- Email: [email protected]
Acute peroxide treatment of rat hippocampal slices induces adenosine-mediated inhibition of excitatory transmission in area CA1
Neuroscience Letters 274 (1999) 91±94 www.elsevier.com/locate/neulet
Acute peroxide treatment of rat hippocampal slices induces adenosine-mediated inhibition of excitatory transmission in area CA1 Susan A. Masino a,*, Michael H. Mesches b, Paula C. Bickford b, Thomas V. Dunwiddie b a
Department of Pharmacology and Neuroscience Program, B138 UCHSC, 4200 E. 9th Avenue, Denver, CO 80262, USA b Department of Pharmacology and VA Medical Center, C236 UCHSC, 4200 E. 9th Avenue, Denver, CO 80262, USA Received 23 July 1999; received in revised form 16 August 1999; accepted 18 August 1999
Abstract Brief exposure to conditions that generate free radicals inhibits synaptic transmission in hippocampal slices, most likely via a presynaptic mechanism. Because other physiologically stressful conditions that generate free radicals, such as hypoxia or ischemia, stimulate the release of adenosine from brain slices, we determined whether increases in extracellular adenosine mediate the presynaptic inhibition of excitatory transmission induced by peroxide treatment. Simultaneous addition of hydrogen peroxide (0.01%) and ferrous sulfate (100 mM) resulted in a .80% decrease in synaptic potentials recorded in the CA1 region of hippocampal slices of adult male rats. Treatment with theophylline (200 mM), a non-selective adenosine receptor antagonist, or 8-cyclopentyl-1, 3-dipropylxanthine (100 nM), a selective adenosine A1 receptor antagonist, prior to and during hydrogen peroxide superfusion prevented the inhibition. These results demonstrate that acute exposure to hydrogen peroxide induces an adenosine-mediated decrease in synaptic transmission in hippocampal slices. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Theophylline; 8-Cyclopentyl-1, 3-dipropylxanthine; Adenosine; Hydrogen peroxide; Ferrous sulfate; CA1; Hippocampus; Excitatory postsynaptic potential; Adenosine receptors; A1 receptors
Reactive oxygen-derived free radicals are present in all tissues and their levels are maintained within normal limits by a variety of enzymes and antioxidants. Numerous studies have reported that increased levels of free radicals are associated with a variety of pathological conditions, including both acute neurotrauma and chronic neurodegenerative disease [7,12,17] and free radical species might contribute to the neuronal damage. Previous studies have explored the acute electrophysiological effects of free radical-generating conditions in vitro. Electrophysiological recordings from the CA1 region of the rat hippocampus have demonstrated that adding hydrogen peroxide (H2O2) or a combination of hydrogen peroxide and ferrous sulfate (FeSO4) to the superfusion buffer, which generates highly reactive hydroxyl radicals, results in an inhibition of synaptic potentials [9,13±15]. This has been observed both with extracellular ®eld recordings [9,13] and intracellular recordings of pyra* Corresponding author. Tel.: 11-303-315-7584 fax: 11-303315-4814. E-mail address: [email protected] (S.A. Masino)
midal cells [14]. As the response to direct iontophoretic application of either glutamate or g aminobutyric acid was unaffected during the peroxide superfusion, the effect of peroxide on hippocampal synaptic potentials was concluded to be presynaptic [14]. Concentrations of adenosine suf®cient to tonically activate adenosine receptors are normally found in the extracellular space in brain, and adenosine is a potent neuromodulator in many brain regions [5]. In the CA1 region of the hippocampus, one of the major actions of extracellular adenosine is a presynaptic inhibition of excitatory glutamatergic transmission, an effect mediated by adenosine A1 receptors [5]. Increased adenosine release is a common response to many physiologically stressful situations, such as hypoxia [21], hypoglycemia [2,8] and increased temperature [11]. Given that adenosine release is commonly induced by metabolic stress, and excessive oxygen-derived free radicals can result from and/or contribute to cell stress, the aim of the present study was to determine whether the inhibition of neurotransmission induced
0304-3940/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 69 3- X
92
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
by superfusion of brain slices with hydrogen peroxide is similarly mediated by increases in the extracellular concentrations of adenosine. Transverse hippocampal slices were obtained from 6±8 week-old Sprague±Dawley rats using standard procedures previously described [11]. Rats were decapitated, the brains rapidly removed into ice-cold arti®cial cerebrospinal ¯uid (aCSF) and the hippocampus dissected free from the surrounding tissue. 400 mm slices were made on a Sorvall TC-2 tissue chopper and incubated at 32.58C. The aCSF used for dissection, incubation and submerged, superfused recordings contained (in mM): NaCl 126.0, KCl 3.0, MgSO4 1.5, CaCl2 2.4, NaH2PO4 1.2, NaHCO3 25.9, d-glucose 11 and was saturated with 95% O2 / 5%CO2. Slices were incubated undisturbed for 90 min before electrophysiological recording. Slices were placed on a nylon net in the recording chamber and superfused continuously (2.0 ml/min) with aCSF bubbled with humidi®ed with 95%O2 /5%CO2. Extracellular ®eld excitatory postsynaptic potentials (fEPSPs) were recorded from the CA1 region of the stratum radiatum using glass micropipettes (10±15 MV) ®lled with 3 M NaCl. A concentric bipolar stimulating electrode stimulated the Schaffer collateral and commissural ®bers in the stratum radiatum every 30 s. In some experiments a pair of synaptic stimuli with an interstimulus interval of 70 ms was delivered every 30 s. The stimulation intensity was adjusted such that the baseline fEPSP amplitude was between 0.7 and 1.2 mV. Data were recorded via an AC ampli®er, digitized and stored for later analysis. After establishing a baseline fEPSP, slices were either superfused with hydrogen peroxide (0.01%) and ferrous sulfate (100 mM), or pretreated with theophylline (a nonselective competitive adenosine receptor antagonist; 200 mM) and then superfused with H2O2/FeSO4. In initial experiments, application of hydrogen peroxide alone produced more variable results or a longer onset to a decrease in the fEPSP, as has been reported previously [13]. Thus, ferrous sulfate was added simultaneously with the hydrogen peroxide to increase tissue iron and promote formation of the hydroxyl radical via the Fenton reaction. Application of ferrous sulfate alone has no effect on synaptic transmission ([13] and unpublished data). All drugs were applied for a minimum of 10 min, or until a stable response was achieved for at least 5 min. There was only one application of H2O2/FeSO4 per slice due to the potential long term effects on membrane integrity due to lipid peroxidation. Theophylline, hydrogen peroxide (30% w/ w), ferrous sulfate and all aCSF constituents were obtained from Sigma (St. Louis, MO). 8-Cyclopentyl-1, 3-dipropylxanthine (DPCPX) was obtained from Research Biochemicals (Natick, MA). Statistical analyses included Student's two-tailed paired or unpaired t-tests where appropriate and the F-test for analysis of sample variance. Co-application of hydrogen peroxide (0.01%) and ferrous sulfate (100 mM) resulted in an 81 ^ 3:8% (mean ^ SEM;
n 6) decrease in the fEPSP. This decrease usually reversed upon washout (Fig. 1), but in some cases the fEPSP did not recover immediately (Fig. 3B). Fig. 1 illustrates the fEPSP recorded in stratum radiatum from a single slice prior to, during and following superfusion with peroxide. In this case the fEPSP decreased rapidly during the peroxide treatment and there was a nearly complete recovery following washout. In four slices tested using this protocol, the fEPSP recovered to 82 ^ 6.5% of the original baseline 35 min after removal of the H2O2/FeSO4. As evidenced by the individual traces of the synaptic response, the amplitude of the fEPSP decreased (Fig. 1; trace (a) vs. trace (b)) and the variance of the responses increased significantly (F 4:069; P , 0:05) upon addition of the H2O2/ FeSO4. In conjunction with this pronounced decrease in the fEPSP, the ratio between the amplitude of the second EPSP response (R2) and the ®rst (R1) increased signi®cantly during the peroxide treatment. The average ratio (R2/R1) during baseline recording was 1:35 ^ 0:03 and increased during H2O2/FeSO4 superfusion to 1:55 ^ 0:09 (n 6; P , 0:05). This increased synaptic facilitation during paired pulse stimulation is typically observed with presynaptic inhibition of transmitter release, previously hypothesized to underlie the fEPSP decrease observed with peroxide. Adenosine exerts its effects on synaptic transmission in the hippocampus primarily via presynaptic inhibition of excitatory transmission and an increase in paired pulse facilitation is observed either during the application of exogenous adenosine [5,6], or during an increase in endogenous adenosine [2]. To determine whether the inhibition during peroxide treatment was due speci®cally to activation of presynaptic
Fig. 1. The effect of hydrogen peroxide/ferrous sulfate (H2O2/ FeSO4) superfusion on fEPSP amplitude. Superfusion with H2O2/FeSO4 (duration of superfusion indicated by the bar along the abscissa) caused a profound decrease in the fEPSP and in this slice, the fEPSP response recovered slowly and almost completely upon washout. Synaptic responses (average of six to eight sweeps) before, during and after the H2O2/FeSO4 superfusion are indicated by (a±c) and superimposed at (d). Traces shown are the ®rst response to a 70 ms paired-pulse stimulation. Scale bars, 1 mV and 10 ms.
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
Fig. 2. Theophylline antagonism of the response to H2O2/FeSO4. Pretreatment with the competitive adenosine receptor antagonist theophylline (200 mM; duration of superfusion indicated by dashed line) prevented the normal decrease in the fEPSP induced by H2O2/FeSO4. In most slices, theophylline completely prevented any decrease in response to H2O2/FeSO4 (average change 1:0 ^ 4:7%, n 3). Synaptic responses before, during and after the H2O2/FeSO4 are indicated by (a±c) and superimposed at d. Scale bars, 1 mV and 10 ms.
adenosine receptors, in some experiments the adenosine receptor antagonist theophylline (200 mM) was applied prior to the H2O2/FeSO4 superfusion. Theophylline alone produced an increase in the fEPSP (16 ^ 3:5%; n 3), re¯ecting antagonism of the inhibitory effect of endogenous adenosine [5]. After theophylline superfusion for 10 min, H2O2/FeSO4 was applied in the continued presence of theophylline. Theophylline completely prevented any decrease of the fEPSP due to the peroxide treatment (average decrease in the presence of theophylline: 1:0 ^ 4:7%, n 3; P , 0:0001 compared with decrease in H2O2/FeSO4 alone). An example of the fEPSP response to H2O2/FeSO4 in the presence of theophylline is shown in Fig. 2. In a separate set of experiments, a selective adenosine A1 receptor antagonist, DPCPX (100 nM), was applied. As with theophylline, DPCPX alone produced an increase in the fEPSP (32 ^ 6:4%; n 4). DPCPX also signi®cantly reduced the inhibition observed upon application of the H2O2/FeSO4 (average decrease in the presence of DPCPX: 15 ^ 10%, n 4; P , 0:0001 compared with decrease in H2O2/FeSO4 alone). In addition to preventing the decrease in the amplitude of the fEPSP when applied prior to the peroxide treatment, theophylline reversed the inhibition even in the continued presence of H2O2/FeSO4 (Fig. 3A), or when the amplitude of the fEPSP remained depressed after the washout of peroxide from the slice (Fig. 3B). In both cases, theophylline completely reversed the depression of the fEPSP. In four slices in which the fEPSP was inhibited by 84 ^ 5:0% by H2O2/ FeSO4, the amplitude of the response following theophylline superfusion was not signi®cantly different from the original baseline (0:98 ^ 0:17 mV vs. 1:06 ^ 0:09 mV, respectively, n 4; P . 0:05).
93
Although the depressant effect of H2O2/FeSO4 on hippocampal synaptic responses is well-established [9,13,14] and a presynaptic mechanism underlying this depression had been inferred [14], the speci®c mechanism by which transmitter release was inhibited has been unknown. In the present study we show that presynaptic activation of adenosine A1 receptors is the primary cause of synaptic inhibition during peroxide treatment. The adenosine A1 receptor is responsible for synaptic inhibition in other paradigms, such as anoxia [19], metabolic inhibition [20] and increased temperature [11], which all result in a similar adenosinemediated decrease in synaptic transmission. Thus, the acute depressant effect of peroxide appears to re¯ect a marked increase in the concentration of endogenous adenosine in the extracellular space, acting on A1 receptors to inhibit glutamate release. It should be noted that free radical generation has multiple cellular consequences [4] and can result in pathological changes such as apoptosis or necrosis [3]; at this point, the potential role of adenosine in these effects of peroxide is not clear. A previous study determined that sodium and potassium currents were not modi®ed and thus were not involved in the
Fig. 3. Theophylline reversal of the inhibition induced by H2O2/ FeSO4. Addition of theophylline (200 mM) either during superfusion with H2O2/FeSO4 (A), or following washout (B) completely reversed the synaptic depression induced by H2O2/FeSO4. This was the case even in slices that did not recover during the initial part of the washout period (cf. Fig. 3B with the recovery in Fig. 1). In general, theophylline returned the fEPSP amplitude to approximately the original baseline.
94
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
reduction of synaptic potential induced by peroxide treatment [14]. The increase in the threshold for activation of Ca 21 spikes reported by Pellmar [14] is also consistent with increased extracellular adenosine, because we have reported that adenosine increases the threshold for Ca 21 spikes [16] and other groups have shown more directly an adenosine A1 receptor-mediated inhibition of hippocampal Ca 21 channel activity [1]. It is not surprising that adenosine mediates the synaptic inhibition induced by free radicals, because increased extracellular adenosine is a common response to many stressful conditions. Free radicals might be intermediaries in some of these situations by contributing to the generation of extracellular adenosine and the resulting presynaptic inhibition via adenosine A1 receptors. There are likely multiple short and long term consequences of free radical generation in hippocampal slices and our data suggest only that adenosine might mediate some of the acute effects of peroxide. The incomplete recovery of the fEPSP that was occasionally observed could be due to the continued presence of adenosine, incomplete washout of the H2O2/FeSO4, continued presence of the free radical species in the tissue, or other effects of the treatment on synaptic transmission unrelated to adenosine receptors. As with hypoxia or ischemia, the increased extracellular adenosine release during superfusion with H2O2/FeSO4 may be neuroprotective, perhaps by reducing metabolic demand and limiting excitotoxicity [10,18]. However, it is not known whether presynaptic inhibition of glutamate release by adenosine would interact in any way with some of the other major actions of free radicals, such as lipid peroxidation and subsequent loss of membrane integrity. Any neuroprotective effect of the increased adenosinemediated presynaptic inhibition during free radical stress remains to be determined. [1] Ambrosio, A.F., Malva, J.O., Carvalho, A.P. and Carvalho, C.M., Inhibition of N-,P/Q- and other types of Ca 21 channels in rat hippocampal nerve terminals by the adenosine A1 receptor. Eur. J. Pharmacol., 340 (1997) 301±310. [2] Calabresi, P., Centonze, D., Pisani, A. and Bernardi, G., Endogenous adenosine mediates the presynaptic inhibition induced by aglycemia at corticostriatal synapses. J. Neurosci., 17 (1997) 4509±4516. [3] Clement, M.V., Ponton, A. and Pervaiz, S., Apoptosis induced by hydrogen peroxide is mediated by decreased superoxide anion concentration and reduction of intracellular milieu. FEBS Lett., 440 (1998) 13±18. [4] Cochrane, C.G., Cellular injury by oxidants. Am. J. Med., 91 (1991) 23S±30S.
[5] Dunwiddie, T.V., The physiological role of adenosine in the central nervous system. Int. Rev. Neurobiol., 27 (1985) 63± 139. [6] Dunwiddie, T.V. and Haas, H.L., Adenosine increases synaptic facilitation in the in vitro rat hippocampus: evidence for a presynaptic site of action. J. Physiol. (Lond.), 369 (1985) 365±377. [7] Farooqui, A.A. and Horrocks, L.A., Lipid peroxides in the free radical pathophysiology of brain diseases. Cell. Mol. Neurobiol., 18 (1998) 599±608. [8] Fowler, J.C., Purine release and inhibition of synaptic transmission during hypoxia and hypoglycemia in rat hippocampal slices. Neurosci. Lett., 157 (1993) 83±86. [9] Fowler, J.C., Hydrogen peroxide opposes the hypoxic depression of evoked synaptic transmission in rat hippocampal slices. Brain Res., 766 (1997) 255±258. [10] Fredholm, B.B., Adenosine and neuroprotection. Int. Rev. Neurobiol., 40 (1997) 259±280. [11] Masino, S.A. and Dunwiddie, T.V., Temperature-dependent modulation of excitatory transmission in hippocampal slices is mediated by extracellular adenosine. J. Neurosci., 19 (1999) 1932±1939. [12] Morrison, B., Saatman, K.E., Meaney, D.F. and McIntosh, T.K., In vitro central nervous system models of mechanically induced trauma: a review. J. Neurotrauma, 15 (1998) 911±928. [13] Pellmar, T., Electrophysiological correlates of peroxide damage in guinea pig hippocampus in vitro. Brain Res., 364 (1986) 377±381. [14] Pellmar, T.C., Peroxide alters neuronal excitability in the CA1 region of guinea-pig hippocampus in vitro. Neuroscience, 23 (1987) 447±456. [15] Pellmar, T.C., Gilman, S.C., Keyser, D.O., Lee, K.H., Lepinski, D.L., Livengood, D. and Myers Jr, L.S., Reactive oxygen species on neural transmission. Ann. N.Y. Acad. Sci., 738 (1994) 121±129. [16] Proctor, W.R. and Dunwiddie, T.V., Adenosine inhibits calcium spikes in hippocampal pyramidal neurons in vitro. Neurosci. Lett., 35 (1983) 197±201. [17] Rosler, M., Retz, W., Thome, J. and Riederer, P., Free radicals in Alzheimer's dementia: currently available therapeutic strategies. J. Neural Transm., 54 (1998) 211±219. [18] Rudolphi, K.A., Schubert, P., Parkinson, F.E. and Fredholm, B.B., Adenosine and brain ischemia. Cerebrovasc. Brain Metab. Rev., 4 (1992) 346±369. [19] Zhu, P.J. and Krnjevic, K., Anoxia selectively depresses excitatory synaptic transmission in hippocampal slices. Neurosci. Lett., 166 (1994) 27±30. [20] Zhu, P.J. and Krnjevic, K., Adenosine release mediates cyanide-induced suppression of CA1 neuronal activity. J. Neurosci., 17 (1997) 2355-2364. [21] Zhu, P.J. and Krnjevic, K., Endogenous adenosine on membrane properties of CA1 neurons in rat hippocampal slices during normoxia and hypoxia. Neuropharmacology, 36 (1997) 169-176.
Acute peroxide treatment of rat hippocampal slices induces adenosine-mediated inhibition of excitatory transmission in area CA1 Susan A. Masino a,*, Michael H. Mesches b, Paula C. Bickford b, Thomas V. Dunwiddie b a
Department of Pharmacology and Neuroscience Program, B138 UCHSC, 4200 E. 9th Avenue, Denver, CO 80262, USA b Department of Pharmacology and VA Medical Center, C236 UCHSC, 4200 E. 9th Avenue, Denver, CO 80262, USA Received 23 July 1999; received in revised form 16 August 1999; accepted 18 August 1999
Abstract Brief exposure to conditions that generate free radicals inhibits synaptic transmission in hippocampal slices, most likely via a presynaptic mechanism. Because other physiologically stressful conditions that generate free radicals, such as hypoxia or ischemia, stimulate the release of adenosine from brain slices, we determined whether increases in extracellular adenosine mediate the presynaptic inhibition of excitatory transmission induced by peroxide treatment. Simultaneous addition of hydrogen peroxide (0.01%) and ferrous sulfate (100 mM) resulted in a .80% decrease in synaptic potentials recorded in the CA1 region of hippocampal slices of adult male rats. Treatment with theophylline (200 mM), a non-selective adenosine receptor antagonist, or 8-cyclopentyl-1, 3-dipropylxanthine (100 nM), a selective adenosine A1 receptor antagonist, prior to and during hydrogen peroxide superfusion prevented the inhibition. These results demonstrate that acute exposure to hydrogen peroxide induces an adenosine-mediated decrease in synaptic transmission in hippocampal slices. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Theophylline; 8-Cyclopentyl-1, 3-dipropylxanthine; Adenosine; Hydrogen peroxide; Ferrous sulfate; CA1; Hippocampus; Excitatory postsynaptic potential; Adenosine receptors; A1 receptors
Reactive oxygen-derived free radicals are present in all tissues and their levels are maintained within normal limits by a variety of enzymes and antioxidants. Numerous studies have reported that increased levels of free radicals are associated with a variety of pathological conditions, including both acute neurotrauma and chronic neurodegenerative disease [7,12,17] and free radical species might contribute to the neuronal damage. Previous studies have explored the acute electrophysiological effects of free radical-generating conditions in vitro. Electrophysiological recordings from the CA1 region of the rat hippocampus have demonstrated that adding hydrogen peroxide (H2O2) or a combination of hydrogen peroxide and ferrous sulfate (FeSO4) to the superfusion buffer, which generates highly reactive hydroxyl radicals, results in an inhibition of synaptic potentials [9,13±15]. This has been observed both with extracellular ®eld recordings [9,13] and intracellular recordings of pyra* Corresponding author. Tel.: 11-303-315-7584 fax: 11-303315-4814. E-mail address: [email protected] (S.A. Masino)
midal cells [14]. As the response to direct iontophoretic application of either glutamate or g aminobutyric acid was unaffected during the peroxide superfusion, the effect of peroxide on hippocampal synaptic potentials was concluded to be presynaptic [14]. Concentrations of adenosine suf®cient to tonically activate adenosine receptors are normally found in the extracellular space in brain, and adenosine is a potent neuromodulator in many brain regions [5]. In the CA1 region of the hippocampus, one of the major actions of extracellular adenosine is a presynaptic inhibition of excitatory glutamatergic transmission, an effect mediated by adenosine A1 receptors [5]. Increased adenosine release is a common response to many physiologically stressful situations, such as hypoxia [21], hypoglycemia [2,8] and increased temperature [11]. Given that adenosine release is commonly induced by metabolic stress, and excessive oxygen-derived free radicals can result from and/or contribute to cell stress, the aim of the present study was to determine whether the inhibition of neurotransmission induced
0304-3940/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 9 9) 00 69 3- X
92
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
by superfusion of brain slices with hydrogen peroxide is similarly mediated by increases in the extracellular concentrations of adenosine. Transverse hippocampal slices were obtained from 6±8 week-old Sprague±Dawley rats using standard procedures previously described [11]. Rats were decapitated, the brains rapidly removed into ice-cold arti®cial cerebrospinal ¯uid (aCSF) and the hippocampus dissected free from the surrounding tissue. 400 mm slices were made on a Sorvall TC-2 tissue chopper and incubated at 32.58C. The aCSF used for dissection, incubation and submerged, superfused recordings contained (in mM): NaCl 126.0, KCl 3.0, MgSO4 1.5, CaCl2 2.4, NaH2PO4 1.2, NaHCO3 25.9, d-glucose 11 and was saturated with 95% O2 / 5%CO2. Slices were incubated undisturbed for 90 min before electrophysiological recording. Slices were placed on a nylon net in the recording chamber and superfused continuously (2.0 ml/min) with aCSF bubbled with humidi®ed with 95%O2 /5%CO2. Extracellular ®eld excitatory postsynaptic potentials (fEPSPs) were recorded from the CA1 region of the stratum radiatum using glass micropipettes (10±15 MV) ®lled with 3 M NaCl. A concentric bipolar stimulating electrode stimulated the Schaffer collateral and commissural ®bers in the stratum radiatum every 30 s. In some experiments a pair of synaptic stimuli with an interstimulus interval of 70 ms was delivered every 30 s. The stimulation intensity was adjusted such that the baseline fEPSP amplitude was between 0.7 and 1.2 mV. Data were recorded via an AC ampli®er, digitized and stored for later analysis. After establishing a baseline fEPSP, slices were either superfused with hydrogen peroxide (0.01%) and ferrous sulfate (100 mM), or pretreated with theophylline (a nonselective competitive adenosine receptor antagonist; 200 mM) and then superfused with H2O2/FeSO4. In initial experiments, application of hydrogen peroxide alone produced more variable results or a longer onset to a decrease in the fEPSP, as has been reported previously [13]. Thus, ferrous sulfate was added simultaneously with the hydrogen peroxide to increase tissue iron and promote formation of the hydroxyl radical via the Fenton reaction. Application of ferrous sulfate alone has no effect on synaptic transmission ([13] and unpublished data). All drugs were applied for a minimum of 10 min, or until a stable response was achieved for at least 5 min. There was only one application of H2O2/FeSO4 per slice due to the potential long term effects on membrane integrity due to lipid peroxidation. Theophylline, hydrogen peroxide (30% w/ w), ferrous sulfate and all aCSF constituents were obtained from Sigma (St. Louis, MO). 8-Cyclopentyl-1, 3-dipropylxanthine (DPCPX) was obtained from Research Biochemicals (Natick, MA). Statistical analyses included Student's two-tailed paired or unpaired t-tests where appropriate and the F-test for analysis of sample variance. Co-application of hydrogen peroxide (0.01%) and ferrous sulfate (100 mM) resulted in an 81 ^ 3:8% (mean ^ SEM;
n 6) decrease in the fEPSP. This decrease usually reversed upon washout (Fig. 1), but in some cases the fEPSP did not recover immediately (Fig. 3B). Fig. 1 illustrates the fEPSP recorded in stratum radiatum from a single slice prior to, during and following superfusion with peroxide. In this case the fEPSP decreased rapidly during the peroxide treatment and there was a nearly complete recovery following washout. In four slices tested using this protocol, the fEPSP recovered to 82 ^ 6.5% of the original baseline 35 min after removal of the H2O2/FeSO4. As evidenced by the individual traces of the synaptic response, the amplitude of the fEPSP decreased (Fig. 1; trace (a) vs. trace (b)) and the variance of the responses increased significantly (F 4:069; P , 0:05) upon addition of the H2O2/ FeSO4. In conjunction with this pronounced decrease in the fEPSP, the ratio between the amplitude of the second EPSP response (R2) and the ®rst (R1) increased signi®cantly during the peroxide treatment. The average ratio (R2/R1) during baseline recording was 1:35 ^ 0:03 and increased during H2O2/FeSO4 superfusion to 1:55 ^ 0:09 (n 6; P , 0:05). This increased synaptic facilitation during paired pulse stimulation is typically observed with presynaptic inhibition of transmitter release, previously hypothesized to underlie the fEPSP decrease observed with peroxide. Adenosine exerts its effects on synaptic transmission in the hippocampus primarily via presynaptic inhibition of excitatory transmission and an increase in paired pulse facilitation is observed either during the application of exogenous adenosine [5,6], or during an increase in endogenous adenosine [2]. To determine whether the inhibition during peroxide treatment was due speci®cally to activation of presynaptic
Fig. 1. The effect of hydrogen peroxide/ferrous sulfate (H2O2/ FeSO4) superfusion on fEPSP amplitude. Superfusion with H2O2/FeSO4 (duration of superfusion indicated by the bar along the abscissa) caused a profound decrease in the fEPSP and in this slice, the fEPSP response recovered slowly and almost completely upon washout. Synaptic responses (average of six to eight sweeps) before, during and after the H2O2/FeSO4 superfusion are indicated by (a±c) and superimposed at (d). Traces shown are the ®rst response to a 70 ms paired-pulse stimulation. Scale bars, 1 mV and 10 ms.
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
Fig. 2. Theophylline antagonism of the response to H2O2/FeSO4. Pretreatment with the competitive adenosine receptor antagonist theophylline (200 mM; duration of superfusion indicated by dashed line) prevented the normal decrease in the fEPSP induced by H2O2/FeSO4. In most slices, theophylline completely prevented any decrease in response to H2O2/FeSO4 (average change 1:0 ^ 4:7%, n 3). Synaptic responses before, during and after the H2O2/FeSO4 are indicated by (a±c) and superimposed at d. Scale bars, 1 mV and 10 ms.
adenosine receptors, in some experiments the adenosine receptor antagonist theophylline (200 mM) was applied prior to the H2O2/FeSO4 superfusion. Theophylline alone produced an increase in the fEPSP (16 ^ 3:5%; n 3), re¯ecting antagonism of the inhibitory effect of endogenous adenosine [5]. After theophylline superfusion for 10 min, H2O2/FeSO4 was applied in the continued presence of theophylline. Theophylline completely prevented any decrease of the fEPSP due to the peroxide treatment (average decrease in the presence of theophylline: 1:0 ^ 4:7%, n 3; P , 0:0001 compared with decrease in H2O2/FeSO4 alone). An example of the fEPSP response to H2O2/FeSO4 in the presence of theophylline is shown in Fig. 2. In a separate set of experiments, a selective adenosine A1 receptor antagonist, DPCPX (100 nM), was applied. As with theophylline, DPCPX alone produced an increase in the fEPSP (32 ^ 6:4%; n 4). DPCPX also signi®cantly reduced the inhibition observed upon application of the H2O2/FeSO4 (average decrease in the presence of DPCPX: 15 ^ 10%, n 4; P , 0:0001 compared with decrease in H2O2/FeSO4 alone). In addition to preventing the decrease in the amplitude of the fEPSP when applied prior to the peroxide treatment, theophylline reversed the inhibition even in the continued presence of H2O2/FeSO4 (Fig. 3A), or when the amplitude of the fEPSP remained depressed after the washout of peroxide from the slice (Fig. 3B). In both cases, theophylline completely reversed the depression of the fEPSP. In four slices in which the fEPSP was inhibited by 84 ^ 5:0% by H2O2/ FeSO4, the amplitude of the response following theophylline superfusion was not signi®cantly different from the original baseline (0:98 ^ 0:17 mV vs. 1:06 ^ 0:09 mV, respectively, n 4; P . 0:05).
93
Although the depressant effect of H2O2/FeSO4 on hippocampal synaptic responses is well-established [9,13,14] and a presynaptic mechanism underlying this depression had been inferred [14], the speci®c mechanism by which transmitter release was inhibited has been unknown. In the present study we show that presynaptic activation of adenosine A1 receptors is the primary cause of synaptic inhibition during peroxide treatment. The adenosine A1 receptor is responsible for synaptic inhibition in other paradigms, such as anoxia [19], metabolic inhibition [20] and increased temperature [11], which all result in a similar adenosinemediated decrease in synaptic transmission. Thus, the acute depressant effect of peroxide appears to re¯ect a marked increase in the concentration of endogenous adenosine in the extracellular space, acting on A1 receptors to inhibit glutamate release. It should be noted that free radical generation has multiple cellular consequences [4] and can result in pathological changes such as apoptosis or necrosis [3]; at this point, the potential role of adenosine in these effects of peroxide is not clear. A previous study determined that sodium and potassium currents were not modi®ed and thus were not involved in the
Fig. 3. Theophylline reversal of the inhibition induced by H2O2/ FeSO4. Addition of theophylline (200 mM) either during superfusion with H2O2/FeSO4 (A), or following washout (B) completely reversed the synaptic depression induced by H2O2/FeSO4. This was the case even in slices that did not recover during the initial part of the washout period (cf. Fig. 3B with the recovery in Fig. 1). In general, theophylline returned the fEPSP amplitude to approximately the original baseline.
94
S.A. Masino et al. / Neuroscience Letters 274 (1999) 91±94
reduction of synaptic potential induced by peroxide treatment [14]. The increase in the threshold for activation of Ca 21 spikes reported by Pellmar [14] is also consistent with increased extracellular adenosine, because we have reported that adenosine increases the threshold for Ca 21 spikes [16] and other groups have shown more directly an adenosine A1 receptor-mediated inhibition of hippocampal Ca 21 channel activity [1]. It is not surprising that adenosine mediates the synaptic inhibition induced by free radicals, because increased extracellular adenosine is a common response to many stressful conditions. Free radicals might be intermediaries in some of these situations by contributing to the generation of extracellular adenosine and the resulting presynaptic inhibition via adenosine A1 receptors. There are likely multiple short and long term consequences of free radical generation in hippocampal slices and our data suggest only that adenosine might mediate some of the acute effects of peroxide. The incomplete recovery of the fEPSP that was occasionally observed could be due to the continued presence of adenosine, incomplete washout of the H2O2/FeSO4, continued presence of the free radical species in the tissue, or other effects of the treatment on synaptic transmission unrelated to adenosine receptors. As with hypoxia or ischemia, the increased extracellular adenosine release during superfusion with H2O2/FeSO4 may be neuroprotective, perhaps by reducing metabolic demand and limiting excitotoxicity [10,18]. However, it is not known whether presynaptic inhibition of glutamate release by adenosine would interact in any way with some of the other major actions of free radicals, such as lipid peroxidation and subsequent loss of membrane integrity. Any neuroprotective effect of the increased adenosinemediated presynaptic inhibition during free radical stress remains to be determined. [1] Ambrosio, A.F., Malva, J.O., Carvalho, A.P. and Carvalho, C.M., Inhibition of N-,P/Q- and other types of Ca 21 channels in rat hippocampal nerve terminals by the adenosine A1 receptor. Eur. J. Pharmacol., 340 (1997) 301±310. [2] Calabresi, P., Centonze, D., Pisani, A. and Bernardi, G., Endogenous adenosine mediates the presynaptic inhibition induced by aglycemia at corticostriatal synapses. J. Neurosci., 17 (1997) 4509±4516. [3] Clement, M.V., Ponton, A. and Pervaiz, S., Apoptosis induced by hydrogen peroxide is mediated by decreased superoxide anion concentration and reduction of intracellular milieu. FEBS Lett., 440 (1998) 13±18. [4] Cochrane, C.G., Cellular injury by oxidants. Am. J. Med., 91 (1991) 23S±30S.
[5] Dunwiddie, T.V., The physiological role of adenosine in the central nervous system. Int. Rev. Neurobiol., 27 (1985) 63± 139. [6] Dunwiddie, T.V. and Haas, H.L., Adenosine increases synaptic facilitation in the in vitro rat hippocampus: evidence for a presynaptic site of action. J. Physiol. (Lond.), 369 (1985) 365±377. [7] Farooqui, A.A. and Horrocks, L.A., Lipid peroxides in the free radical pathophysiology of brain diseases. Cell. Mol. Neurobiol., 18 (1998) 599±608. [8] Fowler, J.C., Purine release and inhibition of synaptic transmission during hypoxia and hypoglycemia in rat hippocampal slices. Neurosci. Lett., 157 (1993) 83±86. [9] Fowler, J.C., Hydrogen peroxide opposes the hypoxic depression of evoked synaptic transmission in rat hippocampal slices. Brain Res., 766 (1997) 255±258. [10] Fredholm, B.B., Adenosine and neuroprotection. Int. Rev. Neurobiol., 40 (1997) 259±280. [11] Masino, S.A. and Dunwiddie, T.V., Temperature-dependent modulation of excitatory transmission in hippocampal slices is mediated by extracellular adenosine. J. Neurosci., 19 (1999) 1932±1939. [12] Morrison, B., Saatman, K.E., Meaney, D.F. and McIntosh, T.K., In vitro central nervous system models of mechanically induced trauma: a review. J. Neurotrauma, 15 (1998) 911±928. [13] Pellmar, T., Electrophysiological correlates of peroxide damage in guinea pig hippocampus in vitro. Brain Res., 364 (1986) 377±381. [14] Pellmar, T.C., Peroxide alters neuronal excitability in the CA1 region of guinea-pig hippocampus in vitro. Neuroscience, 23 (1987) 447±456. [15] Pellmar, T.C., Gilman, S.C., Keyser, D.O., Lee, K.H., Lepinski, D.L., Livengood, D. and Myers Jr, L.S., Reactive oxygen species on neural transmission. Ann. N.Y. Acad. Sci., 738 (1994) 121±129. [16] Proctor, W.R. and Dunwiddie, T.V., Adenosine inhibits calcium spikes in hippocampal pyramidal neurons in vitro. Neurosci. Lett., 35 (1983) 197±201. [17] Rosler, M., Retz, W., Thome, J. and Riederer, P., Free radicals in Alzheimer's dementia: currently available therapeutic strategies. J. Neural Transm., 54 (1998) 211±219. [18] Rudolphi, K.A., Schubert, P., Parkinson, F.E. and Fredholm, B.B., Adenosine and brain ischemia. Cerebrovasc. Brain Metab. Rev., 4 (1992) 346±369. [19] Zhu, P.J. and Krnjevic, K., Anoxia selectively depresses excitatory synaptic transmission in hippocampal slices. Neurosci. Lett., 166 (1994) 27±30. [20] Zhu, P.J. and Krnjevic, K., Adenosine release mediates cyanide-induced suppression of CA1 neuronal activity. J. Neurosci., 17 (1997) 2355-2364. [21] Zhu, P.J. and Krnjevic, K., Endogenous adenosine on membrane properties of CA1 neurons in rat hippocampal slices during normoxia and hypoxia. Neuropharmacology, 36 (1997) 169-176.