Roles of α- and β-estrogen receptors in mouse social recognition memory: Effects of gender and the estrous cycle

Hormones and Behavior 59 (2011) 114–122

Contents lists available at ScienceDirect

Hormones and Behavior j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y h b e h

Roles of α- and β-estrogen receptors in mouse social recognition memory: Effects of gender and the estrous cycle G. Sánchez-Andrade, K.M. Kendrick ⁎ Laboratory of Molecular Signalling, Cognitive and Systems Neuroscience Group, The Babraham Institute, Babraham, Cambridge CB22 3AQ, UK

a r t i c l e

i n f o

Article history: Received 11 May 2010 Revised 28 October 2010 Accepted 30 October 2010 Available online 4 November 2010 Keywords: Estrogen receptor Estrous cycle Learning Neural plasticity Olfaction Social recognition memory

a b s t r a c t Establishing clear effects of gender and natural hormonal changes during female ovarian cycles on cognitive function has often proved difficult. Here we have investigated such effects on the formation and long-term (24 h) maintenance of social recognition memory in mice together with the respective involvement of α- and β-estrogen receptors using α- and β-estrogen receptor knockout mice and wildtype controls. Results in wildtype animals showed that while females successfully formed a memory in the context of a habituation/ dishabituation paradigm at all stages of their ovarian cycle, only when learning occurred during proestrus (when estrogen levels are highest) was it retained after 24 h. In α-receptor knockout mice (which showed no ovarian cycles) both formation and maintenance of this social recognition memory were impaired, whereas β-receptor knockouts showed no significant deficits and exhibited the same proestrus-dependent retention of memory at 24 h. To investigate possible sex differences, male α- and β-estrogen receptor knockout mice were also tested and showed similar effects to females excepting that α-receptor knockouts had normal memory formation and only exhibited a 24 h retention deficit. This indicates a greater dependence in females on α-receptor expression for memory formation in this task. Since non-specific motivational and attentional aspects of the task were unaffected, our findings suggest a general α-receptor dependent facilitation of memory formation by estrogen as well as an enhanced long-term retention during proestrus. Results are discussed in terms of the differential roles of the two estrogen receptors, the neural substrates involved and putative interactions with oxytocin. © 2010 Elsevier Inc. All rights reserved.

Introduction Apart from their key roles in reproductive behaviors estrogens can influence a number of neural growth, plasticity, learning and memory functions (Maggi et al., 2004). They can, for example, alter the organization and structure (Frankfurt, 1990; Wooley et al., 1990) and function (Gibbs, 1996; Kendrick, 1981; Wooley and McEwen, 1994) of brain areas involved in learning, such as the amygdala and hippocampus. Nevertheless, establishing robust correlations between physiological changes in, or exogenous treatment with, estrogen and cognitive performance has been more difficult. The problem is made more complex since estrogens have a variety of genomic and nongenomic actions (Hall et al., 2001; Toran-Allerand et al., 1999; Wise et al., 2001) and there are issues as to whether dynamic changes in physiological concentrations are important or just overall basal ones (Dohanich, 2002). Under the circumstances it is not surprising that attempts to establish estrogenic effects on cognitive performance have provided conflicting results, varying in both direction and magnitude (Dohanich, 2002). Discrepancies also arise with varying

⁎ Corresponding author. E-mail address: [email protected] (K.M. Kendrick). 0018-506X/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2010.10.016

hormonal treatments (Gibbs, 1997) and the use of diverse learning tests involving different memory systems (for review see Daniel, 2006). A few studies have investigated exogenous or endogenous estrogen effects on social recognition memory in female rodents. In rats, a proestrus facilitation of long-term (5 h) social recognition memory has been reported, but only following vaginocervical stimulation (Larrazolo-Lopez et al., 2008). Exogenous estradiol (E2) treatment has been reported to prolong social recognition to 2 h in ovariectomized group-housed rats (Hlinak, 1993) and 24 h in ovariectomized mice (Tang et al., 2005). These effects are also dependent upon hormone dose and length of treatment. A recent study has also found enhanced habituation in a habituation/dishabituation social recognition paradigm in ovariectomized rats treated with estrogen and progesterone (Spiteri and Ågmo, 2009). With other learning paradigms no specific proestrus facilitation effects have been reported, although in rats impaired spatial memory has been found in either proestrus (Warren and Juraska, 1997) or estrus (Healy et al., 1999) stages. In estrus, mice impaired learning in a footshock avoidance paradigm or spatial learning in the Morris maze (Frick and Berger-Sweeney, 2001) have also been reported. It is possible, however, that these results may have more to do with increased sensitivity to stress during proestrus/estrus in these types of

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

tasks. Dose-related effects of exogenous treatments with E2 on spatial memory and other learning and memory tasks have also been reported, sometimes with contradictory results (Di Paolo et al., 1985; Frick et al., 2002; Fugger et al., 1998; Leuner et al., 2004; Rissanen et al., 1999; Sawada et al., 1998). This dose-dependency could reflect an adaptation of the brain to naturally occurring fluctuations in estrogen levels during the ovarian cycle or pregnancy (Cyr et al., 2002; Rado et al., 1970). Differing hormonal levels may also influence learning strategies involved (Korol, 2004; Zurkovsky et al., 2007), with exogenous treatments producing high estrogen levels being associated with impaired learning whereas those producing lower ones with its facilitation (Holmes et al., 2002; Wide et al., 2004). A further consideration is the roles of each of the two nuclear estrogen receptors, alpha (αER) and beta (βER), which are differentially expressed in brain areas associated with cognitive processing (Mitra et al., 2003; Osterlund et al., 1998; Shughrue et al., 1997a,b). This is further complicated by the fact that expression patterns of estrogen receptors can vary with different hormonal states (Alves et al., 1998; Greco et al., 2001; Nomura et al., 2003; Osterlund et al., 1998; Patisaul et al., 1999). Both male and female mice lacking functional expression of the βER show impaired learning on hippocampal-dependent spatial tasks such as the Morris water maze (Rissman et al., 2002), and do not show estradiol-dependent facilitation of long-term potentiation (Liu et al., 2008). Also, βER agonists can facilitate hippocampal long-term potentiation (LTP) and performance on spatial memory tasks in mice and rats (Liu et al., 2008). On the other hand, αER knockout mice do not show spatial memory deficits and have normal estradiol-dependent facilitation of LTP, and αER agonists do not facilitate spatial memory in rats or mice (Liu et al., 2008; Rissman 2008). Indeed, in some respects the two receptors may even be antagonistic in terms of estrogenic actions on hippocampal function and learning (Rissman 2008). On the other hand, both αER and βER knockout male and female mice have been shown to have deficits in a short-term memory social recognition task based on a habituation/dishabituation paradigm (Choleris et al., 2003, 2004), although those in βER knockout mice are slightly less severe (Choleris et al., 2006). In the social recognition memory task estrogen, through activation of its receptors, is proposed to influence memory through modulating oxytocin (OT) and its receptor (OTR), with OT knockout animals showing similar social recognition memory deficits (Choleris et al., 2003, 2006; Ferguson et al., 2000). From these findings an interacting network involving four genes coding for OT, OTR, αER and βER has been proposed (Choleris et al., 2003, 2004). Estrogens may be acting on the OT system at a number of levels: through βER they regulate the production of OT in the hypothalamic paraventricular nucleus (Nomura et al., 2002; Shughrue et al., 1999, 2002), and through activation of αER they drive the transcription of OTR in the amygdala (Dekloet et al., 1986; Shughrue et al., 1999, 2002; Young et al., 1998). In the olfactory pathways involved in rodent social recognition, olfactory stimuli detected by the main and vomeronasal receptors are initially processed by the main and accessory olfactory bulbs which then project to the cortical and medial amygdala where OT via the OTR has been shown to be important for formation of social recognition memory (Choleris et al., 2007; Ferguson et al., 2001). There are also α and βER (Mitra et al., 2003) and OT and OTR (Broad et al., 1993, 1999; Yoshimura et al., 1993) in the main olfactory bulb and the action of OT within the olfactory bulb is important from prolonging the duration of social recognition following vaginocervical stimulation in proestrus female rats (Larrazolo-Lopez et al., 2008). Previous studies using social recognition memory paradigms to investigate the roles of ERα and β have only used habituation/ dishabituation testing protocols (Choleris et al., 2003, 2006; Gheusi et al., 1994; Imwalle et al., 2002) where a short-term memory (b60 min) for individuals is tested, and no ovarian cycle-dependent effects have been established. Possible gender differences in dependency on the

115

different receptors have also not been considered. In mice, the duration of social recognition memory can be as long as 7 days in group-housed animals and that protein synthesis and cyclic AMP responsive element binding protein (CREB) function are necessary for the long-term form of this social memory (Kogan et al., 2000; Richter et al., 2005). Its consolidation also has two protein synthesis-dependent phases (Wanisch and Wotjak, 2008). The present studies have therefore aimed to establish the influence of gender and the ovarian cycle on the formation and duration of social recognition memory and the respective roles of α and βERs for both the short-term and long-term components of this form of memory. Materials and methods Animals All animal experiments received local Ethical Committee approval and were carried out under license in full compliance with the UK Animals (Scientific Procedures) Act, 1986. Homozygous αERKO and βERKO and wildtype control (C57/Bl6 × 129SV) mice were bred and housed in a full specific pathogen free barrier facility at The Babraham Institute. Founders of both ERKO colonies were kindly provided by Dr. Korach, NIH-NIEHS (USA). The ERKO mice had been backcrossed 10× onto a C57/Bl6 background. Mice were housed under temperature and humidity-controlled conditions with a 12 h light–12 h dark cycle (lights on at 07.00 h), with food (irradiated CRM(P) diet supplied by SDS, UK) and water available ad libitum. The diet used has very low levels of phytoestrogens (Genisten — 141 ppm, Daidzein — 78 ppm, and Coumestrol — undetectable) and well below those which might have produced any biological activity (Owens et al., 2003). Transgenic animals were bred from heterozygous male and females (trios of 1 male and 2 females) and were genotyped by PCR of tail biopsy DNA using a standard protocol (Couse et al., 2003). After weaning, around 31 days old, mice were housed in same-sex groups of between two to five animals in standard racks of M3 plastic cages containing enrichment aids (nesting material and fun tunnels) and were cleaned-out once a week. 16 αERKO and 13 αWT littermate, 18 βERKO and 14 βWT littermate intact adult male mice and 13 αERKO and 28 αWT littermate, 32 βERKO and 33 βWT littermate intact adult female mice were used in the experiments. Due to limited availability of ERKO mice and the fact that they might not be optimal socially attractive stimuli (Ågmo et al., 2008; Kavaliers et al., 2004) we used 24 male and 38 female gonadally intact C57/Bl6 × 129SV adult mice as stimulus animals (this strain has been successfully used as stimulus animals in this paradigm in our lab). At testing, all animals were 3–11 months old and test animals were weighed and handled for at least five days before testing commenced. Their cages were cleaned on the day prior to testing in order to avoid any stress effects on the test day and also to ensure a similar home cage background social odor intensity environment. Smears were made daily (between 08.00 and 09.30) with sterile saline-soaked cotton swabs (toothpicks with the blunt end wrapped in a small amount of cotton wool) being used to harvest cells from the vaginal opening (swab sticks are gently twisted in the vaginal opening to remove cells and then rolled across glass slides for histological staining). Vaginal smears were taken from animals for at least five days before testing to confirm the presence of an ovarian cycle. Females that did not show a complete cycle during this period of smear testing were not used (about 1 in 8 females). Stages of the ovarian cycle were determined by microscopic analysis of vaginal cytology after staining with 1% toluidine blue (w/v) and using the guidelines in Allen (1922). Proestrus was determined by predominantly round nucleated epithelial cells that could be either dispersed or clumped, estrus by primarily the presence of only non-nucleated cornified cells and diestrus/metaestrus by a predominance of leukocyte cells with a few scattered epithelial

116

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

cells. Females were then divided into three groups according to cycle stage: diestrus (including metaestrus), proestrus and estrus. Since αERKO females are unfertile, and in a prolonged anestrus stage, there was only one group of αERKO female mice. Vaginal smears were also taken from these females to confirm the presence of a continuous diestrus-like stage. Females were used for testing no less than 1.5 h after taking the vaginal smears to allow sufficient time for any potential increased excitability in the olfactory bulb following vaginal stimulation to subside (Guevara-Guzman et al., 2000). Social recognition test The protocol consisted of two consecutive days of testing. On day one there was a habituation–dishabituation test (h–d) and on day two a discrimination test. On both days test animals were initially habituated for 10 min to the testing arena, a cubic (33 × 33 × 25 cm) white plastic container with fresh sawdust covering the bottom. Social recognition was tested using sedated (Hypnorm–Hypnovel 6 mg/kg i.p. VetaPharma UK) same-sex C57/Bl6 × 129SV adult mice placed into the centre of the testing arena where the test mouse was present. Stimulus animals usually remained anaesthetized for the duration of the h–d tests, however if they did show any signs of coming round before they were completed they were given a third of the original Hypnorm–Hypnovel dose. On the h–d component of the test, test mice were repeatedly exposed to a sedated stimulus mouse for 1 min: four times (habituation) with the same stimulus animal and then in a final 5th trial (dishabituation) with a different novel stimulus mouse. There was a 10 min interval between each 1 min trial where no stimulus animal was present although the test animal remained in the arena throughout. After the five trials all mice were returned to their home cages overnight. On day two, 24 h after the h–d test, test mice were simultaneously exposed to two sedated stimulus mice (placed around 10 cm apart and at the same orientation) for 2 min, the familiar (used in the habituation trials) and a fresh unfamiliar one. Long-term memory retention was considered to be present if a proportionately higher investigation time of the unfamiliar over the familiar stimulus mouse occurred. All animals were returned to their home cages after testing. Stimulus animals used for the h–d and discrimination tests came from different home cages and different breeding pairs. No stimulus animal was used for more than five test animals. Where female stimulus animals were used no attempt was made to establish the stage of their estrus cycle, with diestrus and metaestrus stages being the most probable. All trials were videotaped using an overhead camera and the time the test animal spent investigating the stimulus animal was scored subsequently using handheld timers. Scoring was done blind of genotype and independently by two trained people. The investigation time data used for analysis was taken as the average of their two cumulative time scores. Social investigation was considered to occur when test animals showed any direct oro-nasal contact with any part of the body of the stimulus mouse and also any periods where the testmouse approached closely (~1 cm) to the stimulus animal and showed sniffing directed towards it. We did not distinguish specifically the times spent investigating different body regions since overall experimental animals showed similar patterns with head, anogenital and flank areas always being sampled. Statistical analysis Data from each test day was analyzed separately. The h–d for females was initially analyzed using a 3-way ANOVA with repeated measures for trial and with factors stage and genotypes. Similar analysis was done for the 24 h discrimination data although in this case the repeated measures were for stimulus animal familiarity. Since αERKO females did not display an ovarian cycle and were in constant anestrus analysis comparisons between αERKO and αWT

females used two 2-way ANOVA and repeated measures for trial/ familiarity. One was done using genotype as a factor and another with group as a factor (three cycle stages of the αWT females (estrus, diestrus and proestrus) and the anestrus αERKO females). Analysis of individual measures within genotypes and cycle stages was then carried out using one or 2-way ANOVAs followed by post-hoc Holm– Sidak tests. Data for males, that had no stage factor, were analyzed using a 2-way ANOVA with repeated measures for either trial or familiarity and with genotype as a factor. Analysis of individual measures within genotypes was then carried out using one-way ANOVAs followed by post-hoc Holm–Sidak tests. All statistical tests were carried out using PASW version 18 (SPSS inc) or SigmaPlot 11 software. Results Wildtype and ERKO females and their ovarian cycles The ovarian cycle of laboratory mice is erratic in duration and therefore timing of each stage is unpredictable. The female mice in this study had cycles of anything from 4 to 10 days long, lasting on average 5 days. Diestrus could last 2 to 5 days and in some cases it lasted 6 or more days — these latter animals with long diestrus periods were removed from the study. Estrus lasted for 1 or 2 days, although in exceptional cases it was sometimes as long as 3 days. The proestrus stage had a more consistent duration of ~1 day. Both male and female αERKO mice display highly atrophied reproductive organs and are infertile (Couse et al., 2003). Female αERKO mice are anovulatory and acyclic (Lindzey and Korach, 1997). In accordance with this, the vaginal cytology study of our animals showed αERKO females having a constant predominance of leukocytes with few scattered cornified epithelial cells, i.e. a persistent anestrus-like state. On the other hand, disruption of ERβ has less severe effects on reproduction with females, described as being ‘sub-fertile’, achieving fewer pregnancies, and with fewer embryos than their WT littermates (Krege et al., 1998). Our βERKO mice showed cyclic vaginal cytology similar to WT females, suggesting the presence of a normal ovarian cycle. Social recognition memory in αERKO and βERKO female mice βERKO females were not significantly impaired compared to βWT animals in either h–d or 24 h retention aspects of the social recognition tests. All females, regardless of their ovarian cycle stage and genotype, showed very similar patterns of h–d (stage: F2,57 = 0.311, p = 0.734; genotype: F1,57 = 1.378, p = 0.245 and stage × genotype: F2,57 = 2.379, p = 0.102; trial: F4,228 = 79.695, p b 0.001; trial × stage: F8,228 = 0.660, p = 0.704; trial× genotype: F4,228 = 0.873, p = 0.481; trial× stage × genotype: F8,228 = 0.989, p = 0.439; Fig. 1) with significantly increased investigation of the unfamiliar, novel animal in trial 5 (trial 4 vs trial 5 with novel animal — βWT: estrus, diestrus and proestrus p b 0.001 in all cases; βERKO: estrus, diestrus and proestrus p b 0.001 in all cases; Fig. 1). On the 24 h discrimination test, both genotypes showed a significant ovarian cycle effect (discrimination × stage, F2,59 = 10.174, p b 0.001; discrimination F1,59 = 1.441, p = 0.235 and stage: F1,59 = 2.763, p = 0.071; Fig. 1), with only the animals in proestrus investigating the unfamiliar animal more than the familiar one (βWT: p = 0.013 and βERKO: p = 0.002; Fig. 1), indicating that the ability to discriminate the familiar animal was still present. βERKO and βWT females in the estrus stage of their cycle spent similar amounts of time investigating both familiar and unfamiliar individuals (βWT: p = 0.225 and βERKO: p = 0.577; Fig. 1) during the 24 h discrimination test, indicating that they had not maintained their ability to recognise the familiar animal. During diestrus, βERKO animals actually showed a small significant increase in investigation of the familiar compared with the unfamiliar stimulus animal (p= 0.046) suggesting some degree of recognition. However, since this was not also seen in βWT animals (p = 0.654; Fig. 1)

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

a Mean investigation time (s)

60

βWT

117

** *

50

##

40

++ 30 20

estrus diestrus

10

proestrus

0 1

2

3

4

5

Trial number

Mean investigation time (s)

b

βERKO

Familiar

Unfamiliar

Stimulus animal

**

60

**

##

50 40

++

+

30 20 10 0 1

2

3

4

Trial number

5

Familiar

Unfamiliar

Stimulus animal

Fig. 1. Social recognition memory in (a) βERKO wildtype (n = 33) (top) and (b) βERKO homozygous (bottom) (n = 32) female mice across the ovarian cycle (diestrous, proestrous and estrous stages). Left panel: mean ± sem investigation times in habituation–dishabituation component of the test where females were presented with an anaesthetized adult stimulus female for 4 successive 1 min periods (habituation) followed by a novel female on the 5th trial (dishabituation). Inter-trial interval = 10 min. Right panel: mean ± sem investigation times for a 24 h discrimination test where the test animal was presented simultaneously with the same familiar stimulus female encountered during the habituation trials and another novel, unfamiliar female. 3 or 2-way ANOVA effect of trial (for h–d) or discrimination (24 h test) — *p b 0.05, **p b 0.01. ##p b 0.01 vs trial 5 with the novel male (Holm–Sidak) and 24 h discrimination comparison within genotype, ++p b 0.01, + p b 0.05 (Holm–Sidak).

the importance of this finding is questionable and difficult to interpret. Stress and anxiety might possibly have affected the motivation to investigate the unfamiliar animal in this case although it had clearly not done so during the dishabituation test. There was no influence of genotype (βERKO homozygous vs wildtype—genotype: F2,59 = 1.916, p = 0.172; Fig. 1) either on the discrimination (discrimination × genotype: F1,59 = 0.131, p = 0.719) or on the ovarian cycle effect (stage × genotype F2,59 = 0.670, p = 0.516; discrimination × stage × genotype: F2,59 = 1.784, p = 0.177) showing that βERKO animals were indistinguishable from WT controls in their performance of this task across the ovarian cycle. Overall, αERKO and αWT females showed a trial effect (F4,26 = 9.363, p b 0.001) and a significant interaction with trial × genotype (F4,26 = 2.805, p = 0.046; Fig. 2). However, only αWT females showed a significantly decreased social investigation response across trials 1–4 and then increased social investigation when presented with the novel animal at trial 5 (trial 4 vs 5: αWT-diestrus, p b 0.05; estrus, p b 0.01; proestrus, p b 0.05). The αERKO females failed to show any h–d behavior, spending similar amounts of time investigating the stimulus animal in all five trials (trial 4 vs 5, p N 0.05, ns). This effect was, however, not dependent upon the cycle stage of the αWTs (group × trial F12,108 = 1.479, p = 0.158). For the 24 h discrimination test, αWT females showed an overall discrimination between the familiar and unfamiliar mice (F1,27 = 6.553, p = 0.016; Fig. 2). Social discrimination however was cycle-stage dependent (discrimination × group F3,27 =3.366, p=0.033), with only proestrus females spending significantly more time investigating the

unfamiliar animal, than the familiar one (p b 0.05). Diestrus and estrus females spent similar amounts of time investigating both stimulus animals (both p N 0.05 ns). αERKO mice also failed to discriminate between stimulus animals at the 24 h social recognition test (pN 0.05 ns). The difference on 24 h retention was not due to genotype (discrimination × genotype F1,29 = 3.134, p = 0.087 and genotype F1,29 = 1.568, p= 0.220). In order to help establish whether the proestrus-dependent effect on 24 h retention of the social recognition memory was entirely due to the animals being in this stage of the cycle during memory formation in the h–d test, a further analysis was carried out. In the two wildtype groups of animals used in the study, seven (3 αWT and 4 βWT) were tested for h–d during diestrus and for 24 h recall during proestrus. Despite being in proestrus during the 24 h recall test, these animals did not show any evidence of memory retention (paired t-test, t6 = −0.610, p = 0.564; mean ± sem investigation times — familiar= 35.2± 3.8 s, unfamiliar = 38.8 ± 5.8 s). It seems unlikely therefore that high estrogen levels during proestrus facilitate discrimination at the recall stage in this recognition paradigm. Social recognition memory in αERKO and βERKO male mice On the h–d test both αERKO and βERKO male mice showed behaviors similar to their respective wildtype (WT) littermates (trial ×genotype: αER F4,108 =1.244, p= 0.297 and βER F4,120 = 0.196, p =0.916; Fig. 3). All males (αERKO, βERKO and their WT littermates) showed significant habituation to the repeatedly presented stimulus mouse and increased

118

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

a

αWT

**

Mean investigation time (s)

60

* ##

50 40 30

+ 20

oestrus diestrus proestrus

10 0 1

2

3

4

5

b Mean investigation time (s)

60

Familiar

Unfamiliar

Stimulus animal

Trial number αERKO

50 40 30 20 10 0 1

2

3

4

Trial number

5

Familiar

Unfamiliar

Stimulus animal

Fig. 2. Social recognition memory in (a) αERKO wildtype (n = 28) (top) and (b) αERKO homozygous (n = 13) (bottom) female mice across the ovarian cycle (diestrous, proestrous and estrous stages) for the wildtypes and for αERKO in anestrous (since they do not show cycles). Left panel: mean ± sem investigation times in the habituation–dishabituation component of the test where females were presented with an anaesthetized adult stimulus female for 4 successive 1 min periods (habituation) followed by a novel female on the 5th trial (dishabituation). Inter-trial interval = 10 min. Right panel: mean ± sem investigation times for the 24 h discrimination test where the test animal was presented simultaneously with the same familiar stimulus female encountered during the habituation trials and another novel, unfamiliar female. 3 or 2-way ANOVA effect of trial (for h–d) or discrimination (24 h test) — *p b 0.05, **p b 0.01. ##p b 0.01 vs trial 5 with the novel male (Holm–Sidak) and 24 h discrimination comparison within genotype, + p b 0.05 (Holm–Sidak).

investigation to a new animal (αER trial: F4,108 = 25.107, p b 0.001 and genotype: F1,27 =0.003, p= 0.956, trial 4 vs trial 5 with novel animal — αWT p= 0.015, αERKO p=0.002; βER trial: F4,120 = 44.11, p b 0.001 and genotype: F1,30 =0.039, p= 0.845, trial 4 vs trial 5 with novel animal — βWT pb 0.001, βERKO p b 0.001). During the 24 h social recognition test, both βERKO and βWT males showed similar increased investigation of the unfamiliar mouse compared to the familiar one (discrimination: F1,30 = 8.535, p = 0.007, genotype F1,30 = 0.757, p = 0.391 and discrimination × genotype: F1,30 = 0.757, p = 0.391; Fig. 3). On the other hand, αWT and αERKO mice showed a significant difference on their discrimination of familiar and unfamiliar mice (discrimination: F1,27 = 13.281, p = 0.001, discrimination × genotype: F 1,27 = 4.719, p = 0.039). While αWT male mice showed a 24 h social recognition by spending significantly more time investigating the unfamiliar stimulus mouse (p b 0.001), αERKO ones failed to do so (p = 0.281), spending the same amount of time investigating both animals. Overall αERKO mice spent slightly more time engaged investigating both of the stimulus animals although this did not quite achieve significance (genotype F1,27 = 3.931, p = 0.058). Discussion Results show for the first time that the long-term (24 h) recall of social recognition memory in female mice is influenced by the stage of their ovarian cycle. Only animals learning during the proestrus stage successfully maintained the memory at 24 h, implying a potential estrogen effect. There does not appear to be any proestrus effect during recall per se since animals tested for 24 h recall during this

stage of the cycle showed no evidence for improved discrimination. This may provide a mechanism whereby a more enduring memory can be formed for individuals encountered during social and sexual interactions taking place when the female can conceive. Females approaching and during estrus, a biologically important time, need to be more efficient at managing their social exchanges, so they can engage more easily in reproductive behavior when the environment predicts success (Morgan et al., 2004). If a socially familiar environment will improve reproductive (Dobson and Baudoin, 2002) and rearing success (Konig, 1994a,b), females will need to remember familiar female conspecifics before engaging in reproductive behaviors. Our results also confirm previous studies using αERKO mice reporting that the αER is required for formation of normal short-term social recognition memory in females (Choleris et al., 2003, 2004, 2006). They differ however in that no evidence for a short-term memory formation deficit was found in βERKO mice (Choleris et al., 2003, 2004, 2006). In the studies shown here, βERKO females also had a proestrus facilitation of long-term memory retention, implying that only the αER is important for this. Finally our results showed that there is a sex difference in dependence upon the αER for social recognition memory in this task. Male αERKO mice, unlike females, showed normal short-term memory formation but, like females, they had impaired recall at 24 h. Ovarian cycle-dependent effects on social recognition memory Our results showing a clear proestrus-dependent facilitation of the duration of an olfactory recognition memory in mice are, to the best of

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

a

119

**

50

Mean investigation time (s)

** 40

##

30 20

αWT αERKO

10 0 1

2

3

4

5

Familiar

Trial number

b

** 50

Mean investigation time (s)

Unfamiliar

Stimulus animal

**

##

40

+ +

30 20

βWT βΕΡΚΟ

10 0

1

2

3

4

Trial number

5

Familiar

Unfamiliar

Stimulus animal

Fig. 3. Social recognition memory in (a) αERKO homozygous (n = 16) and wildtype (n = 13) male mice (top) and (b) βERKO homozygous (n = 18) and wildtype (n = 14) male mice (bottom). Left panel: mean ± sem investigation times in habituation–dishabituation component of the test where males were presented with an anaesthetized adult stimulus male for 4 successive 1 min periods (habituation) followed by a novel male on the 5th trial (dishabituation). Inter-trial interval = 10 min. Right panel: mean ± sem investigation times for the 24 h discrimination test where the test animal was presented simultaneously with the same familiar stimulus male from the habituation trials and another novel, unfamiliar male. 3 or 2-way ANOVA effect of trial (for h–d) or discrimination (24 h test) — *p b 0.05, **p b 0.01, discrimination × genotype †p b 0.05. #p b 0.05 vs trial 5 with the novel male (Holm–Sidak) and 24 h discrimination comparison within genotype, ++p b 0.01, + p b 0.05 (Holm–Sidak).

our knowledge, novel. Facilitation of a longer-term (5 h compared to 2 h) retention of social recognition memory has already been reported in proestrus rats, but only after administration of vaginocervical stimulation (Larrazolo-Lopez et al., 2008). In this however there was no effect of proestrus (Larrazolo-Lopez et al., 2008) or estrus (Engelmann et al., 1998) per se. Similarly, exogenous estradiol (E2) treatment has been shown to prolong social recognition to 2 h in ovariectomized group-housed rats (Hlinak, 1993) and up to 24 h in ovariectomized mice (Tang et al., 2005), although these effects were also dependent upon dose and length of treatment. The enhanced long-term recognition memory was not the result of greater exposure to the olfactory stimulus since ovarian cycle stage had no effect on overall levels of investigation. Also, females spent similar large amounts of time (N80% of the trial) investigating the sedated female mouse during the initial (first ever) encounter. These observations would tend to suggest that there are no major anxiety, perception, attention or motivation variations in investigating a novel intruder across the cycle which might have led to increased recognition duration during proestrus. One might have anticipated that significant changes in any of these components would have led to either reductions or increases in investigation. However, without more extensive tests we cannot entirely rule out some contribution from them. Our use of anaesthetized adults of the same sex as the test animal probably also helped keep attention and motivation levels high since it allowed us to use social stimuli promoting the most

interest while avoiding the problems of having a behaving stimulus (e.g. fighting, chasing, etc.), reducing opportunities to perform olfactory investigation. Previously (data not shown), we tested using juveniles (both sexes) or anaesthetized males as stimulus and found that although they were initially arousing and were thoroughly investigated, females' investigation times dropped dramatically after the first trial. Overall investigation times achieved here, were similar to those reported in studies using this paradigm in conscious animals but where stimulus animals were deliberately chosen which did not provoke fighting (Ferguson et al., 2000; Imwalle et al., 2002). The long-term preservation of social recognition memory when learning is during proestrus (when circulating levels of estrogen are at their highest) may be due to estrogen facilitating neural plasticity changes during learning and/or subsequent consolidation. Since there were no significant cycle-dependent effects on either the habituation or dishabituation phases of the task, it might be argued that initial memory formation was unaffected. However, given that the αERKO females failed to show an h–d behavior, and that ovarian cycledependent effects are likely to be mediated by αERs, it is still possible that the strength of initial memory formation may be increased when learning is during proestrus. In relation to subsequent consolidation, anisomycin-sensitive consolidation of long-term recognition memory has been reported to have two stages at 0–3 h and 6–18 h postlearning (Wanisch and Wotjak, 2008). Proestrus females would have high estrogen levels during both of these consolidation stages. Thus,

120

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

estrogen could be acting particularly on recognition memory consolidation although this might be contributed to by a more robust initial memory formation, which was not revealed by the simple investigation time measures taken during the h–d task. Importance of α- and β-estrogen receptors for formation of social recognition memory Lack of functional expression of the βER gene had no consequence on the short-term social memory in either male or female mice, as seen on the h–d phase of the paradigm. By contrast, αERKO mice showed a gender difference on this short-term social recognition. While αERKO males had a normal h–d response, this was significantly impaired in αERKO females. This is unlikely to be the result of some kind of olfactory perception deficit in αERKO females because their investigation durations were similar to wildtype animals. Previous studies have also shown that αERKO mice have normal olfactory perception in that they can find hidden food. They also appear to have a normal functioning accessory olfactory system, in the context of pregnancy block (Wersinger and Rissman, 2000) and normal sensory perception in other modalities (Fugger et al., 1998). A previous study by Imwalle et al. (2002) also reported a normal habituation response in male αERKO mice although no significant subsequent dishabituation. This difference may have been due to the fact that in our study adult male stimulus animals were used as stimuli rather than ovariectomized females. Potentially male odors may have been of greater interest to males than those of ovariectomized females. However, Imwalle et al. (2002) used 8 habituation trials rather than 4 and perhaps this longer duration in conjunction with using less attractive stimulus animals might have resulted in the αERKO males becoming generally demotivated to investigate even a new stimulus animal. Disruption of the αER gene has previously shown sexually dimorphic effects on non-sexual behaviors. Anxiety and fear, for example, are higher in female αERKO mice (Ogawa et al., 1998) but lower in males compared to their WT controls (Imwalle et al., 2002). This increased anxiety could, in turn, inhibit social behaviors in general (Imwalle et al., 2002). In other learning tests activation of the αER has been shown to have a gender-specific effect on performance in the Morris water maze (Fugger et al., 1998). Overall these findings suggest that there may be some organizational differences between the sexes occurring during development. It is possible that the gender differences seen here could be vasopressin (AVP) or OT related. AVP acts on the lateral septum to modulate social recognition (Bielsky et al., 2004; Engelmann and Landgraf, 1994; Engelmann et al., 1998) and is much more abundant in the lateral septum of the male compared with the female rat (De Vries et al., 1981). Our h–d findings in females in the social recognition task differ from the ones initially reported by Choleris et al. (2003, 2004) where they found profound social recognition impairments in αERKO, βERKO and OTKO females. However, in a further study using a social discrimination protocol they found that OT, αER and βER were important, but to different degrees. When given a choice, OTKO and αERKO female mice had “completely impaired” social discrimination, while in βERKO animals it was only reduced (Choleris et al., 2006). These latter findings are closer to ours, although we found no impairments at all in βERKO animals. These differences between our results and those of the Choleris group may reflect the different experimental protocols used. They used a conscious ovariectomized female mouse in a perforated Plexiglas cylinder as a stimulus to prevent direct interaction between the two animals. This however restricts sampling of olfactory cues to odors passing through the holes during the tests and the cylinder containing the stimulus mouse was removed in between them. In our paradigm both direct and indirect sampling of odors would have occurred during the tests and residual odors from the stimulus animal

would have remained in the bedding in the arena throughout the whole period of testing, even though the stimulus animal was only actually present for four 1-minute trials. Thus in our paradigm odor sampling duration would have been longer and animals would additionally have had access to direct odor cues. Social investigation involves close inspection including direct physical contact and licking of the anogenital region, implying that both non-volatile and volatile odors may be being utilized as recognition cues when direct contact is permitted (Ferguson et al., 2000; Insel and Fernald, 2004; Luo et al., 2003; Steel and Keverne, 1985). Single neuron recordings in the accessory olfactory bulb of freely behaving male mice have found that neuronal firing is strongly modulated by physical contact with anaesthetized conspecifics (male and female) (Luo et al., 2003). Therefore, a possible explanation for why we failed to find any βERdependent effects on h–d of social recognition may be because the containment cylinder paradigm made the task much harder by reducing both the quality and quantity of the odor stimuli received. Thus βER-dependent effects in this task might only be revealed when mice are exposed to sub-optimal recognition conditions. There is some evidence that formation of social recognition memory can involve the entorhinal cortex/hippocampus (Sànchez-Andrade and Kendrick, 2009) and other studies have shown that βERs play a more important role in learning tasks involving the hippocampus (Rissman, 2008). It is therefore possible that the role of βERs and the hippocampus becomes of greater importance in some specific recognition paradigms, or as task complexity increases.

Importance of α- and β-estrogen receptors for 24 h retention of social olfactory memory The social recognition paradigm has commonly been used as a test of short-term memory not lasting more than 120 min (Dluzen et al., 1998; Engelmann et al., 1998). Early experiments used individuallyhoused test animals in order to create a sense of territoriality. It was, however, this social isolation that limited the memory duration, since only group-housed mice showed long-term (N24 h) social recognition (Kogan et al., 2000). Our results show that while lack of a βER does not affect this longer-term social olfactory memory, lack of a αER completely prevents it in males even though they have normal h–d responses. Estrogen may therefore help prolong long-term olfactory recognition memory via activation of αER in both sexes. Although estrogen treatment can recover impaired 24 h retrieval of inhibitory avoidance learning in ovariectomized αERKO mice (Fugger et al., 2000), and gonadally intact αERKO males and females have chronically elevated concentrations of estrogen (Couse et al., 2003), these elevated estrogen concentrations clearly did not help with 24 h retention in our social recognition test. Exogenous estrogen treatment may have differential effects on different learning and memory tasks through its two receptors. For example, deletion of αER but not βER impaired performance of ovariectomized females on an inhibitory avoidance (emotional memory) task at 24 h which could be rescued by estradiol treatment (Fugger et al., 2000). However, in the Morris water maze, estradiol treatment had a dose-dependent effect on ovariectomized WT and βERKO females, but no effect on αERKO ones, (Fugger et al., 1998; Rissman et al., 2002). Likewise βER, but not αER, selective agonists improved performance of rats in the water maze, although it was the other way round for the inhibitory avoidance task (Rhodes and Frye, 2006). Hippocampal-dependent social transmission of food preference is also improved in ovariectomized female mice by βER but not αER agonists (Clipperton et al., 2008). From these findings it has been suggested that αER is needed for amygdala-dependent emotional learning, whereas hippocampal-dependent spatial learning requires βER (Rissman 2008). Estrogen can also influence different learning

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

strategies and its effect on cognition may be task-specific (Korol, 2004; Zurkovsky et al., 2007). Olfactory memory formation and short-term social recognition require the involvement of several brain areas, the olfactory bulb (OB), piriform and entorhinal cortices, medial amygdala and hippocampus (Ferguson et al., 2001; Richter et al., 2005). Oxytocin in the medial amygdala of mice is important for short-term social recognition memory. Infusion of OT into this region in OTKO mice recovers social recognition while OT antagonists impair it in wildtype animals (Ferguson et al., 2001). However, OT might also be acting at the level of the OB since infusion of an OT antagonist into the OB also prevents enhancement of social recognition by stimulation of the vagina and cervix in proestrus animals (Larrazolo-Lopez et al., 2008). OT infused in the OB can also extend the duration of the social memory to 120 min (Dluzen et al., 2000). Interestingly, conditional OT receptor knockout mice where expression is completely spared in the amygdala, but only partially in the OB, recognise their species but not individuals. Animals with a total knockout of the gene throughout development are more severely impaired (Lee et al., 2008) although can recognise different outbred mouse strains (Takayanagi et al., 2005). Together these results suggest that OT acts both in the OB and amygdala to facilitate social recognition, although other brain regions may also be involved. How and where might estrogens be acting in this olfactory recognition network? Estrogens promote transcription of OTR in the medial amygdala via ERα (Dekloet et al., 1986; Shughrue et al., 1999, 2002; Young et al., 1998). ERα is also expressed in the OB (Mitra et al., 2003), as well as OT and OTR (Broad et al., 1993, 1999; Larrazolo-Lopez et al., 2008; Yoshimura et al., 1993), and estradiol has also been shown to increase OT activity in this region (Broad et al., 1993). Estrogen may therefore be acting via ERα to facilitate retention of olfactory recognition memory by altering the expression of both OT and its receptor in the medial amygdala and the OB. This is supported by a recent study showing that viral vector reduced expression of ERα in the medial amygdala of estradiol-treated ovariectomized female rats prevented normal social recognition memory (Spiteri et al., 2010). At this stage it is still difficult to determine the precise role for estrogen and OT in specific control of olfactory social recognition memory. While animals with deficient ER and OT signalling often show profound deficits in social recognition memory these appear to be influenced to some extent by different testing paradigms, the type of stimulus animal required to be recognised and the brain regions targeted. It seems reasonable to hypothesize that neither ER nor OT signalling is essential for social recognition memory per se although they clearly have much stronger modulatory influences on the neural substrates processing social odors and their rewarding properties in comparison with other non-social kinds of olfactory cues. It is also clear that αERs, in conjunction with OT, probably play the major role in influencing social recognition memory in paradigms which primarily involve the medial amygdala and OB. However, in some social recognition contexts there may also be some involvement of βERs at the level of the hippocampus. In summary, the studies presented here show that there are clear ovarian-cycle-dependent influences on the duration of olfactory social recognition memory in female mice. The proestrus stage of the cycle, when estrogen levels are at their highest and animals are most receptive to mating males, is when memory for another individual's odors is most enduring. In females lacking the αER memory formation itself is impaired, whereas in males while the memory is formed it is not maintained 24 h later. In contrast to some other studies neither male nor female βERKO animals showed social recognition memory impairments, and βERKO females showed the normal enhanced duration of memory during proestrus. Thus while different paradigms or task difficulty might result in a greater involvement of the βER in memory formation, the proestrus-dependent facilitation of social recognition memory duration we have observed would appear to exclusively involve the αER.

121

References Ågmo, A., Choleris, E., Kavaliers, M., Pfaff, D.W., Ogawa, S., 2008. Social and sexual incentive properties of estrogen receptor alpha, estrogen receptor beta, or oxytocin knockout mice. Genes Brain Behav. 7, 70–77. Allen, E., 1922. The oestrus cycle of the mouse. Am. J. Anat. 30, 297–371. Alves, S.E., Weiland, N.G., Hayashi, S., McEwen, B.S., 1998. Immunocytochemical localization of nuclear estrogen receptors and progestin receptors within the rat dorsal raphe nucleus. J. Comp. Neurol. 391, 322–334. Bielsky, I.F., Hu, S.B., Szegda, K.L., Westphal, H., Young, L.J., 2004. Profound impairment in social recognition and reduction in anxiety-like behaviour in vasopressin V1a receptor knockout mice. Neuropsychopharmacology 29, 483–493. Broad, K.D., Kendrick, K.M., Sirinathsinghji, D.J.S., Keverne, E.B., 1993. Changes in oxytocin immunoreactivity and mRNA expression in the sheep brain during pregnancy, parturition and lactation and in response to oestrogen and progesterone. J. Neuroendocrinol. 5, 435–444. Broad, K.D., Levy, F., Evans, G., Kimura, T., Keverne, E.B., Kendrick, K.M., 1999. Previous maternal experience potentiates the effect of parturition on oxytocin receptor mRNA expression in the paraventricular nucleus. Eur. J. Neurosci. 11, 3725–3737. Choleris, E., Gustafsson, J.A., Korach, K.S., Muglia, L.J., Pfaff, D.W., Ogawa, S., 2003. An estrogen dependent four-gene micronet regulating social recognition: a study with oxytocin and estrogen receptor-alpha and beta knockout mice. Proc. Natl Acad. Sci. USA 100, 6192–6197. Choleris, E., Kavaliers, M., Pfaff, D.W., 2004. Functional genomics of social recognition. J. Neuroendocrinol. 16, 383–398. Choleris, E., Ogawa, S., Kavaliers, M., Gustafsson, J.A., Korach, K.S., Muglia, L.J., Pfaff, D.W., 2006. Involvement of estrogen receptor α, β and oxytocin in social discrimination: a detailed behavioural analysis with knockout female mice. Genes Brain Behav. 5, 528–539. Choleris, E., Little, S.R., Mong, J.A., Puram, S.V., Langer, R., Pfaff, D.W., 2007. Microparticle-based delivery of oxytocin antisense DNA in the medial amygdala blocks social recognition in female mice. Proc. Natl Acad. Sci. USA 104, 4670–4675. Clipperton, A.E., Spinato, J.M., Chernets, C., Pfaff, D.W., Choleris, E., 2008. Differential effects of estrogen receptor alpha and beta specific agonists on social learning of food preferences in female mice. Neuropsychopharmacology 33, 2362–2375. Couse, J.F., Yates, M.M., Walker, V.R., Korach, K.S., 2003. Characterization of the hypothalamic–pituitary–adrenal axis in estrogen receptor (ER) null mice reveals hypergonadism and endocrine sex reversal in females lacking ER alpha but not ER beta. Mol. Endocr. 17, 1039–1053. Cyr, M., Calon, F., Morisette, M., Di Paolo, T., 2002. Estrogenic modulation of brain activity: implications for schizophrenia and Parkinson's disease. J. Psychiatry Neurosci. 27, 12–27. Daniel, J.M., 2006. Effects of oestrogen on cognition: what have we learned from basic research? J. Neuroendocrinol. 18, 787–795. De Vries, G.J., Buijs, R.M., Swaab, D.F., 1981. Ontogeny of the vasopressinergic neurons of the suprachiasmatic nucleus and their extrahypothalamic projections in the rat brain — presence of a sex difference in the lateral septum. Brain Res. 218, 67–78. Dekloet, E.R., Voorhuis, D.A.M., Boschma, Y., Elands, J., 1986. Estradiol modulates density of putative oxytocin receptors in discrete rat-brain regions. Neuroendocrinology 44, 415–421. Di Paolo, T., Rouillard, C., Bedard, P., 1985. 17-Beta-estradiol at a physiological dose acutely increases dopamine turnover in rat brain. Eur. J. Pharmacol. 117, 197–203. Dluzen, D.E., Muraoka, S., Engelmann, M., Landgraf, R., 1998. The effects of infusion of arginine vasopressin and oxytocin or their antagonists into the olfactory bulb upon social recognition responses in male rats. Peptides 19, 999–1005. Dluzen, D.E., Muraoka, S., Engelmann, M., Ebner, K., Landgraf, R., 2000. Oxytocin induces preservation of social recognition in male rats by activating α-adrenoreceptors of the olfactory bulb. Eur. J. Neurosci. 12, 760–766. Dobson, F.S., Baudoin, C., 2002. Experimental tests of spatial association and kinship in monogamous mice (Mus spicilegus) and polygynous mice (Mus musculus domesticus). Can. J. Zool. 80, 980–986. Dohanich, G.P., 2002. Gonadal steroids, learning, and memory. In: Pfaff, D.W., Arnold, A.P., Etgen, A.M., Fahrbach, S.E., Rubin, R.T. (Eds.), Hormones, Brain and Behavior, vol. 2. Academic Press (Elsevier Science), San Diego, pp. 265–327. Engelmann, M., Landgraf, R., 1994. Microdialysis administration of vasopressin into the septum improves social recognition in Brattleboro rats. Physiol. Behav. 55, 145–149. Engelmann, M., Ebner, K., Wotjak, C.T., Landgraf, R., 1998. Endogenous oxytocin is involved in short-term olfactory memory in female rats. Behav. Brain Res. 90, 315–321. Ferguson, J.N., Young, L.J., Hearn, E.F., Matzuk, M.M., Insel, T.R., Winslow, J.T., 2000. Social amnesia in mice lacking the oxytocin gene. Nature 25, 284–288. Ferguson, J.N., Aldag, J.M., Insel, T.R., Young, L.J., 2001. Oxytocin in the medial amygdala is essential for social recognition in the mouse. J. Neurosci. 21, 8278–8285. Frankfurt, M., 1990. Gonadal steroids modify dendritic spine density in ventromedial hypothalamic neurons. Neuroendocrinology 51, 530–535. Frick, K.M., Berger-Sweeney, J., 2001. Spatial reference memory and neocortical neurochemistry vary with the estrous cycle in C57Bl/6 mice. Behav. Neurosci. 115, 229–237. Frick, K.M., Fernandez, S.M., Bulinski, S.C., 2002. Estrogen replacement improves spatial reference memory and increases hippocampal synaptophysin in aged female mice. Neuroscience 115, 547–558. Fugger, H.N., Cunningham, S.G., Rissman, E.F., Foster, T.C., 1998. Sex differences in the activational effect of ERα on spatial learning. Horm. Behav. 34, 163–170. Fugger, H.N., Foster, T.C., Gustafsson, J.A., Rissman, E.F., 2000. Novel effects of estradiol and estrogen receptor alpha and beta on cognitive function. Brain Res. 883, 258–264.

122

G. Sánchez-Andrade, K.M. Kendrick / Hormones and Behavior 59 (2011) 114–122

Gheusi, G., Bluthé, R.-M., Goodall, G., Dantzer, R., 1994. Social and individual recognition in rodents: methodological aspects and neurobiological bases. Behav. Proc. 33, 59–88. Gibbs, R.B., 1996. Fluctuations in the relative levels of choline acetyltransferase mRNA in different regions of the rat basal forebrain across the estrous cycle: effects of estrogen and progesterone. J. Neurosci. 16, 1049–1055. Gibbs, R.B., 1997. Effects of estrogen on basal forebrain cholinergic neurons vary as a function of dose and duration of treatment. Brain Res. 757, 10–16. Greco, B., Allegretto, E.A., Tetel, M.J., Blaustein, J.D., 2001. Coexpression of ER beta with ER alpha and progestin receptor proteins in the female rat forebrain: effects of estradiol treatment. Endocrinology 142, 5172–5181. Guevara-Guzman, R., Barrera-Mera, B., De La Riva, C., Kendrick, K.M., 2000. Release of classical transmitters and nitric oxide in the rat olfactory bulb, evoked by vaginocervical stimulation and potassium, varies with the oestrus cycle. Eur. J. Neurosci. 12, 80–88. Hall, J.M., Couse, J.F., Korach, K.S., 2001. The multifaceted mechanisms of estradiol and estrogen receptor signalling. J. Biol. Chem. 276, 36869–36872. Healy, S.D., Braham, S.R., Braithwaite, V.A., 1999. Spatial working memory in rats: no differences between the sexes. Proc. R. Soc. Lond. Biol. B 266, 2303–2308. Hlinak, Z., 1993. Social recognition in ovariectomized and estradiol-treated female rats. Horm. Behav. 27, 159–166. Holmes, M.M., Wide, J.K., Galea, L.A., 2002. Low levels of estradiol facilitate, whereas high levels of estradiol impair, working memory performance on the radial arm maze. Behav. Neurosci. 116, 928–934. Imwalle, D.B., Scordalakes, E.M., Rissman, E.F., 2002. Estrogen receptor alpha influences socially motivated behaviours. Horm. Behav. 42, 484–491. Insel, T.R., Fernald, R.D., 2004. How the brain processes social information: searching for the social brain. Ann. Rev. Neurosci. 27, 549–579. Kavaliers, M., Ågmo, A., Choleris, E., Gustafsson, J.A., Korach, K.S., Muglia, L.J., Pfaff, D.W., 2004. Oxytocin and estrogen receptor alpha and beta knockout mice provide discriminably different odor cues in behavioural assays. Genes Brain Behav. 3, 189–195. Kendrick, K.M., 1981. Effects of testosterone and the oestrous cycle on neuronal refractory periods and firing rates of stria terminalis in the female rat. Exp. Brain Res. 44, 331–336. Kogan, J.H., Frankland, P.W., Silva, A.J., 2000. Long-term memory underlying hippocampus-dependent social recognition in mice. Hippocampus 10, 45–56. Konig, B., 1994a. Fitness effects of communal rearing in house mice — the role of relatedness versus familiarity. Anim. Behav. 48, 1449–1457. Konig, B., 1994b. Components of lifetime reproductive success in communally and solitarily nursing house mice — a laboratory study. Bevah Ecolo Sociociol 34, 275–283. Korol, D.L., 2004. Role of estrogen in balancing contributions from multiple memory systems. Neurobiol. Learn. Mem. 82, 309–323. Krege, J.H., Hodgin, J.B., Couse, J.F., Enmark, E., Warner, M., Mahler, J.F., Sar, M., Korach, K.S., Gustafsson, J.A., Smithies, O., 1998. Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc. Natl Acad. Sci. USA 95, 15677–15682. Larrazolo-Lopez, A., Kendrick, K.M., Aburto-Arciniega, M., Arriaga-Avila, V., Morimoto, S., Frias, M., Guevara-Guzman, R., 2008. Vaginocervical stimulation enhances social recognition memory in rats via oxytocin release in the olfactory bulb. Neuroscience 152, 585–593. Lee, H.-J., Caldwell, H.K., Macbeth, A.H., Tolu, S.G., Young, W.S., 2008. A conditional knockout mouse line of the oxytocin receptor. Endocrinology 149, 3256–3263. Leuner, B., Mendolia-Loffredo, S., Shors, T.J., 2004. High levels of estrogen enhance associative memory formation of ovariectomized females. Psychoneuroendocrinology 29, 883–890. Lindzey, J., Korach, K.S., 1997. Developmental and physiological effects of estrogen receptor gene disruption in mice. Trends Endocr. Met. 8, 137–145. Liu, F., Day, M., Muniz, L.C., Bitran, D., Arias, R., Revilla-Sanchez, R., Grauer, S., Zhang, G., Kelley, C., Pulito, V., Sung, A., Mervis, R.F., Navarra, R., Hirst, W.D., Reinhart, P.H., Marquis, K.L., Moss, S.J., Pangalos, M.N., Brandon, N.J., 2008. Activation of estrogen receptor-b regulates hippocampal synaptic plasticity and improves memory. Nat. Neurosci. 11, 334–343. Luo, M., Fee, M.S., Katz, L.C., 2003. Encoding pheromonal signals in the accessory olfactory bulb of behaving mice. Science 299, 1196–1201. Maggi, A., Ciana, P., Belcredito, S., Vegeto, E., 2004. Estrogens in the nervous system: mechanisms and non reproductive functions. Ann. Rev. Physiol. 66, 291–313. Mitra, S.W., Hoskin, E., Yudkovitz, J., Pear, L., Wilkinson, H.A., S.H., Pfaff, D.W., Ogawa, S., Rohrer, S.P., Schaeffer, J.M., McEwen, B.S., Alves, S.E., 2003. Immunolocalization of estrogen receptor beta in the mouse brain: comparison with estrogen receptor alpha. Endocrinology 144, 2055–2067. Morgan, M.A., Schulkin, J., Pfaff, D.W., 2004. Estrogens and non-reproductive behaviors related to activity and fear. Neurosci. Biobehav. Rev. 28, 55–63. Nomura, M., McKenna, E., Korach, K.S., Pfaff, D.W., Ogawa, S., 2002. Estrogen receptorbeta regulates transcript levels for oxytocin and arginine vasopressin in the hypothalamic paraventricular nucleus of male mice. Mol. Brain Res. 109, 84–94. Nomura, M., Korach, K.S., Pfaff, D.W., Ogawa, S., 2003. Estrogen receptor beta (ER beta) protein levels in neurons depend on estrogen receptor alpha, (ER alpha) gene expression and on its ligand in a brain region-specific manner. Mol. Brain Res. 110, 7–14. Ogawa, S., Eng, V., Taylor, J., Lubahn, D.B., Korach, K.S., Pfaff, D.W., 1998. Roles of estrogen receptor-alpha gene expression in reproduction-related behaviors in female mice. Endocrinology 139, 5070–5081. Osterlund, M.K., Kuiper, G., Gustafsson, J.A., Hurd, Y.L., 1998. Differential distribution and regulation of estrogen receptor-alpha and -beta mRNA within the female rat brain. Mol. Brain Res. 54, 175–180.

Owens, W., Asby, J., Odum, J., Onyon, L., 2003. The OECD program to validate the rat uterotrophic bioassay. Phase 2: dietary phytoestrogen analyses. Env. Health Perspect. 111, 1559–1567. Patisaul, H.B., Whitten, P.L., Young, L.J., 1999. Regulation of estrogen receptor beta mRNA in the brain: opposite effects of 17 beta-estradiol and the phytoestrogen, coumestrol. Mol. Brain Res. 67, 165–171. Rado, A., Crystle, C.D., Townsley, J.D., 1970. Concentration of estrogens in maternal peripheral plasma in late pregnancy, during labor and post-partum. J. Clin. Endocr. Met. 30, 497–503. Rhodes, M.E., Frye, C.A., 2006. ERβ-selective SERMs produce mnemonic-enhancing effects in the inhibitory avoidance and water maze tasks. Neurobiol. Learn. Mem. 85, 183–191. Richter, K., Wolf, G., Engelmann, M., 2005. Social recognition memory requires two stages of protein synthesis in mice. Learn. Mem. 12, 407–413. Rissanen, A., Puolivali, J., van Groen, T., Riekkinen, P.J., 1999. In mice tonic estrogen replacement therapy improves non-spatial and spatial memory in a water maze task. NeuroReport 10, 1369–1372. Rissman, E.F., 2008. Roles of oestrogen receptors α and β in behavioural neuroendocrinology: beyond yin/yang. J. Neuroendocrinol. 20, 873–879. Rissman, E.F., Heck, A.L., Leonard, J.E., Shupnik, M.A., Gustafsson, J.-A., 2002. Disruption of the estrogen receptor β gene impairs spatial learning in female mice. Proc. Natl Acad. Sci. USA 99, 3996–4001. Sànchez-Andrade, G., Kendrick, K.M., 2009. The main olfactory system and social learning in mammals. Behav. Brain Res. 200, 323–335. Sawada, H., Ibi, M., Kihara, T., Urushitani, M., Akaike, A., Shimohama, S., 1998. Estradiol protects mesencephalic dopaminergic neurons from oxidative stress-induced neuronal death. J. Neurosci. Res. 54, 707–719. Shughrue, P.J., Lane, M.V., Merchenthaler, I., 1997a. Comparative distribution of estrogen receptor-alpha and -beta mRNA in the rat central nervous system. J. Comp. Neurol. 388, 507–525. Shughrue, P.J., Scrimo, P., Lane, M.V., Askew, R., Merchenthaler, I., 1997b. The distribution of estrogen receptor-beta mRNA in forebrain regions of the estrogen receptor-alpha knockout mouse. Endocrinology 138, 5649–5652. Shughrue, P.J., Lane, M.V., Merchenthaler, I., 1999. Biologically active estrogen receptorbeta: evidence from in vivo autoradiographic studies with estrogen receptor alphaknockout mice. Endocrinology 140, 2613–2620. Shughrue, P.J., Askew, G.R., Dellovade, T.L., Merchenthaler, I., 2002. Estrogen-binding sites and their functional capacity in estrogen receptor double knockout mouse brain. Endocrinology 143, 1643–1650. Spiteri, T., Ågmo, A., 2009. Ovarian hormones modulate social recognition in female rats. Physiol. Behav. 98, 247–250. Spiteri, T., Musatov, S., Ogawa, S., Ribeiro, A., Pfaff, D.W., Ågmo, A., 2010. The role of the estrogen receptor α in the medial amygdala and ventromedial nucleus of the hypothalamus in social recognition, anxiety and aggression. Behav. Brain Res. 210, 211–220. Steel, E., Keverne, E.B., 1985. Effect of female odor on male hamsters mediated by the vomeronasal organ. Phys. Behav. 35, 195–200. Takayanagi, Y., Yoshida, M., Bielsky, I.F., Ross, H.E., Kawamata, M., Onaka, T., Yanagisawa, T., Kimura, T., Matzuk, M.M., Young, L.J., Nishimori, K., 2005. Pervasive social deficits, but normal parturition in oxytocin receptor-deficient mice. Proc. Natl Acad. Sci. USA 102, 16096–16101. Tang, A.C., Nakazawa, M., Romeo, R.D., Reeb, B.C., Sisti, H., McEwen, B.S., 2005. Effects of long-term estrogen replacement on social investigation and social memory in ovariectomized C57BL/6 mice. Horm. Behav. 47, 350–357. Toran-Allerand, C.D., Singh, M., Setalo, G., 1999. Novel mechanisms of estrogen action in the brain: new players in an old story. Front. Neuroendocrinol. 20, 97–121. Wanisch, K., Wotjak, C.T., 2008. Time course and efficiency of protein synthesis inhibition following intracerebral and systemic anisomycin treatment. Neurobiol. Learn. Mem. 90, 485–494. Warren, S.G., Juraska, J.M., 1997. Spatial and non-spatial learning across the rat's estrous cycle. Behav. Neurosci. 111, 259–266. Wersinger, S.R., Rissman, E.F., 2000. Oestrogen receptor alpha is essential for female-directed chemo-investigatory behavior but is not required for the pheromone-induced luteinizing hormone surge in male mice. J. Neuroendocrinol. 12, 103–110. Wide, J.K., Hanratty, K., Ting, J., Galea, L.A., 2004. High level estradiol impairs and low level estradiol facilitates non-spatial working memory. Behav. Brain Res. 155, 45–53. Wise, P.M., Dubal, D.B., Wilson, M.E., Rau, S.W., Bottner, M., Rosewell, K.L., 2001. Estradiol is a protective factor in the adult and aging brain: understanding of mechanisms derived from in vivo and in vitro studies. Brain Res. Rev. 37, 313–319. Wooley, C.S., McEwen, B.S., 1994. Estradiol regulates hippocampal dendritic spine density via an N-methyl-D-aspartate-receptor-dependent mechanism. J. Neurosci. 14, 7680–7687. Wooley, C.S., Gould, E., Frankfurt, M., McEwen, B.S., 1990. Naturally-occurring fluctuation in dendritic spine density on adult hippocampal pyramidal neurons. J. Neurosci. 10, 4035–4039. Yoshimura, R., Kiyama, H., Kimura, T., Araki, T., Maeno, H., Tanizawa, O., Tohyama, M., 1993. Localization of oxytocin messenger ribonucleic acid in the rat brain. Endocrinology 133, 1239–1247. Young, L.J., Wang, Z.X., Donaldson, R., Rissman, E.F., 1998. Estrogen receptor alpha is essential for induction of oxytocin receptor by estrogen. NeuroReport 9, 933–936. Zurkovsky, L., Brown, S.L., Boyd, S., Fell, J.A., Korol, D.L., 2007. Estrogen modulates learning in female rats by acting directly at distinct memory systems. Neuroscience 144, 26–37.