Trans-spinal direct current stimulation modifies spinal cord excitability through synaptic and axonal mechanisms

Abstract

The spinal cord is extremely complex. Therefore, trans-spinal direct current stimulation (tsDCS) is expected to produce a multitude of neurophysiological changes. Here, we asked how tsDCS differentially affects synaptic and nonsynaptic transmission. We investigated the effects of tsDCS on synaptically mediated responses by stimulating the medullary longitudinal fascicle and recording responses in the sciatic nerve and triceps and tibialis anterior muscles. Response amplitude was increased during cathodal-tsDCS (c-tsDCS), but reduced during anodal-tsDCS (a-tsDCS). After-effects were dependent on the frequency of the test stimulation. c-tsDCS-reduced responses evoked by low-frequency (0.5 Hz) test stimulation and increased responses evoked by high-frequency (400 Hz) test stimulation. a-tsDCS had opposite effects. During and after c-tsDCS, excitability of the lateral funiculus tract (LFT) and dorsal root fibers was increased. However, a-tsDCS caused a complex response, reducing the excitability of LFT and increasing dorsal root fiber responses. Local DC application on the sciatic nerve showed that the effects of DC on axonal excitability were dependent on polarity, duration of stimulation, temporal profile (during vs. after stimulation), orientation of the current direction relative to the axon and relative to the direction of action potential propagation, distance from the DC electrode, and the local environment of the nervous tissue. Collectively, these results indicate that synaptic as well as axonal mechanisms might play a role in tsDCS-induced effects. Therefore, this study identified many factors that should be considered in interpreting results of DCS and in designing tsDCS-based interventions.

Keywords: Direct current; excitability; trans‐spinal.

© 2014 The Authors. Physiological Reports published by Wiley Periodicals, Inc. on behalf of the American Physiological Society and The Physiological Society.

Figures

Figure 1.

Experimental setup (MLF and LFT). (A) Experiment 1 setup. Top panel: Schematic illustration of MLF stimulation (8‐mm posterior and 4‐mm ventral from bregma). Middle panel: Examples of sciatic nerve responses evoked by a single 0.5 mA pulse applied at the left MLF. Note that two responses were distinguished: short and long latency. Bottom panel: Examples of sciatic nerve potentials evoked by stimulating the left MLF with a train of five pulses (0.2 mA; 400 Hz). Right: Examples of muscle twitches simultaneously recorded from left TA and TS muscles. Two twitches (1 and 2) were evoked in response to a single‐pulse test stimulation despite inhibition of TS muscle tension. High‐frequency test stimulation of LMF produced a single twitch in TA muscle (1) and reduced TS muscle tension, on which a twitch (1) was superimposed. (B) Experiment 2 setup: Stimulation (S) was delivered at the LFT, and recording sites were at LFT (R1) and the sciatic nerve (R2). The tsDC electrode location is colored cyan. Muscle twitch force was recorded from TS and TA. Bottom: Examples of spinal tract potential trace (red) and sciatic nerve potential trace (blue). Waves are numbered. Lower traces represent the concurrent twitch forces recorded form TS and TA muscles.

Figure 2.

Experiment 3 setup (sciatic nerve). A. Single electrode: the DC electrode (DC) was a flat stainless steel plate (width, 7 mm) that was glued onto insulating material (silicone rubber). A reference electrode (Ref.) was attached to a flap of abdominal skin (insert). A hook electrode, placed between the muscle (3 mm) and the DC electrode (4 mm), was used to record nerve CAP (nCAP). The reference electrode was a needle inserted into the hindpaw skin. A bipolar concentric stimulating electrode was located 4 mm from the DC electrode. The nerve underneath both the stimulating and recording electrodes was placed on insulating material. Petroleum jelly (Pet.J) was mixed with silicone oil and applied, as seen in A, to create three isolated chambers: stimulating electrode chamber, DC electrode chamber, and recording electrode chamber. B. b1 is an example of a recording that includes an overlay of muscle twitch (MT) and nerve CAP (nCAP). b2 shows a graded nCAP series with increasing intensity to test stimulation current; therefore, the CAP amplitude was used as an indicator of nerve excitability. b3 is an example of nerve activity during DC application. DC was kept below threshold. C. Parallel electrodes: The general setup is similar to the single‐electrode setup, except that DC was applied using two plates running parallel to the nerve. The plates (length, 7 mm) were glued to silicone rubber, and petroleum jelly was used to create a chamber connecting the two plates. This chamber was filled with Ringer solution (RS). Great care was taken to keep DC electrodes completely insulated from the animal's body. Perpendicular electrodes: DC plates (width, 3.5 mm) were placed perpendicular to the nerve.

Figure 3.

Immediate effects of tsDCS on MLF sciatic nerve‐evoked responses. (A) Single pulse. Left panel: Examples of sciatic nerve responses recorded before stimulation (blue, baseline), during c‐tsDCS (black) and a‐tsDCS (red), and immediately after offset (blue). Right: Concurrent muscle responses. Note that c‐tsDCS increased while a‐tsDCS decreased sciatic nerve potentials and concurrent muscle twitches. Numbers mark the two muscle twitches (1 and 2). Note that TS twitch 2 was not evoked during a‐tsDCS. (B) Box plot showing latency changes of short‐ and long‐latency sciatic nerve potentials. Short‐ and long‐latency responses were shortened during c‐tsDCS and prolonged during a‐tsDCS. (C) High frequency train (five pulses). Left: Sciatic nerve potential was significantly increased during c‐tsDCS and significantly decreased during a‐tsDCS. Right: Concurrent muscle twitches. (D) Box plot showing latency of sciatic nerve potentials. c‐tsDCS significantly shortened latency, but a‐tsDCS had no effect. *P < 0.01.

Figure 4.

a‐transspinal direct current stimulation modified activation of synaptic and nonsynaptic responses. Sciatic nerve potentials and muscle responses were evoked by LFT stimulation. (A) Comparison of mean responses of spinal waves. (B) Comparison of mean responses of sciatic waves. (C) Comparison of mean twitch force of TS and TA muscles. All lines depict the average response at baseline (BL). (D) Of the three spinal waves (1, 2 and 3), only wave 3 significantly correlated with muscle twitch force. Similarly, only sciatic wave 2 significantly correlated with muscle force. Data represent mean ± SEM.

Figure 5.

c‐transspinal direct current stimulation modified activation of synaptic and nonsynaptic responses. (A) Summary plot showing all spinal waves that were significantly amplified by c‐tsDCS. Horizontal lines mark the averages at baseline (BL). (B) Summary plots showing that sciatic nerve waves were amplified during and after c‐tsDCS. (C) Summary plot showing that TS and TA twitch force was significantly increased during and after c‐tsDCS. All values above the horizontal lines were statistically significant. Data represent mean ± SEM.

Figure 6.

Effects of single‐sub‐DC electrode stimulation on sciatic nerve excitability. (A) Examples of CAP traces recorded from nerve segment lie in front of the sub‐DC electrode. (B) Test stimulation at the nerve segment in front of the subDCS electrode. a‐sDCS enhanced nerve excitability for 25 min, and c‐sDCS depressed nerve excitability for 25 min following current offset. Note that the direction of excitability was reversed after subDCS compared to during subDCS. (C) Test stimulation proximal to the DC electrode. a‐subDCS enhanced nerve excitability for 15 min, and c‐subDCS depressed nerve excitability for at least 25 min following current offset. (D) Nerve excitability changes could be reversed by applying the opposite polarity. Top: Experimental outline. Bottom: Summary plot showing that applying a current with opposite polarity could reverse the effect of the previous current. As evident in the summary plot, a‐subDCS not only reversed c‐subDCS‐induced depression, but its effect significantly exceeded that of baseline. Similarly, c‐subDCS not only reversed a‐subDCS‐induced enhancement, but reduced CAP significantly from baseline. Data represent means ± SEM. *P < 0.05 from baseline; **P < 0.05 from the corresponding subDCS condition.

Figure 7.

Effects of parallel subDC electrode arrangement on sciatic nerve excitability. In A–D, between‐electrode test stimulation was used. (A,B) When the nerve was centered between the two sDC electrodes, both lateral to medial subDCS and medial to lateral subDCS caused significant inhibition. (C) When the nerve was moved closer to the anode, the inhibition was enhanced. (D) When the nerve was moved closer to the cathode, the excitability was enhanced. In E–H, between‐electrode test stimulation was used. (E,F) When the nerve was centered between the two sDC electrodes, both lateral to medial subDCS and medial to lateral subDCS caused enhancement that was linearly related to current strength. (G) When the nerve was moved closer to the anode, subDCS significantly enhanced excitability. (H) Conversely, when the nerve was moved closer to the cathode, subDCS significantly inhibited nerve excitability. (I) Long‐lasting effects of parallel sDC electrode arrangement on sciatic nerve excitability. Insets on the right of the figure show the different experimental setup. When the nerve was centered between the two sDC electrodes, lateral to medial subDCS caused inhibition during subDCS. However, nerve excitability increased significantly after subDCS offset, and this enhancement lasted at least 25 min. When the nerve was moved closer to the cathode, nerve excitability decreased significantly during subDCS and for 10 min after subDCS offset, then increased significantly from 15 to 25 min. When the nerve was moved closer to the anode, nerve excitability increased significantly during sDC and lasted at least 25 min after subDCS offset. Data represent means ± S.E.M. *P < 0.05 relative to baseline.

Figure 8.

Long‐lasting effects of perpendicular subDCS electrode arrangement on sciatic nerve excitability. (A) Proximal test stimulation. Distal to proximal subDCS caused inhibition that was evident 25 min after current offset. Proximal to distal subDCS caused enhancement that lasted at least 25 min after current offset. (B) Between‐electrode test stimulation. Distal to proximal subDCS caused enhancement that lasted at least 25 min. Proximal to distal subDCS caused inhibition during subDCS, but significantly enhanced CAP after subDCS offset. Insets on the top of the figure show the experimental setup. Data represent means ± SEM. *P < 0.05 relative to baseline.