Pathological changes of distal motor neurons after complete spinal cord injury
- 227 Downloads
Traumatic spinal cord injury (SCI) causes serious disruption of neuronal circuits that leads to motor functional deficits. Regeneration of disrupted circuits back to their original target is necessary for the restoration of function after SCI, but the pathophysiological condition of the caudal spinal cord has not been sufficiently studied. Here we investigated the histological and biological changes in the distal part of the injured spinal cord, using a mice model of complete thoracic SCI in the chronic stage (3 months after injury). Atrophic changes were widely observed in the injured spinal cord both rostral and caudal to the lesion, but the decrease in area was mainly in the white matter in the rostral spinal cord while both the white and gray matter decreased in the caudal spinal cord. The number of the motor neurons was maintained in the chronic phase of injury, but the number of presynaptic boutons decreased in the lumbar motor neurons caudal to the lesion. Using laser microdissection, to investigate gene expressions in motor neurons caudal to the lesion, we observed a decrease in the expressions of neuronal activity markers. However, we found that the synaptogenic potential of postsynapse molecules was maintained in the motor neurons after SCI with the expression of acetylcholine-related molecules actually higher after SCI. Collectively, our results show that the potential of synaptogenesis is maintained in the motor neurons caudal to the lesion, even though presynaptic input is decreased. Although researches into SCI concentrate their effort on the lesion epicenter, our findings suggest that the area caudal to the lesion could be an original therapeutic target for the chronically injured spinal cord.
KeywordsSpinal cord injury Chronic phase Laser microdissection Cell-selective gene expression Synaptogenic potential Motor neurons
α-amino-3-hydroxy-5-methyl-4-isoxazole propanoic acid receptor
Calcium channel voltage-gated channel auxiliary subunit gamma-2
Ca2+/calmodulin-dependent protein kinase-2a
Central nervous system
Fluorescence-activated cell sorting
Guanylate kinase-associated protein
Glutamate receptor-interacting protein-1
Kinesin superfamily protein
Myelin basic protein
Postsynaptic density protein-95
Synapse associated protein-97
Spinal cord injury
Vesicular acetylcholine transporter
Vesicular glutamate transporters-2
Traumatic spinal cord injury (SCI) is a severely debilitating injury with permanent motor/sensory dysfunction that imposes considerable mental and economic burdens  and leads to a marked reduction in the quality of life . Patients with incomplete SCI experience a recovery in function that can begin immediately after injury and continue for several weeks after injury [3, 4] while the functional recovery in patients with complete SCI is severely limited. Ongoing research into the treatment for SCI has led to the development of numerous strategies to improve the impairment of incomplete SCI, but the unfortunate truth is that the treatment options for patients with complete SCI are virtually nonexistent [5, 6]. Therefore, there is a great demand for developing treatment strategies for complete SCI; but before we can treat complete SCI, we need to understand the pathophysiological condition of the spinal cord following complete SCI.
After SCI, axons and dendrites that lost connection to their original neural pathways degenerate in a process called Wallerian degeneration or axonal dieback, which starts at the site of injury and proceeds in a direction away from the injury epicenter [7, 8]. Previous reports exploring the pathological events that occur in the sites remote to the lesion in the chronic phase of SCI [9, 10] provide information that is vital when considering future strategies for establishing treatment approaches for SCI. Given that there is evidence demonstrating that the synaptic connections of lumbar motor neurons affect locomotor function even after incomplete thoracic SCI , it is clear that the distal neuronal components should not be considered irrelevant. Indeed, a better understanding of the synaptic pathology of the motor neurons distal to the lesion may turn out to be a critical component in the development of novel treatments to promote axonal connectivity and bring about meaningful functional recovery.
In this study, we investigated the histological and biological changes that occur distal to the lesion in a severe thoracic SCI mice model. Utilizing laser microdissection (LMD) to isolate single cells for study, we examined the synaptogenic potential of motor neurons in lumbar spinal cords (which are remote to the lesion) and demonstrated that even though the synaptic input decreases in distal motor neurons after severe SCI, the synaptogenic potential is maintained. Our findings confirm that the neural components caudal area to the lesion undergo pathological change after severe SCI, but contrary to expectations, our results demonstrate that the area caudal to the lesion could be a potential target to improve functional outcome after SCI.
Severe SCI leads to atrophic changes of the spinal cord distal to the lesion
The descending presynaptic inputs to lumbar motor neurons decrease in the chronic phase of SCI
Cell-specific gene expression analysis reveals decreased neuronal activity in lumbar motor neurons of the chronically injured spinal cord
Lumbar motor neurons maintain expression of post-synaptic molecules in the chronic phase of SCI
Motor neurons in the chronically injured spinal cord exhibit altered expression of factors involved in synapse formation
The expression of neurotrophic factor receptors were also examined because they have been shown to affect synapse formation . In the isolated motor neurons, the gene expression of c-Ret (tyrosine kinase receptor), GFRa1 (GDNF family receptor a), and p75 (NGF receptor) were significantly increased in the Tx3m group compared to the Naive group, while the gene expression of TrkA (Tropomyosin-receptor-kinase A), TrkB, and TrkC were comparable between the two groups (Fig. 8c). Although the significance of these changes are not immediately apparent, changes in the expression of axon guidance molecules and neurotrophic factor receptors may reflect an altered potential for synaptogenesis in the lumbar motor neurons of the chronically injured spinal cord.
Cholinergic activity is maintained in lumbar motor neurons in the chronic phase of SCI
In this study, we reported three significant findings concerning the spinal cord in the chronic phase after thoracic SCI in comparison to a normal spinal cord. First, the size of the spinal cord decreases not only at the widely-known are near the injury epicenter, but also in a wide range extending rostrally and caudally. Second, the number of the presynaptic boutons and expression of neuronal activity markers are significantly lower in lumbar motor neurons, illustrating a decreased state of activity that occurs secondary to the decrease in descending synaptic input. Third, the lumbar motor neurons maintain the expression of postsynaptic molecules and cholinergic activity even though they are in a semi-dormant state. Notably, LMD method provided us with reliable data of cell-selective gene expressions in lumbar motor neurons regarding synaptogenic potential, axonal guidance molecules, neurotrophic factor receptors, and axonal transport molecules. The lumbar motor neurons maintained both in number and in size even in the chronic phase of SCI, and these neurons would not be dormant after SCI and possibly restore the neuronal activity by raising the synaptic input to distant caudal part from the lesion site.
Our results revealed a decrease in the white matter both rostral and caudal to the lesion. In the rostral spinal cord, there was a significant reduction of the dorsal column in the rostral spinal cord, which corresponds to the CST. The prominent decrease of the dorsal column area suggests that the retrograde dieback of the CST fibers demonstrated by Seif and colleagues  may be a major factor in the reduction of the rostral white matter area. There was a similar reduction of the dorsal column in the caudal spinal cord, although not as prominent, and we believe that the anterograde degeneration of axons (also known as Wallerian degeneration) is responsible for this change. When we examined the gray matter, there was a significant decrease in the area of the gray matter caudal to the lesion, but no change was detected in the rostral spinal cord. Since the number and size of the motor neurons were not affected by the spinal cord injury, we believe this difference is due to the degeneration of axonal fibers within the gray matter. Tracing experiments have demonstrated the presence of CST fibers within the gray matter [24, 25, 26], presumably from the dorsal column to the ventral horn where they form synapses with secondary motor neurons. The collective loss of CST fibers within the gray matter at all levels of the lumbar spinal cord due to anterograde degeneration may lead to a reduction of the gray matter caudal to the lesion. In the spinal cord rostral to the lesion, the CST fibers maintain connection with their targets and do not degenerate. Furthermore, since ascending sensory fibers transverse the gray matter at the level they enter the spinal cord and do not significantly cross into the gray matter on their path to the brainstem [27, 28, 29], the gray matter rostral to the lesion should not be largely affected.
Immunohistological analysis of the area distal to the injury confirmed that the number and size of lumbar motor neurons were unchanged even in the chronic phase. Although the number of presynaptic boutons in lumbar motor neurons were significantly decreased (Fig. 3), it is noteworthy that transection of the spinal cord did not lead to the complete loss of presynaptic molecules in the distal part of the lesion. In fact, we observed spatial connectivity between the presynaptic and postsynaptic molecules in lumbar motor neurons. These findings support the concept that the endogenous neuronal circuits remain intact within the distal spinal cord even though the circuits to the proximal neural pathways are severely disrupted . Given that the synaptic transmission within the distal spinal cord is maintained , therapeutic interventions that bridge the impaired circuitry of the lesion could be a potential strategy to overcome the neural roadblock of the injury site.
Both vGluT2-positive and Bassoon-positive synaptic boutons decreased in the distal spinal cord after a transection injury. Considering that a recent report demonstrated that excitatory neurons directly contribute to motor function , the decrease in excitatory vGluT-positive synapses may be a potential intervention target to treat spinal cord injury. We did not specifically look into the changes that occur in the inhibitory synapses, but the decrease in Bassoon-positive synapses suggests a parallel decrease in inhibitory synapses [33, 34]. Although Bassoon has been used as a pan-synaptic marker that includes both excitatory and inhibitory synapses , Bassoon immunoreactivity encompasses most inhibitory synapses and gives it a more inhibitory profile.
The gene expression profile of the lumbar motor neurons from the Tx3m group revealed a significant decrease in the repulsive axon guidance molecules Ephrin-A2 and Ephrin-B2 and the attractive axon guidance molecule Netrin-G1, while the neurotrophic factors c-Ret, GFRα1, and p75 increased in the Tx3m group compared to the naive group. The decrease in repulsive axon guidance molecules and the increase in neurotrophic factors can be interpreted as the physiological reaction of lumbar motor neurons to the transection injury, attempting to promote axonal regeneration of descending tracts to their targeted lumbar motor neurons [36, 37]. The significance of the decreased Netrin-G1 is elusive, but suggests a potential therapeutic target to promote axonal regeneration in the injured spinal cord . The role of axon guidance molecules has not been sufficiently studied in the injured spinal cord , and future studies are required to investigate their function during the regenerative process after spinal cord injury.
Rehabilitation programs such as physical therapy and occupational therapy are established treatment strategies for SCI patients [40, 41], that have been reported to be effective even for complete SCI patients . Physiological exercise after SCI is reported to activate the reorganization of neuronal network in the distal part of the lesion , with passive locomotive training even inducing recovery of motor function in cats with complete SCI . The effects of electrical stimulation are believed to induce functional recovery after SCI through a similar process , but the mechanisms underlying the restorative effect of rehabilitation for SCI remains elusive. Since our results demonstrate that the expressions of postsynaptic molecules are maintained in motor neurons distal to the lesion, rehabilitation program may promote synaptic reorganization of lumbar motor neurons that rewires the connection with proximal circuits. Our findings are in line with other studies that strongly suggest that the stimulation of muscles and peripheral nerves could be the catalyst for a reorganization of the neural network to the distal spinal cord.
Contrary to our expectation, the synaptogenic potential of lumbar motor neurons were maintained in the chronic phase of SCI. This suggests that if a means to transmit neural signals across the SCI lesion site was established, the conduction of proximal signals from the brain to peripheral systems may be reestablished. Indeed, Lu and colleagues demonstrated that the neural stem cells transplanted into the injured spinal cord formed synaptic connections with lumbar motor neurons distal to the lesion . This may also imply that we should reconsider the ideal location to transplant neural stem cells into the injured spinal cord; the lesion epicenter may not necessarily be the ideal place for stem cell transplantation when considering the synaptic connectivity with the motor neurons of the distal spinal cord.
In this study, LMD was an indispensable tool to analyze cell-selective gene expression in lumbar motor neurons. By reliably collecting only motor neurons, we were able to obtain critical information pertaining to the synaptogenic factors, axonal guidance molecules, neurotrophic factor receptors, and axonal transport molecules of the lumbar motor neurons. Previously, fluorescence-activated cell sorting (FACS) has thus far been the prevailing method for cell-specific analyses . However, LMD procedure provides certain advantages over FACS. First, LMD enables the collection of neuronal cells without disrupting their distinctive structures, such as long axons and dendrites. This has tremendous value, because the majority of synaptic mRNAs are localized throughout the axonal and dendritic terminals which are hundreds of micrometers distant from the neuronal cell body . Furthermore, FACS requires enzymatic dissociation to isolate neuronal cells from spinal cords and eliminate myelin and extracellular matrix protein, but this enzymatic dissociation often results in a loss of cellular mRNA and proteins . We acknowledge the advantage of FACS in analyzing blood cells and culture cells, but in the analysis of cells integrated into a cellular matrix or a neuronal network, the ability to analyze the gene expression of cells in their original environment and morphology with intact cellular connections is a tremendously useful tool, especially for the investigation of neuronal cells in the CNS.
In conclusion, we clarified the pathologic changes of the lumbar spinal cord in the chronic phase of complete thoracic SCI. The lumbar motor neurons remote to the lesion received less synaptic input and exhibited lower neuronal activity. However, our findings reveal that these motor neurons are not dormant and persistently express postsynaptic molecules even in the chronic phase of SCI. Indeed, the cholinergic potential of lumbar motor neurons were even higher than in naive animals, suggesting that these neurons are primed and ready for any input that comes their way. Our findings suggest that the neural pathways distal to the SCI lesion may be relatively intact, and that what is desperately desired is a dependable means to restore synaptic input to the lumbar motor neurons.
Adult female C57/BL6 mice were anesthetized via an intraperitoneal injection of pentobarbital (75 mg/kg) and received a complete spinal cord transection at the 9th thoracic level (n = 30). Through a longitudinal incision, the paraspinal muscles were sharply dissected and a T9 laminectomy was performed to expose the spinal cord. The spinal cord was completely transected at the T9 level with microscissors (including the dura mater). The cut ends of the spinal cord were lifted with small forceps to assure complete transection. The muscles and skin were subsequently sutured in layers. All animal studies were conducted in accordance with the institutional guidelines and regulations for animal experiments and were approved by the Committee of Ethics on Animal Experiment in Faculty of Medicine, Kyushu University.
After the mice were transcardially fixed with 4% paraformaldehyde, the spinal cord was removed, dehydrated and embedded in OCT compound. The frozen tissue was cut in the sagittal or axial plane in 16 μm sections. For immunostaining, the sections were stained overnight at 4 °C with primary antibodies against MBP (myelin marker, 1:200, rat, Chemicon), NF200 (neuronal fiber marker, 1:200, rabbit, Sigma-Aldrich), NF200 (1:200, mouse, Sigma-Aldrich), 5-HT (serotonergic marker, 1:200, goat, ImmunoStar), VGluT2 (presynaptic marker, 1:200, guinea pig, Synaptic Systems), PSD95 (postsynaptic marker, 1:200, rabbit, Invitrogen), Hu (neuronal marker, 1:1000, human, a gift from Dr. Robert Darnell, The Rockefeller University, New York, NY, USA), Bassoon (pan-presynaptic marker, 1:200, mouse, Stressgen), NeuN (neuronal marker, 1:1000, mouse, Chemicon), c-fos (neuronal activity marker, 1:200, rabbit, Santa Cruz Biotechnology), or ChAT (1:100, goat, Chemicon). Then the sections were then incubated at room temperature for 1 h with Alexa Fluor-conjugated secondary antibodies (1:200, Invitrogen). Nuclear counterstaining was obtained using Hoechst 33342 (Molecular Probes).
The injured spinal cords were resected at 3 months after injury, immediately frozen in dry ice/hexane, and stored at − 80 °C. The tissues were sectioned into 16 μm thick slices using a cryostat at − 20 °C and were mounted on PEN membrane slides. The sections were then fixed in ice-cold acetone for 2 min. Nissl-stained motor neurons in the ventral horn that were over 20 μm in diameter were selected based on morphological criteria [50, 51] and dissected with the LMD 6500 system (Leica Microsystems) into a microcentrifuge tube cap placed directly beneath the section. The tube cap was filled with 75 μl of RLT buffer (Qiagen). For each sample, 500 cells were dissected from one series of axial sections.
Total RNA was isolated from the spinal cord using an RNeasy Micro Kit (Qiagen). For complementary DNA (cDNA) synthesis, the reverse transcriptase reaction was performed using a Prime Script first-strand cDNA Synthesis Kit (Takara Bio). Quantitative RT-PCR was performed using primers specific to the genes of interest (Additional file 4: Table S1) and SYBR Premix Dimmer Eraser (Takara Bio). The data were normalized to glyceraldehyde-3-phosphate dehydrogenase. RT-PCR was conducted using a Thermocycler (Biometra, Göttingen, Germany) and products were detected by electrophoresis and ethidium bromide staining.
Image acquisition and quantitative analysis
All images were obtained using a LSM510 laser scanning microscope system (Zeiss) or a BZ-9000 digital microscope system (Keyence). To quantify the number of neurons, we performed immunostaining with an anti-Hu (neuronal marker) antibody in serial sections and counted the number of immunoreactive cells that were present 6 mm distant from the lesion epicenter. The motor neurons (cell diameter > 20 μm) in the ventral horn were selected under the microscope based on morphological criteria . For quantification of the presynaptic boutons (Bassoon and VGluT2) in lumbar motor neurons, immunostaining of sagittal sections was performed using each antibody. The algorithms for counting the number of synaptic boutons were performed as described previously [52, 53]. Briefly, 40 optical slices of axial sections at × 1000 magnification were acquired 0.4 μm intervals and reconstructed into a single image, and immunopositive synaptic boutons in each motor neuron were counted. Two motor neurons were analyzed from each mouse (8 neurons from 4 mice in total) and the total number of synaptic boutons observed in each motor neuron was counted and presented as a bar graph. Only neurons with a nucleus and proximal neurite processes that were confirmed by focusing up and down through the thickness of the section were included in the cell counts.
The Wilcoxon rank sum test was used to compare the medians of the data for quantitative RT-PCR as well as the area of spinal cord, the area of immunopositive area, and the number of synaptic boutons. In all statistical analyses, significance was defined as P < 0.05. The values for groups are presented as the average ± SEM. All statistical analyses were carried out using the JMP software program (version 13; SAS Institute).
This study was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (JSPS) Grant Numbers JP16H05450, JP16K15668, Overseas Research Fellowships for KY; and Japan Agency for Medical Research and Development (AMED) under Grant Number JP18gm6210003; and a grant from the General Insurance Association of Japan.
Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional files.
KY designed and performed most of the experiments with technical help from KKubota, KKobayakawa, TS, MH, and KKijima. TM, HK, YO, and YN provided experimental support and ideas for the project. SO designed the studies, supervised the overall project, and performed the final manuscript preparation. All authors read and approved the final manuscript.
All surgical procedures and experimental manipulations were approved by the Committee of Ethics on Animal Experiment in Faculty of Medicine, Kyushu University. Experiments were conducted in accordance with the institutional guidelines and regulations for animal experiments.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 2.Kirchberger I, Biering-Sorensen F, Charlifue S, Baumberger M, Campbell R, Kovindha A, Ring H, Sinnott A, Scheuringer M, Stucki G. Identification of the most common problems in functioning of individuals with spinal cord injury using the International Classification of Functioning, Disability and Health. Spinal Cord. 2010;48:221–9.CrossRefGoogle Scholar
- 35.tom Dieck S, Sanmarti-Vila L, Langnaese K, Richter K, Kindler S, Soyke A, Wex H, Smalla KH, Kampf U, Franzer JT, et al. Bassoon, a novel zinc-finger CAG/glutamine-repeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol. 1998;142:499–509.CrossRefGoogle Scholar
- 49.Garg A, Houlihan DD, Aldridge V, Suresh S, Li KK, King AL, Sutaria R, Fear J, Bhogal RH, Lalor PF, Newsome PN. Non-enzymatic dissociation of human mesenchymal stromal cells improves chemokine-dependent migration and maintains immunosuppressive function. Cytotherapy. 2014;16:545–59.CrossRefGoogle Scholar
- 50.Hilton JB, Mercer SW, Lim NK, Faux NG, Buncic G, Beckman JS, Roberts BR, Donnelly PS, White AR, Crouch PJ. CuII (atsm) improves the neurological phenotype and survival of SOD1G93A mice and selectively increases enzymatically active SOD1 in the spinal cord. Sci Rep. 2017;7:42292.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.