FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2096164027> ?p ?o ?g. }

• W2096164027 endingPage "982" @default.
• W2096164027 startingPage "972" @default.
• W2096164027 abstract "Article11 February 2011free access Dynamics and allosteric potential of the AMPA receptor N-terminal domain Madhav Sukumaran Madhav Sukumaran Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Maxim Rossmann Maxim Rossmann Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Indira Shrivastava Indira Shrivastava Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Anindita Dutta Anindita Dutta Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Ivet Bahar Ivet Bahar Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Ingo H Greger Corresponding Author Ingo H Greger Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Madhav Sukumaran Madhav Sukumaran Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Maxim Rossmann Maxim Rossmann Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Indira Shrivastava Indira Shrivastava Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Anindita Dutta Anindita Dutta Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Ivet Bahar Ivet Bahar Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA Search for more papers by this author Ingo H Greger Corresponding Author Ingo H Greger Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK Search for more papers by this author Author Information Madhav Sukumaran1,‡, Maxim Rossmann1,‡, Indira Shrivastava2, Anindita Dutta2, Ivet Bahar2 and Ingo H Greger 1 1Neurobiology Division, MRC Laboratory of Molecular Biology, Cambridge, UK 2Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA ‡These authors contributed equally to this work *Corresponding author. Neurobiology Division, MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 0QH, UK. Tel.: +44 122 340 2173; Fax: +44 122 340 2310; E-mail: [email protected] The EMBO Journal (2011)30:972-982https://doi.org/10.1038/emboj.2011.17 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Glutamate-gated ion channels (ionotropic glutamate receptors, iGluRs) sense the extracellular milieu via an extensive extracellular portion, comprised of two clamshell-shaped segments. The distal, N-terminal domain (NTD) has allosteric potential in NMDA-type iGluRs, which has not been ascribed to the analogous domain in AMPA receptors (AMPARs). In this study, we present new structural data uncovering dynamic properties of the GluA2 and GluA3 AMPAR NTDs. GluA3 features a zipped-open dimer interface with unconstrained lower clamshell lobes, reminiscent of metabotropic GluRs (mGluRs). The resulting labile interface supports interprotomer rotations, which can be transmitted to downstream receptor segments. Normal mode analysis reveals two dominant mechanisms of AMPAR NTD motion: intraprotomer clamshell motions and interprotomer counter-rotations, as well as accessible interconversion between AMPAR and mGluR conformations. In addition, we detect electron density for a potential ligand in the GluA2 interlobe cleft, which may trigger lobe motions. Together, these data support a dynamic role for the AMPAR NTDs, which widens the allosteric landscape of the receptor and could provide a novel target for ligand development. Introduction Binding of L-glutamate to ionotropic glutamate receptors (iGluRs) initiates excitatory neurotransmission in vertebrate central nervous systems. This process is mediated by a series of conformational transitions, ultimately resulting in opening of the ion channel and depolarization of the post-synaptic membrane (Traynelis et al, 2010). Ionotropic GluRs are arranged as dimers of dimers into receptor tetramers (Sobolevsky et al, 2009). The extracellular portion of each subunit consists of two domains, the ligand-binding domain (LBD) and the N-terminal domain (NTD), which resemble bacterial periplasmic-binding proteins (PBPs), ancient bilobate structures evolutionarily selected to capture ligand (O'Hara et al, 1993; Quiocho and Ledvina, 1996; Madden, 2002). Within subunit dimers, these domains are arranged as two-fold symmetric pairs of protomers, each consisting of two lobes, the upper and lower lobes (UL and LL; Armstrong and Gouaux, 2000; Mayer, 2005; Clayton et al, 2009; Jin et al, 2009). L-Glutamate docks to the membrane proximal LBD, which triggers lobe closure and initiation of the gating cascade. In AMPA-type iGluRs (AMPARs), the number of glutamate molecules bound to the receptor (up to four) determines open/closed-channel states and gives rise to complex gating properties, which ultimately shape excitatory signalling. AMPAR gating kinetics are modulated further by alternative RNA processing within the LBD and by a variety of drugs targeting the LBD dimer interface and binding cleft, respectively (Traynelis et al, 2010) (Lomeli et al, 1994; Mosbacher et al, 1994; Jin et al, 2005). Through their capacity to strengthen AMPAR transmission, small-molecule LBD modulators have entered clinical trials as cognitive enhancers (Lynch, 2002; Bowie, 2008; Ward et al, 2010). The second, membrane-distal extracellular portion, the NTD, is structurally related to bacterial leucine-binding protein (Trakhanov et al, 2005), and also closely resembles the ligand-binding cores (LBCs) of natriuretic peptide receptors (He et al, 2005) and type-C G-protein-coupled receptors (GPCRs), including the type-B γ-aminobutyric receptor (GABABR) and the metabotropic glutamate receptors (mGluR1–8; Pin et al, 2003). In mGluRs, glutamate binding within the interlobe cleft triggers a ∼30° interlobe closure motion and a rearrangement of the dimer interface, which initiates G-protein signalling (Kunishima et al, 2000; Tsuchiya et al, 2002). To date, ligand binding to this distal domain in iGluRs has been associated exclusively with the NMDA-type receptors (NMDARs). Zn2+ docking to NR-2 subunit NTDs results in a downregulation of channel activity, presumably via closure of the NTD clamshell (Karakas et al, 2009; Hansen et al, 2010). The opposite effect, an increase in NMDAR open probability, was achieved by wedging the cleft open, implying bi-directional control of channel activity via NTD clamshell motions (Gielen et al, 2009). An allosteric path, originating in the ligand-binding cleft, successively transmitted through the NTD–LBD linker region, the LBD dimer interface and down to the channel gate, has been suggested for NMDARs (Gielen et al, 2008, 2009). Ifenprodil and related NTD-targeting drugs modulate NMDARs and thereby further enrich the functional spectrum of these ion channels (Mony et al, 2009; Hansen et al, 2010). By contrast, AMPAR NTD ligands have not been described, and recent structural data of GluA2 and GluK2 NTDs in the nonNMDAR subfamily have been interpreted to rule out a signalling capacity for this domain (Jin et al, 2009; Kumar et al, 2009). Also, whereas in mGluRs, the LLs of the clamshell are free to rotate upward in response to ligand binding (Kunishima et al, 2000), the LLs in GluA2 and GluK2 NTDs appear constrained due to dimeric packing. As a result, the NTD has been suggested to function purely as a rigid subunit assembly device in nonNMDA receptors (Jin et al, 2009; Kumar et al, 2009). In this study, we present new high-resolution AMPAR NTD structures and the first analysis of their structure-encoded dynamics, which reveal (i) electron density within the GluA2-binding pocket and thus potential ligand-binding capacity for AMPAR NTDs, (ii) a structurally labile GluA3 dimer interface, which facilitates interprotomer rearrangements; and (iii) an intrinsic ability of the protomers themselves to undergo clamshell-like motions, similar to other PBPs. Normal mode analysis (NMA) based on the anisotropic network model (ANM) (Atilgan et al, 2001; Bahar et al, 2010a, 2010b) suggests that classic clamshell motions are more prominent in GluA3 due to unconstrained LLs (similar to mGluR LBCs), but can also be discerned in the more tightly packed GluA2 NTD. NMA further demonstrates that iGluR NTD global motions resemble those of mGluR LBCs. In sum, AMPAR NTDs may have mGluR-like signalling capacity. Our data uncover an allosteric potential for AMPAR NTDs. Modulation via the NTD would widen the functional spectrum of AMPARs and potentially opens a currently unexplored target for ligand development. Results The GluA3 NTD features unconstrained LLs, resembling mGluRs The iGluR NTD comprises the most distal portion of the receptor (Figure 1A and B) and is believed to interact with presynaptic components and secreted factors, including pentraxins in AMPARs (Hansen et al, 2010; see also Figure 6). Contrary to its well-established allosteric potential in NMDARs, this domain has been suggested to merely act as a rigid subunit assembly module in nonNMDARs (based on GluA2 and GluK2 structures). However, the assembly characteristics of the AMPAR NTDs show unexpected diversity with GluA2 and GluA3 lying at functional extremes (Rossmann et al, 2011; Sukumaran et al, 2011). The GluA3 NTD features the weakest homodimeric affinity in solution and harbours conspicuous sequence variations in the LL interface (Figure 1D). To better understand the biology of this elusive domain, we targeted GluA3 for X-ray crystallographic studies. Figure 1.GluA2 and GluA3 NTDs differ structurally. (A) Left: Topology of an iGluR subunit. The NTD segment is denoted as a green curve and the transmembrane segments as grey columns. Right: Structure of the bipartite GluA2 NTD dimer (PDB 3HSY). The two chains/protomers are coloured green and cyan. Upper and lower lobes (UL, LL) are denoted and their respective interprotomer interfaces are circled. Secondary structural elements contributing to the LL interface are labelled. (B) Structure of the GluA3 NTD (dimer I), with the two protomers coloured red and blue. The UL dimer interface analogous to GluA2 is circled, and the LL interface is shown by a box and an arrow indicating the increased space between the LLs, compared with GluA2. Segments homologous to the GluA2 LL interface segments (from A) are labelled. (C) Lower lobe packing markedly differs between GluA2 and GluA3 NTDs. The lower lobe interface of GluA2 (green) and GluA3 (red) are shown after aligning common secondary structure segments. Note the significantly closer packing of the GluA2 LL interface. Also shown are arginines from GluA3 that project into the interface; this unfavourable electrostatic interaction may contribute to the increased interlobe distance. (D) Sequence conservation in the NTD LL of the AMPA and kainate subfamilies. Different background colours indicate different conservation patterns; for example, conserved sites (columns) within a subfamily are coloured red. Residues that project across the interface are denoted with asterisks (*). Note the markedly higher conservation of the LL interface within the kainate subfamily. See also Supplementary Figures S1 and S2. Download figure Download PowerPoint GluA3 NTD crystals diffracted to 2.2 Å. The structure was solved by molecular replacement using GluA2 (PDB 3HSY; Greger et al, 2009) as a search probe; two dimers (I and II) are present in the asymmetric unit (Supplementary Table I). Overall, the architecture of the bilobed protomer and packing across the UL interface in dimer I was highly similar to GluA2 (root-mean-square displacement (RMSD) 0.6 Å; Figure 1B; Supplementary Figure S1). The most striking difference is a repositioning of the LLs, in which GluA3 dimer I are widely separated, up to 8 Å relative to the spacing between the GluA2 LLs (Figure 1B; Supplementary Figure S2C). The LL arrangement observed in the GluA3 NTD bears a striking resemblance to mGluR1 and the natriuretic peptide receptor LBCs, where signalling via flexible LLs is well established (Kunishima et al, 2000; He et al, 2001; Tsuchiya et al, 2002). In fact, GluA3 and mGluR1 show a very similar degree of LL separation (Supplementary Figure S2). Thus, unlike GluA2, in GluA3 the LLs are not constrained by dimeric packing, but have greater freedom to move and may thus propagate signal. It is worth pointing out that ligand-independent clamshell motions have been deduced from experimental data in NMDARs (Gielen et al, 2009); a related scenario mGluA3 (see below). The structure also provides an immediate explanation for the relatively low GluA3 NTD dimer affinity measured in solution (Rossmann et al, 2011). A closer examination of the LL interface reveals that, contrary to GluA2, the GluA3 LL interface is largely polar in dimer I, which was not anticipated previously from sequence alignments (Jin et al., 2009). In particular, Arg163 and Arg184 project towards the interface (Figure 1C) generating positive electrostatic potential (Supplementary Figure S3A); charge repulsion presumably contributes to the increased lobe separation seen in dimer I. Arg163 is replaced by hydrophobic residues in the other AMPAR subunits—in GluA2 Ile157 takes its place and engages Ala148 of the opposite protomer in hydrophobic contacts (Figure 1D). GluK1–3 kainate receptors also harbour Arg at this position (Figure 1D); however, the positive charge is shielded effectively by Glu186 and Glu192 (Kumar et al, 2009). Interestingly, Figure 1D also shows that in kainate receptors the LLs are well conserved, in apparent contrast to AMPARs. We conclude that the previously described ‘locked’ GluA2 dimer, which is also seen in the GluK2 kainate receptor (Jin et al, 2009; Kumar et al, 2009), is not universally found in all nonNMDARs. Interprotomer rearrangements in the GluA3 NTD NTD-driven allostery involves the dimer interface in NMDARs (Hansen et al, 2010) and in the analogous mGluRs, where a large-scale reorientation of dimeric contacts were observed crystallographically (Kunishima et al, 2000). A similar picture is seen with the ANP receptor (He et al, 2001). Similarly, interfacial rearrangements of the membrane-proximal LBD in AMPARs couple between active and non-active, desensitized states (Mayer and Armstrong, 2004; Mayer, 2005; Armstrong et al, 2006). Comparison of two GluA3 quaternary conformations observed in the crystal structures, dimer I and dimer II, show large differences, suggesting that GluA3 protomers also possess the ability to adopt alternative quaternary forms. Dimer II features a counter-rotation along an axis perpendicular to the dimer interface, relative to dimer I, resulting in rearrangements across both the UL and LL dimer interfaces (Figure 2B). Indeed, in a different crystal form of the GluA3 NTD, we find an additional dimer form, dimer III (PDB 3P3W), alongside the original dimer I configuration (Figure 2A; Supplementary Table I), underscoring the fact that the GluA3 NTD can adopt multiple quaternary structures. As dimer I is found in both crystal forms, it seems to be energetically favoured. Although new contacts between the LLs are formed in dimers II and III, contacts within the UL interface are diminished, which may result in an overall less stable and thus more heterogeneous interface (Supplementary Table II and Figure S4A). The relevance of GluA3 dimer II could be assessed in solution: side chains of M150 in the apposing LLs come into close proximity to one another (Supplementary Figure S4B). Conservative mutation of this position to Cys, M150C, resulted in a greater proportion of crosslinked dimer on non-reducing SDS–PAGE (Supplementary Figure S4B), suggesting that dimer II is accessible in solvent, even under more dilute, non-crystallographic conditions. Figure 2.GluA3 NTD crystal structures exhibit different protomerprotomer packings and interfacial contacts. (A) GluA3 crystallizes in three distinct dimeric forms. The dimeric arrangements of each form are shown from above and from the side, with the molecular surfaces of one protomer from dimers I, II and III coloured red, brown and yellow, respectively. (B) Dimers I, II and III are related by rigid-body motions of their protomers. The grey protomers from panel A have been aligned to within 0.5 Å RMSD of each other, whereas the second (coloured) protomer in each structure is left free. The resulting superposition is shown in the inset (the side view from panel A) and turned ∼90°, looking onto the packing surface of each dimer in the main panel. Translational and rotational shifts between interface helices (UL: B and C, LL: E and F) are shown. Note there is a large (∼16 Å) difference between the packing of LL helices in dimer I (red) and II (brown), whereas dimer III (yellow) assumes an intermediate position. The shifts in the UL are indicated for rotation between dimers I and II for helices B and C. The structural difference is more accentuated in the LL, due to intraprotomer structural variabilities. (C) Superposition from B, viewed from the bottom. The two-fold symmetry axis and the plane of the interface are shown as a circle and a dashed line, respectively. Download figure Download PowerPoint Overall, the three alternative quaternary conformations I, II and III provide examples of dimeric rearrangements that are energetically accessible to the GluA3 NTD. As shown in Figure 2B, interprotomer translational and rotational displacements up to 16 Å and ∼11° are observed, whereas no significant intraprotomer changes are apparent (RMSDs 0.45–0.52 Å, when superimposing main chain Cα atoms). As shown below, NMA reveals that the repositioning of the two protomers with respect to each other is enabled by the top-ranking, or softest, normal mode intrinsically accessible to the GluA3 dimeric architecture, and that the observed structures could represent snapshots along this readily accessible mode of motion. The GluA2 LLs are not tightly packed and exhibit structural variabilities The structural variabilities observed in GluA3 prompted us to analyse and compare the GluA2 NTD dimer interface, where the LLs appear constrained by dimeric packing (Figure 1). Protein interfaces have been classified into functionally relevant and those generated by crystal packing, on the basis of the physicochemical properties intrinsic to the interface (Bahadur et al, 2004; Bordner and Abagyan, 2005). One property that reliably discriminates between stable biological interactions versus nonspecific and comparatively weak crystal-packing interfaces is the local atomic contact density (LD). Nonspecific interfaces have been shown to have an LD below 40, whereas values above indicate stable, biologically relevant packing interfaces (Bahadur et al, 2004). Using this method, we investigated the highest resolution structures of GluA2 and GluA3 currently available (PDB 3HSY, 3O21). As expected from the crystal structures, the GluA3 dimer interfaces showed extensive variability. For example, we find that the LD is 42.7 in the UL interface of dimer I, but is <5 in the LL (Figure 3A and Supplementary Figure S4A), indicating that the dimer I UL represents a biologically relevant interface. In contrast, dimer II showed reduced LDs for both UL and LL (approximately 31–32; Supplementary Figure 4A), suggesting that both interfaces in dimer II are less stable than the dimer I UL. Because of lower resolution (4.2 Å), dimer III was not subjected to atomic level analysis, but due to its similar arrangement to dimer II, we expect that dimer III will also be less stable than dimer I. We examined other interface parameters studied by Bahadur et al, (2004), including solvent-accessible surface area, hydrophobicity and evolutionary conservation, which also indicated that the UL and LL interfaces of dimer II are less stable than the UL of dimer I (Supplementary Table II). Figure 3.The LLs in GluA2 and GluA3 NTDs exhibit fewer interfacial contacts and larger structural variabilities compared with the ULs. (A) View onto the dimer interface. Atoms making contacts across the dimer interface within 4.5 Å are depicted as spheres on the molecular surface of the respective monomer. Spheres are coloured by number of contacts from blue (1 contact) to red (⩾7 contacts). Local contact density (LD) is also noted for each interface. (B) Structural variations from known GluA2 NTD structures. All GluA2 NTD crystal structures at resolution of 2.5 Å or better (PDB 3HSY, 2WJW, 3H5V) were analysed for different backbone conformations. UL and LL cores from each chain were aligned to a reference chain (3HSY chain B) to within 0.5 Å RMSD and deviations were measured for backbone atoms; these backbone atom deviations are mapped back onto the 3HSY chain B structure and coloured on a logarithmic scale from 0.1 Å (blue) to 10 Å (red). Whereas loops in the UL were largely invariant, specific loops in the LL showed deviations of up to 0.1 Å. Inset: Using the above alignment and superposition algorithm, segments from 3H5V (αF and αG) and 2WJW (αI) exhibiting the most extreme displacements from 3HSY chain B are highlighted in grey. See also Supplementary Figure S4. Download figure Download PowerPoint Interestingly, when extended to GluA2, we find that these parameters similarly point to relatively weak GluA2 LL contacts, whereas the UL interface classifies as stronger and biologically relevant (Supplementary Table II). The atomic packing densities in the UL are ∼2-fold larger than those computed for the LL, with an LD of 43.5 versus 21.5 (Figure 3A). In addition, hydrophobicity and evolutionary conservation are markedly reduced between LLs (Supplementary Table II), whereas the solvent content is increased (two- to three-fold), signifying more polar, less stable contacts between the LLs compared with the UL interface (Supplementary Figure S3A, B; Dey et al, 2010). In sum, this analysis concludes that the LL interface in GluA2 makes weaker contacts compared with the UL interface. Therefore, the GluA2 LLs could accommodate rearrangements. In order to examine potential flexibility in the GluA2 LL, we compiled published GluA2 NTD data sets (7 chains ⩽2.5 Å resolution) and quantified relative displacements of the main chains (Materials and methods). While the cores of each structure could be fit to within 0.5 Å RMSD and rigid-body motions of the two lobes could be disqualified, isolated backbone segments in the LL showed displacements up to ∼10 Å (coloured red in Figure 3B). Those ‘flexible’ portions include αF, αG, αI plus attached loops: αF and αG showed rotational motions of up to 47° and 21°, respectively, and alternate loop conformations encompassed RMSDs of up to 3.6 Å, whereas segments in the UL were mostly invariant (RMSD 0.2–0.3 Å; Figure 3B; inset). Together, this analysis suggests dynamics for GluA2 LL segments, in addition to the interprotomer dynamic potential suggested above. To characterize further the hierarchy of motions intrinsically accessible to GluA2 and GluA3, we next performed an NMA of these structures. Two hierarchical scales of global oscillation in AMPAR NTDs—interlobe cleft closure and interprotomer counter-rotation The dynamic potential of AMPAR NTDs was analysed by NMA using the Gaussian Network Model (GNM; Bahar et al, 1997; Haliloglu et al, 1997) and the ANM (Atilgan et al, 2001). Motions (or global modes) near the native state, resulting from structure-encoded residue fluctuations, can be simulated with elastic network models (Materials and methods). Normal mode analysis performed on these elastic networks assesses the magnitude and direction of residue fluctuations and predicts the most probable collective motions. This approach allowed us to extract the global modes of motions robustly encoded by each biomolecular architectures (Bahar et al, 2010b). We focused on the most probable, top-ranking or (‘softest’) modes of motion (modes 1–3) accessible to AMPAR NTDs. The softest modes lie at the lowest frequency end of the mode spectrum; they are usually distinguished by their high degree of collectivity and provide insights into the cooperative mechanisms relevant to biological function (Bahar et al, 2010a, 2010b). Calculations were performed for both monomeric and dimeric arrangements of GluA2, GluA3 and mGluR1. The structural differences between GluA2 and GluA3 NTDs can be described at two hierarchical levels as a first approximation: interprotomer packing rearrangements and interlobe (intraprotomer) clamshell motions originating from the high mobility of the LLs. As shown below, the observed structures do not just represent static snapshots, but undergo cooperative fluctuations (or interconversions) between each other. The predisposition of the AMPAR NTDs to undergo these two levels of movements (intra- and interprotomer) renders them competent to transmit conformational changes to downstream portions of the receptor. Regarding intraprotomer motions, PBP-like opening/closing of the two clamshell lobes, which is well established for mGluR1 (Kunishima et al, 2000) and other PBPs (Quiocho and Ledvina, 1996), is also readily accessed by the GluA2- and GluA3 NTD monomers in slowmode 2, as shown for GluA3 in Figure 4A (and for GluA2 and mGluR1 in Supplementary Figure S5A). This common trait is robustly defined by the bilobate structure of each protomer. The occurrence probabilities (reflected by the reciprocal eigenvalues/frequencies associated with these eigenmodes; see Materials and methods) are comparable in the three GluRs, except for a higher predisposition of mGluR1 to undergo these motions. Figure 4.Global dynamics of GluA2 and GluA3 NTDs. (A) ANM predicts GluA2 and GluA3 NTDs to adopt classical clamshell motions. One of the dominant modes of motion predicted by ANM simulations for the GluA3 NTD monomer is a classical clamshell opening/closing with a large range of motion. The major deformation is an opening of the cleft (cleft opening angle shown as bars) along the ‘hinge’ axis; open (green) and closed (pink) states of GluA3 are shown. (B) Deformations within the dimer assembly are a bit different. The dominant mode is now an anti-correlated motion between the monomers along the axis denoted by the dashed red line, and manifests as a counter-rotation when viewed from the direction indicated by the black arrow. (C) Mobility profile of GluA2, GluA3 and GluK2 NTDs. Residue-specific fluctuations in mode 1 are shown for GluA2 (red), GluA3 (blue) and GluK2 (green; PDB 3H6G) NTDs. The correlation coefficients between the mobility distributions are as follows: 0.82 between GluA2 and GluA3, 0.86 between GluA2 and GluK2 and 0.94 between GluA3 and GluK2. Secondary structural segments that exhibit large fluctuations in the lower lobe (αF, αG and αI) are labelled. Upper and lower lobes are identified as brown and green bars below the x axis. (D) Dimer assemblies also show mobility. A GluA2 NTD dimer is shown with residues coloured by magnitude of the fluctuations from the first 10 modes of GNM, from least (blue) to most mobile (red). Note that the lower lobe is more mobile than the upper lobe, with the putative output region contacting the LBD exhibiting the most mobility. See also Supplementary Figure S5. Download figure Download PowerPoint In the dimeric context, a new interprotomer mode of motion dominates: an anticorrelated movement of the protomers, which is essentially imparted by the softest (i.e., the highest probability or lowest frequency) mode 1 in both GluA2 and GluA3. In this mode, the two protomers undergo almost rigid-body counter-rotations about the central axis (Figure 4B). It is this mode that allows GluA3 to sample the different quaternary conformations (dimers I, II and III), in which the protein was crystallized (Figure 2B), and allows mGluR1 to transition between its resting and active states (Kunishima et al, 2000). Interestingly, this mode also allows the dimer assemblies of both GluA2 and GluA3 to access the conformational landscape of mGluR1 (discussed below). Moreover, we find that the intraprotomer clamshell-type motions described for the monomers above (Figure 4A and Supplementary Figure S5A) are also accessible in the dimer, in mode 3. The associated eigenvalues are 0.47 (GluA2), 0.35 (GluA3) and 0.23 (mGluR1), revealing that GluA2 is the stiffest and mGluR1 the most mobile for this clamshell-like motion. Therefore, the NTD dimer maintains, and presumably exploits, the intrinsic propensities of the individual monomers. GluA3 can transit into mGluR1 conformations Motivated by the similarity of the intrinsic mobility encoded by the dimeric structures of GluA2 and GluA3, we examined if these motions could allow access to the mGluR1 apo form (PDB 1EWT). The apo form of mGluR1 features a large (∼70°) rotation about the dimeric interface and is the mGluR structure structurally most different to the iGluR NTDs (RMSD of ∼14 Å); furthermore, the apo form exhibits the functionally relevant dimeric rearrangem" @default.
• W2096164027 created "2016-06-24" @default.
• W2096164027 creator A5005390297 @default.
• W2096164027 creator A5010544263 @default.
• W2096164027 creator A5031167105 @default.
• W2096164027 creator A5034468058 @default.
• W2096164027 creator A5064978137 @default.
• W2096164027 creator A5081814496 @default.
• W2096164027 date "2011-02-11" @default.
• W2096164027 modified "2023-10-16" @default.
• W2096164027 title "Dynamics and allosteric potential of the AMPA receptor N-terminal domain" @default.
• W2096164027 cites W1481557920 @default.
• W2096164027 cites W1572640345 @default.
• W2096164027 cites W1656913045 @default.
• W2096164027 cites W1922786159 @default.
• W2096164027 cites W1945680072 @default.
• W2096164027 cites W1963733435 @default.
• W2096164027 cites W1967499250 @default.
• W2096164027 cites W1969298446 @default.
• W2096164027 cites W1974628023 @default.
• W2096164027 cites W1980970141 @default.
• W2096164027 cites W1983509986 @default.
• W2096164027 cites W1986191025 @default.
• W2096164027 cites W1986923087 @default.
• W2096164027 cites W1991588013 @default.
• W2096164027 cites W1994286350 @default.
• W2096164027 cites W2000300989 @default.
• W2096164027 cites W2000840258 @default.
• W2096164027 cites W2004360551 @default.
• W2096164027 cites W2004801897 @default.
• W2096164027 cites W2005913521 @default.
• W2096164027 cites W2006304581 @default.
• W2096164027 cites W2008093161 @default.
• W2096164027 cites W2011092574 @default.
• W2096164027 cites W2023343746 @default.
• W2096164027 cites W2024138709 @default.
• W2096164027 cites W2037852973 @default.
• W2096164027 cites W2038840577 @default.
• W2096164027 cites W2039593712 @default.
• W2096164027 cites W2041525670 @default.
• W2096164027 cites W2043263252 @default.
• W2096164027 cites W2051942403 @default.
• W2096164027 cites W2055956723 @default.
• W2096164027 cites W2058017150 @default.
• W2096164027 cites W2063987809 @default.
• W2096164027 cites W2066926304 @default.
• W2096164027 cites W2067763399 @default.
• W2096164027 cites W2069663555 @default.
• W2096164027 cites W2077202970 @default.
• W2096164027 cites W2077814219 @default.
• W2096164027 cites W2081659500 @default.
• W2096164027 cites W2087797354 @default.
• W2096164027 cites W2090166619 @default.
• W2096164027 cites W2101017457 @default.
• W2096164027 cites W2108700551 @default.
• W2096164027 cites W2109184569 @default.
• W2096164027 cites W2110893691 @default.
• W2096164027 cites W2111886211 @default.
• W2096164027 cites W2113670233 @default.
• W2096164027 cites W2113994268 @default.
• W2096164027 cites W2116498327 @default.
• W2096164027 cites W2121166097 @default.
• W2096164027 cites W2124134368 @default.
• W2096164027 cites W2130060890 @default.
• W2096164027 cites W2134216529 @default.
• W2096164027 cites W2137689860 @default.
• W2096164027 cites W2140899588 @default.
• W2096164027 cites W2141184730 @default.
• W2096164027 cites W2142595184 @default.
• W2096164027 cites W2144081223 @default.
• W2096164027 cites W2152193893 @default.
• W2096164027 cites W2161724422 @default.
• W2096164027 cites W2162327153 @default.
• W2096164027 cites W2163851282 @default.
• W2096164027 cites W2164739461 @default.
• W2096164027 cites W2166442349 @default.
• W2096164027 cites W4239748070 @default.
• W2096164027 doi "https://doi.org/10.1038/emboj.2011.17" @default.
• W2096164027 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/3049213" @default.
• W2096164027 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/21317871" @default.
• W2096164027 hasPublicationYear "2011" @default.
• W2096164027 type Work @default.
• W2096164027 sameAs 2096164027 @default.
• W2096164027 citedByCount "53" @default.
• W2096164027 countsByYear W20961640272012 @default.
• W2096164027 countsByYear W20961640272013 @default.
• W2096164027 countsByYear W20961640272014 @default.
• W2096164027 countsByYear W20961640272015 @default.
• W2096164027 countsByYear W20961640272016 @default.
• W2096164027 countsByYear W20961640272017 @default.
• W2096164027 countsByYear W20961640272018 @default.
• W2096164027 countsByYear W20961640272019 @default.
• W2096164027 countsByYear W20961640272020 @default.
• W2096164027 countsByYear W20961640272021 @default.
• W2096164027 countsByYear W20961640272022 @default.
• W2096164027 crossrefType "journal-article" @default.
• W2096164027 hasAuthorship W2096164027A5005390297 @default.
• W2096164027 hasAuthorship W2096164027A5010544263 @default.

• next