Adenosine A2 receptors modulate hippocampal synaptic transmission via a cyclic-AMP-dependent pathway

Pergamon

PII:

Neuroscience Vol. 84, No. 1, pp. 59–69, 1998 Copyright ? 1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(97)00504-6

ADENOSINE A2 RECEPTORS MODULATE HIPPOCAMPAL SYNAPTIC TRANSMISSION VIA A CYCLIC-AMP-DEPENDENT PATHWAY K. KESSEY* and D. J. MOGUL*†‡ *Department of Neurobiology & Physiology, Northwestern University, Evanston, IL 60208, U.S.A. †Department of Biomedical Engineering, Northwestern University, 2145 Sheridan Road, Evanston, IL 60208, U.S.A. Abstract––Blockade of adenosine A2 receptors has been shown to significantly reduce the level of tetanus-induced long-term potentiation in area CA1 of rat hippocampus [Kessey K. et al. (1997) Brain Res. 756, 184–190; Sekino Y. et al. (1991) Biochem. biophys. Res. Commun. 181, 1010–1014]. In the present study, the effects of A2 receptor activation and blockade on the modulation of normal synaptic transmission and tetanus-induced long-term potentiation were examined at the Schaffer–CA1 synapse in rat hippocampal slices. A2 receptor activation reversibly enhanced synaptic transmission evoked by low-frequency test pulses as measured by the dendritic field excitatory postsynaptic potential. In the presence of A1 receptor blockade, A2 activation further enhanced the excitatory postsynaptic potential, while A2 receptor blockade resulted in a reversible decrease of the excitatory postsynaptic potential. The A2a receptor agonist, CGS21680, had no effect on the excitatory postsynaptic potential, suggesting that tonic activation of A2b receptors contributes to synaptic transmission under normal physiological conditions. Furthermore, we investigated the contribution of A2 receptors to the level of tetanus-induced long-term potentiation. Under control conditions, a single tetanus potentiated the excitatory postsynaptic potential by 63.5% relative to baseline 30 min post-tetanus. In contrast, tetanus-induced long-term potentiation during A2 blockade was 21.3%. A2 receptor activation increased the level of tetanus-induced long-term potentiation to 90.2%. Because A2 receptors are known to stimulate cyclic-AMP accumulation, the possible involvement of cyclic-AMP was examined. Forskolin, a direct adenylate cyclase activator, and 8-bromo-cyclic-AMP, a membrane-permeable analog of cyclic-AMP, were able to reconstitute tetanusinduced long-term potentiation during A2 receptor blockade; however, the inactive analog 1,9-dideoxyforskolin had no effect, indicating that the effects of A2 activation on synaptic transmission were mediated largely through the regulation of intracellular cyclic-AMP. Because A1 receptors exert an opposing effect on synaptic transmission relative to A2 receptors, these results suggest that the stoichiometry of A1 versus A2 receptor activation appears to play an important role in the modulation of normal synaptic transmission and long-term potentiation in the CA1 region of the hippocampus. ? 1998 IBRO. Published by Elsevier Science Ltd. Key words: purinergic, CA1, long-term potentiation, adenosine receptors, neurotransmission.

neurotransmitter release,10,11,14 and possibly inhibition of postsynaptic excitability.19,41 In contrast, activation of adenosine A2 receptors has been directly shown to exert an excitatory influence on normal synaptic transmission in the hippocampus by A2 activation,38 or more indirectly demonstrated via adenosine-mediated changes induced in a dosedependent manner.34,35 Long-term potentiation (LTP) is a long-lasting increase in synaptic efficacy that can be evoked by high-frequency stimulation of afferent pathways.2 A2 receptor activation has been shown to be an important component necessary for the induction of tetanus-induced LTP. Furthermore, blockade of A2 receptors has been shown to significantly reduce tetanus-induced LTP in the hippocampus.24,40 Because extracellular adenosine is found in regions throughout the brain, we sought to explore what role this purine may play, working through selective

Increases in the rate of neuronal firing can significantly elevate levels of extracellular adenosine in the brain.32 In the hippocampus, exogenously applied adenosine and adenosine analogs acting through adenosine receptors have been shown to exert both inhibitory and excitatory effects on cellular and synaptic activity.1,25,37,41 The inhibitory effects of adenosine appear to be mediated by adenosine A1 receptors via reduction of presynaptic calcium influx49 and ‡To whom correspondence should be addressed. Abbreviations: ACSF, artificial cerebrospinal fluid; AMPA, á-amino-3-hydroxy-5-methyl-4-isoxazolepropionate; 8Br-cAMP, 8-bromo-cyclic-AMP; cAMP, cyclic-AMP; CPT, 8-cyclopentyl-1,3-dimethylxanthine; DMPX, 3,7dimethyl-1-propargylxanthine; DPMA, N6-[2-(3,5dimethyoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine; EPSP, excitatory postsynaptic potential; LTP, long-term potentiation; NMDA, N-methyl--aspartate; PKA, cAMP-dependent protein kinase. 59

60

K. Kessey and D. J. Mogul

adenosine receptors, on synaptic transmission. Specifically, the present work was designed to further examine the influence of A2 receptors on neurotransmission. Our results indicate that the degree of A2 receptor activation can exert a potent modulatory influence on the levels of both normal synaptic transmission elicited by low-frequency test pulses as well as tetanus-induced LTP. Furthermore, because changes in intracellular cyclic-AMP (cAMP) levels have been demonstrated to be critical for LTP induction,4,8,18,42 we investigated whether the linkage between the levels of A2 receptor activation and final LTP expression were mediated by a cAMPdependent cascade. EXPERIMENTAL PROCEDURES

Hippocampal slices were obtained from 22- to 35-day-old Sprague–Dawley rats (Harlan Sprague Dawley, Inc., Indianapolis, IN, U.S.A.). Animals were anesthetized with isoflurane by inhalation then decapitated. The brain was rapidly excised, hemisected at the interhemispheric fissure and placed in cold (4)C) artificial cerebrospinal fluid (ACSF) containing (in mM): NaCl 124, NaHCO3 24, -glucose 10, MgSO4 1.3, NaH2PO4 1.25, KCl 3, CaCl2 2.4 and gassed with 95% O2/5% CO2 (pH 7.4). Both hippocampi were dissected free and 400-µm transverse slices were cut using a vertical tissue chopper (Stoelting). Slices were stored for at least 1 h in the bubbled ACSF at room temperature (approximately 22)C) and transferred as needed to a submersion chamber maintained at 30)C. Extracellular recordings were made using glass microelectrodes (2–4 MÙ) pulled from borosilicate glass capillaries and filled with NaCl (2 M). Recording electrodes for field excitatory postsynaptic potentials (EPSPs) were placed in the stratum radiatum of the CA1 region. Orthodromic stimulation was provided by placement of a twisted Tefloncoated platinum bipolar electrode in the stratum radiatum near the CA3/CA1 border. Low-frequency stimulation, defined as ‘‘normal’’ synaptic transmission, was delivered at 0.033 Hz. Tetanic stimulation consisted of a 1-s train at 100 Hz. Constant-current stimulation (100–400 µA) pulses were delivered by a current stimulus isolator. For each slice experiment the stimulus intensity and duration were selected as those required to achieve approximately 50% of the maximum EPSP slope response. The stimulus duration was not changed within an experiment. Recording signals were amplified using a DAM80 amplifier with active headstage (World Precision Instruments, Sarasota, FL, U.S.A.), filtered between 0.1 Hz and 3 kHz, and were sampled at 10 kHz by a Digidata 1200 (Axon Instruments, Foster City, CA, U.S.A.). Experiments were controlled and analysed using the Axobasic software (Axon Instruments). The maximum slope of the initial negative deflection for each extracellular field measurement was used to quantify each dendritic EPSP. Materials Adenosine receptor pharmacological agents [CGS21680, 8-cyclopentyl-1,3-dimethylxanthine (CPT), 3,7-dimethyl-1propargylxanthine (DMPX) and N6-[2-(3,5-dimethyoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine (DPMA)] and cAMP agents [8-bromo-cAMP (8-Br-cAMP) and 1,9dideoxy-forskolin] were purchased from Research Biochemicals International, Inc. (Natick, MA, U.S.A.) with forskolin purchased from Sigma (St Louis, MO, U.S.A.). CPT was dissolved in 0.1 N NaOH, DMPX and 8-BrcAMP in ACSF, and CGS21680, DPMA, forskolin and 1,9-dideoxy-forskolin in dimethylsulfoxide. Final concen-

trations of dimethylsulfoxide never exceeded 0.01%, a concentration at which no effects of the solvent alone were observed. RESULTS

A2 receptors contribute to the enhancement of synaptic responses during low-frequency stimulation The initial series of experiments examined to what degree selective activation of A2 receptors altered synaptic transmission. Figure 1A shows the results of an experiment in which dendritic EPSPs in CA1 were measured in response to low-frequency test pulses during superfusion of the A2 agonist DPMA (10 nM). The Ki for DPMA is 142 nM at A1 receptors and 4.4 nM at A2 receptors.5 In this experiment, as well as all other slices tested (n=7), DPMA increased the EPSPs recorded by a mean of 37.8% (see Table 1). This effect of DPMA on the EPSP was mostly blocked by superfusion with the A2 receptor antagonist DMPX (10 µM), yielding a mean increase of only 3.4% (P<0.005; comparison of DPMA effect on EPSP with and without DMPX), supporting the involvement of A2 receptors in the augmentation of synaptic transmission (Fig. 1B). DMPX shows approximately equal activity at both A2a and A2b receptor subtypes.39 Because no selective agonists for the A2b receptor exist, we examined the involvement of A2a receptors using the highly selective A2a agonist CGS21680 on synaptic transmission under similar conditions. Figure 1C shows that CGS21680 (20 nM) had no effect on the EPSP (n=6), suggesting that the increase in synaptic transmission in response to DPMA was due to A2b receptor activation. Although exogenous A2 activation can potentiate synaptic transmission, the question remains as to whether any portion of the baseline control response is due to basal A2 activation. Figure 2 shows experimental protocols in which the apparent contributions of A2 receptors to normal synaptic transmission were investigated. Because the A2 blocker, DMPX, also shows significant affinity for the A1 receptor (Ki=11&3 µM for A2 vs Ki=45&4 µM for A1),44 we applied either DMPX or DPMA during continuous A1 blockade. In the presence of CPT (0.1 µM), exposure to DMPX (10 µM) resulted in a net decrease in the EPSP (P<0.01, t-test) during lowfrequency test pulses (Fig. 2A). In contrast, under similar conditions of A1 receptor inhibition, DPMA (10 nM) enhanced the EPSP slope (Fig. 2B) over that observed in CPT alone (see Table 1; P<0.01, t-test, comparison of the effects on EPSP of A2 activation vs blockade in CPT). A2 receptors modulate the level of expression of tetanus-induced long-term potentiation Blockade of A2 receptors blocks or reduces tetanus-induced LTP; thus, we sought to quantify the role of these receptors on LTP modulation. A1 receptor activation itself exerts an inhibitory effect on

Adenosine A2 receptors modulate neurotransmission

61

Fig. 1. Activation of A2 (but not A2a) receptors increases synaptic transmission under conditions of low-frequency stimulation (0.033 Hz) in intact hippocampal slices. (A) A typical experiment showing the increase in synaptic transmission when a hippocampal slice was bathed in control solution containing the A2 agonist DPMA. After a stable 15-min baseline in control medium, slices were perfused for 10 min with DPMA (10 nM; shown as solid bar) followed by washout. Under these conditions, A2 receptor activation resulted in a reversible increase in field EPSP slope (see Table 1). (B) Application of DPMA in the presence of the A2 antagonist, DMPX (10 µM), produces no significant change in the average baseline EPSP response (n=6; bars indicate S.E.M.). (C) Using the same stimulus protocol as in A, exposure of a slice to the A2a agonist, CGS21680 (20 nM), did not alter the field EPSP.

synaptic transmission;12 hence, the effects of A2 receptor inhibition and activation on tetanus-induced LTP were examined during complete A1 receptor

blockade. After a stable baseline recording of at least 10 min in control ACSF with low-frequency test pulses (Fig. 3), slices were superfused with ACSF

62

K. Kessey and D. J. Mogul

Table 1. The change in the excitatory postsynaptic potential during low-frequency stimulation in response to various A1 and A2 receptor agents

Experimental conditions DPMA (10 nM) only DPMA (10 nM) in the presence of continuous CPT (0.1 µM) perfusion DMPX (10 µM) in the presence of continuous CPT (0.1 µM) perfusion CGS21680 (10–30 nM)

Changes in EPSP slope: mean&S.E.M. (n) +37.8&10.9% (7) +15.2&5.6% (7) "8.1&4.9% (8) no change (6)

For experiments using DPMA and CGS21680 alone, baseline was taken to be pre-drug EPSP level (at least 10 min in control solution). When experiments were performed in combinations of CPT and DPMA or DMPX, the baseline was taken to be the EPSP slope in the presence of CPT alone. For each experimental condition, the mean change in EPSP slope was calculated and presented as shown. Numbers in parentheses indicate the number of slices in each experimental group.

containing CPT alone or in combination with an A2 agonist or antagonist until a new stable EPSP slope was obtained. In order to prevent saturating the level of LTP which can result from any prior drug-induced augmentation of the field EPSP and to normalize responses, the stimulus intensity was decreased to reduce the EPSP back to the control level. Following tetanic stimulation (shown as vertical arrows), the resulting changes in EPSPs were monitored for at least 30 min. LTP was measured as the percentage change in EPSP slope 30 min following tetanization compared to the average response during the 10 min prior to tetanus. Control experiments in which a tetanus was delivered in the presence of CPT (0.1 µM) alone yielded a level of LTP of 163.5&9.1% (mean&S.E.M., n=5) of baseline (Fig. 3A). In comparison, Fig. 3B displays the results of a typical experiment used to determine the effects of A2 receptor inhibition on LTP. Hippocampal slices were perfused with ACSF containing CPT (0.1 µM) and DMPX (10 µM). Tetanus-induced LTP under these conditions was 121.3&1.9% (n=6). In contrast, when A2 receptors were activated by perfusing slices with CPT (0.1 µM) and DPMA (10 nM), the EPSP slope was 190.2&10.5% (n=6) of baseline 30 min following tetanic stimulation (Fig. 3C). Table 2 summarizes the results of A2 receptor modulation on tetanus-induced LTP. These results suggest that activation of A2 receptors during A1 blockade can significantly modulate the level of tetanus-induced LTP in the hippocampus (P<0.0001, one-way ANOVA).

cAMP formation,17,28,45 we examined the possibility that the effects of A2 receptors on synaptic transmission were mediated by a cAMP cascade (Fig. 4). If the effects of these receptors on the expression of LTP are dependent on activation of a cAMP pathway, then one would expect that agents that elevate the intracellular levels of this second messenger should restore the level of expression of LTP during A2 receptor blockade. To test this hypothesis, the effects of either forskolin, 1,9-dideoxy-forskolin or 8-Br-cAMP in the presence of CPT and DMPX on LTP were examined. Forskolin, at a typical concentration (50 µM) used in other studies to stimulate cAMP,6,7,29,48 significantly restored the level of potentiation (142.4%&3.1%; n=5) compared to the combination of CPT and DMPX (P<0.002, t-test). To exclude the possibility that this restoration of LTP was through mechanisms independent of cAMP, we performed similar experiments using 1,9dideoxy-forskolin (50 µM), the inactive analog which does not activate cAMP. In contrast to the effects observed with forskolin, the level of LTP attained with 1,9-dideoxy-forskolin did not differ significantly from CPT and DMPX (119.0&2.7%; n=4; P>0.5). The effects of 8-Br-cAMP, a membrane-permeable analog of cAMP, were also studied to further establish a role for cAMP in mediating the observed effects of A2 receptors on LTP expression. As in the previously described experiments, slices were exposed to 8-Br-cAMP in the presence of CPT and DMPX. Following tetanic stimulation, the level of LTP attained was significantly higher than was the case with CPT and DMPX alone (181.2&4.3%; n=5; P<0.0001). Figure 5 compares these results showing mean and standard error bars. Because both forskolin and 8-Br-cAMP reconstitute LTP in the presence of CPT and DMPX but the inactive analog 1,9-dideoxy-forskolin had no significant effect on the level of LTP, these results strongly suggest that the effects of A2 receptors on mediating the final level of LTP expression involve a cAMP cascade. DISCUSSION

Our data yielded the following conclusions. (1) Adenosine A2 receptors contribute to normal synaptic transmission during low-frequency test pulses. (2) Under conditions of A1 receptor blockade, A2 receptor activation and blockade can significantly increase and decrease the level of LTP, respectively, following a single tetanus. (3) The observed modulatory role of A2 receptors on LTP appears to be mediated, at least in part, by their stimulatory influence on cAMP.

The modulatory effects of A2 receptors are mediated by a cyclic-AMP cascade

A2 receptors and synaptic transmission

Because cAMP appears to play a critical role in LTP8,18,36,42,50 and A2 receptor activation increases

Previously published evidence has demonstrated that endogenously produced adenosine tonically

Adenosine A2 receptors modulate neurotransmission

63

Fig. 2. A2 receptor activation potentiates the slope of field EPSPs in the absence of A1 receptor activity during low-frequency stimulation. (A) A single experiment showing the effects of A2 receptor blockade on EPSP slope during A1 receptor blockade. After a 20-min perfusion with control medium to establish a baseline response, each slice was exposed to control medium containing CPT (0.1 µM; shown as a solid bar) for 30 min, resulting in an increase in the EPSP response. In the presence of the A1 antagonist, slices were superfused with DMPX (10 µM) for 30 min followed by washout. Exposure to the A2 antagonist in the presence of continued CPT exposure resulted in a decrease in EPSPs. In all slices tested (n=8), a similar decrease in EPSP slope was observed during DMPX perfusion in the presence of CPT. (B) Under similar conditions of A1 receptor blockade, bath application of the A2 agonist DPMA (10 nM) for 20 min resulted in a further potentiation of the slope of EPSPs beyond that of CPT alone. Dashed lines in both A and B are the baseline response in CPT alone.

depresses hippocampal neurons by activating A1 receptors.15,22 It is unlikely that the inhibitory influence of these receptors is through activation of inhibitory GABAergic synapses, because adenosine26,51 and adenosine agonists14 do not appear to modulate GABA release in the hippocampus. In addition to its inhibitory effects, adenosine has also been shown to exert an excitatory influence on population spikes mediated by A2 receptors.38 Consistent with such an excitatory role, we found that activation

of A2 receptors enhances normal synaptic transmission. Our result in Fig. 2A showing that A2 blockade during low-frequency test pulses reduces the baseline EPSP suggests that tonic activation of A2 receptors by endogenous adenosine may play a modulatory role in synaptic transmission under normal physiological conditions. Although the precise mechanism underlying the contribution of A2 receptors to synaptic transmission has not yet been determined, arguments for either a pre- or

64

K. Kessey and D. J. Mogul

Fig. 3. A2 receptors modulate the level of LTP expression. In all three panels, a stable baseline EPSP was recorded during perfusion of control ACSF during low-frequency stimulation for at least 10 min. Slices were then superfused with ACSF containing CPT alone or in combination with DMPX or DPMA until a new EPSP level was obtained. The stimulus intensity was decreased to equalize pre- and post-drug EPSP levels. Following tetanic stimulation (100 Hz for 1 s; shown as vertical arrow), the resulting changes in EPSP slope were monitored for at least 30 min. LTP was measured as the percentage change in EPSP slope following tetanus compared to the average EPSP slope 10 min prior to tetanization. (A) Control experiment showing the effects of superfusion of CPT on the level of potentiation following tetanic stimulation. Exposure to CPT resulted in an LTP level of 163.5%&9.1% (mean&S.E.M., n=5). (B) When slices were bathed in solution containing both CPT and DMPX (10 µM), the level of potentiation following tetanic stimulation decreased to 121.3%&1.9% (n=6). (C) In the presence of CPT and DPMA, the level of LTP was increased to 190.2%&10.5% (n=6).

postsynaptic modification are possible. First, a presynaptic locus involving an increase in neurotransmitter release is suggested by work which shows that

A2 receptor activation enhances Ca2+ conductance through voltage-dependent Ca2+ channels in a cAMP-dependent manner in hippocampal CA3

Adenosine A2 receptors modulate neurotransmission

65

Table 2. A comparison of the magnitude of tetanus-induced long-term potentiation in the presence of the various combinations of test drugs EPSP slope (% of baseline) 30 min post-tetanus: mean&S.E.M. (n)

Experimental conditions CPT CPT CPT CPT CPT CPT

(0.1 µM) only (0.1 µM)+DMPX (0.1 µM)+DPMA (0.1 µM)+DMPX (0.1 µM)+DMPX (0.1 µM)+DMPX

(10 µM) (10 nM) (10 µM)+forskolin (50 µM) (10 µM)+1,9-dideoxy-forskolin (50 µM) (10 µM)+8-Br-cAMP (10 µM)

163.5&9.1 (5) 121.3&1.9 (6) 190.2&10.5 (6) 142.4&3.1 (5) 119.0&2.7 (4) 181.2&4.3 (5)

The mean&S.E.M. for each group experiment was calculated. LTP in the presence of a combination of CPT and DMPX or DPMA was significantly different from CPT alone (P<0.0001, one-way ANOVA). Significant differences (P values) between protocols with A2 receptor agents were compared to CPT alone using Tukey’s HSD test.

neurons.16,33 Because the excitatory effects of A2 receptors on synaptic transmission are dependent on their influence on cAMP, the possibility that A2 receptors increase transmitter release is further supported by work done by Chavez-Noriega and Stevens,6,7 which shows that elevation of cAMP increases neurotransmitter release. However, more recent evidence has pointed to an entirely postsynaptic locus for long-term changes to synaptic transmission in this region.30,49 á-Amino-3hydroxy-5-methyl-4-isoxazolepropionate (AMPA)/ kainate receptors have been shown to be phosphorylated by the cAMP-dependent protein kinase (PKA),46,47 resulting in an increase in the responsiveness of these receptors to glutamate20 and perhaps the activation of silent AMPA receptors.23,27,31 An A2 receptor-induced increase in cAMP could potentially increase phosphorylation of AMPA/kainate receptors by PKA. Hence, A2 receptors may modulate the sensitivity of AMPA/kainate receptors to glutamate through a cAMP-dependent cascade. An increase in PKA activity could also modulate Ca2+ current through voltage-dependent Ca2+ channels through phosphorylation and subsequently modulating the excitability of postsynaptic neurons.25,43 Further investigation is required to discriminate between a pre- and/or postsynaptic site of action by A2 receptors. A2 receptors, cyclic-AMP and long-term potentiation A1 and A2 receptors are known to decrease and increase, respectively, cAMP accumulation in different preparations.17,28,45 The role of cAMP as the messenger underlying the ability of A2 receptors to modulate LTP is significant because previous studies have demonstrated a potential role for cAMP in the induction of LTP in the hippocampus. In the CA1 region of the hippocampus, induction of LTP requires calcium influx primarily through N-methyl-aspartate (NMDA) receptors;9 the biochemical consequences of such an increase in intracellular Ca2+ levels include an increase in cAMP levels and a

subsequent augmentation of Ca2+ currents through voltage-dependent Ca2+ channels.8 Interestingly, activation of such high-threshold Ca2+ channels by very-high-frequency tetanus (200 Hz) has also been shown to induce LTP.21 Elevated cAMP levels following a tetanus may be important for the induction of LTP in the hippocampus, since the level of cAMP is increased 1 min, but not 30 min, following tetanusinduced LTP in the dentate gyrus.42 In CA1, a similar increase in cAMP following LTP-inducing tetanus can be suppressed by blocking NMDA receptor activity,8 which also blocks the induction of LTP.3 A critical role for cAMP in CA1 tetanus-induced early phase LTP has recently been reported, in addition to its requirement for the late phase of LTP.4,18 Forskolin and dibutyryl cAMP have been shown to increase both synaptic and cell body responses in CA1 of rat hippocampus.6,36 Tetanic stimulation occludes the increase in responses caused by dibutyryl cAMP,36 suggesting identical mechanisms for both forms of potentiation. We show that LTP in the presence of CPT, DMPX and 1,9-dideoxy-forskolin is not significantly different from LTP induced in CPT and DMPX. Both forskolin and 8-Br-cAMP restore a significant fraction of LTP under similar conditions, suggesting that cAMP accumulation occurring through either forskolin, 8-Br-cAMP or A2 activation ultimately converges on the same pathway to modulate LTP. Because the level of cAMP in CA1 neurons rises significantly following tetanization of afferent pathways and decreases to baseline levels shortly thereafter, a role for cAMP in the influence of A2 receptors in LTP expression raises the interesting possibility that the level of LTP expression is determined during or shortly following tetanic stimulation by the level of cAMP accumulated. We have shown previously that the critical time period for the effect of A2 receptor blockade on LTP is during induction and not the maintenance phase.24 The biochemical consequences of such an increase in cAMP levels could potentially include an increase in AMPA

66

K. Kessey and D. J. Mogul

Fig. 4. The modulatory effects of A2 receptors on LTP are mediated by a cAMP cascade. The potential role of a cAMP cascade was examined using forskolin (50 µM), 1,9-dideoxy-forskolin (50 µM) and 8-Br-cAMP (10 µM). (A) When slices were bathed in solution containing forskolin, CPT and DMPX, the level of LTP achieved following a single tetanus was significantly increased (142.4&3.08%; n=5) compared to that obtained in the presence of CPT and DMPX alone (P<0.002, t-test) as shown in Fig. 3B. (B) However, when slices were perfused with 1,9-dideoxy-forskolin (50 µM), the inactive analog of forskolin, the subsequent level of potentiation following a single tetanus was not significantly different from that obtained in CPT and DMPX (119.0&2.66%; n=4; P>0.5). (C) In the presence of 8-Br-cAMP (10 µM), the membrane-permeable analog of cAMP, the level of LTP obtained under similar experimental conditions was 181.2&4.26% (n=5), which was significantly higher than that observed during CPT and DMPX perfusion (P<0.0001).

Adenosine A2 receptors modulate neurotransmission

67

Fig. 5. A comparison of experiments showing the mediatory role of cAMP in the effects of A2 receptors on LTP. Bars indicate means&S.E.M. DMPX significantly reduced the level of LTP in the presence of CPT when compared to that observed in the presence of CPT alone. Both forskolin and 8-Br-cAMP were able to significantly restore the level of LTP during continuous CPT and DMPX exposure (P<0.0003 and P<0.0001 respectively, t-test) in comparison to CPT and DMPX. However, 1,9-dideoxy-forskolin had no significant effect on the level of LTP (P=0.5002, t-test). Together, these data indicate that the effects of A2 receptors on LTP are mediated by changes in intracellular cAMP levels.

receptor responsiveness to glutamate or activation of silent AMPA receptors, both of which could contribute to expression of LTP.23,27,31 In this paper we show that the influence of A2 receptors on LTP involves a cAMP-dependent cascade under conditions which excluded contributions from the adenosine A1 receptor. Although the inhibitory influence of A1 receptors on LTP has not been shown to involve a reduction in cAMP accumulation, such a mechanism is not unreasonable. Additionally, A1 receptors could potentially decrease cAMP accumulation indirectly through their inhibition of NMDA activity,13 probably by hyperpolarizing postsynaptic neurons. It has been shown previously that blockade of A2 receptors significantly inhibits induction of LTP.24,40 It is likely that both the blockade of A2 receptors with DMPX and the activation of A1 receptors by high concentrations of adenosine released as a result of high-frequency stimulation, together, contribute to the observed inhibition of tetanus-induced LTP.

CONCLUSION

Together with the cited work, our results suggest that a combination of A1 activation and A2 blockade may inhibit LTP by decreasing cAMP levels. The extent of cAMP accumulation necessary for LTP induction may depend, in part, on the balance between A1 and A2 receptor activation. This implies that, under normal conditions, relatively high concentrations of adenosine released during increased activity of afferent pathways activate A2 receptors which are able to compensate for A1 receptor inhibition of cAMP levels and subsequently contribute to the induction of LTP. Such a mechanism would be most evident under conditions in which a differential distribution of these adenosine receptors exists in hippocampal pyramidal neurons. Increases in neuronal activity as well as pathological conditions such as ischemia and epilepsy can induce an increase in levels of extracellular adenosine in the brain. Hence, the differential effects of adenosine receptor activation

68

K. Kessey and D. J. Mogul

in modulating synaptic transmission may play an important role both under normal physiological conditions as well as during diseased states.

Acknowledgements—The authors are extremely indebted to Teng Ji for her assistance in the laboratory. This work was supported by the NIH (NS31764) and the Whitaker Foundation.

REFERENCES

1. 2. 3. 4. 5.

6. 7. 8.

9. 10. 11.

12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

Alzheimer C., Rohrenbeck J. and ten Bruggencate G. (1991) Adenosine depresses induction of LTP at the mossy-fiber–CA3 synapse in vitro. Brain Res. 543, 163–165. Bliss T. V. P. and Lomo T. (1973) Long-lasting potentiation of synaptic transmission in dentate area of rabbit following stimulation of the perforant path. J. Physiol. 232, 331–356. Bliss T. V. P. and Collingridge G. L. (1993) A synaptic model of memory: long-term potentiation in the hippocampus. Nature 361, 31–39. Blitzer R. D., Wong T., Nouranifar R., Iyengar R. and Landau E. M. (1995) Postsynaptic cAMP pathway gates early LTP in the hippocampal CA1 region. Neuron 15, 1403–1414. Bridges A. J., Bruns R. F., Ortwine D. F., Priebe S. R., Szotek D. L. and Trivedi B. K. (1988) N6-[2-(3,5Dimethoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine and its uronamide derivatives. Novel adenosine agonists with both high affinity and high selectivity for the adenosine A2 receptor. J. med. Chem. 31, 1282–1285. Chavez-Noriega L. E. and Stevens C. F. (1992) Modulation of synaptic efficacy in field CA1 of the rat hippocampus by forskolin. Brain Res. 574, 85–92. Chavez-Noriega L. E. and Stevens C. F. (1994) Increased transmitter release at excitatory synapses produced by direct activation of adenylate cyclase in rat hippocampal slices. J. Neurosci. 14, 310–317. Chetkovich D. M., Gray R., Johnston D. and Sweatt J. D. (1991) N-Methyl--aspartate receptor activation increases cAMP levels and voltage gated Ca2+ channel activity in area CA1 of hippocampus. Proc. natn. Acad. Sci. U.S.A. 88, 6467–6471. Collingridge G. L., Kehl S. J. and McLennan H. (1983) Excitatory amino acids in synaptic transmission in the Schaffer collateral–commissural pathway of the rat hippocampus. J. Physiol. 334, 33–46. Corradetti R., Lo Conte G., Moroni F., Padssani M. B. and Pepeu G. (1984) Adenosine decreases aspartate and glutamate release from rat hippocampal slices. Eur. J. Pharmac. 104, 19–26. Cunha R. A., Ribeiro J. A. and Sebastiao A. M. (1994) Purinergic modulation of the evoked release of [3H]acetylcholine from the hippocampus and cerebral cortex of the rat: role of ectonucleotidases. Eur. J. Neurosci. 6, 33–42. de Mendonca A. and Ribeiro J. A. (1993) Adenosine inhibits the NMDA receptor-mediated excitatory postsynaptic potential in the hippocampus. Brain Res. 606, 351–356. de Mendonca A. and Ribeiro J. A. (1994) Endogenous adenosine modulates long term potentiation in the hippocampus. Neuroscience 62, 385–390. Dolphin A. C. and Archer E. R. (1983) Adenosine antagonist inhibits and cyclic AMP analogue enhances the release of glutamate but not GABA from slices of rat dentate gyrus. Neurosci. Lett. 43, 49–54. Dunwiddie T. V., Hoffer B. J. and Fredholm B. B. (1981) Alkyl-xanthines elevate hippocampal excitability: evidence for a role of endogenous adenosine. Naunyn-Schmiedeberg’s Arch. Pharmac. 316, 326–330. Fleming K. M. and Mogul D. J. (1997) Activation of adenosine A3 receptors potentiates a calcium current in hippocampal CA3 neurons. Neuropharmacology 36, 353–362. Fredholm B. B., Jonzon B. and Lindstrom K. (1983) Adenosine receptor mediated increases and decreases in cyclic AMP in hippocampal slices treated with forskolin. Acta physiol. scand. 117, 461–463. Frey U., Huang Y.-Y. and Kandel E. R. (1993) Effects of cAMP stimulate a late stage of LTP in hippocampal CA1 neurons. Science 260, 1661–1664. Greene R. W. and Haas H. L. (1985) Adenosine actions on CA1 pyramidal neurones in rat hippocampal slices. J. Physiol. 366, 119–127. Greengard P., Jen J., Nairn A. C. and Stevens C. F. (1991) Enhancement of glutamate response by cAMP-dependent protein kinase in hippocampal neurons. Science 253, 1135–1138. Grover L. M. and Teyler T. J. (1990) Two components of long-term potentiation induced by different patterns of afferent activation. Nature 347, 477–479. Haas H. L. and Greene R. W. (1988) Endogenous adenosine inhibits hippocampal CA1 neurones: further evidence from extra- and intracellular recording. Naunyn-Schmiedeberg’s Arch. Pharmac. 337, 561–565. Isaac J. T., Nicoll R. A. and Malenka R. C. (1995) Evidence for silent synapses: implications for the expression of LTP. Neuron 15, 427–434. Kessey K., Trommer B. L., Overstreet L., Ji T. and Mogul D. J. (1997) A role for adenosine A2 receptors in the induction of long-term potentiation in the CA1 region of rat hippocampus. Brain Res. 756, 184–190. Kullmann D. M., Perkel D. J., Manabe T. and Nicoll R. A. (1992) Ca2+ entry via postsynaptic voltage-sensitive Ca2+ channels can transiently potentiate excitatory synaptic transmission in the hippocampus. Neuron 9, 1175–1183. Lambert N. A. and Teyler T. J. (1991) Adenosine selectively depresses excitatory synaptic transmission in area CA1 of the rat hippocampus. Neurosci. Lett. 122, 50–52. Liao D., Hessler N. A. and Malinow R. (1995) Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice. Nature 375, 400–404. Londos C. and Wolff J. (1977) Two distinct adenosine-sensitive sites of adenylate cyclase. Proc. natn. Acad. Sci. U.S.A. 74, 5482–5486. Lopez-Garcia J. C., Arancio O., Kandel E. R. and Baranes D. (1996) A presynaptic locus for long-term potentiation of elementary synaptic transmission at mossy fiber synapses in culture. Proc. natn. Acad. Sci. U.S.A. 93, 4712–4717. Manabe T. and Nicoll R. A. (1994) Long-term potentiation: evidence against an increase in transmitter release probability in the CA1 region of the hippocampus. Science 265, 1888–1892.

Adenosine A2 receptors modulate neurotransmission 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

69

Maren S., Tocco G., Standley S., Baudry M. and Thompson R. F. (1993) Postsynaptic factors in the expression of long-term potentiation (LTP): increased glutamate receptor binding following LTP induction in vivo. Proc. natn. Acad. Sci. U.S.A. 90, 9654–9658. Mitchell J. B., Lupica C. R. and Dunwiddie T. V. (1993) Activity-dependent release of endogenous adenosine modulates synaptic responses in the rat hippocampus. J. Neurosci. 13, 3439–3447. Mogul D. J., Adams M. E. and Fox A. P. (1993) Differential activation of adenosine receptors decreases N-type but potentiates P-type Ca2+ current in hippocampal CA3 neurons. Neuron 10, 327–334. Nishimura S., Mohri M., Okada Y. and Mori M. (1990) Excitatory and inhibitory effects of adenosine on the neurotransmission in the hippocampal slices of guinea pig. Brain Res. 525, 165–169. Okada Y., Sakurai T. and Mori M. (1992) Excitatory effect of adenosine on neurotransmission is due to increase of transmitter release in the hippocampal slices. Neurosci. Lett. 142, 233–236. Pockett S., Slack J. R. and Peacock S. (1993) Cyclic AMP and long-term potentiation in the CA1 region of rat hippocampus. Neuroscience 52, 229–236. Sciotti V. M., Park T. S., Berne R. M. and Van Wylen D. G. L. (1993) Changes in extracellular adenosine during chemical or electrical brain stimulation. Brain Res. 613, 16–20. Sebastiao A. M. and Ribeiro J. A. (1992) Evidence for the presence of excitatory A2 adenosine receptors in the rat hippocampus. Neurosci. Lett. 138, 41–44. Sebastiao A. M. and Ribeiro J. A. (1996) Adenosine A2 receptor-mediated excitatory actions on the nervous system. Prog. Neurobiol. 48, 167–189. Sekino Y., Ito K., Miyakawa H., Kato H. and Kuroda Y. (1991) Adenosine (A2) antagonist inhibits induction of long-term potentiation of evoked synaptic potentials but not of the population spike in hippocampal CA1 neurons. Biochem. biophys. Res. Commun. 181, 1010–1014. Siggins G. R. and Schubert P. (1981) Adenosine depression of hippocampal neurons in vitro: an intracellular study of dose-dependent actions on synaptic and membrane potentials. Neurosci. Lett. 23, 55–60. Stanton P. K. and Sarvey J. M. (1985) The effect of high frequency stimulation and norepinephrine on cyclic AMP levels in normal versus norepinephrine-depleted rat hippocampal slices. Brain Res. 358, 343–348. Takahashi T. and Momiyama A. (1993) Different types of calcium channels mediate central synaptic transmission. Nature 366, 156–158. Ukena D., Shamim M., Padgett W. and Daly J. W. (1986) Analogs of caffeine with selectivity for A2 adenosine receptors. Life Sci. 39, 743–750. VanCalker D., Muller M. and Hamprecht B. (1979) Adenosine regulates via two different types of receptors, accumulation of cyclic AMP in cultured brain cells. J. Neurophysiol. 33, 999–1005. Wang L., Salter M. W. and MacDonald J. F. (1991) Regulation of kainate receptors by cAMP-dependent protein kinase and phosphotases. Science 253, 1132–1135. Wang L., Taverna F. A., Huang X., MacDonald J. F. and Hampson D. R. (1993) Phosphorylation and modulation of a kainate receptor (GluR6) by cAMP dependent protein kinase. Science 259, 1173–1175. Weisskopf M. G., Castillo P. E., Zalutsky R. A. and Nicoll R. A. (1994) Mediation of hippocampal mossy fiber long-term potentiation by cyclic AMP. Science 265, 1878–1882. Wu L. and Saggau P. (1994) Adenosine inhibits evoked synaptic transmission primarily by reducing presynaptic calcium influx in area CA1 of hippocampus. Neuron 12, 1139–1148. Wu Z. L., Thomas S. A., Villacres E. C., Xia Z., Simmons M. L., Chavkin C., Palmiter R. D. and Storm D. R. (1995) Altered behavior and long-term potentiation in type I adenylyl cyclase mutant mice. Proc. natn. Acad. Sci. U.S.A. 92, 220–224. Yoon K. W. and Rothman S. M. (1991) Adenosine inhibits excitatory but not inhibitory synaptic transmission in the hippocampus. J. Neurosci. 11, 1375–1380. (Accepted 26 September 1997)