Mouse lines, genotyping and ethical statement
Two mouse models were used in the study. The 16p11.2 mouse model corresponds to the Del(7Sult1a1-Spn)6Yah mouse model[20], noted Del/+ here. The line was kept on a pure C57BL/6N inbred genetic background. The deletion allele was identified by PCR using primers Fwd1 (5’-CCTGTGTGTATTCTCAGCCTCAGGATG-3’) and Rev2 (5’-GGACACACAGGAGAGCTATCCAGGTC-3’) to detect a specific band of 500 bp while the wild-type allele was identified using Fwd1 and Rev1 (5’ -GGACACACAGGAGAGCTATCCAGGTC- 3’) primers to detect the presence of a 330 bp fragment. The PCR program was: 95 °C / 5 min; 35× (95 °C / 30 s, 65 °C / 30 s, 70 °C / 1 min), 70 °C / 5 min.
The Kctd13em2(IMPC)Ics knock-out mouse was generated by the CRISPR / Cas9 technology[27] in the C567BL/6N genetic background (Sup Fig.1A). Two pairs of sgRNAs, one pair located upstream and the other pair downstream of the target region, were selected to delete exons 3 and 4 of the gene. Both pairs of sgRNAs (showing a cut) and Cas9 mRNA were microinjected into fertilized eggs derived from super-ovulated sexually immature C57BL/6N female mice (4–5 weeks olds). Injected embryos cultured in vitro were implanted in the oviducts of pseudo-pregnant females. The deletion of Kctd13 in Kctd13em2(IMPC)Ics (noted here Kctd13+/-) was confirmed by PCR using primers Ef (5’-ACCTCTTAGCTGGGCATGCTAAATT-3’) and Xr (5’-AGCCTATGCTAACTATTATCACAGG-3’) and the sequence of the deleted fragment. The PCR reaction gave deletion and wild-type products of 429 and 668 bp long, respectively (Sup Fig.1B). The PCR program was: 94°C / 5 min, 35 X (94°C / 30 sec; 60°C / 30 sec; 72°C / 30 sec), 72°C / 5 min. This set of primers was also used for genotyping (Sup Fig.1B). The model was validated by detecting the decreased level of KCTD13 in the mutant hippocampi compared to control (Sup Fig.1C). All the mouse models are available through the Infrafrontier European repository or the International Mouse Phenotyping consortium.
Chronic fasudil treatment
In this study, we developed a protocol for a pre-clinical treatment (Figure 1) with the drug fasudil hydrochloride or HA1077 (F4660, LC laboratories Boston, MA, USA). At weaning, control wild type littermates and heterozygous male mice, either Kctd13+/-or Del/+, were taken from several litters and housed in groups of 4-2 individuals in ventilated cages (Green Line, Techniplast, Italy), where they had free access to water and diet (D04 chow diet, Safe, Augy, France). Animal bedding (Poplar litter, AB 3 autoclavable, AniBed, Pontvallain, France) was changed once a week. At 11 weeks, animals were transferred from the animal facility to the phenotyping area. The temperature was kept at 21±2 °C, and the light cycle was controlled as 12 h light and 12 h dark (lights on at 7 am).
At 12 weeks of age, three independent cohorts of mice for each line, with wild-type (wt) and mutant littermates, were subjected to a battery of behavioural tests (see below) for 2 weeks. Then, 14-week-old mice were randomly divided into 2 groups: one treated with fasudil administrated orally ad libitum in drinking water to reach a dose of 100mg/kg/day and a second with no treatment. The dose was estimated at 152.7 mg in a feeding bottle (250 ml) changed twice a week, taking into account the previous study in which we verified that the drinking volume was about 4.6ml/day/mouse [25]. Four weeks after the beginning of the treatment, 18-week-old mice were challenged once again with the same battery of behavioural tests (see below) and kept under the same treatment condition (Figure 1). The experiments were conducted blindly for genotype as recommended by the ARRIVE guidelines [28, 29]. All animals injured by their cage companions were excluded from the behavioural tests at the time when they were seen. A second batch of three independent cohorts were processed similarly but without behaviour tests for the molecular analysis of the hippocampal region from treated and non-treated mice. In this case, treated animals started the fasudil treatment at the age of 12 weeks, for 6 weeks. Samples were quickly harvested from 18-week-old mice after euthanasia by cervical dislocation and snap frozen for molecular analyses.
Behavioural analysis
We used three tests that previously unravelled robust phenotypes in the three 16p11.2 mouse models [18-20]: the open field for the exploration activity, novel object location and novel object recognition for learning and memory in mice. The novel object location (NOL) memory task stimulates the parahippocampal cortex, the entorhinal cortex and the hippocampus [26] whereas the novel object recognition (NOR) memory task is based on the innate preference of rodents to explore novelty involving the perirhinal and entorhinal cortex and the hippocampus.
For the open field (OF), mice were tested in an automated arena (44.3 x 44.3 x 16.8 cm) made of PVC with transparent walls and a black floor, and covered with translucent PVC (Panlab, Barcelona, Spain). The arena was divided into central and peripheral regions (8 cm peripheral zone and 28 cm central zone) and homogeneously illuminated at 150 Lux. Each mouse was placed on the periphery of the open field and allowed to explore the apparatus freely for 30 min while 16x16 infrared captors (located at the periphery at two different levels) recorded the horizontal and vertical position of the mouse. During each session we measured the total distance travelled, evaluated the habituation of the animal over time, by splitting the data in 10-minute intervals, and assessed vertical activity through the number of rears.
The NOL memory task was carried out in the same open field arena as previously described. On the first day, mice were habituated to the arena for 30 min at 150 Lux. On the following day, animals went through an acquisition trial during the first 10 min in which they were presented individually to 2 identical objects A. Each object was placed 10 cm away from each of the corners on the north side of the box. The exploration time of objects A (when the animal’s snout was directed towards the object at a distance ≤1 cm) was manually recorded with a chronometer. Minimum exploration time was set to 3 s, and mice that did not reach this criterion or show any interest for one object were excluded from the study. A 10-min retention trial (second trial) was conducted 5 min later, when one of the familiar objects was displaced to a novel location (B) on the south side and the exploration time (t) of the two objects was recorded for 10 min. The selection of the new object (right or left) was counter balanced across animals. We used two identical cylindrical objects of black colour with a white circle on top. In this session, the minimum exploration time was also set to 3 s, and mice that did not reach this criterion or show any interest (0 s of exploration) in one object were excluded from the study. We verified that no preference was seen during the exploration of the left and right object. The recognition index (RI) was defined as (tB / (tA + tB) × 100). An RI of 50% corresponds to the chance level and a significantly higher RI reflects good recognition of which object was moved in between the two sessions.
The NOR test was performed in a circular open field of white PVC with opaque walls and floor 30 cm high and 50 cm diameter. On the first and second days, each mouse was habituated to the arena for 15 minutes at 60 Lux. The following day, we started the NOR sessions. First, each animal was individually subjected to a 10 minute acquisition trial for the presentation of two identical objects A (either marble or dice) placed at the northeast or northwest of the open field arena. The exploration time of both objects A was recorded. 3 hours later (retention delay in home cages), a 10 minute retention trial (second trial) was performed. One of the identical objects A was replaced with a novel object B at the same position. The new object location was randomly determined for each animal and overall balance assessed. The exploration time of the two objects (familiar object and novel object) was recorded using the NOLDUS video tracking software. The recognition index (RI) was defined as (tB / (tA + tB) × 100). An RI of 50% corresponds to a chance level and a significantly higher RI reflects good recognition memory. Any mice that did not explore the objects for more than 3 seconds during the acquisition trial or the retention trial or show any interest for one object were excluded from the analysis.
Western blot
Fresh hippocampal tissues were isolated by rapid decapitation/dissection of naive mice and snap frozen. Then, they were lysed in ice-cold sonication buffer supplemented with Complete™ Protease Inhibitor Cocktail (Roche). Individual samples were disaggregated, centrifuged at 4°C for 30 minutes at 14000 rpm, diluted in 4X Laemmli sample buffer containing β-mercaptoethanol (Bio-Rad, France), and incubated at 95 °C for 5 min. Protein concentration was determined by the PierceTM BCA Protein Assay Kit (23225, Thermo Fisher Scientific, Strasbourg). The samples were diluted with sample buffer such that 30 µg of protein were loaded per lane onto 15% polyacrylamide gel. Gels were run and then transferred to nitrocellulose membranes by the Trans-Blot® Turbo™ Transfer System (BioRad, France) using the MIXED MW Bio-Rad Preprogrammed Protocol. Then they were blocked in 5% BSA, 1 X Tris-buffered saline, 0.1% Tween 20 (TBS-T) and incubated with primary antibody for 10 minutes. Membranes were washed in TBS-T followed by a 10-minute secondary antibody incubation using an HRP conjugated Goat anti-Rabbit IgG (A16096, Invitrogen, France) at 2:10,000 using the SNAP i.d.® 2.0 Protein Detection System (C73105, Merck). This apparatus has a vacuum-driven technology and a built-in flow distributor that actively drives reagents through the membrane.
Total levels of KCTD13, RHOA protein and Myosin Light Chain and its phosphorylation product were analysed using Western Blot. Proteins were visualized with Amersham™ Imager 600. Signals were quantified using ImageJ and analysed using Microsoft Excel and GraphPad Prism. We used the following primary antibodies: KCDT13 (dilution 1:250, HPA043524, Atlas Antibodies, Bromma, Sweden), RHOA (dilution 1:1,000; cat #2117, Cell Signaling, USA), MLC (dilution 1:1,000;cat # 8505, Cell Signaling Technology Europe, B.V., Leiden, The Netherlands) and pMLC (dilution 1:1,000; Thr18/Ser19 cat #3674, Cell signalling, Boston, MA, USA). The ratio of protein (or phosphorylated protein) level was estimated against control β-actin protein level (detected with a mouse monoclonal Anti-β-Actin−Peroxidase antibody (A3854 Sigma)) and normalized to untreated wt sample mean.
Statistical analysis
The statistical analysis was carried out using standard statistical procedures available on the SigmaPlot software (Systat software, San Jose, USA). All outliers were identified using the Grubbs' test from the GraphPad calculator (GraphPad Software, San Diego) or ROUT method with a Q value of 1% from GraphPad Prism 7.01 (GraphPad Software, San Diego) for data with nonlinear regression. We estimated a priori the power of the different variables, scored during the behavioural analysis, to test our H0 hypothesis, taking into account our previous analysis of the Del/+ model [20]. A posteriori we verified that the sample size was sufficient to have a statistical power of 100% for the object location and recognition memory at both ages (Sup. Table 1). Nevertheless the powers of the open field variable “distance travelled” were lower at 12 compared to 18 weeks, certainly due to a change in variability. Based on the power calculation we did not consider the results of the rearing activity from the open field. Data from the behavioural characterization of Kctd13+/- and 16p11.2 Del/+ mouse models were analysed by the Student t-test (see sup. tables 2 and table 3 for a summary). One sample t-test was used to compare recognition index values to the set chance level (50%). Data from post-treatment behavioural phenotyping of both genetic lines were analysed using one-way or two-way ANOVA followed by Tukey’s post-hoc test whenever data presented normal distribution and equal variance. Otherwise, we used the non-parametric Kruskal-Wallis one-way analysis of variance and the Mann-Whitney U test. One sample t-test was also used to compare recognition index values to the set chance level (50%). Western blot data were analysed using a Kruskal-Wallis one-way analysis of variance test between groups followed by a Mann-Whitney U test or Student t-test depending on data distribution. Data are represented as the mean ± SEM and the significant threshold was p < 0.05.