We evaluated by in situ nick end labeling the presence of apoptotic glial cells in the spinal cord of rats which have sustained a moderate and severe compression injury at the level of T8-9, resulting in a severe but reversible paraparesis and irreversible paraplegia, respectively. In a previous investigation we found apoptotic glial cells (oligodendrocytes) in the immediate vicinity of the primary lesion (T7 and T10). The present study was designed to evaluate the extent of such cells in the spinal cord even at long distances away from the primary injury. Rats sustaining a moderate and severe compression injury and surviving 4 and 9 days showed a significant increase in the number of apoptotic glial cells at the T1, T5, T7, T12 and L2 levels. At the T10 level the elevation was significant only after day 9. There was no significant increase in the number of these cells at 4 h and 1 day after moderate and severe compression. In general, the apoptotic cells were most often seen in segments adjacent to the compression. They were randomly located in the ventral, lateral and dorsal tracts but were rarely present in the gray matter of the cord. In conclusion, compression trauma to rat spinal cord induces signs of apoptosis in glial cells, presumably oligodendrocytes of the long tracts. This newly discovered type of secondary injury is widely distributed in the damaged spinal cord and occurs even at long distances remote from the initial compression injury. Apoptotic cell death of oligodendrocytes will induce myelin degeneration and cause additional disturbances of axonal function. This cell damage may be a target for future therapy since it occurs after a delay and chemical compounds are now available by which apoptotic cell death can be modified.

We evaluated by immunohistochemistry the presence of beta-amyloid precursor protein (beta APP) and ubiquitin-like material which may accumulate in axons of the human spinal cord subjected to injury. Autopsy material was obtained from nine cases with different types of trauma: breech delivery with neonatal spinal injury, compression of the cord induced by fractures of the vertebral column, haematomas or intradural meningioma. The posttrauma period ranged from 10 days to several years. The spinal cord of six control cases without evidence of injury presented beta APP immunoreactivity in nerve cell bodies and in a few axonal profiles but not in dendrites. Seven of the nine cases with spinal cord trauma showed an accumulation of beta APP-immunoreactive material in axons of the longitudinal tracts at the site of the injury. Five cases presented similar axonal immunoreactivity in the grey matter of the cord. Ubiquitin-like immunoreactivity was present in expanded axons in cases with spinal cord injury. Cases with spinal cord trauma thus present beta APP-immunoreactive axons particularly of the longitudinal tracts in the same way as in trauma to rat spinal cord and in various brain injuries. The aggregation of beta APP-immunoreactive material indicates disturbed axonal transport of beta APP. Accumulation of ubiquitin-like immunoreactive material in expanded axons at the site of trauma may be one prerequisite for degradation of abnormal proteins by the ubiquitin-mediated proteolytic pathway.

Growth-associated protein 43 (GAP43) is one compound used to indicate growth of axonal endings during development and regeneration, particularly of peripheral neurons. Using immunohistochemistry, we have studied the expression of GAP43 in the spinal cord of rats subjected to mild, moderate or severe compression injury and used neurofilament immunostaining to demonstrate axonal injuries. Samples removed from the compressed T8-9, the cranial T7 and the caudal T10 segments were studied at 4 h, 24 h, 4 days and 9 days after injury. Control rats showed a moderate immunostaining of neurons in dorsal root ganglia, weak staining of ventral motor neurons and, with the exception of the corticospinal tracts, a weak staining in some axons of the longitudinal tracts of the cord. Injury in the compressed region led to increased GAP43 immunoreactivity in axons of normal and expanded size. This occurred particularly 1-4 days after injury and normalized 9 days thereafter. More marked immunostaining was present in the cranial and caudal segments. The corticospinal tracts never showed such staining. The increase of GAP43 immunostaining is presumably caused by disturbed axonal transport from neurons with the capacity to synthesize and transport the GAP43 antigen. Transported material may thus be available for regeneration of axons, but this source of material may vary between different classes of axons within the cord.

This immunohistochemical study evaluated Fas and Fas ligand (FasL) in the rat nervous system and their changes in the spinal cord subjected to compression. Normal spinal cord showed a low level of Fas and FasL immunoreactivity in the white matter except in the corticospinal tracts. Fas and FasL immunoreactivity seemed to be located in axons and their myelin sheaths. Other regions of the nervous system did not show immunoreactivity to Fas and FasL. Moderate and severe compression injury of the spinal cord resulted in a reduction of Fas and FasL immunoreactivity in the white matter of injured T8-9 segments at 4 h and a complete loss at 1 day after trauma. This was seen even in the remaining white matter. In contrast, increased immunoreactivity to Fas and FasL was present in the cranial T7, caudal T10 (moderate injury) and T12 (severe injury) segments at day 4 with most intense staining were seen at day 9 after trauma. Increased Fas and FasL immunoreactivity may have pathophysiological implications for the development of secondary injuries after trauma to the spinal cord. Fas-FasL interactions may for instance be involved in apoptosis of oligodendrocytes which occurs as a delayed phenomenon after trauma to the spinal cord. The integrity of myelin sheaths may in this way be jeopardized by apoptosis of oligodendrocytes.

The ubiquitin-mediated proteolytic pathway is an important mode of protein degradation in various tissues. Since breakdown of proteins may occur in axons after injury we evaluated the presence of ubiquitin-like immunoreactive material in rat spinal cord following compression injury of mild, moderate and severe degrees at T8-9 level, resulting in no neurological deficit, reversible paraparesis and paraplegia of the hind limbs, respectively. Rats with mild to severe compression injury surviving 1-4 days showed numerous, intensely immunoreactive expanded axons at the site of compression. The labelled axons were randomly distributed in the longitudinal tracts but they were never found in the corticospinal tracts. No labelling was detected by 9 days after injury. In addition, the presence of labelled axons was investigated in the T7 and the T10 segments from rats with moderate compression. No labelling was seen in T7, but in T10 segments many immunoreactive axons were present. Control rats did not show immunoreactive axons in the spinal cord. Neurons of dorsal root ganglia, trigeminal ganglia and of the grey matter of the spinal cord were immunoreactive. Cerebral cortical neurons did not show ubiquitin expression. Thus, compression of the rat spinal cord causes a transient accumulation of ubiquitin-like immunoreactive material in axonal swellings. Even though the dynamics of ubiquitin conjugates are not fully understood, the observed axonal accumulation presumably reflects arrested anterograde axonal transport of protein chiefly derived from neurons of dorsal root ganglia and the local neurons of the spinal cord. The presence of ubiquitin in damaged axons is one prerequisite for degradation of abnormal proteins by the ubiquitin-mediated proteolytic pathway, which may be activated in reactive axonal swellings.