Reduction of Ischemic Damage in NGF-Transgenic Mice: Correlation with Enhancement of Antioxidant Enzyme Activities

Neurobiology of Disease 6, 180–189 (1999) Article ID nbdi.1999.0240, available online at http://www.idealibrary.com on

Reduction of Ischemic Damage in NGF-Transgenic Mice: Correlation with Enhancement of Antioxidant Enzyme Activities Christelle Gue´gan,* Ire`ne Ceballos-Picot,† Elisabeth Chevalier,‡ Annie Nicole,† Brigitte Onte´niente,‡ and Brigitte Sola§ *Laboratoire de Neurosciences, Universite´ de Caen, CNRS UMR 6551, 14074 Caen, France; †CNRS URA 1335, Ho ˆ pital Necker, 75743 Paris 15, France; ‡INSERM U421-IM3, UFR de Me´decine, 94010 Cre´teil, France; and §Faculte´ de Me´decine, Universite´ de Caen, UPRES-EA 2128, 14032 Caen, France Received September 23, 1998; accepted for publication February 16, 1999

If permanent focal ischemia is induced by middle cerebral artery occlusion (MCAO), neurons within the infarcted territory die by necrosis and apoptosis (or programmed cell death). We have previously shown, using a mouse strain transgenic (tg) for the nerve growth factor (NGF) gene, that tg mice have consistently smaller infarcted areas than wild-type (wt) animals, correlated with upregulated NGF synthesis and impaired apoptotic cell death. We studied, in wt and tg mice subjected to MCAO, the activities of several antioxidant enzymes and the synthesis of the proteins of the Bcl-2 family. Our results show that the antiapoptotic Bcl-2 protein and glutathione peroxidase are recruited after MCAO. NGF-tg mice also had an intrinsic resistance to oxidative stress because their basal copper zinc superoxide dismutase (SOD) and glutathione transferase activities were high. Additionally, manganese SOD activity increased in NGF-tg mice after MCAO, correlating strongly with the resistance of these mice to apoptosis. r 1999 Academic Press Key Words: superoxide dismutase; glutathione metabolism; Bcl-2 family; nerve growth factor; transgenic mice; cerebral ischemia; apoptosis.

occlusion (MCAO), it was shown that neurons within the infarcted area die by necrosis and apoptosis (Gue´gan et al., 1996). We have also previously reported that mice transgenic (tg) for NGF had consistently smaller (by 40%) areas of infarcted cortical territory than control mice 24 h after occlusion. NGF protein is upregulated in cortical areas after ischemic damage, and its synthesis was even higher in tg mice and was correlated with the observed neuroprotection. The decrease in cortical infarction was accompanied by a dramatic decrease in the number of apoptotic cells underlying the cross-talk between NGF and apoptosis in vivo (Gue´gan et al., 1998a). We further investigated the neuronal protection mediated by NGF, by analyzing the levels of proteins of the Bcl-2 family after MCAO, by immunohistochemistry and Western blotting. NGF exerted its neuroprotective effects, preventing apoptotic death in vitro, by

INTRODUCTION Nerve growth factor (NGF) promotes neuron survival during embryonic development and under various physiological and pathological conditions, including Alzheimer’s disease and hypoglycemic, excitotoxic, and ischemic lesions (Lindvall et al., 1994; Gao et al., 1997). Indeed, the neuroprotective effect in vivo of NGF after intracerebroventricular NGF injection or implantation of NGF-producing fibroblasts has been demonstrated with hippocampal neurons after transient experimental ischemia (Shigeno et al., 1991; Pechan et al., 1995). Furthermore, endogenous NGF synthesis, increased by a monoamine oxidase B inhibitor or a ␤2 adrenergic receptor agonist, protects rat and mouse cortical tissues from permanent ischemic damage (Semkova et al., 1996a,b). In a mouse model of permanent focal ischemia induced by middle cerebral artery

180

0969-9961/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

181

Mechanisms of Neuroprotection after Cerebral Ischemia

upregulating the Bcl-2 protein (Katoh et al., 1996). NGF also protects cells in in vitro models from oxidative damage by stimulating copper zinc superoxide dismutase (CuZn SOD), catalase, and glutathione peroxidase (GPx) activities (Pan & Perez-Polo, 1993; Sampath et al., 1994; Mattson et al., 1995). The use of transgenic animals and animals with knockout mutations has confirmed the protective role of SOD and GPx in models of focal cerebral ischemia (Yang et al., 1994; Kondo et al., 1997; Weisbrot-Lefkowitz et al., 1998). Finally, NGF reverses the decrease, in vivo, of catalase activity and increases SOD and GPx activities in the brain of aged rats (Nistico` et al., 1992). The activity of CuZn and manganese (Mn) SOD, GPx, and glutathione transferase (GSH-T) has been investigated in wt and tg mice subjected to MCAO, at various times after occlusion. We found that both the antiapoptotic protein Bcl-2 and the antioxidant enzyme GPx were recruited in wt and tg mice. In addition, in tg mice, the basal activities of both CuZn SOD and GSH-T were higher than those of wt animals, suggesting that tg mice were more resistant to oxidative stress. Mn SOD activity was also upregulated as a function of time after MCAO, suggesting that Mn SOD was involved in the neuroprotection observed in NGF-tg mice. The parallel between these results and the decrease in apoptotic cell death in this model suggests that Mn SOD is involved in the resistance to ischemia-induced apoptosis.

METHODS MCAO Procedure and Determination of Infarct Volume All studies were conducted according to EC legislation concerning animal care. Male NGF-tg mice homozygous for the transgene or wt mice of the same genetic background (C57Bl/6 ⫻ DBA/2)F1 weighing 20–25 g were anesthetized with chloral hydrate (500 mg/kg). Under low-power magnification, a skin incision was made vertically between the eye and the ear. The parotid gland and surrounding soft tissues were pushed downward and the underlying temporalis was incised. The tissue was retracted until the middle cerebral artery was visible through the surface of the skull. A craniotomy was performed with a small round burr, the dura was opened, and the MCA was electrocoagulated with a bipolar diathermy. The temporalis and the parotid gland were replaced and the incision was sutured. During the surgical procedure, and until

recovery from anesthesia, the body temperature of the mice was maintained at 37–38°C. Under these surgical conditions, the mortality rate was 8% at 3 h, 0% at 24 h, and 18% at 7 days. At various times after MCAO (3 h, 24 h, and 7 days), animals were sacrificed and infarcted areas were measured as previously described (Gue´gan et al., 1998a). Briefly, frozen ischemic brains were embedded in histological medium and cut into coronal sections (15 µm) from the most caudal to the most rostral part of the infarct. Sections were stained with cresyl violet and the infarcted areas were measured with a RAG 200 image analyzer (Biocom). The size of the infarcted area of each section was calculated by subtracting the area of healthy tissue contralateral and ipsilateral to the lesion. The ischemic areas of wt and tg mice were compared statistically using a two-way ANOVA and Student’s t test. Determination of Antioxidant Enzyme Activity Mn SOD, CuZn SOD, GPx, and GSH-T activities were determined in right and left cortices obtained from unoperated wt and tg mice and in the contra- and ipsilateral cortices of wt and tg mice sacrificed 3, 6, and 24 h after MCAO (n ⫽ 6 for each group). Cortices were homogenized in 50 mM NaH2PO4, pH 7.4, and centrifuged at 15,000g for 1 h at 4°C. Enzyme activities in the supernatant were determined in triplicate on the automatic Cobias-Bio. Enzyme activity and protein content determinations were carried out as described in detail elsewhere (Ceballos-Picot et al., 1992). Total SOD activity was determined from the inhibition of the autooxidation of pyrogallol, as described by Marklund and Marklund (1974). Mn SOD activity was determined under the same conditions with addition to the assay buffer of 2 mM KCN for 15 min to inhibit CuZn SOD activity. CuZn SOD activity was determined either by subtracting Mn SOD activity from total activity or by Mn SOD protein precipitation using ethanol/chloroform (0.25/0.15 vol/vol) for 30 min on ice (CeballosPicot et al., 1996). Similar results were obtained with both methods. One unit of SOD activity was defined as 50% inhibition of pyrogallol auto-oxidation (U/mg of protein). Statistical analysis was carried out by oneway ANOVA followed by a Student two-tailed t test for comparison of values. Values of P ⬍ 0.05 were considered statistically significant. Immunohistochemical Analysis Immunohistochemistry was performed on brain sections obtained from unoperated wt and tg mice, and Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

182 from animals sacrificed 3, 6, and 24 h (n ⫽ 3, for each group) after MCAO, by intracardiac perfusion of a buffered 4% paraformaldehyde solution. The fixed brains were embedded in paraffin blocks and cut into 6 µm-thick coronal sections. Immunohistochemistry was carried out on adjacent sections using the avidin–biotin– peroxidase method (ABC; Vector Laboratories). Rabbit polyclonal antibodies directed against Bcl-2 (Santa Cruz Biotechnologies, Inc., No. sc-783), Bax (Upstate Biotechnologies, Inc., 06–499), Bcl-X (Santa Cruz Biotechnologies, Inc., No. sc-634), and Bak (Upstate Biotechnologies Inc., 06–536) were used at final dilutions of 1/100, 1/500, 1/200, and 1/200, respectively. The polyclonal anti-Bcl-X antibody recognizes Bcl-XS and Bcl-XL proteins. The peroxidase reaction was visualized with diaminobenzidine (Sigma) as a substrate, with or without subsequent hematoxylin counterstaining. The specificity of the reaction was tested by incubating sections with biotinylated secondary antibody only (goat anti-rabbit IgG; Vector Laboratories).

Western Blot Analysis Contralateral and ipsilateral cortices were obtained from unoperated wt or tg animals or from animals sacrificed 3, 6, or 24 h after MCAO (n ⫽ 3 for each group). They were homogenized in 1 ml of ice-cold buffer containing 50 mM Tris, pH 7.4, 150 mM NaCl, 0.5% Triton X-100, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 5 µl protease inhibitor cocktail (Sigma) and centrifuged at 4°C for 30 min at 12,000 g. Total protein was pelleted and protein concentration determined using the Protein Assay Kit II (Bio-Rad). Proteins (50 µg) were separated by electrophoresis in a 12% SDS–polyacrylamide gel. The proteins were then blotted onto PVDF membranes (Immobilon, Millipore) at 400 mA for 2 h at 4°C and the membranes were blocked by incubation for 1 h at room temperature in Tris-buffered saline containing 0.05% Tween 20 (TBS-T) plus 5% nonfat milk powder. The membranes were then washed with TBS-T. Membranes were incubated at room temperature for 2 h with primary antibodies diluted in TBS-T/5% milk. The primary antibodies were as for immunohistochemical analysis, with the following dilutions: Bax (1/500), Bcl-X (1/500), Bcl-2 (1/500), and Bak (1/1,000). A goat polyclonal anti-Bad antibody (Santa Cruz Biotechnologies, Inc., No. sc942G) was also used at the final dilution of 1/500. Incubation with the secondary antibody (anti-rabbit IgG peroxidase conjugate 1/80,000 dilution or antiCopyright r 1999 by Academic Press All rights of reproduction in any form reserved.

Gue´gan et al.

goat IgG peroxidase conjugate 1/8,000 dilution; Sigma) was carried out at room temperature for 1 h. Chemiluminescent substrate (NEN; Life Science) was applied and blots were placed against X-Omat AR5 film (Kodak). Western blots were analyzed by densitometry, using an optical density scanner.

RESULTS Following MCAO, mice had an ischemic lesion which was strictly ipsilateral and restricted to the temporoparietal cortex (Fig. 1). The infarcted area, detectable as early as 30 min after injury, expanded outwards from the core to reach a maximum volume 24 h after occlusion. Comparison of infarct sizes for the wt and tg groups, 24 h after the occlusion, showed that tg mice had smaller infarcts than wt mice (Table 1), affecting the rostrocaudal axis (Fig. 2). We integrated the infarct sizes over the distance between each level and this showed that 40% less of the brain volume was injured in tg mice, this difference lasting for up to 1 week (Gue´gan et al., 1998a). We have previously reported that in tg mice, NGF exerts its protective role by impairing apoptotic cell death associated with ischemia. Indeed, the density of apoptotic profiles in tg mice was 30% lower than that in wt mice within the infarcted territory (Gue´gan et al., 1998a).

FIG. 1. Cresyl violet staining of a coronal section of ischemic brain. Stained coronal section corresponding to A2750 µm from the interaural line (according to Lehmann’s atlas, 1974) obtained from a wt mice sacrificed 24 h after MCAO. The infarcted area was identified by the pallor of the damaged cortical tissue relative to the surrounding healthy tissue.

183

Mechanisms of Neuroprotection after Cerebral Ischemia

Immunohistochemistry of Bcl-2 Family Members The level of Bcl-2, Bax, Bak, and Bcl-X proteins was analyzed in brain sections by immunohistochemistry. In all experiments, sections incubated with the secondary antibody were strictly negative (data not shown). As previously reported (Gue´gan et al., 1998b), significant levels of Bcl-2 were not detected in control brain or in the contralateral cortex of wt ischemic brain. However, 3 and 24 h after occlusion, Bcl-2-positive neurons, mostly located in the border zone of the infarct, were present. Neurons producing Bcl-2 protein were likely to survive after ischemia; their number and their location were similar in wt and tg mice (data not shown). Bcl-X protein was present in subcortical and cortical structures in unoperated mice (Fig. 3A1). The location of Bcl-X was unchanged in the contralateral cortex of operated mice 3 and 24 h after MCAO (Fig. 3A2). Neurons within the infarcted territory still contained Bcl-X protein 3 and 24 h after occlusion (Figs. 3A3 and 3A4). No difference in Bcl-X distribution or level was identified in wt and tg groups. A strong basal level of Bax immunostaining was detected in normal control brains (Fig. 3B1). Bax protein was present in the hippocampus, neocortex, and striatum. Bax immunostaining in the contralateral (Fig. 3B2) and ipsilateral cortices (Figs. 3B3 and 3B4) of operated mice did not change, regardless of the time after MCAO. In the infarcted zone, Bax protein was present in neurons already engaged in an apoptotic process, since they had lost neurites and had a shrunken cell body (Figs. 3B3 and 3B4). No difference in the

TABLE 1 Measurement of Infarcted Areas of Seven Coronal Levels in wt and tg Mice Coronal level

wt

tg

A1300 A2100 A2750 A3250 A4150 A4750 A5650

0.96 ⫾ 0.82 3.43 ⫾ 0.73 5.67 ⫾ 1.02 7.96 ⫾ 1.25 8.91 ⫾ 1.35 7.43 ⫾ 1.01 4.10 ⫾ 0.96

0.31 ⫾ 0.50 1.71 ⫾ 0.50* 2.68 ⫾ 0.77*** 4.08 ⫾ 0.87** 5.40 ⫾ 0.98** 4.28 ⫾ 0.97** 2.76 ⫾ 0.97*

Note. Infarcted areas (in mm2 ) were measured in wt (n ⫽ 5) and tg (n ⫽ 6) animals 24 h after MCAO. Means ⫾ SD are reported for each coronal level according to Lehmann’s atlas (1974). For statistical analysis, two-way ANOVA, followed by Student’s t test was used. Significant differences between wt and tg mice are indicated as follow: *P ⬍ 0.05; **P ⬍ 0.002; ***P ⬍ 0.001.

FIG. 2. Schematic representation of ischemic areas in wt and tg mice 24 h after MCAO. The ischemic areas (in black) on the following stereotaxic levels are shown from the back to the front: A1300, A2100, A2750, A3250, A4150, A4750, and A5650. See footnote to Table 1.

spatiotemporal pattern of Bax synthesis was observed between wt and tg mice. Bak was detected in control brains (Fig. 3C1) and in hippocampus, neocortex, and striatum. Ischemia caused no change in the immunoreactivity pattern observed, regardless of the type of cortex analyzed (Figs. 3C2 and 3C3), the time after MCAO (Figs. 3C3 and 3C4), or mouse group (data not shown).

Western Blotting of Bcl-2 Family Proteins We performed a series of Western blots, analyzed by densitometry, to detect more precisely any difference in the pattern of synthesis of Bcl-2 family members in time and space and between wt and tg animals. The patterns of Bcl-2 family protein synthesis were the same in contra- and ipsilateral cortices, so only ipsilateral cortices are shown in Fig. 4. The analysis of Bcl-2 levels showed no significant difference as a function of time or between mouse groups (Fig. 4 and data not shown). Two bands of 30 and 23 kDa, respectively, corresponding to Bcl-XL and Bcl-XS, were detected by Western blotting. Their ratio was not changed by ischemia. In fact, no change in the levels of Bcl-X, Bax, Bak, and Bad proteins was detected between the two mouse groups at any time point (Fig. 4).

Analysis of Glutathione-Related Enzyme Activities For both wt and tg mice, no difference in GSH-T activity over time was observed between contra- and Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

FIG. 3. Immunodetection of Bcl-2-related proteins in coronal brain sections. Anti-Bcl-X (A), anti-Bax (B), and anti-Bak (C) antibodies were used on the left cortex of unoperated mice (1), on the contralateral cortex (2), and on the ipsilateral cortex, within the ischemic zone at 3 (3) and 24 h (4) after occlusion. Magnification ⫻400, insets ⫻1000.

184

Mechanisms of Neuroprotection after Cerebral Ischemia

FIG. 4. Western blot analysis of Bcl-2, Bcl-X, Bax, Bak, and Bad proteins. Protein samples isolated from the ipsilateral cortices of wt and tg mice, unoperated or sacrificed 3, 6, or 24 h after MCAO, were subjected to SDS–PAGE and immunoblotted with anti-Bcl-2, antiBcl-X, anti-Bax, anti-Bak, and anti-Bad antibodies. The sizes of the corresponding proteins were determined using standard molecular weight markers. The size of each protein in kilodaltons is indicated on the left-hand side.

ipsilateral cortices. However, GSH-T activity was significantly higher in tg than in wt mice (t test, P ⬍ 0.05) (Fig. 5). As previously reported (Gue´gan et al., 1998b), GPx activity increased after 6 h and continued to increase until 24 h after occlusion in both contra- and ipsilateral cortices in wt mice. This pattern of GPx activity was not observed in tg mice, but the basal level of GPx in tg mice was close to the maximum activity reached after ischemia in wt mice (Fig. 5).

Determination of Superoxide Dismutase Activity In wt mice, no differences in CuZn and Mn SOD activity were observed, regardless of the cortex analyzed and the time after occlusion (Fig. 6). The tg mouse group had higher basal CuZn SOD activity than wt mice, which was not affected by MCAO. Mn SOD activity increased with time after occlusion, in both contra- and ipsilateral cortices in tg mice (t test, P ⬍ 0.05) (Fig. 6).

DISCUSSION In a model of permanent focal ischemia, NGF-tg mice subjected to MCA occlusion consistently had 40%

185 smaller infarcted areas than wt mice. The inhibition of the spatiotemporal progression of the infarct was correlated with the upregulation of NGF synthesis and the impairment of apoptotic cell death (Gue´gan et al., 1998a). We investigated the cellular mechanisms of neuroprotection by studying the levels of Bcl-2 family proteins in wt and tg mice sacrificed at various times after occlusion. Bcl-2 protein levels were stable, as shown by Western blotting, throughout ischemia, but we have previously reported the selective induction of Bcl-2 around the infarcted area (Gue´gan et al., 1998b). This difference in results is likely due to the small number of Bcl-2-immunoreactive neurons. The Bcl-2-positive neurons are located in the so-called penumbra, in which neurons can recover after injury. Our results are consistent with previous reports in a transient focal ischemia model (Chen et al., 1995) and in the same model of permanent ischemia (Gillardon et al., 1996; Asahi et al., 1997). In the permanent ischemia model, according to our data, no significant change in the expression of bcl-X was observed (Asahi et al., 1997). We used a polyclonal antibody directed against both the antiapoptotic Bcl-XL and proapoptotic Bcl-XS forms. The spatial location of Bcl-X-immunoreactive neurons suggests that the neurons within the infarcted area that underwent apoptosis probably produced the Bcl-XS form, whereas neurons in the healthy part of the brain produced the Bcl-XL form. Synthesis of the proapoptotic Bax protein is associated with the cell death of vulnerable neurons in global (Krajewski et al., 1995; Chen et al., 1996) and permanent focal ischemia in rats (Gillardon et al., 1996). However, although Bax protein was present within the vulnerable neurons of the cortical areas, we detected Bax in neurons of various cerebral structures. In sympathetic neurons, Bak, like Bax, is sufficient to induce apoptosis in the absence of another cell death stimulus (Martinou et al., 1998), and Bak protein levels are upregulated in Alzheimer’s disease (Kitamura et al., 1998). Like Bax, Bak is constitutively present in brain structures. Neurons destined to die within the ischemic zone contained Bak, but no change in Bak level was detected during ischemia. To our knowledge, this is the first report concerning the expression of the proapoptotic Bak protein in mice subjected to ischemic damage. The involvement of Bcl-2 and Bcl-X as antiapoptotic factors and of Bax as a proapoptotic molecule has been reported in several paradigms of neuronal death (MacManus & Linnik, 1997), but their specific effects in ischemia are still unknown. Of the Bcl-2 family members we studied, only Bcl-2 expression changed after MCA occlusion. Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

186

Gue´gan et al.

FIG. 5. Histograms of GSH-T and GPx activities. GSH-T activity is expressed in micromoles of CDNB conjugated per gram of protein per minute. GPx activity is expressed in nanomoles of NADPH oxidized per minute per milligram of protein. Values are means ⫾ SEM. *P ⬍ 0.05 with a one-way ANOVA followed by a t test.

However, as the effects of Bcl-2 members are mediated via their interactions (homo- or heterodimerization), we cannot exclude the possibility that changes of partners occur during ischemic neuronal death. Bax has been shown to disrupt Bcl-XL/Bcl-2 interactions leading to ischemic cell death in transient global ischemia (Antonawich et al., 1998). In vitro, NGF directly exerts its neuroprotective action, preventing neuronal-like cell death by upregulating Bcl-2 protein (Katoh et al., 1996). We found no changes in the levels of proteins from the Bcl-2 family in NGF-tg mice subjected to cerebral ischemia, suggesting that reduction of infarct was mediated by the dimerization of Bcl-2-related proteins or by other cellular mechanisms. The brain is particularly susceptible to deficits in blood supply. During ischemia, ATP production deCopyright r 1999 by Academic Press All rights of reproduction in any form reserved.

creases at the same time as mitochondrial dysfunction occurs, leading to oxidative stress (for review, see Peuchen et al., 1997). Several enzymes are involved in cell defense, especially CuZn SOD and Mn SOD, which metabolize the superoxide radical, and GPx, a key protective enzyme against reactive oxygen species. In addition to its role in detoxifying xenobiotics, GSH-T, which is mostly located in the glial compartment but is also present in neurons (Cammer et al., 1989; Johnson et al., 1993), has peroxidase activity (Peuchen et al., 1997). NGF may protect cells against oxidative stress by increasing, in particular, CuZn SOD and GPx activities, in vitro and in vivo in the brains of aged rats (Nistico` et al., 1992; Pan & Perez-Polo, 1993; Sampath et al., 1994; Mattson et al., 1995). Following permanent focal ischemia, CuZn SOD and GSH-T

Mechanisms of Neuroprotection after Cerebral Ischemia

187

FIG. 6. Histograms of CuZn and Mn SOD activities. Mn and CuZn SOD activities are expressed in units per milligram of protein. Values are means ⫾ SEM. *P ⬍ 0.05 with a one-way ANOVA followed by a t test.

activities were unchanged in wt and NGF-tg mice. Thus, although, CuZn SOD-tg mice were resistant to reperfusion injury after focal ischemia, the size of the brain infarction was not reduced after permanent focal cerebral ischemia (Chan et al., 1993; Yang et al., 1994; our unpublished results). These data suggest that CuZn SOD and GSH-T enzymes were not recruited via the neuroprotective pathways induced by the damage caused by permanent ischemia. However, NGF-tg mice had intrinsically high resistance to oxidative stress. As the basal level of GPx activity was close to the maximum level reached after ischemia in wt mice, the higher level of GPx activity, although not statistically significant in unoperated tg versus wt animals, may be involved in the neuroprotection observed after ischemia in tg mice. Thus, our data demonstrate the involvement of GPx in the neuroprotective pathways set up after permanent or transient focal ischemia as

previously described (Gue´gan et al., 1998b; WeisbrotLefkowitz et al., 1998). Mn SOD activity was stable in wt mice, whereas it increased in NGF-tg mice after MCA occlusion, suggesting an essential role for Mn SOD in neuroprotection against ischemic cell death. The Mn SOD enzyme, present in the mitochondrial membrane, protects cortical neurons from NMDA-mediated toxicity (GonzalesZulueta et al., 1998). In a model of transient focal ischemia, the rate of death of cortical cells was lower in Mn SOD-tg mice (Keller et al., 1998). Moreover, mutant mice with Mn SOD deficiency have exacerbated cerebral infarction following permanent focal ischemia (Murakami et al., 1998). NGF-tg mice had higher basal resistance to oxidative stress, but the reduction in apoptosis in these mice may result from an antiapoptotic function of Mn SOD. During the effector phase of apoptosis, mitochondrial transmembrane potential is Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

188 disrupted. Keller et al. (1998) reported, recently, that mitochondrial transmembrane potential is maintained in neural cells overproducing Mn SOD. They suggest that mitochondrial functions mediated by Mn SOD are pivotal in the regulation of neuronal apoptosis. As NGF-tg mice had much lower levels of apoptotic cell death, it is possible that this effect resulted from the enhancement of Mn SOD activity during ischemia. Further investigations are required to understand the relationships between NGF, Mn SOD, and neuronal apoptosis.

ACKNOWLEDGMENTS C.G. is a Fellow of Ministe`re de l’Education Nationale, de la Recherche et de la Technologie. The authors thank Johan Van Beek for his help with the densitometric scanning procedure.

REFERENCES Antonawich, F. J., Krajewski, S., Reed, J. C., & Davies, J. N. (1998) Bcl-xl Bax interaction after transient global ischemia. J. Cereb. Blood Flow Metab. 18, 882–886. Asahi, M., Hoshimaru, M., Uemura, Y., Tokime, T., Kojima, M., Otsuka, T., Matsuura, N., Aoki, T., Shibahara, K., & Kikuchi, H. (1997) Expression of interleukin-1␤ converting enzyme gene family and bcl-2 gene family in the rat brain following permanent occlusion of the middle cerebral artery. J. Cereb. Blood Flow Metab. 17, 11–18. Cammer, W., Tansey, F., Abramovitz, M., Ishigaki, S., & Listowsky, I. (1989) Differential localization of glutathione-S-transferase Yp and Yb subunits in oligodendrocytes and astrocytes of rat brain. J. Neurochem. 52, 876–883. Ceballos-Picot, I., Nicole, A., Cle´ment, M., Bourre, J.-M., & Sinet, P.-M. (1992) Age-related changes in antioxidant enzymes and lipid peroxidation in brains of control and transgenic mice overexpressing copper–zinc superoxide dismutase. Mutat. Res. 275, 281–293. Ceballos-Picot, I., Witko-Sarsat, V., Merad-Boudia, M., Nguyen, A. T., The´venin, M., Jaudon, M. C., Zingraff, J., Verger, C., Jungers, P., & Descamps-Latscha, B. (1996) Glutathione antioxidant system as a marker of oxidative stress in chronic renal failure. Free Radical Biol. Med. 21, 845–853. Chan, P. H., Kamii, H., Yang, G., Gafni, J., Epstein, C. J., Carlson, E., & Reola, L. (1993) Brain infarction is not reduced in SOD-1 transgenic mice after a permanent focal cerebral ischemia. NeuroReport 5, 293–296. Chen, J., Graham, S. H., Chan, P. H., Lan, J., Zhou, R. L., & Simon, R. P. (1995) bcl-2 is expressed in neurons that survive focal ischemia in the rat. NeuroReport 6, 394–398. Chen, J., Zhu, R. L., Nakayama, M., Kawaguchi, K., Jin, K., Stetler, R. A., Simon, R. P., & Graham, S. H. (1996) Expression of the apoptosis-effector gene, Bax, is up-regulated in vulnerable hippocampal CA1 neurons following global ischemia. J. Neurochem. 67, 64–71. Gao, W.-Q., Weil, R. J., Dugich-Djordjevic, M., & Lu, B. (1997) The therapeutic potentials of neurotrophic factors for diseases of the nervous system. Exp. Opin. Ther. Pat. 7, 325–338.

Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

Gue´gan et al. Gillardon, F., Lenz, C., Waschke, K. F., Krajewski, S., Reed, J. C., Zimmermann, M., & Kuschinsky, W. (1996) Altered expression of Bcl-2, Bcl-X, Bax and c-Fos colocalizes with DNA fragmentation and ischemic cell damage following middle cerebral artery occlusion in rats. Mol. Brain Res. 40, 254–260. Gonzalez-Zulueta, M., Ensz, L. M., Mukhina, G., Lebovitz, R. M., Zwacka, R. M., Engelhardt, J. F., Oberley, L. W., Dawson, V. L., & Dawson, T. M. (1998) Manganese superoxide dismutase protects nNOS neurons from NMDA and nitric oxide-mediated neurotoxicity. J. Neurosci. 18, 2040–2055. Gue´gan, C., Onte´niente, B., Makiura, Y., Merad-Boudia, M., CeballosPicot, I., & Sola, B. (1998a) Reduction of cortical infarction and impairment of apoptosis in NGF-transgenic mice subjected to permanent focal ischemia. Mol. Brain Res. 55, 133–140. Gue´gan, C., Ceballos-Picot, I., Nicole, A., Kato, H., Onte´niente, B., & Sola, B. (1998b) Recruitment of several neuroprotective pathways after permanent focal ischemia in mice. Exp. Neurol., 154, 371–380. Johnson, J. A., Barbary, A. E., Kornguth, S. E., Brugge, J. F., & Siegel, F. L. (1993) Glutathione-S-transferase isoenzymes in rat brain neurons and glia. J. Neurosci. 13, 2013–2023. Katoh, S., Mitsui, Y., Kitani, K., & Suzuki, T. (1996) Nerve growth factor rescues PC12 cells from apoptosis by increasing amount of bcl-2. Biochem. Biophys. Res. Commun. 229, 653–657. Keller, J. N., Kindy, M. S., Holtsberg, F. W., St Clair, D. K., Yen, H.-C., Germeyer, A., Steiner, S. M., Bruce-Keller, A. J., Hutchins, J. B., & Mattson M. P. (1998) Mitochondrial manganese superoxide dismutase prevents neural apoptosis and reduces ischemic brain injury: Suppression of peroxynitrite production, lipid peroxidation and mitochondrial dysfunction. J. Neurosci. 18, 687–697. Kitamura, Y., Shimohama, S., Kamoshima, W., Ota, T., Matsuoka, Y., Nomura, Y, Smith, M. A., Perry, G., Whitehouse, P. J., & Taniguchi, T. (1998) Alteration of proteins regulating apoptosis, Bcl-2, Bcl-x, Bax, Bak, Bad, ICH-1 and CPP32, in Alzheimer’s disease. Brain Res. 780, 260–269. Kondo, T., Reaume, A. G., Huang, T.-T., Carlson, E., Murakami, K., Chen, S. F., Hoffman, E. K., Scott, R. W., Epstein, C. J., & Chan, P. H. (1997) Reduction of CuZn-superoxide dismutase activity exacerbates neuronal cell injury and edema formation after transient focal cerebral ischemia. J. Neurosci. 17, 4180–4189. Krajewski, S., Mai, J. K., Krajewska, M., Sikorska, M., Mossakowski, M. J., & Reed, J. C. (1995) Upregulation of Bax protein levels in neurons following cerebral ischemia. J. Neurosci. 15, 6364–6376. Lehmann, A. (1974) Atlas Ste´re´otaxique du Cerveau de la Souris. Presses du CNRS, Paris. Lindvall, O., Kokaia, Z., Bengzon, J., Elme´r, E., & Kokaia, M. (1994) Neurotrophins and brain insults. Trends Neurosci. 17, 490–496. MacManus, J. P., & Linnik, M. D. (1997) Gene expression induced by cerebral ischemia: An apoptotic perspective. J. Cereb. Blood Flow Metab. 17, 815–832. Marklund, S., & Marklund, G. (1974) Involvement of superoxide anion radical in the autoxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur. J. Biochem. 47, 469–474. Martinou, I., Missotten, M., Fernandez, P.-A., Sadoul, R., & Martinou, J.-C. (1998) Bax and Bak proteins require caspase activity in sympathetic neurons. NeuroReport 9, 15–19. Mattson, M. P., Lovell., M. A., Furukawa, K., & Markesbery, W. R. (1995) Neurotrophic factors attenuate glutamate-induced accumulation of peroxides, elevation of intracellular Ca⫹⫹ concentration, and neurotoxicity and increase antioxidant enzyme activities in hippocampal neurons. J. Neurochem. 65, 1740–1751. Murakami, K., Kondo, T., Kawase, M., Li, Y., Sato, S., Chen, S. F., & Chan, P. H. (1998) Mitochondrial susceptibility to oxidative stress

Mechanisms of Neuroprotection after Cerebral Ischemia exacerbates cerebral infarction that follows permanent focal ischemia in mutant mice with manganese superoxide dismutase deficiency. J. Neurosci. 18, 205–213. Nistico`, G., Ciriolo, M. R., Fiskin, K., Iannone, M., de Martino, A., & Rotilio, G. (1992) NGF restores decrease in catalase activity and increases superoxide dismutase and glutathione peroxidase activity in the brain of aged rats. Free Radical Biol. Med. 12, 177–181. Pan, Z., & Perez-Polo, J. R. (1993) Role of nerve growth factor in oxidant homeostasis: Glutathione metabolism. J. Neurochem. 61, 1713–1721. Pechan, P. A., Yoshida, T., Panahian, N., Moskowitz, M. A., & Breakefield, X. O. (1995) Genetically modified fibroblasts producing NGF protect hippocampal neurons after ischemia in the rat. NeuroReport 6, 669–672. Peuchen, S., Bolan˜os, J. P., Heales, S. J. R., Almeida, A., Duchen, M. R., & Clark, J. B. (1997) Interrelationships between astrocyte function, oxidative stress and antioxidant status within the central nervous system. Prog. Neurobiol. 52, 261–281. Sampath, D., Jackson, G. R., Werrbach-Perez, K., & Perez-Polo, J. R. (1994) Effects of nerve growth factor on glutathione peroxidase and catalase in PC12 cells. J. Neurochem. 62, 2476–2479.

189 Semkova, I., Wolz, P., Schilling, M., & Krieglstein, J. (1996a) Selegiline enhances NGF synthesis and protects central nervous system neurons from excitotoxic and ischemic damage. Eur. J. Pharmacol. 315, 19–30. Semkova, I., Schilling, M., Henrich-Noack, P., Rami, A., & Krieglstein, J. (1996b) Clenbuterol protects mouse cerebral cortex and rat hippocampus from ischemic damage and attenuates glutamate neurotoxicity in cultured hippocampal neurons by induction of NGF. Brain Res. 717, 44–54. Shigeno, T., Mima, T., Takakura, K., Graham, D. I., Kato, G., Hashimoto, Y., & Furukawa S. (1991) Amelioration of delayed neuronal death in the hippocampus by nerve growth factor. J. Neurosci. 11, 2914–2919. Weisbrot-Lefkowitz, M., Reuhl, K., Perry, B., Chan, P. H., Inouye, M., & Mirochnitchenko, O. (1998) Overexpression of human glutathione peroxidase protects transgenic mice against focal cerebral ischemia/reperfusion damage. Mol. Brain Res. 53, 333–338. Yang, G., Chan, P. H., Chen, J., Carlson, E., Chen, S. F., Weinstein, P., Epstein, C. J., & Kamii H. (1994) Human copper–zinc superoxide dismutase transgenic mice are highly resistant to reperfusion injury after focal cerebral ischemia. Stroke 25, 165–170.

Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.