Our research efforts this past year were focused on two subprojects, elucidation of the mechanisms underlying postsynaptic receptor regulation and the characterization of presynaptic proteins and vesicle trafficking in sensory cells of the inner ear. Our first subproject involves a characterization of glutamate receptor synaptic organization and the regulation of synaptic glutamate receptor expression. This past year significant progress has been made in understanding the delivery of AMPA receptors to the synapse, in characterizing the assembly of AMPA receptors in hippocampal pyramidal cells, in understanding the intracellular organization of the NMDA receptor NR1 splice variants, in demonstrating distinct mechanisms for the expression of synaptic and nonsynaptic NMDA receptors, and in characterizing the endocytosis of NMDA receptors from the plasma membrane. For the AMPA receptor subunit GluR1, it has been shown previously that its C-terminus interacts with the PDZ domains of SAP97. We demonstrated that SAP97 interacts with GluR1 early in its biosynthetic pathway, and likely plays a minor role in synaptic stabilization. Little SAP97 is associated with synaptic GluR1. Therefore, our results are consistent with a model in which SAP97 interacts with newly synthesized GluR1 and acts as a chaperone in the delivery of GluR1 to the postsynaptic membrane. We have shown that pyramidal cells of the hippocampus contain two populations of AMPA receptors, those made up of GluR1 and GluR2 and those made up of GluR2 and GluR3, showing that the neuron is capable of regulating the assembly of two distinct complexes and targeting them to different cellular locations. To investigate in more detail the assembly mechanism, we have determined the fate of the remaining subunits after removing one of the key subunits, GluR2. If GluR1 and GluR3 cannot assemble with each other in pyramidal neurons, the remaining receptors, after removal of GluR2, would be homomeric GluR1 and GluR3. Homomeric AMPA receptors are functional. However, in the hippocampus of GluR2 deleted mice, we found that most of the GluR1 and GluR3 are now assembled together. Our results demonstrate the critical importance of GluR2 in establishing the correct number and composition of AMPA receptors in the hippocampus. A project on the NMDA receptor NR1 subunit, described in last year?s annual report, was completed this year. The NR1 subunit has four different C-termini dependent on alternative splicing. We identified an ER retention signal in one of the alternatively spliced cassettes, C1, which is present in the majority of NR1 subunits. Therefore, most of the NR1 is retained in the ER until it assembles with NR2 subunits. However, we also found that subunits containing C1 as well as the C2? cassette were capable of cell surface expression without assembly. Therefore, the C2? cassette can override the ER retention of C1. We demonstrated that a motif of C2?, a PDZ interacting domain on the C-terminus, is required. Thus, this motif, presumably through interaction with a PDZ protein, overrides the ER retention of C1 and mediates surface expression of the subunit. The functional significance of this mechanism is being investigated. We are also investigating whether a similar mechanism is involved in the ER retention of the NR2 subunits. Since the NMDA receptor performs a critical function at the synapse, it is important to understand how the number and composition of receptors at the synapse is regulated. In addition to being at the synapse, some NMDA receptors are extrasynaptic, especially in a developing neuron. We hypothesized that these two populations of receptors are regulated differently and would respond differently to changes in the total cellular amounts of NMDA receptor subunits expressed in the neuron. NMDA receptor subunits were over-expressed by transfection in cultured cerebellar granule cells and the number of functional NMDA receptors at synaptic and extrasynaptic sites was determined physiologically. Over-expression of NR2 subunits, but not NR1 (splice variants with either C2 or C2? cassettes), increased the total number of NMDA receptors present on the cell surface. This is consistent with our previous data showing a greater supply of NR1 subunits than NR2 and suggests that the availability of NR2 controls the total number of NMDA receptors produced. Over-expression of NR1 or NR2 subunits did not change the number of synaptic NMDA receptors. These results show that extrasynaptic receptors are influenced simply by the amount of NR2 subunits produced, but that synaptic receptors are controlled by additional mechanisms. A likely candidate for the synaptic control is a PDZ protein such as PSD-95 which interacts with the C-terminus of NR2. To address this, cells were transfected with NR2 from which the PDZ interacting domain was removed. This changed the number of extrasynaptic receptors, but not synaptic receptors, and shows that the PDZ interacting domain of NR2 is not required for extrasynaptic expression. While over-expression of wild type NR2 did not change the number of synaptic NMDA receptors, the transfected receptors were synthesized and reached the synapse, based on the pharmacological properties of the receptor. PDZ-deleted receptors did not reach the synapse. We also showed that NMDA receptors are endocytosed using a clathrin dependent mechanism. Although AMPA receptors were shown to cycle in and out of the membrane, it was generally thought that NMDA receptors were stable and did not do so. We identified a tyrosine-based motif that is important to the internalization of NR2B; such motifs facilitate endocytosis by binding to an AP-2 adaptor complex. We also found that interaction of NR2B with PSD95 through its C-terminal PDZ interaction domain stabilized the receptor complex on the cell surface. Therefore, our results suggest that the NMDA receptor is capable of internalization but their interaction with PSD95 and related molecules is responsible for stabilizing the receptor complex at the synapse. Our second subproject addressed proteins associated with neurotransmitter release and vesicle tracking in hair cells. Neurotransmitter release in hair cells differs from that which occurs at conventional synapses in a number of ways including the number of vesicles released, the sustainability of the release, and the presence of a presynaptic body thought to facilitate vesicle positioning on the plasma membrane. To identify novel proteins that may be involved in release of neurotransmitter from hair cells, we carried out a yeast two hybrid screen using the SNARE proteins, syntaxin 1, VAMP1 and SNAP-25 as baits. We have now characterized a protein, which interacts with syntaxin, that we call ocsyn. We find that ocsyn is concentrated in a novel structure in the apical part of inner hair cells. Using a series of markers to intracelluar organelles, we conclude that ocsyn is associated with the subapical compartment, a specialized form of recycling endosome found in polarized cells. This specialized organelle highlights the importance of vesicle trafficking in the apical region of hair cells. We conclude that ocsyn plays a role in organizing SNARE proteins on intracellular membranes in the apical region of hair cells.
Showing the most recent 10 out of 70 publications
