Pii: S0304-3940(01)01961-9

Although hippocampal long-term potentiation (LTP) is generally assumed to be a cellular mechanism of learning and memory, there has not been de®nitive evidence for this hypothesis. In the present study, therefore, we addressed the possible relationship between spatial learning ability and LTP by using rats with bilateral ®mbria±fornix lesions. The animals were tested for spatial performance in spontaneous alternation behaviors with further in vivo investigation of LTP. The behavioral parameters of spatial memory showed a signi®cant correlation with LTP in the dentate gyrus, but we found no evidence for a linkage with LTP in the CA1 region. Thus, LTP in the dentate gyrus may be important for spatial cognitive ability. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Long-term potentiation; Fimbria±fornix; CA1; Dentate gyrus; Learning and memory; Spontaneous alternation behavior; Rat Long-term potentiation (LTP), which is a long-lasting increase in synaptic strength following brief high-frequency stimulation of afferent ®bers in the hippocampus, is the primary experimental model of synaptic plasticity that may underlie learning and memory [12]. Although a wide variety of pharmacologic and genetic studies have implied a linkage of cognitive ability with changes in hippocampal LTP, loss or enhancement of LTP is not universally correlated with learning de®cits or surplus [10,13,16]. This discrepancy may be attributable to simple comparisons between different species of animals that received the same treatment. Indeed, most studies using the same animals con®rmed a signi®cant correlation between task performance and LTP induction [5,15]. However, these studies employed isolated hippocampal slices to assess LTP. In the in vitro conditions, the behaviors of LTP may not naturally re ̄ect the hippocampal physiology. Furthermore, most of previous studies did not address the regional difference in LTP contributions, while several distinct excitatory pathways in the hippocampus display LTP. To ensure a reliable comparison between learning ability and LTP, therefore, the present study assessed in vivo LTP of the dentate gyrus and the CA1 region soon after measuring task performance of rats. The ®mbria/fornix (FF) is one of the principal ®ber tracts in the brain and provides the major afferent and efferent systems for the hippocampus, connecting it with the diencephalon, striatum, basal forebrain, and prefrontal cortex [4]. Bilateral FF lesions in animals are known to produce impairments of spatial working memory [3,14]. Therefore, to enhance a divergence in the ability for spatial performance in the present study, some of the animals received treatment of FF transection before the experiments. The animal experiments were performed according to the Japanese Pharmacological Society guide for the care and use of laboratory animal. Subjects were male Wistar/ST rats (SLC, Shizuoka, Japan) weighing from 210 to 250 g at the time of surgery. The animals were deeply anesthetized by intraperitoneal injection of 50 mg/kg pentobarbital, and then placed in a stereotaxic headholder. FF lesions were made bilaterally by inserting a razor blade (6.0 mm in width) into the brain (5.0 mm in depth at a position 1.1 mm posterior to bregma). The sham operation was performed with the same surgical procedure except for the blade insertion of 1.0 mm in depth. The intact rats did not receive any surgery. The behavior experiment was conducted in a Y-shaped Neuroscience Letters 307 (2001) 159±162 0304-3940/01/$ see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0304-3940(01)01961-9 www.elsevier.com/locate/neulet * Corresponding author. Tel./fax: 181-3-5841-4784. E-mail address: [email protected] (Y. Ikegaya). maze 14 ^ 5 days after the surgery. At this time, body weights of the animals were not different among groups: 303.9 ^ 8.9 g in intact rats, 295.5 ^ 13.1 g in Sham-operated rats, and 279.3 ^ 8.5 g in rats with FF lesions (F…2; 44† ˆ 1:37, P ˆ 0:3: one-way analysis of variance (ANOVA)). The three trough-shaped arms (190 mm in width, 672 mm in length, 240 mm in depth) were separated by angles of 1208. Following a 12-h fast, a rat naive to the maze was placed at the end of one arm of the apparatus and allowed to explore the maze for a period of 10 min or up to 12 entries into the arms. Entry was counted when the hind limbs completely entered the arm. Any three consecutive choices of three different arms were considered as an alternation. The percentage of alternation was determined by dividing the total number of alternations by the total number of choices minus two. The memory component in this task is that the rat must remember which arm is more recently visited in order to alternate. After the behavioral experiments, the rats were allowed free access to pellet chow and water for 60 min. They were then anesthetized with urethane (1 g/kg, i.p.) and a-chloralose (25 mg/kg, i.p.), and ®xed in a stereotaxic frame. To record ®eld excitatory postsynaptic potentials (fEPSP), a tungsten recording electrode was inserted into the dentate molecular layer (3.5 mm posterior, 2.0 mm lateral to bregma) or the CA1 stratum radiatum (3.8 mm posterior, 3.0 mm lateral to bregma), and a bipolar stainless steel stimulating electrode was placed along the perforant pathway (8.1 mm posterior, 4.4 mm lateral to bregma) or the Schaffer collaterals (3.8 mm posterior, 2.2 mm lateral to bregma) [11]. Test stimulation (80-ms duration) was applied at intervals of 30 s, and its intensity was adjusted to produce an fEPSP with a slope that was about 50% of maximum. In order to induce LTP, theta-burst stimulation consisting of ®ve burst-like trains (ten pulses at 200 Hz) at 5 Hz was applied four times every 30 s to the perforant pathway or the Schaffer collaterals. After the electrophysiological experiments, the lesion procedure was validated with Nissl staining and histochemical detection of acetylcholine esterase (AChE) activity. The brains were sagittally cryosectioned at 20-mm thickness. The sections were ®xed with 4% paraformaldehyde and 0.2% picric acid for 24 h. Nissl staining was conducted with 10-min incubation with 0.1% cresyl fast violet. AChE activity was detected by a modi®ed thiocholine method of Patre et al. [6]. We found a successful transection of the FF tract (Fig. 1A,B), which always resulted in severely decreased AChE activity in the hippocampus without obvious changes in other regions such as the striatum, the thalamus, or the cerebral cortex (Fig. 1C,D) whereas FF lesions per se induced no apparent loss of hippocampal neurons (Fig. 1A,B). Intact rats placed in Y-shaped apparatus displayed an excellent alteration ratio in arm choices (Fig. 2A). Similar results were obtained for rats that received sham operations. In rats with FF lesions, however, the alternation behaviors were almost collapsed, the ratio of which was close to a putative chance level of 50% (F…2; 44† ˆ 9:37, Q…3; 44† ˆ 5:85, P , 0:01 vs. intact group: Tukey's test following one-way ANOVA) (Fig. 2A). No difference in total moving distance during the test was found among group (F…2; 44† ˆ 0:86, P ˆ 0:4: one-way ANOVA). Any groups of rats did not display either preference for a particular arm or variance in the number of arm choices (F…2; 44† ˆ 0:41, P ˆ 0:7: two-way ANOVA). These results suggest that the FF lesions impair task performance relevant to spatial working memory without affecting sensorimotor functions. To determine whether LTP is altered in rats with FF lesions, fEPSPs evoked in the perforant pathway-DG synapses were recorded (Fig. 2B). The baseline fEPSP slopes before LTP induction were not signi®cantly different among the groups in vivo (F…2; 44† ˆ 2:56, P ˆ 0:1: oneway ANOVA). When theta-burst stimulation was applied to the perforant pathway of intact rats, fEPSP was abruptly enhanced, and LTP was induced in the all cases tested. Similar results were obtained for sham-operated rats. The average percentage of fEPSP slope at 50±60 min after thetaburst stimulation was 132.7 ^ 6.97% in intact rats and 129.7 ^ 7.08% in sham rats. In rats with FF lesions, fEPSPs were facilitated following the same theta-burst stimulation but gradually returned to baseline by 40 min. The average of fEPSP slope at 50±60 min was 104.2 ^ 7.36%, which was signi®cantly less than that of intact rats (F…2; 22† ˆ 4:70, Q…3; 44† ˆ 3:97, P , 0:05: Tukey's test following oneway ANOVA). Using animals that had not received the K. Nakao et al. / Neuroscience Letters 307 (2001) 159±162 160 Fig. 1. Representative bright-®eld microscopic images of sagittal brain sections with Nissl staining (A,B) and histochemical AChE staining (C,D) in intact rats (A,C) and rats with FF lesions (B,D). Arrowheads in the panel B indicate the lesioned site. Arrowheads in the panel D enclose the hippocampus, which does not show evident AChE activity following the FF lesions.