Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease with no cure. To develop effective treatments for this devastating disease, an appropriate strategy for targeting the molecule responsible for the pathogenesis of ALS is needed. We previously reported that mutant SOD1 protein causes motor neuron death through activation of ASK1, a mitogen-activated protein kinase kinase kinase. Additionally, we recently developed K811 and K812, which are selective inhibitors for ASK1. Here, we report the effect of K811 and K812 in a mouse model of ALS (SOD1G93A transgenic mice). Oral administration of K811 or K812 significantly extended the life span of SOD1G93A transgenic mice (1.06 and 1.08% improvement in survival). Moreover, ASK1 activation observed in the lumbar spinal cord of mice at the disease progression stage was markedly decreased in the K811- and K812-treated groups. In parallel, immunohistochemical analysis revealed that K811 and K812 treatment inhibited glial activation in the lumbar spinal cord of SOD1G93A transgenic mice. These results reinforce the importance of ASK1 as a therapeutic target for ALS treatment.
Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by the selective loss of upper and lower motor neurons. Although riluzole is the only approved drug for the treatment of ALS, it only slightly increases the life span of the patients by an average of 2–3 months (1,2). Therefore, the development of new effective medicines for ALS is necessary. Although 90% of ALS cases are sporadic, ∼10% cases are genetically inherited, and mutations in some genes are known to be causative for ALS (3–5). Mutations in the Cu, Zn superoxide dismutase (SOD1) gene are one of the causative agents of ALS (6,7), and considerable effort has been devoted to unraveling how mutant SOD1 proteins (SOD1mut) cause ALS. Transgenic mice expressing SOD1mut are widely accepted as animal models for ALS because they recapitulate the hallmarks of human ALS (8,9).
Apoptosis signal-regulating kinase 1 (ASK1) is a member of the mitogen-activated protein kinase kinase kinase (MAP3K) family (10). ASK1 is activated by various stresses and plays pivotal roles in a wide range of cellular responses. Recent studies have shown that dysregulation of the ASK1 signaling pathway leads to various diseases, including cancer, infections and neurodegenerative diseases (11,12). Therefore, ASK1 has received much attention as a potential therapeutic target. We previously reported that ASK1 is crucial for disease progression in SOD1-related ALS (13). Moreover, we showed that SOD1mut causes motor neuron death through ER stress-induced ASK1 activation. Furthermore, we crossed SOD1G93A transgenic mice with ASK1−/− mice and showed that deletion of ASK1 extends the life span of SOD1G93A transgenic mice. These results suggest that ASK1 is a potential drug target for ALS. In this study, we report for the first time the effect of ASK1 inhibitors on ALS model mice.
Specificity of K811 and K812
From a Kyowa Hakko Kirin's small compound library focusing on kinases, we identified two types of ASK1 inhibitors, K811 and K812 (14) (Fig. 1A). To examine the specificity of K811 and K812, their IC50 values against each kinase were determined by a kinase panel assay. K811 presented an IC50 value of 6 nm for ASK1 and also displayed high IC50 values for other kinases (Table 1 and Supplementary Material, Table S1). Similar findings were obtained for K812 (IC50 value for ASK1: 6 nm). These results demonstrate that K811 and K812 are selective inhibitors for ASK1. We should note that the specificity of K812 is lower than that of K811. Moreover, kinome-wide profiling data (15) also confirmed the excellent selectivity of K811 for ASK1 (Fig. 1B and Supplementary Material, Fig. S1).
|IC50 (nm)||IC50 (nm)|
|IC50 (nm)||IC50 (nm)|
Effect of K811 and K812 in an in vitro ALS model
First, we confirmed the effect of the ASK1 inhibitors in cultured cells using NSC34 motor neuron cells. Treatment with K811 or K812 significantly inhibited the H2O2-induced activation of endogenous ASK1 in NSC34 cells (Fig. 2A). To examine side effects of ASK1 inhibitors on ER stress-ASK1 axis, NSC34 or primary spinal cord cultures were treated with Tunicamycin or Thapsigargin, and the effects of ASK1 inhibitors on ER stress were examined. As a result, ASK1 inhibitors hardly affected the levels of ER stress markers PERK and Herp (Supplementary Material, Fig. S2). Next, we examined the effect of ASK1 inhibitors on SOD1mut-induced motor neuron death using an in vitro ALS model. We previously reported that lentivirus-mediated exogenous expression of SOD1mut, but not SOD1WT, induced motor neuron death in primary spinal cord cultures derived from E12.5 mouse embryos. In this study, we employed primary spinal cord cultures derived from Hb9-GFP mice, and the number of GFP-positive motor neurons was automatically counted using an image analyzer (Fig. 2B and C). As we reported previously, the number of motor neurons in the spinal cord cultures expressing SOD1G93A was significantly decreased compared with those expressing SOD1WT (Fig. 2D and E). In contrast, the number of Hoechst-positive cells, which represented the number of total cells, predominantly consisting of glial cells, was not affected by expression of SOD1G93A or SOD1WT (Fig. 2F and G). These results indicate that SOD1G93A was selectively toxic to motor neurons. Importantly, treatment with K811 or K812 completely inhibited SOD1G93A-induced primary motor neuron death (Fig. 2D and E), indicating that K811 and K812 protect motor neurons from SOD1mut-induced toxicity.
Effect of K811 and K812 on an in vivo ALS model
We examined pharmacokinetic properties of K811 and K812 in male C57BL/6J mice and found that these compounds were useful as in vivo tools to investigate ASK1 biology (Table 2). Thus, we investigated the effect of ASK1 inhibitors on SOD1G93A transgenic mice, which are known as an in vivo model for SOD1-related ALS. Because we previously showed that ASK1 is crucial for the disease progression of SOD1G93A transgenic mice (13), we hypothesized that ASK1 inhibitors could alleviate the progression of ALS pathogenesis. To examine whether ASK1 inhibitors are effective for the treatment of ALS, we administered K811 or K812 to SOD1G93A transgenic mice and evaluated the survival time. Daily oral administration of each ASK1 inhibitor (K811, 100 mg/kg/day; K812, 30 mg/kg/day) was started at the age of 28 weeks, which is around the time of disease onset. Note that we adopted 100 mg/kg as a dose of K811 because the exposure of K811 in vivo was lower than that of K812 (Table 2). As a result, the K811- and K812-treated groups showed a significant extension of survival time compared with the placebo-treated group (Fig. 3A). The average survival times were 252.6 ± 4.9 days (±SEM) for the placebo group, 268.6 ± 3.4 days for the K811 group and 273.3 ± 5.7 days for K812-treated group. K811 and K812 extended the survival of SOD1G93A mice by 1.06 and 1.08%. Next, we assessed the inhibitory effect of K811 and K812 on ASK1 activation in vivo. The whole spinal cords from SOD1G93A transgenic mice treated with ASK1 inhibitors or placebo were analyzed by immunoblot (IB). ASK1 was highly activated in the spinal cord from the placebo-treated group compared with non-transgenic mice (Fig. 3B). In contrast, spinal cords from K811- or K812-treated group showed considerable decrease in ASK1 activation (Fig. 3B). These results suggest that K811 and K812 alleviate the disease progression of ALS by inhibiting the activation of ASK1 in the spinal cord.
|Compound||Tmax (h)||Cmax (nm/mL)||AUC0-tb (nm h/mL)|
|Compound||Tmax (h)||Cmax (nm/mL)||AUC0-tb (nm h/mL)|
aAdministered at a dose of 30 mg/kg, p.o.
bCalculated from 0 to 24 h for K811, and 0 to 7 h for K812.
Effect of K811 and K812 on glial cells
Finally, we assessed how K811 and K812 alleviate the ALS pathogenesis of SOD1G93A transgenic mice. To examine the effect of ASK1 inhibitors in vivo, we analyzed the motor neurons in the lumbar spinal cord (L3) using Nissl staining. The number of motor neurons was significantly higher in the K811- and K812-treated groups than in the placebo-treated group (Fig. 4A–D and M). This result indicates that K811 and K812 may protect motor neurons from SOD1mut-mediated motor neuron death in vivo. Next, because glial cells are shown to be involved in the pathogenesis of ALS, we examined the effect of K811 and K812 on glial cells.
A few glial fibrillary acidic protein (GFAP)-positive cells or ionized calcium binding adaptor molecule 1 (Iba1)-positive cells were observed in the lumbar spinal cord of non-transgenic mice (Fig. 4E and I). In contrast, we detected a considerable increase in the number of GFAP- and Iba1-positive cells in the lumbar spinal cord of placebo-treated SOD1G93A transgenic mice (Fig. 4F and J). Importantly, we observed markedly decreased numbers of GFAP- and Iba1-positive cells in the K811- and K812-treated groups compared with the placebo-treated group (Fig. 4G, H, K and L). These results suggest that K811 and K812 may inhibit the activation of glial cells.
Despite extensive efforts to overcome ALS, there is no prophylactic or curative treatment for ALS (16). One of the reasons for this slow progress is that the molecular mechanism of ALS remains enigmatic. To effectively treat this devastating disease, we believe that the development of molecular mechanism-based therapies should be advanced.
We previously reported that SOD1mut specifically interacts with Derlin-1, a component of the endoplasmic reticulum-associated degradation (ERAD) machinery (13,17,18). This interaction was found to trigger ER stress-mediated ASK1 activation through the disruption of ERAD and lead to motor neuron death. Here, based on this molecular mechanism, we showed the potential of the ASK1 inhibitors K811 and K812, as a new therapeutic strategy for ALS. Several lines of evidence showed that dysregulated activation of ASK1 occurs in in vitro or in vivo ALS models (13,19–22), suggesting that ASK1 is a therapeutic target for ALS. Interestingly, an ASK1 inhibitor NQDI-1 and a p38 MAPK inhibitor MW069 were shown to rescue the axonal transport defect caused by SOD1G85R-YFP (23). Moreover, specific expression of constitutively activated Akt3 in spinal cord motor neurons was shown to prevent neuronal loss through the phosphorylation of ASK1 Ser83, which is associated with the inhibition of ASK1 kinase activity (24). These reports and our present data suggest that targeting ASK1-MAPK signaling pathway is beneficial for ALS treatment.
ALS has been found to be a non-cell-autonomous disease in which glial cells such as astrocytes, microglia and oligodendrocytes play a crucial pathogenic role (25–27). In this study, we found that K811 and K812 markedly inhibited the glial activation in the lumbar spinal cord of SOD1G93A transgenic mice, although it remains unknown whether K811 and K812 inhibit ASK1 activation in glial cells. Thus far, there is no evidence that ASK1 contributes to the non-cell-autonomous cell death in ALS. However, ASK1 was found to be involved in the activation of microglia in various diseases such as multiple sclerosis (28) and optic nerve injury (29), suggesting that ASK1 may cause the aberrant activation of glial cells in ALS. Further study will reveal the role of ASK1 in the non-cell-autonomous cell death mechanism for ALS.
Increasing evidence suggests that aberrant activation of ASK1 is involved in various diseases (11,30). ASK1 is a stress-responsive MAP3K and plays pivotal roles when cells are exposed to stress. Moreover, ASK1−/− mice showed no abnormalities under normal conditions (31), suggesting that ASK1 inhibitors may have a low rate of side effects for normal cells and that ASK1 is an attractive drug target in a variety of diseases. Indeed, long-term (100 days from 28 weeks of age) oral administration of K811 or K812 did not cause apparent toxicity in non-transgenic mice (data not shown). It has been reported that ASK1 inhibitors could contribute to the treatment of some diseases in vivo, such as gastric cancer (32) and multiple sclerosis (28). Further investigations into the availability of ASK1 inhibitors are expected to be continued in the future.
In conclusion, we reported for the first time the effect of the ASK1 inhibitors K811 and K812 in an ALS model. K811 and K812 are found to extend the life span of SOD1G93A transgenic mice through inhibition of ASK1 activation in the spinal cord. K811 and K812 could be promising seed compounds for the treatment of ALS.
Materials and Methods
All experiments using animals except pharmacokinetic properties in vivo in this study were performed according to the guidelines provided by the Institutional Animal Care Committee of the Graduate School of Pharmaceutical Sciences at the University of Tokyo (Tokyo, Japan). All animal studies of Pharmacokinetic properties in vivo were performed in accordance with Standards for Proper Conduct of Animal Experiments at Kyowa Hakko Kirin Co., Ltd. under the approval of the company's Institutional Animal Care and Use Committee (protocol number 08-091).
Antibodies to SOD1 (the polyclonal SOD100 antibody, Enzo Life Science), ASK1 (EP553Y, Abcam), GFAP (DAKO), Iba1 (Wako), a-tubulin (Harlan) and Flag (1E6, Wako) were purchased. The rabbit polyclonal antibody against phospho-ASK1 was described previously (33).
NSC34 cells were maintained in Dulbecco's modified Eagle medium (DMEM) containing 10% 4.5 mg/ml glucose supplemented with fetal bovine serum (FBS, Biowest) and 100 units/ml penicillin (Meiji Seika Pharma) in 5% CO2 at 37°C.
Wild-type, SOD1G93A transgenic (B6.Cg-Tg(SOD1*G93A)dl1Gur/J) and Hb9-GFP transgenic (B6.Cg-Tg(Hlxb9-GFP)1Tmj/J) mice were obtained from the Jackson Laboratory. Wild-type and SOD1G93A transgenic were housed in a specific pathogen-free (SPF) room. Hb9-GFP transgenic mice were housed in a non-SPF room. For K811 and K812 administration, SOD1G93A transgenic mice were transferred to the non-SPF room at the age of 25 weeks. The mice were genotyped by polymerase chain reaction using the following primers: SOD1G93A transgenic mice, 5′ CATCAGCCCTAATCCATCTGA 3′ and 5′ CGCGACTAACAATCAAAGTGA 3′; Hb9-GFP transgenic mice, 5′ AGTGCTTCAGCCGCTACC 3′ and 5′ GAAGATGGTGCGCTCCTG 3′.
K811 and K812 were dissolved in 10% DMSO (Sigma) and 10% chremophore EL (Sigma) immediately before use. K811 (100 mg/kg/day) or K812 (30 mg/kg/day) was administered orally to SOD1G93A mice by a disposable stomach tube (Fuchigami) from 28 weeks of age, and the administration was continued daily until the endpoint. The endpoint was defined as the time when the mice could no longer right themselves after being placed on their back. Survival analysis was performed using Kaplan–Meier method followed by Gehan-Breslow-Wilcoxon test (*P < 0.05). Only male mice were used in this study to control for possible gender differences.
Primary motor neuron death assay
The primary spinal cord culture and the motor neuron death assay were performed as described previously (13) with some modifications. Briefly, whole spinal cords from E12.5 Hb9-GFP transgenic mouse embryos were resected and incubated in 0.05% trypsin for 15–20 min at 37°C. Cells were dissociated by gentle pipetting and plated in culture dishes coated with 0.1% poly(ethyleneimine) solution (Sigma) at a density of 4 × 105 cells/well in DMEM/F12 Ham's medium (Sigma) supplemented with G-5 supplement (Life Technologies). Three hours later, an eight-fold volume of neurogrowth medium (2% horse serum, 10 µg/ml bovine serum albumin, 10 µg/ml insulin, 26 ng/ml sodium selenite, 100 µg/ml conalbumin, 13 ng/ml progesterone, 20 µg/ml hydrocortisone, 100 U/ml penicillin-streptomycin, 20 ng/ml triiodotyronine, 0.1 ng/ml brain-derived neurotrophic factor, 10 ng/ml ciliary neurotrophic factor and 0.1 ng/ml neurotrophin-3 in DMEM/F12 Ham's medium) was added. The culture medium was replaced with fresh neurogrowth medium at 24 and 72 h after seeding. At 120 h after seeding, cultured spinal cords were infected with lentivirus for 96 h. The ASK1 inhibitor was added at 24 h after infection. The GFP-positive motor neurons were automatically counted using Cellomics ArrayScan® VTI HCS Reader (Thermo Scientific) or Cellomics CellInsight (Thermo Scientific). Motor neurons were counted using an optimized Neuronal Profiling BioApplication protocol. Recombinant lentiviruses encoding Flag-SOD1WT and Flag-SOD1G93A were constructed as described previously (13).
NSC34 cells were seeded at 2.0 × 105 cells/ml in 12-well plates coated with Cellmatrix® type I-C (Nitta Gelatin). After 48 h of incubation, cells were stimulated and lysed with pre-chilled lysis buffer (20 mm Tris–HCl pH 7.5, 150 mm NaCl, 10 mm EDTA, 1% Triton X-100, 5 μg/ml leupeptin, 1 mm phenylmethylsulfonyl fluoride, 8 mm NaF, 12 mm beta-glycerophosphate, 1 mm Na3VO4, 1.2 mm Na2MoO4 and 2 mm imidazole). Spinal cords were isolated 4 h after compound administration. Lumbar spinal cords of mice were isolated after perfusion with PBS and homogenized in pre-chilled lysis buffer using a Polytron homogenizer (KINEMATICA). After centrifugation of the lysates at 17 700g at 4°C for 15 min, the supernatants were resolved on SDS–PAGE and electroblotted onto polyvinylidine difluoride membranes (Pall). After blocking with 5% skim milk in TBS-T (50 mm Tris–HCl pH 8.0, 150 mm NaCl and 0.05% Tween 20), the membranes were probed with antibodies. The proteins were detected using the ECL system.
Nissl staining and immunohistochemistry
Mice were perfused with PBS followed by 4% PFA in PBS. The lumbar spinal cords at levels L3–L4 were excised and fixed with 4% PFA in PBS for 16 h. The lumbar spinal cords were then incubated in 30% sucrose for 24 h, embedded in CryoMount I (Muto PureChemicals), and L3 spinal cords were sectioned (20 μm) using a Leica CM 3050 S cryostat. For Nissl staining, cryosections were stained with 1% cresyl violet (Chroma-Gesellshaft) solution in 0.1% acetic acid, washed with 100% ethanol and xylene and mounted onto the slide with Mount-Quick Tube (Daido Sangyo). Each section was photographed using a Leica DM 4000B microscope, and the motor neurons were manually counted. Only large neurons with a diameter larger than 20 μm in the ventral horn of one side of the spinal cord were included in cell counts. For immunohistochemistry, cryosections were washed with PBS and 100% ethanol, then blocked with 2% FBS for 2 h at room temperature. Each section was incubated with the primary antibody solution for 12–16 h.
After three washes with PBS, sections were incubated for 60 min with the secondary antibody solution at room temperature. Sections were then washed three times with PBS and mounted onto the slides. All images were obtained using a Leica TCS SP5 confocal laser scanning microscope.
IC50 determinations by cell-free kinase activity assays
The inhibitory effects of K811 and K812 against 61 kinases that included 20 Tyrosine kinases and 41 Serine/Threonine kinases including ASK1 were determined commercially using the Carna Biosciences (Kobe, Japan) kinase profiling service. ATP concentration was approximately set to the Km value for each kinase. The IC50 values for test compounds were calculated by concentration–response curve fitting using three-variable logistical equations within XLfit (model205) with the curve bottom constrained to 0 and the top constrained to 100 by nocodazole alone treatment.
In order to investigate the selectivity of K811 against 451 human kinases and disease relevant mutant variants, we employed KINOMEscan (DiscoveRx) (15).
Pharmacokinetic properties in vivo
Pharmacokinetic properties of compounds were studied in male C57BL/6J mice (n = 3 for K811 and n = 2 for K812) following single oral (30 mg/kg) dosing. Oral doses were prepared at a concentration of 3 mg/ml in 10 vol% DMSO and 10 vol% Cremophor EL aqueous solution and administered in a volume of 10 ml/kg. At 0.5, 1, 2, 4, 8 and 24 h after administration of K811 or at 0.5, 1, 2, 4 and 7 h after administration of K812, blood samples were collected from tail vein. The blood samples were centrifuged to obtain the plasma fractions, which were stored at −20°C until analysis. Plasma samples (10 μl) were precipitated with phosphate-buffered saline (pH 7.4, 20 μl), DMSO (10 μl) and acetonitrile containing an internal standard (80 μl), stirred and centrifuged. The supernatant was mixed with 0.05 vol% formic acid, and the mixture was analyzed by LC-MS/MS employing atmospheric pressure chemical ionization to determine plasma drug levels. Pharmacokinetic parameters were determined using the mean data from the mice at each time-point. The area under the concentration–time curve from time zero to the last data point (AUC0-t) was calculated using the trapezoidal method. The maximum plasma concentration (Cmax) was obtained directly from the observed values.
This work was supported by KAKENHI from the Japan Society for the Promotion of Science and the Ministry of Education, Culture, Sports, Science and Technology (MEXT); Global Center of Excellence Program; the ‘Understanding of molecular and environmental bases for brain health’ conducted under the Strategic Research Program for Brain Sciences by MEXT; the Advanced research for medical products Mining Programme of the National Institute of Biomedical Innovation (NIBIO); the Nakabayashi Trust for ALS Research and ALS Foundation, Japan ALS Association.
We thank Ayako Watanabe (the University of Tokyo) and Hisae Kadowaki (University of Miyazaki, Japan) for technical assistance. We thank Kohsuke Takeda (Nagasaki University, Japan), Atsushi Matsuzawa (Tohoku University, Japan) and all of the members of the Cell Signaling Laboratory for valuable discussions.
Conflict of Interest statement. Hidenori Ichijo undertook a consultancy assignment for Kyowa Hakko Kirin Co., Ltd.