Keywords

1 Introduction

We have previously documented that a 40–50 % reduction in the transverse area of the cervical cord represents the critical threshold for favorable postoperative neurological recovery [1, 2]. Chronic compression of the spinal cord results in atrophy and loss of anterior horn cells associated with irreversible spinal cord damage [3–5]. Clinically, however, even in cases with severe compression detected on radiological examination, decompression may result in favorable recovery from a profound paresis. This unexpected outcome may be explained by the survival of a group of motoneurons even in the presence of considerable mechanical compression [1, 2].

In order to elucidate the mechanisms of neuronal reserve and survival of anterior horn cells, we used the tiptoe-walking Yoshimura (twy/twy) mouse, a model of spinal cord compression [6]. We examined morphological changes in spinal accessory motoneurons, expression of brain-derived neurotropic factor (BDNF) and neurotrophin (NT)-3 [7], and neuronal cell death [8], which have been used as markers of motoneuron survival and neuronal plasticity [9–11].

1.1 Topographic Morphology of Wheat Germ Agglutinin-Horseradish Peroxidase-Labeled Spinal Accessory Motoneurons

Representative photomicrographs showing WGA-HRP-labeled accessory motoneurons in the medial cell pool of the ICR and twy mice are shown in Fig. 9.1, and results of morphological analysis of these motoneurons are summarized in Table 9.1. At sites A and B of the twy mice, the mean area of the neuron soma was significantly larger than in ICR mice (P < 0.05), where the proportion of the number of WGA-HRP-labeled motoneurons at site C, the compression level, was significantly lower. The total length of dendrites at sites A and B was significantly longer than in the corresponding levels of ICR mice (P < 0.05).

Fig. 9.1
figure 1

Photomicrographs showing wheat germ agglutinin-horseradish peroxidase-labeled spinal accessory motoneurons. (a) A motoneuron at C2–C3 segments in a representative control ICR mouse; (b) site A in the twy mouse, the most rostral level; (c) site B, the level immediately rostral to the compressive lesion; (d) site C, the site of mechanical compression at C2–C3 cord segments. The top side in each panel represents the dorsal aspect of the cord. Scale bar = 50 μm (Reprinted, with permission, from [7])

Full size image
Table 9.1 Results of morphological measurements of neuronal soma and dendrites of the wheat germ agglutinin-horseradish peroxidase-labeled spinal accessory motoneurons in the medial cell pool of the twy and control ICR mice
Full size table

1.2 Three-Dimensional Computerized Display of Spinal Accessory Motoneurons

In the twy mice, at site B, the level immediately rostral to the compressive lesion, the elongated dendrites extended rostrally and ventrally as if the motoneuron avoided posterior compression (arrow, Fig. 9.2c). At site C dendrites were shorter (Fig. 9.2d)

Fig. 9.2
figure 2

Three-dimensional computer display of the dendritic arborization of the wheat germ agglutinin-horseradish peroxidase-labeled spinal accessory motoneurons (abscissa lateral direction, ordinate anteroposterior direction, and z-axis longitudinal direction). The compression force is applied in the x–y plane marked by arrows in (c) and (d). (a) Motoneuron at C2–C3 segment in the control ICR mouse; (b) site A in the twy mouse, the most rostral level; (c) site B, the level immediately rostral to the compressive lesion; (d) site C, the compression level at the C2–C3 cord segments (Reprinted, with permission, from [7])

Full size image

1.3 Localization of Brain-Derived Neurotropic Factor Immunoreactivity in the Anterior Gray Matter and Its Relationship to the Magnitude of Compression

At site B, BDNF immunoreactivity in the anterior horn cells of twy mice was significantly higher than in ICR and site A in twy mice (Fig. 9.3c). More severe compression at site C (TRAS < 50 %; n = 8) correlated with increased BDNF immunoreactivity in anterior horn cells at site B. Also, more labeled non-neuronal astroglia- and microglia-like cells were detected at sites B and C.

Fig. 9.3
figure 3

Photomicrographs showing localization of immunoreactivity of brain-derived neurotropic factor in the anterior gray horn of a representative ICR mouse (a) and twy mouse (b–d) with a transverse remnant area of 60 %. (b) Site A; (c) site B; (d) site C. (a′, d′) Negative controls for ICR and twy mice, respectively. Top posterior aspect of the spinal cord. Scale bars = 50 μm (Reprinted, with permission, from [7])

Full size image

1.4 Localization of Neurotrophin-3 Immunoreactivity in the Anterior Gray Horn and Its Relationship to the Magnitude of Compression

At site B in twy mice, immunostaining for NT-3 V4 antibody was significant in the anterior horn cells, with enhanced immunoreactivity in the neuron soma and dendrites (Fig. 9.4c) compared with those seen at site A in twy and ICR mice.

Fig. 9.4
figure 4

Photomicrographs showing localization of immunoreactivity of neurotrophin-3 in a representative control ICR mouse (a) and twy mouse with a transverse remnant area of 55 % (b–d). (b) Site A; (c) site B; (d) site C. (a′, d′) Negative controls for ICR and twy mice, respectively. Top, dorsal aspect of the cord. Scale bars = 50 μm (Reprinted, with permission, from [7])

Full size image

1.5 Topographic and Quantitative Analysis of Brain-Derived Neurotropic Factor and Neurotrophin-3 Immunoreactivity

BDNF was detected as a band with a molecular weight of 17–19 kDa (Fig. 9.5a) and NT-3 as a 30 kDa band (Fig. 9.5b). At site B in twy mice, BDNF and NT-3 expression was 70.7 % (lane 6 in Fig. 9.5a) and 72.7 % (lane 6 in Fig. 9.5b), respectively (P < 0.05) when compared with that at site C. Immunoreactivity of both neurotrophins increased significantly at the level distal to the compressed segment (lane 8).

Fig. 9.5
figure 5

Photomicrograph showing the expression of brain-derived neurotropic factor (a) and neurotrophin-3 (b) by Western blot analysis. The lane number shows each level of the spinal cord in the ICR (lanes 1–4) and twy (lanes 5–8) mice; CA refers to the hippocampus. Lanes 1 and 5 in site A, the level most rostral to the C1 ventral root; lanes 2 and 6 in site B, the level immediately rostral to compression between C1 ventral and C2 dorsal roots; lanes 3 and 7 site C, between C2 and C3 dorsal roots (the compressed level in the twy mouse); lanes 4 and 8, the level between C3 and C4 dorsal roots (Reprinted, with permission, from [7])

Full size image

1.6 TUNEL Staining in the twy/twy Mouse Spinal Cord

Figure 9.6 shows the topographic distribution of TUNEL-positive cells in the chronically compressed spinal cord of moderately and severely compressed twy/twy mouse examined by the TUNEL method. No TUNEL-positive cells were identified in both the gray and white matters of the control ICR mouse spinal cord. In contrast, TUNEL-positive cells were found in the gray and white matter in the twy/twy mouse with moderate compression. In comparison, fewer TUNEL-positive cells are found in the severe compression group particularly in the anterior horn, though these cells were abundant particularly in the anterior columns.

Fig. 9.6
figure 6

Photomicrographs of terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nick-end labeling (TUNEL) staining in representative control ICR mice (upper photos) and twy/twy mice with moderate compression (middle photos) or severe compression (lower photos). Left column: photomicrographs taken by a roupe (×3). The white rectangular area (anterior horn of the gray matter) and black rectangular area (anterior column of the white matter) are enlarged in the same row in the middle and right columns, respectively. Black arrows: representative TUNEL-positive cells. TUNEL-positive cells in the gray and white matter are noted in twy/twy mouse. Scale bars = 50 μm

Full size image

2 Discussion

Using the twy mice, we have previously demonstrated a significant reduction in the number of anterior horn cells at the level of mechanical compression [6, 7]. A progressively smaller number of these cells was noted when TRAS was 70 % of the normal value, reaching an asymptote approximately below 50 %. Interestingly, we also found increased population of spinal accessory motoneurons at levels rostral to the C1 vertebra (sites free of direct mechanical compression) [12]. The latter finding led us to speculate that a group of motoneurons may translocate rostrally to avoid chronic mechanical compression. Rostral to the site of mechanical compression (site B), we found enlargement of neuron soma and dendritic elongation of the WGA-HRP-labeled accessory motoneurons, which correlated significantly with deterioration of TRAS values measured at the site of compression. Somal size and dendritic arborization increase in order to facilitate synaptic connection [13] representing enhanced synaptic circuits [14, 15] to compensate the compromised function of spinal cord motoneurons.

Nerve growth factors, such as BDNF and NT-3, promote development of neurite outgrowth, growth, and differentiation of neurons, as well as their survival [16]. Our study found increased BDNF and NT-3 immunoreactivities in anterior horn cells, in close association with increased neuronal density of spinal accessory motoneurons when TRAS ranged between 50 % and 70 %. We postulate that considerable amounts of neuropeptides are transported retrogradely [17–19] and/or from the brain [20] in order to increase neuronal activity as well as axonal transport. It is known that astroglial cells produce BDNF [21] and other nerve growth factors [22]. In twy mice, when TRAS was 55 % of the control, a number of infiltrated astroglial cells, positive for BDNF immunostaining, were detected at sites rostral to the site of compression, related to increased number of motoneurons: another possible mechanism to enhance neuronal repair and regeneration. The present study demonstrated increased numbers of TUNEL-positive neurons in the gray matter and oligodendrocytes in the white matter of the spinal cord of the twy/twy mouse with progressive mechanical compression, which probably contributes to axonal degeneration and demyelination in the twy/twy mouse spinal cord with severe compression [8].

3 Conclusion

WGA-HRP-labeled accessory motoneurons expressed significant BDNF and NT-3 immunoreactivities that correlated with enlargement of neuronal soma and extensive dendritic arborization at levels rostral to the site of mechanical compression. Astroglial cells at these levels may play a role in neuronal reserve and survival. There is a significant correlation between the proportion of apoptotic neurons and oligodendrocytes in the compressed area of the spinal cord and the magnitude of cord compression.