B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Research Report

Prenatal restraint stress impairs learning and memory andhippocampal PKCbeta1 expression and translocation inoffspring rats

Jie Wua,b, Tian-Bao Songa,b,⁎, Yuan-Jie Lia, Kan-Sheng Heb, Ling Geb, Li-Rong Wangb

aKey Laboratory of Environment and Genes Related to Diseases of Ministry of Education, Xi'an Jiaotong University School of Medicine,Xi'an Shaanxi, 710061, PR ChinabDepartment of Human Anatomy, Histology and Embryology, Xi'an Jiaotong University School of Medicine, Xi'an Shaanxi, 710061, PR China

A R T I C L E I N F O

⁎ Corresponding author. Department of HumaShaanxi, 710061, PR China. Fax: +86 29 82655

E-mail address: songtbao@mail.xjtu.edu.c

0006-8993/$ – see front matter © 2007 Elsevidoi:10.1016/j.brainres.2007.01.024

A B S T R A C T

Article history:Accepted 6 January 2007Available online 12 January 2007

Prenatal stress results in various learning, behavioral and emotional alterations observedin later life. However, the mechanisms underlying these effects of prenatal stress are notfully understood. In the present study we examined the impact of prenatal stress (anunpredictable restraint stress) during gestational days 13 to 20 on the performance inMorris water maze and passive avoidance training in 1- and 3-month-old rat offspring.The expression and translocation/activation of protein kinase C (PKC) beta1 in thehippocampus of prenatally stressed offspring were also investigated. One-month-oldfemale and male and 3-month-old female prenatally stressed offspring showed longerlatency to find the platform and used the inefficient search strategy in the water mazetask and showed lower memory score in the passive avoidance training compared withcontrols. The expression of PKCbeta1 protein and mRNA in the hippocampus ofprenatally stressed offspring was dramatically weakened. In the control offspringhippocampus, passive avoidance training induced the PKCbeta1 translocation from thecytosol to the membrane, which, however, was not observed in prenatally stressedoffspring. Our results suggest that deficient signal transduction of PKCbeta1 in thehippocampus resulting from prenatal restraint stress may play an important role in theimpairment of learning and memory abilities of offspring.

© 2007 Elsevier B.V. All rights reserved.

Keywords:Prenatal stressLearning and memoryPKCbeta1HippocampusRat

1. Introduction

Increasing evidence in humans indicates a high incidenceof developmental lags and behavioral/emotional problemsin children born from mothers who experienced chronic,psychological stress during pregnancy (Glover and O'Con-nor, 2002; Mulder et al., 2002). Animal studies have alsoshown that prenatal stress leads to impairments in brain

n Anatomy, Histology an032.n (T.-B. Song).

er B.V. All rights reserved

structure, function and behavior in the offspring (Wein-stock, 1997; McEwen, 2000; Blanchard et al., 2001). Learningand memory, as an important function of the brain, havebeen impaired in prenatally stressed offspring, such asslower learning in the water maze, and deficits in delayedalteration learning, passive avoidance conditioning andoperant conditioning (Lordi et al., 1997; Fleming et al.,2002).

d Embryology, Xi'an Jiaotong University School of Medicine, Xi'an

.

206 B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

Although prenatally stressed animals exhibit increasedsecretion of adrenocorticotrophic hormone and corticosterone(McCormick et al., 1995;Weinstock et al., 1998; Schneider et al.,2004), indicating elevated neuroendocrine responses to stress,the mechanism by which prenatal stress affects learning andmemory is still not fully understood. Protein kinase C (PKC) is aphospholipid-dependent enzyme that exhibits a central rolein activity-dependent neuronal plasticity and the early stageof hippocampal long-term potentiation (LTP) (Ben-Ari et al.,1992). PKC is composed of at least 12 different isoforms andenzyme activation is often accompanied by enzyme translo-cation from the cytosol to the plasma membrane. PKCsignaling pathways are strong candidates for mediating thebiochemical changes that underlie learning since specific PKCsubstrates play critical roles in neuronal physiology andactivation of PKC can increase release of specific neurotrans-mitters from neurons in the hippocampus and throughout thenervous system (Majewski and Iannazzo, 1998). The hippo-campus has an important role in many types of learning andmemory. Some evidence indicates that PKC pathways areinvolved in hippocampal-mediated learning. PKC appears tobe activated in hippocampal cells following learning (Berna-beu et al., 1995; Cammarota et al., 1997; Colombo et al., 1997)and translocated from the cytosol to the membrane (Routten-berg et al., 2000). PKCbeta1 is one of behaviorally relevant PKCisoforms that was demonstrated to participate in the earlysynaptic events responsible for the acquisition and consolida-tion of an inhibitory avoidance learning (Paratcha et al., 2000).Furthermore, prenatal exposure to heroin blocked the trans-location of PKCbeta1 in the hippocampus of mice (Huleiheland Yanai, 2006). Taken together, PKCbeta1 might play a rolein the learning impairment observed in prenatally stressedanimals.

In this study, we first investigated the effects of prenatalrestraint stress on spatial learning and memory and on theability to learn and avoid an electric shock of 1- and 3-month-old offspring rats. We next measured the expression ofPKCbeta1 in the hippocampus of the prenatally stressed andcontrol offspring. Finally, we examined the activity ofPKCbeta1 after passive avoidance training to evaluate hippo-campal PKCbeta1 translocation/activation of animals in bothgroups. Our study sought to reveal a possible mechanism bywhich prenatal stress impairs learning and memory of theoffspring.

2. Results

2.1. Morris water maze

The prenatally stressed and control offspring rats wereseparately tested at the age of 1 and 3 months for spatiallearning in awatermaze. The average latencies in 9 sessions ofvarious groups are shown in Fig. 1. All animals were able toimprove their performance (F8,496=173.457, P<0.01) as shownby the reduction of escape latencies. The average latenciesof theprenatally stressedgroupswere significantly longer thanthose of the control groups in 1-month-old female (F1,15=33.85,P<0.01) and male (F1,15=16.73, P<0.01), and 3-month-oldfemale rats (F1,16=4.591, P<0.05), however, there was no

significant difference in 3-month-old male rats (F1,16=3.509,P>0.05).

At the beginning of training, marginal and randomstrategies were mainly used by rats in searching the platform.With training, tendency and straight strategies were adopted.Although some differences in the search-escape strategy wereobserved in 1-month-old male and female and 3-month-oldfemale offspring, the 1-month-old female ones were morepersuasive (Fig. 2). The occurrence times of the straightstrategy, the most efficient one, in prenatally stressed 1-month-old female animals was significantly reduced (P<0.05).

2.2. Passive avoidance conditioning

One- and three-month-old offspring rats were trained in thestep-through passive avoidance procedure, and the retentionsessionwas performed after 24 h. The results (Table 1) indicatethat the differences in T1 values were not significant betweenthe control and prenatally stressed offspring of each age andsex subgroups. Except 3-month-old male rats (U=24.5,P>0.05), the memory scores (T2–T1) of prenatally stressedrats in other groups were significantly lower than those of thecontrol rats (U=10.5, P<0.01 in 1-month-old female rats;U=17, P<0.01 in 1-month-old male rats; U=15, P<0.01 in 3-month-old female rats). In addition, the animals entering thedark compartment in the prenatally stressed groups duringthe retrieval test were significantly more than those in thecontrol groups (χ2=6.6, P<0.01 in 1-month-old female rats;χ2=5.51, P<0.05 in 1-month-old male rats; χ2=4.8, P<0.05 in 3-month-old female rats), except for 3-month-old male off-spring (χ2=0.8, P>0.05).

2.3. Expression of PKCbeta1 protein and mRNA

PKCbeta1 immunoreactivity is widely distributed in thehippocampus. In the control offspring rats, the neuropil ofthe stratum oriens and stratum radiatum was moderatelystained in the hippocampal CA2 region, but heavily stained inthe CA1 andCA3 regions (Fig. 3A). The crowdedpyramidal cellsof the CA3 and someof granular cells of the dentate gyrusweremoderately labeled with a peripheral soma and processstaining (Figs. 3C, D), whereas cell staining in the CA1 andCA2wasweak. The distribution of PKCbeta1 immunoreactivityin the prenatally stressed rats was similar to that of controls,however, the staining was very weak in all regions examined(Fig. 3B). PKCbeta1 mRNA detected by the in situ hybridizationwas mainly located in the cytoplasm of pyramidal cells of theCA1–CA3 regions and in the granular cells of the dentate gyrus.The PKCbeta1 mRNA signal was weaker in the prenatallystressed group vs. the control group in all hippocampal regions(Fig. 4). There were no marked differences in PKCbeta1immunostaining and mRNA signal between male and femaleand between 1-month-old and 3-month-old offspring.

2.4. Translocation of PKCbeta1 after passive avoidancetraining

Western blot analysis displayed similar results in both 1- and3-month-old offspring. The levels of PKCbeta1 protein inTriton-soluble membrane fractions of the hippocampus in the

Fig. 2 – The times of various search strategies in 36 trials (9 sessions, 4 trials/session) used by offspring rats in the Morriswater maze test. One-month-old female (A) and male (B), and 3-month-old female (C) prenatally stressed rats tend to useinefficient random strategy compared with the respective control rats (*P<0.05, **P<0.01). There is no significant difference inusing search strategies between 3-month-old male prenatally stressed and control rats (D).

Fig. 1 – The latency for offspring rats finding the platform in the Morris water maze test. The latency per testing sessionrepresents the average of four trials of all animals in each group. Generally, the latencies of all rats in both control and prenatallystressed groups are reduced with training (P<0.01). The latency of 1-month-old female (n=9) (A) and male (n=9) (B) and3-month-old female (n=10) (C) prenatally stressed rats is significantly longer than that of respective control rats (n=8 each)(P<0.01, P<0.01, and P<0.05, respectively). However, there is no significant difference between 3-month-old male (D)prenatally stressed rats (n=10) and control rats (n=8).

207B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

Table 1 –Memory score (Mean±SD) of prenatally stressedand control offspring rats in the passive avoidance test

Group n Latency(T1) (s)

Memory score(T2–T1) (s)

Animals enteringdark compartment

C1F 10 9.10±1.50 270.10±13.98 2S1F 12 13.25±1.59 109.67±34.03⁎⁎ 9⁎⁎C1M 10 8.40±1.16 278.80±13.24 1S1M 12 15.08±14.19 160.75±29.66⁎⁎ 7⁎C3F 10 10.40±1.35 283.40±4.36 2S3F 10 20.40±4.59 163.10±31.70⁎⁎ 6⁎C3M 10 10.10±1.67 250.80±17.03 4S3M 10 13.34±1.70 167.10±35.90 6

C1F and S1F, the control and prenatally stressed 1-month-oldfemale groups; C1M and S1M, the control and prenatally stressed1-month-old male groups; C3F and S3F, the control and prenatallystressed 3-month-old female groups; C3M and S3M, the control andprenatally stressed 3-month-old male groups.⁎P<0.05, ⁎⁎P<0.01 compared with the control groups in the same ageand sex.

208 B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

1- and 3-month-old training control rats were increased by66.0±12.2% and 55.6±5.6%, respectively, both significantlyhigher (P<0.05) than those in the non-training control rats,indicating a translocation of PKCbeta1 (Fig. 5A). There was nosignificant difference in the level of PKCbeta1 protein inTriton-soluble membrane fractions of the hippocampusbetween the training and non-training prenatally stressedrats of both 1- and 3-month-old groups (Fig. 5B). No effect ofpassive avoidance training on PKCbeta1 level in cytosolfractions of the hippocampus was found in the trainingcontrol and prenatally stressed rats.

3. Discussion

It is suggested from animal studies that prenatal stress mayresult in a general susceptibility to psychopathology, leading

Fig. 3 – Immunohistochemical staining for PKCbeta1 in the hippimmunoreactivity is intensely located in the stratum oriens (o) an(arrows) of the CA3 (C) and granular cells (arrows) of the dentatedramatically reduced in the prenatally stressed rat (B). Scale bar

to behavioral alterations observed throughout postnatal life(Huizink et al., 2004). Numerous prenatal stressors, such asfoot shock, cold water swim, heat stress and restraint orplacement in a novel environment, are applied in such animalstudies. In the present study prenatal restraint stress, which islikely to mimic the psychosocial stress that humans encoun-ter in daily life, was utilized during the late gestation, a periodcritical for synaptogenesis. We found that 1-month-oldprenatally stressed offspring, no matter female or male,showed longer latency to find the platform in the watermaze task and lower memory score in the passive avoidancetraining compared with controls. These results indicate thatprenatal restraint stress impairs the offspring's abilities forspatial learning and memory tasks, and memory retention fornoxious stimuli. However, in 3-month-old offspring, sucheffect was observed only in females, suggesting that prenatalstress-induced behavioral impairment may be reversible andthat the memory ability of adult male offspring has recoveredin a sense. Although mild prenatal stress and stress of shortduration are shown to facilitate brain development andlearning performance in offspring (Fujioka et al., 1999, 2001),most studies on prenatal stress of long duration during thelate pregnancy have revealed changes in brain morphologyand impaired learning and memory of offspring (Lemaire etal., 2000; Coe et al., 2003). Furthermore, the sex and agedifferences in learning and memory deficits induced byprenatal stress have also been reported in the literature withdifferent results. Prenatal stress significantly affected theperformance of female juvenile rats in the step-through typeof passive avoidance procedure (Gue et al., 2004) and caused ahigh anxiety level in female offspring (Bowman et al., 2004).Jiang et al. (2004) found that prenatal exposure to a magneticfield induced impaired performance of female rats at a specificage in theMorris watermaze. These findings are similar to ourresults. Lordi et al. (1997), however, found that the memorycapability of offspring from stressed mothers in passiveavoidance conditioning was impaired, but only adult maleoffspring were involved. It seems that different results are

ocampus of 3-month-old offspring rats. PKCbeta1d stratum radiatum (r) of the CA1–CA3 (A) and pyramidal cellsgyrus (D) in the control rat. However, the staining is=500 μm (A and B), 50 μm (C and D).

Fig. 4 – In situ hybridization microphotographs of the hippocampus showing expression of PKCbeta1 mRNA in 1-month-oldoffspring rats. PKCbeta1 mRNA signals are mainly located in cytoplasm of pyramidal cells (inset in A) of the CA1–CA3 andgranular cells of the dentate gyrus. The signals are apparently stronger in the control rat (A) than that in the prenatally stressedrat (B). Scale bar=500 μm (A and B), 25 μm (inset).

209B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

present in different studies. Maybe, it is because of thedifference in starting time, duration and prediction degree ofprenatal stress, the age and sex of offspring, and even thestrain of animals.

The mechanism for impaired learning and memory ofprenatally stressed offspring needs further investigation, and

Fig. 5 – Expression of PKCbeta1 in hippocampal cytosol andTriton-soluble membrane fractions of offspring rats afterpassive avoidance training. (A) The control group. (B) Theprenatally stressed group. The upper parts of both (A) and (B)show PKCbeta1 protein bands detected by Western blot, andthe lower parts are the bar graphs showing the relativechanges in PKCbeta1 protein levels. The protein levels in thenon-training rats are ascribed 100%. Each bar represents theaverage of three separate assays. *P<0.05 compared with thenon-training rats (n=5 in each subgroup).

the hippocampus is the most studied target brain region.Learning and memory of spatial and contextual informationdepend on the hippocampus (McEwen, 1999, 2000). Manyresearches have demonstrated that prenatal stress induces adecrease in dendritic arborization and density of dendriticspines and synapses (Hayashi et al., 1998; Ishiwata et al., 2005;Barros et al., 2006), a reduction in cell proliferation and brain-derived neurotrophic factor content (VanDenHove et al., 2006)and a lifespan reduction of neurogenesis and volume in thehippocampus of offspring animals (Lemaire et al., 2000; Coe etal., 2003). Moreover, the sex difference in hippocampalimpairments of prenatally stressed offspring has also beenfound, showing that prenatally stressed females, but notmales, had a decrease in the number of hippocampal neurons(Schmitz et al., 2002; Zhu et al., 2004). These data seem to welllink the poor behavioral performance to altered hippocampalmorphology and biochemistry. However, the signaling path-way is largely unknown. Increasing evidence indicates thatprenatal stress impairs LTP in the hippocampus of offspringand PKC pathway is involved in hippocampal LTP and certainforms of learning and memory (Bernabeu et al., 1995; Yang etal., 2006). Thus our study focused on hippocampal PKCbeta1,which is a behaviorally relevant isoform of PKC and not yetwidely studied in the field of prenatal stress. Our result showedthat prenatal stress diminished the expression of PKCbeta1 atbothproteinandmRNA levels in thehippocampusof offspring,especially in the CA3, CA1 and dentate gyrus, the regionsimportant in the entorhinal–hippocampal trisynaptic circuit(Jones, 1993). Unexpectedly, the sex and age difference inPKCbeta1 expressionwasnot obvious in our study.Thismaybedue to limited sensitivity of the methods used. On the otherhand, the control of the behavior is complex and multiplefactors such as glucocorticoids, excitatory amino acids and N-methyl-D-aspartate receptors are involved individually orcooperatively (McEwen, 1999). For example, the high responseof the hypothalamic–pituitary–adrenal axis to prenatal stressis more marked and prolonged in females than in males(McCormick et al., 1995; Richardson et al., 2006), which mightaffect the behavior of offspring independently of the PKCpathway. Nevertheless our finding suggests that PKCbeta1downregulation is, at least in part, responsible for impairedlearning and memory of prenatally stressed offspring.

Basically, PKC is assumed to be present in an inactive formin the cytosol. A further intracellular calciummobilization and

210 B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

stimulation of phospholipid lead to its translocation to themembrane, where it develops a physiologic activity. Accord-ingly, the translocation of PKC can reflect its activation.Translocation of PKC from cytosol to membrane has beenreported after LTP and inhibitory avoidance learning inchicks and rats (Burchuladze et al., 1990; Bernabeu et al.,1995; Cammarota et al., 1997). The deficits of learning andmemory which occurred in aging rats were accompanied bythe impaired translocation of PKC (Battaini et al., 1995).Therefore, we presumed that, besides generally decreasedexpression of PKCbeta1 in the hippocampus, prenatal stressmight also interfere with translocation/activation ofPKCbeta1 during the learning procedure. In order to testthis hypothesis, we investigated the content of hippocampalPKCbeta1 in cytosol and Triton-soluble membrane fractionsby Western blot analysis in the prenatally stressed andcontrol offspring 30 min after passive avoidance training.Our results indicate that, after training hippocampalPKCbeta1 in the control, but not in prenatally stressed,offspring is redistributed and the membrane-associatedPKCbeta1 content is increased. This finding is consistentwith a previous study, which showed that step-downinhibitory avoidance training resulted in a selective increasein the content of PKCbeta1 isozyme in hippocampal synapticmembrane, and bilateral microinjection of a selectiveinhibitor of PKCbeta1 isozyme into the CA1 of the dorsalhippocampus produced amnesia when given before or aftertraining (Paratcha et al., 2000). Moreover, the translocationof PKCbeta could not be induced in the hippocampus ofmice prenatally exposed to heroin, which was associatedwith behavioral deficits, such as impaired spatial discrimi-nation learning (Shahak et al., 2003; Huleihel and Yanai,2006). Thus our results of passive avoidance trainingindicate that prenatal stress results in a deficit in hippo-campal PKCbeta1 translocation/activation during learningand memory.

In summary, our results not only demonstrate a sex-and age-dependent impairment in learning and memory ofprenatally stressed offspring rats but also provide evidencethat PKCbeta1 signaling is related to the behavioral defi-cits. Both the decreased expression and impaired transloca-tion/activation of hippocampal PKCbeta1 induced by prenatalrestraint stress may be an important profile contributingto the impairment in learning and memory ability in theoffspring.

4. Experimental procedures

4.1. Animals and treatment

Sprague–Dawley rats, provided by the Xi'an Jiaotong Univer-sity School of Medicine, were housed with free access to foodand water and in 12 h light/12 h dark, 20–22 °C. The care andtreatment of animals were approved by the Institution AnimalCare and Research Advisory Committee at the Xi'an JiaotongUniversity School of Medicine. Three mature females and twomales were put together in a cage overnight and the vaginalsmear was examined on the following morning. The presencein the smear of both vaginal cells typical of the estrous stage

and spermatozoids indicated day 1 of pregnancy. Individualpregnant rats were separated in a plastic cage and either leftundisturbed or stressed. Restraint stress was done by placingthe rat in a transparent plastic tube (6 cm in diameter andadjustable length) for 45 min, 3 times/day at random intervalsduring gestational days 13 to 20.

After weaning offspring rats of the same sex were grouphoused. Only a single offspring from each litter was randomlyselected to use in the following behavioral and neurochemicalanalyses.

4.2. Morris water maze

Offspring rats were divided into 8 groups: the prenatallystressed 1-month-old female and male (n=9 each), and 3-month-old female andmale (n=10 each); the control 1-month-old female and male (n=8 each), and 3-month-old female andmale groups (n=8 each).

The equipment for thewatermaze is the same as describedby Jiang et al. (2004). For the place navigation (spatial learningacquisition) test, animals were subjected to two test sessions(four trials each) per day for 4.5 consecutive days. Each trialconsisted of placing the rat in water facing the wall of the poolat one of the four starting locations (North, East, South andWest) in a random order. The rat was allowed to search theplatform for a maximum of 120 s. If an animal did not find theplatform in 120 s, it was gently lifted up and placed onto theplatform for 5 s before being returned to the cage. The escapelatency (the duration for finding the platform) and swim pathwere automatically recorded by a video/computer system andthe software developed by the Institute of Psychology, ChineseAcademy of Sciences. The average latency of each sessionwascalculated from the four trials of each rat in the group. Theswimpaths aredefinedas 4 swimstrategies by the software: (1)marginal, swimming along the pool edge; (2) random, swim-ming randomly; (3) tendency, swimming around toward theplatform area; (4) straight, swimming straight toward theplatform. The first two strategies are inefficient and the lastone is the most efficient in searching for the platform. Thetimes of each strategy used by each rat in 36 trials wererecorded and average times in each group were thencalculated.

4.3. One-trail passive avoidance conditioning

Offspring were divided into 8 groups in the sameway as in thewatermaze and trained a one-trial passive avoidance task andtested for retention 24 h later. The experiment was conductedin a shuttle box separated into a lighted compartment and adark one by a wall with a hole permitting animals to pass fromone compartment to the other. The dark compartment isequipped with a grid floor, to which scrambled footshocks(30 V, 4 mA and 50 Hz) can be delivered.

On the first testing day the rat was put into the lightedcompartment, and they then entered the dark compartmentby themselves. When the whole body of the rat entered thedark compartment, the footshock was delivered for about 5 suntil the animal escaped to the lighted compartment, andwas allowed to stay there for 1 min before it was returned toits cage. The step-through latency (T1) that elapsed from the

211B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

moment the rat entered the lighted compartment to themoment when it entered the dark compartment wasmeasured. After 24 h, the animals were subjected to aretrieval test given exactly in the same way as the initialconditioning, except that the rats did not get electric shockswhen entering the dark compartment. The step-throughlatency (T2) was recorded (up to 300 s). The number ofanimals entering the dark compartment was also counted.The memory score (T2–T1) would reflect the ability toremember the nociceptive experience. If the T2–T1 differenceis zero or small, this indicates that the animals haveforgotten their initial nociceptive experience (Lordi et al.,1997, 2000).

4.4. Immunohistochemistry

One- and three-month-old control (n=8 each) and stressed(n=10 each) offspring (half males and half females) weredeeply anesthetized with amobarbital, and all efforts weremade to minimize the suffering of animals. Animals wereperfused transcardially with 4% paraformaldehyde in 0.1 Mphosphate buffer (PB, pH 7.4), and brains removed andimmersed in the same fixative overnight. Serial coronalsections containing the dorsal hippocampus were cut at40 μm with a vibratome. Sections from each rat wereimmunostained free-floating with a polyclonal antibody toPKCbeta1 (1:400, Santa Cruz) and the avidin–biotin–peroxidasecomplex (ABC, Vector) method. PKCbeta1-immunoreactiveproduct was visualized in a chromogen solution containing0.05% diaminobenzidine (DAB, Sigma) and 0.01% H2O2 in 0.1 MPB. Substitution of normal rabbit serum for PKCbeta1 antibodyin the negative control completely eliminated the immuno-histochemical staining.

4.5. In situ hybridization

Animal grouping and perfusion were the same as forimmunohistochemistry except that the fixative solutioncontained 0.1% diethyl pyrocarbonate (Sigma). Serial coronalparaffin sections containing the hippocampus were cut at8 μm. Dewaxed and dehydrated sections were first pretreatedwith proteinase K (Sigma) followed by incubation withprehybridization solution. A synthetic oligonucleotide ofPKCbeta1 mRNA labeled with digoxigenin (Santa Cruz) wasused as a probe, which was applied to tissue sectionsovernight at 37 °C. After washing in a series of standard salinecitrate (SSC), sections were incubated with biotinylatedsecondary antibody (Santa Cruz) to digoxigenin and ABC,respectively. Coloration reaction was developed in DAB–H2O2

solution containing nickel ammonium sulfate. Incubation ofsections in the prehybridization solution without the probewas taken as the negative control.

4.6. Western blot

Thirty minutes after passive avoidance training in the firsttesting day, twenty 1- and 3-month-old offspring weresacrificed by decapitation under deep anesthesia. Twentycontrol and prenatally stressed offspring not subjected topassive avoidance training were also used. The offspring

rats in each age were divided into 4 groups (n=5 each): thenon-training control, training control, non-training stressedand training stressed. The brain was rapidly removed, andhippocampus dissected, frozen in liquid nitrogen, andstored at −80 °C. Frozen hippocampus was homogenizedin 20 mM Tris–HCl buffer (pH7.4) containing 0.32 M sucrose,2 mM EDTA, 0.5 mM EGTA, 0.2 mM phenylmethylsulfonylfluoride and 20 μg/ml leupeptin and centrifuged at 26,000×gfor 30 min at 4 °C. The supernatant represented the cytosolfraction, while the pellet was resuspended in the samevolume of buffer containing 0.2% Triton X-100 for 45 min at4 °C, sonicated and centrifuged. This supernatant repre-sented the Triton-soluble membrane fraction. The totalprotein concentration in both cytosol and Triton-solublemembrane fractions was determined with the Bradfordmethod. The samples (30 μg protein/lane) were electrophor-esed on 10% SDS–PAGE, electroblotted to nitrocellulosemembranes, blocked with Tris buffered saline (TBS) contain-ing 0.05% Tween 20 (TBST) and 5% non-fat dry milk for 2 h.Then membranes were incubated with antibody to PKCbeta1(1:100 in TBST containing 3% bovine serum albumin) over-night at 4 °C followed by alkaline phosphatase conjugatedgoat anti-rabbit IgG (1:500, Santa Cruz) for 1 h. Immunor-eactive bands were visualized with 5-bromo-4-chloro-3-indolyl phosphate and nitro-blue tetrazolium (Sigma).

Coomassie blue-stained gels were used to assess equalloading of protein into gel lanes. A digital gel image analysissystem was used for semi-quantification of PKCbeta1immunoreactivity. PKCbeta1 immunoreactivity of the non-training control and stressed groups was taken as astandard (100%) and that of the training control andstressed groups expressed as a percentage of the standard,respectively.

4.7. Statistical analysis

All data were expressed as mean±SD. The escape latency inthewatermazewas analyzedwith repeated-measure analysisof variance (ANOVA). Treatment, age and sex were the threebetween-subject factors. Mann–Whitney non-parametric testwas applied to analyze swimming strategies. In passiveavoidance conditioning, comparisonofT1wasmadeaccordingto the Wilcoxon test and comparison of (T2–T1) to the Mann–Whitney tests. The difference in number of animals enteringdark compartment was analyzed by chi square test. Therelative expression levels of PKCbeta1 inWestern blot analysiswere analyzed by the t test. Difference was regarded assignificant at P<0.05.

Acknowledgments

This work was supported by a grant from the Key Cultiva-tion Program of Xi'an Jiaotong University. We are mostgrateful to Prof. Tai-Zhen Han (Department of Physiologyand Pathophysiology, Xi'an Jiaotong University School ofMedicine) for her help in the water maze and to Prof. DwightC. German (Department of Psychiatry, University of TexasSouthwestern Medical Center) for assistance with theEnglish grammar.

212 B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

R E F E R E N C E S

Barros, V.G., Duhalde-Vega, M., Caltana, L., Brusco, A., Antonelli,M.C., 2006. Astrocyte–neuron vulnerability to prenatal stress inthe adult rat brain. J. Neurosci. Res. 83, 787–800.

Battaini, F., Elkabes, S., Bergamaschi, S., Ladisa, V., Lucchi, L.,De Graan, P.N., Schuurman, T., Wetsel, W.C., Trabucchi, M.,Govoni, S., 1995. Protein kinase C activity, translocation, andconventional isoforms in aging rat brain. Neurobiol. Aging 16,137–148.

Ben-Ari, Y., Aniksztejn, L., Bregestovski, P., 1992. Protein kinase Cmodulation of NMDA currents: an important link for LTPinduction. Trends Neurosci. 15, 333–339.

Bernabeu, R., Izquierdo, I., Cammarota, M., Jerusalinsky, D.,Medina, J.H., 1995. Learning-specific, time-dependent increasein [3H]phorbol dibutyrate binding to protein kinase C inselected regions of the rat brain. Brain Res. 685, 163–168.

Blanchard, R.J., McKittrick, C.R., Blanchard, D.C., 2001. Animalmodels of social stress: effects on behavior and brainneurochemical system. Physiol. Behav. 73, 261–271.

Bowman, R.E., MacLusky, N.J., Sarmiento, Y., Frankfurt, M.,Gordon, M., Luine, V.N., 2004. Sexually dimorphic effects ofprenatal stress on cognition, hormonal responses, and centralneurotransmitters. Endocrinology 145, 3778–3787.

Burchuladze, R., Potter, J., Rose, S.P., 1990. Memory formation inthe chick depends onmembrane-bound protein kinase C. BrainRes. 535, 131–138.

Cammarota, M., Paratcha, G., Levi de Stein, M., Bernabeu, R.,Izquierdo, I., Medina, J.H., 1997. B-50/GAP-43 phosphorylationand PKC activity are increased in rat hippocampalsynaptosomal membranes after an inhibitory avoidancetraining. Neurochem. Res. 22, 499–505.

Coe, C.L., Kramer, M., Czeh, B., Gould, E., Reeves, A.J., Kirschbaum,C., Fuchs, E., 2003. Prenatal stress diminishes neurogenesis inthe dentate gyrus of juvenile rhesus monkeys. Biol. Psychiatry54, 1025–1034.

Colombo, P.J., Wetsel, W.C., Gallagher, M., 1997. Spatialmemory is related to hippocampal subcellularconcentrations of calcium-dependent protein kinase Cisoforms in young and aged rats. Proc. Natl. Acad. Sci.U. S. A. 94, 14195–14199.

Fleming, A.S., Kraemer, G.W., Gonzalez, A., Lovic, V., Rees, S., Melo,A., 2002. Mothering begets mothering: the transmission ofbehavior and its neurobiology across generations. Pharmacol.Biochem. Behav. 73, 61–75.

Fujioka, T., Sakata, Y., Yamaguchi, K., Shibasaki, T., Kato, H.,Nakamura, S., 1999. The effects of prenatal stress on thedevelopment of hypothalamic paraventricular neurons in fetalrats. Neuroscience 92, 1079–1088.

Fujioka, T., Fujioka, A., Tan, N., Chowdhury, G.M., Mouri, H.,Sakata, Y., Nakamura, S., 2001. Mild prenatal stress enhanceslearning performance in the non-adopted rat offspring.Neuroscience 103, 301–307.

Glover, V., O'Connor, T.G., 2002. Effects of antenatal stress andanxiety: implications for development and psychiatry. Br. J.Psychiatry 180, 389–391.

Gue, M., Bravard, A., Meunier, J., Veyrier, R., Gaillet, S., Recasens,M., Maurice, T., 2004. Sex differences in learning deficitsinduced by prenatal stress in juvenile rats. Behav. Brain Res.150, 149–157.

Hayashi, A., Nagaoka, M., Yamada, K., Ichitani, Y., Miake, Y.,Okado, N., 1998. Maternal stress induces synaptic loss anddevelopmental disabilities of offspring. Int. J. Dev. Neurosci. 16,209–216.

Huizink, A.C., Mulder, E.J.H., Buitelaar, J.K., 2004. Prenatal stressand risk for psychopathology: specific effects or induction ofgeneral susceptibility? Psychol. Bull. 130, 115–142.

Huleihel, R., Yanai, J., 2006. Disruption of the development of

cholinergic-induced translocation/activation of PKCisoforms after prenatal heroin exposure. Brain Res. Bull. 69,174–181.

Ishiwata, H., Shiga, T., Okado, N., 2005. Selective serotoninreuptake inhibitor treatment of early postnatal mice reversestheir prenatal stress-induced brain dysfunction. Neuroscience133, 893–901.

Jiang, M.L., Han, T.Z., Pang, W., Li, L., 2004. Gender- andage-specific impairment of rat performance in theMorris watermaze following prenatal exposure to an MRI magnetic field.Brain Res. 995, 140–144.

Jones, R.S., 1993. Entorhinal–hippocampal connections: aspeculative view of their function. Trends Neurosci. 16,58–64.

Lemaire, V., Koehl, M., Le Moal, M., Abrous, D.N., 2000. Prenatalstress produces learning deficits associated with aninhibition of neurogenesis in the hippocampus. Proc. Natl.Acad. Sci. U. S. A. 97, 11032–11037.

Lordi, B., Protais, P., Mellier, D., Caston, J., 1997. Acute stressin pregnant rats: effects on growth rate, learning, andmemory capabilities of the offspring. Physiol. Behav. 62,1087–1092.

Lordi, B., Patin, V., Protais, P., Mellier, D., Caston, J., 2000. Chronicstress in pregnant rats: effects on growth rate, anxiety andmemory capabilities of the offspring. Int. J. Psychophysiol. 37,195–205.

Majewski, H., Iannazzo, L., 1998. Protein kinase C: a physiologicalmediator of enhanced transmitter output. Prog. Neurobiol. 55,463–475.

McCormick, C.M., Smythe, J.W., Sharma, S., Meaney, M.J., 1995.Sex-specific effects of prenatal stress onhypothalamic–pituitary–adrenal responses to stress and brainglucocorticoid receptor density in adult rats. Brain Res. Dev.Brain Res. 84, 55–61.

McEwen, B.S., 1999. Stress and hippocampal plasticity. Annu. Rev.Neurosci. 22, 105–122.

McEwen, B.S., 2000. Effects of adverse experiences for brainstructure and function. Biol. Psychiatry 48, 721–731.

Mulder, E.J., Robles de Medina, P.G., Huizink, A.C., Van den Bergh,B.R., Buitelaar, J.K., Visser, G.H., 2002. Prenatal maternal stress:effects on pregnancy and the (unborn) child. Early Hum. Dev.70, 3–14.

Paratcha, G., Furman, M., Bevilaqua, L., Cammarota, M., Vianna,M., de Stein, M.L., Izquierdo, I., Medina, J.H., 2000. Involvementof hippocampal PKCbeta1 isoform in the early phase ofmemory formation of an inhibitory avoidance learning. BrainRes. 855, 199–205.

Richardson, H.N., Zorrilla, E.P., Mandyam, C.D., Rivier, C.L., 2006.Exposure to repetitive versus varied stress during prenataldevelopment generates two distinct anxiogenic andneuroendocrine profiles in adulthood. Endocrinology 147,2506–2517.

Routtenberg, A., Cantallops, I., Zaffuto, S., Serrano, P., Namgung,U., 2000. Enhanced learning after genetic overexpression of abrain growth protein. Proc. Natl. Acad. Sci. U. S. A. 97,7657–7662.

Schmitz, C., Rhodes, M.E., Bludau, M., Kaplan, S., Ong, P., Ueffing, I.,Vehoff, J., Korr, H., Frye, C.A., 2002. Depression: reducednumber of granule cells in the hippocampus of female, but notmale, rats due to prenatal restraint stress. Mol. Psychiatry 7,810–813.

Schneider, M.L., Moore, C., Kraemer, G.W., 2004. Moderate levelalcohol during pregnancy, prenatal stress, or both andlimbic–hypothalamic–pituitary–adrenocortical axis responseto stress in rhesus monkeys. Child Dev. 75, 96–109.

Shahak, H., Slotkin, T.A., Yanai, J., 2003. Alterations in PKCgammain themouse hippocampus after prenatal exposure to heroin: alink from cell signaling to behavioral outcome. Brain Res. Dev.Brain Res. 140, 117–125.

213B R A I N R E S E A R C H 1 1 4 1 ( 2 0 0 7 ) 2 0 5 – 2 1 3

Van Den Hove, D.L.A., Steinbusch, H.W.M., Scheepens, A., Van deBerg, W.D.J., Kooiman, L.A.M., Boosten, B.J.G., Prickaerts, J.,Blanco, C.E., 2006. Prenatal stress and neonatal rat braindevelopment. Neuroscience 137, 145–155.

Weinstock, M., 1997. Does prenatal stress impair coping andregulation of hypothalamic–pituitary–adrenal axis? Neurosci.Biobehav. Rev. 21, 1–10.

Weinstock, M., Poltyrev, T., Schorer-Apelbaum, D., Men, D.,McCarty, R., 1998. Effect of prenatal stress on plasma

corticosterone and catecholamines in response to footshock inrats. Physiol. Behav. 64, 439–444.

Yang, J., Han, H., Cao, J., Li, L., Xu, L., 2006. Prenatal stress modifieshippocampal synaptic plasticity and spatial learning in youngrat offspring. Hippocampus 16, 431–436.

Zhu, Z., Li, X., Chen, W., Zhao, Y., Li, H., Qing, C., Jia, N., Bai, Z., Liu,J., 2004. Prenatal stress causes gender-dependent neuronal lossand oxidative stress in rat hippocampus. J. Neurosci. Res. 78,837–844.