Journal of

http://jop.sagepub.com/

Anti-amnesic and neuroprotective potentials of the mixed muscarinic /sigma 1 (σ1) ANAVEX2-73, a novel aminotetrahydrofuran derivative Vanessa Villard, Julie Espallergues, Emeline Keller, Alexandre Vamvakides and Tangui Maurice J Psychopharmacol published online 9 September 2010 DOI: 10.1177/0269881110379286

The online version of this article can be found at: http://jop.sagepub.com/content/early/2010/09/08/0269881110379286

Published by:

http://www.sagepublications.com

On behalf of:

British Association for Psychopharmacology

Additional services and information for Journal of Psychopharmacology can be found at:

Email Alerts: http://jop.sagepub.com/cgi/alerts

Subscriptions: http://jop.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

Downloaded from jop.sagepub.com by guest on September 15, 2010 J Psychopharmacol OnlineFirst, published on September 9, 2010 as doi:10.1177/0269881110379286

Original Paper

Anti-amnesic and neuroprotective potentials of the mixed muscarinic receptor/sigma1 (p1) Journal of Psychopharmacology ligand ANAVEX2-73, a novel 0(0) 1–17 ! The Author(s) 2010 aminotetrahydrofuran derivative Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0269881110379286 jop.sagepub.com Vanessa Villard1,2,3, Julie Espallergues1,2,3, Emeline Keller1,2,3, Alexandre Vamvakides4 and Tangui Maurice1,2,3

Abstract

Tetrahydro-N,N-dimethyl-2,2-diphenyl-3-furanmethanamine hydrochloride (ANAVEX2-73) binds to muscarinic and sigma1 (s1) receptors with affinities in the low micromolar range. We characterized its anti-amnesic and neuroprotective potentials in pharmacological and pathological models. Spatial working was evaluated using spontaneous alternation in the Y-maze and non-spatial memory using passive avoidance procedures. ANAVEX2-73 (0.01–3.0 mg/kg i.p.) alleviated the - and -induced impairments. ANAVEX2-73 (300 mg/kg) also reversed the learning deficits in mice injected with Ab25-35 , a non-transgenic Alzheimer’s disease model. When the was injected simultaneously with Ab25-35, 7 days before the tests, it blocked the appearance of learning impairments. This protective activity was confirmed since

ANAVEX2-73 blocked the Ab25-35-induced oxidative stress in the hippocampus. This effect was differentially sensitive to the muscarinic scopolamine or the s1 antagonist BD1047, confirming the mixed muscarinic/s1 pharmacological action. Finally, its unique demethyl metabolite, ANAVEX19-144, was also effective and ANAVEX2-73 presented a longer duration of action, effective 12 h before Ab25-35, than its related compound ANAVEX1-41. The neuroprotective activity of ANAVEX2-73, its mixed /s1 activity, its low active dose range and its long duration of action together reinforce its therapeutic potential in Alzheimer’s disease.

Keywords

ANAVEX2-73, muscarinic receptors, sigma-1 protein, learning and memory, dizocilpine, scopolamine, amyloid Ab25-35 peptide

Introduction , in the of patients with Alzheimer’s disease The central cholinergic systems are involved in learning (AD). A well-characterized impairment of cholinergic neu- and memory processes. They represent valuable targets for rons within the septo-hippocampic pathway and nucleus drug therapy aiming at improving the cognitive deficits basalis magnocellularis is involved in the rapid loss of learn- associated with or neurodegenerative diseases. ing and memory in patients with AD (Erme et al., 1992; Cholinergic blocking agents such as scopolamine provoke, Perry et al., 1978). In particular, formation of Ab when injected systemically in rodents, impairments of learn- impairs the coupling of M1 receptors with G-proteins ing and memory, and pro-cholinergic , especially cho- (Fisher, 2000; Fisher et al., 2003). Present strategies in AD linesterase inhibitors, can ameliorate them durably (Fibiger, aim at improving or at least maintaining central cholinergic 1991; Mohammed, 1993). Moreover, specific acetylcholine functions, particularly by treatments with receptor subtypes are differentially involved in the effects of inhibitors. This allows a symptomatic alleviation of the cog- acetylcholine on cognition. Blockade of the muscarinic M1 nitive deficits, but also putatively an effective neuroprotec- subtype of is particularly associated tion. Several cholinesterase inhibitors, including tacrine, with memory impairments (Bymaster et al., 1993; Ohno et al., 1994). M2 receptor antagonists may be associated 1 with improvements in memory, including attenuation of the Inserm, U. 710, 34095 Montpellier, France. 2 effects due to M1 receptor blockade, by increasing acetylcho- University of Montpellier 2, 34095 Montpellier, France. 3EPHE, 75017 Paris, France. line release in forebrain structures (Baratti et al., 1993; 4Anavex life Sciences, Pallini, Greece. Bymaster et al., 1993; Vamvakides, 2002b). Cholinergic systems are very sensitive to the neurodegen- Corresponding author: erative associated with the presence of deposits of Dr Tangui Maurice, INSERM U 710, EPHE, University of Montpellier 2, c.c. aggregated amyloid-b (Ab) proteins and neurofibrillary tan- 105, place Euge`ne Bataillon, 34095 Montpellier cedex 5, France gles, composed of abnormal and hyper-phosphorylated Tau Email: [email protected]

Downloaded from jop.sagepub.com by guest on September 15, 2010 2 Journal of Psychopharmacology 0(0)

donepezil and attenuated Ab1-40-orAb25–35- Tetrahydro-N,N-dimethyl-2,2-diphenyl-3-furanmethanamine induced toxicity in rodent models (Arias et al., 2004; hydrochloride (ANAVEX2-73) is a related aminotetrahydro- Kihara et al., 2004; Meunier et al., 2006; Svensson and furan derivative also presenting a mixed muscarinic/s1 Nordberg, 1998). The therapeutic potential of M1 receptor protein profile, but with a better selectivity for the s1 subtype selective , directly activating post-synaptic M1 mus- as compared with s2 sites (Vamvakides, 2002a). Indeed, bind- carinic receptors in persistent cholinergic , has also ing analyses of the compound performed by CEREP (Paris, been reported (Fisher, 2000). France), showed an IC50 ¼ 860 nM for s1 and no affinity for New ways for enhancing brain cholinergic functions s2 receptors. Moreover, the screening profile performed by are also in development. Among them, ligands activating the MDS PharmaServices (King of Prussia, PA, USA) showed sigma-1 (s1) protein have been reported to act as anti-amnesic micromolar affinities for muscarinic M1–M4 receptors and neuroprotective agents (Maurice and Su, 2009; Maurice (3.3–5.2 mL), sodium channel site 2 (5.1 mM), and NMDA et al., 1994, 1997, 1999; Monnet and Maurice, 2006). The s1 receptors (8.0 mL). Negligible interaction with 60 other protein is an intracellular neuromodulatory protein recently receptor and assays was found (data not shown). identified as a ligand-operated molecular chaperone, present In the present study, we first examined the anti-amnesic prop- at -associated endoplasmic reticulum (ER) erties of ANAVEX2-73 in mice in which learning abilities membranes (Hayashi and Su, 2007). Anti-amnesic s1 protein have been impede using a systemic treatment with the mus- activators show potent anti-amnesic properties partly by facil- carinic non-selective antagonist scopolamine, the NMDA itating acetylcholine release in the forebrain cortex (Matsuno receptor antagonist dizocilpine, or in mice that received a et al., 1992, 1993). of s1 protein also modulates central administration of Ab25–35 peptide. Short-term spatial intracellular mobilization (Hayashi et al., 2000), memory was evaluated using the spontaneous alternation channel activation (Aydar et al., 2002; Martina et al., 2007) performance in the Y-maze and long-term non-spatial and membrane recomposition, particularly within lipid rafts memory using the step-through type passive avoidance pro- microdomains (Hayashi and Su, 2001). In turn, s1 protein cedure. Second, we examine the neuroprotective activity of agonists, including (þ)-SKF-10047, (þ)-, PRE- ANAVEX2-73 by injecting the drug intraperitoneally (i.p.) 084 or SA4503, have been shown to attenuate the learning at the same time as the intracerebroventricularly (i.c.v.) injec- impairments induced by scopolamine, by the N-methyl- tion of Ab25-35 peptide, 7 days before the behavioural or bio- D- (NMDA) receptor non-competitive antagonist chemical analyses. We analysed the learning deficits and dizocilpine (Maurice et al., 1994; Matsuno et al., 1997; Senda oxidative stress induction in the hippocampus. We used sco- et al., 1996, 1997), or in mice treated with the Ab25–35 peptide, a polamine and BD1047 as muscarinic and s1 receptor antag- non-transgenic model of AD (Maurice et al., 1998; Meunier onists, respectively, to analyse the involvement of these et al., 2006). The particular localization of s1 protein at the targets in the drug effect. Third, we analysed the activity of ER and function as sensor/regulator of Ca2þ homeostasis by the main metabolite of ANAVEX2-73 and ANAVEX1-41 regulating activity of inositol-1,2,4 trisphophate receptors and their putative involvement in the pharmacological (Hayashi and Su, 2001; Hayashi et al., 2000) may partly action of the parent drugs. explain the neuroprotective effects mediated by selective ago- nists. Indeed, several selective s1 protein agonists were reported to significantly protect neuronal damages in vitro or Materials and methods in vivo. For instance, PRE-084 prevented the cell death in pri- Animals mary neuronal cultures exposed to Ab25–35 peptide (Marrazzo et al., 2005) or the cell loss, oxidative stress or learning and Male Swiss OF-1 mice (Depre´, St Doulchard, France), aged memory deficits observed in mice exposed to Ab25–35 (Meunier 7–9 weeks and weighing 32 2 g were used in this study. et al., 2006). Animals were housed in plastic cages in groups. They had We previously described a new series of aminotetrahydro- free access to food and water, except during behavioural furan compounds showing prominent anti-amnesic, experiments, and they were kept in a regulated environment , and neuroprotective activities (23 1C, 40–60% humidity) under a 12 h light/dark cycle (Espallergues et al., 2007; Vamvakides, 2002a, 2002b; Villard (light on at 8:00 a.m.). Experiments were carried out between et al., 2009). Among them, tetrahydro-N,N-dimethyl-5, 9:00 a.m. and 5:00 p.m., in an experimental room within the 5-diphenyl-3-furanmethanamine hydrochloride (ANAVEX1- animal facility. Mice were habituated 30 min before each 41) is an M1 muscarinic acetylcholine receptor and an experiment. All animal procedures were conducted in strict M2/M3 receptor antagonist, with a median efficacy of 1 mM adherence of European Union Directive of 24 November (for M1, M2) and 0.3 mM (for M3). ANAVEX1-41 is a potent 1986 (86-609). anti-amnesic drug, acting not only through muscarinic activ- ity but also through s1 receptors, since its ameliorating effects Drugs and administration procedures against the scopolamine-induced learning deficits were blocked by pre-treatment with the s1 protein antagonist Tetrahydro-N,N-dimethyl-2,2-diphenyl-3-furanmethanamine BD1047 or an antisense oligodeoxynucleotide targeting the hydrochloride (ANAVEX2-73, Figure 1), tetrahydro-N, s1 protein (Espallergues et al., 2007). The compound also N-dimethyl-5,5-diphenyl-3-furanmethanamine hydrochloride protected against Ab25-35-induced toxicity in mice, measured (ANAVEX1-41, Figure 1), and their demethyl metabolites at the morphological, biochemical and behavioural levels (respectively, ANAVEX19-144 and ANAVEX2-140, Figure 1) (Villard et al., 2009). were all synthesized in the laboratory (Anavex Life

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 3

into all three arms on consecutive trials. The number of the CH3 HCI total possible alternations was therefore the total number of N CH CH3 arm entries minus two, and the percentage of alternation 3 N HCI was calculated as (actual alternations / total possible alterna- tions) 100. Animals performing less than eight arm entries CH3 O in 8 min were discarded from experimental groups. No ani- O mal reached this exclusion criteria in this study. The com- pounds were administered 30 min before the session or 10 min before dizocilpine/scopolamine, given 20 min before the session. ANAVEX2-73 ANAVEX1-41

Step-through passive avoidance test. Non-spatial long- term memory was assessed either using the step-through type CH3 CH3 NH NH passive avoidance procedure, as described elsewhere (Meunier et al., 2006), or the step-down type, as described elsewhere HCI (Maurice et al., 1996, 1998; Villard et al., 2009). The step- HCI O through type apparatus consisted of an illuminated compart- O ment with white polyvinylchloride walls (15 20 15 cm high), a darkened compartment with black polyvinylchloride walls (15 20 15 cm high) and a grid floor. A guillotine door separated each compartment. A 60 W lamp positioned ANAVEX19-144 ANAVEX2-140 40 cm above the apparatus lit the white compartment during the experimental period. Scrambled foot shocks (0.3 mA for 3 s) were delivered to the grid floor using a shock generator Figure 1. Developed formulae of aminotetrahydrofuran derivatives scrambler (Lafayette Instruments, Lafayette, MA, USA). The ANAVEX2-73 and ANAVEX1-41 and their unique demethyl metabolites, guillotine door was initially closed during the training session. ANAVEX19-144 and ANAVE2-140, respectively. Each mouse was placed into the white compartment. After 5 s the door was raised. When the mouse entered the darkened compartment and placed all its paws on the grid floor, the door was gently closed and the scrambled foot shock was Sciences, Greece). N-[2-(3,4-dichlorophenyl)ethyl]-N-methyl- delivered for 3 s. The step-through latency, i.e. the latency 2-(dimethylamino) hydrochloride (BD1047) was spent to enter the dark compartment, and the number of from Tocris (Bristol, UK). All other materials, including sco- vocalizations were recorded. The number of vocalizations polamine hydrobromide, were purchased from Sigma-Aldrich did not differ among groups, indicating that shock sensitivity (St Quentin-Fallavier, France). Drugs were solubilized in was unaffected by the i.c.v. or i.p. treatments (data not physiological saline solution and administered i.p. or subcu- shown). Animals that did not enter within the 3–30 s time taneously (s.c.) in a volume of 100 mL per 20 g body weight. range were discarded from experimental groups. Attrition The amyloid b [25–35] peptide (Ab25-35) and scrambled accounted for less than 2% of the animals and was indepen- Ab25-35 peptide (Sc.Ab) were from NeoMPS (Strasbourg, dent of the treatment. France) or Genepep (Prades-le-Lez, France). They were dis- The retention test was carried out 24 h after training. Each solved in sterile distilled water at a concentration of 3 mg/mL mouse was placed again into the white compartment. After 5 s and stored at 20C until use. Before injection, peptides were the door was raised. The step-through latency was recorded aggregated by incubation at 37C for 4 days. They were up to 300 s. Animals entered the darkened compartment or administered i.c.v. in a final volume of 3 mL per mouse, as were placed into it and the escape latency, i.e. the time spent previously described (Maurice et al., 1996, 1998; Meunier to return into the white compartment, was also measured up et al., 2006; Villard et al., 2009). Animals were used at to 300 s. The two parameters were measured although they do day 7 after injection for spontaneous alternation and at not rely on similar memory stimuli. The step-through latency day 8 for passive avoidance training. involves reinforced stimuli and is the direct measure of pas- sive avoidance behaviour. The escape latency relies on a sup- Behavioural procedures plementary sensory information, the contact with the grid floor that per se activates specific retrieval pathways, but Spontaneous alternation performances. Each mouse, includes an interfering information: the absence of electric naive to the apparatus, was placed at the end of one arm in shock in this compartment during the retention session. a Y-maze (three arms, 50 cm long, 60 separate) and allowed This parameter is more reliably measured in active avoidance to move freely through the maze during a single 8-min ses- paradigms and may, in our case, have led to less sensitive sion. The series of arm entries, including possible returns differences among groups. The compounds were administered into the same arm, was recorded visually. An entry was 30 min before the training session or 10 min before dizocil- defined as the penetration of the hind paws of the animal at pine/scopolamine, given 20 min before the training session. least 2 cm into the arm. An alternation was defined as entries No injection was performed before the retention session.

Downloaded from jop.sagepub.com by guest on September 15, 2010 4 Journal of Psychopharmacology 0(0)

Step-down passive avoidance test. The apparatus con- non-parametric ANOVA (H values), followed by the sisted of a transparent acrylic cage (30 cm 30 cm 40 cm Dunn’s multiple comparisons test. Percentages of animals- high) with a grid floor, inserted in a soundproof outer box to-criterion in the step-down passive avoidance procedure (35 cm 35 cm 90 cm high). A 15 W lamp lighted the cage were analysed using a Fisher’s exact test. For reading clarity, during the experimental period. A wooden platform ANOVA data are all reported in the figure legends. The level (4 cm 4cm 4 cm) was fixed at the centre of the grid of statistical significance was p < 0.05. floor. Intermittent electric shocks (1 Hz, 500 ms, 40 V DC) were delivered to the grid floor using an isolated pulse stim- ulator (Model 2100, AM Systems, Everett, WA, USA). The Results test consisted of two training sessions, at a 90-min time inter- Anti-amnesic effects of ANAVEX2-73 in val, and a retention session, carried out 24 h after the first pharmacological models of amnesia training. During training sessions, each mouse was placed on the platform. When it stepped down and placed its four ANAVEX2-73 was first tested against the learning impair- paws on the grid floor, shocks were delivered for 15 s. Step- ments induced in mice by the muscarinic receptor antagonist down latency and the numbers of vocalizations and flinching scopolamine. Mice were administered with ANAVEX2-73, in reactions were measured. Shock sensitivity was evaluated by the 0.1–3 mg/kg i.p. range, 10 min before receiving scopol- adding these two numbers. None of the treatments used in the amine, 0.5 mg/kg s.c. Animals were then tested for spontane- present study significantly affected the shock sensitivity. ous alternation performances or trained in the step-through Animals that stepped down before 3 s has elapsed or that type passive avoidance procedure (Figure 2). Scopolamine did not step down within 30 s were discarded (less than 5% induced a highly significant decrease in alternation perfor- of the mice). Animals that did not step down within 60 s mance (Figure 2A) and an increase in the number of arm during the second session were considered as remembering entries (Figure 2B). The pre-administration of ANAVEX2- the task and were taken off, without receiving further electric 73 led to a dose-dependent attenuation of the scopolamine- shocks. The retention test was performed in a similar manner induced alternation deficit, significant at 1 and 3 mg/kg as training, except that the shocks were not applied to the grid (Figure 2A). The drug failed to affect the locomotor increase floor. Each mouse was again placed on the platform, and the (Figure 2B). In the step-through passive avoidance test, pre- latency was recorded, with an upper cut-off time of 300 s. Two training administration of scopolamine also provoked highly parametric measures of retention were analysed: the latency significant learning impairments, measured through a and the number of animals reaching the avoidance criterion, decrease in latency (Figure ) and increase in escape latency defined as correct if the latency measured during the retention (Figure 2D). The pre-treatment with ANAVEX2-73 attenu- session was greater than threefold the latency showed by ated the impairments of both step-through latency, dose- the animal during the second training and at least greater dependently and highly significantly at doses higher than than 60 s. 0.3 mg/kg (Figure 2C), and escape latency, highly significantly at 0.3 and 3 mg/kg (Figure 2D). Lipid peroxidation measures ANAVEX2-73 was then tested against the learning impairments induced in mice by the non-competitive Mice were sacrificed by decapitation and were rapidly NMDA receptor antagonist dizocilpine. Mice were adminis- removed, the hippocampus dissected out, weighed and kept in tered with ANAVEX2-73, in the 0.1–3 mg/kg i.p. range, liquid nitrogen until assayed. After thawing, the hippocampus 10 min before receiving dizocilpine, 0.15 mg/kg i.p. was homogenized in cold methanol (1/10 w/v), centrifuged at Dizocilpine induced a highly significant decrease in alterna- 1000 g for 5 min and the supernatant collected. Homogenate tion performance (Figure 3A) and an augmentation of explo- was added to a solution containing FeSO4 1 mM, H2SO4 ration, shown by a highly significant increase in the total 0.25 M, xylenol orange 1 mM, and incubated for 30 min at number of arm entries during the session (Figure 3B). Pre- room temperature. Absorbance was measured at 580 nm administration of ANAVEX2-73 led to a bell-shaped attenu- (A5801), and 10 mL of cumene hydroperoxide (CHP) 1 mM ation of the dizocilpine-induced alternation deficit, significant was added to the sample and incubated for 30 min at room at the 0.3 and 1 mg/kg doses (Figure 3A). The drug failed to temperature, to determine the maximal oxidation level. significantly attenuate the locomotor increase, but a tendency Absorbance was measured at 580 nm (A5802). The level of was noted (Figure 3B). In the long-term memory test, pre- lipid peroxidation was determined as CHP equivalents training administration of dizocilpine also provoked highly according to: CHP eq. ¼ A5801/A5802 [CHP (nmol)] dilu- significant learning deficits, measured through a decrease in tion, and expressed as CHP eq. per wet tissue weight. step-through latency (Figure 3C) and increase in escape latency (Figure 3D). Pre-treatment with ANAVEX2-73 led Statistical analyses to a bell-shaped, but significant, attenuation of the decrease in step-through latency, at 0.3 and 1 mg/kg (Figure 3C). Alternation percentages and biochemical data were expressed Escape latency results appeared less clear, but a significant as mean S.E.M. They were analysed using one-way reduction of the dizocilpine-induced increase was measured ANOVA (F values), followed by the Dunnett’s post-hoc mul- at 0.1 mg/kg (Figure 3D). tiple comparison test. Passive avoidance latencies were The compound shows a moderate affinity for the s1 expressed as median and interquartile range, since an upper receptor, but with a marked selectivity towards s2 sites. cut-off time was set. They were analysed a Kruskal–Wallis We therefore investigated the involvement of the s1 receptor

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 5

(A) 75 (B) 60

70 50

65 40 60 30 55 20 50 Number of arm entries 10 Alternation performance (%) 45

40 0 8 8 8 13 13 14

(D) 180 (C) 300

150 240 120 180 90 120 60 Escape latency (s) Step-through latency (s) 60 30

0 0 998131314 V 0.5 0.5 0.5 0.5 0.5 Scopolamine V 0.5 0.5 0.5 0.5 0.5 VV 0.1 0.3 1 3 ANAVEX 2-73 V V 0.1 0.3 1 3 Drugs (mg/kg ip)

Figure 2. Anti-amnesic effect of ANAVEX2-73 on the learning impairments in scopolamine-treated mice: (A) alternation performances and (B) total number of arm entries in the Y-maze test; (C) step-through latency and (D) escape latency in the passive avoidance test. ANAVEX2-73 (0.1–3.0 mg/kg i.p.) or the saline vehicle solution (V) was administered 30 min before the test, or 10 min before scopolamine (0.5 mg/kg s.c.), which was administered

20 min before the Y-maze session or passive avoidance training. The number of mice per group is indicated below the columns in (B, D). F(5,63) ¼ 6.17, p ¼ 0.0001 in (A); F(5,63) ¼ 2.56, p < 0.05 in (B); H ¼ 33.61, p < 0.0001 in (C); H ¼ 22.95, p < 0.001 in (D). * p < 0.05, ** p < 0.01 vs. the V-treated group;# p < 0.05, ## p < 0.01 vs. the scopolamine-treated group; Dunnett’s test in (A, B); Dunn’s test in (C, D).

in the anti-amnesic effect of ANAVEX2-73 against dizocil- decrease in step-through latency, but significantly blocked pine-induced learning impairments. The s1 receptor antago- the beneficial effects of ANAVEX2-73 at 0.3 or 1 mg/kg nist BD1047 was administered, at 1 mg/kg i.p., (Figure 4C). Results were less clear in terms of escape simultaneously with ANAVEX2-73. As shown in Figure latency (Figure 4D). 4A, BD1047 failed to affect the dizocilpine-induced alterna- tion deficit, but significantly blocked the anti-amnesic effect of ANAVEX2-73, at both the 0.3 and 1 mg/kg i.p. doses. In parallel, BD1047 tended to block the mild attenuation Anti-amnesic effects of ANAVEX2-73 in a induced by ANAVEX2-73 on dizocilpine-induced increase non-transgenic model of amyloid toxicity in number of arm entries (Figure 4B), particularly when co-administered with ANAVEX2-73 (0.3 mg/kg). This is an Muscarinic and/or s1 receptor ligands have been shown to interesting observation, unusual with selective s1 receptor be promising compounds in AD. We therefore examined ligands and therefore putatively involving its cholinergic whether ANAVEX2-73 could reverse the learning impair- activity. In the passive avoidance procedure, the BD1047 ments induced in mice, one week after i.c.v. injection of co-treatment also failed to affect the dizocilpine-induced Ab25-35 peptide. Mice were first administered i.c.v. with the

Downloaded from jop.sagepub.com by guest on September 15, 2010 6 Journal of Psychopharmacology 0(0)

(A) 75 (B) 50

70 40 65

60 30

55 20 50 10 Number of arm entries (%) Alternation (%) performance 45

40 0 14 15 20 14 8 15

(D) 150 (C) 300

120 240

90 180

120 60 Escape latency (s) Step-through latency (s) 60 30

0 0 11 12 12 18 18 18 V 0.15 0.15 0.15 0.15 0.15 Dizocilpine V 0.15 0.15 0.15 0.15 0.15 V V 0.1 0.3 1 3 ANAVEX 2-73 V V 0.1 0.3 1 3 Drugs (mg/kg ip)

Figure 3. Anti-amnesic effect of ANAVEX2-73 on the learning impairments in dizocilpine-treated mice: (A) alternation performances and (B) total number of arm entries in the Y-maze test; (C) step-through latency and (D) escape latency in the passive avoidance test. ANAVEX2-73 (0.1–3.0 mg/kg i.p.) or the saline vehicle solution (V) was administered 30 min before the test, or 10 min before dizocilpine (0.15 mg/kg i.p.), which was administered 20 min before the Y-maze session or passive avoidance training. The number of mice per group is indicated below the columns in (B, D). F(5,93) ¼ 5.83, p ¼ 0.0001 in (A); F(5,93) ¼ 2.59, p < 0.05 in (B); H ¼ 31.48, p < 0.0001 in (C); H ¼ 27.25, p < 0.0001 in (D). * p < 0.05, ** p < 0.01 vs. the V-treated group; # p < 0.05 vs. the dizocilpine-treated group; Dunnett’s test in (A, B); Dunn’s test in (C, D).

scrambled Ab (Sc.Ab, 9 nmol/mouse). On day 7 after injec- treatment led to a bell-shaped attenuation of Ab25-35-induced tion, mice were tested for spontaneous alternation (Figure alternation deficit, significant at 0.3 mg/kg (Figure 5C). The 5A). On day 8, they were trained for step-through passive drug failed to affect the number of arm entries (Figure 5D). avoidance with retention session being performed on day 9 ANAVEX2-73 also significantly attenuated the Ab25-35- (Figure 5B). ANAVEX2-73 was administered 30 min before induced decrease of step-through latency, at 0.3 and 1 mg/ the 8-min session in the Y-maze or the passive avoidance kg (Figure 5E), and Ab25-35-induced increase of escape training session. Neither the central injection of Sc.Ab pep- latency, at 0.1–1 mg/kg (Figure 5F). The dose response tide, nor the subsequent treatment with ANAVEX2-73 tended to be bell shaped, with an attenuation of the effect affected alternation performance in the Y-maze test (Figure at the highest dose tested, 3 mg/kg. 5A) or step-through latency (Figure 5B) in the passive avoid- ance test. When mice were centrally injected with Ab25-35 pep- tide (9 nmol), they showed a highly significant alteration of Neuroprotective effect of ANAVEX2-73 in the A25-35 alternation (Figure 5C), with no change in locomotion non-transgenic model of amyloid toxicity (Figure 5D). They also exhibited a significant decrease of step-through latency (Figure 5E) and a highly significant The neuroprotective activity of ANAVEX2-73 was then increase of escape latency (Figure 5F). The ANAVEX2-73 investigated in the Ab25-35 model of amyloid toxicity by

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 7

(A) 75 (B) 50 70

65 40

60 30

55 20 50 10 Alternation (%) performance 45 of arm number entries Total

40 0 14 12 10 10 10 12 10

(D) 150 (C) 300

240 120

180 90

120 60 Escape latency (s) 60 30 Step-through latency (s)

0 0 8 8 8 10 10 10 10 V 0.150.15 0.15 0.15 0.15 0.15 Dizocilpine V 0.150.15 0.15 0.15 0.15 0.15 VVV 0.3 0.3 1 1 ANAVEX2-73 V VV0.3 0.3 11 VV1 V1 V1 BD1047 V VVV1 1 1 Drugs (mg/kg ip)

Figure 4. Blockade by the s1 receptor antagonist BD1047 of the anti-amnesic effect of ANAVEX2-73: (A) alternation performances and (B) total number of arm entries in the Y-maze test; (C) step-through latency and (D) escape latency in the passive avoidance test. BD1047 (1.0 mg/kg i.p.), ANAVEX2-73 (0.3, 1.0 mg/kg i.p.) or the saline vehicle solution (V) was administered 10 min before dizocilpine (0.15 mg/kg i.p.), which was administered 20 min before the Y-maze session or first passive avoidance training. The number of mice per group is indicated below the columns in

(B, D). F(6,77) ¼ 5.79, p < 0.0001 in (A); F(6,77) ¼ 4.90, p < 0.001 in (B); H ¼ 30.56, p < 0.0001 in (C); H ¼ 22.95, p < 0.001 in (D). ** p < 0.01 vs. the V-treated group; # p < 0.05 vs. the dizocilpine-treated group; o p < 0.05, oo p < 0.01 vs. the (ANAVEX2-73 þ dizocilpine)-treated group; Dunnett’s test in (A, B); Dunn’s test in (C, D). administration of the drug simultaneously with the peptide, (Figure 6C). In terms of animals-to-criterion, a significant i.e. 7 days before the behavioural tests or biochemical mea- increase in the percentage was measured at doses of 100 mg/ sures, as previously used for other drugs (Meunier et al., 2006; kg and higher. It must be noted that dose-response curves Villard et al., 2009). Animals were, from here, tested for pas- did not appear bell shaped, as usually observed for such sive avoidance responses in the step-down procedure in order responses. to directly compare with the responses observed for the The effect of ANAVEX2-73 on Ab25-35-induced toxicity related compound ANAVEX1-41 (Villard et al., 2009). As was characterized at the biochemical level. The Ab25-35- shown in Figure 6, pre-treatment with ANAVEX2-73 led to induced oxidative stress was evaluated in the hippocampus a dose-dependent prevention of the appearance of Ab25-35- formation by measuring the level of lipid peroxidation. induced alternation deficits, with significant effects observed As shown in Figure 7, Ab25-35 provoked a þ 100% increase for the doses of 10 mg/kg and higher (Figure 6A). A moderate in lipid peroxidation levels that was dose-dependently but significant effect on locomotion was noted for the reversed by ANAVEX2-73, with significant differences doses between 10 and 100 mg/kg (Figure 6B). In the passive observed for the 0.3 and 1 mg/kg doses. These doses appeared avoidance test, the Ab25-35-induced learning deficits were thus to be effective neuroprotective doses at the morpholog- significantly attenuated at doses of 30 mg/kg and higher ical, biochemical and behavioural levels of analysis.

Downloaded from jop.sagepub.com by guest on September 15, 2010 8 Journal of Psychopharmacology 0(0)

Drug Drug

ScAb 7 days Behav test Ab 7 days Behav test (A) 80 (C) 75 (D) 50

75 70 40 70 65 30 65 60 20 60 55 10

55 Number of arm entries Alternation (%) performance 50 Alternation (%) performance 50 0 16 9 9 10 10 9 10 16 25 10 10 17 16 17 14

(B) 300 (E) 300 (F) 60 50 240 240 40 180 180 30 120 120 20 Escape latency (s) 60 60 10 Step-through latency (s) Step-through latency (s) 0 0 0 17 10 10 101010 10 17 18 12 12 12 12 12 15 Sc.Ab Sc.Ab Ab25-35 Sc.Ab Ab25-35 V 0.010.03 0.1 0.3 1 3 VV 0.1 0.03 0.1 0.03 1 3 VV0.01 0.03 0.10.3 1 3 ANAVEX2-73 (mg/kg ip) ANAVEX2-73 (mg/kg ip) ANAVEX2-73 (mg/kg ip)

Figure 5. Anti-amnesic effect of ANAVEX2-73 against the learning impairments in Ab25-35 peptide-treated mice: (A, C) alternation performances and (D) total number of arm entries in the Y-maze test; (B, E) step-through latency and (F) escape latency in the passive avoidance test.

Mice were administered with scrambled Ab25-35 or Ab25-35 peptide (9 nmol i.c.v.), 7 days before the tests. ANAVEX2-73 (0.01–3.0 mg/kg i.p.) or the saline vehicle solution (V) was administered 30 min before the Y-maze session or passive avoidance training. Panels (A) and (B) show the lack of effect of ANAVEX2-73 in Sc.Ab-treated mice. The number of mice per group is indicated below the columns in (A, B, D, F). F(6,72) ¼ 2.08, p > 0.05 in (A); H ¼ 5.05, p > 0.05 in (B); F(7,116) ¼ 3.19, p < 0.01 in (C); F(7,116) ¼ 1.26, p > 0.05 in (D); H ¼ 24.76, p < 0.001 in (E); H ¼ 16.38, p < 0.05 in (F). * p < 0.05, ** p < 0.01 vs. the Sc.Ab-treated group; # p < 0.05 vs. the Ab25-35-treated group; Dunnett’s test in (A, C, D); Dunn’s test in (B, E, F).

Moreover, co-administration of BD1047 completely pre- responses (Figure 8A), but attenuated the effects on the passive vented the protective effect of ANAVEX2-73 (1 mg/kg; avoidance responses, significantly at the lower dose and non- Figure 7) on lipid peroxidation, indicating the s1 receptor significantly at the higher dose of ANAVEX2-73 (Figure 8B, involvement in this effect of the drug. 8C). Scopolamine, injected at 0.5 mg/kg significantly blocked ANAVEX2-73 acts both as a potent muscarinic compound, the protective effect of 1 mg/kg ANAVEX2-73 on alternation with Ki values lower than 500 nM for all M1–M4 subtypes and behaviour (Figure 8D) and on passive avoidance response as a moderate s1 receptor agonist, but selective towards s2 (Figure 8E, 8F). A non-significant difference was noted subtypes. In order to confirm the mixed involvement of mus- between the (ANAVEX2-73 þ Ab25-35)-treated and (scopol- carinic and s1 targets in the pharmacological mechanisms amine þ ANAVEX2-73 þ Ab25-35)-treated groups in terms of underlying its neuroprotective action, we co-administered step-down latency (Figure 8E), but significance was reached in ANAVEX2-73 with either BD1047 or scopolamine, simulta- terms of animal-to-criterion (Figure 8F). These experiments neously with the Ab25-35 peptide and examined the extent of confirmed a mixed involvement of muscarinic and s1 systems learning deficits after 7 days in the Y-maze and step-down in the ANAVEX2-73 effect, suggesting the involvement of passive avoidance procedures (Figure 8). BD1047, injected at muscarinic receptors in both short- and long-term , 1 mg/kg, failed to affect the protective effect of ANAVEX2-73, while the s1 activation seem selectively involved in long-term injected at 0.3 or 1 mg/kg, on spontaneous alternation memory processes.

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 9

Drug Aβ 7 days Behav test (A) (B) 75 50

## ## 70 40 # # # ∗ 65 30

60 ∗∗ ∗∗ 20

55 Number of arm entries 10 Alternation (%) performance

50 0 21 19 10 9 17 11 10 10 10

(C) (D) 300 100

240 # ## 80 ∗ ∗ 180 ∗∗ # # ## 60 # # ∗ ∗ ∗∗ 120 40 ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ Step-down latency (s) Step-down 60 Animals-to-criterion (%) ∗∗ ∗∗ 20 ∗∗

0 0 14 13 10 10 12 11 12 15 12 ScAβ Aβ25-35 ScAβ Aβ25-35 V V 1 3 10 301003001000 V V 1 3 10 30 1003001000 ANAVEX 2-73 (μg/kg i.p.) ANAVEX 2-73 (μg/kg i.p.)

Figure 6. Neuroprotective effect of ANAVEX2-73 against the learning impairments in Ab25-35 peptide-treated mice: (A) alternation performances and (B) total number of arm entries in the Y-maze test; (C) step-down latency and (D) percentage of animals-to-criterion in the passive avoidance test. Mice were administered with ANAVEX2-73 (0.001–1.0 mg/kg i.p.) or the saline vehicle solution (V) immediately before the i.c.v. injection of scrambled Ab25-

35 or Ab25-35 peptide (9 nmol i.c.v.), 7 days before the tests. The number of mice per group is indicated below the columns in (B, D). F(8,116) ¼ 5.61, p < 0.0001 in (A); F(8,116) ¼ 2.66, p < 0.05 in (B); H ¼ 26.82, p < 0.001 in (C). * p < 0.05, ** p < 0.01 vs. the Sc.Ab-treated group; # p < 0.05,

## p < 0.01 vs. the Ab25-35-treated group; Dunnett’s test in (A, B); Dunn’s test in (C); Fisher’s exact test in (D).

(Figure 9B) and animals-to-criterion (Figure 9C). A dose- Activity of the demethyl metabolite of ANAVEX2-73 dependent prevention of Ab25-35-induced oxidative stress in the hippocampus was also measured, with a significant effect Preliminary studies revealed that aminotetrahy- at the highest dose tested (Figure 10A). ANAVEX2-140, on drofuran compounds are solely transformed to a major the contrary, failed to attenuate the spontaneous alternation metabolite by demethylation of the tertiary amine group. deficits (Figure 9D), passive avoidance deficits (Figure 9E, ANAVEX2-73 will therefore lead to ANAVEX19-144, and 9F) or increase in lipid peroxidation levels (Figure 10B) ANAVEX1-41 to ANAVEX2-140 (Figure 1). The metabo- induced by Ab25-35, in the dose range tested. These experi- lites were tested in the 30–1000 mg/kg i.p. dose range for ments suggest that the main metabolite of ANAVEX2-73 is their putative neuroprotective activities, after simultaneous also an active neuroprotective compound, in contrast to the administration with Ab25-35 peptide. ANAVEX19-144 pre- ANAVEX1-41 metabolite. vented the Ab25-35-induced alternation deficits at the highest One of the main consequences of this observation is doses tested, with a significant effect at 0.3 mg/kg (Figure 9A). that the duration of action of the compound could be The metabolite prevented significantly the passive avoidance extended. To test this hypothesis, both ANAVEX2-73 and deficits at the highest dose tested, both in terms of latency ANAVEX1-41 were administered 6 h or 12 h before the

Downloaded from jop.sagepub.com by guest on September 15, 2010 10 Journal of Psychopharmacology 0(0)

et al., 2007; Vamvakides, 2002b), but with a lower affinity for the s1 protein and a better selectivity towards s2 sites. 250 ∗∗ It could also be compared with the pharmacological profile of ∗∗ ∗∗ donepezil, which is not only a cholinomimetic through its 200 ∗∗ οο inhibition of acetylcholinesterase but also a s1 protein agonist ∗∗ (Kato et al., 1999; Maurice et al., 2006). Both ANAVEX1-41 150 and donepezil have been reported to act in vivo through both targets, in particular as concerns their neuroprotective actions ## against amyloid toxicity (Ishikawa et al., 2009; Meunier et al., 100 ## 2006; Villard et al., 2009). ANAVEX2-73 was therefore expected to present anti-amnesic and neuroprotective activi- 50 ties. In the first part of this study, we indeed report that Lipid peroxidation (% of control) Lipid peroxidation ANAVEX2-73 attenuated the learning deficits induced by scopolamine or dizocilpine (see Table 1). The compound 0 9 10 5 6 9 10 10 4 was effective in the spontaneous alternation and passive Aβ25-35 avoidance responses, two procedures assessing different memory processes, namely spatial working memory for the ANAVEX2-73 VV0.01 0.03 0.1 0.3 11 Y-maze test and long-term non-spatial memory for the pas- BD1047 1 sive avoidance test. The compound was active at doses Drugs (mg/kg ip) around 0.3 and 1 mg/kg. It appeared more effective in revers- ing scopolamine-induced long-term memory than short-term memory deficits, but equally effective in reversing all deficits Figure 7. Neuroprotective effect of ANAVEX2-73 against Ab -induced 25-35 in the dizocilpine model. These observations are in agreement toxicity in the hippocampus of mice: lipid peroxidation, 7 days after Ab 25- with a pharmacological effect involving both the muscarinic injection. Mice were administered i.p. with saline vehicle solution (V) 35 and s components. Muscarinic agonists, and particularly or ANAVEX2-73 (300 mg/kg) immediately before being administered i.c.v. 1 at the M1 receptor, are known to reverse the scopolamine- with Ab peptide (9 nmol). Lipid peroxidation levels were measured on 25-35 induced amnesia by a direct competitive interaction (Malviya hippocampal extracts using the modified FOX reaction. The number of et al., 2008; Nakahara et al., 1989, 1990). On the other hand, animals per group is indicated below the columns. ** p < 0.01 vs. the we observed that a pre-treatment with BD1047, a selective s1 (Sc.AbþV)-treated group; ## p < 0.01 vs. the (Ab25-35þV)-treated group; oo receptor antagonist, blocked the anti-amnesic effects of p < 0.01 vs. (Ab þANAVEX2-73, 1 mg/kg)-treated group; Dunnett’s 25-35 ANAVEX2-73 in the dizocilpine model, at its two main test. active doses and on both short-term and long-term memory procedures. This effect is usually considered as the most i.c.v. injection of Ab25-35 peptide. The most active doses for adapted to determine the in vivo agonist/antagonist activity each compound were selected, i.e. 0.3 and 1 mg/kg for at the s1 site. Moreover, this observation is coherent with the ANAVEX2-73 and 0.1 and 0.3 mg/kg for ANAVEX1-41. more limited effect of ANAVEX2-73 on dizocilpine- versus After 1 week, animals were tested for the behavioural scopolamine-induced deficits. Indeed, the behavioural efficacy responses. As shown in Figure 11, ANAVEX2-73 signifi- of s1 ligands has been shown to be highly dependent upon cantly prevented the spontaneous alternation deficits at steroidal tonus and particularly (Bergeron et al., both doses when it was injected 6 h before the peptide or 1999; Maurice et al., 1999; Urani et al., 2001). The high acute at the lowest dose when injected 12 h before the peptide stress-reinforcing passive avoidance training provokes a sig- (Figure 11A). A similar profile was observed on passive nificant increase in the brain level of the steroid, acting as an avoidance responses (Figure 11B, 11C). ANAVEX1-41 efficient endogenous s1 antagonist. Stress conditions are tended to attenuate the spontaneous alternation deficits at therefore known to limit the efficacy of the s1 agonist the dose of 0.1 mg/kg when injected 6 h before Ab25-35, and (Urani et al., 2001). at 0.3 mg/kg when injected 12 h before the peptide, but none The differential efficacy of ANAVEX2-73 in reversing of the effect reached significance versus the Ab25-35-treated scopolamine-induced deficits in the Y-maze and passive group (Figure 11D). Similarly, the responses in passive avoid- avoidance was surprising, considering (i) the particular ance failed to show a significant effect of the ANAVEX1-41 involvement of muscarinic receptors in short-term memory treatment (Figure 11E, 11F). These experiments confirmed processes (Ohno et al., 1994), and (ii) the previous observa- that ANAVEX2-73 exhibits a longer duration of action tion that ANAVEX1-41 was still effective in attenuating the than its related compound ANAVEX1-41, an effect likely alternation deficits observed in s1 antisense-treated mice due to the efficacy of its main metabolite, ANAVEX19-144. administered with scopolamine (Espallergues et al., 2007). ANAVEX2-73 was, on the contrary, expected to be equally, Discussion if not more, effective in reversing alternation deficits and pas- sive avoidance deficits in the scopolamine model. This obser- ANAVEX2-73 is a new aminotetrahydrofuran derivative; vation deserves additional experiments, particularly in s1 which acts both as a muscarinic receptor ligand and a s1 knockout mice. protein agonist. It therefore shares the pharmacological pro- ANAVEX2-73 was also effective in reversing the learning file of its related compound ANAVEX1-41 (Espallergues impairments observed 1 week after the i.c.v. injection of

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 11

(A) (B) (C) 75 100 # 300 # 90 70 # 240 80 ∗ ## ## ## 70 65 ## ∗ ∗∗ ∗∗ ∗∗ 180 ∗ 60 ∗∗ ο # 60 ∗∗ 50 ∗∗ ο 120 40 55 ∗∗ 30 Animals-to-criterion (%)

Step-down latency (s) Step-down ∗∗ ∗∗ 20 50 60 ∗∗ ∗∗ Alternation (%) performance 10 45 0 0 13 13 10 13 10 12 11 13 13 10 13 10 12 11 ScAβ Aβ25-35 ScAβ Aβ25-35 ScAβ Aβ25-35 VVV0.3 0.3 1 1 ANAVEX2-73 VVV0.3 0.3 1 1 ANAVEX2-73 VVV0.3 0.3 1 1 VV1VV1 1 BD1047 VV1VV1 1 BD1047 VV1VV1 1

(D) (E) (F) 75 100 300 # # 90 70 # 240 80 ## ## ## ∗∗ ## 70 ∗ 65 ο ∗∗ ## ∗∗ ∗ ∗∗ 60 ∗∗ ∗∗ 180 ∗ 60 50 ∗∗ 120 40 ο ∗∗ 55 30 Animals-to-criterion (%)

Step-down latency (s) Step-down ∗∗ 20 ∗∗ 50 60 Alternation (%) performance 10 45 0 0 13 13 16 13 11 12 9 13 13 16 13 11 12 9 ScAβ Aβ25-35 ScAβ Aβ25-35 ScAβ Aβ25-35 VVV0.3 0.3 1 1 ANAVEX2-73 VVV0.3 0.3 1 1 ANAVEX2-73 VVV0.3 0.3 11 V V 0.5 VV0.5 0.5 Scopolamine V V 0.5 VV0.5 0.5 Scopolamine V V 0.5VV 0.5 0.5 Drugs (mg/kg ip) Drugs (mg/kg ip) Drugs (mg/kg ip)

Figure 8. Effects of the co-administration of (A–C) the s1 receptor antagonist BD1047 or (D-F) the muscarinic receptor antagonist scopolamine on the neuroprotective effect of ANAVEX2-73. (A, D) alternation performances in the Y-maze test, and (B, E) step-down latency and (C, F) percentage of animals to criterion in the passive avoidance test. BD1047 (1 mg/kg i.p.), scopolamine (0.5 mg/kg i.p.), ANAVEX2-73 (0.3, 1.0 mg/kg i.p.) or the saline vehicle solution (V) was administered immediately before Sc.Ab or Ab25-35 peptide (9 nmol). Animals were examined for alternation 7 days after injections and trained in the passive avoidance test 8 days after injections. The number of mice per group is indicated below the columns in (A, B, D, E).

F(6,81) ¼ 3.13, p < 0.01 in (A); H ¼ 39.7, p < 0.0001 in (B); F(6,86) ¼ 3.09, p < 0.01 in (D); H ¼ 28.1, p < 0.0001 in (E). ** p < 0.01 vs. the (Sc.AbþV)- o treated group; # p < 0.05, ## p < 0.01 vs. the (Ab25-35þV)-treated group; p < 0.05 vs. the (ANAVEX2-73þdizocilpine)-treated group; Dunnett’s test in (A, D); Dunn’s test in (B, E); Fisher’s exact test in (C, F).

Ab25-35 peptide. In the rodent brain, Ab25-35 peptide induces However, and as observed for ANAVEX1-41 (see Villard biochemical changes, morphological alterations and beha- et al., 2009), the efficacy of ANAVEX2-73 in vivo is clearly vioural impairments reminiscent of AD physiopathology unrelated to its in vitro s1 affinity in the low micromolar (Maurice et al., 1996; Villard et al., 2009). In particular, range, and this synergy could be explained either by a good Ab25-35-treated mice showed spontaneous alternation and of the compound in terms of protein binding passive avoidance deficits clearly related to alterations in cho- and brain/plasma ratio or, pharmacologically, by considering linergic and corticolimbic systems (Maurice the acetylcholine/glutamate/s1 protein interactions. Such a et al., 1996; Olariu et al., 2001). ANAVEX2-73, administered mixed muscarinic/s1 ligand is indeed expected to notably before the behavioural procedures, reversed the Ab25-35- (i) increase, in the cortex and hippocampus, the glutamatergic induced deficits at the doses of 0.3 and 1 mg/kg. These obser- transmission and long-term potentiation, partly by activa- vations confirmed that ANAVEX2-73 is an effective anti- tion of the s1 protein that facilitates glutamate release and amnesic drug in pharmacological models of amnesia but activates glutamate receptors (Dong et al., 2005; Monnet appearing, however, slightly less potent than its related com- et al., 1992); (ii) increase acetylcholine release, via pre-synap- pound ANAVEX1-41 (Espallergues et al., 2007; Villard et al., tic s1 protein-mediated and M2 autoreceptor-mediated effects 2009). This order of is, remarkably, in accordance (Matsuno et al., 1992, 1993; Quirion et al., 1994; Vamvakides, with the drug affinities towards s1 and muscarinic sites. 2002b); (iii) activate phospholipase C by muscarinic receptor

Downloaded from jop.sagepub.com by guest on September 15, 2010 12 Journal of Psychopharmacology 0(0)

(A) (D) 75 75

70 # 70

65 ∗ 65 ∗ ∗∗ 60 ∗∗ ∗ 60 ∗∗ ∗ ∗∗ 55 55

50 50 Spontaneous alternation (%) 45 45 81678810 16 17 14 14 15 15 (B) (E) 300 300

240 240

180 180

## 120 120 ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗

Step-down latency (s) Step-down 60 60 ∗∗ ∗∗ ∗∗

0 0 7 11 8 8 8 10 16 16 15 15 14 15

(C) (F) 100 100

80 80

60 # 60

40 40 ∗∗ ∗∗ ∗∗ ∗∗ 20 ∗∗

Animals-to-criterion (%) 20 ∗∗ ∗∗ ∗∗ ∗∗ 0 0 ScAβ Aβ25-35 ScAβ Aβ25-35 V V 0.03 0.1 0.3 1 V V 0.03 0.1 0.3 1

ANAVEX19-144 (mg/kg i.p.) ANAVEX2-140 (mg/kg i.p.)

Figure 9. Dose-response effects of ANAVEX19-144 (A–C) and ANAVEX2-140 (D–F) on Ab25-35-induced learning deficits and oxidative stress in mice: alternation performances (A, D) in the Y-maze test; step-down latency (B, E) and animals-to-criterion (C, F) in the passive avoidance test. Mice were administered i.p. with vehicle solution (saline, V) or ANAVEX compounds (0.03–1.0 mg/kg) immediately before ScAb or Ab25-35 peptide (9 nmol). Spontaneous alternation was measured on day 7, passive avoidance response on days 8–9. The number of animals per group is indicated below the columns. ANOVA: F(5,56) ¼ 2.70, p < 0.01 in (A); H ¼ 32.7, p < 0.0001 in (B); F(5,90) ¼ 5.41, p < 0.01 in (D); H ¼ 21.7, p < 0.001 in (E). * p < 0.05, ** p < 0.01 vs. (Sc.AbþV)-treated group; # p < 0.05, ## p < 0.01 vs. the (Ab25-35þV)-treated group; Dunnett’s test in (A, D), Dunn’s test in (B, E), Fisher’s exact test in (C, F).

activation but amplified by s1 protein-mediated activity learning and memory processes and are markedly affected (Felder, 1995; Morin-Surun et al., 1999); and (iv) induce by amyloid peptides. Indeed, Ab25-35 and Ab1-40 peptides InsP3 formation and activation of ER InsP3 receptors, again have been shown to disrupt muscarinic receptor coupling amplified by s1 protein activation (Ferna´ndez de Sevilla, 2008; to G proteins that mediate induction of phosphoinositide Hayashi et al., 2000). These transduction mechanisms sustain accumulation and calcium release (Kelly et al., 1996).

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 13

(A) (B) ∗∗ 180 ∗∗ 180 ∗ ∗ ∗ ∗ 160 160 ∗ 140 140 120 ## 120 100 100 80 80 60 60 40 40 Lipid peroxydation (% of Ctl) Lipid peroxydation 20 20 0 0 10 10 3 10 10 10 8128 88 8 ScAβ Aβ25-35 ScAβ Aβ25-35 V V 0.03 0.1 0.3 1 V V 0.03 0.1 0.3 1

ANAVEX19-144 (mg/kg i.p.) ANAVEX19-144 (mg/kg i.p.)

Figure 10. Dose-response effects of ANAVEX19-144 (A) and ANAVEX2-140 (B) on Ab25-35-induced oxidative stress in mice: lipid peroxidation levels in hippocampus extracts. Mice were administered i.p. with vehicle solution (saline, V) or ANAVEX compounds (0.03–1.0 mg/kg) immediately before ScAb or Ab25-35 peptide (9 nmol). Animals were sacrificed on day 9 for hippocampus extracts preparation. The number of animals per group is indicated below the columns. ANOVA: F(6,68) ¼ 4.04, p < 0.01 in (A); F(5,51) ¼ 4.27, p < 0.01 in (B). * p < 0.05, ** p < 0.01 vs. (Sc.AbþV)-treated group; ## p < 0.01 vs. the (Ab25-35þV)-treated group; Dunnett’s test.

Moreover, Ab25-35 has also been reported to be directly dependent transcriptional activation of nuclear factor kB, involved in the impairment of phosphoinositide signalling associated with an intra-neuronal redox regulation by inhibiting the and GTP(g)S-stimulated PInsP2 (Meunier and Hayashi, 2010). It has been demonstrated phospholipase C activity (Strosznajder et al., 1999). that the above effects, e.g. reactive oxygen species generation The second part of this study examined the neuroprotec- and anti-apoptotic effects, were also obtained by stimulation tive activity of ANAVEX2-73 and included elements to of muscarinic receptors (Fisher et al., 2003; Mangelus et al., assess the pharmacological potential of the compound (see 2001). Activation of the s1 protein is therefore putatively Table 1). Ab toxicity has been shown to follow a very com- directly and potently involved in oxidative response. This plex pattern. Pathological generation of Ab peptides, or hypothesis must be analysed in neurons, particularly to be their application in vitro in cell cultures or in vivo into the reconciled within the context of mitochondrial dysfunction rodent brain, results in calcium deregulation, and ER and and deregulation of activity of involved in oxygen mitochondrial stress, leading to oxidative stress and apopto- radical generation and reduction, e.g. superoxide dismutase. sis. These direct effects are accompanied by a massive The compound therefore appeared less potent than its inflammatory response of the surrounding tissue and sec- related compound ANAVEX1-41 (Villard et al., 2009). ondary excitotoxic processes due to glutamate release in This lower efficacy was expected regarding the lower damaged areas (Selkoe, 1991; Villard et al., 2009; Yamada in vitro affinities measured for the compound on either mus- and Nabeshima, 2000). Toxicity therefore rapidly results in carinic or s1 binding assays. Moreover, as observed for dendritic shrinkage, synaptic loss and eventually cell death. ANAVEX1-41, the anti-amnesic and neuroprotective doses We therefore addressed reliable consequences of this toxic- are closely related. A pharmacological study, by pre-admin- ity, such as lipid peroxidation levels, as a consequence of the istration of either scopolamine or BD1047, showed that both cellular oxidative stress and the learning impairments as a drugs affected the beneficial effects induced by ANAVEX2- behavioural outcome of altered synaptic efficiency. We 73 in either the short-term or long-term memory responses. observed that when the compound was pre-administered These observations remain preliminary, and more precise before the peptide, it prevented the appearance of Ab25-35- studies could be envisaged with antisense or siRNA treat- induced learning deficits in the Y-maze and passive avoid- ments down-regulating durably the muscarinic or s1 targets, ance procedures at doses of 30 mg/kg and higher. The but they confirmed the mixed pharmacological profile of the appearance of oxidative stress was also prevented at the drug. This series of experiments, as a whole, indicated that dose of 0.3–1 mg/kg. The anti-apoptotic and anti-oxidant ANAVEX2-73, although presenting moderate in vitro affin- activity of ANAVEX2-73 may rely both on its muscarinic ities for muscarinic or s1 sites, shows a significant neuro- and s1 activities. Indeed, recent cellular biology studies protective activity against amyloid toxicity. With an active addressing the trophic and neuroprotective function of s1 dose of 0.3 mg/kg, the compound is equally effective as the protein clearly evidenced the stimulation of the anti-apopto- reference compounds donepezil or PRE-084, in this para- tic factor Bcl-2 by s1 agonists, via reactive oxygen species- digm (Meunier et al., 2006).

Downloaded from jop.sagepub.com by guest on September 15, 2010 14 Journal of Psychopharmacology 0(0)

(A) Aβ25-35 - 6 h Aβ25-35 - 12 h (D) Aβ25-35 - 6 h Aβ25-35 - 12 h 75 75 ## # 70 # 70 ∗ ∗ 65 65 ∗∗ ∗ ∗∗ ∗∗ 60 ∗∗ 60 ∗∗ 55 55

50 50

Spontaneous alternation (%) 45 Spontaneous alternation (%) 45 12 12 12 12 14 15 14 11 12 12 12 11 14 15 13 16

(B)## (E) 300 300

240 240

# 180 # 180

120 120 ∗ ∗ ∗∗ ∗ 60 60 ∗∗ Step-down latency (s) Step-down ∗∗ latency (s) Step-down ∗∗ 0 0 13 13 11 12 12 12 12 12 13 13 12 12 12 12 12 12

(C) (F) 100 100

80 80

60 # 60

40 40

∗ ∗ ∗ ∗∗∗ Animals-to-criterion (%) 20 Animals-to-criterion (%) 20

0 0 ScAβ Aβ25-35 ScAβ Aβ25-35 ScAβ Aβ25-35 ScAβ Aβ25-35 V V 0.3 1 V V 0.3 1 V V 0.3 1 V V 0.3 1

ANAVEX2-73 (mg/kg i.p.) ANAVEX2-73 (mg/kg i.p.)

Figure 11. Duration of the neuroprotective effects of ANAVEX2-73 or ANAVEX1-41 against the learning impairments in Ab25-35 peptide-treated mice: (A, D) alternation performances in the Y-maze test; (B, E) step-down latencies and (C, F) animals-to-criterion in the passive avoidance test. Mice were administered with ANAVEX2-73 (0.3, 1.0 mg/kg i.p.), ANAVEX1-41 (0.1, 0.3 mg/kg i.p.) or the saline vehicle solution (V) either 6 h or 12 h before the i.c.v. injection of Ab25-35 peptide (9 nmol i.c.v.), 7 days before the tests. The number of mice per group is indicated below the columns. F(3,47) ¼ 8.88, p ¼ 0.0001 for (Ab25-35 -6 h), F(3,53) ¼ 3.87, p < 0.05 for (Ab25-35 -12 h) in (A); H ¼ 8.17, p < 0.05 for Ab25-35 -6 h, H ¼ 22.5, p < 0.0001 for (Ab25-35 -12 h) in (B); F(3,46) ¼ 5.52, p < 0.01 for (Ab25-35 -6 h), F(3,57) ¼ 2.80, p < 0.05 for (Ab25-35 -12 h) in (D); H ¼ 8.22, p < 0.05 for (Ab25-35 -6 h), H ¼ 17.4, p < 0.001 for (Ab25-35 -12 h) in (E). * p < 0.05, ** p < 0.01 vs. the (Sc.AbþV)-treated group; # p < 0.05, ## p < 0.01 vs. the (Ab25-35þV)-treated group; Dunnett’s test in (A, D); Dunn’s test in (B, E); Fisher’s exact test in (C, F).

The last part of this study addressed the efficacy of by measuring the learning and memory deficits and lipid the drug metabolite. Both lead aminotetrahydrofuran com- peroxidation increase after Ab25-35 and the different drug pounds, ANAVEX1-41 as well as ANAVEX2-73, present a treatments, that only ANAVEX19-144, the ANAVEX2-73 single metabolite by demethylation of their tertiary metabolite, was active, but not ANAVEX2-140, the amine. The bioavailability and time-course of the parent com- ANAVEX1-41 metabolite. Therefore, ANAVEX2-73 may pounds and their metabolites is still under investigation, but act as both a drug and a pro-drug for ANAVEX19-144. we analysed their efficacy in the same paradigm. We observed, This mechanism is expected to considerably prolong the

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 15

Table 1. Summary of the pharmacological data: actives doses in beha- Funding vioural tests and in the lipid peroxidation level measure for ANAVEX2-73 or the metabolites, in the different anti-amnesic or neuroprotection This work was supported by collaboration contracts (#06122, procedures tested #07438) between Anavex Life Sciences (Pallini, Greece) and Inserm (Paris, France). Active dose (mg/kg i.p.)

Spont. Passive Lipid Conflict of interest Procedure alternation avoidance peroxidation AV is an employee of ANAVEX Life Sciences. TM is a Anti-amnesic effects: member of the scientific advisory board of ANAVEX Life ANAVEX2-73 Sciences. Scopolamine 1.0, 3.0 0.1–3.0 Dizocilpine 0.3, 1.0 0.3, 1.0 References Ab25-35 0.3 0.3, 1.0 Arias E, Ales E, Gabilan NH, Cano-Abad MF, Villarroya M, Garcia Neuroprotection: AG, et al. (2004) Galantamine prevents apoptosis induced by ANAVEX2-73 b-amyloid and thapsigargin: Involvement of nicotinic acetylcho- Ab25-35 (20 min) 0.01–1.0 0.03–1.0 0.3–1.0 line receptors. 46: 103–114. Ab25-35 (6 h) 0.3, 1.0 0.3, 1 Aydar E, Palmer CP, Klyachko VA and Jackson MB (2002) The Ab25-35 (12 h) 0.3 0.3 sigma receptor as a ligand-regulated auxiliary potassium channel ANAVEX19-144 subunit. 34: 399–410.

Ab25-35 (20 min) 0.3 1 1.0 Baratti CM, Opezzo JW and Kopf SR (1993) Facilitation of memory ANAVEX2-140 storage by the acetylcholine M2 muscarinic receptor antagonist a a a AF-DX 116. Behav Neural Biol 60: 69–74. Ab25-35 (20 min) — — — Bergeron R, de Montigny C and Debonnel G (1999) Pregnancy reduces aNot effective at doses up to 1 mg/kg. brain sigma receptor function. Br J Pharmacol 127: 1769–1776. Bymaster FP, Heath I, Hendrix JC and Shannon HE (1993) Comparative behavioral and neurochemical activities of choliner- duration of action of the drug and, in a final series of gic antagonists in rats. J Pharmacol Exp Ther 267: 16–24. experiments, we observed that ANAVEX2-73 is active when Dong Y, Fu YM, Sun JL, Zhu YH, Sun FY and Zheng P (2005) injected 6 h or 12 h before Ab25-35, in contrast to ANAVEX1- Neurosteroid enhances glutamate release in rat prelimbic cortex 41. This result indeed suggested a longer duration of action via activation of a1- and s1 receptors. Cell Mol Life Sci for ANAVEX2-73. These experiments deserve to be strength- 62: 1003–1014. ened by precise analyses after repeated or chronic administra- Erme M, Geula C, Ransil BJ and Mesulam MM (1992) The acute tion of the compounds. The pharmacological profile of and effects upon cholinergic axons or intracere- ANAVEX19-144 has also to be precisely established. It is brally injected b-amyloid in the rat brain. Neurobiol Aging 13: 553–559. particularly puzzling that ANAVEX2-73 appeared more Espallergues J, Lapalud P, Christopoulos A, Avlani VA, Sexton PM, active at the lowest dose when injected 12 h before Ab25-35 Vamvakides A, et al. (2007) Involvement of the sigma1 (s1) recep- than when injected 6 h before the peptide. We have no expla- tor in the anti-amnesic, but not antidepressant-like, effects of the nation for this discrepancy from the present study, and a aminotetrahydrofuran derivative ANAVEX1-41. Br J Pharmacol precise pharmacodynamic study must be conducted. The reg- 152: 267–279. ulation of muscarinic receptors and related transduction sys- Felder CC (1995) Muscarinic acetylcholine receptors: signal transduc- tems has to be analysed after chronic co-stimulation of s1 tion through multiple effectors. FASEB J 9: 619–625. receptors. For instance, the coupling of muscarinic receptors Ferna´ndez de Sevilla D, Nu´n˜ ez A, Borde M, Malinow R and Bun˜ oW to G-proteins and the muscarinic receptor-stimulated phos- (2008) Cholinergic-mediated IP3-receptor activation induces long- phoinositide hydrolysis has been shown to be affected during lasting synaptic enhancement in CA1 pyramidal neurons. J Neurosci 28: 1469–1478. ageing and in neurodegenerative diseases (Pedigo, 1994), and Fibiger HC (1991) Cholinergic mechanisms in learning, memory and these responses have been reported to be highly sensitive to dementia: a review of recent evidence. Trends Neurosci 14: long-term s1 protein stimulation (Romero et al., 2000). 220–223. In summary, the present studies showed the anti-amnesic Fisher A (2000) Therapeutic strategies in Alzheimer’s disease: M1 and neuroprotective potential of a new aminotetrahydrofuran muscarinic agonists. Jpn J Pharmacol 84: 101–112. derivative, against amyloid toxicity. These observations Fisher A, Pittel Z, Haring R, Bar-Ner N, Kliger-Spatz M, Natan N, confirmed the synergy between muscarinic and s1 sites in the et al. (2003) M1 muscarinic agonists can modulate some of the pharmacological profile of these drugs, since the compound hallmarks in Alzheimer’s disease: implications in future therapy. presents only moderate, in the micromolar range, affinities J Mol Neurosci 20: 349–356. for its targets but a sub-mg/kg active dose in vivo. Moreover, Hayashi T and Su TP (2001) Regulating ankyrin dynamics: Roles of sigma-1 receptors. Proc Natl Acad Sci U S A 98: 491–496. the major metabolite of the compound, demethylated Hayashi T and Su TP (2007) Sigma-1 receptor chaperones at the ANAVEX2-73/ANAVEX19-144, is equally active as its ER-mitochondrion interface regulate Ca2þ signaling and cell sur- parent molecule and therefore a long duration has been vival. Cell 131: 596–610. 2þ observed. Altogether, ANAVEX2-73 appears as a lead candi- Hayashi T, Maurice T and Su TP (2000) Ca signaling via sigma1- date of particular interest as a neuroprotective and putatively receptors: novel regulatory mechanism affecting intracellular disease-modifying agent in AD. Ca2þ concentration. J Pharmacol Exp Ther 293: 788–798.

Downloaded from jop.sagepub.com by guest on September 15, 2010 16 Journal of Psychopharmacology 0(0)

Ishikawa M, Sakata M, Ishii K, Kimura Y, Oda K, Toyohara J, et al. amnesia in mice through a common mechanism.

(2009) High occupancy of sigma1 receptors in the human brain 83: 413–428. after single oral administration of donepezil: a positron emission Meunier J and Hayashi T (2010) Sigma-1 receptors regulate Bcl-2 tomography study using [11C]SA4503. Int J Neuropsychopharmacol expression by reactive oxygen species-dependent transcriptional 12: 1127–1131. regulation of nuclear factor kB. J Pharmacol Exp Ther 332: Kato K, Hayako H, Ishihara Y, Marui S, Iwane M and Miyamoto M 388–397. (1999) TAK-147, an acetylcholinesterase inhibitor, increases cho- Meunier J, Ieni J and amd Maurice T (2006) The anti-amnesic and

line acetyltransferase activity in cultured rat septal cholinergic neuroprotective effects of donepezil against amyloid b25-35 pep- neurons. Neurosci Lett 260: 5–8. tide-induced toxicity in mice involve an interaction with the s1 Kelly JF, Furukawa K, Barger SW, Rengen MR, Mark RJ, Blanc receptor. Br J Pharmacol 149: 998–1012. EM, et al. (1996) Amyloid b-peptide disrupts carbachol-induced Mohammed AH (1993) Effects of cholinesterase inhibitors on learn- muscarinic cholinergic signal transduction in cortical neurons. ing and memory in rats: a brief review with special reference to Proc Natl Acad Sci U S A 93: 6753–6758. THA. Acta Neurol Scand 49(Suppl 1): 13–15.

Kihara T, Sawada H, Nakamizo T, Kanki R, Yamashita H, Maelicke Monnet FP and Maurice T (2006) The sigma1 protein as a target for A, et al. (2004) Galantamine modulates nicotinic receptor and the non-genomic effects of neuro(active)steroids: Molecular, blocks Abeta-enhanced glutamate toxicity. Biochem Biophys Res physiological, and behavioral aspects. J Pharmacol Sci 100: Commun 325: 976–982. 93–118. Malviya M, Kumar YC, Asha D, Chandra JN, Subhash MN and Monnet FP, Debonnel G and de Montigny C (1992) In vivo Rangappa KS (2008) Muscarinic receptor 1 agonist activity of electrophysiological evidence for a selective modulation of novel N-arylthioureas substituted 3-morpholino deriva- N-methyl-D-aspartate-induced neuronal activation in rat CA3 tives in Alzheimer’s presenile dementia models. Bioorg Med Chem dorsal hippocampus by sigma ligands. J Pharmacol Exp Ther 16: 7095–7101. 261: 123–130. Mangelus M, Kroyter A, Galron R and Sokolovsky M (2001) Morin-Surun MP, Collin T, Denavit-Saubie´M, Baulieu E-E and

Reactive oxygen species regulate signalling pathway induced by Monnet FP (1999) Intracellular sigma1 receptor modulates phos- M1 muscarinic receptors in PC12M1 cells. J Neurochem 76: pholipase C and protein kinase C activities in the brainstem. Proc 1701–1711. Natl Acad Sci U S A 96: 8196–8199. Marrazzo A, Caraci F, Salinaro ET, Su TP, Copani A and Nakahara N, Fujise N, Kawanishi G and Mizobe F (1990) Central Ronsisvalle G (2005) Neuroprotective effects of sigma-1 receptor muscarinic activities of an M1-selective agonist: preferential effect agonists against b-amyloid-induced toxicity. Neuroreport 16: on reversal of amnesia. Brain Res 507: 172–175. 1223–1226. Nakahara N, Iga Y, Saito Y, Mizobe F and Kawanishi G (1989) Martina M, Turcotte ME, Halman S and Bergeron R (2007) The Beneficial effects of FKS-508 (AF102B), a selective M1 agonist, sigma-1 receptor modulates NMDA receptor synaptic transmis- on the impaired working memory in AF64A-treated rats. Jpn J sion and plasticity via SK channels in rat hippocampus. J Physiol Pharmacol 51: 539–547. 578: 143–157. Ohno M, Yamamoto T and Watanabe S (1994) Blockade of hippo- Matsuno K, Matsunaga K and Mita S (1992) Increase of extracellu- campal M1 muscarinic receptors impairs working memory perfor- lar acetylcholine level in rat frontal cortex induced by (þ)N-allyl- mance of rats. Brain Res 650: 260–266. normetazocine as measured by brain microdialysis. Brain Res 575: Olariu A, Tran M-H, Yamada K, Mizuno M, Hefco V and 315–319. Nabeshima T (2001) Memory deficits and increased emotionality

Matsuno K, Matsunaga K, Senda T and Mita S (1993) Increase in induced by b-amyloid25-35 are correlated with the reduced acetyl- extracellular acetylcholine level by s ligands in rat frontal cortex. release and altered phorbol dibutyrate binding in the hip- J Pharmacol Exp Ther 265: 851–859. pocampus. J Neural Transm 108: 1065–1079. Matsuno K, Senda T, Kobayashi T, Okamoto K, Nakata K and Pedigo NW Jr (1994) receptor plasticity in aging.

Mita S (1997) SA4503, a novel cognitive enhancer, with s1 recep- Life Sci 55: 1985–1991. tor agonistic properties. Behav Brain Res 83: 221–224. Perry EK, Tomlinson BE, Blessed G, Bergmann K, Gibson PH and Maurice T and Su TP (2009) The of sigma-1 receptors. Perry RH (1978) Correlation of cholinergic abnormalities with Pharmacol Ther 124: 195–206. senile plaques and mental test scores in senile dementia. Br Med Maurice T, Hiramatsu M, Itoh J, Kameyama T, Hasegawa T and J 2: 1457–1459. Nabeshima T (1994) Behavioral evidence for a modulating role of Quirion R, Richard J and Wilson A (1994) Muscarinic and nico- s ligands in memory processes. I. Attenuation of dizocilpine tinic modulation of cortical acetylcholine release monitored by (MK-801)-induced amnesia. Brain Res 647: 44–56. in vivo microdialysis in freely moving adult rats. Synapse 17: Maurice T, Junien JL and Privat A (1997) 92–100. sulfate attenuates dizocilpine-induced learning impairment in Romero G, Pe´rez MP, Carceller A, Monroy X, Farre´AJ and Guitart

mice via s1-receptors. Behav Brain Res 83: 159–164. X (2000) Changes in phosphoinositide signalling activity and Maurice T, Lockhart BP and Privat A (1996) Amnesia induced in levels of the subunit of G(q/11) protein in rat brain induced

mice by centrally administered beta-amyloid peptides involves by E-5842, a sigma1 receptor ligand and potential atypical anti- cholinergic dysfunction. Brain Res 706: 181–193. psychotic. Neurosci Lett 290: 189–192. Maurice T, Meunier J, Feng B, Ieni J and Monaghan DT (2006) Selkoe DJ (1991) The molecular pathology of Alzheimer’s disease.

Interaction with sigma1 protein, but not N-methyl-D-aspartate Neuron 6: 487–498. receptor, is involved in the pharmacological activity of donepezil. Senda T, Matsuno K, Kobayashi T and Mita S (1997) Reduction J Pharmacol Exp Ther 317: 606–614. of the scopolamine-induced impairment of passive-avoidance Maurice T, Phan VL, Urani A, Kamei H, Noda Y and Nabeshima T performance by s receptor agonist in mice. Physiol Behav 61: (1999) Neuroactive neurosteroids as endogenous effectors for the 257–264.

sigma1 (s1) receptor: pharmacological evidence and therapeutic Senda T, Matsuno K, Okamoto K, Kobayashi T, Nakata K and opportunities. Jpn J Pharmacol 81: 125–155. Mita S (1996) Ameliorating effect of SA4503, a novel s1 receptor Maurice T, Su TP and Privat A (1998) Sigma1 (s1) receptor agonists agonist, on memory impairments induced by cholinergic dysfunc- and neurosteroids attenuate b25-35-amyloid peptide-induced tion in rats. Eur J Pharmacol 315: 1–10.

Downloaded from jop.sagepub.com by guest on September 15, 2010 Villard et al. 17

Strosznajder JB, Zambrzycka A, Kacprzak MD and Strosznajder RP Vamvakides A (2002b) of the tetrahydro-N, (1999) Amyloid b peptide 25–35 modulates hydrolysis of phos- N-dimethyl-5,5-diphenyl-3-furanemethanamine, a putative noot- phoinositides by membrane phospholipase(s) C of adult brain ropic, anti-epileptic and antidepressant compound. Ann Pharm Fr cortex. J Mol Neurosci 12: 101–109. 60: 415–422. Svensson AL and Nordberg A (1998) Tacrine and donepezil attenu- Villard V, Espallergues J, Keller E, Alkam T, Nitta A, Yamada K, ate the neurotoxic effect of Ab25-35 in rat PC12 cells. Neuroreport et al. (2009) Antiamnesic and neuroprotective effects of the ami- 9: 1519–1522. notetrahydrofuran derivative ANAVEX1-41 against amyloid

Urani A, Roman FJ, Phan VL, Su TP and Maurice T (2001) The b25-35-induced toxicity in mice. 34: antidepressant-like effect induced by sigma1-receptor agonists and 1552–1566. neuroactive steroids in mice submitted to the forced swimming Yamada K and Nabeshima T (2000) Animal models of Alzheimer’s test. J Pharmacol Exp Ther 298: 1269–1279. disease and evaluation of anti-dementia drugs. Pharmacol Ther Vamvakides A (2002a) Anticonvulsant and forced swim anti- 88: 93–113. immobility effects of tetrahydro-N,N-dimethyl-2,2-diphenyl- 3-furanemethanamine (AE37): common action mechanism? Ann Pharm Fr 60: 88–92.

Downloaded from jop.sagepub.com by guest on September 15, 2010