Glutamate Receptor Interacting Proteins Enhance The Neurotoxicity Mediated By AMPA Receptors

Stroke, because of occlusion or rupture of an artery, is a leading cause ofdeath or infirmity in developed countries. As stroke, lack of blood flow to thebrain (ischemia) results in excessive release of the neurotransmitter glutamateinto the extracellular space. For over two decades, it has been known that theglutamate release activates ionotropic glutamate receptors, causing calciumentry, excitotoxicity, and neuronal death in vulnerable brain regions. Manyaspects are remained yet including the specific role of different types of Ca2+-permeable glutamate receptor in precipitating cell death and why neurons insome regions of the brain,such as hippocampal CA1 pyramidal neurons, striatalspiny neurons, and cortical projection neurons, are more vulnerable to glutamatemediated excitotoxicity. In the central nervous system (CNS), activation ofionotropic glutamate receptors allows a rapid influx of ions across the neuronalmembrane, mediating fast excitatory synaptic transmission. Ionotropicglutamate receptors are also critical for synaptic plasticity , learning andmemory. However, excessive activation of glutamate receptors in pathologicalconditions, such as anoxia or ischemia, precipitates excitotoxic neuronal death in vulnerable neuronal types by allowing an excessive influx of Ca2+, Na+, andZn2+ into neurons. There are three main classes of ionotropic glutamate receptor:NMDA, AMPA, and kainate receptors. Because all NMDA receptors arepermeable to calcium, it is focused on the role of these receptors in ischemicinjury in early studies. More recently, however, the hotspot is gradually shiftedto a critical role played by Ca2+-permeable AMPA receptors. AMPA receptors(AMPARs) are heteromeric assemblies of subunits GluR1– GluR4. AMPARslacking GluR2 subunits are permeable to Ca2+ and Zn2+. Ca2+-impermeability inGluR2-containing receptors is specified by the presence of an arginine (R)residue present in transmembrane domain 2 of this subunit instead of theglutamine residue (Q) in the GluR1, GluR3, and GluR4 subunits. Although thevast majority of GluR2 subunits in the brain contain arginine at this site, thisphenotype is not encoded at the genomic level but is the result of RNA editingof the GluR2 pre-mRNA in the nucleus by the enzyme adenosine deaminase 2(ADAR2), which catalyzes site-selective deamination of adenosine in thegenomic glutamine codon (CAG) to an inosine, giving an arginine codon (CIG)in the mRNA. This position is nearly 100% edited in the brain, so the vastmajority of GluR2-containing AMPARs in the CNS containing GluR2(R) areCa2+-impermeable. In addition, Q/R editing also influences subunit assembly,membrane trafficking, and synaptic targeting of GluR2-containing receptors.The observation that GluR2 expression is specifically attenuated in vulnerableCA1 pyramidal neurons after transient ischemia led to the''GluR2 hypothesis,''which proposed that decreased availability of Ca2+-impermeable GluR2(R)protein after a neurological insult would result in augmented formation of Ca2+-permeable AMPARs and consequently increased neurotoxic Ca2+ (and possiblyZn2+) influx. Remaining more controversial are the molecular mechanisms that underlie the postischemic downregulation of GluR2(R) protein and, the role ofGluR2 Q/R editing in this process. Peng promoted further understanding ofthe mechanisms that increase AMPAR Ca2+ permeability in vulnerable CA1hippocampal cells after transient forebrain ischemia. By combination ofelectrophysiology and single-cell RT-PCR, the authors correlated GluR2 Q/Rediting efficiency with the Ca2+ permeability of AMPARs in single neuron.Postischemic CA1 pyramidal cells exhibited high levels of heterogeneity in Q/Rediting (7%–98%) compared with sham controls and other less-vulnerablehippocampal cell types (CA3 pyramidal or dentate granule cells), which allexhibited over 95% editing efficiency, suggesting that GluR2Q/R editing inCA1 pyramidal neurons is particularly vulnerable to ischemic insult.Importantly, the impaired GluR2 Q/R editing correlated closely with the Ca2+-permeability of AMPAR channels (assessed from the reversal potential of theAMPAR current) in these cells.Failure to detect alterations in levels of GluR2protein, GluR2 Q/R editing in ischemia in early research may reflect the use ofwhole hippocampal tissue rather than RNA derived from singleelectrophysiologicallycharacterised cells.The postischemic reduction inADAR2-dependent GluR2 Q/R editing results in accumulation of uneditedGluR2 pre-mRNA in the nucleus, which downregulates both GluR2 mRNA andprotein levels, while the remaining GluR2(Q) proteinis preferentially assembledto produce Ca2+-permeable AMPARs that are exported to synapses. Furtherinvestigaton is needed to address these differences.The AMPA receptor subunits GluR2/3 can bind specifically to a kind ofPDZ proteins, termed as glutamate receptor-interacting protein (GRIP), AMPAreceptor-binding protein (ABP), and protein interacting with C kinase 1 (PICK1).Both GRIP1 and GRIP2 contain seven PDZ domains without other recognizable motif, ABP resembles GRIP in primary sequence; it differs from GRIP mostnotably in lacking the C-terminal seventh PDZ domain. GRIP1 also binds akinesin that functions in transport of dendritic vesicles along microtubules indendrites, showing that the scaffolding protein determines the direction oftransport .When the kinesin was bound to a member of the JIP group of scaffoldproteins, JSAP, it moved toward the axon, while when it was bound to GRIP1, itmoved toward the dendrite.GRIP1 can directly interact and steer kinesin heavychains to dendrites of neurons as a motor for AMPA receptors.So GRIP1 andGRIP2 may play important role on exporting GluR2/3 to synapses during anoxiaand ischemia.To clarify the mechanisms of the GRIP1 and GRIP2 for exportingGluR2/3 to synapses during anoxia and ischemia will facilitate to find effectiveantagonist to anoxia and ischemia.In this study, focal cerebral infarction model of adult rat were establishedbased on the principle of photochemical initiation of thrombosis, the rats weredivided into 5 groups according to the timecourse after cerebral local ischemia.Anti GRIP1 and GRIP2 antibodies were used via immunohistochemistry toanalysis the differences between the normal and the injury cerebral and thedifferences among every time point after ischemia.The results show that necrosishappened around the focus of injury region.And we can also find the GRIP1 andGRIP2 labeled neurons and immunoreactive nerve fibers in the ischemicpenumbra. Neurons in the cerebral of 6h and 1d groups are more heavily labeledthan others, cells labeled by GRIP1 and GRIP2 antibodys were coincide with theischemic penumbra. It suggests that the expression of GRIP1 and GRIP2increased after irradiation, GRIPs may be play a role on the mechanisms ofischemia injury. Because of the excitotoxicity of AMPA receptors and theinteraction between GRIP1/2 and AMPA receptors, the relationship between GRIP1/2 and AMPA receptors was further investigated next. Double-labelimmunocytochemistry was used demonstrating that the GluR2,GluR3 werecolocalized with GRIP1/2 in cerebral neuron, specially in the ischemicpenumbra and V pyramidal cells, and the colocalization happeded more at 6h and1d after the irradiation. Furthermore, using Western blotting, we study therelationship the levels of GRIP1 in ischemic penumbra and the levels ofmembrane GluR2/3, we found that the elevated levels of GRIP1 was coincidewith the accumulating of GluR2/3 after rat brain ischemia.When the levels ofGRIP1 were highest during 6h~12h after rat brain ischemia, the levels ofmembrane GluR2/3 reached the highest level suggesting that the accumulatingof membrane GluR2/3 may be induced by GRIPs. We demonstrated for the firsttime that ischemia can induce the expression of GRIP1 and GRIP2 on theoligendendrocytes in corpus callosum implying an important role of GRIP1 andGRIP2 in brain ischemia.