www.e l sev i e r . com/ loca te /eu roneu ro

European Neuropsychopharmacology (2011) 21, 261–273

Insulin receptor signaling in rat hippocampus: A studyin STZ (ICV) induced memory deficit modelRahul Agrawal a, Ethika Tyagi a, Rakesh Shukla a, Chandishwar Nath b,⁎

a Division of Pharmacology, Central Drug Research Institute, CSIR, Lucknow-226001, Indiab Division of Toxicology, Central Drug Research Institute, CSIR, Lucknow-226001, India

Received 2 May 2010; received in revised form 17 September 2010; accepted 30 November 2010

⁎ Corresponding author. Scientist G, DDrug Research Institute (CDRI), P.O.India. CDRI Communication no 7870. Tefax: +91 522 2623405, 2623938.

E-mail address: cnathcdri@rediffm

0924-977X/$ - see front matter © 201doi:10.1016/j.euroneuro.2010.11.009

KEYWORDSDonepezil;Hippocampus;Insulin receptor signaling;Melatonin;Memory;Streptozotocin

Abstract

Brain insulin receptors (IRs) have been suggested as an important regulatory factor for cognitivefunctions but the involvement of IR signaling in memory deficit associated with neurodegen-erative conditions is not yet explored. In the present study, IR gene expression was studied byRT-PCR and signaling pathways by immunoblotting in CA1, DG and CA3 subregions of hippocampusin intracerebroventricular (ICV) administered streptozotocin (STZ, 3 mg/kg twice) inducedmemory deficit model in rat. The effect of pre- and post-treatment of donepezil (5 mg/kg po)

and melatonin (20 mg/kg po) on signaling pathways were studied. Effect of LY294002 (ICV), a PI3Kinase inhibitor, was also investigated on memory functions and Akt phosphorylation. Anincreased IR expression (both gene and protein), phosphorylation of Shc, Erk1/2, IRS-1 and Akt inCA1 and CA3 region of P2M fraction was observed after training as compared to control. STZtreated rats showed memory deficit and significant decrease in IR expression, phosphorylation ofIRS-1 and Akt only in CA3 region as compared to trained group which were reversed by pre andpost-treatment of melatonin but donepezil was effective only against memory deficit. LY294002(3 mM) treatment showed delayed learning and decrease in Akt phosphorylation. This studysuggests that IR expression and its signaling pathways in hippocampal CA1 and CA3 regions areinvolved in memory functions and STZ (ICV) induced memory deficit. Hippocampal IR systemmight be playing an important role in regulation of memory functions, however only IR/IRS-1/Aktpathway in CA3 region is associated with STZ induced memory deficit.© 2010 Elsevier B.V. and ECNP. All rights reserved.

ivision of Toxicology, CentralBox 173, Lucknow-226001,l.: +91 522 2612411 18x4420;

ail.com (C. Nath).

0 Elsevier B.V. and ECNP. All r

igh

1. Introduction

Recently, the brain insulin/insulin receptors (IRs) system isreceiving major attention in regulation of brain functions.Insulin receptors (IR) are distributed in the brain regions and

ts reserved.

262 R. Agrawal et al.

differ from peripheral IR in molecular weight of both α and βsubunits (Zhao et al., 1999). The presence of IR in thehippocampus suggests its functional involvement in cognition(Dou et al., 2005). Moreover, deregulation of brain IR hasbeen linked to the pathogenesis of age-related neurodegen-erative disorders such as Alzheimer's (Frolich et al., 1998)and Parkinson's disease (Takahashi et al., 1996).

The molecular cascades downstream from IR are com-posed of a large number of signaling molecules includinginsulin receptor substrates (IRSs) and Src homology 2 (SH2)and SH2-SH3 domain-containing proteins. Among the diversesignaling pathways of IR, insulin receptor substrate-1 (IRS-1)/PI-3 kinase/phosphoinositide-dependent kinase (PDK)/protein kinase B (PKB/Akt) and the SH2 and collagencontaining protein (Shc)/growth factor receptor-boundprotein-2 (Grb2)/mitogen-activated protein (MAP) kinasepathways in brain have been suggested for learning andmemory functions (Zhao et al., 2004).

A number of signaling molecules have been implicated inlearning and memory functions in different species, such asMAPK activity is involved in associative learning (Atkins et al.,1998), spatial learning in the Morris water maze task and armradial maze (Selcher et al., 1999; Zhao et al., 1999), inhibitoryavoidance learning (Izquierdo et al., 2000) and contextual andauditory fear conditioning (Schafe et al., 1999; Selcher et al.,1999). It has been suggested that MAPKmediates changes in thelong-term storage of information in the brain, a process thatrequires participation of gene regulation and expression. Themolecular events, such as activation of PKC and Akt/PKB Ser/Thr kinase, may also play a role in memory storage. Takentogether, these studies suggest that IR signaling has a role in theregulation of memory but the involvement of IR signaling inmemory deficit linked with neurodegenerative conditions is notyet explored.

Intracerebroventricular (ICV) administration of strepto-zotocin (STZ) in rat has been described as an appropriateexperimental model for dementia close to humans as bothare characterized by progressive deterioration of memory,cerebral glucose and energy metabolism (Lannert and Hoyer,1998). Grünblatt et al. (2007) demonstrated brain insulinsystem dysfunction in this model, which may lead to anincrease in hyperphosphorylated tau-protein concentrationin hippocampus. Recently, we had reported the decrease inIR protein expression in hippocampus in STZ (ICV) inducedmemory deficit (Agrawal et al., 2009).

Hippocampus is the key area to regulate memory functions.It has also been reported that STZ (ICV) causes chronicreductions in glucose and glycogen metabolism in hippocampus(Plaschke and Hoyer, 1993). Recent studies have shown theregional specificity of hippocampus in memory processes. Theinvolvement of CA3 seems to be important at the earliest stageof acquisition, presumably for developing instant representa-tion of a context, whereas the CA1 and dentate gyrus (DG) werecritically involved in retrieving contextual memory after a longtime period (Lee and Kesner, 2004). Inspite of differentfunctional role of hippocampal CA1, DG and CA3 subregions inmemory functions, the role of IR and its signaling present inthese areas has not been investigated in relation to learning andmemory functions.

Zhao et al. (1999) previously reported the traininginduced changes in IR in subregions of hippocampus whichis associated with short-term memory. In our previous study

(Agrawal et al., 2009), we had reported the effect ofdonepezil, an anti-dementia drug and melatonin, an antiox-idant on IR in STZ induced memory deficit in brain areas. Butstudies on IR expression and its signaling in hippocampalsubregions associated with state of memory deficit andeffect of anti-dementia and antioxidant drugs are lacking.

Therefore, the present study was planned to investigatethe involvement of IR and its signaling pathways in CA1, DGand CA3 subregions of hippocampus in memory deficit modelof rat induced by STZ (ICV). Further, we have also includedthe study on pre- and post-treatment of donepezil, an anti-dementia drug and melatonin, an antioxidant on thesesignaling pathways in STZ (ICV) induced memory impairment.

2. Experimental procedures

2.1. Animals

The experiments were carried out in adult male Sprague Dawley (SD)rats (220–250 g). They were kept in polyacrylic cage and maintainedunder standard housing condition (room temperature 24–27 °C andhumidity 60–65%) with 12 h light and dark cycle. There were 5 rats ineach group. The food in form of dry pellets and water were availablead libitum. The rats were procured from the Laboratory AnimalServices Division of Central Drug Research Institute, Lucknow, India.Experiments were performed as per internationally followed ethicalstandards, after clearance from Institutional Animal Ethics Commit-tee of Central Drug Research Institute which is approved fromCommittee for the Purpose of Control and Supervision of Experi-ments on Animals, Government of India on animal experimentation.

2.2. Materials

The bio-chemicals streptozotocin (STZ), LY294002, dimethylsulph-oxide (DMSO), melatonin, chloral hydrate, sodium chloride (NaCl),potassium chloride (KCl), magnesium chloride (MgCl2), calciumchloride (CaCl2), dextrose, bovine serum albumin (BSA), sucrose,HEPES, and protease inhibitor cocktail [containing 4-(2-aminoethyl)benzenesulfonyl fluoride (AEBSF), pepstatin A, E-64, bestatin,leupeptin and aprotinin] were purchased from Sigma-Aldrich, USA.Donepezil was purchased from Hetero Drugs Ltd., India.

2.3. Intracerebroventricular (ICV) injection

STZ (3 mg/kg) was administered intracerebroventricularly (ICV)twice at an interval of 48 h to produce memory impairment. Ratswere anaesthetized with Chloral hydrate (300 mg/kg ip) andmounted in stereotaxic apparatus (Stoelting co. USA). STZdissolved in artificial CSF was injected slowly by microlitre syringe(Hamilton) in a volume of 10 μl on each side using the coordinates:0.8 mm posterior to bregma; 1.5 mm lateral to saggital suture;3.6 mm ventral (Agrawal et al., 2009). In CSF treated group,artificial CSF: 147 mM NaCl; 2.9 mM KCl; 1.6 mM MgCl2; 1.7 mMCaCl2 and 2.2 mM dextrose was injected in the same volume at aninterval of 48 h.

In another set of experiments, LY294002 (PI3-K inhibitor),dissolved in 3% dimethylsulphoxide (DMSO), was administeredintracerebroventricularly (ICV) at the doses of 0.3, 1 and 3 mM in avolume of 10 μl on each side. In vehicle treated group, 3% DMSO wasinjected in the same volume.

2.4. Drug administration

Donepezil (5 mg/kg po) or melatonin (20 mg/kg po) was adminis-tered orally using feeding cannula from the day of 1st STZ injection

263Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

till the 14th day to study the preventive effect of these drugs(Pre-treatment schedule). In another set, donepezil (5 mg/kg po) ormelatonin (20 mg/kg po) was administered 1 h prior to daily sessionof trials that were given from 14th to 20th days after 1st STZinjection for therapeutic effect (Post-treatment schedule). The doseof donepezil and melatonin was selected on the basis of our previousreport in which memory improving ability and significant anti-cholinesterase and antioxidant property has been demonstratedrespectively (Agrawal et al., 2009). The groups in the study with STZmodel were: control (no treatment), Trained (vehicle treated), CSF,STZ, donepezil pre-treated (DON+STZ), donepezil post-treated(STZ+DON), melatonin pre-treated (MEL+STZ) and melatonin post-treated (STZ+MEL). All the groups were subjected to the Morriswater maze except control (no treatment).

2.5. Blood glucose levels

Blood was collected by tail prick on 14th day from 1st STZ injectionand glucose level was checked by Accu-Check Sensor Comfortglucostrips (Roche Diagnostics India Pvt. Ltd.) in control, trained,CSF, STZ, donepezil pre-treated (DON+STZ), donepezil post-treated(STZ+DON), melatonin pre-treated (MEL+STZ) and melatonin post-treated (STZ+MEL) groups.

2.6. Morris water maze (MWM) test

Learning and memory of rats were tested in Morris water maze on14th day from 1st STZ injection (Morris, 1984). In LY294002 andDMSO treated group, memory was tested after 24 h of ICV injection.The Morris water maze consisted of a large circular pool of 120 cmdiameter, 50 cm height, filled to a depth of 30 cm with watertemperature 25±1 °C, which was placed in a darkened room. Withinthe pool, a submerged black colored round platform of 8 cmdiameter was placed 1 cm below surface of water. The water wascolored by a non-toxic black dye to hide the platform location.During the experiment the rat was placed in the water to find out thehidden platform to escape from the swimming. Same startingposition was used on each trial. The rats were given a maximum timeof 120 s (cut off time) to find the hidden platform and were allowedto stay on it for 30 s. Rats failing to find the platform were gentlyguided on to the platform only in 1st trial (acquisition trial). Theanimals were given a daily session of 4 trials per day for 7 days.During a session, latency times to reach the platform were recordedin each trial and mean latency time was calculated. Significantdecrease in latency time in subsequent sessions (retention) fromthat of 1st session (acquisition) was considered as successfullearning.

2.7. Tissue collection

Rats were sacrificed by perfusion through heart with ice-cold normalsaline under ether anesthesia on 21st day after 1st STZ injection.LY294002 and DMSO treated rats were sacrificed after 24 h of Morriswater maze test. The brain was carefully removed from each rat andkept over a glass plate placed on ice and hippocampus was subdissected into CA1, CA3, and dentate gyrus (DG) tissues according toLein et al. (2004).

2.8. Sample preparation for Immunoblotting

The cytosolic and crude synaptic membrane fractions of hippocam-pal tissue were prepared by centrifugation according to Dou et al.(2005). The different hippocampal brain tissues were homogenizedin precooled Buffer A (0.32 M sucrose, 2 mM HEPES, pH 7.4, and5% w/v protease inhibitor cocktail) with Teflon micro-pestle.The homogenate was diluted to 10% (w/v) with Buffer A, andcentrifuged at 1000×g for 5 min. The supernatant (S1) was

preserved, whereas the pellet (P1) was washed with Buffer A andcentrifuged at 1000×g again for 5 min. The supernatant wascombined with preserved S1 and centrifuged at 10,000×g for20 min. The supernatant (S2) from this step was used as thecytosolic fraction and pellet (P2) was osmotically lysed (10 ml/goriginal tissue) on ice for 30 min in Buffer B (2 mM HEPES, pH 7.4 and5% w/v protease inhibitor cocktail) and then centrifuged at20,000×g for 30 min. The pellet (P2M) from this step was collectedas a crude synaptic membrane fraction, redissolved in Buffer B, andstored at −80 °C until use.

2.8.1. SDS-PAGE and ImmunoblottingProtein was estimated according to Lowry et al. (1951) using

bovine serum albumin (BSA) as standard. Equal protein from P2M andS2 fractions of each sample were individually loaded for electro-phoresis, followed by transfer onto a nitrocellulose membrane. Themembrane was incubated overnight at 4 °C in blocking buffer, toprevent non-specific binding of antibodies to the membrane.Following the incubation, the membrane was washed with washingbuffer (1× TBS, pH 7.6 containing 0.1% Tween 20; 1× TBST) andincubated with primary antibody such as anti-IR, anti-phospho-IRS-1,anti-phospho-Akt, anti-phospho-Shc, anti-phospho-Erk1/2 (all fromSanta Cruz, USA) for 2 hr at room temperature, washed again with 1×TBST and then incubated with secondary antibody (Santa Cruz, USA)conjugated with horseradish peroxidase for 1 h at room tempera-ture. Following washes with 1× TBST, membranes were reacted withchemiluminescent reagents (Pierce Biotechnology, inc.) and ex-posed to an x-ray film. Membranes were stripped with β-mercap-toethanol stripping solution (62.5 mM Tris–HCl, pH 6.8; 2% SDS, and100 mM β-mercaptoethanol) and then re-probed with an antibodythat recognizes β-actin or anti-IRS-1 or anti-Akt or anti-Shc (all fromSanta Cruz, USA) or anti-Erk1/2 (Cell Signaling Technology, USA).Relative optical density of bands was analyzed using AlphaImager geldocumentation system.

2.9. Gene expression studies

Total RNA was extracted from brain tissues using TRIZOL Reagent(Sigma co. USA) by a single step RNA isolation method. Followingprecipitation, RNA was resuspended in RNAse-free water and itsconcentration was estimated by absorbance at 260-nm wavelength.RNA samples were stored at −80 °C until analysis.

Single-stranded cDNA was generated from 5 μg of total cellularRNA using Oligo-(dT)-primers and RevertAid cDNA synthesis kit(Fermentas). Briefly, 5 μg of total RNA was uncoiled by heating(70 °C for 5 min) in DNA thermal cycler (Bioer XP cycler) and thenreverse transcribed into complementary DNA (cDNA) in a 20 μlreaction mixture that contained 200 U Moloney murine leukemiavirus reverse transcriptase (M-MuLV-RT), 0.5 μg oligo-(dT)-primer,1 μl Ribolock RNase Inhibitor (20 U/μl), 2 μl of a 10-mM (each) dNTPmix, 4 μl 5× reaction buffer.

The resultant cDNA (4 μl) was amplified in a 20-μl reactionvolume containing 1 U Taq polymerase, 200 μM (each) dNTP mix,2 mM MgCl2, 2 μl 10× Taq buffer and specific primers (MWG biotech,Germany). The polymerase chain reaction mixture was amplified in aDNA thermal cycler (Bioer XP cycler) through a 35-cycle at thespecifications described in Table 1.

Polymerase chain reaction products were detected by electro-phoresis on a 1% agarose gel containing ethidium bromide. Locationof predicted products was confirmed by using 100-bp ladder(Fermentas) as a standard size marker. The intensity of polymerasechain reaction products was measured using AlphaImager geldocumentation system. The signals for examinated mRNAs werenormalized with that of the actin mRNA from each sample and theresults were expressed as PCR-product/actin mRNA ratio.

Table 1 Sequences of primers used in experiments andproducts size.

Primer Sequence Annealingtemperature[°C]

bp

β-actin 5′-GGCTGTGTTGTCCCTGTAT-3′5′-CCGCTCATTGCCGATAGTG-3′

55.7 352

Insulinreceptor

5′-CTCTGTCCGCATCGAGAA-3′5′-CCACACACTGACCATCCA-3′

50 372

264 R. Agrawal et al.

2.10. Statistical analysis

Results are expressed as mean ±S.E.M. and the statistical analysis wasdone by one-way analysis of variance (ANOVA), followed by theNewman–Keuls test to determine the significance of difference amongvarious groups. The p-valueb0.05 was considered as statisticallysignificant.

3. Results

3.1. Blood glucose levels

There was no significant difference (F(7,32)=0.8908, pN0.05)between the blood glucose levels (mg/dl) among control (107.8±3.72), trained (106.0+2.88), CSF (104.4+5.95), STZ (110.2+8.40),donepezil pre-treated (DON+STZ; 109.8±3.98), donepezil post-

0

20

40

60

80

100

Esc

ape

late

ncy

tim

e (s

ec)

session 1 session 2 session 3 sessio

session 1 session 2 session 3 sessio

TRA

**

** **** **

**

0

10

20

30

40

50

60

70

80

90

Esc

ape

late

ncy

tim

e (s

ec)

DON+STZ STZ+DON

*

*******

****

**

****

a

b

Figure 1 Comparison of escape latency times of (A) trained (TRA)post-treated (STZ+DON), melatonin pre-treated (MEL+STZ) and me*Significant difference (*Pb0.05, **Pb0.01) in latency time from ses

treated (STZ+DON; 112.0±2.12), melatonin pre-treated (MEL+STZ; 119.4±6.07) and melatonin post-treated (STZ+MEL; 107.6±2.50) groups on 14th day after 1st ICV injection.

3.2. Morris water maze (MWM) test in STZ (ICV) anddrug treated groups

The trained (F(6,28)=10.42, pb0.0001) and CSF (F(6,28)=7.346,pb0.0001) group had shown improved Morris water mazeacquisition performance i.e. decreased escape latency to findthe hidden platform from the first to last day of training wherethe latency time on session 3, 4, 5, 6 and 7 were significantlylower than that of the session 1. In STZ treated group (F(6,28)=0.4266, pN0.05), there was no significant change in latencytime to reach the platform in retention trials (Session 2–7) incomparison to the acquisition trial (Session 1), showing a poorerlearning performance due to ICV-STZ infusion (Fig. 1a), but preand post-treatment of donepezil (F(6,28) =7.081, p=0.0001;F(6,28) = 4.132, p = 0.0043) and melatonin (F(6,28) = 7.175,p=0.0001; F(6,28) =20.04, pb0.0001) significantly decreasedthe latency time from session 3 as compared to session 1,showing memory improvement over STZ group (Fig. 1b).

3.3. IR gene expression by RT-PCR in STZ(ICV) induced memory deficit

In CA1 (F(3,16) = 7.873, p=0.0019) and CA3 (F(3,16) = 16.92,pb0.0001) regions, the IR gene expression was found to be

n 4 session 5 session 6 session 7

n 4 session 5 session 6 session 7

CSF STZ

** ** ** **

MEL+STZ STZ+MEL

******

******

**

********

, CSF, STZ and (B) donepezil pre-treated (DON+STZ), donepezillatonin post-treated (STZ+MEL) rats in Morris water maze test.sion 1; ANOVA (one-way) followed by Newman–Keuls test.

265Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

significantly higher in trained and CSF treated group as comparedto control whereas STZ (ICV) showed significant decrease in IRexpression only in CA3 region in comparison to trained groups.There were no significant changes observed in IR gene expressionin any of the group in DG (F(3,16)=1.036, pN0.05) region (Fig. 2).

3.4. Insulin receptor (IR) protein expression in STZ(ICV) induced memory deficit

In P2M fraction, trained and CSF treated group showedsignificant increase in IR protein expression in CA1 (F(3,16) =5.607, p=0.0080) and CA3 (F(3,16) =6.210, p=0.0053) regions ascompared to control. STZ (ICV) treated group showed thesignificant decrease in IR protein expression only in CA3 regionin comparison of trained groups. In DG (F(3,16)=0.1750, pN0.05)region, no significant change was observed in IR proteinexpression in any of the group (Fig. 3a).

In S2 fraction, the significant decrease was found in IRprotein expression only in CA3 (F(3,16)=8.654, p=0.0012) regionin trained and CSF treated group as compared to control butthere were no significant changes found in CA1 (F(3,16) =1.330,pN0.05) and DG (F(3,16)=0.5980, pN0.05) in any of the group.STZ (ICV) did not show any significant change in IR expression inany region in this fraction (Fig. 3b).

3.5. Molecules of IR/Shc/Erk pathway in STZ (ICV)induced memory deficit

In P2M fraction, trained and CSF treated group showedsignificant increase in phosphorylation of Shc and Erk1/2 in

Figure 2 IR gene expression examined by RT-PCR in CA1, DGand CA3 subregions of hippocampus. Each lane represents (1)control (CON), (2) trained (TRA), (3) CSF and (4) STZ treatedgroups; Lane M is a 100 bp-DNA ladder indicating the size of PCRproducts; IR gene expressions were normalized with actin.#Significant difference from control, $significant difference fromtrained and CSF (# $ Pb0.05, ## $$ Pb0.01); ANOVA (one-way)followed by Newman–Keuls test.

Figure 3 Insulin receptor (IR) protein expressions in (A) mem-brane (P2M) and (B) cytosolic (S2) fraction of CA1, DG and CA3subregions of hippocampus in control (CON), trained (TRA), CSF, andSTZ treated groups. The IR protein expressions were normalizedwith actin. #Significant difference from control, $significantdifference from trained and CSF (# $ Pb0.05, ## $$ Pb0.01);ANOVA (one-way) followed by Newman–Keuls test.

CA1 (F(3,16) =5.352, p=0.0096; F(3,16) =5.527, p=0.0085) andCA3 (F(3,16) =7.484, p=0.0024; F(3,16)=5.323, p=0.0098) regionsas compared to control whereas STZ (ICV) did not show anysignificant change in any region in comparison to trainedgroups. There were no significant changes observed in Shc andErk1/2 phosphorylation in any of the group in DG (F(3,16) =0.4581, pN0.05; F(3,16) =0.7622, pN0.05) region (Fig. 4a and b).

In S2 fraction, the significant increase in phosphorylation ofShc and Erk1/2 was found only in CA3 (F(3,16) =6.018, p=0.0060;F(3,16)=4.695, p=0.0155) region in trained and CSF treatedgroup as compared to control whereas STZ (ICV) did not show

Figure 4 Phosphorylation of (A) Shc and (B) Erk1/2 protein inmembrane (P2M) fraction of CA1, DG and CA3 subregions ofhippocampus in control (CON), trained (TRA), CSF and STZtreated groups. Phosphorylated Shc and Erk1/2 were normalizedwith total Shc and Erk1/2 respectively. #Pb0.05, ##Pb0.01Significant difference from control; ANOVA (one-way) followedby Newman–Keuls test.

Figure 5 Phosphorylation of (A) Shc and (B) Erk1/2 protein incytosolic (S2) fraction of CA1, DG and CA3 subregions ofhippocampus in control, trained, CSF and STZ treated groups.Phosphorylated Shc and Erk1/2 were normalized with total Shcand Erk1/2 respectively. #Pb0.05, ##Pb0.01 Significant differencefrom control; ANOVA (one-way) followed by Newman–Keuls test.

266 R. Agrawal et al.

any significant change in any region as compared to trainedgroups. In CA1 (F(3,16)=0.2840, pN0.05; F(3,16) =0.4028, pN0.05)and DG (F(3,16)=1.509, pN0.05; F(3,16)=1.339, pN0.05), nosignificant was observed in Shc and Erk1/2 phosphorylation inany of the group (Fig. 5a and b).

3.6. Molecules of IR/IRS-1/Akt pathway in STZ (ICV)induced memory deficit

In P2M fraction, phosphorylation of IRS-1 and Akt was foundsignificantly increased in trained and CSF treated group ascompared to control in CA1 (F(3,16)=5.331, p=0.0097; F(3,16)=9.418, p=0.0008) and CA3 (F(3,16)=9.414, p=0.0008; F(3,16)=12.22,

p=0.0002) regions. The significant decrease of IRS-1 and Aktphosphorylation was found in STZ (ICV) group in CA3 region only incomparison to trained groups. In DG (F(3,16)=3.040, pN0.05; F(3,16)=3.260, p=0.0491), there was no significant change found in IRS-1and Akt Phosphorylation in any of the group (Fig. 6a and b).

In S2 fraction, there was no significant change found inphosphorylation of IRS-1 in CA1 (F(3,16)=0.4387, pN0.05), DG(F(3,16)=0.2801, pN0.05) and CA3 (F(3,16)=0.5543, pN0.05) regionsin any group whereas phosphorylation of Akt was found to besignificantly increased in trained and CSF treated group in CA3(F(3,16)=17.94, pb0.0001) region only as compared to control whichwas found to be reversed by STZ (ICV) in comparison to trainedgroups. There was no significant change found in Akt phosphoryla-

Figure 6 Phosphorylation of (A) IRS-1 and (B) Akt protein inmembrane (P2M) fraction of CA1, DG and CA3 subregions ofhippocampus in control (CON), trained (TRA), CSF, and STZtreated groups. Phosphorylated IRS-1 and Akt were normalizedwith total IRS-1 and Akt respectively. #Significant difference fromcontrol, $significant difference from trained and CSF (# $ Pb0.05,## $$ Pb0.01); ANOVA (one-way) followed by Newman–Keuls test.

Figure 7 Phosphorylation of (A) IRS-1 and (B) Akt protein incytosolic (S2) fraction of CA1, DG and CA3 subregions ofhippocampus in control (CON), trained (TRA), CSF, and STZtreated groups. Phosphorylated IRS-1 and Akt were normalizedwith total IRS-1 and Akt respectively. #Significant difference fromcontrol, $significant difference from trained and CSF (# $ Pb0.05,## $$ Pb0.01); ANOVA (one-way) followed by Newman–Keuls test.

267Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

tion in CA1 (F(3,16)=1.554, pN0.05) and DG (F(3,16)=0.6632, pN0.05)region in any of the group (Fig. 7a and b).

As the significant changes were found only in IR/IRS-1/Aktpathway after STZ (ICV) treatment, therefore we had selectedthis pathway to study the effect of donepezil and melatonin onSTZ induced changes in different hippocampal subregions.

3.7. Effect of donepezil and melatonin on IR geneexpression by RT-PCR in STZ (ICV) induced memorydeficit

Pre and post-treatment of donepezil and melatonin did notproduce any significant change in IR gene expression in CA1

(F(4,20) =0.4027, pN0.05) and DG (F(4,20) =0.7882, pN0.05)regions whereas pre-treatment of melatonin showed significantincrease in IR gene expression in CA3 (F(4,20) =4.574, p=0.0087)region as compared to STZ treated group (Fig. 8).

3.8. Effect of donepezil and melatonin on IR proteinexpression in STZ (ICV) induced memory deficit

In P2M fraction, pre and post-treatment with donepezil andmelatonin did not show any significant change in IR proteinexpression in CA1 (F(4,20)=1.736, pN0.05) and DG (F(4,20)=0.7489, pN0.05) regions whereas in CA3 (F(4,20) = 8.117,p=0.0005) region only melatonin (pre and post-treatment)

Figure 8 IR gene expression examined by RT-PCR in CA1, DGand CA3 subregions of hippocampus. Each lane represents(1) STZ, (2) donepezil pre-treated (DON+STZ), (3) donepezilpost-treated (STZ+DON), (4) melatonin pre-treated (MEL+STZ)and (5) melatonin post-treated (STZ+MEL) groups; Lane M is a100 bp-DNA ladder indicating the size of PCR products; IR geneexpressions were normalized with actin. *Pb0.05, **Pb0.01significant difference from STZ treated group; ANOVA (one-way)followed by Newman–Keuls test.

Figure 9 Insulin receptor (IR) protein expressions in (A) mem-brane (P2M) and (B) cytosolic (S2) fraction of CA1, DG andCA3 subregions of hippocampus in STZ, donepezil pre-treated(DON+STZ), donepezil post-treated (STZ+DON), melatonin pre-treated (MEL+STZ), andmelatonin post-treated (STZ+MEL) groups.The IR protein expressions were normalized with actin. *Pb0.05,**Pb0.01 significant difference from STZ treated group; ANOVA(one-way) followed by Newman–Keuls test.

268 R. Agrawal et al.

showed significant increase in IR protein expression ascompared to STZ treated group (Fig. 9a).

In S2 fraction, pre and post-treatment of donepezil andmelatonin did not show any significant change in IR proteinexpression in CA1 (F(4,20)=1.163, pN0.05), DG (F(4,20)=0.2068,pN0.05) and CA3 (F(4,20) =0.9622, pN0.05) regions as comparedto STZ treated group (Fig. 9b).

3.9. Effect of donepezil and melatonin on moleculesof IR/IRS-1/Akt pathway in STZ (ICV) inducedmemory deficit

In P2M fraction, both pre and post-treatment with donepeziland melatonin did not show any significant change inphosphorylation of IRS-1 and Akt in CA1 (F(4,20) = 1.653,pN0.05; F(4,20) = 0.6724, pN0.05) and DG (F(4,20) = 0.1846,pN0.05; F(4,20) =2.585, pN0.05) regions whereas melatonin preand post-treatment showed the significant increase in IRS-1 andAkt phosphorylation in CA3 (F(4,20) =9.196, p=0.0002; F(4,20)=9.373, p=0.0002) region as compared to STZ treated group(Fig. 10a and b).

In S2 fraction, both pre and post-treatment of donepezil andmelatonin had not shown any significant change in phosphor-ylation of IRS-1 and Akt in CA1 (F(4,20) =0.1179, pN0.05; F(4,20)=0.8790, pN0.05) and DG (F(4,20) =0.3762, pN0.05; F(4,20) =0.1409, pN0.05) regions as compared to STZ treated group. InCA3 (F(4,20)=0.4398, pN0.05; F(4,20)=6.323, p=0.0019) region,there was no significant change was found in IRS-1 phosphor-

ylation by donepezil and melatonin pre and post-treatmentwhereas Akt phosphorylation was found to be significantlyincreased by melatonin but not by donepezil (Fig. 11a and b).

Figure 10 Phosphorylation of (A) IRS-1 and (B) Akt protein inmembrane (P2M) fraction of CA1, DG and CA3 subregions ofhippocampus in STZ, donepezil pre-treated (DON+STZ), donepezilpost-treated (STZ+DON), melatonin pre-treated (MEL+STZ) andmelatonin post-treated (STZ+MEL) groups. Phosphorylated IRS-1and Akt were normalized with total IRS-1 and Akt respectively.*Pb0.05, **Pb0.01 significant difference from STZ treated group;ANOVA (one-way) followed by Newman–Keuls test.

Figure 11 Phosphorylation of (A) IRS-1 and (B) Akt protein incytosolic (S2) fraction of CA1, DG and CA3 subregions ofhippocampus in STZ, donepezil pre-treated (DON+STZ), donepezilpost-treated (STZ+DON), melatonin pre-treated (MEL+STZ) andmelatonin post-treated (STZ+MEL) groups. Phosphorylated IRS-1and Akt were normalized with total IRS-1 and Akt respectively.*Pb0.05, **Pb0.01 significant difference from STZ treated group;ANOVA (one-way) followed by Newman–Keuls test.

269Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

3.10. Effect of LY294002 on memory functions inMorris water maze test

The lower dose (0.3 and 1 mM) (F(6,28)=8.988, pb0.0001; F(6,28)=16.92, pb0.0001) of LY294002 showed the significant decrease inlatency time from session 3 as compared to session 1 whereas with

higher dose (3 mM) (F(6,28)=12.54, pb0.0001), the significantdecrease was observed in latency time from session 5. Vehicle(F(6,28)=14.06, pb0.0001) treated group showed the significantdecrease in latency time from session 2 as compared to session 1(Fig. 12).

0

20

40

60

80

100

120

Esc

ape

late

ncy

tim

e (s

ec)

Session 1 Session 2 Session 3 Session 4 Session 5 Session 6 Session 7Vehicle LY294002 0.3mM LY294002 1mM LY294002 3mM

*

*

**

******

*

****

**

**

****

*

**

**

* * *

Figure 12 Comparison of escape latency times of vehicle (DMSO) and LY294002 (0.3, 1 and 3 mM) treated rats in Morris water mazetest. *Significant difference (*Pb0.05, **Pb0.01) in latency time from session 1; ANOVA (one-way) followed by Newman–Keuls test.

270 R. Agrawal et al.

3.11. Effect of LY294002 on Akt phosphorylation

In P2M fraction, vehicle treatment did not show any significantchange in Akt phosphorylation as compared to trained group inall the regions whereas LY294002 (3 mM) treatment showedsignificant decrease in Akt phosphorylation in CA1 (F(2,12)=50.97, pb0.0001) and CA3 (F(2,12)=30.10, pb0.0001) but not inDG (F(2,12)=1.775, pN0.05) region as compared to vehicletreated group (Fig. 13a).

In S2 fraction, no significant change was found in Aktphosphorylation in vehicle treated group as compared to trainedgroup whereas LY294002 (3 mM) treatment showed significantdecrease in Akt phosphorylation in CA1 (F(2,12)=9.320, p=0.0036),DG (F(2,12)=5.386, p=0.0214) and CA3 (F(2,12)=17.37, pb0.001)region as compared to vehicle treated group (Fig. 13b).

4. Discussion

In the present study, the involvement of IR signaling inhippocampal CA1, CA3 and DG sub regions, which containhigh level of IRs, was investigated in rats with controlmemory, impaired memory and treated with anti-dementiaand antioxidant drugs.

In control trained rats, IR mRNA expression in thehippocampal CA1 and CA3 sub regions were up regulatedindicating an increased synthesis of IR in these areas duringlearning and memory process while IR expression remainedunaffected in DG. Zhao et al. (1999) reported the traininginduced up-regulation of IR mRNA in the CA1 and DG regionsof hippocampus associated with short-term memory, how-ever, in the present study, IR expression remained unaffect-ed in DG, an important region for synaptic plasticity and longterm potentiation (LTP), which may be associated with long-term memory (Saxe et al., 2006).

IRs and its downstream molecules undergo internalizationfrom cell surface and recycle back to the surface. Therefore,changes in IR and signaling molecules were investigated inboth the membrane (P2M) and cytosolic (S2) fractions ofhippocampal areas. An increased IR protein expression inmembrane (P2M) fraction was seen in CA1 and CA3 region.Increased IR protein expression in CA3 region was accompa-

nied by the reduction in cytosolic (S2) fraction that may bedue to the translocation of IR to the synaptic membrane aftertraining. Zhao et al. (1999) showed by immunohistochemicalstudies that IR became more concentrated in the nucleus anddendrites of the CA1 neurons after water maze training.However, in the present study, there was no significantchange in cytosolic (S2) fraction of CA1. These observationsindicate that alterations in IR expression in membranefraction of CA1 and CA3 may be linked with learning andmemory.

To further understand the downstream molecularmechanism, signaling pathways of IR in both fractions(P2M and S2) of CA1, DG and CA3 hippocampal sub regionswere studied. Several molecular pathways of IR signalinghave been identified, among which IRS-1/Akt and Shc/Erkpathways in brain have been suggested for learning andmemory functions (Zhao et al., 2004). A significantincrease in IR/IRS-1/Akt pathway, including IRS-1 and Aktphosphorylation was observed after water maze training inmembrane fraction of both CA1 and CA3 regions. Increasesin Akt phosphorylation are consistent with the up-regula-tion of Akt gene expression in spatial learning reportedby Robles et al. (2003). However, Shc and Erk1/2 showeda persistent activation in only in CA3 region in bothmembrane and cytosolic fraction after learning on Morriswater maze. Zhao et al. (2004) suggested that sustainedactivation of Erk1/2 may play a role in consolidation oflong-term memory through activation of gene transcrip-tion. Since, the hippocampal sub regions are known toperform differently in learning and memory process(Daumas et al., 2005), this may be a factor for lack of asimilar pattern of changes in IR signaling in CA1 and CA3hippocampal sub regions. These observations point out theIR signaling pathways are associated with learning andmemory.

To gain better insight of IR signaling in memory functions,we extended our study to state of memory deficit in STZ (ICV)treated rats. Rats treated with sub-diabetogenic dose of STZ(ICV) showed memory impairment in Morris water maze testas after repetitive trial sessions, time taken by rats to reachthe hidden platform remained insignificantly different fromacquisition session.

Figure 13 Phosphorylation of Akt protein in (A) membrane(P2M) and (B) cytosolic (S2) fraction of CA1, DG and CA3subregions of hippocampus in trained (TRA), vehicle (DMSO) andLY294002 (3 mM) treated groups. Phosphorylated Akt wasnormalized with total Akt. *Pb0.05, **Pb0.01 significantdifference from DMSO treated group; ANOVA (one-way) followedby Newman–Keuls test.

271Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

In present study a significant decrease in IR protein inmembrane fraction as well as mRNA expression level wasfound in CA3 region of hippocampus after STZ (ICV)treatment. IRS-1 and Akt phosphorylation was found to bedecreased after STZ treatment only in CA3 region, suggesting

the involvement of CA3 region in memory deficit induced bySTZ as there were no significant changes observed in CA1 andDG regions. Phosphorylation of Shc and Erk1/2 was also foundunaltered after STZ treatment. These results indicate thatIR/IRS-1/Akt pathway is involved in STZ (ICV) inducedmemory deficit. To confirm the involvement of IR/IRS-1/Akt pathway in memory deficit, we used LY294002, aninhibitor of PI3 kinase, which showed delayed learning anddecreased levels of Akt phosphorylation in hippocampal subregions, indicating the facilitatiry role of IR/IRS-1/Aktpathway in learning process. Therefore, suppression of IR/IRS-1/Akt pathway in CA3 may be a key factor for memoryimpairment observed with STZ (ICV).

Furtherwe studied the effect of donepezil, an anti dementiadrug and melatonin, an antioxidant on IR/IRS-1/Akt signalingpathway in STZ induced memory deficit. Pre-and post-treatmentwith donepezil andmelatonin significantly protectedrats fromSTZ inducedmemorydeficit. However, thesedrugs didnot show similar effect on IR/IRS-1/Akt pathway.

STZ induced decrease in IR mRNA and protein expressionin CA3 region was reversed by melatonin but not bydonepezil. Similarly, STZ induced decrease in phosphoryla-tion of IRS-1 and Akt downstream molecules in CA3 regionwas also antagonized by melatonin, whereas donepeziltreatment in both schedules did not show any significantchange. The results indicate that antioxidant rather thancholinergic drugs may be more effective against STZ inducedalteration in IR signaling pathway in CA3 region of hippo-campus. Earlier, we have demonstrated that brain IRs weremore susceptible to oxidative stress than to cholinergicactivity (Agrawal et al., 2009).

It is known that essential information route of memorypasses from CA3 to CA1 (Liu, 2003). Since in this study duringtraining both IR/Shc/Erk and IR/IRS-1/Akt pathways in CA3and CA1 were enhanced while in STZ treated memory deficitrats only IR/IRS-1/Akt pathway in CA3 was affected and bothpathways in CA1 and IR/Shc/Erk pathway in CA3 remainsunaltered. This suggests the possibility of involvement of IR/Shc/Erk pathway as a link between CA3 to CA1 as explainedin schematic diagram given below (Fig. 14).

Thus, the study provided evidences for involvement ofhippocampal IR in learning and memory process. Theincrease in IR and its both signaling pathways in CA1 andCA3 appeared as more relevant to memory while STZ inducedmemory deficit was associated with decrease in IR and IR/IRS-1/Akt pathway only in CA3. It may be suggested thattargeting of brain IR and IR/IRS-1/Akt pathway may providenewer therapeutic approaches for memory disorders.

Role of the funding source

In house funding from Central Drug Research Institute (CDRI),Lucknow, India.

Contributors

Rahul Agrawal: study conduction, data control, statistical analysis,data interpretation, writing the manuscript.

Ethika Tyagi: study conduction, data interpretation, review ofthe manuscript.

Rakesh Shukla: data interpretation, review of the manuscript.

Learning(MWM)

CA3 CA1

IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

STZ (ICV) CA3 CA1IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

Learning(MWM)

CA3 CA1

IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

STZ (ICV) CA3 CA1IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

Learning(MWM)

CA3 CA1

IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

STZ (ICV) CA3 CA1IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

Learning(MWM)

CA3 CA1

IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

STZ (ICV) CA3 CA1IR/Shc/Erk1/2

IR/IRS-1/Akt

IR/Shc/Erk1/2

IR/IRS-1/Akt

Memory deficit

Memory

Increase Decrease No Change

Figure 14 Schematic diagram of involvement of hippocampal IR/Shc/Erk pathway in learning and memory.

272 R. Agrawal et al.

Chandishwar Nath: protocol development, study planning,analysis and writing of the paper.

All contributors have contributed and have approved the finalmanuscript.

Conflict of interest

There is no conflict of interest for any of the contributing authors.

Acknowledgements

Senior Research Fellowship (SRF) for RA and ET from Council ofScientific and Industrial Research (CSIR), New Delhi, India isgratefully acknowledged.

References

Agrawal, R., Tyagi, E., Shukla, R., Nath, C., 2009. A study of braininsulin receptors, AChE activity and oxidative stress in ratmodel of ICV STZ induced dementia. Neuropharmacology 56,779–787.

Atkins, C.M., Selcher, J.C., Petraitis, J.J., Trzaskos, J.M., Sweatt, J.D.,1998. The MAPK cascade is required for mammalian associativelearning. Nat. Neurosci. 1, 602–609.

Daumas, S., Halley, H., Francés, B., Lassalle, J.M., 2005. Encoding,consolidation, and retrieval of contextual memory: differentialinvolvement of dorsal CA3 and CA1 hippocampal subregions.Learn. Mem. 12, 375–382.

Dou, J.T., Chen, M., Dufour, F., Alkon, D.L., Zhao, W.Q., 2005.Insulin receptor signaling in long-term memory consolidationfollowing spatial learning. Learn. Mem. 12, 646–655.

Frolich, L., Blum-Degen, D., Bernstein, H.G., Engelsberger, S.,Humrich, J., Laufer, S., Muschner, D., Thalheimer, A., Turk, A.,Hoyer, S., Zochling, R., Boissl, K.W., Jellinger, K., Riederer, P.,1998. Brain insulin and insulin receptors in aging and sporadicAlzheimer's disease. J. Neural Transm. 105, 423–428.

Grünblatt, E., Salkovic-Petrisic, M., Osmanovic, J., Riederer, P.,Hoyer, S., 2007. Brain insulin system dysfunction in streptozo-

tocin intracerebroventricularly treated rats generates hyper-phosphorylated tau protein. J. Neurochem. 101, 757–770.

Izquierdo, L.A., Barros, D.M., Medina, J.H., Izquierdo, I., 2000.Novelty enhances retrieval of one-trial avoidance learning in rat1 or 31 days after training unless the hippocampus is inactivatedby different receptor antagonists and enzyme inhibitors. Behav.Brain Res. 117, 215–220.

Lannert, H., Hoyer, S., 1998. Intracerebroventricular administrationof streptozotocin causes long-term diminutions in learning andmemory abilities and in cerebral energy metabolism in adult rats.Behav. Neurosci. 112, 1199–1208.

Lee, I., Kesner, R.P., 2004. Differential contributions of dorsalhippocampal subregions to memory acquisition and retrieval incontextual fear-conditioning. Hippocampus 14, 301–310.

Lein, E.S., Zhao, X., Gage, F.H., 2004. Defining a molecular atlas ofthe hippocampus using dna microarrays and high-throughput insitu hybridization. J. Neurosci. 24, 3879–3889.

Liu, Y., 2003. Electrophysiology experiment on neuron networkdynamical system analysis and memory decoding. Phys. Exp. Proj.1–14.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951.Protein measurement with the Folin phenol reagent. J. Biol.Chem. 193, 265–275.

Morris, R., 1984. Development of water maze procedure for studyingspatial learning in the rat. J. Neurosci. Meth. 11, 47–60.

Plaschke, K., Hoyer, S., 1993. Action of the diabetogenic drugstreptozotocin on glycolytic and glycogenolytic metabolism in adultrat brain cortex and hippocampus. Int. J. Dev. Neurosci. 11,477–483.

Robles, Y., Vivas-Mejia, P.E., Ortiz-Zuazaga, H.G., Felix, J., Ramos, X.,Pena de Ortiz, S., 2003. Hippocampal gene expression profiling inspatial discrimination learning. Neurobiol. Learn. Mem. 80, 80–95.

Saxe, M.D., Battaglia, F., Wang, J.W., Malleret, G., David, D.J.,Monckton, J.E., Garcia, A.D.R., Sofroniew, M.V., Kandel, E.R.,Santarelli, L., Hen, R., Drew, M.R., 2006. Ablation of hippocampalneurogenesis impairs contextual fear conditioning and synapticplasticity in the dentate gyrus. Proc. Natl Acad. Sci. USA 103,17501–17506.

Schafe, G.E., Nadel, N.V., Sullivan, G.M., Harris, A., LeDoux, J.E.,1999. Memory consolidation for contextual and auditory fearconditioning is dependent on protein synthesis, PKA, and MAPkinase. Learn. Mem. 6, 97–110.

273Insulin receptor signaling in rat hippocampus: A study in STZ (ICV) induced memory deficit model

Selcher, J.C., Atkins, C.M., Trzaskos, J.M., Paylor, R., Sweatt, J.D.,1999. A necessity for MAP kinase activation in mammalian spatiallearning. Learn. Mem. 6, 478–490.

Takahashi, M., Yamada, T., Tooyama, I., Moroo, I., Kimura, H.,Yamamoto, T., Okada, H., 1996. Insulin receptor mRNA in thesubstantia nigra in Parkinson's disease. Neurosci. Lett. 204, 201–204.

Zhao, W., Chen, H., Xu, H., Moore, E., Meirii, N., Quon, M.J., Alkon,D.L., 1999. Brain insulin receptors and spatial memory. J. Biol.Chem. 274, 34893–34902.

Zhao, W.Q., Chen, H., Quon, M.J., Alkon, D.L., 2004. Insulin andinsulin receptor in experimental models of learning and memory.Eur. J. Pharmacol. 490, 71–81.