Knockout of Katnal2 Leads to Autism-like Behaviors and Developmental Delay in Zebrafish

KATNAL2 mutations have been associated with autism spectrum disorder (ASD) and other related neurodevelopmental disorders (NDDs) such as intellectual disability (ID) in several cohorts. KATNAL2 has been implicated in brain development, as it is required for ciliogenesis in Xenopus and is required for dendritic arborization in mice. However, a causative relationship between the disruption of Katnal2 function and behavioral defects has not been established. Here, we generated a katnal2 null allele in zebrafish using CRISPR/Cas9-mediated genome editing and carried out morphological and behavioral characterizations. We observed that katnal2-/- embryos displayed delayed embryonic development especially during the convergence and extension (CE) movement. The hatched larvae showed reduced brain size and body length. In the behavioral tests, the katnal2-/- zebrafish exhibited reduced locomotor activity both in larvae and adults; increased nocturnal waking activity in larvae; and enhanced anxiety-like behavior, impaired social interaction, and reduced social cohesion in adults. These findings indicate an important role for katnal2 in development and behavior, providing an in vivo model to study the mechanisms underlying the ASD related to KATNAL2 mutations.


Introduction
Autism spectrum disorder (ASD) represents a group of genetically and clinically heterogeneous neurodevelopmental disorders. The hallmarks of ASD mainly include social communication deficits, restricted interests, and repetitive behavior [1,2]. The global prevalence of ASD has increased steadily in recent years, and it is the leading psychiatric disorder in children under 5 years of age [3]. Currently, the pathogenesis of ASD remains poorly understood. The causes of ASD are complex and include both genetic variants and environmental factors, in which the genetic factors account for up to 80% of disease occurrence [4]. Sequencing projects of large ASD cohorts have identified many candidate genes associated with ASD; however, further studies are required to pinpoint their functions in brain development and behavior [5].
In humans, Katanin Catalytic Subunit A1 Like 2 (KATNAL2, NCBI gene ID: 83473) is located on chromosome 18q21. 1. There have been a number of studies showing that KATNAL2 is an ASD risk gene. Initially, two de novo mutations in the KATNAL2 gene were separately reported in autistic probands in the Simons simplex collection [6,7]. Subsequently, KATNAL2 was identified as a statistically significant candidate gene for ASD (FDR ≤ 0.01) as more de novo variants were reported [8,9]. The association between KATNAL2 and nervous system development is then repeatedly validated by sequencing studies in various populations [6][7][8][9][10][11][12][13][14][15]. In total, 49 mutations in the KATNAL2 gene (27 missense and 22 loss of function mutations) were identified in patients with ASD and neurodevelopmental disorders (NDDs).
KATNAL2 is proposed to be a microtubule (MT)-severing ATPase based on structural prediction. It has been reported that KATNAL2 plays important roles in cellular processes involving cytokinesis, MT dynamics, cell-cycle progression and ciliogenesis [16][17][18]. It is known that microtubule severance and transport play a critical role in neuron formation [19]. Nevertheless, there has been a limited number of studies related to KATNAL2 function in the nervous system. In neonatal mouse, knockout of Katnal2 leads to decreased dendritic arborization [20]. In Xenopus embryos, morpholino-mediated knockdown or genetic knockout of katnal2 causes abnormalities during embryonic development and organogenesis, including reduced brain and eye size [17]. In addition, two other Katanin proteins, KATNAL1 and KATNB1, were also found to affect the development of the nervous system [21][22][23]. These observations suggest that KATNAL2 is essential for early brain development. However, the role of KATNAL2 regarding animal behaviors has not been examined in previous models.
Zebrafish katnal2 shares 56% identity with human KATNAL2. The zebrafish Katnal2 protein contains the same major functional domain as the human homolog. Therefore, it is feasible to use zebrafish as a model organism to study the function of katnal2 gene. To examine the impact of katnal2 loss on behaviors, especially these closely related to ASD, we generated a katnal2 loss-of-function mutation in zebrafish using the CRISPR/Cas9 geneediting technology and then performed a series of morphological analyses and behavioral tests. We found that the katnal2 -/zebrafish exhibited developmental delay, microcephaly, a disturbed sleep pattern, elevated anxiety level and impaired social behaviors, which are reminiscent of key ASD characteristics. These data establish a causative relationship between katnal2 loss and behavioral defects, providing an in vivo model for further basic and translational studies.

Generation of Katnal2 -/-Zebrafish
To examine the role of katnal2 in regulating animal behavior, we generated a zebrafish knockout model using the CRISPR/Cas9 system ( Figure 1A). We first determined the temporal and spatial expression pattern of katnal2 by in situ hybridization. katnal2 is maternally expressed at the early embryonic stage and enriched in the developing brain at 24 h post-fertilization (hpf), indicating that it may function in neurodevelopment ( Figure 1B). We designed and tested three guide RNAs and eventually selected the one targeting a 23-base sequence in exon 6. Using an F3 screening strategy, we isolated a katnal2 mutant allele carrying a deletion of 2 bp and an insertion of 1 bp, resulting in frameshift and truncation of the protein at amino acid 174. The mutation abolished the important AAA+ domain of the Katnal2 protein ( Figure 1A). Moreover, the mRNA levels of katnal2 were significantly decreased in mutant zebrafish larvae at various stages, suggesting that the mutation induced nonsense-mediated decay of the katnal2 mRNA ( Figure 1C). Thus, these results indicate that we successfully generated a genetic knockout allele of katnal2 (katnal2 -/-).

Katnal2 -/-Zebrafish Displayed Developmental Delay during Embryogenesis
We first examined the role of katnal2 during embryonic development by morphological assessment of katnal2 -/and time-matched katnal2 +/+ embryos. The katnal2 -/embryos exhibited an obvious developmental delay phenotype. At 4.3 hpf, when epiboly was initiated in wild-type embryos with the yolk doming into the blastoderm, there was still a flat border between the blastoderm and the yolk in katnal2 -/embryos (Figure 2A). At 5 hpf, the blastoderm became thinner in katnal2 +/+ embryos, whereas it was much thicker in katnal2 -/embryos (Figure 2A,B). From 6 to 10 hpf, the epiboly progression in katnal2 -/embryos was also delayed, judged by the slowed spreading of the blastoderm toward the vegetal pole (Figure 2A,C,D). As a result, when katnal2 +/+ embryos completed gastrulation at 10 hpf, katnal2 -/embryos only reached at about 90% epiboly (Figure 2A). At 12 hpf, all katnal2 -/embryos completed epiboly and presented a shortened anteroposterior axis compared to time-matched katnal2 +/+ embryos (Figure 2A). Loss of katnal2 not only affected epiboly but also affected the convergence and extension (CE) movements ( Figure 2E-G). Upon completion of epiboly, the width of the neural plate of katnal2 -/embryos was significantly wider than katnal2 +/+ embryos, as determined by dlx3b mRNA expression ( Figure 2E,G). Moreover, the katnal2 -/embryos hatched later than the katnal2 +/+ embryos ( Figure 2H). Subsequently, we measured the body length, distance between the eyes, distance between the optic tecta, and area of forebrain and midbrain at 4 dpf to assess the brain development upon loss of katnal2. Compared to katnal2 +/+ larvae, the body length, width between the eyes, distance between the optic tecta, and area of forebrain and midbrain of katnal2 -/zebrafish were significantly decreased ( Figure 3A-F). Taken together, these data indicate that katnal2 is essential for proper embryonic and larval development in zebrafish.

Katnal2 -/-Zebrafish Displayed Developmental Delay during Embryogenesis
We first examined the role of katnal2 during embryonic development by morphological assessment of katnal2 -/-and time-matched katnal2 +/+ embryos. The katnal2 -/-embryos exhibited an obvious developmental delay phenotype. At 4.3 hpf, when epiboly was initiated in wild-type embryos with the yolk doming into the blastoderm, there was still a flat border between the blastoderm and the yolk in katnal2 -/-embryos (Figure 2A). At 5 hpf, the blastoderm became thinner in katnal2 +/+ embryos, whereas it was much thicker in 2E,G). Moreover, the katnal2 -/embryos hatched later than the katnal2 +/+ embryos ( Figure 2H). Subsequently, we measured the body length, distance between the eyes, distance between the optic tecta, and area of forebrain and midbrain at 4 dpf to assess the brain development upon loss of katnal2. Compared to katnal2 +/+ larvae, the body length, width between the eyes, distance between the optic tecta, and area of forebrain and midbrain of katnal2 -/-zebrafish were significantly decreased ( Figure 3A-F). Taken together, these data indicate that katnal2 is essential for proper embryonic and larval development in zebrafish.

Katnal2 -/-Larvae Displayed Abnormal Sleep States and Impaired Locomotion Activity
Since human KATNAL2 mutations have been associated with ASD, we sought to determine whether the loss of function of katnal2 leads to alterations in behaviors. We first examined larval circadian rhythms by measuring sleep length, sleep bouts, and average waking activity under day-night cycles for 2 days between 5 and 7 dpf. Remarkably, katnal2 -/-larvae showed a decreased average sleep length, increased sleep bouts, and decreased sleep length per bout at night ( Figure 4A,C-E). However, the sleep latency during the day-night switch was not significantly altered in katnal2 -/-larvae ( Figure 4F). The average waking activity of katnal2 -/-larvae was significantly higher than that of katnal2 +/+ larvae at night, whereas it was comparable to that of katnal2 +/+ larvae during the daytime ( Figure 4B,G). These data show that the loss of katnal2 causes sleep disturbance.
We also examined the responses evoked by short-time light changes (light/dark switch, 100 lx for brightness, and 0 lx for dark). After a 30 min habituation period, each larva was recorded for 30 min over three light/dark cycles (each consisting of 5 min in light and 5 min in dark setting per cycle). Compared with katnal2 +/+ larvae, katnal2 -/-larvae moved significantly shorter total distances in all periods during the test ( Figure 4H,I). The transition from light to dark caused a sudden increase in movement velocity, while the transition from dark to light caused a sudden decrease. katnal2 -/-larvae showed a delayed response pattern to changes in illumination ( Figure 4H). These data suggest that loss of katnal2 impairs zebrafish larval locomotion activity and attenuates the response to stimuli.

Katnal2 -/-Larvae Displayed Abnormal Sleep States and Impaired Locomotion Activity
Since human KATNAL2 mutations have been associated with ASD, we sought to determine whether the loss of function of katnal2 leads to alterations in behaviors. We first examined larval circadian rhythms by measuring sleep length, sleep bouts, and average waking activity under day-night cycles for 2 days between 5 and 7 dpf. Remarkably, katnal2 -/larvae showed a decreased average sleep length, increased sleep bouts, and decreased sleep length per bout at night ( Figure 4A,C-E). However, the sleep latency during the day-night switch was not significantly altered in katnal2 -/larvae ( Figure 4F). The average waking activity of katnal2 -/larvae was significantly higher than that of katnal2 +/+ larvae at night, whereas it was comparable to that of katnal2 +/+ larvae during the daytime ( Figure 4B,G). These data show that the loss of katnal2 causes sleep disturbance.
We also examined the responses evoked by short-time light changes (light/dark switch, 100 lx for brightness, and 0 lx for dark). After a 30 min habituation period, each larva was recorded for 30 min over three light/dark cycles (each consisting of 5 min in light and 5 min in dark setting per cycle). Compared with katnal2 +/+ larvae, katnal2 -/larvae moved significantly shorter total distances in all periods during the test ( Figure 4H,I). The transition from light to dark caused a sudden increase in movement velocity, while the transition from dark to light caused a sudden decrease. katnal2 -/larvae showed a delayed response pattern to changes in illumination ( Figure 4H). These data suggest that loss of katnal2 impairs zebrafish larval locomotion activity and attenuates the response to stimuli.

Adult Katnal2 -/-Zebrafish Displayed Impaired Locomotion Activity and Elevated Anxiety Level
The katnal2 -/-allele is viable and fertile, allowing behavioral tests in the adult stage.
We performed an open field test to examine the locomotion activity and anxiety level of adult zebrafish ( Figure 5A). A significant reduction in the total distance and swimming velocity was observed in katnal2 -/-zebrafish compared to katnal2 +/+ ( Figure 5C,D). We further calculated the time spent in the central and peripheral zone for both genotypes.

Adult Katnal2 -/-Zebrafish Displayed Impaired Locomotion Activity and Elevated Anxiety Level
The katnal2 -/allele is viable and fertile, allowing behavioral tests in the adult stage.
We performed an open field test to examine the locomotion activity and anxiety level of adult zebrafish ( Figure 5A). A significant reduction in the total distance and swimming velocity was observed in katnal2 -/zebrafish compared to katnal2 +/+ (Figure 5C,D). We further calculated the time spent in the central and peripheral zone for both genotypes.  To further confirm that loss of katnal2 causes higher anxiety, we also performed the novel tank test and the light-and-dark box test. When dropped to a novel tank, we found that katnal2 -/-zebrafish spent more time in the bottom zone and less time in the middle zone than katnal2 +/+ zebrafish, again suggesting an elevated anxiety level (Figure5B,G,H). Similar to the result of the open field test, katnal2 -/-zebrafish displayed reduced total distance traveled and swimming velocity in the novel tank. In addition, katnal2 -/-zebrafish displayed more freezing time than katnal2 +/+ zebrafish ( Figure 5I). In the light-and-dark box test, katnal2 -/-zebrafish displayed a significantly lower frequency of swimming from the dark zone into the light zone and more freezing time in the light zone ( Figure 5J-L). Taken together, these data suggest that katnal2 loss of function mutation leads to lower locomotion activity and a higher anxiety level, which correlates with the observations in the larvae. To further confirm that loss of katnal2 causes higher anxiety, we also performed the novel tank test and the light-and-dark box test. When dropped to a novel tank, we found that katnal2 -/zebrafish spent more time in the bottom zone and less time in the middle zone than katnal2 +/+ zebrafish, again suggesting an elevated anxiety level ( Figure 5B,G,H). Similar to the result of the open field test, katnal2 -/zebrafish displayed reduced total distance traveled and swimming velocity in the novel tank. In addition, katnal2 -/zebrafish displayed more freezing time than katnal2 +/+ zebrafish ( Figure 5I). In the light-and-dark box test, katnal2 -/zebrafish displayed a significantly lower frequency of swimming from the dark zone into the light zone and more freezing time in the light zone ( Figure 5J-L). Taken together, these data suggest that katnal2 loss of function mutation leads to lower locomotion activity and a higher anxiety level, which correlates with the observations in the larvae.

Katnal2 -/-Zebrafish Exhibited ASD-like Behaviors
Deficit in social communications is one of the key features of ASD; therefore, we examined the social activities of the katnal2 -/allele. In the three tank tests, katnal2 +/+ zebrafish spent most of the time in the social zone and were only occasionally present in the middle or empty zone. In contrast, katnal2 -/zebrafish spent significantly less time in the social zone and comparatively more time in the middle and empty zone ( Figure 6A-D). In the mirror test, the katnal2 -/zebrafish spent less time in the mirror zone where they could approach their own reflections ( Figure 6E,F). Taken together, these data suggest that katnal2 -/zebrafish show reduced interest in social interaction. Deficit in social communications is one of the key features of ASD; therefore, we examined the social activities of the katnal2 -/-allele. In the three tank tests, katnal2 +/+ zebrafish spent most of the time in the social zone and were only occasionally present in the middle or empty zone. In contrast, katnal2 -/-zebrafish spent significantly less time in the social zone and comparatively more time in the middle and empty zone ( Figure 6A-D). In the mirror test, the katnal2 -/-zebrafish spent less time in the mirror zone where they could approach their own reflections ( Figure 6E,F). Taken together, these data suggest that katnal2 -/-zebrafish show reduced interest in social interaction.  Zebrafish are natural schooling fish. To test the group behavior of the katnal2 -/allele, we performed the shoaling test to assess the social cohesion among homogeneous groups of zebrafish. In this assay, adult katnal2 +/+ or katnal2 -/zebrafish were placed in the testing tank in groups of eight fish ( Figure 6G). The average inter-fish distance, maximum inter-fish distance and minimum inter-fish distance was measured every 20 s for all pair combinations. katnal2 +/+ zebrafish typically swam in the same direction and maintained a short inter-fish distance at all timepoints, whereas the katnal2 -/zebrafish exhibited increased average inter-fish distance as a significant number of zebrafish swimming away from the group, showing a decreased social cohesion behavior ( Figure 6H-J). Repetitive and stereotyped behavior is another core symptom of ASD. Compared to the controls, katnal2 -/zebrafish exhibited a significantly higher frequency of stereotypical behaviors, including repetitive big circling, small circling, walling and cornering ( Figure 6K-L). Taken together, these data showed that knockout of katnal2 in zebrafish leads to phenotypes reminiscent of the core features of ASD.

Discussion
Although human KATNAL2 has been identified as a candidate gene for ASD, so far, there have been no reported behavioral studies in Katnal2 knockout animals. In this study, we generated a katnal2 loss-of-function allele in zebrafish using the CRISPR/Cas9 geneediting technology, revealed the morphological abnormity of katnal2 -/zebrafish at the early developmental stage, and characterized the behavioral changes of mutants both in larval stage and in adulthood.
The katnal2 -/embryos exhibited delayed epiboly initiation and progression, resulting in reduced elongation of the anteroposterior axis. Katnal2 is a paralogue of the microtubule severing enzymes Katna1 and Katnal1; therefore, it is proposed to possess the microtubulesevering capacity [16,24]. KATNAL2 is associated with a variety of MT-dependent cellular processes [16]. Katnal2 knockdown can result in defective ciliogenesis, inefficient cell division, increased cell and nuclear volume, formation of abnormal multipolar mitotic spindles, mitotic defects, increased MT acetylation, and cell-cycle alterations [16,17]. Furthermore, a previous study has shown that microtubule arrays formed in yolk syncytial layer are critical for driving vegetal migration of the yolk syncytial nuclei [25]. Hence, katnal2 loss of function in mutant embryos may affect the structure and dynamics of microtubules, which are critical for cell migrations. We propose that the developmental delay of katnal2 -/embryos is due to the defects in the microtubule network.
The decreased distance between the eyes, decreased distance between the optic tecta, and reduced area of forebrain and midbrain indicate that katnal2 is essential for proper brain development, which is consistent with observations in the katnal2 knockdown Xenopus model and katnal2 knockout rodent model [17,20]. The role of katnal2 in neurodevelopment and development control could relate to the role of katnal2 in multiple MT-dependent cellular processes such as cell division and cell migration. It will be of great importance to further study the cellular mechanisms underlying the critical roles of katnal2 in brain development.
In recent years, the zebrafish has emerged as a valuable animal model in ASD research [26,27]. Many behavioral tests have been developed in zebrafish models, including the evaluation of social interaction, social cohesion, inhibitory avoidance, repetitive stereotyped behavior, environmental adaptation, the fear and anxiety response, novelty-seeking, and aggression [28,29]. We therefore performed behavioral tests and observed behavioral abnormalities in the katnal2 -/allele. The katnal2 -/larvae displayed impaired locomotor activity, which may be directly related to the developmental delay or neurodevelopment abnormalities. Interestingly, katnal2 -/larvae showed more fragmented sleep time and a marked increase in waking activity at night. The selectively enhanced nocturnal waking activity may suggest that katnal2 deficiency caused an imbalance in excitatory and inhibitory signaling in mutant fish, which has been reported in a previous study [30]. In the open field test, katnal2 -/zebrafish displayed decreased distance moved, velocity, and time to explore the center zone. In the novel tank test, katnal2 -/zebrafish showed a reduction in distance moved and time spent in the middle zone as well as an increase in time spent in the bottom zone. In addition, the freezing time was also increased. In the three-tank social interaction test, the open field test, the shoaling test, and the mirror test, the katnal2 -/zebrafish exhibited core ASD-like behaviors, including impaired social interaction and cohesion as well as repetitive or stereotyped behaviors. Unfortunately, detailed phenotypical data were not reported for patients carrying KATNAL2 mutations in large-scale sequencing projects. We cannot make a direct comparison between the patient symptoms and the zebrafish mutant phenotypes. However, the defects we observed, such as developmental delay, anxiety, locomotion defects and sleep disturbance, are typical co-morbidities besides the core features of ASD. Our findings could provide clues for clinicians to carefully and thoroughly examine the phenotypes of patients carrying KATNAL2 mutations. Additionally, high-throughput drug-screening methods have been established for zebrafish larvae [31,32]. Since the katnal2 mutants exhibit abnormal morphology and behaviors as early as the larval stage, it is feasible to perform large-scale screening of small molecules to identify therapeutic candidates. Therefore, the knockout model we generated would help facilitate further studies related to the molecular mechanism of Katnal2 and drug screening.

Zebrafish Culture
Wild-type AB strain zebrafish were purchased from the China Zebrafish Resource Center. Adult fish were maintained in a water cycling system at 28.5 • C with a 14 h-light/10 h-dark circadian cycle. Embryos were raised in E3 medium in a 28.5 • C incubator. All zebrafish protocols used in this study were approved by the Animal Care and Use Committee of the animal core facility at Huazhong University of Science and Technology.

Generation of Katnal2 Mutant Zebrafish
The katnal2 -/zebrafish were generated using CRISPR/Cas9, as previously described [33,34]. A mixture of 600 pg of Cas9 mRNA and 80 pg of guide-RNA was microinjected into one cell stage of fertilized zebrafish eggs. Efficiencies of the guide-RNAs were validated by PCR and Sanger sequencing, and a guide-RNA targeting exon 6 (sequence, 5 -GTCTCCTATCATCAGGAACG-3 ) was selected. The homozygous katnal2 -/was isolated in a standard F3 screen. The F3 katnal2 +/+ littermates were maintained and used as controls in all experiments.

Whole Mount Immunofluorescence
To perform whole-mount immunofluorescence with acetylated α-tubulin (T7451, Sigma, St. Louis, MO, USA), 4 dpf larvae were fixed in Dent's solution (80% methanol in DMSO) overnight at 4 • C. After rehydration with decreasing methanol series (75%, 50%, and 25%) in PBS, the samples were bleached with bleach solution (0.05 g KOH in 9 mL PBST and 1 mL H 2 O 2 ), permeabilized with proteinase K (10 µg/mL), post-fixed with 4% PFA, and washed twice with IF buffer (1% BSA in PBST). After incubation with blocking solution (10% FBS in IF buffer) for 1 h at room temperature, the samples were incubated with the acetylated α-tubulin in blocking solution overnight at 4 • C. Finally, the larvae were incubated with the Alexa Fluor 488 s antibodies (1:500, A10680, Invitrogen, Carlsbad, CA, USA) in blocking solution for 2 h at room temperature and washed with IF buffer for twice. Measurements of the distance between optic tecta were performed as previously described [36,37].

Larval Locomotion Activity Test
A larval locomotion test was performed as previously described [38]. Free swimming 5 dpf larvae were habituated in 96-well square plates, with one animal per well, in the observation chamber of the Danio Vision tracking system (Noldus, Wageningen, The Netherlands), and videos were recorded for 60 min. After 30 min of habituation in light conditions, each larva was recorded for a total of 30 min with three dark/light cycles, each of which consists of 5 min of dark and 5 min of light. The light intensity was 100 lx and the frame rate was 25/s. The locomotion of each larva was tracked and analyzed by EthoVision XT7 software (Noldus, Wageningen, The Netherlands). For analysis of the locomotor activity, raw data were converted into the velocity (cm/s) of each larva per 30 s time-bin.

Sleep and Waking Activity of Zebrafish Larvae
The sleep and waking activity were measured as previously described [39][40][41]. Freeswimming 4.5 dpf larvae were habituated in 96-well square plates, with one larva per well, in the observation chamber of the Danio Vision tracking system for one night. After 10 h of habituation, each larva was recorded under controlled lighting conditions (10 h of dark and 14 h of light cycles) from 5 dpf. Six or more consecutive seconds of non-movement were classified as sleep, and all other bouts were classified as being awake. Sleep bouts, sleep bout durations, the total sleep time, average sleep (sleep ratio), sleep latency, and average waking activity were calculated.

Open Field Test
All behavioral experiments of adult zebrafish were conducted during the same time frame each day. The adult zebrafish used for behavioral tests were 6-month-old male fish. The open field test was performed as previously described [40]. The open field test was performed in a water tank of 20 × 20 × 10 cm. Individual fish were placed in the center zone and allowed to freely swim inside the tank. Videos were recorded from the top viewpoint of the tank for 15 min. The distance moved, the average velocity, and the time spent in the center zone and the peripheral zone were calculated.
To analyze repetitive and stereotyped behaviors, we conducted the open field test in a zebrafish home tank with a 10 L volume. Videos were recorded for 12 min after 5 min of habituation. In this test, we used a double-blind method to evaluate the swimming patterns within each 15 s. Repetitive behaviors including walling, cornering, small circling, and big circling were scored.

Novel Tank Assay
The novel tank test was performed according to a previously described protocol [42]. Each zebrafish was placed in a transparent tank of 28 × 20 × 5 cm dimensions. The backside of the tank was covered with a light plate to aid video recording. The tester fish was dropped into the tank from the top. Videos were recorded for 12 min from the lateral viewpoint of the tank. For data analysis, we artificially divided the tank into top, middle, and bottom zones. The time spent in each zone as well as the freezing time during the test were quantified.

Light Dark Box Assay
The light-dark preference test was performed based on a protocol modified from a previous study [43]. The light-and-dark box test was performed in a water tank with inner dimensions measuring 21 × 10 × 7.5 cm. In this experiment, half of the mating tank was taped with black tape and the other half was illuminated. Video recordings started after an acclimation time of 5 min, for the tester fish to explore the tank. Videos were recorded from the top viewpoint of the tank for 6 min. The frequency entering into the light zone and the total freezing time in the light zone were quantified.

Social Interaction Assay
The social interaction assay was performed in a transparent tank divided into 3 sections with two transparent barriers as previously described [44]. A group of three zebrafish was placed in the left side, and a single tester zebrafish was placed in the middle section. The right side of the tank was kept empty. The second section with dimensions measuring 20 ×15 × 13 cm was further divided into three equal-sized subzones; the zone nearest to the social cue was designated "social zone", the second zone was designated "middle zone", and the third zone was designated "empty zone". Behavioral recordings typically started after an acclimation of 5 min. Videos were recorded from the lateral viewpoint of the tank for 15 min. The time spent in each zone was then calculated.

Mirror Assay
The mirror test was performed as the protocol was modified from a previous study [45]. The tank was transparent and with dimensions measuring 28 × 20 × 5 cm. The backside of the tank was covered with a light plate to aid data recording. A mirror was inserted into one side of the tank, and the region in which the tester fish could touch the mirror was designated as a "mirror zone" (3 cm in width, the same to the average body length of adult zebrafish). Videos were recorded from the lateral viewpoint of the tank for 15 min. The time spent in the "mirror zone" was calculated.

Shoaling Assay
The zebrafish is a schooling fish that typically prefers to stay close to its congeners [45,46]. For the shoaling assay, 8 six-month-old zebrafish were placed together in a transparent tank with dimensions measuring 40 × 40 × 15 cm and monitored by video tracking as previously described [33]. Video was recorded for 5 min after a 25 min adaptive phase. The distance between any two zebrafish was measured and analyzed.

Statistical Analysis
Statistical analysis was performed using GraphPad Prism 7 software (GraphPad, La Jolla, CA, USA). In all experiments, comparisons between katnal2 +/+ and katnal2 -/zebrafish were performed with two-tailed unpaired Student's t-tests. Data are presented as mean ± S.E.M. In all tests, p < 0.05 was considered to be statistically significant. In the figures, * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001, and ns indicates no significance.

Conclusions
In this study, we generated and characterized a katnal2 knockout allele in zebrafish. The katnal2 -/zebrafish displayed autism-like behavioral characteristics as well as developmental delay. These results established a causative relationship between katnal2 loss and behavior defects, providing supporting evidence for the association between the KATNAL2 variants and ASD in humans.