FULL-LENGTH ORIGINAL RESEARCH
Anxiety and locomotion in Genetic Absence Epilepsy Rats from Strasbourg (GAERS): Inclusion of Wistar rats as a second control *†Jos e Eduardo Marques-Carneiro, *†Jean-Baptiste Faure, *Brigitte Cosquer, †Estelle Koning, †Arielle Ferrandon, *Anne Pereira de Vasconcelos, *Jean-Christophe Cassel, and †‡Astrid Nehlig Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
SUMMARY
Eduardo Jose Marques-Carneiro, PhD student at the University of Strasbourg and Federal University of Sao Paulo.
Objective: The Genetic Absence Epilepsy Rats from Strasbourg (GAERS) is a genetic model, derived from Wistar rats by selective breeding. In all previous studies, GAERS were compared to their paired selected strain not expressing spike-and-wave discharges (SWDs), namely nonepileptic controls (NECs). Because the occurrence/ absence of SWDs is of polygenic origin, some other traits could have been selected along with occurrence/absence of SWDs. Therefore, we explored the importance of using a second control group consisting in Wistar rats, the strain of origin of GAERS, in addition to NECs, on locomotion and anxiety in GAERS. Methods: A test battery encompassing home-cage, open-field, beam-walking and elevated plus-maze evaluations was used. In addition, stereologic analyses were performed to assess the volume of thalamus, amygdala, and hippocampus. The occurrence/absence of SWDs was determined in all three strains by electroencephalography (EEG) recording. Results: When compared to NECs and Wistars, GAERS displayed lower exploratory activity and fastened habituation to novelty. In the plus-maze, scores of GAERS and Wistars were similar, but NECs appeared significantly less anxious (possibly in association with increased amygdala volume); evidence for weaker anxiety in NECs was also found in the open-field evaluation. The volumetric study revealed increased thalamic volume in GAERS compared to both control groups. SWDs were present in all GAERS and in 80% of Wistars. Significance: Compared to the original Wistar strain as an additional control group, the selective breeding that generated the GAERS has no incidence on anxiety-related behavior, conversely to the selection of SWD suppression in NECs, in which anxiety is attenuated. These findings point to the importance of using a second control group composed of Wistar rats in studies characterizing the behavioral profile of GAERS. Thereby, possible confusions between occurrence/absence of SWDs and other features that come along with selection and/or differential brain development induced by the genetic mutations are reduced. KEY WORDS: Absence epilepsy, Activity, Anxiety, Second control group, Stereologic based volumetry.
Accepted June 23, 2014; Early View publication July 24, 2014. *Laboratory of Cognitive and Adaptive Neuroscience (LNCA), Faculty of Psychology, UMR 7364, University of Strasbourg – CNRS, Strasbourg, France; †Faculty of Medicine, INSERM U 666, University of Strasbourg, Strasbourg, France; and ‡Hospital Necker, INSERM U 1129, University Paris Descartes, Paris, France Address correspondence to Jean-Christophe Cassel, Laboratoire de Neurosciences Cognitives et Adaptatives, UMR 7364 University of Strasbourg – CNRS, Faculty of Psychology, 12 rue Goethe, F-67000 Strasbourg, France. E-mail: [email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
Spike-and-wave discharges (SWDs) characterize the electrographic pattern typically observed during absence seizures. SWDs occur spontaneously in many rat strains including inbred strains such as Fischer 344, Brown Norway, and dark agouti,1–3 and outbred strains such as Wistar, Sprague-Dawley, and Long-Evans.3–7 Almost 30% of Wistars present SWDs,8and this percentage increases with aging, reaching 100% by 84–94 weeks.9
1460
1461 Anxiety and Locomotion in GAERS In the 1980s, two independent groups reported the presence of spontaneous SWDs in two different strains derived from selective breeding of Wistar rats. First, in Strasbourg, a selection among Wistar rats that spontaneously expressed SWDs led, after four generations, to a strain named Genetic Absence Epilepsy Rats from Strasbourg (GAERS), in which 100% of the subjects display SWDs. Simultaneously, from the same Wistar strain, rats that never expressed SWDs allowed the development of a control strain, nonepileptic controls (NECs).7 The second strain with spontaneous SWDs, the Wistar Albino Glaxo from Rijswijk (WAG/Rij) was inbred from Wistar rats, but the occurrence of SWDs was observed only once the rats were fully inbred.2 For these reasons, WAG/Rij do have their own control group devoid of SWDs. Therefore, studies on WAG/Rij use either the August Copenhagen Irish rats (ACIs), which do not express any SWD,10 or more often Wistars as control groups. GAERS, WAG/Rij, and NECs are animal models resulting from multigenic selection.11,12 This multigenic selection might have more features than simply the presence or absence of SWDs. This means that the choice of the control group can have a big impact on the interpretation of the results. For example, some studies on WAG/Rij used ACI rats, whereas others used Wistars as controls. Using a strain without any SWD or a strain in which some rats present SWDs might influence the data and their interpretation. In studies regarding Wag/Rij, the use of two control strains—one inbred nonepileptic (ACI) and one outbred (Wistar)—allows the comparison of the effects of genotype (epileptic vs. nonepileptic and inbred vs. outbred) and the possible consequences on multigenic selection and inbreeding on features other than only genetic absence seizures.13–15 The same paradigm might be used with GAERS. However, until now no study on GAERS used the original Wistar strain as a second control group; only NECs were used. An example of a possible interaction is the significantly higher level of anxiety observed in GAERS as compared to NECs,16,17 whereas no difference in anxiety levels were found in WAG/Rij compared to outbred Wistars.18,19 This shows that the multigenic selection at the origin of those strains might have more features than only the presence or absence of SWDs. Thus, an additional control group comprising Wistars would be critical to ensure that strain differences in anxiety levels reflect genuine behavioral or biologic differences and are not biased by consequences of the genetic selection. The present study aimed to evaluate anxiety and locomotion in GAERS compared to either NECs or Wistars from a standard colony. This study was paired with stereologybased measurements of the volume of amygdala, thalamus, and hippocampus in all rats. The objective was to clarify whether NECs can be considered an appropriate and sufficient control group in studies on GAERS.
Materials and Methods Animals Male GAERS (n = 12), NECs (n = 11) from the Institut Jean Roget (Grenoble, France), and Wistar rats (n = 12) from CERJ (L’Arbresle, France) were used for this experiment. Throughout its duration, temperature (21° 23°C) and lighting (12 h/12 h cycle; light period from 7 a.m. to 7 p.m.) were controlled. The rats were placed in individual cages (42 9 26 9 15 cm) with food and water ad libitum. Experiments were performed in accordance with the rules of the European Community Council directive of September 22, 2010 (2010–63) and the French Department of Agriculture (2013–118; License #67–97 for AN, #67–7 for APV, and #67–215 for JCC).The protocol was approved by the ethical animal research committee board of the University of Strasbourg (CREMEAS - Comite Regional d’Ethique en Matiere d’Experimentation Animale de Strasbourg # AL/97/104/ 02/13). Constant efforts were made to minimize discomfort or pain in the animals. The experiments were performed following the principles outlined in the ARRIVE (Animal Research: Reporting on In Vivo Experiments) guidelines and the Basel declaration (http://www. basel-declaration.org). The 3R (Reduce, Refine, Replace) concept has been considered when planning the experiments. Behavioral testing began 2 months after the arrival of rats in the laboratory and was performed between the age of 3 and 6 months according to a timeline that follows the order of presentation below. Behavioral tests Behavioral testing was started 2 months after arrival of the animals. Locomotor activity in the home cage Locomotor activity was measured in the home cages of all rats to avoid any interference of stress. The cages were placed on shelves in a dedicated room. Each cage was traversed by two infrared light beams targeted on two photocells, 4.5 cm above floor level and separated by 28 cm. A 3 h period allowed habituation to the novelty of the test situation.20 The number of longitudinal cage crossings was recorded for each rat by a computer over 48 h. Open field This test taxes an animal’s activity in a novel environment, as well as habituation, anxiety, and exploration. A square open field surrounded by walls that were high enough to prevent rats from jumping out (65 9 65 9 43 cm) was used. The acrylic glass floor was divided into 25 squares of equal size. The rat was placed in the center of the device. The number of squares crossed (total, Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1462 J. E. Marques-Carneiro et al. central, and peripheral) and the total time spent in the center of the device were quantified in 2-min bins over 10 consecutive min. Before each trial, the device was cleaned with 100% alcohol. Sensorimotor coordination in the beam-walking test Assessment of sensorimotor coordination was performed by placing each rat on a 3 cm wide and 200 cm long wooden beam, divided into four segments of 50 cm, and elevated 80 cm above floor level. The beam communicated with the home cage at one end. All rats were trained over 4 days on progressively longer portions of the beam. During the first session, rats were placed on the beam at 50 cm from the goal box (their home cage) for five consecutive trials. In the following sessions, rats had to run on progressively longer portions of the beam (increased by 50 cm). On the fifth day, all rats were tested for three consecutive 200 cm trials, and their performance was scored by an experimenter. For each virtual 50 cm segment traveled on the beam, the score was 1 when the rat traversed the segment with all paws on the upper surface of the beam. A score of 0 was attributed for each segment on which the rat slipped or placed its toes on the side of the beam. The overall score was calculated by adding the scores of the three assays (maximal score = 12). Elevated plus-maze This test uses a device made of two open and two enclosed arms crossing in their middle. It measures the anxiety of the animals based on the contrast between exploratory behavior and tendency to avoid open, enlightened and elevated spaces. A maze made of two black open arms without side walls (50 9 10 9 1.5 cm), two arms closed by side walls (50 9 10 9 40 cm), and a central platform (10 9 10 cm) was placed at 73 cm above the ground. Four halogen lights illuminated the maze (30 lux over open arms and 5 lux over closed arms). The behavior of all rats was recorded using a video camera placed above the maze and monitored by a computer, allowing the experimenter to see the animal’s behavior on a screen. Staying in its home cage, the rat was brought to the test room 5 min before the start of the test (habituation period). To start testing, the rat was placed in the center of the maze, with head toward a closed arm. The rat was free to explore the device during 5 min. The experimenter recorded the displacements using the keyboard (an entry in every part of the maze being assigned to a key on the keyboard). After each trial, the device was cleaned with 100% alcohol. The variables measured were the number of entries in open and closed arms and the time spent in these arms. The percentage of entries and time spent in open arms (open arms/[open arms + closed arms] 9 100) were analyzed to minimize a bias due to possible differences in locomotor activity. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
Electroencephalographic recording GAERS, NECs, and Wistars were anesthetized by an injection of a mixture of ketamine (80 mg/kg, i.p.) and xylazine (30 mg/kg, i.p). The rats were equipped with four single-contact electrodes over the frontoparietal cortex (anteroposterior [AP] 2, mediolateral [ML] 2.5 mm from bregma), as described previously.21 All animals were given a 1-week postsurgical recovery time. Then, they were habituated to the experimental setting over 2 days (placed in recording boxes, the first day without connection to the recording system and the second day with connection). A digital acquisition computer-based system (Coherence, Deltamed, France) was used for the continuous recording of electroencephalography (EEG) activity in freely moving animals. For each recording, the animals were habituated for 15 min to the recording cage. Recording lasted for 2 h (the 20 first and the last 40 min were excluded from the analysis). This procedure was repeated for 5 days, and the first and last days were excluded from the final analysis. To avoid any interference with the sleep-wake cycle, all recordings were carried out in the morning, starting at 10:00 a.m. The number and cumulative duration of SWDs were calculated and represent the mean of three 1-h recording periods. In Wistars, values were calculated only from rats that expressed SWDs. Histologic and stereologic studies (estimation of volume) At completion of behavioral testing and EEG recordings, all rats were killed with an overdose of pentobarbital (200 mg/kg), transcardially perfused with 50 ml saline followed by 50 ml of 0.1 M phosphate-buffered 4% paraformaldehyde (4°C). After extraction, the brains were postfixed for about 4 h and transferred into a 0.1 M phosphate buffered 20% sucrose solution, in which they were kept for 36–40 h at 4°C. The brains were then quickly frozen and cut into 50-lm–thick coronal sections using a cryostat. Every sixth section was sampled using systematic random sampling according to stereologic principles,22 mounted on slides, and stained with cresyl violet. Regions of interest (ROI), that is, the amygdala, hippocampus, thalamus, as well as the whole brain were delineated using a microscope (Leica DM5500B; Leica Microsystems SAS, Nanterre, France), a 2.59 objective, and the Mercator Software (Explora Nova, La Rochelle, France), based on the Paxinos and Watson rat brain atlas.23 Digital live microscope images were visualized using a high-resolution Optronic Microfire camera. The boundaries of amygdala considered all nuclei (bregma 1.80 mm to bregma 3.36 mm). The boundaries of hippocampus were determined by the appearance of the anterior CA3 and dentate gyrus (DG) (bregma 1.80 mm) and the disappearance of the DG at the posterior level (bregma 6.84 mm). The thalamus was considered as a whole (bregma 1.80 mm to bregma 4.68 mm). The whole brain was taken from bregma 1.80 mm to bregma 6.84 mm.23
1463 Anxiety and Locomotion in GAERS Volume estimates were obtained using Cavalieri’s method,24 that is, by summing the area of each region in each image and multiplying it by the section + interval thickness (300 lm). The precision of the estimates is described by the coefficient of error (CE) that in all regions of interest ranged between 0.02 and 0.06. Stereologic data analysis: because a significant difference in brain volume (Table S1) was evidenced between strains, we used data normalized for brain size for each individual region. Statistical analysis Analyses of variance (ANOVAs) were performed. A single “Group” factor design was used for beam-walking and plus-maze scores, a “Group X Interval” design for locomotor activity in the open field, and a “Group X Time” or “Group X Period” design for home cage activity. A post hoc Newman-Keuls test was used to compare different groups after ANOVA and when required. The data concerning the characteristic features of SWDs in GAERS and Wistars were compared with a Student’s t-test for independent samples. All statistical analyses were performed using Statistica software (Statsoft Inc, Maisons-Alfort, France).
Results Behavioral tests Locomotor activity in the home cage One rat from each group was excluded from data analysis because of technical problems in the reliability of infrared sensors. Data from 11 GAERS, 10 NEC, and 11 Wistars were analyzed. Habituation (first 3 hours): For the number of cage crossings (Fig. 1A), ANOVA showed a significant overall Group effect (F2,29 = 3.797, p < 0.05) due to lower activity in NECs compared to Wistars (p < 0.05). ANOVA also showed a significant effect of Time (F5,145 = 102.95, p < 0.001), reflecting a reduction in the overall activity over time. Finally, ANOVA showed a Group X Time interaction (F10,145 = 2.80, p < 0.01) due to activity levels that were significantly lower in GAERS than in Wistars (p < 0.05), but higher than in NECs (p < 0.01) during the first hour. Basal activity: ANOVA did not reveal a significant Group effect during the light period (F2,29 = 0.2921, p = 0.75), compared to the dark period (F2,29 = 4.458, p < 0.05) (Fig. 1B). During the dark period, activity was significantly reduced in GAERS and NECs compared to Wistars (p < 0.05). There was no significant effect of Period or Group X Period interaction. Open field Habituation: ANOVA showed a Group effect associated with lower overall activity in GAERS compared to both control strains (p < 0.001). However, GAERS showed a higher
A
B
Figure 1. Home cage activity. (A) Habituation to a novel environment reflected by activity levels over the first 3 h and shown in six 30min bins. NECs showed lower activity levels than Wistars during the first 30 min. Habituation occurred in all groups, as shown by comparable reduction of activity over time. (B) Basal activity during day and night periods. Day activity was similar in all groups. During the night, NECs and GAERS were significantly less active than Wistars. *p < 0.05, statistically significant difference from Wistars, +p < 0.05, statistically significant intrastrain difference. Epilepsia ILAE
activity than the two control groups during the first two min, but subsequently their activity decreased markedly (Fig. 2A). During the fourth interval (6–8 min), NECs were more active than Wistars. These differences explain the significant Group X Interval interaction (F8,128 = 14.55, p < 0.001). In addition, the activity of Wistars and GAERS gradually decreased over time, which was not the case in NECs. Rearings: ANOVA showed a significant Group effect (F2,32 = 47.12, p < 0.001) due to a lower number of rearings in GAERS compared to Wistars (p < 0.001) and NECs (p < 0.001) (Fig. 2B). Activity in the central area: ANOVA showed a significant Group effect (F2,32 = 9.95, p < 0.001) due to higher activity scores in NECs compared to Wistars (p < 0.01) and GAERS (p < 0.001) (Fig. 2C). Activity in the periphery: ANOVA showed a significant Group effect (F2,32 = 12.65, p < 0.001) due to lower activity scores in GAERS compared to Wistars (p < 0.001) and NECs (p < 0.001) (Fig. 2D). Beam-walking In this test (Fig. 3A), ANOVA showed a significant Group effect (F2,32 = 60.50, p < 0.001) due to impaired performance in Wistars compared to GAERS (p < 0.001) and NECs (p < 0.001). Performance was related inversely to the rats’ body weight (Fig. 3B) and most probably reflected a body weight–induced bias, with Wistars being much heavier than GAERS and NECs (p < 0.001). Elevated plus-maze Concerning the time spent in open arms (Fig. 4A), ANOVA showed a significant Group effect (F2,32 = 14.94, p < 0.001) that was due to a significantly longer time in Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1464 J. E. Marques-Carneiro et al.
A
B
C
A
B
D Figure 3. Sensorimotor coordination assessed in the beam-walking test. (A) Mean score of each group in the test. Wistars exhibited a lower performance compared to NECs and GAERS. (B) Average body weight in three strains measured at the end of the beamwalking test. Wistars were significantly heavier than NECs and GAERS. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
Figure 2. Behavior in the open field. (A) GAERS were more active during the first 2 min and less active during the last 8 min in comparison with the two control groups. NECs showed a relatively constant activity, and the activity of Wistars decreased over time. (B) Number of rearings accounting for exploratory activity. GAERS showed significantly fewer rearings than both control groups. (C) Anxietylike behavior is reflected by avoidance of the central area of the open field. NECs spent significantly more time in the center than did GAERS and Wistars. (D) Locomotor activity is reflected by peripheral activity. GAERS showed lower activity in the periphery than the two other strains. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
NECs compared to GAERS (p < 0.001) and Wistars (p < 0.001). ANOVA also revealed a significant Group effect (F2,32 = 13.41, p < 0.001) on the number of entries in open arms (Fig. 4B); this number was larger in NECs than in GAERS (p < 0.05) or Wistars (p < 0.001). The number of entries in open arms was also larger in GAERS than in Wistars (p < 0.05). EEG recordings Because NEC rats showed no EEG evidence of SWDs, only GAERS and Wistar rats were considered for EEG Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
analyses. As expected, 100% of GAERS presented SWDs, whereas 9 of 11 Wistars exhibited SWDs. The day-to-day variability of SWD features is shown in Figure S1 (Supplemental data). The mean number, total duration, and mean duration of seizures of the 11 GAERS and 9 Wistars over three recording sessions are presented in Figure 5. The t-test showed a significant difference in the total number of seizures (t20 = 4.35, p < 0.001; Fig. 5A), total seizure duration (t20 = 10.96, p < 0.001; Fig. 5B), and mean duration of a seizure (t20 = 9.62, p < 0.001; Fig. 5C). Volumetric analysis In line with the significant difference in body weight between the three groups, the whole brain volume was smaller in NECs and GAERS than in Wistars (Table S1). To standardize measurements, the statistical analysis of volumetric measurements was performed on the ratio between the volume of each structure and the whole brain volume. Amygdala: ANOVA showed a significant Group effect on amygdala volume (F2,29 = 7.35, p < 0.01), which was larger in NECs compared to Wistars (p < 0.01) and GAERS (p < 0.05) (Fig. 6A). Thalamus: ANOVA showed a significant Group effect (F2,29 = 5.09, p < 0.05) on thalamic volume, which was significantly larger in GAERS compared to Wistars (p < 0.05) and NECs (p < 0.05) (Fig. 6B).
1465 Anxiety and Locomotion in GAERS group made of rats from an outbred strain, that is, their common strain of origin, Wistar rats. A
B
Figure 4. Anxiety-like behavior assessed in the elevated plus-maze test. (A) Percentage of time spent in the open arms. NEC rats spent significantly more time in open arms than GAERS and Wistars, indicating reduced anxiety-like behavior. (B) Percentage of open arm entries. This percentage was significantly larger in NECs than in Wistars and GAERS, reflecting reduced anxiety in NECs. The number of open arm entries was higher in GAERS than in Wistars. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
Hippocampus: ANOVA showed no significant Group effect on hippocampal volume (F2,29 = 0.61, p = 0.55) (Fig. 6C).
Discussion Both GAERS and NECs were derived from the Wistar strain by progressive genetic selection. Selective breeding is used to enhance specific phenotypes or traits.25 However, by using this technique, either genetic variants that contribute to the selected trait (e.g. SWD) rise quickly to fixation (homozygosity), or many other variants may become fixed during inbreeding.25 The SWD-related phenotype in GAERS has a polygenic inheritance.11 Nothing is known, however, about the possible consequences of polygenic inheritance changes unrelated to the SWD phenotype or to its absence in NECs. Thus, in studies comparing only GAERS to NECs, it is impossible to distinguish whether the differences observed are due to genetic changes underlying SWDs or to other genetic changes that accompany the selection of both strains. Therefore, we compared GAERS to their usual control strain, NECs, but also, for the first time, to a second control
Locomotion and sensorimotor coordination Locomotor abilities such as basal and exploratory activity, sensorimotor coordination, and novel environment habituation were assessed because a difference at this level could have biased other measurements. All three strains exhibited similar levels of basal activity in the home cage in the lights-on period, during which all behavioral testing was performed. This means that for all other tests, the performance in the different groups was most probably not affected by basal activity differences. However, during the night and hence the activity period, NECs and GAERS were slightly but significantly less active than Wistars. When exposed to higher novelty in the open field, GAERS showed reduced exploration scores compared to either Wistars or NECs, a difference particularly marked for rearing and peripheral activity. Of interest, when compared to Wistars, WAG/Rij also showed decreased exploratory activity in the open field.18 In addition, in the open field, GAERS showed faster habituation than the two other strains, whereas NECs hardly showed evidence of habituation. However, over a period of 90 min in the home cage activity test, all three strains showed habituation. In the beam-walking test, GAERS and NECs showed good sensorimotor abilities. The narrow width of the bar and the overweight condition of Wistars did not allow a reliable assessment of their coordination. However, evidence for sensorimotor ability in Wistars was reflected independently by their good performance in the Morris water maze (data not shown). The rats were tested on the bar at 22 weeks of age. In accordance with the literature and our own observations, at that age Wistars have reached a mean weight of 420 g, whereas GAERS and NECs are 25–40% lighter.26,27 This is most likely a trait selected along with the absence/presence of SWDs. Anxiety Characterized by somatic, emotional, cognitive, and behavioral components, anxiety can be considered an adaptive response to environmental pressure. However, when anxiety becomes excessive, overwhelming, or uncontrollable, it is considered a disorder. The amygdala and ventral hippocampus are the most important brain regions regulating anxiety-related responses.28 In addition, emotional disorders reflected by pathologic anxiety are often comorbid to epilepsy.29 Recently, a significantly higher anxiety level in GAERS compared to NECs was reported.16,17 Our study showed also increased anxiety in GAERS compared to NECs in the open-field and plus-maze evaluations. However, in these tests, GAERS did not differ from Wistars. These observations reflect that the genetic selection of NECs affected anxiety level, which does not seem to be the case in Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1466 J. E. Marques-Carneiro et al.
A
B
C
D
Figure 5. Electroencephalographic recordings in GAERS and Wistars. Significant differences were recorded in (A) number, (B) total duration, and (C) mean duration of seizures between GAERS (n = 11) and Wistars (n = 9). (E) EEG traces showing an example of SWD in GAERS and Wistars. As expected, NECs showed no SWD. *p < 0.05, statistically significant difference between GAERS and Wistars. Epilepsia ILAE
A
B
C
Figure 6. Stereology-based volumetric analysis. (A) Amygdala volume relative to the whole brain was larger in NECs than in Wistars and GAERS. (B) Thalamic volume relative to the whole brain was larger in GAERS than in Wistars and NECs. (C) Hippocampal volume was similar in all strains. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1467 Anxiety and Locomotion in GAERS GAERS. Thus, the use of GAERS to study anxiety as a psychiatric comorbidity in epilepsy, as proposed by Bouilleret et al.,17 might not be indicated. In agreement with our data, Sarkisova and Luijtelaar19 showed no difference in the anxiety level between WAG/ Rij and Wistars. Altogether, our data show that because of the genetic selection for the absence of SWDs, NECs should not be used as sole controls when characterizing GAERS. Volumetric measurements Bouilleret et al.17 suggested that the difference of anxiety-like behavior between GAERS and NECs could be related to a difference in the volume of the amygdala. We found an increased volume of amygdala in NECs compared to Wistars or GAERS. This observation is at variance with the data reported by Bouilleret et al.17 In the latter study,17 volumes were assessed using magnetic resonance imaging (MRI), which does not allow delineation of boundaries of the structures as precisely as histology does. In addition, the MRI measures of the previous study encroached onto entorhinal and piriform cortices,17 which might explain the discordant data between the two studies. Moreover, the difference in the amygdala size between both studies could also be explained by the fact that the study by Bouilleret et al.17 used female while we used male subjects, and the level of anxiety is sex dependent.30 In better agreement with our results, no significant difference in the amygdala volume was reported in patients with absence epilepsy compared to normal control subjects.31 Of interest, a smaller volume of the amygdala was found in individuals with a high anxiety level.32,33 Hence, the larger volume of the amygdala measured in NECs might account for their decreased anxiety-like behavior in the open-field and elevated plus-maze tests. It has been suggested that the presence of abnormalities in thalamocortical networks would underlie idiopathic generalized epilepsy, including absence seizures.34 Some studies found increased cortical volume in patients with absence epilepsy.35 Other studies reported neuroanatomic abnormalities in the thalamus, including thalamic atrophy in adolescent and young adult patients34 and children,36 but thalamic hypertrophy was reported in adult patients.37 Our results are consistent with the latter study, as we found a significantly larger thalamic volume in GAERS than in both control groups. Finally, we did not find any significant volumetric difference in the hippocampus, whatever the group. Spike-and-wave discharges To determine the presence/absence, the frequency, and the duration of SWDs, we recorded EEG in the three groups at the end of the behavioral assessments. As could be expected, all GAERS and no NEC presented SWDs. However, about 80% of Wistars displayed SWDs. The
latter observation is in line with literature data reporting a percentage of affected individuals ranging from around 30% in young adult Wistars8 to up to 100% at 84–94 weeks.9 However, in Wistars, the number of SWDs was two times lower, the total seizure duration six times, and the mean seizure duration four times shorter than in GAERS. In conclusion, it appears that GAERS show a faster habituation to a novel environment but explore less than NECs and Wistars. GAERS appear as anxious as Wistars. On the contrary, NECs appear less anxious than GAERS/ Wistars, an observation that might fit with their significantly larger amygdala volume. These data provide a unique opportunity for studying the genetic features underlying differences in anxiety behavior. In addition, the main conclusion of this work is the importance and benefit of a second control group composed of Wistar rats in studies on GAERS. Limitations of the study The data obtained here should be considered with some caution, because not all three strains originate from the same breeder. Indeed, although GAERS and NEC rats were born and raised in the same environmental conditions, hence having presumably received a similar quality of maternal care and attention from animal keepers, Wistars were provided by a commercial breeder. Therefore, and although laboratory animal breeding and care have both to scrupulously respect national and international statutory constraints, it cannot be excluded that prenatal gestation and early postnatal rearing conditions may have differed between Wistars and the two other groups of rats in a way that could also partly affect the herein reported difference in anxiety-like behavior. Indeed, several studies have shown that differences in, for example, maternal care, or simply in the type of bedding material used in cages could affect the level of anxietylike behavior in the offspring as adults.38–40 Moreover, early life stress, but a priori stronger than differences in breeding, could affect epileptogenesis.41,42 It is, however, of interest to note that in any type of measurement performed here, Wistars never differed from the two genetically modified strains. Indeed, sometimes they differed from NECs and sometimes from GAERS, while the consequences of early life environment reported in the literature are usually concordant.38–40 This indicates that early life environment cannot be the only cause of the differences recorded in this study.
Acknowledgments This work was supported by the French Ministry of Research, the Institut National de la Sante et de la RechercheMedicale (INSERM U 666), the Centre National de la RechercheScientifique (CNRS-UNISTRA UMR 7364), and the Universite de Strasbourg (UNISTRA). GAERS and NECs were a gift from Antoine Depaulis, GIN, Grenoble, France. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1468 J. E. Marques-Carneiro et al.
Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Coenen AM, Drinkenburg WH, Inoue M, et al. Genetic models of absence epilepsy, with emphasis on the WAG/Rij strain of rats. Epilepsy Res 1992;12:75–86. 2. van Luijtelaar EL, Coenen AM. Two types of electrocortical paroxysms in an inbred strain of rats. Neurosci Lett 1986;70:393–397. 3. Willoughby JO, Mackenzie L. Nonconvulsive electrocorticographic paroxysms (absence epilepsy) in rat strains. Lab Anim Sci 1992;42:551–554. 4. Buzsaki G, Laszlovszky I, Lajtha A, et al. Spike-and-wave neocortical patterns in rats: genetic and aminergic control. Neuroscience 1990;38:323–333. 5. Buzsaki G, Smith A, Berger S, et al. Petit mal epilepsy and parkinsonian tremor: hypothesis of a common pacemaker. Neuroscience 1990;36:1–14. 6. Semba K, Szechtman H, Komisaruk BR. Synchrony among rhythmical facial tremor, neocortical ‘alpha’ waves, and thalamic non-sensory neuronal bursts in intact awake rats. Brain Res 1980;195:281–298. 7. Vergnes M, Marescaux C, Micheletti G, et al. Spontaneous paroxysmal electroclinical patterns in rat: a model of generalized nonconvulsive epilepsy. Neurosci Lett 1982;33:97–101. 8. Danober L, Deransart C, Depaulis A, et al. Pathophysiological mechanisms of genetic absence epilepsy in the rat. Prog Neurobiol 1998;55:27–57. 9. Van Luijtelaar EL, Ates N, Coenen AM, et al. Role of L-type calcium channel modulation in non-convulsive epilepsy in rats. Epilepsia 1995;36:86–92. 10. Inoue M, Peeters BW, van Luijtelaar EL, et al. Spontaneous occurrence of spike–wave discharges in five inbred strains of rats. Physiol Behav 1990;48:199–201. 11. Rudolf G, Bihoreau MT, Godfrey RF, et al. Polygenic control of idiopathic generalized epilepsy phenotypes in the Genetic Absence Rats from Strasbourg (GAERS). Epilepsia 2004;45:301–308. 12. Gauguier D, Van Luijtelaar G, Bihoreau MT, et al. Chromosomal mapping of genetic loci controlling absence epilepsy phenotypes in the WAG/Rij rat. Epilepsia 2004;45:908–915. 13. Tolmacheva EA, Van Luijtelaar G, Chepurnov SA, et al. Cortical and limbic excitability in rats with absence epilepsy. Epilepsy Res 2004;62:189–198. 14. Tolmacheva EA, Oitzl MS, Van Luijtelaar G, et al. Sress, glucocorticoids and absences in a genetic epilepsy model. Horm Behav 2012;61:706–710. 15. Mares P, Tolmacheva EA. Comparison of cortical epileptic afterdischarges in immature genetic absence epilepsyWAG/Rij rats with those in two other strains (ACI and Wistar). Epilepsia 2007;48:164–167. 16. Jones NC, Salzberg MR, Kumar G, et al. Elevated anxiety and depressive-like behavior in a rat model of genetic generalized epilepsy suggesting common causation. Exp Neurol 2008;209:254–260. 17. Bouilleret V, Hogan RE, Velakoulis D, et al. Morphometric abnormalities and hyperanxiety in genetically epileptic rats: a model of psychiatric comorbidity? NeuroImage 2009;45:267–274. 18. Sarkisova KY, Kulikov MA. Behavioral characteristics of WAG/Rij rats susceptible and non-susceptible to audiogenic seizures. Behav Brain Res 2006;166:9–18. 19. Sarkisova KY, Van Luijtelaar G. The WAG/Rij strain: a genetic animal model of absence epilepsy with comorbidity of depression. Prog Neuropsychopharmacol Biol Psychiatry 2011;35:854–876. 20. Galani R, Duconseille E, Bildstein O, et al. Effects of room and cage familiarity on locomotor activity measures in rats. Physiol Behav 2001;74:1–4. 21. Vergnes M, Marescaux C, Depaulis A, et al. Spontaneous spike and wave discharges in thalamus and cortex in a rat model of genetic petit mal-like seizures. Exp Neurol 1987;96:127–136. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
22. Gundersen HJG, Jensen EBV, Ki^eu K, et al. The efficiency of systematic sampling in stereology – reconsidered. J Microscopy 1999;193:199–211. 23. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 6th ed. New York: Academic Press; 2007. 24. Gundersen HJG, Jensen EBV. Stereological estimation of the volumeweighted mean volume of arbitrary particles observed on random sections. J Microscopy 1985;138:127–142. 25. Patil CK, McCarroll SA. Of rats and men. Cell 2013;154:481–483. 26. Dezsi G, Ozturk E, Stanic D, et al. Ethosuximide reduces epileptogenesis and behavioral comorbidity in the GAERS model of genetic generalized epilepsy. Epilepsia 2013;54:635–643. 27. Vasconcellos LS, Alberti LR, Romeiro JR, et al. Influence of cholestatic jaundice on the weight variance in an experimental model. Rev Col Bras Cir 2012;39:502–507. 28. Rosen B, Schulkin J. From normal fear to pathological anxiety. Psychol Rev 1998;105:325–350. 29. Kanner AM. Epilepsy, depression and anxiety disorders: a complex relation with significant therapeutic implications for the three conditions. J Neurol Neurosurg Psychiatry 2013;84:1. 30. Donner NC, Lowry CA. Sex differences in anxiety and emotional behavior. Pflugers Arch 2013;465:601–626. 31. Cohen AS, Daley M, Siddarth P, et al. Amygdala volumes in childhood absence. Epilepsy Behav 2009;16:436–441. 32. Alemany S, Mas A, Goldberg X, et al. Regional gray matter reductions are associated with genetic liability for anxiety and depression: an MRI twin study. J Affect Disord 2013;149:175–181. 33. Massana G, Serra-Grabulosa JM, Salgado-Pineda P, et al. Amygdalar atrophy in panic disorder patients detected by volumetric magnetic resonance imaging. NeuroImage 2003;19:80–90. 34. Chan CH, Briellmann RS, Pell GS, et al. Thalamic atrophy in childhood absence epilepsy. Epilepsia 2006;47:399–405. 35. Woermann FG, Sisodiya SM, Free SL, et al. Quantitative MRI in patients with idiopathic generalized epilepsy. Evidence of widespread cerebral structural changes. Brain 1998;121:1661–1667. 36. Luo C, Xia Y, Li Q, et al. Diffusion and volumetry abnormalities in subcortical nuclei of patients with absence seizures. Epilepsia 2011;52:1092–1099. 37. Betting LE, Mory SB, Lopes-Cendes I, et al. MRI volumetry shows increased anterior thalamic volumes in patients with absence seizures. Epilepsy Behav 2006;8:575–580. 38. Masis-Calvo M, Sequeira-Cordero A, Mora-Gallegos A, et al. Behavioral and neurochemical characterization of maternal care effects on juvenile Sprague-Dawley rats. Physiol Behav 2013;118:212–217. 39. Sakhai SA, Preslik J, Francis DD. Influence of housing variables on the development of stress-sensitive behaviors in the rat. Physiol Behav 2013;120:156–163. 40. Weaver IC, Meaney MJ, Szyf M. Maternal care effects on the hippocampal transcriptome and anxiety-mediated behaviors in the offspring that are reversible in adulthood. Proc Natl Acad Sci USA 2006;103:3480–3485. 41. Pearce PS, Friedman D, Lafrancois JJ, et al. Spike-wave discharges in adult Sprague-Dawley rats and their implications for animal models of temporal lobe epilepsy. Epilepsy Behav 2014;32:121–131. 42. Kumar G, Jones NC, Morris MJ, et al. Early life stress enhancement of limbic epileptogenesis in adult rats: mechanistic insights. PLoS ONE 2011;6:e24033.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Day-to-day variation in (A) total duration, (B) mean duration, and (C) number of seizures of Wistars and GAERS over the three one-hour periods of recordings. Table S1. Stereological-based volumetric analysis of amygdala, thalamus and hippocampus of GAERS, NECs and Wistars.
Anxiety and locomotion in Genetic Absence Epilepsy Rats from Strasbourg (GAERS): Inclusion of Wistar rats as a second control *†Jos e Eduardo Marques-Carneiro, *†Jean-Baptiste Faure, *Brigitte Cosquer, †Estelle Koning, †Arielle Ferrandon, *Anne Pereira de Vasconcelos, *Jean-Christophe Cassel, and †‡Astrid Nehlig Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
SUMMARY
Eduardo Jose Marques-Carneiro, PhD student at the University of Strasbourg and Federal University of Sao Paulo.
Objective: The Genetic Absence Epilepsy Rats from Strasbourg (GAERS) is a genetic model, derived from Wistar rats by selective breeding. In all previous studies, GAERS were compared to their paired selected strain not expressing spike-and-wave discharges (SWDs), namely nonepileptic controls (NECs). Because the occurrence/ absence of SWDs is of polygenic origin, some other traits could have been selected along with occurrence/absence of SWDs. Therefore, we explored the importance of using a second control group consisting in Wistar rats, the strain of origin of GAERS, in addition to NECs, on locomotion and anxiety in GAERS. Methods: A test battery encompassing home-cage, open-field, beam-walking and elevated plus-maze evaluations was used. In addition, stereologic analyses were performed to assess the volume of thalamus, amygdala, and hippocampus. The occurrence/absence of SWDs was determined in all three strains by electroencephalography (EEG) recording. Results: When compared to NECs and Wistars, GAERS displayed lower exploratory activity and fastened habituation to novelty. In the plus-maze, scores of GAERS and Wistars were similar, but NECs appeared significantly less anxious (possibly in association with increased amygdala volume); evidence for weaker anxiety in NECs was also found in the open-field evaluation. The volumetric study revealed increased thalamic volume in GAERS compared to both control groups. SWDs were present in all GAERS and in 80% of Wistars. Significance: Compared to the original Wistar strain as an additional control group, the selective breeding that generated the GAERS has no incidence on anxiety-related behavior, conversely to the selection of SWD suppression in NECs, in which anxiety is attenuated. These findings point to the importance of using a second control group composed of Wistar rats in studies characterizing the behavioral profile of GAERS. Thereby, possible confusions between occurrence/absence of SWDs and other features that come along with selection and/or differential brain development induced by the genetic mutations are reduced. KEY WORDS: Absence epilepsy, Activity, Anxiety, Second control group, Stereologic based volumetry.
Accepted June 23, 2014; Early View publication July 24, 2014. *Laboratory of Cognitive and Adaptive Neuroscience (LNCA), Faculty of Psychology, UMR 7364, University of Strasbourg – CNRS, Strasbourg, France; †Faculty of Medicine, INSERM U 666, University of Strasbourg, Strasbourg, France; and ‡Hospital Necker, INSERM U 1129, University Paris Descartes, Paris, France Address correspondence to Jean-Christophe Cassel, Laboratoire de Neurosciences Cognitives et Adaptatives, UMR 7364 University of Strasbourg – CNRS, Faculty of Psychology, 12 rue Goethe, F-67000 Strasbourg, France. E-mail: [email protected] Wiley Periodicals, Inc. © 2014 International League Against Epilepsy
Spike-and-wave discharges (SWDs) characterize the electrographic pattern typically observed during absence seizures. SWDs occur spontaneously in many rat strains including inbred strains such as Fischer 344, Brown Norway, and dark agouti,1–3 and outbred strains such as Wistar, Sprague-Dawley, and Long-Evans.3–7 Almost 30% of Wistars present SWDs,8and this percentage increases with aging, reaching 100% by 84–94 weeks.9
1460
1461 Anxiety and Locomotion in GAERS In the 1980s, two independent groups reported the presence of spontaneous SWDs in two different strains derived from selective breeding of Wistar rats. First, in Strasbourg, a selection among Wistar rats that spontaneously expressed SWDs led, after four generations, to a strain named Genetic Absence Epilepsy Rats from Strasbourg (GAERS), in which 100% of the subjects display SWDs. Simultaneously, from the same Wistar strain, rats that never expressed SWDs allowed the development of a control strain, nonepileptic controls (NECs).7 The second strain with spontaneous SWDs, the Wistar Albino Glaxo from Rijswijk (WAG/Rij) was inbred from Wistar rats, but the occurrence of SWDs was observed only once the rats were fully inbred.2 For these reasons, WAG/Rij do have their own control group devoid of SWDs. Therefore, studies on WAG/Rij use either the August Copenhagen Irish rats (ACIs), which do not express any SWD,10 or more often Wistars as control groups. GAERS, WAG/Rij, and NECs are animal models resulting from multigenic selection.11,12 This multigenic selection might have more features than simply the presence or absence of SWDs. This means that the choice of the control group can have a big impact on the interpretation of the results. For example, some studies on WAG/Rij used ACI rats, whereas others used Wistars as controls. Using a strain without any SWD or a strain in which some rats present SWDs might influence the data and their interpretation. In studies regarding Wag/Rij, the use of two control strains—one inbred nonepileptic (ACI) and one outbred (Wistar)—allows the comparison of the effects of genotype (epileptic vs. nonepileptic and inbred vs. outbred) and the possible consequences on multigenic selection and inbreeding on features other than only genetic absence seizures.13–15 The same paradigm might be used with GAERS. However, until now no study on GAERS used the original Wistar strain as a second control group; only NECs were used. An example of a possible interaction is the significantly higher level of anxiety observed in GAERS as compared to NECs,16,17 whereas no difference in anxiety levels were found in WAG/Rij compared to outbred Wistars.18,19 This shows that the multigenic selection at the origin of those strains might have more features than only the presence or absence of SWDs. Thus, an additional control group comprising Wistars would be critical to ensure that strain differences in anxiety levels reflect genuine behavioral or biologic differences and are not biased by consequences of the genetic selection. The present study aimed to evaluate anxiety and locomotion in GAERS compared to either NECs or Wistars from a standard colony. This study was paired with stereologybased measurements of the volume of amygdala, thalamus, and hippocampus in all rats. The objective was to clarify whether NECs can be considered an appropriate and sufficient control group in studies on GAERS.
Materials and Methods Animals Male GAERS (n = 12), NECs (n = 11) from the Institut Jean Roget (Grenoble, France), and Wistar rats (n = 12) from CERJ (L’Arbresle, France) were used for this experiment. Throughout its duration, temperature (21° 23°C) and lighting (12 h/12 h cycle; light period from 7 a.m. to 7 p.m.) were controlled. The rats were placed in individual cages (42 9 26 9 15 cm) with food and water ad libitum. Experiments were performed in accordance with the rules of the European Community Council directive of September 22, 2010 (2010–63) and the French Department of Agriculture (2013–118; License #67–97 for AN, #67–7 for APV, and #67–215 for JCC).The protocol was approved by the ethical animal research committee board of the University of Strasbourg (CREMEAS - Comite Regional d’Ethique en Matiere d’Experimentation Animale de Strasbourg # AL/97/104/ 02/13). Constant efforts were made to minimize discomfort or pain in the animals. The experiments were performed following the principles outlined in the ARRIVE (Animal Research: Reporting on In Vivo Experiments) guidelines and the Basel declaration (http://www. basel-declaration.org). The 3R (Reduce, Refine, Replace) concept has been considered when planning the experiments. Behavioral testing began 2 months after the arrival of rats in the laboratory and was performed between the age of 3 and 6 months according to a timeline that follows the order of presentation below. Behavioral tests Behavioral testing was started 2 months after arrival of the animals. Locomotor activity in the home cage Locomotor activity was measured in the home cages of all rats to avoid any interference of stress. The cages were placed on shelves in a dedicated room. Each cage was traversed by two infrared light beams targeted on two photocells, 4.5 cm above floor level and separated by 28 cm. A 3 h period allowed habituation to the novelty of the test situation.20 The number of longitudinal cage crossings was recorded for each rat by a computer over 48 h. Open field This test taxes an animal’s activity in a novel environment, as well as habituation, anxiety, and exploration. A square open field surrounded by walls that were high enough to prevent rats from jumping out (65 9 65 9 43 cm) was used. The acrylic glass floor was divided into 25 squares of equal size. The rat was placed in the center of the device. The number of squares crossed (total, Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1462 J. E. Marques-Carneiro et al. central, and peripheral) and the total time spent in the center of the device were quantified in 2-min bins over 10 consecutive min. Before each trial, the device was cleaned with 100% alcohol. Sensorimotor coordination in the beam-walking test Assessment of sensorimotor coordination was performed by placing each rat on a 3 cm wide and 200 cm long wooden beam, divided into four segments of 50 cm, and elevated 80 cm above floor level. The beam communicated with the home cage at one end. All rats were trained over 4 days on progressively longer portions of the beam. During the first session, rats were placed on the beam at 50 cm from the goal box (their home cage) for five consecutive trials. In the following sessions, rats had to run on progressively longer portions of the beam (increased by 50 cm). On the fifth day, all rats were tested for three consecutive 200 cm trials, and their performance was scored by an experimenter. For each virtual 50 cm segment traveled on the beam, the score was 1 when the rat traversed the segment with all paws on the upper surface of the beam. A score of 0 was attributed for each segment on which the rat slipped or placed its toes on the side of the beam. The overall score was calculated by adding the scores of the three assays (maximal score = 12). Elevated plus-maze This test uses a device made of two open and two enclosed arms crossing in their middle. It measures the anxiety of the animals based on the contrast between exploratory behavior and tendency to avoid open, enlightened and elevated spaces. A maze made of two black open arms without side walls (50 9 10 9 1.5 cm), two arms closed by side walls (50 9 10 9 40 cm), and a central platform (10 9 10 cm) was placed at 73 cm above the ground. Four halogen lights illuminated the maze (30 lux over open arms and 5 lux over closed arms). The behavior of all rats was recorded using a video camera placed above the maze and monitored by a computer, allowing the experimenter to see the animal’s behavior on a screen. Staying in its home cage, the rat was brought to the test room 5 min before the start of the test (habituation period). To start testing, the rat was placed in the center of the maze, with head toward a closed arm. The rat was free to explore the device during 5 min. The experimenter recorded the displacements using the keyboard (an entry in every part of the maze being assigned to a key on the keyboard). After each trial, the device was cleaned with 100% alcohol. The variables measured were the number of entries in open and closed arms and the time spent in these arms. The percentage of entries and time spent in open arms (open arms/[open arms + closed arms] 9 100) were analyzed to minimize a bias due to possible differences in locomotor activity. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
Electroencephalographic recording GAERS, NECs, and Wistars were anesthetized by an injection of a mixture of ketamine (80 mg/kg, i.p.) and xylazine (30 mg/kg, i.p). The rats were equipped with four single-contact electrodes over the frontoparietal cortex (anteroposterior [AP] 2, mediolateral [ML] 2.5 mm from bregma), as described previously.21 All animals were given a 1-week postsurgical recovery time. Then, they were habituated to the experimental setting over 2 days (placed in recording boxes, the first day without connection to the recording system and the second day with connection). A digital acquisition computer-based system (Coherence, Deltamed, France) was used for the continuous recording of electroencephalography (EEG) activity in freely moving animals. For each recording, the animals were habituated for 15 min to the recording cage. Recording lasted for 2 h (the 20 first and the last 40 min were excluded from the analysis). This procedure was repeated for 5 days, and the first and last days were excluded from the final analysis. To avoid any interference with the sleep-wake cycle, all recordings were carried out in the morning, starting at 10:00 a.m. The number and cumulative duration of SWDs were calculated and represent the mean of three 1-h recording periods. In Wistars, values were calculated only from rats that expressed SWDs. Histologic and stereologic studies (estimation of volume) At completion of behavioral testing and EEG recordings, all rats were killed with an overdose of pentobarbital (200 mg/kg), transcardially perfused with 50 ml saline followed by 50 ml of 0.1 M phosphate-buffered 4% paraformaldehyde (4°C). After extraction, the brains were postfixed for about 4 h and transferred into a 0.1 M phosphate buffered 20% sucrose solution, in which they were kept for 36–40 h at 4°C. The brains were then quickly frozen and cut into 50-lm–thick coronal sections using a cryostat. Every sixth section was sampled using systematic random sampling according to stereologic principles,22 mounted on slides, and stained with cresyl violet. Regions of interest (ROI), that is, the amygdala, hippocampus, thalamus, as well as the whole brain were delineated using a microscope (Leica DM5500B; Leica Microsystems SAS, Nanterre, France), a 2.59 objective, and the Mercator Software (Explora Nova, La Rochelle, France), based on the Paxinos and Watson rat brain atlas.23 Digital live microscope images were visualized using a high-resolution Optronic Microfire camera. The boundaries of amygdala considered all nuclei (bregma 1.80 mm to bregma 3.36 mm). The boundaries of hippocampus were determined by the appearance of the anterior CA3 and dentate gyrus (DG) (bregma 1.80 mm) and the disappearance of the DG at the posterior level (bregma 6.84 mm). The thalamus was considered as a whole (bregma 1.80 mm to bregma 4.68 mm). The whole brain was taken from bregma 1.80 mm to bregma 6.84 mm.23
1463 Anxiety and Locomotion in GAERS Volume estimates were obtained using Cavalieri’s method,24 that is, by summing the area of each region in each image and multiplying it by the section + interval thickness (300 lm). The precision of the estimates is described by the coefficient of error (CE) that in all regions of interest ranged between 0.02 and 0.06. Stereologic data analysis: because a significant difference in brain volume (Table S1) was evidenced between strains, we used data normalized for brain size for each individual region. Statistical analysis Analyses of variance (ANOVAs) were performed. A single “Group” factor design was used for beam-walking and plus-maze scores, a “Group X Interval” design for locomotor activity in the open field, and a “Group X Time” or “Group X Period” design for home cage activity. A post hoc Newman-Keuls test was used to compare different groups after ANOVA and when required. The data concerning the characteristic features of SWDs in GAERS and Wistars were compared with a Student’s t-test for independent samples. All statistical analyses were performed using Statistica software (Statsoft Inc, Maisons-Alfort, France).
Results Behavioral tests Locomotor activity in the home cage One rat from each group was excluded from data analysis because of technical problems in the reliability of infrared sensors. Data from 11 GAERS, 10 NEC, and 11 Wistars were analyzed. Habituation (first 3 hours): For the number of cage crossings (Fig. 1A), ANOVA showed a significant overall Group effect (F2,29 = 3.797, p < 0.05) due to lower activity in NECs compared to Wistars (p < 0.05). ANOVA also showed a significant effect of Time (F5,145 = 102.95, p < 0.001), reflecting a reduction in the overall activity over time. Finally, ANOVA showed a Group X Time interaction (F10,145 = 2.80, p < 0.01) due to activity levels that were significantly lower in GAERS than in Wistars (p < 0.05), but higher than in NECs (p < 0.01) during the first hour. Basal activity: ANOVA did not reveal a significant Group effect during the light period (F2,29 = 0.2921, p = 0.75), compared to the dark period (F2,29 = 4.458, p < 0.05) (Fig. 1B). During the dark period, activity was significantly reduced in GAERS and NECs compared to Wistars (p < 0.05). There was no significant effect of Period or Group X Period interaction. Open field Habituation: ANOVA showed a Group effect associated with lower overall activity in GAERS compared to both control strains (p < 0.001). However, GAERS showed a higher
A
B
Figure 1. Home cage activity. (A) Habituation to a novel environment reflected by activity levels over the first 3 h and shown in six 30min bins. NECs showed lower activity levels than Wistars during the first 30 min. Habituation occurred in all groups, as shown by comparable reduction of activity over time. (B) Basal activity during day and night periods. Day activity was similar in all groups. During the night, NECs and GAERS were significantly less active than Wistars. *p < 0.05, statistically significant difference from Wistars, +p < 0.05, statistically significant intrastrain difference. Epilepsia ILAE
activity than the two control groups during the first two min, but subsequently their activity decreased markedly (Fig. 2A). During the fourth interval (6–8 min), NECs were more active than Wistars. These differences explain the significant Group X Interval interaction (F8,128 = 14.55, p < 0.001). In addition, the activity of Wistars and GAERS gradually decreased over time, which was not the case in NECs. Rearings: ANOVA showed a significant Group effect (F2,32 = 47.12, p < 0.001) due to a lower number of rearings in GAERS compared to Wistars (p < 0.001) and NECs (p < 0.001) (Fig. 2B). Activity in the central area: ANOVA showed a significant Group effect (F2,32 = 9.95, p < 0.001) due to higher activity scores in NECs compared to Wistars (p < 0.01) and GAERS (p < 0.001) (Fig. 2C). Activity in the periphery: ANOVA showed a significant Group effect (F2,32 = 12.65, p < 0.001) due to lower activity scores in GAERS compared to Wistars (p < 0.001) and NECs (p < 0.001) (Fig. 2D). Beam-walking In this test (Fig. 3A), ANOVA showed a significant Group effect (F2,32 = 60.50, p < 0.001) due to impaired performance in Wistars compared to GAERS (p < 0.001) and NECs (p < 0.001). Performance was related inversely to the rats’ body weight (Fig. 3B) and most probably reflected a body weight–induced bias, with Wistars being much heavier than GAERS and NECs (p < 0.001). Elevated plus-maze Concerning the time spent in open arms (Fig. 4A), ANOVA showed a significant Group effect (F2,32 = 14.94, p < 0.001) that was due to a significantly longer time in Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1464 J. E. Marques-Carneiro et al.
A
B
C
A
B
D Figure 3. Sensorimotor coordination assessed in the beam-walking test. (A) Mean score of each group in the test. Wistars exhibited a lower performance compared to NECs and GAERS. (B) Average body weight in three strains measured at the end of the beamwalking test. Wistars were significantly heavier than NECs and GAERS. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
Figure 2. Behavior in the open field. (A) GAERS were more active during the first 2 min and less active during the last 8 min in comparison with the two control groups. NECs showed a relatively constant activity, and the activity of Wistars decreased over time. (B) Number of rearings accounting for exploratory activity. GAERS showed significantly fewer rearings than both control groups. (C) Anxietylike behavior is reflected by avoidance of the central area of the open field. NECs spent significantly more time in the center than did GAERS and Wistars. (D) Locomotor activity is reflected by peripheral activity. GAERS showed lower activity in the periphery than the two other strains. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
NECs compared to GAERS (p < 0.001) and Wistars (p < 0.001). ANOVA also revealed a significant Group effect (F2,32 = 13.41, p < 0.001) on the number of entries in open arms (Fig. 4B); this number was larger in NECs than in GAERS (p < 0.05) or Wistars (p < 0.001). The number of entries in open arms was also larger in GAERS than in Wistars (p < 0.05). EEG recordings Because NEC rats showed no EEG evidence of SWDs, only GAERS and Wistar rats were considered for EEG Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
analyses. As expected, 100% of GAERS presented SWDs, whereas 9 of 11 Wistars exhibited SWDs. The day-to-day variability of SWD features is shown in Figure S1 (Supplemental data). The mean number, total duration, and mean duration of seizures of the 11 GAERS and 9 Wistars over three recording sessions are presented in Figure 5. The t-test showed a significant difference in the total number of seizures (t20 = 4.35, p < 0.001; Fig. 5A), total seizure duration (t20 = 10.96, p < 0.001; Fig. 5B), and mean duration of a seizure (t20 = 9.62, p < 0.001; Fig. 5C). Volumetric analysis In line with the significant difference in body weight between the three groups, the whole brain volume was smaller in NECs and GAERS than in Wistars (Table S1). To standardize measurements, the statistical analysis of volumetric measurements was performed on the ratio between the volume of each structure and the whole brain volume. Amygdala: ANOVA showed a significant Group effect on amygdala volume (F2,29 = 7.35, p < 0.01), which was larger in NECs compared to Wistars (p < 0.01) and GAERS (p < 0.05) (Fig. 6A). Thalamus: ANOVA showed a significant Group effect (F2,29 = 5.09, p < 0.05) on thalamic volume, which was significantly larger in GAERS compared to Wistars (p < 0.05) and NECs (p < 0.05) (Fig. 6B).
1465 Anxiety and Locomotion in GAERS group made of rats from an outbred strain, that is, their common strain of origin, Wistar rats. A
B
Figure 4. Anxiety-like behavior assessed in the elevated plus-maze test. (A) Percentage of time spent in the open arms. NEC rats spent significantly more time in open arms than GAERS and Wistars, indicating reduced anxiety-like behavior. (B) Percentage of open arm entries. This percentage was significantly larger in NECs than in Wistars and GAERS, reflecting reduced anxiety in NECs. The number of open arm entries was higher in GAERS than in Wistars. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE
Hippocampus: ANOVA showed no significant Group effect on hippocampal volume (F2,29 = 0.61, p = 0.55) (Fig. 6C).
Discussion Both GAERS and NECs were derived from the Wistar strain by progressive genetic selection. Selective breeding is used to enhance specific phenotypes or traits.25 However, by using this technique, either genetic variants that contribute to the selected trait (e.g. SWD) rise quickly to fixation (homozygosity), or many other variants may become fixed during inbreeding.25 The SWD-related phenotype in GAERS has a polygenic inheritance.11 Nothing is known, however, about the possible consequences of polygenic inheritance changes unrelated to the SWD phenotype or to its absence in NECs. Thus, in studies comparing only GAERS to NECs, it is impossible to distinguish whether the differences observed are due to genetic changes underlying SWDs or to other genetic changes that accompany the selection of both strains. Therefore, we compared GAERS to their usual control strain, NECs, but also, for the first time, to a second control
Locomotion and sensorimotor coordination Locomotor abilities such as basal and exploratory activity, sensorimotor coordination, and novel environment habituation were assessed because a difference at this level could have biased other measurements. All three strains exhibited similar levels of basal activity in the home cage in the lights-on period, during which all behavioral testing was performed. This means that for all other tests, the performance in the different groups was most probably not affected by basal activity differences. However, during the night and hence the activity period, NECs and GAERS were slightly but significantly less active than Wistars. When exposed to higher novelty in the open field, GAERS showed reduced exploration scores compared to either Wistars or NECs, a difference particularly marked for rearing and peripheral activity. Of interest, when compared to Wistars, WAG/Rij also showed decreased exploratory activity in the open field.18 In addition, in the open field, GAERS showed faster habituation than the two other strains, whereas NECs hardly showed evidence of habituation. However, over a period of 90 min in the home cage activity test, all three strains showed habituation. In the beam-walking test, GAERS and NECs showed good sensorimotor abilities. The narrow width of the bar and the overweight condition of Wistars did not allow a reliable assessment of their coordination. However, evidence for sensorimotor ability in Wistars was reflected independently by their good performance in the Morris water maze (data not shown). The rats were tested on the bar at 22 weeks of age. In accordance with the literature and our own observations, at that age Wistars have reached a mean weight of 420 g, whereas GAERS and NECs are 25–40% lighter.26,27 This is most likely a trait selected along with the absence/presence of SWDs. Anxiety Characterized by somatic, emotional, cognitive, and behavioral components, anxiety can be considered an adaptive response to environmental pressure. However, when anxiety becomes excessive, overwhelming, or uncontrollable, it is considered a disorder. The amygdala and ventral hippocampus are the most important brain regions regulating anxiety-related responses.28 In addition, emotional disorders reflected by pathologic anxiety are often comorbid to epilepsy.29 Recently, a significantly higher anxiety level in GAERS compared to NECs was reported.16,17 Our study showed also increased anxiety in GAERS compared to NECs in the open-field and plus-maze evaluations. However, in these tests, GAERS did not differ from Wistars. These observations reflect that the genetic selection of NECs affected anxiety level, which does not seem to be the case in Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1466 J. E. Marques-Carneiro et al.
A
B
C
D
Figure 5. Electroencephalographic recordings in GAERS and Wistars. Significant differences were recorded in (A) number, (B) total duration, and (C) mean duration of seizures between GAERS (n = 11) and Wistars (n = 9). (E) EEG traces showing an example of SWD in GAERS and Wistars. As expected, NECs showed no SWD. *p < 0.05, statistically significant difference between GAERS and Wistars. Epilepsia ILAE
A
B
C
Figure 6. Stereology-based volumetric analysis. (A) Amygdala volume relative to the whole brain was larger in NECs than in Wistars and GAERS. (B) Thalamic volume relative to the whole brain was larger in GAERS than in Wistars and NECs. (C) Hippocampal volume was similar in all strains. *p < 0.05, statistically significant difference from Wistars, #p < 0.05, statistically significant difference from NECs. Epilepsia ILAE Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1467 Anxiety and Locomotion in GAERS GAERS. Thus, the use of GAERS to study anxiety as a psychiatric comorbidity in epilepsy, as proposed by Bouilleret et al.,17 might not be indicated. In agreement with our data, Sarkisova and Luijtelaar19 showed no difference in the anxiety level between WAG/ Rij and Wistars. Altogether, our data show that because of the genetic selection for the absence of SWDs, NECs should not be used as sole controls when characterizing GAERS. Volumetric measurements Bouilleret et al.17 suggested that the difference of anxiety-like behavior between GAERS and NECs could be related to a difference in the volume of the amygdala. We found an increased volume of amygdala in NECs compared to Wistars or GAERS. This observation is at variance with the data reported by Bouilleret et al.17 In the latter study,17 volumes were assessed using magnetic resonance imaging (MRI), which does not allow delineation of boundaries of the structures as precisely as histology does. In addition, the MRI measures of the previous study encroached onto entorhinal and piriform cortices,17 which might explain the discordant data between the two studies. Moreover, the difference in the amygdala size between both studies could also be explained by the fact that the study by Bouilleret et al.17 used female while we used male subjects, and the level of anxiety is sex dependent.30 In better agreement with our results, no significant difference in the amygdala volume was reported in patients with absence epilepsy compared to normal control subjects.31 Of interest, a smaller volume of the amygdala was found in individuals with a high anxiety level.32,33 Hence, the larger volume of the amygdala measured in NECs might account for their decreased anxiety-like behavior in the open-field and elevated plus-maze tests. It has been suggested that the presence of abnormalities in thalamocortical networks would underlie idiopathic generalized epilepsy, including absence seizures.34 Some studies found increased cortical volume in patients with absence epilepsy.35 Other studies reported neuroanatomic abnormalities in the thalamus, including thalamic atrophy in adolescent and young adult patients34 and children,36 but thalamic hypertrophy was reported in adult patients.37 Our results are consistent with the latter study, as we found a significantly larger thalamic volume in GAERS than in both control groups. Finally, we did not find any significant volumetric difference in the hippocampus, whatever the group. Spike-and-wave discharges To determine the presence/absence, the frequency, and the duration of SWDs, we recorded EEG in the three groups at the end of the behavioral assessments. As could be expected, all GAERS and no NEC presented SWDs. However, about 80% of Wistars displayed SWDs. The
latter observation is in line with literature data reporting a percentage of affected individuals ranging from around 30% in young adult Wistars8 to up to 100% at 84–94 weeks.9 However, in Wistars, the number of SWDs was two times lower, the total seizure duration six times, and the mean seizure duration four times shorter than in GAERS. In conclusion, it appears that GAERS show a faster habituation to a novel environment but explore less than NECs and Wistars. GAERS appear as anxious as Wistars. On the contrary, NECs appear less anxious than GAERS/ Wistars, an observation that might fit with their significantly larger amygdala volume. These data provide a unique opportunity for studying the genetic features underlying differences in anxiety behavior. In addition, the main conclusion of this work is the importance and benefit of a second control group composed of Wistar rats in studies on GAERS. Limitations of the study The data obtained here should be considered with some caution, because not all three strains originate from the same breeder. Indeed, although GAERS and NEC rats were born and raised in the same environmental conditions, hence having presumably received a similar quality of maternal care and attention from animal keepers, Wistars were provided by a commercial breeder. Therefore, and although laboratory animal breeding and care have both to scrupulously respect national and international statutory constraints, it cannot be excluded that prenatal gestation and early postnatal rearing conditions may have differed between Wistars and the two other groups of rats in a way that could also partly affect the herein reported difference in anxiety-like behavior. Indeed, several studies have shown that differences in, for example, maternal care, or simply in the type of bedding material used in cages could affect the level of anxietylike behavior in the offspring as adults.38–40 Moreover, early life stress, but a priori stronger than differences in breeding, could affect epileptogenesis.41,42 It is, however, of interest to note that in any type of measurement performed here, Wistars never differed from the two genetically modified strains. Indeed, sometimes they differed from NECs and sometimes from GAERS, while the consequences of early life environment reported in the literature are usually concordant.38–40 This indicates that early life environment cannot be the only cause of the differences recorded in this study.
Acknowledgments This work was supported by the French Ministry of Research, the Institut National de la Sante et de la RechercheMedicale (INSERM U 666), the Centre National de la RechercheScientifique (CNRS-UNISTRA UMR 7364), and the Universite de Strasbourg (UNISTRA). GAERS and NECs were a gift from Antoine Depaulis, GIN, Grenoble, France. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
1468 J. E. Marques-Carneiro et al.
Disclosure None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
References 1. Coenen AM, Drinkenburg WH, Inoue M, et al. Genetic models of absence epilepsy, with emphasis on the WAG/Rij strain of rats. Epilepsy Res 1992;12:75–86. 2. van Luijtelaar EL, Coenen AM. Two types of electrocortical paroxysms in an inbred strain of rats. Neurosci Lett 1986;70:393–397. 3. Willoughby JO, Mackenzie L. Nonconvulsive electrocorticographic paroxysms (absence epilepsy) in rat strains. Lab Anim Sci 1992;42:551–554. 4. Buzsaki G, Laszlovszky I, Lajtha A, et al. Spike-and-wave neocortical patterns in rats: genetic and aminergic control. Neuroscience 1990;38:323–333. 5. Buzsaki G, Smith A, Berger S, et al. Petit mal epilepsy and parkinsonian tremor: hypothesis of a common pacemaker. Neuroscience 1990;36:1–14. 6. Semba K, Szechtman H, Komisaruk BR. Synchrony among rhythmical facial tremor, neocortical ‘alpha’ waves, and thalamic non-sensory neuronal bursts in intact awake rats. Brain Res 1980;195:281–298. 7. Vergnes M, Marescaux C, Micheletti G, et al. Spontaneous paroxysmal electroclinical patterns in rat: a model of generalized nonconvulsive epilepsy. Neurosci Lett 1982;33:97–101. 8. Danober L, Deransart C, Depaulis A, et al. Pathophysiological mechanisms of genetic absence epilepsy in the rat. Prog Neurobiol 1998;55:27–57. 9. Van Luijtelaar EL, Ates N, Coenen AM, et al. Role of L-type calcium channel modulation in non-convulsive epilepsy in rats. Epilepsia 1995;36:86–92. 10. Inoue M, Peeters BW, van Luijtelaar EL, et al. Spontaneous occurrence of spike–wave discharges in five inbred strains of rats. Physiol Behav 1990;48:199–201. 11. Rudolf G, Bihoreau MT, Godfrey RF, et al. Polygenic control of idiopathic generalized epilepsy phenotypes in the Genetic Absence Rats from Strasbourg (GAERS). Epilepsia 2004;45:301–308. 12. Gauguier D, Van Luijtelaar G, Bihoreau MT, et al. Chromosomal mapping of genetic loci controlling absence epilepsy phenotypes in the WAG/Rij rat. Epilepsia 2004;45:908–915. 13. Tolmacheva EA, Van Luijtelaar G, Chepurnov SA, et al. Cortical and limbic excitability in rats with absence epilepsy. Epilepsy Res 2004;62:189–198. 14. Tolmacheva EA, Oitzl MS, Van Luijtelaar G, et al. Sress, glucocorticoids and absences in a genetic epilepsy model. Horm Behav 2012;61:706–710. 15. Mares P, Tolmacheva EA. Comparison of cortical epileptic afterdischarges in immature genetic absence epilepsyWAG/Rij rats with those in two other strains (ACI and Wistar). Epilepsia 2007;48:164–167. 16. Jones NC, Salzberg MR, Kumar G, et al. Elevated anxiety and depressive-like behavior in a rat model of genetic generalized epilepsy suggesting common causation. Exp Neurol 2008;209:254–260. 17. Bouilleret V, Hogan RE, Velakoulis D, et al. Morphometric abnormalities and hyperanxiety in genetically epileptic rats: a model of psychiatric comorbidity? NeuroImage 2009;45:267–274. 18. Sarkisova KY, Kulikov MA. Behavioral characteristics of WAG/Rij rats susceptible and non-susceptible to audiogenic seizures. Behav Brain Res 2006;166:9–18. 19. Sarkisova KY, Van Luijtelaar G. The WAG/Rij strain: a genetic animal model of absence epilepsy with comorbidity of depression. Prog Neuropsychopharmacol Biol Psychiatry 2011;35:854–876. 20. Galani R, Duconseille E, Bildstein O, et al. Effects of room and cage familiarity on locomotor activity measures in rats. Physiol Behav 2001;74:1–4. 21. Vergnes M, Marescaux C, Depaulis A, et al. Spontaneous spike and wave discharges in thalamus and cortex in a rat model of genetic petit mal-like seizures. Exp Neurol 1987;96:127–136. Epilepsia, 55(9):1460–1468, 2014 doi: 10.1111/epi.12738
22. Gundersen HJG, Jensen EBV, Ki^eu K, et al. The efficiency of systematic sampling in stereology – reconsidered. J Microscopy 1999;193:199–211. 23. Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 6th ed. New York: Academic Press; 2007. 24. Gundersen HJG, Jensen EBV. Stereological estimation of the volumeweighted mean volume of arbitrary particles observed on random sections. J Microscopy 1985;138:127–142. 25. Patil CK, McCarroll SA. Of rats and men. Cell 2013;154:481–483. 26. Dezsi G, Ozturk E, Stanic D, et al. Ethosuximide reduces epileptogenesis and behavioral comorbidity in the GAERS model of genetic generalized epilepsy. Epilepsia 2013;54:635–643. 27. Vasconcellos LS, Alberti LR, Romeiro JR, et al. Influence of cholestatic jaundice on the weight variance in an experimental model. Rev Col Bras Cir 2012;39:502–507. 28. Rosen B, Schulkin J. From normal fear to pathological anxiety. Psychol Rev 1998;105:325–350. 29. Kanner AM. Epilepsy, depression and anxiety disorders: a complex relation with significant therapeutic implications for the three conditions. J Neurol Neurosurg Psychiatry 2013;84:1. 30. Donner NC, Lowry CA. Sex differences in anxiety and emotional behavior. Pflugers Arch 2013;465:601–626. 31. Cohen AS, Daley M, Siddarth P, et al. Amygdala volumes in childhood absence. Epilepsy Behav 2009;16:436–441. 32. Alemany S, Mas A, Goldberg X, et al. Regional gray matter reductions are associated with genetic liability for anxiety and depression: an MRI twin study. J Affect Disord 2013;149:175–181. 33. Massana G, Serra-Grabulosa JM, Salgado-Pineda P, et al. Amygdalar atrophy in panic disorder patients detected by volumetric magnetic resonance imaging. NeuroImage 2003;19:80–90. 34. Chan CH, Briellmann RS, Pell GS, et al. Thalamic atrophy in childhood absence epilepsy. Epilepsia 2006;47:399–405. 35. Woermann FG, Sisodiya SM, Free SL, et al. Quantitative MRI in patients with idiopathic generalized epilepsy. Evidence of widespread cerebral structural changes. Brain 1998;121:1661–1667. 36. Luo C, Xia Y, Li Q, et al. Diffusion and volumetry abnormalities in subcortical nuclei of patients with absence seizures. Epilepsia 2011;52:1092–1099. 37. Betting LE, Mory SB, Lopes-Cendes I, et al. MRI volumetry shows increased anterior thalamic volumes in patients with absence seizures. Epilepsy Behav 2006;8:575–580. 38. Masis-Calvo M, Sequeira-Cordero A, Mora-Gallegos A, et al. Behavioral and neurochemical characterization of maternal care effects on juvenile Sprague-Dawley rats. Physiol Behav 2013;118:212–217. 39. Sakhai SA, Preslik J, Francis DD. Influence of housing variables on the development of stress-sensitive behaviors in the rat. Physiol Behav 2013;120:156–163. 40. Weaver IC, Meaney MJ, Szyf M. Maternal care effects on the hippocampal transcriptome and anxiety-mediated behaviors in the offspring that are reversible in adulthood. Proc Natl Acad Sci USA 2006;103:3480–3485. 41. Pearce PS, Friedman D, Lafrancois JJ, et al. Spike-wave discharges in adult Sprague-Dawley rats and their implications for animal models of temporal lobe epilepsy. Epilepsy Behav 2014;32:121–131. 42. Kumar G, Jones NC, Morris MJ, et al. Early life stress enhancement of limbic epileptogenesis in adult rats: mechanistic insights. PLoS ONE 2011;6:e24033.
Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Day-to-day variation in (A) total duration, (B) mean duration, and (C) number of seizures of Wistars and GAERS over the three one-hour periods of recordings. Table S1. Stereological-based volumetric analysis of amygdala, thalamus and hippocampus of GAERS, NECs and Wistars.
