Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 1999 Dec 1;521(Pt 2):451–466. doi: 10.1111/j.1469-7793.1999.00451.x

Effects of adenosine receptors on the synaptic and EPSP-spike components of long-term potentiation and depotentiation in the guinea-pig hippocampus

Satoshi Fujii *, Yoichiro Kuroda *, Ken-ichi Ito *, Kennya Kaneko *, Hiroshi Kato *
PMCID: PMC2269672  PMID: 10581315

Abstract

  1. Long-term potentiation (LTP) of synaptic efficacy comprises two components: a synaptic component consisting of increased field excitatory postsynaptic potentials (EPSPs), and a component consisting of a larger population spike amplitude for a given EPSP size (E-S potentiation). In hippocampal CA1 neurons, delivery of three weak bursts (5 pulses at 100 Hz, 20 min intervals) induced LTP in both the EPSP and E-S components. In the same cells, reversal of LTP (depotentiation, DP) in the field EPSP and the E-S component was achieved by delivering three trains of low-frequency stimuli (LFS; 200 pulses at 1 Hz, 20 min intervals).

  2. The effects of adenosine A1 and A2 receptor antagonists on the synaptic and E-S components of LTP and DP in CA1 neurons were studied by perfusing guinea-pig hippocampal slices with either 8-cyclopentyltheophylline (8-CPT) or CP-66713.

  3. When bursts or LFS were applied to CA1 inputs in the presence of the A1 receptor antagonist 8-CPT, the field EPSP was enhanced in LTP and attenuated in DP, while the E-S relationship was not significantly affected in either LTP or DP.

  4. When similar experiments were performed using the A2 receptor antagonist CP-66713, the field EPSP was blocked in LTP, but facilitated in DP, while E-S potentiation was enhanced during both LTP and DP.

  5. The results show that A1 and A2 adenosine receptors modulate both the synaptic and E-S components of the induction and reversal of LTP. Based on these results, we discuss the key issue of the contribution of these receptors to the dynamics of neuronal plasticity modification in hippocampal CA1 neurons.

Long-term potentiation (LTP) is persistent synaptic enhancement induced by a brief period of high-frequency electrical stimulation (HFS) of afferents (Bliss & Lømo, 1973; Bliss & Gardner-Medwin, 1973). In addition to LTP, another type of synaptic plasticity known as depotentiation (DP) has been reported, in which low-frequency stimulation (LFS) of afferents effectively reverses a pre-established LTP, both in vivo (Stäubli & Lynch, 1990; Doyle et al. 1997) and in vitro (Fujii et al. 1991; Bashir & Collingridge, 1994).

LTP consists of a synaptic component which increases field excitatory postsynaptic potentials (EPSPs), and a component that results in a larger population spike amplitude for a given EPSP size (potentiation of EPSP- spike coupling or E-S potentiation) (Andersen et al. 1983). E-S potentiation was noted in the earliest reports of LTP (Bliss & Lømo, 1973; Bliss & Gardner-Medwin, 1973) and has often been observed in the conventional synaptic LTP (Kairiss et al. 1987; Chavez-Noriega et al. 1989; Hess & Gustafsson, 1990; Tomasulo et al. 1991; Tomasulo & Ramirez, 1993). In hippocampal CA1 neurons, DP also consists of a synaptic component and an E-S component, both of which are effectively reversed by LFS towards pre-tetanic control levels (Fujii et al. 1997a). It is obvious that any long-term modification of the synaptic event will change the output of the neuron in response to a given excitatory input, and, as a consequence, there will be a change in discharge pattern in the neuronal network.

Glutamate, the main transmitter of hippocampal neurons, is released by input stimulation, and is believed to be an essential factor in the induction of LTP (Bliss & Collingridge, 1993) or DP (Fujii et al. 1991) in CA1 neurons. In addition, depending on the stimulation frequency, adenosine 5′-triphosphate (ATP) and adenosine derivatives, common constituents of synaptic vesicles (White, 1978), are released as co-transmitters at hippocampal CA1 synapses (Schubert et al. 1976; Wieraszko et al. 1989). It is generally accepted that there are at least two major classes of adenosine receptor, A1 and A2, originally proposed on the basis of adenylyl cyclase studies by Van Claker et al. (1979) and Londos et al. (1980). A1 receptors are known to be present at high levels in hippocampal CA1 neurons (Fastbom et al. 1987) and A2 receptors have also recently been shown to be present in these cells (Sebastião & Ribeiro, 1992; Cunha et al. 1994).

8-Cyclopentyltheophylline (8-CPT) is a potent antagonist of adenosine A1 receptors (Bruns et al. 1987). Dunwiddie & Fredholm (1989) have shown that, in rat hippocampal neurons, 1.0 μm 8-CPT has a significant effect on the A1 receptor-mediated decrease in cyclic AMP formation, but has almost no effect on the A2 receptor-mediated increase, indicating the specific antagonism of A1 receptors by 1.0 μm 8-CPT. In the case of CA1 neurons of rat hippocampal slices, it has been reported that 8-CPT (1.0-2.0 μm) enhances the baseline responses in the EPSP and the population spike, but decreases homosynaptic post-tetanic depression, induced by 5 Hz stimulation, in the EPSP and the population spike (Sekino & Koyama, 1992). Two micromolar 8-CPT also enhances the baseline EPSP slope and decreases heterosynaptic post-tetanic depression of the EPSP in CA1 neurons in guinea-pig hippocampal slices (Manzoni et al. 1994).

4-Amino-8-chloro-1-phenyl-(1,2,4)-triazolo(4,3a) quinoxaline (CP-66713) is a potent A2 receptor antagonist (the Ki values for A1 and A2 receptors are > 10 μm and 22 nm, respectively) (Sarges et al. 1990). For CA1 neurons of rat hippocampal slices, O'Kane & Stone (1998) have shown that 10 μm CP-66713 has no effect on either the baseline population spike or the baseline EPSP, while activation of A1 receptors by a selective receptor agonist results in a decrease of both.

Since LTP and DP are frequency-dependent synaptic plasticities (Collingridge et al. 1988; Fujii et al. 1991), endogenous adenosine, released during HFS or LFS and acting via A1 or A2 receptors, could contribute to LTP or DP in hippocampal CA1 neurons. Arai et al. (1990) showed that activation of A1 receptors interrupts LTP development in the field EPSP in these cells, while Larson et al. (1993) showed that DP in the field EPSP can be blocked by an A1 receptor antagonist, and suggested the involvement of A1 receptor activation in a synaptic component of DP. Kessey et al. (1997) showed that activation of A2 receptors contributes to the induction of LTP in the field EPSP in CA1 neurons. These results suggest that endogenous adenosine, acting via A1 or A2 receptors, modulates a synaptic component of both LTP and DP in CA1 neurons.

Previous studies have shown that, in hippocampal CA1 neurons, endogenous adenosine modulates the E-S component of both LTP and DP. Also using CA1 neurons, Sekino et al. (1991) showed that 10 μm CP-66713 prevents LTP induction in terms of the EPSP, but has no effect on the population spike; they suggested that activation of A2 receptors enhances the formation of LTP in the EPSP, but fails to influence the formation of LTP in the population spike, resulting in attenuation of E-S potentiation. Furthermore, we have shown that, in CA1 neurons, application of 10 μm CP-66713 during LFS facilitates DP in the field EPSP and inhibits that in the population spike, and concluded that activation of A2 receptors inhibits DP in the EPSP, facilitates DP in the population spike and consequently attenuates E-S potentiation (Fujii et al. 1992).

However, the effect of activation of adenosine A1 and A2 receptors on these components during LTP or DP has not been studied in detail. In this study, we perfused hippocampal slices with 1 μm 8-CPT as an adenosine A1 receptor antagonist, or with 10 μm CP-66713, as an adenosine A2 receptor antagonist, during HFS or LFS, and evaluated the effects on both the synaptic and E-S components of LTP and DP.

METHODS

Adult male Hartley guinea-pigs (250-300 g) were killed by decapitation. Experiments were performed in accordance with ethical guidelines of Yamagata University on the use and welfare of animals. The hippocampi were quickly removed and cut into transverse slices of 500 μm thickness using a rotor slicer (Dosaka DK-7700, Kyoto, Japan). Slices were preincubated at 30-32°C for a minimum of 1 h in a 95 % O2-5 % CO2 atmosphere in the standard solution before being placed in a 1 ml capacity recording chamber. The standard solution was composed of (mm): NaCl, 124; KCl, 5.0; NaH2PO4, 1.25; MgSO4, 2.0; CaCl2, 2.5; NaHCO3, 22.0; and D-glucose, 10.0. The slices were completely submerged in the solution, which was perfused continuously at a rate of 2-3 ml min−1. The temperature in the recording chamber was maintained at 30-32°C.

After immobilizing the slice in the recording chamber, a bipolar stimulating electrode was placed in the stratum radiatum to stimulate the input pathways to the CA1 neurons. A recording electrode was placed in the pyramidal cell body layer and another in the stratum radiatum of the CA1 region, from which the population spike and the field EPSP, respectively, were derived. Figure 1B shows typical population spikes and field EPSPs recorded simultaneously in these layers.

Figure 1. Bidirectional plasticities in the field EPSP and population spike in hippocampal CA1 neurons.

Figure 1

Open in a new tab

A, typical example of the time course of the changes in the slope of the field EPSP (S-EPSP, ○) and in the amplitude of the population spike (A-PS, •). Three bursts (downward arrows, 5 pulses at 100 Hz) at 20 min intervals resulted in summation of the potentiations. The potentiated responses were reduced by the delivery of LFS (horizontal bars, 1 Hz, 200 pulses) at 20 min intervals. EPSP-spike (E-S) potentiation is shown by a larger potentiation in the A-PS than in the S-EPSP in this and subsequent time course examples. B, simultaneous recordings of the field EPSP and population spike in CA1 neurons made before, and after, bursts or LFS, at the times indicated in A. Three successive responses were averaged and plotted.

At the beginning of each experiment, a stimulus-response curve (SRC) was established by increasing the stimulus intensity and measuring the amplitude of the population spike (A-PS) and the initial slope of the field EPSP (S-EPSP). Based on the input-output function of the SRC, the strength of the stimulus was adjusted to elicit a population spike with an amplitude of half the maximum (mean ±s.e.m. = 50.2 ± 2.7 % of the maximum, n = 48), and was fixed at this level throughout the experiments. To evaluate changes in both the field EPSP and the population spike, the S-EPSP and A-PS were measured using a microcomputer.

After checking the stability of the responses to a 20 s test stimulus, a burst stimulation (5 pulses at 100 Hz) was delivered every 20 min to elicit LTP (arrows in Figs 1, 2 and 3A-C). To produce DP, a train of LFS (200 pulses at 1 Hz, in Figs 1 and 5A-C) was started 20 min after a tetanus (100 pulses at 100 Hz) and applied three times at 20-25 min intervals. After delivery of tetanus or LFS, the test stimulus was repeated every 20 s and responses were recorded for a minimum of 60 min. SRCs were again determined 20 min after tetanus and 60 min after the final HFS or the final LFS by delivering the stimulus at a different intensity of 0.2 Hz.

Figure 2. LTP induced by five sequential burst stimulations.

Figure 2

Open in a new tab

Burst stimulations were applied to CA1 inputs in the standard solution at 20 min intervals. A, typical example of the time course of the changes in the S-EPSP (○) and A-PS (•). The downward arrows indicate burst stimulation (5 pulses at 100 Hz). In this example, each response remained potentiated for 60 min after the final burst stimulation. B, input strength-response curves (SRCs) for the A-PS and S-EPSP (left graph) and EPSP-population spike (E-S) curves (right graph) before and after the burst stimulations recorded in A. The magnitudes of the A-PS response before and after LTP were recorded at the times indicated by a and b in A; the values for the S-EPSP were recorded at the times indicated by c and d in A. In the left graph, the SRCs of the S-EPSP and A-PS 60 min after the fifth burst stimulation (LTP, ▴) are shifted to the left compared with those recorded 20 min before the first burst stimulation (control, •). Lines were fitted by eye. In the right graph, the A-PS is plotted against the S-EPSP produced in response to different stimulus strengths. The E-S curve obtained 60 min after the final burst (LTP, ▴) shows a leftward shift compared with that recorded 20 min before the first burst stimulation (control, •), i.e. E-S potentiation. Lines were fitted by third order polynomial curves. C, time course of the mean percentage change in the S-EPSP (○, n = 5) and A-PS (•, n = 5). LTP in both the S-EPSP and A-PS was saturated by delivery of the third burst stimulation. Values are means ±s.e.m. in this and subsequent time courses.

Figure 3. Effects of adenosine A1 and A2 receptor antagonists on LTP induction.

Figure 3

Open in a new tab

A, time course of LTP in the S-EPSP (○) and A-PS (•) as a percentage of pre-tetanic control levels obtained from 8 experiments. Three sequential bursts (upward arrows, 100 Hz, 5 pulses) resulted in the summation of potentiation in both the S-EPSP and A-PS to induce LTP. After each burst, the increase in the S-EPSP was always smaller than that in the A-PS, leading to E-S potentiation. B, effects of the adenosine A1 receptor antagonist 8-CPT on LTP induced by repetitive burst stimuli. Time course of the mean percentage change in the S-EPSP (○, n = 8) and A-PS (•, n = 8). Three burst stimuli were delivered in the presence of 1 μm 8-CPT (horizontal bars); the first burst stimulation saturated LTP in S-EPSP and A-PS, and the LTP was maintained for 60 min after the third burst. C, effects of the adenosine A2 receptor antagonist CP-66713 on LTP induction. Application of 10 μm CP-66713 (horizontal bars) inhibited LTP induction in the S-EPSP (○, n = 6), but not in the A-PS (•, n = 6).

Figure 5. Effects of adenosine A1 and A2 receptor antagonists on DP.

Figure 5

Open in a new tab

A, time course of DP induced in the standard solution. LTP was induced by tetanus (upward arrow, 100 Hz, 100 pulses). After a further 20 min, LFS (horizontal bars, 1 Hz, 200 pulses) was repeated three times at intervals of 20-25 min, reducing LTP in the S-EPSP (○) and A-PS (•) towards pre-tetanic control levels. B, effects of the adenosine A1 receptor antagonist on DP. Application of 1 μm 8-CPT during LFS (upper horizontal bars) inhibited DP in both the S-EPSP and A-PS. C, effects of the adenosine A2 receptor antagonist on DP. Application of 10 μm CP-66713 during LFS (upper horizontal bars) facilitated DP in the S-EPSP and inhibited DP in the A-PS. E-S potentiation was enhanced by repetitive LFS. LFSs were applied in the presence of the standard solution (n = 6), 1 μm 8-CPT (n = 6) or 10 μm CP-66713 (n = 5).

To assess the change in the E-S relationship during LTP or DP, we plotted the EPSP-population spike (E-S) curves, in which the values for the S-EPSP and A-PS in response to the different stimulus strengths used for the SRCs were plotted against the corresponding S-EPSPs (Figs 4B and 6B). The E-S relationship was fitted by third order polynomial curves using graph software for Macintosh PC (Cricket Graph version 1.3.2; Computer Associates International Incorporated, Islandia, NY, USA).

Figure 4. Effects of adenosine A1 and A2 receptor antagonists on E-S relationship during LTP.

Figure 4

Open in a new tab

A, representative SRCs for S-EPSP and A-PS 60 min after the third burst stimulation (LTP, ▴) compared with those recorded 20 min before the first burst stimulation (control, •). The left, middle, and right graphs, respectively, show typical examples of the SRCs before, and after, LTP induced in the presence of the standard solution, 1 μm 8-CPT and 10 μm CP-66713. The A-PS and the S-EPSP recorded before, and after, LTP are indicated as a and c, and b and d, respectively. Lines were fitted by eye. B, E-S curves 20 min before the first burst stimulation (control, •) and 60 min after the third burst stimulation (LTP, ▴). The left, middle and right graphs, respectively, show typical examples of the E-S curves before and after LTP induced in the presence of the standard solution, 1 μm 8-CPT, or 10 μm CP-66713. The S-EPSP and the A-PS in response to the different stimulus strengths shown in A were converted into the relationship between the S-EPSP (transverse axis) and the A-PS (vertical axis). Lines were fitted by third order polynomial curves. A shift to the left, or E-S potentiation, can be seen after LTP in the absence (Standard solution) and in the presence of the adenosine A1 and A2 antagonists (8-CPT and CP-66713). C, the change in the input-output relationship was determined as the ratio of the percentage change in the A-PS relative to that in the S-EPSP (A-PS/S-EPSP ratio). This ratio was measured 15-20 min after each burst (First, Second or Third burst) or 50-60 min after the third burst (Final level) in the presence of the standard solution (n = 8), 1 μm 8-CPT (n = 8) or 10 μm CP-66713 (n = 6). In these and subsequent histograms, the values are expressed as means ±s.e.m.*P < 0.05, **P < 0.01, significant difference (Student's two-tailed t test) between control (Standard solution) and CP-66713-treated slices.

Figure 6. Effects of adenosine A1 and A2 receptor antagonists on E-S relationship during DP.

Figure 6

Open in a new tab

A, SRCs of the S-EPSP and A-PS 20 min before (control, •), 20 min after the tetanus (LTP, ▴), and 60 min after the LFS (DP, □). The left, middle, and right graphs, respectively, show typical examples of the SRCs before and after tetanus, and after the LFS applied in the presence of the standard solution, 1 μm 8-CPT and 10 μm CP-66713. The A-PS before and after tetanus, and after the LFS, is indicated by a, b and c, respectively, and the S-EPSP by d, e and f, respectively. B, E-S curves 20 min before the first burst stimulation (control, •), 20 min after the tetanus (LTP, ▴) and 60 min after the LFS (DP, □). The left, middle, and right graphs, respectively, show typical examples of the E-S curves before and after tetanus, and after the LFS applied in the presence of the standard solution, 1 μm 8-CPT and 10 μm CP-66713. The S-EPSP and the A-PS in response to the different stimulus strengths shown in A were converted into the relationship between the S-EPSP (transverse axis) and the A-PS (vertical axis). Lines were fitted by third order polynomial curves. Subsequent LFSs in the standard solution, or in the presence of 8-CPT, returned the tetanus-induced leftward shift of the E-S curve towards the pre-tetanic control level. However, subsequent LFSs in CP-66713 induced a further shift to the left, or E-S potentiation. C, the A-PS/S-EPSP ratio measured 15-20 min after tetanus (LTP), 15-20 min after each LFS (First, Second or Third LFS) and 50-60 min after the third LFS (Final level). The LFSs were applied in the presence of the standard solution (n = 6), 1 μm 8-CPT (n = 6) or 10 μm CP-66713 (n = 5). **P < 0.01, significant difference between DP in control and CP-66713-treated slices.

The changes in responses following burst stimulation (LTP) or LFS (DP) were calculated as follows: (1) the percentage change in response after establishment of LTP, [(y/x) × 100]; and (2) the percentage change in LTP response after LFS [(z/x) × 100], where x, y and z are, respectively, the average value 10-20 min prior to the delivery of HFS (a tetanus or a burst stimulation), that 15-20 min after each HFS or 50-60 min after the final HFS, and that 15-20 min after each LFS or 50-60 min after the final LFS.

In general, LTP is associated with a greater increase in the population spike than can be accounted for by potentiation of the EPSP. To statistically measure the E-S relationship in LTP or DP, the ratio (percentage change in A-PS/percentage change in S-EPSP; A-PS/S-EPSP ratio) was determined 15-20 min after each HFS and 50-60 min after the final HFS ((y/x) for A-PS/(y/x) for S-EPSP), or 15-20 min after each LFS and 50-60 min after the final LFS ((z/x) for A-PS/(z/x) for S-EPSP). The final level of the A-PS/S-EPSP ratio in LTP (Fig. 4C) or DP (Fig. 6C), respectively, was determined from (b/a)/(d/c), as shown in Fig. 4A, or from (c/a)/(f/d), as shown in Fig. 6A.

The percentage changes in S-EPSP and A-PS and the change in the A-PS/S-EPSP ratio during LTP or DP were compared between control slices and slices perfused with either 1 μm 8-CPT or 10 μm CP-66713. The test reagents were applied 10 min before the tetanus and replaced by the standard solution just after the tetanus, or applied 3-5 min before, and during, LFS, then replaced by the standard solution just after the end of LFS. In addition, three bursts were given at 20 min intervals in the presence of the N-methyl-D-aspartate (NMDA) glutamate receptor antagonist d,L-2-amino-5-phosphonovalerate (AP5, 10 μm), and changes in S-EPSP and A-PS measured 50-60 min after the third burst stimulation. 8-CPT and AP5 were supplied by Research Biochemicals Inc. (Natic, MA, USA), while CP-66713 was supplied by Pfizer Inc. (Groton, CT, USA).

All values are given as means ±s.e.m. (%); n refers to the number of slices. The results were analysed for statistical significance (P < 0.05 or < 0.01) using Student's t test (two-tailed).

RESULTS

Bidirectional neural plasticities in hippocampal CA1 neurons

We induced homosynaptic LTP in CA1 neurons in the standard solution using HFS, then reduced it using LFS. Field EPSP and population spike responses were simultaneously recorded in the area of the activated synapses in the stratum radiatum and in the cell body layer of the hippocampal CA1 region, respectively. Figure 1 shows the time course of a typical example of bidirectional changes in responses (A) and sample waveforms (B). Single burst stimulation (5 pulses at 100 Hz, indicated by the arrows in Fig. 1A) of the Schaffer collateral-commissural pathway resulted in an increase in the S-EPSP and in the A-PS. When three bursts were applied at 20 min intervals, summation of potentiation of the responses was seen. A train of LFS (200 pulses at 1 Hz, horizontal bars in Fig. 1A), subsequently applied to the same input pathway, reduced the potentiation of both responses. Potentiation of both the S-EPSP and A-PS was reversed to near the pre-tetanic control level by three LFS trains at intervals of 20 min. Using the same protocol, the same bidirectional changes in responses were seen in all slices tested in the standard solution (n = 4). Bidirectional synaptic plasticities have also been reported in vivo in hippocampal CA1 neurons (Heynen et al. 1996).

LTP induced by burst stimulations in hippocampal CA1 neurons

As shown in Fig. 1A, each single burst stimulation, composed of five pulses at 100 Hz, induced a small potentiation in responses in hippocampal CA1 neurons. In the standard solution, sequential delivery of bursts resulted in summation of these small potentiations to a large LTP. To characterize such changes in response, five bursts were applied at intervals of 20 min and the magnitude of the potentiation measured. An example of the time course of LTP is shown in Fig. 2A. Sequential delivery of burst stimulations to the CA1 inputs evoked a marked increase in both the S-EPSP and A-PS, which was maintained for more than 1 h.

In the representative slice shown, the SRCs for A-PS and S-EPSP (left graph in Fig. 2B) measured 60 min after the fifth burst (▴) were shifted to the left compared with the control SRCs (•). The right graph of Fig. 2B shows the EPSP-population spike (E-S) curves, in which the S-EPSP and the A-PS in response to the different stimulus strengths plotted in the left graph are converted into the E-S relationship. A larger potentiation in population spike for a given EPSP size is called E-S potentiation and is expressed as a leftward shift in the E-S curve. The representative E-S curve shown in the right graph of Fig. 2B was shifted to the left, indicating that E-S potentiation was induced following sequential delivery of five burst stimulations.

A summary of the results from five experiments is shown in Fig. 2C. Single burst stimulation of the CA1 inputs in slices in standard solution evoked a robust increase in the S-EPSP and A-PS. Following the first, second and third bursts, the S-EPSP increased after each burst by 23.5, 12.3 and 14.2 %, respectively, and the A-PS by 26.5, 22.0 and 22.5 % (mean values, n = 5). Fifteen to 20 min after the third burst, the S-EPSP and the A-PS were 150.0 ± 7.5 and 171.0 ± 12.9 %, respectively, of the original control levels (n = 5, means ±s.e.m.). However, delivery of a fourth and fifth burst did not cause any further significant increase in the S-EPSP and the A-PS, which were, respectively, 153.6 ± 12.6 and 173.0 ± 17.6 % of control levels 50-60 min after the fifth burst (n = 5). This suggests that burst stimulation, repeated three times in the standard solution, saturates the potentiation of responses in both the S-EPSP and A-PS.

We then applied three bursts to the CA1 neuron inputs at intervals of 20 min and induced LTP in the standard solution. The mean LTP of S-EPSP and A-PS for eight slices is shown in the time course plot (Fig. 3A). In standard solution, the S-EPSP and A-PS were 150.0 ± 6.0 and 167.9 ± 9.6 % of control levels, respectively, 15-20 min after the third burst (n = 8); 50-60 min after the third burst, these values were not significantly different at 148.2 ± 4.0 and 166.0 ± 8.0 % of control levels (n = 8), indicating that the LTP was stable for at least 60 min after the third burst. Since these values were also not significantly different from those measured 50-60 min after a fifth burst of stimulation (n = 5; Fig. 2C), we concluded that, in CA1 neurons, burst stimulation repeated three times in the standard solution saturates potentiation in both the S-EPSP and A-PS. The LTP in both the A-PS and S-EPSP induced in the standard solution is also shown in the representative SRCs (Fig. 4A, Standard solution). The SRCs for A-PS and S-EPSP, measured 60 min after the tetanus (▴), were shifted to the left compared with the control SRCs (•).

In general, LTP is associated with a greater increase in the A-PS than can be accounted for by potentiation of the EPSP, resulting in a leftward shift of the E-S curve. In eight slices tested (Fig. 4B, Standard solution), the leftward shift of the E-S curve after the final burst showed potentiation of the E-S component. To statistically evaluate the E-S relationship in LTP, we measured the A-PS/S-EPSP ratio 15-20 min after each burst stimulation or 50-60 min after the third burst. For LTP induced in the standard solution, the A-PS/S-EPSP ratio was 1.03 ± 0.039 15-20 min after the first burst stimulation, and increased to 1.17 ± 0.081 50-60 min after the third burst (n = 8; Fig. 4C). After each burst stimulation, E-S potentiation was induced and was enhanced by the delivery of a further burst.

The LTP induced by the delivery of burst stimulation was dependent on activation of NMDA glutamate receptors- Ca2+ channels, as LTP induction in both the S-EPSP and A-PS was blocked in the presence of the NMDA receptor antagonist AP5 (10 μm; data not shown). When three bursts were applied at intervals of 20 min to six slices in the presence of 10 μm AP5 and the S-EPSP and A-PS measured 50-60 min after the third burst, the respective values were 102.2 ± 4.1 and 106.0 ± 4.3 % of the control level.

Effects of an adenosine A1 receptor antagonist on LTP

Adenosine A1 receptor antagonists are known to enhance the excitability of CA1 hippocampal neurons (Dunwiddie et al. 1981). In the present study, transiently increased responses were seen when 1 μm 8-CPT was applied to slices for 10 min while delivering the test stimuli at 0.05 Hz; the responses started to increase almost 5 min after the beginning of 8-CPT perfusion and declined to the control level within 50-60 min. Fifteen to 20 min after wash-out of 8-CPT, the S-EPSP and A-PS were, respectively, 145.6 ± 6.1 and 160.0 ± 8.3 % of the control level (n = 7); however, 50-60 min after removal of 8-CPT, these values were significantly reduced (P < 0.01) to 111.6 ± 4.0 and 105.7 ± 4.3 %, respectively, of the control level (n = 7; data not shown).

When 1 μm 8-CPT was applied during a burst stimulation, LTP induction in CA1 neurons was facilitated in both the S-EPSP and A-PS. Figure 3B shows the average time course plots for eight experiments, normalized to the mean values obtained during the 10 min prior to the initial 8-CPT application. The first burst stimulation delivered in the presence of 8-CPT increased the S-EPSP and A-PS to 170.9 ± 6.0 and 186.1 ± 10.0 %, respectively, of the control level (n = 8); both increases were significantly greater (P < 0.01) than those seen after the first burst stimulation delivered in the standard solution. However, in the presence of 8-CPT, very little additional increase in the S-EPSP and A-PS was seen following a second or third burst, the values measured 50-60 min after the third burst being 189.0 ± 9.6 and 190.7 ± 12.2 %, respectively, of the control level (n = 8). This indicates that, in the presence of an adenosine A1 receptor antagonist, potentiation in both the S-EPSP and A-PS was almost saturated by the first burst of stimulation.

If the effect of 8-CPT was maintained and was included in the level of response 50-60 min after the third burst, the percentage increase in LTP would be overestimated. Taking this into consideration (an 11.6 and 5.7 % increase in the S-EPSP and A-PS, respectively), the magnitude of the saturated LTP can be re-calculated as 178.0 ± 9.7 % for the S-EPSP and 185.2 ± 12.3 % for the A-PS (n = 8), the former still being significantly larger (P < 0.05) than the value of 148.2 ± 4.0 % seen for the S-EPSP of the LTP in the standard solution. This indicates that the level of LTP saturation was increased in the S-EPSP by application of an adenosine A1 receptor antagonist. The representative SRCs showed a leftward shift in both the S-EPSP and A-PS 60 min after the final bursts (Fig. 4A, 8-CPT), also indicating that LTP was induced in both. Thus, we conclude that activation of adenosine A1 receptors inhibited LTP induction of the field EPSP by decreasing the magnitude of the potentiation for a single burst stimulation and by decreasing the level of LTP saturation.

The E-S curve of a typical example showed a leftward shift 60 min after the final bursts (Fig. 4B, 8-CPT), indicating that, in CA1 neurons, LTP induced in the presence of 8-CPT is accompanied by E-S potentiation. However, the A-PS/S-EPSP ratio 15-20 min after each burst stimulation in the eight slices tested was not significantly different from that seen in the standard solution (Fig. 4C). The A-PS/S-EPSP ratio measured 50-60 min after the third burst in the eight slices was 1.10 ± 0.071, which is not significantly different from the value seen in the absence of 8-CPT (Fig. 4C). This indicates that the adenosine A1 receptor antagonist did not affect the development of E-S potentiation during LTP. We therefore suggest that, in CA1 neurons, activation of A1 receptors attenuates LTP in the field EPSP without affecting the E-S relationship.

Effects of an adenosine A2 receptor antagonist on LTP

When 10 μm CP-66713, an adenosine A2 receptor antagonist, was applied to slices for 10 min while delivering the test stimuli at 0.05 Hz, no change in responses was seen for 50-60 min: the S-EPSP and A-PS 15-20 min after wash-out of CP-66713 were, respectively, 100.1 ± 6.2 and 102.6 ± 6.8 % of the control level, while the values 50-60 min after wash-out were 101.1 ± 3.2 and 104.1 ± 3.8 % (n = 6; data not shown).

When three bursts were delivered in the presence of 10 μm CP-66713, LTP was induced in the population spike, but not in the field EPSP. Figure 3C shows the time course plots for the S-EPSP and A-PS for six slices, normalized to the mean value obtained during the 10 min prior to the initial CP-66713 application. In the presence of CP-66713, the first, second and third bursts resulted in changes of 10.8, 8.7 and -2.6 % in the S-EPSP and of 50.5, 17.1 and 8.9 % in the A-PS (mean values, n = 6). Fifteen to 20 min after the first burst stimulation, the S-EPSP and A-PS were 110.8 ± 2.4 and 130.4 ± 3.8 %, respectively, of the control level (n = 6), representing a significant decrease in the magnitude of the potentiation in the S-EPSP (P < 0.01), without a significant change in the A-PS compared with the control LTP. Fifty to 60 min after the third burst, the S-EPSP remained at 111.3 ± 4.6 % of the control level, while the A-PS increased to 172.5 ± 10.3 % of the control level (n = 6), representing a significant decrease in the magnitude of LTP in the S-EPSP (P < 0.01), without a significant change in the A-PS compared with the control LTP. These results indicate that application of the adenosine A2 receptor antagonist blocked LTP induction of S-EPSP by decreasing the magnitude of potentiation for a single burst stimulation and by decreasing the level of LTP saturation.

The SRCs of a typical example (Fig. 4A, CP-66713) revealed that the leftward shift of the SRC for the S-EPSP was inhibited, while the SRC for the A-PS was shifted to the left when the tetanus was delivered in the presence of CP-66713. This result also indicates that the adenosine A2 receptor antagonist blocked LTP induction of S-EPSP. We therefore suggest that, in CA1 neurons, activation of adenosine A2 receptors facilitates LTP induction of the field EPSP by increasing the magnitude of potentiation for a single burst stimulation and by increasing the level of LTP saturation.

The E-S curve of a typical example (Fig. 4B, CP-66713) showed a leftward shift 60 min after the final bursts, indicating that E-S potentiation was induced by the sequential delivery of burst stimulations in the presence of CP-66713. The A-PS/S-EPSP ratio was then measured in the six slices to statistically evaluate E-S potentiation in LTP. The A-PS/S-EPSP ratio was dramatically increased after burst stimulations applied in the presence of CP-66713, the first, second and third bursts causing an increase of 0.35, 0.05 and 0.22, respectively (n = 6; Fig. 4C). The A-PS/S-EPSP ratio 50-60 min after the third burst was 1.63 ± 0.12 (n = 6), significantly higher (P < 0.01) than the value seen in the standard solution (Fig. 4C). This indicates that the adenosine A2 receptor antagonist significantly enhanced E-S potentiation during LTP.

Depotentiation of LTP in hippocampal CA1 neurons

In hippocampal CA1 neurons, LFS has been shown to reverse homosynaptic LTP in both the field EPSP and the population spike (depotentiation of LTP; DP) (Fujii et al. 1991). In six slices, LFS (200 pulses at 1 Hz) was applied three times at 20-25 min intervals after tetanus (100 Hz, 100 pulses). The summarized data, presented in Fig. 5A, show that DP summation was observed in the S-EPSP (○) and A-PS (•). The magnitudes of the tetanus-induced LTP were 151.0 ± 7.1 and 165.1 ± 7.9 % of the pre-tetanic control level for the S-EPSP and A-PS, respectively (n = 6; Fig. 5A). The first, second and third LFS trains, respectively, reduced the S-EPSP by 23.0, 5.6 and 4.0 %, and the A-PS by 24.6, 14.6 and 7.6 %, respectively (mean values, n = 6; Fig. 5A). The final level of the responses measured 50-60 min after the end of the third LFS was 118.1 ± 4.0 and 118.4 ± 5.8 % of the pre-tetanic control level for the S-EPSP and A-PS, respectively (n = 6). The representative SRCs (Fig. 6A, Standard solution) revealed that the leftward shift in the A-PS was reversed after LFS, consistent with the change in the S-EPSP.

In CA1 neurons, DP induced in the standard solution is accompanied by a reduction of the E-S potentiation in the tetanus-induced LTP. Subsequent LFSs returned the tetanus-induced leftward shift of the E-S curve towards the pre-tetanic control level (Fig. 6B, Standard solution). The A-PS/S-EPSP ratio measured 15-20 min after the tetanus was 1.10 ± 0.054 (n = 6; Fig. 6C). The first, second and third LFSs changed this ratio by -0.05, 0.14 and 0.02, respectively (n = 6; Fig. 6C), and the final value measured 50-60 min after the end of the third LFS was 0.99 ± 0.047 (n = 6; Fig. 6C), indicating definite cancellation of tetanus-induced E-S potentiation.

Effects of an adenosine A1 receptor antagonist on DP of LTP

The effects of an adenosine A1 receptor antagonist on DP in hippocampal CA1 neurons is shown in the time course of the A-PS and S-EPSP (Fig. 5B) and in the representative SRCs (Fig. 6A, 8-CPT). When 1 μm 8-CPT was applied during LFS, DP in both the field EPSP and population spike was attenuated (Fig. 5B). The tetanus-induced LTP was 148.6 ± 5.2 and 171.9 ± 5.7 % of pre-tetanic control levels for the S-EPSP and A-PS, respectively (n = 6). Each time following LFS and 8-CPT wash-out, both the S-EPSP and A-PS returned almost to the pre-established LTP levels (Fig. 5B). The responses measured 50-60 min after the third LFS remained at 148.3 ± 6.0 and 159.6 ± 2.5 % of the pre-tetanic control levels for the S-EPSP and A-PS, respectively (n = 6), both of which were significantly greater (P < 0.01) than the values seen in the absence of 8-CPT.

Taking into account the remaining effect of 8-CPT included in the response 50-60 min after the third LFS, the percentage level of the responses measured 50-60 min after the third LFS can be re-calculated as 137.9 ± 5.9 and 154.2 ± 2.4 % for the S-EPSP and A-PS, respectively (n = 6), both of which were still significantly greater (P < 0.05 and < 0.01, respectively) than the values seen in the absence of 8-CPT; the former value does not differ significantly from the pre-established LTP level. The representative SRCs (Fig. 6A, 8-CPT) show that LFS applied in the presence of 8-CPT had little effect on the leftward shift in the SRCs established in the A-PS and S-EPSP after tetanus. These results indicate that, in CA1 neurons, the A1 receptor antagonist blocked DP in the S-EPSP (see Fujii et al. 1997b).

In CA1 neurons, DP induced in the presence of 8-CPT is accompanied by a reduction of E-S potentiation in the tetanus-induced LTP. When 1 μm 8-CPT was applied to the slices during LFS (Fig. 6B, 8-CPT), the tetanus-induced leftward shift of the E-S curve was returned towards the pre-tetanic control conditions. The A-PS/S-EPSP ratio 15-20 min after tetanus was 1.17 ± 0.076, and was reduced after sequential delivery of LFS to 1.10 ± 0.061 (n = 6; Fig. 6C). However, the A-PS/S-EPSP ratios measured 15-20 min after each LFS were not significantly different from the values seen in the standard solution (Fig. 6C) and the A-PS/S-EPSP ratio 50-60 min after the third LFS was 1.10 ± 0.059 (n = 6), not significantly different from the value seen in the absence of 8-CPT (Fig. 6C). These results indicate that application of the A1 receptor antagonist during LFS did not affect the E-S relationship in DP. We therefore suggest that, in CA1 neurons, activation of A1 receptors facilitates DP in the field EPSP without affecting the E-S relationship.

Effects of an adenosine A2 receptor antagonist on DP of LTP

The effects of CP-66713 on DP are shown in time course plots of the A-PS and S-EPSP (Fig. 5C) and the representative SRCs (Fig. 6A, CP-66713). When 10 μm CP-66713 was applied to the slices during LFS, DP was facilitated in the field EPSP, but attenuated in the population spike (Fig. 5C). The tetanus-induced LTP was 147.3 ± 6.2 and 169.7 ± 6.3 % of the pre-tetanic control level for the S-EPSP and A-PS, respectively (n = 5). In the presence of CP-66713, the first, second and third LFS trains, respectively, reduced the S-EPSP by 21.6, 13.6 and 8.7 % and the A-PS by 13.8, 3.1 and 6.6 % (n = 5). The final responses measured 50-60 min after the end of the third LFS were 104.1 ± 2.2 and 146.0 ± 6.0 % of the pre-tetanic control levels for the S-EPSP and A-PS, respectively (n = 5), the former showing a significant decrease (P < 0.05), and the latter a significant increase (P < 0.01), compared with the values for the control DP. The SRCs (Fig. 6A, CP-66713) show that application of CP-66713 during LFS maintained the leftward shift of the SRC of the A-PS, but returned the SRC of the S-EPSP to the pre-tetanic control level. The result that CP-66713, applied during LFS, facilitates DP in the S-EPSP indicates that activation of adenosine A2 receptors inhibits DP in the EPSP in CA1 neurons.

The A2 receptor antagonist, by inhibiting DP in the A-PS, but facilitating that in the S-EPSP, consequently increased E-S potentiation. As shown in Fig. 6B, tetanus applied in the standard solution induced LTP in both the A-PS and S-EPSP, leading to a leftward shift of the E-S curve. A further leftward shift of the E-S curve was induced by application of 10 μm CP-66713 during LFS (Fig. 6B, CP-66713). The A-PS/S-EPSP ratio measured 20 min after the tetanus was 1.16 ± 0.075 (n = 5). When 10 μm CP-66713 was applied during LFS, delivery of three LFSs at 20-25 min intervals increased this ratio by 0.08, 0.14 and 0.068, respectively (n = 5; Fig. 6C). The A-PS/S-EPSP ratio 50-60 min after the third LFS was 1.41 ± 0.082 (n = 5; Fig. 6C), significantly greater (P < 0.01) than the value seen in the absence of the antagonist (Fig. 6C). These results indicate that application of the A2 receptor antagonist during LFS enhanced E-S potentiation during DP. We therefore suggest that, in CA1 neurons, activation of A2 receptors attenuates DP in the field EPSP decreasing the E-S potentiation during DP.

DISCUSSION

LTP and DP both have a synaptic component and an E-S component represented, respectively, as the change in the S-EPSP and the A-PS/S-EPSP ratio. The effects of A1 and A2 receptor antagonists on the S-EPSP and the A-PS/S-EPSP ratio in LTP and DP are summarized in Fig. 7A, with enhancement and inhibition indicated, respectively, by upward- and downward-pointing arrows and no change indicated by a horizontal bar. Since receptor antagonists were used in these experiments, the effects of activation of each receptor can be indicated by an arrow in the opposite direction, as shown in Fig. 7B. Thus, it can be logically expected that activation of A1 receptors will inhibit LTP and enhance DP in the synaptic component without any significant change in the E-S component, while activation of A2 receptors will enhance LTP and decrease DP in the synaptic component, while significantly reducing E-S potentiation.

Figure 7. Effects of antagonism or activation of adenosine receptors on LTP and DP.

Figure 7

Open in a new tab

A, summary of the effects of A1 and A2 receptor antagonists on the S-EPSP and the A-PS/S-EPSP ratio in LTP or DP in hippocampal CA1 neurons. B, summary of the effects of endogenous adenosine, acting via A1 or A2 receptors, on the S-EPSP and the A-PS/S-EPSP ratio in LTP or DP in hippocampal CA1 neurons. Enhancement and inhibition of each factor is indicated, respectively, by upward- and downward-pointing arrows and no change is indicated by a horizontal bar.

Possible roles of activation of adenosine receptors in neuronal plasticities in hippocampal CA1 neurons

Figure 8 shows a schematic representation of one possible role of activation of adenosine receptors, on the EPSP, the population spike and their relationship during LTP and DP in CA1 neurons. Since the A1 receptor antagonist significantly increased the size of LTP saturated in the S-EPSP (Fig. 3B), activation of A1 receptors reduces the level of LTP saturation in the EPSP. However, the A1 receptor antagonist did not affect the A-PS/S-EPSP ratio in LTP and DP (Fig. 7A) and, therefore, activation of A1 receptors also reduces the level of LTP saturation in the population spike in parallel with that in the EPSP (Fig. 8A). Thus, we suggest (Fig. 8A) that, in CA1 neurons, activation of A1 receptors reduces the range of potentiation of the EPSP and the population spike, during which process LTP induction is inhibited, but DP facilitated; adenosine also attenuates potentiation of both the input (EPSP) and output (population spike) of CA1 neurons and levels off neuronal excitability.

Figure 8. Possible roles of activation of adenosine receptors in the modulation of the synaptic and E-S components during LTP and DP in hippocampal CA1 neurons.

Figure 8

Open in a new tab

The synaptic and E-S components are represented as the change in the EPSP and the A-PS/S-EPSP ratio in Fig. 7B, respectively. A, activation of A1 receptors reduces the level of LTP saturation in the EPSP (EPSP saturation level). Since the A-PS/S-EPSP ratio is not affected, the level of LTP saturation in the population spike (PS saturation level) is reduced. Thus, the range of potentiation from baseline to saturation level is reduced in both the EPSP and population spike. In the narrowed range, LTP induction is inhibited, but DP facilitated, in both the EPSP and population spike. B, activation of A2 receptors increases the level of LTP saturation and widens the range of potentiation in the EPSP, within which LTP induction is facilitated and DP is inhibited. Until saturation of LTP in the EPSP, fine regulation of the population spike in response to changes in the EPSP could be possible due to the decrease in E-S potentiation. HFS, delivery of high-frequency stimulation; LFS, delivery of low-frequency stimulation. An increase or decrease in response is indicated, respectively, by the small upward- and downward-pointing arrows.

Since the A2 receptor antagonist significantly decreased the size of LTP saturated in the S-EPSP (Fig. 3C), activation of the A2 receptors increases the saturation level of LTP in the EPSP (Fig. 8B). Thus, we suggest (Fig. 8B) that activation of A2 receptors increases the range of the potentiation of the EPSP (the input of CA1 neurons) during which LTP induction is facilitated while DP is inhibited.

In contrast, since the A2 receptor antagonist significantly increased the A-PS/S-EPSP ratio during LTP and DP (Fig. 7A), activation of A2 receptors reduces the E-S potentiation during LTP and DP (Fig. 8B). This decrease in E-S potentiation might play a role, e.g. as a compensation function for the input-output relationship. If E-S potentiation remained high after LTP induction, LTP in the population spike would be saturated by the increase in the EPSP, and a decrease in the ratio might therefore prevent this. In synapses in which A2 receptors are activated, fine regulation of the output in response to changes in input could be possible, due to the decrease in E-S potentiation during synaptic potentiation.

Possible roles of activation of adenosine receptors in the mechanisms of LTP and DP in the synaptic component

It is widely believed that an increase in postsynaptic Ca2+ concentration ([Ca2+]i) through NMDA receptor-Ca2+ channels is a necessary step in the induction of LTP (Bliss & Collingridge, 1993). In our study, activation of the NMDA receptor was shown to be involved in the burst-induced LTP, since this LTP was sensitive to 10 μm AP5 applied during the theta-burst stimulation. Furthermore, an increase in [Ca2+]i through voltage-gated Ca2+ channels (VGCCs) is thought to contribute to LTP induction.

Ito et al. (1995) examined the involvement of P-type VGCCs in the AP5-sensitive LTP in guinea-pig hippocampal CA1 neurons. In their study, the VGCC blocker ω-agatoxin IVA (60 nm, P-type channel blocker), which had no effect on synaptic transmission, suppressed 78 % of the theta-burst-induced LTP in the EPSP. Since activation of adenosine A2 receptors has been shown to potentiate P-type channels in the mammalian brain (Yawo & Chuhama, 1993; Mogul et al. 1993; Umemiya & Berger, 1994), it is possible that application of the A2 receptor antagonist suppressed LTP induction in the synaptic component by inhibiting P-type Ca2+ channels in CA1 neurons.

On the other hand, activation of the A1 receptor has been reported to inhibit N-type calcium channels in the mammalian brain (Yawo & Chuhama, 1993), although reduced synaptic transmission due to inhibition of calcium influx through N-type calcium channels in response to agonist-mediated adenosine A1 receptor activation (Mogul et al. 1993; Yawo & Chuhama, 1993) made it difficult to study the involvement of this type of Ca2+ channel in LTP induction. However, Zhang & Schmidt (1998) recently demonstrated that C-kinase activation, which is associated with LTP induction, uncouples A1 receptor activation from N-type Ca2+ channel inhibition. Thus, it is possible that adenosine A1 receptor activation is involved in LTP induction by inhibiting Ca2+ influx through N-type Ca2+ channels.

On the basis of adenylyl cyclase activation, adenosine receptors have been classified into A1 and A2 subtypes; activation of adenosine A1 receptors inhibits adenylyl cyclase and thereby reduces cyclic AMP formation, while activation of adenosine A1 receptors has the opposite effect (Dunwiddie & Fredholm, 1989; Lupica et al. 1990). Since the cyclic AMP pathway is thought to be one of the second messenger systems involved in the mechanisms responsible for LTP formation (Ghosh & Greenberg, 1995) at the mossy fibre-CA3 synapse (Weisskopf et al. 1994) and at the synapse in CA1 region (Blitzer et al. 1995), it is also possible that a decrease or increase in intracellular cyclic AMP levels, via activation of A1 or A2 receptors at pre- or postsynaptic sites, affects the signalling mechanisms leading to LTP induction of the synaptic component.

A rise in postsynaptic calcium concentration through NMDA receptor-Ca2+ channels is involved in DP in hippocampal CA1 neurons as application of 50 μm AP5 during LFS inhibited LTP reversal (Fujii et al. 1991). However, the results of the present study indicate that activation of A1 or A2 receptors respectively facilitated or attenuated DP in the EPSP. Although the mechanism of DP is still unknown, activation of adenosine A1 or A2 receptors during LFS, which modulates Ca2+ entry through the VGCCs or intracellular cyclic AMP levels, may be involved in the signalling mechanisms leading to DP in the synaptic component.

Possible roles of adenosine in the mechanism of E-S potentiation

Two hypotheses regarding the mechanism of the E-S component have been proposed. The first involves a reduced inhibitory drive: E-S potentiation may occur when the inhibitory synapses are depressed following tetanus (Kairiss et al. 1987; Chavez-Noriega et al. 1989; Tomasulo et al. 1991; Tomasulo & Ramirez, 1993). The alternative hypothesis is a modification of the firing threshold of the excitatory neurons (Taube & Schwartzkroin, 1988; Hess & Gustafsson, 1990).

A2 receptor activation facilitates whole-cell calcium currents by acting predominantly through P-type calcium channels in the synapses of inhibitory interneurons, while A1 receptor activation predominantly inhibits N-type calcium channels in the somata of the interneurons (Umemiya & Berger, 1994). Although the involvement of these VGCCs in the mechanism of E-S potentiation is still unknown, it is possible that A2 receptor activation following burst stimulation enhances Ca2+ influx through P-type calcium channels at the synapses of inhibitory interneurons, resulting in long-term enhancement of inhibitory drive and decreased E-S potentiation during LTP or DP.

An alternative suggestion is that A2 receptor activation following burst stimulation at the somata triggers a long-term decrease in the postsynaptic firing threshold and decreased E-S potentiation. It has been reported that P-type calcium channels contribute to the action potential-evoked [Ca2+]i increase in the somata of hippocampal CA1 pyramidal cells, while another VGCC contributes to the action potential-evoked [Ca2+]i increase in the proximal apical dendrite (Markram et al. 1995; Miura et al. 1997). However, it is still unknown whether Ca2+ influx at the somata through P-type calcium channels triggers a long-term decrease in the postsynaptic firing threshold.

The different types of modification of the S-EPSP and the A-PS/S-EPSP ratio caused by A1 or A2 receptor antagonists (Fig. 7A) may indicate the independent mechanisms which underlie the synaptic and E-S components in LTP, but further studies are required in this area.

Concluding remarks

Input activities to postsynaptic neurons determine which adenosine receptors are activated. This idea is based on the following four facts. (1) Depending on input activity, ATP and adenosine derivatives are released from the presynaptic terminals into the synaptic cleft (Schubert et al. 1976; White, 1978; Wieraszko et al. 1989). (2) The affinity of adenosine for A1 receptors is higher (nanomolar range) than that for A2 receptors (micromolar range) (Bruns et al. 1980; Londos et al. 1980). (3) The distribution density of A1 receptors is higher than that of A2 receptors (Fastbom et al. 1987; Sebastião & Ribeiro, 1992; Cunha et al. 1994). (4) Activation of A2 receptors in hippocampal CA1 neurons reduces the activity of A1 receptors due to cross-talk between A2 and A1 receptors (Cunha et al. 1994).

It is therefore possible that, depending on the degrees of release, up-take and diffusion, the concentration of adenosine and its derivatives will be higher in the high input synapses and lower in the low input synapses. Because the affinity of adenosine for A1 receptors is higher than that for A2 receptors, both A1 and A2 receptors can be activated in the high input synapses, and A1 receptors can be activated in the low input synapses. In the high input synapses, A2 receptors may be predominantly activated due to cross-talk between A1 and A2 receptors.

Our results have shown that activation of adenosine A1 receptors blocks LTP induction, but facilitates DP, while activation of adenosine A2 receptors facilitates LTP induction, but inhibits DP in the synaptic component. Since the input gradient may correspond to the activation of the different receptor types, endogenous adenosine and its derivatives, acting via A1 receptors, may reduce synaptic potentiation by attenuating LTP induction and facilitating DP in low input synapses. On the other hand, endogenous adenosine, acting via A2 receptors, may maintain synaptic potentiation by facilitating LTP induction and by inhibiting DP in high input synapses. Thus, it is possible that the modulatory role of endogenous adenosine is to make a clear contrast between synaptic plasticities through the input activity-dependent organization.

References

  1. Andersen P, Sundberg SH, Sveen O, Swann JW, Wigström H. Possible mechanisms for long-lasting potentiation of synaptic transmission in hippocampal slices from guinea-pigs. The Journal of Physiology. 1983;302:463–482. doi: 10.1113/jphysiol.1980.sp013256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arai A, Kessler M, Lynch G. The effects of adenosine on the development of long-term potentiation. Neuroscience Letters. 1990;119:41–44. doi: 10.1016/0304-3940(90)90750-4. [DOI] [PubMed] [Google Scholar]
  3. Bashir ZI, Collingridge GL. An investigation of depotentiation of long-term potentiation in the CA1 region of the hippocampus. Experimental Brain Research. 1994;100:437–443. doi: 10.1007/BF02738403. [DOI] [PubMed] [Google Scholar]
  4. Blitzer RD, Wong T, Nouranifar R, Lyengar R, Landau EM. Postsynaptic cAMP pathway gates early LTP in hippocampal CA1 region. Neuron. 1995;15:1403–1414. doi: 10.1016/0896-6273(95)90018-7. [DOI] [PubMed] [Google Scholar]
  5. Bliss TVP, Collingridge GL. A synaptic model of memory: long-term potentiation in the hippocampus. Nature. 1993;361:31–39. doi: 10.1038/361031a0. [DOI] [PubMed] [Google Scholar]
  6. Bliss TVP, Gardner-Medwin AR. Long-lasting potentiation of synaptic transmission in the dentate area of the unanaesthetized rabbit following stimulation of the perforant path. The Journal of Physiology. 1973;232:357–374. doi: 10.1113/jphysiol.1973.sp010274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bliss TVP, Lømo T. Long-lasting potentiation of synaptic transmission in dentate area of the anaesthetized rabbit following stimulation of the perforant path. The Journal of Physiology. 1973;232:331–356. doi: 10.1113/jphysiol.1973.sp010273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bruns RF, Daly JW, Snyder SH. Adenosine receptors in brain membranes: Binding of N6-cyclohexyl[3H]adenosine and 1,3-dimethyl-8-[3H]phenyl-xanthine. Proceedings of the National Academy of Sciences of the USA. 1980;77:5547–5551. doi: 10.1073/pnas.77.9.5547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bruns RF, Fergus JH, Badger EW, Bristol JA, Santay LA, Hartman JD, Hays SJ, Huang CC. Binding of the A1-selective adenosine antagonist 8-cyclopentyl-1,3-dipropylxanthine to rat brain membranes. Naunyn-Schmiedeberg's Archives of Pharmacology. 1987;335:59–63. doi: 10.1007/BF00165037. [DOI] [PubMed] [Google Scholar]
  10. Chavez-Noriega LE, Bliss TVP, Halliwell JV. The EPSP-spike (E-S) component of long-term potentiation in the rat hippocampal slice is modulated by GABAergic but not cholinergic mechanisms. Neuroscience Letters. 1989;104:58–64. doi: 10.1016/0304-3940(89)90329-7. [DOI] [PubMed] [Google Scholar]
  11. Collingridge GL, Herron CE, Lester RAJ. Frequency dependent N-methyl-D-aspartate receptor-mediated synaptic transmission in rat hippocampus. The Journal of Physiology. 1988;399:301–312. doi: 10.1113/jphysiol.1988.sp017081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cunha RA, Johansson B, van der Ploeg I, Sebastião AM, Ribeiro JA, Fredholm BB. Evidence for functionally important adenosine A2a receptors in the rat hippocampus. Brain Research. 1994;649:208–216. doi: 10.1016/0006-8993(94)91066-9. [DOI] [PubMed] [Google Scholar]
  13. Doyle CA, Cullen WK, Rowan MJ, Anwyl R. Low-frequency stimulation induces homosynaptic depotentiation but not long-term depression of synaptic transmission in the adult anaesthetized and awake rat hippocampus in vivo. Neuroscience. 1997;77:75–85. doi: 10.1016/s0306-4522(96)00427-7. [DOI] [PubMed] [Google Scholar]
  14. Dunwiddie TV, Fredholm BB. Adenosine A1 receptors inhibit adenylate cyclase activity and neurotransmitter release and hyperpolarize pyramidal neurons in rat hippocampus. Journal of Pharmacology and Experimental Therapeutics. 1989;249:31–37. [PubMed] [Google Scholar]
  15. Dunwiddie TV, Hoffer BJ, Fredholm BB. Alkylxanthines elevate hippocampal excitability. Evidence for a role of endogenous adenosine. Naunyn-Schmiedeberg's Archives of Pharmacology. 1981;316:326–330. doi: 10.1007/BF00501365. [DOI] [PubMed] [Google Scholar]
  16. Fastbom J, Pazos A, Palacios JM. The distribution of adenosine A1-receptors and 5′-nucleotidase in the brain of some commonly used experimental animals. Neuroscience. 1987;22:813–826. doi: 10.1016/0306-4522(87)92961-7. [DOI] [PubMed] [Google Scholar]
  17. Fujii S, Kuroda Y, Ito K-I, Kaneko K, Sasaki H, Kato H. The roles of endogenous adenosine, acting via A1 and A2 receptors, in the induction and reversal of long-term potentiation in guinea pig hippocampal slices CA1 neurons. In: Okada Y, editor. The Role of Adenosine in the Nervous System. Amsterdam: Elsevier Science; 1997a. pp. 127–134. [Google Scholar]
  18. Fujii S, Saito K, Miyakawa H, Ito K, Kato H. Reversal of long-term potentiation (depotentiation) induced by tetanus stimulation of the input to CA1 neurons of guinea pig hippocampal slices. Brain Research. 1991;555:112–122. doi: 10.1016/0006-8993(91)90867-u. [DOI] [PubMed] [Google Scholar]
  19. Fujii S, Sekino Y, Kuroda Y, Sasaki H, Ito K-I, Kato H. 8-Cyclopentyltheophylline, an adenosine A1 receptor antagonist, inhibits the reversal of long-term potentiation in hippocampal CA1 neurons. European Journal of Pharmacology. 1997b;331:9–14. doi: 10.1016/s0014-2999(97)01024-8. [DOI] [PubMed] [Google Scholar]
  20. Fujii S, Wakizaka A, Sekino Y, Kuroda Y, Ito K-I, Miyakawa H, Kato H. Adenosine A2 receptor antagonist facilitates the reversal of long-term potentiation (depotentiation) of evoked postsynaptic potentials but inhibits that of population spikes in hippocampal CA1 neurons. Neuroscience Letters. 1992;148:148–150. doi: 10.1016/0304-3940(92)90826-s. [DOI] [PubMed] [Google Scholar]
  21. Ghosh A, Greenberg M. Calcium signaling in neurons: Molecular mechanisms and cellular consequences. Science. 1995;268:239–247. doi: 10.1126/science.7716515. [DOI] [PubMed] [Google Scholar]
  22. Hess G, Gustafsson B. Changes in the field excitatory postsynaptic shape induced by tetanization in the guinea-pig hippocampal slice. Neuroscience. 1990;37:61–69. doi: 10.1016/0306-4522(90)90192-7. [DOI] [PubMed] [Google Scholar]
  23. Heynen AJ, Abraham WC, Bear MF. Bidirectional modification of CA1 synapses in the adult hippocampus in vivo. Nature. 1996;381:163–166. doi: 10.1038/381163a0. [DOI] [PubMed] [Google Scholar]
  24. Ito K-I, Miura M, Furuse H, Zhixiong C, Kato H, Yasutomi D, Inoue T, Mikoshiba K, Kimura T, Sakakibara S, Miyakawa H. Voltage-gated Ca2+ channel blockers, ω-AgaIVA and Ni2+, suppress the induction of θ-burst induced long-term potentiation in guinea-pig hippocampal CA1 neurons. Neuroscience Letters. 1995;183:112–115. doi: 10.1016/0304-3940(94)11127-5. [DOI] [PubMed] [Google Scholar]
  25. Kairiss E, Abraham W, Bilkey D, Goddard G. Field potential evidence for long-term potentiation of feed-forward inhibition in the rat dentate gyrus. Brain Research. 1987;401:87–94. doi: 10.1016/0006-8993(87)91167-x. [DOI] [PubMed] [Google Scholar]
  26. Kessey K, Trommer BL, Overstreet LS, Ji T, Mogul DJ. A role for adenosine A2 receptors in the induction of long-term potentiation in the CA1 region of rat hippocampus. Brain Research. 1997;756:184–190. doi: 10.1016/s0006-8993(97)00148-0. [DOI] [PubMed] [Google Scholar]
  27. Larson J, Xiao P, Lynch G. Reversal of LTP by theta frequency stimulation. Brain Research. 1993;600:97–102. doi: 10.1016/0006-8993(93)90406-d. [DOI] [PubMed] [Google Scholar]
  28. Londos C, Cooper DMF, Wolff J. Subclasses of external adenosine receptors. Proceedings of the National Academy of Sciences of the USA. 1980;77:2551–2554. doi: 10.1073/pnas.77.5.2551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Lupica CR, Cass AW, Zahniser NR, Dunwiddie TV. Effects of adenosine A2 receptor agonist CGS 21680 on in vitro electrophysiology, cAMP formation and dopamine release in rat hippocampus and striatum. Journal of Pharmacology and Experimental Therapeutics. 1990;252:1134–1141. [PubMed] [Google Scholar]
  30. Manzoni OJ, Manabe T, Nicoll RA. Release of adenosine by activation of NMDA receptors in the hippocampus. Science. 1994;265:2098–2101. doi: 10.1126/science.7916485. [DOI] [PubMed] [Google Scholar]
  31. Markram HP, Johannes H, Sakmann B. Dendritic calcium transients evoked by single back-propagating action potentials in rat neocortical pyramidal neurons. The Journal of Physiology. 1995;485:1–20. doi: 10.1113/jphysiol.1995.sp020708. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Miura M, Yoshioka M, Miyakawa H, Kato H, Ito K-I. Properties of calcium spikes revealed during GABAA antagonism in hippocampal CA1 neurons from guinea pig. Journal of Neurophysiology. 1997;78:2269–2279. doi: 10.1152/jn.1997.78.5.2269. [DOI] [PubMed] [Google Scholar]
  33. Mogul D, Adams ME, Fox AP. Differential activation of adenosine receptors decreases N-type but potentiates P-type Ca2+ current in hippocampal CA3 neurons. Neuron. 1993;10:327–334. doi: 10.1016/0896-6273(93)90322-i. [DOI] [PubMed] [Google Scholar]
  34. O'Kane EM, Stone TW. Interaction between adenosine A1 and A2 receptor-mediated responses in the rat hippocampus in vitro. European Journal of Pharmacology. 1998;362:17–25. doi: 10.1016/s0014-2999(98)00730-4. [DOI] [PubMed] [Google Scholar]
  35. Sarges R, Howard HR, Browne RG, Lebel LA, Seymour PA, Koe BK. 4-Amino[1,2,4]triazolo[4,3-a]quinoxalines. A novel class of potent adenosine receptor antagonists and potential rapid-onset antidepressants. Journal of Medicinal Chemistry. 1990;33:2240–2254. doi: 10.1021/jm00170a031. [DOI] [PubMed] [Google Scholar]
  36. Schubert P, Lee K, West M, Deadwyler S, Lynch G. Stimulation-dependent release of 3H-adenosine derivatives from central axon terminals to target neurones. Nature. 1976;260:541–542. doi: 10.1038/260541a0. [DOI] [PubMed] [Google Scholar]
  37. Sebastião AM, Ribeiro JA. Evidence for the presence of excitatory A2 adenosine receptors in the rat hippocampus. Neuroscience Letters. 1992;138:41–44. doi: 10.1016/0304-3940(92)90467-l. [DOI] [PubMed] [Google Scholar]
  38. Sekino Y, Ito K, Miyakawa Y, Kato H, Kuroda Y. Adenosine (A2) antagonist inhibits induction of long-term potentiation of evoked synaptic potentials but not of the population spike in hippocampal CA1 neurons. Biochemical and Biophysical Research Communications. 1991;18:1010–1014. doi: 10.1016/0006-291x(91)92037-k. [DOI] [PubMed] [Google Scholar]
  39. Sekino Y, Koyama I. Selective inhibition of homosynaptic depression in a tetanized pathway by an adenosine a1 blocker in the CA1 region of rat hippocampal slice. Neuroscience Letters. 1992;148:109–113. doi: 10.1016/0304-3940(92)90816-p. [DOI] [PubMed] [Google Scholar]
  40. Stäubli U, Lynch G. Stable depression of potentiated synaptic responses in the hippocampus with 1–5 Hz stimulation. Brain Research. 1990;513:113–118. doi: 10.1016/0006-8993(90)91096-y. [DOI] [PubMed] [Google Scholar]
  41. Taube JS, Schwartzkroin PA. Mechanisms of long-term potentiation: EPSP/spike dissociation, intradendritic recordings, and glutamate sensitivity. Journal of Neuroscience. 1988;8:1632–1644. doi: 10.1523/JNEUROSCI.08-05-01632.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tomasulo R, Levy W, Steward O. LTP-associated EPSP/spike dissociation in the dentate gyrus: GABAergic and non-GABAergic components. Brain Research. 1991;561:27–34. doi: 10.1016/0006-8993(91)90745-h. [DOI] [PubMed] [Google Scholar]
  43. Tomasulo R, Ramirez J. Activity-mediated changes in feed-forward inhibition in the dentate commissural pathway: relationship to EPSP/spike dissociation in the converging perforant path. Journal of Neuroscience. 1993;69:165–173. doi: 10.1152/jn.1993.69.1.165. [DOI] [PubMed] [Google Scholar]
  44. Umemiya M, Berger AJ. Activation of adenosine A1 and A2 receptors differentially modulates calcium channels and glycinergic synaptic transmission in rat brainstem. Neuron. 1994;13:1439–1446. doi: 10.1016/0896-6273(94)90429-4. [DOI] [PubMed] [Google Scholar]
  45. Van Calker D, Müller M, Hamprecht B. Adenosine regulates via two different types of receptors, the accumulation of cyclic AMP in cultured brain cells. Journal of Neurochemistry. 1979;33:999–1005. doi: 10.1111/j.1471-4159.1979.tb05236.x. [DOI] [PubMed] [Google Scholar]
  46. Weisskopf MG, Castillo PE, Zalutsky RA, Nicoll RA. Mediation of hippocampal mossy fiber long-term potentiation by cAMP. Science. 1994;265:1878–1882. doi: 10.1126/science.7916482. [DOI] [PubMed] [Google Scholar]
  47. White TD. Release of ATP from a synaptosomal preparation by elevated extracellular K+ and by veratridine. Journal of Neurochemistry. 1978;30:329–336. doi: 10.1111/j.1471-4159.1978.tb06534.x. [DOI] [PubMed] [Google Scholar]
  48. Wieraszko A, Goldsmith G, Seyfried TN. Stimulation-dependent release of adenosine triphosphate from hippocampal slices. Brain Research. 1989;485:244–250. doi: 10.1016/0006-8993(89)90567-2. [DOI] [PubMed] [Google Scholar]
  49. Yawo H, Chuhama N. Preferential inhibition of ω-conotoxin-sensitive presynaptic Ca2+ channels by adenosine autoreceptors. Nature. 1993;365:256–258. doi: 10.1038/365256a0. [DOI] [PubMed] [Google Scholar]
  50. Zhang C, Schmidt JT. Adenosine A1 receptors mediate retinotectal presynaptic inhibition: uncoupling by c-kinase and role in LTP during regeneration. Journal of Neurophysiology. 1998;79:501–510. doi: 10.1152/jn.1998.79.2.501. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES