Contents 1.1 G Protein Coupled Receptors (GPCRs) 4 1.1.1 GPCR Structure 4 1.1.2 GPCR signaling 7 1.1.3 GPCR interaction with arrestin 8 1.1.4 Theoretical models of GPCR activation 10 126.96.36.199 The ternary complex model 10 188.8.131.52 The Extended Ternary Complex Model 11 184.108.40.206 Multiple Conformational state model of GPCR 15 1.2 Molecular Mechanisms of GPCR Activation 18 1.2.1 The Ionic Lock Switch (The E/DRY motif) 19 1.2.2 The 3-7 Lock Switch 21 1.2.3 The CWxPY motif (Transmission Switch) 21 220.127.116.11 Cys6.47 22 18.104.22.168 Trp6.48 23 22.214.171.124 Pro6.50 25 126.96.36.199 Tyr6.51 26 1.2.4 Tyrosine Toggle Switch (The NPxxY motif) 28 1.3 Chemokines and Chemokine Receptors 29 1.3.1 Chemokines 29 1.3.2 Chemokine Receptors 30 1.3.3 Chemokine Binding Mechanism and two-step model of chemokine receptor activation 31 1.4 The CCR5 Chemokine Receptor 32 1.4.1 CCR5 Structure 32 1.4.2 Endogenous ligands of CCR5 34 1.4.3 Human Immunodeficiency Virus (HIV)-Coreceptor Function of CCR5 35 1.4.4 CCR5 Signaling Pathways 38 1.4.5 CCR5 regulation 39 1.4.6 Constitutive Activity in CCR5 40 1.4.7 Mechanism of Activation of CCR5 41 188.8.131.52 The DRY Motif 41 184.108.40.206 The FWxPY Motif 42
The CCR5 chemokine receptor (CCR5) is a G protein-coupled receptor (GPCR) that is found on the surface of leucocytes. CCR5 is a receptor for chemokines involved in trafficking leucocytes to sites of inflammation and also acts as a receptor for chemokines during the transition from innate to adaptive immunity (Olson and Ley, 2002; Fish and Wong 2003; Lagane et al., 2005; Glass et al., 2005; Oliveira et al., 2014). In the immune system chemokines act as signalling proteins that are released by immune cells during inflammation.The CCR5 receptor regulates chemotaxis and the effector functions of T lymphocytes, dendritic cells and macrophages (Opperman, 2004; Oliveira et al., 2014). The natural immune system chemokine ligands that activate CCR5 are CCL3-L1; CCL3 formerly known as macrophage inflammatory protein (MIP-1α); CCL4, formerly MIP-1β, and CCL5, formerly regulated upon activation, normal T cell expressed and secreted (RANTES). Activation of the CCR5 receptor with any of these ligands results in cellular signalling through cytosolic G proteins as well as signalling through G protein-independent pathways (Aramori et al., 1997; De Corno et al., 2001; Barmania and Pepper, 2013, Steen et al., 2014). G proteins are cytosolic heterotrimeric proteins that activate cellular signalling cascades through effector enzymes or ion channels. CCR5 also plays a central role in human immunodeficiency virus (HIV) infection by acting as a co-receptor during entry of the virus into CD4+ T cells. Studies have shown that individuals who have a 32 base pair deletion in the CCR5 gene are resistant or partially immune to HIV infection. The delta-32 mutation in the CCR5 gene results in the expression of non-functional CCR5 receptor. Individuals with the CCR5 Δ-32 mutation produce misfolded CCR5 receptor protein and as a result express low levels or no cell surface CCR5 receptor, making them resistant to infection with HIV strains that use CCR5 as a co-receptor during entry (Blanpain, 2002; de Silva and Stumpf, 2004). 1.1 G Protein Coupled Receptors (GPCRs) GPCRs are a large family of membrane proteins that have been identified in five of the six kingdoms of life including Bacteria, Protozoa, Plantae, Fungi and Animalia (Sciioth and Fredriksson, 2005). In the human genome GPCRs make up the largest family of membrane proteins and they are the targets of more than 501.1.1 GPCR Structure GPCRs are subdivided into six subfamilies: family A, rhodopsin-like receptors; family B, secretin and adhesion receptors; family C, metabotropic glutamate receptors; family D, the fungal mating pheromone receptors; family E, cyclic AMP receptors and family F, frizzled/smoothened receptors. The rhodopsin-like receptor family has the largest number of members of the known 802 GPCRs and chemokine receptors belong to this family (Lagerstrom and Schioth, 2008). Members of the rhodopsin-like receptor family have a low overall amino acid sequence homology, which is restricted to highly conserved amino acid residues in TM segments. This suggests that these conserved amino acid residues have important roles in maintaining either structural or functional integrity of these receptors (Gether, 2000). Ballesteros and Weinstein developed a numbering system that facilitates the comparison of equivalent amino acid residues of different receptors belonging to the rhodopsin-like receptor family (Ballesteros and Weinstein, 1995). In the Ballesteros and Weinstein numbering scheme the most conserved amino acid residue in a TM helix is given the number 50 and all the other amino acid residues in this TM segment are numbered according to their positions relative to this conserved residue (Ballesteros and Weinstein, 1995). Each amino acid residue is identified by three numbers. The first number indicates the transmembrane helix in which the amino acid is located, the second number shows the position of the amino acid residue relative to the reference most conserved residue. The last number is the sequence position of the amino acid residue in the polypeptide chain of the receptor. To illustrate this number system, the amino acid tryptophan at position 248 in the polypeptide chain of CCR5 is found in TM6 and is designated Trp6.48(248), because it precedes the most conserved residue in TM6, Pro6.50(250) by two residues.
Figure 1.1: The general structure of Family A (Rhodopsin-like) GPCRs. The cartoon shows the TM segments numbered 1-7. The highly conserved amino acid residues that are important for receptor integrity and function are shown as white circles. The 2 conserved cysteine residues, which form a disulphide bridge between the first and second extracellular loop, the DRY motif at the cytoplasmic side of TM3and CWxPY motif in TM6 have been implicated in receptor activation, the NPxxY motif in TM7 is important in activation and the cysteine residue in the C-terminal tail is important for receptor regulation. Image taken from Gether, 2000.
In addition to the most highly conserved residues in each TM domain, GPCRs have conserved amino acid sequence motifs as shown in Fig1.1. These include the Asp-Arg-Tyr (DRY) motif at the cytoplasmic side of TM3, the Cys-Trp-x-Pro-Tyr (CWxPY) motif in TM6 and the Asn-Pro-x-x-Tyr (NPxxY) motif in TM7 (Gether, 2000; Khorana et al., 2007). These highly conserved motifs and amino acid residues likely play crucial roles in receptor structure by stabilizing protein folding and in enabling receptor function by facilitating transmembrane motion associated with the receptor activation mechanism (Khorana et al., 2007).
1.1.2 GPCR signaling GPCRs are dynamic proteins that exist in a thermodynamic equilibrium between an inactive state (R) and an active state (R*). Agonists are ligands that stimulate physiological responses. Agonist ligands either preferentially bind to the active receptor conformation or induce the active conformation after binding the receptor. This stabilizes the receptor in the active state and drives the equilibrium towards R* (Samama et al., 1980). In the active receptor conformation, G protein binding sites on the intracellular face of the receptor are exposed. This leads to the coupling of the receptor to and activation of G proteins (Samama et al., 1993; Sprang, 2011). Antagonists are ligands that do not produce physiological responses when they bind to receptors but they instead block agonist-induced responses. Antagonists do not preferentially bind to any receptor state and therefore do not affect the equilibrium between R and R* (Brink et al., 2004). Inverse agonists, on the other hand, selectively bind to the inactive receptor conformation and this drives the equilibrium towards R (Brink et al., 2004). Heterotrimeric G proteins are composed of α, β and γ subunits and they act as molecular switches that activate GPCR-dependent cellular signalling. There are four main families of G proteins, named after the four families of the α subunits of G proteins (Hurowitz et al., 2000; Purves et al, 2001). The four families of G proteins are Gi/Go, Gq, Gs and G12. Members of the Gi protein family inhibit activity of adenylyl cyclase (AC) and activate phosphodiesterase 6 (Hurowitz et al, 2000). Examples of receptors that use the Gi protein family include chemokine receptors, rhodopsin, α2-adrenoreceptors, histamine H3 and H4, and dopamine D2-like receptors (Lodish et al., 2000; Purves et al, 2001). G proteins of the Gs family are involved in the activation AC and these G proteins are associated with β-adrenoreceptors, dopamine D1-like receptors and histamine H2 receptor. The Gq family of G proteins is involved in the activation of phospholipase C (PLC) and they activate histamine H1 receptor, muscarinic M1, M2 and M3 receptors (Qin et al., 2011; Purves, 2001). The G12 family proteins are involved in the activation of the Rho family of GTPases (Lodish et al., 2000; Purves et al, 2001). In the activated state of the G protein the α subunit is bound to GTP while in the inactive state the α subunit is bound to GDP (Sprang, 2011). In the inactive GDP-bound state, all three G protein subunits are bound together. The activated GPCR acts as a guanine nucleotide exchange factor and stimulates the exchange of GDP bound on the α subunit of the G protein for GTP. On binding GTP, the Gα subunit dissociates from the Gβγ subunit heterodimer. The free GTP-Gα and Gβγ subunits regulate the activities of enzymatic effectors such as AC and PLC to generate second messengers (Luttrell, 2006). Second messengers, in turn, control the activities of key signalling molecules involved in effector responses, such as chemotaxis in the case of the CCR5 chemokine receptor. The Gα subunit has an intrinsic ability to hydrolyse GTP to GDP and this is important in regulating the signalling cascade. Once the Gα subunit has hydrolysed GTP to GDP, it then combines with Gβγ subunits and returns the G protein to the inactive state (Hurowitz et al., 2000; Purves et al, 2001). 1.1.3 GPCR interaction with arrestin The binding of agonist ligands activates receptors, but ultimately leads to events that result in reduced receptor responsiveness (Sexton and Tilakarante, 2005). Molecular mechanisms that control receptor signalling, desensitization and resensistization ultimately regulate the activities of GPCRs (Ferguson, 2001). Desensitization is when there is a reduction in receptor responsiveness to the same amount of agonist ligand over time and this process is an important feedback mechanism that protects against both acute and chronic overstimulation (Ferguson, 2001). Desensitization can either be homologous or heterologous desensitization. Homologous desensitization is the reduced responsiveness of a particular receptor, when that receptor has been exposed to an agonist ligand over a period of time. On the other hand stimulation by an agonist that leads to unresponsiveness of other, even unoccupied, receptors is termed heterologous desensitization. The underlying mechanism of desensitization of GPCRs involves phosphorylation of intracellular domains. On binding agonists, GPCRs become substrates for phosphorylation by G protein-coupled receptor kinases (GRKs). GRKs phosphorylate serine or threonine residues in the third ICL or the carboxyl terminus of the agonist-occupied receptor. GRK phosphorylation increases the affinity of the receptor for β-arrestins, which bind to the GPCR and prevent further binding of G proteins (Sexton and Tilakarante, 2005). Arrestins are small cytoplasmic proteins that regulate GPCR activity. Homologous desensitization is exhibited by the ability of GRK to phosphorylate only agonist-occupied receptors. The binding of arrestins only to agonist occupied GRK-phosphorylated GPCRs inhibits further coupling of the GPCRs with G proteins and thus desensitizing the receptor (Kelly et al., 2008). In a fluorescence resonance energy transfer (FRET) study to detect interaction of arrestins and β2-adrenergic receptor it was observed that removal of an agonist from the receptor resulted in a rapid dissociation of the arrestins from the GRK phosphorylated receptor. This therefore highlighted that agonist occupancy of the receptor is important for binding arrestins (Krasel et al., 2005; Kelly et al, 2008). In addition to desensitization of receptors, arrestins are also important for trafficking of internalized receptors. Receptor sequestration or endocytosis, of GPCRs is the internalization of receptors and this occurs subsequent to desensitization. The binding of β-arrestin to the receptor, after GRK-mediated GPCR phosphorylation, results in agonist-induced endocytosis of many GPCRs (Ferguson, 2001). β-arrestins have two motifs that enable them to function as adapter proteins that link the phosphorylated-GPCR-arrestin complex to components of the clathrin-dependent endocytotic machinery and results in the internalization of the receptor. (Lefkowitz and Luttrell, 2002). The internalized receptor is can either be dephosphorylated and recycled back to the cell surface or it can be packaged into lysosomes for degradation. In CCR5, desensitization also involves the internalization of the receptors (Oppermann, 2004; Escola et al., 2010). After internalization, some GPCRs such as may be resensitized by a process that involves dissociation of ligands and dephosphorylation of the receptors, followed by recycling of receptors back to the cell membrane. In other receptors, such as angiotensin receptors, the β-arrestin forms a stable complex with the receptor and this targets the β-arrestin-receptor complex for degradation. Subsequent membrane expression of the receptor depends on synthesis of new receptors (Luttrell, 2006). 1.1.4 Theoretical models of GPCR activation Receptor theories are concepts that are used to explain the mechanisms of action of ligands on receptors and the signaling events that occur following this interaction. Early researchers recognized that in order for a ligand to elicit an intracellular response, the ligand must be able to bind and activate the receptor (Perez and Karnik, 2005). Various models have been proposed to explain how the interaction of ligands and receptors induces cellular responses. 220.127.116.11 The ternary complex model The Ternary Complex Model (TCM), shown in Fig 1.2A, proposes that only the receptor-G protein complex can bind an agonist ligand with high affinity. The ligand-bound receptor is then able to activate G proteins in the cytoplasm and trigger signaling cascade mechanisms (De Lean et al., 1980). The model postulates that there are two forms of the receptor: the high and low affinity states. The high affinity receptor state can interact with the agonist ligand leading to the formation of a high affinity ternary complex with the G protein. It also states that agonist ligands have an intrinsic activity that is correlated to the ability of the agonist to promote the formation and stabilization of the high affinity complex (De Lean et al., 1980). The model implies that in the absence of an agonist ligand the receptor cannot couple to the G protein. One of the disadvantages of this model is that it cannot account for constitutively active receptors, which are able to activate G protein-mediated signalling in the absence of bound ligand (Samama et al., 1993). 18.104.22.168 The Extended Ternary Complex Model Samama and co-workers generated a mutant of the β2-adrenergic receptor, which differed from the wild-type β2-adrenergic receptor in the following ways: the mutant receptor showed increased basal activity, increased affinity for agonist ligands and increased intrinsic activity of partial agonist ligands (Samama et al., 1993). These findings could not be explained by the ternary complex model, so it was modified to the extended ternary complex model (ETCM). The ETCM (Fig 1.2B) proposes that the receptor exists in a dynamic equilibrium between the inactive state, R and the active state, R*: R R*, and it states that only the R* state can bind to G protein such that LR*G is the only possible ternary complex formed (L refers to agonist) (Samama et al., 1993). In the absence of an agonist ligand the equilibrium is shifted towards R, but there still remains a fraction of receptors in the R* state. R* has a higher affinity for agonist and binding of the agonist stabilizes R*and consequently shifts the equilibrium to R*. The ETCM model gave rise to a central question on the mechanism that controls conformational change of receptors. The important question was, do different receptor conformations already exist in the absence of a ligand such that a ligand complementary receptor conformation is selected to mediate signal transduction on addition of a ligand, or do ligands induce conformational changes in the receptors (Changeux and Edelstein, 2011). Fig 1.2B below illustrates both mechanisms of the receptor conformational change, with the equilibrium constants βK showing the conformational selection pathway and βJ the induced fit pathway. The induced fit theory hypothesizes that GPCRs adopt activated conformations that mediate signalling transduction on binding agonist ligands (Changeux and Edelstein, 2011). Mutational studies in GPCRs however resulted in the generation of receptors that can activate signal transduction in the absence of agonist ligands. These mutational studies provided evidence of the existence of different receptor conformations and support for the conformational selection model of GPCR activation.
Figure 1.2: Two models for GPCR systems. A. The ternary complex model, TCM, proposes that the interaction between receptor (R), ligand (L) and G protein (G) results in the formation of a high affinity ternary complex which activates the G protein. B. The extended ternary complex model, ETCM, proposes that only the active receptor state (R*) can interact with the G protein either spontaneously (to form R*G) or through ligand binding (to form LR*G) (Samama et al., 1993).
The mutant β2-adrenergic receptor that Samama and co-workers used was generated by replacing four amino acid residues in the C-terminal of the third intracellular loop with the corresponding residues of α1B-adrenergic receptor. The mutant receptor displayed activation of the effector protein in the absence of any agonist and was therefore able to activate the G protein in the absence of bound agonist (Samama et al., 1993). It was therefore interpreted that the mutant receptor shifted the equilibrium between R and R* towards R*. The mutation in α1B-adrenergic receptor spontaneously generated receptors stabilized in the R* conformations that displayed agonist-independent activation of signal transduction (Samama et al., 1993). Constitutive activity of GPCRs is defined as the ability to produce measurable second messenger in the absence of an agonist. GPCRs that display constitutive activity are believed to have a large number of receptors in the R* state, that stimulate measurable second messenger production (Seifert and Wenzel-Seifert, 2002). When compared with cells that do not have GPCRs, cells that express constitutively active GPCRs have increased basal G protein and effector system activities, as evidenced by the production of second messengers in the absence of agonist stimulation. Recent molecular dynamics studies by Novikov et al., (2014) using orthosteric ligands of the β-2 adrenergic receptor showed that a small fraction of the receptor exists in the active conformation ensemble in the absence of agonist ligands. These studies also indicated that movements of TM6 resulted in the spontaneous activation of the receptor. Novikov et al., went on to propose that their results showed that β adrenergic receptor exists as an ensemble of receptor conformations and that ligand binding shifts the equilibrium towards a particular ensemble of states rather than causing an induced fit. A receptor conformation ensemble is a population of the receptor that exists in similar conformations that is different from other conformations of the receptor at a particular time. Some mutant GPCRs that display increased basal activity in the absence of agonist stimulation. Such GPCR mutants are termed constitutively active mutants (CAMs) (Caluser et al., 2002). It has been proposed that mutations that increase the basal activities of receptors drive the R R* equilibrium towards the R* state. CAM GPCRs have increased numbers of R* state receptors that can interact with the G protein and activate effector systems in the absence of an agonist ligand (Clauser et al., 2002). Mutagenesis studies have been used to show that the highly conserved amino acid residues in GPCRs are important for stabilizing receptor conformations. Mutating the highly conserved amino acid residues of the E/DRY motif on the cytoplasmic face of TM3 in some family A GPCRs, such as rhodopsin, results in the generation of CAM receptors. Such mutations disrupt the highly conserved interaction between arginine and the aspartate or glutamate of the motif. This releases the conformational constrains that stabilize the inactive receptor conformation (Ballesteros et al., 2002). Mutagenesis has also been used to show that the residues of the DRY motif of α1B-adrenergic receptor that play a crucial role for the activation of the receptor (Greasely et al., 2002). Also, mutant receptors of the chemokine C5A receptor generated by random saturation mutagenesis of each TM clearly identified TM3 and TM6 as hotspots for constitutive activity (Geva et al., 2000). This evidence suggests that CAMs release the conformational constraints that stabilize GPCRs in the inactive state. Kobilka and co-workers showed that CAMs of the β2 adrenoreceptor had increased agonist affinity, an elevated basal level of adenylyl cyclase and GTPase activity, and a higher maximal agonist-stimulated adenylyl cyclase and GTPase activity than the WT β2 adrenoreceptor (Kobilka et al., 1997). However, the expression of the functional CAMs was considerably lower than that of the WT β2 adrenoreceptor and the authors attributed this to the fact that CAM proteins are structurally unstable (Samama et al., 1993, Kobilka et al., 1997). Incubation of CAM receptor expressing cells with either an agonist or antagonist which stabilized intracellular CAM receptor protein resulted in a marked increase in the cell surface levels of the CAMs. This suggested that the decreased expression of the CAMs was not a result of constitutive activation and therefore constitutive down-regulation, but rather as a result of enhanced degradation of a structurally unstable protein (Kobilka et al., 1997). Constitutive desensistization of GPCRs results from spontaneous phosphorylation of receptors that are then internalized in the absence of agonist ligands. Constitutively desensistized receptors usually show inappropriate phosphorylation, localization in endosomes or clathrin coated pits, a high degree of uncoupling from G proteins and a high affinity for arrestins (Barak et al., 2001). The naturally occurring mutation Arg137His of the DRY motif of the vasopressin 2 receptor (V2R) a class A GPCR results in constitutively desensitized receptors (Barak et al., 2001). The Arg137His mutation of V2R generates receptors that can bind vasopressin but cannot active AC. Microscopic studies with immunostained human embryonic kidney (HEK) 293 cells that expressed the mutant V2R Arg137His receptor showed that a large portion of the receptors were not on the cell surface but in the cytoplasm. Biochemical assays on the internalized V2R Arg137His receptors showed that these mutant receptors were constitutively phosphorylated (Barak et al., 2001). These studies provided evidence of receptors that are constitutively desensistized. The Arg to His mutation of the DRY motif of other class A GPCRs such as the β2-adrenergic receptor, dopamine D1-like receptor and the α1B-adrenergic receptor has also been associated with constitutive desensitization (Barak et al., 2001). The constitutive activity of GPCRs led Weiss et al. (1996) to propose a revised ECTM which they called the cubic ternary complex. The model stated that receptors in the inactive state can interact with G proteins. In this model, the inactive receptor conformation that interacts with G proteins cannot elicit intracellular signalling. This therefore gave rise to the idea that a receptor can exist in multiple conformational states (Weiss et al., 1996). 22.214.171.124 Multiple Conformational state model of GPCR Structurally diverse agonist ligands can bind to the same receptor and activate different G protein-mediated signalling pathways. The multiple conformational states model proposes that receptors exist as an ensemble of different conformations and that each agonist ligand can bind to a specific ensemble of active receptor conformations (Perez and Karkin, 2005). The model states that a receptor can have more than one activated R* state. A study done by Spengler et al., with the pituitary adenylyl cyclase activating polypeptide (PACAP) receptor, showed the existence of agonist-specific receptor states. In the study, two structurally related agonists PACAP-27 and PACAP-38 stimulated adenylyl cyclase with equal potencies. However, only PACAP-38 could induce IP production through the phospholipase C (PLC) pathway (Spengler et al., 1993). The study showed that although the two PACAP agonists demonstrated equal efficacy on adenylyl cyclase only PACAP-38 stimulated PLC with a high potency. This therefore showed that the two agonists stabilized two different receptor conformations, one that can activate adenylyl cyclase and a different one that can activate PLC. Mutagenesis studies were used to further investigate the existence of multiple receptor conformations. A cysteine residue located one helical turn below the Asp (D) of the DRY-motif of α1B-adrenergic receptor was mutated to phenylalanine. This mutation generated constitutively active mutants that could couple to G proteins in the absence of an agonist. The mutant receptors selectively activated the PLC pathway but not the phospholipase A2 pathway (Perez et al., 1996). These two pathways are coupled to two different G proteins in COS-1 cells (Perrez et al., 1993). More studies with multiple G protein coupling were used to demonstrate the existence of multiple signaling receptor states in the dopamine receptor. Gazi et al., (2003) selectively expressed four different G proteins in insect cells that also expressed the human dopamine D2 Long (D2L) receptor. They observed that different agonists activated the four individual G proteins in different ways. This indicated that the selective activation of G proteins by the D2L receptor depended on different conformations of the receptor stabilized by different agonists (Gazi et al., 2003). Protean agonists are ligands that can act as either an agonist or an inverse agonist for different responses at the same GPCR (Kenakin, 1997). Protean agonists support the existence of multiple distinct receptor conformations. If a GPCR system displays no constitutive activity, that is most of the receptors are in the inactive conformation, then a partial agonist will increase signalling activity by shifting the equilibrium of receptors from the R state towards the R* state and this is observed as positive agonism. However, to observe protean agonism, an agonist should be applied to a system were a large portion of the receptors are already in a spontaneously active conformation, that is constitutively active. If the agonist decreases signalling activity by shifting the equilibrium from the highly efficacious constitutively active state to the agonist-induced active conformation of lower efficacy this is termed negative agonism (Kenakin, 2001). This negative or inverse agonism suggests that the receptor is capable of forming at least two active receptor conformations, the spontaneously formed constitutively active conformation and agonist-induced active conformation of lower efficacy (Kenakin, 2001; Kenakin, 2002; Perez and Karkin, 2005). In recombinant histamine H3 receptors expressed in Chinese hamster ovary cells, proxyfan a ligand that binds H3 receptors with high affinity, acted as a protean agonist. The natural ligand of the H3 receptor is histamine and ciproxifan is an inverse agonist. H3 receptors show no constitutive activity with regard to the mitogen activation protein kinase (MAPK) responses. However, Proxyfan acted as a partial agonist for H3 receptor activation of MAPK responses, which were 301.2 Molecular Mechanisms of GPCR Activation Computer-based molecular modeling has been used to study the structural basis of GPCR activation. A combination of theoretical (molecular dynamic simulations) and experimental methods (crystallography, nuclear magnetic resonance and specific spectroscopic methods) are currently being used to understand the changes to receptor structure that occur on binding ligands. These methods are also being used to answer the question of whether ligand-receptor interactions occur through an ensemble of receptor conformations or via induced fit mechanisms (Trzaskowski et al., 2012). GPCRs have been crystallized in inactive conformations and more recently in active conformations (Wu et al., 2010, Rasmussen et al., 2011, Cherezov et al., 2007, Warne et al., 2011, Lebon et al., 2011, Tan et al., 2013). Comparing inactive and active crystal structures can indicate the differences between the inactive conformation and the active conformation although they do not explain how the change occurs. Crystal structures of GPCRs that have been solved include: bovine rhodopsin, opsin, human H1R, human dopamine D3 receptor, human β2 adrenergic receptor bound to either agonists or inverse agonists, turkey β1 adrenergic receptor bound to agonists or antagonist, adenosine A2A receptor bound to agonists or inverse agonists, the chemokine receptor CXCR4 bound to antagonists (Wu et al., 2010, Rasmussen et al., 2011, Cherezov et al., 2007, Warne et al., 2011, Lebon et al., 2011) and CCR5 bound to maraviroc (Tan et al., 2013). Comparison of agonist-bound crystal structures with inverse agonist-bound structures indicated that agonist ligands induce or stabilize changes in the TM domains that are associated with receptor activation. On binding agonist ligands conserved residues in TM domains form new interhelical interactions that stabilize the active receptor conformation (Mayevu et al., 2014; Deupi et al., 2012; Deupi and Standfuss, 2011; Trzaskowski et al., 2012; Ventkatakrishnan et al., 2013). The molecular mechanism of activation of GPCRs has been shown to occur through a series of conformational changes stabilized by changes in discrete intramolecular interactions called molecular switches. A molecular switch is a structural element that changes conformation between the inactive and active conformation. Molecular switches are composed of structural motifs, which are amino acid residues that show a high degree of conservation among the different rhodopsin-like GPCRs and act as functional microdomains in the receptor activation mechanism (Trzaskowski et al., 2012). Based on crystal structures and nuclear magnetic resonance (NMR) structural studies, molecular switches have been characterized in rhodopsin and a few other GPCRs (Ahuja and Smith, 2009). The integration of experimental observations and biophysical data into mechanistic schemes for receptor structure and function has been done through molecular modeling (Ballesteros et al., 1998). 1.2.1 The Ionic Lock Switch (The E/DRY motif) Intramolecular interactions in the TM domains of GPCRs serve to constrain and stabilize the receptors in the inactive conformation. Random mutations or site-specific mutations can disrupt these intramolecular interactions and release the constraints that stabilize the inactive conformations to generate constitutively active mutant (CAM) receptors (Lattion et al., 1999; Parnot et al., 2002) The motif that has been most studied in GPCRs is the “DRY” motif consisting of highly conserved Asp/Glu 3.49(D/E)-Arg3.50(R)-Tyr3.51(Y) at the cytosolic end of TM3. In the crystal structure of rhodopsin, the side chain of the Arg3.50 residue, makes an ionic interaction with an adjacent Asp/Glu3.49 residue and with the side chain of Asp/Glu6.30 residue, at the cytosolic end of TM6. This ionic interaction between these residues is enabled by the negative charge of the side chains of residues Asp/Glu3.49 and Asp/Glu6.30 and the positive charge of the Arg3.50 side chain. Metarhodopsin II is an activated intermediate between the inactivated rhodopsin conformation and activated opsin conformation. In the crystal structure of metarhodopsin II, Asp3.49 is protonated, and this protonation disrupts the ionic interaction of Asp3.49, Arg3.50 and Glu6.30 (Arnis et al., 1994; Scheer et al., 1996). In the crystal structure of the active conformation opsin, the side chain of Arg3.50 interacts with the highly conserved Tyr5.58 in TM5, as well as directly with the Gα protein (Rosenkilde et al., 2012). Comparison of the intramolecular interactions that the DRY motif makes in the inactive crystal with the active crystals show the ionic lock stabilizes the inactive conformation and the Arg3.50-Tyr5.58 interaction stabilizes the active conformation. The change in orientation of the Arg3.50 constitutes an Arg-switch that leads to activation of GPCRs (Ballesteros et al., 1998). Mutagenesis studies showed the importance of the DRY motif in stabilizing active and inactive conformations before the crystal structures were determined. The wild type histamine H2 receptor is constitutively active. Mutation of the Asp3.49(115) residue to uncharged Ala (D3.49(115)A) or Asn (D3.49(115)N) in this receptor increased the constitutive activity (Alewijnse et al., 2000). The CAMs exhibited characteristic high agonist binding affinity and increased signaling properties and high structural instability. This showed the residue Asp3.49 forms intramolecular interactions that stabilize the inactive conformation of the receptor. The study also looked at the effects of mutating the positively charged Arg3.50(116) to a neutral Ala, R3.50(116)A, and to a negatively charged Asp, R3.50(116)D. These mutations led to a decrease in signal transduction and a significant decrease in the levels of mutant receptor expressed on the cell surface when compared to the wild type receptor (Alewijnse et al., 2000). This observation demonstrated that Arg3.50 is important for proper protein folding during receptor biogenesis and cell surface expression of the receptor. Decreased cell surface receptors due to Arg3.50 mutation to Ala or Asp consequently decreased signalling as evidenced by decreased basal second messenger production of the mutant H2 receptors. The ionic lock has been observed in only a few crystal structures including the dopamine D3 receptor (Chien et al., 2010) and adenosine A2A receptor (Dore et al., 2011). In the CXCR4 chemokine receptor, the first chemokine receptor to be crystallized, there is no Glu6.30(268) and no ionic lock exists between TM3 and TM6 (Wu et al., 2010). However, in the crystal structures of GPCRs in which the ionic lock has not been shown, there are residues in these crystal structures that can form strong interactions between TM3 and TM6 (Trzaskowski et al., 2012). The CCR5 chemokine receptor does not have an acidic residue at position 6.30. However the CCR5 protein that was crystallized had a stabilizing mutation of Ala6.33 to Asp introduced that allowed formation of a salt bridge between Arg3.50 and Asp6.33. This “ionic-lock” stabilized the receptor in an inactive conformation (Tan et al., 2013). 1.2.2 The 3-7 Lock Switch The 3-7 lock is a link between Glu3.28 and protonated Schiff base of Lys7.43 in rhodopsin. During receptor activation, resulting from retinal isomerization in response to light, this 3-7 link is broken. In other receptors such as the opioid receptors and aminergic receptors, a similar activation mechanism is thought to exist, but the switch is composed of Tyr7.43 and Asp3.32 (Chien et al., 2010). Opening of the 3-7 switch was suggested to be the first step in the activation of rhodopsin and possibly one of the first switches to be activated in biogenic amine receptors (Kim et al., 2004). 1.2.3 The CWxPY motif (Transmission Switch) In TM6 of family A GPCRs there is a highly conserved Pro6.50 residue that is surrounded by a group of similarly conserved amino acid residues, which constitute the CWxPY motif (Cysteine-Tryptophan-x-Proline-Tyrosine). The relative percentage conservation of the residues of the CWxPY motif in class A GPCRs is Cys6.47-7126.96.36.199 Cys6.47 Crystal structures have shown that Cys6.47 has a conserved non-covalent interaction with a residue at position 7.45 in TM7 that stabilizes both inactive and active conformational states (Venkatakrishnan et al., 2013). The residue in position 7.45 is mostly conserved as Asn, 67Comparison of the inactive and active crystal structures of β2 adrenergic receptor and adenosine A2a receptor show that the side chain of Cys6.47 does not change its rotameric state (Olivella et al., 2013). The crystal structures of the β2 adrenergic receptor in the inactive conformation and in the active conformation show that Cys6.47 form interactions that modulates the hydrogen bond formation between Asp2.50 and Asn7.49. In the inactive crystal structure of β2 adrenergic receptor, Asn7.49 hydrogen bonds with Arg7.45 through a water molecule. In the active crystal structure of β2 adrenergic receptor, Cys6.47 releases Asn7.45 from the Asn7.45 hydrogen bond; this allows it to form an interaction with Asp2.50 that stabilizes the active conformation (Olivella et al., 2013). These different intramolecular interactions of Asn7.45 between the inactive crystal structure and the active crystal structure of β2 adrenergic receptor suggest that Cys6.47 modulates the interaction of Asp2.50 and Asn7.49 which stabilizes the active conformation. 188.8.131.52 Trp6.48 In the rotamer toggle switch proposed by Shi et al (2002), ligand induced changes in the orientation of the side chain of Trp6.48(286) resulted in the modulation of the pro-kink in TM6 (Shi et al., 2002). Observations made in biophysical and computational studies done in β2 adrenegic receptor showed that changes in rotameric conformations of Trp6.48(286) were associated with receptor activation. Schwartz and coworkers went on to propose that, on receptor activation, Trp6.48(287) rotated into an active rotamer conformation in which the indole side chain made an aromatic stacking interaction with Phe6.44 and Phe6.52(290) that stabilized the receptor in an active conformation (Schwartz et al., 2006). Mutational studies on Trp6.48(248) in various class A GPCRs have shown variable consequences for receptor expression, constitutive activity and uncoupling, supporting a role of the residue in structure and activation. Mutagenesis studies done with the constitutively active ghrelin receptor, GPR119, showed that the Trp6.48(248)Ala mutation resulted in 37As observed with Cys6.47, a comparison of both inactive and active state crystal structures of β2 adrenergic receptor and adenosine A2a receptor show that the side chain of Trp6.48 does not change its rotameric state as previously hypothesized (Olivella et al., 2012). However the crystal structures show that in the transition from an inactive receptor state to an activated receptor state the Trp6.48 side chain moves together with the protein backbone and this unwinds TM6 at the Pro kink causing an opening at the intracellular end of this helix (Xu et al., 2011; Katritch and Abaygan, 2011). Agonist-induced rotation of TM6 is initiated in different ways in different receptors. In active crystal structure of adenosine A2a receptor, agonist ligands make contact with Trp6.48, causing the residue to make a 2Å movement as TM6 rotates during activation of the receptor. In the active crystal structure of β2 adrenergic receptor, Trp6.48 does not make contact with agonist, but rather movement of this residue occurs indirectly due to agonist induced changes in interactions of the “transmission switch” which consists of Ile121(3.40) Leu/Phe/Val/Met5.51 and Phe282(.6.44) (Xu et al., 2011; Katritch and Abaygan, 2011). The activation process of rhodopsin and adenosine receptors is associated with a global conformational change of TM6 which is caused by changes in the orientation of the indole side chain of Trp6.48. In the inactive conformation, the side chain of Trp6.48 is shown to be stabilized by a water molecule in the proximity of TM7. On activation the indole ring of Trp6.48 swings away from TM7 towards TM5 were it forms an aromatic stacking with Phe5.48 (Schwartz et al., 2009; Standfuss et al., 2011). In the crystal structures of ground state rhodopsin the indole ring of Trp6.48 is tightly packed against retinal whereas in the active crystal structures this indole ring is 3.6Å away from this position (Standfuss et al., 2011). These interactions of the side chain of Trp6.48 stabilize either the active receptor conformation or the inactive receptor conformation. In the crystal structure of CCR5 bound to maraviroc, the phenyl group of maraviroc makes hydrophobic interactions with Trp6.48 and these interactions are important for stabilizing the CCR5 receptor in an inactive conformation. The interaction of the maraviroc phenyl and Trp6.48 locks the residue and prevents the movement of TM6 that is associated with receptor activation (Tan et al., 2013). 184.108.40.206 Pro6.50 In TM6 of class A GPCRs the highly conserved Pro6.50 of the CWxPY motif creates a kink in the α-helix (Schwartz et al., 2009; Standfuss et al., 2011). The rotamer toggle switch mechanism of activation proposed that agonist ligand binding to a receptor induced changes in the rotameric state Trp6.48(286) resulted in the straightening of the pro-kink in TM6 (Shi et al., 2002). The straightening of the pro-kink was hypothesized to be associated with helical movements that occur during receptor activation (Shi et al., 2002). Mutating Pro6.50(254) to Ala in the prostacyclin receptor resulted in mutant receptors that had severely impaired agonist binding and receptor activation functions (Stitham et al., 2002). These observations provided evidence of the importance of Pro6.50 in receptor activation. Further support of the important role of Pro6.50 in GPCR activation came from site-directed mutagenesis studies of human calcitonin receptor. The Pro6.50(326)Ala mutation resulted in a calcitonin mutant receptor that had reduced signaling function (Bailey and Hay, 2007). Comparison of inactive and active state crystal structures of rhodopsin show that the angle of the characteristic bend of TM6 does not change but rather changes occur in the structure of retinal leading to the rotation of TM6 (Choe et al., 2011; Standfuss et al., 2011). Similar outward movements of the cytosolic end of TM6 from the core of helix bundle to expose G protein binding sites are observed in the comparison of the inactive and the nanobody stabilized active structures of β2AR (Rasmussen et al., 2011). This shows that the angle of the Pro-kink in TM6 does not change, but rather it is the rotation of the whole TM6 that is associated with activation and because of the Pro-kink this movement is opens the cytosolic end of TM6. Residues that surround Pro6.50 are Cys6.47, Tyr6.51 and Pro7.38 and these residues interact with the a water molecule in the vicinity of Pro6.50 through hydrogen bonds in both active and inactive state crystal structures of rhodopsin, β2 adrenergic and adenosine receptors. It is these interactions of the residues of the CWxPY motif that are thought to stabilize the Pro kink of TM6 in both inactive and active receptor conformations (Standfuss et al., 2011). 220.127.116.11 Tyr6.51 Class A GPCRs contain a residue with an aromatic side chain/…. At position 6.51 (Phe Comparison of crystal structures show that residues of TM3, TM6 and TM7 form a conserved ligand binding pocket in class A GPCRs (Venkatakrishnan et al., 2013). In TM6 residues at position 6.48 and 6.51 make contact with many different ligands in different class A GPCRs (Venkatakrishnan et al., 2013). During activation of rhodopsin there is evidence of an increased level of contact between the ligand retinal and Tyr6.51 which is associated with the rotation of TM6 (Deupi et al., 2012). Retinal is the natural ligand of rhodopsin that captures light and changes it into chemical signals during vision (Zhou et al., 2012). On absorption of light, retinal undergoes isomerization from a 11-cis state to an all trans state and this is followed by conformational changes in the structure of rhodopsin. Before activation, 11 cis-retinal forms a covalent bond with Lys7.43(296) and hydrophobic interactions with with residues Met5.42(207), Phe5.43(208), Phe5.47(212) of TM5; Trp6.48(265) and Tyr6.51(268) of TM6 in the ligand binding pocket (Zhou et al., 2012). Tyr6.51(268) forms new interactions with residues in TM5 during the transition from rhodopsin to opsin (Venkatakrishnan et al., 2013). In the serotonin 5-HT2a receptor position 6.51 is conserved as Phe and mutational studies in this receptor where Phe6.51 was mutated to Leu resulted in a drastic decrease in the binding affinity of receptor for agonist ligands derived from phenethylamines mescaline (Braden et al., 2006). Further evidence of the role of a residue in position 6.51 in ligand binding comes from modeling studies of human histamine H4 receptor (H4R) studies. In the modeling studies of the histamine H4 receptor Lim et al., showed that the imidazole group of histamine forms an aromatic interaction with Tyr6.51 as shown in Fig 1.3 below (Lim et al., 2008; Lim et al., 2010). These studies proved that residues at position 6.51 occupy the ligand binding pocket and form interactions with ligands in the ligand binding pocket. This shows that residues of the CWxPY motif also play a crucial role in the ligand binding pockets of many GPCRs, where their interactions with ligands leads to stabilization of either active or inactive receptor conformations
Figure 1.3: Shows the interactions histamine (magenta atoms) make in the human H4R model binding pocket. The imidazole group of histamine (blue) makes an aromatic stacking interaction with Tyr6.51. Picture taken from Lim et al., 2010. In class A GPCRs the CWxPY motif plays prominent roles in receptor cell surface expression, ligand binding and receptor activation. However in CCR5, the roles of the FWxPY motif in proper protein folding and expression, agonist binding and receptor activation has not yet fully been explored. 1.2.4 Tyrosine Toggle Switch (The NPxxY motif) The NPxxY motif in TM7 of class A GPCRs has been shown in inactive crystal to form a hydrogen bond network that stabilizes the inactive structures. In the ground state structure of rhodopsin a water mediated hydrogen bond network that involves Trp6.48(265) of TM6, Asp2.50(83) in TM2, Ser7.45(298) and Asn7.49(302) of the NPxxY motif in TM7 stabilizes the inactive conformation of the receptor (Standfuss et al., 2011). Isomerization of retinal breaks the water mediated link of Trp6.48(265) in TM6 and Ser7.45(298) in TM7 and this consequently leads to the rearrangements of the residues of the NPxxY motif (Standfuss et al., 2011; Trzaskowski et al., 2012). The side chain of Tyr7.53 of the NPxxY motif in TM 7 has been shown to rotate between three different states in X-ray crystal structures of activated GPCRs. In the inactive rhodopsin, the aromatic side chain of Tyr7.53 interacts with the conserved Phe7.60 in helix 8 by forming an aromatic stacking. However, in the inactive forms of A2A adenosine receptor and in squid rhodopsin, Tyr7.53 has a different rotamer conformation. In these structures the side chain of Tyr7.53 points upwards, towards the centre of the receptor and makes hydrogen bond interactions with a water molecule (Schwartz et al., 2009). In the activated receptor opsin complexed to a Gα protein the side chain of Tyr7.53(306) makes water mediated hydrogen network with Arg3.50(135) in TM3 and Asn8.37(310) in TM8. In the inactive conformation the Tyr7.53(306) rotamer switch interacts with helix 8 and in the active conformation the rotamer switch is stabilized by hydrophobic interactions with residues of TM6 (Schwartz et al., 2009; Standfuss et al., 2011; Trzaskowski et al., 2012). 1.3 Chemokines and Chemokine Receptors 1.3.1 Chemokines Chemokines are chemotactic cytokines that function as chemoattractants of leucocytes during inflammation and are essential for the host response to injury and infection. Chemokines induce cell migration and activation, by binding to specific GPCRs on target cells (Luster, 1998). Chemokines are small proteins (8-10 kilodaltons) that share 20-501.3.2 Chemokine Receptors Chemokine receptors are members of the rhodopsin-like GPCR family. Chemokine receptors are typically 340-370 amino acids in length and are classified according to the type of chemokine they bind (Rosenkilde et al., 2012). Chemokine receptors CXCR1 to CXCR7 bind CXC chemokines, whereas receptors CCR1 to CCR11 bind CC chemokines (Comerford and McColl, 2011). There are also the non-functional (silenced) chemokine receptors, D6 and DUFFY, CCX CKR receptors, and a few virally expressed receptors (Charo and Ransohoff, 2006). The N-terminus and ECLs of chemokine receptors are responsible for ligand binding (Blanpain et al., 1999; Ai and Liao, 2002; Limatola et al., 2005). The C-terminus and the ICLs are responsible for transmitting the signals through G protein binding (Cheng, 2000; Sabroe, 2005). Chemokine receptors have the highly conserved DRY motif which is involved in receptor activation. The so-called “silenced” receptors (D6 and DUFFY) lack the DRY motif and this is thought to be one of the reasons for their non-functionality (Borroni et al., 2006; Hansell et al., 2011). Signalling pathways of most chemokine receptors are blocked by treatment of cells with pertussis toxin (PTx). PTx catalyses the addition of ADP-ribose molecules to the α subunit of Gi/o proteins, locking the G proteins in an inactive state and effectively inhibiting the interaction of G proteins with GPCRs. This therefore shows that chemokine receptors activate the Gi family of heterotrimeric proteins, which inhibit production of cAMP. Gi/o proteins activate a broad spectrum of signalling pathways, ranging from the activation of PLC, the phosphate 3’phosphoinositol kinase (PI3K) to the mitogen activated protein kinase (MAPK cascade). Chemokine receptors regulate the activity of phospholipase C through the activation of the βγ subunit of the Gi protein. The activated G protein causes a phospholipase C mediated breakdown of phosphotidylinositol-4,5-bisphosphate to produce second messengers inositol-1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). IP3 mobilizes intracellular calcium, while DAG activates protein kinase C (PKC) (van Acker et al., 1996). 1.3.3 Chemokine Binding Mechanism and two-step model of chemokine receptor activation Chemokines activate their receptors through a putative two step mechanism. The first step involves the binding of the chemokine to the receptor N-terminus and extracellular loops (ECLs) to a designated site I. The binding of a chemokine ligand to the chemokine receptor recognition site I involves ionic interactions between the positively charged residues of the ligand and the negatively charged residues at the N-terminus and ECLs of the receptor (Fernandez and Lolis, 2002; Colvin et al., 2004; Allen et al., 2007). The second interaction is the binding of the N-terminal region of chemokines preceding the first disulphide motif with the residues in the ECLs and TM domains that form of site II the receptor; and this interaction is necessary for receptor activation (Fernandez and Lolis, 2002; Rajagopalan and Rajarathnam, 2006). Chemokine ligands have a flexible N-terminal part known as the triggering domain which interacts with the ECLs and/or TM domains resulting in receptor activation (Monteclaro and Charo, 1996; Wu et al., 1996; Scholten et al., 2012). Evidence for this, was obtained when Clark-Lewis et al., truncated the N-termini of chemokines. This led to a loss of agonist activity of the ligand, while the ability to bind the receptor with high affinity was retained (Clark-Lewis et al., 2003). The agonist, CXCL8, of the chemokine receptor CXCR1 binds to the receptor using this classical two step model. The N-loop of CXCL8 interacts with the N-terminus of CXCR1 (site I). This then leads to the interaction of CXCR1 with Glu4 of CXCL8 (site II) that triggers receptor activation (Sarmiento et al., 2011). 1.4 The CCR5 Chemokine Receptor 1.4.1 CCR5 Structure The CCR5 chemokine receptor contains 352 amino acids (Fig1.5) with a calculated molecular mass of 40.6 kDa (Oppermann, 2004; Raport et al., 1996). CCR5 is most closely related to CCR2, with 71The CCR5 chemokine receptor has the highly conserved amino acid motifs that are found in family A GPCRs including the DRY motif at the cytoplasmic side of TM3. In CCR5 Asp3.49 forms an ionic interaction with an adjacent Arg3.50. However, like all other chemokine receptors CCR5 has a positively charged residue at position 6.30 which does not interaction with Arg3.50. The Asp3.49-Arg3.50 interaction is important in stabilizing CCR5 receptor structure and for cell surface expression of the receptor. Mutating Asp3.49 to Ala resulted in reduced cell surface levels of CCR5 (Folefoc et al., 2010) and this may be due increased degradation of unstable receptor protein or increased internalization of the cell surface receptors (Folefoc et al., 2010). The Arg3.50Asn mutation in CCR5 also resulted in decreased cell surface levels of the receptor and this was due to increased phosphorylation and internalization of the mutant receptors (Lagane et al, 2005). (the NPxxY motif in TM7. However, position 6.47 which is conserved as Cys in most GPCRs is substituted with Phe in CCR5 and many other chemokine receptors (Ref……). Thus, the CWxPY motif in TM6 is conserved as the FWxPY motif in most chemokine receptors.
Figure 1.5: A two-dimensional structure of the CCR5 sequence. The CCR5 structure shown has the characteristics of class A GPCRs. It consists of the extracellular N-terminus, 7TMs and the intracellular C-terminus. The highly conserved DRY motif and NPxxY motif are highlighted in green. In this study we looked at the highly conserved FWxPY motif in TM6 which is highlighted in blue.
Sequence alignment studies show approximately 70CCR5 also contains the highly conserved NPxxY motif in TM7 that has been shown to be important for ligand-induced receptor activation in other class A GPCRs (Fig 1.5). The Asn residue of this motif plays important role during receptor activation by interacting with Phe6.47 and Trp6.48 in the formation of aromatic stacks that stabilize the active receptor conformation (Jongejan et al., 2005; Trzaskowski et al., 2012). The C-terminal tails of chemokine receptors like all family A receptors have phosphorylation sites (Oppermann, 2004). The phosphorylation sites serve important regulatory purposes for receptor desensitization and cell surface expression. The CWxPY motif located in TM6 at the bottom of the ligand-binding pocket of most family A GPCRs has been shown to be important for maintaining receptor structure and in receptor function (Ahuja and Smith, 2009). In the crystal structure of CCR5 bound to maraviroc, the FWxPY motif also forms part of the ligand-binding pocket of maraviroc. Trp6.48(248) and Tyr6.51(251) are part of the residues that interact with maraviroc in the binding pocket to stabilize the inactive conformation of CCR5 (Tan et al., 2013). 1.4.2 Endogenous ligands of CCR5 The chemokine receptor CCR5 is important for directing chemotaxis of activated T cells and the natural chemokine ligands of CCR5 are CCL3, CCL4 and CCL5. CCL3, CCL4 and CCL5 are inflammatory β-chemokines that have two adjacent cysteine residues near the N-terminus. These chemokines are produced in high concentrations during inflammation and induce the migration of monocytes, natural killer cells and dendritic cells to sites of injury or infection (Murphy et al., 2000; Bachelerie et al., 2014; Charo et al., 2013). CCR5 agonist ligands are also involved in enhancing the activation of T cell responses and the production of antigen specific T cells (Taub et al., 1996).
1.4.3 Human Immunodeficiency Virus (HIV)-Coreceptor Function of CCR5 HIV-1 entry into target cells requires sequential interaction of the surface subunit (gp120) of the viral envelope glycoprotein with cell surface CD4 and a GPCR acting as a coreceptor, namely CCR5 or CXCR4 (Alkhatib et al., 1996; Deng et al., 1996; Dragic et al., 1996; Doranz et al., 1996). HIV-1 viral strains designated X4 use the CXCR4 coreceptor for infection while the R5 viral strains use the CCR5 coreceptor. Dual tropic strains (R5/X4 viruses) use either CCR5 or CXCR4 coreceptors to infect cells. During the early stages of infection viruses that use the CCR5 coreceptor dominate and persist through infection. Variants that use CXCR4 or dual tropic strains emerge late in infection and the switch from R5 tropic to X4 tropic strains is sign of worsening prognosis (Lagane., 2011). The importance of CCR5 in HIV-1 infection is seen in genetic polymorphisms in the human population that accounts for the different susceptibilities to infection and variable progression to AIDS after HIV infection (Gonzalez et al., 2005; McDermott et al., 1998). The most significant of these polymorphisms is the 32 base pair deletion of CCR5 called CCR5∆32. In the homozygous state, this mutation renders individuals highly resistant to HIV infection due to the functional loss of this viral coreceptor (Dean et al., 1996; Liu et al., 1996). Individuals heterozygous for the CCR5∆32 mutation have reduced susceptibity to infection and if they get infected they progress slowly to AIDS (Blanpain et al., 2002; Dean et al., 1996; Hoffman et al., 1997). There are also other genetic polymorphisms of CCR5 that are associated with reduced HIV acquisition and disease progression. Segmental duplication of CCR5 ligand MIP-1α, results in increased expression of the ligand and reduced CCR5 cell surface expression (Gonzalez et al., 2005). HIV entry into the target cell involves an intricate and sequential sequence of events with the aim of delivering the viral RNA into the target cell cytoplasm. Coreceptor binding triggers a conformational change in the viral Env that mediates membrane fusion and delivery of the viral payload into the target cell. Inside the cell, the viral capsid is uncoated, revealing the viral genome as shown in Fig 1.6. The viral RNA is then reverse transcribed into a double-strand DNA that is transported to the nucleus of the host cell where it integrates into a host genome (Wilen et al., 2012). The integrated provirus is then transcribed and translated by host machinery to generate a polyprotein that is processed to form virions that bud off from the host cell and enable additional rounds of replication (Wilen et al., 2012) (Fig 1.6).
Figure1.6:The reproductive cycle of HIV. (a) HIV virus binds to a CD4 and chemokine co-receptor (CCR5/CXCR4) on cell surface. (b) Fusion of virus and target cell membranes. (c) Reverse transcription of viral RNA. (d) Intergration of viral genome into host cell genome. (e) Transcription of viral RNA (f) Processing of expressed viral proteins. (g) Packaging of new viruses. (h) Budding off of viruses from host cell membrane. HIV entry begins with the virion binding to CD4 molecules (Fig 1.10). The CD4-binding site is located at the interface of the outer and inner domains of gp120 (Kwong et al., 1998). CD4 binding to this highly conserved site initiates a sequence of structural rearrangements in Env that reveal and create the co-receptor binding site. The coreceptor binding sites includes, the V3 loop and the bridging sheet (Suphaphiphat et al., 2003).
Figure 1.7: Sequence of events in HIV entry. The entry process involves three key steps. First, viral gp120 binds to CD4. Second, binding of the coreceptor. Third, gp41 subunits allow the fusion of viral and host cell membranes. Taken from Wilen et al., 2012.
Exposure of the V3 loop allows the tip or crown of V3 to interact with the ECL2 of the coreceptor. A second key gp120-coreceptor interaction involves sulfated tyrosines present in the CCR5 N-terminal domain with the base of the V3 loop and the four-stranded bridging sheet (Wilen et al., 2012). Substantial evidence supports that the four CCR5 N-terminus sulfonated tyrosines at positions 3, 10, 14 and 15 are indispensible for efficient viral entry (Farzan et al., 1998; Rabut et al., 1998). Inhibition of tyrosine sulfation decreases the ability of CCR5 to engage HIV (Farzan et al., 1998). Gp120 neutralizing antibodies have been shown have sulfated tyrosines at the site of antigen contact. These antibodies require sulfation for activity, are more effective in the presence of CD4 and compete with N-terminal sulfated CCR5 peptides for gp120 binding (Choe et al., 2003). Coreceptor binding causes further conformational changes that result in the exposure of the hydrophobic N-terminus of gp41 fusion peptide. The gp41 ectodomain contains two helical regions 36 amino acid residues in length. The N-terminal helical regions (HR-N) from each gp41 form a triple-stranded coil. During fusion the three C-terminal helical regions (HR-C) fold back and pack into grooves formed the external interface of the three HR-N domains in an antiparallel fashion, this results in a highly stable six-helix bundle (6HB) (Wilen et al., 2012). The prebundle complex initiates early pore formation and the 6HB stabilizes and facilitates the expansion of the fusion pore (Markosyan et al., 2003). HIV-1 Env, like chemokine ligands, interacts with CCR5 with different consequences. Colin et al., (2013) showed that different CCR5 agonists stabilized distinct receptor conformations which activated different signaling pathways (Colin et at., 2013). They went on to show that [125I]-CCL3 and [35S]-gp120 interact with different CCR5 receptor conformations (Colin et al., 2013). Although HIV-1 Env interacts with the N-terminus and ECL2 like chemokines and activates CCR5 signaling it is not clear whether the active conformation of CCR5 is needed for viral entry. The recent emergence of resistant HIV-1 strains that use CCR5 receptors occupied by small molecules antagonists to gain entry into target cells has shown that HIV-1 can infect cells through different conformational states of CCR5 (Berro et al., 2011). 1.4.4 CCR5 Signaling Pathways Chemokines and their receptors play a pivotal role in the initiation and amplification of the immune response. Stimulation of CCR5 expressing leucocytes by agonist chemokine ligands results in the activation of chemotaxis and secretion of cytokines (Lagane et al., 2005; Flanagan, 2014). Different chemokine ligands activate different CCR5 signaling pathways and also CCR5 activates different signaling pathways in different cell types (Flanagan, 2014; Shi et al., 2007; Mueller et al., 2002). This suggests that there are different CCR5 conformations that activate different signaling pathways. In CCR5-expressing leucocytes, stimulation of the chemotaxis by CCL5 is completely inhibited by treatment of the cells with PTx (Thelen, 2001). PTX catalyses the addition of ADP-ribose moieties to the α subunit of Gi/o proteins (Mangmool and Kurose, 2011). This ADP-ribosylation process of the α subunit of the Gi/o protein family inhibits the G proteins from interacting with their cognitive GPCRs (Mangmool and Kurose, 2011). Chemokine receptors, including CCR5, stimulate intracellular Ca2+ flux and chemotaxis and this process is inhibited by PTx (Mellado et al., 2001). Stimulation of CCR5 with chemokine agonist ligands does not result in the production of IP (Aramori et al., 1997). Chemokine treatment of HEK cells that only express CCR5 receptors and no G proteins does not result in IP production. HEK Gqi cells are cells that stably express the chimeric G protein, which consists of the α subunit of a Gq complex whose 5 carboxyl terminal amino acids have been replaced with those of Gi. (Folefoc et al., 2010). This chimeric G protein enables receptors such as CCR5 that normally activate the Gi/o family of G proteins to activate PLC and stimulate IP production (Folefoc et al., 2010). 1.4.5 CCR5 regulation Chemokine receptor activation of G proteins is regulated by phosphorylation, desensitization and internalization processes of the receptor. Constitutive or agonist-induced activation of CCR5 result in GRK-dependent phosphorylation and β-arrestin binding to the receptor. The binding of β-arrestin induces rapid internalization of the receptor; therefore CAMs can stimulate constitutive desensitization and internalization (Flanagan, 2014). Homologous and heterologous desensitization of chemokine receptors is achieved by GRKs. GRKs are divided into three subfamilies: GRK1 and 7; GRK2 and 3; GRK4 to 6 (Lefkowitz et al., 1998). GRK2 and 3 as well as PKC are important for the phosphorylation of CCR5 (Oppermann et al., 1999). The GRK- and PKC-mediated phosphorylation sites are four serine residues at positions 336, 337, 342 and 349 on the C-terminal of CCR5 (Fig. 1.7) (Oppermann et al., 1999). Phosphorylation of the serine residues on the C-terminal tail of the CCR5 results in high affinity binding of β-arrestins-1 and -2 to ligand occupied receptors (Oppermann, 2004, Flanagan, 2014). The DRY motif was also identified as a binding site of β-arrestins (Oppermann, 2004, Flanagan, 2014). Binding of β-arrestins to the phosphorylated sites and the DRY motif sterically hinder interaction of the receptor with G proteins leading to desensitization and subsequently internalization of the CCR5 receptor. β arrestins initiate endocytosis of CCR5 through clathrin-coated pits by acting as adaptor proteins that bind phosphorylated receptors and both the clathrin heavy chain and the β2-adaptin subunit of the heterotetrameric AP-2 adaptor complex (Oppermann, 2004; Flanagan, 2014). Agonist activated CCR5 was also shown to undergo β-arrestin/clathrin-independent internalization which involves the association of palmitoylated cysteine residue on the C-terminal tail with cholesterol-enriched raft microdomains (Oppermann, 2004; Flanagan, 2014). After the receptor has been internalized, CCR5 accumulate in perinuclear recycling endosomes and returns to the plasma membrane in a dephosphorylated form (Oppermann, 2004; Flanagan, 2014). The regulation cell surface chemokine receptor activities is an important mechanism of regulating inflammatory processes and immune responses. 1.4.6 Constitutive Activity in CCR5 Chemokine receptors CCR5 and CXCR2 have been reported to exhibit constitutive active activity and pertussis toxin was shown to block the coupling of agonist-free receptors to Gi proteins (De Voux et al., 2012, Folefoc et al., 2010, Chen et al., 2000). When compared to untransfected HEK 293 cells, HEK293 cells that express wild type CCR5 receptor show spontaneous coupling of the receptor to G proteins (Lagane et al., 2005; Folefoc et al., 2010). Wild type CCR5 does not have the ionic lock formed by Asp3.49 of the DRY motif and Glu6.30 in TM6. Asp3.49 is negatively charged and therefore it can interact with the positively charged Glu6.30 in TM6. The ionic lock interaction stabilizes the inactive receptor conformation (Deupi and Standfuss, 2011). CCR5 has a basic Arg residue at position 6.30 therefore it cannot form an ionic lock with Asp3.49. The lack of an ionic lock in CCR5, may be the reason of the constitutive activity of the receptor. The crystal structure of CCR5 bound to maraviroc has a modified “ionic lock” that was made by mutating Ala at position 6.33 to Asp. This allowed the receptor to form a salt bridge between Arg3.50 and Asp6.33. This modification stabilized the inactive conformation of CCR5 (Tan et al., 2013). 1.4.7 Mechanism of Activation of CCR5 18.104.22.168 The DRY Motif The role of the DRY motif in the activation mechanism of CCR5 has been investigated. As noted earlier the wild type CCR5 chemokine receptor is a constitutively active receptor. Chemokine receptors, including CCR5, contain a positively charged residue at position 6.30 and hence the ionic lock between Arg3.50 and a negatively charged residue at 6.30 does not exist. The CCR5 receptor has the conserved Asp3.49(125) and Arg3.50(126) residues, but instead of the negatively charged Asp6.30 it has a positively charged Arg residue in this position, which is more likely to sterically repel Arg3.50(126) (Springael et al., 2007). Neutralization of the charged Arg residue at position 6.30 by mutating it to Ala decreases the constitutive activity of CCR5 by half. The introduction of putative corrective conditions for ionic locking by mutating arginine to hydrogen bonding amino acid glutamine (Arg6.30Gln) or an acidic residue (Arg6.30Asp or Glu) results in a decrease of constitutive activity to undetectable levels (Springael et al., 2007). These results suggest that the presence of Arg6.30(223) instead of Asp/Glu3.60 in the classical DRY lock, may account for the basal levels of constitutive activity of the wild type receptor (Springael et al., 2007). In the crystal structure of CCR5 bound to maraviroc the inactive conformation of helix V1 was stabilized by the thermostabilizing mutation Ala6.33(233)Asp. This mutation created a salt bridge between Arg3.50 and Ala6.33(233)Asp that locked the receptor in an inactive conformation (Tan et al., 2013). Disregarding the lack of an ionic lock in chemokine receptors, the DRY-motif is still important in receptor activation as shown in CCR5 were mutation of Arg3.50 to neutral Asn disrupted both basal activity and chemokine induced Gαi protein coupling despite a retained affinity for MIP-1β (Lagane et al. 2005). The Arg3.50(156)Asn mutation resulted in a mutant receptor that was constitutively phosphorylated and had a higher internalization efficiency than the wild type CCR5 receptor (Lagane et al., 2005). Mutation of Asp3.49(125) to Val, had no effect on either the constitutive or agonist-induced activity of CCR5 (Lagane et al., 2005). However mutating Asp3.49(125) to Ala, resulted in decreased cell surface expression of CCR5, suggesting that this residue may be important for receptor biogenesis (De Voux et al., 2009). 22.214.171.124 The FWxPY Motif Ligand-induced activation of CCR5 and other rhodopsin-like receptors involve conformational changes. The highly conserved amino acid motifs in TMs are important for the transition from R to R*. Most rhodopsin-like GPCRs have a highly conserved CWxPY (Cys-Trp-any amino acid-Pro-Try) motif in TM6. However in CCR5 and other CC chemokine receptors the motif is modified to FWAPY (Phe-Trp-Ala-Pro-Try). As mentioned earlier, in class A GPCRs Cys6.47 is important for receptor biogenesis, structure and function. In CCR5, the role of Phe6.47 in receptor structure and function has not been investigated. Trp6.48(248) is another highly conserved amino acid that is tightly packed against retinal in ground-state rhodopsin and has been identified as important for GPCR activation through mutagenesis studies. Comparisons of inactive and active crystal structures of rhodopsin have shown that Trp6.48 undergoes conformational changes that accompany receptor activation when TM6 rotates away from the core of the helix bundle opening the G protein binding site (Pellisier et al., 2009). In CCR5, Garcia-Perez and co-workers mutated Trp6.48(248) to Ala and observed that this alteration drastically affected CCR5 receptor expression and [125I]CCL3 and [35S]gp120 binding (Garcia-Perez et al., 2011). ). In the crystal structure of CCR5 bound to maraviroc, residues Trp6.48(248) and Tyr6.51(251) are part of the ligand binding pocket. These two residues make aromatic interactions with the phenyl group of maraviroc that stabilize the inactive conformation of the receptor (Tan et al., 2013). In the molecular model of CCR5 bound to HIV-1 gp120 V3 loop, Tyr6.51(251) is shown to be consistently hydrogen bound to the carbonyl group of Arg18 of the V3 loop (Tamamis and Floudas, 2014). Mutational studies in CCR5 showed that when Trp6.48(248) was substituted with Ala in CCR5, the mutant receptor generated was poorly expressed on the cell surface and had highly impaired binding affinity to both [125I]CCL3 and [35S]gp120 (Garcia-Perez et al., 2011). Mutations of Tyr6.51(251) to Ala, Phe or Ile resulted in mutant receptors that had reduced binding affinity for [125I]CCL3, [35S]gp120 and maraviroc (Garcia-Perez et al., 2011). This evidence suggests that Trp6.48(248) and Tyr6.51(251) is are important residues for ligand binding. However the role of the FWxPY motif in CCR5 structure and function has not been fully elucidated. In this study, I used site-directed mutagenesis to investigate the role of these amino acid residues in the activation mechanism of CCR5. I used codon-optimized CCR5 to enhance expression of the mutant receptor constructs in mammalian cells. The amino acids Phe6.47(247), Trp6.48(248), Pro6.50(250) and Tyr6.51(251) were mutated and the expression of the mutant receptors assessed by ELISA. The experimental results show that the residues of the FWxPY motif are important for the cell surface expression of CCR5. Pro6.50(250) is important for both basal activity of CCR5 and the transition from an inactive conformation to an active conformation. Hydrogen bonding properties of Phe6.47(247) are important in the transition of CCR5 from an inactive conformation to a ligand-induced active conformation. Aromatic and hydrophobic interactions at position 6.51 are important in stabilizing ligand-induced active conformations of CCR5.
Owen Karimanzira · Created · Last updated over 1 year ago ·