• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. Nov 15, 2001; 537(Pt 1): 141–149.
PMCID: PMC2278933

Factors explaining heterogeneity in short-term synaptic dynamics of hippocampal glutamatergic synapses in the neonatal rat

Abstract

  1. Quantal release from single hippocampal glutamatergic (CA3-CA1) synapses was examined in the neonatal rat during a 10 impulse, 50 Hz stimulus train. These synapses contain a single release site only, thus allowing for an analysis of frequency facilitation/depression at the single release site level.
  2. These synapses displayed a considerable heterogeneity with respect to short-term synaptic dynamics, from a pronounced facilitation to a pronounced depression. Facilitation/depression was the same whether evaluated using the magnitude or the probability of occurrence of the postsynaptic response. This result suggests that postsynaptic factors, such as desensitisation, play little role.
  3. Release probabilities initially and late during the train were uncorrelated. Initially, release is determined by the number of immediately release-ready vesicles and by the probability of releasing such vesicles (Pves). Within the first five stimuli this vesicle pool is depleted. The deciding factor for release is thereafter the rate at which new vesicles can be recruited for release, rather than Pves.
  4. Heterogeneity in facilitation/depression among the synapses was strongly correlated with heterogeneity in initial Pves but not with that of the immediately release-ready vesicle pool. Thus, the main factors deciding short-term synaptic dynamics are heterogeneity in initial Pves and in vesicle recruitment rate among the synapses.

During a brief burst of repetitive activation the ability of a synapse to affect the postsynaptic cell changes, becoming either facilitated or depressed (Zucker, 1989; Thomson, 2000). Such short-term dynamics are of functional importance, allowing a synapse to act either as a low- or as a high-pass filter to incoming high-frequency activation. Such variation in facilitation/depression behaviour can also be seen among glutamatergic synapses within the same region such as the Schaffer collateral synapses on CA1 pyramidal neurones in the hippocampus (Dobrunz & Stevens, 1997). The question is, what may underlie this heterogeneity among the hippocampal synapses? Facilitation is commonly believed to result from an increased vesicle release probability, e.g. via ‘residual calcium’ (Katz & Miledi, 1968; Zucker, 1996; Neher, 1998). Depression, on the other hand, is attributed mainly to either vesicle depletion (Liley & North, 1953; Elmqvist & Quastel, 1965), or to a reduction in vesicle release probability (Betz, 1970; Hsu et al. 1996; Bellingham & Walmsley, 1999; Thomson & Bannister, 1999; Wu & Borst, 1999; Waldeck et al. 2000). In addition, postsynaptic receptor desensitisation may contribute to depression (Trussell et al. 1993).

It has been suggested that heterogeneity in initial release probability among Schaffer collateral synapses is decided by a size variation of the readily releasable pool of vesicles (Dobrunz & Stevens, 1997). However, it would seem that such a factor could not underlie much heterogeneity in facilitation/depression behaviour. On the face of it, a larger pool should allow for less initial depression by lasting longer. However, a larger pool would also increase release probability (Dobrunz & Stevens, 1997) thereby becoming depleted relatively faster. Moreover, facilitation is not accounted for. The question of what underlies heterogeneity in facilitation/depression among these synapses thus appears still unanswered.

In a recent study (Hanse & Gustafsson, 2001b), using neonatal rats, release from single Schaffer collateral synapses containing a single release site was examined. That study determined the relative contribution of vesicle release probability and vesicle pool size to the initial release probability. It also demonstrated that vesicle release probability is an important factor in deciding heterogeneity in release probability. Moreover, the immediately releasable pool was found to be substantially smaller than that estimated by Dobrunz & Stevens (1997). In the present report we have examined the facilitation/depression behaviour of such single release sites, in response to brief burst activation, to determine what shapes short-term synaptic dynamics at these sites.

METHODS

Slice preparation and electrophysiological recording

Hippocampal slices were prepared from 1- to 7-day-old Wistar rats. The rats were killed by decapitation in accordance with the guidelines of the local ethical committee for animal research. The brains were removed and placed in ice-cold solution composed of (mm): 124 NaCl, 3.0 KCl, 2 CaCl2, 6 MgCl2, 1.25 NaH2PO4, 26 NaHCO3 and 10 glucose. Transverse hippocampal slices (300 μm) were cut using a vibrating tissue slicer (Campden Instruments) and transferred to a holding chamber and stored at 28 °C. For recording, slices were individually transferred to a recording chamber where they were perfused (2 ml min−1) at 30–32 °C. The extracellular solution contained (mm): 124 NaCl, 3.0 KCl, 4 CaCl2, 4 MgCl2, 1.25 NaH2PO4, 26 NaHCO3, 10 glucose and 0.02 bicuculline methiodide, or 0.1 picrotoxin. The GABAA antagonists bicuculline and picrotoxin were used to examine glutamatergic transmission in isolation. The increased excitability in the preparation due to the presence of such drugs was counteracted by using higher than normal Ca2+ or Mg2+ concentrations, but keeping the ratio at unity. Nonetheless, these higher-than-normal concentrations are likely to cause somewhat higher-than-normal release probabilities, and consequently, less/ more frequency facilitation/depression (Huang et al. 1988).

Whole-cell patch-clamp recordings were performed from visually identified CA1 pyramidal cells. The pipette solution contained (mm): 95 caesium gluconate, 20 TEACl, 10 NaCl, 5 QX-314, 4 Mg-ATP, 0.4 Na-GTP, 0.2 EGTA and 10 Hepes (adjusted to pH 7.3 with CsOH). Some of the experiments were performed in the perforated patch-clamp mode. A glass pipette filled with extracellular solution was used for stimulation in the stratum radiatum. The stimulation pattern consisted of 10 impulse, 50 Hz trains evoked at 0.2 Hz.

Analysis and calculations of release probability

Procedures for whole-cell recording (minimal stimulation) tests for axonal excitation, and detection of EPSCs have been described previously (Hanse & Gustafsson, 2001a). Several findings suggested that the minimal stimulation protocol used consistently resulted in the activation of a single release site when applied to the neonatal hippocampal slice preparation (see also Hsia et al. 1998). Most importantly, the EPSC amplitude (excluding failures) was found to be independent of variations in release probability during the burst (10 impulse, 50 Hz) stimulation (Hanse & Gustafsson, 2001a). This result strongly argues against a release from more than one release site, as that would have resulted in larger EPSC amplitudes at positions of higher release probability during the burst (due to a higher proportion of multiple releases). Importantly, this result also argues against the release of more than one vesicle at a time from a single release site (Triller & Korn, 1982; Stevens & Wang, 1995; Dobrunz & Stevens, 1997; Liu et al. 1999; Matveev & Wang, 2000; Hanse & Gustafsson, 2001a). This is because AMPA receptors at these synapses do not appear to be saturated by the release of a single vesicle of glutamate (Liu et al. 1999; Umemiya et al. 1999; McAllister & Stevens, 2000; Hanse & Gustafsson, 2001a).

From these functional considerations we conclude that the synaptic currents evoked in the present study are unitary quantal responses, i.e. coming from a single release site releasing at most one vesicle at a time. Release probability (P) is thus directly obtained as 1 - failure rate (Hanse & Gustafsson, 2001b, see Fig. 1A). P at such a single release site should then rely on two factors: (i) the probability of releasing an individual release-ready (primed) vesicle (Pves), and (ii) the number of such vesicles (pool), such that P = 1 - (1 - Pves)pool (Hanse & Gustafsson, 2001b). Release probability to the first stimulus in the burst (P1) will then be determined by Pves1 and by the size of the preprimed pool (the number of vesicles that had entered the immediately releasable (primed) state during the 5 s interburst interval). The procedures for determining Pves1 and the size of the preprimed pool can be found in Hanse & Gustafsson (2001b). That paper describes the preprimed pool as small, variable from trial to trial and rapidly depleted such that replenishment of the pool starts within the course of the (10 impulse, 50 Hz) burst. Pves1 was determined from trials at which only one vesicle was released prior to replenishment. In other words, Pves1 was defined as the release probability of the release site when the preprimed pool contains one primed vesicle. In the equation P = 1 - (1 - Pves)pool, which in the present study is used to calculate Pves for later stimulus positions (see below), all primed vesicles are assumed to act independently with the same value of Pves. This notion is supported by the finding that release probability (P1) was about twice Pves1 when two primed vesicles were released (Hanse & Gustafsson, 2001). It was also tested by a Monte Carlo simulation of release from identical and independent vesicles (according to the equation P = 1 - (1 - Pves)pool) and a binomially distributed intertrial variation in the number of primed vesicles (resulting from a dynamic equilibrium between unprimed and primed vesicles as indicated in Hanse & Gustafsson, 2001b; see also Matveev & Wang, 2000). Such simulation produced a P2 value that was independent of whether there is release in position 1, or not. This is the experimental result we have previously reported (Hanse & Gustafsson, 2001b), an agreement consistent with the notion of independent and identical vesicles.

Figure 1
Heterogeneity in short-term synaptic dynamics

To estimate Pves for later stimulus positions in the burst two different methods were used to differentiate between releases from the preprimed pool and from vesicles recruited during the burst. A first method was to determine the stimulus position when the presence of a second release event has zero influence on the value of P1. Vesicles released at this, or later, stimulus position(s) could not have been present at the beginning of the train, i.e. they could not have belonged to the preprimed pool (Hanse & Gustafsson, 2001b). Release at positions prior to this position was then assumed to originate from the preprimed pool only. A second method was to use the average release probability curves of synapses that appear to lack a preprimed pool, i.e. the ‘low-frequency mute’ synapses (Hanse & Gustafsson, 2001b). It was assumed that replenishment has the same kinetics in mute and in non-mute synapses. The temporal characteristics of release of preprimed vesicles in non-mute synapses was then obtained in isolation by subtracting the release probability curves for these synapses with the average release probability curve from the mute ones (Fig. 3Ab). Before subtraction, the magnitude of the curve from mute synapses was adjusted to agree with the mean release probability at stimulus positions 8–10 (P8–10) of the release probability curve for any given non-mute synapse.

Figure 3
Heterogeneity in facilitation/depression and its relation to initial release probability

Both of these two descriptions of release from the preprimed pool were used to calculate Pves for a later position in the train (Fig. 5) using the general relation Pves(n) = 1 - (1 - Pn)1/pool(n). The size of the primed pool at that position (pool(n)) was obtained by subtracting what had been released prior to that position from the estimated value of the preprimed pool. As long as the pool at such a position, e.g. position 2, was > 1, Pves2 was calculated from 1 - (1 - P2)1/preprimedpool - P1, where P2 is the release probability at position 2. When the pool became less than 1, maximally one vesicle remains in the pool. The expression then simplifies to that of the release probability for that position divided by the pool value.

Figure 5
Activity-dependent adaptation of vesicle release probability

The analysis was performed using custom software written in Igor Pro (Wavemetrics, Lake Oswego, OR, USA). Unless otherwise indicated, data are presented as means ± s.e.m. Student's t test was used to determine statistical significance.

RESULTS

Facilitation/depression during burst activation

Minimal afferent stimulation in the stratum radiatum was used to elicit release from single synapses on CA1 pyramidal cells, these synapses containing a single release site (see Methods). Stimulation trials using 10 impulse, 50 Hz trains at 0.2 Hz produced a release of quantal EPSCs in the pyramidal cells, as illustrated from one experiment in Fig. 1A. When averaged for any given synapse over many trials (about a hundred), such stimulation produced an EPSC pattern that varied considerably among the synapses (n = 52). Figure 1B exemplifies with three different synapses the spectrum of EPSC changes during the train; large facilitation (a), essentially no change (b), and large depression (c). To quantify this behaviour, the last three EPSCs in the train were (for each synapse) averaged, and a ratio was computed between this value and the magnitude of the first EPSC in the train. Most of these ratios were below 1 (Fig. 1C), indicating that depression predominates among these synapses. In some of the synapses (n = 9), facilitation was infinite since no release in the first position was observed (low-frequency mute synapses; see Hanse & Gustafsson, 2001b). When plotted against age of the animal (postnatal days 1–7) the ratio did not correlate with age (r = 0.05; P < 0.05; n = 43; synapses with infinite ratio excluded). This result suggests that heterogeneity in facilitation/depression behaviour among the synapses develops early.

Depression is not explained by postsynaptic factors

Since single release sites were examined, releasing at most a single vesicle at a time, the influence of postsynaptic factors can be directly assessed. Such factors, e.g. desensitisation, will produce a discrepancy between changes occurring during the train in the averaged EPSC magnitude and in the release probability. For each synapse the magnitudes of the averaged EPSC and of the release probability for each position in the train was normalized with respect to those of the first position. Values for all synapses (except the low frequency mute ones) were thereafter averaged. Figure 2 demonstrates that there is no discrepancy between these average changes in release probability and EPSC magnitude. Thus, for this kind of burst activation of these synapses, postsynaptic factors play no role in the facilitation/ depression behaviour (see also Hjelmstad et al. 1999; Hanse & Gustafsson, 2001a).

Figure 2
Postsynaptic factors do not contribute to short-term synaptic depression

Facilitation/depression

Relation to initial release probability

Facilitation/ depression appearing during the burst stimulation is thus a consequence of changes in release probability alone. Initial release probability (P1) is decided by the number of vesicles available for release when the first action potential arrives (preprimed pool), and the release probability of the vesicles in that instant (Pves1), i.e. P1 = 1 - (1 - Pves1)poolsize (Hanse & Gustafsson, 2001b). However, the preprimed pool in these synapses is small and will be depleted within the first five stimuli of the 10 impulse train stimulation (Hanse & Gustafsson, 2001b, see also Fig. 4A). Release later in the train will then depend on vesicles recruited to a primed state during the course of activation, and the Pves of these vesicles. Initial and late release may then depend on different underlying factors, and are thus not necessarily correlated.

Figure 4
The initial vesicle release probability (Pves1), but not the size of the preprimed pool, determines facilitation/depression behaviour

The synapses (excluding the low-frequency mute synapses) were separated into two groups according to their P1 values. Figure 3Aa shows that these two populations, despite a near threefold difference in P1, had the same release probabilities from about the third/ fourth stimulus positions. Synapses with high P1 thus displayed large depression (filled circles), whereas those with low P1, on average, showed little change in release probability (open circles). To quantify late release the average release probability for positions 8–10 in the train (P8–10) were used. When plotting P8–10 against P1 no positive correlation was observed (r = −0.19; P < 0.05; n = 43, data not shown). Thus, factors that control the initial release do not appear to control the late release. The release probability curve averaged from the low-frequency mute synapses (n = 9) lacking initial release is shown in Fig. 3Ab.

As expected from the late/initial EPSC ratios in Fig. 1C, the P8–10/P1 ratio varied much among the synapses, from considerably more than one to almost zero, most values centred at 0.1–0.8. It displayed an overall inverse relation to P1 (Fig. 3B).

Dependence on vesicle release probability and preprimed pool

Heterogeneity in P1 among these synapses is explained both by a variation of the preprimed pool and of Pves1, albeit to a somewhat greater extent by the latter (Hanse & Gustafsson, 2001a). When plotting P8–10/P1 against these two factors, a clear difference emerged. Whereas the variation in P8–10/P1 is uncorrelated with the size of the preprimed pool (Fig. 4A), there is a rather strict inverse relation with Pves1 such that the greater the Pves1, the larger the depression (Fig. 4B). It would thus seem that Pves1 has a considerable influence on P1, but has little influence (in a positive direction) on late release. In fact, there was a significant (P < 0.001) negative correlation as shown by the plot of P8–10 against Pves1 (Fig. 4D). On the other hand, the size of the preprimed pool appears to influence the late release to about the same extent that it does the initial release. This influence can be seen in Fig. 4C where P8–10 is plotted against the preprimed pool size.

Late release

Release later in the train will depend on vesicles recruited to a primed state during the course of activation, and the Pves of these vesicles (see above). However, if the relation between Pves and recruitment is such that vesicles will be released faster than they are recruited, the exact value of Pves will be of little importance. Late release would then depend on recruitment only. The question of whether late release is rate limited by recruitment has been explored by estimating Pves for positions later than the first one and comparing those values with the actual late release probability.

Estimation of Pves for later positions

Pves for later positions in the train was calculated using two ‘extreme’ scenarios, as described in Methods. The first calculation assumes that there is no release of newly recruited vesicles. Such Pves determinations were only performed for positions 2–4 (for different synapses). The results are illustrated in Fig. 5A in such a way that synapses with similar Pves1 values have been grouped together. This graph suggests that the large heterogeneity in Pves at the first position is no longer present for the succeeding positions, Pves for all groups converging towards a value of about 0.4. To possibly overcome the problem of recruitment we made use of the ‘mute’ synapses demonstrating no release at the first (and sometimes the second) position in the train. Our previous analysis (Hanse & Gustafsson, 2001b) suggested that these synapses lack a preprimed pool, and the release at each position in the burst will then be from newly recruited vesicles. Assuming that such recruitment also takes place in synapses in which a preprimed pool is present, a release probability curve for preprimed pool vesicles only can be constructed, as described in Methods. Figure 5B shows that this method resulted in a very similar behaviour of Pves to that obtained above (Fig. 5A), albeit to a somewhat lower ‘adapted’Pves value, of about 0.3.

If we assume that Pves values estimated in this manner from release of preprimed vesicles also hold true for the newly primed ones two conclusions can be drawn. First that Pves cannot be responsible for any heterogeneity in late release since it becomes equalised among the synapses. Second that Pves is not a limiting factor for late release probability since its value of 0.3–0.4 exceeds that of the actual late release probability, which is generally below 0.2 (Fig. 4C and D), on average 0.13. In other words, there should on average only be about half a primed vesicle present at each stimulus event. Vesicles are then primed and released one by one limited by the recruitment rate. During the later part of the burst (between stimulus positions 5 and 10, Fig. 3A) release probability was rather stable indicating equilibrium between recruitment and release. Late release probability would then be a measure of the overall recruitment rate to the release site. The average overall recruitment rate during stimulation is then about 5 vesicles s−1 release site−1 (0.1 per stimulus, 50 Hz). The size of the preprimed pool may be assumed to be correlated with the number of vesicle docking/priming sites at each release site. A measure of the recruitment rate into each such site may then be estimated from the P8–10 value divided by the average preprimed pool. Figure 6A shows that such calculated recruitment rates varied considerably among the synapses. This heterogeneity was significantly correlated with that of Pves1, such that synapses with large Pves1 had lower recruitment rates (Fig. 6B).

Figure 6
Recruitment rate during steady state release

DISCUSSION

The present study has analysed facilitation/depression of synaptic release at Schaffer collateral synapses in the neonatal rat during a brief (0.2 s) high-frequency afferent activation. The synapses examined were single synapses containing a single release site that released at most one vesicle at a time, and the release-ready vesicles within a single release site seem to operate independently with the same release probability (Pves) (see Methods). The analysis suggests that the short-term synaptic dynamics in most of the synapses are determined by two factors alone: (i) the vesicle release probability value at the onset of the train stimulation and (ii) the rate of recruitment of new vesicles at the later part of the train. These two factors show a considerable variation among the synapses, and do also co-vary in such a manner that a high initial Pves is connected with a lower recruitment rate. In this manner there arises a considerable heterogeneity in short-term synaptic dynamics among these synapses.

The analysis also suggests an additional factor that causes heterogeneity in short-term synaptic dynamics in the synaptic population. In about one-sixth of the synapses there was no initial release and thus an infinite facilitation. These synapses may lack a preprimed pool (Hanse & Gustafsson, 2001b), and their large facilitation is thus a consequence of recruitment only. Since these ‘low-frequency mute synapses’ constitute more than a negligible part of the synaptic population, a shift of facilitation/depression behaviour observed in a population response may then rely on more than a change in Pves.

Present ideas regarding facilitation/depression in synaptic release during a brief high-frequency afferent activation are generally focused on activity-dependent changes in what determines release probability to the first stimulus in the train. That is, they are focused on changes in Pves and on the continuing depletion of the release-ready (prior to activation onset) pool of vesicles (Dobrunz & Stevens, 1997; Thomson, 2000). The present analysis, based on release from a single release site, suggests a somewhat different idea. The idea is the existence of two separate controls of release, an initial one and a later one. The scenario is as follows: at each release site there are several vesicle docking/priming sites, some of which will have become occupied by vesicles primed for release during a period of inactivity. The first stimulus of the afferent activation will release one of these vesicles with a probability determined by the number of such preprimed vesicles and the value of Pves1. These preprimed vesicles are few, on average one (Hanse & Gustafson, 2001B), and this immediately release-ready pool is depleted already within the first few stimuli. Release thereafter is therefore not dependent on the degree of depletion of such a pool. Instead, it depends on vesicles that become newly recruited to a primed state, within 100 ms after the initiation of the afferent activation. Such a rapid recruitment during action potential-evoked release, albeit with higher stimulation frequencies (300 versus 50 Hz), has also been described for calyx synapses (Wang & Zucker, 1998; Wu & Borst, 1999). These vesicles will be primed and released one by one since the estimated Pves during this later phase is about twice that of the actual release probability. The probability of late releases is then limited, and thus decided, by recruitment rather than by Pves. This means that facilitation/ depression behaviour is not a consequence of activity-dependent changes occurring in what determines release to the first stimulus in the train, i.e. in Pves and in the preprimed pool. Instead, this behaviour is decided by the relative efficacy of two different controls of release.

Our experimental data showed that the facilitation/ depression behaviour was well correlated with the value of Pves1, but not with the size of the preprimed pool. This can be explained within the context of the above scenario. Thus, whereas Pves will be an important factor deciding heterogeneity of initial release it will probably not be so with respect to the late release. This is because heterogeneity among the synapses in Pves seemed largely to vanish with successive stimuli in the train. Furthermore, the value of this ‘adapted’Pves seems large enough, compared to actual release probability, for Pves not to be an important limiting factor for late release. On the other hand, the fact that facilitation/depression behaviour is not correlated with the size of the preprimed pool suggests that this pool must be of about equal importance for both initial and late release probability. For the initial release the size of the preprimed pool is a direct determinant of release probability. For the late release it seems plausible that the (average) size of the preprimed pool is a measure of the number of sites to which vesicles can be docked and primed, and become recruited to. Thus, it appears that the size of the immediately releasable pool is of no consequence for short-term synaptic dynamics. An alternative view with respect to late release is that it is produced by a subpopulation of preprimed vesicles that have a very low Pves, for example, due to these vesicles being located away from the most active regions of calcium influx. These vesicles will then not participate in initial release but will be released later on due to, for example, residual calcium increasing their Pves. However, as discussed previously (Hanse & Gustafson, 2001b) our analysis of mute synapses does not suggest that their lack of initial release is due to a population of vesicles with very low Pves1. Rather, considering the trial-to-trial variation of the preprimed pool, this pool seems more likely to be a subpopulation of a docked pool of vesicles in dynamic equilibrium between a primed and a non-primed state.

As reasoned above, the (average) size of the preprimed pool may be related to the number of recruitment sites within a release site. A measure of the recruitment rate to each single vesicle docking/priming site is then obtained by dividing the release probability in the later part of the train by that pool size. This measure of recruitment rate was found to vary considerably among the synapses. Interestingly it correlated well in a negative manner with Pves1. Since Pves1 and preprimed pool size are uncorrelated (Hanse & Gustafsson, 2001b), this negative correlation will accentuate the depression during burst stimulation in synapses with a large Pves1. This negative correlation can be seen to be functionally meaningful in the sense that it accentuates the phasic contra-tonic release behaviour of a synapse. Whether such a negative correlation can be mechanistically explained by a common factor is, however, not clear (see below).

What determines Pves?

The present analysis has resulted in a number of suggestions about how Pves should behave in these synapses, none of which can be mechanistically accounted for at present. Pves should vary widely among the synapses, but only with respect to the first stimulus during an afferent activation. Thereafter, Pves should be able to increase in synapses with low Pves and to decrease in those with a high Pves, and to reach much the same ‘adapted’ value in all synapses. This adapted Pves value for release from the preprimed pool should also hold for the release of vesicles recruited during the course of the stimulation. This adapted value should also be large enough not to make Pves a factor deciding late release probability. It may be argued that the value of adapted Pves is of no concern since whatever this value recruited vesicles will, sooner or later, be released. However, since priming is likely to be a reversible process (see also Voets, 2000; Hanse & Gustafsson, 2001b) a too low adapted Pves value will decrease late release probability by allowing newly primed vesicles to become unprimed.

Differences among the synapses in the number of voltage-gated sodium channels in the terminal membrane may underlie variations in Pves by affecting the amount of active depolarisation of the terminal (Prakriya & Mennerick, 2000). Other possibilities are differences in the number or characteristics of functional calcium channels in proximity to the release site (Koester & Sakmann, 2000), or in the release machinery itself. At present we have no understanding of which of these mechanisms may operate here, or why the initial difference is replaced by a later convergence. What may seem reasonable is that the mechanism that underlies the heterogeneity in Pves1 is also the factor that is regularised later on. In that sense a modification of the release machinery itself, rather than of sodium and calcium currents, seems more likely. An activity-dependent, but release-independent, decrease in Pves, possibly explained by an activity-dependent change in the calcium sensor at the release site, has also been described in other synapses (Betz, 1970; Hsu et al. 1996; Bellingham & Walmsley, 1999; Thomson & Bannister, 1999; Wu & Borst, 1999; Waldeck et al. 2000). The wide variation in Pves in the hippocampal synapses should then relate to an initial difference in such a calcium sensor, a difference that rapidly disappears with high-frequency activity. Since the variation in Pves1 seems instrumental in producing much of the heterogeneity in facilitation/depression behaviour among the synapses, this ‘adaptation’ in Pves may be seen as a return to a ‘natural’ release state, being rather the same among the synapses. The heterogeneity in initial Pves can then be seen as a functional adaptation to create differential dynamics of release in the synaptic population.

Acknowledgments

This project was supported by the Swedish Medical Research Council (project number 12600 and 05180), the Swedish Society of Medicine, Harald Jeanson's Foundation, The Royal Society of Arts and Sciences in Göteborg, Magnus Bergvall's Foundation and Adlerbertska Research Foundation.

References

  • Bellingham MC, Walmsley B. A novel presynaptic inhibitory mechanism underlies paired pulse depression at a fast central synapse. Neuron. 1999;23:159–170. [PubMed]
  • Betz WJ. Depression of transmitter release at the neuromuscular junction of the frog. Journal of Physiology. 1970;206:629–644. [PMC free article] [PubMed]
  • Dobrunz LE, Stevens CF. Heterogeneity of release probability, facilitation, and depletion at central synapses. Neuron. 1997;18:995–1008. [PubMed]
  • Elmqvist D, Quastel DM. A quantitative study of end-plate potentials in isolated human muscle. Journal of Physiology. 1965;178:505–529. [PMC free article] [PubMed]
  • Hanse E, Gustafsson B. Quantal variability at glutamatergic synapses in area CA1 of the rat neonatal hippocampus. Journal of Physiology. 2001a;531:467–480. [PMC free article] [PubMed]
  • Hanse E, Gustafsson B. Vesicle release probability and pre-primed pool at glutamatergic synapses in area CA1 of the rat neonatal hippocampus. Journal of Physiology. 2001b;531:481–493. [PMC free article] [PubMed]
  • Hjelmstad GO, Isaac JT, Nicoll RA, Malenka RC. Lack of AMPA receptor desensitization during basal synaptic transmission in the hippocampal slice. Journal of Neurophysiology. 1999;81:3096–3099. [PubMed]
  • Hsia AY, Malenka RC, Nicoll RA. Development of excitatory circuitry in the hippocampus. Journal of Neurophysiology. 1998;79:2013–2024. [PubMed]
  • Hsu SF, Augustine GJ, Jackson MB. Adaptation of Ca2+-triggered exocytosis in presynaptic terminals. Neuron. 1996;17:501–512. [PubMed]
  • Huang YY, Gustafsson B, Wigstrom H. Facilitation of hippocampal long-term potentiation in slices perfused with high concentrations of calcium. Brain Research. 1988;456:88–94. [PubMed]
  • Katz B, Miledi R. The role of calcium in neuromuscular facilitation. Journal of Physiology. 1968;195:481–492. [PMC free article] [PubMed]
  • Koester HJ, Sakmann B. Calcium dynamics associated with action potentials in single nerve terminals of pyramidal cells in layer 2/3 of the young rat neocortex. Journal of Physiology. 2000;529:625–646. [PMC free article] [PubMed]
  • Liley AW, North KAK. An electrical investigation of effects of repetitive stimulation on mammalian neuromuscular junction. Journal of Neurophysiology. 1953;16:509–527. [PubMed]
  • Liu G, Choi S, Tsien RW. Variability of neurotransmitter concentration and nonsaturation of postsynaptic AMPA receptors at synapses in hippocampal cultures and slices. Neuron. 1999;22:395–409. [PubMed]
  • McAllister AK, Stevens CF. Nonsaturation of AMPA and NMDA receptors at hippocampal synapses. Proceedings of the National Academy of Sciences of the USA. 2000;97:6173–6178. [PMC free article] [PubMed]
  • Matveev V, Wang XJ. Implications of all-or-none synaptic transmission and short-term depression beyond vesicle depletion: a computational study. Journal of Neuroscience. 2000;20:1575–1588. [PubMed]
  • Neher E. Vesicle pools and Ca2+ microdomains: new tools for understanding their roles in neurotransmitter release. Neuron. 1998;20:389–399. [PubMed]
  • Prakriya M, Mennerick S. Selective depression of low-release probability excitatory synapses by sodium channel blockers. Neuron. 2000;26:671–682. [PubMed]
  • Stevens CF, Wang Y. Facilitation and depression at single central synapses. Neuron. 1995;14:795–802. [PubMed]
  • Thomson AM. Molecular frequency filters at central synapses. Progress in Neurobiology. 2000;62:159–196. [PubMed]
  • Thomson AM, Bannister AP. Release-independent depression at pyramidal inputs onto specific cell targets: dual recordings in slices of rat cortex. Journal of Physiology. 1999;519:57–70. [PMC free article] [PubMed]
  • Triller A, Korn H. Transmission at a central inhibitory synapse. III. Ultrastructure of physiologically identified and stained terminals. Journal of Neurophysiology. 1982;48:708–736. [PubMed]
  • Trussell LO, Zhang S, Raman IM. Desensitization of AMPA receptors upon multiquantal neurotransmitter release. Neuron. 1993;10:1185–1196. [PubMed]
  • Umemiya M, Senda M, Murphy TH. Behaviour of NMDA and AMPA receptor-mediated miniature EPSCs at rat cortical neuron synapses identified by calcium imaging. Journal of Physiology. 1999;521:113–122. [PMC free article] [PubMed]
  • Waldeck RF, Pereda A, Faber DS. Properties and plasticity of paired-pulse depression at a central synapse. Journal of Neuroscience. 2000;20:5312–5320. [PubMed]
  • Wang C, Zucker RS. Regulation of synaptic vesicle recycling by calcium and serotonin. Neuron. 1998;21:155–167. [PubMed]
  • Voets T. Dissection of three Ca2+-dependent steps leading to secretion in chromaffin cells from mouse adrenal slices. Neuron. 2000;28:537–545. [PubMed]
  • Wu LG, Borst JG. The reduced release probability of releasable vesicles during recovery from short-term synaptic depression. Neuron. 1999;23:821–832. [PubMed]
  • Zucker RS. Short-term synaptic plasticity. Annual Review of Neuroscience. 1989;12:13–31. [PubMed]
  • Zucker RS. Exocytosis: a molecular and physiological perspective. Neuron. 1996;17:1049–1055. [PubMed]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...