Brain Research Protocols 2 Ž1997.

53–58

Protocol

A method for potentiating Renshaw cell activity in humans

R. Mazzocchio ) , A. Rossi
Unita` di Malattie del Sistema Motorio e Scienze del MoÕimento, Istituto di Scienze Neurologiche, UniÕersita` di Siena, Siena, Italy
Accepted 8 July 1997

Abstract

Owing to the introduction of a special electrophysiological method w3x it has been possible to study spinal recurrent inhibition in
humans. The method, however, is indirect and, being based on an H reflex technique, can only be tested in motor nuclei from which a
large monosynaptic response can be obtained. We have developed a complementary method by which pharmacological stimulation of
Renshaw cells is obtained w6x. It exploits the central cholinergic properties of L-acetylcarnitine w18x, a substance which most likely acts
potentiating the synaptic drive of the motoneurone collaterals w7x, known to be the main source of excitation of Renshaw cells w15,20x.
The use of L-acetylcarnitine has allowed to establish the validity of the original methodology w6x and to confirm the presence of recurrent
inhibition, tested either by the H reflex technique w11–13x or, if not possible, by the PSTH technique andror rectified averaged EMG
analysis w1,4x, in many limb motor nuclei. Thus, it can be expected that L-acetylcarnitine may be used as an independent means for
identifying changes in motoneuronal activity related or attributed to the influence of Renshaw cells. q 1997 Elsevier Science B.V.

Themes: Motor systems and sensorimotor integration

Topics: Spinal cord and brainstem
Keywords: Recurrent inhibition; Renshaw cell; Spinal cord; H reflex; L-Acetylcarnitine; Motor control

1. Type of research 3. Materials

Ø Study of the functional properties of spinal recurrent
inhibition of homonymous and heteronymous a-
motoneurones in lower and upper limb muscles.
Ø Study of the interactions of Renshaw cells with neu-
ronal elements other than a-motoneurones Že.g., Ia
inhibitory interneurones..
Ø Study of segmental and descending influences on Ren-
shaw cell activity.
Ø Study of the variations in the amount of recurrent
inhibitory activity in patients with motor disorders.
4. Detailed procedure
Ø Special equipment. A two-channel electrical stimulator
ŽGrass S88, West Warwick, RI, USA., two constant
current units ŽGrass CCUI., electrodes for transcuta-
neous nerve stimulation and for recording electromyo-
graphic activity.
Ø Drugs. L-Acetylcarnitine ŽNicetile w . from Sigma-Tau
ŽPomezia, Italy..

2. Time required This section describes the electrophysiological and

The time required to run the protocol is about 90 min. It
includes a period Žbefore L-acetylcarnitine administration.
during which the parameters of stimulation are set to
produce the appropriate motoneurone responses on which
the effect of the drug is tested.
pharmacological procedure for testing recurrent inhibition
in humans. The procedure was approved by the Local
Ethical Committee.
4.1. Electrophysiological method

)
Corresponding author. Present address: Istituto di Neurochirurgia,
Universita` di Siena, Viale Bracci, I-53100 Siena, Italy. Fax: q39 Ž577.
58-6153; E-mail: mazzocchio@unisi.it
4.1.1. Placement of stimulating and recording electrodes
The subjects are healthy adult volunteers from whom
informed consent is obtained. They are seated and relaxed

1385-299Xr97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.

PII S 1 3 8 5 - 2 9 9 X Ž 9 7 . 0 0 0 2 8 - 7
54 R. Mazzocchio, A. Rossi r Brain Research Protocols 2 (1997) 53–58

either with the elbow semiflexed Ž1208., the forearm
pronated and the hand immobilized or with the examined
leg semiflexed at the hip Ž1208., the knee flexed to 1608,
the ankle in 1108 plantar flexion and the foot fastened to a
footplate. For the lower limb, stimulation of the posterior
tibial nerve Žrectangular pulses of 0.5–1 ms duration. is
obtained by placing the active electrode Žcathode. in the
shape of a half-ball Ž2 cm in diameter. in the popliteal
fossa with the reference electrode Žanode. fixed to the
anterior aspect of the knee. For the upper limb, stimulation
of the median nerve is obtained through bipolar electrodes
placed 2 cm apart in the cubital fossa. The same stimulat-
ing electrodes provide the conditioning and testing stimuli.
EMG responses Ždirect maximum motor response – M max
wave, and reflex responses – H1 and HX reflexes. are
recorded by pairs of non-polarizable disc electrodes Ž0.9
cm in diameter. placed over the mid-bellies of soleus and
wrist flexor muscles, respectively, and measured in terms
of the amplitude of muscle potential.
4.1.2. Estimation of recurrent inhibition in humans
The methodology was originally described by Bussel
and Pierrot-Deseilligny w3x. A first stimulus ŽS1. is applied
so as to obtain an H reflex ŽH1.. This is the conditioning
stimulus because it produces a motoneurone discharge
activating Renshaw cells orthodromically via the axon
collateral. Five to 10 s later, a second stimulus ŽS2. such
as to elicit a maximal M wave is applied to the same nerve
through an independent channel of stimulation. Ten sec-
onds later, S1 and S2 are given together with an interstim-
ulus interval of 10–12 ms. When so combined, a new H
reflex response is observed after the M wave, the HX test
reflex ŽFig. 1A.. The size of the HX test reflex is strictly
dependent on the size of the H1 reflex, as shown in Fig. 2.
The relationship between the H1 and HX reflex is such that
increasing the intensity of the conditioning stimulus, and
therefore the size of the H1 reflex, will cause the HX reflex
to initially grow in parallel with the H1 reflex. Beyond a
certain point, however, the HX will progressively become
smaller with increasing H1 amplitudes. This decay phase
of the HX reflex is due to the growing inhibitory effect of
Renshaw cells on parent a-motoneurones w3x. Briefly, the
evidence is the following. Ža. The acute administration of
X
L-acetylcarnitine has no effect on H reflexes elicited by
conditioning discharges of small amplitude, i.e., prior to
the onset of the decay phase, whereas an additional in-
hibitory effect is observed on HX reflexes produced by H1
reflexes of intermediate and maximum amplitude w6x. The
ability of L-acetylcarnitine to further decrease the size of
HX is a function of the number of motoneurones involved
in the H1 conditioning discharge; the stronger the condi-
tioning discharge, and therefore the amount of Renshaw
cells recruited via the recurrent collateral, the stronger the
inhibitory effect on the motoneurone pool w6,7x. Žb. In all
the experimental conditions in which a modulation of
recurrent inhibition has been reported, significant changes
in the size of the HX have been observed only within the
decay phase of the H1rHX curve w8x.
Thus, to ensure that Ža. changes in HX size are depen-
dent on variation in the amount of recurrent inhibition, and

Fig. 1. Comparison between soleus EMG responses before ŽA., 45 min ŽB. and 80 min ŽC. after the acute administration of L-acetylcarnitine. The EMG
responses from the soleus muscle were obtained by electrical stimulation of the posterior tibial nerve ŽPTN. at the popliteal fossa. 1 s isolated conditioning
PTN stimulus ŽS1. producing an H1 reflex of nearly maximal size. 2 s isolated test PTN stimulus ŽS2., supramaximal to the a-motor fibres, causing a
X
maximal motor response ŽM ma x .. 3 s combined conditioning and test stimuli ŽS1-S2. at 10 ms interval producing an H reflex response after the M max .
 R. Mazzocchio, A. Rossi r Brain Research Protocols 2 (1997) 53–58 55

Žb. an adequate number of Renshaw cells is activated, the

size of the conditioning H1 reflex should be about 80% of
its maximum amplitude. Larger H1 reflexes may give rise
to very small HX reflexes or even produce suppression of
the response. In these cases, no sizeable estimation of
recurrent inhibition is possible.

4.2. Pharmacological data

L-Acetylcarnitine has a close structural similarity to

acetylcholine w10x, it acts on the same receptor sites as
acetylcholine w2x and it has a central cholinergic effect in
animals w18x. This action is largely mediated via nicotinic
receptors since it is preferentially antagonized by dihydro-
b-erythroidine w18x. L-Acetylcarnitine produces a delayed
and prolonged facilitation of neuronal responses to acetyl-
choline potentiating cholinergic transmission considerably
more than a simple summation of effects can do w18x. It is
suspected that L-acetylcarnitine, beside exerting a direct
effect, may also act as a donor of acetyl groups, shuttling
them from mitochondria to the cytosol where they can be

X
Fig. 3. The time course of soleus ŽA. and wrist flexor ŽB. H reflex,
expressed as a percentage % of the maximal motor response M ma x ., is
Ž . Ž
plotted as a function of the time from the onset of L-acetylcarnitine
X
infusion Žhorizontal bar at the bottom.. Time 0 corresponds to control H
X
values obtained before L-acetylcarnitine infusion. The value of the H
amplitude on which the effect of the drug is tested is indicated by the
arrows in Fig. 2A,B. Each point is the mean of 12 measurements. Vertical
bars represent 1 S.D. of the mean.

used for acetylcholine synthesis w19x. L-Acetylcarnitine has

a low acute toxicity ŽLD50 s 1.5–3 grkg i.v. in the rat.
and can be safely administered in humans w16x. Plasma
concentrations of L-acetylcarnitine increase quickly after
intravenous administration and rapidly decline within 4 h
returning to base values within 12 h w5x.

4.3. Experimental protocol

Two intravenous infusion sets are prepared. One is

connected with a 100 ml bottle of sterile saline in which
L-acetylcarnitine hydrochloride Ž30 mgrkg. has been dis-
solved. The other is connected to another bottle containing
sterile saline only. Both sets are connected to the subject
X
Fig. 2. Pattern of variations of the soleus ŽA. and wrist flexor ŽB. H test
by a two-way tap. A continuous intravenous infusion of
reflex after a conditioning H1 discharge in a representative subject. The saline is started at very slow rate. After the subject has
X
amplitude of the H reflex is plotted against that of the H1 reflex, both completely relaxed, the appropriate stimulation parameters
expressed as a percentage Ž%. of the maximal motor response ŽM ma x .. are set. Maximal M waves, H1 reflexes of near-maximal
X
The decay phase is comprised between the maximum size of the H reflex
X size and the respective HX reflexes are recorded and stored
and the smallest H size possible following a conditioning H1 reflex of
maximum amplitude. Each point is the mean of five measurements. The
for later analysis. Time is run in parallel with the onset of
identity line represents the theoretical curve which would be obtained if L-acetylcarnitine administration which is made to last 15–
X
H equalled H1. 20 min Žabout 100 dropsrmin.. During this period, fewer
56 R. Mazzocchio, A. Rossi r Brain Research Protocols 2 (1997) 53–58
responses can be collected Ževery 5 min. just to make sure
that the sizes of the H1 reflex and of the maximum M
wave remain constant throughout. At the end of L-acetyl-
carnitine administration, saline infusion only is resumed
for at least 60 min. EMG responses are continuously
recorded and stored until the end of the experiment.
4.4. Data analysis
The amplitude of H reflexes is automatically measured
Žpeak to peak. by computer. The HX reflexes occurring
before and within every 10 min from L-acetylcarnitine
administration are averaged and the results plotted as a
function of the time elapsed from the onset of drug infu-
sion, in correspondence with the centre of each time
interval. For example, the mean value at 50 min comprises
those HX reflexes collected between 45 and 55 min. The HX
reflexes are expressed as a percentage of the maximum M
wave. The HX reflexes evoked before drug infusion repre-
sent the control values and are plotted at time 0. Thus, the
changes in HX amplitude induced by L-acetylcarnitine are
evaluated by comparing pre- and post-drug infusion values
Žsee Fig. 1.. The mean and standard error of the mean are
computed for all single HX reflexes evoked in a given time
interval, and compared by two-tailed Student’s t-test. Usu-
ally, there is no need for plotting the time course of the H1
reflex and of the maximum M wave, since their amplitudes
are not significantly affected by L-acetylcarnitine ŽFig. 1..
It is, however, important to check that this is so throughout
the experiment Žsee Section 6.1..
4.5. AdÕerse reactions
No adverse reaction has been observed during and after
L-acetylcarnitine infusion. A burning sensation at the injec-
tion site and a cooling sensation along the arm during drug
infusion are quite common. Rarely may nausea and vomit-
ing about 50–60 min after drug infusion occur.
5. Results
Fig. 2 illustrates the typical pattern of variations in the
soleus ŽFig. 2A. and wrist flexor ŽFig. 2B. HX reflex size
as the conditioning H1 reflex amplitude is increased by
augmenting the strength of the conditioning stimulus ŽS1..
At low H1 reflex amplitude, the HX closely follows H1 up
to a certain point, coinciding with HX maximum value,
beyond which, as the H1 is increased, there is a progres-
sive decay in the size of the HX reflex. To test the effects
of L-acetylcarnitine on the soleus and wrist flexor HX
reflex, the intensity of S1 is adjusted to have an H1
conditioning reflex about 80% of its maximum amplitude.
About 30 min after injecting L-acetylcarnitine ŽFig. 3., a
decrease in the amplitude of the HX reflex becomes evi-
dent. The peak of the effect is reached between 50 and 60
min from the onset of drug infusion. The effect is usually
over by 70–80 min when the amplitude of HX attains the
values observed before injecting the drug Žtime 0.. Such
delayed and prolonged effect may be compatible with an
indirect facilitation of cholinergic transmission Žsee Sec-
tion 4.2..
Using the infusion parameters above specified, L-
acetylcarnitine is able to enhance the activity of Renshaw
cells for about 30 min during which the influence of these
interneurones on their target cells can be studied.
6. Discussion
Control experiments described elsewhere have ruled out
that the effect of L-acetylcarnitine on H reflexes may also
be mediated through changes in motoneurone afterhyper-
polarization, post-activation depression, presynaptic Ia in-
hibition, Ib inhibition and neuromuscular transmission
w6,9,12x. Therefore, the decrement of the HX reflex after
L-acetylcarnitine administration is produced by a specific
phasic enhancement of Renshaw cell response. Moreover,
it has been shown that the recurrent inhibitory activity of
Renshaw cells on target cells can be elicited by L-acetyl-
carnitine in the absence of motor discharge w9,14,17x.
6.1. Troubleshooting
One general problem concerns the possible reduction of
the size of the maximum M wave and, consequently, of the
H reflexes after L-acetylcarnitine administration. This is
very rare and never below 10% of the control value. It may
be related to a dose-dependent effect on neuromuscular
transmission or to a minor anticholinergic action w2x. A
more specific problem may be met using the paired H
reflex technique for testing recurrent inhibition in motor
nuclei other than the soleus. Although the conditioning
stimulus S1 evokes an H1 conditioning reflex of maximum
amplitude, this may be insufficient to produce a decay
phase of the HX reflex on which to test the effect of
L-acetylcarnitine ŽFig. 4A.. This is an important point
since the efficacy of L-acetylcarnitine is directly related to
the strength of recurrent inhibition w6,7x.
6.2. AlternatiÕe and support protocols
The best way to deal with the first problem is to repeat
the experiment using a lower dose of L-acetylcarnitine Ž15
mgrkg.. This is usually sufficient to produce changes in
recurrent activity without affecting neuromuscular trans-
mission. The second problem can be overcome by intro-
ducing a second conditioning stimulus preceding S1 by
3–5 ms. The intensity of this stimulus is below the thresh-
old for obtaining the H1 reflex; however, its effect will be
a subliminal facilitation of the motoneurone pool so that
S1 will recruit many more motoneurones thus producing a
 R. Mazzocchio, A. Rossi r Brain Research Protocols 2 (1997) 53–58
Fig. 4. A: as in Fig. 2B. In this case, however, it is not possible to obtain
X
a decay phase of the H size even with maximal H1 amplitudes. B: an
additional increase in the size of the conditioning H1 amplitude Žfrom
35% to 50% of the M ma x . is obtained by introducing a second condition-
ing stimulus Žbelow the threshold for the H reflex. preceding the stimulus
ŽS1. for the H1 by 3–5 ms Ži.e., evoking the homonymous facilitation of
X
the H reflex.. This is enough to produce a decay phase of the H reflex,
making thus possible the assessment of recurrent inhibition.
significant increase in the size of the H1 reflex w4,11,12x.
This is usually enough to elicit a decay phase of the HX
reflex Žsee Fig. 4B..
7. Quick procedure
7.1. Measuring the effect of L-acetylcarnitine
A. Dissolve 30 mgrkg of L-acetylcarnitine hydrochlo-
ride in 100 ml of sterile saline; start an intravenous infu-
sion of saline from a separate bottle at a very slow rate.
B. Place pairs of surface stimulating Žtibial or median
nerve. and recording Žsoleus or wrist flexor muscles.
electrodes.
C. Use two independent channels of stimulation con-
necting them to two constant-current isolation units. Link
the two units together so that the same stimulating elec-
trodes provide the conditioning and testing stimuli. Adjust
the intensity so as to obtain a nearly maximal H reflex
from channel 1, a maximal M response from channel 2 and
an HX reflex response from both channels using an inter-
stimulus interval of 10–15 ms.
57
D. Start the measurements and collect about 60 re-
sponses for each parameter. If the responses are stable, run
the time and start L-acetylcarnitine infusion Ž100
dropsrmin.. Record the time for each response you col-
lect.
E. Resume saline infusion only at the end of L-acetyl-
carnitine administration; keep collecting and recording all
responses for the next 60 min.
F. Stop the measurements and saline infusion; save the
data.
7.2. Viewing the results
G. Average the HX reflexes occurring before and within
every 10 min from L-acetylcarnitine infusion.
H. Plot the results as a function of the time elapsed
from the onset of L-acetylcarnitine administration.
8. Essential literature references
Original papers: Refs. w3,6,18x.
Review papers: Ref. w20x.
Acknowledgements
We should like to thank all the volunteers who partici-
pated in this study. L-Acetylcarnitine was kindly supplied
by Sigma-Tau Co., Pomezia ŽItaly.. This work was in part
financed by grants from the Italian MURST Ž60%..
References
w1x C. Aymard, B. Decchi, R. Katz, C. Lafitte, A. Penicaud,
´ S. Raoul,
A. Rossi, Recurrent inhibition between motor nuclei innervating
opposing wrist muscles in the human upper limb, J. Physiol. 499
Ž1997. 267–282.
w2x K. Blum, E. Seifter, J. Seifter, The pharmacology of D- and L-carni-
tine and L-acetylcarnitine: comparison with choline and acetyl-
choline, J. Pharmacol. Exp. Ther. 178 Ž1971. 331–338.
w3x B. Bussel, E. Pierrot-Deseilligny, Inhibition of human motoneu-
rones, probably of Renshaw origin, elicited by an orthodromic motor
discharge, J. Physiol. 269 Ž1977. 319–339.
w4x R. Katz, R. Mazzocchio, A. Penicaud, A. Rossi, Distribution of
homonymous and heteronymous recurrent inhibition in the human
upper limb, Acta Physiol. Scand. 149 Ž1993. 183–198.
w5x A. Marzo, E. Arrigoni Martelli, R. Urso, M. Rocchetti, V. Rizza,
J.G. Kelly et al., Metabolism and disposition of intravenously ad-
ministered acetyl-L-carnitine in healthy volunteers, Eur. J. Clin.
Pharmacol. 37 Ž1989. 59–63.
w6x R. Mazzocchio, A. Rossi, Further evidence for Renshaw inhibition
in man. A combined electrophysiological and pharmacological ap-
proach, Neurosci. Lett. 106 Ž1989. 131–136.
w7x R. Mazzocchio, M. Schieppati, C. Scarpini, A. Rossi, Enhancement
of recurrent inhibition by intravenous administration of L-acetyl-
58 R. Mazzocchio, A. Rossi r Brain Research Protocols 2 (1997) 53–58
carnitine in spastic patients, J. Neurol. Neurosurg. Psychiatry 53
Ž1990. 321–326.
w8x R. Mazzocchio, A. Rossi, J.C. Rothwell, Depression of Renshaw
recurrent inhibition by activation of corticospinal fibres in human
upper and lower limb, J. Physiol. 481 Ž1994. 487–498.
w9x R. Mazzocchio, A. Rossi, Involvement of spinal recurrent inhibition
in spasticity. Further insight into the regulation of Renshaw cell
activity, Brain 120 Ž1997. 991–1003.
w10x K.W. Reed, W.J. Murray, E.B. Roche, Acetylcarnitine and choliner-
gic receptors, J. Pharmacol. Sci. 69 Ž1980. 1065–1068.
w11x A. Rossi, R. Mazzocchio, Presence of homonymous recurrent inhibi-
tion in motoneurones supplying different lower limb muscles in
humans, Exp. Brain Res. 84 Ž1991. 367–373.
w12x A. Rossi, R. Mazzocchio, Renshaw recurrent inhibition to motoneu-
rones innervating proximal and distal muscles of the human upper
and lower limbs, in: L. Jami, E. Pierrot-Deseilligny, D. Zytnicki
ŽEds.., Muscle Afferents and Spinal Control of Movement, Perga-
mon, Oxford, 1992, pp. 313–319.
w13x A. Rossi, A. Zalaffi, A. Decchi, Heteronymous recurrent inhibition
from gastrocnemius muscle to soleus motoneurones in humans,
Neurosci. Lett. 169 Ž1994. 141–144.
w14x A. Rossi, B. Decchi, A. Zalaffi, R. Mazzocchio, Group Ia non-re-
ciprocal inhibition from wrist extensor to flexor motoneurones in
humans, Neurosci. Lett. 191 Ž1995. 205–207.
w15x R.W. Ryall, Cholinergic transmission in the spinal cord, in: R.A.
Davidoff ŽEd.., Handbook of the Spinal Cord, Dekker, New York,
1983, pp. 203–240.
w16x M. Schieppati, I. Gritti, R. Mazzocchio, A. Rossi, M. Mancia,
Motoneurone recurrent inhibition is enhanced by L-acetylcarnitine in
humans, Electromyogr. Clin. Neurophysiol. 29 Ž1989. 73–80.
w17x M. Schieppati, I. Gritti, C. Romano’, Recurrent and reciprocal
inhibition of the human monosynaptic reflex shows opposite changes
following intravenous administration of acetylcarnitine, Acta Phys-
iol. Scand. 143 Ž1991. 27–32.
w18x E. Tempesta, L. Janiri, C. Pirrongelli, Stereospecific effects of
acetylcarnitine on the spontaneous activity of brainstem neurones
and their responses to acetylcholine and serotonin, Neuropharmacol-
ogy 24 Ž1985. 43–50.
w19x H.L. White, P.W. Scates, Acetyl-L-carnitine as a precursor of acetyl-
choline, Neurochem. Res. 15 Ž1990. 597–601.
w20x U. Windhorst, On the role of recurrent inhibitory feedback in motor
control, Prog. Neurobiol. 49 Ž1996. 517–587.