at SciVerse ScienceDirect

Neuropharmacology 63 (2012) 1042e1050

Contents lists available

Neuropharmacology

journal homepage: www.elsevier .com/locate/neuropharm

Anti-amnesic effect of neurosteroid PREGS in Ab25e35-injected mice through s1receptor- and a7nAChR-mediated neuroprotection

Rong Yang a,b, Lei Chen a,**, Haofei Wang a, Bingzhong Xu a, Hidekazu Tomimoto c, Ling Chen a,b,*

aDepartment of Physiology, Nanjing Medical University, Nanjing 210029, Chinab State Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing 210029, ChinacDepartment of Neurology, Mie University Graduate School of Medicine, Edobashi 2-174, Tsu, Mie, Japan

a r t i c l e i n f o

Article history:Received 6 April 2011Received in revised form16 July 2012Accepted 19 July 2012

Keywords:Pregnenolone sulfate (PREGS)b-Amyloid (Ab)a7 nicotinic acetylcholine receptor(a7nAChR)Sigma-1 receptor (s1R)

Abbreviations: AD, Alzheimer’s disease; Ab,dimethoxybenzylidene)-anabaseine dihydrochloridelinyl)-8-phenyl-1(4H)-benzopyran-4-one hydrochloriMAPK, mitogenic activated protein kinase; NMDAr, NNE100, N,N-dipropyl-2-[4-methoxy-3-(2-phenylethoxchloride; ERK, extracellular signal-related kinase; Pkinase; PREG, pregnenolone; PREGS, pregnenolomorpholinethyl)-1-phenylcyclohexanecarboxylate hdiamino-2,3-dicyano-1-4-bis [2-aminophynylthio]nicotinic acetylcholine receptor; GABAAR, g-aminobu* Corresponding author. Laboratory of Reproducti

Physiology, Nanjing Medical University, HanzhongChina. Tel.: þ86 25 86862878; fax: þ86 25 86260332** Corresponding author. Tel.: þ86 25 86862878; fa

E-mail addresses: lingchen@njmu.edu.cn (L. Chen),

0028-3908/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.neuropharm.2012.07.035

a b s t r a c t

A single intracerebroventricular injection of b-amyloid 25e35 peptide (Ab25e35) (9 nmol/mouse) inducesthe spatial cognitive deterioration and approximately 50% loss of pyramidal cells in hippocampal CA1region within 1 week. The present study focused on exploring the effects of neurosteroid pregnenolonesulfate (PREGS), in comparison with the selective agonists of sigma-1 receptor (s1R) and a7 nicotinicacetylcholine receptor (a7nAChR), on the cognitive deficits and the death of pyramidal cells in Ab25e35-mice. Herein, we reported that the administration of PREGS (1e100 mg/kg) for 7 days after Ab25e35-injection could dose-dependently ameliorate the cognitive deficits and attenuate the apoptosis ofpyramidal cells. Either the s1R antagonist NE100 or the a7nAChR antagonist MLA could block theneuroprotection of PREGS in Ab25e35-mice. Both the s1R agonist PRE084 and the a7nAChR agonist DMXBcould mimic the PREGS-neuroprotection against the Ab25e35-neurotoxicity. The neuroprotection ofPRE084 was attenuated by MLA, but the DMXB-action was insensitive to NE100. The neuroprotection ofPREGS, PRE084 or DMXB was blocked by the phosphatidylinositol-3-kinase (PI3K) inhibitor LY294002,whereas only the effect of PREGS or PRE084 was sensitive to the MAPK/ERK kinase (MEK) inhibitorU0126. PREGS prevented Ab25e35-inhibited Akt (Serine/threonine kinase) phosphorylation leading toincrease in caspase-3 activity, which was s1R- and a7nAChR-dependent. By contrast, PREGS-rescuedreduction of extracellular signal-related kinase-2 (ERK2) phosphorylation in Ab25e35-mice onlyrequired the activation of s1R. Blockage of PREGS-neuroprotection by LY294002 significantly attenuatedits anti-amnesic effect in Ab25e35-mice. The findings indicate that the anti-amnesic effects of PREGS inAb25e35-mice depend on the s1R- and a7nAChR-mediated neuroprotection.

� 2012 Elsevier Ltd. All rights reserved.

b-amyloid; DMXB, 3-(2,4-; LY294002, 2-(4-morpho-de; MLA, methyllycaconitine;-methyl-D-aspartate receptor;y)phenyl] ethylamine hydro-I3K, phosphatidylinositol-3-ne sulfate; PRE084, 2-(4-ydrochloride; U0126, 4-butadiene; a7nAChR, a7

tyric acid type A receptor.ve Medicine, Department ofRoad 140, Nanjing 210029,.x: þ86 25 86260332.chenl@njmu.edu.cn (L. Chen).

All rights reserved.

1. Introduction

Alzheimer’s disease (AD) is characterized by a progressiveneurodegeneration leading to dementia and death. The accumu-lation and deposition of b-amyloid (Ab) within the brain is thoughtto be the primary force driving the pathogenesis of AD. Ab has beenreported to disrupt neuronal cells by calcium dyshomeostasis(Nelson et al., 2007) and enhancement of excitotoxicity (Rothmanand Olney, 1995) and apoptosis (Forloni et al., 1993). At sites ofAb-aggregation, hippocampal neuronal cells die leading to therapid loss of learning and memory (Yan et al., 1996).

Cholinergic system is very sensitive to the Ab-neurotoxicity. TheAb-induced dysfunction of a7 nicotinic acetylcholine receptor(a7nAChR) results in the impairment of spatial memory (Chenet al., 2006, 2010). The a7nAChR agonists are able to attenuatethe Ab-neurotoxicity and glutamate toxicity in cultured neurons(Kihara et al., 1997; Zamani et al., 1997). The treatment with DMXB,

R. Yang et al. / Neuropharmacology 63 (2012) 1042e1050 1043

a selective a7nAChR agonist, ameliorates the Ab25e35-induceddeficits in spatial memory (Chen et al., 2010). Furthermore, nico-tine not only inhibits beta-amyloidosis (Hellström-Lindahl et al.,2004) but also prevents the accumulation of Ab in APP transgenicmice (Liu et al., 2007). On the other hand, the activation of sigma-1receptor (s1R) may induce important neuroprotection against theAb25e35-neurotoxicity (Li et al., 2010). The selective s1R agonists arepotent neuroprotective drugs as observed in excitotoxicity models(Maurice and Lockhart, 1997; Nakazawa et al., 1998) and Ab-induced toxicity in cortical neurons in vitro (Marrazzo et al., 2005).Meunier et al. (2006) have reported the potent anti-amnesic andneuroprotective effects of donepezil against Ab25e35-neurotoxicitythrough its cholinergic and s1R agonistic properties.

Steroids which are synthesized within the central or peripheralnervous system have been named “neurosteroids” (Baulieu, 1997).Recently, the correlation between decreased levels of neurosteroidsand neuronal degeneration in AD patients has been paid closeattention (Naylor et al., 2010). Pregnenolone sulfate (PREGS),a steroid synthesized de novo in the brain, is thought to relate withcognitive performance in senescent animals (Schumacher et al.,2008). Deficient cognitive performance in aged rats can be cor-rected by intrahippocampal injection of PREGS (Vallée et al., 1997).In addition to as a negative modulator of GABAA receptors (Akket al., 2001) and positive allosteric modulator of N-methyl-D-aspartate receptor (NMDAr, Maurice et al., 2006), PREGS has beendemonstrated to enhance the function of a7nAChR (Chen andSokabe, 2005) and s1R (Monnet et al., 1995). Collectively, it isspeculated that PREGS can antagonize the Ab-neurotoxicitythrough its a7nAChR and s1R agonistic actions.

Ab1e42, a major constituent of senile plaques, is well known tobe one of the candidates causing memory loss, because numerousstudies have demonstrated a significant correlation between thenumber of senile plaques and the degree of cognitive deficits in ADbrains. Among the Ab fragments studied so far, peptide bearing the11 amino acids (25e35) (Ab25e35) is the shortest fragment of Abprocessed in vivo by brain proteases (Kubo et al., 2002). This peptideretains the ability to self-aggregate and mediates the toxicity of thefull-length peptide, though it lacks a hydrophobic C terminalsequence of five amino acids as compared to Ab1e40 (Burdick et al.,1992). It has been proposed that Ab25e35 represents the biologicallyactive region of Ab1e42 (Pike et al., 1995). Experiments usingtransgenic and gene targeting mouse models have shown a closeassociation between excess amounts of Ab and the deficits inlearning and memory. Thus, two nontransgenic rodent models ofAD created by intracerebroventricular (i.c.v.) infusion of Ab1e40/42(Chen et al., 2006) or Ab25e35 (Maurice et al., 1996) have beenwidely used to analyze the morphological and behavioral conse-quences of Ab-neurotoxicity in vivo. Consistent with the Ab1e40/42infusion (i.c.v.) in rats, the injection (i.c.v.) of Ab25e35 in miceinduces, within 1 or 2 weeks after administration, histological andbiochemical changes in cholinergic system (Kowall et al., 1991,1992). A single injection (i.c.v.) of either Ab1e40/42 (Wu et al., 2008)or Ab25e35 (Chen et al., 2010) in rats and mice can impair theinduction of long-term potentiation (LTP) in hippocampal CA1region. A single injection (i.c.v.) of aggregated Ab25e35 (3 nmol/mouse) in mice induces amnesia in many kinds of behaviorexperiments such as in Y-maze, step-down type passive avoidanceand Morris Water maze, although the low-dose of Ab25e35 does notproduce neuronal cell loss (Maurice et al., 1996; Wang et al., 2007).By contrast, one single injection (i.c.v.) of Ab25e35 at high dose of9 nmol/mouse can cause the death of hippocampal neuronal cells.Therefore, the present study focused on exploring the anti-amnesicand neuroprotective effects of PREGS, in comparison withthe selective agonists of s1R and a7nAChR, after a single injectionof Ab25e35 (9 nmol/mouse) in mice by examining the spatial

memory behavioral, hippocampal morphological and biochemicalchanges. Our results showed that the administration of PREGSin Ab25e35-mice exerts a potent anti-amnesic effect throughs1R- and a7nAChR-mediated PI3KeAkt and ERK neuroprotectivemechanisms.

2. Materials and methods

2.1. Subjects

The present studies were approved by Animal Care and Ethical Committee ofNanjing Medical University. All procedures were in accordance with the guidelinesof Institute for Laboratory Animal Research of the Nanjing Medical University. Malemice (ICR, Oriental Bio Service Inc., Nanjing), weighing 20e25 g (8 weeks old) at thebeginning of the experiment, were used throughout the study. All animals werehoused in a light controlled room under a 12-h lightedark cycle starting at AM 7:00and kept at a temperature of 23 �C in the Animal Research Center of NanjingUniversity. They received food and water ad libitum. All efforts were made tominimize animal suffering and to reduce the number of animals used.

2.2. Preparation of an animal model of Alzheimer’s disease

Ab25e35 (Sigma, St. Louis, MO, USA) was dissolved in distilled water at theconcentration of 3 mM. “Aggregated” Ab25e35 was obtained by incubating at 37 �Cfor 4 days according to previous report (Maurice et al., 1996) and then diluted to thefinal concentrationwith saline just before the experiment. Mice were anaesthetizedwith intraperitoneal (i.p.) injection of chloral hydrate (400 mg/kg) and placed ina sterotactic device (Kopf Instruments, Tujunga, CA). For a single injection (i.c.v.) ofAb25e35, a 28-G stainless-steel needle (Plastics One, Roanoke, VA) was inserted intolateral ventricular (0.3 mm posterior, 1.0 mm lateral, and 2.5 mmventral to bregma),and then the “aggregated” Ab25e35 (9 nmol/3 ml/mouse) was injectedwith a stepper-motorized micro-syringe (Stoelting, Wood Dale, IL, USA) at a rate of 0.5 ml/min. Theinjection site was confirmed in preliminary experiments by injecting Indian ink.Control mice were given an equal volume of vehicle.

2.3. Drug administration

PREGS, when intraperitoneally (i.p.) injected, may cross bloodebrain barrier andcan be taken up by the brain (Higashi et al., 2003). PREGS and PREG were dissolvedin dimethyl sulfoxide (DMSO), and then diluted in sesame oil to a final concentrationof 1.0% DMSO, because the high dose of PREGS (100 mg/kg) could not be dissolved in1.0% DMSO diluted by saline. PREGS (1e100 mg/kg) or PREG (20 mg/kg) wassubcutaneously (s.c.) injected at 100 ml once daily. s1R agonist PRE084 (1.0 mg/kg)and a7nAChR agonist DMXB (5 mg/kg, Taisho Pharmaceuticals, Tokushima, Japan)were dissolved by 0.9% saline and all of these drugs were intraperitoneally injectedonce daily on 1e7 days after Ab25e35-injection. NMDAr antagonist MK801 (2 mg/kg,i.p.) and s1R antagonist NE100 (3 mg/kg, i.p. Taisho Pharmaceutical Co. Ltd. Tokyo,Japan) were dissolved by distilled water, and a7nAChR antagonist methyl-lycaconitine (MLA) (0.1 nmol/mouse, i.c.v.), MAPK/ERK kinase (MEK) inhibitorU0126 (0.3 nmol/mouse, i.c.v.) and PI3K inhibitor LY294002 (0.3 nmol/mouse, i.c.v.)were dissolved in DMSO, and then in 0.9% saline to a final concentration of 1% DMSO.These antagonists were injected 30 min before treatment with PREGS or agonists.Chemicals, unless stated, all came from Sigma Chemical Company (St Louis, MO,USA). For repeated injection (i.c.v.) of drugs, a 26-G stainless-steel guide cannula(Plastics One, Roanoke, VA) was implanted into the right lateral ventricle (0.3 mmposterior to bregma, 1.0 mm lateral, and 2.3 mm ventral) and anchored to the skullwith four stainless steel screws and dental cement. The drugs were prepared freshlyon the day of experiment and were injected using a 28-G stainless-steel needlecombining with stepper-motorized micro-syringe (Stoelting, Wood Dale, IL, USA) ata rate of 0.5 ml/min (final volume 3 ml/mouse). Control mice were given an equalvolume of vehicle.

2.4. Behavioral analysis

For the Morris water maze task, a pool (diameter 90 cm) made of black-coloredplastic was prepared, with water temperature maintained at 20 � 1 �C. Swimmingpaths were analyzed using a computer systemwith a video camera (AXIS-90 Target/2, Neuroscience). In the hidden platform test, the platform (7 cm in diameter) wassubmerged 1 cm below thewater surface. On the first and the last day of water mazetraining, the swimming speed was assessed in the absence of the platform. Micewere given 90 s to reach the hidden platform. Four starting positions were used andeach mouse was trained with four trials per day. After reaching the platform, themouse was allowed to remain on it for 10 s. If the mouse did not find the platformwithin 90 s, the trial was terminated and the animal was put on the platform for 10 s.The water maze task was consecutively performed on day 3e7 after Ab25e35-injection. Average swimming speed (cm/sec) and latency (sec) to reach the plat-form were scored on all trials and analyzed where appropriate.

0

10

20

30

40

50 **

Lat

ency

(sec

)(o

n da

y 5

afte

r tr

aini

ng )

control mice Aβ

25-35 mice

PR

EG

PR

EG

S

10

20

30

40

50

Lat

ency

(sec

)(o

n da

y 5

afte

r tr

aini

ng)

1 5 10 20 40 100

Dose of PREGS (mg/kg)

1 2 3 4 5

20

40

60

80

Lat

ency

(sec

)

Time post-training (day)

control mice Aβ

25-35 mice

Aβ25-35

/PREGS

0

10

20

30

40

50

60

##

**

Lat

ency

(sec

)(o

n da

y 5

afte

r tr

aini

ng )

control mice Aβ

25-35 mice

PR

EG

S

A

B C

Fig. 1. Beneficial effect of PREGS on behavioral performance in Ab25e35-mice. (A) TheMorris water test revealed that the latency to reach the hidden-platform was pro-longed in Ab25e35-mice as compared with that in control mice. Treatment with PREGS(20 mg/kg) decreased the prolongation of latency in Ab25e35-mice. The latency wasrecorded from day 3 to day 7 after Ab25e35-injection. The histogram shows the latencyon day 5 after training. (B) The dose-dependent curve shows the effect of PREGS(1e100 mg/kg, s.c.) on the latency in Ab25e35-mice. PREGS exerted the maximal effectat dose of 20 mg/kg. (C) Treatment with PREG (20 mg/kg) had no effect on theprolongation of latency in Ab25e35-mice. **P < 0.01 vs. control mice, ##P < 0.01 vs.vehicle-treated Ab25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e10501044

2.5. Histological examination

Histological examination of hippocampal CA1 regionwas assessed on day 7 afterAb25e35-injection. Mice were anesthetized with chloral hydrate (400mg/kg, i.p.) andperfused with ice-cold phosphate-buffered saline (PBS) followed by 4% para-formaldehyde. Brains were removed and immersed in fixative (4 �C overnight), andthen processed for paraffin embedding. Coronal sections (5 mm) were cut from thelevel of hippocampus for toluidine blue and TUNEL staining.

In toluidine blue staining, the pyramidal cells in hippocampal CA1 region wereidentified using a conventional light microscope (Olympus DP70) with a 60�objective, healthy CA1 pyramidal neurons showed a round cell body with a plainlystained nucleus. Density of surviving neurons was counted in 6 sections per mouseand expressed as the number of cells per mm length along the hippocampal CA1pyramidal layer (Cai et al., 2008).

For terminal deoxynucleotide transferase-mediated dUTP nick end labeling(TUNEL) staining, sections were deparaffinized in xylene, rehydrated througha series of graded alcohols. Subsequently, the sections were permeabilized usingproteinase K (20 mg/ml) and then incubated in 3% H2O2 in 0.1 M PBS for 30 min inorder to quench endogenous peroxidase activity. TUNEL reaction performed using InSitu Cell Death Detection Kit (Roche, Mannheim, Germany). Briefly, sections wereincubated with TUNEL reaction mixture for 60 min at 37 �C. The sections weretreated with Converter-POD solution and visualized by 3,30-diaminobenzine (DAB,Chromagen Kit, Vector Laboratories) for 10 min. For positive controls, sections wereincubated with DNase I for 10 min at 15e25 �C to induce DNA strand breaks, prior tolabeling procedure. For negative controls, sections were incubated with label solu-tion only (without terminal transferase) instead of TUNEL reaction mixture. Thesections were observed using an Olympus (DP70) microscope with a 60� objective.

2.6. Western blot analysis

All mice were decapitated under deep anesthesia with ethyl ether on day 7 afterAb25e35-injection. The hippocampus was taken quickly and homogenized in a lysisbuffer (50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 5 mM EDTA, 10 mM NaF, 1 mMsodium orthovanadate, 1% Triton X-100, 0.5% sodium deoxycholate, 1 mM phenyl-methylsulfonyl fluoride and protease inhibitor cocktail Complete; Roche, Man-nheim, Germany). Total proteins (20 mg) were separated by SDS-polyacrylamide gelelectrophoresis (SDS-PAGE) and transferred to a polyphorylated difluoride (PVDF)membrane. The membranes were incubated with 5% nonfat dried milk for 60 min,and then were incubated with a rabbit monoclonal anti-phosphor ERK1/2 antibody(1:1000, Cell Signaling, Beverly, MA), a rabbit monoclonal anti-phosphor Akt anti-body (1:1000), or a rabbit monoclonal anti-caspase-3 antibody (1:1000) at 4 �Covernight. After being washed with TBST, the membranes were incubated with anHRP-labeled secondary antibody and developed using the ECL detection Kit(Amersham Biosciences, Piscataway, NJ). Following visualization, the blots werestripped by stripping buffer (Restore, Pierce Chemical Co, Rockford IL) for 15 min, re-blocked with 5% nonfat dried milk for 60 min, then incubated with anti-ERK1/2antibody (1:1000), anti-Akt (1:1000) or anti-action (1:1000, Chemicon Interna-tional). Western blot bands were scanned and analyzed with the image analysissoftware package, National Institutes of Health Image. Samples were collected fromthe Ab25e35-infused hemisphere of 3 mice as a set of western blot analysis. Thesummarized data represent the average of 3 experimental sets (n ¼ 9 mice).

2.7. Data analysis

Data were retrieved and processed using the software Micro Origin 6.1 (OriginLab, Northampton, MA, USA). All of data were presented as means � standard error(SE). The significance was tested by ANOVA followed by Bonferroni’s post-hoc testfor multiple comparisons. Statistical analyses were performed using the Stata 7.0software program (STATA Corporation, USA). P values less than 0.05 were acceptedas statistically significant.

3. Results

3.1. PREGS improves Ab25e35-impaired spatial memory

The results of Morris water maze test (Fig. 1A) showed thata single injection of “aggregated” Ab25e35 (9 nmol/mouse) signifi-cantly increased the escape-latency to reach the hidden platformcompared to control mice (P < 0.01 on day 5 after training, n ¼ 16),whereas it did not affect the swimming speed (control:14.71 �1.44 cm/s; Ab25e35-mice: 13.07 � 1.63 cm/s; P > 0.05). Thetreatment with PREGS in Ab25e35-mice could attenuate theprolongation of escape-latency in a U-shaped dose-dependentmanner (P < 0.01, n ¼ 16; Fig. 1B). Notably, PREGS at 20 mg/kgshowed the most profound effect against the Ab25e35-impaired

spatial cognition, so this dose was used in the following experi-ments. By contrast, the administration of PREG (20 mg/kg), a non-sulfated form of PREGS, had no effect on the impairment of spatialmemory in Ab25e35-mice (P > 0.05, n ¼ 16; Fig. 1C).

3.2. PREGS reduces apoptosis of pyramidal cells in Ab25e35-mice

To explore mechanisms concerning the anti-amnesic effect ofPREGS, we examined pyramidal cells in hippocampal CA1 region.The results showed that the number of pyramidal cells was reducedapproximately 50% of control mice on day 7 after Ab25e35-injection(P< 0.01, n¼ 8; Fig. 2A). The treatment with PREGS in Ab25e35-micemarkedly attenuated the loss of pyramidal cells (P < 0.01, n ¼ 8),but PREG had no such effect (P > 0.05, n ¼ 8). Compared with thecontrol mice (left panel), the number of apoptotic cells (TUNELpositive cells) was significantly increased in Ab25e35-mice (centralpanel), which could be attenuated by the administration of PREGS(right panel; Fig. 2B). The results indicate that PREGS can protectneuronal cells against Ab-neurotoxicity.

3.3. s1R and a7nAChR are involved in PREGS-neuroprotection

To determine the targets of PREGS-neuroprotection, we inves-tigated the involvement of s1R, NMDAr and a7nAChR using theselective antagonists of the receptors. The results showed that thepre-treatment with the s1R antagonist NE100 almost completelyblocked the neuroprotection of PREGS in Ab25e35-mice (P < 0.01,n ¼ 8; Fig. 3A), while the a7nAChR antagonist MLA partially antag-onized the PREGS effects (P < 0.05, n ¼ 8). The NMDAr antagonistMK801 did not affect the neuroprotection of PREGS in Ab25e35-mice

control Aβ25-35 Aβ25-35/PREGS

control

A 25-35/PREGS

A 25-35

0

100

200

300

400

500

##

**

Surv

ivin

g ce

lls (m

m-1

)

control mice Aβ

25-35 mice

PR

EG

PR

EG

S

PR

EG

S

PR

EG

B

A

Fig. 2. Neuroprotection of PREGS on Ab25e35-induced neuronal death on day 7 after Ab25e35-injection. (A) Toluidine blue staining was used to evaluate the death of pyramidal cellsin hippocampal CA1 region. Bar graph represents the number of surviving pyramidal cells. (B) TUNEL staining in hippocampal CA1 region detected the pyramidal cell apoptosis.Note that the administration of PREGS significantly reduced the apoptosis of Ab25e35-induced pyramidal cells. Black arrows indicate apoptosis pyramidal cells. Scale bars ¼ 50 mM.**P < 0.01 vs. control mice; ##P < 0.01 vs. vehicle-treated Ab25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e1050 1045

(P > 0.05, n ¼ 8), although it could reduce the death of pyramidalcells in Ab25e35-mice treated with vehicle (P < 0.05, n ¼ 8).

To confirm whether s1R or a7nAChR is involved in the PREGS-neuroprotection, the s1R agonist PRE084 or the a7nAChR agonistDMXB was administered as the same protocol as the treatmentwith PREGS. We observed that the treatment with either PRE084 orDMXB in Ab25e35-mice could protect the pyramidal cells (P < 0.01,n ¼ 8; Fig. 3B and C). Interestingly, the PRE084-neuroprotectioncould be blocked by MLA (P < 0.01, n ¼ 8; Fig. 3B), whereas theDMXB-neuroprotection was insensitive to NE100 (P > 0.05, n ¼ 8;Fig. 3C). The results indicate that PREGS targets s1R and a7nAChR toprevent the Ab-neurotoxicity.

3.4. Involvement of PI3eAkt or ERK on s1R- and a7nAChR-mediatedneuroprotection in Ab25e35-mice

Earlier in vitro study (Kihara et al., 2004) found that a7nAChR-mediated neuroprotection was related to PI3KeAkt signalingpathway rather than ERK1/2. Our results showed that in Ab25e35-mice, the PI3K inhibitor LY294002 could block the neuro-protection produced by PREGS, DMXB or PRE084 (P < 0.01, n ¼ 8;Fig. 4A). By contrast, the MEK inhibitor U0126 could attenuate theneuroprotective effects of PREGS and PRE084 (P < 0.05, n ¼ 8;Fig. 4B), but not the effect of DMXB (P > 0.05, n ¼ 8). The admin-istration of LY294002 or U0126 failed to affect the number ofpyramidal cells in hippocampal CA1 region in control mice(P > 0.05, n ¼ 8; Fig. 4A and B).

The phosphorylation level of Akt (p-Akt) in hippocampus ofAb25e35-mice was lower than that in control mice (P < 0.05;Fig. 5A). Treatment with PREGS could rescue the reduction of p-Aktin Ab25e35-mice (P < 0.01), which was sensitive to NE100 and MLA

(P < 0.01). The proapoptotic protein caspase-3 activity wasmeasured by cleavage of the caspase-3 substrate. Elevation of thehippocampal caspase-3 activity was found in Ab25e35-mice(P < 0.01; Fig. 5B). PREGS significantly attenuated the increase incaspase-3 activity in Ab25e35-mice (P < 0.01), which was alsodependent on s1R (P < 0.05) and a7nAChR (P < 0.01). In addition,the phosphorylation level of hippocampal ERK2 (p-ERK2) waslower in Ab25e35-mice than that in control mice (P < 0.05; Fig. 5C).The treatment with PREGS in Ab25e35-mice could rescue thedecreased level of p-ERK2 (P < 0.01), which was blocked by NE100(P < 0.01), but not by MLA (P > 0.05).

3.5. Anti-amnesic effect of PREGS neuroprotection in Ab25e35-mice

To further explore whether the anti-amnesic effect of PREGSresults from its neuroprotection, the PI3K inhibitor LY294002 wasused to block the neuroprotection of PREGS in Ab25e35-mice(Fig. 4A). The results showed that the application of LY294002alone did not affect the spatial cognitive performance in controlmice (P > 0.05, n ¼ 16; Fig. 6), but the treatment with LY294002significantly attenuated the anti-amnesic effect of PREGS on theprolongation of escape-latency in Ab25e35-mice (P < 0.01, n ¼ 16).

4. Discussion

One earlier study (Maurice et al., 1996) reported that thetreatmentwith PREGS could attenuate cognitive deficits induced bya low-dose of Ab25e35 (3 nmol/mouse) without the loss of neuronalcells. The high dose of Ab25e35 (9 nmol/mouse) caused the cognitiveimpairment associated with the apoptosis of neuronal cells(Meunier et al., 2006). The results in the present study showed that

0

50

100

150

200

250

300

350

400

#*

Surv

ivin

g ce

lls (m

m-1

)

A25-35

mice A

25-35 mice/PREGS

##

ML

A

NE

100

MK

801

ML

A

NE

100

MK

801

NE

100

0

50

100

150

200

250

300

350

** **

Surv

ivin

g ce

lls (m

m-1)

Aβ25-35 mice Aβ25-35 mice/DMXB

NE

100

0

50

100

150

200

250

300

350

**

Surv

ivin

g ce

lls (m

m-1)

Aβ25-35 mice Aβ25-35 mice/PRE084

++

ML

A

ML

A

A

B C

Fig. 3. a7nAChR and s1R are involved in the neuroprotection of PREGS in Ab25e35-mice. (A) The a7nAChR antagonist MLA, the s1 receptor antagonist NE100 and the NMDArantagonist MK801 were administrated at 30 min before the treatment of PREGS. Bar graph shows the number of surviving pyramidal cells of hippocampal CA1 region in Ab25e35-mice. (B & C) Neuroprotective effects of the s1R agonist PRE084 and the a7nAChR agonist DMXB on hippocampal CA1 pyramidal cells. MLA could block the neuroprotection ofPRE084, but NE100 has no effect on DMXB-neuroprotection. *P < 0.05 and **P < 0.01 vs. vehicle-treated Ab25e35-injected mice; #P < 0.05 and ##P < 0.01 vs. PREGS-treated Ab25e35-mice; þþP < 0.01 vs. PRE084-treated Ab25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e10501046

the administration of PREGS could improve the cognitive deficitsand attenuate the apoptosis of neuronal cells. The neuroprotectionof PREGS against Ab-neurotoxicity depended on the activation ofs1R- and a7nAChR to cascade the PI3KeAkt and ERK signalingpathways. Because the anti-amnesic effect of PREGS was sensitiveto the PI3K inhibitor, the findings give an indication that PREGSimproves the cognitive deficits in Ab25e35-mice through not onlyrescuing dysfunction of neuronal system but also protecting theneuronal cells.

4.1. Targets of PREGS-neuroprotection against Ab25e35 toxicity

The findings in the present study give a clear indication that thePREGS-neuroprotection against Ab25e35-induced apoptosisdepends on the functions of s1R and a7nAChR. This conclusion isdeduced mainly from the following observations: (1) the neuro-protection of PREGS in Ab25e35-mice was blocked by the antago-nists of s1R and a7nAChR, but not the NMDAr antagonist; (2) theagonists of either s1R or a7nAChR could mimic the neuroprotectionof PREGS to reduce the death of pyramidal cells in Ab25e35-mice; (3)the PREGS-effect on rescuing PI3KeAkt and ERK signals in Ab25e35-mice was also sensitive to the antagonist of s1R or a7nAChR.Because PREGS has been known to activate s1R (Meyer et al., 2002)and a7nAChR (Chen and Sokabe, 2005), it is indicated that PREGSdirectly targets s1R and a7nAChR, respectively. Consistent with theneuroprotective action of PRE084 in Ab25e35-mice, the s1R agonistattenuates cell death in cultured cortical neurons coincubated withAb25e35, and this effect was reversed by the selective s1R

antagonist NE100 (Marrazzo et al., 2005). Moreover, in themodel ofAD-type amnesia induced by low-dose of Ab25e35, which involvesboth cholinergic and glutamatergic neurotransmission throughNMDAr, the s1R agonists and PREGS attenuate amnesia in a bell-shaped manner. Interestingly, our results showed that the s1R-mediated neuroprotection in Ab25e35-mice was blocked by thea7nAChR antagonist, while the a7nAChR-action was insensitive tothe blockade of s1R. Thus, another possibility is that the PREGS-induced activation of s1R further modulates the function ofa7nAChR. Detailed analysis concerning the correlation betweens1R and a7nAChR remains to be done in the future works. One theother hand, the reduction of BDNF may be one of the major path-ological conditions in AD. In hippocampus of APP/PS1 mice, thereduced BDNF could be rescued by the application of PREGS (Xuet al., 2012). BDNF activates a variety of signal molecules,including Ras/ERK and PI3K/Akt, all of which are required forsurvival of neuronal cells. The activation of s1R has been reportedto selectively enhance the BDNF signaling and up-regulate theBDNF mRNA in the rat brain (Yagasaki et al., 2006). In addition,activation of s1 receptors facilitates neuroprotection or neuronalrecovery through regulating the localization of growth factorreceptors in lipid rafts (Takebayashi et al., 2004).

4.2. Possible mechanisms of PREGS neuroprotection againstAb25e35-toxicity

A large body of evidence emerges that Ab25e35 causes thenecrosis of pyramidal cells in hippocampal CA1 region (Mamiya

0

100

200

300

400

**

Surv

ivin

g ce

lls (m

m-1

)

CO Aβ

25-35 mice/PREGS

Aβ25-35

mice/DMXB Aβ

25-35 mice/PRE084

U01

26

U01

26

U01

26

U01

26

0

100

200

300

400

** ****

Surv

ivin

g ce

lls ( m

m-1

) CO Aβ

25-35 mice/PREGS

Aβ25-35

mice/DMXB Aβ

25-35 mice/PRE084

LY

2940

02

LY

2940

02

LY

2940

02LY

2940

02

A

B

Fig. 4. PREGS-neuroprotection in Ab25e35-mice is via s1R- and a7nAChR-mediated ERKor PI3keAkt signaling pathways. (A) PI3K inhibitor LY294002 antagonized PREGS-,PRE084- and DMXB-neuroprotection on pyramidal cells of hippocampal CA1 region inAb25e35-mice. (B) MEK inhibitor U0126 blocked the neuroprotection of PREGS andPRE084. Neither LY294002 nor U0126 had influence on the number of pyramidal cellsin control mice. *P < 0.05 and **P < 0.01 vs. corresponding PREGS- or agonist-treatedAb25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e1050 1047

et al., 2004; Stepanichev et al., 2005, 2006). The Ab25e35-neurotoxicity in hippocampal CA1 pyramidal cells is knownthrough down-regulating ERK and Akt phosphorylation (Danielset al., 2001; Jin et al., 2005). Nicotinic receptor-mediated neuro-protection in the neurodegenerative disease models is related tothe activation of PI3KeAkt signaling pathway (Shimohama, 2009).Several acetylcholine esterase inhibitors, including donepezil andgalantamine, have been reported to attenuate the Ab25e35-toxicity(Arias et al., 2004; Svensson and Nordberg, 1998) through up-regulating a7nAChR-mediated PI3KeAkt signaling (Kihara et al.,2004). Our results in this study indicate that the a7nAChR agonistprevents the Ab25e35-downregulated PI3KeAkt signaling to reducethe caspase-3 activity. Furthermore, the activation of a7nAChR canprevent the Ab25e35-impaired a7nAChR (Chen et al., 2010).Consistent with the report by Kihara et al. (2004), we observed thatthe a7nAChR-mediated neuroprotection in Ab25e35-mice wasindependent of ERK signaling, whereas the s1R-mediated neuro-protection depended on the activation of ERK. Translocation ofactivated ERK into the nucleus (Cai et al., 2008) is necessary for thephosphorylation of cAMP response element binding protein (CREB)that elevates the expression of the pro-survival protein Bcl-2(Meller et al., 2005). In addition, the Ab-neurotoxicity has beenshown to induce disturbance of the endocytoplasmic reticulum

homeostasis and activation of stress-responsive genes, such as grp78 or grp 94 (Yu et al., 1999; Ghribi et al., 2004). The activation ofs1R inhibits NMDA-induced nitric oxide synthase activity to reducenitric oxide production (Kurata et al., 2004). Hayashi and Su (2005)reported that the activation of s1R can inhibit the recomposition ofintracellular compartments andmembrane composition to preventthe Ab25e35-triggered mitochondrial and ER stress.

We observed that the NMDAr blocker MK801 could prevent theAb25e35-neurotoxicity, suggesting that the activation of NMDAr isinvolved in Ab25e35-induced neurodegeneration. Notably, the s1R-dependent PREGS-neuroprotection in Ab25e35-micewas insensitiveto the administration of MK801. Monnet et al. (1995), using theNMDA-induced [3H]norepinephrine release from rat hippocampalslices, reported that PREGS could inhibit the NMDA-response. Theactivation of s1R exerts neuroprotective effects against theischemic injury through reducing ischemia-induced rise in [Ca2þ]i(Katnik et al., 2006). Kume et al. (2002) have provided supportingevidence that the s1R agonist reduces Ca2þ influx through NMDAr.On the other hand, PREGS at micromolar concentration has alsobeen reported to directly enhance NMDA-evoked current in rathippocampal neurons (Irwin et al., 1992; Wu et al., 1991). Thus,PREGS exerts two different effects on regulating NMDAr function:(i) a direct facilitation of NMDAr activation; (ii) an indirect s1R-mediated reduction of NMDA response. Although PREGS has beenreported to enhance NMDA-induced excitotoxicity in culturedhippocampal neurons (Weaver et al., 2000), our results indicatethat the activation of s1R by PREGS can inhibit NMDAr response toreduce the death of neuronal cells in Ab25e35-mice. Further studiesare needed to evaluate whether the administration of PREGS inAb25e35-mice can lead to a s1R-dependent decrease in the NMDA-induced Ca2þ influx in hippocampal CA1 pyramidal cells.

4.3. Anti-amnesic action of PREGS neuroprotection

The injection (i.c.v.) of “aggregated” Ab25e35 at dose of 3 or9 nmol/mouse impairs the short-term memory in spontaneousalternation behavior, the long-term memory in step-down typepassive avoidance and water-maze, although only injection of9 nmol/mouse Ab25e35 causes the death of hippocampal neuronalcells (Mamiya et al., 2004; Maurice et al., 1996; Stepanichev et al.,2006). Our results in the present study showed that the admin-istration of PREGS could exert a potent anti-amnesic effect inAb25e35 (9 nmol/mouse)-mice. Blockage of PREGS-neuroprotectionby the PI3K inhibitor LY294002 significantly attenuated its anti-amnesic effect in Ab25e35-mice, while the administration ofLY294002 did not affect the spatial cognitive performance incontrol mice. Meunier et al. (2006) have reported the anti-amnesicand neuroprotective effects of s1R agonist in Ab25e35 (9 nmol/mouse)-mice. The results give a clear indication that the neuro-protection of PREGS against Ab25e35-toxicity exerts a potent anti-amnesic effect. Maurice et al. (1998) reported that the Ab25e35(3 nmol/mouse)-induced deficits in alternation behavior andpassive avoidance could be improved by the administration ofPREGS. Wang et al. (2006) reported that the sequence 12e28 of Abwas critical for its selective and high-affinity binding to a7nAChR.We observed that the injection of Ab25e35 (3 nmol/mouse) leads tothe dysfunction of a7nAChR and the impairment of LTP induction(Chen et al., 2010). The treatment with the a7nAChR agonistDMXB can protect a7nAChR from Ab25e35-induced damage toimprove the spatial cognition and the induction of LTP. Similarly,the cholinesterase inhibitor tacrine and the nicotinic receptoragonist (�)-nicotine could attenuate the Ab25e35 (3 nmol/mouse)-induced deficits in alternation behavior, passive avoidance andplace water-maze (Maurice et al., 1996). In addition, PREGSameliorates the Ab25e35 (3 nmol/mouse)-induced deficits in

Procaspase-3

Cleaved caspase-3

-actinPREGS

p-ERK2

ERK2 PREGS

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

++

##

++

* p-A

kt/A

kt

control mice Aβ

25-35 mice

ML

A

NE

100

PREGS

p-Akt

AktPREGS

NE

100M

LA

PREGS0

2

4

6

8

10

+

**

##

casp

ase

-3/p

ro-c

aspa

se -3

control mice Aβ

25-35 mice

++

ML

A

NE

100

0.00

0.25

0.50

0.75

1.00

1.25

1.50

++

##

*

p-E

RK

2/E

RK

2

control mice Aβ

25-35 mice

PREGSN

E10

0ML

A

A

B

C

Fig. 5. PREGS prevents the Ab25e35-impaired Aktecaspase-3 and ERK signaling pathway in hippocampal CA1 region. (A & B) PREGS inhibited the decrease of p-Akt and theactivation of caspase-3 induced by Ab25e35-injection, which was sensitive to NE100 and MLA. (C) PREGS prevented Ab25e35-decreased p-ERK2, which could be blocked by NE100.The densitometric values for p-Akt, caspase-3 and p-ERK2 were first normalized by the protein amounts of Akt, pro-caspase-3 and ERK2, respectively, and then were normalizedagain by the basal values (in control mice) of p-Akt, caspase-3 and p-ERK2, respectively. The data are expressed as the means � SE. *P < 0.05, **P < 0.01 vs. control mice; ##P < 0.01vs. vehicle-treated Ab25e35-injected mice; þP < 0.05, þþP < 0.01 vs. PREGS-treated Ab25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e10501048

learning and memory, which is sensitive to the s1R antagonisthaloperidol or BMY-14802 (Maurice et al., 1998). The s1R agonistsare known to be potent modulators of acetylcholine release, bothin vitro and in vivo (Junien et al., 1991; Kato et al., 1999). Recently,

1 2 3 4 5

20

40

60

80

Time post-training (day)

Lat

ency

(sec

)

control micecontrol mice/LY294002Aβ25-35

miceAβ25-35

mice/PREGSAβ25-35mice/PREGS/LY294

Fig. 6. Effect of PI3K inhibitor LY294002 on anti-amnesic effect of PREGS in Ab25e35-mice. Llatency Ab25e35-mice treated with PRGES. The latency of Morris Water maze test was recordeafter training. **P < 0.01 vs. control mice, ##P < 0.01 vs. vehicle-treated Ab25e35-injected m

the s1R agonist is reported to attenuate cognitive deficits in theAD model of Ab25e35-injection with cholinergic loss (Antoniniet al., 2011). Therefore, PREGS targets a7nAChR and s1R to exertthe anti-amnesic effect in Ab25e35 (9 nmol/mouse)-mice through

002

0

10

20

30

40

50

60

++

##

**

control mice Aβ

25-35 mice

Aβ25-35

mice/PREGS

Lat

ency

(se

c)(o

n da

y 5

afte

r tr

aini

ng )

LY

2940

02

LY

2940

02

Y294002 had no effect on the latency in control mice, but it significantly prolonged thed from day 3 to day 7 after Ab25e35-injection. The histogram shows the latency on day 5ice, þþP < 0.01 vs. PREGS-treated Ab25e35-mice.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e1050 1049

not only protecting the neuronal cells but also improving thefunction of cholinergic system.

The reduction of PREGS has been found in AD patients’ brainscompared with that of age-matched non-demented controls(Weill-Engerer et al., 2002), indicating that the administration ofPREGS might exert a neuroprotective effect against the Ab-neuro-toxicity. However, the significance of PREGS-neuroprotection hasrecently been put into question because of the failure to detectsignificant levels of PREGS within the brain and plasma of rats andmice (Schumacher et al., 2008). The P450scc and hydroxysteroidsulfotransferase are highly expressed in hippocampal neurons(Kimoto et al., 2001). Although available direct analytical methodshave failed to detect levels of PREGS in brain tissue, it cannotexclude the local formation of biologically active PREGS withinspecific nervous system. In addition, numerous studies havereported that the neuromodulatory and neuropharmacologicaleffects of PREGS on the hippocampal synaptic transmission andsynaptic plasticity. The synthesis and release of PREGS have beenreported to be dependent on postsynaptic Ca2þ influx by NMDAr(Shibuya et al., 2003). Thus, one explanation is that the localconcentration of PREGS around the neurons, particularly insynaptic clefts in vivo, may be greater than that in brain tissues(Caldeira et al., 2004).

In summary, the a7nAChR agonists or the s1R agonists havebeen reported to attenuate the Ab-neurotoxicity and glutamatetoxicity in APP/PS1 mice, Ab1e42-rats and Ab25e35-mice, which areable to improve the Ab-induced impairment of learning andmemory. This study provides a new insight that PREGS targetinga7nAChR and s1R against Ab-neurotoxicity may exert a powerfulanti-amnesic effect.

Acknowledgments

This work was supported by grants for PCSIRT (IRT0631), NSFC(30872725; 81071027), DFMEC (200803120006) and (BK2011029)to Ling Chen, and NSFC (30900577) to Lei Chen.

We declare that there is no competing financial that could beconstrued as influencing the results or interpretation of thereported study.

References

Antonini, V., Marrazzo, A., Kleiner, G., Coradazzi, M., Ronsisvalle, S., Prezzavento, O.,Ronsisvalle, G., Leanza, G., 2011. Anti-amnesic and neuroprotective actions ofthe sigma-1 receptor agonist (-)-MR22 in rats with selective cholinergic lesionand amyloid infusion. J. Alzheimers Dis. 24, 569e586.

Arias, E., Ales, E., Gabilan, N.H., Cano-Abad, M.F., Villarroya, M., Garcia, A.G.,López, M.G., 2004. Galantamine prevents apoptosis induced by b-amyloid andthapsigargin: involvement of nicotinic acetylcholine receptors. Neuropharma-cology 46, 103e114.

Akk, G., Bracamontes, J., Steinbach, J.H., 2001. Pregnenolone sulfate block ofGABA(A) receptors: mechanism and involvement of a residue in the M2 regionof the alpha subunit. J. Physiol. 532, 673e684.

Baulieu, E.E., 1997. Neurosteroids: of the nervous system, by the nervous system, forthe nervous system. Rec. Prog. Horm. Res. 52, 1e32.

Burdick, D., Soreghan, B., Kwon, M., Kosmoski, J., Knauer, M., Henschen, A., Yates, J.,Cotman, C., Glabe, C., 1992. Assembly and aggregation properties of syntheticAlzheimer’s A4/beta amyloid peptide analogs. J. Biol. Chem. 26, 546e554.

Cai, W., Zhu, Y., Furuya, K., Li, Z., Sokabe, M., Chen, L., 2008. Two different molecularmechanisms underlying progesterone neuroprotection against ischemic braindamage. Neuropharmacology 55, 127e138.

Caldeira, J.C., Wu, Y., Mameli, M., Purdy, R.H., Li, P.K., Akwa, Y., Savage, D.D.,Engen, J.R., Valenzuela, C.F., 2004. Fetal alcohol exposure alters neurosteroidlevels in the developing rat brain. J. Neurochem. 90, 1530e1539.

Chen, L., Sokabe, M., 2005. Presynaptic modulation of synaptic transmission bypregnenolone sulfate as studied by optical recordings. J. Neurophysiol. 94,4131e4144.

Chen, L., Wang, H., Zhang, Z., Li, Z., He, D., Sokabe, M., Chen, L., 2010. DMXB (GTS-21)ameliorates the cognitive deficits in beta amyloid25e35-injected mice throughpreventing the dysfunction of alpha7 nicotinic receptor. J. Neurosci. Res. 8,1784e1794.

Chen, L., Yamada, K., Nabeshima, T., Sokabe, M., 2006. Alpha7 Nicotinic acetylcho-line receptor as a target to rescue deficit in hippocampal LTP induction in beta-amyloid infused rats. Neuropharmacology 50, 254e268.

Daniels, W.M., Hendricks, J., Salie, R., Taljaard, J.J., 2001. The role of the MAP-kinasesuperfamily in beta-amyloid toxicity. Metab. Brain Dis. 16, 175e185.

Forloni, G., Chiesa, R., Smiroldo, S., Verga, L., Salmona, M., Tagliavini, F., Angeretti, N.,1993. Apoptosis mediated neurotoxicity induced by chronic application ofb amyloid fragment 25e35. NeuroReport 4, 523e526.

Ghribi, O., Herman, M.M., Pramoonjago, P., Spaulding, N.K., Savory, J., 2004. GDNFregulates the A beta-induced endoplasmic reticulum stress response in rabbithippocampus by inhibiting the activation of gadd 153 and the JNK and ERKkinases. Neurobiol. Dis. 6, 417e427.

Hayashi, T., Su, T.P., 2005. The potential role of sigma1 receptors in lipid transportand lipid raft reconstitution in the brain: implication for drug abuse. Life Sci. 77,1612e1624.

Hellström-Lindahl, E., Court, J., Keverne, J., Svedberg, M., Lee, M., Marutle, A.,Thomas, A., Perry, E., Bednar, I., Nordberg, A., 2004. Nicotine reduces A beta inthe brain and cerebral vessels of APPsw mice. Eur. J. Neurosci. 19, 2703e2710.

Higashi, T., Sugitani, H., Yagi, T., Shimada, K., 2003. Studies on neurosteroids XVI.Levels of pregnenolone sulfate in rat brains determined by enzyme-linkedimmunosorbent assay not requiring solvolysis. Biol. Pharm. Bull. 26, 709e711.

Irwin, R.P., Maragakis, N.J., Rogawski, M.A., Purdy, R.H., Farb, D.H., Paul, S.M., 1992.Pregnenolone sulfate augments NMDA receptor mediated increases in intra-cellular Ca2þ in cultured rat hippocampal neurons. Neurosci. Lett. 141, 30e34.

Jin, Y., Yan, E.Z., Fan, Y., Zong, Z.H., Qi, Z.M., Li, Z., 2005. Sodium ferulate preventsamyloid-beta-induced neurotoxicity through suppression of p38 MAPK andupregulation of ERK-1/2 and Akt/protein kinase B in rat hippocampus. ActaPharmacol. Sin 26, 943e951.

Junien, J.L., Roman, F.J., Brunelle, G., Pascaud, X., 1991. JO1784, a novel sigma ligand,potentiates [3H]acetylcholine release from rat hippocampal slices. Eur. J.Pharmacol. 200, 343e345.

Katnik, C., Guerrero, W.R., Pennypacker, K.R., Herrera, Y., Cuevas, J., 2006. Sigma-1receptor activation prevents intracellular calcium dysregulation in corticalneurons during in vitro ischemia. J. Pharmacol. Exp. Ther. 319, 1355e1365.

Kato, K., Hayako, H., Ishihara, Y., Marui, S., Iwane, M., Miyamoto, M., 1999. TAK-147,an acetylcholinesterase inhibitor, increases choline acetyltransferase activity incultured rat septal cholinergic neurons. Neurosci. Lett. 260, 5e8.

Kihara, T., Sawada, H., Nakamizo, T., Kanki, R., Yamashita, H., Maelicke, A.,Shimohama, S., 2004. Galantamine modulates nicotinic receptor and blocks Ab-enhanced glutamate toxicity. Biochem. Biophys. Res. Commun. 325, 976e982.

Kihara, T., Shimohama, S., Sawada, H., Kimura, J., Kume, T., Kochiyama, H., Maeda, T.,Akaike, A., 1997. Nicotinic receptor stimulation protects neurons against b-amyloid toxicity. Ann. Neurol. 42, 159e163.

Kimoto, T., Tsurugizawa, T., Ohta, Y., Makino, J., Ho, Tamura, Hojo, Y., Takata, N.,Kawato, S., 2001. Neurosteroid synthesis by cytochrome p450-containingsystems localized in the rat brain hippocampal neurons: N-methyl-D-aspar-tate and calcium-dependent synthesis. Endocrinology 142, 3578e3589.

Kowall, N.W., Beal, M.F., Busciglio, J., Duffy, L.K., Yanker, B.A., 1991. An in vivo modelfor the neurodegenerative effects of beta amyloid and protection by substanceP. Proc. Natl. Acad. Sci. U. S. A. 88, 7247e7251.

Kowall, N.W., McKee, A.C., Yankner, B.A., Beal, M.F., 1992. In vivo neurotoxicity ofbeta-amyloid [beta(1e40)] and the beta(25e35) fragment. Neurobiol. Aging 13,537e542.

Kubo, T., Nishimura, S., Kumagae, Y., Kaneko, I., 2002. In vivo conversion of race-mized beta-amyloid ([D-Ser 26]A beta 1e40) to truncated and toxic fragments([D-Ser 26]A beta 25e35/40) and fragment presence in the brains of Alz-heimer’s patients. J. Neurosci. Res. 70, 474e483.

Kume, T., Nishikawa, H., Taguchi, R., Hashino, A., Katsuki, H., Kaneko, S., Minami, M.,Satoh, M., Akaike, A., 2002. Antagonism of NMDA receptors by sigma receptorligands attenuates chemical ischemia-induced neuronal death in vitro. Eur. J.Pharmacol. 455, 91e100.

Kurata, K., Takebayashi, M., Morinobu, S., Yamawaki, S., 2004. Beta-estradiol,dehydroepiandrosterone, and dehydroepiandrosterone sulfate protect againstN-methyl-D-aspartate-induced neurotoxicity in rat hippocampal neurons bydifferent mechanisms. J. Pharmacol. Exp. Ther. 311, 237e245.

Li, L., Xu, B., Zhu, Y., Chen, L., Sokabe, M., Chen, L., 2010. DHEA preventsAbeta(25e35)-impaired survival of newborn neurons in the dentate gyrusthrough a modulation of PI3KeAkt-mTOR signaling. Neuropharmacology 59,323e333.

Liu, Q., Zhang, J., Zhu, H., Qin, C., Chen, Q., Zhao, B., 2007. Dissecting the signalingpathway of nicotine-mediated neuroprotection in a mouse Alzheimer diseasemodel. FASEB J. 21, 61e73.

Mamiya, T., Asanuma, T., Kise, M., Ito, Y., Mizukuchi, A., Aoto, H., Ukai, M., 2004.Effects of pre-germinated brown rice on beamyloid protein induced learningand memory deficits in mice. Biol. Pharm. Bull. 27, 1041e1045.

Marrazzo, A., Caraci, F., Salinaro, E.T., Su, T.P., Copani, A., Ronsisvalle, G., 2005.Neuroprotective effects of sigma-1 receptor agonists against beta-amyloid-induced toxicity. NeuroReport 16, 1223e1226.

Maurice, T., Grégoire, C., Espallergues, J., 2006. Neuro(active)steroids actions at theneuromodulatory sigma1 (sigma1) receptor: biochemical and physiologicalevidences, consequences in neuroprotection. Pharmacol. Biochem. Behav. 84,581e597.

Maurice, T., Su, T.P., Privat, A., 1998. Sigma1 (s1) receptor agonists and neuro-steroids attenuate b25e35-amyloid peptide-induced amnesia in mice througha common mechanism. Neuroscience 83, 413e428.

R. Yang et al. / Neuropharmacology 63 (2012) 1042e10501050

Maurice, T., Lockhart, B.P., 1997. Neuroprotective and anti-amnesic potentials ofsigma (sigma) receptor ligands. Prog. Neuropsychopharmacol. Biol. Psychiatry21, 69e102.

Maurice, T., Lockhart, B.P., Privat, A., 1996. Amnesia induced in mice by centrallyadministered b-amyloid peptides involves cholinergic dysfunction. Brain Res.706, 181e193.

Meunier, J., Ieni, J., Maurice, T., 2006. The anti-amnesic and neuroprotective effectsof donepezil against amyloid b25e35 peptide-induced toxicity in mice involvean interaction with the s1 receptor. Br. J. Pharmacol. 149, 998e1012.

Meyer, D.A., Carta, M., Partridge, L.D., Covey, D.F., Valenzuela, C.F., 2002. Neuro-steroids enhance spontaneous glutamate release in hippocampal neurons.Possible role of metabotropic sigma1-like receptors. J. Biol. Chem. 277,28725e28732.

Meller, R., Minami, M., Cameron, J.A., Impey, S., Chen, D., Lan, J.Q., Henshall, D.C.,Simon, R.P., 2005. CREB-mediated Bcl-2 protein expression after ischemicpreconditioning. J. Cereb. Blood Flow Metab. 25, 234e246.

Monnet, F.P., Mahe, V., Robel, P., Baulieu, E.E., 1995. Neurosteroids, viasigma receptors modulate the [3H]norepinephrine release evoked by N-methyl-D-aspartate in the rat hippocampus. Proc. Natl. Acad. Sci. U. S. A. 92,3774e3778.

Nakazawa, M., Matsuno, K., Mita, S., 1998. Activation of sigma1 receptor subtypeleads to neuroprotection in the rat primary neuronal cultures. Neurochem. Int.32, 337e343.

Naylor, J.C., Kilts, J.D., Hulette, C.M., Steffens, D.C., Blazer, D.G., Ervin, J.F., Strauss, J.L.,Allen, T.B., Massing, M.W., Payne, V.M., Youssef, N.A., Shampine, L.J., Marx, C.E.,2010. Allopregnanolone levels are reduced in temporal cortex in patients withAlzheimer’s disease compared to cognitively intact control subjects. Biochim.Biophys. Acta 1801, 951e959.

Nelson, O., Tu, H., Lei, T., Bentahir, M., de Strooper, B., Bezprozvanny, I., 2007.Familial Alzheimer disease-linked mutations specifically disrupt Ca2þ leakfunction of presenilin 1. J. Clin. Invest. 117, 1230e1239.

Pike, C.J., Walencewicz-Wasserman, A.J., Kosmoski, J., Cribbs, D.H., Glabe, C.G.,Cotman, C.W., 1995. Structureeactivity analyses of beta-amyloid peptides:contributions of the beta 25e35 region to aggregation and neurotoxicity.J. Neurochem. 64, 253e265.

Rothman, S.M., Olney, J., 1995. Excitotoxicity and the NMDA receptorestill lethalafter eight years. Trends Neurosci. 18, 57e58.

Schumacher, M., Liere, P., Akwa, Y., Rajkowski, K., Griffiths, W., Bodin, K., Sjövall, J.,Baulieu, E.E., 2008. Pregnenolone sulfate in the brain: a controversial neuro-steroid. Neurochem. Int. 5, 522e540.

Shibuya, K., Takata, N., Hojo, Y., Furukawa, A., Yasumatsu, N., Kimoto, T., Enami, T.,Suzuki, K., Tanabe, N., Ishii, H., Mukai, H., Takahashi, T., Hattori, T.A., Kawato, S.,2003. Hippocampal cytochrome P450s synthesize brain neurosteroids whichare paracrine neuromodulators of synaptic signal transduction. Biochim. Bio-phys. Acta 1619, 301e316.

Shimohama, S., 2009. Nicotinic receptor-mediated neuroprotection in neurode-generative disease models. Biol. Pharm. Bull. 32, 332e336.

Stepanichev, M.Y., Moiseeva, Y.V., Lazareva, N.A., Gulyaeva, N.V., 2005. Studies of theeffects of fragment (25e35) of b-amyloid peptide on the behavior of rats ina radial maze. Neurosci. Behav. Physiol. 35, 511e518.

Stepanichev, M.Y., Zdobnova, I.M., Zarubenko, I.I., Lazareva, N.A., Gulyaeva, N.V.,2006. Studies of the effects of central administration of b-amyloid peptide

(25e35): pathomorphological changes in the hippocampus and impairment ofspatial memory. Neurosci. Behav. Physiol. 36, 101e106.

Svensson, A.L., Nordberg, A., 1998. Tacrine and donepezil attenuate the neurotoxiceffect of Ab25e35 in rat PC12 cells. NeuroReport 9, 1519e1522.

Takebayashi, M., Hayashi, T., Su, T.P., 2004. Sigma-1 receptors potentiate epidermalgrowth factor signaling towards neuritogenesis in PC12 cells: potential relationto lipid raft reconstitution. Synapse 53, 90e103.

Vallée, M., Mayo, W., Darnaudéry, M., Corpéchot, C., Young, J., Koehl, M., Le Moal, M.,Baulieu, E.E., Robel, P., Simon, H., 1997. Neurosteroids: deficient cognitiveperformance in aged rats depends on low pregnenolone sulfate levels in thehippocampus. Proc. Natl. Acad. Sci. U. S. A. 94, 14865e14870.

Wang, D., Noda, Y., Zhou, Y., Mouri, A., Mizoguchi, H., Nitta, A., Chen, W.,Nabeshima, T., 2007. The allosteric potentiation of nicotinic acetylcholinereceptors by galantamine ameliorates the cognitive dysfunction in betaamyloid25-35 i.c.v.-injected mice: involvement of dopaminergic systems.Neuropsychopharmacology 32, 1261e1271.

Wang, R.M., Yang, F., Zhang, Y.X., 2006. Preconditioning-induced activation of ERK5is dependent on moderate Ca2þ influx via NMDA receptors and contributes toischemic tolerance in the hippocampal CA1 region of rats. Life Sci. 79,1839e1846.

Weaver, C.E., Land, M.B., Purdy, R.H., Richards, K.G., Gibbs, T.T., Farb, D.H., 2000.Geometry and charge determine pharmacological effects of steroids on N-methyl-D-aspartate receptor-induced Ca(2þ) accumulation and cell death.J. Pharmacol. Exp. Ther. 293, 747e754.

Weill-Engerer, S., David, J.P., Sazdovitch, V., Liere, P., Eychenne, B., Pianos, A.,Schumacher, M., Delacourte, A., Baulieu, E.E., Akwa, Y., 2002. Neurosteroidquantification in human brain regions: comparison between Alzheimer’s andnondemented patients. J. Clin. Endocrinol. Metab. 87, 5138e5143.

Wu, F.S., Gibbs, T., Farb, D.H., 1991. Pregnenolone sulfate: a positive allostericmodulator at the NMDA receptor. Mol. Pharmacol. 40, 333e336.

Wu,M.N., He, Y.X., Guo, F., Qi, J.S., 2008. Alpha4beta2nicotinic acetylcholine receptorsare required for the amyloid beta protein-induced suppression of long-termpotentiation in rat hippocampal CA1 region in vivo. Brain Res. Bull. 77, 84e90.

Xu, B., Yang, R., Chang, F., Chen, L., Xie, G., Sokabe, M., Chen, L., 2012. NeurosteroidPREGS protects neurite growth and survival of newborn neurons in thehippocampal dentate gyrus of APPswe/PS1dE9 mice. Curr. Alzheimer Res. 9,361e372.

Yagasaki, Y., Numakawa, T., Kumamaru, E., Hayashi, T., Su, T.P., Kunugi, H., 2006.Chronic antidepressants potentiate via sigma-1 receptors the brain-derivedneurotrophic factor-induced signaling for glutamate release. J. Biol. Chem.281, 12941e12949.

Yan, S.D., Chen, X., Fu, J., Chen, M., Zhu, H., Roher, A., Slattery, T., Zhao, L.,Nagashima, M., Morser, J., Migheli, A., Nawroth, P., Stern, D., Schmidt, A.M., 1996.RAGE and amyloid-beta peptide neurotoxicity in Alzheimer’s disease. Nature382, 685e691.

Yu, Z., Luo, H., Fu, W., Mattson, M.P., 1999. The endoplasmic reticulum stress-responsive protein GRP78 protects neurons against excitotoxicity andapoptosis: suppression of oxidative stress and stabilization of calciumhomeostasis. Exp. Neurol. 155, 302e314.

Zamani, M.R., Allen, Y.S., Owen, G.P., Gray, J.A., 1997. Nicotine modulates theneurotoxic effect of b-amyloid protein (25e35) in hippocampal cultures. Neu-roReport 8, 513e517.