Paired associative stimulation induces change in presynaptic inhibition of Ia terminals in wrist flexors in humans

Abstract

Enhancements in the strength of corticospinal projections to muscles are induced in conscious humans by paired associative stimulation (PAS) to the motor cortex. Although most of the previous studies support the hypothesis that the increase of the amplitude of motor evoked potentials (MEPs) by PAS involves long-term potentiation (LTP)-like mechanism in cortical synapses, changes in spinal excitability after PAS have been reported, suggestive of parallel modifications in both cortical and spinal excitability. In a first series of experiments (experiment 1), we confirmed that both flexor carpi radialis (FCR) MEPs and FCR H reflex recruitment curves are enhanced by PAS. To elucidate the mechanism responsible for this change in the H reflex amplitude, we tested, using the same subjects, the hypothesis that enhanced H reflexes are caused by a down-regulation of the efficacy of mechanisms controlling Ia afferent discharge, including presynaptic Ia inhibition and postactivation depression. To address this question, amounts of both presynaptic Ia inhibition of FCR Ia terminals (D1 and D2 inhibitions methods; experiment 2) and postactivation depression (experiment 3) were determined before and after PAS. Results showed that PAS induces a significant decrease of presynaptic Ia inhibition of FCR terminals, which was concomitant with the facilitation of the H reflex. Postactivation depression was unaffected by PAS. It is argued that enhancement of segmental excitation by PAS relies on a selective effect of PAS on the interneurons controlling presynaptic inhibition of Ia terminals.

Fig. 1.

Recruitment curves of flexor carpi radialis (FCR) H reflex (A), M waves (B), and motor-evoked potentials (MEPs) (C) obtained before (continuous line–T0) and after (dashed line–T1) paired associative stimulation (PAS)₂₀ in 1 representative subject. Each dot represents the mean (±SD) of 10 H reflexes, M waves, or MEPs. Values are expressed in percentage of M_max. Sigmoid curve-fitting analysis was used to plot the H reflex recruitment curve. In this subject, PAS₂₀ resulted in an increase of both H reflexes and MEP amplitudes without concomitant change of M waves. Right: waveforms of H reflex and M wave collected at 1.1 × motor threshold (MT) (top) and of MEPs collected at 1.2 × rMT (bottom).

Fig. 2.

Box plots representing average variations, obtained in 12 healthy volunteers, of slope parameter m (A), S₅₀ (B), H_threshold (C), H_max (D), M wave AUC (E), and MEP AUC (F) before (open box plots–T0) and after (gray box plots–T1) delivery of PAS₂₀. For each box plot, the boundary of the box closest to abscissa indicates the 25th percentile, the plain line within the box indicates the median, and the boundary of the box farthest from abscissa indicates the 75th percentile. Whiskers above and below the box indicate the minimum and maximum values observed, respectively (*P < 0.05).

Fig. 3.

Modeling of the FCR H reflex recruitment curves obtained before (continuous line–T0) and after (dashed line–T1) PAS₂₀ for the whole population (12 healthy volunteers). Hypothetical H reflex recruitment curves were built using averaged values of individual H_threshold, slope parameter m, S₅₀, and H_max values. PAS₂₀ induced significant increase of both the slope parameter m and the plateau value (H_max) without concomitant significant change of S₅₀ or H_threshold.

Fig. 4.

Effect of PAS₂₀ on presynaptic inhibition of FCR Ia terminals. A: wiring diagram of experimental set-up used to assess presynaptic inhibition of FCR Ia terminals and example of waveforms collected without (continuous line) and with (dashed line) a conditioning radial nerve stimulation delivered 15 ms before the median nerve stimulation in 1 representative subject. B and C: mean amount of D1 inhibition (10–30 ms, conditioning test interval, 12 subjects) (B) and of D2 inhibition (50–500 ms, conditioning-test interval, 6 subjects) (C) obtained before (open bars–T0) and after (black bars–T1) PAS₂₀. The ordinate shows the amount of inhibition calculated as [100 – conditioned FCR H reflex (as a percentage of the control reflex size)], and the abscissa shows the conditioning test interval (ms). At each condition, 20 control and 20 conditioned H reflexes were averaged for each participant. Bars indicate SD (*P < 0.05). D and E: individual results of the modulation of AUCs by PAS₂₀ of D1 inhibition (D, dark gray bars) and D2 inhibition (E, dark gray bars), FCR H reflex curve (black bars) and MEP (middle gray bars). Percentage of modulation by PAS₂₀ was calculated as {[AUC (T1) – AUC (T0)]/AUC (T0)} × 100.

Fig. 5.

Effect of PAS₂₀ on postactivation depression. A: wiring diagram of experimental set-up used to assess postactivation depression of FCR Ia terminals and example of waveforms, collected at 0.5 (continuous line) and 0.125 Hz (dashed line). B: mean amount of postactivation depression (11 subjects) obtained before (open bars–T0) and after (black bars–T1) PAS₂₀. The ordinate shows the FCR H reflex amplitude (expressed as a percentage of M_max), and the abscissa shows the frequency of the median nerve stimulation (Hz). At each condition, 20 H reflexes were averaged for each participant. Bars indicate SD. C: individual results (11 subjects) of the modulation of AUCs by PAS₂₀ of postactivation depression (dark gray bars), FCR H reflex curve (black bars), and MEP (middle gray bars). Note that subject 3 was discarded from the analysis because the amplitude of the H reflex at 0.1 Hz before and after PAS mismatched. Percentage of modulation by PAS₂₀ was calculated as {[AUC (T1) – AUC (T0)]/AUC (T0)} × 100.