Neuroprotective Effect of Eicosapentaenoic Acid in Hippocampus of Rats Exposed to γ-Irradiation*

Exposure to irradiation leads to detrimental changes in several cell types. In this study we assessed the changes induced in hippocampus by exposure of rats to whole body irradiation; the findings reveal that irradiation leads to apoptotic cell death in hippocampus, and as a consequence, long term potentiation in perforant path-granule cell synapses is markedly impaired. The evidence is consistent with the view that irradiation induced an increase in reactive oxygen species and that this leads to stimulation of the stress-activated protein kinase, JNK, and activation of the transcription factor, c-Jun. Consequent upon activation of JNK, a cascade of cell signaling events was stimulated that ultimately resulted in apoptosis, as suggested by parallel increases in cytochromec translocation, caspase-3 activation, poly(ADP-ribose) polymerase cleavage, and terminal dUTP nick-end labeling staining. Treatment of rats with eicosapentaenoic acid inhibited the irradiation-induced increase in reactive oxygen species production and the subsequent cellular signaling events, suggesting that oxidative stress triggered apoptotic cell death in the hippocampus of rats exposed to irradiation. Significantly, when the compromise in cell viability induced by irradiation was prevented by eicosapentaenoic acid, long term potentiation was sustained in a manner similar to that in the sham-treated control group.

It has been recognized for some time that neuronal tissue is susceptible to oxidative stress because of its high oxygen consumption and modest antioxidant defenses (1). These features are coupled with a high concentration of iron (2) and high concentrations of polyunsaturated fatty acids, which are easily oxidized and known to generate oxygen radicals following oxidation (3). Reactive oxygen species (ROS), 1 in particular the hydroxyl radical, can lead to functional alterations in lipids, proteins, and nucleic acids, and an accumulation of ROS is considered to be one factor that contributes to neurodegenerative changes, for example in Parkinson's disease (4) and Alzheimer's disease (5). Indeed recent data from studies undertaken in this laboratory have indicated that there is an increase in ROS accumulation in the hippocampus prepared from aged, compared with young, rats and that this is accompanied by an impairment in long term potentiation (LTP) in perforant path-granule cell synapses (6). Further studies have revealed that activities of two mitogen-activated protein (MAP) kinases, c-Jun NH 2 -terminal kinase (JNK) and p38 MAP kinase, are increased in vivo when ROS accumulation is increased, for example in hippocampal tissue prepared from aged rats (7) or rats injected intracerebroventricularly with the proinflammatory cytokine, interleukin-1␤ (IL-1␤) (8). Significantly, dietary manipulation with -3 fatty acids, which restored the ability of aged rats to sustain LTP, reversed the age-related activation of JNK and p38, which accompanied the deficit in LTP (7).
An increase in ROS production occurs following irradiation (9), and therefore, it is not surprising to find that at least some effects of exposure of cells to ROS are mimicked by exposure of cells to ionizing radiation. For example, exposure of guinea pigs to ␥-irradiation is associated with impaired LTP in CA1 slices (10 -12), and H 2 O 2 has been shown to inhibit LTP in CA1 in guinea pig slices (13). Similarly, LTP is diminished in dentate gyrus in vivo in rats which received H 2 O 2 intracerebroventricularly (14) and in IL-1␤-treated and aged rats (7,15) in which an increase in hippocampal ROS accumulation has been described. It has also been shown that the activities of both JNK and p38 are increased by irradiation (16,17) as well as by H 2 O 2 (7). Interestingly, it has been reported that exposure of Jurkat T cells to ␥-irradiation led to increased JNK activation (16), and its activation (but not activation of either p42 MAP kinase or p38) was correlated with apoptotic changes (18). Several reports have implicated the JNK pathway in apoptosis in a variety of cell types (for example, PC12 cells (19,20), Neuro2A cells (20,21), SHSY5Y cells (20)), and in a dopaminergic cell line (22). It is considered that JNK-stimulated phosphorylation of the transcription factor, c-Jun, represents an important step in the cascade of events leading to apoptosis (23)(24)(25). The JNK signaling pathway is modulated by JNK-interacting protein (JIP), which was originally identified as a cytoplasmic inhibitor of JNK (26) and is now recognized as a scaffold protein which, when overexpressed in cells, prevents translocation of JNK to the nucleus and consequently inhibits JNK-mediated gene expression (27). Significantly, the mouse homologue of JIP, islet brain 1 (IB1), has been shown to be highly expressed in hippocampus (28).
We have reported that age-related deficits in LTP and the accompanying changes in IL-1␤ concentration, ROS accumulation, and JNK and p38 activation can be prevented by dietary manipulation with essential fatty acids, for example the -3 fatty acid, docosahexaenoic acid (7,29). Here we investigate the possibility that its precursor, eicosapentaenoic acid (EPA), which has been shown to have anti-inflammatory properties (30,31) and which inhibited the age-related increase in IL-1␤ in hippocampus (32), might be effective in protecting hippocam-pal neurons from damage induced by whole body irradiation. We present evidence that indicates that hippocampal tissue prepared from rats exposed to whole body irradiation exhibited increases in ROS accumulation and JNK activation and that these changes were coupled with evidence of apoptosis and with impairment in LTP. The data indicate that treatment with EPA prevented these changes and therefore acted as a neuroprotective and antiapoptotic agent.

EXPERIMENTAL PROCEDURES
Animals-In this experiment 24 male Wistar rats were divided into four groups of 6. The animals were housed in groups of 6 in the BioResources Unit, Trinity College, Dublin, Ireland, under a 12-h light schedule, with the ambient temperature controlled between 22 and 23°C. Food and water intake was measured daily for 1 week, and at the end of this period, 12 rats were assigned to a group that was fed on normal laboratory chow supplemented with ethyl EPA (2% v/w, dissolved in corn oil; Laxdale Ltd., UK) for 4 weeks. The remaining 12 rats received control diet (laboratory chow to which only corn oil was added). Rats were offered 100% of their average daily food intake so that the full daily allowance of EPA would be ingested. Sufficient diet was prepared for 2-3 days at a time. Food and water intake did not vary between groups, and there was no significant difference in daily food and water intake before and after dietary modifications were made. Rats were maintained under veterinary supervision for the duration of this experiment.
At the end of this 4-week period, half of rats were exposed to whole body irradiation (10 gray at a rate of 10 gray/min; Nordion Gammacell Cesium Irradiator). The remaining rats were sham-irradiated. Rats were monitored for the 4 days following irradiation and were then assessed for their ability to sustain LTP.
Induction of LTP in Perforant Path-Granule Cell Synapses in Vivo-LTP was induced as described previously (29). Rats were anesthetized by intraperitoneal injection of urethane (1.5 g/kg) and placed in a head holder in a stereotaxic frame. A window of skull was removed to allow placement of recording and stimulating electrodes in the molecular layer of the dentate gyrus (2.5 mm lateral and 3.9 mm posterior to bregma) and perforant path, respectively (angular bundle, 4.4 mm lateral to lambda). The depth of the electrodes was adjusted to obtain maximal responses in the cell body region. Test shocks at the rate of 1/30 s were delivered to the perforant path; base-line recordings were allowed to stabilize for about 15 min, and then electrophysiological recording commenced and continued for 10 min before and 40 min after tetanic stimulation (3 trains of stimuli; 250 Hz for 200 ms; intertrain interval 30 s). At the end of the electrophysiological recording period, rats were killed by decapitation; the hippocampus was removed, dissected on ice, and cross-chopped into slices (350 ϫ 350 m) using a McIlwain tissue chopper. The time taken to prepare slices from the time of death was 2.5-3.5 min. All samples were frozen separately in 1 ml of Krebs solution (composition of Krebs in mM: NaCl 136, KCl 2.54, KH 2 PO 4 1.18, MgSO 4 ⅐7H 2 O 1.18, NaHCO 3 16, glucose 10, CaCl 2 1.13) containing 10% dimethyl sulfoxide as described previously (29,33). For analysis, thawed slices of tissue were rinsed 3 times in fresh Krebs buffer and homogenized in fresh lysis buffer (20 mM HEPES (pH 7.4); 10 mM KCl; 1.5 mM MgCl 2 ; 1 mM EGTA; 1 mM EDTA; 1 mM dithiothreitol; 0.1 mM phenylmethylsulfonyl fluoride; 2 g/ml leupeptin; 2 g/ml aprotinin; 200 mM sucrose), except in the case of tissue that was used for analysis of ROS, which was homogenized in 40 mM Tris-HCl (pH 7.4).
Analysis of Reactive Oxygen Species-The formation of reactive oxygen species was assessed in the impure synaptosomal preparation, P 2 , as described previously (34). The method relies on oxidation of the non-fluorescent probe, 2Ј,7Ј-dichlorofluorescin diacetate, by reactive oxygen species, to the highly fluorescent 2Ј,7-dichlorofluorescein. To assess reactive oxygen species production, synaptosomes were prepared as described previously (29,33) from hippocampal slices obtained from ␥-irradiated and sham-irradiated rats treated with either diet. The resultant pellet was resuspended in 1 ml of ice-cold 40 mM Tris buffer (pH 7.4), and aliquots (1 ml) of homogenate were incubated with 2Ј,7Јdichlorofluorescin diacetate (10 l; final concentration 5 M; from a stock solution of 500 M in methanol; Molecular Probes) at 37°C for 15 min. To terminate the reaction, the dye-loaded synaptosomes were centrifuged at 13,000 ϫ g for 8 min. The pellet was resuspended in 1.5 ml of ice-cold 40 mM Tris buffer (pH 7.4), and fluorescence was monitored at a constant temperature of 37°C at 488 nm excitation (bandwidth 5 nm) and 525 nm emission (bandwidth 20 nm). Reactive oxygen species formation was quantified from a standard curve of 2Ј,7-dichlo-rofluorescein in methanol (range 0.05 to 1 M). Results were expressed as nmol/mg tissue corrected for protein.
Analysis of JNK Phosphorylation, c-Jun Phosphorylation, JIP-1 Expression, Cytochrome c Translocation, and PARP Cleavage-JNK phosphorylation, cytochrome c translocation, PARP cleavage, c-Jun phosphorylation, and JIP-1 expression were analyzed in samples prepared from hippocampal tissue using a method described previously (35). In the case of JNK, PARP, c-Jun, and JIP tissue homogenates were diluted to equalize for protein concentration, and aliquots (100 l, 2 mg/ml) were added to 100 l of sample buffer (Tris-HCl, 0.5 mM (pH 6.8); glycerol 20%; SDS, 2%; ␤-mercaptoethanol, 5%; bromphenol blue, 0.05% w/v), boiled for 5 min, and loaded onto gels (10% SDS for PARP, c-Jun, and JIP and 12% for JNK). In the case of cytochrome c, the cytosolic fraction was prepared by homogenizing slices of hippocampus in lysis buffer and incubating for 20 min on ice and centrifuging (15,000 ϫ g for 10 min at 4°C). The supernatant (i.e. cytosolic fraction) was suspended in sample buffer to a final concentration of 300 g/ml, boiled for 3 min, and loaded (6 g/lane) onto 12% gels. In all cases proteins were separated by application of 30 mA constant current for 25-30 min, transferred onto nitrocellulose strips (225 mA for 75 min), and immunoblotted with the appropriate antibody. To assess JNK activity, proteins were immunoblotted with an antibody that specifically targets phosphorylated JNK (1:300 in Tris-buffered saline/Tween (0.05% Tween 20) containing 0.1% bovine serum albumin; Santa Cruz Biotechnology) for 2 h at room temperature. To assess cleavage of PARP, we immunoblotted with an antibody (1:500 in phosphate-buffered saline (PBS)/ Tween (0.1% Tween 20) containing 2% non-fat dried milk) raised against the epitope corresponding to amino acids 764 -1014 of PARP of human origin (Santa Cruz Biotechnology Inc.). To assess cytochrome c, a rabbit polyclonal antibody (1:250 in PBS/Tween containing 2% non-fat dried milk; Santa Cruz Biotechnology) raised against recombinant protein corresponding to amino acids 1-104 of cytochrome c was used. In the case of JIP-1, we immunoblotted with a goat polyclonal antibody (1:1000 in Tris-buffered saline/Tween containing 0.1% bovine serum albumin; Santa Cruz Biotechnology) raised against a peptide mapped at the carboxyl terminus of JIP-1 of mouse origin for 2 h. To assess phosphorylation of c-Jun, a mouse monoclonal antibody (1:200 in PBS/ Tween containing 2% non-fat dried milk; Santa Cruz Biotechnology) raised against a peptide corresponding to amino acids 56 -69 of c-Jun of human origin was used. Immunoreactive bands were detected as follows: peroxidase-conjugated anti-mouse IgG (Sigma) and SuperSignal chemiluminescence (Pierce) for JNK and c-Jun, peroxidase-conjugated anti-rabbit IgG (Sigma) and SuperSignal (Pierce) for PARP, peroxidaseconjugated anti-rabbit IgG (Sigma) and ECL (Amersham Biosciences) for cytochrome c and peroxide-labeled anti-goat IgG (Vector Laboratories) and SuperSignal (Pierce) for JIP. All immunoblots were exposed to film and processed using a Fuji x-ray processor. Bands were quantified using densitometry (ZERO-Dscan Image Analysis System, Scanalytics, Inc.).
The effects of H 2 O 2 and VIP were also assessed on the activity of JNK. Hippocampal slices were homogenized in either Krebs solution or VIP (10 M) and incubated for 20 min at 37°C with O 2 in Krebs solution in the presence or absence of H 2 O 2 (200 M). The analysis proceeded as described above.
The effects of H 2 O 2 and VIP were also assessed on the activity of caspase-3. Hippocampal slices were homogenized in either Krebs solution or VIP (10 M) and incubated for 20 min at 37°C with O 2 in Krebs solution in the presence or absence of H 2 O 2 (200 M). The analysis of caspase-3 activity proceeded as described above.
Dissociation of Cells and Analysis of TUNEL Staining-Dissociated cells were prepared from slices of hippocampus as described previously (36). Briefly, slices were incubated for 45 min in HEPES-buffered Krebs solution (pH 7.4; composition in mM: NaCl, 145; KCl, 5; MgCl 2 , 1; CaCl 2 , 2; Mg 2 SO 4 , 1; KH 2 PO 4 , 1; glucose, 10; HEPES, 30) containing trypsin (1 mg/ml) and DNase (1600 units/liter) protease X (1 mg/ml), protease XIV (1 mg/ml), and DNase (1600 units/liter). Washed slices were resuspended in 750 l of PBS, triturated with a fire-polished Pasteur pipette, and passed through a nylon mesh filter to remove tissue clumps. Aliquots (100 l) of the cell suspension were spun onto glass slides at 600 rpm for 2 min (Shandon Cytospin 3), and cells were fixed in methanol and stored at room temperature for later use. Apoptotic cell death was assessed using the DeadEnd Colorimetric Apoptosis Detection System (Promega). The cytospun cells were fixed for 25 min with 4% paraformaldehyde, washed twice for 5 min in PBS, permeabilized with proteinase K, and washed. Cells were immersed for 10 min in equilibration buffer and incubated at 37°C for 60 min with TDT mix, containing TdT enzyme (25 units/l), biotinylated nucleotide mix, and equilibration buffer. The reaction was terminated by incubating the cells in sodium chloride/sodium citrate buffer for 15 min at room temperature. Cells were washed three times in PBS to remove unincorporated biotinylated nucleotides, and endogenous peroxidases were blocked by incubating in the presence of 0.3% H 2 O 2 for 5 min. Cells were washed, incubated with streptavidin horseradish peroxidase (1:500 in PBS) for 30 min at room temperature, and washed again, and staining was developed by addition of the chromogene 3,3Ј-diaminobenzidine tetrahydrochloride and H 2 O 2 for 20 min. By using this procedure, apoptotic nuclei were stained dark brown. Negative cells were counter-stained with toluidine blue, dehydrated through alcohol to xylene, and mounted in distyrene plasticizer in xylene and viewed under ϫ40 magnification.
Statistical Analysis-Data are expressed as the mean Ϯ S.E. A one-way analysis of variance (ANOVA) was performed to determine whether there were significant differences between conditions. When this analysis indicated significance (at the 0.05 level), post hoc Student's Newmann-Keuls test analysis was used to determine which conditions were significantly different from each other. Fig. 1 demonstrates that LTP was impaired in rats that were exposed to irradiation; the mean percentage change (Ϯ S.E.) in excitatory postsynaptic potential slope in the 2 min immediately following tetanic stimulation (compared with that in the 5 min immediately prior to stimulation) was 127.99% (Ϯ 3.61), which was significantly decreased compared with the corresponding value in sham-irradiated rats, 148.18% (Ϯ7.31; p Ͻ 0.05; ANOVA; n ϭ 6 in both groups). The mean percentage changes in the last 5 min of the experiment were 134.17 (Ϯ3.60) and 98.82% (Ϯ12.51) in sham-irradiated and irradiated rats, respectively (p Ͻ 0.001; n ϭ 6 in both groups). The mean percentage changes in the 2 min immediately following tetanic stimulation and in the last 5 min of the experiment were 155.83 (Ϯ6.23) and 120.45% (Ϯ1.14) in sham-irradiated rats that were treated with EPA for 4 weeks prior to analysis of LTP; the corresponding values in irradiated rats that were treated with EPA were 146.22 (Ϯ5.01) and 119.38% (Ϯ1.16). Although irradiation resulted in a significant attenuation of LTP (p Ͻ 0.001; ANOVA), the inhibitory effect of irradiation was reversed by treatment with EPA so that there was a significant difference in the response to tetanic stimulation in control-treated and EPA-treated irradiated rats (p Ͻ 0.001; ANOVA). However, it should be noted that the degree of LTP sustained by controltreated, sham-irradiated rats was greater than EPA-treated irradiated rats (p Ͻ 0.05; ANOVA).

RESULTS
Mean ROS accumulation was similar in control-treated and EPA-treated sham-irradiated rats but was significantly enhanced in hippocampal tissue prepared from irradiated rats compared with sham-irradiated rats ( Fig. 2a; p Ͻ 0.05; ANOVA). Treatment with EPA reversed the irradiation-induced change, so that this value was similar to the values in tissue prepared from sham-irradiated rats.
The irradiation-stimulated changes in ROS accumulation were mirrored by the changes in JNK activation. Thus JNK activation was similar in hippocampal tissue prepared from control-treated and EPA-treated sham-irradiated rats (Fig.  2b), and although it was significantly increased in tissue prepared from control-treated, irradiated rats (p Ͻ 0.05; ANOVA), this increase in JNK activation was prevented in EPA-treated rats. The sample immunoblot illustrates the irradiation-induced increase in JNK activation (lane 3) compared with the tissue prepared from the EPA-treated irradiated (lane 4) or sham-irradiated (lane 2) rats or from control-treated shamirradiated rats (lane 1). JNK activation has been shown to trigger phosphorylation of c-Jun in a number of cell types, and therefore it was predicted that it would induce a similar change in hippocampus. Fig. 2c shows one sample immunoblot that indicates that c-Jun phosphorylation was enhanced in hippocampal tissue prepared from control-treated irradiated rats (lane 3) compared with tissue prepared from rats in the other treatment groups. Assessments of the mean data obtained from densitometric analysis revealed that the increase in c-Jun phosphorylation induced by irradiation in control-treated rats was significantly different from the mean value in the shamirradiated control group (p Ͻ 0.05; ANOVA). Treatment with EPA suppressed the irradiation-induced increase in c-Jun phosphorylation.
Phosphorylation of c-Jun by JNK has been shown to be modified by JIP, and the current data indicate that although expression of JIP was similar in hippocampal tissue prepared from both sham-irradiated groups of rats (see lanes 1-3; Fig.  2d), it was markedly enhanced in hippocampal tissue prepared from irradiated rats that were treated with EPA (lane 4; Fig.  2d; p Ͻ 0.05; ANOVA compared with the mean values in any of the other three groups).
Cytochrome c translocation was examined by investigating expression of the protein in the cytosol (Fig. 3a). The data indicate that there was an irradiation-induced increase in cytochrome c translocation (compare lane 3 with each of the other lanes), and analysis of mean data indicates that this increase was significantly different from that in any of the other groups (p Ͻ 0.01; ANOVA). Caspase-3 activation is one consequence of an increase in cytochrome c expression in cytosol, and the data presented in Fig. 3b indicate that activity of caspase-3 paralleled the changes in cytochrome c. Thus enzyme activity was similar in both treatment groups of sham-irradiated rats, but significantly enhanced in hippocampal tissue prepared from control-treated irradiated rats (p Ͻ 0.05; ANOVA). EPA treatment prevented the irradiation-induced increase so that mean caspase-3 activity was similar in this group and the shamirradiated groups of rats.

FIG. 1. Tetanic stimulation led to induction and maintenance of LTP in both groups of sham-irradiated rats.
Whereas LTP was markedly attenuated in irradiated rats that received control treatment, rats that were treated with EPA successfully expressed LTP. Values are expressed as the mean percentage change in excitatory postsynaptic potential (Epsp) slope (Ϯ S.E.; n ϭ 5 or 6) relative to the slope in the 5 min preceding tetanic stimulation.
PARP cleavage was increased with irradiation as reflected by the decreased expression of the 116-kDa fragment of the protein (Fig. 4a); this decrease is illustrated in the sample immunoblot (lane 3) and is reflected in the mean data obtained from densitometric analysis, which revealed that the irradiated induced increase in PARP cleavage was significantly different from that in hippocampal tissue prepared from control-treated (lane 1) or EPA-treated (lane 2) sham-irradiated rats (p Ͻ 0.01; ANOVA). EPA treatment suppressed the irradiation-induced change (lane 4).
Cytospun cells prepared from hippocampus of each of the treatment groups were assessed for TUNEL staining. Fig. 4b indicates that the percentage of TUNEL-positive cells relative to the control was significantly greater in hippocampal preparations obtained from control-treated irradiated rats compared with sham-irradiated rats (p Ͻ 0.001; ANOVA); EPA treatment prevented this change.
The data are consistent with the idea that the increase in ROS accumulation stimulated by irradiation triggered the cascade of cellular events leading to cell death, and the effect was mediated by JNK phosphorylation. Fig. 5a indicates that H 2 O 2 increased JNK activation (p Ͻ 0.05; ANOVA) and that VIP, which has been shown to inhibit JNK activation in certain cells, also abrogated H 2 O 2 induced JNK activation in hippocampus. Fig. 5b shows that H 2 O 2 increased caspase-3 activity in hippocampus (p Ͻ 0.05; ANOVA) and that inhibition of JNK by VIP prevented this effect, suggesting that increased ROS accumulation by stimulating JNK can lead to cell death. It seems reasonable to propose that the impairment in LTP is a consequence of the irradiation-induced cell death in hippocampus and that the key event that leads to cell death is ROS accumulation. DISCUSSION We set out (a) to establish the effects of irradiation on LTP in perforant path-granule cell synapses in vivo, (b) to identify the cellular events stimulated by exposure to irradiation that might underlie the impairment in LTP, and (c) to assess the effectiveness of EPA treatment in reversing irradiation-induced changes.
We report that both the early and later responses of perforant path-granule cell synapses to tetanic stimulation were markedly attenuated in rats that were exposed to irradiation, supporting the previous finding that irradiation inhibits LTP in guinea pig CA1 in vitro (10 -12). Significantly prior treatment with EPA prevented the inhibitory effect of irradiation on LTP; thus EPA-treated rats sustained LTP in a manner that was similar to the sham-irradiated cohort.
In an effort to identify the mechanism by which EPA might exert its beneficial effects, we assessed ROS accumulation in hippocampal tissue prepared from the same rats in which LTP analysis was performed, because ROS accumulation has  1 and 2 in (b and c)). Irradiation significantly increased these measures in control-treated rats (p Ͻ 0.05; ANOVA; n ϭ 6; see lane 3 in b and c), but the stimulatory effect was attenuated in EPA-treated rats (lane 4 in b and c) so that values were similar to those observed in sham-irradiated rats. d, JIP-1 expression was similar in hippocampal tissue prepared from both groups of sham-irradiated rats (compare lanes 1 and 2) and the control-treated irradiated group (lane 3) but was significantly enhanced in tissue prepared from EPA-treated irradiated rats (p Ͻ 0.05; ANOVA; n ϭ 6; see lane 4).
been shown previously to increase following irradiation (37) and because LTP is impaired when ROS accumulation is increased (6,15,38,39). We found that, in parallel with the impairment in LTP, ROS accumulation was significantly enhanced in hippocampal tissue prepared from irradiated rats that did not receive EPA. Data have consistently suggested that an increase in ROS is associated with an attenuation of LTP. For example, H 2 O 2 inhibits LTP in dentate gyrus in vivo (14) and in CA1 in vitro (13). Indeed a negative correlation between ROS concentration in hippocampus and ability of rats to sustain LTP has been described (39). Moreover, overexpression of superoxide dismutase, which leads to formation of H 2 O 2 , is associated with impaired LTP in CA1 in vitro, and LTP is restored in slices prepared from these transgenic mice by addition of catalase (40), which metabolizes H 2 O 2 to O 2 and H 2 O. The mechanism by which irradiation leads to an increase in ROS is not known; however, it has been shown that exposure of thymocytes to irradiation leads to metabolism of arachidonic acid by lipoxygenase (41), and this action can lead to the generation of ROS. It has also been reported that irradiation increases generation of eicosanoids (42) including prostaglandin E 2 (42,43), which is produced by the action of cyclooxygenase on arachidonic acid. The role of eicosanoids in the inflammatory response is well documented (44), and therefore these findings are consistent with the idea that irradiation induces inflammation. There are multiple possible ways in which EPA may act, some of them being related to its known anti-inflammatory actions (30,31). At the molecular level, the effect might be related to inhibition of phospholipases such as phospholipase A 2 (45), inhibition of cyclooxygenases and lipoxygenases (46), reduction of pro-inflammatory cytokine synthesis (47,48), activation of PPAR receptors (49 -51), or inhibition of transcription factor synthesis (52). In addition to this, radiation damages membranes by destroying their polyunsaturated fatty acid components, and the EPA may act in part by providing a substrate for membrane repair (53). Recent data from this laboratory indicated that the age-related decrease in LTP, which was accompanied by an increase in ROS accumulation, was also reversed by EPA treatment (32). In that case, the failure of rats to sustain LTP was also associated with an increase in hippocampal concentration of IL-1␤. The finding that EPA treatment suppressed the age-related increase in IL-1␤ concentration is further evidence of its anti-inflammatory effect and supports previous results obtained in other tissues (see Ref. 54).
Activation of JNK was markedly increased in hippocampal tissue prepared from rats exposed to irradiation, coupling this effect with the increase in ROS accumulation and the decrease in LTP. Both irradiation (16 -18) and ROS (55)(56)(57) enhance JNK activation in a variety of cells; here we demonstrate that the H 2 O 2 -induced increase in JNK activation in hippocampus was attenuated by the nonspecific JNK inhibitor (58) VIP. Consistent with the idea that the primary action of EPA treatment is to inhibit the increase in ROS accumulation stimulated by irradiation is the observation that the downstream increase in JNK activation is also suppressed by EPA treatment. The pattern of activation of the transcription factor, c-Jun, exactly mirrored that of JNK activation in the different treatment groups providing support for the idea that they are activated in tandem by irradiation and protected from activation by EPA. It is of interest that EPA increased JIP expression in hippocampal tissue prepared from rats exposed to irradiation (but not sham-irradiated rats); the mechanism underlying this effect was not probed in this study, but this change represents one factor that may contribute to the inhibitory effect of EPA on c-Jun activation. It is possible that this effect of EPA contributes to its antiinflammatory actions.
One relatively early event in apoptosis is disruption of the mitochondrial transmembrane potential that is associated with cytochrome c translocation from mitochondria to cytosol (59). The finding that cytochrome c translocation is increased in hippocampal tissue prepared from rats that were exposed to ␥-irradiation suggests that irradiation induces mitochondrial dysfunction. It is likely that this is a consequence of the irradiation-induced increase in ROS because in vitro analysis revealed that exposure of tissue to H 2 O 2 increased cytochrome c translocation (32). This finding is supported by previous data which indicated that H 2 O 2 increased cytochrome c translocation in Jurkat T cells within 2 h of treatment (60), and both observations are consistent with the finding that oxidative stress disrupts mitochondrial transmembrane potential (59).
It has been suggested that cytochrome c translocation triggers a cascade of reactions characterized by caspase activation (61)(62), and a key element of this cascade is the interaction of cytochrome c with apoptotic protease-activating factor-1 leading to the activation of caspase-9 (63). One of the consequences of this sequence of events is an increase in activation of caspase-3 (64, 65). Indeed activation of caspase-3 is considered to be a reliable marker of apoptosis, and the finding that lel changes in both measures and that EPA treatment reversed the irradiation-induced changes in both cytochrome c translocation and caspase-3 activation. The data suggest that the increase in caspase-3 activity that occurred in hippocampal tissue prepared from irradiated rats was a consequence of the primary irradiation-stimulated increase in ROS accumulation because enzyme activity was increased in vitro by incubation in the presence of H 2 O 2 . VIP, which blocked H 2 O 2induced increase in JNK, also inhibited the H 2 O 2 -induced increase in caspase-3 activity indicating that the stimulatory effect of ROS on caspase-3 was dependent on JNK activation. These data can reasonably be interpreted as evidence that JNK activation plays a role in inducing apoptosis in hippocampus; findings that favor this hypothesis have been presented previously in several cell types. It is significant that EPA administration that led to increased JIP expression was accompanied by an attenuation in the irradiation-induced activation of JNK and the subsequent apoptosis. The mitochondrial protein, Bcl-2, is one substrate for JNK, and it has been shown that phosphorylation of this protein, which plays a role in maintenance of mitochondrial membrane integrity (and therefore prevents translocation of mitochondrial enzymes like cytochrome c), results in its inactivation (69,70). This represents one mechanism by which JNK activation might lead to apoptosis.
One substrate for caspase-3 is the DNA repair enzyme PARP which, when cleaved, loses its ability to affect repair, and therefore PARP cleavage has also been considered to be a reliable indicator of apoptosis (71,72). We found that PARP cleavage was increased by irradiation but that EPA treatment prevented this effect, probably as a consequence of the upstream effects of the treatment. In parallel with the irradiation-induced increase in PARP cleavage and consistent with the idea that it is an indicator of apoptosis is the finding that TUNEL staining, which is indicative of nicked DNA (a consequence of decreased DNA repair), was enhanced in hippocampal cells prepared from irradiated, compared with sham-irradiated, rats. Evidence from a number of earlier studies has indicated that irradiation induces apoptosis, for example in B lymphocytes (73), whereas it results in marked cell loss in hippocampus (74). EPA treatment inhibited the irradiationinduced increase in TUNEL staining; it is proposed that this action of EPA stems from its ability to prevent the irradiationinduced increase in ROS accumulation.
The data demonstrate that irradiation induces a cascade of events in hippocampus which ultimately leads to cell death; a pivotal element in this cascade is activation of JNK induced by ROS accumulation. It is likely that apoptosis leads to a deterioration in several functions; in this study we assessed the effect on LTP and found that it was markedly impaired by irradiation. A particularly significant finding of the present study is that EPA treatment prevented the irradiation-induced impairment in LTP. The data are consistent with the idea that this results from its ability to suppress the increase in ROS accumulation and the subsequent indicators of apoptosis that are induced by irradiation.