Hydrogen peroxide is selectively toxic to immature murine neurons in vitro

Neuroscience Letters 231 (1997) 17–20

Hydrogen peroxide is selectively toxic to immature murine neurons in vitro Rebecca E. Mischel a ,*, Young S. Kim b, R. Ann Sheldon c, Donna M. Ferriero a , c a

Department of Pediatrics, University of California San Francisco, San Francisco, CA 94143, USA Department of Medicine, University of California San Francisco, San Francisco, CA 94143, USA c Department of Neurology, University of California San Francisco, San Francisco, CA 94143, USA b

Received 1 April 1997; received in revised form 3 July 1997; accepted 7 July 1997

Abstract Hydrogen peroxide (H2O2) accumulates during hypoxic-ischemic brain injury and may mediate neurotoxicity in the immature brain. To determine whether H2O2 causes maturation specific neurotoxicity, primary neuronal cultures were exposed to H2O2 (25, 50, 100 mM) for 5 min or 24 h during in vitro development, and toxicity was assessed. Immature neurons incurred marked and dose dependent injury after both brief and prolonged H2O2 exposures, and marked dose dependent death following prolonged H2O2 exposures. Mature neurons incurred marked injury following prolonged but not brief H2O2 exposures, and were relatively resistant to H2O2 induced death following both brief and prolonged exposures. Thus, H2O2 is selectively toxic to immature neurons in vitro. Neuronal vulnerability to H2O2 during in vivo development is unknown and warrants investigation.  1997 Elsevier Science Ireland Ltd. Keywords: Neonatal; Hypoxia-ischemia; Neurotoxicity; Development; Superoxide dismutase; Catalase; Glutathione peroxidase

Excitotoxins and free radicals are well established mediators of hypoxic-ischemic injury in both the immature and mature brain [16]. Hypoxic-ischemic induced excitotoxicity leads to increased production of superoxide, a free radical which is converted by CuZn-superoxide dismutase (SOD-1) to hydrogen peroxide (H2O2). H2O2 is a reactive oxygen metabolite which mediates toxicity directly by attacking cellular constituents and indirectly by generating other free radicals and amplifying excitotoxicity [11]. H2O2 is reduced by glutathione peroxidase (GSH-Px) or catalase to water. This enzymatic pathway for free radical scavenging constitutes the cell’s most important antioxidant defense system during hypoxia-ischemia. Several recent studies have provided evidence that H2O2 may be a critical mediator of neurotoxicity in the setting of hypoxia-ischemia. For example, Down syndrome neurons overexpress SOD-1, generate increased levels of H2O2, and undergo premature death [1]. These neurons are rescued from premature death in vitro by catalase supplementation

* Corresponding author. University of California San Francisco, Neonatal Brain Disorders Laboratory, Box 0114, San Francisco, CA 94143-0114, USA. Tel.: +1 415 5025820; fax: +1 415 5025821.

[1], suggesting that excessive H2O2 production may be responsible for the neurotoxicity. In addition, although immature neurons in mixed cell cultures are relatively resistant to glutamate mediated excitotoxicity, they are vulnerable to H2O2 toxicity [19]. This suggests that H2O2 may have a uniquely prominent role in mediating neurotoxicity in the developing brain. To determine whether H2O2 causes maturation specific neurotoxicity, primary mouse cortical neuronal cultures were exposed to H2O2 (25, 50, and 100 mM) for either brief (5 min) or prolonged (24 h) durations during in vitro development, and neurotoxicity was assessed. Primary neuronal cortical cultures were prepared using minor modifications of the method employed by Choi et. al. [6], and maintained for prolonged periods with astrocyte conditioned media (ACM). Astrocyte cultures were derived from newborn CD1 mice according to the method of Yu [20], and ACM was prepared by exposing enriched media to confluent astrocyte cultures for 24–48 h. This process depletes glutamate and supplements the media with astrocyte-derived growth factors important for neuronal survival and development [15]. For neuronal cultures, neocortices were aseptically dissected from 16 day old fetal mice with hippocampi, basal ganglia, and meninges carefully

0304-3940/97/$17.00  1997 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3940 (97 )0 0531-4

18

R.E. Mischel et al. / Neuroscience Letters 231 (1997) 17–20

removed, and placed in Hank’s buffered salt solution. Cells were mechanically and enzymatically dissociated with trypsin, centrifuged, resuspended in ACM, and plated on polylysine (0.05 mg/ml) coated 96-well culture dishes. Cells were incubated at 5% CO2, 95% humidity at 37°C. After 16 h in vitro, cultures were exposed to 10 mM cytosine arabinoside (Ara-C) for 24 h to arrest cell division. Fresh ACM was added to neuronal cultures after Ara-C exposure and biweekly thereafter. Purity of neuronal cultures was assessed by immunocytochemistry using the Vectastain avidin-biotin-peroxidase system. Neurons were identified by neuron-specific enolase (1:500) immunoreactivity and astrocytes were identified by glial fibrillary acidic protein

Fig. 1. Phase contrast photomicrograph of primary mouse cortical neuronal cultures. (A) Neuron-specific enolase immunocytochemistry identifies neurons in pure neuronal culture at 21 DIV. (B) Glial fibrillary acidic protein immunocytochemistry identifies astrocytes in pure neuronal culture at 21 DIV (,5% astrocyte contamination). (C) Immature neurons in culture at 6 DIV lack nNOS expression and are not stained by NADPH-d histochemistry. (D) Mature neurons in culture at 20 DIV express nNOS and are stained by NADPH-d histochemistry. Scale bar, 100 mm.

(1:3000) immunoreactivity [7]. Neuronal nitric oxide synthase (nNOS) expression, an indicator of neuronal maturity, was assessed by nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) histochemistry [14]. H2O2 induced neurotoxicity was assessed 24 h after H2O2 exposure was initiated. Neuronal injury was measured by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, which reflects mitochondrial enzyme function. Neuronal death was assessed under phase contrast microscopy by Trypan blue (TB) dye exclusion and neuronal morphology. For the MTT assay, neuronal cultures were plated at 250 000 cells/ml and MTT was added to the cultures at a final concentration of 5 mg/ml. Optical density was read at 540 nm, and the percent of H2O2 induced neuronal injury was calculated relative to control and completely lysed wells [17]. For TB exclusion, neuronal cultures were plated at 100 000 cells/ml, incubated with TB at 0.4%, and washed in ACM [10]. For the TB test, 500 cells/well were counted and the reported percent neuronal death in experimental conditions was corrected for death in control wells. Statistical significance was tested by analysis of variance (ANOVA) for repeated measures or factorial ANOVA with Fisher post-hoc analysis. Statistical significance was set at P , 0.05. Primary neuronal cultures with .95% neuronal purity were maintained for up to 21 days in vitro (DIV) (Fig. 1A,B), and developmental acquisition of nNOS expression was documented (Fig. 1C,D). Sister cultures from five separate neuronal dissections were exposed to H2O2 at 25, 50, and 100 mM for either 5 min or 24 h at either 6 or 20 DIV. All experimental conditions were performed in triplicate wells. Neuronal injury in immature neurons (6 DIV) as assessed by the MTT assay was marked at all H2O2 concentrations tested for both 5 min and 24 h exposures (Fig. 2A), and was dose dependent for both brief and prolonged exposures (P , 0.05 by ANOVA for repeated measures). Neuronal death as assessed by TB dye exclusion and neuronal morphology was low in immature neurons following brief H2O2 exposures, but more marked following prolonged H2O2 exposures (Fig. 3A). In immature neurons, there was a trend toward increased neuronal death with increased H2O2 dose following brief exposures and dose dependent death following prolonged H2O2 exposures (P , 0.05 by ANOVA for repeated measures). Neuronal injury in mature neurons as assessed by the MTT assay was relatively low following brief H2O2 exposures and was not dose dependent. Although the neuronal injury observed in mature neurons following prolonged exposure to 50 mM H2O2 was not statistically increased from that seen following 25 mM H2O2, there was significantly increased neuronal injury at 100 mM H2O2 (P , 0.05 by ANOVA for repeated measures) (Fig. 2B). Neuronal death as assessed by TB dye exclusion and neuronal morphology was low in mature neurons following both brief and prolonged H2O2 exposures (Fig. 3B).

R.E. Mischel et al. / Neuroscience Letters 231 (1997) 17–20

19

Fig. 2. H2O2 induced neuronal injury as assessed by the MTT assay. (A) Neuronal injury in immature neurons at 6 DIV following H2O2 exposure. (B) Neuronal injury in mature neurons at 20 DIV following H2O2 exposure. Hatched bar represents neuronal injury following 5 min H2O2 exposure. Solid bar represents neuronal injury following 24 h H2O2 exposure. Data is presented as the mean ± SEM (n = 5). + P , 0.05, *P , 0.05.

Although death in mature neurons tended to increase with increased H2O2 dose, the increases did not reach statistical significance. Neuronal injury following brief H2O2 exposures was increased in immature neurons relative to that seen in mature neurons at all doses tested (P , 0.05 by factorial ANOVA) (Fig. 2). Following prolonged H2O2 exposures, neuronal injury was increased in immature neurons relative to that seen in mature neurons at 25 and 50 mM H2O2 (P , 0.05 by factorial ANOVA), but at 100 mM H2O2 there was a similar high degree of injury in both immature and mature neurons (Fig. 2). Following brief H2O2 exposures, immature neurons tended to sustain greater neuronal death than did mature neurons, but these increases were not statistically significant (Fig. 3). Following prolonged H2O2 exposures, there was increased neuronal death in immature neurons relative to that seen in mature neurons at each of the doses tested (P , 0.05 by factorial ANOVA) (Fig. 3). Neuronal death was much less extensive than neuronal injury in each of the experimental paradigms except in immature

neurons following prolonged H2O2 exposure when both death and injury were extremely high (≥94%). Thus, H2O2 is selectively toxic to immature cortical neurons in culture. Immature neurons in culture lack expression of N-methyl-d-aspartate (NMDA) receptors and nNOS, both established mediators of hypoxic-ischemic injury, whereas mature neurons have full expression [7]. Thus, H2O2 may exert its neurotoxic effects through mechanisms independent of NMDA receptor activation or nitric oxide production. It has recently been demonstrated that the ratio of SOD-1 to GSH-Px activity is high in the developing brain whereas these enzymes have equivalent activity in the mature brain [2,3,8,12,18]. Thus, there may be a developmentally regulated imbalance between H2O2 generating and H2O2 detoxifying enzymes in the neonatal brain which results in the accumulation of H2O2 in the neurons during hypoxia-ischemia. The role of H2O2 mediated neurotoxicity in the developing nervous system is further substantiated by the finding that SOD-1 overexpression (3-fold increase) protects adult

Fig. 3. H2O2 induced neuronal death as assessed by TB dye exclusion and neuronal morphology. (A) Neuronal death in immature neurons at 6 DIV following H2O2 exposure. (B) Neuronal death in mature neurons at 20 DIV following H2O2 exposure. Hatched bar represents neuronal death following 5 min H2O2 exposure. Solid bar represents neuronal death following 24 h H2O2 exposure. Data is presented as the mean ± SEM (n = 5). *P , 0.05.

20

R.E. Mischel et al. / Neuroscience Letters 231 (1997) 17–20

mice from hypoxic-ischemic brain injury but renders newborn mice more vulnerable to damage [9,13]. In SOD-1 transgenic mice, increased H2O2 production during hypoxia-ischemia is not matched by increased activity of H2O2 detoxifying enzymes in the immature brain [4,5]. H2O2 has recently been found to accumulate in the neonatal transgenic mouse brain following hypoxia-ischemia, and appears to be responsible for the increased brain injury [5]. Thus, the observed increased vulnerability of immature neurons to H2O2 toxicity in vitro may be due to a developmentally regulated inadequacy of H2O2 detoxifying enzymes, and the relative resistance of mature neurons to H2O2 toxicity may be related to the mature phenotypic expression of antioxidant enzymes. Although primary neuronal cultures are simplified models of cerebral hypoxia-ischemia, they enable investigation of specific processes involved in the complex, multifactorial pathogenesis of this injury. Furthermore, neurons developing in vivo and in vitro share important characteristics including developmentally regulated vulnerability to excitotoxicity and maturation-dependent expression of antioxidant enzymes. The similarities between neurons developing in vivo and in vitro suggest that studies of H2O2 induced neuronal injury in culture may provide insights into the unique pathogenesis of perinatal hypoxic-ischemic brain injury. Definition of the specific mediators of perinatal hypoxic-ischemic brain injury may ultimately enable the development of effective clinical therapies in the management of this common and often devastating condition.

[5]

[6] [7]

[8]

[9]

[10]

[11] [12]

[13]

[14]

[15]

This work was supported in part by a grant to R.E.M. (the W.H. Tooley Memorial Fund), and grants to D.M.F. (R01NS33997, P20NS32553). [1] Busciglio, J. and Yankner, B.A., Apoptosis and increased generation of reactive oxygen species in Down’s syndrome neurons in vitro, Nature, 378 (1995) 776–779. [2] Buard, A., Clement, M. and Bourre, J.M., Developmental changes in enzymatic systems involved in protection against peroxidation in isolated rat brain microvessels, Neurosci. Lett., 141 (1992) 72– 74. [3] Ceballos-Picot, I., Nicole, A., Clement, M., Bourre, J.M. and Sinet, P.M., Age-related changes in antioxidant enzymes and lipid peroxidation in brains of control and transgenic mice overexpressing copper-zinc superoxide dismutase, Mutat. Res., 275 (1992) 281– 293. [4] Chetkovich, H.F., Epstein, C.J. and Ferriero, D.M., Catalase activity

[16]

[17]

[18]

[19]

[20]

after perinatal hypoxia-ischemia in Cu/Zn SOD transgenic mice, Soc. Neurosci. Abstr., 22 (1996) 1430A. Chetkovich, H.J.F., Ditelberg, J.S., Chen, S., Chan, P., Epstein, C.J. and Ferriero, D.M., Glutathione peroxidase activity decreases and hydrogen peroxide accumulates after perinatal hypoxia-ischemia in SOD 1 overexpressing mice, Pediatr. Res., (1997) in press. Choi, D.W., Maulucci-Gedde, M. and Kriegstein, A.F., Glutamate neurotoxicity in cortical cell culture, J. Neurosci., 7 (1987) 357–368. Dawson, V.L., Dawson, T.M., Bartley, D.A., Uhl, G.R. and Snyder, S.H., Mechanisms of nitric oxide-mediated neurotoxicity in primary brain cultures, J. Neurosci., 13 (1993) 2651–2661. DeHaan, J.B., Tymms, M.J., Cristiano, F. and Kola, I., Expression of copper/zinc superoxide dismutase and glutathione peroxidase in organs of developing mouse embryos, fetuses, and neonates, Pediatr. Res., 35 (1994) 188–196. Ditelberg, J.S., Sheldon, R.A., Epstein, C.J. and Ferrriero, D.M., Brain injury after perinatal hypoxia-ischemia is exacerbated in copper/zinc superoxide dismutase transgenic mice, Pediatr. Res., 39 (1996) 204–208. Ferriero, D.M., Sheldon, R.A., Black, S.M. and Chuai, J., Selective destruction of nitric oxide synthase neurons with quisqualate reduces damage after hypoxia-ischemia in the neonatal rat, Pediatr. Res., 38 (1995) 912–918. Halliwell, B. and Gutteridge, J.M.C., Free Radicals in Biology and Medicine, Clarendon Press, Oxford, 1989, pp. 79–81, 86. Houdou, S., Kuruta, H., Hasegawa, M., Konomi, H., Takashima, S., Suzuki, Y. and Hashimoto, T., Developmental immunohistochemistry of catalase in the human brain, Brain Res., 556 (1991) 267–270. Kinouchi, H., Epstein, C.J., Mizui, T., Carlson, E., Chen, S.F. and Chan, P.H., Attenuation of focal cerebral ischemic injury in transgenic mice overexpressing CuZn superoxide dismutase, Proc. Natl. Acad. Sci. USA, 88 (1991) 11158–11162. Koh, J. and Choi, D.W., Vulnerability of cultured cortical neurons to damage by excitotoxins: differential susceptibility of neurons containing NADPH-diaphorase, J. Neurosci., 8 (1988) 2153–2163. Koops, A., Kappler, J., Junghans, U., Kuhn, G., Kresse, H. and Muller, H.W., Cultured astrocytes express biglycan, a chondroitin/ dermatan sulfate proteoglycan supporting the survival of neocortical neurons, Brain Res. Mol. Brain Res., 41 (1996) 65–73. Palmer, C., Hypoxic-ischemic encephalopathy: therapeutic approaches against microvascular injury, and role of neutrophils, PAF, and free radicals, Clin. Perinatol., 22 (1995) 481–517. Piani, D., Spranger, M., Frei, K., Schaffner, A. and Fontana, A., Macrophage-induced cytotoxicity of NMDA receptor positive neurons involves EAA rather than reactive oxygen intermediates and cytokines, Eur. J. Immunol., 22 (1992) 2429–2436. Takashima, S., Kuruta, H., Mito, T., Houdou, S., Konomi, H., Yao, R. and Onodera, K., Immunohistochemistry of superoxide dismutase-1 in developing human brain, Brain Dev., 12 (1990) 211–213. Whittemore, E.R., Loo, D.T., Watt, J.A. and Cotman, C.W., A detailed analysis of hydrogen peroxide-induced cell death in primary neuronal culture, Neuroscience, 67 (1995) 921–932. Yu, B.P., Cellular defenses against damage from reactive oxygen species, Physiol. Rev., 74 (1995) 139–162.