|
|
||||||||
Department of Neurology and Division of Neuroscience, Baylor College of Medicine, Houston, Texas 77030
| |
ABSTRACT |
|---|
|
|
|---|
Schulz, Paul E. and Jill C. Fitzgibbons. Differing mechanisms of expression for short- and long-term potentiation. J. Neurophysiol. 78: 321-334, 1997. Long-term potentiation (LTP) is a use-dependent form of synaptic plasticity that is of great interest as a cellular mechanism that may contribute to memory storage. It is the sustained phase of population excitatory postsynaptic potential induced by high-frequency stimulation (HFS). HFS can also induce short-term potentiation (STP), a decremental potentiation lasting ~15 min. It has been unclear whether STP is simply a reversible form of LTP elicited by subthreshold stimuli or whether it is an independently expressed form of synaptic plasticity. We have attempted to clarify the relationship between LTP and STP in the extracellular recording technique in area CA1 of the adult rat hippocampal slice preparation to test four predictions of the hypothesis that LTP and STP are expressed via the same mechanism. First, occluding LTP expression should block STP expression. Saturating LTP under six different conditions, however, did not occlude STP expression. Second, occluding STP expression should occlude LTP expression. The partial or full occlusion of STP by two maneuvers (increasing the stimulus intensity used for HFS or applying 3-isobutyl-1-methylxanthine), however, did not occlude LTP expression. Third, LTP increases and decreases paired-pulse facilitation (PPF), and STP should have the same effect. STP did not change PPF, however. The first three results, then, suggest that STP and LTP are expressed via different mechanisms. Fourth, STP should be maximal near the LTP induction threshold, and then decrease above it. Surprisingly, STP was maximal at or very close to the LTP induction threshold, but it did not decrease above this threshold. This relationship suggests the possibility that STP and LTP share an induction step(s). What is the function of the independently expressed STP? We find that LTP can be induced by two HFSs, each of which is subthreshold for LTP, if the second is given during STP from the first. This suggests that STP can temporarily lower the LTP induction threshold. Three lines of evidence, then, suggest that STP and LTP may be expressed via different mechanisms; however, the proximity of STP saturation to LTP induction suggests that they may share an induction step(s). STP may also have the very important function of temporarily lowering the LTP induction threshold. Finally, these data suggestion caution in interpreting LTP data obtained <20-30 min after HFS, because they may be contaminated by STP, which appears to have different underlying mechanisms.
Long-term potentiation (LTP) is a use-dependent form of synaptic plasticity that is of great interest as a cellular mechanism that may contribute to memory formation (Brown et al. 1988 Preparation of hippocampal slices
Brains of adult Sprague-Dawley rats (50-150 g) were quickly removed and placed in iced saline. The hippocampi were dissected out and 400 µm-thick slices were made perpendicular to the septotemporal axis with a Vibratome (Technical Products International). Slices were transferred to a Haas-type (Haas et al. 1979 Extracellular recordings
Microelectrodes were pulled from 1.5-mm-OD glass tubing with the use of a Flaming/Brown micropipette puller (Sutter Instrument) and were filled with 750 mM NaCl (resistance 1-5 M Data analysis
To obtain suitable pEPSPs, an attempt was made to clearly identify the fiber volley as being separate from the initial pEPSP slope. pEPSP traces were analyzed by obtaining the maximum initial pEPSP slope occurring after the fiber volley. The maximum slope was obtained by linear regression of the points obtained over a range of 0.4-0.6 ms on the initial slope. To avoid bias, the same time points on the initial slope of the pEPSP were analyzed throughout the experiment. Any experiments in which a change occurred in fiber volley or stimulus artifact were excluded from analysis. PPF was calculated as the slope of the second pEPSP minus the first, divided by the first. Statistical tests were performed with the use of standard methods (Zar 1984 We tested the following four predictions of the hypothesis that STP is a reversible form of LTP that is expressed via the same mechanisms. Occluding LTP expression will occlude STP. Occluding STP will occlude LTP. STP will have the same effect on PPF that LTP does (Buonomano and Merzenich 1996 Occluding LTP does not occlude STP expression
If STP is a reversible form of LTP, then maximizing LTP expression should decrease the expression of STP. This would occur because the conversion of STP to LTP with each successive HFS would leave fewer sites available to express STP. This prediction was tested by saturating LTP and examining its effect on STP. HFS was administered repeatedly at 20-min intervals until LTP was saturated (Fig. 1A, initial pEPSP = 1.3 mV, high-calcium saline, LTP = 304%, baseline = 100%, n = 4). Despite the occlusion of LTP, STP expression persisted. In fact, the absolute STP magnitude increased across sequential episodes of HFS; for example, the STP associated with the seventh HFS is 0.5 mV/ms (2.8
MAXIMIZING STP EXPRESSION.
One explanation for not detecting a decrease in STP as LTP was saturated is that maximal STP may not have been elicited under the conditions of study. For a number of reasons, it may then be difficult to detect a decrease in STP. Thus we sought conditions under which greater STP might be elicited. Two maneuvers hypothesized to enhance STP were increasing bath calcium and increasing the stimulus intensity used for HFS. Both should enhance postsynaptic calcium influx during HFS, which has been reported to increase STP (Malenka 1991
MAXIMIZING LTP EXPRESSION.
A second explanation for the lack of effect of LTP saturation on STP is that not enough LTP was elicited. Several measures were taken, however, to maximize LTP magnitude. First, LTP was studied under conditions thought to increase it, such as testing in the presence of picrotoxin and increased extracellular calcium. Second, the magnitude of LTP elicited at saturation was compared with that elicited by other groups, and they were found to be similar. And third, after LTP was saturated by our standard induction protocol, we tested whether additional LTP could be elicited by HFS with a greater intensity or a longer duration. As noted above, neither maneuver elicited additional LTP. We conclude that there are no obvious ways to elicit additional LTP.
PTP VERSUS STP.
A third explanation for the lack of effect of LTP saturation on STP is that the decremental potentiation under study was actually PTP instead of STP. This possibility was addressed by examining the time course of PTP in the normal- and high-calcium salines. HFS was given in the presence of 100 µM D,L-APV, which blocks both STP and LTP. In normal calcium, the duration of PTP was very brief, in agreement with previous observations (Malenka 1991 Occluding STP does not occlude LTP expression
A second prediction of the hypothesis that STP and LTP are expressed via the same mechanism is that STP occlusion will occlude LTP. It was already observed that STP is partially occluded by increasing the stimulus intensity used for HFS, and that this does not occlude LTP (Fig. 2, B and D). We sought another circumstance under which STP could be occluded to observe its effect on LTP.
Differential effects of STP and LTP on PPF
A third prediction of the hypothesis that STP and LTP are expressed via the same mechanism is that they should have the same effect on the expression of PPF. PPF is an increase in a second EPSP when it follows shortly after a first, which is thought to be presynaptically mediated. LTP alters PPF in a very predictable way: it increases or decreases PPF depending on the initial PPF value (Buonomano and Merzenich 1996
STP saturates at, or very near, the LTP induction threshold
Tests of the first three predictions of the hypothesis that STP and LTP are similarly expressed suggested the opposite result, i.e., that they are, in fact, independently expressed. The fourth prediction has three parts: 1) STP has a lower threshold than LTP, 2) increases in the duration or intensity of HFS will increase STP up to the LTP induction threshold, and 3) HFS beyond that will be associated with decreases in STP. Those decreases will occur as more synapses develop the persistent changes of LTP, leaving fewer synapses to express STP. But if the hypothesis is wrong and the STP cascade is independent of LTP, then STP will continue to increase with each HFS beyond the LTP threshold. This is because the induction of LTP with one HFS will produce greater depolarization during the next HFS. The three parts of this prediction were tested with the use of two protocols.
STP lowers the LTP induction threshold
If STP is, in fact, an independently expressed form of potentiation, it raises the question of why two different NMDA-dependent forms of synaptic plasticity are necessary in the hippocampus and what their respective roles are. One hypothesis regarding the function of STP is that it might allow two subthreshold stimuli to induce LTP if they occur near each other in time. This was initially considered to be unlikely because of a report indicating that frequent HFS could actually decrease the probability of LTP induction (Huang et al. 1992
There are three main findings of this study. First, three lines of evidence suggest that STP and LTP are expressed via different mechanisms: occluding LTP does not occlude STP expression, occluding STP does not occlude LTP expression, and STP and LTP have different effects on PPF. Second, this study demonstrates that STP saturation occurs either at, or very close to, the LTP induction threshold. This could hardly be coincidental, and suggests that STP and LTP may share an initial induction step(s). Finally, this study demonstrates that the pairing of two HFSs, which individually only induce STP, may result in the induction of LTP. This suggests that STP lowers the LTP threshold for a period of time, which may have very important consequences for synaptic efficacy.
Differing mechanisms of expression
It has been suggested that if LTP and STP are expressed via the same mechanism, then LTP saturation ought to occlude STP expression (Stevens 1993 STP and LTP may share induction steps
The fourth prediction of the hypothesis that STP and LTP are similarly expressed had three parts. First, the threshold for STP induction will be lower than for LTP. Second, as HFS intensity or duration is increased, STP magnitude will increase up to the LTP induction threshold. Once the threshold is met, then, rather than increasing the reversible STP, LTP will be expressed. And third, further HFSs above the LTP threshold will decrease STP because they will produce more LTP, thereby leaving fewer release sites available to express STP.
STP lowers the LTP induction threshold
Because STP appears to be an independently expressed form of synaptic plasticity, it must have a specific function. We investigated the possibility that its acts to raise or lower the LTP induction threshold. It has already been shown that frequent, repeated (5 times) induction of STP can raise the LTP threshold (Huang et al. 1992
![]()
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
; Madison et al. 1991
; Teyler and DiScenna 1984
). LTP of Schaffer collateral fibers in hippocampal area CA1 is defined as the sustained potentiation, with a duration of >15 min, that can be induced by a short burst of high-frequency stimulation (HFS). The induction of LTP by 100 Hz HFS is blocked by the N-methyl-D-aspartate (NMDA) receptor antagonist amino-phosphono-valerate (APV).
). The second decremental phase is blocked by APV, has a duration of ~10-15 min, and is referred to as short-term potentiation (STP) (Anwyl et al. 1989
; Kauer et al. 1988
; Malenka 1991
).
; Malenka 1991
) and requires an increase in postsynaptic intracellular calcium (Malenka et al. 1988
). STP reversibly occludes LTP, and vice versa (Asztely et al. 1991
; Collingridge et al. 1991
; Gustafsson et al. 1989
; Huang et al. 1992
; Kauer et al. 1988
). Both are associated with increases in the amplitude of miniature synaptic currents (Manabe et al. 1992
). And the degree to which STP decays versus the remaining stable LTP varies under different conditions, which suggests that STP can be converted to LTP to different extents (Gustafsson and Wigström 1990
; Gustafsson et al. 1989
; Malenka and Nicoll 1990
).
). NMDA-induced STP is inhibited in high-calcium saline, whereas LTP is not (McGuinness et al. 1991a
). An STP-like phenomena is induced by NMDA application alone (Kauer et al. 1988
), but it has been difficult, although not impossible (Collingridge et al. 1991
), to induce LTP through NMDA application. And second-messenger involvement may differ: protein kinase C activation enhances LTP (Lovinger and Routtenberg 1988
), but inhibits STP (Asztely et al. 1990
; McGuinness et al. 1991b
), and multiple kinase inhibitors block LTP (Malenka et al. 1989
; Malinow et al. 1988
, 1989
), leaving an STP-like phenomenon (Malenka et al. 1989
; Malinow et al. 1989
; O'Dell et al. 1991
; Schuman and Madison 1991
). The kinase blocking experiments appear to argue strongly for differing induction steps for STP and LTP. It is not clear, however, whether the decremental potentiation induced in the presence of kinase inhibitors is the same as STP. At a minimum, STP and kinase-inhibitor-induced decremental potentiation differ markedly in duration (10-20 min vs. 30-60 min). In addition, even if they are identical, the kinase results could still be consistent with similar mechanisms of expression. LTP expression, for example, may require a kinase for stabilization. In the absence of an activated kinase, then, decremental potentiation (STP), mediated by the same mechanism, would result.
, 1992
).
![]()
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
) interface recording chamber (Medical Systems) at 32.5 ± 0.1°C (mean ± SE) and maintained with the use of standard procedures (Schulz 1997
; Schulz et al. 1994
, 1995
).
-aminobutyric acid-A-mediated inhibition. The salines were gassed with 95% O2-5% CO2. The bathing solution sometimes contained 100 µM D,L-APV (Research Biochemicals, Natick, MA) or 10 µM 3-isobutyl-1-methylxanthine (IBMX), as indicated.
). Stimulation was given at 20-s intervals via bipolar, Teflon-coated platinum stimulating electrodes (WPI stimulus generator and isolator, stimulus duration 50 µs). Extracellular population excitatory psotsynaptic potential (pEPSP) recordings were made from stratum radiatum in area CA1. The data were filtered at 5 kHz and recorded on-line with either a DEC-11/23 or a NeXT computer with software written in Basic.
).
![]()
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
; Kleschevnikov et al. 1997
; Schulz et al. 1994
). And, STP will be maximal near the LTP induction threshold.
2.3 = 0.5) versus 0.3 mV/ms for the first HFS (1.4
1.1 = 0.3).

View larger version (27K):
[in a new window]
FIG. 1.
Occlusion of long-term potentiation (LTP) does not occlude the expression of short-term potentiation (STP). A: high-frequency stimulation (HFS) was administered repeatedly at a single stimulus intensity that eventually resulted in LTP saturation (Satn; slope increase = 204%;n = 4, high-calcium solution, initial population excitatory postsynaptic potential (pEPSP) = 1.3 mV, mean ± SE). Inset: raw voltage traces from a typical experiment (each is an average of several traces). B: pEPSP slopes of the 1st 7 HFSs in A were normalized (pre-HFS slope = 100%), the time of HFS was set equal to 0, and slopes are superimposed. C: to compare STP across HFS, LTP (slope 15 min after HFS) was subtracted from all points in each curve in B and is plotted in C (see text for example). C demonstrates that normalized STP is of constant magnitude and duration from baseline through LTP saturation, despite occlusion of LTP, which suggests differing mechanisms of expression for STP and LTP.

View larger version (26K):
[in a new window]
FIG. 2.
LTP occlusion under other conditions also does not occlude STP expression. HFS was administered repeatedly with a single stimulus intensity that initially yielded pEPSP amplitudes of 1.3 mV (A,n = 7 and C, n = 12) or 2.6 mV (B, n = 8 and D, n = 7), either in normal-calcium (A and B) or moderate-calcium (C and D) saline. Data are displayed as in Fig. 1C. All axes have same scale for convenience of comparison. Under each condition, the STP elicited by all HFS was of similar magnitude and duration. Less STP was expressed, however, with HFS at 2.6 mV (B and D) vs. 1.3 mV (A, C, and Fig. 1). Even when less STP was expressed, though, LTP magnitude was identical (compare distances of baseline points below - - -, which is equivalent to LTP magnitude).

View larger version (22K):
[in a new window]
FIG. 3.
STP magnitude is independent of LTP magnitude. HFS was given repeatedly with a 1.3-mV (
,
,
) or a 2.6-mV (
,
,
) pEPSP in 4 different solutions. Resulting STP and LTP amplitudes were averaged across experiments and are plotted. LTP was defined as potentiation present 15-20 min after HFS. STP was defined as an average of the 3 highest points after HFS, excluding those containing posttetanic potentiation (PTP). Points at right on X-axis are initial HFSs under each condition, which produce the most LTP. Points at left are subsequent HFSs that produce less LTP as saturation is approached (saturation = 0 on the X-axis). Horizontality of regression lines indicates that STP and LTP magnitude are independent.
).

View larger version (22K):
[in a new window]
FIG. 4.
A: effect of bath calcium on STP magnitude was tested by eliciting STP in the normal-calcium (n = 15), moderate-calcium (n = 12), or high-calcium (n = 7) salines (initial pEPSP magnitudes = 1.3 mV), which produced 35%, 28%, and 37% LTP (displayed as in Fig. 1C). Despite increasing bath calcium 3-fold, there was little difference in STP, except for the 1st 3 points after HFS, which may be due to PTP. B: stimulus intensity used for HFS was increased after LTP saturation to determine whether additional LTP or STP could be elicited. LTP was saturated (LTP saturation, n = 13, moderate-calcium saline, baseline pEPSP = 1.3 mV). Then HFS was administered with an increased stimulus intensity (average increase in pEPSP slope = 70%), and stimulus intensity was returned to baseline for recording (Increased SI). Finally, HFS was administered at the baseline intensity again (original SI). No additional LTP or STP was elicited. C: HFS with (
, n = 4) and without (
) amino-phosphono-valerate (APV) indicates that PTP has a duration of only 10 s in normal saline, so that its contribution to decremental potentiation can be ignored.
). Then stimulus intensity was increased by an average of 70%, an additional HFS was administered, and stimulus intensity was returned to baseline for recording. No additional STP or LTP was elicited (
). Another HFS at the original stimulus intensity was given to verify that the slice was still capable of producing the original STP magnitude (
). The effect of increasing the duration of HFS was examined next. HFS was given for 1 s at 100 Hz, and was repeated three times for a total of 300 stimuli (vs. 50 stimuli over 2 s in the original protocol). Three conditions were tested: two slices from Fig. 1, six slices from Fig. 2C, and two slices from Fig. 2D (data not shown). No additional STP or LTP was elicited.
) used sub-LTP-threshold stimuli, whereas we used supra-LTP-threshold stimuli. An explanation for our results, then, may be that when LTP is elicited, STP is already maximal. Increasing calcium influx further, then, would not produce additional STP. This important possibility is explored in greater detail later. We concluded here thatmaximal STP is apparently elicited by our LTP induction protocol.
), only being present during the first point measured 5-15 s after HFS (Fig. 4C,
, n = 5). In high calcium, PTP has a duration of 30 s-2 min (data not shown, n = 5). Thus all of the decremental potentiation observed after the first minute or two is STP and not PTP.
), IBMX was washed in, and an additional HFS was given (2nd
). It elicited LTP despite completely occluding STP. After IBMX washout (3rd
), STP returned to normal. The first two HFSs from Fig. 5A are normalized in Fig. 5B1. The normalized STP elicited after IBMX washout is shown in Fig. 5B2. The occlusion of STP by IBMX is specific, because when STP is produced by one HFS under control conditions, it is never occluded by a second HFS given at the same intensity and duration (n = 102). These IBMX results were recently confirmed in part (Blitzer et al. 1995
).

View larger version (18K):
[in a new window]
FIG. 5.
Occluding STP expression does not occlude LTP expression. A: STP and LTP were elicited by the 1st HFS (
). Twenty minutes later,3-isobutyl-1-methylxanthine (IBMX) was washed in and out without stimulating the slice. Ten minutes later, stimulation was restarted. After a suitable baseline, HFS was given again, and LTP was elicited without STP. HFS 20 and 40 min later elicited STP again. B1: 1st and 2nd HFSs from A are normalized and superimposed. They demonstrate a loss of STP despite preservation of LTP. B2: 3rd HFS from A is normalized and demonstrates that STP returned as IBMX washed out. That IBMX can block STP without occluding LTP suggests differing mechanisms of expression for STP and LTP. Note that IBMX partially or completely blocked STP only 7 of 18 times, so that no conclusion can be drawn regarding the mechanism underlying STP expression.
; Kleschevnikov et al. 1997
; Schulz 1997
; Schulz et al. 1994
, 1995
). PPF averaged across all slices remains constant, however, as demonstrated by many groups (Arai et al. 1994
; Ghijsen and Da Silva 1991
; Manabe et al. 1993
; McNaughton 1982
; Muller and Lynch 1989
; Schulz et al. 1995
; Zalutsky and Nicoll 1990
).
). The slopes of the first and second paired stimuli are shown over time in Fig. 6D. PPF for that slice is shown over time in Fig. 6E, where the slope of Stim 1 increased from 0.66 to 1.08 with the first six HFSs, and the slope of Stim 2 increased from 0.86 to 1.34 so that PPF decreased from 30% to 24%. HFS to other slices produced an increase or a decrease in PPF with LTP (Fig. 6F). The pattern that emerged to explain the increases or decreases in PPF is that greater initial PPF is associated with decreases in PPF (Fig. 6F, right) and greater LTP. And lesser initial PPF is associated with increases in PPF (Fig. 6F, left) and less LTP.

View larger version (20K):
[in a new window]
FIG. 6.
Differential effects of STP and LTP on paired-pulse facilitation (PPF). A and B: STP was elicited in isolation from LTP (A) and had no effect on average PPF (B, n = 8). C: PPF for individual slices also did not change. Slices were tested in moderate calcium (open circles, n = 3) or high calcium (
, n = 8). Slope of linear regression line is
0.02(r2 = 0.08, P > 0.20). D: in contrast, LTP routinely elicits changes in PPF as shown in the example where both paired pulses are plotted. E: PPF for slice in D decreased as LTP was induced. F: changes in PPF with LTP occur in a regular way (n = 100). Slope of linear regression line is
0.70 (P < 0.001). Data in D-F are reproduced from Schulz et al. (1995)
. Differential effects of STP and LTP on PPF suggest differing mechanisms of expression.
). The lack of change in average PPF with STP (Fig. 6B) could be misleading if PPF both increased and decreased as it does with LTP (Fig. 6F). To address this possibility, the change in PPF associated with STP in each individual slice is plotted versus initial PPF in Fig. 6C (
). PPF values were obtained >2 min after HFS so that they are not contaminated by PTP. PPF was also examined during STP in the moderate-calcium saline; those data are plotted in Fig. 6C as open circles. A linear regression line for all points is drawn and follows the equation y =
0.02x + 0.07 (n = 11, r2 = 0.08, P > 0.20). The relatively horizontal slope of
0.02 contrasts with that of Fig. 6F, where the slope is
0.70. STP and LTP have very different effects on PPF, then, suggesting for a third time that STP and LTP have differing mechanisms of expression.

View larger version (16K):
[in a new window]
FIG. 7.
STP saturates at, or very close to, LTP induction threshold. A: slice was followed with 1 stimulus intensity, but was given HFS with successively larger intensities that produced 0.3- to 2.1-mV pEPSPs. Initial HFS produced no change in pEPSP. Greater intensities led to STP alone (labeled 1.1 and 1.3). And still greater intensities led to STP with LTP (1.5 and 1.7) until LTP was saturated (2.1). B: normalizing STP for HFS with the 1.3-, 1.5-, and 1.9-mV pEPSPs demonstrates that STP increased up to the LTP induction threshold (1.5 mV), after which it remained constant.

View larger version (31K):
[in a new window]
FIG. 8.
STP saturates at, or very close to, LTP induction threshold. A: HFS of successively longer duration produced an increase in STP and then a decrease. B: maximum STP was elicited at LTP induction threshold (0.10 s), and less STP was elicited by earlier (0.02 s) or later (1.50 s) HFS. C and D: another example of a decrease in STP with longer HFS (0.5 s). E: an additional HFS was given to the slice in C for 0.5 s with APV present. PTP elicited had a duration of only 1 min, indicating that the decremental potentiation in C is due to STP, not PTP. F: STP magnitude is plotted vs. HFS number for slices given HFS for successively longer durations(n = 13). LTP thresholds differed, so slices are aligned by assigning a value of 0 on the X-axis to threshold HFS. HFS before and after have negative and positive numbers, respectively. STP magnitude is area under potentiation curves derived by normalizing them and subtracting LTP. Maximum STP occurs near LTP induction threshold.

View larger version (16K):
[in a new window]
FIG. 9.
Depression of STP by long-duration HFS is reversible. HFS for 0.2 s elicited STP. Another HFS for 1.0 s elicited LTP and partially occluded STP. A final HFS for 0.4 s elicited greater STP. This indicates that the depression of STP by the 1.0-s HFS was not due to permanent occlusion of STP by LTP, as would occur if STP and LTP were similarly expressed.

View larger version (25K):
[in a new window]
FIG. 10.
Decrease in STP with longer-duration HFS is input specific. A: HFS of successively longer duration produced an increase in STP and then a decrease. Nonetheless, there was no change in a 2nd independent input, indicating that decreases are input specific. Slope of input 1 (
) and amplitude of input 2 (
) are plotted. B: average of 3 similar experiments is plotted.
). Nonetheless, this hypothesis was tested by giving a single HFS that induced STP without LTP (Fig. 11, A and B). Then the same HFS was given twice with a separation of 1 min (Fig. 11A) or 3 min (Fig. 11B). Surprisingly, LTP was often elicited by the second HFS if it was given while STP elicited by the first was still present. Superimposing the results of the two HFS versus the one HFS, averaged over all experiments, demonstrates that two sub-LTP-threshold HFSs can produce LTP (Fig. 11C).

View larger version (17K):
[in a new window]
FIG. 11.
STP lowers LTP induction threshold. A and B: a single HFS only induced STP. Two HFSs of same intensity and duration, separated by 1 (A) or 3 (B) min, often resulted in LTP. C: superimposition of normalized group data for1-min separations demonstrates LTP after paired HFS (
), but not single HFS (
).

View larger version (11K):
[in a new window]
FIG. 12.
Induction of LTP by paired HFS, which individually only induced STP, is not due to a longer total duration of HFS. A: HFS for 0.1, 0.3, and 0.5 s did not elicit LTP. Two HFSs for 0.1 s delivered 1 min apart, however, did. This occurred despite total HFS duration (0.2 s) being less than for the 0.3- or 0.5-s HFS. B: experiment in A was performed in reverse order. Two HFS for 0.05 s induced LTP, whereas another HFS for the same total duration (0.1 s) did not. Two more HFS for 0.05 s induced a small additional amount of LTP.
![]()
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
). Inspection of the figures in many studies suggests that the routine induction of LTP by one or several HFSs does not have an obvious affect on STP. But concluding that LTP saturation has no effect on STP expression requires more rigorous testing. Accordingly, many measures were taken here to maximize both STP and LTP expression so that if LTP were to occlude STP expression, it would be detected. These measures included saturating LTP in a number of different salines, with the use of two different initial stimulus intensities, giving HFS with an increased stimulus intensity and duration after LTP was thought to be saturated, separately analyzing subsets of slices that had the most LTP, comparing the magnitudes of LTP at saturation to those previously reported, and testing for even small changes in STP by normalizing and superimposing the results of each HFS. Despite intensive study, however, we were unable to find any condition under which LTP saturation, by a constant-intensity and -duration HFS, led to an occlusion of STP expression. This suggests that STP and LTP are expressed via different mechanisms.
; Schulz et al. 1994
, 1995
). The lack of effect of STP on PPF (Colino et al. 1992
), the positive effect of LTP on PPF in individual slices (Buonomano and Merzenich 1996
; Kleschevnikov et al. 1997
), and the absence of effect of LTP on PPF when it is averaged across slices (Arai et al. 1994
; Ghijsen and Da Silva 1991
; Manabe et al. 1993
; McNaughton 1982
; Muller and Lynch 1989
; Zalutsky and Nicoll 1990
) have been independently corroborated. Note that the lack of effect of STP on PPF does not necessarily indicate that the mechanism of STP expression resides postsynaptically, because there may be presynaptic mechanisms that do not effect PPF. Nonetheless, their differential effects on PPF indicate that STP and LTP are not identically expressed.
), and both are APV sensitive (Fig. 4C) (Anwyl et al. 1989
; Reymann and Matthies 1989
; Reymann et al. 1989
). But both findings could be explained by similar or dissimilar mechanisms of induction.
). Nonetheless, we investigated whether administering only a single STP-inducing HFS might lower the LTP threshold for a short period of time. Our data indicate that, in fact, the STP induced by a single HFS lowers the LTP induction threshold for at least a few minutes so that a second, previously subthreshold HFS can produce LTP (Fig. 11). STP, then, may act as a coincidence detector. Such a role for STP has important implications for our understanding of synaptic plasticity and needs to be considered in the neural modeling of changes in synaptic weights.
| |
ACKNOWLEDGEMENTS |
|---|
We thank J. Rexer and R. Varghese for technical assistance, M. Haque and R. Gray for computer assistance, and S. Williams and D. Johnston for review of the manuscript.
This work was supported by National Institutes of Health grants to P. E. Schulz.
| |
FOOTNOTES |
|---|
Address for reprint requests: P. E. Schulz, Department of Neurology, NB-302, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030.
Received 15 November 1996; accepted in final form 14 March 1997.
| |
REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Rosato-Siri, A. Cattaneo, and E. Cherubini Nicotine-induced enhancement of synaptic plasticity at CA3-CA1 synapses requires GABAergic interneurons in adult anti-NGF mice J. Physiol., October 15, 2006; 576(2): 361 - 377. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. O'Boyle, V. Do, B. E. Derrick, and B. J. Claiborne In Vivo Recordings of Long-Term Potentiation and Long-Term Depression in the Dentate Gyrus of the Neonatal Rat J Neurophysiol, February 1, 2004; 91(2): 613 - 622. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. O Mann and S. A Greenfield Novel modulatory mechanisms revealed by the sustained application of nicotine in the guinea-pig hippocampus in vitro J. Physiol., September 1, 2003; 551(2): 539 - 550. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Volianskis and M. S Jensen Transient and sustained types of long-term potentiation in the CA1 area of the rat hippocampus J. Physiol., July 15, 2003; 550(2): 459 - 492. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Mereu, M. Fa, L. Ferraro, R. Cagiano, T. Antonelli, M. Tattoli, V. Ghiglieri, S. Tanganelli, G. L. Gessa, and V. Cuomo Prenatal exposure to a cannabinoid agonist produces memory deficits linked to dysfunction in hippocampal long-term potentiation and glutamate release PNAS, April 15, 2003; 100(8): 4915 - 4920. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Schulz, A. D. McIntosh, M. R. Kasten, B. Wieringa, and H. F. Epstein A Role for Myotonic Dystrophy Protein Kinase in Synaptic Plasticity J Neurophysiol, March 1, 2003; 89(3): 1177 - 1186. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. A. Hoffman, R. Sprengel, and B. Sakmann Molecular dissection of hippocampal theta-burst pairing potentiation PNAS, May 28, 2002; 99(11): 7740 - 7745. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. L. Morgan and T. J. Teyler Electrical Stimuli Patterned After the Theta-Rhythm Induce Multiple Forms of LTP J Neurophysiol, September 1, 2001; 86(3): 1289 - 1296. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Mereu, M. Cammalleri, M. Fà, W. Francesconi, P. Saba, M. Tattoli, L. Trabace, A. Vaccari, and V. Cuomo Prenatal Exposure to a Low Concentration of Carbon Monoxide Disrupts Hippocampal Long-Term Potentiation in Rat Offspring J. Pharmacol. Exp. Ther., August 1, 2000; 294(2): 728 - 734. [Abstract] [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |