0006-Ch05.pdf

Investigating the

Synaptic Control of

Human Motoneurons:

New Techniques,

Analyses, and Insights

from Animal Models

Randall K. Powers and Kemal S. Türker

CONTENTS

5.1 Introduction

5.2 Motor Unit Isolation

5.3 Stimulation Techniques

5.3.1 Electrical Stimulation

5.3.2 Mechanical Stimulation

5.3.3 Cortical Stimulation

5.3.4 Microneurography

5.4 Analyzing Afferent Effects on Motor Unit Discharge

5.4.1 Surface Electromyogram (SEMG)

5.4.2 Intramuscular Multi-Unit EMG

5.4.3 Single Motor Unit Discharge

5.4.4 Alternative Methods of Estimating Synaptic Effects on Single

Motor Units

5.5 Insights from Work on Animal Models

5.5.1 Relation of Synaptic Potential Time Course to Effects

on Firing Probability

5.5.2 Predictions from Threshold-Crossing Models

5.5.3 Using Injected Current Transients to Mimic Synaptic Potentials

5.5.3.1 Effects of Simulated Excitatory Synaptic Inputs on

Motoneuron Discharge

5.5.3.2 Effects of Simulated Inhibitory Synaptic Inputs on

Motoneuron Discharge

5.5.3.3 Effects of Background Motoneuron Discharge Rate

5.5.3.4 Effects of Background Noise

5.5.4 White-Noise Analysis of Spike Encoding in Motoneurons

5.6 Summary

References

5.1 INTRODUCTION

The contraction of a group of muscle ﬁbers is directly dependent upon the activity

of the motoneuron connected to those ﬁbers, and there is a one-to-one relation

between the occurrence of action potentials in the motoneuron axon and those in its

innervated muscle ﬁbers. For this reason, the effects of activity in afferents, inter-

neurons, and descending ﬁbers on motor behavior are dependent upon their ability

to modulate the discharge of motoneurons. The unique importance of the motoneuron

in motor control was recognized by Sherrington, 1 who introduced the concept of the

motoneuron as the ﬁnal common path between activity in the central nervous system

and movement. The one-to-one relation between the occurrence of action potentials

in a motoneuron and in the muscle ﬁbers they innervate makes it possible to record

the discharge behaviour of single motoneurons using intramuscular electrodes. Con-

sequently, motoneurons are the only CNS neurons whose individual discharge behav-

iour can be recorded in human subjects during the execution of normal movements.

The ability to record the activity of human motor units has made it possible to

draw inferences regarding the organization of synaptic inputs to human motoneurons

during various motor tasks, in both healthy subjects and in subjects suffering from

a variety of nervous system disorders. These inferences are often based on the effects

of stimulating peripheral or descending afferents on motoneuron discharge proba-

bility. It is generally accepted that excitatory synaptic inputs lead to an increase in

discharge probability, whereas inhibitory synaptic inputs lead to a decrease in dis-

charge probability. However, the exact relation between the time course of a synaptic

potential and its effects on motoneuron discharge probability is still a matter of some

controversy. Nonetheless, there have been a number of methodological and theoret-

ical advances in the last 20 years that have led to an improved understanding of the

synaptic control of human motoneurons. In the following review, we focus on four

different advances related to measuring and interpreting synaptic effects on the

discharge of human motoneurons: (1) improvements in isolating the activity of single

motor units; (2) new techniques for stimulating different afferent inputs; (3) new

methods of analyzing afferent effects on motor unit discharge; and (4) recent insights

from animal models.

5.2 MOTOR UNIT ISOLATION

Although it is possible to record the activity of single motor units from the muscle

surface, the contribution of any motor unit to the surface electromyogram (SEMG)

depends upon its size and its distance from the surface electrodes. Therefore, whereas

action potentials from large motor units situated close to the electrodes appear as

large spikes on the SEMG, action potentials from small motor units situated in the

deep portions of the muscle can be in the noise level.

Insertion of needle or wire electrodes into muscle is commonly used to record

the activity of a more limited sample of motor units. The number of motor unit

action potentials (MUAPs) recorded using intramuscular electrodes depends upon

the extent of the uninsulated (active) area. 2 As the size of the active area of recording

is increased, so do the number of units recorded. Recording from a large number

of motor units is a useful technique to get the overall picture of a muscle’s activity

patterns. Furthermore, quantiﬁcation of multi-unit recording is easier and more

reliable than the SEMG records (see 5.3 ) . It is also possible to record the activity

of one or a few MUAPs using wire or needle electrodes with a smaller active

recording area. 2 From the few-unit record, it is possible to isolate one of the MUAPs

and study its response to a given stimulus.

Selection of a single MUAP from a few-unit record is not straightforward but

requires experience and special electronic circuits or computer programs. The most

common criterion used to discriminate the units is their peak amplitude. This is

achieved with simple level crossing circuitry. However, the units often may have

similar amplitudes but differ in the time taken to reach to the maximal amplitude

(frequency component of the MUAP). It is possible in some cases to produce an

amplitude difference between such units using bandpass ﬁltering. In other cases, some

peculiarities of the MUAP shapes can be used to distinguish different units. Time-

amplitude window circuits can be used for selecting such units. This technique uses a

combination of the slope of the leading edge and the amplitude of the unit and may

require two experienced persons (four hands!) working very closely together.

The most powerful approaches for discriminating single motor units use some

sort of computer-based pattern recognition or template-matching algorithm to char-

acterize the entire MUAP waveform. This has been described by a number of

laboratories, 3–7 and commercial waveform discriminators are now widely available.

These various motor unit recognition programs often use the regular nature of motor

unit discharge together with interaction of the computer operator to minimize missed

or incorrectly classiﬁed MUAPs. One of the ﬁrst approaches to sorting multiple

motor units utilized several different recording channels to improve discrimination, 3

since two motor units that exhibited similar shapes on a single channel often were easily

differentiated on one of the other channels. This approach has recently been used to

allow single unit discrimination based on an array of surface EMG electrodes. 8

Motor unit recognition is generally performed ofﬂine but in some cases 5 can be

used online based on a short epoch of previously collected data. In the latter case,

the operator ﬁrst sets a threshold voltage for incoming unit signals. This threshold

needs to be above the noise level but below the peak of the MUAP of interest. Whenever

the input signal crosses this threshold, the system is triggered; i.e., it will now consider

a millisecond or so of data on each side of the threshold crossing as a potential spike.

The template/s is established before the experimental run begins by running a few

seconds of data containing MUAPs of interest. The operator chooses one (or a few)

MUAP waveforms and stores them in the computer’s memory as the templates.

During the experimental run, the system is triggered each time the incoming

signal exceeds the threshold. A millisecond of the incoming signal on each side of

the trigger is then compared with the previously established templates. If any of the

templates match the incoming MUAP shape, the computer accepts that point as the

discharge of the unit. The discharge times of the motor unit can be saved and used

in variety of the analysis programs to estimate the synaptic potential that is developed

on a motoneuron by the stimulus (see 5.4 ) .

5.3 STIMULATION TECHNIQUES

Stimulating one afferent ﬁber type without affecting others has been a great challenge

for a number of years. Without invasive techniques, completely selective activation

of a single type of afferent ﬁber is generally not possible. However, given sufﬁcient

attention to stimulus parameters, it is often possible to apply stimuli that predomi-

nantly activate one class of afferent ﬁbers.

5.3.1 E LECTRICAL S TIMULATION

Electrical stimuli can be applied with ease to the skin to study the synaptic connection

of the skin afferents to motoneurons. Usually, the sensory perception threshold (T)

of the subject is used to standardize the stimulus intensity. Intensities of 1 to 3T are

used to stimulate low-threshold afferents that convey information about touch sen-

sation. 9,10 When the stimulus intensity is increased, however, as well as stimulating

the low-threshold afferent ﬁbers, the stimulus current begins to activate the higher

threshold afferents that convey information about squeeze, pinprick, and pain to the

central nervous system. Even though some methods can be developed to study the

stimulation of one type of afferents by working during the application of pressure

or local anaesthetics to the region, such experiments are technically challenging and

the results are difﬁcult to interpret. 11

Electrical stimuli can also be applied near mixed or muscle nerves in order to

directly activate afferent axons. This technique is most commonly used to activate

primary muscle spindle (Ia) afferents, since these afferents have relatively low

electrical thresholds and make monosynaptic excitatory connections onto motoneu-

rons innervating the same and synergist muscles. 12 Reproducible activation of a given

population of Ia afferents requires careful placement of the stimulating electrodes

and relatively brief (0.5 ms) stimulus pulses. 12 To ensure that a constant proportion

of Ia afferents is activated, the stimulus intensity is generally increased to a level

that will directly activate a small proportion of motoneuron axons, producing an M-

wave. A constant amplitude M-wave is then taken as evidence that the effective

stimulus strength is constant. 12 Stimuli that are suprathreshold for producing an M-

wave will also activate Golgi tendon organ (Ib) afferent ﬁbers in muscle nerves and

low-threshold cutaneous afferents in mixed nerves, complicating the interpretation

of the reﬂex effects of electrical nerve stimulation. 13

5.3.2 M ECHANICAL S TIMULATION

Mechanical stimulation is more difﬁcult to standardize than electrical stimulation

since its application requires ﬁne control of the duration, the intensity, the shape,

FIGURE 5.1 The effect of the stimulus proﬁle on the masseteric reﬂex in one subject. The

left column illustrates the force proﬁles and the right column is the CUSUM of the ipsilateral

masseter muscle SEMG. The effective force applied to the tooth is indicated in Newtons.

Stimuli were delivered at time zero. In all cases, the number of stimuli used was 150; the

push force was about 2.5 N; all had contraction levels of about 10% MVC. Note that the

probability of eliciting an excitatory reﬂex increases with the increase in the smoothness of

the force delivery. Also note that when the preload is not applied (bottom trace in left column),

even when using the smoothest force proﬁle, the speed of the force delivery increases (not

shown, cf. 14) inducing an inhibitory reﬂex response.

and the exact location of the stimulus. Furthermore, the proﬁle of the stimulus shape,

including the rate of rise of the stimulus, is difﬁcult to standardize since it changes

with the physical relationship between the stimulating probe and the stimulated area.

If this relationship is not ﬁxed, the stimulus proﬁle and the rate of rise of the stimulus

will also change. It is, therefore, best to physically connect the probe to the area to

be stimulated either by gluing, or by applying some preload to take up the slack and

the space between the probe and the area to be stimulated. The existence of a preload

minimizes the fast component of the force and, hence, is more likely to deliver the

exact stimulus proﬁle that is speciﬁed by the wave generating system (see Fig. 5.1 ).

The exact proﬁle of the delivered forceis likely to determine the relative acti-

vation of different types of mechanoreceptors, which in some cases have opposite

reﬂex effects. An example is given below where an upper incisor tooth of an

individual was stimulated using a probe attached to a vibrator. Various mechanical

force proﬁles were delivered to the tooth. Some of the proﬁles had sharp edges (top

two proﬁles in the left column of Fig. 5.1 ) that preferentially activate periodontal

mechanoreceptors that are particularly sensitive to abrupt changes in force, and

which induce inhibition in jaw closing muscles. However, when the stimulus proﬁle

is smoothed (third waveform from top in the left column in Fig. 5.1 ), the applied

force preferentially activates receptors with excitatory actions on jaw closing mus-

cles. The same force proﬁle, however, induces an inhibitory reﬂex response when

no preload is applied to take up the slack between the stimulating probe and the

tooth. This is due to the fact that the probe now moves rapidly to take up the slack

and the abrupt change in applied force (not shown here; see Figure 3 in Reference 14)

causes an inhibitory reﬂex.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: