Effect of Sox10 on remyelination of the hippocampus in cuprizone‐induced demyelinated mice

Abstract Objective The low number of oligodendrocytes (OLs) in the hippocampus of patients with schizophrenia suggests that hippocampal demyelination is changed in this condition. Sox10 is expressed throughout OL development. The effect of Sox10 on myelin regeneration is unknown. This study aimed to analyze changes in Sox10 expression in the hippocampus and its regulatory role in hippocampal myelin regeneration in a mouse model of demyelination. Methods Mice were fed 0.2% cuprizone (CPZ) for six weeks to establish the acute demyelinating model (CPZ mice). Behavioral changes of these mice were assessed via open field and tail suspension tests. The ultrastructure of the myelin sheaths in the hippocampus was observed by transmission electron microscopy. The expression levels of myelin sheath‐related proteins and the transcription factor Sox10 were detected via immunohistochemistry and Western blots. Furthermore, Sox10‐overexpressing adeno‐associated virus was injected into the hippocampus after establishing the demyelinating model to investigate effects of Sox10 on remyelination. Results CPZ mice showed abnormal behavioral changes, a large number of pathological changes in the myelin sheaths, and significantly reduced protein expression of the myelin sheath markers myelin basic protein and proteolipid protein. This confirmed that the demyelinating model was successfully established. Meanwhile, the protein expression of the oligodendrocyte precursor cell marker neural/glial antigen 2 (NG2) increased, whereas Sox10 expression decreased. After Sox10 overexpression in the hippocampus, the abnormal behavior was improved, the ultrastructure of the myelin sheaths was restored, and the expression of myelin sheath protein was reversed. NG2 expression was upregulated. Conclusion Overexpression of Sox10 promotes hippocampal remyelination after CPZ‐induced acute demyelination.


| INTRODUC TI ON
Schizophrenia is a common mental illness with unknown etiology.
Over the last decade, an increasing number of studies have confirmed that neuropathological changes due to myelin sheath injury contribute to schizophrenia . Clinical evidence suggests that abnormal hippocampal structure and function are closely related to schizophrenia (Papiol et al., 2017). Compared with healthy individuals, patients with schizophrenia feature smaller hippocampal volumes but no loss of neurons. Hippocampal network function is disrupted in schizophrenia and these disruptions are associated with relational memory ability, suggesting that resting hippocampal network modularity may be an important marker of neuropathology in schizophrenia (Avery, Rogers, & Heckers, 2018).
Interestingly, the expression of myelin basic protein (MBP) is reduced in the hippocampus of a demyelination animal model of schizophrenia (Cumberland, Palliser, Rani, Walker, & Hirst, 2017). A quantitative stereological study further confirmed that the number of oligodendrocytes (OLs) is decreased in the hippocampus of patients with schizophrenia (Schmitt et al., 2009). This result suggests that hippocampal demyelination occurs in schizophrenia. Genomic studies have shown that OLs and myelin-related genes in the hippocampus of patients with schizophrenia are abnormally expressed (Katsel, Davis, & Haroutunian, 2005). Therefore, the change in hippocampal demyelination may be a pathological basis of schizophrenia. OLs are myelin-forming cells in the central nervous system (CNS). Apoptosis or death of OLs leads to the demyelination of a large number of nerve fibers, resulting in the massive loss of axons and nerve cells and blocked or disordered signal transduction (Takahashi, Sakurai, Davis, & Buxbaum, 2011). But the precise mechanism remains unclear. The research had found that sex-determining region Y-related HMG box 10 (Sox10) is a transcription factor that is expressed throughout the development of OLs. Its main role is to influence the terminal differentiation of oligodendrocyte precursor cells (OPCs) and myelination. However, the effects of Sox10 on myelin regeneration are unknown. Cuprizone (CPZ) can induce demyelination in the CNS and has been widely used for studying the myelin abnormality hypothesis of schizophrenia (Benardais et al., 2013). Therefore, we used CPZ to establish an acute demyelinating mouse model and observed the changes of Sox10 expression in the hippocampus and its regulatory effect on hippocampal remyelination. It may provide a new therapeutic target for the treatment of myelin sheath injury in schizophrenia.

| Animals
Six weeks old healthy ICR male mice (18-22 g) were provided by the Experimental Animal Center of Ningxia Medical University. The mice were allowed to drink water and eat ad libitum at control ambient temperature (22°C). All experiments were carried out in accordance with the National Institutes of Health Guide for Care and were approved by the Experimental Animal Ethics Committee of Ningxia Medical University (Ethical No. 2014-014).

| Preparation of the demyelinating model
The demyelinating model was prepared by feeding mice with 0.2% CPZ (Sigma-Aldrich) for six weeks (Wang et al., 2020). The control group was fed a normal diet.

| Stereotactic injection
Mice were randomly divided into six groups (n = 15/group). Targeted injection into the hippocampus was performed with normal saline (NS), adeno-associated virus-green fluorescent protein (AAV-GFP), and Sox10 overexpressing adeno-associated virus (AAV-Sox10). The groups of mice and treatments were as follows: (a) Control + NS, CPZ + AAV-GFP, and (f) CPZ + AAV-Sox10. Mice were anesthetized intraperitoneally with chloral hydrate (4%, 0.01 ml/g) before the intracranial stereotaxic injection of AAV or NS through the mouse stereotaxic apparatus (RWD LIFE SCIENCE). Details of the stereotaxic injection were described previously (Keiser, Chen, & Davidson, 2018). Briefly, we made a longitudinal incision to expose the bregma, set this point as zero, and determined the hippocampus region from this point (AP ± 2 mm, ML ± 2 mm, DV ± 2 mm). Each side of the hippocampus was injected with 1 μl for five minutes. Then, the needle was retained for five minutes before it was slowly retracted. The same method was performed on the other side. One to two needles were used to stitch the wound. The mice were sacrificed after being fed with normal food for two weeks.

| Open field test
The open field test (OFT) was designed to observe the autonomous and exploratory behavior of experimental animals in a new environment. Activity time around the corner and center time indicates avoidance and anxiety levels, respectively (Fischer et al., 2015). The box was 30 cm high and 50 cm long at the bottom. A digital camera was set at the top. Mice were placed in the open field laboratory in advance to allow adaptation to the environment for two hours.
Before the test began, the box was swabbed with 75% ethanol and dried with a dry cloth. The mice (n = 15/group) were placed with fixed orientation toward one side of the box wall and left to explore the environment freely for five minutes. We recorded the time in the central area, time in the corner area, time in the side area, and the number of defecations . After each mouse experiment ended, the experimenters cleaned the defecation and swabbed the open box again to prevent the smell of one mouse affecting the next test.

| Tail suspension test
The tail suspension test (TST) was designed to assess the level of depression in mice (Jiang et al., 2018). The tail of mice (n = 15/group) was clamped and suspended so that the mice were hanging upside down while their head was about 10 cm away from the bottom, facing the camera. The experiment took six minutes, and the camera system recorded the immobility of the mice lasting four minutes (Chatterjee, Jaiswal, & Palit, 2012). The immobility time is the time when the mice gave up their struggle and expressed disappointment.
This time reflects the emotional changes of the mice.

| Transmission electron microscopy
The mice (n = 3/group) were anesthetized by intraperitoneal injection of 4% chloral hydrate at 0.01 ml/g and then perfused with 4% paraformaldehyde. The hippocampus was exfoliated and immersed in 2% glutaraldehyde for two hours, followed by 0.1 M sodium dimethyl arsenate solution for two hours (three times), soaked in 1% osmium tetroxide for two hours, and sodium dimethyl arsenate solution for 15 min (two times). The tissues were dehydrated in ethanol and propylene oxide series. Afterward, the specimen was immersed in complete embedding solution at 35°C for six hours, placed at 42°C overnight, and then at 60°C for 48 hr (Kim, Sohn, Kim, & Kim, 2018).
After solidification and sectioning, the ultrastructure of hippocampal myelin was observed via transmission electron microscopy (TEM).

| Immunohistochemistry
Paraffin sections (n = 6/group) were successively placed into xylene, high-to-low concentration alcohol, and then repaired antigen.

| Western blot
Total protein was extracted from hippocampal tissues (n = 6/group), and the concentration of the protein samples was detected via the BCA protein assay kit (KeyGEN BioTECH). Proteins were separated using SDS-PAGE and then electroblotted onto PVDF membranes.
The membranes were incubated with 5% skimmed milk for two hours. Anti-mouse monoclonal antibody Sox10 (1:1,000), anti-rabbit polyclonal antibody NG2 (1:1,000), anti-rabbit polyclonal antibody PLP (1:1,000), and anti-mouse monoclonal antibody MBP (1:1,000) as the primary antibodies were added to the membrane at room temperature for one hour and then overnight at 4°C. Then, the membranes were incubated with the secondary antibodies at room temperature for one hour. The bands were visualized using the enhanced chemiluminescence method. The band intensity was analyzed using Image J software.

| Statistical analysis
SPSS18.0 software was used for statistical analysis of the data.
Measurement data with normal distribution assessed via a normality test were expressed as mean ± standard error of the mean (SEM).
Abnormally distributed data were described as median ± quartile distance. For comparison of multiple groups of data, ANOVA was used for differential analysis if the data were normal and the variance was homogeneous. The Bonferroni test was used for comparison between the two groups. The nonindependent sample Kruskal-Wallis test was used if the variances were not equal. Statistical significance was considered at p < .05.

| Effect of CPZ on the behavior of mice and the ultrastructure of hippocampal myelination
In the OFT, mice of the CPZ group spent more time in the central area (p < .01; Figure 1a) and less in the corner and side area than control mice (p < .01; Figure 1b,c). No significant difference in the number of defecations was found ( Figure 1d). In the TST, the immobile time of the CPZ group was significantly increased compared with control mice (p < .05; Figure 1e). TEM data showed that the hippocampal myelin lamina of the control group was uniform in thickness and closely arranged, and the organelles inside the axon were normal in shape and size. In contrast, a large number of pathological myelin sheaths occurred in the CPZ group, which was manifested by myelin edema, increased space between myelin sheaths and axons, and vacuoles. The plate was separated and fractured, the myelin sheaths were disintegrated, and the myelin sheath structure disappeared (Figure 1f).

| Overexpression of Sox10 promotes the expression of myelin sheath-related protein in CPZ mice
No significant difference in the expression of myelin marker protein was found in the control group after the different treatments. Two weeks after the removal of CPZ, the expression of the myelin marker proteins in the CPZ group was still lower than that in the control group. The CPZ group had a large number of adenomatous polyposis coli (APC, also known as CC-1)-positive cells after overexpressing Sox10 (p < .01; Figure 5a,b). The expression of MBP and PLP increased after overexpression of Sox10 (p < .01; Figure 5c-g).

| D ISCUSS I ON
Schizophrenia is a common psychiatric disease whose etiology has not yet been fully revealed. It often occurs in young adults, impairing thinking, perception, emotion, and behavior, and can be incompatible with mental activities and the environment (Cui  (Kaefer et al., 2019) which also reflects the importance of the hippocampus for studying schizophrenia. The hippocampus is highly susceptible to injury by many injury factors (Kutlu & Gould, 2016). Myelin formation begins before birth and continues after birth until adolescence.
Schizophrenia is more common in adolescence (Gur et al., 2019). Therefore, the molecular mechanism of pathological changes in myelination of the hippocampus is important for exploring the pathogenesis of schizophrenia. In demyelinated patients, Sox10 expression is decreased and OPC differentiation is blocked (Zeis, Howell, Reynolds, & Schaeren-Wiemers, 2018). Our study results showed that the expression of Sox10 significantly decreased in the CPZ mice. To reveal the regulatory role of Sox10 in hippocampal myelin regeneration, we injected Sox10-overexpressing AAV into the hippocampus region of mice via a stereotaxic apparatus and confirmed overexpression of Sox10 in the hippocampus through immunohistochemistry and Western blots. We further studied the effect of Sox10 overexpression on hippocampal remyelination in CPZ mice. In this study, we aimed to explore a way to promote myelin regeneration.  (Pusic, Mitchell, Kunkler, Klauer, & Kraig, 2015), confirming the abnormal behavioral changes in CPZinduced demyelinated mice. However, the molecular mechanism of behavioral changes is not clear. This is the first study available on the effects of Sox10 on behavior in the CPZ mouse model. Our data showed that overexpression of Sox10 improved the degree of stress and depression in the CPZ-induced demyelinated mice. A study by Zendedel (Zendedel, Beyer, & Kipp, 2013) reported the relationship between remyelination and axonal injury and found that the animal's motor function began to recover when CPZ was removed after CPZ was fed for five weeks but the animal's motor performance was still impaired even six months after CPZ withdrawal. We overexpressed Previous studies on promoting endogenous remyelination focused on how to enhance the survival of OPCs and stimulate their proliferation, migration, and differentiation into myelin cells (Grade, Bernardino, & Malva, 2013;Kastriti & Adameyko, 2017). In our study, OPCs were found around the demyelinating lesions in the CNS. However, they all remained at a static stage, unable to proliferate and differentiate into mature OLs, and thus unable to form a new myelin sheath to wrap the exposed axons. NG2, a marker of OPCs, is used to reflect the presence of OPCs. Our study showed that the expression of NG2 increased in CPZ mice. Furthermore, the number of OPCs increased after demyelination, which may be the repair response of the body to pathological demyelination. However, despite the increase in OPCs, the expression of MBP and other myelin markers still decreased. This indicates that OPCs after demyelination have not effectively differentiated into mature oligodendrocytes. OPCs cannot differentiate and repair the myelin sheath, which is considered as the major obstacle to successful repair of demyelinating lesions (Dulamea, 2017). In this study, after overexpression of Sox10, the expression levels of myelin maturation markers CC-1, MBP, and PLP significantly increased. Additionally, the expression of NG2 increased in demyelinated mice. This result indicates that Sox10 not only promotes the differentiation of OPCs but also promotes the production of OPCs through other ways, which can supplement the consumption of OPCs caused by differentiation to solve the problem of insufficient remyelination. A previous study has shown that myelin sheaths can be fully restored after the removal of CPZ for 3-5 weeks (Wergeland, Torkildsen, Myhr, Mork, & Bo, 2012). In our study, Sox10 was overexpressed immediately after the demyelinating model was established and this promoted OLs to secrete myelin protein again after two weeks at the same level as the control group.
This result indicated that after the removal of CPZ, Sox10 significantly enhanced the expression of myelin protein to support self-repair of the myelin sheaths.
It should be noted that our study has examined only the myelin marker proteins CC-1, MBP, and PLP and the OPC marker NG2 to investigate the effect of Sox10 on hippocampal remyelination. We did not explore Sox10 signaling pathways. Future studies will be needed to address this.

| CON CLUS ION
In summary, in the CPZ-induced acute demyelinating mouse model, the expression of Sox10 and the expression of myelin marker proteins CC-1, MBP, and PLP were downregulated. Overexpression of Sox10 significantly improved the behavior, myelin morphology, myelin marker protein expression, and the number of OPCs in CPZ-induced demyelinated mice. Therefore, Sox10 plays an important role in regulating the regeneration of hippocampal myelin sheaths. It also provides a new therapeutic target for the treatment of myelin sheath injury in schizophrenia.

ACK N OWLED G M ENT
We thank the Lab of Human Anatomy, School of Basic Medical Sciences, Ningxia Medical University for technical and faculty support.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

AUTH O R S CO NTR I B UTI O N
YS performed the experiments. JD performed the data analyses and drafted the manuscript. QXH assisted with the experiments. QRM, QL, CZ, and HWL collected and interpreted the data. JL conceived the study. All authors read and approved of the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.