- Email: [email protected]
Neuroprotective effects of interleukin-10 and tumor necrosis factor-α against hypoxia-induced hyperexcitability in hippocampal slice neurons
Neuroscience Letters 416 (2007) 236–240
Neuroprotective effects of interleukin-10 and tumor necrosis factor-␣ against hypoxia-induced hyperexcitability in hippocampal slice neurons Maria E. Burkovetskaya, Sergei G. Levin, Oleg V. Godukhin ∗ Institute of Theoretical and Experimental Biophysiscs of the Russian Academy of Sciences, Pushchino, Moscow Region 142290, Russia Received 26 September 2006; received in revised form 25 November 2006; accepted 8 December 2006
Abstract In our previous experiments we have demonstrated that repeated exposures of rat hippocampal slices to brief episodes of hypoxia induce a sustained decrease in the threshold of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons. The aim of this study was to investigate the comparative effects of interleukin-10 (IL-10) and tumor necrosis factor-␣ (TNF-␣) on the hyperexcitability of CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. The method of field potentials measurement in CA1 region of hippocampal slices have been described in our previous work [O. Godukhin, A. Savin, S. Kalemenev, S. Levin, Neuronal hyperexcitability induced by repeated brief episodes of hypoxia in rat hippocampal slices: involvement of ionotropic glutamate receptors and L-type Ca2+ channels, Neuropharmacology 42 (2002) 459–466]. The principal results of our work are summarized as follow. Pro-inflammatory cytokine TNF-␣ (0.8, 4 and 20 ng/ml) and anti-inflammatory cytokine IL-10 (1 and 10 ng/ml) significantly reduced the hyperexcitability in CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. The neuroprotective effects of IL-10 and TNF-␣ against the hypoxia-induced hyperexcitability were mediated by anti-hypoxic actions of these cytokines through, possibly, mechanism of preconditioning. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Hippocampal slices; Hypoxia; Hyperexcitability; Tumor necrosis factor-alpha; Interleukin-10; Population spike; Epileptiform activity
It is known that hippocampal formation, especially CA1 region, is highly sensitive to hypoxia/ischaemia, and these conditions can induce long-lasting functional modifications leading to a decrease in the threshold of seizure activity generation [9,20]. In our previous experiments we have demonstrated that repeated exposures of rat hippocampal slices to brief episodes of hypoxia induce a sustained decrease in the threshold of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons [6]. This hypoxia-induced hyperexcitability in CA1 pyramidal neurons critically depends on functional activities of L-type voltagegated Ca2+ -channels and ionotropic glutamate receptors but not GABAA and GABAB receptors [6,11]. There is some evidence for the involvement of pro-inflammatory cytokine tumor necrosis factor-␣ (TNF-␣) and antiinflammatory cytokine interleukin-10 (IL-10) in experimental
∗
Corresponding author. Tel.: +7 4967 739177; fax: +7 4967 330553. E-mail addresses: [email protected], [email protected] (O.V. Godukhin). 0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.12.059
models of seizures and in human epilepsy [1,7,8,18,19,21]. However, the functional roles of these cytokines in seizure activity either unknown (for IL-10) or controversial (for TNF-␣). Expression of IL-10 is elevated during the course of most major diseases in the brain and promotes survival of neurons and all glial cells by blocking the effects of pro-inflammatory cytokines, including TNF-␣, IL-1, and IL-6 [19]. Furthermore, IL-10 blocks the inhibitory effect of IL-1 on long-term potentiation in the hippocampus [10]. Although the functions of IL-10 in the brain are most often associated with increased animal survival, in certain cases the potent anti-inflammatory property of IL-10 leads to the progression of disease and death [19]. Currently, the functional role of IL-10 in seizures is unknown. The role of pro-inflammatory cytokine TNF-␣ in the seizure activity development is controversial. Some authors have demonstrated that TNF-␣ prolonged epileptiform discharges in amygdala kindled rats [16]. While the other investigators have shown that TNF-␣ inhibited kainic acid-induced seizures in mice, and this neuroprotective action against seizures is mediated by neuronal p75 receptors [2]. It is proposed that this
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
cytokine can cause both detrimental and neuroprotective effects on neuronal activity, depending on its local concentration, the type of target brain cells, the receptor subtypes and the experimental models of epileptogenesis [1,2,22]. In our study, we investigated the comparative effects of IL-10 and TNF-␣ on the hyperexcitability of CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. All experiments were carried out with male Wistar rats (60–70 days old; n = 60). The use of animals was in accordance with the UK Animals (Scientific Procedures) Act 1986. Transverse hippocampal slices (250–300 m thick) were prepared with a tissue chopper and placed into a recording chamber (submersion type). Slices were superfused at 2.5 ml/min with the artificial cerebrospinal fluid (ACSF) maintained at 32 ◦ C. The ACSF composition was (mM): 124 NaCl, 3 KCl, 1.25 KH2 PO4 , 2 MgSO4 , 2 CaCl2 , 26 NaHCO3 , 10 d-glucose; pH 7.4 was adjusted with 95% O2 /5% CO2 . Slices were allowed to recover for 5 h before data collection. This incubation was necessary because the literature findings have demonstrated that the level of some cytokines was relatively high in hippocampal slices incubated for shorter periods [15]. The method of field potentials measurement in CA1 region of hippocampal slices have been described in our previous works [6,11]. Briefly, population spikes (PSs) of CA1 pyramidal neurons in stratum pyramidale were recorded with glass microelectrode (2–5 M) in response to electrical stimulation (0.1 ms, 50–350 A) of Schaffer collateral/commissural fibers (SCH). PS amplitude (mV) were measured for series of 7 separate single current pulses with increasing intensity (from minimum to maximum values for PS generation) applied at 10 s interval. A delay of 20 min separated each group of 7 stimuli from the next. Three hypoxic episodes (3 min duration each with the 10 min interval) were produced by switching from ACSF equilibrated with 95% O2 /5% CO2 to ACSF equilibrated with 95% N2 /5% CO2 . The effectiveness of hypoxic episode (Th ) was evaluated by employing the following formula (Fig. 2A): Th = the time point of 50% recovery of PS amplitude (Thr ) − the time point of 50% depression of PS amplitude (Thd ). The appearance of multiple PSs in the PS response to single electrical stimulus was taken as indication of the development of epileptiform activity. Two parameters of such activity were measured: (1) the stimulus intensity (A) of the appearance of the additional (second) PS was characterized as the threshold of generation of an additional PS in the PS response (TASG), and (2) the number of PSs in the PS response (NPS) measured for the current intensity that induced an additional PS before hypoxic episodes or in the same time points (−20 to 0 min) in the control experiments (without hypoxic episodes applied). IL-10 (1 and 10 ng/ml; from Chemicon, catalog # IL035) and TNF-␣ (0.8, 4 and 20 ng/ml; from Chemicon, catalog # GF046) were applied for 10 min before and together with hypoxic episodes (the total time—40 min). All electrophysiological data were digitized at 20 kHz and analyzed using a computer with software developed in house for
237
the measurements of PS amplitude, TASG and NPS. All values in the Results section are given as mean ± S.E.M. The effects of treatments were analyzed by one-way analysis of variance (ANOVA) followed by multiple-comparisons tests (Scheffe’s tests). A P ≤ 0.05 was considered as significant. Under normal conditions (without hypoxic episodes applied), IL-10 (1 and 10 ng/ml) and TNF-␣ (0.8, 4 and 20 ng/ml) did not change the amplitude of PS and the values of TASG and NPS (data is not shown). Hypoxic episodes abolished the PSs (during episodes) (Fig. 1A) and induced a post-hypoxic sustained decrease in the TASG (data is not shown) and an increase in the NPS (Fig. 1B). Fig. 1B illustrates the effects of IL-10 (1 ng/ml) and TNF-␣ (0.8 ng/ml) on the time courses of NPS produced by repeated brief hypoxic episodes. These cytokines significantly depressed an increase in the NPS in a dose-dependent manner. Fig. 1C shows that IL-10 (1 ng/ml) more effectively depressed a hypoxia-induced increase in the NPS than IL-10 (10 ng/ml): 1 ng/ml—on 60% and 10 ng/ml—on 38.5% (140 min after hypoxia). In contrast to IL-10, the depressive effect of TNF␣ had a reversed dose–response relationship (Fig. 1C). TNF-␣ more effectively depressed a hypoxia-induced increase in the NPS in the concentration of 20 ng/ml than in concentrations of 4 and 0.8 ng/ml: 0.8 ng/ml—on 33.3%, 4 ng/ml—on 45% and 20 ng/ml—on 55% (140 min after hypoxia). However, from the above-mentioned results and the findings received in our previous work [12] it is not clear whether these cytokines depress the mechanisms of hypoxiainduced hyperexcitability (the production of multiple PSs) expressed after hypoxic episodes or depress the effectiveness of hypoxia on functional activity of hippocampal slice tissue during hypoxic episode (antihypoxic action) leading to an abolishment of post-hypoxic hyperexcitability. To examine this issue, we determine the effects of IL-10 and TNF-␣ on the time courses of PS amplitude depression during hypoxic episodes. Fig. 2A demonstrates the time courses of PS amplitude depression during: single hypoxic episode, single hypoxic episode + IL-10 (1 ng/ml) and single hypoxic episode + TNF-␣ (0.8 ng/ml). The effectiveness of hypoxic episodes (Th ) were: (1) single hypoxic episode − 293 ± 10 s; (2) single hypoxic episode + IL-10 (1 ng/ml) − 218 ± 11 s and (3) single hypoxic episode + TNF-␣ (0.8 ng/ml) − 189 ± 27 s. Thus, these results indicated that both cytokines significantly reduce the depressive effect of hypoxia on PS amplitude. This neuroprotective function of IL-10 and TNF-␣ against the hypoxia-induced depression of PS amplitude looks like preconditioning. Fig. 2B demonstrates that there are significant delays in the time points of a 50% depression of PS amplitude (Thd ) during the second and the third hypoxic episodes relatively to the Thd during the first episode (preconditioning by the first episode against the second and the third hypoxic episodes): Thd (1 episode) = 56 ± 4 s; Thd (2 episode) = 89 ± 5 s (P < 0.05); Thd (3 episode) = 100 ± 7 s (P < 0.05). However, an application of IL-10 or TNF-␣ prolonged the Thd during the first hypoxic episode (preconditioning by the cytokines against the first hypoxic episode) (Fig. 2C and D). The values of Thd during the first, second and third hypoxic episodes under the cytokine treatment were approximately the same: (1) IL-10: 1
238
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
Fig. 1. The effects of TNF-␣ and IL-10 on the development of the PS bursts in CA1 pyramidal neurons induced by hypoxic episodes. (A) Representative examples of typical PS responses before (0 min), during (Hypoxic episode), and 140 min (140 min) after 3 × 3 min episodes of hypoxia, hypoxia + TNF-␣ (0.8 ng/ml) and hypoxia + IL-10 (1 ng/ml). (B) Time courses of the NPS values before (−20 to 0) and after (+40 to +140) episodes (↓↓↓) of hypoxia (n = 10), hypoxia + IL-10 (1 ng/ml) (n = 10) and hypoxia + TNF-␣ (0.8 ng/ml) (n = 10). The symbol ‘*’ indicate the P < 0.05 significance level. (C) IL-10- and TNF-␣-induced dose-dependent depressions of a post-hypoxic increase in the NPS (100%) at 140 min after hypoxic episodes.
episode − 79 ± 4 s; 2 episode − 85 ± 3 s; 3 episode − 84 ± 4 s; (2) TNF-␣: 1 episode − 110 ± 16 s; 2 episode − 115 ± 11 s; 3 episode − 114 ± 14 s; (Fig. 2C and D). Some findings indicate that the preconditioning effects of IL-10 and TNF-␣ against transient hypoxia are very plausible because the intracellular signaling processes involved in ischaemic/hypoxic preconditioning and cellular response to IL-10 and TNF-␣ share the same steps, for example mitogen-
activated protein kinase (MAPK)/extracellular regulated kinase (ERK) [1,4,5,19]. There is a direct evidence that some cytokines can mediate ischaemic tolerance in hippocampal CA1 neurons [13]. In most studies, preconditioning required several hours to develop. However, some authors demonstrated that 1 min anoxic episode improved the recovery of PSs after a second anoxic episode separated by 10 min of reoxygenation in CA1 region of hippocampal slices [14]. These findings are in the
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
239
Fig. 2. Time courses of PS amplitude depression during hypoxic episodes (the solid lines on the plots). (A) Time courses of PS amplitude depressions during the first hypoxic episode (n = 10), the first hypoxic episode + TNF-␣ (0.8 ng/ml) (n = 10) and the first hypoxic episode + IL-10 (1 ng/ml) (n = 10). (B) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia. (C) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia + TNF-␣ (0.8 ng/ml). (D) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia + IL-10 (1 ng/ml).
line of our results. In our experiments, the TNF-␣- and IL-10induced neuroprotective effects against 3 min hypoxic episode were observed in 10 min after application of these cytokines to the hippocampal slice medium. Although brief cerebral ischaemia, or cerebral hypoxia, serve as prototypical preconditioning stimuli, many exogenously delivered agents (for example, inflammatory cytokines, metabolic inhibitors and anaesthetics) can also induce pre-
conditioning effect [5]. To induce neuroprotection, the preconditioning stimuli must activate transduction pathways that initiate the adaptive response. Although dependent in part on the nature of the preconditioning stimulus, members of these transduction pathways for which there is strong general support include mitogen-activated protein kinases (late preconditioning) and ATP-sensitive K+ channels (rapid preconditioning) [4,5,14].
240
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
A surprising findings in our work was that pro-inflammatory cytokine TNF-␣ and anti-inflammatory cytokine IL-10 cause the similar neuroprotective effect on the hypoxia-induced hyperexcitability in CA1 hippocampal slice neurons. Recent data indicates that TNF-␣ can cause both detrimental and neuroprotective effects on neuronal activity [1,2,22]. These effects can be mediated by two distinct receptors, TNFR1 and TNFR2 [1]. Some results suggest that TNF-␣ can regulate neuronal circuit homeostasis inducing an increase and a decrease in surface expression of AMPA and GABAA receptors, respectively, via TNFR1 [17]. In addition, TNF-␣ can protect cultured neurons against glucose-deprived-injury and amino acid toxicity [3]. The function of anti-inflammatory cytokine IL-10 are mainly associated with promoting of cell survival in response to insults [19]. Thus, in certain circumstances, IL-10 and TNF-␣ can act in similar neuroprotective manner. Thus, our findings showed that neuroprotective effects of IL10 and TNF-␣ against the hyperexcitability in CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slice are mediated by anti-hypoxic actions of these cytokines through, possibly, mechanism of preconditioning. Acknowledgement This work was supported by grants from the Russian Foundation of Basic Research (no. 05-04-48907). References [1] B.C. Albensy, Potential roles for tumor necrosis factor and nuclear factorkB in seizure activity, J. Neurosci. Res. 66 (2001) 151–154. [2] S. Balosso, T. Ravizza, C. Perego, J. Peschon, I.L. Campbell, M.G. De Simoni, A. Vezzani, Tumor necrosis factor-␣ inhibits seizures in mice via p75 receptors, Ann. Neurol. 57 (2005) 804–812. [3] B. Cheng, S. Christakos, M.P. Mattson, Tumor necrosis factors protect neurons against metabolic-excitotoxic insults and promote maintenance of calcium homeostasis, Neuron 12 (1994) 139–153. [4] V.L. Dawson, T.M. Dawson, Neuronal ischaemic preconditioning, TIPS 21 (2000) 423–424. [5] J.M. Gidday, Cerebral preconditioning and ischaemic tolerance, Nat. Neurosci. 7 (2006) 437–448. [6] O. Godukhin, A. Savin, S. Kalemenev, S. Levin, Neuronal hyperexcitability induced by repeated brief episodes of hypoxia in rat hippocampal slices: involvement of ionotropic glutamate receptors and L-type Ca2+ channels, Neuropharmacology 42 (2002) 459–466.
[7] K.R. Goldstein, R. Bhatt, B.E. Barton, S.S. Zalcman, P. Rameshwar, A. Siegel, Effects of hemispheric lateralization and site specificity on immune alterations induced by kindled temporal lobe seizures, Brain Behav. Immun. 16 (2002) 706–719. [8] J.L. Jankowsky, P.H. Patterson, The role of cytokines and growth factors in seizures and their sequelae, Prog. Neurobiol. 63 (2001) 125– 149. [9] F.E. Jensen, C. Wang, C.E. Stafstrom, Z. Liu, C. Geary, M.C. Stevens, Acute and chronic increases in excitability in rat hippocampal slices after perinatal hypoxia in vivo, J. Neurophysiol. 79 (1998) 73–81. [10] A. Kelly, A.A. Lynch, A.E. Vereker, Y. Nolan, P. Queenan, P.E. Whittaker, L.A.J. O’Neill, M.A. Lynch, The anti-inflammatory cytokine, interleukin10, blocks the inhibitory effect of IL-1 on long term potentiation, J. Biol. Chem. 276 (2001) 45564–45572. [11] S.G. Levin, O.V. Godukhin, Hyperexcitability of CA1 pyramidal neurons induced by repeated brief episodes of hypoxia in the hippocampal slices: involvement of GABAA and GABAB receptors, Russ. Physiol. J. 91 (2005) 581–586. [12] S.G. Levin, O.V. Godukhin, Protective effect of interleukin-10 on the development of epileptiform activity induced by brief hypoxic episodes in the rat hippocampal slices, Zh. Vyssh. Nerv. Deiat. 56 (2006) 379– 383. [13] T. Ohtsuki, C.A. Ruetzler, K. Tasaki, J.M. Hallenbeck, Interleukin-1 mediates induction of tolerance to global ischaemia in gerbil hippocampal CA1 neurons, J. Cereb. Blood Flow Metab. 16 (1996) 1137–1142. [14] M.A. Perez-Pinzon, J.M. Born, J.M. Centeno, Calcium and increase excitability promote tolerance against anoxia in hippocampal slices, Brain Res. 833 (1999) 20–26. [15] H. Schneider, F. Pitossi, D. Balschun, A. Wagner, A. Del Rey, A neuromodulatory role of interleukin-1 in the hippocampus, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 7778–7783. [16] A.A. Shandra, L.S. Godlevsky, R.S. Vastyanov, A.A. Oleinik, V.L. Konovalenko, E.N. Rapoport, N.N. Korobka, The role of TNF-␣ in amygdale kindled rats, Neurosci. Res. 42 (2002) 147–153. [17] D. Stellwagen, E.C. Beattie, J.Y. Seo, R.C. Malenka, Differential regulation of AMPA receptor and GABA receptor trafficking by tumor necrosis factor␣, J. Neurosci. 25 (12) (2005) 3219–3228. [18] R. Straussberg, J. Amir, L. Harel, I. Punsky, H. Bessler, Pro- and antiinflammatory cytokines in children with febrile convulsions, Pediatr. Neurol. 24 (2001) 49–53. [19] K. Strle, J.-H. Zhou, W.-H. Shen, S.R. Broussard, R.W. Johnson, G.G. Freund, R. Dantzer, K.W. Kelly, Interleukin-10 in the brain, Crit. Rev. Immunol. 21 (5) (2001) 427–449. [20] V. Tancredi, Long-lasting changes in synaptic excitability induced by anoxia in the rat hippocampus, Progr. Neuro-Pychopharmacol. 21 (1997) 211–232. [21] A. Vezzani, D. Moneta, C. Richichi, C. Perego, M.G. De Simoni, Functional role of pro-inflammatory and anti-inflammatory cytokines in seizures, Adv. Exp. Med. Biol. 548 (2004) 123–133. [22] C.X. Wang, A. Shuaib, Involvement of inflammatory cytokines in central nervous system injury, Prog. Neirobiol. 67 (2002) 161–172.
Neuroprotective effects of interleukin-10 and tumor necrosis factor-␣ against hypoxia-induced hyperexcitability in hippocampal slice neurons Maria E. Burkovetskaya, Sergei G. Levin, Oleg V. Godukhin ∗ Institute of Theoretical and Experimental Biophysiscs of the Russian Academy of Sciences, Pushchino, Moscow Region 142290, Russia Received 26 September 2006; received in revised form 25 November 2006; accepted 8 December 2006
Abstract In our previous experiments we have demonstrated that repeated exposures of rat hippocampal slices to brief episodes of hypoxia induce a sustained decrease in the threshold of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons. The aim of this study was to investigate the comparative effects of interleukin-10 (IL-10) and tumor necrosis factor-␣ (TNF-␣) on the hyperexcitability of CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. The method of field potentials measurement in CA1 region of hippocampal slices have been described in our previous work [O. Godukhin, A. Savin, S. Kalemenev, S. Levin, Neuronal hyperexcitability induced by repeated brief episodes of hypoxia in rat hippocampal slices: involvement of ionotropic glutamate receptors and L-type Ca2+ channels, Neuropharmacology 42 (2002) 459–466]. The principal results of our work are summarized as follow. Pro-inflammatory cytokine TNF-␣ (0.8, 4 and 20 ng/ml) and anti-inflammatory cytokine IL-10 (1 and 10 ng/ml) significantly reduced the hyperexcitability in CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. The neuroprotective effects of IL-10 and TNF-␣ against the hypoxia-induced hyperexcitability were mediated by anti-hypoxic actions of these cytokines through, possibly, mechanism of preconditioning. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Hippocampal slices; Hypoxia; Hyperexcitability; Tumor necrosis factor-alpha; Interleukin-10; Population spike; Epileptiform activity
It is known that hippocampal formation, especially CA1 region, is highly sensitive to hypoxia/ischaemia, and these conditions can induce long-lasting functional modifications leading to a decrease in the threshold of seizure activity generation [9,20]. In our previous experiments we have demonstrated that repeated exposures of rat hippocampal slices to brief episodes of hypoxia induce a sustained decrease in the threshold of stimulus-evoked epileptiform discharges in CA1 pyramidal neurons [6]. This hypoxia-induced hyperexcitability in CA1 pyramidal neurons critically depends on functional activities of L-type voltagegated Ca2+ -channels and ionotropic glutamate receptors but not GABAA and GABAB receptors [6,11]. There is some evidence for the involvement of pro-inflammatory cytokine tumor necrosis factor-␣ (TNF-␣) and antiinflammatory cytokine interleukin-10 (IL-10) in experimental
∗
Corresponding author. Tel.: +7 4967 739177; fax: +7 4967 330553. E-mail addresses: [email protected], [email protected] (O.V. Godukhin). 0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.12.059
models of seizures and in human epilepsy [1,7,8,18,19,21]. However, the functional roles of these cytokines in seizure activity either unknown (for IL-10) or controversial (for TNF-␣). Expression of IL-10 is elevated during the course of most major diseases in the brain and promotes survival of neurons and all glial cells by blocking the effects of pro-inflammatory cytokines, including TNF-␣, IL-1, and IL-6 [19]. Furthermore, IL-10 blocks the inhibitory effect of IL-1 on long-term potentiation in the hippocampus [10]. Although the functions of IL-10 in the brain are most often associated with increased animal survival, in certain cases the potent anti-inflammatory property of IL-10 leads to the progression of disease and death [19]. Currently, the functional role of IL-10 in seizures is unknown. The role of pro-inflammatory cytokine TNF-␣ in the seizure activity development is controversial. Some authors have demonstrated that TNF-␣ prolonged epileptiform discharges in amygdala kindled rats [16]. While the other investigators have shown that TNF-␣ inhibited kainic acid-induced seizures in mice, and this neuroprotective action against seizures is mediated by neuronal p75 receptors [2]. It is proposed that this
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
cytokine can cause both detrimental and neuroprotective effects on neuronal activity, depending on its local concentration, the type of target brain cells, the receptor subtypes and the experimental models of epileptogenesis [1,2,22]. In our study, we investigated the comparative effects of IL-10 and TNF-␣ on the hyperexcitability of CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slices. All experiments were carried out with male Wistar rats (60–70 days old; n = 60). The use of animals was in accordance with the UK Animals (Scientific Procedures) Act 1986. Transverse hippocampal slices (250–300 m thick) were prepared with a tissue chopper and placed into a recording chamber (submersion type). Slices were superfused at 2.5 ml/min with the artificial cerebrospinal fluid (ACSF) maintained at 32 ◦ C. The ACSF composition was (mM): 124 NaCl, 3 KCl, 1.25 KH2 PO4 , 2 MgSO4 , 2 CaCl2 , 26 NaHCO3 , 10 d-glucose; pH 7.4 was adjusted with 95% O2 /5% CO2 . Slices were allowed to recover for 5 h before data collection. This incubation was necessary because the literature findings have demonstrated that the level of some cytokines was relatively high in hippocampal slices incubated for shorter periods [15]. The method of field potentials measurement in CA1 region of hippocampal slices have been described in our previous works [6,11]. Briefly, population spikes (PSs) of CA1 pyramidal neurons in stratum pyramidale were recorded with glass microelectrode (2–5 M) in response to electrical stimulation (0.1 ms, 50–350 A) of Schaffer collateral/commissural fibers (SCH). PS amplitude (mV) were measured for series of 7 separate single current pulses with increasing intensity (from minimum to maximum values for PS generation) applied at 10 s interval. A delay of 20 min separated each group of 7 stimuli from the next. Three hypoxic episodes (3 min duration each with the 10 min interval) were produced by switching from ACSF equilibrated with 95% O2 /5% CO2 to ACSF equilibrated with 95% N2 /5% CO2 . The effectiveness of hypoxic episode (Th ) was evaluated by employing the following formula (Fig. 2A): Th = the time point of 50% recovery of PS amplitude (Thr ) − the time point of 50% depression of PS amplitude (Thd ). The appearance of multiple PSs in the PS response to single electrical stimulus was taken as indication of the development of epileptiform activity. Two parameters of such activity were measured: (1) the stimulus intensity (A) of the appearance of the additional (second) PS was characterized as the threshold of generation of an additional PS in the PS response (TASG), and (2) the number of PSs in the PS response (NPS) measured for the current intensity that induced an additional PS before hypoxic episodes or in the same time points (−20 to 0 min) in the control experiments (without hypoxic episodes applied). IL-10 (1 and 10 ng/ml; from Chemicon, catalog # IL035) and TNF-␣ (0.8, 4 and 20 ng/ml; from Chemicon, catalog # GF046) were applied for 10 min before and together with hypoxic episodes (the total time—40 min). All electrophysiological data were digitized at 20 kHz and analyzed using a computer with software developed in house for
237
the measurements of PS amplitude, TASG and NPS. All values in the Results section are given as mean ± S.E.M. The effects of treatments were analyzed by one-way analysis of variance (ANOVA) followed by multiple-comparisons tests (Scheffe’s tests). A P ≤ 0.05 was considered as significant. Under normal conditions (without hypoxic episodes applied), IL-10 (1 and 10 ng/ml) and TNF-␣ (0.8, 4 and 20 ng/ml) did not change the amplitude of PS and the values of TASG and NPS (data is not shown). Hypoxic episodes abolished the PSs (during episodes) (Fig. 1A) and induced a post-hypoxic sustained decrease in the TASG (data is not shown) and an increase in the NPS (Fig. 1B). Fig. 1B illustrates the effects of IL-10 (1 ng/ml) and TNF-␣ (0.8 ng/ml) on the time courses of NPS produced by repeated brief hypoxic episodes. These cytokines significantly depressed an increase in the NPS in a dose-dependent manner. Fig. 1C shows that IL-10 (1 ng/ml) more effectively depressed a hypoxia-induced increase in the NPS than IL-10 (10 ng/ml): 1 ng/ml—on 60% and 10 ng/ml—on 38.5% (140 min after hypoxia). In contrast to IL-10, the depressive effect of TNF␣ had a reversed dose–response relationship (Fig. 1C). TNF-␣ more effectively depressed a hypoxia-induced increase in the NPS in the concentration of 20 ng/ml than in concentrations of 4 and 0.8 ng/ml: 0.8 ng/ml—on 33.3%, 4 ng/ml—on 45% and 20 ng/ml—on 55% (140 min after hypoxia). However, from the above-mentioned results and the findings received in our previous work [12] it is not clear whether these cytokines depress the mechanisms of hypoxiainduced hyperexcitability (the production of multiple PSs) expressed after hypoxic episodes or depress the effectiveness of hypoxia on functional activity of hippocampal slice tissue during hypoxic episode (antihypoxic action) leading to an abolishment of post-hypoxic hyperexcitability. To examine this issue, we determine the effects of IL-10 and TNF-␣ on the time courses of PS amplitude depression during hypoxic episodes. Fig. 2A demonstrates the time courses of PS amplitude depression during: single hypoxic episode, single hypoxic episode + IL-10 (1 ng/ml) and single hypoxic episode + TNF-␣ (0.8 ng/ml). The effectiveness of hypoxic episodes (Th ) were: (1) single hypoxic episode − 293 ± 10 s; (2) single hypoxic episode + IL-10 (1 ng/ml) − 218 ± 11 s and (3) single hypoxic episode + TNF-␣ (0.8 ng/ml) − 189 ± 27 s. Thus, these results indicated that both cytokines significantly reduce the depressive effect of hypoxia on PS amplitude. This neuroprotective function of IL-10 and TNF-␣ against the hypoxia-induced depression of PS amplitude looks like preconditioning. Fig. 2B demonstrates that there are significant delays in the time points of a 50% depression of PS amplitude (Thd ) during the second and the third hypoxic episodes relatively to the Thd during the first episode (preconditioning by the first episode against the second and the third hypoxic episodes): Thd (1 episode) = 56 ± 4 s; Thd (2 episode) = 89 ± 5 s (P < 0.05); Thd (3 episode) = 100 ± 7 s (P < 0.05). However, an application of IL-10 or TNF-␣ prolonged the Thd during the first hypoxic episode (preconditioning by the cytokines against the first hypoxic episode) (Fig. 2C and D). The values of Thd during the first, second and third hypoxic episodes under the cytokine treatment were approximately the same: (1) IL-10: 1
238
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
Fig. 1. The effects of TNF-␣ and IL-10 on the development of the PS bursts in CA1 pyramidal neurons induced by hypoxic episodes. (A) Representative examples of typical PS responses before (0 min), during (Hypoxic episode), and 140 min (140 min) after 3 × 3 min episodes of hypoxia, hypoxia + TNF-␣ (0.8 ng/ml) and hypoxia + IL-10 (1 ng/ml). (B) Time courses of the NPS values before (−20 to 0) and after (+40 to +140) episodes (↓↓↓) of hypoxia (n = 10), hypoxia + IL-10 (1 ng/ml) (n = 10) and hypoxia + TNF-␣ (0.8 ng/ml) (n = 10). The symbol ‘*’ indicate the P < 0.05 significance level. (C) IL-10- and TNF-␣-induced dose-dependent depressions of a post-hypoxic increase in the NPS (100%) at 140 min after hypoxic episodes.
episode − 79 ± 4 s; 2 episode − 85 ± 3 s; 3 episode − 84 ± 4 s; (2) TNF-␣: 1 episode − 110 ± 16 s; 2 episode − 115 ± 11 s; 3 episode − 114 ± 14 s; (Fig. 2C and D). Some findings indicate that the preconditioning effects of IL-10 and TNF-␣ against transient hypoxia are very plausible because the intracellular signaling processes involved in ischaemic/hypoxic preconditioning and cellular response to IL-10 and TNF-␣ share the same steps, for example mitogen-
activated protein kinase (MAPK)/extracellular regulated kinase (ERK) [1,4,5,19]. There is a direct evidence that some cytokines can mediate ischaemic tolerance in hippocampal CA1 neurons [13]. In most studies, preconditioning required several hours to develop. However, some authors demonstrated that 1 min anoxic episode improved the recovery of PSs after a second anoxic episode separated by 10 min of reoxygenation in CA1 region of hippocampal slices [14]. These findings are in the
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
239
Fig. 2. Time courses of PS amplitude depression during hypoxic episodes (the solid lines on the plots). (A) Time courses of PS amplitude depressions during the first hypoxic episode (n = 10), the first hypoxic episode + TNF-␣ (0.8 ng/ml) (n = 10) and the first hypoxic episode + IL-10 (1 ng/ml) (n = 10). (B) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia. (C) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia + TNF-␣ (0.8 ng/ml). (D) Time courses of PS amplitude depressions during the first (1 episode, n = 10), the second (2 episode, n = 10) and the third (3 episode, n = 10) episodes of hypoxia + IL-10 (1 ng/ml).
line of our results. In our experiments, the TNF-␣- and IL-10induced neuroprotective effects against 3 min hypoxic episode were observed in 10 min after application of these cytokines to the hippocampal slice medium. Although brief cerebral ischaemia, or cerebral hypoxia, serve as prototypical preconditioning stimuli, many exogenously delivered agents (for example, inflammatory cytokines, metabolic inhibitors and anaesthetics) can also induce pre-
conditioning effect [5]. To induce neuroprotection, the preconditioning stimuli must activate transduction pathways that initiate the adaptive response. Although dependent in part on the nature of the preconditioning stimulus, members of these transduction pathways for which there is strong general support include mitogen-activated protein kinases (late preconditioning) and ATP-sensitive K+ channels (rapid preconditioning) [4,5,14].
240
M.E. Burkovetskaya et al. / Neuroscience Letters 416 (2007) 236–240
A surprising findings in our work was that pro-inflammatory cytokine TNF-␣ and anti-inflammatory cytokine IL-10 cause the similar neuroprotective effect on the hypoxia-induced hyperexcitability in CA1 hippocampal slice neurons. Recent data indicates that TNF-␣ can cause both detrimental and neuroprotective effects on neuronal activity [1,2,22]. These effects can be mediated by two distinct receptors, TNFR1 and TNFR2 [1]. Some results suggest that TNF-␣ can regulate neuronal circuit homeostasis inducing an increase and a decrease in surface expression of AMPA and GABAA receptors, respectively, via TNFR1 [17]. In addition, TNF-␣ can protect cultured neurons against glucose-deprived-injury and amino acid toxicity [3]. The function of anti-inflammatory cytokine IL-10 are mainly associated with promoting of cell survival in response to insults [19]. Thus, in certain circumstances, IL-10 and TNF-␣ can act in similar neuroprotective manner. Thus, our findings showed that neuroprotective effects of IL10 and TNF-␣ against the hyperexcitability in CA1 pyramidal neurons induced by brief episodes of hypoxia in the rat hippocampal slice are mediated by anti-hypoxic actions of these cytokines through, possibly, mechanism of preconditioning. Acknowledgement This work was supported by grants from the Russian Foundation of Basic Research (no. 05-04-48907). References [1] B.C. Albensy, Potential roles for tumor necrosis factor and nuclear factorkB in seizure activity, J. Neurosci. Res. 66 (2001) 151–154. [2] S. Balosso, T. Ravizza, C. Perego, J. Peschon, I.L. Campbell, M.G. De Simoni, A. Vezzani, Tumor necrosis factor-␣ inhibits seizures in mice via p75 receptors, Ann. Neurol. 57 (2005) 804–812. [3] B. Cheng, S. Christakos, M.P. Mattson, Tumor necrosis factors protect neurons against metabolic-excitotoxic insults and promote maintenance of calcium homeostasis, Neuron 12 (1994) 139–153. [4] V.L. Dawson, T.M. Dawson, Neuronal ischaemic preconditioning, TIPS 21 (2000) 423–424. [5] J.M. Gidday, Cerebral preconditioning and ischaemic tolerance, Nat. Neurosci. 7 (2006) 437–448. [6] O. Godukhin, A. Savin, S. Kalemenev, S. Levin, Neuronal hyperexcitability induced by repeated brief episodes of hypoxia in rat hippocampal slices: involvement of ionotropic glutamate receptors and L-type Ca2+ channels, Neuropharmacology 42 (2002) 459–466.
[7] K.R. Goldstein, R. Bhatt, B.E. Barton, S.S. Zalcman, P. Rameshwar, A. Siegel, Effects of hemispheric lateralization and site specificity on immune alterations induced by kindled temporal lobe seizures, Brain Behav. Immun. 16 (2002) 706–719. [8] J.L. Jankowsky, P.H. Patterson, The role of cytokines and growth factors in seizures and their sequelae, Prog. Neurobiol. 63 (2001) 125– 149. [9] F.E. Jensen, C. Wang, C.E. Stafstrom, Z. Liu, C. Geary, M.C. Stevens, Acute and chronic increases in excitability in rat hippocampal slices after perinatal hypoxia in vivo, J. Neurophysiol. 79 (1998) 73–81. [10] A. Kelly, A.A. Lynch, A.E. Vereker, Y. Nolan, P. Queenan, P.E. Whittaker, L.A.J. O’Neill, M.A. Lynch, The anti-inflammatory cytokine, interleukin10, blocks the inhibitory effect of IL-1 on long term potentiation, J. Biol. Chem. 276 (2001) 45564–45572. [11] S.G. Levin, O.V. Godukhin, Hyperexcitability of CA1 pyramidal neurons induced by repeated brief episodes of hypoxia in the hippocampal slices: involvement of GABAA and GABAB receptors, Russ. Physiol. J. 91 (2005) 581–586. [12] S.G. Levin, O.V. Godukhin, Protective effect of interleukin-10 on the development of epileptiform activity induced by brief hypoxic episodes in the rat hippocampal slices, Zh. Vyssh. Nerv. Deiat. 56 (2006) 379– 383. [13] T. Ohtsuki, C.A. Ruetzler, K. Tasaki, J.M. Hallenbeck, Interleukin-1 mediates induction of tolerance to global ischaemia in gerbil hippocampal CA1 neurons, J. Cereb. Blood Flow Metab. 16 (1996) 1137–1142. [14] M.A. Perez-Pinzon, J.M. Born, J.M. Centeno, Calcium and increase excitability promote tolerance against anoxia in hippocampal slices, Brain Res. 833 (1999) 20–26. [15] H. Schneider, F. Pitossi, D. Balschun, A. Wagner, A. Del Rey, A neuromodulatory role of interleukin-1 in the hippocampus, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 7778–7783. [16] A.A. Shandra, L.S. Godlevsky, R.S. Vastyanov, A.A. Oleinik, V.L. Konovalenko, E.N. Rapoport, N.N. Korobka, The role of TNF-␣ in amygdale kindled rats, Neurosci. Res. 42 (2002) 147–153. [17] D. Stellwagen, E.C. Beattie, J.Y. Seo, R.C. Malenka, Differential regulation of AMPA receptor and GABA receptor trafficking by tumor necrosis factor␣, J. Neurosci. 25 (12) (2005) 3219–3228. [18] R. Straussberg, J. Amir, L. Harel, I. Punsky, H. Bessler, Pro- and antiinflammatory cytokines in children with febrile convulsions, Pediatr. Neurol. 24 (2001) 49–53. [19] K. Strle, J.-H. Zhou, W.-H. Shen, S.R. Broussard, R.W. Johnson, G.G. Freund, R. Dantzer, K.W. Kelly, Interleukin-10 in the brain, Crit. Rev. Immunol. 21 (5) (2001) 427–449. [20] V. Tancredi, Long-lasting changes in synaptic excitability induced by anoxia in the rat hippocampus, Progr. Neuro-Pychopharmacol. 21 (1997) 211–232. [21] A. Vezzani, D. Moneta, C. Richichi, C. Perego, M.G. De Simoni, Functional role of pro-inflammatory and anti-inflammatory cytokines in seizures, Adv. Exp. Med. Biol. 548 (2004) 123–133. [22] C.X. Wang, A. Shuaib, Involvement of inflammatory cytokines in central nervous system injury, Prog. Neirobiol. 67 (2002) 161–172.