effects of steroid hormones on (na+, k+)-atpase activity inhibition-induced amnesia on the...

Pharmacological Research 49 (2004) 151–159

Effects of steroid hormones on (Na+, K+)-ATPase activityinhibition-induced amnesia on the step-through passive

avoidance task in gonadectomized miceTomoaki Satoa,∗, Koh-ichi Tanakaa, Yoshiko Ohnishia, Toyonori Teramotoa,

Masahiro Irifuneb, Takashige Nishikawaaa Department of Applied Pharmacology, Kagoshima University Graduate School of Medical and Dental Sciences,

Sakuragaoka, Kagoshima 890-8544, Japanb Department of Anesthesiology, Hiroshima University Dental Hospital, Kasumi, Minami-ku, Hiroshima 734-8533, Japan

Accepted 1 September 2003

Abstract

Inhibition of sodium-potassium adenosine 5′-triphosphatase ((Na+, K+)-ATPase) activity causes edema and cell death in the cen-tral nervous system, and impairment of learning and memory. Several sex steroid hormones have a protective effect against neuronalcell damage and the hypofunction of learning and memory. To examine the possible roles and mechanism of action of steroid hor-mones against amnesia induced by ouabain, an inhibitor of (Na+, K+)-ATPase, gonadectomized male mice were administrated ouabain(0.1�g per mouse) intracisternally (i.cist.), and the learning and memory abilities of the mice were assessed by a step-through passiveavoidance task. Subcutaneous (s.c.) administration of 17�-estradiol (�E2; 10�g kg−1) or testosterone (TES; 1 mg kg−1) improved thememory impairment induced by ouabain, while administration of dihydrotestosterone (1 mg kg−1) or corticosterone (COR) (1 mg kg−1)did not. Treatment with the estradiol receptor antagonists, tamoxifen (TAM) (10 mg kg−1; s.c. or 0.1�g; i.cist.) and 4-hydroxytamoxifen(10 mg kg−1; s.c.), or the androgen receptor antagonist, cyproterone (10 mg kg−1; s.c. or 1�g; i.cist.), did not influence the protectiveeffect of �E2 or TES on ouabain-induced amnesia. Moreover, we studied the effects of several free radical scavengers—17�-estradiol(10�g kg−1; s.c.),�-tocopherol (VE: 200 mg kg−1; per os (p.o.), ascorbic acid (VC: 200 mg kg−1; p.o.), or VE+ VC (200 mg kg−1 each;p.o.)—on ouabain-induced amnesia, and compared those effects with that of�E2. The administration of free radical scavengers had nosignificant effect on memory impairment. These results indicate that�E2 and TES ameliorate the amnesia induced by inhibition of (Na+,K+)-ATPase activity, and that the protective effect of�E2 is caused by a non-genomic, rather than a genomic effect or a radical scavengingaction. Additionally, the ameliorative effect of TES does not appear to involve free radical scavenging, but its aromatization to estrogencould contribute to the non-genomic action of�E2.© 2003 Elsevier Ltd. All rights reserved.

Keywords:Ouabain; Estradiol; Testosterone; Learning and memory; Non-genomic effect

1. Introduction

Several lines of evidence suggest that once ischaemiaand/or postischaemic reperfusion occurs, the ensuing de-ficiency of ATP [1,2], production of reactive oxygenspecies[3,4], or release of excitatory amino acids (EAA)[5,6] causes impairment of the central nervous system(CNS). Such toxicity, which resembles that caused by is-chaemic/postischaemic disease, can be reproduced by theadministration of ouabain, a (Na+, K+)-ATPase inhibitor

∗ Corresponding author. Tel.:+81-99-275-6162; fax:+81-99-275-6168.E-mail address:tomsato@dentb.hal.kagoshima-u.ac.jp (T. Sato).

[7]. (Na+, K+)-ATPase, the so-called sodium pump, reg-ulates intracellular ion homeostasis[5] and the release ofEAA neurotransmitters[1,5]. In a previous study, we sug-gested that deficiency of ATP or the production of reactiveoxygen species inhibits the activity of (Na+, K+)-ATPase[8,9], and it was reported that the inhibition of this enzymecauses the release of EAA and functional impairment ofbrain tissues[5]. In this way, (Na+, K+)-ATPase appears tobe critical for neuronal cell function and survival. Indeed,Lees and co-workers[5,10] showed that several inhibitorsof (Na+, K+)-ATPase caused spongiform degeneration ofthe brain and neuronal cell death, induced by EAA, in rathippocampus. Moreover, Gibbs and co-workers[11,12] in-

152 T. Sato et al. / Pharmacological Research 49 (2004) 151–159

dicated that intracisternal (i.cist.) administration of ouabain,a specific (Na+, K+)-ATPase inhibitor, caused impairmentof learning and memory in a passive avoidance task. Takentogether, these studies suggest that a relationship existsbetween the impairment of learning and memory and theinhibition of (Na+, K+)-ATPase.

Several lines of experimental evidence suggest that17�-estradiol (�E2) protects neurons in vitro against celldeath from oxidative stress[13], and prevents the impair-ment of learning and memory[14]. In fact, oxidative stressand lipid peroxidation are thought to play a significant rolein cerebral ischemia[1,3,4] and the resulting hypofunctionof learning and memory[1,13]. Some steroids can act asfree radical scavengers, and the neuroprotective effects ofsteroids have often been attributed to their antioxidant prop-erties. In addition, there is evidence suggesting that�E2directly modulatesN-methyl-d-aspartate (NMDA)[15] re-ceptors. The aforementioned functions of steroid hormonesare referred to as non-genomic actions[15]. Some effects ofsteroid hormones, such as the protection against neuronalcell death and vascular dementia, are thought to involvethese non-genomic functions. We investigated the protectiveeffects of several steroid hormones against the impairmentof learning and memory induced by inhibition of (Na+,K+)-ATPase activity, and attempted to determine a possiblemechanism for the protective effects of steroid hormones.We also examined the effects of several steroid receptor an-tagonists on (Na+, K+)-ATPase inhibitor-induced amnesiaand compared those with the effects of the radical scav-engers. In this study, the steroid hormones that were used tomodulate the impairment of learning and memory inducedby ouabain were 17�-estradiol, 17�-estradiol, testosterone,dihydrotestosterone, and corticosterone.

2. Materials and methods

2.1. Animals

Male ddY mice (Kyudou, Ltd., Kumamoto, Japan) weigh-ing 25–35 g were used. The animals were housed with freeaccess to standard food (Clea Japan, Inc.) and water in anair-conditioned room at a temperature of 24±1◦C, humidityof 50±10%, and with a constant 12-h light/dark cycle (lighton between 7:00 and 19:00 h). All behavioral experimentswere carried out between 9:00 and 17:00 h. All procedureswere approved by the Committee of Animal Experimenta-tion of the Dental School of Kagoshima University.

2.2. Drugs

The following drugs were used throughout the experi-ments: ouabain, cyproterone (CYP) and 4-hydroxytamoxifen(HYT), purchased from Sigma Chemical Co (St. Louis,MO, USA); �E2, 17�-estradiol (�E2), COR, TAM, VE,and VC, from Naclai Tesque, Inc. (Kyoto, Japan); and TES,

5�-androstan-17�-ol-3-one (dihydrotestosterone, DHT),from Fluka Chemical AG (Buchs, Switzerland).

Ouabain was dissolved in saline, and injected i.cist. at adose of 0.01–1�g per mouse per day (injections were givenin a volume of 10�l, consisting of 1�l ouabain/saline so-lution diluted the vehicle described below) for 3 days be-fore the training trial (Train.), by the method of Ueda et al.[16]. When ouabain and other steroid antagonists were ad-ministrated simultaneously, the drugs were co-injected with10�l of vehicle (1�l of ouabain in saline plus the vehicledescribed below). All steroids (�E2, �E2, TES, DHT, andCOR) and steroid antagonists (TAM, HYT, and CYP) weredissolved in dimethyl sulfoxide (DMSO; Nacalai Tesque),and suspended in propylene glycol (Nacalai Tesque). Thefinal concentration of DMSO was 0.1% (s.c.) or 0.01%(i.cist.), and at this concentration, DMSO did not influ-ence the response latency. Intracisternal injections were per-formed once daily at a volume of 10�l in vehicle consistingof 0.01% DMSO and about 10% propylene glycol in saline;the composition of the vehicle for s.c. injections consisted of0.1% DMSO and roughly 10% propylene glycol in saline.VE was suspended in propylene glycol, and then further di-luted with saline. VC was dissolved in saline and suspendedin propylene glycol. In short, the vehicles used for peroraladministration consisted of 90% propylene glycol in saline.

Mice were injected with steroid hormones s.c. at a dose of10�g kg−1 per day (�E2 and�E2) or 100–1000�g kg−1 perday (TES, DHT or COR), at 120, 96, 72, 48, or 24 h beforethe Train. of the step-through task. All steroid antagonists(TAM, HYT, and CYP) were injected s.c. or i.cist. at dosesof 10 mg kg−1 (s.c.) or 0.01–0.5�g (i.cist.) at the same timeas the steroid injections. VE and VC were administratedorally (p.o.) at a dose of 200 mg kg−1 per day, at 120, 96,72, 48, 24 h before the Train. of the step-through passiveavoidance task.

2.3. Gonadectomy

Male mice were gonadectomized (GOX) or sham-operated8 days before the acquisition phase of the step-through test.The animals were anesthetized with sodium pentobarbital(35 mg kg−1, i.p.), and the skin of the scrotum was washedwith 0.5% povidone iodine. The testes were removed withthe epididymis, and the skin of the scrotum was closed withsilk sutures. The same procedure was performed on thesham-operated mice, except the testes and epididymis werenot removed.

2.4. Behavioral tests

In the present study, a two-compartment step-through typepassive avoidance apparatus was used to measure learningand memory performance of mice. The box was divided intobright (9 cm× 9 cm× 36 cm) and dark (26 cm× 26 cm×36 cm) compartments by a guillotine door. The bright com-partment was illuminated by a white glow light. In the Train.,

T. Sato et al. / Pharmacological Research 49 (2004) 151–159 153

a mouse was placed in the illuminated compartment. Aftera habituation period in the illuminated compartment (30 s),the guillotine door was raised to allow the mouse to enterthe dark chamber. When the mouse’s hind legs entered thedark chamber, the guillotine door was closed and an elec-trical foot shock (0.25 mA, 20 V) was delivered through thegrid floor for 2 s. The time that elapsed prior to entry into thedark compartment (latency) was recorded. At 24 h after theTrain., the mice were submitted to a step-through-latencytrial (STL). In the STL, the mouse was placed in the brightcompartment and the latency time for entry into the darkcompartment was recorded for up to 300 s.

2.5. Locomotor activity

Locomotor activity was measured using four circularactivity cages that were equipped with photocell sensorunits mounted on the outer wall. Locomotor activity studieswere conducted in a separate group of mice that receivedstep-through training and the same injection protocol asdescribed above. The mice were placed one at a time in theactivity cages before step-through training and activity wasrecorded for 20 min.

2.6. Statistical analysis

Statistical analysis was performed using one-wayanalysis of variance (ANOVA) with post hoc tests(Bonferroni’s/Dunn’s post hoc test). The results in the textand figures are expressed as the means± S.E.M.

3. Results

3.1. Effect of GOX on learning and memory in thestep-through passive avoidance task

In the first experiment, we examined the effect of GOXon learning and memory in the step-through passive avoid-ance task (Fig. 1). We confirmed that the latency times forentry into the dark compartment on the Train. and STL trialsdid not differ significantly between GOX and sham-operatedmice at 8 days post-surgery, and consequently, we usedGOX-mice in all subsequent experiments.

3.2. Effect of intracisternal ouabain administrationon learning and memory in the step-through passiveavoidance task in GOX-mice

The effects of varying concentrations of ouabain (0.1�gper mouse) on learning and memory in the passive avoidancetask are shown inFig. 2. There was a significant differenceamong the four ouabain-treated groups (F(3, 31) = 8.40;P < 0.01) in the Train. Post hoc test showed that ouabainat a dose of 0.1�g significantly reduced the latency time onthe STL (P < 0.01). At a dose of 0.01�g, ouabain had a

(contr

STLTrain.

Fig. 1. The latency time of gonadectomized (GOX) or sham-gona-dectomized (Sham) male mice on the Train. and STL in the passiveavoidance task. Male mice were gonadectomized or sham-operated 8 daysbefore the Train. The longitudinal axis represents the latency time (s) toenter the dark compartment. Each column represents the mean± S.E.M.

of 10 mice. Train.: training trial, STL: step-through latency trial.

STLTrain.

Vehicle 0.01 µg 0.1 µg 1 µg

Ouabain

Fig. 2. Effect of intracisternally administered ouabain, a specific (Na+,K+)-ATPase inhibitor on step-through passive avoidance task performancein GOX-mice. GOX-mice were treated with ouabain or vehicle 3 daysbefore the Train. Values are mean± S.E.M. for 10 animals per group.∗∗P < 0.01 vs. the latency time on STL in the vehicle group (by posthoc test).

tendency to reduce the latency time on the retention trial,but the effect was not statistically significant (P = 0.10).Ouabain at 1�g did not have a significant effect on theSTL; however, the latency time on the Train. tended to belonger for the 1�g ouabain group as compared with theother groups (F(3, 31) = 2.00; P = 0.13). Overall, themost pronounced amnesic effect of ouabain was observedat a dose of 0.1�g, so this dose was used in all subsequentexperiments.

3.3. Effects of steroid hormones on ouabain-inducedamnesic and ouabain-untreated mice in the step-throughpassive avoidance test

The effects of steroid hormones on ouabain-induced am-nesia (Fig. 3A), differed significantly among the treatedgroups (F(6, 59) = 2.67; P < 0.05). Post hoc test showedthat pretreatment with�E2 (10�g kg−1) significantly in-hibited (P < 0.01) the ouabain-induced amnesic effect onlatency time for the STL. TES, a testicular steroid hor-mone, did not have a significant effect on ouabain-inducedamnesia at a dose of 0.1 mg kg−1, but showed a signif-icant protective effect at 1 mg kg−1 (P < 0.05). DHT, anon-aromatizable androgen, at doses of 0.1 and 1 mg kg−1

did not significantly inhibit ouabain-induced impairmentof learning and memory. The administration of COR (0.1or 1 mg kg−1), an endogenous adrenocortical steroid hor-mone in rats and mice, did not have a significant effect onouabain-induced amnesia. Additionally, the administrationof �E2 (10�g kg−1) and TES (1 mg kg−1) did not have a

β (10

(A) STLTrain.

GOX and Ouabain 0.1 µg

(B) STLTrain.

without ouabain 0.1 µg

Fig. 3. Effects of steroid hormones on ouabain-induced amnesia (A) and on untreated Sham and GOX-mice in the step-through passive avoidance task.Steroid pretreatment was given 5 days before the Train. The mice were gonadectomized and injected intracisternally with 0.1 �g ouabain (A), or treatedwith same volume of vehicle without ouabain (B). Values are mean ± S.E.M. for 5–10 animals per group. ∗P < 0.05; ∗∗P < 0.01 vs. STL in the vehiclegroup (by post hoc test). �E2 = 17�-estradiol, TES: testosterone, DHT: dihydrotestosterone, COR: corticosterone.

significant effect in sham-operated or GOX-mice that werenot treated with ouabain (Fig. 3B).

3.4. Effects of different routes of administration andconcentrations of TAM on step-through passive avoidancein GOX-mice

TAM is a non-steroidal estrogen antagonist that binds tocytosolic estrogen receptors and has low toxicity [17]. TAMhas also been reported to inhibit protein kinase C [18] and tohave antioxidant properties [19]. TAM has a limited abilityto cross the blood–brain barrier and is therefore unlikely tocause central nervous system disturbances [20]. However, ithas been reported that s.c. administration of TAM at dosesof more than 7 mg kg−1 causes alterations in cytosolic ornuclear receptors of the hypothalamus and pituitary gland[17]. Therefore, in the next set of experiments, we examinedthe effects of TAM at varying doses and by several differ-ent injection methods (Fig. 4). The administration of TAMs.c. at a dose of 10 mg kg−1 did not have a significant effecton latency time in the step-through passive avoidance task.However, there was a significant difference among the i.cist.groups (F(4, 38) = 2.65; P < 0.05). Intracisternal adminis-tration of TAM at a dose of 1 �g showed a tendency to lowerlatency time on the STL (P = 0.06), and at higher dosesTAM (5 �g) clearly decreased the latency (P < 0.01). Thiseffect of TAM at the over 1 �g dose may have been causedby inhibition of PKC, and so accordingly, TAM was admin-istered by i.cist. injection at a dose 0.1 �g in all subsequentexperiments.

STLTrain.

Fig. 4. Effects of different injection methods and four concentrations ofTAM on step-through passive avoidance task performance in GOX-mice.All mice were gonadectomized, and given either vehicle or TAM. TAMwas injected either subcutaneously (s.c.; 10 mg kg−1) or intracisternally(i.cist.; 0.01–5 �g). Values are mean ± S.E.M. for 8–10 animals pergroup. ∗∗P < 0.05 vs. STL in the vehicle (i.cist.) group (by post hoctest). TAM: tamoxifen.

3.5. Effect of TAM and HYT on the reversal ofouabain-induced learning and memory impairment byβE2

We studied the influence of TAM (i.cist.) and HYT (s.c.)on the recovery from ouabain-induced memory impairmentby �E2 (Fig. 5). Although ANOVA revealed a significantdifference between groups (F(5, 52) = 2.44; P < 0.05),the following post hoc test showed that the treatment withTAM at 0.1 �g (i.cist.) did not affect the �E2-induced im-provement of latency time on the STL. In addition, HYTis a metabolite of tamoxifen, and is a more potent estrogenantagonist than TAM [21]. However, this more potent an-tagonist at a dose of 10 mg kg−1 (s.c.) did not inhibit therecovery from ouabain-induced amnesia caused by �E2.

3.6. Influence of cyproterone treatment on the protectiveeffect of TES against ouabain-induced impairmentof learning and memory in the step-through passiveavoidance task

As shown in Fig. 3, TES at a dose of 1 mg kg−1 pro-tected against ouabain-induced amnesia in the step-throughpassive avoidance task. Since the hypothalamus is capa-ble of aromatizing a small proportion of testosterone to

Fig. 5. Influence of TAM and HYT on the protective effect of �E2against ouabain-induced amnesia in the step-through passive avoidancetask. All mice were gonadectomized, injected with 0.1 �g ouabain i.cist.,and given either vehicle, �E2 (10 �g kg−1; s.c.), TAM (0.1 �g; i.cist.),HYT (10 mg kg−1; s.c.), �E2 + TAM, or �E2 + HYT. Neither TAM norHYT influenced the latency time on the retention trial of �E2-treated mice(see Section 3 for details). Values are mean ± S.E.M. for 8–12 animalsper group. ∗P < 0.05 vs. STL in the vehicle group (by post hoc test).�E2 = 17�-estradiol, TAM: tamoxifen, HYT: 4-hydroxytamoxifen.

estradiol [22] or androstenedione to oestrone [23], the pro-tective effect of TES could result from its aromitizationto an estrogen. This possibility was further suggested bythe observation that DHT, a non-aromatizable androgen,did not protect against ouabain-induced amnesia (Fig. 3).To further address this possibility, we studied whether theprotective effect of TES against ouabain-induced amnesiawas mediated by androgen receptors.

Fig. 6 shows the influence of cyproterone, an antiandro-gen, on the protective effect of TES against ouabain-inducedimpairment of learning and memory in the step-through pas-sive avoidance task. Cyproterone was given s.c. at a doseof 10 mg kg−1, which was reported to antagonize the ac-tion of 1 mg kg−1 TES [24], and the i.cist. doses used were0.1 and 1 �g per mouse. The s.c. or i.cist. administration ofcyproterone alone did not affect ouabain-induced amnesiain the step-through passive avoidance task, and cyproteroneadministered s.c. or i.cist. in combination with TES did notsignificantly modify the protective action of TES againstouabain-induced amnesia. Cyproterone at a dose of 0.1 �g(i.cist.) also had no effect on ouabain-induced amnesia orthe protective effect of 1 mg kg−1 TES (data not shown).

Fig. 6. Influence of cyproterone treatment on the protective effect ofTES against ouabain-induced impairment of learning and memory in thestep-through passive avoidance task. All mice were gonadectomized andinjected with 0.1 �g ouabain i.cist., and then given TES (1 mg kg−1; s.c.),CYP (10 mg kg−1; s.c. or 1 �g; i.cist.) or TES+CYP (either s.c. or i.cist.).Both doses of CYP failed to affect the recovery from ouabain-inducedamnesia by TES (see Section 3 for details). Values are mean ± S.E.M.

for 8–10 animals per group. ∗P < 0.05 vs. STL in the vehicle group (bypost hoc test). TES: testosterone, CYP: cyproterone.

3.7. The effect of TAM on the protective action of TESagainst ouabain-induced amnesia in the step-throughpassive avoidance task

As mentioned above, the protective effect of TES suggeststhat estradiol, which can be generated by the aromatizationof TES, could affect ouabain-induced amnesia via estradiolreceptors. We, therefore, examined the effect of TAM on theprotective action of TES against ouabain-induced amnesia inthe step-through passive avoidance task. As shown in Fig. 7,0.1 �g TAM (i.cist.) did not influence the protective actionof TES (1 mg kg−1).

3.8. The effects ofαE2 and several antioxidant drugson ouabain-induced amnesia in the step-through passiveavoidance task

As mentioned in Section 1, the protective effects of �E2and TES may result from their antioxidant properties. We,therefore, examined the effect of �E2, a steroisomer of�E2 that lacks estrogen activity but which has free radi-cal scavenging properties (see Section 4), and compared

Fig. 7. The effect of TAM on TES-mediated protection againstouabain-induced amnesia in the step-through passive avoidance task. Allmice were gonadectomized and injected with 0.1 �g ouabain i.cist., andgiven TES (1 mg kg−1; s.c.) or TES+TAM (0.1 �g; i.cist.). TAM did nothave a significant effect on the recovery from ouabain-induced amnesiaby TES (see Section 3 for details). Values are mean ± S.E.M. for 8–10animals per each group. TES: testosterone, TAM: tamoxifen.

this with the effects of several antioxidant drugs againstouabain-induced amnesia in the step-through passive avoid-ance task (Fig. 8). There was no significant differenceamong the various antioxidant drug-treated groups. In con-trast to the effect of �E2 (10 �g kg−1, Fig. 3), �E2 at thesame dose did not reduce ouabain-induced amnesia. Over-all, the effects of the antioxidants were not significant;however, there was a non-significant trend (P = 0.17) forinhibition of ouabain-induced amnesia at high doses of VE(200 mg kg−1, p.o.), but not of VC (200 mg kg−1, p.o.). Oth-erwise, there were no significant effects of administrationof VE alone or in combination with VC.

3.9. The effects of steroids and antioxidant drugs onlocomotor activity in GOX-mice treated with ouabain

In order to examine the non-cognitive influence of thedrugs, locomotion was measured in GOX injected withouabain (Table 1). ANOVA revealed no significant differ-ences between the groups.

4. Discussion

The present study showed that there was no signifi-cant difference in step-through latency between Sham andGOX-mice, and that the administration of �E2 and TES

Fig. 8. The effects of �E2 and several antioxidant drugs againstouabain-induced amnesia in the step-through passive avoidance task. Allmice were gonadectomized and injected with 0.1 �g ouabain i.cist., andgiven �E2 (10 �g kg−1; s.c.), VE (200 mg kg−1; p.o.), VC (200 mg kg−1;p.o.), or VE+VC. The vehicles were administrated s.c. (vehicle 1) or p.o.(vehicle 2). None of the radical scavengers had a significant effect againstouabain-induced amnesia in comparison with the corresponding controls(see Section 3 for details). Values are mean±S.E.M. for 8–10 animals pergroup. �E2 = 17�-estradiol, VE = �-tocopherol, VC = ascorbic acid.

did not affect latency in either Sham or GOX-mice. Sim-ilarly, �E2 TES did not affect locomotor activity in theGOX-mice injected with ouabain. If the administration of�E2 or TES increased the animals’ electric sensitivity, thenthe latency of the �E2 and TES group on the STL should

Table 1Effects of steroid hormones, a steroid antagonist, and free radical scav-engers on locomotor activity in GOX-mice administered ouabain

Treatment Locomotor counts (10 min−1)

0–10 min 10–20 min

Vehicle 157.4 ± 19.5 107.2 ± 18.6�E2 (10 �g kg−1) 160.2 ± 25.3 99.0 ± 14.0�E2 (10 �g kg−1) + TAM (0.1 �g) 141.4 ± 28.1 99.4 ± 11.0TES (1 mg kg−1) 141.6 ± 26.5 91.6 ± 16.5TES (1 mg kg−1) + CYP (1 �g) 129.0 ± 20.1 86.4 ± 12.0TES (1 mg kg−1) + TAM (0.1 �g) 149.2 ± 34.3 83.0 ± 16.0�E2 (10 �g kg−1) 150.8 ± 30.5 87.0 ± 14.6VE (200 mg kg−1) 131.8 ± 25.9 85.8 ± 8.3VC (200 mg kg−1) 150.2 ± 32.2 90.8 ± 7.3VE (200 mg kg−1)

+ VC (200 mg kg−1)147.4 ± 30.1 99.8 ± 18.6

The locomotor activity of mice was measured 20 min before the trainingsession of the step-through test. Each value represents the mean ±S.E.M.

of five mice.

have increased as compared with that of the Sham andGOX group’s latency. This data suggests that the poten-tial non-cognitive effects of �E2 or TES, such as changesin footshock sensitivity and motivation or sensory/motoraspects of the task, were in fact minimal.

It is already reported that �E2 enhanced memory per-formance in male rats [25], and our results clearly demon-strate that the sex steroid hormones, �E2 and TES blocked,while the adrenocorticosteroid COR failed to inhibitouabain-induced impairment of learning and memory inGOX-mice.

It has been previously shown that the administration ofouabain produces necrotic lesions, characterized by missiveinvasion of foaming macrophages in the hippocampus [5,10]and memory deficits [26]. Additionally, ouabain impairs theintermediate stage of Gibbs et al. three-stage model of mem-ory formation [27], and as mentioned above, the brain le-sions induced by ouabain may be caused by the release ofEEA, although the precise mechanism of ouabain-inducedamnesia is unknown. In the present study, we found thatouabain at a dose of 0.1 �g decreased the latency on thestep-through passive avoidance task. Although 1 �g ouabaindid not effect STL, it did have significant effect on the Train.trial. Furthermore, ouabain at a dose of 1 �g tended to de-crease locomotor activity, while 5 �g increased the latencyon the memory performance task even though locomotionwas clearly inhibited (data not shown). We did not use thedose of 1 �g in this study since high doses of ouabain maycause brain lesions via the release of EAA, and could resultin the increase of latency by non-cognitive processes.

The effect of COR on the CNS has been generally por-trayed as enhancing memory formation [28,29]. Our datadid not show clear enhancement of ouabain-induced amne-sia by COR. In the study, in order to make direct compar-isons TES was administered at the same doses as those usedfor COR. Although the doses of COR used in this study mayhave been somewhat low, doses of 3–7 mg kg−1 per day aregenerally considered to result in high, stress levels of CORafter injection [30].

Although we intended to elucidate the mechanism un-derlying the protective effects of �E2 and TES againstouabain-induced impairment of learning and memory, wewere unable to do so in the present study. However, wecan speculate on some of the possible mechanisms. �E2at a dose of 10 �g kg−1 blocked ouabain-induced impair-ment, and we examined whether the protective effect wasmediated by estradiol receptors. In short, we examinedthe influence of TAM and HYT (a more potent antago-nist than TAM) on the protective effect of �E2 againstouabain-induced amnesia. However, both TAM (0.1 �g permouse i.cist. or 10 mg kg−1 s.c.) and HYT (10 mg kg−1 s.c.)failed to block the protective effect of �E2. Since estrogenreceptor antagonists did not modulate the effect of �E2,our results suggest that the protective effect of �E2 maybe caused by a non-genomic effect [31,32]. However, it isalso possible that the concentrations of TAM and HYT that

were used in the present study may have been insufficientto cause antagonism of the genomic receptors of estradiol.As mentioned above, s.c. injection of TAM or HYT mayaffect the pituitary and hypothalamus but not regions re-lated to learning and memory, such as the hippocampus andamygdala. Therefore, it was difficult to interpret the nega-tive data by the results of the s.c. injections. Additionally,since 1 or 5 �g (i.cist.) of TAM alone reduced the latencytime on the step-through task (Fig. 4), and doses in ex-cess of 10 �g (i.cist.) inhibited locomotor activity (data notshown), we could not sufficiently examine the relationshipbetween �E2 and the genomic antagonists (TAM and HYT).However, if the protective effect of �E2 was induced by anon-genomic action, the ameliorating effect of �E2 againstthe ouabain-induced amnesia would caused by membranereceptors (not classical estrogen receptors). Therefore, theactivation of the membrane receptors may induce the fa-cilitation of adenylate cyclase–cAMP signaling pathway[33] or inositol triphosphate signaling pathway, which arecoupled with G-protein receptors [34]. Consequently, theMAP kinase pathway [33,35] may activate some growthfactor receptors [35,36], and the activation of receptors mayelicit the consolidation of synaptic connectivity [37]. Thesefactors may be critical for neuronal health and survivalfrom ouabain-induced toxicity and may also play a role inlearning and memory formation in their own right.

Except for its free radical scavenging action, �E2 is a bi-ologically and physiologically inactive stereoisomer of �E2.Albaladejo et al. [38] reported that �E2 has lower affin-ity than �E2 for classical estrogen receptors in uterine andMtTF4 tumor tissues. In the present study, the �E2 did notaffect ouabain-induced amnesia. Since �E2 and �E2 shouldhave similar free radical scavenging properties, our resultssuggest that the protective effect of �E2 on ouabain-inducedamnesia is not attributable to free radical scavenging. To de-termine the influence of free radicals induced by the adminis-tration of ouabain, we also studied the effects of other antiox-idants (i.e. VC and VE). Many studies have suggested thatVE and VC can alleviate ischaemia-induced neuronal death[39,40] and protect against �-amyloid neurotoxicity [41].However, as shown in Fig. 8, VE and VC (200 mg kg−1)given alone or in combination did not significantly pro-tect against ouabain-induced amnesia. It was, however, dif-ficult to determine the optimal doses for VE and VC. The200 mg kg−1 dose of VE and VC used in the present studywas chosen based on the doses used in previous studies. Forexample, VE at doses of 50–200 mg kg−1 is effective in min-imizing hypoxia-induced synaptic transmission failure [42]and ozone-induced memory deficits [43]. In addition, it isreported that VC at 100–200 mg kg−1 has a protective effectagainst oxidative stress [44,45]. Therefore, our results using�-estradiol and antioxidants suggest that the protective ef-fect of �-estradiol on ouabain-induced amnesia did not arisefrom free radical scavenging. Although we observed a slighttrend towards inhibition of ouabain-induced amnesia by VEpretreatment, we were unable to conclude that the protec-

tive effect of �E2 is caused by its free radical scavengingaction.

Although 1 mg kg−1 of TES protects the ouabain-inducedamnesia, DHT (the final metabolite of testosterone asidefrom nonaromatizable androgen) had no effect. Addition-ally, the protection by TES against ouabain-induced amnesiawas not antagonized by either CYP (an androgen antagonist)or TAM. These results suggest that the protective effect ofTES does not involve its function as an androgen and raisesthe possibility that TES is aromatized to an estrogen, whichsubsequently affects the impairment of learning and mem-ory by a non-genomic function that does not appear to bemediated by free radical scavenging. If the protective effectof TES does involve such a mechanism, then the underlyingmechanisms are likely similar to those of �E2.

As mentioned previously, the impairment of memoryby ouabain could be mediated by excitatory amino acids[5], and several studies have shown that �E2 protectsagainst excitatory amino acid toxicity [15,46]. For exam-ple, Weaver et al. [15] suggested that �E2 protects againstNMDA-induced neuronal death by directly inhibiting theNMDA receptor in vitro, and that the protective effectof �E2 involves a non-genomic action of estrogen. Ourresults may be consistent with those of Weaver et al. More-over, is it also possible that �E2 directly increased (Na+,K+)-ATPase activity. To our knowledge, there is no priorexperimental data to suggest a direct effect of �E2 on theactivity of (Na+, K+)-ATPase in brain tissues; however,Golden et al. [47] and Dzurba et al. [48] found that estradioldirectly stimulates the activity or improves the kinetics of(Na+, K+)-ATPase in human erythrocytes and in rat cardiacsarcolemma, respectively. Our assertion is highly specula-tive and whether the protective action of �E2 is caused bynon-genomic mechanisms, and identification of the precisesite of action of �E2, awaits further investigation.

The present study showed that estrogen and testosteroneare capable of reversing learning and memory impairmentcaused by the administration of ouabain. Since ouabain is aspecific inhibitor of (Na+, K+)-ATPase, estrogen and testos-terone may rescue an inhibition of (Na+, K+)-ATPase asso-ciated with cerebral ischaemic diseases and the deficiencyof ATP [1].

References

[1] Pulsinelli W. Pathophysiology of acute ischaemic stroke. Lancet1992;339:533–6.

[2] Pulsinelli WA, Duffy TE. Regional energy balance in rat brain aftertransient forebrain ischemia. J Neurochem 1983;40:1500–3.

[3] Flamm ES, Demopoulos HB, Seligman ML, Poser RG, RansohoffJ. Free radicals in cerebral ischemia. Stroke 1978;9:445–7.

[4] Kogure K, Watson BD, Busto R, Abe K. Potentiation of lipid per-oxides by ischemia in rat brain. Neurochem Res 1982;7:437–54.

[5] Lees GJ. Inhibition of sodium-potassium ATPase: a potentiallyubiquitous mechanism contributing to central nervous system neu-ropathology. Brain Res Rev 1991;16:283–300.

[6] Rothman SM, Olney JW. Excitotoxicity and the NMDA receptor.Trend Neurosci 1987;10:299–302.

[7] Veldhuis WB, van der Stelt M, Delmas F, Gillet B, Veldink GA,Vliegenthart JF, et al. In vivo excitotoxicity induced by ouabain, aNa+/K+-ATPase inhibitor. J Cereb Blood Flow Metab 2003;23:62–74.

[8] Sato T, Kamata Y, Irifune M, Nishikawa T. Inhibition of purified(Na+, K+)-ATPase activity from porcine cerebral cortex by NOgenerating drugs. Brain Res 1995;704:117–20.

[9] Sato T, Kamata Y, Irifune M, Nishikawa T. The inhibitory ef-fect of several nitric oxide-generating compounds on purified (Na+,K+)-ATPase activity from porcine cerebral cortex. J Neurochem1997;68:1312–8.

[10] Lees GJ, Leong W. Brain lesions induced by specific and non-specificinhibitors of sodium-potassium ATPase. Brain Res 1994;649:225–33.

[11] Bell GA, Gibbs ME. Unilateral storage of monocular engram inday-old chick. Brain Res 1977;124:263–70.

[12] Gibbs ME, Jeffrey PL, Austin L, Mark RF. Separate biochemicalactions of inhibitors of short- and long-term memory. PharmacolBiochem Behav 1973;1:693–701.

[13] Chiueh C, Lee S, Andoh T, Murphy D. Induction of antioxidativeand antiapoptotic thioredoxin supports neuroprotective hypothesis ofestrogen. Endocrine 2003;21:27–31.

[14] Hurn PD, Brass LM. Estrogen and stroke—a balanced analysis.Stroke 2003;34:338–41.

[15] Weaver Jr CE, Park-Chung M, Gibbs TT, Farb DH. 17�-Estradiolprotects against NMDA-induced excitotoxicity by direct inhibitionof NMDA receptors. Brain Res 1997;761:338–41.

[16] Ueda H, Amano H, Shiomi H, Takagi H. Comparison of the analgesiceffects of various opioid peptides by a newly devised intracisternal in-jection technique in conscious mice. Eur J Pharmacol 1979;56:265–8.

[17] Wiseman H. Tamoxifen: importance of understanding the anticanceraction of tamoxifen at a molecular level. New York: Wiley; 1994.p. 1–7.

[18] Bignon E, Pons M, Gilbert J, Nishizuka Y. Multiple mechanismsof protein kinase C inhibition by triphenylacrylonitrile antiestrogens.FEBS Lett 1990;271:54–8.

[19] Wiseman H. Tamoxifen: antioxidant action of tamoxifen and re-lated compounds in membranes and lipoprotein particles: role in theanticancer, anti-multidrug resistance and cardioprotective actions oftamoxifen. New York: Wiley; 1994. p. 119–59.

[20] Somjen D, Waisman A, Kaye AM. Tissue selective action of tamox-ifen methiodide, raloxifene and tamoxifen on creatine kinase B activ-ity in vitro and in vivo. J Steroid Biochem Mol Biol 1996;59:389–96.

[21] Kupfer D, Mani C, Lee CA, Rifkind AB. Induction oftamoxifen-4-hydroxylation by 2,3,7,8-tetrachlorodibenzo-p-dioxin(TCDD), �-naphthoflavone (�-NF), and phenobarbital (PB) in avianliver: identification of P450 TCDDAA as catalyst of 4-hydroxylationinduced by TCDD and �-NF. Cancer Res 1994;54:3140–4.

[22] Weiz J, Gibbs C. Conversion of testosterone and androstenedioneto estrogens in vitro by the brain of female rats. Endocrinology1974;94:616–20.

[23] Naftolin F, Ryan KJ, Petro Z. Aromatization of androstenedione bythe anterior hypothalamus of adult male and female rats. Endocrinol-ogy 1972;90:295–8.

[24] Roy EJ, Wade GN. Role of estrogens in androgen-induced sponta-neous activity in male rats. J Comp Physiol Psychol 1975;89:573–9.

[25] Luine V, Rodriguez M. Effects of estradiol on radial arm maze per-formance of young and aged rats. Behav Neural Biol 1994;62:230–6.

[26] Watts ME, Mark RF. Separate actions of ouabain and cycloheximideon memory. Brain Res 1971;25:420–3.

[27] Gibbs ME, Andrew RJ, Ng KT. Hemispheric lateralization of memorystages for discriminated avoidance learning in the chick. Behav BrainRes 2003;139:157–65.

[28] Bowman RE, Zrull MC, Luine VN. Chronic restraint stress en-hances radial arm maze performance in female rats. Brain Res2001;904:279–89.

[29] Liu L, Tsuji M, Takeda H, Takeda K, Matumiya T. Adrenocorticalsuppression blocks the enhancement of memory storage produced byexposure to psychological stress in rats. Brain Res 1999;821:134–40.

[30] Sandi C, Venero C, Guaza C. Novelty-related rapid locomotor effectsof corticosterone in rats. Eur J Neurosci 1996;8:794–800.

[31] Moss RL. Estrogen nontranscriptional signaling pathway. RecentProg Horm Res 1997;52:33–70.

[32] Rupprecht R. Neuroactive steroids: neuroactive steroids: mechanismsof action and neuropsychopharmacological perspectives. Psychoneu-roendocrinology 2003;28:139–68.

[33] Beyer C, Ivanova T, Karolczak M, Kuppers E. Cell type-specificity ofnonclassical estrogen signaling in the developing midbrain. J SteroidBiochem Mol Biol 2002;81:319–25.

[34] Kelly MJ, Qiu J, Wagner EJ, Ronnekleiv OK. Rapid effects of estro-gen on G protein-coupled receptor activation of potassium channelsin the central nervous system (CNS). J Steroid Biochem Mol Biol2002;83:187–93.

[35] Lu Q, Ebling H, Mittler J, Baur WE, Karas RH. MAP kinase mediatesgrowth factor-induced nuclear translocation of estrogen receptor �.FEBS Lett 2002;516:1–8.

[36] Mendez P, Azcoitia I, Garcia-Segura LM. Estrogen receptor � formsestrogen-dependent multimolecular complexes with insulin-likegrowth factor receptor and phosphatidylinositol 3-kinase in the adultrat brain. Mol Brain Res 2003;112:170–6.

[37] Yankova M, Hart SA, Woolley CS. Estrogen increases synaptic con-nectivity between single presynaptic inputs and multiple postsynapticCA1 pyramidal cells: a serial electron-microscoic study. Proc NatlAcad Sci USA 2001;98:3525–30.

[38] Albaladejo V, Pharaboz MO, Morel Y, Andre J. Properties of theestrogen receptors contained in the MtTF4 tumor whose growth isinhibited by estradiol: 17�-estradiol, 17�-estradiol and DNA bind-ings. J Steroid Biochem 1986;25:29–36.

[39] Hara H, Kato H, Kogure K. Protective effect of �-tocopherol onischemic neuronal damage in the gerbil hippocampus. Brain Res1990;510:335–8.

[40] Yamamoto M, Shima T, Uozumi T, Sogabe T, Yamada K, KawasakiT. A possible role of lipoid peroxidation in cellular damages causedby cerebral ischemia and the protective effect of �-tocopherol ad-ministration. Stroke 1983;14:977–82.

[41] Harkany T, Mulder J, Sasvari M, Abraham I, Konya C, Zarandi M,et al. N-Methyl-d-aspartate receptor antagonist MK-801 and radi-cal scavengers protect cholinergic nucleus basalis neurons against�-amyloid neurotoxicity. Neurobiol Dis 1999;6:109–21.

[42] Vlkolinsky R, Stolc S. Effects of stobadine, melatonin, and otherantioxidants on hypoxia/reoxygenation-induced synaptic transmissionfailure in rat hippocampal slices. Brain Res 1999;850:118–26.

[43] Guerrero AL, Dorado-Martinez C, Rodriguez A, Pedroza-Rios K,Borgonio-Perez G, Rivas-Arancibia S. Effects of vitamin E onozone-induced memory deficits and lipid peroxidation in rats. Neu-roreport 1999;10:1689–92.

[44] Hasan MY, Alshuaib WB, Fahim MA. The impact of vitamin Con diabetes induced alterations at murine neuromuscular junction.Endocr Res 2002;28:49–59.

[45] Sanchez GM, Re L, Giuliani A, Nunez-Selles AJ, DavisonGP, Leon-Fernandez OS, et al. Mangiferin and selected antioxi-dants against TPA-induced biomolecules oxidation and peritonealmacrophage activation in mice. Pharmacol Res 2000;42:565–73.

[46] Regan RF, Guo Y. Estrogens attenuate neuronal injury due tohemoglobin, chemical hypoxia, and excitatory amino acids in murinecortical cultures. Brain Res 1997;764:133–40.

[47] Golden GA, Mason RP, Tulenko TN, Zubenko GS, Rubin RT. Rapidand opposite effects of cortisol and estradiol on human erythrocyteNa+, K+-ATPase activity: relationship to steroid intercalation intothe cell membrane. Life Sci 1999;65:1247–55.

[48] Dzurba A, Ziegelhöffer A, Vrbjar N, Styk J, Slezák J. Estradiolmodulates the sodium pump in the heart sarcolemma. Mol CellBiochem 1997;176:113–8.

effects of steroid hormones on (na+, k+)-atpase activity inhibition-induced amnesia on the...

Documents

digital amnesia-infographic

huyssen escape from amnesia

learning, memory & amnesia

atpase active-site electrostatic interactions control the...

atpase and lumac

suramin inhibits hsp104 atpase and disaggregase activity...

hippocampal amnesia - ucl

city of amnesia

60´ ferretti amnesia

memory, learning and amnesia

amnesia omniture training

organic amnesia

atpase dataset -> nj in figtree. atpase dataset -> muscle ->...

running head: amnesia and future thinking...

disorders of memory amnesia & animal models. lecture outline...

amnesia ikki

amnesia lemon kush

dementia and amnesia

amnesia tdd

severe anterograde amnesia temporally graded retrograde...