Critical role of calpain in spinal cord degeneration in Parkinson's disease

Abstract

While multiple molecular mechanisms contribute to midbrain nigrostriatal dopaminergic degeneration in Parkinson's disease (PD), the mechanism of damage in non-dopaminergic sites within the central nervous system, including the spinal cord, is not well-understood. Thus, to understand the comprehensive pathophysiology underlying this devastating disease, postmortem spinal cord tissue samples (cervical, thoracic, and lumbar segments) from patients with PD were analyzed compared to age-matched normal subjects or Alzheimer's disease for selective molecular markers of neurodegeneration and inflammation. Distal axonal degeneration, relative abundance of both sensory and motor neuron death, selective loss of ChAT(+) motoneurons, reactive astrogliosis, microgliosis, increased cycloxygenase-2 (Cox-2) expression, and infiltration of T cells were observed in spinal cord of PD patients compared to normal subjects. Biochemical analyses of spinal cord tissues revealed associated inflammatory and proteolytic events (elevated levels of Cox-2, expression and activity of μ- and m-calpain, degradation of axonal neurofilament protein, and concomitantly low levels of endogenous inhibitor - calpastatin) in spinal cord of PD patients. Thus, pathologically upregulated calpain activity in spinal cords of patients with PD may contribute to inflammatory response-mediated neuronal death, leading to motor dysfunction. We proposed calpain over-activation and calpain-calpastatin dysregulation driving in a cascade of inflammatory responses (microglial activation and T cell infiltration) and degenerative pathways culminating in axonal degeneration and neuronal death in spinal cord of Parkinson's disease patients. This may be one of the crucial mechanisms in the degenerative process.

Keywords: axonal degeneration; calpain; inflammation; neurodegeneration.

Figure 1

TUNEL with NeuN and ChAT suggested neuronal degeneration is involved in motoneuronal demise in PD spinal cord. (A) Merged images of TUNEL (red) and NeuN (green) in cervical and thoracic spinal cord sections (5 μm) of neurologically normal subjects (n = 4) and PD patients (n = 12). Co-localization (yellow) of TUNEL with NeuN IR shown by arrows signify neuronal death in dorsal and ventral horns of PD spinal cord; this is absent in normal subjects. (B) Merged images of TUNEL (red) and ChAT (green) in cervical and thoracic spinal cord of normal subjects (n = 4) and PD patients (n = 12) demonstrate damaged ventral motoneurons (yellow) in PD spinal cord, which was absent in normal subjects; magnification 200 ×.

Figure 2

Reduced axonal protein NF-L with elevated deNFP profiles in postmortem PD spinal cord. (A-B) Representative immunoblots and bar graphs corresponding to the 68 kDa band showed % change of NF-L IR in cervical (n = 6) and thoracic (n = 8) sections of PD spinal cord compared to neurologically normal subjects (n = 4). Reduced NF-L IR (68 kDa) was found in cervical spinal cord, whereas changes in thoracic were not significant. Of note, the monoclonal antibody against NF-L distinctly recognized three more bands in all normal samples analyzed, which were markedly reduced in both cervical and thoracic PD spinal cord tissues. As positive controls, significant loss of NF-L was seen in multiple sclerosis (MS) (n = 2) and Huntington’s disease (HD) (n = 2). (C-D) Representative photomicrographs of deNFP immunofluorescence staining and corresponding quantification of pixels indicate prominent increase of deNFP in cervical and thoracic spinal cord of PD patients (n = 12), compared to normal subjects (n = 4). Data are expressed as mean ± S.E.M; *p ≤ 0.05 denotes statistically significant difference between spinal cords of PD patients and normal subjects; magnification 200 ×.

Figure 3

Associated inflammatory cascades in spinal cord degeneration in PD. (A) Marked activation of astroglial cells (GFAP) and microglial cells (Iba-1) in PD spinal cord were found via immunofluorescent staining. Images show increased GFAP IR (upper panel) and Iba-1 IR (lower panel) in lumbar PD spinal cord (n = 12) compared to AD (n = 3); profound reactive gliosis was seen in MS and HD spinal cord, represented as positive reference. (B) Double immunostaining of Iba-1 with pan-axonal NFP (SMI312) shown in ventral horn (upper panel) and white matter (lower panel) of lumbar PD spinal cord indicating presence of activated microglia in both white and gray matter. (C-D) Pro-inflammatory Cox-2 protein levels were more expressed in PD than in normal spinal cord. Immunoblots and corresponding quantification of the 72-70 kDa band showing % change of Cox-2 IR in cervical (n = 6) and thoracic (n = 8) spinal cord sections of PD patients compared to those of normal subjects (n = 4); *p ≤ 0.05. (E) CD3 immunostaining showed T cell infiltration in cervical PD spinal cord (n = 8) compared to normal (n = 4).

Figure 4

Calpain and calpastatin profiles were altered at protein levels only. PD spinal cord (n = 6) relative to normals (n = 4) was tested with RT-PCR and Western blotting. (A, B) Transcriptional expression of m-calpain (131 bp), and calpastatin (188 bp) in normal and PD spinal cord; β-actin (198 bp) and 18SrRNA (245 bp) were used as internal controls for each sample. Corresponding bar graphs indicated no significant alteration in transcriptional activation/repression of calpain or calpastatin in PD compared to normals (*p ≤ 0.05). (C, D) Immunoblots demonstrated increased levels of 80 kDa m- and μ-calpain in PD spinal cord relative to normal subjects. The active forms of 76 kDa m- and μ-calpain were also detected in PD spinal cord, but significantly less in normal subjects. Levels of the 110 kDa calpastatin were markedly diminished in PD spinal cords relative to normals. Protein loading control was verified with 42 kDa β-actin. (E, F) SBDP immunoblots and quantified data show remarkably higher levels of 145 kDa calpain specific SBDP and 120 kDa caspase-3 specific SBDP in PD spinal cord compared to normals; *p ≤ 0.05.