Phosphorylated CRMP1, axon guidance protein, is a component of spheroids and is involved in axonal pathology in amyotrophic lateral sclerosis

Abstract

In amyotrophic lateral sclerosis (ALS), neurodegeneration is characterized by distal axonopathy that begins at the distal axons, including the neuromuscular junctions, and progresses proximally in a "dying back" manner prior to the degeneration of cell bodies. However, the molecular mechanism for distal axonopathy in ALS has not been fully addressed. Semaphorin 3A (Sema3A), a repulsive axon guidance molecule that phosphorylates collapsin response mediator proteins (CRMPs), is known to be highly expressed in Schwann cells near distal axons in a mouse model of ALS. To clarify the involvement of Sema3A-CRMP signaling in the axonal pathogenesis of ALS, we investigated the expression of phosphorylated CRMP1 (pCRMP1) in the spinal cords of 35 patients with sporadic ALS and seven disease controls. In ALS patients, we found that pCRMP1 accumulated in the proximal axons and co-localized with phosphorylated neurofilaments (pNFs), which are a major protein constituent of spheroids. Interestingly, the pCRMP1:pNF ratio of the fluorescence signal in spheroid immunostaining was inversely correlated with disease duration in 18 evaluable ALS patients, indicating that the accumulation of pCRMP1 may precede that of pNFs in spheroids or promote ALS progression. In addition, overexpression of a phospho-mimicking CRMP1 mutant inhibited axonal outgrowth in Neuro2A cells. Taken together, these results indicate that pCRMP1 may be involved in the pathogenesis of axonopathy in ALS, leading to spheroid formation through the proximal progression of axonopathy.

Keywords: ALS; CRMP; axon guidance; axonopathy; motor neuron; neurofilament.

Figure 1

Specificity of anti–pThr509-CRMP1 antibody. (A) Sequence comparison of the C-terminal domain in human and mouse CRMP1. Underlined letters indicate the antigen peptides comprising the anti–pThr509-CRMP1 antibody. (B) Phosphorylation of Thr509 in CRMP1 by Cdk5. HEK 293T cells were transfected with human CRMP1 and Cdk5. The anti–pThr509-CRMP1 antibody successfully recognized pThr509 in human CRMP1. (C) Immunoblotting of brain lysates of WT and Crmp1 knockout KO mice with anti–pThr509-CRMP1 antibody. pThr509-CRMP1 signals are lost in Crmp1 KO mice. (D) DAB labeling of hippocampi of WT and Crmp1 KO mice with an anti–pThr509-CRMP1 antibody. The lower panels show enlargement of the area enclosed by the dotted line in the upper panels. Brown DAB chromogen is observed in the dendrite-rich molecular layer of the hippocampal dentate gyrus in WT mice, but is not seen in Crmp1 KO mice. Scale bars = 100 μm (top); 20 μm (bottom). DG, dentate gyrus; G, granule cell layer; M, molecular layer.

Figure 2

pThr509-CRMP1 staining in lumbar spinal cords of ALS patients. (A) Immunostaining of pThr509-CRMP1 in lumbar spinal cords of patients with ALS (left) and disease controls (right). There were many pThr509-CRMP1–positive oval or round structures (arrows) in ALS patients but rarely in disease controls. The bottom figures show enlarged views of the enclosed squares in the top figures. Scale bars = 100 μm (top); 30 μm (bottom). (B) Double immunofluorescence labeling of pThr509-CRMP1 and pNFs, the latter of which are a major component of spheroids. Co-localization of both proteins indicated that pThr509-CRMP1 is also a component of spheroids. Scale bar = 50 μm. (C) Bodian silver staining and pThr509-CRMP1 immunostaining of consecutive sections in ALS spinal cords. Spheroids are positive for both stains. Scale bar = 50 μm. (D) Although the epitope of the anti–pThr509-CRMP1 antibody was derived from a mouse sequence, a phosphopeptide with a human sequence effectively blocked the immunostaining signals of spheroids. Scale bar = 50 μm. (E) Alkaline phosphatase treatment attenuated pThr509-CRMP1–positive signals of spheroids. Scale bar = 50 μm. (F) The number of pThr509-CRMP1–positive spheroids per anterior horn area (mm²) in ALS patients and disease controls. *p < 0.05.

Figure 3

Time-dependent expression relationships between pThr509-CRMP1 and pNFs in spinal spheroids. (A) Double immunofluorescence labeling of spinal spheroids with anti–pThr509-CRMP1 antibody (green) and anti-pNF antibody (red) in two ALS patients with short (3 months) or long (48 months) disease duration. Scale bar = 30 μm. (B) Mean fluorescence intensity of pThr509-CRMP1 (blue) and pNFs (red) in spheroids of the two ALS patients with a different disease duration (n = 18). (C) The pThr509-CRMP1: pNF fluorescence ratio in spheroids was significantly and negatively correlated with disease duration. (D–F) The fluorescence values of pThr509-CRMP1 (D) and pNFs (E), and the pThr509-CRMP1: pNF ratio (F) in each spheroid, in 18 ALS patients according to short (7.67 ± 3.08 (3–11) months), medium (21.3 ± 5.50 (14–30) months), and long (41.0 ± 11.0 (31–61) months) disease duration. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Figure 4

Effect of CRMP1 phosphorylation on neurite outgrowth. (A) Representative images of neurite elongation in Neuro2A cells transfected with either WT-CRMP1 (left) or the phospho-mimicking CRMP1 mutant Thr509Asp-CRMP1 (right). Scale bar = 50 μm. (B) Quantitative analysis of neurite lengths. The five longest neurites were assessed in each of five fields using ImageJ software. The experiments were repeated three times and a total of 75 neurites were evaluated in each transfection. ****p < 0.0001.

Figure 5

Schematic diagram showing the hypothetical mechanism by which axonal function in ALS is influenced by CRMP1 phosphorylation mediated by Sema3A–Cdk5 signaling. Non-phosphorylated CRMP1 in healthy individuals maintains the NMJ and contributes to axon elongation by virtue of the preserved function of the distal axon (see Figure 4). In ALS, (1) enhanced Sema3A signaling in the NMJ leads to CRMP1 phosphorylation by Cdk5, and this phosphorylation is responsible for aspects of distal axonal dysfunction such as NMJ denervation and inhibition of axonal outgrowth (see Figure 4); (2) pCRMP1 mediates the progression of distal-to-proximal axonopathy (dying back), at least partly through axonal transport dysfunction; (3) pCRMP1 accumulates as spheroids in proximal axons due to transport dysfunction; and (4) pNF accumulation is promoted by axonal transport dysfunction and the presence of pCRMP1 as an obstacle.