Neuroscience Letters 371 (2004) 1–5
Protective effects on neuronal cells of mouse afforded by ebselen against oxidative stress at multiple steps Takumi Satoh a,∗ , Kumiko Ishige b , Yutaka Sagara c a
c
Department of Welfare Engineering, Faculty of Engineering, Iwate University, Ueda 4-3-5, Morioka, Iwate 020-8551, Japan b Department of Pharmacology, College of Pharmacy, Nihon University, Narashino, Chiba 274-8555, Japan Department of Neurosciences, Faculty of Medicine, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093, USA Received 1 April 2004; received in revised form 22 April 2004; accepted 23 April 2004
Abstract Ebselen (2-phenyl-1,2-benzisoselenazol-3[2H]-one) mimics the activity of glutathione peroxidase [Biochem. Pharmacol. 33 (1984) 3235], acts as a substrate for thioredoxin reductase [Proc. Natl. Acad. Sci. U.S.A. 99 (2002) 8579]. The present study focused on the cellular mechanism of its action against oxidative stress by using HT22 cells, a mouse neuroblastoma of hippocampal origin. Ebselen protected HT22 cells against death induced by glutamate and hydrogen peroxide but not against that by tumor necrosis factor ␣. Oxidative glutamate toxicity is initiated by depletion of total glutathione, and ebselen inhibited the decrease in glutathione and increased its basal level. Although glutamate increased intracellular levels of reactive oxygen species (ROS), ebselen suppressed their increase. Ebselen reduced the basal levels of ROS when it was applied in control cells. Ebselen also removed ROS from cells that had accumulated a level of them. The compound had a significant trolox equivalent activity concentration value in a cell-free system, suggesting that it has a direct ROS-scavenging capacity. Finally, ebselen-induced heme oxygenase-1 (HO-1) protein. These results indicate that ebselen protects neuronal cells against the oxidative stress at multiple steps, including an increase in glutathione, a ROS-scavenging activity and the induction of HO-1 protein. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Ebselen; Reactive oxygen species; Glutathione; Heme oxygenase-1; HT22 cells
Oxidative stress, defined as the accumulation of reactive oxygen species (ROS) caused by enhancement of ROS production or by suppression of ROS destruction, plays a pivotal role in neurodegeneration associated with ischemia, trauma, and other neurodegenerative diseases [3]. Accumulation of ROS in neurons results in lipid peroxidation, protein oxidation, DNA damage, and finally cell death [3]. Thus, the inhibition of ROS accumulation is one of the potential targets of drug development for neuroprotection. Ebselen (2-phenyl-1,2-benzisoselenazol-3[2H]-one, Fig. 1A), originally developed as a mimetic of glutathione peroxidase [11], may be a promising candidate. In human and animal model
Abbreviations: DCF, 2 ,7 -dichlorodihydrofluorescein diacetate; ␥-GCL, ␥-glutamylcysteine ligase; HO-1, heme oxygenase-1; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; ROS, reactive oxygen species; TEAC, trolox equivalent activity concentration value; TNF␣, tumor necrosis factor ␣ ∗ Corresponding author. Tel.: +81-19-621-6039; fax: +81-91-621-6039. E-mail address:
[email protected] (T. Satoh).
studies, ebselen was shown to protect neurons against brain ischemia [12]. Other studies clarified that ebselen has cytoprotective effects in vitro models [19]. HT22 cells have been a useful model for studying the mechanism of oxidative glutamate toxicity [4]. Excitotoxicity is not involved in the death since glutamate receptor antagonists did not have any effects [4]. Rather, glutamate inhibits cysteine uptake, which subsequently leads to depletion of glutathione levels [4], increased production of ROS and elevated Ca2+ levels [13]. In the present study, we addressed the issue of what cellular mechanism is responsible for the neuroprotective effects of ebselen in cultured neuronal cells. We found that at least three anti-oxidative actions are involved in the neuroprotective effects afforded by ebselen: (1) an increase in the glutathione level, (2) a ROS-scavenging activity, and (3) the induction of HO-1 protein. Ebselen, tumor necrosis factor ␣ (TNF␣) (Calbiochem, San Diego, CA, USA), glutamate, actinomycin D (WAKO, Osaka, Japan), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT; Research Organics, Cleveland, OH, USA), anti-HO-1 polyclonal antibody (SPA895,
0304-3940/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2004.04.055
2
T. Satoh et al. / Neuroscience Letters 371 (2004) 1–5
Fig. 1. (A) Chemical structure of ebselen. (B–D) Inhibition of cell death by ebselen. Morphologies of the cells were obtained after 10 h incubation with 5 mM glutamate in the absence (C) or presence (D) of 5 M ebselen. (B) showed control cells.
Stressgen, Vict., Canada), peroxidase-conjugated anti-rabbit IgG antibody (Biorad, Hercules, CA, USA), anti-actin monoclonal antibody (Oncogene Research Products, San Diego, CA, USA), peroxidase-conjugated anti-mouse IgM antibody (Calbiochem, San Diego, CA, USA), and 2 ,7 -dichlorodihydrofluorescein diacetate (H2 DCF-DA; Molecular Probe, Eugene, OR, USA) were used in this study. Other chemicals were purchased from Sigma (St. Louis, MO, USA). HT22 cell culture and MTT assay were performed as described previously [14]. Western blotting was performed as described previously [13]. Total glutathione, ROS levels and Trolox Equivalent Activity Concentration (TEAC) were measured as described previously [13].
In the present study, we found that ebselen almost completely blocked the death of HT22 cells induced by 5 mM glutamate (Fig. 1B and D). 1–5 M ebselen protected HT22 cells against oxidative glutamate toxicity (Fig. 2A), but not against TNF␣-induced neuronal death (Fig. 2B). Ebeselen also prevented the death of HT22 cells induced by hydrogen peroxide (data not shown). Thus, the inhibition of cell death by ebselen seems to be specific to that induced by oxidative stress. Since selen and ebselen were reported to mimic glutathione peroxidase [7,11], ebselen may attenuate the metabolism of glutathione. To examine the possibility, total glutathione (reduced and oxidized) was measured (Fig. 3).
Fig. 2. Neuroprotection by ebselen. (A) The cells were exposed to various concentrations of ebselen for 30 min; vehicle (squares), 5 mM glutamate (circles) or 10 mM glutamate (triangles) was added; the cells were incubated for a further 24 h. Cell viability was then assessed by the MTT assay. (B) After cells had been exposed to ebselen as in (A), vehicle (squares), 0.01 ng/ml TNF␣ (circles), 0.1 ng/ml TNF␣ (triangles) or 1.0 ng/ml TNF␣ (diamonds) was added simultaneously with actinomycin D (0.1 g/ml); the cells were then incubated for 24 h. The values, which represent the percentage of the control MTT activity, are means ± S.D. (n = 4). Significance of difference between no ebselen and ebselen at any concentrations was determined by ANOVA (∗ P < 0.01).
T. Satoh et al. / Neuroscience Letters 371 (2004) 1–5
Fig. 3. Effects of ebselen on cellular levels of glutathione. HT-22 cells were treated with 5 M ebselen or/and 5 mM glutamate for the indicated times, and the cellular levels of total glutathione were measured. Ebselen and glutamate were added simultaneously. Squares, ebselen alone; diamonds, ebselen + glutamate; triangles, glutamate. In this experiment, the glutathione level of the control sample (36.5 ± 2 nmol/mg protein) was set at 100%. Similar results were obtained from at least three independent experiments.
Consistent with previous reports, total glutathione was decreased within 5–7 h after the addition of glutamate (triangles in figure). Ebselen completely inhibited the decrease in total glutathione caused by glutamate (diamonds in figure). Ebselen also increased the basal levels 2.0- to 3.0-fold (squares in figure). These results indicate that ebselen could increase the basal levels of glutathione and cancel the depletion induced by glutamate. To investigate the effect of ebselen on intracellular ROS levels, we performed FACS analysis by use of a ROS-sensitive fluorescence indicator, H2 DCF-DA [13]. The incubation of HT22 cells with glutamate for 12 h increased the intracellular ROS to several 10-fold (black and brown lines in Fig. 4A). However, the presence of ebselen (Ebselen + 5 mM Glu) significantly reduced the glutamate-elevated ROS levels to below the basal level (purple line in Fig. 4A). The presence of ebselen (Ebselen) alone reduced the ROS level to an even lower level (green line in
3
Fig. 4A). These results suggest that ebselen reduced basal level of ROS and inhibited the ROS burst during the oxidative glutamate toxicity However, it was unclear whether ebselen removed ROS directly by itself or by an indirect action. To answer this question, we added ebselen at 11 h after glutamate and incubated the cells further for 1 h. In this case, glutathione was completely depleted, ROS had fully accumulated, and all of the cells were committed to death when ebselen was added. Thus, a direct ROS-scavenging activity could be detected by this experiment. The delayed addition of ebselen at 11 h after glutamate (11 h Glu + 1 h Ebselen in Fig. 4B) reduced to a significant extent the ROS levels in comparison with that levels in cells incubated with glutamate alone (5 mM Glu in Fig. 4B). Thus, ebselsen could remove ROS from cells that had fully accumulated ROS elicited by glutamate, suggesting that ebselen has a ROS-scavenging capacity. To obtain a direct evidence that ebselen has a ROS-scavenging capacity in a cell-free system, we performed a TEAC assay. In this assay, ebselen was compared with Trolox, a water-soluble Vitamin E analog, in terms of its ability to suppress the radical cation of 2,2 -azinobis(3-ethylbenzothiazoline 6-sulfonate) in an aqueous solution. Ebselen had a significant TEAC value (0.32 ± 0.06). In this experiment, cycloheximide and apomorphin as negative and positive controls [8] were 0.00 ± 0.04 and 1.71 ± 0.17, respectively. These results thus indicate that ebselen can directly scavenge ROS. Recently, HO-1 was identified as a previously unrecognized regulator of intracellular ROS and neuronal resistance to oxidative stress [2,10,18]. Thus, the induction of HO-1 protein was examined (Fig. 5). Ebselen significantly induced HO-1 protein in HT22 cells at similar concentrations similar to those protected HT22 cells (lanes 1–4). The pattern was not affected by the presence of glutamate (lanes 5–8). In the present study, we proposed that ebselen has at lease three anti-oxidative activities: (1) activity of increase in glutathione levels, (2) a ROS-scavenging capacity, and
Fig. 4. Effects of ebselen on ROS levels. In (A) HT-22 cells were untreated (Control) or treated either with 5 M eblselen alone (Ebselen) or with 5 mM glutamate for 12 h in the absence (Glu) or presence of 5 M ebselen (Ebselen + 5 mM Glu). In (B), HT-22 cells were incubated for 11 h and then exposed to 5 M eblselen for 1 h (1 h Ebselsen), were incubated with 5 mM glutamate for 12 h (5 mM Glu), or were treated with 5 mM glutamate for 11 h and then 5 M ebselen was added for 1 h (11 h Glu + 1 h Ebselen). Similar results were obtained from at least three independent experiments.
4
T. Satoh et al. / Neuroscience Letters 371 (2004) 1–5
Fig. 5. Induction of HO-1 by various concentrations of ebselen in the absence or the presence of 5 mM of glutamate. HT-22 cells were incubated with vehicle (lanes 1 and 5), ebselen 1 M (lanes 2 and 6), ebselen 2 M (lanes 3 and 7), ebselen 5 M (lanes 4 and 8) for 24 h in the absence (lanes 1–4) or presence (lanes 5–8) of 5 mM glutamate, and then the lysates were probed with anti-HO-1 (upper panel) or anti--actin (lower panel) antibody.
(3) induction of HO-1 protein. Ebselen protected HT22 cells against oxidative stress but not against TNF␣-induced cell death (Figs. 1 and 2). Ebselen increased the basal level of total glutathione and inhibited the decrease in it by glutamate (Fig. 3). It reduced the basal level of intracellular ROS and inhibited the increase by glutamate (Fig. 4A) as well as removed ROS from cells that had accumulated them (Fig. 4B). Also, ebselen possessed a direct ROS-scavenging capacity as indicated by a significant TEAC value. Finally, ebselen induced HO-1 protein in HT22 cells at concentrations similar to those protected the cells (Fig. 5). Ebselen increased the level of total glutathione (reduced and oxidized) in HT22 cells (Fig. 3). Since the oxidative glutamate toxicity is initiated by depletion of total glutathione [4], the inhibition of depletion of total glutathione should provide neuroprotection. This effect of ebselen may be due to (1) mimicking of glutathione by ebselen (reducing the consumption of glutathione), (2) induction of ␥-glutamylcysteine ligase (␥-GCL: increasing the synthesis) or (3) increasing in glutathione reductase activity (increasing the reduced type). Since ebselen is known to mimic glutathione peroxidase [11], to induce ␥-GCL protein, and to activate glutathione reductase in cardiac myocytes [7], it could increase the total glutathione levels. Sagara et al. [13] reported that some of the tyrophostins increased glutathione levels by inducing ␥-GCL protein and had a neuroprotective effect on HT22 cells. Since the 5 -flanking region of the ␥-GCL gene possesses an anti-oxidant responsive element [6], ebselen as well as tyrophostins may induce the synthase through activation of this element. Although we did not measure the relative contribution of each mechanism, each of them may play a cooperative role in increasing the level of total glutathione. However, we do not consider that an increase in glutathione alone is sufficient for explaining the neuroprotection by ebselen and propose that its direct ROS-scavenging activities should play a role in the neuroprotection for the following reasons: (1) ebselen removed ROS from cells that had fully accumulated them; (2) ebselen had a significant TEAC value. Ebselen should provide neuroprotective effects on HT22 cells by inhibiting both of these phases by increas-
ing the glutathione level and expressing ROS-scavenging activity. Ebselen induced HO-1 protein in HT22 cells (Fig. 5). Since expression of HO-1 protein is regulated by ARE [1], the upregulation of ␥-GCL and HO-1 may be the same mechanism. HO cleaves heme molecules at the ␣-meso carbon bridge and produces the open tetrapyrrole biliverdin, CO and Fe2+ [9]. Especially biliverdin and bilirubin (a product from biliverdin due to biliverdin reductase action) were reported to protect CNS neurons [5] and HT22 cells [18] against oxidative stress possibly through an anti-oxidative pathway. HO-1 protein induced by low-molecular weight compounds might play an important role in the neuroprotection of CNS neurons [17]. Previously, we found that NEPP11, a neurotrophin-like compound, protected HT22 cells against oxidative glutamate toxicity and potently induced HO-1 protein [15,16,18]. Transfer of the HO-1 gene to HT22 cells protected the cells against oxidative glutamate toxicity [18]. Neuroprotective effects of HO-1 protein were also reported by other groups. For example, cerebellar granule neurons from mice overexpressing HO-1 resisted glutamate-toxicity by decreasing the levels of ROS [2]. These results suggest that the HO-1 protein protected neurons against oxidative stress. In conclusion, we propose the increase in glutathione, ROS-scavenging activity and HO-1 induction should be the main events responsible for the neuroprotective effect of ebselen.
References [1] J. Alam, J.L. Cook, Transcriptional regulation of the heme oxygenase-1 gene via the stress response element pathway, Curr. Pharm. Des. 9 (2003) 2499–2511. [2] K. Chen, K. Gunter, M.D. Maines, Neurons overexpressing heme oxygenase-1 resist oxidative stress-mediated cell death, J. Neurochem. 75 (2000) 304–313. [3] J.T. Coyle, P. Puttfarcken, Oxidative stress, Science 262 (1993) 689– 695. [4] J.B. Davis, P. Maher, Protein kinase C activation inhibits glutamate-induced cytotoxicity in a neuronal cell line, Brain Res. 652 (1994) 169–173. [5] S. Dore, M. Takahashi, C.D. Ferris, L.D. Hester, D. Guastella, S.H. Snyder, Birilubin, formed by activation of heme oxygenase-2, protects neurons against oxidative stress injury, Proc. Natl. Acad. Sci. U.S.A. 96 (1999) 2445–3450. [6] S. Duffy, A. So, T.H. Murphy, Activation of endogenous antioxidant defenses in neuronal cells prevents free radical-mediated damage, J. Neurochem. 71 (1998) 69–77. [7] S. Hoshida, K. Aoki, M. Nishida, N. Yamashita, J. Igarashi, M. Hori, T. Kuzuya, M. Tada, Effects of preconditioning with ebselen on glutathione metabolism and stress protein expression, J. Pharmacol. Exp. Ther. 281 (1997) 1471–1475. [8] K. Ishige, D. Schubert, Y. Sagara, Flavonoids protect neuronal cells from oxidative stress by three distinct mechanisms, Free Radic. Biol. Med. 30 (2001) 433–446. [9] M.D. Maines, The heme oxygenase system: a regulation of second messenger gases, Ann. Rev. Pharmacol. Toxicol. 37 (1997) 517–554. [10] M.D. Maines, N. Panahian, The hemeoxyenase system and cellular defense mechanisms: do HO-1 and HO-2 have different functions?
T. Satoh et al. / Neuroscience Letters 371 (2004) 1–5
[11]
[12]
[13] [14]
[15]
Hypoxia, in: R.C. Roach, et al. (Eds.), From Genes to the Bedsides, Kluwer Academic/Plenum Publishers, New York, 2001, pp. 249–272. A.E. Muller, E. Cadenas, P. Graf, H. Sies, A novel biologically active seleno-organic compound. I. Glutathione peroxidase-like activity in-vitro and antioxidant capacity of PZ 51 (ebselen), Biochem. Pharmacol. 33 (1984) 3235–3239. A. Ogawa, T. Yoshimoto, H. Kikuchi, K. Sano, I. Saito, T. Yamaguchi, H. Yasuhara, Ebselen in acute middle cerebral artery occlusion: a placebo-controlled, double-blind clinical trial, Cerebrovasc. Dis. 9 (1999) 112–118. Y. Sagara, K. Ishige, C. Tsai, P. Maher, Tyrphostins protect neuronal cells from oxidative stress, J. Biol. Chem. 277 (2002) 36204–36215. T. Satoh, D. Nakatsuka, Y. Watanababe, N. Nagata, H. Kikuchi, S. Namura, Neuroprotection by MEK/ERK kinase inhibition against oxidative stress in a mouse neuronal cell line and rat primary cultured neurons, Neurosci. Lett. 288 (2000) 163–166. T. Satoh, K. Furuta, K. Tomokiyo, D. Nakatsuka, M. Miura, H. Hatanaka, K. Ikuta, M. Suzuki, Y. Watanabe, Facilitatory roles of novel compounds designed from cyclopentenone prostaglandins on the neurite outgrowth-promoting activities of NGF, J. Neurochem. 75 (2000) 1092–1102.
5
[16] T. Satoh, K. Furuta, K. Tomokiyo, S. Namura, D. Nakatsuka, Y. Sugie, Y. Ishikawa, H. Hatanaka, M. Suzuki, Y. Watanabe, Neurotrophic actions of novel compounds designed from cyclopentenone prostaglandins, J. Neurochem. 77 (2001) 50–62. [17] T. Satoh, K. Furuta, M. Suzuki, Y. Watanabe, Neurite outgrowthpromoting prostaglandins that act as neuroprotective agents against brain ischemia and may enhance recovery of higher neuronal functions, in: H. Kikuchi (Ed.), Strategenic Medical Science Against Brain Attack, Springer-Verlag, Tokyo, pp. 78–93. [18] T. Satoh, M. Baba, D. Nakatsuka, Y. Ishikawa, H. Aburatani, K. Furuta, T. Ishikawa, H. Hatanaka, M. Suzuki, Y. Watanabe, Role of heme oxygenase-1 protein in the neuroprotective effects by cyclopentenone prostaglandin derivatives as a sustained phase of neuronal survival promoting mechanism under oxidative stress, Eur. J. Neurosci. 17 (2003) 2249–2255. [19] M. Yoshizumi, T. Kogame, Y. Suzaki, Y. Fujita, M. Kyaw, K. Kirima, K. Ishizawa, K. Tsuchiya, S. Kagami, T. Tamaki, Ebselen attenuates oxidative stress-induced apoptosis via the inhibition of the c-Jun N-terminal kinase and activator protein-1 signalling pathway in PC12 cells, Br. J. Pharmacol. 136 (2002) 1023– 1032.