Utility of somatosensory and motor-evoked potentials in reflecting gross and fine motor functions after unilateral cervical spinal cord contusion injury

Rong Li , Zu-Cheng Huang, Hong-Yan Cui Zhi-Ping Huang, Jun-Hao Liu,Qing-An Zhu, , Yong Hu ,

Abstract Fine motor skills are thought to rely on the integrity of ascending sensory pathways in the spinal dorsal column as well as descending motor pathways that have a neocortical origin. However, the neurophysiological processes underlying communication between the somatosensory and motor pathways that regulate fine motor skills during spontaneous recovery after spinal cord contusion injury remain unclear. Here, we established a rat model of cervical hemicontusive injury using C5 laminectomy followed by contusional displacement of 1.2 mm (mild injury)or 2.0 mm (severe injury) to the C5 spinal cord. Electrophysiological recordings were performed on the brachial muscles up to 12 weeks after injury to investigate the mechanisms by which spinal cord pathways participate in motor function. After spinal cord contusion injury,the amplitudes of somatosensory and motor-evoked potentials were reduced, and the latencies were increased. The forelimb open field locomotion test, grooming test, rearing test and Montoya staircase test revealed improvement in functions. With increasing time after injury,the amplitudes of somatosensory and motor-evoked potentials in rats with mild spinal cord injury increased gradually, and the latencies gradually shortened. In comparison, the recovery times of somatosensory and motor-evoked potential amplitudes and latencies were longer,and the recovery of motor function was delayed in rats with severe spinal cord injury. Correlation analysis revealed that somatosensoryevoked potential and motor-evoked potential parameters were correlated with gross and fine motor function in rats with mild spinal cord contusion injury. In contrast, only somatosensory-evoked potential amplitude was correlated with fine motor skills in rats with severe spinal cord injury. Our results show that changes in both somatosensory and motor-evoked potentials can reflect the changes in gross and fine motor functions after mild spinal cord contusion injury, and that the change in somatosensory-evoked potential amplitude can also reflect the change in fine motor function after severe spinal cord contusion injury. This study was approved by the Animal Ethics Committee of Nanfang Hospital, Southern Medical University, China (approval No. NFYY-2017-67) on June 11, 2017.

Key Words: central nervous system; motor-evoked potential; motor function; regeneration; repair; somatosensory-evoked potential; spinal cord; spinal cord injury

Introduction

Upper limb motor skills are necessary for individuals with cervical spinal cord injury (SCI), and rehabilitation of these skills following SCI facilitates a substantive improvement in independent living (Anderson et al., 2007; Ding et al.,2019). The loss of proprioception can lead to deficits in fine motor control, which results from impaired integration of cortical sensory and motor signals (Akitake et al., 2015).However, the neurophysiological processes underlying the coordinated regulation of fine motor skills via links between the somatosensory and motor pathways during spontaneous recovery after SCI remain unclear.

Corticospinal circuits that descend from the motor cortex play an important role in the regulation of motor neuron activity, as has been reported by many classical studies(Isomura et al., 2009; Peters et al., 2017). Skilled voluntary movements (such as reaching and grasping) are generally considered to primarily rely on neural circuits located in the motor cortex, while stereotyped motor functions (such as locomotion and posture) are regulated by spinal motor circuits(Alstermark and Isa, 2012; Wang et al., 2018). However, these traditional views were challenged by the identification of the mechanisms underlying voluntary fine motor recovery after SCI in studies investigating behavioral deficits after limited interruption of the dorsal column in rodents and monkeys (Qi et al., 2014; Inácio et al., 2016). Some studies demonstrated that a preserved sensory pathway after partial or complete dorsal column lesion can facilitate reactivation of the cortex and behavioral recovery (Varej?o and Filipe, 2007; Qi et al.,2014). Even with a normal corticospinal tract, loss of motor control may occur when proprioception is severely impaired by dysfunction of the sensorimotor network (Kato and Izumiyama, 2015). Additionally, a recent study showed that the modulation of walking speed, which is dependent on muscle activity and strength, requires proprioceptive feedback from the muscle spindle; in the absence of this feedback, rats cannot walk as fast as normal (Mayer et al., 2018). Similarly,a mouse study showed that animals without muscle spindle feedback could execute basic locomotor behaviors, but exhibited poor performance in precise motor tasks (Takeoka et al., 2014). These studies highlight the key role of sensory feedback in voluntary fine motor control, and underscore its importance in the development of interventions that enhance motor recovery after SCI.

Regardless of whether the SCI is incomplete or complete at any level of the cord, motor and sensory functions can spontaneously recover to varying degrees in both humans and animals because of the plasticity of the nervous system(Li et al., 2017; Wei et al., 2018; Harnie et al., 2019). This can involve various changes in the neuraxis, including alterations in the properties of the preserved neuronal circuit, axon collateral sprouting, and synaptic rearrangement (Fouad and Tse, 2008; Buen et al., 2017; Li et al., 2017; García et al.,2019). However, it is difficult to evaluate the contribution of these plastic processes that occur in specific sensory and motor systems to functional recovery through behavioral assessments alone. Electrophysiological examinations provide an objective and quantitative method to separately evaluate these plasticity-induced electrophysiological changes in sensory and motor pathways, and have been applied as a routine tool for intraoperative spinal cord monitoring (Hu et al., 2011). Therefore, complementary assessments of sensory and motor pathways after SCI, such as electrophysiology,can expand our understanding of their common and distinct contributions to voluntary sensorimotor control.

In present study, we hypothesized that fine motor skills are associated with somatosensory and motor pathway recovery after SCI. To investigate whether sensory feedback regulation is involved in the control of fine motor function, we applied multiple behavioral analyses of gross to fine motor skills combined with serial recordings of motor-evoked potentials(MEPs) and somatosensory-evoked potentials (SEPs) following dorsal SCI in rats.

Materials and Methods

Animals

A total of 32 adult male Sprague-Dawley rats (weighing ～320 g,aged 3–4 months) were purchased from Guangdong Medical Laboratory Animal Center, Foshan, China (license No. SYXK(Yue) 2018-0002), and were randomly assigned to one of the following three groups: 1.2-mm SCI (n= 12, defined as mild),2-mm SCI (n= 12, defined as severe), or sham (laminectomy,n= 8). The rats were acclimated in an environmentally controlled breeding room at 22°C and indoor humidity(30–50%) on a reverse light cycle (12/12 hours) with standard chow and filtered water for 7 days and fasted for 12 hours before the experiment. All procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH Publications No. 8023, revised 1978), and were approved by the Animal Ethics Committee of Nanfang Hospital, Southern Medical University, China (approval No.NFYY-2017-67) on June 11, 2017.

Cervical hemi-contusive SCI

All rats underwent a surgical procedure as previously described (Huang et al., 2018). Briefly, a C5 laminectomy was conducted to expose the cervical spinal cord, followed by immobilization from C4–6 with a customized vertebral clamp to a stereotaxic frame (68902; RWD Biomed Inc., Shenzhen,China). A 1.5-mm diameter impactor was placed over the dura at 0.25 mm right of the midline, and lowered vertically until the initial contact force reached 0.01 N. A target contusion displacement of 1.2 mm (mild injury) or 2.0 mm (severe injury)at 500 mm/s was applied to the C5 spinal cord by a servoelectromagnetic material testing machine (ElectroPuls E1000;Instron, Canton, MA, USA). Contusion displacement, speed and peak force during the time of impact were recorded and analyzed with Instron software (V1.5.6; Instron WaveMatrix,Instron).

Electrophysiological evaluation

SEP and MEP recordings were performed prior to SCI, followed by recordings collected at 10 minutes, 3 days, 1 , 2, 4, 8 and 12 weeks after injury. The MEP and SEP latencies were defined as the time from the stimulus to the first positive or negative deflection from baseline, and the amplitude was defined as the maximum amplitude between the maximum positive and negative peaks (Figure 1). A electrophysiological monitoring protocol was used for forelimb acquisition of SEPs and MEPs using a 16-channel monitoring system (YRKJ-G2008; YiRui Technology Co., Ltd., Zhuhai, China).

SEPs were recorded from two 1.2 mm needle electrodes (NES-1500/13/0.4; Friendship Medical Electronics Co., Ltd., Xi’an,China) inserted subdermally in the C3/4 region of the skull according to international 10–20 system electrode placement.The reference electrode was inserted subcutaneously over the skull at the frontal midline. Constant-current stimulation(5 mA, 0.1 ms) was applied to the median nerves via a pair of stimulating needle electrodes at a frequency of 5.3 Hz. SEP signals were bandpass-filtered between 30 and 3000 Hz, and 300 responses were averaged for each recording session.MEPs were evoked by constant electrical stimulation via two 1.2 mm needle electrodes (NE-S-1500/13/0.4; Friendship Medical Electronics Co., Ltd.) inserted subcutaneously over the approximate location of the motor cortex corresponding to C3–4. Single-trial MEPs were obtained with a current intensity of 30–50 mA, a pulse duration of 1 ms, and a frequency of 300 Hz. MEPs were recorded using subdermal needle electrode pairs from the brachioradialis muscle, and signals were filtered at a bandpass frequency between 30 and 3000 Hz.

Behavioral testing

All behavioral tests were performed prior to SCI, followed by recordings collected at multiple time points up to 12 weeks after injury. Gross motor function was detected using forelimb open field test, grooming and rearing test, while fine motor skills were evaluated by Montoya staircase test.

Forelimb open field test

The forelimb locomotor rating scale scored from 1 to 17 was developed according to behavioral changes observed after unilateral injury (Singh et al., 2014). Briefly, rats were placed in an enclosure, and were allowed to move freely in the open field. A quantitative scale was used to measure movements of the shoulder, elbow and wrist joints, as well as forepaw position and digit placement in a minimum of three locomotor tests within 5 minutes. The higher the score, the better the motor function.

Grooming test

The grooming test was used to assess forelimb function according to the maximal area each forelimb independently touched during grooming (Plunet et al., 2008). Briefly, points were assigned with a 6-point scoring system on a scale ranging from the inability to touch the head or face (score of 0) to the ability to touch the head area behind the ears (score of 5).

Rearing test

The rearing test was employed to evaluate the coordination of the forelimbs. Briefly, video recordings were made of forelimb movement during exploratory activity in a transparent cylinder. A mirror placed behind the cylinder at an angle enabled to record movements of the forelimbs in all directions. The preference for use of one or both limbs was calculated as the percentage of total limb movement during wall climbing and landing. All animals were estimated for not less than 20 exploratory movements along the wall and at least 10 landing movements in each session.

Montoya staircase test

Forelimb reaching and grasping performance was assessed using the Montoya staircase test (Plunet et al., 2008). During this test, animals are freely allowed to reach down for colored pellets placed within six wells numbered 1 to 6 from near to far distance in a descending staircase on each side.Color-coded pellets were placed on each level of the well on the staircase to keep track of which pellets were obtained during scoring. During baseline and post-injury assessment,the pellets remaining in each well after 15 minutes were counted. Success rates were measured as the number of pellets eaten and the largest numbered well reached. The pellets on larger numbered well the rats eat, the greater distance they reach.

Histological evaluation

Histological quantifications were performed to demonstrate the effect of tissue preservation on pathway conductivity and motor function. Twelve weeks following injury, animals were deeply anesthetized by intraperitoneal injection of sodium pentobarbital (80 mg/kg). Transcardial perfusion was performed with 200 mL of 0.01 M phosphate-buffered saline,sequentially followed by 200 mL 4% paraformaldehyde. Spinal cord tissue at the lesion site was removed and post-fixed in 4% paraformaldehyde overnight. Samples were cryoprotected at 4°C in ascending concentrations of sucrose solution (12%,18% and 24% sucrose, in phosphate-buffered saline solution,24 hours each). A 10-mm spinal cord sample was sliced into 20-μm sections using a cryostat (CM1950; Leica, Wetzlar,Germany). Sections were stained with Eriochrome Cyanine to identify pathological changes in gray and white matter areas.A microscope (Axioplan 2; Zeiss, Oberkochen, Germany) was used to obtain histological images. The damaged area of the spinal cord was measured using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Statistical analysis

All statistical analyses were performed using SPSS 20.0(IBM, Armonk, NY, USA). Between-group differences in SEP and MEP parameters were assessed using a multiple factor repeated-measures analysis of variance followed by Tukey’s honestly significant test. Following the one-way analysis of variance, Bonferronipost hoctests between groups were used to analyze the tissue lesion area and spared white and gray matter areas. An independent linear correlation analysis(Pearson’s correlation coefficient) was performed between the individual behavioral outcomes and electrophysiological parameters. A value ofP＜ 0.05 was considered statistically significant. All errors are given as the standard error (SEM).

Results

Contusion parameters in unilateral cervical SCI

No significant differences in impact velocity were found between the 1.2 and 2-mm SCI groups (P＞ 0.05). There was a significant increase in peak force during impact with increasing displacement of injury (P＜ 0.05). A summary of the contusion biomechanical parameters are presented inAdditional Table 1.

Electrophysiological differences in rats with different severities of unilateral cervical SCI

SEP

Representative evoked potential waveform changes acrosstime after cervical SCI in rats (Additional Figure 1). The SEPs in the 1.2 and 2-mm SCI groups showed extensive waveform changes compared with baseline. SEP amplitude and latency for the two sides in the 1.2 and 2-mm SCI groups differed significantly at all time points following injury (P＜ 0.05;Figure 2). The SEPs ipsilateral to the injury side were abolished in all rats with SCI at 10 minutes post-injury. At 3 days, SEP was partially restored in all 12 rats in the 1.2-mm SCI group,whereas 9 of the 12 rats in the 2-mm SCI group showed SEP activity by 1 week post-injury (Figure 2A). The amplitude and latency of SEPs recorded from the forelimbs on the opposite side of the injury exhibited a fast and substantial recovery to preoperative levels after a slight reduction post-injury (Figure 2BandD).

Moreover, both SEP latency and amplitude showed a significant but distinct trend toward recovery over time in the two injury groups (Figure 2AandC). Notably, SEP latencies in the two SCI groups were significantly different from baseline to 8 weeks post-injury (P＜ 0.01), but the significance was not maintained for the remaining time points. In contrast,the SEPs ipsilateral to the injury side in the 1.2-mm SCI group showed noticeably greater amplitude compared with the 2.0-mm SCI group for all recording days, except immediately postinjury (P＜ 0.01). Although the SEP amplitude recovered much better in the 1.2-mm SCI group compared with the 2-mm SCI group, it did not recover to preoperative levels.

Figure 1｜Illustrations for parameter measurements of SEP (upper) and MEP (lower) response waveforms in rats with unilateral cervical spinal cord contusion.

MEP

As with the SEPs, MEPs in the contralateral forelimb showed a fast and substantial recovery to pre-injury levels after slight amplitude reduction and prolongation in latency (Figure 2FandH). MEP amplitude and latency were significantly different between the two sides in both SCI groups at all days following injury (P＜ 0.01;Figure 2E–H).

MEP amplitude and latency varied significantly with injury degree within each group (P＜ 0.001;Figure 2EandG).Twenty minutes after SCI, there was complete elimination of MEP activity in both injury groups. Rats subjected to the 1.2-mm SCI achieved a slight recovery in MEP activity by day 3, whereas rats with a 2-mm SCI exhibited recovery of MEP activity starting 1 week after injury (Figure 2E). MEP latencies were significantly longer in the two SCI groups compared with the sham group for all recording days (P＜ 0.001), and displayed substantial recovery over time, with better recovery in the 1.2-mm SCI group than the 2.0-mm SCI group. MEP amplitude also showed significant recovery in both SCI groups 4 weeks post-injury (1.2-mm SCI group: from no response to 434.60 ± 78.22 μV,P＜ 0.001; 2-mm SCI group: from no response to 413.54 ± 81.80 μV,P＜ 0.001). However, MEP amplitudes were not significantly different between the injury groups 4 weeks after injury (1.2-mm SCI group: 434.60 ±78.22 μV; 2-mm SCI group: 413.54 ± 81.80 μV;P＞ 0.05). At 12 weeks, MEP amplitude reduction was significantly greater on the injured side, even though both the contralateral and ipsilateral spinal cords achieved a degree of recovery after hemi-contusive injury.

Behavioral difference in rats with different severities of unilateral cervical SCI

Forelimb open field locomotion test

All injured animals presented a clubbed ipsilateral forepaw during locomotion post-SCI. From 1 to 8 weeks post-injury, all injured animals exhibited a noticeable increasing trend in the forelimb locomotor rating scale score. At 8 weeks post-injury,all animals entered a stable stage and demonstrated relatively little improvements up to week 12. The forelimb locomotor rating scale scores varied with the degree of injury, and the scores in the 2-mm SCI group were greater than those in the 1.2-mm SCI group at all time points (P＜ 0.01;Figure 3).

Grooming test

At all time points after SCI, almost all animals demonstrated normal contralateral grooming behavior (data not shown),but exhibited obvious ipsilateral deficits. Significant functional improvements were displayed by the affected forelimbs in the 1.2- and 2-mm SCI groups from 2 to 8 weeks post-SCI (P＜ 0.01),although they maintained only a limited ability to initiate grooming behavior from behind the ears. The grooming scores in the 2-mm SCI group were lower than those in the 1.2-mm SCI group at all time points following injury (P＜ 0.01;Figure 4).

Rearing test

All groups showed a near-complete loss in independent use of the affected paw (ipsilateral) during rearing post-injury (Figure 5A) and used the ipsilateral paw mostly in combination with the contralateral paw (Figure 5B). There was a significant increase in single limb usage to 37.76 ± 2.51% in the 1.2-mm SCI group from 2 to 8 weeks (P＜ 0.001). However, no increase in usage frequency of the ipsilateral paw was noted in the 2.0-mm SCI group over time, and the differences between the two SCI groups were still significant at 12 weeks post-injury (Figure 5A).

Montoya staircase test

After C5 hemi-contusion injury, rats showed severe impairments not only in retrieval success with the ipsilateral paw, but also in maximum distance reached (Figure 6). From 2 to 8 weeks after 1.2-mm SCI, rats demonstrated a marked increase in reaching scores, which remained stable until 12 weeks post-injury (P＜ 0.01;Figure 6A). However, no recovery of reaching deficit was observed in the 2-mm SCI group.Increased maximum distance of reach was observed in rats with 1.2-mm SCI over time, while animals in the 2-mm SCI group could only retrieve pellets from well number 2, and occasionally, well number 3 (Figure 6B).

Histological features in rats with different severities of unilateral cervical SCI

After SCI, all rats developed typical cystic lesions and ipsilateraltissue atrophy (Figure 7A–C). Rats that received an extended 2-mm SCI showed greater ipsilateral tissue damage at the epicenter compared with the 1.2-mm SCI group (P＜ 0.001),and the average lesion areas in the 1.2 and 2-mm SCI groups were 34.2 ± 3.8% and 57.9 ± 6.8% of the hemicord ipsilateral to the injury, respectively (Figure 7D). At the epicenter in the 1.2-mm SCI group, damaged tissue included the total dorsal funiculus and parts of the dorsolateral funiculus(Figure 7A), whereas in the 2-mm SCI group, areas of spared tissue included a thin margin in the ventral funiculus. The gray matter in the dorsal and ventral horns was completely damaged (Figure 7B). The spared area of white matter at the injury epicenter (i.e., ipsilateral side to injury) in the 1.2-mm SCI group was significantly larger than that in the 2-mm SCI group (3.39 ± 0.29 mm2vs. 1.37 ± 0.19 mm2,P＜ 0.01).

Correlation between MEP/SEP and forelimb behavioral scores in rats with different severities of unilateral cervical SCI

The correlations between MEP/SEP and behavior scores were evaluated separately for the 1.2 and 2-mm SCI groups (Figure 8). When evaluating impairment in gross motor function, the latency and amplitude of SEPs and latency of MEPs in the 1.2-mm SCI group were significantly correlated with behavioral outcomes (r= –0.80, –0.72, –0.78 (P＜ 0.05), respectively,for correlation between forelimb locomotor rating scale and SEP latency, SEP amplitude and MEP latency;r= 0.77, 0.78,0.81 (P＜ 0.05), respectively, for correlation between rearing and SEP latency, SEP amplitude and MEP latency;r= –0.73,–0.74, –0.61 (P＜ 0.05), respectively, for correlation between staircase and SEP latency, SEP amplitude and MEP latency). In contrast, these functions were not well correlated in the 2-mm SCI group (P＞ 0.05;Figure 8). Similarly, in the evaluation of fine motor skills with the staircase test, when the degree of SCI was low, a significant correlation was identified between MEP/SEP and the number of pellets consumed, and when the degree of damage was severe, only the SEP amplitude was still correlated (r= 0.67,P＜ 0.05;Figure 8). In general,we noted that the amplitude of MEPs did not correlate well with behavioral outcomes, which may be a result of the large variation in MEP amplitude.

Figure 2｜Forelimb electrophysiological data of rats with unilateral cervical SCI.

Figure 3｜Ipsilateral forelimb locomotor functional recovery is affected by the degree of displacement of the contusion in rats with SCI (forelimb open field test).

Figure 4｜Ipsilateral forelimb function after C5 hemicontusion differing in displacement extent (grooming test).

Figure 5｜Forelimb usage during rearing in C5 hemicontused SCI rats(rearing test).

Figure 6｜Effect of C5 hemicontusion and displacement of the contusion on ipsilateral forelimb reaching and grasping performance (Montoya staircase test).

Discussion

Most rehabilitative interventions emphasize the improvement of motor function, but rarely address sensory functions and sensorimotor control. To maximize recovery and develop effective intervention strategies, the effect of the sensory system on the execution of fine motor control underlying spontaneous recovery after SCI needs to be clarified. In the current study, combined examination of SEP and MEP and changes in clinical neurological functions, ranging from gross to fine motor skills, were systematically analyzed in a rat model of SCI. In general, significant recovery of SEPs and MEPs were accompanied by considerable recovery of gross and fine motor skills in rats with mild SCI. In comparison,rats with severe SCI showed significant MEP recovery, but lower SEP recovery, and while these animals had the ability to execute basic motor tasks, their fine motor skills remained substantially impaired. These findings indicate that not only is the improvement of motor pathways crucial to the recovery of fine motor function, but that the sensory conduction system also plays a key role in the regulation of motor control.Moreover, the restoration of sensory signaling pathways in the spinal cord resulted in a better recovery of fine motor function than of gross motor function, which suggests that sensory regulation may be primarily involved in the feedback control of fine movements rather than gross motor function.

Like primates, the digital dexterity of rats in reaching for and grasping small pieces of food seems to be affected by the integrity of the dorsal spinal ascending sensory pathway and the descending motor pathway originating from the neocortex(Kanagal and Muir, 2007). In the present study, we found that the recovery of SEPs was associated with the recovery of fine motor function. Specifically, MEP and SEP responses recovered to more than 80% of baseline in rats with mild injury, which was followed by a marked increase in forelimb usage during wall placements, grooming, overground locomotion, and paw reaching and grasping. However, rats with severe SCI only showed recovery in locomotor ability and joint motion,and showed no recovery of skilled reaching and grasping functions; this was accompanied by significant improvements in MEPs, but a 70% reduction in SEP amplitude. This suggests that sensory pathway in the dorsal column are involved in providing and regulating unique sensory inputs for certain complex behavioral tasks or fine motor skills. Furthermore,the loss of sensory feedback is thought to be related to the reduction and elimination of the ability to distinguish tactile textures, touch frequencies, and the directions of moving tactile stimuli (Ballermann et al., 2001; Kanagal and Muir,2007). Additionally, we found that MEP and SEP parameters were correlated with gross and fine motor functions in rats with mild injury. In contrast, only SEP amplitude was correlated with behavioral outcome in the severe SCI group,suggesting that SEP amplitude could be a potential indicator of fine motor function.

Behaviorally, spinal dorsal column disruption in rats generates clear motor deficits. Similar to previous studies (Walker et al., 2016), our model exhibited sustained behavioral deficits in the ipsilateral forelimb, with parenchymal damage mainly confined to the ipsilateral side. Motor function deficits after injury were well predicted by the impact parameters, whereby a greater contusion displacement resulted in more severe neurological impairment. Moreover, impaired elbow aiming during limb movement resulted in some imprecise reaches,but less skilled actions, such as joint motion of the limbs, were not significantly affected during locomotion and grooming.However, rats with severe SCI also performed poorly in the rearing test, which may result from impaired feedback control of postural support during vertical exploration (Herter et al.,2015; Kang et al., 2018). In addition, the forepaws on the ipsilateral side of the lesion almost picked up food on the top two wells, without continuing to search along the stairs.In contrast, the forelimbs contralateral to the injury side palpated the well as if searching for the pellets. This indicates that rats with dorsal column lesions failed to realize their failure in detecting food targets. A similar deficit was reported in the probe test, where no food targets were presented,although the rats still reached out and checked for food in their paws (Anderson et al., 2007). These findings reveal that somatosensory feedback is crucial for limb targeting and adjusting movements related to grasping and retracting the forelimbs to bring food to the mouth.

Numerous examples of plasticity have been observed in a multitude of studies on motor recovery in human patients and animal models after SCI (Dunham et al., 2010; Streijger et al., 2013; Yang et al., 2020; Zheng et al., 2020). Dynamic electrophysiological evaluation of sensory and motor pathways has also been performed in various experimental rat models of SCI (Cao et al., 2005; Morris et al., 2017). In the current study,neither SEPs nor MEPs were detectable immediately after cervical SCI, but showed different levels of recovery in the two experimental groups over time, in concordance with previous findings in rats with unilateral graded contusion (Cao et al.,2005). The evoked potentials of rats with mild injury showed earlier signal activity and more obvious recovery compared with rats with severe injury. However, the electrophysiological improvement plateaued after 4 weeks and failed to return to the preoperative level. These results indicate that the rats had a limited ability to establish new connections in the dorsal column system. The reorganization that does occur is not sufficient to completely reactivate the deprived parts of the sensorimotor system (Schlag et al., 2001).

The mechanism underlying the compensation or recovery of interrupted pathways following SCI is not yet completely understood. However, several studies have demonstrated that the spinal cord has the capacity for plasticity and repair following SCI, especially via the lateral sprouting of axons.Indeed, axonal sprouting has been found in corticospinal tracts, reticulospinal tracts and somatosensory pathways following various types of SCI (Raineteau et al., 2002; Bareyre et al., 2005). Tissue staining in the present study revealed that extensive spinal cord tissue destruction was limited to one side, and resulted in white matter demyelination, a reduction in the number of neurons, and cavitation in the gray matter. Although only a small number of axons remained in the dorsal column, we observed considerable functional and electrophysiological recovery in both injury groups.While we did not provide data on plasticity, a previous study showed that the absence of a pathway on one side can be compensated for by the crossover of spontaneous sprouts of axons from the intact tracts on the other side at different levels (Filli et al., 2014).

A limitation of this study is that only general histological changes were observed; quantitative histological evaluation,such as immunofluorescence or immunohistochemical staining, will be carried out in future studies on the underlying mechanisms. In addition, collateral sprouting in single pathways cannot be independently determined from histological findings, and further functional magnetic resonance imaging and neuroanatomical imaging studies are needed to more fully clarify the compensatory mechanisms.

In summary, we demonstrated correlations between electrophysiological findings and motor functional assessments after unilateral cervical spinal cord contusion in rats. A combination of SEP and MEP recordings can be used to evaluate gross and fine motor functional changes after mild contusion injury to the spinal cord. SEP is a potential tool for evaluating fine motor function in both mild and severe contusion injury. Notably, our findings imply that the ascending sensory pathway in motor regulation has a greater impact on the spontaneous recovery of deprived fine motor skills than gross motor function in the forelimbs.

Author contributions:Study conception and design: YH, QAZ;experimental implementation: RL, ZCH, JHL, ZPH; data analysis/interpretation: HYC; manuscript writing: RL; manuscript review and edit:YH, QAZ. All authors contributed to this study and approved the final manuscript.

Conflicts of interest:The authors declare that they have no conflict of interests.

Financial support:This work was supported by the National Natural Science Foundation of China, No. 81871768 (to YH); Chinese Academy of Medical Sciences Initiative for Innovative Medicine of China, No. 2016-I2M-2-006 (to HYC); and High Level-Hospital Program, Health Commission of Guangdong Province, China, No. HKUSZH201902011 (to YH). The funders had no roles in the study design, conduction of experiment,data collection and analysis, decision to publish, or preparation of the manuscript.

Institutional review board statement:The study was approved by the Animal Ethics Committee of Nanfang Hospital, Southern Medical University, China (approval No.NFYY-2017-67) on June 11, 2017.

Copyright license agreement:The Copyright License Agreement has been signed by all authors before publication.

Data sharing statement:Datasets analyzed during the current study are available from the corresponding author on reasonable request.

Plagiarism check:Checked twice by iThenticate.

Peer review:Externally peer reviewed.

Open access statement:This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix,tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Additional files:

Additional Table 1:The impact parameters of unilateral cervical spinal cord injury in rats.

Additional Figure 1:Representative SEP and MEP response waveforms up to 12 weeks after 1.2 mm and 2 mm cervical spinal cord injury from 2 sample rats.

Figure 7｜Histological changes in rats with different severities of unilateral cervical spinal cord injury.

Figure 8｜Intergroup correlations between electrophysiological assessment and behavioral outcome assessed with Pearson’s correlation analysis.