Using the CRISPR-Cas9 system, Septin-14 floxed mice were generated from the Transgenic Mouse Models Core of National Taiwan University, while the Sox2-Cre mice from the Jackson Laboratory (Hayashi et al., 2002; Chen et al., 2019). Mice with germline deletion of Septin-14 intron 3–5 [i.e., SEPT14 knockout (KO) mice] were obtained by crossing the Septin-14 floxed and Sox2-Cre mice (Supplementary Figure 1). Since the primers, namely, 5VF1: 5′-TTCCAA TTGATTTTGTCTGTGTCA-3′, 5VR1: 5′-AAAATGCCATGATAAACC CCAGG-3′, 3VR1: 5′-CGGTAGGAGACATGATAGCCAAG-3′, were used to identify the WT and SEPT14 KO alleles, the base pairs of recognized WT and KO were 746 and 579, respectively. This study was performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals revised in 2011. All procedures were approved by the Local Animal Care Committee at National Cheng Kung University College of Medicine (NCKUCM No. 107172). After their weaning (Day 21 postpartum), male and female WT and SEPT14 KO mice were group-housed in plastic cages (2–5 per cage) in a temperature- and humidity-controlled colony room on a 12-h light/dark cycle with lights on at 07:00. Mice had access to food (Purina Mouse Chow, Richmond, IN, United States) and tap water ad libitum throughout the experiments. Since object location and sociability and social novelty preference tasks were minor in their stress-provoking nature, a total of 45 (20 WT and 23 SEPT14 KO) mice received object location, followed by sociability and social novelty preference task, at a 2-h interval at the same day. These mice, then, received inhibitory training and testing for the following 2 days. Nonetheless, it was of importance to note that WT and KO mice were not repeatedly used for the remaining behavioral tasks or bioassays.

Total RNA isolated from brain tissues was extracted using Direct-zol RNA MiniPrep kit (Zymo Research), and cDNA was synthesized using PrimeScript™ RT reagent Kit (TaKaRa). cDNA was amplified by PCR using specific primers (GAPDH sense 5′-AAGGTCATCCCAGAGCTGAA-3′; GAPDH antisense 5′-TGCCTGCTTCACCACCTTCT-3′; SEPT14 sense 5′-ATGGCGGAAAAACCAACTAACA-3′; SEPT14 antisense 5′-GGAGGTAGGCTTCAAACTGA-3′).

Sept14 BaseScope probe targeting 297–446 of NM_028826.2 was designed by Advanced Cell Diagnostics. Sept14 mRNA in the paraformaldehyde-fixed brain section was detected by BaseScope according to the manufacturer’s instructions (BaseScope™ Detection Reagent Kit-RED User Manual, Advanced Cell Diagnostics). The images were acquired using a VENTANA DP 200 slide scanner (Roche).

To assess mice’ social recognition and novelty, a custom-made three-compartment chamber (42 cm × 21 cm × 21 cm) was used (Liao et al., 2018). In brief, the test consisted of three sessions, including habituation, sociability, and social novelty preference sessions. In the 10-min habituation session, female and male WT and SEPT14 KO mice were individually subjected to the center compartment (12 cm × 21 cm × 21 cm) starting unrestrained navigation in the chamber [consisting of a center and two side compartments (15 cm × 21 cm × 21 cm)]. In the sociability session, an unfamiliar, same-sex C57BL/6 mouse (stranger 1) was placed in one randomly chosen side compartment enclosed in a wire cage (7 cm × 7 cm × 10 cm) that allowed for air exchange and likely olfactory, auditory, and visual interactions. An empty, but otherwise identical, wire cage was placed in the opposite side compartment. Mice were allowed to explore the chamber for social preference for a 10-min period. After the conclusion of the sociability session, the mice were immediately returned to the center compartment, and the doors to the two side compartments were blocked. With stranger 1 retained in its original side compartment, another unfamiliar, same-sex mouse (stranger 2) was placed in the previously empty wire cage in the opposite side compartment. Starting the social novelty preference session, both doors to the side compartments were unblocked and mice were allowed to explore the entire chamber for their novelty preference for another 10-min period. Mouse position and motion were tracked using a top-mounted charge-coupled device connected to a computer, which was recorded and analyzed using Logitech Webcam Software. The exploration time spent (in second) in close proximity (less than 1 cm) of the mouse nose directing to the empty, stranger 1 and 2’s wire cages were recorded. Likewise, individual mouse stranger 1 and stranger 2 preference ratios (i.e., time spent with stranger 1/total time spent with empty and stranger 1 and time spent with stranger 2/total time spent with stranger 1 and 2) were obtained.

The object location task consisted of 30-min habituation, 10-min training, and a 10-min test session (Tzeng et al., 2016). Regardless of sex and SEPT14 KO, mice first received the habituation session individually freely navigating in an empty Plexiglas box (46 cm × 26 cm × 21 cm) with black walls and a bright yellow floor in a dimly lit (<40 lux) laboratory for 30 min. Two teacups, being placed upside down, were then placed in the opposite corners sharing the same wall, with the cup border 6.8 cm away from the walls of the box in the same dimly lit laboratory for the 10-min training session. In this training session, mice were allowed to freely explore in the box and the time mice spent exploring two cups was recorded. Mice were, then, transported to holding cages for an approximate 5-min waiting period following the training session. Starting the test session, the mice were returned to the box with one cup moved to a diagonally opposite corner, with a cup border also 6.8 cm away from two adjacent walls. The recognition percentage, which was the ratio of the time spent exploring the cup in the new corner over the total time spent exploring both cups, was used to determine the object location memory. It was of importance to note that boxes and cups were thoroughly cleaned between adjacent sessions to stop the likely build-up of olfactory cues.

The stress-related navigation was gaged in a custom-made Plexiglas chamber with a white floor and dark walls (41 cm × 41 cm × 30 cm). A 13-W incandescent light was used 60 cm above the chamber to provide illumination at the bottom center (900 lux) throughout the test. Their time spent in the defined central zone (20 cm × 20 cm) and the remaining zone were recorded using the Noldus Ethovision XT video tracking system (Noldus Information Technology, Beijing, China). In addition, mice running speed and navigation distance were also obtained using such system-harboring software.

A custom-made chamber with two compartments was used, consisting of a footshock-delivering compartment and an adjoining insulated observational compartment (Liao et al., 2021b). Three same-sex C57BL/6 mice served as demonstrators, receiving a daily footshock protocol (pseudorandomly scheduled 15 footshocks in total, 0.5 mA in intensity, lasting 2 s each, delivered in 30 min) in the shock-delivering compartment (a 24-cm long, trough-shaped metal alley) and a cotton tip impregnated with vanilla extract was taped onto the compartment ceiling throughout the footshock protocol. Individual female and male WT and SEPT14 KO mice were the observer mice being subjected to the adjoining observational compartment (a 16-cm long, trough-shaped metal alley) throughout the demonstrators’ daily footshock protocol. The shock-delivering compartment was separated from its adjoining observational compartment by a transparent and perforated Plexiglas divider, allowing observer mice’ visual, auditory and olfactory cue reception (Liao et al., 2021b). The aforementioned daily footshock sessions were administered for 3 consecutive days. Approximately 24 h following the conclusion of the training session, freezing baseline and vanilla odor-induced freezing responses (FRs) were assessed in a novel test environment to reveal the magnitudes of novelty- and observational fear conditioning-induced FRs, respectively. A 10-min freezing baseline was first obtained in a transparent test chamber (28.5 cm × 17.5 cm × 11.5 cm) to reveal observers’ differences in their novelty-related FR. These observer mice were then individually subjected to a 30-min vanilla odor-induced FR test in another chamber (28.5 cm × 17.5 cm × 11.5 cm), wherein a cotton tip impregnated with vanilla extract was taped on the ceiling (Liao et al., 2021b). An FR was defined as a complete cessation of all movement except for respiration. FRs were sampled every 10 s throughout the 30-min videotaping and were analyzed by an experienced rater blind to the groups. Importantly, the mouse serum corticosterone (CORT) assay was done following the first observational fear conditioning as previously mentioned (Liao et al., 2021b).

The custom-made step-through IA apparatus consisted of a trough-shaped metal alley (bottom width = 1.7 cm) divided by a sliding door into an illuminated safe compartment (17 cm in length with a transparent lid) and a dark shock delivery compartment (24 cm in length with an opaque dark brown lid) (Tzeng et al., 2012). A shock generator (SMSCK, Kinder Scientific Com LLC, Poway, CA, United States) was connected to the alley floor of the dark compartment. In PA training, mice were individually placed at the far end of the illuminated compartment facing away from the door. As the mouse turned around and entered the dark compartment, the door was manually closed and a 0.5-mA footshock was given for 3 s after a 3-s delay. The mouse was left in the dark compartment for another 10 s, then removed from the apparatus and returned to the home cage. The latency of dark compartment entry in the training was used as a baseline for each mouse. Approximately 24 h later, mice were individually placed into the illuminated compartment, and the latency to step into (with all four paws) the dark compartment was recorded as the measure of retention time indexing IA memory magnitude.

Mice were suspended upside down by the tip of their tail on a custom-made horizontal rod approximately 45 cm away from the bench (Chuang et al., 2011). When mice were held in such an upside-down position, the immobile posture was used to index their giving-up responses in the face of inescapable stress. Accordingly, the duration of immobility (remained to be immobile for at least 1 s) in a 6-min test was videotaped and analyzed by a rater blind to the mouse groups.

To label newly mitotic cells in hippocampal DG, Ki67-staining methods were used in both sexes of WT and SEPT14 KO mice (Wan et al., 2019). In brief, mice were deeply anesthetized with sodium pentobarbital and transcardially perfused with ice-cold 0.1 M phosphate-buffered saline (PBS, pH adjusted to 7.4), followed by 4% paraformaldehyde in ice-cold 0.1 M PBS. Brains were removed and postfixed in a 4% paraformaldehyde solution overnight at 4°C and subsequently cryoprotected in a 30% sucrose solution for 48 h at 4°C. Coronal sections at 20 μm in thickness were made using a microtome (Shandon Cryotome E, Runcorn, Cheshire, United Kingdom). The paraformaldehyde-fixed brain section was stained with rabbit anti-mouse Ki67 (1:200, Chemicon, Temecula, CA, United States) antibody for newly mitotic cells. The immunohistochemical staining was made using the vectastain ABC system (Vector Laboratories, Burlingame, CA, United States). Thus, the brain sections were further incubated with Rhodamine (TRITC) conjugated goat anti-rabbit (1:200, Chemicon, Temecula, CA, United States) secondary antibody and imaged with a Zeiss fluorescent microscope. Staining was quantified in the dorsal (bregma: −1.06 to −2.06 mm) and ventral (bregma: −3.08 to −3.80 mm) hippocampal DG (Franklin and Paxinos, 1997). It was important to note that a Ki67-labeled spot was regarded as a newly proliferated cell by using the standard that its location was aligned with the internal border of the granule cell layer. Ki67-positive cells in every seventh section were added and then divided by the slide selection ratio to obtain the total number of labeled cells for the defined range of the dorsal and ventral hippocampal DG.

Mice were deeply euthanized by intraperitoneal chloral hydrate (40 mg/kg, Sigma, St. Louis, MO, United States), followed by sodium pentobarbital (80 mg/kg, SCI Pharmtech, Inc., Taoyuan, Taiwan). Mice, then, underwent rapid decapitation, and their brains were removed and placed on the dorsal surface of a glass dish sitting on crushed ice. Coronal brain slices (at 1 mm in thickness) were made using a mouse brain slicer matrix (Zivic Instruments, Pittsburgh, PA, United States). Brain tissues containing the dorsal and ventral hippocampus were dissected as previously reported (Deng et al., 2010). These tissue samples were homogenized in an ice-cold lysis buffer containing a protease inhibitor cocktail (Roche, Basel, Switzerland). Homogenates were centrifuged at 12,000 × g for 10 min at 4°C, and the protein concentrations of supernatants were determined by the Bradford method (Bio-Rad Protein Assay, BioRad Laboratories, Hercules, CA, United States) using bovine serum albumin as standards. Samples were heated for 5 min at 95°C in a 2 × sample buffer. The electrophoretically separated proteins were transferred to PVDF membranes (Perkin-Elmer™ Life Sciences, Boston, MA, United States) and blocked with 5% non-fat milk in phosphate saline-Tween-20. The membranes were then incubated overnight at 4°C with the primary antibodies, including rabbit-anti-mouse GluA1 (1:1,000, Merck Millipore) and GluA2 (1:2,000, Merck Millipore), mGluR5 (1:10,000, Abcam), and anti-β actin (1:10,000) antibody (Merck Millipore; Middlesex County, MA, United States). Following incubation with goat anti-rabbit secondary antibody (1:5,000) from Jackson ImmunoResearch Laboratories (West Grove, PA, United States), the blots were developed with the ECLTM Western blot detection kit (Perkin-Elmer™ Life Sciences, Boston, MA, United States). Band density of the protein of interest was quantified using ImageJ image analysis software and expressed as the relative intensity of the respective protein of interest against β actin.

Using male and female WT mice’ training baseline and test latency, a paired Student’s t-test was employed first, serving as a priori test, to assess the validity of IA training. A two-way (sex × genotype) ANOVA was then used to reveal the group differences in IA memory magnitudes, followed by Bonferroni’s post hoc tests if appropriate. However, two-way (empty vs. stranger 1 × group and stranger 1 vs. stranger 2 × group) ANOVAs were used to reveal group differences in sociability and social novelty preference, followed by Bonferroni’s post hoc tests if appropriate. Another two-way (sex × genotype) ANOVA was used to reveal group differences in preference ratios followed by Bonferroni’s post hoc tests if appropriate. In addition, a two-way (sex × genotype) ANOVA was used to analyze group differences in total exploration time spent on this task. Likewise, two-way (sex × genotype) ANOVAs were exploited otherwise to assess sex and SEPT14 KO effects on object location memory, tail suspension immobility time, novelty-induced FR, observational fear conditioning magnitude, observational learning-induced CORT secretion, open field traveling distance, velocity, and time spent in arena center, dorsal, and ventral hippocampal DG stem cell proliferation and protein expression followed by Bonferroni’s post hoc tests if appropriate. All the level of statistical significance was set at p < 0.05.

We first generated SEPT14-deficient mice using the CRISPR/Cas9 KO strategy (Supplementary Figure 1A). Deletion of SEPT14 was verified using PCR (Supplementary Figure 1B). Given the expression of SEPT14 in the testes and brain (Shinoda et al., 2010), RT-PCR was used to examine the SEPT14 expression in these two tissues. We found SEPT14 transcript was abolished in the SEPT14 KO hippocampus, cerebellum (Supplementary Figure 1C), prefrontal cortex (data not shown), and testes (data not shown). A previous study has shown that SEPT14 is expressed in the cortical plate of the developing cerebral cortex (Shinoda et al., 2010). Analysis of the raw data from large-scale single-cell RNA-sequencing work (Yao et al., 2021) indicates that the anterolateral motor cortex and hippocampal region express low levels of Sept14. Given these findings, we used BaseScope in situ hybridization to examine the regional expression pattern of Sept14 mRNA in the adult mouse brain. Sept14 mRNA was detected in the DG, hippocampal CA1–CA3 (Figures 1A,C–H), and cerebral cortex (Supplementary Figures 2A,C–E) in WT mice. In contrast, no signal for Sept14 mRNA was detected in the hippocampus (Figures 1B,I–N) and cerebral cortex (Supplementary Figures 2B,F–H) in SEPT14 KO mice, supporting the specificity of the Sept14 mRNA probe used for BaseScope. The body weight and brain weight were not significantly different between female WT and SEPT14 KO mice. The brain weight was not different between WT and KO mice in males. There were no gross abnormalities in the brain in the KO mice. The body weight was slightly increased while the brain/body weight ratio was slightly decreased in the 8-week old male SEPT14 KO mice (Supplementary Figure 3). Some KO mice have been observed up to 32 weeks of age. There were no obvious health problems in these mice. Both SEPT14^–/^– male and female mice were fertile (data not shown). It seems SEPT14 is not essential for development, fertility, or viability.

Regardless of sex, WT mice’ latency to enter the dark compartment in the IA test was significantly longer than it was in the training baseline [t(38) = 10.87, p < 0.0001], suggesting the validity of our IA training protocol. While SEPT14 KO did not seem to affect the latency to enter the dark compartment in female mice, male SEPT14 KO mice displayed enhanced latency to enter the dark compartment [F(1,39) = 4.983, p = 0.0314], suggesting that SEPT14 KO is associated with elevated aversion-related memory exclusively in male mice (Figure 2).

Male and female WT and SEPT14 KO mice exhibited comparable GluA1 and GluA2 levels in tissues containing dorsal and ventral hippocampus (Supplementary Figure 4). A two-way (sex × genotype) ANOVA revealed that there was an interaction effect of sex and genotype on dorsal hippocampal mGluR5 level [F(1,12) = 12.17, p = 0.0045]. Post hoc tests further revealed that male SEPT14 KO mice exhibited greater mGluR5 levels as compared to same-sex WT mice in tissues containing the dorsal hippocampus (Figure 3A). Likewise, an interaction effect of sex and genotype was noticed on the ventral hippocampal mGluR5 level [F(1,12) = 6.681, p = 0.0239]. Post hoc tests further indicated that male SEPT14 KO mice had higher mGluR5 levels in the ventral hippocampus as compared to same-sex WT mice (Figure 3B).

A two-way (sex × genotype) ANOVA revealed that neither sex nor genotype affected novelty-induced freezing baseline in a 10-min test. Male and female WT mice demonstrated comparable odor-induced FRs (Figure 4A). However, a main effect of genotype was noted on odor-induced FR [F(1,30) = 45.48, p < 0.0001], suggesting that male SEPT14 KO mice displayed lower vanilla odor-induced FRs as compared to their same-sex WT controls regardless of their sexes (Figure 4B). While male and female WT mice had comparable serum CORT levels immediately after the first round of observational learning, male and female SEPT14 KO mice had lower serum CORT levels as compared to their same-sex WT mice [F(1,12) = 19.54, p = 0.0008] (Figure 4C). These results, taken together, suggest that SEPT14 KO appears to dampen observation-related stress-induced CORT secretion and observational fear learning and memory magnitude in both male and female mice.

Male and female WT mice had indistinctive numbers of Ki67-positive cells in the dorsal and ventral hippocampal DG (Figure 5A). SEPT14 KO and WT mice had an indistinctive number of newly proliferated cells in the dorsal hippocampal DG in male and female mice (Figure 5A). In contrast, SEPT14 KO mice had greater newly proliferated cells than that same-sex WT mice in ventral hippocampal DG, regardless of sex [F(1,16) = 23.17, p = 0.0002] (Figure 5B).

A two-way (sex × genotype) ANOVA revealed that the two sexes of SEPT14 KO mice seemed to exhibit significantly less time spent navigating in the lighted center as compared to their same-sex WT ones [F(1,12) = 21.43, p = 0.0006] (Figures 6A,B). Regardless of genotypes (WT and SEPT14 KO), both male and female mice demonstrated comparable navigation distance (Figure 6C) and velocity (Figure 6D) in the stress-related navigation task.

A two-way ANOVA revealed that mice showed indistinctive recognition ratios in test sessions regardless of their sexes and genotypes, suggesting that sex or SEPT14 KO did not affect spatial location memory (Figure 7A). A two-way (sex × genotype) ANOVA revealed that male mice seemed to exhibit greater immobility time than female mice in the tail suspension task [F(1,32) = 20.36, p < 0.0001] (Figure 7B). Nonetheless, SEPT14 KO did not affect such immobility time in either sex.

Regardless of sex and genotype, all four groups of mice spent a higher exploration duration on stranger 1 than on an empty cage [F(1,78) = 112.9, p < 0.0001], suggesting neither sex nor SEPT14 KO affected mice’ sociability (Figure 8A). Interestingly, male mice spent higher total exploration duration in empty and stranger 1 cages as compared to female mice [F(1,39) = 45.31, p < 0.0001]. The latter results imply that female mice seem to exhibit stronger neophobia. Likewise, a two-way (group × genotype) ANOVA revealed that all mice spent higher exploration duration on stranger 2 than on previously met stranger 1 [F(1,78) = 198.7, p < 0.0001] (Figure 8B). Interestingly, male mice spent a higher total exploration duration toward stranger 1 and 2 cages as compared to female mice [F(1,39) = 4.798, p = 0.0345], suggesting females’ stronger neophobia. The implications of these findings, taken together, suggest that sex and SEPT14 KO do not seem to affect sociability or social novelty preference, while female mice appear to have stronger neophobia.