Article Figures & Data
Figures
- Figure 1.
Schematic of the rVRG-PhMN-diaphragm circuit. Schematic of intact (A), ipsilateral regeneration of ablated rVRG fibers (B), and contralateral rVRG fiber sprouting (C). Blue lines represent descending rVRG axons and red dotted lines represent regrowing axons. D, Experimental timeline: C2 hemisection was performed on all animals, and from days 3 to 24 after injury, animals were treated with twice-daily subcutaneous injections of either DMSO or PAP4. Two weeks following injury, animals received either ipsilateral or contralateral brainstem injections of AAV2-mCherry. Four days before killing, CMAP recordings were taken for each animal and CTB injections into the ipsilateral hemidiaphragm were performed. At eight weeks after injury, EMG recordings were taken, and then animals were killed to collect brain and spinal cord tissue samples.
- Figure 2.
Increased mTOR signaling in rVRG neurons. At eight weeks after hemisection, cresyl violet staining of both transverse (A) and sagittal (B) sections shows complete dorsal-ventral and medial-lateral ablation of the spinal cord. No difference was observed between DMSO and PAP4 groups in rostral-to-caudal lesion size (C); scale bars: 1 mm (A), 1 mm (B). Fluorescent microscopy of coronal brainstem sections shows AAV2-mCherry injections into the rVRG (D); higher magnification images reveal robust expression of mCherry and neuronal morphology of transduced cells in rVRG (E, F); scale bars: 1 mm (D), 500 μm (E), and 100 μm (F). mCherry+ coronal brainstem sections immunostained with NeuN and pS6K show co-localization of pS6K in NeuN+ neurons within the rVRG (G–J); scale bar: 100 μm. At eight weeks after injury, immunofluorescence imaging shows that PAP4-treated animals had significantly higher levels of pS6K in ipsilateral rVRG neurons compared to DMSO-treated controls (K, L); scale bar: 100 μm; total fluorescence per mCherry+ cell was averaged for each animal and the mean of each group is represented in M. Immunofluorescence imaging revealed no change in GFAP+ glial scar size at eight weeks after hemisection between DMSO (N) and PAP4 (O) groups; scale bar: 200 μm. Yellow dotted lines represent the rostral and caudal borders of the glial scar at the lesion site. Quantification of glial scar width shown in P; n = 10 sagittal sections per animal and n = 3 animals per group for quantification. *p < 0.05.
- Figure 3.
Diaphragm function was significantly restored by PAP4 following C2 hemisection. At eight weeks after injury, EMG recordings were assessed from uninjured, DMSO-treated hemisection, and PAP4-treated hemisection groups. EMG amplitudes were recorded at three subregions of the hemidiaphragm: dorsal, medial, and ventral regions (D). Both DMSO-treated and PAP4-treated groups were compared to intact (laminectomy-only) animals. Representative EMG traces from the ventral region of the diaphragm show that compared to laminectomy-only animals (A, A’), DMSO-treated rats (B, B’) had a significant decrease in inspiratory EMG burst amplitude at eight weeks after injury. Animals treated with PAP4 (C, C’) showed significant recovery of EMG amplitude following injury. Quantification of EMG recordings at the three hemidiaphragm regions (E). Ventral subregion EMG burst frequency was unaffected by PAP4 treatment (F). EMG duty cycle was increased by PAP4 (G). *p < 0.05. **p < 0.01. ***p < 0.001. ns, not significant p ≥ 0.05.
- Figure 4.
Functional and morphologic innervation of the diaphragm was unaffected by PAP4. CMAP recordings obtained from the ipsilateral hemidiaphragm show no differences in amplitude among laminectomy-only (A), DMSO (B), and PAP4 (C) groups. Similarly, no differences were noted in CMAP amplitudes obtained in the contralateral hemidiaphragm among laminectomy-only (A’), DMSO (B’), and PAP4 (C’) groups. Quantification of CMAP amplitudes showed no differences among groups or between ipsilateral and contralateral hemidiaphragm (D). Representative confocal z-stack image showing labeling of the diaphragm NMJ for the pan-axonal marker SMI-312 (green; F, H), the synaptic vesicle marker SV2 (green; F, H), and the marker of postsynaptic nicotinic acetylcholine receptors α-bungarotoxin (red; G, H). Nearly all NMJs in the ipsilateral hemidiaphragm remained fully innervated at eight weeks after hemisection; scale bar 50 μm. Quantification of the dorsal, medial, and ventral portions of the hemidiaphragm showed no differences in the percentage of fully-innervated, partially-denervated, or completely-denervated NMJs between the DMSO and PAP4 groups (E); n = 3 and n = 4 rats for DMSO and PAP4, respectively; 16 tissue sections per animal analyzed.
- Figure 5.
Ipsilateral rVRG axons regenerated through the lesion and formed putative synaptic connections with PhMNs. At eight weeks after injury, sagittal sections of the cervical spinal cord were examined for growth of mCherry-labeled axons originating in the ipsilateral rVRG at multiple locations relative to the lesion (A). Dotted boxes denote locations of images in subsequent panels, red lines represent rVRG axons, and green triangles represent PhMNs. Representative images of the lesion site reveal mCherry-labeled rVRG fibers were able to regenerate into and through the lesion and then back into the intact caudal spinal cord in PAP4-treated rats (D, F, H, J), but not the DMSO-only controls (C, E, G, I); scale bars: 100 μm (C, D), 40 μm (E–J). Quantification of mCherry-labeled rVRG axon growth in 100 μm rostral-to-caudal bins using the rostral end of the lesion as the starting point (B). Confocal imaging show that regenerating mCherry+ rVRG axons (L, N, P, R; arrowheads in L) formed putative VGLUT2+ excitatory synaptic connections (M, N, Q, R) with CTB+ PhMNs retrogradely-labeled from the ipsilateral hemidiaphragm (K, N, O, R); scale bars: 50 μm (K–N), 30 μm (O–R). Orthogonal projection shows mCherry+/VGLUT2+ excitatory rVRG axon terminals located directly presynaptic to the soma of CTB+ PhMNs in the C3 spinal cord (R; arrowheads in O–R). Dotted box in panel N denotes location of panels O–R. No mCherry-labeled rVRG fibers were found at all in the ipsilateral PhMN pool of DMSO-treated controls (data not shown). *p < 0.05.
- Figure 6.
Sprouting of spared fibers originating in the contralateral rVRG. At eight weeks after injury, sagittal sections of the cervical spinal cord were examined for growth of mCherry-labeled axons originating in the contralateral rVRG at level C3, C4, and C5 (A). Representative images show no difference between DMSO control (A) and PAP4 (B) in the density of mCherry-labeled rVRG fibers in the C3 ventral horn ipsilateral to the lesion; yellow arrowheads in A, B denote mCherry+ rVRG axons; scale bar: 100 μm. Quantification of the number of contralateral-originating rVRG fibers per area present within the ipsilateral ventral horn at levels C3–C5 (C). Distances from the lesion correspond to spinal cord levels: +1.5 mm to C3 (A, B), +3.0 mm to C4, +4.5 mm to C5; n = 4 rats per group; n = 12 tissue sections analyzed per animal. PAP4 treatment increased the number of mCherry-labeled axons originating in the contralateral rVRG within the contralateral C3 ventral horn (D); n = 3 rats per group; nine tissue sections analyzed per animal. *p < 0.05.
- Figure 7.
PAP4 promoted 5-HT axon growth and synaptic innervation of PhMNs caudal to the lesion. Compared to DMSO-only (A–C), PAP4 (D–F) enhanced the density of serotonergic axons directly surrounding CTB-labeled PhMNs at levels C3–C5; scale bar: 100 μm. Quantification demonstrates that PAP4 treatment increased the number of 5-HT+ axonal profiles (G), total length of 5-HT+ axonal profiles (H), and the integrated intensity of 5-HT immunostaining (I) compared to DMSO. PAP4 also increased the number of 5-HT+ axon profiles surrounding CTB-labeled PhMNs in the contralateral spinal cord (J). We also assessed the number of putative synaptophysin+ (L) synaptic connections between 5-HT axons (M) and CTB-labeled PhMNs (K) at levels C3–C5 using single-Z section analysis to establish direct apposition of presynaptic 5-HT+/synaptophysin+ axon terminals and postsynaptic CTB+ PhMNs (N). Compared to DMSO-only, PAP4 animals significantly increased the number of putative synaptic contacts between 5-HT axons and CTB-labeled PhMNs in the C3–C5 spinal cord ipsilateral to the hemisection (O). *p < 0.05.
- Figure 8.
Effects of re-lesion on PAP4-induced restoration of diaphragm function. At eight weeks after hemisection, we performed a re-lesion through the C2 injury site of both DMSO-only and PAP4 groups. Representative traces of EMG recordings from the ventral hemidiaphragm show no change in EMG amplitude recordings between DMSO pre-re-lesion and DMSO post-re-lesion groups (A, B). There was a significant loss of inspiratory burst amplitude following the re-lesion (D) in the PAP-treated group compared to the recordings immediately before re-lesion in the same animal (C). Quantification of inspiratory EMG bursts in the dorsal, medial and ventral subregions of the ipsilateral hemidiaphragm before and after re-lesion (E). *p < 0.05. **p < 0.01. ***p < 0.001. ns, not significant p ≥ 0.05.
In this issue
