Life Sciences 71 (2002) 2625 2631

www.elsevier.com/locate/lifescie

The effects of melatonin and Ginkgo biloba extract on memory loss

and choline acetyltransferase activities in the brain of rats infused
intracerebroventricularly with h-amyloid 140
F. Tang *, S. Nag1, S.Y.W. Shiu, S.F. Pang2
Department of Physiology, Faculty of Medicine, The University of Hong Kong, Hong Kong, China
Received 11 March 2002; accepted 3 June 2002

Abstract
Intraventricular infusion of rats with h-amyloid for 14 days resulted in memory deficit in the water maze as well
as decreases in choline acetyltransferase activities and somatostatin levels in the cerebral cortex and hippocampus.
These changes were not altered by daily intraperitoneal injection of 20 mg/Kg melatonin. Orally administered
Ginkgo biloba extract, however, partially reversed the memory deficit and the decrease in choline acetyltransferase
activities in the hippocampus. The latter treatment failed to reverse the decrease in somatostatin levels. The results
indicate that orally administered Ginkgo biloba extract can protect the brain against h-amyloid-induced changes
which lead to memory deficit through its effect on the cholinergic system.
D 2002 Elsevier Science Inc. All rights reserved.
Keywords: Cholinergic; Ginkgo biloba; melatonin; somatostatin; water-maze

Introduction
In our previous studies, intracerebroventricular (i.c.v.) infusion of rats with h-amyloid for 14 days
results in memory deficit in the Morris water maze as well as decreases in choline acetyltransferase
*

Corresponding author. Tel.: +852-28199-269; fax: +852-28559-730.

E-mail address: [email protected] (F. Tang).
1
Present address: Department of Anatomy and Physiology, Meharry Medical College, 1005 D.B. Todd Boulevard,
Nashville, TN-37208, USA.
2
Present address: CK Life Sciences International Inc., 2 Dai Fu Street, Tai Po Industrial Estate, New Territories, Hong
Kong, China.
0024-3205/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved.
PII: S 0 0 2 4 - 3 2 0 5 ( 0 2 ) 0 2 1 0 5 - 7

2626

F. Tang et al. / Life Sciences 71 (2002) 26252631

(ChAT) and somatostatin (SRIF) levels in the cerebral cortex and hippocampus [1,2]. As an anti-oxidant
[3], melationin has been shown to protect against hippocampal damage by kianic acid [4] and quinolinic
acid [5] by preventing the formation of lipid peroxidation products. Melatonin and its related compounds
have also been shown to protect neurons against the toxic effects of h-amyloid [6]. Furthermore, it has
been postulated that the loss of intraventricular fluid melatonin may lead to the neuropathology of
Alzheimers disease [7]. Similarly Ginkgo biloba extract EGb761 has been shown to protect neurons
from beta amyloid toxicity [8] and to enhance short-term memory in both young and aged rats [9]. In
addition, Gingko biloba has been shown to improve cognitive functions in Alzheimer patients [10]. It is
therefore of interest to look at the effects of melatonin and Ginkgo biloba extract in our h-amyloid
infused rat model.

Materials and Methods

Male Sprague-Dawley rats (3 months of age, weighing 250300 gm) were obtained from the
Laboratory Animal Unit of the University of Hong Kong. They were kept in a temperature controlled
room (22 jC) under a 12L:12D cycle (light on at 600 h) and were given free access to food and water.
In the first experiment, there were 3 groups: (1) vehicle infused and vehicle injected (veh + veh), (2)
h-amyloid infused and vehicle injected (Abeta + veh) and (3) h-amyloid infused plus daily i.p.
injection of 20 mg/Kg melatonin (Abeta + Mel). In the second experiment, there were 4 groups: (1)
vehicle infused plus oral vehicle (veh + veh), (2) vehicle infused plus oral Ginkgo biloba extract, 100
mg/Kg (veh + Egb), (3) h-amyloid infused plus oral vehicle (Abeta + veh) , and (4) h-amyloid
infused plus oral Egb, 100 mg/Kg (Abeta + Egb). The Ginkgo biloba extract was a generous gift from
CV Technologies, Canada and contained 24% flavoid glycosides and 6% terpenoids. In both
experiments, h-amyloid infusion were continued for 14 days and melatonin or Ginkgo biloba extract
was also given daily in the afternoon throughout the same period. The preparation of h-amyloid and
surgery for continual i.c.v. infusion of h-amyloid by Alzet mini osmotic pumps (Alza Corporation,
Palo Alto, CA, U.S.A.) have been described [1]. Oral administration of Ginkgo biloba extract (50 mg
in 1 ml water ) in suspension was by the use of a stomach tube.
The water maze tasks for spatial reference memory has also been described [1,2]. It lasted for 5
consecutive days, starting on postoperative day 10. One block of visible platform trials and one
block of acquisition trials (with the platform submerged in milky water) were administered on
postoperative day 10. Two blocks each of acquisition trials were given on days 1113. Escape
latency (the time the rat took to swim to the platform) was measured. On day 14 the rats were
given one probe trial in which the time spent in the target quadrant during a 60 second swimming
period was noted. The rats were decapitated on post-operative day 15 and their cortices and
hippocampi dissected out on ice to be stored at 70 jC. Choline acetyltransferase (ChAT) activities
were measured by a radioenzymatic assay [2] and somatostatin levels determined by a radioimmunoassay [11] using an anti-serum generously supplied by Dr. J. Hong (Research Triangle Park,
NC, U.S.A.). The protein contents were estimated using the protein reagents from Biorad (Hercules,
CA, U.S.A.). The data of the first experiment were subjected to one-way analysis of variance
(ANOVA) while the two-way ANOVA was used for those of second experiments. These were
followed by a post hoc LSD test for differences between groups. The statistical significance level
was set at p < 0.05.

F. Tang et al. / Life Sciences 71 (2002) 26252631

2627

Results
In the first experiment (i.e. i.p. melatonin, Fig. 1), ANOVA of the mean escape latencies per 4-trial
block demonstrated significant effects of Group [(F(2,21) = 7.639, p < 0.001], and Blocks [F(6,126) =
10.316, p < 0.01] and LSD post hoc test revealed that veh+ veh group significantly differed from the
Abeta+ veh and Abeta+ Mel groups (p < 0.01). Abeta+ veh group was not different from Abeta+ Mel
group. In the probe test, there was a significant effect of group [F(2,23) = 6.505, p < 0.01]. The LSD
post hoc test showed significant differences between veh + veh and Abeta + veh or Abeta + Mel groups.
Abeta + veh was not different from Abeta + Mel group. The ChAT values for the cortex in the veh + veh,
Abeta + veh and Abeta + Mel groups were 0.86 F 0.07, 0.54 F 0.06 and 0.64 F 0.10 respectively,
and for the hippocampus were 0.64 F 0.04, 0.41 F 0.02 and 0.33 F 0.03 respectively. Similarly, the
SRIF values for the cortex in the veh + veh, Abeta + veh and Abeta + Mel groups were 4.42 F 0.68,
1.87 F 0.24 and 1.96 F 0.39 respectively, and for the hippocampus were 2.30 F 0.41, 1.07 F 0.33
and 1.29 F 0.23 respectively. h-Amyloid infusion led to significant reduction in ChAT [F(2,24 = 3.589,
p < 0.05 for cortex; F(2,24) = 2.26, p < 0.01 for hippocampus] and in SRIF [F(2,24) = 9.633, p < 0.01
for cortex; F(2,24) = 3.881, p < 0.05 for hippocampus). The LSD post hoc test showed that veh + veh
group significantly differed from the other 2 groups (p < 0.05). Again Abeta + veh did not not differ
from Abeta + Mel group.
In the experiment with Ginkgo biloba extract (Fig. 2), the performance of the Abeta + veh group was
selectively impaired in the acquisiton phase. A 2  2  7 (Abeta  Egb  Blocks of 4 trials) splitplot ANOVA of the escape latency revealed a significant effect of Blocks [F(6, 168) = 38.14, p < 0.01]
and of Abeta [F(1,28) = 29.31, p < 0.01]. Egb had no effect. However, there was a significant
interaction between Abeta and Egb [F(1.28) = 13,42, p < 0.01] suggesting a restoration of performance
in Abeta + Egb group as indicated by their lower escape latencies compared to Abeta + veh group. Probe
test analysis further supported the above interpretation. The memory of the Abeta + veh group was
selectively impaired (spent less time in the target quadrant) which can be seen as a significant effect of
Abeta [F(1,32) = 17.44, p < 0.01]. Egb alone had no effect whereas there was a significant interaction

Fig. 1. Water-maze performance from Experiment 1. (A) Mean escape latency for the visible platform task, expressed as a
function of trials. (B) Mean escape latency for the hidden platform task, expressed as a function of blocks of four trials. (C)
Performance in the probe test, expressed as time spent in the target quadrant.

2628

F. Tang et al. / Life Sciences 71 (2002) 26252631

Fig. 2. Water-maze performance from experiment 2. See Fig. 1 for details. Significant difference between Abeta + Egb and
Abeta + veh groups emerged from 4th Block onward.

between these two factor [F(1,32) = 7.71, p = 001). This interaction suggests that Egb restored the
memory of Abeta-treated animals so that they spent more time searching for the platform in the
appropriate quadrant. Again h-amyloid infusion led to a significant reduction in ChAT [F(1,32) = 45.11,
p < 0.01 for cortex and F(1,32) = 28.13, p < 0.01 for hippocampus] and in SRIF [F(1,32) = 18.82, p <
0.01 for the cortex and F(1,32) = 35.78, p < 0.01 for hippocampus] (Table 1). Egb had significant effect
on ChAT in the cortex [F(1,32) = 6.23, p = 0.019] and in the hippocampus [F(1,32) = 4.58, p = 0.041].
There was also a significant interaction between Abeta and Egb in hipocampal ChAT (F(1.32) = 7.19,
p = 0.012], showing that Egb significantly increased ChAT level or prevented them from declining in the
Abeta treated-rats. The LSD post hoc test showed that Abeta + veh group significantly differed from veh
+ veh and veh + Egb groups (p < 0.005] in both ChAT and SRIF. The LSD test also showed that ChAT
in the hippocampus in Abeta + Egb group was significantly higher than Abeta + veh (p = 0.002)
although it was still lower than in veh + veh (p = 0.033). As to the ChAT in the cortex, the only effect of
Egb was to increase it in veh + Egb group compared with veh + veh group (p = 0.035). Egb had no effect
on SRIF.

Table 1
Neurochemical measurements in cerebral cortex and hippocampus of i.c.v. Abeta/veh + oral Egb/veh treated rats
veh + veh

Abeta + veh

veh + Egb

Abeta + Egb

ChAT (U/mg protein)

Cortex
Hippocampus

1.27 F 0.10
0.48 F 0.02

0.68 F 0.06
0.33 F 0.02

1.57 F 0.12
0.47 F 0.02

0.86 F 0.09
0.42 F 0.02

SRIF (ng/mg protein)

Cortex
Hippocampus

2.87 F 0.50
3.32 F 0.37

1.46 F 0.20
1.27 F 0.27

2.80 F 0.26
2.65 F 0.25

1.51 F 0.17
1.39 F 0.18

Data are presented as mean F S.E.M. n = 8 for all groups.

F. Tang et al. / Life Sciences 71 (2002) 26252631

2629

Discussion
The result of this study indicates that i.p. melatonin had no effect while orally given Ginkgo biloba
extract restored memory to some degree and increased ChAT activities towards normal only in the
hippocampus. Preliminary data from i.c.v. administered melatonin (20 Ag/kg B.W) are not different from
those obtained by i.p. injection. From the results of oral Ginkgo biloba extract on ChAT, one may
conclude that the cholinergic system in the hippocampus is more important than that in the cortex for
memory, at least in the Morris water-maze system, which is insensitive to lesions of the entorhinal cortex
[12]. This is in line with the current understanding of the importance of the hippocampus in memory
function [13]. Whether the effect was mediated by an increase in cholinergic activities in the existing
fibres or by an increase in the number of cholinergic fibres requires further study.
What are the mechanisms behind the effects of oral Ginkgo biolba extract on ChAT and memory loss?
Ginkgo biloba extract could have protected the cholinergic neurons against h-amyloid toxicity through
its anti-oxidant effects [14]. Furthermore, Ginkgo biloba extract could enhance memory through an
increase in cholinergic [15] and serotonergic [14] neurotransmission and increase in brain circulation.
The negative results obtained with i.p. and i.c.v. given melatonin were unexpected because melatonin
has been shown to have anti-oxidant effect [3] and to protect neural tissues against h-amyloid toxicity in
vitro [6]. Melatonin MT2 (Mel 1b) receptors has also been found in the hippocampus [16]. This
discrepancy may be because the anti-oxidant effect of melatonin may not be the only major factor
involved and because in vitro and in vivo situations are different. In this connection, it is pertinent to
note that Ginkgo biloba has been shown to have in vivo neuromodulatory effects by increasing the gene
expression of a number of proteins associated with brain function [17].
The lack of effect of oral Ginkgo biloba extract on somatostatin levels may explain why the memory
deficit was not completely reversed as somatostatin has been suggested to play some role in memory
[18,19]. Alternatively, it could be just because the ChAT activities (an indicator of cholinergic activities)
in the hippocampus were not completely restored to normal. The significance of the cholinergic system
in memory is well-established [12]. The result here cannot resolve this point.
The beneficial effects of Gingko biloba extract on the memory of h-amyloid infused rats are in line
with some of the findings in clinical trials which show a positive effect of Gingko extract on
improvement of congitive functions [20,21]. These positive clinical findings, however, have not been
confirmed by a later study by van Dongen et al. [22].
Although the results of this study do raise a number of issues, it leaves little doubt as to the
beneficial effects of oral Ginkgo biloba extract on the cholinergic system and memory loss in our rat
model. We must, however, emphasize that the h-amyloid infused rat model is at best a partial model of
Alzheimers disease. There are quite a few intrinsic differences. Firstly, the time frames are different
because our model is based on a short time of infusion (i.e. 14 days), which is not long enough for the
development of neuronal loss or neural degeneration, a characteristic feature in Alzheimers disease.
The second limitation is that the increase in h-amyloid may not be the only factor involved in the
development of Alzheimers disease although it is likely to be the key factor. The discrepancy between
this study and some of the clinical trial with negative results may be explained by these differences and
also by the much higher dose of the Gingko extract used in the rat model than in the clinical trials
(some twenty to fifty times higher in the rat). Regardless of the limitations stated above, the rat model
may still be useful for the study of Alzheimers disease especially if the duration of h-amyloid infusion
can be increased.

2630

F. Tang et al. / Life Sciences 71 (2002) 26252631

Acknowledgements
We are grateful to the Executive Panel for the Review of the Preclinical Departments, the University
of Hong Kong and the Vice-Chancellor Development Fund for Traditional Chinese Medicine, for their
financial support, Dr. John Hong, Research Triangle Park, NC, U.S.A. for his generous gift of
somatostatin anti-serum, the CV Technologies, Canada, for the gift of Ginkgo biloba extract and Dr.
L.W. Chu, Division of Geriatrics, Department of Medicine, Queen Mary Hospital, the University of
Hong Kong, Hong Kong, China, for his useful discussion.

References
[1] Nag S, Yee BK, Tang F. Reduction in somatostatin and substance P levels and choline acetyltransferase activity in the
cortex and hippocampus of the rat after chronic intracerebroventricular infusion of h-amyloid (1 40). Brain Research
Bulletin 1999;50:251 62.
[2] Nag S, Tang F. The effect of age on the response of the rat brain to continuous h-amyloid infusion. Brain Research
2001;889:303 7.
[3] Reiter RJ, Tan DX, Qi W, Manchester LC, Kaarbownik M, Calvo JR. Pharmacology and Physiology of melatonin in the
reduction of oxidative stress in vivo. Biological Signals and Receptors 2000;9:160 71.
[4] Tan DX, Manchester LC, Reiter RJ, Qi W, Kim SJ, El-Sokkay GH. Melatonin protects hippocampal neurons in vivo
against kainic acid-induced damage in mice. Journal of Neuroscience Research 1998;54:382 9.
[5] Behan WM, Mcdonald M, Darlington LG, Stone TW. Oxidative stress as a mechanism for quinolinic acid-induced
hippocampal damage: protection by melatonin and deprenyl. British Journal of Pharmacology 1999;128:1754 60.
[6] Chyan YJ, Poeggeler B, Omar RA, Chain DG, Frangione B, Ghiso J, Pappolla MA. Potent neuroprotective properties
against Alzheimer beta-amyloid by an endogenous melatonin-related indole structure, indole-3-propionic acid. Journal of
Biological Chemistry 1999;274:21931 42.
[7] Maurizi CP. Loss of intraventricular fluid melatonin can explain the neuropathology of Alzheimers disease. Medical
Hypotheses 1997;49:153 98.
[8] Dore S, Bastianetto S, Kar S, Quiron R. Protective and rescuing abilities of IGF-1 and some putative free radical
scavengers against beta-amyloid-inducing toxicity in neurons. Annuals of New York Academy of Science 1999;890:
356 64.
[9] Wirth S, Stemmelin J, Will B, Christen Y, Di-Scala G. Facilitative effects of EGb761 on olfactory recognition in young and
aged rats. Pharmacology Biochemistry and Behaviour 2000;65:321 6.
[10] Oken BS, Storzbach DM, Kaye JA. The efficacy of Gingko biloba on cognitive function in Alzheimers disease. Archives
of Neurology 1998;55:1409 15.
[11] Wang ZP, Man SY, Tang F. Age-related changes in the contents of neuropeptides in the rat brain and pituitary. Neurobiology of Aging 1993;14:529 34.
[12] Pouget B, Wellzl H, Guber MK, Broersen L, Veenam CL, Feldm J, Rawling JN, Yee BK. The effects of NMDA induced
retrohippocampal lesions on performance of four spatial memory tasks known to be sensitive to hippcampal damage in the
rat. European Journal of Neuroscience 1999;11:123 40.
[13] Blokland A. Acetylcholine: a neurotransmitter for learning and memory? Brain Research Review 1996;21:285 300.
[14] Ravel N, Chabaud P, Mannino S, Vigouroux M, Gervais R, Droix-Lefaix MT. Ginkgo biloba extract and short-term
memory. In: Christen Y, Droy-Lefaix MT, Macias-Nunez JF, editors. Effect of Ginkgo biloba extract (EGb671) on
neuronal plasticity. Paris: Elsevier; 1996. p. 35 43.
[15] Winter JC, Timineri D. The discriminitive stimulus properties of EGB761, an extract of Ginkgo biloba. Pharmacology
Biochemistry and Behaviour 1999;62:543 7.
[16] Reppert SM, Godson C, Mahle CD, Weaver DR, Slaugenhaupt SA, Gusella JF. Molecular characterization of a second
melatonin receptor expressed in human retina and brain: the Mel 1b melatonin receptor. Proceedings of National Academy
of Science USA 1995;92:8734 8.

F. Tang et al. / Life Sciences 71 (2002) 26252631

2631

[17] Watanabe CM, Wolffram S, Ader D, Rimbach G, Packer L, Macquire JJ, Schultz PG, Gohil K. The in vivo neuromodulatory effects of the herbal medicine gingko biloba. Proceedings of the National Academy of Science USA
2001;98:6577 80.
[18] Guillou JL, Micheaou J, Jafford R. The opposite effects of cysteamine on the acquisition of the different tasks in mice are
associated with bidirectional testing-induced changes in hippocampal adenyl cyclase acivity. Behavioral Neuroscience
1998;112:900 8.
[19] Matsuoka N, Yamazaki M, Yamaguchi I. Changes in brain somatostatin in memory-deficient rats: comparison with
cholinergic markers. Neuroscience 1995;66:617 26.
[20] LeBars PL, Katz MM, Berman N, Itil TM, Freedman AM, Schaltzberg AF. A placebo-controlled, double-blind, randomized trial of an extract of Ginkgo biloba for dementia. JAMA 1997;278:1327 32.
[21] Kanowski S, Herrman WM, Stephen K, Wierich W, Horr R. Proof of efficacy of the Ginkgo biloba special extract Egb 761
in outpatients suffering from mild to moderate primary degenerative dementia of the Alzheimer type or multi-infarct
dementia. Pharmacopsychiatry 1996;294:47 56.
[22] van Dongen MC, van Rossum E, Kessels AG, Sielhorst HJ, Knipschild PG. The efficacy of Ginkgo for elderly people with
dementia and age-associated memory impairment: new results of a randomized clinical trial. Journal of the American
Geriatric Society 2000;48:1183 94.