in Yeast Are Active in Mammalian Cells-The authenticity of the activated conformation of the CCR5-CAMs selected in yeast was determined by analysis of signaling in mammalian cells. CHO transfectants expressing CCR5-WT or the T82P, T82K, or T82G mutants were prepared by lipofection and cell sorting. Staining of the transfectants with 227R, a mAb to the N-terminal extracellular domain of CCR5, revealed that all were expressed on the surface of CHO cells at similar levels, as shown in Fig. 2B. The autonomous signaling activity of the CCR5-CAMs was determined in [ 35 S]GTP␥S binding experiments (Fig. 2C). This response was inhibited by pertussis toxin (data not shown). Membrane fractions from CHO transfectants expressing CCR5(T82P) or CCR5(T82K) have basal levels of [ 35 S]GTP␥S binding that were ϳ2-3-fold higher than that of transfectants expressing the WT receptor or CCR5(T82G). Exposure to incremental concentrations of RANTES demonstrated that CCR5(T82P) was more sensitive to ligand activation than the WT receptor and the other CCR5 mutants in some experiments (Fig. 2C).
The Role of TXP in Autonomous Signaling is Common to CCR5 and CCR2-Since the TXP sequence in TM2 is highly conserved in receptors for C-C and C-X-C chemokines, the analogous Thr residue in other C-C receptors was substituted with Pro or Lys to determine whether it conferred autonomous signaling. As shown in Fig. 3A, introduction of Thr 3 Pro or Thr 3 Lys conversions in CCR1, CCR3, and CCR4 had no effect on the expression of ␤-galactosidase activity in yeast cells containing a FUS1-lacZ reporter gene. In contrast, yeast cells containing CCR2(T94K), but not CCR2(T94P) demonstrated elevated basal levels of ␤-galactosidase activity.
CCR2(T94K) was expressed in CHO transfectants to establish that the conformation induced by this mutation also conferred autonomous coupling to G proteins in mammalian cells. Transfectants expressing CCR2-WT or the T94K mutation were prepared by lipofection and cell sorting. was only about 5-fold (Fig. 4C). When MCP-2 was used to displace 125 I-MIP-1␣, its binding to CCR5(T82G) was similar to CCR5-WT (6.3 nM versus 4.0 nM). In contrast CCR5(T82P) showed about 3-fold decrease of affinity for this ligand. The T82K mutation resulted in an almost complete loss of binding of MCP-2 (about 2 M) (Fig. 4D). Despite some quantitative differences, for every ligand the T82K mutation, which has the highest constitutive activity in yeast, had the most significant impact on the binding, the order of magnitude of the effect for each ligand being MCP-2 Ͼ RANTES Ͼ MIP-1␣ Ͼ MIP-1␤.
The binding of MCP-1 to the CCR2-CAM was determined in radioligand binding experiments. There was no difference in the ability of cold MCP-1 to displace the binding of 125 I-MCP-1 to CCR2(T94K) or the WT receptor (Fig. 4E).
Impact of Constitutive Activity on Ligand-induced Signaling by CCR5 and CCR2 CAMs-The ability of the CCR5 and CCR2 CAMs to mediate signal transduction following ligand binding was determined in calcium flux experiments using CHO transfectants with matched levels of cell surface expression, as shown in Fig. 5A. Mobilization of cytosolic calcium was detected in transfectants expressing CCR5-WT following exposure to 0.3 nM RANTES, with peak responses evident at 10 nM RANTES. The sensitivity of transfectants expressing the CCR5-CAMs, CCR5(T82P) and CCR5(T82K) was slightly decreased, with calcium mobilization detectable when incubated with 1.0 nM RAN-TES. The calcium flux detectable in CCR5(T82G) transfectants following exposure to 0.3 nM RANTES was slightly less than observed in those expressing the WT receptor. Whereas the maximum amplitude of calcium mobilization mediated by CCR5(T82P) and CCR5(T82G) was similar to that of the WT receptor, peak signaling levels in CCR5(T82K) transfectants was less than observed with the WT control.
The mobilization of cytosolic calcium ions in CCR2 and the CCR2-CAM is also shown in Fig. 5A. The magnitude of the calcium response to 0.3 nM MCP-1 was slightly less in CCR2(T94K) than in the WT receptor. In contrast, the peak response observed following exposure to 10 nM and 100 nM MCP-1 was higher in transfectants expressing CCR2(T94K) than in those expressing the WT receptor.
The induction of chemotaxis in transfectants expressing CCR5 or CCR2 CAMs was compared with the response of cells expressing the respective WT receptors. As shown in Fig. 5B, the migration of CCR5 and CCR5(T82P) transfectants induced by RANTES was similar. The response of transfectants expressing CCR5(T82G) was less than that of those expressing the WT receptor and CCR5(T82K) transfectants lacked a significant chemotactic response to RANTES. There was no significant difference between the ability of CCR2 and CCR2(T94K) transfectants to undergo directed migration in response to MCP-1. DISCUSSION CCR5 has been the target of intense mutational analysis to elucidate the mechanisms for interactions with its physiologic and pathologic ligands, RANTES, MIP-1␣, MIP-1␤, MCP-2, and M-tropic envelope glycoproteins, respectively. Residues that are necessary for ligand binding and coreceptor activities have been identified in extracellular domains and helical bundles (14 -16, 20, 21, 30). To gain insight into the mechanism for receptor activation, CCR5 was expressed in Saccharomyces cerevisiae coupled to pheromone responsive HIS3 and lacZ reporter genes via a hybrid G␣ protein subunit. CCR5 signaling induced by RANTES resulted in complementation of histidine auxotrophy of host yeast strains by expression of the FUS1-HIS3 reporter gene, or of ␤-galactosidase activity by FUS1-lacZ. Access to the power of yeast genetics enabled the deriva- tion of mutants with an autonomous signaling phenotype conferred by HIS3 correction of histidine synthesis defects in yeast grown in the absence of ligand in histidine-free medium. The mutation that conferred autonomous signaling in yeast and mammalian cells involved Thr-82 in the highly conserved TXP motif in TM2. This mutation was also associated with constitutive signaling activity in CCR2, the receptor most closely related to CCR5 at the genetic level, but not other chemokine receptors. This suggests that this family of receptors may transmit signaling induced by chemokine binding through different mechanisms.
The TXP motif is conserved in TM2 near the first extracellular loop of all chemokine receptors, except mouse CCR7, mouse and human CXCR5, and human XCR1 (27). Its role in CCR5 signaling was explored based on the high level of conservation and the prediction that the Pro residues disrupted the helical nature of TM2. Conversion of Pro-84 to Ala dramatically decreased ligand binding and signaling and substitution of Thr-82 with Ser, Val, Ala, or Cys had minimal effects on ligand binding, but impaired signaling, most notably with Val or Ala. These loss of function studies implicate a role for the TXP motif in maintaining receptor conformation involved in formation of the ligand binding site in the case of Pro-84, but suggest that Thr-82 may play a role in regulating the equilibrium between inactive and active states. This could consist of stabilization of the orientation of TM helices resulting from replacement of a polar residue with hydrophobic amino acids. A naturally occurring mutation in TM2, Ala-73 to Val has been shown to be associated with an increase in CCR5 binding affinity to RANTES and increased sensitivity to the induction of chemotaxis by this ligand (31,32). TAK-779 is a small molecule antagonist of CCR5. Alanine scanning mutagenesis revealed that several residues from TM2 participate in the formation of a binding pocket for this inhibitor, including Thr-82 (33). It was also established that interaction of TM2 with TM3 and possibly with TM7 is critical for CCR5 function (30,34). The importance of TM2 in constitutive activation was shown for AT 1A receptor which autonomous signaling resulted from mutation of Asp-111 to Ser and was dependent on TM2 translational and in part rotational movement (35,36). Mutations in TM2 that cause constitutive activation of human rhodopsin have been reported to be responsible for retinal diseases (reviewed in Ref. 37). Some of these mutations occur in the GGF motif, which is similarly positioned in TM2 and is thought to have a structural function similar to the TXP motif of chemokine receptors (27).
The finding that the random conversion of Thr-82 to Pro confers autonomous signaling implicates this region of TM2 in reorientation of TM helices involved in signal transduction in CCR5. The presence of a Pro residue in this site is associated with a disruption of the helical nature of TM2 due to conformational restraints imposed by this residue, as is evident from the high resolution structure of the prototypic GPCR, bovine rhodopin (38). It is likely that the presence of a second Pro residue in this motif induces additional structural restrictions, resulting in shortening of TM2 and/or alteration of its tilt in the lipid bilayer of the plasma membrane and alteration of the exposed cytoplasmic loops. These structural alterations probably result in release of structural constraints that keep the receptor in an inactive conformation and allow it to spontaneously reach an activated state in the absence of ligand. However, binding of RANTES to this mutant has the potential to stabilize its activated conformation as suggested by the relatively well-preserved capacity to mediate agonist-induced transient signaling in calcium flux and chemotaxis experiment. Compared with the Thr 3 Pro mutation, the Thr 3 Lys substitution translates into a strongly enhanced constitutive activity in yeast and a dramatically impaired responsiveness to the natural agonists in mammalian cells paralleled with similar decrease in affinity for all the natural agonists of CCR5. The alteration of the polarity at this site by introduction of a positively charged residue resultes in high constitutive activity that can hardly be further enhanced by RANTES or MIP-1␣. Thus, it is likely that the conformational shift associated with the presence of a positively charged residue is similar to that associated with ligand activation.
Insight into the structural basis for the mechanism by which ligand binding to extracellular domains of GPCRs induces conformation of cytoplasmic domains that are permissive for engagement of heterotrimeric G proteins and GTP hydrolysis is limited. Biophysical and structural analysis of rhodopsin activation reveals a shift that opens the cytoplasmic aspect of the hydrophobic core and exposes novel surfaces for interaction with G proteins (reviewed in Ref. 39). The re-orientation of helical bundles was shown to primarily involve TM3 and TM6, which are located centrally in the hydrophobic core. It is predicted that the helical re-orientation opens a cleft in the hydrophobic core that includes exposure of the inner aspect of TM2, which is predicted to have a more peripheral location in rhodopsin. Mutation of a Pro residue in TM6 in bovine rhodopsin has been reported in retinitis pigmentosa, indicating that alteration of a helical kink alters helical mobility required for shift from the inactive to active conformation (40). Change in TM2 orientation resulting from Thr to Pro or Lys substitutions are likely to induce conformational alterations that modify its interaction with TM3, thereby promoting shift to an active state.
Constitutively active mutants of GPCRs have been associated with human diseases and induced in vitro by mutagenesis (reviewed in Ref. 41). While they have been reported to naturally occur in TM2 in rhodopsin (reviewed in Ref. 37), more directed mutations introduced in TM2 of other GPCRs have resulted in impaired functions of the receptors, constitutive activity or hypersensitity to cognate ligands (42)(43)(44). However, this hydrophobic helix is not a frequent site for CAMs. High frequency mutagenesis of the complement factor 5a receptor (C5aR), which is a member of the chemoattractant receptor family, associated CAMs with amino acid conversions in TM6 (16 out of 22 CAMs), TM3 (5/22), and TM1 (1/22), but not TM2, TM4, TM5, or TM7 (45,46). Since this receptor does not contain a TXP motif in TM2 it is likely that the members of the chemokine receptors have different mechanisms for stabilizing the inactive conformation and regulating the dynamic equilibrium with the active state. While substitution of Asn-119 with Ser or Ala resulted in autonomous activation of CXCR4, this mutation did not confer constitutive activity to CXCR2 (28). Similarly, while conversion of Thr-82 to Pro or Lys was associated with autonomous signaling in CCR5 and CCR2, these substitutions did not alter the signaling phenotype of other C-C chemokine receptors containing the TXP motif in yeast. Even in the case of the two genetically closely related receptors CCR5 and CCR2 (71% identity at the protein level and more than 85% in the TM domains), the same Thr to Lys mutation resulted in qualitatively similar but quantitatively different levels of constitutive activity. The apparent level of constitutive activity in yeast was much higher for the CCR5(T82K) mutant and correlated with a much more profound impairment of the ligandinduced signaling in CHO cells as assessed by both calcium flux and chemotaxis experiments. It is also likely that GPCRs may assume multiple stable conformations in the dynamic equilibrium between inactive and active states. As reported for CXCR4, substitution of critical residues with different amino acids was associated with different levels of G protein coupling in mammalian cells (28). Moreover, some CAMs had ligand binding similar to the wild-type receptor, whereas others had decreased ligand affinities, indicative of shifts in the conformation of extracellular domains involved in binding sites (47). The CXCR4-CAMs were found to be phosphorylated and chronically desensitized while AT 1A -CAM had a basal level of phosphorylation similar to WT receptor even upon ligand stimulation (28,48). The variation in biologic responses of the CAMs, especially the T82K, compared with the WT receptor could be explained by the combined effects of a reduced affinity for CCR5 ligands and desensitization of an already active conformation of the receptor. It has been shown that CAMs of GPCRs induced by different mutations may have multiple activated conformations, which may be constitutively phosphorylated and internalized or undergo endocytosis only following ligand stimulation (49).
Our findings provide new insights into understanding of signaling mechanism of CCR5 and activation of GPCRs in general. Based on the similarities between CCR5 and CCR2 activation we propose that TM2 plays a crucial role in the signal transduction mechanism of these chemokine receptors.
CCR5 is a front line coreceptor of HIV-1. CCR2 is involved in atherosclerosis, asthma and other inflammatory diseases. Therefore both of these receptors are targets for inhibition with small molecule antagonists. CAMs of CCR5 and CCR2 can be a useful tool to dissect the mechanism of available inhibitors and provide approaches of design and screening for new generation of antagonists.