Pharmacological treatment promoting remyelination enhances motor function after internal capsule demyelination in mice

Multiple sclerosis (MS) is an inflammatory demyelinating disease of the central nervous system characterized by remyelination failure, axonal degeneration, and progressive worsening of motor functions. Animal models of demyelination are frequently used to develop and evaluate therapies for MS. We recently reported that focal internal capsule (IC) demyelination in mice with lysophosphatidylcholine injection induced acute motor deficits followed by recovery through remyelination. However, it remains unknown whether the IC demyelination mouse model can be used to evaluate changes in motor functions caused by pharmacological treatments that promote remyelination using behavioral testing and histological analysis. In this study, we examined the effect of clemastine, an anti-muscarinic drug that promotes remyelination, in the mouse IC demyelination model. Clemastine administration improved motor function and changed forepaw preference in the IC demyelinated mice. Moreover, clemastine-treated mice showed increased mature oligodendrocyte density, reduced axonal injury, an increased number of myelinated axons and thicker myelin in the IC lesions compared with control (PBS-treated) mice. These results suggest that the lysophosphatidylcholine-induced IC demyelination model is useful for evaluating changes in motor functions following pharmacological treatments that promote remyelination.


Introduction
Multiple sclerosis (MS) is an immune-mediated demyelinating disease of the central nervous system (CNS) characterized by impaired remyelination, axonal degeneration, and the progressive loss of motor functions (Compston and Coles, 2008;Dutta and Trapp, 2014;Lassmann et al., 2012;Reich et al., 2018). Mouse models of remyelination are frequently used to evaluate drug efficacy and pharmacological mechanisms of action for MS (Huang et al., 2011;Imeri et al., 2021). However, mouse models are currently not available for investigating if the pharmacological enhancement of remyelination promotes functional recovery.
Recently, we showed that the mouse model of lysophosphatidylcholine (LPC)-induced internal capsule (IC) demyelination can be used to evaluate motor functions and remyelination (Yamazaki et al., 2021a). The IC is a major white matter tract of motor neurons that projects from the cerebral cortex to the spinal cord, and this corticospinal pathway regulates limb movements. In patients with MS and white matter stroke, the corticospinal tracts are damaged by demyelination of the IC white matter, resulting in hemiparesis-like motor dysfunction (Lee et al., 2000). However, it remains unknown if the focal IC demyelination model can be used to evaluate the ability of drugs.
In this study, we examined the effect of clemastine on motor function after IC demyelination. Clemastine is an anti-muscarinic drug that has been shown to promote oligodendrocyte differentiation and remyelination (Green et al., 2017;Mei et al., 2014). Clemastine administration after IC demyelination improved motor function and ameliorated asymmetric motor paralysis. Furthermore, the IC in the clemastine-treated mice showed increased oligodendrocyte density, and remyelination was significantly enhanced. Together, our results suggest that the mouse IC demyelination model is useful for examining the effects of drugs on remyelination-mediated motor function.

Animals
Twelve-week-old male C57BL/6J mice were purchased from Japan SLC (Hamamatsu, Japan) or Clea Japan (Fujinomiya, Japan) and maintained in the animal facility of Jichi Medical University. The mice were kept in standard cages, with fewer than six animals per cage, at 20-25 • C under a 12/12-h light/dark cycle. All animal experiments were performed in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines. All efforts were made to reduce the number of animals needed. All experiments were approved by the Institutional Animal Care and Use Committee at Jichi Medical University (approval number 19034-05) and were performed in accordance with the committee's guidelines on the care and use of animals.

Focal demyelination of the internal capsule and Neutral red (NR) injection
Demyelination was induced by injection of 1% LPC (Sigma-Aldrich Japan, Tokyo, Japan) in phosphate-buffered saline (PBS) into the internal capsule. Mice were injected with an anesthetic drug mixture (0.3 mg/kg medetomidine, 4.0 mg/kg midazolam, and 5.0 mg/kg butorphanol) and placed on the stereotaxic frame (SR-5M-TH, Narishige, Tokyo, Japan) using the mouse ear bar (EB-3B, Narishige, Tokyo, Japan). After positioning the needle at the bregma after the mouse was fixed on the stereotactic frame, the tip of the needle was illuminated with a light source (LA-HDI108A, Hayashi-Repic, Tokyo, Japan). Stereotaxis was performed according to a previously described protocol with slight modifications (Yamazaki et al., 2021a). Briefly, LPC or black ink was stereotaxically injected into the right internal capsule (anteroposterior: − 1.5 mm from the bregma; mediolateral: +2.5 mm from the midline; dorsoventral: − 4.0, − 3.7 and − 3.4 mm from the dura). A Hamilton syringe with a needle (33 gauge) connected to the micro injector (IMS-3; Narishige) was inserted into the brain and maintained in place for 3 min before injection of LPC (total volume of 3 μl; injection of 1 μl at each depth). The interval between injections was 2 min. The needle was kept in place for 10 min after the last injection to reduce reflux along the needle track. We injected black ink into the IC, and cross sections of the brain comprising the needle path were examined to confirm the injection site. The mice were sacrificed at 1, 10 or 14 days post lesion (dpl). In some experiments, 500 μl 1% NR (10 mg/ml; Sigma-Aldrich) in PBS was injected intraperitoneally (i.p) to identify the focal lesion, as described previously (Baydyuk et al., 2019;Yamazaki et al., 2021aYamazaki et al., , 2021b.

Immunohistochemistry
Immunofluorescence staining of cryosections with FluoroMyelin green was performed as described previously (Yamazaki et al., 2018). Mice were injected with the anesthetic drug mixture and perfused transcardially with 2 mL/g body weight of 4% (w/v) paraformaldehyde in PBS (PFA/PBS). The brains were then rapidly dissected and immediately immersed in 4% PFA/PBS at 4 • C for overnight post fixation. For cryoprotection, the fixed brains were immersed overnight in 15% (w/v) sucrose/PBS, pH 7.4, at 4 • C, followed by overnight immersion in 30% (w/v) sucrose/PBS, pH 7.4, at 4 • C, before freezing in OCT compound (Sakura Finetek, Tokyo, Japan). Cryosections (12-μm-thick) were prepared using a cryostat (CM3050; Leica Microsystems), collected on coated glass slides (Matsunami, Osaka, Japan), and stored at − 30 • C. For FluoroMyelin staining, the cryosections were incubated with Fluo-roMyelin Green (1:50;Thermo Fisher Scientific,Waltham,MA,USA) in PBS, pH 7.4, containing 0.1% Triton X-100 (TX) for 1 h at room temperature (RT). For immunofluorescence staining, cryosections were permeabilized and incubated with blocking buffer (0.3% TX/Tris-buffered saline (TBS) and 10% normal goat serum) for 1 h, and then incubated overnight at 4 • C with primary antibodies in the same buffer. Thereafter, the cryosections were incubated for 1 h with primary antibodies at RT, washed three times in 0.1% TX/TBS, and then incubated for 3 h at RT with secondary antibodies and 1 μg/ml Hoechst 33342 (Thermo Fisher Scientific) for labeling nuclei. Images were captured on a confocal microscope (FV1000; Olympus, Tokyo, Japan).

Behavioral tests
The grip strength test is a simple and non-invasive method for evaluating muscle force in vivo in rodents (Brooks and Dunnett, 2009;Takeshita et al., 2017). The grip strength was measured using a grip strength meter (GMP-100; Melquest, Toyama, Japan). Mice were placed on the metal wire mesh for a few minutes and allowed to grab the mesh. Then, mice were gently pulled back, and the grip strength (force) was recorded in five trials per animal at 1-min intervals. The highest and lowest scores were removed, and the final grip strength was determined by calculating the average score of the remaining three trials.
The cylinder test was performed to examine forepaw function during exploratory behavior (Blasi et al., 2015;Starkey et al., 2005). The cylinder test was performed according to a previously published protocol (Yamazaki et al., 2021a). Briefly, the mice were placed in a cylinder (25 cm in height, 12 cm in diameter) for 7 min, and the number of touches of the wall of the cylinder with the right, left or both forepaws was recorded at 10 dpl, and the ratio of usage of each forepaw was calculated.
The wire hanging test was performed to evaluate motor function and motor paralysis (Feng et al., 2020;Villani et al., 2020) according to our previously described protocol (Yamazaki et al., 2021a). Briefly, the mouse was allowed to hang from a 2-mm-thick rod with both forepaws. Hanging time was counted from the start of placement on the rod to the time the mouse fell off. Three trials were conducted per mouse, and the longest hanging time was determined for each mouse and averaged for each group.

Electron microscopy (EM) analysis
EM analysis was performed as previously described (Yamazaki et al., 2021b). In brief, 2 h after the NR injection, mice at 14 dpl were anesthetized and transcardially perfused with PBS and then with 2% PFA/2.5% glutaraldehyde in 0.1 M phosphate buffer (PB; pH 7.4). Brain tissues were post-fixed in the same fixative at 4 • C overnight. For transmission electron microscopy (TEM), NR-labeled lesions were dissected and post-fixed in cold 2% OsO 4 in 0.1 M PB for 90 min, dehydrated in a graded ethanol series, and embedded in Quetol 812 epoxy resin (Nisshin EM Co., Tokyo, Japan). The resin was incubated at 60 • C for two nights to ensure polymerization. Ultrathin sections were prepared with an ultramicrotome (Ultracut UCT, LEICA) and stained with uranyl acetate and lead citrate. Images were captured by TEM (HT7700; Hitachi High-Tech, Tokyo Japan).

Statistical analysis
All quantitative analyses are presented as mean ± standard error of the mean (SEM). Demyelinated areas in confocal microscopy images were determined using Fluoview software (Olympus). In the EM analysis, we counted the number of myelinated axons in low magnification images of the lesioned internal capsule in two or three mice from each group (N = 12 each for the PBS and Clemastine groups, N = 8 for the Contralateral). The diameters were calculated from cross sections of axons and fibers (axon + myelin) in high magnification images of the lesioned internal capsule in two or three mice in each group using Fiji. The G-ratio was calculated from the axon and fiber diameters using Microsoft Excel software (Osanai et al., 2022). (N = 321 axons for PBS, N = 361 axons for Clemastine, N = 231 axons for Contralateral). Statistical analyses were performed using Prism 7 (GraphPad Software). The normality of the data was tested with the D'Agostino-Pearson or Shapiro-Wilk normality test. Statistical comparisons were performed using Student's t-test or one-way ANOVA, followed by Tukey-Kramer test. The significance levels are indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

LPC-induced IC demyelination causes motor deficits
To confirm the injection site, we injected black ink into the IC. In the brain section, we could observe the injected ink in the IC (Fig. S1). Previously, we demonstrated that focal IC demyelination induces asymmetric motor dysfunction in the wire hanging test and the cylinder test (Yamazaki et al., 2021a). First, as an additional sensitive behavioral test for the IC demyelination model, we conducted the grip strength test at 10 dpl ( Fig. 1A and B) (Brooks and Dunnett, 2009;Takeshita et al., 2017). We found that IC-demyelinated mice displayed significantly decreased grip strength compared with before LPC injection ( Fig. 1C; Pre, 0.2308 ± 0.0082 kg; 10 dpl, 0.2098 ± 0.0041 kg; t(4) = 3.88, P = 0.0178), suggesting that the grip strength test can detect motor functional impairment caused by IC damage. To confirm the demyelination in the IC, NR labeling was used in mice with motor deficits. NR was injected (i.p.) 2 h before sacrifice (Baydyuk et al., 2019;Yamazaki et al., 2021a). NR labeling was observed in the IC (Fig. 1D and E). In contrast, no NR labeling was detected in the contralateral IC ( Fig. 1D and E). Next, we performed FluoroMyelin and immunofluorescence staining to observe the demyelinated lesion at 10 dpl ( Fig. 1F and G). The demyelinated lesion and myelin debris were detected and activated microglia/macrophage had accumulated in the IC ( Fig. 1F and G). These results suggest that grip strength is correlated with demyelination-associated motor deficits. Therefore, the grip strength test is useful for evaluating motor deficits in the LPC-induced focal IC demyelination mouse model.

Clemastine improves motor function after IC demyelination
In the LPC-induced demyelination model, lesion formation starts a few days after focal injection, and the demyelination peaks at 7 dpl. However, it is known that demyelination and oligodendrocyte precursor cell (OPC) recruitment into the lesion are already detected at 3 dpl (Plemel et al., 2018). After demyelination, OPCs migrate to the demyelinated lesions and differentiate into mature oligodendrocytes. Thereafter, remyelination is induced spontaneously ( Fig. 2A). Based on these observations on the time course of demyelination, mice were treated with the drug from 3 dpl to optimize the effect of drug treatment on remyelination.
To determine whether the IC demyelination mouse model can be used to evaluate drugs for the ability to promote remyelination, we tested the effect of clemastine on motor function after IC demyelination. Double immunofluorescence images of coronal sections from LPC-injected mice at 10 dpl using antiionized calcium binding adapter molecule 1 (Iba1; green) and anti-myelin basic protein (MBP; red).

Clemastine changes forepaw preference in the cylinder test after IC demyelination
It has been reported that IC demyelination induces asymmetric motor deficits in MS (Lee et al., 2000;Maimone et al., 1991). We also demonstrated in our previous study that forepaw preference is changed by asymmetric motor dysfunction after IC demyelination in mice (Yamazaki et al., 2021a). To assess the forepaw use preference between PBS-and clemastine-treated mice, we conducted the cylinder test after LPC injection into the right IC (Fig. 2E-G). At 10 dpl, the percentage of right forepaw usage was significantly decreased by clemastine treatment (Fig. 2H; PBS, 34.97 ± 3.659%; Clemastine, 22.92 ± 3.207%; t(16) = 2.476, P = 0.0124). In contrast, the percentage of left forepaw usage was Fig. 2. Behavioral testing of PBS-and clemastinetreated mice after LPC-induced IC demyelination at 10 dpl. (A) Experimental design of the clemastine injection study. After LPC injection into the IC, mice were intraperitoneally injected with PBS or clemastine once a day from 3 to 9 dpl. Behavioral tests were performed at 10 dpl to assess motor recovery. (B) Quantification of body weight in each group of mice (PBS, n = 9; Clemastine, n = 9). (C) Comparison of the average grip strength between PBS-and clemastine-treated mice (PBS, n = 9; Clemastine, n = 9). The grip strength was significantly increased by clemastine treatment. (D) Comparison of the longest hanging time between PBS-and clemastine-treated mice (PBS, n = 9; Clemastine, n = 9). The longest hanging time was significantly increased by clemastine treatment. (E-G) Images of the cylinder test using the right forepaw (E), left forepaw (F) or both forepaws (G) for weight support. (H-J) Quantification of the ratio of the usage of the right forepaw only (H), left forepaw only (I) or both forepaws (J) in PBS-and clemastine-treated mice after right IC injury (PBS, n = 9; Clemastine, n = 9). Impaired left forepaw usage following demyelination of right internal capsule was ameliorated by clemastine treatment. Mean ± SEM are shown as bars and lines. *P < 0.05. Student's ttest.

Clemastine reduces axonal injury in demyelinated lesions
Axonal dystrophy has been shown to be involved in motor deficits after demyelination (Mei et al., 2016). Recently, we demonstrated that the reduction of axonal injury in lesions of the IC demyelination model is associated with recovery through remyelination (Yamazaki et al., 2021a). To determine whether clemastine reduces axonal injury, we examined the demyelinated axons by immunofluorescence staining using SMI32 and NF antibodies (Fig. 4A-D). In LPC-injected lesions, we detected SMI32-positive signals in both PBS-treated and clemastine-treated mice at 10 dpl, corresponding to the period of demyelination, while SMI32-positive axons were not observed in the contralateral (uninjured) IC (Fig. 4A-D). Furthermore, axonal swellings were observed in the lesions of both PBS-treated and clemastine-treated mice ( Fig. 4C and D). However, SMI32/NF-double positive signals tended to decrease after clemastine treatment ( Fig. 4E; PBS, 0.6237 ± 0.03127; Clemastine, 0.5733 ± 0.01312; t(5) = 1.657, P = 0.0792). These results show that clemastine treatment reduces axonal injury.

Clemastine promotes remyelination
To evaluate the state of remyelination after clemastine treatment, we performed EM analysis. PBS (control) or clemastine was administered (i. p.) into mice once daily from 3 to 12 dpl for EM observation (Fig. 5A). During the remyelination process, proliferating OPCs differentiate into mature oligodendrocytes that then produce myelin in demyelinated areas. We used EM to accurately assess myelin thickness in the remyelinating areas after drug treatment at 14 dpl. We used NR labeling, a method for visualizing white matter lesions, to examine IC demyelination after LPC injection (Baydyuk et al., 2019;Yamazaki et al., 2021a). NR was injected (i.p.) 2 h before sacrifice of PBS or clemastine-treated mice at 14 dpl. NR-labeled lesions were dissected for resin embedding, and myelin structure was examined by TEM. Compared with the contralateral IC, the LPC-injected IC displayed demyelination in each mouse (Fig. 5B). These results suggest that NR was incorporated into CNS lesions, and that NR labeling can be used in combination with EM studies for the detection of focal demyelinating lesions. Next, we counted the number of myelinated axons to compare the levels of remyelination in lesions of PBS-and clemastine-treated mice. Fourteen days after LPC injection, myelinated axons were significantly decreased in LPC-induced demyelinated lesions in PBS-and clemastinetreated mice compared with those in the contralateral IC. However, these myelinated axons were increased in lesions of clemastine-treated mice compared with PBS-treated mice ( Fig. 5C; PBS, 10.12 ± 1.079/ 100 μm 2 ; Clemastine, 14.5 ± 1.274/100 μm 2 ; Contralateral, 26.91 ± 2.097/100 μm 2 ). The G-ratio is defined as the ratio of the axonal diameter (D AX ) to the myelinated fiber diameter (D MF ) of the sheath (Gratio = D AX /D MF ) (Stikov et al., 2015). The G-ratio is frequently used to evaluate myelin thickness in animal models of demyelination/remyelination (Fig. 5D). The G-ratio was significantly increased in LPC-induced demyelinated lesions in PBS-and clemastine-treated mice compared with that in the contralateral IC. However, the G-ratio was significantly reduced by clemastine treatment in the lesion (Fig. 5E and F; PBS, 0.7313 ± 0.002889; Clemastine, 0.7138 ± 0.002906; Contralateral, 0.6956 ± 0.003284), suggesting that clemastine promotes remyelination in the LPC-induced IC demyelination model.

Discussion
In this study, we investigated whether the IC demyelination mouse model can be used to assess motor function associated with remyelination following clemastine treatment. We found that motor recovery was promoted by clemastine treatment and was associated with accelerated remyelination. Furthermore, NR labeling was successfully used for EM analysis of the CNS demyelinating lesions, which demonstrated that clemastine treatment enhances motor function in the mouse model of IC demyelination. Therefore, this model may be used to evaluate drugs for demyelinating CNS diseases as well as to elucidate the underlying pathogenetic processes.
Clemastine was identified as a candidate drug for MS that promoted OPC differentiation and myelination using the micropillar array highthroughput screening platform (Mei et al., 2014). Recently, clemastine was reported to promote functional recovery in hypoxic brain injury and spinal cord injury models via the M1 muscarinic receptor (Cree et al., 2018;Du et al., 2022). Previous reports show that accelerated remyelination by deletion of the M1 muscarinic receptor in oligodendrocytes prevents axonal dystrophy in the experimental autoimmune encephalomyelitis model (Mei et al., 2016). However, the relationship between remyelination and functional recovery following demyelinating injury remains unclear. Our findings here show that clemastine treatment promotes remyelination, improves asymmetric motor dysfunction, and reduces axonal injury. Our results are important for understanding pathologies associated with demyelination, such as MS, white matter stroke and spinal cord injury. In addition, pharmacological treatments promoting remyelination may have therapeutic potential for neurodegenerative diseases associated with white matter injury. Furthermore, the results demonstrate that both enhancement of motor function and remyelination can be successfully evaluated using the IC demyelination mouse model. Therefore, the IC demyelination mouse model can be used to evaluate changes in motor functions associated with remyelination after drug treatment.
The LPC-induced demyelination mouse model has been used for basic research. Previous reports show that functional deficits in visualevoked potentials (VEPs) and visual cliff test are induced by LPC injection into the optic nerve and optic chiasm (Baradaran et al., 2020;Ebrahim-Tabar et al., 2020;Niknam et al., 2019). Visual impairment caused by optic nerve demyelination is a common symptom in MS patients. Therefore, LPC injection into the optic nerve is appropriate for investigating visual functions and histological recovery after optic nerve demyelination. Motor dysfunction is a major symptom of MS. However, common symptoms such as motor deficits and paralysis are usually not observed in the corpus callosum or spinal cord of most demyelinating mouse models. Thus, LPC-induced IC demyelination is useful for evaluating the effects of drugs on motor functional recovery.
EM analysis was performed in NR-labeled lesions in this study. EM analysis is necessary for evaluating myelin structure and axonal morphology, and is critical for confirming demyelination and axonal pathological changes (Franklin and Ffrench-Constant, 2017;Ineichen et al., 2021). However, a critical drawback of EM analysis is the limited size of analyzed areas, because small tissue samples need to be selected in most cases for sectioning, metal staining, and observation at high magnification. NR labeling can facilitate the selection of small areas undergoing demyelination for sample preparation, because the lesions can be readily identified by macroscopic observation. NR labeling not only allows the rapid confirmation of demyelination, but also drastically saves time and effort looking for demyelinating lesions in EM analysis. The number of myelinated axons and G-ratio are indicators of the level of remyelination after demyelination. Notably, we demonstrate that clemastine promotes remyelination by EM observation. However, the major trajectory from the cerebral cortex passes through the IC and terminates on motor and interneurons in the spinal cord. Therefore, many axons from some types of neurons are projecting from several directions. Thus, a drawback of EM observation is that axons of different diameters are observed on sections of IC lesions.
Previous reports show that IC damage in MS patients is correlated with asymmetric motor dysfunction and paralysis (Bitsch et al., 2000;Lee et al., 2000). Recently, a randomized clinical trial of MS patients demonstrated the efficacy of clemastine for the treatment of chronic demyelinating injury in MS (Green et al., 2017). Evaluating drug candidates such as clemastine using animal models is important for developing new therapies for demyelinating diseases. In the current model, focal injection of LPC into the IC induced asymmetric motor deficits at the peak of demyelination. This was followed by remyelination-associated motor functional improvement. Thus, we may be able to adjust the timing of drug treatment to prevent demyelination and/or promote remyelination to optimize functional recovery.

Conclusion
In this study, we demonstrated that the IC demyelination mouse model can be used not only for histological analysis, but also for motor functional assessment, after drug treatment. Notably, the motor functional deficits caused by internal capsule demyelination were improved by pharmacological treatment promoting remyelination. In future studies, the IC demyelinating mouse model may be used to complement drug screens to evaluate drugs for MS in the preclinical stage. Taken together, our findings suggest that this animal model and the NR labeling method are powerful tools for the study of MS pathogenesis as well as drug development.  Electron microscopy (EM) analysis of PBSand clemastine-treated mice after LPC-induced IC demyelination at 14 dpl. (A) Experimental design of the clemastine injection experiment. After LPC injection into the IC, mice were treated with PBS or clemastine by i.p. injection once a day from 3 to 12 dpl. NR was injected 2 h before sacrifice at 14 dpl for the detection of demyelinating lesions. Lesions labeled with NR were dissected and observed by EM. (B) Representative EM images of the contralateral IC (left panel) and ipsilateral IC (right panel) in PBS-and clemastine-treated mice at 14 dpl. Scale bars, 2 μm.
(D) The G-ratio is defined as the ratio of the axonal diameter (D AX ) to the myelinated fiber diameter (D MF ) of the sheath (G-ratio = D AX /D MF ). (E) Quantification of the G-ratio in PBS-and clemastinetreated mice at 14 dpl (axons in visual fields: PBS = 321, Clemastine = 361, Contralateral = 231). (F) A scatterplot of G-ratios of individual axons. The imaged areas were prepared from each of two or three mice; mean ± SEM are shown as bars and lines. ***P < 0.001, ****P < 0.0001, one-way ANOVA, Tukey-Kramer test.