PTEN inhibition promotes robust growth of bulbospinal respiratory axons and partial recovery of diaphragm function in a chronic model of cervical contusion spinal cord injury

Abstract

High spinal cord injury (SCI) leads to persistent and debilitating compromise in respiratory function. Cervical SCI not only causes the death of phrenic motor neurons (PhMNs) that innervate the diaphragm, but also damages descending respiratory pathways originating in the rostral ventral respiratory group (rVRG) located in the brainstem, resulting in denervation and consequent silencing of spared PhMNs located caudal to injury. It is imperative to determine whether interventions targeting rVRG axon growth and respiratory neural circuit reconnection are efficacious in chronic cervical contusion SCI, given that the vast majority of individuals are chronically-injured and most cases of SCI involve contusion-type damage to the cervical region. We therefore employed a rat model of chronic cervical hemicontusion to test therapeutic manipulations aimed at reconstructing damaged rVRG-PhMN-diaphragm circuitry to achieve recovery of respiratory function. At a chronic time point post-injury, we systemically administered: an antagonist peptide directed against phosphatase and tensin homolog (PTEN), a central inhibitor of neuron-intrinsic axon growth potential; an antagonist peptide directed against receptor-type protein tyrosine phosphatase sigma (PTPσ), another important negative regulator of axon growth capacity; or a combination of these two peptides. PTEN antagonist peptide (PAP4) promoted partial recovery of diaphragm motor activity out to nine months post-injury (though this effect depended on the anesthetic regimen used during recording), while PTPσ peptide did not impact diaphragm function after cervical SCI. Furthermore, PAP4 promoted robust growth of descending bulbospinal rVRG axons caudal to the injury within the denervated portion of the PhMN pool, while PTPσ peptide did not affect rVRG axon growth at this location that is critical to control of diaphragmatic respiratory function. In conclusion, we find that, when PTEN inhibition is targeted at a chronic time point following cervical contusion, our non-invasive PAP4 strategy can successfully promote significant regrowth of damaged respiratory neural circuitry and also partial recovery of diaphragm motor function.

Keywords: Axon; Breathing; Cervical; Chronic; Circuit; Contusion; Diaphragm; PTEN; PTPsigma; Regeneration; Respiratory; SCI; Spinal cord injury; Sprouting.

Figure 1:. Histological analysis of cervical hemicontusion SCI.

Representative images of the lesion epicenter for DMSO-only, PAP4, PTPσ peptide, and PAP4 + PTPσ peptide treated groups. 40 μm spinal cords sections were cresyl violet-stained, and images at the lesion epicenter were obtained using an All-in-One Fluorescence Microscope BZ-X810 (objective 4X).

Figure 2:. Experimental timeline.

Rats received C4 hemicontusion on day-0. Starting at 8 weeks post-contusion SCI, rats were administered DMSO-only, PAP4, PTP-sigma peptide, or both peptides subcutaneously twice-daily for 21 days. At 33 weeks post-injury, rats underwent intra-medullary injection of AAV2-mCherry into ipsilesional rVRG. CMAP recordings were conducted at week 36 post-injury, and CTB tracer was injected intrapleurally at this same time point. At 34 weeks after SCI, diaphragm EMG was performed, followed by transcardial perfusion and tissue harvesting.

Figure 3:. PAP4 promoted partial recovery of diaphragm EMG amplitude.

(A) Representative traces of diaphragm electromyography (EMGdia) recordings ipsilateral to injury from dorsal diaphragm subregion for DMSO-only, PAP4, PTPσ peptide, and PAP4 + PTPσ peptide treated groups. Quantification of EMGdia amplitudes for ventral, medial and dorsal subregions of diaphragm ipsilateral to injury under ketamine/xylazine anesthesia (B) or isoflurane anesthesia (C). * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, One Way ANOVA (Fisher LSD Method) was used for all graphs.

Figure 4:. PAP4 and PTPσ peptide did not impact functional diaphragm innervation.

(A) Representative diaphragm compound motor action potentials (CMAPs) recorded ipsilateral to injury for DMSO-only, PAP4, PTPσ peptide, and PAP4 + PTPσ peptide treated groups. Black arrow denotes a representative CMAP response. Black bar denotes the 0.5 ms stimulation of the phrenic nerve. (B) Quantification of CMAP amplitudes for the different groups. p > 0.05 for all comparisons, Kruskal-Wallis One Way Analysis of Variance on Ranks (Dunn’s Method), H = 0.336 with 3 degrees of freedom (p = 0.953).

Figure 5:. PAP4 promoted robust growth of rVRG axons in the phrenic nucleus.

(A) AAV2-mCherry anterograde tracer was injected into ipsilesional rostral ventral respiratory group (rVRG). Labeled mCherry+ rVRG axons are observed around CTB-labeled phrenic motoneurons located in C3-C5 spinal cord (B), as well as in the lesion (C). Representative images are displayed for lesion epicenter (D-G) and ventral horn caudal to injury (H-L). Scale bar: 50μm (D-L). Quantification of mCherry+ axon profiles (M) and total length (N) ipsilaterally for the area caudal to injury. * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, One Way ANOVA (Fisher LSD Method).

Figure 6:. Morphological analysis of diaphragm neuromuscular junction (NMJ) innervation.

Representative images illustrating examples of intact (I) and partially denervated (PD) NMJs in ipsilesional hemidiaphragm (A and B), with enlargement images showing intact (C), partially denervated (D) and completely denervated (E) NMJs. Scale bar: 30μm (A-B), 15μm (C-E). Quantification of intact NMJs (F, I, L), partially denervated NMJs (G, J, M), and completely denervated NMJs (H, K, N) in ventral, medial and dorsal subregions. * p < 0.05, *** p < 0.001, One Way ANOVA (Fisher LSD Method) for percentage of intact NMJs in the medial subregion (G); ^# p < 0.05, Kruskal-Wallis One Way Analysis of Variance on Ranks (Dunn’s Method). Addiitonal information regarding statistical tests is presented in Table 2.

Figure 7:. PTPσ peptide induced formation of atypical denervated neuromuscular junctions (NMJs).

Representative atypical NMJ morphologies in ipsilesional hemidiaphragm of animals treated with PTPσ peptide (A and B). Scale bar: 15μm (A), 30μm (B). C. Yellow arrows denote thick SV2+/SMI-312+ pre-terminal phrenic motor axons (green) overlaying the sites of postsynaptic α-bungarotoxin+ nicotinic acetylcholine receptors but failing to make a synaptic connection (red). Quantification of these atypical NMJs, expressed as percentage of total NMJs across the entire hemidiaphragm. * p < 0.05, Kruskal-Wallis One Way Analysis of Variance on Ranks (Dunn’s Method), H = 18.901 with 3 degrees of freedom (p ≤ 0.001).