Environmental enrichment has minimal effects on behavior in the Shank3 complete knockout model of autism spectrum disorder

Abstract Introduction Several studies have supported the use of enriched environments to prevent the manifestation of ASD‐like phenotypes in laboratory rodents. While the translational value of such experiments is unknown, the findings have been relatively consistent across many different models. Methods In the current study, we tested the effects of early environmental enrichment on a mouse model of ASD with high construct validity, the Shank3 ∆e4–22 mice our laboratory previously generated and characterized. Results Contrary to previous reports, we found no benefits of enriched rearing, including no change in repetitive self‐grooming or hole‐board exploration. Instead, we found that early environmental enrichment increased anxiety‐like behavior in all mice regardless of genotype and decreased motor performance specifically in wild‐type mice. Conclusions Although using a different enrichment protocol may have rescued the phenotypes in our mouse model, these results suggest that a “one‐size fits all” approach may not be the best when it comes to behavioral intervention for ASD and underscores the need for effective pharmaceutical development in certain genetic syndromes with severe symptom presentation.

Although the type of enrichment and the developmental stages during exposure varied considerably, all these previous studies reported improvements on at least one behavioral outcome and seldom reported any adverse effects of enrichment. Moreover, some of the models utilized previously lack construct validity in terms of a known, highly penetrant cause of ASD, such as a genetic mutation that has been consistently linked to the disorder in humans.
However, to date, no study has explored the role of the environment on behavioral phenotypes in Shank3 mutant mice.
In the present study, we tested the effect of early environmental enrichment on some of the most robust behavioral phenotypes that we previously reported in our complete Shank3 knockout model (∆e4-22) of ASD. Although ASD involves impaired social communication, we did not observe a strong social phenotype in our previous characterization of these mice using available behavioral assays  and therefore focused on repetitive behaviors and comorbidities, such as anxiety-like behavior and motor impairments.
We found that, contrary to previous findings using other mouse models of ASDs, early environmental enrichment did not prevent the manifestation of behaviors that resemble repetitive behaviors in Shank3 ∆e4-22 mice: self-grooming and restricted head poking on the hole-board task. We also found that enrichment decreased motor performance on the rota-rod task specifically in wild-type mice and increased anxiety-like behavior in all mice, regardless of genotype.

| Animals
Shank3 ∆e4-22 mice were previously generated and characterized by our laboratory and were maintained on a C57BL/6J background after backcrossing for at least eight generations .
All experiments were conducted with protocols approved by the Institutional Animal Care and Use Committee at Duke University.

| Rearing conditions
Similar to a previous study utilizing a mouse model of Rett syndrome, mice were placed in enriched environment early in development (starting at postnatal day 10, P10) in an attempt to maximize effectiveness of the enrichment paradigm (Lonetti et al., 2010). In the enrichment condition, starting at P10, mice were housed with two lactating dams and two litters per cage, whereas in the standard condition, mice were housed with one lactating dam and one litter per cage.

| Behavioral testing
Three cohorts of approximately 45 Shank3 ∆e4-22 (+/+, +/−, and −/−) mice were tested in a battery of assays in order to assess the effects of environment on anxiety-like behavior, motor function, and stereotypy at 8-10 weeks of age. The experimenter was blind to both genotype and rearing conditions until data analysis. The assays were performed in order they are described, but not every cohort was put through every test. See Table 1 for details of each cohort, including the numbers of mice in each experimental group and the tests performed. Both male and female mice were used for all experiments presented in this study.

| Zero maze
Mice were introduced into a closed portion of the maze and were given 5 min of free exploration under dim (40-60 lux) illumination. Activity was scored by Ethovision XT 7 (Noldus Information Technologies) using a high-resolution camera suspended 180 cm above the center of the maze. Tracking profiles were generated by Ethovision XT software and were used to measure the time each mouse spent in the open portions of the maze.

| Open field
Activity in the open field was measured over 1 hr in an automated Omnitech Digiscan apparatus (AccuScan Instruments, Columbus, OH). AccuScan software scored the total distance travelled and the time spent in the center of the apparatus.

| Rota-rod
Motor performance was assessed on a steady-speed rota-rod (Med-Associates) set to 20 rotations per minute. Each mouse attempted four trials with an intertrial interval of 30 min. The latency to fall off the apparatus was recorded. If a mouse displayed three successive passive rotations, this was also counted as a fall.
Each trial ended after 5 min, and any mouse that successfully remained on the rod at the end of the trial was recorded as a latency of 300 s.

| Grooming
Individual animals were acclimated to clean home cages for 5 min prior to filming (MediaRecorder 2; Noldus Information Technologies).
Mice were filmed for 10 min, and grooming behavior was handscored using Observer 9 XT (Noldus Information Technologies).

| Hole-board
Mice were allowed 5 min of exploration on a 16-hole-board apparatus. Animals were filmed with a digital video camera and handscored for the numbers of nose pokes and the location of each nose poke. Back-to-back nose pokes were defined as when the animal made two or more consecutive visits to the same hole.

| Statistical analysis
Graphs were produced, and statistical analysis was performed in GraphPad Prism 7. For the rota-rod data, a repeated-measures ANOVA was performed and each of the six groups (three genotypes × two rearing conditions) was independently compared with each other group with Tukey's multiple comparison post hoc test.
For all other tests, a two-way ANOVA for genotype and rearing condition was performed. Significant differences in genotype were followed up with Tukey's multiple comparison post hoc test. Statistical significance was defined as p < 0.05.

| Shank3 complete knockout mice display repetitive behaviors which are not ameliorated by early environmental enrichment
Consistent with our previous reports, the Shank3 ∆e4-22 −/− mice engaged in increased amounts of repetitive self-grooming, compared to both Shank3 ∆e4-22 +/− and +/+ mice (Figure 1a). There was no effect of environment on the expression of this behavior.
Similarly, the Shank3 ∆e4-22 −/− mice engaged in repetitive behavior, in terms of increased back-to-back pokes compared to +/+ mice, on the hole-board task and this phenotype was not affected by the rearing conditions of the mice (Figure 1b).

| Early environmental enrichment has a negative impact on motor performance which is specific to wild-type mice
In our previous characterization of the Shank3 complete knockouts, we found the most severe motor performance deficits on the F I G U R E 3 Enriched rearing decreases motor performance in wild-type mice (a) All +/+ mice raised under standard conditions performed perfectly on the 20 r.p.m. steady-speed rota-rod. However, +/+ mice raised in enriched environments performed significantly worse than +/+ raised in standard cages (RMANOVA, main effect of genotype/environment group, p < 0.05; Tukey's multiple comparisons +/+ standard vs. +/+ enriched, p < 0.05). n = 14-15 per group. (b) There was no statistically significant effect of environment for the +/− mice. n = 17-23 per group. (c) There was no statistically significant effect of environment for the −/− mice. n = 10-11 per group. *signifies p < 0.05, **signifies p < 0.01, ***signifies p < 0.001, and n.s. stands for not significant TA B L E 2 Summary of key findings from previous studies utilizing environmental enrichment and rodent models of ASD Male and female mice lacking the µ-opioid receptor gene (Oprm1 −/−) Enriched mice were housed with an additional lactating female from approximately 1 week before birth to weaning Positive effects: • Enrichment increased body weight of all mice at PND8, but this normalized by weaning. • There was a significant effect of environment such that enriched mice (of both genotypes) spent more time interacting with the social stimulus in the juvenile social approach-avoidance test compared to standard-housed mice. • For male mice specifically, there was also an effect of enrichment on adult social behavior. While mutant mice spent less time investigating an intruder mouse regardless, enrichment increased the investigation time in both wild types and mutants. Phenotypes that failed to improve: • While enrichment decreased the number of PND8 USVs in wild-type mice, mutants had decreased numbers of USVs compared with wild type, and this was not affected by enrichment. Exacerbations: • Not reported Reynolds et al. (2013) Male BTBR inbred mouse strain The mice were placed in enriched housing in groups of 8 at 7 weeks of age for 30 days. The enrichment cage was a three-floor dog kennel with various toys that were changed every 5 days Positive effects: • BTBR mice self-groom significantly more than C57BL/6 mice, and this phenotype was rescued by environmental enrichment Phenotypes that failed to improve: • Not reported Exacerbations: Male mice exposed prenatally to VPA One week after weaning, mice were in enriched environments for four weeks. This consisted of a larger cage (65 × 35 × 30 cm) filled with toys that were repositioned twice per week Positive effects: • VPA-exposed mice housed in standard conditions spent more time in the closed arms of the elevated plus maze compared with controls, but enriched VPA-exposed mice were similar to controls. • VPA-exposed mice housed in standard conditions spent less time sniffing a stimulus in a social interaction test compared to controls, but enriched VPA-exposed mice were similar to controls. • VPA-exposed mice housed in standard conditions showed deficits in novel object recognition, but enriched VPA-exposed mice were similar to controls. Phenotypes that failed to improve: • VPA-exposed mice were hypoactive, reared less, and had fewer center crossings in the open field irrespective of environmental condition. Exacerbations: • Not reported TA B L E 2 (Continued) (Continues) steady-speed variation of the rota-rod. All fifteen of the +/+ mice that were raised in standard conditions performed perfectly throughout all four trials, and this was not true for mice in any other experimental condition (Figure 3a-c). By repeated-measures ANOVA, there was a significant effect of experimental group on motor performance; however, the only significant differences in post hoc analysis were between +/+ mice raised under standard conditions and +/+ mice raised in enriched environments (Figure 3a).

| D ISCUSS I ON
Our results indicate that raising Shank3 mutant mice in enriched environments has little effect on their behavior, which is somewhat surprising, given that environmental enrichment has been shown to alleviate behavioral phenotypes in a number of other rodent models of ASDs (Favre et al., 2015;Garbugino et al., 2016;Kerr et al., 2010;Kondo et al., 2008;Lacaria et al., 2012;Lonetti et al., 2010;Nag et al., 2009;Oddi et al., 2015;Restivo et al., 2005;Reynolds et al., 2013;Schneider et al., 2006;Yamaguchi et al., 2017). A summary of previous findings along with findings from the current study is provided in Table 2.
Particularly, striking was that although we modeled our enrichment paradigm after the one described in a previous study that found positive effects of enrichment on a mouse model of Rett syndrome (Lonetti et al., 2010) Male rats prenatally exposed to valproic acid (VPA) Rats were enriched from PND 23-123. Enriched rats were in larger cages in groups of six and were given access to various toys. Enriched environments were either "predictable" or "unpredictable," which simply meant whether or not the toys were changed at all. In the "unpredictable" condition the toys were changed twice weekly Positive effects: • VPA-exposed rats housed in standard conditions had increased fear conditioning responses, but VPA-exposed rats raised in "unpredictable" enriched environments were similar to controls. However, VPA-exposed rats raised in "predictable" enriched environments had an impairment in fear conditioning. Phenotypes that failed to improve: • VPA-exposed rats had increased repeated entries in the Y-maze task regardless of housing condition. Exacerbations: • VPA-exposed rats housed in the "predictable" enriched environments had higher sociability in the three-chamber test compared to control rats housed in similar conditions. However, this appears to be mostly because of a decrease in sociability in the enriched controls compared to standard-housed controls. • VPA-exposed rats housed in "predictable" enriched environments spent more time in the open arms of the elevated plus maze compared to the other groups TA B L E 2 (Continued) 12-16 months (Neul et al., 2010). Another possible explanation is that because in our attempt to remain blind to experimental con- We did not observe any sex differences in this study or in our previous characterization of the Shank3 knockout mice.
Although human males are more likely to be diagnosed with ASD compared to females, there is no apparent sex bias for SHANK3 It remains to be determined whether humans are more sensitive to smaller changes in dosage of the SHANK3 protein, whether the highly controlled environment of the laboratory prevents the expression of ASD-like phenotypes in heterozygotes, or whether the current methods used for assessing behavior in rodents prevent us from detecting more subtle changes most relevant to ASD.

ACK N OWLED G M ENTS
We