The MuSK agonist antibody protects the neuromuscular junction and extends the lifespan in C9orf72-ALS mice

Abstract

The disassembly of the neuromuscular junction (NMJ) is an early event in amyotrophic lateral sclerosis (ALS), ultimately leading to motor dysfunction and lethal respiratory paralysis. The hexanucleotide GGGGCC repeat expansion in the C9orf72 gene is the most common genetic mutation, and the dipeptide repeat (DPR) proteins have been shown to cause neurodegeneration. While no drugs can treat ALS patients efficiently, new treatment strategies are urgently needed. Here, we report that a MuSK agonist antibody alleviates poly-PR-induced NMJ deficits in C9orf72-ALS mice. The HB9-PR^F/F mice, which express poly-PR proteins in motor neurons, exhibited impaired motor behavior and NMJ deficits. Mechanistically, poly-PR proteins interacted with Agrin to disrupt the interaction between Agrin and Lrp4, leading to attenuated activation of MuSK. Treatment with a MuSK agonist antibody rescued NMJ deficits, and extended the lifespan of C9orf72-ALS mice. Moreover, impaired NMJ transmission was observed in C9orf72-ALS patients. These findings identify the mechanism by which poly-PR proteins attenuate MuSK activation and NMJ transmission, highlighting the potential of promoting MuSK activation with an agonist antibody as a therapeutic strategy to protect NMJ function and prolong the lifespan of ALS patients.

Keywords: C9orf72; MuSK; amyotrophic lateral sclerosis; dipeptide repeat proteins; neuromuscular junction.

Graphical abstract

Figure 1

Neuromuscular structure is aberrant in HB9-PR^F/F mice (A) Diagram of the generation of HB9-PR^F/F mice. (B) Representative images showing the distribution of GFP-poly-PR proteins in lower motor neurons in the spinal cord. (C) Representative images showing the distribution of GFP-poly-PR proteins in upper motor neurons in the motor cortex. (D) The body weight of Ctrl and HB9-PR^F/F mice at the age of 30 days. Ctrl group, n = 7 mice; HB9-PR^F/F group, n = 11 mice. Data are represented as mean ± SEM. Student’s t test, ∗∗∗p < 0.001. (E) Grip strength of four limbs in Ctrl and HB9-PR^F/F mice at the age of 30 days. Ctrl group, n = 7 mice; HB9-PR^F/F group, n = 5 mice. Data are represented as mean ± SEM. Student’s t test, ∗p < 0.05. (F) The rotarod test in Ctrl and HB9-PR^F/F mice. Ctrl group, n = 7 mice; HB9-PR^F/F group, n = 7 mice. Data are represented as mean ± SEM. Student’s t test, ∗∗p < 0.01. (G) Representative images of NMJs in the TA muscle at the age of 30 days. The muscle fibers were stained whole mount with α-bungarotoxin (α-BTX) (red) to label AChR clusters and with NF/synapsin-1 (green) to label nerve terminals. (H and I) Quantification of NMJs in (G), including AChR area (H) and relative AChR intensity (I). Ctrl group, n = 133 NMJs from 9 mice; HB9-PR^F/F group, n = 233 NMJs from 9 mice. Data are represented as mean ± SEM. Student’s t test, ∗∗∗p < 0.001. (J) Poly-PR was detectable in the conditioned medium. The N2A cells were transfected with constructs expressing poly-GR or poly-PR. The conditioned medium was concentrated by immunoprecipitation and subjected to immunoblotting analysis. (K) Representative images of NMJs in the TA muscle. The muscle fibers were stained whole mount with α-BTX (red) to label AChR clusters and with GFP (green) to label poly-PR proteins. 3D representation of GFP-poly-PR (blue), and AChR clusters (red) generated from a z stack before being processed into an orthogonal projection. X, Y, and Z dimensions can be observed. The fluorescence-intensity profiles of GFP-poly-PR (blue) and AChR clusters (red) in the right region were obtained along the while line. (L) Quantification of GFP intensity in (K). Ctrl group, n = 105 NMJs from 3 mice; HB9-PR^F/F group, n = 113 NMJs from 4 mice. Data are represented as means ± SEM. Student’s t test, ∗∗∗p < 0.001.

Figure 2

Poly-PR impairs MuSK activation and AChR clustering (A) C2C12 myotubes were incubated with or without Agrin overnight in the presence of 5–10 μM poly-GR or poly-PR. AChR clusters were revealed by α-BTX staining. (B) Quantification of AChR clusters per 100 μm in (A). Data are from 4 independent experiments. n (Ctrl group) = 86 myotubes, n (Agrin group) = 53 myotubes, n (PR group) = 79 myotubes, n (GR group) = 53 myotubes. Data are represented as mean ± SEM. One-way ANOVA, ∗∗∗p < 0.001. (C) C2C12 myotubes were treated with Agrin for 0.5 h in the presence of 5–10 μM poly-GR or poly-PR. MuSK was isolated by immunoprecipitation and MuSK phosphorylation was revealed by anti-p-tyrosine (4G10) antibody. (D) Quantification of p-tyrosine intensity in (C). n = 3 independent experiments. Data are represented as mean ± SEM. One-way ANOVA, ∗p < 0.05. (E) HEK293T cells were transfected with Flag-Agrin and GFP-poly-GR or GFP-poly-PR, and immunoprecipitated with anti-Flag antibody. The immunoprecipitated complex was probed with indicated antibodies. (F) Quantification of GFP intensity in (E). n = 3 independent experiments. Data are represented as mean ± SEM. Two-way ANOVA, ∗p < 0.05. (G) Poly-PR disrupts the interaction between Agrin and Lrp4. Anti-HA beads were conjugated with conditioned medium of HEK293T cells containing HA-Agrin and were then incubated with conditioned medium of HEK293T cells expressing Flag-Lrp4-ecd and poly-GR or poly-PR peptide, respectively. The immunoprecipitated complex was probed with indicated antibodies. (H) Quantification of Lrp4 intensity in (G). n = 3 independent experiments. Data are represented as mean ± SEM. One-way ANOVA, ∗∗p < 0.01.

Figure 3

MuSK agonist antibody rescues poly-PR-induced impairment in MuSK activation and AChR clustering (A) C2C12 myotubes were treated with Agrin or MuSK agonist X-17 for 0.5 h. MuSK was isolated by immunoprecipitation with anti-MuSK antibody and MuSK phosphorylation was revealed by anti-p-tyrosine antibody. (B) C2C12 myotubes were incubated with Agrin or X-17 overnight in the presence of poly-GR or poly-PR. AChR clusters were revealed by α-BTX staining. (C) Quantification of AChR clusters per 100 μm in (B). Data are from 3 independent experiments. n (Ctrl with Agrin group) = 55 myotubes, n (PR with Agrin group) = 79 myotubes, n (GR with Agrin group) = 61 myotubes, n (Ctrl with X-17 group) = 66 myotubes, n (PR with X-17 group) = 72 myotubes, and n (GR with X-17 group) = 64 myotubes. Data are represented as mean ± SEM. One-way ANOVA, ∗∗∗p < 0.001. (D) C2C12 myotubes were treated with Agrin or X-17 for 0.5 h in the presence of poly-GR or poly-PR. MuSK was isolated by immunoprecipitation with anti-MuSK antibody and MuSK phosphorylation was revealed by anti-p-tyrosine antibody. (E) Quantification of p-tyrosine intensity in (D). n = 3 independent experiments. Data are represented as mean ± SEM. Student’s t test, ∗p < 0.05, ∗∗p < 0.01.

Figure 4

Poly-PR proteins attenuate neuromuscular transmission in mice (A) Diagram of injection of poly-PR proteins and X-17. The synthesized poly-PR proteins were injected into the TA muscles of WT mice. Poly-PR proteins were injected into the right hindlimb, and PBS was injected into the left hindlimb every 3 days, and X-17 antibody was administered to the mice by intraperitoneal injection every 7 days. (B) Grip strength of hindlimbs injected with PBS or poly-PR proteins. n = 10 mice per group. Data are represented as mean ± SEM. Student’s t test, ∗p < 0.05, ∗∗∗p < 0.001. (C) CMAPs were recorded in TA muscles in response to a train of 10 submaximal stimuli at different frequencies. The first stimulus response was normalized as 100%. Ten representative CMAP traces in 40 Hz, shown stacked in succession for better comparison. (D) CMAP amplitudes of the tenth stimulation at different stimulation frequencies. Ctrl group, n = 8 mice; PR group, Ctrl with X-17 group, and PR with X-17 group, n = 5 mice. Data are represented as mean ± SEM. two-way ANOVA, ∗p < 0.05, ∗∗p < 0.01. (E and F) Reduced CMAP amplitudes at 40 Hz in poly-PR-injected mice (E) and CMAP amplitudes of the tenth stimulation at 40 Hz (F). Ctrl group, n = 8 mice; PR group, Ctrl with X-17 group, and PR with X-17 group, n = 5 mice. Data are represented as mean ± SEM. Data in (E) were analyzed by two-way ANOVA, ∗p < 0.05. Data in (F) were analyzed by one-way ANOVA, ∗p < 0.05.

Figure 5

Poly-PR proteins impair AChR clustering and MuSK activation in mice (A) Representative images of NMJs in the TA muscle. The muscle fibers were stained whole mount with α-BTX (red) to label AChR clusters and with NF/synapsin-1 (green) to label nerve terminals. (B–D) Quantification of NMJs in (A), including AChR area (B), relative AChR intensity (C), and percentage of denervation (D). Data in (B), Ctrl group, n = 52 NMJs from 4 mice; PR group, n = 95 NMJs from 3 mice; PR with X-17, n = 98 NMJs from 3 mice. Data in (C), Ctrl group, n = 87 NMJs from 4 mice; PR group, n = 94 NMJs from 3 mice; PR with X-17, n = 133 NMJs from 4 mice. Data in (D), Ctrl group, n = 3 mice; PR group, n = 3 mice; PR with X-17, n = 3 mice. Data are represented as mean ± SEM. One-way ANOVA, ∗∗∗p < 0.001. (E) Representative images of NMJs in the TA muscle. The muscle fibers were stained whole mount with α-BTX (red) to label AChR clusters and with anti-p-tyrosine (green) to label tyrosine phosphorylation proteins. (F) Quantification of p-tyrosine intensity in (E). n = 35 NMJs from 3 mice per group. Data are represented as mean ± SEM. One-way ANOVA, ∗p < 0.05, ∗∗∗p < 0.001. (G) MuSK in muscle was isolated by immunoprecipitation with anti-MuSK antibody and MuSK phosphorylation was revealed by anti-p-tyrosine antibody. (H) Quantification of p-tyrosine intensity in (G). Data in every group are from 3 mice. Data are represented as mean ± SEM. Paired t test, ∗p < 0.05, ∗∗p < 0.01.

Figure 6

MuSK agonist X-17 rescues NMJ deficits and extends lifespan in HB9-PR^F/F mice (A) Diagram of administration of X-17 by intraperitoneal injection into HB9-PR^F/F mice. X-17 antibody was administered to the mice by intraperitoneal injection every 7 days. (B) Grip strength of four limbs in control, HB9-PR^F/F mice, and X-17-treated HB9-PR^F/F mice. Ctrl group, n = 14 mice; HB9-PR^F/F group, n = 6 mice; X-17-treated HB9-PR^F/F group, n = 6 mice. Data are represented as mean ± SEM. One-way ANOVA, ∗∗p < 0.01. (C) The rotarod test in control, HB9-PR^F/F mice, and X-17-treated HB9-PR^F/F mice. Ctrl group, n = 7 mice; HB9-PR^F/F group, n = 5 mice; X-17-treated HB9-PR^F/F group, n = 5 mice. Data are represented as mean ± SEM. One-way ANOVA, ∗p < 0.05, ∗∗p < 0.01. (D) Representative images of NMJs in the TA muscle in control, HB9-PR^F/F mice, and X-17-treated HB9-PR^F/F mice. The muscle fibers were stained whole mount with α-BTX (red) to label AChR clusters and with NF/synapsin-1 (green) to label nerve terminals. (E–G) Quantification of NMJs in (D), including AChR area (E), relative AChR intensity (F), and percentage of denervation (G). Data in (E), Ctrl group, n = 548 NMJs from 16 mice; HB9-PR^F/F group, n = 370 NMJs from 14 mice; X-17-treated HB9-PR^F/F group, n = 266 NMJs from 5 mice. Data in (F), Ctrl group, n = 589 NMJs from 15 mice; HB9-PR^F/F group, n = 484 NMJs from 13 mice; X-17-treated HB9-PR^F/F group, n = 381 NMJs from 6 mice. Data in (G), Ctrl group, n = 6 mice; HB9-PR^F/F group, n = 6 mice; X-17-treated HB9-PR^F/F group, n = 6 mice. Data are represented as mean ± SEM. One-way ANOVA, ∗∗∗p < 0.001. (H) Representative images of choline acetyltransferase (ChAT)-positive motor neurons in the spinal cord of control, HB9-PR^F/F mice, and X-17-treated HB9-PRF/F mice. The lumbar spinal cord slices from mice were stained with anti-ChAT antibody to reveal motor neurons in the anterior horn. (I) Quantification of ChAT-positive motor neurons in (H). Ctrl group, n = 5 mice; HB9-PR^F/F group, n = 5 mice; X-17-treated HB9-PR^F/F group, n = 5 mice. Data are represented as mean ± SEM. One-way ANOVA, ∗p < 0.05. (J) Kaplan-Meier survival curves after birth of HB9-PR^F/F mice and X-17-treated HB9-PR^F/F mice. Ctrl group, n = 12 mice; HB9-PR^F/F group, n = 11 mice; X-17-treated HB9-PR^F/F group, n = 11 mice. Log rank test between HB9-PR^F/F group and X-17-treated HB9-PR^F/F group, ∗∗p < 0.01. (K) The mean survival time of HB9-PR^F/F mice, and X-17-treated HB9-PR^F/F mice. n = 11 mice per group. Data are represented as mean ± SEM. Student’s t test, ∗∗p < 0.01.