Short-Term Plasticity in Hindlimb Motoneurons of Decerebrate Cats
DAVID J. BENNETT, HANS HULTBORN, BRENT FEDIRCHUK, AND MONICA GORASSINI
Department of Medical Physiology, The Panum Institute, University of Copenhagen, Blegdamsvej 3,
Copenhagen N, Denmark
potentials and bursting associated with these calcium channels. The facilitation of L-type Ca 2/ channels can be evoked
either by synaptic excitation or intracellular pulses; thus the
depolarization per se seems to be an essential factor, although a number of neurotransmitters and intracellular messengers may modulate the plateau properties (Russo et al.
1997).
Recently, Svirskis and Hounsgaard (1997) have investigated whether a similar depolarization-induced facilitation
occurred for the plateau potentials in turtle motoneurons.
Indeed, they demonstrated a powerful voltage-dependent facilitation of the plateau potential and the underlying inward
current with interstimulus intervals of õ4 s. In turtle motoneurons plateau potentials are mediated by L-type Ca 2/
channels, as on the dorsal horn cells (Hounsgaard and Kiehn
1989). In cat motoneurons a similar noninactivating (possibly calcium) inward current is likely involved in plateau
potentials (Hounsgaard et al. 1988; Hultborn and Kiehn
1992; Schwindt and Crill 1980a–c, 1982, 1984).
In the course of our recent study of plateaus in motoneurons in decerebrate cats, we noticed a facilitation of plateaus
with repetitive activation, similar to that described in the
turtle (referred to as warm-up) (Bennett et al. 1998). Initially, we saw this warm-up phenomenon as a nuisance and
waited long enough between trials ( ú20 s) to avoid it. However, we later realized that the effects of warm-up were so
strong that they must be functionally important. The objective of the present paper is thus to describe warm-up in
cat motoneurons, first as it occurs with intracellular current
injection and second, with more natural activation produced
from repeated (sinusoidal) muscle stretches. Although the
main part of this study involved intracellular recordings, we
also found that the effects of warm-up were easily seen in
gross EMG recordings. This finding may open new possibilities for investigating the question of whether plateaus are
present in awake animals and humans (Eken and Kiehn
1989, 1992; Gorassini et al. 1998).
INTRODUCTION
METHODS
When identical stimuli are repeatedly delivered to dorsal
horn neurons, there is an increase in the duration and rate
of discharge with each repetition (related to the wind-up
phenomenon in the central sensitization to pain), and this
can in part be attributed to a facilitation of the postsynaptic
L-type Ca 2/ channels (in rat: Morisset and Nagy 1996; in
turtle: Russo and Hounsgaard 1994, 1996a). This facilitation
is short lasting (5 s) and results in a facilitation of plateau
Intracellular recordings were made in hindlimb motoneurons of
14 decerebrate cats with approval from the local ethical committee.
The detailed methods are described in a companion paper (Bennett
et al. 1998). Plateaus were studied both by activation with intracellular current injection (triangular ramps; Figs. 1 and 2), and by
stretch reflex activation (Fig. 4; sinusoidal stretching of the triceps
surae muscle). After activation, the plateaus were often deactivated
by hyperpolarization from intracellular current or nerve stimulation
[e.g., common peroneal (CP) nerve stimulation]. The threshold
2038
0022-3077/98 $5.00 Copyright q 1998 The American Physiological Society
/ 9k2d$$oc14 J-790-7
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
Bennett, David J., Hans Hultborn, Brent Fedirchuk, and Monica Gorassini. Short-term plasticity in hindlimb motoneurons of
decerebrate cats. J. Neurophysiol. 80: 2038–2045, 1998. Cat hindlimb motoneurons possess noninactivating voltage-gated inward
currents that can, under appropriate conditions, regeneratively produce sustained increments in depolarization and firing of the cell
(i.e., plateau potentials). Recent studies in turtle dorsal horn neurons and motoneurons indicate that facilitation of plateaus occurs
with repeated plateau activation (decreased threshold and increased
duration; this phenomenon is referred to as warm-up). The purpose
of the present study was to study warm-up in cat motoneurons.
Initially, cells were studied by injecting a slow triangular current
ramp intracellularly to determine the threshold for activation of
the plateau. In cells where the sodium spikes were blocked with
intracellular QX314, plateau activation was readily seen as a sudden jump in membrane potential, which was not directly reversed
as the current was decreased (cf. hysteresis). With normal spiking,
the plateau activation (the noninactivating inward current) was
reflected by a steep and sustained jump in firing rate, which was
not directly reversed as the current was decreased (hysteresis).
Repetitive plateau activation significantly lowered the plateau activation threshold in 83% of cells (by on average 5 mV and 11 Hz
with and without QX314, respectively). This interaction between
successive plateaus (warm-up) occurred when tested with 3- to
6-s intervals; no interaction occurred at times ú20 s. Plateaus
initiated by synaptic activation from muscle stretch were also facilitated by repetition. Repeated slow muscle stretches that produced
small phasic responses when a cell was hyperpolarized with intracellular current bias produced a larger and more prolonged responses (plateau) when the bias was removed, and the amplitude
and duration of this response grew with repetition. The effects of
warm-up seen with intracellular recordings during muscle stretch
could also be recorded extracellularly with gross electromyographic (EMG) recordings. That is, the same repetitive stretch as
above produced a progressively larger and more prolonged EMG
response. Warm-up may be a functionally important form of shortterm plasticity in motoneurons that secures efficient motor output
once a threshold level is reached for a significant period. Finally,
the finding that warm-up can be readily observed with gross EMG
recordings will be useful in future studies of plateaus in awake
animals and humans.
WARM-UP OF PLATEAU POTENTIALS
2039
for the first activation of a plateau was compared with the threshold
for subsequent activations.
Intracellular recordings were made under conditions of muscle
paralysis (using pancuronium bromide; 0.6 mg/h), so gross electromyographic (EMG) responses to muscle stretch could not be
measured simultaneously. However, because the effects of the paralysis lasted õ60 min, we were able to record EMG shortly after
making intracellular recordings. EMG was recorded with pairs of
flexible stainless steel wires (Cooner AS632) inserted into the
muscle with a 22-gauge needle. Means { SD are quoted in the
text. Statistical differences were tested with the Student’s t-test.
Terminology
As a point of terminology, the term ‘‘plateau’’ is used rather
broadly to indicate that the noninactivating voltage-gated inward
currents are activated (see review Hultborn and Kiehn 1992).
These currents can be activated regeneratively and produce a sustained depolarization or increment in firing rate, thus the term
plateau. The term plateau is however misleading because it implies
that the potential and firing rate is somehow fixed. This is not the
case (see Bennett et al. 1998; Hounsgaard et al. 1988); plateau
activation simply means that an additional depolarizing current
is present, just as if a constant current were injected through a
microelectrode. We could have used the phrase ‘‘noninactivating
voltage-gated inward current’’ instead of ‘‘plateau,’’ but for brevity
and historic reasons we use the latter. Also, we will use the term
‘‘warm-up’’ to refer to the phenomena of facilitation of plateau
potentials with repeated activation in general, without regard to
the underlying mechanism.
/ 9k2d$$oc14 J-790-7
RESULTS
Effects of repeated intracellular current injection
In the results described below, we only discuss the properties of hindlimb motoneurons that were capable of exhibiting
plateaus during intracellular current injection, as described
by Bennett et al. (1998). In Fig. 1 we illustrate the frequency-current ( f-I) relations for a motoneuron with a relatively high initial threshold for plateau activation during triangular ramp current injections. In each ramp, the current
was increased to activate the plateau (at steep rise in frequency; upward arrows), and then decreased to deactivate
the plateau (to 05 nA; downward arrows; note counterclockwise hysteresis and self-sustained firing indicative of plateau) (see Bennett et al. 1998) (Fig. 2). The current ramps
were repeated with 3- to 4-s intervals between each response
to investigate their interactions. In this example, the first
current ramp initiated the plateau at a threshold frequency
of Ç35 Hz (asterisk in Fig. 1A). On the second ramp, the
plateau threshold was at only 25 Hz (Fig. 1B). Furthermore,
firing was sustained for lower current levels on the descending phase of the triangular current ramp, compared with
the first ramp. A third ramp (Fig. 1C) lowered the plateau
threshold even more, to the point where it was initiated just
after recruitment. That is, the firing rate increased steeply at
recruitment, and the cell remained on the plateau for most
of the up and downward swing of the injected current. Fur-
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
FIG. 1. Warm-up with repeated intracellular
plateau activation. Firing rates during ramp current injections into a tibial motoneuron plotted
against current (triangular current ramps as in
Fig. 2, but at 8.2 nA/s). Threshold for plateau
onset indicated with an asterisk; arrows indicate
direction of time. A–C: responses to 3 successive
ramps, one following immediately after the other
(3- to 4-s pause in firing between each ramp
response). Note the drop in plateau threshold
with each successive ramp (warm-up). D: superimposed responses from A–C, with only firing
rate during upward current ramp shown. Note the
overlap in firing in the 3 cases before the plateau
was activated (at õ25 Hz).
2040
BENNETT, HULTBORN, FEDIRCHUK, AND GORASSINI
ther ramps produced little change in the plateau threshold,
i.e., a steady-state threshold was reached so the responses
looked as in Fig. 1C, with repeated activation (see also
Fig. 4, described below). This progressive facilitation of the
plateaus with repeated activation we refer to as warm-up
(see discussion for its relation to warm-up in other cells).
Figure 2 illustrates a similar experimental protocol to examine warm-up for another cell with a somewhat lower
threshold for plateau activation. In this case the first two
triangular current injections (ramps 1 and 2) were kept subthreshold to plateau activation. That is, when the current was
first injected into such neurons, the membrane potential and
firing rate initially increased and decreased relatively linearly
with increasing and decreasing current [current ramp 1, Fig.
2A; cf. ‘‘primary slope’’ (see definition in Bennett et al.
1998], and a steep jump in frequency did not occur (cf. no
plateau activation). An identical response was obtained
when the ramp was repeated 5 s later (ramp 2; shown in
Fig. 3A). That is, no facilitation in the response occurred
/ 9k2d$$oc14 J-790-7
with repetition when firing was initially confined to these
low rates, well below the plateau threshold.
The effects of repeated intracellular current ramps were
very different when the plateau was activated (Fig. 2, B and
C). For example, when a slightly larger current ramp was
applied to the same cell, then a plateau was activated with
an initial threshold of 12 nA and 20 Hz (ramp 3 in Fig. 2B,
plateau activated at steep jump in frequency; see arrow in
Fig. 3 B), and was followed by hysteresis (in the frequencyand voltage-current plots; i.e., F-I and V -I plots) and selfsustained firing. Another ramp that followed this by a 5-s
interval (ramp 4) activated a plateau again, but at a much
lower threshold current of 3 nA. This threshold was so low
that firing had not begun at this time, and the plateau activation can be seen as a steep jump in the membrane potential,
rather than frequency (see arrow in Fig. 2C; ramp 4), followed by hysteresis in the V -I plot and self-sustained firing.
The first time the plateau was activated (ramp 3), there
was a vertical shift in F-I plot (labeled ‘‘on plateau’’ in Fig.
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
FIG. 2. Effect of repeated current injection with and
without plateaus activation. Intracellular recording from soleus motoneuron during 4 ramp current injections, one following immediately after the other, with a 5-s pause in
firing between each response (A–C). First and 2nd ramp
responses were nearly identical (see Fig. 3 A for comparison), so only the 1st is shown. All ramps at 5.8 nA/s.
Spikes clipped to amplify signal. A, left column: 1st current
ramp, up to 10 nA (subthreshold for plateau at this time).
Right column: membrane potential replotted on same vertical scale, but as a function of current during up and down
phases of ramp. Likewise, firing rate is plotted against current, with solid and open circles for up and down phases
of ramp, respectively (rate computed from average number
of spikes in 0.5-nA intervals). Arrows indicate direction
of time. B: 3rd current ramp, but with a higher peak current,
which activates a plateau, as seen by hysteresis in plots of
potential and frequency against current [voltage-current
(V -I) and frequency-current (F-I), right column]. C: 4th
current ramp, which activated the plateau at a much lower
threshold before firing began (cf. warm-up), as seen by
steep rise in potential at arrow (compare with reference
line drawn in band C), sustained firing and hysteresis in
V -I plot.
WARM-UP OF PLATEAU POTENTIALS
2041
3C), as if there were a steady current injected into the cell
(cf. inward current and plateau; see methods). The subsequent current ramp (ramp 4) caused firing that started directly on this vertically shifted region of the F-I plot (marked
‘‘on plateau’’ in Fig. 3C), presumably because the plateau
(inward current) was activated before firing began and remained on throughout.
Of the 23 cells studied 87% showed effects of warm-up
as just described, with a significant drop in threshold after
warm-up of 11.1 { 8.5 Hz (mean { SD; P õ 0.05) when
a steady state was reached within two to three current ramps
(as in Figs. 1 and 2, respectively). This warm-up was found
when we tested plateau activations at 3- to 6-s intervals. We
did not systematically test for the maximum interval where
warm-up occurred, but we found that warm-up did not occur
at intervals longer than 10–20 s. Our present material was
not sufficient to show any correlations between cell type and
the occurrence of warm-up. We also investigated warm-up
in cells with the spiking mechanism blocked by injecting
QX314 through the recording microelectrode (Bennett et al.
1998). Again, we found that the plateau threshold dropped
significantly with repeated activation; on average by 5.3 {
/ 9k2d$$oc14 J-790-7
4.0 mV [P õ 0.05, 71% of cells studied (5/7) showed
warm-up].
Stretch reflex activated plateaus and warm-up
To examine the functional relevance of the warm-up phenomenon, we investigated repeated plateau activation in motoneurons of the triceps surae muscle during repeated stretch
of this muscle. As we have shown in the companion paper
(Bennett et al. 1998), the threshold for plateau activation is
close to the recruitment level (i.e., the spike threshold),
when the motoneurons are synaptically activated in this way
(i.e., without additional current injection) (Bennett et al.
1998). Therefore it was most convenient to study the plateaus and warm-up in these motoneurons following QX314
injection to block spiking. When these cells were initially
hyperpolarized, and the muscle was stretched, there were
phasic depolarizations during stretch, and not during muscle
shortening. We suppose that this depolarization reflected the
synaptic excitation per se, from the Ia afferents (under curarizated conditions), relatively unaffected by nonlinear membrane properties (see Bennett et al. 1998), because the depo-
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
FIG. 3. Superimposed F-I plots during repeated intracellular activation. A: F-I plot for 1st and 2nd current
ramp responses for cell in Fig. 2, with direction of time
indicated by arrows. Solid and open circles for up and
down phases of ramp, respectively. Note responses identical in each case. B: F-I plot for 3rd ramp. The plateau
was activated at point shown and continued for the rest
of the up (solid symbols) and down (open symbols) phase
of the triangular ramps. C: F-I plots for 1st (triangles)
and 4th (squares) current ramps, which respectively
caused firing with the plateau entirely activated (on plateau) and not activated (off plateau), as indicated by the
superimposed F-I plot from the 3rd ramp (solid line).
2042
BENNETT, HULTBORN, FEDIRCHUK, AND GORASSINI
to also apply a steady depolarizing current bias to study them
(cf. Bennett et al. 1998). In this situation the current bias was
initially held just below the level to activate plateaus during
muscle stretch and then stepped up slightly to enable plateau
activation by muscle stretch (Fig. 4B). In the example in Fig.
4B, a 1-nA step increase in current triggered a gradual increase
and prolongation of the stretch responses, over three stretch
cycles, i.e., a warm-up. The prolonged depolarization (seconds
long) after each stretch cycle suggests that the plateau was
being activated, as before. In this situation the steady-state
response to stretch did not involve a tonic activation of the
plateau. Instead the plateau dropped off over 2–3 s after each
stretch (e.g., right half of Fig. 4B). This tendency for the
plateaus to not last (tonically) was also seen in cells that did
not exhibit warm-up and was associated with high recruitment
threshold cells that required large depolarizing currents to activate the plateaus (Bennett et al. 1998).
The gradual potentiation of the plateau, seen in Fig. 4,
was in part due to a residual depolarization left after each
phasic depolarization to the plateau level (marked with
arrows). This is in contrast to the situation described for Fig.
1D, where firing rate (and thus depolarization) remained
relatively unchanged between plateau activation. One difference in the present situation is that the cell was not actively
hyperpolarized between plateau activations. However, in
separate trials (not shown) we found that strong, transient
hyperpolarization (greater than 015 nA) of the cell between
stretch-evoked plateaus did not eliminate warm-up; thus the
effect of warm-up was not simply the accumulation of depolarization mentioned above.
Warm-up reflected in the EMG responses to muscle
stretch
Because warm-up was seen in a high percentage of motoneurons during stretch activation, its combined effect over
FIG. 4. Warm-up with repeated muscle stretch. A: intracellular recording from relatively low recruitment
threshold gastrocnemius-soleus (GS) motoneuron during
sinusoidal stretching [5.5 nA rheobase, 81 ms afterhyperpolarization (AHP) half-amplitude duration]. Firing
inactivated with intracellular QX314, and estimation of
firing threshold before inactivation indicated by horizontal line. Cell initially held hyperpolarized by 01-nA
current bias. When this current was removed, the stretch
caused progressively larger and more prolonged responses that ended in a tonic plateau activation ( far
right). B: same as A, but in high-threshold GS motoneuron (9 nA rheobase, 24 ms AHP half-amplitude duration). Initially, cell held depolarized with 8-nA bias current (subthreshold for plateau). When this current was
increased by 1 nA, warm-up was seen, as in A, but the
plateau did not remain tonically active between muscle
stretches. Arrows indicate sustained responses following
stretch.
/ 9k2d$$oc14 J-790-7
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
larizations were below the threshold for plateau activation.
Also, note that the responses remained the same with repeated stretches (i.e., no signs of warm-up at this level of
depolarization; see 5 stretch responses of left of Fig. 4A).
When the hyperpolarizing current was removed abruptly,
the responses to stretch grew progressively with repeated
stretching. For example, in Fig. 4A, the first stretch after
removal of the 01-nA bias current produced a phasic response and a sustained depolarization during shortening
(arrow). The second stretch produced an even larger and
more sustained response and so on. Finally, after the fourth
stretch the membrane potential remained depolarized (tonically; on a plateau; Fig. 4A, at right). The gradually increasing and more prolonged responses to the first four stretch
cycles (described above) likely reflected transient activations of the plateau, and the effect of these activations appeared to accumulate to potentiate subsequent plateaus, as
we have seen with the intracellular current injection (Figs.
1–3). Such warm-up during repeated stretch activation also
occurred without QX314, an example of which is shown in
Fig. 7 of Bennett et al. (1998). In total, 83% of cells tested
(10/12) showed the effects of warm-up during stretch, similar to the incidence of warm-up with intracellular current
injection.
In cells with a relatively low recruitment threshold, such
as in Fig. 4A, no intracellular bias current was needed to
activate the plateau and the subsequent warm-up. We also
know from measurements before the QX314 took effect,
that the initial recruitment level was approximately at the
horizontal dashed line indicated in Fig. 4A; thus this cell
would have normally been recruited by the stretch reflex,
and the plateau initiation would have occurred subthreshold,
as we described before (Bennett et al. 1998).
In cells with a higher recruitment threshold the muscle
stretch could not by itself activate plateaus, and thus we had
WARM-UP OF PLATEAU POTENTIALS
2043
FIG. 5. Warm-up of electromyographic (EMG) response to muscle stretch. EMG recorded from soleus
muscle during continuous sinusoidal stretching of the
whole triceps surae, as in Fig. 4. Initially, motoneuron
pool was silenced with a long inhibitory nerve stimulation
[common peroneal (CP) nerve; 5T, 100 Hz, during bar].
Note that the motor-unit responses (EMG) to stretch only
grew slowly after the pool was silenced, with a similar
time course to warm-up seen in single motoneurons
(Fig. 4).
DISCUSSION
We have shown that plateau potentials in cat motoneurons
are facilitated by a preceding activation, a phenomenon referred to as warm-up. The effects of warm-up on plateaus
are short term, lasting several seconds after a plateau activation, and thus constitute an example of short-term plasticity
in the postsynaptic membrane, similar to that described in
turtle dorsal horn neurons and motoneurons (Russo and
Hounsgaard 1994, 1996a,b; Russo et al. 1997; Svirskis and
Hounsgaard 1997). Warm-up can be seen from synaptic
input as well as current injected into the cell body. Importantly, its effects are also reflected in the extracellular records
/ 9k2d$$oc14 J-790-7
recorded with conventional gross EMG. In the following we
wish to address 1) the mechanism underlying warm-up, 2)
the relation between the reduction in plateau thresholds by
warm-up (present study) and synaptic excitation (Bennett et
al. 1998), and finally 3) the possible functional implications.
Mechanisms of warm-up
As referred to in the introduction, plateau potentials in cat
motoneurons are assumed to depend on L-type Ca 2/ currents
or similar noninactivating inward currents. Thus the observed warm-up likely involves a direct or indirect facilitation of these inward currents. Similar warm-up effects have
been analyzed both in turtle motoneurons (Svirskis and
Hounsgaard 1997) and turtle dorsal horn neurons (Russo
and Hounsgaard 1994, 1996a,b; Russo et al. 1997) and rat
dorsal horn neurons (Morisset and Nagy 1996). Rosso and
Hounsgaard found that facilitation of plateaus occurred with
and without a buildup of residual depolarization between
each plateau activation, and argued that the latter situation
would indicate that there was a direct facilitation of the
L-type Ca 2/ channels. In cat motoneurons we also saw evidence for a similar facilitation with (Fig. 4) and without
(Fig. 1 and see text) a residual depolarization (or increase
in firing rate) between plateau activations. In comparison
with facilitation of L-type Ca 2/ channels in other systems
(Dolphin 1996), it is noteworthy that the warm-up in dorsal
horn neurons and motoneurons occurs at a relatively low
level of depolarization and has a longer duration (several
seconds). This would imply that it is of importance in the
physiological range of membrane potentials and would influence synaptic integration over a period of several seconds.
Since plateaus were first described in decerebrate cat motoneurons, it was known that their activation is slow ( ú100
ms, see introduction) (Hounsgaard et al. 1988), although
the mechanism behind this was unclear. It is therefore interesting to consider the possibility that the slow plateau onset
may result from the same mechanism as warm-up (i.e., slow
kinetics intrinsic to the channels may be involved) as described by Svirskis and Hounsgaard (1995, 1997). That is,
a depolarizing pulse may initially only influence a fraction
of the Ca 2/ channels associated with the plateau, and when
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
many motor units should be reflected in the activation of
the whole muscle. We thus measured gross EMG during
stretch after making intracellular recordings as above. To do
this we waited 60 min for the muscle paralysis (pancuronium
bromide) to wear off, during which time we inserted EMG
wires. Figure 5 illustrates the results, with the same continuous sinusoidal stretch as before. Initially, many units were
typically seen in the EMG recordings; some fired phasically
and others in a sustained manner between the stretches ( left
of Fig. 5). To study warm-up, we silenced the entire motoneuron pool with a stimulation train applied to the CP nerve
during the sinusoidal stretching, as shown in Fig. 5 (thick
bar). This stimulation typically produced a gap in firing of
most motor units for Ç6–8 s, thus long enough for the
effects of previous warm-up to be reduced. Following the
stimulation, the first cycle only activated a few motor units
phasically. Subsequent stretch cycles activated more motor
units, and the firing lasted for longer with each stretch, until
by the fourth stretch cycle sustained firing occurred in some
units. Eventually, a steady state was reached similar to that
before the CP nerve stimulation. These findings are consistent with the effects of warm-up with stretch seen with intracellular recordings. One concern was that the CP stimulation
itself caused inhibition that lasted for many seconds after it
was turned off. However, this is unlikely, because we used
a similar CP stimulation during intracellular recordings and
found no hyperpolarization beyond the stimulation period
(e.g., see Fig. 6C in Bennett et al. 1998).
2044
BENNETT, HULTBORN, FEDIRCHUK, AND GORASSINI
these channels have sufficient time to be facilitated (warmup), they will depolarize the cell further, activate further
Ca 2/ channels, and thus set up a slow regenerative depolarization, limited in speed, as warm-up is. This mechanism
would also explain why stronger stimuli produce more rapid
plateau activations (more channels initially affected). Also,
when the stimulus is sufficiently weak, it may only cause a
partial activation of the plateau, and only on subsequent
activations will the plateau be fully activated, as we have
seen.
Synaptic activation of plateaus
Functional implications
Because we have demonstrated that warm-up can occur
in individual motoneurons during repeated muscle stretch
(e.g., Fig. 4), it is logical to suppose that this should be
reflected in the behavior of the entire motoneuron pool, as
measured with EMG and muscle force. Measurements of
EMG, such as in Fig. 5, support this important conclusion.
It should be kept in mind, however, that the fusimotor system
was blocked during the intracellular recordings where we
initially studied warm-up, and not during the EMG recordings (cf. muscle paralysis). However, the relevant point
in either situation is that the afferent input does not change
from stretch cycle to cycle, so the changes can be attributed
to warm-up. During intracellular recording, this condition
was met, because at hyperpolarized levels the responses to
stretch were identical with each stretch cycle (Fig. 4, left).
During EMG recording this condition was also likely to be
met, particularly because previous studies indicate that in
decerebrate cats Ia afferents respond similarly on each
stretch cycle, suggesting that fusimotor drive does not
change during repeated muscle stretch (e.g., Fig. 8 of Bennett et al. 1996).
Warm-up could thus play an important part in normal
motoneuron function during any slow repetitive movements.
Considering the effects that we have seen (Figs. 4 and 5),
/ 9k2d$$oc14 J-790-7
We thank L. Grondahl and I. Kjær for expert technical assistance.
Funding was provided by the Danish Medical Research Council, the
Michaelsen Foundation, the Alberta Heritage Foundation for Medical Research (for D. Bennett), the Canadian Medical Research Council (for B.
Fedirchuk), and a grant from the Danish Research Academy (for M. Gorassini).
Address for reprint requests: D. Bennett, 513 HMRC, Division of Neuroscience, University of Alberta, Edmonton, Alberta T6G 2S2, Canada.
Received 26 September 1997; accepted in final form 17 June 1998.
REFERENCES
BENNETT, D. J., DE SERRES, S. J., AND STEIN, R. B. Regulation of soleus
muscle spindle sensitivity in decerebrate and spinal cats during postural
and locomotor activities. J. Physiol. (Lond.) 495.3: 835–850, 1996.
BENNETT, D. J., HULTBORN, H., FEDIRCHUK, B., AND GORASSINI, M. Synaptic activation of plateaus in hindlimb motoneurons of decerebrate cats.
J. Neurophysiol. 80: 2023–2037, 1998.
DOLPHIN, A. C. Facilitation of Ca 2/ current in excitable cells. Trends Neurosci. 19: 35–43, 1996.
EKEN, T. AND KIEHN, O. Bistable firing properties of soleus motor units in
unrestrained rats. Acta Physiol. Scand. 136: 383–394, 1989.
EKEN, T. AND KIEHN, O. Discontinuous changes in discharge patterns of
human motor units (Abstract). J. Physiol. (Lond.) 452: 277P, 1992.
GORASSINI, M., BENNETT, D. J., SIU, M., AND YANG, J. Reduction in recruitment threshold of human motor units by repeated muscle activation. Soc.
Neurosci. Abstr. 23: 2094, 1997.
GORASSINI, M., BENNETT, D. J., AND YANG, J. F. Self-sustained activation
of human motor units. Neurosci. Lett. 247: 13–16, 1998.
HOUNSGAARD, J., HULTBORN, H., JESPERSEN, B., AND KIEHN, O. Bistability
of alpha-motoneurones in the decerebrate cat and in the acute spinal cat
after intravenous 5-hydroxytryptophan. J. Physiol. (Lond.) 405: 345–
367, 1988.
HOUNSGAARD, J. AND KIEHN, O. Serotonin-induced bistability of turtle mo-
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
The finding that depolarization from intracellular or synaptic input can cause warm-up and thus lower the plateau
threshold leads to the possibility that warm-up might, in
part, underlie the lowered plateau threshold seen with tonic
synaptic excitation (tonic EPSPs) in the companion paper
(Fig. 2 of Bennett et al. 1998). Two situations need to be
considered to rule out this possibility. First, the effect of the
repeated plateau activation in the companion paper probably
did not in itself cause a lowering of the plateau threshold,
for as we have already stated, there were long intervals ( ú20
s) between plateau activations, and warm-up has only shortterm effects ( Ç5 s). Second, the tonic EPSPs could have
caused warm-up, even though they did not by themselves
cause plateau activation (or firing). Such a subthreshold
warm-up is certainly conceivable, because Russo and Hounsgaard (1996a) have reported that the warm-up can occur
‘‘just subthreshold’’ to the plateau. In our case we have not
found that subthreshold activations cause warm-up (Figs. 3
and 4), although this issue clearly warrants more specific
investigation.
we find it remarkable that progressively larger and more
prolonged motor outputs are not more commonly produced
during repetitive tasks. Likely, in such tasks as locomotion,
the effect of warm-up (if present) must somehow be taken
into account in the generation of rhythmic drive to the motoneurons. Also, because plateau activation and warm-up are
both processes that occur relatively slowly, they would not
likely be involved in brief activations (e.g., tendon tap or
H-reflex).
Recent evidence from human motor-unit recordings suggests that plateaus may occur in humans (cf. self-sustained
firing) (Eken and Kiehn 1992; Gorassini et al. 1998). The
phenomenon of warm-up, seen in extracellular recordings
(EMG) in the present study, should provide a useful new
tool for studying plateaus in humans (and awake animals).
Indeed, Gorassini et al. (1997, 1998) have recently shown
that the duration of sustained motor-unit firing following
phasic muscle vibration progressively increases on repeated
activation of the motor unit (with vibration), as for the cat
motoneurons in Figs. 4 and 5. Warm-up may also explain
the lowering of recruitment thresholds seen in human motor
units during repetitive activation (Gorassini et al. 1997,
1998; Suzuki et al. 1990). In summary, together the effects
of repeated depolarization (warm-up) and synaptic activation (companion paper; Bennett et al. 1998) seen in cat
motoneurons provide powerful mechanisms of changing the
threshold and gain of the motoneuron. These effects, if present in humans, are likely to influence the effectiveness of
recruitment and force production in many motor tasks, especially those that involve repetitive movements.
WARM-UP OF PLATEAU POTENTIALS
/ 9k2d$$oc14 J-790-7
SCHWINDT, P. C. AND CRILL, W. E. Properties of a persistent inward current
in normal and TEA-Injected motoneurons. J. Neurophysiol. 43: 1700–
1724, 1980b.
SCHWINDT, P. C. AND CRILL, W. E. Effects of barium on cat spinal motoneurons studied by voltage clamp. J. Neurophysiol. 44: 827 – 846,
1980c.
SCHWINDT, P. C. AND CRILL, W. E. Factors influencing motoneuron rhythmic firing: results from a voltage-clamp study. J. Neurophysiol. 48: 875–
890, 1982.
SCHWINDT, P. C. AND CRILL, W. E. Membrane properties of cat spinal motoneurons. In: Handbook of the Spinal Cord, edited by R. Davidoff. New
York: Dekker, 1984, p. 199–242.
SUZUKI, S., HAYAMI, A., SUZUKI, M., WATANABE, S., AND HUTTON, R. S.
Reductions in recruitment force thresholds in human single motor
units by successive voluntary contractions. Exp. Brain Res. 82: 227 –
230, 1990.
SVIRSKIS, G. AND HOUNSGAARD, J. Depolarization-induced facilitation of
L-type Ca 2/ channels underlies slow kinetics of plateau potentials in
turtle motoneurons. Soc. Neurosci. Abstr. 21: 65, 1995.
SVIRSKIS, G. AND HOUNSGAARD, J. Depolarization-induced facilitation of
plateau generating currents in ventral horn neurons in the turtle spinal
cord. J. Neurophysiol. 79: 45–50, 1998.
09-17-98 13:52:34
neupas
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 15, 2017
toneurones caused by a nifedipine-sensitive calcium plateau potential. J.
Physiol. (Lond.) 414: 265–282, 1989.
HULTBORN, H. AND KIEHN, O. Neuromodulation of vertebrate motor neuron
membrane properties. Curr. Opin. Neurobiol. 2: 770–775, 1992.
MORISSET, V. AND NAGY, F. Modulation of regenerative membrane properties by stimulation of metabotropic glutamate receptors in rat deep dorsal
horn neurons. J. Neurophysiol. 76: 2794–2798, 1996.
RUSSO, R. E. AND HOUNSGAARD, J. Short-term plasticity in turtle dorsal
horn neurons mediated by L-type Ca 2/ channels. Neuroscience 61: 191–
197, 1994.
RUSSO, R. E. AND HOUNSGAARD, J. Plateau-generating neurones in the dorsal horn in an in vitro reparation of the turtle spinal cord. J. Physiol.
(Lond.) 493: 39–54, 1996a.
RUSSO, R. E. AND HOUNSGAARD, J. Burst-generating neurones in the dorsal
horn in an in vitro preparation of the turtle spinal cord. J. Physiol. (Lond.)
493: 55–66, 1996b.
RUSSO, R. E., NAGY, F., AND HOUNSGAARD, J. Modulation of plateau properties in dorsal horn neurones in a slice preparation of the turtle spinal
cord. J. Physiol. (Lond.) 499: 459–474, 1997.
SCHWINDT, P. C. AND CRILL, W. E. Role of a persistent inward current in
motoneuron bursting during spinal seizures. J. Neurophysiol. 43: 1296–
1318, 1980a.
2045