Portal de Periódicos da CAPES

Excitement about Inhibitory Presynaptic Terminals

2015; Cell Press; Volume: 85; Issue: 6 Linguagem: Inglês

10.1016/j.neuron.2015.03.006

1097-4199

David Vandael, Claudia Espinoza, Péter Jónás,

Lipid Membrane Structure and Behavior

Based on extrapolation from excitatory synapses, it is often assumed that depletion of the releasable pool of synaptic vesicles is the main factor underlying depression at inhibitory synapses. In this issue of Neuron, using subcellular patch-clamp recording from inhibitory presynaptic terminals, Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar show that at Purkinje cell-deep cerebellar nuclei neuron synapses, changes in presynaptic action potential waveform substantially contribute to synaptic depression. Based on extrapolation from excitatory synapses, it is often assumed that depletion of the releasable pool of synaptic vesicles is the main factor underlying depression at inhibitory synapses. In this issue of Neuron, using subcellular patch-clamp recording from inhibitory presynaptic terminals, Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar show that at Purkinje cell-deep cerebellar nuclei neuron synapses, changes in presynaptic action potential waveform substantially contribute to synaptic depression. Short-term synaptic depression is of critical importance for the processing of information in neuronal networks. Synaptic depression may, for example, underlie low-pass filtering, habituation, removal of redundant correlations, and dynamic input compression (Abbott and Regehr, 2004Abbott L.F. Regehr W.G. Nature. 2004; 431: 796-803Crossref PubMed Scopus (1142) Google Scholar). Thus, it is important to completely understand the phenomenon of synaptic depression, including the underlying molecular and cellular mechanisms. In principle, several distinct mechanisms could contribute to depression, including failures in axonal action potential propagation, changes in presynaptic action potential waveform (Geiger and Jonas, 2000Geiger J.R.P. Jonas P. Neuron. 2000; 28: 927-939Abstract Full Text Full Text PDF PubMed Scopus (459) Google Scholar), inactivation of presynaptic Ca2+ channels (Xu and Wu, 2005Xu J. Wu L.G. Neuron. 2005; 46: 633-645Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar), depletion of the releasable pool of synaptic vesicles (Dobrunz and Stevens, 1997Dobrunz L.E. Stevens C.F. Neuron. 1997; 18: 995-1008Abstract Full Text Full Text PDF PubMed Scopus (924) Google Scholar, Schneggenburger et al., 1999Schneggenburger R. Meyer A.C. Neher E. Neuron. 1999; 23: 399-409Abstract Full Text Full Text PDF PubMed Scopus (456) Google Scholar), inhibition of release by presynaptic metabotropic receptors, and desensitization of postsynaptic ionotropic receptors (Figure 1A). It is generally thought that depletion of the vesicle pool is the major factor underlying synaptic depression. However, this view is mainly based on the extensive analysis of various excitatory synapses (Dobrunz and Stevens, 1997Dobrunz L.E. Stevens C.F. Neuron. 1997; 18: 995-1008Abstract Full Text Full Text PDF PubMed Scopus (924) Google Scholar, Schneggenburger et al., 1999Schneggenburger R. Meyer A.C. Neher E. Neuron. 1999; 23: 399-409Abstract Full Text Full Text PDF PubMed Scopus (456) Google Scholar). Whether it can be extrapolated to inhibitory synapses remains unclear. Early experimental analysis of inhibitory synaptic transmission in hippocampus and cerebellum eliminated, one by one, several candidate mechanisms of synaptic depression. For example, depression at hippocampal basket cell-granule cell synapses (the main output synapses of fast-spiking, parvalbumin-expressing interneurons; Hu et al., 2014Hu H. Gan J. Jonas P. Science. 2014; 345: 1255263Crossref PubMed Scopus (0) Google Scholar) and Purkinje cell-deep cerebellar nucleus (PC-DCN) synapses (the main output synapses of cerebellar Purkinje cells; Telgkamp et al., 2004Telgkamp P. Padgett D.E. Ledoux V.A. Woolley C.S. Raman I.M. Neuron. 2004; 41: 113-126Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar) is not primarily generated by pool depletion, since changes in release probability via variation in extracellular Ca2+ concentration only have subtle effects on the extent of depression (Kraushaar and Jonas, 2000Kraushaar U. Jonas P. J. Neurosci. 2000; 20: 5594-5607PubMed Google Scholar, Telgkamp et al., 2004Telgkamp P. Padgett D.E. Ledoux V.A. Woolley C.S. Raman I.M. Neuron. 2004; 41: 113-126Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar). Failure of action potential propagation does not seem to contribute, since the axons of inhibitory neurons propagate action potentials highly reliably (Khaliq and Raman, 2005Khaliq Z.M. Raman I.M. J. Neurosci. 2005; 25: 454-463Crossref PubMed Scopus (121) Google Scholar, Hu et al., 2014Hu H. Gan J. Jonas P. Science. 2014; 345: 1255263Crossref PubMed Scopus (0) Google Scholar). Inactivation of presynaptic Ca2+ channels also cannot be the main mechanism, because inhibitory synapses typically rely on P/Q-type channels for the initiation of transmitter release (Hu et al., 2014Hu H. Gan J. Jonas P. Science. 2014; 345: 1255263Crossref PubMed Scopus (0) Google Scholar), and these channels show only minimal inactivation or even facilitation during repetitive activation. Finally, desensitization of postsynaptic receptors cannot be involved, since depression primarily appears as a presynaptic change (Kraushaar and Jonas, 2000Kraushaar U. Jonas P. J. Neurosci. 2000; 20: 5594-5607PubMed Google Scholar). These results illustrate the sometimes frustrating nature of a rigorous scientific approach. We falsify several hypotheses underlying a phenomenon, until the remaining possibilities are almost impossible to test. Moving forward in this situation then requires new experimental strategies. This is exactly the approach taken by Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar. To address the mechanisms of synaptic depression at GABAergic synapses, they approached the problem in the most direct possible way: by recording from inhibitory terminals. To perform these experiments, they focused on the PC-DCN synapse. This experimental system seems highly suitable, since the presynaptic terminals are large enough to enable subcellular patch-clamp recording (∼3 μm). Furthermore, this synapse shows robust depression of inhibitory postsynaptic currents during repetitive stimulation (Telgkamp et al., 2004Telgkamp P. Padgett D.E. Ledoux V.A. Woolley C.S. Raman I.M. Neuron. 2004; 41: 113-126Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar). Using this system, they were able to perform simultaneous measurements of presynaptic voltage, presynaptic capacitance, and postsynaptic current. This almost sounds too nice to be true, but there is a price to pay: currently, the experimental strategy used is primarily applicable to inhibitory terminals under culture conditions, although the authors offer a couple of additional experiments in acute slices. Using this technique, Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar performed several “dream experiments” at an inhibitory synapse. First, they measured stimulus-evoked capacitance changes in presynaptic terminals. As the specific capacitance of biological membranes is constant (∼1 μF cm−2), these measurements provide a quantitative assay of the number of vesicles fusing with the plasma membrane. The main advantage of capacitance measurements is that they provide a perfectly linear assay of exocytosis, in contrast to postsynaptic currents, which may exhibit nonlinearities because of saturation and desensitization of transmitter receptors. Using this approach, they found that the size of the releasable pool was surprisingly large, while the release probability per vesicle was small. Consequently, there is only minimal contribution of pool depletion to synaptic depression. Imaging experiments with synaptopHluorins further corroborated this conclusion. Second, they directly measured the voltage changes in the presynaptic terminal preceding transmitter release, which were previously inaccessible to quantitative measurements. They showed that the action potential in the presynaptic terminal undergoes waveform changes during repetitive stimulation. These changes involve a reduction in peak amplitude and an increase in the half duration. Previous work at excitatory terminals showed that the relation between action potential waveform and presynaptic Ca2+ inflow can be complex. Reduction in action potential amplitude reduces the total presynaptic Ca2+ inflow, whereas action potential broadening enhances it (Bischofberger et al., 2002Bischofberger J. Geiger J.R.P. Jonas P. J. Neurosci. 2002; 22: 10593-10602PubMed Google Scholar; Figure 1B). In inhibitory terminals, the effects of amplitude reduction predominated, resulting in an attenuation of Ca2+ inflow and release probability. Hence, an activity-dependent reduction in the amplitude of the presynaptic action potential substantially contributed to synaptic depression. Interestingly, action potential waveform changes occurred specifically in presynaptic terminals, and not in adjacent axons. Thus, the mechanism is implemented in a presynaptic terminal-specific manner. Finally, they probed the mechanisms underlying the activity-dependent changes in presynaptic action potential amplitude. They found that the excitability of the inhibitory presynaptic terminals is low. The voltage-gated Na+ channel density was small, whereas the K+ channel density was large, resulting in a small Na+-to-K+ channel ratio. Pharmacological experiments further suggested that the presynaptic K+ channels were primarily assembled from subunits of the Kv1 family (Kv1.1 and Kv1.3). As Kv1 channels have a low activation threshold, they may be ideally suited to regulate the amplitude of the presynaptic action potential. Taken together, these results suggest a novel mechanism underlying synaptic depression: an activity-dependent reduction in the amplitude of the presynaptic action potential, caused by a low excitability of the presynaptic terminal. Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar also performed additional experiments at inhibitory synapses in the hippocampus, suggesting that the new mechanism may be generalizable to at least some cortical inhibitory synapses. More work on inhibitory terminals of different GABAergic interneuron subtypes will be required to explore how far the generalization can go. The results may trigger a déjà vu: the idea that changes in presynaptic action potential waveform may regulate synaptic strength was quite popular in the early scientific literature, especially for invertebrate synapses (Byrne and Kandel, 1996Byrne J.H. Kandel E.R. J. Neurosci. 1996; 16: 425-435Crossref PubMed Google Scholar). Furthermore, dynamic changes in the presynaptic action potential waveform were previously documented at excitatory presynaptic terminals (Geiger and Jonas, 2000Geiger J.R.P. Jonas P. Neuron. 2000; 28: 927-939Abstract Full Text Full Text PDF PubMed Scopus (459) Google Scholar). However, in these cases activity-dependent broadening of the presynaptic action potential enhanced presynaptic Ca2+ inflow, and thereby facilitated transmitter release (Byrne and Kandel, 1996Byrne J.H. Kandel E.R. J. Neurosci. 1996; 16: 425-435Crossref PubMed Google Scholar, Geiger and Jonas, 2000Geiger J.R.P. Jonas P. Neuron. 2000; 28: 927-939Abstract Full Text Full Text PDF PubMed Scopus (459) Google Scholar). In contrast, in inhibitory PC-DCN synapses, activity-dependent amplitude reduction of the presynaptic action potential reduced presynaptic Ca2+ inflow, and therefore depressed transmitter release (Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). At inhibitory PC-DCN synapses, depression is prevented by a large vesicular pool, but re-introduced at a level upstream of the release machinery. What could be the functional advantage of such a weird design? As Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar speculate, such an organization may convey several advantages. First, it may facilitate reliable synaptic transmission, preventing pool depletion during high-frequency activity. This will be relevant under physiological conditions, because Purkinje cells fire high-frequency trains of action potentials in vivo in a wide range of behavioral conditions (Yang and Lisberger, 2014Yang Y. Lisberger S.G. Nature. 2014; 510: 529-532Crossref PubMed Scopus (126) Google Scholar). Second, it may allow a flexible implementation of target cell-specific differences in synaptic dynamics, using different presynaptic Na+-to-K+ channel ratios at PC-DCN synapses versus, for example, synapses on vestibular nucleus neurons or other Purkinje cells (Kawaguchi and Sakaba, 2015Kawaguchi S.-y. Sakaba T. Neuron. 2015; 85 (this issue): 1273-1288Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar). Third, the large releasable pool may provide room for the regulation of inhibitory synaptic strength by neuromodulators or long-term plasticity, while synaptic depression would be maintained under all of these conditions (Hefft et al., 2002Hefft S. Kraushaar U. Geiger J.R.P. Jonas P. J. Physiol. 2002; 539: 201-208Crossref PubMed Scopus (67) Google Scholar). Short-term synaptic depression at excitatory synapses appears to serve several complex functions in neuronal networks (Abbott and Regehr, 2004Abbott L.F. Regehr W.G. Nature. 2004; 431: 796-803Crossref PubMed Scopus (1142) Google Scholar). But what could be the function of depression at inhibitory synapses in the network? In hippocampal microcircuits, neuromodulators, such as oxytocin or cholecystokinin, depolarize fast-spiking, parvalbumin-expressing GABAergic interneurons. Under these conditions, short-term depression will ensure the reciprocity of changes in tonic and phasic inhibition: while tonic inhibition is enhanced, phasic inhibition would be reduced (Owen et al., 2013Owen S.F. Tuncdemir S.N. Bader P.L. Tirko N.N. Fishell G. Tsien R.W. Nature. 2013; 500: 458-462Crossref PubMed Scopus (234) Google Scholar). Alterations in the temporal structure of inhibition will, in turn, enhance stimulus-evoked and reduce background activity in excitatory principal neurons. Thus, depression in inhibitory synapses may be critically important for setting the signal-to-noise ratio in neuronal networks (Owen et al., 2013Owen S.F. Tuncdemir S.N. Bader P.L. Tirko N.N. Fishell G. Tsien R.W. Nature. 2013; 500: 458-462Crossref PubMed Scopus (234) Google Scholar). Control of Inhibitory Synaptic Outputs by Low Excitability of Axon Terminals Revealed by Direct RecordingKawaguchi et al.NeuronFebruary 26, 2015In BriefBy directly recording from the axon and terminal of a cultured cerebellar Purkinje cell, Kawaguchi and Sakaba demonstrate that action potential amplitudes are attenuated around axon terminals upon high-frequency activation, leading to short-term depression. Full-Text PDF Open Archive