Tetratricopeptide repeat domain 9A modulates anxiety-like behavior in female mice

Tetratricopeptide repeat domain 9A (TTC9A) expression is abundantly expressed in the brain. Previous studies in TTC9A knockout (TTC9A−/−) mice have indicated that TTC9A negatively regulates the action of estrogen. In this study we investigated the role of TTC9A on anxiety-like behavior through its functional interaction with estrogen using the TTC9A−/− mice model. A battery of tests on anxiety-related behaviors was conducted. Our results demonstrated that TTC9A−/− mice exhibited an increase in anxiety-like behaviors compared to the wild type TTC9A+/+ mice. This difference was abolished after ovariectomy, and administration of 17-β-estradiol benzoate (EB) restored this escalated anxiety-like behavior in TTC9A−/− mice. Since serotonin is well-known to be the key neuromodulator involved in anxiety behaviors, the mRNA levels of tryptophan hydroxylase (TPH) 1, TPH2 (both are involved in serotonin synthesis), and serotonin transporter (5-HTT) were measured in the ventromedial prefrontal cortex (vmPFC) and dorsal raphe nucleus (DRN). Interestingly, the heightened anxiety in TTC9A−/− mice under EB influence is consistent with a greater induction of TPH 2, and 5-HTT by EB in DRN that play key roles in emotion regulation. In conclusion, our data indicate that TTC9A modulates the anxiety-related behaviors through modulation of estrogen action on the serotonergic system in the DRN.

As our previous study showed the TTC9A negatively regulates estrogen action in mammary development, we tested the hypothesis whether the effect of TTC9A gene deletion on anxiety is also related to its effect on estrogen response in female mice. This would be done by examining the effect of estrogen on anxiety-like behaviors in ovariectomized (OVX) mice. After ovariectomy, mice were given a 10-day recovery period and subsequently injected with either sesame oil (CTRL) or 17-β -estradiol benzoate (EB) in sesame oil. Twenty four hours after injection, two-way ANOVA showed no significant difference in the interaction, genotype, and treatment in the duration of time-spent in the open-arm (F (1,18) < 1.25, p = n.s.); however, TTC9A −/− OVX-EB mice showed an increase in closed-arm duration (p = 0.02) in the elevated plus-maze (Fig. 5A,B). Analysis of the number of entries to closed or open arm showed no significant difference in interaction, genotype, and treatment (closed: F (1,21) < 0.36, open: F (1,22) < 2.66), all p = n.s.; data not shown). In cylinder test, two-way ANOVA with post-hoc analysis also revealed no significant difference in genotype (F (1,(19)(20) < 0.412), p = n.s.), treatment (F (1,(19)(20) < 0.179, p = n.s.) and their interactions (F (1,(19)(20) < 1.266), p = n.s.) for immobility and rearing behaviors in the cylinder environment (Fig. 5C,D). Forty eight hours after injection, in the home-cage emergence test, two-way ANOVA revealed a significant difference in interaction (F (1,21) = 17.05, p = 0.0005), but not in genotype or treatment (F (1,21) = 0.29 and F (1,21) = 0.01, respectively, all p = n.s.). Post-hoc analysis showed a significant reduction on the escape latency (p < 0.01) in TTC9A +/+ OVX-EB group as compared to the TTC9A +/+ OVX-CTRL animals, indicating an EB-induced anxiolytic effect in the home-cage emergence test. There was also a significant decrease of escape latency in the TTC9A −/− OVX-CTRL mice compared to TTC9A +/+ OVX-CTRL mice (p < 0.01). Interestingly, we found an increase of escape latency in the home-cage emergence test in TTC9A −/− OVX-EB treated mice compared to (B) novel-cage exploration test; (C,D) rat-cage exploration test); memory function ((E,F) object recognition test); hedonia ((G) sucrose preference test); and behavioral despair ((H) forced swimming test) in male mice. Note, despite TTC9A −/− increased rearing behavior (marginal effect; p = 0.041) in the novel-cage exploration test, no significant differences were found in the overall anxiety, memory function, hedonia, and behavioral despair measures. The data were presented as mean ± S.E.M. Indication: *Represents p < 0.05, ***Represents p < 0.005. in the open-field test with increased immobility duration and time-spent in the corner zone (C,F); as well as increased immobility and reduced rearing frequency in the novel-cage exploration test (G,H). Interestingly, there was an overall reduction on social behavior in the social interaction test (I), and no significant change was found on the sucrose preference test which measured the level of hedonia (J). The data were presented as mean ± S.E.M. Indication: *Represents p < 0.05, **Represents p < 0.01, ***Represents p < 0.005. Scientific RepoRts | 6:37568 | DOI: 10.1038/srep37568 TTC9A −/− OVX-CTRL and TTC9A +/+ OVX-EB animals (p < 0.05). In tail-suspension test, two-way ANOVA revealed a significant difference between treatments (F (1,23) = 6.61, p = 0.017), but no difference between genotypes or interactions (F (1,23) = 0.96 and F (1,23) = 3.90, respectively, all p = n.s.). Post-hoc analysis showed a significant increase of tail-suspension immobility in the TTC9A −/− OVX-EB treated mice as compared to the TTC9A −/− OVX-CTRL (p < 0.01) and TTC9A +/+ OVX-EB (p < 0.05) animals. This suggests that TTC9A −/− mice are susceptible to estrogen-induced anxiety-like behaviors (Fig. 5E,F).
Effect of TTC9A gene KO on the expression of TPH and 5-HTT mRNA in TTC9A −/− mice. It has been reported that EB treatment increases the expression of TPH1 and TPH2 in the dorsal raphe nucleus, which In the TTC9A +/+ animals, EB treatment reversed the anxiety-like behavior in the home-cage emergence test, while no difference was observed in the immobility measures of tail-suspension test. The data were presented as mean ± S.E.M. Indication: # Significant difference from the respective group of OVX-CTRL treated mice; *Significant difference from the TTC9A +/+ OVX-CTRL mice; and † Significant difference from the TTC9A −/− OVX-CTRL treated mice; (p < 0.05).
Scientific RepoRts | 6:37568 | DOI: 10.1038/srep37568 in turn regulates serotonin synthesis 22,23 . As serotonergic system is fundamentally important in anxiety disorders 24,25 , we investigated if the changes in mRNA expression levels of TPH1, TPH2, and 5-HTT in the vmPFC and the DRN of TTC9A −/− mice would be the reason for increased anxiety-like behaviors.

Discussion
While TTC9A mRNA expression was highly detected in the brain, its function with estrogen in the regulation of mood and anxiety behavior has not been reported. The findings from the present study demonstrated that TTC9A −/− induced anxiogenic behavior with significant increase of escape latency in the home-cage emergence test and increased immobility in the open-field test in the female, but not in the male, mice. Although we observed comparable escape latencies for male TTC9A +/+ , male TTC9A −/− , and female TTC9A −/− in the home-cage emergence test, we cannot interpret that both the TTC9A +/+ and TTC9A −/− male mice were highly anxious as other behavioral tests showed no significant difference in their exploratory and anxiety-related behaviors. Furthermore, most rodent models demonstrate lower anxiety-like behavior in females, compared to males in the light-dark box test 26 , elevated plus-maze 27 , and elevated t-maze tests 28 . Given that there were no major differences between the TTC9A +/+ and TTC9A −/− male mice, and we argue that TTC9A −/− induced anxiogenic behavior is specific to female mice.
Our observation of anxiogenic behavior in female TTC9A −/− mice was further supported by a remarkable decrease of exploratory rearing behavior in both the novel-cage exploration and rat-cage behavior tests, as well as increase of immobility behavior in the open-field and rat-cage behavior tests. In addition, there was also a significant decrease in the social behavior of the TTC9A −/− female mice. It is of note that while we found no remarkable difference in the time-spent of animals in the center zone of the open-field, we showed a significant increase of immobility and time-spent in the corner zone in open-field for TTC9A −/− female mice, which would indicate anxiety-like behavior. A possible explanation for this phenomenon could be due to a 'minimal' or 'low' level of anxiety response induced in the TTC9A −/− female mice which could possibly result in no significant changes in the open-field center zone duration. This observation is strongly supported by our previous studies in which we found no significant difference in the center zone of the open-field after serotonergic drugs treatment (e.g. buspirone, escitalopram) in a low-anxiety condition as compared to the high-anxiety condition 25,29 . Regardless, our findings provide evidence that TTC9A gene has an important role in modulating the anxiety-like behavior.
Reports have suggested that in peri-and post-menopausal period, the incidence of mood and anxiety disorder is relatively higher in women due to estrogen deficiency 3 . Estrogen replacement therapy has shown a beneficial effect in the peri-and post-menopause women in clinical studies including randomized double-blind trials 20,30 . Similarly, anxiety-and depressive-like behaviors can also be found in animal models with OVX, and this effect can be reversed by administration of EB regimen [31][32][33] . This behavioral alteration in female animals with reduced estrogen after OVX is generally considered a well-validated animal model for human post-menopausal mood-related disorder 34,35 . Consistent with the reported studies, EB treatment reduced the escape latency in TTC9A +/+ OVX mice. However, we observed no depressive-like behavior in the tail-suspension test in the TTC9A +/+ OVX-CTRL mice. This could be due to the short duration of EB treatment. Reports have suggested that some effects on behavior can only be observed several months after ovariectomy 36,37 .
Since our previous data suggest that TTC9A negatively regulates estrogen action 15,21 , we investigated the effect of estrogen on the behaviors of TTC9A −/− mice following OVX. The anxiety-like behavior in TTC9A −/− mice following ovariectomy was generally similar to the TTC9A +/+ mice except for the decrease of the escape latency in TTC9A −/− mice in the home-cage emergence test. This suggests some functional interaction between TTC9A and estrogen, which is supported by the evidence, that TTC9A +/+ mice exhibited decrease in anxiety in the home-cage emergence test in response to EB, and that TTC9A −/− mice showed increased anxiety in both the home-cage emergence test and in the tail-suspension test after EB treatment. This implies that TTC9A modulates estrogen action on anxiety and depressive behaviors such that its absence leads to estrogen to exert an anxiogenic effect.
The serotonergic system is important in mood and anxiety disorders and effective antidepressant therapy commonly uses drugs that increase the extracellular concentration of 5-HT 6,38 . Serotonergic neurons are found in the DRN and project their fibers to the forebrain regions including the prefrontal cortex and limbic structures 39,40 . TPH1 and TPH2 are rate limiting enzymes for 5-HT synthesis and it has been reported that estrogen increased the expression of TPH1 and TPH2 in the DRN and this regulation is mediated by ERβ through an estrogen response element in the 5′ untranslated region 22,41 . Our results show a trend of increased TPH1, TPH2 and 5-HTT expression in DRN in response to EB but the observation is not statistically significant. TTC9A gene deletion significantly increased the expression of TPH2 and 5-HTT in the DRN in response to EB. EB treatment also inhibited the expression of TPH2 and 5-HTT in vmPFC of TTC9A −/− mice and this is not observed in TTC9A +/+ mice. In addition, TTC9A −/− OVX mice expressed higher levels of TPH1, TPH2 and 5-HTT in vmPFC than TTC9A +/+ OVX mice. Together, TTC9A appears to influence estrogen action on gene expression such that in its absence, estrogen is more active in repressing (in the case on vmPFC) or activating (in the case on DRN) the genes related to 5-HT synthesis or transmission.
Concerning the 5-HT mediated regulation of anxiety, there are several discrepancies of its gene expression within the DRN and vmPFC for mood and anxiety-like behaviors. However, our results corroborate previous findings that showed increased 5-HTT expression in the DRN 42 , and reduced TPH synthesis and 5-HTT function in the vmPFC in animal models of depression 8,9 . Furthermore, the increase of TPH 43,44 , with lower levels of 5-HT and its metabolite (5-hydroxyindoleacetic acid) in the DRN (or brainstem), as well as decreased 5-HTT binding in the ventral PFC, have also been reported in patients with major depression and suicidal behaviors [44][45][46] . The plausible explanation for an increase of 5-HTT and TPH expression in the DRN of anxiogenic mice could be that it resulted from the upregulation of homeostatic responses to the deficiency in 5-HT release, dysfunctional receptor activation, and/or hypoactive 5-HT synthesizing enzymes 43,44,46 . Nonetheless, the involvement of serotonergic system for anxiety-like behaviors is complex due to different existing neuronal subpopulations in the DRN and their multi-synaptic interactions. Nevertheless, the effects of TTC9A −/− on estrogen regulation of TPH1, TPH2 and 5-HTT which influences anxiety-like behaviors will remain an interesting topic for future investigation.
In conclusion, our study showed for the first time that absence of TTC9A gene induced anxiety-like behavior in mice, and this response was substantially abolished after ovariectomy. We confirmed that TTC9A −/− OVX mice exhibited significantly higher level of anxiety and behavioral despair in response to EB than TTC9A +/+ OVX mice. In addition, TTC9A −/− heightened EB inhibition of TPH1, TPH2 and 5-HTT in vmPFC and strengthened EB induction of TPH2 and 5-HTT in DRN. Taken together, these findings provide novel insight into the role of TTC9A on estrogen modulation in the mood and anxiety disorders.

Methods
Subjects. 5 month-old mice were socially housed in the individual-ventilated cages on corncob bedding under controlled temperature (about 24-26 °C), humidity (60-70%), and 12/12-h light/dark cycle (lights off at 19:00 h). Food and water were available ad libitum. All experimental procedures and methods were performed in accordance with the relevant guidelines and regulations, and approval from the Institutional Animal Care and Use Committee (IACUC), Nanyang Technological University, Singapore (Reference Number: ARF-SBS/NIE-A 0169 AZ).
Generation of TTC9A knockout mice. The detailed description of generation of the TTC9A −/− mice has been previously described 15,21 . In brief, the TTC9A targeting vector was generated by replacing the TTC9A exon 1 with a loxp flanked neomycin cassette, and electroporated into R1 mouse ES cells. The TTC9A heterozygous ES clones were microinjected into 8 cell stage mouse embryos isolated from C57BL/6J. The TTC9A chimeras were crossed to C57BL6/6J mice to generate TTC9A heterozygous mice. These TTC9A heterozygous mice were crossed to obtain the TTC9A homozygous mice.
Experimental design. Two systematic behavioral studies were used to evaluate the mood -and anxiety-related behaviors, and memory function in both the male and female TTC9A −/− and TTC9A +/+ mice. In behavioral study 1, home-cage emergence test, novel-cage exploration test, and rat-cage behavior test were used to measure the anxiety and rearing behaviors, object recognition test was applied to measure the short-and long-term memory functions; sucrose preference test was used to measure hedonic activity; and finally forced swimming test was used to assess behavioral despair ( Fig. 2A). Throughout the behavioral experiments, estrous cycle was determined by vaginal swab cytology and only females in diestrus stage were included in the analysis.
Drug administration. 17-β -Estradiol benzoate (EB; Sigma-Aldrich, Missouri, USA) was dissolved in the sesame oil with 0.25% benzyl alcohol and injected subcutaneously at a dosage 10 μ g/kg in both the TTC9A −/− and TTC9A +/+ female mice. The selection of this dosage was based upon previous studies of dose-dependent effects of EB treatment on anxiety level in OVX rodents which producing physiological plasma levels of estradiol and anxiolytic behavior 48 h later 47,48 . The behavioral testing was conducted in accordance with the order of time after EB injection.
In behavioral study 2, a new set of female mice were subject to OVX procedure followed by treatment of either control (CTRL; injection of sesame oil with 0.25% benzyl alcohol) or estradiol benzoate (EB). Firstly, mice were assessed for their anxiety and rearing behaviors using the open-field test, novel-cage exploration test, social interaction test, and sucrose preference test. Ten days after recovery from OVX surgery, all animals were injected with either CTRL or EB, and subjected to behavioral testing on day 20 and 21. In this paradigm, mice were assessed for anxiety-related behaviors in the elevated plus-maze (24 h), cylinder test (26 h), and home-cage emergence test (48 h), and finally a tail-suspension test (52 h) was used to measure behavioral despair. Behavioral testing conducted 24 h and 48 h after EB injection were based on previous studies demonstrating its promising effects on the behavioral responses and EB-induced synaptic plasticity changes [47][48][49] . The experimental design is presented in a schematic representation in Fig. 4A.

Surgical procedures.
A detailed description of the surgical procedure has been previously reported 50 . In brief, the animals were ovariectomized and anesthetized using a combination of ketamine (75 mg/kg) and xylazine (5 mg/kg) injected subcutaneously. After surgery, animals had a 10-day recovery period.
Behavioral testing and evaluation. All testing and analysis of animal behaviors were performed by researchers who were blind to the treatments and condition of testing. The tests for behavioral study 1 were conducted during 20:00-00:00, while the tests for behavioral study 2 after OVX were performed in the time order of 24 h, 26 h, 48 h, 52 h after injections during 20:00-01:00. Mice were given at least 30 min acclimatization in the room before testing. After each test the arena or set-up was cleaned with 70% ethanol and dried before the next mouse was tested.
Home-cage emergence test. The procedure of home-cage emergence test was performed as previously described with minor modifications 51 . In brief, a novel mouse home-cage was placed on a platform and the lid of the home-cage was removed. A grid was placed over the edge of the home-cage in order to make it easier for the mouse to leave the home-cage. The experimenters measured the escape latency (time taken for the mouse to Scientific RepoRts | 6:37568 | DOI: 10.1038/srep37568 climb out of its cage onto the grid). If the mouse did not escape from its home-cage within 10 min, the session was ended, and the mouse was given a score of 600 s.
Novel-cage exploration test. The standard novel-cage exploration test was performed as previously described 52 . Mice were introduced into a standard plastic cage filled with fresh corncob bedding. The number of exploratory rearing was counted during a 5 min period by offline recorded video observation.
Rat-cage behavior test. The test was conducted in a standard rat cage (25 × 35 × 25 cm), with an open top and a dark floor. The behavior of each mouse was video recorded. The duration of immobility and frequency of rearing were analyzed for the total 5 min observation of the mice spent in the arena.
Object recognition test. The object recognition test was performed as previously described with minor modifications 53 . The test consists of a habituation phase, acquisition phase, and finally the test phases for short-term memory (1 h after acquisition phase), and long-term memory (24 h after short-term memory test) functions. In the habituation phase, mice explored an open-field arena (40 × 40 × 40 cm 3 ) for 10 min. Next day, during acquisition phase, mice were presented with two identical objects (object A + object A) in the same open-field environment for a total duration of 3 min. After 1 h, mice were tested in the same arena with one familiar object and one novel object (object A + object B) in the short-term memory phase. On the next day (24 h later), the familiar object was replaced by another novel object in the long-term memory test phase (object C + object B). The positions of objects were counter-balanced across subjects and testing phase. The duration spent on exploration in each object was video recorded. The total duration of exploration time was considered as the sum of time spent at both objects. Discrimination index was calculated as follows, the exploration time (T) of the novel object minus the familiar object, then dividing this value by the total amount of exploration of the novel and familiar objects [DI = (T N − T F )/(T N + T F )] 54 .
Sucrose preference test. The sucrose preference test was performed as previously described with minor modifications 51,55 . One day prior to the test, animals were exposed to drink 1% sucrose solution and water for 1 h, and subsequently followed by food and water deprivation for 14 h. After 14 h fasting, mice were again given a choice between two bottles of 1% sucrose solution and water, and tested for the total duration of 1 h. To prevent side preference for water or sucrose consumption, the two bottles were counter-balanced during the pre-test and test sessions. The sucrose and water intake were calculated from the total amount sucrose solution or water consumed (ml) by weighing the bottle pre-and post-testing session. The relative sucrose consumption was corrected for body weight of the animals (ml/kg). The sucrose preference was calculated as a percentage of the sucrose consumption with respect to the total amount of liquid intake. The animals were tested individually during the sucrose consumption, and they were placed back into their respective group-housing cage immediately after the test.
Forced swimming test. The testing was carried out using a transparent Plexiglas cylinder (height 30 cm × diameter 10 cm). The cylinder was filled with tap water (25 ± 1 °C) to a depth of 10 cm. The testing procedure was performed over two consecutive days as previously described 51,56 . In a pre-test session, each mouse was placed in the water for 15 min. The following day, the mouse was again tested in the cylinder containing water for 10 min and the behavior was recorded by a digital camera. The duration of the following behaviors was measured: "immobility" (no movements or small and infrequent movements performed solely to maintain the nose above the water), "swimming" (active swimming with the forepaws), and "climbing" (scratching of cylinder walls using both forepaws and hind paws).
Open-field test. The test was conducted in an enclosed square, made of white plastic box (40 × 40 × 40 cm 3 ). Behavioral testing started after the mouse had been individually placed in the middle of the arena. A trial was stopped after 5 min and the behavior of each mouse was video recorded. The duration of immobility and time-spent in the different zones (center, periphery and corner) of the open-field were measured and analyzed as previously described with minor modifications 29,57 .
Social interaction test. The social interaction behavior was conducted in a Plexiglas rectangle box (25 × 35 × 25 cm) with a wire net (30 × 40 cm) as a separator between the two animals. Animals were allowed to freely interact in the arena and the time-spent interacting was video recorded 58 .
Elevated plus-maze test. The apparatus consisted of two open-arms (40 × 5 cm) and two closed-arms (40 × 5 × 10 cm) that extended from a common central platform (5 × 5 cm). The apparatus was elevated at a height of 50 cm from the floor. The mouse was placed in the central platform and individually tested for 5 min. The duration of time-spent and frequency of entry into the open-and closed-arms were measured.
Cylinder test. All mice were tested in the transparent Plexiglas cylinder (height 30 cm × diameter 10 cm) for a total of 5 min duration. The behavior of each mouse was video recorded and the frequency of rearing was analyzed.
Tail-suspension test. All mice were subjected to the tail-suspension procedure as previously described 59 for a total duration of 5 min. The mice were suspended by hanging their tails with adhesive tape in an individual compartment of 40 cm from the floor. The duration of immobility was measured when mice were completely motionless.
Statistical analysis. Statistical analysis were performed using GraphPad Prism 7. All data are presented as mean ± S.E.M. Tests done in behavior study 1 were analyzed using independent sample t-tests. A repeated-measures analysis was used to measure the within-and between-subject components in the object recognition test. Tests done before EB injection in behavior study 2 were analyzed using independent sample t-tests. Tests done after EB injection were analyzed using a two-way ANOVA with LSD post-hoc test. Real-time PCR results were analyzed using two-way ANOVA with LSD post-hoc. All p-values < 0.05 were considered significant. Outliers were identified and removed using ROUT method.