Activities of the primary and supplementary motor areas increase in preparation and execution of voluntary muscle relaxation: an event-related fMRI study

Abstract

Brain activity associated with voluntary muscle relaxation was examined by applying event-related functional magnetic resonance imaging (fMRI) technique, which enables us to observe change of fMRI signals associated with a single motor trial. The subject voluntarily relaxed or contracted the right upper limb muscles. Each motor mode had two conditions; one required joint movement, and the other did not. Five axial images covering the primary motor area (M1) and supplementary motor area (SMA) were obtained once every second, using an echoplanar 1.5 tesla MRI scanner. One session consisted of 60 dynamic scans (i.e., 60 sec). The subject performed a single motor trial (i.e., relaxation or contraction) during one session in his own time. Ten sessions were done for each task. During fMRI scanning, electromyogram (EMG) was monitored from the right forearm muscles to identify the motor onset. We calculated the correlation between the obtained fMRI signal and the expected hemodynamic response. The muscle relaxation showed transient signal increase time-locked to the EMG offset in the M1 contralateral to the movement and bilateral SMAs, where activation was observed also in the muscle contraction. Activated volume in both the rostral and caudal parts of SMA was significantly larger for the muscle relaxation than for the muscle contraction (p < 0.05). The results suggest that voluntary muscle relaxation occurs as a consequence of excitation of corticospinal projection neurons or intracortical inhibitory interneurons, or both, in the M1 and SMA, and both pre-SMA and SMA proper play an important role in motor inhibition.

Figures

Fig. 1.

Surface EMGs recorded from the bilateral forearm muscles during fMRI scanning (R_mv task). Despite the conspicuous signals caused by radio frequency pulses on the record, the EMG offset (arrow) can be identified, which happens to coincide with the 30th scan in this particular session. ECR, Extensor carpi radialis muscle; FCU, flexor carpi ulnaris muscle.

Fig. 2.

Three box-car functions used in the present analysis for representing premotor, motor, and postmotor phases. Motor phase contains three scans: one coinciding with the EMG onset or offset and the preceding two. Each time series is convolved with a Gaussian-shaped hemodynamic response function to produce three regressors (expected hemodynamic responses). ⊗ denotes convolution.

Fig. 3.

Activated areas for the isometric muscle relaxation (R_iso) and contraction (C_iso) of the right hand (A) and time course of signal change in the contralateral primary motor area (Lt M1) and supplementary motor area (Lt SMA) for the R_iso (B), obtained from a single subject.A, Activated areas, which showed a significant transient increase of activity time-locked to the motor trial, are superimposed on the subject’s own anatomic MRI. Brighter gray colorrepresents higher Z score. The illustrated slice corresponds to an axial plane of z = 60 mm in the standard space (Talairach and Tournoux, 1988). Activation is observed in the left M1 and bilateral SMAs for both the R_iso and C_iso.CS, Central sulcus. B, Averaged signals across 10 sessions at the voxel showing the maximum Zscore in the left M1 and SMA are represented by solid lines. Each dot represents data from a single trial at each sampling point. The vertical lineindicates the offset of EMG activity. Clear transient increase of activity is observed even in a single trial and is highly reproducible.

Fig. 4.

Activated areas (A) and time course of the hemodynamic response (B) in each task. A, Spatially normalized SPM {Z} maps were superimposed across eight subjects on an axial plane (z = 56 mm) of the averaged anatomic MRI from all the subjects. Note that this figure does not represent the direct statistical analysis across multiple subjects, instead the illustrated areas indicate the overlap in three or more of eight subjects. The right side of the brain is shown on the right side of the image. Brighter yellow color represents a larger number of subjects showing significant activation in the pixel. The majority of subjects show common activation in the contralateral (left) M1 and bilateral SMAs for the R_mv and R_iso.B, Averaged signal changes at the local maxima in the left M1 and SMA across the subjects who showed significant activation during each task. Vertical interrupted lines represent the onset of motor trial. A similar signal change is observed in the four motor tasks in the left M1 and SMA. Note that the number of subjects is different among tasks and among regions.

Fig. 5.

Activation on the medial frontal wall during the R_iso and C_iso in a single subject. The illustrated slice corresponds to a sagittal plane of x = 0 mm in the standard space (Talairach and Tournoux, 1988). Activated area over the medial frontal wall is larger in the R_iso than in the C_iso.

Fig. 6.

Volume of the activated areas during each task (mean ± SD) in the bilateral pre-SMA (A), SMA proper (B), and the contralateral (left) M1 (C). The pre-SMA and SMA proper show significantly larger activation for the relaxation mode than for the contraction mode but no difference between the two conditions (i.e., movement vs isometric). By contrast, the activated volume in the left M1 is significantly larger for the isometric condition than for the movement condition, whereas there was no difference between the two modes (i.e., relaxation vs contraction). *p < 0.05 by two-factors ANOVA.