Acetylcholinesterase Modulates Stress-Induced Motor Responses Through Catalytic and Noncatalytic Properties

AMNE

BoseMetRsrttsCf

Kc

Sm(Migsbinou“hg

n“2pwnm

F

A

R

0d

cetylcholinesterase Modulates Stress-Inducedotor Responses Through Catalytic andoncatalytic Properties

lla H. Sklan, Amit Berson, Klara R. Birikh, Amos Gutnick, Or Shahar, Shai Shoham, and Hermona Soreq

ackground: Cholinergic neurotransmission notably participates in stress-induced motor responses. Here we report the contributionf alternative splicing of acetylcholinesterase (AChE) pre-mRNA to modulate these responses. More specifically, we inducedtress-associated hypofunction of dopaminergic, mainly D2 dopamine receptor–mediated neurotransmission by haloperidol andxplored stress induced hyperlocomotion and catalepsy, an extreme form of immobility, induced in mice with AChE deficiencies.ethods: Conditional transgenic (Tet/AS) mice were created with tetracycline-induced antisense suppression of AChE gene

xpression. Locomotion and catalepsy times were measured in Tet/AS and strain-matched control mice, under open-field exposurehreat and under home-cage safety.esults: In vitro, NGF-treated PC12 cells failed to extend neurites upon Tet/AS suppression. In vivo, Tet/AS but not control mice showed

tress-associated hippocampal deposits of heat-shock protein 70 and GRP78 (BiP), predicting posttranscriptional changes in neuronaleactions. Supporting this notion, their striatal cholinergic neurons demonstrated facilitated capacity for neurite extension, attributinghese in vivo changes in neurite extension to network interactions. Tet/AS mice presented stress-induced hyperlocomotion. Moreover,he dopamine antagonist haloperidol induced longer catalepsy in threatened Tet/AS than in control mice. When returned to home-cageafety, Tet/AS mice showed retarded release from catalepsy.onclusions: Acetylcholinesterase modulates stress-induced motor responses and facilitates resumption of normal motor behavior

ollowing stress through both catalytic and noncatalytic features.

ey Words: AChE, antisense, catalepsy, neuroleptics, stress, tetra-ycline-controlled transgenics

tress stimuli elicit motor response patterns that are mostadequate to handle specific types of stresses (Eilam 2005;Graeff 1994). Under threat stress, these response patterns

ay take an extreme form such as freezing, running, or fightinge.g., when a predator threat is proximal; Eilam 2005; Gray andcNaughton 2000). Alternatively, moderation reactions may

nvolve careful risk assessment, prior learning, and experience-uided motor behavior (Gray and McNaughton 2000). When thetressful situation is over, normal motor behavior patterns shoulde retrieved. Excessive intensity of emotional reactions maynitiate nonadaptive motor responses and impair such return toormal motor patterns, yielding stress-associated psychopathol-gies. Exploring brain mechanisms regulating motor behaviornder stress, which has been referred to collectively as theemotional–motor” interface (Phelps and LeDoux 2005), canence shed new light on these stress-induced psychopatholo-ies.

The emotional–motor interface spans several levels in theervous system. It may involve subcortical circuitries that engagequick and dirty” motor response patterns (Keay and Bandler001; Phelps and LeDoux 2005) or activate cortical and hip-ocampal circuitries involving the striatum (Haber et al 2000),hich allow moderated response patterns and the return toormal motor behavior when stress is over. Several neurotrans-itters likely participate in this process and thus contribute to the

rom the Department of Biological Chemistry (EHS, AB, KRB , AG, OS, HS),Institute of Life Sciences, Hebrew University of Jerusalem; and ResearchDepartment (SS), Herzog Memorial Hospital, Jerusalem, Israel.

ddress reprint requests to Hermona Soreq, Ph.D., Department of BiologicalChemistry, Institute of Life Sciences, Hebrew University of Jerusalem,Givat Ram, Jerusalem, 91904, Israel; E-mail: [email protected].

eceived August 3, 2005; revised March 23, 2006; accepted March 23, 2006.

006-3223/06/$32.00oi:10.1016/j.biopsych.2006.03.080

selection of motor response patterns under stress. For example,Neuropeptide Y moderates the endocrine and motor compo-nents of stress reactions through incompletely understood mech-anisms (Heilig 2004). The contributions of specific neurotrans-mitters to the emotional–motor interface under stress werehistorically extended from investigations of the hypothalamic–pituitary axis (HPA). Thus, corticotropin-releasing factor (CRF)increases locomotor behavior in a familiar test environment butsuppresses it in a novel test environment (Heilig et al 1994). Thiscontext-dependent selection of motor response patterns parallelsCRF activation of multiple brain regions (Imaki et al 1993), whichis refractory to the removal of the pituitary gland, indicating aprimary but only partially explained organizational role of CRF instress-elicited motor responses. Importantly, forebrain cholin-ergic neurons carrying CRF receptors project to cortical andhippocampal regions (Sauvage and Steckler 2001) and intraven-tricular CRF administration increases acetylcholine (ACh) releasefrom cholinergic terminals in these regions (Day et al 1998).Moreover, increased ACh release in cortex and hippocampuspotentiates the responsiveness of nicotinic receptors on principalneurons in these brain regions to incoming messages (Cox et al1994), attributing a “stress-arousal” capacity to the cholinergiccontribution to the emotional–motor interface. Other classicalstress-associated neurotransmitters, such as norepinephrine, alsoinduce ACh release at cortical and hippocampal brain regions(Vizi 1980), presenting the cholinergic contribution to “stress-arousal” as a ubiquitous component of these pathways. Hip-pocampal cholinergic neurotransmission was shown to be im-portant in fear-induced immobility (Miyakawa et al 2001;Overstreet et al 1986; Power and McGaugh 2002; Takahashi andGoh 1996); however, it remained unknown whether cholinergicneurotransmission might also moderate the intensity of motorreactions under stress and enable switching back to normalmotor behavior when the stressful situation is over.

The ACh hydrolyzing enzyme acetylcholinesterase (AChE) maymodulate both excitatory and pacifying influences on stress re-

BIOL PSYCHIATRY 2006;60:741–751© 2006 Society of Biological Psychiatry

pTraccrmtascmppsse2amhtRiftea

euotpsAo

diita(sctC1u(t1dr1eHiprt

742 BIOL PSYCHIATRY 2006;60:741–751 E.H. Sklan et al

w

re-mRNA (Meshorer and Soreq 2006; Soreq and Seidman 2001).wo distinct 3= splice variants of AChE mRNA may be potentiallyelevant to the emotional modulation of motor reactions. Theselternative transcripts all encode a 543 amino-acids long commonore domain through exons 2, 3, and 4 (the “classical” exon 1ontains a 5= untranslated region of AChE mRNA) but differ in theegions encoding the C-terminus (Soreq and Seidman 2001). Theain transcript found in the brain and in neuromuscular junctions is

he “synaptic” form (AChE-S, also called T for “tailed”; Massoulié etl 2005). AChE-S potentially modulates cholinergic neurotransmis-ion in all of the cholinergic pathways relevant to the emotionalontribution to motor activity. “Read-through” AChE (AChE-RRNA) is a splice variant retaining “pseudointron” number 4 in therocessed transcript that modifies the C terminus of its proteinroduct. AChE-R mRNA is produced in response to psychologictress and drug-induced AChE inhibition (Kaufer et al 1998; Me-horer et al 2002; Perrier et al 2005). The prominent stress-inducedxpression of AChE-R in cortex and hippocampus (Cohen et al002) and its expression in amygdala (Sternfeld et al 2000) suggestunique contribution to mechanisms of emotional modulation ofotor activity. Supporting this notion, transgenic overexpression ofuman AChE-R potentiates stress-induced behavioral inhibitionhrough intracellular complex formation with the scaffold proteinACK1 and protein kinase C beta2 (Birikh et al 2003). Furthermore,ntracellular complex formation of AChE-R with these partnersacilitates hippocampal LTP and accentuates contextual fear condi-ioning (Nijholt et al 2004). In addition, AChE-R overexpressionxerts neuritogenic actions that attenuate hippocampal age-associ-ted neurodeterioration (Sternfeld et al 2000).

To explore the stress-moderating capacity of AChE-R, wengineered a tetracycline-controlled antisense system that atten-ates AChE overproduction at will, interfering with the capacityf the mouse ACHE gene to respond to stressful stimuli byranscriptional activation and shifted alternative splicing. We em-loyed this system to determine the functional consequences ofuch attenuation on the morphologic and neuritogenic effects ofChE-R in cultured cells and on the behavioral stress reactionsf engineered mice in vivo.

Importantly, the behavioral paradigm developed was especiallyesigned to probe a specific aspect of the striatal emotional–motornterface (Haber et al 2000), namely, cholinergic–dopaminergicnteractions (Calabresi et al 2000; Kawaguchi et al 1995). Thus,hreat stress typically results in immobility. In the striatum, itssociates with both reduced dopamine release in the striatumKatoh et al 1996) and increased immediate early gene expres-ion (Perrotti et al 2004). Chronic administration of CRF similarlyauses reduced dopamine release in the striatum and increaseshe propensity to haloperidol-induced catalepsy (Izzo et al 2005).atalepsy, which is an extreme form of immobility (De Ryck et al980), is sensitive to stress (Antelman et al 1992) and is contin-ous with the fear-induced freezing immobility responseBarykina et al 2004). Striatal cholinergic neurons are critical forhe manifestation of haloperidol-induced catalepsy (Dains et al996; Klemm 1985), reflecting the fact that reduced striatalopamine neurotransmission enhances striatal cholinergic neu-otransmission (Calabresi et al 2000; DeBoer and Abercrombie996). Dopamine D2 receptors have been linked to the motorffects of stress (Cabib et al 1998; Horger and Roth 1996).aloperidol, with a high ratio of D2/D1 antagonism induces

mmediate early gene expression with a distribution that highlyarallels that of stress (Perrotti et al 2004). Varying the level ofisk–safety by comparing open-field versus home-cage condi-

ww.sobp.org/journal

interactions in the emotional–motor interface and its moderationby AChE. Specifically, we tested whether attenuating the stress-induced AChE production may remove a moderating influence,resulting in exacerbation of catalepsy under threat and retarda-tion of the return to normal motor patterns under safety.

Methods and Materials

Teracycline-Controlled Transgenic MiceThe laboratory animal experiments, approved by the Hebrew

University’s Committee for Animal Experimentation, involvedinjecting the XhoI-HindIII DNA fragment from the pTRE-ASplasmid or the ScaI-XhoI fragment of pTet-ON (Clontech, Moun-tain View, California) into the pronuclei of fertilized eggs of CB6inbred mice, according to standard procedures. The tetracycline-controlled AS-AChE (pTRE-AS) construct was obtained by re-cloning nt 1728-1832 of mouse AChE DNA (exon 6, GeneBankaccession No.NM_009599) into the EcoRI site of the pTREplasmid. Two lines of transgenic mice were generated perconstruct (four lines all together). F1 mice carrying the TetONtrans-activator gene were mated with transgenic F1 mice fromone of the AS lines to create composite transgenic F2 mice, onwhich experiments were preformed.

Drug AdministrationDoxycyline hydrochloride (2 mg/mL, Sigma, St. Louis, Missouri),

dissolved in 5% sucrose, was supplied in drinking water,changed every third day for 6 consecutive days. Bacterial lipo-polysaccharide (LPS; E. coli O55, Disco Labs, Detroit, Michigan),dissolved in saline, was administered intraperitoneally (IP; 50�g/kg). Haloperidol (Janssen Pharmaceuticals, Beerse, Belgium,5 mg/mL) was diluted in saline and injected IP in a volume of.1 mL. The AChE catalytic activity was determined as detailedpreviously (Sklan et al 2004).

Measuring Locomotion and CatalepsySeveral stress paradigms were considered, including restraint

stress or placing mice in a water bath with no exit option; however,restraint stress is not amenable for measuring the impact on motorresponse patterns because the subject is not free to move. Likewise,swimming motor patterns do not allow observations of locomotionand freezing behavior that are both relevant to the normal repertoireof behavior in rodents. Also, when a subject is released fromrestraint or removed from a water bath the motor activity observedis already not “under stress.” Therefore, we selected the paradigm of“acute” exposure to a brightly lit novel open field, a situation thatinvolves threat (McNaughton and Gray 2000) and in which motoractivity can be manifested in its entire range, from immobility tohyperlocomotion.

All behavioral tests were videotaped for subsequent quantifi-cation. Two-month-old binary Tet/AS transgenic and CB6 femalecontrol mice were injected IP with saline (.9%, .1 mL). Twentyminutes postinjection, the mice were placed in a black squareacrylic 100 � 100 � 50 cm open field and allowed 5 min toexplore, after which the time spent in locomotion was measured.The next day, the mice were injected with haloperidol, and thisprocedure was repeated, with locomotion patterns video moni-tored. To measure catalepsy, we placed the mice with theirforepaws leaning on an elevated platform and with their hindlegs on the open-field floor. Catalepsy was defined as remainingin this posture, and its duration was measured (catalepsy time).For data management, we used a two-factor analysis of variancesubjected to post hoc analyses offered by the Statistica software

S

as2(rFu1ow(gTiac(d

N

abIftapf

R

TC

acAcpps

T

B

L

C

C

s ted inH

E.H. Sklan et al BIOL PSYCHIATRY 2006;60:741–751 743

taining ProceduresGRP78/GFAP (glucose-regulated protein 78/antiglial fibrillary

cidic protein) staining solutions contained .001% soybean tryp-in inhibitor (Sigma type IIS), in .3% Triton X-100, .05% Tween-0, 2% normal goat serum and 2% normal donkey serumstandard solution), mouse GFAP (Sigma, Israel; 1:100), andabbit anti-GRP78 (SPA-826, Stressgene, Victoria, Canada, 1:100).or iHSP70 (inducible heatshock protein 70)/GFAP staining, wesed a similar Solution St with mouse-anti-GFAP (Sigma), diluted:100, and rabbit anti-iHSP70 (SPA-812, Stressgene; 1:100). Sec-ndary antibody solutions involved incubation in Solution Stith donkey serum, donkey antirabbit conjugated to fluorescein

AP182F, Chemicon, Temecula, California) diluted 1:100, andoat antimouse conjugated with tetra-methyl-rhodamine (Sigma,7782) diluted 1:800. Choline acetyltransferase (ChAT) staining

nvolved goat anti-ChAT diluted 1:200 and biotinylated donkeyntigoat diluted 1:400, both antibodies purchased from Chemi-on. Nicotinamide adenine dinucleotide phosphate diaphoraseNADPH) activity was histochemically detected as previouslyescribed (Kawaguchi et al 1995).

eurite MeasurementsWe measured neurites derived from both cultured cells and

ttained neurons in brain sections. Length from the cell bodyorders to the terminal end of the neurites was measured usingmage-Pro software (Media Cybernetics, Silver Spring, Maryland)or all processes in 15 neurons per section or 50 per culture. Theotal length of processes per neuron, the maximal process length,nd the slope of the best-fit linear regression enabled us toredict the average length of the processes per neuron as aunction of the number of processes involved, using Statistica.

esults

et-Mediated AS Suppression of AChE Expression in Culturedell Lines

To manipulate AChE gene expression transiently, we clonedn antisense (AS) DNA fragment encoding 100 nt of AChEcRNAomplementary to the evolutionarily conserved exon 6 of rodentChE mRNA (Figure 1A) into a tetracycline-responding DNAonstruct (Tet/AS). The Tet/AS sequence we employed is ex-ressed in the nucleus and targeted toward exon 6 of there-AChE mRNA transcript, destroying much of the newly tran-

able 1. Transgene Effects on Motor Behavior

ehavior Factor

ocomotion, Open Field, dosage2 mg/kg

Tet/AS F(1Dox F(1Tet/AS � Dox F(1Halop. F(1Halop. � Tet/AS F(1Halop. � Dox F(1Halop. � Tet/AS � Dox F(1

atalepsy Leaning Time, HomeCage, dosage 2 mg/kg

Tet/AS F(1Dox F(1Tet/AS � Dox F(1

atalepsy Leaning Time, HomeCage, dosage 2.5 mg/kg

Tet/AS F(1Dox F(1Tet/AS � Dox F(1

Results of analysis of variance with two factors (Tet/AS transgene and Dignificance. Specific post hoc Newman-Keuls (NK) comparisons are presenalop, haloperidol; ns, nonsignificant.

cribed sequences. In cultured PC12 cells or in live mice under

stress, shifted alternative splicing yielded preferred destructionby this Tet/AS construct of the AChE-RmRNA transcript, provid-ing a conditional tool for testing the physiologic significance ofthis cholinergic feedback response to stress.

First, we tested the effects of this transgene on AChE geneexpression. To this end, the AS or empty vehicle was cotrans-fected with CMV-controlled AChE-R or AChE-S expression con-structs (Sternfeld et al 1998) to the CHO-TetOff cells (Figure 1B;Gossen and Bujard 1992), which do not express endogenousAChE. The AS sequence used was previously shown to suppressAChE expression in the rat-derived PC12 cells (Grifman et al1998). Compatible with this capacity, the Tet/AS vector, but notthe ineffective “empty” vehicle, reduced AChE activity in CHO-TetOff cells overexpressing AChE-R by 40% (Fig 1C). Addingdoxycycline (Dox, a tetracycline derivative) restored AChE-Rexpression to full-range levels, thus demonstrating Dox depen-dence (Fig 1C, left). In contrast, lowering the AS:AChE-R ratio didnot change the extent of inhibition but demonstrated that Tetexerted a significant contribution to AChE gene expression.

To test for potential “leakiness” of the Tet/AS effect, we stablytransfected PC12-TetON cells (Clontech, Figure1B) with theAS-expressing construct. Constitutive suppression of the AChEactivity of PC12 cells (by ca. 60%) reflected promiscuous “leak-age” of the tetracycline control system, but adding Dox furtherreduced cellular AChE levels to about 10%–20% of the parentcells (Figure 1C, right, Student’s t test, p � .01). Immunoblotanalysis confirmed that AChE protein levels were predictablylower (Figure 1D); however, this was independent of addingDox, once more reflecting leakage of the system. Most pro-nounced was the decrease in AChE-R-specific immunolabeling(to 44% of the PC12 control in the more efficient line AS2),indicating a splice variant preference for the Tet/AS effect. Thispreference was compatible with our previous findings (Grifmanet al 1998), thus supporting our results.

AChE-R Suppression Impairs NGF-Induced Neuritogenesisin Tet-AS PC12 Cells

Previous studies attributed an evolutionarily conserved non-enzymatic role to AChE in neuritogenesis (Dori et al 2005;Grifman et al 1998; Layer and Willbold 1995; Sharma et al 2001).We therefore examined nerve growth factor (NGF)-mediatedneurite extension in the Dox-treated Tet/AS and untreated

F Value Post Hoc NK Tests, p � .05

10.43; p � .0039 Tet/AS � CB615.48; p � .0007 Dox � no-Dox5.49; p � .0286 Tg � Dox � CB6 � Dox95.97; p � .0000 Halop. � Saline8.86; p � .0070 Greater reduction from baseline in Tet/AS mice.6.13; p � .0215 Greater reduction from baseline under Dox.9.28; p � .0059 Due to Tet/AS � Dox baseline5.38; p � .0281 Tet/AS � CB67.36; p � .0115 Dox � no-Dox5.38; p � .0281 Tet/AS � Dox � CB6 � Dox10.09; p � .0055 Tet/AS � CB60.61; p � .4458 ns1.02; p � .3258 ns

e presented as the F ratio with degrees of freedom followed by the level ofwhich a significant difference is accepted at the 0.05 level of significance.

,22) �,22) �,22) �,22) �,22) �,22) �,22) �,27) �,27) �,27) �,17) �,17) �,17) �

ox) ar

control PC12 cells, lacking synaptic ACh neurotransmission

www.sobp.org/journal

(Pitpwct(cwias

FsaTTamtPmpewvTrtA

744 BIOL PSYCHIATRY 2006;60:741–751 E.H. Sklan et al

w

Figure 2). Nondifferentiated Tet/AS cells maintained a typicalC12 morphology, suggesting that the constitutive “leakiness”-nduced reduction in their AChE activity did not appreciably affectheir basic cytoacrchitecture. When grown in the presence of NGF,arent PC12 cells displayed a neurite extension that was facilitatedhen Dox was administered, extending neurites 65 � 8 �m long,ompared with 46 � 2 �m in parent line cells (Figure 2A), whereashe average width of neurites was reduced from 7–8 to 5–6 �mFigure 2B, p � .01, Student’s t test). In Tet/AS Dox-treated PC12ells, however, average neurite length, but not width, decreasedithin 3 days in culture to 25 � 5 �m (Figure 2A and 2D),

ndicating that the Dox-dependent suppression of neuritogenicctivity (Figure 2E) might be associated with AChE-R suppres-

igure 1. Designing and testing the AS sequence. (A) The target AS site:hown are nucleotide numbers (GeneBank No. NM_009599) in human (H)nd rat/mouse (R/M) acetylcholinesterase (AChE) splice variants. Bottom:arget AChE-S mRNA includes two rat–mouse mismatches (gray). (B) Theet system: a tetracycline-responsive element (TRE) contains 7 tetO copiesnd the minimal cytomegalovirus promoter (PminCMV). The pTetOff plas-id expresses tTA, a fusion of TetR to the VP16 herpes simplex virus activa-

ion domain. tTA binds the tetO sequences, activating transcription fromhCMV*-1 in the absence of Tet and turns it off by Tet in a dose-dependentanner. The “reverse” rTetR TetOn system activates transcription in the

resence of doxycycline (Dox). (C) AS represents the inhibition of AChExpression in cultured cells. CHO-TetOff cells were transiently cotransfectedith the AS sequence of 1A and with AChE-S (S) or AChE-R (R) encoding or

ehicle (TRE) plasmids in the indicated ratios (left). Two independent PC12-etON cell lines were stably transfected with the AS construct (AS1, AS2,ight). (D) Immunoblot analysis of cell extracts with anti-N-ter, reacting withhe AChE N-terminus common to all variants or anti-ARP, specific for theChE-R C-terminus. Exp: expressed Tet/AS.

ion.

ww.sobp.org/journal

Inducible In Vivo Suppression of AChE ActivitySuppression of AChE’s hydrolytic activity was first tested in

naïve binary transgenic mice expressing the TetON regulator andthe AS sequence, both under CMV promoter control (Tet/AS mice,n � 6). Enzyme activity was markedly diminished in the cortexfrom 89% � 2 to 48% � 7 in the intestine and from 70.5% � 2.7to 46.6% � 3.6, compared with the parallel tissues of the parentCB6 strain mice given 2 mg/mL Dox in their drinking water for6 consecutive days (Figure 3A, compare white bars to black bars,n � 9). This reflected an in vivo leakage of the tetracyclinecontrol. Next, we tested the cortical and intestinal AChE activityin the binary transgenic animals (n � 6) given the same Doxtreatment. Activity decreased even more, down to ca. 40% of theparent strain activity (Figure 3A,B, p � .05, Student’s t test),reflecting tetracycline-induced suppression.

Injecting bacterial LPS (50 �g/kg) induces endotoxic stress,which consequently elevates plasma AChE-R levels (Cohen et al2003). In LPS-injected Tet/AS mice, we observed no AChE-R in

Figure 2. Nerve growth factor (NGF)-induced differentiation of parent andTet/AS cell lines. (A) In the presence of NGF, parent PC12 cells displayedneurite extension that was facilitated under doxycycline (Dox). (B) Doxtreatment reduced neurite width. (C) Tet/AS expression abrogated the NGF-induced elongation process and caused neurite shortening. (D) Tet/AS didnot change the process width. Insets: Camera lucida drawings of control andtreated cells. (E) Schematic drawings of the summarized results. p values

peaA

TP

ahTmpc(ptd(tcTf

CN

m

FibdfisPfaLT(

E.H. Sklan et al BIOL PSYCHIATRY 2006;60:741–751 745

lasma, unlike Dox-treated CB6 control or transgenic mice withxcess AChE-R (Figure 3C and 3D). This was compatible with thessumption that the effects of the Tet/AS transgene attenuateChE-R overproduction in animals subjected to stressful stimuli.

et/AS Suppression Facilitates the Accumulation of Stressroteins in the Mouse Brain

FVB/N transgenic mice constitutively overexpressing AChE-Rre protected from aging-associated neuronal accumulation ofeat-shock proteins (Sternfeld et al 2000). This suggested thatet/AS mice, with impaired AChE-R overproduction, would beore susceptible than their parent strain mice for stress neuro-athology. To challenge this prediction, we immunolabeledlusters of the GRP78 (BiP), a hallmark of endoplasmic reticulumER) stress responses (Kaufman 1999) and iHSP70 in the hippocam-al striatum radiatum and in the upper layer of the olfactoryubercle–piriform cortex. BiP and iHSP70 clusters were embed-ed in astrocyte-enriched layers in both wildtype control miceFigure 4A) and Tet/AS mice (Figure 4B). Interestingly, the size ofhese clusters was larger in Tet/AS (Figure 4D) mice than in CB6ontrols (Figure 4C) and their numbers were much larger in theet/AS piriform cortex (Figure 4, compare E–F), U � 6, p � .05or iHSP70 and U � 3, p � .03 for BiP, Mann–Whitney U Test).

holinergic Neurons in Tet/AS Mice Display Facilitatedeurite Extension

The transgene-induced neuropathology suggested putative

igure 3. Inducible in vivo suppression. Acetylcholinesterase (AChE) activityn the cortex (A) and intestine (B) of transgenic Tet/AS or parent CB6 miceefore or after administering doxycycline (Dox). (C) The experimental para-igm: lipopolysaccharide (LPS) induces AChE overexpression. Dox intensi-es Tet/AS expression, which destroys part of the AChE pre-mRNA tran-cripts, thus reducing AChE activities. (D) Prevention of the LPS response.lasma samples from control CB6, TgR mice overexpressing AChE-R (Stern-

eld et al 2000), and Tet/AS mice were separated by a nondenaturing polycrylamide gel electrophoresis after administering Dox or after exposure toPS (or both). TgR mice overexpressing AChE-R served as a positive control.he migration distances of the AChE tetramers (gray, AChE-S), dimers

white, AChE-E), and monomers (black, AChE-R) are indicated with arrows.

odulation of neuronal plasticity. Also, apart from its catalytic

function, AChE-R was shown to exert morphogenic, nonenzy-matic activity (reviewed by Soreq and Seidman 2001). To test thefunctional consequences of reducing AChE-R in the stress-susceptible brain areas of Tet/AS mice, we stained cholinergicand nitric oxide (NO) interneurons in straital sections from sex-and age-matched Dox-treated Tet/AS mice and CB6 controls(Figure 5A). Within the striatum of Tet/AS and CB6 mice, the totalneurite length (55–60 �m) and the number of processes/cell(2–2.5) were similar for the two strains and cell types. Interest-ingly, the maximal process length was somewhat shorter forcholinergic neurons (from 144 �m in CB6 to 111 �m in Tet/ASmice) but longer for NO neurons in Tet/AS mice (129 �m vs.159 �m), suggesting that this effect was selective. In both cases,the average neurite length was directly correlated with thenumber of neurites per cell. Neurite extension is thought toreflect a response to interneuronal contacts formed with preex-isting processes (Valtorta and Leon 1999). In view of the neuri-togenic properties of ACh (Zheng et al 1994), we tested thepossibility that cholinergic interneurons in the Tet/AS striatumrespond more readily than control neurons to neuritogenicinputs. Indeed, a linear regression analysis predicting the totallength of processes as a function of the number of processes perneuron, revealed that cholinergic interneurons in Tet/AS mice

Figure 4. Acetylcholinesterase (AChE) suppression increases stress proteinclusters. Tet/AS and CB6 brain sections were immunostained with antibodiesagainst iHSP70 (inducible heatshock protein) and GRP78 (glucose-regulatedprotein (red). Counterstaining was with GFAP (green). (A, B) GRP78 clusters inthe hippocampal CA1 region of doxycycline (Dox)-treated CB6 (top) and Tet/ASmice (bottom). (E, F) Cluster numbers in 100 �m2 areas within the hippocam-pus CA1 and piriform cortex of Dox-treated CB6 and Tet/AS mice (n � 6 micefrom each strain, five areas from each section). Asterisks denote statistically

www.sobp.org/journal

eiSm

afvpbwtt

H

Tn

Flflssp(nttiteararwbc

746 BIOL PSYCHIATRY 2006;60:741–751 E.H. Sklan et al

w

xtend even longer processes than those of CB6 controls with anncrease in the number of preexisting processes [two-sided,tudent’s t test comparison t (df � 10) � 2.6, p � .05, a nonpara-etric test, Mann–Whitney U test, p � .05, Figure 5B and 5C] .The fact that Tet/AS cholinergic interneurons maintained

pparently normal neurite lengths but displayed modified plasticeatures suggested that AChE–ACh relationships might be in-olved (Figure 5D). AChE expressed in cholinergic neuronsromotes neurite growth by virtue of its morphogenic featuresut also degrades ACh through its enzymatic activity. Neuronsith depleted AChE-R would hence promote growth less effec-

ively but maintain their neurite length by reacting more readilyo ACh neurotransmission than neurons of matched control mice.

yperlocomotion Effects of Tet/AS SuppressionTo test for potential changes in the adaptive response of

et/AS mice to stress stimuli, we injected these mice intraperito-eally with saline, which activates stress-responsive brain re-

igure 6. Induction of excess locomotion and cata-epsy in stressed Tet/AS mice. (A) The experimentalow chart, see text for details. (B) Locomotion mea-ured in the CB6 control and Tet/AS mice followingaline (1) and/or haloperidol injections (2); n � 8 miceer group. (C) Catalepsy observed in a novel open field

1) and in the home cage after the mice received theoted amounts of haloperidol (2). Asterisks note statis-

ically significant differences (values shown in paren-heses). (D) The conceptual prediction. Injection stressnduces acetylcholinesterase (AChE) overexpressionhat is largely prevented by the Tet/AS system (1). Thislevates acetylcholine levels, creating a Da/ACh imbal-nce (2). Haloperidol (inset) blocks the D2 dopamineeceptors (3), consequently exacerbating this imbal-nce and inducing catalepsy (4). (E) Video cinematog-aphy (see Supplement 1) of the cataleptic response inildtype CB6 and Tet/AS mice in an open field (top) orack at the home cage (bottom). The clocks show totalatalepsy time (red).

ww.sobp.org/journal

gions similar to those activated by immobilization, footshock, orforced-swim stresses and induces neuronal AChE overproduc-tion (Birikh et al 2003). Owing to the impaired production ofAChE in the Tet/AS mice, high ACh levels should be maintainedunder stress because of suppressed ACh hydrolysis. Therefore,we postulated that a partial blockade of this stress responsewould take place in Dox-untreated Tet/AS mice because of theleakage of the TetOn system, that greater prevention would takeplace in Dox-treated Tet/AS mice and that such preventionwould disrupt the Da–ACh balance.

First, we tested the transgene effects on stress-induced loco-motion. Injecting saline prolonged the locomotion time to aboutthe same extent in CB6 and uninduced Tet/AS mice. Thisindicated that the limited constitutive suppression created by theTetON leakage did not affect stress-induced locomotion. Trans-gene activation by administering Dox doubled the injection-induced locomotion time of Tet/AS but not CB6 mice (Figure 6 Aand 6B1; Table 1), thus demonstrating a marked interaction

Figure 5. Subtle morphologic Tet/AS-inducedchanges in striatal cholinergic neurons. (A) Represen-tative striatal interneurons from sex- and age-matchedTet/AS mice and CB6 controls stained with the notedantibodies. (B) Mean and maximal number of pro-cesses and the cumulative length of processes per cellfor each cell population. (C) Linear regression analysisshows a modified association of cumulative neuritelength with the total number of neurites in striatal in-terneurons of Tet/AS and CB6 mice (15 cells per sec-tion, p � .05, Student’s t test). (D) The predicted mech-anism. Acetylcholinesterase (AChE) promotes neuriteextension through noncatalytic features, whereas itsenzyme activity limits the incoming acetylcholine(ACh) gradient, consequently suppressing its capacityto exert neurite growth. In Tet/AS mice, AChE suppres-sion decreases the nonenzymatic component of neu-ritogenesis, but the decreased AChE hydrolytic activityenhances the ACh-induced neuritogenic effect.

bswl

i[ir9dToso

Ti

tRceits

ltpfmuohop7tttDe

2omct6sS[tewr(

D

tl

E.H. Sklan et al BIOL PSYCHIATRY 2006;60:741–751 747

etween Dox and the transgene [F (1,22) � 5.49, p � .03]. Thisuggests that the massive attenuation of AChE overproduction,hich follows Dox treatment, reduced the normal inhibition over

ocomotion in Tet/AS mice subjected to stress.Next, we tested the effect of haloperidol. In all mouse groups,

njecting haloperidol sharply reduced locomotor activityF (1,22) � 8.86, p � .0005] in an open-field test compared withnjecting saline. This decrease in locomotion, however, waselatively limited in CB6 mice after administering Dox [F (1,22) �.28, p � .06]. The general motor effect produced is probablyependent on the infection status of the animal (Figure 6B2).hus, Dox by itself may suppress the haloperidol-induced effectn locomotion, likely because of the reduced inflammatorytatus under suppressed bacterial infection that is the direct effectf this antibiotic (Lee et al 2004).

et/AS Expression Prolongs Haloperidol-Induced Catalepsyn Home Cages

Stressful insults are known to induce dopaminergic hyperac-ivation, which is pivotal for preventing catalepsy (Bowers andoth 1972). Others have shown that cholinergic reactions areausally involved in such dopaminergic hyperactivation (Picconit al 2003). We hypothesized that the cholinergic reactionsnvolved require adaptive changes in ACHE gene expression andhat the Tet/AS mice, incapable of such adaptive reaction, woulduccumb to prolonged, stress-induced catalepsy.

To induce catalepsy, we placed the mice with their forepawseaning on an elevated platform while their hind legs remained onhe open-field floor. Catalepsy was defined as maintaining thisosture, and its duration was measured. In the open field andollowing an injection of 2 mg/kg haloperidol, both Tet/AS and CB6ice maintained a cataleptic state for about 8 sec. This suggests thatnder the stressful conditions of open-field exposure, the inductionf catalepsy was refractory to the doxycycline used. Increasing thealoperidol dose to 2.5 mg/kg prolonged the cataleptic responsef CB6 mice to 35 � 12 and 55 � 19 sec in the absence andresence of Dox, with significantly longer periods of 79 � 34 and6 � 20 sec in Tet/AS mice � Dox (Figure 6C1, Table 1). Thus,he mild suppression of AChE in the Tet/AS mice was sufficiento prolong their open-field catalepsy in a manner dependent onhe inherited “leakage” of the Tet/AS system but refractory toox administration, suggesting both a threshold and a saturationffect for this phenotype.

When returned to their home cage, Tet/AS mice receivingmg/kg haloperidol under Dox, but not Dox-untreated Tet/ASr CB6 mice � Dox, displayed appreciable catalepsy that lastedore than a few seconds [F (1,27) � 5.38, p � .03, Table 1]. In

ontrast, after receiving 2.5 mg/kg haloperidol, CB6 mice re-urned to their home cage and displayed a catalepsy reaction for� 8 and 8 � 10 sec and Tet/AS mice for 36 � 21 and 43 � 20

ec for untreated and Dox-fed mice, respectively (Figure 6C2 andupplement 1). Analysis of variance was again significantF (3,17) � 7.13; p � .0026], supporting the notion of ahreshold effect. Thus, AChE-R inhibition emerged as a pivotallement for terminating the emotional arousal associatedith stress, its absence resulting in catalepsy even after

eturning the animals to their familiar home-cage environmentFigure 6D).

iscussion

Taking advantage of a conditional gene regulation model inransgenic mice, we addressed the impact of maladaptive cho-

tor activity, including drug-induced catalepsy. More specifically,we tested the functional consequences of suppressed stress-induced AChE-R production by a tetracycline-regulated antisenseconstruct and explored the modulation of motor response pat-terns under stress, aiming to examine whether cholinergic neu-rotransmission may moderate the intensity of motor reactionsunder stress and enable switching back to normal motor behav-ior when the stressful situation is over. Activation of an antisenseconstruct directed at the ACHE gene resulted in specific changesof motor patterns: hyperlocomotion under open-field stress andexacerbated cataleptic reactions that persisted after return tosafety. These findings indicate that cholinergic neurotransmis-sion, in addition to its ability to enhance stress reactions, canmoderate motor response patterns under stress, most likely viainvolvement of AChE-R. The underlying mechanisms are dis-cussed in the paragraphs that follow.

Tet-AS Treatment Activates Cellular Stress Responsesin Piriform Cortex and Hippocampus

Microscopic examination throughout the brain did not revealdifferences in cell number or density in transgenic Tet/AScompared with control mice; however, differences were detectedin neuritic morphology in restricted regions of the hippocampus,piriform cortex, and striatum and are likely involved in theobserved behavioral phenomena.

In the piriform cortex and hippocampus, clusters of grainswere detected by staining for stress response proteins includingheat shock protein 70 and glucose-regulated protein 78 (GRP78),indicating endoplasmic reticulum stress. The ER chaperoneGRP78 (BiP), induced in response to stress-induced ER accumu-lation of unfolded proteins, negatively regulates three proximalsensors. In the presence of unfolded proteins, GRP78 dissociatesfrom the sensors and binds to the unfolded proteins, thusreleasing the sensors from inhibition, inducing the expression ofUPR genes, and decreasing general translation (Harding et al2000; Yoshida et al 2001). This activates the unfolded proteinresponse, resulting in degradation of misfolded proteins (Formanet al 2003). The accumulation of GRP78 clusters in the Tet/ASbrain thus reflects a chronic ER stress response, and the murineGRP78 and iHSP70 deposits might mimic the deposition ofextracellular matrix molecules presumed to precede amyloidplaque formation and cerebral amyloidosis resulting from im-paired IRE1 function (Katayama et al 1999). Importantly, theTet/AS neuropathology, which involves both the hippocampusand piriform cortex, resembles a parallel phenotype noted withadvancing age (Jucker et al 1994). Compatible with this, agedmice are in general much more sensitive to catalepsy than youngmice (Hoskins et al 1991).

Our findings suggest that antisense-driven reduction ofAChE activity increased the availability of ACh in the piriformcortex and hippocampus. This leads to enhanced sensitivity toglutamatergic neurotransmission in these brain regions (Coxet al 1994), inducing both local and distal consequences. Thelocal consequences involve increased oxidative stress, likelyresponsible for the observed ER stress. The distal conse-quences may include increased glutamatergic output from thehippocampus and piriform cortex to the basal ganglia (Yangand Mogenson 1985). The increase in glutamatergic neuro-transmission may further be responsible for the neuriticmorphology phenomena in the striatum. Compatible with thisconclusion, hippocampal glutamatergic transmission to theventral striatum has earlier been shown to modulate motor

www.sobp.org/journal

SN

NtmdsaeneeUdoArhcwnnsrsoot

ARA

r(cpalecoeciaa(

atYecitbcmsp

748 BIOL PSYCHIATRY 2006;60:741–751 E.H. Sklan et al

w

uppression of AChE Gene Expression Modulates Cholinergiceurite Growth in the Striatum

Dendrites of striatal cholinergic interneurons express various-methyl-D-aspartate (NMDA) receptor subtypes. This suggests

hat the increased glutamatergic neurotransmission in Tet-ASice could affect neurite extension in cholinergic neurons,epending on several factors (e.g., stage of development andubtypes of glutamate receptors; Lipton and Kater 1989). Inddition, chronic stress conditions were shown to affect cholin-rgic neurite growth, likely reflecting impairments in the dy-amic balance between the neuritogenic effects of ACh (Zhengt al 1994) and the intrinsic neuritogenesis induced by AChExpressed in these neurons (Layer 1995; Sternfeld et al 1998).nder exposure to anticholinesterases, increased AChE-R pro-uction is causally involved with neuroprotection from poison-us chemical warfare agents (Soreq and Seidman 2001). ReducedChE-R could affect neuritogenesis in two opposing ways,eflecting its catalytic and morphogenic functions: reduced AChydrolysis would facilitate neurite extension resulting from in-reased ACh signaling, whereas lower AChE-R protein levelsould limit its noncatalytic morphogenic capacity to induceeurite extension. This could maintain balanced cumulativeeuritogenic contributions, with the increased power of AChignals resulting from reduced hydrolysis compensating for theeduced nonenzymatic neuritogenic activities of the antisense-uppressed AChE. In cultured Tet/AS PC12 cells, in the absencef ACh signals, the length but not width of NGF-induced neuriteutgrowth was markedly suppressed after Dox was adminis-ered, supporting this hypothesis.

ttenuated AChE Gene Expression Promotes Extreme Motoreactions and Enhanced Cataleptic Responses tontipsychotic Drugs

We found chronic suppression of AChE levels to intensifyesponses to both threatening stress and antipsychotic drugshaloperidol). This is compatible with the hypothesis that in-reased AChE-R production is pivotal for modulating the com-lex brain pathways governing motor behavior. Both the “syn-ptic” and the AChE-R splice variants are widely distributed inimbic brain regions and in the basal ganglia and are particularlynriched in the striatum (Birikh et al 2003). Limbic pathwaysonverging on the striatum may be central to stress modulationf motor behavior (Gray and McNaughton 2000; Groenewegent al 1987). Our findings support the notion that under optimalonditions, these pathways navigate behavior smoothly betweenmmobility and locomotion. Under conditions in which thebility to produce AChE-R is diminished (e.g., in aged individu-ls), navigation of behavior would loose its smooth featuresFigure 7).

Earlier studies on the effects of stress on catalepsy yieldedpparently contradictory results. Thus, some studies concludedhat stressful experiences attenuate catalepsy (Chopde et al 1995;ntema and Korf 1987), whereas others found stressful experi-nces to promote it (Antelman et al 1992). Apparently, certainonditions, such as acute physical stress (cold exposure, restraintmmobilization), associate with attenuation of catalepsy. In con-rast, other conditions, for example, when threat of an aversiveut distal stimulus is imminent (psychological stress), enhanceatalepsy. Furthermore, different neurotransmitters and proteinsay be involved in these conditions. Thus, under acute physical

tress, adrenal hormones attenuate catalepsy, whereas under

ww.sobp.org/journal

glutamate–ACh and ACh/Da interactions in the striatum, whichmay facilitate catalepsy.

When an animal is under direct physical attack, its main

Figure 7. Acetylcholinesterase (AChE)-R mediates motor responses understress. (A) AChE activity staining in the mouse brain (after Franklin andPaxinos 1997). Importantly, this staining does not discriminate betweendistinct splice variants of the AChE protein. Nevertheless, it highlights thefact that the limbic regions project into the striatum. Pathways (markedgreen) from the AChE-rich cingulate cortex, hippocampus (HPC), and amyg-dala (Amg) reach the AChE-rich striatum (Str). The substantia nigra parscompacta (SN), also enriched in AChE, projects dopaminergic (DA) signals(blue) into the striatum. Our findings are compatible with the hypothesisthat adaptive cortical motor reactions (red arrow) require AChE-R in thehippocampus and striatum. (B) The circled enlarged neuron shows theconvergence of these multiple neurotransmission signals onto the neuritesof the striatal cholinergic neurons. (C) Drawing showing the gradualchanges in neuronal AChE-R levels is based on our previous (Birikh et al2003; Sternfeld et al 2000) and present data. The highest levels of AChE-R,such as those induced under stress, are associated with well-balanced shiftsfrom mobility to immobility. Low AChE-R levels (e.g., in the current Tet/ASmodel) are associated with extreme locomotion or catalepsy, depending onthe acetylcholine–DA balance (central drawing). Specifically, our data sug-gest that AChE-R mediates motor responses in subjects faced with an envi-ronmental threat (e.g., exposure to danger). Under normal dopaminergicneurotransmission conditions, AChE-R moderates the tendency for hyper-locomotion. Given abnormally low or drug-suppressed dopaminergic neu-rotransmission, AChE-R moderates the tendency to suppress movement(catalepsy). Both extremes are revealed under conditions of progressiveTet/AS suppression of AChE-R synthesis.

chances for survival involve struggle and escape, rather than the

f(iltdFimff(ettr

rhineeublbtun

C

icm(miineccmeamatt

Aftevaonttp

E.H. Sklan et al BIOL PSYCHIATRY 2006;60:741–751 749

reezing reactions of catalepsy; however, when danger is distalthreat), suppression of locomotion and even freezing maymprove chance of survival. Supporting this notion, animals canearn to expect danger and display catalepsy even without beingreated with haloperidol (Iwata and Mikuni 1980). This catalepsyisappears as soon as rats are returned to their home cage.urthermore, catalepsy can be enhanced with repeated admin-stration of haloperidol (Fregoso-Aguilar et al 2002). Catalepsyay hence be perceived as being on the same continuum with

reezing, as evident from a strain of rats in which fear-inducedreezing results in catalepsy without haloperidol treatmentBarykina et al 2004). Taken together, these converging lines ofvidence suggest that AChE-R modulates a risk-assessment sys-em, which engages inputs from the hippocampal formation tohe striatum (Groenewegen et al 1987) and moderates motoreactions to distal threats.

Importantly, striatal cholinergic interneurons have not beeneported in electrophysiologic studies to fire during motor be-avior. Rather, cholinergic neurons are notably positioned tonnervate and modulate synaptically striatal projection outputeurons, which in turn fire during motor behavior (Kawaguchit al 1995). Nevertheless, we found that manipulation of cholin-rgic gene expression prolonged the release from catalepsypon transition from threat to safety. This highlights the contri-ution of cholinergic reactions to the stream of cortical andimbic information reporting environmental conditions (Cala-resi et al 2000; Kawaguchi et al 1995; Pollack 2001). It remainso be further explored how switching between motor patternsnder stress is modulated by components of striatal cholinergiceurotransmission.

ontext-Dependent ConsiderationsAlthough this study emphasizes cholinergic–dopaminergic

nteractions, contributions of other neurotransmitters should beonsidered, in particular, glutamate. Glutamatergic neurotrans-ission acting on striatal cholinergic neurons via metabotropic

Bell et al 2002) and NMDA receptors (Calabresi et al 2000) mayediate their combined contribution to catalepsy. Thus, block-

ng of glutamatergic NMDA receptors antagonizes haloperidol-nduced catalepsy (Yoshida et al 1991). Furthermore, dopami-ergic D2 receptors , localized on cholinergic neurons (Alcantarat al 2003; Berlanga et al 2005), antagonize the glutamatergic-to-holinergic signals (Pisani et al 2000), and haloperidol-induced-fos expression involves increased NMDA-mediated neurotrans-ission (Yanahashi et al 2004). Different types of stress therefore

ngage distinct brain regions and neurotransmitters and mayctivate these systems differentially through various cellular andolecular circuits (e.g., receptor subtypes) within those brain

reas and pathways. This activation may, in turn, guide distinctypes of motor behaviors (such as catalepsy vs. hyperlocomo-ion).

The rapid yet long-lasting stress-induced overproduction ofChE-R (Meshorer et al 2002), is tightly controlled by the shift

rom stress to anxiety reactions (Sklan et al, in preparation). Inhis context, stress-induced catalepsy in rodents resembles thextrapyramidal side effects that occur in humans exposed toarious antipsychotic drugs (Janowsky et al 1994), some of whichlso function as AChE inhibitors (Muller et al 2002). In view ofur findings, the occurrence of extrapyramidal symptoms ineurologic patients with Parkinson’s disease or psychiatric pa-ients receiving antipsychotic agents might be causally related toheir capacity to respond to stress stimuli by increasing AChE-R

pressing �25% of neuronal AChE reduced stress memories butdid not impair motor behavior (Nijholt et al 2004). This supportsthe notion that larger AChE-R reductions, such as those enabledby the Tet/AS system, are required for inducing extreme changesin motor reactions (Scheme I). In 1973, Oliver Sacks stated: “Wecould only guess at the relative importance of various determi-nants in this catastrophic reaction” (Sacks 1999, p. 216). Furtherstudies would be required to gain insight into the dynamics ofthe cataleptic process when adapting to the exposure to dopa-minergic antagonists and establish its putative association withthe AChE-R feedback reaction to stressful stimuli and the cessa-tion of such stimuli in humans.

This work was supported by the Israel Science Fund (GrantNos. 618/02 to H.S. and 484/02 to S.S.) and European UnionGrants Nos. LSHM-CT-2003-503330 and LSH-2004-1.1.5-3-EURASNET (to HS).

Supplementary data associated with this article can be found,in the online version, at doi:10.1016/j.biopsych.2006.03.080.

Alcantara AA, Chen V, Herring BE, Mendenhall JM, Berlanga ML (2003): Localiza-tion of dopamine D2 receptors on cholinergic interneurons of the dorsalstriatum and nucleus accumbens of the rat. Brain Res 986:22–29.

Antelman SM, Kocan D, Knopf S, Edwards DJ, Caggiula AR (1992): One briefexposure to a psychological stressor induces long-lasting, time-depen-dent sensitization of both the cataleptic and neurochemical responsesto haloperidol. Life Sci 51:261–266.

Barykina NN, Alekhina TA, Chugui VF, Petrenko OI, Popova NK, Kolpakov VG(2004): Correlation between cataleptic freezing and prepulse inhibitionof the startle reflex in rats. Neurosci Behav Physiol 34:413– 416.

Bell MI, Richardson PJ, Lee K (2002): Functional and molecular characteriza-tion of metabotropic glutamate receptors expressed in rat striatal cho-linergic interneurones. J Neurochem 81:142–149.

Berlanga ML, Simpson TK, Alcantara AA (2005): Dopamine D5 receptor local-ization on cholinergic neurons of the rat forebrain and diencephalon: Apotential neuroanatomical substrate involved in mediating dopaminer-gic influences on acetylcholine release. J Comp Neurol 492:34 – 49.

Birikh KR, Sklan EH, Shoham S, Soreq H (2003): Interaction of “readthrough”acetylcholinesterase with RACK1 and PKCbeta II correlates with intensi-fied fear-induced conflict behavior. Proc Natl Acad Sci U S A 100:283–288.

Bowers MB Jr, Roth RH (1972): Interaction of atropine-like drugs with do-pamine-containing neurones in rat brain. Br J Pharmacol 44:301–306.

Cabib S, Giardino L, Calzá L, Zanni M, Mele A, Puglisi-Allegra S (1998): Stresspromotes major changes in dopamine receptor densities within themesoaccumbens and nigrostriatal systems. Neuroscience 84:193–200.

Calabresi P, Centonze D, Gubellini P, Pisani A, Bernardi G (2000): Acetylcho-line-mediated modulation of striatal function. Trends Neurosci 23:120 –126.

Chopde CT, Hote MS, Mandhane SN, Muthal AV (1995): Glucocorticoids attenu-ate haloperidol-induced catalepsy through adrenal catecholamines. J Neu-ral Trans 102:47–54.

Cohen O, Erb C, Ginzberg D, Pollak Y, Seidman S, Shoham S, et al (2002):Neuronal overexpression of “readthrough” acetylcholinesterase is asso-ciated with antisense-suppressible behavioral impairments. Mol Psychi-atry 7:874 – 885.

Cohen O, Reichenberg A, Perry C, Ginzberg D, Pollmacher T, Soreq H,Yirmiya R (2003): Endotoxin-induced changes in human working anddeclarative memory associate with cleavage of plasma “readthrough” ace-tylcholinesterase. J Mol Neurosci 21:199–212.

Cox CL, Metherate R, Ashe JH (1994): Modulation of cellular excitability inneocortex: Muscarinic receptor and second messenger-mediated ac-tions of acetylcholine. Synapse 16:123–136.

Dains K, Hitzemann B, Hitzemann R (1996): Genetics, neuroleptic responseand the organization of cholinergic neurons in the mouse striatum.J Pharmacol Exp Ther 279:1430 –1438.

Day JC, Koehl M, Le Moal M, Maccari S (1998): Corticotropin-releasing factoradministered centrally, but not peripherally, stimulates hippocampal

www.sobp.org/journal

D

D

D

E

F

F

F

G

G

G

G

G

H

H

H

H

H

H

I

I

I

J

J

K

K

750 BIOL PSYCHIATRY 2006;60:741–751 E.H. Sklan et al

w

e Ryck M, Schallert T, Teitelbaum P (1980): Morphine versus haloperidolcatalepsy in the rat: A behavioral analysis of postural support mecha-nisms. Brain Res 201:143–172.

eBoer P, Abercrombie ED (1996): Physiological release of striatal acetylcho-line in vivo: Modulation by D1 and D2 dopamine receptor subtypes.J Pharmacol Exp Ther 277:775–783.

ori A, Cohen J, Silverman WF, Pollack Y, Soreq H (2005): Functional manip-ulations of acetylcholinesterase splice variants highlight alternativesplicing contributions to murine neocortical development. Cereb Cortex15:419 – 430.

ilam D (2005): Die hard: A blend of freezing and fleeing as a dynamicdefense—implications for the control of defensive behavior. NeurosciBiobehav Rev 29:1181.

orman MS, Lee VM, Trojanowski JQ (2003): “Unfolding” pathways in neuro-degenerative disease. Trends Neurosci 26:407– 410.

ranklin KBJ, Paxinos G (1997): The Mouse Brain in Stereotaxic Coordinates.San Diego, CA: Academic Press.

regoso-Aguilar T, Urióstegui T, Zamudio S, de la Cruz F (2002): The differ-ential effect of haloperidol and repetitive induction on four immobilityresponses in mouse and guinea pig. Behav Pharmacol 13:253–260.

ossen M, Bujard H (1992): Tight control of gene expression in mammaliancells by tetracycline-responsive promoters. Proc Natl Acad Sci U S A 89:5547–5551.

raeff FG (1994): Neuroanatomy and neurotransmitter regulation of defen-sive behaviors and related emotions in mammals. Braz J Med Biol Res27:811– 829.

ray JA, McNaughton N (2000): The Neuropsychology of Anxiety: An Inquiryinto the Functions of the Septo-Hippocampal System, 2nd ed. Oxford, En-gland: Oxford University Press.

rifman M, Galyam N, Seidman S, Soreq H (1998): Functional redundancy ofacetylcholinesterase and neuroligin in mammalian neuritogenesis. ProcNatl Acad Sci U S A 95:13935–13940.

roenewegen HJ, Vermeulen-Van der Zee E, te Kortschot A, Witter MP(1987): Organization of the projections from the subiculum to the ven-tral striatum in the rat. A study using anterograde transport of Phaseolusvulgaris leucoagglutinin. Neuroscience 23:103–120.

aber SN, Fudge JL, McFarland NR (2000): Striatonigrostriatal pathways inprimates form an ascending spiral from the shell to the dorsolateralstriatum. J Neurosci 20:2369 –2382.

arding HP, Zhang Y, Bertolotti A, Zeng H, Ron D (2000): Perk is essential fortranslational regulation and cell survival during the unfolded proteinresponse. Mol Cell 5:897–904.

eilig M (2004): The NPY system in stress, anxiety and depression. Neuropep-tides 38:213–224.

eilig M, Koob GF, Ekman R, Britton KT (1994): Corticotropin-releasing factorand neuropeptide Y: Role in emotional integration. Trends Neurosci 17:80 – 85.

orger BA, Roth RH (1996): The role of mesoprefrontal dopamine neurons instress. Crit Rev Neurobiol 10:395– 418.

oskins B, Peeler DF, Lawson K, Barnes AM, Ho IK (1991): Effects of haloper-idol on motor and cognitive functioning in aged mice. Brain Res Bull27:279 –281.

maki T, Shibasaki T, Hotta M, Demura H (1993): Intracerebroventricularadministration of corticotropin-releasing factor induces c-fos mRNA ex-pression in brain regions related to stress responses: Comparison withpattern of c-fos mRNA induction after stress. Brain Res 616:114 –125.

wata N, Mikuni N (1980): EEG change in the conscious rat during immobilityinduced by psychological stress. Psychopharmacology 71:117–122.

zzo E, Sanna PP, Koob GF (2005): Impairment of dopaminergic systemfunction after chronic treatment with corticotropin-releasing factor.Pharmacology Biochem Behav 81:701.

anowsky DS, Overstreet DH, Nurnberger JI Jr (1994): Is cholinergic sensitiv-ity a genetic marker for the affective disorders? Am J Med Genet 54:335–344.

ucker M, Walker LC, Kuo H, Tian M, Ingram DK (1994): Age-related fibrillardeposits in brains of C57BL/6 mice. A review of localization, stainingcharacteristics, and strain specificity. Mol Neurobiol 9:125–133.

atayama T, Imaizumi K, Sato N, Miyoshi K, Kudo T, Hitomi J, et al (1999):Presenilin-1 mutations downregulate the signalling pathway of the un-folded-protein response. Nat Cell Biol 1:479 – 485.

atoh A, Nabeshima T, Kuno A, Wada M, Ukai R, Kameyama T (1996):Changes in striatal dopamine release in stress-induced conditioned sup-

ww.sobp.org/journal

Kaufer D, Friedman A, Seidman S, Soreq H (1998): Acute stress facilitateslong-lasting changes in cholinergic gene expression. Nature 393:373–377.

Kaufman RJ (1999): Stress signaling from the lumen of the endoplasmicreticulum: Coordination of gene transcriptional and translational con-trols. Genes Dev 13:1211–1233.

Kawaguchi Y, Wilson CJ, Augood SJ, Emson PC (1995): Striatal interneurones:Chemical, physiological and morphological characterization. TrendsNeurosci 18:527–535.

Keay KA, Bandler R (2001): Parallel circuits mediating distinct emotionalcoping reactions to different types of stress. Neurosci Biobehav Rev 25:669.

Klemm WR (1985): Evidence for a cholinergic role in haloperidol-inducedcatalepsy. Psychopharmacology 85:139 –142.

Layer PG (1995): Nonclassical roles of cholinesterases in the embryonicbrain and possible links to Alzheimer disease. Alzheimer Dis AssocDisord 9:29 –36.

Layer PG, Willbold E (1995): Novel functions of cholinesterases in develop-ment, physiology and disease. Prog Histochem Cytochem 29:1–99.

Lee KS, Jin SM, Kim SS, Lee YC (2004): Doxycycline reduces airway inflamma-tion and hyperresponsiveness in a murine model of toluene diisocya-nate-induced asthma. J Allergy Clin Immunol 113:902–909.

Lipton SA, Kater SB (1989): Neurotransmitter regulation of neuronal out-growth, plasticity and survival. Trends Neurosci 12:265–270.

Massoulié J, Bon S, Perrier N, Falasca C (2005): The C-terminal peptides ofacetylcholinesterase: Cellular trafficking, oligomerization and functionalanchoring. Chem Biol Interact 3:157–158.

McNaughton N, Gray JA (2000): Anxiolytic action on the behavioural inhibi-tion system implies multiple types of arousal contribute to anxiety. JAffect Disord 61:161.

Meshorer E, Erb C, Gazit R, Pavlovsky L, Kaufer D, Friedman A, et al (2002):Alternative splicing and neuritic mRNA translocation under long-termneuronal hypersensitivity. Science 295:508 –512.

Meshorer E, Soreq H (2006): Virtues and woes of AChE alternative splicing instress-related neuropathologies. Trends Neurosci 29:216 –224.

Miyakawa T, Yamada M, Duttaroy A, Wess J (2001): Hyperactivity and intacthippocampus-dependent learning in mice lacking the M1 muscarinicacetylcholine receptor. J Neurosci 21:5239 –5250.

Muller TC, Rocha JB, Morsch VM, Neis RT, Schetinger MR (2002): Antidepres-sants inhibit human acetylcholinesterase and butyrylcholinesterase ac-tivity. Biochim Biophys Acta 1587:92–98.

Nijholt I, Farchi N, Kye M, Sklan EH, Shoham S, Verbeure B, et al (2004): Stress-induced alternative splicing of acetylcholinesterase results in enhanced fearmemory and long-term potentiation. Mol Psychiatry 9:174–183.

Overstreet DH, Janowsky DS, Gillin JC, Shiromani PJ, Sutin EL (1986): Stress-induced immobility in rats with cholinergic supersensitivity. Biol Psychi-atry 21:657– 664.

Perrier NA, Salani M, Falasca C, Bon S, Augusti-Tocco G, Massoulié J (2005):The readthrough variant of acetylcholinesterase remains very minorafter heat shock, organophosphate inhibition and stress, in cell cultureand in vivo. J Neurochem 94:629 – 638.

Perrotti LI, Hadeishi Y, Ulery PG, Barrot M, Monteggia L, Duman RS, Nestler EJ(2004): Induction of deltaFosB in reward-related brain structures afterchronic stress. J Neurosci 24:10594 –10602.

Phelps EA, LeDoux JE (2005): Contributions of the amygdala to emotion pro-cessing: From animal models to human behavior. Neuron 48:175–187.

Picconi B, Centonze D, Håkansson K, Bernardi G, Greengard P, Fisone G, etal (2003): Loss of bidirectional striatal synaptic plasticity in L-DOPA-induced dyskinesia. Nature Neurosci 6:501–506.

Pisani A, Bonsi P, Centonze D, Calabresi P, Bernardi G (2000): Activation ofD2-like dopamine receptors reduces synaptic inputs to striatal cholin-ergic interneurons. J Neurosci 20:RC69.

Pollack AE (2001): Anatomy, physiology, and pharmacology of the basalganglia. Neurol Clin 19:523.

Power AE, McGaugh JL (2002): Cholinergic activation of the basolateralamygdala regulates unlearned freezing behavior in rats. Behav Brain Res134:307.

Sacks OW (1999): Awakenings. London: Picador.Sauvage M, Steckler T (2001): Detection of corticotropin-releasing hor-

mone receptor 1 immunoreactivity in cholinergic, dopaminergic andnoradrenergic neurons of the murine basal forebrain and brainstemnuclei—potential implication for arousal and attention. Neuroscience

S

S

S

S

S

S

T

E.H. Sklan et al BIOL PSYCHIATRY 2006;60:741–751 751

harma KV, Koenigsberger C, Brimijoin S, Bigbee JW (2001): Direct evidencefor an adhesive function in the noncholinergic role of acetylcholinester-ase in neurite outgrowth. J Neurosci Res 63:165–175.

klan EH, Lowenthal A, Korner M, Sklan EH, Lowenthal A, Korner M, Ritov Y,Landers DM, Rankinen T, et al (2004): Anxiety scores in the HeritageFamily Study associate with expression variabilities and polymorphismsin the acetylcholinesterase paraoxonase locus. Proc Natl Acad Sci U S A101:5512–5517.

oreq H, Seidman S (2001): Acetylcholinesterase—new roles for an old actor.Nat Rev Neurosci 2:294 –302.

ternfeld M, Ming G, Song H, Sela K, Timberg R, Poo M, Soreq H (1998):Acetylcholinesterase enhances neurite growth and synapse develop-ment through alternative contributions of its hydrolytic capacity, coreprotein, and variable C termini. J Neurosci 18:1240 –1249.

ternfeld M, Shoham S, Klein O, Flores-Flores C, Evron T, Idelson GH, et al(2000): Excess “read-through” acetylcholinesterase attenuates but the“synaptic” variant intensifies neurodeterioration correlates. Proc NatlAcad Sci U S A 97:8647– 8652.

vensson A, Carlsson ML, Carlsson A (1995): Crucial role of the accumbensnucleus in the neurotransmitter interactions regulating motor control inmice. J Neural Trans Gen Sec 101:127–148.

akahashi LK, Goh CS (1996): Presynaptic muscarinic cholinergic receptors

Valtorta F, Leon IC (1999): Molecular mechanisms of neurite extension.Philos Trans R Soc Lond B Biol Sci 354:387–394.

Vizi ES (1980): Modulation of cortical release of acetylcholine by noradrena-line released from nerves arising from the rat locus coeruleus. Neuro-science 5:2139 –2144.

Yanahashi S, Hashimoto K, Hattori K, Yuasa S, Iyo M (2004): Role of NMDAreceptor subtypes in the induction of catalepsy and increase in Fosprotein expression after administration of haloperidol. Brain Res 1011:84 –93.

Yang CR, Mogenson GJ (1985): An electrophysiological study of the neuralprojections from the hippocampus to the ventral pallidum and thesubpallidal areas by way of the nucleus accumbens. Neuroscience 15:1015–1024.

Yntema OP, Korf J (1987): Transient suppression by stress of haloperidolinduced catalepsy by the activation of the adrenal medulla. Psychophar-macology 91:131–1345.

Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K (2001): XBP1 mRNA isinduced by ATF6 and spliced by IRE1 in response to ER stress to producea highly active transcription factor. Cell 107:881– 891.

Yoshida Y, Ono T, Kizu A, Fukushima R, Miyagishi T (1991): Striatal N-methyl-D-aspartate receptors in haloperidol-induced catalepsy. Eur J Pharmacol203:173–180.

Zheng JQ, Felder M, Connor JA, Poo MM (1994): Turning of nerve growthcones induced by neurotransmitters. Nature 368:140 –144.

www.sobp.org/journal