Natural Progression of Spinal Cord Transection Injury and Reorganization of Neural Pathways

Abstract

The spinal cord injury (SCI) transection model accurately represents traumatic laceration and has been widely used to study the natural history and reorganization of neuropathways and plasticity in the central nervous system (CNS). This model is highly reproducible, which makes it ideal for studying the progression of injury as well as endogenous recovery and plasticity in the CNS. Five experimental groups of transection injury were designed: left hemitransection; right hemitransection; double hemitransection; complete transection injuries; and laminectomy-only control. We used somatosensory evoked potentials (SSEPs) as an objective electrophysiological assessment tool and motor behavior testing (Basso, Beattie, and Bresnahan [BBB] scoring) to functionally assess the neural pathways post-injury. Histological examinations were carried out to investigate the extent of injury and spinal cord morphological changes. Significant (p < 0.05) electrophysiological changes were observed and were verified by an increase in SSEP amplitude in somatosensory cortices for all four injury groups during days 4 and 7 post-injury. Degree of plasticity among the groups was distinguished by changes in SSEP amplitude and BBB scores. Our results support our previous published findings (using a contusive model of SCI), which shows that the reorganization of neuropathways and plasticity persist in time and are not transient phenomena. SSEPs are a reliable tool to assess the functionality of neural pathways and their projections to higher CNS structures such as the cortices. They enable us to determine residual function and the changes within the CNS post-injury and consistently track these events over time. The results from our study provide supporting evidence for the presence of neuronal network reorganization and plasticity in the CNS after transection SCI.

Introduction

Spinal cord injury (SCI) leads to the disconnection between the brain, spinal cord, and peripheral nerves, resulting in paralysis below the injury. Evidence from recent studies indicates that the large-scale functional (plasticity) and neuroanatomical (reorganization) changes in the central nervous system (CNS), which occur after SCI, are caused by the damage to axons and neural pathways, which consequently disrupts neuronal networks.^1
–3 Such neuroanatomical reorganization has also been reported to constitute mechanisms of functional reorganization and appears to play a critical role in rehabilitation and functional recovery after injury to the CNS.^4

–7 For instance, short-term plasticity has been demonstrated in patients with incomplete SCI, resulting in the increased sensitivity of the somatosensory cortex and corticospinal tracts.⁸ These phenomena can be measured through the electrical activity of specific neurons within their representative area of the cortex, especially during the very acute phase of injury, where their activity is more prominent.⁹ However, the functional adaptations that occur as a consequence of plasticity and neuropathway reorganization along the spinal cord and within the brain are still poorly understood. Further investigation into post-SCI compensatory mechanisms will be important in enabling the design of better therapeutic strategies for neurological recovery.

Plasticity is often described as the dynamic potential of the CNS to reorganize and reflects the nonstatic nature of neuronal circuits.^6,10 Experience-dependent plasticity is considered the basis of behavioral changes, altering extended neural networks and local synapses in response to environmental stimuli and pathological conditions.¹¹ This provides neurons—which may have been either part of existing or inactive circuitry—with new response properties and can contribute to long-term cortical reorganization.^6,12 Such neuroplasticity can refer to the reorganization of neural pathways or alterations in neural networks. Anatomically, new synapse formation and synaptic rearrangement occur in both the cortical and subcortical regions in response to a SCI. The spinal cord also adapts through the collateral sprouting of spared axons across the injury site.^4,13 With regard to plasticity in neural networks, several studies have shown the adaptive expansion of cortical representations into areas rendered nonfunctional because of corresponding injuries.^5,14

Electrophysiology studies suggest that network plasticity likely occurs because of the close association between the endogenous compensation and recovery mechanisms between the spinal cord and the brain.¹⁰ In our previous studies, we reported a significant increase in somatosensory evoked potential (SSEP) response from both forelimb cortices (contralateral and ipsilateral to injury) after a T8 moderate unilateral contusion SCI.¹⁵ We also identified increased oxygenation in the ipsilateral cortex post-injury using BOLD-fMRI (blood-oxygen-level–dependent/functional magnetic resonance imaging).⁹ These time-dependent changes within the brain appear to be indicative of cortical adaptation post-SCI. These could be described as either an increased connectivity between the left and right hemispheres or the formation of new intraspinal circuits. Additionally, Ghosh and colleagues⁵ have reported that axons originating from the ipsilateral cortex sprout across the site of transection, which consequently innervate spinal sections below the injury and create new ipsilateral forelimb representation in the cortex. Moreover, an enhanced representation of forelimb sensory neurons into the adjacent hindlimb cortex was confirmed through BOLD-fMRI, retrograde and anterograde tracing.

In this study, we aimed to investigate the natural progression of spinal cord transection injury and illustrate the reorganization of neural pathways in a rodent model. This experiment was designed to monitor the electrophysiological changes related to axons, as well as the reorganization of neural pathways and plasticity within the CNS. Four groups of transection injury were investigated: left T8 hemitransection (group 1); right T10 hemitransection (group 2); left T8 and right T10 double hemitransection (group 3); and T8 complete transection (group 4). We compared single hemitransection injuries between the first and second injury groups to investigate the existence of a similar compensatory response in the cortex. To further enable us to compare these compensatory responses, we also designed more severe double hemitransection and complete transection injury groups 3 and 4.

Methods

Animals

Female adult Sprague-Dawley (200–225 g) rats were used. We designed four groups (n = 5 in each group) of experimental injuries: left T8 hemisection; right T10 hemisection; T8 complete transection; left T8 and right T10 double hemisection; and one control group (n = 5; laminectomy without injury). All procedures reported in this experiment were approved by the Institutional Animal Care and Use Committee at the National University of Singapore.

Electrode implantation

To induce general anesthesia, rats were intraperitoneally anesthetized with a mixture of ketamine (75 mg/kg) and xylazine (10 mg/kg) at 0.2 mL/100 g. A midline incision through the skin of the head was made and the skull was exposed and cleaned. Animals were implanted with five transcranial stainless steel screw electrodes (VA E363/20/SPC; Plastics One, Inc., Roanoke, VA), corresponding to right and left forelimb and hindlimb somatosensory cortices, such that they made light contact with the dura mater without compressing the brain. A micro drill designed for research applications (MD Ideal Micro drill; Cellpoint Scientific Inc., Gaithersburg, MD) was used to drill four burr holes at 2.5 mm posterior and 2.8 mm lateral to the bregma for hindlimbs and 0.2 mm posterior and 3.8 mm lateral to the bregma for forelimbs. A fifth hole was made in the right hemisphere at 3.0 mm lateral to lambda for reference. Carboxylate dental cement (IL Jet Denture Repair Package; Lang Dental Manufacturing Co., Inc., Wheeling, IL) was subsequently applied to hold the screw electrodes in place and seal the exposed skull. Electrode connections were inserted into a pedestal to enable recording of long-term SSEPs precisely from the same area.

Injury

General anesthesia as reported above was used during all surgical procedures. Limb withdrawal to pinch stimuli and the corneal reflex were monitored in order to ascertain that the animal was adequately anesthetized. Laminectomy was performed to expose the dorsal surface of the spinal cord. Care was taken to maintain the integrity of the dura mater. A size 11 stainless steel disposable scalpel (Swann-Morton, Sheffield, UK) was used to carry out the transection injuries. For the hemitransections, the midline on the spinal cord dorsal pathway of the spinal cord was identified. The scalpel fully penetrated perpendicularly and moved laterally to induce injury on either the right or the left or a combination of the two for double hemitransection. The full transection was achieved by cutting the segment of the spinal cord transversely from left to right. A Nikon microscope (SMZ745 T; Nikon Corporation, Tokyo, Japan) was used to ensure consistency in inducing the injury among all rats within the same group. Rats in the control group underwent laminectomy only.

Post-operative care

Analgesic buprenorphine (0.06 mg/kg) and antibiotic gentamicin (8 mg/kg) at 0.2 mL/100 g volume were administered subcutaneously twice a day for 5 days after transection injury. Animals' bladders were expressed twice-daily until rats regained their ability to urinate.

Electrophysiological recording

SSEP baseline recordings were carried out 1 week after electrode implantation using our electrophysiological setup. This setup included a Tucker-Davis Technologies (TDT; Tucker-Davis Technologies Inc., Alachua, FL) workstation, which comprised a head-stage amplifier (RA4LI), a pre-amplifier (RA4PA), and a Bioamp processor (RZ5). An isolated current stimulator (Letchworth DS3; Digitimer Ltd., Welwyn Garden City, UK) was used with two stainless steel subdermal needle electrodes (RI Safelead F-E3-48; Grass Technologies, West Warwick, RI) to deliver single stimulus pulses of 3.5 mA and 200 μsec. Needle electrodes were inserted into the right forelimb of rat subjects to stimulate the right median nerve. The stimulator was then individually triggered using a train of 150 transistor-transistor logic (TTL) pulses from the Bioamp processor at 0.5 Hz. Simultaneously, SSEP data from skull electrodes were collected through the TDT workstation in 1-sec epochs at 4882-Hz sampling rate, synchronized to the TTL trigger pulse. Subdermal needle electrodes were then swapped to the left forelimb, and the stimulation and recording procedure was repeated. In total, each forelimb received 150 consecutive positive monophasic current pulses, with a corresponding 150 samples of SSEPs of 1-sec length collected from the right and left somatosensory cortices. The above procedure used to collect SSEP baselines was also used to collect SSEPs on days 4 and 7 post-injury.

For SSEP recording, we used a mixture of 1.5% isoflurane (Singapore Aerrane Isoflurane; Baxter Healthcare [Asia], Singapore) and 98% oxygen given at a flow rate of 1.3 L/min using a Patterson Scientific Versa II isoflurane vaporizer (Patterson Scientific, Foster City, CA). Anesthesia was maintained through a rodent-sized anesthesia mask, connected to a diaphragm with a C-pram circuit designed to administer and evacuate the gas. Rats were moved onto a heating pad within a Faraday cage to maintain the temperature of the animals at 37 ± 0.5°C during recording.

Basso, Beattie, and Bresnahan motor behavioral assessment

We used a 21 score Basso, Beattie and Bresnahan (BBB) locomotion rating scale to assess the motor behavior of rats pre- and post-injury. Rats were placed in a 90-cm plastic open field for 4 min. Two blinded examiners proficient in the method evaluated and scored animals individually on days 4, 7, 14, and 21 (D4, D7, D14, and D21) post-injury. Motor behavior assessment preceded SSEP recording on D4 and D7. Further details can be found in studies by Basso and colleagues,^16,17 Agrawal and colleagues,¹⁸ and Bazley and colleagues.¹⁹

Somatosensory evoked potential analysis

From the recorded raw SSEPs for each limb stimuli (in all the groups), we considered only sweeps with ≥50% energy of the maximal energy of overall sweeps. After this, we averaged the SSEPs for the baseline, D4, and D7 signals. Consequently, we determined the first negative (N1) and second positive (P2) peaks. The energy of baseline and post-injury signals were determined by summing the amplitude square of all samples in the N1–P2 segment. An example of an averaged SSEP signal is presented in Figure 1A. The same process was used to compute the corresponding baseline signals pre-injury for comparison. Figure 1B shows a comparison of the averaged SSEPs for baseline, D4, and D7 for one rat in group 1.

FIG. 1.

An example of somatosensory evoked potentials (SSEPs). (A) Stacked pile plot of single sweeps are shown on the left and the first negative (N1) and second positive (P2) peaks are shown on the right. We developed an averaged SSEP signal from the single sweeps and determined the SSEP amplitude, which is indicated by the vertical dotted line. (B) An example of a baseline, as well as day 4 (D4) and day 7 (D7) recordings, are given for a rat in group 1 (left hemitransection). Signals represent recordings from the right hemisphere after left forelimb stimulation.

Histology

At the end of the observation period, spinal cords were harvested from control and injured rats. Animals were deeply anesthetized and perfused with freshly prepared 4% paraformaldehyde (PFA). After perfusion, spinal cords were carefully dissected, post-fixed in 4% PFA, and dehydrated overnight in 30% sucrose at 4°C. We used a Nikon microscope (SMZ745 T) to capture images of the harvested segments at 1 × magnification to show the gross structure of the spinal cords similar to what has been reported previously.²⁰ The thoracic region of spinal cords were embedded in OCT^© and sectioned longitudinally at 10-μm thickness using a cryostat. Spinal cord sections were stained with hematoxylin and eosin (H&E) and visualized with a Leica (CLS150X; Leica Microsystems, Wetzlar, Germany) bright-field microscope at 5 × magnification to observe the location and extent of transection injuries.

Statistical analysis

For SSEPs, energies of the signals recorded post-injury were statistically compared to baseline to determine the relative variation of SSEP signals across the post-injury period^21
–23 using the Student's t-test. The t-test compared D4 to baseline and D7 to baseline. It was performed using the null hypothesis that data from D4 to D7 and baseline are independent random samples from normal distributions with equal means and equal, but unknown, variances against the alternative that the means are not equal. For this test, p values <0.05 were considered statistically significant. For the BBB assessment, we first statistically compared the BBB scores of the left hindlimb and the right hindlimb for each of the four groups using the Student's t-test. Next, we also statistically compared the motor behavior scores of the same hindlimbs for groups 1 and 2, followed by the scores from the same hindlimb for groups 3 and 4. For these comparisons, p values <0.05 were considered statistically significant.

Results

Our experimental groups are as follows: left hemitransection injury at T8 for group 1; right hemitransection injury at T10 for group 2; left and right double hemitransection injury at T8 and T10 for group 3; and complete transection injury at T8 for group 4. The control group received only laminectomy (no injury) at T8.

Somatosensory evoked potentials

We used SSEPs as an objective assessment of neuronal responses in the somatosensory cortex to forelimb electric stimulation.^24

–27 SSEPs were recorded from both left and right somatosensory cortices, with animals undergoing both left and right forelimb stimulation pre-injury as a baseline and then on D4 and D7 post-injury. SSEPs from the experimental groups are presented in Tables 1 –4 and Figures 2 –5. The question mark (?) in the body of Figures 2 –5 corresponds to the cortical area being investigated. We monitored the changes to SSEPs related to axonal reorganization at and around the site of transection, as well as to plasticity in the cortices.

FIG. 2.

Amplitude at days 4 (D4) and 7 (D7) post-injury compared to the normalized baseline (dashed line) in animals that underwent T8 left hemitransection. (A) Amplitude changes in the right (contralesional) forelimb somatosensory cortex after left forelimb stimulation. (B) Amplitude changes in the left (ipsilesional) forelimb somatosensory cortex after left forelimb stimulation. (C) Amplitude changes in the right (contralesional) hindlimb somatosensory cortex after left forelimb stimulation. (D) Dorsal view of T8 left spinal cord hemitransection injury and schematic of the stimulation and recording paradigm for group 1. Question mark (?) symbols correspond to the area being investigated. Error bars correspond to one standard deviation.

FIG. 3.

Amplitude at days 4 (D4) and 7 (D7) post-injury compared to the normalized baseline (dashed line) in animals that underwent right T10 hemitransection. (A) Amplitude changes in the left (contralesional) forelimb somatosensory cortex after right forelimb stimulation. (B) Amplitude changes in the right (ipsilesional) forelimb somatosensory cortex after right forelimb stimulation. (C) Amplitude changes in the left (contralesional) hindlimb somatosensory cortex after right forelimb stimulation. (D) Dorsal view of T10 right hemitransection and schematic of the stimulation and recording paradigm for group 2. Question mark (?) symbols correspond to the area being investigated. Error bars correspond to one standard deviation.

FIG. 4.

Amplitude at days 4 (D4) and 7 (D7) post-injury compared to the normalized baseline (dashed line) in animals that underwent double hemitransection (T8 and T10). (A) Average amplitude changes against baseline in the contralateral forelimb somatosensory cortex after separate left and right forelimb stimulations, with somatosensory evoked potentials recorded from the corresponding contralateral forelimb somatosensory cortex. (B) Average amplitude changes after separate left (LF) and right forelimb (RF) stimulations, but recording in the corresponding contralateral hindlimb somatosensory cortex. (C) Dorsal view of T8 left transection and T10 right hemitransection and schematic of the stimulation and recording paradigm for group 3. RF and LF stimulation and recording was carried out separately. Question mark (?) symbols correspond to the area being investigated. Error bars correspond to one standard deviation.

FIG. 5.

Amplitude at days 4 (D4) and 7 (D7) post-injury compared to the normalized baseline (dashed line) in animals that underwent complete transection at T8. (A) Average amplitude changes against baseline in the contralateral forelimb somatosensory cortex after separate left (LF) and right forelimb (RF) stimulations, with somatosensory evoked potentials recorded from the corresponding contralateral forelimb somatosensory cortex. (B) Average amplitude changes after separate LF and RF stimulations, but recording in the corresponding contralateral hindlimb somatosensory cortex. (C) Dorsal view of T8 complete transection and schematic of the stimulation and recording paradigm for group 4. RF and LF stimulation and recording was carried out separately. Question mark (?) symbols correspond to the area being investigated. Error bars correspond to one standard deviation.

Table 1.

Left Hemitransection (T8), with the Left Forelimb (FL) Stimulated

Day	Comparison	Percent amplitude increase from baseline	p value	Standard deviation
4	LF stimulation/ Cont. FL response	124.40	0.004^*	47.40
7	LF stimulation/ Cont. FL response	124.20	0.006^*	53.04
4	LF stimulation/ Ipsi. FL response	170.39	0.028^*	113.55
7	LF stimulation/ Ipsi. FL response	50.55	0.081	39.18
4	LF stimulation Cont. HL response	113.81	0.011^*	57.43
7	LF stimulation Cont. HL response	103.72	0.007^*	31.62

Contralateral (Cont.) forelimb and hindlimb (HL) somatosensory cortex showed a significant increase in somatosensory evoked potential (SSEP) amplitude (compared to baseline) at post-injury days 4 and 7. Ipsilateral (Ipsi.) forelimb somatosensory cortex showed a significant SSEP amplitude increase at post-injury day 4. ^* p < 0.05.

Table 2.

Right Hemitransection (T10), with the Right Forelimb (FL) Was Stimulated

Day	Comparison	Percent amplitude increase from baseline	p value against baseline	Standard deviation
4	RF stimulation/ Cont. FL response	127.60	0.012^*	65.19
7	RF stimulation/ Cont. FL response	186.00	0.003^*	62.29
4	RF stimulation/ Ipsi. FL response	188.66	0.030^*	128.17
7	RF stimulation/ Ipsi. FL response	314.89	0.002^*	104.02
4	RF sti ulation/ Cont. HL response	74.12	0.015^*	29.10
7	RF stimulation/ Cont. HL response	164.12	0.001^*	42.83

Contralateral (Cont.) forelimb, hindlimb (HL), and ipsilateral (Ipsi.) forelimb somatosensory cortex showed a significant increase in somatosensory evoked potential amplitude (compared to baseline) at post-injury days 4 and 7. ^* p < 0.05.

Table 3.

Double Hemitransection Group (T8 and T10), with Left and Right Forelimbs (FL) Stimulated

Day	Comparison	Percent amplitude increase from baseline	p value against baseline	Standard deviation
4	RF stimulation/ Cont. FL response	129.50	0.012^*	47.72
7	RF stimulation/ Cont. FL response	212.00	0.042^*	60.04
4	LF stimulation/ Cont. FL response	143.33	0.009^*	23.09
7	LF stimulation/ Cont. FL response	164.67	0.021^*	128.72
4	RF stimulation/ Cont. HL response	89.05	0.029^*	44.99
7	RF stimulation/ Cont. HL response	253.05	0.018^*	145.49
4	LF stimulation/ Cont. HL response	140.30	0.015^*	77.08
7	LF stimulation/ Cont. HL response	268.15	0.015^*	147.08

Contralateral (Cont.) forelimb and hindlimb (HL) somatosensory cortex showed a significant increase in somatosensory evoked potential amplitude (compared to baseline) at post-injury days 4 and 7. ^* p < 0.05.

Table 4.

Complete Transection Was Carried Out at T8 and Left and Right Forelimbs (FL) Were Stimulated

Day	Comparison	Percent amplitude increase from baseline	p value against baseline	Standard deviation
4	RF stimulation/ Cont. FL response	212.00	0.003^*	70.50
7	RF stimulation/ Cont. FL response	136.67	0.049^*	15.28
4	LF stimulation/ Cont. FL response	176.67	0.011^*	32.15
7	LF stimulation/ Cont. FL response	323.60	0.025^*	207.64
4	RF stimulation/ Cont. HL response	164.37	0.000^*	31.06
7	RF stimulation/ Cont. HL response	195.37	0.002^*	63.75
4	LF stimulation/ Cont. HL response	176.39	0.001^*	31.27
7	LF stimulation/ Cont. HL response	168.89	0.004^*	17.56

Comparisons of forelimb SSEPs between D4 and D7 post-injury against baseline showed significant (p < 0.05) differences in amplitude in all groups. A significant increase in the amplitude of the corresponding contralateral forelimb cortex after ipsilesional forelimb stimulation was observed in hemitransection groups 1 and 2 as seen in Tables 1 and 2 and Figures 2 and 3 (panels A–C), respectively. Moreover, the response in the contralateral hindlimb somatosensory cortex to ipsilesional forelimb stimulation was also observed to be significantly increased despite the presence of injury in the spinal cord. An increase in contralateral hindlimb and forelimb SSEP amplitude was observed in animals that underwent double hemitransection at T8 and T10 (Table 3; Fig. 4A–C). Similarly, in the group of animals that underwent a complete transection at T8, there was a significant increase in both contralateral hindlimb and forelimb somatosensory cortex amplitude after left and right forelimb stimulation at D4 and D7 post-injury (Table 4; Fig. 5A–C). The presence of the increased amplitude of SSEPs recording from the hindlimb somatosensory cortex after forelimb stimulation suggests the possible increase in forelimb cortical representation into the hindlimb cortex.

Motor behavior

Our motor behavior results (Fig. 6) show the mean BBB scores for the four experimental groups and one control over 21 days. BBB scores are rated from 0 to 21, with 0 representing no joint movement, locomotion, and stepping ability and 21 being normal locomotion. All rats had a score of 21 pre-injury, as noted at baselines 1 and 2 (B1 and B2). Hemitransection groups (groups 1 and 2) showed scores similar to the “late-stage recovery,” double hemitransection (group 3) developed BBB scores in the “intermediate recovery” range, and the BBB score of complete transection group (group 4) remained in the “early recovery” stage. Groups of injury were analyzed statistically for differences in motor behavior between the left and right hindlimbs. Groups 1, 2, and 3 did not show statistically significant differences between the left and right hindlimbs from D4 to D21. Results showed statistically significant differences between left and right hindlimbs for group 4 (p = 0.034) on D4, but not for D7–D21. We also analyzed the statistical significance between injury groups using both right hindlimbs and left hindlimbs. There was not enough evidence that there was a significant difference between groups 1 and 2 when comparing their respective hindlimbs, from D4 to D21. There was a statistically significant difference in motor behavior between groups 3 and 4 on D4 for the right hindlimb (p = 0.0089), but there was no evidence of significant difference between groups 3 and 4 on D4 for the left hindlimb. The difference was significant between groups 3 and 4 for both hindlimbs on D7 and D14 (p[D7/right hindlimb] = 0.017; p[D7/left hindlimb] = 0.012; p[d14/right hindlimb] = 0.002; p[D14/left hindlimb] = 0.0001).

FIG. 6.

Transection injury Basso, Beattie, and Bresnahan scale (BBB) scores averaged by group. The early recovery stage is at the bottom, the intermediate stage is in the middle, and the late recovery stage is at the top. BBB scores are shown for both pre-injury (baseline B1 and baseline B2) and 21 days post-injury (D4, D7, D14, and D21). D, day.

Histology

Figure 7A shows the gross anatomical structures of representative samples of the spinal cord from each group, with yellow arrows marking the region of transection injury. This is to confirm the extent and precise location of the SCI. H&E staining, as shown in Figure 7B, indicates the histological damage of the different transection as well as its parenchymal extent above and below the site of transection.

FIG. 7.

Representative images of T8 left, T10 right, double hemitransection (T8 left and T10 right), and complete T8 transection regions of stained spinal cord are shown in (A) (gross anatomy, 1 × magnification) and (B) (hematoxylin and eosin staining, 5 × magnification).

Discussion

In this study, the effects of complete and hemitransection injuries at T8 and T10 of the spinal cord were examined in an established CNS injury model. Moreover, we utilized a novel approach by using objective electrophysiological measures and motor behavior assessments to illustrate the reorganization of neural pathways within the injured CNS.

SSEPs are an excellent objective assessment tool to reveal the changes in the functionality from the peripheral to the CNS. SSEPs are also a valid diagnostic tool in medicine that can be used to study both the natural history of a transection injury and its subsequent endogenous recovery. We have adapted a modified version of transection injury, where no insulating material was placed after transection in the spinal cord to make our research study clinically relevant. This methodology would enable us to clearly demonstrate how endogenous recovery could relate to plasticity, reorganization, and regeneration within the spinal cord, as well as to any changes in the cortices. The combination of SSEPs and a true replicate model of injury found in humans strengthen our findings regarding the compensatory mechanism in the CNS.

Our investigation using four groups of hemitransection and complete transection injury has shown that there is a significant increase in SSEP amplitudes using only a forelimb stimulus in the corresponding contralateral cortex, indicating the reorganization of neural pathways within the lower CNS structures and the development of more permanent plasticity in the cortex. These results are in line with our previous publication using SSEP assessment in a contusion model of spinal cord, where we have shown an increase in evoked potential response in the corresponding cortices after contusive SCI for 4 weeks.⁹

Interestingly, this increase was observed in the ipsilateral forelimb cortex and in the corresponding contralateral hindlimb cortex. This suggests the presence of a compensatory mechanism that may involve the reorganization of the neuropathways within the spinal cord at and around the site of transection. These mechanisms may also involve plasticity in the cortex, whereby cortical networks adapt to generate new functionality and/or adopt silent neuronal networks in the corresponding somatosensory areas where the ascending inputs are absent. These effects can be observed as early as D4 post-injury. Within our four groups of injury, the full transection group was comparable to the severe contusion injury, whereas the double hemitransection group was similar to moderate contusion.^18,28 The other two groups of hemitransection on the right and left had similar assessments as mild contusive injury. Such a gradation in injury severity was also reflected in the reported BBB scores.

Cortical representations of uninjured pathways have been shown to extend into cortical areas that have lost their corresponding ascending inputs.²⁹ Such a robust expansion of the somatosensory representation has been shown to be stable even up to 2 years post-injury.³⁰ Earlier studies have reported that, after T9 complete transection, there is a significant increase in fMRI signal within the hindlimb somatosensory cortex despite no sensory input from the corresponding limb after forelimb stimulation. Further, BOLD signals indicated that hindlimb territory developed a permanent responsiveness to contralateral forelimb stimulation, indicating true plasticity.³¹ Moreover, Ghosh and colleagues⁴ also used fMRI to reveal that the expansion of the forelimb sensory representations into the corresponding hindlimb cortex after bilateral dorsal hemitransection was not a temporary adaptation and could indicate the presence of rewired neurons.

Although our results and the study by Ghosh and colleagues⁴ both demonstrate that these adaptive reorganizations and plasticity occur within weeks, previous studies also show that changes could be detected within hours in the corresponding contralesional hindlimb cortex. They also found that the forelimb cortex produces an immediate bilateral increase in response to forelimb stimuli in both the contralateral and ipsilateral cortex.^32,33 Another SCI study found that this increased evoked response and suppressed spontaneous response also occurred in the complete transection model.³⁴ These changes may be the cause or the precursor of the plasticity we have observed in this study.

Future work would involve comparing contusion and transection injury and assessing how SSEPs differ between the two injury models in a larger cohort of animals. We would also include more-advanced anatomical and anatomical-functional assessment tools in order to study the different aspects of reorganization and plasticity.

Footnotes

Acknowledgments

This work was supported by National University of Singapore (NUS) grants R-175-000-122-112 (Tier 1; principal investigator [PI]: A. All), R-175-000-121-133 (Departmental Start-Up Fund; PI: A. All), and R-175-000-121-733 (Singapore Institute for Neurotechnology Seed Fund; PI: A. All) and NUHS Clinician Research Grant FY2014 (PI: A. All; R-175-000-133-733), National University of Singapore. It was also supported by a Maryland stem cell research fund grant proposal (2013-MSCRFII-0109-00; PI: A. All).

Author Disclosure Statement

No competing financial interests exist.

References

Blesch

, and Tuszynski

M.H.

(2009). Spinal cord injury: plasticity, regeneration and the challenge of translational drug development. Trends Neurosci. 32, 41–47.

Ding

, Kastin

A.J.

, and Pan

(2005). Neural plasticity after spinal cord injury. Curr. Pharm. Des., 11, 1441–1450.

Onifer

S.M.

, Smith

G.M.

, and Fouad

(2011). Plasticity after spinal cord injury: relevance to recovery and approaches to facilitate it. Neurotherapeutics, 8, 283–293.

Ghosh

, Haiss

, Sydekum

, Schneider

, Gullo

, Wyss

M.T.

, Mueggler

, Baltes

, Rudin

, Weber

, and Schwab

M.E.

(2010). Rewiring of hindlimb corticospinal neurons after spinal cord injury. Nat. Neurosci., 13, 97–104.

Ghosh

, Sydekum

, Haiss

, Peduzzi

, Zorner

, Schneider

, Baltes

, Rudin

, Weber

, and Schwab

M.E.

(2009). Functional and anatomical reorganization of the sensory-motor cortex after incomplete spinal cord injury in adult rats. J. Neurosci., 29, 12210–12219.

Keck

, Mrsic-Flogel

T.D.

, Vaz Afonso

, Eysel

U.T.

, Bonhoeffer

, and Hubener

(2008). Massive restructuring of neuronal circuits during functional reorganization of adult visual cortex. Nat. Neurosci., 11, 1162–1167.

Isa

, and Nishimura

(2014). Plasticity for recovery after partial spinal cord injury—hierarchical organization. Neurosci. Res., 78, 3–8.

Chisholm

A.E.

, Peters

, Borich

M.R.

, Boyd

L.A.

, and Lam

(2015). Short-term cortical plasticity associated with feedback-error learning after locomotor training in a patient with incomplete spinal cord injury. Phys. Ther., 95, 257–266.

Bazley

F.A.

, Maybhate

, Tan

C.S.

, Thakor

N.V.

, Kerr

, and All

A.H.

(2014). Enhancement of bilateral cortical somatosensory evoked potentials to intact forelimb stimulation following thoracic contusion spinal cord injury in rats. IEEE Trans. Neural Syst. Rehabil. Eng., 22, 953–964.

10.

Nardone

, Holler

, Brigo

, Seidl

, Christova

, Bergmann

, Golaszewski

, and Trinka

(2013). Functional brain reorganization after spinal cord injury: systematic review of animal and human studies. Brain Res. 1504, 58–73.

11.

Moxon

K.A.

, Oliviero

, Aguilar

, and Foffani

(2014). Cortical reorganization after spinal cord injury: always for good?. Neuroscience, 283, 78–94.

12.

Murata

, Higo

, Hayashi

, Nishimura

, Sugiyama

, Oishi

, Tsukada

, Isa

, and Onoe

(2015). Temporal plasticity involved in recovery from manual dexterity deficit after motor cortex lesion in macaque monkeys. J. Neurosci., 35, 84–95.

13.

Silva

N.A.

, Sousa

, Reis

R.L.

, and Salgado

A.J.

(2014). From basics to clinical: a comprehensive review on spinal cord injury. Prog. Neurobiol., 114, 25–57.

14.

Raineteau

, and Schwab

M.E.

(2001). Plasticity of motor systems after incomplete spinal cord injury. Nat. Rev. Neurosci., 2, 263–273.

15.

Bazley

F.A.

, All

A.H.

, Thakor

N.V.

, and Maybhate

(2011). Plasticity associated changes in cortical somatosensory evoked potentials following spinal cord injury in rats. Conf. Proc. IEEE Eng. Med. Biol. Soc., 2011, 2005–2008.

16.

Basso

D.M.

, Beattie

M.S.

, and Bresnahan

J.C.

(1995). A sensitive and reliable locomotor rating scale for open field testing in rats. J. Neurotrauma, 12, 1–21.

17.

Basso

D.M.

, Beattie

M.S.

, and Bresnahan

J.C.

(1996). Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Exp. Neurol., 139, 244–256.

18.

Agrawal

, Sherman

, Maybhate

, Gorelik

, Kerr

D.A.

, Thakor

N.V.

, and All

A.H.

(2010). Slope analysis of somatosensory evoked potentials in spinal cord injury for detecting contusion injury and focal demyelination. J. Clin. Neurosci., 17, 1159–1164.

19.

Bazley

F.A.

, Hu

, Maybhate

, Pourmorteza

, Pashai

, Thakor

N.V.

, Kerr

C.L.

, and All

A.H.

(2012). Electrophysiological evaluation of sensory and motor pathways after incomplete unilateral spinal cord contusion. J. Neurosurg. Spine, 16, 414–423.

20.

DePaul

M.A.

, Lin

C.Y.

, Silver

, and Lee

Y.S.

(2015). Peripheral nerve transplantation combined with acidic fibroblast growth factor and chondroitinase induces regeneration and improves urinary function in complete spinal cord transected adult mice. PLoS One, 10, e0139335.

21.

Kortelainen

, Vipin

, Thow Xin

, Mir

, Thakor

, Al-Nashash

, and All

(2014). Effect of isoflurane on somatosensory evoked potentials in a rat model. Conf. Proc. IEEE Eng. Med. Biol. Soc., 2014, 4286–4289.

22.

Mir

, Al-Nashash

, Kerr

, Thakor

, and All

(2010). Histogram based quantification of spinal cord injury level using somatosensory evoked potentials. Conf. Proc. IEEE Eng. Med. Biol. Soc., 2010, 4942–4945.

23.

Sherman

D.L.

, Wuyyuru

, Brooke

M.J.

, Zhang

H.X.

, Sepkuty

J.P.

, Thakor

N.V.

, Natarajan

, and All

A.H.

(2010). Spinal cord integrity monitoring by adaptive coherence measurement. J. Neurosci. Methods, 193, 90–99.

24.

Al-Nashash

, Fatoo

N.A.

, Mirza

N.N.

, Ahmed

R.I.

, Agrawal

, Thakor

N.V.

, and All

A.H.

(2009). Spinal cord injury detection and monitoring using spectral coherence. IEEE Trans. Biomed. Eng., 56, 1971–1979.

25.

Agrawal

, Sherman

, Thakor

, and All

(2008). A novel shape analysis technique for somatosensory evoked potentials. Conf. Proc. IEEE Eng. Med. Biol. Soc., 2008, 4688–4691.

26.

Agrawal

, Thakor

N.V.

, and All

A.H.

(2009). Evoked potential versus behavior to detect minor insult to the spinal cord in a rat model. J. Clin. Neurosci., 16, 1052–1055.

27.

All

A.H.

, Agrawal

, Walczak

, Maybhate

, Bulte

J.W.

, and Kerr

D.A.

(2010). Evoked potential and behavioral outcomes for experimental autoimmune encephalomyelitis in Lewis rats. Neurol. Sci., 31, 595–601.

28.

Agrawal

, Kerr

, Thakor

N.V.

, and All

A.H.

(2010). Characterization of graded multicenter animal spinal cord injury study contusion spinal cord injury using somatosensory-evoked potentials. Spine, 35, 1122–1127.

29.

Sydekum

, Baltes

, Ghosh

, Mueggler

, Schwab

M.E.

, and Rudin

(2009). Functional reorganization in rat somatosensory cortex assessed by fMRI: elastic image registration based on structural landmarks in fMRI images and application to spinal cord injured rats. Neuroimage, 44, 1345–1354.

30.

Jain

, Qi

H.X.

, Collins

C.E.

, and Kaas

J.H.

(2008). Large-scale reorganization in the somatosensory cortex and thalamus after sensory loss in macaque monkeys. J. Neurosci., 28, 11042–11060.

31.

Endo

, Spenger

, Tominaga

, Brene

, and Olson

(2007). Cortical sensory map rearrangement after spinal cord injury: fMRI responses linked to Nogo signalling. Brain, 130, 2951–2961.

32.

Yague

J.G.

, Foffani

, and Aguilar

(2011). Cortical hyperexcitability in response to preserved spinothalamic inputs immediately after spinal cord hemisection. Exp. Neurol., 227, 252–263.

33.

Yague

J.G.

, Humanes-Valera

, Aguilar

, and Foffani

(2014). Functional reorganization of the forepaw cortical representation immediately after thoracic spinal cord hemisection in rats. Exp. Neurol., 257, 19–24.

34.

Aguilar

, Humanes-Valera

, Alonso-Calvino

, Yague

J.G.

, Moxon

K.A.

, Oliviero

, and Foffani

(2010). Spinal cord injury immediately changes the state of the brain. J. Neurosci., 30, 7528–7537.