Using a weight drop SCI model, the present study established the temporal and spatial profiles of motoneuron and glial cell loss after SCI, demonstrated that apoptosis is the common mechanism for different types of cell death after SCI, and characterized and quantitated apoptosis at different distances from the injury epicenter. The ability of the catalytic antioxidant MnTBAP and of the combination MnTBAP + L-NA to protect against SCI-induced cell loss and apoptosis were explored. The effectiveness of MnTBAP alone to attenuate neurological dysfunction was also evaluated.
Comparing our temporal and spatial profile of motoneuron loss following SCI in the present study (Figure 2A) with the profile of total neuron loss that we reported previously , we found that motoneuron loss had the same pattern as neuron loss: increase over time and decrease over distance from the epicenter. However, the significant loss of motoneurons occurred earlier and spread away from the epicenter quicker than total neuron loss. From 48 h until a week following SCI, motoneuron loss did not significantly differ by distance, indicating that most motoneuron loss occurred within this period. Although motoneurons are located in the ventral gray matter, farther from the impact site than total neurons, motoneuron loss was more severe than total neuron loss (Figure 2A). Our quantitative profile of motoneuron loss extended the information reported hitherto  from 24 h to one week.
Compared with the neuron and motoneuron loss in the ventral gray matter, our temporal and spatial profile of glial loss in the ventral white matter (Figure 2B) showed a different pattern: a delayed (starting at 24 h post-SCI), much diminished and more concentrated at the epicenter (at 0 - 1.5 mm from the epicenter) and quickly decreased (by 72 h post-SCI) due to proliferation. One report  found significant glia loss in the ventromedial white matter at 15 min after thoracic SCI. Our observation of significant glial loss starting at 24 h post-SCI may be attributed to the differences in the impact site, as Bresnahan et al.  indicated that a similar impact force produces shorter, more truncated lesion sites at lumbar levels with less involvement of the white matter than in thoracic lesions, as determined by the three-dimensional structure of lesion sites . The injury site in most previous studies of SCI was at vertebrae T9 [51, 63] or T10 [6, 13, 52–54]. Our profiles of cell loss and apoptosis were established from cord injured at vertebra T13, equal to lumber L2 of the cord within the lumbar enlargement. This added new information for a severe injury in the lumber enlargement where the hind limb motoneurons are located. Cell loss in the lumbar enlargement is more related to residual locomotor function than corresponding loss at other sites .
Apoptosis is characterized by changes such as cell surface blebbing, cell shrinking, formation of apoptotic bodies, nuclear chromatin condensation, compaction of the cytoplasmic organelles, nuclear lobation and a typical ladder-like electrophoretic pattern of DNA cleaved into nucleosome-size fragments of 180-200 bp or multiples thereof . DNA fragmentation can be observed morphologically by TUNEL staining and quantitated biochemically by ELISA. Numerous endogenous agents formed upon initial insult can cause apoptosis through different pathways, many of which converge at the caspase activation cascade [65, 66]. Active caspase-3 is the downstream executioner of the caspase cascade . Caspase-3 activation follows SCI [6, 68–70]. The discovery that some apoptotic cells do not necessarily go through DNA fragmentation and that cells with DNA fragmentation may die by necrosis led to the use of caspase-3 activation as another indicator of apoptosis. Therefore, in addition to characterizing DNA fragmentation by TUNEL, DNA laddering, and ELISA, this study also detected caspase-3 activation to further characterize apoptosis. Although apoptosis has been reported frequently after SCI, as cited in the Background section, few reports have qualitatively and quantitatively studied SCI-induced apoptosis in the lumbar enlargement.
In our previous study, using TUNEL staining as a marker of possible apoptosis, we obtained the temporal and spatial profile of total TUNEL-positive cells. Using TUNEL and NSE double staining, we also established the profile of TUNEL-positive neurons in the lumbar enlargement area after SCI . The present study, further morphologically characterized neuronal and glial apoptosis in the lumbar enlargement using the sections selected at the peak time and distance of apoptosis according to our previously established profiles. TUNEL or active caspase-3 as apoptosis markers were double stained with specific cell markers (NSE, ChAT and GFAP). The TUNEL-positive or active caspase-3-positive neurons, motoneurons and astrocytes in the double-stained sections in the ventral gray matter of injured spinal cords were observed at 24 h (for TUNEL + cell markers) and 12 h (for active fragment p20 of caspase-3 + cell marker) post-SCI (Figures 3 and 4), indicating apoptosis of neurons, motoneurons and astrocytes. Neuronal and glial apoptosis was confirmed by TEM, a gold standard to determine apoptosis based on its specific morphology. Figure 5 illustrates the different stages of neuronal and glial apoptosis by observation of their ultra-structural changes in the ventral gray matter in spinal cord sections removed at 24 h post-SCI, as described in the Results. Motoneuron apoptosis have been reported following compression SCI [69, 71], but few reports are available of motoneuron apoptosis after contusion SCI. Our characterization provides important evidence for motoneuron apoptosis after contusion SCI in the lumbar enlargement. Using TUNEL staining for DNA fragmentation and casepase-3 activation as two indicators of cellular apoptosis, our data support the notion that apoptosis is a common mechanism for neurons, motoneurons and glial cells to die following contusion SCI.
Apoptotic DNA fragmentation was also characterized biochemically in the spinal cord tissue by DNA laddering and quantitated by ELISA at 48 h (when the maximum number of TUNEL-positive cells appeared) post-SCI. Agarose gel electrophoresis and autoradiography indicated that the intensity of DNA fragment ladders was higher in the epicenter than 1 cm away from the epicenter, and no fragmented DNA bands were observed at 2 cm away from the epicenter or in the sham-operated sections. Compared with the size markers, these DNA fragments were confirmed to have characteristic oligonucleosome-length fragments at intervals of about 180 bp (Figure 6A). ELISA quantitation demonstrated that DNA fragmentation decreased with distance from the epicenter. The values in injured animals for each section were 2- to 10-fold that of the corresponding sections in sham-operated rats (Figure 6B). The decline in values over the distance from the epicenter as measured by ELISA was comparable to the decline in intensity of DNA laddering over the same distance. The fact that fragmentation was detected in tissue 2 cm from the epicenter and in the sham controls by ELISA, but no bands were observed in these samples by electrophoresis reflects the greater sensitivity of ELISA over DNA laddering. We also demonstrated that the caspase-3-like protease activity was significantly higher at the epicenter than 1 cm away from the epicenter or in the sham control (Figure 6C). Lack of caspase-3 activity at 1 cm from the epicenter may be attributed to the early measurement of caspase-3 activation at 4 h post-SCI.
Apoptosis is important in regulating normal development and maintaining tissue homeostasis in the adult. However, too much or too little apoptosis may result in pathological disorders. Therefore, there should be a basal level of apoptosis in the normal cord. This basal level was measured in the present study by ELISA in the sham-operated cord (Figure 6B). As mentioned in the Methods, to establish the temporal and spatial profile of cellular apoptosis, the TUNEL-positive cells were counted from 0 to 4 mm caudal from the epicenter at 1 h to 1 week post-SCI. At 4 mm, TUNEL positive cells appeared only at 24, 48 and 72 h post-SCI without significant differences from the sham control . The significantly higher level of DNA fragmentation measured by ELISA in the injured cord relative to the sham-operated cord was found even at 2 cm caudal from the epicenter. This indicates a greater sensitivity of ELISA, impossible by morphological observation of TUNEL-positive cells at such distance. Clearly, morphological examination of TUNEL-positive and caspase-3-positive neurons, motoneurons and astrocytes, and TEM identification are feasible for characterized apoptosis in different types of cells at the cellular level, whereas biochemical analysis of DNA fragmentation and caspase activation has great value in determining the spread of apoptosis from the epicenter. So the results of biochemical analysis provide a great addition to the morphological observation at longer distances from the epicenter.
We previously demonstrated that MnTBAP (i.p.) significantly reduced neuron death at sections 1–2.5 mm rostral and 1 mm caudal from the epicenter . The present study demonstrated that antioxidant treatment with a combination of MnTBAP + L-NA significantly reduced total cell and motoneuron loss in the gray matter of the cord and glial cell loss in the ventromedial white matter of the cord compared with vehicle-treated rats (Figure 7). Significant increases in the number of glial cells in the white matter by the combination treatment was closer to the epicenter (at sections 0 – 1.0 mm from the epicenter) compared with the increases of total cells (at sections 1.5 – 2.5 mm from the epicenter) and motoneurons (at sections 2.0 and 2.5 mm from the epicenter) in the gray matter. This is owing to the fact that most cells near the epicenter in the gray matter of the cord have been lost and cannot benefit from the treatment, whereas the glial loss was much slighter in the ventral white matter, so that many glia are still alive even in the epicenter, to be counted to evaluate the effectiveness of the treatment. Attenuation of total cell and motoneuron loss by the combination suggests that RS contribute to cell death after SCI.
We previously demonstrated that MnTBAP (i.p.) significantly reduced the TUNEL-positive neurons at 1 mm caudal from the epicenter . In the present study, we demonstrated that the combination of MnTBAP + L-NA significantly reduced the numbers of TUNEL-positive cells in the gray and white matter of the cord as compared with the vehicle treatment: an approximately 2.4 times reduction in the gray matter and approximately 10 times reduction in the white matter (Figure 8). The more effective reduction of TUNEL-positive cells in white matter compared with those in gray matter might be attributable to the much slighter cell loss in the ventral white matter as shown in Figure 7. We further demonstrated that MnTBAP alone (i.p.) significantly reduced the number of TUNEL-positive cells in the gray matter of the cord at sections 0, 1 and 2 mm rostral from the epicenter, with the number of TUNEL-positive cells in the vehicle-treated sections approximately 3, 2 and 2.5 times higher than those in the corresponding sections in the MnTBAP-treated group (Figure 9). The reports that treatment with antioxidants, RS scavengers and iron-chelators all reduce RS production, oxidative stress and apoptotic cell death after SCI links RS with apoptotic cell death in SCI [10–13, 69, 70, 72–75]. The impressive reduction in apoptosis by MnTBAP or its combination found in this study strongly supports the notion of the correlation of RS overproduction and apoptosis in SCI and provides further evidence of the causal relationship between RS overproduction and necrotic and apoptotic cell losses. The present results ― together with our previous finding that the combination or MnTBAP alone significantly reduces oxidation and nitration of proteins and MLP, cell death, and neurological deficits after SCI in rats [18, 21, 38–40, 43] ― support the sequence RS → oxidative stress → cell death → neuronal dysfunction.
Using behavioral tests, we recently reported that, with its lower ability than MP to penetrate the BSB, 10 mg/kg MnTBAP administered (i.p.) 4 h post-SCI followed by one half of the first dose 2 h later significantly increased BBB and inclined plane scores compared to vehicle treatment. Post-SCI treatment with MnTBAP is significantly more effective than MP for improving functional recovery after SCI . In the present study, we demonstrated that 15 min pre- and 6 h post-SCI treatment with MnTBAP significantly increased the scores of BBB and beam walk tests, and increased the angle of inclined plane compared with the vehicle-treated group. Our pre- and post-SCI treatment by MnTBAP strongly support the candidacy of MnTBAP for SCI treatment. However, the scores for all treatments in injured animals were significantly worse compared to sham controls (p < 0.001 for all), indicating that pharmaceutical treatments alone cannot improve function to normal levels.
The BBB test is a standard behavioral test for estimating the time course of recovery after SCI and for evaluating the efficiency of treatment. An injury force of 25 g.cm (10 g weight drop 2.5 cm down onto the exposed cord) was popularly used for the contusion injury, and the BBB scores of rats corresponding to this injury force are from 0 to 6 – 10 overtime [6, 13, 51]. In the present study, the contusion force was 12.5 g.cm and the BBB scores of rats were from 0 to approximately 15 overtime. A similar BBB scores (0 to approximately 14) was reported by Hyun et al.  using the 12.5 g.cm injury force on T9. Using the IH device, Springer et al.  reported that a 150 kilodyne injury force (which gives an injury comparable to that produced by a 12.5 g.cm weight drop with the NYU device) contused on T10, produced BBB scores of 0 to 12. Apparently, the BBB scores are correlated with the injury force and our results are comparable with other reports with similar injury force. This is also supported by our recent publication in which a 12.5 g.cm injury force was used on T10 .
We have demonstrated in vivo that RS administered into an uninjured rat spinal cord at a SCI-induced concentration and duration can directly oxidize major cellular components, thereby inducing necrotic and apoptotic cell death and neurological deficits, and that the removal of RS by MnTBAP significantly reduces the RS-induced damage [27–31] - directly and unequivocally demonstrating RS-induced oxidation to the major cellular components as an important pathway for cell death after SCI. Since the oxidative defense enzymes (such as superoxide dismutase, catalase, glutathione peroxidase, thioredoxin and others) are all susceptible to RS-induced oxidative damage, they may suffer function-distorting oxidative modification, in turn causing more RS production to form a negative feedback amplification of the oxidative damage. MnTBAP scavenges SCI-induced RS, protecting against oxidative damage to these enzymes and thereby restoring their enzymatic activities. This process reverses the negative to positive feedback amplification. RS can also cause injury by acting as intracellular death signals that lead to changes in the expression of proteins. RS may be active in different cell death pathways, including the caspase activation cascade. Therefore the removal of RS by MnTBAP reduces apoptotic cell death. RS can also act as modulators of the redox state, and the right dose of MnTBAP treatment to remove overproduced RS helps to maintain the redox balance, thereby avoiding further damage. Whether RS are the initiators or the intracellular messengers in cell death pathways after CNS injury, MnTBAP provides an opportunity for treatment by removing RS. In addition to catalytically scavenging a wide range of RS, other mechanisms may also contribute to the beneficial effects of metalloporphyrins, such as modulating RS-based redox signaling pathways and regulating cellular transcription activity . It has been reported that MnTBAP attenuated the nuclear translocation of apoptosis-inducing factor and the subsequent DNA fragmentation induced by ROS after permanent focal cerebral ischemia in mice . Therefore, in addition to its enzymatic catalytic scavenging activities, more complicated mechanisms may be involved in our in vivo finding that MnTBAP reduced cell death including apoptosis and improved functional recovery after SCI.