A sensory signal related to left-right symmetry modulates intra- and interlimb cutaneous reflexes during locomotion in intact cats

Abstract

Introduction: During locomotion, cutaneous reflexes play an essential role in rapidly responding to an external perturbation, for example, to prevent a fall when the foot contacts an obstacle. In cats and humans, cutaneous reflexes involve all four limbs and are task- and phase modulated to generate functionally appropriate whole-body responses.

Methods: To assess task-dependent modulation of cutaneous interlimb reflexes, we electrically stimulated the superficial radial or superficial peroneal nerves in adult cats and recorded muscle activity in the four limbs during tied-belt (equal left-right speeds) and split-belt (different left-right speeds) locomotion.

Results: We show that the pattern of intra- and interlimb cutaneous reflexes in fore- and hindlimbs muscles and their phase-dependent modulation were conserved during tied-belt and split-belt locomotion. Short-latency cutaneous reflex responses to muscles of the stimulated limb were more likely to be evoked and phase-modulated when compared to muscles in the other limbs. In some muscles, the degree of reflex modulation was significantly reduced during split-belt locomotion compared to tied-belt conditions. Split-belt locomotion increased the step-by-step variability of left-right symmetry, particularly spatially.

Discussion: These results suggest that sensory signals related to left-right symmetry reduce cutaneous reflex modulation, potentially to avoid destabilizing an unstable pattern.

Keywords: cutaneous reflexes; interlimb coordination; locomotion; modulation; split-belt treadmill.

FIGURE 1

Experimental set-up and spatiotemporal parameters. (A) Experimental design illustrating a cat walking on a split-belt treadmill. The superficial radial (SR) and peroneal (SP) nerves were electrically stimulated. (B) The stimulated and recorded limbs are displayed in red and black, respectively. (C) Spatial and temporal parameters of the left fore- and hindlimb during a locomotor cycle divided in 10 phases (0–100%). LHST, left hindlimb stance; RHST, right hindlimb stance.

FIGURE 2

Reflex analysis. (A) We tagged cycles as stimulated (S) when a stimulus fell within the cycle or control (C) if it was not preceded by a stimulated cycle. (B) We averaged control cycles for the different muscles normalized to the ipsilateral sartorius (iSRT) onset. Each normalized cycle was separated into 10 bins to provide a baseline locomotor EMG (blEMG) in each phase. (C) Stimulated cycles (black traces) were averaged into 1 of 10 bins and superimposed on the blEMG (gray traces). This allowed us to determine positive (in red) and negative (in blue) responses. (D) Onsets and offsets of responses, defined as a prominent positive or negative deflection away from the blEMG, were determined visually. In each phase, the blEMG occurring in the same time window as the response was subtracted from the response in the stimulated cycles to provide a net reflex value. This value is then divided by the blEMG occurring in the same time window giving N1 and P3 amplitudes for this muscle (ipsilateral soleus).

FIGURE 3

Modulation of the locomotor pattern during tied-belt and split-belt locomotion in a single cat. The figure shows EMG activities for selected forelimb (BB and ECU) and hindlimb (SRT and LG) muscles bilaterally along with the stance phases (thick black bars) for all four limbs. The vertical and horizontal scales are the same for a given muscle in all four panels. L, left; R, right; LG, lateral gastrocnemius; SRT, anterior sartorius; ECU, extensor carpi ulnaris; BB, biceps brachii; RFST, right forelimb stance; LFST, left forelimb stance; RHST, right hindlimb stance; LHST, left hindlimb stance.

FIGURE 4

Temporal and spatial parameters during tied-belt and split-belt locomotion across animals. (A) Cycle, stance, and swing durations are shown for the left forelimb and hindlimb. (B) Step and stride lengths are shown for the left forelimb and hindlimb. Each bar is the mean ± SD from 8 cats. Asterisks indicate significant differences between locomotor conditions (pairwise comparisons): *P < 0.05; **P < 0.01; ***P < 0.001.

FIGURE 5

Phase intervals between limb pairs during tied-belt and split-belt locomotion in a single cat and across animals. Circular plots show phase intervals expressed in degrees around the circumference while cycle durations are plotted in radii. (A) Each data point represents a locomotor cycle in a single cat. (B) Each data point represents the average of 7–21 locomotor cycles from the eight cats. Straight lines starting from the center of the circle to the circumference display mean directions for each locomotor condition. P-values in bold indicate whether mean directions between two locomotor conditions were significantly different for group data (Watson-Williams’ test).

FIGURE 6

Gap intervals for forelimb and hindlimb couplings during tied-belt and split-belt locomotion in a single cat and across animals. Circular plots show gap intervals expressed in degrees around the circumference stride lengths are plotted in radii. (A) Each data point represents a locomotor cycle in a single cat. (B) Each data point represents the average of 7–21 locomotor cycles from the eight cats. Straight lines starting from the center of the circle to the circumference display mean directions for each locomotor condition. P-values in bold indicate whether mean directions between two locomotor conditions were significantly different for group data (Watson-Williams’ test).

FIGURE 7

Spatiotemporal coordination for forelimb and hindlimb couplings during tied-belt and split-belt locomotion across animals. Histograms show temporal and spatial symmetry indexes in the four locomotor conditions across animals. Smaller percentage values indicate greater left-right symmetry between forelimb/hindlimb pairs. Each vertical bar is the mean ± SD from eight cats. Asterisks indicate significant differences between conditions (pairwise comparisons): *P < 0.05; **P < 0.01; ***P < 0.001.

FIGURE 8

Intra- and interlimb reflexes in a forelimb muscle during tied-belt and split-belt locomotion. Each panel shows reflex responses in 10 phases of the cycle evoked in the extensor carpi ulnaris (ECU) in a single cat by (A) stimulating the homonymous superficial radial (SR) nerve, (B) the crossed SR, (C) the homolateral superficial peroneal (SP) nerve, and (D) the diagonal SP in the four locomotor conditions. Black traces are averaged cycles that received a stimulation (n = 4–26 stimuli per phase) while gray traces are averaged cycles (blEMG) without stimulation (n = 88–140 cycles). Scale bars are shown in arbitrary units (a.u.) and are the same across phases and conditions for each limb. Aligned vertically on the right side of each panel is the averaged rectified EMG of the ECU in the normalized cycle. At the bottom of each panel, the amplitude of short- and longer-latency responses is shown for the group within a normalized cycle to illustrate phase-dependent modulation. SR stimulation evoked homonymous (n = 12 P1/N1 in 8/8 cats and n = 9 P2 in 6/8 cats) and crossed (n = 8 P2 in 5/8 cats) responses while SP nerve stimulation evoked homolateral (n = 9 P1 in 7/8 cats and n = 12 N2 in 8/8 cats) and diagonal (n = 7 P2 in 5/8 cats) responses. Horizontal bars at the top represent the period of ECU activity for each locomotor condition (n = 7–19 control cycles) for pooled data.

FIGURE 9

Phase-dependent modulation of forelimb cutaneous reflexes during tied-belt and split-belt locomotion. Homonymous, crossed, homolateral, and diagonal responses are shown for biceps brachii (BB), triceps brachii (TRI), latissimus dorsi (LD), extensor (ECU), and flexor (FCU) carpi ulnaris muscles. The limbs stimulated and recorded in the cat diagram are displayed in red and black, respectively. Each black trace represents averaged stimulated cycles (n = 6–32 cycles) for a locomotor condition during a muscle’s period of mid-activity or mid-inactivity. The four locomotor conditions are superimposed for comparisons in a representative cat and optimized for display. Evoked responses are highlighted in red for positive and blue for negative. The fraction for each response indicates the proportion of pooled data evoking the same pattern of response in all locomotor conditions.

FIGURE 10

Forelimb reflex modulation across locomotor conditions for the group. Modulation indexes are shown for short- and longer-latency responses for all five forelimb muscles in the four locomotor conditions. Each bar represents the mean ± SD for pooled data. Note that only responses present in the four locomotor conditions in the same cat were pooled. P-values comparing conditions are indicated (main effect of repeated-measures ANOVA). Asterisks indicate significant differences between conditions (pairwise comparisons): *P < 0.05; **P < 0.01. BB, biceps brachii; TRI, triceps brachii; LD, latissimus dorsi; ECU, extensor carpi ulnaris; FCU, flexor carpi ulnaris.

FIGURE 11

Intra- and interlimb reflexes in a hindlimb muscle during tied-belt and split-belt locomotion. Each panel shows reflex responses in 10 phases of the cycle evoked in the soleus (SOL) in a single cat by (A) stimulating the homonymous superficial peroneal (SP) nerve, (B) the crossed SP, (C) the homolateral superficial radial (SR) nerve, and (D) the diagonal SR in the four locomotor conditions. Black traces are averaged cycles that received a stimulation (n = 4–21 stimuli per phase) while gray traces are averaged cycles (blEMG) without stimulation (n = 74–137 cycles). Scale bars are shown in arbitrary units (a.u.) and are the same across phases and conditions for responses in each limb. Aligned vertically on the right side of each panel is the averaged rectified EMG of the SOL in the normalized cycle. At the bottom of each panel, the amplitude of short- and longer-latency responses is shown for the group within a normalized cycle to illustrate phase-dependent modulation. SP nerve stimulation evoked homonymous (n = 8 P1/N1 in 5/8 cats, n = 6 P2 in 5/8 cats and n = 8 P3 in 5/8 cats) and crossed (n = 6 P2 in 4/8 cats and n = 8 N3 in 6/7 cats) responses while SR nerve stimulation evoked homolateral (n = 6 P2 in 4/8 cats and n = 8 N3 in 5/8 cats) and diagonal (n = 7 N2 in 4/8 cats and n = 5 P3 in 3/8 cats) responses. Horizontal bars at the top represent the period of SOL activity for each locomotor condition (n = 7–19 control cycles) for pooled data.

FIGURE 12

Phase-dependent modulation of hindlimb cutaneous reflexes during tied-belt and split-belt locomotion. Homonymous, crossed, homolateral, and diagonal responses are shown for anterior sartorius (SRT), semitendinosus (ST), vastus lateralis (VL), iliopsoas (IP), biceps femoris posterior (BFP), biceps femoris anterior (BFA), lateral gastrocnemius (LG), soleus (SOL), medial gastrocnemius (MG), and tibialis anterior (TA) muscles. The limbs stimulated and recorded in the cat diagram are displayed in red and black, respectively. Each black trace represents averaged stimulated cycles (n = 5–32 cycles) for a locomotor condition during a muscle’s period of mid-activity or mid-inactivity. The four locomotor conditions are superimposed for comparisons in a representative cat and optimized for display. Evoked responses are highlighted in red for positive and blue for negative. The fraction for each response indicates the proportion of pooled data evoking the same pattern of response in all locomotor conditions.

FIGURE 13

Hindlimb reflex modulation across locomotor conditions for the group. Modulation indexes are shown for short- and longer-latency responses for all ten forelimb muscles in the four locomotor conditions. Each bar represents the mean ± SD for pooled data. Note that only responses present in the four locomotor conditions in the same cat were pooled. P-values comparing conditions are indicated (main effect of repeated-measures ANOVA). Asterisks indicate significant differences between conditions (pairwise comparisons): *P < 0.05; **P < 0.01. SRT, anterior sartorius; ST, semitendinosus; VL, vastus lateralis; IP, iliopsoas; BFP, biceps femoris posterior; BFA, biceps femoris anterior; LG, lateral gastrocnemius; SOL, soleus; MG, medial gastrocnemius; TA, tibialis anterior.