DBP2: Regulation of binding to PSD-95 and its relation to AMPA receptor trafficking

Back to table

A. Collaborating Investigators: Mary Kennedy,1 Terry Sejnowski,2 Tom Bartol,2 I. Bahar,3 J Faeder3

B. Institutions: 1Caltech, 2Salk, 3Pitt

C. Funding Status of Project: Caltech Institutional Funds

D. Driving relationship between TR&D2 and DBP2: Electrophysiologically observed changes in synaptic strength that follow Ca2+ influx can have a variety of underlying biochemical mechanisms, and usually involve the concerted action of several partially redundant regulatory mechanisms.33,34 It is essential to fully define the biochemical mechanisms that produce long-term potentiation (LTP) and long-term depression (LTD) in order to fully understand brain function. This DBP will test and explore one possible underlying mechanism for regulation of the number of AMPA receptors (AMPARs) that are stabilized at the postsynaptic site. A well-supported hypothesis states that a 3-step mechanism is involved in the initial recruitment of AMPARs to cortical and hippocampal excitatory synapses: exocytosis at extra/perisynaptic sites, lateral diffusion to the synapse, and a subsequent rate-limiting diffusional trapping”.35 Steps 1 and 3 are tightly regulated by synaptic activity. We now have a reasonably good idea about the particular proteins that contribute to the regulation of these steps, but we don’t yet understand the precise dynamics that organize these steps into a well-regulated system that integrates the functions of synapses, neurons, and the brain.


Fig VII.2. Schematic representation of neuroligin, SynGAP, LRTTM and TARP, that bind to PSD-95 PDZ domains, and regulate AMPAR stabilization near the PSD.

Specifically, this DBP tests the hypothesis that trapping of AMPARs in the postsynaptic density (PSD) is regulated by competition for binding to the PDZ domains of the postsynaptic scaffold protein PSD-95. We will employ MCell36 and two sets of experiments to study how competition among four key PSD proteins plays a critical role in determining the number of AMPARs at a postsynaptic site. The four proteins are neuroligin (NLG1),37 transmembrane AMPAR-regulatory protein (TARP),38,39 leucine-rich-repeat transmembrane protein (LRRTM),40 and synaptic GTPase activating protein (synGAP)41,42 (Fig VII.2). Two of these, LRRTM and TARP, bind to AMPAR and are believed to participate in trapping of AMPARs in the PSD. The objectives also test the hypothesis, based on recent findings in the Kennedy lab, that the phosphorylation of synGAP by Ca2+/CaMKII alters the balance of competition among these proteins (see Fig I.3 in the Overall Section).43 A central consequence is that modulating the binding affinity of these proteins to PDZ domains contributes to activity-dependent changes in synaptic strength.

E. Resulting Innovations: This project will drive development of the spatial extensions to BioNetGen Language (BNGL)44 for MCell in TR&D2 in two ways: (i) the competitive interactions among the molecules for the PDZ domains results in a reaction mechanism with a high degree of combinatorial complexity and (ii) the trans-synaptic complexes between neurexin and NLG1 and LRRTMs45 (Fig VII.2) will be represented as structures that bridge between pre- and postsynaptic cells.

F. Methods and Procedures: Aim 1: We will measure the affinities of the carboxyl terminal tails of NLG1, TARP-ɣ8, and LRRTM2 for the three PDZ domains of PSD-95 by BIAcore surface plasmon resonance (SPR) methods. This objective will provide necessary parameters for the model proposed in Aim 2. Affinities for binding of synGAP to PDZ domains, before and after phosphorylation of synGAP by CaMKII, have already been measured in the Kennedy lab. Thus, methods are in place to carry out this objective. The interaction of PDZ domains with AMPAR, NLG1 and SynGAP C-terminal motif QTRV, as well as the neuroligin-neurexin interactions will be analyzed using the methods developed in TR&D1, toward characterizing the molecular basis of the competition of these proteins for binding to PSD-95.

Aim 2: We will use MCell to simulate diffusion of NLG1, TARPs, and LRRTMs in the membrane of postsynaptic spines, and their interactions with PSD-95 immobilized near the membrane within the PSD. Prior evidence indicates that synGAP may be immobilized in the PSD by associations in addition to its binding to the PDZ domains of PSD-95. Therefore, we will simulate diffusion of synGAP and its dynamic association with immobilized PSD-95 under three conditions; diffusing freely in the cytosol, tethered to the spine membrane, and tethered to the membrane within the volume of the PSD. We will assess the dynamics of the competition of these proteins for binding to PDZ domains of PSD-95. For example, we will measure the equilibrium mixtures of the four proteins in the PSD before and after phosphorylation of synGAP (which reduces its affinity for synGAP ~10 fold.) We will use the experimental measurements of binding affinities already obtained for synGAP and to be gathered in Aim 1, as parameters in the simulations. We will include in the simulations binding of LRRTMs and NLGs to presynaptic neurexins across the synaptic cleft (the affinity of which has been measured by other labs). One goal will be to establish a proof of principle that competition for binding to PDZ domains of PSD-95 is physiologically significant given the range of experimentally determined parameters and numbers of proteins. A second will be to measure how changes in the affinities for PDZ domains of PSD-95, for example reduced affinity of phosphorylated synGAP, can alter the complement of proteins immobilized in the PSD.

Aim 3: We will carry out experiments to measure changes in the ratios of PSD-95 to synGAP, NLG1, LRRTM1/2, and TARP-ɣ8 in PSDs and in PSD-95 complexes isolated from neuronal cultures before and after induction of chemLTP, a pharmacological treatment that mimics physiological induction of LTP. The amounts of the proteins will be measured by quantitative immunoblotting with specific antibodies. We will test the hypothesis that synGAP plays a central role in regulation of the composition of the PSD-95 complex by comparing these measured ratios among neurons cultured from synGAP homozygotes, synGAP heterozygotes, and wild type litter mates. These experiments will also be used as a first test of whether results of the simulations proposed in Aim 2 reflect an observable physiological process.


Copyright © 2018 National Center for Multiscale Modeling of Biological Systems. All Rights Reserved.