Insights into a molecular switch that gates sensory neuron synapses during habituation in Aplysia

Abstract

This review focuses on synaptic depression at sensory neuron-to-motor neuron synapses in the defensive withdrawal circuit of Aplysia as a model system for analysis of molecular mechanisms of sensory gating and habituation. We address the following topics: 1. Of various possible mechanisms that might underlie depression at these sensory neuron-to-motor neuron synapses in Aplysia, historically the most widely-accepted explanation has been depletion of the readily releasable pool of vesicles. Depletion is also believed to account for synaptic depression at long interstimulus intervals in a variety of other systems. 2. Multiple lines of evidence now indicate that vesicle depletion is not an important contributing mechanism to synaptic depression at Aplysia sensory neuron-to-motor neuron synapses. More generally, it appears that vesicle depletion does not contribute substantially to depression that occurs with those stimulus patterns that are typically used in studying behavioral habituation. 3. Recent evidence suggests that at these sensory neuron-to-motor neuron synapses in Aplysia, synaptic depression is mediated by an activity-dependent, but release-independent, switching of individual release sites to a silent state. This switching off of release sites is initiated by Ca2+ influx during individual action potentials. We discuss signaling proteins that may be regulated by Ca2+ during the silencing of release sites that underlies synaptic depression. 4. Bursts of 2-4 action potentials in presynaptic sensory neurons in Aplysia prevent the switching off of release sites via a mechanism called "burst-dependent protection" from synaptic depression. 5. This molecular switch may explain the sensory gating that allows animals to discriminate which stimuli are innocuous and appropriate to ignore and which stimuli are more important and should continue to elicit responses.

Figure 1

Strong and weak sensory neuron-to-motor neuron synaptic connections undergo synaptic depression with an identical time course, but differ in their paired-pulse ratios. Inset: The inverse relationship between paired-pulse ratio (PPR) and initial EPSP amplitude. Curve is hyperbolic function from Jiang and Abrams (1998), which was fit to empirical paired-pulse ratios for non-depressed sensory neuron synapses. The paired-pulse ratio was measured with a 50 millisecond interval between the two pulses. (PPR = amplitude of EPSP₂/ampltude of EPSP₁.) Note, that, in contrast to initially weak synapses, initially strong synapses show relatively little paired-pulse facilitation, suggesting that these stronger synapses have higher release site probabilities. A. Synaptic depression for strong and weak synapses. Synaptic connections are grouped according to initial EPSP amplitudes as either strong (> 8 mV) or weak (< 8 mV) (mean EPSP on trial #1: 22.3 ± 2.1 for strong synapses and 5.7 ± 0.3 mV for weak synapses). In each experiment, EPSP amplitude is normalized to the amplitude of the EPSP on trial #1. Mean amplitude of all the EPSPs on trial #12 was 36 ± 3% of the initial amplitude. (A 15 second ISI was used in all experiments measuring synaptic depression in this and subsequent figures.) B. Increasing release by broadening the sensory neuron action potential with the K⁺ channel blocker 3,4-DAP does not affect the rate of synaptic depression. Superfusing abdominal ganglia with 5 µM 3,4-DAP prior to and during experiments resulted in an approximately 3-fold increase in the duration of the sensory neuron action potential, and approximately a 50% increase in the amplitude of the EPSPs. (Reprinted from Gover et al., 2002.)

Figure 2

Synaptic depression is not accompanied by a substantial reduction in release probability, as indicated by the paired-pulse ratio. Paired-pulse ratios of large and small synapses tested once, either without depression (NON-DEPRESSED) or 15 seconds after a series of 15 single action potentials (DEPRESSED). The average initial EPSP amplitudes were 14.8 mV and 4.4 mV for the large and small synapses, respectively, in the nondepressed group; for synapses in the depressed group, the average EPSP amplitudes before depression were 16.7 mV and 5.7 mV for the large and small synapses, respectively. Both before and after depression, the large synapses (>8 mV) differed significantly from the small synapses (<8 mV) in their paired-pulse ratios. After synaptic depression the paired-pulse ratio for large synapses increased significantly; nevertheless, although the depressed EPSP amplitudes were approximately the same as the small synapses prior to depression, the paired-pulse ratios remained much smaller. In the case of the small synapses, despite comparable synaptic depression, there was not a significant change in the paired-pulse ratios. Thus, overall there was not a sufficient decrease in release probability to account for the synaptic depression. (Modified from Jiang and Abrams, 1998.)

Figure 3

Simulated synaptic depression as a result of vesicle depletion. The simulated synapses were representations of a synaptic connection between a single siphon sensory neuron and a single motor neuron, consisting of 40 release sites (active zones); each release site was functionally independent. In these Monte Carlo simulations, vesicle release at each site occurred in a stochastic manner. In A, strong and weak synapses differed in their initial number of readily releasable vesicles. In B, strong and weak synapses differed in the release probability of individual vesicles (P_ves). During each simulation, the per vesicle release probability and the number of active release sites remained constant. In both A, B, there was univesicular release, in which at a release site, after an initial vesicle release event, further release was inhibited for 5 msec. Similar results were obtained with limited multivesicular release, in which after an initial release event, the release probability of the release site was reduced by a factor of 0.66 and then recovered exponentially with a time constant of 3 msec. The per vesicle release probability was selected to achieve in A, a per site release probability of 0.9 had there been 20 releasable vesicles, and in B, a per site release probability of 0.25 and 0.75 for P_ves Low and P_ves High, respectively, with 6 releasable vesicles. Average initial numbers of quanta released were: for A, 11.4 for 3 vesicles and 32.6 for 15 vesicles; and for B, 11.0 for P_ves Low and 29.7 for P_ves High. Note that in A, strong synapses underwent depression at a slower rate than weak synapses, whereas in B, strong synapses underwent depression at a faster rate than weak synapses. In other simulation experiments, P_ves values were two- and four-fold smaller; with these lower P_ves values, strong and weak synapses still depressed at different rates, although with the smallest P_ves values tested, synaptic depression was extremely modest because minimal depletion occurred. (Reprinted from Gover et al., 2002.)

Figure 4

Simulated depression as a result of stimulus-dependent decrement in the number of active release sites. In these models, N_site decremented exponentially with each presynaptic action potential. During each simulation, the number of releasable vesicles and the per vesicle release probability remained constant. In A, strong and weak synapses differed in the number of releasable vesicles. In B, strong and weak synapses differed in the per vesicle release probability. (Reprinted from Gover et al., 2002.)

Figure 5

Limited recovery of depressed sensory neuron synaptic connections with rest. A. Persistent depression of sensory neuron-to-motor neuron synaptic connections 40 minutes after induction of synaptic depression. Sensory neurons were initially activated 15 times. After induction of synaptic depression (i.e. after the 15th stimulus, synapses were rested 40 minutes and again stimulated repetitively. Note, there was no apparent recovery of depressed sensory neuron connections. Note also that no additional depression was induced by the subsequent stimulation after rest. B. After depression develops at sensory neuron synapses, EPSPs show partial recovery after a 100 seconds period of rest. After the 15th stimulus, synapses were rested 100 seconds and then sensory neurons were stimulated once more. In both A and B, during the series of 15 stimuli, the ISI was 15 sec. (Reprinted from Gover et al., 2002.)

Figure 6

Ca²⁺ influx and PKC toggle the sensory neuron release sites between active and silent states, depending on the temporal pattern of Ca²⁺ influx. PKC serves as a sensitive detector of the presynaptic firing pattern. A: During a single action potential (trace in A2), Ca²⁺ influx initiates the switching off of release sites. This switching off changes the release state of synaptic vesicles (SV), shown as a dissociation of the previously docked SNARE proteins in A3. (Although the nature of the synaptic switch is not fully understood, the small G protein Arf appears to play a central role.) Note, in A2, Ca²⁺ binding to the C2 domain of PKC initiates the translocation of the kinase to the membrane, but PKC remains associated with the membrane only briefly (for approximately 200 milliseconds). B: Bursts of action potentials prevent the switching off of sensory neuron release sites. Ca²⁺ influx during the first action potential in the burst (B2) causes translocation of PKC, [shown with one Ca²⁺ ion bound to the C2 domain (red dot)]. As in A2, Ca²⁺ initiates the switching off of the release site. If a second action potential occurs within 200 milliseconds (B3), PKC is still associated with the plasma membrane and binds an additional Ca²⁺ ion (which is coordinated by the phospholipid and the C2 domain). When the additional Ca²⁺ ion binds, PKC is activated (shown by conformational change and yellow asterisk) and its association with the membrane is stabilized. Phosphorylation by PKC interrupts the silencing mechanism, via an unknown target protein. The silencing process must require at least a couple of hundred milliseconds to be completed or consolidated; within this time window, PKC activation can completely prevent the depression process. In B4, the synapse remains in the active state. Thus, in parallel, Ca²⁺ initiates the switching off of release sites, which is evident as synaptic depression, and the translocation and activation of PKC, which prevents synaptic depression.