Chronic, closed‐loop, cervical epidural stimulation elicits plasticity in diaphragm motor output and upregulates spinal neurotrophic factor gene expression

Abstract

Electrical spinal stimulation after spinal cord injury (SCI) improves functional outcomes. However, there's an unmet need to understand if epidural stimulation has a lasting impact on motor systems in people with chronic SCI. We developed a closed‐loop stimulation protocol (endogenous respiratory output from diaphragm EMG‐triggered spinal stimulation with bilateral stim electrodes at C4) in freely behaving rats. We hypothesized that long‐term (20 h/day, 4 d), closed‐loop epidural stimulation elicits plasticity in diaphragm EMG. Since neurotrophic factor expression is upregulated after spinal stimulation in other systems, and is associated with mechanisms of spinal respiratory motor plasticity, we hypothesized that stimulation upregulates neurotrophic factor gene expression in the cervical spinal cord. Four rat groups were studied: 1) C2 hemisection (C2HS) + stim (n = 3); 2) C2HS + no stim (n = 3); 3) no injury + stim (n = 2); and 4) sham controls (n = 2). Increased current during motor evoked potential testing increased area under the curve (AUC) of the stimulus‐triggered average in all rats (range: 50–750mA or max tolerated; ~0.5Hz inspiratory‐triggered). However, after 4 d of stim, a dramatic AUC potentiation was observed in 2 of 3 C2HS rats (667% and 239% increase from 1d – 4d, 520mA vs. 6% in the non‐responder) in ipsilesional diaphragm. Contralesional AUC was also potentiated (565% and 199% increase from 1d‐4d, 520mA vs. −6% in the non‐responder). Ipsilesional peak‐to‐peak amplitude increased after stimulation (2.9mV at 4d vs. 0.5mV at 1d) with additional peaks appearing in longer time domains, suggesting potential for interneuronal recruitment. qPCR neurotrophic factor and receptor mRNA (VEGFA & VEGFR2, BDNF & TrkB, NT3 & TrkC) revealed that stim after injury upregulated VEGFA (1.06 ± 0.77 fold vs. 0.948 in injury only) and BDNF (1.53 ± 0.35 fold vs. 1.00 in injury only) in C3–C5 ventral spinal cord. Dorsal spinal VEGFA, BDNF and TrkB were downregulated after injury in C1–C2 (0.42 fold vs. 1.00 in sham; 0.43 fold vs. 1.01 in sham; and 0.56 fold vs 1.00 in sham, respectively). However, stim post‐injury partially restored expression of VEGFA (0.60 ± 0.02 fold), BDNF (0.64 ± 0.02 fold) and TrkB (0.74 ± 0.06 fold). This was surprising as the stimulation site is several segments below the injury. Finally, in dorsal C1–C3, C2HS downregulated both NT3 and TrkC after injury (NT3: 0.43 fold vs.1.00 in sham; TrkC: 0.43 fold vs. 1.01 in sham) with a partial recovery of NT3 (0.64 fold ± 0.02) and TrkC (0.64 fold ± 0.02) post‐stim. In C3–C5 ventral horn, stim upregulated TrkC mRNA (1.01 ± 0.05 vs. 0.95), suggesting a possible substrate for axonal growth. Collectively these preliminary results demonstrate the first known evidence of spinal respiratory plasticity associated with long‐term spinal stimulation in an awake, freely‐behaving animal. Results from gene expression assays lay the groundwork for essential mechanistic studies of epidural stimulation.Support or Funding InformationCraig Nielsen FoundationThis abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.