Seizure: European Journal of Epilepsy
Volume 16, Issue 1 , Pages 59-68, January 2007

Short-term social recognition memory deficit and atypical social and physiological stressor reactivity in seizure-susceptible El mice

Department of Psychology, Boston College, 140 Commonwealth Avenue, Chestnut Hill, MA 02467, USA

Received 15 August 2006; received in revised form 22 October 2006; accepted 25 October 2006. published online 20 November 2006.

Article Outline

Summary 

The present studies characterize working memory capabilities in the El mouse model of epilepsy using a species-typical social recognition memory task. As the El mouse exhibits a stress hyper-reactivity phenotype, the impact of hypertonic saline consumption, a memory modulatory treatment, upon social recognition performance was also examined. The hypotheses under test were: (1) that seizure susceptible El mice would perform poorly in the short-term working memory task relative to seizure resistant ddY controls, and (2) that the behavioral and neural responses to stressor exposure would be atypical in El mice. Results revealed a short-term working memory deficit and altered reactivity to social, environmental, and physiological stressors in El mice. In Experiment 1, El mice exhibited poor sociability and decreased olfactory investigation times, both anxiogenic-like traits, compared to ddY controls. In Experiment 2, El mice exhibited poor working memory performance compared to capable performance in ddY controls. Social recognition memory in ddY mice was abolished, however, by salt-loading whereas El mice were unaffected by exposure to this physiological stressor. In Experiment 3, all salt-loaded mice exhibited enhanced brain stress neuropeptide (corticotropin releasing factor–CRF) content, and salt-loaded El mice exhibited a 70% reduction in handling-induced seizures. These findings suggest that El mice exhibit high emotionality as well as atypical reactions to stressor exposure, and that these characteristics impact social working memory performance and seizure susceptibility.

Keywords: Learning, Memory, Stress, Epilepsy, Mouse, Social recognition

 

Back to Article Outline

Introduction 

Human epilepsy and seizure management using anti-epileptic drugs are often accompanied by some degree of cognitive decline, typically expressed as learning or memory deficits. Cognitive deficits have developmental significance in diagnosing seizure susceptibility,1, 2, 3 are relevant for seizure prognosis,4 and likely reveal much about hippocampal brain mechanisms that underlie seizure onset.5, 6 Similarly, certain anticonvulsant drugs are prized for their ability to suppress seizure incidence or severity without exerting learning and memory-related side effects.7 Finally, children with epilepsy, as a group, have a greater risk for developing learning problems as co-morbid disorders.8 Clinical evidence for co-morbidity of seizure and learning disorders is consistent with findings in animal models of epilepsy.9

Current animal models of epilepsy suggest that derangements in basic non-associative forms of learning and memory are related to seizure onset. For example, rats treated with the convulsant pilocarpine exhibit no signs of habituation, a simple non-associative form of learning and memory, suggesting the presence of multi-task amnesia which is a characteristic of complex partial status epilepticus.10 Similarly, seizure-prone El mice differ from ddY controls in their locomotor habituation to a novel environment within the same 10-min session and across the 3 days of testing, with El mice exhibiting higher activity in both cases.11 In the case of between-session trials, the lack of diminished activity over time in El mice suggests a basic deficit in non-associative learning capacity. Taken together, these studies provide evidence that changes over time in reactivity to a particular environmental context can serve as a sensitive marker for cognitive ability in multiple animal models of seizure susceptibility. However, performance in tasks sensitive to learning/memory phenomena is notoriously susceptible to a wide variety of non-specific motoric and motivational influences.12 Thus, it would be desirable to assess learning/memory capabilities in a seizure-prone animal model of epilepsy using an unbiased, naturalistic dependent measure of working memory with built-in manipulation checks to guard against non-specific alterations in task performance.13 The present studies were therefore designed to characterize working memory capabilities in the El mouse model of epilepsy using the species-typical social recognition memory task.

Social recognition in rodents is charterized by a pattern of olfactory investigative behaviors, particularly sniffing and nosing the ano-genital region of the stimulus animal.14 Adult rodents exposed during a brief test to unfamiliar, conspecific juveniles spend less time investigating the juveniles following a delay and subsequent re-exposure to the very same stimulus animals. The contrast between the amount of social investigation on exposure to an unfamiliar juvenile provides a quantitative index of non-specific motoric, motivational, or affective components relative to the memory-specific component measured in the familiar juvenile condition. As the El mouse exhibits a stress hyper-reactivity phenotype,15 the present studies also examined the impact of exposure to a physiological stressor, namely, hyper-osmotic saline consumption,16 on social recognition memory performance and brain stress neuropeptide (corticotropin-releasing factor–CRF 17) levels. The specific hypotheses under test were that seizure-susceptible El mice would perform poorly in the short-term working memory task relative to non-susceptible ddY controls and that behavioral and neural responses to stressor exposure would be atypical in the El mouse strain.

Back to Article Outline

Methods 

Animals 

Mice of the El (n=88) and ddY (n=96) strains were descendants of animals obtained from J. Suzuki at the Tokyo Institute of Psychiatry (breeding stock for these strains was kindly provided by Dr. Thomas Seyfried). Experiment 1 assessed working memory performance in El (n=20, 12 females and 8 males) and ddY (n=24, 12 females and 12 males) strain mice. Experiment 2 explored the impact of a physiological stressor on working memory performance using El (n=16, 8 females and 8 males) and ddY (n=16, 8 females and 8 males) strain mice. Experiment 3 examined the impact of a physiological stressor on seizure susceptibility and brain stress neuropeptide activation using El (n=16, 11 females and 5 males) and ddY (n=16, 7 females and 9 males) strain mice. All adult mice were at least 120 days of age at the time of testing and a matched set (strain×gender) of juvenile mice between 25 and 35 days of age were employed as stimulus animals in the social recognition memory studies. Mice were group-housed using a reverse 12-h light/dark cycle (lights off 1000–2200h) and colony temperature/humidity of 21°C/48%. For Experiment 1, mice were provided with food (ProLab 3000, LabDiets, Richmond, IN) and water ad libitum, except during social recognition testing. Routine husbandry was conducted weekly, except when experiments were in progress, and involved minimal handling. Seizure susceptibility induced by handling for routine husbandry was logged for mice of the El strain so that matched treatment groups could be selected. For Experiment 3, only seizure-susceptible El mice were selected. All testing was conducted during the dark phase of the circadian cycle, when mice are awake and active. All experimental procedures described were approved by the Institutional Animal Care and Use Committee of Boston College.

Experiment 1—Assessment of social recognition memory performance in El mice 

Social recognition memory testing 

Data collection was performed under low-light conditions with minimal background noise in a dedicated room separate from the colony. Adult mice were placed singly in holding cages and allowed 30min to acclimate to the novel test setting prior to data collection. The protocol for social recognition testing was adapted from Current Protocols in Neuroscience 18 and employed two conditions of social interaction: (1) a familiar juvenile condition, and (2) a novel juvenile condition. Using a within subjects design, one juvenile condition was implemented on the first day, and following 24h, the second juvenile condition was tested. Experimental mice were paired with different juveniles from day-to-day to prevent carryover effects of juvenile familiarity. In Trial 1 of the familiar juvenile condition, each naïve adult test mouse was paired with a conspecific juvenile. In Trial 2, the adult mouse was re-exposed to the same juvenile with which they were paired initially. Trials 1 and 2 were separated by a 30-min inter-exposure interval, a delay over which unimpaired rodents typically exhibit competent performance in this task.19 In the unfamiliar juvenile condition, each adult mouse was paired with a novel juvenile mouse during both Trials 1 and 2. Each trial consisted of a 4-min social interaction that was recorded using a video camera equipped with an infrared light source for low light conditions. Video recordings of social interaction, quantified as the total time that the adult experimental mouse engaged in olfactory investigation of the juvenile mouse, were scored by two treatment-blind observers to obtain a cumulative duration of investigation score. After testing, mice were returned to group housing with fresh bedding. The social recognition memory phenotyping protocol of Experiment 1 was replicated twice using separate groups of naïve mice and the results pooled.

Experiment 2—Social recognition performance of El mice exposed to a physiological stressor 

Salt loading procedure 

Mice were housed singly in standard polypropylene cages 1 week prior to the study in order to facilitate recording of body weight, food intake, and fluid intake. For control mice, water was available ad libitum, whereas experimental mice were supplied with a hyper-osmotic solution of 2% (w/v) sodium chloride solution 16 as their only source of fluid. Fluid intake was monitored 24 and 48h after salt solution exposure and food intake/body weight were recorded before and after exposure. After 48h, all mice were tested in the familiar juvenile condition of the social recognition memory task as described for Experiment 1 using a 120-min inter-exposure interval, a long-delay condition in which performance is altered by salt-loading in normal rodents.16 The stress-induced social recognition memory performance protocol of Experiment 2 was replicated twice using separate groups of naïve mice and the results pooled.

Experiment 3–Seizure susceptibility and brain stress neuropeptide activation following exposure to a physiological stressor 

Salt loading procedure 

Mice were maintained in their original housing conditions (singly or in groups of 2–4) in standard polypropylene cages in order to avoid triggering of handling-induced seizures in El mice. Both control and salt-loading groups were matched for gender and housing condition. For control mice, water was available ad libitum, whereas experimental mice were supplied with a hyper-osmotic solution of 2% (w/v) sodium chloride 16 as their only source of fluid. Fluid intake was monitored at 24 and 48h post-exposure. After 48h, all mice were tested for seizure susceptibility induced by handling.

Handling-induced seizure susceptibility (HISS) testing 

Seizure susceptibility was tested using the HISS protocol, which involves repetitive handling that reproduces the stress associated with routine cage changes. Each mouse was picked up by the tail and suspended 10–15cm above the home cage floor for 30s and then placed in a clean cage for 120s. The mouse was then suspended for an additional 15s before being returned to its home cage. Generalized seizures were identified by a loss of postural equilibrium, an erect forward-arching Straub tail, and head, limb, or chewing automatisms. Mice that displayed other signs of seizures such as vocalizations or twitching, but that did not progress to a generalized seizure were not considered seizure susceptible. Regular husbandry was discontinued for the week prior to HISS testing in order to insure a high frequency of seizures in El mice.

Perfusion and brain harvest 

Two hours after the HISS testing, a subset of control and salt-loaded mice (n=4/strain) were anesthetized and perfused intracardially with 0.1M phosphate-buffered saline, followed by 4% paraformaldehyde. All perfusions were performed during the dark-phase of the light-dark cycle between 1200 and 1800. Each brain was removed and then postfixed in 4% paraformaldehyde for 24–48h. The brains were then transferred into 30% sucrose solution until sinking (24–72h). Brains were sectioned (40μm) using a cryostat prior to histochemical analysis.

Nissl staining 

Tissue sections were mounted onto gelatinized slides, rinsed in ethanol and distilled water and then stained using 0.13% cresyl violet in an acetic acid buffer. Stained slides were rinsed in ethanol, cleared in xylene, and coverslipped.

CRF immunohistochemistry 

Immunocytochemical detection of r/h CRF (1–41) peptide was performed using a kit obtained from Peninsula Laboratories (San Carlos, CA). After rinsing in 50% ethanol and quenching of endogenous peroxidase activity with a 3% hydrogen peroxide solution, tissue sections were placed in a 15% normal goat serum blocking solution. Sections were incubated with the rabbit anti-CRF antibody (1:1000) overnight at 4°C. On the second day, sections were incubated with a goat, anti-rabbit secondary antibody (1:1000) for 1h and then a streptavidin-HRP conjugate for 30min. Sections were stained using a DAB chromagen and a hematoxylin counterstain.

Brain mapping microscopy 

Sections were photographed using an RT color Spot camera (Diagnostic Instruments Inc, Sterling Heights, MI) mounted on a Zeiss bright-field microscope. Nissl positive and CRF immunoreactive cells were counted using the IP Lab image analysis software (Scanalytics, Fairfax, VA). Analysis was completed without knowledge of strain or treatment condition. Nissl stained and CRF labeled cells were counted in the striatal fundus, paraventricular nucleus of the hypothalamus, basolateral nucleus of the amygdala, paraventricular nucleus of the thalamus and lateral septum. Sites were selected for analysis based upon a series of cFos mapping studies performed previously in El and ddY strain mice,20 by coarse examination of areas of intense staining by a treatment-blind observer, and by the known distribution of CRF-containing nuclei in rodent seizure models.21 Specific brain regions and nuclei of interest were identified with the help of a standard mouse brain atlas.22

Data analysis 

For Experiment 1, 2×2×2 mixed factor analyses of variance (ANOVAs) were performed for investigation times and difference scores with gender and strain as between subjects factors and time as a within subjects factor. For Experiments 2 and 3, 2×2×2×2 mixed factor ANOVAs were performed with gender, strain and treatment as between subjects factors and time as a within subjects factor. Simple main effect analyses were conducted when appropriate to determine individual group differences, and comparisons were considered significant when p<0.05.

Back to Article Outline

Results 

In Experiment 1, there was a main effect of strain on social investigation time on Trial 1 [F(1,80)=29.8, p<0.001] and on investigation time difference scores [F(1,80)=17.7, p<0.001] indicating that ddY mice had higher cumulative times on the first trial (Fig. 1) and higher difference scores (Fig. 2). There was a main effect of juvenile condition on investigation time difference scores [F(1,80)=4.8, p<0.05] indicating that ddY mice recognized familiar juveniles but not novel juveniles. There were no main effects of gender on investigation times or time difference scores although a strain by gender interaction effect was noted for Trial 1 [F(1,80)=25.8, p<0.001] as female mice of the ddY strain spent more time exploring juveniles than males (Fig. 3). There was also a strain by gender by condition interaction effect [F(1,80)=4.3, p<0.05]. This three-way interaction was due in part to the fact that ddY females demonstrated larger positive investigation time differences in the familiar juvenile condition as compared to El females, whose times were lower.

  • View full-size image.
  • Figure 1. 

    Social investigation times (mean±SEM) on the first and second trial of Experiment 1 in El and ddY mice tested in the social recognition memory task with familiar juveniles. Experimental mice were exposed to the same juvenile on each of the two trials so that the juvenile was familiar on Trial 2. The results reflect poor sociability and deficient short term working memory performance for El mice in contrast to relatively greater sociability and capable memory performance for ddY controls. *p<0.05 relative to the Trial 1.

  • View full-size image.
  • Figure 2. 

    Social investigation difference scores (mean±SEM) in El and ddY mice tested in the social recognition memory task of Experiment 1 with familiar and unfamiliar juveniles. In the familiar juvenile condition, experimental mice were exposed to the same juvenile on each of the two trials so that the juvenile was familiar on Trial 2. In the unfamiliar juvenile condition, experimental mice were exposed to novel juveniles on both trials. The results reflect capable short term working memory performance for ddY controls without non-specific performance or motivational deficiencies in contrast to the poor working memory performance for El mice. *p<0.05 relative to a mean of zero.

  • View full-size image.
  • Figure 3. 

    Social investigation times (mean±SEM) on the first trial in El and ddY mice tested in the social recognition memory task of Experiment 1. Experimental mice were exposed to an unfamiliar juvenile during a 4min test. The results reflect enhanced investigation of juveniles by ddY females relative to ddY males; no such difference emerged in the El strain. *p<0.05 relative to ddY males.

In Experiment 2, there was a strain by gender interaction effect on social investigation time [F(1,24)=5.8, p<0.05]. Simple main effect analyses revealed a significant difference in social investigation time from Trial 1 to Trial 2 in ddY subjects in the control group, but not in the ddY-salt loading or El treatment groups (Fig. 4). There was a main effect of gender [F(1,23)=6.0, p<0.05], but not strain or treatment on food intake with males eating more than females. Similarly, there was a main effect of gender [F(1,23)=9.7, p<0.05], but not strain or treatment, on fluid intake with males drinking more than females. There were main effects of strain [F(1,24)=19.6, p<0.05] and gender [F(1,24)=16.3, p<0.05], but not treatment, on body weight such that ddY and male mice were heavier than their El and female counterparts, respectively. Thus, salt loading did not impact significantly any appetitive or body weight measure (Table 1).

  • View full-size image.
  • Figure 4. 

    Social investigation times (mean±SEM) on the first and second trial of Experiment 2 in control or salt-loaded El and ddY mice tested in the social recognition memory task with familiar juveniles. Experimental mice were exposed to the same juvenile on each of the two trials so that the juvenile was familiar on Trial 2. The results reflect poor sociability and short term working memory performance for El mice in contrast to capable memory performance for ddY controls. Social recognition memory performance in ddY mice was abolished, however, by the salt-loading treatment. *p<0.05 relative to Trial 1.

Table 1. In Experiment 2, which assessed the impact of a physiological stressor on working memory performance, initial and final body weights, food intake, fluid intake for the first 24h, fluid intake for the second 24h, and total fluid consumption were measured in control and salt-loaded El (n=16, 8 females and 8 males) and ddY (n=16, 8 females and 8 males) strain mice (all mean±SEM)
StrainTreatment conditionInitial body weight (g)Final body weight (g)Food intake (g)First 24-h consumption (g)Final 24-h consumption (g)Total fluid intake (g)
ddYControl44.4±2.142.7±1.914.4±2.15.4±0.47.9±0.713.2±0.9
ddYSalt-loading45.6±1.442.6±1.311.0±1.45.8±2.28.0±0.813.7±1.2

ElControl35.7±3.832.9±3.112.6±1.17.2±1.89.5±1.216.7±1.9
ElSalt-loading37.0±2.434.0±2.510.8±1.18.1±1.011.7±1.319.8±2.1

The salt loading treatment did not impact significantly any of the appetitive or body weight measures.

In Experiment 3, there were no main effects of strain, condition or gender on total fluid intake level per cage or per individual mouse (data not shown). Consistent with Experiment 2 findings, these results suggest that a 48h period of exposure to hypertonic saline did not have a detectable impact on fluid intake. However, tail suspension handling resulted in a significantly lower percentage [χ2=3.8, p<0.05] of seizure occurrence in salt-loaded (22%, 2 of 9) than in control (71%, 5 of 7) El mice. These results suggest that a 48h period of exposure to hypertonic saline was sufficient to reduce seizure frequency. Analysis of Nissl stained brains revealed a significant [F(1,3)=19.3, p<0.05] main effect of treatment with exposure to hypertonic saline suppressing the cell count in paraventricular hypothalamic nucleus in both mouse strains relative to tap water controls (Table 2). Similarly, analysis of CRF content revealed significant main effects of treatment with exposure to hypertonic saline increasing CRF content in both the paraventicular hypothalamic nucleus [F(1,3)=9.7, p<0.05] and the lateral septum [F(1,3)=10.5, p<0.05] in both mouse strains relative to tap water controls (Table 2). These results suggest that a 48h period of exposure to hyptertonic saline was sufficient to induce compensatory changes in brain cell density and neurochemistry which were stressful in nature.

Table 2. For Experiment 3, which assessed the impact of a physiological stressor on seizure susceptibility and brain activation, a subset of control subjects (ddY mice: n=2, 1 female and 1 male, El mice: n=2, 1 female and 1 male) and a subset of subjects exposed to the salt-loading treatment (ddY mice: n=2, 1 female and 1 male, El mice: n=2, 1 female and 1 male) were sacrificed for brain mapping analysis
Nissl Staining
Region of Interest
StrainTreatmentNAcPVTPVNAmygSept
ddYControl1697±21829±1771388±1091865±3022007±562
ddYSalt-Loaded1517±2531106±601050±23*1320±4601709±323
ElControl1526±851792±1381647±881319±1691598±166
ElSalt-Loaded1291±791278±1121296±138*1154±1161470±69

CRF Staining
Region of Interest
StrainTreatmentNAcPVTPVNAmygSept
ddYControl2595±2732592±4982160±1662437±2172491±144
ddYSalt-Loaded2545±1462406±4482480±43*2418±2262831±151*
ElControl2307±772075±1651915±1011697±1512051±120
ElSalt-Loaded2551±22513±1562499±145*2730±1632531±227*

Exposure to hypertonic saline suppressed the cell count in paraventricular hypothalamic nucleus in both ddY and El strain mice relative to tap water controls and increased CRF content in the paraventricular hypothalamic nucleus and the lateral septum in both mouse strains relative to tap water controls. * p<0.05 relative to tap water controls.

Back to Article Outline

Discussion 

El mice exhibited in the present studies a short-term working memory deficit as well as altered reactivity to social, environmental, and physiological stressors. In Experiment 1, El mice demonstrated poor sociability and decreased olfactory investigation times, both anxiogenic-like traits, as compared to ddY controls. In Experiment 2, El mice revealed poor working memory performance in contrast to the capable working memory performance of ddY controls. Social recognition memory performance in ddY mice was abolished by salt-loading whereas El mice were unresponsive to stressor exposure. Finally, in Experiment 3 all mice exhibited enhanced brain stress neuropeptide (CRF) content following salt-loading and El mice exposed to hypertonic saline experienced a blunted reactivity to handling-induced seizures. Taken together, these findings suggest that El mice exhibit high emotionality as well as atypical reactions to stressor exposure, and that these characteristics impact social working memory performance and seizure-susceptibility.

The present studies examined a seizure-prone strain of mouse in a social context together with conspecific juveniles, thus providing a naturalistic measure of learning/memory in which spontaneous social behavior is monitored. In the ddY control strain, capable non-associative habituation learning was suggested by the marked decrease in social investigation time of familiar juveniles from the first to the second trial. However, in the El strain there was no such evidence of learning in spite of continued sensitivity to decreases in social investigation from the roughly 40s/4min trial juvenile exploration baseline of the El strain in the present studies (Fig. 1). The mechanism for lack of short term social working memory performance in El mice is unknown at the present time. One possibility is that sensory deficits in El mice led to poor social recognition memory performance. In particular, castration produces less persistence in social exploration, suggesting that social memory appears to be chemosensorily mediated by olfactory cues and gonadal hormones.19 However, El mice do not appear to have an impaired sense of smell in an olfactory discrimination task,23 and it can be suggested that some yet to be determined motivational or performance aspect of the social recognition memory task is instead deficient.

The result of Experiment 1 in which female ddY mice were more investigatory than males suggests that it may be adaptive for a female mouse to engage in social investigatory behavior for a longer duration or with more intensity in order to ascertain whether an unfamiliar intruder constitutes a threat. Accordingly, female rodents are reported to exhibit a 50% longer working memory span in the social recognition memory task compared to males.19 In comparison to male rats, females show less persistence in investigating juvenile conspecifics and yet retained social memories over longer delay intervals.24 Regarding mechanism, administration of oxytocin receptor agonists facilitated social recognition in male, but not female, rats and oxytocin receptor antagonists interfere with normal social recognition in female, but not male, rats 25 thus suggesting a sexually dimorphic role of oxytocin in rat social recognition.26, 27 Consistent with these findings, vasopressinergic neurons appear to mediate female versus male olfactory functions differently since the vasopressin receptor antagonist, dPTur(Me)VP, blocks social recognition in intact male, but not female, rodents.14 These studies provide a neurohypophyseal hormone mechanism for gender differences in normal rodent social recognition memory performance. Since the paraventricular hypothalamic nucleus is the source of neurohypophyseal hormones such as vasopressin, the present finding that two days of salt-loading altered cell density and CRF content in the paraventricular nucleus is consistent with the possibility that salt-loading altered social recognition performance in ddY mice via this mechanism. In addition, increased glial infiltration of hippocampal regions in the El brain, without obvious neuronal loss or synaptic rearrangements,28 provides evidence of neurodevelopmental perturbations in another prime substrate for mammalian learning and memory.29 Similarly, differential immediate-early gene activation arises in the El hippocampus over development with higher activation at an early age that normalizes as the El mouse enters adulthood.20 As social recognition performance was completely disrupted in both male and female El mice in the present studies, future studies can explore these neurohypophyseal and hippocampal memory modulatory circuits for clues to the primary neurological deficit in El mouse brain.

Seizure-prone El strain mice, on average, exhibit less social investigatory behavior than seizure-resistant ddY control strain mice. Video recordings of social investigation in El mice reveal a range of anti-social behaviors including lack of interest in an unfamiliar cagemate to persistent olfactory investigation of inanimate portions of the cage for most of a 4-min investigation period. Consistent with these findings, one study suggests that El mothers exhibit an overabundance of motoric activities that compete with crouching/nursing and pup retrieval behaviors required for viability of offspring.30 A general decrease in social investigation can be interpreted as an anxiogenic-like effect in rodents 31 and a specific decrease in social interaction among El mice has been documented previously.32 Consistent with evidence for high emotionality in El mice, behavioral, endocrine and neural indices of stressor exposure reveal heightened arousal in El mice relative to ddY controls and identify activation of the stress neuropeptide CRF as one component of the coping response to environmental challenge in El mice.33 Thus, the present decrease in Trial 1 social investigation times in El relative to ddY mice likely reflects a general anxiogenic-like phenotype in El mice expressed in an affiliative behavior context. Clinical results demonstrating co-morbidity of social cognition deficits and epilepsy are consistent with this result. For instance, clinical studies of pediatric epilepsy employing children with complex partial seizures and absence epilepsy demonstrate lower social competence scores compared to healthy control subjects.8

In an attempt to extend the finding of spontaneous emotionality exhibited by El mice, the present studies also investigated the effect of explicit stressor exposure on performance of ddY and El mice in the social recognition memory task. The present studies employed a physiological stressor, salt-loading, with the intention of exerting a gradual physiological departure from homeostasis over a period of 48h. Importantly, El mice could be exposed to the salt loading stressor passively and without any requirement for human handling or contact which would confound subsequent seizure susceptibility testing.34 No significant differences from the first to the second social investigation trial were found in either salt-loaded ddY or El mice, suggesting that the salt-loading treatment abolished capable social working memory performance in high functioning ddY control mice. In prior studies, injection of hypertonic saline to adult male rats following exposure to a juvenile decreased social exploration of the same juvenile presented two hours later.16 That is, exposure to the salt loading stressor during memory consolidation enhanced social recognition performance in a retention test completed under conditions of poor, long delay learning. The difference in the direction of salt-loading induced change in social recognition performance in these two studies may be due to procedural or empirical differences. For example, in the rat study the salt-loading treatment was administered after the first exposure to the juvenile,16 whereas in the present studies mice were exposed to the salt-loading treatment for 48h before the first exposure to the juvenile. Alternatively, baseline performance in the rat studies was poor whereas the performance of ddY mice over a 120min delay in the present studies was optimal suggesting that the salt-loading stressor produced a classical, biphasic memory modulatory effect.12 The essential point is that upon stressor exposure, the behavioral response of El mice was atypical.

Tail suspension handling, a noninvasive manipulation that is part of routine husbandry when lifting an animal by the tail for cage transfer, has been documented to exert multifaceted physiological and endocrine stress responses in normal rodents.35 A significant portion of a population of El mice beyond the age of post-natal day 70–80 will exhibit seizures in response to tail suspension handling. The surprising result of the present studies was that the normative 70% seizure frequency in El mice produced by tail suspension handling was attenuated to a frequency of 20% by exposure to hypertonic saline over a period of 48h. One possibility is that long-term neural adaptation to the salt-loading stressor fatigued stress reactive brain circuits by the time mice were exposed subsequently to the normally potent handling stressor. This in turn would suggest that salt loading altered brain circuits linked to seizure onset, termed “intrinsic fabricators.”36 Important clues for identifying the intrinsic fabricator mechanism(s) by which salt-loading in the present studies altered seizure susceptibility in El mice are provided by the quantification of brain cell density and CRF content. Hypertonic saline suppressed paraventricular hypothalamic nucleus (PVN) cell density in both ddY and El strains and increased CRF content in the PVN and the lateral septum in both mouse strains relative to tap water controls. These findings are consistent with the report that chronic salt loading in rats induced CRF expression in oxytocin (OT) neurons of the PVN and the supraoptic nucleus (SON) and upregulated CRF1 receptors in both vasopressin (VP) and OT-containing neurons.37 The increased CRF expression in hypothalamic OT and VP neurons in combination with the increase in CRF receptor levels suggest that CRF plays a role in fluid homeostasis in response to osmotic stimulation.37 Similarly, an increase in CRF content in vasopressinergic magnocellular neurons of the PVN and SON is reported in salt loaded rats.38 The regulatory role of VP, a hormone secreted by the hypothalamus and stored in the posterior pituitary, in fluid homeostasis has been well studied; in response to decreased levels of fluid intake, VP stimulates water conservation and reabsorption in the kidneys. Thus, the present increase in CRF content in PVN suggests that salt-loaded mice were chronically stressed, and that the neural response to stressor exposure was physiologically adaptive. The lateral septum is another brain area in which CRF levels are increased by administration of the seizure initiator, kainic acid,21 and this brain area is well known to mediate fear learning and other stress coping responses.39, 40 Taken together, these studies suggest that the role CRF plays in the regulation of physiological fluid homeostasis in rats under conditions of environmental flux is consistent with adaptive changes to salt-loading in ddY and El mice.

Further support for the hypothesis that an atypical response to stressor exposure can alter seizure susceptibility in El mice is provided by studies of the role of CRF in social withdrawal.32 Central injection of CRF-SAP, a conjugate of CRF linked with the neurotoxin saporin, resulted in an increase in social exploration upon exposure to an adult conspecific so that El mice performed a social interaction task in a manner comparable to normal ddY controls.32 These findings suggest that El mice express constitutively elevated sensitivity to CRF, resulting in oversensitivity to environmental challenge. Concurrent studies also employing CRF-SAP injections in El mice using additional behavioral endpoints, such as odor reactivity, also support these conclusions.23 Thus, the present studies provide evidence for stress dysregulation in El mice that contributes both to deficient performance of untrained, adaptive behaviors such as social recognition as well as neural and behavioral pathology associated with seizures.

Back to Article Outline

Acknowledgments 

This research was supported by a Research Incentive Grant from Boston College, and a grant from C.U.R.E. (SCH).

Back to Article Outline

References 

  1. Holmes GL, Khazipov R, Ben-Ari Y. Seizure-induced damage in the developing human: relevance of experimental models. Prog Brain Res. 2002;135:321–334
  2. Stafstrom CE. Assessing the behavioral and cognitive effects of seizures on the developing brain. Prog Brain Res. 2002;135:377–390
  3. Sogawa Y, Monokoshi M, Silveira DC, Cha BH, Cilio MR, McCabe BK, et al. Timing of cognitive deficits following neonatal seizures: relationship to histological changes in the hippocampus. Brain Res Dev Brain Res. 2001;131:73–83
  4. Akman CI, Hu Y, Fu DD, Holmes GL. The influence of cognitive reserve on seizure-induced injury. Epilepsy Behav. 2003;4:435–440
  5. Hannesson DK, Wallace AE, Pollock M, Corley S, Mohapel P, Corcoran ME. The relation between extent of dorsal hippocampal kindling and delayed-match-to-place performance in the Morris water maze. Epilepsy Res. 2004;58:145–154
  6. Cilio MR, Sogawa Y, Cha BH, Liu X, Huang LT, Holmes GL. Long-term effects of status epilepticus in the immature brain are specific for age and model. Epilepsia. 2003;44:518–528
  7. Ali A, Ahmad FJ, Pillai KK, Vohora D. Evidence of the antiepileptic potential of amiloride with neuropharmacological benefits in rodent models of epilepsy and behavior. Epilepsy Behav. 2004;5:322–328
  8. Caplan R, Sagun J, Siddarth P, Gurbani S, Koh S, Gowrinathan R, et al. Social competence in pediatric epilepsy: insights into underlying mechanisms. Epilepsy Behav. 2005;6:218–228
  9. Brace HM, Jefferys JG, Mellanby J. Long-term changes in hippocampal physiology and learning ability of rats after intrahippocampal tetanus toxin. J Physiol. 1985;368:343–357
  10. Krsek P, Mikulecka A, Druga R, Hlinak Z, Kubova H, Mares P. An animal model of nonconvulsive status epilepticus: a contribution to clinical controversies. Epilepsia. 2001;42:171–180
  11. McFadyen-Leussis MP, Heinrichs SC. Seizure-prone EL/Suz mice exhibit physical and motor delays and heightened locomotor activity in response to novelty during development. Epilepsy Behav. 2005;6:312–319
  12. Martinez JL, Thompson KA, McFadyen-Leussis M, Heinrichs SC. In:  Buccafusco J editors. Peptide and steroid hormone receptors as drug targets for enhancement of learning and memory performance. Basel: Birkhauser; 2003;p. 2004
  13. Heinrichs SC, Seyfried TN. Behavioral seizure correlates in animal models of epilepsy: a road map for assay selection, data interpretation, and the search for causal mechanisms. Epilepsy Behav. 2006;8:5–38
  14. Dantzer R. Vasopressin, gonadal steroids and social recognition. Prog Brain Res. 1998;119:409–414
  15. Heinrichs SC, Turek S, Orefice L, Forcelli P. A diathesis-stress model of seizure susceptibility in El/Suz mice: endocrine and electroencephalographic hyperreactivity to human contact. In: Society for Neuroscience Annual Meeting, Atlanta, Georgia, 2006.
  16. Bluthe RM, Koob GF, Dantzer R. Hypertonic saline mimics the effects of vasopressin and social recognition in rats. Behav Pharmacol. 1991;2:513–516
  17. Heinrichs SC, Koob GF. Corticotropin-releasing factor in brain: a role in activation, arousal, and affect regulation. J Pharmacol Exp Ther. 2004;311:427–440
  18. Winslow JT. Mouse social recognition and preference. In:  Crawley J,  Gerfen CR,  Rogawski MA,  Sibley DR,  Skolnick P,  Wray S editor. Current protocols in neuroscience. New York: Wiley; 2003;p. 1–16Ch. 8.16
  19. Sawyer TF, Hengehold AK, Perez WA. Chemosensory and hormonal mediation of social memory in male rats. Behav Neurosci. 1984;98:908–913
  20. McFadyen-Leussis M, Heinrichs SC. Handling of EL/Suz mice elicits differential fos activation in the locus ceruleus during development. In: Society for Neuroscience Annual Meeting, San Diego, California, 2004.
  21. Piekut DT, Phipps B. Increased corticotropin-releasing factor immunoreactivity in select brain sites following kainate elicited seizures. Brain Res. 1998;781:100–113
  22. Paxinos G, Franklin KBJ. The mouse brain in stereotaxic coordinates. 2nd ed.. San Diego: Academic Press; 2001;
  23. Pascual, J. and Heinrichs, S. C. Olfactory neophobia and seizure susceptibility phenotypes in an animal model of epilepsy are normalized by impairment of brain corticotropin releasing factor, submitted for publication.
  24. Bluthe RM, Dantzer R. Social recognition does not involve vasopressinergic neurotransmission in female rats. Brain Res. 1990;535:301–304
  25. Engelmann M, Ebner K, Wotjak CT, Landgraf R. Endogenous oxytocin is involved in short-term olfactory memory in female rats. Behav Brain Res. 1998;90:89–94
  26. Bielsky IF, Hu SB, Szegda KL, Westphal H, Young LJ. Profound impairment in social recognition and reduction in anxiety-like behavior in vasopressin V1a receptor knockout mice. Neuropsychopharmacology. 2004;29:483–493
  27. Bielsky IF, Young LJ. Oxytocin, vasopressin, and social recognition in mammals. Peptides. 2004;25:1565–1574
  28. Drage MG, Holmes GL, Seyfried TN. Hippocampal neurons and glia in epileptic EL mice. J Neurocytol. 2002;31:681–692
  29. Fenoglio KA, Brunson KL, Baram TZ. Hippocampal neuroplasticity induced by early-life stress: Functional and molecular aspects. Front Neuroendocrinol. 2006;27:180–192
  30. Bond TL, Drage M, Heinrichs SC. Seizure-prone EL mice exhibit deficits in pup nursing and retrieval assessed using a novel method of maternal behavior phenotyping. Epilepsy Behav. 2003;4:57–64
  31. File SE, Seth P. A review of 25 years of the social interaction test. Eur J Pharmacol. 2003;463:35–53
  32. Turner LH, Lim CE, Heinrichs SC. Anti-social and seizure susceptibility phenotypes in an animal model of epilepsy are normalized by impairment of brain corticotropin releasing factor. Epilepsy Behav, in press.
  33. Forcelli P, Orefice LL, Heinrichs SC. Neural, endocrine and electroencephalographic hyperreactivity to human contact: a diathesis-stress model of seizure susceptibility in El mice, submitted for publication.
  34. Todorova MT, Burwell TJ, Seyfried TN. Environmental risk factors for multifactorial epilepsy in EL mice. Epilepsia. 1999;40:1697–1707
  35. Balcombe JP, Barnard ND, Sandusky C. Laboratory routines cause animal stress. Contemp Top Lab Anim Sci. 2004;43:42–51
  36. Jobe PC, Browning RA. The serotonergic and noradrenergic effects of antidepressant drugs are anticonvulsant, not proconvulsant. Epilepsy Behav. 2005;7:602–619
  37. Imaki T, Katsumata H, Miyata M, Naruse M, Imaki J, Minami S. Expression of corticotropin releasing factor (CRF), urocortin and CRF type 1 receptors in hypothalamic-hypophyseal systems under osmotic stimulation. J Neuroendocrinol. 2001;13:328–338
  38. Young WS. Corticotropin-releasing factor mRNA in the hypothalamus is affected differently by drinking saline and by dehydration. FEBS Lett. 1986;208:158–162
  39. Radulovic J, Fischer A, Katerkamp U, Spiess J. Role of regional neurotransmitter receptors in corticotropin-releasing factor (CRF)-mediated modulation of fear conditioning. Neuropharmacology. 2000;39:707–710
  40. Price ML, Kirby LG, Valentino RJ, Lucki I. Evidence for corticotropin-releasing factor regulation of serotonin in the lateral septum during acute swim stress: adaptation produced by repeated swimming. Psychopharmacology (Berl). 2002;162:406–414

PII: S1059-1311(06)00199-3

doi:10.1016/j.seizure.2006.10.006

Seizure: European Journal of Epilepsy
Volume 16, Issue 1 , Pages 59-68, January 2007

Article Tools

* Image Usage & Resolution