Synaptic trafficking of glutamate receptors by MAGUK scaffolding proteins

Synaptic transmission underlies every aspect of brain function. Excitatory synapses, which release the neurotransmitter glutamate, are the most numerous type of synapse in the brain. The trafficking of α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)-type glutamate receptors to and from these synapses controls the strength of excitatory synaptic transmission. However, the underlying mechanisms controlling this trafficking have remained elusive. Recent studies, drawing from advances in molecular biology and electrophysiology techniques, have established an essential role for a family of synaptic scaffolding molecules, known as membrane associate guanylate kinases (MAGUKs), in this trafficking process. These studies highlight the remarkable orchestration of AMPA-type glutamate receptor synaptic trafficking by multiple MAGUKs at different synapses within the same neuron and at different developmental stages.

Introduction

Human behavior and cognition are predicated on coherent communication between trillions of neurons assembled in brain circuits. Although the blueprint of the neural circuitry underlying simple conserved behaviors is encoded in our genes, a large measure of who we are is based on our ability to learn and remember new knowledge. How does the brain acquire new knowledge, and what are the molecular and cellular mechanisms underlying the storage of this knowledge?

The brain possesses a remarkable ability to respond to everyday experiences by modifying the strength of communication between neurons at individual synapses, a process thought to encode new knowledge about the world. Thus, recent efforts have been focused on elucidating the cellular and molecular mechanisms that determine the strength of synaptic transmission. In this regard, much attention has focused on understanding the biology of excitatory synapses in the mammalian hippocampus, a brain region long known to be necessary for the formation of new memories [1] [Box 1(a)]. At synapses between CA3 and CA1 hippocampal pyramidal neurons, presynaptic release of the neurotransmitter glutamate activates primarily two types of ionotropic receptors, AMPAR and N-methyl-D-aspartate (NMDAR) receptors [Box 1 (c,d)]. Whereas AMPARs carry most of the depolarizing current responsible for moment-to-moment, ‘basal’, synaptic transmission, calcium influx through NMDARs is required to trigger changes in synaptic transmission strength, a process known as ‘synaptic plasticity’. Increasing evidence suggests that dynamic regulation of the number of synaptic AMPARs underlies changes in synaptic transmission strength, which raises the fundamental question: what determines the number of glutamate receptors at the synapse?

Answering this question requires consideration of the synapse as a specialized cell junction for the transmission of information between neurons. Much like any other cell, therefore, a neuron must be able to solve the cell biological problem of spatially and temporally organizing the signal transduction machinery necessary for cell-to-cell communication. Although many solutions to this challenge have evolved, the role of protein scaffolding has emerged as a dominant mode of multiprotein complex assembling at excitatory synapses. Indeed, in addition to the high concentration of neurotransmitter receptors clustered on the postsynaptic side of the synapse, there are numerous scaffolding proteins that contain modular protein–protein interaction motifs capable of directly binding to short amino acid sequences present in interacting proteins. Through these modular interactions, scaffolding proteins are thought to tether receptors and intracellular signaling complexes, forming macromolecular protein complexes essential for synaptic transmission.

The prototypical scaffolding protein present at excitatory synapses is synapse-associated protein-90 (SAP-90)/postsynaptic density protein of 95 kDa (PSD-95), a member of a family of proteins collectively known as membrane associate guanylate kinases (MAGUKs). Early insights into the potential involvement of MAGUKs in determining the number of synaptic glutamate receptors came from overexpression experiments in heterologous systems, in which it was found that the PSD-95/Discs large/zona occludens-1 (PDZ) domains of PSD-95 bind to the C-terminal tails of NMDA receptor type-2 (NR2) subunits 2, 3 and cluster them on the membrane surface. This observation led to the proposal that PSD-95 might be involved in the synaptic localization of NMDARs [4]. Similar biochemical approaches have been used to identify interacting partners for other synaptic MAGUKs and infer the biological function of these scaffolding proteins from the constellation of possible protein–protein interactions 5, 6, 7, 8, 9, 10, 11. Here, we discuss recent findings using an alternative approach to examine whether MAGUKs have a role in determining the number of glutamate receptors at the synapse. This approach combines molecular biology to vary the expression levels of MAGUKs while monitoring the consequences of these manipulations on synaptic function using electrophysiology to measure synaptic transmission directly. This approach frames the question within the cell biological context of the neuron as a functional unit within a brain circuit, and thus provides insights from the unique functional perspective of how synapses transmit and encode information in the brain.