APP/PS1 transgenic mice (stock number 004462) were purchased from the Jackson Laboratory and maintained by breeding with C57BL/6 mice. Male APP/PS1 transgenic mice and the age-matched wild-type (WT) mice were used in all experiments. Animal care and surgical procedures were approved by the Institutional Animal Care and Use Committee of Soochow University in accordance with international laws.

Fifteen-month-old APP/PS1 transgenic mice received a diet of standard laboratory chow supplemented with clemastine (10 mg/kg/day) (sodium salt, Tocris Bioscience) for 2 months (Li et al., 2021). The transgenic mice and the WT mice received the same chow without clemastine.

Lysates from the cortex and hippocampus were extracted in lysis buffer [50-mM Tris (pH 7.4), 150-mM sodium chloride, 1% TritonX-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate] containing protease and phosphatase inhibitor cocktail (Roche, Switzerland). Lysates were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, then transferred to a polyvinylidene fluoride membrane (Millipore, Darmstadt, Germany). The membrane was blocked in 5% bovine serum albumin dissolved in Tris-buffered saline containing Tween 20 (TBST) (20-mM Tris-HCL pH 7.6, 150-mM sodium chloride, and 0.1% Tween 20) for 1 h at room temperature (RT) and then was incubated with the appropriate primary antibodies in 5% bovine serum albumin diluted in TBST for overnight at 4°C. The membranes were washed with TBST and incubated with horseradish peroxidase-conjugated secondary antibodies (Sigma-Aldrich, United States) for 1 h at RT. The immunoreacted proteins were visualized using ECL Western Blotting Detection Reagents (Pierce, United States). Semiquantitative analysis of band density was performed in ImageJ¹ for the protein levels measurement. A list of the primary antibodies used in this study can be found in Supplementary Table 1.

The 17-month-old mice were anesthetized with 1% pentobarbital and perfused with phosphate-buffered saline (PBS), followed by fixation with 4% paraformaldehyde (PFA) overnight. After that, the brain tissues were dehydrated and embedded in an optimal cutting temperature compound. Coronal cryosections of hippocampal tissue and the cortex were collected. In contrast, for the cultured cells, the cells were rinsed with cold PBS before being fixed with 4% PFA for 15 min on ice. In immunostaining, the brain sections or the fixed cells were washed with PBS containing 0.5% Triton and blocked with immunostaining blocking buffer (P0102, Beyotime) for 1 h. Subsequently, the sections are incubated with primary antibodies overnight at 4°C. After rinsing in PBS, the sections are incubated with a secondary antibody for 1 h at RT. The sections are subsequently mounted with the mounting medium containing 4,6-diamidino-2-phenylindole (Vector Laboratories). Images were captured under a Leica confocal microscope LSM700.

A novel object recognition test was performed in a 40 × 40 × 40-cm open arena. The behavior of the mouse was recorded by a camera above and analyzed in the software ANY-maze. Two identical objects were placed on opposite sides in the familiarization phase. In a 10-min time frame, the time of the mice exploring the two objects was recorded. An object preference index is then calculated, which equals the time exploring one of the identical objects/time exploring the identical object pairs × 100%. If the mice show no difference on the object preference index, then one of the objects is replaced with a novel object, and the test phase is carried out. Recognition index is calculated accordingly, which equals the time exploring novel object/(time exploring novel object + time exploring familiar object) × 100%.

In the learning phase, the mice took four trials per day for six consecutive days. A different starting position was used on each trial. The duration of one trial was 90 s for the mice to find the platform. Escape latencies (time spent swimming from the start point to the target) and path length (the distance from the start point to the platform), as well as swimming speed, were recorded. The escape latencies in the following training day were analyzed (escape latency in the following day/escape latency in the first day) and labeled as a learning trend. For probe trials, the platform was removed after the last trial of the learning phase. Mice were tested 24 h later to assess memory consolidation. The time spent in the target quadrant within 60 s was recorded. The latency to the first target site was measured, and the numbers of platform-site crossovers were recorded.

The central nervous system myelin isolation method is based on the previous publication (Erwig et al., 2019). Adult Sprague–Dawley rats were perfused with PBS, and the brain was homogenized in 0.32-M sucrose. The brain lysates were laid on 0.85-M sucrose and centrifuged at 75,000 × g for 30 min. The crude myelin presented at the interface was harvested and resuspended in dH₂O, then centrifuged at 75,000 × g for 30 min. The myelin pellet after the centrifuge is collected and washed once with ddH₂O at 12,000 × g for 15 min. The myelin pellet after the centrifuge was resuspended in 0.32-M sucrose, laid on 0.85-M sucrose, and centrifuged at 75,000 × g for 30 min. The myelin pellet was collected after the centrifuge and resuspended in ddH₂O. Centrifuged at 75,000 × g for 15 min, the purified myelin pellet is collected and resuspended in TBS and stored at –80°C. All the isolation procedures were carried out on ice or at 4°C during the centrifuges.

The oligodendrocyte cell line (OLN) cells were treated with the isolated myelin (10 μg/ml) for 30 min to induce cellular senescence in immunostainings or treated with the isolated myelin (30 μg/ml) for 4 h in Western blot analysis. After that, the cells were treated with 3-μM clemastine (dissolved in dimethyl sulfoxide) for 12 h. The cells treated with dimethyl sulfoxide is served as control. Then, cells were fixed in 4% PFA for 15 min on ice.

Senescence-associated β-galactosidase (SA-β-gal) staining was performed for both the fixed cells and the brain cryosections according to the manufacturer’s instructions (CellEvent^TM Senescence Green Detection Kit, Invitrogen), whereas the sections were subsequently subjected for immunostaining of platelet-derived growth factor-α (PDGFRα). Images were captured under a Leica confocal microscope LSM700.

All the imaging data were processed in Image J and subsequently analyzed in SPSS 25.0 and GraphPad Prism 8.0. All the data are shown as mean ± standard error of the mean. One-way analysis of variance was used to compare the difference between multiple groups. Unpaired Student’s t-test was used to analyze the difference between the two groups. P-value indicates significance, and P < 0.05 indicates statistical significance.

To calculate the density of OPCs and oligodendrocytes (OLs) (PDGFRα or CC1 immuno-positive cells per square millimeter), the number of the cells was quantified. Then, it is divided by the surface area of the brain section (square millimeter) to calculate the cell density. The percentage of the Aβ covering area is calculated in a similar way. The Aβ covering area is measured, and then, it is divided by the surface area of the brain section (square millimeter). To quantify the percentage of PDGFRα⁺ SA-β-gal⁺ and PDGFRα⁺ p21⁺ cells, the number of double-positive cells and the total PDGFRα⁺ cells were quantified. Then, the number of double-positive cells is divided by the total PDGFRα⁺ cells to calculate the percentage.

To investigate if the treatment of clemastine could have a therapeutic effect on aged APP/PS1 mice, clemastine was orally administrated to 15-month-old APP/PS1 mice for 2 months. We first examined the impact of clemastine treatment on cognitive functions to see if it can rescue the cognitive deficits of the 17-month-old APP/PS1 mice. To evaluate the memory of the mice, we performed both a novel object recognition test and Morris water maze. In a novel object recognition test, we confirmed that the age-matched WT mice and APP/PS1 mice treated with or without clemastine show no difference on preference index during the familiarization phase (Figure 1A). In contrast, in the test phase, we found that APP/PS1 mice show reduced identification index in comparison with the WT mice (Figure 1B), and the treatment of clemastine rescues the index value (Figure 1B). This result indicates that the short-term memory deficits of the aged APP/PS1 mice can be rescued by clemastine treatment. In the Morris water maze, the APP/PS1 mice display longer escape latency than the WT mice during the learning phase, indicating that the learning ability is impaired for the APP/PS1 mice (Figure 1C). In the probe trial, despite that the swimming distance shows no differences (Figure 1D), the APP/PS1 mice spent a shorter time in the target quadrant and swam crossover the target site less often than the WT mice, which indicates the memory retention is impaired for the APP/PS1 mice (Figures 1E,F). However, the treatment of clemastine did not alter the performance of the APP/PS1 mice in the trial (Figures 1C–F). As all the groups showed no differences in swimming speed, the swimming ability is not a factor affecting the performances of the mice (Figure 1G). Together, these data suggest that clemastine treatment can rescue the short-term memory impairment but not the memory retention deficits or learning ability impairment of the aged APP/PS1 mice.

Next, we examined the accumulation of Aβ in the brain to evaluate the brain neuropathological burden. A previous study showed that large Aβ plaques were formed in APP/PS1 mice at 10 months of age, which became even larger and covered more area at 17 months of age (Rupp et al., 2011). Consistent with the previous finding, by immunostaining against Aβ of 17-month-old APP/PS1 mice brain, we also observed that large Aβ plaques were formed in both the cortex and hippocampus (Figure 2A). In contrast, treatment of clemastine significantly reduced the numbers and the covering areas of Aβ plaques (Figure 2). For the cortex, on average, there was over a 50% reduction of the Aβ plaque numbers and the covering area of it (Figures 2B,C). In contrast, for the hippocampus, there was approximately 30% reduction of the Aβ plaque numbers and the covering area of it (Figures 2D,E). These data show that the brain neuropathological burden was greatly alleviated upon the chronic treatment of clemastine for only 2 months, even at the late stage of the disease.

To understand the mechanisms of the neurological amelioration effects of clemastine on APP/PS1 mice, we first investigated the impact of clemastine on modulating the proliferation of OPCs. By immunolabeling PDGFRα, a marker of OPCs (Li et al., 2017), we found that there was a significant decline of the PDGFRα immuno-positive(⁺) cells of the APP/PS1 mice in comparison with the WT mice in both the hippocampus and the cortex (Figures 3A–C), whereas the treatment of clemastine restored the number of PDGFRα⁺ cells to the level of WT mice (Figures 3A–C). Consistent with that, through Western blot analysis, we also observed an increase of PDGFRα level for the APP/PS1 mice treated with clemastine (Figures 3D–G). Therefore, clemastine prevents the loss of OPCs in the brains of aged APP/PS1 mice. By immunolabeling differentiated OLs using the CC1 clone against adenomatous polyposis coli (Bin et al., 2016), we observed an over 50 and 30% decline of CC1⁺ cells in the cortex and hippocampus, respectively, of the APP/PS1 mice in comparison with the WT mice (Figure 4), whereas the treatment of clemastine increased the numbers of CC1⁺ OLs in those two regions (Figure 4). Thereby, the clemastine treatment prevents the loss of OPCs and OLs in the brains of the aged APP/PS1 mice.

We wonder if an increase in myelin formation accompanies the increased numbers of OPCs and OLs. By immunolabeling myelin basic protein (MBP), the main protein component of myelin (Liu et al., 2021), we found that compared with the WT mice, the MBP immunofluorescence intensity was decreased for the APP/PS1 mice (Figures 5A–C), which was reverted by the treatment of clemastine (Figures 5A–C). Consistently, in Western blot analysis, we also detected the MBP protein level increased upon the treatment of clemastine (Figures 5D–G). Thus, the clemastine treatment induced myelin formation in the aged APP/PS1 mice. On the other hand, we evaluated myelin degeneration by immunolabeling degraded myelin basic protein complex (dMBP), which specifically reflects the degenerating myelin (Matsuo et al., 1997). A significant increase of dMBP signal was observed for the APP/PS1 mice in comparison with WT mice in the hippocampus, whereas the clemastine treatment reduced the signal of dMBP (Figures 5H,I), indicating the treatment that protected OL MBP from degenerating. Therefore, clemastine treatment enhanced the myelin formation and prevented its degradation in aged APP/PS1 mice.

A strong correlation between Aβ plaque and the senescence of OPCs surrounding it was observed in the APP/PS1 mice (Zhang et al., 2019). Selectively removing senescent OPCs reduces the Aβ deposition and ameliorates the cognitive deficits of the mice (Zhang et al., 2019), emphasizing a causal role of senescent OPCs in AD pathogenesis. Therefore, we wonder if the therapeutic effects of clemastine on modulating Aβ deposition and cognition are mediated by preventing the senescence of the glial cells. To directly measure the effects of clemastine on cellular senescence, we used an OLN cell line derived from primary glial cells (Richter-Landsberg and Heinrich, 1996). To induce cellular senescence, we added myelin debris as an external stressor to mimic the overwhelming degradation of myelin under the pathological condition that contributes to the senescence of glial cells (Safaiyan et al., 2016). Also, we measured the activity of the lysosomal enzyme β-galactosidase and the expression of p21, two markers of cellular senescence to detect the senescent cells (Lee et al., 2006; Jurk et al., 2012). The myelin debris induced OLN senescence, as high levels of β-galactosidase activity (SA-β-gal⁺) and the accumulation of p21 (p21⁺) were detected in a portion of the cells (Figures 6A–D). Interestingly, the treatment of clemastine significantly reduced both the number of SA-β-gal⁺ cells (Figures 6A,C) and p21⁺ cells (Figures 6B,D) to a basal level comparable with the non-treated control cells. Thus, clemastine can prevent the senescence of glial cells.

Previously, we reported clemastine enhances autophagy while suppressing the mTOR pathway in vitro and in vivo (Li et al., 2021), both of which exhibit strong interplay with cellular senescence (Kang et al., 2011; Kwon et al., 2017; Park et al., 2020). Especially, suppression of mTOR plays pivotal roles in preventing cellular senescence in a way either dependent or independent autophagy (Carroll and Korolchuk, 2017; Carroll et al., 2017; Park et al., 2020). Thus, we further examined whether clemastine altered mTOR activity in myelin debris-treated OPCs. Through Western blot analysis, we found myelin debris treatment not only induced the upregulation of p21 and p16, the markers of cellular senescence, but also increased the levels of phosphorylated mTOR (p-mTOR, Ser2448) (Figures 6E,F), indicating increased mTOR activity. In contrast, the treatment of clemastine reduced the levels of both p21 and p16, as well as p-mTOR to be comparable with the control (Figures 6E,F). Moreover, consistent with the previous observation (Li et al., 2021),we also found that both mTOR and p-mTOR levels were increased in the hippocampus of the APP/PS1 mice, whereas the treatment of clemastine reduced the protein levels to be comparable with the WT mice (Figure 6G). Therefore, clemastine prevents the senescence of glial cells and the upregulation of mTOR activity concomitantly.

To further corroborate the senescence prevention effect of clemastine on OPCs, we measured the number of PDGFRα⁺ OPCs that were positive for SA-β-gal or p21 in the hippocampus of the mice brain. In comparison with the WT mice, there was a significant increase in the percentage of PDGFRα⁺ OPCs that was positive for SA-β-gal (Figures 7A,C) or p21 (Figures 7B,D) for the APP/PS1 mice, whereas the treatment of clemastine significantly reduced the percentage of the cells (Figure 7). This result indicates that the OPCs in the APP/PS1 mice were more susceptible to be senescent, whereas clemastine can effectively protect the OPCs from senescence. Together, our data suggest that clemastine prevents the senescence of OPCs in the aged APP/PS1 mice, which may contribute to the prevention of Aβ deposition in the aged APP/PS1 mice brain.