Systemic administration of defined extracts from Withania somnifera (Indian ginseng) and Shilajit differentially affects cholinergic but not glutamatergic and GABAergic markers in rat brain

~

Pergamon

Neurochem. Int. Vol. 30, No. 2, pp. 181-190, 1997 Copyright 6© 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved P l h SO 197-0186(96)00025-3 0197~)186/97 $17.00+0.00

SYSTEMIC ADMINISTRATION OF DEFINED EXTRACTS FROM WITHANIA SOMNIFERA (INDIAN GINSENG) AND SHILAJIT DIFFERENTIALLY AFFECTS CHOLINERGIC BUT NOT GLUTAMATERGIC AND GABAERGIC MARKERS IN RAT BRAIN R E I N H A R D SCHLIEBS,*§ A N D R I ~ L I E B M A N N , * S A L I L K. B H A T T A C H A R Y A , t A S H O K K U M A R , * S H I B N A T H GHOSAL:~ a n d V O L K E R B I G L * *Paul Flechsig Institute for Brain Research, Department of Neurochemistry, University of Leipzig, D-04109, Leipzig, Germany; tDepartment of Pharmacology, Institute of Medical Sciences, Banaras Hindu University, Varanasi, 221005, India; ~Department of Pharmaceutics, Institute of Technology, Banaras Hindu University, Varanasi, 221005, India

(Received 2 January 1996; accepted 13 March 1996) Abstract--Although some promising results have been achieved by acetylcholinesterase inhibitors, an effective therapeutic intervention in Alzheimer's disease still remains an important goal. Sitoindosides VIIX, and withaferin-A, isolated from aqueous methanol extract from the roots of cultivated varieties of Withania somnifera (known as Indian Ginseng), as well as Shilajit, a pale-brown to blackish brown exudation from steep rocks of the Himalaya mountain, are used in Indian medicine to attenuate cerebral functional deficits, including amnesia, in geriatric patients. The present investigation was conducted to assess whether the memory-enhancing effects of plant extracts from Withania somnifera and Shilajit are owing to neurochemical alterations of specific transmitter systems. Therefore, histochemistry to analyse acetylcholinesterase activity as well as receptor autoradiography to detect cholinergic, glutamatergic and GABAergic receptor subtypes were performed in brain slices from adult male Wistar rats, injected intraperitoneally daily with an equimolar mixture of sitoindosides VII-X and withaferin-A (prepared from Withania somnifera) or with Shilajit, at doses of 40 mg/kg of body weight for 7 days. Administration of Shilajit led to reduced acetylcholinesterase staining, restricted to the basal forebrain nuclei including medial septum and the vertical limb of the diagonal band. Systemic application of the defined extract from Withania somnifera, however, led to differential effects on AChE activity in basal forebrain nuclei: slightly enhanced AChE activity was found in the lateral septum and globus pallidus, whereas in the vertical diagonal band AChE activity was reduced following treatment with sitoindosides VII-X and withaferin-A. These changes were accompanied by enhanced M rmuscarinic cholinergic receptor binding in lateral and medial septum as well as in frontal cortices, whereas the M2-muscarinic receptor binding sites were increased in a number of cortical regions including cingulate, frontal, piriform, parietal and retrosplenial cortex. Treatment with Shilajit or the defined extract from Withania somnifera affected neither GABAA and benzodiazepine receptor binding nor NMDA and AMPA glutamate receptor subtypes in any of the cortical or subcortical regions studied. The data suggest that Shilajit and the defined extract from Withania somnifera affect preferentially events in the cortical and basal forebrain cholinergic signal transduction cascade. The drug-induced increase in cortical muscarinic acetylcholine receptor capacity might partly explain the cognition-enhancing and memory-improving effects of extracts from Withania somnifera observed in animals and humans. Copyright © 1996 Elsevier Science Ltd

Alzheimer's disease is characterized by a loss of memory a n d cognition as well as disturbances in higher

cortical functions leading to severe d e m e n t i a a n d death. Post m o r t e m n e u r o p a t h o l o g i c a n d biochemical studies have d e m o n s t r a t e d selective a n d severe losses of cholinergic n e u r o n s in the basal forebrain complex. Therefore, drugs t h a t e n h a n c e cholinergic activity have been investigated as potential therapeutic agents in the t r e a t m e n t of Alzheimer's disease. A l t h o u g h some promising results have been achieved by admin-

§Author to whom all corresondence should be addressed. Tel: +49-341-9725733; Fax: +49-341-2114492. Abbreviations: ACHE, acetylcholinesterase; mAChR, muscarinic acetylcholine receptor; WS, equimolar mixture of sitoindosides VII-X and withaferin-A, prepared from Withania somnifera. 181

182

R. Schliebs et al.

istration of acetylcholinesterase inhibitors including physostigmine or tacrine, an effective therapeutic intervention in Alzheimer's disease still remains to be found. Withania somnifera (L) Dun., popularly known as Indian Ginseng, is used in Indian medicine to attenuate cerebral function deficits, including amnesia in geriatric patients (Ghosal et al., 1989). The beneficial effects in treating elderly people and the obvious absence of any toxicity or side effects of the plant extracts, usually individually prescribed without any standard recipe, have initiated the search for the active compounds of plant extracts from Withania somnifera. Investigations on the active chemical constituents of the plant led to the isolation of withaferinA (Lavie et al., 1965), which was found to exhibit immunosuppressive activities (Fiigner, 1973). The compound represented the first member of a new class of phytosteroids, the withanolides (Lavie et al., 1965). Later the occurence of O-glycosylated withanolides, named sitoindosides, was reported in extracts from the roots of several cultivated varieties of Withania somnifera (Ghosal et al., 1988). Unlike withaferin-A, which is chemically a withanolide aglycone, the Oglycosylated forms of the compound, namely sitoindosides VII-X and withaferin-A, administered in an equimolar mixture to rats, produced immunostimulation (Bhattacharya et al., 1987; Chattopadhyay et al., 1987a,b; Ghosal, 1985) and promoted learning acquisition and memory retention in both young and old rats (Ghosal et al., 1989). More recently, sitoindosides from Withania somnifera were shown to reverse cognitive deficits induced by ibotenic acid, concomitant with attenuation of cholinergic deficits induced by these agents (Bhattacharya et al., 1995a). Shilajit exhibits another remedy used in the treatment of memory deficits, including that seen in senile and presenile dementias (Bhattacharya and Ghosal, 1992; Ghosal, 1993; Ghosal et al., 1993). A pale-brown to blackish-brown exudation from steep rocks (height: 1000-5000 m), Shilajit can be collected throughout the Himalayan mountains from Arunachal Pradesh in the East to Kashmir in the West (Ghosal, 1990). There is experimental evidence that Shilajit contains active chemical constituents that enhance learning acquisition and memory in rats (Ghosal et al., 1993) and induce an increase in oxidative free radical scavenger activity in rat striatum and frontal cortex (Bhattacharya et al., 1995b). Oxygenated dibenzo-~pyrones, in combination with medium molecular weight fulvic acids, have been found to be the major active constituents of Shilajit (Ghosal, 1990, 1993; Ghosal et al., 1993). The efficacy of native Shilajit can

be augmented by further processing the water-soluble fraction of the tritiurated substance suspended in distilled water. This processed Shilajit (Ghosal et al., 1993), as well as a preparation consisting of a mixture of an extract by ethyl acetate and medium molecular weight fulvic acids, was found to be more effective in promoting learning and memory of rats than the unprocessed native substance (Ghosal et al., 1993). To reveal whether the learning and memoryimproving effects of a defined extract from Withania somnifera and Shilajit are due to alterations in neurotransmission of a particular transmitter system, both histochemistry for acetylcholinesterase (ACHE) and quantitative receptor autoradiography for a number of transmitter receptor subtypes of the cholinergic, glutamatergic and GABAergic system were performed in brain slices from drug-treated rats. Neurotransmitter receptors are one of the decisive links in the chain of synaptic information processing, and they can respond to drug-induced alterations in neuronal activity by adaptive mechanisms like sub- and supersensitivity or down-regulation (Schwartz et al., 1983). This study was designed to screen for drug-induced changes in receptor binding and AChE staining through the whole brain, including basal forebrain nuclei, hippocampal formation, and cerebral cortical areas.

MATERIAL AND METHODS

Materials

[N-methyl-aH]pirenzepine (specific activity 3.03 TBq/mmol), [2,3-dipropylamino-aH]AF-DX384 (4.44 TBq/mmol), [methyl-aH]hemicholinium-3 diacetate (5.27 TBq/mmol), [3-3H](+)MK-801 (951 GBq/ mmol), and [5-3H]CNQX (577 GBq/mmol) were purchased from New England Nuclear, DuPont, Germany. [Methylamine-aH]Muscimol (740 GBq/mmol) and [N-methyl-3H] flunitrazepam (3.15 TBq/mmol) were provided by the Radiochemical Centre Amersham-Buchler, Germany. Atropine (SIGMA), MK-801 (RBI), L-glutamate, gamma-aminobutyric acid (SIGMA), tetraisopropyl pyrophosphoramide (iso-OMPA; SIGMA) and chlordiazepoxide (gift from Arzneimittelwerk Dresden) were used for nonspecific binding; all other chemicals used were commercial products of highest purity available. Treatment o f animals and tissue preparation

Male Wistar rats (weighing 220--250 g) were housed a under controlled laboratory environment with a 12-

Cholinergic receptors after treatment with sitoindosides Shilajit h light-dark cycle, in groups of five, in standard laboratory cages with ad libitum access to food and water. They were injected daily, intraperitoneally, with either an equimolar mixture of sitoindosides VII-X and withaferin-A (WS) or with Shilajit at 40 mg/kg of body weight (injection volume: 0.5 ml) for 7 days. Each experimental group consisted of four Shilajittreated or five WS-treated animals. An equal number of animals received 0.5 ml of the vehicle solution alone and they were considered to be controls. The equimolar mixture of sitoindosides VII-X and withaferinA (WS) was prepared from Withania somnifera roots as described previously (Bhattacharya et al., 1995a). Shilajit was prepared and processed as described previously (Ghosal et al., 1991, 1993), and both WS and Shilajit were used as solutions in distilled water in a concentration of 20 mg/ml. Two hours after the last injection, the rats were sacrificed, the brains were rapidly removed and frozen at - 7 0 ° C . Thick serial coronal sections (12 p.m each) were cut, thaw-mounted on gelatin coated slides and stored at - 2 0 ° C pending histochemical and receptor autoradiographic analysis.

Acetylcholinesterase (ACHE) histochemistry Histochemical staining for acetylcholinesterase (ACHE) was performed in adjacent sections in each cortical level measured according to the method of Andr~ and Lojda (1986). Briefly, air dried cryocut sections were preincubated for 30 min at 37°C in 0.1 M Tris-maleate buffer (pH 5.0) containing 30 ~tM iso-OMPA to inhibit nonacetylcholinesterases. After preincubation sections were incubated in a solution consisting of 1.7 mM acetylthiocholine iodide, 40 mM sodium citrate, 12 mM cupric sulfate, 8 mM potassium ferricyanide, 30 pM iso-OMPA and 75 mM Tris-maleate buffer (pH 5.0) for 60 rain at 37°C. The reaction was stopped by rinsing sections in 0.1 M Trismaleate buffer (pH 5.0) followed by a short dipping in distilled water. Finally, the sections were dehydrated and coverslipped.

Receptor binding assays Ml-muscarinic acetylcholine receptor (mAChR) subtype binding was carried out as described previously (Kumar and Schliebs, 1992) using [3H]pirenzepine as a specific radioligand. Briefly, slides were incubated in humid chambers with 50 mM sodiumpotassium phosphate (pH 7.4) containing [3H]pirenzepine at a final concentration of 9 nM for 1 h at room temperature, followed by two rinses of 1 min each in ice-cold buffer before drying in a light air stream. Nonspecific binding, which was estimated in

183

adjacent sections by adding 100 p.M atropine sulphate to the incubation buffer, and amounted to about 10% of total binding. [3H]AF-DX384 binding to the M2-mAChR subtype was performed as previously described (Schliebs et al., 1994). Briefly, after preincubation for 30 min, slides were incubated for 1 h at room temperature with Krebs buffer (NaC1, 120 raM; MgSO4, 1.2 mM; KHzPO4, 1.2 mM; NaHCO3, 25 mM, CaCI2, 2.5 mM, KC1, 4.7 raM; D-glucose, 5.6 mM; pH 7.4) containing 15 nM [3H]AF-DX384. After incubation, slides were given four rinses of 3 min each in ice-cold 50 mM Tris-HC1 (pH 7.4) followed by a short dipping in icecold distilled water before drying in a light air stream. Nonspecific binding which was estimated in adjacent sections by adding 100 ~tM atropine sulphate to the incubation buffer, amounted to about 5% of total binding. [3H]muscimol binding to GABAA receptors was carried according to the method of Kumar and Schliebs (1993). Slides were preincubated three times for 5 min each in ice-cold 50 mM Tris-citrate buffer containing 2.5 mM CaC12 (pH 7.4) to remove endogenous GABA. After preincubation, the dried slides were covered with 50 mM Tris-citrate containing 2.5 mM CaCl2 and 34 nM [3H]muscimol (pH 7.4) and incubated in humid chambers for 30 min at 4°C. Following incubation, slides were rinsed three times for 3 s each with ice-cold buffer followed by one quick rinse with glutaraldehyde (2.5% in acetone) before drying in a light air stream. Nonspecific binding, which was estimated in adjacent sections by adding 100 ~tM unlabelled G A B A to the incubation buffer, amounted to about 30% of total binding. For [3H]flunitrazepam binding to benzodiazepine receptors (Kumar and Schliebs, 1993), sections were preincubated three times for 5 min each in ice-cold 50 mM Tris-HC1 (pH 7.4). After drying, slides were immediately flooded with an incubation solution consisting of 50 mM Tris-HC1 (pH 7.4) and 9 nM [3H]flunitrazepam. After 1 h of incubation at room temperature, slides were given three rinses of 1 min each in ice-cold buffer before drying in a light air stream. Nonspecific binding, determined in adjacent sections, co-incubated with 100 ~tM chlordiazepoxide, normally did not exceed 5% of total binding. [3H]MK-801 binding to N M D A receptors was carried out according to the method of Kumar et al. (1994). Sections were prewashed for 45 min in 50 mM Tris-acetate buffer (pH 7.4) at 4°C and blown dry under a stream of air at room temperature before the binding assay was performed. After preincubation, the dried slides were covered with 50 mM Tris-acetate

184

R. Schliebs et al.

(pH 7.4) containing 50 nM [3H]MK-801 and incubated in humid chambers for 2 h at room temperature. Following incubation, slides were dipped quickly into ice-cold buffer, then rinsed for 80 min in ice-cold buffer and blown dry under warm air. Nonspecific binding, which was estimated in adjacent sections by adding 100 gM unlabeled MK-801 to the incubation buffer, represented < 5% of total binding. For [3H]CNQX binding to A M P A receptors (Kumar et al., 1994), sections were prewashed for 45 min in 50 mM Tris-acetate buffer (pH 7.2) at 4°C and blown dry under a stream of warm air prior to the binding assay. After preincubation, the dried slides were flooded with 50 mM Tris-HC1 (pH 7.2) containing 48 nM [3H]CNQX and incubated in humid chambers for 45 rain at 2°C. Following incubation, sections were rinsed quickly three times with ice-cold buffer for 3 s each, and blown dry under warm air. Nonspecific binding was estimated in adjacent sections in the presence of 1 mM L-glutamate in the incubation buffer, represented < 5% of total binding. Preliminary experiments were performed with each radioligand to characterize binding sites and to determine optimal binding and washing conditions (reaching equilibrium at the time indicated; binding was displaceable by unlabeled drugs used; checking for optimal buffer systems and rinsing times).

Receptor autoradiography The labeled and dried tissue sections were apposed to tritium-sensitive film (3H-Hyperfilm, Amersham, Bucks, U.K.) at 4°C, together with slides containing standards of known radioactivity level (3H-microscale, Amersham, Bucks, U.K.). After exposure for 2-4 weeks (depending on the radioligand applied), the films were developed with a Kodak D19 developer for 5 min at 20~'C, fixed, rinsed and dried.

Evaluation of autoradioframs Quantitative analysis of the autoradiograms was done on a video camera-based, computer assisted imaging device (SIGNUM) using the autoradiographic software package Image Pro Plus, version 1.1. For calibration of grey values (optical density) 3Hmicroscale standards (Amersham, Bucks, U.K.) coexposed with labeled sections were used. The density of the receptor binding sites was calculated from the mean grey level determined in the corresponding tissue region using a calibration curve plotted from the radioactivity of the tissue standards (kBq/mg tissue equivalent) and the densitometrically determined optical density values of the respective auto-

radiograms. Receptor densities were expressed as fmol of specifically bound radioligand per milligram of tissue (for details of quantification, see also Kumar and Schliebs, 1992). For evaluation of cortical regions all over the brain coronal sections at selected distances from the b r e g m a - - +3.7, + 1.2, - 0 . 7 , - 1.3, - 3 . 3 , - 5 . 3 mm (according to the atlas of Zilles, 1985)-- were used. The levels of cryocutting were selected to include data analysis for all cortical areas that receive a prominent cholinergic innervation from the basal forebrain, basal forebrain nuclei, striatum and hippocampal formation. Particularly, optical density readings were performed in the following regions: accumbens, subfields of hippocampal formation including dentate gyrus and subiculum, frontal cortices (Frl, Fr2, Fr3, according to the atlas of Zilles, 1985), parietal cortices (Parl, Par2), temporal cortices (Tel, Te2, Te3), occipital cortices (OC 1, OC2), forelimb and hindlimb area, piriform, cingulate, entorhinal and retrosplenial cortex, as well as corpus striatum, globus pallidus, lateral and medial septum, and vertical and horizontal limb of the diagonal band. The data obtained were corrected for non-specific binding. Measurements were made on three consecutive sections from each animal. The corresponding data obtained from 4-5 animals in each experimental group were averaged. For measuring the cortical laminar distribution of receptor binding sites, the variation of grey levels along a 1 mm thick band over the entire cortical depth (from the pial surface to the cortex white matter boundary) was determined (binding density thick profile) and corrected for nonspecific binding. Measurements were made on 3 5 sections from each animal and the data were averaged. The mean binding profiles through the cortex, from at least five different animals belonging to each experimental group, were estimated and expressed graphically.

Statistical analysis A one-way analysis of variance (ANOVA) was used to examine differences in the neurochemical parameters measured between brain region and coronal level and between brain regions of control and experimental animals, followed by a two-tailed Student's ttest. RESULTS To check whether the function of the cholinergic transmission is affected by administration of WS and/or Shilajit, brain sections were stained for AChE and semiquantitatively evaluated by image analysis.

Cholinergic receptors after treatment with sitoindosides Shilajit The relative changes in AChE staining (expressed as percentage change over the corresponding control region) in a number of brain regions following treatment with WS and Shilajit are presented in Fig. 1. The data revealed that in rats chronically injected with WS, the enzyme staining was significantly increased in the lateral septum and globus pallidus, by 20 and 12% (P<0.05, two-tailed Student's t-test), respectively, whereas in the vertical diagonal band a 12% decrease in AChE activity was observed as compared with the corresponding control regions (Fig. 1). No effect of WS on AChE staining in other cortical and subcortical regions studied could be detected (Fig. 1). Rats injected with Shilajit for 7 days demonstrated a decline in AChE staining in the medial septum and vertical diagonal band by about 20% (P<0.05) as compared with vehicle-injected control animals, whereas in the temporal cortex 2 a significant 20% increase was observed (Fig. 1).

185

Binding levels of M~- and M2-mAChR, N M D A and AMPA as well as GABAA and benzodiazepine receptors were determined in consecutive coronal brain sections at six selected distances from the bregma, and the autoradiograms obtained were evaluated by quantitative image analysis. Typical autoradiograms obtained by labeling mAChR subtypes are shown in Fig. 2. Daily injections of WS over 7 days resulted in enhanced levels in Mj-mAChR binding sites in the frontal cortex (Fr2 and Fr3) by about 20% (P<0.05) as compared with corresponding control values, but other cortical regions were not affected by the treatment (Fig. 3). Although the M]-mAChR binding levels in the basal forebrain nuclei are relatively low compared to the cerebral cortex and hippocampal formation (Fig. 2), a significant increase in M~-mAChR binding was observed in the lateral and medial septum by 43 and 48% (P<0.05), respectively, following treatment with WS (Fig. 3).

Acetylcholinesterase 140 130

--

**

120

o t0

i' 9o eo g

7 6O lid

ill Brain

[] [B

regions

ws Shilajit

Fig. 1. Effect of systemic application of active constituents of Withania somnifera (WS, black columns) and Shilajit (grey columns) for 7 days on acetylcholinesterase (ACHE) staining in selected rat brain regions. Brain coronal sections at various distances from the bregma were used for histochemical staining and the stained sliceswere evaluated by image analysis. Data represent the mean values obtained from five animals and are given as relative changes expressed as a percentage over the corresponding brain region from the vehicle-injectedcontrol rat. Acb, accurnbens;Cing, cingulum; CPu, caudate putamen; DG, dentate gyrus; Ent, entorhinal cortex; Fr, frontal cortex; GP, globus pallidum; HDB, horizontal limb of the diagonal band; HL, hind limb area; FL, forelimb area; LS, lateral septum; MS, medial septum; OCC, occipital cortex; Par, parietal cortex; Pir, piriform cortex; PrS, presubiculum; RSG/RSA, retrosplenial cortex; S, subiculum; Te, temporal cortex; VDB, vertical limb of the diagonal band. *P<0.05, **P<0.02 vs. corresponding control region, two-tailed Student's t-test.

186

R. Schliebs et al.

Fig. 2. Representative examples of autoradiograms obtained from rat brain cryocut sections labeled for (A) M~ and (B) Mz-muscarinic acetylcholine receptors using [3H]pirenzepine and [3H]AF-DX384, respectively, as specific radioligands.

['H]Pirenzepine Binding Sites 400 350 300 Q m

'5 2SO 2..

~

150

O

100

iiil tii,

Z

.~. I~ m

I,-

Brain regions

Control W$ [] Shilajit Fig. 3. Effect of systemic application of active constituents of Withania somnifera (WS, black columns) and Shilajit (grey columns) for 7 days on Ml-muscarinic receptor binding in selected rat brain regions as compared with vehicle-injected control rats (white columns) using [3H]pirenzepine as specific radioligand. Brain coronal sections at various distances from the bregma were used for receptor autoradiography and the autoradiograms obtained were evaluated by image analysis. Data represent the mean values obtained from five animals and are given as fmoles of specifically bound [3H]pirenzepine per milligram tissue equivalent. For abbreviations, see legend to Fig. 1. *P < 0.05, **P < 0.02 vs. corresponding control region, two-tailed Student's t-test.

Cholinergic receptors after treatment with sitoindosides Shilajit M2-mAChR binding was enhanced in a number of cortical, but not subcortical, regions following chronic injections of WS, as presented in Fig. 4. Enhancements in Mz-mAChR after WS were seen in the cingulate cortex (by 23%, P<0.05), in frontal cortex (about 22%, P<0.05), in the hindlimb and forelimb areas (22 %, P < 0.05), in the parietal cortex (17 %, P < 0.05), as well as in the piriform and retrosplenial cortex (17% and 12%, respectively, P<0.05; Fig. 4). In contrast, treatment of rats with Shilajit for 7 days affected neither M1- nor Mz-mAChR binding in any of the cortical or subcortical regions studied (Figs 3 and 4). In order to elucidate whether the effects of drug administration on cortical muscarinic acetylcholine receptors are due to changes in a particular cortical layer, binding profiles through the depth of the cortical area along a 1 mm thick band were determined. A representative laminar pattern of Mz-mAChR binding sites through the parietal cortex is presented graphically in Fig. 4. The comparison of binding profiles through the parietal cortex of both control and exper-

187

imental animals revealed that the administration of WS affects Mz-mAChR binding preferentially in the upper cortical layers (Fig. 5). Daily single administrations of WS or Shilajit to rats for 7 days affected neither GABAA and benzodiazepine receptor binding nor N M D A and A M P A glutamate receptor subtypes in any of the cortical or subcortical regions studied (data not shown). DISCUSSION The aim of this study was to show whether the learning and cognition-enhancing effects detected after subchronic systemic administration of WS and Shilajit to rats (Bhattacharya et al., 1995a; Ghosal et al., 1989) can be related to changes in neurochemical transmission. The study was designed as a thorough screening of receptor markers of the major transmitter systems presumably involved in learning and memory mechanisms, in order to elucidate which transmitter systems are affected by the drugs tested. At this stage

[ ' H ] A F - D X 384 Binding Sites 350 300

]

2SO

g 2oo

"= 150 100 50

nlii,l'l[lilliiiiiil ~n

r~mns

Control • WS [] Shilajit Fig. 4. Effect of systemic application of active constituents of Withania somnifera (WS, black columns) and Shilajit (grey columns) for 7 days on Mz-muscarinic receptor binding in selected rat brain regions as compared with vehicle-injected control rats (white columns) using [3H]AF-DX384 as specific radioligand. Brain coronal sections at various distances from the bregma were used for receptor autoradiography and the autoradiograms obtained were evaluated by image analysis. Data represent the mean values obtained from five animals and are given as fmoles of specifically bound [3H]AF-DX384 per milligrams of tissue equivalent. For abbreviations, see legend to Fig. 1. *P<0.05, **P<0.02, ***P<0.01 vs. corresponding control region, two-tailed Student's t-test.

188

R. Schliebs et al.

300

] 100 ~Q[~

[~ ~ [~ [~ [~ [~

~

m_J _

D~

. . . . .

I i

O+

I

0

20

•

40

60

80

100

Cortical thickness (%)

[]

Control

Withania somnifera

Fig. 5. Representative laminar distribution of M2-muscarinic acetylcholine receptor binding sites through the whole cortical depth of the parietal cortex of rats treated with active constituents of Withania somnifera (WS, m) or with the vehicle solution alone (control; [5]) as obtained by image analysis. of investigation we are unable to state whether the alterations in receptor binding are owing to changes in maximum receptor number and/or receptor affinity. However, owing to the adaptive properties of receptors to respond to altered neuronal activity by superand sub-sensitivity, the changes in receptor binding following drug treatment should reflect the functional alterations in a particular transmission system. It is well known that the basal forebrain cholinergic system plays an important role in cortical arousal and normal cognitive function, and cortical cholinergic dysfunction has been implicated in cognitive deficits that occur in Alzheimer's disease (see, e.g. Nordberg, 1992). Indeed, the most prominent changes following application of the compounds studied were observed in the cholinergic system, but the data also clearly demonstrated differences in the mode of action of WS and Shilajit. While administration of Shilajit led to reduced AChE staining restricted to the the basal forebrain nuclei, including medial septum and the vertical limb of the diagonal band, the application of WS resulted in differential changes in both AChE activity restricted to some basal forebrain nuclei and in M~and M2-mAChR binding sites in cortical and/or basal forebrain regions.

To our knowledge, there are no previous studies investigating the neurochemical consequences of systemic administration of Shilajit. However, there are indications that Shilajit might act as free radical scavenger in rat striatum and frontal cortex (Bhattacharya et al., 1995b). Here we have found that Shilajit slightly reduces AChE activity in basal forebrain nuclei. Changes in brain AChE activity as a consequence of drug treatment could be due to either cholinergic cell loss, inhibition of the enzyme by the drug or by suppressing AChE expression in the cell body. Shilajitinduced cholinergic cell loss can be excluded, and a possible inhibitory effect of Shilajit on AChE activity must await further biochemical analysis. However, when interpreting the effect of Shilajit we also have to take into consideration that AChE is not exclusively associated with cholinergic neurons (Eckenstein et al., 1988). AChE can also be localized on GABAergic cells, which are intermingled with cholinergic neurons in all basal forebrain nuclei. GABAergic cells in the medial septum receive a cholinergic input via muscarinic cholinergic receptors, and membrane-bound AChE is known to be localized near to mAChR sites. Thus the Shilajit-induced decrease in medial septal AChE might also affect the cholinergic control of

Cholinergic receptors after treatment with sitoindosides Shilajit GABAergic neurons in the medial septum, a region which provides the major source for the GABAergic afferents to the hippocampal formation. The decreased AChE activity found in basal forebrain nuclei after Shilajit administration did not have any impact on cortical cholinergic receptors, as seen in animals treated with WS. So the question arises whether the duration of the treatment with Shilajit used in this study was long enough to produce any significant neurochemical effect. Possiblybly, more pronounced effects of Shilajit on the cholinergic system could be seen after longer periods of treatment. The mode of neurochemical action of defined extracts from Withania somnifera still is not known. Administration of WS led to differential effects on AChE activity in basal forebrain nuclei: slightly enhanced AChE activity was found in the lateral septurn and globus pallidus, whereas in the vertical diagonal band AChE activity was reduced following WS treatment. These changes were accompanied by enhanced MI-mAChR binding in septal nuclei. At this stage of the investigation we can only speculate that the change in basal forebrain mAChR binding is a consequence of altered cholinergic activity as indicated by the drug-induced changes in AChE activity in the basal forebrain. However, increased M2- and, partly, M~-mAChR binding was also observed in a number of neocortical regions including frontal, piriform, parietal, retrosplenial and cingulate cortex. Taking into consideration the adaptive properties of receptors to respond to altered synaptic transmitter level by becoming supersensitive, the enhanced cortical M2-mAChR binding suggests a reduced cortical cholinergic activity induced by WS administration. However, this contradicts to the absence of any change in cortical AChE activity. So the increased M2-mAChR binding could be the result of the drug itself, presumably by affecting the binding site and thus inducing an enhanced mAChR capacity at the synaptic membrane surface. From a number of experiments it is well known that blocking cortical muscarinic cholinergic receptors by cholinergic antagonists like atropine or scopolamine impairs learning and memory behaviour in rats (see, e.g., Dunnett et al., 1991). So the increased muscarinic receptor density following WS treatment might be vice-versa responsible for the cognition-enhancing effects of WS observed in humans and animals. In earlier studies it has already been demonstrated that the O-glycosylated derivatives of withanolides (e.g. sitoindosides I-VIII including those present in Withania somnifera) produced immunostimulant effects in rats and mice (Bhattacharya et al., 1987; Chattopadhyay

189

et al., 1987a; Ghosal et al., 1989). Systemic application of these glycowithanolides induced alertness and arousal in animals and exhibited significant anti-stress activity (Bhattacharya et al., 1987). Whether these immunomodulatory effects of Withania somnifera are additionally involved in the changes in some cholinergic markers following treatment with WS cannot be clarified from this study but must await further analysis. The changes in markers of the neocortical cholinergic system found in brains of Alzheimer patients are complemented by alterations in glutamate and G A B A receptors in the cerebral cortex (for review see, e.g., Carlson et al., 1993; Nordberg, 1992), suggesting an important influence of the cholinergic basal forebrain system on glutamatergic and GABAergic transmission in the cerebral cortex. This is emphasized by recent lesion studies demonstrating changes in N M D A , A M P A and GABAA receptor binding in cortical regions displaying reduced cholinergic activity resulting from ibotenic acid (RoBner et al., 1994) or cholinergic immunolesion (Rol3ner et al., 1995). However, neither compound tested in this study showed any effect on G A B A and glutamate receptor subtypes in any of the cortical and subcortical regions studied. Obviously, the changes in the cholinergic neurotransmission following treatment with WS or Shilajit were not effective enough to influence other associated cortical transmitter systems which might also be due to the relatively short duration of drug treatment. At present, it is too early to understand the cellular and molecular mechanisms that underlie the therapeutic actions of WS and Shiljait. All the conclusions regarding the mode of action of WS and Shilajit must remain necessarily tentative at this stage of investigation. However, the data clearly demonstrate that WS and Shilajit differentially affect some events in cortical and basal forebrain cholinergic signal transduction cascade. The drug-induced increase in cortical mAChR capacity might explain the cognition-enhancing and memory-improving effects of extracts from Withania somnifera observed in animals and humans. So drugs that induce cortical muscarinic receptor enhancements without severely affecting other cholinergic events might represent a useful therapeutical alternative in treating demential disorders compared with the use of AChE inhibitors, which severely interfere with the cholinergic transmission and produce a number of toxic side effects. But regardless of possible explanations, more detailed pharmacological and neurochemical studies are needed to specify any therapeutic advantage that plant extracts from Withania

190

R. Schliebs et al.

somnifera m i g h t have over other cholinergic drugs currently in use to treat Alzheimer's disease.

Ghosal S. (1990) Chemistry of Shilajit, an immunomodulatory Ayurvedic rasayan. Pure Appl. Chem. (IUPAC) 62, 1285-1288.

Ghosal, S. (1993) Shilajit, its origin and vital significance. In: Acknowledgements--The authors gratefully acknowledge the Traditional Medicine (Mukherjee B., Ed.), pp. 308-319. expert technical assistance of Mrs Renate Jendrek. This work ' Oxford-IBH, New Dehli, India. was supported by a grant to R.S. from the Bun- Ghosal S., Lal J., Jaiswal A. K. and Bhattacharya S. K. (1993) Effects of Shilajit and its active constituents on desministerium far Forschung und Technik, no. FKZ 01 ZZ learning and memory in rats. Phytother. Res. 7, 29-34. 9130/2.8. S.K.B. was the recipient of a short-term senior Ghosal S., Lal J., Singh S. K., Goel R. K., Jaiswal A. K. and fellowship granted by the Deutscher Akademischer AusBhattacharya S. K. (1991) Shilajit part 9. The need for tauschdienst (DAAD). formulation of shilajit by its isolated active constituents. Phytother. Res. 5, 211-216. Ghosal S., Lal J., Srivastava R., Bhattacharya S. K., UpaREFERENCES dhyay S. N., Jaiswal A. K. and Chattopadhyay U. (1989) Immunomodulatory and CNS effects of sitoindosides IX Andrh J. and Lojda Z. (1986) A histochemical method for and X, two new glycowithanolides from Withania somnithe demonstration of acetylcholinesterase activity using fera. Phytother. Res. 3, 201-206. semipermeable membranes. Histochemistry 84, 575-579. Bhattacharya S. K. and Ghosal S. (1992) Effect of Shilajit Ghosal S., Kaur R. and Srivastava R. S. (1988) Sitoindosides IX and X, new glycowithanolides from Withania somnion rat brain monoamines. Phytother. Res. 6, 163-164. fera. Ind. J. Natural Prod. (Sagar) 4, 12-13. Bhattacharya S. K., Kumar A. and Ghosal S. (1995) Effects of glycowithanolides from Withania somnifera on an ani- Kumar A. and Schliebs R. (1992) Postnatal development of cholinergic receptors, protein kinase C and dihymal model of Alzheimer's disease and perturbed central dropyridine-sensitive calcium antagonist binding in rat vischolinergic markers of cognition in rats. Phytother. Res. ual cortex. Effect of visual deprivation. Int. J. Devl. 9, I 1(~113. Neurosci. 10, 491-504. Bhattacharya S. K., Sen A. P. and Ghosal S. (1995) Effects Kumar A. and Schliebs R. (1993) Postnatal ontogeny of of Shilajit on biogenic free radicals. Phytother. Res. 9, 56GABAA and benzodiazepine receptors in individual layers 59. of rat visual cortex and the effect of visual deprivation. Bhattacharya S. K., Goel R. K., Kaur R. and Ghosal S. Neurochem. Int. 23, 99-106. (1987) Anti-stress activity of sitoindosides VII and VIII, Kumar A., Schliebs R. and Bigl V. (1994) Postnatal develnew acylsterylglucosides from Withania somnifera. Phyopment of NMDA, AMPA, and kainate receptors in inditother. Res. 1, 32-37. vidual layers of rat visual cortex and the effect of Carlson M. D., Penney J. B. Jr and Young A. B. (1993) monocular deprivation. Int. J. DevL Neurosci. 12, 31-41. NMDA, AMPA, and benzodiazepine binding site changes Lavie, D., Glotter, E. and Shro, Y. (1965) Constituents of in Alzheimer's disease visual cortex. Neurobiol Agin9 14, Withania somnifera Dun., the structure of withaferin-A. J. 343-352. Chem. Soc. 7517-7531. Chattopadhyay S., Chaudhuri S. and Ghosal S. (1987) Acti- Nordberg A. (1992) Neuroreceptor changes in Alzheimer's vation of peritoneal macrophages by sitoindoside IV, an disease. Cerebrovasc. Brain Met. Rev. 4, 303 328. anti-ulcerogenic acysteryl glucoside from Musa paraRoBner S., Schliebs R. and Bigl V. (1994) Ibotenic acid lesion disiaca. Planta Med. 54, 16-19. of nucleus basalis magnocellularis differentially affects Chattopadhyay U., Das S., Guha S. and Ghosal S. (1987) cholinergic, glutamatergic and GABAergic markers in Activation of lymphocytes of normal and tumour-bearing cortical rat brain regions. Brain Res. 668, 85-99. mice by mangiferin, a naturally occurring glucoRoBner S., Schliebs R. and Bigl V. (1995) 192IgG-saporinsylxanthone. Cancer Lett. 37, 293-299. induced immunotoxic lesions of cholinergic basal foreDunnett S. B., Everitt B. J. and Robbins T. W. (1991) The brain system differentially affect glutamatergic and basal forebrain-cortical cholinergic system: interpreting GABAergic markers in cortical rat brain regions. Brain the functional consequences of excitotoxic lesions. Trends Res. 696, 165-176. Neurosci. 14, 495-501. Schliebs R., Fiest T., RoBner S. and Bigl V. (1994) Receptor Eckenstein F. P., Baughman R. W. and Quinn J. (1988) An function in cortical rat brain regions after lesion of nucleus anatomical study of cholinergic innervation in rat cerebral basalis. J. Neural Transm., Suppl. 44, 195-208. cortex. Neuroscience 25, 457-474. Schwartz J. C., Cortes C. L., Rose C., Quach T. T. and Fiigner A. (1973) Hemmung immunologisch bedingter Pollard H. (1983) Adaptive changes of neurotransmitter Entziindungen durch das Pflanzensteroid Withaferin A. receptor mechanisms in the central nervous system. Progr. Arzneim.-Forsch. (Drug Res.) 23, 932-935. Brain Res. 58, 117-129. Ghosal S. (1985) Steryl glycosides and acyl steryl glycosides Zilles, K. (1985) The Cortex of the Rat. A Stereotaxic Atlas. from Musa paradisiaca. Phytochemistry 24, 1807-1810. Springer, Berlin.