Effects of Hint1 deficiency on emotional‐like behaviors in mice under chronic immobilization stress

Abstract Background Histidine triad nucleotide‐binding protein 1 (HINT1) is regarded as a haplo‐insufficient tumor suppressor and is closely associated with diverse neuropsychiatric diseases. Moreover, HINT1 is related to gender‐specific acute behavior changes in schizophrenia and in response to nicotine. Stress has a range of molecular effects in emotional disorders, which can cause a reduction in brain‐derived neurotrophic factor (BDNF) expression in the hippocampus, resulting in hippocampal atrophy and neuronal cell loss. Methods This study examined the role of HINT1 deficiency in anxiety‐related and depression‐like behaviors and BDNF expression in the hippocampus under chronic immobilization stress, and investigated whether the sex‐specific and haplo‐insufficient effects exist in emotional‐like behaviors under the same condition. Results In a battery of behavior tests, the results of the control group, not exposed to stress, showed that knockout (KO) and heterozygosity (HT) of Hint1 had anxiolytic‐like and antidepression‐like effects on the male and female mice. However, both male and female Hint1‐KO mice showed elevated anxiety‐related and antidepression‐like behavior under chronic immobilization stress; moreover, both male and female Hint1‐HT mice displayed elevated anxiety‐related behavior and increased depression‐like behavior under chronic immobilization stress. There were no significant differences in general locomotor activity between Hint1‐KO and ‐HT mice and their wild‐type (WT) littermates. Hint1‐KO mice under basal and chronic immobilization stress conditions expressed more BDNF in the hippocampus than did Hint1‐HT and WT mice; overall, there were no significant sex differences in emotional‐like behaviors of Hint1‐KO and ‐HT mice. Additionally, Hint1‐HT mice showed haplo‐insufficient effects on emotional‐like behaviors under basic conditions, rather than under chronic immobilization stress. Conclusions Both male and female HINT 1 KO and HT mice had a trend of anxiolytic‐like behavior and antidepression‐like behavior at control group. However, both male and female HINT1 KO mice showed elevated anxiety‐related and antidepression‐like behavior under chronic immobilization stress; moreover, both male and female HINT1 HT mice displayed elevated anxiety‐related behavior and increased depression‐like behavior under chronic immobilization stress.


| INTRODUCTION
Mood and anxiety disorders are among the most common mental health conditions, resulting in functional disability, and major depressive disorder (MDD), with a lifetime risk of 20%-25% in women and 7%-12% in men has already created an enormous financial burden for modern human societies. Approximately 60% of people with MDD have comorbid anxiety disorder (Belmaker & Agam, 2008;Krishnan & Nestler, 2008). Comorbid anxiety and depression generally involve more severe symptoms, a continuous course of illness, and greater functional disability than either anxiety or depression alone (Funke et al., 2016;Moscati, Flint, & Kendler, 2016). Although these psychiatric disorders involve complex interactions between genetic and environmental factors, the etiology and the underlying molecular mechanism are still largely unclear (Finn, Rutledge-Gorman, & Crabbe, 2003;Gispen-de Wied & Jansen, 2002). Pharmacological agents used as antidepressants, such as selective serotonin reuptake inhibitors, are effective treatments, but up to 20% of patients completely fail to respond to standard interventions, and nearly 60% cannot achieve an adequate response (Block & Nemeroff, 2014). Moreover, a large proportion of patients have a poor prognosis; thus, it is urgent to explore the nosogenesis of anxiety and depression further at the molecular and cellular level, in order to develop more effective treatments.
Early life stress events are closely related to depression, and stressful events seem to play an important role in the onset, progression, and treatment effects of MDD and anxiety disorder (Fone & Porkess, 2008;Kim et al., 2011). Stress activates neuroendocrine systems and affects the activity of the relative brain area, causing physiological and behavioral responses (Chrousos, 2009;Ulrich-Lai & Herman, 2009).
Interestingly, in recent studies, accumulating evidence from human postmortem analysis, research on mouse behavior, and anatomical studies has demonstrated that HINT1 is associated with diverse neuropsychiatric diseases. In a microarray analysis, HINT1 mRNA expression was lower in the dorsolateral prefrontal cortex of patients who suffer from schizophrenic and bipolar disorder than healthy controls (Elashoff et al., 2007;Vawter et al., 2001Vawter et al., , 2004. HINT1 is also associated with abuse of drugs, such as nicotine (Jackson et al., 2011).
Studies on the distribution of HINT1 in the brain have shown that it is expressed in neurons of the cerebral cortex and limbic system, which are anatomically related to mood disorders (Baumann & Bogerts, 1999;Benes & Berretta, 2001;Liu et al., 2008;Zhang & Reynolds, 2002).
Knockout of Hint1 in mice has antidepressant and anxiolytic-like effects and elevated corticosterone levels in plasma (Barbier & Wang, 2009). However, Varadarajulu and colleagues have reported that Hint1-KO mice showed increased anxiety-like behavior (Varadarajulu et al., 2011). Additionally, the correlations between schizophrenia and HINT1 are sex specific (Chen et al., 2008;Jackson et al., 2012;Vawter et al., 2004), and nicotine-mediated acute behavior changes also show different effects in males and females (Jackson et al., 2012).
Nonetheless, it remains unknown whether HINT1 involved in the regulation of anxious and depressed behaviors under chronic immobilization stress is sex specific or haplo insufficient.
We speculated that HINT1 may be related to BDNF expression in the limbic system. However, it is not yet known whether HINT1 knockout can affect BDNF expression in the hippocampus under chronic stress conditions. Thus, we considered whether Hint1 deficiency would have effects on the emotional-like behaviors of mice, as well as on the expression of BDNF in the hippocampus under chronic immobilization K E Y W O R D S brain-derived neurotrophic factor, chronic immobilization stress, haplo-insufficient effect, hippocampus, histidine triad nucleotide-binding protein 1, sex specific stress (CIS), and whether such effects would be influenced by gender and gene dosage (haplo-insufficient effects). We used male and female mice that were homozygous mutant (KO), heterozygous mutant (HT), or wild type (WT) for Hint1 in a CIS model. We then used a battery of behavioral tests to examine changes in emotional-like behavior, and also explored BDNF protein expression in the hippocampus of these mice.

| Animal
The Hint1-KO mouse model was previously generated in the Columbia University Health Science animal facility by Haiyang li and Bernard Weinstein as described (Li et al., 2006). Mice with three genotypes (homozygous mutant, heterozygous mutant, and their WT littermates) were the offspring of heterozygous breeding. The genetic background of the Hint1-KO mice, -HT mice, and their WT littermates was initially 96% 129SvJ (Su et al., 2003). Mice were housed 4-6/cage and allowed free access to water and food under standard laboratory conditions, in a temperature-controlled (temperature: 23°C) environment, with a 12-hr light/dark cycle, with lights on 8:00 a.m. Male and female mice of 10-12 weeks, of the three genotypes, including Hint1-KO mice, -HT mice, and their WT littermates were used in this study.
We observed that Hint1-KO and -HT mice display normal physiological growth and development and inconspicuous behaviors under home-cage conditions. There were no significant differences in body weight in Hint1-KO, -HT, or WT mice, which were weighed weekly from week 4 to week 12 (data are not shown).
All experimental protocols were approved by the Animal Care and Use Committee of Xi , an Jiaotong University.

| Genotype identification
To identify genotypes of all types of progeny, genomic DNA was extracted from the tails of the mice, at the time of weaning (about 4 weeks), and was analyzed by PCR as described (Li et al., 2006). The forward primer 5′-GCC TGA AGA ACG AGA TCA GC-3′ and the reverse primer 5′-CGC CCC AGT TAG TTA GT CAG-3′ generated a 285-bp product from the mutant allele, while the forward primer 5′-GCC CCC TGT AAA GTG GAG AC-3′ and the reverse primer 5′-CGC CCC AGT TAG TTA GT CAG-3′ produced a 339-bp product from the wild-type allele. Heterozygous mice produced both the 285-bp and 339-bp product.

| Chronic immobilization stress
In order to create a reproducible animal model of chronic stressinduced anxiety-like and depression-like behaviors, mice of six groups (3 genotypes × 2 genders) were subjected to CIS, and were restrained in a plastic cylinder with a height of 12 cm and diameter of 3 cm for 1 hr in the afternoon once daily, for 20 days, while the six negative control groups (3 genotypes × 2 genders) were handled in the same manner as the stressed animals, but without exposure to CIS. The 12 groups consisted of six experimental groups (3 genotypes × 2 genders) and six negative control groups without any stress (3 genotypes × 2 genders). There were 7-9 mice for each group; these animals were exposed to a battery of behavioral tests, starting on the 15th day of immobilization stress.

| Behavioral testing
In order to investigate the effects of Hint1knockout on emotional-like behaviors in mice under CIS, adult male and female Hint1-KO mice, -HT mice, and their WT littermates were exposed to a number of paradigms estimating locomotor activity, anxiety-related behavior, and depressionlike behavior (Cryan & Holmes, 2005). The animals were brought to the testing room at least 30 min before each experiment. There were 12 groups, as described above, for a series of behavior tests. For each group, behavioral tests were performed in the following order: Open-field test (OFT), Elevated plus maze (EPM), Tail suspension test (TST), and Forced swimming test (FST). All the behavioral tests were conducted between 9 a.m. and 12 a.m., and were sorted by stress intensity from weak to strong, with a 48-hr interval among the adjacent tests, and with a 12-hr interval after the last immobilization. The results of all behavior tests were recorded and the relevant data were automatically collected using a video-tracking system (SMART, Panlab SL, Barcelona, Spain).

| Open-field test
The OFT was used to measure locomotor activity and explorative behavior, by assessing the rodents' drive to explore unfamiliar environments. Briefly, mice were placed individually into a 45 × 45 × 45 cm open-field chamber and tested for 1 hr. The data were recorded by SMART. The open field was divided into a central field (center, 20 × 20 cm) and a peripheral field. The total distance travelled in 1 hr was measured. After each trial, the chamber was thoroughly cleaned with water containing detergents, followed by 80% ethanol, and dried with a tissue to avoid olfactory influences on subsequent tests.

| Elevated plus maze
The EPM utilized a plus-shaped apparatus to detect anxiety-related behavior and was performed as described elsewhere (Lister, 1987).

| Tail suspension test
Mice were suspended by their tails to the edge of a shelf 80 cm above the ground. The tail was secured to the shelf by adhesive tape placed approximately 1 cm from the tip of the tail. Immobility time was recorded using SMART during a 6-min session.

| Forced swimming test
Depression-like and stress-coping behavior was measured by means of the FST (Cryan & Holmes, 2005;Touma et al., 2008). Briefly, mice were placed in a Plexiglas cylinder (25 cm height × 15 cm diameter) containing 15 cm of water (22 ± 1°C) for 6 min. The behavior of mice and the immobility time during the 6-min session were recorded using a video-tracking system (SMART).

| Protein extraction and Western blot
Total proteins in mice hippocampi were extracted using RIPA Lysis Buffer (Beyotime, Guangzhou, China) and the concentration of proteins was determined using the BCA protein assay kit (Pierce, Rockford, USA) according to the manufacturer's instruction. Western blotting assays were performed as previously described (Zhang, Fang, & Wang, 2015). Primary antibodies were anti-BDNF (dilution 1:1,000; #sc-546, Santa Cruz, Dallas, TX, USA) and antiβ-actin (dilution 1:1,000; #sc-8432, Santa Cruz). The protein expression was visualized using enhanced chemiluminescence (Millipore, Billerica, MA, USA). Images were captured using the ChemiDoc XRS imaging system (Bio-Rad, Hercules, CA, USA) and Quantity One image software was used for the densitometric analysis of each band. β-actin was used as an internal loading control.

| Statistical analysis
All data were expressed as mean ± standard error of mean (SEM), and we used SPSS18.0 software for data processing and analysis. For data of OFT of distance moved every 10 min, repeated-measures ANOVA with between-subjects factors was used to analyze within-subject differences (control vs. CIS) and between-subject differences (genotype), statistical differences in other behavioral studies were performed using three-way analysis of variance (ANOVA) with treatment, sex, and genotype as the between-subject factors, and significant results were further analyzed using a post hoc multiple comparison (Bonferroni post hoc test). Statistical significance was set at p < .05.
Similarly, there were no statistically significant differences with the time spend in the center zoom (%) in the first 10 min and distance moved every 10 min. Therefore, we found no significant differences in Hint1 deficiency and CIS on locomotor activity in OFT (Figure 1).

| Effects of Hint1 deficiency and CIS on the anxiety-related behavior
The EPM test was applied to detect anxiety-related behavior in mice with complete or partial loss of Hint1 under basal or CIS con-
In the post hoc multiple comparison analysis, the immobility time
These results indicated that there was no significant gender differ- There were no significant main effects of gender (F 1,72 = 0.365, p = .547), but significant main effects of genotype (F 2,72 = 19.636, p < .0001) and significant main effects of intervention (F 1,72 = 8.839, p = .004) and significant genotype × intervention interactions (F 2,72 = 5.835, p = .004). In addition, both male and female Hint1-KO mice exhibited a slightly decreased immobility time under basal conditions, and the immobility time was significantly decreased under CIS. Hint1-HT female mice had a decreased immobility time under basal conditions, but displayed an increased immobility time under CIS. Values represent mean ± SEM (n = 7-9 for each group). *p < .05; **p < .01. Control: basal conditions, with no stress; CIS, chronic immobilization stress. KO, Hint1-knockout mice; HT, Hint1-heterozygous mice; WT, wild type with their WT littermates under CIS. Both male and female Hint1-HT mice showed an increased trend for depression-like behavior under CIS.

| BDNF expression in hippocampus of Hint1-KO and -HT mice and their littermates
After finishing of the last behavior tests (FST), mice were sacrificed and total proteins were extracted from mouse hippocampi for Western blotting analysis of BDNF expression. The statistical analysis of BDNF expression showed that there were significant main effects of gender (F 1,72 = 5.394, p = .022), significant main effects of genotype (F 2,72 = 10.615, p < .0001), significant main effects of intervention (F 1,72 = 7.965, p = .006), and no significant gender × genotype × intervention interactions (F 2,72 = 0.062, p = .940).
Our results ( Figure 5) showed that Hint1-KO mice, under both basic and CIS conditions, exhibited higher expression of BDNF in the hippocampus than Hint1-HT mice and their WT littermates, which correlated with the reduced depression-like behavior of Hint1-KO.
Hint1-HT mice also showed a higher level of BDNF expression than their WT littermates, but a lower level than that of Hint1-KO mice under basic as well as under CIS conditions.  (Barbier & Wang, 2009). Previous explanations for these contradictory findings included different test settings, such as light intensities, the size of the compartments, the age and time of testing, etc. (Varadarajulu et al., 2011). Our findings of Hint1-KO mice under basal conditions, with no stress, were in accordance with the findings of Wang. However, under CIS conditions, our results supported those of Varadarajulu. Accumulating evidence has shown that different stressful events play an essential role in the onset and progression of many psychiatric disorders, such as anxiety disorders and MDD (Charney & Manji, 2004;Lupien et al., 2009). Thus, these findings indicate that the degree of stress intensity may exert pivotal roles in mood behaviors of Hint1-KO mice.

This study investigated whether
Early studies have demonstrated that HINT1 is a haplo-insufficient tumor suppressor (Li et al., 2006). Whether haplo-insufficient effects are also present in emotional-like behavior remains unclear. Under basal conditions, both full and partial deletion of Hint1 in mice tended to have anxiolytic-like and antidepressant-like effects on behavior.
After CIS, Hint1-HT mice showed elevated anxiety-related behavior and increased depression-like behavior, while Hint1-KO mice displayed elevated anxiety-related behavior and reduced depression-like behavior. Therefore, our results showed that haplo-insufficient effects may exist in the emotional-like behaviors of Hint1-HT mice under basal conditions, but not under CIS conditions. Moreover, we propose that the antidepressant-like effects of knockout of Hint1 in mice represent the genotype rather than the stress effect. The HPA axis commonly dysfunctions under stress conditions, and is involved in mood disorders (Dedovic, Duchesne, Andrews, Engert, & Pruessner, 2009;Jo, Zhang, Emrich, & Dietrich, 2015;Schule, Baghai, Eser, & Rupprecht, 2009). The finding of a trend toward anxiolytic-like effects in the elevated anxiety-like behavior in Hint1-KO mice after CIS indicates that CIS plays an important role in changes in anxiety-related behaviors, and that the underlying mechanism may involve activation of the HPA axis by stress. However, after CIS, the emotional-like behaviors of Hint1-HT mice changed markedly from a trend toward anxiolytic-like effects and antidepressant-like effects, to increased anxiety-related and depression-like behavior, implying that the underlying mechanism may be more complicated. Firstly, these results indicate that Hint1-HT mice are more sensitive to a CIS stimulus for totally reverting emotional-like behaviors from a tendency toward anxiolytic-like effects and antidepressive-like effects, to the elevated anxiety-related and increased depression-like behavior. However, Hint1-HT mice had a higher expression of BDNF than their WT littermates, while it was lower than that of Hint1-KO mice, not only under the basic condition but also under CIS. This indicated that the expression of BDNF in the hippocampus of Hint1-HT mice may represent the effect of the gene (Hint1) and that other underlying molecular alterations under CIS may participate in these emotional-like behavior changes. Secondly, Hint1-HT mice in vivo still retain a certain amount of expression of HINT1 protein, thus, there may have been a physiological compensatory effect in the process of mouse growth and development.
Furthermore, it remains unclear why reduced expression of HINT1 protein in Hint1-HT mice in interaction with a stress stimulus caused changes in emotional-like behavior, and this requires further study.
There are gender differences between men and women in terms of the occurrence, symptoms, and therapeutic responses of stress-related mental disorders, such as MDD, which is twice as common in females than in males (Marcus et al., 2005). However, there are no gender differences in the rates of MDD at the puberty (Dalla, Pitychoutis, Kokras, & Papadopoulou-Daifoti, 2010), which suggests that age may exert an important effect in the mental disorders with a gender difference.
However, in our studies, we did not find significant sex differences overall in the emotional-like responses to HINT1 KO and HT mice.
The age of mice may exert important effects on emotional-like behaviors that do not show a sex difference. Moreover, the neurobiology of mental disorders, such as anxiety and depression itself, is still not fully understood. Furthermore, there is still lack of relative basic research of developing animal models and drug treatments that are sensitive to sex differences. It is necessary to further investigate the mechanism of HINT1 associations with gender differences in mental disorders.
Stress can activate a large array of molecular effects in the brain, especially in the hippocampus, which then generate functional, structural, molecular, and behavior alterations (McEwen, 2001;Sapolsky, 2003).
Current studies have implicated that depression is not only viewed as abnormalities in the monoaminergic systems but is also involved in the impairment of neural plasticity, neurotrophic factors, and dendritic complexity in the hippocampus (Duman & Monteggia, 2006), and BDNF gene transcriptional activity can influence the neuronal growth, plasticity, and neurogenesis, which is conducive to the therapeutic response to antidepressants. Our Western blot results showed that Hint1-KO mice not only under basic conditions but also under chronic immobilization stress exhibited higher expression of BDNF in hippocampus compared with Hint1-HT mice and their WT littermates, which F I G U R E 5 Brain-derived neurotrophic factor (BDNF) expression in the hippocampus of Hint1-KO and -HT mice and their WT littermates under basal conditions or under CIS, as determined by Western blotting. Representative images of BDNF expression in hippocampus under basal conditions in male mice (a) and female mice (b); representative image of BDNF expression in the hippocampus under CIS in male mice (c); and female mice (d); densitometric analyses of the relative expression of BDNF in the hippocampus of male mice (e); and female mice (f) by QuantityOne image analysis software. The BDNF results were normalized to β-actin, and the value in the WT group under basal conditions (control group) was set as 100%. The statistical analysis of BDNF expression showed that there were significant main effects of gender (F 1,72 = 5.394, p = .022), significant main effects of genotype (F 2,72 = 10.615, p < .0001), significant main effects of intervention (F 1,72 = 7.965, p = .006), and no significant gender × genotype × intervention interactions (F 2,72 = 0.062, p = .940). In addition, Hint1-KO mice not only in basic condition but also under chronic immobilization stress exhibited higher expression of BDNF in hippocampus than Hint1-HT mice and their WT littermates; Hint1-HT mice had higher levels of expression of BDNF than their WT littermates, but lower than that of Hint1-KO mice, not only under basic condition but also under CIS. Values represent mean ± SEM (n = 7-9 for each group). *p < .05; **p < .01. Control: basal conditions, with no stress; CIS, chronic immobilization stress. KO, Hint1-knockout mice; HT, Hint1-heterozygous mice; WT, wild type agreed with the finding that Hint1-KO mice displayed antidepressionlike behavior, not only under basal conditions but also under CIS.
Early studies have showed that PKC activity in brain lysates of Hint1-KO mice was higher than in their WT littermates, and PKCγ protein isoform expression increased in Hint1-KO mice (Varadarajulu et al., 2011). Moreover, elevated abundances of PKCγ were also detected in Hint1 haplo-insufficiency tissue such as hippocampus, the cortex, etc. . Some studies have revealed that intracellular signal transduction plays an important role in the pathophysiology of depression. One key set of mechanisms implicate kinases, such as protein kinases A (PKA) and C (PKC) (Pandey & Dwivedi, 2005). Protein kinases are critical elements of stimulus-response coupling, which can phosphorylate the transcriptional factor cyclic AMP response element-binding protein (CREB) (Hyman & Nestler, 1996). Thus, we speculated that elevated PKC activity in Hint1 (Barbier et al., 2007). It is possible that such variation in dopamine transmission may exert important role in HINT1-related emotionallike behaviors alterations after CIS; therefore, the molecular mechanism still needs further clarification.