Behavioural Brain Research 167 (2006) 42–48

Research report

Heterozygous mice with Ric-8 mutation exhibit impairedspatial memory and decreased anxiety

Tambet Tonissooa, Sulev Koksb,c,∗, Riho Meiera,Sirli Raudb,c, Mario Plaasa, Eero Vasarb,c, Alar Karisa

a Department of Integrative Zoology, University of Tartu, 46 Vanemuise St., Tartu 51014, Estoniab Department of Physiology, University of Tartu, 19 Ravila St., Tartu 50411, Estonia

c Centre of Molecular and Clinical Medicine, University of Tartu, 19 Ravila St., Tartu 50411, Estonia

Received 21 June 2005; received in revised form 21 August 2005; accepted 22 August 2005Available online 10 October 2005

Abstract

Ric-8 is a guanine nucleotide exchange factor for a subset of G� proteins and it is required to maintain G�q and the G�s pathways in functionalstate. In adult mice Ric-8 is expressed in regions involved in the regulation of behavior (neocortex, cingulate cortex and hippocampus). As Ric-8irw ed thel salt nces int for anxia©

K

1

tzRfbapRassr

C-d

andnesis

vousthat

ingranialglia,

rtex,llum

sentf

0d

s shown to regulate neuronal transmitter release, the aim of present study was to perform behavioral analysis ofric-8 mutant. Homozygous (−/−)ic-8 mutant mice are not viable and die in early embryonic development, therefore for behavioral analysis heterozygous (+/−) ric-8 mutant miceere used. We found decreased anxiety ofric-8 heterozygous mice in light–dark compartment test where mutant mice significantly avoid

ight compartment. In spatial learning paradigm (Morris water maze) the performance ofric-8 (+/−) mice was impaired. Namely, in the reverest,ric-8 (+/−) mice exhibited significant delay to find the hidden platform compared to wild-type (wt) littermates. We did not find differehe behavioral tests reflecting the motor abilities of mice (motor activity, rota-rod). Therefore, described alterations seem to be specificetynd spatial learning. Based on these results we can conclude the importance ofric-8 in the regulation of memory and emotional behavior.2005 Elsevier B.V. All rights reserved.

eywords: Ric-8 (synembryn); Targeted mutagenesis; Spatial memory; Anxiety; Mice

. Introduction

Ric-8 (Synembryn) is a novel 63-kDa cytoplasmic proteinhat is conserved in vertebrates[8]. Biochemical characteri-ation of Ric-8A/synembrin (rat homologs of theC. elegansIC-8) revealed that it is a potent guanine nucleotide exchange

actor for subset of G� proteins. Ric-8A interacts with GDP-ound G� subunits, causing release of GDP and formation ofstable, nucleotide-free transition state complex with the G�

roteins. GTP binding to G� disrupts the complex, releasingic-8A and the activated G� protein[14]. Ric-8 interacts withll three major classes of G�’s that are involved in synapticignaling [2,6,8,9,14]. The function of Ric-8 in the nervousystem is associated with G�o-G�q signaling network, whichegulates neurotransmitter secretion inC. elegans by control-

∗ Corresponding author. Tel.: +372 7374 335; fax: +372 7374 332.E-mail address: Sulev.Koks@ut.ee (S. Koks).

ling the production and consumption of diacylclycerol[4,8,11].Epistasis analysis ofric-8 null mutant worm suggests that RI8 is required to maintain both the G�q priming pathway anthe neuronal G�s pathway in a functional state[13]. RIC-8 alsoplays role in mitotic spindle alignment, nuclear migrationother centrosome-mediated events during early embryogein C. elegans [1,2,9].

Although the importance of Ric-8 in theC. elegans nervoussystem is appreciated, its function in the mammalian nersystem is poorly known. In our earlier study, we showedtemporal and spatial expression profile of the murineric-8(synembryn) is restricted mainly to the neural cells. Durdevelopment of nervous system Ric-8 is expressed in the cganglia, neural tube, sympathetic chain, dorsal root ganvomeronasal organ and endolymphatic sac[16]. In adult micebrain it is expressed in distinct regions, like in the neococingulate cortex, corpus striatum, hippocampus, cerebeand pineal gland[16]. Based on these results the aim of prestudy was to elucidate the role ofric-8 in the regulation o

166-4328/$ – see front matter © 2005 Elsevier B.V. All rights reserved.

oi:10.1016/j.bbr.2005.08.025

T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48 43

behavior. For that purpose we used mutant mice lackingric-8gene. Mice with homozygous (−/−) deletion of ric-8 geneare not viable and they die in early stages of development.Therefore, for behavioral analysis we used heterozygous (+/−)ric-8 mutant mice. In present study, we demonstrate thatric-8(+/−) animals showed higher anxiety compared to wild-type(wt) mice andric-8 (+/−) mice exhibited deficit in spatiallearning paradigm (Morris water maze). In the other tests,reflecting the motor abilities of mutant mice did not express anydifferences compared to wt littermates. Behavioral changes inric-8 (+/−) mutant mice are not caused by unspecific behavioralsuppression or activation as changes present in memory andanxiety tests. Taken together, these results support the role ofric-8 in the regulation of emotional behavior and memory.

2. Materials and methods

2.1. Construction of ric-8 targeting vector and generation ofmutant mice

Targeting vector was constructed to replace first 5 exons ofric-8 genewith the �-geo cassette (Fig. 1). Mouseric-8 gene was isolated from mousegenomic cos library derived from AB1 ES cells. A 9.3 kb XbaI fragment con-taining exons 1–9 was subcloned in pBluescript KS (Stratagene) and 2.7 kbNcoI fragment (exons 1–5) was replaced by�-geo reporter gene. In com-plete targeting vector�-geo sequence was introduced under control of theric-8 regulatory sequences. Herpes simplex virus-thymidine kinase gene was

used as a negative selection marker. Targeting vector was linearized and elec-troporated into E14.1 ES cells yielding 5 clones with homologous recombi-nation out of 150 double selection resistant colonies. Positive clones wereidentified by PCR using a sense primer specific for theric-8 gene (5′-TAAGGTCAACGTGTTCTGAG-3′) and an antisense primer specific for the�-geo cassette (5′-GTGCTGCAAGGCGATTAAGTTG-3′). Chimeric mice weregenerated by injecting the recombinant ES cells into C57BL/6J blastocysts.Yolk-sac DNA of embryos and tail DNA of newborn mice were genotypedby allele-specific PCR using primers 5′-CCAGCCCAAGCCTTCCAG-3′, 5′-CGCATCGTAACCGTG-3′ and 5′-CACACCCCAGCCGAGTTG-3′.

2.2. Animals

In the present study, male and female mice of C57Bl/6J and heterozygousric-8 (+/−) back-crosses (seven generations) were used. In every experiment,mice from each genotype were littermates and of isogenic genetic background.The age of the animals was 18–24 weeks at the beginning of the experiments.Animals were housed in the same groups of four mice per cage and 2-weeksperiod for adaptation was given before the testing began. The animals weremaintained under a 12-h light/12-h dark cycle at relative humidity 50–60% at21± 1◦C, with food and water available ad libitum.

2.3. Behavioral testing

In present study were analyzedric-8 (+/−) mice in the battery of behavioraltests. Tests were performed in following order (light–dark box, motility box,rota-rod, Morris water maze). Animals were brought into experimental room1 h before the experiment. All behavioral experiments were performed between11:00 and 19:00. Since behavioral experiments lasted 6–8 h, precautions weretaken to control possible daily fluctuations in exploratory behavior of animals.

Fag

ig. 1. Targeting strategy forric-8 (A). The geo gene cassette was inserted in-framrrows represent exons, vertical lines indicate restriction enzyme sites. Abbrevene. The mutant mice genotype was verified by allele-specific PCR (B).

e intoric-8 inition codon and TK gene was added for negative selection. Numberediations: geo, nlsLacZ gene and neomycin-resistant gene cassette; tk, thymidine kinase

44 T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48

Therefore, experiments were always performed in randomized order. That is, wtmice were always used in parallel with genetically modified littermates.

2.4. Light–dark exploration

Light–dark test is an unconditioned test of anxiety-like behavior designedfor mice[3]. The Plexiglas box (45 cm× 20 cm× 20 cm) was divided into twoparts: 2/3 was brightly illuminated (∼1000 lx) by a 60 W light bulb fixed 30 cmabove the floor, 1/3 was painted black, covered by a lid and separated from thewhite compartment with a partition containing an opening (13 cm× 5 cm) at thefloor level. The animal was placed in the centre of the dark compartment facingtoward the opening, and the latency to move to the light, the time spent in thelight compartment and the number of transitions between the two compartmentswere measured. Additionally, we measured the number of rearing in the lightpart. The duration of the test was 10 min beginning from the first entry to thelight (the test was terminated if this time was 300 s).

2.5. Motility test

For the study of locomotor activity the animals were placed singly intothe photoelectric motility boxes (448 mm× 448 mm× 450 mm) connected toa computer (TSE Technical & Scientific Equipment GMBH, Germany). Theillumination level of the transparent test boxes was∼750 lx. After removing amouse from the box the floor was cleaned by using 5% alcohol solution. Timein movement (s), the total distance of movement (m), the number of rearing andcorner entries were registered during the 30 min observation period.

2.6. Rota-rod

miceT lasticc inings til thefi testinw andt

2

h mingp wates rfacev fillingT timeo of thep m anr 60 st al wat p untn uctedd s 5 h.T movet ut 18a alyzeb BH,G in thz ditiot asureT t ringc

2

la behav

ioral studies were analyzed using one-way, two-way or three-way analysis ofvariance (ANOVA). Two-factorial ANOVA was performed using genotype andgender as a independent variables for detecting the respective effects. The dataof males and females were pooled where the effect of gender was not significant.

3. Results

3.1. Generation of ric-8 mutant mice

To study the function of the nucleotide exchange factor Ric-8during embryonic development and in the adult mice, we gen-erated mice with a mutantric-8 gene (Fig. 1A). The targetedallele was verified by allele-specific PCR (Fig. 1B). Homozy-gous (−/−) ric-8 mutant mice are not viable and die in earlyembryonic development during gastrulation. Neither develop-mental nor morphological defects were observed in the heterozy-gousric-8 (+/−) embryos. Adult heterozygousric-8 (+/−) micewere viable and fertile and did not present visible morphologi-cal alterations. Brain and body weight was similar to wild-typelittermates.

3.2. Light–dark anxiety test

Light–dark compartment test is used to investigate anxiety-like behavior in rodents. Wild-type (wt; malen = 18; femalen = 18) andric-8 heterozygous mutant (+/−; malen = 18; femalen sig-n thanr ,P :F ee -wayA ,PL art-m e( :FF ders( :F dera sem

3

anysp

3

an oft s

Rota-rod model enables to explore coordinated locomotor functions inhe rota-rod apparatus (produced by workshop at University of Tartu) is pylindrical rod 8 cm in diameter, rotation speed was 10 rpm. A 1 min traession was given to mice on the rota-rod. Motor performance (time unrst fall) was registered during a 2 min session on 3 consecutive days. Reas performed 1 week later. In each trial, the time until falling off the rod

he number of fallings, were recorded.

.7. Morris water maze (MWM)

We used Morris water maze test to study spatial learning abilities inric-8eterozygous mutant mice. The system consisted of a white circular swimool (Ø 120 cm), an escape platform (Ø 10 cm) plunged 0.5 cm under theurface in the centre of one of imaginary quadrants, and a computer inteideo tracing system. Water was made opaque with small amount of latexhe mice were released to swim in random positions facing the wall. Thef reaching the escape platform was measured in every trial. Findinglatform was defined when the mouse climbed onto the escape platforemained there for at least 5 s. When the platform was not found withinhe mouse was guided to the platform and kept there for 15 s. The animaken to its home cage and was allowed to dry up under an infrared lamext trial. Two training sessions, both consisting of three trials, were condaily. Time interval between trials was 3 min and between training blockhe platform remained in a constant location for 3 days and was thereafter

o the opposite quadrant for next 2 days. A transfer test was conducted abofter 6th and 10th training session. The swim-paths were recorded and any the water maze software (TSE, Technical & Scientific Equipment GMermany). The spatial memory of mice was estimated by the time spent

one near the platform and in the zones of three remaining quadrants. In adhe swimming distances and speed and also the thigmotaxis were mehigmotaxis was defined as the time spent swimming within the outermosovering 30.6% of the water maze area.

.8. Statistical analysis

Data on the figures are expressed as mean values± S.E.M. Statisticanalyses were conducted by SPSS for Windows software. Results of

.

g

rd.

d,sil

dhd

en,d.

-

= 18) mice were tested. In this test wt mice madeificantly more transitions between two compartmentsic-8 (+/−) mice (two-way ANOVA: genotype:F3,58= 11.15< 0.01; gender:F3,58= 16.78, P < 0.01; genotype× gender3,58= 0.01, P = 0.92; Fig. 2A). Heterozygous mutant micxpressed significant avoidance of light compartment (twoNOVA: genotype:F3,58= 6.11,P < 0.01; gender:F3,58= 18.55< 0.01; genotype× gender:F3,58= 0.006,P = 0.93,Fig. 2B).atency, time spent until the first entry to the light compent, was in favor ofric-8 (+/−) animals compared to wt mic

two-way ANOVA: genotype:F3,58= 4.90, P < 0.05; gender3,58= 0.56,P = 0.45; genotype× gender:F3,58= 0.89,P = 0.34,ig. 2C). Rearing in the light part was different between gentwo-way ANOVA: genotype:F3,58= 15.37,P < 0.01; gender3,58= 10.42,P < 0.01; significant interaction between gennd genotype:F3,58= 4.25;P < 0.05,Fig. 2D). These findingstablished a increased anxiety-like behavior inric-8 (+/−)ice.

.3. Rota-rod test and locomotor activity

In the rota-rod test and motility boxes we did not findignificant differences in the behavior ofric-8 (+/−) mice com-ared to wt littermates.

.4. Morris water maze

Thirty six wild-type mice (wt; malen = 18; femalen = 18)nd thirty six ric-8 heterozygous mutant mice ((+/−); male= 18; female n = 18) were tested. The escape latency

he heterozygousric-8 (+/−) mice and their wt littermate

T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48 45

Fig. 2. Light–dark exploration test (data expressed as means± S.E.M.). In this panel experimental parameters of male and female mice are compared separatelyaccording to their genotype. (A) Transitions between two compartments. (B) Time spent in the lit compartment, (C) latency (time spent until the first entry to thelight side) and number of rearings (D). Wt (malen = 15; femalen = 15) andric-8 (+/−) (malen = 15; femalen = 15) mice were tested. Abbreviations: m, male; f,female.* P < 0.05 and** P < 0.01 between the wt andric-8 (+/−) mice.

across the training trials decreased considerably and simi-larly (multiple-ANOVA: gender× genotype× training session:F5,355= 1.06,P = 0.37; gender× training session:F5,355= 0.77,P = 0.56; genotype× training session:F5,355= 1.13, P = 0.34;gender× genotype:F1,71= 0.04,P = 0.82; gender:F1,71= 14.48,P < 0.01; genotype:F1,71= 5.61, P < 0.01; training session:F5,355= 34.79,P < 0.01;Fig. 3A and B, sessions 1–6). Effect oftraining session indicates that bothric-8 (+/−) and wt animalslearned to find the platform. The significant difference (P < 0.05)between genotype occurred in training session 5, where wt micefound the platform faster thanric-8 (+/−) mice. In order toestimate the selectivity of spatial navigation two transfer testswere carried out. In the first transfer test, after six sessions oftraining (18 trials) the navigation strategy was similar in bothgroups (Fig. 3C and D). The testing was continued with reversallearning by placing the platform into a new quadrant, oppositeto the previous. In the new situation wt mice displayed betteradjustment and found escape platform faster thanric-8 (+/−)mice (multiple-ANOVA: gender× genotype× training session:F3,213= 0.36,P = 0.78; gender× training session:F3,213= 1.16,P = 0.32; significant genotype and training session interac-tion: F3,213= 2.67, P < 0.05; gender× genotype:F1,71= 1.24,P = 0.26; gender:F1,71= 2.77, P = 0.97; significant genotypeeffect: F1,71= 47.0, P < 0.01; training session:F3,213= 51.63,P < 0.01; Fig. 3A and B, sessions 7–10). The second trans-fer test was performed after 10th training session (30 trials).T ret ormi A:gP

Fig. 3E and F). Furthermore the thigmotactic behavior was notdifferent in transfer tests. Taken together these data suggest, thatduring the initial learning periodric-8 (+/−) mice do not differfrom wt littermates in spatial learning. However, during reversaltraining and testing, wt mice showed significantly better perfor-mance. This difference could be related to the altered learningabilities.

4. Discussion

Main purpose of described study was to characterizeric-8mutant mice. Mice with homozygous (−/−) deletion ofric-8gene are not viable and die early in gastrulation stage of devel-opment. InC. elegans ric-8 reduction-of-function mutants wereshown to exhibit embryonic lethality[9]. Detailed mechanismcausing prenatal death ofric-8 (−/−) mice embryos is not yetclear. In our earlier experiments withric-8 (+/−) mice embryoswe showed that the expression of Ric-8 is first detectable inthe cranial ganglia at E9.5. Later, at E10.5–E12.0 Ric-8 is alsoexpressed in the sympathetic trunk, spinal ganglia, ventral partof the neural tube and in the developing brain, additionally Ric-8was detected in the vomeronasal organ, endolymphatic sac andlens[16]. Moreover, to characterizeric-8 expression level dur-ing the early stages of embyogenesis (at E4.5–E9.5) RT-PCRwas used. Results of RT-PCR experiment revealed thatric-8 isexpressed over the given period but is stronger at stages E4.5 andE mayp viouss dlea iatede

he heterozygousric-8 (+/−) mice spent significantly moime swimming in the opposite quadrant (contained platfn initial learning) than the wt animals (two-way ANOVender× genotype:F1,71= 1.12,P = 0.72; gender:F1,71= 1.47,= 0.22; significant genotype effect:F1,71= 7.51, P < 0.01;

7.5 (unpublished data). These findings indicate that Ric-8lay a role in early stages of mouse embryogenesis. Pretudies have shown that Ric-8 is important in mitotic spinlignment, nuclear migration and other centrosome-medvents during early embryogenesis inC. elegans [1,2,9]. Adult

46 T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48

Fig. 3. Water maze test of spatial learning and memory (data expressed as means± S.E.M.). Escape latencies of male (A) and female mice (B) during trainingsessions (three trials for each session). Training sessions 1–6 represent initial learning and 7–10 reversal learning. Arrowheads illustrate the locations of transfer tests(18 h after 6th and 10th training session). In the first transfer test, the navigation strategy was similar in all groups (males (C) and females (D)). Second transfer testafter reversal learning (males (E) and females (F)). Heterozygousric-8 (+/−) mice spent significantly more time swimming in the opposite quadrant than wt animals.Wt (malen = 18; femalen = 18) andric-8 (+/−) (malen = 18; femalen = 18) mice were tested. Abbreviations: TR, trained; OP, opposite; L/R, left/right quadrants ofmaze;* P < 0.05 and** P < 0.01 between the wt andric-8 (+/−) mice.

heterozygousric-8 (+/−) mice were viable and fertile and didnot present visible morphological alterations.

We initiated series of behavioral tests aimed to detect phe-notypes related to motor activity, coordination and balance,exploratory activity and anxiety, learning and memory. Thesebehavioral tests were chosen according to the expression profileof Ric-8 in the adult mouse brain. The highest Ric-8 expres-sion levels were found in the following areas: neocortex, cin-gulate cortex, caudate putamen, hippocampus, cerebellum andpineal gland[16]. To minimize the variability in the results ofbehavioral studies, groups of animals for behavioral tests werematched by age, gender and genetic background. Littermateshomozygous for either wt or heterozygous mutantric-8 genewere tested. We backcrossed theric-8 (+/−) for seven genera-tions onto the C57Bl/6 genetic background to establish breedingpairs to generate the mice for the current study. Such approachensures that majority of mice used in tests were genetically assimilar as possible.

It was shown that inC. elegans reduction-of-function muta-tions of ric-8 result in strong neuronal phenotypes includingdecreased locomotion and body flexion[8]. The experiments

with G�q and G�o (as major interaction partners of Ric-8) indi-cate, that the absence of G�q in mutant mice, results in motorcoordination deficit, which bounds up with cerebellar dysfunc-tion [12], and mice with G�o (−/−) genotype display severemotor control impairment[5]. Results from our studies sug-gest that heterozygousric-8 (+/−) mice are not different fromwt counterparts with respect to locomotor activity, coordinationand balance. Results from Morris water maze test confirmedintact motor behavior, because swimming speed and climbingability to the platform was not notably affected inric-8 (+/−)mice.

Ric-8 is expressed in the pyramidal and dentate granule cellsof the hippocampus (Fig. 4) [16]. One of the interacting partnersof Ric-8, G�q/G�11, is co-localized with mGluR5 in hippocam-pal pyramidal cells and is primarily localized in the postsynapticextrajunctional membrane[15]. Experiments with mutant miceshowed, that G�q plays a crucial role in the mGluR-dependentLTD because it was absent in Gaq (−/−) mice but exhibitednormal LTP[7]. These findings raised a possibility that Ric-8signaling is involved in the synaptic transmission in the hip-pocampus and in the spatial learning and memory. The Morris

T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48 47

Fig. 4. Ric-8 expression in the adult heterozygousric-8 (+/−) mouse brain using X-gal staining. (A) Cross-section of brain indicates expression in the pyramidal anddentate granule cells of hippocampus. (B) Sagital section of brain show marked expression in the hippocampus, ependymal cell layer of the lateral ventricle, granularcell layer of cerebellum and cinglate cortex. Abbreviations: DG, dentate gyrus; CA, cornu Ammonis field of hippocampus; GrC, granular cell layer of cerebellum;LVe, ependymal cell layer of the lateral ventricle; CG, cingulate cortex. Bars: 300�m (A) and 900�m (B).

water maze test was used to analyze spatial learning and mem-ory in ric-8 heterozygous mice. We found that escape latency ofheterozygousric-8 (+/−) mice and their wt littermates across thetraining trials decreased considerably and similarly. Navigationstrategy in the first transfer test was also similar in both groups.These results indicate thatric-8 (+/−) mice were able to learnthe spatial procedure to locate the hidden platform. Similarly,behavioral pattern ofric-8 (+/−) mice during training session(a 2 min session on 3 consecutive days and retesting 1 weeklater) in rota-rod test did not differ significantly from wt litter-mates. Significant differences betweenric-8 (+/−) and wt micewere observed in the reversal training of the reference memorywater maze task. Mutantric-8 (+/−) mice were less efficientto find the platform in novel position. This finding reveals thatric-8 (+/−) mice could have defective acquisition of new spa-tial information. Additionally the heterozygousric-8 (+/−) micespent significantly more time swimming in the opposite quad-rant (contained platform in initial learning) than wt animals,indicating a poor cognitive flexibility and impaired extinction.Interestingly, in hippocampus, Ric-8 is co-localized with G�q,which is required for LTD[7]. As LTD is essential for intactworking memory, it is reasonable to suggest thatric-8 (+/−)mice have deficit in extinction due to aberrant LTD[10]. How-ever, this hypothesis should be verified later.

Light–dark exploration test is a common method to investi-gate anxiety-like behavior in rodents. In L/D test wt mice mades andtm tiveb , wtm hesr(

ic-8i Ric8 ablet ei atiam port nalb

Acknowledgements

The study was funded by grants 4887 and 6232 fromthe Estonian Science Foundation, target-based project number0182584Bs03 from the Estonian Ministry of Education andScience and by Centre of Molecular and Clinical Medicine,University of Tartu. We thank Sulev Kuuse and Merly Malvafor excellent technical assistance.

References

[1] Afshar K, Willard FS, Colombo K, Johnston CA, McCudden CR,Siderovski DP, et al. RIC-8 is required for GPR-1/2-dependent Gal-pha function during asymmetric division ofC. elegans embryos. Cell2004;119:219–30.

[2] Couwenbergs C, Spilker AC, Gotta M. Control of embryonic spin-dle positioning and Galpha activity byC. elegans RIC-8. Curr Biol2004;14:1871–6.

[3] Crawley J, Goodwin FK. Preliminary report of a simple animal behaviormodel for the anxiolytic effects of benzodiazepines. Pharmacol BiochemBehav 1980;13:167–70.

[4] Hajdu-Cronin YM, Chen WJ, Patikoglou G, Koelle MR, SternbergPW. Antagonism between G(o)alpha and G(q)alpha in Caenorhabdi-tis elegans: the RGS protein EAT-16 is necessary for G(o)alpha sig-naling and regulates G(q)alpha activity. Genes Dev 1999;13:1780–93.

[5] Jiang M, Boulay G, Spicher K, Peyton MJ, Brabet P, Birnbaumer L, etal. Inactivation of the G alpha i2 and G alpha o genes by homologousrecombination. Receptors Channels 1997;5:187–92.

A,a andol and

ientepres-CA1

yn):n the

-1early

:1649–

[ am-Eur J

ignificantly more transitions between two compartmentshe time spent in the lit compartment was longer thanric-8 (+/−)ice. Moreover, the number of rearings (indicator of exploraehavior) in the light part was different between gendersale mice performed more rearings than others groups. T

esults established an increased anxiety-like behavior inric-8+/−) mice.

In conclusions, the murine nucleotide exchange factor Rs important during early embryogenesis. The absence of

causes embryonic lethality. In present study, we wereo show thatric-8 heterozygous (+/−) mutant animals havncreased anxiety-like behavior and modest impaired sp

emory in reversal paradigm. These findings suggest imance of Ric-8 in the regulation of memory and emotioehavior.

e

-

l-

[6] Klattenhoff C, Montecino M, Soto X, Guzman L, Romo X, Garcia Met al. Human brain synembryn interacts with Gsalpha and Gqalphis translocated to the plasma membrane in response to isoproterencarbachol. J Cell Physiol 2003;195:151–7.

[7] Kleppisch T, Voigt V, Allmann R, Offermanns S. G(alpha)q-deficmice lack metabotropic glutamate receptor-dependent long-term dsion but show normal long-term potentiation in the hippocampalregion. J Neurosci 2001;21:4943–8.

[8] Miller KG, Emerson MD, McManus JR, Rand JB. RIC-8 (Synembra novel conserved protein that is required for G(q)alpha signaling iC. elegans nervous system. Neuron 2000;27:289–99.

[9] Miller KG, Rand JB. A role for RIC-8 (Synembryn) and GOA(G(o)alpha) in regulating a subset of centrosome movements duringembryogenesis in Caenorhabditis elegans. Genetics 2000;15660.

10] Nakao K, Ikegaya Y, Yamada MK, Nishiyama N, Matsuki N. Hippocpal long-term depression as an index of spatial working memory.Neurosci 2002;16:970–4.

48 T. Tonissoo et al. / Behavioural Brain Research 167 (2006) 42–48

[11] Nurrish S, Segalat L, Kaplan JM. Serotonin inhibition of synaptic trans-mission: Galpha(0) decreases the abundance of UNC-13 at release sites.Neuron 1999;24:231–42.

[12] Offermanns S, Hashimoto K, Watanabe M, Sun W, Kurihara H, Thomp-son RF, et al. Impaired motor coordination and persistent multipleclimbing fiber innervation of cerebellar Purkinje cells in mice lackingGalphaq. Proc Natl Acad Sci USA 1997;94:14089–94.

[13] Reynolds NK, Schade MA, Miller KG. Convergent, RIC-8-DependentG{alpha} signaling pathways in the Caenorhabditis elegans synapticsignaling network. Genetics 2005;169:651–70.

[14] Tall GG, Krumins AM, Gilman AG. Mammalian Ric-8A (synembryn)is a heterotrimeric Galpha protein guanine nucleotide exchange factor.J Biol Chem 2003;278:8356–62.

[15] Tanaka J, Nakagawa S, Kushiya E, Yamasaki M, Fukaya M, Iwanaga T,et al. Gq protein alpha subunits Galphaq and Galpha11 are localized atpostsynaptic extra-junctional membrane of cerebellar Purkinje cells andhippocampal pyramidal cells. Eur J Neurosci 2000;12:781–92.

[16] Tonissoo T, Meier R, Talts K, Plaas M, Karis A. Expression of ric-8(synembryn) gene in the nervous system of developing and adult mouse.Gene Expr Patterns 2003;3:591–4.