The innervation and organization of motor units in a series-fibered human muscle: the brachioradialis

Abstract

We studied the innervation and organization of motor units in the brachioradialis muscle of 25 normal human subjects. We recorded intramuscular EMG signals at points separated by 15 mm along the proximodistal muscle axis during moderate isometric contractions, identified from 27 to 61 (mean 39) individual motor units per subject using EMG decomposition, and estimated the locations of the endplates and distal muscle/tendon junctions from the motor-unit action potential (MUAP) propagation patterns and terminal standing waves. In three subjects all the motor units were innervated in a single endplate zone. In the other 22 subjects, the motor units were innervated in 3-6 (mean 4) distinct endplate zones separated by 15-55 mm along the proximodistal axis. One-third of the motor units had fibers innervated in more than one zone. The more distally innervated motor units had distinct terminal waves indicating tendonous termination, while the more proximal motor units lacked terminal waves, indicating intrafascicular termination. Analysis of blocked MUAP components revealed that 19% of the motor units had at least one doubly innervated fiber, i.e., a fiber innervated in two different endplate zones by two different motoneurons, and thus belonging to two different motor units. These results are consistent with the brachioradialis muscle having a series-fibered architecture consisting of multiple, overlapping bands of muscle fibers in most individuals and a simple parallel-fibered architecture in some individuals.

Fig. 1.

Estimation of motor unit (MU) architecture. Intramuscular EMG signals were analyzed to identify individual MUs and average their MU action potential (MUAP) waveforms at evenly spaced recording sites (filled circles) along the proximodistal axis of the brachioradialis muscle (BR). The top traces show a raster plot of 1 MUAP; the bottom plots show the same signals superimposed and scaled to better illustrate the standing components. (In this and the following figures, MUAPs are plotted with negative polarity upward.) The MUAP onset (o) and terminal wave (t) were identified by inspection. The latency of the MUAP spike was estimated at each recording site (filled squares), and lines of propagation were drawn to interpolate them (gray lines). The MUAP endplate (e) was estimated from the point of intersection of the lines of propagation (open circles), which usually occurred at a latency of ∼2 ms after the MUAP onset (i). The distal MUAP termination (T) was estimated from the point at which the distal line of propagation intersected the latency of the terminal wave (open squares).

Fig. 2.

Raster plots of 2 MUAPs from a single-endplate-zone muscle. Both MUAPs had 2 propagating waves that originated from the same endplate zone located at 20 mm. See Fig. 1 for an explanation of the symbols. The dotted lines indicate the MUAP onsets and terminal waves. The inset shows the terminal waves of MU 2.1 superimposed to highlight their constant latency and change of sign. The architecture of each MU is shown schematically to the left of each raster plot, with e indicating the endplate and T the muscle tendon junction. The MUAP of MU 2.2 had several late components. The potentials marked a were volatile satellite potentials from doubly innervated branched fibers (bf) and were propagating proximally from a branching point at ∼70 mm (dashed gray line). The potentials marked b were satellite potentials that did not exhibit blocking. The diagram on the right summarizes the location of all the endplates found in this subject.

Fig. 3.

Raster plots of 4 MUAPs from a muscle with multiple endplate zones. Each MUAP had a different propagation pattern. MU 3.1 had only a single propagating wave that originated proximal to the electrode at 0 mm. Its endplate was estimated by extrapolation to be at −15 mm. MU 3.2 had three waves: a pair originating from an endplate at 40 mm, and a single wave originating from an endplate at −15 mm. MUs 3.3 and 3.4 had 2 waves originating from endplates at 55 and 80 mm, respectively. Only MU 3.4 had a detectable terminal wave. The diagram at right summarizes the locations of all the endplates found in this subject.

Fig. 4.

Distribution of endplate zones for each subject, ordered by the location of the most distal endplate zone. Each line extends to the distalmost termination detected in the subject. Subject 15 is not shown because the locations of the endplate zones could not be reliably determined. In subject 17, one endplate zone was atypically long (*).

Fig. 5.

Raster plots of 2 MUAPs from the same muscle recorded at 2 different depths. MU 5.1 was from a deeper MU and had a distinct terminal wave indicating a distal termination at ∼75 mm. MU 5.2 was from a more superficial MU and showed propagation to at least 105 mm.

Fig. 6.

Doubly innervated muscle fibers. A: MUAPs of 2 MUs that exhibited interdependent shape irregularity at the 3 recording sites indicated by the filled circles. B: MUAP spikes from the 3 recording sites after 1-kHz high-pass filtering. In each column, the top traces show several occurrences of the normal spike, the middle traces show several occurrences of the spike when the volatile component was blocked, and the bottom trace shows the waveform of the volatile component, obtained by averaging and subtracting the top 2 traces. The similarity of the volatile components at each recording site attests to their having come from the same muscle fiber. C: blocking diagrams at the 3 recording sites. Each symbol indicates one discharge of both MUs, plotted as a function of the time interval between the MUAP onsets. Open circles: volatile component of MU 6.2 delayed. Upper x's: volatile component of MU 6.2 blocked; lower x's: volatile component of MU 6.1 blocked; filled circles: volatile component of MU 6.1 delayed; w₁+w₂: combined blocking window. D: schematic diagram of a doubly innervated fiber. The arrows illustrate an action potential collision when MU 6.1 discharges a few milliseconds before MU 6.2. The action potential associated with MU 6.1 (white arrow) reaches the electrode at 0 mm but is blocked from reaching the other electrodes. Although only 1 fiber is shown in the diagram, the volatile components observed at the 3 different recording sites may have come from 3 different fibers.

Fig. 7.

Doubly innervated muscle fiber for which only 1 parent MU was well seen. A: the MUAPs of MU 7.1 and MU 7.2 exhibited interdependent shape irregularity at the 75-mm recording site. MU 7.2 was not well seen by the electrodes. B: blocking diagram, as in Fig. 6. Since the onset of MUAP 7.2 could not be detected, the blocking behavior is plotted as a function of the interval between the times of occurrence of the stable components of the 2 MUAPs observed at 75 mm. The scatter in the open circles reflects a large intercomponent jitter resulting from the long propagation distance between the 30- and 75-mm recording sites.