All the procedures were approved by the Animal Experiment Committee of the University of Toyama (Authorization No. A2023MED-05 and A2016OPR-3) and conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals of the University of Toyama. Nlgn3hse line and Nlgn3mf line mice with a pure C57BL/6N genetic background had been reported previously [15]. Mice were housed in a room under a 12 h light/dark cycle (lights on at 7:00 a.m.) at 23 ± 3 °C and 30–60% humidity. Food and water were provided ad libitum. Wild-type (WT) and mutant mice were generated by crossing heterozygous female mice with WT male mice. All behavioral analyses were performed on male mice only. For general behavioral testing, the male offspring of mating pairs were weaned around one month, genotyped, and housed 4 (two pairs of WT and mutant mice) per cage. For the five-trial social novelty test, male offspring mice were weaned at postnatal day (P) 21-P23 and randomly assigned 3–5 animals per cage.

Behavioral test battery was carried out with male mice (19 Nlgn3hse mutant mice and 18 littermate WT mice, and 19 Nlgn3mf mutant mice and 17 littermate WT mice) and started at 9–11 weeks of age. The behavioral test battery included general health and neurological screening, light/dark transition test, open field test, elevated plus maze test, hot plate test, rotarod test, startle response/prepulse inhibition, and Porsolt forced swimming test. All the behavioral testing was performed between 8:30 a.m. and 18:30 p.m. Prior to all experiments, mice were left undisturbed in the testing room for at least 30 min to allow acclimation. After each trial of experiment the apparatus was thoroughly cleaned with hypochlorous water to eliminate any scent to prevent giving a bias as olfactory cue to a next subject. All the behavioral testing except the light/dark test, was performed with illumination level 100 lx. The detailed procedures for each behavioral testing are as follows. After the behavioral test battery, same batches of mice were subjected to Barnes maze test and contextual and cued fear conditioning with 0.3 mA footshock paradigm (please see below).

General health and neurological screening: Health status including body weight, rectal temperature, and neuromuscular strength was examined at the first day of the series of the behavioral testing. Neuromuscular strength was examined by the grip strength test and wire hang test as described [16]. A grip strength meter (O’Hara & Co., Tokyo, Japan) was used to assess forelimb grip strength. Each mouse was tested three times and the greatest value measured was used for statistical analysis. A box (215 × 22 × 23 cm) with a wire mesh grid (10 × 10 cm) on its top (O’Hara & Co., Tokyo, Japan) was used for the wire hang test. Latency to fall was recorded with a 60 s cutoff time.

Light/dark transition test: Light/dark transition test was conducted as previously reported [17]. Mice were placed into the dark chamber and allowed to move freely in the light (380 ± 20 lx) and dark chamber through the opening in between for 10 min. The total number of transitions between chambers, time spent in each side, latency to the first transmission to the light chamber, and distance traveled in each chamber were recorded.

Open field test: Locomotor activity was measured in an open field apparatus (40 × 40 × 30 cm; Accuscan Instruments, Columbus, OH, USA) as described [18]. Mice were placed into left corner of the apparatus and allowed to move freely for 120 min. Total distance traveled, vertical activity (rearing measured by counting the number of photobeam interruptions), time spent in the center (20 × 20 cm) of the open field area, and the stereotypic counts were recorded using VersaMax system (Accuscan Instruments, Columbus, OH, USA).

Elevated pulse maze test: The elevated plus maze test was conducted as described [19]. The elevated plus maze (O’Hara & Co., Tokyo, Japan) consisted of four arms (25 × 5 cm) arranged in plus shape with a 5 × 5 cm square in the center. Two of the arms were closed with 15 cm high walls and other two were open without walls. The closed and open arms were alternately arranged in the plus maze and the maze was elevated to a height of 55 cm above from floor. Each mouse was placed on the center of the maze facing one of the closed arms. Mouse behavior was recorded during a 10 min test period. The time spent in the open and closed arm, the number of entries into the arms, and the distance traveled were recorded.

Hot plate test: The hot plate test was conducted as described [20]. Mice were placed on a 55.0 ± 0.3 °C hot plate (Columbus Instruments, Columbus, OH, USA), and latency to the first front-paw response (rubbing the paws) was recorded.

Porsolt forced swim test: The subject was placed in a cylindrical container (12 cm in diameter) of water (12 cm in depth and 23–24 °C in temperature) in the illuminated chamber (800–850 lx) and allowed to freely move for 10 min. Total distance traveled and percentage of immobility were measured. The same test was repeated on the following day.

Acoustic startle response and prepulse inhibition tests: The acoustic startle responses and prepulse inhibition tests were conducted using startle reflex measurement system (O’Hara & Co., Tokyo, Japan) as described [18]. The subject mouse was put in the test equipment, a Plexiglas cylinder and the cylinder was placed in the test apparatus with background noise level at 70 dB for 10 min for habituation. A test session consisted of 6 trial types: two types for startle stimulus only trials, and four types for prepulse inhibition trials. The duration of white noise that was used as the startle stimulus was 40 ms for all trial types. The startle response was recorded for 140 ms (measuring the response every 1 ms) starting with the onset of the prepulse stimulus. The peak startle amplitude recorded during the 140 ms sampling window was used as the dependent variable. The intensity of startle stimulus was 110 or 120 dB. The prepulse sound was presented 100 ms before the startle stimulus, and its intensity was 74 or 78 dB. Four combinations of prepulse and startle stimuli were employed (74–110. 78–110, 74–120, and 78–120). Six blocks of the 6 trial types were presented in pseudorandom order such that each trial type was presented once within a block. The average intertrial interval was 15 s (range: 10–20 s). The startle amplitude and percentage of prepulse inhibition was measured.

Five-trial social novelty test was carried out with 26 Nlgn3hse mutant mice and 22 littermate WT mice, and 26 Nlgn3mf mutant mice and 17 littermate WT mice, essentially according to the previous report [12]. Subject mice (P26-P32) and younger social stimulus mice (WT, P21-P28) were employed in this study. The stimulus mice were marked with a yellow spot on the neck by bleaching under anesthesia to discriminate from the subject mice under the video recording. The experimental arena was a transparent acrylic resin box (30 × 15 × 30 cm), filled with 450 ml of homecage bedding (Paper-clean, Japan SLC, Japan) and illuminated at 100 lx. Subject mice were gently introduced into the box and allowed to acclimate for 15 min. For the first trial, a stimulus mouse (referred to as “Stranger-1”) was introduced into the cage, permitting free interaction between the mice for 4 min. This procedure was replicated for four consecutive trials with 5-min intervals in between, allowing the subject mice to be acquainted with the stimulus mouse. During the fifth trial, a novel mouse (Stranger-2) was introduced instead of the “Stranger-1”. Every subject mouse was temporarily placed in a numbered cage after the test. The subject mice were ear-punched following the completion of all tests. The punched-out samples were stored at − 20 °C and genotyped after manual video analysis. Social interaction behavior was manually scored from TIFF image stacks (4 frame/s) before the genotyping. The social interaction duration included the time for nose-to-nose sniffing, anogenital sniffing, allogrooming, other forms of non-aggressive physical contact, and following the stimulus mouse.

After ear-punching and a week’s rest, mutant and littermate WT mice at P33-P40 (23 Nlgn3hse mutant mice and 26 littermate WT mice, and 24 Nlgn3mf mutant mice and 15 littermate WT mice) were subjected to a sCPP test [21]. The apparatus is a 30 × 30 cm clear acrylic resin box, equally divided into two chambers by a central sliding partition having a circular opening, illuminated at 100 lx. Each chamber was filled with 1 cm of different novel bedding (Alpha-dri, Shepherd Specialty Papers, Watertown, TN, USA; Kaytee Soft Granules [discontinued], Petco, Irvine, CA, USA; or Care-feeaz, Hamri Co., Ibaraki-ken, Japan). For the pre-test, mice were gently placed in the apparatus for 30 min of free exploration. Following the pre-test, mice were group-housed in a home cage with “bedding A” for social conditioning over 24 h. Subsequently, the mice were separated and individually housed on "bedding B" for isolated conditioning, for another 24 h. On the third day, the apparatus was set up with both conditioned beddings, and the mice were allowed free exploration for 30 min. The left and right positions of two types of bedding were alternated for each mouse during the tests; however, all mice were consistently introduced from the right side. For each individual, the bedding positions remained the same from the pre-test to the post-test. The bedding used for social and isolated contexts was alternated between each group-housed cage, to ensure the counterbalance. The duration that the subjects spent on the “social context” was recorded and analyzed using an ImageJ plug-in [22].

The Barnes maze test was conducted with mice used for behavioral test battery on a white circular platform, 1.0 m in diameter, with 12 holes equally placed around the perimeter (O’Hara & Co., Tokyo, Japan) elevated 75 cm from the floor as previously described [20]. The platform was illuminated with 1200 lx lighting and one of the 12 holes leads to a black Plexiglas escape box (17 × 13 × 7 cm). The particular hole which leads to the escape box was assigned to each subject as a target. Location of the target hole were evenly assigned to WT and mutant mice. Prior to the test, the subject had habituation trial to become familiar with the maze and the escape box. Each trial test started with emergence of the subject by retracting the wall around the subject at the center of the platform and ended when the subject entered the escape box through their assigned target hole or 5 min elapsed. The latency and distance for the subject to reach the target hole and the number of times that the subject visited incorrect holes were measured in all the tests. Two to three trials per day were conducted for 3 consecutive days. On day 4, the mice received probe trial test that conducted without the escape box for 3 min. During the probe trial test, staying time around the target was measured to confirm that this spatial task was acquired based on navigation by distal environment room cues. Mice were left undisturbed until receiving next probe trials. On day 28, the mice once again received a probe trial test to check remote memory. The dwell time spent wandering around each hole was recorded using Image BM software.

Radial maze test was carried out with 14 Nlgn3hse mutant mice and 14 littermate WT mice, and 20 Nlgn3mf mutant mice and 20 littermate WT mice. The automated eight-arm radial maze apparatus (O’Hara & Co., Tokyo, Japan) was used as previously described [23]. The maze consisted of a regular octagonal platform connecting eight narrow arms (9 × 40 cm) made of white plastic and the walls (25 cm high) made of transparent plastic and illuminated at 100 lx. Program-controlled automatic gates were set between the platform and the arms. At the distal end of each arm, a food pellet well (1.4 cm deep and 1.4 cm in diameter) was set, equipped with a photodetector to automatically record pellet intake. Four distinct cues were hung at the corners of the ceiling. The position of the maze and the direction of the arms remained consistent throughout the experiment. One week before the training phase, the mice were subjected to a food restriction (8 g/day for 4 mice) and were weighed daily to reduce their weight to 80–85% of their initial weight, ensuring they remained hungry without compromising health. From the eighth day, mice explored the maze and consumed scattered pellets for 30 min as a pre-training. Subsequently, they underwent advanced pre-training to take a food pellet from each food well, repeating this process eight times until all arms had been traversed. Once these pre-training trials were completed, the actual maze acquisition trials began. In the spatial working memory task, each of the eight arms contained a food pellet. Initially, the mouse was positioned on the central platform with all gates closed. Each trial began with all gates being simultaneously opened, allowing the subject to explore and consume the food pellets. An "entry" was defined as the entering beyond 5 cm from the platform into a specific arm, resulting in the closure of the others seven gates. Once the mouse returned to the central platform, the remaining gate closed, confining the mouse to the central platform for a 5-s interval. Thereafter, all gates were simultaneously reopened, allowing the mouse to make its next choice. The trial ended when the subject had consumed all the food pellets or when 25 min had elapsed. The mice underwent one trial daily. Two weeks later, once they reached a plateau in their food collection speed and accuracy, the difficulty level increased. The first interval following the consumption of the fourth food pellet was termed “delay-after-4th”. The delay-after-4th was sequentially set at 30 s, 120 s, and 300 s, with each duration lasting for two days. All other intervals were kept at 5 s. Metrics including arm choice, time to collect pellets, travel distance, number of different arms chosen within the first eight choices, revisits, and omission errors were automatically recorded.

Contextual fear conditioning using 0.5 mA footshock and fear memory extinction test were conducted with 10 Nlgn3hse mutant mice and 9 littermate WT mice, and 14 Nlgn3mf mutant mice and 17 littermate WT mice as previously described[24, 25]. Mice were handled for 1 min per day for 1 week, prior to the experiment. Contextual fear conditioning was conducted in a small conditioning chamber made of transparent plastic surrounded by a sound-attenuating chest (O’Hara & Co., Tokyo, Japan). Mice were placed in the conditioning chamber and two footshocks (0.5 mA, 2 s) were delivered at 58 and 118 s after entry to the chamber. Twenty-four hours later, mice were re-exposed to the conditioned chamber for 30 min without receiving a footshock again for extinction training. The initial 5 min of the extinction session was also referred to as "Test 1", to assess the contextual fear memory. After another twenty-four hours, the mice underwent a 5-min "Test 2" without footshock to evaluate the outcome of the extinction training. Percentage of freezing time was measured using ImageFZ software with image capture rate of 1 frame/s.

Contextual and cued fear conditioning using 0.3 mA footshock were conducted with mice used for behavioral test battery as described previously [24]. The tests consisted of three sessions, conditioning (Day 1), and contextual and cued tests (Day 2 and 30). In the conditioning session, a 55 dB white noise, which served as the conditioned stimulus (CS), was played for 30 s from 120, 240, and 360 s. During the last 2 s of the tone, a footshock of 0.3 mA was delivered as the unconditioned stimulus (US). Each mouse received three CS-US pairings with 2 min interstimulus interval. Contextual testing was conducted 24 h after conditioning (Day 2). The mice were placed in the same chamber that contextual conditioning was taken place and monitored for freezing for 5 min. Cued testing with altered context was conducted 3 h after contextual test in a triangular box (35 × 35 × 40 cm) made of white opaque Plexiglas with illumination level 30 lx. Freezing behavior was assessed during a 3 min free exploration, followed by a 3 min presentation of the tone. Contextual and cued tests were conducted again at Day 30 to assess remote memory. Percentage of freezing time was measured using ImageFZ software with image capture rate of 2 frame/s.