Life-span characterization of epilepsy and comorbidities in Dravet syndrome mice carrying a targeted deletion of exon 1 of the Scn1a gene

Objective: Dravet Syndrome (DS) is a catastrophic form of paediatric epilepsy associated with multiple comorbidities mainly caused by mutations in the SCN1A gene. DS progresses in three different phases termed febrile, worsening and stabilization stage. Mice that are haploinsufficient for Scn1a faithfully model each stage of DS, although various aspects have not been fully described, including the temporal appearance and sex differences of the epilepsy and comorbidities. The aim of the present study was to investigate the epilepsy landscape according to the progression of DS and the long-term co-morbidities in the Scn1a(+/-) tm1Kea DS mouse line that are not fully understood yet. Methods: Male and female F1. Scn1a(+/+) and F1. Scn1a(+/-) tm1Kea mice were assessed in the hyperthermia model or monitored by video electroencephalogram (vEEG) and wireless video-EEG according to the respective stage of DS. Long-term comorbidities were investigated through a battery of behaviour assessments in ~6 month-old mice. Results: At P18, F1. Scn1a(+/-) tm1Kea mice showed the expected sensitivity to hyperthermia-induced seizures. Between P21 and P28, EEG recordings in F1. Scn1a(+/-) tm1Kea mice combined with video monitoring revealed a high frequency of SRS and SUDEP. Power spectral analyses of background EEG activity also revealed that low EEG power in multiple frequency bands was associated with SUDEP risk in F1. Scn1a(+/-) tm1Kea mice during the worsening stage of DS. Later, SRS and SUDEP rates stabilized and then declined in F1. Scn1a(+/-) tm1kea mice. Incidence of SRS ending with death in F1. Scn1a(+/-) tm1kea mice displayed variations with the time of day and sex, with female mice displaying higher numbers of severe seizures resulting in greater SUDEP risk. F1. Scn1a(+/-) tm1kea mice ~6 month-old displayed fewer behavioural impairments than expected including hyperactivity, impaired exploratory behaviour and poor nest building performance. Significance: These results reveal new features of this model that will optimize use and selection of phenotype assays for future studies on the mechanisms, diagnosis, and treatment of DS. E,F,G,H, no sex difference was observed on these parameters. I,J,K,L, Subgroup analysis of male (n=11) and female (n=10) F1. Scn1a(+/-) tm1kea mice revealed no significant difference in any anxiety-like behaviour parameters. Mann-Whitney test, Whisker plots, median IQR spontaneous locomotion in the Open field and E,F,G autism-like behaviour in the open field and three chamber test. Similarly, no significant difference was found between male and female F1. Scn1a(+/-) tm1kea mice in parameters related to anxiety in the H, elevated plus maze, I, Light dark box or memory deficits in J, Y- maze spontaneous, K, Y-maze forced and L, in the novel object recognition. Similar performance between male and female F1. Scn1a(+/-) tm1kea mice was observed in the M, nest building assessment and N,O, exploratory behaviour in the marble burying test. (A,B,C,,E,J,K,L) t-test, mean Two-way ANOVA, Scheirer-Ray Hare test, Mann-Whitney test and (M) Two-way repeated measures ANOVA, abstract Main findings of Epilepsy and behaviour phenotypes in F1.Scn1a(+/-) tm1kea mice. Febrile stage : At P18, males and females F1.Scn1a(+/-) tm1kea mice similarly show sensitivity to hyperthermia-induced seizures at temperatures ~39.4 C to 41.2 C, with duration falling into two categories (short ~10s and long duration ~20s). Worsening stage: From P20 to P28/32, F1.Scn1a(+/-) tm1kea mice experience a high frequency of SRS and SUDEP. Interestingly, females displayed higher frequency of severe SRS and SUDEP than males F1.Scn1a(+/-) tm1kea mice. Stabilization stage: After P28-P32, SRS and SUDEP rates stabilized and then declined, nevertheless females F1.Scn1a(+/-) tm1kea mice remained experiencing higher frequency of SRS ending with hindlimb extension and SUDEP than males. Lastly, F1.Scn1a(+/-) tm1kea mice ~ 6 month old displayed some behaviour impairments caraterized by hyperactivity, impaired exploratory behaviour and poor nest building performance. No sex differences were observed in the behaviour phenotype of F1.Scn1a(+/-) tm1kea mice.


INTRODUCTION
Dravet Syndrome (DS) is a rare, intractable and catastrophic form of childhood epilepsy with an estimated incidence of 1 in 17,700 to 40,000 births . Nearly 90% of DS patients carry de novo heterozygous mutations in the SCN1A gene leading to haploinsufficiency of the type 1 voltage-gated sodium channel α subunit (Nav1.1) . Loss of Nav1.1 protein mainly affects parvalbumin-expressing interneurons, causing disruption of excitation and inhibition balance in several neuronal circuits (Yu et al., 2006). Recent studies show this impairment is transient, however, and the mechanisms by which SCN1A deficiency contributes to cognitive and other phenotypes remains incompletely understood . DS is primarily characterized by hyperthermia sensitivity, spontaneous recurrent seizures (SRS) and premature death. Apart from epilepsy, a global development delay, hyperactivity, intellectual disability, and autistic like-behaviour may be also present . The behavioural impairments and epilepsy phenotype emerge in an age-dependent manner according to three different stages of DS termed 'febrile', 'worsening' and 'stabilization' stage . Thus, a comprehensive understanding of DS features in each disease stage is crucial for successful preclinical development and evaluation of new targeted therapies or biomarker discovery.

Hyperthermia-induced seizures
Hyperthermia-induced seizure threshold assay was performed at P18 as previously described . First, the mouse was gently hand-restrained in a supine position with tail lifted. Then, a temperature probe (RET-4, physitemp) covered with Vaseline was inserted into the rectum and taped on the tail, to keep it in place throughout the procedure. Later, animals were placed into a Plexiglass box with an infrared heat lamp (HL-1, physitemp, Clifton, New Jersey) positioned above and the rectal probe attached to a TCAT-2DF thermocontroller (physitemp, Clifton, New Jersey). Mice were held at 37.5°C for 5 min to become accustomed to the chamber and then core body temperature was gradually elevated by 0.5°C every 2 min until a seizure occurred or 42.5 °C was reached. If the mouse had a seizure, the heating process was stopped immediately to cooled down the mouse to 37°C in a cold metal surface.
Animals not presenting seizures were held for 3 min until 42.5°C and then the heat lamp was turned off. Mice remained 5 minutes into the chamber for observation of any eventual late occurrence of seizure before they are removed, cooled down and considered seizure free. Seizure severity was classified according to the Racine scale scoring system with few modifications .

Seizure semiology
All seizures detected in each stage of DS were scored according to the Racine scale with a few modifications, as reported: . No behaviour changes (0), mouth and facial movements (1), head nodding (2), unilateral forelimb clonus (3), bilateral forelimb clonus with rearing (4), rearing and falling (loss of posture) (5), wild running or jumping (6) and Tonic hindlimb extension possibly leading to death (7). J o u r n a l P r e -p r o o f

Behaviour phenotype
Naïve 5 to 6-month-old mice were submitted to a battery of behavioural assessments. Spatial reference, working and recognition memory was assessed with the Y maze and novel object recognition test, respectively .
Spontaneous locomotion, stereotyped behaviour, and anxiety-like behaviour was investigated in the open field. Further investigation related to anxiety-like behaviour was performed in the elevated plus maze and light dark box test as previously described (Almeida Silva et al., 2020). Autism-like behaviour and motor coordination were assessed in the three chamber and rotarod test, respectively . Finally, innate exploratory behaviour and welfare parameters were assessed through marble burying and the nest building task . Behavioural tests were performed with multiple rest day intervals and according to the increasing order of interventional complexity.

Statistical analyses
The normality of the data was analysed using D'Agostino and Pearson's omnibus normality test. Data were analysed using unpaired two-tailed Student t test, Mann-Whitney U test, Kaplan-Meier method, Spearman's rank-order correlation, Two-way repeated measures ANOVA, one and two-way analysis of variance (ANOVA), Fisher's exact test, Wilcoxon signed-rank test, Friedman test or Permutation test followed by Genotype differences were investigated using a mix of male and female animals in each group F1.Scn1a(+/+) and F1.Scn1a(+/-) tm1Kea mice. Further analyses on sex differences related to the epilepsy phenotype were focused on male vs female F1.Scn1a(+/-) tm1Kea mice, since no seizures were observed in F1.Scn1a(+/+) mice across the three different stages of the disease. Sex influence on behaviour phenotype was investigated performing a two-way analysis of variance (ANOVA) considering sex a factor with the respective groups F1.Scn1a(+/+) (male/female) and F1.Scn1a(+/-) tm1Kea (male/female) mice. Finally, sub-group analyses comparing male vs female F1.Scn1a(+/-) tm1Kea mice were performed to completely rule out sex influence on behaviour phenotype. The specific statistical test used for each experiment are indicated in the figure legends. Data are expressed as mean ± SEM or median with interquartile range (IQR), as appropriate. Differences between groups were considered statistically significant when p<0.05. Experiments and data were analysed blind to genotype.

RESULTS
3.1 F1.Scn1a(+/-) tm1kea P18 mice display sensitivity to hyperthermia-induced seizures during the febrile stage of DS Febrile seizures as a result of sensitivity to hyperthermia are a hallmark of DS onset.
Spontaneous seizures were least severe at P21. From P22 to P25 F1.Scn1a(+/-) tm1kea mice experienced the most severe SRS ( Figure 3F). The total number of SRS with severity 6 and 7 observed in females F1.Scn1a(+/-) tm1kea mice was twice that in males (p = 0.015 and p = 0.041 respectively, Figure 3G). We also observed that SRS with severity 6 and 7 were of a longer duration than SRS with severity 5 (p = 0.016; p = 0.0008, Figure 3H), and a strong positive correlation was found between SRS duration and seizure severity (r = 0.7019, p < 0.0001, Figure 3I). Next, we investigated the time J o u r n a l P r e -p r o o f Journal Pre-proof of day during which SRS were most likely to occur. Figure 3J shows a plot of 107 SRS recorded in F1.Scn1a (+/-) tm1kea mice during monitoring between P21 and P28 according to the time of day. Interestingly, the number of SRS peaked just before the light-dark cycle lights went off (20:00 h). However, no difference was found on the number of SRS when dividing the day into 8 h segments ( Figure 3K) and the average number of SRS over the course of a full day did not differ between light and dark phases ( Figure   3L).

F1.Scn1a(+/-) tm1kea mice display higher incidence of SUDEP and reduction in background EEG power during the worsening stage of DS
SUDEP is a prominent feature of DS especially during the worsening stage of the disease (stage 2). A Kaplan-Meier plot of deaths revealed that F1.Scn1a (+/-) tm1kea mice display a critical period of SUDEP risk from P22 to P25, with a survival rate of 20% at P28 (p = 0.0003, Figure 4A). There was no sex difference in the occurrence of SUDEP in F1.Scn1a(+/-) tm1kea EEGimplanted mice between P21 to P28 ( Figure 4B). All sudden deaths experienced by F1.Scn1a(+/-) tm1kea mice were preceded by a severe SRS following a stereotypical progression. All pre-SUDEP SRS began with forelimb clonus, rearing and loss of balance/posture (GTCS-Racine scale 5) followed by wild running or jumping (Racine scale 6) ending with tonic hindlimb extension and possibly death (Racine scale 7) (Movie S3). No difference in the number of hindlimb extension seizures was found in any period of the day ( Figure 4C). However, SUDEP incidence was higher in the last period of the day (16:00-00:00) (p = 0.003, Figure 4D). EEG power spectral analyses in the interictal period revealed low EEG power in multiple frequency bands in the period of high SUDEP incidence from P22 to P24/P25 in F1.Scn1a(+/-) tm1kea mice ( Figure   4E,F,G). Consistently, no SUDEP or reduction in EEG background power was observed at P21 in F1.Scn1a(+/-) tm1kea mice. Figure 4 H,I,J shows representative EEG traces of F1.Scn1a(+/+) and F1.Scn1a(+/-) tm1kea mice respectively. As can been observed, SUDEP in F1.Scn1a(+/-) tm1kea mice was preceded by a severe SRS ending with hindlimb extension in both sexes.

F1.Scn1a(+/-) tm1kea mice
J o u r n a l P r e -p r o o f SRS and SUDEP tend to reduce across adulthood during the stabilization stage of DS . Thus, we next equipped F1.Scn1a mice with implantable EEG telemetry units to track the occurrence of SRS and SUDEP in young adult mice (P36-P49). Monitoring during this period detected SRS in only 1 out of 5 F1.Scn1a(+/-) tm1kea mice ( Figure 5A). J o u r n a l P r e -p r o o f P0 -6 month old confirmed that more than 50% of F1.Scn1a(+/-) tm1kea mice experience SUDEP from between P20 to P32 which later on tend to stabilize. H, female F1.Scn1a(+/-) tm1kea mice displayed higher number of SUDEP when compared to male F1.Scn1a(+/-) tm1kea mice. (B) Wilcoxon signed-rank test, Whisker plots, median IQR, (C) Fisher's exact test, (G,H) Log-rank Mantel-Cox test *p<0.05, **p<0.01.
As expected, no seizures were observed in F1.Scn1a(+/+) control mice. Furthermore, all SRS observed over this period were less severe and did not reach the maximum score on the Racine scale ( Figure 5B) and no deaths were recorded ( Figure 5C). Figure   5D,E and F show representative EEG traces of one F1.Scn1a(+/+), one F1.Scn1a(+/-) tm1kea mice (without SRS) and one F1.Scn1a(+/-) tm1kea mouse experiencing a SRS respectively. Later, using another batch of animals, we investigated the long-term survival in naïve mice. A Kaplan Meier plot revealed that the premature death commences at P20 culminating in overall mortality of ~40% at the end of 6 months (p = 0.002, Figure 5G). More than half of the deaths (56.25%) occurred within a short interval between P20 to P32 (worsening stage). Thereafter, the occurrence dropped, tending to stabilize over the period until 6 months ( Figure 5G). Interestingly, a subgroup analysis of SUDEPs found a sex difference with higher mortality rates in female F1.Scn1a(+/-) tm1kea mice (p = 0.035, Figure 5H).

DISCUSSION
In the present study we provide the most comprehensive assessment of the phenotypes over the lifetime in the F1.Scn1a(+/-) tm1kea model of DS and identify a number of features including testing age and sex that may improve the design and execution of therapeutic and biomarker studies using this DS model. DS is commonly classified in three different stages accordingly to the clinical manifestations . The "febrile stage" of DS is marked by high incidence of hyperthermia sensitivity often resulting in prolonged seizures . Temperature sensitivity is a conserved feature of mouse models of DS and here we confirmed that F1.Scn1a(+/-) tm1kea mice present a reduced threshold to the hyperthermia-induced seizures at early developmental stages  Here we found that hyperthermia-induced seizures in P18 F1.Scn1a(+/-) tm1kea mice fall into two types (short ~10s and long duration ~20s) with distinct behaviour which was not influenced by the sex or temperature reached during the hyperthermia challenge.
Such a distinct seizure phenotype has not been previously reported in DS mouse models indicating that it is a specific feature only observed in Scn1a(+/-) tm1kea mouse line.
In children with Dravet, the "febrile stage" is followed by a "worsening stage" that extends up to the fifth year of life with increasing SRS frequency and severity. This has also been observed from P21 to P28/P30 in the Scn1a+/-knockout mouse model . Similarly, our results reveal that F1.Scn1a(+/-) tm1kea mice present a high number of severe SRS over this stage. We also observed that the frequency of SRS was higher in female than male F1.Scn1a(+/-) tm1kea mice. Moreover, the number of the most severe SRS featuring wild running or jumping (score 6) and hindlimb extension (score 7) were twice as high in female compared to male F1.Scn1a(+/-) tm1kea mice. Although such sex differences in SRS have not been previously reported in DS mouse models and patients, a previous study in F1.Scn1a(+/-) tm1kea mice reported higher female SUDEP rates in DS mice J o u r n a l P r e -p r o o f . Thus, these findings indicate a distinct epilepsy phenotype according to the sex of F1.Scn1a(+/-) tm1kea mice.
There is increasing evidence that circadian rhythms affect brain excitability . Previous studies showed that Scn1a R1407X/+ DS mice experience a peak of SRS ending with death between 18:00 to 19:00 (before the lights went out), . Here, F1.Scn1a(+/-) tm1kea mice also experience a peak of SRS before the lights went off (19:00-20:00) with a higher incidence of SRS ending with death from 16:00 to 00:00. Thus, the time of day may have an influence on SRS followed by SUDEP in F1.Scn1a(+/-) tm1kea mice. While the mechanism is unknown, it is possible that network effects of clock-related genes may intersect with the defects arising from Scn1a loss. Therefore, pre-clinical studies using F1.Scn1a(+/-) tm1kea mice should take account a proper monitoring of SRS, particularly if non-continuous monitoring is planned.
The worsening stage of DS in humans is marked by a high incidence of SUDEP, typically preceded by a severe GTCS .
Here, EEG and video monitoring revealed that all deaths experienced by F1.Scn1a(+/-) tm1kea mice were preceded by a severe GTCS ending with hindlimb extension. SUDEP incidence was highest from P21 to P28, around 80%, and this had no sex bias.
Interestingly, neither SUDEP nor changes in background EEG patterns were observed at the beginning of the worsening stage (P21). In contrast, SUDEPs observed from P22 to P25 were associated with loss of EEG power during the interictal period in multiple frequency bands including beta, gamma and alpha. This matches findings in Scn1a A1783V missense mutation mice that showed low EEG power correlated with the risk of premature death during the worsening stage of DS . Thus, these findings not only characterise the second stage of DS in F1.Scn1a(+/-) tm1kea mice but J o u r n a l P r e -p r o o f also reinforce that a severe GTCS and low background EEG power may be an immediate risk factor for SUDEP.
The severity and frequency of seizures in humans and mouse models of DS is known to decline in adulthood, often replaced by escalating co-morbidities . EEG telemetry recordings in F1.Scn1a(+/-) tm1kea mice confirmed this, with a lower incidence of SRS and SUDEP into adulthood (P36-P49). In naïve F1.Scn1a(+/-) tm1kea mice followed for 6 months, we found survival rates of ~60%. More than half of deaths occurred in a short period of time from P20 to P32, tending to decline thereafter until P180 (6 months). We observed a much lower survival rate in EEG implanted compared to non-instrumented F1.Scn1a(+/-) tm1kea mice from P21-P28 (~20% vs ~80%, respectively). These findings indicate that instrumentation or daily handling may exacerbate the epilepsy phenotype and should be take into consideration when planning preclinical studies. In general, our results confirm the critical SUDEP period occurs from P21 to ~P30, followed by a mitigation period thereafter . Together, these findings not only characterize the third stage of DS but also reinforces that the epilepsy phenotype in F1.Scn1a(+/-) tm1kea mice progresses in an age-dependent manner.
A previous study using F1.Scn1a(+/-) tm1kea mice found that females are more susceptible to SUDEP . We found a similar increased incidence of SUDEP in naïve female F1.Scn1a(+/-) tm1kea mice. This is probably because of the higher incidence of seizures. The mechanism of increased susceptibility to SRS and SUDEP in female F1.Scn1a(+/-) tm1kea mice is unknown. Factors such as fluctuations in sex hormones or GABA turnover have been found to differ between male and female rodents . Further studies are needed to define the causes of J o u r n a l P r e -p r o o f sex differences in DS phenotypes over the different stages of DS in the Scn1a(+/-) tm1kea mouse model. DS is also accompanied by multiple long term comorbidities including hyperactivity, social impairments, anxiety and cognitive decline resulting in a poor quality of life . We report the first comprehensive assessment of long-term behavioural phenotypes of F1.Scn1a(+/-) tm1kea mice during the stabilization stage of DS. Hyperactivity is one of the most consistent findings across different ages of development in individuals with DS . F1.Scn1a(+/-) tm1kea mice exhibited hyperactivity when exposed to novel environments but had normal motor coordination. This is in agreement with the higher total distance travelled reported in the open field in 8 week old male F1.Scn1a(+/-) tm1kea mice . Autistic features including repetitive behaviour and social deficits are also observed in DS (Han et al., 2012;Ito et al., 2013;. We found that F1.Scn1a(+/-) tm1kea mice exhibited a prominent stereotyped behaviour in the open field arena which may be an autistic-like trait. However, F1.Scn1a(+/-) tm1kea mice did not exhibit social deficits in the three chamber test. Thus, the repetitive behaviours observed in F1.Scn1a(+/-) tm1kea mice may reflect hyperactive rather than autistic features, as concluded for Scn1a WT/A1783V mice . DS patients are reported to have anxiety and especially cognitive deficits  and anxiety-like behaviour phenotypes are present in 6-8 weeks old F1.Scn1a(+/-) tm1kea mice . Similarly, 6 weeks old F1.Scn1a(+/-) tm1kea mice showed impaired reference and working memory in the radial arm maze . In contrast, 8 weeks old F1.Scn1a(+/-) tm1kea mice may display a better novel object recognition memory J o u r n a l P r e -p r o o f when compared to WT mice and just slight decline in spatial memory in the Barnes maze test . Here we consistently found that ~6 month old F1.Scn1a(+/-) tm1kea mice did not present any anxiogenic phenotype across the peripheral regions of the open-field, elevated plus maze or light dark box assessment. Furthermore, ~6 month old F1.Scn1a(+/-) tm1kea mice did not display spatial reference, working or recognition memory deficits in the forced/spontaneous Y maze or NOR assessment respectively. Together, these results suggest that F1.Scn1a(+/-) tm1kea mice may experience transient behaviour changes that do not persist into the later stages of life, as reported for the pathophysiology of DS in young adult F1.Scn1a(+/-) tm1kea mice . Most DS patients experience a lifelong debilitating condition affecting their daily activities and resulting in a poor quality of life . In mice, the innate tendency to hide objects in the marble test and nest building activity are relevant measures to compare an individual's performance in daily life activities . Indeed, ~6 month old F1.Scn1a(+/-) tm1kea mice showed impaired exploratory behaviour as indicated by a reduction in the number of buried and displaced marbles. In addition, F1.Scn1a(+/-) tm1kea mice displayed a poor performance in the nest building assessment.
These results suggest that this epileptic encephalopathy is compromising the welfare and quality of life of F1.Scn1a(+/-) tm1kea mice as similarly observed in DS patients.

Genotyping of F1.Scn1a(+/-) tm1Kea mice
Tissue was collected by tail snipping before P7. DNA was extracted and subject to PCR using KAPA2G Fast Genotyping Mix (KAPA Biosystems, UK). Primers used for genotyping are described in the table below.

Acute video-EEG recordings of SRS and death
At P21, another cohort of mice were injected via i.p with buprenorphine (0.3mg/ml) and placed in an adapted stereotaxic frame under anaesthesia (isoflurane/oxygen 5% for induction and 3% for maintenance). Body temperature was maintained by a feedback-controlled heat blanket. After topical application of EMLA cream 5%, a midline scalp incision was performed, and three screw electrodes were implanted and secured with dental cement and special glue. The screw electrodes were placed bilaterally to the midline over the cerebral cortex followed by the reference electrode positioned over the nasal sinus. After surgery, animals were immediately placed in an incubator at 33°C and monitored for 30 min. Once fully recovered, mice single housed were connected to the lead socket of a swivel commutator, which was connected to a brain monitor amplifier for EEG digital recordings. Gel diet was added

Video-EEG monitoring of SRS and death
Immediately after acute vEEG recordings, single housed mice in their home cages were transferred to a room equipped with a high resolution, infrared video cameras (Hikvision). Continuous digital videos were recorded at 30 fps and stored in a Dell PC workstation. Video's recording was offline review at 16x speed using VSplayer  . When a mouse was found dead in a J o u r n a l P r e -p r o o f cage, the video was reviewed to determine time of death and whether it was preceded by a severe GTCS ending with full hindlimb extension.

Long term continuous video-EEG recordings of SRS, SUDEP and survivals analyses in naïve mice
At P36, another group of mice were implanted with telemetry devices for long term vEEG recordings. The telemetry device (Model: F20-EET, Data Systems International) was inserted subcutaneously along the dorsal flank. Four electrodes were connected to skull-mounted screws, 2 references electrodes were placed in the nasal sinus, one electrode was placed over the right cerebral cortex and another electrode placed over the left cerebral cortex. Continuous vEEG monitoring (telemetry system; Data Sciences International, USA) was performed 24/7 for 2 weeks until P49. Finally, another batch of naïve mice was used to monitor the survival rates from P0 to 6 months of age, period that long-term comorbidities assessment started as described below.

Behavioral experiments
Animals reaching 5 to 6 months of age were submitted to a battery of behavioral tests designed with multiple resting days interval and according to increasing order of invasiveness. 2 weeks before starting the behavioral assessment, animals were moved to the animal unit equipped with all behavioral apparatus to get familiarized with new environment. 1 week before starting the battery of behavioral tests, all mice were handled during 5 min in three different days with interval of 24 hours in the behaviour unit to get familiarized with the researcher and environment. After each behaviour session, apparatus and objects used were cleaned with 70% ethanol to avoid any potential bias. All behavioral sessions were video recorded by an overhead Logitech J o u r n a l P r e -p r o o f Webcam C270 (720p) and analysed with ANY-maze video tracking system (Version 6.32) when appropriated.

Forced alternation test (Y-maze)
Spatial reference memory was assessed as previously described . The mouse was placed into the end of the start arm, facing the wall and away from the centre. In the trial, mouse was allowed to explore during 5 min the two arms of the Y-maze, while entry into the third arm was blocked. After the sample trial, the mouse was returned to its home cage for a 60 min inter-trial interval. In the trial 2, the block in arm 3 was removed and the mouse was again placed into the start arm, and then allowed to access all three arms of the maze. If a mouse climbed on the maze wall, it was immediately returned into the abandoned maze arm. An arm entry was recorded when 85% of a mouse's body entered the arm. Time in Novel Arm [%] was defined as the time spent in the novel arm divided by the time spent in all arms during the first minute of the retrieval trial T2. Forced Alternation [%] was defined as the percent of mice entering first the novel arm during T2. Mice with less than three arm entries in the first minute of T2 were excluded from the analysis.

Novel object recognition
Recognition memory was measured using the novel object recognition task as previously described (Sik et al., 2003;. During habituation ( Video recordings were manually offline reviewed by an experienced researcher.
Exploration was defined as sniffing or touching the objects with the nose and/or forepaws. Sitting on or turning around the objects was not considered exploratory behaviours. Data acquired during the habituation stage of NOR in the open field was used quantify the spontaneous locomotion and anxiety-like behaviour.

Spontaneous alternation test (Y-maze)
Spatial working memory test was conducted using a symmetrical Y-maze (35 cm x 5 cm x 60 cm) with the floor and walls made of white and clear Plexiglass respectively as previously described.  For this test, the Y-maze was rotated by 45° and the distal cues (geometrical marks) were differed between the forced alternation tests. During a single 5 min trial, mice freely explored all three arms of the Y-maze containing different geometrical marks (black and/or white) on the wall of each J o u r n a l P r e -p r o o f arm. The start arm was varied across genotypes to avoid placement bias. Mice that climbed on the maze walls were immediately returned to the abandoned arm.
Spontaneous alternation [%] was defined as consecutive entries in 3 different arms (ABC), divided by the number of possible alternations (total arm entries minus 2). Mice with less than 8 arm entries during the 5-min trial were excluded from the analysis.

Three chamber tests of social preference
The three-chamber apparatus (Ugo Basile Sociability Cage/code 46503) is a transparent plexiglass box (60 x 40 x 22 (h) cm) divided into three identical chambers separated by two transparent partitions. Each partition has a square opening (5 x 5 cm) in the bottom centre that offer free access to any partitions. Three chambered social preference test consists in 3 sessions respectively named as habituation, sociability and social novelty session performed as previously described with modifications . Test mice were confined in the centre chamber at the beginning of each session. In the habituation session, a test mouse was placed in the centre of the threechamber unit to freely explore the area during 10 min. Then, for the sociability session, a stranger age-and sex-matched C57BL/6J mouse (S1) that had never been exposed to the test mouse was placed in one of the two wire cups leaving the opposite wire cup empty. The location of the S1 alternated between the two side chambers across test mice. During the social novelty phase, a stranger mouse (S2) was enclosed in the wire cup that had been empty during the sociability phase. Thus, the test mouse would have the choice between a mouse that was already familiar (S1) and a new stranger mouse (S2). Exploration of an enclosed mouse or a wire cup was defined as when a test mouse oriented toward the wire cup with the distance between the nose and wire cup less than 1 cm. C57BL/6J mouse were habituated in the three chamber apparatus 1 week before J o u r n a l P r e -p r o o f the tests during 10 minutes into the wire cup in three different days to avoid any potential stress that could interfere in the test day.

Elevated plus maze (EPM)
The

Marble burying
Marble burying was used to assess the innate exploratory behaviour of mice . The test was conducted in opaque Perspex cages (35 cm × 16,5 cm × 14 cm) containing normal bedding (5 cm deep) overlaid with 18 glass marbles (15 mm diameter) equidistant in a 3 × 6 arrangement. For testing, a single mouse was placed in the cage with the lid closed for 30 min. At the end of the trial, the mouse was removed The light/dark task was performed as previously described (Takao and Miyakawa, 2006  variable.

Rotarod
Motor coordination and balance were evaluated on rotarod (Harvard apparatus model LE 8500), accelerating from 4 to 40 rpm in 300 s . The mouse ability to maintain balance and keep pace with a rotating road were evaluate during 5 min over three trials with interval of 15 min. Latency to fall was recorded for each trial.

Nest building
Single-housed mice were transferred into a new cage with 100g of soft cob bedding containing 10g of nest-building material cut into square (Whatman filter paper, J o u r n a l P r e -p r o o f Journal Pre-proof grade 201, Sigma). After 6, 24, and 48 h, the weight of intact nest material was assessed to determine the nest building ability as previously described . Tables   Table S1. Primers used for genotyping of F1.Scn1a(+/-) tm1Kea mice.