Silent synapse

Silent synapse refers to a synaptic contact between two neurons where a presynaptic action potential fails to evoke a detectable postsynaptic signal. A synapse can be presynaptically silent if the action potential invading the presynaptic bouton or terminal fails to evoke release of neurotransmitter. However, the term is more commonly used to describe a postsynaptically silent excitatory synapse, where glutamate release fails to elicit a detectable AMPA receptor-mediated response. At such synapses an NMDA receptor-mediated signal may nevertheless be detected if the postsynaptic membrane is depolarized sufficiently to relieve the Mg²⁺-mediated blockade of these receptors.

[edit] Presynaptically silent (‘mute’) synapse

At a presynaptically silent synapse the neurotransmitter release probability is so low that no response can be elicited via postsynaptic receptors. This situation probably occurs during development at many synapses, and was described at the neuromuscular junction long before NMDA-only synaptic transmission in the mammalian brain. Among evidence for presynaptically silent synapses is the finding that some terminals in the crayfish neuromuscular junction are devoid of presynaptic dense bodies (Jahromi and Atwood, 1974; Atwood and Wojtowicz, 1999). High-resolution optical methods have been used to resolve among multiple release sites within individual Drosophila neuromuscular junctions, where short-term facilitation is expressed in part by recruitment of previously silent sites (Peled and Isacoff, 2011). In the mammalian brain, some evidence exists that activity-dependent 'unsilencing' of glutamatergic synapses is mediated by changes in vesicle kinetics (Gasparini et al., 2000). Interestingly, a role for nicotinic receptors in unsilencing has been reported at both glutamatergic and cholinergic synapses (Maggi et al., 2003; Krishnaswamy and Cooper, 2009). Functionally mute synapses have also been reported at GABAergic synapses that appear to be tonically depressed by presynaptic CB1 receptors (Losonczy et al., 2004).

[edit] Experimental evidence for NMDA-only synapses

[edit] Coefficient of variation analysis

The first evidence that glutamatergic synapses in the hippocampus could signal exclusively via NMDA receptors came from analyzing the trial-to-trial amplitude fluctuations of excitatory postsynaptic currents (EPSCs) recorded in CA1 pyramidal neurons in acute rodent brain slices (Kullmann, 1994). The two components of excitatory transmission evoked by stimulating presynaptic axons (Schaffer collaterals of CA3 pyramidal neurons) were isolated by sequentially clamping the postsynaptic membrane potential at a negative value (around -70 mV) to ensure that NMDA receptors were blocked by Mg²⁺ ions, and then at a positive value (around +40 mV) in the presence of AMPA receptor blockers to reveal NMDA receptor-mediated signaling. If both AMPA and NMDA receptors were present at all synapses, the variability of each component of the postsynaptic signal, expressed as the coefficient of variation (CV), should be approximately equal. This is because CV (the ratio of standard deviation to mean amplitude) is mainly determined by trial-to-trial variability in the number of vesicles of glutamate released from the presynaptic terminal. Instead, CV was found to be consistently higher for the AMPA receptor-mediated component of EPSCs than for the NMDA receptor-mediated component.

Probabilistic transmission at most synapses can be described as a binomial process, where the average number of quanta of transmitter released (quantal content) is given by the product of \(n\ ,\) the number of release sites, and \(p\ ,\) the average probability at each site. The variance of such a process is equal to \(np(1-p)Q^2\ ,\) where \(Q\) is the quantal amplitude, while the mean is \(npQ\ .\) CV is thus \(\sqrt{(1-p)/np}\ .\) The larger CV of AMPA than NMDA receptor-mediated EPSCs can therefore be explained if \(n\) is effectively lower for AMPA than NMDA receptors. That is, postsynaptic AMPA receptors do not sense the release of glutamate from as many presynaptic sites as do NMDA receptors. Thus, a fraction of synapses appear to be ‘silent’ with respect to AMPA receptors.

[edit] Minimal stimulation

Figure 1: Discrepancy between failure rates of EPSCs mediated by AMPA and NMDA receptors. EPSCs were recorded in a rat dentate granule cell in response to minimal stimulation of presynaptic axons. Twenty successive responses are shown superimposed in each case. When holding the neuron at a negative potential to keep NMDA receptors blocked (left), AMPA receptor-mediated EPSCs were only observed in a small minority of trials. When the neuron was held at a positive potential to relieve the Mg²⁺ block of NMDA receptors (right), most trials resulted in an EPSC with slower kinetics typical of NMDA receptors. (Data from M.-Y. Min and D.M. Kullmann)

Subsequent studies provided more direct evidence for a discrepancy between AMPA and NMDA receptor-mediated signaling. By recording the response to glutamate released from only a few synapses activated by very weak electrical stimuli, the frequency of failures of transmission was compared at two different postsynaptic membrane potentials (Isaac et al., 1995; Liao et al., 1995). When the membrane was clamped at a negative potential a higher rate of failures was observed than when recording at a positive potential, again consistent with NMDA receptors sensing glutamate release at a larger number of synapses than AMPA receptors. Minimal stimulation has revealed silent synapses in numerous areas, including the developing thalamocortical projection (Isaac et al., 1997), rodent spinal cord (Bardoni et al., 1998; Li and Zhuo, 1998), frog optic tectum (Wu et al., 1996), and zebra finch cortex (Bottjer, 2005).

[edit] Paired recordings

It is difficult to be certain that minimal stimulation reliably activates the same axons during the course of an experiment. A definitive demonstration that silent synapses are not an artifact of stimulus drift came from paired recordings where the response to action potentials in a single presynaptic neuron was monitored (Montgomery et al., 2001). In this situation again, some functional connections were shown to be mediated exclusively by NMDA receptors: action potentials in the presynaptic neuron evoked a postsynaptic signal that could only be detected when the postsynaptic neuron was depolarized.

[edit] Mechanisms

Figure 2: Alternative models for silent synapses. The most widely accepted model to explain NMDA-only synaptic transmission is that these receptors are absent from the postsynaptic density (or non-functional) at a proportion of synapses (A). An alternative hypothesis to explain NMDA-only transmission is that glutamate is released in such a manner that it reaches too low a concentration to activate AMPA receptors (B). The higher affinity of NMDA receptors for glutamate would explain why these receptors are able to mediate transmission. A third hypothesis holds that, because of their high sensitivity, NMDA receptors are able to detect spillover of glutamate from neighboring synapses (C).

Several mechanisms have been proposed to account for silent synapses.

[edit] ‘Deaf’ synapse

The first proposed explanation for the discrepancy in AMPA and NMDA receptor-mediated signaling was that AMPA receptors were absent or non-functional at a subset of synapses. Such synapses can be considered ‘deaf’ with respect to glutamate release, but nevertheless functional when the Mg²⁺ block of NMDA receptors is relieved (Kullmann, 1994). This remains the dominant model for silent synapses to date. It is consistent with immuno-EM data showing that the density of AMPA receptors varies far more among synapses than that the density of NMDA receptors (Nusser et al., 1998).

Although there is strong evidence for dynamic AMPA receptor trafficking during development and activity-dependent plasticity, this only lends indirect support to the deaf synapse model. Some observations on the co-localization of AMPA receptors, and/or effects of competitive or non-competitive blockers of glutamate receptors imply that this model may not account for all silent synapses (Balland et al., 2008; Choi et al., 2000; Kullmann et al., 1996). Two alternative explanations for silent synapses rely on the fact that NMDA receptors have a much higher affinity for glutamate than AMPA receptors: when tested with different concentrations of glutamate at steady-state, NMDA receptors are approximately 100-fold more sensitive than AMPA receptors (Patneau and Mayer, 1990).

[edit] ‘Whispering’ synapse

An EPSC exclusively mediated by NMDA receptors could, in principle, result from a synaptic glutamate concentration transient that was insufficient to activate AMPA receptors. A possible drawback of this model is that the difference in sensitivity of NMDA and AMPA receptors at steady state does not necessarily predict how they respond to a brief transient of glutamate as would be expected to occur in the synaptic cleft (Kullmann, 1999). Indeed, the binding rates for the two receptors are quite similar, and the different sensitivities to steady state glutamate concentrations are more a reflection of the desensitization and unbinding rates. Nevertheless, if glutamate were released relatively slowly, for instance via a narrow pore linking the vesicle to the synaptic cleft, the glutamate concentration transient could still be sufficient to activate NMDA but not AMPA receptors (Choi et al. 2000; Renger et al., 2001).

[edit] Spillover

A further alternative explanation for silent synapses is that glutamate diffuses from neighboring synapses, and therefore reaches NMDA receptors as a relatively slow wave of neurotransmitter, insufficient to activate AMPA receptors (Asztely et al., 1997; Kullmann et al., 1996). Ultrastructural studies show that, although the average nearest-neighbor distance between synapses in CA1 pyramidal cells of the rat hippocampus is approximately 0.5 micrometer (Rusakov and Kullmann, 1998), there is considerable variability and some synapses can be even closer together, and often without astrocytic processes separating them. Most of these neighboring synapses are likely to be on spines belonging to different cells. Thus, a proportion of NMDA receptors may well ‘eavesdrop’ on other synapses. NMDA receptors mediating crosstalk may not even be exclusively synaptic, because these receptors also occur in the extrasynaptic membrane, albeit at a lower concentration. Indeed, some evidence exists that some NMDA receptors, especially those incorporating GluN2 subunits, are shared among synapses activated by non-overlapping populations of axons (Scimemi et al., 2004). In the cerebellar cortex glutamate spillover has been shown to be sensed not only by NMDA receptors but even, under certain conditions, by AMPA receptors (Carter and Regehr, 2000).

[edit] Implications for long-term potentiation

Figure 3: Unsilencing with LTP. Different mechanisms underlying silent synapses lead to different interpretations of the selective increase in quantal content of AMPA receptor-mediated EPSCs. Postsynaptically silent ‘deaf’ synapses may be unsilenced by exocytosis (A) or lateral translocation (B) of a cluster of AMPA receptors into the synapse. A ‘whispering’ synapse may be unsilenced by a change in the mode of exocytosis (C). If glutamate spillover occurs onto ‘eavesdropping’ NMDA receptors at a synapse with a low release probability, the release probability at this synapse may be increased (D). Growth of a postsynaptic spine towards a bouton may also unsilence a synapse by reducing the diffusional distance from the release site (E).

Much of the attention given to the silent synapse phenomenon stems from its consequences for understanding the early changes that occur during NMDA receptor-dependent long-term potentiation (LTP). LTP is expressed through an increase in the average quantal content detected by AMPA receptors, with a far smaller change in NMDA receptor-mediated signaling (Kullmann, 1994). According to the deaf synapse model, this is most simply explained by insertion of clusters of AMPA receptors at synapses that, under baseline conditions, were only equipped with NMDA receptors (Kullmann, 1994; Isaac et al., 1995; Liao et al., 1995). Such insertion could either occur via postsynaptic fusion of a vesicle containing AMPA receptors, or via lateral translocation of a cluster from a reserve of extrasynaptic AMPA receptors. However, alternative explanations for silent synapses prompt different explanations for the increase in AMPA receptor-mediated quantal content with LTP. If glutamate is released slowly via a fusion pore, then LTP may be associated with a change in the mode of release, by conversion to rapid exocytosis, which could activate both NMDA and AMPA receptors (Choi et al., 2000; Renger et al., 2001). If silent synapses represent inter-synaptic spillover, LTP might occur as a presynaptic increase in release probability at an ‘eavesdropping’ synapse (Kullmann et al., 1996). Both these explanations require a retrograde signal leading from postsynaptic NMDA receptor-dependent cascade (most likely consequent to Ca²⁺ influx). A final possibility is that a postsynaptic structural change occurs, such that the diffusional distance from presynaptic release site to the postsynaptic AMPA and NMDA receptors decreases.

[edit] Development of silent synapses

NMDA-only silent synapses appear to be more abundant early in development (Durand et al., 1996; Wu et al., 1996; Baba et al., 2000). This has been interpreted as reflecting maturation of synapses with sequential expression of NMDA, followed by AMPA, receptors, possibly through an LTP-like process. However, in the early post-natal rodent hippocampus, repetitive synaptic stimulation has been shown to cause a disappearance of AMPA receptor-mediated signals without a corresponding loss of NMDA receptor-mediated signaling, implying that active silencing occurs (Xiao et al., 2004). In contrast, in the spinal cord, repetitive stimulation while the postsynaptic neuron was held at a positive potential, led to the appearance of an AMPA receptor component, that is, unsilencing (Baba et al., 2000). The mechanisms underlying these changes, and their relationship to LTP and long-term depression, remain incompletely understood, as do their role in the normal maturation of excitatory synaptic transmission.

[edit] References

Asztely, F., Erdemli, G., and Kullmann, D. M. (1997). Extrasynaptic glutamate spillover in the hippocampus: dependence on temperature and the role of active glutamate uptake. Neuron 18: 281-93.
Atwood, H. L., and Wojtowicz, J. M. (1999). Silent Synapses in Neural Plasticity: Current Evidence. Learning & Memory 6: 542 -571.
Baba, H., Doubell, T. P., Moore, K. A., and Woolf, C. J. (2000). Silent NMDA receptor–mediated synapses are developmentally regulated in the dorsal horn of the rat spinal cord. J. Neurophysiol. 83: 955 -962.
Balland, B., Lachamp, P., Kessler, J., and Tell, F. (2008). Silent Synapses in Developing Rat Nucleus Tractus Solitarii Have AMPA Receptors. J. Neurosci. 28: 4624-4634.
Bardoni, R., Magherini, P. C., and MacDermott, A. B. (1998). NMDA EPSCs at Glutamatergic Synapses in the Spinal Cord Dorsal Horn of the Postnatal Rat. J. Neurosci. 18: 6558-6567.
Bottjer, S. W. (2005). Silent synapses in a thalamo-cortical circuit necessary for song learning in zebra finches. J. Neurophysiol 94, 3698-3707.
Carter, A. G., and Regehr, W. G. (2000). Prolonged Synaptic Currents and Glutamate Spillover at the Parallel Fiber to Stellate Cell Synapse. J. Neurosci. 20, 4423-4434.
Choi, S., Klingauf, J., and Tsien, R. W. (2000). Postfusional regulation of cleft glutamate concentration during LTP at 'silent synapses'. Nat. Neurosci 3: 330-336.
Durand, G. M., Kovalchuk, Y., and Konnerth, A. (1996). Long-term potentiation and functional synapse induction in developing hippocampus. Nature 381: 71-75.
Gasparini S., Saviane C., Voronin L.L., Cherubini E. (2000). Silent synapses in the developing hippocampus: lack of functional AMPA receptors or low probability of glutamate release? Proc Natl Acad Sci U S A. 97: 9741-6.
Isaac, J. T., Crair, M. C., Nicoll, R. A., and Malenka, R. C. (1997). Silent synapses during development of thalamocortical inputs. Neuron 18, 269-280.
Isaac, J. T., Nicoll, R. A., and Malenka, R. C. (1995). Evidence for silent synapses: implications for the expression of LTP. Neuron 15: 427-434.
Jahromi, S. S., and Atwood, H. L. (1974). Three-dimensional ultrastructure of the crayfish neuromuscular apparatus. J. Cell Biol. 63: 599-613.
Krishnaswamy, A., and Cooper, E. (2009). An activity-dependent retrograde signal induces the expression of the high-affinity choline transporter in cholinergic neurons. Neuron 61: 272-286.
Kullmann, D. M. (1994). Amplitude fluctuations of dual-component EPSCs in hippocampal pyramidal cells: implications for long-term potentiation. Neuron 12: 1111-20.
Kullmann, D. M. (1999). Excitatory synapses. Neither too loud nor too quiet. Nature 399: 111-2.
Kullmann, D. M., Erdemli, G., and Asztély, F. (1996). LTP of AMPA and NMDA receptor-mediated signals: evidence for presynaptic expression and extrasynaptic glutamate spill-over. Neuron 17: 461-74.
Li, P., and Zhuo, M. (1998). Silent glutamatergic synapses and nociception in mammalian spinal cord. Nature 393: 695-698.
Liao, D., Hessler, N. A., and Malinow, R. (1995). Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice. Nature 375: 400-404.
Losonczy, A., Biró, A. A., and Nusser, Z. (2004). Persistently active cannabinoid receptors mute a subpopulation of hippocampal interneurons. Proc. Natl. Acad. Sci. U.S.A 101: 1362-1367.
Maggi, L., Le Magueresse, C., Changeux, J., and Cherubini, E. (2003). Nicotine activates immature "silent" connections in the developing hippocampus. Proc. Natl. Acad. Sci. U.S.A 100: 2059-2064.
Montgomery, J. M., Pavlidis, P., and Madison, D. V. (2001). Pair recordings reveal all-silent synaptic connections and the postsynaptic expression of long-term potentiation. Neuron 29: 691-701.
Nusser, Z., Lujan, R., Laube, G., Roberts, J. D., Molnar, E., and Somogyi, P. (1998). Cell type and pathway dependence of synaptic AMPA receptor number and variability in the hippocampus. Neuron 21: 545-559.
Patneau, D.K. and Mayer M.L. (1990). Structure-activity relationships for amino acid transmitter candidates acting at N-methyl-D-aspartate and quisqualate receptors. J Neurosci. 10: 2385-99.
Peled E.S. and Isacoff E.Y. (2011). Optical quantal analysis of synaptic transmission in wild-type and rab3-mutant Drosophila motor axons. Nat Neurosci 14: 519-26.
Renger, J. J., Egles, C., and Liu, G. (2001). A developmental switch in neurotransmitter flux enhances synaptic efficacy by affecting AMPA receptor activation. Neuron 29: 469-484.
Rusakov, D. A., and Kullmann, D. M. (1998). Extrasynaptic glutamate diffusion in the hippocampus: ultrastructural constraints, uptake, and receptor activation. J Neurosci 18: 3158-70.
Scimemi, A., Fine, A., Kullmann, D. M., and Rusakov, D. A. (2004). NR2B-containing receptors mediate cross talk among hippocampal synapses. J Neurosci 24: 4767-77.
Xiao, M., Wasling, P., Hanse, E., and Gustafsson, B. (2004). Creation of AMPA-silent synapses in the neonatal hippocampus. Nat. Neurosci 7: 236-243.

Internal references

Harel Z. Shouval (2007) Models of synaptic plasticity. Scholarpedia, 2(7):1605.

[edit] Recommended reading

Atwood, H. L., and Wojtowicz, J. M. (1999). Silent Synapses in Neural Plasticity: Current Evidence. Learning & Memory 6: 542 -571.
Groc, L., Gustafsson, B., and Hanse, E. (2006). AMPA signalling in nascent glutamatergic synapses: there and not there! Trends Neurosci. 29: 132-139.
Kerchner GA, Nicoll RA. (2008). Silent synapses and the emergence of a postsynaptic mechanism for LTP. Nat Rev Neurosci. 9:813-25.
Kullmann, D.M. (2003). Silent synapses: what are they telling us about LTP? Philos Trans R Soc Lond B Biol Sci. 358:727-33.
Voronin, L. L., and Cherubini, E. (2004). 'Deaf, mute and whispering' silent synapses: their role in synaptic plasticity. J. Physiol. (Lond.) 557, 3-12.

[edit] External links

Author's laboratory