Plasticity-dependent, full detonation at hippocampal mossy fiber–CA3 pyramidal neuron synapses
- Institute of Science and Technology Austria, Austria
- Oregon Health and Science University, United States
Figures
Figure 1
Properties of single unitary EPSPs at the hippocampal mossy fiber–CA3 pyramidal neuron synapse.
(A, B) Infrared videomicrograph (A) and schematic illustration (B) showing a paired recording from a mossy fiber bouton (tight-seal, bouton-attached configuration) and a synaptically connected CA3 pyramidal neuron (whole-cell configuration). This recording configuration enabled the specific, noninvasive stimulation of a single presynaptic terminal. (C) Stimulation of a presynaptic terminal in the bouton-attached configuration. Top, short pulses of increasing amplitude (50–600 mV, 50-mV steps); center, presynaptic action currents; bottom, unitary EPSCs. Black, subthreshold; red, suprathreshold stimuli and corresponding responses. Note that the appearance of a presynaptic action current is tightly associated with the generation of a postsynaptic response in the CA3 pyramidal cell. (D) Traces of unitary EPSPs evoked by a single presynaptic stimulus to the mossy fiber bouton (top, 700 mV, 0.1 ms, same recording as in C). Ten consecutive EPSPs are shown superimposed (gray; 20 s repetition interval) overlayed with the average (black). The decay time course of the average EPSP was fit with a monoexponential function (red; decay τ = 166 ms). Note that the slow EPSP decay will promote temporal summation. (E–I) Summary graphs for the proportion of failures (E), latency (F), peak amplitude (G), 20–80% rise time (H), and decay time constant (I) of the average unitary EPSP across recordings (6 to 8 pairs). Circles indicate data from single experiments; bars indicate mean ± SEM.
Figure 2
Presynaptic facilitation contributes to 'conditional detonation' at the hippocampal mossy fiber–CA3 pyramidal neuron synapse.
(A) Pyramidal neuron EPSPs and action potentials evoked by a short train of stimuli delivered to the presynaptic bouton (top, three stimuli delivered at 50 Hz). Ten consecutive postsynaptic responses are shown superimposed (40 s repetition interval). Note that the first presynaptic stimulus was unable to discharge the postsynaptic neuron, whereas the third stimulus reliably initiated firing. (B) Probability of postsynaptic action potential initiation as a function of presynaptic stimulus number (50-Hz stimulation of bouton) for experiments like that shown in A (points connected by lines for clarity, eight pairs). (C) EPSCs recorded in the pyramidal neuron in response to 50-Hz stimulation of the presynaptic bouton, showing marked facilitation of transmitter release (five consecutive EPSCs shown superimposed, 20 s repetition interval, same recording as in A). (D, E) Summary graphs for absolute (D) and normalized EPSC peak amplitude (E) from experiments like that shown in C (13 pairs). Error bars indicate SEM.
Figure 3
PTP converts mossy fiber synapses from a subdetonation into a full detonation mode.
(A) EPSPs and action potentials in a pyramidal neuron evoked by stimulation of a presynaptic bouton (three stimuli delivered at 50 Hz) before (left, ‘control’) and after (right, ‘post-HFS’) a single high-frequency stimulation period (HFS; 100 stimuli delivered at 100 Hz; ten and nine consecutive traces shown superimposed for control and post-HFS, respectively, 20 s repetition interval). After HFS, single presynaptic stimuli were sufficient to discharge the postsynaptic neuron. Red traces were recorded during the 100 s time period defined as PTP (22–122 s after HFS). (B) Summary plot of normalized EPSP1 maximum slope versus experimental time for experiments like that shown in A. HFS produced an enhancement of EPSP slope which decayed back to baseline (gray, monoexponential curve, τ = 89 s). Inset: example traces of unitary EPSPs in control (black) and PTP (red, 22 s after HFS) periods from the experiment in A. Circles indicate points of maximum slope, dashed lines indicate the corresponding tangential lines (slope, 2.9 and 21.5 mV ms-1, respectively). (C) Mean EPSP1 maximum slope during control and PTP periods. Open circles connected by lines show data from individual experiments. Bars illustrate mean ± SEM. (D) Summary plot of action potential probability during the first stimulus versus experimental time. HFS produced an enhancement of detonation, which slowly decayed back to baseline (gray, monoexponential curve, τ = 67 s). Dashed horizontal lines in B and D indicate baseline values. (E) Probability of action potential initiation in the pyramidal neuron as a function of presynaptic stimulus number (50-Hz stimulation of bouton) for control (black) and PTP (red) periods (seven pairs in B–E). PTP significantly increased the probability of postsynaptic action potential initiation. Error bars indicate SEM. Time interval used for quantification of PTP effects in C and E is indicated by red horizontal bars in B and D.
Figure 4
Synaptic computations enabled by plasticity-dependent, full detonation in the hippocampal mossy fiber network.
(A) Synaptic computations in the dentate gyrus–CA3 cell network in a subdetonation regime. Action potential initiation in CA3 pyramidal cells requires spatial summation (e.g. activation of multiple granule cells; left) or temporal summation (e.g. repetitive activation of a single granule cell, right). In the spatial summation scenario, mossy fiber transmission will implement a logic ‘and’ operation. Specific wiring rules in the network could be generated by structural plasticity (Galimberti et al., 2006). Activation of multiple distal perforant path synapses may substitute for the activation of single proximal mossy fiber inputs. (B) Synaptic computations in the dentate gyrus–CA3 cell network in a full detonation regime (e.g. during PTP). Action potential initiation in CA3 pyramidal cells is possible after a single action potential in a single granule cell (left). Thus, plasticity-dependent detonation at mossy fiber synapses will enable a logic ‘or’ operation and allow a single highly specific cue to trigger the encoding, storage, or recall of complex information in CA3 pyramidal neurons (Quiroga et al., 2005; Wilson and McNaughton, 1993). Additionally, a single spike in a single granule cell may not only activate a single CA3 pyramidal cell, but rather an ensemble of ~15 CA3 cells. Thus, plasticity-dependent detonation at mossy fiber synapses will contribute to the generation of ensemble activity in the hippocampal network. GC, granule cell; CA3, cornu ammonis 3; black, inactive cells; red, active cells. Traces represent single action potentials and action potential trains, respectively.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Plasticity-dependent, full detonation at hippocampal mossy fiber–CA3 pyramidal neuron synapses
eLife 5:e17977.
https://doi.org/10.7554/eLife.17977
