B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

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

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

Research Report

Effects of early weaning and social isolation on the expressionof glucocorticoid and mineralocorticoid receptor and11β-hydroxysteroid dehydrogenase 1 and 2 mRNAs in thefrontal cortex and hippocampus of piglets

R. Polettoa, J.P. Steibelb, J.M. Siegforda, A.J. Zanellaa,⁎aAnimal Behavior and Welfare Group, Department of Animal Science, Michigan State University,1287 Anthony Hall, East Lansing, MI 48824, USAbDepartment of Animal Science, Michigan State University, 1205 Anthony Hall, East Lansing, MI 48824, USA

A R T I C L E I N F O

⁎ Corresponding author. Fax: +1 517 353 1699.E-mail address: zanella@msu.edu. (A.J. Za

0006-8993/$ – see front matter © 2005 Elsevidoi:10.1016/j.brainres.2005.10.001

A B S T R A C T

Article history:Accepted 2 October 2005Available online 3 November 2005

Pigs weaned at young ages show more abnormal and aggressive behaviors and cognitivedeficits compared to later weaned pigs. We investigated the effects of age, weaning and/orsocial isolation on the expression of genes regulating glucocorticoid response[glucocorticoid receptor (GR), mineralocorticoid receptor (MR), 11β-hydroxysteroiddehydrogenases 1 and 2 (11β-HSD1 and 11β-HSD2)] in the frontal cortex andhippocampus. Early- (EW; n = 6) and conventionally-weaned (CW; n = 6) piglets wereweaned at 10 and 21 days after birth, respectively. Non-weaned (NW) piglets of both ages(NW; n = 6/group) remainedwith their dams. Immediately before euthanasia, half of CW, EWand NW animals were socially isolated for 15 min at 12 (EW, NW) and 23 (CW, NW) days ofage. Differences in amounts of 11β-HSD1, 11β-HSD2, GR and MR mRNA were determined byquantitative real-time RT-PCR and data subjected to multivariate linear mixed modelanalysis. When compared with NW piglets at 12 days of age, the hippocampi of EW pigletsshowed decreased gene expression (P b 0.01). Social isolation decreased gene expression(P b 0.05) in the frontal cortex of all piglets. Twelve-day-old piglets showed higher MRmRNAin the frontal cortex (P b 0.01) and lower 11β-HSD2 and GR mRNA (P b 0.05) in thehippocampus compared to 23-day-old animals. Results indicate that EW affected thehippocampus of piglets at 12 days of age, while social isolation affected frontal cortexregardless of age. These results may be correlated with behavioral and cognitive changesreported in EW piglets.

© 2005 Elsevier B.V. All rights reserved.

Theme:Neural basis of behaviorTopic:Stress

Keywords:PigSus scrofaStressWelfareGene expressionQ-RT-PCR

1. Introduction

Early maternal separation has been extensively studied andshown to cause long-term psycho-physiological effects onbrain and behavior of humans and animals (Kuhn and

nella).

er B.V. All rights reserved

Schanberg, 1998; Sanchez et al., 2001). Maternal separationinterferes with the proper development of both psychobiolog-ical and neuroendocrine regulatory mechanisms in thedeveloping brain (Kanitz et al., 2004; Kaufman et al., 2000;Plotsky et al., 2001). Moreover, maternal separation affects the

.

Table 1 – Average change in relative expression ofstress-responsive genes in the frontal cortex andhippocampus of weaned compared with non-weanedpiglets

Brain area Weaningstatus

Average foldchange

SEM(limits)

Pvalue

Frontal cortex EW 1.35 [0.913,1.914]

NS

CW −1.48 [−2.196,−0.808]

NS

Hippocampus EW −2.52 [−3.337,−1.244]

b0.01

CW 1.50 [1.1299,1.904]

NS

Note. Positive and negative values indicate higher and lower geneexpression, respectively, in weaned compared with non-weanedpiglets at 12 and 23 days of age (EW = early-weaned, CW =conventionally weaned). SEM = standard error of the mean.NS = not significant.

37B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

individual's behavior by eliminating the predictability andcontrollability provided by the mother–infant interaction(Plotsky et al., 2001). The early weaning of piglets can providea useful model for studying early maternal separation effectson behavior and neuroendocrine mechanisms. Previousstudies revealed that piglets weaned at young ages, forinstance, at less than 21 days of age show abnormal behaviors,such as belly-nosing (Gardner et al., 2001; Orgeur et al., 2001)and intensified aggression (Hohenshell et al., 2000; Yuan et al.,2004) later in life when compared to piglets weaned at laterages. Early-weaned piglets, when exposed to 15 min of socialisolation, also show cognitive deficits compared to non-socially isolated animals (Laughlin and Zanella, 2002; Souzaand Zanella, 2004). The behavioral outcomes and cognitivedeficits reported may result from modifications in respon-siveness of neuronal mechanisms during a sensitive period ofbrain development (Plotsky et al., 2001; Tuchscherer et al.,2004). An age-dependent sensitivity to environmental manip-ulations (Hemsworth and Barnett, 1992; Hemsworth et al.,1986) and brain exposure to stress hormones, such ascorticosteroids, has been reported to occur in piglets duringthe first weeks after birth (Weaver et al., 2000).

Hippocampus and frontal cortex functions include cogni-tion and behavioral organization (Cao et al., 2004; Kesner et al.,1996, 2004; Ongur and Price, 2000). Basal concentrations ofendogenous corticosteroid hormones like glucocorticoids (GC)are essential for maintaining these functions (Erickson et al.,2003). However, hippocampus and frontal cortex can beaffected by stressful events since both areas contain highdensity binding sites for corticosteroids. Thus, high concen-trations of corticosteroids in the hippocampus during braindevelopment cause behavioral and cognitive deficits that maynot be reversible (McEwen, 1997), as well as cognitive deficitsin the frontal cortex by altering neuronal morphology andfunction (Cook and Wellman, 2004; Mizoguchi et al., 2004).

Stressful challenges, such as early weaning and socialisolation, are responsible for activation of the stress axisresulting in increased GC levels (Hay et al., 2001; Hohenshell etal., 2000; Kanitz et al., 2004). One of the primary outcomes ofincreased central GC action is to trigger a negative feedbacksystem which suppresses further activation of the hypotha-lamic–pituitary–adrenal (HPA) axis, preventing serious neuro-nal damage directly caused by GC (Lee et al., 2002). Therefore,at early brain developmental stages, the disruption of the GCnegative feedback system, which is characterized by de-creased GR expression in the hippocampus and frontal cortex,may be responsible for brain dysfunction (Mizoguchi et al.,2003, 2004). The GC action is mediated by mineralocorticoid(MR) and glucocorticoid receptors (GR) which bind GCs withdifferent affinities in the brain (Erickson et al., 2003). Miner-alocorticoid receptors, when activated, buffer the activation ofthe stress response by preventing GC from binding to GR(Perreau et al., 1999), which once heavily bound mediatesvarious neuronal changes (Herman, 1993; McEwen, 1997).Glucocorticoids indirectly regulate the activity of the enzymes11β-hydroxysteroid dehydrogenases type 1 (11β-HSD1) andtype 2 (11β-HSD2), which modulate GC action (Holmes et al.,2003). When acting as dehydrogenases, both enzymes protectthe brain from the deleterious effects of excessive GC byrendering them inactive (Seckl, 1997). Thus, during early

stages of development, the disruption of the HPA axis maycompromise the developing brain and be accountable forsome aspects of cognitive impairment and abnormal beha-viors (Mizoguchi et al., 2003, 2004).

Therefore, our research focus is to understand the molecu-lar events associated with the cognitive deficits reportedpreviously inearly-weanedpiglets subjected to social isolation,a short-term stressor (Laughlin and Zanella, 2002; Souza andZanella, 2004).Wehypothesize that spatial and social cognitiveimpairments experienced by early-weaned piglets are preced-ed by aberrant expression of stress-sensitive genes in thefrontal cortex and hippocampus. Our aim was to investigatethe effects of weaning at two ages, 10 and 21 days, on 11β-HSD1, 11β-HSD2, GR and MR mRNA abundance in the frontalcortex and hippocampus of piglets, when the animals wereeither socially isolated or kept with their littermates.

2. Results

The piglets studied in this experiment were weaned at twoages. Early-weaned animalswereweaned at 10 days of age andsamples collected at 12 days of age. Conventionally-weanedanimals were weaned at 21 days after birth, and frontal cortexand hippocampus collected at 23 days of age.When examiningmRNA levels for 11β-HSD1, 11β-HSD2, GR and MR using real-time RT-PCR, a significant effect of weaning on gene expres-sion was observed only in the hippocampus of early-weanedanimals (interaction of age by weaning; P = 0.004). Early-weaned piglets showed suppressed expression of the fourtested genes in the hippocampus by an average fold of −2.52(P = 0.004), when comparedwith non-weaned piglets at 12 daysof age (Table 1). Conversely, no significant changes inexpressionof the four genesweredetected in thehippocampusof conventionally-weaned piglets when compared with non-weaned piglets at 23 days of age (Table 1). Interestingly, whenexamining the mRNA abundance for 11β-HSD1, 11β-HSD2, GRand MR in the frontal cortex, there were no evidences ofsignificant differences in mRNA levels in either early-weaned

Fig. 1 – Relative gene expression in the frontal cortex of12-day-old compared with 23-day-old piglets. Positivefold changes indicate higher mRNA levels for therespective genes in the frontal cortex of 12-day-oldpiglets. The error bars represent the standard errors ofthe means (SEM). **P b 0.01.

38 B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

compared with non-weaned piglets at age 12 or in conven-tionally-weaned compared with non-weaned piglets at age 23(Table 1). The interaction of age byweaningwas not significantfor this brain area (P N 0.1).

At 12 or 23 days of age, a group of either weaned (EWI andCWI) or non-weaned (NWI) piglets were exposed to 15 min ofsocial isolation. Social isolation significantly suppressed 11β-HSD1, 11β-HSD2, GR and MR mRNA by a mean fold of −1.94(P b 0.05) in the frontal cortex of both 12- and 23-day-oldpiglets,regardless of whether the animals were weaned or non-weaned (Table 2). The interaction of gene with social isolationwas not significant (P = 0.32). Intriguingly, for all four genestested in the hippocampus, no effects of social isolation wereobserved, and no significant interactions of social isolationwith either age or weaning were detected (P N 0.1; Table 2).

Exclusively looking at age effects, 12-day-old pigletsshowed significantly higher levels of MR mRNA (P = 0.007) inthe frontal cortexwhen compared to 23-day-old piglets (Fig. 1).Additionally, simple effect comparisons of age within geneindicated that 12-day-old piglets had lower expression of 11β-HSD2 (P = 0.03) and GR (P = 0.0009) in the hippocampus (Fig. 2)when compared to 23-day-old piglets (interaction of gene byage; P = 0.006).

3. Discussion

Our data illustrate distinct regulation of mRNA expression for11β-HSD1 and 2, GR and MR in the frontal cortex andhippocampus of piglets. Interestingly, early weaning affectedthe studied genes in the hippocampus of piglets but did notaffect frontal cortex expression of the same genes. Whereas,15 min of social isolation affected expression of genesexamined in the frontal cortex of piglets, regardless of age,but did not affect expression in the hippocampus. Both brainareas are involved in cognitive function, control of learningandmemory and behavioral organization processes (McEwen,1997; Radley et al., 2004).

As described in other species, maternal separation alsocauses activation of the HPA axis in piglets as demonstratedby studies which reported that weaning increased cortisollevels in piglets when performed at different ages (Hay et al.,2001; Hohenshell et al., 2000; Kanitz et al., 2004). Our resultsshowed that piglets weaned at 10 days of age (early-weaned)had consistently suppressed mRNA abundance for 11β-HSD1and 2, GR andMR in the hippocampus at 48 h post-weaning (12days of age) compared to non-weaned piglets. However, there

Table 2 – Effect of social isolation on the average mRNAabundance for all tested genes in the brain areasexamined

Brain area Average fold change SEM (Limits) P value

Frontal cortex −1.94 [−2.553, −1.092] b0.05Hippocampus 1.15 [0.942, 1.438] NS

Note. Positive and negative fold changes indicate higher and lowermRNA abundance, respectively, in socially isolated compared withnon-socially isolated animals regardless of weaning status. SEM =standard error of the mean. NS = not significant.

may be a confounding and (or) additive effect of age withweaning. When exclusively considering age effects, 12-day-old piglets were found to have lower levels of 11β-HSD2 andGR gene expression than 23-day-old animals. The decreasedexpression of GR mRNA in this brain area may suggest thatyounger animals may experience difficulties in reacting toraised GC levels and consequently may be less competent inactivating the negative feedback to regulate HPA axis function.The disruption of the GC negative feedback system is relatedwith decreased GR expression in hippocampus and may beassociated with brain dysfunction (Mizoguchi et al., 2003,2004). Furthermore, the decreased levels of 11β-HSD2 mRNAmay indicate greater vulnerability of younger piglets to highGC levels since this enzyme inactivates circulating GC (Holmeset al., 2003). Other studies showed that, during the earlypostnatal period, 11β-HSD2 expression correlates with GRexpression in brain regions where GCs strongly affect neuro-nal division, growth and maturation (Robson et al., 1998).

The suppression of GR expression found in the hippocam-pus of early-weaned piglets partially supports the work ofKanitz et al. (1998) who found a decrease in GR binding in thehippocampus of piglets 4 days afterweaning (at 35 days of age),compared to 2 h before weaning. Decreased GR expression in

Fig. 2 – Relative gene expression in the hippocampus of12-day-old compared with 23-day-old piglets. Positiveand negative fold changes indicate higher and lowermRNA levels of the tested genes in the hippocampus of12-day-old piglets respectively. The error bars representthe standard errors of the means (SEM). *P b 0.05, ***P b 0.001.

39B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

the hippocampus suggests potential attenuation of negativefeedback, which is responsible for suppressing further activa-tion of HPA axis and GC synthesis (Mizoguchi et al., 2003). Ourresults are also in agreement with findings from a previousstudy in which rats exposed to 24 h of maternal deprivationdeveloped reduced MR and GR mRNA in the hippocampus(Vazquez et al., 1996). Moreover, we observed a suppression ofthe mRNA levels for dehydrogenases 11β-HSD1 and 2 in thehippocampus of early-weaned piglets. It is important tomention that the regulation of mRNA for both enzymes inthe brains of either weaned or socially isolated piglets has notbeen previously reported. Based on studies performed mostlywith rodents, it is known that both enzymes play a role in GCmetabolism, protecting the brain from the deleterious effectsof excessive GC (Seckl, 1997). This indicates that the hippo-campus of early-weaned piglets may be exposed to greaterconcentrations of active GC, whichmay be directly responsiblefor negative effects on cognition, due mostly to hippocampalneuronal death (Antonawich et al., 1999; Lee et al., 2002).

Twelve-day-old piglets also showed significantly higherMRmRNA levels in the frontal cortex when compared with 23-day-old piglets. Unfortunately, limited information is knownabout the role of MR in the frontal cortex, but Diorio et al.(1993) reported GR and MR binding in the frontal cortex of ratsand suggested that the frontal cortex mediates an inhibitoryeffect of GC on stress-inducedHPA axis activity. Therefore, it isplausible to argue that the function of MR in the frontal cortexmay be similar to that found in other brain areas in mediatingGC negative feedback (De Kloet et al., 1993; Diorio et al., 1993).In the hippocampus, the beneficial effects of GC in neuronaltransmission and integrity are mediated by MR binding (DeKloet et al., 1993). No differences in expression of either GR orthe dehydrogenases were detected in the frontal cortex of 12-compared to 23-day-old animals.

In addition to early weaning, the impact of 15 min of socialisolation, a short-term stressor, was tested in this experiment.We limited social isolation to only 15 min because our goalwas to minimize nutritional deprivation and repeat the sameprotocol used in a previous study which reported cognitivedeficits in early-weaned piglets following social isolation(Laughlin and Zanella, 2002; Souza and Zanella, 2004). Fifteenminutes of social isolation significantly suppressed expres-sion of 11β-HSD1 and 2, GR and MR mRNAs in the frontalcortex, while no effect was observed in the hippocampus. Inthis case, we could speculate that social isolation, primarily apsychological stress, encompasses social separation fromlittermates and dam and exposure to a novel environment(Hennessy et al., 2004). Previous research using longer periodsof social isolation showed different patterns of gene expres-sion in both frontal cortex and hippocampus of piglets. Forinstance, Kanitz et al. (2004) reported that 2 h of daily socialisolation from 3 to 11 days of age in piglets caused changes inbehavioral, neuroendocrine and immune regulation andproduced long-term effects on HPA axis activity. Furthermore,according to Lapiz et al. (2003), social isolation, when appliedas an early adverse event for long periods, causes long-termeffects on the behavior and neurobiology in the hippocampusand frontal cortex by altering, for instance, central neuro-transmitter functions. Short-term periods of social isolationmay not be enough to cause changes in the hippocampal

mRNA abundance for genes examined. Previous studiesperformed in piglets showed that repeated social isolationsignificantly altered GR binding in the hippocampus (Kanitz etal., 2004; Tuchscherer et al., 2004).

When comparing the differential responses to weaningversus social isolation observed, one should consider that theweaning period consisted of 48 h of “stress”, while 15 min ofsocial isolation could be characterized as an acute stressor.Therefore, the intensity and duration of the stress stimuli maybe accountable for the differences in regulation of geneexpression in the frontal cortex versus the hippocampus.However, the reasons for observed differences between frontalcortex and hippocampus, in regard to regulation of expressionof 11β-HSD1 and 2, GR and MR, in response to distinctstressors (SI and EW) are unclear. Both brain areas are targetsof GC and are intrinsically involved in stress responseregulation (Meaney et al., 1996). However, it is intriguing thatthe negative feedback effects of GC on HPA axis activity in thefrontal cortex of rats vary according to the nature of the stress(Diorio et al., 1993).

In summary, our results indicate that early weaningaffected hippocampal expression of 11β-HSD1 and 2, GR andMR in piglets at 12 days of age, while no effects were observedin conventionally-weaned piglets. Thus, piglets at 10 days ofagemay bemore vulnerable toweaning stress than 21-day-oldanimals. Furthermore, 15 min of social isolation suppressedthe expression of all four tested genes examined in the frontalcortex, but not in the hippocampus, regardless of whether thepiglets were 12 or 23 days old. These changes may becorrelated to some behavioral and cognitive deficits previous-ly reported in piglets weaned, for instance, at 12 days of ages.Thus, early maternal separation in piglets may have the samenegative effects as observed in other species (Mirescu et al.,2004; Sanchez et al., 2001), but further studies should addresswhether changes in mRNA abundance found in this studywould also be confirmed at the level of receptor binding andenzyme activity in the brain.

4. Experimental procedures

4.1. Animals

Twenty-four Large White female piglets from six litters were usedin this experiment. Females were used as the experimentalanimals instead of males because male piglets are subjected tocastration on the first days after birth, which is an adverseexperience that may interfere with later stress response. Theanimals were housed at the Swine Teaching and Research Center,Michigan State University. Housing and feeding routines followedstandard swine production practices. Animal housing and exper-imental procedures were approved by the Michigan State Univer-sity All-University Committee on Animal Use and Care.

Four littermates from each litter were randomly assigned toone of four treatment groups belonging either to early orconventional weaning model. The early weaning model con-sisted of early-weaned control (EWC; n = 3), early-weanedsocially isolated (EWI; n = 3), non-weaned control (NWC; n = 3)and non-weaned socially isolated (NWI; n = 3) treatments. Theconventional weaning model consisted of conventionally-weaned control (CWC; n = 3), conventionally-weaned sociallyisolated (CWI; n = 3), non-weaned control (NWC; n = 3) and non-weaned socially isolated (NWI; n = 3) treatments. Early-weaned

40 B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

piglets (EW) were weaned 10 days after birth, and the littermateswere kept together until day 12 after birth. Conventionally-weaned piglets (CW) were weaned 21 days after birth, andweaned littermates were kept together until day 23 after birth.Non-weaned animals were left with their dams until days 12 or23. On days 12 or 23 after birth, piglets assigned to EWI and CWI,respectively, and NWI groups, were socially isolated for 15 minand immediately euthanized. Social isolation was performed ina pen located in the same room as the dams and littermates butwith no visual contact. Control animals (EWC, CWC and NWC)were then taken from their pens and immediately euthanized.Euthanasia was preceded by induction of general anesthesiawith 5% isoflurane using a Moduflex Coaxial anesthesiamachine (Dispomed, Quebec, Canada). Immediately after reach-ing a profound general anesthesia state, euthanasia wasperformed using an intracardiac injection of Sodium Pentobar-bital (86 mg/kg; Fatal Plus®, Vortech Pharmaceuticals, USA).

4.2. Tissue preparation and RNA extraction

Brains were removed from all animals within 10 min ofeuthanasia. Frontal cortices and hippocampi from all animalswere dissected out of both brain hemispheres and placed inRNAlater™ (Ambion, Austin, TX) to protect RNA integrity andthen snap-frozen in liquid nitrogen. Samples were then stored at−80 °C. The RNAlater™ was removed from the samples accordingto manufacturer's guidelines. Total RNA from individual frontalcortex and hippocampus samples was isolated using Ribo PureKit (Ambion) following the manufacturer's instructions. Quantityand purity of isolated RNA samples were analyzed using an RNA6000 Nano Chip on an Agilent 2100 Bioanalyzer (AgilentTechnologies, Wilmington, DE).

4.3. Quantitative real-time RT-PCR

Quantitative real-time reverse transcriptase polymerase chainreaction (Q-RT-PCR) was performed to examine the relativeabundance of 11β-HSD1, 11β-HSD2, GR and MR mRNA in thefrontal cortex and hippocampus of piglets. Q-RT-PCR was per-formed as described previously (Gibson et al., 1996; Heid et al.,1996). A total of 2 μg of total RNA from each sample was reversetranscribed with oligo (dT)18 and SuperScript II reverse transcrip-tase (Invitrogen Life Technologies Corp., Carlsbad, CA). An ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Rockland,DE) was used to determine quality and quantity of the resultingcomplementary DNA. Primer sequences for Q-RT-PCR weredesigned using Primer Express 2.0 Software (Applied Biosystems)and synthesized by Qiagen (Valencia, CA). Primer sequences fortested and control genes are shown in Table 3.

A total of 30 ng of template cDNA and 12.5 μl of SYBR Green(Applied Biosystems, Foster City, CA) were employed in each real-time reaction. Forward and reverse primer sequences for eachgene were mixed at a final volume concentration of 5 μM andappropriately added to each reaction for amplification purpose.Sus scrofa 18S ribosomal RNA and beta actin were selected ascontrol genes for frontal cortex and hippocampus samples,

Table 3 – Primer sequences used for Q-RT-PCR to measure genpiglets

Gene name GenBank accession no. Forward prim

11β-HSD 1 AF414124 GCAGCTTTGCGC11β-HSD 2 AF414125 GCGAAAGCTTCCGR AF141371 GATCATGACCGCMR U88893 GCGCAATCTGGABeta actin AF054837 CTCCTTCCTGGG18S AY265350 GGCTCATTAAAT

respectively, and used for normalization purposes. All Q-RT-PCR reactions were performed in triplicate using template fromindividual animals in each reaction. A relative standard curvewas used as the Q-RT-PCR quantification method (Johnson et al.,2000; Livak, 1997). The standard curve was constructed using thefollowing amounts of cDNA (in duplicate): 95 ng, 40 ng, 10 ng, 0.1ng and 0.01 ng. A single control sample was chosen to be used astemplate for the standard curve, which was constant for eachtissue and for each group of samples (12 or 23 days). Standardcurves for the control genes and genes of interest wereincorporated into every run. Q-RT-PCR was performed andanalyzed on an ABI Prism 7000 Sequence Detection System(Applied Biosystems).

4.4. Q-RT-PCR data analysis

The relative standard curve quantification software yielded 3expression values for the tested and reference genes in eachsample. These three technical replicates were averaged, and theresult for each tested genewas divided by the corresponding valueof the control to obtain the normalized gene expression in eachbiological sample. For further analysis, the log transformation ofthe normalized quantity for each sample was considered as theresponse variable. The expression of the four genes was analyzedsimultaneously using a multivariate linear mixed model (Littell etal., 1996). Social isolation, age (12 and 23 days), weaning (weanedand non-weaned), gene (11β-HSD1, 11β-HSD2, GR and MR) and allpossible interactions were set as fixed effects. Litter was consid-ered as a random effect. All the (co)variances were estimated, andthe model assumptions (e.g. normality) were assessed. The mainor simple effect contrasts were computed depending on thesignificance of the higher order interactions. The main effects arerepresented by the average difference in expression of twoconditions. For instance, the difference of all socially isolatedversus non-socially isolated individuals is averaged across genes,weaning and age. The simple effects are represented by thedifferences between the two groups within two levels of anotherfactor. For instance, the simple effect of age within gene is thedifference between the expression at two ages within each geneaveraged across social isolation andweaning. Fold changes (ratios)were obtained by back transforming the linear estimates. The backtransformation of the SEM was obtained by taking the anti-log ofthe mean plus or minus the standard error of the mean (SEM). Forpresentation of the results inwhich the gene expressionwas lowerin the first treatment compared to the second, the ratio (b1) wasdivided into −1 to give a negative relative expression value (−1/ratio). Fold changes showing P b 0.05 were considered statisticallysignificant. Fold change values are presented as the means withthe corresponding SEM.

Acknowledgments

We thank Dr. George W. Smith for editing the manuscript. Wealso thank Michigan State University Swine Teaching and

e expression in the hippocampus and frontal cortex of

er (5′ → 3′ end) Reverse primer (5′ → 3′ end)

CAAGAG GATTTCCCAGCAGGAGAGAAGTCCACTGAAC AGGGTCTGTTTGGGCTCATGACTCAACATG TTGCCTTTGCCCATTTCACAGGTTCTTC CGGGAGCGTGTTTTTTAAGCCATGGA CGCACTTCATGATCGAGTTGACAGTTATGGTTCCT AGCTCTAGAATTACCACAGTTATCCAAG

41B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

Research Center for providing animals for this study andCenter for Animal Functional Genomics for essential technicalassistance during the development of this work. This researchwas funded by the Michigan Agricultural Experiment Stationand the United States Department of Agriculture/NationalResearch Initiative Grant # 2001-35204-10810 (A.Z.).

R E F E R E N C E S

Antonawich, F.J., Miller, G., Rigsby, D.C., Davis, J.N., 1999.Regulation of ischemic cell death by glucocorticoids andadrenocorticotropic hormone. Neuroscience 88, 319–325.

Cao, L., Jiao, X., Zuzga, D.S., Liu, Y., Fong, D.M., Young, D., During,M.J., 2004. VEGF links hippocampal activity with neurogenesis,learning and memory. Nat. Genet. 36, 827–835.

Cook, S.C., Wellman, C.L., 2004. Chronic stress alters dendriticmorphology in rat medial prefrontal cortex. J. Neurobiol. 60,236–248.

De Kloet, E.R., Sutanto, W., van den Berg, D.T., Carey, M.P., vanHaarst, A.D., Hornsby, C.D., Meijer, O.C., Rots, N.Y., Oitzl, M.S.,1993. Brain mineralocorticoid receptor diversity: functionalimplications. J. Steroid Biochem. Mol. Biol. 47, 183–190.

Diorio, D., Viau, V., Meaney, M.J., 1993. The role of the medialprefrontal cortex (cingulate gyrus) in the regulation ofhypothalamic–pituitary–adrenal responses to stress.J. Neurosci. 13, 3839–3847.

Erickson, K., Drevets, W., Schulkin, J., 2003. Glucocorticoidregulation of diverse cognitive functions in normal andpathological emotional states. Neurosci. Biobehav. Rev. 27,233–246.

Gardner, J.M., de Lange, C.F., Widowski, T.M., 2001. Belly-nosing inearly-weaned piglets is not influenced by diet quality or thepresence of milk in the diet. J. Anim. Sci. 79, 73–80.

Gibson, U.E., Heid, C.A., Williams, P.M., 1996. A novel method forreal time quantitative RT-PCR. Genome Res. 6, 995–1001.

Hay, M., Orgeur, P., Levy, F., Le Dividich, J., Concordet, D., Nowak,R., Schaal, B., Mormede, P., 2001. Neuroendocrineconsequences of very early weaning in swine. Physiol. Behav.72, 263–269.

Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Real timequantitative PCR. Genome Res. 6, 986–994.

Hemsworth, P.H., Barnett, J.L., 1992. The effects of early contactwith humans on the subsequent level of fear of humans inpigs. Appl. Anim. Behav. Sci. 35, 83–90.

Hemsworth, P.H., Barnett, J.L., Hansen, C., Gonyou, H.W., 1986.The influence of early contact with humans on subsequentbehavioral response of pigs to humans. Appl. Anim. Behav. Sci.15, 55–63.

Hennessy, M.B., Deak, T., Schiml-Webb, P.A., Wilson, S.E.,Greenlee, T.M., McCall, E., 2004. Responses of guinea pigpups during isolation in a novel environment mayrepresent stress-induced sickness behaviors. Physiol. Behav.81, 5–13.

Herman, J.P., 1993. Regulation of adrenocorticosteroid receptormRNA expression in the central nervous system.Cell. Mol. Neurobiol. 13, 349–372.

Hohenshell, L.M., Cunnick, J.E., Ford, S.P., Kattesh, H.G.,Zimmerman, D.R., Wilson, M.E., Matteri, R.L., Carroll, J.A.,Lay Jr., D.C., 2000. Few differences found between early- andlate-weaned pigs raised in the same environment. J. Anim. Sci.78, 38–49.

Holmes, M.C., Yau, J.L., Kotelevtsev, Y., Mullins, J.J., Seckl, J.R., 2003.11 Beta-hydroxysteroid dehydrogenases in the brain: twoenzymes two roles. Ann. N. Y. Acad. Sci. 1007, 357–366.

Johnson, M.R., Wang, K., Smith, J.B., Heslin, M.J., Diasio, R.B., 2000.Quantitation of dihydropyrimidine dehydrogenase expression

by real-time reverse transcription polymerase chain reaction.Anal. Biochem. 278, 175–184.

Kanitz, E., Manteuffel, G., Otten, W., 1998. Effects of weaning andrestraint stress on glucocorticoid receptor binding capacity inlimbic areas of domestic pigs. Brain Res. 804, 311–315.

Kanitz, E., Tuchscherer, M., Puppe, B., Tuchscherer, A., Stabenow,B., 2004. Consequences of repeated early isolation in domesticpiglets (Sus scrofa) on their behavioural, neuroendocrine, andimmunological responses. Brain Behav. Immun. 18, 35–45.

Kaufman, J., Plotsky, P.M., Nemeroff, C.B., Charney, D.S., 2000.Effects of early adverse experiences on brain structure andfunction: clinical implications. Biol. Psychiatry 48, 778–790.

Kesner, R.P., Hunt, M.E., Williams, J.M., Long, J.M., 1996. Prefrontalcortex and working memory for spatial response, spatiallocation, and visual object information in the rat. Cereb. Cortex6, 311–318.

Kesner, R.P., Lee, I., Gilbert, P., 2004. A behavioral assessment ofhippocampal function based on a subregional analysis.Rev. Neurosci. 15, 333–351.

Kuhn, C.M., Schanberg, S.M., 1998. Responses to maternalseparation: mechanisms and mediators. Int. J. Dev. Neurosci.16, 261–270.

Lapiz, M.D., Fulford, A., Muchimapura, S., Mason, R., Parker, T.,Marsden, C.A., 2003. Influence of postweaning social isolationin the rat on brain development, conditioned behavior, andneurotransmission. Neurosci. Behav. Physiol. 33, 13–29.

Laughlin, K., Zanella, A.Z., 2002. Weaning age impairs spatialmemory of pigs at increased but not basal levels of stress.J. Anim. Sci. 80, 83.

Lee, A.L., Ogle, W.O., Sapolsky, R.M., 2002. Stress and depression:possible links to neuron death in the hippocampus.Bipolar Disord. 4, 117–128.

Littell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D. (Eds.),1996. SAS System for Mixed Models. SAS Institute Inc., Cary(647 pp).

Livak, K.J., ABI Prism 7700 Sequence Detection System. UserBulletin no. 2. PE Applied Biosystems, 1997, AB website,bulletin reference: 4303859B 777802-002.http://docs.appliedbiosystems.com/pebiodocs/04303859.pdf.

McEwen, B.S., 1997. Hormones as regulators of brain development:life-long effects related to health and disease. Acta Paediatr.Suppl. 422, 41–44.

Meaney, M.J., Diorio, J., Francis, D., Widdowson, J., LaPlante, P.,Caldji, C., Sharma, S., Seckl, J.R., Plotsky, P.M., 1996. Earlyenvironmental regulation of forebrain glucocorticoid receptorgene expression: implications for adrenocortical responses tostress. Dev. Neurosci. 18, 49–72.

Mirescu, C., Peters, J.D., Gould, E., 2004. Early life experience altersresponse of adult neurogenesis to stress. Nat. Neurosci.7, 841–846.

Mizoguchi, K., Ishige, A., Aburada, M., Tabira, T., 2003. Chronicstress attenuates glucocorticoid negative feedback:involvement of the prefrontal cortex and hippocampus.Neuroscience 119, 887–897.

Mizoguchi, K., Ishige, A., Takeda, S., Aburada, M., Tabira, T., 2004.Endogenous glucocorticoids are essential for maintainingprefrontal cortical cognitive function. J. Neurosci.24, 5492–5499.

Ongur, D., Price, J.L., 2000. The organization of networks within theorbital and medial prefrontal cortex of rats, monkeys andhumans. Cereb. Cortex 10, 206–219.

Orgeur, P., Hay, M., Mormede, P., Salmon, H., Le Dividich, J., Nowak,R., Schaal, B., Levy, F., 2001. Behavioural, growth and immuneconsequences of early weaning in one-week-old large-whitepiglets. Reprod. Nutr. Dev. 41, 321–332.

Perreau, V., Sarrieau, A., Mormede, P., 1999. Characterization ofmineralocorticoid and glucocorticoid receptors in pigs:comparison of Meishan and Large White breeds.Life Sci. 64, 1501–1515.

42 B R A I N R E S E A R C H 1 0 6 7 ( 2 0 0 6 ) 3 6 – 4 2

Plotsky, P.M., Sanchez, M.M., Levine, S., 2001. Intrinsic andextrinsic factors modulating physiological cope systemsduring development. In: Broom, D.M. (Ed.), Dahlem Workshopon Coping with Challenge: Welfare in Animals IncludingHumans. Dahlem Univ. Press, Dahlem, pp. 169–196.

Radley, J.J., Sisti, H.M., Hao, J., Rocher, A.B., McCall, T., Hof, P.R.,McEwen, B.S., Morrison, J.H., 2004. Chronic behavioralstress induces apical dendritic reorganization inpyramidal neurons of the medial prefrontal cortex.Neuroscience 125, 1–6.

Robson, A.C., Leckie, C.M., Seckl, J.R., Holmes, M.C., 1998.11 Beta-hydroxysteroid dehydrogenase type 2 in the postnataland adult rat brain. Brain Res. Mol. Brain Res. 61, 1–10.

Sanchez, M.M., Ladd, C.O., Plotsky, P.M., 2001. Early adverseexperience as a developmental risk factor for laterpsychopathology: evidence from rodent and primate models.Dev. Psychopathol. 13, 419–449.

Seckl, J.R., 1997. 11beta-Hydroxysteroid dehydrogenase in thebrain: a novel regulator of glucocorticoid action?Front. Neuroendocrinol. 18, 49–99.

Souza, A.S., Zanella, A.J., 2004. Can acute stress impair theability of weaned piglets to recognize a familiar conspecific?In: Hanninem, L., Valros, A. (Eds.), Proceedings of the 38thInternational Congress of the ISAE. International Society forApplied Ethology, Helsinki, p. 119.

Tuchscherer, M., Kanitz, E., Puppe, B., Tuchscherer, A., Stabenow,B., 2004. Effects of postnatal social isolation on hormonal andimmune responses of pigs to an acute endotoxin challenge.Physiol. Behav. 82, 503–511.

Vazquez, D.M., Van Oers, H., Levine, S., Akil, H., 1996. Regulation ofglucocorticoid and mineralocorticoid receptor mRNAs in thehippocampus of the maternally deprived infant rat. Brain Res.731, 79–90.

Weaver, S.A., Aherne, F.X., Meaney, M.J., Schaefer, A.L., Dixon,W.T., 2000. Neonatal handling permanently altershypothalamic–pituitary–adrenal axis function, behaviour,and body weight in boars. J. Endocrinol. 164, 349–359.

Yuan, Y., Jansen, J., Charles, D., Zanella, A.J., 2004. The influence ofweaning age on post-mixing agonistic interactions in growingpigs. Appl. Anim. Behav. Sci. 88, 39–46.