Skip to main content
NCBI home page
CNS Neuroscience & Therapeutics logoLink to CNS Neuroscience & Therapeutics
. 2018 Apr 14;24(12):1163–1174. doi: 10.1111/cns.12855

Repurposing carbamazepine for the treatment of amyotrophic lateral sclerosis in SOD1‐G93A mouse model

Jing‐Jing Zhang 1,2,3, Qin‐Ming Zhou 4, Sheng Chen 5,✉, Wei‐Dong Le 1,2,6,✉
PMCID: PMC6489874  PMID: 29656576

Summary

Aims

To investigate the effect and mechanisms of carbamazepine (CBZ) on the onset and progression of amyotrophic lateral sclerosis (ALS) in SOD1‐G93A mouse model.

Methods

Starting from 64 days of age, SOD1‐G93A mice were orally administered with CBZ at 200 mg/kg once daily until death. The disease onset and life span of SOD1‐G93A mice were recorded. Motor neurons (MNs) in anterior horn of spinal cord were quantified by Nissl staining and SMI‐32 immunostaining. Hematoxylin and eosin (H&E), nicotinamide adenine dinucleotide hydrogen (NADH), modified Gomori trichrome (MGT), and α‐bungarotoxin‐ATTO‐488 staining were also performed to evaluate muscle and neuromuscular junction (NMJ) damage. Expressions of aggregated SOD1 protein and autophagy‐related proteins were further detected by Western blot and immunofluorescent staining.

Results

Carbamazepine treatment could delay the disease onset and extend life span of SOD1‐G93A mice by about 14.5% and 13.9%, respectively. Furthermore, CBZ treatment reduced MNs loss by about 46.6% and ameliorated the altered muscle morphology and NMJ. Much more interestingly, mechanism study revealed that CBZ treatment activated autophagy via AMPK‐ULK1 pathway and promoted the clearance of mutant SOD1 aggregation.

Conclusion

Our findings uncovered the therapeutic effects of CBZ against disease pathogenesis in SOD1‐G93A mice, indicating a promising clinical utilization of CBZ in ALS therapy.

Keywords: amyotrophic lateral sclerosis, autophagy, carbamazepine, SOD1‐G93A mice

1. INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterized by loss of upper and lower motor neurons (MNs), with an incidence of 2‐3 persons per 100 000.1 About 10% of patients with ALS are inherited due to genetic mutation, while the remaining 90% of cases are sporadic.2 The median survival of patients is 2‐3 years from symptom onset.1 So far, there are only 2 FDA‐approved drugs available for the treatment of ALS, including glutamate antagonist riluzole and antioxidant edaravone. The potential causes for ALS are multifactorial and not fully understood. Multiple cellular events contribute to the pathobiology of ALS, including oxidative stress, mitochondrial dysfunction, excitotoxicity, protein aggregation, impaired axonal transport, neuroinflammation, and dysregulated RNA signaling.3

Autophagy is involved in the clearance of long‐lived protein aggregation and organelles,4 and it is a highly conserved and tightly regulated cellular self‐degradative process, which degrades long‐lived proteins through sequestrating them by double‐membrane vesicles called autophagosomes. Then, the autophagosomes are trafficked to fuse with lysosomes or endosomes before merging into lysosomes, which could hydrolysis the sequestrated proteins.5 In ALS, several lines of evidence have identified mutant superoxide dismutase 1 (SOD1), transactivation response DNA‐binding protein of 43 kDa (TDP43), FUS RNA binding protein (FUS), and the translational product of intronic repeats in the gene C9ORF72 as the major constituents of ALS‐linked misfolded proteins aggregates,6, 7 which might activate autophagy. It has been observed that autophagic markers are upregulated in MNs of both ALS patients and animal models.8, 9 Furthermore, several ALS‐causing genes including optineurin (OPTN), sequestosome 1 (SQSTM1), valosin‐containing protein (VCP), and dynactin subunit 1 (DCTN1) have been identified to play important roles in autophagy system.10, 11 Increasing evidence has supported that activation of autophagy might play a protective role in animal models. Lithium has been shown to influence ALS progression in SOD1‐G93A mice through autophagy activation.12 Trehalose, an enhancer of mammallian target of rapamycin (mTOR)‐independent autophagy, could delay ALS onset and reduce MNs loss in SOD1‐G93A mice.13 In contrast, it has been reported that mTOR‐dependent activation of autophagy resulted in MNs loss and survival reducing in SOD1‐G93A mice,14 which may be due to inhibition of mTOR affect other physiological functions of mTOR signaling in neurons including axon guidance, dendritic spine morphogenesis, and dendritic development.15 Therefore, the mTOR‐independent activation of autophagy may serve as an ideal strategy for ALS therapy.

Carbamazepine (CBZ) is a well‐known anti‐epileptic drug used in clinical practice for more than 4 decades. It has been reported that CBZ can stimulate autophagy by decreasing the intracellular level of inositol.16, 17 We have found that long‐term CBZ treatment exhibited a protective effect in a mouse model of Alzheimer's disease possibly through enhancing the autophagic flux.18 Based on the above‐mentioned findings, in this study, we conducted experiments to investigate the potential impacts and mechanisms of CBZ on the onset and progression of ALS in SOD1‐G93A mouse model.

2. MATERIALS AND METHODS

2.1. Transgenic mice and treatment

Transgenic (TG) SOD1‐G93A mice overexpressing high‐level human mutant SOD1‐G93A gene were originally obtained from Jackson Laboratories (B6SJL‐Tg‐SOD1‐G93A‐1Gur/J, 002726). This mouse model has been reported to become paralyzed in hindlimbs on average 99.27 ± 1. 79 days, while female mice show later onsets than male mice and death at 135.77 ± 2.93 days.19 The mice were maintained as hemizygotes by crossing TG males with wild‐type (WT) females of the same B6SJL genetic background. The genotype of the TG mice was identified by PCR according to our previous report.14 Carbamazepine (ab141785, Abcam) was dissolved in dimethylsulfoxide (DMSO), and then, the mixture was diluted with water to a final concentration of 10 mg/mL (DMSO concentration 4%). The maximum applicable dose of CBZ is 20 mg/kg in human, taking into account the higher ratio of surface area to body weight and previous study,16 we chose 200 mg/kg as the therapeutic dose of CBZ for mice. No significant side effect under this dosage was observed throughout the course of treatment. Thirty‐six female TG mice were randomized into CBZ‐treated group (TG‐CBZ) and DMSO‐treated group (TG‐DMSO) with 18 mice per group. In TG‐CBZ group, mice were orally administered with CBZ at 200 mg/kg once daily (via gavage, 0.2 mL/10 g body weight). In TG‐DMSO group, mice were treated with an equal volume of 4% DMSO once daily. Although SOD1‐G93A mice have no obvious symptoms at 60 days of age, there had been some mild pathological changes in spinal cord including minor MNs loss.20 Therefore, CBZ treatment was started from 64 days (on the 10th weeks) of age until the animal's death. Simultaneously, 36 age‐ and sex‐matched wild‐type (WT) littermates were divided into WT‐CBZ and WT‐DMSO groups. In each group, twelve mice were used for assessing the disease onset and life span. The rest of the mice were sacrificed at 120 days of age, and their tissues were subjected to Western blot analysis (3 mice per group) or pathological staining analysis as described below. The body weight of all mice was measured every 4 days from the first drug administration to death. All experiments were conducted according to the principles established for the care and use of laboratory animals by the National Institutes of Health and were approved by the Animal Care Committee of Dalian Medical University.

2.2. Assessment of disease onset and life span

Disease onset was assessed by rotarod test according to our previous study.21 After 1‐week training session, the rotarod test was performed on a 4‐cm rod once every other day at a constant speed of 20 rpm for a duration of 5 minutes from 70 days of age until disease onset. If a mouse fell down from the apparatus within 5 minutes, the other 2 tests will be performed after 20‐minutes rest. The mice were recorded as a disease onset when the longest run time was less than 5 minutes.

As for the life span, the time of animals death was defined as the day when mice could no longer right themselves when they were placed on their back for 30 seconds.22 Mice were then sacrificed to reduce further pain from respiratory failure according to animal care guidelines.

2.3. Frozen sections preparation and MNs counting

Three mice in each group were used for the preparation of frozen sections. The mice were anesthetized and perfused with 4% paraformaldehyde (PFA) in phosphate buffer (PBS). The lumbar spinal cord was removed and fixed with 4% PFA overnight at 4°C, followed by dehydrating with 15% and 30% sucrose solution for 24 hour, respectively. Then, the tissues were embedded by optimal cutting temperature (OCT) and cut into serial sections of 10 μm with cryostat microtome. For MNs counting, 1 of every 4 slices and a total of fifty slices were selected for Nissl staining. The MNs in the anterior horns of both sides were counted by an independent investigator who was blinded to genotype and group according to the criterion as described in our previous study.22 We also counted the number of MNs by immunofluorescent staining with SMI‐32 antibody. Twenty slices were collected per animal at an interval of 10 slices. MNs in anterior horns at both sides per slice were counted.

2.4. Pathological analysis of skeletal muscle

The fresh gastrocnemius muscle was dissected and frozen in n‐hexane at −70°C cooled in liquid nitrogen for 2 minutes. Cross sections (10 μm) were cut using cryostat microtome at −20°C and were stained with hematoxylin and eosin (H&E), nicotinamide adenine dinucleotide hydrogen (NADH), or modified Gomori trichrome (MGT) according to our previously described protocols.21 The neuromuscular junction (NMJ) was stained by α‐bungarotoxin‐ATTO‐488 (1:1000, Alomone, B‐100‐AG) and photographed by fluorescent microscope (Olympus BX53). Ten images at a final magnification of 200 ×  were counted in each mouse. ImageJ software was used to measure the area of gastrocnemius muscle fibers and the percent of type I fiber in 10 visual fields of each mouse. The number of NMJ was counted by a researcher blinded to the experimental design.

2.5. Extraction of soluble and insoluble proteins

According to our previously reported protocol,13 lumbar spinal cord were removed and homogenized in ice‐cold lysis buffer (10 mmol/L Tris‐HCl pH 8.0, 1 mmol/L EDTA pH 8.0, and 100 mmol/L NaCl, 1% NonidetP40 (NP40), and 50 mmol/L iodoacetamide) containing protease inhibitor cocktail and PMSF by sonication. After being centrifuged at 130 000 g for 15 minutes at 4°C, the homogenates were divided into supernatant and pellet fraction. The supernatant was reserved as soluble protein for future Western blotting analysis. The pellet was resuspended in lysis buffer by sonication and centrifuged as previous step. Then, the remaining pellet was ultrasonically dissolved in resuspension buffer (10 mmol/L Tris‐HCl, pH 8.0, 1 mmol/L EDTA, 100 mmol/L NaCl, 0.5% NP40, 0.5% deoxycholic acid (Sigma, D6750), and 2% SDS) and reserved as insoluble proteins for future analyzing by Western blot.

2.6. Western blot analysis

After mixing with the loading buffer and boiling at 100°C for 5 minutes, equal amounts of proteins were separated by 8%‐12% sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE). Polyvinylidene fluoride (PVDF) membranes were used for protein transfer and then incubated with corresponding primary antibodies overnight at 4°C after blocking by 5% bovine serum or skim milk. The primary antibodies included anti‐SOD1 (1:2000, Abcam, ab16831), anti‐LC3 (1:1000, Sigma, L7543), anti‐p62 (1:1000, CST, 5114), anti‐Beclin1 (1:1000, CST, 3738), anti‐mTOR (1:1000, Abcam, ab32028), anti‐P‐mTOR (1:1000, Abcam, ab107268), anti‐AKT (1:1000, CST, 4691), anti‐P‐AKT (1:1000, CST, 9611), anti‐p70s6k (1:1000, CST, 2708), anti‐P‐P70S6K (1:1000, CST, 9208), anti‐AMPK (1:1000, CST, 5831), anti‐p‐AMPK (1:1000, CST, 2535), anti‐p‐ULK1 (ser757) (1:1000, CST, 14202), anti‐p‐ULK1 (ser555) (1:1000, CST, 5869), anti‐GAPDH (1:5000, CST, 2118s), and anti‐β‐actin (1:10000, Sigma, a5441). After washing with Tris‐buffered saline‐Tween (TBST) for 3 times (10 minutes each time), the anti‐rabbit or anti‐mouse HRP‐linked secondary antibodies were added. The membranes were incubated at room temperature for 1 hour and then were washed with TBST for 3 times (10  minutes each time). The chemical luminescence was assessed using FluorChem Q system (ProteinSimple, USA)

2.7. Immunofluorescent staining

The frozen sections were heated at 55°C for 30 minutes, washed with PBS for 3 times (5 minutes/time), and blocked in 5% bovine serum in PBS‐0.3% Triton X‐100 for 1 hour. After incubating with primary antibodies against SMI‐32 antibody (1:1000, Abcam, ab8135), LC3B (1;200, Sigma, L7543), P62 (1;100, CST, 5114), or SOD1 (1;200, Abcam, ab16831) overnight at 4°C, the sections were washed with PBS for 3 times and incubated with Alexa Fluor 594‐conjugated goat anti‐rabbit secondary antibody (1:2000, CST, 8889s) for 60 minutes at room temperature. After washing 3 times with PBS, coverslips were partially dried and mounted onto slides using fluorescence mounting medium. Microphotographs were analyzed with a fluorescent microscope (BX53, Olympus).Mean number of LC3 puncta/cell out of 20 MNs per mouse was calculated by a person blinded to groups. The mean fluorescence density of SOD1 or P62 was analyzed with ImageJ software.

2.8. Statistical analysis

All data were expressed as the mean ± SEM and analyzed with SPSS 22.0. The comparisons of survival and disease onset among groups were statistically analyzed by Kaplan‐Meier survival analysis. Other data distribution type was explored using a Shapiro‐Wilk test. If the data meet normal distribution and the variances between groups were equal, 1‐way ANOVA with or without Bonferroni post hoc multiple comparisons was used for statistic analysis; otherwise, nonparametric tests (Mann‐Whitney U test or Kruskal‐Wallis 1‐way ANOVA test) were used. A value of P < .05 was considered statistically significant. P < .0125 was used as significance threshold for Bonferroni post hoc multiple comparisons.

3. RESULTS

3.1. Carbamazepine treatment delayed the disease onset and extended life span of SOD1‐G93A mice

Weight loss is a frequent feature of ALS. While WT mice exhibited a gradual increase in body weight, the SOD1‐G93A mice showed an inhibited body weight gain. Carbamazepine treatment delayed this body weight loss of TG mice as compared with TG‐DMSO group (Figure 1A) and reduced mean body weight loss between initial weight and death weight by about 18.1% (Figure 1B). Moreover, compared with TG‐DMSO group, a significant delay of disease onset was observed in TG‐CBZ group as assessed by rotarod test (115.83 ± 1.37 days vs 101.17 ± 1.20 days, P < .001; Figure 1C). Carbamazepine treatment delayed the disease onset by about 14.5% (Figure 1D). Carbamazepine treatment also significantly extended the life span of SOD1‐G93A mice by about 13.9% (153.75 ± 2.76 days vs 134.9 ± 3.46 days, P < .001; Figure 1E,F). There was no significant difference in disease duration among 2 groups (37.92 ± 2.64 days vs 33.50 ± 2.22 days, P = .215).

Figure 1.

Figure 1

Open in a new tab

Carbamazepine treatment delayed the disease onset and extended life span of SOD1‐G93A mice. A, The body weight curves in the 4 groups of mice. B, Mean body weight loss between initial weight and death weight. C, Probability of onset was compared by Kaplan‐Meier statistic analysis. D, Mean age at disease onset. E, Probability of survival was compared by Kaplan‐Meier statistic analysis. F, Mean age at death. The values were presented as mean ± SEM, n = 12 in each group, **P < .001

3.2. Carbamazepine treatment alleviated the MNs loss in the anterior horn of SOD1‐G93A mice

The loss of MNs in anterior horn of spinal cord is a major pathological feature of ALS. We evaluated the number of MNs in lumbar spinal cord section of each group by Nissl staining. Consistent with the data of disease progression, there was approximately 70.7% loss of MNs in TG mice as compared to WT mice. More importantly, CBZ treatment rescued MNs loss by about 46.6% in TG mice, although no statistical significance was reached (Figure 2A,B). The immunofluorescent staining of SMI‐32 also indicated an alleviating effect of CBZ against MNs loss in anterior horn of SOD1‐G93A mice (Figure 2C,D).

Figure 2.

Figure 2

Open in a new tab

Carbamazepine treatment alleviated the MNs loss. A, Photographs of Nissl‐stained MNs in anterior horn of spinal cord in 4 groups. B, The number of MNs in lumbar spinal cord in different groups (both sides of 50 slices were counted in each mouse). C, MNs were stained with SMI‐32 antibody in lumbar spinal cord in different groups. D, Quantification of SMI‐32‐positive MNs in both sides of 1 slices in lumbar spinal cord. *P < .0125, **P < .001, n = 3 in each group, scale bar = 50 μm

3.3. Carbamazepine treatment ameliorated morphological damage of muscle and protected NMJ in SOD1‐G93A mice

Besides body weight decline and MN loss, muscle atrophy is also the main manifestation of ALS. To confirm the therapeutic potential of CBZ against ALS, we further analyzed the morphological changes of gastrocnemius muscle using H&E staining and MGT staining. Compared to WT mice, H&E staining showed significant angular muscle atrophy, grouped atrophic fibers, and central nuclei in TG mice. These morphological deteriorations could be alleviated by CBZ treatment (Figure 3A). The average muscular fiber area in TG‐CBZ group was significantly larger than that in TG‐DMSO group (1238.1 ± 53.8 μm2 vs 748.6 ± 37.8 μm2, P < .001; Figure 3B). modified Gomori trichrome staining also showed a decreased muscle atrophy in TG mice treated with CBZ (Figure 3C), indicating a protective effect of CBZ on muscular fiber. Then, we assessed the metabolic muscle phenotype by determining the proportion type I fibers using NADH staining. Our data revealed that the type I fibers (dark blue) in TG mice were significantly increased. There was lower ratio of type I fiber in TG‐CBZ group than that in TG‐DMSO group, although difference did not reach statistical significance (Figure 3D,E). These data suggested that CBZ treatment had a protective role on muscle morphology in SOD1‐G93A mice.

Figure 3.

Figure 3

Open in a new tab

Carbamazepine (CBZ) treatment ameliorated morphological damage of muscle (A). H&E staining of gastrocnemius muscle sections in different groups, significantly angular muscle atrophy, grouped atrophic fibers, and central nuclei (white arrow) were found in TG‐DMSO group, and CBZ ameliorated these pathological changes. B, The average fiber area of gastrocnemius muscle in different groups. C, modified Gomori trichrome staining in different groups. D, NADH staining in different groups. E, The percent of type I muscle fiber. F, The neuromuscular junction (NMJ) staining with a‐bungarotoxin in different groups. G, The average number of NMJs per field in different groups. n = 3 in each group. The values were presented as mean ± SEM, *P < .0125, **P < .001, Scale bar = 50 μm

Neuromuscular junction destruction is an early pathological change in SOD1‐G93A mice.23 We labeled NMJs on cross section of gastrocnemius muscle using α‐bungarotoxin. While SOD1‐G93A mice showed a significant NMJs degeneration compared with WT littermates, the number of NMJs in TG‐CBZ group was better preserved than that in TG‐DMSO group (9.06 ± 0.49 vs 6.17 ± 0.35, P = .005; Figure 3F,G).

3.4. Carbamazepine treatment reduced mutant SOD1 protein aggregation which was associated with autophagy activation

Mutant SOD1 aggregates are involved in the pathogenesis of ALS,24 and reducing mutant SOD1 protein aggregation is regarded as an effective strategy to treat ALS.25 Carbamazepine have been reported to promote the degradation of mutant α1‐antitrypsin Z protein,16 so we hypothesized possible changes of SOD1 protein aggregation in spinal cord of those CBZ‐treated TG mice. As expected, immunofluorescent staining showed that the expression level of SOD1 in MNs of TG mice was significantly higher than that of WT mice. Carbamazepine treatment could significantly reduce the expression of SOD1 in TG mice (Figure 4A,B). Further analysis of aggregated and monomeric SOD1 protein using Western blot also confirmed the immunofluorescent staining results. Strong expression of monomeric SOD1 protein was observed in all TG mice, but monomeric and aggregated SOD1 proteins were significantly reduced in CBZ‐treated TG mice (Figure 4C,D).

Figure 4.

Figure 4

Open in a new tab

Effect of CBZ on the SOD1 aggregation in SOD1‐G93A mice. A, Immunostaining of SOD1 in MNs, scale bar = 10 μm. B, Quantitative analysis of SOD1 density in MNs (*P < .0125, **P < .001). C, Western blot analysis of aggregated and monomeric SOD1 protein levels. D, Quantitative analysis of aggregated and monomeric SOD1 protein levels (*P < .05). n = 3 in each group. The values were presented as mean ± SEM

Considering the degradation functions of autophagy in eliminating abnormal aggregation of proteins, we further examined the expressions of autophagy‐related proteins to determine whether the CBZ‐induced decrease in mutant SOD1 aggregates was associated with autophagy activation. We firstly tested LC3 conversion by Western blot. Carbamazepine treatment could increase LC3 expression in WT and TG mice, as compared with those DMSO‐treated controls (Figure 5A,B). However, the expression change in LC3‐II between TG‐CBZ and TG‐DMSO groups was lack of statistical significance, which might be due to the improved autophagic flux. To prove this hypothesis, we next examined the expression of Beclin1 and p62. The expression of Beclin1 in TG‐CBZ group was 121% higher than that in TG‐DMSO group, while p62 in TG‐CBZ group was 77% lower than that in TG‐DMSO group (Figure 5A,C,D). We also detected the changes in LC3 and P62 expression by immunofluorescent staining. Consistent with Western blot data, compared to WT‐DMSO group, LC3‐II puncta per MN was apparently increased in WT‐CBZ, TG‐DMSO, and TG‐CBZ group, but no significantly difference was observed between TG‐DMSO and TG‐CBZ groups (Figure 5E,F). Notably, immunofluorescent staining also showed a significant reduction in p62 expression in TG mice after CBZ treatment (Figure 5G,H). These results suggested that CBZ was able to activate autophagy and improve autophagic flux, and the decrease in aggregated SOD1 protein may be associated with the activation of autophagy.

Figure 5.

Figure 5

Open in a new tab

Carbamazepine regulated autophagic flux in SOD1‐G93A mice. A, Western blot analysis of protein level of LC3, Beclin1, and P62 in 4 groups. Quantitative analysis protein level of LC3II (B), Beclin1 (C), and P62 (D). E, Immunostaining of LC3 in the MNs. F, Quantitative analysis of LC3 puncta. G, Immunostaining of P62 in the MNs. H, Quantitative analysis of P62 density. n = 3 in each group. The values were presented as mean ± SEM, *P < .0125, **P < .001, Scale bar = 10 μm

3.5. Carbamazepine treatment activated autophagy through AMPK‐ULK1 (Ser555) pathway

Inhibition of mTOR signaling is an important initiator of autophagy. Here, we investigated the potential involvement of mTOR pathway in CBZ‐activating autophagy by testing the alteration of p‐AKT/AKT and p‐p70s6k/p70s6k, which are upstream signaling molecules and downstream substrates of mTOR, respectively. Our data clearly revealed that the ratios of both p‐AKT/AKT (Figure 6A,B) and p‐p70s6k/p70s6k (Figure 6E,F) were obviously decreased in TG mice, compared with WT mice, indicating an inhibited mTOR signaling in SOD1‐G93A mouse model. Interestingly, no significant difference was observed between CBZ‐treated mice and control mice in both WT and TG groups (Figure 6C,D).

Figure 6.

Figure 6

Open in a new tab

Carbamazepine activated autophagy by AMPK‐ULK1 pathway. Western blots and quantitative analysis for p‐AKT/AKT (A,B), P‐mTOR/mTOR (C,D), P‐P70S6K/P70S6K (E, F), P‐ULK1 (ser757) (G,H), AMPK (I,J), P‐AMPK (K,L), and P‐ULK1 (ser555) (M,N) protein level in each group. n = 3 in each group. The values were presented as mean ± SEM, *P < .0125

Mammallian target of rapamycin could inhibit autophagy through phosphorylating Unc‐51‐like autophagy‐activating kinase 1 (ULK1) at Ser757. Consistent with the data of p‐AKT/AKT and p‐p70s6k/p70s6k, we found that the expression of P‐ULK1 (Ser757) was decreased in TG mice. Carbamazepine treatment also caused no change in the P‐ULK1 (Ser757) expression (Figure 6G,H). All these results indicated that CBZ induced autophagy via mTOR‐independent mechanisms.

Except the mTOR‐dependent pathways, the activation of 5′‐AMP‐activated protein kinase (AMPK) directly induces autophagy by phosphorylating ULK1 at Ser555. Therefore, we next examined the level of AMPK and P‐AMPK (Thr172). Western blotting analysis indicated that AMPK level was reduced in TG mice at the age of 120 days. Moreover, there was an increase in AMPK in TG and WT mice after CBZ treatment, although no significant difference was observed (Figure 6I,J). Interestingly, we noted an increase in p‐AMPK after CBZ treatment, especially in TG mice (Figure 6K,L). Similarly, the p‐ULK1 (Ser555) level was significantly increased in WT and TG mice treated with CBZ, suggesting CBZ might induce autophagy via AMPK‐ULK1 (Ser555)‐Beclin1 signaling pathway.

4. DISCUSSION

A large number of promising drugs have been preclinically discovered and showed successful therapeutic effects in various ALS animal models, but most of them failed to demonstrate clinical efficacy. Single‐factor or single‐target treatments such as antioxidants alone seem unlikely to stop ALS progression due to the multifactorial nature of ALS.26 Multifactorial combination therapy or multitargets therapy may be a choice for ALS treatment such as riluzole, which not only modulates glutamatergic transmission, but also modulates sodium channel current and inhibits neurotransmitter release.27 As an old drug for the treatment of epilepsy, CBZ also shows neuroprotective effects through a variety of mechanisms including inhibition of glutamate release, antagonizing glutamate receptor and modulating signal transduction.28 We revealed that CBZ treatment could not only delay the disease onset and extend life span of SOD1‐G93A mice, but also alleviate the MNs loss and morphological damage of muscle and protect neuromuscular junction. Moreover, many failed studies only focused on survival impact but lack the result of potential impact on quality of life. These findings indicated that CBZ may have the potential to improve the quality of life for patients with ALS through preserving MNs and functions of muscles when treatment is initiated at an early stage of ALS. Moreover, CBZ has been widely used for treatment of seizure disorders and neuropathic pain, and it is easy to start a phase II clinical trial to assess its effect on patients with ALS.

Misfolded protein aggregation is a major pathological feature of ALS, which is detected in spinal cord even at the presymptomatic stage, and may propagate via cell‐to‐cell transmission.29, 30 Accumulation of mutant SOD1 aggregates has been proposed to result in MNs loss by gain of function toxicity and inhibition of a multitude of cellular function including axonal transport, mitochondrial function, and protein homeostasis.31 Clearance of mutant SOD1 aggregates played a protective role in reducing MNs damage.13, 32, 33 We found CBZ treatment reduced monomeric and aggregated mutant SOD1 proteins in lumbar spinal cord, which might contribute to CBZ‐induced neuroprotection in ALS mice. Consistent with our findings, CBZ treatment has been reported to promote the clearance of abnormal protein aggregation in several other diseases, such as hepatic fibrosis and fibrinogen storage disease.16, 34

Autophagy is a main pathway for degradation of SOD1 aggregates. Moreover, a dysregulated autophagy has been observed in ALS.10 Increased expression levels of LC3‐II and p62 have been found in spinal cord of SOD1‐G93A mice, and electron microscopy observation has also revealed accumulations of autophagic vacuoles in this TG animal model of ALS.14 In our present study, CBZ treatment could reduce p62 accumulation and increase expression of Beclin1. P62 could bind ubiquitylated protein aggregates and deliver them to the autophagosomes. Meanwhile, p62 is also constantly degraded via nonselective autophagy through its LIR domain that binds to LC3 on autophagosome membranes.35 Decrease in p62 protein can reflect activation of autophagic flux. Beclin1 enables the recruitment of other autophagy proteins involved in nucleation and maturation of the autophagosome, and the deletion of Beclin1 caused earlier SOD1 aggregation and markedly shortened survival in ALS mice.36 Overexpression of Beclin1 in injured neuronal cells resulted in greater LC3II/I conversion and cell viability, lower levels of apoptosis, and higher BCL‐2 expression.37 Based on the above results, the unconspicuous increase in LC3II may be associated with the increase in autophagy flux. CBZ can activate autophagy and increase autophagic flux in SOD1‐G93A mice, possibly leading to the clearance of SOD1 aggregation.

Mammallian target of rapamycin‐related signaling pathway plays important roles in autophagy regulation. Mammallian target of rapamycin activation could inhibit autophagy through phosphorylating ULK1 at Ser757 and disrupting the interaction between ULK1 and AMPK, which is essential for autophagosome biogenesis.38 Mammallian target of rapamycin activity could be induced by AKT through suppressing the repressive action of tuberous sclerosis complex (TSC) by phosphorylating TSC2. Meanwhile, mTOR induces the phosphorylation of ribosomal protein S6 kinase (S6K) and eukaryotic translation initiation factor 4E‐binding protein 1 (4E‐BP1).39 In addition to regulating autophagy, mTOR pathway also promotes lipid biosynthesis and regulates glucose metabolism and mitochondrial function. Inhibition of mTOR pathway may have negative impacts on age‐related diseases.40 We found CBZ treatment did not affect the expression of the key proteins in mTOR pathway including p‐mTOR, p‐AKT, and p‐P70S6K, especially the phosphorylation level of ULK1 at Ser757 site, suggesting mTOR‐independent mechanisms might be involved in CBZ‐induced autophagy.

AMPK could directly activate autophagy by phosphorylating ULK1 at multiple sites including Ser555. Loss of AMPK or ULK1 resulted in aberrant accumulation of the autophagy adaptor p62 In mammals.41 In ALS mice, the phosphorylation levels of AMPK were reduced in spinal cord at the end stages of the disease.42 Synergetic activation of AMPK could improve autophagic flux and extend survival in ALS mice.43, 44 Consistent with our present data, previous studies have also reported that CBZ treatment could induce AMPK activation and promote ULK1 phosphorylation at Ser555 site through inositol‐1,4,5‐triphosphate (IP3) depletion.16, 17 Carbamazepine also prevent calcium overloading in mitochondria and increase expression of Beclin1 and ATG7 by suppressing calpain activation.45

Of course, neuronal hyperexcitability was also an important feature of ALS. It has been reported that the faster firing cluster MNs appeared hyperpolarization at 2‐3 months of age and lost at 4 months of age in SOD1‐G85R mice.46 Carbamazepine bound to voltage‐gated sodium channel and prevent repetitive and sustained firing of neurons, which may be another reason for its neuroprotective effect in ALS mice. Further study with electrophysiology tests may provide experimental evidence to better reveal the multifactorial nature of CBZ, especially its ameliorating activities on muscle function and body weight loss and improvement in the quality of life. However, we should note that our research still has some limitations to clinical translation. Clinical treatments always start after symptom onset, and a successful postonset treatment in preclinical models will be easier to translate into clinical therapies. Further studies are required to better evaluate the therapeutic effects of CBZ in ALS mice when treatment is initiated after the onset of obvious symptoms.

In conclusion, CBZ treatment showed a therapeutic potential in SOD1‐G93A mouse model of ALS. Carbamazepine is capable of activating autophagy through AMPK‐ULK1 pathway, reducing mutant SOD1 aggregation, protecting MNs and muscles. Clinical trials are encouraged to further validate its therapeutic benefits for patients with ALS.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

ACKNOWLEDGMENT

This work is supported by the National Natural Science Foundation of China (81430021, 81771521 and 81671241), the Program for Liaoning Innovative Research Team in University (LT2015009), the Scientific Research Fund of Liaoning Provincial Education Department (L2015145), the Liaoning Science and Technology Project (2015225008), and Shanghai New Youth Science and Technology Award (15QA1403000).

Zhang J‐J, Zhou Q‐M, Chen S, Le W‐D. Repurposing carbamazepine for the treatment of amyotrophic lateral sclerosis in SOD1‐G93A mouse model. CNS Neurosci Ther. 2018;24:1163–1174. 10.1111/cns.12855

Contributor Information

Sheng Chen, Email: mztcs@163.com.

Wei‐Dong Le, Email: wdle@sibs.ac.cn.

REFERENCES

  • 1. Al‐Chalabi A, Hardiman O. The epidemiology of ALS: a conspiracy of genes, environment and time. Nat Rev Neurol. 2013;9:617‐628. [DOI] [PubMed] [Google Scholar]
  • 2. Taylor JP, Brown RH Jr, Cleveland DW. Decoding ALS: from genes to mechanism. Nature. 2016;539:197‐206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Ferraiuolo L, Kirby J, Grierson AJ, Sendtner M, Shaw PJ. Molecular pathways of motor neuron injury in amyotrophic lateral sclerosis. Nat Rev Neurol. 2011;7:616‐630. [DOI] [PubMed] [Google Scholar]
  • 4. Ciechanover A, Kwon YT. Degradation of misfolded proteins in neurodegenerative diseases: therapeutic targets and strategies. Exp Mol Med. 2015;47:e147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Tyedmers J, Mogk A, Bukau B. Cellular strategies for controlling protein aggregation. Nat Rev Mol Cell Biol. 2010;11:777‐788. [DOI] [PubMed] [Google Scholar]
  • 6. Al‐Chalabi A, Jones A, Troakes C, King A, Al‐Sarraj S, van den Berg LH. The genetics and neuropathology of amyotrophic lateral sclerosis. Acta Neuropathol. 2012;124:339‐352. [DOI] [PubMed] [Google Scholar]
  • 7. Marangi G, Traynor BJ. Genetic causes of amyotrophic lateral sclerosis: new genetic analysis methodologies entailing new opportunities and challenges. Brain Res. 2015;1607:75‐93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Li L, Zhang X, Le W. Altered macroautophagy in the spinal cord of SOD1 mutant mice. Autophagy. 2008;4:290‐293. [DOI] [PubMed] [Google Scholar]
  • 9. Sasaki S. Autophagy in spinal cord motor neurons in sporadic amyotrophic lateral sclerosis. J Neuropathol Exp Neurol. 2011;70:349‐359. [DOI] [PubMed] [Google Scholar]
  • 10. Cipolat Mis MS, Brajkovic S, Frattini E, Di Fonzo A, Corti S. Autophagy in motor neuron disease: key pathogenetic mechanisms and therapeutic targets. Mol Cell Neurosci. 2016;72:84‐90. [DOI] [PubMed] [Google Scholar]
  • 11. Navone F, Genevini P, Borgese N. Autophagy and neurodegeneration: insights from a cultured cell model of ALS. Cells. 2015;4:354‐386. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Fornai F, Longone P, Cafaro L, et al. Lithium delays progression of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A. 2008;105:2052‐2057. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Zhang X, Chen S, Song L, et al. MTOR‐independent, autophagic enhancer trehalose prolongs motor neuron survival and ameliorates the autophagic flux defect in a mouse model of amyotrophic lateral sclerosis. Autophagy. 2014;10:588‐602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Zhang X, Li L, Chen S, et al. Rapamycin treatment augments motor neuron degeneration in SOD1(G93A) mouse model of amyotrophic lateral sclerosis. Autophagy. 2011;7:412‐425. [DOI] [PubMed] [Google Scholar]
  • 15. Switon K, Kotulska K, Janusz‐Kaminska A, Zmorzynska J, Jaworski J. Molecular neurobiology of mTOR. Neuroscience. 2017;341:112‐153. [DOI] [PubMed] [Google Scholar]
  • 16. Hidvegi T, Ewing M, Hale P, et al. An autophagy‐enhancing drug promotes degradation of mutant alpha1‐antitrypsin Z and reduces hepatic fibrosis. Science. 2010;329:229‐232. [DOI] [PubMed] [Google Scholar]
  • 17. Schiebler M, Brown K, Hegyi K, et al. Functional drug screening reveals anticonvulsants as enhancers of mTOR‐independent autophagic killing of mycobacterium tuberculosis through inositol depletion. EMBO Mol Med. 2015;7:127‐139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Li L, Zhang S, Zhang X, et al. Autophagy enhancer carbamazepine alleviates memory deficits and cerebral amyloid‐beta pathology in a mouse model of Alzheimer's disease. Curr Alzheimer Res. 2013;10:433‐441. [DOI] [PubMed] [Google Scholar]
  • 19. Pfohl SR, Halicek MT, Mitchell CS. Characterization of the contribution of genetic background and gender to disease progression in the SOD1 G93A mouse model of amyotrophic lateral sclerosis: a meta‐analysis. J Neuromuscul Dis. 2015;2:137‐150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Zhang X, Chen S, Li L, Wang Q, Le W. Decreased level of 5‐methyltetrahydrofolate: a potential biomarker for pre‐symptomatic amyotrophic lateral sclerosis. J Neurol Sci. 2010;293:102‐105. [DOI] [PubMed] [Google Scholar]
  • 21. Song L, Gao Y, Zhang X, Le W. Galactooligosaccharide improves the animal survival and alleviates motor neuron death in SOD1G93A mouse model of amyotrophic lateral sclerosis. Neuroscience. 2013;246:281‐290. [DOI] [PubMed] [Google Scholar]
  • 22. Zhou QM, Zhang JJ, Li S, Chen S, Le WD. n‐butylidenephthalide treatment prolongs life span and attenuates motor neuron loss in SOD1G93A mouse model of amyotrophic lateral sclerosis. CNS Neurosci Ther. 2017;23:375‐385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Campanari ML, Garcia‐Ayllon MS, Ciura S, Saez‐Valero J, Kabashi E. Neuromuscular junction impairment in amyotrophic lateral sclerosis: reassessing the role of acetylcholinesterase. Front Mol Neurosci. 2016;9:160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Salehi M, Nikkhah M, Ghasemi A, Arab SS. Mitochondrial membrane disruption by aggregation products of ALS‐causing superoxide dismutase‐1 mutants. Int J Biol Macromol. 2015;75:290‐297. [DOI] [PubMed] [Google Scholar]
  • 25. Gregoire S, Glitzos K, Kwon I. Suppressing mutation‐induced protein aggregation in mammalian cells by mutating residues significantly displaced upon the original mutation. Biochem Eng J. 2014;91:196‐203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Bond L, Bernhardt K, Madria P, Sorrentino K, Scelsi H, Mitchell CS. A metadata analysis of oxidative stress etiology in preclinical amyotrophic lateral sclerosis: benefits of antioxidant therapy. Front Neurosci. 2018;12:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Dorst J, Ludolph AC, Huebers A. Disease‐modifying and symptomatic treatment of amyotrophic lateral sclerosis. Ther Adv Neurol Disord. 2018;11:1756285617734734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Johannessen LC. Antiepileptic drugs in non‐epilepsy disorders: relations between mechanisms of action and clinical efficacy. CNS Drugs. 2008;22:27‐47. [DOI] [PubMed] [Google Scholar]
  • 29. Wei R, Bhattacharya A, Hamilton RT, Jernigan AL, Chaudhuri AR. Differential effects of mutant SOD1 on protein structure of skeletal muscle and spinal cord of familial amyotrophic lateral sclerosis: role of chaperone network. Biochem Biophys Res Commun. 2013;438:218‐223. [DOI] [PubMed] [Google Scholar]
  • 30. Sugaya K, Nakano I. Prognostic role of “prion‐like propagation” in SOD1‐linked familial ALS: an alternative view. Front Cell Neurosci. 2014;8:359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Kitamura A, Inada N, Kubota H, et al. Dysregulation of the proteasome increases the toxicity of ALS‐linked mutant SOD1. Genes Cells. 2014;19:209‐224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Yung C, Sha D, Li L, Chin L‐S. Parkin protects against misfolded SOD1 toxicity by promoting its aggresome formation and autophagic clearance. Mol Neurobiol. 2016;53:6270‐6287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Jablonski AM, Lamitina T, Liachko NF, et al. Loss of RAD‐23 protects against models of motor neuron disease by enhancing mutant protein clearance. J Neurosci. 2015;35:14286‐14306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Puls F, Goldschmidt I, Bantel H, et al. Autophagy‐enhancing drug carbamazepine diminishes hepatocellular death in fibrinogen storage disease. J Hepatol. 2013;59:626‐630. [DOI] [PubMed] [Google Scholar]
  • 35. Moscat J, Karin M, Diaz‐Meco MT. p62 in cancer: signaling adaptor beyond autophagy. Cell. 2016;167:606‐609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Tokuda E, Brannstrom T, Andersen PM, Marklund SL. Low autophagy capacity implicated in motor system vulnerability to mutant superoxide dismutase. Acta Neuropathol Commun. 2016;4:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Wang Z‐Y, Lin J‐H, Muharram A, Liu W‐G. Beclin‐1‐mediated autophagy protects spinal cord neurons against mechanical injury‐induced apoptosis. Apoptosis. 2014;19:933‐945. [DOI] [PubMed] [Google Scholar]
  • 38. Kim J, Kundu M, Viollet B, Guan KL. AMPK and mTOR regulate autophagy through direct phosphorylation of Ulk1. Nat Cell Biol. 2011;13:132‐141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Heras‐Sandoval D, Perez‐Rojas JM, Hernandez‐Damian J, Pedraza‐Chaverri J. The role of PI3K/AKT/mTOR pathway in the modulation of autophagy and the clearance of protein aggregates in neurodegeneration. Cell Signal. 2014;26:2694‐2701. [DOI] [PubMed] [Google Scholar]
  • 40. Johnson SC, Rabinovitch PS, Kaeberlein M. mTOR is a key modulator of ageing and age‐related disease. Nature. 2013;493:338‐345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Egan DF, Shackelford DB, Mihaylova MM, et al. Phosphorylation of ULK1 (hATG1) by AMP‐activated protein kinase connects energy sensing to mitophagy. Science. 2011;331:456‐461. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Watanabe S, Hayakawa T, Wakasugi K, Yamanaka K. Cystatin C protects neuronal cells against mutant copper‐zinc superoxide dismutase‐mediated toxicity. Cell Death Dis. 2014;5:e1497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Mancuso R, del Valle J, Modol L, et al. Resveratrol improves motoneuron function and extends survival in SOD1(G93A) ALS mice. Neurotherapeutics. 2014;11:419‐432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Coughlan KS, Mitchem MR, Hogg MC, Prehn JH. “Preconditioning” with latrepirdine, an adenosine 5’‐monophosphate‐activated protein kinase activator, delays amyotrophic lateral sclerosis progression in SOD1(G93A) mice. Neurobiol Aging. 2015;36:1140‐1150. [DOI] [PubMed] [Google Scholar]
  • 45. Kim JS, Wang JH, Biel TG, et al. Carbamazepine suppresses calpain‐mediated autophagy impairment after ischemia/reperfusion in mouse livers. Toxicol Appl Pharmacol. 2013;273:600‐610. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Hadzipasic M, Tahvildari B, Nagy M, Bian M, Horwich AL, McCormick DA. Selective degeneration of a physiological subtype of spinal motor neuron in mice with SOD1‐linked ALS. Proc Natl Acad Sci U S A. 2014;111:16883‐16888. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from CNS Neuroscience & Therapeutics are provided here courtesy of Wiley

RESOURCES