Energetics of the cleft closing transition and the role of electrostatic interactions in conformational rearrangements of the glutamate receptor ligand binding domain

The ionotropic glutamate receptors are localized in the pre- and postsynaptic membrane of neurons in the brain. Activation by the principal excitatory neurotransmitter glutamate allows the ligand binding domain to change conformation, communicating opening of the channel for ion conduction. The free energy of the GluR2 S1S2 ligand binding domain (S1S2) closure transition was computed using a combination of thermodynamic integration and umbrella sampling modeling methods. A path that involves lowering the charge on E705 was chosen to clarify the role of this binding site residue. A continuum electrostatics approach in S1S2 is used to show E705, located in the ligand binding cleft, stabilizes the closed conformation of S1S2 via direct interactions with other protein residues, not through the ligand. In the closed conformation, in the absence of a ligand, S1S2 is somewhat more closed than what has been reported in X-ray structures. A semiopen conformation has been identified which is characterized by disruption of a single cross-cleft interaction and differs only slightly in energy from the fully closed S1S2. The fully open S1S2 conformation exhibits a wide energy well and shares structural similarity with the apo S1S2 crystal structure. Hybrid continuum electrostatics/MD calculations along the chosen closure transition pathway reveal solvation energies, and electrostatic interaction energies between two lobes of the protein increase the relative energetic difference between the open and closed conformational states. By analyzing the role of several cross-cleft contacts as well as other binding site residues, we demonstrate how S1S2 interactions facilitate formation of the closed conformation of the GluR2 ligand binding domain.     

Structure of the S1S2 glutamate binding domain of GLuR3

Glutamate receptors are the most prevalent excitatory neurotransmitter receptors in the vertebrate central nervous system. Determining the structural differences between the binding sites of different subtypes is crucial to our understanding of neuronal circuits and to the development of subtype specific drugs. The structures of the binding domain (S1S2) of the GluR3 (flip) AMPA receptor subunit bound to glutamate and AMPA and the GluR2 (flop) subunit bound to glutamate were determined by X‐ray crystallography to 1.9, 2.1, and 1.55 Å, respectively. Overall, the structure of GluR3 (flip) S1S2 is very similar to GluR2 (flop) S1S2 (backbone RMSD of 0.30 ± 0.05 for glutamate‐bound and 0.26 ± 0.01 for AMPA‐bound). The differences in the flip and flop isoforms are subtle and largely arise from one hydrogen bond across the dimer interface and associated water molecules. Comparison of the binding affinity for various agonists and partial agonists suggest that the S1S2 domains of GluR2 and GluR3 show only small differences in affinity, unlike what is found for the intact receptors (with the exception of one ligand, Cl‐HIBO, which has a 10‐fold difference in affinity for GluR2 vs. GluR3). Proteins 2009. © 2008 Wiley‐Liss, Inc.     

Molecular dynamics simulations of the ligand-binding domain of the ionotropic glutamate receptor GluR2.

Ionotropic glutamate receptors are essential for fast synaptic nerve transmission. Recent x-ray structures for the ligand-binding (S1S2) region of the GluR2 alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-sensitive receptor have suggested how differences in protein/ligand interactions may determine whether a ligand will behave as a full agonist. We have used multiple molecular dynamics simulations of 2-5 ns duration to explore the structural dynamics of GluR2 S1S2 in the presence and absence of glutamate and in a complex with kainate. Our studies indicate that not only is the degree of domain closure dependent upon interactions with the ligand, but also that protein/ligand interactions influence the motion of the S2 domain with respect to S1. Differences in domain mobility between the three states (apo-S1S2, glutamate-bound, and kainate-bound) are surprisingly clear-cut. We discuss how these changes in dynamics may provide an explanation relating the mechanism of transmission of the agonist-binding event to channel opening. We also show here how the glutamate may adopt an alternative mode of binding not seen in the x-ray structure, which involves a key threonine (T480) side chain flipping into a new conformation. This new conformation results in an altered pattern of hydrogen bonding at the agonist-binding site.     

Physical location of the site for N-acetyl-L-glutamate, the allosteric activator of carbamoyl phosphate synthetase, in the 20-kilodalton COOH-terminal domain

Mammalian liver mitochondrial carbamoyl phosphate synthetase, a polypeptide of 160 kDa, is activated allosterically by N-acetyl-L-glutamate. The analogue of this activator N-(chloroacetyl)-L-[14C]glutamate has been found to serve as a photoaffinity label for this enzyme. The specificity was demonstrated by the drastic reduction in the radioactivity bound to the protein when (a) an excess of unlabeled acetylglutamate was present during the irradiation and (b) the enzyme was replaced by pyruvate kinase, an enzyme that is not affected by acetylglutamate. The labeling was due to the photoactivation of the chloroacetyl group since there was no labeling under equal conditions with acetyl[14C]glutamate. To localize the binding site, limited proteolysis was used. Trypsin cleaves carbamoyl phosphate synthetase into complementary NH2- and COOH-terminal fragments of about 140 and 20 kDa, respectively [Powers-Lee, S. G., & Corina, K. (1986) J. Biol. Chem. 261, 15349-15352], but only the latter was found to be labeled. Similarly, of the various fragments generated by elastase, only two, of 20 and 120 kDa, contain the COOH terminus [see Powers-Lee and Corina (1986) above] and were found to be labeled. Thus, the binding site for acetylglutamate is within 20 kDa from the COOH terminus. This excludes the possibility that the acetylglutamate binding site evolved from an ancestral substrate site for glutamine: this substrate binds to the small subunit of the Escherichia coli enzyme, which is homologous to the NH2-terminal domain of the rat liver enzyme. Exhaustive tryptic digestion of photolabeled carbamoyl phosphate synthetase yielded a single radioactive peak, suggesting that the labeling is restricted to a single minimal tryptic peptide.     

Conformational change in the human glucocorticoid receptor induced by ligand binding is altered by mutation of isoleucine 747 by a threonine

Limited proteolysis experiments were performed to study conformation changes induced by ligand binding on in vitro produced wild-type and I747T mutant glucocorticoid receptors. Dexamethasone-induced conformational changes were characterized by two resistant proteolysis fragments of 30 and 27 kDa. Although dexamethasone binding affinity was only slightly altered by the I747T substitution (Roux, S., Térouanne, B., Balaguer, P., Loffreda-Jausons, N., Pons, M., Chambon, P., Gronemeyer, H., and Nicolas, J.-C. (1996) Mol. Endocrinol. 10, 1214-1226), higher dexamethasone concentrations were required to obtain the same proteolysis pattern. This difference was less marked when proteolysis experiments were conducted at 0 degrees C, indicating that a step of the conformational change after ligand binding was affected by the mutation. In contrast, RU486 binding to the wild-type receptor induced a different conformational change that was not affected by the mutation. Analysis of proteolysis fragments obtained in the presence of dexamethasone or RU486 indicated that the RU486-induced conformational change affected the C-terminal part of the ligand binding domain differently. These data suggest that the ligand-induced conformational change occurs via a multistep process. In the first step, characterized by compaction of the ligand binding domain, the mutation has no effect. The second step, which stabilizes the activated conformation and does not occur at 4 degrees C, seems to be a key element in the activation process that can be altered by the mutation. This step could involve modification of the helix H12 position, explaining why the conformation induced by RU486 is not affected by the mutation.     

Binding site and interlobe interactions of the ionotropic glutamate receptor GluK3 ligand binding domain revealed by high resolution crystal structure in complex with (S)-glutamate

Ionotropic glutamate receptors (iGluRs) are involved in excitatory signal transmission throughout the central nervous system and their malfunction is associated with various health disorders. GluK3 is a subunit of iGluRs, belonging to the subfamily of kainate receptors (GluK1–5). Several crystal structures of GluK1 and GluK2 ligand binding domains have been determined in complex with agonists and antagonists. However, little is known about the molecular mechanisms underlying GluK3 ligand binding properties and no compounds displaying reasonable selectivity towards GluK3 are available today. Here, we present the first X-ray crystal structure of the ligand binding domain of GluK3 in complex with glutamate, determined to 1.6 Å resolution. The structure reveals a conserved glutamate binding mode, characteristic for iGluRs, and a water molecule network in the glutamate binding site similar to that seen in GluK1. In GluK3, a slightly lower degree of domain closure around glutamate is observed compared to most other kainate receptor structures with glutamate. The volume of the GluK3 glutamate binding cavity was found to be of intermediate size between those of GluK1 and GluK2. The residues in GluK3 contributing to the subfamily differences in the binding sites are primarily: Thr520, Ala691, Asn722, Leu736 and Thr742. The GluK3 ligand binding domain seems to be less stabilized through interlobe interactions than GluK1 and this may contribute to the faster desensitization kinetics of GluK3.     

Oligomerization and ligand-binding properties of the ectodomain of the alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptor subunit GluRD.

The extracellular part of ionotropic glutamate receptor (iGluR) subunits can be divided into a conserved two-lobed ligand-binding domain ("S1S2") and an N-terminal approximately 400-residue segment of unknown function ("X domain") which shows high sequence variation among subunits. To investigate the structure and properties of the N-terminal domain, we have now produced affinity-tagged recombinant fragments which represent the X domain of the GluRD subunit of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-selective glutamate receptors either alone or covalently linked to the ligand-binding domain ("XS1S2"). These fragments were expressed in insect cells as secreted soluble proteins and were recognized by a conformation-specific anti-GluRD monoclonal antibody. A hydrodynamic analysis of the purified fragments revealed them to be dimers, in contrast to the S1S2 ligand-binding domain which is monomeric. The X domain did not bind radiolabeled AMPA or glutamate nor did its presence affect the ligand binding properties of the S1S2 domain. Our findings demonstrate that the N-terminal domain of AMPA receptor can be expressed as a soluble polypeptide and suggest that subunit interactions in iGluR may involve the extracellular domains.     

Coupled control of desensitization and gating by the ligand binding domain of glutamate receptors

The kinetics of ligand gated ion channels are tuned to permit diverse roles in cellular signaling. To follow high-frequency excitatory synaptic input, postsynaptic AMPA-type glutamate receptors must recover rapidly from desensitization. Chimeras between AMPA and the related kainate receptors demonstrate that the ligand binding domains alone control the lifetime of the desensitized state. Mutation of nonconserved amino acids in the lower lobe (domain 2) of the ligand binding domain conferred slow recovery from desensitization on AMPA receptors, and fast recovery on kainate receptors. Single-channel recordings and a correlation between the rate of deactivation and the rate of recovery across panels of mutant receptors revealed that domain 2 also controls ion channel gating. Our results demonstrate that the same mechanism that ensures fast recovery also sharpens the response of AMPA channels to synaptically released glutamate.     

Kainate induces various domain closures in AMPA and kainate receptors

Ionotropic glutamate receptors are key players in fast excitatory synaptic transmission within the central nervous system. These receptors have been divided into three subfamilies: the N-methyl-d-aspartic acid (NMDA), 2-amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA) and kainate receptors. Kainate has previously been crystallized with the ligand binding domain (LBD) of AMPA receptors (GluA2 and GluA4) and kainate receptors (GluK1 and GluK2). Here, we report the structures of the kainate receptor GluK3 LBD in complex with kainate and GluK1 LBD in complex with kainate in the absence of glycerol. Kainate introduces a conformational change in GluK3 LBD comparable to that of GluK2, but different from the conformational changes induced in GluA2 and GluK1. Compared to their domain closures in a glutamate bound state, GluA2 and GluK1 become more open and kainate induces a domain closure of 60% and 62%, respectively, relative to glutamate (100%). In GluK2 and GluK3 with kainate, the domain closure is 88% and 83%, respectively. In previously determined structures of GluK1 LBD in complex with kainate, glycerol is present in the binding site where it bridges interlobe residues and thus, might contribute to the large domain opening. However, the structure of GluK1 LBD with kainate in the absence of glycerol confirms that the observed domain closure is not an artifact of crystallization conditions. Comparison of the LBD structures with glutamate and kainate reveals that contacts are lost upon binding of kainate in the three kainate receptors, which is in contrast to the AMPA receptors where similar contacts are seen. It was revealed by patch clamp electrophysiology studies that kainate is a partial agonist at GluK1 with 36% efficacy compared to glutamate, which is in between the published efficacies of kainate at GluK2 and AMPA receptors. The ranking of efficacies seems to correlate with LBD domain closures.     

Crystal structures of the GluR5 and GluR6 ligand binding cores: molecular mechanisms underlying kainate receptor selectivity

Little is known about the molecular mechanisms underlying differences in the ligand binding properties of AMPA, kainate, and NMDA subtype glutamate receptors. Crystal structures of the GluR5 and GluR6 kainate receptor ligand binding cores in complexes with glutamate, 2S,4R-4-methylglutamate, kainate, and quisqualate have now been solved. The structures reveal that the ligand binding cavities are 40% (GluR5) and 16% (GluR6) larger than for GluR2. The binding of AMPA- and GluR5-selective agonists to GluR6 is prevented by steric occlusion, which also interferes with the high-affinity binding of 2S,4R-4-methylglutamate to AMPA receptors. Strikingly, the extent of domain closure produced by the GluR6 partial agonist kainate is only 3 degrees less than for glutamate and 11 degrees greater than for the GluR2 kainate complex. This, together with extensive interdomain contacts between domains 1 and 2 of GluR5 and GluR6, absent from AMPA receptors, likely contributes to the high stability of GluR5 and GluR6 kainate complexes.     

On the binding determinants of the glutamate agonist with the glutamate receptor ligand binding domain

Ionotropic glutamate receptors (GluRs) are ligand-gated membrane channel proteins found in the central neural system that mediate a fast excitatory response of neurons. In this paper, we report theoretical analysis of the ligand-protein interactions in the binding pocket of the S1S2 (ligand binding) domain of the GluR2 receptor in the closed conformation. By utilizing several theoretical methods ranging from continuum electrostatics to all-atom molecular dynamics simulations and quantum chemical calculations, we were able to characterize in detail glutamate agonist binding to the wild-type and E705D mutant proteins. A theoretical model of the protein-ligand interactions is validated via direct comparison of theoretical and Fourier transform infrared spectroscopy (FTIR) measured frequency shifts of the ligand's carboxylate group vibrations [Jayaraman et al. (2000) Biochemistry 39, 8693-8697; Cheng et al. (2002) Biochemistry 41, 1602-1608]. A detailed picture of the interactions in the binding site is inferred by analyzing contributions to vibrational frequencies produced by protein residues forming the ligand-binding pocket. The role of mobility and hydrogen-bonding network of water in the ligand-binding pocket and the contribution of protein residues exposed in the binding pocket to the binding and selectivity of the ligand are discussed. It is demonstrated that the molecular surface of the protein in the ligand-free state has mainly positive electrostatic potential attractive to the negatively charged ligand, and the potential produced by the protein in the ligand-binding pocket in the closed state is complementary to the distribution of the electrostatic potential produced by the ligand itself. Such charge complementarity ensures specificity to the unique charge distribution of the ligand.     

Mechanisms for activation and antagonism of an AMPA-sensitive glutamate receptor: crystal structures of the GluR2 ligand binding core

Crystal structures of the GluR2 ligand binding core (S1S2) have been determined in the apo state and in the presence of the antagonist DNQX, the partial agonist kainate, and the full agonists AMPA and glutamate. The domains of the S1S2 ligand binding core are expanded in the apo state and contract upon ligand binding with the extent of domain separation decreasing in the order of apo > DNQX > kainate > glutamate approximately equal to AMPA. These results suggest that agonist-induced domain closure gates the transmembrane channel and the extent of receptor activation depends upon the degree of domain closure. AMPA and glutamate also promote a 180 degrees flip of a trans peptide bond in the ligand binding site. The crystal packing of the ligand binding cores suggests modes for subunit-subunit contact in the intact receptor and mechanisms by which allosteric effectors modulate receptor activity.     

The SH3 domain of HS1 protein recognizes lysine-rich polyproline motifs

HS1 is a protein involved in erythroid proliferation and apoptotic cell death, containing several structurally significant motifs including a C-terminal SH3 domain. HPK1 is a member of the Ste20-related kinase family, which contains four proline-rich sequences and is constitutively associated with HS1 in hematopoietic cells. Recombinant fusion protein GST-SH3_HS1 was expressed to assess the binding properties of 16 peptides derived from the HPK1 proline-rich regions. The binding affinities were determined by non-immobilized ligand interaction assay by circular dichroism. Our results revealed that the classical PxxPxK class II binding motif is not sufficient to induce the interaction with the GST-SH3_HS1 domain, an event dependent on the presence of additional basic residue(s) located at the C-terminus of the PxxPxK motif: Lys₋₅ in P2 peptide and Lys₋₈ in P4c peptide. Lys replacement by Arg residues decreases the ligand binding affinity. The finding that both SH3_HS1 domain and full-length HS1 protein bind to P2 peptide with similar affinity demonstrates that the whole protein sequence does not affect the interaction properties of the domain. In silico models of SH3_HS1 as a complex with P2 or P4c highlight the domain residues that interact with the recognition determinants of the peptide ligand and that cooperate in the complex stabilization.     

Agonist-induced isomerization in a glutamate receptor ligand-binding domain. A kinetic and mutagenetic analysis

Agonist binding to glutamate receptor ion channels occurs within an extracellular domain (S1S2) that retains ligand affinity when expressed separately. S1S2 is homologous to periplasmic binding proteins, and it has been proposed that a Venus flytrap-style cleft closure triggers opening of glutamate receptor ion channels. Here we compare the kinetics of S1S2-agonist binding to those of the periplasmic binding proteins and show that the reaction involves an initial rapid association, followed by slower conformational changes that stabilize the complex: "docking" followed by "locking." The motion detected here reflects the mechanism by which the energy of glutamate binding is converted into protein conformational changes within S1S2 alone. In the intact channel, these load-free conformational changes are harnessed and possibly modified as the agonist binding reaction is used to drive channel opening and subsequent desensitization. Using mutagenesis, key residues in each step were identified, and their roles were interpreted in light of a published S1S2 crystal structure. In contrast to the Venus flytrap proposal, which focuses on motion between the two lobes as the readout for agonist binding, we argue that smaller, localized conformational rearrangements allow agonists to bridge the cleft, consistent with published hydrodynamic measurements.     

Binding of a photoaffinity analogue of pyridoxal to pyridoxal kinase

The binding of pyridoxal analogues to the structural domains of pyridoxal kinase was studied by fluorescence spectroscopy and chromatographic techniques. Two fragments of 24 and 16 kDa, arising from limited proteolysis of the native enzyme, were separated by ion-exchange chromatography and used for binding studies with pyridoxal oxime. Fluorometric titrations yielded dissociation constants of 6 and 12.4 MicroM for pyridoxal oxime bound to the native enzyme and 24-kDa fragment, respectively. 4-(4-Azido-2-nitrophenyl)-pyridoxamine, a new photolabeling reagent, binds irreversibly to the kinase with concomitant loss of catalytic activity. The modified kinase (2.1 mol label/mol dimer) yields two fragments upon limited proteolysis with chymotrypsin. The two fragments were separated by reverse-phase HPLC and SDS/polyacrylamide gel electrophoresis. Radiolabeled ligand was detected only in the 24-kDa fragment. It is postulated that the pyridoxal binding site is located in the 24-kDa structural domain.     

Evaluation of the membrane-spanning domain of ClC-2

The ClC family of chloride channels and transporters includes several members in which mutations have been associated with human disease. An understanding of the structure-function relationships of these proteins is essential for defining the molecular mechanisms underlying pathogenesis. To date, the X-ray crystal structures of prokaryotic ClC transporter proteins have been used to model the membrane domains of eukaryotic ClC channel-forming proteins. Clearly, the fidelity of these models must be evaluated empirically. In the present study, biochemical tools were used to define the membrane domain boundaries of the eukaryotic protein, ClC-2, a chloride channel mutated in cases of idiopathic epilepsy. The membrane domain boundaries of purified ClC-2 and accessible cysteine residues were determined after its functional reconstitution into proteoliposomes, labelling using a thiol reagent and proteolytic digestion. Subsequently, the lipid-embedded and soluble fragments generated by trypsin-mediated proteolysis were studied by MS and coverage of approx. 71% of the full-length protein was determined. Analysis of these results revealed that the membrane-delimited boundaries of the N- and C-termini of ClC-2 and the position of several extramembrane loops determined by these methods are largely similar to those predicted on the basis of the prokaryotic protein [ecClC (Escherichia coli ClC)] structures. These studies provide direct biochemical evidence supporting the relevance of the prokaryotic ClC protein structures towards understanding the structure of mammalian ClC channel-forming proteins.     

Steroid binding activity is retained in a 16-kDa fragment of the steroid binding domain of rat glucocorticoid receptors

The steroid binding domain of the rat glucocorticoid receptor is considered as extending from amino acids 550 to 795. However, such a synthetic protein (i.e. amino acids 547-795; Mr approximately 31,000) has been reported to show very little affinity for the potent synthetic glucocorticoid dexamethasone. We now disclose that digestion of steroid-free rat glucocorticoid receptors with low concentrations of trypsin yields a single species, of Mr = 16,000, that is specifically labeled by dexamethasone 21-mesylate. This 16-kDa fragment retains high affinity binding for [3H]dexamethasone that is only approximately 23-fold lower than that seen with the intact 98-kDa receptor. Analysis of the protease digestion patterns obtained both with trypsin and with lysylendopeptidase C allowed us to deduce the proteolytic cleavage maps of the receptor with these enzymes. From these protease maps, the sequence of the 16-kDa fragment was identified as being threonine 537 to arginine 673. These results show that glucocorticoid receptor fragments smaller than 34 kDa do bind steroids and that the amino acids Thr537-Arg673 constitute a core sequence for ligand binding within the larger steroid binding domain. The much slower kinetics in generating the 16-kDa fragment from affinity-labeled receptors suggests that steroid binding causes a conformation change in the receptor near the cleavage sites.     

Interaction of fMet-tRNAfMet and fMet-AMP with the C-terminal domain of Thermus thermophilus translation initiation factor 2.

Two polypeptides resistant against proteolytic digestion were identified in Thermus thermophilus translation initiation factor 2 (IF2): the central part of the protein (domains II/III), and the C-terminal domain (domain IV). The interaction of intact IF2 and the isolated proteolytic fragments with fMet-tRNAfMet was subsequently characterized. The isolated C-terminal domain was as effective in binding of the 3' end of fMet-tRNAf Met as intact IF2. N-Formylation of Met-tRNAfMet was required for its efficient binding to the C-terminal domain. This suggests that the interaction between the C-terminal domain and the 3' end of fMet-tRNAfMet is responsible for the recognition of fMet-tRNAfMet by IF2 during translation initiation. Moreover, it was demonstrated that fMet-AMP is a minimal ligand of IF2. fMet-AMP inhibits fMet-tRNAfMet binding to IF2 as well as the activity of IF2 in the stimulation of ApUpG-dependent ribosomal binding of fMet-tRNAf Met. Specific interaction of fMet-AMP with IF2 was demonstrated by 1H-NMR spectroscopy. These findings indicate that fMet-AMP and the 3' terminal fMet-adenosine of fMet-tRNAfMet use the same binding site on the C-terminal domain of IF2 and imply that the interaction between the C-terminal domain and the 3' end of fMet-tRNAfMet is primarily responsible for the fMet-tRNAfMet binding and recognition by IF2.