New directions in synaptic and network plasticity – a move away from NMDA receptor mediated plasticity

Abstract

In the field of synaptic transmission, the NMDA-preferring subtype of glutamate receptor is the fulcrum for the most widely studied forms of activity-dependent plasticity: NMDA receptor-dependent long lasting potentiation (LTP) and long lasting depression (LTD) of excitatory synaptic transmission. Although this form of plasticity is found at a large number of central synapses, it is most commonly studied at the Schaffer collateral–CA1 pyramidal cell synapse of the hippocampal formation. Indeed studies of plasticity at this synapse have dominated this field since its original description more than 20 years ago. A quick scan of the literature tells us that much has been learned about the mechanism(s) underlying this form of plasticity and its role in a wide number of developmental, physiological and pathophysiological processes. However, despite being the centre of intense research focus there remains little concensus regarding all of the nuts and bolts responsible for its induction and expression, indicating that much is still to be learned. In addition to these more widely studied forms of plasticity the last decade has seen a crescendo of research into novel mechanisms of plasticity whose mechanisms and properties markedly differ from the ‘classical’ NMDAR-dependent forms. Indeed it seems at times as if virtually every ion channel, synapse, cell or circuit is subject to some form of plastic change that can endure for extended periods of time. These forms of plasticity can arise via alterations in the basic properties of synaptic transmission, and can lower or raise the threshold for other forms of plasticity, alter cell intrinsic excitability, homeostatically control network function, and lead to structural and developmental maturation; they may underlie a number of aberrant pathophysiological conditions. Thus, the classic view of associative, Hebbian-types of plasticity as a central mechanism for learning and memory is slowly being augmented by myriad new and disparate mechanisms, which undoubtedly are also essential for a large number of physiological and pathophysiological processes. To highlight recent progress in this field, The Journal of Physiology organized a symposium entitled ‘New directions in synaptic and network plasticity – a move away from NMDA receptor mediated plasticity’ at the 2007 Society for Neuroscience Meeting in San Diego, California. This symposium brought together leading luminaries in this emerging field, all of whom work on different but exciting aspects of these new forms of plasticity. This issue of The Journal brings together Symposium Reports from each of these speakers. Bloodgood & Sabatini (2008) highlight their recent studies on the downstream consequences of glutamate receptor activation in single spines of hippocampal neurons. Using state of the art imaging techniques they illustrate that the amplitude of synaptic events and their related spine head Ca2+ transients are established by a complex signalling cascade involving the regulation of postsynaptic non-glutamate receptor ion channels. Specifically, voltage- and Ca2+-gated ion channels located within the spines themselves open in response to glutamate receptor signalling, which can then act as a feedback loop for self-regulation of subsequent synaptic signals. These data underscore an important observation that synaptic depolarization of an individual spine must be of a sufficient magnitude (tens of millivolts) to trigger opening of voltage-dependent Ca2+ channels. Importantly, the interplay between these postsynaptic voltage-gated channels and synaptic glutamate receptors offer a large number of potential sites for modulation of up- and downstream events. Traditionally, mechanisms of long lasting plasticity have been studied at glutamate receptor synapses made onto excitatory principal cells. However, three reviews in this issue discuss recent reports of plasticity within inhibitory interneuron circuits. Kullmann & Lamsa (2008) discuss recent data showing that glutamate receptor mediated synaptic events onto local circuit GABA-containing inhibitory interneurons also possess novel mechanism(s) of plasticity (see also Pelkey & McBain in this issue). Nugent & Kauer (2008) then report that the GABAergic synaptic output of inhibitory interneurons in the ventral tegmentum possesses a hitherto unexplored mechanism of plasticity under tight regulation by opioid receptor activation. Kullmann & Lamsa (2008) describe two forms of plasticity recently observed in distinct interneuron subtypes. One of these is a so-called ‘anti-Hebbian’ form of plasticity observed at GluR2-lacking, Ca2+ permeable AMPA receptors in a subset of hippocampal interneurons. Rapid synaptic signalling at glutamate synapses onto interneurons is often controlled by the presence of Ca2+-permeable AMPA receptors, and unlike glutamate receptors on principal neurons, which are typically Ca2+-impermeable, these AMPA receptors are highly rectifying and pass primarily inward current, due to a voltage-dependent block by intracellular polyamines. Consistent with the known properties of these receptors, this form of plasticity is blocked by postsynaptic depolarization but enhanced by hyperpolarization. These authors then discuss their data in the context of other forms of Hebbian plasticity known to exist at glutamate synapses made onto hippocampal interneurons. The ventral tegmentum area is a brain region intimately involved in reward behaviour and importantly is the major focus of research into mechanisms of addiction. Recent work from the Kauer lab and others has highlighted that some mechanisms of addiction are likely to be associated with novel forms of synaptic plasticity. Nugent & Kauer (2008) discuss their recent data illustrating a long-lasting plasticity of GABAergic synapses in the ventral tegmental area. Specifically, this new form of GABA synapse plasticity is heterosynaptic and is triggered not by activity at the GABA synapses onto dopamine neurons, but rather by activation of glutamatergic inputs onto the same cells. This excitatory input then triggers a retrograde transmitter (most likely via a nitric oxide signalling cascade) that ultimately leads to a strengthening of inhibitory input onto the very same cells. Importantly this form of plasticity can be prevented by a single in vivo administration of morphine 24 h earlier and suggests a model whereby prior opioid receptor activation interferes with nitric oxide signalling cascades to prohibit long-lasting plasticity of GABAergic transmission. The authors then discuss these observations in the context of other examples of plasticity of GABAergic synapses. Metabotropic glutamate receptors (mGluRs) have long been considered auto- or heteroreceptors that primarily regulate pre- and postsynaptic excitability. When localized to presynaptic terminals, activation of mGluRs typically reduce transmitter release probability. However, Pelkey & McBain (2008) highlight work at the mossy fibre–CA3 system describing a metaplastic role for a Group III mGluR, mGluR7b, in triggering bi-directional plasticity. Importantly mGluR7 is expressed preferentially on mossy fibre axon terminals apposed to interneurons and is largely absent at the larger mossy fibre boutons innervating principal cells of the CA3 hippocampus. The presence or absence of mGluR7 dictates whether synaptic strength will weaken or strengthen in response to high frequency afferent mossy fibre activity, respectively. In addition to their role in regulating neuronal pre- and postsynaptic excitability, mGluRs also have an impact on local and global protein synthesis. Thus, Group 1 mGluRs signalling via their associated Gq second messenger cascade can have a major influence over neuronal structure as well as regulating protein synthesis. Dolen & Bear (2008) discuss their recent observation that indicates a pivotal role for the Group 1 mGluR, mGluR5, in fragile X syndrome. Fragile X syndrome results from a mutation that leads to a transcriptional silencing of the FMR-1 gene, which encodes the fragile X mental retardation protein (FMRP). FMRP usually functions as a negative regulator of protein synthesis and can counter mGluR5's positive regulation of protein synthesis. Consequently, it is suggested that in the absence of sufficient FMRP, exaggerated protein synthesis triggered via mGluR'5 signalling can account for many of the features of fragile X syndrome. Of particular importance, a reduction of mGluR5 signalling reverses a number of CNS related fragile X phenotypes pointing to the possibility that intervention of mGluR activity may be a tenable therapeutic strategy for treatment of this debilitating disorder. As mentioned above, activity at a number of surface ionotropic and metabotropic receptors can have major implications for determining or influencing neuron structural development and plasticity. The so-called ‘synaptotrophic’ hypothesis suggests that synaptogenesis is an orderly series of hierarchical processes that is dynamic, and influenced by ongoing activity. Cline & Haas (2008) provide a scholarly re-evaluation of this hypothesis and highlight that a better understanding of the mechanisms underlying this complicated multifaceted interplay between activity and development is crucial for our appreciation of neuronal development. As this brief overview shows, the field of neuronal plasticity is a lively and engaging one. There is hardly a week that passes without some new aspect or mechanism of plasticity being postulated in our peer-reviewed journals. Progress is often marred by our inability to reach concensus on the underlying principles for these myriad processes, but this is ultimately the stimulus to delve deeper into each and every form of plasticity. Vive le difference!