Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 10 (1), 5815

Descending Motor Circuitry Required for NT-3 Mediated Locomotor Recovery After Spinal Cord Injury in Mice

Affiliations

Descending Motor Circuitry Required for NT-3 Mediated Locomotor Recovery After Spinal Cord Injury in Mice

Qi Han et al. Nat Commun.

Abstract

Locomotor function, mediated by lumbar neural circuitry, is modulated by descending spinal pathways. Spinal cord injury (SCI) interrupts descending projections and denervates lumbar motor neurons (MNs). We previously reported that retrogradely transported neurotrophin-3 (NT-3) to lumbar MNs attenuated SCI-induced lumbar MN dendritic atrophy and enabled functional recovery after a rostral thoracic contusion. Here we functionally dissected the role of descending neural pathways in response to NT-3-mediated recovery after a T9 contusive SCI in mice. We find that residual projections to lumbar MNs are required to produce leg movements after SCI. Next, we show that the spared descending propriospinal pathway, rather than other pathways (including the corticospinal, rubrospinal, serotonergic, and dopaminergic pathways), accounts for NT-3-enhanced recovery. Lastly, we show that NT-3 induced propriospino-MN circuit reorganization after the T9 contusion via promotion of dendritic regrowth rather than prevention of dendritic atrophy.

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. NT-3 failed to improve functional recovery after T9 transection.
a Diagram illustrates the experimental design. A complete spinal cord transection was made at the T9 vertebral level to transect all descending pathways. AAV-NT-3 or AAV-GFP (control) was injected bilaterally into sciatic nerves to retrogradely transport the NT-3 to lumbar MNs. BDA was bilaterally injected into cervical segments of C5–C6 to anterogradely label the cervical descending dPST at 6 wpi. A representative image of the boxed area (right) shows the lesion site, recognized by GFAP staining, in which BDA-labeled (BDA+) dPST axons were completely stopped at the rostral lesion border. Scale bar = 200 μm. Compass: R, rostral; C, caudal; D, dorsal; V, ventral. b Representative images show lumbar MN dendritic complexity from sham and T9 transected mice treated with either AAV-GFP or AAV-NT-3. Scale bar = 200 μm. c Curve plot represents the sholl analysis of lumbar MN dendritic distribution in each experimental group. Thick lines represent the average distribution of n = 3–4 animals per group; shaded area, 95% confidence intervals. d Line plot indicates the BMS score in three experimental groups, before and after the T9 transection, at different time points. Note that mice receiving AAV-NT-3 treatment after the T9 transection show increased MN dendritic arbors, but no locomotor improvement as compared to mice receiving AAV-GFP. Data are presented as mean ± SEM; n = 5–7 biologically independent animals per group. Two-way ANOVA followed by Tukey’s multiple comparison tests. AAV adeno-associated virus, NT-3 neurotrophin-3, GFP green fluorescent protein, BDA biotinylated dextran amines, dPST descending propriospinal tract, GFAP glial fibrillary acidic protein, MNs motoneurons. Source data are provided as a Source Data file.
Fig. 2
Fig. 2. NT-3 enhanced cortical and rubral control of hindlimb EMG response.
a Schematic diagram shows EMG potentials recorded in contralateral gastrocnemius muscle in response to electrical stimulation in the motor cortex of mice at 6 weeks after a T9 contusion. Representative examples of EMG traces were evoked by motor-cortex stimulation in the sham, SCI + AAV-GFP, and SCI + AAV-NT-3 treatment groups after SCI. The dotted line indicates the onset of EMG response in an AAV-NT-3-treated mouse; the arrows indicate the EMG latency. b Schematic diagram illustrates EMG activity in each group at 1 day after bilateral pyramidotomy. c Schematic diagram shows EMG potentials recorded in the contralateral gastrocnemius muscle in response to electrical stimulation in the red nucleus of mice at 6 weeks after the T9 contusion. Representative examples of EMG traces were plotted for each animal group. d Schematic diagram illustrates the EMG activities in each animal group at 1 day following bilateral lateral hemisection to transect the RST at the C5 vertebral level. e Quantitative analysis of cortico-stimulation-induced EMG amplitudes and latencies of the sham (44 stimulation sites from 6 mice), SCI + AAV-GFP (57 stimulation sites from 8 mice), and SCI + AAV-NT-3 (50 stimulation sites from 7 mice) animal groups. f Quantitative analysis of rubro-stimulation-induced EMG amplitudes and latencies in the sham (44 stimulation sites from 5 mice), SCI + AAV-GFP (49 stimulation sites from 6 mice), and SCI + AAV-NT-3 (45 stimulation sites from 6 mice) animal groups. The black dots represent independent stimulation sites from each group. The yellow lines from violin plots indicate mean ± SD. The dashed lines in histograms indicate the mean line from each group. ***P< 0.001. One-way ANOVA followed by Tukey’s multiple comparison test. EMG electromyography, CST corticospinal tract, RST rubrospinal tract. Source data are provided as a Source Data file.
Fig. 3
Fig. 3. Residual dPST responded to NT-3 by increasing its terminal sprouting and TrkC expression.
a Diagram illustrates BDA injections into cervical segments (C5–C6) to label the cervical dPST axons projecting to the lumbar spinal cord following the T9 contusion. b Representative black-to-white intensity scales demonstrating residual dPST axonal distribution in the ventral horn of the lumbar spinal cord (boxed area) after the T9 contusion in each experimental group. c Higher magnification of boxed area shows close apposition of BDA+ dPST axons with lumbar MNs (GFP+), which were more frequently seen in the SCI + AAV-NT-3 group than the SCI + AAV-GFP group. Scale bar = 40 μm. d Bar plot indicates the relative integrated density/mm2 of BDA+ dPST fibers in the lumbar ventral horn between groups; n = 4–6 mice per group (3–5 sections per animal). e Confocal z-stacks demonstrate that TrkC was expressed in the lumbar spinal cord, with some TrkC co-localization with BDA+ dPST terminals in the AAV-NT-3-treated group (arrowheads). Scale bar = 3 μm. f Bar plot shows TrkC relative integrated density/mm2 and individual values for each animal (3–5 sections per animal). Violin plot represents the ratio of double-labeled puncta (TrkC and BDA) to BDA relative integrated density in the corresponding sections. The yellow lines from the violin plots indicate mean ± SD. Data are presented as box plots with center lines indicating medians, boxes representing 25th to 75th percentiles, also known as the interquartile range (IQR), and whiskers representing data points within 1.5 times the IQR; n = 4–6 biologically independent animals per group. *P < 0.05; **P < 0.01; ***P < 0.001. One-way ANOVA followed by Tukey’s post-hoc test. dPST descending propriospinal tract, TrkC tropomyosin receptor kinase C. Source data are provided as a Source Data file.
Fig. 4
Fig. 4. Thoracic dPNs functionally contributed to NT-3-mediated locomotor recovery.
a A schematic drawing shows the experimental design. The contusive mice with either AAV-GFP or AAV-NT-3 treatment received two viral injections, i.e. HiRet-TRE-EGFP-eTeNT into the L3–L4 segments and Tet-On/AAV into the T5–T7 segments at 6 wpi. Three weeks later, Dox was administered to induce the expression of eTeNT. Locomotor function was evaluated using BMS, grid walking, and rotarod assessments before, during and after Dox administration. b A diagram shows the strategy to silence the thoracic (T5–T7) dPNs with dual virus injections. Insets of the boxed area indicate the transverse view of Tet-On/AAV injections and a representative image of GFP+ dPNs infected by dual viruses. Scale bar = 100 μm. Bar plot reveals the total number of dual-virus infected, GFP+ dPNs in T5–T7 spinal segments among three groups with individual values for each animal. Data are presented as box plots with center lines indicating medians, boxes representing 25th to 75th percentiles, and whiskers representing data points within 1.5 times the IQR. n = 9–10 animals per group. ***P < 0.001; one-way ANOVA followed by Tukey’s post-hoc test. c Line plots show changes of hindlimb locomotor function with grid walking and rotarod testing before, during and after Dox administration. Data are presented as mean ± SEM. n = 9–10 biologically independent animals per group. *P < 0.05, **P< 0.01, ***P < 0.001 (SCI + AAV-GFP vs SCI + AAV-NT-3); two-way ANOVA followed by Tukey’s multiple comparisons test. Bar plots detail alterations in locomotor behaviors, with individual animals from each experimental group, in response to Dox treatment (n = 9–10 animals per group). *P < 0.05; **P < 0.01; one-way ANOVA followed by Tukey’s post-hoc test. d Scatter plots indicating the correlations of the number of double-infected GFP+ dPNs with Dox-induced locomotor deficits in grid walking and rotarod assessments. The middle lines in scatter plots indicate the regression lines and the shades represent 95% confidence intervals. SCI spinal cord injury, BMS Basso Mouse Scale, Dox doxycycline, wpi weeks post-injury. Source data are provided as a Source Data file.
Fig. 5
Fig. 5. NT-3 promoted propriospino-MN reorganization after staggered double hemisection.
a Schematic shows unilateral injection of BDA (red circle) into the T9 (right) spinal cord in mice with T7 (right) and T12 (left) lateral hemisections. b Representative image shows an anti-GFAP stained horizontal section of the spinal cord after staggered hemisections. Arrowheads indicate the midline. Scale bar = 600 μm. c Representative images show BDA+ dPST axonal distribution in transverse sections of the lumbar spinal cord (L3–L4) in different experimental groups. Box plots indicate BDA+ dPST fiber density index (left side relative to the right side of the spinal cord) and mean density of dPST fibers per area of left L3–L4. Scale bar = 500 μm. d High magnification images show the BDA+ dPST axons innervating CTB-labeled (CTB+) lumbar MNs (left L3–L4). Arrowheads indicate the appositions between dPST axons and MNs in a 3D view. Box plots indicate the mean density of dPST fiber per area and the number of colocalized puncta in the left ventral horn of L3–L4. Scale bar = 20 μm. e Schematic shows the tcMMEP response recorded from left gastrocnemius muscle, were evoked by a magnetic field applied to the motor cortex. Representative tcMMEP signals are plotted among three groups at 6 weeks after the staggered lesions. The plots report the average latency and peak-to-peak amplitude of tcMMEPs among three groups. f Schematic and representative stacked traces show the trajectory of left hindlimb endpoints between groups at 6 weeks after staggered lesions. Box plots indicate average step lengths and heights between groups. Data are presented as box plots with center lines indicating medians, boxes representing 25th to 75th percentiles, and whiskers representing data points within 1.5 times the IQR; n = 5–6 biologically independent animals per group; *P < 0.05; **P < 0.01; ***P < 0.001. One-way ANOVA followed by Tukey’s post-hoc test. tcMMEPs transcranial magnetic motor-evoked potentials, CTB cholera toxin subunit B. Source data are provided as a Source Data file.
Fig. 6
Fig. 6. NT-3 induced neuroplasticity of lumbar MNs following contusion.
a Bar graph shows local NT-3 expression levels, measured by ELISA, in lumbar cord before and after contusions followed by AAV-NT-3 treatment. Data are presented as mean ± SEM; n = 3–4 mice per each time point; ***P< 0.001 vs baseline (before SCI). One-way ANOVA followed by Tukey’s post-hoc test. b Representative images show synapse-like contacts (arrowheads) between BDA+ dPST axons and CTB+ lumbar MNs in sham and contusive mice with AAV-NT-3 post-treatment at 2 and 4 wpi. Scale bar = 10 μm. c Quantitative analysis of the number of triple-labeled appositions between BDA+ dPST, CTB+ MNs, and synaptophysin per each MN per section among three groups. Dots in violin plots represent triple appositions from 3–4 mice in each group. The yellow lines indicate mean ± SD. **P < 0.01, ***P < 0.001; one-way ANOVA followed by Tukey’s post-hoc test. d Representative images show lumbar MN dendritic complexity in the three experimental groups. Curve plot represents sholl analysis of MN morphology among groups. Thick lines represent the average distributions of MN dendrites for each experimental group (n = 3 mice per group); shaded areas represent 95% confidence intervals. Line plot details the difference in the dendritic complexity of MNs at distance from somas between 50 and 160 μm. Data are presented as mean ± SEM. Two-way ANOVA followed by Tukey’s post-hoc test. e Representative images show MN dendritic complexity in sham and T9 contusion mice with AAV-NT-3 pre-treatment and MN dendritic morphology examined at 2 and 4 wpi. Curve plot represents sholl analysis of MN dendritic distribution in the three groups. Thick lines represent the average distributions of MN dendrites among groups (n = 3 mice per group); shaded areas represent 95% confidence intervals. Line plot details the difference in the dendritic complexity of lumbar MNs at distances from somas between 50 and 160 μm. Data are presented as mean ± SEM. Two-way ANOVA followed by Tukey’s post-hoc test. Source data are provided as a Source Data file.

Similar articles

See all similar articles

References

    1. Kiehn O. Decoding the organization of spinal circuits that control locomotion. Nat. Rev. Neurosci. 2016;17:224–238. doi: 10.1038/nrn.2016.9. - DOI - PMC - PubMed
    1. Ferreira-Pinto MJ, Ruder L, Capelli P, Arber S. Connecting circuits for supraspinal control of locomotion. Neuron. 2018;100:361–374. doi: 10.1016/j.neuron.2018.09.015. - DOI - PubMed
    1. Song J, Ampatzis K, Bjornfors ER, El Manira A. Motor neurons control locomotor circuit function retrogradely via gap junctions. Nature. 2016;529:399–402. doi: 10.1038/nature16497. - DOI - PubMed
    1. Marder E, Bucher D. Central pattern generators and the control of rhythmic movements. Curr. Biol. 2001;11:R986–R996. doi: 10.1016/S0960-9822(01)00581-4. - DOI - PubMed
    1. Lemon RN. Descending pathways in motor control. Annu. Rev. Neurosci. 2008;31:195–218. doi: 10.1146/annurev.neuro.31.060407.125547. - DOI - PubMed
Feedback