Mapping the membrane topology and extracellular ligand binding domains of the retinol binding protein receptor

STRA6 is a multitransmembrane domain protein not homologous to any other proteins with known function. It functions as the high-affinity receptor for plasma retinol binding protein (RBP) and mediates cellular uptake of vitamin A from the vitamin A-RBP complex. Consistent with the diverse roles of vitamin A and the wide tissue expression pattern of STRA6, mutations in STRA6 are associated with severe pathological phenotypes in humans. The structural basis for STRA6's biochemical function is unknown. Although computer programs predict 11 transmembrane domains for STRA6, its topology has never been studied experimentally. Elucidating the transmembrane topology of STRA6 is critical for understanding its structure and function. By inserting an epitope tag into all possible extracellular and intracellular domains of STRA6, we systematically analyzed the accessibility of each tag on the surface of live cells, the accessibility of each tag in permeabilized cells, and the effect of each tag on RBP binding and STRA6-mediated vitamin A uptake from the vitamin A-RBP complex. In addition, we used a new lysine accessibility technique combining cell-surface biotinylation and tandem-affinity purification to study a region of the protein not revealed by the epitope tagging method. These studies not only revealed STRA6's extracellular, transmembrane, and intracellular domains but also implicated extracellular regions of STRA6 in RBP binding.     

Assembly and ligand binding properties of the water-soluble extracellular domains of the glutamate receptor 1 subunit

High resolution structural studies of models of glutamate receptors (GluRs) have been limited to monomeric models of the ligand-binding site. To obtain oligomeric models of glutamate receptors that can reveal more complete structural information, we examined the assembly and ligand binding properties of two truncated versions of the GluR1 subunit. The first version, GluR1-WS, consisted of only the N-terminal extracellular segment (Ala(1)-Glu(520)) bridged by a synthetic linker to the second extracellular domain (Asn(615)-Gly(790)). The second version, GluR1-M1, consisted of the first N-terminal extracellular domain (Ala(1)-Glu(520)) bridged by a synthetic linker to a second segment containing the second extracellular domain, the third transmembrane domain, and the intracellular C-terminal domain (Asn(615)-Leu(889)). When expressed in Xenopus oocytes, GluR-WS was secreted and water-soluble; GluR1-M1 was displayed on the surface of oocytes. GluR1-WS exhibited a velocity sedimentation profile that was consistent with assembly of homooligomers and bound the glutamate receptor agonist alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid with high affinity. These findings show that the extracellular domains of GluR1 that are sufficient for ligand binding apparently are sufficient for subunit assembly and might be a suitable target for structural studies of a water-soluble GluR1 oligomer.     

Molecular characterization of CCR6: Involvement of multiple domains in ligand binding and receptor signaling

The CC chemokine receptor 6 (CCR6) is selectively expressed on memory T cells, B cells, and dendritic cells and appears to be involved in the initiation of a memory immune response. The only chemokine ligand for CCR6 is CCL20/MIP-3. In the present study, we attempted to define the extracellular domains (ECDs) of CCR6 responsible for CCL20/MIP-3 binding using a domain-swapping approach in which the ECDs of CCR6 were substituted with the corresponding CCR5 domains to generate various CCR6/CCR5 chimeras. These chimeras were tested for receptor expression, ligand binding, and functional activity as evaluated by calcium flux and chemotaxis. All chimeras showed respectable surface expression; however only one, substituted with extracellular loop 1 from CCR5, showed reduced functional activity. The general failure of functionality of the CCR6/CCR5 chimeras may imply that characteristics of each ECD are critical for coordination among all the ECDs of CCR6. Additionally, of interest, a chimera containing all of the ECDs from CCR5 in the context of CCR6 neither responded to CCR5 ligands nor served as a coreceptor for macrophage-tropic HIV-1. These results suggest that not only ECDs but also transmembrane and intracellular domains of CCR5 are involved in both ligand binding and coreceptor activity.     

Model of glycoprotein hormone receptor ligand binding and signaling

Studies described here were initiated to develop a model of glycoprotein hormone receptor structure and function. We found that the region that links the lutropin receptor leucine-rich repeat domain (LRD) to its transmembrane domain (TMD) has substantial roles in ligand binding and signaling, hence we term it the signaling specificity domain (SSD). Theoretical considerations indicated the short SSDs in marmoset lutropin and salmon follitropin receptors have KH domain folds. We assembled models of lutropin, follitropin, and thyrotropin receptors by aligning models of their LRD, TMD, and shortened SSD in a manner that explains how substitutions in follitropin and thyrotropin receptors distant from their apparent ligand binding sites enable them to recognize lutropins. In these models, the SSD is parallel to the concave surface of the LRD and makes extensive contacts with TMD outer loops 1 and 2. The LRD appears to contact TMD outer loop 3 and a few residues in helices 1, 5, 6, and 7. We propose that signaling results from contacts of the ligands with the SSD and LRD that alter the LRD, which then moves TMD helices 6 and 7. The positions of the LRD and SSD support the notion that the receptor can be activated by hormones that dock with these domains in either of two different orientations. This would account for the abilities of some ligands and ligand chimeras to bind multiple receptors and for some receptors to bind multiple ligands. This property of the receptor may have contributed significantly to ligand-receptor co-evolution.     

TrkA immunoglobulin-like ligand binding domains inhibit spontaneous activation of the receptor

The extracellular region of the nerve growth factor (NGF) receptor, TrkA, contains two immunoglobulin (Ig)-like domains that are required for specific ligand binding. We have investigated the possible role of these two Ig-like domains in receptor dimerization and activation by using different mutants of the TrkA extracellular region. Deletions of each Ig-like domain, of both, and of the entire extracellular region were made. To probe the structural constraints on ligand-independent receptor dimerization, chimeric receptors were generated by swapping the Ig-like domains of the TrkA receptor for the third or fourth Ig-like domain of c-Kit. We also introduced single-amino-acid changes in conserved residues within the Ig-like domains of TrkA. Most of these TrkA variants did not bind NGF, and their expression in PC12nnr5 cells, which lack endogenous TrkA, promoted ligand-independent neurite outgrowth. Some TrkA mutant receptors induced malignant transformation of Rat-1 cells, as assessed by measuring proliferation in the absence of serum, anchorage-independent growth, and tumorigenesis in nude mice. These mutants exhibited constitutive phosphorylation and spontaneous dimerization consistent with their biological activities. Our data suggest that spontaneous dimerization of TrkA occurs when the structure of the Ig-like domains is altered, implying that the intact domains inhibit receptor dimerization in the absence of NGF.     

Structural domains of the insulin receptor and IGF receptor required for dimerisation and ligand binding

We investigated structural requirements for dimerisation and ligand binding of insulin/IGF receptors. Soluble receptor fragments consisting of N-terminal domains (L1/CYS/L2, L1/CYS/L2/F0) or fibronectin domains (F0/F1/F2, F1/F2) were expressed in CHO cells. Fragments containing F0 or F1 domains were secreted as disulphide-linked dimers, and those consisting of L1/CYS/L2 domains as monomers. None of these proteins bound ligand. However, when a peptide of 16 amino acids from the alpha-subunit C-terminus was fused to the C-terminus of L1/CYS/L2, the monomeric insulin and IGF receptor constructs bound their respective ligands with affinity only 10-fold lower than native receptors.     

Dynamic stabilization of nuclear receptor ligand binding domains by hormone or corepressor binding

We have developed a novel assembly assay to examine structural changes in the ligand binding domain (LBD) of the thyroid hormone receptor (TR). Fragments including the first helix of the TR LBD interact only weakly with the remainder of the LBD in the absence of hormone, but this interaction is strongly enhanced by the addition of either hormone or the corepressor NCoR. Since neither the ligand nor the corepressor shows direct interaction with this helix, we propose that both exert their effects by stabilizing the overall structure of the LBD. Current models of activation of nuclear hormone receptors focus on a ligand-induced allosteric shift in the position of the C-terminal helix 12 that generates the coactivator binding site. Our results suggest that ligand binding also has more global effects that dynamically alter the structure of the receptor LBD.     

Mutating the four extracellular cysteines in the chemokine receptor CCR6 reveals their differing roles in receptor trafficking,ligand binding,and signaling

CCR6 is the receptor for the chemokine MIP-3 alpha/CCL20. Almost all chemokine receptors contain cysteine residues in the N-terminal domain and in the first, second, and third extracellular loops. In this report, we have studied the importance of all cysteine residues in the CCR6 sequence using site-directed mutagenesis and biochemical techniques. Like all G protein-coupled receptors, mutating disulfide bond-forming cysteines in the first (Cys118) and second (Cys197) extracellular loops in CCR6 led to complete elimination of receptor activity, which for CCR6 was also associated with the accumulation of the receptor intracellularly. Although two additional cysteines in the N-terminal region and the third extracellular loop, which are present in almost all chemokine receptors, are presumed to form a disulfide bond, this has not been demonstrated experimentally for any of these receptors. We found that mutating the cysteines in the N-terminal domain (Cys36) and the third extracellular loop (Cys288) neither significantly affected receptor surface expression nor completely abolished receptor function. Importantly, contrary to several previous reports, we demonstrated directly that instead of forming a disulfide bond, the N-terminal cysteine (Cys36) and the third extracellular loop cysteine (Cys288) contain free SH groups. The cysteine residues (Cys36 and Cys288), rather than forming a disulfide bond, may be important per se. We propose that CCR6 forms only a disulfide bond between the first (Cys118) and second (Cys197) extracellular loops, which confines a helical bundle together with the N-terminus adjacent to the third extracellular loop, creating the structural organization critical for ligand binding and therefore for receptor signaling.     

Mapping ligand binding domains in chimeric fibroblast growth factor receptor molecules. Multiple regions determine ligand binding specificity

Fibroblast growth factors (FGFs) mediate essential cellular functions by activating one of four alternatively spliced FGF receptors (FGFRs). To determine the mechanism regulating ligand binding affinity and specificity, soluble FGFR1 and FGFR3 binding domains were compared for activity. FGFR1 bound well to FGF2 but poorly to FGF8 and FGF9. In contrast, FGFR3 bound well to FGF8 and FGF9 but poorly to FGF2. The differential ligand binding specificity of these two receptors was exploited to map specific ligand binding regions in mutant and chimeric receptor molecules. Deletion of immunoglobulin-like (Ig) domain I did not effect ligand binding, thus localizing the binding region(s) to the distal two Ig domains. Mapping studies identified two regions that contribute to FGF binding. Additionally, FGF2 binding showed positive cooperativity, suggesting the presence of two binding sites on a single FGFR or two interacting sites on an FGFR dimer. Analysis of FGF8 and FGF9 binding to chimeric receptors showed that a broad region spanning Ig domain II and sequences further N-terminal determines binding specificity for these ligands. These data demonstrate that multiple regions of the FGFR regulate ligand binding specificity and that these regions are distinct with respect to different members of the FGF family.     

Disparity between expression of transferrin receptor ligand binding and non-ligand binding domains on human lymphocytes

We compared transferrin receptor (TfR) expression on human peripheral blood lymphocytes (PBL) activated by phorbol myristate acetate (PMA) or L-phytohemagglutinin (LPHA) using two techniques: (1) 125I-iron-saturated transferrin (FeTf) binding, (2) reactivity with monoclonal anti-TfR antibodies--OKT9 and B3/25. These monoclonal antibodies do not block FeTf binding, and therefore bind to TfR domains separate from the ligand binding site. Unstimulated PBL bound fewer than 1,000 molecules of 125I-FeTf per cell, and less than 5% of cells expressed TfR antigens detected by OKT9 or B3/25. 125I-FeTf binding and antibody binding increased in parallel on LPHA-activated PBL. After exposure to LPHA for 72 hr, 125I-FeTf binding increased 100-fold to 10(5) molecules per cell and greater than 50% of cells expressed TfR antigens. By contrast, PMA activation of PBL markedly increased binding of OKT9 and B3/25 but not the binding of 125I-FeTf. Cell surface expression of TfR antigens seen by OKT9 and B3/25 did not differ between LPHA- and PMA-activated PBL. However, after 72 hr with PMA, 125I-FeTf binding increased only 6-fold and consistently remained at less than 10(4) molecules per cell. Therefore, PMA induced a disparity between expression of TfR ligand binding domains and immunological domains at the cell surface. Cell proliferation assessed by fluorescent DNA analysis was similar in cultures stimulated by LPHA or PMA. These data indicate that lymphoid cells may possess a mechanism for modulating TfR expression in which down-regulation of FeTf binding occurs without receptor internalization. Alternatively, it is possible that this observation may reflect a membrane perturbation effect of PMA.     

Coactivator binding promotes the specific interaction between ligand and the pregnane X receptor

The pregnane X receptor (PXR) detects the presence of a wide variety of endogenous and xenobiotic compounds, and is a master regulator of the expression of genes central to drug metabolism and excretion. We present the 2.0A crystal structure of the human PXR ligand-binding domain (LBD) in complex with the cholesterol-lowering compound SR12813 and a 25 amino acid residue fragment of the human steroid receptor coactivator-1 (SRC-1) containing one LXXLL motif. PXR crystallizes as a homodimer in the asymmetric unit in this structure and possesses a novel alpha2 helix adjacent to its ligand-binding cavity. The SRC-1 peptide forms two distinct helices and binds adjacent to the ligand-dependent transactivation AF-2 helix on the surface of PXR. In contrast with previous PXR structures, in which SR12813 bound in multiple orientations, the small SR12813 agonist in this structure binds in a single, unique orientation within the receptor's ligand-binding pocket and contacts the AF-2 helix. Thermal denaturation studies reveal that the SR12813 ligand and SRC-1 coactivator peptide each stabilize the LBD of PXR, and that together they exert an additive effect on the stability of the receptor. These results indicate that the binding of coactivator to the surface of PXR limits the ability of this promiscuous receptor to "breathe" and helps to trap a single, active conformation of SR12813. They further reveal that specificity is required for PXR activation.     

Contribution to ligand binding by multiple carbohydrate-recognition domains in the macrophage mannose receptor.

The extracellular portion of the macrophage mannose receptor is composed of several cysteine-rich domains, including a fibronectin type II repeat and eight segments related in sequence to Ca(2+)-dependent carbohydrate-recognition domains (CRDs) of animal lectins. Expression of portions of the receptor in vitro, in fibroblasts and in bacteria, has been used to determine which of the extracellular domains are involved in binding and endocytosis of ligand. The NH2-terminal cysteine-rich domain and the fibronectin type II repeat are not necessary for endocytosis of mannose-terminated glycoproteins. CRDs 1-3 have at most very weak affinity for carbohydrate, so the carbohydrate binding activity of the receptor resides in CRDs 4-8. CRD 4 shows the highest affinity binding and has multispecificity for a variety of monosaccharides. However, CRD 4 alone cannot account for the binding of the receptor to glycoproteins. At least 3 CRDs (4, 5, and 7) are required for high affinity binding and endocytosis of multivalent glycoconjugates. In this respect, the mannose receptor is like other carbohydrate-binding proteins, in which several CRDs, each with weak affinity for single sugars, are clustered to achieve high affinity binding to oligosaccharides. In the mannose receptor, these multiple weak interactions are achieved through several active CRDs in a single polypeptide chain rather than by oligomerization of polypeptides each containing a single CRD.     

Ligand-induced heterodimerization between the ligand binding domains of the Drosophila ecdysteroid receptor and ultraspiracle.

The insect ecdysteroid receptor consists of a heterodimer between EcR and the RXR-orthologue, USP. We addressed the question of whether this heterodimer, like all other RXR heterodimers, may be formed in the absence of ligand and whether ligand promotes dimerization. We found that C-terminal protein fragments that comprised the ligand binding, but not the DNA binding domain of EcR and USP and which were equipped with the activation or DNA binding region of GAL4, respectively, exhibit a weak ability to interact spontaneously with each other. Moreover, the heterodimer formation is greatly enhanced upon administration of active ecdysteroids in a dose-dependent manner. This was shown in vivo by a yeast two-hybrid system and in vitro by a modified electromobility shift assay. Furthermore, the EcR fragment expressed in yeast was functional and bound radioactively labelled ecdysteroid specifically. Ligand binding was greatly enhanced by the presence of a USP ligand binding domain. Therefore, ecdysteroids are capable of inducing heterodimer formation between EcR and USP, even when the binding of these receptor proteins to cognate DNA response elements does not occur. This capability may be a regulated aspect of ecdysteroid action during insect development.     

Retention of ligand binding activity by the extracellular domain of the IL-1 receptor

The IL-1R on murine T cells is an 80-kDa cell surface glycoprotein which binds both IL-1 alpha and IL-1 beta. We have recently isolated a cDNA clone encoding this molecule. From the primary sequence mature receptor is predicted to be a 557 residue integral membrane protein with a 319 residue carbohydrate-rich extracellular region. We have constructed a cDNA clone encoding this region of the protein (residues 1 to 316). Expression of this cDNA in HeLa cells leads to secretion of a soluble IL-1 alpha binding protein into the culture medium. Quantitative binding experiments with the truncated receptor show that it possesses IL-1 binding properties which are indistinguishable from those of full length IL-1R. Gel filtration chromatography experiments show that a complex can be formed between a single truncated receptor molecule and a single IL-1 alpha molecule.     

Plasticity of PDZ domains in ligand recognition and signaling

The PDZ domain is a protein-protein interacting module that plays an important role in the organization of signaling complexes. The recognition of short intrinsically disordered C-terminal peptide motifs is the archetypical PDZ function, but the functional repertoire of this versatile module also includes recognition of internal peptide sequences, dimerization and phospholipid binding. The PDZ function can be tuned by various means such as allosteric effects, changes of physiological buffer conditions and phosphorylation of PDZ domains and/or ligands, which poses PDZ domains as dynamic regulators of cell signaling. This review is focused on the plasticity of the PDZ interactions.     

Molecular determinants of glucagon receptor signaling

A 29-amino acid polypeptide hormone, glucagon has been one of the most prolific models in the study of hormone action. The key biologic function of glucagon is to counterbalance the actions of insulin and maintain a normal level of serum glucose. Diabetes mellitus can thus be considered a bihormonal disorder with an excess of glucagon contributing to the hyperglycemic state. The effects of glucagon are mediated by the glucagon receptor, which is itself a prototypical member of a distinct category called family B receptors within the G protein-coupled superfamily of seven-helical transmembrane receptors (GPCRs). At the structural level, the peptide ligands of family B receptors are highly homologous, in particular in the N-terminal region of the molecules. The mechanism by which highly homologous peptide ligands selectively recognize their receptors involves distinct molecular interactions that are gradually being elucidated. This review focuses on structural determinants of the glucagon receptor that are important for its activity with respect to interaction with its ligand and G proteins. Information about the glucagon receptor is presented within the context of what is known about other members of the family B GPCRs.     

New insights into receptor ligand binding domains from a novel assembly assay

Previous studies have demonstrated that hormone binding stabilizes the ligand binding domain (LBD) of the nuclear hormone receptors against proteolysis. We have confirmed and extended this observation using a newly developed assembly assay. In this assay, the LBD is divided into two parts, of which one includes the first helix of this domain and the other corresponds to the remainder of the LBD. Several independent criteria demonstrate that these two fragments can assemble into a functional LBD in the presence of a ligand, but not in its absence, and that this is a reflection of the stabilizing effect of ligand. We have also used this assay to demonstrate that binding of the nuclear receptor corepressor NCoR can directly stabilize the LBD. Overall, these results highlight the dynamic nature of the LBD and suggest that current models for activation based solely on allosteric effects on the C-terminal helix may be too limited.     

Normal ligand binding and signaling by CD47 (integrin-associated protein) requires a long range disulfide bond between the extracellular and membrane-spanning domains.

CD47 is a unique member of the Ig superfamily with a single extracellular Ig domain followed by a multiply membrane-spanning (MMS) domain with five transmembrane segments, implicated in both integrin-dependent and -independent signaling cascades. Essentially all functions of CD47 require both the Ig and MMS domains, raising the possibility that interaction between the two domains is required for normal function. Conservation of Cys residues among CD47 homologues suggested the existence of a disulfide bond between the Ig and MMS domains that was confirmed by chemical digestion and mapped to Cys(33) and Cys(263). Subtle changes in CD47 conformation in the absence of the disulfide were suggested by decreased binding of two anti-Ig domain monoclonal antibodies, decreased SIRPalpha1 binding, and reduced CD47/SIRPalpha1-mediated cell adhesion. Mutagenesis to prevent formation of this disulfide completely disrupted CD47 signaling independent of effects on ligand binding, as assessed by T cell interleukin-2 secretion and Ca(2+) responses. Loss of the disulfide did not affect membrane raft localization of CD47 or its association with alpha(v)beta(3) integrin. Thus, a disulfide bond between the Ig and MMS domains of CD47 is required for normal ligand binding and signal transduction.     

Role of extracellular disulfide-bonded cysteines in the ligand binding function of the beta 2-adrenergic receptor

Evidence is presented for a role of disulfide bridging in forming the ligand binding site of the beta 2-adrenergic receptor (beta AR). The presence of disulfide bonds at the ligand binding site is indicated by "competitive" inhibition by dithiothreitol (DTT) in radioligand binding assays, by specific protection by beta-adrenergic ligands of these effects, and by the requirement of disulfide reduction for limit proteolysis of affinity ligand labeled receptor. The kinetics of binding inhibition by DTT suggest at least two pairs of disulfide-bonded cysteines essential for normal binding. Through site-directed mutagenesis, we indeed were able to identify four cysteines which are critical for normal ligand binding affinities and for the proper expression of functional beta AR at the cell surface. Unexpectedly, the four cysteines required for normal ligand binding are not those located within the hydrophobic transmembrane domains of the receptor (where ligand binding is presumed to occur) but lie in the extracellular hydrophilic loops connecting these transmembrane segments. These findings indicate that, in addition to the well-documented involvement of the membrane-spanning domains of the receptor in ligand binding, there is an important and previously unsuspected role of the hydrophilic extracellular domains in forming the ligand binding site.