Article Figures & Data
Figures
- Figure 1.
Western blot analysis of synapse-associated proteins in lumbar spinal cord tissue from intact, SCI and control, vehicle- or NSC23766-treated animals. Quantification was performed on three blots for each protein. To normalize for variation between blots, β-actin control protein level was set to 100% in each lane and proteins of interest were compared across treatment groups. A, Phosphorylated Rac1 protein was probed in the three treatment groups with a visible band at ∼21 kDa. B, Significantly increased levels of phosphorylated Rac1 were found after SCI plus veh treatment compared with uninjured controls. Post-SCI levels were reduced with intrathecal NSC23766. C, D, No significant difference in the levels of total Rac (∼21 kDa) was observed between the treatment groups. E, F, PSD-95 band is visible at ∼95 kDa (E), and significantly higher levels were observed in SCI plus veh compared with intact levels (F). NSC23766 treatment also significantly reduced PSD-95 levels below that of SCI plus veh and intact animals. G, A cortactin band was observed at ∼80 kDa along with breakdown products. H, No differences were observed for levels of cortactin between groups. I, J, We also observed a trend, which did not reach statistical significance, toward an increase in the levels of synaptophysin after SCI. Graphs are mean ± SEM. *p < 0.05; **p < 0.01.
- Figure 2.
Coronal sections of spinal cord tissue from the lumbar enlargement were Golgi-stained and examined for dendritic spines. A, A representative neuron whose soma (black arrow) is located in lamina 5. B, A magnified view of the neuron identified in A (inset). C, A ∼50 μm dendritic segment from the intact group (see C′, inset). D, After SCI, there is an increase in spine density and the appearance of larger spine head structures. E, NSC23766 reduces the number of spines and decreases the overall dimension of spines after SCI. F, High-power images of two representative thin-shaped spines found on deep lamina neurons in the spinal cord dorsal horn. G, Mushroom spines appeared with swollen head structures with narrow spine necks. C′–E′, Inset from white boxes shows visible thin and mushroom spines distributed along a 10 μm span of the dendrite from each treatment group. Scale bar is the same for C–E and C′–E′. H, All dendritic spines from five neurons randomly chosen from each treatment group were measured for spine length. SCI plus veh treatment increased the length of spines compared with intact animals; spine length was significantly decreased with NSC23766 treatment (*p < 0.001). I, SCI plus veh treatment also increased the spine head diameter of dendritic spines compared with intact animals. NSC23766 treatment significantly decreased the spine head diameter compared with both SCI plus veh treated animals and intact, uninjured animals (*p < 0.001). Scale bars: A, 500 μm; B, 100 μm; C–E, 10 μm; C′–E′, 5 μm; F, G, 1 μm.
- Figure 3.
Neurolucida reconstructions reveal the location of identified neurons (A–C), the complete spatial distribution and density of thin- and mushroom-shaped spines. A′, An identified neuron has several primary dendritic branches arising from the soma. A 50 μm length of the primary dendrite from this neuron is shown in A″, which reveals a single mushroom spine (red dot) and multiple thin spines (blue dots) along its length. B′, A neuron from an SCI plus veh treated animal. B″, A sample 50 μm length of the primary dendrite shows a higher density of thin spines and mushroom spines. C′, A neuron reconstructed from an SCI and NSC23766 Rac1 inhibitor-treated animal. C″, A significant decrease in thin and mushroom spine density is observed on the dendrites from NSC23766-treated animals. Dendrites in A″, B″, and C″ are oriented with bottom closest to soma. Scale bars: A–C, 500 μm; A′–C′, 50 μm; A″–C″, 10 μm.
- Figure 4.
Analysis of reconstructed neurons reveals differences in spine density (A–C), spine distribution (D–F), and spine shape (G, H) in the three treatment groups. A–C, Total spine density (A), thin-shaped spine density (B), and mushroom-shaped spine density (C) increased after SCI plus veh treatment compared with intact animals. A–C, NSC23766 treatment reduces total spine, thin spines, and mushroom spine density after SCI (*p < 0.05; **p < 0.001). D, The distribution of spines shifts significantly after SCI. Significantly more spines are observed more proximal to the soma after SCI and vehicle treatment compared with intact animal group (*p < 0.001). NSC23766 Rac1 inhibitor treatment significantly reduced the spine density at the 350 μm distance compared with intact animals (**p < 0.001). E, Similarly, thin spines increase in density in regions more proximal to the soma after SCI plus veh treatment compared with intact animals. NSC23766 decreases this density distribution so that it is not significantly different compared with intact animals. F, SCI results in a significant increase in mushroom spines in two region, 50 and 100 μm, distances from the soma compared with intact animals (*p < 0.001). No significant differences were observed between intact and SCI plus NSC23766-treated animals. G, Summary diagram demonstrating the increase in spine density and the proximal shift in spine distribution after SCI. Sample size for neurons analyzed for each spine category in A–F is shown in parentheses in A–C. Graphs are mean ± SEM.
- Figure 5.
NEURON simulation demonstrates that SCI-induced dendritic spine remodeling (E) alters the input–output function of a postsynaptic neuron. A, Presynaptic input onto a dendritic spine is modeled using an α-function to simulate the kinetics of an excitatory AMPA synapse (τ max conductance, 5 ms; gmax, 0.05 μS). B, The shape of the spine influences the EPSP recorded at the base of the dendritic spine neck (as shown in D) in intact, SCI plus veh, and SCI plus NSC23766 (as shown in D; H-H channels in gray region head compartment). C, Post-SCI changes in spine density and distribution (Fig. 4) result in a second action potential. NSC23766 decreased spine density and resulted in output similar to output cells from intact animals.
- Figure 6.
Intrathecal administration of NSC23766 attenuated peripherally evoked activity in WDR units from SCI animals 31 d after injury. A, A representative unit from an intact animal displaying evoked responses to stimuli (PB, 1.44 g/mm2, 538 g/mm2, 0.39 g, 1.01 g, and 20.8 g refer to phasic brush, pressure, pinch, and von Frey filaments of increasing intensities applied for 10 s). B, After SCI, the peristimulus histogram shows increased evoked responses to all peripheral stimuli. Evoked discharged rates were between 45 and 60 impulses/s. Phasic brush, pressure, and pinch resulted in elevated responses. von Frey filament stimulation resulted in a graded increase. C, Peripherally evoked responses of all sampled WDR units decreased significantly after NSC23766 treatment. SCI plus veh WDR unit peristimulus histogram responses (gray) are overlaid with SCI plus NSC23766 peristimulus histogram responses (black). D, SCI significantly (*p < 0.05; **p < 0.001) increased the evoked responses to all peripheral stimuli compared with intact animals. NSC23766 significantly (*p < 0.05; **p < 0.001) reduced SCI-induced increases in all peripherally evoked responses. Evoked responses with NSC23766 remained significantly higher for phasic brush and 1.01 and 20.8 g/mm2 von Frey filament stimulation (*p < 0.05; **p < 0.001) compared with intact. Graphs are mean ± SEM.
- Figure 7.
Behavioral assays for pain and locomotor function. Intact animals demonstrated expected levels for tactile allodynia (A), thermal hyperalgesia (B), and displayed competent locomotor function (C). The dashed line represents group mean for intact stimulus thresholds. After SCI, however, stimulus threshold for hindpaw withdrawal decreased (open circles) in response to both tactile and thermal stimulus intensities. Intrathecal NSC23766 administration over 3 d progressively restored tactile and thermal thresholds compared with SCI plus veh (*p < 0.05; **p < 0.01). No significant difference in locomotor competence between SCI plus veh and SCI plus NSC23766 animals (p = 0.266). Compared with intact animals, both performed poorly (**p < 0.001). Graphs are mean ± SEM.
Tables
Dorsal–ventral (μm) Medial–lateral Total dendrite length (μm) No. of primary dendrites Length of primary dendrite (μm) Primary dendrites with secondary branches (%) Intact 39.2 ± 8.5 24.7 ± 6.8 1372 ± 755 5.3 ± 2.0 50.2 ± 63 54.7 ± 28.7 SCI + veh 38.3 ± 14.2 21.5 ± 5.2 1419 ± 528 4.8 ± 1.2 44.7 ± 54.7 54.6 ± 26.6 SCI + anti-Rac1 38.2 ± 6.5 21.7 ± 5.0 1550 ± 539 4.6 ± 0.9 35.6 ± 35 62.4 ± 20.1 Data are shown as mean ± SD.
Supplemental Data
Files in this Data Supplement:
- supplemental material - Supplemental Legend
- supplemental material - Supplemental Movie 1
- supplemental material - Supplemental Movie 2
- supplemental material - Supplemental Movie 3
