Address correspondence and reprint requests to M. Nakajima, Department of Pharmaceutical Pharmacology, School of Clinical Pharmacy, College of Pharmaceutical Sciences, Matsuyama University, 4-2 Bunkyo-cho, Matsuyama 790-8578, Ehime, Japan. E-mail: email@example.com
The neural crest is a unique structure in vertebrates. Wnt1-cre and Wnt1-GAL4 double transgenic (dTg) mice have been used in a variety of studies concerning neural crest cell lineages in which the Cre/loxP or GAL4/UAS system was applied. Here, we show psychiatric disorder-related behavioral abnormalities and histologic alterations in a neural crest-derived brain region in dTg mice. The dTg mice exhibited increased locomotor activity, decreased social interaction, and impaired short-term spatial memory and nesting behavior. The choline acetyltransferase- and vesicular glutamate transporter 2-immunoreactive habenulointerpeduncular fiber tracts that project from the medial habenular nucleus of the epithalamus to the interpeduncular nucleus of the midbrain tegmentum appeared irregular in the dTg mice. Both the medial habenula nucleus and the interpeduncular nucleus were confirmed to be derived from the neural crest. The findings of this study suggest that neural crest-derived cells have pathogenic roles in the development of psychiatric disorders and that the dTg mouse could be a useful animal model for studying the pathophysiology of mental illness such as autism and schizophrenia. Scientists that use the dTg mice as a cre-transgenic deleter line should be cautious in its possible toxicity, especially if behavioral analyses are to be performed.
In vertebrate development, the neural crest originates at the most dorsal region of the neural tube. The crest is divided into four main regions: the cranial neural crest, the trunk neural crest, the vagal and sacral neural crest, and the cardiac neural crest, which is a subregion of the vagal neural crest. The head is largely the product of the cranial neural crest. Neural crest cells delaminate from the dorsal neural tube and migrate to various regions and differentiate. Derivatives of the neural crest include peripheral nervous system cells, pigment cells, bone and cartilage cells of the face (Gilbert 1949), as well as cranial cells, such as a part of skull, meninges, and brain cells (Jiang et al. 2000and Jiang et al. 2002). Thus, neural crest cells are potentially involved in the pathophysiology of various disorders, such as Alagille syndrome (Humphreys et al. 2012), Fryns syndrome (Alkuraya et al. 2005), and DiGeorge syndrome (Shprintzen et al. 2005).
Although many candidate genes for psychiatric disorders have been identified (Stefansson et al. 2002; Craddock et al. 2005; Won et al. 2012), the pathophysiology of psychiatric disorders remains unknown. Brain developmental disturbance is implicated in psychiatric disorders such as autism and schizophrenia (Rapoport et al. 2005; Geschwind 2008), and recent research suggests an importance of analysis at the level of brain circuitry in addition to molecular pathways (Geschwind and Levitt 2007; Alarcón et al. 2008; Rünker et al. 2011; White and Hilgetag 2011).
Wnt1-cre and Wnt1-GAL4 double transgenic (dTg) mice are a deleter line, in which Cre recombinase and GAL4 transcriptional activator are expressed in early embryonic stages under the control of Wnt1 regulatory sequences (Rowitch et al. 1999). As the Wnt1 promoter/enhancer is activated in the neural crest cell lineage, this mouse line is bred, in most cases, for crest cell lineage-targeted genetic manipulations in Cre/loxP or GAL4/UAS systems (Rowitch et al. 1999; Dietrich et al. 2009; Utikal et al. 2009). In this study, we report behavioral and pathologic abnormalities in the dTg mice. This mutant exhibits psychiatric disorder-related behavioral abnormalities and abnormal cholinergic and glutamatergic fiber tracts in a neural crest-derived brain region. On the basis of these findings, we propose that irregular development of the neural crest-derived cells is involved in the pathogenesis of psychiatric disorders.
Materials and methods
All mice were maintained under a controlled temperature and photoperiod (23°C, 12 h light and 12 h dark) with food and water provided ad libitum. All experimental procedures followed the Guideline for Animals Experimentation prepared by the Animal Care and Use Committee of Matsuyama University.
The dTg mice (Rowitch et al. 1999) were obtained from the Jackson Laboratory (stock number 003829, Bar Harbor, ME, USA). Rosa-cre-LacZ-reporter mice (Rosa26-LacZ floxed mice, Mao et al. 1999) were also obtained from Jackson Laboratory (stock number 003504). The dTg mice were backcrossed at least five generations to C57BL/6J Jms Slc (JAPAN SLC; Shizuoka, Japan) mice and used at the age of 6 to 7 weeks, except the histologic studies were performed using Rosa-cre-LacZ-reporter mice with a C57BL/6 and 129 hybrid background. All control mice consisted of transgene negative littermates.
Mice were housed three to five mice per cage. Cages were made of polycarbonate and were 17 cm deep × 28 cm wide × 12 cm high. The lighting condition during the open field, social interaction, Morris water maze, and elevated-plus maze tests was 2 lux under indirect lighting. A total of 17 litters of mice were used for behavioral testing. Testing was carried out during 10 : 00–16 : 00 for each experiment.
Nest building test
We evaluated nesting behavior as reported previously (Ballard et al. 2002; Miyakawa et al. 2003; Hiroi et al. 2005). Mice were singly housed in individual cages with food and water available ad libitum. Tissue paper made of pulp (Ellemoi, Kamisyoji Co., Ehime, Japan) was folded three times and placed in each cage at 17 : 00–18 : 00 on Day-1 of the nest building test, and nest build-up was assessed the next morning (9 : 00–10 : 00 on Day-2). The nest depth was measured at its highest point in the cage, excluding at the edge regions of the cage. After measurement, the nest was immediately removed, and new folded tissue paper was placed in the cage at 17 : 00–18 : 00 on Day-2. On the following day (Day-3), the depth of the nest build-up was again assessed.
Open field test
Locomotor activity was evaluated using the open field test. Each mouse was placed individually in the center of an open field apparatus (70 × 70 × 50 cm) and the behavior was monitored with a CCD camera connected to a personal computer for 10 min. Total distance traveled and number of entries into the center area were analyzed using ANY-maze Video Tracking System (BrainScienceIdea Co. Ltd., Osaka, Japan). The center area was defined as the area at least 17.5 cm from the walls of the open field apparatus.
Social interaction test
The social interaction test was used to evaluate social behavior in the mice. Mice were individually paired with an unfamiliar age- and sex-matched, and genotype-different Wt and dTg mice in an open field apparatus (70 × 70 × 50 cm) and allowed to explore freely for 10 min. Total number and duration of sniffing and following were counted and measured. As Wt and dTg mice did not exhibit social behaviors such as allogrooming, fighting, and huddling in the open field apparatus, we selected sniffing and following as indices for the social interaction test.
Morris water maze test
The Morris water maze test was used to evaluate spatial learning and memory. The water maze was a circular pool 110 cm in diameter containing a submerged (1 cm below the water surface) platform (10 cm in diameter) to which the mouse could escape. The temperature of the water in the pool was 22°C. Mouse position in the maze was tracked using the ANY-maze Video Tracking System. Each mouse was given five trials daily for 4 days. Each trial was performed with a maximum duration of 60 s with intertrial intervals of at least 5 min. If a mouse did not find the submerged platform, it was guided to the platform and allowed to remain on it for 10 s. On Day-5, the mice were given three trials, and then the submerged platform was removed and a 60-s probe trial was conducted 2 h after the third training trial.
Elevated-plus maze test
Anxiety-related behavior was evaluated using the elevated-plus maze test as previously described (Lister 1987). The plus maze apparatus was made of polyvinyl chloride and consisted of two open arms 30 × 5 cm and two enclosed arms 30 × 5 × 15 cm. The arms extended from a central 5 × 5 cm platform. The apparatus was elevated 40 cm above the ground. At the start of the test, the mouse was individually placed in the center of the maze, facing one of the open arms. Number of entries into each arm and cumulative time spent in each arm were recorded for 5 min and analyzed using the ANY-maze Video Tracking System.
Adult mice were anesthetized with pentobarbital (90 mg/kg) and perfused with 4% paraformaldehyde in phosphate-buffered saline. Newborn mice were decapitated. The brains were removed and immersed in the fixative for 2–4 days at 4°C. After sequential treatment in a graded series of 10%, 20%, and 30% sucrose in phosphate-buffered saline at 4°C, the brains were embedded in optimal cutting temperature compound (Sakura Finetechnical, Tokyo, Japan) and sectioned at a thickness of 30 μm on a cryostat.
Immunostaining was performed with the following reagents: goat anti-choline acetyltransferase (ChAT; #AB144P, Millipore, Billerica, MA, USA), rabbit anti-substance P (#20064; ImmunoStar, Hudson, WI, USA), rabbit anti-vesicular glutamate transporter 2 (VGLUT2; #BML-GZ1060, Enzo Life Sci, Farmingdale, NY, USA), rabbit anti-β-galactosidase (#A11132, Invitrogen, Carlsbad, CA, USA), donkey anti-goat-Alexa 488 (#A11055; Invitrogen), goat anti-rabbit Alexa 568 (#A11036; Invitrogen), ENVISION plus (rabbit; #K4003, DAKO, Carpinteria, CA, USA), donkey normal serum (#017-000-121; Jackson Laboratory), and goat normal serum (#S-1000; Vector Laboratories, Burlingame, CA, USA). The immunohistochemistry procedures were confirmed with a negative control in which the primary antibodies were omitted. Fluorescent images were obtained with confocal microscopy (LMS700; Carl Zeiss, Jena, Germany).
For cresyl violet staining, sections were immersed in 0.1% cresyl violet solution (MUTO Pure Chemicals, Tokyo, Japan) and destained in 95% ethanol containing a small quantity of acetic acid.
Brains were dissected out and processed for a modified Golgi-Cox staining as described by the manufacturer (Rapid GolgiStain Kit, FD NeuroTechnologies, Ellicott City, MD, USA). Sections were obtained using a cryostat and the stained sections were analyzed using an All-in-One fluorescence microscope (Biozero Bz-800; KEYENCE, Osaka, Japan) with a Z-stack function.
Statistical analysis was performed using a two-tailed Student's t-test. A p-value of less than 0.05 was considered significant. All the data are presented as mean ± SEM.
In the course of breeding the mice, we observed differences in the nesting behavior between Wt and dTg mice. While most Wt mice built up their nest overnight, dTg mice had significantly less nest building activity than the Wt mice among both sexes (Fig. 1).
As impaired nesting behavior is observed in many mutant mice exhibiting psychotic behaviors (Ballard et al. 2002; Miyakawa et al. 2003; Hiroi et al. 2005), we next examined locomotor activity of the mice in an open field test. Based on the center entry index, the dTg mice of both sexes exhibited significantly increased locomotor activity (Fig. 2b). Distance traveled by the dTg mice was increased in the female mice, but not the male mice (Fig. 2a). Male and female mice exhibited different responses with regard to the amount of time spent in the center of the open field. Male dTg mice spent more time in the center of the open field than male Wt mice (Wt 42.7 ± 3.8 s (n = 12), dTg 57.0 ± 4.1 s (n = 10), p = 0.019), whereas female mice showed no difference in the time spent in the center between the two genotypes (Wt 51.6 ± 4.4 s (n = 22), dTg 70.9 ± 9.8 s (n = 22), p = 0.081).
During a 10-min social interaction test, male and female mice paired with an unfamiliar age- and sex-matched Wt and dTg mice exhibited different responses. The dTg male mice exhibited significantly decreased social interactions, based on the total number and duration of the interactions. Social interaction behavior did not differ significantly between the Wt and dTg female mice (Fig. 3).
To examine the spatial learning and memory performance of dTg mice, we conducted the Morris water maze test. The mice were given five trials daily to learn the location of a submerged platform in the water maze for four consecutive days. On Day-5, the mice were given three trials and a probe test was conducted within 2 h after the third training trial. During the 4 days of training, dTg and Wt mice showed no significant differences in their escape latencies to find the submerged platform (data not shown). In the probe test to examine the time spent in the target quadrant and in the other three quadrants on Day-5, female, but not male, dTg mice were significantly impaired compared to controls (Fig. 4), suggesting that short-term spatial memory was impaired in the dTg female mice.
To evaluate anxiety-related behavior, we performed the elevated-plus maze test. The dTg mice and Wt mice did not differ in the duration of time spent in the open arms or the number of open arm entries (data not shown). Male and female mice exhibited different responses with regard to the amount of time spent in the center of the maze. Male dTg mice spent more time in the center of the maze than male Wt mice (Wt 27.0 ± 2.3 s (n = 7), dTg 42.6 ± 7.6 s (n = 5), p = 0.047), whereas female mice showed no difference in the time spent in the center between the two genotypes (Wt 32.9 ± 5.3 s (n = 10), dTg 39.4 ± 6.1 s (n = 7), p = 0.44), suggesting that the male dTg mice had slight anxiety-related symptoms.
These findings together indicate that dTg mice exhibit psychiatric disorder-related behavioral abnormalities. We next investigated whether any brain regions of the dTg mice exhibited pathologies.
To specify the distribution of the neural crest-derived cells in the brain, dTg mice were crossed with Rosa-cre-lacZ-reporter mice. Wnt1-cre/Wnt1-GAL4/floxed-lacZ mice were subjected to immunohistochemistry using the anti-β-galactosidase antibody. β-Galactosidase-positive cells were detected in the cerebellum, midbrain, medial habenula, and meninges (Fig. 5). This finding indicates that these brain regions derive from the neural crest, which specifically expresses Wnt1 protein, consistent with previous reports describing the distribution of neural crest-derived cells using different experimental systems (Danielian et al. 1998; Jiang et al. 2002; Bach et al. 2003; Creuzet 2009). Gross anatomic analysis using cresyl violet staining revealed no apparent pathologic abnormalities in neural crest-derived brain regions, such as the cerebellum, midbrain, and medial habenula (Fig. 6). We did not detect obvious abnormalities using Golgi staining in various brain regions, including the cerebral cortex, hippocampus, striatum, midbrain, and cerebellum (Fig. 7) of the dTg mice.
Previous studies suggest possible involvement of the habenula in the pathogenesis of schizophrenia in human and animal models (Sandyk 1992; Lecourtier et al. 2004). To examine the involvement of the medial habenula for the behavioral abnormality of the dTg mice, we performed histopathologic analyses. Immunohistochemistry using anti-ChAT and anti-substance P antibodies revealed no apparent morphologic anomalies in the neural crest-derived medial habenula cholinergic or substance P neurons of the dTg mice (Fig. 8a and d). Cholinergic fibers from neurons of the medial habenula nucleus to the interpeduncular nucleus also showed no morphologic abnormalities in the retroflexus fasciculus of the mutants (Fig. 8b and e). The cholinergic fibers, however, appeared abnormal in the interpeduncular nucleus (Fig. 8c and f). The interpeduncular nucleus is a region in the midbrain that derives from the neural crest.
Double immunostaining of the interpeduncular nucleus using anti-ChAT and anti-VGLUT2 antibodies revealed that cholinergic and glutamatergic fibers from the medial habenula nucleus were disordered in female dTg mice (Fig. 9, upper and medial lines). In particular, fibers doubly stained with anti-ChAT and anti-VGLUT2 antibodies revealed robustly irregular tracts in the interpeduncular nucleus of the mutants (Fig. 9, lower line of merged image). Similar defects were observed in the fibers of male dTg mice (data not shown).
The findings of this study suggest that neural crest-derived cells have a role in the development of psychiatric disorders. Although there are no reports describing a direct relation of the development of neural crest-derived cells and mental illness, dysfunction of the medial habenula, which derives from the neural crest and projects via the habenulointerpeduncular fibers to other brain areas, is suggested to be involved in the onset of human psychiatric disorders (Sandyk 1992; Caputo et al. 1998; Ranft et al. 2010). Psychiatric disorder-related behavioral abnormalities are also observed in adult habenula-lesioned rat (Lecourtier et al. 2004; Lecourtier and Kelly 2005). Our present results provide a genetic animal model for psychiatric disorders, in which the progenitor cells of the habenula and/or interpeduncular neurons might be injured by Cre recombinase and/or GAL4 transcription factor over-expressed in the neural crest during early embryonic stages, such as embryonic days 9.5–11.5 (Rowitch et al. 1999). This model animal, the dTg mouse, exhibits psychiatric disorder-related behavioral abnormalities, such as increased locomotor activity, decreased social interaction, and impaired short-term spatial memory. The behavioral abnormalities observed in mice with induced genetic alterations are similar to those in patients with congenital psychiatric disorders (Stefansson et al. 2002; Tabuchi et al. 2007; Jaaro-Peled 2009; Schmeisser et al. 2012). Furthermore, the habenulointerpeduncular fibers in the neural crest-derived brain region are disorganized in the mutant mice. These findings support the view that pathologic changes of the habenulointerpeduncular pathway contribute to the induction of psychiatric disorders and that mutations in genes related to neural crest cell development could induce etiologic changes underlying psychiatric disorders. We cannot completely rule out that genome modification at the integration site of the transgene by itself affects brain development in dTg mice and/or that defects in brain regions other than the habenulointerpeduncular pathway cause the phenotypes of the mutant.
Our results revealed behavioral abnormalities in dTg mice, a deleter line. Several other reports have described the toxicity of transgene products in deleter lines (Lee et al. 2006; Yuan et al. 2011). Lee et al. showed the toxicity of Cre recombinase over-expressed in the pancreas under the control of a short fragment of the rat insulin II gene promoter. Yuan et al. demonstrated aberrant toxic effects of the recombinase protein expressed by the powerful mouse mammary tumor virus (MMTV) promoter in mammary tissues. These findings emphasize the need for strict controls in experiments using the Cre/loxP and GAL4/UAS systems. In most cases, scientists confirm and/or demonstrate that the results obtained from animals genetically manipulated with deleter lines are because of the gene alterations intended by the experimental systems. In some cases, however, the descriptions or demonstrations are missing adequate controls (Schmidt-Supprian and Rajewsky 2007).
The incidence of a psychiatric disorder, schizophrenia, is approximately 1% and men and women suffer from this disorder with equal frequency. The phenotypes of this illness, however, are expressed differently between the sexes. Women with schizophrenia show a distinct symptom profile and a later age at onset. Premenopausal women tend to have a better response to typical anti-psychotics compared with men and post-menopausal women (Canuso and Pandina 2007; Kirkbride et al. 2012). In case of another psychiatric disorder, autism, sex differences in the prevalence are well known: the ratio of men to women is about four to one (Volkmar et al. 1993). This disorder is characterized by impaired social interaction and communication involving abnormal repetitive and restrictive behaviors (Won et al. 2012). In this study, we performed behavioral analyses of both male and female dTg mice and found conspicuous sex differences in some behavioral tests. In the social interaction test, male dTg mice showed severely decreased social interactions, but the female mice did not (Fig. 3). On the other hand, we demonstrated histologic abnormalities in the interpeduncular nucleus of the dTg mice. Although all the tissue sections showing disordered fiber tracts presented in Fig. 9 were prepared from female mice, irregular fiber tracts were equally detected in sections from male dTg mice (data not shown). These findings indicate that dTg mice might be an excellent model for studying sex differences in the pathophysiology of mental illness such as autism and schizophrenia.
There are many reports suggesting that psychiatric disorders have their origins in the disturbance of neurodevelopment (Rapoport et al. 2005; Geschwind 2008). The behavioral and pathologic observations in this study using dTg mice support the neurodevelopmental disturbance hypotheses for the pathogenesis of psychiatric disorders, because over-expression of the induced transgenes under the control of Wnt1 regulatory sequences is essentially restricted to the dorsal regions during early embryonic stages in the dTg mice (Rowitch et al. 1999).
We propose a possible mechanism underlying the psychiatric disorder-related behavior of the dTg mice. The interpeduncular nucleus extends processes to the dorsal tegmental area such as the raphe nucleus. Because most of the serotonergic neurons are found within the raphe nucleus and these serotonergic neurons have widespread projections throughout the entire brain (Dahlström and Fuxe 1964), various serotonergic functions are likely because of these serotonergic neurons in the raphe nucleus. Alterations in serotonergic activity could be responsible for the functional dopamine dysregulation observed in schizophrenia (Kapur et al. 2005). This study demonstrated abnormalities of the cholinergic and glutamatergic fiber tracts from the medial habenula neurons in the interpeduncular nucleus of the mutant. These abnormalities cause the loss of efferent connections with the interpeduncular nucleus. The disturbed wiring in the interpeduncular nucleus could lead to the breakdown of functions of the serotonergic and dopaminergic systems and result in behavioral abnormalities in the dTg mutant. To ascertain the reliability of this proposed mechanism, the neuronal connectivity and distribution of these neurotransmitters should be investigated in the brains of these mutants in future studies.
We thank Keiko Matsuda, Yasuyoshi Fukunaga, Naho Miyauchi, and Takuya Kawamura for helpful discussion. We also thank Drs. Miwako Kobayashi and Ichiro Matsuoka for technical assistance. The authors declare no competing financial interests. Author contributions: M.N., S.O., and Y.F. designed the research; H.M., C.N., M.T., and S.O. performed the research; M.N. and Y.F. wrote the manuscript.