- Email: [email protected]
Adenosine promotes neuronal recovery from reactive oxygen species induced lesion in rat hippocampal slices
Neuroscience Letters 339 (2003) 127–130 www.elsevier.com/locate/neulet
Adenosine promotes neuronal recovery from reactive oxygen species induced lesion in rat hippocampal slices Cla´udia G. Almeida, Alexandre de Mendonc¸a*, Rodrigo A. Cunha, J. Alexandre Ribeiro Laboratory of Neurosciences, Faculty of Medicine of Lisbon, Avenue Prof Egas Moniz, 1649-028 Lisbon, Portugal Received 8 October 2002; received in revised form 5 December 2002; accepted 18 December 2002
Abstract Reactive oxygen species (ROS) are believed to be involved in the pathogenesis of several neurological disorders. We now tested whether the endogenous neuroprotective substance, adenosine, attenuates the cell damage induced by ROS. In rat hippocampal slices, the xanthine oxidase (40 mU/ml) plus xanthine (1 mM) (X/XO) system produced a 27.8 ^ 7.3% (n ¼ 3) increase in ROS, measured by fluorimetry with 20 ,70 -dichlorodihydrofluorescein, a 246.9 ^ 18.4% (n ¼ 6) increase in the release of tritiated adenosine, and a decrease in synaptic transmission that fully recovered after washout. In the presence of the adenosine A1 receptor selective antagonist, 1,3-dipropyl-8cyclopentylxanthine (100 nM), X/XO induced a similar inhibition, however synaptic transmission only recovered to 70.7 ^ 5.8% of control (n ¼ 5). The blockade of A2A receptors was devoid of effect (n ¼ 4). Adenosine is released by ROS-generating systems, and attenuates the deleterious cellular consequences of ROS through A1 receptor activation. q 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Adenosine; Neuroprotection; Reactive oxygen species; Hippocampus; A1 receptors; A2A receptors; Dipropylcyclopentylxanthine
Cell damage induced by of reactive oxygen species (ROS) is presumably involved in acute neurological disorders such as stroke [5], as well as in chronic neurodegenerative disorders [2]. The possibility of attenuating the deleterious effects of free radicals on nervous tissue thus appears an interesting approach to ameliorate those neurological disorders. Adenosine is an endogenous purine with an important modulatory role on neuronal activity [10], and broad neuroprotective properties exerted in experimental models of hypoxia, ischemia, excitotoxicity and trauma [9]. However, it is not known whether adenosine can also protect nerve cells against the deleterious actions of ROS. Interestingly, the free radical generating system, hydrogen peroxide plus ferrous sulfate, causes a transient inhibition of synaptic transmission, which is blocked by the adenosine receptor antagonist, theophylline, and thus presumably due to released adenosine in hippocampal slices [13]. Another free radical generating system, xanthine plus xanthine oxidase, is also able to induce the release of purines from hippocampal slices [3]. Since the generation of ROS is able to induce the release of adenosine, we hypothesize that adenosine might represent * Corresponding author. Tel./fax: þ 351-21-7936787. E-mail address: [email protected] (A. de Mendonc¸a).
an endogenous feedback protective system to limit the damage induced by ROS. We now tested this hypothesis by studying whether endogenous adenosine could promote recovery of synaptic transmission in hippocampal slices subjected to ROS-induced damage. The experiments were performed on hippocampal slice preparations 400 mm thick, taken from male Wistar rats 5– 8 weeks old, superfused with an artificial cerebrospinal fluid (aCSF) solution composed of (mM): NaCl 124, KCl 3, NaH2PO4 1.25, MgSO4 1, CaCl2 2, NaHCO3 26, glucose 10, pH 7.4, equilibrated with a 95% O2/5%CO2 gas mixture, at 30.5 8C [21]. To induce oxidative stress the free radical generating system xanthine oxidase (XO; 40 mU/ml, Sigma) plus xanthine (X; 1 mM, Sigma), which predominantly generates the superoxide radical [12], was applied for 45 min. Another free radical generating system was also tested, FeSO4 (100 mM, applied during 30 min), which by the Fenton reaction induces the formation of the hydroxyl radical [22]. Intracellular oxidative stress was evaluated with the probe 20 ,70 -dichlorodihydrofluorescein diacetate (DCFH2-DA; Molecular Probes, USA). The hippocampal slices were loaded with 20 mM DCFH2-DA for 20 min in the dark [18], rinsed for 30 min with aCSF, and the slice fluorescence measured for 10 min (baseline), exciting the probe at 502
0304-3940/03/$ - see front matter q 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(02)01478-7
128
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
nm and measuring the emission at 550 nm, with a Fluoromax-2 spectrometer. Oxidative stress was then induced and slice fluorescence continuously monitored. Release of endogenous adenosine was measured by scintillation spectrophotometry after loading the hippocampal slices with [2 – 3H]adenosine (37 MBq/ml; 1 mCi/ml; from Amersham International, UK) [8]. Evoked field excitatory postsynaptic potentials (fEPSP) were recorded in the stratum radiatum of the CA1 area evoked by electrical stimulation of the Schaffer collateral/commissural fibers [21]. Field EPSPs (the averages of eight consecutive potentials, 1.0 –1.5 mV amplitude and 0.5 –0.8 mV.ms21 slope) were quantified as the slope of the initial phase of the potential with the LTP 2.21 software [1]. The slope data are shown as the percentage change from the average slope of the fEPSP taken during the 10 min that preceded the application of the ROS generating system. The drugs, 1,3-dipropyl-8-cyclopentylxanthine (DPCPX; from Sigma), 5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine (SCH 58261; kindly supplied by Dr E. Ongini, Schering-Plough, Milan), and DL-2-amino-5-phosphonopentanoate (AP5; from Sigma), were allowed an equilibrium period of 30 min of superfusion prior to the application of the ROS generating system. Data are expressed as mean ^ SEM values, of n experiments. Statistical analysis was performed using the two-tailed Student’s t-test, at a level of significance of P , 0:05. The intracellular ROS production induced by X/XO application was studied in the hippocampal slices by using a fluorimetric assay with the probe DCFH2-DA. There was a rapid and steady increase in fluorescence that attained a maximum of 32.8 ^ 5.1% at the end of the 45 min superfusion with X/XO (n ¼ 3, P , 0:05). In the presence of the selective adenosine A1 receptor antagonist, DPCPX (100 nM), there was a maximal increase in fluorescence levels induced by X/XO (21.1 ^ 3.8% increase, n ¼ 5, P , 0:05), which was not significantly different from that observed in control conditions. We next investigated whether the generation of ROS could elicit an increase in the extracellular levels of adenosine. Superfusion with X/XO induced an increase in the release of adenosine which had a maximum, 246.9 ^ 18.4% (n ¼ 6, P , 0:05), about 15 min after the application of X/XO, and afterwards decreased but remained above basal levels at least until 1 h after beginning X/XO application (Fig. 1). In the electrophysiological experiments, X/XO (superfused for 45 min; xanthine itself elicited 4.3 ^ 0.9% increase in fEPSP slope, n ¼ 3) caused a rapid decrease in the fEPSP slope. After 14 min the depression was apparently stable and had a value of 62.0 ^ 9.1% (n ¼ 5, P , 0:05, Fig. 2A,B). The removal of X/XO allowed the virtual recovery of synaptic transmission from the insult (to 2.7 ^ 11.4% of the fEPSP before X/XO) in usually 40 min.
When the same protocol was performed in the presence of DPCPX (100 nM), X/XO also elicited a marked decrease of fEPSP slope, 66.4 ^ 7.8% (n ¼ 5, P , 0:05). There was no significant difference between the depressions of the synaptic transmission induced by X/XO in the presence and in the absence of DPCPX. However, in the washout period, the recovery of synaptic transmission in the presence of DPCPX was only partial, reaching 70.7 ^ 5.8% of control (n ¼ 5, P , 0:05 compared with the fEPSP before X/XO). This recovery was significantly different (P , 0:05) from the virtually complete recovery observed in control conditions in the absence of DPCPX. We next investigated whether the selective adenosine A2A receptor antagonist, SCH 58261, could modify the effects of X/XO on synaptic transmission. In the presence of SCH 58261 (100 nM), superfusion with X/XO (45 min) induced a marked decrease, 96.9 ^ 2.9% (n ¼ 4, P , 0:05) of the fEPSP slope, similar to control conditions (94.5 ^ 0.5%, n ¼ 4, P , 0:05; Fig. 2C). After the removal of X/XO, synaptic transmission took about 40 min to recover to 77.4 ^ 13.3% (of the fEPSP before X/XO, n ¼ 4, P , 0:05) in the presence of SCH 58261 and to 85.4 ^ 15.7% (of the fEPSP before X/XO, n ¼ 4, P , 0:05) in control conditions. There were no significant differences either in the depression or in the recovery of fEPSP obtained in the presence or in the absence of the A2A receptor antagonist. We next tested whether the impairment in recovery after X/XO observed in the presence of DPCPX could involve the activation of the N-methyl-D -aspartate (NMDA) subtype of glutamate receptors. This was not the case, since recovery from X/XO was still decreased (70.7 ^ 4.9% of the fEPSP before X/XO, n ¼ 4, P , 0:05) in the presence of DPCPX (100 nM) plus the NMDA receptor selective antagonist, AP5 (50 mM), as compared with the recovery in the presence of DPCPX (100 nM) only (67.5 ^ 5.5% of the fEPSP before X/XO, n ¼ 4, P , 0:05, Fig. 3). We also tested another substance as a source of ROS,
Fig. 1. Effect of ROS production with X plus XO on the release of tritium from hippocampal slices labeled with [2-3H]adenosine. X (1 mM) and XO (40 mU/ml) were applied after 8 min (indicated by the bar) of sample collection. The ROS produced evoked a rise in the release of tritium. Tritium outflow is presented as percentage change in relation to baseline (0%). The results are the mean ^ SEM of 2–6 independent experiments (performed in duplicate).
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
129
Fig. 3. Effect of blockade of the NMDA subtype of glutamate receptors on the impaired recovery of synaptic transmission after X/XO application, in the presence of the selective adenosine A1 receptor antagonist, DPCPX. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in the presence of DPCPX (100 nM) (†) and in the presence of DPCPX (100 nM) plus the NMDA receptor selective antagonist, AP5 (50 mM, W). Upon washout of X/XO, there was no difference in the recovery due to the presence of AP5. The results are the mean ^ SEM of four independent experiments.
Fig. 2. Effect of ROS production with X/XO on synaptic transmission. (A) Effect of ROS on synaptic transmission in the presence and in the absence of a selective antagonist of A1 receptors. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in control conditions (†) and in the presence of the selective adenosine A1 receptor antagonist, DPCPX (100 nM; W). Upon washout of X/XO, there was a complete recovery under control conditions, but only a partial recovery in the presence of DPCPX. Please note that the difference between control conditions and in the presence of DPCPX (100 nM) was reached immediately after the removal of X/XO. The results are the mean ^ SEM of five independent experiments. (B) Individual fEPSP representative of phase 1, 2 and 3 as depicted in (A). (C) Effect of ROS on synaptic transmission in the presence and in the absence of a selective antagonist of A2A receptors. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in control conditions (†) and in the presence of the selective adenosine A2A receptor antagonist, SCH 58261 (100 nM; W). Upon washout of X/XO, there was a recovery both under control conditions and in the presence of SCH 58261. The results are the mean ^ SEM of four independent experiments.
FeSO4. During superfusion of FeSO4 (100 mM) for 30 min and wash-out for 60 min there was no significant modification in the fEPSP slope (3.2 ^ 8.1% change at the end of washout, n ¼ 6), whereas in experiments performed in the presence of DPCPX (100 nM), application of FeSO4 (100 mM) induced a fast decrease in the synaptic transmission, which remained until the end of the experiment (19.5 ^ 2.0% decrease at the end of washout, n ¼ 4, P , 0:05, Fig. 4).
The production of ROS is known to inhibit synaptic transmission in hippocampal slices [17] and in invertebrate synapses [6]. This impairment appears to involve a presynaptic mechanism, since the postsynaptic responses to the iontophoretical application of neurotransmitter persist in the presence of marked depression of transmission [16], and ROS decrease evoked excitatory amino acid release in cortical synaptosomes [11]. Although in another study the inhibition of synaptic transmission caused by ROS was due to endogenous adenosine [13], and we indeed observed a large increase in the extracellular levels of adenosine by applying X/XO, we found that the depression of synaptic transmission was not attenuated by either adenosine A1 or A2A receptor antagonists. Presumably the X/XO system generates large extracellular amounts of the superoxide
Fig. 4. Effect of ROS production with FeSO4 on synaptic transmission in the presence and in the absence of the selective adenosine A1 receptor antagonist, DPCPX. During superfusion of FeSO4 (100 mM) for 30 min (indicated by the bar) and wash-out for 60 min there was no significant modification in the fEPSP (†). In the presence of DPCPX (100 nM, W) the application of FeSO4 (100mM) induced a sustained decrease in the synaptic transmission. The results are the mean ^ SEM of four independent experiments.
130
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
radical, which are not immediately detected by the intracellular fluorescent probe, and rapidly induce protein oxidation and lipid peroxidation, interfering with the activity of ion channels, transporters, and other crucial membrane constituents at the synaptic level [14], and prevailing over the inhibitory effects of released adenosine on synaptic transmission. Remarkably, although released adenosine did not contribute to the depression of synaptic transmission, it could determine the subsequent recovery. With the ROS generating system X/XO, we observed a significant impairment of recovery when A1 receptors were blocked, so that endogenous adenosine was able to promote rescue of the synaptic transmission by a further 30% until full recovery. In a similar way, FeSO4 could only induce irreversible depression of synaptic transmission when adenosine A1 receptors were blocked. Interestingly, upon hypoxic conditions, in which ROS are produced, the recovery of synaptic transmission also requires the presence of adenosine acting upon A1 receptors [20]. In the present study, adenosine did not limit the formation of ROS, so it must attenuate the deleterious cellular consequences of produced ROS. Adenosine may modulate the cellular metabolic status, refraining the neuronal metabolic activity and preventing irreversible lesions [7]. Also, the important role of adenosine in decreasing the release of glutamate and other excitatory amino acids [4], could avoid the vicious cycle generated by ROS-induced glutamate release [15] and glutamate-induced ROS formation [19]. However, in contrast with what happens in hypoxia, in which adenosine enhances recovery by inhibiting NMDA receptors [20], the facilitation of recovery by adenosine after ROS lesion did not involve NMDA receptors. The present findings that the neuronal lesion induced by ROS may be reversible, provided released adenosine is allowed to act upon A1 receptors, reinforces the importance of the neuroprotective role for this nucleoside.
Acknowledgements The authors thank the colleagues of the laboratory for their comments and Prof. Joa˜o Malva, Center for Neuroscience, Coimbra, for the use of the spectrometer. Supported by a grant from Reitoria Universidade Lisboa and a master fellowship SFRH/BM/2096/2000 to C.A. from FCT.
References [1] W.W. Anderson, G.L. Collingridge, The LTP Program: a data acquisition program for on-line analysis of long-term potentiation and other synaptic events, J. Neurosci. Methods 108 (2001) 71–83. [2] M.F. Beal, Aging, energy, and oxidative stress in neurodegenerative diseases, Ann. Neurol. 38 (1995) 357–366.
[3] R.M. Broad, N. Fallahi, B.B. Fredholm, Nitric oxide interacts with oxygen free radicals to evoke the release of adenosine and adenine nucleotides from rat hippocampal slices, J. Auton. Nerv. Syst. 81 (2000) 82– 86. [4] S.P. Burke, J.V. Nadler, Regulation of glutamate and aspartate release from slices of the hippocampal CA1 area: effects of adenosine and baclofen, J. Neurochem. 51 (1988) 1541–1551. [5] W. Cao, J.M. Carney, A. Duchon, R.A. Floyd, M. Chevion, Oxygen free radical involvement in ischemia and reperfusion injury to brain, Neurosci. Lett. 88 (1988) 233 –238. [6] C.A. Colton, J. Yao, Y. Grossman, D.L. Gilbert, The effect of xanthine/xanthine oxidase generated reactive oxygen species on synaptic transmission, Free Radical Res. Commun. 14 (1991) 385 –393. [7] R.A. Cunha, Adenosine as a modulator and as a homeostatic regulator in the nervous system: different roles, different sources and different receptors, Neurochem. Int. 38 (2001) 107–125. [8] R.A. Cunha, E.S. Vizi, J.A. Ribeiro, A.M. Sebastia˜o, Preferential release of ATP and its extracellular catabolism as a source of adenosine upon high- but not low-frequency stimulation of rat hippocampal slices, J. Neurochem. 67 (1996) 2180–2187. [9] A. de Mendonc¸a, A.M. Sebastia˜o, J.A. Ribeiro, Adenosine: does it have a neuroprotective role after all?, Brain Res. Rev. 33 (2000) 258 –274. [10] T.V. Dunwiddie, S.A. Masino, The role and regulation of adenosine in the central nervous system, Annu. Rev. Neurosci. 24 (2001) 31– 55. [11] S.C. Gilman, M.J. Bonner, T.C. Pellmar, Effect of oxidative stress on excitatory amino acid release by cerebral cortical synaptosomes, Free Radical Biol. Med. 15 (1993) 671 –675. [12] E.W.I. Kellogg, I. Fridovich, Liposome oxidation and erythrocyte lysis by enzimatically generated superoxide and hydrogen peroxide, J. Biol. Chem. 252 (1977) 6721–6728. [13] S.A. Masino, M.H. Meshes, P.C. Bickford, T.V. Dunwiddie, Acute peroxide treatment of rat hippocampal slices induces adenosinemediated inhibition of excitatory transmission in area CA1, Neurosci. Lett. 274 (1999) 91–94. [14] M.P. Mattson, Modification of ion homeostasis by lipid peroxidation: roles in neuronal degeneration and adaptive plasticity, Trends Neurosci. 20 (1998) 53 –57. [15] D.E. Pellegini-Giampetro, G. Cherici, M. Alesiani, V. Carla`, F. Moroni, Excitatory amino acid release from rat hippocampal slices as a consequence of free-radical formation, J. Neurochem. 51 (1988) 1960–1963. [16] T.C. Pellmar, Electrophysiological correlates of peroxide damage in guinea pig hippocampus in vitro, Brain Res. 364 (1986) 377–381. [17] T.C. Pellmar, Use of brain slices in the study of free-radical actions, J. Neurosci. Methods 59 (1995) 93–98. [18] A.C. Rego, C.R. Oliveira, Alteration of nitric oxide synthase activity upon oxidative stress in cultured retinal cells, J. Neurosci. Res. 51 (1998) 627 –635. [19] I.J. Reynolds, T.G. Hastings, Glutamate induces the production of reactive oxygen species in cultured forebrain neurons following NMDA receptor activation, J. Neurosci. 15 (1995) 3318–3327. [20] A.M. Sebastia˜o, A. de Mendonc¸a, T. Moreira, J.A. Ribeiro, Activation of synaptic NMDA receptors by action potential-dependent release of transmitter during hypoxia impairs recovery of synaptic transmission on reoxygenation, J. Neurosci. 21 (2001) 8564–8571. [21] A.M. Sebastia˜o, T.W. Stone, J.A. Ribeiro, The inhibitory adenosine receptor at the neuromuscular junction and hippocampus of the rat: antagonism by 1,3,8-substituted xanthines, Br. J. Pharmacol. 101 (1990) 453 –459. [22] G.J. Wilde, A.K. Pringle, P. Wright, F. Iannotti, Differential vulnerability of the CA1 and CA3 subfields of the hippocampus to superoxide and hydroxyl radicals in vitro, J. Neurochem. 69 (1997) 883–886.
Adenosine promotes neuronal recovery from reactive oxygen species induced lesion in rat hippocampal slices Cla´udia G. Almeida, Alexandre de Mendonc¸a*, Rodrigo A. Cunha, J. Alexandre Ribeiro Laboratory of Neurosciences, Faculty of Medicine of Lisbon, Avenue Prof Egas Moniz, 1649-028 Lisbon, Portugal Received 8 October 2002; received in revised form 5 December 2002; accepted 18 December 2002
Abstract Reactive oxygen species (ROS) are believed to be involved in the pathogenesis of several neurological disorders. We now tested whether the endogenous neuroprotective substance, adenosine, attenuates the cell damage induced by ROS. In rat hippocampal slices, the xanthine oxidase (40 mU/ml) plus xanthine (1 mM) (X/XO) system produced a 27.8 ^ 7.3% (n ¼ 3) increase in ROS, measured by fluorimetry with 20 ,70 -dichlorodihydrofluorescein, a 246.9 ^ 18.4% (n ¼ 6) increase in the release of tritiated adenosine, and a decrease in synaptic transmission that fully recovered after washout. In the presence of the adenosine A1 receptor selective antagonist, 1,3-dipropyl-8cyclopentylxanthine (100 nM), X/XO induced a similar inhibition, however synaptic transmission only recovered to 70.7 ^ 5.8% of control (n ¼ 5). The blockade of A2A receptors was devoid of effect (n ¼ 4). Adenosine is released by ROS-generating systems, and attenuates the deleterious cellular consequences of ROS through A1 receptor activation. q 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Adenosine; Neuroprotection; Reactive oxygen species; Hippocampus; A1 receptors; A2A receptors; Dipropylcyclopentylxanthine
Cell damage induced by of reactive oxygen species (ROS) is presumably involved in acute neurological disorders such as stroke [5], as well as in chronic neurodegenerative disorders [2]. The possibility of attenuating the deleterious effects of free radicals on nervous tissue thus appears an interesting approach to ameliorate those neurological disorders. Adenosine is an endogenous purine with an important modulatory role on neuronal activity [10], and broad neuroprotective properties exerted in experimental models of hypoxia, ischemia, excitotoxicity and trauma [9]. However, it is not known whether adenosine can also protect nerve cells against the deleterious actions of ROS. Interestingly, the free radical generating system, hydrogen peroxide plus ferrous sulfate, causes a transient inhibition of synaptic transmission, which is blocked by the adenosine receptor antagonist, theophylline, and thus presumably due to released adenosine in hippocampal slices [13]. Another free radical generating system, xanthine plus xanthine oxidase, is also able to induce the release of purines from hippocampal slices [3]. Since the generation of ROS is able to induce the release of adenosine, we hypothesize that adenosine might represent * Corresponding author. Tel./fax: þ 351-21-7936787. E-mail address: [email protected] (A. de Mendonc¸a).
an endogenous feedback protective system to limit the damage induced by ROS. We now tested this hypothesis by studying whether endogenous adenosine could promote recovery of synaptic transmission in hippocampal slices subjected to ROS-induced damage. The experiments were performed on hippocampal slice preparations 400 mm thick, taken from male Wistar rats 5– 8 weeks old, superfused with an artificial cerebrospinal fluid (aCSF) solution composed of (mM): NaCl 124, KCl 3, NaH2PO4 1.25, MgSO4 1, CaCl2 2, NaHCO3 26, glucose 10, pH 7.4, equilibrated with a 95% O2/5%CO2 gas mixture, at 30.5 8C [21]. To induce oxidative stress the free radical generating system xanthine oxidase (XO; 40 mU/ml, Sigma) plus xanthine (X; 1 mM, Sigma), which predominantly generates the superoxide radical [12], was applied for 45 min. Another free radical generating system was also tested, FeSO4 (100 mM, applied during 30 min), which by the Fenton reaction induces the formation of the hydroxyl radical [22]. Intracellular oxidative stress was evaluated with the probe 20 ,70 -dichlorodihydrofluorescein diacetate (DCFH2-DA; Molecular Probes, USA). The hippocampal slices were loaded with 20 mM DCFH2-DA for 20 min in the dark [18], rinsed for 30 min with aCSF, and the slice fluorescence measured for 10 min (baseline), exciting the probe at 502
0304-3940/03/$ - see front matter q 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(02)01478-7
128
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
nm and measuring the emission at 550 nm, with a Fluoromax-2 spectrometer. Oxidative stress was then induced and slice fluorescence continuously monitored. Release of endogenous adenosine was measured by scintillation spectrophotometry after loading the hippocampal slices with [2 – 3H]adenosine (37 MBq/ml; 1 mCi/ml; from Amersham International, UK) [8]. Evoked field excitatory postsynaptic potentials (fEPSP) were recorded in the stratum radiatum of the CA1 area evoked by electrical stimulation of the Schaffer collateral/commissural fibers [21]. Field EPSPs (the averages of eight consecutive potentials, 1.0 –1.5 mV amplitude and 0.5 –0.8 mV.ms21 slope) were quantified as the slope of the initial phase of the potential with the LTP 2.21 software [1]. The slope data are shown as the percentage change from the average slope of the fEPSP taken during the 10 min that preceded the application of the ROS generating system. The drugs, 1,3-dipropyl-8-cyclopentylxanthine (DPCPX; from Sigma), 5-amino-7-(2-phenylethyl)-2-(2-furyl)-pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine (SCH 58261; kindly supplied by Dr E. Ongini, Schering-Plough, Milan), and DL-2-amino-5-phosphonopentanoate (AP5; from Sigma), were allowed an equilibrium period of 30 min of superfusion prior to the application of the ROS generating system. Data are expressed as mean ^ SEM values, of n experiments. Statistical analysis was performed using the two-tailed Student’s t-test, at a level of significance of P , 0:05. The intracellular ROS production induced by X/XO application was studied in the hippocampal slices by using a fluorimetric assay with the probe DCFH2-DA. There was a rapid and steady increase in fluorescence that attained a maximum of 32.8 ^ 5.1% at the end of the 45 min superfusion with X/XO (n ¼ 3, P , 0:05). In the presence of the selective adenosine A1 receptor antagonist, DPCPX (100 nM), there was a maximal increase in fluorescence levels induced by X/XO (21.1 ^ 3.8% increase, n ¼ 5, P , 0:05), which was not significantly different from that observed in control conditions. We next investigated whether the generation of ROS could elicit an increase in the extracellular levels of adenosine. Superfusion with X/XO induced an increase in the release of adenosine which had a maximum, 246.9 ^ 18.4% (n ¼ 6, P , 0:05), about 15 min after the application of X/XO, and afterwards decreased but remained above basal levels at least until 1 h after beginning X/XO application (Fig. 1). In the electrophysiological experiments, X/XO (superfused for 45 min; xanthine itself elicited 4.3 ^ 0.9% increase in fEPSP slope, n ¼ 3) caused a rapid decrease in the fEPSP slope. After 14 min the depression was apparently stable and had a value of 62.0 ^ 9.1% (n ¼ 5, P , 0:05, Fig. 2A,B). The removal of X/XO allowed the virtual recovery of synaptic transmission from the insult (to 2.7 ^ 11.4% of the fEPSP before X/XO) in usually 40 min.
When the same protocol was performed in the presence of DPCPX (100 nM), X/XO also elicited a marked decrease of fEPSP slope, 66.4 ^ 7.8% (n ¼ 5, P , 0:05). There was no significant difference between the depressions of the synaptic transmission induced by X/XO in the presence and in the absence of DPCPX. However, in the washout period, the recovery of synaptic transmission in the presence of DPCPX was only partial, reaching 70.7 ^ 5.8% of control (n ¼ 5, P , 0:05 compared with the fEPSP before X/XO). This recovery was significantly different (P , 0:05) from the virtually complete recovery observed in control conditions in the absence of DPCPX. We next investigated whether the selective adenosine A2A receptor antagonist, SCH 58261, could modify the effects of X/XO on synaptic transmission. In the presence of SCH 58261 (100 nM), superfusion with X/XO (45 min) induced a marked decrease, 96.9 ^ 2.9% (n ¼ 4, P , 0:05) of the fEPSP slope, similar to control conditions (94.5 ^ 0.5%, n ¼ 4, P , 0:05; Fig. 2C). After the removal of X/XO, synaptic transmission took about 40 min to recover to 77.4 ^ 13.3% (of the fEPSP before X/XO, n ¼ 4, P , 0:05) in the presence of SCH 58261 and to 85.4 ^ 15.7% (of the fEPSP before X/XO, n ¼ 4, P , 0:05) in control conditions. There were no significant differences either in the depression or in the recovery of fEPSP obtained in the presence or in the absence of the A2A receptor antagonist. We next tested whether the impairment in recovery after X/XO observed in the presence of DPCPX could involve the activation of the N-methyl-D -aspartate (NMDA) subtype of glutamate receptors. This was not the case, since recovery from X/XO was still decreased (70.7 ^ 4.9% of the fEPSP before X/XO, n ¼ 4, P , 0:05) in the presence of DPCPX (100 nM) plus the NMDA receptor selective antagonist, AP5 (50 mM), as compared with the recovery in the presence of DPCPX (100 nM) only (67.5 ^ 5.5% of the fEPSP before X/XO, n ¼ 4, P , 0:05, Fig. 3). We also tested another substance as a source of ROS,
Fig. 1. Effect of ROS production with X plus XO on the release of tritium from hippocampal slices labeled with [2-3H]adenosine. X (1 mM) and XO (40 mU/ml) were applied after 8 min (indicated by the bar) of sample collection. The ROS produced evoked a rise in the release of tritium. Tritium outflow is presented as percentage change in relation to baseline (0%). The results are the mean ^ SEM of 2–6 independent experiments (performed in duplicate).
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
129
Fig. 3. Effect of blockade of the NMDA subtype of glutamate receptors on the impaired recovery of synaptic transmission after X/XO application, in the presence of the selective adenosine A1 receptor antagonist, DPCPX. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in the presence of DPCPX (100 nM) (†) and in the presence of DPCPX (100 nM) plus the NMDA receptor selective antagonist, AP5 (50 mM, W). Upon washout of X/XO, there was no difference in the recovery due to the presence of AP5. The results are the mean ^ SEM of four independent experiments.
Fig. 2. Effect of ROS production with X/XO on synaptic transmission. (A) Effect of ROS on synaptic transmission in the presence and in the absence of a selective antagonist of A1 receptors. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in control conditions (†) and in the presence of the selective adenosine A1 receptor antagonist, DPCPX (100 nM; W). Upon washout of X/XO, there was a complete recovery under control conditions, but only a partial recovery in the presence of DPCPX. Please note that the difference between control conditions and in the presence of DPCPX (100 nM) was reached immediately after the removal of X/XO. The results are the mean ^ SEM of five independent experiments. (B) Individual fEPSP representative of phase 1, 2 and 3 as depicted in (A). (C) Effect of ROS on synaptic transmission in the presence and in the absence of a selective antagonist of A2A receptors. Superfusion with X/XO for 45 min (indicated by the bar) induced a marked decrease of the fEPSP both in control conditions (†) and in the presence of the selective adenosine A2A receptor antagonist, SCH 58261 (100 nM; W). Upon washout of X/XO, there was a recovery both under control conditions and in the presence of SCH 58261. The results are the mean ^ SEM of four independent experiments.
FeSO4. During superfusion of FeSO4 (100 mM) for 30 min and wash-out for 60 min there was no significant modification in the fEPSP slope (3.2 ^ 8.1% change at the end of washout, n ¼ 6), whereas in experiments performed in the presence of DPCPX (100 nM), application of FeSO4 (100 mM) induced a fast decrease in the synaptic transmission, which remained until the end of the experiment (19.5 ^ 2.0% decrease at the end of washout, n ¼ 4, P , 0:05, Fig. 4).
The production of ROS is known to inhibit synaptic transmission in hippocampal slices [17] and in invertebrate synapses [6]. This impairment appears to involve a presynaptic mechanism, since the postsynaptic responses to the iontophoretical application of neurotransmitter persist in the presence of marked depression of transmission [16], and ROS decrease evoked excitatory amino acid release in cortical synaptosomes [11]. Although in another study the inhibition of synaptic transmission caused by ROS was due to endogenous adenosine [13], and we indeed observed a large increase in the extracellular levels of adenosine by applying X/XO, we found that the depression of synaptic transmission was not attenuated by either adenosine A1 or A2A receptor antagonists. Presumably the X/XO system generates large extracellular amounts of the superoxide
Fig. 4. Effect of ROS production with FeSO4 on synaptic transmission in the presence and in the absence of the selective adenosine A1 receptor antagonist, DPCPX. During superfusion of FeSO4 (100 mM) for 30 min (indicated by the bar) and wash-out for 60 min there was no significant modification in the fEPSP (†). In the presence of DPCPX (100 nM, W) the application of FeSO4 (100mM) induced a sustained decrease in the synaptic transmission. The results are the mean ^ SEM of four independent experiments.
130
C.G. Almeida et al. / Neuroscience Letters 339 (2003) 127–130
radical, which are not immediately detected by the intracellular fluorescent probe, and rapidly induce protein oxidation and lipid peroxidation, interfering with the activity of ion channels, transporters, and other crucial membrane constituents at the synaptic level [14], and prevailing over the inhibitory effects of released adenosine on synaptic transmission. Remarkably, although released adenosine did not contribute to the depression of synaptic transmission, it could determine the subsequent recovery. With the ROS generating system X/XO, we observed a significant impairment of recovery when A1 receptors were blocked, so that endogenous adenosine was able to promote rescue of the synaptic transmission by a further 30% until full recovery. In a similar way, FeSO4 could only induce irreversible depression of synaptic transmission when adenosine A1 receptors were blocked. Interestingly, upon hypoxic conditions, in which ROS are produced, the recovery of synaptic transmission also requires the presence of adenosine acting upon A1 receptors [20]. In the present study, adenosine did not limit the formation of ROS, so it must attenuate the deleterious cellular consequences of produced ROS. Adenosine may modulate the cellular metabolic status, refraining the neuronal metabolic activity and preventing irreversible lesions [7]. Also, the important role of adenosine in decreasing the release of glutamate and other excitatory amino acids [4], could avoid the vicious cycle generated by ROS-induced glutamate release [15] and glutamate-induced ROS formation [19]. However, in contrast with what happens in hypoxia, in which adenosine enhances recovery by inhibiting NMDA receptors [20], the facilitation of recovery by adenosine after ROS lesion did not involve NMDA receptors. The present findings that the neuronal lesion induced by ROS may be reversible, provided released adenosine is allowed to act upon A1 receptors, reinforces the importance of the neuroprotective role for this nucleoside.
Acknowledgements The authors thank the colleagues of the laboratory for their comments and Prof. Joa˜o Malva, Center for Neuroscience, Coimbra, for the use of the spectrometer. Supported by a grant from Reitoria Universidade Lisboa and a master fellowship SFRH/BM/2096/2000 to C.A. from FCT.
References [1] W.W. Anderson, G.L. Collingridge, The LTP Program: a data acquisition program for on-line analysis of long-term potentiation and other synaptic events, J. Neurosci. Methods 108 (2001) 71–83. [2] M.F. Beal, Aging, energy, and oxidative stress in neurodegenerative diseases, Ann. Neurol. 38 (1995) 357–366.
[3] R.M. Broad, N. Fallahi, B.B. Fredholm, Nitric oxide interacts with oxygen free radicals to evoke the release of adenosine and adenine nucleotides from rat hippocampal slices, J. Auton. Nerv. Syst. 81 (2000) 82– 86. [4] S.P. Burke, J.V. Nadler, Regulation of glutamate and aspartate release from slices of the hippocampal CA1 area: effects of adenosine and baclofen, J. Neurochem. 51 (1988) 1541–1551. [5] W. Cao, J.M. Carney, A. Duchon, R.A. Floyd, M. Chevion, Oxygen free radical involvement in ischemia and reperfusion injury to brain, Neurosci. Lett. 88 (1988) 233 –238. [6] C.A. Colton, J. Yao, Y. Grossman, D.L. Gilbert, The effect of xanthine/xanthine oxidase generated reactive oxygen species on synaptic transmission, Free Radical Res. Commun. 14 (1991) 385 –393. [7] R.A. Cunha, Adenosine as a modulator and as a homeostatic regulator in the nervous system: different roles, different sources and different receptors, Neurochem. Int. 38 (2001) 107–125. [8] R.A. Cunha, E.S. Vizi, J.A. Ribeiro, A.M. Sebastia˜o, Preferential release of ATP and its extracellular catabolism as a source of adenosine upon high- but not low-frequency stimulation of rat hippocampal slices, J. Neurochem. 67 (1996) 2180–2187. [9] A. de Mendonc¸a, A.M. Sebastia˜o, J.A. Ribeiro, Adenosine: does it have a neuroprotective role after all?, Brain Res. Rev. 33 (2000) 258 –274. [10] T.V. Dunwiddie, S.A. Masino, The role and regulation of adenosine in the central nervous system, Annu. Rev. Neurosci. 24 (2001) 31– 55. [11] S.C. Gilman, M.J. Bonner, T.C. Pellmar, Effect of oxidative stress on excitatory amino acid release by cerebral cortical synaptosomes, Free Radical Biol. Med. 15 (1993) 671 –675. [12] E.W.I. Kellogg, I. Fridovich, Liposome oxidation and erythrocyte lysis by enzimatically generated superoxide and hydrogen peroxide, J. Biol. Chem. 252 (1977) 6721–6728. [13] S.A. Masino, M.H. Meshes, P.C. Bickford, T.V. Dunwiddie, Acute peroxide treatment of rat hippocampal slices induces adenosinemediated inhibition of excitatory transmission in area CA1, Neurosci. Lett. 274 (1999) 91–94. [14] M.P. Mattson, Modification of ion homeostasis by lipid peroxidation: roles in neuronal degeneration and adaptive plasticity, Trends Neurosci. 20 (1998) 53 –57. [15] D.E. Pellegini-Giampetro, G. Cherici, M. Alesiani, V. Carla`, F. Moroni, Excitatory amino acid release from rat hippocampal slices as a consequence of free-radical formation, J. Neurochem. 51 (1988) 1960–1963. [16] T.C. Pellmar, Electrophysiological correlates of peroxide damage in guinea pig hippocampus in vitro, Brain Res. 364 (1986) 377–381. [17] T.C. Pellmar, Use of brain slices in the study of free-radical actions, J. Neurosci. Methods 59 (1995) 93–98. [18] A.C. Rego, C.R. Oliveira, Alteration of nitric oxide synthase activity upon oxidative stress in cultured retinal cells, J. Neurosci. Res. 51 (1998) 627 –635. [19] I.J. Reynolds, T.G. Hastings, Glutamate induces the production of reactive oxygen species in cultured forebrain neurons following NMDA receptor activation, J. Neurosci. 15 (1995) 3318–3327. [20] A.M. Sebastia˜o, A. de Mendonc¸a, T. Moreira, J.A. Ribeiro, Activation of synaptic NMDA receptors by action potential-dependent release of transmitter during hypoxia impairs recovery of synaptic transmission on reoxygenation, J. Neurosci. 21 (2001) 8564–8571. [21] A.M. Sebastia˜o, T.W. Stone, J.A. Ribeiro, The inhibitory adenosine receptor at the neuromuscular junction and hippocampus of the rat: antagonism by 1,3,8-substituted xanthines, Br. J. Pharmacol. 101 (1990) 453 –459. [22] G.J. Wilde, A.K. Pringle, P. Wright, F. Iannotti, Differential vulnerability of the CA1 and CA3 subfields of the hippocampus to superoxide and hydroxyl radicals in vitro, J. Neurochem. 69 (1997) 883–886.