- Email: [email protected]
The protective effect of ganstigmine against amyloid beta25–35 neurotoxicity on chicken cortical neurons is independent from the cholinesterase inhibition
Neuroscience Letters 341 (2003) 181–184 www.elsevier.com/locate/neulet
The protective effect of ganstigmine against amyloid beta25 – 35 neurotoxicity on chicken cortical neurons is independent from the cholinesterase inhibition Manfred Windischa,*, Birgit Hutter-Paiera, Lidija Jerkovica, Bruno Imbimbob, Gino Villettib a JSW-Research Forschungslabor GmbH., Rankengasse 28, A-8020 Graz, Austria Chiesi Farmaceutici S.p.A., Department of Pharmacology, Via Palermo, 26/A, 43100 Parma, Italy
b
Received 13 December 2002; received in revised form 16 January 2003; accepted 16 January 2003
Abstract Ganstigmine (CHF2819), a novel genserine derived acetylcholinesterase inhibitor and its enantiomer CHF3360 have been investigated for neuroprotective activity in two different in vitro assay systems using isolated cortical neurons from 9 day old chicken embryos. In the first in vitro model cells were lesioned by growth factor deprivation for 8 days achieved by reduced serum supplementation (2%) to the tissue culture medium. In the second lesion model neurodegeneration due to the addition of pre-aggregated b-amyloid25 – 35 has been achieved. Neuronal viability of treated neurons evaluated with the 3-(4,5-dimethylthiazol-2-yl)-2,5,diphenyl tetrazolium bromid reduction assay was compared to that of untreated control cells. In a dose range between 1.0 and 10.0 mM both compounds significantly prevent progressive neuronal cell death due to growth factor deprivation. Furthermore Ganstigmine and its enantiomer in concentrations between 0.1 and 3 mM also significantly decrease neurodegeneration achieved by the addition of b-amyloid25 – 35 by approximately 50%. Dose response curves of both substances were identical concerning effect size and concentration. Because CHF3360 does not show any acetylcholine inhibitor activity in the applied dose range it is concluded that Ganstigmine provides significant neuroprotection independent from its cholinergic activity. q 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ganstigmine; Alzheimer’s disease; Acetylcholinesterase inhibitor; Growth factor; Amyloid toxicity; Acetylcholinesterase and butyrylcholinesterase activity
The majority of drugs approved for the treatment of symptoms of Alzheimer’s disease (AD) belong to the class of cholinesterase inhibitors [3]. All of them achieve reproducible but rather modest improvement of cognitive performance, global function and activities of daily living in patients suffering from AD [6]. The modulation of amyloid precursor protein (APP) processing by stimulation of muscarinic receptors stimulated the hope that these drugs can provide disease modification, slowing down the progression of neurodegeneration. Data from clinical studies, even after long-term use of cholinesterase inhibitors, do not clearly support an influence on disease progression [4]. Therefore new cholinesterase inhibitors are under development which should provide additional neuroprotective activity. Such substances could achieve an * Corresponding author. Tel.: þ 43-316-765114; fax: þ43-316-7651144. E-mail address: [email protected] (M. Windisch).
acute cognitive improvement accompanied by long-term stabilization of the patients. Ganstigmine (CHF2819) is a novel geneserine derivative with acetylcholinesterase (AChE) inhibitory activity [11]. The drug produces marked increase in acetylcholine levels in the hippocampus of both young adult and aged rats and it attenuates scopolamine induced amnesia. In addition it provides increase of hippocampal 5-hydroxy tryptamine (5HT) levels. Due to this concomitant increase of acetylcholine and 5-HT levels in the hippocampus it is expected that CHF2819 can also positively influence depressive symptoms which are frequently reported in AD patients [13]. It was of interest to investigate whether this substance exerts additional effects on neurons subjected to disease relevant lesion conditions. In all experiments the CHF2819 enantiomer CHF3360 – lacking cholinesterase inhibitor activity and having a diminished butyrylcholinesterase (BChE) activity as shown here – was used as control to
0304-3940/03/$ - see front matter q 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(03)00125-3
182
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
assess the contribution of the cholinergic activity to neuroprotection. Two mechanisms which are discussed as factors possibly responsible for onset of neurodegeneration in AD were used to induce neuronal death in primary cultures of embryonic chicken cortical neurons. The first simulates disturbances in neurotrophic factor availability as a causative factor for onset of disease achieved by reduced serum supplementation to the tissue culture medium. This manipulation induces progressive apoptotic cell death. In a second series of experiments neurons were damaged by the addition of pre-aggregated b-Amyloid25 – 35 to the tissue cultures on day 8 in vitro, destroying approximately 50% of the neurons within 48 h of incubation. For evaluation of the inhibitory effects of CHF2819 and of its enantiomer on AChE as well as on BChE the modified Ellman’s method was used [2]. The source for BChE was human serum, while in case of AChE human recombinant AChE from embryonic kidney-cells was used. The principle of this method is based on the production of thiocholine from acetylthiocholine-iodide or butyrylthiocholine-iodide by AChE or BChE (both from Sigma), respectively. Thiocholine reacts with the reagent 5,50 -dithiobis 2-nitrobenzoic acid (DTNB), the so-called Ellman’s reagent, to 5-mercaptothiocholine-2-nitrobenzoate 5thio-2-nitrobenzoate to produce a yellow color. The rate of hydrolysis by AChE or BChE has been monitored spectrophotometrically. Appropriate amounts of the enzymes AChE or BChE, 500 mM of the substrate acetylthiocholine-iodide or butyrylthiocholine-iodide and 300 mM of the Ellman’s reagent were diluted in phosphate buffer pH 8 (for AChE) or pH 7.2 (for BChE) and were freshly prepared and used within 6 h. To each well of a 96 well plate 50 ml of CHF2819 and of its enantiomer solution as well as 100 ml DTNB and 100 ml enzyme were added. For the so-called buffer control (negative control), instead of enzyme 100 ml of the appropriate buffer was added. After 10 min of incubation at 37 8C directly in the plate-reader, 50 ml substrate, either acetylthiocholine-iodide or butyrylthiocholine-iodid, was added to each well. To avoid temperature variation, this substrate was preheated at 37 8C for about 15 min. Immediately after addition of the substrate, enzyme activity measurement was performed in the plate-reader in 10 cycles at 405 nm and within 8.5 min. While CHF2819 has the potential to inhibit BChE (IC50 is the concentration needed to inhibit 50% of the enzyme activity: 68 mM) but especially AChE (IC50: 1 mM) its enantiomer CHF3360 very slightly inhibits BChE (IC50: 175 mM) but not AChE (IC50: 298 mM). These results show that the affinity of the enantiomer CHF3360 to the AChE receptor is extremely low. The primary cultures of embryonic cortical neurons have been described in detail elsewhere [8]. In short, for the growth factor withdrawal assay brains were taken from chicken embryos on embryonic day E9 since in this stage of the development the brain almost exclusively contains nerve cells and less than 5% glia [10].
The hemispheres have been collected and cleaned from all loose tissue and remaining meningeal membranes. Thereafter they have been dissociated mechanically and then the isolated neurons have been seeded into 96-well tissue culture plates pre-coated with poly-D -lysin. The pure neurons were maintained in Eagle’s minimum essential medium with 1 g glucose/l and 2 mM L -glutamine supplementation and 0.01% gentamycin. Only 2% fetal calf serum was added to simulate conditions of reduced trophic factor availability. The preparation of neurons for the Ab25 – 35 lesion assay was identical with the exception that brains have been obtained on embryonic day 8 and that the neurons have been maintained in Dulbecco’s modified Eagle’s medium containing 4,5 g glucose/l, 2 mM L -glutamine, 5% NU-serum supplementation and 0.01% gentamycin. The lesion was performed by addition of 10 mM Ab25 – 35 which has been pre-aggregated for at least 72 h at room temperature. Unlesioned neurons were used as positive controls to assess the normal survival rate. For all experiments the neurons have been maintained at 37 8C, 95% relative humidity and 5% CO2. CHF2819 and the enantiomer CHF3360 have been dissolved in tissue culture medium and added to the neurons at the first day in vitro in dosages ranged from 0.1 – 300 mM final concentrations. The substances or for control purpose phosphate buffered saline remained with the cells until end of the experiment. Then the viability of the remaining neurons has been measured with the colorimetric MTT (3(4,5-dimethylthiazol-2-yl)-2,5,diphenyl tetrazolium bromid) reduction assay assessing the metabolic activity of surviving cells. MTT is cleaved by mitochondrial dehydrogenases to produce dark blue formazan crystals. Their concentration is proportional to the number of viable cells and can be measured with a plate reader at 570 nm [7]. In the growth factor withdrawal assay both substances show identical neuroprotective/neurotrophic potency (Fig. 1), increasing the neuronal survival rate almost 3-fold at a dose of 3 mM compared to the untreated control cells. Also the 1 mM and the 10 mM dosages result in slight neuroprotection; all other investigated dosages do not show any drug specific effect. On the other side from these data it can be concluded that the substances have no neurotoxic properties over the whole range of applied dosages. In the Ab25 – 35 lesion assay again both substances display identical dose response protecting neurons from the dose range of 0.1– 3 mM (Fig. 2). Addition of Ab25 – 35 to the tissue cultures decreased viability to about 50% in the untreated control cells. CHF2819 and CHF3360 increased the viability to 74.4% of the unlesioned controls, what means that it protected 50% of the neurons from Ab25 – 35 toxicity. In contrast to the growth factor withdrawal assay dosages of 10 mM and higher resulted in additional neuronal damage. However, considering the IC50 of CHF2819 for cholinesterase inhibition of 0.47– 1 mM [11] these high concentrations applied to the tissue cultures are far above
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
183
Fig. 1. Effects of CHF2819 (grey) and its enantiomer (CHF3360; hatched) in an apoptosis inducing low serum assay in vitro. Chick primary cortical neurons have been maintained under low serum (2% FCS) conditions for 8 days. Neuronal viability of surviving neurons has been obtained with the MTT assay. Y-axis: optical density [OD] measurement at 570 nm; x-axis: substance concentrations investigated whereby 0 stands for the control raised under the same low serum conditions but without CHF2619 or CHF3360. Values are shown as mean þ SEM (n . 12). Statistical differences versus control were calculated with Student’s t-test: *P # 0:05; **P # 0:005; ***P # 0:001.
the therapeutic concentration. Because already with the lowest dosage investigated a significant protective activity has been obtained it can be speculated that a neuroprotective potential against Ab toxicity can be achieved already in dose ranges without a relevant cholinesterase inhibition. It is remarkable that CHF2819 and its enantiomer show almost identical dose response profiles in two different lesion models. Since the enantiomer CHF3360 does not inhibit AChE in the investigated dose range at all it can be clearly concluded that the neuroprotective/neurotrophic activity of these substances is completely independent from their cholinesterase inhibition. It would be of substantial interest to explore in detail the molecular pathways resulting in neuroprotection. So far there are no reports published about neurotrophic activity of other cholinesterase inhibitors in a growth factor withdrawal model which has been used in the current experiments. However, a comparison to galantamine (data not shown) in our laboratory displayed also a slight effect but only in high (300 mM) concentration. However, it is evident that Huperzine A and Donepezil can protect
pheochromocytoma 12 (PC12) cells against oxygen glucose deprivation lesion [18], simulating conditions of ischemic brain damage. It is claimed that these drugs are reducing radical stress by decreasing superoxide-dismutase activity. Both substances show equal efficacy in spite of considerable differences in their cholinesterase inhibitor activity. An alternative explanation for the protection by Huperzine A could be its interaction with N-methyl-D -aspartate (NMDA) glutamate subtype receptor inhibiting in a non competitive way the ion channel PCP site [5]. Inhibition of NMDA receptor has been also reported for tetrahydroaminoacridine as shown in tissue culture experiments using murine cortical cell cultures exposed to 500 mM NMDA [1]. Also for this drug an interaction with the phencyclidine receptor has been speculated. Huperzine A displays also protection against hydrogen peroxide induced injury in rat PC12 cells [16]. In these experiments it was reported that the drug is elevating the activities of catalase and glutathione peroxidase. But the anti-apoptotic effect has been also related in a drug induced overexpression of apoptosis related genes [14]. The protection against Ab toxicity is of particular interest for
Fig. 2. Neuroprotective effects of CHF2819 (grey) and its enantiomer (CHF3360; hatched) obtained in a b-amyloid lesion assay in vitro. Chick primary cortical neurons have been lesioned at the 8th day in vitro with 10 mM b-amyloid25 – 35 for 48 h. Neuronal viability of surviving neurons has been obtained with the MTT assay. Y-axis: optical density [OD] measurement at 570 nm; x-axis: substance concentrations whereby C stands for the unlesioned control [black column] and 0 for the lesioned control [white column] raised under the same low serum conditions but without CHF2619 or CHF3360. Values are shown as mean þ SEM (n . 12). Statistical differences versus lesioned control were calculated with Student’s t-test: ***P # 0:001.
184
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
drugs used to treat AD. There is a vast body of evidence claiming that disturbances in the processing of APP producing highly agreeable Ab fragments to be the leading pathogenetic factor of AD [9]. CHF2819 and the enantiomer CHF3360 efficiently protect against the toxic properties of aggregated amyloid in the current investigation. Also Tacrine, Donepezil and Huperzine A can attenuate the neurotoxic effects of Ab25 – 35 in rat PC12 cells [12,17]. In spite of differences in performing the lesion and different dosages of Ab25 – 35 a similar extent of neuronal damage has been reported in these papers comparable to our results. Tacrine and Donepezil protect PC12 cells in a wide dose range, but the effect is also depending on the applied Ab25 – 35 concentration, resulting in a modest effect after addition of 1 mM Ab25 – 35 [12]. In the present study primary neurons were exposed to 1 mM of Ab25 – 35 resulting in a 50% reduction of cell viability. The achieved protection resulting in a survival rate of more than 75% advocates for a higher effect size of CHF2819 and its derivative. Compared to the discussed compounds the effect of Huperzine A is different because only 1 mM and 10 mM dosages increase cell viability [15]. Also in this assay Huperzine effects are related to its anti-oxidative properties whereas it is speculated that Tacrine and Donepezil achieve neuroprotection via interaction with nicotinic receptors. Obviously CHF2819 displays the significant neuroprotection via mechanisms independent from its cholinergic effect. The results obtained in the growth factor withdrawal assay indicate that both geneserine derivatives are able to mimic activities of trophic factors. It can be speculated that the anti-amyloid effect of these compounds results from interaction with trophic factor signaling pathways. It is of importance to note that the efficacious dosages are within the therapeutic range of CHF2819. Therefore an influence on disease progression beyond the acute cognitive effect of cholinesterase inhibition can be expected in AD patients. The missing relation between cholinergic effect and neuroprotection opens a wider range of possible therapeutic use of geneserine derivatives.
References [1] C.J. Davenport, H. Monyer, D.W. Choi, Tetrahydroaminoacridine selectively attenuates NMDA receptor-mediated neurotoxicity, Eur. J. Pharmacol. 154 (1988) 73 –78. [2] G.L. Ellman, K.D. Courtney, U.V. Andres Jr, R.M. Featherstone, A
[3] [4] [5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
new and rapid colorimetric determination of acetylcholinesterase activity, Biochem. Pharmacol. 7 (1961) 88 –95. E. Giacobini, Cholinesterase inhibitor therapy stabilizes symptoms of Alzheimer disease, Alz. Dis. Assoc. Disord. 14 (2000) 3–10. E. Giacobini, Is anti-cholinesterase therapy of Alzheimer’s disease delaying progression?, Aging Clin. Exp. Res. 13 (3) (2001) 247 –254. R.K. Gordon, S.V. Nigam, J.A. Weitz, J.R. Dave, B.P. Doctor, H.S. Ved, The NMDA receptor ion channel: a site for binding of Huperzine A, J. Appl. Toxicol. 21(Suppl. 1) (2001) 47 –51. J. Grutzendler, J.C. Morris, Cholinesterase inhibitors for Alzheimer’s disease, Drugs 61 (1) (2001) 41 –52. M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill, J. Immunol. Methods 119 (1989) 203– 210. B. Hutter-Paier, E. Grygar, M. Fru¨hwirth, I. Temmel, M. Windisch, Further evidence that Cerebrolysin protects cortical neurons from neurodegeneration in vitro, J. Neural. Transm., Suppl. 53 (1998) 363 –372. R. Le, L. Cruz, B. Urbanc, R.B. Knowles, K. Hsiao-Ashe, K. Duff, M.C. Irizarry, H.E. Stanley, B.T. Hyman, Plaque-induced abnormalities in neurite geometry in transgenic models of Alzheimer disease: implications for neural system disruption, J. Neuropathol. Exp. Neurol. 60 (8) (2001) 753 –758. B. Pettmann, J.C. Louis, L. Sensenbrenner, Morphological and biochemical maturation of neurons cultured in the absence of glial cells, Nature 28 (1979) 378– 380. C. Pietra, G. Villetti, P.T. Bolzoni, M. Delcanale, G. Amari, P. Caruso, M. Civelli, Cholinesterase inhibitory activity of new Geneserine derivatives, in: K. Iqbal, D.F. Swaab, B.B. Winblad, H.M. Wisnievski (Eds.), Alzheimer’s Disease and Related Disorders, John Wiley & Sons Ltd, Chichester, 1999, pp. 689–696. A.L. Svensson, A. Nordberg, Tacrine and donepezil attenuate the neurotoxic effect of A beta(25 – 35) in rat PC12 cells, NeuroReport 9 (1998) 1519–1522. L. Trabace, T. Cassano, L. Steardo, C. Pietra, G. Villetti, K.M. Kendrick, V. Cuomo, Biochemical and neurobehavioral profile of CHF2819, a novel, orally active acetylcholinesterase inhibitor for Alzheimer’s disease, J. Pharmacol. Exp. Ther. 294[1 (2000) 187 –194. R. Wang, X.Q. Xiao, X.C. Tang, Huperzine A attenuates hydrogen peroxide-induced apoptosis by regulating expression of apoptosisrelated genes in rat PC12 cells, NeuroReport 12 (12) (2001) 2629–2634. X.Q. Xiao, R. Wang, Y.F. Han, X.C. Tang, Protective effects of huperzine A on b-amyloid(25 – 35) induced oxidative injury in rat pheochromocytoma cells, Neurosci. Lett. 286 (3) (2000) 155 –158. X.Q. Xiao, J.W. Yang, X.C. Tang, Huperzine A protects rat pheochromocytoma cells against hydrogen peroxide-induced injury, Neurosci. Lett. 275 (1999) 73–76. X.Q. Xiao, H.Y. Zhang, X.C. Tang, Huperzine A ttenuates amyloid beta-peptide fragment 25 – 35-induced apoptosis in rat cortical neurons via inhibiting reactive oxygen species formation and caspase-3 activation, J. Neurosci. Res. 67 (2002) 30–36. J. Zhou, Y. Fu, X.C. Tang, Huperzine A and donepezil protect rat pheochromocytoma cells against oxygen-glucose deprivation, Neurosci. Lett. 306 (1–2) (2001) 53–56.
The protective effect of ganstigmine against amyloid beta25 – 35 neurotoxicity on chicken cortical neurons is independent from the cholinesterase inhibition Manfred Windischa,*, Birgit Hutter-Paiera, Lidija Jerkovica, Bruno Imbimbob, Gino Villettib a JSW-Research Forschungslabor GmbH., Rankengasse 28, A-8020 Graz, Austria Chiesi Farmaceutici S.p.A., Department of Pharmacology, Via Palermo, 26/A, 43100 Parma, Italy
b
Received 13 December 2002; received in revised form 16 January 2003; accepted 16 January 2003
Abstract Ganstigmine (CHF2819), a novel genserine derived acetylcholinesterase inhibitor and its enantiomer CHF3360 have been investigated for neuroprotective activity in two different in vitro assay systems using isolated cortical neurons from 9 day old chicken embryos. In the first in vitro model cells were lesioned by growth factor deprivation for 8 days achieved by reduced serum supplementation (2%) to the tissue culture medium. In the second lesion model neurodegeneration due to the addition of pre-aggregated b-amyloid25 – 35 has been achieved. Neuronal viability of treated neurons evaluated with the 3-(4,5-dimethylthiazol-2-yl)-2,5,diphenyl tetrazolium bromid reduction assay was compared to that of untreated control cells. In a dose range between 1.0 and 10.0 mM both compounds significantly prevent progressive neuronal cell death due to growth factor deprivation. Furthermore Ganstigmine and its enantiomer in concentrations between 0.1 and 3 mM also significantly decrease neurodegeneration achieved by the addition of b-amyloid25 – 35 by approximately 50%. Dose response curves of both substances were identical concerning effect size and concentration. Because CHF3360 does not show any acetylcholine inhibitor activity in the applied dose range it is concluded that Ganstigmine provides significant neuroprotection independent from its cholinergic activity. q 2003 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Ganstigmine; Alzheimer’s disease; Acetylcholinesterase inhibitor; Growth factor; Amyloid toxicity; Acetylcholinesterase and butyrylcholinesterase activity
The majority of drugs approved for the treatment of symptoms of Alzheimer’s disease (AD) belong to the class of cholinesterase inhibitors [3]. All of them achieve reproducible but rather modest improvement of cognitive performance, global function and activities of daily living in patients suffering from AD [6]. The modulation of amyloid precursor protein (APP) processing by stimulation of muscarinic receptors stimulated the hope that these drugs can provide disease modification, slowing down the progression of neurodegeneration. Data from clinical studies, even after long-term use of cholinesterase inhibitors, do not clearly support an influence on disease progression [4]. Therefore new cholinesterase inhibitors are under development which should provide additional neuroprotective activity. Such substances could achieve an * Corresponding author. Tel.: þ 43-316-765114; fax: þ43-316-7651144. E-mail address: [email protected] (M. Windisch).
acute cognitive improvement accompanied by long-term stabilization of the patients. Ganstigmine (CHF2819) is a novel geneserine derivative with acetylcholinesterase (AChE) inhibitory activity [11]. The drug produces marked increase in acetylcholine levels in the hippocampus of both young adult and aged rats and it attenuates scopolamine induced amnesia. In addition it provides increase of hippocampal 5-hydroxy tryptamine (5HT) levels. Due to this concomitant increase of acetylcholine and 5-HT levels in the hippocampus it is expected that CHF2819 can also positively influence depressive symptoms which are frequently reported in AD patients [13]. It was of interest to investigate whether this substance exerts additional effects on neurons subjected to disease relevant lesion conditions. In all experiments the CHF2819 enantiomer CHF3360 – lacking cholinesterase inhibitor activity and having a diminished butyrylcholinesterase (BChE) activity as shown here – was used as control to
0304-3940/03/$ - see front matter q 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0304-3940(03)00125-3
182
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
assess the contribution of the cholinergic activity to neuroprotection. Two mechanisms which are discussed as factors possibly responsible for onset of neurodegeneration in AD were used to induce neuronal death in primary cultures of embryonic chicken cortical neurons. The first simulates disturbances in neurotrophic factor availability as a causative factor for onset of disease achieved by reduced serum supplementation to the tissue culture medium. This manipulation induces progressive apoptotic cell death. In a second series of experiments neurons were damaged by the addition of pre-aggregated b-Amyloid25 – 35 to the tissue cultures on day 8 in vitro, destroying approximately 50% of the neurons within 48 h of incubation. For evaluation of the inhibitory effects of CHF2819 and of its enantiomer on AChE as well as on BChE the modified Ellman’s method was used [2]. The source for BChE was human serum, while in case of AChE human recombinant AChE from embryonic kidney-cells was used. The principle of this method is based on the production of thiocholine from acetylthiocholine-iodide or butyrylthiocholine-iodide by AChE or BChE (both from Sigma), respectively. Thiocholine reacts with the reagent 5,50 -dithiobis 2-nitrobenzoic acid (DTNB), the so-called Ellman’s reagent, to 5-mercaptothiocholine-2-nitrobenzoate 5thio-2-nitrobenzoate to produce a yellow color. The rate of hydrolysis by AChE or BChE has been monitored spectrophotometrically. Appropriate amounts of the enzymes AChE or BChE, 500 mM of the substrate acetylthiocholine-iodide or butyrylthiocholine-iodide and 300 mM of the Ellman’s reagent were diluted in phosphate buffer pH 8 (for AChE) or pH 7.2 (for BChE) and were freshly prepared and used within 6 h. To each well of a 96 well plate 50 ml of CHF2819 and of its enantiomer solution as well as 100 ml DTNB and 100 ml enzyme were added. For the so-called buffer control (negative control), instead of enzyme 100 ml of the appropriate buffer was added. After 10 min of incubation at 37 8C directly in the plate-reader, 50 ml substrate, either acetylthiocholine-iodide or butyrylthiocholine-iodid, was added to each well. To avoid temperature variation, this substrate was preheated at 37 8C for about 15 min. Immediately after addition of the substrate, enzyme activity measurement was performed in the plate-reader in 10 cycles at 405 nm and within 8.5 min. While CHF2819 has the potential to inhibit BChE (IC50 is the concentration needed to inhibit 50% of the enzyme activity: 68 mM) but especially AChE (IC50: 1 mM) its enantiomer CHF3360 very slightly inhibits BChE (IC50: 175 mM) but not AChE (IC50: 298 mM). These results show that the affinity of the enantiomer CHF3360 to the AChE receptor is extremely low. The primary cultures of embryonic cortical neurons have been described in detail elsewhere [8]. In short, for the growth factor withdrawal assay brains were taken from chicken embryos on embryonic day E9 since in this stage of the development the brain almost exclusively contains nerve cells and less than 5% glia [10].
The hemispheres have been collected and cleaned from all loose tissue and remaining meningeal membranes. Thereafter they have been dissociated mechanically and then the isolated neurons have been seeded into 96-well tissue culture plates pre-coated with poly-D -lysin. The pure neurons were maintained in Eagle’s minimum essential medium with 1 g glucose/l and 2 mM L -glutamine supplementation and 0.01% gentamycin. Only 2% fetal calf serum was added to simulate conditions of reduced trophic factor availability. The preparation of neurons for the Ab25 – 35 lesion assay was identical with the exception that brains have been obtained on embryonic day 8 and that the neurons have been maintained in Dulbecco’s modified Eagle’s medium containing 4,5 g glucose/l, 2 mM L -glutamine, 5% NU-serum supplementation and 0.01% gentamycin. The lesion was performed by addition of 10 mM Ab25 – 35 which has been pre-aggregated for at least 72 h at room temperature. Unlesioned neurons were used as positive controls to assess the normal survival rate. For all experiments the neurons have been maintained at 37 8C, 95% relative humidity and 5% CO2. CHF2819 and the enantiomer CHF3360 have been dissolved in tissue culture medium and added to the neurons at the first day in vitro in dosages ranged from 0.1 – 300 mM final concentrations. The substances or for control purpose phosphate buffered saline remained with the cells until end of the experiment. Then the viability of the remaining neurons has been measured with the colorimetric MTT (3(4,5-dimethylthiazol-2-yl)-2,5,diphenyl tetrazolium bromid) reduction assay assessing the metabolic activity of surviving cells. MTT is cleaved by mitochondrial dehydrogenases to produce dark blue formazan crystals. Their concentration is proportional to the number of viable cells and can be measured with a plate reader at 570 nm [7]. In the growth factor withdrawal assay both substances show identical neuroprotective/neurotrophic potency (Fig. 1), increasing the neuronal survival rate almost 3-fold at a dose of 3 mM compared to the untreated control cells. Also the 1 mM and the 10 mM dosages result in slight neuroprotection; all other investigated dosages do not show any drug specific effect. On the other side from these data it can be concluded that the substances have no neurotoxic properties over the whole range of applied dosages. In the Ab25 – 35 lesion assay again both substances display identical dose response protecting neurons from the dose range of 0.1– 3 mM (Fig. 2). Addition of Ab25 – 35 to the tissue cultures decreased viability to about 50% in the untreated control cells. CHF2819 and CHF3360 increased the viability to 74.4% of the unlesioned controls, what means that it protected 50% of the neurons from Ab25 – 35 toxicity. In contrast to the growth factor withdrawal assay dosages of 10 mM and higher resulted in additional neuronal damage. However, considering the IC50 of CHF2819 for cholinesterase inhibition of 0.47– 1 mM [11] these high concentrations applied to the tissue cultures are far above
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
183
Fig. 1. Effects of CHF2819 (grey) and its enantiomer (CHF3360; hatched) in an apoptosis inducing low serum assay in vitro. Chick primary cortical neurons have been maintained under low serum (2% FCS) conditions for 8 days. Neuronal viability of surviving neurons has been obtained with the MTT assay. Y-axis: optical density [OD] measurement at 570 nm; x-axis: substance concentrations investigated whereby 0 stands for the control raised under the same low serum conditions but without CHF2619 or CHF3360. Values are shown as mean þ SEM (n . 12). Statistical differences versus control were calculated with Student’s t-test: *P # 0:05; **P # 0:005; ***P # 0:001.
the therapeutic concentration. Because already with the lowest dosage investigated a significant protective activity has been obtained it can be speculated that a neuroprotective potential against Ab toxicity can be achieved already in dose ranges without a relevant cholinesterase inhibition. It is remarkable that CHF2819 and its enantiomer show almost identical dose response profiles in two different lesion models. Since the enantiomer CHF3360 does not inhibit AChE in the investigated dose range at all it can be clearly concluded that the neuroprotective/neurotrophic activity of these substances is completely independent from their cholinesterase inhibition. It would be of substantial interest to explore in detail the molecular pathways resulting in neuroprotection. So far there are no reports published about neurotrophic activity of other cholinesterase inhibitors in a growth factor withdrawal model which has been used in the current experiments. However, a comparison to galantamine (data not shown) in our laboratory displayed also a slight effect but only in high (300 mM) concentration. However, it is evident that Huperzine A and Donepezil can protect
pheochromocytoma 12 (PC12) cells against oxygen glucose deprivation lesion [18], simulating conditions of ischemic brain damage. It is claimed that these drugs are reducing radical stress by decreasing superoxide-dismutase activity. Both substances show equal efficacy in spite of considerable differences in their cholinesterase inhibitor activity. An alternative explanation for the protection by Huperzine A could be its interaction with N-methyl-D -aspartate (NMDA) glutamate subtype receptor inhibiting in a non competitive way the ion channel PCP site [5]. Inhibition of NMDA receptor has been also reported for tetrahydroaminoacridine as shown in tissue culture experiments using murine cortical cell cultures exposed to 500 mM NMDA [1]. Also for this drug an interaction with the phencyclidine receptor has been speculated. Huperzine A displays also protection against hydrogen peroxide induced injury in rat PC12 cells [16]. In these experiments it was reported that the drug is elevating the activities of catalase and glutathione peroxidase. But the anti-apoptotic effect has been also related in a drug induced overexpression of apoptosis related genes [14]. The protection against Ab toxicity is of particular interest for
Fig. 2. Neuroprotective effects of CHF2819 (grey) and its enantiomer (CHF3360; hatched) obtained in a b-amyloid lesion assay in vitro. Chick primary cortical neurons have been lesioned at the 8th day in vitro with 10 mM b-amyloid25 – 35 for 48 h. Neuronal viability of surviving neurons has been obtained with the MTT assay. Y-axis: optical density [OD] measurement at 570 nm; x-axis: substance concentrations whereby C stands for the unlesioned control [black column] and 0 for the lesioned control [white column] raised under the same low serum conditions but without CHF2619 or CHF3360. Values are shown as mean þ SEM (n . 12). Statistical differences versus lesioned control were calculated with Student’s t-test: ***P # 0:001.
184
M. Windisch et al. / Neuroscience Letters 341 (2003) 181–184
drugs used to treat AD. There is a vast body of evidence claiming that disturbances in the processing of APP producing highly agreeable Ab fragments to be the leading pathogenetic factor of AD [9]. CHF2819 and the enantiomer CHF3360 efficiently protect against the toxic properties of aggregated amyloid in the current investigation. Also Tacrine, Donepezil and Huperzine A can attenuate the neurotoxic effects of Ab25 – 35 in rat PC12 cells [12,17]. In spite of differences in performing the lesion and different dosages of Ab25 – 35 a similar extent of neuronal damage has been reported in these papers comparable to our results. Tacrine and Donepezil protect PC12 cells in a wide dose range, but the effect is also depending on the applied Ab25 – 35 concentration, resulting in a modest effect after addition of 1 mM Ab25 – 35 [12]. In the present study primary neurons were exposed to 1 mM of Ab25 – 35 resulting in a 50% reduction of cell viability. The achieved protection resulting in a survival rate of more than 75% advocates for a higher effect size of CHF2819 and its derivative. Compared to the discussed compounds the effect of Huperzine A is different because only 1 mM and 10 mM dosages increase cell viability [15]. Also in this assay Huperzine effects are related to its anti-oxidative properties whereas it is speculated that Tacrine and Donepezil achieve neuroprotection via interaction with nicotinic receptors. Obviously CHF2819 displays the significant neuroprotection via mechanisms independent from its cholinergic effect. The results obtained in the growth factor withdrawal assay indicate that both geneserine derivatives are able to mimic activities of trophic factors. It can be speculated that the anti-amyloid effect of these compounds results from interaction with trophic factor signaling pathways. It is of importance to note that the efficacious dosages are within the therapeutic range of CHF2819. Therefore an influence on disease progression beyond the acute cognitive effect of cholinesterase inhibition can be expected in AD patients. The missing relation between cholinergic effect and neuroprotection opens a wider range of possible therapeutic use of geneserine derivatives.
References [1] C.J. Davenport, H. Monyer, D.W. Choi, Tetrahydroaminoacridine selectively attenuates NMDA receptor-mediated neurotoxicity, Eur. J. Pharmacol. 154 (1988) 73 –78. [2] G.L. Ellman, K.D. Courtney, U.V. Andres Jr, R.M. Featherstone, A
[3] [4] [5]
[6] [7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
new and rapid colorimetric determination of acetylcholinesterase activity, Biochem. Pharmacol. 7 (1961) 88 –95. E. Giacobini, Cholinesterase inhibitor therapy stabilizes symptoms of Alzheimer disease, Alz. Dis. Assoc. Disord. 14 (2000) 3–10. E. Giacobini, Is anti-cholinesterase therapy of Alzheimer’s disease delaying progression?, Aging Clin. Exp. Res. 13 (3) (2001) 247 –254. R.K. Gordon, S.V. Nigam, J.A. Weitz, J.R. Dave, B.P. Doctor, H.S. Ved, The NMDA receptor ion channel: a site for binding of Huperzine A, J. Appl. Toxicol. 21(Suppl. 1) (2001) 47 –51. J. Grutzendler, J.C. Morris, Cholinesterase inhibitors for Alzheimer’s disease, Drugs 61 (1) (2001) 41 –52. M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill, J. Immunol. Methods 119 (1989) 203– 210. B. Hutter-Paier, E. Grygar, M. Fru¨hwirth, I. Temmel, M. Windisch, Further evidence that Cerebrolysin protects cortical neurons from neurodegeneration in vitro, J. Neural. Transm., Suppl. 53 (1998) 363 –372. R. Le, L. Cruz, B. Urbanc, R.B. Knowles, K. Hsiao-Ashe, K. Duff, M.C. Irizarry, H.E. Stanley, B.T. Hyman, Plaque-induced abnormalities in neurite geometry in transgenic models of Alzheimer disease: implications for neural system disruption, J. Neuropathol. Exp. Neurol. 60 (8) (2001) 753 –758. B. Pettmann, J.C. Louis, L. Sensenbrenner, Morphological and biochemical maturation of neurons cultured in the absence of glial cells, Nature 28 (1979) 378– 380. C. Pietra, G. Villetti, P.T. Bolzoni, M. Delcanale, G. Amari, P. Caruso, M. Civelli, Cholinesterase inhibitory activity of new Geneserine derivatives, in: K. Iqbal, D.F. Swaab, B.B. Winblad, H.M. Wisnievski (Eds.), Alzheimer’s Disease and Related Disorders, John Wiley & Sons Ltd, Chichester, 1999, pp. 689–696. A.L. Svensson, A. Nordberg, Tacrine and donepezil attenuate the neurotoxic effect of A beta(25 – 35) in rat PC12 cells, NeuroReport 9 (1998) 1519–1522. L. Trabace, T. Cassano, L. Steardo, C. Pietra, G. Villetti, K.M. Kendrick, V. Cuomo, Biochemical and neurobehavioral profile of CHF2819, a novel, orally active acetylcholinesterase inhibitor for Alzheimer’s disease, J. Pharmacol. Exp. Ther. 294[1 (2000) 187 –194. R. Wang, X.Q. Xiao, X.C. Tang, Huperzine A attenuates hydrogen peroxide-induced apoptosis by regulating expression of apoptosisrelated genes in rat PC12 cells, NeuroReport 12 (12) (2001) 2629–2634. X.Q. Xiao, R. Wang, Y.F. Han, X.C. Tang, Protective effects of huperzine A on b-amyloid(25 – 35) induced oxidative injury in rat pheochromocytoma cells, Neurosci. Lett. 286 (3) (2000) 155 –158. X.Q. Xiao, J.W. Yang, X.C. Tang, Huperzine A protects rat pheochromocytoma cells against hydrogen peroxide-induced injury, Neurosci. Lett. 275 (1999) 73–76. X.Q. Xiao, H.Y. Zhang, X.C. Tang, Huperzine A ttenuates amyloid beta-peptide fragment 25 – 35-induced apoptosis in rat cortical neurons via inhibiting reactive oxygen species formation and caspase-3 activation, J. Neurosci. Res. 67 (2002) 30–36. J. Zhou, Y. Fu, X.C. Tang, Huperzine A and donepezil protect rat pheochromocytoma cells against oxygen-glucose deprivation, Neurosci. Lett. 306 (1–2) (2001) 53–56.