Persistent Effects of Delayed Treatment with Nefiracetam on the Water Maze Task in Rats with Sustained Cerebral Ischemia

Materials and Methods

Animals.

Male Wistar rats (Charles River Japan Inc., Atsugi, Japan), weighing 180 to 220 g, were maintained in a room with a 12-h light/12-h dark cycle at a temperature of 23 ± 1°C and a humidity of 55 ± 5% throughout the experiment. The animals had free access to food and water according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the Guideline of Experimental Animal Care issued by the Prime Minister's Office of Japan. All efforts were made to minimize suffering of the animals, to reduce the number of animals used, and to use alternatives to in vivo techniques, if available. The study protocol was approved by the Committee of Animal Care and Welfare of Tokyo University of Pharmacy and Life Science.

Microsphere Embolism.

Microsphere-induced cerebral embolism was performed by the method described previously (Miyake et al., 1993) with some modification. In brief, after anesthetization of animals with 35 mg/kg sodium pentobarbital i.p., the right external carotid and pterygopalatine arteries were temporarily occluded with strings. Seven hundred microspheres (47.5 ± 0.5 μm in diameter, NEN-005; PerkinElmer Life Sciences, Boston, MA), suspended in 20% dextran solution, were injected into the right internal carotid artery through a needle connected to a polyethylene catheter (3 French; Atom Co., Tokyo, Japan). The needle was then removed, and the puncture wound was repaired with surgical glue (Aron α A, Sankyo Co., Tokyo, Japan). Finally, the strings occluding the right external carotid and pterygopalatine arteries were released. After the temporal occlusion of the arteries, the blood flow was reestablished within 2 to 3 min to the areas supplied by the right external carotid and pterygopalatine arteries. Sham-operated (sham) rats were prepared in a similar manner by injecting vehicle without microspheres.

Neurological Deficits.

Fifteen hours after the operation, the behavior of the operated rats was scored on the basis of paucity of movement, truncal curvature, and forced circling during locomotion according to the criteria described previously (Miyake et al., 1993). The score of each item (neurological deficit) was rated from 3 to 0 (3, very severe; 2, severe; 1, moderate; 0, little or none). The rats with a total score of 7 to 9 points (ME rats) were used in the present study. The neurological deficits were determined at 10:00 AM every day up to either day 19 or day 3.

Water Maze Test.

The water maze test was performed according to the method described previously (Nagakura et al., 2002a). The test was started on day 7 after the operation. ME and sham animals were tested in the water maze by using a three trials/day regimen. To eliminate rats that could not swim due to microsphere embolism-induced injury, we performed a habituation study by placing the surgically treated rats in a pool with a diameter of 100 cm on day 6 after the operation. There were no animals that could not swim in the habituation test for the first series of experiments. The water maze apparatus (model TARGET/2; Neuroscience Co., Tokyo, Japan) consisted of a circular pool with a diameter of 170 cm, which was filled with water to a 30-cm depth. The water temperature was maintained at 23 ± 1°C. A hidden transparent circular platform with a diameter of 12 cm was placed 1.5 cm below the surface of the water and kept in a constant position in the center of one of the four quadrants of the pool. The animals were released from three randomly assigned start locations (excluding the platform-containing quadrant). When a rat mounted the platform, it was kept there for 30 s. If the rat did not reach the platform, it was transferred onto the platform by hand and kept there for 30 s. Data collection was automated by an online video-tracking device designed to track an object in a field. Escape latency (the time to climb onto the platform) and swimming speed (the distance that the animals swam divided by escape time) were determined for each trial with a behavioral tracing analyzer (BAT-2; Neuroscience Co.). The cutoff time for each trial was set at 180 s. Each trial was performed with an intertrial interval of approximately 1 h.

The animals that showed stroke-like symptoms and the sham animals were subjected to the water maze test on days 7 to 9 with or without nefiracetam treatment to examine the acquisition of spatial memory.

The retention test was performed on day 17 after the operation to determine whether the animals could retain the spatial navigation ability in the hidden platform test even 1 week after cessation of the drug treatment. The regimen and starting point used for this task were the same as those conducted on day 9.

The contraposition test was performed on day 18 to examine the overall cognition and short-term or working memory by assessing the ability of the animals to search for and learn a new location of the platform. The platform was moved from its previous position (the 3rd quadrant of the tank) to the opposite position (the first quadrant), and the hidden platform test was carried out by releasing the animals from the starting point of either the 2nd or 4th quadrant alternatively.

On day 19, the visible platform test was performed to examine the ability of spatial navigation of the operated animals. The platform having a striped black-and-white flag was set at 1 cm above the water surface.

Brain Membrane and Cytosolic Preparations.

To determine the protein levels of AC-I and PKA subunits, we prepared membrane fractions from animals sacrificed at 15 h after microsphere embolism (before administration of the drug), at day 3, and at day 19 after the visible platform test. The isolated brain regions were homogenized in buffer A (20 mM HEPES, pH 7.4, 0.25 M sucrose, 0.3 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 1 mM EGTA, and 1 mM MgCl₂). The homogenates were centrifuged at 1000g for 10 min. The supernatant fluid was collected and centrifuged at 100,000g for 20 min to pellet the membrane fraction. The pellets were then resuspended in the same buffer as described above and stored at −80°C until used. The supernatant obtained after the second centrifugation was used as the cytosolic fraction. Determination of protein concentrations was conducted with a protein assay kit according to the method of Bradford (1976).

Western Immunoblot Analysis.

Samples containing 5 to 50 μg of protein were separated on an 8 or 10% polyacrylamide gel and transferred to a polyvinylidene difluoride membrane (Immobilon polyvinylidene difluoride; Millipore Corporation, Bedford, MA). The following primary antibodies were used: anti-AC-I antibody (1:1000; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and anti-PKA RIIα, Cα, or Cβ antibodies (1:1000; Santa Cruz Biotechnology, Inc.). After incubation with the appropriate corresponding secondary antibodies, the blots were developed by the enhanced chemiluminescence detection method (Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). Quantification of the immunoreactive bands was performed by using an Image Analysis System (NIH 1.61). Care was taken to ensure that bands to be quantified were in the linear range of response. To minimize between-blot variability, we applied an aliquot of pooled “control” membranes to one lane of every gel and calculated the immunolabeling of samples relative to this standard. Immunoblotting of AC-I, PKA RIIα, PKA Cα, and PKA Cβ was also performed after preabsorption of these antibodies with their respective synthetic complementary peptides according to the method of Yamamoto et al. (1996). These synthetic peptides completely abolished each corresponding band (data not shown).

Experimental Protocol and Delayed Treatment with Nefiracetam.

The experimental protocol used in the present study is shown in Fig. 1. Microspheres were injected into rats on day 0. After assessment of stroke-like symptoms of microsphere-injected rats, the ME or sham-operated animals were randomly divided into four groups, i.e., nefiracetam-treated and untreated sham groups and nefiracetam-treated and untreated ME groups. Randomization of the operated animals was performed according to the result of the toss of a coin. Nefiracetam (10 mg/kg/day) was administered orally from 15 h after the operation, once daily to the agent-treated group, whereas vehicle (0.5% carboxyl methylcellulose) was administered to the untreated group. Drug treatment on each day was completed by 2 h before either the water maze test for assessment of learning and memory function or tissue sampling for immunochemical examination. This treatment was conducted until day 9 in the first series of experiments or until day 3 in the second series. The dose used in the present study, 10 mg/kg/day, was based on the data of Hara and Ogawa (1990) and those obtained in our preliminary study: treatment with 3 mg/kg/day nefiracetam shortened the escape latency to a lesser degree than that with 10 mg/kg/day.

The water maze test was performed from days 7 to 9 after the operation. The daily administration of nefiracetam was stopped after day 9. On day 17, the retention test was performed and on day 18, the contraposition test was conducted. On day 19, the animals were subjected to the visible water maze test and then their brain samples were isolated for determination of AC-I and PKA proteins.

In another set of experiments, the ME and sham animals were subjected to determination of AC-I and PKA proteins at 15 h (before administration of the agent) and day 3 after the operation. The examination at 15 h after the operation was designed to determine the profile of neurochemical variables before treatment and that on day 3 after the operation, to determine whether nefiracetam might have an effect on neurochemical parameters in the ischemic brain. Our previous studies showed that cerebral blood flow and metabolic parameters of the ME animals were most severely impaired on day 3 (Takeo et al., 1992;Miyake et al., 1993).

Statistical Analysis.

The results were presented as means ± S.E.M. The data of neurochemical variables of nefiracetam-untreated sham-operated (S), nefiracetam-treated sham-operated (SN), nefiracetam-untreated microsphere-embolized (ME), and nefiracetam-treated microsphere-embolized (MN) animals were evaluated by using two-way analysis of variance (ANOVA) followed by post hoc Fisher's protected least significant difference (PLSD)t test. The escape latency of the water maze task was analyzed by using two-way ANOVA for repeated measures followed by post hoc Fisher's PLSD. Differences with a probability of 5% or less were considered to be significant (P < 0.05).

Results

Operation

In the present study, microspheres were injected into 45 rats. Thirty-four of the surviving rats (76%) showed stroke-like symptoms with a total score of 7 to 9 points, and five animals (11%) showed the symptoms of stroke with the score of less than 7 points. Six rats (13%) died before all examinations were completed. Fifteen ME rats (seven for the ME group and eight for the MN group) and 14 sham rats (seven for the S group and seven for the SN group) were used in the first series of experiments for the water maze test. Among the eight MN rats, one animal was eliminated due to a failure in reaching onto the platform in the visible platform test. Nineteen ME rats (seven for ME and seven for MN groups in the day 3 experiments, and five for ME in the 15-h experiment) and 19 sham rats were used in the second series of experiments for examination of changes in proteins involved in the signal transduction. None of the sham rats (14 in the first series of experiments and 19 in the second series of experiments) died during the experiments.

Neurological Deficits

Changes in neurological deficits of the ME and MN rats are depicted in Fig. 2. The neurological deficits gradually decreased starting 3 days after the operation. No significant improvement in the nefiracetam-treated ME animals was seen when these animals were compared with the untreated ME ones. Sham animals, regardless of treatment with the agent or not, did not reveal any neurological deficits throughout the experiment.

Effects of Nefiracetam on the Water Maze Task

Learning and Memory Function (Acquisition).

In the first series of experiments, the ME, MN, S, and SN animals were subjected to the water maze test (acquisition) on days 7 to 9 after the operation (Fig. 3). There were significant differences in the escape latency by groups (F_3,24 = 23.0; P < 0.0001; n = 7 each) and by days (F_8,192 = 16.2; P < 0.001; n = 7 each). The groups by day interaction tended to be significant (P = 0.0874). Analysis by Fisher's PLSD showed that the escape latency of the ME rats was significantly lengthened compared with that of the sham-operated rats from the second trial of day 7 to the third trial of day 9 (P < 0.05). The MN rats showed a significant attenuation in the prolongation of the escape latency compared with the ME rat at the third trials of days 8 and 9 (P < 0.05).

The swimming speed (the distance that the animals swam divided by the escape time) in the water maze task was similar in all groups examined (Table 1).

Retention Test.

The retention test was performed on day 17 with the same regimen as that on day 9 (Fig.4). There were significant differences in the escape latency by groups (F_3,24 = 12.4; P < 0.0001; n = 7 each). Fisher's PLSD revealed that the escape latency of the ME rat from the first to third trials in the retention test was significantly lengthened compared with that of the S or SN rats (P < 0.0001). The MN rats showed a significant attenuation in the prolongation of the escape latency compared with the ME rats from the first to third trials (P < 0.05).

Contraposition Test.

On day 18, the water maze task was changed by placing the platform in the contraposition quadrant in the pool (Fig. 4). The statistical analysis by two-way ANOVA revealed a significant difference of the escape latency by groups (F_3,24 = 19.6; P < 0.0001; n = 7 each) and by times (F_3,72 = 16.2; P < 0.01; n = 7 each). Fisher's PLSD showed that the escape latency of the ME rat was significantly lengthened compared with that of the sham rat from the first trial to fourth trial (P < 0.01). The MN rat revealed a significant shortening of the escape latency (P < 0.05) compared with the ME rat at the first, second, and fourth trials.

Visible Platform Test.

On day 19, the visible platform test was performed. There were no significant changes in the swimming speed among these groups (Table 1). One rat that belonged to the MN group did not get onto the platform within 60 s, and thus it was eliminated from the study. The escape latency of the MN rats was slightly, but significantly, lengthened compared with that of the untreated ME animals.

Effects of Nefiracetam on AC and PKA Subunit Proteins

We examined changes in the AC-I and PKA subunits of S, SN, ME, and MN rats at 15 h, day 3, and day 19 after the operation.

On day 19 in the first series of experiments, the animals were sacrificed after the visible platform test, and AC-I in the membrane fraction and PKA subunits in the cytosol fraction of the right cerebral cortex and hippocampus were determined. A significant decrease in the AC-I protein in the cerebral cortex and in the hippocampus of the ME rat was seen (Fig. 5). Furthermore, PKA catalytic subunit Cβ (β cat) and PKA regulatory subunit RIIα (IIα reg) in the cytosol of the right cerebral cortex and hippocampus of the ME animal were decreased (Fig. 6). The delayed treatment with nefiracetam almost completely reversed the decreases in these three variables in both regions of the ME rat. There were no differences in these protein contents between S and SN rats.

In another set of experiments, we determined the AC-I in the membrane fraction and PKA proteins in the cytosolic fraction of the right cerebral cortex and hippocampus of the ME and sham animals at 15 h and day 3 after the operation.

At 15 h after the operation in the second series of experiments, the protein level of AC-I in the membrane fraction of the right hippocampus of the ME rat was decreased significantly and that of the right cerebral cortex tended to be decreased (Fig.7). In contrast, no significant changes in the protein levels of PKA Cα, Cβ, and RIIα were seen.

On day 3, a significant decrease in the AC-I protein level in the membrane fraction (Fig. 8) and one in the PKA Cβ level in the cytosolic fraction of the right cerebral cortex and hippocampus (Fig. 9) were seen in the ME rat. In contrast, no changes in the PKA Cα and PKA RIIα in either brain region of the ME rat were seen. The delayed treatment with nefiracetam partially prevented the decrease in AC-I protein (P < 0.05) and significantly attenuated the decrease in the PKA Cβ subunit in the cytosolic fraction of the right cerebral cortex and hippocampus of the ME rat (P < 0.05).

We also examined these protein levels in the left hemisphere. There were no significant changes in AC-I and PKA subunits in the cerebral cortex and hippocampus of the left hemispheres of the ME and sham rats.

Discussion

In the present study, microsphere embolism induced a failure in spatial memory function of the rats when they were tested on days 7 to 9 after the operation. There are several reports concerning learning and memory dysfunction in the water maze test in ischemic models such as four-vessel ligation (Hodges, 1996; Imanishi et al., 1997) and middle cerebral artery occlusion (Yonemori et al., 1999; Modo et al., 2000). Although the degree of the severity in learning and memory dysfunction would not compare simply with that of other models, the learning and memory dysfunction of the ME animal seems to be no less, or rather more, long-lasting than that of the ischemia/reperfusion models mentioned above, probably due to permanent embolism in the ME animal.

The delayed treatment with nefiracetam reduced the failure in the spatial memory function in the water maze test, suggesting that this drug is capable of improving learning and memory function impaired by sustained cerebral ischemia. We conducted the retention test on day 17 after the operation to examine whether the effect of treatment may persist after cessation of the drug treatment. Notably, nefiracetam-treated ME rats revealed a significant shortening of the escape latency in the retention test. This finding suggests that this drug is capable of exerting a persistent effect on learning and memory function or of preventing the loss of spatial navigation ability due to ischemic damage in animals with sustained cerebral ischemia. Furthermore, at the first trial in the contraposition test on day 18 where the platform had been moved to the opposite quadrant, nefiracetam shortened the escape latency of the ME animal, suggesting that the overall process, including search strategy, of the drug-treated animal may be superior to that of the untreated animal. The shortened escape latency at the fourth trial in the contraposition test predicted better acquisition of short-term memory in the nefiracetam-treated ME animal.

The question arises as to the nature of the mechanism underlying the nefiracetam-mediated improvement of learning and memory function by nefiracetam in ischemia-induced amnesic animals. In previous studies, we reported the possible involvement of the AC/cAMP/PKA signal transduction pathway in the learning and memory dysfunction of the ME animal (Nagakura et al., 2002b,c). Particularly, alterations in AC-I activity and protein well correlated with the impairment of the spatial memory function in the ME animal (Nagakura et al., 2002a). Furthermore, these pathophysiological alterations were attenuated by treatment with a phosphodiesterase III inhibitor, rolipram, which augmented the cellular cAMP content (Nagakura et al., 2002b). Thus, we focused on the role of this system in the nefiracetam-exerted improvement of learning and memory function in our ischemic animals. We found that, among variables examined, only a small, but significant, decrease in the level of AC-I protein in the right hippocampus occurred at 15 h after the microsphere embolism (before administration of the agent). The ME animal on day 3 revealed greater decreases in AC-I and PKA Cβ proteins in the hippocampus and cerebral cortex, whereas neither PKA Cα nor PKA RIIα protein level had decreased. This finding suggests that impairment of the signal transduction system is gradually developed in the hippocampus and cerebral cortex along with time after microsphere embolism. The delayed treatment with nefiracetam for 3 days attenuated these decreases, suggesting that nefiracetam prevented the ME-induced decrease in these protein levels. On day 19, a reduction in PKA RIIα protein, apart from the decreases in AC-I and PKA Cβ proteins, was detected, suggesting additional damage to the PKA regulatory subunit by 19 days. The delayed treatment from day 1 to 9 reversed these decreases almost completely on day 19, when administration of nefiracetam had already been halted. The results suggest that this agent is capable of exerting a persistent effect on the learning and memory function, probably by preventing the ischemia-induced degradation of proteins involved in the AC/cAMP/PKA signal transduction.

One might attribute the decreases in the levels of membranous AC-I protein and cytosolic PKA Cβ, and PKA RIIα proteins to a drop in the number of neurons or of an increase in the proliferation of glial cells seen after cerebral ischemia. However, we cannot simply attribute the decrease in these proteins to ischemia-induced global loss of neurons or gliosis, because no decrease in PKA Cα protein in the cytosolic fraction of the ME animal was detected even 19 days after microsphere embolization. Also, no alterations in other proteins in this signal transduction system, such as AC-V/VI, AC-VIII, and PKA regulatory units were detected under similar experimental conditions in this model (Nagakura et al., 2002b). Thus, whether AC subtypes and PKA subunits after ME alter or not, might depend upon the difference in the vulnerability of each protein to sustained ischemia.

Several investigators postulated a possible connection of the AC/cAMP/PKA signal transduction system with the nuclear CREB in normal and pathophysiological animals (Abel et al., 1997; Bernabeu et al., 1997; Cammarota et al., 2000). Because nefiracetam prevented the microsphere embolism-induced decrease in the levels of AC-I and PKA subunit proteins involved in the signal transduction system even at day 19, nefiracetam is capable of protecting the AC/cAMP/PKA signal transduction system from ischemic damage even in the absence of treatment as long as the therapeutic dose of this agent has been previously administered for a few days.

Several observations, including disruption of long-term memory for fear conditioning and spatial learning in CREB knockout mice (Bourtchuladze et al., 1994) and disruption of long-term spatial memory by the intrahippocampal infusion of antisense oligonucleotides to CREB (Guzowski and McGaugh, 1997), suggest the possible connection of CREB to learning and memory function. We also observed a decrease in phosphorylated CREB protein in the ME rat, which was associated with the spatial memory dysfunction in the water maze task (Nagakura et al., 2002b), and its reversal by treatment with rolipram (Nagakura et al., 2002c). Taken together, the data suggest that nefiracetam-mediated protection of the AC/cAMP/PKA signal transduction and the following CREB-mediated transcription may play an important role in the improvement of the spatial memory function in sustained cerebral ischemic animals.

In summary, nefiracetam improved the spatial memory function of sustained cerebral ischemic animals in association with the prevention of the ischemia-induced damage to the AC/cAMP/PKA signal transduction, although the exact site of action of nefiracetam in this system remains unclear. The present study has proposed the possible mechanism of the effect of nefiracetam on learning and memory function.