Genomic and Epigenomic Evaluation of Electrically Induced Exercise in People With Spinal Cord Injury: Application to Precision Rehabilitation

Abstract

Objective: Physical therapists develop patient-centered exercise prescriptions to help overcome the physical, emotional, psychosocial, and environmental stressors that undermine a person's health. Optimally prescribing muscle activity for people with disability, such as a spinal cord injury, is challenging because of their loss of volitional movement control and the deterioration of their underlying skeletal systems. This report summarizes spinal cord injury-specific factors that should be considered in patient-centered, precision prescription of muscle activity for people with spinal cord injury. This report also presents a muscle genomic and epigenomic analysis to examine the regulation of the proliferator-activated receptor γ coactivator 1α (PGC-1α) (oxidative) and myostatin (hypertrophy) signaling pathways in skeletal muscle during low-frequency (lower-force) electrically induced exercise versus higher-frequency (higher-force) electrically induced exercise under constant muscle recruitment (intensity).

Methods: Seventeen people with spinal cord injury participated in 1 or more unilateral electrically induced exercise sessions using a lower-force (1-, 3-, or 5-Hz) or higher-force (20-Hz) protocol. Three hours after the exercise session, percutaneous muscle biopsies were performed on exercised and nonexercised muscles for genomic and epigenomic analysis.

Results: We found that low-frequency (low-force) electrically induced exercise significantly increased the expression of PGC-1α and decreased the expression of myostatin, consistent with the expression changes observed with high-frequency (higher-force) electrically induced exercise. Further, we found that low-frequency (lower-force) electrically induced exercise significantly demethylated, or epigenetically promoted, the PGC-1α signaling pathway. A global epigenetic analysis showed that >70 pathways were regulated with low-frequency (lower-force) electrically induced exercise.

Conclusion: These novel results support the notion that low-frequency (low-force) electrically induced exercise may offer a more precise rehabilitation strategy for people with chronic paralysis and severe osteoporosis. Future clinical trials are warranted to explore whether low-frequency (lower-force) electrically induced exercise training affects the overall health of people with chronic spinal cord injury.

Keywords: Disability; Hypertrophy; Molecular; Muscle; Plasticity; Rehabilitation.

© The Author(s) 2021. Published by Oxford University Press on behalf of the American Physical Therapy Association. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Figures

Figure 1

Illustration of bone quality after spinal cord injury. Following a spinal cord injury, bone within paralyzed extremities, particularly the femur and tibia, loses bone mineral density, with a concurrent change in bone microarchitecture. The estimated range of bone density (gray shading) rapidly decreases during the first year after injury. Bone mineral density begins to level off in subsequent years, leaving underlying bone in the paralyzed extremities extremely osteoporotic and at high risk of injury. Figure was adapted from data presented by Dudley-Javoroski and Shields.

Figure 2

Force-frequency relationship. (A) Representative example of force, as measured at the ankle, elicited during a bout of exercise using supramaximal electrical stimulation at various stimulation frequencies. As the stimulation frequency increases, the time between stimulus pulses decreases. As stimulus pulses become closer together, the stimulated muscle does not have sufficient time to fully recover from the previous stimulus pulse, resulting in a temporal summation of force and a fused muscle contraction. (B) As stimulation frequency increases, the elicited force increases. Higher-force contractions require bone to be of sufficient quality (upper photograph) to ensure that injuries do not occur. However, using low stimulation frequencies results in low-force contractions to lower the risk of injury in osteoporotic bone (lower photograph). Figure was created from unpublished force profiles collected using the device setup presented by Shields and Chang and Petrie et al and consistent with force profiles published in those studies.,

Figure 3

Acute gene expression. (A) Mean proliferator-activated receptor γ coactivator 1α (PGC-1α) fold change 3 hours after exercise for the unfused contractions (1, 3, and 5 Hz) and fused contractions (20 Hz). There was a significant effect of exercise relative to the nonexercised control limb (P < .001), but no significant difference was detected between the unfused (1-, 3-, and 5-Hz) and fused (20-Hz) exercises for PGC-1α expression. (B) Mean myostatin (MSTN) fold change 3 hours after exercise for the unfused contractions (1, 3, and 5 Hz) and fused contractions (20 Hz). There was a significant effect of exercise relative to the nonexercised control limb (P < .03), but no significant difference was detected between the unfused (1-, 3-, and 5-Hz) and fused (20-Hz) exercises for MSTN expression. † = significant change for 1, 3, 5, and 20 Hz relative to control condition.

Figure 4

Acute epigenetics. (A) Mean proliferator-activated receptor γ coactivator 1α (PGC-1α) methylation beta value (% methylated) 3 hours after an unfused (3-Hz) exercise in a subset of 4 participants. There was a significant demethylation effect of the unfused (3-Hz) exercise relative to the nonexercised control limb (P = .006). (B) Mean myostatin (MSTN) methylation β value (% methylated) 3 hours after an unfused (3-Hz) exercise in a subset of 4 participants. There was no significant methylation effect of the unfused (3-Hz) exercise relative to the nonexercised control limb (P = .20). However, noteworthy was the still relative observed increase in the methylation of MSTN after exercise in this small sample subset. † = significance.

Figure 5

Epigenetic connectivity network. (A) Base connectivity map for the 1925 pathways used for the gene set enrichment analysis. Each node represents a pathway defined within the Reactome knowledge base, which defines various biological processes that are categorized into 26 broad categories, indicated by node color. (B) A total of 74 pathways were found to be demethylated 3 hours after completion of a dose of low-force (3-Hz) exercise relative to the baseline (nonexercised) limb in people with a spinal cord injury. (C) The 74 demethylated pathways represent 19 of the 26 broad categories. Of note is that most of the pathways were associated with the signal transduction, gene expression and transcription, immune system, and metabolism categories.