Abstract
Spinal cord injury (SCI) leads to wide-spread neurodegeneration across the neuroaxis. We explored trajectories of surface morphology, demyelination and iron concentration within the basal ganglia-thalamic circuit over 2 years post-SCI. This allowed us to explore the predictive value of neuroimaging biomarkers and determine their suitability as surrogate markers for interventional trials. Changes in markers of surface morphology, myelin and iron concentration of the basal ganglia and thalamus were estimated from 182 MRI datasets acquired in 17 SCI patients and 21 healthy controls at baseline (1-month post injury for patients), after 3, 6, 12, and 24 months. Using regression models, we investigated group difference in linear and non-linear trajectories of these markers. Baseline quantitative MRI parameters were used to predict 24-month clinical outcome. Surface area contracted in the motor (i.e. lower extremity) and pulvinar thalamus, and striatum; and expanded in the motor thalamus and striatum in patients compared to controls over 2-years. In parallel, myelin-sensitive markers decreased in the thalamus, striatum, and globus pallidus, while iron-sensitive markers decreased within the left caudate. Baseline surface area expansions within the striatum (i.e. motor caudate) predicted better lower extremity motor score at 2-years. Extensive extrapyramidal neurodegenerative and reorganizational changes across the basal ganglia-thalamic circuitry occur early after SCI and progress over time; their magnitude being predictive of functional recovery. These results demonstrate a potential role of extrapyramidal plasticity during functional recovery after SCI.
Similar content being viewed by others
Introduction
Spinal cord injury (SCI) leads to permanent functional deficits below the level of injury. Neurorehabilitation can foster sensorimotor recovery, but often only partial improvements can be achieved. Recovery is paralleled by a cascade of progressive neurodegenerative changes affecting the pyramidal, sensory and limbic system1,2; its magnitude predicting functional recovery1,2. In addition, compensatory changes within the extrapyramidal system might contribute to recovery as shown in the non-human primate model of SCI3,4. For example, impaired information flow within the basal ganglia-thalamic circuit5 and the motor cortex6 was associated with abnormal activation patterns and increased functional connectivity in pallido-thalamocortical loops6. However, in SCI patients, the role of the extrapyramidal system in sensorimotor recovery is understudied7.
To characterize structural trajectories of surface area and microstructural parameters, we modelled the MRI measures in terms of linear rate of change (reflecting degeneration/plasticity) and non-linear rate of change (reflecting acceleration/deceleration). In particularly, T1-weighted volumes were used to track trajectories of vertex-wise surface area contractions and expansions of the basal ganglia and thalamic subnuclei over 2-years. Myelin-sensitive magnetization transfer saturation (MT) and longitudinal relaxation rate (R1), a well as iron-sensitive effective transverse relaxation rate (R2*)8,9 were used to characterize associated microstructural changes within the same regions. Finally, early MRI markers of basal ganglia and thalamic subnuclei changes at 1-month post-SCI were used to predict recovery at 2-years.
Methods
Subjects
17 SCI patients were recruited at the Balgrist University Hospital and 21 healthy controls from the local neighborhood between July 2010 and July 2016. All patients had a subacute SCI (< 3 months) without brain injury, and all participants had no pre-existing neurological, mental or medical disorders affecting functional outcome. A subset of this cohort was reported previously10,11,12,13.
All participants underwent a comprehensive MRI protocol. In total, 182 appointments were attended by 38 participants (17 SCI patients, 21 healthy controls), at baseline (1-month post-SCI) and at 3, 6, 12, and 24 months post-SCI. Follow-ups were performed successfully in 82.4%, 100%, 100% and 76.5% of the patients, respectively in 100%, 100%, 95.2% and 100% of the controls. At the same time points, patients were also clinically assessed including 1) the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) protocol for motor, light-touch and pin-prick score, and 2) the Spinal Cord Independence Measure (SCIM) to measure daily life independence. The mean age between patients (mean age 43.8 years) and controls (mean age 35.7) did not differ (Mann–Whitney test, p = 0.12).
The study protocol was in accordance with the Declaration of Helsinki and was approved by the local ethics committee of Zurich “Kantonale Ethikkommission Zurich, KEK” (EK-2010–0,271). All participants provided written informed consent before participation.
Image acquisition
Participants were scanned with a 3 T Magnetom Verio or the upgraded Skyra Fit MRI scanner (Siemens Healthcare, Erlangen, Germany) operated with a 16-channel radiofrequency receive head and neck coil and radiofrequency body transmit coil.
At each time point, the MRI measurement consisted of two protocols: (1) an optimized 3 dimensional high-resolution T1-weighted Magnetization Prepared Rapid Acquisition Gradient-Echo (MPRAGE) sequence14 was used to segment the basal ganglia structures, and (2) a multi-parametric mapping protocol based on multi-echo 3D FLASH sequences15 was used to extract underlying microstructural parameters.
For the T1-weighted MPRAGE sequence, 176 sagittal partitions were acquired with a resolution of 1 mm isotropic within a nominal scan time of 9.04 min (field of view (FOV) = 224 × 256mm2, repetition time (TR) = 2420 ms, echo time (TE) = 4.18 ms, inversion time = 960 ms, flip angle (FA) = 9°, readout bandwidth = 150 Hz/pixel).
The MPM protocol consisted of 3 volumes including a T1-weighted image (TR = 25 ms, FA = 23°), a proton density (PD)-weighted image (TR = 25 ms, FA = 4°), and a magnetization transfer (MT)-weighted image (TR = 37 ms, FA = 9°) with 1 mm isotropic resolution. Total scan time was 23 min (FOV = 240 × 256mm2, readout bandwidth = 480 Hz/pixel) (for further details see10,11,12,13).
Image analysis
Segmentation of basal ganglia based on T1-weighted MPRAGE images
Thalamus, striatum, and globus pallidus were automatically identified using a the MAGeT Brain algorithm16,17. In this approach, an a priori defined atlas is first registered to a representative subset of the dataset (21 subjects) using nonlinear transformations18. This set of subjects then serves as a set of templates and all other subjects are warped to these templates using non-linear registration, resulting in 21 candidate segmentations for each subject. Nonlinear registration was performed using a version of Automatic Normalization Tools (ANTs) registration that is compatible with the minc file format (https://github.com/vfonov/mincANTS). After data quality assessment of all attended scans, this resulted in a total number of MPRAGE scans analyzed of 20/17 (controls/patients) at baseline, 21/14 at 3-month, 21/17 at 6-month, 20/17 at 12-month and 21/13 at 24-month.
Vertex-wise surface area estimation
A quantitative surface-based methodology16 was used to estimate the shape of each structure at study. Atlas to subject transformations were applied to 3D surface models of each labelled atlas structure, resulting in 21 candidate surfaces for each subject. The final surface was developed by computing the median location of each vertex across the 21 candidate surfaces. Second, transformations from the 21 candidate segmentations were averaged within each subject and then across the template library to reduce noise16. Then, for each vertex, a third of the surface of each connecting triangle is assigned to the vertex surface area. A surface-base diffusion-smoothing kernel was then applied (5 mm and 3 mm for the thalamus/striatum and globus pallidus, respectively).
Microstructural analysis
MT-weighted, PD-weighted, and T1-weighted volumes were used to estimate quantitative MT, R1, and R2* maps as described previously15. R1 maps were corrected for radiofrequency transmit field inhomogeneities with UNICORT15. For each subject and time point, MT, R1, and R2* maps were coregistered to their respective T1-weighted image using a 6 parameter registration with a normalized mutual information cost function (https://github.com/CobraLab/minc-toolkit-extras/blob/master/bestlinreg_g). After coregistration, segmentations of the thalamus, striatum, and globus pallidus (including the subnuclei) obtained from MAGeT Brain were used as a binary mask to extract mean values of MT, R1, and R2* for each subject and time point.
The MPM protocol was not acquired in 3 controls at any time point and in 3 patients/1 healthy control at baseline, 6 patients at 3-month, 1 patient at 6-month, and in 4 patients at 24-month. After data quality assessment, this results in a total number of scans analyzed of 17/12 (patients/controls) at baseline, 11/17 at 3-month, 16/18 at 6-month, 17/17 at 12-month, and 13/18 at 24-month.
Statistics
R (R core team, 2015) was used for all statistical analysis. Clinical measures and vertex-wise shape measures were analyzed using mixed-model regression as it permits the inclusion of multiple measurements per person at different time points and at irregular intervals between measurements. We first compared rates of change of these measures between patients and healthy controls by including a group × time (reflecting degeneration/plasticity) and a group x squared time (reflecting acceleration/deceleration) interaction. For clinical measures, time was modelled on a log scale to account for non-linear recovery patterns. Age and gender were included to accommodate potential confounding effects of no interest. We also performed a within patient analysis to look at rates of change of vertex wise measurements between cervical and thoracic injuries. To do this, we performed a similar analysis to that described above, but with grouping being either cervical or thoracic as opposed to patient vs. control. To identify in which specific nuclei group differences were observed, we superimposed surface-labels of the cytoarchitectonic atlas for the thalamus19 on our thalamic template. Second, in those structures that showed group differences in surface area over time, we explored underlying differences in microstructure (R1, MT, R2*) using the same models. Associations between quantitative MRI parameters at 1-month (baseline) and clinical recovery at 2-years were investigated using linear regression models adjusted for age, gender and clinical baseline scores. All vertex-wise analyses were corrected for multiple comparisons using the false discovery rate (FDR) at 10%.
Results
Demographics
We included a total of 17 SCI patients and 21 healthy controls. The mean age of patients did not differ from controls (p > 0.05). The baseline scan was acquired 1.4 ± 0.5 months after injury and the remaining scans at 3.5 ± 1.5, 7.1 ± 1.9, 13.3 ± 3.1 and 28.7 ± 5.1 months. Over 2-years, patients recovered by 2.76 points per log month on their lower extremity motor score [p = 0.001, 95% CI (1.07 4.45)], by 0.91 point per log month on their upper extremity motor score [p = 0.046, 95% CI (0.02 1.80)], and by 7.89 points per log month on their SCIM score (p < 0.001, 95% CI (5.03 10.75)) (Table 1). Light-touch and pin-prick score remained unchanged over time (p = 0.152, p = 0.790). For subject wise plots of trajectory of each clinical score, see Supplementary Information (Clinical Trajectories).
Time course of thalamic changes
Both thalami showed a greater linear rate in surface area contractions in patients compared to controls over time within the inferior parts of the ventral anterior nuclei (VAN), and surface expansion within its superior parts (Fig. 1, Suppl. Table 1). In the left thalamus, these changes were levelling-off over time (Suppl. Fig. 1). In the ventral lateral nuclei (VLN), linear surface contractions were observed in its inferior parts, which levelled-off over time in the right nuclei. Greater linear rate in surface area expansions were found within the superior parts of the VLN in patients compared to controls over time. Greater linear rate in surface area expansions was evident within the ventral posterior nuclei (VPN) in patients compared to controls over time, which rate of change decelerated within the right VPN. Both pulvinar nuclei showed greater rate in linear surface contractions in patients compared to controls, in particular within the lateral and the medial pulvinar. Microstructural changes are summarized in Table 2. No significant difference in the rate of change were observed between patients with cervical or thoracic injuries.
Time course of striatal changes
In the striatum, patients showed bilateral surface area contractions and expansions over time in the caudate and the putamen compared to controls. Specifically, greater linear rate in surface area contractions were evident medially within the central to posterior part of the bilateral caudate (Fig. 2A), which levelled-off over time in patients compared to controls (Suppl. Fig. 2). Greater linear rate in expansions of surface area were observed in the bilateral inferior part of the anterior caudate (Fig. 2B), bilaterally levelling-off over time. Greater linear rate in expansions of surface area was observed in the inferior parts of the ventral caudate (i.e. nucleus accumbens; Fig. 2C), which showed a levelling-off over time at the left side only. Greater linear rate of surface area contractions was evident in the left lateral part of the anterior caudate (Fig. 2D). Within the putamen, greater rate of liner contractions of surface area were observed in the medial central (Fig. 2E) as well as the lateral anterior putamen, bilaterally (Fig. 2F). The surface area of the right lateral anterior putamen showed a levelling-off over time in patients compared to controls. In addition, surface area contractions were observed in the left medial posterior (Fig. 2G), left medial anterior (Fig. 2H) and left lateral superior posterior putamen (Fig. 2I). Expansions in surface area were detected within the left medial anterior (Fig. 2J) and lateral anterior putamen (Fig. 2K). Expansions within the left anterior putamen levelled-off over time in patients compared to controls. Microstructural changes are summarized in Table 2. No significant difference in the rate of change were observed between patients with cervical or thoracic injuries.
Time course of pallidal changes
The surface area analysis demonstrated no changes over time in patients compared to controls in the globus pallidus. Microstructural changes are summarized in Table 2. No significant difference in the rate of change were observed between patients with cervical or thoracic injuries.
Relationship of MRI parameters at 1-month post-SCI and clinical recovery at 2-years
Through running linear models at each vertex, we found that surface area measurements at baseline were significantly related to lower extremity motor score at the 2-year timepoint. Specifically, a cluster of voxels showing increased surface area expansions within the central caudate and the lateral central-posterior putamen at 1-month were associated with better lower extremity motor score at 2-years, adjusted for baseline score (Fig. 3).
Discussion
This study characterizes neurodegenerative and reorganizational changes across the basal ganglia-thalamic circuit after SCI using advanced neuroimaging measures, in vivo. We demonstrate a dynamic pattern of neurodegenerative and reorganizational changes in the basal ganglia-thalamus circuitry within the first 2-years post-SCI, its magnitude predicting functional recovery. These observations illustrate the enduring neurodegenerative and neuroplastic processes induced by a SCI and highlight a progressive diaschisis across the neuroaxis.
Time course of thalamic changes
We have previously shown in the same patient cohort that a focal injury to the spinal cord produces neurodegenerative changes to the sensorimotor and limbic system10,11,13. Moreover, the atrophic thalamus showed also an accumulation of iron content, however the exact location of these changes was not identified13,20. This study shows that SCI leads to bilateral surface area contractions in the inferior parts of the VAN and VLN, while surface area expansions were evident in the superior parts of the VAN, VLN and VPN. The changes in all of these nuclei were paralleled by decreases in myelin-sensitive R1 over time.
Both, the VAN and VLN receive input from the globus pallidus and project to the premotor and motor cortex21. In contrast, the VPN receives input from the lemniscus medialis (transmitting light touch, vibration, and proprioception) and from the spinothalamic tract (transmitting pain and heat sensation)21. All of these nuclei are somatotopically organized21 with distal muscles represented dorsal to those for proximal limb portions; and more cranial representations were found more anterior22. This suggests that the contracting surface area (i.e. local volume decreases) within the lower limb representing areas in motor thalamic nuclei are likely neurodegenerative changes while surface area expansions (i.e. local volume increases) in upper limb representing areas of sensorimotor thalamic nuclei could be interpreted as compensatory changes on the assumption that performance improvements during training23,24 or neurorehabilitation24,25 induces grey matter volume increase.
Possible biological substrates for local volume decreases (i.e. surface area contractions) might include changes in the number of neurons induced by trans-synaptic degeneration, neuronal death, demyelination, synaptic pruning, or changes in the number of glial cells26. Axonal sprouting and activity-dependent plasticity27 might account for volume increases (i.e. surface area expansions). In particular, compensatory use of the upper limbs might drive expansion of cortical M1 hand area into output-deprived leg area28 as a consequence of the rewired hind limb neurons onto cervical motor circuits28; the same mechanism might also be occurring in the basal ganglia. However, surface area changes are not specific to any single pathological process. Here we show that changes in myelin-sensitive R129 values might be interpreted as ongoing demyelination in the sensorimotor nuclei of the thalamus.
The lateral and medial pulvinar nuclei also showed surface area contractions over time. These nuclei are involved in visual attention regulation, and also coordinate sensory input with its corresponding motor response30. As most SCI patients have motor and sensory deficits, it seems likely that the coordination between these two modalities is also altered31. In addition, SCI patients rely more on their visual input in order to coordinate limb movements due to impaired proprioception32.
Time course of striatal changes
Surface area contractions were also observed in striatal regions involved in sensorimotor and cognition processing, and surface area expansions in regions involved in cognition, emotion and reward processing4,33. These changes were paralleled by decreases in myelin-sensitive R1 and MT. Interestingly, we observed bilateral surface area contraction within the medial and lateral dorsal part of the motor putamen, and as the putamen consists of two sets of somatotopic representation34. This indicates loss of afferents from both the primary motor cortex and the supplementary motor area. This is in line with previous studies, showing that both of these structures show altered function and structure after SCI20,35. Remarkably, we also observed surface area expansions in the ventral striatum (i.e. nucleus accumbens), a region known to be involved in recovery of the dexterous finger movements after experimental SCI4,36.
Time course of pallidal changes
We observed decreases in myelin-sensitive R1 changes in the globus pallidus. Since the globus pallidus is the relay station between the striatum and the thalamus, trans-synaptic degeneration of afferents and efferents of the primary motor cortex and the premotor cortex may primarily lead to changes in the striatum and the thalamus.
Effect of rostro caudal level of spinal injury on neurodegeneration
Analyzing thoracic and cervical SCI patients conjointly allows to address (i) to what extent the level of lesion and consequently the extent and kind of neurological-functional impairment are related to secondary structural changes within the extrapyramidal system, and (ii) whether some secondary effects are independent of the amount and level of spinal cord damage. As in other neurological disorders, even a very local and focal lesion can have remote effects, i.e. diaschisis)37,38. The fact that we did not detect a relationship between lesion level and extrapyramidal changes indicates that these remote structural changes are likely trans-synaptic and might be a general effect due to the loss of voluntary control over the limbs. This is also reflected by the association between early surface area expansion and lower limb recovery.
Relationship of MRI parameters at 1-month post-SCI and clinical recovery at 2-years
Clinical recovery occurs most rapidly within the first 6 months and is fostered through neurorehabilitation by promoting cortical, subcortical and spinal cord neural circuit reorganisation39,40. Interestingly, early surface area expansions within the motor striatum were associated with better 2-year lower extremity motor score. Although all patients of the present study were admitted for neurorehabilitation within one months after the injury, there is an inherent variability (depending on lesion level and extent of impairment) of how much therapy each individual patient received already at this early timepoint. Thus it remains speculative if the expansion of surface area is indicative of compensatory plasticity—a phenomena also observed in Multiple sclerosis41 and Parkinson’s disease42 or whether it is a co-occurrence. However, our study shows the clinical viability of MRI based structural shape measures for monitoring and predicting recovery after traumatic SCI. For example these neuroimaging biomarkers could be used to assess the influence of therapeutic interventions (e.g. levodopa (L-Dopa)) on training effects druing rehabilaition not only after SCI43 but also stroke44 or other disorders where the extrapyramidal system is affected. Its future application holds promise to provide a tool to improve our understanding of the disease mechanism after SCI and might render extrapyramidal neuroimaging biomarkes. These new insights will enable us to better predict individual recovery trajectories and identify those patients who could profit from targeted interventions, such as levodopa. However, further investigations are required to assess the interplay between function, the underlying neuropathology and treatment response.
Limitations
First, although we identified the approximate location of changes within the striatum and the globus pallidus, a histology-based atlas of the striatum and the globus pallidus is currently not available. Next, the upgrade of the MRI scanner might potentially be a confounding factor, but this was mitigated by the fact that all participants were scanned before and after the upgrade. Although the mean age of the control cohort was not significantly different from the patient cohort we included age as a (nuisance) covariate, and its inclusion made no material difference to the results (e.g. the relevant coefficient for the subject indicator terms in the repression models). Finally, the applied MPM protocol, consisting of MT, R1 and R2* measures, provides only indirect measures of myelin45 and iron46 but not specific to processes such as demyelination, degeneration and iron content.
Conclusion
This study provides evidence for progressive extrapyramidal macro- and microstructural changes after SCI, indicating large scale, trans-neuronal remodeling of primarily unaffected brain regions. Extrapyramidal plasticity early after SCI are associated with long-term motor recovery. This study therefore provides unbiased, quantitative readouts of nuclei-specific extrapyramidal changes which are clinically eloquent.
References
Freund, P. et al. MRI in traumatic spinal cord injury: from clinical assessment to neuroimaging biomarkers. Lancet Neurol. 18, 1123–1135 (2019).
David, G. et al. Traumatic and nontraumatic spinal cord injury: pathological insights from neuroimaging. Nat. Rev. Neurol. https://doi.org/10.1038/s41582-019-0270-5 (2019).
Zaaimi, B., Edgley, S. A., Soteropoulos, D. S. & Baker, S. N. Changes in descending motor pathway connectivity after corticospinal tract lesion in macaque monkey. Brain 135, 2277–2289 (2012).
Sawada, M. et al. Function of the nucleus accumbens in motor control during recovery after spinal cord injury. Science 350, 98–101 (2015).
Cramer, S. C., Lastra, L., Lacourse, M. G. & Cohen, M. J. Brain motor system function after chronic, complete spinal cord injury. Brain 128, 2941–2950 (2005).
Min, Y.-S. et al. Alteration of resting-state brain sensorimotor connectivity following spinal cord injury: a resting-state functional magnetic resonance imaging study. J. Neurotrauma 32, 1422–1427 (2015).
Lemon, R. N. Descending pathways in motor control. Annu. Rev. Neurosci. 31, 195–218 (2008).
Weiskopf, N., Mohammadi, S., Lutti, A. & Callaghan, M. F. Advances in MRI-based computational neuroanatomy. Curr. Opin. Neurol. 28, 313–322 (2015).
Tabelow, K. et al. hMRI: a toolbox for quantitative MRI in neuroscience and clinical research. Neuroimage 194, 191–210 (2019).
Freund, P. et al. MRI investigation of the sensorimotor cortex and the corticospinal tract after acute spinal cord injury: a prospective longitudinal study. Lancet Neurol. 12, 873–881 (2013).
Grabher, P. et al. Tracking sensory system atrophy and outcome prediction in spinal cord injury. Ann. Neurol. 78, 751–761 (2015).
Seif, M., Ziegler, G. & Freund, P. Progressive ventricles enlargement and cerebrospinal fluid volume increases as a marker of neurodegeneration in patients with spinal cord injury: a longitudinal magnetic resonance imaging study. J. Neurotrauma https://doi.org/10.1089/neu.2017.5522 (2018).
Ziegler, G. et al. Progressive neurodegeneration following spinal cord injury. Neurology 90, e1257–e1266 (2018).
Tardif, C. L., Collins, D. L. & Pike, G. B. Sensitivity of voxel-based morphometry analysis to choice of imaging protocol at 3 T. Neuroimage. 44, 827–838 (2009).
Weiskopf, N. et al. Quantitative multi-parameter mapping of R1, PD(*), MT, and R2(*) at 3T: a multi-center validation. Front. Neurosci. 7, 95 (2013).
Chakravarty, M. M. et al. Performing label-fusion-based segmentation using multiple automatically generated templates. Hum. Brain Mapp. 34, 2635–2654 (2013).
Pipitone, J. et al. Multi-atlas segmentation of the whole hippocampus and subfields using multiple automatically generated templates. Neuroimage 101, 494–512 (2014).
Chakravarty, M. M. et al. Comparison of piece-wise linear, linear, and nonlinear atlas-to-patient warping techniques: analysis of the labeling of subcortical nuclei for functional neurosurgical applications. Hum. Brain Mapp. 30, 3574–3595 (2009).
Chakravarty, M. M., Bertrand, G., Hodge, C. P., Sadikot, A. F. & Collins, D. L. The creation of a brain atlas for image guided neurosurgery using serial histological data. Neuroimage 30, 359–376 (2006).
Jutzeler, C. R. R. C. R. et al. Association of pain and CNS structural changes after spinal cord injury. Sci. Rep. 6, 18534 (2016).
Nambu, A. Somatotopic organization of the primate basal ganglia. Front. Neuroanat. 5, 26 (2011).
Vitek, J. L., Ashe, J., DeLong, M. R. & Alexander, G. E. Physiologic properties and somatotopic organization of the primate motor thalamus. J. Neurophysiol. 71, 1498–1513 (1994).
Draganski, B. et al. Neuroplasticity: changes in grey matter induced by training. Nature 427, 311–312 (2004).
Sehm, B. et al. Structural brain plasticity in Parkinson’s disease induced by balance training. Neurobiol. Aging 35, 232–239 (2014).
Villiger, M. et al. Relationship between structural brainstem and brain plasticity and lower-limb training in spinal cord injury: a longitudinal pilot study. Front. Hum. Neurosci. 9, 254 (2015).
Hains, B. C., Black, J. A. & Waxman, S. G. Primary cortical motor neurons undergo apoptosis after axotomizing spinal cord injury. J. Comp Neurol. 462, 328–341 (2003).
Raineteau, O. & Schwab, M. E. Plasticity of motor systems after incomplete spinal cord injury. Nat. Rev. Neurosci. 2, 263–273 (2001).
Ghosh, A. et al. Rewiring of hindlimb corticospinal neurons after spinal cord injury. Nat. Neurosci. 13, 97–104 (2010).
Lutti, A., Dick, F., Sereno, M. I. & Weiskopf, N. Using high-resolution quantitative mapping of R1 as an index of cortical myelination. Neuroimage 93, 176–188 (2014).
Cappe, C., Morel, A., Barone, P. & Rouiller, E. M. The thalamocortical projection systems in primate: an anatomical support for multisensory and sensorimotor interplay. Cereb. Cortex 19, 2025–2037 (2009).
Ionta, S. et al. Spinal cord injury affects the interplay between visual and sensorimotor representations of the body. Sci. Rep. 6, 20144 (2016).
Curt, A., Van Hedel, H. J. A., Klaus, D. & Dietz, V. Recovery from a spinal cord injury: significance of compensation, neural plasticity, and repair. J. Neurotrauma 25, 677–685 (2008).
Arsalidou, M., Duerden, E. G. & Taylor, M. J. The centre of the brain: Topographical model of motor, cognitive, affective, and somatosensory functions of the basal ganglia. Hum. Brain Mapp. 34, 3031–3054 (2013).
Tokuno, H. et al. Corticostriatal projections from distal and proximal forelimb representations of the monkey primary motor cortex. Neurosci. Lett. 269, 33–36 (1999).
Hou, J.-M. et al. Brain sensorimotor system atrophy during the early stage of spinal cord injury in humans. Neuroscience 266, 208–215 (2014).
Nishimura, Y. et al. Neural Substrates for the Motivational Regulation of Motor Recovery after Spinal-Cord Injury. PLoS ONE 6, e24854 (2011).
Carrera, E. & Tononi, G. Diaschisis: past, present, future. Brain 137, 2408–2422 (2014).
Freund, P. et al. Embodied neurology: An integrative framework for neurological disorders. Brain 139, 1855–1861 (2016).
Courtine, G. et al. Recovery of supraspinal control of stepping via indirect propriospinal relay connections after spinal cord injury. Nat. Med. 14, 69–74 (2008).
Bareyre, F. M. et al. The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats. Nat. Neurosci. 7, 269–277 (2004).
Hier, D. B. & Wang, J. Reduced cortical surface area in multiple sclerosis. Neurol. Res. 29, 231–232 (2007).
Worker, A. et al. Cortical thickness, surface area and volume measures in Parkinson’s disease, multiple system atrophy and progressive supranuclear palsy. PLoS ONE 9, e114167 (2014).
Maric, O., Zörner, B. & Dietz, V. Levodopa therapy in incomplete spinal cord injury. J. Neurotrauma 25, 1303–1307 (2008).
Lokk, J., Salman Roghani, R. & Delbari, A. Effect of methylphenidate and/or levodopa coupled with physiotherapy on functional and motor recovery after strok—a randomized, double-blind, placebo-controlled trial. Acta Neurol. Scand. 123, 266–273 (2011).
Schmierer, K., Scaravilli, F., Altmann, D. R., Barker, G. J. & Miller, D. H. Magnetization transfer ratio and myelin in postmortem multiple sclerosis brain. Ann. Neurol. 56, 407–415 (2004).
Langkammer, C. et al. Quantitative MR imaging of brain iron: a postmortem validation study. Radiology 257, 455–462 (2010).
Acknowledgements
We thank all the participants for taking part in this study, as well as the staff of the Department of Radiology and Neurology at the University Hospital Balgrist for their help in acquiring the MR images. In addition, we thank Jürgen Germann for valuable discussions, and Gabriel A. Devenyi for his technical assistance. A previous version of this paper was published as part of the thesis “Tracking structural changes in the spinal cord and brain as biomarkers for interventions after spinal cord injury” by Dr Eveline Huber (2018). This study was made possible by support from ‘The Oxford McGill ZNZ Partnership in the Neurosciences’ which provided a funding for mobility and training. This study was further supported by Wings for Life, Austria (WFL-CH-007/14), the International Foundation for Research in Paraplegia (IRP- P158), the European Union's Horizon 2020 research and innovation programme under the grant agreement No 681094 (‘NISCI’) and the Swiss State Secretariat for Education, Research and Innovation (SERI) under contract number 15.0137, the Clinical Research Priority Program (CRPP) Neurorehab UZH, the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013)/(ERC grant agreement no. 616905), the Federal Ministry of Education and Research (BMBF) under support code 01EW1711A & B in the framework of ERA-NET NEURON. The Wellcome Trust Centre for Neuroimaging is supported by core funding from the Wellcome Trust 0915/Z/10/Z. P.F. is funded by a SNF Eccellenza Professorial Fellowship grant (PCEFP3_181362/1). We thank SciNet for usage of Canada’s largest supercomputer platform, which is funded by the Canada Foundation for Innovation, NSERC, the Government of Ontario, Fed Dev Ontario, and the University of Toronto. MMC receives salary support from the Fonds de Recherche Québec Santé (Junior 2 program) and research funding from Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council of Canada, and the McGill University Healthy Brain for Healthy Lives program (a Canada First Research Excellence Fund).
Author information
Authors and Affiliations
Contributions
E.H., M.C., N.W. and P.F. were involved in the concept and design of the study. Analysis of data was conducted by E.H. and R.P. Data were discussed and interpreted by E.H., R.P., M.H., M.C. and P.F. E.H. and P.F. wrote the manuscript. All authors reviewed the manuscript and approved the final version.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Huber, E., Patel, R., Hupp, M. et al. Extrapyramidal plasticity predicts recovery after spinal cord injury. Sci Rep 10, 14102 (2020). https://doi.org/10.1038/s41598-020-70805-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-020-70805-5
- Springer Nature Limited
This article is cited by
-
Machine intelligence identifies soluble TNFa as a therapeutic target for spinal cord injury
Scientific Reports (2021)