Tuesday, 11 December 2012


Structure and function of glutamate receptor amino terminal domains by Furukawa
Glutamate m ediated excitatory transmission is elicited by actions of metabotropic glutamate receptors (mGluRs) and ionotropic glutamate receptors (iGluRs) classified as GPCR and ligand-gated ion channels respectively. 4 subfamilies of iGluRs are AMPA receptors (GluA1-GluA4), kainate receptors (GluK1-Glu5) NMDA receptors (GluN1, GluN2A-D, GluN3A-B) and delta receptors (GluD1 and GluD2).  AMPa, kainate and NMDA receptors at synaptic and extrasynatpic sites determines amplitude and kinetics of excitatory postsynaptic currents.

All iGluR subunits are composed of 4 domains, ATD, ligand-binding domain (LBD) TM domain (TMD) and C-terminal domain (CTD). ATD has most divergent primary sequences.

iGluR LBDs have bi-lobed clamshell-like architecture.  Opening and closing of LBD clamshell structures are coupled to gating.  non-NMDA receptor and NMDA receptor LBDs form homodimers in crystals. A cluN1-GluN2 heterdimer has been observecd. Dimer interface regulates speed of deactivation and extent of desensitisation.

A Glu2A AMPAR crystal structure was completed.  2 conformers (A/C and B/D types) were present in 4 subunits with a crossover at ATD and LBD. This causes staggering of ATD and LBD dimers (A-B and C-D dimers at ATD and AD- and B-C dimers at LBD).

NMDA receptor ATDs are different from non-NMDA receptor ATDs. ATD is structurally most diverse region among iGluR subunits.

iGluR ATD is a site of interaction with EC proteins and cis or trans-synaptic proteins. The AMPA receptor ATD interacts with N-cadherin either cis or trans and promotes formation of dendritic spines.

Excitatory view of a receptor by Wollmuth and Traynelis
Sobolevsky's structure is AMPA rceptor from rat. It is made up of 4 GluA2 subunits identical in aa sequence. It compasses 3 structural/functional domains. 2 domains are on external side of cell membrane: modulatory amino-terminal domain (ATD) and the ligand-binding domain (LBD) with clamshell-like arrangement.  The 3rd is TM domain (TMD) which forms the ion channel. 

ATD and LBD are arranged as dimers, a key structural/functional motif for receptor function. The agonist(glutamate) recognition site is occupied by a competitive agonist.  It is in the clam shell formed by LBD,  The domains have 2fold symmetry relative to axis perpendicular to cell membrane. TMD has 4 fold symmetry.

Unexpectedly, domain swapping and crossover occurs between subunits.  Homotetrameric GluA2 has 2 conformationally distinct pairs of subunits, A/C and B/D.  At level of ATD, the dimer pairs are A-B and C-D. Inter-pair interactions between B and D.  At level of LBD, the dimer pairs are A-D and B-C, with inter-apir connections between A and C. This pairwise arrangement is abolished in TMD, with 4 independent but equivalent subunits have 4fold symmetry.

Core of GluR ion channel (TM helix1, M2 pore loop and TM helix M3) share structural similarity and homology with permeation pore in K+ channels.  Like K+ channels, they have an additional peripheral TM helix, the M4 segment that associates with ion channel core of adj subunit.

Conserved residues at apex of TM gating helix M3 like inner helices in K+ channel are in close proximity and form gate to block ion permeation through closed channel.

X-ray structure, symmetry and mechanism of an AMPA-subtype glutamate receptor by Sobolevsky
Architecture and symmetry
GluA2 receptor is shaped like capital Y. 3 major domains are arranged in layers. TMDs form ion channel and define the narrow base. ATDs are splayed outward at top of Y.  LBDs in complex with agonist are sandwiched between ion channel adn ATDs .

2 fold axis of molecular symmetry perpendicular to membrane plane. The ion channel domain has a 4fold axis of rotational symmetry.

Extracellular domains
ATD is implicated in receptor assembly, trafficking and localisation. It forms 2 distinct types of subunit-subunit contacts. Within each ATD dimer, extensive subuinit- subunit contacts (A-B or C-D). There is an interface between ATD dimers on overall axis of 2fold symmetry mediated by residues on L2 lobes of B and D subunits. In ATD layer, B/D and A/C subunits are proximal and distal to overall 2fold axis of symmetry.

At LBD layer each agonist binding domain is a partner in dimers. These dimers interact across 2fold axis. Due to subunit crossover between ATD and LBD layers, local LBD dimers re formed by A-D and B-C subunits. A and C subunits are proximal and B and D subunits are distal to overall 2fold axis. Within a LBD dimer multiple contacts between domain 1 of each subunit.  Lodged in clamshell of each LBd in bound antagonist. This proves that agonist/competitive antagonist bindings site is within and not between subunits

Transmembrane domain
4 GluA2 crys subunits arrange TM domains around an axis of 4 fold rotational symmetry.  Each subunit has 3 TM helices, M1, M3 and M4, a central porelike helix, M2, and a polypeptide pore-lining loop. M1 is on exterior of ion channel domain. Within pore lies M 2helix.  M3 helices line inside of ion channel domain. M4 is on exterior of ion channel domain, connected to S2 segment of LBD by 2 turns of helix and a short extended region of polypeptide. Extensive subunit-subunit interactions between TM segments with M4 segment of one subunit making interactions with TM domains of afj subunit.

Ion channel
In competitive antagonist-bound state, ion channel adopts closed conformation by crossing of M3 helices. Crossing of helices occurs near highly conserved SYTANLAAF motif.  Ala 622 in GluA2 subunit participates in close contacts with M3 helix of a neighbouring subunit. This indicates that introducing bulky residues can destabilise tight helix crossing associated with resting, closed state of receptor, leading to constitutively open ion channels. Ion channel permeation pathway is occluded above the SYTANLAAF motif by a pair of Met629 on A/C subunits protruding side chains towards centre of pore. When mets are mutated receptor desensitisation is perturbed.

Mechanism of activation
Antagonist binding stabilises binding domain clamshell in open conformation. Places TM -associated linker regions closest together. Binding ull agonists eg glutamate or ampa, results in closure of clamshell by movement of domain 2 closer to domain 1. Closure of clamshells pulls apart M3 helices at bundle crossing, opens ion channel. It is the fundamental conformation change that transmits energy assoc with agonist binding to work required to open ion channel.

Mechanism of desensitisation
AMPA and kainate receptors have rapid and profound desensitisation or ion channel closure after receptor activation. LBD D1-D1 interface ruptures. Entire binding domain subunits rotate. This allows D2 domains and linkers to ion channel to adpot a closed statelike conformation.  Rearranging D1-D1 LBD interface during desensitation demands movemtns of ATDs and ATD-LBD linkers.  Distance between and within ATD dimers may change providing mechanisms for binding of ions and small molecules to ATD of NMDA receptors can modulate receptor fuction. Ligands alter conformation of ATD clamshell and propagate conformational changes throughout receptor.

Treynalis 2010
AMPA receptors get pickled by Jackson and Nicoll
To mediate fast synaptic communication in brain, AMPA rceptors require TARP auxiliary proteins. AMPAR interact with TARPs. They regulate surface expresion and biophysical properties of AMPARs.  Schwenk et al described interaction between AMPARs and another family of TM proteins called cornichons. Cornichons influence intracellular traffifcking and gating activity of AMPARs.

Stargaxin, prototypical TARP is essential for surface expression of AMPARs and for targeting them to synapses in granule cells of cerebellum. Stargazin is called γ2. TARP family are widely expressed in CNS and intimately involved with AMPARs.

TARP proteins localise to synapses through motifs in Carboxy terminus that bind PDZ domain of scaffolding proteins eg PSD-95 in postsynaptic neurons. TARPS modulate AMPAR gating and pharmacology. They slow channel deactivation and desensitisation. They enhance single-channel conductance. They convert partial agonist kainate into full agonist. They cause competitive antagonist CNQX to act as aprtial agonist.

Schwenk detected CNIH-2 and CNIH-3. These members of mammalian CNIH family are homologous to cornichon proteins.

Schwenk posits that a small proportion (30%) of AMPARs associate with TARPs while remaining 70% form complexes with CNIHs.

Like TARPS, CNIHs are widely distributed in brain. Expressed in principal neurons, interneurons and glial cells in hippocampus, cerebellum and neocortex.  CNIHs are absent in berebellar granule cells where surface expression and synaptic targeting of AMPARs relies on γ2.

Like TARPs, CNIHs modulate AMPAR trafficking and slow deactivation and desensitisation kinetics of receptors. This enhances charge transfer associated with synaptic events.  Magniture of CNIH's effect on AMPAR kinectic outstrps that of γ2.

Beyond TARPs: the growing list of auxiliary AMPAR subunits by Guzman and Jonas
Gating of postsynaptic glutamate receptors of AMPARs determines time course of excitatory postsynaptic current (EPSC) in central neurons. If glu concentration transient in synaptic cleft is brief, the EPSC decayt approches deactivation time constant of receptors (time course of channel closure after agonist removal). If glu pulse in synpatic cleft is long, EPSC decay approaches desensitisation time constant( time course of chanel closure in maintained presence of glu). In many synapses EPSC decay timc course is intermediate.

Evidence suggests that auxiliary subunits fine-tune AMPAR gating. Stargazer mouse exhibits seizures and cerebellar ataxia. Loss of stargazin causes this phenotype/ γ-2, 3, 4, 5, 7 and 8 are auxiliary subunits of AMPARs. When coexpressed with principal subunits TARPs promote surface expression of AMPARs. They regulate AMPAR gating, prolonging deactivation and desensitisation of AMPARs in parallel. They affect pore properties, reducing sensitivity to intracellular polyamines and increasing single=channel conductnace.

TARPs are not only auxiliary AMPAR subunits. Cornichon=related proteins (CNIHs) are ocmponents of AMPAR protein microcomplex in brain.  They prolong seactivation and desensiation of AMPArs without noticeable effect on time course of receory from desensitisation.

Von Engelhardt identified another new protein assoc with AMPAR subunits, CKAMP44. Unlike TARPs and CNIHs, CKAMP44 has a single putative TM segment. It contains several cysteine residues presumable forming a cystine know similar to peptide toxins. When coexpressed with AMPAR subunits. CKAMP44 only minimally alters AMPAR surface expression. It affects deactivation and desensitisation time constant in a unique way, prolonging deactivation while accelerating desensitisation. Most modulators of AMPAR gating prolong both deactivation and desensitisation.  It slows recovery of AMPArs from recovery from desensitisation unlike TARPS (accelerate recovery) and CNIHs (no effect). CKAMP44 effects are opposite to TARPs and cornichons. It may be reciprocal regulation of AMPAR gating in the brain.

In hippocampus, strongest epxression of CKAMP44 in dentate gyrus granule cells.  TARP γ-2 is enriched in cerebellar granule cells and CNIHs is expessed through brain. Different expression pattern. FLAG-tagging suggest that CKAMP44 is concentrated at synapses.  This subcellular distribution is similar to that of TARPs but different from CNIHs which appear to be in extra-synaptic plasma membrane areas. It is possible that TARPs and CKAMP44 regulate properties of postsynaptic receptors. CNIHs may modify extrasynaptic receptors, making them more responsive to glutamate spillover.

Using CKAMP44 knockout and overexpression, von Engelhardt showed hat CKAMP44 prolongs decay time course of EPSC. It affects short term dynamics of excitatory synaptic transmission at medial and lateral perforant path synapses on hppocamapl granule cells shifting paired-pulse ration towards depression. Postsynaptic factors eg AMPAR desensitisation, contrbute to paired-pulse depression at these synapses.

A-to-I RNA Editing: EFfects on proteins key to neural excitability by Rosenthal and Seeburg
RNA editing by adenosine deamination diversifies proteome. Introduces codon changes and generate sturcturally and functionally diff isoforms of proteins.  Isoforms cannot be divined from genomic sequences.  Most editing sites occur in non-coding sequence.

RNA Editting in mammals: transmitter and voltage-gated ion channels 
AMPA receptors feature an edit critical for survival. 
AMPAR are Glu-activated cation channels. Mediate bulk of fast synaptic excitatory neurotransmision in mammalian/vertebrate brain. These receptors are assembled from subunits CluA1-4 encoded by 4 related genes into tetramers.  Primary transcripts of gene for GluA2 subunit undergo A-to-I editting at a CAG codon for gluatmine. This glu particupates in lining ion channel pore. It is conserved across subinits CluA1, 3 and 4. Only GluA2 carried edited codon CIG. GluA2 contributes an arginine instead of glutamine to channel lining. Arginine at this critical position renders channel impermeable to Ca2+. Decreases singe=channel conductance of activated ion channel compared to GluA2-less AMPAR.

Q/R site is towards 3'-end of Grai2 (gene encoding GluA2) exon 11. in primary transcrupts this region forms an imperfect ds structure with a short downstream sequence essential for Q/R site editting. A few hundred nt into intron 11. Cis-acting exon complementary sequences (ECS) have been found round many other edits in diverse species.

When one of the GluA2 alleles in mouse has its ECS deleted to prevent Q/R site editting of transcripts, severe epilepsy and premature death.
Figure 1. Edited Mammalian AMPA Receptors Are Impermeable to Ca2+Depicted are two versions of a heteromeric AMPA receptor, each showing two of the four subunits that make up a functional receptor. The transmembrane regions of the subunits are shown as cylinders, the re-entrant channel loop with the typical α-helical segment and the functionally critical Q/R site line the ion channel. Extracellular and intracellular subunit portions are sketched. The filled dot in the extracellular region of GluA2, between transmembrane segments 3 and 4, denotes the R/G edit (see text). The receptor version depicted on the left corresponds to the most prevalent AMPA channel in the brain, composed of the subunits GluA1 and GluA2, the latter edited in the Q/R site of the channel segment M2. The version on the right is the same channel except that the Q/R site of GluA2 is unedited, thus having the exonically encoded GluA2 sequence. This receptor is probably never expressed normally and can only be generated by gene manipulation. The characteristic property differences of the two AMPA receptor versions are listed below the channels, along with consequences on circuits and CNS disease for the unedited receptor. A role for the unedited form in sporadic ALS is presently an attractive hypothesis under debate.

Circuit changes in forebrain causing epilepsy may be related to elevated Ca2+ influx through receptors containing uneditted GluA2 subunits. It is possible the phenotype caused by mutation is due to greater tetramerisation and trafficking potential of Q/R site-unedited GluA2 subunits.

ADAR2 is responsible for Q/R site editting of GluA2 transcripts. Global knockout of ADAR2 causes early postnatal death of mice. This can be prevented by making mice homozygous for Gria2 alleles that carry a codon for arginine instead of glutamine for Q/R site.

Mammalian Kv1.1 editing fine tunes channel inactivation
For VGPC, timing is critical.  K+ channels are tetramers composed of 4 pore-forming α subunits. They are sometimes joined by accessory β subunits. After channel opens in response to depolarisation inactivation particle diffuses thru one of 4 large cytoplasmic portals and docks in a scpacious internal vestibule.n  It binds below the selectivity filter and may block ion flow, removing channel from equation.

After membran returns to rest, inactviation particle is free to unbind and return to cytoplasm.  After inactivation particle unbinds, channel passes thru open state. It continues to conduct ions before gate closes with normal deactivation rpcoess. This allows channel to be recruited into action during the next depolarisation.  Binding kinetics are determined by access to its receptor. Unbinding kinetics are determined by how tightly it binds.  Slow unbinding rates exaggerate AP's after hyperpolarisation phase due to transient passage through open state before closing. This limits repetitive firing.

In mammals, K+ channel fast inactivation is a target for regulation by mRNA editing. One edit change in mouse brain mRNA is Ile  to Val in codon 400, a highly conserved position in 6th TM along ion conductance pathway. Mutations reduced block by quarternary amines by 400fold.  Onset of inactivation was largely unchanged. Recovery from inactivation was 20 times faster. Inactivation particle's rate of release from its receptor may increase. I400V edit removes a methyl group.

Figure 2. Edited Mammalian Potassium Channels Recover More Quickly from InactivationFast inactivation in voltage-dependent potassium channels is caused by a tethered inactivation particle, which enters the channel's inner vestibule after opening and plugs the ion conduction pathway by binding to a receptor through a hydrophobic interaction. In the case of human Kv1.1, the inactivation particle is attached to a β subunit. RNA editing reduces the hydrophobicity of the inactivation particle's receptor, allowing the particle to unbind more rapidly. The dashed arrow indicates a slower rate. The channel's gate, in the open position, is shown in black
Holmgren substituted a cysteine at position 400.  They modified hydrophobicity or bulk at site. They showed that hydrophobicity at position 400 was principal determinant of recovery 

Developing a complete pharmacology for AMPA receptors: a perspective of subtype-specific ligands by Fleming and England
AMPAR ligand-binding sites
5 diff ligand-binding regions identified on AMPARs. Each subunit has an extracellular ATD and a LBD. 3 additional ligand-binding regions are at interface between subinits, between LBDs, between EC domain and TM esgments and within pore. 4 categories of ligands that target these sites have been developed: 1) competitive agonists and antagonists, 2) positive allosteric modulators, 3) negative allosteric modulators,  4) pore blockers.

Competitive agonists and antagonists control gating of the channel.  As individual AMPAR subunits each have a LBD, receptors can accommodate 4 ligands. Number and type of bound agonist influences conductance level of channel.

Positive allosteric modulators bind at interface between subnits with 1 or 2 molecules per interfce. They potentiate AMPAR currents by slowing desensitisation and/or deactivation of receptor.

Negative allosteric modulators bind at interface of adjacent subunits between the EC LBD and the channel TM domains. This inhibits conformational change that leads to channel opening.

Pore blockers bind pore of open receptors, inhibiting flow of ions through channel.  Each assembled receptor with only one pore binds a single blocker.

Subtype-selective ligands for AMPARs
Positive allosteric modulators
Add caption
Positive allosteric modulators, ampakines, potentiate AMPAR-mediated currents by attenuating receptor desensitisation and/or deactivation. Have been pursued for potential as cognition enhancing or nootropic drugs.  Eg benzamides, benzothiadiazines. Show low to moderate (<10fold br="br"> Cyclothiazide slows desensitisation. It is more efficacious at flip isoforms.

Aniracetam and PEPA slow deactivation potentiated flop over flip isoforms.

Negative allosteric modulators
Non-competitive antagonists of AMPARs.

Pore blockers
Polyamine toxins are selective for a subset of AMPARs, channels lacking GluA2 subunit. Mutagenesis studies suggest a single pore-lining residue causes selectivity among polyamine toxins for GluA2-lacking receptors. Edited GluA2 subinits carrt a positively charged Arginine at Q/R site which is thought to repel positively charged poylamines rather than uncharged glutamine (Q) in other subunits.

a, Domain organization of NMDA receptor subunits. Both NR1 and NR2 consist of N- (ATD) and C- (CTD) terminal domains, a transmembrane domain (TM) and a S1S2 ligand-binding core. The ligand-binding core can be isolated by tethering S1 and S2 with a Gly–Thr (GT) dipeptide linker. b, NMDA receptors form tetrameric channels comprising two copies each of NR1 and NR2. Shown here are the two possible modes of dimerization at the S1S2 ligand-binding cores, assuming that the subunits are organized in a NR1–NR1–NR2–NR2 arrangement25. Thick black lines between the S1S2 domains indicate the formation of a particular homo- or heterodimer interface. ATD and CTD have been omitted for clarity.
Subunit arrangement and function in NMDA receptors by Furukawa
NMDA receptors are heteromatic ion channels. For activation require binding of glycine and glutamate to NR1 and NR2 subunits respectively.  Slow channel opening and deactivation.  Influx of cations initiates signal transduction cascares crucial to learning and memory. NR2A-glutamate complex defines determinants of glutamate and NMDA recognition. NR1-NR2A heterdimer suggests mechanism for ligand-induced ion channel opening.  NR1-NR2A heterodimer is the functional unit in tetrameric NMDA receptors. Tyrosin 535 of NR1 in the subunit interface modulates rate of ion channel deactivation.

NMDA receptors are unusual ligand-gated ion channels. Activation requires not only binding of 2 agonists, glycine and glyutamate, but also relief of Mg2+ block by membrane depolarisation.  When NMDA receptors open, cation influx inc Ca2+. Ca2+ permeats through NMDA receptor ion channels and initiates signal transduction cascades which modulate synaptic strength.

Rates at which responses of NMDA receptors rise (activate) and decline (deactivate)on application and removal of agonists are slower than of nonNMDA receptor.  Slow deactivation rate governs duration of excitatory postsynaptic potential, a measure of strength og synaptic signalling.  Integration of chemical and electrical stimuli by NMDA receptors into a Ca2+ signal is crucial for activity dependent synaptic plasticity which inderpins learning and memonry.

NMDA receptor dysfucuntion is implicated in stroke, Parkinsons', Huntingdon's and schizophrenia.

NMDA assemble as heteromers of glycine-binding NR1, glutamate-binding NR2 and glycine-binding NR3 subunits. nonNMDA receptors eg AMPA and kainate receptors can form functional homotreameric channels activated only by glutamate. but NMDA receptors require assembly of 2 copies each of NR1 and NR2 and/or NR3.

On cell and developmental stage, 1 of the 4 NR2 subunits (A-D) combines with a splice variant of NR1 subunit, yeilding receptors with distinct deactivation kinetics.  NR1 apparent affinity for glycine depends on identity of cosasembled NR2 subunit. May be allosteric coupling beteen NR1 and NR2.  For a particular combination of NR1 and NR2 subunits show negative cooperativity, giving rise to a glycine-dependent for of receptor desensitisation.  Crystal structure of NMDA receptor was taken

Glutamate-binding site of NR2A
Comparing agonist binding site of NR2 with corresponding site of GluR2, difference is a negatively charged residue that participates in binding positively charged amino group of agonist.   In NR2a the residue is D731. In non-NMDA receptors it is a glutamate. As aspartate in NR2A is one methylene shorter than glutamate in non-NMDA receotirs, no salt bridge between D731 and amino group of agonist glutamate as there is in non-NMDA receptors.  In NR2a amino group of agonist forms whatermediated H bonds to E413 and Y761.  As NR2 have srhoter asprtate residue, NMDA can be modelled into agonist-binding opcket by displacing water molecule.

NR2a binding site is distrinct from non-NMDA receptors as there is a van der Waals contact between γ-carboxylate group of glutamate and Y730, a residue conserved among NR2 subunits.  This contact and interdomain N bond between Y730 and E414 may cause high-affinity binding of glutamate to NMDA receptors. Mutation of equivalent tyrosine in NR2B to alanin increases effector concentration for half-maximum response for glutamate 450 fold.

ab, Normalized traces of glutamate- (a) or glycine- (b) induced currents for NR1 Y535S, Y535W, Y535L and wild-type (WT) combined with wild-type NR2A. Insets show the fit of the wild-type current decays to a double exponential equation after a 3-ms application of ligand (traces are shown in black and equation fits in red; see Methods). The typical open-tip response is roughly 500 micros for a rise from 10 to 90%. cd, Time constant values (tau) and percentages of the fast components of deactivation for glutamate- and glycine-induced currents calculated from the double exponential fit. Bars show mean plusminus s.d. from ten different patches. Note that the y axis for tau is on a log scale. e, Proposed mechanism by which Y535 of the NR1 subunit (red) slows the deactivation of NMDA receptors. The aromatic side chain of Y535 binds to a primarily hydrophobic pocket at the hinge region of the NR2A subunit, stabilizing the activated, glutamate-bound conformation. A single heterodimer with S1S2 and transmembrane domains is shown for clarity.

Origin and evolution of synapses by Ryan and Grant
Schematic comparing the Drosophila melanogaster NR2 containing NMDA receptor (a) and the mouse NR2 containing NMDA (N-methyl-D-aspartate) receptors (b). The PDZ binding domains at the carboxy-terminus of both D. melanogaster NMDA receptor 2 (NR2; SVL) and mouse NR2 (ESDV) are indicated. Note that the mouse NR2 intracellular domain is five times larger than that of D. melanogaster. The diagrams compare the evolution of NMDA receptor signalling complexity in D. melanogaster and mouse showing protein–protein interactions at the NR2 carboxy-termini. The only established protein interaction site on theD. melanogaster NR2 carboxy-terminus is the interaction of Dlg (discs large homologue; SAP97 orthologue) (Bayes A. and S.G.N.G., unpublished data) through the PDZ binding domain. The vertebrate NR2B carboxy-terminus has numerous established primary and secondary interacting proteins (zoom out) and therefore a greater degree of NMDA receptor signalling complexity. Furthermore, the number of potential interactions of the NMDA receptor with MAGUK (membrane-associated guanylate kinase) components differ between protostome and deuterostome synapses. In the case of protostomes only one such interaction can occur, between NR2 and the protostome MAGUK, Dlg. Because of gene family expansion in chordates there are four available NR2 subunits (NR2A–NR2D) and four MAGUKs (PSD95, SAP102, PSD93 and SAP97). Three of the MAGUK paralogues can interact with any of the four NR2 subunits, making twelve potential deuterostome NR2–MAGUK interactions. As NMDA receptors are considered to be tetramers that contain two NR2 subunits, the existence of tri-heteromeric NMDA receptor channel further increases this combinatorial complexity

A frontier in the understanding of synaptic plasticity: solving the structure of postsynaptic density
The Postsynaptic Density (PSD) is a multi-protein complex. It positions signalling molecules for induction of long-term potentiaion (LTP) and depression (LTD) of synaptic strength. They are thought to underlie memory formation. A structural model of PSD is emerging.

The postsynaptic density (PSD) is a protein dense specialization attached to the postsynaptic membrane. PSDs were originally identified by electron microscopy as an electron-dense region at the membrane of a postsynaptic neuron. The PSD is in close apposition to the presynapticactive zone and ensures that receptors are in close proximity to presynaptic neurotransmitter release sites. PSDs vary in size and composition among brain regions and have been studied in great detail at glutamatergic synapses. Hundreds of proteins have been identified in the postsynaptic density including glutamate receptorsscaffold proteins, and many signaling molecules.

The PSD has been proposed to concentrate and organize neurotransmitter receptors in the synaptic cleft. The PSD also serves as a signaling apparatus. For instance kinases and phosphatases in the PSD are activated and released from the PSD to change the activity of proteins located in the spine or are transported to the nucleus to affect protein synthesis. Some of the features of the PSD are similar to the neuromuscular junctionand other cellular junctions, as the PSD has been modeled as a specialized cellular junction that allows for rapid, asymmetrical signaling.

Many proteins in the PSD are involved in the regulation of synaptic function. Key among these, are postsynaptic density-95 (PSD95), neuroligin (a cellular adhesion molecule), NMDA receptorsAMPA receptors, calcium/calmodulin-dependent protein kinase II and actin. As protein detection technologies have increased in sensitivity, such as with improvements in mass spectrometry techniques, more numerous proteins have been attributed to the PSD. Current estimates are greater than several hundred proteins are found at PSDs among brain regions and during different states of development and synaptic activity. PSDs also contain cell adhesion molecules and a diverse set of other signaling proteins. Many of the PSD proteins contain PDZ domains.

High frequency tetanic stimulation of Schaffer collateral axons projected by CA3 neurons causes increase in excitatory postsynaptic potential elicited in CA1 neurons with which they form synapse. This lasts for several hours and is called long term potentiation.

Low-frequency stimulation leads to long-lasting weakening of same synaptic population called long-term depression (LTD).

Together LTP and LTD at Schaffer collateral-CA1 synapses constitute prototypical form of synaptic plasticity.  Signalling in a proteinaceous specialisation of dendritic spine called PSD is central to induction of LTP and LTD.

EM of thinly sectioned hippocampal neurons indicates that a PSD at head of thin dendritic spine is disc-shaped. Surface area is 0.07 micrm2. Thickness is 25nm. PSD positions glutamate receptors across from presynaptic glutamate release sites. It links receptors to intracellular signalling cascades.  It is a locus for mutations causing neurological disease.

Mutations in PSD scaffold proteins Shank 2 and Shank3 are associated with Autism Spectrum Disorders.  Mutations in LG11 and ADAM22 are related to epilepsy.

Ca2+ entry into PSD through NMDA-type glutamate receptors is often required for LTP and LTD.  Trafficking and regulation of open probability/single-channel conductance of AMPARs is fundamental to changes in strength of many synapses.

During LTP, Ser831 of GluR1 AMPARs is phosphorylated by CaMKII and protein kinase C increaseing single-channel conductance. Ser831 phosophorylation is opposed by photphotases, thought to include protein phosphatase 1 acting downstream of Ca2+/CaM-activated protein phosphatase 2B (PP2B). PP2B also dephosphorylates cAMP-dependent protein kinase A phosphorylation site Ser 845 during LTD. This decreases open probability of AMPARs and leads to their trafficking out of PSD.

It is likely that spatio-temporal subtleties in Ca2+ signal determine direction of plasticity  Second messengers eg Ca2+ and cAMP are elevated in cellular microdomains. A second messenger-responsive enzyme will not be activated unless it is positionted in such a microdomain. Substrates must also be in proximity to the microdomain or they will not be acted on.  

3 factors that determine whether a second messenger-dependent signalling event falls in a microdomain are a) size of microdomain as determined by signal amplitude, eg high frequency tetanic stimulation enables maximal Ca2+ entry through NMDARs and thus activates signalling enzymes over a larger volume b) position of enzymes and substrates in relation to the second messenger generation/entry location eg anchoring protein AKAP79 positions PKA and PP2B for bidirectional phosphoregulation of AMPAR GluR1 residue Ser845. c) duration of elevation in second messenger concentration since architecture of signalling microdomain may be regulated by second messenger. 

Neuroligins and neurexins link synaptic function to cognitive disease by Thomas Südhof
Neurexins and neuroligins are synaptic cell-adhesion molecules that connect presynaptic and
postsynaptic neurons at synapses, mediate signalling across the synapse, and shape the properties of neural networks by specifying synaptic functions. In humans, alterations in genes encoding neurexins or neuroligins have recently been implicated in autism and other cognitive diseases, linking synaptic cell adhesion to cognition and its disorders.

Vertebrate neurexins (NRXNs) and neurpligins (NLGNs) are synaptic cell-adhesion molecules.  They probably bind to each other and interact with intracellular proteins (PDZ-domain proteins).  In mice lacking NRXNs and NLGNs deficit in synaptic transmission. 

Neurexins are polymorphic synaptic receptors
Venom of black widow spider contains α-latrotoxin. α-latrotoxin binds presynaptic receptors. Induces massive release of neurotransmitters.  NRXNs are receptors for α-latrotoxin. 

NRXNs are type I membrane proteins. 2 types: α- and β-latrotoxins. 

a, The structure of an excitatory synapse and the putative locations of neurexins (NRXNs) and neuroligins (NLGNs) in the synapse. A presynaptic varicosity containing synaptic vesicles is shown on the left, and a postsynaptic spine with a postsynaptic density containing neurotransmitter receptors on the right. b, The NRXN–NLGN junction, including selected presynaptic and postsynaptic binding proteins: CASK, VELIs and MINTs on the presynaptic side62; and PSD95 (which binds to AMPA-type glutamate receptors through its first PDZ domain, and to NLGNs through its third PDZ domain64), GKAP and SHANKs on the postsynaptic side. Note that a proportion of NRXNs and CASK could be also postsynaptic, and that SHANKs might also be presynaptic. C, carboxyl terminus; CAMK, Ca2+/calmodulin-dependent protein kinase domain of CASK; CHO, carbohydrate-attachment sequence; E, epidermal growth factor (EGF)-like domain; GUK, guanylate-kinase domain; L, LNS (laminin, NRXN, sex-hormone-binding globulin) domain; N, amino terminus; P, PDZ domain; S, SH3 domain. c, Alternative splicing of NRXNs and NLGNs. alpha-NRXNs contain five canonical splice sites (1 to 5), and beta-NRXNs contain two (4 and 5). Splice site 1 is C-terminal to the first EGF-like domain. Splice sites 2, 3 and 4 are at similar positions in the second, fourth and sixth LNS domains, respectively. Splice site 5 is between the glycosylated CHO sequence and the transmembrane region. Most alternative splicing involves the insertions of small evolutionarily conserved sequences, except for at splice site 5. Splicing at this site in NRXN2 involves a large insertion (191 residues), and in NRXN3 it involves at least 16 variants, some of which include stop codons and thus produce secreted isoforms of NRXN3 (ref. 35). NLGNs contain only two sites of alternative splicing, and one of these sites, site B, is present only in NLGN1.

No comments:

Post a Comment