Respiratory axon regeneration in the chronically injured spinal cord

Abstract

Promoting the combination of robust regeneration of damaged axons and synaptic reconnection of these growing axon populations with appropriate neuronal targets represents a major therapeutic goal following spinal cord injury (SCI). A key impediment to achieving this important aim includes an intrinsic inability of neurons to extend axons in adult CNS, particularly in the context of the chronically-injured spinal cord. We tested whether an inhibitory peptide directed against phosphatase and tensin homolog (PTEN: a central inhibitor of neuron-intrinsic axon growth potential) could restore inspiratory diaphragm function by reconnecting critical respiratory neural circuitry in a rat model of chronic cervical level 2 (C2) hemisection SCI. We found that systemic delivery of PTEN antagonist peptide 4 (PAP4) starting at 8 weeks after C2 hemisection promoted substantial, long-distance regeneration of injured bulbospinal rostral Ventral Respiratory Group (rVRG) axons into and through the lesion and back toward phrenic motor neurons (PhMNs) located in intact caudal C3-C5 spinal cord. Despite this robust rVRG axon regeneration, PAP4 stimulated only minimal recovery of diaphragm function. Furthermore, re-lesion through the hemisection site completely removed PAP4-induced functional improvement, demonstrating that axon regeneration through the lesion was responsible for this partial functional recovery. Interestingly, there was minimal formation of putative excitatory monosynaptic connections between regrowing rVRG axons and PhMN targets, suggesting that (1) limited rVRG-PhMN synaptic reconnectivity was responsible at least in part for the lack of a significant functional effect, (2) chronically-injured spinal cord presents an obstacle to achieving synaptogenesis between regenerating axons and post-synaptic targets, and (3) addressing this challenge is a potentially-powerful strategy to enhance therapeutic efficacy in the chronic SCI setting. In conclusion, our study demonstrates a non-invasive and transient pharmacological approach in chronic SCI to repair the critically-important neural circuitry controlling diaphragmatic respiratory function, but also sheds light on obstacles to circuit plasticity presented by the chronically-injured spinal cord.

Keywords: Breathing; Cervical; Chronic; Diaphragm; PTEN; Regeneration; Regrowth; Respiratory; SCI; Spinal cord injury.

Figure 1.. PAP4 promoted partial recovery of diaphragm motor function in chronic cervical SCI.

(A) Timeline of experimental design. (B-C) Representative traces showing impact of DMSO-only or PAP4 on raw EMG activity (top row) and integrated EMG activity (bottom row). At 24 weeks post-injury, EMG recordings were assessed at 3 sub-regions of hemi-diaphragm (dorsal, medial and ventral sub-regions) from C2 hemisection rats that received either DMSO-only (B) or PAP4 (C). Quantification of integrated EMG amplitude at ventral (D), medial (E), and dorsal (F) sub-regions. Quantification of inspiratory burst frequency (G) and burst duration (H) in the ventral subregion. n = 12 animals for DMSO-only; n= 9 for PAP4; n = 4 for uninjured control. Data represent mean ± SEM, *P < 0.05 by Kruskal–Wallis ANOVA.

Figure 2.. PAP4 did not affect functional innervation of the diaphragm.

Representative CMAP recordings obtained from ipsilateral or contralateral hemi-diaphragm of DMSO-only (A-B) and PAP4 (C-D) groups. (E) Quantification of CMAP amplitudes showed no differences between DMSO-only and PAP4 groups or between ipsilateral and contralateral hemi-diaphragm. n = 6 animals per group, one-way ANOVA.

Figure 3.. PAP4 did not affect morphological innervation of the diaphragm.

Whole-mount immunohistochemistry of diaphragm was used to label post-synaptic nicotinic acetylcholine receptors with AlexaFluor-conjugated alpha-bungarotoxin (red) and pre-synaptic phrenic motor axons all the way to their terminals with an antibody against neurofilament (SMI-312, green) and a second antibody against synaptic vesicle protein 2 (SV2, also green). As shown in representative confocal z-stacks, nearly all NMJs in ipsilateral hemi-diaphragm were intact in C2 hemisection animals receiving either DMSO-only (A-F) or PAP4 (G-L). As shown in higher magnification images, intact NMJs were characterized by complete apposition of pre- and post-synaptic labeling. There were no differences between rats treated with DMSO-only and PAP4 in percentages of intact (M), completely-denervated (N) partially-denervated NMJs (O), or multiply-innervated (P) NMJs, or percentage of NMJs with thin pre-terminal axons (Q). n = 5 rats for DMSO-only, n = 5 for PAP4. P > 0.05 by Kruskal–Wallis ANOVA.

Figure 4.. PAP4 induced long-distance regeneration of bulbospinal rVRG axons.

(A) Sagittal section of cervical spinal cord showing: the columnar organization of CTB-labeled PhMNs extending from C3 to C5; the locations (i.e. rostral-caudal distances relative to hemisection site) of rVRG axon growth analysis; and location of representative images in panels B-G. Scale bar in A: 250 μm. Representative images of sagittal sections from DMSO-only and PAP4 treated rats injected with AAV2-mCherry into ipsilateral rVRG: (B, E) rostral to lesion site, (C, F) lesion site, (D, G) caudal to lesion site. Yellow arrows in panels B-G denote mCherry-labeled rVRG axons. Dotted yellow lines in panels C and F denote rostral and caudal lesion-intact borders. Scale bar: 100 μm. (H) To assess rVRG axon regeneration, we selectively transduced neurons within rVRG ipsilateral to hemisection with AAV2-mCherry anterograde tracer. We also intrapleurally injected Cholera Toxin B Subunit (CTB) to selectively label PhMN cell bodies within the cervical spinal cord. (I) Quantification shows number of ipsilateral-originating mCherry-labeled rVRG axons at different locations in DMSO-only and PAP4 group using the rostral end of the lesion as the rostral-caudal starting point. Indicated differences were comparisons between DMSO-only and PAP4 groups at each distance (n = 7-8 animals per group, *P < 0.05, Kruskal-Wallis ANOVA). (J-L) Pearson Correlation was used to examine correlation between degree of rVRG axon regeneration at C3 (J), C4 (K) or C5 (L) levels and EMG amplitude in corresponding ventral, medial and dorsal diaphragm subregions, respectively (correlation coefficient for linear fit, ventral: R² = 0.95, P < 0.001; medial: R² = 0.11, P = 0.46; dorsal: R² = 0.12, P = 0.44, n = 8 animals for DMSO-only, n = 7 for PAP4).

Figure 5.. Regenerating rVRG axons formed few putative excitatory synapses with PhMNs.

We assessed number of putative synaptic connections specifically between mCherry/DsRed double-positive rVRG axons and CTB-labeled PhMN cell bodies at level C3 with confocal acquisition of z-stacks and quantification of rVRG axon-PhMN contacts using single-Z section analysis to establish direct apposition of pre-synaptic VGLUT2⁺/mCherry⁺ axon terminals and post-synaptic CTB⁺ PhMNs. (A-D) We observed no putative rVRG-PhMN excitatory synapses in DMSO-only control. (E-H) On the contrary, we observed putative rVRG-PhMN excitatory synapses with PAP4. (H) Orthogonal projection shows mCherry⁺/VGLUT2⁺ excitatory rVRG axon terminals located directly presynaptic to the soma of a CTB⁺ PhMN (example putative connections denoted by arrowheads). Scale bar: 100 μm. (I) Pearson Correlation showed a strong positive correlation between EMG amplitude in ipsilateral ventral hemi-diaphragm and numbers of these rVRG-PhMN excitatory synaptic connections per PhMN in C3 portion of the PhMN pool (correlation coefficient for linear fit, R² = 0.98, P < 0.001, n = 8 animals for DMSO-only, n = 7 for PAP4). (J) Number of putative rVRG-PhMN excitatory synapses per PhMN in PAP4-treated SCI animals (n = 7 rats) was only a small fraction of normal density of these synapses in uninjured rats (n = 4 rats, P = 0.03 versus PAP4, Kruskal–Wallis ANOVA) at the same C3 location (DMSO-only, n = 8 rats, P = 0.61 for DMSO-only versus PAP4, Kruskal–Wallis ANOVA). (K) Quantification of putative excitatory synapses from mCherry⁻/VGLUT2⁺ non-rVRG inputs onto ipsilateral CTB⁺ PhMNs at C3 (P = 0.56 between DMSO-only and PAP4 groups, one-way ANOVA; P > 0.05 uninjured group versus DMSO-only or PAP4, one-way ANOVA, n = 4 animals for uninjured group, n = 8 for DMSO-only, n = 7 for PAP4).

Figure 6.. PAP4 did not promote sprouting of spared rVRG axons.

We injected AAV2-mCherry tracer into contralateral rVRG (B) and then quantified number of mCherry⁺ rVRG axon profiles directly surrounding CTB⁺ PhMNs on the side of hemisection (A, C). In all injured rats in both groups, we observed significant innervation of the phrenic nucleus on the injured side (D-I). Compared to DMSO-only control (D-F), PAP4 (G-I) did not enhance sprouting of spared rVRG fibers at C3 (J), C4 (K) or C5 (L). Scale bars: 250 μm (A); 100 μm (C); 20 μm (D-I), n = 3 rats for uninjured, n = 5 for DMSO-only, n = 5 for PAP4, P = 0.40 for C3, P = 0.37 for C4, P = 0.46 for C5; comparisons of DMSO-only versus PAP4. DMSO-only and PAP4 injury groups were not different than the uninjured control group in mCherry⁺ axon profile measurements at all sub-regions (J-L; P = 0.54-0.84 for the various comparisons, Kruskal–Wallis ANOVA).

Figure 7.. PAP4 did not promote sprouting of modulatory serotonergic axon input to the PhMN pool.

We assessed density of serotonergic axon innervation of the PhMN pool by performing immunohistochemistry for serotonin (5-HT) and quantifying total length of 5-HT axons directly surrounding CTB-labeled PhMNs on the injury side. Sagittal sections of cervical spinal cord show that, compared to DMSO-only control (A), PAP4 (B) did not impact 5-HT axon sprouting within the phrenic nucleus at C3 (C), C4 (D) or C5 (E) at 24 weeks post-injury (n = 8 rats for DMSO-only, n = 9 for PAP4, P = 0.86 for C3, P = 0.56 for C4, P = 0.15 for C5, comparisons of DMSO-only versus PAP4; one-way ANOVA). DMSO-only and PAP4 injury groups were not different than the uninjured control group (n = 3) in total 5-HT axon length at all subregions (C-E; P = 0.18-0.74 for the various comparisons, one-way ANOVA). Scale bar: 100 μm. Arrowheads indicate 5-HT⁺ axons.

Figure 8.. Relesion ablated PAP4-induced recovery of diaphragm function.

At 24 weeks after C2 hemisection, a surgical relesion was performed through the injury site of both DMSO-only and PAP4 treated groups. Representative EMG recordings from ventral hemi-diaphragm show a significant loss of inspiratory burst amplitude following relesion in PAP4-treated animals (C-D), but not in DMSO-only rats (A-B). Quantification of inspiratory EMG burst amplitude in ventral (E), medial (F) and dorsal (G) subregions of ipsilateral hemi-diaphragm before and after relesion in DMSO-only and PAP4 treated cohorts, n = 4 rats per condition. Asterisks indicate P < 0.05 by Kruskal–Wallis ANOVA.