- Email: [email protected]
Effects of nonsaponin fraction of red ginseng on learning deficits in aged rats
Physiology & Behavior 82 (2004) 345 – 355
Effects of nonsaponin fraction of red ginseng on learning deficits in aged rats Hiroaki Kurimotoa, Hisao Nishijoa,b, Teruko Uwanoa, Hidetoshi Yamaguchia, Yong-Mei Zhonga, Kazuko Kawanishic, Taketoshi Onoa,b,* a
System ‘Emotional Science’, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Sugitani 2630, Toyama 930-0194, Japan b CREST, JST, Tokyo, Japan c Department of Pharmacognosy, Kobe Pharmaceutical University, Kobe 658-8558, Japan Received 9 October 2003; received in revised form 11 February 2004; accepted 2 April 2004
Abstract Previously we reported that oral application of red ginseng significantly ameliorated learning deficits in aged rats and young rats with hippocampal lesions. In the present study, we investigated the effects of the nonsaponin fraction of red ginseng on learning deficits in aged rats in behavioral studies and those on long-term potentiation (LTP) in the hippocampal CA3 subfield in young rats in electrophysiological studies. In the behavioral studies, three groups of rats [aged rats with and without oral administration of the nonsaponin fraction of red ginseng and young rats] were tested with the three types of spatial-learning task [distance movement task (DMT), random-reward place search task (RRPST), and place-learning task (PLT)] in a circular open field. The results in the DMT and RRPST indicated that motivational and motor activity was not significantly different among the three groups of rats. However, performance of the aged rats without nonsaponin was significantly impaired in the PLT when compared with the young rats. Treatment with nonsaponin significantly ameliorated deficits in place-navigation learning in the aged rats in the PLT. In the electrophysiological studies, effects of nonsaponin on the LTP in the CA3 subfield of the hippocampal slices were investigated in vitro. Pretreatment with nonsaponin significantly augmented the increase in population spike amplitudes in the CA3 subfield after LTP induction. These results suggest that the nonsaponin fraction of red ginseng contains important substances to improve learning and memory in aged rats and that this amelioration by nonsaponin might be attributed partly to augmentation of LTP in the CA3 subfield. D 2004 Elsevier Inc. All rights reserved. Keywords: Place learning; Memory; CA3 subfield; LTP; Aged rats; Red ginseng; Nonsaponin
1. Introduction Aged animals do not learn as well as young animals [1 – 6]. Several lines of evidence have implicated the hippocampal formation (HF) as a possible locus for spatial memory deficits in aged animals [2– 5]. The HF is most susceptible to senescence-related pathological changes, such as those in senile dementia and human Alzheimer’s disease [7 –11]. It has been suggested that age-related impairments of HF-dependent forms of memory may be caused by altered synaptic plasticity mechanisms in the HF, including longterm potentiation (LTP) (see a review by Barnes [12]). LTP * Corresponding author. System ‘Emotional Science’, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Sugitani 2630, Toyama 930-0194, Japan. Tel.: +81-76-434-7220; fax: +81-76-4345013. 0031-9384/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2004.04.001
has been proposed as a candidate neural model for memory storage; evidence to support the LTP model of memory storage includes a study in which pharmacological or genetic manipulations that impaired LTP induction also impaired acquisition of HF-dependent behavior [13]. Numerous herbs have been used in traditional Asian medicine to treat neuropsychological disorders associated with aging and brain injury such as stroke. It is reported that some of traditional medicines including red ginseng have significant therapeutic effects in stroke [14]. Prescriptions of herbs including ginseng or ginseng extract alone have significant effects on neurological and psychiatric symptoms in aged humans [15] and on psychomotor functions in healthy subjects [16]. Ginseng is composed of two major fractions, saponin and nonsaponin fraction. It is reported that the saponin fraction attenuated an increase in striatal dopamine induced by amphetamine [17], and that purified
346
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
saponin, such as ginsenosides (Rb1, Rg1, etc.), inhibited the acetylcholine-evoked Na + and Ca2 + influxes into the cells [18]. Furthermore, ginsenosides decreased the NMDA receptor bindings in the parietal cortex, and elevated GABA receptor bindings in almost all regions of the frontal cortex [19]. On the other hand, the nonsaponin fraction is reported to inhibit thromboxane A2 production [20]. It has been reported that the nonsaponin fraction included sugars, nucleic acids, nucleotides, and amino acids [21 –24]. We previously reported that red ginseng ameliorated learning and memory deficits in aged rats and young rats with HF lesions [6]. Some pharmacological and behavioral studies in animals reported that ginseng had stimulating or suppressing effects on the central nervous system [25 – 28]. The saponin fraction including various ginsenosides inhibited neuronal responses to excitatory transmitters or stress [18,29 –31]. These previous results suggest that fractions of red ginseng other than the saponin fraction, i.e., nonsaponin fraction, might have stimulatory effects on learning and cognitive functions as well as LTP. The aim of the present study is to investigate effects of the nonsaponin fraction of red ginseng on learning and memory deficits in aged rats and LTP in the CA3 subfield of the HF.
2. Materials and methods 2.1. Behavioral experiments 2.1.1. Subjects Young (10 – 12 weeks, n = 6) and aged (28 – 32 months, n = 14) male Fisher 344 (Charles River, Japan) rats were used. All rats were given food and water ad libitum in a clear cage and handled on three consecutive days before start of the experiments. The housing area provided a temperature-controlled environment (21 – 23 jC) under a 12/12-h light cycle (on at 0800 h, off at 2000 h). These rats were divided into three groups: aged + nonsaponin (n = 9), aged + water (n = 5), and young + water (n = 6). These rats received surgery for implantation of electrodes for intracranial self-stimulation (ICSS) in the lateral hypothalamic area (LHA) (see Surgery for ICSS). After recovery from the surgery, the rats were tested with the three spatial paradigms (see Behavioral Tests). All experimental protocols were approved by the Animal Care and Use Committee of Toyama Medical Pharmaceutical University and conformed to NIH Guidelines on the Humane Care and Use of Laboratory Animals. 2.1.2. Surgery for ICSS Electrodes for ICSS were implanted in the medial forebrain bundle (MFB) at the level of the LHA under anesthesia (sodium pentobarbital 40 mg/kg ip). The coordinates were 4.2 mm posterior to bregma, 1.2 mm lateral to midline, and 8.5 mm below the cranial bone according to the atlas of Paxinos and Watson [32].
2.1.3. ICSS training After 1 week of recovery from electrode implantation, the rats were screened to self-stimulate in an operant chamber equipped with a lever on one wall. Each lever press triggered the delivery of a 0.5-s train of 0.3-ms negative, square wave pulses at 100 Hz. The current intensity for ICSS was determined to produce 40– 70 lever presses/min in the operant chamber. Any rat for which the threshold exceeded 300 AA was excluded from the study. All rats used in the present study met this criterion. The rats were trained for ICSS for 8 days (30 min/day). 2.1.4. Experimental setup for behavioral testing For the three spatial tasks, an open field (150-cm diameter) was used. The open field was enclosed by a black curtain. A charge-coupled device (CCD) video camera viewed the open field containing a rat from top center to signal the rat’s position in Cartesian coordinates (Fig. 1Aa). The video signal was sent to a conventional TV monitor and a digital interface, which sent the X and Y coordinates of a miniature light bulb attached to the head of the rat through an RS-232C serial port to a microcomputer (Color Tracking System, MATYS). The microcomputer plotted the trail of the rat, compared the rat’s behavior with preset criteria, and gated ICSS delivery as reward to the MFB at the level of the LHA from a stimulator when the criteria were met. A schematic diagram of the top view of cues in an open field is shown in Fig. 1Ab. Under usual conditions, an electric bulb at the 3 o’clock position and a speaker at the 9 o’clock position on the ceiling of the enclosure were turned on. 2.1.5. Behavioral tests After ICSS training, the rats were first tested in the open field to measure spontaneous locomotor activity, then tested in the three kinds of spatial tasks in the open field. In each of the following three spatial tasks, the small electric bulb on the head of the rat was turned on at the start of a trial, and a train of ICSS current was delivered to activate the rat. Each trial was terminated after 50 rewards had been delivered or 10 min had elapsed, whichever occurred first. Each session consisted of three trials, and the rats were given training of one session/day. 2.1.5.1. Spontaneous activity. Spontaneous locomotor activity was measured in the open field on the first training day in the distance movement task (DMT). A rat was put on the center of the open field. Spontaneous locomotor activity of the rat was monitored by the CCD camera, and distance traveled by the rat for 10 min was computed. Then, the rat was returned to its home cage for 5 min and subsequently trained with the DMT. From the second training day, the rat was trained only with the DMT. 2.1.5.2. Distance movement task. A rat learned to move in the open field on this task (Fig. 1Ba). A computer program computed the cumulative distance traveled by the rat from
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
347
when it entered the reward place, which was then made inactive (changed to thin line circle). After a 5-s interval, the reward place was moved to a different location and reactivated. The rat was trained until it could move more than 2500 cm/trial at least in two trials/session on three successive sessions.
Fig. 1. Scheme of experimental setup (A) and behavioral paradigm (B). (A) Schematic illustration of the experimental setup (a). An open field (150-cm diameter) containing a rat was viewed from top center by a CCD camera with brightness-tracking interface that signaled the rat’s position in Cartesian coordinates. The open field was enclosed by a black curtain. The video signal was directly sent to a conventional TV monitor and the tracking interface sent the X and Y coordinates of a miniature electric light bulb attached to the head of the rat every 33 ms through an RS-232C serial port to a microcomputer. The microcomputer plotted the trail of the rat and gated ICSS delivery as reward from a stimulator. Schematic diagram of top view of cues in an open field (b). Under usual conditions, the incandescent lamp at the 3 o’clock position and the speaker at the 9 o’clock position on the ceiling of the enclosure were turned on. (B) Schema of the three behavioral paradigms. In DMT (a), a computer program computed distance traveled from a trial. A rat acquired ICSS rewards if it moved the fixed distance (i.e.,100, 120, 140, and 160 cm). In RRPST (b), a computer program delimited a circular reward place (small circle, 90-cm diameter) at some randomly selected coordinate. The rat was rewarded with ICSS when it entered the reward place, which was then made inactive. After a 5-s interval, the reward place was moved to a different nonoverlapping location and reactivated. In PLT (c), the rat received rewarding ICSS in two target areas (two small circles; 40-cm diameter) when it returned to one reward place after a visit to another reward place. Large circles, open field; S, location of the rat at start of session; line, locomotion trial of the rat.
the trail. The rat acquired ICSS rewards if it moved the fixed distance (i.e., 100, 120, 140, 160 cm). The predetermined fixed distance to acquire ICSS reward was progressively increased by 20 cm from 100 cm to 160 cm when the rat acquired 50 ICSS rewards within 10 min of one trial. If the rat could not get 50 rewards, the trial was also terminated at the end of the 10 min. If the rat could not acquire 50 rewards within consecutive three sessions, the distance was also increased by 20 cm from the fourth session. When the rat passed the criterion in the DMT (i.e., distance of 160 cm), training in the DMT was completed. Thus, rats learned association of ICSS rewards with movements in the DMT. 2.1.5.3. Random-reward place search task (RRPST). In this protocol, rats must learn to navigate randomly in the open field. A computer program delimited a circular reward place (90 cm diameter, thick line circle); its center was chosen at random within a square circumscribed around the open field (Fig. 1Bb). The rat was rewarded with ICSS
2.1.5.4. Place learning task (PLT). Two 40-cm-diameter reward places were located diametrically opposite to one another in the open field. The rat was rewarded in both reward places, when it returned to one of them after a visit to the other one (Fig. 1Bc). So in this task the rat was required to learn and memorize the places where ICSS rewards were delivered. When the rat acquired 50 rewards within 10 min, the trial was terminated, and if the rat could not acquire 50 rewards, the trial was also terminated at the end of 10 min. The rats were trained for 21 days within this protocol. 2.1.6. Administration of the nonsaponin fraction The nonsaponin fraction of red ginseng powder (Ginseng radix rubra, Seikansho, Korea Tabacco & Ginseng), which was extracted from red ginseng powder with water and freeze-dried, was used. The nonsaponin fraction was suspended in distilled water (10 mg/ml). Suspension of the nonsaponin fraction (50 mg/kg/day) (aged + nonsaponin group) or water (5 ml/kg/day) (aged + water and young + water groups) was administered orally 2 h before the training/behavioral testing, and daily after the fourth day of ICSS training. 2.1.7. Data analysis A personal computer (PC-98, NEC) was programmed to continuously monitor (1) the number of lever presses in the operant chamber, (2) spontaneous locomotor activity in the open field, and (3) the number of ICSS rewards, distance traveled, and duration of each trial in the three spatial tasks. Each data set obtained in the tests other than the PLT (i.e., number of bar presses for ICSS, spontaneous locomotor activity, number of training sessions for rats to pass the criteria in the DMT and RRPST, and number of rewards and distance traveled in the RRPST) was analyzed by one-way analysis of variance (ANOVA) and a post hoc test (Newman –Keuls test). The number of rewards and distance traveled in the PLT were averaged in each rat in each experimental day (session). Then, the grand mean of each parameter in each group was computed in each experimental day. These data sets were analyzed by twoway ANOVA (Group Day) and a post hoc test (Newman –Keuls test). The significant level for statistical analysis was P < .05. 2.2. Electrophysiological experiments 2.2.1. Slice preparation Young (4– 6 weeks) Fisher 344 rats (n = 20) were used. The brain slices were prepared according to methods
348
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
previously developed [33 –36]. The rat was decapitated under ether anesthesia, and the brain was rapidly removed and chilled in an oxygenated (95% O2 –5% CO2) incubation medium [artificial cerebrospinal fluid (ACSF)] at 4 jC for about 1 min. The ACSF included the following composition (in mM): 124 NaCl, 5.00 KCl, 1.24 KH2PO4, 1.30 MgSO4, 2.60 CaCl2, 26.0 NaHCO3, and 10 glucose. The pH of this ACSF was adjusted at 7.4. The HF slices (400 Am thick) were prepared using a micro slicer (DTK1000, Dosaka EM, Kyoto, Japan), and preincubated in oxygenated ACSF at 25 jC in a holding chamber for at least 1 h. One slice was then transferred to a submerged recording chamber that was controlled at 32.0 F 1.0 jC by a temperature controller (Biowarmer, Narishige, Tokyo, Japan), and continuously perfused with the medium at a rate of 2.0– 3.0 ml/min. Nonsaponin fraction was added to the ACSF (0.4 – 12 Ag/ml), which was perfused continuously for 10 min from 20 to 30 min before tetanic stimulation. 2.2.2. Extracellular recording and LTP-induction protocols A standard, saline-filled, glass microelectrode (3 –8 MV) was placed in the pyramidal cell layer of the CA3 subfield to record extracellularly population spike. Bipolar stimulating electrodes were placed in the granular cell layer of the dentate gyrus. A schema of an HF slice with both stimulating and recording electrodes is shown in Fig. 2. The test electrical stimuli (0.2 –0.5 Hz, duration 0.1 ms), intensity of which was adjusted to give 30 – 50% of maximal population spike amplitudes, were applied throughout all the experiments. Eight successive population spikes were averaged and recorded every 10 min. Tetanic stimulation was applied at 100 Hz for 1 s using the same stimulation electrode in the dentate gyrus 20 min after cessation of perfusion with nonsaponin fraction. The population spikes were recorded for 20 min before and for 210 min after start of perfusion with nonsaponin fraction.
Fig. 2. Schematic representation of the electrophysiological experiment. A recording electrode was placed in the pyramidal layer of the CA3 region to record population spike. A stimulating electrode was placed in the granule cell layer of the dentate gyrus to stimulate the mossy fibers.
2.2.3. Data analysis The amplitude of the population spike was measured as voltage from negative to positive peaks. The amplitudes of the population spikes were normalized as the percent amplitudes, relative to the baseline averaged population spike amplitudes recorded during 20 min before perfusion with nonsaponin fraction. The magnitude of LTP in each preparation was also expressed as percent increase in population spike amplitude relative to the averaged baseline responses before perfusion with nonsaponin fraction. These data were analyzed by two-way ANOVA (Group Time) and a post hoc test (Newman – Keuls test). The significant level for statistical analysis was set at P < .05.
3. Results 3.1. Behavioral experiment 3.1.1. ICSS training Fig. 3Aa and Ab shows the mean current intensity and mean number of bar presses for ICSS on the last training day of ICSS in the operant chamber, respectively. There were no significant differences in the mean current intensity [one-way ANOVA: F(2,17) = 0.889, P>.05] nor in the mean number of bar presses among these three groups of the rats [one-way ANOVA: F(2,17) = 1.775, P>.05]. These results suggest that motivational levels of the three groups of animals were similar. 3.1.2. Performance in the DMT and RRPST Fig. 3B shows the spontaneous activity. The spontaneous activity was significantly larger in the aged + water group than the aged + nonsaponin group [Newman – Keuls test ( P < .01) after one-way ANOVA ( P < .05)]. Fig. 3C shows the mean number of training sessions for the rats to reach the criteria in the DMT. There were no significant differences in the mean number of training sessions among these three groups [one-way ANOVA: F(2,17) = 2.536, P>.05]. Fig. 3Da shows the mean number of training sessions for the rats to reach the criteria in the RRPST. There were no significant differences in the training sessions among the three groups of rats [one-way ANOVA: F(2,17) = 2.160, P>.05]. These results indicated that all three groups of the rats similarly learned the simple motor tasks. Fig. 3Db and Dc shows the mean number of ICSS rewards per trial and distance traveled per trial, respectively, in the last session when the rats passed the criteria. There were no significant differences in the number of rewards [one-way ANOVA: F(2,17) = 2.678, P>.05] nor in the distance traveled among the three groups of rats [one-way ANOVA: F(2,17) = 0.734, P>.05]. This evidence further confirmed the finding that there were no significant differences among the three groups of rats in parameters correlated with simple motor/motivational functions.
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
349
Fig. 3. Comparison of parameters of ICSS (A), spontaneous locomotor activity (B), and performance on the DMT (C) and RRPST (D) among the three groups of rats. (A, B) No significant differences in the mean current intensity for ICSS (Aa), and the mean number of bar pressing (Ab) were observed among the three groups of rats (one-way ANOVA, P >.05). However, spontaneous locomotor activity of aged + water rats was larger than that of aged + nonsaponin rats (Newman – Keuls test after one-way ANOVA, P < .01) (B). (C) There were no significant differences in the mean number of training sessions for the rats to pass the criteria in the DMT among the three groups of rats (one-way ANOVA, P >.05). (D) There were no significant differences in the mean number of sessions to pass the criteria in the RRPST (a) and the mean number of rewards (b) and mean distance (c) per trial on the last session of the RRPST (one-way ANOVA, P >.05). Open bars, young + water rats; closed bars, aged + water rats; shaded bars, aged + nonsaponin rats.
3.1.3. Performance in the PLT Fig. 4 illustrates typical examples of trails of rats in the three groups on the 1st, 7th, and 21st day in the PLT. A young + water rat moved randomly on the 1st day (Aa). Training improved navigation performance of the young + water rat; it navigated to display constant and straight trails connecting between the two reward areas on the 7th and 21st day (Ab, c). The trails of an aged + water rat were similarly random on the 1st day (Ba). However, it did not show straight trails connecting between the two reward areas, but navigated along a wall of the open field on the 7th day (Bb). On the 21st day the rat displayed shortcut trails connecting the two reward areas although the trails were not stable (Bc). On the other hand, the trails of an aged + nonsaponin rat were similar to those of the young +
water rat on the 7th and 21st day except that the trails were not stable in the aged + nonsaponin rat (Cb, c). Its trails were intermediate between those of the young + water and aged + water rat. Fig. 5 shows the mean number of rewards acquired (A) and distance traveled (B) per trial by the rats in each group over 21 days navigation learning. In the young + water group, the number of rewards and distance increased rapidly to nearly maximum values from the 1st to 8th and from 1st to fourth day, respectively, and then maintained the values. On the other hand, the number of rewards and distance increased gradually with learning in the aged + nonsaponin group, but more gradually in the aged + water group. Statistical analyses by two-way ANOVA (Group Day) indicated that there were significant main effects among groups
350
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
A. Young+water a. 1st day
b . 7th day ϑ ϑϑ ϑϑϑϑϑϑϑϑ ϑ ϑϑ ϑϑϑ ϑϑ
ϑ ϑϑ ϑϑ
ϑ ϑ ϑ ϑ ϑ
c. 21st day
t=600 s n=10 d=4415 cm
ϑ ϑ ϑϑϑ ϑ ϑϑ ϑϑ ϑϑ ϑϑ ϑϑϑ
ϑ ϑϑϑϑ ϑϑϑ ϑ ϑϑϑϑϑϑϑϑϑϑϑ ϑ ϑ
t=554.4 s n=50 d=8558 cm
ϑ ϑϑϑ ϑϑϑϑϑ ϑϑϑϑ ϑϑ
t=357.4 s n=50 d=8243 cm
Fig. 4. Typical examples of trials of rats of the three groups in the PLT. (A – C) Trials of young + water rat (A), aged + water rat (B), and aged + nonsaponin rat (C) on Days 1 (a), 7 (b), and 21 (c). Note that a young + water rat in Ab and an aged + nonsaponin rat in Cb navigated with shortcut trails between two reward areas, while an aged + water rat in Bb navigated randomly in the open field on the 7th day. t, trial duration (s); n, number of rewards acquired; d, distance traveled (cm).
in both the number of rewards [ F(2,354) = 0.0117, P < .001] and distance [ F(2,354) = 98.959, P < .001]. Post hoc analyses indicated that both the number of rewards and distance were significantly larger in the young + water group than the aged + nonsaponin group (Newman – Keuls test, P < .05), and significantly larger in the aged + nonsaponin group than the aged + water group (Newman – Keuls test, P < .05). When the data between the aged + nonsaponin and aged + water group were compared in each three successive days, the number of rewards was significantly larger in the aged + nonsaponin group than the aged + water group from the second to the last stages [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)], but did not significantly differ in the initial stage of navigation learning (i.e., between first to third) [Newman – Keuls test ( P >.05) after two-way ANOVA ( P < .05)]. On the other hand, distance traveled was significantly larger in the aged + nonsaponin group than the aged + water group through all the stages [Newman– Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. However, both the number of
rewards and the distance traveled did not significantly differ between the young + water and aged + nonsaponin group through all the stages [Newman– Keuls test ( P >.05) after two-way ANOVA ( P < .05)] except the third stage in which the number of rewards was significantly larger in the young + water group than aged + nonsaponin group [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. This pattern of performance differences among the three groups of rats indicated that treatment with the nonsaponin fraction of red ginseng significantly improved performance of the aged rats in the PLT to a degree comparable to that of the young rats. 3.1.4. Electrophysiological experiment Population spikes in the CA3 pyramidal cell layer in response to test electrical stimulation of the mossy fibers were extracellularly recorded (Fig. 2). Amplitude of population spikes in the CA3 pyramidal cell layer reflects synaptic functions between the CA3 pyramidal cells and the mossy fibers. Fig. 6Aa shows raw data traces of
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
351
Fig. 5. Comparison of performance on the PLT among the three groups of rats. (A) Mean number of rewards acquired per trial. (B) Mean distance traveled per trial. Mean values of the three trials in each day (session) for each of the three groups are shown. Filled squares, young + water rats; filled circles, aged + water rats; filled triangles, aged + nonsaponin rats; a – c, significant difference by Newman – Keuls test after two-way ANOVA ( P < .05).
population spikes in the CA3 pyramidal cell layer before tetanic stimulation; single population spike was elicited by single electrical stimulation of the mossy fibers. Ten minutes after tetanic stimulation, population spike amplitude increased to 140% of the baseline amplitude (Fig. 6Ab). Fifty and 90 min after titanic stimulation, it decayed to 108% and 105% of the baseline amplitude, respectively (Fig. 6Ac, d). On the other hand, perfusion with the nonsaponin fraction at a concentration of 4 Ag/ml for 10 min increased the amplitude to 150%, 135%, and 128% of the baseline amplitude 10, 50, and 90 min after tetanic stimulation, respectively (Fig. 6Bb, c, d). Fig. 7 shows mean percent population spike amplitudes before and after tetanic stimulation with the control medium (ACSF) and the three concentrations of nonsaponin fraction. There were no significant main effects of group [two-way ANOVA: F(3,48) = 0.465, P >.05] nor significant interaction between group and time [two-way ANOVA: F(6,48) = 1.342, P >.05] during 20 min before application of nonsaponin fraction. Furthermore, there were no significant main effects of group [two-way ANOVA: F(3,48) = 1.334, P >.05] nor significant interaction between group and time [two-way ANOVA: F(6,48) = 0.564, P >.05] during 20 min between cessation of nonsaponin application and tetanic stimulation. These results indicated that there were no significant differences in percent population spike ampli-
tudes among the four groups during 20 min before application of nonsaponin fraction and during 20 min between cessation of nonsaponin application and tetanic stimulation. However, there was a significant main effect of group during 80 min from 20 to 100 min after tetanic stimulation [two-way ANOVA: F(3,144) = 16.278, P < .05]. Post hoc tests indicated that perfusion with nonsaponin fraction at 4 Ag/ml significantly enhanced the amplitudes of population spike compared with the control medium (ACSF) (Newman –Keuls test, P < .01). When the data were compared in each three successive points after LTP induction, percent amplitudes were also significantly larger in the nonsaponin fraction than the control medium: (1) significantly larger in the nonsaponin fraction at 4 Ag/ml than the control during the three stages from + 20 to + 100 min [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)], (2) significantly larger in the nonsaponin fraction at 0.4 Ag/ml than the control in the first stage from + 20 to + 40 min [Newman –Keuls test ( P < .05) after twoway ANOVA ( P < .05)], and (3) significantly larger in the nonsaponin fraction at 12 Ag/ml than the control in the second stage from + 50 to + 70 min [Newman– Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. However, percent amplitudes did not significantly differ in the fourth [two-way ANOVA: main effect, F(3,48) = 1.292, P >.05;
352
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
Fig. 6. Examples of raw data traces of population spikes before and after LTP induction in the CA3 subfield of the HF. (A, B) Population spikes in response to mossy fiber stimulation were recorded before (a) and 10 (b), 50 (c), and 90 min (d) after tetanic stimulation in control (ACSF) (A) and nonsaponin-treated slices (B). Numbers expressed as percentages indicate normalized amplitudes relative to the mean amplitude during the baseline period for 20 min. Each test stimulus to mossy fiber was applied at each arrow mark. Calibration: vertical, 0.5 mV; horizontal, 10 ms.
interaction, F(6,48) = 0.181, P >.05] and fifth stages [twoway ANOVA: main effect, F(3,48) = 0.371, P >.05; interaction, F(6,48) = 0.168, P >.05] among the four groups.
These results indicated that the nonsaponin fraction selectively enhanced LTP processes, but not usual synaptic transmission.
Fig. 7. Time course of population spike amplitudes in the CA3 subfield of the HF in nontreated and nonsaponin-fraction-treated slices. Ordinate, percent amplitude of population spikes relative to the mean amplitude during the baseline period; abscissa, minutes after tetanic stimulation (TS). Each point represents the mean F S.E.M. The numbers in parentheses indicate those of the tested slices. Nonsaponin fraction was perfused for 10 min from 30 to 20 min, as shown by the filled strip. Open circles, control (ACSF) slices; filled squares, 0.4 Ag/ml nonsaponin-fraction-treated slices; filled triangles, 4 Ag/ml nonsaponinfraction-treated slices; filled diamonds, 12 Ag/ml nonsaponin-fraction-treated slices. (a) Nonsignificant differences among the four groups by two-way ANOVA ( P >.05); (b – d) significant difference by Newman – Keuls test after two-way ANOVA ( P < .05).
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
4. Discussion
353
4.2. Effects of the nonsaponin fraction on LTP in the CA3 subfield
4.1. Effects of nonsaponin fraction on spatial learning The HF plays an important role in spatial learning and memory [37,38]. In previous studies, we showed that there were no differences in behavioral performance in the DTM and RRPST between the rats with and without HF lesions [6,39]. However, the rats with HF lesions displayed significant learning and memory deficits in the PLT [6,39]. These results suggest that cognitive demands to perform the tasks are larger in the PLT than the DMT and RRPST although all these three tasks require similar motor (locomotion) and motivational functions to acquire same rewards. In the present study, the nonsaponin fraction of red ginseng significantly ameliorated place-learning deficits in aged rats in the PLT. It has been reported that ginseng extract activates spontaneous motor activity during the dark period in aged rats [40]. This finding suggests that amelioration of performance decline in aged rats with ginseng extract might be attributed to enhanced motor activity induced by ginseng extract. It is not clear whether the nonsaponin fraction of red ginseng activates spontaneous motor activity. However, the nonsaponin fraction did not increase spontaneous locomotor activity in the aged rats. Furthermore, there were no differences in performance in the DMT and RRPST between the aged + water and aged + nonsaponin rats in the present study. These results suggest that the nonsaponin fraction of red ginseng ameliorated cognitive deficits but not motor activity in aged rats. Administration of ginseng extract has been reported to influence learning and memory functions in animal studies [6,41 –43]. Furthermore, ginsenosides Rb1 and Rg1, main components of the saponin fraction of ginseng, have been reported to improve learning and memory functions in animals [44,45]. It is reported that Rb1 facilitated release of acetylcholine in HF slices [46]. Therefore, it is plausible that ginseng saponin enhances learning and memory functions. However, other studies reported that the saponin fraction including various ginsenosides inhibited neuronal responses to excitatory transmitters or stress [18,29 – 31], suggesting an inhibitory influence of the saponin fraction on learning and memory. In contrast with the saponin fraction, treatment with the nonsaponin fraction ameliorated performance decline in aged rats in the PLT in the present study. We previously reported that red ginseng at a dose of 100 mg/kg ameliorated performance decline in aged rats in the PLT [6]. The amelioration effects of the nonsaponin fraction at a dose of 50 mg/kg in the present study were comparable to or stronger than those of red ginseng at a dose of 100 mg/ kg in the previous study. Since red ginseng powder includes an even percentage of saponin and nonsaponin components (Dr. K. Kawanishi, personal communication), these results suggest that at least the nonsaponin fraction of red ginseng contains important substances to improve learning and memory in aged rats.
The mechanisms required for LTP induction and maintenance in the HF are not fully understood. It is reported that LTP is induced by a rise in presynaptic Ca2 + and requires an increased release of excitatory amino acid from mossy fiber terminals [47]. On the other hand, LTP in the mossy fiber – CA3 pathway was abolished in mice lacking the synaptic vesicle protein Rab3A, supporting the essential role of the presynaptic component in expression of the LTP in this pathway [48]. A recent study reported a central role of presynaptic kainate receptors in LTP induction in the mossy fiber – CA3 synapses in which kainate receptors facilitated glutamate release [49]. Furthermore, A-opioid receptors may play a role in induction of mossy fiber LTP as a result of their actions at the presynaptic elements [50]. Interestingly, activation of presynaptic opioid receptors altered calcium influx at the mossy fiber synapses [51]. Collectively, this evidence suggests that LTP induction in the mossy fiber –CA3 pathway requires activation of various molecular mechanisms in the presynaptic terminals of the mossy fibers to increase glutamate release. Therefore, the nonsaponin fraction of red ginseng might affect specific molecular components in the presynaptic terminals of the mossy fibers. Further studies are necessary to confirm or disprove this idea. 4.3. Relationships between learning improvement and LTP induction The present electrophysiological studies indicated that the nonsaponin fraction of red ginseng significantly augmented the LTP of population spikes in the mossy fiber – CA3 system, without effects on the basal responses recorded before tetanic stimulation. This finding suggests that the nonsaponin fraction acted selectively on the LTP-related synaptic process in the mossy fiber – CA3 system without effects on low frequent, normal neurotransmission. Previous extensive studies suggest that the mossy fiber – CA3 subfield system plays an important role in learning. For example, lesions of the CA3 subfield impaired spatial memory acquisition [52]. In addition, it is reported that LTP in the mossy fiber –CA3 region coincided well with the advance of learning [53]. Aged rats displayed distinct morphological changes in various regions of the brain, especially in the HF [54]. It has been reported that the degree of cognitive impairments was highly correlated with degenerative structural changes in the HF, in particular with the changes in the CA3 pyramidal cells [54]. Therefore, improvement of spatial learning in aged + nonsaponin rats might be attributed to augmentation of LTP in the CA3 subfield by the nonsaponin fraction. However, the nonsaponin fraction was tested using young but not aged rats in the present electrophysiological study. It has been reported that the other nootropic drugs, for example anir-
354
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
acetam [55,56] and FK960 [57,58], augmented LTP in the CA3 subfield of young rats and ameliorated memory impairments in aged rats. These previous and present results strongly suggest that the nonsaponin fraction has similar electrophysiological effects on the HF of the aged rats. Since the HF synaptic functions in aged rats differ from those in young rats [12], further electrophysiological studies using aged rats are necessary to determine whether this suggestion is correct. In conclusion, the present results suggest that the nonsaponin fraction of red ginseng contains important substances to improve learning and memory functions in aged rats, and these amelioration effects of the nonsaponin fraction might be attributed partly to augmentation of LTP in the mossy fiber – CA3 synapses.
Acknowledgements We thank Korea Tabacco & Ginseng, which provided red ginseng. This work was supported partly by the MEXT and JSPS Grants-in-Aid for Scientific Research (15500282, 15029219, 12210009).
References [1] Barnes CA. Memory deficits associated with senescence: neurophysiological and behavioral study in the rat. J Comp Physio Psychol 1979;93:74 – 104. [2] Barnes CA, McNaughton BL. An age comparison of the rates of acquisition and forgetting of spatial information in relation to longterm enhancement of hippocampal synapses. Behav Neurosci 1985; 99:1040 – 8. [3] Gallagher M, Pelleymounter MA. An age-related spatial learning deficit: choline uptake distinguishes ‘‘impaired’’ and ‘‘unimpaired’’ rats. Neurobiol Aging 1988;9:363 – 9. [4] Rapp PR, Amaral DG. Individual differences in the cognitive and neurobiological consequences of normal aging. Trends Neurosci 1992;15:340 – 5. [5] Rapp PR, Gallagher M. Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc Natl Acad Sci U S A 1996;93:9926 – 30. [6] Zhong Y-M, Nishijo H, Uwano T, Tamura R, Kawanishi K, Ono T. Red ginseng ameliorated place navigation deficits in young rats with hippocampal lesions and aged rats. Physiol Behav 2000;69:511 – 25. [7] Carr DB, Goate A, Phild D, Morris JC. Current concepts in the pathogenesis of Alzheimer’s disease. Am J Med 1997;103(3A):3S – 10S. [8] Corey-Bloom J, Sabbagh MN, Bondi MW, Hansen L, Alford MF, Masliah E, et al. Hippocampal sclerosis contributes to dementia in the elderly. Neurology 1997;48:154 – 60. [9] Heinonen O, Soininen H, Sorvari H, Kosunen O, Paljarvi L, Koivisto E, et al. Loss of synaptophysin-like immunoreactivity in the hippocampal formation is an early phenomenon in Alzheimer’s disease. Neuroscience 1995;64:375 – 84. [10] Mizutani T, Kashara M. Degeneration of the intrahippocampal routes of the perforant and alvear pathways in senile dementia of Alzheimer type. Neurosci Lett 1995;184:141 – 4. [11] Ott BR, Saver JL. Unilateral amnesic stroke. Six new cases and a review of the literature. Stroke 1993;24:1033 – 42. [12] Barnes CA. Long-term potentiation and the ageing brain. Philos Trans Royal Soc Lond Series B: Biol Sci 2003;358:765 – 72.
[13] Chen C, Tonegawa S. Molecular genetic analysis of synaptic plasticity, activity-dependent neural development, learning, and memory in the mammalian brain. Annu Rev Neurosci 1997;20:157 – 84. [14] Gong X, Sucher NJ. Stroke therapy in traditional Chinese medicine (TCM): prospects for drug discovery and development. Trends Pharmacol Sci 1999;20:191 – 6. [15] Terasawa K, Shimada Y, Kita T, Yamamoto T, Tosa H, Tanaka N, et al. Choto-san in the treatment of vascular dementia: a double-blind, placebo-controlled study. Phytomedicine 1997;4:15 – 22. [16] D’Angelo L, Grimaldi R, Caravaggi M, Marcoli M, Perucca E, Lecchini S, et al. A double-blind, placebo-controlled clinical study on the effect of a standardized ginseng extract on psychomotor performance on healthy volunteers. J Ethnopharmacol 1986;16:15 – 22. [17] Hallady AK, Yu YL, Palmer J, Oh KW, Wagner GC. Acute and chronic effects of ginseng total saponin and amphetamine on fixedinterval performance in rats. Planta Med 1999;65:162 – 4. [18] Tachikawa E, Kudo K, Kashimoto T, Takahashi E. Effects of ginseng saponins on receptor stimulation-responses. Nippon Yakurigaku Zasshi 1997;110:126 – 31. [19] Kim HS, Hwang SL, Nah SY, Oh S. Changes of [3H]MK-801, [3H]muscimol and [3H]flunitrazepam binding in rat brain by the prolonged ventricular infusion of ginsenoside Rc and Rg1. Pharmacol Res 2001;43:473 – 9. [20] Park HJ, Rhee MH, Park KM, Nam KY, Park KH. Effects of nonsaponin fraction from Panax ginseng on cGMP and thromboxane A2 in human platelet aggregation. J Ethnopharmacol 1995;49:157 – 62. [21] Hiyama C, Miyai S, Yoshida H, Yamasaki K, Tanaka O. Application of high-speed liquid chromatography and dual wave-length thin-layer chromatograph-densitometry to analysis of crude drugs: nucleosides and free bases of nucleic acids in ginseng roots. Yakugaku Zasshi 1978;98:1132 – 7. [22] Okuda H. Biological activities non-saponin fractions in Korean red ginseng. In: Proc. ’99 Korean – Japan Ginseng Symposium. Seoul, Korea: The Korean Society of Ginseng; 1999. p. 59 – 69. [23] Park JD, Kim YS, Park KM, Shin HJ, Jeong TC, Lee YH, et al. Biological activities of water soluble nonsaponin constituents from Korean red ginseng (Panax ginseng C.A. Meyer). In: Proc. ’99 Korean – Japan Ginseng Symposium. Seoul, Korea: The Korean Society of Ginseng; 1999. p. 191 – 209. [24] Takaku T, Han KL, Kameda K, Ninomiya H, Okuda H. Production of arginyl – fructosyl – glucose during processing of red ginseng. J Tradit Med 1996;13:118. [25] Petkov V. Effect of ginseng on the brain biogenic monoamines and 3V, 5V-AMP system: experiments on rats. Arzneimittel-Forschung/Drug Res 1978;28:388 – 93. [26] Saito H, Yoshida Y, Takagi K. Effects of Panax ginseng root on exhaustive exercise in mice. Jpn J Pharmacol 1974;24:119 – 27. [27] Saito H, Tsuchiya M, Naka S, Takagi K. Effects of Panax ginseng root on conditioned avoidance response in rats. Jpn J Pharmacol 1977; 27:509 – 16. [28] Zhang Y, Saito H, Nishiyama N. Improving effects of DX-9386, a traditional Chinese medicinal prescription, on thymectomy-induced impairment of learning behaviors in mice. Biol Pharm Bull 1994; 17:1199 – 205. [29] Dou DQ, Zhang YW, Zhang L, Chen YJ, Yao XS. The inhibitory effects of ginsenosides on protein tyrosine kinase activated by hypoxia/reoxygenation in cultured human umbilical vein endothelial cells. Planta Med 2001;67:19 – 23. [30] Kim DH, Jung JS, Suh SO, Min SK, Son BK, Park JH, et al. Inhibition of stress-induced plasma corticosterone levels by ginsenosides in mice: involvement of nitric oxide. Neuroreport 1998;9:2261 – 4. [31] Lee LH, Choi S, Kim JH, Kim JK, Kim JI, Nah SY. Effects of ginsenosides on carbachol-stimulated formation of inositol phosphates in rat cortical cell cultures. Neurochem Res 2003;28:1307 – 13. [32] Paxisons G, Watson C. The rat brain in stereotaxic coordinates. San Diego (CA): Academic Press; 1986. [33] Fukuda M, Ono T, Nishino H, Sasaki K. Independent glucose effects
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
[34]
[35] [36]
[37] [38] [39]
[40]
[41]
[42]
[43] [44]
[45]
[46]
on rat hypothalamic neurons: an in vitro study. J Auton Nerv Syst 1984;10:373 – 81. Kimura T, Tamura R, Kurimoto H, Ono T. Effects of T-588, a newly synthesized cognitive enhancer, on hippocampal CA1 neurons in rat brain tissue slices. Brain Res 1999;831(1 – 2):175 – 83. Nakamura K, Ono T, Nishijo H, Tamura R. Action of neurotropin on rat hypothalamic neurons in tissue slices. Brain Res Bull 1990;24:811 – 7. Yamamoto C. Activation of hippocampal neurons by mossy fiber stimulation in thin brain sections in vitro. Exp Brain Res 1972;14: 423 – 35. Morris RGM, Garrud P, Rawlins JNP, O’Keefe J. Place navigation is impaired in rats with hippocampal lesion. Nature 1982;297:681 – 3. O’Keefe J, Nadel L. The hippocampus as a cognitive map. Oxford (UK): Clarendon Press; 1978. Hori E, Uwano T, Tamura R, Miyake N, Nishijo H, Ono T. Effects of a novel arginine-vasopressin derivative, NC-1900, on the spatial memory impairment of rats with transient forebrain ischemia. Cogn Brain Res 2002;13:1 – 15. Watanabe H, Ohta H, Imamura L, Asakura W, Matoba Y, Matsumoto K. Effect of Panax ginseng on age-related changes in the spontaneous motor activity and dopaminergic nervous system in the rat. Jpn J Pharmacol 1991;55:51 – 6. Nitta H, Matsumoto K, Shimizu M, Ni XH, Watanabe H. Panax ginseng extract improves the performance of aged Fischer 344 rats in radial maze task but not operant brightness discrimination task. Biol Pharm Bull 1995;18:1286 – 8. Nitta H, Matsumoto K, Shimizu M, Ni XH, Watanabe H. Panax ginseng extract improves the scopolamine-induced disruption of 8-arm radial maze performance in rats. Biol Pharm Bull 1995;18:1439 – 42. Zhao R, McDaniel WF. Ginseng improves strategic learning by normal and brain-damaged rats. Neuroreport 1998;9:1619 – 24. Mook-Jung I, Hong HS, Boo JH, Lee KH, Yun SH, Cheong MY, et al. Ginsenoside Rb1 and Rg1 improve spatial learning and increase hippocampal synaptophysin level in mice. J Neurosci Res 2001;63: 509 – 15. Yamaguchi Y, Haruta K, Kobayashi H. Effects of ginsenosides on impaired performance induced in the rat by scopolamine in a radialarm maze. Psychoneuroendocrinology 1995;20:645 – 53. Benishin CG, Lee R, Wang LC, Liu HJ. Effects of ginsenoside Rb1 on central cholinergic metabolism. Pharmacology 1991;42:223 – 9.
355
[47] Maeda T, Kaneko S, Akaike A, Satoh M. Direct evidence for increase in excitatory amino acids release during mossy fiber LTP in rat hippocampal slices as revealed by the patch sensor methods. Neurosci Lett 1997;224:103 – 6. [48] Castillo PE, Janz R, Sudhof TC, Tzounopoulos T, Malenka RC, Nicoll RA. Rab3A is essential for mossy fiber long-term potentiation in the hippocampus. Nature 1997;388:590 – 3. [49] Lauri SE, Bortolotto ZA, Bleakman D, Ornstein PL, Lodge D, Isaac JT, et al. A critical role of a facilitatory presynaptic kainite receptor in mossy fiber LTP. Neuron 2001;32:697 – 709. [50] Derrick BE, Weinberger SB, Martinez Jr JL. Opioid receptors are involved in an NMDA receptor-independent mechanism of LTP induction at hippocampal mossy fiber – CA3 synapses. Brain Res Bull 1991;27:219 – 23. [51] Gannon RL, Terrian DM. U50,488H inhibits dynorphin and glutamate release from guinea pig mossy fiber terminals. Brain Res 1991;548: 242 – 7. [52] Handelmann GE, Olton DS. Spatial memory following damage to hippocampal CA3 pyramidal cells with kainic acid: impairment and recovery with preoperative training. Brain Res 1981;217: 41 – 58. [53] Ishihara K, Mitsuno K, Ishikawa M, Sasa M. Behavioral LTP during learning in rat hippocampal CA3. Behav Brain Res 1997;83:235 – 8. [54] Kadar T, Silbermann M, Brandeis R, Levy A. Age-related structural changes in the rat hippocampus: correlation with working memory deficiency. Brain Res 1990;512:113 – 20. [55] Satoh M, Ishihara K, Iwama T, Takagi H. Aniracetam augments, and midazolam inhibits, the long-term potentiation in guinea-pig hippocampal slices. Neurosci Lett 1986;68:216 – 20. [56] Petkov VD, Mosharrof AH, Petkof VV, Kehayov RA. Age-related differences in memory and in the memory effects of nootropic drugs. Acta Physiol Pharmacol Bulg 1990;16:28 – 36. [57] Matsuoka N, Satoh M. FK960, a novel potential anti-dementia drug, augments long-term potentiation in mossy fiber – CA3 pathway of guinea-pig hippocampal slices. Brain Res 1998;794:248 – 54. [58] Yamazaki M, Matsuoka N, Maeda N, Ohkubo Y, Yamaguchi I. FK960 N-(4-acetyl-1-piperazinyl)-p-fluorobenzamide monohydrate ameliorates the memory deficits in rats through a novel mechanism of action. J Pharmacol Exp Ther 1996;279:1157 – 73.
Effects of nonsaponin fraction of red ginseng on learning deficits in aged rats Hiroaki Kurimotoa, Hisao Nishijoa,b, Teruko Uwanoa, Hidetoshi Yamaguchia, Yong-Mei Zhonga, Kazuko Kawanishic, Taketoshi Onoa,b,* a
System ‘Emotional Science’, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Sugitani 2630, Toyama 930-0194, Japan b CREST, JST, Tokyo, Japan c Department of Pharmacognosy, Kobe Pharmaceutical University, Kobe 658-8558, Japan Received 9 October 2003; received in revised form 11 February 2004; accepted 2 April 2004
Abstract Previously we reported that oral application of red ginseng significantly ameliorated learning deficits in aged rats and young rats with hippocampal lesions. In the present study, we investigated the effects of the nonsaponin fraction of red ginseng on learning deficits in aged rats in behavioral studies and those on long-term potentiation (LTP) in the hippocampal CA3 subfield in young rats in electrophysiological studies. In the behavioral studies, three groups of rats [aged rats with and without oral administration of the nonsaponin fraction of red ginseng and young rats] were tested with the three types of spatial-learning task [distance movement task (DMT), random-reward place search task (RRPST), and place-learning task (PLT)] in a circular open field. The results in the DMT and RRPST indicated that motivational and motor activity was not significantly different among the three groups of rats. However, performance of the aged rats without nonsaponin was significantly impaired in the PLT when compared with the young rats. Treatment with nonsaponin significantly ameliorated deficits in place-navigation learning in the aged rats in the PLT. In the electrophysiological studies, effects of nonsaponin on the LTP in the CA3 subfield of the hippocampal slices were investigated in vitro. Pretreatment with nonsaponin significantly augmented the increase in population spike amplitudes in the CA3 subfield after LTP induction. These results suggest that the nonsaponin fraction of red ginseng contains important substances to improve learning and memory in aged rats and that this amelioration by nonsaponin might be attributed partly to augmentation of LTP in the CA3 subfield. D 2004 Elsevier Inc. All rights reserved. Keywords: Place learning; Memory; CA3 subfield; LTP; Aged rats; Red ginseng; Nonsaponin
1. Introduction Aged animals do not learn as well as young animals [1 – 6]. Several lines of evidence have implicated the hippocampal formation (HF) as a possible locus for spatial memory deficits in aged animals [2– 5]. The HF is most susceptible to senescence-related pathological changes, such as those in senile dementia and human Alzheimer’s disease [7 –11]. It has been suggested that age-related impairments of HF-dependent forms of memory may be caused by altered synaptic plasticity mechanisms in the HF, including longterm potentiation (LTP) (see a review by Barnes [12]). LTP * Corresponding author. System ‘Emotional Science’, Faculty of Medicine, Toyama Medical and Pharmaceutical University, Sugitani 2630, Toyama 930-0194, Japan. Tel.: +81-76-434-7220; fax: +81-76-4345013. 0031-9384/$ – see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2004.04.001
has been proposed as a candidate neural model for memory storage; evidence to support the LTP model of memory storage includes a study in which pharmacological or genetic manipulations that impaired LTP induction also impaired acquisition of HF-dependent behavior [13]. Numerous herbs have been used in traditional Asian medicine to treat neuropsychological disorders associated with aging and brain injury such as stroke. It is reported that some of traditional medicines including red ginseng have significant therapeutic effects in stroke [14]. Prescriptions of herbs including ginseng or ginseng extract alone have significant effects on neurological and psychiatric symptoms in aged humans [15] and on psychomotor functions in healthy subjects [16]. Ginseng is composed of two major fractions, saponin and nonsaponin fraction. It is reported that the saponin fraction attenuated an increase in striatal dopamine induced by amphetamine [17], and that purified
346
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
saponin, such as ginsenosides (Rb1, Rg1, etc.), inhibited the acetylcholine-evoked Na + and Ca2 + influxes into the cells [18]. Furthermore, ginsenosides decreased the NMDA receptor bindings in the parietal cortex, and elevated GABA receptor bindings in almost all regions of the frontal cortex [19]. On the other hand, the nonsaponin fraction is reported to inhibit thromboxane A2 production [20]. It has been reported that the nonsaponin fraction included sugars, nucleic acids, nucleotides, and amino acids [21 –24]. We previously reported that red ginseng ameliorated learning and memory deficits in aged rats and young rats with HF lesions [6]. Some pharmacological and behavioral studies in animals reported that ginseng had stimulating or suppressing effects on the central nervous system [25 – 28]. The saponin fraction including various ginsenosides inhibited neuronal responses to excitatory transmitters or stress [18,29 –31]. These previous results suggest that fractions of red ginseng other than the saponin fraction, i.e., nonsaponin fraction, might have stimulatory effects on learning and cognitive functions as well as LTP. The aim of the present study is to investigate effects of the nonsaponin fraction of red ginseng on learning and memory deficits in aged rats and LTP in the CA3 subfield of the HF.
2. Materials and methods 2.1. Behavioral experiments 2.1.1. Subjects Young (10 – 12 weeks, n = 6) and aged (28 – 32 months, n = 14) male Fisher 344 (Charles River, Japan) rats were used. All rats were given food and water ad libitum in a clear cage and handled on three consecutive days before start of the experiments. The housing area provided a temperature-controlled environment (21 – 23 jC) under a 12/12-h light cycle (on at 0800 h, off at 2000 h). These rats were divided into three groups: aged + nonsaponin (n = 9), aged + water (n = 5), and young + water (n = 6). These rats received surgery for implantation of electrodes for intracranial self-stimulation (ICSS) in the lateral hypothalamic area (LHA) (see Surgery for ICSS). After recovery from the surgery, the rats were tested with the three spatial paradigms (see Behavioral Tests). All experimental protocols were approved by the Animal Care and Use Committee of Toyama Medical Pharmaceutical University and conformed to NIH Guidelines on the Humane Care and Use of Laboratory Animals. 2.1.2. Surgery for ICSS Electrodes for ICSS were implanted in the medial forebrain bundle (MFB) at the level of the LHA under anesthesia (sodium pentobarbital 40 mg/kg ip). The coordinates were 4.2 mm posterior to bregma, 1.2 mm lateral to midline, and 8.5 mm below the cranial bone according to the atlas of Paxinos and Watson [32].
2.1.3. ICSS training After 1 week of recovery from electrode implantation, the rats were screened to self-stimulate in an operant chamber equipped with a lever on one wall. Each lever press triggered the delivery of a 0.5-s train of 0.3-ms negative, square wave pulses at 100 Hz. The current intensity for ICSS was determined to produce 40– 70 lever presses/min in the operant chamber. Any rat for which the threshold exceeded 300 AA was excluded from the study. All rats used in the present study met this criterion. The rats were trained for ICSS for 8 days (30 min/day). 2.1.4. Experimental setup for behavioral testing For the three spatial tasks, an open field (150-cm diameter) was used. The open field was enclosed by a black curtain. A charge-coupled device (CCD) video camera viewed the open field containing a rat from top center to signal the rat’s position in Cartesian coordinates (Fig. 1Aa). The video signal was sent to a conventional TV monitor and a digital interface, which sent the X and Y coordinates of a miniature light bulb attached to the head of the rat through an RS-232C serial port to a microcomputer (Color Tracking System, MATYS). The microcomputer plotted the trail of the rat, compared the rat’s behavior with preset criteria, and gated ICSS delivery as reward to the MFB at the level of the LHA from a stimulator when the criteria were met. A schematic diagram of the top view of cues in an open field is shown in Fig. 1Ab. Under usual conditions, an electric bulb at the 3 o’clock position and a speaker at the 9 o’clock position on the ceiling of the enclosure were turned on. 2.1.5. Behavioral tests After ICSS training, the rats were first tested in the open field to measure spontaneous locomotor activity, then tested in the three kinds of spatial tasks in the open field. In each of the following three spatial tasks, the small electric bulb on the head of the rat was turned on at the start of a trial, and a train of ICSS current was delivered to activate the rat. Each trial was terminated after 50 rewards had been delivered or 10 min had elapsed, whichever occurred first. Each session consisted of three trials, and the rats were given training of one session/day. 2.1.5.1. Spontaneous activity. Spontaneous locomotor activity was measured in the open field on the first training day in the distance movement task (DMT). A rat was put on the center of the open field. Spontaneous locomotor activity of the rat was monitored by the CCD camera, and distance traveled by the rat for 10 min was computed. Then, the rat was returned to its home cage for 5 min and subsequently trained with the DMT. From the second training day, the rat was trained only with the DMT. 2.1.5.2. Distance movement task. A rat learned to move in the open field on this task (Fig. 1Ba). A computer program computed the cumulative distance traveled by the rat from
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
347
when it entered the reward place, which was then made inactive (changed to thin line circle). After a 5-s interval, the reward place was moved to a different location and reactivated. The rat was trained until it could move more than 2500 cm/trial at least in two trials/session on three successive sessions.
Fig. 1. Scheme of experimental setup (A) and behavioral paradigm (B). (A) Schematic illustration of the experimental setup (a). An open field (150-cm diameter) containing a rat was viewed from top center by a CCD camera with brightness-tracking interface that signaled the rat’s position in Cartesian coordinates. The open field was enclosed by a black curtain. The video signal was directly sent to a conventional TV monitor and the tracking interface sent the X and Y coordinates of a miniature electric light bulb attached to the head of the rat every 33 ms through an RS-232C serial port to a microcomputer. The microcomputer plotted the trail of the rat and gated ICSS delivery as reward from a stimulator. Schematic diagram of top view of cues in an open field (b). Under usual conditions, the incandescent lamp at the 3 o’clock position and the speaker at the 9 o’clock position on the ceiling of the enclosure were turned on. (B) Schema of the three behavioral paradigms. In DMT (a), a computer program computed distance traveled from a trial. A rat acquired ICSS rewards if it moved the fixed distance (i.e.,100, 120, 140, and 160 cm). In RRPST (b), a computer program delimited a circular reward place (small circle, 90-cm diameter) at some randomly selected coordinate. The rat was rewarded with ICSS when it entered the reward place, which was then made inactive. After a 5-s interval, the reward place was moved to a different nonoverlapping location and reactivated. In PLT (c), the rat received rewarding ICSS in two target areas (two small circles; 40-cm diameter) when it returned to one reward place after a visit to another reward place. Large circles, open field; S, location of the rat at start of session; line, locomotion trial of the rat.
the trail. The rat acquired ICSS rewards if it moved the fixed distance (i.e., 100, 120, 140, 160 cm). The predetermined fixed distance to acquire ICSS reward was progressively increased by 20 cm from 100 cm to 160 cm when the rat acquired 50 ICSS rewards within 10 min of one trial. If the rat could not get 50 rewards, the trial was also terminated at the end of the 10 min. If the rat could not acquire 50 rewards within consecutive three sessions, the distance was also increased by 20 cm from the fourth session. When the rat passed the criterion in the DMT (i.e., distance of 160 cm), training in the DMT was completed. Thus, rats learned association of ICSS rewards with movements in the DMT. 2.1.5.3. Random-reward place search task (RRPST). In this protocol, rats must learn to navigate randomly in the open field. A computer program delimited a circular reward place (90 cm diameter, thick line circle); its center was chosen at random within a square circumscribed around the open field (Fig. 1Bb). The rat was rewarded with ICSS
2.1.5.4. Place learning task (PLT). Two 40-cm-diameter reward places were located diametrically opposite to one another in the open field. The rat was rewarded in both reward places, when it returned to one of them after a visit to the other one (Fig. 1Bc). So in this task the rat was required to learn and memorize the places where ICSS rewards were delivered. When the rat acquired 50 rewards within 10 min, the trial was terminated, and if the rat could not acquire 50 rewards, the trial was also terminated at the end of 10 min. The rats were trained for 21 days within this protocol. 2.1.6. Administration of the nonsaponin fraction The nonsaponin fraction of red ginseng powder (Ginseng radix rubra, Seikansho, Korea Tabacco & Ginseng), which was extracted from red ginseng powder with water and freeze-dried, was used. The nonsaponin fraction was suspended in distilled water (10 mg/ml). Suspension of the nonsaponin fraction (50 mg/kg/day) (aged + nonsaponin group) or water (5 ml/kg/day) (aged + water and young + water groups) was administered orally 2 h before the training/behavioral testing, and daily after the fourth day of ICSS training. 2.1.7. Data analysis A personal computer (PC-98, NEC) was programmed to continuously monitor (1) the number of lever presses in the operant chamber, (2) spontaneous locomotor activity in the open field, and (3) the number of ICSS rewards, distance traveled, and duration of each trial in the three spatial tasks. Each data set obtained in the tests other than the PLT (i.e., number of bar presses for ICSS, spontaneous locomotor activity, number of training sessions for rats to pass the criteria in the DMT and RRPST, and number of rewards and distance traveled in the RRPST) was analyzed by one-way analysis of variance (ANOVA) and a post hoc test (Newman –Keuls test). The number of rewards and distance traveled in the PLT were averaged in each rat in each experimental day (session). Then, the grand mean of each parameter in each group was computed in each experimental day. These data sets were analyzed by twoway ANOVA (Group Day) and a post hoc test (Newman –Keuls test). The significant level for statistical analysis was P < .05. 2.2. Electrophysiological experiments 2.2.1. Slice preparation Young (4– 6 weeks) Fisher 344 rats (n = 20) were used. The brain slices were prepared according to methods
348
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
previously developed [33 –36]. The rat was decapitated under ether anesthesia, and the brain was rapidly removed and chilled in an oxygenated (95% O2 –5% CO2) incubation medium [artificial cerebrospinal fluid (ACSF)] at 4 jC for about 1 min. The ACSF included the following composition (in mM): 124 NaCl, 5.00 KCl, 1.24 KH2PO4, 1.30 MgSO4, 2.60 CaCl2, 26.0 NaHCO3, and 10 glucose. The pH of this ACSF was adjusted at 7.4. The HF slices (400 Am thick) were prepared using a micro slicer (DTK1000, Dosaka EM, Kyoto, Japan), and preincubated in oxygenated ACSF at 25 jC in a holding chamber for at least 1 h. One slice was then transferred to a submerged recording chamber that was controlled at 32.0 F 1.0 jC by a temperature controller (Biowarmer, Narishige, Tokyo, Japan), and continuously perfused with the medium at a rate of 2.0– 3.0 ml/min. Nonsaponin fraction was added to the ACSF (0.4 – 12 Ag/ml), which was perfused continuously for 10 min from 20 to 30 min before tetanic stimulation. 2.2.2. Extracellular recording and LTP-induction protocols A standard, saline-filled, glass microelectrode (3 –8 MV) was placed in the pyramidal cell layer of the CA3 subfield to record extracellularly population spike. Bipolar stimulating electrodes were placed in the granular cell layer of the dentate gyrus. A schema of an HF slice with both stimulating and recording electrodes is shown in Fig. 2. The test electrical stimuli (0.2 –0.5 Hz, duration 0.1 ms), intensity of which was adjusted to give 30 – 50% of maximal population spike amplitudes, were applied throughout all the experiments. Eight successive population spikes were averaged and recorded every 10 min. Tetanic stimulation was applied at 100 Hz for 1 s using the same stimulation electrode in the dentate gyrus 20 min after cessation of perfusion with nonsaponin fraction. The population spikes were recorded for 20 min before and for 210 min after start of perfusion with nonsaponin fraction.
Fig. 2. Schematic representation of the electrophysiological experiment. A recording electrode was placed in the pyramidal layer of the CA3 region to record population spike. A stimulating electrode was placed in the granule cell layer of the dentate gyrus to stimulate the mossy fibers.
2.2.3. Data analysis The amplitude of the population spike was measured as voltage from negative to positive peaks. The amplitudes of the population spikes were normalized as the percent amplitudes, relative to the baseline averaged population spike amplitudes recorded during 20 min before perfusion with nonsaponin fraction. The magnitude of LTP in each preparation was also expressed as percent increase in population spike amplitude relative to the averaged baseline responses before perfusion with nonsaponin fraction. These data were analyzed by two-way ANOVA (Group Time) and a post hoc test (Newman – Keuls test). The significant level for statistical analysis was set at P < .05.
3. Results 3.1. Behavioral experiment 3.1.1. ICSS training Fig. 3Aa and Ab shows the mean current intensity and mean number of bar presses for ICSS on the last training day of ICSS in the operant chamber, respectively. There were no significant differences in the mean current intensity [one-way ANOVA: F(2,17) = 0.889, P>.05] nor in the mean number of bar presses among these three groups of the rats [one-way ANOVA: F(2,17) = 1.775, P>.05]. These results suggest that motivational levels of the three groups of animals were similar. 3.1.2. Performance in the DMT and RRPST Fig. 3B shows the spontaneous activity. The spontaneous activity was significantly larger in the aged + water group than the aged + nonsaponin group [Newman – Keuls test ( P < .01) after one-way ANOVA ( P < .05)]. Fig. 3C shows the mean number of training sessions for the rats to reach the criteria in the DMT. There were no significant differences in the mean number of training sessions among these three groups [one-way ANOVA: F(2,17) = 2.536, P>.05]. Fig. 3Da shows the mean number of training sessions for the rats to reach the criteria in the RRPST. There were no significant differences in the training sessions among the three groups of rats [one-way ANOVA: F(2,17) = 2.160, P>.05]. These results indicated that all three groups of the rats similarly learned the simple motor tasks. Fig. 3Db and Dc shows the mean number of ICSS rewards per trial and distance traveled per trial, respectively, in the last session when the rats passed the criteria. There were no significant differences in the number of rewards [one-way ANOVA: F(2,17) = 2.678, P>.05] nor in the distance traveled among the three groups of rats [one-way ANOVA: F(2,17) = 0.734, P>.05]. This evidence further confirmed the finding that there were no significant differences among the three groups of rats in parameters correlated with simple motor/motivational functions.
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
349
Fig. 3. Comparison of parameters of ICSS (A), spontaneous locomotor activity (B), and performance on the DMT (C) and RRPST (D) among the three groups of rats. (A, B) No significant differences in the mean current intensity for ICSS (Aa), and the mean number of bar pressing (Ab) were observed among the three groups of rats (one-way ANOVA, P >.05). However, spontaneous locomotor activity of aged + water rats was larger than that of aged + nonsaponin rats (Newman – Keuls test after one-way ANOVA, P < .01) (B). (C) There were no significant differences in the mean number of training sessions for the rats to pass the criteria in the DMT among the three groups of rats (one-way ANOVA, P >.05). (D) There were no significant differences in the mean number of sessions to pass the criteria in the RRPST (a) and the mean number of rewards (b) and mean distance (c) per trial on the last session of the RRPST (one-way ANOVA, P >.05). Open bars, young + water rats; closed bars, aged + water rats; shaded bars, aged + nonsaponin rats.
3.1.3. Performance in the PLT Fig. 4 illustrates typical examples of trails of rats in the three groups on the 1st, 7th, and 21st day in the PLT. A young + water rat moved randomly on the 1st day (Aa). Training improved navigation performance of the young + water rat; it navigated to display constant and straight trails connecting between the two reward areas on the 7th and 21st day (Ab, c). The trails of an aged + water rat were similarly random on the 1st day (Ba). However, it did not show straight trails connecting between the two reward areas, but navigated along a wall of the open field on the 7th day (Bb). On the 21st day the rat displayed shortcut trails connecting the two reward areas although the trails were not stable (Bc). On the other hand, the trails of an aged + nonsaponin rat were similar to those of the young +
water rat on the 7th and 21st day except that the trails were not stable in the aged + nonsaponin rat (Cb, c). Its trails were intermediate between those of the young + water and aged + water rat. Fig. 5 shows the mean number of rewards acquired (A) and distance traveled (B) per trial by the rats in each group over 21 days navigation learning. In the young + water group, the number of rewards and distance increased rapidly to nearly maximum values from the 1st to 8th and from 1st to fourth day, respectively, and then maintained the values. On the other hand, the number of rewards and distance increased gradually with learning in the aged + nonsaponin group, but more gradually in the aged + water group. Statistical analyses by two-way ANOVA (Group Day) indicated that there were significant main effects among groups
350
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
A. Young+water a. 1st day
b . 7th day ϑ ϑϑ ϑϑϑϑϑϑϑϑ ϑ ϑϑ ϑϑϑ ϑϑ
ϑ ϑϑ ϑϑ
ϑ ϑ ϑ ϑ ϑ
c. 21st day
t=600 s n=10 d=4415 cm
ϑ ϑ ϑϑϑ ϑ ϑϑ ϑϑ ϑϑ ϑϑ ϑϑϑ
ϑ ϑϑϑϑ ϑϑϑ ϑ ϑϑϑϑϑϑϑϑϑϑϑ ϑ ϑ
t=554.4 s n=50 d=8558 cm
ϑ ϑϑϑ ϑϑϑϑϑ ϑϑϑϑ ϑϑ
t=357.4 s n=50 d=8243 cm
Fig. 4. Typical examples of trials of rats of the three groups in the PLT. (A – C) Trials of young + water rat (A), aged + water rat (B), and aged + nonsaponin rat (C) on Days 1 (a), 7 (b), and 21 (c). Note that a young + water rat in Ab and an aged + nonsaponin rat in Cb navigated with shortcut trails between two reward areas, while an aged + water rat in Bb navigated randomly in the open field on the 7th day. t, trial duration (s); n, number of rewards acquired; d, distance traveled (cm).
in both the number of rewards [ F(2,354) = 0.0117, P < .001] and distance [ F(2,354) = 98.959, P < .001]. Post hoc analyses indicated that both the number of rewards and distance were significantly larger in the young + water group than the aged + nonsaponin group (Newman – Keuls test, P < .05), and significantly larger in the aged + nonsaponin group than the aged + water group (Newman – Keuls test, P < .05). When the data between the aged + nonsaponin and aged + water group were compared in each three successive days, the number of rewards was significantly larger in the aged + nonsaponin group than the aged + water group from the second to the last stages [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)], but did not significantly differ in the initial stage of navigation learning (i.e., between first to third) [Newman – Keuls test ( P >.05) after two-way ANOVA ( P < .05)]. On the other hand, distance traveled was significantly larger in the aged + nonsaponin group than the aged + water group through all the stages [Newman– Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. However, both the number of
rewards and the distance traveled did not significantly differ between the young + water and aged + nonsaponin group through all the stages [Newman– Keuls test ( P >.05) after two-way ANOVA ( P < .05)] except the third stage in which the number of rewards was significantly larger in the young + water group than aged + nonsaponin group [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. This pattern of performance differences among the three groups of rats indicated that treatment with the nonsaponin fraction of red ginseng significantly improved performance of the aged rats in the PLT to a degree comparable to that of the young rats. 3.1.4. Electrophysiological experiment Population spikes in the CA3 pyramidal cell layer in response to test electrical stimulation of the mossy fibers were extracellularly recorded (Fig. 2). Amplitude of population spikes in the CA3 pyramidal cell layer reflects synaptic functions between the CA3 pyramidal cells and the mossy fibers. Fig. 6Aa shows raw data traces of
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
351
Fig. 5. Comparison of performance on the PLT among the three groups of rats. (A) Mean number of rewards acquired per trial. (B) Mean distance traveled per trial. Mean values of the three trials in each day (session) for each of the three groups are shown. Filled squares, young + water rats; filled circles, aged + water rats; filled triangles, aged + nonsaponin rats; a – c, significant difference by Newman – Keuls test after two-way ANOVA ( P < .05).
population spikes in the CA3 pyramidal cell layer before tetanic stimulation; single population spike was elicited by single electrical stimulation of the mossy fibers. Ten minutes after tetanic stimulation, population spike amplitude increased to 140% of the baseline amplitude (Fig. 6Ab). Fifty and 90 min after titanic stimulation, it decayed to 108% and 105% of the baseline amplitude, respectively (Fig. 6Ac, d). On the other hand, perfusion with the nonsaponin fraction at a concentration of 4 Ag/ml for 10 min increased the amplitude to 150%, 135%, and 128% of the baseline amplitude 10, 50, and 90 min after tetanic stimulation, respectively (Fig. 6Bb, c, d). Fig. 7 shows mean percent population spike amplitudes before and after tetanic stimulation with the control medium (ACSF) and the three concentrations of nonsaponin fraction. There were no significant main effects of group [two-way ANOVA: F(3,48) = 0.465, P >.05] nor significant interaction between group and time [two-way ANOVA: F(6,48) = 1.342, P >.05] during 20 min before application of nonsaponin fraction. Furthermore, there were no significant main effects of group [two-way ANOVA: F(3,48) = 1.334, P >.05] nor significant interaction between group and time [two-way ANOVA: F(6,48) = 0.564, P >.05] during 20 min between cessation of nonsaponin application and tetanic stimulation. These results indicated that there were no significant differences in percent population spike ampli-
tudes among the four groups during 20 min before application of nonsaponin fraction and during 20 min between cessation of nonsaponin application and tetanic stimulation. However, there was a significant main effect of group during 80 min from 20 to 100 min after tetanic stimulation [two-way ANOVA: F(3,144) = 16.278, P < .05]. Post hoc tests indicated that perfusion with nonsaponin fraction at 4 Ag/ml significantly enhanced the amplitudes of population spike compared with the control medium (ACSF) (Newman –Keuls test, P < .01). When the data were compared in each three successive points after LTP induction, percent amplitudes were also significantly larger in the nonsaponin fraction than the control medium: (1) significantly larger in the nonsaponin fraction at 4 Ag/ml than the control during the three stages from + 20 to + 100 min [Newman – Keuls test ( P < .05) after two-way ANOVA ( P < .05)], (2) significantly larger in the nonsaponin fraction at 0.4 Ag/ml than the control in the first stage from + 20 to + 40 min [Newman –Keuls test ( P < .05) after twoway ANOVA ( P < .05)], and (3) significantly larger in the nonsaponin fraction at 12 Ag/ml than the control in the second stage from + 50 to + 70 min [Newman– Keuls test ( P < .05) after two-way ANOVA ( P < .05)]. However, percent amplitudes did not significantly differ in the fourth [two-way ANOVA: main effect, F(3,48) = 1.292, P >.05;
352
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
Fig. 6. Examples of raw data traces of population spikes before and after LTP induction in the CA3 subfield of the HF. (A, B) Population spikes in response to mossy fiber stimulation were recorded before (a) and 10 (b), 50 (c), and 90 min (d) after tetanic stimulation in control (ACSF) (A) and nonsaponin-treated slices (B). Numbers expressed as percentages indicate normalized amplitudes relative to the mean amplitude during the baseline period for 20 min. Each test stimulus to mossy fiber was applied at each arrow mark. Calibration: vertical, 0.5 mV; horizontal, 10 ms.
interaction, F(6,48) = 0.181, P >.05] and fifth stages [twoway ANOVA: main effect, F(3,48) = 0.371, P >.05; interaction, F(6,48) = 0.168, P >.05] among the four groups.
These results indicated that the nonsaponin fraction selectively enhanced LTP processes, but not usual synaptic transmission.
Fig. 7. Time course of population spike amplitudes in the CA3 subfield of the HF in nontreated and nonsaponin-fraction-treated slices. Ordinate, percent amplitude of population spikes relative to the mean amplitude during the baseline period; abscissa, minutes after tetanic stimulation (TS). Each point represents the mean F S.E.M. The numbers in parentheses indicate those of the tested slices. Nonsaponin fraction was perfused for 10 min from 30 to 20 min, as shown by the filled strip. Open circles, control (ACSF) slices; filled squares, 0.4 Ag/ml nonsaponin-fraction-treated slices; filled triangles, 4 Ag/ml nonsaponinfraction-treated slices; filled diamonds, 12 Ag/ml nonsaponin-fraction-treated slices. (a) Nonsignificant differences among the four groups by two-way ANOVA ( P >.05); (b – d) significant difference by Newman – Keuls test after two-way ANOVA ( P < .05).
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
4. Discussion
353
4.2. Effects of the nonsaponin fraction on LTP in the CA3 subfield
4.1. Effects of nonsaponin fraction on spatial learning The HF plays an important role in spatial learning and memory [37,38]. In previous studies, we showed that there were no differences in behavioral performance in the DTM and RRPST between the rats with and without HF lesions [6,39]. However, the rats with HF lesions displayed significant learning and memory deficits in the PLT [6,39]. These results suggest that cognitive demands to perform the tasks are larger in the PLT than the DMT and RRPST although all these three tasks require similar motor (locomotion) and motivational functions to acquire same rewards. In the present study, the nonsaponin fraction of red ginseng significantly ameliorated place-learning deficits in aged rats in the PLT. It has been reported that ginseng extract activates spontaneous motor activity during the dark period in aged rats [40]. This finding suggests that amelioration of performance decline in aged rats with ginseng extract might be attributed to enhanced motor activity induced by ginseng extract. It is not clear whether the nonsaponin fraction of red ginseng activates spontaneous motor activity. However, the nonsaponin fraction did not increase spontaneous locomotor activity in the aged rats. Furthermore, there were no differences in performance in the DMT and RRPST between the aged + water and aged + nonsaponin rats in the present study. These results suggest that the nonsaponin fraction of red ginseng ameliorated cognitive deficits but not motor activity in aged rats. Administration of ginseng extract has been reported to influence learning and memory functions in animal studies [6,41 –43]. Furthermore, ginsenosides Rb1 and Rg1, main components of the saponin fraction of ginseng, have been reported to improve learning and memory functions in animals [44,45]. It is reported that Rb1 facilitated release of acetylcholine in HF slices [46]. Therefore, it is plausible that ginseng saponin enhances learning and memory functions. However, other studies reported that the saponin fraction including various ginsenosides inhibited neuronal responses to excitatory transmitters or stress [18,29 – 31], suggesting an inhibitory influence of the saponin fraction on learning and memory. In contrast with the saponin fraction, treatment with the nonsaponin fraction ameliorated performance decline in aged rats in the PLT in the present study. We previously reported that red ginseng at a dose of 100 mg/kg ameliorated performance decline in aged rats in the PLT [6]. The amelioration effects of the nonsaponin fraction at a dose of 50 mg/kg in the present study were comparable to or stronger than those of red ginseng at a dose of 100 mg/ kg in the previous study. Since red ginseng powder includes an even percentage of saponin and nonsaponin components (Dr. K. Kawanishi, personal communication), these results suggest that at least the nonsaponin fraction of red ginseng contains important substances to improve learning and memory in aged rats.
The mechanisms required for LTP induction and maintenance in the HF are not fully understood. It is reported that LTP is induced by a rise in presynaptic Ca2 + and requires an increased release of excitatory amino acid from mossy fiber terminals [47]. On the other hand, LTP in the mossy fiber – CA3 pathway was abolished in mice lacking the synaptic vesicle protein Rab3A, supporting the essential role of the presynaptic component in expression of the LTP in this pathway [48]. A recent study reported a central role of presynaptic kainate receptors in LTP induction in the mossy fiber – CA3 synapses in which kainate receptors facilitated glutamate release [49]. Furthermore, A-opioid receptors may play a role in induction of mossy fiber LTP as a result of their actions at the presynaptic elements [50]. Interestingly, activation of presynaptic opioid receptors altered calcium influx at the mossy fiber synapses [51]. Collectively, this evidence suggests that LTP induction in the mossy fiber –CA3 pathway requires activation of various molecular mechanisms in the presynaptic terminals of the mossy fibers to increase glutamate release. Therefore, the nonsaponin fraction of red ginseng might affect specific molecular components in the presynaptic terminals of the mossy fibers. Further studies are necessary to confirm or disprove this idea. 4.3. Relationships between learning improvement and LTP induction The present electrophysiological studies indicated that the nonsaponin fraction of red ginseng significantly augmented the LTP of population spikes in the mossy fiber – CA3 system, without effects on the basal responses recorded before tetanic stimulation. This finding suggests that the nonsaponin fraction acted selectively on the LTP-related synaptic process in the mossy fiber – CA3 system without effects on low frequent, normal neurotransmission. Previous extensive studies suggest that the mossy fiber – CA3 subfield system plays an important role in learning. For example, lesions of the CA3 subfield impaired spatial memory acquisition [52]. In addition, it is reported that LTP in the mossy fiber –CA3 region coincided well with the advance of learning [53]. Aged rats displayed distinct morphological changes in various regions of the brain, especially in the HF [54]. It has been reported that the degree of cognitive impairments was highly correlated with degenerative structural changes in the HF, in particular with the changes in the CA3 pyramidal cells [54]. Therefore, improvement of spatial learning in aged + nonsaponin rats might be attributed to augmentation of LTP in the CA3 subfield by the nonsaponin fraction. However, the nonsaponin fraction was tested using young but not aged rats in the present electrophysiological study. It has been reported that the other nootropic drugs, for example anir-
354
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
acetam [55,56] and FK960 [57,58], augmented LTP in the CA3 subfield of young rats and ameliorated memory impairments in aged rats. These previous and present results strongly suggest that the nonsaponin fraction has similar electrophysiological effects on the HF of the aged rats. Since the HF synaptic functions in aged rats differ from those in young rats [12], further electrophysiological studies using aged rats are necessary to determine whether this suggestion is correct. In conclusion, the present results suggest that the nonsaponin fraction of red ginseng contains important substances to improve learning and memory functions in aged rats, and these amelioration effects of the nonsaponin fraction might be attributed partly to augmentation of LTP in the mossy fiber – CA3 synapses.
Acknowledgements We thank Korea Tabacco & Ginseng, which provided red ginseng. This work was supported partly by the MEXT and JSPS Grants-in-Aid for Scientific Research (15500282, 15029219, 12210009).
References [1] Barnes CA. Memory deficits associated with senescence: neurophysiological and behavioral study in the rat. J Comp Physio Psychol 1979;93:74 – 104. [2] Barnes CA, McNaughton BL. An age comparison of the rates of acquisition and forgetting of spatial information in relation to longterm enhancement of hippocampal synapses. Behav Neurosci 1985; 99:1040 – 8. [3] Gallagher M, Pelleymounter MA. An age-related spatial learning deficit: choline uptake distinguishes ‘‘impaired’’ and ‘‘unimpaired’’ rats. Neurobiol Aging 1988;9:363 – 9. [4] Rapp PR, Amaral DG. Individual differences in the cognitive and neurobiological consequences of normal aging. Trends Neurosci 1992;15:340 – 5. [5] Rapp PR, Gallagher M. Preserved neuron number in the hippocampus of aged rats with spatial learning deficits. Proc Natl Acad Sci U S A 1996;93:9926 – 30. [6] Zhong Y-M, Nishijo H, Uwano T, Tamura R, Kawanishi K, Ono T. Red ginseng ameliorated place navigation deficits in young rats with hippocampal lesions and aged rats. Physiol Behav 2000;69:511 – 25. [7] Carr DB, Goate A, Phild D, Morris JC. Current concepts in the pathogenesis of Alzheimer’s disease. Am J Med 1997;103(3A):3S – 10S. [8] Corey-Bloom J, Sabbagh MN, Bondi MW, Hansen L, Alford MF, Masliah E, et al. Hippocampal sclerosis contributes to dementia in the elderly. Neurology 1997;48:154 – 60. [9] Heinonen O, Soininen H, Sorvari H, Kosunen O, Paljarvi L, Koivisto E, et al. Loss of synaptophysin-like immunoreactivity in the hippocampal formation is an early phenomenon in Alzheimer’s disease. Neuroscience 1995;64:375 – 84. [10] Mizutani T, Kashara M. Degeneration of the intrahippocampal routes of the perforant and alvear pathways in senile dementia of Alzheimer type. Neurosci Lett 1995;184:141 – 4. [11] Ott BR, Saver JL. Unilateral amnesic stroke. Six new cases and a review of the literature. Stroke 1993;24:1033 – 42. [12] Barnes CA. Long-term potentiation and the ageing brain. Philos Trans Royal Soc Lond Series B: Biol Sci 2003;358:765 – 72.
[13] Chen C, Tonegawa S. Molecular genetic analysis of synaptic plasticity, activity-dependent neural development, learning, and memory in the mammalian brain. Annu Rev Neurosci 1997;20:157 – 84. [14] Gong X, Sucher NJ. Stroke therapy in traditional Chinese medicine (TCM): prospects for drug discovery and development. Trends Pharmacol Sci 1999;20:191 – 6. [15] Terasawa K, Shimada Y, Kita T, Yamamoto T, Tosa H, Tanaka N, et al. Choto-san in the treatment of vascular dementia: a double-blind, placebo-controlled study. Phytomedicine 1997;4:15 – 22. [16] D’Angelo L, Grimaldi R, Caravaggi M, Marcoli M, Perucca E, Lecchini S, et al. A double-blind, placebo-controlled clinical study on the effect of a standardized ginseng extract on psychomotor performance on healthy volunteers. J Ethnopharmacol 1986;16:15 – 22. [17] Hallady AK, Yu YL, Palmer J, Oh KW, Wagner GC. Acute and chronic effects of ginseng total saponin and amphetamine on fixedinterval performance in rats. Planta Med 1999;65:162 – 4. [18] Tachikawa E, Kudo K, Kashimoto T, Takahashi E. Effects of ginseng saponins on receptor stimulation-responses. Nippon Yakurigaku Zasshi 1997;110:126 – 31. [19] Kim HS, Hwang SL, Nah SY, Oh S. Changes of [3H]MK-801, [3H]muscimol and [3H]flunitrazepam binding in rat brain by the prolonged ventricular infusion of ginsenoside Rc and Rg1. Pharmacol Res 2001;43:473 – 9. [20] Park HJ, Rhee MH, Park KM, Nam KY, Park KH. Effects of nonsaponin fraction from Panax ginseng on cGMP and thromboxane A2 in human platelet aggregation. J Ethnopharmacol 1995;49:157 – 62. [21] Hiyama C, Miyai S, Yoshida H, Yamasaki K, Tanaka O. Application of high-speed liquid chromatography and dual wave-length thin-layer chromatograph-densitometry to analysis of crude drugs: nucleosides and free bases of nucleic acids in ginseng roots. Yakugaku Zasshi 1978;98:1132 – 7. [22] Okuda H. Biological activities non-saponin fractions in Korean red ginseng. In: Proc. ’99 Korean – Japan Ginseng Symposium. Seoul, Korea: The Korean Society of Ginseng; 1999. p. 59 – 69. [23] Park JD, Kim YS, Park KM, Shin HJ, Jeong TC, Lee YH, et al. Biological activities of water soluble nonsaponin constituents from Korean red ginseng (Panax ginseng C.A. Meyer). In: Proc. ’99 Korean – Japan Ginseng Symposium. Seoul, Korea: The Korean Society of Ginseng; 1999. p. 191 – 209. [24] Takaku T, Han KL, Kameda K, Ninomiya H, Okuda H. Production of arginyl – fructosyl – glucose during processing of red ginseng. J Tradit Med 1996;13:118. [25] Petkov V. Effect of ginseng on the brain biogenic monoamines and 3V, 5V-AMP system: experiments on rats. Arzneimittel-Forschung/Drug Res 1978;28:388 – 93. [26] Saito H, Yoshida Y, Takagi K. Effects of Panax ginseng root on exhaustive exercise in mice. Jpn J Pharmacol 1974;24:119 – 27. [27] Saito H, Tsuchiya M, Naka S, Takagi K. Effects of Panax ginseng root on conditioned avoidance response in rats. Jpn J Pharmacol 1977; 27:509 – 16. [28] Zhang Y, Saito H, Nishiyama N. Improving effects of DX-9386, a traditional Chinese medicinal prescription, on thymectomy-induced impairment of learning behaviors in mice. Biol Pharm Bull 1994; 17:1199 – 205. [29] Dou DQ, Zhang YW, Zhang L, Chen YJ, Yao XS. The inhibitory effects of ginsenosides on protein tyrosine kinase activated by hypoxia/reoxygenation in cultured human umbilical vein endothelial cells. Planta Med 2001;67:19 – 23. [30] Kim DH, Jung JS, Suh SO, Min SK, Son BK, Park JH, et al. Inhibition of stress-induced plasma corticosterone levels by ginsenosides in mice: involvement of nitric oxide. Neuroreport 1998;9:2261 – 4. [31] Lee LH, Choi S, Kim JH, Kim JK, Kim JI, Nah SY. Effects of ginsenosides on carbachol-stimulated formation of inositol phosphates in rat cortical cell cultures. Neurochem Res 2003;28:1307 – 13. [32] Paxisons G, Watson C. The rat brain in stereotaxic coordinates. San Diego (CA): Academic Press; 1986. [33] Fukuda M, Ono T, Nishino H, Sasaki K. Independent glucose effects
H. Kurimoto et al. / Physiology & Behavior 82 (2004) 345–355
[34]
[35] [36]
[37] [38] [39]
[40]
[41]
[42]
[43] [44]
[45]
[46]
on rat hypothalamic neurons: an in vitro study. J Auton Nerv Syst 1984;10:373 – 81. Kimura T, Tamura R, Kurimoto H, Ono T. Effects of T-588, a newly synthesized cognitive enhancer, on hippocampal CA1 neurons in rat brain tissue slices. Brain Res 1999;831(1 – 2):175 – 83. Nakamura K, Ono T, Nishijo H, Tamura R. Action of neurotropin on rat hypothalamic neurons in tissue slices. Brain Res Bull 1990;24:811 – 7. Yamamoto C. Activation of hippocampal neurons by mossy fiber stimulation in thin brain sections in vitro. Exp Brain Res 1972;14: 423 – 35. Morris RGM, Garrud P, Rawlins JNP, O’Keefe J. Place navigation is impaired in rats with hippocampal lesion. Nature 1982;297:681 – 3. O’Keefe J, Nadel L. The hippocampus as a cognitive map. Oxford (UK): Clarendon Press; 1978. Hori E, Uwano T, Tamura R, Miyake N, Nishijo H, Ono T. Effects of a novel arginine-vasopressin derivative, NC-1900, on the spatial memory impairment of rats with transient forebrain ischemia. Cogn Brain Res 2002;13:1 – 15. Watanabe H, Ohta H, Imamura L, Asakura W, Matoba Y, Matsumoto K. Effect of Panax ginseng on age-related changes in the spontaneous motor activity and dopaminergic nervous system in the rat. Jpn J Pharmacol 1991;55:51 – 6. Nitta H, Matsumoto K, Shimizu M, Ni XH, Watanabe H. Panax ginseng extract improves the performance of aged Fischer 344 rats in radial maze task but not operant brightness discrimination task. Biol Pharm Bull 1995;18:1286 – 8. Nitta H, Matsumoto K, Shimizu M, Ni XH, Watanabe H. Panax ginseng extract improves the scopolamine-induced disruption of 8-arm radial maze performance in rats. Biol Pharm Bull 1995;18:1439 – 42. Zhao R, McDaniel WF. Ginseng improves strategic learning by normal and brain-damaged rats. Neuroreport 1998;9:1619 – 24. Mook-Jung I, Hong HS, Boo JH, Lee KH, Yun SH, Cheong MY, et al. Ginsenoside Rb1 and Rg1 improve spatial learning and increase hippocampal synaptophysin level in mice. J Neurosci Res 2001;63: 509 – 15. Yamaguchi Y, Haruta K, Kobayashi H. Effects of ginsenosides on impaired performance induced in the rat by scopolamine in a radialarm maze. Psychoneuroendocrinology 1995;20:645 – 53. Benishin CG, Lee R, Wang LC, Liu HJ. Effects of ginsenoside Rb1 on central cholinergic metabolism. Pharmacology 1991;42:223 – 9.
355
[47] Maeda T, Kaneko S, Akaike A, Satoh M. Direct evidence for increase in excitatory amino acids release during mossy fiber LTP in rat hippocampal slices as revealed by the patch sensor methods. Neurosci Lett 1997;224:103 – 6. [48] Castillo PE, Janz R, Sudhof TC, Tzounopoulos T, Malenka RC, Nicoll RA. Rab3A is essential for mossy fiber long-term potentiation in the hippocampus. Nature 1997;388:590 – 3. [49] Lauri SE, Bortolotto ZA, Bleakman D, Ornstein PL, Lodge D, Isaac JT, et al. A critical role of a facilitatory presynaptic kainite receptor in mossy fiber LTP. Neuron 2001;32:697 – 709. [50] Derrick BE, Weinberger SB, Martinez Jr JL. Opioid receptors are involved in an NMDA receptor-independent mechanism of LTP induction at hippocampal mossy fiber – CA3 synapses. Brain Res Bull 1991;27:219 – 23. [51] Gannon RL, Terrian DM. U50,488H inhibits dynorphin and glutamate release from guinea pig mossy fiber terminals. Brain Res 1991;548: 242 – 7. [52] Handelmann GE, Olton DS. Spatial memory following damage to hippocampal CA3 pyramidal cells with kainic acid: impairment and recovery with preoperative training. Brain Res 1981;217: 41 – 58. [53] Ishihara K, Mitsuno K, Ishikawa M, Sasa M. Behavioral LTP during learning in rat hippocampal CA3. Behav Brain Res 1997;83:235 – 8. [54] Kadar T, Silbermann M, Brandeis R, Levy A. Age-related structural changes in the rat hippocampus: correlation with working memory deficiency. Brain Res 1990;512:113 – 20. [55] Satoh M, Ishihara K, Iwama T, Takagi H. Aniracetam augments, and midazolam inhibits, the long-term potentiation in guinea-pig hippocampal slices. Neurosci Lett 1986;68:216 – 20. [56] Petkov VD, Mosharrof AH, Petkof VV, Kehayov RA. Age-related differences in memory and in the memory effects of nootropic drugs. Acta Physiol Pharmacol Bulg 1990;16:28 – 36. [57] Matsuoka N, Satoh M. FK960, a novel potential anti-dementia drug, augments long-term potentiation in mossy fiber – CA3 pathway of guinea-pig hippocampal slices. Brain Res 1998;794:248 – 54. [58] Yamazaki M, Matsuoka N, Maeda N, Ohkubo Y, Yamaguchi I. FK960 N-(4-acetyl-1-piperazinyl)-p-fluorobenzamide monohydrate ameliorates the memory deficits in rats through a novel mechanism of action. J Pharmacol Exp Ther 1996;279:1157 – 73.