Effects of chronic cuff electrode implantation and stimulation of the rat tibial nerve: ultrastructural analysis of myelinated axons and their microenvironment

Numerous experimental data point to therapeutic effects of electrical stimulation of peripheral nerves. Stimulation of low‑threshold proprioceptive afferents to motoneurons (MNs) innervating the ankle extensor muscles of the rat increases glutamatergic Ia MN inputs and spinal and muscle expression of neurotrophin‑3, instrumental for proprioceptive afferent‑MN connections. We aimed to examine morphological consequences of the cuff electrode implanted around the tibial nerve used for chronic stimulation which may cause unwanted effects through its long‑lasting contact with the nerve. We addressed also the questions if stimulation of low‑threshold proprioceptive afferents may contribute to neural damage, affect capillary vessels and macrophage morphology reflecting functional state of the nerve. Their consequences on axonal and nerve ultrastructure have not been previously evaluated. To examine the effect of cuff electrodes and of 7‑day low‑threshold stimulation applied in the 4th week post‑implantation on the tibial nerve, electrodes were implanted bilaterally, and a stimulation was added unilaterally. Non‑implanted group was the control. The counts of myelinated axons, their cross‑sectional area (>1 μm2) and circularity were measured on photomicrographs captured with use of light microscope from semi‑thin Epon sections stained with toluidine blue. Ultrastructural observations with TEM included myelin, axoplasm, Remak fibers, blood vessels, macrophages and fibroblasts. Four weeks after implantation of the cuff, the mean cross‑sectional area of the nerve, and counts, area and diameter of the myelinated fibers were lower compared to control. The largest fibers located in the external parts of the main nerve branches and dispersed in small branches were changed the most. No significant difference in either parameter between unstimulated (NS) and stimulated (S) nerve was found. Axons in Remak bundles were dispersed comparably in NS and S nerves. Thinly myelinated fibers demonstrating features of demyelination with remyelination were identified. We provide evidence that long-lasting presence of the cuff electrode is not neutral for nerve structure, causing mild shrinkage of large myelinated axons and impairment of organization of unmyelinated fibers. Since we demonstrated that low‑threshold proprioceptive stimulation of the tibial nerve results in a stable Hoffmann reflex and induces synaptic plasticity by increasing glutamatergic input to MNs, we conclude that the applied protocol maintains an adequate margin of safety.