Mice

C57BL/6J and BALB/cByJ male mice were obtained from the National Laboratory Animal Center in Taiwan. C57BL/6N mice were purchased from BioLASCO Taiwan company, Ltd. With the exception of C57BL/6N mice, which were used for stereotactic surgery, C57BL/6J was used as residents for all experiments. Mice used in social hierarchy experiments and distressed social partners were housed in pairs, whereas mice in all other experiments were individually housed. While all C57BL/6 residents were used for only one experiment, BALB/cByJ mice, housed in groups of four, were repeatedly used as intruders from 8 weeks old until they reached one year of age. All animal procedures followed institutional guidelines established and approved by the Institutional Animal Care and Use Committee of National Tsing Hua University. The strains, mouse numbers and manipulations for each experiment were listed in Table S1.

Ablation of the main olfactory epithelium (MOE)

2,6-dichlorobenzonitrile (dichlobenil) has been widely used to damage olfactory epithelium [29, 30]. Mice received intraperitoneal injections of dichlobenil (Sigma, D57558) at a concentration of 50 mg/mL, dissolved in dimethyl sulfoxide (DMSO) (Sigma-Aldrich, V900090), at a dosage of 100 µg/g body weight. The injections were administrated on days 1, 3 and 5 prior to the experimentation [31]. Control animals were injected with DMSO only.

Mouse behavioral assaysEstablishment and identification of social rank (Fig. S1)

Eight-week-old male mice were housed in pairs for 1 week to form a social hierarchy (Hierarchy Establishment). We modified the standard resident-intruder assay to identify the social ranks of two resident mice (Hierarchy Identification). A BALB/cByJ intruder mouse with ablated MOE was introduced into the cage for a 10-minute interaction. The resident that carried out more than one attack on the intruder was identified as the dominant male, whereas the resident that showed no aggression was identified as the subordinate. Pairs of mice that both showed aggression were not used. Pairs of mice in which neither showed aggression were tested again the following day; if both males continued not showing aggression, they were also excluded. Identified dominant and subordinate males were used separately for the standard resident-intruder assay (Behavioral Testing).

Tube test behavioral assay

The tube test was performed based on a previous study [32]. Mice were habituated to the procedure room for 1 h on two consecutive days and were tested on the third day. A tube test trial was carried out involving two mice that were simultaneously released at opposite ends of a clear Plexiglas tube (3.75 cm diameter, 60 cm length) and then ran toward the middle. When a mouse retreated and placed all four paws outside the tube, the trial was considered over, and that mouse was classified as the loser. Four consecutive trials were conducted for each pair. A mouse was defined as dominant if it won 3 (75%) or 4 (100%) out of 4 trials, while a subordinate mouse was defined by winning 0 or 25% of the trials. Pairs without a winner (50%) were removed. The interior of the tube was cleaned with 70% ethanol after each pairwise trial.

Standard resident-intruder assay

The resident-intruder assay was carried out as previously described [12]. In brief, with the exception of pair-housed males for social hierarchy or distressed social partners, all resident males were individually housed in standard rack cages for a minimum of 7 days. Prior to the assay, mice were moved to the procedure room for a period of 1-hour on two consecutive days to acclimate the environment. On the third day, following a 1-hour habituation period in procedure room, an intruder was introduced into the home cage of a resident mouse. Their interaction was digitally recorded for 10 min. Both the habituation and the assay were conducted under red light conditions to minimize visual cues. All intruders used in experiments were adult BALB/cByJ treated with dichlobenil to ablate the MOE and eliminate aggression.

The recorded videos were subsequently analyzed manually with the assistance of software BORIS to obtain the total time of aggression (tail rattling, attack biting, wrestling and chasing) [33], allogrooming (mouthing, licking, nibbling but not barbering) and social investigation (encompassing all social interactions excluding aggression, allogrooming and mounting) exhibited by each resident. Parameters such as latency (time until the first occurrence) and number of bouts (instances of continuous action) for each behavior were also calculated. For mice that did not engage in a specific behavior within the 10-minute assay period, the latency was denoted as 600 s. In addition, to capture the recipients’ responses during interaction, we recorded their stationary behaviors (defined as no movement for more than 3 s) and categorized them as either freezing-like (upright posture, hunched back, or motionless immediately after fleeing) or non-freezing-like behaviors [34,35,36].

Assays with intruders covered with foreign materials

Mineral oil (100 µl) or glue from a glue stick was spread onto one side of an intruder with MOE ablation before its use in a resident-intruder assay, which was carried out as described above.

Assays for evaluation of cleaning efficiency

Red poster paint (Pentel, Scarlet Lake No. 37) was carefully applied to a designated area on one side of an intruder approximately 2.4 × 1.1 cm2 in size. The intruder was anesthetized by intraperitoneal injection of 20 mg/kg Zoletil 50 (Virbac) to prevent self-grooming which could interfere with the paint marking. Following the application of the paint, a drying period of 10 min was allowed. Subsequently, the intruder, with dry paint, was introduced into a cage either with or without residents for a duration of 10 min. After the completion of the assay, a photograph was taken, and the intensity of the red poster paint color was measured using ImageJ software.

Two-choice assay

Two stimuli, mineral oil (100 µl, Sigma, M5904) or glue from a glue stick (Simbalion, GS104), and 100 µl 1× PBS (as a control) were individually applied to 1 × 1−inch squares of odorless blotting paper (Fisher Scientific 05-714-4). The papers were affixed to opposite walls of an arena approximately 7.5 cm from the bedding, using double-sided adhesive tape. Following a 1-hour habituation period in the procedure room, a mouse was introduced into the arena and given 10 min to explore the papers. The behaviors of mice during exploration were digitally recorded under red light conditions. The recording was subsequently used to score the investigation time of each stimulus. Investigation time was defined as the time during which the nose of the mouse was within 0.5 cm of the blotting paper.

Distressed social partners

The mice were housed in pairs for a minimum of 1 week prior to the experiment. Two days before the assay, the mice underwent a 1-hour habituation period each day in the procedure room. On the day of the test, the mice would once again be moved into the procedure room for 1-hour habituation period prior to the experimental trials. For distressed partners, 30 min prior to the experimental trials, mice would be taken out of their home cage and put into a 50mL centrifuge tube with breath holes and positioned stilly outside their home cage [26]. After the 30-minute restraint period, the mice would be returned to their home cage for behavioral observation. In the control (Unstressed) group, 30 min before the experiment trial commenced, the mice would be simply taken out of their home cage and placed into a neutral clean cage before being transferred back to their home cage for the behavior trials. Following the return of the mice to their home cage, the interaction between the testing mice (without manipulation) and their distressed/unstressed partners would be recorded for 10 min and subsequently analyzed manually with the assistance of software BORIS [33].

Excitotoxicity lesion

To induce a neural lesion, ibotenic acid (IBO) at a concentration of 10 mg/mL in 10xPBS was bilaterally injected into either the MeApd (ML ± 2.10, AP-1.50, DV-5.40) or VMHvl (ML ± 0.70, AP-1.60, DV-5.55) in a volume of 200nL [37]. For the control group, 200nL of 10x PBS was injected at the same target site. Following the injection, the mice were given a minimum of 7 days to recover before behavioral testing. The efficacy of the lesion was confirmed through immunohistochemical (IHC) staining.

Stereotaxic surgical procedures

Because the stereotaxic surgery (injection of PBS only) itself had a significant impact on C57BL/6J mouse behaviors, including reducing aggression and increasing allogrooming behaviors (Fig. S2A-D), in order to compare the behaviors of mice with and without lesion, we had to switch to C57BL/6N mice, whose behaviors were not affected significantly by the surgery (Fig. S2E-H). At 8 weeks of age, mice were anesthetized with isoflurane (5% induction, 1–1.5% maintenance) and placed in a stereotaxic platform. Ophthalmic ointment (Puralube) was applied at the beginning to protect the eyes. The head skin was sterilized with 75% ethanol, followed by careful removal using scissors. Two 1 mm micro-injection holes were drilled into the skull, and the micro-injection needle was inserted into the desired location. Once the injection was completed, the needle was withdrawn, and the holes were sealed with bone wax. To ensure stability, the head was sealed with Tempron (GC company, Japan). At the end of the procedure, carprofen (50 mg/mL, RIMADYL®) at a dose of 4.6 mg/kg and ampicillin at dose of 20 mg/kg by body weight were administrated subcutaneously to alleviate discomfort and prevent infection. Following the surgery, the mice were individually housed and carefully monitored to ensure proper recovery before further behavioral testing.

Immunochemistry

The Brains were collected and pre-fixed with 4% paraformaldehyde in PBS for 24 h. After pre-fixation, the brains were cut into 50 μm thick sections using a vibratome. The brain slices were then blocked in a solution containing 3% BSA (Sigma-Aldrich, A3059) in 0.3% PBST, PBS plus Triton X-100 (Sigma-Aldrich, V900502), for an hour. Subsequently, the sections were incubated with the primary antibody Anti-NeuN Antibody (Sigma-Aldrich, MAB377) at a dilution of 1:500 for neuron cell staining in the blocking solution for 16 h at 4℃. After 3X wash in 0.1% PBST for 30 min at room temperature, the sections were incubated with the secondary antibody Goat Anti-Mouse IgG H&L (Alexa Fluor® 594, Abcam, ab150120) at a dilution of 1:1000 in the blocking solution for 2.5 h at room temperature. After 2X wash in 0.1% PBST and 1X wash in PBS each for 10 min at room temperature, sections were incubated with Hoechst 33342, Trihydrochloride, Trihydrate (Invitrogen™, H3570) at a dilution of 1:10000 for 10 min at room temperature. Finally, the sections were washed in 3X in PBS for 10 min each, mounted on glass slides, and sealed with ProLong™ Diamond Antifade Mountant (Invitrogen™, P36970) for imaging. All images were captured by NIS-Element AR 5.21.03 with Nikon ECLIPSE Ni microscope and Cool SNAP HQ2 camera from Photometrics®. Regions of interest were circled manually and based on Mouse Brain Atlas as reference [38]. Cell numbers were counted using Image J software following the same protocol and criteria [39].

Children participants

Approved by the National Tsing Hua University Research Ethics Committee (REC No.11206HT093), we sent research recruitment letters to 6 preschools in the Hsinchu area in Taiwan, soliciting newly enrolled children to participate in this study. A total of 118 parents’ informed consent was obtained, including 67 boys and 51 girls, with an average age of 52.31months (47 ~ 57 months).

Procedure

We collected one year of longitudinal data for this study. In the 4th week after the preschools started, we asked the preschool teachers to fill in the three scales after they carefully observed the daily behaviors of these newly admitted young children. After one school year (12 months later), the class teachers were asked to fill in the same scales again according to the children’s daily behaviors.

Children behavioral assessmentsAggressive behavior

The Aggressive Behavior Subscale of the Preschool Social Behavior Scale [40], revised from the Preschool Social Behavior Scale—Teacher Form [41], was used in this study. The scale includes 6 items in Overt Aggression Subscale (e.g. The child will kick or hit others) and 5 items in Relational Aggressive Behavior Subscale (e.g. The child will ask other children not to play with certain child). The applicable age of this scale is 3.5 ~ 5.5 years old. The scale was filled out by the class teachers based on their daily observation of the social interaction between the target child and their peers. The scale is a continuous 5-point Likert scale, ranging from “never like this”, “rarely like this”, “sometimes”, “usually” to “always”, giving 1 ~ 5 points. The total score was converted to the z score for statistical analysis. The higher the sum score of the items is, the more overt and relational aggression the children have.

Prosocial behavior

Preschool Prosocial Behavior Scale was used in this study [42]. This scale has 25 items and describes the four dimensions of children’s prosocial behaviors, including sharing (e.g. S/He will give her/his favorite items, such as stickers, picture cards, etc., to her/his friends.), helping (e.g. S/He will proactively provide assistance to children who have difficulties in work or activities.), caring (e.g. S/He comforts other children when they cry or get hurt.) and cooperation (e.g. S/He can work with other children to get things done.). Class teachers were asked to fill in the scale based on observations of children’s daily behavior. This scale adopts a 5-point Likert scale, ranging from 1 to 5 points from “never”, “rarely”, “sometimes”, “usually”, and “always”. The total score was converted to the z score for statistical analysis. The higher the total score is for each dimension, the greater the performance of prosocial social behavior.

Executive function

The Taiwanese Traditional-Chinese Childhood Executive Functioning Inventory was used to assess the self-inhibitory control of young children [43]. This scale has a national norm for ages from 4 to 12 years old. This scale is a five-point scale, from 1 to 5 points, divided into “completely incorrect”, “mostly incorrect”, “partially correct”, “correct”, and “completely correct”. Teachers were asked to choose appropriate behavior descriptions after observing children’s daily performance (e.g. Even if he is ordered to stop, it is still difficult for him to stop immediately during activities. For example, he always jumps a few more times or plays on the computer for a while after being told to stop.). Because this scale is presented with reverse questions, the total score was reversed and converted to the z score for the statistical analysis. The higher the total score is, the greater the ability of inhibitory self-control for the children.

Statistical analyses

All statistics were carried out using GraphPad Prism 8.3 and SPSS 22.0 software. Shapiro-Wilk normality test was used to analyze the distribution of the data. For normally distributed data, Unpaired t-test (two-tailed) was used to compare two independent groups. Paired t-test (two-tailed) was applied to compare two dependent groups. Pearson correlation (with two-tailed significance testing) was applied to analyze correlation between poster paint intensity and allogrooming. For data not normally distributed, Mann‒Whitney test (two-tailed) was used for two independent data sets. Wilcoxon signed-rank test (two-tailed) was used for two dependent groups. Non-linear Spearman regression was used for correlation analyses. The relationship between social ranks identified through the intruder assay and the tube test was examined using Fisher’s Exact Test. Controlling inhibitory self-control to analyze the effect of aggression on prosocial behavior was done by hierarchical regression analysis. Data are shown as the mean ± standard error of the mean (SEM), along with individual data points.