Glutamatergic synapses are structurally and biochemically complex because of multiple plasticity processes: long-term potentiation, long-term depression, short-term potentiation and scaling

Abstract

Synapses are complex because they perform multiple functions, including at least six mechanistically different forms of plasticity. Here, I comment on recent developments regarding these processes. (i) Short-term potentiation (STP), a Hebbian process that requires small amounts of synaptic input, appears to make strong contributions to some forms of working memory. (ii) The rules for long-term potentiation (LTP) induction in CA3 have been clarified: induction does not depend obligatorily on backpropagating sodium spikes but, rather, on dendritic branch-specific N-methyl-d-aspartate (NMDA) spikes. (iii) Late LTP, a process that requires a dopamine signal (and is therefore neoHebbian), is mediated by trans-synaptic growth of the synapse, a growth that occurs about an hour after LTP induction. (iv) LTD processes are complex and include both homosynaptic and heterosynaptic forms. (v) Synaptic scaling produced by changes in activity levels are not primarily cell-autonomous, but rather depend on network activity. (vi) The evidence for distance-dependent scaling along the primary dendrite is firm, and a plausible structural-based mechanism is suggested.Ideas about the mechanisms of synaptic function need to take into consideration newly emerging data about synaptic structure. Recent super-resolution studies indicate that glutamatergic synapses are modular (module size 70-80 nm), as predicted by theoretical work. Modules are trans-synaptic structures and have high concentrations of postsynaptic density-95 (PSD-95) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor. These modules function as quasi-independent loci of AMPA-mediated transmission and may be independently modifiable, suggesting a new understanding of quantal transmission.This article is part of the themed issue 'Integrating Hebbian and homeostatic plasticity.'

Keywords: CaMKII; long-term depression; long-term potentiation; metaplasticity; postsynaptic density; synaptic scaling.

Figure 1.

Kinetics and localization of the AMPAR current during an mEPSC. (a) Model (red) accounts for the extremely rapid rise time of the mEPSC, as measured in the dendrite close to the site of generation. (b) Model shows that most AMPA channels opened during a mEPSC are within a hotspot (100 nm radius) that can be a small fraction of the total PSD size (pink); three examples illustrate variation in PSD size. Adapted from [67].

Figure 2.

Model of early and late phases of LTP (see figure 1 caption for definition of structural elements) at a partially silent synapse. In the early phase of LTP, the AMPA-silent module is made functional by a coupled process in which there is both a change in vesicle release mode (from kiss and run to full fusion mode) and the addition of a hyperslot containing many AMPA channels to the postsynaptic membrane. This mode change is represented by a change in the size of the fusion pore; the higher cleft concentration of glutamate achieved after full fusion enhanced the activation of local AMPA channels. The coupled presynaptic and postsynaptic changes convert an AMPA-silent module into a functional one and thereby increase the apparent quantal content. By a different vesicle-mediated process, LTP induces a transient elevation of extrasynaptic GluA1 (GluR1), leading to a transient increase in the number of filled GluR1+ slots and therefore to a transient increase in quantal size. In the slower phase of LTP, the synapse grows through a trans-synaptic process that adds an AMPA-silent module to the synapse. This model can also explain distance-dependent scaling and why residual AMPA currents are equal distally and proximally after GluA1 KO: it is assumed that GluA1 extrasynaptically is in a simple equilibrium with GluA1 bound to GluA1-binding slots (green) within the synapse. Because measurements show that extrasynaptic GluA1 is greater in distal dendrites than proximal dendrites [44,45], the density of slots occupied by GluA1 distally will be greater than proximally, thereby producing distance-dependent scaling of quantal amplitude. Note that slot proteins are assumed be grouped into hyperslots; the addition of a hyperslot after LTP induction would account for the observed sudden large increases in AMPA-mediated currents (adapted from [69]).

Figure 3.

Evidence for functional modules within the PSD. (a) Super-resolution studies show PSD-95 sub-clusters in spines. (b) Size distribution of sub-clusters (filled circle) and PSDs. Inset shows distribution of number of sub-clusters within the synapse (a,b are adapted with permission from [77]). (c) Model incorporating nascent zone conversion and growth during the synaptic plasticity associated with LTP (at time zero). Colours represent: axonal bouton (green), active zone (AZ; red), nascent zones (NZs; aqua), dense core vesicles (DCV; dark grey), docked vesicles (dark blue), and non-docked small synaptic vesicles (SSV) located within 94 nm of the presynaptic membrane (white) (adapted with permission from [79]).