The unique distribution of neurotransmitter receptors and their subtypes within a single cell and throughout the brain requires highly selective intracellular targeting mechanisms. My laboratory studies the regulation of glutamate receptor trafficking and localization using a combination of biochemical and molecular techniques. Glutamate receptors are the major excitatory neurotransmitter receptors in the mammalian brain and are a diverse family with many different subtypes. The ionotropic glutamate receptors include AMPA, NMDA, and kainate receptor subtypes, each of which are formed from a variety of subunits. The metabotropic glutamate receptors (mGluR1-8) are G protein-coupled receptors (GPCRs), which are assembled as homodimers. We focus on defining subunit-specific mechanisms that regulate the synaptic localization and functional regulation of glutamate receptors. These mechanisms include posttranslational modifications such as phosphorylation and ubiquitination, as well as protein-protein interactions. A major focus of the lab is the study of the molecular mechanisms regulating the trafficking of NMDA receptors, which are multi-subunit complexes (GluN1;NR2A-D;NR3A-B). We have made significant progress in the detailed characterization of the synaptic expression of NMDARs and the role of NR2A and NR2B in receptor trafficking and synaptic expression. NMDA receptors are removed from synapses in an activity- and calcium-dependent manner, via casein kinase 2 (CK2) phosphorylation of the PDZ-ligand of the GluN2B subunit (S1480). We find that the NR2B subunit, and not NR2A, is specifically phosphorylated by CK2 and phosphorylation of NR2B increases in the second postnatal week and is important in the subunit switch (NR2B to NR2A), which takes place in many cortical regions during development and in response to activity. These data support unique contributions of the individual NMDA receptor subunits to NMDA receptor trafficking and localization. However, how synaptic activity drives this process remains unclear because CK2 is a constitutively active kinase, which is not directly regulated by calcium. We recently demonstrated that activated CaMKII couples GluN2B and CK2 to form a tri-molecular complex and increase CK2-mediated phosphorylation of GluN2B S1480. In addition, a GluN2B mutant, which contains an insert to mimic the GluN2A sequence and cannot bind to CaMKII, displays reduced S1480 phosphorylation and increased surface-expression. Importantly, we find that although disrupting GluN2B/CaMKII binding reduces synapse number, it increases synaptic-GluN2B content. Therefore, the GluN2B/CaMKII association controls synapse density and PSD composition in an activity-dependent manner, including recruitment of CK2 to remove GluN2B from synapses. Our studies have shown that a single point mutation in the GluN2B C-terminus (E1479Q) totally blocks CK2 phosphorylation of S1480 and results in significant increases in synaptic GluN2B. In collaboration with the NIMH Transgenic Core Facility we are currently generating a line of genetically-altered mice: a knock-in mouse expressing a point-mutated non-phosphorylatable GluN2B subunit (GluN2B E1479Q). This knock-in mouse will allow us to examine the precise regulation of GluN2B S1480 phosphorylation in neurons, in vivo, and without the requirement of exogenous protein overexpression. Because it is anticipated that these animals will show an impaired developmental GluN2 subunit switch (Sanz-Clemente et al, 2010), they will be a valuable tool for understanding how this process contributes to the refinement of neuronal connections. We have also investigated the role of posttranslational modifications, such as ubiquitination and phosphorylation, on AMPA receptor trafficking. We found that the first intracellular loop domain (Loop1) of GluA1, a previously overlooked region within AMPA receptors, is critical for receptor targeting to synapses, but not for delivery of receptors to the plasma membrane. We identified a CaMKII phosphorylation site (S567) in the GluA1 Loop1, which is phosphorylated in vitro and in vivo. Furthermore, we show that S567 is a key residue that regulates Loop1-mediated AMPA receptor trafficking, revealing a unique mechanism for targeting AMPA receptors to synapses to mediate synaptic transmission. Because this S567 is a relatively weak CaMKII substrate in contrast to its substrate residue in the GluA1 C-terminus (Ser831), and that the first half of the region is moderately conserved between subunits, we sought to identify other putative kinases. We performed a bioinformatics analysis of AMPARs and found that CK2 was a good candidate to phosphorylate the intracellular loop1 region of AMPAR subunits GluA1 and GluA2. Using in vitro kinase assays, we determined that CK2 phosphorylates the GluA1 and GluA2 intracellular loop1 region, but not their C-termini. Site-directed mutagenesis combined with an in vitro kinase assays revealed the presence of two CK2-phosphorylated serine residues in the GluA1 intracellular loop1 region, including S567 and a more robust substrate for CK2, S579. To investigate a role for CK2 in AMPAR trafficking, we reduced the endogenous expression of CK2 using an shRNA against the regulatory subunit CK2 beta, and we detected a reduction of GluA1 surface expression, whereas GluA2 was unchanged. Importantly, the expression of GluA1 phosphodeficient mutant (S579A) in hippocampal neurons displayed reduced surface expression. Therefore, our study identifies CK2 as a regulator of GluA1 surface expression by phosphorylating the intracellular loop1 region. Ubiquitination is a post-translational modification that dynamically regulates the synaptic expression of many proteins. However, very few of the ubiquitinating enzymes implicated in the process have been identified. In a screen to identify transmembrane RING domain-containing E3 ubiquitin ligases that regulate surface expression of AMPARs, we identified RNF167 and RNF112. Previously, we have demonstrated that RNF167 regulates excitatory synaptic transmission. Interestingly, we now find that RNF112 is a brain-specific functional GTPase, as well as E3 ligase. We have now named it neurolastin (RNF112/Znf179) because it is most closely related to the dynamin superfamily GTPase, atlastin. Neurolastin is the first identified protein with a unique domain organization harboring both GTPase and RING domains. We have demonstrated that neurolastin has the ability to hydrolyze GTP to mono-phosphate (GMP) and the GTPase activity is involved in the maintenance of dendritic spine density. In addition neurolastin leads to an increase in the density of dendritic spines on hippocampal neurons, whereas expression of the GTPase activity mutant did not affect the spine density. Subsequently, we also observed a significant decrease in the frequency of mEPSCs in hippocampal slices from knockout mice indicating a marked reduction in the number of functional synapses in the absence of neurolastin. These results indicate that neurolastin affects synaptic function by regulating synaptogenesis and spine maintenance.
Showing the most recent 10 out of 34 publications
